Plant and Cell Physiology

Igawa, Tomoko;Ochiai-Fukuda, Tetsuko;Takahashi-Ando, Naoko;Ohsato, Shuichi;Shibata, Takehiko;Yamaguchi, Isamu;Kimura, Makoto

doi: 10.1093/pcp/pch195pmid: 15564518

Abstract TAXI-I (Triticum aestivumxylanase inhibitor I) is a wheat grain protein that inhibits arabinoxylan fragmentation by microbial endo-β-1,4-xylanases used in the food industry. Although TAXI was speculated to be involved in counterattack against pathogens, there is actually no evidence to support this hypothesis. We have now demonstrated the presence of TAXI family members with isolation of two mRNA species, Taxi-III and Taxi-IV. At the nucleotide sequence level, Taxi-III and Taxi-IV were 91.7% and 92.0% identical, respectively, to Taxi-I, and Taxi-III and Taxi-IV were 96.8% identical. Accumulation of Taxi-III/IV transcripts was most evident in roots and older leaves where transcripts of Taxi-I were negligible. When challenged by fungal pathogens Fusarium graminearum and Erysiphe graminis, the concentrations of Taxi-III/IV transcripts increased significantly. In contrast, the increases in Taxi-I transcripts in response to these pathogens were rather limited. Both Taxi-I and Taxi-III/IV were strongly expressed in wounded leaves. The upstream region of Taxi-III contained W boxes and GCC boxes, which are sufficient to confer pathogen and wound inducibility on promoters. Recombinant TAXI-III protein inhibited Aspergillus niger and Trichoderma sp. xylanases: it was also active against some spelt xylan-induced xylanases of F. graminearum. These features suggest that some, but not all, TAXI-type xylanase inhibitors have a role in plant defense. (Received August 10, 2004; Accepted September 12, 2004) Introduction Plant pathogens secrete an array of cell-wall-degrading enzymes as offensive arsenals to penetrate the host cell wall and to release the nutrients from their building blocks. Representatives of such enzymes include cellulases (e.g. endo-β-1,4-glucanase), pectinases [e.g. pectin lyase, polygalacturonase (PG)] and endo-β-1,4-xylanase, which are considered to be virulence factors for the pathogens (Walton 1994). These classes of enzymes are also produced by plants themselves and participate in many physiological processes such as the mobilization of reserves from storage organs and remodeling of the plant cell walls during growth and development. Hence, a highly specific and organized regulation of cell wall depolymerase activities is important for various physiological processes in plants. Plants appear to have evolved wall depolymerase-inhibiting proteins that help to limit pathogen colonization, either directly by acting as inhibitors of the pathogen’s virulence factors (i.e. depolymerases) or indirectly by ensuring production of elicitor-active polysaccharides that might have been extensively degraded in their absence. Well-known examples of such plant inhibitors are the PG-inhibiting proteins (PGIPs), which are found in a variety of dicotyledonous plants and pectin-rich monocotyledonous plants (e.g. onion and leek) (De Lorenzo et al. 2001). Other examples include a tomato inhibitor protein that inhibits the activity of the xyloglucan-specific endoglucanase of Aspergillus aculeatus (Qin et al. 2003). A protein that inhibits the pectin lyase of Rhizoctonia solani was reportedly discovered in sugar beet (Bugbee 1993), but a detailed characterization has yet to be reported. Recently, proteins that inhibit family 11 endo-β-1,4-xylanases were found in wheat flour (McLauchlan et al. 1999, Gebruers et al. 2001). These inhibitor proteins (which belong to different families designated XIP and TAXI) were purified to homogeneity and the corresponding genes (i.e. Xip and Taxi-I, respectively) were cloned and characterized (Elliott et al. 2002, Fierens et al. 2003). However, these proteins were identified in the search for causative substances that reduce the activity of xylanases used in food processing (e.g. breadmaking, pasta processing, gluten-starch separation, preparation of animal feeds), and it is not yet known whether these inhibitors are involved in counterattack against pathogens in wheat plants (Sansen et al. 2004). Here we report on the cloning, structural characterization and expression analyses of new genes that belong to the TAXI family. Results Cloning of the Taxi-I gene from wheat cv. Norin 61 We first isolated a complete cDNA of Taxi from wheat cv. Norin 61 based on an assembled contig of wheat expressed sequence tags (ESTs), which included the sequence encoding the 10 kDa domain of the TAXI polypeptides reported previously (see Fig. 1A) (Gebruers et al. 2001). The missing 5′ and 3′ portions of these overlapping ESTs were amplified from cDNA synthesized from total RNA of immature embryo (3 weeks after anthesis) by rapid amplification of cDNA ends (RACE). The cloned cDNA (accession number AB114626) encodes a protein of 402 amino acids and is identical to the recently reported Taxi-I sequence (AJ438880) of wheat cv. Estica (Fierens et al. 2003). Using primers Isp(For) and Isp(Rev), we were also able to amplify Taxi-I from the cDNA of other wheat cultivars including Norin 12 and Florida. Their nucleotide sequences were identical to those of cultivars Norin 61 and Estica, suggesting that Taxi-I is well conserved among different wheat cultivars. Cloning and characterization of Taxi-III and Taxi-IV Southern analysis using Taxi-I as a probe revealed multiple hybridization bands on a blot of DNA digested with restriction enzymes that do not cut the probe, which is indicative of the presence of a multigene TAXI family in hexaploid wheat (Fig. 1B). To gain insight into the structure and function of the TAXI family genes, we isolated the genes of other family members using degenerate primers TAXI-F1 and TAXI-R2 (represented as green arrows in Fig. 1A; see Table 1 for primer sequences). Closely related genes were successfully amplified from cDNA of root where the Taxi transcripts were most abundant (see next section). Sequencing a dozen reverse transcription-PCR (RT-PCR) products allowed discrimination of the original sequence variations of the TAXI family genes from PCR errors. Among these new family members, we focused on the two most abundant genes, designated Taxi-III (eight clones) and Taxi-IV (three clones). The 5′ and 3′ regions of these genes were cloned by RACE with primers listed in Table 1 (represented as red and blue arrows between the middle and lower bars in Fig. 1A), and at least six independent clones were sequenced for each cloned RACE product. On the basis of the assembled cDNA sequences, the entire coding regions of Taxi-III (1,206 bp) and Taxi-IV (1,227 bp) were directly amplified by RT-PCR with primers listed in Table 1 (represented as black arrows above the middle and lower bars in Fig. 1A), and their sequences were further confirmed by sequencing the RT-PCR products (accession numbers AB114627 and AB114628 for Taxi-III and Taxi-IV, respectively). When a comparison was made with the entire coding region of Taxi-I, Taxi-III and Taxi-IV were found to contain substitutions of 100 and 96 nucleotides (i.e. 91.7% and 92.0% identity), respectively. More than half of the nucleotide substitutions resulted in amino acid sequence diversification in Taxi-III and Taxi-IV (Fig. 1C). The nucleotide sequences of the genes are more similar to each other between the two new Taxi genes (i.e. 96.8% identity between Taxi-III and Taxi-IV). With reference to the crystal structure of TAXI-I complexed with Aspergillus niger xylanase (Sansen et al. 2004), the ratios of non-synonymous to synonymous substitutions were calculated at the sites involved in the interaction (i.e. a region comprising residues Ala-291 to Ala-317 of TAXI-I, which contains a loop that covers the deep substrate-binding cleft of xylanase). Interestingly, the dN/dS ratio (ω) among TAXI family members was >1 when these sites were compared between Taxi-I and Taxi-III (ω = 1.33). The result suggests that some members of the TAXI family adaptively coevolved with xylanases. Such features are similar to those of pgip, which is organized into a complex multigene family as a result of co-evolution with the pathogen’s PG genes (Leckie et al. 1999, Stotz et al. 2000). TBASTN search of TAXI (as of August 2004) revealed the presence of several homologs from other cereals including barley (Hordeum vulgare), rye (Secale cereale) and rice (Oryza sativa). Phylogenetic analysis (Fig. 1D) showed that TAXI-I is more closely related to a rye protein (CAE46333) than TAXI-III and TAXI-IV from wheat. Taxi-III and Taxi-IV (TAXI-III/IV-type genes) may reasonably be classified as a separate subfamily within the TAXI family members as increasing number of homologs are identified from graminaceous crops in the future. Northern blot analysis of the Taxi genes The Taxi transcript levels were determined by Northern blot analysis using RNA isolated from various parts of plants at different developmental stages. The full-length Taxi-I riboprobe was used to probe blots of total RNA isolated from scutella (SC), roots (R), shoots (S) and leaves (L1–L4 for the first to fourth leaves) at days 3, 5, 10 and 24 after seed germination. As shown in Fig. 2A, an apparently single band of 1.4 kb mRNA was detected in roots at all stages investigated. The transcripts were also found at significant levels in older leaves (L1 and L2 at day 24, which were partially etiolated during growth on wet paper), and at decreased levels in scutella (SC at day 3), younger shoots (S at day 3), and elongating third and fourth leaves (L3 and L4 at day 24). However, substantially no transcripts were detected from old scutella (SC at day 5), elongating shoots (S at day 5) and younger leaves (L1 and L2 at day 10) on the Northern blot. We next examined whether Taxi genes show pathogen-inducible expression patterns. Northern blot analyses were carried out with total RNA isolated from wheat spikes infected by a model cereal pathogen, Fusarium graminearum, which causes a devastating disease called scab or Fusarium head blight (FHB) in wheat and barley (Trail et al. 2003). Expression levels of Taxi were examined in glume, lemma, palea and ovary at 24 h after Fusarium infection. As shown in Fig. 2B, a high level of Taxi mRNA was present in both infected (Fg) and uninfected (mock) ovary (O), but no transcripts were detected in uninfected glume (G), lemma (L) or palea (P). After challenge inoculation with F. graminearum, induction of Taxi expression was observed in lemma, palea and ovary, which are known to be sites of infection and pathways for the fungus. Although the transcript levels of Taxi were slightly reduced at 48 h (data not shown) compared with those at 24 h, the general expression pattern is similar to that of defense response genes that are expressed during the early stages of infection in wheat (Pritsch et al. 2000). Taxi gene expression was also examined in wheat leaves infected with the powdery mildew fungus Erysiphe graminis f. sp. tritici. As shown in Fig. 2C, transcriptional induction of the Taxi genes by E. graminis was demonstrated in the leaf sample collected 5 d after fungal inoculation. We further investigated responses of plants to wounding since mechanical injury often induces expression of defense-related genes. Young leaves (first leaves at day 5) were cut into leaf segments 3 mm in length with a razor blade and placed on wet paper for up to 48 h. As shown in Fig. 2D, Northern blot analysis demonstrated that wounding significantly increased the levels of Taxi transcripts in leaves within 24 h. Expression analysis of individual Taxi genes using RT-PCR/cleaved amplified polymorphic sequence (CAPS) Since the Taxi-I probe hybridized to multiple gene family members on a Southern blot, the RNA blot data represent the sum of differential expression of the TAXI family genes. It is not known whether these individual genes are differentially regulated under various developmental and stress conditions in the manner demonstrated by Northern blot analysis. To characterize the transcripts of individual Taxi genes in more detail, we carried out CAPS analysis using a transversion from A to C (deleting a SphI site from Taxi-I) or G to C/C to G (adding a NcoI site to Taxi-I) that occurs in the coding regions of Taxi-III and Taxi-IV (see Fig. 3A); expression of Taxi-I can be distinguished from that of the TAXI-III/IV-type genes by digestion of the semi-quantitative RT-PCR products with these restriction enzymes. When consensus primers Tcons-F and Tcons-R (labeled ‘CONS’ in Fig. 3) were used for amplification of the Taxi family members at different developmental stages, RT-PCR products sensitive to SphI digestion (i.e. Taxi-I) were found in younger shoots at day 3 (Fig. 3B; 422 bp/94 bp bands in lane 3), and to a much lesser amount, in scutella at days 3 and 5 (Fig. 3B; lanes 1 and 4) and roots at days 5 and 10 (Fig. 3B; lanes 5 and 7). However, the transcript of Taxi-I was not detected from leaves at all stages investigated (Fig. 3B; lanes 8, 9 and 11–14) or root at day 24 (Fig. 3B, lane 10). Despite the absence of the Taxi-I transcript in leaves, a small amount of RT-PCR products resistant to NcoI digestion was found in leaves (Fig. 3B; 516 bp bands in lanes 25–27, which were not split into two fragments even after 5-fold dilution of the DNA), suggesting that as yet uncloned family members of Taxi are expressed in these tissues. The transcripts of Taxi-III/IV accumulated in response to F. graminearum in infected lemma/palea, as demonstrated by the presence of the cDNA amplified with consensus primers (Fig. 3C; a 516 bp SphI-resistant band). Expression of Taxi-I was also induced upon pathogen infection, but the amount of the transcript was not significant compared with that of Taxi-III/IV (Fig. 3C; 422 bp/94 bp SphI-sensitive minor bands). The use of primers T101/102-F and T101/102-R (labeled ‘III/IV’) resulted in the amplification of PCR products that are completely resistant to SphI digestion (Fig. 3C; a 592 bp band in the left panel), suggesting that these primers are specific to the TAXI-III/IV-type genes and can be used in monitoring their expression. The presence of an extremely faint NcoI-resistant band may thus represent additional minor members of the TAXI-III/IV-type genes (Fig. 3C; a 592 bp band in the right panel). The transcript levels of Taxi-III and Taxi-IV were comparable, as assessed by MluI digestion of the cDNA (Fig. 3A legend). Induction of Taxi-III/IV, but not Taxi-I, was also observed in leaves after challenge inoculation with E. graminis (Fig. 3D). Again, a small amount of additional TAXI-III/IV-type genes was detected in the infected leaves (Fig. 3D, a 592 bp NcoI-resistant band). Wounding induced expression of both Taxi-I (Fig. 3E; 422 bp/94 bp SphI-sensitive bands and a 516 bp NcoI-resistant band) and Taxi-III/IV (Fig. 3E; 516 bp and 592 bp SphI-resistant bands and 312 bp/204 bp and 337 bp/255 bp NcoI-sensitive bands). Quantification of the Taxi transcripts To more accurately determine the induction levels of Taxi-I and Taxi-III/IV in response to the pathogens, we evaluated the amount of the transcripts by competitive RT-PCR. In contrast to Northern blot analysis, this sensitive RT-PCR method allowed identification and quantification of the Taxi transcripts in pathogen-uninfected tissues such as lemma/palea (Fig. 4A; upper and lower panels of ‘mock’) and leaves (Fig. 4B; lower panel of ‘mock’). By this assay, only the transcript of Taxi-I failed to be detected from uninfected healthy leaves (Fig. 4B; upper panel of ‘mock’). Although expression of Taxi-I was approximately 2-fold higher in Fusarium-uninfected lemma/palea compared with Taxi-III/IV (Fig. 4A; compare upper and lower panels of ‘mock’), there was a marked difference in the degree of transcriptional activation by the pathogen between Taxi-I and Taxi-III/IV. While Taxi-I expression was induced 2.5-fold, the concentration of Taxi-III/IV transcripts was induced approximately 20-fold (Fig. 4A; compare ‘mock’ and ‘Fg’). Expression of the Taxi genes was also induced by infection with the powdery mildew fungus. Again, the TAXI-III/IV-type genes responded much more significantly than Taxi-I; the expression level of Taxi-III/IV was increased approximately 25-fold in leaves by the pathogen (Fig. 4B; compare lower panels of ‘mock’ and ‘Eg’), and the amount of the transcripts was approximately 10-fold higher than that of Taxi-I (Fig. 4B; compare upper and lower panels of ‘Eg’). When leaves were injured, the amount of Taxi-I transcript was comparable to that of Taxi-III/IV (Fig. 4B; compare upper and lower panels of ‘wound’), indicating that the responses of these genes are different upon pathogen infection and wounding. Cloning and analysis of the Taxi-III promoter The genomic DNA upstream of the Taxi-III cDNA was amplified by thermal asymmetric interlaced (TAIL)-PCR using an arbitrary degenerate (AD) primer AD3 and (nested) specific primers taxi-TAIL1, taxi-TAIL2 and taxi-TAIL3 (Table 1) in a consecutive manner. The DNA sequence of the PCR product obtained by primer taxi-TAIL3 overlapped with the 5′-RACE product, suggesting that the cloned DNA fragment indeed represents a region upstream of Taxi-III. Further confirmation came from the observation that a PCR product containing this region was obtained by genomic PCR with primers 101-PROFOR (5′-end sequence of the TAIL-PCR product; see Fig. 5) and 101sp(Rev) (non-translated sequence of the cDNA downstream of the Taxi-III stop codon; see Table 1). Sequence analysis of the PCR product (AB178471) revealed that the coding region of Taxi-III is not interrupted by introns. In the 5′ upstream region of Taxi-III, there were perfect direct repeat sequences (underlined) of 14 nucleotides (–836 to –849 and –794 to –807) and 13 nucleotides (–111 to –123 and –72 to –84), and an imperfect palindrome sequence (boxed) of 22 nucleotides (–1,007 to –1,028; mismatched bases are shown in lower-case letters). The region did not contain other sequences that may form secondary structures. As expected from the expression pattern, consensus sequences of cis-acting elements implicated in pathogen- and wound-inducible gene expression were found in the Taxi-III promoter region. As shown in Fig. 5, this putative promoter region has GCC box (GCCGCC) sequences (Hart et al. 1993, Ohme-Takagi and Shinshi 1995) at –884 to –889 and –816 to –822 (AGC box sequence comprising ‘AGCCGCC’), and W box [(C/T)TGAC(C/T)] sequences (Rushton et al. 1996) at –404 to –409 and –253 to –258. These plant-specific cis-acting elements are found in the promoters of many pathogenesis-related (PR) genes and are known as the DNA-binding sites of ethylene-responsive factor (ERF) and WRKY transcription factor (TF), respectively. In addition, we found the core sequence (TGACG) of activation sequence-1 (as-1) (Qin et al. 1994), a functionally important element of a subgroup of salicylic acid (SA)-inducible defense genes, at –555 to –559, and consensus sequences of type I Myb-binding sites [(T/C)AAC(T/G)G] (Yang and Klessig 1996) at –1,027 to –1,032, –177 to –182, and –142 to –147. These cis-acting elements are often found in promoters of defense-related genes with pathogen and wound inducibility. There was no sequence that perfectly matched that of the promoter consensus, but the region at –101 to –104 (doubly underlined) is likely to be a candidate TATA box sequence considering its distance from the 5′-end of the RACE product (i.e. a putative transcription initiation site). Expression and characterization of recombinant TAXI-III Initially, we tried to obtain the recombinant TAXI-III protein by using a T7 promoter expression system in Escherichia coli. However, the expression level of Taxi-III appears to be extremely low and after induction with isopropyl-1-thio-β-d-galactopyranoside (IPTG), newly synthesized protein bands were not detected on an SDS-PAGE gel. When the sequence of Taxi-III was analyzed by the Graphical Codon Usage Analyzer on the worldwide web (http://gcua.schoedl.de/index.html), it revealed significant differences in codon use compared with the codon usage table of E. coli: among the 380 codons corresponding to mature TAXI-III protein, 36 and 46 of them are classified as ‘very few used codons’ (i.e. threshold of 10%) and ‘few used codons’ (i.e. threshold of 20%), respectively. These codons are scattered throughout the coding region of the gene. To facilitate purification of recombinant TAXI-III, we constructed a codon-optimized Taxi-III gene for expression in E. coli (see Supplementary Fig. S1). Since our initial experiments with the conventional T7 expression system resulted in the production of inclusion bodies that could not be regenerated by various methods (e.g. Refolding CA kit; TaKaRa Bio Inc., Kusatsu, Japan), we tested a fusion system to enhance the solubility of TAXI-III. The mature protein region of the synthetic gene (1,140 bp long) was fused to the C-terminus of the Nus·Tag™ protein on pET-43.1a(+). The resulting vector, pET-43.1a-ΔSPsyn.Taxi-III (see Supplementary Fig. S2), was transformed to E. coli Origami (DE3) for production of the NUS::TAXI-III protein. The fusion protein comprising 938 amino acid residues was recovered from the soluble fraction (Fig. 6A) when expressed in strain Origami (DE3) lacking both thioredoxin reductase and glutathione reductase genes. Partially purified NUS::TAXI-III protein was used for the inhibition assay. As shown in Fig. 6B, A. niger xylanase M4 and Trichoderma sp. xylanase were inhibited by the NUS::TAXI-III protein; the xylanase activities were completely lost with increasing amount of the inhibitor protein. In contrast, NUS::TAXI-III inhibited only a small fraction of spelt xylan-induced xylanase activities of F. graminearum. The result suggests that the active xylanase fraction of F. graminearum comprises enzyme mixtures that contain both TAXI-III-sensitive and -insensitive isozymes. The Nus·Tag™ protein itself did not affect the xylanase activities of A. niger, Trichoderma sp. or F. graminearum (data not shown). Discussion Unlike in dicotyledonous plants, the pectic content of cell walls is relatively low in graminaceous crops and grasses. The predominant hemicellulosic components of cereal cell wall matrices are heteroxylans consisting mainly of arabinoxylans, and for this reason, endo-β-1,4-xylanases appear to be more important than pectinases (e.g. PGs) for cereal pathogens. In this study, we have cloned two new xylanase inhibitor genes, Taxi-III and Taxi-IV, from wheat cv. Norin 61, and compared their sequences with that of Taxi-I from the same cultivar. Compared with Taxi-I, the coding region of these genes showed extensive sequence divergence. While the response of Taxi-I against pathogens was rather small, expression of Taxi-III and Taxi-IV (TAXI-III/IV-type genes) was significantly induced upon infection by these pathogens. The transcripts of Taxi-III/IV were also detected in wounded leaves and healthy roots. Recombinant TAXI-III protein completely inhibited A. niger and Trichoderma sp. xylanases: it was also active against certain isozymes of F. graminearum xylanase. Induction by wounding is reminiscent of the known classes of inducible defense-related cell wall proteins such as hydroxyproline-rich glycoproteins (Corbin et al. 1987), PR proteins (Mauch et al. 1988) and PGIPs (Bergmann et al. 1994). Expression in healthy roots is similar to the pre-existing defense mechanism described previously for the ethylene-inducible PR genes (Ohme-Takagi and Shinshi 1995); plants constitutively express them in tissues vulnerable to attack by pathogens (i.e. non-infected roots that are surrounded by soil phytopathogenic microorganisms). The induction level of Taxi-III/IV (i.e. 20- to 25-fold of induction) was also comparable to that of class I basic β-1,3-glucanase (Vögeli-Lange et al. 1994), which belongs to this PR gene family. Furthermore, the promoter of Taxi-III contained binding sites (GCC box) of ERF, which regulates the expression of ethylene-inducible PR genes. There were other consensus binding sequences of major TFs that have roles in plant defense. These TFs include WRKY, bZIP (basic domain/Leu zipper) TF family TGA and Myb1, which bind to W box, as-1, and Myb-binding sequences, respectively. Among these cis-acting elements, the W box and GCC box are sufficient to confer pathogen and wound inducibility on promoters. All of these features suggest that TAXI-III/IV-type genes function in the resistance of plants to fungal attack. In contrast, the response of Taxi-I to pathogen was not so significant. Compared with Taxi-III/IV, a different expression pattern was observed, and it may be possible that Taxi-I is implicated in a physiological process during growth and development. However, expression of cereal xylanase genes appears to be mostly restricted to the aleurone layer (Banik et al. 1997, Caspers et al. 2001) with the exception of the anther-specific xylanase genes, maize ZmXYN1 and barley HvXYN2 (Wu et al. 2002). The distribution of the Taxi-I transcripts is different from that of cereal xylanase genes. Besides, a sequence-based classification (e.g. hydrophobic cluster analyses and amino acid sequence conservation in the catalytic sites) of the endo-β-1,4-xylanases identified to date indicates that plant xylanases belong to family 10 (Simpson et al. 2003), against which TAXI shows no inhibitory activity (Gebruers et al. 2001). The physiological significance of Taxi-I remains to be elucidated in future studies. In host-pathogen relationships, pathogenicity and disease resistance are directly influenced by antagonistically interacting proteins of the hosts and pathogens (e.g. attacking wall-degrading enzymes and their inhibitors) (Rausher 2001). Such systems provide the opportunity for molecular coevolution of genes of attack and defense, as demonstrated in the interaction of PGs and PGIPs (Stotz et al. 2000). These genes exist in large multigene families and undergo rapid adaptive diversification for new recognition specificities, as do the plant disease resistance genes (R-genes) encoding pathogen detection proteins in gene-for-gene interactions (Wang et al. 1998). In this context, the suggested adaptive evolution, together with the pathogen-responsive expression pattern, agrees with the interaction of TAXI-III with the pathogen’s xylanase(s). The adaptive forces may have stimulated diversification of Taxi with nearly identical biochemical characteristics but different inhibition specificities to better counteract pathogen infection (Wang et al. 1998). Although the importance of xylanases for cereal pathogens has been suggested from several physiological studies (Cooper et al. 1988, Braun and Rodrigues 1993), conclusive evidence for the proposed roles of xylanases (i.e. their involvement in penetration, pathogen ramification, plant defense induction and symptom expression) has been difficult to obtain (Walton 1994). This is attributed to the extensive redundancy of the xylanase genes in the genome of cereal pathogens. For example, a triple mutant of the maize pathogen Cochliobolus carbonum (three family 11 xylanases deleted) (Apel-Birkhold and Walton 1996), as well as a double mutant of the rice pathogen Magnaporthe grisea (each family 10 and family 11 xylanase deleted) (Wu et al. 1997), still retained full virulence against their respective hosts. This may suggest that the pathogens are equipped with multiple different copies of xylanases to evade inhibition by specific members of TAXI. Consistent with this, TAXI-III protein inhibited only a portion of spelt xylan-induced xylanase activities of F. graminearum. Other members of TAXI may be responsible for inhibition of the pathogen’s redundant xylanases. Recently, the complete genome sequence of F. graminearum has been released into the public domain (Broad Institute Fusarium graminearum Database; http://www.broad.mit.edu/annotation/fungi/fusarium/). This database significantly facilitates comprehensive identification of the xylanase genes of the pathogen. Thus, the availability of the TAXI-III/IV-type genes provides an experimental basis for a search of the inhibitor-sensitive target xylanases by employing the use of genetic or biophysical approaches (e.g. yeast two-hybrid techniques, surface plasmon resonance). In addition, the possible involvement of TAXI-III/IV-type genes in defense (i.e. conversely, the role of the pathogen’s xylanases) can be clarified by analyzing the host-pathogen interactions of transgenic wheat lines overexpressing these genes. Since there are only a limited number of transgenes that could alleviate the problems associated with FHB (Rebmann et al. 1991, Balzi et al. 1994, Castagnaro et al. 1994, Gautier et al. 1994, Kimura et al. 1998, Alexander et al. 1999, Dickman et al. 2001, Takahashi-Ando et al. 2002), TAXI-III/IV-type genes may be added to the list of candidate genes that deserve to be tested by transgenic experiments. Materials and Methods Plant and fungal materials Triticum aestivum L. cv. Norin 61 (MAFF GenBank Plant; accession number JP 21313) was used throughout this study. Wheat cultivars Norin 12 and Florida were used for comparison of the Taxi-I sequence. For the extraction of total RNA at different developmental stages, wheat seeds were placed on wet papers in plastic dishes and germinated under constant light for up to 24 d at 26°C. Each tissue was collected at the desired time point and stored at –80°C until used for the extraction of total RNA. For infection with F. graminearum, flowering spikelets were inoculated with the fungus (IFO 5269) by injecting 5 µl of macroconidial suspension (2×105 conidia ml–1) or water (uninfected control) into the floral cavity between lemma and palea (Pritsch et al. 2001). For infection of the powdery mildew fungus E. graminis f. sp. tritici, the leaves maintaining the sporulating fungal inocula were lightly tapped before use to remove older conidia. Young wheat seedlings in a chamber were inoculated with the high-viability conidia by blowing the fresh inocula at a spore density of approximately 10 conidia mm–2 (Stadnik and Buchenauer 2000). Cloning of the Taxi genes Total RNA was prepared from each tissue using the RNeasy Plant Mini kit (Qiagen, Hilden, Germany). Oligo(dT)-primed cDNA was synthesized from total RNA using the SuperScript First-strand Synthesis system (Invitrogen, Carlsbad, CA, U.S.A.). For sequence determination, the PCR products were cloned in pGEM®-TEasy (Promega, Madison, WI, U.S.A.) and at least six independent clones were analyzed to exclude any errors introduced by PCR. To obtain full-length Taxi cDNA by PCR, LA Taq with GC Buffer I (TakaRa Bio Inc.) was necessary to ensure successful amplification. For cloning of the 5′ and 3′ portions of Taxi-I by RACE, the SMART™ RACE cDNA Amplification kit (Clontech, Palo Alto, CA, U.S.A.) was used (see Table 1 for primers; represented as red and blue arrows above the upper bar in Fig. 1A). A full-length Taxi-I cDNA was amplified by RT-PCR with primers Isp(For) and Isp(Rev) (see Table 1 for primer sequences; represented as black arrows above the upper bar in Fig. 1A), which were designed on the basis of the RACE product sequences. To isolate closely related family members of Taxi-I, degenerate primers TAXI-F1 and TAXI-R2 (Table 1) were designed by the CODEHOP computer program (Rose et al. 1998) (http//blocks.fhcrc.org/codehop.html) using the sequences of Taxi-I and its putative homolog (NM_190072). Portions of the Taxi genes (Fig. 1A; the region between primers shown in green) were amplified from cDNA prepared from roots, and the internal sequences of amplified fragments (i.e. Taxi-III and Taxi-IV) were determined. Primers for RACE of Taxi-III and Taxi-IV (see Table 1) were designed based on the local sequences of those regions with least homology to Taxi-I. Evolutionary analyses The synonymous (dS) and non-synonymous (dN) substitution distances were determined by the method of Nei and Gojobori (1986) on the basis of the sequence alignment of the TAXI-I–A. niger xylanase contact region. For this purpose, the total numbers of synonymous (S) and non-synonymous (N) sites and synonymous (Sd) and non-synonymous (Nd) substitutions were calculated using the MEGA2.1 program (Kumar et al. 2001). The neighbor-joining (NJ) trees of proteins similar to TAXI were constructed using the tree-making program (with the BLOSUM scoring matrix) at DDBJ (http://spiral.genes.nig.ac.jp/homology/welcome-e.shtml). The trees were calculated with 1,000 bootstrap replications and the results (downloaded as MIME type files) were visualized using the Macintosh program TreeView (Page 1996). The phylogenetic analysis was also done using a maximum likelihood method with the program PUZZLE (Strimmer and von Haeseler 1996) at the Pasteur Institute (http://www.pasteur.fr/english.html). Nucleic acid hybridizations For Southern blot analysis, total DNA was isolated from wheat leaves using the Nucleon PhytoPure DNA extraction kit (Amersham Biosciences, Piscataway, NJ, U.S.A.). The digoxigenin (DIG)-labeled Taxi probe (obtained by RT-PCR with primers taxi-I-For. and taxi-I-Rev. in Table 1) was generated using the PCR DIG Probe Synthesis kit (Roche Applied Science GmbH, Mannheim, Germany) and used for the Southern blot analysis. For Northern blot analysis, the Taxi-I riboprobe was prepared from the full-length cDNA [obtained by RT-PCR with primers Isp(For) and Isp(Rev) in Table 1] using the DIG RNA labeling kit (SP6/T7) (Roche Applied Science). Standard hybridization (68°C), washing (68°C) and detection protocols recommended by the manufacturer were used for Southern and Northern blot analyses. RT-PCR/CAPS analysis To amplify the three Taxi genes isolated and sequenced in this study, consensus primers Tcons-F and Tcons-R (Table 1) were used for semi-quantitative RT-PCR. Primers T101/102-F and T101/102-R (Table 1) were used for detection of the TAXI-III/IV-type gene expression: use of primers T101/102-F (anneals to Taxi-IV with 1 bp of mismatch) and T101/102-R was also effective for the amplification of Taxi-IV (data not shown). As a template for the amplification, cDNA synthesized from 3 µg of total RNA was used for each reaction. For semi-quantitative estimation of the transcript (i.e. to remain within the exponential phase of the amplification curve), optimal cycle number was determined by checking the PCR products (at 20, 25 and 30 cycles) on agarose gels by ethidium bromide. The semi-quantitative RT-PCR products (obtained with 25 amplification cycles) were purified from primers and salts with the QIAquick PCR Purification kit (Qiagen) following the manufacturer’s instructions. Aliquots of 5 µl of undiluted DNA, or 5-fold diluted DNA to exclude the possibility of incomplete digestion, were used for the restriction analysis, and the reaction mixtures were separated by 1.2% agarose gel electrophoresis. Equal volumes were loaded on each lane to compare the relative expression levels in various tissues. Competitive RT-PCR Quantitative RT-PCR assays were carried out to evaluate the transcript levels of Taxi family member genes through the use of internal competitor DNA (see Table 1 for primer sequences): completely unrelated sequences (a fragment of pUC18 vector) were amplified by PCR with primers pUC-I-F and pUC-I-R for Taxi-I and primers pUC-101-F and pUC-101-R for Taxi-III/IV, cloned in pGEM®-TEasy and sequenced. The amplified competitor DNA (548 bp) for Taxi-I and Taxi-III/IV differs only in the primer regions at either end of the target DNA (Table 1, underlined). Competitor DNA was prepared by PCR using primer pairs of comp-I-F × comp-I-R and T101/102-F × T101/102-R for Taxi-I and Taxi-III/IV, respectively. To quantify the target cDNA, various amounts of competitor DNA (1, 5, 10, 15, 25, 50, 75, 100, 250 and 500 fg) were added to the reaction mixture (50 µl) with a constant amount of cDNA (synthesized from 3 µg RNA) from various tissues, under the following conditions: 94°C 30 s, 60°C 30 s, 72°C 30 s for 25 cycles. In the PCRs, competitor DNA (548 bp) and target cDNA (592 bp for both Taxi-I and Taxi-III/IV) share identical sequences only at the primer annealing sites. The PCR products were separated on 2.5% agarose gels to compare the band intensities of amplified competitor DNA and cDNA products. The concentrations of Taxi were normalized to the amounts of competitor DNA that showed equal band intensities. TAIL-PCR To allow chromosome walking beyond the known Taxi-III cDNA sequences into the unknown 5′-flanking region, TAIL-PCR was performed using LA Taq with GC Buffer I (TaKaRa Bio Inc.). The primary TAIL-PCR mixtures (20 µl) contained 200 µM dNTP, 0.2 µM specific primer taxi-TAIL1, 5 µM AD primer AD3 and approximately 20 ng of genomic DNA. The secondary amplification reaction (20 µl) contained 125 µM dNTP, 0.2 µM nested specific primer taxi-TAIL2, 5 µM AD primer AD3 and 1 µl of a 1/20 dilution of the primary PCR product. The reaction solution for the tertiary PCR contained 25 µM dNTP, 0.6 µM nested specific primer taxi-TAIL3, 6 µM AD primer AD3 and 0.5 µl of a 1/20 dilution of the secondary PCR product. Thermal conditions for TAIL-PCR were the same as described previously (Liu and Whittier 1995). Construction of a synthetic Taxi-III expression vector for E. coli A synthetic Taxi-III gene was designed using the DNABuilder program (http://cbi.swmed.edu/computation/cbu/DNABuilder.html) on the basis of a codon frequency table of E. coli (http://www.kazusa.or.jp/codon) (see Supplementary Fig. S1). The synthetic gene consisted of two segments (segments 1 and 2), each with an NheI restriction site as shown in Supplementary Fig. S2. Each segment was assembled from sets of complementary oligonucleotides containing 25 bp overlaps so that when annealed, they will serve as primers and templates for DNA synthesis by Taq DNA polymerase (Dillon and Rosen 1990). Twenty-five PCR cycles were carried out using five and seven overlapping oligonucleotides for segments 1 and 2, respectively (see Supplementary Fig. S1 for oligonucleotide sequences). Each PCR product was used as a template for the second PCR of 30 cycles with primers Former-For and Former-Rev (segment 1) and Latter-For and Latter-Rev (segment 2). Primers Former-For and Latter-Rev contain NdeI and BamHI sites at the start codon and downstream of the stop codon, respectively. The PCR products were cloned in pGEM®-TEasy, and inserts with no PCR errors were screened and selected. With regard to the orientation of segment 2 in pGEM®-TEasy, we obtained a clone, pGEM-Seg2, with the PstI site (on the polylinker) and the BamHI site (at the end of the insert) adjacent to each other. Segment 2 was excised from the vector by digestion with NheI and PstI, and cloned into the corresponding sites downstream of segment 1 in pGEM-Seg1, yielding pGEM-syn.Taxi-III. Plant signal peptide (SP) sequence was removed by replacing the NdeI–NheI fragment with a DNA fragment (with NdeI and NheI sites at 5′- and 3′-ends, respectively) that was amplified by PCR with primers 101-SP-For and Former-Rev (i.e. the mature peptide-coding fragment designated ΔSPsyn.Taxi-III). The resulting vector, pGEM-ΔSPsyn.Taxi-III, was digested with NdeI and BamHI, and ΔSPsyn.Taxi-III was cloned in pET-19b (Novagen, Darmstadt, Germany), yielding pET-19b-ΔSPsyn.Taxi-III. However, recombinant TAXI-III was recovered as an inclusion body using this expression system. The coding region of ΔSPsyn.Taxi-III was amplified by PCR from ScaI-linearized pET-19b-ΔSPsyn.Taxi-III using primers Nus-synTAXI-5′ (containing an EcoRI site) and Nus-synTAXI-3′ (containing a PstI site) (see Table 1 for primer sequences). The PCR product was digested with EcoRI and PstI, and inserted into a Nus·Tag™ fusion expression vector, pET-43.1a(+) (Novagen). The resulting vector, pET-43.1a-ΔSPsyn.Taxi-III, was transformed to strain Origami (DE3) of E. coli (Novagen) and used for production of a Nus·Tag™-tagged recombinant protein. Expression and characterization of recombinant TAXI-III The bacterial strain harboring pET-43.1a-ΔSPsyn.Taxi-III (see Supplementary data online) was inoculated on 5 ml of liquid medium (CircleGrow; Qbiogene Inc., Montréal, Canada) containing 50 µg ml–1 ampicillin, 15 µg ml–1 kanamycin and 12.5 µg ml–1 tetracycline with vigorous shaking. After overnight incubation at 37°C, the inoculum (100 µl to 1 ml) was transferred to 100 ml of CircleGrow medium containing 50 µg ml–1 ampicillin and incubated until the OD600 reached 0.4. Then IPTG was added to a final concentration of 1 mM and the culture incubated at ambient temperature. Induction of the recombinant proteins was examined by SDS-PAGE on a 7.5% polyacrylamide gel. After overnight incubation, the bacterial pellets were harvested, suspended in 5 ml of buffer A (20 mM Tris-HCl, pH 7.5), disrupted with an ultrasonic disruptor (Tomy Tech USA Inc., Fremont, CA, U.S.A.), and centrifuged at 12,000×g for 15 min. The supernatant was passed through a 0.22 µm filter (Minisart-Plus; Sartorius AG, Goettingen, Germany) and the recombinant protein was precipitated with 65% saturated ammonium sulfate. During dialysis against buffer A, portions of the NUS::TAXI-III protein were lost as a precipitate but approximately 20% of the inhibition activity was recovered (as assessed by an assay with Trichoderma sp. xylanases). The active fraction was applied to a HiTrap Q HP (5 ml) column (Amersham Biosciences) equilibrated with buffer A, and eluted with a linear gradient of 1 M NaCl in buffer A (in a total volume of 96 ml at a flow rate of 6 ml min–1). The recombinant NUS::TAXI-III (pI = 4.8) was recovered at approximately 0.4 M NaCl. The protein concentration was measured by using a BCA protein assay kit (Pierce Biotechnology Inc., Rockford, IL, U.S.A.) and used for the inhibition assay of xylanases. The inhibition activity of TAXI-III was determined by colorimetric assay using AZO-WAX solution (Megazyme International Ltd., Wicklow, Ireland) as described in the manufacturer’s instructions. Trichoderma sp. xylanases and A. niger xylanase M4 (1,851 U ml–1; used as a standard in this study) were obtained from Megazyme. The partially purified recombinant inhibitor protein (6–120 µg protein in 40 µl buffer A) was mixed with 60 µg of bovine serum albumin (BSA) and 0.04 units of xylanases in 160 µl of 0.1 M sodium acetate buffer, pH 4.6, and the mixture (200 µl) was pre-incubated at room temperature for 20 min. Inclusion of BSA in the reaction mixture was necessary to avoid inactivation of the xylanases by dilution. The reaction was initiated by adding AZO-WAX substrate solution (200 µl) to the enzyme-inhibitor mixture solution. After 30 min of incubation at 40°C, the reaction was terminated by the addition of ethanol (1 ml). After removal of the precipitated non-depolymerized substrate by centrifugation (2,430×g, 5 min), the absorbance of the supernatant was measured at 590 nm against a reaction blank. The residual xylanase activity was determined from a calibration curve, which showed good proportionality in the range 0.005–0.1 U. Partial purification of xylanases from F. graminearum Xylanases were partially purified from xylan-induced culture of F. graminearum IFO 5269 as described previously (Holden and Walton 1992). Briefly, the fungus was inoculated on 200 ml of liquid synthetic low nutrient medium (see http://www.jcm.riken.go.jp/cgi-bin/jcm/jcm_grmd?GRMD=332 for medium composition) supplemented with 0.8% xylan oat spelts (Fluka, Basel, Switzerland) in a 500 ml baffled flask and cultured for 7 d with shaking (101 rpm) at ambient temperature. An equal amount of buffer B (25 mM sodium acetate buffer, pH 5.0) was added to the supernatant, and the diluted sample was filtered through 0.2 µm filters (Durapore; Millipore, Billerica, U.S.A). The sample was passed through a HiTrapQ HP (5 ml) column and then applied to a HiTrap S HP (5 ml) column (Amersham Biosciences) pre-equilibrated with buffer B. After washing the column with two bed volumes of buffer B, xylanases were eluted with a linear gradient of 0–0.8 M KCl in 100 ml of buffer B. Xylanase activities were detected at approximately 0.1–0.3 M KCl. Aliquots of the active fractions (0.04 U in <60 µl) were used for the inhibition assay. Supplementary Material Supplementary material mentioned in the article is available to online subscribers at the journal website www.pcp.oupjournals.org. Acknowledgments This research was supported in part by a grant for ‘Integrated Research on the Safety and Physiological Function of Food’ to M.K. from the Ministry of Agriculture, Fishery, and Forestry of Japan (MAFF). We greatly thank M. Inaba and T. Tokai for their excellent technical supports. 4 Corresponding author: E-mail, [email protected]; Fax, +81-48-462-4394. View largeDownload slide Fig. 1 Structural characterization of the TAXI family genes. (A) Overview of the cloning strategy based on the published TAXI peptide sequences (shown as open boxes on a thick line that represents the entire TAXI-I polypeptide) (Gebruers et al. 2001) and EST sequences (shown as thin lines). Filled and open vertical arrows denote the SP cleavage site and internal cleavage site (generating the 10 kDa and 30 kDa polypeptide of TAXI-I), respectively. Bars represent cloned cDNA of Taxi-I, Taxi-III and Taxi-IV: their coding regions are filled in gray. Horizontal arrows of different colors indicate primers used for the cloning of these Taxi genes; red, primers/nested primers for 5′-RACE of Taxi-I and Taxi-III/IV; blue, primers/nested primers for 3′-RACE of Taxi-I and Taxi-III/IV; green, degenerate primers for RT-PCR of Taxi-III/IV; black, specific primers for RT-PCR of full-length Taxi-I, Taxi-III and Taxi-IV. (B) Southern blot analysis of Taxi. Five micrograms of genomic DNA was digested with BamHI (lane 1), HindIII (lane 2) and XbaI (lane 3), and probed with the DIG-labeled Taxi-I probe. M, λ HindIII marker. (C) Sequence alignment of the TAXI-I, TAXI-III and TAXI-IV polypeptides. Amino acid residues identical to those of TAXI-I are indicated by dots. Dots highlighted in blue denote amino acids unaltered by synonymous substitutions. Residues highlighted in red (with increasing color depth relative to the number of nucleotide substitutions in each codon) indicate non-synonymous substitutions. Filled and open vertical arrows denote the SP cleavage site and internal cleavage site, respectively. (D) Evolutionary relatedness of TAXI homologs. Unrooted phylogenetic trees were constructed using the NJ method and downloaded from the DDBJ Web server. The scale bars indicate the branch length corresponding to the mean number of differences (0.1) per residue along each branch. Bootstrap values supporting the branches connecting the subgroups are indicated. A tree with the same topology was constructed using the maximum likelihood method accomplished with PUZZLE. An enhanced view of closely related proteins from wheat, rye and barley is shown. View largeDownload slide Fig. 1 Structural characterization of the TAXI family genes. (A) Overview of the cloning strategy based on the published TAXI peptide sequences (shown as open boxes on a thick line that represents the entire TAXI-I polypeptide) (Gebruers et al. 2001) and EST sequences (shown as thin lines). Filled and open vertical arrows denote the SP cleavage site and internal cleavage site (generating the 10 kDa and 30 kDa polypeptide of TAXI-I), respectively. Bars represent cloned cDNA of Taxi-I, Taxi-III and Taxi-IV: their coding regions are filled in gray. Horizontal arrows of different colors indicate primers used for the cloning of these Taxi genes; red, primers/nested primers for 5′-RACE of Taxi-I and Taxi-III/IV; blue, primers/nested primers for 3′-RACE of Taxi-I and Taxi-III/IV; green, degenerate primers for RT-PCR of Taxi-III/IV; black, specific primers for RT-PCR of full-length Taxi-I, Taxi-III and Taxi-IV. (B) Southern blot analysis of Taxi. Five micrograms of genomic DNA was digested with BamHI (lane 1), HindIII (lane 2) and XbaI (lane 3), and probed with the DIG-labeled Taxi-I probe. M, λ HindIII marker. (C) Sequence alignment of the TAXI-I, TAXI-III and TAXI-IV polypeptides. Amino acid residues identical to those of TAXI-I are indicated by dots. Dots highlighted in blue denote amino acids unaltered by synonymous substitutions. Residues highlighted in red (with increasing color depth relative to the number of nucleotide substitutions in each codon) indicate non-synonymous substitutions. Filled and open vertical arrows denote the SP cleavage site and internal cleavage site, respectively. (D) Evolutionary relatedness of TAXI homologs. Unrooted phylogenetic trees were constructed using the NJ method and downloaded from the DDBJ Web server. The scale bars indicate the branch length corresponding to the mean number of differences (0.1) per residue along each branch. Bootstrap values supporting the branches connecting the subgroups are indicated. A tree with the same topology was constructed using the maximum likelihood method accomplished with PUZZLE. An enhanced view of closely related proteins from wheat, rye and barley is shown. View largeDownload slide Fig. 2 Northern blot analysis of Taxi transcripts under various developmental and stress conditions. The DIG-labeled full-length Taxi-I riboprobe was used for hybridization. The 1% formaldehyde gel stained with ethidium bromide is shown below each blot to demonstrate equal loading of RNA samples. (A) Expression of Taxi in developing seedlings at days 3, 5, 10 and 24 after seed germination. Total RNA (10 µg) from scutella (SC), roots (R), shoots (S) and leaves (1st leaf, L1; 2nd leaf, L2; 3rd leaf, L3; 4th leaf, L4) were loaded in each lane. (B) Expression of Taxi in wheat spikes upon challenge inoculation with F. graminearum. Flowering spikelets were inoculated with water (mock) or with macroconidial suspensions of F. graminearum (Fg). Each lane contains total RNA (20 µg) isolated from glume (G), lemma (L), palea (P) and ovary (O) at 24 h after infection. (C) Expression of Taxi in wheat leaves upon challenge inoculation with E. graminis. Each lane contains total RNA (20 µg) isolated from water-treated (mock) and infected (Eg) leaves 5 d after inoculation. (D) Induction of Taxi expression by wounding. Twenty micrograms of total RNA from control (C) and wounded (W) leaves of 5-day-old seedlings were collected 24 h (24) and 48 h (48) after the treatment and used for the analysis. View largeDownload slide Fig. 2 Northern blot analysis of Taxi transcripts under various developmental and stress conditions. The DIG-labeled full-length Taxi-I riboprobe was used for hybridization. The 1% formaldehyde gel stained with ethidium bromide is shown below each blot to demonstrate equal loading of RNA samples. (A) Expression of Taxi in developing seedlings at days 3, 5, 10 and 24 after seed germination. Total RNA (10 µg) from scutella (SC), roots (R), shoots (S) and leaves (1st leaf, L1; 2nd leaf, L2; 3rd leaf, L3; 4th leaf, L4) were loaded in each lane. (B) Expression of Taxi in wheat spikes upon challenge inoculation with F. graminearum. Flowering spikelets were inoculated with water (mock) or with macroconidial suspensions of F. graminearum (Fg). Each lane contains total RNA (20 µg) isolated from glume (G), lemma (L), palea (P) and ovary (O) at 24 h after infection. (C) Expression of Taxi in wheat leaves upon challenge inoculation with E. graminis. Each lane contains total RNA (20 µg) isolated from water-treated (mock) and infected (Eg) leaves 5 d after inoculation. (D) Induction of Taxi expression by wounding. Twenty micrograms of total RNA from control (C) and wounded (W) leaves of 5-day-old seedlings were collected 24 h (24) and 48 h (48) after the treatment and used for the analysis. View largeDownload slide Fig. 3 CAPS analyses of the Taxi cDNA obtained by semi-quantitative RT-PCR. (A) Comparison of the partial nucleotide sequences of Taxi-I, Taxi-III and Taxi-IV. Dots and boxes indicate identical nucleotides and restriction enzyme recognition sequences (NcoI and SphI), respectively. Arrows denote consensus primers (Tcons-F and Tcons-R) and primers T101/102-F and T101/102-R used for the RT-PCR/CAPS analyses. The Taxi-IV cDNA lacks an MluI site (bp 821–826 in the figure; site not shown), which can be used to distinguish the transcript of Taxi-III and Taxi-IV. (B) CAPS analysis of the cDNA at different developmental stages. Amplification (before reaching the saturation; i.e. 25 cycles for all experiments through Fig. 3B to Fig. 3E) was carried out with consensus primers (labeled ‘CONS’ hereafter): the RT-PCR products were digested with SphI (lanes 1–14) or NcoI (lanes 15–28). Lanes 1–14 and 15–28 correspond to the Northern blot data of Fig. 2A (i.e. increasing the lane number from left to right). As a size marker, 1 kb Plus DNA Ladder marker (M) from Invitrogen was loaded throughout the experiments (B–E). (C) CAPS analysis of spikelet cDNA upon challenge inoculation with F. graminearum. Amplification was carried out with consensus primers or primers T101/102-F and T101/102-R (labeled ‘III/IV’ hereafter): the RT-PCR products were digested with SphI (left panel) or NcoI (right panel). Template cDNA was prepared from uninfected (mock) or infected (Fg) lemma/palea. (D) CAPS analysis of leaf cDNA upon challenge inoculation with E. graminis. Amplification was carried out with consensus primers or primers T101/102-F and T101/102-R: the RT-PCR products were digested with SphI (left panel) or NcoI (right panel). Template cDNA was prepared from uninfected (mock) or infected (Eg) leaves. (E) CAPS analysis of leaf cDNA after wounding (24 h). Amplification was carried out with consensus primers or primers T101/102-F and T101/102-R: the RT-PCR products were digested with SphI (left panel) or NcoI (right panel). Template cDNA was prepared from control (C) or wounded (W) leaves. View largeDownload slide Fig. 3 CAPS analyses of the Taxi cDNA obtained by semi-quantitative RT-PCR. (A) Comparison of the partial nucleotide sequences of Taxi-I, Taxi-III and Taxi-IV. Dots and boxes indicate identical nucleotides and restriction enzyme recognition sequences (NcoI and SphI), respectively. Arrows denote consensus primers (Tcons-F and Tcons-R) and primers T101/102-F and T101/102-R used for the RT-PCR/CAPS analyses. The Taxi-IV cDNA lacks an MluI site (bp 821–826 in the figure; site not shown), which can be used to distinguish the transcript of Taxi-III and Taxi-IV. (B) CAPS analysis of the cDNA at different developmental stages. Amplification (before reaching the saturation; i.e. 25 cycles for all experiments through Fig. 3B to Fig. 3E) was carried out with consensus primers (labeled ‘CONS’ hereafter): the RT-PCR products were digested with SphI (lanes 1–14) or NcoI (lanes 15–28). Lanes 1–14 and 15–28 correspond to the Northern blot data of Fig. 2A (i.e. increasing the lane number from left to right). As a size marker, 1 kb Plus DNA Ladder marker (M) from Invitrogen was loaded throughout the experiments (B–E). (C) CAPS analysis of spikelet cDNA upon challenge inoculation with F. graminearum. Amplification was carried out with consensus primers or primers T101/102-F and T101/102-R (labeled ‘III/IV’ hereafter): the RT-PCR products were digested with SphI (left panel) or NcoI (right panel). Template cDNA was prepared from uninfected (mock) or infected (Fg) lemma/palea. (D) CAPS analysis of leaf cDNA upon challenge inoculation with E. graminis. Amplification was carried out with consensus primers or primers T101/102-F and T101/102-R: the RT-PCR products were digested with SphI (left panel) or NcoI (right panel). Template cDNA was prepared from uninfected (mock) or infected (Eg) leaves. (E) CAPS analysis of leaf cDNA after wounding (24 h). Amplification was carried out with consensus primers or primers T101/102-F and T101/102-R: the RT-PCR products were digested with SphI (left panel) or NcoI (right panel). Template cDNA was prepared from control (C) or wounded (W) leaves. View largeDownload slide Fig. 4 Competitive RT-PCR of Taxi-I and Taxi-III/IV under stress conditions. The amount of competitor DNA (fg) is indicated above each lane. The estimated amount of the Taxi transcript is shown in bold. Open and filled arrowheads represent Taxi-I (upper panels) and Taxi-III/IV (lower panels), respectively. (A) Quantification of cDNA from uninfected (mock) and Fusarium-infected (Fg) lemma/palea. (B) Quantification of cDNA from control (mock), wounded (wound) and powdery mildew-infected (Eg) leaves. View largeDownload slide Fig. 4 Competitive RT-PCR of Taxi-I and Taxi-III/IV under stress conditions. The amount of competitor DNA (fg) is indicated above each lane. The estimated amount of the Taxi transcript is shown in bold. Open and filled arrowheads represent Taxi-I (upper panels) and Taxi-III/IV (lower panels), respectively. (A) Quantification of cDNA from uninfected (mock) and Fusarium-infected (Fg) lemma/palea. (B) Quantification of cDNA from control (mock), wounded (wound) and powdery mildew-infected (Eg) leaves. View largeDownload slide Fig. 5 The nucleotide sequence of the region upstream of Taxi-III. The sequence was determined by direct sequencing of the PCR product amplified with primers 101-PROFOR (dotted arrow) and 101sp(Rev) (downstream of the Taxi-III stop codon and not included in this sequence). The sequence of primer 101-PROFOR is based on the sequence of six independent clones of the TAIL-PCR product in pGEM®-TEasy. Arrows indicate directions of cis-acting elements, in which highly conserved bases are shaded. See text for details of each cis-acting element. Direct repeats (two pairs of 14 and 13 nucleotides; see text) and imperfect palindrome sequences are underlined and boxed, respectively; lower-case letters in a palindrome indicate mismatched bases. A putative TATA box is doubly underlined. The inverted ATG (white letters in a black background) indicates the translation initiation codon of Taxi-III. Italicized letters represent nucleotide sequences included in the Taxi-III cDNA. View largeDownload slide Fig. 5 The nucleotide sequence of the region upstream of Taxi-III. The sequence was determined by direct sequencing of the PCR product amplified with primers 101-PROFOR (dotted arrow) and 101sp(Rev) (downstream of the Taxi-III stop codon and not included in this sequence). The sequence of primer 101-PROFOR is based on the sequence of six independent clones of the TAIL-PCR product in pGEM®-TEasy. Arrows indicate directions of cis-acting elements, in which highly conserved bases are shaded. See text for details of each cis-acting element. Direct repeats (two pairs of 14 and 13 nucleotides; see text) and imperfect palindrome sequences are underlined and boxed, respectively; lower-case letters in a palindrome indicate mismatched bases. A putative TATA box is doubly underlined. The inverted ATG (white letters in a black background) indicates the translation initiation codon of Taxi-III. Italicized letters represent nucleotide sequences included in the Taxi-III cDNA. View largeDownload slide Fig. 6 Inhibition of xylanases by recombinant TAXI-III. (A) Overexpression of NUS::TAXI-III in E. coli Origami (DE3). After induction by IPTG (1 mM), the bacterial cells were collected and disrupted by a sonicator. The soluble supernatant fraction of bacterial cells was analyzed by 7.5% SDS-PAGE. The gel was stained with Coomassie Brilliant Blue R-250. Compared with the control in which soluble fraction from the Origami (DE3) strain was loaded (lane 1), the stained gel revealed newly synthesized protein bands at 64.9 kDa for the Nus•Tag™ protein (lane 2, arrowhead) from pET-43.1a(+) and 100.6 kDa for the Nus•Tag™ tagged TAXI-III fusion protein (lane 3, arrowhead) from pET-43.1a-ΔSPsyn.Taxi-III. (B) Inhibition of xylanases by NUS::TAXI-III. The inhibition activities (average of triplicate experiments with standard deviations) were measured against 0.04 units of xylanases from A. niger (square), Trichoderma sp. (circle) and F. graminearum (triangle). View largeDownload slide Fig. 6 Inhibition of xylanases by recombinant TAXI-III. (A) Overexpression of NUS::TAXI-III in E. coli Origami (DE3). After induction by IPTG (1 mM), the bacterial cells were collected and disrupted by a sonicator. The soluble supernatant fraction of bacterial cells was analyzed by 7.5% SDS-PAGE. The gel was stained with Coomassie Brilliant Blue R-250. Compared with the control in which soluble fraction from the Origami (DE3) strain was loaded (lane 1), the stained gel revealed newly synthesized protein bands at 64.9 kDa for the Nus•Tag™ protein (lane 2, arrowhead) from pET-43.1a(+) and 100.6 kDa for the Nus•Tag™ tagged TAXI-III fusion protein (lane 3, arrowhead) from pET-43.1a-ΔSPsyn.Taxi-III. (B) Inhibition of xylanases by NUS::TAXI-III. The inhibition activities (average of triplicate experiments with standard deviations) were measured against 0.04 units of xylanases from A. niger (square), Trichoderma sp. (circle) and F. graminearum (triangle). Table 1 Primers used for cloning and CAPS analysis of the TAXI family genes Primer name  Primer sequence  Comments  5′GSP2  5′-CTACAATGGACCTGGCCGAGATGTAGTGCGCGG-3′  Primer for 5′-RACE of Taxi-I  5′nGSP  5′-CCCGACGTTCACCTTGCTCACCGGCTTGCT-3′  Nested primer for 5′-RACE of Taxi-I  3′GSP  5′-GGCCGCAATTCACGCAGTCGATGCCTTACACGC-3′  Primer for 3′-RACE of Taxi-I  3′nGSP  5′-CCCCGCGCACTACATCTCGGCCAGGTCCATTGT-3′  Nested primer for 3′-RACE of Taxi-I  Isp(For)  5′-CCCACAAACAATTCCACGCTCCAT-3′  Forward primer for RT-PCR of full-length Taxi-I  Isp(Rev)  5′-GGACGAATCCACCTGTCGTTAAAC-3′  Reverse primer for RT-PCR of full-length Taxi-I  taxi-I-For.  5′-CTTCCGGTGCTCGCTCCGGTCACC-3′  Forward primer for preparation of Taxi-I probe  taxi-I-Rev.  5′-CGCCGCAACCCGTAAAGTGCGGCAGCCTGCTAA-3′  Reverse primer for preparation of Taxi-I probe  TAXI-F1  5′-GACCCCGGGGTGTGGWSNACNTGYG-3′  Degenerate forward primer for cloning of Taxi-III/IV  TAXI-R2  5′-CCACCATGGAGTTTTTGCCNGTCATNGT-3′  Degenerate reverse primer for cloning of Taxi-III/IV  T101/102-R  5′-CTGCCAGCCCAACGTTCGGCACCC-3′  Primer for 5′-RACE and RT-PCR a of Taxi-III/IV  5′n-101/102  5′-GCGGGTGTTCTCCACTTTGATGGACTTGAG-3′  Nested primer for 5′-RACE of Taxi-III/IV  T101/102-F  5′-ACCGACGGGAATAAACCGGTTAGC-3′  Primer for 3′-RACE and RT-PCR a of Taxi-III/IV  3′n-101/102  5′-CTCAAGTCCATCAAAGTGGAGAACACCCGC-3′  Nested primer for 3′-RACE of Taxi-III/IV  101sp(For)  5′-CCATCTCCGTCCACTGGTGCGAGC-3′  Forward primer for RT-PCR of full-length Taxi-III  101sp(Rev)  5′-GCCGTAGTCGACGAAGTACATTGT-3′  Reverse primer for RT-PCR of full-length Taxi-III  102sp(For)  5′-CCACGGTCCATTGGTGCCAGCAAA-3′  Forward primer for RT-PCR of full-length Taxi-IV  102sp(Rev)  5′-CGTGCAGTAGTCGACAAATACAGT-3′  Reverse primer for RT-PCR of full-length Taxi-IV  Tcons-F  5′-GTGCGCGCCGAGCAAGCTCCTGGC-3′  Consensus primer for RT-PCR/CAPS of Taxi  Tcons-R  5′-TACCCGCCGAGGTTGTTGCCCAGC-3′  Consensus primer for RT-PCR/CAPS of Taxi  pUC-I-F  5′-ACCGACGGGAGCAAGCCGGTGAGCACGCGCGGGG-3′ b  Primer for preparation of competitor DNA of Taxi-I  pUC-I-R  5′-GCCCCAGCTGGACGTTGGGCACCGAGCGGTCGGG-3′ b  Primer for preparation of competitor DNA of Taxi-I  pUC-101-F  5′-ACCGACGGGAATAAACCGGTTAGCACGCGCGGGG-3′ b  Primer for preparation of competitor DNA of Taxi-III/IV  pUC-101-R  5′-CTGCCAGCCCAACGTTCGGCACCCAGCGGTCGGG-3′ b  Primer for preparation of competitor DNA of Taxi-III/IV  comp-I-F  5′-ACCGACGGGAGCAAGCCGGTGAGC-3′  Forward primer for competitive RT-PCR of Taxi-I  comp-I-R  5′-GCCCCAGCTGGACGTTGGGCACCG-3′  Reverse primer for competitive RT-PCR of Taxi-I  taxi-TAIL1  5′-GAAGGGGATGGTGTAGAGGGAGGT-3′  Primer for TAIL-PCR of Taxi-III  taxi-TAIL2  5′-CTTTGGACGACGCCAGCGCAACGA-3′  1st nested primer for TAIL-PCR of Taxi-III  taxi-TAIL3  5′-GATGGTCAGGACCGTGGAATTGTT-3′  2nd nested primer for TAIL-PCR of Taxi-III  AD3  5′-WGTGNAGWANCANAGA-3′  Arbitrary degenerate primer for TAIL-PCR of Taxi-III  101-PROFOR  5′-AGTGAAGTAGCAGAGAGGTGCATA-3′  Primer for amplification of promoter region of Taxi-III  Former-For  5′-CATATGGCACGCGTTTTGCTGCTG-3′  Primer for amplification of segment 1 of synthetic Taxi-III  Former-Rev  5′-GTGCTAGCTACTTGCGCCGGTAAG-3′  Primer for amplification of segment 1 of synthetic Taxi-III  Latter-For  5′-CAAGTTGCTAGCACTCAAAAGGTG-3′  Primer for amplification of segment 2 of synthetic Taxi-III  Latter-Rev  5′-GGATCCTTAGGAACCACAGCCGGT-3′  Primer for amplification of segment 2 of synthetic Taxi-III  101-SP-For  5′-CATATGTTACCTGTACTGGCTCCGGTGACG-3′  Primer for amplification of synthetic Taxi-III without signal peptide  Nus-synTAXI-5′  5′-GCTCTTGAATTCTCAAAAGGATTACCTGTACTG-3′  Forward primer for amplification of synthetic Taxi-III  Nus-synTAXI-3′  5′-TTGTCTGCAGCCGGATCCTTAGGAACC-3′  Reverse primer for amplification of synthetic Taxi-III  Primer name  Primer sequence  Comments  5′GSP2  5′-CTACAATGGACCTGGCCGAGATGTAGTGCGCGG-3′  Primer for 5′-RACE of Taxi-I  5′nGSP  5′-CCCGACGTTCACCTTGCTCACCGGCTTGCT-3′  Nested primer for 5′-RACE of Taxi-I  3′GSP  5′-GGCCGCAATTCACGCAGTCGATGCCTTACACGC-3′  Primer for 3′-RACE of Taxi-I  3′nGSP  5′-CCCCGCGCACTACATCTCGGCCAGGTCCATTGT-3′  Nested primer for 3′-RACE of Taxi-I  Isp(For)  5′-CCCACAAACAATTCCACGCTCCAT-3′  Forward primer for RT-PCR of full-length Taxi-I  Isp(Rev)  5′-GGACGAATCCACCTGTCGTTAAAC-3′  Reverse primer for RT-PCR of full-length Taxi-I  taxi-I-For.  5′-CTTCCGGTGCTCGCTCCGGTCACC-3′  Forward primer for preparation of Taxi-I probe  taxi-I-Rev.  5′-CGCCGCAACCCGTAAAGTGCGGCAGCCTGCTAA-3′  Reverse primer for preparation of Taxi-I probe  TAXI-F1  5′-GACCCCGGGGTGTGGWSNACNTGYG-3′  Degenerate forward primer for cloning of Taxi-III/IV  TAXI-R2  5′-CCACCATGGAGTTTTTGCCNGTCATNGT-3′  Degenerate reverse primer for cloning of Taxi-III/IV  T101/102-R  5′-CTGCCAGCCCAACGTTCGGCACCC-3′  Primer for 5′-RACE and RT-PCR a of Taxi-III/IV  5′n-101/102  5′-GCGGGTGTTCTCCACTTTGATGGACTTGAG-3′  Nested primer for 5′-RACE of Taxi-III/IV  T101/102-F  5′-ACCGACGGGAATAAACCGGTTAGC-3′  Primer for 3′-RACE and RT-PCR a of Taxi-III/IV  3′n-101/102  5′-CTCAAGTCCATCAAAGTGGAGAACACCCGC-3′  Nested primer for 3′-RACE of Taxi-III/IV  101sp(For)  5′-CCATCTCCGTCCACTGGTGCGAGC-3′  Forward primer for RT-PCR of full-length Taxi-III  101sp(Rev)  5′-GCCGTAGTCGACGAAGTACATTGT-3′  Reverse primer for RT-PCR of full-length Taxi-III  102sp(For)  5′-CCACGGTCCATTGGTGCCAGCAAA-3′  Forward primer for RT-PCR of full-length Taxi-IV  102sp(Rev)  5′-CGTGCAGTAGTCGACAAATACAGT-3′  Reverse primer for RT-PCR of full-length Taxi-IV  Tcons-F  5′-GTGCGCGCCGAGCAAGCTCCTGGC-3′  Consensus primer for RT-PCR/CAPS of Taxi  Tcons-R  5′-TACCCGCCGAGGTTGTTGCCCAGC-3′  Consensus primer for RT-PCR/CAPS of Taxi  pUC-I-F  5′-ACCGACGGGAGCAAGCCGGTGAGCACGCGCGGGG-3′ b  Primer for preparation of competitor DNA of Taxi-I  pUC-I-R  5′-GCCCCAGCTGGACGTTGGGCACCGAGCGGTCGGG-3′ b  Primer for preparation of competitor DNA of Taxi-I  pUC-101-F  5′-ACCGACGGGAATAAACCGGTTAGCACGCGCGGGG-3′ b  Primer for preparation of competitor DNA of Taxi-III/IV  pUC-101-R  5′-CTGCCAGCCCAACGTTCGGCACCCAGCGGTCGGG-3′ b  Primer for preparation of competitor DNA of Taxi-III/IV  comp-I-F  5′-ACCGACGGGAGCAAGCCGGTGAGC-3′  Forward primer for competitive RT-PCR of Taxi-I  comp-I-R  5′-GCCCCAGCTGGACGTTGGGCACCG-3′  Reverse primer for competitive RT-PCR of Taxi-I  taxi-TAIL1  5′-GAAGGGGATGGTGTAGAGGGAGGT-3′  Primer for TAIL-PCR of Taxi-III  taxi-TAIL2  5′-CTTTGGACGACGCCAGCGCAACGA-3′  1st nested primer for TAIL-PCR of Taxi-III  taxi-TAIL3  5′-GATGGTCAGGACCGTGGAATTGTT-3′  2nd nested primer for TAIL-PCR of Taxi-III  AD3  5′-WGTGNAGWANCANAGA-3′  Arbitrary degenerate primer for TAIL-PCR of Taxi-III  101-PROFOR  5′-AGTGAAGTAGCAGAGAGGTGCATA-3′  Primer for amplification of promoter region of Taxi-III  Former-For  5′-CATATGGCACGCGTTTTGCTGCTG-3′  Primer for amplification of segment 1 of synthetic Taxi-III  Former-Rev  5′-GTGCTAGCTACTTGCGCCGGTAAG-3′  Primer for amplification of segment 1 of synthetic Taxi-III  Latter-For  5′-CAAGTTGCTAGCACTCAAAAGGTG-3′  Primer for amplification of segment 2 of synthetic Taxi-III  Latter-Rev  5′-GGATCCTTAGGAACCACAGCCGGT-3′  Primer for amplification of segment 2 of synthetic Taxi-III  101-SP-For  5′-CATATGTTACCTGTACTGGCTCCGGTGACG-3′  Primer for amplification of synthetic Taxi-III without signal peptide  Nus-synTAXI-5′  5′-GCTCTTGAATTCTCAAAAGGATTACCTGTACTG-3′  Forward primer for amplification of synthetic Taxi-III  Nus-synTAXI-3′  5′-TTGTCTGCAGCCGGATCCTTAGGAACC-3′  Reverse primer for amplification of synthetic Taxi-III  a RT-PCR includes RT-PCR/CAPS and competitive RT-PCR. b Underlines represent sequences derived from Taxi. View Large Abbreviations as-1 activation sequence-1 CAPS cleaved amplified polymorphic sequence DIG digoxigenin EST expressed sequence tag ERF ethylene-responsive factor FHB Fusarium head blight IPTG isopropyl-1-thio-β-d-galactopyranoside PG polygalacturonase PGIP PG-inhibiting protein PR pathogenesis-related RACE rapid amplification of cDNA ends RT-PCR reverse transcription-PCR SA salicylic acid SP signal peptide TAIL-PCR thermal asymmetric interlaced-PCR TF transcription factor. The nucleotide sequences reported in this paper have been submitted to DDBJ under accession numbers AB114627, AB114628 and AB178471. References Alexander, N.J., McCormick, S.P. and Hohn, T.M. ( 1999) TRI12, a trichothecene efflux pump from Fusarium sporotrichioides: gene isolation and expression in yeast. Mol. Gen. Genet.  261: 977–984. Google Scholar Apel-Birkhold, P.C. and Walton, J.D. ( 1996) Cloning, disruption, and expression of two endo-β 1,4-xylanase genes, XYL2 and XYL3, from Cochliobolus carbonum. Appl. Environ. Microbiol.  62: 4129–4135. Google Scholar Balzi, E., Wang, M., Leterme, S., Van Dyck, L. and Goffeau, A. ( 1994) PDR5, a novel yeast multidrug resistance conferring transporter controlled by the transcription regulator PDR1. J. Biol. Chem.  269: 2206–2214. Google Scholar Banik, M., Li, C.D., Langridge, P. and Fincher, G.B. ( 1997) Structure, hormonal regulation, and chromosomal location of genes encoding barley (1→4)-β-xylan endohydrolases. Mol. Gen. Genet.  253: 599–608. Google Scholar Bergmann, C.W., Ito, Y., Singer, D., Albersheim, P., Darvill, A.G., Benhamou, N., Nuss, L., Salvi, G., Cervone, F. and De Lorenzo, G. ( 1994) Polygalacturonase-inhibiting protein accumulates in Phaseolus vulgaris L. in response to wounding, elicitors and fungal infection. Plant J.  5: 625–634. Google Scholar Braun, E.J. and Rodrigues, C.A. ( 1993) Purification and properties of an endoxylanase from a corn stalk rot strain of Erwinia chrysanthemi. Phytopathology  83: 332–338. Google Scholar Bugbee, W.M. ( 1993) A pectin lyase inhibitor protein from cell walls of sugar beet. Phytopathology  83: 63–68. Google Scholar Caspers, M.P., Lok, F., Sinjorgo, K.M., van Zeijl, M.J., Nielsen, K.A. and Cameron-Mills, V. ( 2001) Synthesis, processing and export of cytoplasmic endo-β-1,4-xylanase from barley aleurone during germination. Plant J.  26: 191–204. Google Scholar Castagnaro, A., Marana, C., Carbonero, P. and Garcia-Olmedo, F. ( 1994) cDNA cloning and nucleotide sequences of α1 and α2 thionins from hexaploid wheat endosperm. Plant Physiol.  106: 1221–1222. Google Scholar Cooper, R.M., Longman, D., Campbell, A., Henry, M. and Lees, P.E. ( 1988) Enzymic adaptation of cereal pathogens to the monocotyledonous primary wall. Physiol. Mol. Plant Pathol.  32: 33–47. Google Scholar Corbin, D.R., Sauer, N. and Lamb, C.J. ( 1987) Differential regulation of a hydroxyproline-rich glycoprotein gene family in wounded and infected plants. Mol. Cell. Biol.  7: 4337–4344. Google Scholar De Lorenzo, G., D’Ovidio, R. and Cervone, F. ( 2001) The role of polygalacturonase-inhibiting proteins (PGIPs) in defense against pathogenic fungi. Annu. Rev. Phytopathol.  39: 313–335. Google Scholar Dickman, M.B., Park, Y.K., Oltersdorf, T., Li, W., Clemente, T. and French, R. ( 2001) Abrogation of disease development in plants expressing animal antiapoptotic genes. Proc. Natl Acad. Sci. USA  98: 6957–6962. Google Scholar Dillon, P.J. and Rosen, C.A. ( 1990) A rapid method for the construction of synthetic genes using the polymerase chain reaction. Biotechniques  9: 298–300. Google Scholar Elliott, G.O., Hughes, R.K., Juge, N., Kroon, P.A. and Williamson, G. ( 2002) Functional identification of the cDNA coding for a wheat endo-1,4-β-d-xylanase inhibitor. FEBS Lett.  519: 66–70. Google Scholar Fierens, K., Brijs, K., Courtin, C.M., Gebruers, K., Goesaert, H., Raedschelders, G., Robben, J., Van Campenhout, S., Volckaert, G. and Delcour, J.A. ( 2003) Molecular identification of wheat endoxylanase inhibitor TAXI-I, member of a new class of plant proteins. FEBS Lett.  540: 259–263. Google Scholar Gautier, M.F., Aleman, M.E., Guirao, A., Marion, D. and Joudrier, P. ( 1994) Triticum aestivum puroindolines, two basic cystine-rich seed proteins: cDNA sequence analysis and developmental gene expression. Plant Mol. Biol.  25: 43–57. Google Scholar Gebruers, K., Debyser, W., Goesaert, H., Proost, P., Van Damme, J. and Delcour, J.A. ( 2001) Triticum aestivum L. endoxylanase inhibitor (TAXI) consists of two inhibitors, TAXI I and TAXI II, with different specificities. Biochem. J.  353: 239–244. Google Scholar Hart, C.M., Nagy, F. and Meins, F., Jr. ( 1993) A 61 bp enhancer element of the tobacco β-1,3-glucanase B gene interacts with one or more regulated nuclear proteins. Plant Mol. Biol.  21: 121–131. Google Scholar Holden, F.R. and Walton, J.D. ( 1992) Xylanases from the fungal maize pathogen Cochliobolus carbonum. Physiol. Mol. Plant Pathol.  40: 39–47. Google Scholar Kimura, M., Kaneko, I., Komiyama, M., Takatsuki, A., Koshino, H., Yoneyama, K. and Yamaguchi, I. ( 1998) Trichothecene 3-O-acetyltransferase protects both the producing organism and transformed yeast from related mycotoxins. Cloning and characterization of Tri101. J. Biol. Chem.  273: 1654–1661. Google Scholar Kumar, S., Tamura, K., Jakobsen, I.B. and Nei, M. ( 2001) MEGA2: molecular evolutionary genetics analysis software. Bioinformatics  17: 1244–1245. Google Scholar Leckie, F., Mattei, B., Capodicasa, C., Hemmings, A., Nuss, L., Aracri, B., De Lorenzo, G. and Cervone, F. ( 1999) The specificity of polygalacturonase-inhibiting protein (PGIP): a single amino acid substitution in the solvent-exposed β-strand/β-turn region of the leucine-rich repeats (LRRs) confers a new recognition capability. EMBO J.  18: 2352–2363. Google Scholar Liu, Y.G. and Whittier, R.F. ( 1995) Thermal asymmetric interlaced PCR: automatable amplification and sequencing of insert end fragments from P1 and YAC clones for chromosome walking. Genomics  25: 674–681. Google Scholar Mauch, F., Hadwiger, L.A. and Boller, T. ( 1988) Antifungal hydrolases in pea tissue. I. Purification and characterization of two chitinases and two β-1,3-glucanases differentially regulated during development and in response to fungal infection. Plant Physiol.  87: 325–333. Google Scholar McLauchlan, W.R., Garcia-Conesa, M.T., Williamson, G., Roza, M., Ravestein, P. and Maat, J. ( 1999) A novel class of protein from wheat which inhibits xylanases. Biochem. J.  338: 441–446. Google Scholar Nei, M. and Gojobori, T. ( 1986) Simple methods for estimating the numbers of synonymous and nonsynonymous nucleotide substitutions. Mol. Biol. Evol.  3: 418–426. Google Scholar Ohme-Takagi, M. and Shinshi, H. ( 1995) Ethylene-inducible DNA binding proteins that interact with an ethylene-responsive element. Plant Cell  7: 173–182. Google Scholar Page, R.D.M. ( 1996) TREEVIEW: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci.  12: 357–358. Google Scholar Pritsch, C., Muehlbauer, G.J., Bushnell, W.R., Somers, D.A. and Vance, C.P. ( 2000) Fungal development and induction of defense response genes during early infection of wheat spikes by Fusarium graminearum. Mol. Plant-Microbe Interact.  13: 159–169. Google Scholar Pritsch, C., Vance, C.P., Bushnell, W.R., Somers, D.A., Hohn, T.M. and Muehlbauer, G.J. ( 2001) Systemic expression of defense response genes in wheat spikes as a response to Fusarium graminearum infection. Physiol. Mol. Plant Pathol.  58: 1–12. Google Scholar Qin, Q., Bergmann, C.W., Rose, J.K., Saladie, M., Kolli, V.S., Albersheim, P., Darvill, A.G. and York, W.S. ( 2003) Characterization of a tomato protein that inhibits a xyloglucan-specific endoglucanase. Plant J.  34: 327–338. Google Scholar Qin, X.F., Holuigue, L., Horvath, D.M. and Chua, N.H. ( 1994) Immediate early transcription activation by salicylic acid via the cauliflower mosaic virus as-1 element. Plant Cell  6: 863–874. Google Scholar Rausher, M.D. ( 2001) Co-evolution and plant resistance to natural enemies. Nature  411: 857–864. Google Scholar Rebmann, G., Mauch, F. and Dudler, R. ( 1991) Sequence of a wheat cDNA encoding a pathogen-induced thaumatin-like protein. Plant Mol. Biol.  17: 283–285. Google Scholar Rose, T.M., Schultz, E.R., Henikoff, J.G., Pietrokovski, S., McCallum, C.M. and Henikoff, S. ( 1998) Consensus-degenerate hybrid oligonucleotide primers for amplification of distantly related sequences. Nucleic Acids Res.  26: 1628–1635. Google Scholar Rushton, P.J., Torres, J.T., Parniske, M., Wernert, P., Hahlbrock, K. and Somssich, I.E. ( 1996) Interaction of elicitor-induced DNA-binding proteins with elicitor response elements in the promoters of parsley PR1 genes. EMBO J.  15: 5690–5700. Google Scholar Sansen, S., De Ranter, C.J., Gebruers, K., Brijs, K., Courtin, C.M., Delcour, J.A. and Rabijns, A. ( 2004) Structural basis for inhibition of Aspergillus niger xylanase by Triticum aestivum xylanase inhibitor-I. J. Biol. Chem.  279: 36022–36028. Google Scholar Simpson, D.J., Fincher, G.B., Huang, A.H.C. and Cameron-Mills, V. ( 2003) Structure and function of cereal and related higher plant (1→4)-β-xylan endohydrolases. J. Cereal Sci.  37: 111–127. Google Scholar Stadnik, M.J. and Buchenauer, H. ( 2000) Inhibition of phenylalanine ammonia-lyase suppresses the resistance induced by benzothiadiazole in wheat to Blumeria graminis f. sp tritici. Physiol. Mol. Plant Pathol.  57: 25–34. Google Scholar Stotz, H.U., Bishop, J.G., Bergmann, C.W., Koch, M., Albersheim, P., Darvill, A.G. and Labavitch, J.M. ( 2000) Identification of target amino acids that affect interactions of fungal polygalacturonases and their plant inhibitors. Physiol. Mol. Plant Pathol.  56: 117–130. Google Scholar Strimmer, K. and von Haeseler, A. ( 1996) Quartet puzzling: a quartet maximum-likelihood method for reconstructing tree topologies. Mol. Biol. Evol.  13: 964–969. Google Scholar Takahashi-Ando, N., Kimura, M., Kakeya, H., Osada, H. and Yamaguchi, I. ( 2002) A novel lactonohydrolase responsible for the detoxification of zearalenone: enzyme purification and gene cloning. Biochem. J.  365: 1–6. Google Scholar Trail, F., Xu, J.R., San Miguel, P., Halgren, R.G. and Kistler, H.C. ( 2003) Analysis of expressed sequence tags from Gibberella zeae (anamorph Fusariumgraminearum). Fungal Genet. Biol.  38: 187–197. Google Scholar Vögeli-Lange, R., Fründt, C., Hart, C.M., Nagy, F. and Meins, F., Jr. ( 1994) Developmental, hormonal, and pathogenesis-related regulation of the tobacco class I β-1,3-glucanase B promoter. Plant Mol. Biol.  25: 299–311. Google Scholar Walton, J.D. ( 1994) Deconstructing the cell wall. Plant Physiol.  104: 1113–1118. Google Scholar Wang, G.L., Ruan, D.L., Song, W.Y., Sideris, S., Chen, L., Pi, L.Y., Zhang, S., Zhang, Z., Fauquet, C., Gaut, B.S., Whalen, M.C. and Ronald, P.C. ( 1998) Xa21D encodes a receptor-like molecule with a leucine-rich repeat domain that determines race-specific recognition and is subject to adaptive evolution. Plant Cell  10: 765–779. Google Scholar Wu, S.C., Ham, K.S., Darvill, A.G. and Albersheim, P. ( 1997) Deletion of two endo-β-1,4-xylanase genes reveals additional isozymes secreted by the rice blast fungus. Mol. Plant-Microbe Interact.  10: 700–708. Google Scholar Wu, S.S., Suen, D.F., Chang, H.C. and Huang, A.H. ( 2002) Maize tapetum xylanase is synthesized as a precursor, processed and activated by a serine protease, and deposited on the pollen. J. Biol. Chem.  277: 49055–49064. Google Scholar Yang, Y. and Klessig, D.F. ( 1996) Isolation and characterization of a tobacco mosaic virus-inducible myb oncogene homolog from tobacco. Proc. Natl Acad. Sci. USA  93: 14972–14977. Google Scholar

Ha, Chan Man;Jun, Ji Hyung;Nam, Hong Gil;Fletcher, Jennifer C.

doi: 10.1093/pcp/pch201pmid: 15564519

Abstract The BLADE-ON-PETIOLE1 (BOP1) gene of Arabidopsis thaliana is required for proper leaf morphogenesis. BOP1 regulates leaf differentiation in a proximal-distal manner, and represses the expression of three class I knotted-like homeobox (knox) genes during leaf formation. Utilizing a map-based approach, we identified the molecular nature of the BOP1 gene, which encodes a BTB/POZ domain protein with ankyrin repeats. BOP1 is a member of a small gene family in Arabidopsis that includes the disease resistance regulatory protein NPR1. Insertions in and around BOP1 cause distinct lesions in leaf morphogenesis, revealing complex regulation of the locus. BOP1 transcripts are initially detectable in embryos, where they specifically localize to the base of the developing cotyledons near the SAM. During vegetative development, BOP1 is expressed in young leaf primordia and at the base of the rosette leaves on the adaxial side. During reproductive development, BOP1 transcripts are detected in young floral buds, and at the base of the sepals and petals. Our results indicate that BOP1 encodes a putative regulatory protein that modulates meristematic activity at discrete locations in developing lateral organs. This is the first report on a plant protein that plays a key role in morphogenesis with the distinctive combinatorial architecture of the BTB/POZ and ankyrin repeat domains. (Received September 3, 2004; Accepted September 15, 2004) Introduction Leaves are the sites of photosynthesis and are the main carbon source for higher plants. Leaves develop as lateral organs from the growing shoot tip, the shoot apical meristem (SAM). In Arabidopsis thaliana, leaf initiation occurs through coordinated changes in the rates and planes of cell division in a small group of founder cells on the SAM periphery (Steeves and Sussex 1989). Following initiation, leaf morphogenesis proceeds with regional patterning of the primordia along the proximal-distal, adaxial-abaxial and central-lateral axes (Engstrom et al. 2004). Establishment of leaf polarity along these three axes begins shortly after organ initiation, and involves developmental gradients that partition the leaf into distinct morphological and anatomical domains. During the final stages of leaf development, cell and tissue specification occurs through coordinated processes of cell division, expansion and differentiation. The genetic regulation of these different stages of leaf formation is under intense investigation, and a number of loci involved in leaf morphogenetic events have been identified. The earliest molecular marker for Arabidopsis leaf initiation is the down-regulation of transcripts from the class 1 knotted-like homeobox (knox) gene SHOOTMERISTEMLESS (STM) from the incipient lateral primordia. STM is expressed throughout the SAM, and is required for the establishment and maintenance of the meristem cell population (Jackson et al. 1994, Long et al. 1996). STM fulfills this function in part through the negative regulation of ASYMMETRIC LEAVES1 (AS1) in the meristem apex, which restricts AS1 transcription to developing leaf primordia (Byrne et al. 2000). AS1 encodes a Myb domain transcription factor that confers leaf founder cell identity (Byrne et al. 2000), and has been shown to interact in yeast with a LOB domain leucine-zipper protein encoded by the AS2 locus (Iwakawa et al. 2002, Lin et al. 2003, Xu et al. 2003b). Loss-of-function as1 and as2 mutants form lobed leaves, occasionally bearing ectopic shoots, in which the STM-related, meristematic class I knox genes BREVIPEDICELLUS (BP) and KNAT2 are ectopically expressed (Byrne et al. 2000, Ori et al. 2000, Semiarti et al. 2001). Exclusion of knox gene activity from organ founder cells by AS1 and AS2 is therefore important in the acquisition of differentiated leaf cell fates. The mutual antagonism between AS1-AS2 and the knox genes therefore establishes a pathway that distinguishes leaf founder cells from SAM cells, which enables the spatial separation and tightly controlled regulation of leaf cell fate. In addition to AS1 and AS2, the BOP1 locus plays a key role in regulating meristematic activity in Arabidopsis leaves through the control of class I knox gene expression (Ha et al. 2003). bop1-1 plants are characterized by the formation of ectopic blade outgrowths along the adaxial side of cotyledon and rosette leaf petioles. These structures are generated through prolonged and clustered cell division in the mutant petioles, and have been interpreted as the result of disrupted cellular differentiation along the proximal-distal leaf axis. Three class I knox genes, BP, KNAT2 and KNAT6, are ectopically expressed in bop1-1 leaves, indicating a role for BOP1 in establishing and/or maintaining the repression of genes that normally promote the meristematic state. Interestingly, the requirement for BOP1 to control meristematic activity is not restricted to leaves, since bop1-1 plants also display ectopic outgrowths on the stem and at the base of floral organs. Here, we report the cloning of the BOP1 gene and the nature of the BOP1 gene product, the characterization of bop1 alleles and the pattern of BOP1 mRNA expression in wild-type Arabidopsis tissues. Results Map-based cloning of the BOP1 gene The BOP1 locus was previously mapped to the lower arm of chromosome 3, approximately 7.28 cM below the TT5 locus (Ha et al. 2003). To refine the BOP1 region, we analyzed 1,148 F2 progeny (2,296 chromosomes) that were homozygous for the bop1-1 allele, and detected tight linkage between the BOP1 and BGL1 loci. Cleaved amplified polymorphic sequence (CAPS) markers were then developed across the BACs F24I3 and F28O9 that cover the BGL1 locus. These markers were used to fine-map the BOP1 locus to an approximately 40 kb region between the CAPS markers F24-21 and F28-58 (Fig. 1A). A complementation testing strategy was used to delimit the position of BOP1 within this region. Overlapping cosmid clones were transformed into bop1-1 mutant plants that had been four-times backcrossed from Ler into the Ws-2 background. bop1-1 plants transformed with cosmid 145A10 showed complemented phenotypes in the T1 and T2 generations, with 94% of T1 plants and 100% of T2 plants displaying a revertant phenotype (Fig. 1A, B). Cosmid 84G23, which overlapped with the 5′ end of 145A10, failed to complement bop1-1, indicating that the 3′ portion of the 145A10 cosmid contained the BOP1 locus. Four annotated genes are located within this region of the 145A10 cosmid (Fig. 1A). Transformation of additional cosmid DNA overlapping 145A10 revealed that cosmids 5G22 and 66J5 failed to complement the bop1-1 mutation, while cosmid 100C24 was sufficient to confer a revertant phenotype. These data narrowed the candidates for BOP1 to a single annotated gene, At3g57130. Comparison of genomic DNA sequences revealed that At3g57130 contains a single nucleotide substitution in bop1-1 relative to wild-type Ler. This mutation alters the final nucleotide of the coding sequence, converting the stop codon to a tyrosine residue and leading to the insertion of four additional amino acids before a new stop codon is encountered (Fig. 2A). Taken together, these data demonstrate that At3g57130 corresponds to the BOP1 gene. Characterization of bop1 alleles The BOP1 gene consists of two exons and a single intron (Fig. 2A). Genomic sequence analysis revealed the presence of a TATA box and an in-frame stop codon upstream of the transcription start site. Rapid amplification of cDNA ends-polymerase chain reaction (RACE-PCR) delineated 215 bp of 5′ untranslated sequence and 211 bp of 3′ untranslated sequence, producing a 2.4 kb mature BOP1 transcript. Database searches revealed the presence of a locus on Arabidopsis chromosome 2 that is highly similar to BOP1 at the nucleotide and amino acid levels. This BOP1-like gene, At2g41370, also consists of two exons and a single intron, and shares approximately 82% nucleotide identity with BOP1. Three additional bop1 alleles have been identified that were generated by insertional mutagenesis. The bop1-2 allele is caused by a T-DNA insertion several kilobases upstream of the BOP1 locus (K. Krolikowski and S. Hake, personal communication). The bop1-3 allele is caused by a T-DNA insertion 446 bp upstream of the translation start site, while the bop1-4 allele is caused by a T-DNA insertion into the second exon of BOP1 (Fig. 2A). The transcription levels of BOP1 were analyzed by reverse transcription-polymerase chain reaction (RT-PCR) to detect differences in expression between wild-type and bop1 mutant plants. BOP1 transcripts are undetectable in bop1-4 mutant plants using primers spanning the BOP1 coding region, demonstrating that bop1-4 is a null allele for the locus. BOP1 transcript levels are slightly reduced in bop1-3 mutant plants compared with the wild type. In this analysis we observed that the BOP1 expression level is increased in bop1-1 plants compared with wild-type plants (Fig. 2B). We interpret this result as an indirect effect of an increase in BOP1-expressing leaf tissue (see below) in bop1-1 mutants. We next examined the bop1-3 and bop1-4 leaf phenotypes and compared them with those of bop1-1 plants. Relative to wild-type Ler leaves (Fig. 3A), bop1-1 leaves are misshapen and display ectopic outgrowths along the petiole and the base of the blade (Fig. 3B). bop1-3 plants, which have slightly reduced BOP1 transcript levels, have a weak phenotype. This phenotype consists of a single ectopic outgrowth from the leaf petiole region in approximately 2–3% of the homozygous plants (Fig. 3C). Plants carrying the bop1-4 RNA null allele also have a weak phenotype, displaying a single ectopic leaf outgrowth and mild leaf lobing in a proportion similar to that in bop1-3 plants (Fig. 3D). Examining bop1-1 in the heterozygous state, we determined that approximately three out of four bop1-1/+ plants formed a small ectopic outgrowth on one or two of their rosette leaves (Fig. 3E, F). This suggests that the bop1-1 allele is slightly semi-dominant with respect to leaf morphogenesis. Finally, we examined the bop1-1/bop1-3 and bop1-1/bop1-4 allele combinations and found that both combinations produced leaf phenotypes that were in the range of those observed in bop1-1 homozgyotes (Fig. 3G, H). Thus, the severity of the bop1 leaf phenotypes does not simply correlate with the level of BOP1 transcription in the various bop1 alleles. Structure of BOP1 and related proteins The BOP1 gene encodes a predicted protein of 467 amino acids with a mass of 51.8 kDa. The predicted BOP1 protein contains several distinct motifs that have been characterized in other plant and animal proteins. A BTB/POZ motif is located in the N-terminal region from residues 17 to 160, and an ankyrin repeat domain is located in the C-terminal region from residues 242 to 368 (Fig. 4A). In addition, two stretches of five and six histidine residues occur at the very C-terminal end of the protein. The bop1-1 stop codon mutation that adds four additional amino acids is predicted to introduce a novel β-sheet at the carboxyl-terminal end of the BOP1 protein (http://www.igb.uci.edu/tools/scratch), and to cause a conformational change in the overall protein structure. An alignment of BOP1 with other BTB/POZ and ankyrin-repeat-containing proteins is shown in Fig. 4. BOP1, the BOP1-like protein encoded by At2g41370 and a putative ortholog from rice (OsBOP1) share the highest degree of amino acid conservation. The predicted product of the At2g41370 gene is 83% similar and 78% identical to BOP1 across the full length of the proteins, but contains 29 additional amino acids. OsBOP1 is 74% similar and 67% identical to Arabidopsis BOP1, although OsBOP1 lacks the two stretches of histidine residues that are present at the carboxyl terminus of Arabidopsis BOP1. OsBOP1 also contains a 24 amino acid insertion in the BTB/POZ domain (Fig. 4B) and a nine amino acid insertion in the first ankyrin repeat (Fig. 4C) that are absent from BOP1 and the other BOP1-like proteins in our analysis. The combinatorial BTB/POZ and ankyrin repeat domain structure is conserved between BOP1 and the Arabidopsis regulatory protein NPR1 (NONEXPRESSOR OF PR GENES1) (Cao et al. 1997, Ryals et al. 1997), also known as NIM1 (Fig. 4B, C). NPR1 controls the onset of systemic acquired resistance (SAR) to a broad spectrum of pathogens, which is normally established after a primary exposure to avirulent pathogens (reviewed in Pieterse and Van Loon 2004). BOP1 shares 47% amino acid similarity and 30% identity with NPR1 across the full length of the proteins. BOP1 also aligns with NML (NIM-like protein) sequences that have been identified in several plant species, for example sharing 33% amino acid identity and 49–52% similarity with the NML1 and NML2 proteins from Lycopersicon esculentum. Three other uncharacterized Arabidopsis genes also encode proteins that are predicted to contain both a BTB/POZ domain and ankyrin repeats. The predicted gene products of At4g19660 and At5g45110 share 31% amino acid identity with BOP1, and At4g26120 is 30% identical across the length of the proteins. The RPT2 protein, which is involved in phototropic signal transduction (Sakai et al. 2000), contains a BTB/POZ domain that shares similarity with BOP1, but lacks an ankyrin repeat domain. Expression pattern of BOP1 To determine the sites of BOP1 activity, we examined the distribution of BOP1 transcript accumulation in wild-type and bop1-1 mutant plants. We isolated total RNA from 14 different tissue types and stages of development, and analyzed the spatial and temporal expression of BOP1 by semi-quantitative RT-PCR. BOP1 transcripts are detected at high levels in vegetative shoot apices, rosette leaves and floral buds, and at lower levels in stem nodes and mature flowers (Fig. 5A). Because the phenotypes of bop1-1 plants indicated an important role for BOP1 in rosette leaf morphogenesis (Ha et al. 2003), we analyzed BOP1 expression temporally in both young and mature rosette leaves, and spatially in the petiole and blade domains. BOP1 appears to be transcribed at the same relative level in whole rosette leaves regardless of age, but is expressed at very high levels in the petiole relative to the blade (Fig. 5A). This is consistent with the bop1-1 phenotype in which ectopic outgrowths develop more readily from the petioles of the rosette leaves. BOP1 is expressed at barely detectable levels in cotyledons and below the level of detection in cauline leaves, despite the observation that bop1-1 plants can form ectopic outgrowths on these organs (see below, Ha et al. 2003). BOP1 transcripts are also barely detectable in roots, and are not detected in stem internode tissue, flower pedicels and siliques (Fig. 5A). In bop1-1 mutant plants, BOP1 transcription is elevated in all tissues compared with the wild type and becomes readily detectable in cotyledons and cauline leaves (Fig. 5B), consistent with the manifestation of phenotypes in these tissues. Next, we analyzed the specific pattern of BOP1 gene expression in wild-type tissues throughout Arabidopsis development. The first stage at which BOP1 transcripts can be reliably detected is the torpedo stage of embryogenesis, when BOP1 is found in two discrete foci at the base of the developing cotyledons, specifically on the adaxial side near the boundary with the shoot apical meristem (Fig. 6A). This expression pattern remains unchanged as the embryos mature (Fig. 6B). After germination, BOP1 transcripts are not observed in the SAM, but are present in the rosette leaf primordia initiating from its flanks (Fig. 6D). As the leaves grow out, BOP1 is expressed strongly yet in a highly restricted pattern at the base of the leaf primordia on the adaxial side adjacent to the SAM (Fig. 6E). This BOP1 leaf expression pattern mimics the cotyledon expression pattern. BOP1 transcripts are also present along the margin of young leaves, where the adaxial and abaxial domains juxtapose (Fig. 6F). In maturing leaves, a low level of BOP1 expression becomes detectable across the adaxial portion of the leaf (Fig. 6D, F). bop1-1 mutant plants form ectopic outgrowths at the base of sepals and petals, indicating that the gene plays a role in flower development (Ha et al. 2003), and indeed BOP1 shows a highly restricted expression pattern during the reproductive phase. BOP1 is absent from the inflorescence meristem but is expressed in early floral buds (Fig. 6H). In floral meristems and maturing flowers, BOP1 mRNA is specifically localized at the base of the developing sepals and petals, beginning as early as stage 3 (Fig. 6H, J). In mature flowers, BOP1 continues to be transcribed at the base of the sepals and petals (Fig. 6K). No expression is detected at any stage using a BOP1 sense probe, except in the leaf stipules (Fig. 6C, G, I). In sum, the vegetative and reproductive expression pattern of BOP1 correlates very precisely with the tissues in which phenotypes are detected in bop1-1 mutant plants (Ha et al. 2003). Discussion We have cloned the BOP1 gene using a map-based strategy, and shown that it encodes a protein with a BTB/POZ domain and ankyrin repeats that is expressed in a highly restricted pattern during Arabidopsis development. Although there are 93 BTB/POZ-containing proteins and more than 240 ankyrin-repeat-containing proteins encoded in the A. thaliana genome alone, only a handful of proteins from Arabidopsis and other higher plants contain both domains. BOP1 represents the second plant protein with this type of combinatorial protein architecture to be functionally characterized, and the first that plays a key role in morphogenesis. Functional domains of the BOP1 protein The amino terminus of BOP1 shows strong similarity to proteins containing a BTB/POZ domain. The BTB/POZ domain is an evolutionarily conserved ~120 amino acid region that was originally named for the DrosophilaBroad-Complex, Tramtrack and Bric-a-brac proteins (BTB) (Zollman et al. 1994), and for two of the main classes of proteins in which it can be found, Pox virus and zinc finger proteins (POZ) (Bardwell and Treisman 1994). The BTB/POZ domain is conserved throughout eukaryotes, but has undergone independent expansion in plants and different animal lineages (Aravind and Koonin 1999). Three-dimensional structural analysis of the BTB/POZ domain shows that it has two distinct parts. The amino-terminal half forms a four-stranded β-sheet with two associated short α-helices, while the carboxyl-terminal portion forms two α-helices separated by an extended region (Aravind and Koonin 1999). The BTB/POZ domain has been shown to mediate protein homodimerization or multimerization (Dhordain et al. 1995), and in some instances can participate in heteromeric interactions (David et al. 1998). Proteins containing this domain are involved in a wide variety of processes, including cytoskeletal organization, transcriptional repression, chromatin remodeling and proteolysis. The BOP1 protein also contains four copies of an ankyrin repeat motif. Ankyrin repeats are tandemly repeated ~33 amino acid modules, and represent one of the most common protein-protein interaction motifs found in nature. Each ankyrin repeat consists of pairs of antiparallel α-helix-loop-helix structures connected by a series of extended β-hairpin motifs (Sedgwick and Smerdon 1999), an arrangement that is conserved in the four ankyrin repeats of BOP1 (Fig. 4C). The ankyrin repeat motif has been found in a diverse group of proteins involved in cytoskeletal organization, transcriptional regulation, cell differentiation, and enzymatic and toxic activities (Sedgwick and Smerdon 1999). Examples of ankyrin-repeat-containing proteins include the cytoskeletal protein ankyrin, 53BP2 (p53 Binding Protein2), the mammalian transcription factor NF-κB and its inhibitor I-κB. When compared with characterized plant proteins, the structure of BOP1 is most similar to that of the Arabidopsis NPR1 protein. NPR1 contains both a BTB/POZ domain and ankyrin repeats (Cao et al. 1997, Ryals et al. 1997), and is required for regulating salicylic acid-dependent gene expression during SAR (Cao et al. 1994). Although NPR1 is known to regulate PR defense gene expression in the nucleus, the protein lacks a bona fide DNA binding domain. However, NPR1 has been shown to interact in yeast two-hybrid assays with members of the TGA subclass of basic leucine zipper (bZIP) transcription factors (Zhang et al. 1999), and to regulate their DNA binding activity (Despres et al. 2003). Recent studies have shown that NPR1 protein is present in the cytoplasm in an oligomeric form mediated by intermolecular disulfide bonds (Mou et al. 2003). Upon SAR induction, NPR1 is reduced to a monomeric form, which is translocated from the cytosolic to the nuclear compartment where it interacts with its TGA transcription factor partners (Pieterse and Van Loon 2004). NPR1 significantly increases the binding of TGA2 to SA-responsive promoter elements in the PR-1 gene (Despres et al. 2000), indicating that NPR1-mediated DNA binding of TGA factors is important for PR gene activation. Interestingly, in addition to its roles in defense and cell death, NPR1 has been shown to promote cell division and/or to suppress endoreduplication during leaf development (Vanacker et al. 2001). Like NPR1, BOP1 protein localizes to both the nucleus and the cytosol (data not shown), although it remains to be seen whether BOP1 interacts with any members of the TGA transcription factor family. Possible mechanisms of BOP1 activity In analyzing the available bop1 mutants, we observed that the level of BOP1 transcription in the various bop1 alleles does not appear to correlate with phenotype severity. bop1-4 RNA null mutant plants, which have a mild phenotype, make no detectable full-length BOP1 mRNA. The deficiency of BOP1 protein in the null mutant may be largely complemented by the activity of the BOP1-like protein encoded by At2g41370 (Fig. 4). In contrast to bop1-4 null mutants, bop1-1 plants have a severe leaf phenotype and increased levels of BOP1 expression. These contradictory observations can be explained by the specific expression pattern of BOP1 at the base and the margin of wild-type leaves. The primary leaves of bop1-1 mutants form ectopic leaf outgrowths, each of which would be expected to express BOP1 at the margin and base. Thus many more cells express BOP1 in bop1-1 plants than in wild-type plants, accounting for the increased expression at the level of the whole plant. Our results indicate that the originally characterized bop1-1 allele confers a weakly semi-dominant phenotype in Arabidopsis rosette leaves, as bop1-1 heterozygous plants have a very weak phenotype (Fig. 3E). The semi-dominant phenotype of bop1-1 could be derived from a haploid deficiency in the heterozygous plants or from a dominant-negative interaction between the mutant allele and the wild-type allele. Since the mutant phenotype of homozygous bop1-1 plants is more severe than that of bop1-4 null mutant plants, haplo-insufficiency is unlikely to be the reason for the semi-dominance of the bop1-1 allele. Thus, we suggest that the mutant BOP1-1 protein may interfere with the normal function of other proteins in the leaf morphogenesis pathway. Mutant BOP1-1 protein, possibly in an altered conformational state due to addition of the C-terminal amino acids, could, for instance, bind to one or more of its regular interaction partners and titrate away its normal function in the cell. A precedent for this hypothesis is the demonstration that chromosomal translocations of the BTB/POZ proteins PLZF and BCL6 result in novel fusion proteins that act in a dominant-negative fashion by sequestering their partner proteins into inactive multimeric complexes (Bardwell and Treisman 1994). The formation of simple leaves by Arabidopsis plants requires the down-regulation of class I knox genes such as STM, BP and KNAT2, since ectopic expression of one or more of these genes in developing leaves causes leaf lobing and compounding (Lincoln et al. 1994, Chuck et al. 1996, Pautot et al. 2001). We have previously shown that BOP1 is necessary to prevent the formation of ectopic, lobed blades along the adaxial side of cotyledons and rosette leaves, and to negatively affect the expression of BP, KNAT2 and KNAT6 in Arabidopsis leaves (Ha et al. 2003). Given that the BOP1 protein contains several motifs that mediate protein-protein interactions, we hypothesize that BOP1 may exert its regulatory effects in leaf primordia by interacting with other leaf-expressed gene products through its BTB/POZ and/or its ankyrin repeat domains. An obvious possibility is that BOP1 interacts with the product of the BOP1-like gene At2g41370, and that their activities may significantly overlap. In bop1-1 plants, for example, mutant BOP1-1 protein might also interfere with the function of the BOP1-like protein, causing phenotypes more severe than those of the bop1-4 null mutant. Other candidates for leaf-specific co-acting factors include both AS1 and AS2, since BOP1 appears to function synergically with these two proteins to control leaf morphogenesis. AS1 and AS2, like BOP1, prevent the ectopic expression of class I knox genes in lateral organ primordia (Byrne et al. 2000, Ori et al. 2000, Semiarti et al. 2001). Thus, these three proteins might act together in a complex that represses knox gene transcription in leaves. However, bop1-1 plants have some cotyledon and leaf phenotypes that are not observed in as1 or as2 plants, suggesting that at least a subset of BOP1 functions is independent of these two genes. Another potential BOP1-interacting protein is the product of the LATERAL ORGAN BOUNDARIES (LOB) gene, which is expressed in a pattern similar to BOP1 in developing leaves (Shuai et al. 2002). Thus BOP1 might interact with LOB or some of the related LBD proteins (Shuai et al. 2002). The combinatorial architecture of the BOP1 protein suggests several possible mechanisms through which BOP1 might regulate leaf morphogenesis. The BTB/POZ domain has been proposed to act as a scaffold for the organization of higher-order cellular structures, such as the cytoskeleton, ubiquitin ligase substrate complexes and chromatin. Some animal BTB/POZ proteins, such as BCL6 and PLZF, interact with transcriptional co-repressors that can recruit histone deacetylases for chromatin remodeling at the target loci (Huynh and Bardwell 1998, Lin et al. 1998). More recently, other BTB/POZ proteins have been shown to act as substrate-specific adaptor proteins for Cullin3 E3 ubiquitin ligases in yeast and Caenorhabditis elegans (Geyer et al. 2003, Pintard et al. 2003, Xu et al. 2003a). Based on the BCL6 and PLZF protein paradigm, one hypothesis is that BOP1 might act as a transcriptional repressor of class I knox gene expression in leaf primordia. Through its BTB/POZ domain, BOP1 could potentially recruit histone deacetylases or other chromatin remodeling proteins to the BP, KNAT2 and KNAT6 promoters to lock them in a transcriptionally inactive state as leaf maturation progresses. Alternatively, BOP1 may act in a proteasome-mediated degradation pathway that targets one or more meristem-promoting factors, or meristem gene-activating factors, for turnover in developing leaves. The ankyrin repeats could mediate either the recruitment of the interaction partners, or the stability of the resulting protein complexes. Future biochemical experiments to determine the interaction partners of BOP1 will help to distinguish between these possibilities. Materials and Methods Plant material and growth conditions The bop1-3 allele (SALK_012994) was obtained from the Arabidopsis Biological Resource Center (ABRC) at Ohio State University (Columbus, OH, U.S.A.). The bop1-4 allele (386G09) was obtained from the Genome Analysis of the Plant Biological System project (GABI-Kat, Cologne, Germany). Seeds were imbibed at 4°C for 3 d before sowing on soil and were grown in a greenhouse under long days (16 h light and 8 h dark) with a day/night temperature cycle of 22°C/18°C. For the genetic analysis of F1 plants, we reciprocally crossed bop1-1 mutant plants with Ler plants, and with bop1-3 and bop1-4 mutant plants. A total of 76 F1 plants between bop1-1 and Ler, 13 F1 plants between bop1-1 and bop1-3, and 12 F1 plants between bop1-1 and bop1-4 were observed for the ectopic outgrowths phenotype. Map-based cloning The bop1-1 mutation was mapped using cleaved amplified polymorphic sequence (CAPS) markers (Konieczny and Ausubel 1993). bop1-1 mutant plants in the Ler ecotype were crossed with the Columbia (Col) ecotype. DNA for mapping studies was prepared from 1,148 individual F2 progeny plants with mutant phenotypes. Additional CAPS markers for fine mapping were developed using Arabidopsis genome sequence data (TAIR, http://www.Arabidopsis.org). The BOP1 nucleotide and deduced protein sequences were used for database searches using the BLAST network service (http://www.ncbi.nlm.nih.gov/BLAST/). The amino acid sequences of BOP1 and homologous proteins were aligned using CLUSTAL W, version 1.8 (Thompson et al. 1994). Alignment of the ankyrin repeats was performed using the consensus sequence described in Sedgwick and Smerdon (1999). CAPS marker sequences developed for this study are available upon request. Complementation of bop1-1 The cosmid clones that contain the BOP1 genomic regions were obtained from the Genomic Arabidopsis Resource Network (GARNet; John Innes Centre, Norwich, UK). Overlapping cosmid clones of GARNet were transformed into Agrobacterium (GV3101 strain), and in turn used to transform bop1-1 mutants using the floral dip method (Clough and Bent 1998). T1 seeds were plated on media supplemented with 40 µg ml–1 kanamycin (Sigma, St. Louis, MO, U.S.A.). Kanamycin resistance and rescued phenotypes were observed after 2 weeks. RACE-PCR Total RNA from 12-day-old Ler seedlings was prepared using RNeasy kit (Qiagen, Valencia, CA, USA). RACE-PCR cDNA was synthesized using GeneRacer™ Kit (Invitrogen, Carlsbad, CA, U.S.A.). The 5′ end sequence of the cDNA was determined by 5′-RACE with primer RBI-5(1) (5′-ATCTAAAGAGAGATCGACGGCTGCA-3′) and a GeneRacer™ 5′ primer. The 3′ end sequence of the cDNA was determined by 3′-RACE with primer RBI-3(1) (5′-GTGATGGGAGAAGGACTCAATCTAGA-3′) and a GeneRacer™ 3′ primer. PCR product was cloned into the pCR“2.1-TOPO“ vector (Invitrogen) and sequenced. RT-PCR For the analysis of Ler wild-type plants, shoot apices, cotyledons, rosette leaf petioles, rosette leaf blades, young rosette leaves and mature rosette leaves were collected 17 d after planting; roots were collected 13 d after planting; cauline leaves, inflorescence stem nodes and internodes, floral buds including the inflorescence meristem, open flowers, pedicels and green siliques were collected 35 d after planting. To monitor BOP1 transcript levels, total RNA was extracted from 14-day-old Ler, bop1-1, bop1-3 and bop1-4 seedlings. Five micrograms of total RNA was reverse transcribed from an oligo(dT15) primer using SuperScript™III Reverse Transcriptase (Invitrogen). One microliter of the first-strand cDNA reaction was used as a template for PCR amplification with BOP1-specific primers (BIa: BOP1-RT-76R; 5′-GTGAATCTGATCCTTCGCAACC-3′, BId: F24-76F; 5′-ATCCAAACTACTTCCGCTCGTG-3′, BIb: 5′-CTAAAATCAAAGGGTACGACA-3′, BIc: 5′-AGAGGCATTGAAGATTTGAGA-3′). Primers to amplify the ArabidopsisTUBULIN4 gene were used as an internal control (Ha et al. 2003). In situ hybridization For probe generation, the full-length BOP1 cDNA sequence was amplified using BOP1 specific primers (BI-IS1: 5′-GTCGACCTCTCTCTTCTTCATCTTCT-3′; BI-IS2: 5′-ACTAGT CAACAATTTAATTAGATATTGAATATG-3′) and cloned into the SalI and SpeI site of the pBluescript KS+ plasmid (Stratagene, La Jolla, CA, U.S.A.). T3 and T7 polymerase (Promega, Madison, WI, U.S.A.) were used to synthesize sense and antisense UTP-digoxigenin-labeled RNA as described (Jackson 1992). Plant fixation and in situ hybridization were performed as described (Jackson 1992). Acknowledgments We acknowledge the Arabidopsis Biological Resource Center, the Genome Analysis of the Plant Biological System project and the Salk Institute for Biological Studies for providing insertion lines, and Genomic Arabidopsis Resource Network for cosmid clones. This work was supported by a fellowship from the Postdoctoral Fellowship Program of the Korea Science and Engineering Foundation (KOSEF) to C.M.H., a grant from the Crop Functional Genomics Frontier Research Program of the Ministry of Science and Technology of Korea to H.G.N. and a United States Department of Agriculture CRIS grant to J.C.F. 4 These authors contributed equally to this work. 5 Corresponding authors: E-mail, [email protected] or [email protected]; Fax, +1-510-559-5678 or +82-54-279-5972. View largeDownload slide Fig. 1 Map-based cloning of BOP1. (A) Map of the BOP1 locus on chromosome 3. F24I3 and F28O9 are BAC clones. The positions of the mapping markers are indicated, and the number of recombination events is shown in parenthesis. The BOP1 locus was mapped to an approximately 40 kb region between the molecular markers F24-21 and F28-58. Cosmid clones that overlap the region of interest are shown. Cosmids that complemented the bop1-1 phenotype are labeled with asterisks. The complementing region contains four open reading frames (arrows). bop1-1 plants have a mutation in the At3g57130 gene. (B) Complementation of the bop1-1 phenotype. Photographs of a wild-type (Ler) plant, a bop1-1 mutant plant and a bop1-1 plant transformed with cosmid 145A10 were taken at 20 d after planting. View largeDownload slide Fig. 1 Map-based cloning of BOP1. (A) Map of the BOP1 locus on chromosome 3. F24I3 and F28O9 are BAC clones. The positions of the mapping markers are indicated, and the number of recombination events is shown in parenthesis. The BOP1 locus was mapped to an approximately 40 kb region between the molecular markers F24-21 and F28-58. Cosmid clones that overlap the region of interest are shown. Cosmids that complemented the bop1-1 phenotype are labeled with asterisks. The complementing region contains four open reading frames (arrows). bop1-1 plants have a mutation in the At3g57130 gene. (B) Complementation of the bop1-1 phenotype. Photographs of a wild-type (Ler) plant, a bop1-1 mutant plant and a bop1-1 plant transformed with cosmid 145A10 were taken at 20 d after planting. View largeDownload slide Fig. 2 BOP1 gene structure and alleles. (A) Schematic representation of the BOP1 locus, showing the bop1-1 mutation and the location of the T-DNA insertions in the bop1-3 and bop1-4 alleles. The exon/intron organization of At3g57130 and the deduced protein structure are shown. The translation start codon (ATG) is designated nucleotide +1, and the stop codon (TAG) is at nucleotide position +2192. At3g57130 contains two exons (box) and one intron (bar). White boxes indicate the untranslated region in the exons, and gray boxes indicate the translated regions. The transcription start sites (–215) were determined by RACE-PCR. The bop1-1 mutant has a single G to C nucleotide substitution in the stop codon, which results in the addition of four amino acids to the predicted protein. Arrows indicate the locations of primers used for RT-PCR analysis. The BTB/POZ and ankyrin repeat domains of the BOP1 protein are indicated. (B) BOP1 transcription in bop1 mutant plants. Total RNA isolated from the indicated 2-week-old plants was analyzed using RT-PCR to monitor the transcript accumulation of BOP1. Each PCR amplification of cDNA was conducted in parallel with specific primer sets for the BOP1 coding region (F and R) and for TUBULIN4 as a control. View largeDownload slide Fig. 2 BOP1 gene structure and alleles. (A) Schematic representation of the BOP1 locus, showing the bop1-1 mutation and the location of the T-DNA insertions in the bop1-3 and bop1-4 alleles. The exon/intron organization of At3g57130 and the deduced protein structure are shown. The translation start codon (ATG) is designated nucleotide +1, and the stop codon (TAG) is at nucleotide position +2192. At3g57130 contains two exons (box) and one intron (bar). White boxes indicate the untranslated region in the exons, and gray boxes indicate the translated regions. The transcription start sites (–215) were determined by RACE-PCR. The bop1-1 mutant has a single G to C nucleotide substitution in the stop codon, which results in the addition of four amino acids to the predicted protein. Arrows indicate the locations of primers used for RT-PCR analysis. The BTB/POZ and ankyrin repeat domains of the BOP1 protein are indicated. (B) BOP1 transcription in bop1 mutant plants. Total RNA isolated from the indicated 2-week-old plants was analyzed using RT-PCR to monitor the transcript accumulation of BOP1. Each PCR amplification of cDNA was conducted in parallel with specific primer sets for the BOP1 coding region (F and R) and for TUBULIN4 as a control. View largeDownload slide Fig. 3 Comparison of the bop1 leaf phenotypes. (A) Wild-type Ler rosette leaf. (B) bop1-1 rosette leaf. (C) bop1-3 rosette leaf. (D) bop1-4 rosette leaf. (E) bop1-1/+ rosette leaf. (F) Higher magnification of the ectopic outgrowth from the bop1-1/+ leaf shown in (E). (G) bop1-1/bop1-3 rosette leaf. (H) bop1-1/bop1-4 rosette leaf. The shape of the rosette leaf reflects the severity of the Bop1– phenotype. Arrows indicate the presence of ectopic outgrowths. View largeDownload slide Fig. 3 Comparison of the bop1 leaf phenotypes. (A) Wild-type Ler rosette leaf. (B) bop1-1 rosette leaf. (C) bop1-3 rosette leaf. (D) bop1-4 rosette leaf. (E) bop1-1/+ rosette leaf. (F) Higher magnification of the ectopic outgrowth from the bop1-1/+ leaf shown in (E). (G) bop1-1/bop1-3 rosette leaf. (H) bop1-1/bop1-4 rosette leaf. The shape of the rosette leaf reflects the severity of the Bop1– phenotype. Arrows indicate the presence of ectopic outgrowths. View largeDownload slide Fig. 4 Sequences similar to BOP1. (A) Amino acid sequence of the BOP1 gene product. The BTB/POZ domain is shaded, the ankyrin repeat region is underlined and the two stretches of histidines are boxed. (B) Amino acid alignment of the BTB/POZ domain in BOP1 and related plant proteins. (C) Amino acid alignment of the ankyrin repeat region. The alignment was performed using the CLUSTALW program. Asterisks indicate identical amino acids, colons indicate the conservative amino acid substitutions and dots indicate semi-conservative amino acid substitutions. The four consecutive ankyrin repeats are indicated by long arrows above the sequence. Protein accession numbers: BOP1, NP_191272; At2g41370, NP_181668; OsBOP, P0466H10; LeNML1, AAT57638; NtNML1, AAT57641; LeNML2, AAT57639; NPR1, AAM65726; RTP2, NP_568989. View largeDownload slide Fig. 4 Sequences similar to BOP1. (A) Amino acid sequence of the BOP1 gene product. The BTB/POZ domain is shaded, the ankyrin repeat region is underlined and the two stretches of histidines are boxed. (B) Amino acid alignment of the BTB/POZ domain in BOP1 and related plant proteins. (C) Amino acid alignment of the ankyrin repeat region. The alignment was performed using the CLUSTALW program. Asterisks indicate identical amino acids, colons indicate the conservative amino acid substitutions and dots indicate semi-conservative amino acid substitutions. The four consecutive ankyrin repeats are indicated by long arrows above the sequence. Protein accession numbers: BOP1, NP_191272; At2g41370, NP_181668; OsBOP, P0466H10; LeNML1, AAT57638; NtNML1, AAT57641; LeNML2, AAT57639; NPR1, AAM65726; RTP2, NP_568989. View largeDownload slide Fig. 5 BOP1 expression pattern. (A) BOP1 expression in wild-type Ler tissues. (B) BOP1 expression in bop1-1 tissues. RT-PCR was performed using RNA isolated from seedling shoot apices 12 d after planting, cotyledons, young rosette leaves (RL), mature rosette leaves (RL), rosette leaf (RL) petiole, rosette leaf (RL) blade, cauline leaves (CL), inflorescence stem nodes, inflorescence stem internodes, floral buds including inflorescence meristem, open flowers, pedicels, green siliques and roots. Each PCR amplification of cDNA was conducted in parallel with specific primer sets for BOP1 (34 cycles) and for TUBULIN4 (29 cycles) as a control. View largeDownload slide Fig. 5 BOP1 expression pattern. (A) BOP1 expression in wild-type Ler tissues. (B) BOP1 expression in bop1-1 tissues. RT-PCR was performed using RNA isolated from seedling shoot apices 12 d after planting, cotyledons, young rosette leaves (RL), mature rosette leaves (RL), rosette leaf (RL) petiole, rosette leaf (RL) blade, cauline leaves (CL), inflorescence stem nodes, inflorescence stem internodes, floral buds including inflorescence meristem, open flowers, pedicels, green siliques and roots. Each PCR amplification of cDNA was conducted in parallel with specific primer sets for BOP1 (34 cycles) and for TUBULIN4 (29 cycles) as a control. View largeDownload slide Fig. 6 BOP1 RNA localization by in situ hybridization. (A) Torpedo stage Ler embryo. (B) Mature embryo. (C) Mature embryo hybridized with a BOP1 sense probe. (D) Central section of a seedling 8 d after planting. (E) Peripheral section of an 8-day-old seedling. (F) Tranverse section of an 8-day-old seedling. (G) Transverse section of an 8-day-old seedling hybridized with a BOP1 sense probe. (H) Central section of an inflorescence meristem and young floral buds. (I) Central section of an inflorescence meristem and young floral buds hybridized with a BOP1 sense probe. (J) Transverse section of an inflorescence meristem and young floral buds. (K) Tangential section of a mature flower. Scale bar = 100 µm (A–C), 50 µm (D–K). View largeDownload slide Fig. 6 BOP1 RNA localization by in situ hybridization. (A) Torpedo stage Ler embryo. (B) Mature embryo. (C) Mature embryo hybridized with a BOP1 sense probe. (D) Central section of a seedling 8 d after planting. (E) Peripheral section of an 8-day-old seedling. (F) Tranverse section of an 8-day-old seedling. (G) Transverse section of an 8-day-old seedling hybridized with a BOP1 sense probe. (H) Central section of an inflorescence meristem and young floral buds. (I) Central section of an inflorescence meristem and young floral buds hybridized with a BOP1 sense probe. (J) Transverse section of an inflorescence meristem and young floral buds. (K) Tangential section of a mature flower. Scale bar = 100 µm (A–C), 50 µm (D–K). Abbreviations BAC bacterial artificial chromosome BTB Broad-Complex/Tramtrack/Bric-a-brac CAPS cleaved amplified polymorphic sequence knox knotted-like homeobox POZ pox virus/zinc finger RACE-PCR rapid amplification of cDNA ends-polymerase chain reaction RT-PCR reverse transcription-polymerase chain reaction SAM shoot apical meristem SAR systemic acquired resistance. References Aravind, L. and Koonin, E.V. ( 1999) Fold prediction and evolutionary analysis of the POZ domain: Structural and evolutionary relationship with the potassium channel tetramerization domain. J. Mol. Biol.  285: 1353–1361. Google Scholar Bardwell, V.J. and Treisman, R. ( 1994) The POZ domain: a conserved protein-protein interaction motif. Genes Dev.  8: 1664–1677. Google Scholar Byrne, M.E., Barley, R., Curtis, M., Arroyo, J.M., Dunham, M., Hudson, A. and Martienssen, R.A. ( 2000) Asymmetric leaves1 mediates leaf patterning and stem cell function in Arabidopsis. Nature  408: 967–971. Google Scholar Cao, H., Bowling, S.A., Gordon, S. and Dong, X. ( 1994) Characterization of an Arabidopsis mutant that is nonresponsive to inducers of systemic acquired resistance. Plant Cell  6: 1583–1592. Google Scholar Cao, H., Glazebrook, J., Clarke, J.D., Volko, S. and Dong, X. ( 1997) The Arabidopsis NPR1 gene that controls systemic acquired resistance encodes a novel protein containing ankyrin repeats. Cell  88: 57–63. Google Scholar Chuck, G., Lincoln, C. and Hake, S. ( 1996) KNAT1 induces lobed leaves with ectopic meristems when overexpressed in Arabidopsis. Plant Cell  8: 1277–1289. Google Scholar Clough, S.J. and Bent, A.F. ( 1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J.  16: 735–743. Google Scholar David, G., Alland, L., Hong, S.H., Wong, C.W., DePinho, R.A. and Dejean, A. ( 1998) Histone deacetylase associated with mSin3A mediates repression by the acute promyelocytic leukemia-associated PLZF protein. Oncogene  16: 2549–2556. Google Scholar Despres, C., Chubak, C., Rochon, A., Clark, R., Bethune, T., Desveaux, D. and Fobert, P.R. ( 2003) The Arabidopsis NPR1 disease resistance protein is a novel cofactor that confers redox regulation of DNA binding activity to the basic domain/leucine zipper transcription factor TGA1. Plant Cell  15: 2181–2191. Google Scholar Despres, C., DeLong, C., Glaze, S., Liu, E. and Fobert, P.R. ( 2000) The Arabidopsis NPR1/NIM1 protein enhances the DNA binding activity of a subgroup of the TGA family of bZIP transcription factors. Plant Cell  12: 279–290. Google Scholar Dhordain, P., Albagli, O., Ansieau, S., Koken, M.H., Deweindt, C., Quief, S., Lantoine, D., Leutz, A., Kerckaert, J.P. and Leprince, D. ( 1995) The BTB/POZ domain targets the LAZ3/BCL6 oncoprotein to nuclear dots and mediates homomerisation in vivo. Oncogene  11: 2689–2697. Google Scholar Engstrom, E.M., Izhaki, A. and Bowman, J.L. ( 2004) Promoter bashing, microRNAs, and Knox genes. New insights, regulators, and targets-of-regulation in the establishment of lateral organ polarity in Arabidopsis. Plant Physiol.  135: 685–694. Google Scholar Geyer, R., Wee, S., Anderson, S., Yates, J., III and Wolf, D.A. ( 2003) BTB/POZ domain proteins are putative substrate adaptors for Cullin 3 ubiquitin ligases. Mol. Cell  12: 783–790. Google Scholar Ha, C.M., Kim, G.-T., Kim, B.C., Jun, J.H., Soh, M.S., Ueno, Y., Machida, Y., Tsukaya, H. and Nam, H.G. ( 2003) The BLADE-ON-PETIOLE1 gene controls leaf pattern formation through the modulation of meristematic activity in Arabidopsis . Development  130: 161–172. Google Scholar Huynh, K.D. and Bardwell, V.J. ( 1998) The BCL-6 POZ domain and other POZ domains interact with the co-repressors N-CoR and SMRT. Oncogene  17: 2473–2484. Google Scholar Iwakawa, H., Ueno, Y., Semiarti, E., Onouchi, H., Kojima, S., Tsukaya, H., Hasebe, M., Soma, T., Ikezaki, M., Machida, C. and Machida, Y. ( 2002) The ASYMMETRIC LEAVES2 gene of Arabidopsis thaliana, required for formation of a symmetric flat leaf lamina, encodes a member of a novel family of proteins characterized by cysteine repeats and a leucine zipper. Plant Cell Physiol.  43: 467–478. Google Scholar Jackson, D. ( 1992) In situ hybridization in plants. In Molecular Plant Pathology: A Practical Approach. Edited by Bowles, D.J., Gurr, S.J. and McPherson, R., pp. 163–174. Oxford University Press, Oxford. Google Scholar Jackson, D., Veit, B. and Hake, S. ( 1994) Expression of maize KNOTTED1 related homeobox genes in the shoot apical meristem predicts patterns of morphogenesis in the vegetative shoot. Development  120: 405–413. Google Scholar Konieczny, A. and Ausubel, F.M. ( 1993) A procedure for mapping Arabidopsis mutations using co-dominant ecotype-specific PCR-based markers. Plant J.  4: 403–410. Google Scholar Lin, R.J., Nagy, L., Inoue, S., Shao, W., Miller, W.H., Jr and Evans, R.M. ( 1998) Role of the histone deacetylase complex in acute promyelocytic leukaemia. Nature  391: 811–814. Google Scholar Lin, W., Shuai, B. and Springer, P.S. ( 2003) The ArabidopsisLATERAL ORGAN BOUNDARIES-domain gene ASYMMETRIC LEAVES2 functions in the repression of KNOX gene expression and adaxial-abaxial patterning. Plant Cell  15: 2241–2252. Google Scholar Lincoln, C., Long, J., Yamaguchi, J., Serikawa, K. and Hake, S. ( 1994) A knotted1-like homeobox gene in Arabidopsis is expressed in the vegetative meristem and dramatically alters leaf morphology when overexpressed in transgenic plants. Plant Cell  6: 1859–1876. Google Scholar Long, J.A., Moan, E.I., Medford, J.I. and Barton, M.K. ( 1996) A member of the KNOTTED class of homeodomain proteins encoded by the STM gene of Arabidopsis. Nature  379: 66–69. Google Scholar Mou, Z., Fan, W. and Dong, X. ( 2003) Inducers of plant systemic acquired resistance regulate NPR1 function through redox changes. Cell  113: 935–944. Google Scholar Ori, N., Eshed, Y., Chuck, G., Bowman, J.L. and Hake, S. ( 2000) Mechanisms that control knox gene expression in the Arabidopsis shoot. Development  127: 5523–5532. Google Scholar Pautot, V., Dockx, J., Hamant, O., Kronenberger, J., Grandjean, O., Jublot, D. and Traas, J. ( 2001) KNAT2: evidence for a link between Knotted-like genes and carpel development. Plant Cell  13: 1719–1734. Google Scholar Pieterse, C.M.J. and Van Loon, L.C. ( 2004) NPR1: the spider in the web of induced resistance signaling pathways. Curr. Opin. Plant Biol.  7: 456–464. Google Scholar Pintard, L., Willis, J.H., Willems, A., Johnson, J.-L.F., Srayko, M., Kurz, T., Glaser, S., Mains, P.E., Tyers, M., Bowerman, B. and Peter, M. ( 2003) The BTB protein MEL-26 is a substrate-specific adaptor of the CUL-3 ubiquitin-ligase. Nature  425: 311–316. Google Scholar Ryals, J.A., Weymann, K., Lawton, K., Friedrich, L., Ellis, D., Steiner, H.Y., Johnson, J., Delaney, T.P., Jesse, T., Vos, P. and Uknes, S. ( 1997) The Arabidopsis NIM1 protein shows homology to the mammalian transcription factor inhibitor IκB. Plant Cell  9: 425–439. Google Scholar Sakai, T., Wada, T., Ishiguro, S. and Okada, K. ( 2000) RPT2: a signal transducer of the phototropic response in Arabidopsis. Plant Cell  12: 225–236. Google Scholar Sedgwick, S.G. and Smerdon, S.J. ( 1999) The ankyrin repeat: A diversity of interactions on a common structural framework. Trends Biochem. Sci.  24: 311–316. Google Scholar Semiarti, E., Ueno, Y., Tsukaya, H., Iwakawa, H., Machida, C. and Machida, Y. ( 2001) The ASYMMETRIC LEAVES2 gene of Arabidopsis thaliana regulates formation of a symmetric lamina, establishment of venation and repression of meristem-related homeobox genes in leaves. Development  128: 1771–1783. Google Scholar Shuai, B., Reynaga-Pena, C.G. and Springer, P.S. ( 2002) The LATERAL ORGAN BOUNDARIES gene defines a novel, plant-specific gene family. Plant Physiol.  129: 747–761. Google Scholar Steeves, T.A. and Sussex, I.M. ( 1989) Patterns in Plant Development. Cambridge University Press, New York. Google Scholar Thompson, J.D., Higgins, D.G. and Gibson, T.J. ( 1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res.  22: 4673–4680. Google Scholar Vanacker, H., Lu, H., Rate, D.N. and Greenberg, J.T. ( 2001) A role for salicylic acid and NPR1 in regulating cell growth in Arabidopsis. Plant J.  28: 209–216. Google Scholar Xu, L., Wei, Y., Reboul, J., Vaglio, P., Shin, T.H., Vidal, M., Elledge, S.J. and Harper, J.W. ( 2003) BTB proteins are substrate-specific adaptors in an SCF-like modular ubiquitin ligase containing CUL-3. Nature  425: 316–321. Google Scholar Xu, L., Xu, Y., Dong, A., Sun, Y., Pi, L., Xu, Y. and Huang, H. ( 2003) Novel as1 and as2 defects in leaf adaxial-abaxial polarity reveal the requirement for ASYMMETRIC LEAVES1 and 2 and ERECTA functions in specifying leaf adaxial identity. Development  130: 4097–4107. Google Scholar Zhang, Y., Fan, W., Kinkema, M., Li, X. and Dong, X. ( 1999) Interaction of NPR1 with basic leucine zipper protein transcription factors that bind sequences required for salicylic acid induction of the PR-1 gene. Proc. Natl Acad. Sci. USA  96: 6523–6528. Google Scholar Zollman, S., Godt, D., Prive, G.G., Couderc, J.-L. and Laski, F.A. ( 1994) The BTB domain, found primarily in zinc finger proteins, defines an evolutionary conserved family that includes several developmentally regulated genes in Drosophila. Proc. Natl Acad. Sci. USA  91: 10717–10721. Google Scholar

Karita, Eri;Yamakawa, Hiromoto;Mitsuhara, Ichiro;Kuchitsu, Kazuyuki;Ohashi, Yuko

doi: 10.1093/pcp/pch158pmid: 15564520

Abstract We previously reported that three types of tobacco calmodulin (CaM) isoforms originated from 13 genes are differently regulated at the transcript and protein levels in response to wounding and tobacco mosaic virus-induced hypersensitive reaction (HR); wound-inducible type I and HR-inducible type III levels increased after wounding and HR, respectively, while type II, whose expression is constitutive and wound responsible, remained unchanged. Here, we show that these CaMs differentially activate target enzymes; rat NO synthase was activated most effectively by type III, moderately by type I and weakly by type II, and plant NAD kinase (NADK) was activated in the inverse order. Furthermore, we found that a suitable Ca2+ concentration differs by type; type II activated NADK at lower Ca2+ of around 0.1 µM, which is the cytosolic concentration in unstimulated cells, type I did so at 1–5 µM, which is the increased Ca2+ concentration in stimulated cells, while type III did not at any Ca2+ level. NADK activation was highest over a pH range of 7.1–6.8 for which the cytosolic pH reportedly changed from 7.5 after being stimulated. Thus, tobacco CaMs, especially type I, effectively activate NADK in stimuli-induced conditions. (Received February 20, 2004; Accepted July 14, 2004) Introduction We studied the resistance mechanism active against wounding and pathogen infection in plants using wounded and tobacco mosaic virus (TMV)-infected tobacco leaves containing the N resistance gene (Nicotiana tabacum L. cv Samsun NN). N gene-dependent formation of necrotic lesions in virus-infected tissue is a kind of programmed cell death responsible for virus enclosure, and is a hypersensitive reaction (HR) that occurs during the plant’s defense response (Goodman and Novacky 1994). It induces both salicylic acid-dependent signaling and jasmonic acid/ethylene-dependent signaling, which mimic wound signaling (Niki et al. 1998). As a HR-responsive gene, we isolated the calmodulin gene (CaM), which encodes a major Ca2+ receptor that activates target enzymes in the presence of Ca2+ (Klee and Vanaman 1982). Although the nature of the downstream signaling of CaM in plant cells is largely unknown, plants have multiple CaM isoforms, which are highly divergent, in contrast to the invariant mammalian CaM (Lee et al. 1995, Takezawa et al. 1995). Therefore, it is expected that multiple plant-specific CaM-mediated signaling pathways exist. Furthermore, a number of studies indicated that the production of reactive oxygen species (ROS) such as O2–, H2O2 and nitric oxide (NO) plays an important role in the defense response, and is controlled in a Ca2+-dependent manner (Nürnberger et al. 1994, Hahlbrock et al. 1995, Delledonne et al. 1998, Park et al. 1998, Grant et al. 2000). On the other hand, change in the concentration of cytosolic free Ca2+ ([Ca2+]cyt) is evoked by various extracellular stimuli including pathogenic infection or elicitors (Sanders et al. 1999, Kadota et al. 2004). Increased [Ca2+]cyt is thought to be required to induce an array of defense responses including ROS production and gene expression via various Ca2+ receptors (Reddy 2001). To determine how Ca2+ triggers such responses, we successfully isolated 13 CaM genes, NtCaM1-13, from TMV-infected or wounded tobacco belonging to three plant-specific types (Yamakawa et al. 2001; Fig. 1A). Among them, type I isoforms encoded by NtCaM1 and 2 showed overall similarities to PCM1, a potato CaM (Takezawa et al. 1995), while type II isoforms, which contain NtCaM3/4/5/6/7/8/9/10/11/12, are highly homologous to soybean SCaM-1. NtCaM13 belongs to type III, and has the most diverse substitutions, which are common to SCaM-4 (Lee et al. 1995). In soybean plant, SCaM-1 and SCaM-4 were characterized by different spectra on activation of target enzyme (Lee et al. 1995, Cho et al. 1998), suggesting different functions. Among the three types of plant CaMs, there is little biochemical information on type I CaM except from a gene expression study on potato PCM1 (Takezawa et al. 1995). To characterize the precise functions of each type of CaM, a comparative study was necessary using all CaMs from the same plant source under the same conditions. In this paper, we compared the biochemical nature of type I, II and III CaMs, using purified representative CaMs, NtCaM1, 3 and 13 (Fig. 1B), respectively. For a target enzyme, we selected plant NAD kinase (NADK), which is suggested to function in the self-defense of plants by providing the cofactor NADP for ROS production by NADPH oxidase (Harding et al. 1997). We found that the enzyme was specifically activated by wound-inducible types I and II, whose ratio of total CaMs increased, but not by type III, whose ratio decreased after wounding (Yamakawa et al. 2001). Furthermore, the activation by types I and II was maximized under elevated Ca2+ concentrations and decreased pH, which are induced after exposure to various stimuli (Sanders et al. 1999, Lebrun-Garcia et al. 2002, Lecourieux et al. 2002). The activity of animal NO synthase (NOS), which is also activated by plant CaMs in vitro, was highest, moderate and lowest by type III, type I and type II, respectively, indicating the substrate preference of the CaM types. Results NADK was activated by NtCaM1 and NtCaM3 but not by NtCaM13 Plant NADK is activated by plant CaMs (Anderson and Cormier 1978, Roberts and Harmon 1992). This study had been conducted using only type II and III CaMs, lacking data on type I CaMs whose structure and expression profiles are different from those of types II and III (Takezawa et al. 1995, Yamakawa et al. 2001). To understand the role of each CaM type in NADK activation, we used representatives of all three plant-specific CaMs (NtCaM1, 3 and 13) and a control animal CaM (bovine brain CaM). Using purified pea NADK as the target enzyme, we assessed which type of CaM is the most potent activator under conditions of 1 mM Ca2+ and pH 8.0. NADK activity was evaluated by quantification of the product, NADP+, via reaction with excess NADP+-dependent glucose 6-phosphate dehydrogenase and oxidation–reduction indicator dyes as described previously (Harmon et al. 1984). Pea NADK was activated most by NtCaM3 and then by NtCaM1 in the presence of 1 mM Ca2+ at pH 8.0, but NtCaM13 hardly activated the enzyme (Fig. 2A). Mammalian CaM had less activity than NtCaM1 and 3. In the absence of Ca2+, no activity was detected for any of the CaMs (data not shown). Vmax, the maximal activity compared with that of the mammalian CaM, and Kact, the concentration of CaM required for half-maximal activity, were 207%, 235% and 9%, and 16, 9 and 26 nM for NtCaM1, NtCaM3 and NtCaM13, respectively (Table 1). NOS was effectively activated by NtCaM13 and then by NtCaM1, but weakly by NtCaM3 NOS is an important CaM-dependent enzyme in animals. We studied the properties of tobacco CaMs using a rat recombinant NOS as a target enzyme according to the citrulline method coupled with thin-layer chromatography (TLC) separation (Kumar et al. 1999). NOS was activated by all tobacco CaM types as well as by mammalian CaM (Fig. 2B). Compared with the control animal CaM, NtCaM13 was the most potent activator, NtCaM1 was the second most potent, while NtCaM3 was the weakest. Without Ca2+, no NOS activity was observed (data not shown). Vmax compared with that for bovine CaM and Kact for NtCaM1, NtCaM3 and NtCaM13 were 98%, 67% and 119%, and 56, 89 and 67 nM, respectively (Table 1). Calcineurin was most activated by NtCaM3 and moderately activated by NtCaM1 and NtCaM13 Calcineurin (CaN), a CaM-dependent protein phosphatase, is a target of CaM in mammals. The activity of bovine brain CaN was determined by a fluorometric method using 4-methyl umbelliferyl phosphate (4MUP) as the substrate (Anthony et al. 1986). In the absence of Ca2+, the enzyme exhibited 35–45% of the activity in the presence of Ca2+ (data not shown). With Ca2+, NtCaM3 increased the activity of bovine CaN 2.7-fold, followed second by bovine CaM, and followed by NtCaM1 and NtCaM13 (Fig. 2C). Vmax compared with that for bovine CaM and Kact for NtCaM1, NtCaM3 and NtCaM13 were 81%, 130% and 74% and 4.6, 4.0 and 7.5 nM, respectively (Table 1). The Ca2+ concentration required for NADK activation is different for different types of CaMs NADK activation as shown in Fig. 2A was analyzed under similar conditions as described previously (Harmon et al. 1984, Lee et al. 1995); at 1 mM Ca2+ and pH 8.0, which is different from real cytosolic conditions found in plant cells. The cytosolic Ca2+ concentration ([Ca2+]cyt) is around 0.1 µM in unstimulated cells, but dynamically increases as a result of stimuli such as red light, touch, heat shock, oxidative stress, hormonal signals. For instance, [Ca2+]cyt transiently rises to 3 µM after elicitor treatment in suspensison-cultured cells (Sanders et al. 1999, Lecourieux et al. 2002). To understand the Ca2+-dependent activation of NADK by each CaM type in intact plant cells after stimulation, we used the three representative tobacco CaM proteins at various Ca2+ concentrations. Because the cytosolic pH in unstimulated plant cells is around 7.5 (Kurkdjian and Guern 1989), the pH of the reaction mixture for this study was adjusted to 7.5. CaM concentration in this study was fixed at 0.5 µM. Although there are few reports on cytosolic CaM concentration in plant, it is said to be 5–20 µM (reviewed by Zielinski 1998). While 0.5 µM CaM is lower than the known concentration, it seemed to be sufficient because NADK activation was saturated at 0.5 µM CaM in Fig. 2A. Under this condition, NtCaM3 could activate NADK at a Ca2+ concentration lower than 1 µM (Fig. 3A). However, NtCaM1 required at least 10 µM Ca2+, and its activating ability reached to the same or a higher degree than that of NtCaM3 at 100 µM Ca2+. NtCaM13 could not activate NADK at any Ca2+ concentration. Ca2+-dependent NADK activation is regulated by pH Next, we examined the effect of pH on NADK activation by CaMs because cytosolic pH is changed by stress such as elicitor treatment. In N-acetylchitooligosaccharide elicitor-treated suspension-cultured rice cells, cytosolic pH changed rapidly from 7.5 to 7.1 (Kuchitsu et al. 1997), and the pH declined to 6.8 from 7.3, 40 min after treatment with 0.3 M NaCl in suspension-cultured tobacco cells (Qiao et al. 2002). Thus it is possible that cytosolic pH decreases after stimuli in vivo. To reproduce the actual cytosolic conditions present after stimuli in plants, we used reaction mixtures at pH 7.1, 6.8 and 6.4 with various Ca2+ concentrations for the analysis of NADK activation by CaMs. Under pH 7.1, NtCaM1 and 3 activated NADK more efficiently than under pH 7.5 (Fig. 3B). NtCaM3 activated it under pH 7.1 at Ca2+ concentrations such as 0.1 µM, and NtCaM1 responded to 1 µM Ca2+, at which NADK activation was not detected under pH 7.5. At pH 6.8, NtCaM3 activation was more efficient than at pH 7.1. NtCaM1 activated NADK similarly to pH 7.1 for almost the same Ca2+ concentrations (Fig. 3C). Under pH 6.4, for which the cytosolic pH was changed from 7.3 after NaCl treatment in suspension-cultured tobacco cells to induce cell death (Qiao et al. 2002), the activating ability of both types of CaMs declined (Fig. 3D). Discussion In contrast to the mammalian system in which only one CaM protein operates, plants have multiple types of CaMs with different protein structures to transduce Ca2+ signals downstream. Using purified NtCaM1, 3 and 13, which belong to plant-specific type I, II and III CaMs, respectively, we analyzed the roles of each type of CaM in target enzyme activation in vitro. Our work presents comparable studies of all three types of plant CaMs and the control animal CaM under experimental conditions mimicking that in the cytosol of healthy unstimulated and stimulated cells. The characterization of type I CaM using NtCaM1 as a representative in comparison with other types of CaMs was first reported in this work, which found that NtCaM1 activates NADK only under stress-induced conditions. From this study, we could understand that the mode of target enzyme activation is different and specific for each type of CaM. Studies on plant NADK activation by the three types of CaMs under various Ca2+ concentrations and pHs gave us important information on the specificity of CaM. Increased Ca2+ concentration after stimuli enhanced NADK activation by NtCaM1 and 3, but no activation by NtCaM13 occurred. Similarly, lowered pH after stimuli also enhanced activation by the NtCaM1 and 3, but no activation by NtCaM13 occurred. In contrast, NtCaM13 most effectively activated another target enzyme, NOS. These results indicate that plants efficiently use three types of CaMs to activate different target enzymes adapted for the circumstances induced by various stimuli. [Ca2+]cyt in plant cells at steady state is around 0.1 µM (Bush 1995). After stress treatment such as elicitor application (Lecourieux et al. 2002) and pathogen infection (Xu and Heath 1998), [Ca2+]cyt increases, triggering stress-responsive signaling pathways used for self-defense. Cytosolic pH is also dynamically altered to bring about acidification after stress treatment (Kurkdjian and Guern 1989). This evidence suggests that Ca2+-induced target enzyme activation in vitro should be studied under similar conditions in unstimulated and stimulated plant cells in vivo to obtain meaningful data. We examined Ca2+/CaM-dependent NADK activation under conditions mimicking the cytosolic conditions of plant cells before and after stimuli. We found that plant NADK was more effectively activated by type I and II NtCaMs under pH 7.1 and 6.8, which are stress-inducible cytosolic pHs, than at pH 7.5, which is the cytosolic pH in unstimulated plant cells. A report by Yamamoto (1966) showed that in crude green leaf extract, NADK activity was highest at pH 6.8 when reacted in buffer at pH 5.2–8.4, which supports our findings. Furthermore, we found that NtCaM3 could activate NADK at Ca2+ concentrations between 0.1 and 1 µM at all pH values tested. Since the [Ca2+]cyt in plant cells at steady state is around 0.1 µM (Bush 1995), NADK activation by type II CaMs would occur constitutively and is enhanced by increased Ca2+ after stimuli, indicating that type II CaMs respond most sensitively to small stimuli. In contrast to type II CaMs, wound-inducible type I CaMs such as NtCaM1 did not respond to Ca2+ concentration lower than 1 µM but did to Ca2+ higher than 10 µM at pH 7.5, implying that type I CaMs do not function in NADK activation in healthy unstimulated cells. Since NtCaM1 responded to low Ca2+ concentrations such as 1 µM at pH 7.1 and 6.8 but not at pH 7.5, it appears that type I CaMs are recruited only in stimulated cells in which [Ca2+]cyt has increased and pH has declined (Fig. 3E). Many reports indicated the involvement of CaMs in plant self-defense responses (Heo et al. 1999, Kim et al. 2002). A transgenic tobacco plant that possesses a mutated CaM, which contains one amino acid substitution crucial for the hyperactivation of NADK, showed enhanced H2O2 production in response to elicitor treatment (Harding et al. 1997). Since ROS such as H2O2 are produced by NADPH oxidase activity, these CaMs likely promote the production of its substrate NADP+/NADPH by activating NADK to convert NAD+ to NADP+. In suspension-cultured tobacco cells treated with cryptogein, a phytopathogenic fungus-derived proteinous elicitor, the pentose phosphate pathway is activated, which enhances NADP+/NADPH conversion (Pugin et al. 1997). The production of O2– by tobacco homologs of NADPH oxidase is stimulated by TMV infection and Ca2+ application (Sagi and Fluhr 2001). At low [Ca2+]cyt in unstimulated cells, only type II CaMs such as NtCaM3 would activate NADK, producing a basal level of NADP(H) almost all of which is consumed by reducing reactions such as photosynthesis in plant cells. When [Ca2+]cyt increases and pH decreases due to a stimulus, NtCaM1 and NtCaM3 may activate NADK more effectively in elevating the NADP(H) level, making available excess NADP(H) for the NADPH oxidase reaction to produce ROS. Thus, type I CaMs including NtCaM1 may trigger Ca2+ signaling for ROS production after stimulation such as by wounding or elicitor treatment (Fig. 4). Generally, plant CaMs have four sets of well-conserved EF-hand motifs for Ca2+ binding. The reason for three types of tobacco CaMs with different abilities to activate NADK at different Ca2+ concentration is a focus of interest. As for plant NADK activation by CaM, it was reported that Lys-30 and Gly-40 in the first EF-hand of soybean SCaM-1, which is the ortholog of NtCaM3, are essential for NADK activation but not for binding to the enzyme. Soybean SCaM-4, which is the ortholog of NtCaM13 and has Glu-30 and Asp-40, does not activate NADK but can bind to the enzyme (Lee et al. 1997). Consistent with this, both NtCaM1 and 3 have Lys-30 and Gly-40 in the first EF-hand, whereas NtCaM13 has Glu-30 and Asp-40(Fig. 1A). This evidence indicates that the difference in these amino acids is not involved in the difference in Ca2+ response between NtCaM1 and NtCaM3. Liao et al. (1996) suggested that the C-terminal hydrophobic region of CaM plays a role in binding to several target proteins including NADK. Since NtCaM1 has a different amino acid sequence from that of NtCaM3 in the C-terminal region rather than around EF-hands (Fig. 1A), different Ca2+ responses for NADK activation may be attributed to the predicted difference in binding affinity between Ca2+/CaM and NADK rather than the affinity between Ca2+ and CaM. In contrast to plant NADK, mammalian NOS was most activated by NtCaM13 (Fig. 2C). Ca2+-dependent NOS-like activity is necessary for the expression of pathogenesis-related (PR) genes following TMV infection in tobacco plants (Durner et al. 1998). We confirmed the results of Heo et al. (1999) for the constitutive expression of PR genes in transgenic tobacco plants overproducing NtCaM13/SCaM-4-type CaMs (unpublished data). Thus, type III CaMs might regulate PR gene expression involved in the defense response against pathogen infection by an unknown mechanism. Recently, Chandok et al. (2003) indicated that a variant of the P protein of glycine decarboxylase had NOS activity, and that it is induced by viral infection. Another study revealed that an Arabidopsis NOS gene encodes a protein with sequence similarity to a NO synthesis-related protein in the snail Helix pomatia (Guo et al. 2003). The evaluations of plant NOS enzymes reported so far and the isolation of new putative plant NOS molecules as CaM binding proteins will help in the elucidation of Ca2+/CaM signaling during self-defense reactions in plants. At present, genes of the target enzymes of tobacco CaMs in self-defense signaling to pathogen infection and wounding have not been identified. One of the candidates is a putative tobacco MAPK phosphatase (NtMKP1) which we isolated as a CaM binding protein (Yamakawa et al. 2004). Wound-induced activation of defense-related MAPKs such as WIPK and SIPK was significantly inhibited in transgenic tobacco plants overproducing NtMKP1, indicating that the MAPK phosphatase is involved in MAPK signaling via a Ca2+/CaM system after wounding or pathogen infection. NtMKP1 has higher binding affinity to NtCaM1 and NtCaM3 than to NtCaM13. An amino acid substitution in the CaM binding domain of NtMKP1 abolished NtCaM1 and NtCaM3 binding. The structure of NtMKP1 is considerably different from that of animal MKPs, suggesting that plants have specific Ca2+/CaM signaling cascades via the MKP/MAPK system which are quite different from those of animals. Plants lack the highly developed immune system that vertebrates have, and vertebrates have only one CaM protein with an identical amino acid sequence. Our data suggest that transcriptional and post-transcriptional control of diverse plant CaM isoforms with a characteristic manner may effectively activate individual target enzymes and transduce the Ca2+ signal downstream via dynamic changes in cytosolic Ca2+ concentration and pH upon internal and external stimulation. The isolation and characterization of other plant-originated target molecules of individual CaMs by direct analysis of CaM-interacting proteins will shed light on the physiological importance of individual CaM isoforms for the self-defense mechanism in plants. Materials and Methods Purification of CaM protein Recombinant CaM proteins NtCaM1, 3 and 13 were produced by Escherichia coli harboring pET-NtCaM1, 3 and 13, respectively, and purified by Ca2+-dependent hydrophobic chromatography as described previously (Fromm and Chua 1992, Yamakawa et al. 2001). As shown in Fig. 1B, the purity of CaM proteins that showed an electrophoretic mobility shift in the presence of Ca2+ was >95% as seen by Coomassie Brilliant Blue (CBB) staining following separation on SDS–polyacrylamide gel. Bovine brain CaM was obtained from Wako (Osaka, Japan), and used as a control for the enzyme assays. Protein concentration was determined by both the Bradford method (Bradford 1976) and the Lowry method (Lowry et al. 1951) using a protein assay kit (Bio-Rad, Hercules, CA, U.S.A.) with BSA as the standard. Purification of NADK Pea NADK was partially purified from pea seedlings (Pisum sativam L. cv. Akabana-tsurunashi-endo) by the method of Muto and Miyachi (1977). Peas were germinated in moistened vermiculite and grown under artificial light without additional nutrients. The following purification procedures were carried out at 4°C. The aerial parts of 12-day-old seedlings (627 g) were macerated with liquid nitrogen and extracted with 3 vol of 25 mM triethanolamine-acetate (pH 7.5) containing 1 mM phenylmethylsulfonyl fluoride, 0.5 M sucrose and 1 mM DTT. The homogenate was squeezed through double-layered Miracloth (Calbiochem, La Jolla, CA, U.S.A.) and centrifuged at 27,000×g for 30 min. A one-tenth volume of 0.7% protamine sulfate solution in 10 mM triethanolamine-acetate buffer (pH 7.5) was added to the supernatant (2,000 ml). After continuous stirring for 15 min, the precipitate was collected by centrifugation at 27,000×g for 15 min. From the precipitate, the enzyme was extracted with 250 ml of 0.2 M Na-acetate buffer (pH 6.0) and 1 µg ml–1 pepstatin A. To the extract, the same volume of 50% (w/w) polyethylene glycol 6,000 solution was added and the mixture was stirred for 30 min. After centrifugation at 39,000×g for 30 min, the precipitate was resuspended in 100 ml of 50 mM Tris–HCl (pH 7.0), 100 mM KCl, 3 mM MgCl2 and 1 mM EGTA. Insoluble materials were removed by centrifugation at 27,000×g for 15 min. The supernatant solution was passed through a DEAE–Sephacel (Amersham Pharmacia Biotech, Buckinghamshire, UK) column (1.6×30 cm) pre-equilibrated with the buffer mentioned above, allowing pea endogenous CaM to be completely adsorbed by the column. The effluent was stocked at –80°C in small portions of 5% glycerol solution and used for NADK assays. NADK assay The NADK assay was performed as described previously (Harmon et al. 1984) in a 0.5-ml solution A containing 50 mM Tricine (pH 8.0), 5 mM MgCl2, 2 mM NAD+, 3 mM ATP, 1 mM CaCl2 or EGTA and various amounts of CaM. The reaction was initiated with 10 µl of freshly thawed, purified NADK stock solution. After incubation for 60 min at 37°C, the reaction was terminated by placing the tubes in boiling water for 3 min. The tubes were then cooled to ambient temperature, and 0.5 ml of solution B containing 50 mM Tricine (pH 8.0), 5 mM MgCl2, 1 mM EGTA, 0.8 mM glucose 6-phosphate, 0.1 mg ml–1 phenazine methosulfate, and 0.15 mg ml–1 2,6-dichlorophenolindophenol was added. The mixture was transferred to a cuvette pre-incubated at 30°C, and 20 µl of glucose 6-phosphate dehydrogenase (6 U ml–1) was added. The decrease in A600 per minute was monitored using a Beckman spectrophotometer (Model DU-7400; Fullerton, CA, U.S.A.) equipped with a temperature controller set at 30°C. The amount of NADP+ produced by the NADK reaction was calculated from a standard curve of NADP+ versus the descending rate of A600. No activation by Ca2+ was found in the absence of exogenous CaM, confirming that the preparation of NADK was free from contamination by pea endogenous CaM (data not shown). To analyze the effect of Ca2+ concentration and pH, solutions of CaM and NADK were dialyzed respectively against 50 mM Tris–HCl (pH 7.5), 3mM MgCl2. Solution C, which contained various amounts of CaCl2 buffered by BAPTA, 50 mM PIPES (pH7.5, 7.1, 6.8, 6.4), 5 mM MgCl2, 2 mM NAD+, 3 mM ATP, 0.5 µM CaM was used in place of solution A. Since concentration of free Ca2+ is depending on pH or other components such as Mg2+ and ATP, we previously used calculation software (BOUND AND DETERMINED; Brooks and Storey 1992) for determination of CaCl2 and BAPTA concentrations in solution C. NOS assay NOS activity was determined by the citrulline assay followed by TLC, as described by Kumar et al. (1999). The freshly prepared reaction mixture (20 µl) consisted of 30 mM HEPES–NaOH (pH 7.0), 2 mM NADPH, 100 µM FAD, 100 µM tetrahydrobiopterin, 1 mM CaCl2 or EGTA, various concentrations of CaM, 0.25 µl of l-[U-14C]arginine (272 mCi mmol–1, 100 µCi ml–1; Moravek Biochemicals, Brea, CA, U.S.A.) and 100 mU of recombinant rat neuronal NOS (Calbiochem). The reaction was carried out at 30°C for 60 min. The NOS reaction was terminated by adding 50 µl of cold methanol. The samples were left on ice for 20 min and centrifuged at 20,000×g for 10 min. An aliquot (10 µl) of the supernatant was spotted onto a silica gel TLC plate (Merck, Darmstadt, Germany), air-dried and subjected to chromatography. The solvent was ammonium hydroxide:chloroform:methanol:water (4 : 1 : 9 : 2). The plate was imaged by a PhosphorImager SI (Amersham Pharmacia Biotech) after exposure for 48 h and the radioactivity of the product, l-[14C]citrulline, was quantified using the ImageQuant 1.1 program (Amersham Pharmacia Biotech). The Rf values for l-arginine and l-citrulline were 0.44 and 0.90, which were those of the standard amino acids stained with ninhydrin (data not shown). Calcineurin assay The activity of a CaM-dependent protein phosphatase, CaN, was determined by a fluorescent assay using 4-MUP (ICN, Costa Mesa, CA, U.S.A.) as the substrate, as reported previously (Anthony et al. 1986). Each assay was conducted in 200 µl of 50 mM Tris–HCl (pH 8.0), 1 mg ml–1 BSA, 0.5 mM DTT, 1 mM MgCl2, 0.3 mM CaCl2 or EGTA, 12.5 nM bovine brain CaN (Upstate Biotechnology, Lake Placid, NY, U.S.A.), with various concentrations of CaM, and 200 µM 4MUP, which was added to start the reaction. After incubation at 37°C for 60 min, the reaction was terminated by the addition of 1 ml of 0.4 M Na2CO3. Fluorescence was monitored in a quartz cuvette (1 cm light path) using a Hitachi fluorescence spectrophotometer (Model F-2500; Tokyo, Japan) with an excitation wavelength of 365 nm and an emission wavelength of 446 nm. To correlate the amount of product, 4-methyl umbelliferone (4MU), with the amount of fluorescence, a standard curve was made by monitoring the fluorescence intensity as a function of the concentration of 4MU. All assays were corrected for the non-enzymic hydrolysis of 4MUP. Specific activity was defined as nmol 4 MU mg–1 min–1. Acknowledgments We thank Drs S. Seo, K. Higashi, S. Katou and Ms Y. Gotoh for helpful advice regarding procedures. This work was supported in part by grants from the COE (Center of Excellence) project and a Grant-in-aid for scientific research on the Molecular Mechanisms of Plant–Pathogenic Microbe Interaction,from the Ministry of Education, Science and Culture, Japan. 5 These two authors contributed equally to this work. 6 Present address: National Agricultural Research Center, Joetsu, Nigata, 943-0193 Japan 7 Corresponding author: E-mail, [email protected]; Fax, +81-298-38-7469. View large Download slide View large Download slide Fig. 1 The characteristics of three types of NtCaMs. (A) Comparison of three plant-specific CaMs and vertebrate CaM amino acid sequences. Dots indicate amino acids identical to those of NtCaM1. Shadows behind the letters indicate four conserved Ca2+ binding motifs, EF-hands. (B) Purified NtCaM proteins prepared as described previously (Fromm and Chua 1992). Four micrograms each of recombinant NtCaM1, NtCaM3 and NtCaM13 protein were subjected to 12% SDS–PAGE in the presence of 7.5 mM CaCl2 (+) or 7.5 mM EGTA (–), and protein in the gel was stained with Coomassie Brilliant Blue R-250. The Ca2+/CaM complex migrated more quickly than CaM alone. Molecular weight marks (M) are indicated on the right. View large Download slide View large Download slide Fig. 1 The characteristics of three types of NtCaMs. (A) Comparison of three plant-specific CaMs and vertebrate CaM amino acid sequences. Dots indicate amino acids identical to those of NtCaM1. Shadows behind the letters indicate four conserved Ca2+ binding motifs, EF-hands. (B) Purified NtCaM proteins prepared as described previously (Fromm and Chua 1992). Four micrograms each of recombinant NtCaM1, NtCaM3 and NtCaM13 protein were subjected to 12% SDS–PAGE in the presence of 7.5 mM CaCl2 (+) or 7.5 mM EGTA (–), and protein in the gel was stained with Coomassie Brilliant Blue R-250. The Ca2+/CaM complex migrated more quickly than CaM alone. Molecular weight marks (M) are indicated on the right. View large Download slide View large Download slide Fig. 2 CaM concentration-dependent activation of three target enzymes. (A) NADK activity was determined by the procedure of Harmon et al. (1984) using NADK purified from pea seedlings with 1 mM Ca2+ and increasing NtCaM1 (filled circles), NtCaM3 (open squares), NtCaM13 (open circles) and bovine CaM (filled triangles), respectively, under pH 8.0. One unit of the activity is defined as the conversion of 1 nmol of NAD+ to NADP+ per 1 min. (B) NOS activity was determined by the citrulline method. Recombinant rat neuronal NOS was reacted with 1 mM Ca2+ and increasing individual CaMs. The activity is shown as a % relative to that for 10–5 M bovine CaM. (C) CaN activity was determined by the procedure of Anthony et al. (1986) using bovine brain CaN with 0.3 mM Ca2+ and increasing individual CaMs. The activity is shown relative to basal activity in the absence of any CaM. Values represent the means of three independent experiments, and error bars indicate ±SD. View large Download slide View large Download slide Fig. 2 CaM concentration-dependent activation of three target enzymes. (A) NADK activity was determined by the procedure of Harmon et al. (1984) using NADK purified from pea seedlings with 1 mM Ca2+ and increasing NtCaM1 (filled circles), NtCaM3 (open squares), NtCaM13 (open circles) and bovine CaM (filled triangles), respectively, under pH 8.0. One unit of the activity is defined as the conversion of 1 nmol of NAD+ to NADP+ per 1 min. (B) NOS activity was determined by the citrulline method. Recombinant rat neuronal NOS was reacted with 1 mM Ca2+ and increasing individual CaMs. The activity is shown as a % relative to that for 10–5 M bovine CaM. (C) CaN activity was determined by the procedure of Anthony et al. (1986) using bovine brain CaN with 0.3 mM Ca2+ and increasing individual CaMs. The activity is shown relative to basal activity in the absence of any CaM. Values represent the means of three independent experiments, and error bars indicate ±SD. View large Download slide View large Download slide Fig. 3 Effects of Ca2+ concentration and pH on NADK activation by NtCaM1, 3 and 13. The assay was conducted with 0.5 µM NtCaM1 (filled circles), NtCaM3 (open squares) and NtCaM13 (open circles), respectively, and increasing Ca2+ under pH 7.5 (A), pH 7.1 (B), pH 6.8 (C) or pH 6.4 (D). One unit of the NADK activity is the same as in Fig. 2. The data are the mean of at least two experiments ±SD. (E) Illustration of NADK activation by NtCaM1 (type I) and NtCaM3 (type II). The activity was enhanced under increased Ca2+ concentration and decreased pH. Crescendo bars indicate the level of NADK activity. View large Download slide View large Download slide Fig. 3 Effects of Ca2+ concentration and pH on NADK activation by NtCaM1, 3 and 13. The assay was conducted with 0.5 µM NtCaM1 (filled circles), NtCaM3 (open squares) and NtCaM13 (open circles), respectively, and increasing Ca2+ under pH 7.5 (A), pH 7.1 (B), pH 6.8 (C) or pH 6.4 (D). One unit of the NADK activity is the same as in Fig. 2. The data are the mean of at least two experiments ±SD. (E) Illustration of NADK activation by NtCaM1 (type I) and NtCaM3 (type II). The activity was enhanced under increased Ca2+ concentration and decreased pH. Crescendo bars indicate the level of NADK activity. View largeDownload slide Fig. 4 Hypothetical model of NADK activation in plant cells. Left, an unstimulated healthy cell in which only NtCaM3 activates NADK conferring NADP(H)-dependent metabolic pathway. Right, a stimulated cell, in which NtCaM3 is further activated and NtCaM1 is newly potentiated for NADK activation by elevated Ca2+ concentration and decreased pH, likely enhancing the defense response through NADPH oxidase. View largeDownload slide Fig. 4 Hypothetical model of NADK activation in plant cells. Left, an unstimulated healthy cell in which only NtCaM3 activates NADK conferring NADP(H)-dependent metabolic pathway. Right, a stimulated cell, in which NtCaM3 is further activated and NtCaM1 is newly potentiated for NADK activation by elevated Ca2+ concentration and decreased pH, likely enhancing the defense response through NADPH oxidase. Table 1 Activation profiles of NADK, NOS and CaN by NtCaM1, 3, 13 and bovine CaM   NADK    NOS    CaN  Target enzyme  Vmax (%)      Vmax (%)      Vmax (%)    NtCaM1  207  ++++    98  ++    81  +  NtCaM3  235  +++++    67  +    130  +++  NtCaM13  9  ±    119  +++    74  +  Bovine CaM  100  ++    100  ++    100  ++    NADK    NOS    CaN  Target enzyme  Vmax (%)      Vmax (%)      Vmax (%)    NtCaM1  207  ++++    98  ++    81  +  NtCaM3  235  +++++    67  +    130  +++  NtCaM13  9  ±    119  +++    74  +  Bovine CaM  100  ++    100  ++    100  ++  The extent of activation was summarized. The number of (+) indicates the extent of activation and (±) denotes little activation. Vmax, the maximal activity compared with that of bovine CaM. View Large Abbreviations CaM calmodulin [Ca2+]cyt cytosolic Ca2+ concentration CaN calcineurin HR hypersensitive reaction NADK NAD+ kinase NO nitric oxide NOS nitric oxide synthase PR pathogenesis-related ROS reactive oxygen species TMV tobacco mosaic virus. References Anderson, J.M. and Cormier, M.J. ( 1978) Calcium-dependent regulator of NAD kinase in higher plants. Biochem. Biophys. Res. Commun.  84: 595–602. Google Scholar Anthony, F.A., Merat, D.L. and Cheung, W.Y. ( 1986) A spectrofluometric assay of calmodulin-dependent protein phosphatase using 4-methylumbelliferyl phosphate. Anal. Biochem.  155: 103–107. Google Scholar Bradford, M.M. ( 1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal. Biochem.  72: 248–254. Google Scholar Brooks, S.P.J. and Storey, K.B. ( 1992) Bound and determined: a computer program for making buffers of defined ion concentrations. Anal. Biochem.  201: 119–126. Google Scholar Bush, D.S. ( 1995) Calcium regulation in plant cells and its role in signaling. Annu. Rev. Plant Mol. Biol.  46: 95–122. Google Scholar Chandok, M.R., Ytterberg, A.J., Van Wijk, K.J. and Klessig, D.F. ( 2003) The pathogen-inducible nitric oxide synthase (iNOS) in plants is a variant of the P protein of the glycine decarboxylase complex. Cell  113: 469–482. Google Scholar Cho, M.J., Vaghy, P.L., Kondo, R., Lee, S.H., Davis, J.P., Rehl, R., Heo, W.D. and Johnson, J.D. ( 1998) Reciprocal regulation of mammalian nitric oxide synthase and calcineurin by plant calmodulin isoforms. Biochemistry  37: 15593–15597. Google Scholar Delledonne, M., Xia, Y., Dixon, R.A. and Lamb, C. ( 1998) Nitric oxide functions as a signal in plant disease resistance. Nature  394: 585–588. Google Scholar Durner, J., Wendehenne, D. and Klessig, D.F. ( 1998) Defense gene induction in tobacco by nitric oxide, cyclic GMP, and cyclic ADP-ribose. Proc. Natl Acad. Sci. USA  95: 10328–10333. Google Scholar Fromm, H. and Chua, N.-H. ( 1992) Cloning of plant cDNAs encoding calmodulin-binding proteins using 35S-labeled recombinant calmodulin as a probe. Plant Mol. Biol. Rep.  10: 199–206. Google Scholar Grant, M., Brown, I., Adams, S., Knight, M., Ainslie, A. and Mansfield, J. ( 2000) The RPM1 plant disease resistance gene facilitates a rapid and sustained increase in cytosolic calcium that is necessary for the oxidative burst and hypersensitive cell death. Plant J.  23: 441–450. Google Scholar Goodman, R.N. and Novacky, A.J. ( 1994) The hypersensitive reaction in plants to pathogens. APS Press, St. Paul, Minnesota, U.S.A. Google Scholar Guo, F.Q., Okamoto, M. and Crawford, N.M. ( 2003) Identification of a plant nitric oxide synthase gene involved in hormonal signaling. Science  302: 100–103. Google Scholar Hahlbrock, K., Scheel, D., Logemann, E., Nürnberger, T., Parniske, M., Reinold, S., Sacks, W.R. and Schmelzer, E. ( 1995) Oligopeptide elicitor-mediated defense gene activation in cultured parsley cells. Proc. Natl Acad. Sci. USA  92: 4150–4157. Google Scholar Harding, S.A., Oh, S.-H. and Roberts, D.M. ( 1997) Transgenic tobacco expressing a foreign calmodulin gene shows an enhanced production of active oxygen species. EMBO J.  16: 1137–1144. Google Scholar Harmon, A.C., Jarrett, H.W. and Cormier, M.J. ( 1984) An enzymatic assay for calmodulins based on plant NAD kinase activity. Anal. Biochem.  141: 168–178. Google Scholar Heo, W.D., Lee, S.H., Kim, M.C., Kim, J.C., Chung, W.S., Chun, H.J., Lee, K.J., Park, C.Y., Park, H.C., Choi, J.Y. and Cho, M.J. ( 1999) Involvement of specific calmodulin isoforms in salicylic acid-independent activation of plant disease resistance responses. Proc. Natl Acad. Sci. USA  96: 766–771. Google Scholar Kadota, Y., Goh, T., Tomatsu, H., Tamauchi, R., Higashi, K., Muto, S. and Kuchitsu, K. ( 2004) Cryptogein-induced initial events in tobacco BY-2 cells: pharmacological characterization of molecular relationship among cytosolic Ca2+ transients, anion efflux and production of reactive oxygen species. Plant Cell Physiol.  45: 160–170. Google Scholar Kim, M.C., Panstrugua, R., Elliott, C., Muller, J., Devoto, A., Yoon, H.W., Park, H.C. Cho, M.J. and Schulze-Lefert, P. ( 2002) Calmodulin interacts with MLO protein to regulate defence against mildew in barley. Nature  416: 447–450. Google Scholar Klee, C.B. and Vanaman, T.C. ( 1982) Calmodulin. Adv. Protein Chem.  35: 213–321. Google Scholar Kuchitsu, K., Yazaki, Y., Sakano, K. and Shibuya, N. ( 1997) Transient cytoplasmic pH change and ion fluxes through the plasma membrane in suspension-cultured rice cells triggered by N-acetylchitooligosaccharide elicitor. Plant Cell Physiol.  38: 1012–1018. Google Scholar Kumar, V.B., Bernardo, A.E., Alshaher, M.M., Buddhiraju, M., Purushothaman, R. and Morley, J.E. ( 1999) Rapid assay for nitric oxide synthase using thin-layer chromatography. Anal. Biochem.  269: 17–20. Google Scholar Kurkdjian, A. and Guern, J. ( 1989) Intercellular pH: measurement and importance in cell activity. Annu. Rev. Plant Mol. Biol.  40: 271–303. Google Scholar Lebrun-Garcia, A., Chiltz, A., Gout, E., Bligny, R. and Pugin, A. ( 2002) Questioning the role of salicylic acid and cytosolic acidification in mitogen-activated protein kinase activation induced by cryptogein in tobacco cells. Planta  214: 792–797. Google Scholar Lecourieux, D., Mazars, C., Pauly, N., Ranjeva, R. and Pugin, A. ( 2002) Analysis and effects of cytosolic free calcium increases in response to elicitors in Nicotiana plumbaginifolia cells. Plant Cell  14: 2627–2641. Google Scholar Lee, S.H., Kim, J.C., Lee, M.S., Heo, W.D., Seo, H.Y., Yoon, H.W., Hong, J.C., Lee, S.Y., Bahk, J.D., Hwang, I. and Cho, M.J. ( 1995) Identification of a novel divergent calmodulin isoform from soybean which has differential ability to activate calmodulin-dependent enzymes. J. Biol. Chem.  270: 21806–21812. Google Scholar Lee, S.H., Seo, H.Y., Kim, J.C., Heo, W.D., Chung, W.S., Lee, K.J., Kim, M.C., Cheong, Y.H., Choi, J.Y., Lim, C.O. and Cho, M.J. ( 1997) Differential activation of NAD kinase by plant calmodulin isoforms: the critical role of domain I. J. Biol. Chem.  272: 9252–9259. Google Scholar Liao, B., Gawienowski, M.C. and Zielinski, R.E. ( 1996) Differential stimulation of NAD kinase and binding of peptide substrates by wild-type and mutant plant calmodulin isoforms. Arch. Biochem. Biophys.  327: 53–60. Google Scholar Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. ( 1951) Protein measurement with the folin phenol reagent. J. Biol. Chem.  193: 265–275. Google Scholar Muto, S. and Miyachi, S. ( 1977) Properties of a protein activator of NAD kinase from plants. Plant Physiol.  59: 55–60. Google Scholar Niki, T., Mitsuhara, I., Seo, S., Ohtsubo, N. and Ohashi, Y. ( 1998) Antagonistic effect of salicylic acid and jasmonic acid on the expression of pathogenesis-related (PR) protein genes in wounded mature tobacco leaves. Plant Cell Physiol.  39: 500–507. Google Scholar Nürnberger, T., Nennstiel, D., Jabs, T., Sacks, W.R., Hahlbrock, K. and Scheel, D. ( 1994) High affinity binding of a fungal oligopeptide elicitor to parsley plasma membranes triggers multiple defense response. Cell  78: 449–460. Google Scholar Park, H.-J., Miura, Y., Kawakita, K., Yoshioka, H. and Doke, N. ( 1998) Physiological mechanisms of a sub-systemic oxidative burst triggered by elicitor-induced local oxidative burst in potato tuber slices. Plant Cell Physiol.  39: 1218–1225. Google Scholar Pugin, A., Frachisse, J.-M., Tavernier, E., Bligny, R., Gout, E., Douce, R. and Guern, J. ( 1997) Early events induced by the elicitor cryptogein in tobacco cells: involvement of a plasma membrane NADPH oxidase and activation of glycolysis and the pentose phosphate pathway. Plant Cell  9: 2077–2091. Google Scholar Qiao, J., Mitsuhara, I., Yazaki, Y., Sakano, K., Gotoh, Y., Miura, M. and Ohashi, Y. ( 2002) Enhanced resistance to salt, cold and wound stress by overproduction of animal cell death suppressors Bcl-xL and Ced-9 in tobacco cells—their possible contribution through improved function of organella. Plant Cell Physiol.  43: 992–1005. Google Scholar Reddy, A.S.N. ( 2001) Calcium: silver bullet in signaling. Plant Sci.  160: 381–404. Google Scholar Roberts, D.M. and Harmon, A.C. ( 1992) Calcium-modulated proteins: targets of intracellular calcium signals in higher plants. Annu. Rev. Plant Physiol. Plant Mol. Biol.  43: 375–414. Google Scholar Sagi, M. and Fluhr, R. ( 2001) Superoxide production by plant homologues of the gp91ph°x NADPH oxidase. modulation of activity by calcium and by tobacco mosaic virus infection. Plant Physiol.  126: 1281–1290. Google Scholar Sanders, D., Brownlee, C. and Harper, J.F. ( 1999) Communicating with calcium. Plant Cell  11: 691–706. Google Scholar Takezawa, D., Liu, Z.H., An, G. and Poovaiah, B.W. ( 1995) Calmodulin gene family in potato: developmental and touch-induced expression of the mRNA encoding a novel isoform. Plant Mol. Biol.  27: 693–703. Google Scholar Xu, H. and Heath, M.C. ( 1998) Role of calcium in signal transduction during the hypersensitive response caused by basidiospore-derived infection of the cowpea rust fungus. Plant Cell  10: 585–597. Google Scholar Yamakawa, H., Katou, S., Seo, S., Mitsuhara, I., Kamada, H. and Ohashi, Y. ( 2004) Plant MAPK phosphatase interacts with calmodulins. J. Biol. Chem.  279: 928–936. Google Scholar Yamakawa, H., Mitsuhara, I., Ito, N., Seo, S., Kamada, H. and Ohashi, Y. ( 2001) Transcriptionally and post-transcriptionally regulated response of 13 calmodulin genes to tobacco mosaic virus-induced cell death and wounding in tobacco plant. Eur. J. Biochem.  268: 3916–3929. Google Scholar Yamamoto, Y. ( 1966) NAD kinase in higher plants. Plant Physiol.  41: 523–528. Google Scholar Zielinski, R.E. ( 1998) Calmodulin and calmodulin-binding proteins in plants. Plant Mol. Biol.  49: 697–725. Google Scholar

Kang, Jiman;Mehta, Sohum;Turano, Frank J.

doi: 10.1093/pcp/pch159pmid: 15564521

Abstract The involvement of the putative glutamate receptor 1.1 (AtGLR1.1) gene in the regulation of abscisic acid (ABA) biosynthesis and signaling was investigated in Arabidopsis. Seeds from AtGLR1.1-deficient (antiAtGLR1.1) lines had increased sensitivity to exogenous ABA with regard to the effect of the hormone on the inhibition of seed germination and root growth. Seed germination, which was inhibited by an animal ionotropic glutamate receptor antagonist, 6,7-dinitroquinoxaline-2,3-[1H,4H]-dione, was restored by co-incubation with an inhibitor of ABA biosynthesis, fluridone. These results confirm that germination in antiAtGLR1.1 lines was inhibited by increased ABA. When antiAtGLR1.1 and WT seeds were co-incubated in fluridone and exogenous ABA, the antiAtGLR1.1 seeds were more sensitive to ABA. In addition, the antiAtGLR1.1 lines exhibited altered expression of ABA biosynthetic (ABA) and signaling (ABI) genes, when compared with WT. Combining the physiological and molecular results suggest that ABA biosynthesis and signaling in antiAtGLR1.1 lines are altered. ABA levels in leaves of antiAtGLR1.1 lines are higher than those in WT. In addition, the antiAtGLR1.1 lines had reduced stomatal apertures, and exhibited enhanced drought tolerance due to deceased water loss compared with WT lines. The results from these experiments imply that ABA biosynthesis and signaling can be regulated through AtGLR1.1 to trigger pre- and post-germination arrest and changes in whole plant responses to water stress. Combined with our earlier results, these findings suggest that AtGLR1.1 integrates and regulates the different aspects of C, N and water balance that are required for normal plant growth and development. (Received March 27, 2004; Accepted July 15, 2004) Introduction Abscisic acid (ABA) is a plant hormone involved in the control of a wide range of physiological processes, primarily seed germination and development. ABA is also associated with plant responses to environmental stresses such as drought, cold and salinity (Zeevaart and Creelman 1988, Koornneef and Karssen 1994). The most extensively studied ABA responses are those that regulate seed maturation, seed germination and stomatal closure (Schroeder et al. 2001, Finkelstein et al. 2002). Progression through embryonic development and the transition to germination is correlated with an increase in seed ABA content (Wang et al. 1998). The coordination of these events in the developing embryo blocks cell division and initiates accumulation of storage reserves that are used later by the germinating seedling (Finkelstein et al. 2002). Seed germination is regulated by many intrinsic signals (Koornneef et al. 2002). For instance, it is known that ABA inhibits seed germination, whereas gibberellin and ethylene antagonize this effect (Ritchie and Gilroy 1998, Lovegrove and Hooley 2000, Rolland et al. 2002). In addition, amino acid catabolism during the early stages of germination plays an important role in controlling development (Below et al. 2000) and may be controlled by ABA (Garciarrubio et al. 1997). Metabolites, such as carbon (C), can also interfere with seed germination by increasing ABA levels during early development (Arenas-Huertero et al. 2000). High concentrations of exogenous sugars (>300 mM) inhibit WT seedling development as evidenced by the failure to develop green, extended cotyledons, or true leaves (Gibson 2000). The ability of sugars to interfere with seed germination has been used to identify mutations, such as sugar insensitive (sis; Laby et al. 2000) and Glc insensitive (gin; Zhou et al. 1998), which exhibit reduced sensitivity to high concentrations of these sugars. Analyses of sis or gin mutants have shown that these mutants share components of the ABA biosynthetic and sugar signaling pathways, as evidenced by the identification of allelic loci in ABA-deficient (aba) and ABA-insensitive (abi) mutants (Rolland et al. 2002). For example, it has been shown that the sis5 (Laby et al. 2000) and gin6 (Arenas-Huertero et al. 2000) mutants are allelic to abi4, while the sis4 (Laby et al. 2000) and gin1 (Rook et al. 2001) mutants are allelic to aba2. Similar to C sensing in germinating seeds, ABA also appears to interact with nitrogen (N) signaling in the control of stomatal opening. Recent work has established that nitric oxide (NO) is a key regulator of stomatal closure (Mata and Lamattina 2001). Nitrate reductase (NR) and nitric oxide synthase (AtNOS1)-mediated NO synthesis in guard cells are required for ABA-induced closure (Desikan et al. 2002, Guo et al. 2003a). In addition, in pepper (Capsicum annuum L.), N deprivation in xylem sap can cause stomatal closure (Dodd et al. 2003). Interestingly, the nitrate transporter, AtNRT1.1 (CHL1), is expressed in Arabidopsis guard cells and it functions in stomatal opening. Analysis of chl1 mutants indicates that the mutation does not affect ABA-induced stomatal closure; thus ABA sensing in the guard cells of chl1 lines appears normal (Guo et al. 2003b). However, ABA titers and the expression of ABA biosynthetic genes (ABA) were not measured in chl1 lines, therefore it is still possible that ABA biosynthesis is activated in the chl1 lines, which results in stomatal closure. Much progress has been made recently in revealing ‘crosstalk’ between C and ABA signaling pathways, and several novel signaling components have been identified (Finkelstein and Gibson 2001, Hetherington 2001). In addition, we have demonstrated that the putative plant glutamate receptor 1.1 gene in Arabidopsis (AtGLR1.1) functions as a regulator of C/N metabolism and a key step in ABA biosynthesis (Kang and Turano 2003). In addition, we demonstrated that AtGLR1.1-deficient Arabidopsis lines (antiAtGLR1.1) had a conditional germination phenotype that was sensitive to different C:N ratios or the iGLR antagonist 6,7-dinitroquinoxaline-2,3-[1H,4H]-dione (DNQX), and exhibited increased ABA concentrations during early development. Thus, we investigated further the interaction between AtGLR1.1 and ABA signaling by combining physiological and molecular approaches. Here, we show that antiAtGLR1.1 lines overproduce ABA and are hypersensitive to ABA, which affects several physiological processes such as seed germination, root length, stomatal opening, water loss and drought tolerance. Furthermore, we analyzed the expression patterns of the ABA and ABA signaling (ABI) genes and show that altered accumulation of the ABA and ABI transcripts in antiAtGLR1.1 lines is consistent with the ABA-related phenotypes. Our results provide strong evidence that AtGLR1.1 functions as a molecular and biochemical link between C/N status and metabolism, ABA metabolism and sensitivity, and response to water stress in Arabidopsis. Results AntiAtGLR1.1 response to ABA Earlier, we demonstrated that 3-day-old antiAtGLR1.1 lines contained eight times more endogenous ABA than did wild-type (WT) lines grown on Murashige and Skoog (MS) media supplemented with 3% sucrose and DNQX (Kang and Turano 2003). Since ABA is considered a key regulator in seed germination and root growth (Finkelstein et al. 2002, Kang et al. 2002), ABA sensitivity was compared in two transgenic (antiAtGLR1.1-1 and antiAtGLR1.1-2) and WT lines on media containing a range of ABA concentrations using two semi-quantitative bioassays. First, to test the effects of ABA on germination, WT and antiAtGLR1.1 seeds were germinated on plates containing 0–6 µM ABA. Second, to test post-germination development, WT and antiAtGLR1.1 seeds were germinated on ABA-free media for 4 d, and then the seedlings were transferred to media supplemented with different concentrations (0–10 µM) of ABA. When the exogenous ABA concentration was 4.0 µM, antiAtGLR1.1 seeds germinated but exhibited mild growth retardation when compared with WT seedlings (Fig. 1A). Without ABA, antiAtGLR1.1 seedlings developed similarly to WT (Fig. 1A). When the exogenous ABA concentration was 2.0 µM or greater the germination of antiAtGLR1.1 lines was significantly decreased when compared with WT (Fig. 1B). For example, the germination of antiAtGLR1.1 was decreased to 40% (antiAtGLR1.1-1) and 30% (antiAtGLR1.1-2) in the presence of 4.0 µM ABA. However, approximately 80% of the WT seeds germinated under the same conditions. Likewise, root growth of the antiAtGLR1.1 lines was also hypersensitive to exogenous ABA (Fig. 1C). There were no differences between the root length of antiAtGLR1.1 and WT lines on MS plates minus ABA (Fig. 1D). However, at 0.5 µM ABA, WT root growth was 72% of the control (MS minus ABA), whereas that of antiAtGLR1.1 plants was 58% and 51%, respectively. When plates were supplemented with 5 µM ABA, the root growth of antiAtGLR1.1 was 22% of that of the control. At the same ABA concentration, WT plants grew at 33% of the 0 µM ABA control. The implication of these findings is that ABA signaling is regulated through AtGLR1.1 to trigger the inhibition of germination and post-germination root development. ABA biosynthesis and sensitivity Previously, we demonstrated that ABA titers were increased in 3-day-old antiAtGLR1.1 lines treated with the iGLR antagonist DNQX, which inhibits seed germination (Kang and Turano 2003). We hypothesized that the inhibition of antiAtGLR1.1 seed germination was due to elevated endogenous ABA. To test this hypothesis, we treated WT and antiAtGLR1.1 lines with fluridone, an inhibitor of carotenoid and ABA biosynthesis (Gamble and Mullet 1986), and DNQX. Here we demonstrate the ability of 100 µM fluridone to reverse the DNQX effect on MS media containing 200 µM DNQX. This fluridone concentration had been shown previously to be effective in reducing the endogenous ABA pool in Phaseolus vulgaris L. (Moreno-Fonseca and Covarrubias 2001) and Arabidopsis (Ullah et al. 2002). As shown previously (Kang and Turano 2003), the germination of seeds from antiAtGLR1.1 lines was inhibited on MS medium supplemented with 3% sucrose and 200 µM DNQX (Fig. 2A). However, the DNQX-mediated inhibition of seed germination and development of antiAtGLR1.1 seedlings was restored by simultaneous incubation with 100 µM fluridone (Fig. 2A). At 50 µM fluridone, the DNQX-mediated inhibition of seed germination was restored to approximately 80% (data not shown). These results strongly suggest that decreased seed germination of antiAtGLR1.1 lines by DNQX was due to elevated endogenous ABA titers, which is consistent with our earlier observations, which showed a 160% increase in endogenous ABA concentrations in the presence of 200 µM DNQX when compared with that of the untreated antiAtGLR1.1 controls (Kang and Turano 2003). To assess ABA sensitivity of the antiAtGRL1.1 lines, we examined the effect of increasing concentrations of exogenous ABA in the presence of fluridone (Fig. 2B). WT and antiAtGLR1.1 seeds were grown on media containing 100 µM fluridone and different ABA concentrations (0–9 µM) in the presence of 3% sucrose. At 3 µM, ABA did not significantly decrease germination of antiAtGLR1.1 or WT seeds. However, at 9 µM ABA, the germination of antiAtGLR1.1 seeds decreased to approximately 25%, whereas at the same ABA concentration nearly 90% of the WT seeds germinated. Since these experiments were conducted in the presence of fluridone, an inhibitor of endogenous ABA, these findings suggest that the antiAtGLR1.1 lines are hypersensitive to exogenous ABA due to alterations in ABA sensing. Expression of ABA-related genes in antiAtGLR1.1 lines The antiAtGLR1.1 lines exhibit ABA hypersensitivity, or ABA-associated phenotypes, suggesting that AtGLR1.1 enhances various aspects of the ABA response. Moreover, we demonstrated previously that the transcript for ABA1 was more abundant in antiAtGLR1.1 plants as compared with WT (Kang and Turano 2003). ABA1 encodes a zeaxanthin epoxidase that controls an early step of ABA biosynthesis by converting zeaxanthin to violaxanthin (Rock and Zeevaart 1991). Thus, to understand the transcriptional regulation of ABA biosynthesis and sensing in antiAtGLR1.1 lines, we analyzed the expression patterns of ABA and ABI genes by RT-PCR. The transcript levels of ABA biosynthetic genes, such as ABA2 and ABA3, were enhanced in antiAtGLR1.1–1 and antiAtGLR1.1–2 lines when compared with WT (Fig. 3). ABA2 encodes a short-chain dehydrogenase/reductase (SDR1), which is involved in the second to last step of ABA biosynthesis (Cheng et al. 2002). ABA3, which encodes a putative molybdenum cofactor sulfurase, is a key regulator of ABA biosynthesis (Bittner et al. 2001, Xiong et al. 2001). Interestingly, the transcripts for ABI1 and ABI2, which encode homologous class 2C protein serine/threonine phosphatases (PP2Cs; Leung et al. 1997) were decreased in antiAtGLR1.1 lines. However, ABI4 and ABI5, which encode transcription factors of the AP2 domain and the basic leucine zipper (bZIP) families, respectively (Finkelstein et al. 1998, Lopez-Molina et al. 2001), were increased in antiAtGLR1.1 lines compared with WT. We also looked at the expression of the ABA- and stress-inducible alcohol dehydrogenase (ADH) (de Bruxelles et al. 1996) and rab18 (Lang and Palva 1992) genes. The ADH and rab18 transcripts increased approximately 40% in the antiAtGLR1.1 lines when compared with levels in WT. Stomatal opening Another well-documented ABA-regulated process is stomatal closure, which minimizes water loss through gas exchange into and out of leaves (Leung and Giraudat 1998). Some ABA signaling mutants, such as abi1 and abi2, exhibit impaired stomatal closure in vegetative tissue (Leung and Giraudat 1998, Schroeder et al. 2001). Therefore, we examined the role of AtGLR1.1 in controlling stomatal aperture in Arabidopsis leaves. Epidermal peels were prepared from leaves of WT and antiAtGLR1.1 plants, and stomatal apertures between the guard cells were observed during the early morning (Fig. 4A). Under normal growth conditions, the stomatal apertures of WT plants were found to be approximately 3.5 times wider than those in antiAtGLR1.1 lines (Fig. 4B). To confirm that the elevated accumulation of the ABA1 (Kang and Turano 2003) ABA2 and ABA3 (Fig. 3) transcripts in the antiAtGLR1.1 lines affected ABA biosynthesis, we measured endogenous ABA concentrations in 30-day-old WT and antiAtGLR1.1 plants (Fig. 4C). The antiAtGLR1.1 lines contained 30% more ABA when compared with WT (WT, 0.203±0.007; antiAtGLR1.1 0.263±0.015, n=3). Combined, these findings suggest that AtGLR1.1 modulates both the ABA biosynthesis and signaling genes and responses. Drought tolerance and water loss of antiAtGLR1.1 lines Since antiAtGLR1.1 lines exhibited increased stomatal closure in epidermal strips, we investigated whether this phenomenon also affected water loss in whole transgenic plants. Both WT and antiAtGLR1.1 plants were grown and irrigated for 21 d, and then subjected to drought stress by terminating irrigation as described in Pei et al. (1998) and Vartanian et al. (1994). To minimize soil evaporation, the soil surface in the pots was covered. After 16 d of drought treatment, WT plants had withered severely and showed complete chlorosis of rosette leaves. However, leaves of antiAtGLR1.1 plants remained green and turgid (Fig. 5A). Water loss from pots containing antiAtGLR1.1 versus WT plants was decreased (Fig. 5B) which is consistent with the observed phenotypes. For example, measurements of changes in soil water content after 10 d of drought stress showed that water loss, due to transpiration, was approximately 21% lower in pots that contained antiAtGLR1.1 plants than in pots containing WT. Discussion In this study, we demonstrate that Arabidopsis lacking AtGLR1.1 are sensitive to elevated exogenous ABA with regard to inhibition of seed germination and root growth (Fig. 1). Although endogenous ABA is essential for the induction of dormancy, the germination of WT Arabidopsis seeds can be suppressed by exogenous (3 µM) ABA (Finkelstein et al. 2002). However, the germination of mature antiAtGLR1.1 seeds was blocked by as little as 1 µM ABA. These results are consistent with the observation that antiAtGLR1.1 lines have increased endogenous ABA levels during early seedling development (Kang and Turano 2003). Furthermore, this hypothesis is consistent with the ability of fluridone to reverse the inhibitory effects of ABA on seed germination (Fig. 2), which indicates that part of the DNQX-mediated inhibition of germination is due to the accumulation of endogenous ABA. However, this interpretation is confounded by the fact that the antiAtGLR1.1 lines have increased sensitivity to exogenous ABA in the presence of fluridone. Combined, these data suggest that ABA-mediated inhibition of seed germination or post-germination growth is regulated by the relative accumulation, or activity, of AtGLR1.1, which results in increased ABA biosynthesis and sensitivity. The physiological results during early development are consistent with the molecular results that were observed in 30-day-old plants, where the accumulation of ABA transcripts was increased and the expression of the ABI genes was consistent with plants with increased ABA sensitivity (Fig. 3; Kang and Turano 2003). Arenas-Huertero et al. (2000) concluded that the inhibitory effect of high Glc on germination and seedling development was mediated by increases in ABA levels during early development. Previously, endogenous ABA concentrations were determined in 3-day-old WT and antiAtGLR1.1 seedlings incubated in liquid culture containing MS medium with 3% Suc and 200 µM DNQX. The antiAtGLR1.1 lines contained eight times more ABA than did WT (Kang and Turano 2003). The observation that 3-day-old antiAtGLR1.1 plants incubated in MS medium with 3% Suc contained much higher ABA than 30-day-old plant grown on soil may be explained by the fact that 3% Suc contributes to an increase in ABA levels during early antiAtGLR1.1 seedling development. Many independently isolated sugar- or Glc-insensitive mutants are allelic to aba or abi mutants (Finkelstein et al. 2002, Rolland et al. 2002). Of the ABA genes tested here, the expression of ABA2 and ABA3 showed a marked difference in their accumulation between WT and antiAtGLR1.1 lines. In addition, we have preliminary results from genome-wide microarray analysis (data not shown) that show an increase in another ABA biosynthetic gene, AAO3 (Seo et al. 2000), in the antiAtGLR1.1 lines. We cannot definitively state whether the increased levels of the ABA biosynthetic transcripts in antiAtGLR1.1 lines are due to increased gene expression or a decrease in transcript turnover during or after germination. Cheng et al. (2002) observed that ABA1, ABA2 and ABA3 transcripts were induced in seedlings after treatment in 2% or 6% Glc, suggesting that high levels of sugar are required to activate ABA genes (Cheng et al. 2002). We observed that the germination of antiAtGLR1.1 seeds was inhibited on MS plates minus inorganic N in the presence of 25 mM Suc but germination was restored with co-incubation of 5 mM NO3–, suggesting that AtGLR1.1 is associated with sugar or nitrogen sensing (Kang and Turano 2003). Interestingly, germination of antiAtGLR1.1 seeds had a similar response to higher Glc (100 mM) under the same conditions (unpublished data), so it appears that an alternative explanation is that antiAtGLR1.1 plants may accumulate unusually high concentrations of endogenous sugars or have increased sensitivity to different sugars, which could stimulate ABA accumulation by activating ABA biosynthesis genes or alter the expression of the ABI genes. This hypothesis awaits further investigation. We found that the transcripts for ABI1 and ABI2 were less abundant in antiAtGLR1.1 lines compared with WT. Null mutations in the PP2Cs, encoded by ABI1 and ABI2, display ABA hypersensitivity (Merlot et al. 2001). Conversely, constitutive overexpression of ABI1 in maize mesophyll protoplasts inhibits ABA action (Sheen 1998), suggesting that PP2Cs are involved in the negative regulation of ABA signaling. This finding implies that a reduction in PPC activity can increase ABA sensitivity, which is consistent with our evidence that the reduced accumulation of the AtGLR1.1 peptide may result in ABA hypersensitivity. The reason for the decreased levels of the ABI1 and ABI2 transcripts may be elevated ABA. The effects of ABA on ABI1 expression are mixed. Hoth et al. (2002) demonstrated that ABA increases the ABI1 and ABI2 transcripts in WT. However, ABI2 is not ABA inducible in the abi1 lines, suggesting that decreased ABI1 activity, such as in the antiAtGLR1.1 lines, will have low ABI2. ABA has also been shown to decrease the ABI1 transcript when compared with controls (Lu et al. 2002); the change in ABI1 expression was not described in the text so it is unclear whether this result is significant. Furthermore, ABA has been shown to either slightly decrease or have no effect on the accumulation of the ABI1 transcript (Price et al. 2003). So in appears that the regulation of ABI1, and ABI2, by ABA has not been clearly defined, perhaps some of the variability in the results is associated with the different developmental stages of the tissues, tissue type, the distinct experimental conditions used in each study and/or the relative levels of each ABI gene. It is also possible that alteration of ABI1 gene expression in the antiAtGLR1.1 lines might be independent of the increase in ABA content. The ABI4 and ABI5 genes have been shown to encode putative transcription factors of the AP2 and bZIP domain factor classes, respectively (Finkelstein et al. 1998, Finkelstein and Lynch 2000). The overexpression of ABI5 in Arabidopsis resulted in ABA-induced (3 µM ABA) reduction in growth and hypersensitivity to sugars in the inhibition of seedling growth, whereas overexpression of ABI4 blocked growth under the same conditions (Brocard et al. 2002). The increased ABI5 transcript levels in antiAtGLR 1.1 lines may be explained by the fact that the increased endogenous ABA levels induce ABI5 expression and result in inhibition of seed germination and development (Fig. 1). This result is consistent with the published molecular observations that ABA transiently increases the accumulation of the ABI5 transcript (Lu et al. 2002, Arroyo et al. 2003) and ABI5 peptide (Lu et al. 2002) in WT young (2- to 10-day-old) seedlings and in roots, where Brocard et al. (2002) showed that ABA treatments resulted in very strong expression of ABI5 in roots, as well as our work here showing that fluridone restored DNQX-mediated inhibition of seed germination. The molecular results, mentioned above, are contradicted by the results of Soderman et al. (2000), which show that ABI4 and ABI5 transcripts do not accumulate in 15-day-old WT seedlings after exposure to ABA treatment. However, in the same paper it was shown that ABI5 is ABA inducible when either ABI3 or ABI4 expression is increased. Since ABI4 is elevated in antiAtGLR1.1 lines it is plausible that ABI5 is ABA inducible in these lines. In addition, elevated ABI5 levels positively affects ABA induction of ABI5 (Brocard et al. 2002) and ABI5 has been proposed to function as a positive autoregulator (Soderman et al. 2000). ABA also transiently increases the accumulation of ABI4 transcripts in 3-day-old WT seedlings (Arroyo et al. 2003, Price et al. 2003), but like results of the accumulation of ABI5, described above, other studies contradict that finding; ABI4 did not accumulate in vegetative tissue of 15-day-old plants (Soderman et al. 2000) after ABA treatment. But the ABI4 transcript accumulates in the abi2-1 mutant (Soderman et al. 2000), suggesting that decreased AB±2 activity results in elevated ABI4. The ABI2 transcript is lower in antiAtGLR1.1 lines, thus it is plausible that ABI4 could be elevated in antiAtGLR1.1 lines. It should be noted that the ABI3 transcript did not accumulate in antiAtGLR1.1 lines, a finding confirmed by 10 sets of experiments utilizing either RT-PCR or genome-wide microarray experiments (data not shown). As described by Soderman et al. (2000), it appears that a role of the ABI genes in the control of the ABA response is a complex regulatory network. Although it appears that ABA can modulate the expression of the ABI genes it is possible that other factors such as sugar can affect the expression of these genes, since the ABI4 and ABI5 (Cheng et al. 2002, Arroyo et al. 2003), but not the ABI1 and ABI2 (Cheng et al. 2002), genes are Glc inducible. It is possible that the induction of ABI4 and ABI5 in the antiAtGLR1.1 lines might be due to high cellular sugar levels or due to the perception of high sugar levels in the plant. This alternative hypothesis may also partially explain the increased sucrose- or Glc-sensitivity that we observe in the antiAtGLR 1.1 lines, but this hypothesis does not explain the significant decrease in the ABI1 and ABI2 transcripts in antiAtGLR1.1 lines. It is plausible that a combination of regulatory factors, as proposed by Soderman et al. (2000), such as elevated ABA and sugar gives raise to a distinct set of ABI responses but the potential role of elevated sugars in the antiAtGLR1.1 lines is speculative and remains to be tested in the future. Stomatal closure is an ABA-controlled process for coping with water-deficit stress. Our data indicate that AtGLR1.1 plays a major role in the control of stomatal closure in guard cells. The stomatal openings of antiAtGLR1.1 lines were smaller than those of WT plants (Fig. 4). In this study, we demonstrated that ABA transcripts increased in 30-day-old plants and we measured ABA levels in 30-day-old WT and antiAtGLR1.1 plants grown in soil. The antiAtGLR1.1 lines contained 30% more ABA when compared with WT. How a relatively small increase in ABA concentration results in a dramatic physiological change remains unclear; one possible explanation is that concentrations of ABA are very high in distinct cells or tissues in the leaf that may affect stomatal closure, i.e. guard cells. Regardless of this apparent discrepancy, the data are consistent, thus we conclude that the reduced stomatal opening is the result of increased endogenous ABA levels in antiAtGLR1.1 lines. In addition, we observed earlier that the transcript for the nitrate transporter CHL1 was less abundant (30% decrease) in antiAtGLR1.1 plants compared with WT. Guo et al. (2003b) demonstrated that CHL1 was expressed and functions in guard cells, and chl1 mutants had reduced stomatal opening. Thus the smaller stomatal apertures in antiAtGLR 1.1 lines may also be due to a decrease in the CHL1 transcript, in addition to increased ABA concentrations. The abi1-1 and abi2-1 (Koornneef et al. 1984) mutants exhibited impaired ABA-induced stomatal closure and wilty phenotypes (Schroeder et al. 2001). Thus the stomata of abi1-1 and abi2-1 do not close in response to exogenous ABA or drought stress (Pei et al. 1997). The abi1-1 and abi2-1 mutants have decreased but detectable amounts of phosphatase activity when compared with WT. However intragenic revertants of the abi1 mutant lack detectable in vitro phosphatase activity and interestingly this appears to result in reduced stomatal apertures, 15–20% smaller than in WT (Gosti et al. 1999). This finding implies that a large or significant reduction in protein phosphatase activity can lead to stomatal closure. In antiAtGLR1.1 lines, the transcripts of ABI1 and ABI2 were significantly decreased to approximately 50% and 60% of the levels observed in WT. Thus, it is possible that the reduction in the ABI1 and ABI2 transcripts assists in triggering stomatal closure in antiAtGLR1.1 leaves as well. Soil water content from pots containing antiAtGLR1.1 plants decreased more slowly during drought treatment than those of WT plants (Fig. 5B). The antiAtGLR1.1 plants also withstood drought stress better than WT (Fig. 5A), suggesting that antiAtGLR1.1 plants have increased tolerance to drought due to decreased water loss. These results are consistent with the observed increased stomatal closure in antiAtGLR1.1 lines. It is possible that a decrease in the CHL1 transcript and/or reduced transcripts for ABI1 and ABI2 assist in triggering the drought adaptive response; because chl1 mutants also displayed drought tolerance and resulted in higher water content (Guo et al. 2003b) and intragenic relevants of abi1 grown in a microphytotron displayed enhanced drought tolerance when compared with WT (Gosti et al. 1999). In addition, the accumulation of the ADH and rab18 transcripts is higher in antiAtGLR1.1 lines, suggesting that induction of ABA- and stress-inducible genes may be required for, or associated with, drought tolerance in these lines. To understand the multiple ABA perception and signaling mechanisms, extensive genetic studies have been performed, and many likely signal intermediates or components have been identified (Leung et al. 1997, Finkelstein et al. 2002). Despite these efforts toward elucidating a clear and linear understanding of ABA signaling, the molecular mechanisms underlying ABA signaling and the resulting physiological responses remain complicated and are best described as a combinatorial network for the control of seed (Finkelstein et al. 2002, Rolland et al. 2002). Recently, ABI4, ABI5 and ABA have been shown to play important roles in physiological responses in vegetative tissue such as mediating the inhibitory effects of NO3– on root development, possibly through changes in sensing the C/N balance (Signora et al. 2001). The link between C/N metabolism and ABA biosynthesis is further supported by our previous observation that genes in these pathways are altered in antiAtGLR1.1 lines (Kang and Turano 2003). The results from this investigation strongly suggest that AtGLR1.1 is involved in the control and/or regulation of ABA biosynthesis, ABA signaling and distinct physiological events (Fig. 6). Taken together, the results from all these investigations raise the exciting possibility that the AtGLR1.1-mediated regulation of C/N metabolism is of fundamental importance for ABA signaling, which may represent the link between C/N status and the regulation of developmental events and/or plant responses to certain environmental cues (Rabe 1990, Bohnert and Sheveleva 1998, Rolland et al. 2002). Materials and methods Plant materials and growth conditions Arabidopsis thaliana ecotype Wassilewskija was maintained at 20–21°C, with 60–70% relative humidity, under cool white fluorescent light (140 µmol of protons m–2 s–1) with a 16-h light/8-h dark cycle. Seeds from antiAtGLR1.1 transgenic lines (Kang and Turano 2003) were used, and their phenotypes were confirmed before use. Plants were grown on MS media supplemented with ABA, fluridone or DNQX and 0.8% (w/v) phytagar. For root growth measurements, plants were germinated and grown in the vertical position. For all other experiments, plants were maintained in the horizontal position. Growth assays Stock solutions (100 µM) of fluridone (Chem Service, West Chester, PA, U.S.A.) were freshly prepared for each experiment. Details of fluridone preparation are described in Ullah et al. (2002). The volume of dissolved fluridone was brought to 500 ml with deionized water. Fluridone was added from the stock solution to the MS medium to reach the indicated final concentration. For most experiments with fluridone, seeds from different Arabidopsis lines were treated with 50% household bleach for 5 min, washed three times with sterile water, plated on MS media and incubated in the dark at 4°C for 48 h. ABA (Sigma) was added to the MS media supplemented with 3% sucrose. To examine ABA sensitivity, WT and mutant seeds were incubated in the dark at 4°C for 24 h and grown as described previously (Kang and Turano 2003). The germination rate was determined 12 d after incubation. Water loss and drought tolerance measurements The drought experiments and estimates of water loss through transpiration were conducted as described by Vartanian et al. (1994) and Pei et al. (1998). Briefly, both WT and antiAtGRL1.1 plants were germinated in individual pots and irrigated for 21 d and then drought stressed by terminating irrigation. Estimates of water loss due to transpiration in WT and antiAtGLR1.1 plants were conducted on plants from the same developmental stage based on the initiation of drought stress (day 0). The soil surface of each pot was covered with plastic to minimize the loss of water through soil evaporation. Measurement of ABA contents ABA content was measured in 30-day-old plants. ABA extraction and determination was performed as described previously (Kang and Turano 2003). Stomatal aperture measurement Stomatal aperture was determined from leaves of 25- to 30-day-old WT or antiAtGLR1.1 plants. Leaves were collected in the early morning. The adaxial face of each leaf was affixed to a strip of transparent tape and epidermal strips were peeled off from the adaxial leaf surface. The strips were then placed directly, adhesive-side down, onto microscope slides and guard cells were viewed at 40× or 100× magnification with a Lieca DC 200 Leica Microsystem microscope. Photographs were taken using a Sony Mavica MVC-FD91 digital camera and stomatal apertures were individually measured using digital software (Grab it, Raleigh, NC, U.S.A.). RNA isolation and RT-PCR analysis Total RNA was extracted from leaves of 30-day-old plants with the RNeasy Plant Mini Kit (Qiagen Inc., Valencia, CA, U.S.A.). To determine the integrity of the RNA and to ensure that equal amounts of RNA were added to each reaction, 1 µg of RNA from each sample was separated via gel electrophoresis in formaldehyde-formamide gels as described by Turano et al. (2002). Total RNA (1 µg) served as the template for RT-PCR using the RT-PCR beads System (Amersham Bioscience Corp., Piscataway, NJ, U.S.A.) with 25 pmol of each transcript specific primer. All primers were commercially synthesized (Bio-Synthesis, Inc., Lewsiville, TX, U.S.A.). The RT-PCR was conducted as follows: 42°C for 15 min, 95°C for 5 min, followed by 25 cycles of 94°C for 30 s, 55°C for 30 s, 72°C for 2 min and a 72°C extension reaction for 2 min. Results were confirmed as described previously by Kang and Turano (2003). The primers used and corresponding accession numbers were as follows: TUB4 (Acc. No. M21415), 5′TUB4 5′-TTGCTGTCTTCGTTTCCCTGG-3′, 3′TUB4 5′-GAGGGTGCCATTGACAACATC-3′; ABA2 (Acc. No. AY082345), 5′ABA2 5′-AGCTTGGACAGCACGGGATACGTG-3′, 3′ABA2 5′-TTGGCGGACAATAAACATTAAACA-3′; ABA3 (Acc. No. AF325457), 5′ABA3 5′-TGGAGGTGAACCATACGGGGAAGA-3′, 3′ABA3 5′-TAGTAGACGATAATACACACAAGC-3′; ABI1 (Acc. No. ATMRABI1), 5′ABI1 5′-GATATCTCCGCCGGAGATGAGATC-3′, 3′ABI1 5′-CATTCCACTGAATCACTTTCCCTC-3′; ABI2 (Acc. No. ATABI2RNA), 5′ABI2 5′-GTTCTTGTTCTGGCGACGGAGC-3′, 3′ABI2 5′-CCATTAGTGACTCGACCATCAAG-3′; ABI4 (Acc. No. AF040959), 5′ABI4 5′-GGATTTTATTGATCCCATCTTGGG-3′, 3′ABI4 5′-CCACCATCTCCTCCGATTCTCTTC-3′, ABI5 (Acc. No. AF334206) 5′ABI5 5′-CCTTGACGAGTTCCAACATGCTTT-3′, 3′ABI5 5′-ACCATAAAGCTGTTGCTGTGCAGC-3′; ADH (Acc. No. M12196) 5′ADH 5′- GCCAGGAGATCATGTGTTGC-3′, 3′ADH 5′-GCACCAGCGATTCTAGCACC-3′, and rab18 (Acc. No. X68042), 5′rab18 5′-TTTGGAACTGGCGGAGGAGCTAGG-3′, 3′rab18 5′-AGCATCATATCCGGATCCCATGCCG-3′. Acknowledgments Portions of this work were funded by a USDA/NRI, Plants Response to the Environment (CGA# 2001-35100-09930 3125) grant to F.J.T. 1 Present address: Department of Biology, Johns Hopkins University, 3400 North Charles Street, Baltimore, MD 21218-2685, USA. 2 Corresponding author: E-mail, [email protected]; Fax, +1-202-994-6100. View largeDownload slide Fig. 1 ABA sensitivity of antiAtGLR1.1 plants. (A) Seeds/seedlings from WT and antiAtGLR1.1 lines were maintained on MS media containing 3% (w/v) Suc and 0.8% (w/v) Phytagar for 7 d in the presence of 0 (left) or 4 µM (right) ABA. (B) ABA dose-dependent response of germination. The germination of WT and antiAtGLR1.1 seeds was scored after a 7-d incubation in the presence of different concentrations of ABA (0–6 µM). Each treatment consists of 24–30 seedlings (n = 4). Small bars represent the SE. The symbols are as follows: WT, solid diamonds; antiAtGLR1.1-1, open triangles and antiAtGLR1.1-2, solid squares. (C) Root growth response to ABA dosage. Seeds were germinated for 4 d on ABA-free medium, and then seedlings were transferred to MS media containing 3% (w/v) Suc in the presence of 0.5 (left) or 5 (right) µM ABA. (D) Root growth response to ABA. Seeds and seedlings were treated as described in C (above) in the presence of different concentrations of ABA (0–10 µM). Plates were photographed and root length was measured individually using digital software. Each treatment consists of six to eight seedlings (n = 3). Small bars represent the SE. The symbols are as follows: WT, solid diamonds; antiAtGLR1.1-1, open triangles and antiAtGLR1.1-2, solid squares. View largeDownload slide Fig. 1 ABA sensitivity of antiAtGLR1.1 plants. (A) Seeds/seedlings from WT and antiAtGLR1.1 lines were maintained on MS media containing 3% (w/v) Suc and 0.8% (w/v) Phytagar for 7 d in the presence of 0 (left) or 4 µM (right) ABA. (B) ABA dose-dependent response of germination. The germination of WT and antiAtGLR1.1 seeds was scored after a 7-d incubation in the presence of different concentrations of ABA (0–6 µM). Each treatment consists of 24–30 seedlings (n = 4). Small bars represent the SE. The symbols are as follows: WT, solid diamonds; antiAtGLR1.1-1, open triangles and antiAtGLR1.1-2, solid squares. (C) Root growth response to ABA dosage. Seeds were germinated for 4 d on ABA-free medium, and then seedlings were transferred to MS media containing 3% (w/v) Suc in the presence of 0.5 (left) or 5 (right) µM ABA. (D) Root growth response to ABA. Seeds and seedlings were treated as described in C (above) in the presence of different concentrations of ABA (0–10 µM). Plates were photographed and root length was measured individually using digital software. Each treatment consists of six to eight seedlings (n = 3). Small bars represent the SE. The symbols are as follows: WT, solid diamonds; antiAtGLR1.1-1, open triangles and antiAtGLR1.1-2, solid squares. View largeDownload slide Fig. 2 Effect of fluridone on DNQX-mediated inhibition of seed germination. (A) Seeds/seedlings from WT and antiAtGLR1.1 lines were maintained on MS media containing 3% (w/v) Suc and 0.8% (w/v) Phytagar in the presence of DNQX (200 µM), a plant-derived iGLR antagonist, plus or minus 100 µM fluridone. Each treatment consists of 24–30 seedlings. Experiments were performed more than three times. (B) Effect of exogenous ABA in WT and antiAtGLR1.1 lines grown on media containing 100 µM fluridone. WT and antiAtGLR1.1 lines were grown on MS media containing 3% (w/v) Suc and 0.8% (w/v) Phytagar in the presence of 100 µM fluridone and different concentrations of ABA (0–9 µM). Each treatment consists of 25–30 seedlings (n = 3). Small bars represent the SE. The designations for the bar graphs are as follows: WT, open bar; antiAtGLR1.1-1, black bar; antiAtGLR1.1-2, gray bar. View largeDownload slide Fig. 2 Effect of fluridone on DNQX-mediated inhibition of seed germination. (A) Seeds/seedlings from WT and antiAtGLR1.1 lines were maintained on MS media containing 3% (w/v) Suc and 0.8% (w/v) Phytagar in the presence of DNQX (200 µM), a plant-derived iGLR antagonist, plus or minus 100 µM fluridone. Each treatment consists of 24–30 seedlings. Experiments were performed more than three times. (B) Effect of exogenous ABA in WT and antiAtGLR1.1 lines grown on media containing 100 µM fluridone. WT and antiAtGLR1.1 lines were grown on MS media containing 3% (w/v) Suc and 0.8% (w/v) Phytagar in the presence of 100 µM fluridone and different concentrations of ABA (0–9 µM). Each treatment consists of 25–30 seedlings (n = 3). Small bars represent the SE. The designations for the bar graphs are as follows: WT, open bar; antiAtGLR1.1-1, black bar; antiAtGLR1.1-2, gray bar. View largeDownload slide Fig. 3 Transcript accumulation of ABA biosynthetic (ABA) and signaling (ABI) genes in WT and antiAtGLR1.1 lines. (A) Representative results of RT-PCR analysis: the experiment was repeated more than three times, with consistent results. RT-PCR analysis was performed with 1 µg of total RNA isolated from the leaves of 30-day-old plants. Details of the RT-PCR conditions, controls, confirmation and primer sequences are provided in Materials and Methods. 1 and 2 represent antiAtGLR1.1-1 and 1-2 lines, respectively. (B) Results from densitometric analyses of the RT-PCR data from three experiments. The band intensity for each of the two transgenic lines was determined and used to calculate the mean intensity for each band. The results were expressed as a percentage of the WT control (100%) ±SD. View largeDownload slide Fig. 3 Transcript accumulation of ABA biosynthetic (ABA) and signaling (ABI) genes in WT and antiAtGLR1.1 lines. (A) Representative results of RT-PCR analysis: the experiment was repeated more than three times, with consistent results. RT-PCR analysis was performed with 1 µg of total RNA isolated from the leaves of 30-day-old plants. Details of the RT-PCR conditions, controls, confirmation and primer sequences are provided in Materials and Methods. 1 and 2 represent antiAtGLR1.1-1 and 1-2 lines, respectively. (B) Results from densitometric analyses of the RT-PCR data from three experiments. The band intensity for each of the two transgenic lines was determined and used to calculate the mean intensity for each band. The results were expressed as a percentage of the WT control (100%) ±SD. View largeDownload slide Fig. 4 Stomatal openings in WT and antiAtGLR1.1 lines. (A) Representative results of Arabidopsis stomata in WT and antiAtGLR1.1 plants. The scale bar represents 10 µm. (B) Stomtal apertures of WT and antiAtGLR1.1 plants. Stomatal openings, between guard cells, were observed in the early morning. 1 and 2 represent antiAtGLR1.1-1 and 1-2 lines, respectively. Each large solid bar indicates a mean of 30–36 measurements with the standard deviations represented by the cross bar. View largeDownload slide Fig. 4 Stomatal openings in WT and antiAtGLR1.1 lines. (A) Representative results of Arabidopsis stomata in WT and antiAtGLR1.1 plants. The scale bar represents 10 µm. (B) Stomtal apertures of WT and antiAtGLR1.1 plants. Stomatal openings, between guard cells, were observed in the early morning. 1 and 2 represent antiAtGLR1.1-1 and 1-2 lines, respectively. Each large solid bar indicates a mean of 30–36 measurements with the standard deviations represented by the cross bar. View largeDownload slide Fig. 5 Drought response in WT and antiAtGLR1.1 lines. (A) WT and antiAtGLR1.1 plants after 16 d of drought treatment. Both WT and antiAtGLR1.1 plants were grown and subirrigated for 21 d, and then drought stressed by termination of irrigation for 16 d. The figure shows one representative pot (out of five) for each line. (B) Estimate of water loss by transpiration over time. Soil water content was measured for each line by weighting the plants and pots every other day during the drought treatment. Details of the experiment can be obtained in Materials and Methods. Three pots were weighed for each time point. Each treatment consists of five plants per pot (n = 3). Small bars represent the SD. The symbols are as follows: WT, solid diamonds; antiAtGLR1.1-1, open triangles and antiAtGLR1.1-2, solid squares. View largeDownload slide Fig. 5 Drought response in WT and antiAtGLR1.1 lines. (A) WT and antiAtGLR1.1 plants after 16 d of drought treatment. Both WT and antiAtGLR1.1 plants were grown and subirrigated for 21 d, and then drought stressed by termination of irrigation for 16 d. The figure shows one representative pot (out of five) for each line. (B) Estimate of water loss by transpiration over time. Soil water content was measured for each line by weighting the plants and pots every other day during the drought treatment. Details of the experiment can be obtained in Materials and Methods. Three pots were weighed for each time point. Each treatment consists of five plants per pot (n = 3). Small bars represent the SD. The symbols are as follows: WT, solid diamonds; antiAtGLR1.1-1, open triangles and antiAtGLR1.1-2, solid squares. View largeDownload slide Fig. 6 Proposed role of AtGLR1.1 in the regulation of ABA biosynthesis and sensing, and the physiological consequences of those events. AtGLR1.1 is negatively affected (Inhibitory) by Suc and positively affected (Stimulatory) by N (Kang and Turano 2003). Negative signals (–), either by antisense, Suc or DNQX, which result in decreased accumulation of the HXK1 transcript, elevated ABA titers in developing seedlings and leaves of mature Arabidopsis, increased accumulation of transcripts associated with ABA biosynthesis (ABA1, ABA2, ABA3 and AAO3) and altered levels of the transcripts associated with ABA sensing (ABI1, ABI2, ABI4 and ABI5). Combined, these events result in physiological changes such as inhibition of seed germination, primary root length and stomatal opening. The latter results in decreased water loss. N, in the form of amino acids, specifically Glu or Gln, positively (+) affects AtGLR1.1 (oval) and restores seed germination even in the presence of Suc or DNQX (Kang and Turano 2003). Large straight solid arrows indicate activation, and solid truncated lines indicate inactivation of genes or processes. Thick dashed lines represent multiple-step processes. Thin dotted lines represent potential signals or pathways that may be mediated through AtGLR1.1. Small arrows before the gene names indicate increased (up) or decreased (down) transcript accumulation. View largeDownload slide Fig. 6 Proposed role of AtGLR1.1 in the regulation of ABA biosynthesis and sensing, and the physiological consequences of those events. AtGLR1.1 is negatively affected (Inhibitory) by Suc and positively affected (Stimulatory) by N (Kang and Turano 2003). Negative signals (–), either by antisense, Suc or DNQX, which result in decreased accumulation of the HXK1 transcript, elevated ABA titers in developing seedlings and leaves of mature Arabidopsis, increased accumulation of transcripts associated with ABA biosynthesis (ABA1, ABA2, ABA3 and AAO3) and altered levels of the transcripts associated with ABA sensing (ABI1, ABI2, ABI4 and ABI5). Combined, these events result in physiological changes such as inhibition of seed germination, primary root length and stomatal opening. The latter results in decreased water loss. N, in the form of amino acids, specifically Glu or Gln, positively (+) affects AtGLR1.1 (oval) and restores seed germination even in the presence of Suc or DNQX (Kang and Turano 2003). Large straight solid arrows indicate activation, and solid truncated lines indicate inactivation of genes or processes. Thick dashed lines represent multiple-step processes. Thin dotted lines represent potential signals or pathways that may be mediated through AtGLR1.1. Small arrows before the gene names indicate increased (up) or decreased (down) transcript accumulation. Abbreviations C carbon DNQX 6,7-dinitroquinoxaline-2,3-[1H,4H]-dione N nitrogen NO nitric oxide NR nitrate reductase AtNOS1 nitric oxide synthase. References Arenas-Huertero, F., Arroyo, A., Zhou, L., Sheen, J. and Leon, P. ( 2000) Analysis of Arabidopsis glucose insensitive mutants, gin5 and gin6, reveals a central role of the plant hormone ABA in the regulation of plant vegetative development by sugar. Genes Dev.  14: 2085–2096. Google Scholar Arroyo, A., Bossi, F., Finkelstein, R.R. and Leon P. ( 2003) Three genes that affect sugar sensing (abscisic acid insensitive 4, abscisic acid insensitive 5, and constitutive triple response 1) are differentially regulated by glucose in Arabidopsis. Plant Physiol.  133: 231–242. Google Scholar Below, B.E., Cazetta, J.O. and Seebauer, J.R. ( 2000) Carbon/nitrogen interactions during ear and kernel development of maize. In Physiology and Modeling Kernel Set in Maize, Special Publication No. 29. Edited by Westgate, M.E. and Boote, K.J. pp. 15–24. Crop Science Society of America, Madison, WI, U.S.A. Google Scholar Bittner, F., Oreb, M. and Mendel, R. ( 2001) ABA3 is a molybdenum cofoactor sulfurase required for activation of aldehyde oxidase and xanthine dehydrogenase in Arabidopsis thaliana. J. Biol. Chem.  276: 40381–40384. Google Scholar Bohnert, H.J. and Sheveleva, E. ( 1998) Plant stress adaptations—making metabolism move. Curr. Opin. Plant Biol.  1: 267–274. Google Scholar Brocard, I.M., Lynch, T.J. and Finkelstein, R.R. ( 2002) Regulation and role of the Arabidopsis abscisic acid-insensitive 5 gene in abscisic acid, sugar, and stress response. Plant Physiol.  129: 1533–1543. Google Scholar Cheng, W.H., Endo, A., Zhou, L., Penney, J., Chen, H.C., Arroyo, A., Leon, P., Nambara, E., Asami, T., Seo, M., Koshiba, T. and Sheen, J. ( 2002) A unique short-chain dehydrogenase/reductase in Arabidopsis glucose signaling and abscisic acid biosynthesis and functions. Plant Cell  14: 2723–2743. Google Scholar de Bruxelles, G.L., Peacock, W.J., Dennis, E.S. and Dolferus, R. ( 1996) Abscisic acid induces the alcohol dehydrogenase gene in Arabidopsis. Plant Physiol.  111: 381–391. Google Scholar Desikan, R., Griffiths, R., Hancock, J. and Neill, S. ( 2002) A new role for an old enzyme: Nitrate reductase mediated nitric oxide generation is required for abscisic acid-induced stomatal closure in Arabidopsis thaliana. Proc. Natl Acad. Sci. USA  99: 16314–16318. Google Scholar Dodd, I.C., Tan, L.P. and He, J. ( 2003) Do increases in xylem sap pH and/or ABA concentration mediate stomatal closure following nitrate deprivation? J. Exp. Bot.  54: 1281–1288. Google Scholar Finkelstein, R.R., Gampalas, S.S. and Rock, S. ( 2002) Abscisic acid signaling seeds and seedlings. Plant Cell  14: S15–S45. Google Scholar Finkelstein, R.R. and Gibson, S.I. ( 2001) ABA and sugar interactions regulating development: cross-talk or voices in a crowd? Curr. Opin. Plant Biol.  5: 26–32. Google Scholar Finkelstein, R.R. and Lynch, T.J. ( 2000) The Arabidopsis abscisic acid response gene ABI5 encodes a basic leucine zipper transcription factor. Plant Cell  12: 599–609. Google Scholar Finkelstein, R.R., Wang, M.L., Lynch, T.J., Rao, S. and Goodman, H.M. ( 1998). The Arabidopsis abscisic acid response locus ABI4 encodes an APETALA2 domain protein. Plant Cell  10: 1043–1054. Google Scholar Gamble, P.E. and Mullet, J.E. ( 1986) Inhibition of carotenoid accumulation and ABA biosynthesis in fluridone-treated dark grown barley. Eur. J. Biochem.  160: 117–121. Google Scholar Garciarrubio, A., Legaria, J.P and Covarrubias, A.A. ( 1997) Abscisic acid inhibits germination of mature Arabidopsis seeds by limiting the availability of energy and nutrients. Planta  203: 182–187. Google Scholar Gibson, S.I. ( 2000) Plant sugar-response pathways. Part of a complex regulatory web. Plant Physiol.  124: 1532–1539 Google Scholar Gosti, F., Beaudoin, N., Serizet, C., Webb, A.R.R., Vartanian, N. and Giraudat, J. ( 1999) ABI1 protein phosphatase 2C is a negative regulator of absicisic acid signaling. Plant Cell  11: 1897–1909. Google Scholar Guo, F.Q., Okamoto, M. and Crawford, N.M. ( 2003) Identification of a plant nitric oxide synthase gene involved in hormonal signaling. Science  302: 100–103. Google Scholar Guo, F.Q., Young, J. and Crawford, N.M. ( 2003) The nitrate transporter AtNRT1.1 (CHL1) functions in stomatal opening and contributes to drought susceptibility in Arabidopsis. Plant Cell  15: 107–117. Google Scholar Hetherington, A.M. ( 2001) Guard cell signaling. Cell  107: 711–714. Google Scholar Hoth, S., Morgante, M., Sanchez, J.P., Hanafey, M.K., Tingey, S.V. and Chua, N.H. ( 2002) Genome-wide gene expression profiling in Arabidopsis thaliana reveals new targets of abscisic acid and largely impaired gene regulation in the abi1-1 mutant. J. Cell Sci.  115: 4891–4900 Google Scholar Kang, J. and Turano, F.J. ( 2003) The putative glutamate receptor 1.1 (AtGLR1.1) functions as a regulator of carbon and nitrogen metabolism in Arabidopsis. Proc. Natl Acad. Sci. USA  100: 6872–6877. Google Scholar Kang, J.Y., Choi, H.I., Im, M.Y. and Kim, S.Y. ( 2002) Arabidopsis basic leucine zipper proteins that mediate stress-responsive absisic acid signaling. Plant Cell  14: 343–357. Google Scholar Koornneef, M., Bentsink, L. and Hilhorst, H. ( 2002) Seed dormancy and germination. Curr. Opin. Plant Biol.  5: 33–36. Google Scholar Koornneef, M. and Karssen, C.M. ( 1994) Seed dormancy and germination. InArabidopsis. Edited by Somerville, C. and Meyerowitz, E. pp. 313–334. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, U.S.A. Google Scholar Koornneef, M., Reuling, G. and Karssen, C.M. ( 1984) The isolation and characterization of abscisic acid-insensitive mutants of Arabidopsis thaliana. Physiol. Plant.  61: 377–383. Google Scholar Laby, R.J., Kincaid, M.S., Kim, D. and Gibson, S.I. ( 2000) The Arabidopsis sugar-insensitive mutants sis4 and sis5 are defective in abscisic acid synthesis and response. Plant J.  23: 587–596. Google Scholar Lang, V. and Palva, E.T. ( 1992) The expression of a rab-related gene, rab18, is induced by abscisic acid during the cold acclimation process of Arabidopsis thaliana (L) Heynh. Plant Mol. Biol.  20: 951–962 Google Scholar Leung, J. and Giraudat, J. ( 1998) Abscisic acid signal transduction. Annu. Rev. Plant Physiol. Plant Mol. Biol.  49: 199–222. Google Scholar Leung, J., Merlot, S. and Giraudat, J. ( 1997) Arabidopsis abscisic acid-insensitive 2 (ABI2) and ABI1 genes encode homologous protein phosphatases 2C involved in abscisic acid signal transduction. Plant Cell  9: 759–771. Google Scholar Lopez-Molina, L., Mongrad, S. and Chua, N.H. ( 2001) A postgermination developmental arrest checkpoint is mediated by abscisic acid and requires the ABI5 transcription factor in Arabidopsis. Proc. Natl Acad. Sci. USA  98: 4782–4787. Google Scholar Lovegrove, A. and Hooley, R. ( 2000) Gibberellin and abscisic acid signalling in aleurone. Trends Plant Sci.  5: 102–110. Google Scholar Lu, C., Han, M.H., Guevara-Garcia, A. and Fedoroff, N.V. ( 2002) Mitogen-activated protein kinase signaling in postgermination arrest of development by abscisic acid. Proc. Natl Acad. Sci. USA  99: 15812–15817. Google Scholar Mata, C.G. and Lamattina, L. ( 2001) Nitric oxide induces stomatal closure and enhances the adaptive plant responses against drought stress. Plant Physiol.  126: 1196–1204. Google Scholar Merlot, S., Gosti, F., Guerrier, D., Vavasseur, A. and Giraudat, J. ( 2001) The ABI1 and ABI2 protein phosphatases 2C act in a negative feedback regulatory loop of the abscisic acid signalling pathway. Plant J.  25: 295–303. Google Scholar Moreno-Fonseca, L.P. and Covarrubias, A.A. ( 2001) Downstream DNA sequences are required to modulate Pvlea-18 gene expression in response to dehydration. Plant Mol. Biol.  45: 501–515. Google Scholar Pei, Z.M., Kuchitsu, K., Ward, J.M., Schwarz, M. and Schroeder, J.I. ( 1997) Differential abscisic acid regulation of guard cell slow anion channels in Arabidopsis wild type and abi1 and abi2 mutants. Plant Cell  9: 409–423. Google Scholar Pei, Z.M., Ghassemian, M., Kwak, C.M., McCourt, P. and Schroeder, J.I. ( 1998) Role of farnesyltransferase in ABA regulation of guard cell anion channels and plant water loss. Science  282: 287–290. Google Scholar Price, J., Li, T.C., Kang, S.G., Na, J.K. and Jang, J.C. ( 2003) Mechanisms of glucose signaling during germination of Arabidopsis. Plant Physiol.  132: 1424–1438. Google Scholar Ritchie, S. and Gilroy, S. ( 1998) Abscisic acid signal transduction in the barley aleurone is mediated by phospholipase D activity. Proc. Natl Acad. Sci. USA  95: 2697–2702. Google Scholar Rabe, E. ( 1990) Stress physiology: The functional significance of the accumulation of nitrogen-containing compounds. J. Hort. Sci.  65: 231–243. Google Scholar Rock, C.D. and Zeevaart, J.A.D. ( 1991) The aba mutant of Arabidopsis thaliana is impaired in epoxy-cartenoid biosynthesis. Proc. Natl Acad. Sci. USA  88: 7496–7499. Google Scholar Rolland, F., Moore, B. and Sheen, J. ( 2002) Sugar sensing and signaling in plant. Plant Cell  14: S185–S205. Google Scholar Rook, F., Corke, F., Card, R., Munz, G., Smith, C. and Bevan, M.W. ( 2001) Impaired sucrose-induction mutants reveal the modulation of sugar-induced starch biosynthetic gene expression by abscisic acid signaling. Plant J.  26: 421–433. Google Scholar Schroeder, J.I., Allen, G.J., Hugouvieux, V., Kwak, J.M. and Waner, D. ( 2001) Guard cell signal transduction. Annu. Rev. Plant Physiol. Plant Mol. Biol.  52: 627–658. Google Scholar Seo, M., Peeters, A.J., Koiwai, H., Oritani, T., Marion-Poll, A., Zeevaart, J.A., Koornneef, M., Kamiya, Y. and Koshiba, T. ( 2000) The Arabidopsisaldehyde oxidase 3 (AAO3) gene product catalyzes the final step in abscisic acid biosynthesis in leaves. Proc. Natl Acad. Sci. USA  97: 12908–12913. Google Scholar Sheen, J. ( 1998) Mutational analysis of protein phosphatase 2C involved in abscisic acid signal transduction in higer plants. Proc. Natl Acad. Sci. USA  95: 975–980. Google Scholar Signora, L., De Smet, I., Foyer, C.H. and Zhang, H. ( 2001) ABA plays a central role in mediating the regulatory effects of nitrate on root branching in Arabidopsis. Plant J.  28: 655–662. Google Scholar Soderman, E.M., Brocard, I.M., Lynch, T.J. and Finkelstein, R.R. ( 2000) Regulation and function of the ArabidopsisABA-insensitive4 gene in seed and abscisic acid response signaling networks. Plant Physiol.  124: 1752–1765. Google Scholar Turano, F.J., Muhitch, M.J., Felker, F.C. and MacMahon, M.B. ( 2002) The putative glutamate receptor 3.2 from Arabidopsis thaliana (AtGLR3.2) is an integral membrane peptide that accumulates in rapidly growing tissues and persists in vascular-associated tissue. Plant Sci.  163: 43–51. Google Scholar Ullah, H., Chen, J.G., Wang, S. and Jones, A.M. ( 2002) Role of a heterotrimeric G protein in regulation of Arabidopsis seed germination. Plant Physiol.  129: 897–907. Google Scholar Vartanian, N., Marcotte, L. and Giraudat, J. ( 1994) Drought rhizogenesis in Arabidopsis thaliana. Plant Physiol.  104: 761–767. Google Scholar Wang, H., Qi, Q., Schorr, P., Cutler, A.J., Crosby, W. and Fowke, L.C. ( 1998) ICK, a cyclin-dependent protein kinase inhibitor from Arabidopsis thaliana interacts with both Cdc2a and CycD3 and its expression is induced by abscisic acid. Plant J.  15: 501–510. Google Scholar Xiong, L., Itshitani, M., Lee, H. and Zhu, J.K. ( 2001) The Arabidopsis LOS5/ABA3 locus encodes a molybdenum cofactor sulfurase and modulates cold stress- and osmotic stress-responsive gene expression. Plant Cell  13: 2063–2083. Google Scholar Zeevaart, J.A. and Creelman, R.A. ( 1988) Metabolism and physiology of abscisic acid. Annu. Rev. Plant Physiol. Plant Mol. Biol.  39: 439–473. Google Scholar Zhou, L., Jang, J.C., Jones, T.L. and Sheen, J. ( 1998) Glucose and ethylene signal transduction crosstalk revealed by an Arabidopsis glucose-insensitive mutant. Proc. Natl Acad. Sci. USA  95: 10294–10299. Google Scholar

Onizuka, Takuo;Endo, Sumiyo;Akiyama, Hideo;Kanai, Shozo;Hirano, Masahiko;Yokota, Akiho;Tanaka, Satoshi;Miyasaka, Hitoshi

doi: 10.1093/pcp/pch160pmid: 15564522

Abstract The operon encoding ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) in the cyanobacterium Synechococcus sp. PCC7002 contains three rbc genes, rbcL, rbcX and rbcS, in this order. Introduction of translational frameshift into the rbcX gene resulted in a significant decrease in the production of large (RbcL) and small (RbcS) subunits of the Rubisco protein in Synechococcus sp. PCC7002 and in Escherichia coli. To investigate the function of the rbcX gene product (RbcX), we constructed the expression plasmid for the rbcX gene and examined the effects of RbcX on the recombinant Rubisco production in Escherichia coli. The coexpression experiments revealed that RbcX had marked effects on the production of large and small subunits of Rubisco without any significant influence on the mRNA level of rbc genes and/or the post-translational assembly of the Rubisco protein. The present rbcX coexpression system provides a novel and useful method for investigating the Rubisco maturation pathway. (Received May 3, 2004; Accepted July 15, 2004) Introduction Oligomeric proteins produced in Escherichia coli cannot always assemble by themselves (Pelham 1988). In some cases, molecular chaperones facilitate production of soluble proteins and enhance the post-translational assembly of these polypeptides into oligomeric structures (Hemmingsen et al. 1988). Ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco), which catalyzes the CO2 fixation reaction in photosynthesis, is composed of eight large (RbcL) and eight small (RbcS) subunits in plants and most photosynthetic prokaryotes (Miziorko and Lorimer 1983). In higher plants, newly synthesized RbcL associates with Rubisco-binding protein (cpn60) (Barraclough and Ellis 1980, Ellis and van der Vies 1988), which is involved in assembly of the hexadecameric (L8S8) structure (Gatenby and Ellis 1990). The Rubisco-binding protein (cpn60) is homologous to the E. coli 60-kDa GroEL chaperonin (Hemmingsen et al. 1988), and the GroEL oligomer, like the Rubisco-binding protein, binds to newly synthesized plant RbcL (Gutteridge and Gatenby 1995). Rubisco from cyanobacteria is an L8S8-form enzyme and its genes are in the rbc operon in which one possible open reading frame, rbcX, is present in the intergenic space between the large (rbcL) and small (rbcS) subunit genes for Rubisco (Larimer and Soper 1993). It was first reported that the rbcX gene of the filamentous cyanobacterium Anabaena sp. strain PCC7120 does not influence the expression levels of recombinant Rubisco in E. coli under conditions favoring maximum E. coli GroEL and GroES synthesis (Larimer and Soper 1993). More recently, the rbcX gene of Anabaena sp. strain CA was found to affect the levels of recombinant Rubisco activity in E. coli strongly, when the chaperonin synthesis was not maximized (Li and Tabita 1997). This suggests chaperone-like functions of the rbcX gene product. We recently found that in other unicellular cyanobacteria, including Synechocystis sp. strain 6803 and Synechococcus sp. PCC7002 (Agmenellum quadruplicatum PR-6), the rbcX gene is juxtaposed with the rbcL and rbcS genes and is likely to be cotranscribed with the rbcL and rbcS genes (Onizuka et al. 2002). This suggests that RbcX is involved in the synthesis and the subsequent assembly of RbcL and RbcS in this organism. The aim of this study with genes from Synechococcus sp. PCC7002 is to gain insight into the mechanism of the involvement of RbcX in the synthesis and assembly of the Rubisco subunits. Results Effect of mutation in rbcX gene on RbcLS production We first examined the influence of the rbcX gene on production of RbcLS in Synechococcus sp. PCC7002. A 2.6-kb DNA fragment containing the rbc operon with a frameshift mutation in the rbcX gene was constructed (Fig. 1A). Fig. 1B shows the resulting amino acid sequences of the mutated RbcX. We transformed Synechococcus sp. PCC7002 wild-type cells using the construct and selected the mutant strain containing the partially inactivated rbcX gene. As shown in Fig. 1C, D, Western blotting revealed that the amount of soluble Rubisco proteins from the mutant strain was decreased compared with that from the wild-type cells. In the same manner, mutation of the rbcX gene lowered the Rubisco activity (Fig. 1E). Because a complete loss of the rbcX gene hampers the viability of cyanobacterial cells, we examined the effects of the rbcX gene on production of RbcLS in E. coli using the rbc expression plasmids. Two rbc expression plasmids, in which the rbc genes were placed under the control of the rbc promoter, were constructed. The one that could express rbcLXS completely was designated pRbcLXS, and the other in which the rbcX gene was inactivated by introducing a translational frameshift was named pRbcLS (Fig. 2). When the rbc genes were expressed in the strain JM109, a large part of the recombinant Rubisco was insoluble. The soluble Rubisco was hardly detected by SDS-PAGE, because its level was significantly lower than the insoluble one. However, subsequent immunoblotting revealed that the recombinant Rubisco was also produced in soluble proteins in the E. coli cells containing the plasmid pRbcLXS (Fig. 3A, B, lanes 1). When the rbcX frameshift plasmid (pRbcLS) was transformed into cells, RbcL and RbcS could hardly be detected in soluble proteins (Fig. 3A, B, lanes 2). Similarly, the disruption of the rbcX gene eliminated the Rubisco activity (Fig. 3C). Moreover, SDS-PAGE of corresponding insoluble proteins revealed that cells containing pRbcLS produced significantly less RbcLS than cells containing pRbcLXS (Fig. 3D). These results suggest that the rbcX gene is involved in a process of RbcLS production and/or subsequent assembly. We also examined the effect of the rbcX gene on mRNA levels of rbc genes in E. coli harboring the above plasmids. As shown in Fig. 3E, the mRNA levels of rbc genes were almost the same in cells containing pRbcLXS and pRbcLS, indicating that the rbcX gene had no effect on the transcription of rbc genes. Effect of RbcX protein on RbcLS production We constructed an rbcX-expression plasmid in which the gene was placed under the control of the rbc promoter and designated it pACYC-rbcX (Fig. 4A). This plasmid was a derivative of pACYC184 and carried the tetracycline resistance gene for selection. The pACYC-rbcX contained a p15A replicon and was compatible with pRbcLS. E. coli cells harboring the plasmid pACYC-rbcX synthesized the soluble RbcX detected by immunoblotting (Fig. 4B), indicating that the plasmid pACYC-rbcX could facilitate studies on the effect of the synthesized RbcX on RbcLS production. To examine the effects of the synthesized RbcX on the production of RbcLS, we used E. coli cells harboring a pair of compatible plasmids for expression of rbcX, pACYC-rbcX, and for expression of rbcLS, pRbcLS, in which the rbcX gene was inactivated by introducing a translational frameshift. As shown in Fig. 5, the cells containing the plasmid encoding the SynechococcusrbcLXS genes (pRbcLXS) produced a large amount of RbcLS (Fig. 5A, B, lanes 1), while the cells containing the pRbcLS lacking the active rbcX gene together with the control plasmid pACYC184 did not (Fig. 5A, B, lanes 2). However, when the rbcLS genes were coexpressed with the rbcX gene, RbcLS production was recovered, suggesting that the RbcX protein could effectively promote production of the RbcLS proteins (Fig. 5A, B, lanes 3). Similar results were obtained when the E. coli cells carrying pRbcLS were cotransformed with the expression plasmid for groE genes (pACYC-T7-GroE) instead of pACYC-rbcX (data not shown). Thus, the RbcX protein and the GroE chaperonins have similar effects on the production of the RbcLS proteins. Similar coexpression experiments were carried out by using E. coli cells harboring a pair of compatible plasmids pACYC-rbcX and pT7LS carrying the inactive rbcX gene whose expression is controlled by the T7 promoter. The effect of the synthesized RbcX is similar to that shown in Fig. 5, although there was a decrease in the yield (data not shown). These results indicate that the RbcX protein could effectively promote production of RbcLS proteins independently of the promoter controlling rbcLS expression. Effect of the RbcX protein on Rubisco assembly To investigate whether the synthesized RbcX promotes the assembly of RbcL and RbcS into a L8S8 form, coexpression experiments were carried out by using E. coli cells harboring a pair of compatible plasmids pACYC-rbcX and pRbcLS. By expressing the rbcX gene with the plasmid pACYC-rbcX, the formation of L8S8 was achieved as revealed by non-denaturing PAGE and immunoblotting (Fig. 6, lane 2). The recombinant L8S8 unit appeared to be identical to the control Rubisco holoenzyme (Fig. 6, lanes 1, 2). On the other hand, the E. coli cells harboring a combination of plasmids pACYC184 (control) and pRbcLS revealed incomplete assembly of the Rubisco holoenzyme (Fig. 6, lane 3). These results indicate that the RbcX protein is involved in the post-translational assembly of RbcLS into a L8S8 holoenzyme. Discussion As a prerequisite to the study of the function of cyanobacterial RbcX, we constructed the coexpression system for the rbc genes from Synechococcus sp. PCC7002 in E. coli. When the recombinant RbcX was synthesized in the E. coli strain, the protein was soluble enough to examine the role of RbcX in production and assembly of the Synechococcus Rubisco proteins. In the present study with the rbcX gene from Synechococcus sp. PCC7002, we found that the introduction of a translational frameshift into the rbcX gene resulted in a significant decrease in the production of large (RbcL) and small (RbcS) subunits of the Rubisco protein in Synechococcus sp. PCC7002 and in E. coli (Fig. 1, 3). mRNA levels of rbcL and rbcS were not affected by the inactivation of the rbcX gene (Fig. 3). Moreover, when the rbcX gene was coexpressed with the rbcLS genes in E. coli, the synthesized RbcX increased the amounts of RbcL and RbcS (Fig. 5), and the L8S8 structure of Rubisco protein was apparently formed (Fig. 6). These findings suggest that the product of the rbcX gene is not a transcription activator, but promotes the RbcLS protein production without regulating the mRNA levels of Rubisco genes, and that the RbcX protein is involved in the synthesis and/or the post-translational assembly of the Synechococcus Rubisco proteins. The difficulty in overexpressing cyanobacterial Rubisco to yield the active enzyme in E. coli cells was overcome by simultaneous oversynthesis of E. coli chaperonin proteins (GroEL and GroES) and Rubisco (Gurevitz et al. 1985, Goloubinoff et al. 1989, Larimer and Soper 1993). In the presence of an excess amount of chaperonin proteins, AnabaenarbcX (Anabaena sp. strain PCC7120) had little effect on the levels of Rubisco activity (Larimer and Soper 1993). On the other hand, the intact rbcX gene of Anabaena sp. strain CA retained the activity of recombinant Rubisco in E. coli limiting the level of chaperonins (Li and Tabita 1997). In the present study, the recombinant RbcX protein promoted the production of RbcLS and the formation of the L8S8 structure of the Rubisco protein, indicating that the low activity of Rubisco observed in other laboratories was due to a failure in the formation of the L8S8-form enzyme of Rubisco in the absence of RbcX. Only the expression of rbcLS was insufficient for the production of RbcLS and assembly into complete oligomeric form in E. coli expressing endogenous chaperonins only. The product of the rbcX gene was required for the production of the Rubisco protein and formation of the L8S8 structure. In the E. coli cellular environment, the RbcX protein had effects apparently similar to those of GroEL and GroES chaperonins in promoting the production of RbcLS protein and assembly of RbcLS into the L8S8 structure, in agreement with the conclusion drawn first by Goloubinoff et al. (1989) using Anacystis Rubisco. Since inactivation of the rbcX gene significantly abolished the production of RbcLS proteins both in supernatant and in pellet fractions in our case, RbcX may have functions in translation of transcripts of rbcLS and/or folding of nascent peptides, preventing the Rubisco proteins from degradation rather than from aggregation. The proposed maturation pathway of Rubisco consists of polypeptide translation, folding, L2 dimerization, tetramerization of L2 dimers into L8 octamers and association of small subunits with the L8 octomers into the hexadecameric L8S8 holoenzyme (Goloubinoff et al. 1989, Fitchen et al. 1990, Lee and Tabita 1990, Paul et al. 1991). Experimentally, we have shown that RbcX is required for assembly of the Rubisco protein in E. coli and possibly in the cyanobacterium, but we have not specified the steps in which the RbcX protein is involved. However, we have not yet determined the chaperonin dependence and efficacy of the RbcX protein for assembly of the L8S8 structure of Rubisco in Synechococcus. Much more work is required to determine the role of RbcX in the maturation of Rubisco in E. coli and also in Synechococcus itself. The rbcX coexpression system described here will provide a useful way to study the translation, folding and assembly of Rubisco and prompts us to investigate whether this rbcX coexpression system is able to improve production of various foreign proteins, including plant Rubisco in E. coli cells. Materials and Methods Materials and strains Restriction enzymes were purchased from Takara Bio (Kyoto, Japan). The DNA sequencing kits were from Applied Biosystems (Norwalk, CT, U.S.A.). All chemicals were purchased from Nacalai Tesque (Kyoto, Japan). The E. coli strain JM109 purchased from Takara Bio was used throughout this study and cultured at 37°C in LB medium (Sambrook et al. 1989). A groE expression vector pACYC-T7-GroE, which harbored the chloramphenicol resistance cassette, was a kind gift from Dr. Hiroshi Yamamoto of the Research Institute of Innovative Technology for the Earth (RITE), Kyoto, Japan. The cyanobacterial strain Synechococcus sp. PCC7002 (A. quadruplicatum PR-6, ATCC 27264) (Buzby et al. 1983, Buzby et al. 1985), obtained from the American Type Culture Collection, was cultured at 30°C in medium A (Tabita et al. 1974) under aeration with 1% CO2. Continuous illumination was provided at 50 µmol photons m–2 s–1 by three FL40SS (37W) fluorescent lamps. Agar plates were prepared using medium A solidified with 1.5% agar and the cells were cultured at 30°C. Plasmid constructions The rbc expression vector, pRbcLXS was made by cloning the rbc operon into pAQJ4-MCS (Akiyama et al. 1998, Akiyama et al. 1999) using the PCR technique. A 2.6-kb fragment in the rbc operon was amplified with Pfx DNA polymerase (Invitrogen, Carlsbad, CA, U.S.A.) for 25 cycles of 94°C denaturation for 15 s, 55°C annealing for 30 s and 68°C extension for 3 min with a final extension time of 3 min. A PCR product including the rbc promoter and rbcLXS was digested with BamHI and XbaI, and cloned into the BamHI and XbaI sites on pAQJ4-MCS. Plasmid pRbcLS was derived from a partial SacI digestion of pRbcLXS, treated with T4 DNA polymerase. The resulting repair of the SacI site located at codon 66 of rbcX produced a frameshift mutation. A 2.3-kb PCR product without the rbc promoter was amplified from pRbcLS as a template, digested with NdeI and XhoI and inserted into the NdeI and XhoI sites on pIVEX2.4b-Nde (Roche Diagnostics GmbH, Penzberg, Germany), constructing plasmid pT7LS. To generate the construct for inactivation of the rbcX gene in Synechococcus sp. PCC7002, a 2.6-kb DNA fragment containing the rbc operon was amplified by PCR from pRbcLS with a frameshift mutation in the rbcX gene. The PCR-amplified DNA fragment was cloned into plasmid pUC18 and a 1.3-kb kanamycin resistance cassette was inserted into a SalI site created between the rbcL and rbcX genes to select the mutant strain. The resulting plasmid was linearized by digestion with BamHI and XbaI, and was used to transform Synechococcus sp. PCC7002 wild-type cells. Plasmid pACYC-rbcX was derived from plasmid pACYC184 (Nippon Gene, Tokyo, Japan) containing a p15A replicon and tetR repressor gene. A 250-bp NcoI–EcoRI fragment containing the rbc promoter and a 400-bp NcoI fragment containing the rbcX gene, amplified by PCR, were inserted into pACYC184 cut with NcoI/EcoRI. All the resulting vectors were confirmed by DNA sequence analysis using a 3100 DNA sequencer (Applied Biosystems). Protein extraction and Rubisco assay The E. coli strains harboring a pair of compatible expression plasmids were grown in LB medium containing 100 µg ml–1 of ampicillin and 25 µg ml–1 of tetracycline or chloramphenicol at 37°C with constant aeration for 6–8 h. The cells were harvested, and whole-cell proteins or insoluble proteins from the same cell number were analyzed by SDS-PAGE and Western blotting. Cells were resuspended in phosphate-buffered saline (PBS) to examine the soluble proteins. Protein extracts were prepared using a FastPrep system (Qbiogene, Carlsbad, CA, U.S.A.) as described by the manufacturer, and centrifuged at 10,000×g for 15 min at 4°C. The activity of Rubisco in an aliquot of the supernatant was determined using the spectrophotometric, enzyme-coupled assay developed by Racker (1962). We defined the specific activity of Rubisco as the amount of enzyme that catalyzes the carboxylation of 1.0 nmol of ribulose bisphosphate per min per mg of total proteins. Protein concentrations were determined by the method of Bradford (1976) with bovine serum albumin as the standard. Western blot analysis An antiserum raised against Synechococcus RbcX was prepared as follows. The PCR-generated DNA fragment, containing the rbcX-coding region, was cloned into the NcoI site of pIVEX2.4b-Nde, which is the N-terminal hexa-His-containing vector. The sequence of the resulting plasmid was verified by DNA sequencing. The expression experiment of the rbcX gene was performed using the RTS 500 instrument and the RTS 500 E. coli HY Kit (Roche Diagnostics GmbH) according to the supplier’s instructions (Martin et al. 2001). The reaction was run with a stirring rate of 120 rpm, at 30°C for 20 h. The content of the reaction mixture (1 ml) was diluted to 10 ml with PBS and purified using B-PER 6 × His spin purification kit (Pierce, Rockfold, IL, U.S.A.). The purity and concentration of recombinant RbcX were estimated using SDS-PAGE, and an antiserum against the purified recombinant RbcX was raised in rabbits. Whole-cell proteins or soluble proteins (1.3 µg) were separated on a 4–20% gradient SDS-polyacrylamide gel. In the case of non-denaturing PAGE, soluble proteins (3.3 µg) were applied on a 4–12% gradient polyacrylamide gel. Separated proteins were transferred onto a nitrocellulose membrane. Anti-RbcL, anti-RbcS (Onizuka et al. 2003) or anti-RbcX antibodies (1 : 10,000 dilution) and anti-rabbit IgG antibody conjugated with horseradish peroxidase (1 : 10,0000 dilution, Santa Cruz Biotechnology, Santa Cruz, CA, U.S.A.) were used as primary and secondary antibodies, respectively. Immunodetection was performed by an enhanced chemiluminescence method as recommended by the manufacturer (Amersham Biosciences, Piscataway, NJ, U.S.A.). RT-PCR Total RNA was extracted from E. coli strains harboring the indicated plasmids using ISOGEN (Nippon Gene). Contaminating DNA was removed completely using a DNA-free kit (Ambion, Austin, TX, U.S.A.). The quality and quantity of RNA were examined by conventional 1% agarose gel electrophoresis and spectrophotometric measurement. Aliquots of total RNA (1.0 µg) were amplified using 0.4 µM of a set of oligonucleotide primers, 1 mM of dNTP-analog mixture (Takara Bio), 0.8 U µl–1 of RNase inhibitor (Takara Bio), 0.1 U µl–1 of AMV Rtase XL (Takara Bio) and 0.1 U µl–1 of AMV-optimized Taq polymerase (Takara Bio) in 50 µl of reaction mixture. A set of oligonucleotide primers for Synechococcus PCC7002 rbc genes were synthesized based on the following nucleotide sequences: rbc sense primer 5′-CCC TCA GCG ACC AGC AAA TC-3′: rbc antisense primer 5′-ACG GGT TTG GTT GGG CTT GT-3′. The size of the fragment amplified was 278 bp. Reactions were aliquoted from a master mix to minimize tube to tube variation. After cDNA synthesis at 50°C for 30 min, PCR amplification was conducted by the protocol of 85°C denaturation for 40 s, 55°C annealing for 40 s and 72°C extension for 1 min. The mRNA levels were compared at a logarithmic amplification of 25 cycles. Amplification products were subjected to electrophoresis with 2.0% agarose gels and stained with ethidium bromide. Acknowledgments We are grateful to Mrs. Mina Usui-Takeshige for DNA sequencing analysis. 4 Corresponding author: E-mail, [email protected]; Fax, +81-467-320414. View largeDownload slide Fig. 1 Effect of partial inactivation of the rbcX gene on RbcLS production in Synechococcus sp. PCC7002. (A) Scheme for the frameshifted inactivation of the rbcX gene. A 1.3-kb kanamycin resistance cassette was inserted at a SalI site created between the rbcL and rbcX genes for selection. Wild-type cells were transformed with this construct to generate the strain containing the partially inactivated rbcX gene. (B) Amino acid sequences of the wild-type RbcX (RbcX) and the mutated RbcX (*RbcX). Translation frameshift was introduced from the SacI site located at codon 66 of the rbcX gene. Identical amino acids between RbcX and *RbcX are indicated in white on a black background. (C) SDS-PAGE of cell lysates, revealed by immunodetection with the antiserum raised against RbcL. Wild-type cells (lane 1) or cells containing the partially inactivated rbcX gene (lane 2) were grown to mid-log phase. Cells were harvested, disrupted and separated into soluble fractions. Soluble proteins were analyzed by Western blot analyses as described in Materials and Methods. (D) SDS-PAGE of cell lysates revealed by immunodetection with an antiserum raised against RbcS. The experiment was carried out as described in (C). (E) Rubisco activities of cell lysates from wild-type cells (1) or cells containing the mutated rbcX gene (2). Data represent mean ± SD obtained from three independent measurements. *Translation frameshift. View largeDownload slide Fig. 1 Effect of partial inactivation of the rbcX gene on RbcLS production in Synechococcus sp. PCC7002. (A) Scheme for the frameshifted inactivation of the rbcX gene. A 1.3-kb kanamycin resistance cassette was inserted at a SalI site created between the rbcL and rbcX genes for selection. Wild-type cells were transformed with this construct to generate the strain containing the partially inactivated rbcX gene. (B) Amino acid sequences of the wild-type RbcX (RbcX) and the mutated RbcX (*RbcX). Translation frameshift was introduced from the SacI site located at codon 66 of the rbcX gene. Identical amino acids between RbcX and *RbcX are indicated in white on a black background. (C) SDS-PAGE of cell lysates, revealed by immunodetection with the antiserum raised against RbcL. Wild-type cells (lane 1) or cells containing the partially inactivated rbcX gene (lane 2) were grown to mid-log phase. Cells were harvested, disrupted and separated into soluble fractions. Soluble proteins were analyzed by Western blot analyses as described in Materials and Methods. (D) SDS-PAGE of cell lysates revealed by immunodetection with an antiserum raised against RbcS. The experiment was carried out as described in (C). (E) Rubisco activities of cell lysates from wild-type cells (1) or cells containing the mutated rbcX gene (2). Data represent mean ± SD obtained from three independent measurements. *Translation frameshift. View largeDownload slide Fig. 2 Structure of rbc expression plasmids used in this study. pRbcLXS contains the rbc promoter followed by SynechococcusrbcL, rbcX and rbcS genes cloned in pAQJ4-MCS (Akiyama et al. 1998, Akiyama et al. 1999). pRbcLS contains a frameshifted rbcX gene between rbcL and rbcS genes under the control of the rbc promoter cloned in pAQJ4-MCS. *Translation frameshift. View largeDownload slide Fig. 2 Structure of rbc expression plasmids used in this study. pRbcLXS contains the rbc promoter followed by SynechococcusrbcL, rbcX and rbcS genes cloned in pAQJ4-MCS (Akiyama et al. 1998, Akiyama et al. 1999). pRbcLS contains a frameshifted rbcX gene between rbcL and rbcS genes under the control of the rbc promoter cloned in pAQJ4-MCS. *Translation frameshift. View largeDownload slide Fig. 3 Expression of rbc genes in E. coli cells containing rbc expression plasmid pRbcLXS (lanes 1) or pRbcLS (lanes 2). (A) SDS-PAGE of cell lysates revealed by immunodetection with an antiserum raised against RbcL. Cells were grown in LB medium under selective conditions. Cultures were grown to mid-log phase. (B) SDS-PAGE of cell lysates revealed by immunodetection with the antiserum raised against RbcS. (C) Rubisco activities of cell lysates. Data represent mean ± SD obtained from three independent measurements. (D) SDS-PAGE of insoluble fractions. The gel was stained with Coomassie brilliant blue. (E) Expression of the rbc mRNA revealed by RT-PCR using rbc primers. View largeDownload slide Fig. 3 Expression of rbc genes in E. coli cells containing rbc expression plasmid pRbcLXS (lanes 1) or pRbcLS (lanes 2). (A) SDS-PAGE of cell lysates revealed by immunodetection with an antiserum raised against RbcL. Cells were grown in LB medium under selective conditions. Cultures were grown to mid-log phase. (B) SDS-PAGE of cell lysates revealed by immunodetection with the antiserum raised against RbcS. (C) Rubisco activities of cell lysates. Data represent mean ± SD obtained from three independent measurements. (D) SDS-PAGE of insoluble fractions. The gel was stained with Coomassie brilliant blue. (E) Expression of the rbc mRNA revealed by RT-PCR using rbc primers. View largeDownload slide Fig. 4 Synthesis of RbcX in E. coli cells containing the rbcX expression plasmid pACYC-rbcX. (A) Structure of the rbcX expression plasmid; p15A, p15A replicon; tetR, tetR repressor gene; rbcP, rbc promoter; rbcX, rbcX gene. (B) Synthesis of RbcX in E. coli cells containing pACYC-rbcX. Cells were grown in LB medium containing 25 µg ml–1 of tetracycline at 37°C. Cultures were grown to mid-log phase and soluble proteins were analyzed by SDS-PAGE, revealed by immunodetection with an antiserum raised against recombinant RbcX. View largeDownload slide Fig. 4 Synthesis of RbcX in E. coli cells containing the rbcX expression plasmid pACYC-rbcX. (A) Structure of the rbcX expression plasmid; p15A, p15A replicon; tetR, tetR repressor gene; rbcP, rbc promoter; rbcX, rbcX gene. (B) Synthesis of RbcX in E. coli cells containing pACYC-rbcX. Cells were grown in LB medium containing 25 µg ml–1 of tetracycline at 37°C. Cultures were grown to mid-log phase and soluble proteins were analyzed by SDS-PAGE, revealed by immunodetection with an antiserum raised against recombinant RbcX. View largeDownload slide Fig. 5 Effect of synthesized RbcX on RbcLS production. Strain JM109 harboring the rbc expression plasmid pRbcLXS (lanes 1), rbc expression plasmid pRbcLS together with pACYC184 (lanes 2), or rbc expression plasmid pRbcLS together with pACYC-rbcX (lanes 3) were grown in LB medium under selective conditions. Cultures were grown to mid-log phase and harvested. Whole-cell proteins were subjected to SDS-PAGE and detected with an antiserum raised against RbcL (A) or RbcS (B). View largeDownload slide Fig. 5 Effect of synthesized RbcX on RbcLS production. Strain JM109 harboring the rbc expression plasmid pRbcLXS (lanes 1), rbc expression plasmid pRbcLS together with pACYC184 (lanes 2), or rbc expression plasmid pRbcLS together with pACYC-rbcX (lanes 3) were grown in LB medium under selective conditions. Cultures were grown to mid-log phase and harvested. Whole-cell proteins were subjected to SDS-PAGE and detected with an antiserum raised against RbcL (A) or RbcS (B). View largeDownload slide Fig. 6 Effect of synthesized RbcX on Rubisco assembly. Strain JM109 co-transformed with plasmids pRbcLS and pACYC-rbcX (lane 2), or plasmids pRbcLS and pACYC184 (lane 3) were grown in LB medium under selective conditions. Cultures were grown to mid-log phase. Cells were harvested, disrupted and separated into soluble fractions. Soluble proteins were analyzed by non-denaturing PAGE with the endogenous Rubisco protein from Synechococcus sp. PCC7002 (lane 1) and Rubisco detection was performed by Western blot analysis as described in Materials and Methods. View largeDownload slide Fig. 6 Effect of synthesized RbcX on Rubisco assembly. Strain JM109 co-transformed with plasmids pRbcLS and pACYC-rbcX (lane 2), or plasmids pRbcLS and pACYC184 (lane 3) were grown in LB medium under selective conditions. Cultures were grown to mid-log phase. Cells were harvested, disrupted and separated into soluble fractions. Soluble proteins were analyzed by non-denaturing PAGE with the endogenous Rubisco protein from Synechococcus sp. PCC7002 (lane 1) and Rubisco detection was performed by Western blot analysis as described in Materials and Methods. Abbreviations Rubisco ribulose-1,5-bisphosphate carboxylase/oxygenase PBS phosphate buffered saline RT reverse transcription. References Akiyama, H., Kanai, S., Hirano, M. and Miyasaka, M. ( 1998) Nucleotide sequence of plasmid pAQ1 of marine cyanobacterium Synechococcus sp. PCC7002. DNA Res.  5: 127–129. Google Scholar Akiyama, H., Kanai, S., Hirano, M. and Miyasaka, M. ( 1999) A novel plasmid recombination mechanism in the marine cyanobacterium Synechococcus sp. PCC7002. DNA Res.  5: 327–334. Google Scholar Barraclough, R. and Ellis, R.J. ( 1980) Protein synthesis in chloroplasts. IX. Assembly of newly-synthesized large subunits into ribulose bisphosphate carboxylase in isolated pea chloroplasts. Biochim. Biophys. Acta  608: 19–31. Google Scholar Bradford, M.M. ( 1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem.  72: 248–254. Google Scholar Buzby, J.S., Porter, R.D. and Stevens, S.E. Jr. ( 1983) Plasmid transformation in Agmenellumquadruplicatum PR-6: Construction of biphasic plasmids and characterization of their transformation properties. J. Bacteriol.  154: 1446–1450. Google Scholar Buzby, J.S., Porter, R.D. and Stevens, S.E. Jr. ( 1985) Expression of the Escherichia colilacZ gene on a plasmid vector in a cyanobacterium. Science  230: 805–807. Google Scholar Ellis, R.J. and van der Vies, S.M. ( 1988) The Rubisco subunit binding protein. Photosynth. Res.  16: 101–115. Google Scholar Fitchen, J.H., Knight, S., Andersson, I., Branden, C.I. and Mcintosh, L. ( 1990) Residues in three conserved regions of the small subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase are required for quaternary structure. Proc. Natl Acad. Sci. USA  87: 5768–5772. Google Scholar Gatenby, A.A. and Ellis, R.J. ( 1990) Chaperone function: the assembly of ribulose bisphosphate carboxylase-oxygenase. Annu. Rev. Cell. Biol.  6: 125–149. Google Scholar Goloubinoff, P., Gatenby, A.A. and Lorimer, G.H. ( 1989) GroE heat-shock proteins promote assembly of foreign prokaryotic ribulose bisphosphate carboxylase oligomers in Escherichia coli. Nature  337: 44–47. Google Scholar Gurevitz, M., Somerville, C.R. and Mcintosh, L. ( 1985) Pathway of assembly of ribulosebisphosphate carboxylase/oxygenase from Anabaena 7120 expressed in Escherichia coli. Proc. Natl Acad. Sci. USA  82: 6546–6550. Google Scholar Gutteridge, S. and Gatenby, A.A. ( 1995) Rubisco synthesis, assembly, mechanism, and regulation. Plant Cell  7: 809–819. Google Scholar Hemmingsen, S.M., Woolford, C., van der Vies, S.M., Tilly, K., Dennis, D.T., Georgopoulos, C.P., Hendrix, R.W. and Ellis, R.J. ( 1988) Homologous plant and bacterial proteins chaperone oligomeric protein assembly. Nature  333: 330–334. Google Scholar Larimer, F.W. and Soper, T.S. ( 1993) Overproduction of Anabaena 7120 ribulose-bisphosphate carboxylase-oxygenase in Escherichia coli. Gene  126: 85–92. Google Scholar Lee, B. and Tabita, F.R. ( 1990) Purification of recombinant ribulose-1,5-bisphosphate carboxylase/oxygenase large subunits suitable for reconstitution and assembly of active L8S8 enzyme. Biochemistry  29: 9352–9357. Google Scholar Li, L.A. and Tabita, F.R. ( 1997) Maximum activity of recombinant ribulose 1,5-bisphosphate carboxylase/oxygenase of Anabaena sp. strain CA requires the product of the rbcX gene. J. Bacteriol.  179: 3793–3796. Google Scholar Sambrook, J., Fritsch, E.F. and Maniatis, T. ( 1989) Molecular Cloning: A Laboratory Manual, 2nd edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Google Scholar Martin, G.A., Kawaguchi, R., Lam, Y., DeGiovanni, A., Fukushima, M. and Mutter, W. ( 2001) High-yield, in vitro protein expression using a continuous-exchange, coupled transcription/translation system. Biotechniques  31: 948–953. Google Scholar Miziorko, H.M. and Lorimer, G.H. ( 1983) Ribulose-1,5-bisphosphate carboxylase-oxygenase. Annu. Rev. Biochem.  52: 507–535. Google Scholar Onizuka, T., Akiyama, H., Endo, S., Kanai, S., Hirano, M., Tanaka, S. and Miyasaka, M. ( 2002) CO2 response element and corresponding trans-acting factor of the promoter for ribulose-1,5-bisphosphate carboxylase/oxygenase genes in Synechococcus sp. PCC7002 found by an improved electrophoretic mobility shift assay. Plant cell Physiol.  43: 660–667. Google Scholar Onizuka, T., Akiyama, H., Endo, S., Kanai, S., Hirano, M., Tanaka, S. and Miyasaka, M. ( 2003) CO2 response for expression of ribulose-1,5-bisphosphate carboxylase/oxygenase genes is inhibited by AT-rich decoy in the cyanobacterium. FEBS Lett.  542: 42–46. Google Scholar Paul, K., Morell, M.K. and Andrews, T.J. ( 1991) Mutations in the small subunit of ribulosebisphosphate carboxylase affect subunit binding and catalysis. Biochemistry  30: 10019–10026. Google Scholar Pelham, H. ( 1988) Coming in from the cold. Nature  332: 776–777. Google Scholar Racker, E. ( 1962) Methods in Enzymology  5: p. 266. Academic Press, New York. Google Scholar Tabita, E.R., Stevens, S.E. Jr. and Quiano, R. ( 1974) D-Riburose 1, 5-diphosphate carboxylase from blue-green algae. Biochem. Biophys. Res. Commun.  61: 45–52. Google Scholar

Mól, Rafal;Filek, Maria;Machackova, Ivana;Matthys-Rochon, Elisabeth

doi: 10.1093/pcp/pch167pmid: 15564523

Abstract The role of ethylene and auxin in stigma-to-ovule signalling was investigated in maize (Zea mays L.). Maturation of the egg cells in an ear was stimulated before actual fertilization by the application of fresh pollen grains or quartz sand to fully receptive stigmas. Ethylene emission by maize ears increased in response to those treatments. Silks and ovaries were involved in ethylene synthesis after pollen or sand was shed over the silks. The content of ethylene precursor [1-aminocyclopropane-1-carboxylic acid (ACC)] increased in both pistil parts soon after pollination. ACC rise was delayed by 4 h in the ovaries, and by 8 h in the silks after mock-pollination with sand. The auxin level increased rapidly in the silks and ovaries after pollination, and it was very high in the pollinated silks due to the high indole-3-acetic acid (IAA) content of pollen grains. IAA rise also appeared in the silks and ovaries after treatment with sand but it was delayed by 8 h. Application of ACC (10 µM) or IAA (6 µM) solutions to non-pollinated silks stimulated maturation of the egg cells. Moreover, the response of the egg cells to pollination was cancelled by l-α-(2-aminoethoxyvinyl)-glycine, α-aminoisobutyric acid or 2,3,5-triiodobenzoic acid applied to the silks before pollination. Thus ethylene synthesis and polar auxin transport in the silks pollinated with fresh pollen were necessary to evoke accelerated differentiation of the egg cells in maize ovules. Differences in pistil responses found between true- and mock-pollination suggest that signalling pathways are at least partially different for the reception of pollen grains and sand crystals on maize stigma. (Received February 25, 2004; Accepted July 16, 2004) Introduction Various effects of pollination on flower development and senescence are known in angiosperms and such post-pollination symptoms as petal wilting, pigmentation changes, ovary and ovule development or style and stamen abscission are mediated by ethylene (Stead 1992, O’Neill 1997, Van Doorn 1997, Van Doorn and Stead 1997). Pollination promotes ethylene biosynthesis in flower organs and accumulation of mRNAs coding for ethylene biosynthetic enzymes (Nichols et al. 1983, Hoekstra and Weges 1986, Pech et al. 1987, O’Neill et al. 1993, Larsen et al. 1995, Clark et al. 1997, Jones and Woodson 1997, Woltering et al. 1997, Bui and O’Neill 1998, Jones and Woodson 1999, Ketsa and Rugkong 2000, Llop-Tous et al. 2000). Ethylene, ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC) and auxin have been proposed to operate as the pollination signals (for review see O’Neill 1997). Indeed, auxin induces the synthesis of ACC synthase (Yang and Hoffman 1984) and some auxin-inducible ACC synthase genes have been found (Yoon et al. 1997, Bui and O’Neill 1998). It appears that auxin and ethylene act in concert during post-pollination events. Interaction of these hormones is also known in the regulation of vegetative growth (Sitbon and Perrot-Rechenmann 1997, Smalle and Van der Straeten 1997, Stepanova and Ecker 2000). Differentiation of embryo sacs within developing ovules in response to pollination has been reported for Phalaenopsis orchid (Zhang and O’Neill 1993). Pollination can also act on specific cells within the embryo sac. In maize, maturation of the egg cells was accelerated during the progamic phase, i.e. after pollination and before actual fertilization (Mól et al. 2000). Other examples of the post-pollination responses of embryo sac cells are known for synergids which, in some species, may degenerate at various steps of the progamic phase (reviewed by Russell 1992). Those studies, nevertheless, were concerned only with the cytological aspect of the response, and no signalling mechanisms were investigated in relation to the cellular events, except for the work on Phalaenopsis. Moreover, investigations of ethylene and other factors affecting flower morphology and senescence have been focused on ornamental plants, mainly carnation and orchids, because of their importance for the flower trade. In cereal inflorescences, ethylene is possibly also involved in early responses to pollination; however, this has not yet been shown. It is only known that late post-pollination processes, such as ear senescence and grain maturation in wheat (Beltrano et al. 1994) or the programmed cell death in maize endosperm (Young et al. 1997), are promoted by ethylene. In the grasses, post-pollination events progress rapidly in comparison with those of other angiosperms. The sequence of events in maize is known from several papers (Heslop-Harrison et al. 1985, Mól et al. 1994, Mól et al. 2000, Mól et al. 2004). In brief, pollen grain hydration occurs 3–5 min after pollen arrival to silk trichomes, and the pollen tube tip appears 10 min later on. Pollen hydration and germination are accompanied by changes of the electric potential in the silk and ovary, and by calcium changes in the stigma trichomes. Numerous pollen tubes enter the stigma about 20 min after pollination, and they reach the transmitting tissue of the silk axis within the first hour following pollination. The pollen tubes grow towards the ovary. A few of them pass the silk base about 6 h after pollination, and turgor loss appears there at the formation of the silk abscission zone. Intra-ovarian growth of the pollen tubes takes some hours until one pollen tube enters the embryo sac 12 h after pollination. The shift in frequency of the egg cell developmental stages starts 8 h after pollination, and 4 h later, at the end of progamic phase, only about 35% of egg cells remain immature in a maize ear. In the research presented here we check whether ethylene and auxin participate in the post-pollination signalling which leads, in maize, to accelerated maturation of the egg cells (see Mól et al. 2000). Ethylene biosynthesis in the ears, as well as the content of ACC and indole-3-acetic acid (IAA) in the silks and ovaries, was investigated. We successfully used ACC and IAA to stimulate egg maturation in non-pollinated ears, and inhibitors of ethylene synthesis or auxin transport to block the egg cell response to pollination. Our results indicate that both hormones are necessary for accelerated differentiation of the egg cells in the pistils pollinated with fresh pollen grains. Moreover, we used mock-pollination with quartz sand to test the hypothesis that mechanical stimulation of the stigma might also play a role in the response elicitation after pollination. Results Silks and ovaries contribute to ethylene synthesis in pollinated maize ears Gas chromatography was used to determine ethylene production by maize female inflorescences in planta after pollination with fresh pollen or mock-pollination with quartz sand. When fresh pollen grains were applied to the receptive silks extending out of the husks, ethylene emission from the ears increased within 1 h after pollination (Fig. 1). Then up to 12 h, the ethylene production in the pollinated ears was high and slightly variable (15–22 ng ml–1 h–1). Quartz sand used to substitute for pollen also caused a rise in ethylene synthesized by the ears showing the first peak 90 min after treatment and the second increase at 7–9 h (Fig. 1). The effect of quartz sand indicates that a mechanical stimulus applied to the maize stigma and comparable to that of pollen was sufficient to induce ethylene synthesis in the ear, although at a slightly lower level compared with the effect of pollen. To define the participation of particular pistil parts in the overall ethylene production, the content of ACC in maize silks and ovaries was determined after pollination or sand application. As early as 20 min after pollination, the ACC level increased in both the silks and the ovaries by about 10 ng (g DW)–1. However, ACC was also found in pollen grains [up to 61 ng (g DW)–1]. In the silks, ACC content remained high [45 ng (g DW)–1] for 4 h, and decreased afterwards (Fig. 2A). In the ovaries, it reached the second maximum at 8 h (Fig. 2B). At 12 h after pollination, i.e. at the end of progamic phase in maize (Mól et al. 1994), ovary ACC reached its lowest level. Different profiles of ACC content were observed in the silks and ovaries after quartz sand application to the silks (Fig. 2A, B). Until 4 h after sand application, the ACC level in the silks was lower than after pollination, and only 8 h after the treatment it increased by 10 ng (g DW)–1. In the ovaries, an ACC rise by 15 ng (g DW)–1 was detected at 4 h after treatment, and another peak appeared 8 h later. These results indicate that both parts of maize pistil, the silk and the ovary, were involved in ethylene biosynthesis after true- or mock-pollination, but with slightly differing kinetics. The initial rise of ethylene synthesis was rapid in the pollinated ears, and coincided with an immediate ACC augmentation. The auxin level rises very high in the silks after pollination During the progamic phase, the IAA content in the pollinated silks increased markedly (Fig. 2C). The first maximum occurred as early as 20 min after pollination when approximately 0.5 µg of IAA was found per g DW of the pollinated silks compared with 6 ng (g DW)–1 in the silks before pollination. The second peak in the silks was even higher, and after 8 h, was twice as high as the first maximum. Maize pollen contained 0.66–2.36 µg (g DW)–1 IAA, and if silk weight and pollen amount are considered (Table 1), the partial contribution of pollen-borne auxin to the initial increase in IAA content in pollinated silks seems possible even if some auxin might degrade in germinating pollen grains as discussed later on. The auxin content increased also in the ovaries of pollinated ears but only to levels about 10 times lower than in the silks: 57±15 ng (g DW)–1 at 20 min and 123±25 ng (g DW)–1 at the end of the progamic phase 12 h after pollination (Fig. 2D). In the experiment with quartz sand applied to the silks instead of pollen, IAA augmentation was lower and slower than in the pollinated ears, both in the silks and the ovaries (Fig. 2C, D). The rapid changes of auxin in pistil tissues were thus specific to pollination. Shortly after pollination, the amount of IAA increased not only in the silks but also in the ovaries. Translocation of 14C-labelled auxin in maize pistils was investigated to check whether IAA transfer from the pollination site might contribute to the rise of auxin content in the ovaries. The distal fragments of the silks, which extended out of the husks, were treated with [14C]IAA, and the radioactivity was subsequently measured in the proximal parts of the silks and in the ovaries (Table 2). Less than 1% and <5% of the radioactivity in tissue extracts was recovered from the proximal silk fragments and the ovaries, respectively. The fact that radioactivity levels detected in the ovaries were rather stable throughout the experiment and higher than in the silks, indicates that pistil tissues enable relatively rapid transfer of some auxin from the stigma. Ethylene precursor or auxin applied to the silks stimulates egg cell maturation in the ovules Our previous report (Mól et al. 2000) showed that maize egg cell differentiation progressed more quickly in a pollinated ear than without pollination. The frequencies of egg developmental stages (A–C; Fig. 3) changed before sperm delivery to the embryo sac, and significantly fewer young stage-A cells appeared 12 h after pollination. Here we found that substituting ACC or IAA for pollination still caused shifts in the frequencies of various egg classes in individual ears at 12 h after treatment (Fig. 4). After hormonal treatments, as well as after pollination, the small and non-vacuolated egg cells (stage A) were at much lower frequencies than larger and vacuolated ones (stages B + C). In non-treated or H2O-treated ears, the frequencies of stage-A egg cells were 58% and 48%, respectively. In pollinated ears, such immature egg cells were found only in 31% of the ovules. The effects of 1 µM ACC and 0.6 µM IAA were not significant when compared with the control treatment with water. At higher concentrations however, stage-A egg cells appeared in 22–29% of the ovules after application of ACC and in 14–15% after application of IAA. The results of significance tests for differences in frequencies of all egg cell stages are given in Table 3. Differences for stage-B cells were usually non-significant, as this stage is an intermediate step in egg cell maturation (Mól et al. 2000). Both ACC (at all concentrations) and IAA (at 6 and 60 µM) increased the percentage of highly vacuolated mature egg cells (stage C). Overall, supplying non-pollinated receptive silks of Zea mays with micromolar solutions of auxin or ACC stimulated maturation of the egg cells to a greater extent than did pollination with fresh pollen. Inhibitors of ethylene synthesis and auxin transport block the egg response to ear pollination In another series of experiments, various inhibitors of hormone synthesis, transport or action were applied to the silks before pollination and frequencies of the egg cell stages were estimated from microscopic observations (Fig. 5, Table 3). Silk pre-treatment with the inhibitors did not markedly affect pollen grain germination or pollen tube growth (data not shown). Blocking of ethylene biosynthesis by stigma treatment with l-α-(2-aminoethoxyvinyl)-glycine (AVG) or α-aminoisobutyric acid (AIB) abolished the effect of pollination, and 12 h after pollination stage-A egg cells were present at high frequencies. Also when 2,3,5-triiodobenzoic acid (TIBA), an auxin transport inhibitor, was applied to the silks, pollination did not accelerate egg cell maturation. After all those treatments, the egg cell frequencies remained basically the same as in non-treated controls (stage A 54–59%, stage B 20–24%, stage C 21–22%). On the other hand, silver thiosulfate (STS), a recognized inhibitor of ethylene action, was not able to cancel the effect of pollination. STS treatment of the silks before pollination gave a similar response to that after the control treatment with water. Moreover, the egg cell response was not affected by TIBA or AIB used prior to sand application to the silks (Fig. 6, Table 3). This confirms that hormonal mechanisms of maize ear response to the fresh pollen and quartz sand are different. IAA polar transport and ACC oxidation in the silks were necessary to mediate signals during genuine pollination but not after mock-pollination with sand. Discussion The acceleration of egg cell maturation in a pollinated maize ear was demonstrated in our former paper (Mól et al. 2000). The reproductive strategy in maize, a wind-pollinated plant, seems to involve triggering of egg cell maturation by pollen arrival at the stigmatic surfaces. A very efficient signalling system must thus act between the maize stigmas (silks) and ovaries where the egg cells wait for sperm delivery within the embryo sacs. Factors involved in this signalling in maize pistils have not been identified. However, a burst of ethylene production in pollinated flowers is well documented in several species (see Introduction). In carnation and tobacco, the first peak of ethylene provides the information about a pollination event, and two subsequent peaks are responsible for corolla wilting or recognition of compatible pollen tubes (Jones and Woodson 1997, De Martinis et al. 2002). We observed that ethylene synthesis was rapidly stimulated in the pollinated ears of Z. mays, and the levels of ethylene precursor (ACC) and auxin (IAA) increased both in the silks and ovaries 20 min after pollination. These early changes are possibly related to the information about pollen arrival. Late increases in ethylene evolution and in auxin content found 4–9 h after pollen or sand application correspond to the formation of an abscission zone at the silk base and further silk wilting (Heslop-Harrison et al. 1985, and our unpublished data). We compared the effects of the fresh maize pollen and quartz sand on ethylene emission from maize ears as well as on the ACC level in the silks and ovaries. Ethylene production started to increase immediately after pollen or sand placement on the silks. The first peak was recorded 1 h after pollination, whereas after sand application, the first peak appeared 30 min later and was lower than after pollination. The level of ethylene precursor increased within 20 min after pollination both in the silks and ovaries. Most probably, the fast ACC rise in the ovaries was not the result of ACC translocation because the pollination site was about 10 cm apart from the ovaries, and ACC was shown to be largely immobile in flower tissues (Woltering et al. 1995, Woltering et al. 1997). When quartz sand was used for mock-pollination, ACC content increased in the ovaries and in the silks with a delay of 4 h and 8 h, respectively. After using sand, the IAA rise was also slower, as mentioned earlier. Moreover, TIBA or AIB treatments of the silks did not hamper egg cell maturation when quartz sand was applied but the same inhibitors were effective after using fresh maize pollen. At present, the mechanisms underlying the differences between true- and mock-pollination are unknown. The clue might lie in IAA and ACC delivered with pollen grains to the silks, as these compounds could be directly involved in signalling at the pollination site. The silks were not supplied with auxin and ethylene precursor when sand was used instead of pollen. Nevertheless, our sand treatment stimulated egg maturation (see Table 3), although the frequency of immature egg cells (stage A) was significantly higher after pollination with sand than with pollen (Mól et al. 2000). Thus, the pure mechanical action of sand application almost substituted for the effect of pollen. Sand and pollen caused similar, but not identical, responses in egg cell populations as well as different post-pollination profiles of auxin and ACC in pistil tissues. Could two independent mechanisms act after true or mock pollination, and one of them, evoked by viable maize pollen, required auxin and ACC on the stigma? We recently found another difference between maize stigma pollinated with sand or pollen: trichomes receiving pollen grains responded with an intracellular Ca2+ increase, and such a reaction of the trichomes did not occur after mock-pollination with sand (Mól et al. 2004). All these facts suggest that signalling pathways are not the same for reception of pollen grains or sand crystals on the maize stigma. Application of ACC solution instead of pollen to the silks stimulated egg cell maturation (see Fig. 4) and ethylene synthesis (data not shown). We found a low ACC level in maize pollen. If not immediately converted to ethylene, it could contribute to ACC found in the silks just after pollination. However, many reports imply that pollination-induced ethylene is derived from endogenous synthesis of ACC rather than from exogenous pollen-borne ACC (reviewed by O’Neill 1997). Pollen grains of Petunia contained a very high concentration of ACC but it was not the precursor and trigger of the early ethylene production on the stigma (Hoekstra and Weges 1986). Also in petunias engineered for low ACC content in the pollen grains, the early ethylene peak in pollinated flowers was similar to the effect of wild-type pollen (Lei et al. 1996). The same effect was observed after cross-pollination of petunia flowers with pollen grains of Erythrina orientalis and Cosmos bipinnatus exhibiting low ACC levels (Hoekstra and Weges 1986). Evidence that ethylene rather than ACC stimulated the egg response in maize appears from the effects of AVG or AIB applied to the silks before pollination. Inhibition of ACC synthesis or ACC oxidation both abolished egg cell maturation in pollinated ears. Our results suggest that a direct action of ethylene at the pollination site was not a prerequisite for further egg cell response. STS, the inhibitor of ethylene action, did not block rapid egg maturation when applied to the apical parts of the silks before pollination. Auxin-induced ACC synthase genes are known from seedlings and flowers (Yoon et al. 1997, Bui and O’Neill 1998) and auxin was postulated to act as another pollination signal (for review see O’Neill 1997). We found that IAA content increased rapidly in the silks and ovaries of maize after pollination with fresh pollen grains but not after quartz sand application. At 20 min after pollination, the auxin level was about 80 times higher in pollinated silks than in unpollinated controls. To explain the initial rise of IAA in the silks it is necessary to consider the quantity of IAA delivered with the pollen (calculated on the basis of weight; see Table 1). Pollen might have provided as much as 33–73% of the auxin found in the pollinated silks. However, the question of how much of pollen IAA can be taken up by maize stigma needs further investigation. When the first auxin rise appeared in the silks, the pollen grains already germinated, and short pollen tubes penetrated into the stigma trichomes. Liu and Lee (1995) determined that in maize pollen grains germinating in vitro, the IAA level declines by 64% due to its degradation by oxidases. If a similar situation appears in vivo, there would still be some pollen-borne auxin left to play a role on the stigma. Blockage of the egg cell response by an inhibitor of polar auxin transport (TIBA) applied to the silks before pollen indicates that auxin is involved in the post-pollination signalling. In our experiment on auxin translocation in maize pistil, only about 4.3% of radiolabelled IAA was transported from stigma to the ovaries. If pollen delivered to the silks was the only source of auxin for the ovaries, one could expect approximately 20 ng IAA per g DW of ovary, i.e. 4.3% of the estimated IAA content in a pollen dose (see Table 1) calculated per g DW of ovary (675±167 mg, n = 8). Such a putative amount of auxin transferred from the pollination site is about 36% of the IAA content found in the ovaries 20 min after pollination (see Fig. 2D). Thus, auxin liberation from conjugates and/or its translocation from other tissues possibly occurred. De novo synthesis of IAA was unlikely in a narrow time window. The rate of polar auxin transport (about 1 cm h–1; Taiz and Zeiger 1998) is too slow for a long-distance effect and the early rise of IAA in maize ovaries. Translocation via phloem had to occur where the transport velocities range from 30 to 150 cm h–1 (Taiz and Zeiger 1998). Phloem is present along the transmitting tissue in maize silks and ends in the ovary wall (Heslop-Harrison et al. 1985, and our unpublished observations). We postulate that some pollen-borne IAA was moved to phloem elements in the silks by a polar transport mechanism, and then a long-distance transport enabled its distribution throughout the pistil. It is conceivable that in new locations, auxin-stimulated ethylene synthesis and various other cytological events were promoted by these hormones, e.g. abscission zone formation in the silk or egg cell maturation in the embryo sac. Taken together, our results demonstrate that the response of maize egg cells to pollination is governed by partially different mechanisms when pollen or sand are applied to the silks. Only the application of fresh pollen grains caused rapid production of ethylene and immediate augmentation of auxin levels in both the silks and ovaries. Whereas after using sand, the first peak of ethylene emission by the ear and the IAA rise in the ovaries were lower and delayed in comparison with true pollination. Obviously, there were no pollen tubes in maize silks after application of sand instead of pollen, and Weterings et al. (2002) found recently that ACC-synthase transcript levels are modulated by the number of pollen tubes growing in tobacco styles. Despite differences in hormonal response, pollen or sand both evoked accelerated maturation of maize egg cells (Mól et al. 2000). We can assume that some yet unidentified factors contributed to the reception of a mechanical signal after the arrival of pollen grains or sand crystals. In another paper (Mól et al. 2004), we show that electric signals propagate through the silk to nucellus after the reception of pollen grains or quartz sand on maize stigma. Thus, ion fluxes could be the basic events in a pollinated stigma. An increase in ethylene synthesis in the inflorescence and hormonal changes in the ovaries are possibly the secondary responses to those primary events. Differential expression of ethylene biosynthetic genes was investigated in flowers of several species (Clark et al. 1997, Ten Have and Woltering 1997, Bui and O’Neill 1998, Jones and Woodson 1999, Llop-Tous et al. 2000). Such molecular studies in Z. mays will be feasible after identification of adequate maize genes. Here we have shown for this species that ethylene metabolism is affected by pollination, and that auxin together with ethylene are involved in stigma-to-ovary signalling. However, the precise sequence of hormonal events in pistil tissues as well as the evidence that auxin and/or ethylene act directly on the cells in maize silk and embryo sac need further investigation. Future cyto-physiological and molecular studies should also help to discriminate between primary and secondary pollination signals. Materials and Methods Plant material and treatments Maize (Z. mays L., line A 188) plants were grown in a growth chamber as described earlier (Mól et al. 2000). Ears were covered with paper bags before silk emergence and used for experiments when the external silk length was 11–13 cm (optimum receptivity). Portions (0.5 ml) of fresh maize pollen (line A188) or quartz sand (Fontainebleau 150–200 µm; Prolabo, Nogent sur Marne, France) were applied to the silks for true- or mock-pollination, respectively. Subsequently, ethylene production, ACC and IAA content or the frequencies of three egg developmental stages were determined (see below) after pollination, sand application and in non-pollinated ears. In a series of experiments, silks extending out of the husks were treated prior to pollination with 0.2 mM solutions of AVG (inhibitor of ACC-synthase) or AIB (ACC analogue blocking ACC oxidase), 0.4 mM STS (inhibitor of ethylene action), 12 µM TIBA (inhibitor of polar auxin transport) or ultrapure water (control). Silks were immersed for 20 min in the solutions contained in 50-ml Falcon tubes attached to the ears with paper tape and then gently shaken to remove liquid. Then entire ears were covered again (during STS treatment, aluminium foil was used instead of paper bags to avoid tissue blackening) and left to dry for 1 h. After this time, hand pollination was applied and 12 h later the silks and ovules were fixed as described in the next paragraph. Similar treatments with TIBA and AIB were also performed before mock-pollination with sand. In another series of experiments, solutions of ACC (1, 10 and 100 µM), IAA (0.6, 6 and 60 µM) or water (control) were applied to the silks for 20 min. The silks were treated as in the experiments with inhibitors but were not pollinated, and 12 h after treatment, ovule-halves were fixed for egg cell counting (see next paragraph). If no pollen, sand or solutions were applied to the silks, they were touched with hands in the same manner as during pollination. Chemicals were purchased from Sigma-Aldrich; STS was prepared directly before use from AgNO3 and Na2O3S2 stock solutions. All experiments were repeated twice. Cytological preparations To prepare material for tissue clearing, ovule-halves were dissected from the ears as described elsewhere (Mól et al. 2000) and fixed for 24 h in FAA (formalin 4% : acetic acid : ethanol 70% = 5 : 5 : 90 by vol.). Nucellar slices containing embryo sacs were cleared in methyl salicylate according to Young et al. (1979). The egg cell stages were determined by differential interference contrast (DIC) microscopy (Biolar PI; PZO, Warsaw, Poland) in 4,173 embryo sacs (for details see Table 3). To check for pollen tube growth after treatments with hormone inhibitors, the silks were fixed in FAA for 24 h, then rinsed in 70% ethanol and water, macerated in 10 M NaOH for 10 min and rinsed twice in water. Pollen tubes were stained with aniline blue for 15 min and observed in glycerol under a fluorescence microscope (Optiphot 2; Nikon, Tokyo, Japan). Agfa Pan APX 25 (Agfa-Gevaert AG, Leverkusen, Germany) was used for microphotography. The two-tailed t-test for the difference between two proportions (a module of Statistica 6.0 software, StatSoft Inc., Tulsa, OK, U.S.A.) was applied to determine significance levels for the differences found between egg cell frequencies after various treatments. Ethylene measurement The rate of ethylene production by maize ears in planta was determined after true- or mock-pollination. Fresh pollen or quartz sand was applied to the silks at optimum receptivity (external length of about 12 cm). In unpollinated controls, the silks were only touched in the same way as during hand pollination. The paper bags protecting young silks against accidental and precocious pollination were then replaced by PVC bags closed around the ears. During the subsequent 12 h period, 1-ml gas samples were collected with a gas syringe from the air inside the bag at 30-min intervals. Ethylene concentration was determined from the withdrawn samples injected into a gas chromatograph (Hewlett Packard 5890 Series II; Hewlett-Packard, Waldbronn, Germany), which was equipped with an alumina capillary column and a flame ionization detector. The experiment was repeated twice. ACC determination The pollinated or sand-treated ears were harvested directly after treatment and then after 4, 8 and 12 h. Their husks were removed, the silks were detached from ovaries and the ovaries were cut off from the ear rachis. As for microscopic observations, the ovaries from both ear tip (underdeveloped ones) and ear base (overgrown ones) were omitted. The samples of pistil tissues were quickly weighed, frozen in liquid nitrogen and stored at –70°C for freeze-drying (ChristAlpha I-5; Fisher, Illkirch, France). These manipulations took about 20 min, thus the real timing for ACC determination was 20, 260, 500 and 740 min after treatment. The freeze-dried samples were stored in liquid N2 until extracted and analysed. Tissue samples were homogenized in 80% ethanol [2 ml (g FW)–1] and centrifuged (5,000×g, 15 min). Supernatants were kept on ice and the extraction was repeated. Combined supernatants of both extractions were evaporated to water phase at 50°C and afterwards the residues were dissolved in 3 ml H2O. Samples were spun at 28,500×g for 5 min and the ACC content was measured in the aqueous phase of tissue extracts according to Lizada and Yang (1979). Briefly, each supernatant was held in a tight vial, and 4 µM ACC (internal standard), 1 µM HgCl2, and a mixture of 5% NaOCl and saturated NaOH (2 : 1, v/v) were injected into the vial through its cap. The mixture was vortexed and incubated for 3 min before a 1.5-ml gas sample was taken from each vial for gas chromatography (Hewlett-Packard 5890 Series II). The released ethylene was measured three times for each sample and the conversion factor determined by comparison with the production of ethylene from the internal standard. All steps were performed at 4°C and experiments were conducted on two series of female inflorescences. The results were expressed on dry weight (DW) basis. IAA determination Tissue samples were collected and stored as described above. Tissue was homogenized in liquid N2 and extracted twice with 80% methanol containing 100 mg litre–1 butylated hydroxytoluene. Tritiated IAA (1.06 TBq mmol–1; Amersham) was added as an internal standard. After centrifugation (4°C, 15 min, 10,000×g) the supernatant was evaporated to the water phase, the same volume of 0.5 M K2HPO4 was added and the extract was partitioned against diethylether. Ether was discarded and the extract was transferred to a Polyclar AT column and washed with 10 ml 0.1 M K2HPO4. The eluate was acidified with 0.3 M H3PO4 to pH 2.7 and partitioned against ether. The ether phase was evaporated, dissolved in 300 µl of mobile HPLC phase and IAA content (per g of DW) was determined using HPLC with fluorimetric detector as described in Eder et al. (1988). Three assays were performed for each time point. Translocation of radiolabelled auxin Silks extending out of the husks (12 cm external length) were dipped in [14C]IAA solution (23×104 dpm) for 20 min as described above for hormonal treatments. Then the ears were harvested, silk parts exposed to [14C]IAA were cut off and the husks were removed. Internal parts of the silks which had no contact with radiolabelled auxin and the ovaries were collected, weighed and frozen in liquid N2. Tissue samples were prepared at 20, 260, 500 and 740 min after [14C]IAA application and for control non-treated ears. For determination of label movement, silk and ovary tissues (triplicate samples of 0.5 g) were homogenized in Bray’s scintillation solution (Akwascynt; BioCare, Warsaw, Poland). Homogenates were supplied with 20 µl H2O2 and left for 24 h for decoloration, and chemiluminescence quenching was achieved 24 h after adding 20 µl of acetic acid to each sample. Radioactivity of the samples was measured by liquid scintillation counter (Beckham LS 5801; Beckham Instruments Inc., Irvine, CA), and the results were recalculated per g of fresh weight (FW) after subtraction of background noise. Acknowledgments We thank Dr. T. Wyka for his comments on the manuscript and for correction of the English. This work was supported by the French Ministry of Foreign Affairs and the KBN Polish State Committee for Scientific Research (projects 98084, 01847UJ, KBN grant 6P04C 09212). 5 Corresponding author: E-mail, [email protected]; Fax, +48-61-829-56-11. View largeDownload slide Fig. 1 Effects of pollination or sand application on ethylene production of maize ears during the progamic phase. The silks at optimum receptivity were pollinated with fresh pollen grains or treated with quartz sand instead of maize pollen. Measurements were conducted during 12 h following manipulations and in control plants. Mean ± SD represent data (n = 4) from two independent experiments. View largeDownload slide Fig. 1 Effects of pollination or sand application on ethylene production of maize ears during the progamic phase. The silks at optimum receptivity were pollinated with fresh pollen grains or treated with quartz sand instead of maize pollen. Measurements were conducted during 12 h following manipulations and in control plants. Mean ± SD represent data (n = 4) from two independent experiments. View largeDownload slide Fig. 2 Levels (mean ± SD) of ethylene precursor (A, B) and auxin (C, D) in the silks (A, C) and ovaries (B, D) of maize ears determined at 4-h intervals after application of fresh pollen grains or quartz sand. At each time point, silk and ovary samples were collected from the same ears for measurements of ACC (n = 6) and IAA (n = 3) in two independent ear sets. The time necessary for sample preparation after treatment was 20 min, and this was the earliest time point compared with non-treated controls. View largeDownload slide Fig. 2 Levels (mean ± SD) of ethylene precursor (A, B) and auxin (C, D) in the silks (A, C) and ovaries (B, D) of maize ears determined at 4-h intervals after application of fresh pollen grains or quartz sand. At each time point, silk and ovary samples were collected from the same ears for measurements of ACC (n = 6) and IAA (n = 3) in two independent ear sets. The time necessary for sample preparation after treatment was 20 min, and this was the earliest time point compared with non-treated controls. View largeDownload slide Fig. 3 Three developmental stages of maize egg cells in the micropylar regions of the embryo sacs. (A) small non-vacuolated egg cell; (B) larger cell with a higher degree of vacuolation; (C) mature egg cell with a large apical vacuole. Synergids (s) and polar or secondary nuclei (asterisks) of the central cell (cc) are visible next to the egg cells (ec); n, nucellus. DIC images from cleared tissues. Bar = 30 µm View largeDownload slide Fig. 3 Three developmental stages of maize egg cells in the micropylar regions of the embryo sacs. (A) small non-vacuolated egg cell; (B) larger cell with a higher degree of vacuolation; (C) mature egg cell with a large apical vacuole. Synergids (s) and polar or secondary nuclei (asterisks) of the central cell (cc) are visible next to the egg cells (ec); n, nucellus. DIC images from cleared tissues. Bar = 30 µm View largeDownload slide Fig. 4 Effects of auxin and ethylene precursor on mean (± SD) frequencies of three egg cell stages (A, B, C) in the ears of maize. Solutions of IAA and ACC at given concentrations were applied to non-pollinated silks at their optimum receptivity, and the egg cell frequencies were determined 12 h after treatment. Pollinated ears and non-treated or water-treated ears were positive and negative controls, respectively. View largeDownload slide Fig. 4 Effects of auxin and ethylene precursor on mean (± SD) frequencies of three egg cell stages (A, B, C) in the ears of maize. Solutions of IAA and ACC at given concentrations were applied to non-pollinated silks at their optimum receptivity, and the egg cell frequencies were determined 12 h after treatment. Pollinated ears and non-treated or water-treated ears were positive and negative controls, respectively. View largeDownload slide Fig. 5 Effects of ethylene and auxin inhibitors on mean (± SD) frequencies of three egg cell stages (A, B, C) in the pollinated ears of maize. Inhibitors of ethylene synthesis (AVG 0.2 mM, AIB 0.2 mM) or an inhibitor of auxin polar transport (TIBA 12 µM) were applied to the receptive silks 1 h before pollination with fresh maize pollen, and the egg cell frequencies were determined 12 h after pollination. Pollinated ears (without or with water pre-treatment) and non-treated ears were positive and negative controls, respectively. View largeDownload slide Fig. 5 Effects of ethylene and auxin inhibitors on mean (± SD) frequencies of three egg cell stages (A, B, C) in the pollinated ears of maize. Inhibitors of ethylene synthesis (AVG 0.2 mM, AIB 0.2 mM) or an inhibitor of auxin polar transport (TIBA 12 µM) were applied to the receptive silks 1 h before pollination with fresh maize pollen, and the egg cell frequencies were determined 12 h after pollination. Pollinated ears (without or with water pre-treatment) and non-treated ears were positive and negative controls, respectively. View largeDownload slide Fig. 6 Mean (± SD) frequencies of three egg cell stages (A, B, C) in maize ears after treatment with ethylene or auxin inhibitors and mock-pollination with quartz sand. An analogue of ACC (AIB 0.2 mM) or an inhibitor of auxin polar transport (TIBA 12 µM) were applied 1 h before placing sand on the receptive silks, and the egg cell frequencies were determined 12 h after mock-pollination. Ears treated with sand (without or with water pre-treatment) and non-treated ears were positive and negative controls, respectively. View largeDownload slide Fig. 6 Mean (± SD) frequencies of three egg cell stages (A, B, C) in maize ears after treatment with ethylene or auxin inhibitors and mock-pollination with quartz sand. An analogue of ACC (AIB 0.2 mM) or an inhibitor of auxin polar transport (TIBA 12 µM) were applied 1 h before placing sand on the receptive silks, and the egg cell frequencies were determined 12 h after mock-pollination. Ears treated with sand (without or with water pre-treatment) and non-treated ears were positive and negative controls, respectively. Table 1 Estimated contribution of pollen-borne IAA to the initial rise in auxin content in the maize silks 20 min after pollination with fresh pollen grains Values  Mean  Minimum  Maximum  IAA in unpollinated silks [ng (g DW)–1]  6.1 ± 1.7 (n = 3)  4.4  7.8  IAA in pollinated silks 20 min after pollen reception (ng g–1 DW)  506 ± 23 (n = 3)  489  522  IAA in fresh maize pollen [ng (g DW)–1]  1,511 ± 774 (n = 4)  665  2,361  Pollen dose applied to silks (mg DW)  203 ± 10 (n = 8)  184  216  Estimated IAA in pollen dose (ng)  307  122  510  Dry weight of pollinated silks (g DW)  1.05 ± 0.24 (n = 8)  0.76  1.34  Estimated pollen-borne IAA in pollinated silks [ng (g DW)–1]  292  161  381  Estimated contribution of pollen-borne IAA to auxin content in pollinated silks (%)  58  33  73  Values  Mean  Minimum  Maximum  IAA in unpollinated silks [ng (g DW)–1]  6.1 ± 1.7 (n = 3)  4.4  7.8  IAA in pollinated silks 20 min after pollen reception (ng g–1 DW)  506 ± 23 (n = 3)  489  522  IAA in fresh maize pollen [ng (g DW)–1]  1,511 ± 774 (n = 4)  665  2,361  Pollen dose applied to silks (mg DW)  203 ± 10 (n = 8)  184  216  Estimated IAA in pollen dose (ng)  307  122  510  Dry weight of pollinated silks (g DW)  1.05 ± 0.24 (n = 8)  0.76  1.34  Estimated pollen-borne IAA in pollinated silks [ng (g DW)–1]  292  161  381  Estimated contribution of pollen-borne IAA to auxin content in pollinated silks (%)  58  33  73  For calculations, experimental data on IAA content and dry weight of the silks and pollen grains were taken, and no rapid degradation of auxin in germinating pollen was assumed. View Large Table 2 Distribution of radiolabelled auxin in the pistils of maize ear Time after [14C]IAA application  Radioactivity (% of applied)     Silks  Ovaries  20 min  0.48 ± 0.06  4.35 ± 0.07  260 min  0.53 ± 0.02  4.73 ± 0.11  500 min  0.84 ± 0.08  3.71 ± 0.07  740 min  0.79 ± 0.05  4.56 ± 0.10  Time after [14C]IAA application  Radioactivity (% of applied)     Silks  Ovaries  20 min  0.48 ± 0.06  4.35 ± 0.07  260 min  0.53 ± 0.02  4.73 ± 0.11  500 min  0.84 ± 0.08  3.71 ± 0.07  740 min  0.79 ± 0.05  4.56 ± 0.10  [14C]IAA was applied to apical fragments of the silks, and radioactivity was measured in the proximal silk fragments and ovaries, which were covered by husks and unexposed to the labelled compound. Mean ± SD (n = 3) is given for silks and ovaries. View Large Table 3 Effects of various treatments on egg cell (EC) frequency (%) in maize ears at the end of progamic phase Treatment  Total EC number  EC frequency (%)      Stage A  Stage B  Stage C  No treatment  306  57.5  21.6  20.9  Pollen a  198  31.3**  27.3 ns  34.3**  Water b  198  48.0*/**  23.2 ns/ns  28.8*/ns  IAA 0.6 µM c  151  43.0**/*/ns  30.5 ns/ns/ns  26.5 ns/ns/ns  IAA 6 µM c  167  15.0**/**/**  25.1 ns/ns/ns  59.9**/**/**  IAA 60 µM c  183  13.7**/**/**  20.2 ns/ns/ns  66.1**/**/**  ACC 1 µM c  266  41.7**/*/ns  18.0 ns/*/ns  40.2**/ns/*  ACC 10 µM c  220  28.6**/ns/**  24.1 ns/ns/ns  47.3**/**/**  ACC 100 µM c  193  21.8**/*/**  24.9 ns/ns/ns  53.4**/**/**  Water + pollen b  310  35.8**/ns  27.4 ns/ns  36.8**/ns  TIBA + pollen d  253  53.8 ns/**/**  23.7 ns/ns/ns  22.5 ns/**/**  AVG + pollen d  282  58.9 ns/**/**  19.8 ns/*/*  21.3 ns/**/**  AIB + pollen d  258  55.0 ns/**/**  23.2 ns/ns/ns  21.7 ns/**/**  STS + pollen d  301  36.5**/ns/ns  20.6 ns/ns/ns  42.8**/ns/ns  Sand a  204  42.2**  27.9 ns  29.9*  Water + sand e  248  39.1**/ns  23.8 ns/ns  37.1**/ns  TIBA + sand f  225  33.8**/ns/ns  23.6 ns/ns/ns  42.7**/**/ns  AIB + sand f  210  30.5**/*/ns  31.9*/ns/ns  37.6**/ns/ns  Treatment  Total EC number  EC frequency (%)      Stage A  Stage B  Stage C  No treatment  306  57.5  21.6  20.9  Pollen a  198  31.3**  27.3 ns  34.3**  Water b  198  48.0*/**  23.2 ns/ns  28.8*/ns  IAA 0.6 µM c  151  43.0**/*/ns  30.5 ns/ns/ns  26.5 ns/ns/ns  IAA 6 µM c  167  15.0**/**/**  25.1 ns/ns/ns  59.9**/**/**  IAA 60 µM c  183  13.7**/**/**  20.2 ns/ns/ns  66.1**/**/**  ACC 1 µM c  266  41.7**/*/ns  18.0 ns/*/ns  40.2**/ns/*  ACC 10 µM c  220  28.6**/ns/**  24.1 ns/ns/ns  47.3**/**/**  ACC 100 µM c  193  21.8**/*/**  24.9 ns/ns/ns  53.4**/**/**  Water + pollen b  310  35.8**/ns  27.4 ns/ns  36.8**/ns  TIBA + pollen d  253  53.8 ns/**/**  23.7 ns/ns/ns  22.5 ns/**/**  AVG + pollen d  282  58.9 ns/**/**  19.8 ns/*/*  21.3 ns/**/**  AIB + pollen d  258  55.0 ns/**/**  23.2 ns/ns/ns  21.7 ns/**/**  STS + pollen d  301  36.5**/ns/ns  20.6 ns/ns/ns  42.8**/ns/ns  Sand a  204  42.2**  27.9 ns  29.9*  Water + sand e  248  39.1**/ns  23.8 ns/ns  37.1**/ns  TIBA + sand f  225  33.8**/ns/ns  23.6 ns/ns/ns  42.7**/**/ns  AIB + sand f  210  30.5**/*/ns  31.9*/ns/ns  37.6**/ns/ns  Total data scored 12 h after treatments in two experimental series are given for developmental stages of maize egg cells (A, B, C; see Fig. 3). **Significant at P < 0.01;* P < 0.05; ns, not significant. Significance of differences is given: a vs. no treatment; b vs. no treatment/vs. pollen; c vs. no treatment/vs. pollen/vs. water; d vs. no treatment/vs. pollen/vs. water + pollen; e vs. no treatment/vs. sand; f vs. no treatment/vs. sand/vs. water + sand. View Large Abbreviations ACC 1-aminocyclopropane-1-carboxylic acid AIB α-aminoisobutyric acid AVG l-α-(2-aminoethoxyvinyl)-glycine IAA indole-3-acetic acid STS silver thiosulfate TIBA 2,3,5-triiodobenzoic acid. References Beltrano, J., Carbone, A., Montaldi, E.R. and Y Guiamet, J.J. ( 1994) Ethylene is a promoter of wheat grain maturation and ear senescence. Plant Growth Regul.  15: 107–112. Google Scholar Bui, A.Q. and O’Neill, S.D. ( 1998) Three 1-aminocyclopropane-1-carboxylate synthase genes regulated by primary and secondary pollination signals in orchid flowers. Plant Physiol.  116: 419–428. Google Scholar Clark, D.G., Richards, C., Hilioti, Z., Lind-Iversen, S. and Brown, K. ( 1997) Effect of pollination on accumulation of ACC synthase and ACC oxidase transcripts, ethylene production and flower petal abscission in geranium (Pelargonium x hortorum L.H. Bailey). Plant Mol. Biol.  34: 855–865. Google Scholar De Martinis, D., Cotti, G., Te Lintel Hekker, S., Harren, F.J.M. and Mariani, C. ( 2002) Ethylene response to pollen tube growth in Nicotiana tabacum flowers. Planta  214: 806–812. Google Scholar Eder, J., Opatrná, J., Zazímalová, E., Kutácek, M. and Cermák, V. ( 1988) Endogenous free IAA levels in the early phase of wheat coleptile development. In Physiology and Biochemistry of Auxins in Plants. Edited by Kutácek, M., Bandurski, R.S. and Krekule, J. pp. 291–294. Academia, Prague. Google Scholar Heslop-Harrison, Y., Heslop-Harrison, J. and Reger, B.J. ( 1985) The pollen–stigma interaction in the grasses. 7. Pollen-tube guidance and the regulation of tube number in Zea mays L. Acta Bot. Neerl.  34: 193–211. Google Scholar Hoekstra, F.A and Weges, R. ( 1986) Lack of control by early pistillate ethylene of the accelerated wilting in Petunia hybrida flowers. Plant Physiol.  80: 403–408. Google Scholar Jones, M.L. and Woodson, R.W. ( 1997) Pollination-induced ethylene in carnation. Role of stylar ethylene in corolla senescence. Plant Physiol.  115: 205–212. Google Scholar Jones, M.L. and Woodson, R.W. ( 1999) Differential expression of three members of the 1-aminocyclopropane-1-carboxylate synthase gene family in carnation. Plant Physiol.  119: 755–764. Google Scholar Ketsa, S. and Rugkong, A. ( 2000) Ethylene production, senescence and ethylene sensitivity of Dendrobium ‘Pompadour’ flowers following pollination. J. Hort. Sci. Biotechnol.  75: 149–153. Google Scholar Larsen, P.B., Ashworth, E.N., Jones, M.L. and Woodson, W.R. ( 1995) Pollination-induced ethylene in carnation: role of pollen tube growth and sexual compatibility. Plant Physiol.  108: 1405–1412. Google Scholar Lei, C.H., Lindstrom, J.T. and Woodson, W.R. ( 1996) Reduction of 1-aminocyclopropane-1-carboxylic acid (ACC) in pollen by expression of ACC deaminase in transgenic petunias. Plant Physiol.  149 (Suppl): 111. Google Scholar Liu, Z.H. and Lee, B.H. ( 1995) Changes of endogenous indole-3-acetic acid, peroxidases, and auxin oxidases during pollen germination in maize. Bot. Bull. Acad. Sin.  36: 53–58. Google Scholar Lizada, C. and Yang, S.F. ( 1979) A simple and sensitive assay for 1-aminocyclopropane-1-carboxylic acid. Anal. Biochem.  100: 140–145. Google Scholar Llop-Tous, I., Barry, C.S. and Gierson, D. ( 2000) Regulation of ethylene biosynthesis in response to pollination in tomato flowers. Plant Physiol.  123: 971–978. Google Scholar Mól, R., Filek, M., Dumas, C. and Matthys-Rochon, E. ( 2004) Cytoplasmic calcium in silk trichomes after pollen grain reception and post-pollination changes of the electric potential in pistil tissues of maize. Plant Sci.  166: 1461–1469. Google Scholar Mól, R., Idzikowska, K., Dumas, C. and Matthys-Rochon, E. ( 2000) Late steps of egg cell differentiation are accelerated by pollination in Zea mays L. Planta  210: 749–757. Google Scholar Mól, R., Matthys-Rochon, E. and Dumas, C. ( 1994) The kinetics of cytological events during double fertilization in Zea mays L. Plant J.  5: 197–206. Google Scholar Nichols, R., Bufler, G., Mor, Y., Fujino, D.W. and Reid, M.S. ( 1983) Changes in ethylene production and 1-aminocyclopropane-1-carboxylic acid content of pollinated carnation flowers. J. Plant Growth Regul.  2: 1–8. Google Scholar O’Neill, S.D. ( 1997) Pollination regulation of flower development. Annu. Rev. Plant Physiol. Plant Mol. Biol.  48: 547–574. Google Scholar O’Neill, S.D., Nadeau, J.A., Zhang, X.S., Bui, A.Q. and Halevy, A.H. ( 1993) Interorgan regulation of ethylene biosynthetic genes by pollination. Plant Cell  5: 419–432. Google Scholar Pech, J.C., Latche, A., Larrigaudiere, C. and Reid, M.S. ( 1987) Control of early ethylene synthesis in pollinated petunia flowers. Plant Physiol. Biochem.  25: 431–437. Google Scholar Russell, S.D. ( 1992) Double fertilization. Int. Rev. Cytol.  140: 357–388. Google Scholar Sitbon, F. and Perrot-Rechenmann, C. ( 1997) Expression of auxin-related genes. Physiol. Plant.  100: 593–605. Google Scholar Smalle, J. and Van der Straeten, D. ( 1997) Ethylene and vegetative development. Physiol. Plant.  100: 443–455. Google Scholar Stead, A.D. ( 1992). Pollination-induced flower senescence: a review. Plant Growth Regul.  11: 13–20. Google Scholar Stepanova, A.N. and Ecker, J.R. ( 2000) Ethylene signaling: from mutants to molecules. Curr. Opin. Plant Biol.  3: 353–360. Google Scholar Taiz, L. and Zeiger, E. ( 1998) Plant Physiology. pp. 263, 552. Sinauer Associates Inc. Publishers, Sunderland. Google Scholar Ten Have, A. and Woltering, E.J. ( 1997) Ethylene biosynthetic genes are differentially expressed during carnation (Dianthus caryophyllus L.) flower senescence. Plant Mol. Biol.  34: 89–97. Google Scholar Van Doorn, W.G. ( 1997) Effects of pollination on floral attraction and longevity. J. Exp. Bot.  48: 1615–1622. Google Scholar Van Doorn, W.G. and Stead, A.D. ( 1997) Abscission of flowers and floral parts. J. Exp. Bot.  48: 821–837. Google Scholar Weterings, K., Pezzotti, M., Cornelissen, M. and Mariani, C. ( 2002) Dynamic 1-aminocyclopropane-1-carboxylate-synthase and -oxidase transcript accumulation patterns during pollen tube growth in tobacco styles. Plant Physiol.  130: 1190–1200. Google Scholar Woltering, E.J., De Vrije, T., Harren, F. and Hoekstra, F.A. ( 1997) Pollination and stigma wounding: same response, different signal? J. Exp. Bot.  48: 1027–1033. Google Scholar Woltering, E.J., Somhorst, D. and Van der Veer, P. ( 1995) The role of ethylene in inter-organ signaling during flower senescence. Plant Physiol.  109: 1219–1225. Google Scholar Yang, S.F. and Hoffman, N.E. ( 1984) Ethylene biosynthesis and its regulation in higher plants. Annu. Rev. Plant Physiol.  35: 155–189. Google Scholar Yoon, I.S., Mori, H., Kim, J.H., Kang, B.G. and Imaseki, H. ( 1997) VR-ACS6 is an auxin-inducible 1-aminocyclopropane-1-carboxylate synthase gene in mungbean (Vigna radiata). Plant Cell Physiol.  38: 217–224. Google Scholar Young, B.A., Sherwood, R.T. and Bashaw, E.C. ( 1979) Cleared-pistil and thick-sectioning techniques for detecting aposporous apomixes in the grasses. Can. J. Bot.  57: 1668–1672. Google Scholar Young, T.E., Gallie, D.R. and DeMason, D.A. ( 1997) Ethylene-mediated programmed cell death during maize endosperm development of wild-type and shrunken2 genotypes. Plant Physiol.  115: 737–751. Google Scholar Zhang, X.S. and O’Neill, S.D. ( 1993) Ovary and gametophyte development are coordinately regulated by auxin and ethylene following pollination. Plant Cell  5: 403–418. Google Scholar

Okamoto, Takashi;Higuchi, Kanako;Shinkawa, Takashi;Isobe, Toshiaki;Lörz, Horst;Koshiba, Tomokazu;Kranz, Erhard

doi: 10.1093/pcp/pch161pmid: 15564524

Abstract In most flowering plants, the female gametophyte develops in an ovule deeply embedded in the ovary. Through double fertilization, the egg cell fuses with the sperm cell, resulting in a zygote, which develops into the embryo. In the present study, we analyzed egg cell lysates by polyacrylamide gel electrophoresis and subsequent mass spectrometry-based proteomics technology, and identified major protein components expressed in the egg cell. The identified proteins included three cytosolic enzymes of the glycolytic pathway, glyceraldehyde-3-phosphate dehydrogenase, 3-phosphoglycerate kinase and triosephosphate isomerase, two mitochondrial proteins, the ATP synthase β-subunit and an adenine nucleotide transporter, and annexin p35. In addition, expression levels of these proteins in the egg cell were compared with those in the early embryo, the central cell and the suspension cell. Annexin p35 was highly expressed only in the egg cell, and glyceraldehyde-3-phosphate dehydrogenase, 3-phosphoglycerate kinase and the adenine nucleotide transporter were expressed at higher levels in egg cells than in central and cultured cells. These results indicate that annexin p35 in the egg cell and zygote is involved in the exocytosis of cell wall materials, which is induced by a fertilization-triggered increase in cytosolic Ca2+ levels, and that the egg cell is rich in an enzyme subset for the energy metabolism. (Received May 19, 2004; Accepted July 18, 2004) Introduction Egg cells in higher plants are highly differentiated haploid cells, which are fertilized with sperm cells and undergo subsequent early embryogenesis. In angiosperms, the female gametophyte, also referred to as the embryo sac or the megagametophyte, develops in an ovule embedded within the ovary. Although among angiosperms the female gametophyte has a variety of forms, the most common consists of seven cells composed of four cell types: one egg cell, one central cell, two synergid cells and three antipodal cells (Huang and Russell 1992, Drews and Yadegari 2002). Upon double fertilization, one sperm cell from the pollen grain fuses with the egg cell, and the resulting zygote develops into an embryo. The central cell fuses with the second sperm cell to form a triploid primary endosperm, which develops into the endosperm (Nawaschin 1898, Guignard 1899, Russell 1992). In general, the composition of cellular proteins differs depending on cell type. For example, mesophyll cells have a large amount of ribulose-1,5-bisphosphate carboxylase/oxygenase for the fixation of carbon dioxide, while the cotyledon cells of non-endospermic seeds such as legume seeds abundantly contain storage globulins and albumins, which supply the nutrient source for hypocotyl growth during seed germination and seedling growth (Bewley and Black 1994). These indicate that the major proteins in such highly differentiated cells reflect the biological function of the cells. Therefore, identification of the major protein components in egg cells will provide basic knowledge of their character. In addition, identification of the major proteins in the egg cell will give a cue for analyzing the mechanisms of female gametogenesis, fertilization and early embryogenesis in higher plants. Unlike in animals and lower plants, higher plant egg cells are located in the embryo sac, which is deeply embedded in ovular tissue. Methods were developed for the isolation of embryo sacs and egg cells in a wide range of higher plant species (for review see Theunis et al. 1991). However, biochemical analyses of egg cells of higher plants at the protein level have not been performed to our knowledge due to the limited amount of such isolated cells. Nevertheless, in maize, routinely 20–40 egg cells can be isolated/experienced by one experimenter per day, and, under optimal conditions up to 60 egg cells can be obtained by one person per day (Kranz 1999). Despite this relatively small amount of plant material, recent advances in proteomics technologies provide the possibility of identifying proteins in such cells. In this study, we detected traceable amounts of proteins in a small number of the egg cells by minimizing the size of gels for one- and two-dimensional polyacrylamide gel electrophoresis and identified major protein components by highly sensitive liquid chromatography with tandem mass spectrometry (LC-MS/MS) technology. We show here that the egg cell abundantly contains three cytosolic glycolytic enzymes, mitochondrial ATP synthase β-subunit, adenine nucleotide translocator and annexin p35, and discuss possible functions of these proteins in the cell. Results Identification of major proteins by SDS-PAGE and subsequent mass spectrometry Isolated egg cells (Fig. 1A) were extensively washed to eliminate contamination of proteins in the enzymic solutions, which were used during isolation of the cells as described in Materials and Methods. Proteins from 75 egg cells were separated by 12.5% SDS-PAGE and the gel was silver stained. Protein bands were successfully detected possibly due to the small-sized gel, in which proteins are concentrated more efficiently than in a normal-sized gel. The band pattern was almost identical in repetitive experiments. By comparing the intensity of protein bands from the egg cells with that of the molecular weight marker co-migrated on the gel, the amount of protein in an egg cell was roughly estimated to be 100–200 pg (data not shown). Major protein bands, assigned as bands 1–7 in Fig. 1B, were excised from the gel, in-gel digested with trypsin, and the resulting peptide mixtures were analyzed by direct nano-flow LC-MS/MS. Two doubly charged peptide ions with m/z 518.34 and 696.38 were observed in band 5. Database analysis of the MS/MS spectrum of the peptide ion with m/z 518.34 showed that it corresponded to the LIISILAHR sequence of maize annexin p35 at residues 33–40 (Table 1). Manual assignment of the fragment ions also yielded the same sequence (Fig. 1C). Likewise, the MS/MS spectrum of the other peptide ion with m/z 696.38 was assigned to the ADPKDEFLSTLR sequence of maize annexin p35 at residues 222–233 (Table 1), confirming that band 5 corresponds to annexin p35, which is thought to be involved in exocytosis and vesicle trafficking in plant cells (Carroll et al. 1998, Battey et al. 1999, Clark et al. 2001). The results obtained from LC-MS/MS analysis for bands 2–6 are summarized in Table 1. The proteins of 39 kDa (band 4) and 42 kDa (band 3) were identified as cytosolic glyceraldehyde-3-phosphate dehydrogenase and 3-phosphoglycerate kinase, respectively, which are known to be responsible for glycolysis (Plaxton 1996). Bands 2 and 6 corresponded to mitochondrial ATP synthase β-subunit and adenine nucleotide translocator, respectively. Although mitochondrial adenine nucleotide translocator was identified on the basis of a single peptide (Table 1), the molecular weight of the protein has been estimated as 30.5 kDa by the electrophoretic mobility in the SDS-PAGE gel (Winning et al. 1992), which is consistent with the mobility of band 6 (Fig. 1B). This supports the possibility that band 6 corresponds to mitochondrial adenine nucleotide translocator. Four of the seven major proteins analyzed were thought to be involved in energy metabolism, such as glycolysis and ATP production/transport, within the cell. Database analysis of the LC-MS/MS spectrum of peptides from band 1 indicated that this protein has a SSVLESLAGISLPR sequence, which is identical to the Arabidopsis hypothetical protein (At1g60500.1) at residues 79–92, however, no maize protein was detected by the database search (data not shown). The protein of band 7 could not be identified. In addition to bands 1–7, an attempt was made to determine the first structure of the protein bands with weak intensity. The gel region between bands 1 and 2 (indicated by blanket, No 8 in Fig. 1B) were excised and trypsin-digested, and the resulting peptides were analyzed with LC-MS/MS. But the proteins could not be identified by our LC-MS/MS system, although the system has extremely high sensitivity. This indicates that only major proteins can be analyzed using LC-MS/MS when proteins from 75 egg cells are used as materials. However, this analytical limitation confirms that the identified proteins listed in Table 1 are not derived from minor proteins overlapping with the major proteins in the gel, but from major proteins themselves. Identification of major proteins by 2D-PAGE and subsequent mass spectrometry Proteins from 180 egg cells were separated by isoelectric focusing and subsequent SDS-PAGE. Protein spots stained with silver were successfully detected, and eight spots were selected for in-gel tryptic digestion and subsequent analysis by LC-MS/MS (Fig. 1D). The profile of protein spots in the gel was similar in repetitive experiments. The results from LC-MS/MS analyses are summarized in Table 2. Spot 1 was determined as cytosolic glyceraldehyde-3-phosphate dehydrogenase, which is identical to band 4 in Fig. 1B, and spot 6 was determined as cytosolic 3-phosphoglycerate kinase, which corresponds to band 3 in Fig. 1A. Spot 5 was identified as cytosolic triosephosphate isomerase, which also belongs to the enzymes of the glycolytic pathway. Calculated molecular mass and isoelectric point of cytosolic triosephosphate isomerase (accession number GI136063) are 27,292 and 5.52, respectively. These values fit the position of the gel where spot 5 was detected (Fig. 1D), supporting the conclusion that the protein spot corresponds to cytosolic triosephosphate isomerase although only a single peptide was detected by LC-MS/MS analysis (Table 2). A doubly charged peptide ion with m/z 617.85 was observed in spot 2, and database analysis of the LC-MS/MS spectrum of this ion showed a KIYETKILVK sequence (data not shown), which is identical to tomato cystatin at residues 226–235 (PIR accession number T06323). However, database analysis did not hit with maize cystatin. The result suggests that spot 2 corresponds to a novel maize cystatin, which has not been identified, or to an unknown maize protein containing the KIYETKILVK sequence. For spots 3, 4, 7 and 8, proteins could not be identified. Some major proteins detected in the SDS-PAGE gel were not detected in that of 2D-PAGE (Fig. 1B, D) probably due to the narrow pI range of isoelectric focusing (pI 4.5–7). Comparison of protein profiles from egg cells with those from early embryos, central and cultured cells The modified silver staining method was approximately five times less sensitive than the conventional method, since fixative in the modified procedures does not contain glutaraldehyde (Taoka et al. 2000). Although 75 egg cells were subjected to SDS-PAGE for subsequent LC-MS/MS analyses in Fig. 1B, 15 egg cells were enough to visualize the proteins in SDS-PAGE gels with conventional silver staining (Fig. 2A–C). The protein profiles of the egg cells in the SDS-PAGE gel were compared with those of two-celled or multicellular embryos produced in vitro to see whether the expression levels of the five identified proteins (bands 2–6 in Fig. 1B and Table 1) change after in vitro fertilization and during early embryogenesis. Annexin p35 was strongly expressed in the egg cells, but largely decreased in the two-celled and multicellular embryos (band 5 in Fig. 2A). Expression levels of the other proteins remained unchanged after fertilization and during early embryogenesis (bands 2–4 and 6 in Fig. 2A). Next, protein profiles were compared between the egg and central cells. In the central cells, the band intensities for cytosolic 3-phosphoglycerate kinase and cytosolic glyceraldehyde-3-phosphate dehydrogenase were weaker than those in the egg cells (bands 3 and 4 in Fig. 2B). Furthermore, the protein corresponding to annexin p35 was hardly detected in the central cells (band 5 in Fig. 2B). Finally, we compared the profile with cultured maize cells, which are neither gametophytic nor embryonic. Annexin p35 was not observed in the cultured cells (band 5 in Fig. 2C). Moreover, cytosolic 3-phosphoglycerate kinase and mitochondrial adenine nucleotide translocator were hardly detected in the cultured cells (bands 3 and 6 in Fig. 2C). Discussion Three cytosolic enzymes for the glycolytic pathway, glyceraldehyde-3-phosphate dehydrogenase, 3-phosphoglycerate kinase and triosephosphate isomerase, and two mitochondrial proteins, an ATPase β-subunit and an adenine nucleotide transporter, and annexin p35 were identified as major proteins in maize egg cells (Tables 1, 2). Of these six proteins, annexin p35 was strongly expressed only in the egg cells (Fig. 2A–C). Annexins are Ca2+ and phospholipid binding proteins, and extensive studies of the proteins in animal cells have shown their multifunctional roles in essential cellular processes such as membrane trafficking, ion transport, mitotic signaling, cytoskeleton rearrangement and DNA replication (reviewed in Gerke and Moss 2002). Plant annexins share the basic properties of Ca2+-dependent membrane binding molecules and are structurally similar to their animal counterparts (Pirck et al. 1994, Clark and Roux 1995, Battey et al. 1996). Exocytosis and the Golgi-mediated secretion of newly synthesized plasma membranes and cell wall materials have been reported as the function of annexin in plant cells (Carroll et al. 1998, Battey et al. 1999, Clark et al. 2001). It has been demonstrated that cell wall formation around the zygote starts 30 s after in vitro fusion of the egg with a sperm cell (Kranz et al. 1995). This rapid formation of the cell wall around the zygote suggests that cell wall materials are stored in the egg cells before fertilization, and are secreted via possible exocytosis after fertilization. It is well known that Ca2+ exerts the regulation of exocytosis in plant and animal cells (Bush 1995, Battey et al. 1999), and Carroll et al. (1998) reported that Ca2+-stimulated exocytosis in root cap cells is enhanced by exogenously applied annexin p35, suggesting that annexin is involved in Ca2+-stimulated exocytosis. It has also been revealed that concentrations of cytosolic Ca2+ in maize egg cell/zygote increase after fertilization (Digonnet et al. 1997) possibly via influxes of extracellular Ca2+ (Antoine et al. 2000). When a fluorescent Ca2+ indicator (Kao et al. 1989) was used to monitor intracellular Ca2+ levels in egg and zygote cells, it was observed that levels reached a maximum 85 s after in vitro fertilization (Digonnet et al. 1997). Annexin p35, existing abundantly in egg cells, might play a role in exocytosis, which is stimulated by fertilization-induced increases in Ca2+ levels in the zygote, for rapid cell wall formation around the zygote. In contrast to animal mitochondria, which respire fatty acids and glycolytically derived pyruvate, plant mitochondria rarely respire fatty acids (reviewed in Plaxton 1996). This indicates that glycolysis is of crucial importance in plants because it is the predominant pathway supplying ‘fuels’ for plant respiration. Recently, it was revealed that seven glycolytic enzymes, including glyceraldehyde-3-phosphate dehydrogenase and triosephosphate isomerase, are associated with the outer membranes of mitochondria, suggesting that such microcompartmentation of glycolysis allows pyruvate to be provided directly into the mitochondrion (Giege et al. 2003). In mitochondria, ATPase synthesizes ATP, which is the principal energy source for the cells, via an H+ gradient between the inner and outer membranes, and the resultant ATP is exchanged with cytosolic ADP by adenine nucleotide transporters (Vignais 1976, Mozo et al. 1995). Giant and polymorphic mitochondria have been observed in egg cells of maize (Faure et al. 1992) and geranium (Kuroiwa and Kuroiwa 1992), indicating that identification of two mitochondrial proteins as major proteins in maize egg may reflect such well-developed mitochondria. Five of the six major egg proteins identified in this study are thought to be involved in the cytosolic and mitochondrial energy production pathways, suggesting that the egg cell has sufficient enzymes and transporters to produce and transport an energy source. After in vitro fusion of the maize egg with a sperm cell, the majority of cytoplasmic organelles migrate towards the zygote nucleus, cell wall is actively formed, and duplication and division of the nucleus occur as part of the early cytological events in the zygote (Kranz et al. 1995). These energy-consuming serial zygotic events might explain why these cells abundantly contain proteins for energy production. Interestingly, it has been reported that glycolysis in the mouse oocyte is activated by fertilization (Urner and Sakkas 1999). Activation of glycolysis may occur in maize zygote after fertilization of the egg cell with the sperm cell. Expression levels of the three glycolytic enzymes and two mitochondrial proteins in the egg cells were identical to those in two-celled and multicellular embryos (Fig. 2A). This is probably the result of early embryogenesis, which requires a large quantity of energy for embryonic development. Glyceraldehyde-3-phosphate dehydrogenase and 3-phosphoglycerate kinase were expressed at a higher level in the egg cells than in the central cells (Fig. 2B), while expression levels of 3-phosphoglycerate kinase and adenine nucleotide transporter was low in the cultured cells (Fig. 2C). These results might indicate that early embryos, as well as egg cells, are rich in the enzyme subset for energy metabolism. However, in the present study, the expression levels of the identified proteins were compared between egg cells and other cell types on the basis of the intensity of the silver-stained band with the same mobility in the SDS-PAGE gels. This will be insufficient to elucidate the specificity of the protein bands. To know whether the protein bands observed in early embryos, central cells and cultured cells correspond to the identified proteins in egg cells, further analysis such as Western blotting with specific antibody or/and RT-PCR should be conducted in the near future. In fact, we have tried to synthesize cDNA from a small number of the cells. To our knowledge, this is the first report to identify the major proteins in angiosperm egg cells. Using the small-sized 2D-PAGE, comparisons of the protein profiles between egg cells and zygotes are currently underway in our laboratories. LC-MS/MS analyses of the protein spots, which are detected only in zygotes or egg cells, will reveal newly synthesized, modified or rapidly degraded proteins in the zygotes. This will provide a novel insight into zygote development and early embryogenesis in higher plants. Materials and Methods lsolation and selection of egg and central cells Ears from the inbred maize (Zea mays) line A188 (courtesy of A. Pryor, CSIRO, Canberra, Australia) were used for isolating egg and central cells. Egg cells were isolated as described previously (Kranz et al. 1991, Kranz 1999). Isolated egg cells were washed four times by transferring the cells into fresh droplets of mannitol solution (650 mosmol kg–1 H2O) on coverslips. Between 10 and 50 isolated egg cells were transferred into a 1 µl droplet of SDS-sample buffer (2% SDS, 25 mM Tris-HCl pH 6.8, 30% glycerol, 5% 2-mercaptoethanol) for SDS-PAGE, or into a 1 µl droplet of lysis buffer [8 M urea, 5% 2-mercaptoethanol, 2% ampholine pH 3.5–10 (Amersham), 2% Nonidet P40] for 2D-PAGE. These samples were stored at –80°C until use. Central cells were isolated according to the method previously described (Kranz et al. 1998). Isolated central cells were washed as above, and four isolated central cells were transferred into a 6 µl droplet of SDS-sample buffer. Electrofusion and culture procedures Sperm and egg cells were isolated from pollen grains and the ears of maize, respectively, as described (Kranz et al. 1991, Kranz 1999). A pair of a sperm and an egg cell protoplast were fused electrically under microscopical observation (Kranz and Lörz 1993). Fusion products were cultured on the transparent, semipermeable membranes of Millicell-CM dishes (diameter 12 mm; Millipore, Bedford, MA, U.S.A.) as described (Kranz et al. 1991). After 40–50 h of culture, two-celled embryos were harvested and washed three times by transferring them into fresh droplets of culture medium. Two to five isolated two-celled embryos were transferred into a 1 µl droplet of SDS-sample buffer. The zygotes developed into multicellular embryos after 3 d of culture, and after washing, three to five of these embryos were transferred into a 1 µl droplet of SDS-sample buffer. One- and two-dimensional electrophoresis According to Laemmli (1970), 12.5% SDS-polyacrylamide gels were prepared in a small mold (50×60×1 mm; Atto, Tokyo, Japan). Seventy five or 15 egg cells dissolved in 6 µl of SDS-sample buffer were applied to the SDS-PAGE gel. The isoelectric focusing gel [8 M urea, 3.5% acrylamide, 0.18% bis-acrylamide, 5% (v:v) ampholine pH 3.5–10 and 2% Nonidet P40] was prepared using a thin glass capillary (length, 5 cm; diameter, 1 mm). One hundred and eighty egg cells dissolved in 6 µl of lysis buffer were applied to the capillary gel. After isoelectric focusing, the proteins in the capillary gel were separated further by 12.5% SDS-PAGE. The proteins in the gel used for in-gel tryptic digestion and subsequent analysis with LC-MS/MS were visualized by modified silver staining according to Taoka et al. (2000). In other cases, the proteins in the SDS-PAGE gel were detected by conventional silver staining procedures (Oakley et al. 1980). Identification of proteins by tandem mass spectrometry Protein bands or spots were excised from the SDS-PAGE gel, in-gel digested with trypsin, and subjected to direct nano-flow LC-MS/MS analysis for protein identification. The chromatography was performed on a nano ESI column (inside diameter, 150 µm × 30 mm) packed with a C18 reversed phase medium (Mightysil-C18, 3 µm; Kanto Chemicals, Tokyo, Japan) using a linear gradient from 0 to 70% acetonitrile in 0.1% formic acid for 35 min at a flow rate of 100 nl/min, and the separated peptides were directly sprayed into a hybrid mass spectrometer equipped with an electrospray source (Q-Tof ultima; Micromass-Waters, Milford, MA, U.S.A.). Electrospray ionization was carried out at a voltage of 1.5 kV, and MS/MS spectra were automatically acquired in data-dependent mode during the entire run. All MS/MS spectra were correlated by the search engine, Mascot program (Matrixscience, London, U.K.), against the non-redundant protein sequence database at the National Center for Biotechnology Information (National Institutes of Health). Each high-scoring peptide sequence was confirmed by manual inspection of the corresponding MS/MS spectrum to ensure that the match was correct. Acknowledgments We thank Marlis Nissen and Petra von Wiegen for their excellent technical help in the isolation of ovular tissues and gametes. We thank Dr. Stefan Scholten for the discussions and the suggestion about the protein analysis. This work was supported in part by Grants-in-Aid from the Ministry of Education, Science, Sport, and Culture of Japan (grants 15031222 to T.K. and 16027242 to T.O.). T.O. was supported by a JSPS Postdoctoral Fellowship for Research Abroad. 4 Corresponding author: E-mail, [email protected]; Fax, +81-426-77-2559. View largeDownload slide Fig. 1 SDS-PAGE and 2D-PAGE of maize egg cell proteins. (A) Isolated egg cell. Bar = 50 µm. (B) Proteins from 75 egg cells were separated by SDS-PAGE followed by a modified silver-staining procedure. Numbers to the right of the arrowheads indicate the protein bands subjected to in-gel tryptic digestion and subsequent LC-MS/MS. Numbers to the right of the bracket indicates the gel region subjected to in-gel tryptic digestion and LC-MS/MS. (C) Identification of annexin p35 as a major protein in maize egg cells. The doubly charged ions of the tryptic peptides (m/z = 518.34) from a major protein in the egg cells (band 5 in Fig. 1A) were analyzed by LC-MS/MS. The amino acid sequences were verified by interpreting the b-type (italics) and y-type (normal text) production series as indicated in the figure. (D) Proteins from 180 egg cells were separated by 2D-PAGE followed by modified silver staining. Numbers around the arrowheads indicate the protein spots subjected to in-gel tryptic digestion and LC-MS/MS. View largeDownload slide Fig. 1 SDS-PAGE and 2D-PAGE of maize egg cell proteins. (A) Isolated egg cell. Bar = 50 µm. (B) Proteins from 75 egg cells were separated by SDS-PAGE followed by a modified silver-staining procedure. Numbers to the right of the arrowheads indicate the protein bands subjected to in-gel tryptic digestion and subsequent LC-MS/MS. Numbers to the right of the bracket indicates the gel region subjected to in-gel tryptic digestion and LC-MS/MS. (C) Identification of annexin p35 as a major protein in maize egg cells. The doubly charged ions of the tryptic peptides (m/z = 518.34) from a major protein in the egg cells (band 5 in Fig. 1A) were analyzed by LC-MS/MS. The amino acid sequences were verified by interpreting the b-type (italics) and y-type (normal text) production series as indicated in the figure. (D) Proteins from 180 egg cells were separated by 2D-PAGE followed by modified silver staining. Numbers around the arrowheads indicate the protein spots subjected to in-gel tryptic digestion and LC-MS/MS. View largeDownload slide Fig. 2 Comparison of the protein profiles of the egg cells with those of the early embryo (A), the central cell (B) and cultured cell (C). (A) Proteins from 15 egg cells (lane 1), 15 two-celled embryos (lane 2) and 15 multicellular embryos (lane 3) were separated by SDS-PAGE. Proteins in the gel were visualized with conventional silver staining. Numbers to the left of the arrowheads are equivalent to those in Fig. 1B. (B) Proteins from 15 egg cells (lane 1) and four central cells (lane 2) were separated by SDS-PAGE, and the gel was silver stained. (C) Proteins from 15 egg cells (lane 1) and maize cultured cells (lane 2; Kranz et al. 1991) were separated by SDS-PAGE followed by silver staining. View largeDownload slide Fig. 2 Comparison of the protein profiles of the egg cells with those of the early embryo (A), the central cell (B) and cultured cell (C). (A) Proteins from 15 egg cells (lane 1), 15 two-celled embryos (lane 2) and 15 multicellular embryos (lane 3) were separated by SDS-PAGE. Proteins in the gel were visualized with conventional silver staining. Numbers to the left of the arrowheads are equivalent to those in Fig. 1B. (B) Proteins from 15 egg cells (lane 1) and four central cells (lane 2) were separated by SDS-PAGE, and the gel was silver stained. (C) Proteins from 15 egg cells (lane 1) and maize cultured cells (lane 2; Kranz et al. 1991) were separated by SDS-PAGE followed by silver staining. Table 1 Major proteins of maize egg cells identified by SDS-PAGE and subsequent tandem mass spectrometry Band number  Protein  Accession number (GI)  Peptide m/z  Charge  Sequence determined  residues  2  Mitochondrial ATP synthase β-chain  114420  705.41  2  VLNTGSPITVPVGR  147–160        729.43  2  TVLIMELINNVAK  235–247  3  Cytosolic 3-phosphoglycerate kinase  28172915  868.01  2  LAAALPEGGVLLLENVR  31–48        694.89  2  ELDYLVGAVANPK  105–117  4  Cytosolic glyceraldehyde-3-phosphate dehydrogenase   6016075  559.66  2  TLLFGEKPVTVFGIR  68–82        749.94  2  VPTVDVSVVDLTVR  237–250  5  Annexin P35  7441507  518.34  2  LIISILAHR  33–40        696.38  2  ADPKDEFLSTLR  222–233  6  Mitochondrial adenine nucleotide translocator  22166  723.89  2  YFPTQALNFAFK  161–172  Band number  Protein  Accession number (GI)  Peptide m/z  Charge  Sequence determined  residues  2  Mitochondrial ATP synthase β-chain  114420  705.41  2  VLNTGSPITVPVGR  147–160        729.43  2  TVLIMELINNVAK  235–247  3  Cytosolic 3-phosphoglycerate kinase  28172915  868.01  2  LAAALPEGGVLLLENVR  31–48        694.89  2  ELDYLVGAVANPK  105–117  4  Cytosolic glyceraldehyde-3-phosphate dehydrogenase   6016075  559.66  2  TLLFGEKPVTVFGIR  68–82        749.94  2  VPTVDVSVVDLTVR  237–250  5  Annexin P35  7441507  518.34  2  LIISILAHR  33–40        696.38  2  ADPKDEFLSTLR  222–233  6  Mitochondrial adenine nucleotide translocator  22166  723.89  2  YFPTQALNFAFK  161–172  View Large Table 2 Major proteins of maize egg cells identified by 2D-PAGE and subsequent tandem mass spectrometry Spot number  Protein  Accession number (GI)  Peptide m/z  Charge  Sequence determined  Residues  1  Cytosolic glyceraldehyde-3-phosphate dehydrogenase  6016075  870.46  2  VIHDNFGIIEGLMTTVHAITATQK  165–188        749.93  2  VPTVDVSVVDLTVR  237–250  5  Cytosolic triosephosphate isomerase  136063  684.36  2  IIYGGSVTAANCK  207–219  6  Cytosolic 3-phosphoglycerate kinase  28172915  868.01  2  LAAALPEGGVLLLENVR  31–48        694.88  2  ELDYLVGAVANPK  105–117        787.42  2  GVTTIIGGGDSVAAVEK  277–293  Spot number  Protein  Accession number (GI)  Peptide m/z  Charge  Sequence determined  Residues  1  Cytosolic glyceraldehyde-3-phosphate dehydrogenase  6016075  870.46  2  VIHDNFGIIEGLMTTVHAITATQK  165–188        749.93  2  VPTVDVSVVDLTVR  237–250  5  Cytosolic triosephosphate isomerase  136063  684.36  2  IIYGGSVTAANCK  207–219  6  Cytosolic 3-phosphoglycerate kinase  28172915  868.01  2  LAAALPEGGVLLLENVR  31–48        694.88  2  ELDYLVGAVANPK  105–117        787.42  2  GVTTIIGGGDSVAAVEK  277–293  View Large Abbreviations !LC-MS/MS liquid chromatography with tandem mass spectrometry. References Antoine, A.F., Faure, J.E., Cordeiro, S., Dumas, C., Rougier, M. and Feijo, J.A. ( 2000) A calcium influx is triggered and propagates in the zygote as a wavefront during in vitro fertilization of flowering plants. Proc. Natl Acad. Sci. USA  97: 10643–10648. Google Scholar Battey, N.H., James, N.C. and Greenland, A.J. ( 1996) cDNA isolation and gene expression of the maize annexins p33 and p35. Plant Physiol.  112: 1391–1396. Google Scholar Battey, N.H., James, N.C., Greenland, A.J. and Brownlee, C. ( 1999) Exocytosis and endocytosis. Plant Cell  11: 643–660. Google Scholar Bewley, J.D. and Black, M. ( 1994) Seed: Physiology of Development and Germination. Plenum Press, New York. Google Scholar Bush, D.S. ( 1995) Calcium regulation in plant cells and its role in signaling. Annu. Rev. Plant Physiol. Plant Mol. Biol.  46: 95–122. Google Scholar Carroll, A.D., Moyen, C., Van Kesteren, P., Tooke, F., Battey, N.H. and Brownlee, C. ( 1998) Ca2+, annexins, and GTP modulate exocytosis from maize root cap protoplasts. Plant Cell  10: 1267–1276. Google Scholar Clark, G.B. and Roux, S.J. ( 1995) Annexins of plant cells. Plant Physiol.  109: 1133–1139. Google Scholar Clark, G.B., Sessions, A., Eastburn, D.J. and Roux, S.J. ( 2001) Differential expression of members of the annexin multigene family in Arabidopsis. Plant Physiol.  126: 1072–1084. Google Scholar Digonnet, C., Aldon, D., Leduc, N., Dumas, C. and Rougier, M. ( 1997) First evidence of a calcium transient in flowering plants at fertilization. Development  124: 2867–2874. Google Scholar Drews, G.N. and Yadegari, R. ( 2002) Development and function of the angiosperm female gametophyte. Annu. Rev. Genet.  36: 99–124. Google Scholar Faure, J.-E., Mogensen, H.L., Kranz, E., Digonnet, C. and Duma, C. ( 1992) Ultrastructural characterization and three-dimensional reconstruction of isolated maize (Zeamaize L.) egg cell protoplasts. Protoplasma  171: 97–103. Google Scholar Gerke, V. and Moss, S.E. ( 2002) Annexins: from structure to function. Physiol. Rev.  82: 331–371. Google Scholar Giege, P., Heazlewood, J.L., Roessner-Tunali, U., Millar, A.H., Fernie, A.R., Leaver, C.J. and Sweetlove, L.J. ( 2003) Enzymes of glycolysis are functionally associated with the mitochondrion in Arabidopsis cells. Plant Cell  15: 2140–2151. Google Scholar Guignard, M.L. ( 1899) Sur les antherozoides et la double copulation sexuelle chez les vegetaux angiosperms. Rev. Gen. Bot.  11: 129–135. Google Scholar Huang, B.-Q. and Russell, S.D. ( 1992) Female germ unit: organization isolation, and function. Int. Rev. Cytol.  140: 233–292. Google Scholar Kao, J.P., Harootunian, A.T. and Tsien, R.Y. ( 1989) Photochemically generated cytosolic calcium pulses and their detection by fluo-3. J. Biol. Chem.  264: 8179–8184. Google Scholar Kranz, E. ( 1999) In vitro fertilization with isolated single gametes. Methods Mol. Biol.  111: 259–267. Google Scholar Kranz, E., Bautor, J. and Lörz, H. ( 1991) In vitro fertilization of single, isolated gametes of maize mediated by electrofusion. Sex Plant Reprod.  4: 12–16. Google Scholar Kranz, E. and Lörz, H. ( 1993) In vitro fertilization with isolated, single gametes results in zygotic embryogenesis and fertile maize plants. Plant Cell  5: 739–746. Google Scholar Kranz, E., von Wiegen, P. and Lörz, H. ( 1995) Early cytological events after induction of cell division in egg cells and zygote development following in vitro fertilization with angiosperm gametes. Plant J.  8: 9–23. Google Scholar Kranz, E., von Wiegen, P., Quader, H. and Lorz, H. ( 1998) Endosperm development after fusion of isolated, single maize sperm and central cells in vitro. Plant Cell  10: 511–524. Google Scholar Kuroiwa, H. and Kuroiwa, T. ( 1992) Giant mitochondria in the mature egg cell of Pelargonium zonale. Protoplasma  168: 184–188. Google Scholar Laemmli, U.K. ( 1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature  227: 680–685. Google Scholar Mozo, T., Fischer, K., Flugge, U.I. and Schmitz, U.K. ( 1995) The N-terminal extension of the ADP/ATP translocator is not involved in targeting to plant mitochondria in vivo. Plant J.  7: 1015–1020. Google Scholar Nawaschin, S. ( 1898) Revision der Befruchtungsvorgange bei Lilium martagon und Fritilla tenella. Bull. Acad. Imp. Sci. St. Petersburg  9: 377–382. Google Scholar Oakley, B.R., Kirsch, D.R. and Morris, N.R. ( 1980) A simplified ultrasensitive silver stain for detecting proteins in polyacrylamide gels. Anal. Biochem.  105: 361–363. Google Scholar Pirck, M., Hirt, H. and Heberle-Bors, E. ( 1994) The cDNA sequence encoding an annexin from Medicago sativa. Plant Physiol.  104: 1463–1464. Google Scholar Plaxton, W. ( 1996) The organization and regulation of plant glycolysis. Annu. Rev. Plant Physiol. Plant Mol. Biol.  47: 185–214. Google Scholar Russell, S.D. ( 1992) Double fertilization. Int. Rev. Cytol.  40: 357–390. Google Scholar Taoka, M., Wakamiya, A., Nakayama, H. and Isobe, T. ( 2000) Protein profiling of rat cerebella during development. Electrophoresis  21: 1872–1879. Google Scholar Theunis, C.H., Pierson, E.S. and Cresti, M. ( 1991) Isolation of male and female gametes in higher plants. Sex. Plant Reprod.  4: 145–154. Google Scholar Urner, F. and Sakkas, D. ( 1999) Characterization of glycolysis and pentose phosphate pathway activity during sperm entry into the mouse oocyte. Bio. Reprod.  60: 973–978. Google Scholar Vignais, P.V. ( 1976) Molecular and physiological aspects of adenine nucleotide transport in mitochondria. Biochim. Biophys. Acta  456: 1–38. Google Scholar Winning, B.M., Sarah, C.J., Purdue, P.E., Day, C.D. and Leaver, C.J. ( 1992) The adenine nucleotide translocator of higher plants is synthesized as a large precursor that is processed upon import into mitochondria. Plant J.  2: 763–773. Google Scholar

Gray, Gordon R.;Villarimo, Alicia R.;Whitehead, Carmen L.;McIntosh, Lee

doi: 10.1093/pcp/pch162pmid: 15564525

Abstract Many metabolic reactions are coupled to NADPH in the mitochondrial matrix, including those involved in thiol group reduction. One enzyme linked to such processes is mitochondrial NADP+-dependent isocitrate dehydrogenase (mtICDH; EC 1.1.1.42), although the precise role of this enzyme is not yet known. Previous work has implicated mtICDH as part of a biochemical mechanism to reductively activate the alternative oxidase (AOX). We have partially purified mtICDH from tobacco (Nicotiana tabacum L. cv. Petit Havana SR1) cell suspension cultures and localized this to a 46-kDa protein on SDS–PAGE, which was verified by peptide sequencing. In the inflorescence of the aroid Sauromatum guttatum Schott (voodoo lily), mtICDH appears to be developmentally regulated, presenting maximal specific activity during the thermogenic period of anthesis when the capacity for AOX respiration is also at its peak. Transgenic tobacco plants were generated that overexpress mtICDH and lines were obtained that demonstrated up to a 7-fold increase in mtICDH activity. In isolated mitochondria, this resulted in a measurable increase in the reductive activation of AOX in comparison with wild type. When examined in planta in response to citrate feeding, a strong conversion of AOX from its oxidized to its reduced form was observed in the transgenic line. These data support the hypothesis that mtICDH may be a regulatory switch involved in tricarboxylic acid cycle flux and the reductive modulation of AOX. (Received June 1, 2004; Accepted July 18, 2004) Introduction The mitochondrial tricarboxylic acid (TCA) cycle, also referred to as the citric acid or Krebs cycle, plays both a bioenergetic and a biosynthetic role, supplying reducing equivalents [NAD(P)H] for electron transport and carbon skeletons, respectively. However, little is known as to how this cycle is regulated in plants in vivo or how it is linked to respiratory electron transport (Wiskich and Dry 1985, Douce and Neuburger 1989, Oliver and McIntosh 1995). In plants, isocitrate dehydrogenases (ICDHs) comprise a multi-isoenzymic family whose members catalyze the oxidative decarboxylation of the six-carbon organic acid isocitrate, to the five-carbon organic acid 2-oxoglutarate and CO2, with NAD+ and/or NADP+ acting as electron acceptor. The NAD+-specific enzyme (NAD-IDH; EC 1.1.1.41) is found only in the mitochondrion and has been typically associated with TCA cycle flux (Oliver and McIntosh 1995). In contrast, several NADP+-specific isoforms (EC 1.1.1.42) exist, located in the cytosol (ICDH1), chloroplasts (ICDH2), mitochondria (mtICDH) and peroxisomes (Cooper and Beavers 1969, Chen and Gadal 1990, Gálvez and Gadal 1995, Hodges et al. 2003). mtICDH has not yet been purified from a plant source and its functional role remains obscure despite studies in maize (Zea mays), potato (Solanum tuberosum L.), sunflower (Helianthus annuus L.) and Norway spruce (Picea abies L. Karst.) (Curry and Ting 1976, Rasmusson and Møller 1990, Attucci et al. 1994, Cornu et al. 1996). However, a possible role in the production of NADPH for redox-regulated cell metabolism has recently been suggested (Møller 2001, Hodges et al. 2003). In plants, many algae and fungi, and certain protozoa, two pathways of mitochondrial electron transport exist from reduced ubiquinone (ubiquinol) to oxygen, producing water as a reduced product. Electron transfer can occur through the cytochrome pathway, which is coupled with the generation of an electrochemical proton gradient and subsequent ATP synthesis (Douce and Neuburger 1989, Siedow and Umbach 1995). However, electron flow can occur through an alternative pathway, distinguished by its resistance to cyanide, an inhibitor of cytochrome pathway electron flow (Vanlerberghe and McIntosh 1997). The capacities of these two mitochondrial electron pathways are regulated such that high respiratory rates may be maintained, even when the cytochrome pathway is inhibited. The terminal oxidase of this pathway, alternative oxidase (AOX), is encoded by the nuclear gene AOX1 and is part of a small gene family in Arabidopsis thaliana, soybean (Glycine max L.) and tobacco (Nicotiana tabacum L.) (Vanlerberghe and McIntosh 1997). AOX is biochemically regulated, existing as a reduced, non-covalently linked dimer or an oxidized, covalently cross-linked homodimer spanning the inner mitochondrial membrane (Umbach and Siedow 1993). The reduction of the intermolecular disulfide bonds at one of two highly conserved Cys residues (Cys126) in the oxidized, less active dimer, results in a more active form of the enzyme, leading to increased AOX activity (Umbach and Siedow 1993, Vanlerberghe et al. 1998, Vanlerberghe et al. 1999). This active form can be stimulated by certain α-keto acids, particularly pyruvate, and thus, may respond to the redox state of the mitochondrial matrix (Vanlerberghe et al. 1995, Vanlerberghe and McIntosh 1996, Vanlerberghe and McIntosh 1997). Matrix NADPH has been implicated in metabolic regulation involving the disulfide reduction of cysteinyl thiol groups via glutathione/glutathione reductase and/or thioredoxin/thioredoxin reductase systems (Levings and Siedow 1995, Vanlerberghe et al. 1995, Mackenzie and McIntosh 1999, Møller 2001). To elucidate the currently undefined role of mtICDH in the mediation of Cys reduction and subsequent covalent modification of AOX we employed three experimental approaches. First, the contributions of the ICDH isoenzymes in stably transformed suspension cells containing reduced levels of AOX were examined. Secondly, the enzymic activity of mtICDH was determined in a developmental system where the regulation of alternative pathway respiration has been extensively characterized. Lastly, tobacco plants were transformed to overexpress a cDNA encoding mtICDH. This report presents our findings in two well-characterized systems for the examination of AOX and the analyses of our transgenic plants. The physiological role of this NADPH-generating enzyme in the reductive activation of AOX and mitochondrial redox status are discussed. Results Analysis of AOX transgenic cell lines To investigate the potential role of mtICDH in the biochemical regulation of AOX, we utilized gradient-purified mitochondria from transgenic cell suspensions of tobacco containing AOX1 in the antisense orientation (AS8). This transgenic cell line with decreased AOX protein demonstrates a decreased capacity for cyanide-resistant respiration in comparison with wild-type (WT) cells, and thus a differential capacity for electron transport through the alternative pathway (Vanlerberghe et al. 1994). In addition, this cell line has been characterized extensively in studies examining various aspects of alternative pathway respiration (Vanlerberghe et al. 1994, Vanlerberghe and McIntosh 1997, Maxwell et al. 1999). Interestingly, the antisense line exhibited a 59% and 41% decrease in mtICDH- and NAD-IDH-specific activity respectively, in comparison with WT activity (Fig. 1A). In contrast, the specific activity of the cytosolic isoenzyme, ICDH1, presented a 2.8-fold increase in the AS8 line in comparison with that of the WT (Fig. 1A). When soluble matrix extracts from the gradient-purified mitochondria were examined by immunoblotting with specific antibodies generated against mtICDH, NAD-IDHa and ICDH1, the abundance of both proteins appeared to correlate well with their specific enzymic activity (Fig. 1B). In cell lines genetically altered in AOX activity, correlations are also observed in mtICDH, NAD-IDH and ICDH1 activity. The results presented in Fig. 1C demonstrate minimal differences in mtICDH, NAD-IDHa and ICDH1 transcript accumulation between the WT and AS8 cell lines. Thus, changes in specific enzymic activity and corresponding protein abundance (Fig. 1A, B) are likely to reflect post-transcriptional regulation. Developmental regulation of mtICDH Developmental changes in mtICDH activity were observed for S. guttatum floral appendix tissue with the specific activity of mtICDH increasing to a peak on D-day during thermogenesis (Fig. 2). This value is 1.7- to 1.4-fold greater than that observed on D–1 or D+2, respectively. This result is consistent with previous reports describing increases in matrix enzymes during the transition from pre-thermogenesis to thermogenesis in Arum maculatum spadices (MacDougall and ap Rees 1991, Chivasa et al. 1999). In addition, previous studies with S. guttatum appendix tissue have demonstrated that both the capacity for alternative pathway respiration and rates of oxygen uptake increase during the D–1 to D-day transition (Elthon et al. 1989). Thus, in S. guttatum, mtICDH activity is increased during the same developmental time frame in which the capacity for alternative pathway respiration and TCA cycle flux have been shown to be elevated in thermogenic tissue. Partial purification of mtICDH and protein sequencing A previous study (Gálvez et al. 1998) reported the isolation of a cDNA encoding a non-cytosolic ICDH (pST5; accession no. X96728) from tobacco (N. tabacum L. cv. Xanthi). To directly confirm that this was a mitochondrial ICDH (mtICDH) we proceeded to partially purify mtICDH from cell suspensions of tobacco and obtain protein sequence. This allowed us to assess the applicability of this cDNA to facilitate the cloning of mtICDH from our tobacco (N. tabacum L. cv. Bright Yellow) cDNA expression library. The protocol we employed for matrix enzyme purification resulted in three protein fractions: a membrane fraction; a soluble, high-molecular mass complex fraction; and a soluble fraction enriched in several matrix enzymes. These three fractions obtained from washed mitochondria in tobacco are distinct in their polypeptide profiles when separated by SDS–PAGE (data not shown). The soluble fraction was then utilized for chromatographic separation. A one-step protocol adopting affinity chromatography on Blue Sepharose®, which we developed, is summarized in Table 1. The tobacco mtICDH was purified 279-fold with a yield of 78%. The specific activity of the partially purified protein was 10,027 mU mg–1 (Table 1). The eluate from the affinity column containing the partially purified mtICDH preparation was analyzed by SDS–PAGE followed by silver staining. A peptide was detected at approximately 45–46 kDa in our gel system, which was tentatively identified as mtICDH, based on the subunit mass of the ICDH1 and ICDH2 isoenzymes from tobacco (data not shown; Gálvez et al. 1994, Gálvez et al. 1996). The band tentatively corresponding to mtICDH was excised for protein sequencing to confirm unequivocally the identity of this polypeptide. The protein sequences of the tryptic fragments were utilized to search the National Center for Biotechnology Information (NCBI; Bethesda, MD, U.S.A.) protein database. Sequence comparisons revealed that we had sequenced over approximately 26% of the pST5 gene product (pST5, accession no. CAA65503), a non-cytosolic ICDH (Gálvez et al. 1998). Thus, direct protein sequence analysis from purified mitochondria and comparison of sequences has allowed us to identify our partially purified protein as the tobacco mitochondrial NADP+-dependent ICDH (mtICDH) and confirmed that the pST5 cDNA encodes mitochondrial ICDH. Taken as a whole, our fragments demonstrated a 96% identity at the amino acid level to the pST5 gene product (data not shown) and we were able to utilize this cDNA for our molecular analysis. PCR amplification of mtICDH and generation of transgenic mtICDH overexpressors A sense-RNA approach was adopted to increase tobacco mtICDH activity using standard protocols as described in Materials and Methods. Utilizing the cDNA sequence of pST5 (Gálvez et al. 1998), we amplified a 1,583-bp fragment that spans the 5′-untranslated region (UTR), the complete coding region of mtICDH and 59 bp of the 3′-UTR, directly from a cDNA expression library of tobacco previously constructed in the laboratory (Vanlerberghe and McIntosh 1994). We inserted the cDNA fragment into the binary vector pGA748 in the sense orientation (Fig. 3A) and transformed tobacco plants with the construct using Agrobacterium-mediated gene transfer. We selected transformants on the basis of kanamycin resistance and regenerated those selected. Using leaves obtained from plants maintained in axenic culture, we used immunoblot analysis to test for expression of the chimeric gene, based on the abundance of mtICDH proteins (Fig. 3B). Twenty independent transgenic plants were identified with increased mtICDH protein abundance and, of these, three lines (PMIS 1–1, 10–1, 12–1) were selected for a more detailed analysis of enzymic activity (Fig. 3C). The most strongly overexpressing transgenic line (PMIS 10–1) displayed a leaf-specific enzymic activity of 223±36 mU mg–1, corresponding to approximately a 7.4-fold increase in the specific activity determined in CTRL 4–1 (WT) tobacco leaves (30±18 mU mg–1) when measured in matrix fractions of gradient-purified mitochondria (Fig. 3C). This plant line was selected for all further analysis. Constitutive overexpression of mtICDH does not appear to affect growth or other ICDH isoenzymes The primary transformant (PMIS 10–1) was clonally propagated as required and no obvious differences in growth phenotype were detected between shoots, roots or leaves of CTRL 4–1 (WT) and PMIS 10–1 plants (Fig. 4A). Mitochondria and RNA were extracted from leaf material and analyzed. The transgenic line PMIS 10–1, which showed a substantial increase in mtICDH-specific activity (Fig. 3C), also showed a specific increase in mtICDH mRNA and protein levels as compared with the untransformed control (CTRL 4–1; Fig. 4B, C). In contrast, the mRNA and protein levels for the mitochondrial NAD+-dependent (NAD-IDHa) or cytosolic (ICDH1) isoenzymes demonstrated minimal differences between CTRL 4–1 and PMIS 10–1 as a result of the insertion of the mtICDH sense gene (Fig. 4B, C). Based on these observations, we concluded that the chimeric mtICDH sense gene specifically increased the activity of the mitochondrial NADP+-dependent isoenzyme without significantly affecting the mitochondrial NAD+-dependent (NAD-IDHa) or cytosolic (ICDH1) isoenzymes. mtICDH mRNA demonstrates a tissue-specific expression pattern Flowers, green leaves and roots of axenically grown tobacco plants were analyzed with respect to mtICDH mRNA expression. Fig. 4C demonstrates that the highest transcript levels were present in flowers, followed by roots and then leaves. However, transcript levels for both NAD-IDHa and ICDH1 were relatively equal for flowers and roots, while leaves presented the lowest levels (Fig. 4C). In contrast, steady-state levels of AOX1 mRNA were consistent in all three organs examined between the WT and transgenic plants (Fig. 4C). These same patterns were observed in both the CTRL 4–1 and PMIS 10–1 lines (Fig. 4C), indicating that the overexpression had not altered tissue-specific expression patterns. Constitutive overexpression of mtICDH in transgenic plants influences AOX abundance and reduction state An increase in mtICDH enzyme levels may serve to increase flux through this mitochondrial enzyme with an accompanying increase in NADPH concentration. If the mitochondrial redox state, in particular NADPH concentration, acts to reductively activate the alternative oxidase, then we may expect to see increases in the reduced form of AOX (Vanlerberghe et al. 1995, Umbach and Siedow 1997, Vanlerberghe et al. 1998, Vanlerberghe et al. 1999). To test this, mitochondria were isolated from CTRL 4–1 and PMIS 10–1 transgenic tobacco lines and analyzed by non-reducing immunoblotting. Results from these analyses show an increased amount of the reduced form of AOX in the mtICDH overexpressor as compared with the control line (Fig. 5A). This difference was apparent, but predictably less than one might initially expect, owing to the previously recognized problem that AOX will undergo oxidation during the isolation procedure of mitochondria (Umbach and Siedow 1997). To circumvent this problem, whole-leaf extracts were subjected to the same analyses (Fig. 5B). These results show a marked difference from mitochondrial extracts, with the control line and three mtICDH-overexpressing lines demonstrating the AOX protein in its reduced form. In addition, it appears that there is a greater amount of AOX present in the mtICDH-overexpressing lines in comparison with the control line (Fig 5B.), although this is difficult to ascertain as this experimental technique allows for a qualitative examination of the relative amount of AOX in the oxidized and reduced forms. RNA gel blot analysis indicated only minimal changes in the level of AOX1 mRNA between the CRTL 4–1 and PMIS 10–1 lines (Fig. 4C). The above results, further supporting a connection between AOX and TCA cycle activities, prompted us to more directly investigate possible mtICDH involvement with AOX modification. We incubated leaves from the mtICDH overexpressor in the presence of antimycin A, an inhibitor of cytochrome pathway respiration and strong inducer of AOX (Vanlerberghe and McIntosh 1992b). In addition, we also incubated leaves with the TCA cycle organic acid citrate, in an attempt to increase flux through NAD-IDH or mtICDH. Leaves from CTRL 4–1 and PMIS 10–1 were incubated in B-5 medium with or without the addition of 20 mM citrate for 8 h or alternatively with 25 µM antimycin A. At the beginning of both experiments AOX was present in its fully reduced form in the control and transgenic lines (Fig. 6A, B). At the end of the 8-h incubation, leaves in B-5 medium demonstrated both oxidized and reduced forms of AOX (Fig. 6A). The presence of oxidized AOX is to be expected based on the duration of the incubation. Exposure to antimycin A for the same time course resulted in AOX being maintained in its fully reduced form in the control leaves (Fig. 6A). In contrast, leaves from PMIS 10–1 presented both oxidized and reduced forms of AOX (Fig. 6A). While antimycin A can drive the conversion of AOX to its reduced and active form (Fig. 6A), this was not observed in our transgenic line with the levels we utilized, which have been used previously for leaf incubations with antimycin A (Fig. 6A; Vanlerberghe and McIntosh 1992b, Vanlerberghe and McIntosh 1994, Vanlerberghe et al. 1995, Vanlerberghe et al. 1997). When this experiment was repeated with citrate, also demonstrated to drive the conversion of AOX from its oxidized to reduced form (Vanlerberghe et al. 1995), interesting results were observed (Fig. 6B). At the end of the 8-h incubation AOX was detected in non-reducing immunoblots from whole-leaf extracts in both the oxidized and reduced forms from leaves of CTRL 4–1 and PMIS 10–1 without citrate addition (Fig. 6B), owing once again to the duration of the incubation. Similar results were observed for the control leaves with citrate (Fig. 6B). In contrast, leaves from PMIS 10–1 that were incubated in the presence of citrate were able to maintain AOX in its reduced and active form (Fig. 6B). Discussion A role for mtICDH in development A proposed function of the alternative respiratory pathway is to maintain TCA cycle flux under developmental or environmental conditions that constrict cytochrome pathway electron flow (Lambers 1982, Vanlerberghe and McIntosh 1997). An experimental connection between AOX respiration and TCA cycle carbon flow has not been documented previously. Our study with S. guttatum indicated that mtICDH activity reaches a peak with maximal alternative pathway respiratory capacity during anthesis, whereas decreased activity occurs post-anthesis when alternative pathway respiratory capacity decreases (Fig. 2; Elthon et al. 1989). Increased activity of several TCA cycle enzymes and subsequent TCA cycle flux have been observed during thermogenesis in other Arum species (MacDougall and ap Rees 1991, Chivasa et al. 1999). The matrix enzymes citrate synthase, aconitase, fumarase and NAD+-dependent malate dehydrogenase increased during this period, while NAD-IDH presented a 24% decrease during the same developmental time frame (MacDougall and ap Rees 1991). This finding supports the role of this enzyme as a flux control point in the TCA cycle (Wiskich and Dry 1985, Oliver and McIntosh 1995, Popova and de Carvalho 1998). The observed increase in mtICDH activity may compensate for the apparent decrease in NAD-IDH activity during anthesis. Thus, mtICDH activity appears to correlate with the developmental switch from cytochrome to alternative pathway respiration during anthesis in S. guttatum. Studies with A. maculatum spadices have demonstrated that the TCA cycle may play an important role during development, as carbon flux through the cycle is increased 26- to 40-fold during thermogenesis in these organs (MacDougall and ap Rees 1991). This finding may indicate that positive feedback of AOX activity affects mtICDH/NAD-IDH activity and/or gene expression as suggested previously (McIntosh et al. 1998, Mackenzie and McIntosh 1999). Physiological significance of mtICDH Plant mitochondria contain two ICDH enzymes: NAD+-dependent (NAD-IDH) and NADP+-dependent (mtICDH; Chen and Gadal 1990, Møller 2001). Based on evidence that citrate/isocitrate conversions act as a branchpoint for metabolic pathways in plant cells, various roles have been proposed for ICDH in nitrogen metabolism (Lancien et al. 1999, Hodges et al. 2003). These proposed ICDH reactions not only provide carbon skeletons for nitrogen assimilation and reducing equivalents for biosynthetic processes, but also support the glyoxylate cycle and gluconeogenesis (Hill et al. 1992, Falk et al. 1998). Investigations utilizing cucumber (Cucumis sativus L.) and Brassica napus have demonstrated that NAD-IDH is regulated by a mechanism in which carbon flux through the decarboxylative portion of the TCA cycle is restricted, further suggesting that NAD-IDH may represent a rate-limiting step and/or a flux-limitation point in plant respiration (Hill et al. 1992, Falk et al. 1998, Popova and de Carvalho 1998). Since NAD-IDH activity is low and just sufficient to account for respiratory rates (Oliver and McIntosh 1995), it is possible that mtICDH contributes a substantial portion of isocitrate oxidation, serving to catalyze a basal carbon flux through the TCA cycle, as suggested previously for potato (Rasmusson and Møller 1990). Our data would support this possibility based on the high mtICDH:NAD-IDH ratios (3- to 5-fold) observed in heterotrophic cell suspensions (Fig. 1A). The aconitase equilibrium strongly favors citrate formation, resulting in relatively low isocitrate concentrations in the mitochondrial matrix and mtICDH has been reported to exhibit 18- to 20-fold lower Km values for substrate in comparison with NAD-IDH (Rasmusson and Møller 1990, Cornu et al. 1996). Thus, given the low isocitrate concentration in the matrix, mtICDH is likely to present the highest activity in vivo (Møller and Rasmusson 1998). The transfer of metabolic redox energy in the plant cell occurs in part via the pyridine nucleotide carriers NAD+ and NADP+, which alternate between their reduced [NAD(P)H] and oxidized [NAD(P)+] forms when exchanging reducing equivalents (Siedow and Umbach 1995, Møller 2001). Traditionally, plant mitochondria were thought to utilize only NAD(H) as reducing equivalents for respiratory metabolism (catabolic reactions), whereas NADP(H) was considered to be associated mainly with biosynthetic (anabolic) reactions in the cytosol and chloroplast. However, it has been demonstrated recently that NADP+ is present in the plant mitochondrial matrix in appreciable amounts (Møller and Rasmusson 1998, Møller 2001) and can be reduced to NADPH by the activity of a number of important enzymes such as mtICDH (EC 1.1.1.42), malic enzyme (EC 1.1.1.39), malate dehydrogenase (EC 1.1.1.37), Δ1-pyrroline-5-carboxylate dehydrogenase (EC 1.5.1.12), glutamate dehydrogenase (EC 1.4.1.3) and methylenetetrahydofolate dehydrogenase (EC 1.5.1.5). The main sinks for mitochondrial NADPH are thought to be folate turnover, fatty acid biosynthesis, NADPH dehydrogenases and the metabolism of reactive oxygen species (Møller and Rasmusson 1998, Møller 2001, Hodges et al. 2003). Thus, matrix NADP(H) levels play a role in several diverse metabolic processes. Mitochondrial redox state and the reductive activation of AOX AOX exists as a reduced (at Cys126), non-covalently linked homodimer or an oxidized, covalently cross-linked homodimer spanning the inner mitochondrial membrane (Umbach and Siedow 1993, Vanlerberghe and McIntosh 1997). A positive correlation between the oxidation of specific TCA cycle substrates (citrate, isocitrate and malate) whose oxidation can produce matrix NADPH and an abundance of reduced AOX has been clearly demonstrated and thus, it was proposed that intramitochondrial reducing power (NADPH) generated by the activity of mtICDH promotes the reduction of AOX to its more active form, possibly mediated by a thioredoxin/thioredoxin reductase system (Levings and Siedow 1995, Vanlerberghe et al. 1995, Vanlerberghe and McIntosh 1997). Under conditions of restricted electron transport through the cytochrome pathway, accumulation of pyruvate and TCA cycle intermediates would result in increased NADPH, which could potentially activate the alternative oxidase. While this would decrease the ATP yield of respiration, it would concomitantly act as a ‘clutch’, whose engagement would permit increased glycolytic and TCA cycle turnover for the production of biosynthetic carbon skeletons (Lambers 1982, Vanlerberghe and McIntosh 1997, Mackenzie and McIntosh 1999). Coupling essential cellular processes, such as the TCA cycle, with energy metabolism (via AOX electron transport), may be provided for the metabolic flexibility required of plants as sessile organisms. This coupling allows effective balancing of cellular NAD(P)H/NAD(P)+ and ATP/ADP ratios, thus linking respiratory energy coupling with demands for carbon skeletons (Vanlerberghe and McIntosh 1997, McIntosh et al. 1998, Popova and de Carvalho 1998, Mackenzie and McIntosh 1999, Møller 2001). Utilizing tobacco cell suspensions altered in AOX1 expression, we have examined the putative role of mtICDH in the biochemical regulation of AOX. Changes in mtICDH- and NAD-IDH-specific activities are positively correlated with altered AOX activity in transgenic tobacco cell lines (Fig. 1A). In contrast, the specific activity of ICDH1, the cytosolic isoenzyme, is negatively correlated with AOX activity (Fig. 1A). This correlation may indicate that a feedback of AOX activity in stably transformed cell lines affects mitochondrial ICDH activities and/or gene expression (McIntosh et al. 1998, Mackenzie and McIntosh 1999). This possibility is supported by the coordinate response of both mtICDH and NAD-IDH in cell lines genetically altered in their capacity for alternative pathway respiration (Fig. 1A). Decreased enzymic activity in the antisense AOX cell line (AS8) could be explained by the fact that lowered NADPH concentration is related to reduced redox equivalents required in the absence of AOX. This is consistent with the decreased mtICDH protein abundance and enzymic activity (Fig. 1A). The observed increase in ICDH1 enzymic activity in the AS8 line (Fig. 1A) may provide reductant for NADPH-requiring enzymes involved in the protection of membranes and electron transfer components from reactive oxygen species as suggested previously (Møller 2001, Hodges et al. 2003). This possibility is particularly intriguing as this antisense cell line produces approximately 2-fold more reactive oxygen than the WT cell line when cytochrome pathway electron flow is inhibited by antimycin A (Maxwell et al. 1999). To fully elucidate the physiological role of mtICDH in the reductive activation of AOX, we adopted a molecular genetic approach by which we generated transgenic plants in tobacco that overexpressed mtICDH. In our strongest overexpressing line (PMIS 10–1), a 7.4-fold increase in mtICDH activity was observed in comparison with control plants (Fig. 3C). This occurred with no visible changes in growth phenotype or alteration of the mitochondrial isoenzyme (NAD-IDH) or the cytosolic isoenzyme (ICDH1) at the mRNA or protein levels (Fig. 4A–C). We also discovered that mtICDH displays a tissue-specific gene expression pattern, with flowers presenting the largest levels of transcript accumulation (Fig. 4C). This finding is consistent with a previous study linking the TCA cycle enzyme citrate synthase with floral development (Landschütze et al. 1995a, Landschütze et al. 1995b). In addition, analysis of total cellular concentrations of NADPH through direct measurement via reverse-phase HPLC has shown that the mtICDH-overexpressing line demonstrates a significant (13-fold) increase in cellular NADPH in comparison with WT control plants (G.R. Gray and L. McIntosh, unpublished data). It is difficult to demonstrate conclusively that the source of this increase is mitochondrial. However, the difference between these two lines is the overexpression of mtICDH and we believe that this increased cellular concentration of NADPH is due primarily to the mitochondrial matrix NADPH content. The increased NADPH concentration and relatively increased amounts of the reduced form of AOX in response to citrate feeding (Fig. 6B) are evidence that flux through mtICDH may be part of a biochemical framework for the post-translational reductive regulation of AOX, related to TCA cycle activity, as proposed previously (Vanlerberghe et al. 1995, Vanlerberghe and McIntosh 1997, Mackenzie and McIntosh 1999). Reductive activation of AOX is mediated by the TCA cycle The overexpresion of mtICDH does not significantly favor accumulation of reduced (active) AOX when examined in isolated mitochondria (Fig. 5A). Therefore, this implies that an intermediate transducer must be involved in the reductive activation of AOX, as suggested previously (Levings and Siedow 1995, Møller and Rasmusson 1998, Møller 2001). Thioredoxin and thioredoxin reductase have been identified in plant mitochondria (Konrad et al. 1996, Banze and Follmann 2000). This system has been characterized recently (Laloi et al. 2001, Balmer et al. 2004) and represents a candidate for the link between mitochondrial redox status (NADPH) and the reductive activation of AOX. Recent analysis of the proteome from the mitochondrial matrix in Arabidopsis has also identified a thioredoxin and thioredoxin reductase (L. McIntosh, G.R. Gray, J. Yu, R. Nickels and C. Wilkerson, unpublished data). In our study, the reduced form of AOX predominates in leaf extracts, in agreement with previous reports for this type of analysis of AOX in leaf tissues (Umbach and Siedow 1997, Vanlerberghe et al. 1998, Vanlerberghe et al. 1999). It has been shown previously that the reduction state of the AOX in whole-tissue extracts is variable, depending on species and/or treatment. In Poa species and Arabidopsis no oxidized form of AOX is detected, while in soybean this is age dependent (Millar et al. 1998, Millenaar et al. 2001, Millenaar et al. 2002, Simons et al. 1999). Interestingly, our transgenic line displays an increase in reduced AOX in comparison with WT (Fig. 5B) with minimal changes in AOX1 mRNA levels (Fig. 4C). The biochemical basis for this increase is currently under investigation. Further examination of the reduction state of AOX by incubating the mtICDH overexpressor with antimycin A or citrate provided interesting results (Fig. 6A, B). While AOX started in its fully reduced form, the overexpressor could not maintain AOX in this form during the 8-h time course of our study, even in the presence of antimycin A at levels known to increase AOX activity in planta (Fig. 6A; Vanlerberghe et al. 1995). In contrast, when this transgenic line was provided with substrate in the form of citrate, AOX was maintained in its reduced form for the duration of the treatment (Fig. 6B) while leaves incubated in nutrient media showed a reversion of AOX to its oxidized form, similar to what we observed in the presence of antimycin A (Fig. 6A, B). These results support the hypothesis that flux through the TCA cycle is related to AOX activation and that mtICDH fulfils this role. This suggests that an additional regulatory mechanism is required to explain TCA cycle modulation of AOX. We propose the following to explain TCA cycle ICDH enzyme switching of metabolism and the reductive activation of AOX. Under steady-state conditions, TCA cycle flux is maintained via the action of NAD-IDH, also supplying reducing equivalents for the mitochondrial electron transport chain. However, in conditions that create a metabolic imbalance between respiratory carbon metabolism and electron transport, TCA cycle flux becomes more dependent on mtICDH and the reduction of NADP+. Our developmental data with S. guttatum, and those present in the literature provide supportive evidence for this metabolic switching (Popova and de Carvalho 1998). The consequence of this switch is increased matrix NADPH, which can then be utilized, possibly via a thioredoxin/thioredoxin reductase system, to reduce AOX and provide maximal capacity of the alternative respiratory pathway. The mtICDH overexpressor, when provided with substrate in the form of citrate, increases flux though this enzyme and generates NADPH to drive the conversion of AOX, already present at elevated levels, from its oxidized to its reduced form (Fig. 6B). In addition, it enables TCA flux to continue under conditions that would otherwise result in adenylate inhibition of the TCA cycle, thereby maintaining the production of carbon skeletons at the expense of ATP formation. This provides a functional role for mtICDH in the mitochondrial matrix. Our results provide a viable experimental link and a possible model for the connection of the induction of AOX activity and TCA cycle flux, first proposed by Lambers (1982) and later suggested to reside with the TCA cycle and mtICDH (Vanlerberghe et al. 1995). It has still to be proved whether or not a thioredoxin/thioredoxin reductase component is the critical connection between NADPH and AOX. We have also found that alteration of mtICDH levels can not only influence the reduction of AOX but also dramatically change the level of NADPH in plant cells, most probably present in mitochondria. It is an interesting challenge to predict how such a dramatic change in the redox poise of a plant cell may alter metabolism and to determine whether previously unknown mitochondrial functions are made more apparent in such a reductive environment. Materials and Methods Plant material and growth conditions Cell suspensions of transgenic tobacco (N. tabacum L. cv. Petit Havana SR1) containing AOX1 in the antisense (line AS8) orientation and WT lines were grown in axenic batch culture as described previously (Vanlerberghe et al. 1994). Cells were harvested in the early log phase (3–4 d after subculture) for enzymic analysis or late log phase (5–6 d after subculture) for protein purification as determined by growth analysis (data not shown). Plants of S. guttatum Schott were maintained at 27±4°C in a glasshouse as described previously (Elthon and McIntosh 1987). The inflorescence consists of a spadix surrounded by a spathe. The day when the upper sterile region of the spadix, the appendix, heats is referred to as ‘D-day’. Other developmental stages are referred to as the number of days before or number of days after ‘D-day’ (D–1, D+2). Plants of WT control (CTRL 4–1) and transgenic (PMIS 1–1, 10–1, 12–1) tobacco (N. tabacum L. cv. Petit Havana SR1) containing mtICDH in the sense orientation under the transcriptional control of the cauliflower mosaic virus (CaMV) 35S promoter were grown at 28°C (day/night) in GA-7 vessels (Magenta, Chicago, IL, U.S.A.) on Murashige–Skoog (MS) complete medium (Gibco, Carlsbad, CA, U.S.A.) with 0.8% (w/v) Phytagar under cool white fluorescent light (150 µmol m–2 s–1) with a 16-h photoperiod in a controlled environment incubator (New Brunswick, Scientific, Edison, NJ, U.S.A.). Transgenic lines were grown in medium supplemented with 100 µg ml–1 kanamycin. Primary transformants and WT control lines were clonally propagated as required. Soluble protein extracts, mitochondrial isolation and fractionation Tobacco cell suspensions were disrupted using a commercial blender; S. guttatum thermogenic appendices were finely diced with a razor blade, followed by grinding in a chilled mortar and pestle. Green leaf tissue was also homogenized using a chilled mortar and pestle with sand. Washed mitochondria were isolated by differential centrifugation as previously described (Boutry et al. 1984, Elthon et al. 1989, Vanlerberghe and McIntosh 1992a). The resulting mitochondria were then gradient purified directly (for enzyme activity assays) as previously described (Douce et al. 1972, Boutry et al. 1984) or suspended in assay buffer (for protein purification) containing 30 mM 3-(N-morpholino)propanesulfonic acid (MOPS, pH 6.8) and 250 mM sucrose and stored frozen at –80°C until use. Washed mitochondria, already subjected to one freeze–thaw cycle (see above) or gradient-purified mitochondria, were diluted with column wash buffer containing 30 mM MOPS (pH 7.5), 2 mM dithiothreitol and 5% (v/v) glycerol to a protein concentration of approximately 1 mg ml–1. Mitochondrial matrix was obtained by sonication essentially following a protocol described previously (Hayes et al. 1991). This resulted in a membrane fraction, a fraction containing soluble complexes of high molecular mass and a soluble matrix fraction. The soluble fraction containing the matrix proteins was concentrated by ultrafiltration using an Ultrafree®-15 apparatus (Millipore, Billerica, MA, U.S.A.; nominal molecular weight limit 10,000) and used for enzyme assays of mtICDH and NAD-IDH and the partial purification of mtICDH. Soluble protein extracts for the enzymic analysis of ICDH1 were obtained by harvesting suspension cells by centrifugation, followed by homogenization in a cytosolic extraction buffer as described (Fieuw et al. 1995). The homogenate was centrifuged at 20,000×g for 5 min and the supernatant removed and subsequently centrifuged for an additional 10 min. The soluble protein fraction was removed and concentrated as described above for matrix fractions. Enzyme activity assays and protein analysis NADP+-dependent (ICDH1 and mtICDH; EC 1.1.1.42) and NAD+-dependent (NAD-IDH; EC 1.1.1.41) ICDH activities were determined essentially as described by Omran and Dennis (1971) and Cox and Davies (1967) respectively, with minor modifications as indicated below. Enzymic activities were assayed at 25°C using a diode array spectrophotometer (DU 7400; Beckman-Coulter, Fullerton, CA, U.S.A.) by monitoring the reduction of NADP+ to NADPH at A340 after non-specific changes at A400 (collected simultaneously) were subtracted. The 1.5 ml reaction mixture contained 50 mM MOPS (pH 7.8 or 7.5 for NADP+- and NAD+-dependent isoenzymes, respectively), 1 mM NAD(P)+, 5 mM MgSO4, 10 mM threo-dSlS-isocitrate and 50 µl of enzyme extract. The reaction was initiated by the addition of substrate (isocitrate). Negative controls, omitting both isocitrate and NAD(P)+, were employed to determine whether other NAD(P)+-consuming enzymes present in the extracts contributed to the measured activity. One milliunit of activity is defined as the reduction of 1 nmol of NAD(P)+ to NAD(P)H per min at 25°C. Protein concentrations were estimated by the Coomassie dye binding method of Bradford (1976) or Lowry et al. (1951) using protein assay kits supplied by Bio-Rad (Hercules, CA, U.S.A.) with bovine serum albumin (fraction V) as a standard. SDS–PAGE and immunoblotting Samples used for polypeptide analyses were solubilized on a protein basis in 2X sample buffer (Laemmli 1970) and heated in a boiling water bath for 90 s prior to electrophoresis. Samples were separated by SDS–PAGE by means of a Mini-PROTEAN II apparatus (Bio-Rad) with a 4% (w/v) stacking gel and a 10% or 15% (w/v) resolving gel with a constant current of 15 mA applied for approximately 1.5 h at room temperature in a discontinuous buffering system (Laemmli 1970). Non-reducing SDS–PAGE analysis was performed as described above with the exception that reductant (β-mercaptoethanol) was omitted from the sample buffer (Umbach and Siedow 1993) and leaf material was flash frozen in liquid nitrogen and ground directly in non-reducing sample buffer (Umbach and Siedow 1997). The polyacrylamide gels were utilized for immunoblotting (see below) or polypeptides were visualized by either Coomassie Blue staining or silver staining using the Silver Stain Plus kit (Bio-Rad) following the manufacturer’s instructions. Polypeptides were electrophoretically transferred (Mini-Trans Blot, Bio-Rad) to nitrocellulose membranes (0.45 µm pore size; Schleicher & Schuell) by applying a constant current of 295 mA for 1.5 h in a transfer buffer described by Towbin et al. (1979) with 0.0375% (w/v) SDS. After blocking with 5% (w/v) reconstituted milk in Tris-buffered saline (TBS; 50 mM Tris-HCl pH 7.6 and 500 mM NaCl) containing 0.5% (v/v) polyoxyethylene-sorbitan monolaurate (TBS-T), the membrane was incubated with a 1 : 3,000, 1 : 5,000 or 1 : 10,000 dilution of a rabbit polyclonal antibody raised against recombinant proteins of ICDH1, mtICDH or the catalytic subunit of NAD-IDH (NAD-IDHa), respectively. Immunoblot analysis for oxidized and reduced AOX was performed using a monoclonal antibody (AOA) raised against the S. guttatum AOX (Elthon et al. 1989) at a 1 : 100 dilution. In all cases, after washing with TBS-T, the polypeptide–primary antibody complexes were incubated with either goat anti-rabbit or goat anti-mouse IgG horseradish peroxidase conjugate (Sigma, St. Louis, MO, U.S.A.) as a secondary antibody at a 1 : 20,000 dilution. The complexes were visualized using the ECL chemiluminescent detection system and Hyperfilm (Amersham, Piscataway, NJ, U.S.A.). Partial purification of mtICDH and protein sequence determination To purify mtICDH from tobacco cell suspensions, a 5-ml aliquot of the concentrated matrix fraction was applied to a Blue Sepharose® high performance affinity column (Hitrap Blue®, Amersham; 1-ml bed volume) equilibrated with 3 bed volumes of column wash buffer. The column was washed with 5 bed volumes of column wash buffer before the bound proteins were eluted with 3 bed volumes of 10 mM NADP+ in column wash buffer using a batch elution procedure. All affinity chromatographic procedures were performed at room temperature with a luerlock syringe. The enzymically active eluant was concentrated by ultrafiltration (Millipore) and stored at –80°C. During the purification procedure, proteins were routinely separated by SDS–PAGE in conjunction with enzyme activity assays to determine the relative purity of mtICDH. Samples for protein sequencing were solubilized and separated by SDS–PAGE using PROTEAN II Xi apparatus as described above. Electrophoresis was performed using a 4% (w/v) stacking gel and a 10% (w/v) resolving gel (1.0 mm thickness) with a constant voltage of 200 V applied for approximately 5 h at 25°C. Polypeptides were visualized with 0.2% (w/v) Coomassie Brilliant Blue R-250 (Sigma) in methanol:acetic acid:water (50 : 7 : 43, by vol.) and destained methanol:acetic acid:water (20 : 7 : 73, by vol.). Band(s) of interest were excised with a sterile razor blade, washed with 50% (v/v) acetonitrile and stored frozen at –80°C. Proteolytic digestion and subsequent protein sequencing were performed at the Harvard Microchemistry Facility (Cambridge, MA, U.S.A.). Putative identification of the tryptic fragments was conducted using the basic local alignment search tool (BLAST; Altschul et al. 1997) for searches of the NCBI protein databases. Amplification of mtICDH by PCR and generation of hybridization probes Cloning of the cDNA encoding mitochondrial NADP+-dependent isocitrate dehydrogenase (mtICDH; accession no. X96728) was achieved by PCR using Pwo DNA polymerase (Roche, Indianapolis, IN, U.S.A.) and oligonucleotide primers designed to regions located in the 5′- and 3′-UTRs of the tobacco (N. tabacum L. cv. Xanthi) sequence (Gálvez et al. 1998). The oligonucleotide pairs consisted of a 19-mer forward (CGAAGCTTGCACGGCGACAAAGACAAT) primer and a 21-mer reverse (GGGAATTCAGGGTATGAGTAGGAAGCGAA) primer, which amplify a 1,583-bp fragment that spans the 5′-UTR, the complete coding region and 59 bp of the 3′-UTR, while introducing 5′-HindIII and 3′-EcoRI restriction-enzyme recognition sites at the respective ends of the amplified DNA (underlined). A cDNA expression library from tobacco (N. tabacum L. cv Bright Yellow) was used as a template (Vanlerberghe and McIntosh 1994). The resulting DNA fragment was cloned in plasmid pSP72 (Promega, Madison, WI, U.S.A.) and introduced into Escherichia coli DH5α cells (Gibco) by heat shock (Sambrook et al. 1989). Insert-containing plasmids were identified by EcoRI–HindIII digestion, followed by sequence analysis to verify the identity of mtICDH. The plasmid harboring mtICDH was designated p1583–5 and used for the generation of a probe for RNA gel blots and the construction of transgenic plants. Plasmid DNA isolated from p1583-5 was subjected to an NcoI digest, which released a 110-bp gene-specific probe spanning the 5′-UTR and transit peptide coding sequence of mtICDH. Specific probes were generated for NAD-IDHa (accession no. X96727) and ICDH1 (accession no. X77944) by PCR using Taq DNA polymerase (Gibco). Oligonucleotide primers were designed to the coding region of NAD-IDHa and the 3′-UTR of ICDH1 from tobacco (N. tabacum L. cv. Xanthi) (Lancien et al. 1998, Lancien et al. 1999). The oligonucleotide primer pairs consisted of a 23-mer forward (AGGGAAAACACTGAAGGAGAGTA) primer and a 21-mer reverse (GCACTCAAAAGCAAAGCAGTT) primer, and a 19-mer forward (TGTCTGGGCAGACAAGAGG) primer and a 17-mer reverse (CTCCAGGCATTAATGTT) primer. These primer pairs amplify a 497-bp fragment from the coding region (bp 548–1,044) of tobacco NAD-IDHa and a 144-bp fragment from the 3′-UTR (bp 1,334–1,477) of tobacco ICDH1, respectively. The resulting PCR products were cloned in the vector pGEM®-T Easy (Promega) and introduced into E. coli XL1-Blue MRF′ cells (Stratagene, La Jolla, CA, U.S.A.) by heat shock (Sambrook et al. 1989). Insert-containing plasmids were identified by EcoRI digestion, followed by sequence analysis. The previously characterized 1,396-bp EcoRI insert from the cDNA clone pAONT1 was used as a hybridization probe for AOX1 (Vanlerberghe and McIntosh 1994). All fragments released as a result of restriction digestion were gel purified (QIAquick; Qiagen, Valencia, CA, U.S.A.) and subsequently used as hybridization probes for RNA gel blots. RNA isolation and gel blot analysis Total RNA was isolated from fresh leaf tissue or frozen tobacco cells using a hot phenol/LiCl procedure as described by Verwoerd et al. (1989). Samples (25 µg total RNA) were denatured, separated overnight on formaldehyde-containing agarose gels and transferred to nitrocellulose according to standard procedures (Sambrook et al. 1989). Blots were hybridized at 51°C essentially as described by Ausubel et al. (1987) with cDNA probes for mtICDH, NAD-IDHa, ICDH1 or AOX1 which were radiolabelled with [α-32P]dATP (specific activity 111 TBq mmol–1; Amersham) using a random-prime labeling kit (Roche; Feinberg and Vogelstein 1983). Following hybridization, the membranes were subjected to several buffer changes of decreasing saline sodium citrate concentrations (2–0.1×) with 0.1% (w/v) SDS, depending on the probe, to remove non-specific hybridization prior to autoradiography at –80°C on Hyperfilm (Amersham) with intensifying screens. Each experiment was repeated at least twice with total RNA isolated from different cultures or plants. Plant transformation, generation and screening The tobacco mtICDH cDNA sequence was placed under the control of the CaMV 35S promoter by sub-cloning the 1,583-bp HindIII–EcoRI insert into the binary expression vector pGA748 (An et al. 1988) in the sense orientation, creating the plasmid pGA1583 (Fig. 3A). This plasmid was transferred into Agrobacterium tumefaciens strain LBA4404 (Gibco) by electroporation. Leaf segments from young leaves of axenically grown tobacco (N. tabacum L. cv Petit Havana SR1) plants were infected with A. tumefaciens harboring pGA1583 by the leaf disc method (Horsch et al. 1988). Material was maintained under controlled environment conditions as described above under ‘Plant material and growth conditions’ on MS complete medium containing 0.8% (w/v) Phytagar, 0.1 mg liter–1 benzyladenine, 0.1 mg liter–1 naphthalene acetic acid, 100 mg liter–1 kanamycin and 500 mg liter–1 carbenicillin for callus formation. The emerging shoots were transferred to GA-7 vessels containing this same medium without hormones or carbenicillin for rooting. Primary transformants were clonally propagated as required. Screening of the transgenic lines was performed by immunoblot analysis. Tissue (0.3 g) was harvested, frozen in liquid nitrogen and lyophilized overnight. Glass beads were added and the tissue disrupted for 10 s at 50,000 rpm by means of a Bead beater (BioSpec, Bartlesuille, OK, U.S.A.) which completely powdered the tissue. To this product, 750 µl of 1.5× sample buffer (Laemmli 1970) was added followed by boiling. Samples were centrifuged briefly to pellet cellular debris and glass beads and the supernatant was removed and placed in a fresh tube. An aliquot (20 µl) of the crude extract was precipitated in 1 ml of 10% (w/v) trichloroacetic acid for 1 h at 4°C. Samples were centrifuged and the subsequent pellet was suspended in 1 M NaOH containing 5% (w/v) SDS. Crude extracts (25 µg) were separated by SDS–PAGE and mtICDH was detected by means of a 1 : 2,000 dilution of a rabbit polyclonal antibody raised against recombinant mtICDH as described above. Leaf incubations with citrate and antimycin A Fresh leaves (2–4 g) from axenically grown plants were floated adaxial side up in Petri dishes containing 25 ml of B-5 nutrient medium (pH 5.7; Gamborg et al. 1968). Tri-sodium citrate (1 M stock in B-5 medium, pH 5.7) was added to a final concentration of 20 mM or 25 µM antimycin A (70 mM stock in 2-propanol; Vanlerberghe et al. 1995). Controls in the antimycin A experiment were supplemented with identical volumes of 2-propanol. The Petri dishes were sealed with porous tape and incubated under low fluorescent light (3 µmol m–2 s–1) at room temperature for 8 h. Leaves were removed, flash frozen in liquid nitrogen and ground directly in non-reducing 2× sample buffer for non-reducing immunoblot analyses of AOX (Umbach and Siedow 1997). Acknowledgments The authors wish to thank Dr. Jonathan Walton, DOE-Plant Research Laboratory and Department of Plant Biology, Michigan State University, for helpful advice on protein purification and use of the HPLC apparatus; Dr. Ann Umbach, Duke University, for advice on AOX non-reducing immunoblots; and Dr. Mark Stitt, Max-Planck-Institut für Molekulare Pflanzenphysiologie for helpful comments on citrate feeding experiments. Antibody to the NAD(P+)-dependent ICDH enzymes was generously provided by Dr. Michael Hodges, Institut de Biotechnologie des Plantes (Centre National de la Recherche Scientifique), Université de Paris-Sud, Orsay. G.R.G. was supported, in part, by a postdoctoral fellowship from the Natural Sciences and Engineering Research Council of Canada (NSERC) during the initial phases of this study. A.R.V. acknowledges support from the Ronald E. McNair Post-Baccalaureate Achievement Program at Michigan State University. This research was supported, in part, by U.S. Department of Energy grant DE-FG02-91ER20021 and National Science Foundation grant IBM 0110768 to L.M. and a NSERC research grant to G.R.G. 4 Corresponding author: E-mail, [email protected]; Fax, +1-306-966-5015. View largeDownload slide Fig. 1 Analysis of ICDH isoenzymes in cell suspensions of WT and transgenic (AS8) tobacco (N. tabacum L.). (A) Specific enzymic activities for mtICDH (WT, 10.12±1.04 mU mg–1; AS8, 4.15±1.24 mU mg–1), NAD-IDH (WT, 2.08±0.40 mU mg–1; AS8, 1.23±0.20 mU mg–1) and ICDH1 (WT, 81.70±6.76 mU mg–1 and AS8, 232.3±16.79 mU mg–1). All enzymic determinations represent mean ± SE; n = 2. (B) Steady-state protein abundance of ICDH isoenzymes determined by immunoblotting. For each sample, 10 µg of protein was loaded and equal loading was confirmed by Coomassie staining (data not shown). (C) Steady-state mRNA accumulation for mtICDH, NAD-IDHa and ICDH1 transcripts. For each sample, 25 µg of total RNA was loaded and equal loading was confirmed by ethidium staining (data not shown). The results presented in (B) and (C) are representative of four separate experiments. Matrix fractions from gradient-purified mitochondria (mtICDH, NAD-IDH) and cytosolic fractions (ICDH1) were utilized for (A) and (B). View largeDownload slide Fig. 1 Analysis of ICDH isoenzymes in cell suspensions of WT and transgenic (AS8) tobacco (N. tabacum L.). (A) Specific enzymic activities for mtICDH (WT, 10.12±1.04 mU mg–1; AS8, 4.15±1.24 mU mg–1), NAD-IDH (WT, 2.08±0.40 mU mg–1; AS8, 1.23±0.20 mU mg–1) and ICDH1 (WT, 81.70±6.76 mU mg–1 and AS8, 232.3±16.79 mU mg–1). All enzymic determinations represent mean ± SE; n = 2. (B) Steady-state protein abundance of ICDH isoenzymes determined by immunoblotting. For each sample, 10 µg of protein was loaded and equal loading was confirmed by Coomassie staining (data not shown). (C) Steady-state mRNA accumulation for mtICDH, NAD-IDHa and ICDH1 transcripts. For each sample, 25 µg of total RNA was loaded and equal loading was confirmed by ethidium staining (data not shown). The results presented in (B) and (C) are representative of four separate experiments. Matrix fractions from gradient-purified mitochondria (mtICDH, NAD-IDH) and cytosolic fractions (ICDH1) were utilized for (A) and (B). View largeDownload slide Fig. 2 Analysis of mtICDH activity during anthesis in S. guttatum. Specific enzymic activities were determined at the developmental stages D–1, 64.67±8.74 mU mg–1; D-day, 108.50±2.12 mU mg–1; and D+2, 78.00±9.90 mU mg–1. All determinations represent mean ±SD; n=2–3. Matrix fractions from gradient-purified mitochondria were utilized. View largeDownload slide Fig. 2 Analysis of mtICDH activity during anthesis in S. guttatum. Specific enzymic activities were determined at the developmental stages D–1, 64.67±8.74 mU mg–1; D-day, 108.50±2.12 mU mg–1; and D+2, 78.00±9.90 mU mg–1. All determinations represent mean ±SD; n=2–3. Matrix fractions from gradient-purified mitochondria were utilized. View largeDownload slide Fig. 3 Overexpression of mtICDH in tobacco (N. tabacum L.). (A) mtICDH cDNA was directionally inserted into the pGA748 binary expression vector under control of the constitutive CaMV 35S promoter. This vector also contains a marker for kanamycin resistance (NPT II). (B) Leaf protein extracts (25 µg) of kanamycin-resistant plants were subjected to an immunoblot screen and equal loading was confirmed by Coomassie staining (data not shown). Bold text indicates the plant lines used for further characterization. (C) Specific enzymic activity of mtICDH in three independently transformed lines (CTRL 4–1, 30.00±17.64 mU mg–1; PMIS 1–1, 86.67±24.03 mU mg–1, PMIS 10–1, 222.67±38.44 mU mg–1; PMIS 12–1, 87.04±28.51 mU mg–1). All enzymic determinations represent means ± SE; n = 2. Representative results are shown in (B) and (C) from the six WT (CTRL) lines and 20 transformed (PMIS) lines screened. Matrix fractions from gradient- purified mitochondria were utilized for (C). 35S, CaMV 35S; LB, left border; NPT II, neomycin phosphotransferase II; NOS-P, nopalin synthase gene promoter; NOS-T, 3′ terminator region of the nopalin synthase gene; RB, right border; T7, transcript 7 terminator. View largeDownload slide Fig. 3 Overexpression of mtICDH in tobacco (N. tabacum L.). (A) mtICDH cDNA was directionally inserted into the pGA748 binary expression vector under control of the constitutive CaMV 35S promoter. This vector also contains a marker for kanamycin resistance (NPT II). (B) Leaf protein extracts (25 µg) of kanamycin-resistant plants were subjected to an immunoblot screen and equal loading was confirmed by Coomassie staining (data not shown). Bold text indicates the plant lines used for further characterization. (C) Specific enzymic activity of mtICDH in three independently transformed lines (CTRL 4–1, 30.00±17.64 mU mg–1; PMIS 1–1, 86.67±24.03 mU mg–1, PMIS 10–1, 222.67±38.44 mU mg–1; PMIS 12–1, 87.04±28.51 mU mg–1). All enzymic determinations represent means ± SE; n = 2. Representative results are shown in (B) and (C) from the six WT (CTRL) lines and 20 transformed (PMIS) lines screened. Matrix fractions from gradient- purified mitochondria were utilized for (C). 35S, CaMV 35S; LB, left border; NPT II, neomycin phosphotransferase II; NOS-P, nopalin synthase gene promoter; NOS-T, 3′ terminator region of the nopalin synthase gene; RB, right border; T7, transcript 7 terminator. View largeDownload slide Fig. 4 Effects of mtICDH overexpression on growth and ICDH isoenzyme profiles in tobacco (N. tabacum L.). (A) Growth phenotype. (B) Steady-state protein abundance in leaf tissue for mtICDH, NAD-IDHa and ICDH1. For each sample, 10 µg of protein was loaded and equal loading was confirmed by Coomassie staining (data not shown). (C) Tissue specificity of steady-state mRNA accumulation for mtICDH, NAD-IDHa and ICDH1 transcripts. For each sample, 25 µg of total RNA was loaded and equal loading was confirmed by ethidium staining (data not shown). The results presented are representative of two or three separate experiments. Matrix fractions from gradient-purified mitochondria (mtICDH, NAD-IDH) and cytosolic fractions (ICDH1) were utilized for (B). View largeDownload slide Fig. 4 Effects of mtICDH overexpression on growth and ICDH isoenzyme profiles in tobacco (N. tabacum L.). (A) Growth phenotype. (B) Steady-state protein abundance in leaf tissue for mtICDH, NAD-IDHa and ICDH1. For each sample, 10 µg of protein was loaded and equal loading was confirmed by Coomassie staining (data not shown). (C) Tissue specificity of steady-state mRNA accumulation for mtICDH, NAD-IDHa and ICDH1 transcripts. For each sample, 25 µg of total RNA was loaded and equal loading was confirmed by ethidium staining (data not shown). The results presented are representative of two or three separate experiments. Matrix fractions from gradient-purified mitochondria (mtICDH, NAD-IDH) and cytosolic fractions (ICDH1) were utilized for (B). View largeDownload slide Fig. 5 Effects of mtICDH overexpression on the reductive activation of AOX in WT (CTRL) and transgenic (PMIS) tobacco (N. tabacum L.). (A) Non-reducing immunoblot analysis of AOX protein from a WT (CTRL) and an mtICDH-overexpressing line (PMIS) under steady-state growth conditions. For each sample, 10 µg of protein from gradient-purified mitochondria was loaded and equal loading was confirmed by Coomassie staining (data not shown). (B) Non-reducing immunoblot analysis of AOX protein in leaf extracts from WT (CTRL) and mtICDH-overexpressing lines (PMIS) under steady-state growth conditions. For each sample 30 µg of protein was loaded and equal loading was confirmed by Coomassie staining (data not shown). The results presented are representative of two or three separate experiments. View largeDownload slide Fig. 5 Effects of mtICDH overexpression on the reductive activation of AOX in WT (CTRL) and transgenic (PMIS) tobacco (N. tabacum L.). (A) Non-reducing immunoblot analysis of AOX protein from a WT (CTRL) and an mtICDH-overexpressing line (PMIS) under steady-state growth conditions. For each sample, 10 µg of protein from gradient-purified mitochondria was loaded and equal loading was confirmed by Coomassie staining (data not shown). (B) Non-reducing immunoblot analysis of AOX protein in leaf extracts from WT (CTRL) and mtICDH-overexpressing lines (PMIS) under steady-state growth conditions. For each sample 30 µg of protein was loaded and equal loading was confirmed by Coomassie staining (data not shown). The results presented are representative of two or three separate experiments. View largeDownload slide Fig. 6 Non-reducing immunoblot analysis of the reductive activation of AOX in leaf extracts from the WT (CRTL 4–1) and an mtICDH-overexpressing line (PMIS 10–1) of tobacco (N. tabacum L.) in response to incubation with (A) 25 µM antimycin A or (B) 20 mM citrate for 8 h. For each sample 30 µg of protein was loaded and equal loading was confirmed by Coomassie staining (data not shown). The results presented are representative of two or three separate experiments. AA, antimycin A. View largeDownload slide Fig. 6 Non-reducing immunoblot analysis of the reductive activation of AOX in leaf extracts from the WT (CRTL 4–1) and an mtICDH-overexpressing line (PMIS 10–1) of tobacco (N. tabacum L.) in response to incubation with (A) 25 µM antimycin A or (B) 20 mM citrate for 8 h. For each sample 30 µg of protein was loaded and equal loading was confirmed by Coomassie staining (data not shown). The results presented are representative of two or three separate experiments. AA, antimycin A. Table 1 Partial purification of mtICDH from cell suspension cultures of tobacco (N. tabacum L.) Purification step  Total activity (mU) a  Total protein (µg)  Specific activity (mU mg–1) b  Recovery (%)  Enrichment (-fold)  Washed mitochondria  478  13,283  36  100  –  Soluble matrix fraction  636  4,657  137  133  4  Blue Sepharose® affinity chromatography  371  37  10,027  78  279  Purification step  Total activity (mU) a  Total protein (µg)  Specific activity (mU mg–1) b  Recovery (%)  Enrichment (-fold)  Washed mitochondria  478  13,283  36  100  –  Soluble matrix fraction  636  4,657  137  133  4  Blue Sepharose® affinity chromatography  371  37  10,027  78  279  Washed mitochondria and matrix were isolated from 5 liters (3.3 kg) of cell suspension cultures. The results presented are representative of two separate experiments. a One mU of activity is defined as the production of 1 nmol of NADPH min–1. b Specific activity is expressed per mg soluble matrix protein. View Large Abbreviations AA antimycin A AOX alternative oxidase CaMV cauliflower mosaic virus D-day day of thermogenesis in the appendix region of the spadix D–1 1 d before D-day D+2 2 d after D-day ICDH isocitrate dehydrogenase ICDH1 cytosolic NADP+-dependent ICDH ICDH2 chloroplastic NADP+-dependent ICDH NAD-IDH mitochondrial NAD+-dependent ICDH NAD-IDHa catalytic subunit of NAD-IDH MOPS 3-(N-morpholino) propanesulfonic acid mtICDH mitochondrial NADP+-dependent ICDH MS Murashige–Skoog SE standard error of the mean TBS Tris-buffered saline TBS-T TBS containing polyoxyethylene-sorbitan monolaurate TCA cycle tricarboxylic acid cycle UTR untranslated region WT wild type. References Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W. and Lipman, D.J. ( 1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res.  25: 3389–3402. Google Scholar An, G., Ebert, P.R., Mitra, A. and Ha, S.B. ( 1988) Binary vectors. In Plant Molecular Biology Manual. Edited by Gelvin, S.B. and Schilperoort, R.A. pp. A3, 1–19. Kluwer Academic Publishers, Dordrecht. Google Scholar Attucci, S., Rivoal, J., Brouquisse, R., Carde, J.-P., Pradet, A. and Raymond, P. ( 1994) Characterization of a mitochondrial NADP-dependent isocitrate dehydrogenase in axes of germinating sunflower seeds. Plant Sci.  102: 49–59. Google Scholar Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A. and Struhl, K. ( 1987) Current Protocols in Molecular Biology. John Wiley & Sons, New York. Google Scholar Balmer, Y., Vensel, W.H., Tanaka, C.K., Hurkman, W.J., Gelhaye, E., Rouhier, N., Jacquot, J.-P., Manieri, W., Schürmann, P., Droux, M. and Buchanan, B.B. ( 2004) Thioredoxin links redox to the regulation of fundamental processes of plant mitochondria. Proc. Natl Acad. Sci. USA  101: 2642–2647. Google Scholar Banze, M. and Follmann, H. ( 2000) Organelle-specific NADPH thioredoxin reductase in plant mitochondria. J. Plant Physiol.  156: 126–129. Google Scholar Boutry, M., Faber, A.-M., Charbonnier, M. and Briquet, M. ( 1984) Microanalysis of plant mitochondrial protein synthesis products: Detection of variant polypeptides associated with cytoplasmic male sterility. Plant Mol. Biol.  3: 445–452. Google Scholar Bradford, M.M. ( 1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal. Biochem.  72: 248–254. Google Scholar Chen, R.D. and Gadal, P. ( 1990) Structure, functions and regulation of NAD and NADP dependent isocitrate dehydrogenases in higher plants and other organisms. Plant Physiol. Biochem.  28: 411–427. Google Scholar Chivasa, S., Berry, J.O., ap Rees, T. and Carr, J.P. ( 1999) Changes in gene expression during development and thermogenesis in Arum. Aust. J. Plant Physiol.  26: 391–399. Google Scholar Cooper, T.G. and Beavers, H. ( 1969) Mitochondria and glyoxysomes from castor bean endosperm. Enzyme constituents and catalytic capacity. J. Biol. Chem.  244: 3507–3513. Google Scholar Cornu, S., Pireaux, J.-C., Gerard, J. and Dizengremel, P. ( 1996) NAD(P)+-dependent isocitrate dehydrogenases in mitochondria purified from Picea abies seedlings. Physiol. Plant.  96: 312–318. Google Scholar Cox, G.F. and Davies, D.D. ( 1967) Nicotinamide-adenine dinucleotide-specific isocitrate dehydrogenase from pea mitochondria. Biochem. J.  105: 729–734. Google Scholar Curry, R.A. and Ting, T.P. ( 1976) Purification, properties and kinetic observation on the isoenzymes of NADP-isocitrate dehydrogenase. Arch. Biochem. Biophys.  176: 501–509. Google Scholar Douce, R., Christensen, E.L. and Bonner, W.D. Jr. ( 1972) Preparation of intact plant mitochondria. Biochim. Biophys. Acta  275: 148–160. Google Scholar Douce, R. and Neuburger, M. ( 1989) The uniqueness of plant mitochondria. Annu. Rev. Plant Physiol.  40: 371–414. Google Scholar Elthon, T.E. and McIntosh, L. ( 1987) Identification of the alternative terminal oxidase of higher plant mitochondria. Proc. Natl Acad. Sci. USA  84: 8399–8403. Google Scholar Elthon, T.E., Nickels, R.L. and McIntosh, L. ( 1989) Mitochondrial events during development of thermogenesis in Sauromatum guttatum (Schott). Planta  180: 82–89. Google Scholar Falk, K.L., Behal, R.H., Xiang, C. and Oliver, D.J. ( 1998) Metabolic bypass of the tricarboxylic acid cycle during lipid mobilization in germinating oilseeds. Regulation of NAD+-dependent isocitrate dehydrogenase versus fumarase. Plant Physiol.  117: 473–481. Google Scholar Feinberg, A.P. and Vogelstein, B. ( 1983) A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem.  132: 6–13. Google Scholar Fieuw, S., Müller-Röber, B., Gálvez, S. and Willmitzer, L. ( 1995) Cloning and expression analysis of the cytosolic NADP+-dependent isocitrate dehydrogenase from potato. Implications for nitrogen metabolism. Plant Physiol.  107: 905–913. Google Scholar Gálvez, S., Bismuth, E., Sarda, C. and Gadal, P. ( 1994) Purification and characterization of chloroplastic NADP-isocitrate dehydrogenase from mixotrophic tobacco cells. Comparison with the cytosolic isoenzyme. Plant Physiol.  105: 593–600. Google Scholar Gálvez, S. and Gadal, P. ( 1995) On the function of the NADP-dependent isocitrate dehydrogenase isoenzymes in living organisms. Plant Sci.  105: 1–14. Google Scholar Gálvez, S., Hodges, M., Decottignies, P., Bismuth, E., Lancien, M., Sangwan, R.S., Dubois, F., LeMaréchal, P., Crétin, C. and Gadal, P. ( 1996) Identification of a tobacco cDNA encoding a cytosolic NADP-isocitrate dehydrogenase. Plant Mol. Biol.  30: 307–320. Google Scholar Gálvez, S., Roche, O., Bismuth, E., Brown, S., Gadal, P. and Hodges, M. ( 1998) Mitochondrial localization of a NADP-dependent isocitrate dehydrogenase isoenzyme by using the green fluorescent protein as a marker. Proc. Natl Acad. Sci. USA  95: 7813–7818. Google Scholar Gamborg, O.L., Miller, R.A. and Ojima, K. ( 1968) Nutrient requirements of suspension cultures of soybean root cells. Exp. Cell Res.  50: 151–158. Google Scholar Hayes, M.K., Luethy, M.H. and Elthon, T.E. ( 1991) Mitochondrial malate dehydrogenase from corn. Purification of multiple forms. Plant Physiol.  97: 1381–1387. Google Scholar Hill, S.A., Grof, C.P.L., Bryce, J.H. and Leaver, C.J. ( 1992) Regulation of mitochondrial function and biogenesis in cucumber (Cucumis sativus L.) cotyledons during early seedling growth. Plant Physiol.  99: 60–66. Google Scholar Hodges, M., Flesch, V., Gálvez, S. and Bismuth, E. ( 2003) Higher plant NADP+-dependent isocitrate dehydrogenases, ammonium assimilation and NADPH production. Plant Physiol. Biochem.  41: 577–585. Google Scholar Horsch, R.B., Fry, J., Hoffmann, N., Neidermeyer, J., Rogers, S.G. and Fraley, R.T. ( 1988) Leaf disc transformation. In Plant Molecular Biology Manual. Edited by Gelvin, S.B. and Schilperoort, R.A. pp. A5, 1–9. Kluwer Academic Publishers, Dordrecht. Google Scholar Konrad, A., Banze, M. and Follmann, H. ( 1996) Mitochondria of plant leaves contain two thioredoxins. Completion of the thioredoxin profile of higher plants. J. Plant Physiol.  149: 317–321. Google Scholar Laemmli, U.K. ( 1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature  227: 680–685. Google Scholar Laloi, C., Rayapuram, N., Chartier, Y., Grienenberger, J.-M., Bonnard, G. and Meyer, Y. ( 2001) Identification and characterization of a mitochondrial thioredoxin system in plants. Proc. Natl Acad. Sci. USA  98: 14144–14149. Google Scholar Lambers, H. ( 1982) Cyanide-resistant respiration: a nonphosphorylating electron transport pathway acting as an energy overflow. Physiol. Plant.  55: 478–485. Google Scholar Lancien, M., Ferrario-Méry, S., Roux, Y., Bismuth, E., Masclaux, C., Hirel, B., Gadal, P. and Hodges, M. ( 1999) Simultaneous expression of NAD-dependent isocitrate dehydrogenase and other Krebs cycle genes after nitrate resupply to short-term nitrogen-starved tobacco. Plant Physiol.  120: 717–725. Google Scholar Lancien, M., Gadal, P. and Hodges, M. ( 1998) Molecular characterization of higher plant NAD-dependent isocitrate dehydrogenase: evidence for a heteromeric structure by the complementation of yeast mutants. Plant J.  16: 325–333. Google Scholar Landschütze, V., Müller-Röber, B. and Willmitzer, L. ( 1995) Mitochondrial citrate synthase from potato: predominant expression in mature leaves and young flower buds. Planta  196: 756–764. Google Scholar Landschütze, V., Willmitzer, L. and Müller-Röber, B. ( 1995) Inhibition of flower formation by antisense repression of mitochondrial citrate synthase in transgenic potato plants leads to a specific disintegration of the ovary tissues of flowers. EMBO J.  14: 660–666. Google Scholar Levings, C.S. III and Siedow, J.N. ( 1995) Regulation by redox poise in chloroplasts. Science  268: 695–696. Google Scholar Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. ( 1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem.  193: 265–275. Google Scholar MacDougall, A.J. and ap Rees, T. ( 1991) Control of the Krebs cycle in Arum spadix. J. Plant Physiol.  137: 683–690. Google Scholar Mackenzie, S. and McIntosh, L. ( 1999) Higher plant mitochondria. Plant Cell  11: 571–585. Google Scholar Maxwell, D.P., Wang, Y. and McIntosh, L. ( 1999) The alternative oxidase lowers mitochondrial reactive oxygen production in plant cells. Proc. Natl Acad. Sci. USA  96: 8271–8276. Google Scholar McIntosh, L., Eichler, T., Gray, G.R., Maxwell, D., Nickels, R. and Wang, Y. ( 1998) Biochemical and genetic controls exerted by plant mitochondria. Biochim. Biophys. Acta  1365: 278–284. Google Scholar Millar, A.H., Atkin, O.K., Menz, R.I., Henry, B., Farquhar, G. and Day, D.A. ( 1998) Analysis of respiratory chain regulation in roots of soybean seedlings. Plant Physiol.  117: 1083–1093. Google Scholar Millenaar, F.F., Gonzàlez-Meler, M.A., Fiorani, F., Welschen, R., Ribas-Carbo, M., Siedow, J.N., Wagner, A.M. and Lambers, H. ( 2001) Regulation of alternative oxidase activity in six wild monocotyledonous species. An in vivo study at the whole root level. Plant Physiol.  126: 376–387. Google Scholar Millenaar, F.F., Gonzalez-Meler, M.A., Siedow, J.N., Wagner, A.M. and Lambers, H. ( 2002) Role of sugars and organic acids in regulating the concentration and activity of the alternative oxidase in Poa annua roots. J. Exp. Bot.  53: 1081–1088. Google Scholar Møller, I.M. ( 2001) Plant mitochondria and oxidative stress: Electron transport, NADPH turnover, and metabolism of reactive oxygen species. Annu. Rev. Plant Physiol. Plant Mol. Biol.  52: 561–591. Google Scholar Møller, I.M. and Rasmusson, A.G. ( 1998) The role of NADP in the mitochondrial matrix. Trends Plant Sci. 3: 21–27. Google Scholar Oliver, D.J. and McIntosh, C.A. ( 1995) The biochemistry of the mitochondrial matrix. In The Molecular Biology of Plant Mitochondria. Edited by Levings, C.S. III and Vasil, I.K. pp. 237–280. Kluwer Academic Publishers, Dordrecht. Google Scholar Omran, R.G. and Dennis, D.T. ( 1971) Nicotinamide adenine dinucleotide phosphate-specific isocitrate dehydrogenase from a higher plant. Plant Physiol.  47: 43–47. Google Scholar Popova, T.N. and de Carvalho, M.A.A.P. ( 1998) Citrate and isocitrate in plant metabolism. Biochim. Biophys. Acta  1364: 307–325. Google Scholar Rasmusson, A.G. and Møller, I.M. ( 1990) NADP-utilizing enzymes in the matrix of plant mitochondria. Plant Physiol.  94: 1012–1018. Google Scholar Sambrook, J., Fritsch, E.F. and Maniatis, T. ( 1989) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, U.S.A. Google Scholar Siedow, J.N. and Umbach, A.L. ( 1995) Plant mitochondrial electron transfer and molecular biology. Plant Cell  7: 821–831. Google Scholar Simons, B.H., Millenaar, F.F., Mulder, L., Van Loon, L.C. and Lambers, H. ( 1999) Enhanced expression and activation of the alternative oxidase during infection of Arabidopsis with Pseudomonas syringae pv tomato. Plant Physiol.  120: 529–538. Google Scholar Towbin, H., Staehelin, T. and Gordon, J. ( 1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl Acad. Sci. USA  76: 4350–4354. Google Scholar Umbach, A.L. and Siedow, J.N. ( 1993) Covalent and noncovalent dimers of the cyanide-resistant alternative oxidase protein in higher plant mitochondria and their relationship to enzyme activity. Plant Physiol.  103: 845–854. Google Scholar Umbach, A.L. and Siedow, J.N. ( 1997) Changes in the redox state of the alternative oxidase regulatory sulfhydryl/disulfide system during mitochondrial isolation: implications for inferences of activity in vivo. Plant Sci.  123: 19–28. Google Scholar Vanlerberghe, G.C., Day, D.A., Wiskich, J.T., Vanlerberghe, A.E. and McIntosh, L. ( 1995) Alternative oxidase activity in tobacco leaf mitochondria. Dependence on tricarboxylic acid cycle-mediated redox regulation and pyruvate activation. Plant Physiol.  109: 353–361. Google Scholar Vanlerberghe, G.C. and McIntosh, L. ( 1992) Lower growth temperature increases alternative pathway capacity and alternative oxidase protein in tobacco. Plant Physiol.  100: 115–119. Google Scholar Vanlerberghe, G.C. and McIntosh, L. ( 1992) Coordinate regulation of cytochrome and alternative pathway respiration in tobacco. Plant Physiol.  100: 1846–1851. Google Scholar Vanlerberghe, G.C. and McIntosh, L. ( 1994) Mitochondrial electron transport regulation of nuclear gene expression. Studies with the alternative oxidase gene of tobacco. Plant Physiol.  105: 867–874. Google Scholar Vanlerberghe, G.C. and McIntosh, L. ( 1996) Signals regulating the expression of the nuclear gene encoding alternative oxidase of plant mitochondria. Plant Physiol.  111: 589–595. Google Scholar Vanlerberghe, G.C. and McIntosh, L. ( 1997) Alternative oxidase: from gene to function. Annu. Rev. Plant Physiol. Plant Mol. Biol.  48: 703–734. Google Scholar Vanlerberghe, G.C., McIntosh, L. and Yip, J.Y.H. ( 1998) Molecular localization of a redox-modulated process regulating plant mitochondrial electron transport. Plant Cell  10: 1551–1560. Google Scholar Vanlerberghe, G.C., Vanlerberghe, A.E. and McIntosh, L. ( 1994) Molecular genetic alteration of plant respiration. Silencing and overexpression of alternative oxidase in transgenic tobacco. Plant Physiol.  106: 1503–1510. Google Scholar Vanlerberghe, G.C., Vanlerberghe, A.E. and McIntosh, L. ( 1997) Molecular genetic evidence of the ability of alternative oxidase to support respiratory carbon metabolism. Plant Physiol.  113: 657–661. Google Scholar Vanlerberghe, G.C., Yip, J.Y.H. and Parsons, H.L. ( 1999) In organello and in vivo evidence of the importance of the regulatory sulfhydryl/disulfide system and pyruvate for alternative oxidase activity in tobacco. Plant Physiol.  121: 793–803. Google Scholar Verwoerd, T.C., Dekker, B.M.M. and Hoekema, A. ( 1989) A small scale procedure for the rapid isolation of plant RNAs. Nucleic Acids Res.  17: 2362. Google Scholar Wiskich, J.T. and Dry, I.B. ( 1985) The tricarboxylic acid cycle in plant mitochondria: Its operation and regulation. In Encyclopedia of Plant Physiology, New Series, vol. 18, Higher Plant Cell Respiration. Edited by Douce, R. and Day, D.A. pp. 281–313. Springer-Verlag, Berlin. Google Scholar

Miyake, Chikahiro;Shinzaki, Yuki;Miyata, Momoko;Tomizawa, Ken-ichi

doi: 10.1093/pcp/pch163pmid: 15564526

Abstract Non-photochemical quenching (NPQ) of Chl fluorescence is a mechanism for dissipating excess photon energy and is dependent on the formation of a ΔpH across the thylakoid membranes. The role of cyclic electron flow around photosystem I (PSI) (CEF-PSI) in the formation of this ΔpH was elucidated by studying the relationships between O2-evolution rate [V(O2)], quantum yield of both PSII and PSI [Φ(PSII) and Φ(PSI)], and Chl fluorescence parameters measured simultaneously in intact leaves of tobacco plants in CO2-saturated air. Although increases in light intensity raised V(O2) and the relative electron fluxes through both PSII and PSI [Φ(PSII)×PFD and Φ(PSI)×PFD] only Φ(PSI)×PFD continued to increase after V(O2) and Φ(PSII)×PFD became light saturated. These results revealed the activity of an electron transport reaction in PSI not related to photosynthetic linear electron flow (LEF), namely CEF-PSI. NPQ of Chl fluorescence drastically increased after Φ(PSII)×PFD became light saturated and the values of NPQ correlated positively with the relative activity of CEF-PSI. At low temperatures, the light-saturation point of Φ(PSII)×PFD was lower than that of Φ(PSI)×PFD and NPQ was high. On the other hand, at high temperatures, the light-dependence curves of Φ(PSII)×PFD and Φ(PSI)×PFD corresponded completely and NPQ was not induced. These results indicate that limitation of LEF induced CEF-PSI, which, in turn, helped to dissipate excess photon energy by driving NPQ of Chl fluorescence. (Received May 5, 2004; Accepted July 20, 2004) Introduction Photon energy absorbed by chloroplasts is transformed in the thylakoid membranes into the biochemical energy (NADPH and ATP) required for CO2 fixation in the Calvin cycle. These transformations are highly efficient when the rate of photosynthesis is limited by the supply of photons to the chloroplast, but become less efficient when photosynthesis is limited by the CO2 supply to ribulose-1,5-bisphosphate (RuBP) carboxylase/oxygenase (Rubisco) or by the regeneration of RuBP in the Calvin cycle. Supply of photon energy to chloroplasts in excess of that consumed by photosynthetic CO2 assimilation causes damage to photosystem (PS) II in the thylakoid membranes, a phenomenon called photoinhibition (Asada 1999, Miyake and Okamura 2003). Under photoinhibitory conditions, the rate of photosynthetic linear electron flow (LEF) is limited by the regeneration rates of both NADP+ and ADP. As a result of this limitation, the photosynthetic electron-transport system becomes filled up with electrons (Melis 1999, Niyogi 2000, Ort and Baker 2002). This accumulation of electrons suppresses the charge separation of photoexcited PSII reaction center Chl, P680, leading to an increase in the concentration of excited singlet P680, 1P680* (Hideg et al. 1994a, Hideg et al. 1994b, Hideg et al. 1998). De-excitation of 1P680* produces excited triplet P680, 3P680*, which reacts rapidly with O2 to produce reactive singlet O2 (1O2). The 1O2 oxidizes D1-protein, a PSII reaction center protein, and inactivates PSII. Furthermore, under conditions where 1P680* accumulates, less P680 is available to accept excitons from photoexcited Chl in the LHCII protein, 1Chl*, and the latter also accumulates. Similar to 1P680*, 1Chl* de-excites to its triplet state and produces 1O2 (Macpherson et al. 1993, Mishra et al. 1994, Miyao 1994). Thylakoid membranes are rich in unsaturated fatty acids, which are very susceptible to oxidative attack by 1O2. Production of 1O2 in the thylakoid membranes stimulates fatty acid peroxidation and destabilizes the membrane bilayer (Asada 1996). Therefore, photo-excitation of P680 in PSII beyond the level required for CO2 assimilation must be avoided to protect PSII from photoinhibition. To minimize PSII photoinhibition, plant chloroplasts dissipate excess photon energy as heat, a protective mechanism observed as non-photochemical quenching (NPQ) of Chl fluorescence (Demmig-Adams and Adams 1996, Niyogi et al. 1998, Niyogi 1999, Yamamoto et al. 1999). In NPQ of Chl fluorescence, violaxanthin de-epoxidase, an enzyme in the xanthophyll cycle, is activated by acidification of the thylakoid lumen and catalyzes de-epoxidation of violaxanthin to zeaxanthin. Zeaxanthin accepts exitons directly from 1Chl* in LHCII protein and dissipates the energy as heat. As a result, Chl P680 is excited less efficiently and less 1O2 accumulates at the PSII reaction center. Furthermore, zeaxanthin also reacts directly with 1O2 and dissipates its energy as heat. Zeaxanthin can also maintain membrane bilayer structures and help to stabilize oxidatively damaged thylakoid membranes. These protective abilities of zeaxanthin are demonstrated by the fact that Arabidopsis mutants deficient in NPQ of Chl fluorescence, npq-1 and npq-4, are more susceptible than the wild type to PSII photoinhibition (Pogson et al. 1998, Niyogi 2000, Pogson and Rissler 2000, Müller et al. 2001). The induction of NPQ of Chl fluorescence requires a pH gradient (ΔpH) across the thylakoid membranes. However, it is still unclear which pathway of electron flow is mainly responsible for forming this ΔpH. At present, two pathways are proposed for the induction of NPQ of Chl fluorescence. The first is an O2-dependent electron flow in the chloroplasts, known as the water-water cycle (WWC; Asada 1999). When electrons flow through the WWC, a ΔpH is formed across the thylakoid membranes, but ATP is not consumed. Therefore, turnover of the WWC would be expected to induce NPQ. However, the WWC itself produces active oxygen and one of its components, ascorbate peroxidase, is inactivated by H2O2 (Shikanai et al. 1998). Thus, under conditions where photosynthesis is suppressed and the WWC is most active, the latter actually self-destructs. That is, under the very conditions where the physiological function of NPQ is most required, the WWC cannot sustain a ΔpH across the thylakoid membranes. The second proposed pathway of electron flow is cyclic electron flow around PSI (CEF-PSI; Heber and Walker 1992, Heber et al. 1992, Heber et al. 1995, Heber 2002, Makino et al. 2002). In CEF-PSI, NADPH or ferredoxin (Fd) is photoreduced at PSI and then donates electrons to the Cytb6/f complex. There, the Q-cycle transfers 1 H+ from the stroma to the lumen for each electron donated, resulting in a ΔpH (Allen 2003). The ability of CEF-PSI to form a ΔpH across the thylakoid membranes is supported by the fact that far-red illumination of intact chloroplasts quenches 9-aminoacridine fluorescence (Miyake et al. 1995) and that photo-excitation of PSI alone is sufficient to induce NPQ of Chl fluorescence in intact leaves (Cornic et al. 2000). CEF-PSI is composed of two main pathways, one Fd dependent and the other NAD(P)H dependent (Asada et al. 1993, Mi et al. 1995). The Fd-dependent pathway of CEF-PSI is considered to be catalyzed by Fd-quinone oxidoreductase (FQR) and is inhibited by the antibiotic antimycin A (Arnon et al. 1954, Arnon 1959, Arnon et al. 1967, Arnon and Chain 1975). Recently, it was reported that the protein pgr5 may also contribute to the FQR-pathway (Munekage et al. 2002). On the other hand, the NAD(P)H-dependent pathway of CEF-PSI is considered to be catalyzed by NAD(P)H dehydrogenase (NDH). An algal mutant deficient in NDH-dependent CEF-PSI could not produce sufficient ATP for the CO2 pump (Mi et al. 1992a, Mi et al. 1992b). These results suggest that NDH-dependent formation of a ΔpH would contribute to photosynthesis. We need to clarify the relationship between CEF-PSI activity and NPQ of Chl fluorescence to determine whether the former contributes to the dissipation of excess photon energy during steady-state photosynthesis. In the present work, we evaluated the relative activity of CEF-PSI from a simultaneous analysis of the quantum yield of both PSII and PSI. From this analysis, we found that both CEF-PSI and NPQ were induced under conditions that limited LEF. That is, the activity of CEF-PSI in the thylakoid membranes was positively correlated with that of NPQ. These results show that CEF-PSI induces NPQ, which then protects PSII from photoinhibition. Results Effect of increased light intensity on NPQ of Chl fluorescence To elucidate the relationships between the rate of O2 evolution [V(O2)], the relative electron flux through PSII [Φ(PSII)×PFD] and NPQ of Chl fluorescence in tobacco leaves, we investigated the effect of light intensity on these parameters. In CO2-saturated air, where photorespiration was suppressed, increases in light intensity stimulated O2 evolution (Fig. 1A). V(O2) approached its maximum, 35 µmol O2 m–2 s–1, at a light intensity of 600 µmol photons m–2 s–1. Φ(PSII)×PFD also increased with increases in light intensity. Φ(PSII)×PFD and V(O2) had a similar dependence on light intensity. This result indicated that V(O2), which reflects the rate of photosynthetic linear electron flow (LEF), could be estimated from Φ(PSII)×PFD in tobacco leaves under non-photorespiratory conditions. Although NPQ of Chl fluorescence increased with increases in light intensity, the response was biphasic (Fig. 1B). NPQ increased drastically at light intensities over 600 µmol photons m–2 s–1, where both V(O2) and Φ(PSII)×PFD were saturated. The dramatic increase in NPQ of Chl fluorescence under high light suggested that an electron flow not related to LEF stimulated the formation of a ΔpH across the thylakoid membranes. Therefore, we hypothesized that the turnover of CEF-PSI (CEF-PSI) was responsible for the ΔpH. Relationship between Φ(PSII)×PFD and Φ(PSI)×PFD To test the above hypothesis, we studied the relationship between the relative electron flux through PSII, Φ(PSII)×PFD, and the relative electron flux through PSI, Φ(PSI)×PFD, by simultaneous measurement of the quantum yields of both PSII and PSI in tobacco leaves. Both Φ(PSII)×PFD and Φ(PSI)×PFD were plotted against light intensity (Fig. 2A). Unlike Φ(PSII)×PFD, Φ(PSI)×PFD did not saturate against light intensity and continued to increase even after Φ(PSII)×PFD became light saturated. In a further experiment, the dependence of both Φ(PSII)×PFD and Φ(PSI)×PFD on light intensity was analyzed in leaves from different tobacco plants. Φ(PSI)×PFD was then plotted against Φ(PSII)×PFD (Fig. 2B). When electron flux through PSII was low, Φ(PSI)×PFD corresponded well with Φ(PSII)×PFD. However, higher light intensities caused Φ(PSI)×PFD to exceed Φ(PSII)×PFD and this discrepancy became even more pronounced after LEF became saturated against light intensity. These results indicated that an electron flow system not related to LEF, namely CEF-PSI, was functioning in PSI and that its activity was manifested as the difference between Φ(PSI)×PFD and Φ(PSII)×PFD. Relationship between CEF-PSI and NPQ of Chl fluorescence From the above results (Fig. 1, 2), it was found that both NPQ and CEF-PSI were induced after LEF became saturated against light intensity. Next, we plotted NPQ against Φ(PSI)/Φ(PSII), to reveal the relationship between NPQ and CEF-PSI (Fig. 3A). The parameter Φ(PSI)/Φ(PSII) reflects the relative activity of CEF-PSI, and becomes larger than 1 when CEF-PSI functions. For Fig. 3A, we replotted the data for light intensities where Φ(PSI) was larger than Φ(PSII) (Fig. 2A, B). NPQ was positively correlated with, and showed a biphasic dependence on Φ(PSI)/Φ(PSII), indicating that CEF-PSI contributed to the formation of the ΔpH responsible for induction of NPQ. NPQ decreases the photoexcitation efficiency of P680 in PSII, leading to the down-regulation of electron flux from PSII to PSI (Niyogi 1999). Under these conditions, then, P700 in PSI should be relatively oxidized. In fact, NPQ showed a positive linear relationship with [P700+]/[P700]total (Fig. 3B). That is, turnover of CEF-PSI oxidized P700 in PSI through the action of NPQ. Effect of temperature on CEF-PSI From the above results, we found that CEF-PSI was induced when the rate of LEF decreased (Fig. 2B). Next, we manipulated the rate of LEF by varying leaf temperature, and studied the resulting relationships between Φ(PSII)×PFD and Φ(PSI)×PFD (Fig. 4). Under CO2-saturated conditions, photosynthesis is limited by the rate of RuBP regeneration (Makino et al. 2003). The rate of RuBP regeneration is, in turn, regulated by the activities of Calvin-cycle enzymes, and the rate of photosynthetic electron transport and/or uptake of inorganic phosphate from the cytosol into chloroplasts (Leegood and Edwards 1996). These activities are temperature dependent, and generally increase with increasing temperature. The temperature dependence of these processes is the molecular basis for the temperature dependence of photosynthesis under CO2-saturated conditions, where increases in temperature stimulate the rate of CO2 fixation (Leegood and Edwards 1996). At all temperatures in our experiment, increases in light intensity led to increases in both Φ(PSII)×PFD and Φ(PSI)×PFD (Fig. 4). At a leaf temperature of 40°C, the dependence of both electron fluxes on light intensity corresponded and no CEF-PSI activity was observed (Fig. 4A). At 25°C, the light-saturation point of Φ(PSII)×PFD decreased compared with that of Φ(PSI)×PFD, and CEF-PSI activity was observed (Fig. 4B). The lower the temperature, the less light was required to induce the activity of CEF-PSI (Fig. 4C, D). Furthermore, decreases in temperature induced NPQ (Fig. 4B–D). That is, the limitation of LEF at low temperatures induced the activity of CEF-PSI (Fig. 4E). These results also indicated that the induction of CEF-PSI would contribute to the activation of NPQ. Discussion We deduced that CEF-PSI contributed to the formation of NPQ of Chl fluorescence from the following result: in CO2-saturated air, which suppresses photorespiration, the simultaneous light saturation of V(O2) and Φ(PSII)×PFD led to induction of NPQ. These results indicated that an electron flow not related to LEF was driving the formation of a ΔpH across the thylakoid membranes. To confirm that this electron flow was CEF-PSI, we measured the quantum yield of both PSII and PSI simultaneously and found that increases in light intensity enhanced Φ(PSI)×PFD even after LEF was saturated. This result showed that CEF-PSI, manifested as the increase in Φ(PSI)/Φ(PSII), was occurring at PSI in the thylakoid membranes. Furthermore, we found that Φ(PSI)/Φ(PSII) was positively correlated with both NPQ of Chl fluorescence and [P700+]/[P700]total. These results imply that CEF-PSI produced a ΔpH across the thylakoid membranes, which drove NPQ of Chl fluorescence and led to oxidation of P700. In short, we found that CEF-PSI is stimulated when LEF is limited. This result is consistent with those of Makino et al. (2002), who found that CEF-PSI was enhanced in rbcS-antisense rice, in which LEF was limited by the decreased rate of RuBP carboxylation. Under these conditions both NADPH and ATP are produced more quickly than they are consumed, resulting in the accumulation of reduced Fd and NADPH in the stroma. Accumulation of reduced Fd and NADPH would in turn stimulate CEF-PSI. This conclusion is supported by the fact that addition of either reduced Fd or NADPH to thylakoid membranes increases the Chl fluorescence yield, demonstrating reduction of the plastoquinone pool and the Cytb6/f complex (Mano et al. 1995). NPQ of Chl fluorescence showed a positive, but non-linear relationship with Φ(PSI)/Φ(PSII). Specifically, NPQ did not begin to increase until Φ(PSI)/Φ(PSII) exceeded 1.2; thereafter, NPQ had a proportional relationship with CEF-PSI/LEF (Fig. 3A). This relationship between NPQ and Φ(PSI)/Φ(PSII) implied the existence of a threshold for full induction of NPQ. This threshold could reflect the requirement for CEF-PSI to decrease the luminal pH to a value low enough for NPQ to operate. Ascorbate, the electron donor for violaxanthin de-epoxidase, has a pKa of 4.1 and must be in the acid form to support NPQ (Bratt et al. 1995, Jahns and Heyde 1999, Cornell and Northwood 2000). Furthermore, violaxanthin de-epoxaidase localized in the lumen must bind to the thylakoid membranes to catalyze the conversion of violaxanthin to zeaxanthin. For this binding to occur, luminal pH must be less than 6 to allow protonation of the many acidic amino acids in the C-terminal region of the protein (Bratt et al. 1995, Jahns and Heyde 1999, Cornell and Northwood 2000, Eskling et al. 2001). In sum, lowering the pH of the lumen below 6 to activate NPQ would require a high rate of CEF-PSI during steady-state photosynthesis. When Φ(PSI)/Φ(PSII) was <1.2, NPQ was induced to a value close to 0.5. This small induction of NPQ is not necessarily dependent on CEF-PSI, because almost the same amount of NPQ was detected at 40°C when there was no CEF-PSI (Fig. 4). For now, the molecular mechanism of CEF-PSI-independent activation of NPQ is unknown and remains to be clarified. At 8°C, NPQ was dramatically induced at a light intensity lower than that required to induce NPQ at higher temperatures (Fig. 4D). Furthermore, the amount of NPQ was larger than would be expected from the data obtained at 25°C (Fig. 3A, 4B). For example, at a light intensity of 330 µmol photons m–2 s–1 and a temperature of 8°C, Φ(PSI)/Φ(PSII) was about 1.24 and NPQ was about 1.5; at 25°C, the same Φ(PSI)/Φ(PSII) was associated with much less NPQ activity (Fig. 3A). The enhancement of NPQ at lower temperatures would be partly due to the limited movement of inorganic phosphate (Pi) from cytosol to chloroplast, resulting in Pi deficiency (Sage et al. 1990, Leegood and Edwards 1996). Consequently, the production of ATP would be suppressed, increasing the ΔpH across the thylakoid membranes. Therefore, at lower temperatures both the induction of CEF-PSI and the limitation of Pi recovery would lead to more NPQ. Until now, the existence of CEF-PSI had been inferred from quantum yield analysis of PSI (Foyer et al. 1990, Harbinson et al. 1990, Harbinson and Foyer 1991, Foyer and Harbinson 1999, Nixon and Mullineaux 2001). However, the relationship between CEF-PSI activity and NPQ of Chl fluorescence had not been determined quantitatively. By focusing on the enhancement of NPQ after light saturation of LEF, we were able to demonstrate that CEF-PSI contributes to the induction of NPQ of Chl fluorescence. Furthermore, we were fortunate to find an assay condition where CEF-PSI could be modulated by varying leaf temperature. That is, our measurements of CEF-PSI were not an artifact, and reflected the physiology of chloroplast photosynthesis. Finally, these results give rise to the rule that CEF-PSI is induced when LEF is limited. Materials and Methods Plant growth conditions Wild-type tobacco plants (Nicotiana tabacum cv. Xanthi) were grown from seed under standard air-equilibrated conditions with 30/25°C, light (16 h)/dark (8 h) cycles, 50–60% relative humidity and 700 µmol photons m–2 s–1 photon flux density. Seeds were sown in a 0.5 dm3 pot containing commercial peat-based compost. Plants were watered daily and fertilized (Hyponex 8-12-6; Hyponex Japan, Osaka, Japan) once a week. All measurements were made using the 11th fully expanded leaf 4 weeks after sowing. Measurements of photosynthetic parameters and collection of leaves were initiated 4 h after the start of the light period. O2 exchange, Chl fluorescence and P700+ absorbance O2 exchange and Chl fluorescence were measured simultaneously. P700+ absorbance was measured sequentially after the Chl fluorescence measurement. All measurements were repeated at least three times using three different plants. Leaf disks (10 mm in diameter) were punched from tobacco leaves and placed in a cuvette (LD2/2; Hansatech Ltd, King’s Lynn, U.K.) equipped with adaptors for glass-fiber optics linked to a PAM Chl fluorometer (PAM-101; Walz, Effeltrich, Germany). The O2-evolution rate was measured in CO2-saturated air under various light intensities and temperatures according to the method of Delieu and Walker (1981). Irradiance was provided by a halogen lamp (KL-1500; Walz) through the above glass-fiber optics. Temperature was controlled by circulating temperature-controlled water through the water-jacket, and leaf temperature was monitored with a calibrated thermocouple. Chl fluorescence was measured with the PAM Chl fluorometer through the same fiber-optics. The steady-state fluorescence yield (Fs) was monitored continuously and a 1,000-ms pulse of saturating light was supplied at intervals of 60 s to determine maximum variable fluorescence (Fm′). The relative quantum yield of PSII [Φ(PSII)] at the steady state was defined as (Fm′–Fs)/Fm′, as proposed by Genty et al. (1989). NPQ of Chl fluorescence was calculated as (Fm/Fm′–1) according to Bilger and Björkman (1994). The rate of electron flux through PSII [Je(PSII)]was determined as described by Harley et al. (1992). According to Genty et al. (1989), Je(PSII) is equal to α × Φ(PSII)×PFD, where α is a constant that depends on the molar ratio of PSII to PSI in the thylakoid membranes and the efficiency of absorption of light by the leaf, and PFD is the intensity of light illuminating the leaves. The value of α was calculated from the rate of the photosynthetic carbon reduction cycle and the Chl fluorescence yield under non-photorespiratory conditions (CO2-saturated air). The net O2-evolution rate (A) can be expressed as A = Vc – 0.5 × Vo – Rd, where Rd (rate of non-photorespiratory day respiration) is the rate of O2 consumption due to processes other than the photorespiratory carbon oxidation cycle, and Vc and Vo are the rates of carboxylation and oxygenation of RuBP, respectively, by Rubisco (von Caemmerer and Farquhar 1981). Under non-photorespiratory conditions, Vc is equivalent to A + Rd and Je(PSII) is equal to 4 × Vc or 4 × (A + Rd). Therefore, α is equal to 4 × (A + Rd)/(PFD×Φ(PSII)). A value for α of 0.44±0.02 (n =4) was calculated from the results in Fig. 1 for the tobacco leaves used in the present study. The absorbance of P700+ was measured with the same PAM Chl fluorometer by exchanging the Chl fluorescence detector unit with an ED-P700DW-E emitter-detector unit (Walz; Backhausen et al. 1998, Holtgrefe et al. 2003). The amplitude of full P700 oxidation was measured in the dark for each leaf before illumination was started. In darkness, P700 is in its reduced state, and full oxidation of P700, [P700]total, was achieved by illumination with far-red light (>700 nm), which excites only PSI. The oxidation of P700, [P700+], was monitored by the change in A810–860. During illumination, the same amount of oxidizable P700 should be available, unless the PSI electron acceptors are already in their reduced state and cannot accept more electrons. During illumination, the fraction of reduced P700, [reduced P700] or PSI acceptor (A–) is determined by short saturating light pulses, which give full oxidation of P700, followed by a ‘dark pulse’, which yields fully reduced P700. The difference between the P700 amplitude in the light and the far-red-induced amplitude determined in the dark-adapted leaf must be attributed to A–. The relative quantum yield of PSI (Φ(PSI)) was calculated as described by Klughammer and Schreiber (1994), Φ(PSI) = [reduced P700]/[P700]total. Acknowledgments This work was partly supported by New Energy and Industrial Technology Development Organization and Ministry of Economy, Trade and Industry, Japan. 1 Corresponding author: E-mail, [email protected]; Fax, +81-774-75-2320. View large Download slide View large Download slide Fig. 1 Effect of light intensity on O2-evolution rate, Φ(PSII)×PFD and NPQ of Chl fluorescence in tobacco leaf disks. (A) O2-evolution rate (closed square) and Φ(PSII)×PFD (open circle) were measured in CO2-saturated air (see Materials and Methods). Light intensity was varied from 0 to 1,800 µmol photons m–2 s–1. (B) NPQ of Chl fluorescence (open circle) was determined simultaneously with Φ(PSII)×PFD. Data are the average (n = 3) of three experiments using three leaf disks prepared from the same tobacco leaf. Vertical bars represent the standard deviation of measurement. View large Download slide View large Download slide Fig. 1 Effect of light intensity on O2-evolution rate, Φ(PSII)×PFD and NPQ of Chl fluorescence in tobacco leaf disks. (A) O2-evolution rate (closed square) and Φ(PSII)×PFD (open circle) were measured in CO2-saturated air (see Materials and Methods). Light intensity was varied from 0 to 1,800 µmol photons m–2 s–1. (B) NPQ of Chl fluorescence (open circle) was determined simultaneously with Φ(PSII)×PFD. Data are the average (n = 3) of three experiments using three leaf disks prepared from the same tobacco leaf. Vertical bars represent the standard deviation of measurement. View large Download slide View large Download slide Fig. 2 Effect of light intensity on Φ(PSII)×PFD and Φ(PSI)×PFD, and the relationship between Φ(PSII)×PFD and Φ(PSI)×PFD in tobacco leaf disks prepared from two different tobacco plants. (A) Φ(PSII)×PFD (open circle) and Φ(PSI)×PFD (closed circle) were determined simultaneously in CO2-saturated air (see Materials and Methods). Data are the average (n = 3) of three experiments using three leaf disks prepared from the same tobacco leaf. Vertical bars represent standard deviation of measurement. (B) Φ(PSII)×PFD and Φ(PSI)×PFD were determined simultaneously at various light intensities (Fig. 1) in leaf disks from different tobacco plants. Each symbol type represents the data series from a single leaf disk. Data are the average (n = 3) of three experiments using three leaf disks prepared from the same tobacco leaf. Vertical and horizontal bars represent the standard deviation of measurement. View large Download slide View large Download slide Fig. 2 Effect of light intensity on Φ(PSII)×PFD and Φ(PSI)×PFD, and the relationship between Φ(PSII)×PFD and Φ(PSI)×PFD in tobacco leaf disks prepared from two different tobacco plants. (A) Φ(PSII)×PFD (open circle) and Φ(PSI)×PFD (closed circle) were determined simultaneously in CO2-saturated air (see Materials and Methods). Data are the average (n = 3) of three experiments using three leaf disks prepared from the same tobacco leaf. Vertical bars represent standard deviation of measurement. (B) Φ(PSII)×PFD and Φ(PSI)×PFD were determined simultaneously at various light intensities (Fig. 1) in leaf disks from different tobacco plants. Each symbol type represents the data series from a single leaf disk. Data are the average (n = 3) of three experiments using three leaf disks prepared from the same tobacco leaf. Vertical and horizontal bars represent the standard deviation of measurement. View large Download slide View large Download slide Fig. 3 Relationships among NPQ of Chl fluorescence, Φ(PSI)/Φ(PSII), and [P700+]/[P700]total in tobacco leaf disks. (A) NPQ of Chl fluorescence was plotted against Φ(PSI)/Φ(PSII). Data were calculated from Fig. 1, 2. (B) [P700+]/[P700]total was determined (see Materials and Methods) simultaneously in the experiments of Fig. 1, 2, and plotted against NPQ. Data are the average (n = 3) of three experiments using three leaf disks prepared from the same tobacco leaf. Vertical and horizontal bars represent standard deviation of measurement. View large Download slide View large Download slide Fig. 3 Relationships among NPQ of Chl fluorescence, Φ(PSI)/Φ(PSII), and [P700+]/[P700]total in tobacco leaf disks. (A) NPQ of Chl fluorescence was plotted against Φ(PSI)/Φ(PSII). Data were calculated from Fig. 1, 2. (B) [P700+]/[P700]total was determined (see Materials and Methods) simultaneously in the experiments of Fig. 1, 2, and plotted against NPQ. Data are the average (n = 3) of three experiments using three leaf disks prepared from the same tobacco leaf. Vertical and horizontal bars represent standard deviation of measurement. View large Download slide View large Download slide View large Download slide View large Download slide View large Download slide Fig. 4 Effect of temperature on Φ(PSII)×PFD, Φ(PSI)×PFD, and NPQ of Chl fluorescence, and the relationship of Φ(PSI)×PFD to Φ(PSII)×PFD in leaf disks from a tobacco leaf. Φ(PSII)×PFD, Φ(PSI)×PFD and NPQ of Chl fluorescence were determined as in Fig. 1, at various leaf temperatures (A, 40°C; B, 25°C; C, 16°C; D, 8°C), except that light intensity was varied from 0 to 1,800 µmol photons m–2 s–1. Symbols are: open circle, Φ(PSII)×PFD; closed circle, Φ(PSI)×PFD; closed triangle, NPQ. (E) Φ(PSI)×PFD plotted against Φ(PSII)×PFD, using data from (A) (boxed cross, 40°C), (B) (open triangle, 25°C), (C) (open square, 16°C) and (D) (open circle, 8°C). Data are the average (n = 3) of three experiments using three leaf disks prepared from the same tobacco leaf. Vertical or horizontal bars represent the standard deviation of measurement. View large Download slide View large Download slide View large Download slide View large Download slide View large Download slide Fig. 4 Effect of temperature on Φ(PSII)×PFD, Φ(PSI)×PFD, and NPQ of Chl fluorescence, and the relationship of Φ(PSI)×PFD to Φ(PSII)×PFD in leaf disks from a tobacco leaf. Φ(PSII)×PFD, Φ(PSI)×PFD and NPQ of Chl fluorescence were determined as in Fig. 1, at various leaf temperatures (A, 40°C; B, 25°C; C, 16°C; D, 8°C), except that light intensity was varied from 0 to 1,800 µmol photons m–2 s–1. Symbols are: open circle, Φ(PSII)×PFD; closed circle, Φ(PSI)×PFD; closed triangle, NPQ. (E) Φ(PSI)×PFD plotted against Φ(PSII)×PFD, using data from (A) (boxed cross, 40°C), (B) (open triangle, 25°C), (C) (open square, 16°C) and (D) (open circle, 8°C). Data are the average (n = 3) of three experiments using three leaf disks prepared from the same tobacco leaf. Vertical or horizontal bars represent the standard deviation of measurement. References Allen, J.F. ( 2003) Cyclic, pseudocyclic and noncyclic photophosphorylation: new links in the chain. Trends Plant Sci.  8: 15–19. Google Scholar Arnon, D.I. ( 1959) Conversion of light into chemical energy in photosynthesis. Nature  184: 10–21. Google Scholar Arnon, D.I., Allen, M.B. and Whatley, F.R. ( 1954) Photosynthesis by isolated chloroplasts. Nature  174: 394–396. Google Scholar Arnon, D.I. and Chain, R.K. ( 1975) Regulation of ferredoxin-catalyzed photosynthetic phosphorylations. Proc. Natl Acad. Sci. USA  72: 4961–4965. Google Scholar Arnon, D.I., Tsujimoto, H.Y. and McSwain, B.D. ( 1967) Ferredoxin and photosynthetic phosphorylation. Nature  214: 562–566. Google Scholar Asada, K. ( 1996) Radical production and scavenging in the chloroplasts. In Photosynthesis and the Environment. Edited by Baker, N.R. pp. 128–150. Kluwer Academic Publisher, Dordrecht, The Netherlands. Google Scholar Asada, K. ( 1999) The water-water cycle in chloroplasts: scavenging of active oxygens and dissipation of excess photons. Annu. Rev. Plant Physiol. Plant Mol. Biol.  50: 601–639. Google Scholar Asada, K., Heber, U. and Schreiber, U. ( 1993) Electron flow to the intersystem chain stromal components cyclic electron flow in maize chloroplasts, as detected in intact leaves by monitoring redox change of P700 and chlorophyll fluorescence. Plant Cell Physiol.  34: 39–50. Google Scholar Backhausen, J.E., Emmerlich, A., Holtgrefe, S., Horton, P., Nast, G., Rogers, J.J.M., Müller-Röber, B. and Scheibe, R. ( 1998) Transgenic potato plants with altered expression levels of chloroplast NADP-malate dehydrogenase: interactions between photosynthetic electron transport and malate metabolism in leaves and in isolated intact chloroplasts. Planta  207: 105–114. Google Scholar Bilger, W. and Björkman, O. ( 1994) Relationship among violaxanthin deepoxidation, thylakoid membranes conformation, non-photochemical chlorophyll fluorescence quenching in leaves of cotton (Gossypium hirsutum L.). Planta  193: 238–246. Google Scholar Bratt, C.E., Arvidsson, P.-O., Carlsson, M. and Åkerlund, H.-E. ( 1995) Regulation of violaxanthin de-epoxidase activity by pH and ascorbate concentration. Photosynth. Res.  45: 16–175. Google Scholar Cornell, R.B. and Northwood, I.C. ( 2000) Regulation of CTP:phosphocholine cytidyltransferase by amphitropism and relocalization. Trends Biol. Sci.  25: 441–447. Google Scholar Cornic, G., Bukhov, N.G., Wiese, C., Bligny, R. and Heber, U. ( 2000) Flexible coupling between light-dependent electron and vectorial proton transport in illuminated leaves of C3 plants. Role of photosystem I-dependent proton pumping. Planta  210: 468–477. Google Scholar Delieu, T. and Walker, D.A. ( 1981) Polarographic measurement of photosynthetic O2 evolution by leaf discs. New Phytol.  89: 165–175. Google Scholar Demmig-Adams, B. and Adams, W.W. III ( 1996) Xanthophyll cycle and light stress in nature: uniform response to excess direct sunlight among higher plant species. Planta  198: 460–470. Google Scholar Eskling, M., Emanuelsson, A. and Åkerlund, H.-E. ( 2001) Enzymes and mechanism for violaxanthin-zeaxanthin conversion. In Regulation of Photosynthesis. Edited by Aro, E.-M. and Andersson, B. pp. 433–452. Kluwer Academic Publisher, Dordrecht, The Netherlands. Google Scholar Foyer, C., Furbank, R., Harbinson, J. and Horton, P. ( 1990) The mechanisms contributing to photosynthetic control of electron transport by carbon assimilation in leaves. Photosynth. Res.  25: 83–100. Google Scholar Foyer, C.H. and Harbinson, J. ( 1999) Relationship between antioxidant metabolism and carotenoids in the regulation of photosynthesis. In The Photochemistry of Carotenoids. Edited by Frank, H.A., Young, A.J., Britton, G. and Cogdell, R.J. pp. 305–325. Kluwer Academic Publisher, Dordrecht, The Netherlands. Google Scholar Genty, B., Briantais, J.-M. and Baker, N.R. ( 1989) The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence. Biochim. Biophys. Acta  990: 87–92. Google Scholar Harbinson, J. and Foyer, C.H. ( 1991) Relationships between the efficiencies of photosystems I and II and stromal redox state in CO2-free air. Plant Physiol.  97: 41–49. Google Scholar Harbinson, J., Genty, B. and Baker, N.R. ( 1990) The relationship between CO2 assimilation and electron transport in leaves. Photosynth. Res.  25: 213–224. Google Scholar Harley, P.C., Loreto, F., Marco, G.D. and Sharkey, T.D. ( 1992) Theoretical considerations when estimating the mesophyll conductance to CO2 flux by analysis of the response of photosynthesis to CO2. Plant Physiol.  98: 1429–1436. Google Scholar Heber, U. ( 2002) Irrungen, Wirrungen? The Mehler reaction in relation to cyclic electron transport in C3 plants. Photosynth. Res.  73: 223–231. Google Scholar Heber, U., Gerst, U., Krieger, A., Neimanis, S. and Kobayashi, Y. ( 1995) Coupled cyclic electron transport in intact chloroplasts and leaves of C3 plants: Does it exist? If so, what is its function? Photosynth. Res.  46: 269–275. Google Scholar Heber, U., Neimanis, S., Siebke, K., Schönknecht, G. and Katona, E. ( 1992) Chloroplast energization and oxidation of P700/plastocyanin in illuminated leaves at reduced level of CO2 or oxygen. Photosynth. Res.  34: 433–447. Google Scholar Heber, U. and Walker, D.A. ( 1992) Concerning a dual function of coupled cyclic electron transport in leaves. Plant Physiol.  100: 1621–1626. Google Scholar Hideg, E., Kálai, T., Hideg, K. and Vass, I. ( 1998) Photoinhibition of photosynthesis in vivo results in singlet oxygen production detection via nitroxide-induced fluorescence quenching in broad bean leaves. Biochemistry  37: 11405–11411. Google Scholar Hideg, E., Spetea, C. and Vass, I. ( 1994) Singlet oxygen production in thylakoid membranes during photoinhibition as detected by EPR spectroscopy. Photosynth. Res.  39: 191–199. Google Scholar Hideg, E., Spetea, C. and Vass, I. ( 1994) Singlet oxygen and free radical production, during acceptor- and donor-induced photoinhibition. Studies with spin trapping EPR spectroscopy. Biochim. Biophys. Acta  1186: 143–152. Google Scholar Holtgrefe, S., Bader, K.P., Horton, P., Scheibe, R., von Schaewen, A. and Backhausen, J.E. ( 2003) Decreased content of leaf ferredoxin changes electron distribution and limits photosynthesis in transgenic potato plants. Plant Physiol.  133: 1768–1778. Google Scholar Jahns, P. and Heyde, S. ( 1999) Dicyclohexylcarbodiimide alters the pH dependence of violaxanthin de-epoxidation. Planta  207: 393–400. Google Scholar Klughammer, C. and Schreiber, U. ( 1994) An improved method, using saturating light pulses, for the determination of photosystem I quantum yield via P700+-absorbance change at 830 nm. Planta  192: 261–268. Google Scholar Leegood, R.C. and Edwards, G.E. ( 1996) Carbon metabolism and photorespiration: temperature dependence in relation to other environmental factors. In Photosynthesis and the Environment. Edited by Baker, N.R. pp. 191–221. Kluwer Academic Publisher, Dordrecht, The Netherlands. Google Scholar Macpherson, A.N., Telfer, A., Barber, J. and Truscott, T.G. ( 1993) Direct detection of singlet oxygen from isolated photosystem II reaction centers. Biochim. Biophys. Acta  1143: 301–309. Google Scholar Makino, A., Miyake, C. and Yokota, A. ( 2002) Physiological functions of the water-water cycle (Mehler reaction) and the cyclic electron flow around PSI in rice leaves. Plant Cell Physiol.  43: 1017–1026. Google Scholar Makino, A., Sakuma, H., Sudo, E. and Mae, T. ( 2003) Differences between maize and rice in N-use efficiency for photosynthesis and protein allocation. Plant Cell Physiol.  44: 952–956. Google Scholar Mano, J., Miyake, C., Schreiber, U. and Asada, K. ( 1995) Photoactivation of the electron flow from NADPH to plastoquinone in spinach chloroplasts. Plant Cell Physiol.  36: 1589–1598. Google Scholar Melis, A. ( 1999) Photosystem-II damage and repair cycle in chloroplasts: what modulates the rate of photodamage in vivo? Trends Plant Sci.  4: 130–135. Google Scholar Mishra, N.P., Francke, C., vam Grokom, H.J. and Ghanotakis, D.F. ( 1994) Destructive role of singlet oxygen during aerobic illumination of the photosystem II core complex. Biochim. Biophys. Acta  1186: 81–90. Google Scholar Mi, H., Endo, T., Ogawa, T. and Asada, K. ( 1995) Thylakoid membrane-bound NADPH-specific pyridine nucleotide dehydrogenase complex mediates cyclic electron transport in the cyanobacterium Synechocystis sp. PCC 6803. Plant Cell Physiol.  36: 661–668. Google Scholar Mi, H., Endo, T., Schreiber, U. and Asada, K. ( 1992) Donation of electrons from cytosolic components to the intersystem chain in the cyanobacterium Synechococcus sp. PCC 7002, as determined by reduction of P700+. Plant Cell Physiol.  33: 1099–1105. Google Scholar Mi, H., Endo, T., Schreiber, U., Ogawa, T. and Asada, K. ( 1992) Electron donation from cytosolic and respiratory flows to the photosynthetic intersystem chain is mediated by pyridine nucleotide dehydrogenase in the cyanobacterium Synechocystis sp. PCC 6803. Plant Cell Physiol.  33: 1233–1237. Google Scholar Miyake, C. and Okamura, M. ( 2003) Cyclic electron flow within PSII protects PSII from its photoinhibition in thylakoid membranes from spinach chloroplasts. Plant Cell Physiol.  44: 457–462. Google Scholar Miyake, C., Schreiber, U. and Asada, K. ( 1995) Ferredoxin-dependent and antimycin A-sensitive reduction of cytochrome b-559 by far-red light in maize thylakoids: Participation of a menadiol-reducible cytochrome b-559 in cyclic electron flow. Plant Cell Physiol.  36: 743–748. Google Scholar Miyao, M. ( 1994) Involvement of active oxygen species in degradation of the D1 protein under strong illumination in isolated subcomplexes of photosystem II. Biochemistry  33: 9722–9730. Google Scholar Müller, P., Li, X.-P. and Niyogi, K.K. ( 2001) Non-photochemical quenching. A response to excess light energy. Plant Physiol.  125: 1558–1566. Google Scholar Munekage, Y., Hojo, M., Meurer, J., Endo, T., Tasaka, M. and Shikanai, T. ( 2002) PGR5 is involved in cyclic electron flow around photosystem I and is essential for photoprotection in Arabidopsis. Cell  110: 361–371. Google Scholar Niyogi, K.K. ( 1999) Photoprotection revisited: genetic and molecular approaches. Annu. Rev. Plant Physiol. Plant Mol. Biol.  50: 333–359. Google Scholar Niyogi, K.K. ( 2000) Safety valves for photosynthesis. Curr. Opin. Plant Biol.  3: 455–460. Google Scholar Niyogi, K.K., Grossman, A.R. and Björkman, O. ( 1998) Arabidopsis mutants define a central role for the xanthophylls cycle in the regulation of photosynthetic energy conversion. Plant Cell  10: 1121–1134. Google Scholar Nixon, P.J. and Mullineaux, C.W. ( 2001) Regulation of photosynthetic electron transport. In Regulation of Photosynthesis. Edited by Aro, E.-M. and Andersson, B. pp. 533–555. Kluwer Academic Publisher, Dordrecht, The Netherlands. Google Scholar Ort, D.R. and Baker, N.R. ( 2002) A photoprotective role for O2 as an electron sink in photosynthesis? Curr. Opin. Plant Biol.  5: 193–198. Google Scholar Pogson, B.J., Niyogi, K.K., Björkman, O. and DellaPenna, D. ( 1998) Altered xanthophylls compositions adversely affect chlorophyll accumulation and non-photochemical quenching in Arabidopsis mutants. Proc. Natl Acad. Sci. USA  95: 13324–13329. Google Scholar Pogson, B.J. and Rissler, H.M. ( 2000) Genetic manipulation of carotenoid biosynthesis and photoprotection. Phil. Trans. R. Soc. Lond. B  355: 1395–1403. Google Scholar Sage, R.F., Sharkey, T.D. and Pearcy, R.W. ( 1990) The effect of leaf nitrogen and temperature on the CO2 response of photosynthesis in the C3 dicot Chenopodium album L. Aust. J. Plant Physiol.  17: 135–148. Google Scholar Shikanai, T., Takeda, T., Yamauchi, H., Sano, S., Tomizawa, K., Yokota, A. and Shigeoka, S. ( 1998) Inhibition of ascorbate peroxidase under oxidative stress in tobacco having bacterial catalase in chloroplasts. FEBS Lett.  428: 47–51. Google Scholar von Caemmerer, S. and Farquhar, G.D. ( 1981) Some relationships between the biochemistry of photosynthesis and the gas exchange of leaves. Planta  153: 376–387. Google Scholar Yamamoto, H.Y., Bugos, R.C. and Hieber, A.D. ( 1999) Biochemistry and molecular biology of the xanthophylls cycle. In The Photochemistry of Carotenoids. Edited by Frank, H.A., Young, A.J., Britton, G. and Cogdell, R.J. pp. 293–303. Kluwer Academic Publisher, Dordrecht, The Netherlands. Google Scholar

Obata, Toshihiro;Araie, Hiroya;Shiraiwa, Yoshihiro

doi: 10.1093/pcp/pch164pmid: 15564527

Abstract We investigated the uptake and bioconcentration of the essential element selenium by a coccolithophorid, Emiliania huxleyi, using [75Se]selenite. The time course of 75Se uptake showed a biphasic pattern, namely a primary phase and a subsequent secondary phase. The primary and secondary phases are due to a rapid selenite uptake process that attained a stationary level within 2 min and a slow Se-accumulation process that continued at a constant rate for 4 h or longer, respectively. Kinetic analysis revealed that the selenite uptake process consists of two components, one saturable and one linearly related to substrate concentration. The Km of the saturable component was 29.8 nM selenite; the uptake activity of this component was suppressed by inhibitors of ATP biogenesis, suggesting that selenite uptake is driven by a high-affinity, active transport system. During a 6-h incubation of cells with [75Se]selenite, 70% of the intracellular 75Se was incorporated into low-molecular-mass compounds (LMCs), and 17% was incorporated into proteins, but [75Se]selenite was barely detectable. A pulse–chase experiment demonstrated that the 75Se that had accumulated in LMCs was transferred into proteins. When the syntheses of amino acids and proteins were each separately inhibited, 75Se incorporation into LMCs and proteins was decreased. These results suggest that E. huxleyi rapidly absorbs selenite, filling a small intracellular pool. Then, Se-containing LMCs are immediately synthesized from the selenite, creating a pool of LMCs that are then metabolized to selenoproteins. (Received May 26, 2004; Accepted July 20, 2004) Introduction Coccolithophorids are unicellular, marine, calcifying algae that are well known as important components of marine ecosystems and contributors to marine primary production. Emiliania huxleyi is a dominant coccolithophorid species; its bloom generates a continuous rain of calcium carbonate to the deep ocean, and the cell debris is partly preserved as sediment (Riebesell et al. 2000, Shiraiwa 2003). Changes in algal growth rates and bloom frequencies in the ocean affect oceanic productivity and the role of algae in climate change and global biogeochemical processes. Therefore, an analysis of the factors that regulate the growth status of algae is fundamental to understanding how the global carbon cycle can be controlled. We previously reported that coccolithophorids have an obligatory growth requirement for a nanomolar level of selenium (Se), although this element has been shown to be toxic for the algae at micromolar concentrations (Danbara and Shiraiwa 1999). Some phytoplankton species have an absolute requirement for Se, while in other species, addition of Se to the culture medium stimulates growth (Doblin et al. 1999). Therefore, Se is a potentially important growth-regulating factor for coccolithophorids and other microalgae in oceanic environments, as described by Doblin et al. (2000). Furthermore, many species of algae concentrate Se in their cells (Baines and Fisher 2001) despite the extremely low concentration of Se in natural seawater (1.7 nM) (Libes 1992). We previously found that E. huxleyi accumulates selenite 1,500-fold in its cells when 10 nM selenite is present in the medium and that its ability to concentrate Se is greater than that of another non-calcifying haptophyte, Isochrysis galbana, or the unicellular green alga Dunaliella tertiolecta (Obata et al. 2004). However, the biochemical pathways and physiological significance of Se accumulation remain unknown. Selenium shares many chemical properties with sulfur, and in general Se is metabolized via the sulfur metabolic pathway (Läuchli 1993). However, the incorporation of Se into sulfur compounds induces a decrease in enzyme activity and the generation of superoxides (Läuchli 1993, Spallholz 1994), making excess Se highly toxic. Nevertheless, Se-specific metabolic pathways are present in some organisms and function to synthesize and metabolize some Se-containing compounds. For example, selenocysteine insertion mechanisms that specifically insert selenocysteine into the reaction center of selenoproteins have been identified in various organisms such as bacteria, archea and mammals, but not yet in yeast or plants (Fu et al. 2002). Selenium is an essential cofactor of selenoenzymes, which catalyze many important redox reactions. In some Se-accumulating plants, such as several species of Astragalus, selenoproteins are not synthesized, but non-protein seleno-amino acids, e.g. Se-methyl selenocysteine and Se-cystathionine, and a seleno-dipeptide, γ-glutamyl-Se-methylselenocysteine, are produced to detoxify accumulated Se in the plants (Neuhierl and Böck 1996, Terry et al. 2000). E. huxleyi exhibits both a growth requirement for Se and a high accumulation activity for Se, suggesting that this alga may have distinctive systems for the metabolism and detoxification of Se. For the present paper, we analyzed the kinetic properties of the uptake of selenite and the subsequent metabolism of accumulated Se in E. huxleyi using [75Se]selenite as a radiotracer. We revealed that E. huxleyi possesses effective mechanisms for the incorporation of Se by the active uptake of selenite and for the metabolism of Se into Se-containing low-molecular-mass compounds (LMCs) and selenoproteins, which may function to maintain the viability of the algal cells. Results Kinetics of selenite uptake We investigated the time course of [75Se]selenite incorporation into E. huxleyi cells in standard conditions (see Materials and Methods). 75Se incorporation into the cells increased rapidly within 30 s of the addition of [75Se]selenite. The [75Se]selenite uptake became slower after 30 s and increased linearly with time after 2 min (Fig. 1). Overall, a biphasic pattern was seen with a primary rapid-uptake phase that ceased within only 1 min (Fig. 1 dashed line) and a secondary slow-uptake phase that continued linearly at a constant rate for 4 h or longer (Fig. 1 dotted line). The rate of the primary selenite uptake during the first 30 s was determined at different concentrations of [75Se]selenite (Fig. 2). The relationship between the rate and the substrate concentration exhibited a biphasic pattern, suggesting the existence of more than one component for [75Se]selenite uptake. At higher selenite concentration, a linear relationship was shown between the uptake rate and selenite concentration (‘linear component’) (Fig. 2 A, dashed line). The curve obtained by the subtraction of values of the linear component from the total uptake showed Michaelis–Menten-type kinetics (‘saturable component’) (Fig 2, dotted line). From the double reciprocal plot analysis, a Km of 29.8 nM external selenite was calculated for the saturable component (Fig. 2B). Effects of light and metabolic inhibitors on the activity of primary selenite uptake To examine the energy requirements for the primary selenite uptake, we assessed the requirement for a photo-energy supply, and we evaluated the effects of several metabolic and energetic inhibitors on selenite uptake during 30 s in E. huxleyi cells. The selenite uptake occurred in both the light and the dark, although the uptake rate in the light was about 20% higher than that in the dark (Table 1). In accordance with the study of Sekino and Shiraiwa (1996), 1 mM KCN, 170 µM carbonyl cyanide 3-chlorophenylhydrazone (CCCP) and 1 mM salicylhydroxamic acid (SHAM) were used in combination to maximally inhibit both the cyanide-sensitive and -insensitive respirations. When the rates of 75Se uptake were determined at both 160 nM and 2.63 µM selenite, the inhibition was more obvious at the lower concentration than at the higher one (Table 1); that is, the saturable component was strongly inhibited but the linear component was not affected (data not shown). CCCP and diethylstilbestrol (DES), inhibitors of ATP biosynthesis, also inhibited Se uptake at 160 nM but not at 2.63 µM selenite (Table 1). These results suggest that the saturable component of selenite uptake is dependent on energy generation by ATP, but the linear component is not. The uptake of selenite at 160 nM was inhibited by 4,4′-diisothiocyanatostilbene-2,2′-disulfonate (DIDS), an inhibitor of anion transport by protein modification; in contrast, DIDS accelerated uptake at 2.63 µM. Analysis of 75Se-labeled compounds during continuous and pulse–chase labeling experiments In continuous labeling experiments, 75Se incorporation into fractions of LMCs, proteins, lipids and polysaccharides and nucleic acids (PS & NA) proceeded linearly with time for 6 h (Fig. 3A). The percent of total 75Se uptake into each fraction was almost constant. The percentages of the 75Se found in the LMC and protein fractions were 70% and 17%, respectively. The percent-incorporations into the lipid and PS & NA fractions were 3.8% and 6.1%, respectively. The pulse–chase experiment was performed by labeling cells with 75Se for 1 h and then removing the [75Se]selenite from the reaction medium to chase the transfer of 75Se among metabolites. During the chase period, the amount of 75Se incorporated into the LMCs rapidly decreased, and conversely, the amount of 75Se-labeled protein increased, creating a mirror image. The amounts of 75Se incorporated into the lipid and PS & NA fractions were maintained nearly constant during the chase period (Fig. 3B). Thin-layer chromatography (TLC) analysis revealed that the LMC fraction was composed of five 75Se-labeled compounds (LMC1-5), but 75SeO32– was barely detectable (Fig. 4). Neither selenomethionine (Se-Met) nor selenocystine had accumulated in the cells (Fig. 4). Effect of metabolic inhibitors on the incorporation of 75Se into metabolites The incorporation of 75Se into the LMC fraction was suppressed by treatment with the amino acid synthesis inhibitors aminooxyacetic acid (AOA), which inhibits aminotransferases, and l-methionine sulfoximine (l-MSO), which inhibits glutamine synthase (Fig. 5A). The total incorporation of 75Se was diminished by about 20% with these inhibitors, but total incorporation was hardly affected with cycloheximide (CHI), an inhibitor of protein synthesis on the 80S ribosome. AOA and l-MSO suppressed mainly the synthesis of [75Se]LMCs, whereas CHI primarily inhibited 75Se protein synthesis and scarcely affected 75Se incorporation into the LMC, lipid and PS & NA fractions (Fig. 5B). Discussion In seawater, Se is present in three different chemical forms: selenite [Se(IV)O32–], selenate [Se(VI)O42–] and organic selenides [e.g. Se(-II)-selenomethionine] (Cutter and Bruland 1984). Selenite and selenate, inorganic forms of Se, represent about 15% and 36% of the Se in seawater, respectively (Cutter and Cutter 2001). Inconsistent with the composition ratio, selenite was found to enhance the growth of coccolithophorids much more effectively than did selenate in our previous study; the optimum concentration of selenite was 1–10 nM in E. huxleyi (Danbara and Shiraiwa 1999). To analyze how E. huxleyi cells use selenite so efficiently, the Se uptake and accumulation processes in E. huxleyi were characterized in this study. The clearly biphasic pattern of the time course of selenite uptake was shown to be composed of a primary rapid phase and a secondary slow phase (Fig. 1). The uptake rate of the primary phase indicates the rate of selenite transport process into the cell, which is the first step in the use of selenite. The kinetic curve showing the relationship between selenite concentration and its uptake rate at primary phase consists of two components, which are the saturable component showing Michaelis–Menten-type kinetics and the non-saturable linear component (Fig. 2). Moreover, inhibitors of mitochondrial respiration and ATP biogenesis diminished the contribution of the saturable component to selenite uptake (Table 1). These results indicate the involvement of an active transport mechanism in the selenite transport process. The active transport is probably mediated by membrane transporters that are located on the cell surface, based on the inhibition by DIDS, a membrane-impermeable inhibitor of anion exchange (Table 1). The active transport exhibited high affinity for selenite but very low maximum velocity, indicating that the active transport system functions predominantly at very low selenite concentrations. Since sulfate is an analogue of selenite, an energy-dependent transporter of sulfate is generally supposed to mediate the active uptake of selenite. However, the Km of well-known sulfate transporters (7–10 µM for sulfate; Terry et al. 2000) is much higher than that of the active transport system of selenite in E. huxleyi (29.8 nM for selenite, in this study). This can be supported by the results that E. huxleyi absorbed selenite efficiently from a medium containing 26.0 mM sulfate and 10 nM selenite without significant competitive inhibition. From these evidences, a selenite-specific transporter seems to mediate selenite transport into E. huxleyi cells. Uptake at selenite concentrations above micromolar was mediated mainly by the linear component (Fig. 2). The low-affinity linear component was not affected by respiratory inhibitors but was accelerated by DIDS (Table 1). These results suggest that the linear component of selenite uptake is a result of passive transport that is regulated or mediated by a protein(s) at the cell surface. Both selenite and sulfite, an analogue of selenite, have been thought to enter cells only by simple diffusion through the plasma membranes of photosynthetic cells (Furihata et al. 1997, Terry et al. 2000). The present study clearly demonstrated E. huxleyi as the first photosynthetic organism proved to have an active selenite transport system. E. huxleyi might have acquired an active system in order to absorb Se at the very low Se concentrations present in seawater. The analysis of our results using the Michaelis–Menten equation demonstrated that 94% of the selenite taken up by cells was carried by an active transport system at the practical selenite concentration in seawater, that is, 0.1 nM (Cutter and Cutter 2001) (Fig. 2). The uptake rate of the secondary uptake phase indicates the rate of synthesis of Se metabolites and their subsequent metabolism to produce macromolecules such as proteins. The synthesis of Se-containing compounds is the critical process that enables E. huxleyi to bioconcentrate Se. The size of the selenite pool in the cells was nearly zero (Fig. 4), as the incorporated selenite was quickly converted to Se compounds, such as LMCs and proteins (Fig. 3). In fact, the selenite that was transported into E. huxleyi cells was found to be metabolized into five kinds of Se-containing, low-molecular-mass, organic compounds, namely, LMC1-5 (Fig. 4). The synthesis of LMCs in E. huxleyi might be a process homologous to the Se-metabolism pathway of land plants in which selenite is rapidly converted to selenodiglutathione (GS-Se-SG), which is reduced to the selenol GS-SeH via a non-enzymic reaction (Terry et al. 2000). The incorporation of 75Se into the LMC fraction decreased when amino acid synthesis was inhibited by AOA and l-MSO (Fig. 5A). This suggests that at least some of the Se-containing LMCs contained seleno-amino acids, other than selenocystine and Se-Met, or perhaps other compounds such as dimethyl selenoniopropionate (DMSeP) or dimethyl diselenide (DMDSe), but LMC1-5 have not yet been identified. The accumulation of Se-containing amino acids such as selenocysteine, Se-methyl selenocysteine and Se-methyl selenomethionine has been reported in Dunaliella primolecta, Chlorella sp. (Bottino et al. 1984) and in Se-accumulating plants (Terry et al. 2000). One of the Se-containing compounds that was observed in E. huxleyi might be DMSeP, which is synthesized from Se-Met; this is a reasonable expectation, because coccolithophorids are known to produce large amounts of dimethyl sulfoniopropionate (DMSP), a sulfur analogue of DMSeP, as a compatible solute for osmoregulation (Wolfe et al. 1997). The pulse–chase experiment clearly demonstrated that 75Se in the Se-containing LMCs was transferred into selenoproteins (Fig. 3B). The incorporation of selenium into proteins is thought to occur via the translation process, given that the incorporation was immediately stopped by the addition of CHI (Fig. 5B). Recently, selenoproteins were found in the unicellular green alga, Chlamydomonas reinhardtii and the dinoflagellate Oxyrrhis marina, suggesting that these algae possess translation machinery corresponding to that in animals (Fu et al. 2002, Novoselov et al. 2002, Osaka et al. 2003). Based on this evidence, it can be assumed that E. huxleyi also possesses animal-type translation machinery and that the functions of synthesized selenoproteins are essential to maintaining the viability of E. huxleyi. In conclusion, Se accumulation in the coccolithophorid E. huxleyi occurs via three processes: (i) uptake by the active transport of selenite into cells from seawater containing only nanomolar selenite, (ii) immediate fixation and accumulation of Se through the synthesis of unidentified LMCs, and (iii) synthesis of selenoproteins from the LMCs (Fig. 6). Total Se accumulation was inhibited by AOA, l-MSO and CHI, making steps (ii) and (iii) rate limiting. Several other organisms have been shown to accumulate Se. Se-accumulating plants synthesize non-protein seleno-amino acids via a plant-specific pathway (Neuhierl and Böck 1996, Terry et al. 2000). Four species of bacteria, Thauera selenatis, Sulfurospirillum barnesii, Bacillus arsenicoselenatis and Bacillus selenitireducens, are able to reduce Se oxyanions to Se(0), and the accumulation of the element is exogenous, occurring outside the cell membrane (Stolz and Oremland 1999). In these cases, however, Se is accumulated primarily for purposes of detoxification, and the accumulated Se is not used as a Se pool for the synthesis of selenoproteins in these organisms. The distinctive Se-accumulation mechanism in E. huxleyi is thought to enable efficient use of the extremely low levels of Se in seawater. By exploiting this machinery, E. huxleyi may be able to maintain active selenoenzymes and thereby maintain its continuous and rapid growth during algal blooms in the ocean. Materials and Methods Organism and culture conditions A coccolithophorid, E. huxleyi (NIES 873), was used in this study. Cells were grown in an artificial seawater, Marine Art SF (Senju Pharmaceutical Co., Osaka, Japan), enriched with Erd-Schreiber’s medium and containing 10 nM sodium selenite instead of soil extracts (MA-ESM; Danbara and Shiraiwa 1999); cells were maintained under constant illumination of 100 µmol m–2 s–1 and at 20°C (standard conditions). Cells were harvested at the linear growth phase and resuspended at a concentration of 1–3×107 cells ml–1 in MA-ESM medium without Se. After preincubation overnight under standard conditions, the cell suspension was used for the experiments below. Selenite uptake assays An aliquot of the cell suspension was preincubated under experimental conditions for 30 min in an L-shaped reaction tube, after which the reaction was initiated by the addition of [75Se]selenite (H2SeO3 in 0.1 M HCl, 111 MBq ml–1, 4,847 GBq g–1; Isotope Products Laboratories, Burbank, CA, U.S.A.). The 75Se-uptake reaction was terminated at intervals by separating the cells from the [75Se]selenite reaction medium using a centrifugation–filtration technique, as follows. Each 200-µl aliquot of the cell suspension was layered onto 60 µl of silicone oil (SH550:SH556, 2 : 1; Dow Corning Toray Silicone Co., Ltd., Tokyo, Japan) that had been layered over a killing solution, composed of 20 µl of 1 M glycine (pH 10.0) containing 0.75% SDS, which had been added first in the bottom of a 0.4-ml microcentrifuge tube (No. 72.700; Assist, Tokyo, Japan). The preparation was immediately centrifuged at 10,000×g for 1 min to sediment the cells, and then the tube was immediately frozen in liquid nitrogen and cut at the middle of the silicone layer. The radioactivity in the bottom fraction was determined using a gamma counter (COBRA II, Packard Instrument Co., Meriden, CT, U.S.A.). 75Se labeling of cells and fractionation of 75Se-labeled compounds For the continuous labeling experiments, E. huxleyi cells were incubated with 1 µM [75Se]selenite under the standard conditions. For the pulse–chase experiments, the cells were incubated with 3 µM [75Se]selenite for 1 h and then harvested by centrifugation. After the cells had been washed with fresh medium, they were immediately resuspended in MA-ESM medium without Se. The 75Se-labeled cells were harvested by centrifugation at 1,600×g for 5 min at 4°C and washed again with fresh medium. The resulting pellets were treated with 1.5 ml of CHCl3-MeOH (2 : 1, v/v) and centrifuged (18,000×g, 30 min, 4°C). When the supernatant (Fraction A, 1.2 ml) was mixed with 0.3 ml of 0.88% KCl solution, it separated into upper and lower layers that contained LMCs and lipids, respectively. The pellet (fraction B) was treated with 0.5 ml of 5% trichloroacetic acid (TCA) in a boiling water bath for 30 min and then centrifuged to produce supernatant and precipitate, which contained PS & NA and proteins, respectively. The levels of radioactivity were measured separately using a gamma counter. TLC analysis After the cells had been incubated with [75Se]selenite for 6 h, they were harvested and fractionated by the above-mentioned method. The LMC fraction obtained was concentrated and subjected to TLC on a silica gel 70 plate (Wako, Osaka, Japan); the plate was developed with phenol-water-acetic acid (83 : 16 : 1, by vol). The positions of Se-Met (Wako), selenocystine (Sigma, St. Louis, MO, U.S.A.) and [75Se]selenite were determined in separate experiments with standard compounds. Se-Met and selenocystine were detected by ninhydrin staining. Radioactive spots on the TLC plates were detected using BAS5000 (Fuji Film, Tokyo, Japan). Acknowledgments This work was funded by a Sasakawa Scientific Research Grant from the Japan Science Society to T.O. (13–381MK in 2001), a Grant-in-Aid for Scientific Research in an exploratory research area from the Japanese Ministry of Education, Culture, Sports, Science, and Technology to Y.S. (15657026 in 2003-4), and a Research Grant for Constructing a Genetic Resource Library of Unidentified Microbes based on Genomic Information (Metagenome Project) from the New Energy and Industrial Technology Development Organization (NEDO) of Japan to Y.S. 1 Corresponding author: E-mail, [email protected]; Fax, +81-29-853-6614. View largeDownload slide Fig. 1 Time course of [75Se]selenite incorporation into E. huxleyi cells. (A) Short-term 75Se uptake at 2.6 µM selenite. (B) Long-term [75Se]uptake at 1 µM selenite. Selenite uptake assays were performed at intervals. The solid line represents total uptake. Dashed and dotted lines represent the primary and secondary phases, respectively. The line of the secondary phase was drawn by calculation from linear regression analysis of total uptake after 2 min. The dashed line was obtained by subtraction of the extrapolated secondary phase from the total uptake. View largeDownload slide Fig. 1 Time course of [75Se]selenite incorporation into E. huxleyi cells. (A) Short-term 75Se uptake at 2.6 µM selenite. (B) Long-term [75Se]uptake at 1 µM selenite. Selenite uptake assays were performed at intervals. The solid line represents total uptake. Dashed and dotted lines represent the primary and secondary phases, respectively. The line of the secondary phase was drawn by calculation from linear regression analysis of total uptake after 2 min. The dashed line was obtained by subtraction of the extrapolated secondary phase from the total uptake. View largeDownload slide Fig. 2 Kinetic analysis of the transport of selenite into E. huxleyi cells as a function of external selenite concentration. (A) The rate of selenite uptake by E. huxleyi cells versus the external selenite concentration. The rates were calculated from the data of the 30-s experimental time. The experimental data (plots) can be fitted by the sum of a saturable component (dotted line) and a linear component (dashed line). Each point and the vertical bars represent the average of three independent measurements and the standard deviation, respectively. (B) Double reciprocal plot of the saturable component (dotted line) in (A). View largeDownload slide Fig. 2 Kinetic analysis of the transport of selenite into E. huxleyi cells as a function of external selenite concentration. (A) The rate of selenite uptake by E. huxleyi cells versus the external selenite concentration. The rates were calculated from the data of the 30-s experimental time. The experimental data (plots) can be fitted by the sum of a saturable component (dotted line) and a linear component (dashed line). Each point and the vertical bars represent the average of three independent measurements and the standard deviation, respectively. (B) Double reciprocal plot of the saturable component (dotted line) in (A). View largeDownload slide Fig. 3 Time courses of the incorporation of 75Se into compounds of E. huxleyi in the continuous (A) and pulse–chase (B) experiments. At intervals, cells were harvested; an aliquot was used for the determination of total uptake by cells (closed circle), and the remaining sample was fractionated into LMC (closed diamond), proteins (open circle), PS & NA (closed triangle) and lipids (open diamond). (A) A continuous 75Se-labeling pattern with 1 µM [75Se]selenite as the substrate. (B) Changes in the 75Se-labeling pattern during the chase period. Cells were pulse-labeled by incubation with 1 µM [75Se]selenite for 1 h. After the pulse-labeling, cells were centrifuged to remove [75Se]selenite and resuspended in fresh MA-ESM medium without Se and then changes in the 75Se-labeling pattern between compounds were chased. Data shown in this figure represent typical data from three separate experiments. View largeDownload slide Fig. 3 Time courses of the incorporation of 75Se into compounds of E. huxleyi in the continuous (A) and pulse–chase (B) experiments. At intervals, cells were harvested; an aliquot was used for the determination of total uptake by cells (closed circle), and the remaining sample was fractionated into LMC (closed diamond), proteins (open circle), PS & NA (closed triangle) and lipids (open diamond). (A) A continuous 75Se-labeling pattern with 1 µM [75Se]selenite as the substrate. (B) Changes in the 75Se-labeling pattern during the chase period. Cells were pulse-labeled by incubation with 1 µM [75Se]selenite for 1 h. After the pulse-labeling, cells were centrifuged to remove [75Se]selenite and resuspended in fresh MA-ESM medium without Se and then changes in the 75Se-labeling pattern between compounds were chased. Data shown in this figure represent typical data from three separate experiments. View largeDownload slide Fig. 4 Autoradiograph of the TLC analysis of 75Se-labeled LMCs obtained from cells of E. huxleyi incubated with [75Se]selenite for 6 h. The LMC fraction (sample), the [75Se]selenite and standard compounds were developed in lanes 1, 2 and 3, respectively. Open triangle, 75Se-labelled compounds in LMC fraction; closed triangle, radioactive spot of 75SeO32– and standards of Se-Met and selenocystine colored by ninhydrin staining. View largeDownload slide Fig. 4 Autoradiograph of the TLC analysis of 75Se-labeled LMCs obtained from cells of E. huxleyi incubated with [75Se]selenite for 6 h. The LMC fraction (sample), the [75Se]selenite and standard compounds were developed in lanes 1, 2 and 3, respectively. Open triangle, 75Se-labelled compounds in LMC fraction; closed triangle, radioactive spot of 75SeO32– and standards of Se-Met and selenocystine colored by ninhydrin staining. View largeDownload slide Fig. 5 Effects of inhibitors of amino acid (A) and protein (B) synthesis on the incorporation of 75Se into various cellular components. (A) Effect of AOA and l-MSO. Cells were treated with AOA (1.0 mM, final concentration, closed diamond) and l-MSO (0.3 mM, final concentration, closed triangle) for 0.5 h before the addition of [75Se]selenite. (B) Effect of CHI. CHI (0.2 mM, final concentration) was added to the medium at 0 h (open diamond) or 2 h (open triangle) after the addition of [75Se]selenite. Arrows, addition of CHI; closed circle, control performed without inhibitors. View largeDownload slide Fig. 5 Effects of inhibitors of amino acid (A) and protein (B) synthesis on the incorporation of 75Se into various cellular components. (A) Effect of AOA and l-MSO. Cells were treated with AOA (1.0 mM, final concentration, closed diamond) and l-MSO (0.3 mM, final concentration, closed triangle) for 0.5 h before the addition of [75Se]selenite. (B) Effect of CHI. CHI (0.2 mM, final concentration) was added to the medium at 0 h (open diamond) or 2 h (open triangle) after the addition of [75Se]selenite. Arrows, addition of CHI; closed circle, control performed without inhibitors. View largeDownload slide Fig. 6 Diagram showing the possible mechanism of selenium bioconcentration by Emiliania huxleyi. LMC1-5, five Se-containing LMCs. View largeDownload slide Fig. 6 Diagram showing the possible mechanism of selenium bioconcentration by Emiliania huxleyi. LMC1-5, five Se-containing LMCs. Table 1 Effects of light and metabolic inhibitors on the rate of primary selenite uptake at 160 nM and 2.63 µM Se under conditions in which the contributions of the active and passive transport systems were dominant Condition  Percent of control    160 nM  2.63 µM  Dark (control)  100  100  Light (100 µmol m–2 s–1)  126  113  Dark + KCN (1 mM), CCCP (170 µM), SHAM (1 mM)  58.9  68.1  Dark + CCCP (170 µM)  59.9  103  Dark + DES (10 µM)  48.6  98.3  Dark + DIDS (500 µM)  50.0  133  Condition  Percent of control    160 nM  2.63 µM  Dark (control)  100  100  Light (100 µmol m–2 s–1)  126  113  Dark + KCN (1 mM), CCCP (170 µM), SHAM (1 mM)  58.9  68.1  Dark + CCCP (170 µM)  59.9  103  Dark + DES (10 µM)  48.6  98.3  Dark + DIDS (500 µM)  50.0  133  The cells were preincubated in darkness with the inhibitors for about 10 min prior to the initiation of the reaction by the addition of [75Se]selenite. The data are the mean values of two or three experiments. View Large Abbreviations AOA aminooxyacetic acid CCCP carbonyl cyanide 3-chlorophenylhydrazone CHI cycloheximide DES diethylstilbestrol DIDS 4,4′-diisothiocyanatostilbene-2,2′-disulfonate DMSeP dimethyl selenoniopropionate ESM Erd-Schreiber’s medium l-MSO l-Methionine sulfoximine MA Marine Art LMC low-molecular-mass compound PS & NA polysaccharides and nucleic acids Se-Met selenomethionine SHAM salicylhydroxamic acid TCA trichloroacetic acid. References Baines, S.B. and Fisher, N.S. ( 2001) Interspecific differences in the bioconcentration of selenite by phytoplankton and their ecological implications. Mar. Ecol. Prog. Ser.  213: 1–12. Google Scholar Bottino, N.R., Banks, C.H., Irgolic, K.J., Micks, P., Wheeler, A.E. and Zingaro, R.A. ( 1984) Selenium-containing amino acids and proteins in marine algae. Phytochemistry  23: 2445–2452. Google Scholar Cutter, G.A. and Bruland, K.W. ( 1984) The marine biogeochemistry of selenium: A re-evaluation. Limnol. Oceanogr.  29: 1179–1192. Google Scholar Cutter, G.A. and Cutter, L.S. ( 2001) Sources and cycling of selenium in the western and equatorial Atlantic Ocean. Deep-Sea Res. II  48: 2917–2931. Google Scholar Danbara, A. and Shiraiwa, Y. ( 1999) The requirement of selenium for the growth of marine coccolithophorids, Emiliania huxleyi, Gephyrocapsa oceanica and Helladosphaera sp. (Prymnesiophyceae). Plant Cell Physiol.  40: 762–766. Google Scholar Doblin, M.A., Blackburn, S.I. and Hallegraeff, G.M. ( 1999) Comparative study of selenium requirements of three phytoplankton species: Gymnodinium catenatum, Alexandrium minutum (Dinophyta) and Chaetoceros cf. tenuissimus (Bacillariophyta). J. Plankton Res.  21: 1153–1169. Google Scholar Doblin, M.A., Blackburn, S.I. and Hallegraeff, G.M. ( 2000) Intraspecific variation in the selenium requirement of different geographic strains of the toxic dinoflagellate Gymnodinium catenatum. J. Plankton Res.  22: 421–432. Google Scholar Fu, L.H., Wang, X.F., Eyal, Y., She, Y.M., Donald, L.J., Standing, K.G. and Ben-Hayyim, G. ( 2002) A selenoprotein in the plant kingdom: Mass spectrometry confirms that an opal codon (UGA) encodes selenocysteine in Chlamydomonas reinhardtii glutathione peroxidase. J. Biol. Chem.  277: 25983–25991. Google Scholar Furihata, T., Pomprasirt, S. and Sakurai, H. ( 1997) Characteristics of sulfite transport by Chlorella vulgaris. Plant Cell Physiol.  38: 398–403. Google Scholar Läuchli, A. ( 1993) Selenium in plants: Uptake, functions and environmental toxicity. Bot. Acta  106: 455–468. Google Scholar Libes, S.M. ( 1992) Oceanic residences times and elemental concentrations in river water and seawater. In An Introduction to Marine Biogeochemistry. Edited by Libes, S.M. p. 683. John Wiley & Sons, Inc., New York. Google Scholar Neuhierl, B. and Böck, A. ( 1996) On the mechanism of selenium tolerance in selenium-accumulating plants – Purification and characterization of a specific selenocysteine methyltransferase from cultured cells of Astragalus bisculatus. Eur. J. Biochem.  239: 235–238. Google Scholar Novoselov, S.V., Rao, M., Onoshko, N.V., Zhi, H., Kryukov, G.V., Xiang, V., Weeks, D.P., Hatfield, D.L. and Gladyshev, V.N. ( 2002) Selenoproteins and selenocysteine insertion system in the model plant cell system, Chlamydomonas reinhardtii. EMBO J.  21: 3681–3693. Google Scholar Obata, T., Sera, K., Futatsugawa, S. and Shiraiwa, Y. ( 2004) Multi-element analysis of marine microalgae using particle-induced X-ray emission (PIXE). Mar. Biotechnol.  In Press. Google Scholar Osaka, T., Beika, A., Hattori, A., Kohno, Y., Kato, K.H. and Mizutani, T. ( 2003) The protozoa dinoflagellate Oxyrrhis marina contains selenoproteins and the relevant translation apparatus. Biochem. Biophys. Res. Commun.  300: 236–240. Google Scholar Riebesell, U., Zondervan, I., Rost, B., Tortell, P.D., Zeebe, R.E. and Morel, F.M.M. ( 2000) Reduced calcification of marine plankton in response to increased atmospheric CO2. Nature  407: 364–367. Google Scholar Sekino, K. and Shiraiwa, Y. ( 1996) Evidence for the involvement of mitochondrial respiration in calcification in a marine coccolithophorid, Emiliania huxleyi. Plant Cell Physiol.  37: 1030–1033. Google Scholar Shiraiwa, Y. ( 2003) Physiological regulation of carbon fixation in the photosynthesis and calcification of coccolithophorids. Comp. Biochem. Physiol. B  136: 775–783. Google Scholar Spallholz, J.E. ( 1994) On the nature of selenium toxicity and carcinostatic activity. Free Radical Bio. Med.  17: 45–64. Google Scholar Stolz, J.F. and Oremland, R.S. ( 1999) Bacterial respiration of arsenic and selenium. FEMS Microbiol. Rev.  23: 615–627. Google Scholar Terry, N., Zayed, A.M., De Souza, M.P. and Tarun, A.S. ( 2000) Selenium in higher plants. Annu. Rev. Plant Physiol. Plant Mol. Biol.  51: 401–432. Google Scholar Wolfe, G.V., Steinke, M. and Kirst, G.O. ( 1997) Grazing-activated chemical defence in a unicellular marine alga. Nature  387: 894–897. Google Scholar