Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Positional Cloning Identifies Lotus japonicus NSP2, A Putative Transcription Factor of the GRAS Family, Required for NIN and ENOD40 Gene Expression in Nodule Initiation

Positional Cloning Identifies Lotus japonicus NSP2, A Putative Transcription Factor of the GRAS... Rhizobia-secreted Nod-factors (NFs) are required for nodulation. In the early developmental process of nodulation, a large number of changes occur in gene expression. Lotus japonicus nsp2 mutants isolated from Gifu B-129 ecotype have defects in nodule initiation and display non-nodulating phenotype. Here, we describe positional cloning of LjNSP2 as a component of the nodulation-specific signaling pathway. LjNSP2 was mapped near the translocation site of chromosome 1 where the recombination is severely suppressed. To circumvent this problem, we introduced Lotus burttii as an alternative crossing partner in place of L. japonicus Miyakojima. The development of the high-resolution map using a total of 11 481 F2 plants, in combination with newly developed DNA markers and construction of BAC library, enabled us to identify the gene responsible for mutant phenotype. LjNSP2 encodes a putative transcription factor of the GRAS family that constitutes a subfamily with Medicago truncatula NSP2. LjNSP2 was expressed in roots and early nodules, but strongly suppressed in matured nodules. The expression analysis of NIN and LjENOD40-1 genes in Ljnsp2 mutants indicates that LjNSP2 functions upstream of these genes. These results suggest that LjNSP2 acts as a transcription factor to directly or indirectly switch on the NF-induced genes required for nodule initiation. Key words: positional cloning; Lotus japonicus; transcription factor; nodule initiation 1. Introduction by rhizobia are responsible for nodule formation and induce a variety of responses in a host-specific manner, The legume–rhizobia symbioses lead to the formation including root hair deformation and cortical cell division, 2þ of novel organs, termed nodules, which arise from division during the early steps of nodulation. Ca spiking in root of cortical cells in the root and the infection of these hair cells is one of the most early responses to NFs. nodules by rhizobia. Rhizobia within nodule cells differ- In the past few years, the phenotypes of an increasing entiate to bacteroids which fix atmospheric nitrogen. number of symbiotically defective mutants have been Lipochitin oligosaccharides, Nod-Factors (NFs) secreted analyzed in the model legumes, Lotus japonicus and Medicago truncatula to dissect the NF signaling 2–4 Communicated by Satoshi Tabata pathway. Several genes required for nodulation have * To whom correspondence should be addressed. Tel./Fax: þ81-3- been identified by positional cloning. L. japonicus 5841-4458, E-mail: [email protected] 5 6 7 NFR1, NFR5 and Medicago LYK3 encode trans- y Present address: Biotechnology Research Center, The University membrane receptor-like serine/threonine kinases with of Tokyo, 1-1-1 Yayoi, Bunkyo, Tokyo 113-8657, Japan The Author 2006. Kazusa DNA Research Institute. The online version of this article has been published under an open access model. Users are entitled to use, reproduce, disseminate, or display the open access version of this article for non-commercial purposes provided that: the original authorship is properly and fully attributed; the Journal and Oxford University Press are attributed as the original place of publication with the correct citation details given; if an article is subsequently reproduced or disseminated not in its entirety but only in part or as a derivative work this must be clearly indicated. For commercial re-use, please contact [email protected] Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 256 Identification of Lotus japonicus NSP2 [Vol. 13, putative extracellular regions similar to LysM domains, the addition of 10 M NF. Representative traces were which are thought to be directly involved in perception selected from at least 10 independent cells. of NF signal. In subsequent signal transduction, 8 9 2.3. Genetic mapping population and L. japonicus SYMRK, CASTOR and POLLUX, 10 11 12 genomic DNA isolation M. truncatula DMI1, DMI2 and DMI3 , and 13 14 L. japonicus CCaMK and Nup133 have been identi- Genetic mapping of the LjNSP2 gene was performed fied as components required for the common symbiosis with F2/F3 population derived from cross between pathway shared between the fungal and bacterial Ljnsp2-1 (Gifu B-129) and the early flowering ecotype 20 21 endosymbiotic systems. The L. japonicus nin mutant Miyakojima MG-20 or Lotus burttii B-303. For AFLP showed normal mycorrhization and early responses analysis, genomic DNA was extracted from 0.1 g leaves following rhizobial inoculation, including root hair using a Qiagen DNeasy plant kit according to the 15 2þ 2þ 16 deformation, Ca influx and Ca spiking. NIN manufacture’s protocol. For PCR markers analysis, was cloned as a putative transcription factor gene and DNA was extracted from one young leaf in 100 mlof shown to be expressed during both early and late stages PEB (200 mM Tris–HCl pH 7.5, 250 mM NaCl, 25 mM of nodule development, suggesting that NIN controls EDTA, 0.5% SDS), precipitated with isopropanol, various developmental aspects of nodulation after the washed and dissolved in 100 ml of TE (pH 8.0). perception of rhizobia in root hairs. However, little is 2.4. High efficiency genome scanning (HEGS)/ known about what signal component relays the signal AFLP or PCR-based screening of from the common symbiotic pathway into the LjNSP2-inked markers nodulation-specific program in L. japonicus. Recently, GRAS family genes, nodulation signaling pathway Genomic DNA (100 ng) was digested with EcoRI 1 (NSP1) and NSP2, have been identified as putative and MseI, ligated to EcoRI- and MseI-adaptors and 2þ transcription factors functioning downstream of Ca preamplified by using EcoRI- and MseI-adaptor primers. 17,18 spiking and CCaMK in Medicago truncatula. Preamplified DNA was prepared at 0.05 mg/ml concentra- Here, we describe the positional cloning and charac- tion before amplification with selective primers. A bulked terization of nodulation-specific LjNSP2 gene encoding segregant analysis was performed to identify markers a plant-specific GRAS protein most similar to MtNSP2, linked to LjNSP2. Bulks were constructed from which might lead to the induction of expression of preamplified DNAs of 10 recessive (Nod ) or dominant genes required for rhizobial infection and early nodule (Nod ) homozygous F2 plants, 4096 selective primer development. combinations of EcoRIþ3/MseIþ3 were screened to identify markers present only in the bulk of dominant homozygotes. Subsequently, LjNSP2-linked HEGS/AFLP markers 2. Materials and methods were excised directly from polyacrylamide gels and cloned with the TOPO TA Cloning Kit (Invitrogen). 2.1. Plant materials The cloned markers were sequenced and primers were The mutant carrying Ljnsp2-1 was isolated from the designed by using the software Primer 3 (Whitehead EMS-mutagenesis experiments of L. japonicus Gifu Institute, Cambridge, MA). SCAR (sequence charac- B-129. The mutant Ljsym35 was isolated from a terized amplified region) markers, which revealed poly- population of Gifu derived from a transposon-/ morphism between L. japonicus accession Gifu and DNA-tagging trial, as a non-tagged culture mutant, Miyakojima MG-20 or L. burttii, were analyzed in 2989 and was kindly provided by Prof. Jens Stougaard or 8472 F2/3 individuals from Miyakojima MG-20 or (Arrhus University, Denmark). L. burttii, respectively. To facilitate the efficiency of electrophoresis of AFLP 2.2. Root hair deformation and assays of or PCR products, the HEGS system was adapted. In 2þ Ca spiking this system, a set of electrophoresis apparatus is Seeds of L. japonicus were germinated and grown on equipped with two sets of 24.5 · 26.5 cm glass plates, BNM agar medium essentially as described previously, each accommodating a gel with 100 lanes and analysis except that the roots were grown between two filter of 400 samples is practicable in single run. After papers (grade 0860; Schleicher and Schull, UK), one of electrophoresis, the gels were stained with Vistra Green which was on the agar surface. Root hair deformation (Amersham Biosciences) and scanned by fluorescent gel was scored as described previously following 16 h scanner (FluorImager 595; Amersham Biosciences). exposure to 10 M NF in 1 ml BNM medium in a 2.5. BAC contig development chamber on a microscope slide. Images were taken with a 2þ digital-camera attached to an inverted microscope. Ca Three-dimensional BAC DNA pools prepared from 9 23 spiking was assayed as described previously, following our BAC library were screened with LjNSP2-linked Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 No. 6] Y. Murakami et al. 257 HEGS/AFLP or SCAR markers as described above. scanner (Storm840, Amersham Bioscience). After strip- HindIII-digested DNA from positive BACs was ping, the same filters were reprobed with the 400 bp fractionated on an agarose gel for fingerprinting and fragment from L. japonicus ubiquitin cDNA as a loading determination of overlaps. BAC ends were sequenced control. using M13 reverse or forward primer and non-repetitive 2.8. Transient expression of the LjNSP2-GFP fusion sequences in BAC were used for chromosome walking to protein in onion epidermal cells screen 3-D BAC pools with PCR or AFLP. The process The 1.5 kb ORF of LjNSP2 was amplified from 188C5 was repeated as needed to complete the BAC contig. BAC DNA using the primers: [5 -ACGCGTCGACA- Polymorphic PCR fragments in BAC sequences were TGGAAATGGATATAGATTGCATCC-3 (SalI-site analyzed to directly score recombinants in F2/F3 underlined), 5 -CATGTCATGAATGCACAATCTGA- populations. TTCTGAAGAAC-3 (BspHI-site underlined)], digested BAC clone 188C5 containing the LjNSP2 gene was with SalIand BspHI, and cloned into pUC18- shotgun-sequenced and then annotated by Rice-GAAS CaMV35SV-sGFP(s65T)-NOS plasmid at the SalI/ (Rice Genome Automated Annotation System) (http:// NcoI sites just upstream of GFP gene. Onion epidermal ricegaas.dna.affrc.go.jp/). cells were bombarded with DNA-coated particles using a Biolistic PDS-1000/He Particle Delivery System (Bio- 2.6. Complementation experiments Rad). About 18–24 h after bombardment, the cells were For complementation, a 6.9 kb PstI fragment, carrying observed using a Bio-Rad Radiance2000 confocal laser only the wild-type LjNSP2 1500 bp ORF and 4486 and scanning microscope. 899 bp of upstream and downstream sequence, respec- The sequence data of LjNSP2 and genomic sequences tively, was cloned into the hairy root transformation (B-129 Gifu) of LjNSP2 have been deposited with the vector, which was made from pCAMBIA1300 by replac- DDBJ data library under accession numbers AB241456 ing the hygromycin-resistant gene with sGFP(S65T). and AB241457, respectively. The resulting LjNSP2 recombinant plasmid was intro- duced into A. rhizogenes LBA1334 by electroporation. 3. Results Hairy root transformation of Ljnsp2-1 mutant was per- formed as described. The plants with transgenic hairy 3.1. Early infection phenotypes of a Ljnsp2 mutant roots were grown in vermiculite pots and inoculated with Mesorhizobium loti TONO. GFP fluorescence and nodule The Ljnsp2-1 mutant of L. japonicus grew normally formations were confirmed 4 weeks after inoculation. in nitrogen-rich compost, and established a normal symbiosis with the mycorrhizal fungus, but did not form nodules under nitrogen limitation when inoculated 2.7. Southern and northern hybridizations with M. loti. The addition of M. loti to Ljnsp2-1 seedlings Genomic DNA was extracted from leaves using CTAB induced no phenotypes, such as root hair curling, infec- method from L. japonicus, and 2.5 mg DNA were tion thread formation and cortical cell division, typically digested by EcoRI, electrophoresed on 0.8% agarose gel seen in the wild-type (data not shown). However, NF did and blotted to nylon membrane (Biodyne A, Pall). induce swelling and branching in root hair tips, although Twelve-day-old plants were inoculated with M. loti at a reduced level compared with the wild-type (Fig. 1A TONO. Infected roots at 4 days post inoculation (dpi) and H). Cytological staining of M. loti expressing lacZ and nodules at 8–32 dpi were immediately frozen in revealed no infection foci and no infection thread forma- liquid nitrogen. Total RNA was isolated from flower, 2þ tion (data not shown). Intracellular Ca spiking, which shoot, root and nodule tissue, and 5 mg aliquots were is induced by NF and has been proposed to be integrated electrophoresed by denaturing agarose gel and blotted as 2þ by a Ca -calmodulin-dependent protein kinase required above. for activation of early nodulation gene expression and LjNSP2 probe (1322 bp) was amplified from 188C5 mycorrhization , was indistinguishable in the mutant BAC DNA with primers 5 -ACTTCCACCACCTCATC- from the wild-type (Fig. 1I). The observed induction 0 0 0 GAC-3 and 5 -ACAAGTCCAAAGGGATGCAG-3 , 2þ of Ca spiking and normal mycorrhization of the and labeled with P using a random primer labeling mutant indicates that the mutation in Ljnsp2 affects a kit (Takara). Hybridization was done at 63 C in Church nodulation-specific signaling component that is down- buffer [0.5 M sodium phosphate, pH 7.2, 7% (w/v) SDS, stream of the common pathway for mycorrhizal and 1 mM EDTA] and the filters were washed once in 2· rhizobial symbioses. SSC containing 0.5% SDS at room temperature for 10 min, and twice in 0.2· SSC, 0.1% SDS at 63 C for 3.2. High-resolution genetic mapping of LjNSP2 15 min. The hybridized membranes were then exposed for 3 days to phosphor imaging plates (Fuji, Tokyo, The mutant allele Ljnsp2-1 (Gifu B-129) was crossed Japan), which were then scanned by a phosphor imaging to the early flowering ecotype Miyakojima MG-20. The Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 258 Identification of Lotus japonicus NSP2 [Vol. 13, Figure 1. L. japonicus root hair responses induced by NF. (A–D) Root hair deformation induced with 10 M NF. Slight swelling and branching were observed also in Ljnsp2-1 mutant (C and D), although to a lesser extent than the wild-type (Gifu B-129; A and B). (E–H) Non-treated 2þ root hairs look similar between the mutant (G and H) and wild-type (E and F). Scale bar: 50 mm. (I) Analysis of Ca spiking in root hairs. Each trace is from a single root hair using seedlings of wild-type (Gifu B-129), or the Ljnsp2 mutant. After about 20 min from injection of the 2þ 8 Ca -sensitive dye Oregon green-dextran to the root hairs, NF was added to the medium at 10 M. Both wild-type (upper) and the mutant (lower) responded positively for the addition of NF in 10–20 min and similar effects were observed with at least 10 independent seedlings. LjNSP2 locus was mapped near translocation site of the pools with AFLP system. Finally, we constructed a short arm of chromosome 1. Since the recombination is 2 Mb physical BAC contig that spans the LjNSP2 locus significantly suppressed in the chromosomal segment, (Fig. 2B). total of 4096 EcoRI/MseI primer combinations were exa- mined for bulked segregant analysis on bulks 10 Nod 3.4. The narrowing of LjNSP2 genomic region using recessive and Nod dominant homozygous F2 plants in L. burttii as an alternative crossing partner the cross of Ljnsp2-1 and Miyakojima MG-20. LjNSP2- and identification of LjNSP2 gene linked AFLP markers from this screen were further analyzed in additional 2989 F2 plants with HEGS/AFLP As no more recombination was found from the system. Nineteen markers are located on the southern population of this cross of Ljnsp2-1 · Miyakojima side of LjNSP2 locus at distances of 0.07 cM and seven MG-20, even with new markers developed from the markers on the northern side of LjNSP2 locus at dis- BAC clones in the contig, we made another cross with tances of 0.45 cM, respectively, while nine markers L. burttii. Among 8 472 F2 progenies from this cross, cosegregated with LjNSP2 locus (Table 1, Fig. 2A). six recombination events were found between the The screened markers were converted into the 11 flanking markers S26d and 183R derived from the BAC polymorphic SCAR markers (Table 2). clone 188C5; this located the mutation within a 130 kb region (Fig. 2B). Among the 5 ORFs (Fig. 2C) predicted from the sequence in this region (excluding transposable 3.3. Construction of a 2 Mb physical BAC contig elements), only one was identified as having a mutation spanning LjNSP2 in the mutant. A 6.9 kb fragment including this ORF At the first step, all SCAR markers were used to complemented the mutant for nodulation in hairy roots screen 3-D BAC DNA pools prepared from our transformed by the Agrobacterium rhizogenes carrying L. japonicus Gifu BAC library. The contigs containing the cloned region (Fig. 3). No nodules were formed using these SCAR markers were extended by the chromosome empty vector, confirming that mutation of this gene walking with the screening of 3-D BAC DNA pool using caused the mutant phenotype. PCR primer combinations based on end sequences of The ORF in this region corresponding to LjNSP2 BAC clones. In several cases, however, we could not encodes a protein belonging to the plant GRAS family 32–34 obtain non-repetitive PCR fragments from BAC end of putative transcription factors and analysis of sequences and screen BAC library. As an alternative the major plant GRAS family protein sequences indicated strategy, AFLP fragments from BAC with EcoRIþ1 that the closest in sequence to LjNSP2 were the selective and MseIþ1 selective primers were searched for M. truncatula and Pisum sativum NSP2 proteins. non-repetitive sequences of BAC inner. Subsequently, AtSCL26 of Arabidopsis was about twice distant from the corresponding EcoRIþ3 selective and MseIþ3 them, and together these proteins made an apparent selective primers were used to screen 3-D BAC DNA subfamily of GRAS proteins (Fig. 4). However, genome Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 No. 6] Y. Murakami et al. 259 Table 1. AFLP markers linked to LjNSP2 locus DNA-binding sequence, second leucine heptad repeat (LHRII); and a Src-homology 2 (SH2)-like domain AFLP marker Primer Approximate Marker type (Fig. 5A). About 15% of the N-terminal region is diver- combination length of the AFLP gent among most GRAS family members but that of fragment (bp) LjNSP2 showed strong homology with the M. truncatula EM117 E-CGA/M-GTG 700 Dominant and Pisum sativam NSP2 proteins suggesting functio- EM140 E-CTT/M-CAA 700 Co-dominant nal equivalence and that they are probably orthologs. EM157 E-GAC/M-GCC 130 Dominant Furthermore, LjNSP2 has a well-conserved SH2-like EM177 E-GCA/M-CAA 550 Dominant domain among GRAS proteins (Fig. 5B). The Ljnsp2-1 EM205 E-GGC/M-TGG 1500 Dominant mutation causes a substitution of a conserved valine (V) EM242 E-TAG/M-CTC 130 Dominant to glutamate (E) in this SH2-like domain (Fig. 5C). EM262 E-TCT/M-ACT 500 Co-dominant Complementation tests with the various L. japonicus nodulation mutants carrying mutations mapped to EM265 E-TCT/M-GGC 250 Co-dominant linkage group 1 revealed that Ljsym35 carried a mutation EM299 E-TTA/M-TGT 250 Dominant allelic to Ljnsp2. DNA hybridizations detected no signal EM341 E-AAT/M-GTT 500 Co-dominant in Ljsym35 (Fig. 6A) and PCR analyses indicated that it EM367 E-ACG/M-GGA 400 Dominant has a deletion of >100 kb around the LjNSP2 gene; this EM390 E-AGC/M-AAA 1000 Co-dominant allele was renamed Ljnsp2-2. EM480 E-CCA/M-GCA 190 Dominant EM527 E-CGT/M-CTC 70 Dominant 3.5. Expression of LjNSP2 during nodulation and in EM582 E-GAC/M-TGG 800 Dominant different organs EM603 E-GCA/M-TTA 800 Dominant RNA hybridization showed that LjNSP2 expression EM713 E-TAC/M-AAA 400 Dominant was detectable in roots but not in shoots and flowers EM784 E-TCT/M-TTC 100 Dominant (Fig. 6B), in contrast to the ubiquitously expressed Medicago NSP2. LjNSP2 was also expressed in infected EM807 E-TGG/M-CGA 90 Dominant roots at 4 dpi and early nodules but strongly suppressed EM011 E-AAG/M-TTC 200 Dominant in matured nodules (Fig. 6C). These expression patterns EM013 E-AAT/M-CCT 120 Dominant of LjNSP2 are similar to that of the putative Nod factor EM116 E-CGA/M-GGA 500 Dominant 5 6 receptor kinase genes, NFR1 and NFR5, and the signal EM160 E-GAG/M-AAT 250 Co-dominant transduction components CASTOR, although the decre- EM190 E-GCT/M-CAA 350 Co-dominant ase in expression of the latter was slight. This contrasts EM261 E-TCG/M-TTG 250 Co-dominant with the expression patterns of the genes encoding the EM283 E-TGG/M-CTG 500 Dominant nodulation signaling pathway components M. truncatula 17 18 NSP1, M. truncatula NSP2 and L. japonicus EM350 E-ACA/M-TCA 250 Dominant NIN (Fig. 6A) whose expression increased following EM466 E-CAT/M-ACT 800 Dominant inoculation. EM130 E-CGT/M-TGG 600 Dominant EM686 E-GTC/M-GAT 600 Dominant 3.6. Expression of NIN and LjENOD40-1 genes in EM163 E-GAG/M-GTC 300 Co-dominant Ljnsp2-2 mutant EM115 E-CGA/M-AGT 400 Co-dominant In the Ljnsp2-2 null mutant, RT–PCR revealed that EM214 E-GGG/M-TTA 850 Dominant even before inoculation with M. loti the expression levels EM231 E-GTT/M-TCT 80 Dominant of early nodulation genes, NIN and LjENOD40-1, were EM802 E-TGG/M-AAT 200 Dominant both <40% of the wild-type (Fig. 7A and B). Upon inoculation of the wild-type with M. loti, NIN and 36 LjNSP2-linked AFLP markers were selected from 3009 F2/F3 plants in the cross of Ljnsp2-1 (Gifu B-129) and LjENOD40-1 increased by up to 65- or 3.5-fold, respec- Miyakojima MG-20. tively. In contrast, the levels of NIN and LjENOD40-1 AFLP fragments were derived from Miyakojima MG-20. transcripts in the Ljnsp2-2 mutant remained low after inoculation reaching only about 5–20% of wild-type levels (Fig. 7A and B). This demonstrates that LjNSP2 sequences around the LjNSP2 and M. truncatula NSP2 function is required, directly or indirectly, for either the revealed no clear co-linearity based on the available data. expression and/or the induction of these early nodulins. The LjNSP2 gene consists of a 1500 bp exon with no introns, encoding a predicted 499 amino acids protein of 3.7. Nuclear localization of LjNSP2 in onion cells 55 kDa, containing the following GRAS family-specific domains: homopolymeric stretches (HPS) of polyE and Although most GRAS family proteins have a puta- polyT; first leucine heptad repeat (LHRI); a VHIID tive nuclear localization sequence (NLS) , PSORT II Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 260 Identification of Lotus japonicus NSP2 [Vol. 13, Figure 2. Positional cloning of the LjNSP2 gene (A) Genetic map of LjNSP2 region. This map is based on the analysis of 3 009 F2 plant of a cross between Ljnsp2-1 and Miyakojima MG-20. These HEGS/AFLP makers were selected from 4 096 of AFLP primer combinations with HEGS system. The markers conversed into polymorphic SCAR markers are underlined. (B) About 2 Mb of BAC contig around LjNSP2 genomic region. The numbers below the markers indicates recombination events detected in 16 944 chromosomes from 8 472 siblings of the heterozygotes from crosses of Ljnsp2-1 · L. burttii . Only overlapping BAC clones spanning LjNSP2 are displayed in this figure. Finally, LjNSP2 region was delimited within a 130 kb region on a 211 kb BAC clone (188C5). (C) Putative genes in LjNSP2 region. ORFs of putative genes (closed arrows) and retro-elements (RE; open arrows) are marked in this 130 kb region. Predicted genes are GTK, glutamine transferase K; P4P5K, phosphatidilinositol-4-phosphate-5-kinase; UP, unknown proteins. There were no introns in the 1 500 bp of the predicted LjNSP2 ORF. analysis did not identify an NLS-like sequence in containing a 5.1 kb fragment of the 5 flanking sequence LjNSP2. However, LjNSP2-GFP fusion delivered into was constructed and introduced into Ljnsp2-2 mutants. onion (Allium cepa) epidermal cells by particle bombard- These constructs complemented the Ljnsp2-2 mutant ment revealed it to be exclusively localized in nuclei, phenotype, indicating that this fusion protein retained although not in nucleoli (Fig. 8A). The mutation in the the LjNSP2 activity for nodulation. However, the SH-2-like domain reduced nuclear localization of Ljnsp2- transgenic roots showed no detectable GFP fluorescence 1-GFP and fluorescence in the cytoplasm became (data not shown). noticeable in at least 30 GFP-expressed cells (Fig. 8B), but not as strong as seen with GFP alone (Fig. 8C). 4. Discussion These observations are consistent with LjNSP2 acting as a transcription factor, and the SH2-like domain So far the only described nodulation-specific mutants facilitating its nuclear localization. of L. japonicus, which completely lack nodules but 5 6 The above LjNSP2-GFP fusion under the control of have normal mycorrhization, are Ljnsp2, nfr1, nfr5 the CaMV 35S promoter or the LjNSP2 promoter and nin. One key difference among these mutants is the Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 No. 6] Y. Murakami et al. 261 Table 2. PCR-based markers linked to LjNSP2 locus 0 0 PCR marker Origin Sequence (5 !3 ) Anealing Length (bp) Marker type temperature ( C) SEM140 EM140 GAATTCCTTCCCGGTTCTTC 65 750 Dominant (M) TTAACAACAGCAACTTCACT SEM299 EM299 GAATTCTTACCGAGTGAGAT 60 250 Dominant (M, b) TTAATGTTTCTGACAAAGGC SEM341 EM341 GAATTCAATGATTGTTGGGA 50 500 Co-dominant (M, b) TTAAGTTGTTCGAATTGATA SEM390 EM390 GAATTCAGCCACAGCCTCTG 50 230 Dominant (M) AACCACAAGTCAAATGCAAC SEM160 EM160 GAATTCGAGAATTGGAGAAG 60 220 Dominant (M) TTAAAATCAAACCCTAACAA SEM190 EM190 GAATTCGCTTGCAAGTAGTG 50 340 Co-dominant (M) TTAACAAGAGTCTCATAATC SEM261 EM261 TTCTCGGCGGTTCCACCAAT 68 290 Dominant (M) GCTTGTTTGAAAGTGGCTTT Co-dominant (b) SEM115 EM115 CCGATTACATTTATGAGTAC 50 350 Co-dominant (M, b) TTAAAGTCGATAGACCGACG SEM163 EM163 GAATTCGAGATCCAGAATCT 60 290 Dominant (M) TTAAGTCTGAACCATTGCTA SEM214 EM214 GAATTCGGGGTGTTACACCTC 65 580 Dominant (M) TTAATTACCCCCTCGTTCTC SEM802 EM802 GAATTCTGGAAGCACGTTGA 60 190 Co-dominant (M) TTAAAATCATCATCTGAGAG s26d 188C5 TCATGTTTAGCGCTCTGATT 60 100 Co-dominant (M, b) GCACCGTAGTACCGTCAGTTTCC S11 188C5 TTGCAGGTTTTCAGGAACATT 60 520 Co-dominant (M, b) GAAGTTCATTGTAACTCTCAAACG 183R 183A2 CATCGTATGTTGAAAAGAAGAATGAT 60 580 Co-dominant (M, b) TCTTCAGTTCTTCCCTTTATAGCC PCR was performed using the indicated primer pairs with the following cycling parameter: 4 min at 94 C; 35 cycles of 30 s at 94 C, 30 s at the temperature indicated, 1 min at 72 C; 10 min at 72 C; and hold at 4 C. PCR fragments were electrophoresed on 13% acrylamide gels with HEGS system. PCR fragments are derived from Miyakojima MG-20. M and b indicates the polymorphism of Miyakojima MG-20 and L. burttii, respectively, compared with Gifu B-129. 2þ induction of Ca spiking, which might be required for even in the population of 3 009 F2/F3 plants of the the early common response in activation of the symbioses cross with Miyakojima MG-20. This corresponds to with both rhizobia and mycorrhizal fungi. The Ljnsp2 24 Mb/cM, 78-fold greater than the average physical 2þ mutants are normal for Ca spiking but the nfr1 and to genetic distance found in our high-density map of the nfr5 mutants are blocked. Recently, NFR1 and NFR5 L. japonicus (Wang et al., manuscript in preparation). were shown to encode LysM-receptor-like-kinases, that This indicates that recombination in the chromosomal 5,6 were predicted to function in NF perception. NIN segment near the translocation site is highly suppressed functions downstream of LjNSP2 as indicated in this in the cross combination of Gifu B-129 · Miyakojima study. It is possible that LjNSP2 is earliest known MG-20. Therefore, we changed the crossing partner from protein executing nodulation-specific gene expression Miyakojima MG-20 to L. burttii. As a result, LjNSP2 2þ from Ca spiking induced through NFR1 and NFR5 in region narrowed up to 130 kb in one BAC clone (188C5) L. japonicus. using the population of 8 472 F2 plants of the cross with We have established a 2 Mb physical BAC contig L. burttii, indicating that L. burttii is significantly useful that spans the LjNSP2 locus (Fig. 2B). However, LjNSP2 as alternative crossing partner of Gifu especially near the region was only closed to the minimum of 14 BAC clones translocation site of chromosome 1. Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 262 Identification of Lotus japonicus NSP2 [Vol. 13, LjNSP2 encodes a GRAS family protein of a putative transcription factor. The phylogenetic analysis revealed that LjNSP2 is most closely related to M. truncatula NSP2 and P. sativum NSP2. The comparison of the Figure 4. Phylogenetic tree of plant GRAS family proteins. Members of the family of plant GRAS proteins aligned using Clustal W are Figure 3. Complementation of Ljnsp2-1 mutant. Roots of the Ljnsp2 shown as a neighbor-joining dendrogram with 1000 bootstrap mutant were transformed by Agrobacterium rhizogenes strain replicates. At, Arabidopsis thaliana; Mt, Medicago truncatula; Ph, LBA1334 carrying appropriate plasmids carrying the 1 500 bp Petunia hybrida; Ps, Pisum sativum; Le, Lycopersicum esculentum; LjNSP2 ORF and 4 486 and 899 bp of upstream and downstream NSP, nodulation signaling pathway; SCR, SCARECROW; SCL, sequences and a transformation indicator 35S-GFP. On inoculation SCARECROW-like; SHR, short-root; HAM, hairy meristem; Ls, with Mesorhizobium loti, the GFP-marked transformants formed lateral suppressor; RGA, repressor of ga1-3; PAT, phytochrome A nodules complementing the mutant phenotype. Scale bar: 1 mm. signal transduction. Figure 5. Domain structure of the LjNSP2 protein. (A) Amino acid sequence of 499 residues and predicted functional domains of LjNSP2. HPS; homopolymeric stretches characteristic for GRAS protein near to the N-terminal, LHR (leucine heptad repeat) 1 and 2; putative leucine zipper, VHIID; putative DNA-binding sites, SH2-like region, GY; GY, Y of which is phosphorylated in STATs, is conserved about 100 residues downstream from SH2-like domain. (B) Comparison of GRAS and STAT family proteins with LjNSP2 in their SH2(-like) domains. STAT2: human STAT2; P52630, Ce-STAT: Caenorhabditis elegans STAT; Z70754, D-STAT: Drosophila stat; Q24151, Dd-STAT: Dictyostelium discoideum STAT; Y13097, GAI: Arabidopsis GAI (gibberellin insensitive); At1g14920, SCR: Arabidopsis SCR (SCARECROW); At3g54220. GAI and SCR are the representatives of plant GRAS family proteins. Conserved amino acids are indicated by red boxes. A missense mutation in Ljnsp2-1 is indicated by arrowhead. Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 No. 6] Y. Murakami et al. 263 AC B Figure 8. Nuclear localization of the LjNSP2-GFP. LjNSP2-GFP (A), Ljnsp2-1-GFP (B) fusion genes and control GFP gene (C), delivered by particle bombardment, were transiently expressed in onion epidermal cells and observed with a laser confocal microscope 24 h after bombardment. The nuclear localization of the Ljnsp2-1 mutant gene product is significantly suppressed compared with LjNSP2. Scale: 10 mm. after rhizobial inoculation. These results suggest that, in contrast with a possible role for M. truncatula NSP2 in another organ development in addition to nodulation, LjNSP2 seems to be specialized in nodule initiation. Figure 6. DNA, RNA hybridization and RT–PCR analyses of LjNSP2. (A) Genomic hybridization with the full-length LjNSP2 The expression of NIN and ENOD40 is induced rapidly ORF probe to the wild-type (WT; Gifu B-129) and a null deletion after rhizobial inoculation and NF-treatment. The NIN mutant Ljnsp2-2. Genomic DNA was digested with EcoRI. The transcripts are detected in different tissues during LjNSP2 homolog seems to be unique in WT, but cannot be detected various nodule stages, such as the dividing cells of in Ljnsp2-2. (B and C) Northern analysis of LjNSP2 expression in various organs (B), and infected roots at 4 dpi of M. loti TONO and the nodule primordia and the nodule vascular bundles. nodules at 8–32 dpi of M. loti TONO, respectively (C). DPI: days The Ljnin mutants are nodulation-minus and blocked in post inoculation. both the infection thread formation and the cortical cell division. ENOD40 is induced in the root pericycle a few hours after rhizobial inoculation, and subsequently in the dividing cortical cells of the root and nodule 36–38 39 primordia. ENOD40 RNAi knock-down lines or the possible co-suppression suppressed nodule primor- dium formation, resulting in very poor nodulation. Despite of their function in nodule initiation, however, little is known about what kinds of transcription factors in the root activate NIN and ENOD40 gene expression in response to rhizobia. In this study, we demonstrate that the induction of NIN and LjENOD40-1 gene expression is clearly cancelled by the Ljnsp2 mutation (Fig. 7). On the basis of these findings, we speculate that LjNSP2 may function as a transcriptional activator to Figure 7. RT–PCR analyses of expression of LjNSP2 and the early directly or indirectly switch on the NIN and LjENOD40- nodulins. (A and B) RT–PCR analysis of the change of expression of 1 gene expression in nodule initiation. In order to address NIN (A) and LjENOD40-1 (B) after M. loti inoculation, in wild-type this issue, identification of the promoter region and and Ljnsp2. Suppression of LjNSP2 expression after inoculation was confirmed, and almost complete loss of induction in NIN and subsequent binding assay using LjNSP2 protein would be LjENOD40-1 were apparent. Ubiquitin is as a loading control. of great importance in future. Relative expression levels were normalized against the amount of Kalo´ et al. reported that M. truncatula NSP2-GFP ubiquitin (set as 10 ). Standard deviations of three independent localizes in the endoplasmic reticulum and nuclear experiments are indicated by error bars. DPI: days post inoculation. envelope and re-localizes into the nucleus rapidly after NF-treatment. In this case, they made a functional mutant phenotypes suggests that the LjNSP2, MtNSP2 C-terminal GFP fusion under control of the constitutive and PsNSP2 genes function at similar or parallel CaMV 35S promoter that was introduced in Medicago positions in the nodulation signal transduction. nsp2 mutant plants. Here, we attempted to detect However, the expression pattern is different between LjNSP2-GFP, -YFP or -DsRED2 fusion in L. japonicus LjNSP2 and M. truncatula NSP2. LjNSP2 is predomi- hairy roots. These fusions could complement the mut- nantly expressed in roots and its expression decreases in ant phenotype but no fluorescence was detected even developed nodules. The expression of M. truncatula under the control of CaMV35S promoter. In place of NSP2 is observed in shoots as well as roots and is induced L. japonicus we delivered LjNSP2-GFP fusion into onion Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 264 Identification of Lotus japonicus NSP2 [Vol. 13, 9. Imaizumi-Anraku, H., Takeda, N., Charpentier, M., et al. epidermal cells by particle bombardment. The fusion 2005, Plastid proteins crucial for symbiotic fungal and exclusively localizes in the nucleus but not in the nuc- bacterial entry into plant roots, Nature, 433, 527–531. lear envelope. Although there is no way to explain the 10. Anee, _ J. M., Kiss, G. B., Riely, B. K., et al. 2004, Medicago peculiar difference of subcellular localization of NSP2 truncatula DMI1 required for bacterial and fungal sym- between Medicago and onion, putative nucleoporins bioses in legumes, Science, 303, 1364–1367. such as NUP133 and NUP85 (Saito et al. unpublished 11. Endre, G., Kereszt, A., Kevei, Z., Mihacea, S., Kalo´, P., data) required for rhizobial and arbuscular mycorrhizal and Kiss, G. B. 2002, A receptor kinase gene regulating symbioses may retain the NSP2 putative transcription symbiotic nodule development, Nature, 417, 962–966. factor in the nuclear envelope of M. truncatula.Most 12. Leevy, _ J., Bres, C., Geurts, R., et al. 2004, A putative 2þ recently, Heckmann et al. reported that non-nodulating Ca and Calmodulin-dependent protein kinase required mutant, SL781-3, carrying an allele of LjNSP2 was found for bacterial and fungal symbiosis, Science, 303, 1361–1364. by TILLING. 13. Tirichine, L., Imaizumi-Anraku, H., Yoshida, S., et al. Acknowledgements: We thank Prof. Jens 2þ 2006, Deregulation of a Ca /calmodulin-dependent kinase Stougaard for kind providing the mutant seeds of leads to spontaneous nodule development, Nature, 441, Ljsym35. We wish to thank Krzyztof Szczyglowski and 1153–1156. Martin Parniske for valuable comments on the manu- 14. Kanamori, N., Madsen, L. H., Radutoiu, S., et al. 2006, A script and Anne Heckman for helpful discussion. This 2þ nucleoporin is required for induction of Ca spiking in work was supported by the fund of Promotion of Basic legume nodule development and essential for rhizobial Research Activities for Innovative Biosciences, and fungal symbiosis, Proc. Natl Acad. Sci. USA, 103, (BRAIN), a grant in aid from the BBSRC (to J.A.D.), 359–364. and a John Innes Foundation studentship and an 15. Schauser, L., Roussis, A., Stiller, J., and Stougaard, J. Overseas Student Award (to H.M.). This work was also 1999, A plant regulator controlling development of symbiotic root nodules, Nature, 402, 191–195. supported in part by Grants-in-Aid for Scientific 16. Miwa, H., Sun, J., Oldroyd, G. E. D., and Downie, J. A. Research on Priority Areas from the Ministry of 2006, Analysis of Nod-factor-induced calcium signaling in Education, Culture, Sports, Science and Technology of root hairs of symbiotically defective mutants of Lotus Japan. japonicus, Mol. Plant-Microbe Interact., 19, 914–923. 17. Smit, P., Raedts, J., Portyanko, V., et al. 2005, NSP1 of the GRAS protein family is essential for REFERENCES rhizobial Nod factor-induced transcription, Science, 303, 1789–1791. 1. Oldroyd, G. E. D. and Downie, J. A. 2004, Calcium, 18. Kalo, P., Gleason, C., Edwards, A., et al. 2005, Nodulation kinases and nodulation signalling in legumes, Nat. Rev. signaling in legumes requires NSP2, a member of the GRAS Mol. Cell Biol., 5, 566–576. family of transcriptional regulators, Science, 303, 2. Catorina, R., Galera, C., de Billy, F., et al. 2000, Four 1786–1789. genes of Medicago truncatula controlling components of a 19. Thykjær, T., Stiller, J., Handberg, K., Jones, J., and Nod factor transduction pathway, Plant Cell, 12, Stougaard, J. 1995, The maize transposable element Ac is 1647–1665. mobile in the legume Lotus japonicus, Plant Mol. Biol., 27, 3. Kawaguchi, M., Imaizumi-Anraku, H., Koiwa, H., et al. 981–993. 2002, Root, root hair, and symbiotic mutants of the model 20. Kawaguchi, M. 2000, Lotus japonicus ‘Miyakojima’ MG-20: legume Lotus japonicus, Mol. Plant-Microbe Interact., 15, an early flowering accession suitable for indoor genetics, 17–26. J. Plant Res., 113, 507–509. 4. Szczyglowski, K., Shaw, R.S., Wopereis, J., et al. 1998, 21. Kawaguchi, M., Pedrosa-Harand, A., Yano, K., et al. Nodule organogenesis and symbiotic mutants of the model 2005, Lotus burttii takes a position of the third corner in legume Lotus japonicus, Mol. Plant-Microbe Interact., 11, the Lotus molecular genetics triangle, DNA Res., 12, 684–697. 69–77. 5. Rodutoiu, S., Madsen, L. H., Madsen, E. B., et al. 2003, 22. Kawasaki, S., Murakami, Y., Imaizumi-Anraku, H., Plant recognition of symbiotic bacteria requires two LysM Shimizu, A., and Mikami, I. 2003, Construction of high- receptor-like kinases, Nature, 425, 585–592. density map, genome library, and saturation mapping of 6. Madsen, L. H., Madsen, E. B., Rodutoiu, S., et al. 2003, nodulation genes, In: Nagata, and Tabata (eds) A receptor kinase gene of the LysM type is involved in Biotechnology in Agriculture and Forestry. Brassicas and legume perception of rhizobial signals, Nature, 425, Legumes, Vol. 52, pp. 183–202. 637–640. 23. Kawasaki, S. and Murakami, Y. 2000, Genome analysis of 7. Limpens, E., Franken, C., Smit, P., Willemse, J., Lotus japonicus, J. Plant Res., 113, 497–506. Bisseling, T., and Geurts, R. 2003, LysM domain receptor 24. Chiu, W., Niwa, Y., Zeng, W., Hirano, T., Kobayashi, H., kinases regulating rhizobial Nod factor-induced infection, and Sheen, J. 1996, Engineered GFP as a vital reporter in Science, 302, 630–633. plants, Curr. Biol., 6, 325–330. 8. Stracke, S., Kistner, C., Yoshida, S., et al. 2002, A plant 25. Diaz, C. L., Melchers, L. S., Hooykaas, P. J. J., receptor-like kinase required for both bacterial and fungal Lugtenberg, B. J. J., and Kijne, J. W. 1989, Root lectin symbiosis, Nature, 417, 959–962. Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 No. 6] Y. Murakami et al. 265 as a determinant of host-plant specificity in the Rhizobium- 34. Tian, C., Wan, P., Sun, S., Li, J., and Chen, M. 2004, legume symbiosis, Nature, 338, 579–581. Genome-wide analysis of the GRAS gene family in rice and 26. Kumagai, H. and Kouchi, H. 2003, Gene silencing by Arabidopsis, Plant Mol. Biol., 54, 519–532. expression of Hairpin RNA in Lotus japonicus roots and 35. Richards, D. E., Peng, J., and Harberd, N. 2000, Plant root nodules, Mol. Plant-Microbe Interact., 16, 663–668. GRAS and metazoan STATs: only family? BioEssays, 22, 27. Murray, M. G. and Thompson, W. F. 1980, Rapid isolation 573–577. of high molecular weight plant DNA, Nucleic Acids Res., 8, 36. Kouchi, H. and Hata, S. 1993, Isolation and characteriza- 4321–4325. tion of novel noduline cDNAs representing genes expressed 28. Kouchi, H., Shimomura, K., Hata, S., et al. 2004, Large- at early stages of soybean nodule development, Mol. Gen. scale analysis of gene expression profiles during early stages Genet., 238, 106–119. of root nodule formation in a model legume, Lotus 37. Yang, W. C., Katinakis, P., Hendriks, P., et al. 1993, japonicus, DNA Res., 11, 263–274. Characterization of GmENOD40, a gene showing novel 29. Chiu, W., Niwa, Y., Zeng, W., Hirano, T., Kobayashi, H., patterns of cell-specific expression during soybean nodule and Sheen, J. 1996, Engineered GFP as a vital reporter in development, Plant J., 3, 573–585. plants, Curr. Biol., 6, 325–330. 38. Minami, E., Kouchi, H., Cohn, J. R., Ogawa, T., and 30. Senoo, K., Solaiman, M. Z., Kawaguchi, M., et al. 2000, Stancey, G. 1996, Expression of the early nodulin, Isolation of two different phenotypes of mycorrhizal ENOD40, in soybean roots in response to various lipo- mutants in the model legume plant Lotus japonicus after chitin signal molecules, Plant J., 10, 23–32. EMS-treatment, Plant Cell Physiol., 41, 726–732. 39. Kumagai, H., Kinoshita, E., Ridge, R. W., and Kouchi, H. 31. Sandal, N., Petersen, T. R., Murray, J., et al. 2006, 2006, RNAi knock-down of ENOD40s leads to significant Genetics of symbiosis in Lotus japonicus: recombinant suppression of nodule formation in Lotus japonicus, Plant inbred lines, comparative genetic maps, and map position of Cell Physiol., 47, 1102–1111. 35 symbiotic loci, Mol. Plant-Microbe Interact., 19, 80–92. 40. Charon, C., Sousa, C., Crespi, M., and Kondorosi, A. 1999, 32. Bolle, C. 2004, The role of GRAS proteins in plant signal Alteration of enod40 expression modifies Medicago trun- transduction and development, Planta, 218, 683–692. catula root nodule development induced by Sinorhizobium 33. Pysh, L., Wysocka-Diller, J. W., Camilleri, C., meliloti, Plant Cell, 11, 1953–1966. Bouchez, D., and Benfey, P. N. 1999, The GRAS gene 41. Heckmann, A. B., Lombardo, F., Miwa, H., et al. 2006, family in Arabidopsis: sequence characterization and basic Lotus japonicus nodulation required two GRAS domain expression analysis of the SCARECROW-LIKE genes, regulators, one of which is functionally conserved in a non- Plant J., 18, 111–119. legume, Plant Physiol., 142, 1739–1750. Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png DNA Research Oxford University Press

Positional Cloning Identifies Lotus japonicus NSP2, A Putative Transcription Factor of the GRAS Family, Required for NIN and ENOD40 Gene Expression in Nodule Initiation

Download PDF
12 pages

Loading next page...
 
/lp/oxford-university-press/positional-cloning-identifies-lotus-japonicus-nsp2-a-putative-0DSHFn1pM3

References (41)

Publisher
Oxford University Press
Copyright
Copyright © 2022 Kazusa DNA Research Institute
ISSN
1340-2838
eISSN
1756-1663
DOI
10.1093/dnares/dsl017
pmid
17244637
Publisher site
See Article on Publisher Site

Abstract

Rhizobia-secreted Nod-factors (NFs) are required for nodulation. In the early developmental process of nodulation, a large number of changes occur in gene expression. Lotus japonicus nsp2 mutants isolated from Gifu B-129 ecotype have defects in nodule initiation and display non-nodulating phenotype. Here, we describe positional cloning of LjNSP2 as a component of the nodulation-specific signaling pathway. LjNSP2 was mapped near the translocation site of chromosome 1 where the recombination is severely suppressed. To circumvent this problem, we introduced Lotus burttii as an alternative crossing partner in place of L. japonicus Miyakojima. The development of the high-resolution map using a total of 11 481 F2 plants, in combination with newly developed DNA markers and construction of BAC library, enabled us to identify the gene responsible for mutant phenotype. LjNSP2 encodes a putative transcription factor of the GRAS family that constitutes a subfamily with Medicago truncatula NSP2. LjNSP2 was expressed in roots and early nodules, but strongly suppressed in matured nodules. The expression analysis of NIN and LjENOD40-1 genes in Ljnsp2 mutants indicates that LjNSP2 functions upstream of these genes. These results suggest that LjNSP2 acts as a transcription factor to directly or indirectly switch on the NF-induced genes required for nodule initiation. Key words: positional cloning; Lotus japonicus; transcription factor; nodule initiation 1. Introduction by rhizobia are responsible for nodule formation and induce a variety of responses in a host-specific manner, The legume–rhizobia symbioses lead to the formation including root hair deformation and cortical cell division, 2þ of novel organs, termed nodules, which arise from division during the early steps of nodulation. Ca spiking in root of cortical cells in the root and the infection of these hair cells is one of the most early responses to NFs. nodules by rhizobia. Rhizobia within nodule cells differ- In the past few years, the phenotypes of an increasing entiate to bacteroids which fix atmospheric nitrogen. number of symbiotically defective mutants have been Lipochitin oligosaccharides, Nod-Factors (NFs) secreted analyzed in the model legumes, Lotus japonicus and Medicago truncatula to dissect the NF signaling 2–4 Communicated by Satoshi Tabata pathway. Several genes required for nodulation have * To whom correspondence should be addressed. Tel./Fax: þ81-3- been identified by positional cloning. L. japonicus 5841-4458, E-mail: [email protected] 5 6 7 NFR1, NFR5 and Medicago LYK3 encode trans- y Present address: Biotechnology Research Center, The University membrane receptor-like serine/threonine kinases with of Tokyo, 1-1-1 Yayoi, Bunkyo, Tokyo 113-8657, Japan The Author 2006. Kazusa DNA Research Institute. The online version of this article has been published under an open access model. Users are entitled to use, reproduce, disseminate, or display the open access version of this article for non-commercial purposes provided that: the original authorship is properly and fully attributed; the Journal and Oxford University Press are attributed as the original place of publication with the correct citation details given; if an article is subsequently reproduced or disseminated not in its entirety but only in part or as a derivative work this must be clearly indicated. For commercial re-use, please contact [email protected] Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 256 Identification of Lotus japonicus NSP2 [Vol. 13, putative extracellular regions similar to LysM domains, the addition of 10 M NF. Representative traces were which are thought to be directly involved in perception selected from at least 10 independent cells. of NF signal. In subsequent signal transduction, 8 9 2.3. Genetic mapping population and L. japonicus SYMRK, CASTOR and POLLUX, 10 11 12 genomic DNA isolation M. truncatula DMI1, DMI2 and DMI3 , and 13 14 L. japonicus CCaMK and Nup133 have been identi- Genetic mapping of the LjNSP2 gene was performed fied as components required for the common symbiosis with F2/F3 population derived from cross between pathway shared between the fungal and bacterial Ljnsp2-1 (Gifu B-129) and the early flowering ecotype 20 21 endosymbiotic systems. The L. japonicus nin mutant Miyakojima MG-20 or Lotus burttii B-303. For AFLP showed normal mycorrhization and early responses analysis, genomic DNA was extracted from 0.1 g leaves following rhizobial inoculation, including root hair using a Qiagen DNeasy plant kit according to the 15 2þ 2þ 16 deformation, Ca influx and Ca spiking. NIN manufacture’s protocol. For PCR markers analysis, was cloned as a putative transcription factor gene and DNA was extracted from one young leaf in 100 mlof shown to be expressed during both early and late stages PEB (200 mM Tris–HCl pH 7.5, 250 mM NaCl, 25 mM of nodule development, suggesting that NIN controls EDTA, 0.5% SDS), precipitated with isopropanol, various developmental aspects of nodulation after the washed and dissolved in 100 ml of TE (pH 8.0). perception of rhizobia in root hairs. However, little is 2.4. High efficiency genome scanning (HEGS)/ known about what signal component relays the signal AFLP or PCR-based screening of from the common symbiotic pathway into the LjNSP2-inked markers nodulation-specific program in L. japonicus. Recently, GRAS family genes, nodulation signaling pathway Genomic DNA (100 ng) was digested with EcoRI 1 (NSP1) and NSP2, have been identified as putative and MseI, ligated to EcoRI- and MseI-adaptors and 2þ transcription factors functioning downstream of Ca preamplified by using EcoRI- and MseI-adaptor primers. 17,18 spiking and CCaMK in Medicago truncatula. Preamplified DNA was prepared at 0.05 mg/ml concentra- Here, we describe the positional cloning and charac- tion before amplification with selective primers. A bulked terization of nodulation-specific LjNSP2 gene encoding segregant analysis was performed to identify markers a plant-specific GRAS protein most similar to MtNSP2, linked to LjNSP2. Bulks were constructed from which might lead to the induction of expression of preamplified DNAs of 10 recessive (Nod ) or dominant genes required for rhizobial infection and early nodule (Nod ) homozygous F2 plants, 4096 selective primer development. combinations of EcoRIþ3/MseIþ3 were screened to identify markers present only in the bulk of dominant homozygotes. Subsequently, LjNSP2-linked HEGS/AFLP markers 2. Materials and methods were excised directly from polyacrylamide gels and cloned with the TOPO TA Cloning Kit (Invitrogen). 2.1. Plant materials The cloned markers were sequenced and primers were The mutant carrying Ljnsp2-1 was isolated from the designed by using the software Primer 3 (Whitehead EMS-mutagenesis experiments of L. japonicus Gifu Institute, Cambridge, MA). SCAR (sequence charac- B-129. The mutant Ljsym35 was isolated from a terized amplified region) markers, which revealed poly- population of Gifu derived from a transposon-/ morphism between L. japonicus accession Gifu and DNA-tagging trial, as a non-tagged culture mutant, Miyakojima MG-20 or L. burttii, were analyzed in 2989 and was kindly provided by Prof. Jens Stougaard or 8472 F2/3 individuals from Miyakojima MG-20 or (Arrhus University, Denmark). L. burttii, respectively. To facilitate the efficiency of electrophoresis of AFLP 2.2. Root hair deformation and assays of or PCR products, the HEGS system was adapted. In 2þ Ca spiking this system, a set of electrophoresis apparatus is Seeds of L. japonicus were germinated and grown on equipped with two sets of 24.5 · 26.5 cm glass plates, BNM agar medium essentially as described previously, each accommodating a gel with 100 lanes and analysis except that the roots were grown between two filter of 400 samples is practicable in single run. After papers (grade 0860; Schleicher and Schull, UK), one of electrophoresis, the gels were stained with Vistra Green which was on the agar surface. Root hair deformation (Amersham Biosciences) and scanned by fluorescent gel was scored as described previously following 16 h scanner (FluorImager 595; Amersham Biosciences). exposure to 10 M NF in 1 ml BNM medium in a 2.5. BAC contig development chamber on a microscope slide. Images were taken with a 2þ digital-camera attached to an inverted microscope. Ca Three-dimensional BAC DNA pools prepared from 9 23 spiking was assayed as described previously, following our BAC library were screened with LjNSP2-linked Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 No. 6] Y. Murakami et al. 257 HEGS/AFLP or SCAR markers as described above. scanner (Storm840, Amersham Bioscience). After strip- HindIII-digested DNA from positive BACs was ping, the same filters were reprobed with the 400 bp fractionated on an agarose gel for fingerprinting and fragment from L. japonicus ubiquitin cDNA as a loading determination of overlaps. BAC ends were sequenced control. using M13 reverse or forward primer and non-repetitive 2.8. Transient expression of the LjNSP2-GFP fusion sequences in BAC were used for chromosome walking to protein in onion epidermal cells screen 3-D BAC pools with PCR or AFLP. The process The 1.5 kb ORF of LjNSP2 was amplified from 188C5 was repeated as needed to complete the BAC contig. BAC DNA using the primers: [5 -ACGCGTCGACA- Polymorphic PCR fragments in BAC sequences were TGGAAATGGATATAGATTGCATCC-3 (SalI-site analyzed to directly score recombinants in F2/F3 underlined), 5 -CATGTCATGAATGCACAATCTGA- populations. TTCTGAAGAAC-3 (BspHI-site underlined)], digested BAC clone 188C5 containing the LjNSP2 gene was with SalIand BspHI, and cloned into pUC18- shotgun-sequenced and then annotated by Rice-GAAS CaMV35SV-sGFP(s65T)-NOS plasmid at the SalI/ (Rice Genome Automated Annotation System) (http:// NcoI sites just upstream of GFP gene. Onion epidermal ricegaas.dna.affrc.go.jp/). cells were bombarded with DNA-coated particles using a Biolistic PDS-1000/He Particle Delivery System (Bio- 2.6. Complementation experiments Rad). About 18–24 h after bombardment, the cells were For complementation, a 6.9 kb PstI fragment, carrying observed using a Bio-Rad Radiance2000 confocal laser only the wild-type LjNSP2 1500 bp ORF and 4486 and scanning microscope. 899 bp of upstream and downstream sequence, respec- The sequence data of LjNSP2 and genomic sequences tively, was cloned into the hairy root transformation (B-129 Gifu) of LjNSP2 have been deposited with the vector, which was made from pCAMBIA1300 by replac- DDBJ data library under accession numbers AB241456 ing the hygromycin-resistant gene with sGFP(S65T). and AB241457, respectively. The resulting LjNSP2 recombinant plasmid was intro- duced into A. rhizogenes LBA1334 by electroporation. 3. Results Hairy root transformation of Ljnsp2-1 mutant was per- formed as described. The plants with transgenic hairy 3.1. Early infection phenotypes of a Ljnsp2 mutant roots were grown in vermiculite pots and inoculated with Mesorhizobium loti TONO. GFP fluorescence and nodule The Ljnsp2-1 mutant of L. japonicus grew normally formations were confirmed 4 weeks after inoculation. in nitrogen-rich compost, and established a normal symbiosis with the mycorrhizal fungus, but did not form nodules under nitrogen limitation when inoculated 2.7. Southern and northern hybridizations with M. loti. The addition of M. loti to Ljnsp2-1 seedlings Genomic DNA was extracted from leaves using CTAB induced no phenotypes, such as root hair curling, infec- method from L. japonicus, and 2.5 mg DNA were tion thread formation and cortical cell division, typically digested by EcoRI, electrophoresed on 0.8% agarose gel seen in the wild-type (data not shown). However, NF did and blotted to nylon membrane (Biodyne A, Pall). induce swelling and branching in root hair tips, although Twelve-day-old plants were inoculated with M. loti at a reduced level compared with the wild-type (Fig. 1A TONO. Infected roots at 4 days post inoculation (dpi) and H). Cytological staining of M. loti expressing lacZ and nodules at 8–32 dpi were immediately frozen in revealed no infection foci and no infection thread forma- liquid nitrogen. Total RNA was isolated from flower, 2þ tion (data not shown). Intracellular Ca spiking, which shoot, root and nodule tissue, and 5 mg aliquots were is induced by NF and has been proposed to be integrated electrophoresed by denaturing agarose gel and blotted as 2þ by a Ca -calmodulin-dependent protein kinase required above. for activation of early nodulation gene expression and LjNSP2 probe (1322 bp) was amplified from 188C5 mycorrhization , was indistinguishable in the mutant BAC DNA with primers 5 -ACTTCCACCACCTCATC- from the wild-type (Fig. 1I). The observed induction 0 0 0 GAC-3 and 5 -ACAAGTCCAAAGGGATGCAG-3 , 2þ of Ca spiking and normal mycorrhization of the and labeled with P using a random primer labeling mutant indicates that the mutation in Ljnsp2 affects a kit (Takara). Hybridization was done at 63 C in Church nodulation-specific signaling component that is down- buffer [0.5 M sodium phosphate, pH 7.2, 7% (w/v) SDS, stream of the common pathway for mycorrhizal and 1 mM EDTA] and the filters were washed once in 2· rhizobial symbioses. SSC containing 0.5% SDS at room temperature for 10 min, and twice in 0.2· SSC, 0.1% SDS at 63 C for 3.2. High-resolution genetic mapping of LjNSP2 15 min. The hybridized membranes were then exposed for 3 days to phosphor imaging plates (Fuji, Tokyo, The mutant allele Ljnsp2-1 (Gifu B-129) was crossed Japan), which were then scanned by a phosphor imaging to the early flowering ecotype Miyakojima MG-20. The Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 258 Identification of Lotus japonicus NSP2 [Vol. 13, Figure 1. L. japonicus root hair responses induced by NF. (A–D) Root hair deformation induced with 10 M NF. Slight swelling and branching were observed also in Ljnsp2-1 mutant (C and D), although to a lesser extent than the wild-type (Gifu B-129; A and B). (E–H) Non-treated 2þ root hairs look similar between the mutant (G and H) and wild-type (E and F). Scale bar: 50 mm. (I) Analysis of Ca spiking in root hairs. Each trace is from a single root hair using seedlings of wild-type (Gifu B-129), or the Ljnsp2 mutant. After about 20 min from injection of the 2þ 8 Ca -sensitive dye Oregon green-dextran to the root hairs, NF was added to the medium at 10 M. Both wild-type (upper) and the mutant (lower) responded positively for the addition of NF in 10–20 min and similar effects were observed with at least 10 independent seedlings. LjNSP2 locus was mapped near translocation site of the pools with AFLP system. Finally, we constructed a short arm of chromosome 1. Since the recombination is 2 Mb physical BAC contig that spans the LjNSP2 locus significantly suppressed in the chromosomal segment, (Fig. 2B). total of 4096 EcoRI/MseI primer combinations were exa- mined for bulked segregant analysis on bulks 10 Nod 3.4. The narrowing of LjNSP2 genomic region using recessive and Nod dominant homozygous F2 plants in L. burttii as an alternative crossing partner the cross of Ljnsp2-1 and Miyakojima MG-20. LjNSP2- and identification of LjNSP2 gene linked AFLP markers from this screen were further analyzed in additional 2989 F2 plants with HEGS/AFLP As no more recombination was found from the system. Nineteen markers are located on the southern population of this cross of Ljnsp2-1 · Miyakojima side of LjNSP2 locus at distances of 0.07 cM and seven MG-20, even with new markers developed from the markers on the northern side of LjNSP2 locus at dis- BAC clones in the contig, we made another cross with tances of 0.45 cM, respectively, while nine markers L. burttii. Among 8 472 F2 progenies from this cross, cosegregated with LjNSP2 locus (Table 1, Fig. 2A). six recombination events were found between the The screened markers were converted into the 11 flanking markers S26d and 183R derived from the BAC polymorphic SCAR markers (Table 2). clone 188C5; this located the mutation within a 130 kb region (Fig. 2B). Among the 5 ORFs (Fig. 2C) predicted from the sequence in this region (excluding transposable 3.3. Construction of a 2 Mb physical BAC contig elements), only one was identified as having a mutation spanning LjNSP2 in the mutant. A 6.9 kb fragment including this ORF At the first step, all SCAR markers were used to complemented the mutant for nodulation in hairy roots screen 3-D BAC DNA pools prepared from our transformed by the Agrobacterium rhizogenes carrying L. japonicus Gifu BAC library. The contigs containing the cloned region (Fig. 3). No nodules were formed using these SCAR markers were extended by the chromosome empty vector, confirming that mutation of this gene walking with the screening of 3-D BAC DNA pool using caused the mutant phenotype. PCR primer combinations based on end sequences of The ORF in this region corresponding to LjNSP2 BAC clones. In several cases, however, we could not encodes a protein belonging to the plant GRAS family 32–34 obtain non-repetitive PCR fragments from BAC end of putative transcription factors and analysis of sequences and screen BAC library. As an alternative the major plant GRAS family protein sequences indicated strategy, AFLP fragments from BAC with EcoRIþ1 that the closest in sequence to LjNSP2 were the selective and MseIþ1 selective primers were searched for M. truncatula and Pisum sativum NSP2 proteins. non-repetitive sequences of BAC inner. Subsequently, AtSCL26 of Arabidopsis was about twice distant from the corresponding EcoRIþ3 selective and MseIþ3 them, and together these proteins made an apparent selective primers were used to screen 3-D BAC DNA subfamily of GRAS proteins (Fig. 4). However, genome Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 No. 6] Y. Murakami et al. 259 Table 1. AFLP markers linked to LjNSP2 locus DNA-binding sequence, second leucine heptad repeat (LHRII); and a Src-homology 2 (SH2)-like domain AFLP marker Primer Approximate Marker type (Fig. 5A). About 15% of the N-terminal region is diver- combination length of the AFLP gent among most GRAS family members but that of fragment (bp) LjNSP2 showed strong homology with the M. truncatula EM117 E-CGA/M-GTG 700 Dominant and Pisum sativam NSP2 proteins suggesting functio- EM140 E-CTT/M-CAA 700 Co-dominant nal equivalence and that they are probably orthologs. EM157 E-GAC/M-GCC 130 Dominant Furthermore, LjNSP2 has a well-conserved SH2-like EM177 E-GCA/M-CAA 550 Dominant domain among GRAS proteins (Fig. 5B). The Ljnsp2-1 EM205 E-GGC/M-TGG 1500 Dominant mutation causes a substitution of a conserved valine (V) EM242 E-TAG/M-CTC 130 Dominant to glutamate (E) in this SH2-like domain (Fig. 5C). EM262 E-TCT/M-ACT 500 Co-dominant Complementation tests with the various L. japonicus nodulation mutants carrying mutations mapped to EM265 E-TCT/M-GGC 250 Co-dominant linkage group 1 revealed that Ljsym35 carried a mutation EM299 E-TTA/M-TGT 250 Dominant allelic to Ljnsp2. DNA hybridizations detected no signal EM341 E-AAT/M-GTT 500 Co-dominant in Ljsym35 (Fig. 6A) and PCR analyses indicated that it EM367 E-ACG/M-GGA 400 Dominant has a deletion of >100 kb around the LjNSP2 gene; this EM390 E-AGC/M-AAA 1000 Co-dominant allele was renamed Ljnsp2-2. EM480 E-CCA/M-GCA 190 Dominant EM527 E-CGT/M-CTC 70 Dominant 3.5. Expression of LjNSP2 during nodulation and in EM582 E-GAC/M-TGG 800 Dominant different organs EM603 E-GCA/M-TTA 800 Dominant RNA hybridization showed that LjNSP2 expression EM713 E-TAC/M-AAA 400 Dominant was detectable in roots but not in shoots and flowers EM784 E-TCT/M-TTC 100 Dominant (Fig. 6B), in contrast to the ubiquitously expressed Medicago NSP2. LjNSP2 was also expressed in infected EM807 E-TGG/M-CGA 90 Dominant roots at 4 dpi and early nodules but strongly suppressed EM011 E-AAG/M-TTC 200 Dominant in matured nodules (Fig. 6C). These expression patterns EM013 E-AAT/M-CCT 120 Dominant of LjNSP2 are similar to that of the putative Nod factor EM116 E-CGA/M-GGA 500 Dominant 5 6 receptor kinase genes, NFR1 and NFR5, and the signal EM160 E-GAG/M-AAT 250 Co-dominant transduction components CASTOR, although the decre- EM190 E-GCT/M-CAA 350 Co-dominant ase in expression of the latter was slight. This contrasts EM261 E-TCG/M-TTG 250 Co-dominant with the expression patterns of the genes encoding the EM283 E-TGG/M-CTG 500 Dominant nodulation signaling pathway components M. truncatula 17 18 NSP1, M. truncatula NSP2 and L. japonicus EM350 E-ACA/M-TCA 250 Dominant NIN (Fig. 6A) whose expression increased following EM466 E-CAT/M-ACT 800 Dominant inoculation. EM130 E-CGT/M-TGG 600 Dominant EM686 E-GTC/M-GAT 600 Dominant 3.6. Expression of NIN and LjENOD40-1 genes in EM163 E-GAG/M-GTC 300 Co-dominant Ljnsp2-2 mutant EM115 E-CGA/M-AGT 400 Co-dominant In the Ljnsp2-2 null mutant, RT–PCR revealed that EM214 E-GGG/M-TTA 850 Dominant even before inoculation with M. loti the expression levels EM231 E-GTT/M-TCT 80 Dominant of early nodulation genes, NIN and LjENOD40-1, were EM802 E-TGG/M-AAT 200 Dominant both <40% of the wild-type (Fig. 7A and B). Upon inoculation of the wild-type with M. loti, NIN and 36 LjNSP2-linked AFLP markers were selected from 3009 F2/F3 plants in the cross of Ljnsp2-1 (Gifu B-129) and LjENOD40-1 increased by up to 65- or 3.5-fold, respec- Miyakojima MG-20. tively. In contrast, the levels of NIN and LjENOD40-1 AFLP fragments were derived from Miyakojima MG-20. transcripts in the Ljnsp2-2 mutant remained low after inoculation reaching only about 5–20% of wild-type levels (Fig. 7A and B). This demonstrates that LjNSP2 sequences around the LjNSP2 and M. truncatula NSP2 function is required, directly or indirectly, for either the revealed no clear co-linearity based on the available data. expression and/or the induction of these early nodulins. The LjNSP2 gene consists of a 1500 bp exon with no introns, encoding a predicted 499 amino acids protein of 3.7. Nuclear localization of LjNSP2 in onion cells 55 kDa, containing the following GRAS family-specific domains: homopolymeric stretches (HPS) of polyE and Although most GRAS family proteins have a puta- polyT; first leucine heptad repeat (LHRI); a VHIID tive nuclear localization sequence (NLS) , PSORT II Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 260 Identification of Lotus japonicus NSP2 [Vol. 13, Figure 2. Positional cloning of the LjNSP2 gene (A) Genetic map of LjNSP2 region. This map is based on the analysis of 3 009 F2 plant of a cross between Ljnsp2-1 and Miyakojima MG-20. These HEGS/AFLP makers were selected from 4 096 of AFLP primer combinations with HEGS system. The markers conversed into polymorphic SCAR markers are underlined. (B) About 2 Mb of BAC contig around LjNSP2 genomic region. The numbers below the markers indicates recombination events detected in 16 944 chromosomes from 8 472 siblings of the heterozygotes from crosses of Ljnsp2-1 · L. burttii . Only overlapping BAC clones spanning LjNSP2 are displayed in this figure. Finally, LjNSP2 region was delimited within a 130 kb region on a 211 kb BAC clone (188C5). (C) Putative genes in LjNSP2 region. ORFs of putative genes (closed arrows) and retro-elements (RE; open arrows) are marked in this 130 kb region. Predicted genes are GTK, glutamine transferase K; P4P5K, phosphatidilinositol-4-phosphate-5-kinase; UP, unknown proteins. There were no introns in the 1 500 bp of the predicted LjNSP2 ORF. analysis did not identify an NLS-like sequence in containing a 5.1 kb fragment of the 5 flanking sequence LjNSP2. However, LjNSP2-GFP fusion delivered into was constructed and introduced into Ljnsp2-2 mutants. onion (Allium cepa) epidermal cells by particle bombard- These constructs complemented the Ljnsp2-2 mutant ment revealed it to be exclusively localized in nuclei, phenotype, indicating that this fusion protein retained although not in nucleoli (Fig. 8A). The mutation in the the LjNSP2 activity for nodulation. However, the SH-2-like domain reduced nuclear localization of Ljnsp2- transgenic roots showed no detectable GFP fluorescence 1-GFP and fluorescence in the cytoplasm became (data not shown). noticeable in at least 30 GFP-expressed cells (Fig. 8B), but not as strong as seen with GFP alone (Fig. 8C). 4. Discussion These observations are consistent with LjNSP2 acting as a transcription factor, and the SH2-like domain So far the only described nodulation-specific mutants facilitating its nuclear localization. of L. japonicus, which completely lack nodules but 5 6 The above LjNSP2-GFP fusion under the control of have normal mycorrhization, are Ljnsp2, nfr1, nfr5 the CaMV 35S promoter or the LjNSP2 promoter and nin. One key difference among these mutants is the Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 No. 6] Y. Murakami et al. 261 Table 2. PCR-based markers linked to LjNSP2 locus 0 0 PCR marker Origin Sequence (5 !3 ) Anealing Length (bp) Marker type temperature ( C) SEM140 EM140 GAATTCCTTCCCGGTTCTTC 65 750 Dominant (M) TTAACAACAGCAACTTCACT SEM299 EM299 GAATTCTTACCGAGTGAGAT 60 250 Dominant (M, b) TTAATGTTTCTGACAAAGGC SEM341 EM341 GAATTCAATGATTGTTGGGA 50 500 Co-dominant (M, b) TTAAGTTGTTCGAATTGATA SEM390 EM390 GAATTCAGCCACAGCCTCTG 50 230 Dominant (M) AACCACAAGTCAAATGCAAC SEM160 EM160 GAATTCGAGAATTGGAGAAG 60 220 Dominant (M) TTAAAATCAAACCCTAACAA SEM190 EM190 GAATTCGCTTGCAAGTAGTG 50 340 Co-dominant (M) TTAACAAGAGTCTCATAATC SEM261 EM261 TTCTCGGCGGTTCCACCAAT 68 290 Dominant (M) GCTTGTTTGAAAGTGGCTTT Co-dominant (b) SEM115 EM115 CCGATTACATTTATGAGTAC 50 350 Co-dominant (M, b) TTAAAGTCGATAGACCGACG SEM163 EM163 GAATTCGAGATCCAGAATCT 60 290 Dominant (M) TTAAGTCTGAACCATTGCTA SEM214 EM214 GAATTCGGGGTGTTACACCTC 65 580 Dominant (M) TTAATTACCCCCTCGTTCTC SEM802 EM802 GAATTCTGGAAGCACGTTGA 60 190 Co-dominant (M) TTAAAATCATCATCTGAGAG s26d 188C5 TCATGTTTAGCGCTCTGATT 60 100 Co-dominant (M, b) GCACCGTAGTACCGTCAGTTTCC S11 188C5 TTGCAGGTTTTCAGGAACATT 60 520 Co-dominant (M, b) GAAGTTCATTGTAACTCTCAAACG 183R 183A2 CATCGTATGTTGAAAAGAAGAATGAT 60 580 Co-dominant (M, b) TCTTCAGTTCTTCCCTTTATAGCC PCR was performed using the indicated primer pairs with the following cycling parameter: 4 min at 94 C; 35 cycles of 30 s at 94 C, 30 s at the temperature indicated, 1 min at 72 C; 10 min at 72 C; and hold at 4 C. PCR fragments were electrophoresed on 13% acrylamide gels with HEGS system. PCR fragments are derived from Miyakojima MG-20. M and b indicates the polymorphism of Miyakojima MG-20 and L. burttii, respectively, compared with Gifu B-129. 2þ induction of Ca spiking, which might be required for even in the population of 3 009 F2/F3 plants of the the early common response in activation of the symbioses cross with Miyakojima MG-20. This corresponds to with both rhizobia and mycorrhizal fungi. The Ljnsp2 24 Mb/cM, 78-fold greater than the average physical 2þ mutants are normal for Ca spiking but the nfr1 and to genetic distance found in our high-density map of the nfr5 mutants are blocked. Recently, NFR1 and NFR5 L. japonicus (Wang et al., manuscript in preparation). were shown to encode LysM-receptor-like-kinases, that This indicates that recombination in the chromosomal 5,6 were predicted to function in NF perception. NIN segment near the translocation site is highly suppressed functions downstream of LjNSP2 as indicated in this in the cross combination of Gifu B-129 · Miyakojima study. It is possible that LjNSP2 is earliest known MG-20. Therefore, we changed the crossing partner from protein executing nodulation-specific gene expression Miyakojima MG-20 to L. burttii. As a result, LjNSP2 2þ from Ca spiking induced through NFR1 and NFR5 in region narrowed up to 130 kb in one BAC clone (188C5) L. japonicus. using the population of 8 472 F2 plants of the cross with We have established a 2 Mb physical BAC contig L. burttii, indicating that L. burttii is significantly useful that spans the LjNSP2 locus (Fig. 2B). However, LjNSP2 as alternative crossing partner of Gifu especially near the region was only closed to the minimum of 14 BAC clones translocation site of chromosome 1. Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 262 Identification of Lotus japonicus NSP2 [Vol. 13, LjNSP2 encodes a GRAS family protein of a putative transcription factor. The phylogenetic analysis revealed that LjNSP2 is most closely related to M. truncatula NSP2 and P. sativum NSP2. The comparison of the Figure 4. Phylogenetic tree of plant GRAS family proteins. Members of the family of plant GRAS proteins aligned using Clustal W are Figure 3. Complementation of Ljnsp2-1 mutant. Roots of the Ljnsp2 shown as a neighbor-joining dendrogram with 1000 bootstrap mutant were transformed by Agrobacterium rhizogenes strain replicates. At, Arabidopsis thaliana; Mt, Medicago truncatula; Ph, LBA1334 carrying appropriate plasmids carrying the 1 500 bp Petunia hybrida; Ps, Pisum sativum; Le, Lycopersicum esculentum; LjNSP2 ORF and 4 486 and 899 bp of upstream and downstream NSP, nodulation signaling pathway; SCR, SCARECROW; SCL, sequences and a transformation indicator 35S-GFP. On inoculation SCARECROW-like; SHR, short-root; HAM, hairy meristem; Ls, with Mesorhizobium loti, the GFP-marked transformants formed lateral suppressor; RGA, repressor of ga1-3; PAT, phytochrome A nodules complementing the mutant phenotype. Scale bar: 1 mm. signal transduction. Figure 5. Domain structure of the LjNSP2 protein. (A) Amino acid sequence of 499 residues and predicted functional domains of LjNSP2. HPS; homopolymeric stretches characteristic for GRAS protein near to the N-terminal, LHR (leucine heptad repeat) 1 and 2; putative leucine zipper, VHIID; putative DNA-binding sites, SH2-like region, GY; GY, Y of which is phosphorylated in STATs, is conserved about 100 residues downstream from SH2-like domain. (B) Comparison of GRAS and STAT family proteins with LjNSP2 in their SH2(-like) domains. STAT2: human STAT2; P52630, Ce-STAT: Caenorhabditis elegans STAT; Z70754, D-STAT: Drosophila stat; Q24151, Dd-STAT: Dictyostelium discoideum STAT; Y13097, GAI: Arabidopsis GAI (gibberellin insensitive); At1g14920, SCR: Arabidopsis SCR (SCARECROW); At3g54220. GAI and SCR are the representatives of plant GRAS family proteins. Conserved amino acids are indicated by red boxes. A missense mutation in Ljnsp2-1 is indicated by arrowhead. Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 No. 6] Y. Murakami et al. 263 AC B Figure 8. Nuclear localization of the LjNSP2-GFP. LjNSP2-GFP (A), Ljnsp2-1-GFP (B) fusion genes and control GFP gene (C), delivered by particle bombardment, were transiently expressed in onion epidermal cells and observed with a laser confocal microscope 24 h after bombardment. The nuclear localization of the Ljnsp2-1 mutant gene product is significantly suppressed compared with LjNSP2. Scale: 10 mm. after rhizobial inoculation. These results suggest that, in contrast with a possible role for M. truncatula NSP2 in another organ development in addition to nodulation, LjNSP2 seems to be specialized in nodule initiation. Figure 6. DNA, RNA hybridization and RT–PCR analyses of LjNSP2. (A) Genomic hybridization with the full-length LjNSP2 The expression of NIN and ENOD40 is induced rapidly ORF probe to the wild-type (WT; Gifu B-129) and a null deletion after rhizobial inoculation and NF-treatment. The NIN mutant Ljnsp2-2. Genomic DNA was digested with EcoRI. The transcripts are detected in different tissues during LjNSP2 homolog seems to be unique in WT, but cannot be detected various nodule stages, such as the dividing cells of in Ljnsp2-2. (B and C) Northern analysis of LjNSP2 expression in various organs (B), and infected roots at 4 dpi of M. loti TONO and the nodule primordia and the nodule vascular bundles. nodules at 8–32 dpi of M. loti TONO, respectively (C). DPI: days The Ljnin mutants are nodulation-minus and blocked in post inoculation. both the infection thread formation and the cortical cell division. ENOD40 is induced in the root pericycle a few hours after rhizobial inoculation, and subsequently in the dividing cortical cells of the root and nodule 36–38 39 primordia. ENOD40 RNAi knock-down lines or the possible co-suppression suppressed nodule primor- dium formation, resulting in very poor nodulation. Despite of their function in nodule initiation, however, little is known about what kinds of transcription factors in the root activate NIN and ENOD40 gene expression in response to rhizobia. In this study, we demonstrate that the induction of NIN and LjENOD40-1 gene expression is clearly cancelled by the Ljnsp2 mutation (Fig. 7). On the basis of these findings, we speculate that LjNSP2 may function as a transcriptional activator to Figure 7. RT–PCR analyses of expression of LjNSP2 and the early directly or indirectly switch on the NIN and LjENOD40- nodulins. (A and B) RT–PCR analysis of the change of expression of 1 gene expression in nodule initiation. In order to address NIN (A) and LjENOD40-1 (B) after M. loti inoculation, in wild-type this issue, identification of the promoter region and and Ljnsp2. Suppression of LjNSP2 expression after inoculation was confirmed, and almost complete loss of induction in NIN and subsequent binding assay using LjNSP2 protein would be LjENOD40-1 were apparent. Ubiquitin is as a loading control. of great importance in future. Relative expression levels were normalized against the amount of Kalo´ et al. reported that M. truncatula NSP2-GFP ubiquitin (set as 10 ). Standard deviations of three independent localizes in the endoplasmic reticulum and nuclear experiments are indicated by error bars. DPI: days post inoculation. envelope and re-localizes into the nucleus rapidly after NF-treatment. In this case, they made a functional mutant phenotypes suggests that the LjNSP2, MtNSP2 C-terminal GFP fusion under control of the constitutive and PsNSP2 genes function at similar or parallel CaMV 35S promoter that was introduced in Medicago positions in the nodulation signal transduction. nsp2 mutant plants. Here, we attempted to detect However, the expression pattern is different between LjNSP2-GFP, -YFP or -DsRED2 fusion in L. japonicus LjNSP2 and M. truncatula NSP2. LjNSP2 is predomi- hairy roots. These fusions could complement the mut- nantly expressed in roots and its expression decreases in ant phenotype but no fluorescence was detected even developed nodules. The expression of M. truncatula under the control of CaMV35S promoter. In place of NSP2 is observed in shoots as well as roots and is induced L. japonicus we delivered LjNSP2-GFP fusion into onion Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 264 Identification of Lotus japonicus NSP2 [Vol. 13, 9. Imaizumi-Anraku, H., Takeda, N., Charpentier, M., et al. epidermal cells by particle bombardment. The fusion 2005, Plastid proteins crucial for symbiotic fungal and exclusively localizes in the nucleus but not in the nuc- bacterial entry into plant roots, Nature, 433, 527–531. lear envelope. Although there is no way to explain the 10. Anee, _ J. M., Kiss, G. B., Riely, B. K., et al. 2004, Medicago peculiar difference of subcellular localization of NSP2 truncatula DMI1 required for bacterial and fungal sym- between Medicago and onion, putative nucleoporins bioses in legumes, Science, 303, 1364–1367. such as NUP133 and NUP85 (Saito et al. unpublished 11. Endre, G., Kereszt, A., Kevei, Z., Mihacea, S., Kalo´, P., data) required for rhizobial and arbuscular mycorrhizal and Kiss, G. B. 2002, A receptor kinase gene regulating symbioses may retain the NSP2 putative transcription symbiotic nodule development, Nature, 417, 962–966. factor in the nuclear envelope of M. truncatula.Most 12. Leevy, _ J., Bres, C., Geurts, R., et al. 2004, A putative 2þ recently, Heckmann et al. reported that non-nodulating Ca and Calmodulin-dependent protein kinase required mutant, SL781-3, carrying an allele of LjNSP2 was found for bacterial and fungal symbiosis, Science, 303, 1361–1364. by TILLING. 13. Tirichine, L., Imaizumi-Anraku, H., Yoshida, S., et al. Acknowledgements: We thank Prof. Jens 2þ 2006, Deregulation of a Ca /calmodulin-dependent kinase Stougaard for kind providing the mutant seeds of leads to spontaneous nodule development, Nature, 441, Ljsym35. We wish to thank Krzyztof Szczyglowski and 1153–1156. Martin Parniske for valuable comments on the manu- 14. Kanamori, N., Madsen, L. H., Radutoiu, S., et al. 2006, A script and Anne Heckman for helpful discussion. This 2þ nucleoporin is required for induction of Ca spiking in work was supported by the fund of Promotion of Basic legume nodule development and essential for rhizobial Research Activities for Innovative Biosciences, and fungal symbiosis, Proc. Natl Acad. Sci. USA, 103, (BRAIN), a grant in aid from the BBSRC (to J.A.D.), 359–364. and a John Innes Foundation studentship and an 15. Schauser, L., Roussis, A., Stiller, J., and Stougaard, J. Overseas Student Award (to H.M.). This work was also 1999, A plant regulator controlling development of symbiotic root nodules, Nature, 402, 191–195. supported in part by Grants-in-Aid for Scientific 16. Miwa, H., Sun, J., Oldroyd, G. E. D., and Downie, J. A. Research on Priority Areas from the Ministry of 2006, Analysis of Nod-factor-induced calcium signaling in Education, Culture, Sports, Science and Technology of root hairs of symbiotically defective mutants of Lotus Japan. japonicus, Mol. Plant-Microbe Interact., 19, 914–923. 17. Smit, P., Raedts, J., Portyanko, V., et al. 2005, NSP1 of the GRAS protein family is essential for REFERENCES rhizobial Nod factor-induced transcription, Science, 303, 1789–1791. 1. Oldroyd, G. E. D. and Downie, J. A. 2004, Calcium, 18. Kalo, P., Gleason, C., Edwards, A., et al. 2005, Nodulation kinases and nodulation signalling in legumes, Nat. Rev. signaling in legumes requires NSP2, a member of the GRAS Mol. Cell Biol., 5, 566–576. family of transcriptional regulators, Science, 303, 2. Catorina, R., Galera, C., de Billy, F., et al. 2000, Four 1786–1789. genes of Medicago truncatula controlling components of a 19. Thykjær, T., Stiller, J., Handberg, K., Jones, J., and Nod factor transduction pathway, Plant Cell, 12, Stougaard, J. 1995, The maize transposable element Ac is 1647–1665. mobile in the legume Lotus japonicus, Plant Mol. Biol., 27, 3. Kawaguchi, M., Imaizumi-Anraku, H., Koiwa, H., et al. 981–993. 2002, Root, root hair, and symbiotic mutants of the model 20. Kawaguchi, M. 2000, Lotus japonicus ‘Miyakojima’ MG-20: legume Lotus japonicus, Mol. Plant-Microbe Interact., 15, an early flowering accession suitable for indoor genetics, 17–26. J. Plant Res., 113, 507–509. 4. Szczyglowski, K., Shaw, R.S., Wopereis, J., et al. 1998, 21. Kawaguchi, M., Pedrosa-Harand, A., Yano, K., et al. Nodule organogenesis and symbiotic mutants of the model 2005, Lotus burttii takes a position of the third corner in legume Lotus japonicus, Mol. Plant-Microbe Interact., 11, the Lotus molecular genetics triangle, DNA Res., 12, 684–697. 69–77. 5. Rodutoiu, S., Madsen, L. H., Madsen, E. B., et al. 2003, 22. Kawasaki, S., Murakami, Y., Imaizumi-Anraku, H., Plant recognition of symbiotic bacteria requires two LysM Shimizu, A., and Mikami, I. 2003, Construction of high- receptor-like kinases, Nature, 425, 585–592. density map, genome library, and saturation mapping of 6. Madsen, L. H., Madsen, E. B., Rodutoiu, S., et al. 2003, nodulation genes, In: Nagata, and Tabata (eds) A receptor kinase gene of the LysM type is involved in Biotechnology in Agriculture and Forestry. Brassicas and legume perception of rhizobial signals, Nature, 425, Legumes, Vol. 52, pp. 183–202. 637–640. 23. Kawasaki, S. and Murakami, Y. 2000, Genome analysis of 7. Limpens, E., Franken, C., Smit, P., Willemse, J., Lotus japonicus, J. Plant Res., 113, 497–506. Bisseling, T., and Geurts, R. 2003, LysM domain receptor 24. Chiu, W., Niwa, Y., Zeng, W., Hirano, T., Kobayashi, H., kinases regulating rhizobial Nod factor-induced infection, and Sheen, J. 1996, Engineered GFP as a vital reporter in Science, 302, 630–633. plants, Curr. Biol., 6, 325–330. 8. Stracke, S., Kistner, C., Yoshida, S., et al. 2002, A plant 25. Diaz, C. L., Melchers, L. S., Hooykaas, P. J. J., receptor-like kinase required for both bacterial and fungal Lugtenberg, B. J. J., and Kijne, J. W. 1989, Root lectin symbiosis, Nature, 417, 959–962. Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022 No. 6] Y. Murakami et al. 265 as a determinant of host-plant specificity in the Rhizobium- 34. Tian, C., Wan, P., Sun, S., Li, J., and Chen, M. 2004, legume symbiosis, Nature, 338, 579–581. Genome-wide analysis of the GRAS gene family in rice and 26. Kumagai, H. and Kouchi, H. 2003, Gene silencing by Arabidopsis, Plant Mol. Biol., 54, 519–532. expression of Hairpin RNA in Lotus japonicus roots and 35. Richards, D. E., Peng, J., and Harberd, N. 2000, Plant root nodules, Mol. Plant-Microbe Interact., 16, 663–668. GRAS and metazoan STATs: only family? BioEssays, 22, 27. Murray, M. G. and Thompson, W. F. 1980, Rapid isolation 573–577. of high molecular weight plant DNA, Nucleic Acids Res., 8, 36. Kouchi, H. and Hata, S. 1993, Isolation and characteriza- 4321–4325. tion of novel noduline cDNAs representing genes expressed 28. Kouchi, H., Shimomura, K., Hata, S., et al. 2004, Large- at early stages of soybean nodule development, Mol. Gen. scale analysis of gene expression profiles during early stages Genet., 238, 106–119. of root nodule formation in a model legume, Lotus 37. Yang, W. C., Katinakis, P., Hendriks, P., et al. 1993, japonicus, DNA Res., 11, 263–274. Characterization of GmENOD40, a gene showing novel 29. Chiu, W., Niwa, Y., Zeng, W., Hirano, T., Kobayashi, H., patterns of cell-specific expression during soybean nodule and Sheen, J. 1996, Engineered GFP as a vital reporter in development, Plant J., 3, 573–585. plants, Curr. Biol., 6, 325–330. 38. Minami, E., Kouchi, H., Cohn, J. R., Ogawa, T., and 30. Senoo, K., Solaiman, M. Z., Kawaguchi, M., et al. 2000, Stancey, G. 1996, Expression of the early nodulin, Isolation of two different phenotypes of mycorrhizal ENOD40, in soybean roots in response to various lipo- mutants in the model legume plant Lotus japonicus after chitin signal molecules, Plant J., 10, 23–32. EMS-treatment, Plant Cell Physiol., 41, 726–732. 39. Kumagai, H., Kinoshita, E., Ridge, R. W., and Kouchi, H. 31. Sandal, N., Petersen, T. R., Murray, J., et al. 2006, 2006, RNAi knock-down of ENOD40s leads to significant Genetics of symbiosis in Lotus japonicus: recombinant suppression of nodule formation in Lotus japonicus, Plant inbred lines, comparative genetic maps, and map position of Cell Physiol., 47, 1102–1111. 35 symbiotic loci, Mol. Plant-Microbe Interact., 19, 80–92. 40. Charon, C., Sousa, C., Crespi, M., and Kondorosi, A. 1999, 32. Bolle, C. 2004, The role of GRAS proteins in plant signal Alteration of enod40 expression modifies Medicago trun- transduction and development, Planta, 218, 683–692. catula root nodule development induced by Sinorhizobium 33. Pysh, L., Wysocka-Diller, J. W., Camilleri, C., meliloti, Plant Cell, 11, 1953–1966. Bouchez, D., and Benfey, P. N. 1999, The GRAS gene 41. Heckmann, A. B., Lombardo, F., Miwa, H., et al. 2006, family in Arabidopsis: sequence characterization and basic Lotus japonicus nodulation required two GRAS domain expression analysis of the SCARECROW-LIKE genes, regulators, one of which is functionally conserved in a non- Plant J., 18, 111–119. legume, Plant Physiol., 142, 1739–1750. Downloaded from https://academic.oup.com/dnaresearch/article/13/6/255/464686 by DeepDyve user on 24 June 2022

Journal

DNA Research – Oxford University Press

Published: Jan 1, 2006

There are no references for this article.