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

Learn More →

The NADPH-cytochrome P450 Reductase Gene from Gibberella fujikuroi Is Essential for Gibberellin Biosynthesis

The NADPH-cytochrome P450 Reductase Gene from Gibberella fujikuroi Is Essential for Gibberellin... THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 279, No. 24, Issue of June 11, pp. 25075–25084, 2004 © 2004 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. The NADPH-cytochrome P450 Reductase Gene from Gibberella fujikuroi Is Essential for Gibberellin Biosynthesis* Received for publication, August 4, 2003, and in revised form, March 16, 2004 Published, JBC Papers in Press, March 22, 2004, DOI 10.1074/jbc.M308517200 ¶ ¶ ¶ Stefan Malonek‡, Maria C. Rojas§, Peter Hedden , Paul Gaskin , Paul Hopkins , and Bettina Tudzynski‡ From the ‡Institut fu ¨ r Botanik der Westfa ¨ lischen Wilhelms-Universita ¨t Mu ¨ nster, Schlossgarten 3, D-48149 Mu ¨ nster, Germany, §Laboratorio de Bioorga ´ nica, Departamento de Quı ´mica, Facultad de Ciencias, Universidad de Chile, Casilla 653, Santiago, Chile, and Rothamsted Research, Harpenden, Herts AL5 2LQ, United Kingdom known as a rich source of gibberellins (GAs), which function as The fungus Gibberella fujikuroi is used for the com- mercial production of gibberellins (GAs), which it pro- hormones in higher plants. The major GA in most strains of duces in very large quantities. Four of the seven GA G. fujikuroi is gibberellic acid (GA ), the biosynthesis of which biosynthetic genes in this species encode cytochrome requires seven genes that are arranged in a gene cluster. Each P450 monooxygenases, which function in association of these genes and the function of the encoded enzymes have with NADPH-cytochrome P450 reductases (CPRs) that recently been fully characterized (3– 8). As well as genes encod- mediate the transfer of electrons from NADPH to the ing a pathway-specific geranylgeranyl diphosphate synthase P450 monooxygenases. Only one cpr gene (cpr-Gf) was (ggs2), ent-copalyldiphosphate/ent-kaurene synthase (cps/ks), found in G. fujikuroi and cloned by a PCR approach. The and GA desaturase (des), the cluster contains four cytochrome encoded protein contains the conserved CPR functional P450 monooxygenase genes (P450-1 to P450-4), which in most domains, including the FAD, FMN , and NADPH binding cases encode multifunctional enzymes that catalyze several motifs. cpr-Gf disruption mutants were viable but biosynthetic steps. showed a reduced growth rate. Furthermore, disruption About 40 different P450 monooxygenases involved in diverse resulted in total loss of GA ,GA , and GA production, 3 4 7 metabolic pathways have been identified in fungi (9). Many of but low levels of non-hydroxylated C -GAs (GA and 20 15 the P450 genes form part of gene clusters that are responsible GA ) were still detected. In addition, the knock-out mu- for the biosynthesis of metabolites, such as aflatoxins (10), tants were much more sensitive to benzoate than the trichothecenes (11), fumonisins (12), and paxillin (13), or for wild type due to loss of activity of another P450 mo- the metabolism of xenobiotics (14). Indeed a large group of the nooxygenase, the detoxifying enzyme, benzoate p-hy- fungal cytochrome P450s have been shown to be involved in the droxylase. The UV-induced mutant of G. fujikuroi, metabolism of drugs and other foreign compounds (14 –16), SG138, which was shown to be blocked at most of the GA whereas others participate in the biosynthesis of intracellular biosynthetic steps catalyzed by P450 monooxygenases, displayed the same phenotype. Sequence analysis of the compounds such as steroids (cited in Ref. 16). Metabolism of mutant cpr allele in SG138 revealed a nonsense muta- xenobiotics, including drugs and toxins, is also an important tion at amino acid position 627. The mutant was comple- function of P450 monooxygenases in animals as is the forma- mented with the cpr-Gf and the Aspergillus niger cprA tion of endogenous compounds such as sterols and fatty acids genes, both genes fully restoring the ability to produce (17). In plants, P450s are implicated in the formation of a broad GAs. Northern blot analysis revealed co-regulated ex- range of metabolites, including the growth hormones gibberel- pression of the cpr-Gf gene and the GA biosynthetic lins and jasmonates, essential components such as lignin, pig- genes P450-1, P450-2, P450-4 under GA production con- ments, and fatty acids, and secondary metabolites such as ditions (nitrogen starvation). In addition, expression of alkaloids, phytoalexins, glucosinolates, phenylpropanoids, and cpr-Gf is induced by benzoate. These results indicate terpenoids. In addition, they are involved in the detoxification that CPR-Gf is the main but not the only electron donor of herbicides and pesticides (18 –20). for several P450 monooxygenases from primary and sec- Eukaryotic non-mitochondrial cytochrome P450 monooxyge- ondary metabolism. nases are membrane proteins that require association with a NADPH-cytochrome P450 oxidoreductase (CPR) for activity. CPRs facilitate the transfer of electrons from NADPH via FAD The Ascomycetes Gibberella fujikuroi mating population 1 and FMN to the prosthetic heme group of the P450 monooxy- (MP) C (1), recently renamed Fusarium fujikuroi (2), is well genase. Interaction of both proteins in the microsomal mem- brane have been reported to occur through charge pairing as * This work was supported by the Deutsche Forschungsgemeinschaft well as by hydrophobic interactions through the N-terminal (Tu101/9-1) and Fondo Nacional de Desarrollo Cientifico y Tecnologico region of CPR (21). In contrast to the many different cyto- Grant 1020140. Rothamstead Research receives grant-aided support from the Biotechnology and Biological Sciences Research Council of the chrome P450 monooxygenases that can be found in a single United Kingdom. The costs of publication of this article were defrayed species, only one CPR-encoding gene is found in most organ- in part by the payment of page charges. This article must therefore be isms. Exceptionally, certain plants (22–24) and some zygomy- hereby marked “advertisement” in accordance with 18 U.S.C. Section cetes (25) possess two or three CPRs, although the physiologi- 1734 solely to indicate this fact. The nucleotide sequence(s) reported in this paper has been submitted TM to the GenBank /EBI Data Bank with accession number(s) AJ576025. To whom correspondence should be addressed. Tel.: 49-251-832- lin; GC-MS, combined gas chromatography-mass spectrometry; ICI, 24801; Fax: 49-251-8323823; E-mail: Bettina.Tudzynski@uni- Imperial Chemical Industries Ltd.; CPR, NADPH-cytochrome P450 muenster.de. oxidoreductase; cpr, disruption mutant; cyt b , cytochrome b ; kb, 5 5 The abbreviations used are: MP, mating population; GA, gibberel- kilobase(s); CM, culture medium; CD, Czapek Dox medium. This paper is available on line at http://www.jbc.org 25075 This is an Open Access article under the CC BY license. 25076 The G. fujikuroi cpr Gene 0.5% MgSO , 0.1% KH PO , and 5.0, 1.0, or 0 g/liter NH NO , cal relevance of the occurrence of multiple CPRs in these 4 2 4 4 3 respectively. organisms is unknown. Fungal cpr genes have been isolated For analysis of cpr-Gf expression with and without benzoate, strain from yeasts (26 –30) and some filamentous fungi, such as As- IMI58289 was cultivated for 3 days in 10% ICI medium on a rotary pergillus niger (31), Phanerochaete chrysosporium (32), Cun- shaker at 28 °C. The mycelium was washed, and 1.5 g (wet weight) each ninghamella (33), Rhizopus nigricans (25), and Coriolus versi- were transferred to 50 ml of 0 or 100% ICI medium with or without (0.5 color (34). Targeted cpr gene disruption to determine the or1mM) benzoate. For GA production, the strains were grown for 7–10 days on a rotary shaker (200 rpm) at 28 °C in 300-ml Erlenmeyer flasks function of these proteins has not yet been described for myce- containing 100 ml of either 20% ICI or optimized production medium lial fungi. containing 6% sunflower oil, 0.05% (NH ) SO , 1.5% corn-steep solids 4 2 4 Few data are available on the regulation of cpr gene expres- (Sigma-Aldrich), and 0.1% KH PO . Benzoate plate tests were per- 2 4 sion in lower eukaryotes. In the yeasts C. maltosa and Candida 2 formed on CM and Czapek Dox (CD) (Sigma-Aldrich) agar with 1 mM tropicalis, the assimilation of n-alkanes is catalyzed by P450 benzoate or without benzoate. monooxygenases. The addition of n-alkanes to the medium DNA and RNA Isolation—Genomic DNA was isolated from lyophi- lized mycelium as described by Doyle and Doyle (43).  DNA from resulted in strong induction of P450 and cpr gene (28, 35), and positive  clones was prepared according to Maniatis et al. (44). Plasmid co-regulation of the cpr and P450 genes has been reported for DNA was extracted using Genomed columns following the manufactur- Saccharomyces cerevisiae (36). In A. niger, the addition of ben- er’s protocol (Genomed, Bad Oeynhausen, Germany). RNA was isolated zoate increased the expression of the benzoate p-hydroxylase using the RNAgents total RNA isolation kit (Promega, Mannheim, (bphA) and cprA gene (14). Germany). The G. fujikuroi CPR would be expected to have a strong PCR—Degenerate primers CPR1 and CPR2 were designed by C. Wasmann (University of Arizona) on the basis of CLUSTAL alignment influence on GA biosynthesis since four P450 monooxygenases of fungal CPRs and kindly provided for cloning the G. fujikuroi cpr-Gf (P450-1-P450-4) are known to be involved in this pathway. gene. PCR reactions contained 25 ng of DNA, 10 ng of each primer, 0.2 Thus, loss of CPR activity should affect the rates of several mM dNTPs and 2 units of Taq polymerase (Red Taq, Sigma-Aldrich) in reactions in the pathway. Such an effect was found for the 50 l. PCR was carried out at 94 °C for 4 min followed by 30 cycles of UV-induced G. fujikuroi mutant SG138, which has lost most of 94 °C for 1 min, 54 °C for 1 min, and 72 °C for 1 min. The resulting the oxidation steps catalyzed by P450 monooxygenases (37). 0.5-kb PCR fragment was used as a probe for screening a genomic -DASH II library (Stratagene Europe, Amsterdam, The Netherlands) Because it is unlikely that all structural P450 monooxygenase of G. fujikuroi m567 at 65 °C. Sequences are: CPR1, 5-AAG YTG CAG genes are mutated by one UV treatment it is possible that the CCY CGC TAC TAY TCS ATC TC-3; CPR2, 5-CTT CCA YTC RTC mutant contains a lesion mutation in the cpr gene. This is CTT GTA SAR GAA RTC CTC-3. confirmed in the present paper, in which we report the cloning, For cloning cpr-Gf of SG138, four primer pairs were used to get sequencing, and targeted gene disruption of the G. fujikuroi cpr overlapping fragments after amplification: P138-1, 5-GTG GCC AAA gene (cpr-Gf) and its effect on GA production. We show by GTT CAT GAT TAG TGC-3;-2,5-TTG CGG ACC ATA GAG TTG TAG TGC-3; -3, 5-TCG CCA AGG AGG GTA AGA-GTC-3; -4, 5- sequence comparison that cpr-Gf contains a mutation in GCT GCC AGG GCG GTT CAT-3;-5,5-AAC CCC TAC ATT GCC SG138, and we were able to restore GA production to the CCT ATC G-3; -6, 5-TCG GCA ACC AAA GAA CAA GAG TG-3; -7, mutant by complementation not only with the cpr-Gf gene but 5-ACA GGC CCC CGC AAT AAG TA-3;-8,5-TGT CGG CAA GTC also with the heterologous A. niger cprA gene. We have also CAT GTC TAA GTG-3. Each fragment is about 850 bp long. For compared the regulation of cpr-Gf with that of the four GA RT-PCR, primers CPR-RT1 and CPR-RT2 were used to amplify frag- biosynthetic monooxygenase genes. ments of about 330 bp including a putative intron: primer CPR-RT1 (1), 5-CAA CCG AGG ATT TCA TGT ACC-3; CPR-RT2 (2), 5-CCC EXPERIMENTAL PROCEDURES TTG GCC TCA GAC ACC-3. For analysis of putative cpr knock-out transformants, the following Fungal Strains—Strains IMI58289 (Commonwealth Mycological In- diagnostic primers were used: for integration at the 5 region of cpr, stitute, Kew, UK) and m567 (Fungal Culture Collection, Weimar, CPR-DF1 (7), 5-CGG GGA TGG AGG CAA GAG AAT GAA-3, and Germany) are GA-producing wild-type strains of G. fujikuroi MP-C PUCH-P (8), 5-CCC TTG GCC TCA GAC ACC-3; for integration at the (anamorph F. fujikuroi). The GA-deficient mutant G. fujikuroi SG138 3 region of cpr, CPR-DF2 (9), 5-GAT CTA CAG ACT TGC TTC TGT (kindly provided by J. Avalos, University of Sevilla, Spain) strain was LGG-3, and PUCH-T (10), 5-TCA ACG CAT ATA GCG CTA GC-3. derived from IMI58289 via UV mutagenesis of spores (37). The gene replacement vector pcpr-Gf was constructed by cloning Bacterial Strains and Plasmids—Escherichia coli strain Top10 the PCR-amplified flanks into vector pUCH2– 8 (45). For amplifying the (Invitrogen) was used for plasmid propagation. Vector pUC19 was 5 and 3 flanks the following primers with introduced restriction sites used to clone DNA fragments carrying the G. fujikuroi cpr gene and were used: Dcpr-1 (3), 5-CGG GAA GTA CAA GGT ACC GTG CAA gene fragments. First, a 2.2-kb BglII fragment of -clone 2–1was AT-3 (underlines indicate the KpnI site); Dcpr-2 (4), 5-TCA ACG AGA cloned into BamHI-restricted pUCI9 (pCPR1A). Then a 3.8-kb XbaI TGT CGA CGT TTT TGT CC-3 (underlines indicate the SalI site); fragment was cloned into pUCI9/XbaI (pCPR1B). pCPR1B was cut Dcpr-3 (5), 5-TCC AAC TTC AAG CTT CCC TCG GAC-3 (underlines with NcoI/SphI. The derived 3-kb fragment was cloned into pCPR1A/ indicate the HindIII site); Dcpr-4 (6), 5-GAT AAC CAA AGA GCT CGT NcoI/SphI to produce vector pcpr-Gf, containing the entire cpr-Gf GGA CAG GT-3 (underlines indicate the SacI site). gene. The vector pNR1 was constructed by cloning the PstI/BamHI Screening the -DASH II Library—About 35,000 recombinant phages fragment of the Streptomyces noursei nat1 gene encoding the nourseo- from a  library prepared from genomic DNA of wild-type G. fujikuroi thricin acetyltransferase (38) into pBluescript II KS. The gene was m567 (46) were plated and transferred to Nylon N membranes (Am- transcribed under the control of the Aspergillus nidulans oliC pro- 2 ersham Biosciences). Hybridization was performed at high stringency moter (39) and terminated by the Botrytis cinerea tub1 terminator. (65 °C). The blots were washed in 2 SSC (1 SSC  0.15 M NaCl and For gene replacement experiments, fragments from the 5- and 3- 0.015 M sodium citrate), 0.1% SDS at 65 °C followed by a wash with noncoding regions of cpr were cloned into the vector pUCH2– 8 (40), 0.1 SSC, 0.1% SDS at 65 °C. Positive recombinant phages were used carrying the hygromycin B resistance marker. for a second round of plaque purification. Media and Culture Conditions—For DNA isolation, fungal strains Southern and Northern Blot Analysis—After digestion with restric- were grown in 100 ml of liquid CM optimized for Fusarium spp. (41) for tion endonucleases and electrophoresis, genomic or  DNA was trans- 3 days at 28 °C on a rotary shaker set at 200 rpm. The mycelium was ferred onto Hybond N filters (Amersham Biosciences). P-Labeled harvested by filtration through a sterile glass filter (G2, Schott Jena, probes were prepared using the random oligomer-primer method (44). Germany), washed with sterile distilled water, frozen in liquid nitrogen, Filters were hybridized at 65 or 56 °Cin5 Denhardt’s solution con- and lyophilized for 24 h. The lyophilized mycelia were ground to a fine taining 5% dextran sulfate (44). Filters were washed at the same powder with a mortar and pestle. For RNA isolation, fungal strains temperature used for hybridization in 2 saline/sodium phosphate/ were grown in 100, 20, or 0% ICI medium (42) containing 8% glucose, EDTA (SSPE), 0.1% SDS and 1 SSPE, 0.1% SDS. Northern blot hybridizations were accomplished by the method of Church and Gilbert (47). The G. fujikuroi rDNA gene was used as a J. van Kan, personal communication. control hybridization probe to confirm RNA transfer. The G. fujikuroi cpr Gene 25077 FIG.1. A, physical map and major sub- clones of gene cpr-Gf. The functional do- mains of the CPR-Gf enzyme and the po- sition of PCR primers are marked. 1, CPR-RT1; 2, CPR-RT2; 3, Dcpr-1; 4, Dcpr-2; 5, Dcpr-3; 6, Dcpr-4. B, strategy for construction of the gene replacement vector pcpr-Gf. 7, CPR-DF1; 8, PUCH-P; 9, CPR-DF2; 10, PUCH-T. H, HindIII; B, BamHI; X, XbaI; S, SalI; E, EcoRI; N, NcoI; Sa, SacII. Sequencing—DNA sequencing of recombinant plasmid clones was ously (8). Compounds were identified by comparison of their mass accomplished with an automatic sequencer LI-COR 4000 (MWG, spectra with those in a spectral library (49). For quantitative analysis of Mu ¨ nchen, Germany). The two strands of overlapping subclones ob- GAs aliquots of the extracts were spiked with [17- H ]GA internal tained from the genomic DNA clones were sequenced using the univer- standards and analyzing using a GCQ instrument (Thermo Electron sal and the reverse primers or specific oligonucleotides obtained from Corp.) as described previously (50), except chromatogram peak areas MWG Biotech (Munich, Germany). DNA and protein sequence align- were obtained from full scans. ments were done with DNA Star (Madison, WI). For phylogenetic anal- Plate Tests with Benzoate—For analysis of benzoate tolerance, yses, gaps were not considered, and corrections for multiple substitu- strains IMI58289, SG138, and transformants T20 and KT-1 were grown tions were applied. Trees were constructed with the program MegAlign for 6 days at 28 °C on CM and CD agar containing 1 mM benzoate. (DNA Star Madison, WI). Transformation of G. fujikuroi—The preparation of protoplasts and RESULTS the transformation procedure were as previously described (46). For Cloning and Targeted Gene Disruption of cpr-Gf—A frag- gene replacement, 10 protoplasts (50 l) of strain IMI58289 were transformed with 10 g of the KpnI/SacI fragment of the gene replace- ment of the cpr-Gf gene with the expected size of 500 bp was ment vector pcpr. For complementation of the mutant strain SG138 amplified by PCR using degenerate primers CPR1 and CPR2 with intact cpr genes, protoplasts were transformed with 10 gofthe derived from the FAD and NADPH binding domains, respec- circular complementation vector pcpr-Gf carrying the G. fujikuroi cpr tively (see Fig. 1A). The fragment exhibited a high degree of gene or pcprA with the cpr gene from A. niger (31). Both plasmids were homology with CPRs from other fungi and served as a probe for co-transformed with pAN7.1 (48), carrying the hygromycin resistance marker. screening the -DASH II genomic library of G. fujikuroi, MP-C. For complementation of the transformant cpr -T20 with the wild- Of the three genomic  clones isolated, one was used for isolat- type cpr-Gf gene, protoplasts were co-transformed with 10 g of each ing the putative full-length gene (Fig. 1A). About a 4000-bp the circular complementation vector pcpr-Gf and pNR1. Transformed sequence including 1000 bp of the 5-non-coding region was protoplasts were regenerated at 28 °C on complete regeneration agar obtained by sequencing in both directions. A 273-bp cDNA (0.7 M sucrose, 0.05% yeast extract, 0.1% (NH ) SO ) containing 120 4 2 4 g/ml hygromycin B (Calbiochem) or 100 g/ml nourseothricin (Werner fragment spanning a putative intron was generated by RT-PCR BioAgents, Jena, Germany) for 6 –7 days. For purification, single conid- using primers CPR-RT1 and CPR-RT2 (Fig. 1A). Comparison of ial cultures were obtained from hygromycin B- or nourseothricin-resist- the genomic and cDNA sequences confirmed the expected in- ant transformants and used for DNA isolation and Southern blot tron of 52 bp. The cpr-Gf gene has been deposited under the analysis. accession number AJ576025. Gibberellin Assays—GA and GA were analyzed by thin layer 3 4/7 chromatography on silica gel eluted with ethyl acetate/chloroform/ace- The deduced amino acid sequence of cpr-Gf was aligned with tic acid (60:40:5). The complete GA complement produced by the differ- known CPRs from taxonomically diverse species and allowed ent strains was determined by GC-MS analysis after extraction from the identification of all functional domains involved in the the culture fluid as already described (8), except that compounds were binding of the cofactors FMN, FAD, and NADPH and the P450 separated on a 30-m  0.32-mm  0.25-m HP-5 WCOT column (Agi- proteins (Fig. 1A). A phylogenetic tree comparing CPR-Gf with lent Technologies) and analyzed using a MAT95XP mass spectrometer (Thermo Electron Corp.) GC-MS conditions were as described previ- CPRs from other species indicates that it is most closely related 25078 The G. fujikuroi cpr Gene were then analyzed by TLC (Fig. 3C) and GC-MS (Table I). Transformants T20 and T49 as well as mutant SG138 do not produce GA ,GA , and GA , the last three products of the 4 7 3 pathway. Interestingly, the deletion mutants and SG138 show a reduction of growth rate on agar plates, which is more sig- nificant on minimal CD agar (Fig. 4C) than on CM (Fig. 4A). The similar characteristics for T20, T49 (data not shown), and SG138 as well as the described loss of P450-catalyzed oxidation steps (37) support our proposal that the latter strain is also affected in cpr-Gf. Identification of the Mutation in SG138 —To confirm our proposal that the UV treatment affected the cpr gene in the mutant SG138, the cpr gene from the mutant was amplified by four primer pairs to give four overlapping fragments, which were cloned, and three independent clones were sequenced in both directions. Comparison with the sequence of the wild-type gene copy confirmed a point mutation in the first position of the codon at amino acid position 627 from C to T resulting in a TGA FIG.2. Phylogenetic tree of cytochrome P450 reductases from stop codon instead of CGA for arginine. The truncated CPR fungi, plants, and animals, based on amino acid sequences. Ac- peptide is, therefore, 83 amino acids shorter than the wild-type cession numbers: A. niger CprA S38427; Arabidopsis thaliana ATR1 CPR enzyme. All CPR proteins analyzed so far contain a S21530; A. thaliana ATR2 S21531; Bombyx mori BAA95684; Cunning- NADPH binding domain consisting of three segments (32). This hamella echinulata AAF89959; C. maltosa N_CPR P50126; Cavia por- is also the case for CPR-Gf (Fig. 1A). In the mutant SG138 a cellus P37039; C. versicolor CPR BAB83588; C. tropicalis N_CPR P37201; Drosophila melanogaster Cpr-P1 NP_477158; Homo sapiens large part of the last segment of the NADPH binding domain is P16435; Musca domestica Q07994; Mus musculus NP_032924; P. chry- missing. sosporium CPR AAG31351; P. chrysosporium CPR2 AAG31350; Rattus Complementation of a Deletion Mutant with the Wild-type norvegicus AAA41683; S. cerevisiae NCPRNP_011908; Schizosaccharo- myces pombe CPR CAA2429; S pombe CPR2 T40056; cpr-Gf Gene—The deletion mutant T20 was co-transformed with the complementation vector pcpr-Gf carrying the wild- type cpr-Gf gene and vector pNR1, with the nourseothricin to CPRA of A.niger (61% identity) followed by the yeast and the resistance gene as selection marker (see “Experimental Proce- other fungal CPRs (Fig. 2). dures”). Two nourseothricin-resistant transformants, KT-1 and To determine the importance of the cpr-Gf gene for GA pro- KT-13, were analyzed for correct integration of the cpr-Gf gene. duction, a gene replacement vector was constructed as shown As shown in Southern blot (data not shown) and Northern blot in Fig. 1B. Two flanking sequences were amplified by PCR analysis (Fig. 5), only KT-1 showed multiple copies of the introducing KpnI and SalI (flank 1) and HindIII and SacI hybridizing wild-type gene and a high transcript level of the (flank 2) restriction sites into the oligonucleotides. The flank- correct size, whereas transformant KT-13 does not contain ing sequences were cloned into the corresponding sites of vector vector pcpr-Gf. Analysis of the GA concentrations showed al- pUCH2– 8 carrying the hygromycin resistance cassette produc- most full restoration of GA production. Analysis of the three ing vector pcpr-Gf (Fig. 1B). The 4.4-kb KpnI/SacI fragment of final products, GA ,GA , and GA , by GC-MS (Table I) dem- 4 7 3 the vector was used for transforming the wild-type IMI58289. onstrated that the activity of all four P450 monooxygenases 72 transformants were analyzed by PCR for homologous inte- was at least partially restored, resulting in production of the gration using the primer pairs 7 (CPR-DF1) and 8 (PUCH-P) normal GA pattern (Fig. 6) and formation of wild-type-like and primer pairs 9 (CPR-DF2) and 10 (PUCH-T) (Fig. 1B). amounts (or even more) of the final product gibberellic acid Three transformants revealed the expected diagnostic bands. (GA ). Furthermore, the growth rate of KT-1 on CM and CD After purification by single-spore isolation, these transfor- agar was comparable with that of the wild-type (Fig. 4, A and mants were analyzed by Southern blot analysis together with C). the wild-type IMI58289 and the putative cpr mutant SG138. Analysis of GA Intermediates in the cpr Mutants—The effect The HindIII-digested DNA was hybridized to flank 2 as probe of the cpr-Gf deletion on each of the P450-catalyzed steps in the (Fig. 1B). For two transformants, T20 and T49, the hybridizing GA biosynthetic pathway was investigated by determining the 2.5-kb wild-type band was replaced by a smaller (1.4 kb) and a full spectrum of intermediates in the cpr mutants (Fig. 6) and bigger (4.4 kb) band (Fig. 3A) due to a tandem integration of quantifying selected intermediates by GC-MS (Table I). The cpr two replacement cassettes into the cpr locus. The tandem in- deletion mutants do not produce GA ,GA ,orGA , the final tegration was confirmed by PCR with primers 8 (PUCH-P) and 3 4 7 products of the GA biosynthetic pathway, but only a very low 9 (CPR-DF2) (Fig. 1B) and sequencing of the fragment. Trans- amount of the non-hydroxylated intermediates GA and GA formant T57 integrated the fragment ectopically and still con- 15 24 together with high levels of ent-kaurene (Fig. 7). The level of tained the wild-type band. GAs in cpr is about 1% of that found in the wild-type strain The deletion of the wild-type copy of cpr in transformants IMI58289. These results indicate a very low activity of ent- T20 and T49 was confirmed by Northern blot analysis. The kaurene oxidase (P450 – 4) and low but significant activities for wild-type, transformant T57, and mutant SG138 contained a GA 7-oxidase (one of the activities of P450-1) and GA 20- transcript of about 2.1 kb, whereas T20 and T49 lost the entire oxidase (P450-2) in the absence of the P450 reductase. In con- coding region of cpr and did not show any cpr transcript trast, 3-hydroxylation, another activity of the P450-1 mo- (Fig. 3B). GA Production and Growth Characteristics of Deletion Mu- nooxygenase, was absent in cpr and, thus, would appear to have an absolute requirement for the P450 reductase. The tants and Strain SG138 —To determine the effect of cpr-Gf deletion on GA production, the wild-type strain, both knock-out spectrum of GAs found in SG138 (Table I) was similar to that in cpr, consistent with our demonstration of a mutation in the mutants, and the putative cpr mutant SG138 were cultivated for 7 days in the synthetic 20% ICI medium. The culture fluids P450 reductase in this strain. The requirement of the 13- The G. fujikuroi cpr Gene 25079 FIG.3. Analysis of gene replace- ment strains. A, Southern blot analysis of the wild-type strain IMI58289, UV mu- tant SG138, and three transformants. Genomic DNA was digested with HindIII and hybridized with the right flanking sequence of the replacement vector pcpr-Gf (see Fig. 1). B, Northern blot analysis. G. fujikuroi rDNA was used as the control. C, thin layer chromatography (TLC) of the wild-type IMI58289 and mu- tant strains after 7 days of cultivation in 20% ICI medium. TABLE I Concentration of GAs in culture filtrates of wild-type, cpr mutants, and cpr-complemented strains Results are expressed in g/ml  S.E.; n  3, except where indicated otherwise. Strain GA GA GA GA GA GA GA GA 3 1 7 4 9 15 24 25 IMI58289 (wild-type) 13.4  0.3 5.75  0.25 1.34  0.09 3.23  0.11 1.26  0.05 0.35  0.02 0.80  0.07 0.50  0.03 cpr-T20 0 0 0.004  0.001 0.017  0.004 0 0.64  0.02 0.17  0.01 0 cpr-cprGf (KT-1) 19.5  0.5 15.1  0.5 1.57  0.07 5.95  0.13 2.21  0.14 0.44  0.03 1.15  0.03 1.32  0.17 SG138 0 0 0.012  0.0005 0.014  0.002 0 0.37  0.02 0.14  0.004 0 b b SG138-cprGf 11.5  0.4 1.50  0.02 1.01  0.06 1.15  0.08 0.54  0.03 0.06  0.007 0.20  0.02 0.39  0.01 c d c c b c SG138-cprA-7 29.5  0.2 6.29  0.13 3.10  0.05 0.56  0.05 0.42  0.03 0.15  0.04 0.35  0.01 1.03  0.06 0.004 g/ml. n  2. n  6. n  5. hydroxylation reaction, catalyzed by P450-3, for the reductase ity of SG138 was also fully restored by complementation with was tested in the cpr mutants by incubating with [ C]GA and the reductase gene from A. niger (cpr-A), although A. niger is analyzing the products by high performance liquid chromatog- not able to produce any GAs (Table I). raphy and GC-MS. Formation of [ C]GA was reduced in the The Specificity of cpr-Gf for GA Biosynthesis Activity—As mutants relative to the wild-type (20% compared with 84%), part of an enquiry into whether or not cpr-Gf is specific for the indicating that the 13-hydroxylase has a partial, but not abso- P450s involved in GA biosynthesis we attempted to determine lute requirement for the P450 reductase. Complementation of if the cpr and GA monooxygenase genes were co-regulated. SG138 with cpr-Gf increased [ C]GA formation from Co-regulation has been reported, for example, for the A. niger [ C]GA to 56%. cprA and benzoate p-hydroxylase (bphA) genes (14). The GA profile and amounts in cpr and SG138 were re- Three of the four GA-specific P450 monooxygenase genes stored to those in the wild-type strain by complementation with (P450-1, P450-2, and P450-4) are known to be regulated by the the reductase gene from G. fujikuroi (Table I and Fig. 6). The general transcription factor AREA (51) and, thus, are highly major products synthesized by the complementation mutants expressed under nitrogen starvation conditions. Therefore, we were the 3-hydroxylated GAs GA and GA plus lower compared the expression pattern of these three P450 genes 3 1 amounts of GA and GA . Interestingly, the GA-producing abil- with that of cpr-Gf. Interestingly, the cpr-Gf gene is co-regu- 4 7 25080 The G. fujikuroi cpr Gene FIG.4. Growth of the cpr-deficient mutants is compromised. The wild-type IMI58289, mutants SG138 and T20, and the comple- mented strain KT-1 were grown on CM and CD agar with and without benzoate. A, SG138. B, IMI58289. C, T20. D, KT-1 FIG.5. Northern blot analysis of the wild-type IMI58289 and two transformants, KT-13 and KT-1, after transformation with the complementation vector pcpr-Gf. lated with P450-1, P450-2, and P450-4; high cpr-Gf transcrip- FIG.6. GC-MS analysis of culture filtrates of the wild-type tion levels were found under nitrogen starvation conditions but (IMI58289), cpr disruption mutant (CPR-T20), and line KT-1, in much less (though higher than for the monooxygenase genes) which T20 has been complemented with the G. fujikuroi CPR with high amounts of nitrogen (Fig. 8A). We investigated the gene. Total ion currents are shown for extracts as methyl esters trim- specificity of the interaction between CPR-Gf and the GA bio- ethylsilyl ethers. Components were identified by comparison of their mass spectra with published data (49) as follows: peak 1, ent-kaurene; synthetic monooxygenases by transforming the mutant SG138 peak 2, ent-kaurenoic acid; peak 3,GA peak 4,GA ; peak 5,GA ; peak 9; 25 24 with the cprA gene of A. niger. Ten hygromycin-resistant trans- 6,GA and 7-hydroxykaurenolide; peak 7,GA ; peak 8,GA ; peak 9, 14 4 7 formants were cultivated under GA production conditions and fujenoic acid, peak 10 GA ; peak 11,GA ; peak 12;GA isolactone; 13 36 3 analyzed for GA content. Three transformants, SG138-cprA-7, peak 13,7, 18-dihydroxykaurenolide; peak 14,GA ; peak 15,GA ; peak 1 3 16, ent-kaurenol; peak 17,GA . Unlabeled peaks are due to compounds -8, and -19, were able to produce GA (e.g. SG138-cprA-7, Table unrelated to GA biosynthesis. The peak at the same retention time as I), demonstrating that CPRA from the GA-non-producing fun- ent-kaurene in the KT-1 extract contains no ent-kaurene. gal species A. niger, which is described as an activator of the benzoate p-hydroxylase, is able to act as electron donor and activator of GA biosynthetic enzymes P450-1-P450-4 in G. fu- sensitivity to this compound due to the lost activation of the jikuroi. However, in contrast to cpr-Gf, cprA was expressed benzoate p-hydroxylase by CPR-Gf. On the other hand, comple- independently of the nitrogen condition in G. fujikuroi mentation of T20 with the wild-type cpr-Gf copy fully restored (Fig. 8B). the high resistance level for benzoate (Fig. 4, B and D). On the basis of these results we speculated that CPR-Gf may These results led us to anticipate induction of cpr-Gf gene also act in much more than the GA biosynthetic pathways. To expression by benzoate, as is the case for cprA in A. niger. The show this, we determined if CPR-Gf is involved in detoxifica- tion of benzoate in a similar way to CPRA in A. niger. We grew addition of benzoate to the medium significantly induced the cpr-Gf transcription level in G. fujikuroi, especially when 1 mM the wild-type strain IMI58289, two cpr-Gf mutants, SG138 and T20, as well as the complemented strain KT-1 on CM and CD benzoate was added (Fig. 9). Interestingly, with benzoate in the agar with or without benzoate. The growth patterns show very medium, cpr-Gf expression is no longer repressed by high clearly that mutation or deletion of cpr-Gf led to an extreme amounts of nitrogen. The G. fujikuroi cpr Gene 25081 FIG.8. A,Northern blot analysis of the wild-type IMI58289 and the mutant SG138 showing regulation of P450 and cpr gene expression by nitrogen. Strains were grown for 3 days in 10% ICI medium, washed, and transferred to 0% ICI (no nitrogen) or 100% ICI (high amounts of nitrogen) medium for 5 or 10 h. B, Northern blot analysis of transformant SG138-K1, carrying the A. niger cprA gene. FIG.7. GA biosynthetic pathways in G. fujikuroi indicating FIG.9. Northern blot analysis of the wild-type IMI58289 show- reactions affected in the cpr-Gf mutant. Products detected in the ing benzoate induction of cpr gene expression. The strain was culture are underlined, absent reactions are marked by an X, undetec- cultivated for 3 days in 10% ICI and then transferred to 0 or 100% ICI ted final products are in parentheses, and the proposed sequence of medium with 0.5 or 1 mM benzoate or without benzoate. reactions are shown with pointed arrows. Participation of CPR in reac- tions after the block at GA -aldehyde was not demonstrated experimentally. ent-7-hydroxykaurenoic acid, followed by ring contraction with the production of GA -aldehyde, its 3-hydroxylation to DISCUSSION GA -aldehyde, and oxidation to GA , in addition to several 14 14 Using degenerate PCR and genomic library screening we side reactions (6). P450-2 converts GA to GA and GA to 14 4 12 isolated a gene from G. fujikuroi with high homology with GA with the loss of carbon-20 (7). Finally, P450-3 is responsi- cytochrome P450 reductase (CPR) genes from other species, ble for the 13-hydroxylation of GA to GA , and, in a minor 7 3 particularly the gene from A. niger (cprA) with which it is 61% pathway, of GA to GA (8). Although not all P450-catalyzed 4 1 identical at the amino acid level. The G. fujikuroi cpr gene steps are completely blocked in the deletion mutants, they are (cpr-Gf) encodes a protein of 713 amino acids containing all the reduced in activity and some reactions, such as 3-hydroxyla- expected domains for binding the prosthetic factors FAD, FMN, tion and loss of C-20, are completely absent in cpr mutants. and NADPH as well as for P450s (32). Targeted disruption of Thus, CPR-Gf may act as electron donor of all four P450 cpr-Gf led to a very substantial loss of GA production, demon- monooxygenases. strating that the reductase is required for normal activity of The small amounts of the non-hydroxylated GA and GA , 15 24 P450s involved in GA biosynthesis. In G. fujikuroi, GA biosyn- and the high level of ent-kaurene found in cpr and SG138 thesis requires four P450 monooxygenases, which are respon- indicate low activities of ent-kaurene oxidase (P450-4), GA sible for more than 10 enzymatic steps. P450 – 4 catalyzes the 7-oxidase (P450-1), and 20-oxidase (P450-2) in the absence of three oxidation steps from ent-kaurene to ent-kaurenoic acid the reductase. The activity of 13-hydroxylase (P450-3), which (5). P450-1 catalyzes the oxidation of ent-kaurenoic acid to could not be assessed from the GA profiles of the cpr mutants 25082 The G. fujikuroi cpr Gene because its substrate is not produced, was assayed by incubat- but only one CPR-encoding gene, with a high degree of identity ing with [ C]GA and shown to have reduced activity in the to CPR-Gf. Therefore, it is likely that G. fujikuroi contains a absence of the reductase. The residual activities of these en- single cpr gene that interacts with each of the P450s. zymes indicate the participation of a second electron transport We analyzed the GA biosynthetic and cpr genes also in the closest related members of the G. fujikuroi species complex protein in GA biosynthesis that would supply electrons only to some of the reactions catalyzed by P450 monooxygenases and consisting of eight mating populations (MP-A to MP-H). Most of these species contain the complete GA gene cluster, but only with less efficiency than the P450 reductase. members of MP-C (F. fujikuroi) are able to produce GAs. The Analysis of the GA content of the UV-induced mutant SG138 loss of GA production capability is due to a set of mutations in suggests that it contains slightly higher P450 activities than the coding and 5-noncoding regions of the GA biosynthetic the cpr lines and may, thus, possess low CPR activity. The GA genes, resulting in an overall amino acid sequence identity of levels in SG138 are similar to cpr, but ent-kaurene does not only 84 –94% in the case of P450 – 4. In contrast to the dispen- accumulate, and traces of GA ,GA ,GA , and GA are pres- 4 7 15 24 sable GA pathway genes, the level of sequence identity between ent. SG138 contains a point mutation in the cpr gene that gives the CPR enzymes is about 98%. These results suggest the a truncated protein that lacks part of the NADPH binding importance of CPR for essential functions of cell metabolism. domain. This mutation in SG138 would, thus, reduce consider- Here we report on the first deletion of a cpr gene in a ably but may not abolish CPR activity. Complementation of filamentous fungus. So far, only in S. cerevisiae has the single cpr with the cpr-Gf gene fully restored GA synthesis, giving cpr gene been successfully deleted without dramatic influence 3-hydroxylated C -GAs (GA ,GA ,GA , and GA ) at similar 19 3 4 1 7 on viability (54), indicating that an alternative electron donor levels than in the wild-type strain. This demonstrates that the must exist. It was suggested that in S. cerevisiae, a cytochrome P450 reductase is the main electron donor to the four GA b (cyt b ) could act as a second important element in the 5 5 biosynthetic P450s in G. fujikuroi. The second electron trans- electron donating system. Deletion of cyt b in the wild type did port pathway is much less effective and can only partially not display a phenotype, whereas disruption of the gene in a compensate for the absence of CPR. cpr strain was lethal, demonstrating that both enzymes can Interestingly, the two activities of the multifunctional complement each other in mutants with single disruptions of P450-1 monooxygenase (6) differ in their dependence on CPR. cpr or b (54). However, there is no additional electron-donat- Although 3-hydroxylase depends absolutely on CPR and is ing system overcoming the double knock out. completely blocked in cpr, the 7-oxidase activity can obtain The exact mechanism by which cyt b interacts with P450 electrons from an alternative source and is still moderately reductase is not yet clear. Numerous studies show that P450 active in the deletion mutant. Our results agree with previous activity can be enhanced by the addition of cyt b in some, but findings about differences in nucleotide co-factor requirements not all reactions (55–57). Human, but not yeast cyt b can for the different activities of P450-1 (52). The GA 20-oxidase selectively augment the rate of steroid hormone hydroxylations (P450-2) (7) does not produce C -GAs by cleavage of C-20 in by more than 10-fold, but this stimulation requires CPR and the absence of the CPR. Instead, the C tricarboxylic acid occurs without electron transfer to or from cyt b (55). product GA was found in cpr together with GAs with inter- In petunia, the product of a cyt b gene, which is expressed mediate oxidation states (alcohol and aldehyde) at C-20 that do exclusively in the flowers, regulates the activity of two P450s not accumulate in the wild type. Production of the C -GAs, involved in the biosynthesis of anthocyanin pigments. Tar- thus, appears to be completely dependent on CPR in contrast to geted inactivation of the b gene resulted in a flower color tricarboxylic acid synthesis. If the rate of C-20 oxidation is change caused by reduction in activity of these two P450s, considerably reduced in the absence of CPR, this may result in but it did not affect other P450s (17). We suggest that in G. the accumulation of enzyme-bound intermediates, which can fujikuroi a cyt b might take over the function of CPR for be hydrolyzed to give the C -GAs detected (7). activation of P450s involved in primary metabolism and, to a Besides the dramatic effect on GA production, mutations in much lesser extent, some of those functions in the dispensa- cpr in the deletion mutants and SG138 also affected the growth ble GA biosynthetic pathway. rate on synthetic medium and to a lesser extent on CM, indi- A possible alternative to general CPRs and cyt b as electron cating that CPR probably acts as electron donor also for P450- donors was found in the fumonisin gene cluster of G. fujikuroi related pathways in primary metabolism, e.g. for metabolism of MP-A (Fusarium verticilloides); the open reading frame of the sterols and fatty acids. However, the effect on primary metab- fumonisin biosynthetic gene FUM6 consists of a P450 gene that olism, especially on CM, is not as strong as might be expected is fused to a cpr gene (12) in which the FMN, FAD, and NADPH if CPR-Gf were the only electron donor associated with P450s binding domains are arranged in the same order as in other in G. fujikuroi. It is also possible that the reduced growth rate CPRs. This unusual enzyme belongs to a family consisting of is due to the accumulation of toxic intermediates of disrupted another fungal and two bacterial enzymes, the Fusarium oxys- secondary metabolite pathways. porum fatty acid -hydroxylase (58), the Bacillus megaterium Although A. niger does not produce GAs and does not contain fatty acid hydroxylase P450 (59), and the Bacillus subtilis BM-3 GA-related P450s, the cprA gene, involved in benzoate detoxi- Yfn1 gene product (A69975). The fusion of a P450 and a CPR into fication in A. niger, fully restored the GA production capacity of one single enzyme (FUM6) is highly unusual for fungal second- the cpr mutant SG138. This result indicates that CPRs act ary metabolite genes. In all the other examples described so far, nonspecifically as general electron donors for P450 monooxy- only typical P450 monooxygenase genes are present in gene genases from different pathways. The recently completed ge- clusters for dispensable metabolites; no CPRs, either as single nome of the Basidiomycetes, P. chrysosporium, revealed the genes or fused to P450s, are present in such clusters (3, 10, 45, presence of only one CPR-encoding gene (CPR, EC 1.6.2.4) and 60). It is not yet clear if the fusion protein is specific for the P450s at least 123 cytochrome P450 monooxygenase genes (53). The in the fumonisin pathway or could complement the functions of genome sequences of Neurospora crassa (www-genome.wi.mit- the general reductase in other pathways. .edu/annotation/fungi/neurospora) and Fusarium graminea- Major functions of P450s in numerous organisms, including rum (www-genome.wi.mit.edu/cgi-bin/annotation/fusarium), a species of the same genus (Fusarium)as G. fujikuroi, revealed 44 and 40 cytochrome P450 monooxygenases, respectively (9), S. Malonek and B. Tudzynski, unpublished results. The G. fujikuroi cpr Gene 25083 Acknowledgments—Strain SG138 was kindly provided by J. Avalos fungi, are the metabolism of xenobiotic drugs and toxins and (University of Sevilla, Spain). We gratefully acknowledge Catherine the assimilation of long chain alkanes as well as the metabo- Wasmann (University of Arizona) for sharing the sequence of the CPR- lism of endogenous compounds, such as sterols and fatty acids specific degenerated primers. We also thank P. Punt (TNO Nutrition (cited in Ref. 16). For Fusarium moniliforme it was shown that and Food Research Institute, Zeist, The Netherlands) for providing the cprA gene and for critical reading of the manuscript. We thank B. Berns the fungus can oxidize propylbenzene and that this reaction for typing the manuscript and J. Schulte for technical assistance. needs molecular oxygen and NADPH as the preferential coen- zyme, suggesting a microsomal cytochrome P450 monooxygen- REFERENCES ase system that contained NADPH-cytochrome P450 reductase 1. Xu, J.-R., Yan, K., Dickman, M. B., and Leslie J. F. (1995) Mol. Plant-Microbe Interact. 8, 74 – 84 (61). Other Fusarium strains, like F. solani, are able to detoxify 2. O’Donnell, K., Cigelnik, E., and Nirenberg, H. I. (1998) Mycologia 90, 465– 493 plant phytoalexins, such as pisatin, by a cytochrome P450 3. Tudzynski, B., and Ho ¨ lter, K. (1998) Fungal Genet. Biol. 25, 157–170 4. Linnemannsto ¨ ns, P., Vo, T., Hedden, P., Gaskin, P., and Tudzynski, B. (1999) enzyme system (62). The loss of CPR would have dramatic Appl. Environ. Microbiol. 65, 2558 –2564 consequences for fungal survival in the environment and for its 5. Tudzynski, B., Hedden, P., Carrera, E., and Gaskin, P. (2001) Appl. Environ. ability to infect plant tissue. Microbiol. 67, 3514 –3522 6. Rojas, M. C., Hedden, P., Gaskin, P., and Tudzynski, B. (2001) Proc. Natl. Many of the P450s are substrate-inducible. To become func- Acad. Sci. U. S. A., 98, 5828 –5834 tionally active, they are dependent on the high electron donat- 7. Tudzynski, B., Rojas, M. C., Gaskin, P., and Hedden, P. (2002) J. Biol. Chem. 277, 21246 –21253 ing activity of CPR under the same induction conditions. Thus, 8. Tudzynski, B., Mihlan, M., Rojas, M. C., Gaskin, P., and Hedden, P. (2003) phenobarbital and numerous other drugs and chemicals induce J. Biol. Chem. 278, 28635–28643 9. Yoder, O. C., and Turgeon, B. G. (2001) Curr. Opin. Plant Biol. 4, 315–321 not only the expression of many drug- and steroid-metabolizing 10. Keller, N. P., and Hohn, T. M. (1997) Fungal Genet. Biol. 21, 17–29 P450s but also that of NADPH-dependent reductase gene in 11. Desjardins, A. E., Hohn, T. M., and McCormick, S. P. (1992) Mol. Plant- Microbe Interact. 5, 214 –222 liver tissue (15). In the subtropical plant Catharanthus roseus, 12. Seo, J.-A., Proctor, R. H., and Plattner, R. D. (2001) Fungal Genet. Biol. 34, where so far only one cpr gene has been identified, expression 155–165 of this gene is co-regulated with that of numerous P450 genes 13. Young, C., McMillan, L., Telfer, E., and Scott, B. (2001) Mol. Microbiol. 39, 754 –764 involved in the biosynthesis of terpenoid indole alkaloids, phe- 14. Van den Brink, J. M., Punt, P. J., van Gorcom, R. F. M., and van den Hondel, nylpropanoids, and in other defense-related compounds (63). C. A. M. J. J. (2000) Mol. Gen. Genet. 263, 601– 609 15. Waxman, D. J., and Azaroff, L. (1992) Biochem. J. 281, 577–592 For A. niger it was shown that expression of both the P450 gene 16. Van den Brink, J. M., van Gorcom, R. F. M., van den Hondel, C. A. M. J. J., and bphA, involved in benzoic acid hydroxylation, and cprA were Punt, P. J. (1998) Fungal Genet. Biol. 23, 1–17 17. de Vetten, N., ter Horst, J., van Schaik, H.-P., de Boer, A., Mol, J., and Koes, induced by the substrate, benzoate. Interestingly, the majority R. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 778 –783 of cprA transcripts after benzoate induction were of a larger 18. Donaldson, R. P., and Luster, D. G. (1991) Plant Physiol. 96, 669 – 674 19. Bollwell, G. P., Bozak, K., and Zimmerlin, A. (1994) Phytochemistry 37, size due to the use of four alternative transcription start points. 1491–1506 In addition, an upstream open reading frame was found (14). 20. Schuler, M. (1996) Crit. Rev. Plant Sci. 15, 235–284 In G. fujikuroi the GA biosynthetic pathway is under control 21. Backes, W. L, and Kelley, R. W. (2003) Pharmacol. Ther. 98, 221–233 22. Lesot, A., Hasenfratz, M. P., Batard, A., Durst, F., and Benveniste, I. (1995) of nitrogen metabolite repression. Nitrogen starvation results Plant Physiol. Biochem. 33, 751–757 in a manyfold increase in transcript level for three of the four 23. Urban, P., Mignotte, C., and Kazmaier, M. (1997) J. Biol. Chem. 272, 19176 –19186 P450s (P450-1, P450-2, and P450-4) and induction of cpr-Gf.To 24. Koopmann, E., and Hahlbrock, K. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, demonstrate that CPR-Gf acts as a general electron donor for 14954 –14959 25. Kunic, B., Truan, G., Breskvar, K., and Pompon, D. (2001) Mol. Genet. Genom- many P450s, we analyzed the expression of cpr-Gf with and ics 265, 930 –940 without benzoate. As in A. niger, the cpr-Gf gene was signifi- 26. Yabusaki, Y., Murakami, H., and Ohkawa, H. (1988) J. Biochem. (Tokyo) 103, cantly induced by benzoate, indicating the existence of a CPR- 1004 –1010 27. Sutter, T. R., and Loper, J. C. (1989) Biochem. Biophys. Res. Commun. 160, Gf-dependent benzoate p-hydroxylase in G. fujikuroi. This was 1257–1266 supported by the finding that cpr mutants are significantly 28. Sutter, T. R., Sanglard, D., and Loper, J. C. (1990) J. Biol. Chem. 265, 16428 –16436 less resistant to benzoate than the wild type and the comple- 29. Miles, J. S. (1992) Biochem. J. 287, 195–200 mented mutant. Interestingly, the 5-non-coding region of 30. Kargel, E., Menzel, R., Honeck, H., Vogel, F., Bohmer, A., and Schunck, W.-H. (1996) Yeast 12, 333–348 cpr-Gf contains two n-alkane-inducible sequence elements, CA- 31. Van den Brink, J. M., van Zeijl, C. J. M., Brons, J. F., van den Hondel, CAT, suggesting a role for CPR also in n-alkane hydroxylation C. A. M. J. J., and van Gorcom, R. F. M. (1995) DNA Cell Biol. 14, 719 –729 32. Yadav, J. S., and Loper, J. C. (2000a) Biochem. Biophys. Res. Commun. 268, (see Ref. 64). In contrast to the cprA of A. niger, benzoate- 345–353 responsive regions were not found in the G. fujikuroi gene. 33. Yadav, J. S., and Loper, J. C. (2000b) Curr. Genet. 37, 65–73 34. Ichinose, H., Wariishi, H., and Tanaka, H. (2002) Appl. Microbiol. Biotechnol. However, these sequence elements might be located at position 59, 658 – 664 1400 to 1600 bp upstream of ATG (14). 35. Ohkuma, M., Masuda, Y., Park, S. M., Ohtomo, R., Ohta, A., and Takagi, M. NADPH-dependent oxidoreductase is the common electron (1995) Biosci. Biotechnol. Biochem. 59, 1328 –1330 36. Turi, T. G., and Loper, J. C. (1992) J. Biol. Chem. 267, 2046 –2056 donor to multiple P450 monooxygenases in all eukaryotic or- 37. Barrero, A. F., Oltra, J. E., Cerda ´ -Olmedo, E. Avalos, J., and Justicia, J. (2001) ganisms. In most fungi and all animals tested so far, only a J. Nat. Prod. (Lloydia) 64, 222–225 38. Kru ¨ gel, H., Fiedler, G., Smith, C., and Baumberg, S. (1993) Gene (Amst.) 127, single cpr gene was identified, indicating involvement of single 127–131 CPR in a multitude of P450-related reactions. In contrast to F. 39. Ward, M., and Turner, G. (1986) Mol. Gen. Genet. 205, 331–338 40. Alexander, N. J., McCormick, S. P., and Hohn, T. M. (1999) Mol. Gen. Genet. verticilloides (G. fujikuroi MP-A), where a catalytically self- 261, 977–984 sufficient P450 gene, Fum6, was found, the GA gene cluster in 41. Pontecorvo, G. V., Roper, J. A, Hemmonns, L. M., Mac Donald, K. D., and G. fujikuroi MP-C consists of four P450 monooxygenase genes Buften, A. W. J. (1953) Adv. Genet. 141, 141–238 42. Geissman, T. A., Verbiscar, A. J., Phinney, B. O., and Cragg, G. (1966) Phy- not fused to cpr genes. Our results indicate that cpr-Gf, which tochemistry 5, 933–947 is not located in the GA gene cluster, encodes the main electron 43. Doyle, J. J., and Doyle, J. L. (1990) Focus (Idaho) 12, 13–15 44. Maniatis, T., Sambrook, J., and Fritsch, E. F. (1989) Molecular Cloning: A donor responsible not only for activation of all four GA-related Laboratory Manual, 2nd Ed., Cold Spring Harbor Laboratory Press, Cold P450 monooxygenases but also for benzoate p-hydroxylase and Spring Harbor Laboratory, NY 45. Alexander, N. J., Hohn, T. M., and McCormick, S. P. (1998) Appl. Environ. at least some P450s involved in primary metabolism. Cloning Microbiol. 64, 221–225 and characterization of the gene encoding the G. fujikuroi 46. Tudzynski, B., Mende, K., Weltring, K. M., Kinghorn, J. R., and Unkles, S. E. cytochrome b (1996) Microbiology 142, 533–539 might provide insights into the possible role of 47. Church, G. M., and Gilbert, W. (1984) Proc. Natl. Acad. Sci. U. S. A. 81, this enzyme as an alternative electron donator and its partic- 1991–1995 ipation in the functional complex of GA P450s. 48. Punt, P. J., Oliver, R. P., Dingemanse, M. A., Pouwels, P. H., and van den 25084 The G. fujikuroi cpr Gene Hondel, C. A. M. J. J. (1987) Gene (Amst.) 56, 117–124 F. P. (1996) J. Biol. Chem. 271, 27438 –27444 49. Gaskin, P., and MacMillan, J. (1992) GC-MS of the Gibberellins and Related 57. Yamazaki, H., Gillam, E. M. J., Dong, M. S., Johnson, W. W., Guengerich, Compounds: Methodology and a Library of Spectra, Cantock’s Enterprises, F. P., and Shimada, T. (1997) Arch. Biochem. Biophys. 342, 329 –337 Bristol, UK 58. Nakayama, N., Takemae, A., and Shoun, H. (1996) J. Biochem. (Tokyo) 119, 50. Coles, J. P., Phillips, A. L., Croker, S. J., Garcı ´a-Lepe, R., Lewis, M. J., and 435– 440 Hedden, P. (1999) Plant J. 17, 547–556 59. Ruettinger, R. T., Wen, L. P., and Fulco, A. J. (1989) J. Biol. Chem. 264, 51. Mihlan, M., Homann, V., Liu, T.-W. D., and Tudzynski, B. (2003) Mol. Micro- 10987–10995 biol. 47, 975–991 60. Hohn, T. M., Desjardins, A. E., and McCormick, S. P. (1995) Mol. Gen. Genet. 52. Urrutia, O., Hedden, P., and Rojas, M. C. (2001) Phytochemistry 56, 505–511 248, 95–102 53. Warrilow, A. G. S., Lamb, D. C., Kelly, D. E., and Kelly, S. L. (2002) Biochem. 61. Uzura, A., Suzuki, T., Katsuragi, T., and Tani, Y. (2001) J. Biosci. Bioeng. 92, Biophys. Res. Commun. 299, 189 –195 580 –584 54. Truan, G., Epinat, J.-C., Rougeulle, C., Cullin, C., and Pompon, D. (1994) Gene 62. Khan, R., Tan, R., Mariscal, A. G., and Straney, D. (2003) Mol. Microbiol. 49, (Amst.) 142, 123–127 117–130 55. Auchus, R. J., Lee, T. C., and Miller, W. L. (1998) J. Biol. Chem. 273, 63. Lopes Cardoso, M. I., Meijer, A. H. Rueb, S., Queiroz Machado, J., Memelink, 3158 –3165 J., and Hoge, J. H. C. (1997) Mol. Gen. Genet. 256, 674 – 681 56. Yamazaki, H., Johnson, W. W., Ueng, Y.-F., Shimada, T., and Guengerich, 64. Yadav, J. S, and Loper, J. C. (1999) Gene (Amst.) 226, 139 –146 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

The NADPH-cytochrome P450 Reductase Gene from Gibberella fujikuroi Is Essential for Gibberellin Biosynthesis

Malonek, Stefan; Rojas, Maria C.; Hedden, Peter; Gaskin, Paul; Hopkins, Paul; Tudzynski, Bettina
Journal of Biological Chemistry – Jun 1, 2004
Download PDF
10 pages

Loading next page...
 
/lp/unpaywall/the-nadph-cytochrome-p450-reductase-gene-from-gibberella-fujikuroi-is-3hmFoMXNhq

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.m308517200
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 279, No. 24, Issue of June 11, pp. 25075–25084, 2004 © 2004 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. The NADPH-cytochrome P450 Reductase Gene from Gibberella fujikuroi Is Essential for Gibberellin Biosynthesis* Received for publication, August 4, 2003, and in revised form, March 16, 2004 Published, JBC Papers in Press, March 22, 2004, DOI 10.1074/jbc.M308517200 ¶ ¶ ¶ Stefan Malonek‡, Maria C. Rojas§, Peter Hedden , Paul Gaskin , Paul Hopkins , and Bettina Tudzynski‡ From the ‡Institut fu ¨ r Botanik der Westfa ¨ lischen Wilhelms-Universita ¨t Mu ¨ nster, Schlossgarten 3, D-48149 Mu ¨ nster, Germany, §Laboratorio de Bioorga ´ nica, Departamento de Quı ´mica, Facultad de Ciencias, Universidad de Chile, Casilla 653, Santiago, Chile, and Rothamsted Research, Harpenden, Herts AL5 2LQ, United Kingdom known as a rich source of gibberellins (GAs), which function as The fungus Gibberella fujikuroi is used for the com- mercial production of gibberellins (GAs), which it pro- hormones in higher plants. The major GA in most strains of duces in very large quantities. Four of the seven GA G. fujikuroi is gibberellic acid (GA ), the biosynthesis of which biosynthetic genes in this species encode cytochrome requires seven genes that are arranged in a gene cluster. Each P450 monooxygenases, which function in association of these genes and the function of the encoded enzymes have with NADPH-cytochrome P450 reductases (CPRs) that recently been fully characterized (3– 8). As well as genes encod- mediate the transfer of electrons from NADPH to the ing a pathway-specific geranylgeranyl diphosphate synthase P450 monooxygenases. Only one cpr gene (cpr-Gf) was (ggs2), ent-copalyldiphosphate/ent-kaurene synthase (cps/ks), found in G. fujikuroi and cloned by a PCR approach. The and GA desaturase (des), the cluster contains four cytochrome encoded protein contains the conserved CPR functional P450 monooxygenase genes (P450-1 to P450-4), which in most domains, including the FAD, FMN , and NADPH binding cases encode multifunctional enzymes that catalyze several motifs. cpr-Gf disruption mutants were viable but biosynthetic steps. showed a reduced growth rate. Furthermore, disruption About 40 different P450 monooxygenases involved in diverse resulted in total loss of GA ,GA , and GA production, 3 4 7 metabolic pathways have been identified in fungi (9). Many of but low levels of non-hydroxylated C -GAs (GA and 20 15 the P450 genes form part of gene clusters that are responsible GA ) were still detected. In addition, the knock-out mu- for the biosynthesis of metabolites, such as aflatoxins (10), tants were much more sensitive to benzoate than the trichothecenes (11), fumonisins (12), and paxillin (13), or for wild type due to loss of activity of another P450 mo- the metabolism of xenobiotics (14). Indeed a large group of the nooxygenase, the detoxifying enzyme, benzoate p-hy- fungal cytochrome P450s have been shown to be involved in the droxylase. The UV-induced mutant of G. fujikuroi, metabolism of drugs and other foreign compounds (14 –16), SG138, which was shown to be blocked at most of the GA whereas others participate in the biosynthesis of intracellular biosynthetic steps catalyzed by P450 monooxygenases, displayed the same phenotype. Sequence analysis of the compounds such as steroids (cited in Ref. 16). Metabolism of mutant cpr allele in SG138 revealed a nonsense muta- xenobiotics, including drugs and toxins, is also an important tion at amino acid position 627. The mutant was comple- function of P450 monooxygenases in animals as is the forma- mented with the cpr-Gf and the Aspergillus niger cprA tion of endogenous compounds such as sterols and fatty acids genes, both genes fully restoring the ability to produce (17). In plants, P450s are implicated in the formation of a broad GAs. Northern blot analysis revealed co-regulated ex- range of metabolites, including the growth hormones gibberel- pression of the cpr-Gf gene and the GA biosynthetic lins and jasmonates, essential components such as lignin, pig- genes P450-1, P450-2, P450-4 under GA production con- ments, and fatty acids, and secondary metabolites such as ditions (nitrogen starvation). In addition, expression of alkaloids, phytoalexins, glucosinolates, phenylpropanoids, and cpr-Gf is induced by benzoate. These results indicate terpenoids. In addition, they are involved in the detoxification that CPR-Gf is the main but not the only electron donor of herbicides and pesticides (18 –20). for several P450 monooxygenases from primary and sec- Eukaryotic non-mitochondrial cytochrome P450 monooxyge- ondary metabolism. nases are membrane proteins that require association with a NADPH-cytochrome P450 oxidoreductase (CPR) for activity. CPRs facilitate the transfer of electrons from NADPH via FAD The Ascomycetes Gibberella fujikuroi mating population 1 and FMN to the prosthetic heme group of the P450 monooxy- (MP) C (1), recently renamed Fusarium fujikuroi (2), is well genase. Interaction of both proteins in the microsomal mem- brane have been reported to occur through charge pairing as * This work was supported by the Deutsche Forschungsgemeinschaft well as by hydrophobic interactions through the N-terminal (Tu101/9-1) and Fondo Nacional de Desarrollo Cientifico y Tecnologico region of CPR (21). In contrast to the many different cyto- Grant 1020140. Rothamstead Research receives grant-aided support from the Biotechnology and Biological Sciences Research Council of the chrome P450 monooxygenases that can be found in a single United Kingdom. The costs of publication of this article were defrayed species, only one CPR-encoding gene is found in most organ- in part by the payment of page charges. This article must therefore be isms. Exceptionally, certain plants (22–24) and some zygomy- hereby marked “advertisement” in accordance with 18 U.S.C. Section cetes (25) possess two or three CPRs, although the physiologi- 1734 solely to indicate this fact. The nucleotide sequence(s) reported in this paper has been submitted TM to the GenBank /EBI Data Bank with accession number(s) AJ576025. To whom correspondence should be addressed. Tel.: 49-251-832- lin; GC-MS, combined gas chromatography-mass spectrometry; ICI, 24801; Fax: 49-251-8323823; E-mail: Bettina.Tudzynski@uni- Imperial Chemical Industries Ltd.; CPR, NADPH-cytochrome P450 muenster.de. oxidoreductase; cpr, disruption mutant; cyt b , cytochrome b ; kb, 5 5 The abbreviations used are: MP, mating population; GA, gibberel- kilobase(s); CM, culture medium; CD, Czapek Dox medium. This paper is available on line at http://www.jbc.org 25075 This is an Open Access article under the CC BY license. 25076 The G. fujikuroi cpr Gene 0.5% MgSO , 0.1% KH PO , and 5.0, 1.0, or 0 g/liter NH NO , cal relevance of the occurrence of multiple CPRs in these 4 2 4 4 3 respectively. organisms is unknown. Fungal cpr genes have been isolated For analysis of cpr-Gf expression with and without benzoate, strain from yeasts (26 –30) and some filamentous fungi, such as As- IMI58289 was cultivated for 3 days in 10% ICI medium on a rotary pergillus niger (31), Phanerochaete chrysosporium (32), Cun- shaker at 28 °C. The mycelium was washed, and 1.5 g (wet weight) each ninghamella (33), Rhizopus nigricans (25), and Coriolus versi- were transferred to 50 ml of 0 or 100% ICI medium with or without (0.5 color (34). Targeted cpr gene disruption to determine the or1mM) benzoate. For GA production, the strains were grown for 7–10 days on a rotary shaker (200 rpm) at 28 °C in 300-ml Erlenmeyer flasks function of these proteins has not yet been described for myce- containing 100 ml of either 20% ICI or optimized production medium lial fungi. containing 6% sunflower oil, 0.05% (NH ) SO , 1.5% corn-steep solids 4 2 4 Few data are available on the regulation of cpr gene expres- (Sigma-Aldrich), and 0.1% KH PO . Benzoate plate tests were per- 2 4 sion in lower eukaryotes. In the yeasts C. maltosa and Candida 2 formed on CM and Czapek Dox (CD) (Sigma-Aldrich) agar with 1 mM tropicalis, the assimilation of n-alkanes is catalyzed by P450 benzoate or without benzoate. monooxygenases. The addition of n-alkanes to the medium DNA and RNA Isolation—Genomic DNA was isolated from lyophi- lized mycelium as described by Doyle and Doyle (43).  DNA from resulted in strong induction of P450 and cpr gene (28, 35), and positive  clones was prepared according to Maniatis et al. (44). Plasmid co-regulation of the cpr and P450 genes has been reported for DNA was extracted using Genomed columns following the manufactur- Saccharomyces cerevisiae (36). In A. niger, the addition of ben- er’s protocol (Genomed, Bad Oeynhausen, Germany). RNA was isolated zoate increased the expression of the benzoate p-hydroxylase using the RNAgents total RNA isolation kit (Promega, Mannheim, (bphA) and cprA gene (14). Germany). The G. fujikuroi CPR would be expected to have a strong PCR—Degenerate primers CPR1 and CPR2 were designed by C. Wasmann (University of Arizona) on the basis of CLUSTAL alignment influence on GA biosynthesis since four P450 monooxygenases of fungal CPRs and kindly provided for cloning the G. fujikuroi cpr-Gf (P450-1-P450-4) are known to be involved in this pathway. gene. PCR reactions contained 25 ng of DNA, 10 ng of each primer, 0.2 Thus, loss of CPR activity should affect the rates of several mM dNTPs and 2 units of Taq polymerase (Red Taq, Sigma-Aldrich) in reactions in the pathway. Such an effect was found for the 50 l. PCR was carried out at 94 °C for 4 min followed by 30 cycles of UV-induced G. fujikuroi mutant SG138, which has lost most of 94 °C for 1 min, 54 °C for 1 min, and 72 °C for 1 min. The resulting the oxidation steps catalyzed by P450 monooxygenases (37). 0.5-kb PCR fragment was used as a probe for screening a genomic -DASH II library (Stratagene Europe, Amsterdam, The Netherlands) Because it is unlikely that all structural P450 monooxygenase of G. fujikuroi m567 at 65 °C. Sequences are: CPR1, 5-AAG YTG CAG genes are mutated by one UV treatment it is possible that the CCY CGC TAC TAY TCS ATC TC-3; CPR2, 5-CTT CCA YTC RTC mutant contains a lesion mutation in the cpr gene. This is CTT GTA SAR GAA RTC CTC-3. confirmed in the present paper, in which we report the cloning, For cloning cpr-Gf of SG138, four primer pairs were used to get sequencing, and targeted gene disruption of the G. fujikuroi cpr overlapping fragments after amplification: P138-1, 5-GTG GCC AAA gene (cpr-Gf) and its effect on GA production. We show by GTT CAT GAT TAG TGC-3;-2,5-TTG CGG ACC ATA GAG TTG TAG TGC-3; -3, 5-TCG CCA AGG AGG GTA AGA-GTC-3; -4, 5- sequence comparison that cpr-Gf contains a mutation in GCT GCC AGG GCG GTT CAT-3;-5,5-AAC CCC TAC ATT GCC SG138, and we were able to restore GA production to the CCT ATC G-3; -6, 5-TCG GCA ACC AAA GAA CAA GAG TG-3; -7, mutant by complementation not only with the cpr-Gf gene but 5-ACA GGC CCC CGC AAT AAG TA-3;-8,5-TGT CGG CAA GTC also with the heterologous A. niger cprA gene. We have also CAT GTC TAA GTG-3. Each fragment is about 850 bp long. For compared the regulation of cpr-Gf with that of the four GA RT-PCR, primers CPR-RT1 and CPR-RT2 were used to amplify frag- biosynthetic monooxygenase genes. ments of about 330 bp including a putative intron: primer CPR-RT1 (1), 5-CAA CCG AGG ATT TCA TGT ACC-3; CPR-RT2 (2), 5-CCC EXPERIMENTAL PROCEDURES TTG GCC TCA GAC ACC-3. For analysis of putative cpr knock-out transformants, the following Fungal Strains—Strains IMI58289 (Commonwealth Mycological In- diagnostic primers were used: for integration at the 5 region of cpr, stitute, Kew, UK) and m567 (Fungal Culture Collection, Weimar, CPR-DF1 (7), 5-CGG GGA TGG AGG CAA GAG AAT GAA-3, and Germany) are GA-producing wild-type strains of G. fujikuroi MP-C PUCH-P (8), 5-CCC TTG GCC TCA GAC ACC-3; for integration at the (anamorph F. fujikuroi). The GA-deficient mutant G. fujikuroi SG138 3 region of cpr, CPR-DF2 (9), 5-GAT CTA CAG ACT TGC TTC TGT (kindly provided by J. Avalos, University of Sevilla, Spain) strain was LGG-3, and PUCH-T (10), 5-TCA ACG CAT ATA GCG CTA GC-3. derived from IMI58289 via UV mutagenesis of spores (37). The gene replacement vector pcpr-Gf was constructed by cloning Bacterial Strains and Plasmids—Escherichia coli strain Top10 the PCR-amplified flanks into vector pUCH2– 8 (45). For amplifying the (Invitrogen) was used for plasmid propagation. Vector pUC19 was 5 and 3 flanks the following primers with introduced restriction sites used to clone DNA fragments carrying the G. fujikuroi cpr gene and were used: Dcpr-1 (3), 5-CGG GAA GTA CAA GGT ACC GTG CAA gene fragments. First, a 2.2-kb BglII fragment of -clone 2–1was AT-3 (underlines indicate the KpnI site); Dcpr-2 (4), 5-TCA ACG AGA cloned into BamHI-restricted pUCI9 (pCPR1A). Then a 3.8-kb XbaI TGT CGA CGT TTT TGT CC-3 (underlines indicate the SalI site); fragment was cloned into pUCI9/XbaI (pCPR1B). pCPR1B was cut Dcpr-3 (5), 5-TCC AAC TTC AAG CTT CCC TCG GAC-3 (underlines with NcoI/SphI. The derived 3-kb fragment was cloned into pCPR1A/ indicate the HindIII site); Dcpr-4 (6), 5-GAT AAC CAA AGA GCT CGT NcoI/SphI to produce vector pcpr-Gf, containing the entire cpr-Gf GGA CAG GT-3 (underlines indicate the SacI site). gene. The vector pNR1 was constructed by cloning the PstI/BamHI Screening the -DASH II Library—About 35,000 recombinant phages fragment of the Streptomyces noursei nat1 gene encoding the nourseo- from a  library prepared from genomic DNA of wild-type G. fujikuroi thricin acetyltransferase (38) into pBluescript II KS. The gene was m567 (46) were plated and transferred to Nylon N membranes (Am- transcribed under the control of the Aspergillus nidulans oliC pro- 2 ersham Biosciences). Hybridization was performed at high stringency moter (39) and terminated by the Botrytis cinerea tub1 terminator. (65 °C). The blots were washed in 2 SSC (1 SSC  0.15 M NaCl and For gene replacement experiments, fragments from the 5- and 3- 0.015 M sodium citrate), 0.1% SDS at 65 °C followed by a wash with noncoding regions of cpr were cloned into the vector pUCH2– 8 (40), 0.1 SSC, 0.1% SDS at 65 °C. Positive recombinant phages were used carrying the hygromycin B resistance marker. for a second round of plaque purification. Media and Culture Conditions—For DNA isolation, fungal strains Southern and Northern Blot Analysis—After digestion with restric- were grown in 100 ml of liquid CM optimized for Fusarium spp. (41) for tion endonucleases and electrophoresis, genomic or  DNA was trans- 3 days at 28 °C on a rotary shaker set at 200 rpm. The mycelium was ferred onto Hybond N filters (Amersham Biosciences). P-Labeled harvested by filtration through a sterile glass filter (G2, Schott Jena, probes were prepared using the random oligomer-primer method (44). Germany), washed with sterile distilled water, frozen in liquid nitrogen, Filters were hybridized at 65 or 56 °Cin5 Denhardt’s solution con- and lyophilized for 24 h. The lyophilized mycelia were ground to a fine taining 5% dextran sulfate (44). Filters were washed at the same powder with a mortar and pestle. For RNA isolation, fungal strains temperature used for hybridization in 2 saline/sodium phosphate/ were grown in 100, 20, or 0% ICI medium (42) containing 8% glucose, EDTA (SSPE), 0.1% SDS and 1 SSPE, 0.1% SDS. Northern blot hybridizations were accomplished by the method of Church and Gilbert (47). The G. fujikuroi rDNA gene was used as a J. van Kan, personal communication. control hybridization probe to confirm RNA transfer. The G. fujikuroi cpr Gene 25077 FIG.1. A, physical map and major sub- clones of gene cpr-Gf. The functional do- mains of the CPR-Gf enzyme and the po- sition of PCR primers are marked. 1, CPR-RT1; 2, CPR-RT2; 3, Dcpr-1; 4, Dcpr-2; 5, Dcpr-3; 6, Dcpr-4. B, strategy for construction of the gene replacement vector pcpr-Gf. 7, CPR-DF1; 8, PUCH-P; 9, CPR-DF2; 10, PUCH-T. H, HindIII; B, BamHI; X, XbaI; S, SalI; E, EcoRI; N, NcoI; Sa, SacII. Sequencing—DNA sequencing of recombinant plasmid clones was ously (8). Compounds were identified by comparison of their mass accomplished with an automatic sequencer LI-COR 4000 (MWG, spectra with those in a spectral library (49). For quantitative analysis of Mu ¨ nchen, Germany). The two strands of overlapping subclones ob- GAs aliquots of the extracts were spiked with [17- H ]GA internal tained from the genomic DNA clones were sequenced using the univer- standards and analyzing using a GCQ instrument (Thermo Electron sal and the reverse primers or specific oligonucleotides obtained from Corp.) as described previously (50), except chromatogram peak areas MWG Biotech (Munich, Germany). DNA and protein sequence align- were obtained from full scans. ments were done with DNA Star (Madison, WI). For phylogenetic anal- Plate Tests with Benzoate—For analysis of benzoate tolerance, yses, gaps were not considered, and corrections for multiple substitu- strains IMI58289, SG138, and transformants T20 and KT-1 were grown tions were applied. Trees were constructed with the program MegAlign for 6 days at 28 °C on CM and CD agar containing 1 mM benzoate. (DNA Star Madison, WI). Transformation of G. fujikuroi—The preparation of protoplasts and RESULTS the transformation procedure were as previously described (46). For Cloning and Targeted Gene Disruption of cpr-Gf—A frag- gene replacement, 10 protoplasts (50 l) of strain IMI58289 were transformed with 10 g of the KpnI/SacI fragment of the gene replace- ment of the cpr-Gf gene with the expected size of 500 bp was ment vector pcpr. For complementation of the mutant strain SG138 amplified by PCR using degenerate primers CPR1 and CPR2 with intact cpr genes, protoplasts were transformed with 10 gofthe derived from the FAD and NADPH binding domains, respec- circular complementation vector pcpr-Gf carrying the G. fujikuroi cpr tively (see Fig. 1A). The fragment exhibited a high degree of gene or pcprA with the cpr gene from A. niger (31). Both plasmids were homology with CPRs from other fungi and served as a probe for co-transformed with pAN7.1 (48), carrying the hygromycin resistance marker. screening the -DASH II genomic library of G. fujikuroi, MP-C. For complementation of the transformant cpr -T20 with the wild- Of the three genomic  clones isolated, one was used for isolat- type cpr-Gf gene, protoplasts were co-transformed with 10 g of each ing the putative full-length gene (Fig. 1A). About a 4000-bp the circular complementation vector pcpr-Gf and pNR1. Transformed sequence including 1000 bp of the 5-non-coding region was protoplasts were regenerated at 28 °C on complete regeneration agar obtained by sequencing in both directions. A 273-bp cDNA (0.7 M sucrose, 0.05% yeast extract, 0.1% (NH ) SO ) containing 120 4 2 4 g/ml hygromycin B (Calbiochem) or 100 g/ml nourseothricin (Werner fragment spanning a putative intron was generated by RT-PCR BioAgents, Jena, Germany) for 6 –7 days. For purification, single conid- using primers CPR-RT1 and CPR-RT2 (Fig. 1A). Comparison of ial cultures were obtained from hygromycin B- or nourseothricin-resist- the genomic and cDNA sequences confirmed the expected in- ant transformants and used for DNA isolation and Southern blot tron of 52 bp. The cpr-Gf gene has been deposited under the analysis. accession number AJ576025. Gibberellin Assays—GA and GA were analyzed by thin layer 3 4/7 chromatography on silica gel eluted with ethyl acetate/chloroform/ace- The deduced amino acid sequence of cpr-Gf was aligned with tic acid (60:40:5). The complete GA complement produced by the differ- known CPRs from taxonomically diverse species and allowed ent strains was determined by GC-MS analysis after extraction from the identification of all functional domains involved in the the culture fluid as already described (8), except that compounds were binding of the cofactors FMN, FAD, and NADPH and the P450 separated on a 30-m  0.32-mm  0.25-m HP-5 WCOT column (Agi- proteins (Fig. 1A). A phylogenetic tree comparing CPR-Gf with lent Technologies) and analyzed using a MAT95XP mass spectrometer (Thermo Electron Corp.) GC-MS conditions were as described previ- CPRs from other species indicates that it is most closely related 25078 The G. fujikuroi cpr Gene were then analyzed by TLC (Fig. 3C) and GC-MS (Table I). Transformants T20 and T49 as well as mutant SG138 do not produce GA ,GA , and GA , the last three products of the 4 7 3 pathway. Interestingly, the deletion mutants and SG138 show a reduction of growth rate on agar plates, which is more sig- nificant on minimal CD agar (Fig. 4C) than on CM (Fig. 4A). The similar characteristics for T20, T49 (data not shown), and SG138 as well as the described loss of P450-catalyzed oxidation steps (37) support our proposal that the latter strain is also affected in cpr-Gf. Identification of the Mutation in SG138 —To confirm our proposal that the UV treatment affected the cpr gene in the mutant SG138, the cpr gene from the mutant was amplified by four primer pairs to give four overlapping fragments, which were cloned, and three independent clones were sequenced in both directions. Comparison with the sequence of the wild-type gene copy confirmed a point mutation in the first position of the codon at amino acid position 627 from C to T resulting in a TGA FIG.2. Phylogenetic tree of cytochrome P450 reductases from stop codon instead of CGA for arginine. The truncated CPR fungi, plants, and animals, based on amino acid sequences. Ac- peptide is, therefore, 83 amino acids shorter than the wild-type cession numbers: A. niger CprA S38427; Arabidopsis thaliana ATR1 CPR enzyme. All CPR proteins analyzed so far contain a S21530; A. thaliana ATR2 S21531; Bombyx mori BAA95684; Cunning- NADPH binding domain consisting of three segments (32). This hamella echinulata AAF89959; C. maltosa N_CPR P50126; Cavia por- is also the case for CPR-Gf (Fig. 1A). In the mutant SG138 a cellus P37039; C. versicolor CPR BAB83588; C. tropicalis N_CPR P37201; Drosophila melanogaster Cpr-P1 NP_477158; Homo sapiens large part of the last segment of the NADPH binding domain is P16435; Musca domestica Q07994; Mus musculus NP_032924; P. chry- missing. sosporium CPR AAG31351; P. chrysosporium CPR2 AAG31350; Rattus Complementation of a Deletion Mutant with the Wild-type norvegicus AAA41683; S. cerevisiae NCPRNP_011908; Schizosaccharo- myces pombe CPR CAA2429; S pombe CPR2 T40056; cpr-Gf Gene—The deletion mutant T20 was co-transformed with the complementation vector pcpr-Gf carrying the wild- type cpr-Gf gene and vector pNR1, with the nourseothricin to CPRA of A.niger (61% identity) followed by the yeast and the resistance gene as selection marker (see “Experimental Proce- other fungal CPRs (Fig. 2). dures”). Two nourseothricin-resistant transformants, KT-1 and To determine the importance of the cpr-Gf gene for GA pro- KT-13, were analyzed for correct integration of the cpr-Gf gene. duction, a gene replacement vector was constructed as shown As shown in Southern blot (data not shown) and Northern blot in Fig. 1B. Two flanking sequences were amplified by PCR analysis (Fig. 5), only KT-1 showed multiple copies of the introducing KpnI and SalI (flank 1) and HindIII and SacI hybridizing wild-type gene and a high transcript level of the (flank 2) restriction sites into the oligonucleotides. The flank- correct size, whereas transformant KT-13 does not contain ing sequences were cloned into the corresponding sites of vector vector pcpr-Gf. Analysis of the GA concentrations showed al- pUCH2– 8 carrying the hygromycin resistance cassette produc- most full restoration of GA production. Analysis of the three ing vector pcpr-Gf (Fig. 1B). The 4.4-kb KpnI/SacI fragment of final products, GA ,GA , and GA , by GC-MS (Table I) dem- 4 7 3 the vector was used for transforming the wild-type IMI58289. onstrated that the activity of all four P450 monooxygenases 72 transformants were analyzed by PCR for homologous inte- was at least partially restored, resulting in production of the gration using the primer pairs 7 (CPR-DF1) and 8 (PUCH-P) normal GA pattern (Fig. 6) and formation of wild-type-like and primer pairs 9 (CPR-DF2) and 10 (PUCH-T) (Fig. 1B). amounts (or even more) of the final product gibberellic acid Three transformants revealed the expected diagnostic bands. (GA ). Furthermore, the growth rate of KT-1 on CM and CD After purification by single-spore isolation, these transfor- agar was comparable with that of the wild-type (Fig. 4, A and mants were analyzed by Southern blot analysis together with C). the wild-type IMI58289 and the putative cpr mutant SG138. Analysis of GA Intermediates in the cpr Mutants—The effect The HindIII-digested DNA was hybridized to flank 2 as probe of the cpr-Gf deletion on each of the P450-catalyzed steps in the (Fig. 1B). For two transformants, T20 and T49, the hybridizing GA biosynthetic pathway was investigated by determining the 2.5-kb wild-type band was replaced by a smaller (1.4 kb) and a full spectrum of intermediates in the cpr mutants (Fig. 6) and bigger (4.4 kb) band (Fig. 3A) due to a tandem integration of quantifying selected intermediates by GC-MS (Table I). The cpr two replacement cassettes into the cpr locus. The tandem in- deletion mutants do not produce GA ,GA ,orGA , the final tegration was confirmed by PCR with primers 8 (PUCH-P) and 3 4 7 products of the GA biosynthetic pathway, but only a very low 9 (CPR-DF2) (Fig. 1B) and sequencing of the fragment. Trans- amount of the non-hydroxylated intermediates GA and GA formant T57 integrated the fragment ectopically and still con- 15 24 together with high levels of ent-kaurene (Fig. 7). The level of tained the wild-type band. GAs in cpr is about 1% of that found in the wild-type strain The deletion of the wild-type copy of cpr in transformants IMI58289. These results indicate a very low activity of ent- T20 and T49 was confirmed by Northern blot analysis. The kaurene oxidase (P450 – 4) and low but significant activities for wild-type, transformant T57, and mutant SG138 contained a GA 7-oxidase (one of the activities of P450-1) and GA 20- transcript of about 2.1 kb, whereas T20 and T49 lost the entire oxidase (P450-2) in the absence of the P450 reductase. In con- coding region of cpr and did not show any cpr transcript trast, 3-hydroxylation, another activity of the P450-1 mo- (Fig. 3B). GA Production and Growth Characteristics of Deletion Mu- nooxygenase, was absent in cpr and, thus, would appear to have an absolute requirement for the P450 reductase. The tants and Strain SG138 —To determine the effect of cpr-Gf deletion on GA production, the wild-type strain, both knock-out spectrum of GAs found in SG138 (Table I) was similar to that in cpr, consistent with our demonstration of a mutation in the mutants, and the putative cpr mutant SG138 were cultivated for 7 days in the synthetic 20% ICI medium. The culture fluids P450 reductase in this strain. The requirement of the 13- The G. fujikuroi cpr Gene 25079 FIG.3. Analysis of gene replace- ment strains. A, Southern blot analysis of the wild-type strain IMI58289, UV mu- tant SG138, and three transformants. Genomic DNA was digested with HindIII and hybridized with the right flanking sequence of the replacement vector pcpr-Gf (see Fig. 1). B, Northern blot analysis. G. fujikuroi rDNA was used as the control. C, thin layer chromatography (TLC) of the wild-type IMI58289 and mu- tant strains after 7 days of cultivation in 20% ICI medium. TABLE I Concentration of GAs in culture filtrates of wild-type, cpr mutants, and cpr-complemented strains Results are expressed in g/ml  S.E.; n  3, except where indicated otherwise. Strain GA GA GA GA GA GA GA GA 3 1 7 4 9 15 24 25 IMI58289 (wild-type) 13.4  0.3 5.75  0.25 1.34  0.09 3.23  0.11 1.26  0.05 0.35  0.02 0.80  0.07 0.50  0.03 cpr-T20 0 0 0.004  0.001 0.017  0.004 0 0.64  0.02 0.17  0.01 0 cpr-cprGf (KT-1) 19.5  0.5 15.1  0.5 1.57  0.07 5.95  0.13 2.21  0.14 0.44  0.03 1.15  0.03 1.32  0.17 SG138 0 0 0.012  0.0005 0.014  0.002 0 0.37  0.02 0.14  0.004 0 b b SG138-cprGf 11.5  0.4 1.50  0.02 1.01  0.06 1.15  0.08 0.54  0.03 0.06  0.007 0.20  0.02 0.39  0.01 c d c c b c SG138-cprA-7 29.5  0.2 6.29  0.13 3.10  0.05 0.56  0.05 0.42  0.03 0.15  0.04 0.35  0.01 1.03  0.06 0.004 g/ml. n  2. n  6. n  5. hydroxylation reaction, catalyzed by P450-3, for the reductase ity of SG138 was also fully restored by complementation with was tested in the cpr mutants by incubating with [ C]GA and the reductase gene from A. niger (cpr-A), although A. niger is analyzing the products by high performance liquid chromatog- not able to produce any GAs (Table I). raphy and GC-MS. Formation of [ C]GA was reduced in the The Specificity of cpr-Gf for GA Biosynthesis Activity—As mutants relative to the wild-type (20% compared with 84%), part of an enquiry into whether or not cpr-Gf is specific for the indicating that the 13-hydroxylase has a partial, but not abso- P450s involved in GA biosynthesis we attempted to determine lute requirement for the P450 reductase. Complementation of if the cpr and GA monooxygenase genes were co-regulated. SG138 with cpr-Gf increased [ C]GA formation from Co-regulation has been reported, for example, for the A. niger [ C]GA to 56%. cprA and benzoate p-hydroxylase (bphA) genes (14). The GA profile and amounts in cpr and SG138 were re- Three of the four GA-specific P450 monooxygenase genes stored to those in the wild-type strain by complementation with (P450-1, P450-2, and P450-4) are known to be regulated by the the reductase gene from G. fujikuroi (Table I and Fig. 6). The general transcription factor AREA (51) and, thus, are highly major products synthesized by the complementation mutants expressed under nitrogen starvation conditions. Therefore, we were the 3-hydroxylated GAs GA and GA plus lower compared the expression pattern of these three P450 genes 3 1 amounts of GA and GA . Interestingly, the GA-producing abil- with that of cpr-Gf. Interestingly, the cpr-Gf gene is co-regu- 4 7 25080 The G. fujikuroi cpr Gene FIG.4. Growth of the cpr-deficient mutants is compromised. The wild-type IMI58289, mutants SG138 and T20, and the comple- mented strain KT-1 were grown on CM and CD agar with and without benzoate. A, SG138. B, IMI58289. C, T20. D, KT-1 FIG.5. Northern blot analysis of the wild-type IMI58289 and two transformants, KT-13 and KT-1, after transformation with the complementation vector pcpr-Gf. lated with P450-1, P450-2, and P450-4; high cpr-Gf transcrip- FIG.6. GC-MS analysis of culture filtrates of the wild-type tion levels were found under nitrogen starvation conditions but (IMI58289), cpr disruption mutant (CPR-T20), and line KT-1, in much less (though higher than for the monooxygenase genes) which T20 has been complemented with the G. fujikuroi CPR with high amounts of nitrogen (Fig. 8A). We investigated the gene. Total ion currents are shown for extracts as methyl esters trim- specificity of the interaction between CPR-Gf and the GA bio- ethylsilyl ethers. Components were identified by comparison of their mass spectra with published data (49) as follows: peak 1, ent-kaurene; synthetic monooxygenases by transforming the mutant SG138 peak 2, ent-kaurenoic acid; peak 3,GA peak 4,GA ; peak 5,GA ; peak 9; 25 24 with the cprA gene of A. niger. Ten hygromycin-resistant trans- 6,GA and 7-hydroxykaurenolide; peak 7,GA ; peak 8,GA ; peak 9, 14 4 7 formants were cultivated under GA production conditions and fujenoic acid, peak 10 GA ; peak 11,GA ; peak 12;GA isolactone; 13 36 3 analyzed for GA content. Three transformants, SG138-cprA-7, peak 13,7, 18-dihydroxykaurenolide; peak 14,GA ; peak 15,GA ; peak 1 3 16, ent-kaurenol; peak 17,GA . Unlabeled peaks are due to compounds -8, and -19, were able to produce GA (e.g. SG138-cprA-7, Table unrelated to GA biosynthesis. The peak at the same retention time as I), demonstrating that CPRA from the GA-non-producing fun- ent-kaurene in the KT-1 extract contains no ent-kaurene. gal species A. niger, which is described as an activator of the benzoate p-hydroxylase, is able to act as electron donor and activator of GA biosynthetic enzymes P450-1-P450-4 in G. fu- sensitivity to this compound due to the lost activation of the jikuroi. However, in contrast to cpr-Gf, cprA was expressed benzoate p-hydroxylase by CPR-Gf. On the other hand, comple- independently of the nitrogen condition in G. fujikuroi mentation of T20 with the wild-type cpr-Gf copy fully restored (Fig. 8B). the high resistance level for benzoate (Fig. 4, B and D). On the basis of these results we speculated that CPR-Gf may These results led us to anticipate induction of cpr-Gf gene also act in much more than the GA biosynthetic pathways. To expression by benzoate, as is the case for cprA in A. niger. The show this, we determined if CPR-Gf is involved in detoxifica- tion of benzoate in a similar way to CPRA in A. niger. We grew addition of benzoate to the medium significantly induced the cpr-Gf transcription level in G. fujikuroi, especially when 1 mM the wild-type strain IMI58289, two cpr-Gf mutants, SG138 and T20, as well as the complemented strain KT-1 on CM and CD benzoate was added (Fig. 9). Interestingly, with benzoate in the agar with or without benzoate. The growth patterns show very medium, cpr-Gf expression is no longer repressed by high clearly that mutation or deletion of cpr-Gf led to an extreme amounts of nitrogen. The G. fujikuroi cpr Gene 25081 FIG.8. A,Northern blot analysis of the wild-type IMI58289 and the mutant SG138 showing regulation of P450 and cpr gene expression by nitrogen. Strains were grown for 3 days in 10% ICI medium, washed, and transferred to 0% ICI (no nitrogen) or 100% ICI (high amounts of nitrogen) medium for 5 or 10 h. B, Northern blot analysis of transformant SG138-K1, carrying the A. niger cprA gene. FIG.7. GA biosynthetic pathways in G. fujikuroi indicating FIG.9. Northern blot analysis of the wild-type IMI58289 show- reactions affected in the cpr-Gf mutant. Products detected in the ing benzoate induction of cpr gene expression. The strain was culture are underlined, absent reactions are marked by an X, undetec- cultivated for 3 days in 10% ICI and then transferred to 0 or 100% ICI ted final products are in parentheses, and the proposed sequence of medium with 0.5 or 1 mM benzoate or without benzoate. reactions are shown with pointed arrows. Participation of CPR in reac- tions after the block at GA -aldehyde was not demonstrated experimentally. ent-7-hydroxykaurenoic acid, followed by ring contraction with the production of GA -aldehyde, its 3-hydroxylation to DISCUSSION GA -aldehyde, and oxidation to GA , in addition to several 14 14 Using degenerate PCR and genomic library screening we side reactions (6). P450-2 converts GA to GA and GA to 14 4 12 isolated a gene from G. fujikuroi with high homology with GA with the loss of carbon-20 (7). Finally, P450-3 is responsi- cytochrome P450 reductase (CPR) genes from other species, ble for the 13-hydroxylation of GA to GA , and, in a minor 7 3 particularly the gene from A. niger (cprA) with which it is 61% pathway, of GA to GA (8). Although not all P450-catalyzed 4 1 identical at the amino acid level. The G. fujikuroi cpr gene steps are completely blocked in the deletion mutants, they are (cpr-Gf) encodes a protein of 713 amino acids containing all the reduced in activity and some reactions, such as 3-hydroxyla- expected domains for binding the prosthetic factors FAD, FMN, tion and loss of C-20, are completely absent in cpr mutants. and NADPH as well as for P450s (32). Targeted disruption of Thus, CPR-Gf may act as electron donor of all four P450 cpr-Gf led to a very substantial loss of GA production, demon- monooxygenases. strating that the reductase is required for normal activity of The small amounts of the non-hydroxylated GA and GA , 15 24 P450s involved in GA biosynthesis. In G. fujikuroi, GA biosyn- and the high level of ent-kaurene found in cpr and SG138 thesis requires four P450 monooxygenases, which are respon- indicate low activities of ent-kaurene oxidase (P450-4), GA sible for more than 10 enzymatic steps. P450 – 4 catalyzes the 7-oxidase (P450-1), and 20-oxidase (P450-2) in the absence of three oxidation steps from ent-kaurene to ent-kaurenoic acid the reductase. The activity of 13-hydroxylase (P450-3), which (5). P450-1 catalyzes the oxidation of ent-kaurenoic acid to could not be assessed from the GA profiles of the cpr mutants 25082 The G. fujikuroi cpr Gene because its substrate is not produced, was assayed by incubat- but only one CPR-encoding gene, with a high degree of identity ing with [ C]GA and shown to have reduced activity in the to CPR-Gf. Therefore, it is likely that G. fujikuroi contains a absence of the reductase. The residual activities of these en- single cpr gene that interacts with each of the P450s. zymes indicate the participation of a second electron transport We analyzed the GA biosynthetic and cpr genes also in the closest related members of the G. fujikuroi species complex protein in GA biosynthesis that would supply electrons only to some of the reactions catalyzed by P450 monooxygenases and consisting of eight mating populations (MP-A to MP-H). Most of these species contain the complete GA gene cluster, but only with less efficiency than the P450 reductase. members of MP-C (F. fujikuroi) are able to produce GAs. The Analysis of the GA content of the UV-induced mutant SG138 loss of GA production capability is due to a set of mutations in suggests that it contains slightly higher P450 activities than the coding and 5-noncoding regions of the GA biosynthetic the cpr lines and may, thus, possess low CPR activity. The GA genes, resulting in an overall amino acid sequence identity of levels in SG138 are similar to cpr, but ent-kaurene does not only 84 –94% in the case of P450 – 4. In contrast to the dispen- accumulate, and traces of GA ,GA ,GA , and GA are pres- 4 7 15 24 sable GA pathway genes, the level of sequence identity between ent. SG138 contains a point mutation in the cpr gene that gives the CPR enzymes is about 98%. These results suggest the a truncated protein that lacks part of the NADPH binding importance of CPR for essential functions of cell metabolism. domain. This mutation in SG138 would, thus, reduce consider- Here we report on the first deletion of a cpr gene in a ably but may not abolish CPR activity. Complementation of filamentous fungus. So far, only in S. cerevisiae has the single cpr with the cpr-Gf gene fully restored GA synthesis, giving cpr gene been successfully deleted without dramatic influence 3-hydroxylated C -GAs (GA ,GA ,GA , and GA ) at similar 19 3 4 1 7 on viability (54), indicating that an alternative electron donor levels than in the wild-type strain. This demonstrates that the must exist. It was suggested that in S. cerevisiae, a cytochrome P450 reductase is the main electron donor to the four GA b (cyt b ) could act as a second important element in the 5 5 biosynthetic P450s in G. fujikuroi. The second electron trans- electron donating system. Deletion of cyt b in the wild type did port pathway is much less effective and can only partially not display a phenotype, whereas disruption of the gene in a compensate for the absence of CPR. cpr strain was lethal, demonstrating that both enzymes can Interestingly, the two activities of the multifunctional complement each other in mutants with single disruptions of P450-1 monooxygenase (6) differ in their dependence on CPR. cpr or b (54). However, there is no additional electron-donat- Although 3-hydroxylase depends absolutely on CPR and is ing system overcoming the double knock out. completely blocked in cpr, the 7-oxidase activity can obtain The exact mechanism by which cyt b interacts with P450 electrons from an alternative source and is still moderately reductase is not yet clear. Numerous studies show that P450 active in the deletion mutant. Our results agree with previous activity can be enhanced by the addition of cyt b in some, but findings about differences in nucleotide co-factor requirements not all reactions (55–57). Human, but not yeast cyt b can for the different activities of P450-1 (52). The GA 20-oxidase selectively augment the rate of steroid hormone hydroxylations (P450-2) (7) does not produce C -GAs by cleavage of C-20 in by more than 10-fold, but this stimulation requires CPR and the absence of the CPR. Instead, the C tricarboxylic acid occurs without electron transfer to or from cyt b (55). product GA was found in cpr together with GAs with inter- In petunia, the product of a cyt b gene, which is expressed mediate oxidation states (alcohol and aldehyde) at C-20 that do exclusively in the flowers, regulates the activity of two P450s not accumulate in the wild type. Production of the C -GAs, involved in the biosynthesis of anthocyanin pigments. Tar- thus, appears to be completely dependent on CPR in contrast to geted inactivation of the b gene resulted in a flower color tricarboxylic acid synthesis. If the rate of C-20 oxidation is change caused by reduction in activity of these two P450s, considerably reduced in the absence of CPR, this may result in but it did not affect other P450s (17). We suggest that in G. the accumulation of enzyme-bound intermediates, which can fujikuroi a cyt b might take over the function of CPR for be hydrolyzed to give the C -GAs detected (7). activation of P450s involved in primary metabolism and, to a Besides the dramatic effect on GA production, mutations in much lesser extent, some of those functions in the dispensa- cpr in the deletion mutants and SG138 also affected the growth ble GA biosynthetic pathway. rate on synthetic medium and to a lesser extent on CM, indi- A possible alternative to general CPRs and cyt b as electron cating that CPR probably acts as electron donor also for P450- donors was found in the fumonisin gene cluster of G. fujikuroi related pathways in primary metabolism, e.g. for metabolism of MP-A (Fusarium verticilloides); the open reading frame of the sterols and fatty acids. However, the effect on primary metab- fumonisin biosynthetic gene FUM6 consists of a P450 gene that olism, especially on CM, is not as strong as might be expected is fused to a cpr gene (12) in which the FMN, FAD, and NADPH if CPR-Gf were the only electron donor associated with P450s binding domains are arranged in the same order as in other in G. fujikuroi. It is also possible that the reduced growth rate CPRs. This unusual enzyme belongs to a family consisting of is due to the accumulation of toxic intermediates of disrupted another fungal and two bacterial enzymes, the Fusarium oxys- secondary metabolite pathways. porum fatty acid -hydroxylase (58), the Bacillus megaterium Although A. niger does not produce GAs and does not contain fatty acid hydroxylase P450 (59), and the Bacillus subtilis BM-3 GA-related P450s, the cprA gene, involved in benzoate detoxi- Yfn1 gene product (A69975). The fusion of a P450 and a CPR into fication in A. niger, fully restored the GA production capacity of one single enzyme (FUM6) is highly unusual for fungal second- the cpr mutant SG138. This result indicates that CPRs act ary metabolite genes. In all the other examples described so far, nonspecifically as general electron donors for P450 monooxy- only typical P450 monooxygenase genes are present in gene genases from different pathways. The recently completed ge- clusters for dispensable metabolites; no CPRs, either as single nome of the Basidiomycetes, P. chrysosporium, revealed the genes or fused to P450s, are present in such clusters (3, 10, 45, presence of only one CPR-encoding gene (CPR, EC 1.6.2.4) and 60). It is not yet clear if the fusion protein is specific for the P450s at least 123 cytochrome P450 monooxygenase genes (53). The in the fumonisin pathway or could complement the functions of genome sequences of Neurospora crassa (www-genome.wi.mit- the general reductase in other pathways. .edu/annotation/fungi/neurospora) and Fusarium graminea- Major functions of P450s in numerous organisms, including rum (www-genome.wi.mit.edu/cgi-bin/annotation/fusarium), a species of the same genus (Fusarium)as G. fujikuroi, revealed 44 and 40 cytochrome P450 monooxygenases, respectively (9), S. Malonek and B. Tudzynski, unpublished results. The G. fujikuroi cpr Gene 25083 Acknowledgments—Strain SG138 was kindly provided by J. Avalos fungi, are the metabolism of xenobiotic drugs and toxins and (University of Sevilla, Spain). We gratefully acknowledge Catherine the assimilation of long chain alkanes as well as the metabo- Wasmann (University of Arizona) for sharing the sequence of the CPR- lism of endogenous compounds, such as sterols and fatty acids specific degenerated primers. We also thank P. Punt (TNO Nutrition (cited in Ref. 16). For Fusarium moniliforme it was shown that and Food Research Institute, Zeist, The Netherlands) for providing the cprA gene and for critical reading of the manuscript. We thank B. Berns the fungus can oxidize propylbenzene and that this reaction for typing the manuscript and J. Schulte for technical assistance. needs molecular oxygen and NADPH as the preferential coen- zyme, suggesting a microsomal cytochrome P450 monooxygen- REFERENCES ase system that contained NADPH-cytochrome P450 reductase 1. Xu, J.-R., Yan, K., Dickman, M. B., and Leslie J. F. (1995) Mol. Plant-Microbe Interact. 8, 74 – 84 (61). Other Fusarium strains, like F. solani, are able to detoxify 2. O’Donnell, K., Cigelnik, E., and Nirenberg, H. I. (1998) Mycologia 90, 465– 493 plant phytoalexins, such as pisatin, by a cytochrome P450 3. Tudzynski, B., and Ho ¨ lter, K. (1998) Fungal Genet. Biol. 25, 157–170 4. Linnemannsto ¨ ns, P., Vo, T., Hedden, P., Gaskin, P., and Tudzynski, B. (1999) enzyme system (62). The loss of CPR would have dramatic Appl. Environ. Microbiol. 65, 2558 –2564 consequences for fungal survival in the environment and for its 5. Tudzynski, B., Hedden, P., Carrera, E., and Gaskin, P. (2001) Appl. Environ. ability to infect plant tissue. Microbiol. 67, 3514 –3522 6. Rojas, M. C., Hedden, P., Gaskin, P., and Tudzynski, B. (2001) Proc. Natl. Many of the P450s are substrate-inducible. To become func- Acad. Sci. U. S. A., 98, 5828 –5834 tionally active, they are dependent on the high electron donat- 7. Tudzynski, B., Rojas, M. C., Gaskin, P., and Hedden, P. (2002) J. Biol. Chem. 277, 21246 –21253 ing activity of CPR under the same induction conditions. Thus, 8. Tudzynski, B., Mihlan, M., Rojas, M. C., Gaskin, P., and Hedden, P. (2003) phenobarbital and numerous other drugs and chemicals induce J. Biol. Chem. 278, 28635–28643 9. Yoder, O. C., and Turgeon, B. G. (2001) Curr. Opin. Plant Biol. 4, 315–321 not only the expression of many drug- and steroid-metabolizing 10. Keller, N. P., and Hohn, T. M. (1997) Fungal Genet. Biol. 21, 17–29 P450s but also that of NADPH-dependent reductase gene in 11. Desjardins, A. E., Hohn, T. M., and McCormick, S. P. (1992) Mol. Plant- Microbe Interact. 5, 214 –222 liver tissue (15). In the subtropical plant Catharanthus roseus, 12. Seo, J.-A., Proctor, R. H., and Plattner, R. D. (2001) Fungal Genet. Biol. 34, where so far only one cpr gene has been identified, expression 155–165 of this gene is co-regulated with that of numerous P450 genes 13. Young, C., McMillan, L., Telfer, E., and Scott, B. (2001) Mol. Microbiol. 39, 754 –764 involved in the biosynthesis of terpenoid indole alkaloids, phe- 14. Van den Brink, J. M., Punt, P. J., van Gorcom, R. F. M., and van den Hondel, nylpropanoids, and in other defense-related compounds (63). C. A. M. J. J. (2000) Mol. Gen. Genet. 263, 601– 609 15. Waxman, D. J., and Azaroff, L. (1992) Biochem. J. 281, 577–592 For A. niger it was shown that expression of both the P450 gene 16. Van den Brink, J. M., van Gorcom, R. F. M., van den Hondel, C. A. M. J. J., and bphA, involved in benzoic acid hydroxylation, and cprA were Punt, P. J. (1998) Fungal Genet. Biol. 23, 1–17 17. de Vetten, N., ter Horst, J., van Schaik, H.-P., de Boer, A., Mol, J., and Koes, induced by the substrate, benzoate. Interestingly, the majority R. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 778 –783 of cprA transcripts after benzoate induction were of a larger 18. Donaldson, R. P., and Luster, D. G. (1991) Plant Physiol. 96, 669 – 674 19. Bollwell, G. P., Bozak, K., and Zimmerlin, A. (1994) Phytochemistry 37, size due to the use of four alternative transcription start points. 1491–1506 In addition, an upstream open reading frame was found (14). 20. Schuler, M. (1996) Crit. Rev. Plant Sci. 15, 235–284 In G. fujikuroi the GA biosynthetic pathway is under control 21. Backes, W. L, and Kelley, R. W. (2003) Pharmacol. Ther. 98, 221–233 22. Lesot, A., Hasenfratz, M. P., Batard, A., Durst, F., and Benveniste, I. (1995) of nitrogen metabolite repression. Nitrogen starvation results Plant Physiol. Biochem. 33, 751–757 in a manyfold increase in transcript level for three of the four 23. Urban, P., Mignotte, C., and Kazmaier, M. (1997) J. Biol. Chem. 272, 19176 –19186 P450s (P450-1, P450-2, and P450-4) and induction of cpr-Gf.To 24. Koopmann, E., and Hahlbrock, K. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, demonstrate that CPR-Gf acts as a general electron donor for 14954 –14959 25. Kunic, B., Truan, G., Breskvar, K., and Pompon, D. (2001) Mol. Genet. Genom- many P450s, we analyzed the expression of cpr-Gf with and ics 265, 930 –940 without benzoate. As in A. niger, the cpr-Gf gene was signifi- 26. Yabusaki, Y., Murakami, H., and Ohkawa, H. (1988) J. Biochem. (Tokyo) 103, cantly induced by benzoate, indicating the existence of a CPR- 1004 –1010 27. Sutter, T. R., and Loper, J. C. (1989) Biochem. Biophys. Res. Commun. 160, Gf-dependent benzoate p-hydroxylase in G. fujikuroi. This was 1257–1266 supported by the finding that cpr mutants are significantly 28. Sutter, T. R., Sanglard, D., and Loper, J. C. (1990) J. Biol. Chem. 265, 16428 –16436 less resistant to benzoate than the wild type and the comple- 29. Miles, J. S. (1992) Biochem. J. 287, 195–200 mented mutant. Interestingly, the 5-non-coding region of 30. Kargel, E., Menzel, R., Honeck, H., Vogel, F., Bohmer, A., and Schunck, W.-H. (1996) Yeast 12, 333–348 cpr-Gf contains two n-alkane-inducible sequence elements, CA- 31. Van den Brink, J. M., van Zeijl, C. J. M., Brons, J. F., van den Hondel, CAT, suggesting a role for CPR also in n-alkane hydroxylation C. A. M. J. J., and van Gorcom, R. F. M. (1995) DNA Cell Biol. 14, 719 –729 32. Yadav, J. S., and Loper, J. C. (2000a) Biochem. Biophys. Res. Commun. 268, (see Ref. 64). In contrast to the cprA of A. niger, benzoate- 345–353 responsive regions were not found in the G. fujikuroi gene. 33. Yadav, J. S., and Loper, J. C. (2000b) Curr. Genet. 37, 65–73 34. Ichinose, H., Wariishi, H., and Tanaka, H. (2002) Appl. Microbiol. Biotechnol. However, these sequence elements might be located at position 59, 658 – 664 1400 to 1600 bp upstream of ATG (14). 35. Ohkuma, M., Masuda, Y., Park, S. M., Ohtomo, R., Ohta, A., and Takagi, M. NADPH-dependent oxidoreductase is the common electron (1995) Biosci. Biotechnol. Biochem. 59, 1328 –1330 36. Turi, T. G., and Loper, J. C. (1992) J. Biol. Chem. 267, 2046 –2056 donor to multiple P450 monooxygenases in all eukaryotic or- 37. Barrero, A. F., Oltra, J. E., Cerda ´ -Olmedo, E. Avalos, J., and Justicia, J. (2001) ganisms. In most fungi and all animals tested so far, only a J. Nat. Prod. (Lloydia) 64, 222–225 38. Kru ¨ gel, H., Fiedler, G., Smith, C., and Baumberg, S. (1993) Gene (Amst.) 127, single cpr gene was identified, indicating involvement of single 127–131 CPR in a multitude of P450-related reactions. In contrast to F. 39. Ward, M., and Turner, G. (1986) Mol. Gen. Genet. 205, 331–338 40. Alexander, N. J., McCormick, S. P., and Hohn, T. M. (1999) Mol. Gen. Genet. verticilloides (G. fujikuroi MP-A), where a catalytically self- 261, 977–984 sufficient P450 gene, Fum6, was found, the GA gene cluster in 41. Pontecorvo, G. V., Roper, J. A, Hemmonns, L. M., Mac Donald, K. D., and G. fujikuroi MP-C consists of four P450 monooxygenase genes Buften, A. W. J. (1953) Adv. Genet. 141, 141–238 42. Geissman, T. A., Verbiscar, A. J., Phinney, B. O., and Cragg, G. (1966) Phy- not fused to cpr genes. Our results indicate that cpr-Gf, which tochemistry 5, 933–947 is not located in the GA gene cluster, encodes the main electron 43. Doyle, J. J., and Doyle, J. L. (1990) Focus (Idaho) 12, 13–15 44. Maniatis, T., Sambrook, J., and Fritsch, E. F. (1989) Molecular Cloning: A donor responsible not only for activation of all four GA-related Laboratory Manual, 2nd Ed., Cold Spring Harbor Laboratory Press, Cold P450 monooxygenases but also for benzoate p-hydroxylase and Spring Harbor Laboratory, NY 45. Alexander, N. J., Hohn, T. M., and McCormick, S. P. (1998) Appl. Environ. at least some P450s involved in primary metabolism. Cloning Microbiol. 64, 221–225 and characterization of the gene encoding the G. fujikuroi 46. Tudzynski, B., Mende, K., Weltring, K. M., Kinghorn, J. R., and Unkles, S. E. cytochrome b (1996) Microbiology 142, 533–539 might provide insights into the possible role of 47. Church, G. M., and Gilbert, W. (1984) Proc. Natl. Acad. Sci. U. S. A. 81, this enzyme as an alternative electron donator and its partic- 1991–1995 ipation in the functional complex of GA P450s. 48. Punt, P. J., Oliver, R. P., Dingemanse, M. A., Pouwels, P. H., and van den 25084 The G. fujikuroi cpr Gene Hondel, C. A. M. J. J. (1987) Gene (Amst.) 56, 117–124 F. P. (1996) J. Biol. Chem. 271, 27438 –27444 49. Gaskin, P., and MacMillan, J. (1992) GC-MS of the Gibberellins and Related 57. Yamazaki, H., Gillam, E. M. J., Dong, M. S., Johnson, W. W., Guengerich, Compounds: Methodology and a Library of Spectra, Cantock’s Enterprises, F. P., and Shimada, T. (1997) Arch. Biochem. Biophys. 342, 329 –337 Bristol, UK 58. Nakayama, N., Takemae, A., and Shoun, H. (1996) J. Biochem. (Tokyo) 119, 50. Coles, J. P., Phillips, A. L., Croker, S. J., Garcı ´a-Lepe, R., Lewis, M. J., and 435– 440 Hedden, P. (1999) Plant J. 17, 547–556 59. Ruettinger, R. T., Wen, L. P., and Fulco, A. J. (1989) J. Biol. Chem. 264, 51. Mihlan, M., Homann, V., Liu, T.-W. D., and Tudzynski, B. (2003) Mol. Micro- 10987–10995 biol. 47, 975–991 60. Hohn, T. M., Desjardins, A. E., and McCormick, S. P. (1995) Mol. Gen. Genet. 52. Urrutia, O., Hedden, P., and Rojas, M. C. (2001) Phytochemistry 56, 505–511 248, 95–102 53. Warrilow, A. G. S., Lamb, D. C., Kelly, D. E., and Kelly, S. L. (2002) Biochem. 61. Uzura, A., Suzuki, T., Katsuragi, T., and Tani, Y. (2001) J. Biosci. Bioeng. 92, Biophys. Res. Commun. 299, 189 –195 580 –584 54. Truan, G., Epinat, J.-C., Rougeulle, C., Cullin, C., and Pompon, D. (1994) Gene 62. Khan, R., Tan, R., Mariscal, A. G., and Straney, D. (2003) Mol. Microbiol. 49, (Amst.) 142, 123–127 117–130 55. Auchus, R. J., Lee, T. C., and Miller, W. L. (1998) J. Biol. Chem. 273, 63. Lopes Cardoso, M. I., Meijer, A. H. Rueb, S., Queiroz Machado, J., Memelink, 3158 –3165 J., and Hoge, J. H. C. (1997) Mol. Gen. Genet. 256, 674 – 681 56. Yamazaki, H., Johnson, W. W., Ueng, Y.-F., Shimada, T., and Guengerich, 64. Yadav, J. S, and Loper, J. C. (1999) Gene (Amst.) 226, 139 –146

Journal

Journal of Biological Chemistry – Unpaywall

Published: Jun 1, 2004

There are no references for this article.