Discovery of a gene cluster for the biosynthesis of novel cyclic peptide compound, KK-1, in Curvularia clavata

Abstract

KK-1, a cyclic depsipeptide with 10 residues produced by a filamentous fungus Curvularia clavata BAUA-2787, is a promising pesticide active compound with high activity against many plant pathogens, especially Botrytis cinerea. As a first step toward the future mass production of KK-1 through synthetic biological approaches, we aimed to identify the genes responsible for the KK-1 biosynthesis. To achieve this, we conducted whole genome sequencing and transcriptome analysis of C. clavata BAUA-2787 to predict the KK-1 biosynthetic gene cluster. We then generated the overexpression and deletion mutants for each cluster gene using our originally developed transformation system for this fungus, and analyzed the KK-1 production and the cluster gene expression levels to confirm their involvement in KK-1 biosynthesis. As a result of these, a region of approximately 71 kb was found, containing 10 open reading frames, which were co-induced during KK-1 production, as a biosynthetic gene cluster. These include kk1B, which encodes nonribosomal peptide synthetase with a domain structure that is consistent with the structural features of KK-1, and kk1F, which encodes a transcription factor. The overexpression of kk1F increased the expression of the entire cluster genes and, consequently, improved KK-1 production, whereas its deletion decreased the expression of the entire cluster genes and almost eliminated KK-1 production, demonstrating that the protein encoded by kk1F regulates the expressions of the other nine cluster genes cooperatively as the pathway-specific transcription factor. Furthermore, the deletion of each cluster gene caused a reduction in KK-1 productivity, indicating that each gene is involved in KK-1 production. The genes kk1A, kk1D, kk1H, and kk1I, which showed a significant decrease in KK-1 productivity due to deletion, were presumed to be directly involved in KK-1 structure formation, including the biosynthesis of the constituent residues. kk1C, kk1E, kk1G, and kk1J, which maintained a certain level of KK-1 productivity despite deletion, were possibly involved in promoting or assisting KK-1 production, such as extracellular transportation and the removal of aberrant units incorporated into the peptide chain.

Keywords: Curvularia clavata; antifungal agent; biosynthetic gene cluster; genome sequence; grey mold; natural fungicide; non-ribosomal peptide; pesticide.

Figure 1

The chemical structure of KK-1. The constituent residues are separated by dotted lines: Lac, lactic acid residue; Pip, pipecolic acid residue; Val, valine residue; Asp, aspartic acid residue; Ile, isoleucine residue; Gly, glycine residue; Tyr, tyrosine residue.

Figure 2

The domain structure of each non-ribosomal peptide synthetase that is possessed by Curvularia clavata BAUA-2787, which was deduced by antiSMASH: C, condensation domain; A, adenylation domain; PCP, peptidyl carrier protein domain; nMT, nitrogen methyl transferase; E, epimerization domain; KS, ketosynthase domain; AT, acyltransferase domain; DH, dehydratase domain; cMT, carbon methyltransferase; KR, ketoreductase domain; TD, terminal reductase domain; NAD, male sterility protein. Only the modular structure of NRPS that is encoded by Ccnps4-3-2 is consistent with the structural characteristics of KK-1.

Figure 3

(A) A fragmentation image of Ccnps4-3-2, namely kk1B, into three ORFs (Ccnps2, Ccnps3, and Ccnps4) using the obtained sequencing data of Curvularia clavata BAUA-2787. (B) The genetic map and predicted gene arrangement of the KK-1 biosynthetic gene cluster (see Table 2 ). The gene kk1B, encoding non-ribosomal peptide synthetase (NRPS), is shown as a white arrow, and kk1F, encoding the transcription factor, is shown as a black arrow.

Figure 4

KK-1 cluster detection by MIDDAS-M. The gene clusters whose expression levels increased at 26 ˚C, the KK-1-producing culture temperature condition, when compared to 37˚C, the non-producing culture temperature condition. The analysis was performed on the data that consisted of the manually connected sequence of the putative KK-1 cluster, which was presumed to have been fragmented into three sequences (see Figure 3 ) and is at the end of all the obtained gene data at the right end of the horizontal axis. The three detected peaks are indicated by (A–C, (A), a fragmented KK-1 cluster containing Ccnps4; (B) a fragmented KK-1 cluster containing Ccnps2; and (C) a manually connected KK-1 cluster of the three fragments.

Figure 5

The susceptibility of the Curvularia clavata BAUA-2787 protoplast to aureobasidin A (AbA) and hygromycin (Hyg). The concentration of each agent in the medium is shown in the pictures.

Figure 6

Agarose gel electrophoresis to confirm the introduction of pAUR316 into Curvularia clavata BAUA-2787. Lane 1, 1 kbp DNA ladder; lane 2, pAUR316 plasmid; lane 3, wild-type strain as the negative control; lane 4, non-transformant (negative result); lane 5-13, pAUR316 transformants; and lane 14, 100 bp DNA ladder.

Figure 7

Analysis of kk1F overexpression strain. (A) The KK-1 productivity of the wild-type and two kk1F overexpression strains (OXkk1F-1, OXkk1F-2) on day 3 and day 7 of the culture. The error bars represent standard deviations (three biological replicates). (B) The reads per kilobase of exon per million mapped reads (RPKM) expression values in log2 scale for the putative KK-1 biosynthetic gene cluster corresponding to the gene IDs from kk1A to kk1J that are shown as shading and the surrounding genes in the above three strains on day 2 and day 4 of the culture. The numbers (1) and (2) indicate biological replicates. The region containing Ccnps2, Ccnps3, and Ccnps4 corresponds to kk1B, which encodes non-ribosomal peptide synthetase.

Figure 8

Analysis of the kk1F deletion strain. (A) Biological assays of the antifungal activity in the culture supernatants of the wild-type and kk1F deletion strains (DELkk1F). The pictures show the Botrytis cinerea colonies grown against paper discs soaked with the culture supernatant of each strain. (B) The reads per kilobase of exon per million mapped reads (RPKM) expression values in log2 scale for the putative KK-1 biosynthetic gene cluster corresponding to the gene IDs from kk1A to kk1J that are shown as shading and the surrounding genes of the wild-type and two kk1F deletion strains (DELkk1F-1, DELkk1F-2) on day 3 of the culture. The numbers (1) and (2) indicate biological replicates. The region containing Ccnps2, Ccnps3, and Ccnps4 corresponds to kk1B, which encodes non-ribosomal peptide synthetase.

Figure 9

The relative KK-1 production of each cluster gene deletion strain when compared to that of the wild-type strain on day 7 of the culture. The abbreviations for the mutants are shown on the horizontal axis: kk1A deletion strain, DELkk1A; kk1B deletion strain, DELkk1B; kk1C deletion strain, DELkk1C; kk1D deletion strain, DELkk1D; kk1E deletion strain, DELkk1E; kk1F deletion strain, DELkk1F; kk1G deletion strain, DELkk1G; kk1H deletion strain, DELkk1H; kk1I deletion strain, DELkk1I; kk1J deletion strain, DELkk1J. The error bars represent standard deviations (three biological replicates). ND, not detected.

Figure 10

The HPLC-MS analysis of the culture extracts of the Curvularia clavata BAUA-2787 strains; (A) wild-type strain; (B) kk1C deletion strain (DELkk1C) as a representative of the KK-1 cluster gene deletion mutants except kk1A; and (C) kk1A deletion stain (DELkk1A). The extracted ion chromatograms are shown for KK-1 ([M+H+] = 1113) and its putative mono-demethylated metabolites (a, b, c, d and e, [M+H+] = 1099). The dotted lines are drawn at the peak intensity of 2.00e7 cps.

Figure 11

The proposed biosynthetic pathway for KK-1. (A) The proposed model for the KK-1 non-ribosomal peptide synthetase assembly line: C, condensation domain; A, adenylation domain; PCP, peptidyl carrier protein domain; nMT, nitrogen methyl transferase; CT, condensation-like domain. (B) The proposed pathway for the biosynthesis of non-amino acids or non-proteinogenic amino acids building blocks.