- Research article
- Open Access
Genomics-driven discovery of a biosynthetic gene cluster required for the synthesis of BII-Rafflesfungin from the fungus Phoma sp. F3723
- Swati Sinha†1Email author,
- Choy-Eng Nge†1,
- Chung Yan Leong1,
- Veronica Ng1,
- Sharon Crasta1,
- Mohammad Alfatah1,
- Falicia Goh1,
- Kia-Ngee Low1,
- Huibin Zhang2,
- Prakash Arumugam1,
- Alexander Lezhava2,
- Swaine L. Chen2, 3,
- Yoganathan Kanagasundaram1,
- Siew Bee Ng1,
- Frank Eisenhaber1, 4 and
- Birgit Eisenhaber1Email author
© The Author(s). 2019
- Received: 5 November 2018
- Accepted: 2 May 2019
- Published: 14 May 2019
Phomafungin is a recently reported broad spectrum antifungal compound but its biosynthetic pathway is unknown. We combed publicly available Phoma genomes but failed to find any putative biosynthetic gene cluster that could account for its biosynthesis.
Therefore, we sequenced the genome of one of our Phoma strains (F3723) previously identified as having antifungal activity in a high-throughput screen. We found a biosynthetic gene cluster that was predicted to synthesize a cyclic lipodepsipeptide that differs in the amino acid composition compared to Phomafungin. Antifungal activity guided isolation yielded a new compound, BII-Rafflesfungin, the structure of which was determined.
We describe the NRPS-t1PKS cluster ‘BIIRfg’ compatible with the synthesis of the cyclic lipodepsipeptide BII-Rafflesfungin [HMHDA-L-Ala-L-Glu-L-Asn-L-Ser-L-Ser-D-Ser-D-allo-Thr-Gly]. We report new Stachelhaus codes for Ala, Glu, Asn, Ser, Thr, and Gly. We propose a mechanism for BII-Rafflesfungin biosynthesis, which involves the formation of the lipid part by BIIRfg_PKS followed by activation and transfer of the lipid chain by a predicted AMP-ligase on to the first PCP domain of the BIIRfg_NRPS gene.
- Cyclic lipodepsipeptide
- Biosynthetic gene cluster
- Combined NRPS/PKS cluster
- NonRibosomal Peptide Synthetase (NRPS)
- Adenylation domain
- Condensation domain
- Phoma species
Despite the general reluctance and the slowly changing attitude of pharmaceutical and biotech industries to explore secondary metabolites of plants and microbes for pharmaceutical applications during the last two decades , more than a third of recently approved medicines are still natural products or have been derived from lead compounds found in living organisms [2–6]. In the field of antibacterial and antifungal compounds, inputs from natural product biology are particularly indispensable. Recent sequencing outputs from many microbes including bacteria and fungi support their potential role as a rich source pool for compounds with broad pharmacological relevance.
Nonribosomal peptides and polyketides represent a large class of natural products. Despite their immense structural and functional diversity, they are synthesized by strikingly similar multimodular enzymes called nonribosomal peptide synthetases (NRPSs) and polyketide synthases (PKSs), respectively [7–9]. The sequences of both types of enzymes, NRPSs and PKSs, consist of modules where each module is thought to be responsible for catalyzing the attachment of a specific substrate on-to the growing chain in an assembly-line like manner [7, 8]. Typically, amino acids (NRPS) or simple carboxylic acids (PKS) are the substrates added by one module. A module consists of essential (core) domains but it is possible that it harbors additional auxiliary domain(s). The immense structural diversity of nonribosomal peptides and polyketides can be achieved by varying the number and/or order of modules with different combinations of both core domains and auxiliary domains [7, 8].
In the case of NRPSs, a typical module has at least three core domains, an adenylation domain (A domain), a peptidyl carrier protein (PCP; also known as thiolation domain, i.e. T domain) and a condensation domain (C domain). The A domain selects and activates the cognate amino acid by adenylation [10, 11]. The activated amino acid adenylate is then transferred to a PCP, which transports the activated intermediate to a C domain . The PCP domain carries a phosphopantetheinyl at a conserved serine (Ser) residue, which supports the transportation of substrates between the active sites of the domains. The C domain finally catalyzes the formation of the peptide bond between the thioester group of the elongating peptide chain from the earlier module with the amino group of the current module. It has been inferred that the enzyme has an acceptor site for the nucleophile and a donor site for the electrophile [13–16].
For PKSs, a typical module contains an acyl carrier protein (ACP), a ketosynthase (KS), an acyltransferase (AT) domain and a thioesterase (Te) domain, which all together catalyse the linear extension of a polyketide intermediate by two carbon atoms [17–19]. The building block of a polyketide chain is usually an acyl-CoA or a carboxyacyl-CoA which is loaded on the AT domain. The AT domain transfers the activated building block chain to the neighbouring ACP domain . The ACP domain, like the PCP domain of NRPSs, is 4′-phosphopantetheinylated and transfers the intermediate to the active site (cysteine) of the upstream KS domain. The KS domain catalyses a Claisen-like condensation reaction by forming a carbon-carbon bond between the thioester carbonyl of the ACP-bound polyketide chain and the alpha carbon atom of the extending unit. This decarboxylative condensation step yields a ketide chain with a β-keto group [8, 21]. The β-keto group is optionally modified by additional auxiliary domains like ketoreductase (KS), dehydratase (DH), or enoyl reductase (ER) domains before the growing polyketide chain is released as a linear or cyclic product via the C-terminal thioesterase (Te; termination) domain [8, 17]. Of note, most of the fungal PKS belong to modular iterative type I-PKS, which is a single, giant protein consisting of only one module using the same set of domains of this module in an iterative manner to build the final polyketide [7, 21, 22].
In this study, we report the discovery of a new cyclic lipodepsipeptide variant, BII-Rafflesfungin, from a marine Phoma species. It structurally resembles Phomafungin but differs in the amino acid composition of the closed ring structural subunit; it also retains the expected antifungal activity. Analysis of the genome sequence identified a combined NRPS-t1PKS biosynthetic gene cluster consistent with a role in the biosynthesis of this new compound.
Sequencing and sequence annotation of the genome from Phoma strain F3723
The discovery of Phomafungin with distinct antifungal activity  without any identification of its biosynthetic genes or enzymes motivated us to analyse the publicly available Phoma genomes: Phoma herbarum (GCA_001599375.1) and, more recently, Phoma koolunga (GCA_004151575.1, GCA_004151145.1, GCA_004151165.1). Unfortunately, no NRPS/PKS was detected that could be speculated as the correct gene cluster for the synthesis of Phomafungin (for details see Additional File 1: Table S1). Because we did not have access to the original Phoma strain from which Phomafungin was isolated, we tested extracts derived from Phoma strains in a large strain collection  for antifungal activity and found that the extract of one strain, F3723, had growth inhibitory activity against Candida albicans (data not shown).
We used PacBio SMRT sequencing to determine the genome sequence of Phoma strain F3723. The 18S rRNA of F3723 is 99.8% identical with 18S rRNA of Phoma MJ76 (E-value = 0.0, Accession: HM590661) confirming the original species annotation in the library . This finding is further supported by the phylogenetic analysis based on the standard protein coding marker gene, beta-tubulin, where F3723 clusters together with all other Phoma species (Additional File 2: Figure S1).
Detailed analysis of the NRPS gene architecture
According to the antiSMASH output, the NRPS gene in the cluster consists of 27 domains organized in 10 modules (M); M1 = PCP1, M2 = C1-A1-PCP2, M3 = C2-A2-PCP3, M4 = C3-A3-PCP4, M5 = C4-A4-PCP5, M6 = C5-A5-PCP6, M7 = C6-A6-PCP7-E1, M8 = C7-A7-PCP8-E2, M9 = A8-PCP9, M10 = CT. The initiation module, M1, consists of a single PCP domain which is sufficient to accept the fatty acid intermediate for further elongation of the peptide chain by subsequent downstream domains [35–38]. Each of the modules M2-M9 is thought to be responsible for catalyzing the attachment of one amino acid to the growing chain of the compound, in total eight amino acids. In accordance with their functionality, modules M2-M9 are expected to be repetitive blocks of C-A-PCP domains. We observed that modules M2-M8 were complete but module 9 (M9) lacked the C domain. Modules 7 and 8 (M7 and M8) harbor additional E domains M7 = C6-A6-PCP7-E1 and M8 = C7-A7-PCP8-E2 besides the typical C-A-PCP domains. The termination module M10 consists of the terminal condensation domain CT which is believed to help in cyclization of the peptide as reported previously in different studies [37, 38]. Unlike bacterial NRPS systems, where the terminating domain is usually a thioesterase domain, fungi typically have this type of specialized CT domain [39–41].
Identification of specificity conferring codes for the adenylation domains reveals that the NRPS cluster is not compatible with the biosynthesis of Phomafungin
Stachelhaus codes for the adenylation domains (A1-A8) of the NRPS gene cluster. The residues are mapped based on positions according to GrsA-Phe numbering. The predicted activated AAs are listed in column 3
Residue positions according to GrsA Phe numbering
Towards the structure of the new compound BII-Rafflesfungin (2)
Since the extract derived from F3723 was previously shown to possess antifungal activity against Candida albicans, we used a C. albicans whole-cell assay (see Methods, “Antifungal assay”) to guide purification of the bioactive compound. This yielded a light yellowish oil, which we denote as BII-Rafflesfungin, with an ESIMS [M + H]+ peak at m/z 1002 and a molecular formula of C44H75N9O17 based on HRESIMS measurement. Notably, the mass of the new compound is 28 units smaller than that reported for Phomafungin, which would be consistent with the absence of two -CH2- groups .
1H (400 MHz) and 13C (100 MHz) NMR data of 2 in DMSO-d6a
δH (J in Hz)
1.20 d (7.1)
7.89 br d (6.9)
1.75 m, 1.93 m
2.22 m, 2.22 m
8.18 d (7.2)
4.50 dt (7.4, 5.6)
2.65 dd (15.5, 5.6), 2.73dd (15.5, 5.6)
8.12 br d (7.4)
7.03 br s, 7.54br s
4.11 q (5.4)
3.68 m, 3.68 m
8.15 br s
3.64 m, 3.64 m
8.21 br s
4.36 dt (7.8, 5.6)
3.65 m, 3.65 m
7.75br d (7.8)
1.07 d (6.3)
7.93br d (8.8)
3.78dd (17.3, 5.5), 3.87dd (17.3, 5.5)
7.96br t (5.5)
2.32 m, 2.35 m
1.03 m, 1.33 m
1.22 m, 1.31 m
1.22 m–1.28 m
1.24 m, 1.24 m
1.24 m, 1.29 m
0.86 t (6.6)
0.83 d (6.6)
To conclude, there is experimentally verified evidence that the Phoma species reported in the present study produces a new cyclic lipodepsipeptide that we named BII-Rafflesfungin. Its planar structure is as shown in (2) in Fig. 1. The related biosynthetic gene cluster was named as ‘BIIRfg’ and the NRPS gene responsible for the biosynthesis of the cyclic peptide core of BII-Rafflesfungin was named as ‘BIIRfg_NRPS’. The sequence of BIIRfg was submitted to Genbank (Accession number MK043052).
Revisiting the Stachelhaus codes of the BIIRfg NRPS adenylation domains: discovery of 6 new Stachelhaus codes
Even without knowing the exact sequence of BII-Rafflesfungin, we were able to predict some new Stachelhaus codes  (see Table 1) from the amino acid composition and the NRPS cluster sequence for BII-Rafflesfungin. The identical signatures for A4, A5, and A6 should code for serine as this is the only amino acid with three occurrences in the sequence. Further, the signature code for the A7 domain is ‘DAQTIMAI’ which is very similar to the code ‘DVQTVMAI’ for the serine-activating domain. Given that serine and threonine are structurally similar with polar uncharged side chains, we predicted that the A7 domain is a threonine-activating domain. In addition, we were able to predict A2 as a glutamic acid-activating domain based on similarity with the known codes . Also, the NRPSPredictor2  suggests the most probable Stachelhaus code to be Glu.
Based on the experimental structural elucidation of the new compound BII-Rafflesfungin, we were able to fill in all the gaps in Table 1 regarding the information about which adenylation domain selects/activates which amino acid. The predicted Stachelhaus codes of the adenylation domains A2 and A7 are in good agreement with the experimental findings. Most importantly, the three identical Stachelhaus codes for adenylation domains A4, A5 and A6 code for the same amino acid serine.
To our best knowledge, none of the new Stachelhaus codes found in our analysis for 6 amino acids (Ala, Glu, Asn, Ser, Thr, Gly; Table 1) was described before. Our result is fully supported by theoretical considerations and experimental findings.
Attempts towards the stereochemistry: sequence analysis of functional subtypes of condensation domains
In order to identify the subtypes of the eight condensation domains and to predict the stereochemistry of the amino acids, we carried out HMMER searches against the HMM models of DCL, LCL, starter and dual type of C domains. The HMM models were created from the alignments provided by Rausch et al. . Unfortunately, the result of this analysis was inconclusive because all the C domains were highly scored by the DCL HMM model (Additional file 5: Table S2). This result suggests that all amino acids seem to be D amino acids. However, there is a contradictory line of evidence. For further investigation, we performed a phylogenetic tree analysis of the functional distribution of all the C domains using MEGA7  (Maximum Likelihood (ML) tree). Surprisingly, all the C domains of the NRPS gene were found bundled with dual type C domains despite of being highly scored by the DCL type HMMER models (Additional file 6: Figure S8).
Of note, the NRPS gene harbors two epimerization domains (M7, Ser; M8, Thr) which typically epimerize the amino acid of the growing peptide chain into D configuration. Both domains seem to be functionally intact: We find the functionally important motif/active site HHxxxDxVSW  conserved in both sequences (Additional file 7: Figure S9). Interestingly, a HMMER search with the sequences of both epimerization domains E1(M7) and E2 (M8) against enzyme classification (EC) profiles [52, 53] resulted in significant hits to aspartate racemaces (EC 126.96.36.199) with E-values of 6.1 e-17 and 4.8 e-15, respectively. Therefore, we speculate that Ser7 and Thr 8 undergo stereoisomerization. But, to conclude, we cannot deduce the exact functional subtypes of the C domains of the NRPS gene and the stereochemistry of the amino acids from sequence-analytic considerations. This issue requires additional experimental clarification that is described below.
Experimental determination of the stereochemistry of BII-Rafflesfungin
In order to determine the absolute configuration of the amino acid residues, BII-Rafflesfungin (2) was hydrolyzed with 6 N HCl and the acid hydrolysate was derivatized with Nα-(2,4-dinitro-5-fluorophenyl)-L- alaninamide (L-FDAA, Marfey’s reagent). The LCMS analysis of L-FDAA of (2) in comparison with the corresponding derivatives of D- and L-amino acid standards (Marfey’s method , see also Methods “Acid hydrolysis and derivatization of BII-Rafflesfungin”) indicated the exclusive presence of L-Ala, D/L-Ser, D-allo-Thr, Gly, L-Asn and L-Glu (Additional file 8: Figure S10). In the case of Ser, both the D- and the L-forms were identified. The observed ratio of L-serine: D-serine was 2:1 in the extracted ion chromatogram (EIC) suggesting the presence of two L-serine and one D-serine amino acids (Additional file 8: Figure S10). Of note, the FDAA-Asn residue was determined by comparison with the retention time of FDAA-Asp instead of FDAA-Asn due to the conversion of asparagine to aspartic acid during acid hydrolysis. We did not determine the absolute configuration of HMHDA.
There is full agreement between the architecture of the BIIRfg_NRPS gene (Fig. 2b) and the experimental findings for the stereochemistry of BII-Rafflesfungin. The NRPS cluster contains three modules (M5, M6, M7) for catalyzing the attachment of the amino acid serine to the growing chain of the compound but only one of them (M8) harbors an epimerization (E) domain. This is a perfect match with the experimental finding that L-serine and D-serine were found in a ratio 2:1. The only other module that contains an epimerization domain is module M8 and this explains the finding of allo-Thr in the D configuration. All the other amino acid residues are found to be in the L configuration. To conclude, based on the combination of the experimental stereochemistry data and the analysis of the NRPS gene architecture, a cyclo-[HMHDA-L-Ala-L-Glu-L-Asn-L-Ser-L-Ser-D-Ser-D-allo-Thr-Gly] lipodepsipeptide structure (2) was established for BII-Rafflesfungin.
The BIIRfg gene cluster contains a single type I polyketide synthase (t1-PKS), an AMP-dependent ligase, and a type II thioesterase domain containing protein in addition to the BIIRfg NRPS gene
Genes located in and adjacent to the BIIRfg biosynthetic gene cluster. The predicted function of the genes is based on the in-house sequence annotation pipeline ANNOTATOR. The best blast hit is reported. The observed E-value and percent identity for each gene are shown in parentheses. The rows in bold show the genes thought to be involved in the biosynthesis of BII-Rafflesfungin
Best blast hit
(E-value, % Identity)
Nonribosomal peptide synthase
[ Fungal sp. 11,243 ]
[ Fusarium aywerte ]
Thioesterase type II (PF00975)
[ Fusarium langsethiae ]
Hypothetical RTA1 (PF04479) domain containing protein
FAD binding domain (PF01565) containing protein
Similar to heme oxygenase
MARVEL domain (PF01284) containing protein
Putative membrane transport protein (Major facilitator superfamily, PF07690)
AMP dependent ligase (PF00501)
[ Fusarium lansethiae ]
Hypothetical P-loop containing ABC transporter protein
Heme binding (PF00141)
To conclude, with a single type I polyketide synthase (t1-PKS), an AMP-dependent ligase, a type II thioesterase domain containing protein, and the described NRPS gene in place, the composition of the BIIRfg gene cluster is consistent with the biosynthesis of the cyclic lipodepsipeptide BII-Rafflesfungin.
BII-Rafflesfungin has anti-fungal activity
We tested the effect of BII-Rafflesfungin on the growth of Candida albicans strain (SC5314) using the Clinical and Laboratory Standards Institute (CLSI) guidelines . The widely used antifungal amphotericin B was used as a positive control. Exponentially-growing Candida albicans cells were treated with either BII-Rafflesfungin or amphotericin B at different concentrations and the growth as measured by OD600 nm was recorded after 48 h. Consistent with published data , amphotericin B completely inhibited the growth of C. albicans cells at 125 nM, BII-Rafflesfungin completely inhibited the growth of C. albicans cells at 32 μM (see Additional File 11: Figure S13).
Biological activity of BII_Rafflesfungin
Fungal strain/Cell line
Candida albicans ATCC 10231
Candida albicans ATCC 90028
Saccharomyces cerevisiae (BY4741)
Aspergillus fumigatus ATCC 46645
Aspergillus brasiliensis ATCC 16404
A549 lung carcinoma cells
HepG2 liver carcinoma cells
To determine whether BII-Rafflesfungin has cytocidal or cytostatic activity, we tested the ability of Candida albicans (SC5314) cells to recover after pre-treatment with either BII-Rafflesfungin (100 μM) or DMSO for 6 h. We used Amphotericin B (10 μM; a cytocidal inhibitor) and Itraconazole (100 μM; a cytostatic inhibitor) as controls. Treatment with all three compounds for 6 h completely inhibited the growth of yeast cells (Additional file 13: Figure S15 A). We then washed off the compound and transferred the cells onto drug-free YPD agar plates. As expected Itraconazole-treated cells were able to grow after their transfer to drug-free YPD agar plates (Additional file 13: Figure S15 B). Similar to cells treated with Amphotericin B, BII-Rafflesfungin-treated cells failed to grow on YPD agar plates indicating that BII-Rafflesfungin has cytocidal activity (Additional file 13: Figure S15 B).
We also evaluated the mammalian cytotoxicity of BII-Rafflesfungin against 2 established cell lines, A549 lung carcinoma cells and HepG2 liver carcinoma cells (Table 4; Fig. S14 (c)). BII-Rafflesfungin was cytotoxic against both cell lines with an IC50 of 16.5 μM and 13.8 μM, respectively.
To conclude, BII-Rafflesfungin has potent broad spectrum anti-fungal activity but moderate level of mammalian cytotoxicity.
With a continuous increase in the number of fungal genomes being sequenced, there is a substantial evidence about the abundance of various secondary metabolites which are synthesized by different NRPS/PKS gene clusters. Several of these fungal secondary metabolites are cyclic lipopeptides like destruxins , aureobasidine , beauvericine , apicidin  or echinocandins . We have now identified another cyclic depsipeptide in a fungal genome, BII-Rafflesfungin, which is a novel analogue of Phomafungin . We identified the NRPS-t1PKS gene cluster, which we name BIIRfg, responsible for the synthesis of BII-Rafflesfungin. The vicinity of BIIRfg_NRPS was explored to identify a polyketide synthase, BIIRfg_PKS gene, an AMP-dependent ligase, and a type II thioesterase (TEII) which completes the BIIRfg cluster for the biosynthesis of BII-Rafflesfungin. Of note, type II thioesterases are often found to be encoded within PKS and NRPS gene clusters; they fulfill diverse functions, including removal of aberrant residues blocking the megasynthase (repair function), and various roles in substrate selection or, release of intermediates and products [56, 57, 63].
Sequence analysis of BIIRfg identified ten modules consisting of 27 domains which were part of a single NRPS gene. The initial module M1 lacks the A domain as well as the C domain which is consistent with the other known NRPS clusters for lipopeptides like echinocandin (EcdA;AFT91378) [35, 37, 64], emericellamide (EasA; XP_660149)  and verlamelin (vlmS;AB862312) . It has been reported that, in such cases, the initial module accepts the activated fatty acid intermediate at the PCP domain for further elongation of the actual peptide by downstream modules .
At the outset, we had no independent information about the structure of BII-Rafflesfungin. Since the specificities of many of the A domains have been biochemically characterized [11, 46, 65] and several prediction algorithms have been implemented to predict the specificity determining codes of A domains [34, 46, 66–72], it is possible to analyse the amino acid sequences of some A domains in order to predict which amino acid would be activated by the corresponding A domain. Therefore, we have investigated the sequences of all the A domains and tried to infer the specificity of binding amino acids from the signature sequence encoded in the A domains. The sequence segments representing A domains are actually Acyl-CoA synthetase NRPS A domains and belong to the Luciferase superfamily of enzymes . Adenylation domains are organized two-partite: (i) a N-terminal core domain and (ii) a C-terminal sub-domain. Several characteristic conserved motifs of A domains (these are mA1:LTYxEL, mA2:LKAGxAYVPID, mA3:LAYxxYTSGTTGxPKG, mA4:FdxS, mA5:NxYGPTE, mA6:GELxIxGxGLARGYW, mA7:YKTGDQ, mA8:GrxDxQVKIRGxRVELEEVE, mA9:LpxYMIP and mA10:NGKIDR) have been well established in various studies [11, 74]. The positioning of the respective substrate α-amino acid is governed by the highly conserved aspartate residue in the A4 motif of the core domain and the lysine residue in the A10 motif of the sub-domain . Originally, it was established that the approximately 100 amino acid residues in the region between core motifs A4 and A5 have ~ 10 critical residues (in the case of the gramicidin S synthetase (GrsA; P0C061), these 10 residues are: D235, A236, W239, T278, I299, A301, A322, I330, C331, also including K517), which represent the signature sequence of an A domain and determine its substrate specificity. But later, Challis et al.  explained that Cys331 and Lys 517 can be excluded from the signature sequence as Cys331 points away from the specificity pocket in GrsA  while Lys 517 is strictly conserved and does not participate in substrate discrimination. Hence, we took only eight residues in consideration as the signature sequence for the adenylation domains. This set of eight amino acids present at such specific locations is commonly referred to as Stachelhaus code . The sequence positions of these codes are determined based on their proximity to the conserved sequence motifs mA4 and mA5.
The collinearity rule for NRPS genes implies that the order of different modules of an NRPS is coherent with the amino acid sequence of the corresponding peptide product . However certain exceptions/variations to this rule have already been reported [77, 78]. The Stachelhaus codes from the A domains’ sequence for BIIRfg_NRPS combined with the collinearity rule helped to predict some A domain – amino acid assignments. The presence of identical signature codes in A4, A5, and A6 coincides with the observation of three serine residues in BII-Rafflesfungin. Two other codes (for A2 and A7) were predicted based on a sequence similarity approach and these predictions were later confirmed by HRESIMS data.
the LCL (donor C acceptor) domain (it catalyses a peptide bond between two L-amino acids),
the DCL domain (it catalyses a peptide bond to link a L-amino acid to the elongation peptide ending with a D-amino acids),
the starter C Domain (it catalyses the acylation of the first amino acid with a β-hydroxyl-carboxylic acid) and,
the dual E/C domain (it catalyses the peptide bond formation and cyclization of cysteine (Cys), serine (Ser) and threonine (Thr) residues.
There are no clear sequence distinctions between LCD or DCD type of condensation domains . Despite being best scored by the DCL HMMs (Additional file 5: Table S2), the ML tree clusters the C domains from BIIRfg_NRPS with the dual type which however is very close to the DCL type. Similar pattern in clustering of DCL and dual type domains along with other types of C domain have been reported earlier  suggesting that these two types of domains are closely related. In fact, biochemical experiments had provided insights that dual C domains have catalytic roles for condensation and epimerization and act as DCL catalysts .
Although all the C domains in BIIRfg_NRPS cluster with the dual type, their amino acid sequences do not show the presence of the ‘HHI/LxxxxGD’ motif which is known to be present in the N terminal region of dual C domains along with the ‘HHxxxDG’ motif [49, 81]. The presence of these different core motifs in the C domains was first described by Crecy-Lagard et al.  and Marahiel et al.  and was subsequently updated by Rausch et al. . As can be seen from the alignment (Fig. 3), all the C domains (with the exception of C1, C7, C8 and C9 (CT)) harbor the conserved ‘HHxxxDG’ motif; C1 and C9 have a ‘SHxxxDG’ motif while C7 and C8 show the presence of an alanine residue in place of the usual glycine residue in a ‘SHxxxDA’ motif. The second His-residue was found conserved in all these condensation domains which is known to be essential for the catalytic function of these domains [83, 84].
Based on all the genomic and experimental information, it is still not trivial to predict the stereochemistry of the amino acids in the final peptide. Nevertheless, it has been reported that traditionally E domains are associated with the modules which add D-amino acids in final peptides like surfactin , verlamelin  and other cyclic lipopeptides like bacillomycin, locillomycin and fengycin . The epimerization domains in the BIIRfg_NRPS gene observed here also harbor the same ‘HHxxxDxVSW’ motif (where the second histidine is functionally important) [49, 79, 87, 88] (see alignment in Additional File 7: Figure S9). In the case of BII-Rafflesfungin, we were able to show that Ser and allo-Thr that are recognized by A6 (M7) and A7 (M8) are in D-configuration; i.e., there are D-Ser6 and D-allo-Thr7 in the peptide product.
Biosynthesis of BII-Rafflesfungin
Mechanism of action of BII-Rafflesfungin
BII-Rafflesfungin displayed antifungal activity against Candida albicans, Saccharomyces cerevisiae and Aspergillus fumigatus. BII-Rafflesfungin’s cytotoxic mode of inhibition suggests that it perturbs the membrane integrity of yeast cells. Similar mode of action was also proposed for Phomafungin, based on its C. albicans fitness test profile . The long hydrophobic chain may facilitate its insertion into the plasma membrane and the polar cyclic peptide unit might affect the membrane structure. For Phomafungin, it was speculated that the primary effect on the plasma membrane is affected by changes in sphingolipid content and Ca2+ concentration . Interestingly, cyclic depsipeptides Syringomycin and Syringopeptins from Pseudomonas inhibit the growth of yeast cells by binding to the plasma membrane [90–92]. Another cyclic depsipeptide, Aureobasidin, produced by the fungus Aureobasidium pullulan affects membrane structure by inhibiting inositol phosphorylceramide (IPC, sphingolipid) synthase, a key enzyme involved in sphingolipid biosynthesis . Chemogenomic profiling, a fitness-based assay involving barcoded deletion strains in yeast, could give insights into BII-Rafflesfungin’s mode-of -action .
We have discovered a biosynthetic gene cluster predicted to synthesize BII-Rafflesfungin, which is a novel analogue of the known compound Phomafungin. The two compounds have the same amino acid composition with the exception of the presence of two homo-serine residues in Phomafungin instead of two common serines in BII-Rafflesfungin. This is the first study stating the presence of a NRPS-t1PKS gene cluster in Phoma sp for the synthesis of a novel analogue of Phomafungin. The compound BII-Rafflesfungin is biologically important as it shows antifungal activity.
The NRPS-t1PKS cluster ‘BIIRfg’ was found to be complete by the software tool antiSMASH except one C domain in the second last module of BIIRfg_NRPS gene and one ACP domain in the BIIRfg_PKS gene which were later identified by in-depth sequence analysis. The order of the amino acids in the final peptide product was experimentally verified using NMR studies and novel specificity conferring codes were predicted. Further studies using the available biochemical assays [95–99] to confirm the amino acid specificity for the A domains would be insightful. The mechanism of synthesis of BII-Rafflesfungin is proposed which involves the formation of the lipid part by BIIRfg_PKS followed by activation and transfer of the lipid chain by the predicted AMP-ligase on to the first PCP domain of BIIRfg_NRPS gene. This initiates the peptide synthesis by BIIRfg_NRPS gene which successively add different amino acids to the growing peptide and in the end releases the final cyclic lipodepsipeptide, BII-Rafflesfungin.
The fungal strain was isolated from a black sea cucumber near the coastal area of Raffles Marina Clubhouse, Singapore. The strain was initially numbered as F3723. The collected black sea cucumber was blended and particles of size 105–210 μm were resuspended in sterile water. Serial dilutions of this suspension were then plated on marine agar supplemented with 50 mg/L tetracycline and allowed to grow at 24 °C for up to 30 days. Fungal colonies growing on the isolation plates were sub-cultured onto marine agar to obtain pure cultures. F3723 was maintained as frozen mycelia in 10% glycerol at -80 °C.
DNA isolation and sequencing
F3723 was grown for 5 days at 24 °C in YMG media (10 g/L malt extract, 4 g/L yeast extract and 4 g/L glucose), and the mycelia were harvested by centrifugation at 4000 g for 10 min. High-quality and molecular weight genomic DNA (> 20 kb) was obtained using the modified CTAB method . Cells were re-suspended in 2X CTAB buffer containing 2% hexadecyltrimethylammonium bromide (CTAB; Sigma), 1.5 M NaCl, 25 mM EDTA, 100 mM Tris-HCl pH 8.8 and 0.1% polyvinylpyrrolidone (Sigma) and homogenised. Proteins were then removed using 1% SDS and proteinase K (25 mg/mL) before treated with RNaseA (10 mg/mL) for the removal of RNA from the total nucleic acids. The DNA was further purified with 25:24:1 phenol chloroform isoamyl alcohol (Sigma), followed by 24:1 chloroform isoamyl alcohol (Sigma). Extracted DNA was re-suspended in 50 μL 10 mM TE buffer pH 8.0 and additional purification step was carried out using Mag-Bind® RxnPure Plus (OMEGA bio-tek) as per manufacturer’s protocol. The quality of DNA was verified using a NanoDropTm 2000 spectrophotometer (Thermo Scientific), Qubit™fluorometric quantitation with Qubit™ 3.0 Fluorometer (Thermo Scientific) and agarose gel electrophoresis. Purified genomic DNA was sheared to approximately 20 kb using a g-Tube (Covaris). A SMRTbell library was prepared according to manufacturer’s instructions, loaded with a MagBead bound library protocol onto two SMRTCells at 0.125 and 0.3 nM, and sequenced using the P5-C3 chemistry on the PacBio RSII instrument (Pacific Biosciences) with a 180 min movie time. Two additional SMRTCells were run at 0.2 and 0.4 nM and sequenced using the P6-C4 chemistry on the PacBio RSII instrument with a 240 min movie time. De novo assembly was performed with the Hierarchical Genome Assembly Process 3 (HGAP3)  in the SMRT Analysis suite (version 2.3) using all default parameters. In total, 288,789 reads were collected, with a mean read length of 13,441 bp, giving a total of 3,881,628,628 bp of raw sequence. De novo assembly of the raw sequence resulted in 122 polished contigs. The total length of these 122 assembled contigs was 35,411,911 bp; the largest contig size was 2,158,956 bp, and the N50 contig length was 978,069 bp. Given the final genome assembly size, therefore, the overall sequencing coverage was ~ 110 × .
BII-Rafflesfungin: extraction and isolation
The culture broths (160 × 50 mL, total 8 L) of Phoma (F3723) were combined and centrifuged to separate the supernatant and the mycelia. The combined mycelia were freeze-dried, extracted two times with MeOH (4 L), filtered and, then, were evaporated to dryness using rotary evaporation. One litre of water was added to the dried methanolic extract (94 g) and charged to a 10 cm × 8 cm Sepra C18-E (50 μm, 65A, phenomenex) column. The column was eluted by isocratic gradient of 20, 50, 80, and 100% aqueous methanol. Each fraction was submitted to antifungal activity testing. The antifungal activity was found to be concentrated in the 80% aqueous methanol fraction, which was then concentrated under reduced pressure and yielded 0.25 g of a partially enriched fraction. The 80% aqueous methanol dried extract was dissolved in 2.5 mL of DMSO and separated by C18 reversed-phase preparative HPLC (solvent A: H2O + 0.1% HCOOH, solvent B: ACN + 0.1% HCOOH; flow rate: 30 mL/min, gradient conditions: 65:35 isocratic for 5 min; 35 to 70% of solvent B over 85 min, followed by 70 to 100% of solvent B over 10 min, and finally isocratic at 100% solvent B for 10 min to give BII-Rafflesfungin (2, 10 mg).
Chemical structural data
The 1H, 13C, HSQC, COSY, and HMBC spectra of the compound are provided in Additional file 4: Figures. S3 to S7.
BII-Rafflesfungin (2) yellowish oil; [α]D + 65.3 (c 0.2, MeOH); UV (MeOH) λmax (log ε) end absorption nm; HRESIMS m/z 1002.5371 (calcd for C44H75N9O17 + H, 1002.5354); 1H and 13C NMR data, see Table 2.
General experimental procedures
JASCO P-2000 digital polarimeter was used to record the optical rotations while GE Healthcare Ultrospec 9000 spectrophotometer was used to obtain the UV spectra.
Bruker DRX-400 NMR spectrometer with Cryoprobe was used to collect the NMR spectra. 5-mm BBI (1H, G-COSY, multiplicity-edited G-HSQC, and G-HMBC spectra) or BBO (13C spectra) probe heads equipped with z-gradients were used.
Agilent 1260 Infinity Preparative-Scale LC/MS Purification System and Agilent 6130B single quadrupole mass spectrometer for LC and LC/MS Systems were used to perform the preparative HPLC analysis.
Agilent UHPLC 1290 Infinity coupled to Agilent 6540 accurate-mass quadrupole time-of-flight (QTOF) mass spectrometer which was equipped with a splitter and an ESI source were used to acquire the HRESIMS and MS/MS spectra. For over 15 min, under standard gradient condition of 100% water with 0.1% formic acid to 100% acetonitrile with 0.1% formic acid, the analysis was performed with a C18 4.6 × 75 mm, 2.7 μm column at flowrate of 2 mL/min. The operating parameters for QTOF were the same as in .
Nα-(2,4-Dinitro-5-fluorophenyl)-L-alaninamide (L-FDAA) and amino acid standards were purchased from Sigma Aldrich except D-allo-threonine which was purchased from Chem Cruz. Aspartic acid and glutamic acid were converted from asparagine and glutamine (Sigma Aldrich), respectively.
Hydrolysis of BII-Rafflesfungin
To a stirred solution of BII-Rafflesfungin (2, 1 mg) and 1 mL of distilled water, 1 mL of 0.01 M NaOH was added. The mixture was stirred for 19 h at room temperature. The reaction mixture was charged on a small column of C-18 silica gel and washed with water, and the linear peptide was then eluted with MeOH. The MeOH fraction was concentrated and dried using rotary evaporator to yield 0.8 mg of the opened form of the linear peptide, as light yellowish oil. HRESIMS m/z 1020.5470 (calcd for C44H77N9O18 + H, 1020.5459). HRESIMS/MS: m/z 945.5151 (calcd for C42H73N8O16, 945.5139), 844.4669 (calcd for C38H66N7O14, 844.4662), 757.4343 (calcd for C35H61N6O12, 757.4342), 681.2686 (calcd for C24H41N8O15, 681.2686), 670.4028 (calcd for C32H56N5O10, 670.4022), 583.3701 (calcd for C29H51N4O8, 583.3701), 552.2255 (calcd for C19H34N7O12, 552.2260), 469.3270 (calcd for C25H45N2O6, 469.3272), 438.1828 (calcd for C15H28N5O10, 438.1831), 340.2844 (calcd for C20H38NO3, 340.2846).
Acid hydrolysis and derivatization of BII-Rafflesfungin
BII-Rafflesfungin (0.4 mg) was dissolved in 0.5 mL of 6 N HCl and heated at 100 °C for 8 h. The reaction mixture was cooled to room temperature and evaporated to dryness. The dried acid hydrolysate was added in 100 μL of H2O, 100 μL of 1% Nα-(2,4-Dinitro-5-fluorophenyl)-L-alaninamide (FDAA) in acetone, and 40 μL of 1 M sodium bicarbonate. The mixture was heated at 40 °C, with stirring for 1 h. The reaction mixture was then cooled to room temperature, neutralized with 20 μL of 2 M HCl, and dried in vacuo. The residue was dissolved in H2O (200 μL) and centrifuged to remove insoluble material before LCMS analysis [Zorbax Eclipse Plus 2.1 × 50 mm 1.8 μm, elution with 20 min linear gradient of 10–45% of solvent B (ACN 0.1% FA)]. The retention time (tR, min) for the standard amino acids DAA-amino acid derivatives were as follows: L-alanine (8.661), D-alanine (10.377), L-serine (6.379), D-serine (6.660), glycine (7.922), L-glutamic acid (7.831), D-glutamic acid (8.807), L-aspartic acid (7.100), D-aspartic acid (7.900), L-threonine (6.775), D-threonine (8.816), L-allo-threonine (6.941), D-allo-threonine (7.845) (Additional file 8: Figure S10).
Species annotation and biosynthetic cluster detection
Blast  based taxonomic analysis of the 18S rRNA was done which supports the species annotation of F3723. In addition, the beta-tubulin gene from F3723 was scanned against NT database using a blastn search. The most similar sequences were selected and incorporated into a multiple sequence alignment using ClustalO . A maximum likelihood tree was generated from the alignment using Mega7 with JTT matrix under discrete gamma distribution with four categories .
The assembled Phoma genome was analysed by a locally installed antiSMASH (version 3.0.3) [34, 104]. Protein coding genes were predicted by using AUGUSTUS (version 3.2.2, default parameters with Fusarium graminearum as training set)  and GeneMark-ES (version4.10, default parameters) [31, 32]. The predicted protein sets were further analysed by ANNOTATOR tools . Manual annotation efforts were helpful to detect missing function predictions.
Phylogenetic analysis of the C domains
All the sequences of the different subtypes of C domains were obtained from the NaPDoS server . The amino acid sequences of all C domains from BIIRfg cluster were aligned with the NaPDos dataset using MUSCLE . The ML tree was constructed using Mega7 with JTT matrix under discrete gamma distribution with four categories .
The Minimal Inhibitory Concentration (MIC) of C. albicans (SC5314) cells was determined using the CLSI guidelines . Cells were treated with various concentrations (0 μM – 64 μM, see Additional file 11: Figure S13) of BII-Rafflesfungin and Amphotericin B (0 nM – 1000 nM, Fig. S13) in RPMI 1640 medium in duplicates in a 96-well microplate. The microplate was photographed after 48 h of incubation at 35 °C. The least concentration of the compound that caused an optically clear well was computed as its MIC.
Antifungal activity was also determined against five strains: three yeasts, Candida albicans ATCC 10231, Candida albicans ATCC 90028 and Saccharomyces cerevisiae BY4741; and two moulds, Aspergillus brasiliensis ATCC 16404 and Aspergillus fumigatus ATCC 46645. The activity was evaluated by determining the IC50 value using a rapid microbroth dilution method in 96- or 384-well microplate format . Logarithmically growing yeast (Candida albicans/ Saccharomyces cerevisiae) cells in Sabouraud Dextrose Broth (SDB) medium were treated with various concentrations (0, 0.47, 0.94, 1.88, 3.75, 7.5, 15 and 30 μM) of BII-Rafflesfungin in triplicates. Cultures of Candida albicans strains ATCC 10231 and ATCC 90028 were incubated at 25 °C and 35 °C, respectively. Saccharomyces cerevisiae cultures were incubated at 30 °C. The growth (OD600nm) was recorded after 24 h incubation.
Frozen stocks of Aspergillus conidial suspensions were diluted in SDB medium to final concentrations of approximately 2.5 × 103 spores/mL (ATCC 16404) or 2.5 × 104 spores/mL (ATCC 46645). These conidial suspensions were then added to various concentrations (3 nM to 100 μM) of BII-Rafflesfungin in 384-well microplates in triplicates. The microplates were subsequently incubated for 72 h at 24 °C prior to OD600nm measurement.
For the cytocidal/cytostatic activity test, C. albicans (SC5314) cells at a starting OD600 nm of 0.6 were incubated with either DMSO or BII-Rafflesfungin (100 μM) or Amphotericin B (10 μM) or Itraconazole (100 μM) in duplicate in YPD medium at 30 °C with shaking (220 rpm). OD600nm values of the cultures after 6 h were obtained. Subsequently, the cells were washed twice, normalized and diluted in YPD medium. Equal volumes (3 μl) were spotted on YPD agar plates. Growth was recorded after incubation of the plates at 30 °C for 2 days.
Cytotoxicity of BII-Rafflesfungin on the A549 human lung carcinoma cell line and the liver carcinoma cell line HepG2 were determined as described before . A549 cells were seeded at 1500 cells per well, and HepG2 cells were seeded at 2500 cells per well in a 384-well microplate. Cells were treated with a series of BII-Rafflesfungin concentrations ranging from 0.8 μM to 100 μM and incubated for 72 h at 37 °C in the presence of 5% CO2. Cytotoxic effect of the compound was measured using the PrestoBlue™ cell viability reagent (Life Technologies). Following incubation of the microplates with the dye for 2 h, the fluorescence reading (Excitation / Emission: 560 nm /590 nm) was recorded using the Tecan Infinite M1000 Pro reader.
The authors acknowledge financial support from Agency for Science, Technology and Research (A*STAR) Singapore but the funding body did not participate in the design of the study and collection, analysis and interpretation of data and in writing the manuscript.
Availability of data and materials
The datasets generated and analysed during the current study are available in GenBank (Accession Number MK043052). Regarding the availability of the strain, please contact Ng Siew Bee (email@example.com).
FG, VN, CYL and SBN cultivated the Phoma sp., isolated the DNA for sequencing and prepared the extract from Phoma sp. fermentation. HZ, AL, and SLC performed sequencing and assembly. CEN, KNL and YK isolated the BII-Rafflesfungin compound from the extract and defined its chemical structure. MA, SC, SBN and PA performed the antifungal assay. SC and SBN tested the cytotoxicity. SS, FE and BE performed the bioinformatics analysis. SBN, YK, PA, FE and BE were involved in the project design. SS, CEN, SLC, PA, YK, SC, SBN, FE and BE took part in writing the manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Harvey AL, Edrada-Ebel R, Quinn RJ. The re-emergence of natural products for drug discovery in the genomics era. Nat Rev Drug Discov. 2015;14:111–29.PubMedView ArticlePubMed CentralGoogle Scholar
- Cragg GM, Grothaus PG, Newman DJ. New horizons for old drugs and drug leads. J Nat Prod. 2014;77:703–23.PubMedView ArticlePubMed CentralGoogle Scholar
- Newman DJ, Cragg GM. Natural products as sources of new drugs over the 30 years from 1981 to 2010. J Nat Prod. 2012;75:311–35.PubMedPubMed CentralView ArticleGoogle Scholar
- Newman DJ, Cragg GM. Drugs and drug candidates from marine sources: an assessment of the current "state of play". Planta Med. 2016;82:775–89.PubMedView ArticlePubMed CentralGoogle Scholar
- Ng SB, Kanagasundaram Y, Fan H, Arumugam P, Eisenhaber B, Eisenhaber F. The 160K natural organism library, a unique resource for natural products research. Nat Biotechnol. 2018;36:570–3.PubMedView ArticlePubMed CentralGoogle Scholar
- Sirota FL, Goh F, Low KN, Yang LK, Crasta SC, Eisenhaber B, Eisenhaber F, Kanagasundaram Y, Ng SB. Isolation and identification of an Anthracimycin analogue from Nocardiopsis kunsanensis, a halophile from a Saltern, by genomic mining strategy. J Genomics. 2018;6:63–73.PubMedPubMed CentralView ArticleGoogle Scholar
- Salomon CE, Magarvey NA, Sherman DH. Merging the potential of microbial genetics with biological and chemical diversity: an even brighter future for marine natural product drug discovery. Nat Prod Rep. 2004;21:105–21.PubMedView ArticlePubMed CentralGoogle Scholar
- Schwarzer D, Marahiel MA. Multimodular biocatalysts for natural product assembly. Naturwissenschaften. 2001;88:93–101.PubMedView ArticlePubMed CentralGoogle Scholar
- Hutchinson CR. Polyketide and non-ribosomal peptide synthases: falling together by coming apart. Proc Natl Acad Sci U S A. 2003;100:3010–2.PubMedPubMed CentralView ArticleGoogle Scholar
- Conti E, Stachelhaus T, Marahiel MA, Brick P. Structural basis for the activation of phenylalanine in the non-ribosomal biosynthesis of gramicidin S. EMBO J. 1997;16:4174–83.PubMedPubMed CentralView ArticleGoogle Scholar
- Stachelhaus T, Mootz HD, Marahiel MA. The specificity-conferring code of adenylation domains in nonribosomal peptide synthetases. Chem Biol. 1999;6:493–505.PubMedView ArticleGoogle Scholar
- Weber T, Baumgartner R, Renner C, Marahiel MA, Holak TA. Solution structure of PCP, a prototype for the peptidyl carrier domains of modular peptide synthetases. Structure. 2000;8:407–18.PubMedView ArticleGoogle Scholar
- Keating TA, Marshall CG, Walsh CT, Keating AE. The structure of VibH represents nonribosomal peptide synthetase condensation, cyclization and epimerization domains. Nat Struct Biol. 2002;9:522–6.PubMedGoogle Scholar
- Samel SA, Schoenafinger G, Knappe TA, Marahiel MA, Essen LO. Structural and functional insights into a peptide bond-forming bidomain from a nonribosomal peptide synthetase. Structure. 2007;15:781–92.PubMedView ArticleGoogle Scholar
- Stachelhaus T, Mootz HD, Bergendahl V, Marahiel MA. Peptide bond formation in nonribosomal peptide biosynthesis. Catalytic role of the condensation domain. J Biol Chem. 1998;273:22773–81.PubMedView ArticleGoogle Scholar
- Weber T, Marahiel MA. Exploring the domain structure of modular nonribosomal peptide synthetases. Structure. 2001;9:R3–9.PubMedView ArticleGoogle Scholar
- Dutta S, Whicher JR, Hansen DA, Hale WA, Chemler JA, Congdon GR, Narayan AR, Hakansson K, Sherman DH, Smith JL, et al. Structure of a modular polyketide synthase. Nature. 2014;510:512–7.PubMedPubMed CentralView ArticleGoogle Scholar
- Khosla C. Harnessing the biosynthetic potential of modular polyketide synthases. Chem Rev. 1997;97:2577–90.PubMedView ArticlePubMed CentralGoogle Scholar
- Smith JL, Skiniotis G, Sherman DH. Architecture of the polyketide synthase module: surprises from electron cryo-microscopy. Curr Opin Struct Biol. 2015;31:9–19.PubMedPubMed CentralView ArticleGoogle Scholar
- Yuzawa S, Backman TWH, Keasling JD, Katz L. Synthetic biology of polyketide synthases. J Ind Microbiol Biotechnol. 2018.Google Scholar
- Chen H, Du L. Iterative polyketide biosynthesis by modular polyketide synthases in bacteria. Appl Microbiol Biotechnol. 2016;100:541–57.PubMedView ArticleGoogle Scholar
- Fujii I. Functional analysis of fungal polyketide biosynthesis genes. J Antibiot (Tokyo). 2010;63:207–18.View ArticleGoogle Scholar
- Bionda N, Pitteloud JP, Cudic P. Cyclic lipodepsipeptides: a new class of antibacterial agents in the battle against resistant bacteria. Future Med Chem. 2013;5:1311–30.PubMedView ArticleGoogle Scholar
- Bionda N, Fleeman RM, de lF-N, Rodriguez MC, Reffuveille F, Shaw LN, Pastar I, Davis SC, Hancock REW, Cudic P. Identification of novel cyclic lipopeptides from a positional scanning combinatorial library with enhanced antibacterial and antibiofilm activities. Eur J Med Chem. 2016;108:354–63.PubMedView ArticleGoogle Scholar
- Herath K, Harris G, Jayasuriya H, Zink D, Smith S, Vicente F, Bills G, Collado J, Gonzalez A, Jiang B, et al. Isolation, structure and biological activity of phomafungin, a cyclic lipodepsipeptide from a widespread tropical Phoma sp. Bioorg Med Chem. 2009;17:1361–9.PubMedView ArticleGoogle Scholar
- Grilley MM, Stock SD, Dickson RC, Lester RL, Takemoto JY. Syringomycin action gene SYR2 is essential for sphingolipid 4-hydroxylation in Saccharomyces cerevisiae. J Biol Chem. 1998;273:11062–8.PubMedView ArticleGoogle Scholar
- Di GD, Camoni L, Marchiafava C, Ballio A. Biological activities of pseudomycin a, a lipodepsinonapeptide from Pseudomonas syringae MSU 16H. Phytochemistry. 1997;45:1385–91.View ArticleGoogle Scholar
- Scaloni A, Dalla SM, Amodeo P, Mannina L, Vitale RM, Segre AL, Cruciani O, Lodovichetti F, Greco ML, Fiore A, et al. Structure, conformation and biological activity of a novel lipodepsipeptide from Pseudomonas corrugata: cormycin a. Biochem J. 2004;384:25–36.PubMedPubMed CentralView ArticleGoogle Scholar
- Singh SB, Ondeyka J, Harris G, Herath K, Zink D, Vicente F, Bills G, Collado J, Platas G, Gonzalez d, V et al.: Isolation, structure, and biological activity of Phaeofungin, a cyclic lipodepsipeptide from a Phaeosphaeria sp. using the genome-wide Candida albicans fitness test. J Nat Prod 2013, 76:334–345.PubMedView ArticleGoogle Scholar
- Stanke M, Diekhans M, Baertsch R, Haussler D. Using native and syntenically mapped cDNA alignments to improve de novo gene finding. Bioinformatics. 2008;24:637–44.PubMedView ArticlePubMed CentralGoogle Scholar
- Lomsadze A, Ter-Hovhannisyan V, Chernoff YO, Borodovsky M. Gene identification in novel eukaryotic genomes by self-training algorithm. Nucleic Acids Res. 2005;33:6494–506.PubMedPubMed CentralView ArticleGoogle Scholar
- Ter-Hovhannisyan V, Lomsadze A, Chernoff YO, Borodovsky M. Gene prediction in novel fungal genomes using an ab initio algorithm with unsupervised training. Genome Res. 2008;18:1979–90.PubMedPubMed CentralView ArticleGoogle Scholar
- Eisenhaber B, Kuchibhatla D, Sherman W, Sirota FL, Berezovsky IN, Wong WC, Eisenhaber F. The recipe for protein sequence-based function prediction and its implementation in the ANNOTATOR software environment. Methods Mol Biol. 2016;1415:477–506.PubMedView ArticlePubMed CentralGoogle Scholar
- Weber T, Blin K, Duddela S, Krug D, Kim HU, Bruccoleri R, Lee SY, Fischbach MA, Muller R, Wohlleben W, et al. antiSMASH 3.0-a comprehensive resource for the genome mining of biosynthetic gene clusters. Nucleic Acids Res. 2015;43:W237–43.PubMedPubMed CentralView ArticleGoogle Scholar
- Cacho RA, Jiang W, Chooi YH, Walsh CT, Tang Y. Identification and characterization of the echinocandin B biosynthetic gene cluster from Emericella rugulosa NRRL 11440. J Am Chem Soc. 2012;134:16781–90.PubMedPubMed CentralView ArticleGoogle Scholar
- Chiang YM, Szewczyk E, Nayak T, Davidson AD, Sanchez JF, Lo HC, Ho WY, Simityan H, Kuo E, Praseuth A, et al. Molecular genetic mining of the Aspergillus secondary metabolome: discovery of the emericellamide biosynthetic pathway. Chem Biol. 2008;15:527–32.PubMedPubMed CentralView ArticleGoogle Scholar
- Huttel W. Structural diversity in echinocandin biosynthesis: the impact of oxidation steps and approaches toward an evolutionary explanation. Z Naturforsch C. 2017;72:1–20.PubMedView ArticlePubMed CentralGoogle Scholar
- Ishidoh K, Kinoshita H, Nihira T. Identification of a gene cluster responsible for the biosynthesis of cyclic lipopeptide verlamelin. Appl Microbiol Biotechnol. 2014;98:7501–10.PubMedView ArticlePubMed CentralGoogle Scholar
- Du L, Lou L. PKS and NRPS release mechanisms. Nat Prod Rep. 2010;27:255–78.PubMedView ArticlePubMed CentralGoogle Scholar
- Kopp F, Marahiel MA. Macrocyclization strategies in polyketide and nonribosomal peptide biosynthesis. Nat Prod Rep. 2007;24:735–49.PubMedView ArticlePubMed CentralGoogle Scholar
- von DH: A survey of nonribosomal peptide synthetase (NRPS) genes in Aspergillus nidulans. Fungal Genet Biol 2009, 46 Suppl 1:S45-S52.Google Scholar
- Eddy SR. Accelerated profile HMM searches. PLoS Comput Biol. 2011;7:e1002195.PubMedPubMed CentralView ArticleGoogle Scholar
- Li W, Rokni-Zadeh H, De VM, Ghequire MG, Sinnaeve D, Xie GL, Rozenski J, Madder A, Martins JC, De MR. The antimicrobial compound xantholysin defines a new group of Pseudomonas cyclic lipopeptides. PLoS One. 2013;8:e62946.PubMedPubMed CentralView ArticleGoogle Scholar
- Mizuno CM, Kimes NE, Lopez-Perez M, Auso E, Rodriguez-Valera F, Ghai R. A hybrid NRPS-PKS gene cluster related to the bleomycin family of antitumor antibiotics in Alteromonas macleodii strains. PLoS One. 2013;8:e76021.PubMedPubMed CentralView ArticleGoogle Scholar
- Katoh K, Standley DM. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol. 2013;30:772–80.PubMedPubMed CentralView ArticleGoogle Scholar
- Challis GL, Ravel J, Townsend CA. Predictive, structure-based model of amino acid recognition by nonribosomal peptide synthetase adenylation domains. Chem Biol. 2000;7:211–24.PubMedView ArticlePubMed CentralGoogle Scholar
- De B, I, de Kock MJ, Yang M, de WP, van Beek TA, Raaijmakers JM: Genome-based discovery, structure prediction and functional analysis of cyclic lipopeptide antibiotics in Pseudomonas species. Mol Microbiol 2007;63:417–28.Google Scholar
- Zhang W, Li Y, Qian G, Wang Y, Chen H, Li YZ, Liu F, Shen Y, Du L. Identification and characterization of the anti-methicillin-resistant Staphylococcus aureus WAP-8294A2 biosynthetic gene cluster from Lysobacter enzymogenes OH11. Antimicrob Agents Chemother. 2011;55:5581–9.PubMedPubMed CentralView ArticleGoogle Scholar
- Rausch C, Hoof I, Weber T, Wohlleben W, Huson DH. Phylogenetic analysis of condensation domains in NRPS sheds light on their functional evolution. BMC Evol Biol. 2007;7:78.PubMedPubMed CentralView ArticleGoogle Scholar
- Kumar S, Stecher G, Tamura K. MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol Biol Evol. 2016;33:1870–4.PubMedPubMed CentralView ArticleGoogle Scholar
- Marahiel MA, Stachelhaus T, Mootz HD. Modular peptide Synthetases involved in nonribosomal peptide synthesis. Chem Rev. 1997;97:2651–74.PubMedView ArticlePubMed CentralGoogle Scholar
- Desai DK, Nandi S, Srivastava PK, Lynn AM. ModEnzA: accurate identification of metabolic enzymes using function specific profile HMMs with optimised discrimination threshold and modified emission probabilities. Adv Bioinforma. 2011;2011:743782.View ArticleGoogle Scholar
- Srivastava PK, Desai DK, Nandi S, Lynn AM. HMM-ModE--improved classification using profile hidden Markov models by optimising the discrimination threshold and modifying emission probabilities with negative training sequences. BMC Bioinformatics. 2007;8:104.PubMedPubMed CentralView ArticleGoogle Scholar
- Salib MN, Molinski TF. Cyclic Hexapeptide dimers, Antatollamides a and B, from the ascidian Didemnum molle. A tryptophan-derived auxiliary for l- and d-amino acid assignments. J Org Chem. 2017;82:10181–7.PubMedPubMed CentralView ArticleGoogle Scholar
- Schubert HL, Blumenthal RM, Cheng X. Many paths to methyltransfer: a chronicle of convergence. Trends Biochem Sci. 2003;28:329–35.PubMedPubMed CentralView ArticleGoogle Scholar
- Claxton HB, Akey DL, Silver MK, Admiraal SJ, Smith JL. Structure and functional analysis of RifR, the type II thioesterase from the rifamycin biosynthetic pathway. J Biol Chem. 2009;284:5021–9.PubMedPubMed CentralView ArticleGoogle Scholar
- Koglin A, Lohr F, Bernhard F, Rogov VV, Frueh DP, Strieter ER, Mofid MR, Guntert P, Wagner G, Walsh CT, et al. Structural basis for the selectivity of the external thioesterase of the surfactin synthetase. Nature. 2008;454:907–11.PubMedPubMed CentralView ArticleGoogle Scholar
- Rex JH: Reference method for broth dilution antifungal susceptibility testing of yeasts : approved standard. Wayne PA: Clinical and Laboratory Standards Institute.; 2008.Google Scholar
- Wang B, Kang Q, Lu Y, Bai L, Wang C. Unveiling the biosynthetic puzzle of destruxins in Metarhizium species. Proc Natl Acad Sci U S A. 2012;109:1287–92.PubMedPubMed CentralView ArticleGoogle Scholar
- Slightom JL, Metzger BP, Luu HT, Elhammer AP. Cloning and molecular characterization of the gene encoding the Aureobasidin a biosynthesis complex in Aureobasidium pullulans BP-1938. Gene. 2009;431:67–79.PubMedView ArticlePubMed CentralGoogle Scholar
- Xu Y, Orozco R, Wijeratne EM, Gunatilaka AA, Stock SP, Molnar I. Biosynthesis of the cyclooligomer depsipeptide beauvericin, a virulence factor of the entomopathogenic fungus Beauveria bassiana. Chem Biol. 2008;15:898–907.PubMedView ArticlePubMed CentralGoogle Scholar
- Jin JM, Lee S, Lee J, Baek SR, Kim JC, Yun SH, Park SY, Kang S, Lee YW. Functional characterization and manipulation of the apicidin biosynthetic pathway in Fusarium semitectum. Mol Microbiol. 2010;76:456–66.PubMedView ArticlePubMed CentralGoogle Scholar
- Kotowska M, Pawlik K. Roles of type II thioesterases and their application for secondary metabolite yield improvement. Appl Microbiol Biotechnol. 2014;98:7735–46.PubMedPubMed CentralView ArticleGoogle Scholar
- Huttel W, Youssar L, Gruning BA, Gunther S, Hugentobler KG. Echinocandin B biosynthesis: a biosynthetic cluster from Aspergillus nidulans NRRL 8112 and reassembly of the subclusters Ecd and Hty from Aspergillus pachycristatus NRRL 11440 reveals a single coherent gene cluster. BMC Genomics. 2016;17:570.PubMedPubMed CentralView ArticleGoogle Scholar
- Stachelhaus T, Marahiel MA. Modular structure of peptide synthetases revealed by dissection of the multifunctional enzyme GrsA. J Biol Chem. 1995;270:6163–9.PubMedView ArticlePubMed CentralGoogle Scholar
- Ansari MZ, Yadav G, Gokhale RS, Mohanty D. NRPS-PKS: a knowledge-based resource for analysis of NRPS/PKS megasynthases. Nucleic Acids Res. 2004;32:W405–13.PubMedPubMed CentralView ArticleGoogle Scholar
- Bachmann BO, Ravel J. Methods for in silico prediction of microbial polyketide and nonribosomal peptide biosynthetic pathways from DNA sequence data. Methods Enzymol. 2009;458:181–217.PubMedView ArticlePubMed CentralGoogle Scholar
- Blin K, Medema MH, Kottmann R, Lee SY, Weber T. The antiSMASH database, a comprehensive database of microbial secondary metabolite biosynthetic gene clusters. Nucleic Acids Res. 2017;45:D555–9.PubMedView ArticlePubMed CentralGoogle Scholar
- guero-Chapin G, Molina-Ruiz R, Maldonado E, de la RG, Sanchez-Rodriguez A, Vasconcelos V, Antunes A. Exploring the adenylation domain repertoire of nonribosomal peptide synthetases using an ensemble of sequence-search methods. PLoS One. 2013;8:e65926.View ArticleGoogle Scholar
- Prieto C, Garcia-Estrada C, Lorenzana D, Martin JF. NRPSsp: non-ribosomal peptide synthase substrate predictor. Bioinformatics. 2012;28:426–7.PubMedView ArticlePubMed CentralGoogle Scholar
- Prieto C. Characterization of nonribosomal peptide Synthetases with NRPSsp. Methods Mol Biol. 2016;1401:273–8.PubMedView ArticlePubMed CentralGoogle Scholar
- Rottig M, Medema MH, Blin K, Weber T, Rausch C, Kohlbacher O. NRPSpredictor2--a web server for predicting NRPS adenylation domain specificity. Nucleic Acids Res. 2011;39:W362–7.PubMedPubMed CentralView ArticleGoogle Scholar
- Gulick AM. Conformational dynamics in the acyl-CoA synthetases, adenylation domains of non-ribosomal peptide synthetases, and firefly luciferase. ACS Chem Biol. 2009;4:811–27.PubMedPubMed CentralView ArticleGoogle Scholar
- Lee TV, Johnson LJ, Johnson RD, Koulman A, Lane GA, Lott JS, Arcus VL. Structure of a eukaryotic nonribosomal peptide synthetase adenylation domain that activates a large hydroxamate amino acid in siderophore biosynthesis. J Biol Chem. 2010;285:2415–27.PubMedView ArticlePubMed CentralGoogle Scholar
- Sussmuth RD, Mainz A. Nonribosomal peptide synthesis-principles and prospects. Angew Chem Int Ed Engl. 2017;56:3770–821.PubMedView ArticlePubMed CentralGoogle Scholar
- Schwarzer D, Finking R, Marahiel MA. Nonribosomal peptides: from genes to products. Nat Prod Rep. 2003;20:275–87.PubMedView ArticlePubMed CentralGoogle Scholar
- Luo C, Liu X, Zhou X, Guo J, Truong J, Wang X, Zhou H, Li X, Chen Z. Unusual biosynthesis and structure of Locillomycins from Bacillus subtilis 916. Appl Environ Microbiol. 2015;81:6601–9.PubMedPubMed CentralView ArticleGoogle Scholar
- Rouhiainen L, Jokela J, Fewer DP, Urmann M, Sivonen K. Two alternative starter modules for the non-ribosomal biosynthesis of specific anabaenopeptin variants in Anabaena (cyanobacteria). Chem Biol. 2010;17:265–73.PubMedView ArticlePubMed CentralGoogle Scholar
- Roongsawang N, Lim SP, Washio K, Takano K, Kanaya S, Morikawa M. Phylogenetic analysis of condensation domains in the nonribosomal peptide synthetases. FEMS Microbiol Lett. 2005;252:143–51.PubMedView ArticlePubMed CentralGoogle Scholar
- Ziemert N, Podell S, Penn K, Badger JH, Allen E, Jensen PR. The natural product domain seeker NaPDoS: a phylogeny based bioinformatic tool to classify secondary metabolite gene diversity. PLoS One. 2012;7:e34064.PubMedPubMed CentralView ArticleGoogle Scholar
- Balibar CJ, Vaillancourt FH, Walsh CT. Generation of D amino acid residues in assembly of arthrofactin by dual condensation/epimerization domains. Chem Biol. 2005;12:1189–200.PubMedView ArticlePubMed CentralGoogle Scholar
- De Crecy-Lagard V, Marliere P, Saurin W. Multienzymatic non ribosomal peptide biosynthesis: identification of the functional domains catalysing peptide elongation and epimerisation. C R Acad Sci III. 1995;318:927–36.PubMedPubMed CentralGoogle Scholar
- Bergendahl V, Linne U, Marahiel MA. Mutational analysis of the C-domain in nonribosomal peptide synthesis. Eur J Biochem. 2002;269:620–9.PubMedView ArticlePubMed CentralGoogle Scholar
- Roche ED, Walsh CT. Dissection of the EntF condensation domain boundary and active site residues in nonribosomal peptide synthesis. Biochemistry. 2003;42:1334–44.PubMedView ArticlePubMed CentralGoogle Scholar
- Kraas FI, Helmetag V, Wittmann M, Strieker M, Marahiel MA. Functional dissection of surfactin synthetase initiation module reveals insights into the mechanism of lipoinitiation. Chem Biol. 2010;17:872–80.PubMedView ArticlePubMed CentralGoogle Scholar
- Luo C, Liu X, Zhou H, Wang X, Chen Z. Nonribosomal peptide synthase gene clusters for lipopeptide biosynthesis in Bacillus subtilis 916 and their phenotypic functions. Appl Environ Microbiol. 2015;81:422–31.PubMedView ArticlePubMed CentralGoogle Scholar
- Bloudoff K, Schmeing TM. Structural and functional aspects of the nonribosomal peptide synthetase condensation domain superfamily: discovery, dissection and diversity. Biochim Biophys Acta. 2017;1865:1587–604.View ArticleGoogle Scholar
- Konz D, Marahiel MA. How do peptide synthetases generate structural diversity? Chem Biol. 1999;6:R39–48.PubMedView ArticlePubMed CentralGoogle Scholar
- Chen L, Yue Q, Zhang X, Xiang M, Wang C, Li S, Che Y, Ortiz-Lopez FJ, Bills GF, Liu X, et al. Genomics-driven discovery of the pneumocandin biosynthetic gene cluster in the fungus Glarea lozoyensis. BMC Genomics. 2013;14:339.PubMedPubMed CentralView ArticleGoogle Scholar
- Bensaci MF, Gurnev PA, Bezrukov SM, Takemoto JY. Fungicidal activities and mechanisms of action of Pseudomonas syringae pv. Syringae Lipodepsipeptide Syringopeptins 22A and 25A. Front Microbiol. 2011;2:216.PubMedPubMed CentralView ArticleGoogle Scholar
- Takemoto JY, Zhang L, Taguchi N, Tachikawa T, Miyakawa T. Mechanism of action of the phytotoxin syringomycin: a resistant mutant of Saccharomyces cerevisiae reveals an involvement of Ca2+ transport. Microbiology. 1991;137:653–9.Google Scholar
- Zhang L, Takemoto JY. Syringomycin stimulation of potassium efflux by yeast cells. Biochimica et Biophysica Acta-Biomembranes. 1989;987:171–5.View ArticleGoogle Scholar
- Aeed PA, Young CL, Nagiec MM, Elhammer AP. Inhibition of inositol phosphorylceramide synthase by the cyclic peptide aureobasidin a. Antimicrob Agents Chemother. 2009;53:496–504.PubMedView ArticleGoogle Scholar
- Giaever G, Flaherty P, Kumm J, Proctor M, Nislow C, Jaramillo DF, Chu AM, Jordan MI, Arkin AP, Davis RW. Chemogenomic profiling: identifying the functional interactions of small molecules in yeast. Proc Natl Acad Sci U S A. 2004;101:793–8.PubMedPubMed CentralView ArticleGoogle Scholar
- Duckworth BP, Wilson DJ, Aldrich CC. Measurement of nonribosomal peptide Synthetase Adenylation domain activity using a continuous hydroxylamine release assay. Methods Mol Biol. 2016;1401:53–61.PubMedPubMed CentralView ArticleGoogle Scholar
- Ishikawa F, Miyamoto K, Konno S, Kasai S, Kakeya H. Accurate detection of Adenylation domain functions in nonribosomal peptide Synthetases by an enzyme-linked immunosorbent assay system using active site-directed probes for Adenylation domains. ACS Chem Biol. 2015;10:2816–26.PubMedView ArticleGoogle Scholar
- Kittila T, Schoppet M, Cryle MJ. Online pyrophosphate assay for analyzing Adenylation domains of nonribosomal peptide Synthetases. Chembiochem. 2016;17:576–84.PubMedView ArticleGoogle Scholar
- Maruyama C, Niikura H, Takakuwa M, Katano H, Hamano Y. Colorimetric detection of the Adenylation activity in nonribosomal peptide Synthetases. Methods Mol Biol. 2016;1401:77–84.PubMedView ArticleGoogle Scholar
- McQuade TJ, Shallop AD, Sheoran A, Delproposto JE, Tsodikov OV, Garneau-Tsodikova S. A nonradioactive high-throughput assay for screening and characterization of adenylation domains for nonribosomal peptide combinatorial biosynthesis. Anal Biochem. 2009;386:244–50.PubMedView ArticleGoogle Scholar
- Allen GC, Flores-Vergara MA, Krasynanski S, Kumar S, Thompson WF. A modified protocol for rapid DNA isolation from plant tissues using cetyltrimethylammonium bromide. Nat Protoc. 2006;1:2320–5.PubMedView ArticleGoogle Scholar
- Chin CS, Alexander DH, Marks P, Klammer AA, Drake J, Heiner C, Clum A, Copeland A, Huddleston J, Eichler EE, et al. Nonhybrid, finished microbial genome assemblies from long-read SMRT sequencing data. Nat Methods. 2013;10:563–9.PubMedView ArticleGoogle Scholar
- Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 1997;25:3389–402.PubMedPubMed CentralView ArticleGoogle Scholar
- Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Soding J, et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol. 2011;7:539.PubMedPubMed CentralView ArticleGoogle Scholar
- Blin K, Wolf T, Chevrette MG, Lu X, Schwalen CJ, Kautsar SA, Suarez Duran HG, de Los Santos ELC, Kim HU, Nave M, et al. antiSMASH 4.0-improvements in chemistry prediction and gene cluster boundary identification. Nucleic Acids Res. 2017.Google Scholar
- Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32:1792–7.PubMedPubMed CentralView ArticleGoogle Scholar
- Balouiri M, Sadiki M, Ibnsouda SK. Methods for in vitro evaluating antimicrobial activity: a review. J Pharm Anal. 2016;6:71–9.PubMedView ArticleGoogle Scholar