Background: Plipastatin, an antifungal lipopeptide, is synthesized by a non-ribosomal peptide synthetase (NRPS) in Bacillus subtilis. However, little information is available on the combinatorial biosynthesis strategies applied in plipasta- tin biosynthetic pathway. In this study, we applied module or individual domain deletion strategies to engineer the plipastatin biosynthetic pathway, and investigated the effect of deletions on the plipastatin assembly line, as well as revealed the synthetic patterns of novel lipopeptides. Results: Module deletion inactivated the entire enzyme complex, whereas individual domain (A/T domain) deletion within module 7 truncated the assembly line, resulting in truncated linear hexapeptides (C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Ala/Val). Interestingly, within the module 6 catalytic unit, the effect of thiolation 16~17 domain deletion differed from that of adenylation deletion. Absence of the T -domain resulted in a nonproductive strain, whereas deletion of the A -domain resulted in multiple assembly lines via module-skipping mechanism, gen- erating three novel types of plipastatin derivatives, pentapeptides ( C β-OHFA-Glu-Orn-Tyr-Thr-Glu), hexapeptides 16~17 (C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Ile), and octapeptides (C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Gln-Tyr-Ile). 16~17 16~17 Conclusions: Notably, a unique module-skipping process occurred following deletion of the A -domain, which has not been previously reported for engineered NRPS systems. This finding provides new insight into the lipopeptides engineering. It is of significant importance for combinatorial approaches and should be taken into consideration in engineering non-ribosomal peptide biosynthetic pathways for generating novel lipopeptides. Keywords: Genetic engineering, Plipastatin synthetase, Novel lipopeptides, Module skipping Background cyclic or branched-cyclic, and often contain non-pro- Microorganisms produce a large variety of small bioac- teinogenic amino acids as building blocks. Amino acids tive peptides, many of which are biosynthesised by mul- can be modified by peptide synthetases through epimeri - tifunctional megasynthetases known as nonribosomal sation, methylation or hydroxylation, resulting in enor- peptide synthetases (NRPSs). Prominent biosurfactants mous structural diversity of peptides that makes many (e.g., surfactin), drugs (e.g., vancomycin and cyclosporin compounds difficult to chemically synthesise [ 3, 4]. A) and fungicides (e.g., iturin A and fengycin) are exam- NRPSs are often composed of a series of specific amino ples of compounds derived from nonribosomal peptide acid-incorporating modules, each consisting of three biosynthesis [1–3]. Compared with the polypeptides pro- catalytic domains ; condensation (C) domains couple duced by ribosomal synthesis, peptides built on NRPSs activated amino acids to the growing peptide chain, ade- are typically shorter (up to about 20 residues), linear, nylation (A) domains selectively activate specific amino acids, and thiolation (T) domains, also called peptidyl carrier proteins (PCPs), tether the activated amino acids and growing peptide chains through the cofactor 4′-phos- *Correspondence: firstname.lastname@example.org College of Food Science and Technology, Nanjing Agricultural University, phopantetheine. The final module usually contains an 1 Weigang, Nanjing 210095, People’s Republic of China © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/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://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Gao et al. Microb Cell Fact (2018) 17:84 Page 2 of 13 additional thioesterase (Te) domain to release products fatty acid chain (C to C ) at the N-terminal residue in 14 18 by cyclisation or hydrolysis. In the plipastatin NRPS [6– the chain (Fig. 1b). Plipastatin is assembled on five giant 8], modules 2, 4, 6 and 9 contain an epimerase (E) domain NRPS multi-enzymes, PPSA, PPSB, PPSC, PPSD, and after the T domain, which converts l-amino acids into PPSE, encoded by the ppsA, ppsB, ppsC, ppsD, and ppsE d-amino acids (Fig. 1a). The lipopeptide plipastatin, also genes, respectively, in the plipastatin synthetase operon known as fengycin, possesses strong biological activity (Fig. 1a). against phospholipase A2 and filamentous fungi . The Because NRPSs have a modular architecture, the order compound has a 10 amino acid core cyclised by an intra- and specificity of modules determines the sequence of molecular ester bond to make an eight-membered ring amino acid residues in the final peptide products. This with a two residue side chain linked with a β-hydroxy organisation is conducive to straightforward strategies ppsA (7.7 kb) ppsB (7.8 kb) ppsC (7.7 kb) ppsD (10.8 kb) ppsE (3.8 kb) Module 6 Module 8 Module 4 Module 10 Module 2 Module 9 Module 5 Module 7 Module 1 Module 3 E E E Ala/ Ile T T T C Te C Glu T C Pro C Gln T Tyr T Glu T Tyr T T C C C C Orn T C C Thr Val PPSA (289 kDa) PPSB (290 kDa) PPSC (287 kDa) PPSD (407 kDa) PPSE (145 kDa) Plipastatin D D L D Tyr Orn Thr Glu Ala/Val C CHCH CO Glu 11-15 2 OH L D L L Ile Tyr Gln Pro Deletion Strategies Mutants Plipastatin derivatives Module 6 BM6 None C Ala/Val i) Module deletion Module 7 C Pro BM7 None Module 6 C16~17β-OHFA-Glu-Orn-Tyr-Thr-Glu BA6 C16~17β-OHFA-Glu-Orn-Tyr-Thr-Glu-Gln-Tyr-Ile Ala/Val ii) A-domain deletion C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Ile 16~17 Module 7 Pro T C BA7 C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Ala/Val 16~17 Module 6 BT6 T None C Ala/Val iii) T-domain deletion Module 7 C Pro BT7 C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Ala/Val 16~17 Fig. 1 a Schematic diagram of the plipastatin biosynthetic system. b Structure of plipastatin. c Deletion strategies: (i) deleting module 6 or 7 in plipastatin NRPS respectively, (ii) Deleting single A or A domain, (iii) deleting single T and T domain. Plipastatin NRPS assembly line contains 6 7 6 7 five subunits, encoded by five genes ppsABCDE. Base on their function, the plipastatin synthetase are divided into 10 modules. Each module is comprised of condensation (C), adenylation (A) and thiolation ( T ) domains. Module 2, 4, 6 and 9 contains an epimerization (E) domian. A thioesterase ( Te) domain is located on the C-teminus of module 10, which is responsible for products cyclisation and hydrolysis Gao et al. Microb Cell Fact (2018) 17:84 Page 3 of 13 for constructing hybrid NRPSs in which functional mod- Plasmids pKS-ΔA6 and pKS-ΔA7 were used to knockout ules (C-A-T or C-A-T-E) or domains are deleted and the respective A-domain, while pKS-ΔT6 and pKS-ΔT7 replaced to produce novel peptide analogs [10–12]. How- were constructed to delete the T-domain of modules 6 or ever, genetic engineering of the plipastatin synthetase 7, respectively. All primers used are listed in Additional complex has not been exploited for the biosynthesis of file 1: Table S2 and were designed with Primer 5.0 based novel lipopeptides, and combinatorial biosynthesis strat- on genes encoding plipastatin synthetase subunits C and egies related to the plipastatin biosynthetic pathway have D (Gene ID 940102 and 940013, respectively). not been widely reported. Surfactin synthetases and dap- For pKS-ΔM6, fragments of DNA upstream and down- tomycin synthetases have been developed into a model stream of module 6 (C-A-T) were amplified by PCR with system for manipulation of NRPSs, and novel compounds primer pairs Module6U-F/R and Module6D-F/R, respec- have been successfully produced via module or domain tively. The 0.75 kb upstream fragment (ppsC: 2365–3114 exchange [13–15]. Based on the crystal structure of the region; coordinates refer to Gene ID 940102 in the ppsC substrate selecting/activating A domain, the specificity- gene cluster) consisted of a portion of the 3′ end of conferring codes in the active sites of A domains have the gene encoding module 5, while the 0.56 kb down- been exploited to modulate specificity in NRPSs using stream fragment (ppsC: 6211–6774 region) included the site-specific mutagenesis and subdomain swapping [16, gene region encoding the T-E linker and a portion of 17]. These successes demonstrate the potential to further E-domain. These two fragments were used as templates explore and exploit plipastatin biosynthesis to build novel for splice overlap extension polymerase chain reaction lipopeptides by genetic engineering. (SOE-PCR) with primers Module6SOE-F/R. A 1.3 kb We recently investigated the production of bioengi- fusion fragment including the upstream and downstream neered plipastatin analogs by thioesterase domain- and regions of module 6 (C-A-T) was obtained and cloned inter-module communication domain-mediated repro- into the SalI-KpnI sites in pKS2 to yield pKS-ΔM6. gramming of the plipastatin NRPS complex [18, 19]. In All inserted fragments were confirmed by sequencing the present study, we attempted to further modify the (Genscript, Nanjing, China). Other deletion plasmids plipastatin amino acid core by deleting complete mod- pKS-ΔM7, pKS-ΔA6, pKS-ΔA7, pKS-ΔT6 and pKS-ΔT7 ules or individual domains, and subsequently evaluated (Additional file 1: Table S1) were constructed by a similar the effect of these changes on the plipastatin assembly protocol. line. Additionally, we coupled the same catalytic unit with deletion of the entire module or single domains Construction of Bacillus subtilis deletion mutants and investigated the relationships between deletions and Each deletion plasmid was introduced into B. subtilis novel lipopeptide derivatives catalysed by plipastatin pB2-L by traditional chemical transformation . The synthetases. pKS-based vector carries a kanamycin resistance (Kn ) marker, and transformants were selected at 30 °C on LB −1 Methods agar with 20 μg mL Kn. The deletion mutants were Strains and culture conditions selected by a two-step replacement recombination pro- Bacterial strains and plasmids used in this study are cedure as described previously [19, 21]. Growth at 37 °C, shown in Additional file 1: Table S1. Bacterial strains were a non-permissive temperature for plasmid replication, routinely grown on Luria–Bertani (LB) agar plates or in in the presence of kanamycin selects for clones in which −1 LB broth at 37 °C. Seed medium contained 5 g L beef the plasmid has been integrated into the chromosome −1 −1 −1 extract, 10 g L peptone, 5 g L yeast extract, 5 g L between the target gene and a homologous sequence on −1 NaCl, and 10 g L glucose. Fermentation medium used the plasmid by a single crossover. Subsequently, a sepa- to produce lipopeptides was prepared as described pre- rate clone of the integrant was cultured in LB medium viously . Escherichia coli DH5α was used for general at 30 °C for 48 h to induce a second crossover event cloning, and JM110 was used for demethylation of plas- and excise the plasmid. Kanamycin-sensitive (Kn ) −1 mids. When required, kanamycin (50 μg mL for E. clones with either the parental or deletion sequence −1 coli or 20 μg mL for Bacillus) was added to the culture were obtained and verified for genotype by PCR analy - medium. sis and sequencing. All deletions maintained the linker regions between the target module/domain and adjacent Construction of deletion plasmids domains. Deletion plasmids used for module or individual domain deletion in the plipastatin synthetase (Fig. 1) were all Lipopeptide production, purification, and identification derivatives of pKS2. Deletions of modules 6 or 7 were Mutant strains from LB agar plates were inoculated performed using plasmids pKS-ΔM6 and pKS-ΔM7. into 20 mL of seed medium, and the whole preculture Gao et al. Microb Cell Fact (2018) 17:84 Page 4 of 13 was inoculated into 200 mL of fermentation medium in enzymes lacking a C-domain in module 6 are also unable 1 L flasks and cultured at 30 °C for 3 days with shaking to produce plipastatin derivatives (data not shown), sug- −1 at 180 r min . Cultures were centrifuged at 5000×g for gesting deletion of the C-domain may severely damage 15 min at 4 °C to remove bacterial cells, and the superna- protein–protein interactions and the overall structural tant was adjusted to pH 2.0 with 6 M HCl to precipitate conformation of the plipastatin NRPS, rendering deletion plipastatin and plipastatin-derived peptides. The pellet mutants unable to synthesise lipopeptides. was then collected and extracted with methanol, and the We wondered whether deletion of the Ala/Val-acti- methanol supernatant was evaporated to dryness and vating domain (A ) or the Pro-activating domain (A ) 6 7 redissolved in 2 mL of methanol. Subsequently, the crude would affect the plipastatin NRPS assembly line. The extract was filtered through a 0.22 μm filter and analysed mutants BA6 (ΔA ) and BA7 (ΔA ) were constructed 6 7 by high-resolution liquid chromatography–electrospray (Fig. 1c), and high-resolution LC–ESI–MS analysis of ionisation–mass spectrometry (LC–ESI–MS) using a crude extracts from BA7 revealed a series of molecu- G2-XS Q-TOF mass spectrometer (Waters, USA). A 5 μL lar mass ions at m/z 980.5579, 994.5729, 1008.5878, aliquot of crude extract was loaded onto a UPLC column and 1022.6014 (Fig. 2), which were consistent with (2.1 × 100 mm ACQUITY UPLC BEH C18 column con- the mass ions of predicted truncated hexapeptides taining 1.7 μm particles), and eluted with a solvent gradi- C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Ala/Val. Using 16~17 ent of 5–95% buffer B for 22 min (buffer A = H O + 0.1% MS/MS spectra of the precursor ions [M + H] at m/z formic acid; buffer B = acetonitrile + 0.1% formic acid) at 980.5579 and 1008.5878 (Fig. 3a, b), b- and y-fragment −1 a flow rate of 0.4 mL min and monitoring at 205 nm. ions were assigned, which confirmed the sequence of the Mass spectrometry was performed using an electro- hexapeptide as C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Ala/ spray source in positive ion mode within a mass range Val. Precursor ions m/z 994.5729 and 1022.6014 with a of 50–1500 m/z. Ionisation was performed with a capil- 14 Da mass difference from m/z 980.5579 and 1008.5878 lary voltage of 2.5 kV, a collision energy of 40 eV, a source were assigned as hexapeptide variants with a C β-OHFA temperature of 120 °C, and a desolvation gas tempera- chain. This result indicated that deletion of the A ture of 400 °C. Data acquisition and processing were per- domain generated completely inactive ppsD and ppsE formed using Masslynx 4.1 (Waters, USA). modules (except for the Te domain), which truncated the plipastatin NRPS complex assembly line to PPSA, PPSB, Results PPSC and Te domain. Effects of module deletion on plipastatin synthetase By contrast, three types of predicted plipastatin deriva- To study the plipastatin biosynthesis pathway with the tives, C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Gln-Tyr-Ile intention of building novel lipopeptides by genetic engi- (1; m/z calcd for C H N O [M + H] 1327.7401, 65 103 10 19 neering, the complete amino acid-incorporating module found 1327.7378), C β-OHFA-Glu-Orn-Tyr-Thr-Glu (2; D + (C-A-T) was deleted. We knocked out module 6 ( Ala/ m/z calcd for C H N O [M + H] 923.5341, found 45 75 6 14 Val-incorporating module) or 7 ( Pro-incorporating 923.5330) and C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Ile (3; module) in B. subtilis pB2-L by homologous recombina- m/z calcd for C H N O [M + H] 1036.6182, found 51 86 7 15 tion, and the corresponding mutants BM6 and BM7 were 1036.6172; Fig. 4) were detected in crude extracts from generated using a markerless module deletion (Fig. 1c). mutant BA6 (ΔA ). These protonated molecular species The new amino acid sequence of subunit PPSC and of putatively assigned plipastatin derivative ions were fur- PPSD were showed in Additional file 2: Sequence S1 and ther subjected to MS/MS analysis, and spectra of the m/z S2, which suggested that the communication mediating 1327.7378 ion (Fig. 5b) yielded a series of b fragment ions (COM) domain docking interaction between PPSC and (m/z 1196.65 → 1033.58 → 905.52 → 776.48 → 675.43 → PPSD was not affected by module deletion. LC–MS anal - 512.37 → 398.29) and y fragment ions (m/z 295.16 → 42 ysis of the fermentation extracts from deletion strains 3.22 → 552.27 → 653.31 → 816.37 → 930.46), as well as failed to detect any plipastatin analogs (data not shown), Orn and Tyr residues (m/z 115.09 and 136.08), consistent indicating that internal module 6 or 7 are essential for with hydrogen adducts of predicted octapeptides with the lipopeptide production. Thus, we further explored the sequence C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Gln-Tyr-Ile effects of individual domains in module 6 and 7 by gener - (Fig. 5a). Similarly, MS/MS spectra of precursor ions at ating additional deletion mutants. m/z 923.5330 and 1036.6172 (Figs. 6b, 7b) showed that all product ions were consistent with the sequences of the Effects of A‑domain deletion on plipastatin synthetase putative pentapeptide C β-OHFA-Glu-Orn-Tyr-Thr-Glu Crystal structures of the PCP-C portion of the complex (Fig. 6a) and hexapeptide C β-OHFA-Glu-Orn -Tyr-Thr- revealed a highly conserved fold and a V-shaped struc- Glu-Ile (Fig. 7a). Additionally, given the variation in the ture [22, 23]. Our previous study showed that hybrid length of fatty acid chains, we also identified homologs Gao et al. Microb Cell Fact (2018) 17:84 Page 5 of 13 Fig. 2 The high-resolution ESI-TOF–MS of hexapeptide ions with retention time (RT ) 12.67–13.01 min produced by mutant strain BA7 of these plipastatin derivatives (m/z 1313.7268, 909.5298 ESI–MS/MS spectrum of [M + H] ions at 980.5557 and and 1022.5976) in extracts from the BA6 mutant in which 1022.6120 m/z, as shown in Additional file 3: Figure S2, the β-OH fatty acid containing 16 carbons is linked to the results of which were consistent with the hexapep- the N-terminus of the peptide product (Additional file 3: tides produced by mutant strain BA7. The LC–ESI–MS/ Figure S1). This result indicated that engineered hybrid MS data indicated that after deleting the T -domain, the plipastatin NRPSs lacking the A -domain could form resulting NRPS assembly line consisting only of PPSA, multiple assembly lines that produce plipastatin analogs. PPSB, PPSC and Te domain could direct the production of the same hexapeptides generated by A -domain dele- Effects of T‑domain deletion on plipastatin synthetase tion mutant. The thiolation (T) domain, also called the peptidyl car - rier protein (PCP), tethers activated substrates to the Discussion growing peptide chain during plipastatin biosynthesis. To Nonribosomal peptide synthetases (NRPSs) direct the investigate the effects of deleting the T-domain and test production of lipopeptides via a thiotemplate mecha- whether deletion of the T-and A-domains in the catalytic nism . An interesting feature of NRPS systems is their module has the same impact, the T-domain of modules 6 modular design, where individual modules act as build- and 7 were deleted, generating mutant strains BT6 (ΔT ) ing blocks for incorporation of single amino acid compo- and BT7 (ΔT ), as outlined in Fig. 1c (strategy iii). Unlike nents present in the final lipopeptide product. Based on the BA6 (ΔA ) mutant, BT6 was unable to produce any known NRPS biosynthetic principles, deletion of mod- lipopeptides related to plipastatin, which indicated that ules in the NRPS assembly could yield lipopeptide ana- deletion of T -domain directly led to the direct inactiva- logs with a decreased ring size, and a small number of tion of the plipastatin NRPS complex. successful in vivo deletions of single modules have been However, high-resolution LC–ESI–MS analysis of reported for NRPS-derived compounds. Mootz et al.  the crude extract from mutant BT7 revealed a series claimed that deleting the leucine-incorporating SrfA-A2 of molecular mass ions at m/z 980.5557, 994.5785, module in the surfactin NRPS yielded a Δ2-surfactin var- 3 5 1008.5961 and 1022.6120. Further characterisation of the iant. Meanwhile, three surfactin variants (ΔLeu , ΔAsp , putative hexapeptide was obtained from analysis of the and ΔLeu ) were generated by knocking out modules 3, (See figure on next page.) Fig. 3 ESI-MS/MS spectra of protonated hexapeptide ions [M + H] at m/z 980.5579 a and m/z 1008.5878 b, acquired in Quadrupole-TOF (Q-TOF) mass spectrometer of crude extract from mutant strain BA7 Gao et al. Microb Cell Fact (2018) 17:84 Page 6 of 13 Gao et al. Microb Cell Fact (2018) 17:84 Page 7 of 13 Fig. 4 a LC–ESI–MS total chromatogram of crude extract from mutant strain BA6. b Chromatogram corresponding to m/z 1327.7401. c Chromatogram corresponding to m/z 923.5341. d Chromatogram corresponding to m/z 1036.6182. e High-resolution ESI–MS of parent ions from compound 1, 2 and 3 5 and 6 of the surfactin NRPS, and the resulting ΔLeu production. This finding indicates that the C domain of displayed antifungal activity . However, in the pre- module 6 and the T -domain are likely to play an impor- sent study, deletion of module 6 or 7 in the plipastatin tant role in the overall structural conformation of the NRPS led to inactive hybrid enzymes unable to form plipastatin NRPS complex, and may be essential moie- novel NRPS assembly lines producing lipopeptides. We ties for protein–protein interactions and efficient com - assume that the spatial arrangement of domains and munication within plipastatin synthetase. By contrast, protein interfaces in the hybrid enzyme complexes are deletion of the T -domain resulted in mutant complexes disturbed, even though the linker regions between adja- that could form linear hexapeptides with the sequence cent domains were retained in the mutant complexes. C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Ala/Val. Thus, we 16~17 u Th s, although surfactin and plipastatin are synthesised conclude that deletion of the same domain coupled with by NRPSs, the module deletion approach that worked for the different catalytic modules can impact the plipasta - surfactin lipopeptide derivatives may not be suitable for tin NRPS system differently, and a similar conclusion was the genetic modification of plipastatin synthetase. There - drawn from the deletion of the A-domain in modules 6 fore, we investigated the effects of deletion of individual and 7. domains on plipastatin synthetases. In module 7 of plipastatin synthetases, the A-domain Our previous study showed that deletion of the is responsible for selecting and activating the substrate C-domain in module 6 inactivated the plipastatin syn- proline. Based on the plipastatin biosynthetic process, thetase complex. Furthermore, deletion of the T -domain deleting the A -domain should generate the ΔPro pli- 6 7 also inactivates the complex, preventing plipastatin pastatin variant. However, high-resolution LC–ESI–MS/ Gao et al. Microb Cell Fact (2018) 17:84 Page 8 of 13 Fig. 5 MS/MS analysis of octapeptide produced by mutant strain BA6. a The sequence of octapeptide and thirteen characteristic product ions b ~ b and y ~ y . b MS/MS spectrum of octapeptide ion [M + H] at m/z 1327.7378 and assignment of key product ions 1 7 2 7 MS analysis of the crude extract from the BA7 mutant five N-terminal residues (Glu-Orn-Tyr-Thr-Glu), which only revealed linear hexapeptides, consistent with the implies that a series of modules located upstream of mod- result of T domain deletion. This result demonstrated ule 6 retain the ability to assemble the precursor penta- that deletion of the A- or T-domain of module 7 do not peptide chain. Furthermore, we postulate a mechanism affect catalysis by upstream modules, but prevented involving module skipping for mutant plipastatin NRPSs, downstream modules from catalysing the extension of as outlined in Fig. 8. The absence of an A -domain results the lipopeptide product, ultimately leading to a trun- in a hybrid biosynthetic system that is more flexible than cated assembly line with PPSA, PPSB, PPSC and Te the native complex, that is able to incorporate the pre- domain that produce hexapeptides with the sequence cursor pentapeptide chain with substrate Gln by skip- C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Ala/Val. ping modules 6 and module 7, then transfer the peptide 16~17 Surprisingly, deletion of the A-domain of module 6, product to the T -domain and continue elongation to which is responsible for selecting and activating the sub- generate octapeptides with the sequence C β -OH 1 6~1 7 1 2 3 4 5 8 9 10 strate alanine or valine, generated a plipastatin hybrid FA- Glu -Or n-T yr - Thr -Gl u-G ln -Tyr -Ile ), before enzyme complex in the BA6 mutant that produced hydrolysing and releasing the final peptide product via three types of plipastatin derivatives; pentapeptides the thioesterase (Fig. 8a). Alternatively, incorporation (C β-OHFA-Glu-Orn-Tyr-Thr-Glu), hexapeptides with substrate Ile could occur by skipping modules 6, 16~17 (C β-OHFA-Glu-Orn-Tyr-Thr-Glu -Ile), and octapep - 7, 8, and 9, then transferring the peptide to the T-domain 16~17 tides (C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Gln-Tyr-Ile; of module 10 to generate linear hexapeptides with the 16~17 1 2 3 4 5 10 Fig. 1c). Comparison of the sequence of the three pli-sequence C β-OHFA-Glu -Orn -Tyr -Thr -Glu -Ile 16~17 pastatin derivatives revealed that they share the same (Fig. 8b). A third option involves transfer directly to Gao et al. Microb Cell Fact (2018) 17:84 Page 9 of 13 Fig. 6 MS/MS analysis of pentapeptide produced by mutant strain BA6. a The sequence of pentapeptide and seven characteristic product ions b ~ b and y ~ y . b MS/MS spectrum of pentapeptide ion [M + H] at m/z 923.5330 and assignment of key product ions 1 4 2 4 the thioesterase (Te) domain to generate linear penta- that engineering NRPSs by deleting the A -domain can 1 2 3 4 5 peptides (C β-OHFA-Glu -Orn -Tyr -Thr -Glu ; yield diverse peptide products from complexes with 16~17 Fig. 8c). Module skipping has been described previously greater biosynthetic potential than originally expected. for hybrid polyketide synthetases [26, 27], and the skip- An consideration in choosing module 6 for deletions ping process was shown to involve passage of the grow- was that it can recruit D-Ala or D-Val. This variability ing polyketide through the skipped module by direct acyl means that its adjacent domains may have more relaxed carrier protein (ACP)-to-ACP transfer . However, substrate specificity. It seemed to be possible to accept to our knowledge, module skipping phenomenon has and transfer the altered amino acid substrates or pep- rarely occurred in NRPS system. There was one report tide intermediates. Another consideration for choos- that module 4 was inactive and skipped completely dur- ing the module 6 and 7 is that the substrate amino acids ing assembly of the myxochromide S peptide core . (D-Ala/Val and L-Pro) they incorporated are location in In this context, it was unexpected that deletion of the the center of the lactone ring in plipastatin structure. We A -domain failed to generate the predicted ΔAla/Val speculate that the module 6 and 7 might have a special plipastatin variant according to linear modular arrange- spatial arrangement in plipastatin synthetase system. Our ment, but instead formed an nonconventional assembly results demonstrated that the site of adenylation domain line through skipped modules 6 and 7, or module 6, 7, in module 6 is special, comparing with module 7. How- 8, and 9 to produce a variety of plipastatin derivatives. ever, little is known about information from intramodule This is the first report of such a process by an engineered protein–protein interaction and substrate recognition. NPRS biosynthesis pathway. This finding demonstrates Whether the deletions of other A-domain located in Gao et al. Microb Cell Fact (2018) 17:84 Page 10 of 13 Fig. 7 MS/MS analysis of hexapeptide produced by mutant strain BA6. a The sequence of hexapeptide and nine characteristic product ions b ~ b 1 5 and y ~ y . b MS/MS spectrum of hexapeptide ion [M + H] at m/z 1036.6172 and assignment of key product ions 2 5 module 2, 3, 4, and 5 will cause module-skipping, which Val) from company of KareBay (Ningbo, China), and needs to further investigations. Additionally, further pro- tested its antimicrobial activity. The results showed that tein structures of full length modules or subunits will this linear hexapeptide was inactive against bacteria, −1 give more detail insight into NRPS architecture for allow- but exhibited a MIC of 62.5 μg mL against Fusarium ing a specific and efficient modification to produce novel oxysporum, Rhizopus stolonifer, and Aspergillus ochra- lipopeptides. ceus. We think it is necessary to improve the production All derivatives identified in the mutant strains were and purity of lipopeptides derivatives, then investigate novel lipopeptides and absent from the wild-type strain. the relationship of their structure-bioactivity in future −1 But, the original plipastatin yield only reaches 10 mg L research. In fact, similar yield reduction has been in B. subtilis pB2-L, which resulted in the production reported in daptomycin and surfactin engineering NRPS. level of derivatives in these experiments were very low. For example, when module and subunit exchanges were The high resolution LC–MS/MS was used to detect and used to engineer daptomycin synthetases, the produc- identify the newly derivatives from the crude extract of tion of novel antibiotics related to daptomycin generally mutant strains. These derivatives were only detected by ranged from about 1 to 50% of control [14, 15]. The yield ion-extraction (shown in Fig. 4). It was difficult to acquire of surfactin derivatives generated by modified the pep - sufficient quantities of purified lipopeptides deriva - tide synthetase were also relatively low [29–31], by mod- tives from the engineering mutant for biological testing ule deletion was about 10% of control, by exchange of by isolation and separation. In previous study , we the leucine-specific domains with heterologous domains chemically synthesized the linear plipastatin derivative of the same specificity were about 0.1–0.5% of control (hexapeptide: C β-OHFA-Glu-Orn-Tyr-Thr-Glu-Ala/ . Since the yields of surfactin and daptomycin could 16 Gao et al. Microb Cell Fact (2018) 17:84 Page 11 of 13 Module 6 Module 8 Module 10 a Module 9 Module 5 Module 7 E E Ala/ Ile T T C C C Te Glu T Pro T Gln T C Tyr T C Ala/Val Val Ile OH S S D 9 Tyr O CONH O D 9 Ile Tyr incorporation of Gln H N glutamine HOOC NH D 9 hydrolysis Tyr Gln by module 8 Glu D 4 Thr Gln Glu D 4 Thr module 6 and 7 Tyr are skipped 5 D 4 Glu Thr Tyr D 2 Orn D 4 Thr Tyr D 2 Orn Glu FA 3 D 2 Tyr Orn Glu FA D 2 Orn Glu FA Linear octapeptide Glu FA Module 6 Module 8 Module 10 b Module 9 Module 5 Module 7 E E Ala/ Ile T T T C T C T C T C Te C Glu Ala/Val Pro Gln Tyr Val Ile OH Glu hydrolysis D 4 Thr incorporation of H N HOOC NH 2 isoleucine D 4 Tyr by module 10 Thr D 2 Orn Tyr module 6, 7, 8 and 9 are skipped D 2 Glu FA Orn Linear hexapeptide Glu FA Module 6 Module 8 Module 10 c Module 9 Module 5 Module 7 E E Ala/ Ile T T T C T C T C T C Te C Glu Ala/Val Pro Gln Tyr Val OH Glu D 4 Thr hydrolysis NH HOOC Tyr D 4 Thr D 2 Orn Tyr Glu FA D 2 Orn Linear pentapeptide Glu FA Fig. 8 An assumed mechanism involving module skipping for mutant plipastatin NRPS without A domain. a Formation of linear octapeptide by complete module 6 and 7 skipping. b Formation of linear hexapeptide by complete module 6, 7, 8 and 9 skipping. c Formation of linear pentapeptide by hydrolysis in advance of thioesterase ( Te) domain Gao et al. Microb Cell Fact (2018) 17:84 Page 12 of 13 −1 reach 1–2 g L by optimizing fermentation param- Additional files eters, the novel derivatives from engineering synthetase can be interesting for pharmaceutical applications, even Additional file 1: Table S1. Strains and plasmids in this study. Table S2. PCR primers used for genetic constructs. though the reduction compared with the wild-type Additional file 2: Sequence S1. The new amino acid sequence of subu- enzyme. It is possible that this approach could be applied nit PPSC. Sequence S2. The new amino acid sequence of subunit PPSD. to industrial plipastatin production for generating sig- Additional file 3: Figure S1. ESI–MS/MS spectra of protonated ions nificant quantities of plipastatin derivatives. Besides, we [M + H] at m/z 1313.7268 (A), m/z 909.5298 (B) and m/z 1022.5976 (C), might anticipate improving the yields by modification acquired in Quadrupole-TOF (Q-TOF) mass spectrometer of crude extract of the regulatory regions and optimizing fermentation isolated from mutant strain BA6. Figure S2. (A) The high-resolution ESI- TOF–MS of hexapeptide ions with retention time (RT ) 12.57 ~ 12.98 min parameters. from crude extract of mutant BT7. ESI–MS/MS spectra of protonated hexapeptide ions [M + H] at m/z 980.5579 (B) and m/z 1022.6120 (C), acquired in Quadrupole-TOF (Q-TOF) mass spectrometer. Conclusions In conclusion, deletion of modules and individual Abbreviations domains were used to engineer plipastatin synthetases NRPS: non-ribosomal peptide synthetase; C-domain: condensation domain; A-domain: adenylation domain; T-domain: thiolation domain; PCP-domain: in B. subtilis, leading to the production of lipopeptide peptidyl carrier protein; E-domain: epimerization domain; Te-domain: thi- analogs. All analogs identified in the mutant strains were oesterase domain; COM-doamin: communication mediating domain; FA: fatty novel lipopeptides and absent from the wild-type strain. acid; LC–ESI–MS/MS: liquid chromatography–electrospray ionisation–mass spectrometry/mass spectrometry. Since none of the products was synthesised as expected. Some synthetic patterns of novel lipopeptides catalyzed Authors’ contributions by engineered NRPS have been found. Deletion of mod LG and XB conceived and designed the experiments. LG and JG performed the experiments. LG, YF and ZM analysed the data. ZL, CZ, HZ and XB contrib- ules resulted in functionally impaired synthetase com- uted reagents/materials/analytical tools. LG and XB wrote the manuscript. All plexes, revealing that the module deletion strategy may authors read and approved the final manuscript. not be suitable for engineering plipastatin synthetase to produce novel plipastatin derivatives. By contrast, Acknowledgements deletion of individual A-/T-domains did not affect the The authors would like to thank members of Enzyme Engineering Laboratory assembly capacity of upstream modules or the hydro in the College of Food Science and Technology at Nanjing Agricultural Univer- sity for their valuable comments and helpful discussions. lytic activity of the downstream thioesterase, which led to a truncated plipastatin assembly line that produced Competing interests truncated linear lipopeptides. Interestingly, a unique The authors declare that they have no competing interests. module-skipping process occurred following deletion of Availability of data and materials the A -domain, which has not been previously reported All data generated during this study are included in this article, and all material for engineered NRPS systems. This finding suggests is available upon request. hybrid enzyme complexes were more flexible and diverse Consent for publication in terms of biosynthetic potential than envisaged based All authors approved publication. on their design. This makes it difficult to design and Ethics approval and consent to participate produce lipopeptide analogs based simply on modular Not applicable. sequences. But, this module skipping phenomenon offers great potential for introducing further structural diver Funding This work was supported by grants from the National Natural Science Founda- sity into nonribosomal peptides through combinatorial tion of China (Grant No. 31271828), the Independent Innovation Program approaches, and should be taken into consideration when of Jiangsu Province [Grant No. CX(16)1058] and Priority Academic Program engineering NRPS biosynthetic pathways. On the other Development of Jiangsu Higher Education Institutions (PAPD). hand, engineering of NRPS to generate new and func- tional derivatives is extremely difficult, and few success - Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- ful cases have reported in last 20 years. In our study, we lished maps and institutional affiliations. successfully obtained five kinds of plipastatin derivatives through deleting module or individual domain. Compar- Received: 7 November 2017 Accepted: 11 May 2018 ing with this two deletion strategies, deletion of individ- ual domains is more effective approach to generate novel plipastatin analogs, which would provide a valuable ref- References erence for exploiting combinatorial biosynthesis of NRPS 1. Kumar A, Johri B. Antimicrobial lipopeptides of Bacillus: natural weap- to produce novel lipopeptides. ons for biocontrol of plant pathogens. In: Satyanarayana TJB, editor. Gao et al. Microb Cell Fact (2018) 17:84 Page 13 of 13 Microorganisms in sustainable agriculture and biotechnology. Berlin: 19. Gao L, Liu H, Ma Z, Han J, Lu Z, Dai C, Lv F, Bie X. Translocation of the Springer; 2012. p. 91–111. thioesterase domain for the redesign of plipastatin synthetase. Sci Rep 2. Baltz RH. Molecular engineering approaches to peptide, polyketide and UK. 2016;6:38467. other antibiotics. Nat Biotechnol. 2006;24:1533–40. 20. Green M, Sambrook J. Molecular cloning: a laboratory manual. 4th ed. 3. Sieber SA, Marahiel MA. Learning from nature’s drug factories: nonriboso- New York: Cold Spring Harbor Laboratory Press; 2012. mal synthesis of macrocyclic peptides. J Bacteriol. 2003;185:7036–43. 21. Zakataeva NP, Nikitina OV, Gronskiy SV, Romanenkov DV, Livshits VA. A 4. Grünewald J, Marahiel MA. Chemoenzymatic and template-directed simple method to introduce marker-free genetic modifications into the synthesis of bioactive macrocyclic peptides. Microbiol Mol Biol R. chromosome of naturally nontransformable Bacillus amyloliquefaciens 2006;70:121–46. strains. Appl Microbiol Biot. 2010;85:1201–9. 5. Sieber SA, Marahiel MA. Molecular mechanisms underlying nonribo- 22. Samel SA, Schoenafinger G, Knappe TA, Marahiel MA, Essen L-O. Structural somal peptide synthesis: approaches to new antibiotics. Chem Rev. and functional insights into a peptide bond-forming bidomain from a 2005;105:715–38. nonribosomal peptide synthetase. Structure. 2007;15:781–92. 6. Tosato V, Albertini AM, Zotti M, Sonda S, Bruschi CV. Sequence comple- 23. Marahiel MA. A structural model for multimodular NRPS assembly lines. tion, identification and definition of the fengycin operon in Bacillus Nat Prod Rep. 2016;33:136–40. subtilis 168. Microbiology. 1997;143:3443–50. 24. Mootz HD, Kessler N, Linne U, Eppelmann K, Schwarzer D, Marahiel MA. 7. Steller S, Vollenbroich D, Leenders F, Stein T, Conrad B, Hofemeister J, Decreasing the ring size of a cyclic nonribosomal peptide antibiotic by Jacques P, Thonart P, Vater J. Structural and functional organization of the in-frame module deletion in the biosynthetic genes. J Amer Chem Soc. fengycin synthetase multienzyme system from Bacillus subtilis b213 and 2002;124:10980–1. A1/3. Chem Biol. 1999;6:31–41. 25. Jiang J, Gao L, Bie X, Lu Z, Liu H, Zhang C, Lu F, Zhao H. Identification of 8. Samel SA, Wagner B, Marahiel MA, Essen L-O. The thioesterase domain of novel surfactin derivatives from NRPS modification of Bacillus subtilis the fengycin biosynthesis cluster: a structural base for the macrocycliza- and its antifungal activity against Fusarium moniliforme. BMC Microbiol. tion of a non-ribosomal lipopeptide. J Mol Biol. 2006;359:876–89. 2016;16:31. 9. Jacques P. Surfactin and other lipopeptides from Bacillus spp. In: Soberón- 26. Wenzel SC, Müller R. Formation of novel secondary metabolites by bacte- Chávez G, editor. Biosurfactants. Berlin: Springer; 2011. p. 57–91. rial multimodular assembly lines: deviations from textbook biosynthetic 10. Baltz RH. Biosynthesis and genetic engineering of lipopeptide antibiotics logic. Curr Opin Chem Biol. 2005;9:447–58. related to daptomycin. Curr Top Med Chem. 2008;8:618–38. 27. Thomas I, Martin CJ, Wilkinson CJ, Staunton J, Leadlay PF. Skipping in 11. Fischbach MA, Walsh CT. Assembly-line enzymology for polyketide and a hybrid polyketide synthase: evidence for ACP-to-ACP chain transfer. nonribosomal peptide antibiotics: logic, machinery, and mechanisms. Chem Biol. 2002;9:781–7. Chem Rev. 2006;106:3468–96. 28. Wenzel SC, Kunze B, Höfle G, Silakowski B, Scharfe M, Blöcker H, Müller R. 12. Baltz RH, Miao V, Wrigley SK. Natural products to drugs: daptomycin and Structure and biosynthesis of myxochromides S1–3 in Stigmatella auran- related lipopeptide antibiotics. Nat Prod Rep. 2005;22:717–41. tiaca: evidence for an iterative bacterial type I polyketide synthase and for 13. Koglin A, Doetsch V, Bernhard F. Molecular engineering aspects for the module skipping in nonribosomal peptide biosynthesis. ChemBioChem. production of new and modified biosurfactants. In: Soberón-Chávez G, 2005;6:375–85. editor. Biosurfactants. Berlin: Springer; 2010. p. 158–69. 29. Schneider A, Stachelhaus T, Marahiel M. Targeted alteration of the sub- 14. Baltz RH. Biosynthesis and genetic engineering of lipopeptides in Strepto- strate specificity of peptide synthetases by rational module swapping. myces roseosporus. Method Enzymol. 2009;458:511–31. Mol Gen Genet. 1998;257:308–18. 15. Nguyen KT, Ritz D, Gu J-Q, Alexander D, Chu M, Miao V, Brian P, Baltz RH. 30. de Ferra F, Rodriguez F, Tortora O, Tosi C, Grandi G. Engineering of peptide Combinatorial biosynthesis of novel antibiotics related to daptomycin. synthetases key role of the thioesterase-like domain for efficient produc- Prod Natl Acad Sci. 2006;103:17462–7. tion of recombinant peptides. J Biol Chem. 1997;272:25304–9. 16. Eppelmann K, Stachelhaus T, Marahiel MA. Exploitation of the selectivity- 31. Stachelhaus T, Schneider A, Marahiel MA. Rational design of peptide conferring code of nonribosomal peptide synthetases for the rational antibiotics by targeted replacement of bacterial and fungal domains. design of novel peptide antibiotics. Biochemistry. 2002;41:9718–26. Science. 1995;269:69–72. 17. Kries H, Niquille DL, Hilvert D. A subdomain swap strategy for reengineer- ing nonribosomal peptides. Chem Biol. 2015;22:640–8. 18. Liu H, Gao L, Han J, Ma Z, Lu Z, Dai C, Zhang C, Bie X. Biocombinatorial synthesis of novel lipopeptides by COM domain-mediated reprogram- ming of the plipastatin NRPS complex. Front Microbiol. 2016;7:1801. Ready to submit your research ? Choose BMC and benefit from: fast, convenient online submission thorough peer review by experienced researchers in your ﬁeld rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions
Microbial Cell Factories – Springer Journals
Published: May 31, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera