Communication Cite This: Biochemistry XXXX, XXX, XXX−XXX
pubs.acs.org/biochemistry
Specificity of Nonribosomal Peptide Synthetases in the Biosynthesis of the Pseudomonas virulence factor Gina L. Morgan, Ashley M. Kretsch, Kevin C. Santa Maria, Savannah J. Weeks,† and Bo Li* Department of Chemistry, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599, United States
Downloaded via KEAN UNIV on August 3, 2019 at 09:18:10 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
ABSTRACT: The Pseudomonas virulence factor (pvf) biosynthetic operon has been implicated in bacterial virulence and signaling. We identified 308 bacterial strains containing pvf homologues that likely produce signaling molecules with distinct structures and biological activities. Several homologues of the nonribosomal peptide synthetase (NRPS), PvfC, were biochemically characterized and shown to activate L-Val or L-Leu. The amino acid selectivity of PvfC and its homologues likely direct pvf signaling activity. We explored the natural diversity of the active site residues present in 92% of the adenylation domains of PvfC homologues and identified key residues for substrate selection and catalysis. Sequence similarity network (SSN) analysis revealed grouping of PvfC homologues that harbor the same active site residues and activate the same amino acids. Our work identified PvfC as a gatekeeper for the structure and bioactivity of the pvf-produced signaling molecules. The combination of active site residue identification and SSN analysis can improve the prediction of aliphatic amino acid substrates for NRPS adenylation domains. Figure 1. pvf-containing strains produce molecules with distinct signaling activities. (A) pvf from P. entomophila L48 (top) and a homologous cluster from B. cenocepacia H111 (bottom). (B) Complementation of L48 Δpvf C::Pmnl-lacZ with cell-free supernatants of pvf-containing strains from left to right: P. entomophila L48 (black), B. cenocepacia HI2424 (dark gray), P. f luorescens Pf0-1 (gray), P. f luorescens WH6 (light gray), and P. syringae pv tomato DC3000 (white). WT and Δpvf C represent the promoter activity of L48::PmnllacZ (diagonal lines, positive control) and L48 Δpvf C::Pmnl-lacZ (dots, background), respectively, without complementation. Significance was calculated using an unpaired t test.
B
acteria can communicate via small molecule signals. For example, pathogenic bacteria produce signaling molecules to coordinate the production of toxins, virulence factors, and other secondary metabolites.1−3 A variety of signaling molecules have been identified, such as the acyl-homoserine lactones. Signaling molecules within the same class, such as N3-oxododecanoyl-homoserine lactone and N-butanoyl-homoserine lactone, can exhibit differential signaling activity to enable specific intra- or interspecies communication.1,2 The Pseudomonas virulence factor (pvf) operon is widely conserved in Pseudomonas and other proteobacteria. pvf was identified through genetic screens to be important for signaling and virulence.4−6 Disruption of pvf in Pseudomonas entomophila L48 and the homologous cluster, mgo, in Pseudomonas syringae pv. syringae UMAF0158 significantly attenuated the virulence of L48 and UMAF0158 toward adult flies and tomato plants, respectively.4−6 pvf homologues are required for the biocontrol activity of Pseudomonas f luorescens strain X7 and the antifungal activity of Burkholderia cenocepacia H111.8 pvf is primarily comprised of four genes: a nonribosomal peptide synthetase (NRPS, pvf C), a diiron N-oxygenase (pvf B), and two genes of unknown function (pvfA and pvf D); a fifth gene (pvf E) is present in some strains such as B. cenocepacia H111 (Figure 1A). We recently characterized the © XXXX American Chemical Society
chemistry of pvf-encoded enzymes from L48 and showed that PvfC selectively activates and incorporates L-Val into three (dihydro)pyrazine N-oxides [(d)PNOs (Figure S1)].9 However, (d)PNOs did not restore the virulence of the pvf deletion strain of L48, suggesting that these molecules are not the final virulence factors.9 Similar chemistry may be involved in the biosynthesis of the pvf-related molecules fragin, an antifungal molecule, and valdiazen, a valinol diazeniumdiolate exhibiting moderate regulatory activity for fragin production in B. Special Issue: Current Topics in Mechanistic Enzymology 2019 Received: April 21, 2019 Revised: June 2, 2019
A
DOI: 10.1021/acs.biochem.9b00360 Biochemistry XXXX, XXX, XXX−XXX
Communication
Biochemistry cenocepacia H111 (Figure S1).8 These findings suggest that PvfC and its homologues play an essential role in incorporating the amino acid precursors into a diversity of small molecules. Here we explored the diversity of small molecules produced by pvf-encoded enzymes and identified important residues for the PvfC amino acid selection. Through genome mining, we identified pvf in the genome of 308 bacterial strains, most of which are Pseudomonas and other proteobacteria (Data Set S1). We assessed the signaling activity of small molecules produced by pvf-encoded enzymes in five representative strains: P. entomophila L48, B. cenocepacia HI2424, P. f luorescens Pf0-1, P. f luorescens WH6, and P. syringae pv. tomato DC3000. In L48, pvf regulates the expression of monalysin (mnl), a secreted pore-forming toxin that contributes to pathogenicity.10 We constructed two transcriptional reporter strains that place the mnl promoter upstream of a lacZ gene in the wild-type P. entomophila L48 (L48::P mnl -lacZ) or the pvf C deletion mutant (L48 Δpvf C::Pmnl-lacZ). Signaling activity was assessed by supplementing cultures of L48 Δpvf C::Pmnl-lacZ, which cannot produce pvf signaling molecules, with the cell-free supernatant of different bacterial strains. Supplementation with the supernatant of wild-type L48 restored the promoter activity of L48 Δpvf C::P mnl-lacZ to wild-type reporter levels, confirming detection of the pvf-produced signaling molecules (Figure 1B). The cell-free supernatant of the B. cenocepacia HI2424 culture also induced lacZ transcription in L48 Δpvf C::Pmnl-lacZ, suggesting that HI2424 produces signaling molecules similar to those of L48 that can be sensed by L48. The supernatant of P. syringae and P. fluorescens strains (Pf0-1 and WH6, respectively) did not induce lacZ transcription (Figure 1B and Figure S2), suggesting that the pvf homologues in these strains produce distinct signaling molecules from L48 and HI2424. These results indicate that pvf-encoded enzymes in different proteobacteria produce at least two distinct variants of signaling molecules. We hypothesized that homologues of the NRPS, PvfC, may incorporate different amino acids into the signaling molecules produced by pvf-encoded enzymes (Figure S1). To test this hypothesis, we studied the adenylating activity of PvfC. NRPSs are large multimodular enzymes that produce complex natural products with diverse biological activity.11−14 PvfC contains three domains: adenylation (A), thiolation (T), and reductase (R). A domains are responsible for NRPS amino acid specificity and typically act as gatekeepers in determining NRPS substrates.15−19 Specificity-conferring codes for A domains comprise 8−10 amino acids in the proximity of the active site that significantly contribute to substrate specificity, such as the 10-amino acid code.20−22 This code is widely used to predict A-domain specificity; however, bioinformatic prediction is often unable to differentiate aliphatic amino acids due to the generally hydrophobic 10-amino acid codes of the A domains for these substrates.21 We analyzed the 10-amino acid codes of the A domain of PvfC homologues in the 308 pvf-containing bacterial strains and identified 14 codes, all of which are predicted to activate LLeu using NRPSpredictor2 (Table 1, Table S3, and Data Set S1).23 This prediction is in contrast to the specificity for L-Val that we previously determined for PvfC from P. entomophila L48.9 Therefore, the prediction is likely unable to distinguish homologues that are specific for L-Val from those for L-Leu. To refine these predictions, we constructed a protein sequence similarity network (SSN) of PvfC homologues from all 308
Table 1. Ten-Amino Acid Codes of a Selection of PvfC Homologues
strains using the Enzyme Function Initiative-Enzyme Similarity tool (Figure 2).24 Homologues with the same 10-amino acid code cluster together with few exceptions. PvfC homologues from phylogenetically related bacteria also form clusters, with
Figure 2. Sequence similarity network of PvfC homologues. Nodes represent PvfC homologues from different pvf-containing strains. Colors represent different 10-amino acid codes assigned in Table 1. The alignment cutoff is 550 (approximately 85% sequence identity). Arrows indicate PvfC homologues from strains examined in this work using the mnl promoter reporter (DC3000 and Pf0-1 are in green clusters, HI2424 and L48 are in red clusters, and WH6 is a blue singleton). B
DOI: 10.1021/acs.biochem.9b00360 Biochemistry XXXX, XXX, XXX−XXX
Communication
Biochemistry the largest group being plant pathogens P. syringae. Similarly, human pathogens Burkholderia and Pandoraea spp. each form their own clusters (Figure 2). In contrast, homologues from P. f luorescens are spread across multiple clusters (Figure 2). We selected several PvfC homologues to biochemically characterize amino acid preference (Figure 3A). Homologues
L48, likely because both Pf0-1 and WH6 strains produce signaling molecules derived from L-Leu while L48 produces and senses signaling molecules derived from L-Val. Furthermore, the A domain of the PvfC homologue from B. cenocepacia HI2424 harbors the same 10-amino acid code as P. entomophila L48 (code 1), whereas the one from P. syringae pv. tomato DC3000 harbors the same 10-amino acid code as Pf0-1 (code 6). The 10-amino acid code analysis and the incell reporter activity suggest that the PvfC homologues from HI2424 and DC3000 activate L-Val and L-Leu, respectively. These results prompted us to investigate the role of the 10amino acid code in determining A-domain specificity for aliphatic amino acids by site-directed mutagenesis. We generated single-point mutations of PvfCPf0‑1 and PvfCL48 based on the natural diversity in the 10-amino acid code of PvfC A domains. Four codes (10-amino acid codes 3, 4, 7, and 8) can be achieved by mutating methionine 429 of PvfCPf0‑1 at position 5 of the 10-amino acid code to cysteine, isoleucine, threonine, or valine (PvfCPf0‑1-M429C, -M429I, -M429T, or -M429V, respectively), while one code (10-amino acid code 13) can be achieved by mutating valine 444 of PvfCL48 at position 5 to isoleucine (PvfCL48-V444I). We characterized the amino acid preference of each mutant by ATP−[32P]PPi exchange (Figures S10−S19). All four PvfCPf0‑1 mutants prefer L-Leu, while PvfCL48-V444I prefers L-Val (Table 2 and Figures S10−S19). The PvfCL48-V444I end-point assay showed significant activity toward L-Ile; however, steady-state kinetic analysis shows that like wild-type PvfCL48 (Figures S4, S6), PvfCL48-V444I is more selective toward L-Val than toward L-Ile, suggesting L-Val remains the preferred substrate (Table 2 and Figures S18 and S19). Our results support that position 5 of the 10-amino acid code is a “wobble-like position”, which does not significantly affect A-domain specificity.20,27 Therefore, PvfC homologues with 10-amino acid code 1 or 13 (represented by PvfCL48 or PvfCL48-V444I, respectively) are likely specific for L-Val, while those with 10-amino acid codes 3−8 (represented by PvfCPf0‑1-M429C, PvfCPf0‑1-M249I, PvfCAWH6, PvfCPf0‑1, PvfCPf0‑1-M429T, and PvfCPf0‑1-M429V, respectively) are likely specific for L-Leu. These codes are found in 92% of the 308 PvfC homologues (Tables S3 and S4, Data Set S1). A total of 77% of these PvfC homologues harbor codes 3−8 and likely incorporate L-Leu, while 23% harbor codes 1 and 13 and likely incorporate L-Val. We further determined the role of the 10-amino acid code in PvfC specificity by swapping the codes for L-Val and L-Leu. The 10-amino acid codes of PvfCL48 and PvfCPf0‑1 differ by four amino acids at positions 2, 3, 5, and 7 (Table 1, codes 1 and 6). Therefore, a PvfCL48-A381S/L384I/V444M/I467L quadruple mutant [PvfCL48-SIML (Figure 4A)] was generated to match the 10-amino acid code for PvfCPf0‑1 that specifies LLeu. Conversely, a PvfCPf0‑1-S366A/I369L/M429V/L452I quadruple mutant (PvfCPf0‑1-ALVI) was generated to match the 10-amino acid code for PvfCL48 that specifies L-Val. An amino acid screen showed that PvfCL48-SIML now prefers LLeu over L-Val (Figure 4B). Steady-state kinetic analysis confirmed a 6-fold higher kcat/KM for L-Leu than for L-Val (Table 2 and Figure S20). However, the KM for L-Leu increased 2-fold, and the overall kcat/KM of PvfCL48-SIML is 2 orders of magnitude lower than that of the wild-type enzyme, indicating a significant decrease in overall catalytic proficiency (Table 2, Table S5, and Figure S20). PvfCPf0‑1-ALVI still prefers L-Leu over L-Val. The overall kcat/KM decreased by 100fold, and no activity was observed toward L-Val (Table 2, Table
Figure 3. PvfCL48 is selective for L-Val. PvfCPf0‑1 is selective for L-Leu. (A) Amino acid adenylation and thiolation activity of PvfC. (B) Steady-state kinetic measurements of PvfCL48 toward L-Val (black) and L-Leu (blue). (C) Steady-state kinetic measurements of PvfCPf0‑1 toward L-Leu (blue) and L-Val (black).
of pvf C were cloned and expressed from three strains, each with a different 10-amino acid code, including P. entomophila L48 (pvf CL48, code 1), P. fluorescens WH6 (pvf CWH6, code 5), and P. fluorescens Pf0-1 (pvf CPf 0‑1, code 6) (Figure S3). Because of the solubility problems of full-length pvf CWH6, it was expressed as a truncate including only the A domain (pvf C-AWH6). The amino acid specificity was determined using ATP−[32P]PPi exchange25,26 against a panel of 20 amino acids. Both PvfCL48 and PvfCPf0‑1 demonstrate. Kinetic promiscuity toward L-Leu and L-Val in an end-point assay (Figures S4 and S5); therefore, steady-state kinetic analysis was performed using L-Leu and L-Val as substrates (Table 2, Figure 3B,C). PvfCL48 prefers L-Val. The kcat/KM is 120-fold higher for L-Val than for L-Leu (Table 2, Figure 3B, and Figure S6). PvfCPf0‑1 and PvfC-AWH6 both prefer L-Leu. The kcat/KM of PvfCPf0‑1 is 540-fold higher for L-Leu than for L-Val (Table 2, Figure 3C, and Figure S7). No activity was observed for PvfC-AWH6 toward L-Val (Figures S8 and S9). The in vitro specificities of the PvfC homologues are consistent with results from in-cell biological assay using the P. entomophila L48 reporter strain (Figure 1B). Neither P. f luorescens Pf0-1 nor WH6 culture supernatant activates the promoter reporter in P. entomophila C
DOI: 10.1021/acs.biochem.9b00360 Biochemistry XXXX, XXX, XXX−XXX
Communication
Biochemistry Table 2. Kinetic Parameters of PvfC Homologues construct PvfCPf0‑1
substrate
KM (mM)
1/KM (M−1)
kcat (s−1)
L-Leu
0.13 ± 0.03 2.4 ± 0.8 0.24 ± 0.01 0.6 ± 0.4 0.17 ± 0.03 N/A 0.31 ± 0.03 2±1 0.42 ± 0.06 1.5 ± 0.4 1.3 ± 0.4 N/A 7±3 1.2 ± 0.8 8±2 3±1 1.5 ± 0.2 3.7 ± 0.7 15 ± 9 1.2 ± 0.7 2.1 ± 0.6
7900 ± 1300 430 ± 70 4100 ± 100 3100 ± 2400 6100 ± 1400 N/A 3200 ± 270 720 ± 470 2500 ± 640 690 ± 70 860 ± 290 N/A 150 ± 10 680 ± 60 130 ± 20 410 ± 150 670 ± 30 280 ± 110 210 ± 100 1000 ± 450 560 ± 80
11.9 ± 0.5 0.40 ± 0.04 17.0 ± 0.2 0.17 ± 0.03 15.2 ± 0.5 N/A 11.6 ± 0.3 0.18 ± 0.03 16.5 ± 0.6 0.19 ± 0.02 1.2 ± 0.1 N/A 0.65 ± 0.1 13.1 ± 0.7 1.3 ± 0.2 0.71 ± 0.09 10.7 ± 0.5 1.2 ± 0.09 2.2 ± 0.8 0.03 ± 0.01 1.6 ± 0.1
L-Val
PvfCPf0‑1-M429C
L-Leu L-Val
PvfCPf0‑1-M429I
L-Leu L-Val
PvfCPf0‑1-M429T
L-Leu L-Val
PvfCPf0‑1-M429V
L-Leu L-Val
PvfCPf0‑1-ALVI
L-Leu L-Val
PvfCL48
L-Leu L-Val L-Ile
PvfCL48-V444I
L-Leu L-Val L-Ile
PvfCL48-SIML
L-Leu L-Val
PvfC-AWH6
L-Leu
kcat/KM (M−1 s−1) 9.2 × 170 7.1 × 280 8.9 × N/A 3.7 × 90 3.9 × 130 920 N/A 93 1.1 × 160 240 7.1 × 320 150 25 760
104 104 104 104 104
104
103
ALVI toward L-Leu provides further evidence for the importance of positions 2, 3, and 7 of the 10-amino acid code in overall enzyme activity. Our study of PvfC and homologues revealed important insight into substrate selection of branched-chain aliphatic amino acids by NRPS A domains. Position 5 of the 10-amino acid code has no effect on substrate preference, while positions 2, 3, and 7 can affect both substrate selectivity and catalytic efficiency. Further studies are needed to define the role of these residues in substrate selection. Our work also suggests that 10-amino acid code analysis has limitations in predicting the precise aliphatic amino acids for NRPS adenylation domains and biochemical characterization is necessary to confirm substrate specificity. SSN analysis showed that PvfC homologues with the same 10-amino acid code group together and homologues specific for L-Val or L-Leu separate into different groups. Thus, SSN in combination with 10-amino acid code analysis can help refine predictions of amino acid substrates for homologous NRPSs. Engineering of NRPS A-domain specificity can modulate the structure and properties of the nonribosomal peptide products,28 as shown for andrimids,29 calcium-dependent peptide antibiotics,30 and luminmides,31 by introducing nonnative or non-natural amino acids.30,32−34 Here, we show that the A domain of the NRPS PvfC is a gatekeeper for the structure and activity of the signaling molecules produced by pvf-encoded enzymes. Future studies will focus on identifying the structures of these signaling molecules and determining if mutated PvfC can incorporate non-native amino acids into the molecules. In summary, our work combines cellular, bioinformatic, and biochemical studies to define the adenylating activities of a family of NRPSs that result in two distinct types of signaling molecules and to identify active site residues of these NRPSs that are essential for substrate specificity. This work advances the understanding of a key biosynthetic step in a widespread signaling pathway.
Figure 4. PvfCL48-SIML is selective for L-Leu. (A) Ten-amino acid code of PvfCL48 compared to that of PvfCL48-SIML. (B) Amino acid screen of PvfCL48-SIML. Activity is relative to the activation of L-Leu.
S5, and Figures S21 and S22). To assess if decreased enzyme activity resulted from changes in protein folding, we analyzed the secondary structures of wild-type PvfCL48 and the SIML mutant and wild-type PvfCPf0‑1 and the ALVI mutant by circular dichroism. The circular dichroism spectra indicate that the quadruple mutants share the same key structural features as their corresponding wild-type enzymes, suggesting the introduction of the mutations did not significantly alter their structures (Figures S23 and S24). The altered amino acid preference of PvfCL48-SIML from LVal to L-Leu suggests that positions 2, 3, and 7 of the 10-amino acid code play a key role in the substrate specificity of PvfCL48. However, the significant decrease in catalytic efficiency suggests that these residues also contribute to enzyme activity and that the substrate preference is dictated by not only the 10amino acid code but also the overall protein sequence and structure.27 The decreased catalytic efficiency of PvfCPf0‑1D
DOI: 10.1021/acs.biochem.9b00360 Biochemistry XXXX, XXX, XXX−XXX
Communication
Biochemistry
■
f luorescens strain X reveals novel genes regulated by glucose. PLoS One 8, No. e61808. (8) Jenul, C., Sieber, S., Daeppen, C., Mathew, A., Lardi, M., Pessi, G., Hoepfner, D., Neuburger, M., Linden, A., Gademann, K., and Eberl, L. (2018) Biosynthesis of fragin is controlled by a novel quorum sensing signal. Nat. Commun. 9, 1297. (9) Kretsch, A. M., Morgan, G. L., Tyrrell, J., Mevers, E., Vallet-Gely, I., and Li, B. (2018) Discovery of (dihydro)pyrazine N-Oxides via genome mining in Pseudomonas. Org. Lett. 20, 4791−4795. (10) Opota, O., Vallet-Gely, I., Vincentelli, R., Kellenberger, C., Iacovache, I., Gonzalez, M. R., Roussel, A., Van Der Goot, F. G., and Lemaitre, B. (2011) Monalysin, a novel β-pore-forming toxin from the Drosophila pathogen Pseudomonas entomophila, contributes to host intestinal damage and lethality. PLoS Pathog. 7, No. e1002259. (11) Robbel, L., and Marahiel, M. A. (2010) Daptomycin, a bacterial lipopeptide synthesized by a nonribosomal machinery. J. Biol. Chem. 285, 27501−8. (12) Cézard, C., Farvacques, N., and Sonnet, P. (2014) Chemistry and biology of pyoverdines, Pseudomonas primary siderophores. Curr. Med. Chem. 22, 165−186. (13) Felnagle, E. A., Jackson, E. E., Chan, Y. A., Podevels, A. M., Berti, A. D., McMahon, M. D., and Thomas, M. G. (2008) Nonribosomal peptide synthetases involved in the production of medically relevant natural products. Mol. Pharmaceutics 5, 191−211. (14) Dose, B., Niehs, S. P., Scherlach, K., Florez, L. V., Kaltenpoth, M., and Hertweck, C. (2018) Unexpected bacterial origin of the antibiotic icosalide: Two-tailed depsipeptide assembly in multifarious burkholderia symbionts. ACS Chem. Biol. 13, 2414−2420. (15) Strieker, M., Tanovic, A., and Marahiel, M. A. (2010) Nonribosomal peptide synthetases: Structures and dynamics. Curr. Opin. Struct. Biol. 20, 234−40. (16) Villiers, B., and Hollfelder, F. (2011) Directed evolution of a gatekeeper domain in nonribosomal peptide synthesis. Chem. Biol. 18, 1290−9. (17) Magarvey, N. A., Fortin, P. D., Thomas, P. M., Kelleher, N. L., and Walsh, C. T. (2008) Gatekeeping versus promiscuity in the early stages of the andrimid biosynthetic assembly line. ACS Chem. Biol. 3, 542−554. (18) Mcerlean, M., Overbay, J., and Van Lanen, S. (2019) Refining and expanding nonribosomal peptide synthetase function and mechanism. J. Ind. Microbiol. Biotechnol. 46, 493−513. (19) Sattely, E. S., Fischbach, M. A., and Walsh, C. T. (2008) Total biosynthesis: In vitro reconstitution of polyketide and nonribosomal peptide pathways. Nat. Prod. Rep. 25, 757−793. (20) Stachelhaus, T., Mootz, H. D., and Marahiel, M. A. (1999) The specificity-conferring code of adenylation domains in nonribosomal peptide synthetases. Chem. Biol. 6, 493−505. (21) Challis, G. L., Ravel, J., and Townsend, C. A. (2000) Predictive, structure-based model of amino acid recognition by nonribosomal peptide synthetase adenylation domains. Chem. Biol. 7, 211−224. (22) Rausch, C., Weber, T., Kohlbacher, O., Wohlleben, W., and Huson, D. H. (2005) Specificity prediction of adenylation domains in nonribosomal peptide synthetases (NRPS) using transductive support vector machines (TSVMs). Nucleic Acids Res. 33, 5799−5808. (23) Rottig, M., Medema, M. H., Blin, K., Weber, T., Rausch, C., and Kohlbacher, O. (2011) NRPSpredictor2–a web server for predicting NRPS adenylation domain specificity. Nucleic Acids Res. 39, W362−7. (24) Gerlt, J. A., Bouvier, J. T., Davidson, D. B., Imker, H. J., Sadkhin, B., Slater, D. R., and Whalen, K. L. (2015) Enzyme function initiative-enzyme similarity tool (EFI-EST): A web tool for generating protein sequence similarity networks. Biochim. Biophys. Acta, Proteins Proteomics 1854, 1019−37. (25) Lee, S. G., and Lipmann, F. (1975) Tyrocidine synthetase system. Methods Enzymol. 43, 585−602. (26) Linne, U., and Marahiel, M. A. (2004) Reactions catalyzed by mature and recombinant nonribosomal peptide synthetases. Methods Enzymol. 388, 293−315. (27) Eppelmann, K., Stachelhaus, T., and Marahiel, M. A. (2002) Exploitation of the selectivity-conferring code of nonribosomal
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biochem.9b00360. pvf-containing strains and 10-amino acid codes (Data Set S1) (XLSX) Detailed experimental procedures and supplemental figures (PDF) Accession Codes
PvfCL48, Q1IGU3; PvfCPf0‑1, Q3KK36; PvfCWH6, E2XJF6.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Bo Li: 0000-0002-8019-8891 Present Address †
S.J.W.: University of Florida College of Medicine, Gainesville, FL 32610. Funding
This work is supported by the National Institutes of Health (R00GM099904 and DP2HD094657 to B.L.), the Rita Allen Foundation, and The University of North Carolina at Chapel Hill. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors thank Peggy Cotter [The University of North Carolina at Chapel Hill (UNC)] for B. cenocepacia HI2424 and the pUC-lacZ/pTNS3 plasmids, Kristin Trippe (U.S. Department of Agriculture) for P. f luorescens WH6, Ashutosh Tripathy (UNC) for help with circular dichroism experiments, and Andrew Chan (UNC) for helpful discussion and reading of the manuscript.
■
REFERENCES
(1) Antunes, L. C., Ferreira, R. B., Buckner, M. M., and Finlay, B. B. (2010) Quorum sensing in bacterial virulence. Microbiology 156, 2271−82. (2) Papenfort, K., and Bassler, B. L. (2016) Quorum sensing signalresponse systems in gram-negative bacteria. Nat. Rev. Microbiol. 14, 576−88. (3) Abisado, R. G., Benomar, S., Klaus, J. R., Dandekar, A. A., and Chandler, J. R. (2018) Bacterial quorum sensing and microbial community interactions. mBio 9, 02331-17. (4) Vallet-Gely, I., Opota, O., Boniface, A., Novikov, A., and Lemaitre, B. (2010) A secondary metabolite acting as a signalling molecule controls Pseudomonas entomophila virulence. Cell. Microbiol. 12, 1666−79. (5) Arrebola, E., Cazorla, F. M., Romero, D., Pérez-García, A., and de Vicente, A. (2007) A nonribosomal peptide synthetase gene (mgoA) of Pseudomonas syringae pv. syringae is involved in mangotoxin biosynthesis and is required for full virulence. Mol. Plant-Microbe Interact. 20, 500−509. (6) Carrion, V. J., Arrebola, E., Cazorla, F. M., Murillo, J., and De Vicente, A. (2012) The mbo operon is specific and essential for biosynthesis of mangotoxin in Pseudomonas syringae. PLoS One 7, No. e36709. (7) Kremmydas, G. F., Tampakaki, A. P., and Georgakopoulos, D. G. (2013) Characterization of the biocontrol activity of Pseudomonas E
DOI: 10.1021/acs.biochem.9b00360 Biochemistry XXXX, XXX, XXX−XXX
Communication
Biochemistry peptide synthetases for the rational design of novel peptide antibiotics. Biochemistry 41, 9718−9726. (28) Winn, M., Fyans, J. K., Zhuo, Y., and Micklefield, J. (2016) Recent advances in engineering nonribosomal peptide assembly lines. Nat. Prod. Rep. 33, 317−47. (29) Evans, B. S., Chen, Y., Metcalf, W. W., Zhao, H., and Kelleher, N. L. (2011) Directed evolution of the nonribosomal peptide synthetase AdmK generates new andrimid derivatives in vivo. Chem. Biol. 18, 601−7. (30) Thirlway, J., Lewis, R., Nunns, L., Al Nakeeb, M., Styles, M., Struck, A. W., Smith, C. P., and Micklefield, J. (2012) Introduction of a non-natural amino acid into a nonribosomal peptide antibiotic by modification of adenylation domain specificity. Angew. Chem., Int. Ed. 51, 7181−4. (31) Bian, X., Plaza, A., Yan, F., Zhang, Y., and Muller, R. (2015) Rational and efficient site-directed mutagenesis of adenylation domain alters relative yields of luminmide derivatives in vivo. Biotechnol. Bioeng. 112, 1343−53. (32) Kries, H., Wachtel, R., Pabst, A., Wanner, B., Niquille, D., and Hilvert, D. (2014) Reprogramming nonribosomal peptide synthetases for ″clickable″ amino acids. Angew. Chem., Int. Ed. 53, 10105−8. (33) Kaljunen, H., Schiefelbein, S. H., Stummer, D., Kozak, S., Meijers, R., Christiansen, G., and Rentmeister, A. (2015) Structural elucidation of the bispecificity of A domains as a basis for activating non-natural amino acids. Angew. Chem., Int. Ed. 54, 8833−6. (34) Niquille, D. L., Hansen, D. A., Mori, T., Fercher, D., Kries, H., and Hilvert, D. (2018) Nonribosomal biosynthesis of backbonemodified peptides. Nat. Chem. 10, 282−287.
F
DOI: 10.1021/acs.biochem.9b00360 Biochemistry XXXX, XXX, XXX−XXX
