AEM Accepted Manuscript Posted Online 10 July 2015 Appl. Environ. Microbiol. doi:10.1128/AEM.01639-15 Copyright © 2015, American Society for Microbiology. All Rights Reserved.
2
Unusual biosynthesis and structure of locillomycins from Bacillus subtilis 916
3
Chuping Luo a,b*, Xuehui Liuc*, Xian Zhoud, Junyao Guo a, John Truongd, Xiaoyu Wanga ,
4
Huafei Zhou a , Xiangqian Li*b, Zhiyi , Chena*
5
a
6
China
7
b
8
223003, China
9
c
10
d
11
2751, Australia
12
Corresponding Authors:
13
*Chuping Luo, E-mail:
[email protected], Tel: +86-0517-89221726,
14
*Xuehui Liu, E-mail:
[email protected], Tel: +86-010-64887727
15
*Xiangqian Li, E-mail:
[email protected], Tel: +86-0517-89221726
16
*Zhiyi Chen, E-mail:
[email protected] , Tel: +86-025-84390393.
1
Institute of Plant Protection, Jiangsu Academy of Agricultural Sciences, Nanjing 210014，
School of Life Science and Chemical Engineering, Huaiyin Institute of Technology, Huaian
Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, China The National Institute of Complementary Medicine, University of Western Sydney, Sydney
17
1
18
ABSTRACT: Three families of Bacillus cyclic lipopeptides – surfactins, iturins and fengycins –
19
have well-recognized potential uses in biotechnology and biopharmaceutical applications. This
20
study outlines the isolation and characterization of locillomycins, a novel family of cyclic
21
lipopeptide produced by B. subtilis 916. Structure elucidation of locillomycins revealed several
22
molecular features not observed in other Bacillus lipopeptides, including a unique nonapeptide
23
sequence and macrocyclization. Locillomycins are active against bacteria and viruses.
24
Biochemical analysis and gene deletions studies have supported the assignment of a 38 kb gene
25
cluster as the locillomycins biosynthetic gene cluster. Interestingly, this gene cluster encodes 4
26
proteins (namely LocA, LocB, LocC and LocD) using a hexamodular nonribosomal peptide
27
synthetase to biosynthesize cyclic nonapeptides. Genome analysis and the chemical structures of
28
the end products indicated that the biosynthetic pathway exhibited two distinct features: (i) a
29
non-linear hexamodular assembly line with three modules in the middle utilized twice while the
30
first and last two modules were used only once; and (ii) several domains are skipped or
31
optionally selected.
32
2
33
In the competition for nutrients, members of the Bacillus genus often produce a vast array
34
of biologically active molecules that potentially inhibit the development of competing organisms.
35
The gram-positive bacterium Bacillus subtilis has an average of 4 – 5% of its genome devoted to
36
antibiotic synthesis and is able to produce more than two dozen antibiotics with an amazing
37
variety of structures.(1) Many of these compounds, which have a peptide origin, are synthesized
38
either ribosomally or non-ribosomally. Among the non-ribosomally generated amphipathic
39
cyclic lipopeptides, surfactins, iturins and fengycins have well-recognized potential applications
40
in biotechnology and biopharmaceutical products due to their antagonistic activities and
41
surfactant properties.(2, 3) Furthermore, the mechanisms behind the observed biocontrol efficacy
42
of the different Bacillus strains have also been well described.(4–6) Lipopeptides are able to
43
induce systemic resistance in plants and facilitate the multicellular behaviors of the producing
44
strains such as swarming motility, biofilm formation and colony morphology.(5–7)
45
Surfactins, iturins and fengycins are synthesized by nonribosomal peptide synthetases
46
(NRPSs) which exhibit a distinct modular architecture.(2, 8–10) A module is typically composed
47
of three core domains with each domain responsible for a certain biochemical reaction.(11)
48
Specifically, the amino acid adenylation domain (A domain) controls the entry of the substrates
49
into the peptide by recognizing and activating a specific amino acid. The thiolation domain (T
50
domain), also referred to as the peptidyl carrier protein (PCP), contains an invariant serine
51
residue which is essential for the binding of a 4'-phosphopantetheine cofactor. An N-terminal
3
52
condensation domain (C domain) is required for the coupling of two consecutively bound amino
53
acids.(12, 13) These three domains constitute a minimal elongation module, the basic repetitive
54
unit of a multi-modular NRPS. Furthermore, modules can be supplemented with domains that
55
catalyze the modification of the activated amino acid, such as N-methylation and epimerization.
56
In some cases, when the first module of a NRPS complex lacks a C-domain, the last module
57
contains a termination thioesterase domain (TE domain) to release the end product.(14) Order
58
and specificity of the modules within the protein template determines the sequence of the product
59
(type A, linear NRPS).(8, 11) Genetic and biochemical analyses have revealed that the modular
60
arrangement of most lipopeptide synthetases is colinear with the amino acid sequences of
61
lipopeptides.(1, 2) This assembly-line arrangement of the conserved catalytic modules and
62
domains provide the means to construct hybrid NRPS used to synthesize new lipopeptide
63
compounds.(15–18) The prospect of creating numerous bioactive lipopeptides by engineering
64
existing lipopeptide synthetases has stimulated the search for new NRPS responsible for
65
lipopeptide synthesis.(19–24)
66
To date, only two reported kinds of biosynthetic machinery within the NRPS assembly line
67
do not conform to the rule of colinearity. These include the type B and type C NRPS which
68
iteratively use all of their modules or certain domains during the assembly of a product,
69
respectively.(25–28) While our understanding of nonlinear NRPS biosynthetic mechanism is
70
limited, it is clear that this mechanism has great potential to introduce structural diversity to the
4
71
secondary metabolites through combinatorial biosynthesis.(25, 33, 36) It is necessary to propose
72
nonlinear NRPS biosynthetic models that can further unravel the details of this biosynthetic
73
mechanism. The gram-positive bacterium B. subtilis serves as a model organism and is
74
intensively used in the heterologous expression of commercial metabolites.(36–38) However,
75
apart from this study, no non-linear NRPS assembly line has been observed in this well-
76
characterized strain.
77
The commercial strain of B. subtilis 916 was isolated from paddy soils of Jurong County,
78
Jiangsu and has been reported to be effective in the biocontrol of plant diseases (39). While
79
NRPSs for LPs production in B. subtilis 916 have been reported recently and the locillomycin
80
also described briefly (40), in this study, we describe in detail how to elucidate the novel
81
structure and unique biosynthesis of this new family of nonribosomal lipopeptides locillomycins
82
and its biological function. In this study, our results strongly suggest that locillomycins revealed
83
molecular features not observed in the three families of surfactins, iturins and fengycins,
84
including a unique nonapeptide sequence and macrocyclization. Locillomycins are active against
85
bacteria and viruses with low hemolytic activity and can be used for therapeutic applications as
86
natural health products. While the end products were straightforward, the proposed biosynthetic
87
pathway of locillomycins contained several atypical aspects that were not predictable by
88
bioinformatics analysis. Our study thus adds to the knowledge regarding non-linear NRPS
5
89
biosynthesis used by B. subtilis and further broadens the potential of strain 916 as a model
90
system for nonlinear NRPS studies.
91
6
92
MATERIALS AND METHODS
93
Cultivation, Isolation and Purification. Luria-Bertani (LB) broth was inoculated with B.
94
subtilis 916 to a concentration of 5.5 × 105 cells/mL. One hundred cultures of 330 mL each in 2-
95
liter Erlenmeyer flasks were incubated with 180 rpm agitation at 28 ºC for three days. After
96
removing the biomass by centrifugation, the broth was titrated to pH 2.8 with concentrated
97
hydrochloric acid and the resulting grey precipitate was extracted with 150 mL MeOH. The
98
MeOH extraction was added with 300 mL H2O and titrated to pH 7 with 5 M NaOH. The
99
extracted impurities were added to an Agilent amino solid phase extraction column and
100
sequentially washed with 50% (v/v) MeOH-H2O, 100% MeOH, 1% (v/v) formic acid-MeOH
101
and 2% (v/v) formic acid-MeOH. The 2% (v/v) formic acid-MeOH elution was concentrated by
102
nitrogen drying to 10 mg/mL and loaded on an Agilent C18 solid phase extraction column and
103
sequentially washed with 40, 42, 46, and 48% (v/v) MeOH-H2O. The 44, 46, and 48% (v/v)
104
MeOH-H2O elutions were concentrated by nitrogen drying to 20 mg/mL. These fractions were
105
further processed by HPLC by repetitive 50 µL injections using a reversed-phase column (RP-
106
18, 5 µM, 4 × 250 mm; Merck) with a 0.5 mL/min flow rate and monitored at 230 nm. Fraction I
107
yielded derivatives A which was collected at an average peak retention time of 9 min. Fraction II
108
yielded derivatives B at a retention time of 13 min. Fraction III yielded derivatives C at a
109
retention time of 18 min. Both eluted with 50% (v/v) CH3CN, 0.5% (v/v) trifluoroacetic acid in
7
110
H2O. All collections were concentrated by nitrogen and vacuum drying and resulted in the
111
purified locillomycin A, B and C derivatives.
112
NMR Structure Determination. All NMR spectra were acquired at 25 ºC on a Bruker
113
AVANCE III 600MHz, an Agilent DD2 500MHz (for 13C detected spectra) or an Agilent DD2
114
600MHz spectrometer equipped with a cold-probe. Locillomycin A (~8 mg) were dissolved in
115
500 µL aqueous solvent (20 mM phosphate buffer, pH 6.5, D2O/H2O, (9:1, v/v)) while ~8 mg of
116
Locillomycin B and C respectively was dissolved in 500 µL CD3OH to make the NMR samples.
117
For each sample, a series of 1D and 2D spectra, including 1H, 13C, DEPT, 1H-1H COSY, 1H-1H
118
TOCSY, 1H-1H ROESY, 1H-13C HSQC, 1H-13C HMBC, were acquired for the structure
119
elucidation. The spin-lock time for each TOCSY experiment was 80 ms and the mixing time for
120
each ROESY was 200 ms. All NMR data were processed and analyzed with MestReNova. The
121
chemical shifts in the aqueous solvent were referenced to the HOD peak at 4.77 ppm while those
122
in the CD3OH were referenced to the peak of the methyl group at 3.30 ppm. Chemical shifts of
123
13
124
for the peptide part was done using the protocol developed by Wuethrich.(46)
125
Gene Disruption in B. subtilis 916 and mutant analysis. In vivo generation of targeted
126
mutations in B. subtilis 916 were achieved by a modified protocol originally developed for B.
127
subtilis 168.9 The knockout plasmids were constructed from vector pUC19 by inserting
128
homologous fragments from B. subtilis 916 genomic DNA and a neomycin cassette from
C were referenced indirectly(48) or to the peak of CD3OH at 49 ppm. Sequential assignment
8
129
pBEST501. An example for disruption of LocD (ΔlocD) is detailed below. A 2.5 kb fragment
130
was amplified by PCR using the primers LocDF and LocDR and cloned into vector pUC19,
131
generating pLocD1. The latter was digested with xbaI, which cuts in the middle of the PCR
132
fragment. Simultaneously, pBEST501 with the same enzyme to obtain the neomycin cassette
133
(~1.3 kb), which was then ligated to pLocD1, resulting in pLocD2. This was subsequently
134
transformed into the naturally competent B. subtilis 916, where it was introduced into the
135
genome via double-crossover homologous recombination. The disruption locD gene was
136
demonstrated in the resistant colonies obtained by PCR with appropriate primers and by
137
Southern hybridization. The locillomycins were extracts from mutants and further analyzed by
138
HPLC/MS as described previously.
139
Heterologous expression and purification of internal adenylation domains. The pET28a
140
expression system was used to clone PCR products in which Nco I and Xho I sites were
141
introduced to by the PCR with 5’-modified primers (Table 1). PCR from chromosomal B. subtilis
142
916 was carried out with 50 µL reaction volumes containing 2.5 U
143
µL 10× Mg-free reaction buffer, 200 µM dNTP, 2.5 mM MgCl2 and 1 µM each of primers.
144
Amplification conditions were: 30 cycles of 95 °C for 30 s, 55 °C for 30 s and 72°C for 3 min,
145
followed by a final extension at 72 °C for 10 min. PCR products were digested with NcoI and
146
XhoI, and cloned into the digested pET28a vector. PCR products cloned into the XhoI site of
147
pET28a results in appending the poly-Histag at the carboxyl end of recombinant proteins.
EXTaq
DNA polymerase, 5
9
148
Transformants of E. coli BL21(DE3) were selected by growth at 28 °C in LB media containing
149
100 µg/mL ampicillin. Cells were induced with 1mM IPTG at an OD600 of 0.6 and allowed to
150
grow for an ad
151
ditional 4 h before being harvested. Purification of the His6-tagged proteins was carried out by
152
Ni2+-affinity chromatography. They were separated using 10% (w/v) SDS–PAGE, stained with
153
Coomassie blue, and analyzed by density scanning using an image analysis system (Bio-Rad).
154
Pyrophosphate Exchange Assay. Amino acid-dependent ATP-sodium pyrophosphate assays
155
were performed by using the spectrophotometric assay furnished by the EnzChek Pyrophosphate
156
Assay Kit (Molecular Probes). A 200 µL reaction contained 75 mM Tris•HCl (pH 7.5), 10 mM
157
MgCl2, 5 mM DTT, 5 mM ATP, 400 mM MesG, 0.2 units purine nucleoside phosphorylase, and
158
0.2 units inorganic pyrophosphatase, 2 mM amino acid. Reaction without amino acid was as the
159
control. Reaction was initiated by addition of enzyme and incubated at 30 °C for 30 min.
160
Reading of absorbance at 360 nm was measured in a Perkin-Elmer Lambda 6-vis
161
spectrophotometer. All reactions were performed in triplicate.
162
Nucleotide sequence accession number. The complete genome sequence of B. subtilis 916 and
163
the gene cluster for biosynthesis locillomycins described in this work have been submitted to
164
GenBank under accession numbers CP009611 and KF866134, respectively.
165
RESULTS AND DISCUSSION
10
166
Identification and in vivo gene disruption of loc gene cluster. Identification and
167
bioinformatics analyses of the loc gene cluster were employed as described in our previous work.
168
(40) The draft genome and complete genome sequence of B. subtilis 916 was recently analyzed
169
(GenBank no. AFSU00000000.1 and CP009611) and four NRPS gene clusters were identified
170
(Figure S1).(41) In addition to the three conventional gene clusters, srf (for surfactins), bmy (for
171
bacillomycin Ls) and fen (for fengycins), the B. subtilis 916 also contains a fourth NRPSs gene
172
cluster, loc (GenBank no. KF866134) (40). Further analysis of the loc open reading frames led to
173
the conclusion that the loc gene cluster was a potential candidate for the biosynthesis of an
174
unknown family of lipopepetides called locillomycins. Interestingly, this particular locus in B.
175
subtilis 916 shares a high degree of homology (>95% identity) with the same locus found in B.
176
subtilis 168 and B. amyloquefaciens FZB42 strains (Figure 1).
177
Four NRPS genes designated locD, A, B and C are contiguous in the loc gene cluster and
178
correspond to the six peptide extension modules (Figure 1). The first gene, locD, encodes a 145.8
179
kDa protein containing the polyketide synthase module, and includes domains with homology to
180
the proteins involved in the synthesis of fatty acids and polyketides: fatty acid acyl-CoA
181
synthetase domain (ACS), acyl carrier protein/thiolation domain (T) and β-keto acyl synthetase
182
domain (KAS). The second gene, locA, encodes a 267.4 kDa protein with only one extension
183
module containing the core elongation domains: C domain, A domain and T domain. Two C
184
domains exist independently in the upstream of the A domain. The third gene, locB, encodes a
11
185
443.1 kDa protein containing three extension modules. The first module contains a predicted
186
epimerization (E) domain, suggesting that the corresponding amino acid appears in the D-
187
configuration. The fourth gene, locC, encoding a 243.9 kDa protein, contains two extension
188
modules and a termination module, which contains a C-terminal TE domain. A similarity
189
analysis also showed that the nucleotide sequence of loc gene cluster has a low similarity ( B > A) ( Table S3 – S5 and Figure S6 – S8). This
251
difference is derived from the different lengths of the long-chain acyl group shown in Figure S10
252
and the method of structural determination will not be detailed here.
253
Unlike the surfactin and iturin lipopeptide families, in which the peptide residues are
254
interlinked with a β-hydroxy or a β-amino fatty acid to form a cyclic lactone or lactam, the
255
locillomycins form an internal lactone ring within the peptidic moiety. This construction is more
256
similar to the fengycin family which also forms a lactone without the participation of a β-
257
hydroxyl fatty acid. Importantly, however, the number of residues in locillomycins is nine rather
258
than ten as observed in the fengycin family (eight out of ten residues form the lactone ring)
259
(Figure 4). These nine-membered cyclic lipopeptides are also different from the iturin and
260
surfactin families, which are seven-membered rings. Furthermore, only one D-amino acid (D-
261
Gln2) is incorporated in the locillomycin family, while all the members of the other three
15
262
families possess more than two D-amino acid residues. The length of the fatty acid chains is
263
similar to other Bacillus lipopeptides.
264
Biochemical investigation of internal adenylation domains. To further elucidate the
265
biosynthetic process of locillomycin, DNA fragments coding the adenylation domains of
266
modules LocA1 (A1), LocB1 (A2), LocB2 (A3), LocB3 (A4), LocC1 (A5), LocC2 (A6) (Figure
267
6) were amplified from chromosomal B. subtilis 916 DNA and cloned into IPTG-inducible
268
expression as described in Experimental section. The constructs were confirmed by sequencing
269
fusion sites. The enzyme fragments were overexpressed in Escherichia coli (E. coli) as His6-
270
tagged proteins and purified by Ni2+-affinity chromatography. All proteins were found within the
271
soluble fraction after French press lysis of E. coli cells, as confirmed from Coomassie blue-
272
stained SDS-polyacrylamide gels (Figure 5A) with calculated molecular masses of 60.0 to 65.0
273
kDa. The purified adenylation domains were biochemically investigated with respect to their
274
activity and specificity in the ATP-PPi-exchange reaction. In order to determine amino acid
275
specificity, each protein was incubated with all constituent amino acids of locillomycins (control
276
without amino acid). The six adenylation domains were found to activate the amino acids
277
corresponding to the constituent amino acids of locillomycins. Specifically, A1 exclusively
278
activated L-Thr; A2 activated L-Gln (100%), L-Asn (80%) and L-Asp (10%); A3 activated L-
279
Asp (100%), L-Asn (15%) and L-Gln (10%); A4 exclusively activated Gly; A5 activated L-Tyr
280
(100%) and L-Val (15%); and A6 activated L-Val (100%) and L-Asn (10%). (The highest
16
281
activity was set at 100%, Fig. 5B) A1, the only adenylation domain of LocA, exclusively
282
activated L-Thr and is likely to be involved in the position 1 amino acid biosynthesis. A2, A3
283
and A4, the three adenylation domains of LocB, are likely to iteratively act to incorporate the
284
2nd-7th amino acids. Particularly, the A2 was able to efficiently activate both L-Gln and L-Asn.
285
A5 and A6, the last two adenylation domains of LocC, optimally activated L-Tyr and L-Val,
286
respectively, which is in agreement with the last two amino acids of locillomycins.
287
Proposed locillomycins biosynthesis pathway. The gene cluster for locillomycin
288
biosynthesis that was identified in this study sets the stage for further delineating the intricate
289
chemical assembly of this family of lipopeptide antibiotics. Based on the chemical structures of
290
locillomycins, the analysis of in vivo gene disruptions, internal adenylation domains experiments,
291
amino acid sequence homology of loc proteins to enzymes and previous knowledge of
292
lipopeptide biosynthetic pathways led to the proposal of a locillomycins biosynthetic pathway
293
(Figure 6). In brief, it’s a pathway where a cyclic lipo-nonapeptide is assembled by
294
multifunctional enzymes of a hexamodular NRPS. In the initial step, the ACS domain couples
295
coenzymeA (CoA) to a long-chain fatty acid, presumably tridecanoic to pentadecanoic, in an
296
ATP-dependent reaction. The activated fatty acid is then transferred to the 4-phosphopantetheine
297
cofactor of the T domain. In the second step, the fatty acid is coupled to the activated threonine
298
directly, catalyzed by one of the condensation domains preceding the A1 domain of the peptide
299
synthetases, with the KAS, two T domains and one C domain skipped (Figure 6). In subsequent
17
300
condensation reactions, the donor part containing the fatty acid and thiolated threonine is
301
connected to the acceptor part containing the A2 domain activated glutamine which is
302
epimerized by the E domain following the A2 before the next extension step (Figure 6). The
303
process continues in a canonical fashion until the A4 domain activated glycine is connected.
304
When this occurs, the whole LocB is used a second time for further extension (Figure 6). This
305
time the A2 activates asparagine rather than glutamine, which is connected to the donor without
306
epimerization. Extension then proceeds through LocB and LocC all the way down to the TE
307
domain, which catalyzes the cyclization of the synthesized linear peptide by connecting the
308
carbonyl group of Val9 and the hydroxyl group of Thr1 sidechain, before releasing the mature
309
cyclic lipopeptide of locillomycins (Figure 6).(47) Particular emphasis should be noted that the
310
first A domain (A1) and the last two A domains (A5, A6) were used only once, whereas the three
311
A domains (A2, A3, A4) in the middle were iteratively used during the biosynthesis (Figure 6).
312
Moreover, the A2 domain in LocB somehow exhibits different specificity in the iteration. Also,
313
the epimerization domain following the A2 domain which is responsible for conversion of L-Gln
314
to D-Gln in the first time is skipped in the second time (Figure 6). The different selectivity of
315
these catalytic domains within the same NRPS assembly line is quite unusual compared to other
316
NRPS assembly lines. Thus, compared to the iterative uses in type B and C NRPS, loc NRPS
317
was designated as a type D NRPS(Figure 6).(25–28)
18
318
Typical lipopeptide biosynthesis involves the KAS domain catalyzing the formation of a β-
319
ketoacyl thioester, which is subsequently converted into a β-substituted fatty acid by a
320
transamination or transhydoxylation reaction. This is followed by a transfer of the fatty acid to a
321
T domain and is coupled to the first activated amino acid. In contrast, with regards to the
322
synthesis of locillomycin, only the ACS domain, not the KAS domain, is responsible for the
323
activation of the fatty acid. The reason for skipping the KAS domain may be due to the lack of
324
any functional domains such as amino transferase or hydroxyl transferase in the downstream of
325
the KAS domain. The function of the redundant T and C domains in the front of the LocA are
326
unclear since one condensation domain preceding the first A domain of the peptide synthetase is
327
enough to couple the fatty acid to the first activated amino acid.(2) While a few of these atypical
328
biochemical features of single module or multiple modules iteration have been previously
329
observed, the locillomycins represent a rare nonlinear NRPS biosynthetic model with a
330
combination of iteration, skipping, alternating and specificity changes (Figure 6).
331
Biological functions of locillomycin. Biological activity assays revealed that locillomycins
332
have moderate antibacterial activities (Figure S11). The antibacterial minimum inhibitory
333
concentrations (MICs) of locillomycin A, B and C were 24.3, 18.8, and 17.6 µg/mL, respectively,
334
for methicillin-resistant Staphylococcus aureus (MRSA), and were 6.3, 5.8, and 5.4 µg/mL,
335
respectively, for Xanthomonas oryzae pv. oryzae. The antiviral results at different locillomycins
336
concentrations revealed that the Porcine Epidemic Diarrhea Virus (PEDV) infection could be
19
337
effectively inhibited using the compound. Particularly, at the concentration of 10 µg/mL,
338
locillomycins reduced the virus copies by 300 fold and was about ten times as efficient as
339
surfactins (Figure S12). Locillomycins also have lower hemolytic activities than surfactins and
340
are presumed to reduce toxicity against eukaryotic cells, which could improve their possible
341
therapeutic application (Figure S12).(3)
342
20
343
Acknowledgements
344
This work was supported by the National Natural Science Foundation of China (grants
345
no.30900929), National High-tech R&D Program of China (2011AA10A201), and Jiangsu
346
Independent Innovation Fund (CX(12)5001) . We are also grateful to L. Du (Department of
347
Chemistry, University of Nebraska-Lincoln) for his critical review and revision.
348
Author Contributions
349
The authors declare no competing financial interest. The manuscript was written through
350
contributions of all authors. All authors have given approval to the final version of the
351
manuscript. §Chuping Luo and Xuehui Liu contributed equally to this work.
352 353
21
354
References
355 356
1.
Stein T. 2005. Bacillus subtilis antibiotics: structures, syntheses and specific functions. Molecular microbiology 56:845–857.
357 358
2.
Ongena M, Jacques P. 2008. Bacillus lipopeptides: versatile weapons for plant disease biocontrol. Trends in microbiology 16:115–125.
359 360
3.
Rodrigues L, Banat IM, Teixeira J, Oliveira R. 2006. Biosurfactants: potential applications in medicine. The Journal of antimicrobial chemotherapy 57:609–618.
361 362 363 364
4.
Romero D, de Vicente A, Rakotoaly RH, Dufour SE, Veening JW, Arrebola E, Cazorla FM, Kuipers OP, Paquot M, Pérez-García A. 2007. The iturin and fengycin families of lipopeptides are key factors in antagonism of Bacillus subtilis toward Podosphaera fusca. Molecular plant-microbe interactions: MPMI 20:430–440.
365 366 367
5.
Ongena M, Jourdan E, Adam A, Paquot M, Brans A, Joris B, Arpigny JL, Thonart P. 2007. Surfactin and fengycin lipopeptides of Bacillus subtilis as elicitors of induced systemic resistance in plants. Environmental microbiology 9:1084–1090.
368 369 370
6.
Zeriouh H, de Vicente A, Pérez-García A, Romero D. 2014. Surfactin triggers biofilm formation of Bacillus subtilis in melon phylloplane and contributes to the biocontrol activity. Environmental microbiology 16:2196–2211.
371 372 373 374
7.
Han Q, Wu F, Wang X, Qi H, Shi L, Ren A, Liu Q, Zhao M, Tang C. 2014. The bacterial lipopeptide iturins induce Verticillium dahliae cell death by affecting fungal signalling pathways and mediate plant defence responses involved in pathogen-associated molecular pattern-triggered immunity. Environmental microbiology 17:1166-1188.
375 376 377 378 379 380
8.
Duitman EH, Hamoen LW, Rembold M, Venema G, Seitz H, Saenger W, Bernhard F, Reinhardt R, Schmidt M, Ullrich C, Stein T, Leenders F, Vater J. 1999. The mycosubtilin synthetase of Bacillus subtilis ATCC6633: A multifunctional hybrid between a peptide synthetase , an amino transferase , and a fatty acid synthase. Proceedings of the National Academy of Sciences of the United States of America 96:13294–13299.
381 382
9.
Shoda M. 2001. Cloning , sequencing , and characterization of the iturin A operon. Journal of bacteriology 183:6265–6273.
22
383 384 385
10.
Tosato V, Albertinij AM, Zotti M, Sondal S, Bruschi CV. 1997. Sequence completion , identification and definition of the fengycin operon in Bacillus subtilis 168. Microbiology 143:3443–3450.
386 387 388 389
11.
Kessler N, Schuhmann H, Morneweg S, Linne U, Marahiel MA. 2004. The linear pentadecapeptide gramicidin is assembled by four multimodular nonribosomal peptide synthetases that comprise 16 modules with 56 catalytic domains. The Journal of biological chemistry 279:7413–7419.
390 391
12.
Role C, The OF, Domain C. 1998. Peptide Bond Formation in Nonribosomal Peptide Biosynthesis. The Journal of biological chemistry 273:22773–22781.
392 393 394
13.
Belshaw PJ, Walsh CT, Stachelhaus T. 1999. Aminoacyl-CoAs as probes of condensation domain selectivity in nonribosomal peptide synthesis. Science (New York, N.Y.) 284:486–489.
395 396
14.
Sieber SA, Marahiel MA. 2003. Learning from Nature’s Drug Factories: Nonribosomal Synthesis of Macrocyclic Peptides. Journal of bacteriology 185:7036–7043.
397 398 399
15.
Mootz HD, Schwarzer D, Marahiel MA. 2000. Construction of hybrid peptide synthetases by module and domain fusions. Proceedings of the National Academy of Sciences of the United States of America 97:5848–5853.
400 401 402 403
16.
Hoe BC, Gorzelnik KV, Yang JY, Hendricks N, Dorrestein PC. 2012. Enzymatic resistance to the lipopeptide surfactin as identified through imaging mass spectrometry of bacterial competition. Proceedings of the National Academy of Sciences of the United States of America 109:13082–13087.
404 405
17.
Zaman M, Toth I. 2013. Immunostimulation by Synthetic Lipopeptide-Based Vaccine Candidates: Structure-Activity Relationships. Frontiers in immunology 4:1–12.
406 407
18.
Sieber SA, Marahiel MA. 2005. Molecular mechanisms underlying nonribosomal peptide synthesis: approaches to new antibiotics. Chemical reviews 105:715–738.
408 409
19.
Robbel L, Marahiel MA. 2010. Daptomycin, a bacterial lipopeptide synthesized by a nonribosomal machinery. The Journal of biological chemistry 285:27501–27508.
410 411 412
20.
Magarvey NA, Haltli B, He M, Greenstein M, Hucul JA. 2006. Biosynthetic pathway for mannopeptimycins, lipoglycopeptide antibiotics active against drug-resistant grampositive pathogens. Antimicrobial agents and chemotherapy 50:2167–2177.
23
413 414 415 416
21.
Mascio CTM, Mortin LI, Howland KT, Van Praagh ADG, Zhang S, Arya A, Chuong CL, Kang C, Li T, Silverman JA. 2012. In vitro and in vivo characterization of CB183,315, a novel lipopeptide antibiotic for treatment of Clostridium difficile. Antimicrobial agents and chemotherapy 56:5023–5030.
417 418 419 420
22.
Miao V, Coëffet-Legal M-F, Brian P, Brost R, Penn J, Whiting A, Martin S, Ford R, Parr I, Bouchard M, Silva CJ, Wrigley SK, Baltz RH. 2005. Daptomycin biosynthesis in Streptomyces roseosporus: cloning and analysis of the gene cluster and revision of peptide stereochemistry. Microbiology 151:1507–1523.
421 422 423 424
23.
Owen JG, Reddy BVB, Ternei MA, Charlop-Powers Z, Calle PY, Kim JH, Brady SF. 2013. Mapping gene clusters within arrayed metagenomic libraries to expand the structural diversity of biomedically relevant natural products. Proceedings of the National Academy of Sciences of the United States of America 110:11797–11802.
425 426 427 428
24.
Rosconi F, Davyt D, Martínez V, Martínez M, Abin-Carriquiry JA, Zane H, Butler A, de Souza EM, Fabiano E. 2013. Identification and structural characterization of serobactins, a suite of lipopeptide siderophores produced by the grass endophyte Herbaspirillum seropedicae. Environmental microbiology 15:916–927.
429 430 431
25.
Felnagle EA, Jackson EE, Chan YA, Podevels AM, Berti AD, Mcmahon MD, Thomas MG. 2008. Nonribosomal Peptide Synthetases Involved in the Production of Medically Relevant Natural Products. molecular pharmaceutics 5:191–211.
432 433 434 435
26.
Tang MC, Fu CY, Tang GL. 2012. Characterization of SfmD as a Heme peroxidase that catalyzes the regioselective hydroxylation of 3-methyltyrosine to 3-hydroxy-5methyltyrosine in saframycin A biosynthesis. The Journal of biological chemistry 287:5112–5121.
436 437 438 439
27.
Dimise EJ, Widboom PF, Bruner SD. 2008. Structure elucidation and biosynthesis of fuscachelins, peptide siderophores from the moderate thermophile Thermobifida fusca. Proceedings of the National Academy of Sciences of the United States of America 105:15311–15316.
440 441 442
28.
Peng C, Pu J, Song L, Jian X, Tang M, Tang G. 2012. Hijacking a hydroxyethyl unit from a central metabolic ketose into a nonribosomal peptide assembly line. Proceedings of the National Academy of Sciences of the United States of America 109:8540–8545.
24
443 444 445
29.
Zhang W, Ostash B, Walsh CT. 2010. Identification of the biosynthetic gene cluster for the pacidamycin group of peptidyl nucleoside antibiotics. Proceedings of the National Academy of Sciences of the United States of America 107:1–6.
446 447 448
30.
Ross AC, Xu Y, Lu L, Kersten RD, Shao Z, Al-Suwailem AM, Dorrestein PC, Qian PY, Moore BS. 2013. Biosynthetic multitasking facilitates thalassospiramide structural diversity in marine bacteria. Journal of the American Chemical Society 135:1155–1162.
449 450
31.
Lautru S, Deeth RJ, Bailey LM, Challis GL. 2005. Discovery of a new peptide natural product by Streptomyces coelicolor genome mining. Nature chemical biology 1:265–269.
451 452
32.
Zhang Q, Pang B, Ding W, Liu W. 2013. Aromatic Polyketides Produced by Bacterial Iterative Type I Polyketide Synthases. ACS Catalysis 3:1439–1447.
453 454 455
33.
Moss SJ, Martin CJ, Wilkinson B. 2004. Loss of co-linearity by modular polyketide synthases: a mechanism for the evolution of chemical deversity. Natural product reports 21:575–593.
456 457 458
34.
Busch B, Ueberschaar N, Sugimoto Y, Hertweck C. 2012. Interchenar retrotransfer of aureothin intermediates in an iterative polyketide synthase module. Journal of the American Chemical Society 134:12382–12385.
459 460 461
35.
Sugimoto Y, Ding L, Ishida K, Hertweck C. 2014. Rational design of modular polyketide synthases: morphing the aureothin pathway into a luteoreticulin assembly line. angewandte chemie-international edition 53:1560–1564.
462 463 464
36.
Gruenewald S, Mootz HD, Stehmeier P, Stachelhaus T. 2004. In vivo Production of Artificial Nonribosomal Peptide Products in the Heterologous Host Escherichia coli. Applied and environmental microbiology 70:3282–3291.
465 466 467
37.
Tsuge K, Inoue S, Ano T, Itaya M, Shoda M. 2005. Horizontal Transfer of Iturin A Operon , itu , to Bacillus subtilis 168 and Conversion into an Iturin A Producer. Antimicrobial agents and chemotherapy 49:4641–4648.
468 469 470
38.
Park SY, Choi SK, Kim J, Oh TK, Park SH. 2012. Efficient production of polymyxin in the surrogate host Bacillus subtilis by introducing a foreign ectB gene and disrupting the abrB gene. Applied and environmental microbiology 78:4194–4199.
471 472
39.
Luo C, Zhou H, Zou J, Wang X, Zhang R, Xiang Y. 2015. Bacillomycin L and surfactin contribute synergistically to the phenotypic features of Bacillus subtilis 916 and
25
473 474
the biocontrol of rice sheath blight induced by Rhizoctonia solani. Applied Microbiology and Biotechnology 99:1897–1910.
475 476 477
40.
Luo C, Liu X, Zhou H, Wang X, Chen Z. 2015. Nonribosomal peptide synthase gene cluster for lipopeptide biosynthesis in Bacillus subtilis 916 and their phenotypic functions. Applied and Environmental Microbiology 81:422–431.
478 479
41.
Wang X, Luo C, Chen Z. 2012. Genome sequence of the plant growth-promoting rhizobacterium Bacillus sp. strain 916. Journal of bacteriology 194:5467–5468.
480 481 482
42.
Borisova SA, Circello BT, Zhang JK, van der Donk WA, Metcalf WW. 2010. Biosynthesis of rhizocticins, antifungal phosphonate oligopeptides produced by Bacillus subtilis ATCC6633. Chemistry & biology 17:28–37.
483 484
43.
Lautru S, Challis GL. 2004. Substrate recognition by nonribosomal peptide synthetase multi-enzymes. Microbiology 150:1629–1636.
485 486 487
44.
Challis GL, Ravel J, Townsend CA. 2000. Predictive, structure-based model of amino acid recognition by nonribosomal peptide synthetase adenylation domains. Chemistry & biology 7:211–24.
488 489
45.
Kunst F, Rapoport G. 1995. Salt stress is an environmental signal affecting degradative enzyme synthesis in Bacillus subtilis. Journal of bacteriology 177:2403–2407.
490
46.
Wuthrich K. 1986. NMR of Proteins and Nucleic Acids.
491 492
47.
Du L, Lou L. 2010. PKS and NRPS release mechanisms. Natural product reports 27:255– 78.
493 494 495 496 497
48.
Markley J, Bax A, Arata Y, Hibers C, Kaptein R, Sykes B, Wright PE, Wuthrich K. 1998. Recommendations for the presentation of NMR structures of proteins and nucleic acids--IUPAC-IUBMB-IUPAB Inter-Union Task Group on the standardization of data bases of protein and nucleic acid structures determined by NMR spectroscopy. Eur J Biochem 256:1–15.
498 499
26
500
Figure legends:
501
Figure 1. Organization of the locillomycins gene cluster and surrounding genes on the B.
502
subtilis 916 chromosome. The same locus of B. subtilis 168 and B. amyloquefaciens FZB42 are
503
also shown for comparison. Genes with high degree of homology between the three strains
504
(>95% identity) are shown in yellow. The biosynthetic clusters for locillomycins and sporulation
505
killing factor are shown in blue and purple, respectively. Other unrelated genes are also shown in
506
green, red and orange, respectively.
507
508
Figure 2. HPLC-MS chromatograms for locillomycins produced by wild-type B. subtilis 916
509
and its mutant ΔlocD. (A) locillomycins A-C were detected in B. subtilis 916 broth cultures, but
510
were below detectable level in ΔlocD. (B) B(a-c) and B(d-f) are the MS spectra of the extracts
511
from mutant ΔlocD and wild-type B. subtilis 916 respectively. Molecular weights of
512
locillomycin-A, -B and -C were detected in B. subtilis 916 broth cultures, but not observed in
513
mutant ΔlocD culture broth.
514
515
Figure 3. Important NOE and HMBC interactions for determination of the attaching site of the
516
acyl group and the cyclic structure of the peptide moiety were illustrated in the structure (upper
517
right) and in the spectra (upper left and lower panel). The ROESY (upper left) and HMBC
27
518
(lower) spectra are also labeled with assigned signals along the sides. The assigned nucleic
519
names of the key residues are underlined and displayed in red, and the red circles showed
520
relevant NOE and HMBC interactions.
521
522
Figure 4. Structural comparison of locillomycins with surfactins, iturins and fengycins. Some of
523
the key bonds are colored red and blue for clarity. The residue numbers are labeled next to the α-
524
carbon. Statistical sequences and configuration for each residue are listed below the formulas.
525
526
Figure 5. Expression, purification and relative substrate specificities of internal adenylation
527
domains (A domains). (A) Expression and purification of the A domains were analyzed on 10%
528
SDS-polyacrylamide slab gels. (B) A domains were investigated in terms of activity in the ATP-
529
PPi exchange reaction with the amino acids of locillomycin and a control without amino acid.
530
The highest activities were set at 100%. The background was below 5%. The specificities of the
531
different domains coincide with the primary structures of the locillomycins.
532
533
Figure 6. Proposed biosynthetic pathway for locillomycins. Nonribosomal peptide synthetase
534
genes (locA, locB and locC) are represented by arrows, and domains encoded by the respective
535
genes are shown underneath; the domains encoded by locB are proposed to act iteratively. The 28
536
FA (fatty acid chain) couples to the peptide by formation of an amide bond with the amino group
537
of the threonine residue, and a nonapeptide is cyclized by formation of a macrolactone between
538
the threonine hydroxyl and the valine carboxylate. The KAS domain is skipped during the
539
synthesis and is labeled with dashed line box while domains optionally used are labeled by a
540
purple box. LocB appeared twice because it is iteratively used and the A2 domain exhibits
541
different specificity in these two rounds. The E domain following the A2 domain also loses its
542
function in the second round.
543
544
29
Table 1: Primers used in this study Primer
Size
Sequence (5’-3’ )
Note
LocDF
30
TTTGCATGCATGAACTATGATTTATCACAT
Used for ΔlocD mutant
LocDR
29
TTTGAGCTCAGTAAAAATGAGAGGCAATT
LocCF
30
TTTAAGCTTAATTCAGCTCTTTATGAAGAA
LocCR
30
TTTGAGCTCATGAAGTTTACTAATGAATAC
Orf1UF
30
TTTAAGCTTGAATAAATAATTCACGGTAAA
Orf1UR
30
TTTTCTAGAATTCAGCTGCTTTATCGTAAG
Orf1DF
30
TTTTCTAGACTGTTGCTTTGTCGCATAATG
Orf1DR
30
TTTGCATGCAAGAGTGAGTTATCCAGTTGA
Orf3F
30
TTTAAGCTTCATTGAACTGAATAAAATGTA
Orf3R
30
TTTGAGCTCTTAGTCTCAATCTCAATGTTT
Orf7F
30
TTTGCATGCATACGATTAATAAAAGATATG
Orf7R
30
TTTGAATTCTTAAATTGTTATATCATCTTT
LocAA1F
32
TTTCCATGGTATCTGAAATAGAAATGATTACG
LocAA1R
30
TTTCTCGAGTTCTTTCTGAATAGCTGTTTG
LocBA2F
33
TTTCCATGGTAAAAGATGTAGAAATTATTACAG
LocBA2R
33
TTTCTCGAGCATTTGTTCTTTTTTATTATTTGG
LocBA3F
29
TTTCCATGGTTTCGGAGATTGATATCACG
LocBA3R
29
TTTCTCGAGGGTTTCCTGAACATCATTCT
LocBA4F
31
TTTGGATCCATCTCTTCTATAGATATCATGA
construction
Used for ΔlocC mutant construction
Used for Δorf1 mutant construction
Used for Δorf3 mutant construction
Used for Δorf1 mutant construction
Used for A1 domain expression in E.coli
Used for A2 domain expression in E.coli
Used for A3 domain expression in E.coli
Used for A4 domain
LocBA4R
32
TTTCTCGAGTTATGTTTCCTGCAAAACTGTTT
expression in E.coli
LocCA5F
28
TTTCCATGGGAGATGTAGGTTTGCTGAC
Used for A5 domain
LocCA5R
33
TTTCTCGAGAACTAATTCTTTTTCAATGTTATT
LocCA6F
32
AAACCATGGTATATCAAATTAACATGATGACT
LocCA6R
30
AAACTCGAGTTTTGTCTCTGTTTCGTTTCT
PEDF
24
CGCAAAGACTGAACCCACTAATTT
expression in E.coli
Used for A6 domain expression in E.coli
Used
for
quantitative
analysis of PEDV PEDR
24
TTGCCTCTGTTGTTACTTGGAGAT
