Journal of Microbiology and Biotechnology

The Korean Society for Microbiology and Biotechnology (한국미생물·생명공학회)
권호 표지
DOI QR코드

DOI QR Code

Biosynthesis of Polymyxins B, E, and P Using Genetically Engineered Polymyxin Synthetases in the Surrogate Host Bacillus subtilis

  • Kim, Se-Yu (Super-Bacteria Research Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB)) ;
  • Park, Soo-Young (Super-Bacteria Research Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB)) ;
  • Choi, Soo-Keun (Super-Bacteria Research Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB)) ;
  • Park, Seung-Hwan (Super-Bacteria Research Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB))
  • Received : 2015.05.12
  • Accepted : 2015.06.08
  • Published : 2015.07.28
Download PDF
⟨ Previous Next ⟩

Abstract

The development of diverse polymyxin derivatives is needed to solve the toxicity and resistance problems of polymyxins. However, no platform has generated polymyxin derivatives by genetically engineering a polymyxin synthetase, which is a nonribosomal peptide synthetase. In this study, we present a two-step approach for the construction of engineered polymyxin synthetases by substituting the adenylation (A) domains of polymyxin A synthetase, which is encoded by the pmxABCDE gene cluster of Paenibacillus polymyxa E681. First, the seventh L-threonine-specific A-domain region in pmxA was substituted with the L-leucine-specific A-domain region obtained from P. polymyxa ATCC21830 to make polymyxin E synthetase, and then the sixth D-leucine-specific A-domain region (A6-D-Leu-domain) was substituted with the D-phenylalanine-specific A-domain region (A6-D-Phe-domain) obtained from P. polymyxa F4 to make polymyxin B synthetase. This step was performed in Escherichia coli on a pmxA-containing fosmid, using the lambda Red recombination system and the sacB gene as a counter-selectable marker. Next, the modified pmxA gene was fused to pmxBCDE on the chromosome of Bacillus subtilis BSK4dA, and the resulting recombinant strains BSK4-PB and BSK4-PE were confirmed to produce polymyxins B and E, respectively. We also succeeded in constructing the B. subtilis BSK4-PP strain, which produces polymyxin P, by singly substituting the A6-D-Leu-domain with the A6-D-Phe-domain. This is the first report in which polymyxin derivatives were generated by genetically engineering polymyxin synthetases. The two recombinant B. subtilis strains will be useful for improving the commercial production of polymyxins B and E, and they will facilitate the generation of novel polymyxin derivatives.

Keywords

Introduction

Polymyxin is an old class of cyclic lipopeptide antibiotics that was discovered in 1947, which has excellent bactericidal activity against many gram-negative bacteria because of its ability to disrupt the cell membrane [5]. Among a number of different polymyxins that have been discovered, polymyxins B and E (also known as colistin) have been used clinically since the late 1950s, but were largely abandoned in the 1970s because of their toxicities, especially their nephrotoxicity [13, 16]. However, despite their toxicities, the emergence of extremely multidrugresistant gram-negative bacteria has forced clinicians to re-instate polymyxins as the last-line therapy for gramnegative infections [12, 33]. Therefore, the development of less toxic polymyxin derivatives would be highly welcome. In the past few decades, total or semisynthesis, as well as modifications, of polymyxins were performed chemically or enzymatically to generate novel derivatives, and the resulting products were effectively used for structurefunction studies that sought clues to reduce the toxicity of polymyxins [1, 9, 19, 25, 30, 31]. However, there are limitations to obtaining diverse derivatives using chemical or enzymatic approaches, and these limitations are related to the structural complexity of polymyxins and the low efficiencies of the approaches. Therefore, the development of a platform that provides tools for the generation of polymyxin derivatives via the genetic engineering of polymyxin synthetase is greatly needed. Additionally, the molecular tools for the genetic manipulation of Paenibacillus polymyxa strains that produce polymyxins are very poor; thus, a surrogate host that is equipped with genetic tools and other characteristics, which are conducive to the introduction and expression of foreign genes while improving product safety, is also required. It is well known that Bacillus subtilis is a generally recognized as safe (GRAS) organism that can fulfill safety requirements, such as ensuring the absence of endotoxins. Well-developed genetic tools, such as transformation via natural competence and the ability to secrete peptides into culture media, also make B. subtilis a promising surrogate host for the expression of foreign nonribosomal peptide synthetase (NRPS) genes [3, 4, 34]. It can also potentially help improve polymyxin production and generate various derivatives through genome engineering.

Many pharmacologically important peptide antibiotics with modular structures, including polymyxins, are produced by NRPSs. Each module of an NRPS can be divided into different domains, such as the adenylation (A), thiolation (T; also referred to as the peptidyl carrier protein), condensation (C), epimerization (E), and termination (TE) domains. The A-domain plays a role in the selection and activation of amino acid monomers [17, 26]. Over the last two decades, many researchers have tried to obtain derivatives of peptide antibiotics by genetically modifying the modules or domains of NRPSs, especially through A-domain swapping. Stachelhaus et al. [27] and Schneider et al. [23] presented a general method for the targeted replacement of A-domains in the srfA operon, and they successfully modified the amino acid sequence of surfactin. Miao et al. [15] developed an NRPS trans-complementation system consisting of the substitution of NRPS subunits to generate novel peptide antibiotics. They generated various daptomycin derivatives (hybrid lipopeptides) by introducing expression plasmids carrying dptD or its substituents from other lipopeptide biosynthetic genes into a daptomycin-nonproducing Streptomyces roseosporus dptD deletion strain. However, polymyxin derivatives have not yet been constructed by genetically modifying a polymyxin synthetase.

Polymyxins are cyclic heptapeptides with a tripeptide side chain that is acylated by a fatty acid at its amino terminus, and they are produced by NRPSs consisting of 10 modules [10, 14, 28]. We previously identified the polymyxin A synthetase gene cluster (pmxABCDE) by whole genome sequencing of P. polymyxa E681, and succeeded in polymyxin A production by heterologous expression of the pmx gene cluster in a surrogate B. subtilis host [2, 8]. Recently, we identified two other pmx gene clusters encoding the polymyxin B and polymyxin E synthetases from P. polymyxa F4 and P. polymyxa ATCC21830, respectively, by genome sequencing [21]. The two pmx gene clusters were shown to have the same organization as that of the E681 strain (Fig. 1A). The pmx gene clusters responsible for the biosynthesis of polymyxins B and E have also been reported by Shaheen et al. [24] and Tambadou et al. [29], respectively.

Fig. 1.The organization of the pmx gene clusters for the biosynthesis of polymyxins A, B, and E in P. polymyxa E681, F4, and ATCC21830, respectively (A), and the module and domain arrangements of the NRPSs for the biosynthesis of polymyxin A (B), polymyxin B (C), and polymyxin E (D).

In this study, we constructed recombinant B. subtilis strains that can produce commercially important polymyxin B and polymyxin E, respectively, by replacing the A-domain regions of the polymyxin A synthetase gene via a two-step approach consisting of targeted modification of the A-domain using a fosmid clone as a template in Escherichia coli, followed by integration and expression of the modified pmx genes in a surrogate B. subtilis host. In our previous study, the recombinant B. subtilis strain BSK4, which can produce polymyxin A without extracellular addition of L-2,4-diaminobutyric acid, was constructed by introducing the ectB gene of P. polymyxa [22], and the B. subtilis BSK4dA strain derived from strain BSK4 by deleting pmxA was used as a host in this study. This work demonstrates that the two-step approach used to engineer polymyxin A synthetase to construct polymyxin B and E synthetases works well and may facilitate the development of novel derivatives of polymyxins.

 

Materials and Methods

Bacterial Strains, Plasmids, Primers, and Culture Conditions

The bacterial strains, plasmids, and PCR primers used in this study are described in Tables 1 and 2. E. coli DH5α was used as a cloning host for recombinant plasmids and as a test strain for assaying the antibacterial activity of polymyxins. E. coli EcNRK, a derivative of the EcNR2 strain, was constructed as described below and used as a cloning host for fosmids. The EcNR2 constructed in George Church’s laboratory (Harvard Medical School, MA, USA) was kindly provided by Professor Duhee Bang (Yonsei University, Seoul, Korea) [32]. B. subtilis strains were grown in LB broth or LB agar medium at 37℃ for general purposes, and in Cal18 broth with shaking at 37℃ to analyze polymyxin production [7]. Spectinomycin (100 μg/ml), chloramphenicol (5 μg/ml for B. subtilis, 20 μg/ml for E. coli), erythromycin (1 μg/ml for B. subtilis, 100 μg/ml for E. coli), kanamycin (40 μg/ml), and ampicillin (100 μg/ml) were used when required. For positive selection of transformants using the counter-selectable marker sacB, sucrose (5% final concentration) was added to the LB medium.

Table 1.Bacterial strains and plasmids used in this study.

Table 2.The underlined sequences indicate the targeted regions for lambda Red recombination.

Construction of the E. coli EcNRK Strain

To construct the E. coli EcNRK strain, the chloramphenicol resistance gene (cat) integrated at the mutS gene of strain EcNR2 was replaced with a kanamycin resistance (km) gene using the lambda Red recombination system. The km gene was obtained from pKD4 by PCR with the P1-muts and P2-muts primer set bearing 45 bp homologous arms that bind to the flanking regions of the cat gene of EcNR2, and it was introduced into the mutS locus of the EcNR2 strain by double-crossover recombination.

Construction of the Fosmid p8H3-em

To construct p8H3-em from fosmid p8H3, which contains the pmxABC genes and a truncated pmxD gene of P. polymyxa strain E681, the erythromycin resistance (em) gene was amplified from pDG1664 with the pmxAup-emF and pmxAup-emR primers bearing 50 bp arms that bind to the upstream region of the pmxA of p8H3. The PCR f ragment containing the em gene (1.1 kb) was inserted upstream of pmxA by double-crossover recombination.

Construction of the B. subtilis BSK4dA Strain

To delete the pmxA region from the pmx gene cluster of the B. subtilis BSK4 strain, the cat gene was amplified from pDG1662 by PCR with primers cmF and cmR, and it was used to replace the pmxA region. The 5’- and 3’-flanking regions of pmxA (1.0 and 1.9 kb, respectively) were amplified from the chromosomal DNA of P. polymyxa E681 by PCR using primer sets pmxdFf2 and pmxdFr2, and pmxA3dF2 and pmxB5dR2, respectively. These two PCR fragments contain sequences that allow them to assemble with a PCR fragment containing the cat gene. In turn, these three PCR fragments were joined by fusion PCR and cloned into the pGEM-T Easy vector (Promega, Madison, WI, USA) to construct pGT-dA, which was subsequently introduced into the pmxA locus of strain BSK4 by homologous recombination to construct strain BSK4dA (Fig. 2). Transf ormation of B. subtilis was conducted as described by Harwood and Cutting [6]

Fig. 2.Schematic diagram showing the strategy for the construction of B. subtilis BSK4dA. The BSK4dA strain was constructed by replacing the pmxA region of the pmx gene cluster on the BSK4 chromosome with the cat gene by homologous recombination.

Construction of the sp-sacB Cassette

The spectinomycin resistance gene (sp) was amplified using primers spF and spR from pDG1730, and the sacB gene was amplified with primers sacF and sacR from B. subtilis genomic DNA. To assemble the two PCR fragments, primer sacF was designed to have a 21 bp overlap at the junction regions. The two PCR fragments were joined by fusion PCR with primers spF and sacR. The constructed sp-sacB cassette was amplified with primers A6spF and A6sacR such that it had 50 bp homologous arms that bind to each end of the gene encoding the sixth A-domain region, and the resulting PCR fragment was termed the A6sp-sacB cassette. The sp-sacB cassette was also amplified with primers A7spF and A7sacR such that it had 50 bp homologous arms that bind to each end of the gene encoding the seventh A-domain region. The resulting PCR fragment was termed the A7sp-sacB cassette.

Screening for Polymyxin-Producing B. subtilis Transformants

E. coli DH5α cells that were grown overnight in 3 ml of LB at 37℃ were mixed with 300 ml of LB agar medium, autoclaved, and cooled to <50℃ to prepare the bioassay plate (EC plate). To test the antimicrobial activity of the B. subtilis BSK4dA transformants, the colonies were transferred onto an EC plate using toothpicks, and the plate was incubated at 37℃ for 15 h. Transformants showing growth inhibition zones were selected for further analysis.

Antibacterial Activity Assay and Electrospray Ionization–Liquid Chromatography Mass Spectrometry (ESI-LC/MS) Analysis

Antibacterial activity against E. coli DH5α was assayed as follows. Recombinant B. subtilis strains were grown in Cal18 medium at 37℃ with vigorous shaking (220 rpm) for 24 h [22]. Following centrifugation at 15,000 ×g for 15 min at 4℃, the supernatant was collected, and a 10 μl aliquot of each supernatant was dropped onto 6 mm paper disks, dried, and placed on an EC plate to observe the growth inhibition zone. The polymyxin in the culture supernatant was analyzed using ESI-LC/MS (Thermo Fisher Scientific, Waltham, MA, USA), as previously described by Park et al. [22].

 

Results and Discussion

As described above, we previously identified the polymyxin A synthetase gene cluster (pmxABCDE) of P. polymyxa E681. The polymyxin A synthetase consists of 10 modules that are encoded by three genes, pmxA, pmxB, and pmxE. Based on the structure of polymyxin, the order of the modules for amino acid assembly during polymyxin synthesis might be PmxE-PmxA-PmxB [2]. Recently, we identified two more pmx gene clusters encoding the polymyxin B and polymyxin E synthetases by genome sequencing of P. polymyxa F4 and P. polymyxa ATCC21830, respectively [21]. As shown in Figs. 1B–1D, the domain structures of the three polymyxin synthetases have a very similar modular organization. Only one difference was found between the 10 A-domains of the polymyxin A and polymyxin E synthetases. Specifically, the A-domain of the seventh position of the former is an L-threonine-specific domain (A7-L-Thr-domain), whereas that of the latter is an Lleucine-specific domain (A7-L-Leu-domain). There were two differences in the sixth and seventh A-domains of polymyxin B synthetase; that is, the D-phenylalaninespecific domain (A6-D-Phe-domain) and A7-L-Leu-domain, compared with the D-leucine-specific domain (A6-D-Leu-domain) and A7-L-Thr-domain of polymyxin A synthetase. These results suggest that polymyxin A synthetase can be genetically engineered to produce polymyxin B or polymyxin E by replacing the A6- and A7- or A7-domain regions of the pmxA gene, respectively.

Here, we describe experiments in which two recombinant B. subtilis strains that produced polymyxin B or polymyxin E were constructed. There were two reasons for choosing this A-domain engineering approach instead of cloning the gene clusters entirely. First, we wished to develop a system for creating diverse polymyxin derivatives by A-domain engineering. Second, it is very difficult to introduce the entire polymyxin gene cluster, which spans about 41 kb, into the chromosome of B. subtilis. The engineering of polymyxin A synthetase to yield polymyxin B or polymyxin E synthetases was performed in two steps. In the first step, the substitution of the target adenylation domain in polymyxin A synthetase with the adenylation domain of the polymyxin B or polymyxin E synthetases was performed using a fosmid clone that contained the pmxA gene as template DNA in E. coli. In the second step, the modified pmx gene was introduced into a polymyxin-nonproducing B. subtilis ΔpmxA host strain.

Preparation of an E. coli Host Strain and Template DNA for Polymyxin A Synthetase Gene Recombineering

Although a variety of molecular tools for the genetic manipulation of B. subtilis are available, the low efficiency of recombination between short homologous nucleotide sequences causes some difficulties. Therefore, in the first step, recombination-mediated genetic engineering (recombineering) of the A-domain gene was performed by homologous recombination using the lambda Red recombination system in E. coli. The EcNRK strain was constructed by modifying the EcNR2 strain as a cloning host for fosmids carrying pmx genes with a cm marker, as described in Materials and Methods. Fosmid p8H3, which contains the pmxABC genes, as well a truncated pmxD gene, of the P. polymyxa E681 strain, was selected from our previous sequencing library, and it was modified into p8H3-em by introducing the em gene into the upstream region of pmxA so that it could serve as a selection marker for transformants in the next step, after introducing the modified pmxA gene into the B. subtilis host. p8H3-em was used as template DNA for the recombineering of the A-domain gene in E. coli EcNRK.

Preparation of a B. subtilis Host Strain for Polymyxin Synthetase Gene Recombineering and Expression

To efficiently construct and select the engineered polymyxin synthetase gene cluster via recombineering into the B. subtilis chromosome, the B. subtilis BSK4dA strain was constructed by modifying the B. subtilis BSK4 strain, which carries the entire pmx gene cluster (pmxABCDE). The 13.5kb 5’ region of the pmxA gene (14.9 kb) of the BSK4 strain was removed, and the cat gene was integrated into the region as described in Materials and Methods (Fig. 2). The BSK4dA strain carrying pmxBCDE was used as a host for the introduction of the modified pmxA gene, as well as for the expression of the modified pmx gene cluster to produce polymyxins. When the modified pmxA gene is fused to the other pmx genes carried on the chromosome of strain BSK4dA, the entire gene cluster will be restored, and the strain will exhibit antibacterial activity against E. coli.

Preparation of Donor DNAs Encoding the A6-D-Phe-Domain and A7-L-Leu-Domain of Polymyxin Synthetase

DNA fragments encoding the A6-D-Phe-domain (1.3 kb) and the A7-L-Leu-domain (1.9 kb) were obtained by PCR using the genomic DNAs of P. polymyxa F4 and P. polymyxa ATCC21830 as templates with the primer sets PA6F2-PE253 and sacredF-sacredR, respectively. The 5’- and 3’-end regions of the two PCR fragments were designed to be highly homologous to those of the sixth or seventh Adomain regions of the polymyxin A synthetase gene; that is, there were more than 95% identities in the nucleotide sequences of the 200–300 bp end regions and 100% identities in narrow regions of about 50 bp. Therefore, lambda Redmediated homologous recombination between these regions was possible. Thus, the two DNA fragments were used to engineer the target A-domains of the polymyxin A synthetase gene.

Recombineering of the A7-Domain Region of the Polymyxin A Synthetase Gene to Construct Polymyxin E Synthetase

To construct polymyxin E synthetase by engineering polymyxin A synthetase, the A7-L-Thr-domain region of P. polymyxa E681 was substituted with the A7-L-Leu-domain region of P. polymyxa ATCC21830, using the pmxA gene in fosmid p8H3-em as template DNA and E. coli EcNRK as a host for the recombineering reaction. This process was conducted in two steps, as shown in Fig. 3A. First, the A7-L-Thrdomain region (1.9 kb) of the pmxA gene was deleted by inserting the A7sp-sacB cassette (3.1 kb), which contains a spectinomycin resistance gene, the sacB gene, a nd 50bp homologous arms to the A7-L-Thr-domain region, using the lambda Red recombination system. A transformant carrying the recombinant fosmid containing the sp-sacB cassette was selected by its resistance to spectinomycin and sensitivity to 5% sucrose, and the recombination event was confirmed by PCR using primers A7F and A7R, which bind to each end of the A7sp-sacB cassette. The resulting recombinant fosmid was designated as p8H3-SS7. Second, the DNA fragment containing the A7-L-Leu-domain region, which was obtained by PCR amplification of the chromosomal DNA of the P. polymyxa ATCC21830 strain, was integrated into p8H3-SS7 by replacing the sp-sacB cassette via doublecrossover recombination (Fig. 3A). The E. coli transformant carrying the recombinant fosmid, named p8H3-A7Leu, was selected based on its sucrose-resistant and spectinomycinsusceptible phenotypes. The replacement of the A7-domain region in the fosmid p8H3-A7Leu was confirmed by PCR and nucleotide sequencing. The fosmid p8H3-A7Leu was introduced into B. subtilis strain BSK4dA by homologous recombination to complete the polymyxin E synthetase gene cluster, and it was also used as a template to construct polymyxin B synthetase, as described below.

Fig. 3.Schematic diagram showing the strategy for the construction of polymyxin B and E synthetases by modifying the specificity of the A6- and A7- or A7-domains of polymyxin A synthetase, respectively. (A) Recombineering of the A7-domain coding region of pmxA was performed in two steps; the A7-L-Thr-domain region was replaced by the sp-sacB cassette, and then the sp-sacB cassette was replaced with a DNA fragment encoding the A7-L-Leu-domain, using the lambda Red recombination system. (B) Recombineering of the A6-D-Leu-domain coding region of pmxA was performed similarly to that described in (A) by introducing a DNA fragment encoding the A6-D-Phe-domain.

Recombineering of the A6-Domain Region to Construct Polymyxin B Synthetase

The construction of the polymyxin B synthetase gene was performed by substituting the A6-D-Leu-domain region of the pmxA gene of P. polymyxa E681 in fosmid p8H3-A7Leu with the A6-D-Phe-domain region of P. polymyxa F4. This process was also conducted in two steps as described above (Fig. 3B). First, the A6-D-Leu-domain region (about 1.3 kb) of the pmxA gene of the fosmid p8H3-A7Leu was deleted by inserting the A6sp-sacB cassette (about 3.1 kb) and confirmed by PCR using primers A6F and A6R. The resulting recombinant fosmid was designated as p8H3-A7Leu-SS6. Second, the DNA fragment containing the A6-D-Phe-domain region, which was obtained from the P. polymyxa F4 strain by PCR, was integrated into p8H3-A7Leu-SS6 by replacing the sp-sacB cassette to construct p8H3-A6Phe-A7Leu (Fig. 3B). The replacement of the A6-domain region in the fosmid was confirmed by PCR and nucleotide sequencing. The resulting fosmid, p8H3-A6Phe-A7Leu, was introduced into the B. subtilis BSK4dA strain to complete the polymyxin B synthetase gene cluster, as described below.

Completion of Polymyxin Synthetase Gene Clusters and Biosynthesis of Polymyxin B and E in B. subtilis

Fosmids p8H3-A7Leu and p8H3-A6Phe-A7Leu containing pmxA genes with altered amino acid sequences in the seventh or sixth and seventh A-domains of the polymyxin A synthetase gene were introduced into B. subtilis BSK4dA harboring intact pmxBCDE genes, but lacking the pmxA gene (Fig. 4). As we previously reported, polymyxin synthetase is encoded by three pmx genes: pmxA, pmxB, and pmxE [2]. The deleted pmxA region of the BSK4dA strain will be restored by homologous recombination with the modified pmxA regions of the fosmids p8H3-A7Leu or p8H3-A6Phe-A7Leu. After separately transforming the BSK4dA strain with the two fosmids, more than 100 transformants (Emr Cms ) were selected and assayed for their antibacterial activity against E. coli, using an EC plate, and we found that about 50% of the Emr Cms transformants showed antibacterial activity in both cases. These results showed that the upstream region of the pmx genes on the BSK4dA chromosome, including the cat gene, were successfully replaced with the modified pmxA gene and em gene, thereby restoring the entire polymyxin synthetase gene cluster in both cases. Some of the B. subtilis transformants that were Emr Cms, but which did not exhibit antibacterial activity, were analyzed by PCR and nucleotide sequencing. It was found that the recombination events occurred in non-target regions, and that some regions of pmx genes are missing in these transformants. This is probably because there is a somewhat high degree of homologies between domains consisting of multimodules of polymyxin synthetase. The recombinant B. subtilis strains obtained by transformation with p8H3-A7Leu and p8H3-A6Phe-A7Leu were named BSK4-PE and BSK4-PB, respectively. We confirmed the correct integration of the modified A-domains into the target sites by PCR and nucleotide sequencing, and the expression of the engineered polymyxin synthetase gene clusters of the two recombinant B. subtilis transformants was confirmed by LC/MS analysis.

Fig. 4.Schematic diagram showing the strategy for the construction of the B. subtilis strains BSK4-PB, BSK4-PE, and BSK4-PP. The three recombinant Bacillus strains were constructed by introducing the modified pmxA genes of the fosmid clones p8H3-A7Leu, p8H3-A6Phe-A7Leu, and p8H3-A6Phe, respectively, into the chromosome of BSK4dA.

The two recombinant strains, BSK4-PB and BSK4-PE, as well as BSK4, were grown in Cal18 medium, and culture supernatants were collected and assayed for antibacterial activity against E. coli DH5α by the disk diffusion method using an EC plate (Fig. 5A). As was expected, all three strains, BSK4, BSK4-PE, and BSK4-PB, showed antibacterial activity against E. coli. The culture supernatant samples were analyzed using an ESI-LC/MS system to detect polymyxins, and the BSK4-PB and BSK4-PE strains were found to produce polymyxin B and polymyxin E, respectively, as was expected (Fig. 5B). The (M+H)+ ion peaks of BSK4-PB and BSK4-PE were detected at m/z 1,203.59 and m/z 1,169.60, respectively, which correspond to the known mass values of polymyxin B1 (m/z 1,203.6) and polymyxin E1 (m/z 1,169.77) [20, 29]. The mass values of the polymyxins produced by BSK4-PB and BSK4-PE were 46.06 and 12.07 greater than that of BSK4 (m/z 1,157.53), respectively. These differences coincide well with the differences in the molecular weights of the amino acids at the sixth and seventh or seventh positions of the modified polymyxins; the mass difference between threonine (119.12) and leucine (131.17) is 12.05, and the mass difference between threonine (119.12) and phenylalanine (165.19) is 46.07. We also succeeded in constructing the B. subtilis BSK4-PP strain, which can produce polymyxin P by engineering of polymyxin A synthetase. As shown in Fig. 5C there is only one difference between the primary structure of polymyxin A and that of polymyxin P; that is, the amino acid of the sixth position of the former is D-leucine, whereas that of the latter is D-phenylalanine. The construction was done by substituting the A6-D-Leu-domain region of the polymyxin A synthetase gene from P. polymyxa E681 with the A6-D-Phe-domain region from P. polymyxa F4 using the same strategy as described above (Fig. 4). The culture supernatant of the BSK4-PP strain also showed antibacterial activity against E. coli, and it was confirmed to produce polymyxin P by ESI-LC/MS (Figs. 5A and 5B). The m/z (1,191.69) of the polymyxin produced by the BSK4- PP strain coincided well with the known molecular mass of polymyxin P (m/z 1,191.9) produced by Paenibacillus polymyxa M-1 [18].

Fig. 5.Analysis of polymyxins synthesized by the recombinant B. subtilis strains BSK4-PB, BSK4-PE, and BSK4-PP. (A) Antibacterial activities of the cell-free culture supernatants of BSK4 and the three recombinant strains against E. coli DH5α. (B) LC/MS data for polymyxins A, B, E, and P produced by B. subtilis strains BSK4, BSK4-PB, BSK4-PE, and BSK4-PP, respectively. (C) The primary structures of the polymyxins A, B, E and P.

To determine the polymyxin productivity of the three recombinant B. subtilis strains producing polymyxins B, E and P, respectively, the antibacterial activities in their culture supernatants were compared with serially diluted standard solutions of polymyxin B. The range of productivities of the strains were estimated to be 200 mg/l or lower levels (data not shown). These levels were obtained without optimizing the expression of the pmx gene clusters or the fermentation conditions; therefore, yield improvements should be possible by further genetic engineering of the early-stage transformants or fermentation development. To improve the environmental safety of the recombinant B. subtilis strains that produce polymyxins, the removal of antibiotic resistance marker genes from the strains is under process.

In our two-step approach, the target A-domain region of P. polymyxa E681 was substituted with a donor A-domain region from P. polymyxa ATCC21830 and/or P. polymyxa F4 on a fosmid containing the pmxA gene, and then the entire pmx gene cluster was restored via recombination of the modified pmxA gene into the chromosome of the B. subtilis BSK4dA strain containing pmxBCDE. The above-mentioned results show that this approach works well, and that we have succeeded in constructing the three recombinant B. subtilis strains producing polymyxins B, E and P. The former two strains produce the commercially important polymyxins B and E that will be useful for improving polymyxin production because of the advantages of using B. subtilis as a host. B. subtilis 168, the parent strain of the BSK4 strain used in this study, is a non-pathogenic strain with GRAS status, and therefore, downstream processing can be simplified. The biosynthesis of surfactin in strain BSK4 was completely blocked by knockout of the srfC gene during the construction of the strain in our previous study [22]. If necessary, we can also block the biosynthesis of other compounds that possibly hinder purification of polymyxins, by disruption of the relevant genes without difficulty. This will be an advantage for producing polymyxin or its derivatives in pure form. The three recombinant B. subtilis strains will also facilitate the generation of novel polymyxin derivatives as a platform for further engineering polymyxin synthetases. The two-step approach might be used to generate modifications in any A-domains of polymyxin synthetases by introducing various A-domain genes, with different amino acid specificities, obtained from other polymyxin synthetase genes or possibly other NRPS genes.

References

  1. Chihara S, Tobita T, Yahata M, Ito A, Koyama Y. 1973. Enzymatic degradation of colistin isolation and identification of α-N-acyl α, γ-diaminobutyric acid and colistin nonapeptide. Agric. Biol. Chem. 37: 2455-2463. https://doi.org/10.1271/bbb1961.37.2455
  2. Choi S-K, Park S-Y, Kim R, Kim S-B, Lee C-H, Kim JF, Park S-H. 2009. Identification of a polymyxin synthetase gene cluster of Paenibacillus polymyxa and heterologous expression of the gene in Bacillus subtilis. J. Bacteriol. 191: 3350-3358. https://doi.org/10.1128/JB.01728-08
  3. Doekel S, Eppelmann K, Marahiel MA. 2002. Heterologous expression of nonribosomal peptide synthetases in B. subtilis: construction of a bi-functional B. subtilis/E. coli shuttle vector system. FEMS Microbiol. Lett. 216: 185-191. https://doi.org/10.1111/j.1574-6968.2002.tb11434.x
  4. Eppelmann K, Doekel S, Marahiel MA. 2001. Engineered biosynthesis of the peptide antibiotic bacitracin in the surrogate host Bacillus subtilis. J. Biol. Chem. 276: 34824-34831. https://doi.org/10.1074/jbc.M104456200
  5. Evans ME, Feola DJ, Rapp RP. 1999. Polymyxin B sulfate and colistin: old antibiotics for emerging multiresistant gram-negative bacteria. Ann. Pharmacother. 33: 960-967. https://doi.org/10.1345/aph.18426
  6. Harwood CR, Cutting SM. 1990. Molecular Biological Methods for Bacillus, p. 33-36. John Wiley & Sons, Chichester, UK.
  7. Jensen K, Østergaard PR, Wilting R, Lassen SF. 2010. Identification and characterization of a bacterial glutamic peptidase. BMC Biochem. 11: 47. https://doi.org/10.1186/1471-2091-11-47
  8. Kim JF, Jeong H, Park S-Y, Kim S-B, Park YK, Choi S-K, et al. 2010. Genome sequence of the polymyxin-producing plant-probiotic rhizobacterium Paenibacillus polymyxa E681. J. Bacteriol. 192: 6103-6104. https://doi.org/10.1128/JB.00983-10
  9. Kline T, Holub D, Therrien J, Leung T, Ryckman D. 2001. Synthesis and characterization of the colistin peptide polymyxin E1 and related antimicrobial peptides. J. Pept. Res. 57: 175-187.
  10. Komura S, Kurahashi K. 1980. Biosynthesis of polymyxin E. III. Total synthesis of polymyxin E by a cell-free enzyme system. Biochem. Biophys. Res. Commun. 95: 1145-1151. https://doi.org/10.1016/0006-291X(80)91592-2
  11. Koyama Y, Kurosasa A, Tsuchiya A, Takakuta K. 1950. A new antibiotic ‘colistin’ produced by spore-forming soil bacteria. J. Antibiot. 3: 457-458.
  12. Li J, Nation RL, Milne RW, Turnidge JD, Coulthard K. 2005. Evaluation of colistin as an agent against multi-resistant gram-negative bacteria. Int. J. Antimicrob. Agents 25: 11-25. https://doi.org/10.1016/j.ijantimicag.2004.10.001
  13. Li J, Nation RL, Turnidge JD, Milne RW, Coulthard K, Rayner CR, Paterson DL. 2006. Colistin: the re-emerging antibiotic for multidrug-resistant gram-negative bacterial infections. Lancet Infect. Dis. 6: 589-601. https://doi.org/10.1016/S1473-3099(06)70580-1
  14. Martin NI, Hu H, Moake MM, Churey JJ, Whittal R, Worobo RW, Vederas JC. 2003. Isolation, structural characterization, and properties of mattacin (polymyxin M), a cyclic peptide antibiotic produced by Paenibacillus kobensis M. J. Biol. Chem. 278: 13124-13132. https://doi.org/10.1074/jbc.M212364200
  15. Miao V, Coëffet-Le Gal M-F, Nguyen K, Brian P, Penn J, Whiting A, et al. 2006. Genetic engineering in Streptomyces roseosporus to produce hybrid lipopeptide antibiotics. Chem. Biol. 13: 269-276. https://doi.org/10.1016/j.chembiol.2005.12.012
  16. Michalopoulos A, Falagas ME. 2008. Colistin and polymyxin B in critical care. Crit. Care Clin. 24: 377-391. https://doi.org/10.1016/j.ccc.2007.12.003
  17. Mootz HD, Schwarzer D, Marahiel MA. 2002. Ways of assembling complex natural products on modular nonribosomal peptide synthetases. Chembiochem 3: 490-504. https://doi.org/10.1002/1439-7633(20020603)3:6<490::AID-CBIC490>3.0.CO;2-N
  18. Niu B, Vater J, Rueckert C, Blom J, Lehmann M, Ru J-J, et al. 2013. Polymyxin P is the active principle in suppressing phytopathogenic Erwinia spp. by the biocontrol rhizobacterium Paenibacillus polymyxa M-1. BMC Microbiol. 13: 137. https://doi.org/10.1186/1471-2180-13-137
  19. Okimura K, Ohki K, Sato Y, Ohnishi K, Sakura N. 2007. Semi-synthesis of polymyxin B (2-10) and colistin (2-10) analogs employing the trichloroethoxycarbonyl (Troc) group for side chain protection of alpha, gamma-diaminobutyric acid residues. Chem. Pharm. Bull. 55: 1724-1730. https://doi.org/10.1248/cpb.55.1724
  20. Orwa J, Govaerts C, Busson R, Roets E, Van Schepdael A, Hoogmartens J. 2001. Isolation and structural characterization of polymyxin B components. J. Chromatogr. A 912: 369-373. https://doi.org/10.1016/S0021-9673(01)00585-4
  21. Park S-H, Kim JF, Lee C, Choi S-K, Jeong H, Kim S-B, et al. 2012. Polymyxin synthetase and gene cluster thereof. US Patent 8,329,430.
  22. Park S-Y, Choi S-K, Kim J, Oh T-K, Park S-H. 2012. Efficient production of polymyxin in the surrogate host Bacillus subtilis by introducing a foreign ectB gene and disrupting the abrB gene. Appl. Environ. Microbiol. 78: 4194-4199. https://doi.org/10.1128/AEM.07912-11
  23. Schneider A, Stachelhaus T, Marahiel M. 1998. Targeted alteration of the substrate specificity of peptide synthetases by rational module swapping. Mol. Gen. Genet. 257: 308-318. https://doi.org/10.1007/s004380050652
  24. Shaheen M, Li J, Ross AC, Vederas JC, Jensen SE. 2011. Paenibacillus polymyxa PKB1 produces variants of polymyxin B-type antibiotics. Chem. Biol. 18: 1640-1648. https://doi.org/10.1016/j.chembiol.2011.09.017
  25. Sharma S, Wu A, Chandramouli N, Fotsch C, Kardash G, Bair K. 1999. Solid-phase total synthesis of polymyxin B1. J. Pept. Res. 53: 501-506. https://doi.org/10.1034/j.1399-3011.1999.00045.x
  26. Sieber SA, Marahiel MA. 2005. Molecular mechanisms underlying nonribosomal peptide synthesis: approaches to new antibiotics. Chem. Rev. 105: 715-738. https://doi.org/10.1021/cr0301191
  27. Stachelhaus T, Schneider A, Marahiel MA. 1995. Rational design of peptide antibiotics by targeted replacement of bacterial and fungal domains. Science 269: 69-72. https://doi.org/10.1126/science.7604280
  28. Storm DR, Rosenthal KS, Swanson PE. 1977. Polymyxin and related peptide antibiotics. Annu. Rev. Biochem. 46: 723-763. https://doi.org/10.1146/annurev.bi.46.070177.003451
  29. Tambadou F, Caradec T, Gagez A-L, Bonnet A, Sopéna V, Bridiau N, et al. 2015. Characterization of the colistin (polymyxin E1 and E2) biosynthetic gene cluster. Arch. Microbiol. 197: 521-532. https://doi.org/10.1007/s00203-015-1084-5
  30. Tsubery H, Ofek I, Cohen S, Eisenstein M, Fridkin M. 2002. Modulation of the hydrophobic domain of polymyxin B nonapeptide: effect on outer-membrane permeabilization and lipopolysaccharide neutralization. Mol. Pharmacol. 62: 1036-1042. https://doi.org/10.1124/mol.62.5.1036
  31. Vaara M, Fox J, Loidl G, Siikanen O, Apajalahti J, Hansen F, et al. 2008. Novel polymyxin derivatives carrying only three positive charges are effective antibacterial agents. Antimicrob. Agents Chemother. 52: 3229-3236. https://doi.org/10.1128/AAC.00405-08
  32. Wang HH, Isaacs FJ, Carr PA, Sun ZZ, Xu G, Forest CR, Church GM. 2009. Programming cells by multiplex genome engineering and accelerated evolution. Nature 460: 894-898. https://doi.org/10.1038/nature08187
  33. Zavascki AP, Goldani LZ, Li J, Nation RL. 2007. Polymyxin B for the treatment of multidrug-resistant pathogens: a critical review. J. Antimicrob. Chemother. 60: 1206-1215. https://doi.org/10.1093/jac/dkm357
  34. Zobel S, Kumpfmüller J, Süssmuth RD, Schweder T. 2015. Bacillus subtilis as heterologous host for the secretory production of the non-ribosomal cyclodepsipeptide enniatin. Appl. Microbiol. Biotechnol. 99: 681-691. https://doi.org/10.1007/s00253-014-6199-0

Cited by

  1. Bacillus subtilis from Soybean Food Shows Antimicrobial Activity for Multidrug-Resistant Acinetobacter baumannii by Affecting the adeS Gene vol.26, pp.12, 2015, https://doi.org/10.4014/jmb.1607.07006
  2. A Highly Efficient CRISPR-Cas9-Mediated Large Genomic Deletion in Bacillus subtilis vol.8, pp.None, 2017, https://doi.org/10.3389/fmicb.2017.01167
  3. Characterization of the Polymyxin D Synthetase Biosynthetic Cluster and Product Profile of Paenibacillus polymyxa ATCC 10401 vol.80, pp.5, 2015, https://doi.org/10.1021/acs.jnatprod.6b00807
  4. Chronicle of a Soil Bacterium: Paenibacillus polymyxa E681 as a Tiny Guardian of Plant and Human Health vol.10, pp.None, 2015, https://doi.org/10.3389/fmicb.2019.00467
  5. Marine Biosurfactants: Biosynthesis, Structural Diversity and Biotechnological Applications vol.17, pp.7, 2015, https://doi.org/10.3390/md17070408
  6. Control of the polymyxin analog ratio by domain swapping in the nonribosomal peptide synthetase of Paenibacillus polymyxa vol.47, pp.6, 2015, https://doi.org/10.1007/s10295-020-02275-7