• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Nov 14, 2006; 103(46): 17462–17467.
Published online Nov 7, 2006. doi:  10.1073/pnas.0608589103
PMCID: PMC1859951

Combinatorial biosynthesis of novel antibiotics related to daptomycin


Daptomycin, a cyclic lipopeptide produced by Streptomyces roseosporus, is the active ingredient of Cubicin (daptomycin-for-injection), a first-in-class antibiotic approved for treatment of skin and skin-structure infections caused by Gram-positive pathogens and bacteremia and endocarditis caused by Staphylococcus aureus, including methicillin-resistant strains. Genetic engineering of the nonribosomal peptide synthetase (NRPS) in the daptomycin biosynthetic pathway was exploited for the biosynthesis of novel active antibiotics. λ-Red-mediated recombination was used to exchange single or multiple modules in the DptBC subunit of the NRPS to modify the daptomycin cyclic peptide core. We combined module exchanges, NRPS subunit exchanges, inactivation of the tailoring enzyme glutamic acid 3-methyltransferase, and natural variations of the lipid tail to generate a library of novel lipopeptides, some of which were as active as daptomycin against Gram-positive bacteria. One compound was more potent against an Escherichia coli imp mutant that has increased outer membrane permeability. This study established a robust combinatorial biosynthesis platform to produce novel peptide antibiotics in sufficient quantities for antimicrobial screening and drug development.

Keywords: cubicin, genetic engineering, nonribosomal peptide, Streptomyces

Daptomycin (Fig. 1A) is a member of the A21978C family of cyclic anionic 13-amino acid lipopeptides produced by a nonribosomal peptide synthetase (NRPS) mechanism in Streptomyces roseosporus (13). During fermentation, three major compounds (A21978C1–3) that have branched 11-, 12-, or 13-carbon chain fatty acids coupled to the N-terminal Trp residue normally are produced, but daptomycin, which has a straight 10-carbon chain fatty acid, is the predominant product if n-decanoic acid is provided during fermentation (3, 4). Daptomycin has potent in vitro bactericidal activity against Gram-positive pathogens, including methicillin-resistant Staphylococcus aureus (MRSA), penicillin-resistant Streptococcus pneumoniae (PRSP), vancomycin-resistant enterococci (VRE), and vancomycin-resistant S. aureus (VRSA) (3, 57). Cubicin (daptomycin-for-injection) was approved by the Food and Drug Administration for the treatment of skin and skin-structure infections caused by Gram-positive pathogens (5, 8) and bacteremia and endocarditis caused by S. aureus, including MRSA (9). Besides linezolid, daptomycin is the only novel antibiotic structural class to be approved in the last four decades.

Fig. 1.
Lipopeptide antibiotics and daptomycin gene cluster. (A) Structurally related lipopeptides daptomycin, A54145 and CDA. (B) Subunit and module organizations ...

Although daptomycin is efficacious for the indications discussed above, its analogs can be optimized for additional clinical uses, e.g., for the treatment of community-acquired pneumonia. A number of semisynthetic derivatives of daptomycin that contained modified lipid tails or side chains coupled through the δ-amino group of ornithine (Orn6) were synthesized, but they represented limited structural diversity, and none had improved pharmacological properties (3, 10). Attempts to synthesize daptomycin derivatives with changes in the amino acid core by chemical (I. Parr, personal communication) and chemoenzymatic (11) methods have been hampered by the lack of commercially available 3-methyl-glutamic acid (for 3mGlu12). The lack of suitable chemical approaches for the total synthesis of daptomycin analogs makes combinatorial biosynthesis an important alternative to generate and scale up derivatives of daptomycin for clinical evaluation.

The cloning and characterization of the biosynthetic gene clusters of daptomycin (12) and related lipopeptides, A54145 (13) and calcium-dependent antibiotic (CDA) (14) (Fig. 1), provided the starting materials for making derivatives of daptomycin. The daptomycin (dpt) biosynthetic gene cluster contains dptA, dptBC, and dptD, coding for the three NRPS subunits needed for incorporation of the 13 aa (Fig. 1B). As in other NRPSs, the daptomycin NRPS subunits are organized into modules (Figs. 1B and and2)2) with catalytic domains. Adenylation (A) domains selectively activate specific amino acids, and condensation (C) domains couple them to the peptide chain. Thiolation (T) domains located after A domains are pantetheinylated to provide an attachment point for the growing peptide chain as well as for the activated amino acids before condensation. Modules that incorporate d-amino acids usually contain epimerase (E) domains after T domains. The multimodular organization of NRPSs typically follows the colinearity rule in which the substrate binding specificity of the series of modules reflects the order of amino acid residues in the final products. Such organization can facilitate the construction of modified NRPSs in which defined functional modules (e.g., CAT or CATE) or domains (e.g., A or AT) are excised and replaced.

Fig. 2.
λ-Red-mediated recombination for construction of hybrid dptBC. The coding region of domains (CAT, in this example) of module d-Ala8 was replaced by an antibiotic resistance cassette flanked by rare restriction sites. The cassette was then excised, ...

After some initial successes in producing novel compounds by manipulation of NRPSs using domain or module exchanges (1518), most recent studies have focused on in vitro characterization of specific domains, substrate specificity, and intermodule communication by using purified proteins including truncated NRPSs (18). Several amino acid residues in the interpeptide docking sites were identified as important for maintaining or preventing functional interactions of NRPS subunits (19, 20). In an in vitro approach, rationally designed cyclic peptides and antibiotics were synthesized by using thioesterase-catalyzed chemoenzymatic biosynthesis (21, 22).

We recently described bioengineered daptomycin analogs in which Kyn13 was substituted with Ile, Val, or Trp by using subunit exchange of the DptD subunit with the heterologous LptD and CdaPS3 subunits from the A54145 and CDA NRPSs, respectively (2325). In the present study, we expanded the modification of the daptomycin amino acid core at residues nonconserved among structurally related lipopeptides by replacing single or multiple modules in the DptBC subunit with modules from daptomycin and A54145 NRPSs, and we combined them with DptD subunit exchanges, disruption of Glu12 methylation (26), and natural lipid side-chain variations to generate a combinatorial library of novel lipopeptides. Characterization of purified compounds yielded valuable information on structural features critical for antibiotic activity and identified candidates for further modification and clinical development.


Experimental System for Combinatorial Biosynthesis.

To carry out combinatorial biosynthesis, we used different strains of S. roseosporus deleted for different sets of daptomycin genes (Table 3, which is published as supporting information on the PNAS web site, and Fig. 3) as hosts to express engineered dptBC genes, and we used heterologous lptD and cdaPS3 in place of dptD, with and without the dptI gene essential for the Glu12 methylation. We used λ-Red-mediated recombination (27) to exchange new domains or modules into the dptBC gene at specific intradomain linker sites (outlined in Figs. 2 and and44).

Fig. 3.
Daptomycin gene content in deletion mutants and expression plasmids. (A) NRPS deletion mutants as expression hosts. (B) Expression plasmids for hybrid NRPS pathways. Relative locations of domain and module exchanges on dptBC are marked by filled squares. ...
Fig. 4.
Amino acid sequence alignment of linkers of modules from the biosynthetic pathways of S. roseosporus daptomycin (Sr), S. fradiae A54145 (Sf), and Streptomyces ...

Control experiments indicated that strain UA431 (ΔdptEFABCDGHIJ mutant) was fully complemented when pDA300 containing dptEFABCDGHIJ was inserted at the [var phi]C31 attB site. Likewise, KN100 (ΔdptBCD mutant) was fully complemented by coexpressing plasmids pKN24 (dptBC) and pRB04 (dptD), which integrate into the chromosome at [var phi]C31 attB and IS117 attB sites, respectively. In KN125 (ΔdptBCDGHIJ mutant), coexpression of pKN24 and pRB04 led to production of a series of compounds containing Glu12 instead of 3mGlu12 (26). Replacement of dptD by cdaPS3 and lptD resulted in the production of novel compounds in which Kyn13 was replaced by Trp and Ile or Val (because of the substrate flexibility of LptD), respectively (23, 24). To introduce and recombine new changes at positions 8 or 11 with those at positions 12 and 13, plasmids with module exchanges in dptBC were introduced into a series of KN100- and KN125-derived strains carrying pRB04 (dptD), pMF23 (cdaPS3), or pMF30 (lptD) (Table 3 and Fig. 3).

Exchange of Homologous Modules Within DptBC.

New domains/modules can be introduced into DptBC between splicing sites at the ends of CAT or CATE modules: two sites are within the T–C linker (TC1 and TC2) and one is within each of the T–E or E–C linkers (TE1 and EC1; Fig. 4), giving four possible configurations for CAT or CATE exchanges. Modules 8 and 11, comprising the CAAlaTE and CASerTE domains, are highly homologous (12), making them good candidates for reciprocal module exchanges. Because leaving the native E domains in place may preserve homologous intermodule and interpeptide interactions, exchanges of CAT were made in pDA300, which carries the entire dpt NRPS gene cluster (Fig. 3B). λ-Red-mediated recombination (27) was used to introduce a tetA gene to delete the DNA sequence coding for CAAlaT domains of the d-Ala8 module between TC2 and TE1 (Figs. 2 and and4).4). The tetA gene then was replaced by a DNA sequence coding for CASerT from the d-Ser11 module (Fig. 2), resulting in pDR2153. A reciprocal construct (pDR2158) in which CASerT of d-Ser11 was replaced by CAAlaT was made in a similar manner. To monitor the effects of inserting restriction sites in the linker region, CAAlaT and CASerT also were reintroduced at positions 8 and 11, respectively.

S. roseosporus UA431 strains expressing either one of these control plasmids produced native lipopeptides at control levels (200–250 mg/liter). When pDR2153 containing CASerT exchanged for CAAlaT was introduced into UA431, lipopeptides with mass ions of 1,636, 1,650, and 1,664 were produced, consistent with the incorporation of d-Ser at position 8. Similarly, when pDR2158 containing CAAlaT exchanged for CASerT was introduced into UA431, three mass ions of 1,604, 1,618, and 1,632 were detected, consistent with the incorporation of d-Ala at position 11. The levels of production of the lipopeptides were 35 mg/liter and 100 mg/liter, respectively, and the purified compounds were as potent as daptomycin against S. aureus (Table 1).

Table 1.
Antibiotic activity of daptomycin analogs

Exchange of Single Heterologous Modules.

To introduce the d-Asn11 module from the A54145 pathway (13) at positions 8 and 11, we made module exchanges in pKN24 expressing only dptBC (Fig. 3B). Fusions at the splicing sites TC1 and TE1 or EC1 generated four CAAsnT or CAAsnTE exchanges at positions 8 and 11 (see Fig. 4). The engineered plasmids were introduced into KN156, lacking only dptBC, and recombinants containing CAT inserted at TC1–TE1 or CATE at TC1–EC1 produced mass ions of 1,677, 1,691, and 1,705, consistent with the incorporation of d-Asn at position 8, and 1,660, 1,674, and 1,688, consistent with the incorporation of d-Asn at position 11. The yields of compounds containing d-Asn8 or d-Asn11 were 19 and 33 mg/liter in CAT exchanges and 6 and 19 mg/liter in CATE exchanges, respectively. Similar results were obtained when d-Asn11 was introduced by using the splicing site TC2 in combination with TE1 or EC1 sites (Fig. 4). These results indicate that functional module exchanges can be made at different sites in the linkers, that linker regions can be modified to accommodate new restrictions sites, and that the native E domains of the d-Ala8 and d-Ser11 modules can catalyze the epimerization of l-Ala, l-Ser, and l-Asn. CAT exchanges that left the native E domain in place seemed to be superior to CATE exchanges, perhaps because the native E domain may provide better protein–protein interactions between adjacent modules in dptBC and between the hybrid dptBC and native dptD proteins. The purified compound containing d-Asn11 had antibiotic activity comparable with that of daptomycin, but that containing d-Asn8 was less active (Table 1).

Exchange of Multiple Heterologous Modules.

To modify several amino acid residues concurrently, four modules (d-Ala8-Asp9-Gly10-d-Ser11) on pKN24 were replaced by modules d-Lys8-OmAsp9-Gly10-d-Asn11 from the LptC subunit of the A54145 NRPS. [In the A54145 producing Streptomyces fradiae, the methoxylation of Asp9 is likely catalyzed by tailoring enzymes (13)]. In addition to the coding region of LptC, the DNA fragment also contained the stop codon of lptB, so the resulting plasmid (pKN69) may express two separate subunits, DptB* (Orn6-Asp7) followed by LptC (d-Lys8-OmAsp9-Gly10-d-Asn11). Purified lipopeptides produced by a recombinant carrying pKN69 had mass ions of 1,717, 1,731, and 1,745, as predicted for analogs containing Lys8, Asp9, Gly10, and Asn11. The results show that the multimodule exchange was successful, and that Asp can be incorporated at position 9 without methoxylation, as the tailoring functions (for hydroxylation and O-methylation) were not provided by S. roseosporus.

The fermentation yield of the hybrid compounds was ≈1 mg/liter, and the purified compound was as active as daptomycin against S. aureus (Table 1). This finding is of particular interest because the substitution of d-Lys8 for d-Ala8 changed the net charge on the molecule from −3 to −2.

Combinatorial Biosynthesis of Lipopeptides with Modifications at Positions 8, 11, 12, and 13, and in the Lipid Tail.

By using the strategy outlined earlier, the module exchanges at positions 8 and 11 in DptBC were combined with amino acid substitutions at positions 12 and 13 by deletion of the methyltransferase DptI and by exchanges of the last subunit (DptD), respectively. In total, 30 hybrid NRPS biosynthetic pathways were constructed by combining five module configurations at positions 8 or 11, three terminal subunits for variations at position 13, and two configurations at position 12 (3mGlu or Glu). Considering that LptD incorporates Ile or Val (albeit at a lower efficiency) at position 13 (23), and DptEF naturally couples three different lipid side chains to initiate lipopeptide biosynthesis, these pathways have the potential of producing from 90 to 120 novel lipopeptides. In fermentation, 21 of 30 strains expressing these pathways produced over 60 novel mass ions. Generally, multiple modifications in the NRPS led to lower production (Table 4, which is published as supporting information on the PNAS web site). The titers for the readily identified compounds varied from ≈100 mg/liter to ≈1 mg/liter. Some of the hybrid NRPSs with multiple changes generated truncated products or failed to produce lipopeptides.

Several compounds were purified and tested for antibiotic activity against S. aureus (Table 1). Compared with the compounds containing single amino acid substitutions at position 11, which had essentially equivalent in vitro potency as daptomycin, those with single exchanges at positions 12 or 13 showed reduced potency. The combined introduction of d-Lys8 and d-Asn11 did not influence activity, but some combinatorial modifications at position 8 or 11 and position 12 (Glu12) and/or 13 (Ile13) resulted in significant reductions in activity.

Antibacterial Spectrum of Novel Lipopeptides.

The five most active compounds were further assayed for antibiotic activity against a panel of Gram-positive and Gram-negative bacteria (Table 2). Except for S. pneumoniae, these compounds were as active as daptomycin against Gram-positive test strains, including mutants with reduced susceptibility to daptomycin. As expected, all were inactive against wild-type Escherichia coli and the maltoporin (ΔlamB) mutant [minimal inhibitory concentration (MIC) >256 μg/ml]. Most also exhibited a low activity (MIC 128 μg/ml) against an E. coli imp mutant that has increased outer membrane permeability to large molecules (28) except CB182290. Interestingly, this analog containing d-Asn11 displayed an increased activity against the imp mutant (MIC 32 μg/ml). The in vivo efficacy of the most potent compounds is an area for investigation in animal models.

Table 2.
Antibacterial profile of daptomycin analogs


Combinatorial engineering of polyketide biosynthetic pathways has been extensively explored (29), but only a small number of successful in vivo exchanges of single modules have been reported for NRPS-derived compounds (1518). In this study, we used combinatorial biosynthesis to generate daptomycin analogs. We demonstrated exchange of single and multiple modules from heterologous pathways, and we combined them with subunit exchanges, blockage of Glu methylation, and natural variation in side-chain lipidation to generate a library of hybrid lipopeptide pathways. Thirty novel NRPS biosynthetic pathways were constructed, and many produced novel active antibiotics in quantities sufficient for unambiguous analytical verification, purification, and antimicrobial characterization. These results are significant because many factors are required for successful engineering of NRPS pathways, including functional protein–protein interactions within and between enzyme subunits, effective peptide coupling, and cyclization of hybrid compounds (1618). Interestingly, in one experiment, the six-module subunit DptBC was successfully replaced by a new arrangement comprising a truncated two-module subunit (DptB*) that was required to interact with a four-module subunit (LptC) to function. There are no apparent dedicated interpeptide docking sequences between LptB and LptC normally (13), so the interaction between DptB* and LptC may function by a different mechanism, perhaps translational frameshifting, to bypass the stop codon. Biochemical studies (30) suggested the native A54145 NRPS is composed of two large and one small subunit, instead of four subunits, consistent with read-through of the stop codon between lptB and lptC.

The building of novel compounds around the scaffold of daptomycin has helped to initiate a deeper understanding of structure–activity relationships in the cyclic peptide core. Although none of the hybrid compounds produced by combinatorial biosynthesis is superior to daptomycin, several have equivalent potencies. It is clear from this and other work (23, 25) that the amino acids present at positions 12 and 13 play key roles in determining antibacterial potency, whereas substitutions at position 11 seem to have little effect (Table 1). Substitutions at position 8 were well tolerated in the case of Ser and Lys (in conjunction with Asn at position 11), but substitution of Asn reduced the potency. The disubstitution of Lys8 and Asn11 was particularly interesting in that it generated a compound with an MIC of 1 μg/ml. This result indicated that although the replacement of the small and hydrophobic d-Ala by the larger, basic d-Lys and d-Ser by d-Asn increases the basic charge from +1 to +2 and net charge from −3 to −2, it does not affect the antibiotic potency significantly in daptomycin-sensitive strains. However, the compound was less active than daptomycin against strains with reduced susceptibility to daptomycin, indicating that the overall charge on the molecule may be an important factor to consider when designing molecules with increased activity against daptomycin-resistant pathogens.

The compounds described in this study represent only a fraction of the possibilities offered by combinatorial biosynthesis. The significant structural diversity already generated can be further expanded by medicinal chemistry toward an even larger set of molecules to explore as potential candidates for clinical development: e.g., more extensive substitutions of the fatty acid tail, which previous studies have shown can have a pronounced effect on potency and toxicity of daptomycin related lipopeptides (2, 3, 10), and modifications of the amino and hydroxyl groups of the introduced residues such as d-Lys8 and d-Ser8, respectively. Introduction of modules specific for modified amino acids together with their tailoring enzymes (e.g., hydroxylase and methyltransferase for Asp) would provide additional options to generate diversity.

In conclusion, this study has demonstrated that combinatorial biosynthesis is an effective approach to generate novel lipopeptides related to daptomycin, an important antibiotic recently approved for clinical use that cannot be modified effectively by medicinal chemistry alone. The combinatorial biosynthesis methodology described here generated many new derivatives with modifications in the core peptide that have provided valuable information on features of the molecule required for good antibacterial activity. The demonstration of functional module exchanges in vivo at different interdomain linker sites should be applicable to other secondary metabolites produced by NRPS or mixed NRPS/polyketide synthetase pathways.

Materials and Methods

Media, Growth Conditions, and General Methods.

Bacterial strains and plasmids are given in Table 3. S. roseosporus strains were grown at 30°C on AS-1 agar or liquid TSB medium (12). Lipopeptides produced in fermentations were identified by electrospray-ionization liquid chromatography-MS, quantified by HPLC using daptomycin as the standard, and purified as described (23). Plasmids were transformed into E. coli by electroporation and transferred to S. roseosporus by conjugation (23). MICs were determined by using microdilution techniques in calcium-supplemented Mueller Hinton broth (MHB) (12). E. coli strains were grown in LB medium (31) at 37°C except when a temperature-sensitive λ-Red plasmid (pKD78 or pKD119) was present. For the latter, cells were grown at 30°C and λ-Red expression was induced for 2 h by l-arabinose (final 0.2% wt/vol) (27). S. aureus, Enterococcus faecalis, Enterococcus faecium, and S. pneumoniae strains were grown at 37°C in MHB. Antibiotics were added into media at the following final concentrations: 100 μg/ml ampicillin (Ap); 50 μg/ml apramycin (Am); 25 μg/ml chloramphenicol (Cm); 50 μg/ml gentamicin (Gen); 50 μg/ml hygromycin (Hyg); 50 μg/ml kanamycin (Kan); 50 μg/ml nalidixic acid (Nal); 50 μg/ml spectinomycin (Spc); 15 μg/ml tetracycline (Tet); and 50 μg/ml thiostrepton (Ts). The following antibiotic resistance cassettes were PCR-amplified and used in plasmid deletions by λ-Red-mediated recombination: aad9(Spcr) [National Center for Biotechnology Information (NCBI) accession no. AP002527], aph(2″)-Ib(Genr) (NCBI accession no. AF207840) available under the control of the phage T5 promoter integrated into the E. coli chromosome (CM141; C. Monahan, personal communication), bla(Apr) (NCBI accession no. J01749), and tetA(Tetr) (NCBI accession no. J01749) available from a multiple-component cassette (K.T.N., unpublished data).

Construction of Deletion Cassettes.

λ-Red-mediated recombination (27) was used to introduce PCR-generated cassettes [Long Template Expand PCR kit (Roche, Indianapolis, IN) or Advantage GC-2 polymerase mix (Clontech, Mountain View, CA)] and delete targeted sequences of the cloned dpt gene cluster. The cassettes may contain a resistance gene [aad9, aph(2″)-Ib, bla, or tetA] alone or in combination with the strong promoter ermEp* (32). Primers (Table 5, which is published as supporting information on the PNAS web site) used for amplification of the deletion cassette were composed of a terminal 45- to 50-nt gene-specific sequence for targeted recombination and a proximal 20-nt universal sequence (single-underlined in Table 5) for portable use of these primers with multiple-resistance gene templates. Some of these cassettes are flanked by engineered restriction sites AvrII or PmeI (double-underlined in Table 5). The PCR products were purified, concentrated to 100 ng/μl, and used to electroporate into arabinose-induced E. coli BW25113 carrying a λ-Red system-expressing plasmid (pKD78 or pKD119) and the targeted Amr plasmid (pCV1, pDA300, pDR2153, pDR2158, or pKN24). Plasmids recovered from transformants grown on AS-1 agar containing Am and Ap, Gen, Spc, or Tet were purified and checked by restriction analysis and sequencing of the fusion sites.

Construction of pDA300 and pKN24 by λ-Red-Mediated Recombination.

pDA300 and pKN24 were constructed by truncation of pCV1, a BAC clone that contains the daptomycin biosynthetic gene cluster (NCBI accession no. AY787762) (12). For construction of pDA300, the region upstream of the dptMN locus (Figs. 1 and and3)3) was replaced by aad9 amplified by the primers Sp6 del3 and Sp6 del4 (Table 5). The region downstream of dptJ (Figs. 1 and and3)3) was replaced by bla using the primers GTC del1 and GTC del2. For construction of pKN24, the region upstream of dptBC (nucleotides 191 to 69,057, NCBI accession no. AY787762) was replaced by an aad9-ermEp* cassette PCR-amplified by using primers Sp6Del-1-2 and dptBC-ermEp from a template (D.A., unpublished data) containing the aad9 gene, followed by the ermEp* promoter and an optimized ribosome binding site (32). The region downstream of dptBC (nucleotides 91,140 to 127,346) then was replaced by the bla gene with primers DptD-3′::amp and primer GTC del2.

Deletion of Domains and Modules by λ-Red-Mediated Recombination.

Deletions at modules d-Ala8 or d-Ser11 were done by using splicing sites in the T–C linker (TC1 or TC2) and in the T–E linker (TE1) or the E–C linker (EC1) (Figs. 2 and and4).4). For TC2–TE1 deletions, the primers Ala8_AvrII-left and Ala8_PmeI-right were used to amplify the deletion tetA cassette. For TC1–TE1 deletions at modules 8 and 11, the primer 8_B-AvrII and 8_CAT-PmeI were used to amplify the aph(2″) gene. As the homologous regions of the primers are almost identical for the d-Ala8 and d-Ser11 linker regions, individual disruptions of both modules were obtained by using the same linear PCR cassette. Because there is a single base mismatch (C6339G) in the T–C linker preceding modules 8 and 11, recombination introduced an Ala to Gly mutation in the T–C linker preceding the d-Ser11 module. For TC1–EC1 deletions at module 8 or 11, the aph(2″) gene was amplified by primer 8_B-AvrII and 8-CATE2-PmeI or 11-CATE2-PmeI. By using a similar protocol, the DNA sequence coding for d-Ala8-Asp9-Gly10-d-Ser11 on pKN24 was replaced by aph(2″) amplified by using primers 8_B-AvrII and 11_SuE-PmeI; the latter anneals to a sequence downstream of the dptBC stop codon.

Gap-Repair Cloning by λ-Red-Mediated Recombination.

The 3.1-kb DNA fragments coding for the CAT domains of d-Ala8 and d-Ser8 were cloned from pCV1 by the “gap-repair” method using λ-Red-mediated recombination (Fig. 2). Primers Ala-8-gap L and Ala-8-gap R were used to amplify the origin of replication and bla region of pBR322. The PCR product has 45 bp at each end that are homologous to the 5′ and 3′ ends of the CAT domains of both d-Ala8 and d-Ser8 modules (which are 100% identical in nucleotide sequence), flanked by unique NheI (5′ end) and HpaI (3′ end) sites. After transformation of this PCR product into arabinose-induced E. coli BW25113 carrying pKD78 and pCV1, recombination between the 45-bp terminal sequences of the linear pBR322 PCR fragment and the homologous regions in pCV1 fills in the gap between the ends of pBR322 with fragments from the dpt gene cluster to generate a circular pBR322 derivative that confers Apr to E. coli. The insert ends were checked by sequencing. To clone modules from the A54145 biosynthetic pathway, pBR322 PCR fragments were transformed into E. coli BW25113 carrying pKD119 and pCB01. For cloning the d-Asn11 module, PCR product was generated by using primers Lpt-N11-B-P13 and Lpt-N11-CAT-P14 (for the TC1–TE1 fragment) or Lpt-N11-CATE2-P14 (for the TC1–EC1 fragment). For cloning d-Lys8-OmAsp9 -Gly10-d-Asn11, PCR primers Lpt-K8-B-P13 and Lpt-N11-SUE-P14 were used. The inserts were excised by NheI and HpaI digestion and gel-purified.

Construction of Hybrid dptBC Plasmids.

The antibiotic cassette that had been inserted to replace modules in pKN24 was excised by digestion with AvrII and PmeI, and the linearized plasmid was ligated to compatible NheI and HpaI restriction sites of the above-described fragments (Fig. 2). The ligation mix was redigested with AvrII to eliminate self-ligated plasmids, dialyzed, and transformed into E. coli DH10B. The final hybrid plasmids were recovered from colonies grown on LB plates containing Am and checked by restriction digestion and sequencing of the fusion sites.

Deletion of dptD and the dptGHIJ Locus in pDA300 Derivatives (pDR2153 and pDR2158).

dptD and dptGHIJ in pDR2153 and pDR2158 were replaced by the aph(2″) cassette using λ-Red-mediated recombination. The cassette was amplified by using primers dptD-del1 and dptD-del3.

Supplementary Material

Supporting Tables:


We dedicate this paper to the memory of Dr. Francis P. (Frank) Tally. We thank C. Monahan (Cubist Pharmaceuticals) for providing E. coli CM141, S. Doekel (Cubist Pharmaceuticals) for providing pCB01, X. He and V. Rajgarhia for fermentation support, D. Kau and S. Wrigley for purification and liquid chromatography–MS analyses of some compounds, N. Cotroneo for MIC determinations, and J. Silverman and J. Alder for comments on the manuscript. This work was supported in part by National Institutes of Health Small Business Innovation Research Grant 2 R44 GM068173-08.


nonribosomal peptide synthetase
calcium-dependent antibiotic
minimal inhibitory concentration.


The authors declare no conflict of interest.


1. Debono M, Barnhart M, Carrell CB, Hoffmann JA, Occolowitz JL, Abbott BJ, Fukuda DS, Hamill RL, Biemann K, Herlihy WC. J Antibiot. 1987;40:761–777. [PubMed]
2. Debono M, Abbott BJ, Molloy RM, Fukuda DS, Hunt AH, Daupert VM, Counter FT, Ott JL, Carrell CB, Howard LC, et al. J Antibiot. 1988;41:1093–1105. [PubMed]
3. Baltz RH, Miao V, Wrigley SK. Nat Prod Rep. 2005;22:717–741. [PubMed]
4. Huber FM, Pieper RL, Tietz AJ. J Biotechnol. 1988;7:283–292.
5. Arbeit RD, Maki D, Tally FP, Campanaro E, Eisenstein BI. Clin Infect Dis. 2004;38:1673–1681. [PubMed]
6. Eisenstein BI. Expert Opin Investig Drugs. 2004;13:1159–1169. [PubMed]
7. Raja A, LaBonte J, Lebbos J, Kirkpatrick P. Nat Rev Drug Discov. 2003;2:943–944. [PubMed]
8. Schriever CA, Fernandez C, Rodvold KA, Danziger LH. Am J Health Syst Pharm. 2005;62:1145–1158. [PubMed]
9. Fowler VG, Boucher HW, Corey GR, Abrutyn E, Karchmer AW, Rupp ME, Levine DP, Chambers HF, Tally FP, Vigliani GA, et al. N Engl J Med. 2006;355:653–665. [PubMed]
10. Hill J, Siedlecki J, Parr I, Morytko M, Yu X, Zhang Y, Silverman J, Controneo N, Laganas V, Li T, et al. Bioorg Med Chem Lett. 2003;13:4187–4191. [PubMed]
11. Grünewald J, Sieber SA, Mahlert C, Linne U, Marahiel MA. J Am Chem Soc. 2004;126:17025–17031. [PubMed]
12. Miao V, Coëffet-Le Gal M-F, Brian P, Brost R, Penn J, Whiting A, Martin S, Ford R, Parr R, Bouchard M, et al. Microbiology. 2005;151:1507–1523. [PubMed]
13. Miao V, Brost R, Chapple J, She K, Coëffet-Le Gal M-F, Baltz RH. J Ind Microbiol Biotechnol. 2006;33:129–140. [PubMed]
14. Hojati Z, Milne C, Harvey B, Gordon L, Borg M, Flett F, Wilkinson B, Sidebottom PJ, Rudd BA, Hayes MA, et al. Chem Biol. 2002;9:1175–1187. [PubMed]
15. Stachelhaus T, Schneider A, Marahiel MA. Science. 1995;269:69–72. [PubMed]
16. Schwarzer D, Finking R, Marahiel MA. Nat Prod Rep. 2003;20:275–287. [PubMed]
17. Finking R, Marahiel MA. Annu Rev Microbiol. 2004;58:453–488. [PubMed]
18. Sieber SA, Marahiel MA. Chem Rev. 2005;105:715–738. [PubMed]
19. Hahn M, Stachelhaus T. Proc Natl Acad Sci USA. 2004;101:15585–15590. [PMC free article] [PubMed]
20. Hahn M, Stachelhaus T. Proc Natl Acad Sci USA. 2006;103:275–280. [PMC free article] [PubMed]
21. Kohli RM, Walsh CT. Chem Commun (Camb) 2003;3:297–307. [PubMed]
22. Grünewald J, Marahiel MA. Microbiol Mol Biol Rev. 2006;70:121–146. [PMC free article] [PubMed]
23. Miao V, Coëffet-Le Gal M-F, Nguyen K, Brian P, Penn J, Whiting A, Steele J, Kau D, Martin S, Gibson T, et al. Chem Biol. 2006;13:269–276. [PubMed]
24. Coëffet-Le Gal M-F, Thurston L, Rich P, Miao V, Baltz RH. Microbiology. 2006;152:2993–3001. [PubMed]
25. Baltz RH, Brian P, Miao V, Wrigley SK. J Ind Microbiol Biotechnol. 2006;33:66–74. [PubMed]
26. Nguyen KT, Kau D, Gu JQ, Brian P, Wrigley S, Baltz RH, Miao V. Mol Microbiol. 2006;61:1294–1307. [PubMed]
27. Datsenko KA, Wanner BL. Proc Natl Acad Sci USA. 2000;97:6640–6645. [PMC free article] [PubMed]
28. Eggert US, Ruiz N, Falcone BV, Branstrom AA, Goldman RC, Silhavy TJ, Kahne D. Science. 2001;294:361–364. [PubMed]
29. Weissman KJ, Leadlay PF. Nat Rev Microbiol. 2005;3:925–936. [PubMed]
30. Wessels P, von Döhren H, Kleinkauf H. Eur J Biochem. 1996;242:665–673. [PubMed]
31. Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. Plainview, NY: Cold Spring Harbor Lab Press; 1989.
32. Bibb MJ, Janssen GR, Ward JM. Gene. 1985;38:215–226. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem Compound links
  • EST
    Published EST sequences
  • Gene (nucleotide)
    Gene (nucleotide)
    Records in Gene identified from shared sequence links
  • MedGen
    Related information in MedGen
  • Nucleotide
    Published Nucleotide sequences
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links
  • Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...