Influence of Lipophilicity on the Antibacterial Activity of Polymyxin Derivatives and on Their Ability to Act as Potentiators of Rifampicin.

Novel polymyxin derivatives are often classified either as having direct activity against Gram-negative pathogens or as compounds inactive in their own right, which through permeabilization of the outer membrane act as potentiators of other antibiotics. Here, we report the systematic investigation of the influence of lipophilicity on microbiological activity (including against strains with reduced susceptibility to polymyxins), potentiation of rifampicin, and in vitro toxicity within a series of next-generation polymyxin nonapeptides. We demonstrate that the lipophilicity at the N-terminus and amino acids 6 and 7 in the cyclic peptide core is interchangeable and that the activity, ability to potentiate, and cytotoxicity all appear to be primarily driven by overall lipophilicity. Our work also suggests that the characterization of a polymyxin molecule as either a direct acting compound or a potentiator is more of a continuum that is strongly influenced by lipophilicity rather than as a result of fundamentally different modes-of-action.

Direct modifications of the cyclic peptide Polymyxin B leading to analogues with enhanced in vitro antibacterial activity.

Synthetic modifications have been made directly to the cyclic peptide core of polymyxin B, enabling the further understanding of structure activity relationships of this antimicrobial peptide. Such modified polymyxins are also substrates for enzymic hydrolysis, enabling the synthesis of a variety of semi-synthetic analogues, resulting in compounds with increased in vitro antibacterial activity.

Design of Next Generation Polymyxins with Lower Toxicity: The Discovery of SPR206.

Polymyxins are an important class of antibiotics for the treatment of bacterial infections due to multidrug resistant Gram-negative pathogens. However, their clinical utility is limited by nephrotoxicity. Here, we report a series of promising next generation polymyxin nonapeptides identified on the basis of our understanding of the relationship of structure with activity, cytotoxicity, and kidney compartment accumulation. We demonstrate that nonapeptides with an amine-containing N-terminal moiety of specific regio- and stereochemistry possess superior in vitro activity, together with lower cytotoxicity compared to polymyxin B. We further demonstrate that compounds with a ß-branched aminobutyrate N-terminus with an aryl substituent offer a promising combination of low cytotoxicity and kidney exposure, leading to low toxicity in the mouse. From this series, SPR206 has been selected as a development candidate.

Antibacterial activity of the novel semisynthetic lantibiotic NVB333 in vitro and in experimental infection models.

NVB333 is a novel semisynthetic lantibiotic derived from the amide coupling of 3,5-dichlorobenzylamine to the C-terminal of deoxyactagardine B. The in vitro activity of NVB333 includes efficacy against clinically relevant pathogens including methicillin-resistant Staphylococcus aureus and vancomycin-resistant Enterococcus spp. NVB333 shows no cross-resistance with other antibiotics tested and a very low propensity for resistance development. After intravenous dosing NVB333 has high exposure in mouse plasma and shows generally improved in vivo activity compared with vancomycin in mouse infection models despite modest MIC values. In thigh infection models, promising efficacy was demonstrated against several strains of S. aureus including methicillin-resistant S. aureus (MRSA) and vancomycin-intermediate S. aureus (VISA) strains, and against Enterococcus faecalis UNT126-3. Area under the concentration curve (AUC)/MIC was shown to be the best predictor of efficacy against S. aureus UNT103-3 with an AUC/MIC of 138 (uncorrected for protein binding) achieving a static effect. NVB333 was also effective in a disseminated infection model where it conferred complete survival from the MRSA strain ATCC 33591. NVB333 showed rather modest lung penetration after intravenous dosing (AUC in lung 2-3% of plasma AUC), but because of very high plasma exposure, therapeutic levels of compound were achieved in the lung. Efficacy at least equal to vancomycin was demonstrated against an MRSA strain (UNT084-3) in a bronchoalveolar infection model. The impressive in vivo efficacy of NVB333 and strong resistance prognosis makes this compound an interesting candidate for development for treating systemic Gram-positive infections.

Generation of an actagardine A variant library through saturation mutagenesis.

The lantibiotic actagardine A is nineteen amino acids in length and comprises three intertwined C-terminal methyllanthionine-bridged rings and an N-terminal lanthionine-bridged ring. Produced by the actinomycete Actinoplanes garbadinensis ATCC 31049, actagardine A demonstrates antibacterial activity against important Gram-positive pathogens. This activity combined with its ribosomal synthesis makes it an attractive target for the generation of lantibiotic variants with improved biological activity. A variant generation system designed to allow the specific substitution of amino acids at targeted sites throughout the actagardine A peptide has been used to generate a comprehensive library by site-directed mutagenesis. With the exception of residues involved in bridge formation, each amino acid in the actagardine A peptide as well as the alanine (ala(0)) at position -1 relative to the mature peptide, has been systematically substituted with all remaining 19 amino acids. A total of 228 mutants have been engineered with 44 produced in good yield. The mutant V15F in particular demonstrates improved activity against a range of notable Gram-positive pathogens including Clostridium difficile, when evaluated alongside actagardine A. The scope of variants generated provides an insight into the flexibility of the actagardine A processing machinery and will undoubtedly assist in future mutational studies.

Organization of the biosynthetic genes encoding deoxyactagardine B (DAB), a new lantibiotic produced by Actinoplanes liguriae NCIMB41362.

Deoxyactagardine B (DAB) is a hitherto unknown type B lantibiotic, produced by Actinoplanes liguriae NCIMB41362. The mature peptide is 19 amino acids in length and structurally analogous to actagardine, differing by two amino acids (V15L and I16V) and the absence of a sulfoxide bond between residues 14 and 19. The biosynthetic genes encoding DAB are clustered, and in addition to the structural gene ligA include genes believed to encode for the proteins responsible for the modification, transport and regulation of DAB synthesis. Surprisingly, despite the presence of a gene that shares significant homology to the monooxygenase garO from the actagardine biosynthetic gene cluster, the oxidized form of DAB has not been detected. A lanA gene encoding the DAB peptide has been introduced into the plasmid pAGvarX and delivered into a strain of Actinoplanes garbadinensis lacking the structural gene for actagardine, garA (A. garbadinensis DeltagarA). Expression of this gene in A. garbadinensis DeltagarA resulted in the production of actagardine B, an oxidized form of DAB.

Biosynthesis of the putative siderophore erythrochelin requires unprecedented crosstalk between separate nonribosomal peptide gene clusters.

The genome of the erythromycin-producing bacterium Saccharopolyspora erythraea contains many orphan secondary metabolite gene clusters including two (nrps3 and nrps5) predicted to govern biosynthesis of nonribosomal peptide-based siderophores. We report here the production by S. erythraea, even under iron-sufficient conditions, of a 2,5-diketopiperazine siderophore candidate we have named erythrochelin. Deletion of the nonribosomal peptide synthetase (NRPS) gene ercD within the nrps5 cluster abolished erythrochelin production. The tetrapeptide backbone of erythrochelin (alpha-N-acetyl-delta-N-acetyl-delta-N-hydroxyornithine-serine-delta-N-hydroxyornithine-delta-N-acetyl-delta-N-hydroxyornithine) suggests an orthodox colinear model for erythrochelin assembly. Curiously, the delta-N-acetyltransferase required for erythrochelin biosynthesis is encoded within a remote NRPS-cluster (nrps1) whose own NRPS contains an inactivating mutation. Disruption of the nrps1 gene mcd abolished erythrochelin biosynthesis, which could then be restored by addition of synthetic L-delta-N-acetyl-delta-N-hydroxyornithine, confirming an unprecedented example of functional crosstalk between nrps clusters.

Dissecting structural and functional diversity of the lantibiotic mersacidin.

Mersacidin is a tetracyclic lantibiotic with antibacterial activity against Gram-positive pathogens. To probe the specificity of the biosynthetic pathway of mersacidin and obtain analogs with improved antibacterial activity, an efficient system for generating variants of this lantibiotic was developed. A saturation mutagenesis library of the residues of mersacidin not involved in cycle formation was constructed and used to validate this system. Mersacidin analogs were obtained in good yield in approximately 35% of the cases, producing a collection of 82 new compounds. This system was also used for the production of deletion and insertion mutants of mersacidin. The outcome of these studies suggests that this system can be extended to produce mersacidin variants with multiple changes that will allow a full investigation of the potential use of modified mersacidins as therapeutic agents.

Organization of the genes encoding the biosynthesis of actagardine and engineering of a variant generation system.

The biosynthetic pathway of the type B lantibiotic actagardine (formerly gardimycin), produced by Actinoplanes garbadinensis ATCC31049, has been cloned, sequenced and annotated. The gene cluster contains the gene garA that encodes the actagardine prepropeptide, a modification gene garM, involved in the dehydration and cyclization of the prepeptide, several putative transporter and regulatory genes as well as a novel luciferase-like monooxygenase gene designated garO. Expression of these genes in Streptomyces lividans resulted in the production of ala(0)-actagardine while deletion of the garA gene from A. garbadinensis generated a strain incapable of producing actagardine. Actagardine production was successfully restored however, by the delivery of the plasmid pAGvarX. This plasmid contains an engineered cassette of the actagardine encoding gene garA and offers an alternative route to generating extensive libraries of actagardine variants. Using this plasmid, an alanine scanning library has been constructed and the mutants analysed. Further modifications include the removal of the novel garO gene from A. garbadinensis. Deletion of this gene resulted in the production of deoxy variants of actagardine, demonstrating that the formation of the sulfoxide group is enzyme catalysed and not a spontaneous chemical modification as previously believed.

A new modular polyketide synthase in the erythromycin producer Saccharopolyspora erythraea.

A previously unidentified set of genes encoding a modular polyketide synthase (PKS) has been sequenced in Saccharopolyspora erythraea, producer of the antibiotic erythromycin. This new PKS gene cluster (pke) contains four adjacent large open reading frames (ORFs) encoding eight extension modules, flanked by a number of other ORFs which can be plausibly assigned roles in polyketide biosynthesis. Disruption of the pke PKS genes gave S. erythraea mutant JC2::pSBKS6, whose growth characteristics and pattern of secondary metabolite production did not apparently differ from the parent strain under any of the growth conditions tested. However, the pke PKS loading module and individual pke acyltransferase domains were shown to be active when used in engineered hybrid PKSs, making it highly likely that under appropriate conditions these biosynthetic genes are indeed expressed and active, and synthesize a novel polyketide product.