De novo molecular drug design benchmarking.

De novo molecular design for drug discovery is a growing field. Deep neural networks (DNNs) are becoming more widespread in their use for machine learning models. As more DNN models are proposed for molecular design, benchmarking methods are crucial for the comparision and validation of these models. This review looks at recently proposed benchmarking methods Fréchet ChemNet Distance, GuacaMol and Molecular Sets (MOSES), and provides a commentary on their future potential applications in de novo molecular drug design and possible next steps for further validation of these benchmarking methods.

Electrochemical Surface-Enhanced Raman Spectroscopy as a Platform for Bacterial Detection and Identification.

The field of bacterial screening is in need of a rapid, easy to use, sensitive, and selective platform for bacterial detection and identification. Current methods of bacterial identification lack time efficiency, resulting in problems for many sectors of society. Surface-enhanced Raman spectroscopy (SERS) has been investigated as a possible candidate for bacterial screening due to its demonstrated ability to detect biological molecules with a high degree of sensitivity. However, the field of bacterial screening using SERS is currently facing limitations such as signal irreproducibility, weak spectra, and difficulty differentiating between strains based on the SERS spectra of bacteria alone. The current study reports on the first ever use of electrochemical surface-enhanced Raman spectroscopy (EC-SERS) for bacterial screening. The results of this study demonstrate the ability of EC-SERS to greatly improve upon the SERS performance for the detection of Gram-positive and Gram-negative bacteria both in terms of improved peak intensities and spectral richness. EC-SERS shows great promise in its ability to advance SERS-based bacterial screening and could potentially be used for more efficient species discrimination at the point-of-need (PON).

Insights into the Mechanism of Action of the Two-Peptide Lantibiotic Lacticin 3147.

Lacticin 3147 is a two peptide lantibiotc (LtnA1 and LtnA2) that displays nanomolar activity against many Gram-positive bacteria. Lacticin 3147 may exert its antimicrobial effect by several mechanisms. Isothermal titration calorimetry experiments show that only LtnA1 binds to the peptidoglycan precursor lipid II, which could inhibit peptidoglycan biosynthesis. An experimentally supported model of the resulting complex suggests that the key binding partners are the C-terminus of LtnA1 and pyrophosphate of lipid II. A combination of in vivo and in vitro assays indicates that LtnA1 and LtnA2 can induce rapid membrane lysis without the need for lipid II binding. However, the presence of lipid II substantially increases the activity of lacticin 3147. Furthermore, studies with synthetic LtnA2 analogues containing either desmethyl- or oxa-lanthionine rings confirm that the precise geometry of these rings is essential for this synergistic activity.

Selvamicin, an atypical antifungal polyene from two alternative genomic contexts.

The bacteria harbored by fungus-growing ants produce a variety of small molecules that help maintain a complex multilateral symbiosis. In a survey of antifungal compounds from these bacteria, we discovered selvamicin, an unusual antifungal polyene macrolide, in bacterial isolates from two neighboring ant nests. Selvamicin resembles the clinically important antifungals nystatin A1 and amphotericin B, but it has several distinctive structural features: a noncationic 6-deoxymannose sugar at the canonical glycosylation site and a second sugar, an unusual 4-O-methyldigitoxose, at the opposite end of selvamicin's shortened polyene macrolide. It also lacks some of the pharmacokinetic liabilities of the clinical agents and appears to have a different target. Whole genome sequencing revealed the putative type I polyketide gene cluster responsible for selvamicin's biosynthesis including a subcluster of genes consistent with selvamicin's 4-O-methyldigitoxose sugar. Although the selvamicin biosynthetic cluster is virtually identical in both bacterial producers, in one it is on the chromosome, in the other it is on a plasmid. These alternative genomic contexts illustrate the biosynthetic gene cluster mobility that underlies the diversity and distribution of chemical defenses by the specialized bacteria in this multilateral symbiosis.

A Rebeccamycin Analog Provides Plasmid-Encoded Niche Defense.

Bacterial symbionts of fungus-growing ants occupy a highly specialized ecological niche and face the constant existential threat of displacement by another strain of ant-adapted bacteria. As part of a systematic study of the small molecules underlying this fraternal competition, we discovered an analog of the antitumor agent rebeccamycin, a member of the increasingly important indolocarbazole family. While several gene clusters consistent with this molecule's newly reported modification had previously been identified in metagenomic studies, the metabolite itself has been cryptic. The biosynthetic gene cluster for 9-methoxyrebeccamycin is encoded on a plasmid in a manner reminiscent of plasmid-derived peptide antimicrobials that commonly mediate antagonism among closely related Gram-negative bacteria.

Variable genetic architectures produce virtually identical molecules in bacterial symbionts of fungus-growing ants.

Small molecules produced by Actinobacteria have played a prominent role in both drug discovery and organic chemistry. As part of a larger study of the actinobacterial symbionts of fungus-growing ants, we discovered a small family of three previously unreported piperazic acid-containing cyclic depsipeptides, gerumycins A-C. The gerumycins are slightly smaller versions of dentigerumycin, a cyclic depsipeptide that selectively inhibits a common fungal pathogen, Escovopsis. We had previously identified this molecule from a Pseudonocardia associated with Apterostigma dentigerum, and now we report the molecule from an associate of the more highly derived ant Trachymyrmex cornetzi. The three previously unidentified compounds, gerumycins A-C, have essentially identical structures and were produced by two different symbiotic Pseudonocardia spp. from ants in the genus Apterostigma found in both Panama and Costa Rica. To understand the similarities and differences in the biosynthetic pathways that produced these closely related molecules, the genomes of the three producing Pseudonocardia were sequenced and the biosynthetic gene clusters identified. This analysis revealed that dramatically different biosynthetic architectures, including genomic islands, a plasmid, and the use of spatially separated genetic loci, can lead to molecules with virtually identical core structures. A plausible evolutionary model that unifies these disparate architectures is presented.

Lassomycin, a ribosomally synthesized cyclic peptide, kills mycobacterium tuberculosis by targeting the ATP-dependent protease ClpC1P1P2.

Languishing antibiotic discovery and flourishing antibiotic resistance have prompted the development of alternative untapped sources for antibiotic discovery, including previously uncultured bacteria. Here, we screen extracts from uncultured species against Mycobacterium tuberculosis and identify lassomycin, an antibiotic that exhibits potent bactericidal activity against both growing and dormant mycobacteria, including drug-resistant forms of M. tuberculosis, but little activity against other bacteria or mammalian cells. Lassomycin is a highly basic, ribosomally encoded cyclic peptide with an unusual structural fold that only partially resembles that of other lasso peptides. We show that lassomycin binds to a highly acidic region of the ClpC1 ATPase complex and markedly stimulates its ATPase activity without stimulating ClpP1P2-catalyzed protein breakdown, which is essential for viability of mycobacteria. This mechanism, uncoupling ATPase from proteolytic activity, accounts for the bactericidal activity of lassomycin.

Biochemical, structural, and genetic characterization of tridecaptin A1, an antagonist of Campylobacter jejuni.

Bacillus circulans NRRL B-30644 (now Paenibacillus terrae) was previously reported to produce SRCAM 1580, a bacteriocin active against the food pathogen Campylobacter jejuni. We have been unable to isolate SRCAM 1580, and did not find any genetic determinants in the genome of this strain. We now report the reassignment of this activity to the lipopeptide tridecaptin A1. Structural characterization of tridecaptin A1 was achieved through NMR, MS/MS and GC-MS studies. The structure was confirmed through the first chemical synthesis of tridecaptin A1, which also revealed the stereochemistry of the lipid chain. The impact of this stereochemistry on antimicrobial activity was examined. The biosynthetic machinery responsible for tridecaptin production was identified through bioinformatic analyses. P. terrae NRRL B-30644 also produces paenicidin B, a novel lantibiotic active against Gram-positive bacteria. MS/MS analyses indicate that this lantibiotic is structurally similar to paenicidin A.

Structural characterization of the highly cyclized lantibiotic paenicidin A via a partial desulfurization/reduction strategy.

Lantibiotics are ribosomally synthesized antimicrobial peptides produced by bacteria that are increasingly of interest for food preservation and possible therapeutic uses. These peptides are extensively post-translationally modified, and are characterized by lanthionine and methyllanthionine thioether cross-links. Paenibacillus polymyxa NRRL B-30509 was found to produce polymyxins and tridecaptins, in addition to a novel lantibiotic termed paenicidin A. A bacteriocin termed SRCAM 602 previously reported to be produced by this organism and claimed to be responsible for inhibition of Campylobacter jejuni could not be detected either directly or by genomic analysis. The connectivities of the thioether cross-links of paenicidin A were solved using a novel partial desulfurization/reduction strategy in combination with tandem mass spectrometry. This approach overcame the limitations of NMR-based structural characterization that proved mostly unsuccessful for this peptide. Paenicidin A is a highly cyclized lantibiotic, containing six lanthionine and methyllanthionine rings, three of which are interlocking.

Response of methicillin-resistant Staphylococcus aureus to amicoumacin A.

Amicoumacin A exhibits strong antimicrobial activity against methicillin-resistant Staphylococcus aureus (MRSA), hence we sought to uncover its mechanism of action. Genome-wide transcriptome analysis of S. aureus COL in response to amicoumacin A showed alteration in transcription of genes specifying several cellular processes including cell envelope turnover, cross-membrane transport, virulence, metabolism, and general stress response. The most highly induced gene was lrgA, encoding an antiholin-like product, which is induced in cells undergoing a collapse of Δψ. Consistent with the notion that LrgA modulates murein hydrolase activity, COL grown in the presence of amicoumacin A showed reduced autolysis, which was primarily caused by lower hydrolase activity. To gain further insight into the mechanism of action of amicoumacin A, a whole genome comparison of wild-type COL and amicoumacin A-resistant mutants isolated by a serial passage method was carried out. Single point mutations generating codon substitutions were uncovered in ksgA (encoding RNA dimethyltransferase), fusA (elongation factor G), dnaG (primase), lacD (tagatose 1,6-bisphosphate aldolase), and SACOL0611 (a putative glycosyl transferase). The codon substitutions in EF-G that cause amicoumacin A resistance and fusidic acid resistance reside in separate domains and do not bring about cross resistance. Taken together, these results suggest that amicoumacin A might cause perturbation of the cell membrane and lead to energy dissipation. Decreased rates of cellular metabolism including protein synthesis and DNA replication in resistant strains might allow cells to compensate for membrane dysfunction and thus increase cell survivability.

The 3D structure of thuricin CD, a two-component bacteriocin with cysteine sulfur to α-carbon cross-links.

Thuricin CD is an antimicrobial factor that consists of two peptides, Trn-α and Trn-ß, that exhibit synergistic activity against drug resistant strains of Clostridium difficile. Trn-α and Trn-ß each possess three sulfur to α-carbon thioether bridges for which the stereochemistry is unknown. This report presents the three-dimensional solution structures of Trn-α and Trn-ß. Structure calculations were performed for the eight possible stereoisomers of each peptide based on the same NMR data. The structure of the stereoisomer that best fit the experimental data was chosen as the representative structure for each peptide. It was determined that Trn-α has L-stereochemistry at Ser21 (α-R), L-stereochemistry at Thr25 (α-R), and D-stereochemistry at Thr28 (α-S) (an LLD isomer). Trn-ß was also found to be the LLD isomer, with L-stereochemistry at Thr21 (α-R), L-stereochemistry at Ala25 (α-R), and D-stereochemistry at Tyr28 (α-S).

Biosynthesis of aminovinyl-cysteine-containing peptides and its application in the production of potential drug candidates.

Bacteria produce a wide array of metabolites to protect themselves from competing microbes. These antimicrobial compounds include peptides with an S-[(Z)-2-aminovinyl]-d-cysteine (AviCys) or S-[(Z)-2-aminovinyl]-(3S)-3-methyl-d-cysteine (AviMeCys) residue, which have been isolated from several different bacterial species. The peptides are structurally diverse: some feature polycyclic backbones, such as the lantibiotic epidermin, and others feature a mostly linear structure, such as cypemycin. Each of the AviCys-containing peptides characterized to date exhibit highly potent biological activities, ranging from antimicrobial activity against methicillin-resistant Staphylococcus aureus (MRSA) to anticancer activity against mouse leukemia cells. The AviCys-containing peptides gallidermin and mutacin 1140 have been suggested as possible treatments of acne and of throat infections, respectively. Unfortunately, their low production yield in fermentation (typically only 10-200 mg/L) remains a major hindrance to the widespread use and clinical testing of AviCys-containing peptides for human therapeutics. Although scientists have made great strides in the total chemical synthesis of polycyclic peptides on solid support, an efficient method to form the AviCys ring has yet to be developed. In light of these difficulties, it may be possible to draw inspiration from the natural biosynthesis of AviCys-containing peptides within the producer organisms. In this Account, we examine the characteristics of the enzymes responsible for constructing AviCys to evaluate possibilities for generating high yields of bioactive AviCys- or AviMeCys-containing peptides for research and clinical use. The gene cluster for the biosynthesis of epidermin has been studied in depth, leading to the proposal for a mechanism of AviCys formation. First, a serine residue upstream of the C-terminus is enzymatically dehydrated to form a dehydroalanine residue. Then, the C-terminal cysteine residue is oxidatively decarboxylated to form an enethiolate, which subsequently cyclizes onto the dehydroalanine to give the AviCys ring. Extensive research on EpiD, the enzyme responsible for the oxidative decarboxylation reaction, has led to its purification and cocrystallization with a model substrate peptide, yielding an X-ray crystal structure. An in vitro assay of the enzyme with a library of synthetic heptapeptides has resulted in the discovery that EpiD has low absolute substrate specificity and can oxidatively decarboxylate a wide variety of C-terminal cysteine-containing peptides. Recently, the gene cluster for the biosynthesis of cypemycin was also identified. Despite certain structural similarities between cypemycin and the lantibiotic peptides, analysis of the biosynthetic genes suggests that cypemycin production is quite different from that of the lantibiotics. In particular, the AviCys residue in cypemycin is formed from two cysteine residues instead of one serine and one cysteine, and the CypD enzyme that catalyzes the oxidative decarboxylation of the C-terminal cysteine shows little homology to EpiD. The knowledge accrued from studying EpiD and CypD could be used to develop a semisynthetic methodology to produce AviCys-containing peptides. In particular, suitable precursor peptides could be synthesized on solid support before being fed to either of these enzymes in vitro to generate the C-terminal AviCys moiety. Exploring the potential of this methodology could lead to the efficient production of epidermin, cypemycin, and analogues thereof.

Chemical synthesis and biological evaluation of gallidermin-siderophore conjugates.

The lantibiotic gallidermin was modified at lysine residues by regioselective attachment of derivatives of pyochelin, agrobactin and desferrioxamine B with the objective of having siderophore receptors of Gram-negative bacteria transport the antibiotic-iron chelator conjugate through the outer membrane. All of the conjugates retained activity against the Gram-positive indicator strain, Lactococcus lactis subsp. cremoris HP. However, testing of the conjugates against several Gram-negative strains yielded unexpected results. Bacteria treated with 100 µM of the conjugates complexed with Fe(3+) grew better than bacteria grown in iron-free media but worse than bacteria grown in the same media supplemented with 10 µM FeCl(3). Although these findings indicate that the conjugates are unable to inhibit the growth of Gram-negative bacteria, they indicate penetration of the outer membrane and provide structure-activity information for design of other lantibiotic conjugates. The synthetic strategy is applicable for linking biomarkers or fluorescence probes to gallidermin for studies on its localization and mode of action. As there are many lantibiotics that operate with unknown mechanisms of action, this chemical approach provides a means to modify such peptides with biomarkers for biological investigations.

The activity of bacteriocins from Carnobacterium maltaromaticum UAL307 against gram-negative bacteria in combination with EDTA treatment.

Bacteriocins from gram-positive bacteria are potent antimicrobial peptides that inhibit pathogenic and food-spoilage bacteria. They are usually ineffective against gram-negative bacteria because they cannot penetrate the outer membrane (OM). Disruption of the OM of some gram-negative bacteria was reported to sensitize them to certain bacteriocins. This study evaluates the activity of three purified bacteriocins [carnocyclin A (CclA), carnobacteriocin BM1 (CbnBM1) and piscicolin 126 (PisA)] produced by Carnobacterium maltaromaticum UAL307, which has been approved for preservation of food in United States and Canada, against three gram-negative bacteria (Escherichia coli DH5α, Pseudomonas aeruginosa ATCC 14207 and Salmonella Typhimurium ATCC 23564). Their efficacy is compared with bacteriocins of other classes: the lantibiotics nisin A (positive control) and gallidermin, and the cyclic peptide subtilosin A (SubA). In combination with EDTA, CclA inhibited both E. coli and Pseudomonas. PisA inhibited Pseudomonas, but CbnBM1 showed weak activity toward Pseudomonas. In comparison, nisin and gallidermin inhibited the growth of all three strains, whereas SubA was active against E. coli and Pseudomonas only at high concentrations. The results reveal that UAL307 bacteriocins can inhibit gram-negative bacteria if the OM is weakened, and that the different classes of bacteriocins in this study exert unique modes of action toward such bacteria.

Thuricin CD, a posttranslationally modified bacteriocin with a narrow spectrum of activity against Clostridium difficile.

The last decade has seen numerous outbreaks of Clostridium difficile-associated disease (CDAD), which presented significant challenges for healthcare facilities worldwide. We have identified and purified thuricin CD, a two-component antimicrobial that shows activity against C. difficile in the nanomolar range. Thuricin CD is produced by Bacillus thuringiensis DPC 6431, a bacterial strain isolated from a human fecal sample, and it consists of two distinct peptides, Trn-alpha and Trn-beta, that act synergistically to kill a wide range of clinical C. difficile isolates, including ribotypes commonly associated with CDAD (e.g., ribotype 027). However, this bacteriocin thuricin CD has little impact on most other genera, including many gastrointestinal commensals. Complete amino acid sequencing using infusion tandem mass spectrometry indicated that each peptide is posttranslationally modified at its respective 21st, 25th, and 28th residues. Solution NMR studies on [(13)C,(15)N] Trn-alpha and [(13)C,(15)N]Trn-beta were used to characterize these modifications. Analysis of multidimensional NOESY data shows that specific cysteines are linked to the alpha-carbons of the modified residues, forming three sulfur to alpha-carbon bridges. Complete sequencing of the thuricin CD gene cluster revealed genes capable of encoding two S'-adenosylmethionine proteins that are characteristically associated with unusual posttranslational modifications. Thuricin CD is a two-component antimicrobial peptide system with sulfur to alpha-carbon linkages, and it may have potential as a targeted therapy in the treatment of CDAD while also reducing collateral impact on the commensal flora.

Isolation of a variant of subtilosin A with hemolytic activity.

Bacillus subtilis produces an anionic bacteriocin called subtilosin A that possesses antibacterial activity against certain gram-positive bacteria. In this study, we uncovered a hemolytic mutant of B. subtilis that produces an altered form of subtilosin A. The mutant bacteriocin, named subtilosin A1, has a replacement of threonine at position 6 with isoleucine. In addition to the hemolytic activity, subtilosin A1 was found to exhibit enhanced antimicrobial activity against specific bacterial strains. The B. subtilis albB mutant that does not produce a putative immunity peptide was more sensitive to both subtilosin A and subtilosin A1. A spontaneous suppressor mutation of albB that restored resistance to subtilosin A and subtilosin A1 was obtained. The sbr (subtilosin resistance) mutation conferring the resistance is not linked to the sboA-alb locus. The sbr mutation does not increase the resistance of B. subtilis to other cell envelope-targeted antimicrobial agents, indicating that the mutation specifically confers the resistance to subtilosins. The findings suggest possible bioengineering approaches for obtaining anionic bacteriocins with enhanced and/or altered bactericidal activity. Furthermore, future identification of the subtilosin-resistant mutation could provide insights into the mechanism of subtilosin A activity.

Approaches to the discovery of new antibacterial agents based on bacteriocins.

The development of antibiotic resistance in pathogenic bacteria has led to a search for novel classes of antimicrobial drugs. Bacteriocins are peptides that are naturally produced by bacteria and have considerable potential to fulfill the need for more effective bacteriocidal agents. In this mini-review, we describe research aimed at generating analogues of bacteriocins from lactic acid bacteria, with the goal of gaining a better understanding of structure-activity relationships in these peptides. In particular, we report recent findings on synthetic analogues of leucocin A, pediocin PA1, and lacticin 3147 A2, as well as on the significance of these results for the design and production of new antibiotics.

The Saccharomyces cerevisiae succinate dehydrogenase does not require heme for ubiquinone reduction.

The coupling of succinate oxidation to the reduction of ubiquinone by succinate dehydrogenase (SDH) constitutes a pivotal reaction in the aerobic generation of energy. In Saccharomyces cerevisiae, SDH is a tetramer composed of a catalytic dimer comprising a flavoprotein subunit, Sdh1p and an iron-sulfur protein, Sdh2p and a heme b-containing membrane-anchoring dimer comprising the Sdh3p and Sdh4p subunits. In order to investigate the role of heme in SDH catalysis, we constructed an S. cerevisiae strain expressing a mutant enzyme lacking the two heme axial ligands, Sdh3p His-106 and Sdh4p Cys-78. The mutant enzyme was characterized for growth on a non-fermentable carbon source, for enzyme assembly, for succinate-dependent quinone reduction and for its heme b content. Replacement of both Sdh3p His-106 and Sdh4p Cys-78 with alanine residues leads to an undetectable level of cytochrome b(562). Although enzyme assembly is slightly impaired, the apocytochrome SDH retains a significant ability to reduce quinone. The enzyme has a reduced affinity for quinone and its catalytic efficiency is reduced by an order of magnitude. To better understand the effects of the mutations, we employed atomistic molecular dynamic simulations to investigate the enzyme's structure and stability in the absence of heme. Our results strongly suggest that heme is not required for electron transport from succinate to quinone nor is it necessary for assembly of the S. cerevisiae SDH.