Trent J. Oman

Follow the leader: The use of leader peptides to guide natural product biosynthesis

The avalanche of genomic information in the past decade has revealed that natural product biosynthesis using the ribosomal machinery is much more widespread than originally anticipated. Nearly all of these compounds are crafted through post-translational modifications of a larger precursor peptide that often contains the marching orders for the biosynthetic enzymes. We review here the available information for how the peptide sequences in the precursors govern the post-translational tailoring processes for several classes of natural products. In addition, we highlight the great potential these leader peptide-directed biosynthetic systems offer for engineering conformationally restrained and pharmacophore-rich products with structural diversity that greatly expands the proteinogenic repertoire.

Sublancin is not a lantibiotic but an S-linked glycopeptide

Sublancin is shown to be an S-linked glycopeptide containing a glucose attached to a Cys residue, establishing a new post-translational modification. The activity of the S-glycosyl transferase was reconstituted in vitro and the enzyme is shown to have relaxed substrate specificity allowing the preparation of analogs of sublancin. Glycosylation is essential for its antimicrobial activity.

In Vitro Mutasynthesis of Lantibiotic Analogues Containing Nonproteinogenic Amino Acids

Lantibiotics are ribosomally synthesized and post-translationally modified peptide antibiotics containing the characteristic thioether cross-links lanthionine and methyllanthionine. To date, no analogues of lantibiotics that contain nonproteinogenic amino acids have been reported. In this study, in vitro-reconstituted lacticin 481 synthetase was used in conjunction with synthetic peptide substrates containing nonproteinogenic amino acids to generate 11 analogues of lacticin 481. These analogues contained sarcosine and aminocyclopropanoic acid in place of Gly5, d-valine at position 6, 4-cyanoaminobutyric acid in place of Glu13, β3-homoarginine at the position of Asn15, N-butylglycine and β-Ala at Met16, naphthylalanine (Nal) at Trp19, 4-pyridynylalanine (Pal) at Phe21, and homophenylalanine (hPhe) at Phe23. Of these analogues, the Trp19Nal and Phe23hPhe mutants provided zones of inhibition larger than the parent compound in agar diffusion assays against the indicator strains Lactococcus lactis HP and Bacillus subtilis 6633. These two compounds also demonstrated improved MIC values against liquid cultures of L. lactis HP.

Insights into the Mode of Action of the Two- Peptide Lantibiotic Haloduracin

Haloduracin, a recently discovered two-peptide lantibiotic composed of the post-translationally modified peptides Halα and Halβ, is shown to have high potency against a range of Gram-positive bacteria and to inhibit spore outgrowth of Bacillus anthracis. The two peptides display optimal activity in a 1:1 stoichiometry and have efficacy similar to that of the commercially used lantibiotic nisin. However, haloduracin is more stable at pH 7 than nisin. Despite significant structural differences between the two peptides of haloduracin and those of the two-peptide lantibiotic lacticin 3147, these two systems show similarities in their mode of action. Like Ltnα, Halα binds to a target on the surface of Gram-positive bacteria, and like Ltnβ, the addition of Halβ results in pore formation and potassium efflux. Using Halα mutants, its B- and C-thioether rings are shown to be important but not required for bioactivity. A similar observation was made with mutants of Glu22, a residue that is highly conserved among several lipid II-binding lantibiotics such as mersacidin.

An Engineered Lantibiotic Synthetase That Does Not Require a Leader Peptide on Its Substrate

Ribosomally synthesized and post-translationally modified peptides are a rapidly expanding class of natural products. They are typically biosynthesized by modification of a C-terminal segment of the precursor peptide (the core peptide). The precursor peptide also contains an N-terminal leader peptide that is required to guide the biosynthetic enzymes. For bioengineering purposes, the leader peptide is beneficial because it allows promiscuous activity of the biosynthetic enzymes with respect to modification of the core peptide sequence. However, the leader peptide also presents drawbacks as it needs to be present on the core peptide and then removed in a later step. We show that fusing the leader peptide for the lantibiotic lacticin 481 to its biosynthetic enzyme LctM allows the protein to act on core peptides without a leader peptide. We illustrate the use of this methodology for preparation of improved lacticin 481 analogues containing non-proteinogenic amino acids.

Haloduracin α Binds the Peptidoglycan Precursor Lipid II with 2:1 Stoichiometry

The two-peptide lantibiotic haloduracin is composed of two post-translationally modified polycyclic peptides that synergistically act on Gram-positive bacteria. We show here that Halα inhibits the transglycosylation reaction catalyzed by PBP1b by binding in a 2:1 stoichiometry to its substrate lipid II. Halβ and the mutant Halα-E22Q were not able to inhibit this step in peptidoglycan biosynthesis, but Halα with its leader peptide still attached was a potent inhibitor. Combined with previous findings, the data support a model in which a 1:2:2 lipid II:Halα:Halβ complex inhibits cell wall biosynthesis and mediates pore formation, resulting in loss of membrane potential and potassium efflux.

The glycosyltransferase involved in thurandacin biosynthesis catalyzes both O- and S-glycosylation

The S-glycosyltransferase SunS is a recently discovered enzyme that selectively catalyzes the conjugation of carbohydrates to the cysteine thiol of proteins. This study reports the discovery of a second S-glycosyltransferase, ThuS, and shows that ThuS catalyzes both S-glycosylation of the thiol of cysteine and O-glycosylation of the hydroxyl group of serine in peptide substrates. ThuS-catalyzed S-glycosylation is more efficient than O-glycosylation, and the enzyme demonstrates high tolerance with respect to both nucleotide sugars and peptide substrates. The biosynthesis of the putative products of the thuS gene cluster was reconstituted in vitro, and the resulting S-glycosylated peptides thurandacin A and B exhibit highly selective antimicrobial activity toward Bacillus thuringiensis.

Non-proteinogenic amino acids in lacticin 481 analogues result in more potent inhibition of peptidoglycan transglycosylation.

Lantibiotics are ribosomally synthesized and post-translationally modified peptide natural products that contain the thioether structures lanthionine and methyllanthionine and exert potent antimicrobial activity against Gram-positive bacteria. At present, detailed modes-of-action are only known for a small subset of family members. Lacticin 481, a tricyclic lantibiotic, contains a lipid II binding motif present in related compounds such as mersacidin and nukacin ISK-1. Here, we show that lacticin 481 inhibits PBP1b-catalyzed peptidoglycan formation. Furthermore, we show that changes in potency of analogues of lacticin 481 containing non-proteinogenic amino acids correlate positively with the potency of inhibition of the transglycosylase activity of PBP1b. Thus, lipid II is the likely target of lacticin 481, and use of non-proteinogenic amino acids resulted in stronger inhibition of the target. Additionally, we demonstrate that lacticin 481 does not form pores in the membranes of susceptible bacteria, a common mode-of-action of other lantibiotics.

NMR structure of the S-linked glycopeptide sublancin 168.

Sublancin 168 is a member of a small group of glycosylated antimicrobial peptides known as glycocins. The solution structure of sublancin 168, a 37-amino-acid peptide produced by Bacillus subtilis 168, has been solved by nuclear magnetic resonance (NMR) spectroscopy. Sublancin comprises two α-helices and a well-defined interhelical loop. The two helices span residues 6–16 and 26–35, and the loop region encompasses residues 17–25. The 9-amino-acid loop region contains a β-S-linked glucose moiety attached to Cys22. Hydrophobic interactions as well as hydrogen bonding are responsible for the well-structured loop region. The three-dimensional structure provides an explanation for the previously reported extraordinary high stability of sublancin 168.

Mode of action and structure–activity relationship studies of geobacillin I

Lanthipeptides are lanthionine- and methyllanthionine- containing peptides that are ribosomally-synthesized and post-translationally modified.1 Lanthipeptides that possess antimicrobial activity are called lantibiotics.2 Lanthionines consist of two alanine residues that are linked through a thioether that connects their β-carbons, and methyllanthionines contain an additional methyl group (Figure 1a). Nisin is the best studied and longest known lantibiotic and has been used as a food preservative for over 50 years.3,4 Nisin displays antibacterial activity against clinically important pathogens such as methicillin-resistant Staphylococcus aureus, vancomycin-resistant Enterococci, Streptococcus pneumoniae, and food-borne pathogens such as Clostridium botulinum, and Listeria monocytogenes.5–7 Despite its use for over 50 years, reports of resistance against nisin have been scarce.8–12 The slow development of resistance may stem from the dual mode-of-action of nisin. Nisin exhibits antimicrobial activity by binding to the pyrophosphate moiety of lipid II (Figure 1b),13,14 a membrane-bound advanced intermediate involved in the biosynthesis of the cell wall. By doing so, nisin inhibits the transglycosylation step in cell wall biogenesis and sequesters lipid II from its functional location.15,16 Furthermore, the nisin-lipid II complex leads to formation of pores in the membrane causing cell death.17 Figure 1 Structures of a) lanthionine and methyllanthionine, b) lipid II, and c) geobacillin I and nisin. The shorthand notation for lanthionine (Lan) and methyllanthionine (MeLan) depicted in panel a is used in panel c. Recently, we characterized two lanthipeptides, geobacillin I and geobacillin II, from the thermophilic bacterium Geobacillus thermodenitrificans NG80-2.18 Geobacillin I contains seven thioether bridges, one dehydroalanine (Dha), and one dehydrobutyrine (Dhb) (Figure 1c). The N-terminal A and B rings of geobacillin I are very similar to the corresponding rings of nisin but the C-terminal structures are very different (Figure 1c). The nisin A and B rings are involved in lipid II binding,13 and hence we anticipated that geobacillin I might also bind lipid II. The three amino acid linker peptide between the C and D rings of nisin has been shown to be indispensable for pore formation activity.17,19–21 For instance, the ΔN20ΔM21 and N20P/M21P mutants of nisin lost pore formation ability against Gram-positive bacteria.17,19 Geobacillin I has only a single amino acid between the C and D rings, similar to the ΔN20ΔM21 mutant of nisin. Thus, based on the available data on nisin, we anticipated that geobacillin I would bind to lipid II, but not form pores in the membrane of Gram-positive bacteria. In this work we tested these expectations experimentally. We first conducted antimicrobial activity assays in liquid medium. In these assays, geobacillin I exhibited a four-fold higher minimal inhibitory concentration (MIC) against Bacillus subtilis ATCC 6633 compared to nisin (Table 1). Flow cytometry was then used to examine changes in the polarization of the bacterial membrane of B. subtilis ATCC 6633 upon incubation with geobacillin I using the membrane potential sensitive dye 3,3′-diethyloxacarbocyanine iodide (DiOC2).22 Incubation with geobacillin I resulted in a significant decrease in mean fluorescence intensity (MFI), similar to the observations when the same experiments were carried out with nisin (Figure 2a and Supplementary Figures S1 and S2). The unexpected ability of geobacillin I to form pores despite the single amino acid linker between rings C and D may be a consequence of the overall differences between the C-terminal region of geobacillin I and nisin (Figure 1c). Figure 2 The effect of geobacillin I and nisin on the membrane integrity of B. subtilis ATCC 6633. (a) Average mean fluorescence intensity (MFI) of triplicate flow cytometry measurements with different concentrations of nisin and geobacillin I using DiOC2 as indicator ... Table 1 Specific activity of nisin, geobacillin I and the geobacillin I analogues generated in this study against B. subtilis ATCC6633. We also investigated the efficiency of pore formation by geobacillin I using propidium iodide (PI), a membrane impermeable fluorescent dye. Upon pore formation or membrane disruption, PI can enter the cell, resulting in an increase in fluorescence intensity because of the interaction of PI with nucleic acids. PI uptake was monitored at nine different concentrations with each experiment conducted in triplicate (Figure 2b and Supplementary Figure S3). The data showed only two-fold lower efficiency in PI uptake for geobacillin I, with IC50 values for nisin at 0.3 μM compared to 0.6 μM for geobacillin I (Figure 2b and Supplementary Figure S3). Previously, site-saturation mutagenesis was performed on the amino acids in the linker between the C- and D-rings of nisin. The antibiotic activity of the nisin mutants N20P, M21V, K22T, and K21A was increased against several pathogenic bacteria.23 Conversely, introduction of a double Pro in this region (N20P/M21P) strongly decreased pore formation;17 this double mutant as well as the single M21P mutant also had strongly reduced antimicrobial activities. 17,23 To evaluate how such mutations would affect the activity of geobacillin I, site-directed mutagenesis was used in this study to replace the naturally occuring Leu19 in geobacillin I with Pro, and to introduce the tripeptide AsnValAla as linker between the C- and D-rings, thus generating a linker sequence that combines two of the mutations in the nisin variants with improved activity. These analogs were generated by co-expression of mutants of the precursor peptide GeoAI with the modification enzymes GeoB and GeoC in Escherichia coli as previously reported for the production of wild type geobacillin I (Supplementary Figure S4; Tables S1 and S2).18 Compared to wild type geobacillin, the analogues with NVA and P as the linker between the C and D rings had eight-fold and two-fold increased MIC values, respectively (Table 1). The ability to induce pore formation by these analogues was also investigated. Although the efficiency of pore formation was strongly reduced, replacement of Leu19 with Pro did not abolish this activity (Figure 2b). Introduction of the amino acid residues NVA in this region region also greatly reduced formation of pores in the bacterial cell membrane by geobacillin I. Thus, mutations in the linker peptide between the C- and D-rings affect the activities of nisin and geobacillin quite differently, suggesting that the detailed mechanism of pore formation by geobacillin I differs from that of nisin. These findings also suggest that the structure of the C-terminus of class I lantibiotics may vary significantly while retaining pore formation activity. The ability of geobacillin I to bind lipid II was investigated next using in vitro inhibition of the transglycosylation reaction catalyzed by penicillin-binding protein 1b (PBP1b) from E. coli. PBP1b uses lipid II as a substrate for glycan polymerization.24 Geobacillin I inhibited PBP1b-catalyzed peptidoglycan formation using 4 μM heptaprenyl lipid II with a half-maximal inhibitory concentration (IC50) of 4.6 ± 0.8 μM (Figure 2c). For comparison, inhibition by nisin under the same conditions displayed an IC50 of 2.9 ± 0.6 μM (Figure 2c). Thus, the inhibitory activity of the two peptides is very similar. Geobacillin I has higher stability at physiological pH compared to nisin.18 Although higher stability for a compound from a thermophile is not unexpected, the higher stability was somewhat surprising because nisin degradation at neutral pH is believed to be caused by non-enzymatic hydrolysis at Dha5,25 a residue that is also present in geobacillin I. We wondered whether the stability and hence antimicrobial activity of geobacillin could be further improved by mutation of Dha5 in light of a previous report that the nisin analog I4K/Dha5F/L6I had higher antimicrobial activity against various bacteria.26 Geobacillin I already has a Lys at position 4, and hence the mutant Dha5F/L6I was generated, but it proved to be only slightly more stable than the wild type geobacillin I (Supplementary Figure S4) while displaying similar MIC values (Table 1). Compared to nisin, ring C of geobacillin I is reduced in size by one amino acid, the region between rings C and D is reduced in length by two amino acids, and ring E is a lanthionine ring as opposed to a methyllanthionine ring (Figure 1c). Furthermore, geobacillin I contains two additional thioether bridges at its C-terminus. However, the antimicrobial activity and the overall mode of action of the two lantibiotics appear to be quite similar: binding to lipid II and pore formation.