Peptides having antimicrobial activity and new enzyme capable of converting L-configured residue in D-configured amino acid in a peptide

Abstract

The present invention relates to a new class of peptides having antibacterial activity and presenting D-amino acids and their uses. It also relates to a new enzyme presenting a peptide epimerase activity in vitro and in vivo, thereby being useful for modifying peptides in order to change the amino acid configuration from L to D.

Claims

1. A peptide 17 to 25 amino acids in length and in which the amino acids [V/I/A] in position 4 and 12 of the sequence as set forth in the following SEQ ID NOs have a D-configuration, wherein: a) the peptide contains SEQ ID NO: 66, 67, or 68; b) the peptide contains SEQ ID NO: 77; c) the peptide contains SEQ ID NO: 78; d) the peptide contains SEQ ID NO: 79; or e) the peptide contains SEQ ID NO: 73 and the peptide has at least one modification selected from the group consisting of N-acetylation, acylation, C-amidation, a (CH.sub.2NH) reduced bond, a (NHCO) retro-inverso bond, a (CH.sub.2-O) methylene-oxy bond, a (CH.sub.2-S) thiomethylene bond, a (CH.sub.2CH.sub.2) carba bond, a (CO—CH.sub.2) cetomethylene bond, a (CHOH—CH.sub.2) hydroxyethylene bond, a (N—N) bound, a E-alcene bond and a —CH═CH-bond.

2. The peptide according to claim 1, wherein the peptide contains SEQ ID NO: 66.

3. The peptide according to claim 1, wherein the peptide contains SEQ ID NO: 67.

4. The peptide according to claim 1, wherein the peptide contains SEQ ID NO: 68.

5. The peptide according to claim 1, wherein the peptide contains SEQ ID NO: 77.

6. The peptide according to claim 1, wherein the peptide contains SEQ ID NO: 78.

7. The peptide according to claim 1, wherein the peptide contains SEQ ID NO: 79.

8. The peptide of claim 1, wherein the peptide is between 17 to 25 amino acids in length, has at least two D-configured amino acids and contains SEQ ID NO: 73, wherein the amino acids [V/I/A] in position 4 and 12 have a D-configuration and the peptide has at least one modification selected from the group consisting of N-acetylation, acylation, C-amidation, a (CH.sub.2NH) reduced bond, a (NHCO) retro-inverso bond, a (CH.sub.2-O) methylene-oxy bond, a (CH.sub.2-S) thiomethylene bond, a (CH.sub.2CH.sub.2) carba bond, a (CO—CH.sub.2) cetomethylene bond, a (CHOH—CH.sub.2) hydroxyethylene bond, a (N—N) bound, a E-alcene bond and a —C═CH-bond.

9. A pharmaceutical or veterinary composition comprising the peptide according to claim 1.

10. A medical device or implant comprising a body having at least one surface coated with or including the peptide of claim 1.

11. A disinfectant or preservative composition comprising the peptide of claim 1.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1. Purification, spectroscopic analysis and activity of YydG. (FIG. 1A) Structure of the yydFGHIJ operon. yydF: Putative peptide, yydG: radical SAM enzyme, yydH: protease, yydIJ ABC-type transporter. (FIG. 1B) Gel electrophoresis analysis of purified YydG expressed in E. coli. (FIG. 1C) UV-visible spectrum of anaerobically reconstituted YydG. (FIG. 1D) Multiple alignments of close YydF homologs found in particular in B. subtilis and Staphylococci species. (FIG. 1E) HPLC analysis of SAM cleavage (257 nm) and (FIG. 1F) peptides produced in reaction (280 nm) after 4 hours incubation under anaerobic conditions in the presence or the absence of sodium dithionite (2 mM). (FIG. 1G) Time course for the production of 5′-dA ( and modified peptides (.diamond-solid.) quantified by reverse phase HPLC and monitored by UV-visible detection (280 nm). YydG (100 μM) was incubated under anaerobic conditions with sodium dithionite (2 mM) in the presence of 1 mM of substrate YydF.sub.18-49. (FIG. 1H & I) HPLC analysis of YydG incubated with (FIG. 1H) YydF or (FIG. 1I) a truncated version of YydF18-49. YydG, after anaerobic reconstitution, was incubated under anaerobic conditions in the presence of DTT (6 mM) and SAM (1 mM) with or without sodium dithionite (2 mM).

(2) FIG. 2—YydG catalyzes H-atom transfer to the peptide backbone. Tryptic peptide mapping and LC-MS analysis of (FIG. 2A) YydF.sub.18-49 or (FIG. 2B) YydF.sub.18-49 after incubation with YydG. Numbers indicate the m/z value for each peptide. Sequence in bold indicates the relevant peptide identified by LC-MS (i.e. Peptide 1: NH.sub.2-GLLDESQK, [M+H].sup.+=931.48; Peptide 2: VNDLWYF [M+2H].sup.2+=592.31; Peptide 3: WILGSGH-Ac, [M+H].sup.+=768.41). (FIG. 2C) LC-MS analysis of the peptide YydF.sub.18-49 after incubation with YydG in deuterated buffer. (FIG. 2D) Tryptic peptide mapping and LC-MS analysis of YydF.sub.18-49 after incubation with YydG in deuterated buffer.

(3) FIG. 3—YydG catalyzes amino acid epimerization. LC-MS/MS analysis of a (FIG. 3A) L-Ile/D-allo-Ile and (FIG. 3B) L-/D-Val (upper traces) compared with the amino acids obtained after incubation of YydF.sub.18-49 with the rSAM enzyme YydG in deuterated buffer (lower traces). The amino acids were derivatized by N-α-(2,4-dinitro-5-fluorophenyl)-L-valinamide (L-FDVA) and detected by LC-MS as FDVA-derivatives.

(4) FIG. 4—Activity of YydG mutants. (FIG. 4A) Sequence of YydG with cysteine residues highlighted. (FIG. 4B) Gel electrophoresis analysis of the purified mutant enzymes. (FIG. 4C) UV-visible spectra of A3 (i.e. AxxxAxxA) (blue trace), C22A (green trace), C222A (red trace) and C223A (purple trace) mutants after anaerobic reconstitution. (FIG. 4D) HPLC analysis of the reaction after incubation of YydF18-49 and the four mutants in the presence of SAM (1 mM) and sodium dithionite. (FIG. 4E) LC-MS analysis of the peptides produced by the C223A mutant and their corresponding masses and sequences.

(5) FIG. 5—Activity of YydF18-49 of YydG. (FIG. 5A) Plate growth inhibition assay of B. subtilis in the presence of YydF.sub.18-49 or the epimierized product YydFb. (FIG. 5B) Growth of Bacillus subtilis in LB medium in the absence of peptide (□) or in the presence of the YydF.sub.18-49 peptide with a free C-terminus: Ac-GLLDESQKLAKVNDLWYFVKSKENRWILGSGH (SEQ ID No 20) containing either no modification (.square-solid.), a D-allo-lle (.diamond-solid.), a D-Val (⋄) or two-epimerized residues: Ac-GLLDESQKLAKVNDLWYF{d-V}KSKENRW{d-I}LGSGH (SEQ ID No 20) (.box-tangle-solidup.). The OD values are the means of 3 independent cultures.

(6) FIG. 6—Multiple alignments of close YydG homologs found in particular in B. subtilis and Staphylococci species.

(7) FIG. 7—Mass spectrometry analysis of the YydF peptide isolated from B. subtilis and named YydF.sub.33-49. LC-MS/MS analysis of (a) the peptide secreted by B. subtilis, (b) synthetic YydF.sub.33-49DD peptide containing two D-amino acids residues and (c) synthetic YydF.sub.33-49 peptide. Relevant ions are indicated. Tables 1 for full assignment. In the peptide sequence, amino acids with a D-configuration are indicated in grey.

(8) FIG. 8—(A) Growth of B. subtilis in liquid LB medium in the presence of YydF.sub.33-49 or YydF.sub.33-49DD. B. subtilis was grown in LB medium alone (.diamond-solid.), in the presence of YydF.sub.33-49 (□) (SEQ ID NO: 76, top) or YydF.sub.33-49DD (.square-solid.) (SEQ ID NO: 75, bottom). Each measurement is the mean of three growth experiments with the SD indicated. The epimerized residues are in black. (b) Growth ratio of B. subtilis in presence of YydF.sub.18-49 (100 μM) or YydF.sub.18-49DD (100, 10 or 1 μM). Ratios were determined by comparison with growth in the absence of peptide. (c) Growth ratio of B. subtilis in presence of YydF.sub.33-49 or YydF.sub.33-49DD (100, 10, 1, 0.1 or 0.01 μM). Ratios were determined by comparison with bacterial growth in the absence of peptide.

(9) FIG. 9—Growth of B. subtilis in liquid LB medium in the presence of YydF.sub.18-49 (□) or after addition (arrow) of the YydF.sub.18-49DD (.square-solid.) after 3 hours of growth. Each measurement is the mean of three growth experiments with the SD indicated.

(10) FIG. 10—Growth of B. subtilis in liquid LB medium in the presence of YydF.sub.33-48 (a), YydF.sub.33_49AA (b) (SEQ ID NO: 66), YydF.sub.33_49vv (c) (SEQ ID NO: 67), YydF.sub.33_49II (d) (SEQ ID NO: 68), Peptide-SA1 (e) (SEQ ID NO: 62), Peptide-SA2 (f) (SEQ ID NO: 63), Peptide-SE (g) (SEQ ID NO: 64) or Peptide-SP (h) (SEQ ID NO: 65) at 100 μM. In bold, epimerized residues.

(11) FIG. 11—Growth of Enterococcus faecalis in liquid BHI medium in the presence of YydF.sub.33-49 (a), YydF.sub.33_49AA (b) (SEQ ID NO: 66), YydF.sub.33_49VV (c) (SEQ ID NO: 67), YydF.sub.33_49II (d) (SEQ ID NO: 68), Peptide-SA1 (e) (SEQ ID NO: 62), Peptide-SA2 (f) (SEQ ID NO: 63), Peptide-SE (g) (SEQ ID NO: 64) or Peptide-SP (h) (SEQ ID NO: 65) at 100 μM. In bold, epimerized residues.

(12) FIG. 12—Growth of Streptococcus agalactiae in liquid BHI medium in the presence of YydF.sub.33-49 (a), YydF.sub.33_49AA (b) (SEQ ID NO: 66), YydF.sub.33_49VV (c) (SEQ ID NO: 67), YydF.sub.33_49II (d) (SEQ ID NO: 68), Peptide-SA1 (e) (SEQ Ill NO: 62), Peptide-SA2 (f) (SEQ ID NO: 63), Peptide-SE (g) (SEQ ID NO: 64) or Peptide-SP (h) (SEQ ID NO: 65) at 100 μM. In bold, epimerized residues.

EXAMPLES

Example 1

(13) Here the inventors showed that the common laboratory strain Bacillus subtilis is able to produce a novel type of bioactive peptides containing D-amino acids despite being of ribosomal origin. This peptide is post-translationally modified by a novel enzyme belonging to the superfamily of radical SAM enzymes. They demonstrated that this novel enzyme uses an unprecedented radical-based mechanism to convert L-Isoleucine and L-valine residues into D-allo-lsoleucine and D-valine. They established that this enzyme generates a 5′-deoxyadenosyl radical to catalyze C.sub.α H-atom abstraction leading to the formation of a carbon-centered radical. Mutagenesis experiments support that this enzyme possesses two essential [4Fe-4S] centers and allow identifying a critical H-atom donor, required for the termination of the catalytic cycle. Finally, in a unique manner, they discovered that the presence of D-amino acids is required for the activity of this bioactive peptide which likely induces LiaRS, a major component of the bacterial cell wall integrity.

(14) The Lia system of Bacillus subtilis is a cell envelope stress module composed of a two-component system (LiaRS) and an inhibitory protein (LiaF). This genetic system is highly conserved among Firmicutes and part of the complex regulatory network orchestrating the cell wall stress response. Although its regulation has been described in great details, its precise physiological role in B. subtilis is not fully understood. LiaRS is specifically and strongly induced by antibiotics targeting the cell wall such as nisin, vancomycin or bacitracin and has thus been developed as a biosensor and high-throughput screen for cell wall antibiotics. Upon antibiotics sensing, LiaRS transduces cell envelop stress signals activating gene expression presumably to maintain cell wall integrity although it does not confer antibiotic resistance.

(15) In an attempt to identify genes involved in LiaRS regulation, a mutagenesis study was undertaken in B. subtilis and led to the discovery of the yydFGHIJ operon (Butcher et al, 2007, J Bacteriol, 189, 8616). This operon shows positive regulation on LiaRS and possesses all the characteristic features of a genetic system encoding a putative peptide (YydF) modified by a radical SAM enzyme and a protease (YydG and YydH respectively) then finally exported in the extracellular medium by an ABC-type transporter (YydIJ) even though none of these components were ever isolated or investigated (FIG. 1A).

(16) Radical SAM enzymes are an emerging family of enzymes catalyzing a large diversity of protein and peptide modifications such as oxidation, unusual methyl transfer reaction or thioether bond formation. They have emerged as major players for the biosynthesis of the so called RiPPs (Ribosomally synthesized and post-translationally modified peptides) being involved in chemically challenging reactions, that no other enzymes are able to perform. To investigate the biological role and the catalytic function of the putative radical SAM enzyme YydG, the inventors over-expressed the protein in E. coli and assayed its activity against the YydF peptide. The purified protein (FIG. 1B) had the distinctive spectroscopic properties of radical SAM enzyme with charge transfer absorption band at 320 and 420 nm (FIG. 1C). Based on a [4Fe-4S].sup.2+ cluster molar extinction coefficient of ∈.sub.410˜15,000 M.sup.−1 cm.sup.−1, the enzyme appears to possess one to two [4Fe-4S] centers after anaerobic reconstitution (FIG. 1C).

(17) Genome mining revealed that yydF and yydG are also present in several Gram-positive pathogens such as Enterococcus faecalis and several Streptococci and Staphylococci including S. agalactiae and S. epidermidis. Sequence alignment of the YydF homologs indicated a putative leader-sequence located in the N-terminus part and a highly conserved motif from the residue 17 to the end of the peptide (FIG. 1D). The inventors assayed the reconstituted enzyme either with the full-length YydF peptide or a truncated form, encompassing the conserved amino acid residues from position 18 to 49 that we called YydF.sub.18-49. As shown (FIG. 1E), under anaerobic and reducing conditions, YydG produced the expected 5′-deoxyadenosine (5′-dA; eluting at 12.3 min) resulting from the S-adenosyl-L-methionine (SAM) homolytic cleavage and also three peptide derivatives: YydF.sub.a, YydF.sub.b & YydF.sub.c eluting at 46, 47.6 & 48 min (FIGS. 1E, F, G and H). The formation of the three peptides was strictly dependent of the presence of sodium dithionite as one-electron donor and similar products were obtained using YydF or YydF.sub.18-49 (FIGS. 1F, G and H). Kinetic analysis of the reaction showed that YydG produced one mole of 5′-dA per mole of modified product and catalyzed several turnovers under in vitro conditions, although uncoupled SAM cleavage occurred as the reaction proceed (FIG. 1G). These results demonstrated that, in vitro, YydG used SAM to modify YydF through a radical-based mechanism. Since YydF.sub.18-49 proved to be a better substrate and was easier to characterize, we decided to use it to identify the modification catalyzed by YydG.

(18) Mass spectrometry inspection of the three peptides formed revealed no mass difference compared with the substrate (YydF.sub.18-49[M+3H].sup.+=1258.92). This was in contrast with all known rSAM enzymes catalyzing peptide or protein post-translational modifications such as such as cross-linking, oxidation or methylation (7, 9, 14-16). Tryptic peptide mapping of the substrate gave three peptides (Peptide 1 [M+H].sup.+=931.48, Peptide 2 [M+H].sup.2+=592.31 and Peptide 3: [M+H].sup.+=768.41) eluting at 22, 27 and 19.2 min) as shown on FIG. 2A. Comparison with the enzymatically modified peptide showed the appearance of two new peptides (i.e. Peptide 2* and Peptide 3*) having the same molecular weight as Peptide 2 & 3 but eluting at 26.5 and 23.5 min, respectively (FIG. 2B). This result supported that YydG had introduced two modifications one located internally (in Peptide 2) and one in the C-terminus end of peptide (in Peptide 3). In all the experiments performed, YydFc was the main product. Tryptic peptide mapping revealed that essential the C-terminus end of the peptide was modified since the Peptide 3*/Peptide 3 ratio was 5 times larger than the Peptide 2*/Peptide 2 ratio (FIG. 2B) indicating that YydG had some preferred sites.

(19) To identify the nature and location of the modifications catalyzed by YydG, the inventors repeated the reaction in >90% deuterated buffer since rSAM enzymes are known to abstract and sometimes exchange, H-atoms during catalysis. In deuterated buffer, YydG produced a similar product pattern with YydFc being always the most abundant product (FIG. 2C). Interestingly, LC-MS analysis of the reaction showed that under these conditions, YydF.sub.a and YydF.sub.c had a molecular weight of [M+3H].sup.3+=1259.24 and YydF.sub.b a molecular weight of [M+3H].sup.3+=1259.6. This corresponded to one and two Dalton units more than the substrate, YydF.sub.18-49([M+3H].sup.3+=1258.92), unambiguously demonstrating that one deuterium atom was incorporated into YydF.sub.a, YydF.sub.c and two deuterium atoms in YydF.sub.b, respectively. Tryptic peptide mapping of the reaction allowed to localize deuterium incorporation exclusively in the Peptide 2* and Peptide 3* whose molecular masses shifted from one Dalton unit (i.e. [M+2H].sup.2+=593.04 and [M+2H].sup.+=769.52) (FIG. 2D). LC-MS/MS fragmentation of these two peptides demonstrated deuterium incorporation on Val.sub.36 in Peptide 2* (as shown by the characteristic ions y.sub.1, y.sub.2+1 and b.sub.7, b.sub.8+1) and on Ile.sub.44 in Peptide 3* (as shown by identification of the ions y.sub.5 and y.sub.6+1). Altogether these results demonstrated that YydG catalyzes the replacement of two peptide H-atoms by two solvent exchangeable H-atoms.

(20) To determine the nature of the modification, the inventors performed acid hydrolysis of the peptide and analyzed its amino acid content, after derivatization with N-α-(2,4-dinitro-5-fluorophenyl)-L-valinamide (L-FDVA), by LC-MS. The YydF.sub.18-49 peptide contains two Val and one Ile residues but five Leu residues which not only have the same molecular weight than Ile but also eluted at similar retention times. Optimized LC-MS/MS conditions allowed, as shown on FIG. 3A, the separation and characterization of L-Ile and L-Leu but also of their D-configured counterparts (D-allo-Ile and D-Leu). Analysis of the enzymatically modified peptides clearly showed that, in addition of the identification of L-Ile and L-Leu, another product eluting at 27.7 min was formed corresponding to D-allo-Ile. Similarly, the analyses of Val residues (FIG. 3B) showed the presence not only of L-Val but also of D-Val eluting at 26 min.

(21) The inventors hence established that YydG is a radical SAM epimerase, the first one shown to be active in vitro on a peptide backbone. The enzyme catalyzed the epimerization of up to ˜35% of the Ile and ˜10% of Val residues. Consistent with this conclusion, when one derivatized the amino acids epimerized by incubating YydG in deuterated buffer, their mass analyses revealed a +1 Da increment (FIG. 3C&D), consistent with the analyses performed on the intact peptide and their tryptic derivatives (FIG. 2C&D).

(22) To definitely assert their identity as D-configured amino acids, the inventors synthesized a YydF.sub.18-49 variant peptide containing one D-Val and one D-allo-Ile in positions 36 and 44, respectively. The tryptic peptide mapping and the amino acid profile of this synthetic peptide perfectly reproduced the ones of the enzymatically modified peptide (data not shown).

(23) Based on these analyses, the inventors were able to assign YydF.sub.a as a peptide containing a D-Val in position 36, YydF.sub.c as peptide containing a D-allo-Ile in position 44 and YydF.sub.b as peptide containing a D-Val and a D-allo-Ile in positions in positions 36 & 44, respectively. Hence YydG, produced a mixture of peptides containing either a single or double modified amino acids, with Ile.sub.44 being the favored substrate (FIG. 2C).

(24) During the epimerization reaction performed in deuterium buffer, if the inventors have established that a solvent-exchangeable H-atom is incorporated into in the peptide backbone, the 5′-dA produced contained no significant labeling as shown by LC-M analysis. These results and the kinetic analysis on FIG. 1G are consistent with YydG producing one 5′-dA radical (5′-dA⋅) to abstract the C.sub.α H-atom of Val.sub.36 or Ile.sub.44 with the concomitant formation of one mole of 5′-dA.

(25) The last questions which remained to be solved, was the origin of exchangeable H-atom introduced during catalysis. Indeed, the carbon-centered radical was unlikely to interact with a buffer component as such highly reactive species must be kept sealed in the enzyme active site. The inventors favored a protein amino acid residue as H-atom donor and radical quencher required to terminate the reaction. Close inspection of the YydG sequence pointed out that, in addition to the three cysteine residues from the radical SAM motif, only six cysteines were present in the sequence (FIG. 4A). Interestingly, two cysteine residues (i.e. Cys22 and Cys223) were adjacent to another cysteine residue, one of which being inside the predicted loop containing the rSAM [4Fe-4S] center. The organization of the five other cysteine residues in the C-terminus end of the protein was reminiscent of motifs involved in the coordination of additional [4Fe-4S] centers in rSAM enzymes. To probe their function, the inventors substituted the three cysteine residues of the CxxxCxxC radical SAM motif, Cys22, Cys222 or Cys223 by alanine residues. The four designed mutants (i.e. A3, C22A, C222A and C223A) were successfully purified although the C222A mutant proved to be recalcitrant to purification and produced partly as a truncated form (FIG. 4B). Spectroscopic analysis showed that, based on its UV-visible spectrum, the AxxxAxxA mutant contained ˜1 [4Fe-4S] center while the amount of [4Fe-4S] center was two-times higher in the C22A and the C223A mutants (FIG. 4C). Importantly, the aerobically purified AxxxAxxA mutant already contained high amounts of iron-sulfur center demonstrating that the presence of [4Fe-4S] center in this mutant was independent of the anaerobic reconstitution. The UV-visible spectra of C22A and C223A mutants perfectly superimpose with the wild-type enzyme (FIG. 1B), supporting the fact that YydG likely contains two [4Fe-4S] centers. The C222A mutant appeared to contain no iron-sulfur center, even after anaerobic reconstitution (FIG. 4C). The C222A absorption maximum was shifted toward 250 nm indicating that the protein was likely miss-folded, as it has been repeatedly reported when cysteine residues involved in [4Fe-4S] coordination are mutated in iron-suflur enzymes including rSAM enzymes. It is thus likely than Cys222, is involved in coordination of the second [4Fe-4S] center present in YydG.

(26) The activity of all the mutants was assayed with the YydF.sub.18-49 substrate (FIG. 4D). As expected, the A3 mutant was unable to convert the peptide substrate and to cleave SAM. The C222A mutant was also totally impaired for enzyme activity. In contrasts, the C22A mutation did not affect the activity of the enzyme and the three epimerized peptides were produced (i.e. YydF.sub.a, YydF.sub.b and YydF.sub.c). The C223A mutation did not either prevent the epimerization activity. However, this enzyme variant produced other peptide derivatives eluting at 10 min, 19.8min and 26 min (FIG. 4E). High-resolution mass spectrometry showed these peptides to have a mass shift of −30.005 Da or −1.032 Da, compared to predicted hydrolytic products. They all contained at their C-terminus or N-terminus ends a truncated Val.sub.36 or Ile.sub.44, the targets of the YydG enzyme (FIG. 4E). Their structure was determined as: Ac-GLLDESQKLAKVNDLWYFVKSKENRWI* (SEQ ID No 53) (peptide G.sub.18-I.sub.44*, [M+2H].sup.2+=1646.3834); and I*LGSGH—NH.sub.2 (SEQ ID No 55) (peptide G.sub.18-I.sub.44*, [M+Na].sup.+=603.2851). The truncation was identified as the loss of the amino acid carboxylic or amino group, resulting from the rupture of either the C.sub.α—N or the C.sub.α—CO bonds, and the addition of an oxygen-atom on the amino acid C.sub.α-atom. These results are reminiscent of the substrate fragmentation obtained with another member of the rSAM enzyme family, the pyruvate formate lyase activase, when the reaction was exposed to molecular oxygen. They also definitively established that YydG generates a carbon-centered radical on the C.sub.α-atom of Val.sub.36 and Ile.sub.44 and that Cys223 plays a critical role for the termination of the reaction.

(27) In light of the previous work of the inventors on another rSAM enzyme, the spore photoproduct lyase (SP lyase), the inventors interpreted the role of Cys223 as the critical H-atom donor. Indeed, while investigating a mutant of the SP lyase, they have shown that in the absence of a suitable protein H-atom donor, the substrate radical intermediate can react with adventitious radical scavengers leading to the formation of various adducts. Here, the stabilized C.sub.α radical, in the absence of the thiol group of Cys223, is free to react with molecular oxygen leading to these unique peptidyl backbone breakages.

(28) Finally, since the YydFGHIJ operon (FIG. 1A) was shown to activate the Lia system in B. subtilis, the inventors assayed the activity of the YydF.sub.18-49 peptide before and after enzyme reaction on various bacterial strains. They evidenced strong inhibition growth of B. subtilis with either the enzymatically epimerized peptide or the synthetic peptide containing D-Val.sub.36 and D-allo-Ile.sub.44. In contrast, the unmodified YydF.sub.18-49 peptide was devoid of activity (FIG. 5A). Similarly, in liquid medium, only the epimerized peptide proved to be active as a strong and persistent inhibition could be measured (FIG. 5B). Further studies will be required to decipher the molecular basis of this inhibition but according to the inventors' knowledge, it is the first time that a naturally epimerized peptide proved to be the active form of a regulatory or antimicrobial bacterial peptide.

(29) The present study demonstrates that peptides containing D-amino acids, called herein Epipeptides, are much more common than previously anticipated in living organisms including the common laboratory bacterium Bacillus subtilis but also many pathogenic species such as Streptococcus agalactiae, Enterococcus faecails or Staphylococcus epidermidis. Unexpectedly, the inventors demonstrated here that D-amino acids appear not only to provide resistance to proteases but are directly involved in bacterial response.

Example 2

(30) A gene, yydF, was proposed in the literature to encode a peptide produced by Bacillus subtilis (SEQ ID No 1). However no proof of its actual synthesis or of any post-translational modification has been reported.

(31) After growth of B. subtilis in a synthetic medium (Buffer solution 5× (Na2HPO4: 17 g ; KH2PO4: 7.5 g, NaCl: 1.25 g; NH4Cl: 2.5 g in 500 mL); Trace element solution: MnCl.sub.2: 20 mg; ZnCl.sub.2: 34 mg; CuCl.sub.2: 8.6mg; CoCl.sub.2: 12mg; Na.sub.2MoO.sub.4 : 12 mg; in 200 mL), the inventors successfully purified a peptide, originating from YydF and encompassing residues 33 to 49 (FIG. 7), as established by mass spectrometry analysis (Table 1). The sequence of the isolated peptide was determined to be (SEQ ID No 61):

(32) TABLE-US-00002 Trp Tyr Phe Val Lys Ser Lys Glu Asn Arg Trp Ile 1               5                   10 Leu Gly Ser Gly His         15 SEQ ID No 61: Sequence of the peptide YydF.sub.33_49

(33) In addition, the inventors determined that the peptide YydF.sub.33-49, produced by B. subtilis contained 2 epimerized residues (i.e. D-amino acids) located in position 36 (Val) and 44 (Ile). The peptide was thus called YydF_.sub.33-49DD. Previous work from the inventors (Example 1) has established that the epimerized residues are the result of the conversion of L-amino acid residues by a unique radical SAM enzyme, YydG, which targets the amino acids C.sub.α-atom. Currently, no such short peptides, containing discreet epimerization, are known to be produced by bacteria.

(34) Because the operon YydFGHIJ, was shown to induce the two component system LiaRS, which among other stimuli, sense the bacterial cell-wall integrity, the inventors searched for a putative bacterial growth inhibition triggered by various YydF peptide derivatives. Initial tests were performed with a peptide encompassing residues 18-49 (YydF.sub.18-49, SEQ ID No 20).

(35) As shown (FIG. 5B), only in the presence of the peptide YydF.sub.18-49DD containing two epimerized residues: Val.sub.36 and Ile.sub.44 (numbering according to SEQ ID No 61), the inventors monitored bacterial growth inhibition. The presence of one epimerized residues or the absence of epimerized residues did not significantly impacted bacterial growth. This demonstrated, for the first time, that a short peptide with epimerized residues can inhibit bacterial growth.

(36) Having established that the presence of two key epimerized residues is critical for the inhibitory properties, the inventors synthesized two peptides corresponding to the sequence of the peptide produced by B. subtilis (SEQ ID No 61). These two peptides contained either only L-amino acid residues (YydF.sub.33-49) or the two critical epimerized residues: D-Val.sub.36 and D-Ile.sub.44 (YydF.sub.33-49DD). Only the YYdF.sub.33-49DD peptide proved to inhibit bacterial growth (FIG. 8a). This novel peptide proved to be 100 times more potent than the YydF.sub.1849DD peptide previously assayed (FIG. 8b&c) with a MIC <1 μM. In addition, the inventors showed that these peptides do not only inhibit bacterial growth but they also induce bacterial cell death. Indeed, if after initial growth for 3 hours, the YydF.sub.18-49 peptide is added at mid-exponential phase (FIG. 9), a clear slow down followed by a decrease of the cell density was measured.

(37) Interestingly, homologs of the YydF peptides are predicted in the genome of several Gram-positive bacteria such as: Salinibacillus aidingensis, Bacillus coagulans, Paenibacillus sp and several pathogenic species such as: Enterococcus faecalis, Enterococcus caccae, Streptococcus agalactiae, Staphylococcus pseudintermedius, Staphylococcus equorum, Staphylococcus condimenti and Staphylococcus epidermidis (FIG. 1D, FIG. 6).

(38) In order to determine if these peptides are bioactive, the inventors synthesized a library of peptides based on the sequences identified in the genomes of Streptococcus and Staphylococcus species. They hypothesized that these peptides should contain the same post-translational modifications as the ones identified in B. subtilis, which means a processed peptide of 17 amino acid residues with two D-amino acids in the positions 4 and 12 (SEQ ID Nos 62-65). The epimerized residues are in bold.

(39) TABLE-US-00003 Streptococcus agalactiae Peptide-SA1 (SEQ ID No 62) WYFVRSSKNRWVAGSAH Streptococcus agalactiae Peptide-5A2 (SEQ ID No 63) WYFVRNSKNRWVAGSAH Staphylococcus equorum Peptide-SE (SEQ ID No 64) WYFVKSKQNRWVVGSGH Staphylococcus pseudintermedius Peptide-SP (SEQ ID No 65) WYFVKSQSNRWIVGSGH

(40) In addition, the inventors also synthesized three unnatural peptides derived from the B. subtilis YydF.sub.33_49 sequence (SEQ ID No 61) but for which the two epimerized residues (i.e. Val.sub.36 and Ile.sub.44) were both substituted by Val, Ile or Ala residues, YydF.sub.33_49VV, YydF.sub.33_49II and YydF.sub.33_49AA, respectively (SEQ ID Nos 66-68). The epimerized residues are in bold.

(41) TABLE-US-00004 (SEQ ID No 66) YYd F.sub.33_49AA WYFAKSKENRWALGSGH (SEQ ID No 67) Yyd F.sub.33_49VV WYFVKSKENRWVLGSGH (SEQ ID No 68) Yyd F.sub.33_49II WYFIKSKENRWILGSGH

(42) These 7 peptides (SEQ ID Nos 62-68) were assayed against B. subtilis and the two representative Gram-positive pathogens: S. agalactiae and E. faecalis. As shown, all peptides were effective against B. subtilis including the peptides with unnatural sequences (FIG. 10). E. faecalis growth was significantly delayed with Peptide-SA1 & Peptide-SA2 and growth totally inhibited with Peptide-SE and the unnatural peptide YydF.sub.33_49AA (FIG. 11). S. agalactiae was inhibited by YydF.sub.33-49DD and the derivatives YydF.sub.33-49AA and YydF.sub.33-49VV but not by YydF.sub.33-49II (FIG. 12). The peptides: Peptide-SA1, Peptide-SA2, Peptide-SE and Peptide-SP were all inhibitors.

(43) The inventors thus demonstrated that short peptides containing two D-amino acid residues are a novel class of inhibitory peptides able to inhibit the growth of several Gram-positive bacteria including relevant pathogens. They are efficient whether added at the beginning or after bacterial growth at mid-exponential phase. Finally, some discreet modifications in the sequence are able to tune the inhibition properties and the specificities at the genera and species level of these peptides allowing the development of targeted antibiotics. In addition, based on the framework of 17 amino acids and the conserved location of two D-amino acids (in position 4 and 12) downstream to aromatic residues (W or Y), the inventors also demonstrated that it is possible to design peptides with unnatural sequences that proved to be effective against all the Gram-positive bacteria assayed.

(44) The bioactive peptides proved to have sequence identity varying from 100 to 58.8% relative to the original YydF.sub.33-49 sequence which means at least 7 amino acid residues could be changed without altering their global inhibition properties. It is thus possible to engineer these peptides in an unprecedented manner to target specific bacterial genera and tune their biological properties.

(45) TABLE-US-00005 # Percent Identity Matrix # 1: Peptide-SA1 100.00 94.12 64.71 70.59 58.82 58.82 64.71 70.59 2: Peptide-SA2 94.12 100.00 58.82 64.71 52.94 52.94 58.82 64.71 3: Peptide-SP 64.71 58.82 100.00 82.35 70.59 76.47 82.35 76.47 4: Peptide-SE 70.59 64.71 82.35 100.00 76.47 76.47 82.35 88.24 5: YydF33_49AA 58.82 52.94 70.59 76.47 100.00 88.24 88.24 88.24 6: YydF33_49II 58.82 52.94 76.47 76.47 88.24 100.00 94.12 88.24 7: YydF33_49DD 64.71 58.82 82.35 82.35 88.24 94.12 100.00 94.12 8: YydF33_49VV 70.59 64.71 76.47 88.24 88.24 88.24 94.12 100.00

(46) Materials and Methods

(47) YydG Expression

(48) The yydG genes was synthesized (Life Technologies) and cloned into a pASK plasmid. The plasmid was expressed in E. coli BL21 (DE3) star (Life Technologies) and protein expression was performed in LB medium containing ampicillin (100 μg.mL-1). After overnight growth at 21° C., the cells were collected and disrupted by ultra-sonication in buffer A (Tris 50 mM, KCl 300 mM, Glycerol 10% pH 7.5). The bacterial suspension was centrifuged at 45,000×g for 1.5 hours and the protein supernatant was loaded onto a Streptactin high capacity (IBA GmbH) column previously equilibrated with buffer A. The YydG protein was eluted with 6mL of buffer A containing desthiobiotine (0.6 mg/mL) further concentrated with Amicon concentrator (Millipore) with a molecular cut-off of 10 kDa.

(49) Enzyme Reconstitution

(50) YydG was reconstituted under anaerobic conditions in a Bactron IV anaerobic chamber. The protein was mixed with 3 mM of DTT at 12° C. during 15 minutes then Na.sub.2S and (NH.sub.4).sub.2Fe(SO.sub.4).sub.2 were added and the solution was incubated at 12° C. during 4 h.

(51) Enzyme Assays

(52) YydG was incubated with 3 mM DTT, 1 mM SAM and 1 mM peptide substrate unless otherwise indicated. Incubations were performed at 25° C. under strict anaerobic conditions and 10 μL aliquots sampled overtime.

(53) HPLC Analysis

(54) HPLC analysis was performed on an Agilent 1200 series infinity equipped with a reversed phase column (LiChroSphere 100 RP-18e 5 μm) (Merck Millipore). A gradient from solvent A (H.sub.2O, 0.1% TFA) to B (80% CH.sub.3CN, 19.9% H.sub.2O, 0.1% TFA) was applied as follow: 0-1 min: 100% A/0% B; 1-45 min: a linear gradient with 1% of solvent B per minute at a flow rate of 1 ml.min.sup.−1. Detection was made at 257 & 278 nm with a diode array detector and by fluorescence (excitation at 278 nm and emission at 350 nm).

(55) Liquid Chromatography—Mass Spectrometry/Mass Spectrometry Analysis

(56) High resolution liquid chromatography—mass spectrometry/mass spectrometry analysis were performed using an LTQ-Orbitrap Discovery mass spectrometer (ThermoFisher) with a nanoelectrospray ion source and an Ultimate 3000 LC system (Dionex). A LTQ mass spectrometer (ThermoFisher) with a nanoelectrospray ion source was used for routine analysis. Peptide analysis was performed on a nanocolumn Pepmap 100 C18 (0.075 by 15 cm, 100 Å, 3 μm).

(57) Inhibition Assay on Solid Medium

(58) An overnight culture of the bacterial strain to be assayed was freshly inoculated to sterile BHI liquid medium. After 4 hours of bacterial growth at 37° C., the medium was diluted to 1/1000 and inoculated into a soft agarose medium pre-heated at 42° C. The agarose medium containing bacteria was overlaid on a previously jellified sterile BHI agarose layer. 200 μg of peptide was spotted onto the plate and bacterial growth proceeded at 37° C.

(59) Inhibition Assay on Liquid Medium

(60) An overnight culture of the bacterial strain to be assayed was freshly inoculated to sterile LB liquid medium. After 4 hours of bacterial growth at 37° C., the medium was diluted to 1/10,000 and inoculated into sterile liquid LB or BHI medium. Peptide solution was added ( 1/100) to a final concentration ranging from 0.01 to 100 μM and OD at 600 nm was recorded continuously using a Tecan microplate reader (Infinite® 200 PRO series).

(61) TABLE-US-00006 TABLE 1 Mass fragments for peptide YydF.sub.33-49 isolated from B. subtilis Sequence b+ b++ y+ y++ W 1 187.08718 94.04753 2107.08779 1054.04783 17 Y 2 350.15051 175.57919 1921.00848 961.00818 16 F 3 497.21892 249.11340 1757.94515 879.47651 15 V 4 596.28734 298.64760 1610.87674 805.94230 14 K 5 724.38230 362.69509 1511.80833 756.40810 13 S 6 811.41433 406.21110 1383.71336 692.36062 12 K 7 939.50929 470.25858 1296.68133 648.84460 11 E 8 1068.55188 534.77988 1168.58637 584.79712 10 N 9 1182.59481 591.80134 1039.54378 520.27582 9 R 10 1338.69592 669.85190 925.50035 463.25436 8 W 11 1524.77523 762.89155 769.39974 385.20381 7 I 12 1637.85930 819.43358 583.32043 292.16415 6 L 13 1750.94336 875.97562 470.23637 235.62212 5 G 14 1807.96483 904.48635 357.15230 179.08009 4 S 15 1894.99685 948.00236 300.13084 150.56935 3 G 16 1952.01832 976.51309 213.09881 107.05334 2 H 17 2089.07723 1045.04255 156.07735 78.54261 1 (M) 2106.07997 (M + H).sup.+ 2107.08779 (M + 2H).sup.2+ 1054.04783 (M + 3H).sup.3+ 703.03451 (M + 4H).sup.4+ 527.52785