Identification and Use of Non-Peptide Analogs of RNAIII-Inhibiting Peptide for the Treatment of Staphylococcal Infections

Abstract

RNAIII inhibiting peptide (RIP) has been established as an effective inhibitor of staphylococcal infections. Non-peptide small molecule analogs based on a pharmacophore conforming to the putative atomic structure of RIP, can be identified by computer screening of that pharmacophore against established libraries of known small molecules other than peptides. One such identified structural analog is hamamelitannin. When tested for effective inhibition of staphylococcal infections, hamamelitannin demonstrated inhibition similar to that exhibited by RIP. Other analogs can be identified and similarly used.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will be described in conjunction with the accompanying drawings, in which:

FIG. 1 is a view of molecular model of RIP prepared on the basis of homology with the crystal structure of a related protein, ribosomal protein L2 within the crystal structure of the 50S ribosomal subunit from the bacterium Deinococcus radiodurans (Protein Data Bank code 1NKW);

FIG. 2 is a representation of the pharmacophore used in the invention to identify potential nonpeptide inhibitory compounds with distances presented in Å units;

FIG. 3 is a Representation of the Chemical Structure of Hamamelitannin.

FIG. 4 is a graphical presentation of Hamamelitannin inhibition of RNAIII production: 2×10⁷ early exponential S. aureus cells containing rnaiii::blaZ fusion construct were grown for 2.5 hrs with increasing amounts of hamamelitannin or RIP. RNAIII levels were determined as lactamase activity (reporter gene product) and denoted as V_max.

FIG. 5 demonstrates graphically the inhibition of S. aureus attachment in vitro by hamamelitannin: S. aureus cells were placed in polystyrene plates and incubated with increasing amounts of hamamelitannin or RIP for 3 hrs at 37° C. without shaking. Attached cells were stained and OD_{595 nm} was determined.

FIG. 6 is a graph presentation of the effectiveness of bacterial graft-associated infection inhibition by hamamelitannin and RIP against challenge by methicillin resistant S. aureus and S. epidermidis (MRSA and MRSE) using pre-incubation.

FIG. 7 is a graph presentation of the effectiveness of bacterial graft-associated infection inhibition by hamamelitannin against challenge by MRSA and MRSE using local treatment (graft soak);

DETAILED DESCRIPTION

Model Building of the RIP Peptide

Short peptides such as RIP do not have a fixed conformation in solution. However, the active conformation of RIP can be deduced from the corresponding sequence segment in RAP since RIP competes with RAP for binding to the same receptor. This suggests that RIP is structurally similar to a segment of RAP and that probably RAP acts as an agonist and RIP as an antagonist of RAP. The sequence of RIP (YSPWTNF) is similar to the sequence of residues 4-9 of RAP (YKPITN). Consequently, it is reasonable to assume that the structure of RIP is very similar to the corresponding segment in RAP, if not entirely identical. Therefore, it would be best to build a model of RIP based on the corresponding segment in the RAP structure. However, a crystal structure or a solution NMR structure of RAP is not available. Fortunately, there is another way to build a model of the active conformation of RIP. A crystal structure of a protein similar to RAP is available. This is the crystal structure of ribosomal protein L2 from Deinococcus radiodurans (PDB code 1 NKW). The structure of protein L2 from S. aureus is expected to be very similar to the structure of L2 from Dienococcus radiodurans because L2 is a highly conserved protein. A model of RIP was built based on the crystal structure of ribosomal protein L2 from Dienococcus radiodurans (PDB code 1 NKW). This homology-built model of RIP was subjected to energy minimization with program CNS (FIG. 1).

Amino acid residues 1, 3, and 5 in RIP are entirely conserved, and in residues 2 and 4 the sequence differences are conservative, i.e. an arginine instead of a serine and a tyrosine instead of a tryptophan. RIP homologs with such conservative amino acid replacements in these positions are known to retain their inhibitory activity.

In Silico Screening for RIP Analogs

Screening for small molecule nonpeptide analogs of RIP was carried out by a computer search with the ISIS software (Integrated Scientific Information System) from Elsevier MDL against the Available Chemicals Database (ACD), a library of 300,000 commercially available small molecule compounds. The model of RIP served as the basis for the search. Our first approach was to carry out similarity searches with the RIP models against the ACD. As this search yielded only peptides it was abandoned. Next we turned to a search of the ACD based on a pharmacophore (FIG. 2).

Definition of a Pharmacophore for a Nonpeptide RIP Analog

The basis for the pharmacophore design was the RIP model. The pharmacophore was defined in terms of distances in the RIP model between aromatic moieties, distances between aromatic moieties and hydrogen donors or acceptors and distances between pairs of hydrogen bond donors/acceptors. Different pharmacophores were used in the search for a suitable RIP analog. FIG. 2 shows the pharmacophore that led to the identification of hamamelitannin as an inhibitor of staphylococcal infections. Distances are indicated in units of Angstroms (Å). This pharmacophore calls for distances of 13-15 Å between the centers of two aromatic rings corresponding to Tyr 1 and Phe 7; distances of 7.5-9.5 Å between the hydroxyl group of Tyr 1 and a hydrogen donor or acceptor corresponding to Ser 2 or Asn 2, consisting of a nitrogen or oxygen atom; distances of 7.5-9.5 Å between the center of the aromatic ring corresponding to Phe 7 and a hydrogen donor or acceptor corresponding to Asn 6, consisting of a nitrogen or oxygen atom; distances of 2.5-4.5 Å between the center of an aromatic ring corresponding to Phe 7 and a hydrogen donor or acceptor corresponding to Thr 5, consisting of a nitrogen or oxygen atom.

Hamamelitannin Inhibits RNAIII Production In Vitro.

To test if hamamelitannin is a quorum sensing inhibitor and thus suppresses RNAIII synthesis, 2×10⁷ cells were incubated with increasing amounts (0-50 μg) of hamamelitannin (or RIP). RNAIII levels were measured as β-lactamase activity as a reporter gene product by the addition of nitrocefin as substrate. As shown in FIG. 4, hamamelitannin inhibits RNAIII synthesis in a dose dependent manner, and is most effective at doses>2 μg/2×10⁷ bacteria (2 nanomoles/10⁷ bacteria or 0.2 picomoles/10³ bacteria). RIP also inhibited RNAIII production in a dose dependent manner, and was most effective in doses of 8 μg/2×10⁷ bacteria (0.5 picomoles/10³ bacteria). Hamamelitannin (2′,5-di-O-galloyl hamamelose) was purchased from Chromadex, Inc. (93% purity, as assessed by HPLC by the manufacturer). Hamamelitannin was dissolved in water at 25 mg/ml and stored at −70° C. until use.

Hamamelitannin Inhibits Cell Attachment In Vitro

To test for the effect of hamamelitannin on bacterial attachment in vitro, S. aureus cells were incubated with 0-50 μg hamamelitannin or RIP in polystyrene plates for 3 hrs at 37° C. Adherent bacteria were stained with methylene blue and OD determined at 595 nm. As shown in FIG. 5, hamamelitannin (or RIP) reduced cell attachment in a dose-dependent manner and were most effective when 2×10⁷ bacteria were grown in the presence of at least 12.5 μg hamamelitannin or RIP (0.4 picomoles/10³ bacteria). Similar results were obtained with S. epidermidis.

Coating with Hamamelitannin Prevents Device-Associated Infections In Vivo in a Rat Model

To measure the amount of hamamelitannin necessary to prevent device-associated infections, bacteria (2×10⁷ MRSA or MRSE) were pre-incubated with increasing amounts of hamamelitannin for 30 min at room temperature. Grafts were implanted and rats were challenged with the pre-incubated bacteria. Seven days later the graft was removed and bacterial load determined. As shown in FIG. 6, bacterial load on the graft decreased with increasing dose of hamamelitannin while bacterial load in the control untreated group was ˜10⁷ CFU/ml. No bacteria was found when either MRSA or MRSE were pre-incubated with >20 μg hamamelitannin.

In the testing reflected in FIG. 6 and summarized above, to ascertain the amount of hamamelitannin/bacteria necessary to potentially prevent device-associated infections, bacteria (2×10⁷ MRSE or MRSA in 150 μl saline) were pre-incubated with hamamelitannin (0, 0.5, 10, 20, 30, 50 μg) for 30 min at room temperature, as noted. Grafts were implanted and rats were challenged with the pre-incubated bacteria. Seven days later the graft was removed and bacterial load determined. As shown in FIG. 6, all 10 control rats included in the untreated control group (without pre-treatment) demonstrated evidence of graft infection, with quantitative culture results showing 5.9×10⁶±1.9×10⁶ colony-forming units (CFU)/ml MRSE. When the bacteria was treated with 0.5, 10, 20, 30 or 50 μg hamamelitannin, MRSE load was 8.4×10³±2.2×10³ for 0.5 ug and 8.7×10¹±6.6×10¹ for 10 μg. No sign of infection was found when bacteria were pre-incubated with 20, 30 or 50 μg hamamelitannin. Similar results were obtained for MRSA, where untreated control group (without pre-treatment) demonstrated evidence of graft infection, with quantitative culture results showing 7.6×10⁷±1.9×10⁷ CFU/ml MRSA, while those treated with 0.5 or 10 μg hamamelitannin only showed 8.8×10³±2.3×10³ and 3.6 10²±0.8×10² CFU/mL, respectively. No sign of infection was detected if bacteria were pre-incubated with 20, 30 or 50 μg hamamelitannin. These results indicate that 10 μg hamamelitannin could prevent an infection caused by 2×10⁷ CFU MRSE or MRSA, which is what was found for RIP.

To test if hamamelitannin could inhibit graft-associated infection, grafts (1 cm² collagen coated Dacron) were soaked for 1 h in increasing hamamelitannin concentrations. The graft was then implanted into the animal, and bacteria injected onto the graft. Seven days later the graft was removed and bacteria on the graft counted. All rats included in the untreated control groups demonstrated evidence of graft infections, with quantitative culture results showing 7.0×10⁶±1.7×10⁶ CFU/ml MRSE and 6.8×10⁷±1.5×10⁷ CFU/ml MRSA. Significant (p<0.05) decrease in bacterial load was found when the grafts were presoaked with hamamelitannin (0-50 mg/L). As shown in FIG. 7, animals challenged with MRSE and grafts pre-soaked with 0.5 mg/L hamamelitannin had 9.0×10⁴±2.6×10⁴ CFU/mL, those soaked with 10 mg/L hamamelitannin had 2.1×10³±0.2×10³ CFU/mL, those soaked with 20 mg/L had 3.7×10¹±1.1×10¹ CFU/mL, and those soaked with 30 or 50 mg/L had no sign of bacterial load and thus had <10 CFU/mL. Very similar results were obtained when animals were challenged with MRSA. Specifically, as shown in FIG. 7, grafts soaked with 0.5, 10, 20, 30 or 50 mg/L hamamelitannin had bacterial load of 9.3×10⁴±2.8×10⁴, 2.6×10³±0.3×10³, 3.7×10¹±1.1×10¹ CFU/mL, respectively and no bacterial load was detected when the graft was pre-soaked in 30 or 50 mg/L hamamelitannin.

As shown, the analog identification method of the invention presents a powerful new tool for identifying agents for the treatment and prevention of virulent staphylococcal infections. RIP has been previously shown to act as a powerful agent suppressing virulence in staphylococcal infections, likely through suppression of normal responses to quorum sensing. Hamamelitannin exhibits similar inhibitory activity. Similar mechanisms may well be in play, in light of the inhibition of RNAIII production shown.

Hamamelitannin (2,5-di-O-galloyl-hamamelose) is an ester of hamamelose (2-hydroxymethyl-D-ribose) with two molecules of gallic acid (FIG. 3). Since gallic acid contains three phenolic functional groups, hamamelitannin is considered a polyphenol. It belongs to the family of tannins, which are plant polyphenols that are used in tanning animal hides into leather.

Hamamelitannin is a natural product found in the bark and the leaves of Hamamelis virginiana (witch hazel), a deciduous shrub native to damp woods in eastern North America and Canada. The concentration of hamamelitannin in the bark and the leaves is 5 and less than 0.04% (w/w), respectively. Witch hazel extracts were used by Native Americans for pain relief, colds and fever, and they are currently used in skin care products and in dermatological treatment of sun burn, irritated skin, atopic eczema as well as to promote wound healing via anti-inflammatory effects. Hamamelitannin also was shown to inhibit tumor necrosis factor α-mediated endothelial cell death at concentrations less than 100 uM. Hamamelitannin, at a minimum concentration of 50 uM, was also found to have a high protective activity against cell damage induced by peroxides or UVB radiation. In addition, some antibacterial properties of witch hazel have been reported, where aqueous extracts of the bark or the leaves inhibited the growth of E. coli, S. aureus, B. subtilis and E. faecalis. In contrast, we have determined that hamamelitannin has no effect on bacterial growth in vitro even at concentrations as high as 2.5 mM per 1000 bacteria, 13,000 times the MIC (minimum inhibitory concentration) of ampicillin to the same S. aureus strain (0.2 uM per 1000 bacteria).

Hamamelitannin and RIP inhibit RNAIII production at minimal doses of 0.2 and 0.5 picomoles per 1000 bacteria, respectively. Hamamelitannin and RIP also inhibit cell attachment in vitro, both at doses equal or greater than 0.4 picomoles per 1000 bacteria.

Hamamelitannin presents a non-peptide small molecule alternative to RIP as an excellent inhibitor of device-associated infections in vivo, in line with its inhibitory effect on RNAIII and cell attachment in vitro. Inhibition of infection is concentration dependent. Grafts pre-soaked with at least 30 mg/L (˜60 uM) hamamelitannin, show no signs of infection even though the animals were challenged with a high bacterial load of 2×10⁷ CFU. These results are similar to those observed with RIP. Device-associated infections are prevented by merely soaking a graft in the hamamelitannin solutions, suggesting that hamamelitannin can be used to coat medical devices to prevent staphylococcal infections, including those caused by drug resistant strains MRSA and MRSE.

Although the present invention has been fully described in conjunction with several embodiments thereof with reference to the accompanying drawings, it is to be understood that various changes and modifications may be apparent to those skilled in the art. Such changes and modifications are to be understood as included within the scope of the present invention as defined by the appended claims, unless specifically and expressly excluded thereby.

Claims

1. A therapeutic composition comprising a non-peptide small molecule analog of RNAIII-Inhibiting peptide (RIP) which inhibits Staphylococcus aureus infections in a mammal, wherein said analog exhibits a pharmacophore of RIP as reflected in the distances between aromatic moieties in said analog, distances between aromatic moieties and hydrogen bond donors, distances between aromatic moieties and hydrogen bond acceptors, distances between pairs of hydrogen bond donors and distances between hydrogen bond acceptors, wherein said analog is present in an amount effective to inhibit said Staphylococcus aureus infection in said mammal, said composition further comprising a pharmaceutically acceptable carrier.

2. The composition of claim 1, wherein said pharmacophore comprises the restraints reflected in FIG. 2.

3. The composition of claim 2, wherein said analog conforms to the pharmacophore of hamamelitannin.

4. The composition of claim 3, wherein said analog is hamamelitannin.

5. The composition of claim 1, with the proviso that the analog is not hamamelitannin.

6. A method of inhibiting staphylococcal infection in a mammal, comprising administering, to a source of said staphylococcal infections, an amount of the composition of claim 1 effective to inhibit said infection.

7. The method of claim 6, wherein said composition is the composition of claim 3.

8. The method of claim 6, wherein said composition comprises, as a RIP analog, hamamelitannin.

9. The method of claim 6, wherein said administering comprises applying said composition to an exposed surface of a device to be inserted into said mammal's body.

10. The method of claim 9, wherein said device is selected from the group consisting of an implantable device, a catheter, a tampon and a bandage.

11. A method of identifying a potential non-peptide small molecule analog of RIP, which inhibits staphylococcal infections in a mammal, comprising: preparing an atomic model of RIP by identifying a pharmacophore reflecting the distances between aromatic moieties in said analog, distances between aromatic moieties and hydrogen bond donors, distances between aromatic moieties and hydrogen bond acceptors, distances between pairs of hydrogen bond donors and distances between hydrogen bond acceptors;Screening said pharmacophore against a library of atomic models of known small molecule non-peptide compounds to identify compounds conforming to the pharmacophore of said model; andtesting any said compounds so identified to determine whether the compounds so identified inhibit the growth of S. aureus or S. epidermidis wherein said tested compounds which inhibit said growth are potential non-peptide small molecule analogs of RIP.