Chimeric endolysin polypeptide

Abstract

The invention relates to the field of medicine, specifically to the field of treatment of conditions associated with Staphylococcus infection. The invention relates to a novel endolysin polypeptide specifically targeting a bacterial Staphylococcus cell. The invention further relates to said endolysin polypeptide for medical use, preferably for treating an individual suffering from a condition associated with Staphylococcus infection.

Claims

1. An endolysin polypeptide that has lytic activity for Staphylococcus, said endolysin polypeptide comprising a polypeptide, wherein the amino acid sequence of the polypeptide has at least 98% sequence identity with SEQ ID NO: 1 wherein the endolysin polypeptide has enhanced lytic activity for Staphylococcus in human serum compared to: the endolysin with the amino acid sequence as set forward in SEQ ID NO: 2, and/or the endolysin with the amino acid sequence as set forward in SEQ ID NO: 3; and wherein the M23 endopeptidase domain and the CHAP domain in the endolysin polypeptide are separated by a linker that comprises at least 13 amino acids.

2. An endolysin polypeptide according to claim 1, wherein the amino acid sequence of the endolysin polypeptide has at least 98% sequence identity with SEQ ID NO: 1.

3.-8. (canceled)

9. A composition comprising the endolysin polypeptide according to claim 1.

10. A composition according to claim 3, further comprising an excipient acceptable for cosmetics.

11. A composition according to claim 3, further comprising a pharmaceutically acceptable excipient.

12. A composition according to claim 9, further comprising an additional active ingredient.

13. (canceled)

14. A method of treatment of a condition associated with infection with a Staphylococcus, comprising administration an endolysin polypeptide according to claim 1.

15. (canceled)

16. The method of treatment according to claim 14, wherein the condition to be treated is selected from: a skin infection, soft tissue infections such as infected diabetic foot ulcers, mastitis, pneumonia, meningitis, endocarditis, Toxic Shock Syndrome (TSS), sepsis, septicaemia, bacteraemia, or osteomyelitis.

17. The method of treatment according to claim 16, wherein the skin infection is selected from the group consisting of pimples, impetigo, boils, furuncles, cellulitis, folliculitis, carbuncles, scaled skin syndrome, atopic dermatitis, and abscesses.

18. The method of treatment according to claim 14, wherein the condition to be treated is selected from the group consisting of bacteraemia, infective endocarditis, prosthetic joint infection, osteomyelitis, indwelling medical device infection and implanted medical device infection.

19. The endolysin polypeptide comprising a polypeptide according to claim 1, wherein the polypeptide has 100% sequence identity with SEQ ID NO: 1.

Description

Figure Legends

[0064] FIG. 1. Killing of S. aureus NR-46543/USA300 JE2 (MRSA) in human serum by peptidoglycan hydrolases.

[0065] Quantitative killing assays (QKAs) were performed for peptidoglycan hydrolase (PGH) constructs ID453, ID557, ID558 in human serum against methicillin-resistant Staphylococcus aureus (MRSA) NR-46543/USA300 JE2 at 37? C. for 30 min at 180 rpm (25 mm orbit). Viable cell counts (CFUs/ml) (y-axis) of S. aureus NR-46543/USA300 JE2 are shown for three PGH constructs in a 2-fold concentration range (x-axis) in nM (320-0.625 nM). The depicted negative control (no addition of PGH) corresponds to the averaged individual controls of the three tested constructs. The limit of detection (LoD) is 200 CFUs/ml (grey dashed line). The y-axis is cut at 160 CFU's/ml. Three biological replicates (n=3) were made. Error bars of averaged controls and constructs represent the standard error of the mean (S.E.M).

[0066] FIG. 2. Killing of S. aureus ATCC 12600 (MSSA) in human serum by peptidoglycan hydrolases.

[0067] Quantitative killing assays (QKAs) were performed for peptidoglycan hydrolase (PGH) constructs ID453, ID557, ID558 in human serum against methicillin-sensitive Staphylococcus aureus (MSSA) ATCC 12600 at 37? C. for 30 min at 180 rpm (25 mm orbit). Viable cell counts (CFUs/ml) (y-axis) of S. aureus ATCC 12600 are shown for three PGH constructs in a 2-fold concentration range (x-axis) in nM (320 nM-0.625 nM). The depicted negative control (no addition of PGH) corresponds to the averaged individual controls of the three tested constructs. The limit of detection (LoD) is 200 CFUs/ml (grey dashed line). The y-axis is cut at 160 CFU's/ml. Three biological replicates (n=3) were made. Error bars of averaged controls and constructs represent the standard error of the mean (S.E.M).

[0068] FIG. 3. Killing of S. aureus ATCC 12598 (MSSA) in human serum by peptidoglycan hydrolases.

[0069] Quantitative killing assays (QKAs) were performed for peptidoglycan hydrolase (PGH) constructs ID453, ID557, ID558 in human serum against methicillin-sensitive Staphylococcus aureus (MSSA) ATCC 12598 (Cowan 1) at 37? C. for 30 min at 180 rpm (25 mm orbit). Viable cell counts (CFUs/ml) (y-axis) of S. aureus ATCC 12598 are shown for three PGH constructs in a 2-fold concentration range (x-axis) in nM (320 nM-0.625 nM). The depicted negative control (no addition of PGH) corresponds to the averaged individual controls of the three tested constructs. The limit of detection (LoD) is 200 CFUs/ml (grey dashed line). The y-axis is cut at 160 CFU's/ml. Three biological replicates (n=3) were made. Error bars of averaged controls and constructs represent the standard error of the mean (S.E.M).

[0070] FIG. 4. Killing of S. epidermidis ATCC 12228/DSM1798 (MSSE) in human serum by peptidoglycan hydrolases.

[0071] Quantitative killing assays (QKAs) were performed for peptidoglycan hydrolase (PGH) constructs ID453, ID557, ID558 in human serum against methicillin-sensitive Staphylococcus epidermidis (MSSE) ATCC 12228/DSM1798 at 37? C. for 30 min at 180 rpm (25 mm orbit). Viable cell counts (CFUs/ml) (y-axis) of S. epidermidis ATCC 12228/DSM1798 are shown for three PGH constructs in a 2-fold concentration range (x-axis) in nM (320 nM-0.625 nM). The depicted negative control (no addition of PGH) corresponds to the averaged individual controls of the three tested constructs. The limit of detection (LoD) is 200 CFUs/ml (grey dashed line). The y-axis is cut at 160 CFU's/ml. Three biological replicates (n=3) were made, except for concentrations 0.625 and 1.25 nM (n=2). Error bars of averaged controls and constructs represent the standard error of the mean (S.E.M).

[0072] FIG. 5. Killing of S. epidermidis ATCC 35984/BB1336 (MRSE) in human serum by peptidoglycan hydrolases.

[0073] Quantitative killing assays (QKAs) were performed for peptidoglycan hydrolase (PGH) constructs ID453, ID557, ID558 in human serum against methicillin-resistant (MRSE) ATCC 35984/BB1336 at 37? C. for 30 min at 180 rpm (25 mm orbit). Viable cell counts (CFUs/ml) (y-axis) of S. epidermidis ATCC 35984/BB1336 are shown for three PGH constructs in a 2-fold concentration range (x-axis) in nM (320 nM-0.625 nM). The depicted negative control (no addition of PGH) corresponds to the averaged individual controls of the three tested constructs. The limit of detection (LoD) is 200 CFUs/ml (grey dashed line). The y-axis is cut at 160 CFU's/ml. Three biological replicates (n=3) were made, except for concentrations 0.625 and 1.25 nM (n=2). Error bars of averaged controls and constructs represent the standard error of the mean (S.E.M).

EXAMPLES

1. Introduction

[0074] Phage lysins are phage-encoded cell wall hydrolases that cleave bonds within the bacterial peptidoglycan, leading to localized dissolution and eventually cell death.

[0075] In short, these peptidoglycan hydrolases (PGHs) are modular with one domain being responsible for specific binding to target cell walls and other parts being responsible for cleaving very specific bonds within the peptidoglycan. The modules may be connected by polypeptide linkers. These enzymes display various degrees of specificity for binding to the target and for cleaving certain bonds that may be absent in other bacterial genera or species. By combining and rearranging different parts form various naturally occurring enzymes, novel molecules with very specific recognition and cleavage patterns may be created allowing highly targeted antimicrobial preparations to be made.

[0076] Here we present the efficacy profiles in human serum of three such molecules (Table 1). Two of those molecules were already published in WO2021213898 (ID557 [SEQ IDNO: 41] and ID558 [SEQ ID NO: 5]), whereas ID453 is a novel molecule presented herein. All three molecules include the CHAP domain of phage Twort, followed by the M23LST domain of the lysostaphin gene of Staphylococcus simulans. Both ID453 and ID557 contain the SH3b domain of the lysostaphin gene whereas ID558 contains a SH3b domain from phage 2638A. The domain and linker origin references are shown in Table 2.

TABLE-US-00003 TABLE 1 Overview of constructs, which were compared in the experiments. ID Construct SEQ ID ID453 CHAPTw(L)_M23LST_SH3bLST SEQ ID NO: 1 ID557 CHAPTw_M23LST_SH3bLST SEQ ID NO: 2 ID558 CHAPTw(L)_M23LST_(L)SH3b2638 SEQ ID NO: 3

[0077] The IDs are listed with the corresponding construct names and domain architecture. ID557 and ID558 are disclosed in WO2021213898 and are listed with the corresponding sequence IDs herein.

TABLE-US-00004 TABLE2 Domainsusedfortheengineeringoftheconstructsandtheirorigins. Domain Organism/Origin SEQIDNO: CHAPTw BacteriophageTwort,endolysingene SEQIDNO:4 M23LST S.simulansbiovarstaphylolyticus, SEQIDNO:5 lysostaphingene SH3bLST S.simulansbiovarstaphylolyticus, SEQIDNO:6 lysostaphingene CH3b2638 Bacteriophage2638A,endolysingene SEQIDNO:7 TwLinker BacteriophageTwort,endolysingene SEQIDNO:8 (SVKKKDTKKKPKPSNRDGINKDK) LSTLinker(KAGGTVTPTPNTG) S.simulansbiovarstaphylolyticus, SEQIDNO:9 lysostaphingene Ale1Linker(SNSTSSSNNNG) S.capitisEPK1,ale-1gene SEQIDNO:10 2638ALinker Bacteriophage2638A,endolysingene SEQIDNO:11 (GKLEVSKAATIKQSDVKQEVKKQEAKQIVKATD)

[0078] The different domains are listed with the organisms they originate from.

[0079] When targeting SBIs, sufficient activity in human serum of candidate molecules is a prerequisite. In this setting efficacy against three S. aureus strains including 1 methicillin resistant strain and against one methicillin resistant and one methicillin sensitive S. epidermidis strains was investigated. Interestingly, despite significant similarity of the three molecules, one of the candidate compounds showed clear superiority over the other two molecules. ID453 performed better than the others, requiring a lower dose to achieve complete killing of target cells in human serum for two of the three S. aureus strains tested and also for one of the two S. epidermidis strains investigated. It was shown to be at least as effective as either of the other compounds on the remaining strains, with a higher killing potency at lower concentrations as compared to the other constructs. For targeting SBIs, ID453 is a superior option.

2. Results

2.1 Quantitative Killing Assays

[0080] Quantitative killing assays (QKAs) were performed at 37? C. and 180 rpm (25 mm orbit) for 30 min in human serum with the peptidoglycan hydrolase (PGH) constructs CHAPTw(L)_M23LST_SH3bLST (ID453), CHAPTw_M23LST_SH3bLST (ID557) and CHAPTw(L)_M23LST_(L)SH3b2638 (ID558) against S. aureus NR-46543/USA300 JE2, S. aureus ATCC 12600, S. aureus ATCC 12598 (Cowan I), S. epidermidis ATCC 12228/DSM1798 and S. epidermidis ATCC 35984/BB1336. The constructs were tested in a 2-fold concentration range (320 nM-0.625 nM). The evaluated results of the QKAs (viable cell counts) of the three constructs in human serum are shown in FIGS. 1-5. Three biological replicates were performed for each construct in each strain. As shown in FIGS. 1-5, all three constructs have staphylolytic activity in each strain within the tested concentration range. Furthermore, a dose-response is apparent. The further to the left a curve is positioned, the more active a construct is.

[0081] ID453 is the construct with the highest killing activity against the methicillin resistant S. aureus NR-46543/USA300 JE2 (FIG. 1) and the methicillin sensitive S. aureus ATCC 12598 (Cowan I) (FIG. 3) strains, followed by ID557. Although the difference in activity is not as pronounced between these two constructs against the methicillin sensitive S. aureus ATCC 12600 (FIG. 2) strain, ID453 still showed higher activity than the other constructs at most of the tested concentrations. In all three strains, ID558 showed the lowest activity.

[0082] The superiority of ID453 is most apparent against the methicillin sensitive S. epidermidis ATCC 12228/DSM1798 (FIG. 4) strain,. In contrast to the results obtained with the S. aureus strains, ID557 is the construct with the lowest activity when tested against S. epidermidis strains. Although ID453 shows similar activity as construct ID558 against the methicillin resistant S. epidermidis ATCC 35984/BB1336 strain at higher concentrations (>20 nM), ID453 is clearly superior at lower concentrations (<20 nM) (FIG. 5). Overall, our data demonstrated superiority of ID453 as compared to the other tested constructs.

3. Materials and Methods

3.1 Materials: Bacteria, Media, Buffers, Devices, Consumables

[0083]

TABLE-US-00005 TABLE 3 Bacteria, which were used for the experimentation. Bacterial Strain Article Nr. Supplier S. aureus NR-46543/USA300 JE2 (MRSA) NR-46543 NARSA, ETH S. aureus ATCC 12600/DSM20231 (MSSA) DSM20231 DSMZ S. aureus ATCC 12598/DSM20372 (MSSA) DSM20372 DSMZ (Cowan I) S. epidermidis ATCC 12228/DSM1798 (MSSE) DSM1798 DSMZ S. epidermidis ATCC 35984/BB1336 (MRSE) ATCC 35984 Berger-B?chi, UZH

[0084] The bacterial strains are listed with the corresponding article number, and supplier.

TABLE-US-00006 TABLE 4 List of media and buffers used for the experimentation. Media/Buffer Component Amount/L pH Protocol Tryptic Soy Broth Caso Bouillon 30 g 7.3 Dissolve 30 g/L of the ready mixed (TSB) medium (ready mixed) powder in dH2O, adjust pH using 10M NaOH. Autoclave. Tryptic Soy Broth Caso Bouillon 30 g 7.3 Dissolve the reagents in dH2O, (TSB) agar plates (ready mixed) 15 g adjust pH using 10 M NaOH. Agar Autoclave. LB agar, ready mixed Tryptone 10 g 7.4 Dissolve 40 g/L of the ready mixed (Square agar plates) Yest Extract 5 g powder in dH2O, adjust pH using NaCl (171 mM) 10 g 10M NaOH. Autoclave. Cool to 55? Agar 15 g C. and pour 70 ml per square plate. 10x Stopping buffer Trisodium 113.75 g [4.55 g] Weigh both components in [To produce ca. 40 citrate 136 g [5.44 g] separate bottles/50 mL Falcon ml] (dihydrate) tubes. Add 905 g [36.2 g] upH2O to Citric acid the trisodium citrate bottle/tube (monohydrate) and dissolve by swirling or vortexing. When the solution is completely dissolved, pour the solution the citric acid bottle/tube and dissolve by swirling or vortexing. Sterile filter with 0.22 ?m PES-membrane filter. Store at 4? C., light protected, and the shelf life is 1 month. 1x Stopping buffer Prepare 1x stopping buffer by making a 1:10 dilution of 10x stopping buffer with upH.sub.2O. Sterile filter with 0.22 ?m PES-membrane filter. Store at 4? C., light protected, and the shelf life is 1 month.

[0085] For each medium and buffer the components with the corresponding amount per litre, the pH and the protocol are listed.

TABLE-US-00007 TABLE 5 Chemicals, which were used for experimentation. Material Article Nr. Supplier Lot Nr. Caso Bouillon (ready mixed) 413820.1210 HuberLab C203081 LB-Agar (ready mixed) A0927.1000KG HuberLab 0927-1/103, Agar-Agar 5210.4 Carl Roth 489289238 Trisodium citrate (dihydrate) S1804-500G Sigma-Aldrich (Merck) BCBX4142 Citric acid (monohydrate) C1909-500G Sigma-Aldrich (Merck) SLCF3211 Human Serum, normal S1-LITER Sigma-Aldrich (Merck) 3717696

[0086] The chemicals are listed with the corresponding article number, supplier and lot number.

TABLE-US-00008 TABLE 6 List of devices and consumables, which were used for the experimentation. Name Model/Article Nr. Brand/Supplier Climo Shaker [Incubator Shaker] ISF1-X Kuhner Incucenter [Incubator] IC80 SalvisLab Water bath 1113 GFL Spectrophotometer Nanodrop Onec Thermo Fisher Scientific Vortex Vortex Genie 2 Scientific Industries Mini-Vortex Mini Vortex Mixer Fisher Scientific Vacuum pump BVC Professional Vacuubrand Mini-Centrifuge Z 130 M Hermle EVOLVE Pipette 1-Channel, 1-10 ?l 3012 Integra EVOLVE Pipette 1-Channel, 20-200 ?l 3016 Integra EVOLVE Pipette 1-Channel, 100-1000 ?l 3018 Integra EVOLVE Pipette 12-Channel, 1-10 ?l 3032 Integra EVOLVE Pipette 12-Channel, 10-100 ?l 3035 Integra VOYAGER Pipette 8-Channel, 50-1250 ?l 4724 Integra VOYAGER Pipette 12-Channel, 5-125 ?l 4732 Integra Semi-micro cuvettes (Greiner Bio-One) 7.613 101 HuberLab 1.5 ml Eppendorf Tubes, Safe-lock 11.3817.01 HuberLab 96-well Plate [F-Bottom, sterile, with lid] 400096 Bioswisstec 25 ml SureFlo Reservoir 4382 Integra 100 ml SureFlo Reservoir 4392 Integra Inoculation loops 7.731 170 HuberLab Filtropur S PES Filter, 0.45 ?m 83.1826 Sarstedt 20 ml Syringe (Codan) 3.7410.08 HuberLab 50 ml/60 ml Syringe (Codan) 3.7414.12 HuberLab Serological Pipettes 10 ml (Greiner Bio-One) 7.607 180 HuberLab Serological Pipettes 50 ml (Greiner Bio-One) 7.768 180 HuberLab 500 ml Bottle-Top 0.22 ?m PES Filter (Nalgene) 29669 Milian 50 ml Falcon Tube 50050 Bioswisstec

[0087] Each device/consumable is listed with its corresponding model, brand, and serial/article number.

3.2 Methods: Protocol and Procedure

3.2.1 Quantitative Killing Assay (QKA)

[0088] In total, three PGH constructs were tested in human serum against three S. aureus and two S. epidermidis strains. The aliquots of the constructs were stored in PCR tubes at ?80? C. (in CIEX PO4 Elution buffer). If not mentioned otherwise, all steps were performed under sterile conditions. A final PGH concentration range of 320 nM, 160 nM, 80 nM, 40 nM, 20 nM, 10 nM, 5 nM, 2.5 nM, 1.25 nM and 0.625 nM was tested against all five strains. The desired final bacterial control was between 1?10.sup.6-1?10.sup.7 CFUs/ml. At least two biological replicates were performed. One 96-well F-bottom plate was used for each PGH construct to be tested.

[0089] Precultures were prepared by inoculating 5 ml tryptic soy broth (TSB) medium with either S. aureus NR-46543/USA300 JE2, S. aureus ATCC 12600, S. aureus ATCC 12598 (Cowan I), S. epidermidis ATCC 12228/DSM1798 or S. epidermidis ATCC 35984/BB1336. Cultures were incubated at 37? C. and 180 revolutions per minute (rpm) (25 mm orbit) overnight (O/N).

[0090] An aliquot of human serum (stored at ?20? C.) was thawed in the water bath at 30? C., filtered using a 0.45 ?m filter and was stored on ice. A 1:25 dilution of the O/N culture was made by mixing 400 ?l O/N culture with 10 ml TSB medium, and the culture was incubated at 37? C. and 180 rpm until an optical density (OD600 nm) of 0.5-0.6 was reached. The culture was then put on ice for ?5 min to stop growth. 1 ml of the culture was adjusted to an OD of 0.51 with TSB medium, transferred to a 1.5 ml reaction tube and placed on ice. While the culture was growing, 160 ?l 1? stopping buffer was provided in each 96-well F-Bottom plate in rows B-H, and plates were stored at 4? C. until use. Once bacterial cultures had been adjusted to the desired OD, suitable aliquots (10-20 ?l aliquots) of the PGH constructs to be tested were thawed on ice and briefly spun down. The human serum was acclimatized to RT. 100 ?l of human serum were provided via reverse pipetting into the wells A1 and A3-A12 of the prepared 96-well F-Bottom plates (one plate per enzyme to be tested). In well A2 of each plate, the enzyme predilution of the respective construct to be tested in human serum (final volume 200 ?l) was prepared to reach a concentration twice as high (640 nM) as the highest final concentration (320 nM) to be tested. The enzyme predilution was mixed well, and a 2-fold dilution series was made by transferring 100 ?l from well A2 to well A3, mixing six times and changing pipette tips, continuing with the same procedure from well A3 to A4 and so on until well A11. The last 100 ?l were discarded from well A11 leaving wells A1 and A12 as negative controls (no addition of enzyme, only addition of bacterial suspension in serum). Each enzyme predilution and dilution series in the plates was prepared no more than 10 min prior to inoculation to minimize interaction of the enzymes with the walls of the plate wells. The bacterial suspension for inoculation was prepared by mixing 500 ?l of the vortexed, OD-adjusted culture with 4.5 ml of human serum (1:10 dilution) in a 25 ml reservoir. Using a manual multichannel pipette, 100 ?l of bacterial suspension from the reservoir were added to wells A1-A12 of the plate, mixed one time by pipetting up and down and the plate was immediately transferred to the Incubator Shaker at 37? C. and 180 rpm for exactly 30 min.

[0091] After the incubation of the plate for exactly 30 min, the plate was removed from the Incubator Shaker, and 20 ?l 10? stopping buffer were immediately added to row A with a multichannel pipette. The suspensions in the wells were homogenized by pipetting up and down eight times to stop further activity of the enzyme. Then, a 5-fold dilution series was made in the provided 1? stopping buffer by transferring 40 ?l from row A to row B, mixing six times, changing pipette tips and then transferring 40 ?l from row B to C and so on until row H. The last 40 ?l were discarded from row H. From each well of the 96-well plate, 5.5 ?l were spot-plated on a pre-dried LB agar square plate. After the spots were dried, the agar plates were placed in an incubator upside down at 37? C. for O/N incubation (S. aureus strains for ?16 h and S. epidermidis strains for ?20 h).

[0092] The following day, colony forming units (CFUs) per spot were counted. CFUs/ml were calculated, and viable cell count and CFUs/ml log reduction across the tested final concentration range (320 nM-0.625 nM) were visualized in graphs (GraphPad Prism 9.2.0). Results of biological replicates were averaged and the standard error of the mean (S.E.M) was calculated and displayed.

REFERENCES

[0093] Antimicrobial Resistance Collaborators. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet. 2022 Feb. 12;399(10325):629-655. doi: 10.1016/S0140-6736(21)02724-0. Epub 2022 Jan. 19. PMID: 35065702; PMCID: PMC8841637. [0094] Kleinschmidt S, Huygens F, Faoagali J, Rathnayake I U, Hafner L M. Staphylococcus epidermidis as a cause of bacteremia. Future Microbiol. 2015; 10(11): 1859-79. doi: 10.2217/fmb.15.98. Epub 2015 Oct. 30. PMID: 26517189. [0095] Guo Y, Song G, Sun M, Wang J, Wang Y. Prevalence and Therapies of Antibiotic-Resistance in Staphylococcus aureus. Front Cell Infect Microbiol. 2020 Mar. 17; 10:107. doi: 10.3389/fcimb.2020.00107. PMID: 32257966; PMCID: PMC7089872. [0096] Grundling, A., Missiakas, D. M. & Schneewind, O., 2006. Staphylococcus aureus Mutants with Increased Lysostaphin Resistance. Journal of Bacteriology, 188(17), pp. 6286-6297. [0097] Kashani, H. et al., 2017. Recombinant Endolysins as Potential Therapeutics against Antibiotic-Resistant Staphylococcus aureus: Current Status of Research and Novel Delivery Strategies. Clinical Microbiology Reviews, 31(1). [0098] Kornd?rfer, I. P. et al., 2006. The Crystal Structure of the Bacteriophage PSA Endolysin Reveals a Unique Fold Responsible for Specific Recognition of Listeria Cell Walls. Journal of Molecular Biology, 364(4), pp. 678-689. [0099] Schmelcher, M. et al., 2015. Evolutionarily distinct bacteriophage endolysins featuring conserved peptidoglycan cleavage sites protect mice from MRSA infection. Journal of Antimicrobial Chemotherapy, 70(5), pp. 1453-1465. [0100] Schmelcher, M., Donovan, D. M. & Loessner, M. J., 2012. Bacteriophage endolysins as novel antimicrobials. Future Microbiology, 7(10), pp. 1147-1171.