The Novel Phage-Derived Antimicrobial Agent HY-133 Is Active against Livestock-Associated Methicillin-Resistant Staphylococcus aureus

ABSTRACT

Livestock-associated methicillin-resistant Staphylococcus aureus (LA-MRSA) isolates are increasingly migrating from livestock into human and animal health care settings. Alternative substances are needed to overcome the drawbacks of currently available drugs used for MRSA eradication. The recombinant bacteriophage endolysin HY-133 has proved to be an active agent against S. aureus. Here, the in vitro activity of HY-133 was studied against a large collection of genetically diverse LA-MRSA isolates revealing its high activity against mecA-, mecB-, and mecC-positive LA-MRSA.

TEXT

Originally classified as a nosocomial pathogen, methicillin-resistant Staphylococcus aureus (MRSA) isolates are increasingly reported as frequent colonizers of animals, mainly of livestock but also of companion animals (1–5). In particular, mecA-based livestock-associated MRSA (LA-MRSA) isolates belonging to clonal complex (CC) 398 have been introduced back into human health care settings during the last decade (3, 6–11). This trend can be observed especially in regions characterized by a high density of livestock industry, such as the border region between Germany and The Netherlands (4, 9, 12–14). Moreover, mecC-based MRSA clonal lineages are widespread among farm, companion, and wildlife animals, and transmission to humans has been described (15–19). Finally, plasmid-borne mecB-mediated methicillin resistance with putative origin from animal-associated Macrococcus species has recently been found in a human MRSA isolate (20).

To contain zoonotic transmission, elimination of MRSA in the animal host is a key approach. However, efficient treatment of animals is challenging due to the wide spread of antibiotic resistance throughout livestock (21–23). The use of phage endolysins is a promising alternative to combat pathogens since, in contrast to conventional antibiotics, phage endolysins profit from high specificity for the target pathogen without affecting the cocolonizing bacterial community and from a low likelihood of inducing bacterial resistance formation (24, 25). HY-133 (HYpharm, Bernried, Germany), a recombinantly produced chimeric endolysin, and its forerunner PRF-119 (Hyglos, Regensburg, Germany) have been shown to efficiently and distinctively impact S. aureus strains (26, 27). Data on their activity against common lineages of LA-MRSA are not available.

In this study, HY-133 was tested on a representative set of diverse LA-MRSA samples comprising mecA- and mecC-carrying MRSA isolates (n = 50 each) and 1 mecB-carrying MRSA isolate belonging to 27 different spa types. Reference groups for evaluation of data from LA-MRSA included non-LA-MRSA and methicillin-sensitive S. aureus (MSSA) isolates (n = 15 each; 19 spa types). LA-MRSA isolates were of human and animal origin from the northwestern part of Germany derived within Dutch-German EUREGIO cross-border research projects (MRSA-net Twente/Münsterland, EurSafety-Health-Net, and SafeGuard MRSA vet-net) (4, 9, 16, 28, 29). Reference groups presented a selection of clinical strains from two German studies covering different spa types (Table 1) (6, 30). In addition, a uniquely reported mecB-positive MRSA isolate was included (20). In vitro activity of HY-133 against S. aureus was evaluated using the broth microdilution method in accordance with Clinical and Laboratory Standards Institute (CLSI) guidelines for evaluation of MICs and minimum bactericidal concentrations (MBCs), as previously described (31). In brief, 2-fold dilution series of HY-133 were prepared in cationic adjusted Mueller-Hinton broth medium (Becton Dickinson, Heidelberg, Germany), with final concentrations ranging from 0.06 to 8 μg/ml. S. aureus strains were inoculated with a final concentration of 5 × 10⁵ CFU/ml. Tests were performed in triplicate, and median values were taken for analysis. S. aureus ATCC 29213 was used as a quality control strain.

TABLE 1.

Antimicrobial activity of HY-133 against a large collection of mecA- and mecC-carrying LA-MRSA and one mecB-carrying MRSA compared with groups of non-LA-MRSA and MSSA

S. aureus group (no. of isolates)	spa types (no. of isolates)	MIC (μg/ml)			MBC (μg/ml)
		MIC50	MIC90	Range	MBC50	MBC90	Range
LA MRSA, mecC (50)	t843 (20); t1736 (5); t1535 (4); t3391, t6521, t9165, t11706 (2 each); t1738, t1773, t5930, t6220, t6292, t6902, t7914, t8842, t9123, t9738, t11120, t11702, t11290 (1 each)	0.12	1	≤0.06–4	0.12	1	≤0.06–4
LA MRSA, mecA (50)	t011, t034 (12 each); t108, t1451 (7 each); t571, t6867 (6 each)	0.25	0.5	≤0.06–1	0.25	0.5	≤0.06–1
MRSA, mecBa (1)	t091 (1)	NCb	NC	NC	NC	NC	NC
Non-LA-MRSA, mecA (15)	t008 (3); t003, t032, t037, t044 (2 each); t002, t007, t504, t019 (1 each)	0.5	1	0.12–1	0.5	1	0.12–1
MSSA (15)	t008 (3); t002 (2); t211, t021, t034, t076, t337, t352, t364, t529, t1029, t2927 (1 each)	0.25	0.5	0.12–1	0.5	0.5	0.12–1
Total (131)		0.25	0.5	≤0.06–4	0.25	0.5	≤0.06–4

^aMIC and MBC, 0.5 μg/ml each; putative livestock origin of the mecB-carrying plasmid (73.3% nucleotide identity with a plasmid of Macrococcus caseolyticus [20], a species colonizing animal skin and recovered from food).

^bNC, not calculable.

The isolates tested showed MICs and MBCs between 0.06 and 4 μg/ml. The MIC₅₀ was 0.25 μg/ml, and the MIC₉₀ was 0.5 μg/ml. Within each group of LA-MRSA isolates, values for MIC and MBC were identical. Comparison between the groups of LA-MRSA isolates showed a slightly wider range of HY-133 MICs and MBCs against mecC-carrying LA-MRSA (0.12 to 4 μg/ml) than against mecA-carrying LA-MRSA (≤0.06 to 1 μg/ml). For the mecB-positive MRSA isolate, median MIC and MBC values were 0.5 μg/ml. Comparison against values obtained from the groups of non-LA-MRSA and MSSA isolates (both 0.12 to 1 μg/ml) altogether showed similar MIC and MBC ranges (Table 1). Across the complete group of strains tested (n = 131), four strains (3.1%) showed MBC values different from the corresponding MICs, with an MBC:MIC ratio of 2 in all cases. However, phenotypic antimicrobial tolerance defined by an MBC:MIC ratio of ≥32 was excluded for all strains (32, 33).

The wide spread of antibiotic resistance in pathogenic bacterial strains calls for alternative MRSA prevention measures and treatment strategies for human and animal patients. However, conservative phage therapy brings with it several drawbacks, including the potential for horizontal gene transfer of virulence and antimicrobial resistance genes but also bacterial resistance mechanisms, such as attachment inhibition, injection blocking of phage DNA, restriction modification, or phage abortive infection (34–38). In Gram-positive bacteria, purified phage-derived enzymes can be applied exogenously, leading to rapid cell lysis and death (39), presenting a clear advantage over whole-phage application. Phage endolysins have the potential for future use as alternative antibacterials to reduce the resistance selection pressure caused by classic antibiotics. For example, their application may overcome the drawbacks of topical mupirocin administration used for nasal MRSA decolonization characterized by a slow bacteriostatic mode of action that requires several treatment courses and triggers resistance development. Even in the presence of sublethal levels of endolysins, an event of resistance formation is unlikely, as the targets in the bacterial cell wall are highly conserved (24, 25, 40). Similar to its precursor PRF-119 (27), HY-133 is composed of the CHAP (cysteine- and histidine-dependent amidohydrolase/peptidase) domain from the endolysin of phage K acting as enzymatic active domain (EAD) and a lysostaphin-derived cell wall-binding domain (CBD) (18, 19). However, in HY-133, the EAD-CBD link has been shortened to increase the enzyme's stability and specificity (26). The MIC values determined across the complete study set of MRSA and MSSA isolates are comparably low, similar to those of HY-133 against African S. aureus complex strains, as shown previously (26).

Comparison of the MICs between the different groups of LA-MRSA and the groups of non-LA-MRSA and MSSA revealed that the MIC ranges of LA-MRSA carrying the methicillin resistance gene mecC appeared to be slightly wider than the MIC ranges of the other groups. The mecC gene shares only 70% identity at the genome level with its homologue mecA and encodes a penicillin-binding protein (PBP) that is 63% identical to the mecA-encoded PBP2a on the amino acid level (19). Nevertheless, it was previously demonstrated in principle that mecC is inducible by oxacillin and mediates β-lactam resistance in S. aureus (41). Kim et al. (42) found that the mecC-encoded PBP had a lower optimum temperature than that of PBP2a. Moreover, the enzyme's structure was shown to collapse when incubated at 37°C, suggesting a loss of activity at higher temperatures. This impaired transpeptidase function might influence the formation of the bacterium's pentaglycine cross bridges, the part of the peptidoglycan layer where the CHAP region of HY-133 is supposed to bind. Thus, an attack of HY-133 on the cell wall may be hampered, explaining a slightly higher MIC of HY-133 in rare cases.

In conclusion, HY-133 shows a broad spectrum of activity against different clonal lineages of mecA-, mecB-, and mecC-carrying LA-MRSA.