A Novel Strategy to Identify Endolysins with Lytic Activity against Methicillin-Resistant Staphylococcus aureus

Abstract

The increasing prevalence of methicillin-resistant Staphylococcus aureus (MRSA) in the dairy industry has become a fundamental concern. Endolysins are bacteriophage-derived peptidoglycan hydrolases that induce the rapid lysis of host bacteria. Herein, we evaluated the lytic activity of endolysin candidates against S. aureus and MRSA. To identify endolysins, we used a bioinformatical strategy with the following steps: (1) retrieval of genetic information, (2) annotation, (3) selection of MRSA, (4) selection of endolysin candidates, and (5) evaluation of protein solubility. We then characterized the endolysin candidates under various conditions. Approximately 67% of S. aureus was detected as MRSA, and 114 putative endolysins were found. These 114 putative endolysins were divided into three groups based on their combinations of conserved domains. Considering protein solubility, we selected putative endolysins 117 and 177. Putative endolysin 117 was the only successfully overexpressed endolysin, and it was renamed LyJH1892. LyJH1892 showed potent lytic activity against both methicillin-susceptible S. aureus and MRSA and showed broad lytic activity against coagulase-negative staphylococci. In conclusion, this study demonstrates a rapid strategy for the development of endolysin against MRSA. This strategy could also be used to combat other antibiotic-resistant bacteria.

Keywords: Staphylococcus aureus; antibiotic resistance; bovine mastitis; endolysin; methicillin-resistant Staphylococcus aureus; prophage.

Figure 1

General information regarding Staphylococcus aureus genomes and putative endolysins. (a) Distribution of S. aureus and methicillin-resistant S. aureus (MRSA). (b) Quantity of putative endolysins observed in the MRSA genomes. (c) Distribution of domain types (single versus multiple domains). (d) Types of domains observed in single-domain putative endolysins. (e) Types of domain combinations observed in multiple-domain putative endolysins. The conserved domains were analyzed using the database from the National Center for Biotechnology Information.

Figure 2

Diversity of putative endolysins with both enzymatically active domains and cell wall binding domains identified from methicillin-resistant Staphylococcus aureus related genomes. (a) Cladogram of putative endolysins. In total, 114 putative endolysins were multi-aligned based on the amino acid sequences using the E-INS-I algorithm on multiple alignments using fast Fourier transform (version 7.505). The resulting alignments were phylogenetically reconstructed using FastTree, and the cladogram was constructed using ggtree in R software (version 4.1.3). Color tiles in the cladogram represent multi-alignment results among 114 putative endolysins. (b) Conserved domain structure of putative endolysins. The gray boxes represent enzymatically active domains and the white boxes show cell wall binding domains. CHAP, cysteine, histidine-dependent amidohydrolases/peptidases.

Figure 3

Multi-alignment of all putative endolysins with an enzymatically active domain and a cell wall binding domain based on amino acid sequences. (a) Multi-alignment of the cysteine, histidine-dependent amidohydrolases/peptidases domain from groups 1 and 2 (Accession: pfam05257 and sequence identity: 82.4%). (b) Multi-alignment of the Amidase-like domains (Accession: Ami_2 (smart00644) and Amidase_3 (pfam01520); sequence identity: 12.7%). (c) Multi-alignment of the SH3-like domain (Accession: SH3_5 (pfam08460), SH3 superfamily (cl17036), and SH3b (smart00287); sequence identity: 22.9%).

Figure 4

Structural characteristics of endolysin LyJH1892. (a) Analysis of endolysin LyJH1892 by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Lane 1 contained a protein molecular weight marker and lane 2 contained purified LyJH1892. (b) The conserved domain of LyJH1892. The gray square represents the enzymatically active domain, and the white square denotes the cell wall binding domain. (c) Three-dimensional model of LyJH1892, as predicted by AlphaFold2 using the Colab Fold notebook. (d) Connolly surface form of LyJH1892 created in ChimeraX 1.3. Blue and red colors represent the most positive and most negative polar activities, respectively.

Figure 5

Lytic activity of LyJH1892 against Staphylococcus aureus (NCCP 16830) at various (a) pH levels, (b) temperatures, (c) NaCl concentrations, and (d) metal ion additions. Data are shown as means ± standard deviation of triplicate assays. In Figure 5d, None, pure LyJH1892; EDTA, purified LyJH1892 after incubating with 5 mM ethylenediaminetetraacetic acid (EDTA); Metal ion (Ca²⁺, Mg²⁺, Mn²⁺, or Zn²⁺), LyJH1892 after adding 10 mM metal ions to EDTA-treated LyJH1892. Bars with different superscript letters (a–c) indicate a significant difference (p < 0.05).

Figure 6

Optimal lytic activity and lytic spectrum of LyJH1892. (a) Dose–response test against Staphylococcus aureus (NCCP 16830). (b) A lytic spectrum of LyJH1892. All lytic tests were performed under optimal conditions of LyJH1892 (pH 9.0, 25 °C, and no addition of NaCl and metal ions). Data are shown as means ± standard deviation of triplicate assays.