This study aimed to elucidate retrospectively the correlations between the genome and phenotype in clinical methicillin-resistant Staphylococcus aureus (MRSA) gentamicin (GEN), clindamycin (CLI), and minocycline (MIN) susceptibility using next-generation sequencing (NGS) technology. Ninety two MRSA strains were isolated from individual inpatients treated in Hiroshima University Hospital, Hiroshima, Japan, extracted for their genomic DNA, and sequenced using an Illumina® MiSeq sequencer to obtain their de novo whole-genome assembly. An in silico analysis using ResFinder was performed to obtain the genomic antimicrobial susceptibility profile which was analyzed together with GEN, CLI, and MIN minimum inhibitory concentration (MIC) levels. This study found aac(6’)aph(2”)+, spc+, ermA+, tetM+ MRSA strains were predominant (42/92) and were shown to exhibit >16 mg/L GEN (40/42), >4 mg/L CLI (26/42), and >8 mg/L MIN MIC levels (30/42). Associations between aac(6’)aph(2”) detections and GEN MIC levels (p <0.001), ermA detections and CLI MIC levels (p <0.001), and tetM detections and MIN MIC levels (p <0.001) were revealed in this study. Correlations between simultaneous detections of aac(6’) aph(2”)-spc-ermA-tetM and GEN MIC levels (φc= 0.398, p <0.001), CLI MIC levels (φc= 0.448, p <0.001), and MIN MIC levels (φc= 0.515, p <0.001) were revealed in this study. The genomicphenotypic correlation analyses in this study provided an insight of a rapid antimicrobial detection in MRSA using in silico genomic antimicrobial susceptibility profiling.

This research was partially supported by a Grant-in-Aid for Scientific Research (A) (No.15H02567) from the Ministry of Education, Culture, Sports, Science, and Technology and those from Ministry of Health, Labor, and Welfare of the Government of Japan.