In total, 100 clinical isolates, including S. pseudintermedius (n = 60) and S. coagulans (n = 40), were obtained from canine superficial pyoderma cases at different veterinary clinics in Japan between June and July 2023 (Supplementary Table 1). The isolates were cultured on the Pearlcore Mannitol Salt Agar (Eiken Chemical Co., Ltd., Tokyo, Japan) at a companion animal-focused bacteriology testing facility (1s Co., Ltd., Kanagawa, Japan). Species-level identification was performed using a MALDI Biotyper (Bruker Daltonics, Billerica, MA, USA) and multiplex PCR targeting the nuc gene (9). As an alternative to oxacillin susceptibility testing, given that mecA is considered a more reliable marker of methicillin resistance in S. pseudintermedius and S. coagulans, we detected this gene via PCR, as previously described (10).

Antiseptic susceptibility testing was performed as described previously (11). Briefly, 20% (w/v) chlorhexidine digluconate solution (Sigma-Aldrich, Saint Louis, MO, USA) and solid chlorhexidine diacetate (Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) were used to prepare two-fold serial dilutions of 0.125–64 μg/mL. A 10% (w/v) aqueous solution of benzalkonium chloride (Alinamin Pharmaceutical Co., Ltd., Tokyo, Japan) was used to prepare dilutions from 0.5 to 12 μg/mL in 0.5 μg/mL increments. This range was selected based on the findings of a previous study (8), in which MIC values between 1 and 4 μg/mL were recorded for most of the assessed isolates. MICs were determined using the broth microdilution method, in accordance with the Clinical and Laboratory Standards Institute guidelines (M07-A11, 11th Edition) (12). Following previously described classification criteria (11), isolates with MIC values ≤ 4.0 μg/mL were categorized as having low MICs, whereas those with higher recorded MICs were categorized as having high MICs.

PCR detection of antiseptic resistance-associated genes (qacA/B, smr) was performed as previously described (8, 13, 14). Gene-specific primers were used for amplification, and their sequences are listed in Supplementary Table 2. Template DNA was prepared as previously described (9) by suspending a single colony in TE buffer containing achromopeptidase (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) and incubating at 55°C for 10 min.

Whole-genome sequencing of S. coagulans SC18 was conducted by Genome-Lead Co., Ltd. (Kagawa, Japan) using a hybrid approach combining Illumina short-read and Nanopore long-read sequencing. DNA was extracted from bacterial cells via enzymatic lysis, as previously described (15). For Nanopore sequencing, genomic DNA (1,000 ng) was processed using a short-read liminator XS (Circulomics, Baltimore, MD, USA), followed by library preparation using a Ligation Sequencing Kit (SQK-LSK110; Oxford Nanopore Technologies, Oxford, UK) and sequencing using the MinION flow cell (FLO-MIN106 R9.41 revD) with the GridION X5 platform (Oxford Nanopore Technologies). Base calling was performed using MinKNOW (v.23.07.5) and Guppy (v.7.0.9) (16), according to the manufacturer’s instructions. Illumina sequencing libraries were prepared using the Illumina DNA Prep (M) Tagmentation Kit (formerly Nextera DNA Flex; Illumina, San Diego, CA, USA), according to the manufacturer’s protocol. Paired-end sequencing (151 × 2 cycles) was performed using the NovaSeq 6,000 instrument (Illumina). Finally, base calling, demultiplexing, and adapter trimming were performed using BCL Convert (v3.9; Illumina).

Sequencing data were quality filtered by removing the low-quality short reads using standard thresholds. Hybrid assembly of the filtered short and long reads was performed using Unicycler (17). Genome completeness and contamination were evaluated using BlobTools (18), and taxonomic classification was confirmed using the Genome Taxonomy Database (19). The genome showed 97.8% average nucleotide identity with S. coagulans strain 1,031,336 based on the Nucleotide-Basic Local Alignment Search Tool (BLASTn) analysis. Genome polishing was conducted using Pilon (20) (short reads) and Racon (21) (long reads), followed by structural validation using BBMap (22).

Assembled plasmid sequences were analyzed using BLASTn searches against the NCBI nucleotide database to identify similarities with previously reported plasmids. Plasmid identification and assessment of phylogenetic relatedness were based on sequence identity and coverage values obtained from the BLAST results.

Next, a core genome alignment-based phylogenetic tree of S. coagulans was constructed using publicly available genome data with the kSNP3.0 algorithm (23), without a reference genome. Genetic population structure analysis was performed via Bayesian hierarchical clustering with the FastBAPS package (v.1.0.8) (24) in R v4.4.2 environment (25) to classify the isolates into genetically similar clusters. The number of clusters (K) was inferred automatically using optimise.baps prior, which selects the most appropriate population structure based on the input SNP alignment. Additionally, a maximum likelihood (ML) phylogenetic tree was constructed using RAxML-NG v.1.0.1 (26), with the best-fit model inferred using ModelTest-NG v.0.1.7 (27) and 100 bootstrap replicates. Subsequently, the ML tree was midpoint-rooted and visualized using FigTree v.1.4.4 (28).

Antimicrobial resistance genes were identified using Abricate (v.1.0.1) (29) and ResFinder (v.4.6.0) (30). Quinolone resistance-determining region mutations were detected using PointFinder (v.4.1.11) (31). Methicillin-resistance gene mecA (NG_047936.1) and antiseptic resistance-associated genes qacA (AB566411.1), qacB (AF535087.1), and smr (NC_005024.1) were identified via in silico analysis using the Nucleotide-Basic Local Alignment Search Tool against the assembled genome sequences.

CGView (v.1.0) (32) was used to visualize the genome structure of S. coagulans SC18, and genome annotation was performed using Prokka (v.1.14.6) (33). In addition to the chromosome, we obtained two small circular contigs carrying replication initiation protein (rep) genes with a GC content of <30%, which is lower than that of staphylococcal chromosomes (30–40%), indicating that these sequences might be plasmids. BLASTn analysis revealed a high similarity to known plasmids, confirming their identification as plasmids. Comparative analysis of smr-positive plasmids was performed using Easyfig v2.2.5 (34), and homologous regions between the plasmids were visualized using gradient gray shading. For comparative analysis, we used pST827 (Z37964.1) and pKSE124-2-5 (AP028327.1), which were the top BLASTn hits identified in the search described in Section 2.6 and represented type III short rolling-circle (RC) replicating plasmids, similar to the plasmid carried by SC18.

Inhibition zone diameters of amoxicillin/clavulanic acid, cephalexin, cefpodoxime proxetil, gentamicin, erythromycin, clindamycin, doxycycline, minocycline, enrofloxacin, chloramphenicol, and sulfamethoxazole/trimethoprim were determined using the KB/VKB discs (Eiken Chemical Co., Ltd.), according to the manufacturer’s instructions, via disc diffusion. Cefovecin disks were provided by Zoetis Japan Co., Ltd. (Tokyo, Japan). Antimicrobial susceptibility was assessed according to the Clinical and Laboratory Standards Institute guidelines (32^nd Edition: M100 (35) and 4th Edition: VET08 (36)) and EUCAST v.12.047 (37). In cases where interpretive criteria were not available, we followed the manufacturer’s instructions (Eiken Chemical Co., Ltd.). The interpretive criteria applied to each antimicrobial agent were as follows: amoxicillin/clavulanic acid, Eiken Chemical Co., Ltd.; cephalexin, Eiken Chemical Co., Ltd.; cefpodoxime proxetil, CLSI VET08, 4th ed.; cefovecin, CLSI VET08, 4th ed.; gentamicin, Eiken Chemical Co., Ltd.; erythromycin, CLSI M100, 32nd ed.; clindamycin, CLSI M100, 32nd ed.; doxycycline, CLSI M100, 32nd ed.; minocycline, EUCAST v12.0; enrofloxacin, CLSI VET08, 4th ed.; chloramphenicol, CLSI M100, 32nd ed.; sulfamethoxazole/trimethoprim, CLSI M100, 32nd ed.

To assess the association between the presence of antiseptic resistance-associated genes and mecA, we performed an analysis using Fisher’s exact test in R v4.4.3. A p-value of less than 0.05 was considered to be indicative of statistical significance.

In 2023, we collected 60 S. pseudintermedius and 40 S. coagulans strains from canine superficial pyoderma cases in Japan (Supplementary Table 1). Among these, 33 S. pseudintermedius and 27 S. coagulans isolates were mecA-positive (Supplementary Table 3). To determine their susceptibility to antiseptic chlorhexidine, MICs were measured using the broth microdilution method (Table 1 and Supplementary Table 4). MIC₉₀ was 1.0 μg/mL for chlorhexidine gluconate and 0.5 μg/mL for chlorhexidine acetate. Importantly, all isolates exhibited low MICs (≤ 2.0 μg/mL) to chlorhexidine, regardless of species or mecA status.

PCR screening for antiseptic resistance-associated genes revealed that one methicillin-susceptible S. coagulans isolate (SC18) carried smr (qacC; Supplementary Table 3). This gene primarily confers reduced susceptibility to quaternary ammonium compounds rather than to chlorhexidine (38). Therefore, susceptibility testing was performed for benzalkonium chloride, for which we recorded a notably low MIC of ≤0.5 μg/mL against SC18 (Supplementary Table 4). The isolates in this study showed an MIC₉₀ of 1.0 μg/mL (Table 2).

To determine the phylogenetic position of SC18, we performed whole-genome sequencing and constructed a phylogenetic tree using the kSNP algorithm incorporating 227 publicly available S. coagulans genomes retrieved from the National Center for Biotechnology Information database (Figure 1). Bayesian population structure analysis was performed using the fastbaps v1.0.8 package in R, based on the core genome SNPs. The optimal number of clusters (K) was inferred automatically using optimise.baps prior, which selects the most appropriate population structure from the data. This analysis grouped all isolates into nine major clusters (A–I). SC18 was classified into cluster D, the largest cluster comprised 48 isolates (Supplementary Table 5).

To analyze the resistance (−associated) genes distribution, local Basic Local Alignment Search Tool searches were performed against all genomes using the reference sequences for mecA (NG_047936.1), qacA (AB566411.1), qacB (AF535087.1), and smr (NC_005024.1). The results revealed that 21.1% (n = 48) of the genomes were mecA-positive, whereas 3.9% (n = 9) carried antiseptic resistance-associated genes (Figure 1 and Supplementary Table 5). Specifically, qacA was detected in 2.6% (n = 6) and smr in 1.3% (n = 3) of the isolates.

Although mecA was predominantly found in phylogenetic clusters B–D, the distribution of antiseptic resistance-associated genes was characterized by a geographical trend, with qacA detected exclusively in isolates from Thailand and smr identified in isolates from Japan, Australia, and Brazil.

Among the nine isolates carrying antiseptic resistance-associated genes (qacA or smr), seven (77.8%) also harbored mecA. Statistical analysis using Fisher’s exact test revealed a significant association between the presence of antiseptic resistance-associated genes and mecA (p = 0.00042, odds ratio = 14.18). All six qacA-positive isolates were mecA-positive (p = 0.000067, odds ratio = ∞), indicating a strong association. In contrast, no significant association was observed between smr and mecA (p = 0.51). These findings provide statistical support for the co-occurrence of antiseptic and antibiotic resistance genes, particularly qacA, although the sample size was limited.

To further characterize SC18, we performed hybrid sequencing to determine the complete genome of S. coagulans SC18. This strain possessed a single circular chromosome and two plasmids, pSC18-1 and pSC18-2, with smr located in pSC18-2 (Figure 2).

pSC18-1 was 3,175 bp in length and encoded a single rep gene and three additional genes predicted to encode hypothetical proteins with unknown functions. BLASTn analysis revealed that pSC18-1 had 94.36% sequence identity and 63% coverage with an unnamed plasmid (CP094737.1) from S. delphini strain IVB6222, which was isolated from a camel in Kenya (39). However, no antimicrobial resistance-or virulence-associated genes were identified in this plasmid.

pSC18-2 is a small plasmid of 2,233 bp containing only smr and rep. This is a type III short RC replicating plasmid previously characterized in staphylococci (40).

Comparative alignment revealed that pSC18-2 shared 70.5% sequence identity with pST827 (Z37964.1), a previously described type III short RC replicating plasmid from Staphylococcus spp. (41). Moreover, pSC18-2 exhibited high sequence similarity (99.7%) to pKSE124-2-5 (AP028327.1), a plasmid isolated from the S. epidermidis KSE124-2 strain obtained from a human in Hiroshima (Japan) in 2016 (Figure 3). Given the high similarity, pSC18-2 is considered genetically identical to pKSE124-2-5; however, we refer to this plasmid as pSC18-2 to enable a clear distinction for comparative analyses.

SC18 did not harbor any additional antimicrobial resistance genes and remained susceptible to clavulanic acid/amoxicillin, cephalexin, cefpodoxime proxetil, cefovecin, gentamicin, erythromycin, clindamycin, doxycycline, minocycline, chloramphenicol, and sulfamethoxazole/trimethoprim (Supplementary Tables 6, 7). However, this strain harbored several mutations in chromosomal quinolone resistance-determining regions, including GyrA S84F, ParC S80I and E84D, and ParE D432E and P535Y, conferring it resistance to enrofloxacin. This was the only antimicrobial-resistant phenotype observed in SC18.