Molecular Epidemiological Characterization of Staphylococcus argenteus Clinical Isolates in Japan: Identification of Three Clones (ST1223, ST2198, and ST2550) and a Novel Staphylocoagulase Genotype XV

Staphylococcus argenteus, a novel emerging species within Staphylococcus aureus complex (SAC), has been increasingly reported worldwide. In this study, prevalence of S. argenteus among human clinical isolates, and their clonal diversity and genetic characteristics of virulence factors were investigated in Hokkaido, the northern main island of Japan. During a four-month period starting from March 2019, twenty-four S. argenteus and 4330 S. aureus isolates were recovered from clinical specimens (the ratio of S. argenteus to S. aureus :0.0055). Half of S. argenteus isolates (n = 12) belonged to MLST sequence type (ST) 2250 and its single-locus variant, with staphylocoagulase genotype (coa-) XId, while the remaining isolates were assigned to ST2198/coa-XIV (n = 6), and ST1223 with a novel coa-XV identified in this study (n = 6). All the isolates were mecA-negative, and susceptible to all the antimicrobials tested, except for an ST2198 isolate with blaZ and an ST2250 isolate with tet(L) showing resistance to ampicillin and tetracyclines, respectively. Common virulence factors in the S. argenteus isolates were staphylococcal enterotoxin (-like) genes sey, selz, sel26, and sel27 in ST2250, selx in ST2198, and enterotoxin gene cluster (egc-1: seg-sei-sem-sen-seo) in ST1223 isolates, in addition to hemolysin genes (hla, hlb, and hld) distributed universally. Elastin binding protein gene (ebpS) and MSCRAMM family adhesin SdrE gene (sdrE) detected in all the isolates showed high sequence identity among them (> 97%), while relatively lower identity to those of S. aureus (78–92%). Phylogenetically, ebpS, sdrE, selx, sey, selw, sel26, and sel27 of S. argenteus formed clusters distinct from those of S. aureus, unlike sec, selz, tst-1, and staphylokinase gene (sak). The present study revealed the prevalence of S. argenteus among clinical isolates, and presence of three distinct S. argenteus clones (ST2250; ST2198 and ST1223) harboring different virulence factors in northern Japan. ST2198 S. argenteus, a minor clone (strain BN75-like) that had been rarely reported, was first identified in Japan as human isolates.


Introduction
Staphylococcus argenteus is a novel species of coagulase-positive staphylococci previously described as a divergent lineage of S. aureus [1], and has been increasingly reported worldwide as an emerging pathogen affecting both humans  and animals [1,25,[27][28][29]. The major characteristic of this lineage is non-pigmented (white) colonies on blood agar plates due to lack of the crtOPQMN gene

Bacterial Isolates, Species Identification
S. aureus and S. argenteus were isolated from various clinical specimens that were brought to the Sapporo Clinical Laboratory Inc., Sapporo, Japan, from both hospitalized patients and outpatients in medical facilities in Hokkaido, during a four-month period starting on 7th March 2019. The clinical specimens were inoculated onto blood agar plates and incubated at 37 • C for 24 h aerobically. Gram-positive, coagulase-positive isolates were collected for further bacteriological identification. Initial screening of S. argenteus was performed by MALDI-TOF mass spectrometry using MALDI Biotyper (BRUKER). Isolates assigned as S.argenteus were confirmed genetically by three methods: (1) PCR targeting crtOPQMN gene operon-deficient region, (2) PCR targeting NRPS [4], and (3) sequence analysis of the partial NRPS gene [4] and nuc gene [1]. The method (1) was designed in the present study and performed by multiplex PCR to amplify crtOPQMN-deficient region (S. argenteus) or crtP (S. aureus) using primers listed in Table S1. In the method (2), non-S. aureus SAC (S. argenteus/S. schweitzeri)-specific amplicon that is 180-bp longer than that of S. aureus was detected. In method (3), full-length nucleotide sequence of nuc and partial NRPS gene were determined by PCR and direct sequencing, followed by alignment with those of S. aureus and S. argenteus using Clustal Omega program (https://www.ebi.ac.uk/Tools/msa/clustalo/).

Genetic Typing, Detection of Virulence Factors and Drug Resistance Genes
For all the isolates, presence of staphylococcal 16S rRNA, nuc, mecA, PVL genes and ACME-arcA (arginine deiminase gene) was examined by multiplex PCR assay as described by Zhang et al. [33]. Sequence type (ST) was determined according to the scheme of multilocus sequencence typing (MLST) (https://pubmlst.org/) [34]. spa type based on sequence of protein A gene X-region was determined by PCR and sequencing as described previously [35] using Ridom SpaServer (http: //spa.ridom.de/index.shtml). Staphylocoagulase genotype (coa-type) was determined by multiplex PCR assay as described previously [36]. For the isolates of which coa-type was not classified by the multiplex PCR, partial staphylocoagulase gene sequences (D1, D2, and the central regions) were determined and their highly similar staphylocoagulase sequences were searched by Basic Local Alignment Search Tool (BLAST: https://blast.ncbi.nlm.nih.gov/Blast.cgi) to assign their coa-types. For untypeable isolates, nucleotide sequences of whole staphylocoagulase gene was determined as described previously [37], subsequently sequence identities of D1 region, D2 and the central (C) region to those of established coa-types were analyzed by using Clustal Omega program.

Phylogenetic Analysis of Virulence Factors
Nucleotide sequences of full-length ORF were determined for staphylocoagulase genes belonging to a novel type, microbial surface component recognizing adhesive matrix molecule (MSCRAMM) family adhesin SdrE gene (sdrE), elastin-binding protein gene (ebpS), SE(-like) genes (selw, selx, sey, selz, sel26, sel27), and staphylokinase gene (sak) by PCR with primers listed in Table S1, followed by Sanger sequencing using the BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA) on an automated DNA sequencer (ABI PRISM 3100). Similarly, full length sequences of enterotoxin gene sec and tst-1 gene were also determined. Phylogenetic dendrograms of these virulence factors were constructed by the maximum likelihood method using the MEGA7 software, together with sequence data of staphylococcal strains available in GenBank database. The dendrograms were statistically supported by bootstrapping with 1000 replicates. Clustal Omega program was used for multiple alignments of amino acid sequences of EbpS. Prediction of transmembrane/hydrophobic region in EbpS was performed by TMpred program (https://embnet.vital-it.ch/software/TMPRED_form.html). All the sequence data of S. argenteus genes (nuc, nrps, sec, selx, sey, selw, selz, sel26, sel27, tst-1, sdrE, ebpS and sak) determined in the present study were deposited in GenBank database under accession numbers shown in Table S2.
Thermostable nuclease gene (nuc) of all the 24 isolates comprised 669 nucleotides that was shorter than that of S. aureus by 18 bp, and showed 98.8-100% identity to those of S. argenteus strains (MSHR1132 T , BN75, XNO62, XNO106, 51183) (Table S3), while 82% to S. aureus nuc gene. Similarly, partial NRPS gene sequences (264 bp) of the 24 isolates had 98.1-100% identity among them and to those of the reported S. argenteus strains, in contrast they were 180-bp longer than that of S. aureus strain NCTC8325 showing 60-61% identity.
Because coa-type of S. argenteus isolates could not be assigned to I through X by the multiplex PCR method, sequence identities of D1 region and D2-C region to those of established coa-types Ia-XIV were analyzed. All the ST2250 isolates were assigned to staphylocoagulase genotype XId, while ST2198 isolates to coa-XIV.
According to the criteria to define coa-type proposed previously [43], staphylocoagulase genes showing > 90% sequence identity in D1 region and D2-C regions are assigned to identical coa-type and subtype within the coa-type, respectively. D1 and D2-C regions of six isolates (SG03, SG06, SG07, SG15, SG17 and SG18) showed less than 88.3% and 82.9% identities to the 14 established coa-types, respectively, while 100% identities to S. argenteus strain SJTU F20214, D7903 and M051_MSHR of which coa-type has not yet been described (Table 2). Therefore, a novel staphylocoagulase genotype XV was created and assigned to all the six ST1223 S. argenteus isolates.

Prevalence of Virulence Factors, Drug Resistance Genes and Antimicrobial Susceptibility
None of the isolates had PVL genes (lukS-PV-lukF-PV) or ACME-arcA genes. Alpha-, beta-, and delta-hemolysin genes (hla, hlb, hld) were detected in all the isolates (Table 1). SE-like genes sey, selz, sel26 and sel27 were detected in mainly ST2250 isolates, while selz was harbored by some ST2198 and ST2113 isolates. selx was found in all the ST2198 isolates and one isolate each of ST2250 and ST1223. All the ST1223 isolates harbored seg-sei-sem-sen-seo (egc-1) and selw. sec and tst-1 genes were detected in three and two isolates of ST2250, respectively. However, all other remaining SE genes and exfoliative toxin genes examined were not detected. Staphylokinase gene (sak) was detected in seven isolates (six ST2250 and one ST2198 isolates). Nine adhesin genes (fib, clfA, clfB, ebpS, eno, fnbA, fnbB, sdrE, icaA) were identified in all the isolates, while most isolates had sdrC and sdrD. Two hemolysis-positive and negative clones (SG05-1 and SG05-2, respectively) from a patient, were classified into the same genotypes (ST2250/coa-XId) and showed identical profiles of virulence factors.
All the isolates were methicillin-sensitive without having mecA, and susceptible to 20 antimicrobials tested, except for two isolates; an ST2198 isolate SG20 with blaZ was resistant to ampicillin (MIC, 4 µg/mL) and an ST2250 isolate SG23 with tet(L) showed resistance to tetracycline (MIC, 64 µg/mL) and intermediate resistance to doxycycline (MIC, 8 µg/mL). None of the isolates harbored other resistance genes tested.

Phylogenetic Analysis of Virulence Factors
To clarify phylogenetic relatedness of staphylocoagulase genes of the novel coa-type XV to those of other types, dendrograms of D1 and D2-C regions of staphylocoagulase genes were constructed (Figure 1a,b, respectively). In D1 region, S. argenteus strains were discriminated into five clusters representing coa-XId, -XII, -XIV, -XV, and an unassigned type. coa-XV D1 region was revealed to be of the same lineage as that of coa-VIa, although sequence identity was 88.3% (Table 2). In contrast, D2-C regions of all the S. argenteus strains including coa-XV clustered in a single phylogenetic group, which was distinctive from those of S. aureus. a coa genotype of these isolates were assigned to XId in our previous study [5]. coa-XV is a novel type identified in this study. NA, not assigned; presumptive coa type XVI was assigned in our previous study [5]; Nucleotide sequences of SG07 and SG15 were not included to this Table because they were 100% identical to that of SG03. SE-like toxin genes selx, selw, and sel27 in S argenteus isolates, were phylogenetically distinct from those of S. aureus with 94.4-98.4% sequence identity, forming isolated clusters ( Figure S1). Similarly, sey and sel26 of S. argenteus formed clusters evidently distinct from S. aureus, while showing 98-99% identity to those of S. aureus. In contrast, selz of S. argenteus was not distinctive from that of S. aureus, likewise sec, tst-1 and sak having 98-100% sequence identity to S. aureus. (Table S4) Elastin binding protein gene (ebpS) was phylogenetically distinct between S. argenteus and S. aureus with only 78-89% identity, while higher sequence identity (>98.9%) was noted among S. argenteus isolates. Despite the lower identity and difference in amino acid length, S. argenteus EbpS contained three hydrophobic domains (H1, H2, and H3) that were located at almost the same positions as in the S. aureus EbpS [44], among which H3 domain was the most conserved between the two species ( Figure S2).
SdrE genes (sdrE) was revealed to have two major lineages I and II, both of which contained S. argenteus clusters (cluster 1 and 2, respectively). ST2250 and ST1223 isolates were grouped into cluster 1, which was closely related to an S. schweitzeri strain, while cluster 2 comprised ST2198 S. argenteus including strains MSHR1132 T and BN75. Within a same cluster, S. argenteus sdrE showed >97% identity, while 88.6-90.2% identity between cluster 1 and 2. Within a same lineage, identity of sdrE sequence between S. aureus and S. argenteus was 90-92.7%. Amino acid sequence alignment of SdrE revealed that TYTFTDYVD motif, which is conserved within MSCRAMMs (ClfA, ClfB, SdrC, SdrD, SdrE) of S. aureus [45], are also found in S. argenteus isolates, although only lineage II (ST2198) strains had a modified motif (TYKFTDYVD, single amino acid difference is underlined) that was reported for Bbp (bone sialoprotein-binding protein, an allelic variant of SdrE) [46] ( Figure S3). In addition, CnaBE3 domain, which is implicated in immunological protection of bacterial cells in mouse model [47] was also highly conserved, except for an ST1223 isolate SG06.

Discussion
Several clonal complexes including more than 60 STs of S. argenteus have been described so far [1,3], and some CCs (CC75 and CC2250) show a wide geographical distribution indicating international spread [48]. The prevalence of S. argenteus reported to date have varied depending on geographical areas, study populations and infection sites. In northern Australia, CC75 S. argenteus accounted for 8% of MSSA and 69% of MRSA isolated from skin infections of indigenous children [1].
Two separate studies in Thailand revealed an S. argenteus prevalence of 19% for community acquired MRSA causing sepsis [11] and 4.1% among invasive staphylococcal infections [10]. In Lao PDR, 6% of presumptive S. aureus isolates from skin and soft tissue infections were identified as S. argenteus [12]. In Myanmar, prevalence of S. argenteus was reported as 0.9% (5/563) from healthy nasal carriers [14] and 2.9% (4/137) [5] in clinical isolates. In contrast, the prevalence seems to be low in European countries (< 1%, Belgium (0.16%) [19], Sweden (0.3%) [21] and Denmark (0.35%) [3], and eastern China (0.7%) [4]. In our present study, the prevalence (ratio of S. argenteus to S. aureus) in the northern Japan was revealed to be 0.55% (0.85% to MSSA), that was lower than Southeast Asian countries but comparable to those of European countries and eastern China. The 24 S. argenteus isolates in our study were derived from 11 cities/towns, and various specimens from inpatients and outpatients of all age groups, suggesting that this species may be widely distributed to general population endemically, despite considerably lower prevalence than S. aureus.
Among S. argenteus, ST2250 has been described as the most widely spread genotype [45], and found to be a dominant clone in Thailand [10,11,24], Denmark [3], and Myanmar [5,14]. In contrast, ST1223 (CC1223) and ST2198 (CC2198) are recognized as minor lineages [3,45], and reported in Thailand [10,11] and Lao PDR [12] at low detection rate. In Japan, food poisoning outbreaks were caused by ST1223 S. argenteus strains [15,16], while a purulent lymphadenitis and bacteremia cases by ST2250 strain [17,26]. Despite the dominance of ST2550, our present surveillance suggested that ST1223 S. argenteus may not be a rare clone in Japan. All the six ST1223 isolates in our study had enterotoxin gene cluster (egc-1), as reported for isolates from the food poisoning [15], and three and one isolates were derived from stools and subdural abscess, respectively. Thus, it is suggested that ST1223 may be related to gastrointestinal and/or invasive infections. Meanwhile, S. argenteus of ST2198 was first reported in Japan in our study. Genetically, most well-characterized ST2198 strain is BN75 which was isolated from feces of a western lowland gorilla in Gabon [27]. This strain has less virulence factor genes and no drug resistance genes, showing susceptibility to all the antimicrobials tested, which was similar characteristics to our present ST2198 isolates.
Along with coa-XII that was assigned to strain MSHR1132 T , S. argenteus strains reported to date belong to at least four coa-types, XId, XII, XIV, and XV, which are scarcely detected in S. aureus. The coa-type is defined by sequence diversity in the D1 and D2 regions, which contain antigenic regions and are associated with prothrombin-binding of staphylocoagulase [49]. Therefore, the presence of the coa-types in S. argenteus distinctive from S. aureus suggests that S. argenteus staphylocoagulase might have evolved through selective pressure with antibodies and/or prothrombin in animal species that is different from humans. This assumption is supported by the finding of S. argenteus isolates in Thailand which are genetically related to livestock-associated S. aureus [24].
In this study, prevalence of SE (-like) genes in S. argenteus isolates was revealed to be generally different depending on the clone. A report in Thailand described that S. argenteus has less virulence factors compared with S. aureus, suggesting its less virulence to humans [11]. In contrast, higher risk of mortality was reported for S. argenteus bacteremia than that of S. aureus [18]. Although it is not certain whether toxins are sole determinants of virulence, it is evident that prevalence of virulence factors is different depending on S. argenteus clone. Thus, virulence of S. argenteus is suggested to be variable by individual clones. Because S. argenteus genome has 87% identity to that of S. aureus [1], and as high as 38% of S. argenteus genes have only 85% homology to S. aureus [3], it is conceivable that S. argenteus may have any unidentified virulence factors, or variants of virulence factors commonly shared with S. aureus.
Phylogenetic analysis revealed that SE (-like) genes, selw, selx, sey, sel26 and sel27 of S. argenteus were genetically distinguished from those in S. aureus, despite showing high sequence identity to S. aureus, suggesting that these genes might have been shared among the two species or present in the most recent common ancestor since long time ago, followed by evolving within individual species. Similar findings were reported for protein A and alpha-hemolysin genes in our previous study [5]. In contrast, sec, selz, tst-1 and sak detected in S. argenteus isolates were almost identical to those of S. aureus, indicating that dissemination of these genes might have occurred more recently.
It was notable in the present study that ebpS and sdrE genes were genetically distinct from those of S. aureus with lower identity for ebpS (78-89%) and slightly high identity for sdrE (90-92%). However, elastin binding protein (EbpS) of S. argenteus was presumed to have three hydrophobic domains at the same positions as S. aureus [44], suggesting that it has the same topology and function as S. aureus EbpS. Similarly, TYTFTDYVD motif and CnaBE3 domain of SdrE were conserved among S. argenteus and S. aureus, suggesting that SdrE of these two species may have the same function.
In summary, the present study revealed the prevalence and clonal diversity of S. argenteus in northern Japan, together with presence of SE(-like) genes and other virulence factors. A novel staphylocoagulase genotype XV was identified for ST1223 clones. S. argenteus isolates harbored ebpS, sdrE, selw, selx, sey, sel26, and sel27 that were phylogenetically distinct from those of S. aureus. This is the first comprehensive report on molecular epidemiological study of S. argenteus clinical isolates in Japan. Identification of emerging S. argenteus strains with three different clones with different profiles of virulence factors highlights the need for continuous surveillance of S. argenteus to understand its clinical significance.