Genetic Characterization of Staphylococcus aureus, Staphylococcus argenteus, and Coagulase-Negative Staphylococci Colonizing Oral Cavity and Hand of Healthy Adults in Northern Japan

The spread of methicillin resistance and virulence among staphylococci in the community poses a public health concern. In this study, we investigated the prevalence of Staphylococcus species colonizing the oral cavity and hand (skin) of healthy university students and their phenotypic and genetic characteristics in northern Japan. Among a total of 332 subjects, 6 and 110 methicillin-resistant and susceptible Staphylococcus aureus (MRSA and MSSA, respectively) isolates were recovered from 105 subjects. MRSA isolates were genotyped as CC5, CC8, CC45, and CC59 with SCCmec-IIa or IV, among which an isolate of ST6562 (single-locus variant of ST8) harbored SCCmec-IVa, PVL genes and ACME-I, which are the same traits as the USA300 clone. ST1223 S. argenteus was isolated from the oral cavity and hand of a single student. Coagulase-negative Staphylococcus (CoNS) was recovered from 154 subjects (172 isolates), and classified into 17 species, with S. capitis being the most common (38%), followed by S. warneri (24%) and S. epidermidis (15%), including nine mecA-positive isolates. S. capitis was differentiated into seven clusters/subclusters, and genetic factors associated with the NRCS-A clone (nsr, tarJ, ebh) were detected in 10–21% of isolates. The colonization of the USA300-like MRSA variant and S. capitis with the traits of the NRCS-A clone in healthy individuals was noteworthy.


Introduction
Staphylococcus inhabits the skin and mucous membrane of humans and animals, forming normal bacterial flora. In terms of coagulase production, the genus Staphylococcus is largely classified into coagulase-positive and -negative staphylococcus (CoPS and CoNS, respectively), and some species with the coagulase-positive/variable trait [1]. CoPS is a major pathogenic group, including Staphylococcus aureus and minor species represented by S. argenteus [2]. S. aureus is the most common pathogen that causes various staphylococcal diseases such as superficial skin infections, pneumonia, bacteremia, food poisoning, and toxic shock syndrome [3]. S. argenteus, which is classified into S. aureus complex (SAC), has been recognized as an emerging pathogen in humans and animals, causing diseases similar to those of S. aureus [4]. Although CoNS represents a less virulent group of Staphylococcus, was investigated to clarify the spread of such virulent strains and also the antagonism with S. aureus in colonization.

Prevalence of Staphylococcal Isolates from Study Subjects
This research was conducted as an observational, cross-sectional study at Health Sciences University of Hokkaido, Ishikari-Tobetsu, Hokkaido, Japan. Eligible subjects were recruited among students who attended annual medical check-up at the university during a period from April to May 2021. Students who participated in this study belonged to six faculties/schools (pharmaceutical science, dentistry, nursing and social services, psychological science, rehabilitation science, dental hygiene school) and graduate schools of the university. From subjects who agreed to participate in this study, oral-cavity samples (saliva) and skin-swab samples from the hand were collected.
A total of 332 students participated in this study. The age rage of the participants was 18-45 years, and the average was 20.5 years. One hundred and eighteen CoPS isolates (116 S. aureus and 2 S. argenteus) were recovered from 105 students (32% of all the subjects), while 172 CoNS isolates were recovered from 154 students (46%) (Tables 1 and 2). Among the 118 CoPS, 101 isolates (86%) were derived from the oral cavity, with 26 isolates (14%) being obtained from both the oral cavity and the hand. The CoNS isolates were mostly recovered from the hand (165/172; 96%). Six MRSA isolates were obtained from the oral cavities of four subjects and the hands of two subjects, indicating the isolation rate of MRSA as 2% for all the subjects and 6% for the CoPS-positive subjects. More MSSA isolates were derived from the oral cavity (n = 84) than the hand (n = 26). S. argenteus was detected in one student (21-year-old, male; 0.3% of all subjects), from both the oral cavity and the hand.
One hundred and ten MSSA isolates were assigned to 12 coa-types, with coa-Vb and VIIb being the most frequent, while coa-IIa, IIIa, VIIa, and Xa were also common ( Table S1). The MSSA isolates were differentiated into 23 STs, among which ST188 (CC1) was the most common, followed by ST12, ST508 (CC45), ST8 and ST15. In this study, we identified four novel STs (ST6921, ST6922, ST6923, ST6924), among which ST6922 was grouped into CC5, and ST6923 was a single-locus variant of ST188.
Most MRSA showed resistance to AMP, ERY, and LVX, and inducible resistance to CLI, while they were susceptible to other antimicrobials including anti-MRSA drugs such as VAN (Table 2). MSSA was susceptible to most of the antimicrobials. Only low resistance rates (4-25%) were observed against AMP, ERY, CLI, GEN, and LVX, more commonly in ST8/ST30, ST30/ST508, ST30, ST508, and ST188/ST432, respectively. The S. argenteus isolates were susceptible to all of the antimicrobials examined.
The CoNS isolates showed generally high susceptibility rates (86-100%) to all of the antimicrobials examined, although a resistance rate of only 38% was found against FOF, due to the higher frequency of FOF resistance in S. capitis and S. warneri (Table S2). Isolates resistant to AMP, ERY, GEN, and LVX were more commonly detected in S. epidermidis and S. warneri.

Genetic Characterization of S. capitis
To understand the clonal diversity of S. capitis, which was the most frequently isolated among the CoNS species, we performed a phylogenetic analysis of arcC. As shown in Figure 1, 66 isolates were discriminated into two clusters 1 and 2, with six subclusters (1a-1f) in cluster 1. By the sequence analysis of the hsp60 gene, 28 isolates were discriminated into S. capitis subsp. capitis (19 isolates) or S. capitis subsp. ureolyticus (9 isolates), though subspecies of other isolates were not identified due to the unsuccessful amplification of hsp60 following the published PCR protocol [34]. hsp60 sequences of S. capitis isolates exhibited >98.7% identity within individual subspecies, while 90.5-92.5% identity between these subspecies (data not shown). Although these subspecies were not evidently distinctive of clusters, subsp. capitis was mostly assigned to cluster1-subcluster 1a, and all other subclusters of cluster 1 and cluster 2 contained subsp. ureolyticus. The nucleotide sequences of tgsC, lanC, nsr, tarJ and ebh that were determined for the representative isolates showed 97-100% identity to those of strain CR01 (data not shown), and were deposited to the GenBank database (Table S3).  ) and S. caprae (26D) that were retrieved from GenBank database are added to the dendrogram. S. capitis clusters 1 and 2 with subclusters 1a-1d are shown on the right. S. capitis subspecies are shown only for isolates whose hsp60 gene could be sequenced. Colors arranged in individual columns in the right side represent detection of mecA, ACME-arcA, tgsC, lanC, nsr, tarJ, ebh in the present study. For the reference strains, presence of these genes is shown for only CR01 marked with filled rhombus. ACME was highly prevalent in the cluster-1 subclusters 1a and 1e, and cluster 2. tgsC was detected in 56 isolates (85% of S. capitis isolates) of all the cluster/subclusters. Among the five bacteriocin genes examined, only the gallidermine biosynthesis cluster gene (lanC) was detected in 31 isolates (47%), mostly in the cluster-1 subclusters 1a and 1d. Next, we attempted to detect the genetic factors associated with the NRCS-A clone, which has been known to have increased pathogenicity to neonates [25][26][27][28][29]. As a result, nsr, tarJ and ebh were identified in 14, 7, and 10 isolates (11-21%), respectively, and at least one of these genes was found in 27 S. capitis isolates (41%). The distribution of these genes was generally different depending on the genetic group; nsr in the cluster-1 subclusters 1a, 1b, and cluster 2; tarJ in the cluster-1 subcluster 1b; ebh in the cluster-1 subclusters 1c-1f. The coexistence of nsr and tarJ was observed in the cluster-1 subcluster 1b, which includes the NRCS-A prototype strain CR01. However, none of the isolates harbored all three genes. The nucleotide sequences of tgsC, lanC, nsr, tarJ and ebh that were determined for the representative isolates showed 97-100% identity to those of strain CR01 (data not shown), and were deposited to the GenBank database (Table S3).

Co-Isolation of S. aureus and CoNS
Among the 116 S. aureus isolates, 107 isolates (92%) were isolated solely from oral cavity or hand samples, without the isolation of CoNS. CoNS was co-isolated with nine S. aureus (1 MRSA, 8 MSSA) isolates and one S. argenteus isolate (Table 4). CoNS coisolated with S. aureus belonged to five species, among which S. capitis and S. warneri were commonly found (three isolates each). Four S. capitis isolates that were co-isolated with MSSA and S. argenteus included those harboring mecA (n = 1), ACME (n = 2), and tgsC (n = 4). Among 66 S. capitis isolates, co-isolation with other Staphylococcus was found in nine isolates, which were all tgsC-positive and included four isolates with the gallidermine synthesis cluster gene.

Discussion
In the present study, we described the current status of staphylococcal colonization in the oral cavity and the skin of healthy adults in northern Japan, and revealed genetic characteristics of S. aureus and S. capitis. The prevalence of S. aureus (32%) and MRSA (2%) in our study, among university students of dentistry and other health science schools, was in line with those previously reported for healthy individuals; the isolation rate of S. aureus and MRSA from dental patients: oral cavity, 6-37% and 0-9%, respectively [35][36][37][38]; and dental students: any site of mouth/nose/skin, 15% and 3%, respectively [21]. Nevertheless, the present isolation rate of S. aureus was slightly lower than that in our latest research in the same study site, for dental patients and staff (44%) [24], suggesting a lower prevalence of S. aureus among healthy individuals unrelated to healthcare settings. Though the incidence of MRSA among S. aureus isolates in the present study (6 among 116 isolates; 5%) was comparable to that in our previous study (3 among 83 isolates; 3.6%), some genetic traits of MRSA were notable. First, two isolates belonged to SCCmec-II-ST5 (two isolates), while the remaining MRSA had SCCmec-IV (four isolates), which is most commonly carried by CA-MRSA [5,6]. ST5-MRSA with SCCmec-II has been predominantly known as HA-MRSA in Japan, eastern Asia and north America, designated the "New York/Japan clone" [39], and has been found to be a major MRSA strain in hospitals and the community in northern Japan [40,41]. In contrast, this strain has not yet been identified from the oral cavity of dental patients/staff or children in our study site [23,24], though it has been detected from community-associated infections [41]. The unexpected detection of colonizing ST5 SCCmec-II MRSA may imply the spread of the dominant HA-MRSA clone to healthy individuals in the community. Other STs or CC of MRSA isolates with SCCmec-IV, i.e., CC8, ST59, ST45, have been described as genotypes of common CA-MRSA clones [42], and detected in our previous studies for clinical isolates [41,43]. However, it was remarkable that ST6562 MRSA having SCCmec-IVa along with PVL genes and ACME-I was again identified, following its first detection in our previous study from the oral cavity of a 66-year-old dental patient [24]. This may suggest the potential dissemination of ST6562 (single-locus variant of ST8) MRSA among the community, despite a still low prevalence. Because the genetic traits of this clone are similar to those of the USA300 clone, which is predominantly CA-MRSA in the US [42], ST6562 is considered a variant of USA300 [24]. Though in Japan, the prevalence of the USA300 clone is still low, 5.1% of blood isolates of S. aureus were considered to be the USA300 clone in our previous study [43], and an increasing trend of this clone was also recently described by other researchers in Japan [44]. Thus, ST6562 is suggested to have emerged as a variant during the spread of USA300 in Japan. Still, the isolation of ST6562 MRSA from patients with infectious disease has not yet been reported. However, because the colonization of this clone may be a potential risk for severe disease due to PVL, as has been known for the USA300 clone, further surveillance may be necessary for colonizing and clinical isolates.
Among the MSSA isolated in the present study, the most commonly detected genotypes were ST5, ST8, ST12, ST15, ST30, ST188, ST508 (CC45), among which ST12, CC45, and ST188 were livestock-associated types [45,46]. Minor types found in this study, ST20, ST72, ST97, ST398, a part of CC5, and CC8 including ST72, were also related to animals [45]. Eventually, at least half of the MSSA isolates were related to animals, which was similarly observed in our previous study for dental patients/staff [24]. Therefore, a considerable part of MSSA colonizing healthy adults was suggested to be derived from animals in the present study, probably due to reduced pathogenicity to humans, while some isolates belonged to virulent clones represented by ST121 [47]. Furthermore, it was remarkable to identify ST1223 MS-S. argenteus from both the oral cavity and the hand of a single subject. The colonization of ST1223 S. argenteus was also described in the previous study in younger dental patients with 8-10-year-olds [24]. Despite a lower prevalence than S. aureus, S. argenteus has been identified among clinical isolates, as well as a cause of food poisoning in Japan [48,49]. Our present and previous findings provided evidence that the oral cavity and the skin could be reservoirs of S. argenteus.
The predominance of S. capitis among CoNS surpassing S. epidermidis was noteworthy, because the S. epidermidis has been the most prevalent CoNS species among those colonizing healthy individuals [17,23,50], as well as clinical isolates [12]. A similar trend of S. capitis, i.e., relatively higher frequency comparable to that of S. epidermidis, was observed in our previous study on colonizing Staphylococcus in dental patients/staff [24]. In addition, the incidence rate of ACME was 55%, which was comparable to that in S. epidermidis (58%) in the present study, and far higher than that in clinical isolates of S. capitis (7%) [12]. Although it is not clear whether this indicates an ecological change in the CoNS species among healthy individuals, it is possible that S. capitis with ACME might have spread as a colonizing strain due to the increased ability to persist on human skin. Furthermore, the production of bacteriocins and 6-TG [33,51] may have also contributed to the spread of S. capitis, overwhelming other CoNS species in the host. This may be supported in our present findings by the rather high incidence rates of tgsC (85%) and the gallidermin synthesis cluster gene (47%), and most of S. capitis isolates (62/66; 94%) were obtained without the co-isolation of S. aureus. However, the inhibiting effect on S. aureus in vivo conditions may not necessarily be explained by S. capitis 6-TG, because S. aureus/S. argenteus isolates from four specimens were co-isolated with S. capitis harboring tgsC.
While S. capitis has been known for its clinical importance as a cause of various infections including endocarditis, bacteremia, prosthetic joint infections, etc., the current crucial issue in public health is the emergence and spread of multidrug-resistant strains as well as the NRCS-A clone that causes neonatal sepsis [25,26,52]. As for traits specific to the NRCS-A clone, in silico prediction revealed the nsr, ebh, and tarJ genes, which were suggested to confer a competitive advantage to this clone in the neonatal gut [27,29]. In the present study, the mecA-positive (MR) rate in S. capitis was low (2/66; 3%). However, one or two of the nsr, tarJ, and ebh genes were detected in 10-21% of isolates. Though nsr was distributed to various genetic clusters/subclusters defined by arcC, the detection of tarJ and ebh was limited to some clusters. Particularly, isolates having both nsr and tarJ were classified into cluster-1 subcluster 1b, to which the prototype of NRCS-A (CR01 strain) was also assigned. Accordingly, it was revealed that S. capitis strains with genetically similar traits to the NRCS-A clone were colonizing in healthy adults. Although the reason for the emergence and worldwide endemicity of the NRCS-A clone is not evident, it is suggested that the distribution of some S. capitis lineages, such as cluster-1 subcluster 1b in the present study, may be related to the occurrence of the NRCS-A clone.
The present study revealed the colonization of USA300-like ST6525 MRSA and S. argenteus, and the presence of NRCS-A-clone-like S. capitis in healthy adults. Because these Staphylococci are considered to increase the risk of infections, further epidemiological surveillance of clinical and colonizing isolates may be necessary.

Study Subjects and Isolation of Staphylococcus
Saliva specimens of subjects were collected from the floor of the mouth by using a sterile cotton swab. A sterile cotton swab that was moistened with normal saline and rubbed on the palms and fingers was used as the hand swab specimen. All the swab samples were directly plated on CHROMagar Staph aureus (Kanto Chemical Industry Co., Ltd., Tokyo, Japan) and aerobically incubated at 37 • C for 48 h. Staphylococcus-like colonies were subcultured on blood agar plates followed by aerobic incubation at 37 • C overnight. For all the isolates grown on the plates, the partial 16S rRNA gene sequence was determined by Sanger sequencing with PCR products (approx. 1500-bp) as described previously [24]. The staphylococcal species of an isolate was identified as that showing >99% identity of the 16S rRNA sequence revealed by BLAST search (https://blast.ncbi. nlm.nih.gov/Blast.cgi, accessed on 31 January 2022). Individual isolates were stored in cryovials (Microbank, Pro-Lab Diagnostics, Richmond Hill, ON, Canada) at -80 • C and recovered when they were analyzed. DNA samples were extracted from cultured bacterial cells by the use of achromopeptidase (FUJIFILM Wako Pure Chemical Corp., Osaka, Japan).
Briefly, 1-2 colonies from pure bacteria culture were dissolved in TNE 100µL in a 1.5 mL tube and were centrifuged at 10,000 rpm for 1 min. The supernatant was removed and 10 µL achromopetidase (10,000 U/mL) were added and mixed by vortex, then heated at 40 • C for 10 min in a water bath. Next, 50 µL each of 0.5 M KOH and 1M Tris-HCl (pH 6.8) were added and mixed by vortex. The 1.5 mL tube was centrifuged at 10,000 rpm for 1 min and the supernatant was used as a DNA template for the PCR reactions. The PCR mixture contained 200 µM dNTP, 0.5 µM of each primer, 1.25 U Ex Taq DNA polymerase (Takara Bio Inc., Shiga, Japan) and its buffer with Mg2+ (final conc. 2 mM), extracted bacterial DNA 1µL (approximately 2-3 ng), and sterile distilled water to a final volume of 25 µL. PCR was performed on a thermal cycler (Gene Atlas, ASTEC, Fukuoka, Japan) with the following conditions: preheating at 94 • C for 2 min, 30 cycles of denaturation at 94 • C for 30 s, annealing at 55 • C for 30 s and extension at 72 • C for 1 min, and a final extension at 72 • C for 5 min. PCR amplicons were analyzed for their product size using electrophoresis on a 1.5% agarose gel and the results were recorded by a gel documentation machine.

Initial Genetic Characterization of Staphylococcal Isolates
For all the isolates, the presence of nuc, mecA, PVL genes, and ACME-associated arcA was confirmed by multiplex PCR assay as described by Zhang et al. [56]. In addition, PCR targeting the nonribosomal peptide synthetase (NRPS) gene was performed as previously described [57], to discriminate non-SAC species (S. argenteus, S. schweitzeri) from S. aureus. For all the methicillin resistant (mecA-positive) isolates, SCCmec type and subtype of SCCmec-IV were determined by multiplex PCR using previously published primers and conditions [58,59]. For all the ACME arcA-positive isolates, ACME type I, II, III, I', and II' was assigned by long-range PCR (LR-PCR) as described previously [12].

Genetic Typing and Analysis of S. aureus, S. argenteus, and S. capitis
The genotype based on the staphylocoagulase gene (coa type) of S. aureus and S. argenteus was determined by sequencing of partial coa (D1, D2, and the central regions), via PCR amplification with coa-ant1 and coa7 primers [60], and the subsequent search for coa type representing a highly similar coa sequence by BLAST. The sequence type (ST) of S. aureus and S. argenteus was determined according to the scheme of multilocus sequencing typing (MLST) [61].
For the genetic discrimination of two subspecies of S. capitis, i.e., subsp. capitis and subsp. ureolyticus, the partial hsp60 gene sequence was determined by direct sequencing of the PCR product with primers described by Kwok et al. [34]. The subspecies was assigned based on high sequence identity (>98%) of the hsp60 gene to that of prototype strains of S. capitis subsp. capitis (ATCC27840) or S. capitis subsp. ureolyticus (ATCC49324) [62].
Because the MLST scheme is not available for S. capitis, clonal diversity of S. capitis was analyzed by arcC, a housekeeping gene encoding carbamate kinase, which is included as one of the loci of the MLST scheme of most staphylococcal species. Partial arcC gene (approx. 500 bp) was amplified by PCR with primers designed in this study (Table S4), and its nucleotide sequence was determined by Sanger sequencing using BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA) on an automated DNA sequencer (ABI PRISM 3100, Applied Biosystems, Foster City, CA, USA). A phylogenetic dendrogram of arcC was constructed by the maximum-likelihood method using the MEGA.X software, together with arcC sequence data of S. capitis reference strains retrieved from GenBank database. For S. capitis isolates, the presence of the following genes was detected by PCR with primers listed in Table S4: bacteriocin genes encoding Nisin J, epidermicin, gallidermine biosynthesis cluster, PSM beta peptidase, and capidermicin; NRCS-A clone-related genes nsr, tarJ, and ebh involved in nisin resistance, teichoic acid biosynthesis, and cell-wall-associated fibronectin binding, respectively. The tgsC gene, which is among the 6-TG biosynthetic genes and located in the middle of the gene cluster [33], was detected by PCR as described previously [63]. The nucleotide sequences of the above-mentioned genes were determined by PCR and direct sequencing.