Abstract
Prevalence and genetic characteristics of methicillin-resistant Staphylococcus aureus (MRSA) and coagulase-negative staphylococci in oral cavity of healthy children were studied in Hokkaido, northern main island of Japan. From saliva of 526 children, a total of 248 staphylococcal isolates comprising S. aureus (n = 143), S. epidermidis (n = 84), S. warneri (n = 13), S. haemolyticus (n = 5), S. hominis (n = 2), and S. intermedius (n = 1) were recovered. Presence of mecA was confirmed in 6.3% of S. aureus, 50% of S. epidermidis, and 7.7% in S. warneri. SCCmec was mostly classified into type IV, and ACME (arginine catabolic mobile element)-arcA was detected in S. epidermidis (23.8%) and S. intermedius. Nine MRSA isolates belonged to staphylocoagulase gene (coa) type Ia, IIa, IIIa, VIIb/sequence type 1 (ST1), ST5, ST8, ST89, ST120, and were negative for PVL (Panton-Valentine leukocidin) genes. These isolates included two clones of emerging community-acquired MRSA (CA-MRSA) that had been described recently in Japan: ST5/SCCmec IVc, which resembles the “Pediatric clone,” and ST8/SCCmec IVl belonging to coa-IIIa/agr-I with sasL gene, designated “CA-MRSA/J” clone. Various enterotoxin genes were found in all the MRSA and some methicillin-susceptible S. aureus (MSSA) isolates examined, while tst-1 was detected in four MRSA isolates. Notably, a variant of elastin-binding protein gene (ebpS-v) was identified in ST120 MRSA and ST45 MSSA isolates, and exfoliative toxin D gene (etd) was detected in an MSSA isolate. The present study revealed the presence of MRSA, including the novel CA-MRSA clones, and high prevalence of methicillin-resistant S. epidermidis in oral cavity of healthy children in Japan.
Introduction
Staphylococcus is one of the most prevalent genera causing a broad variety of infectious diseases in humans, and generally classified into S. aureus and coagulase-negative staphylococci (CNS). S. aureus is the well-known cause of common skin and soft tissue infections; life-threatening diseases including sepsis and necrotizing pneumonia; as well as toxic diseases such as food poisoning and toxic shock syndrome. CNS, which is represented by S. epidermidis, are common pathogens of nosocomial and opportunistic infections associated with indwelling devices. 1 Methicillin-resistant S. aureus (MRSA) and methicillin-resistant CNS (MR-CNS) are an important cause of hospital-acquired (HA) infections.1,2 Methicillin resistance of staphylococci is conferred by mecA gene, which encodes an alternative penicillin-binding protein (PBP2a) with low affinity to β-lactams, and is carried within a transmissible genetic element, staphylococcal cassette chromosome mec (SCCmec). SCCmec in MRSA is genetically highly diverse in structural organization and has been classified into at least 13 SCCmec types (I-XIII),3,4 among which types I, II, III, IV, and V are commonly found in HA-MRSA and/or community-associated MRSA (CA-MRSA). 5
Staphylococci constitute normal flora of skin and mucous membrane in humans, with CNS being most widely distributed. S. aureus has been documented to colonize approximately 30% of healthy individuals asymptomatically in their nasal cavity, generating persistent and non-persistent carriers.6,7 Colonization of S. aureus as well as CNS, particularly methicillin-resistant strains, is considered to increase the risk of invasive diseases, including bacteremia, in hospital settings.6,8 Associated with worldwide recognition of CA-MRSA as a public health issue since the late 1990s, the spread of MRSA as well as MR-CNS among the community has been also another growing concern.5,9 Accordingly, the prevalence of colonizing methicillin-resistant staphylococci in nasal cavity has been investigated for healthy adults, children, health care workers, and food handlers, in many countries.10–14 Although the anterior nares are considered the primary ecological niche for staphylococcus, much less study has been done on staphylococcal colonization in oral cavity. The oral cavity is known to serve as a reservoir of S. aureus for lower respiratory infections as well as cross-infections to others.15,16 It was also revealed that S. aureus may inhabit the oral cavity and oropharynx, irrespective of simultaneous colonization in the nasal cavity. 13 Therefore, the colonization in oral cavity deserves more attention as a mode that mediates spreading of CA-MRSA and MR-CNS in a community.
In Japan, prevalence of MRSA and MR-CNS in nasal cavity of children in 2001–2003 was described to be 3.8–4.3% and 28–30%, respectively, revealing a high degree of their genetic diversity.17,18 Oral colonization of S. aureus in children was studied in only one report showing persistence and increase in MRSA for a 6-year period. 19 However, detailed genotypes of oral staphylococcal isolates and their genetic traits have not yet been analyzed.
The present study was conducted to determine the prevalence of staphylococci derived from oral cavity and their methicillin resistance in healthy children in Hokkaido, northern main island of Japan. The isolated MRSA and methicillin-susceptible S. aureus (MSSA) strains were further analyzed for their genotypes, prevalence of genes encoding virulence factors, adhesins, and drug resistance determinants to understand their features and potential as a cause of infections in a community.
Materials and Methods
Bacterial isolates and susceptibility testing
Staphylococcal isolates collected from oral cavity of 526 healthy children aged 3–6 years who had dental check-up at nursery schools and kindergartens from June 2013 to July 2015 were analyzed. Saliva samples were allowed to drip from the lower lip into a sterile test tube. A sterile pipette was then used to transfer a 100 μL sample to mannitol salt agar plates for bacterial culture, followed by incubation at 37°C for 48 hours aerobically. Species of bacterial isolates was identified biochemically with BBL Crystal Gram-Positive ID Kit (Becton Dickinson Microbiology Systems, Cockeysville, MD) or genetically with multiplex PCR (M-PCR) targeting 16S rRNA and thermonuclease genes.20,21 When staphylococcal species were not identified by the above methods, a partial sequence of 16S rRNA gene was determined with primers Epsilon F and 1510R. 22 The presence of mecA, PVL (Panton-Valentine leukocidin) gene, and ACME (arginine catabolic mobile element)-arcA genes was investigated for all MRSA and MR-CNS isolates with M-PCR assay as described by Zhang et al. 21 Individual isolates were stored in Microbank (Pro-Lab Diagnostics, Richmond Hill, ON, Canada) at −80°C and recovered when they were to be analyzed. Antimicrobial susceptibility was measured by broth microdilution test using Dry Plate Eiken DP32 (Eiken, Tokyo, Japan). MICs of 18 antimicrobial agents (oxacillin, ampicillin, cefazolin, cefmetazole, flomoxef, imipenem, gentamicin, arbekacin, erythromycin, clindamycin, vancomycin, teicoplanin, linezolid, fosfomycin, levofloxacin, cefoxitin, and trimethoprim/sulfamethoxazole) were measured and resistance was judged according to break points mentioned in the Clinical Laboratory Standards Institute guidelines. 23 This study was approved by the research ethics committee of the Health Sciences University of Hokkaido, Japan (No. 2014-003, 2015-026), and saliva samples were collected from children after obtaining consent from their guardians.
Genotyping
Staphylocoagulase gene (coa) type of S. aureus was determined with M-PCR assay as described previously. 24 For the isolates in which coa type was not classified into I–X by the M-PCR, partial coa sequences (D1, D2 regions) were determined to search for highly similar coa sequences through BLAST search (http://blast.ncbi.nlm.nih.gov/Blast.cgi) for assignment of coa type, as described previously.25,26 For methicillin-resistant isolates with mecA, SCCmec type and subtype of SCCmec IV were determined with M-PCR using previously published primers and conditions.27–29 Sequence type (ST) of S. aureus was determined according to the scheme of multilocus sequencing typing, 30 and accessory gene regulator (agr) group was assigned by the PCR with specific primers as previously decribed. 31 ACME type was determined for all the ACME-arcA-positive isolates by PCR profiling analysis as described previously.32,33
Detection and analysis of virulence factor and drug resistance genes
Prevalence of genes encoding various toxins (enterotoxins, exfoliative toxins, TSST-1), virulence factors, and adhesins in S. aureus, and antimicrobial resistance-associated proteins in S. aureus and MR-CNS was analyzed with multiplex or uniplex PCRs, using primers described previously.29,34,35 Nucleotide sequences of seb, sec, toxic shock syndrome toxin-1 (TSST-1) gene, exfoliative toxin D (etd), epidermal cell differentiation inhibitor B (edinB), and variant of elastin-binding protein gene (ebpS-v) were determined as described in the previous studies,35,36 by direct sequencing with the PCR products, using the BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems, Foster City, CA) on an automated DNA sequencer (ABI PRISM 3100). Phylogenetic dendrogram of ebpS/ebpS-v genes was constructed by maximum likelihood method using the MEGA.6 software package. Multiple alignment of elastin-binding protein (EbpS) amino acid sequences determined in the present study and those retrieved from the GenBank database was performed with the ClustalW program (www.genome.jp/tools-bin/clustalw/) available in the GenomeNet Database Resources.
The nucleotide sequences of tsst-1 (isolate ID: H59-3), ebpS-v (isolate ID: F17, Y74), and etd-edinB (isolate ID: Y86-2) were deposited in the GenBank database under accession numbers MG717461–MG717464, respectively.
Results
A total of 248 staphylococcal isolates were recovered from oral cavity of 242 healthy children; a single isolate was obtained from 236 children, while two isolates with different species or genetic traits (presence of mecA and/or ACME) were collected from six children (total 12 isolates; Supplementary Table S1. The isolates were identified as S. aureus (n = 143), S. epidermidis (n = 84), S. warneri (n = 13), S. haemolyticus (n = 5), S. hominis (n = 2), and S. intermedius (n = 1). Overall rate of methicillin resistance was 21.0% (52/248), and the highest mecA-positive rate was seen in S. epidermidis (50%; 42/84), followed by S. aureus (6.3%; 9/143) and S. warneri (7.7%; 1/13) (Table 1). Incidence rates of MRSA and methicillin-resistant S. epidermidis (MRSE) among the study subjects were 1.7% (9/526) and 7.8% (41/526), respectively. (Two MRSE isolates were obtained from one subject [Supplementary Table S1]) SCCmec type was assigned to IV or V, with type IV being dominant in both MRSA and MRSE. SCCmec subtype IVa was the most common in MRSE (n = 16), while three subtypes (IVa, IVc, IVl) were detected in MRSA. ACME-arcA was detected in only S. epidermidis (23.8%; same rate in mecA-positive and negative isolates), while PVL genes were not detected in any staphylococcal isolates. ACME of S. epidermidis isolates was differentiated into three types (I, I′, II), with ACME-II being dominant in methicillin-susceptible isolates.
Frequencies of Isolates with mecA (SCCmec Type) and Arginine Catabolic Mobile Element Among Different Staphylococcal Species
ACME, arginine catabolic mobile element; NT, non-typeable.
The nine MRSA isolates were assigned to four coa types (Ia, IIa, IIIa, VIIb), among which coa type IIa was found in four isolates (Table 2). Two isolates with SCCmecIVl were classified into coa type IIIa. Among the 11 coa types detected for MSSA, types IVa and VIIb were the most common, followed by Xa, IIIa, and Vb. Genotypes (ST and agr type), prevalence of toxins/virulence factors/adhesins, and drug resistance genes with antimicrobial susceptibility patterns were analyzed for all the MRSA isolates and 11 MSSA with coa types IIa, IIIa, IVa, VIIb, and Xa (Table 3). MRSA was classified into five STs (ST1, ST5, ST8, ST89, and ST120) with either of four agr types. ST8 was identified in both MRSA and MSSA isolates with coa-IIIa/agr-I. All the MRSA isolates were resistant to AMP and FOX, and harbored blaZ, while five and three isolates with erm gene and aminoglycoside-modifying enzyme gene (aac(6′)-Ie-aph(2″)-Ia) showed resistance to ERY and GEN, respectively. In contrast, MSSA was generally more susceptible to antimicrobials with less numbers of resistance genes than MRSA.
Staphylocoagulase (coa) Genotypes and SCCmec Types in Staphylococcus aureus Isolates
NT, non-typeable.
Genetic Characteristics, Antimicrobial Resistance Profiles, Prevalence of Drug Resistance Genes, and Virulence Factor Genes in the 20 MRSA/MSSA Isolates Analyzed
Resistance to individual antimicrobial agent was judged according to the guidelines of Clinical Laboratory Standards Institute (CLISI). For antimicrobials whose resistance is not defined by CLISI guidelines, EUCAST breakpoints (Staphylococcus spp.: FOF, >32 μg/mL) and the following definitions (MIC) were employed to determine resistance to S. aureus: ABK, >4 μg/mL. None of the strains showed resistance to cefazoline, cefmetazole, flomoxef, arbekacin, imipenem, minocycline, vancomycin, teicoplanin, linezolid, and trimethoprim-sulfamethoxazole.
The following genes were undetectable in any of the strains: tet(K), tet(L), tet(M), ermB, aph(3') IIIa.
The following genes were detected in all strains: hla, hlb, hld, icaA, icaD, ebpS. A variant of ebpS (ebpS-v) was detected in F-17 and Y-74. The following genes were not detected in any strain: see, set, ses, lukM, etb, scn, bap.
AMP, ampicillin; CLI, clindamycin; ERY, erythromycin: FOF, fosfomycin; FOX, cefoxitin; GEN, gentamycin; LVX, levofloxacin; NT, non-typeable; OXA, oxacillin.
All the MRSA isolates had two or more enterotoxin genes. Four MRSA isolates (coa-IIa, VIIb) and seven MSSA isolates (coa-IIa, IVa, VIIa) harbored enterotoxin gene cluster (egc; seg-sei-sem-sen-seo-seu). Four ST5/ST8 MRSA and three ST30 MSSA had tsst-1 whose sequences were identical to that of ST8 CA-MRSA strain NN50 (GenBank accession no. AB679717). Two ST8 SCCmec-IVl MRSA with tsst-1 possessed also sasL, a gene encoding a novel cell surface protein, 37 which was detected also in an ST8 MSSA isolate (F54-1). A variant of ebpS (ebpS-v) lacking internal 180-nucleotide sequence was identified in an MRSA isolate F17 (ST120, CC121) and MSSA isolate Y74 (ST45). Phylogenetically, ebpS-v of ST120 and ST45 isolates clustered with those of ST121 strains reported previously in lineage III (Supplementary Fig. S1). 38 As found in ST121 strain Y12 reported in Myanmar, 35 deduced amino acid (aa) sequences of F17/Y74 EbpS revealed internal deletion of 60 aa and insertion of alanine at position 97 compared with intact EbpS (Supplementary Fig. S2). MSSA isolate Y86-2, which belongs to coa-IIa/ST2895 (CC25), had etd-edinB. Deduced amino acid sequence of exfoliative toxin D exhibited one amino acid difference from those reported previously for strains TY114 39 and DK-B3 36 (Supplementary Fig. S3), while edinB showed identical sequences to these strains (data not shown).
Antimicrobial resistance and prevalence of resistance genes were analyzed for 14 MRSE isolates with different SCCmec types with or without ACME, and a single isolate of methicillin-resistant S. warneri (Table 4). All the isolates harbored blaZ and showed resistance to OXA and AMP, and most isolates were resistant to ERY, while resistance to GEN, FOF, LVX, and CLI was found in some isolates. Relatedness of resistance profile of the MR-CNS to SCCmec types, the presence of ACME and its genotype was not evident.
Genetic Characteristics, Antimicrobial Resistance Profiles, and Prevalence of Drug Resistance Genes in the 15 Methicillin-Resistant Coagulase-Negative Staphylococcus Isolates
Resistance to individual antimicrobial agent was judged according to the guidelines of Clinical Laboratory Standards Institute (CLISI). For antimicrobials whose resistance is not defined by CLISI guidelines, EUCAST breakpoints (Staphylococcus spp.: FOF, >32 μg/mL) and the following definition (MIC) was employed to determine resistance for CNS: ABK, >4 μg/mL. None of the strains showed resistance to cefazoline, cefmetazole, flomoxef, arbekacin, imipenem, minocycline, vancomycin, teicoplanin, linezolid, and trimethoprim-sulfamethoxazole.
The following genes were undetectable in any of the strains: tet(K), tet(L), tet(M), ermA, ermB, ermC, aph(3')-IIIa.
Discussion
The present study first revealed the prevalence and genetic traits of MRSA and MR-CNS in oral cavity of healthy Japanese children. While nasal colonization of MRSA in humans has been investigated by numerous studies, only a limited number of reports are available for MRSA from oral cavity and pharynx.13,16,19,40 In studies conducted in dental hospitals, overall prevalence of MRSA was 9% in oral cavity and perioral region of patients in the United Kingdom, 16 and 2–13% in tongue swabs of healthy children aside from dental diseases in Japan. 19 Among healthy school children in Italy, MRSA was detected from nasal isolates, but not from oropharyngeal isolates. 40 Our present study, showing the isolation rate of S. aureus (27.2%; 143/526) comparable with that of the Italian study (25.9%), 40 revealed that the overall MRSA prevalence was 1.7%, which may be slightly lower than those reported for nasal colonization in healthy children: 2.3% (95% CI = 1.8–2.7) 10 and 3.9% (95% CI = 1.8–6.1) 11 from meta-analyses. Isolation rate of S. aureus is also described as higher from nasal cavity than oropharynx. 40 The higher rates of S. aureus/MRSA in nasal cavity are considered to be due to more exposures to open air than oral cavity, as well as the presence of non-persistent, transient carriers. 7 However, in a specific population, higher prevalence and longitudinal colonization (1-year period) of S. aureus in oropharynx than nare were observed. 41 Therefore, it remains to be elucidated whether oral colonization represents persistent career state of S. aureus/MRSA.
Type IV SCCmec was dominant in MRSA and MRSE in the present study, as reported for those from nasal cavity.10,18 In addition, common genotypes (ST, SCCmec, etc.) of locally predominant HA-MRSA are also frequently detected for nasal-colonizing isolates, for example, SCCmecIV-ST59 in Taiwan, 42 SCCmecIV-ST72 in Korea, 43 and ST5-SCCmecII in Japan, 17 suggesting the transmission of endemic MRSA from health care settings to the community. In the present study, however, uncommon genetic traits for Japanese HA-MRSA, that is, ST5/SCCmecIVc/coa-II/agr-II and ST8/SCCmecIVl/coa-IIIa/agr-I, were identified.
In Japan, MRSA with genotypes ST5/SCCmecIIa/coa-IIa, a pandemic clone designated “New York/Japan clone,,” 44 has been predominant in hospitals and community.29,34 In contrast, ST5/SCCmecIV belongs to a genotype of “pediatric clone” (USA800) that has been distributed to mainly the United States and South America,44,45 and its detection was very rare in our previous studies in Japan.29,34,46 Nevertheless, a recent study has shown that ST5/SCCmecIV was the most frequent genotype in CA-MRSA in a northern prefecture in Japan, 47 and several MRSA isolates with ST5/SCCmecIV/agr-II were isolated from feces of healthy neonates in Spain. 48 Identification of ST5/SCCmecIV MRSA in healthy children in our study may suggest an affinity of this clone to children and/or changing epidemiology of CA-MRSA in Japan. The presence of egc and tsst-1 in this clone may also pose a concern about pathogenic potential with increased virulence.
The PVL-negative ST8-MRSA-SCCmec IVl clone is designated ST8 CA-MRSA/J, which is characterized as coa-III/agr-I, having mostly tsst-1, sec, sel, and sep, and a novel cell wall-anchored surface protein (SasL or CWASP/W) encoded by sasL (spj) located in J1 region of SCCmec IVl.37,49,50 The ST8 CA-MRSA/J clone has been detected in central and western Japan and Hong Kong since 2003 and was also identified among CA-MRSA in northern Japan in our previous study. 29 Among hospitalized infants, nasal colonization of this clone was reported in Tokyo. 51 In our present study, the ST8 CA-MRSA/J clone was first isolated from healthy children, suggesting its prevalence in the community. The two ST8/SCCmec IVl isolates harbored sec3, sei, sel, and tsst-1, but slight difference was found in the presence of drug resistance genes and leukocidin/haemolysin genes. Interestingly, ST8 MSSA/coa-III/agr-I isolate with sasL was detected, suggesting its possible involvement in the occurrence of CA-MRSA/J. Prevalence and genetic diversity of this clone should be carefully monitored in both health care settings and community.
EbpS is one of the adhesins on cell surface that binds to host cellular matrix factors involved in biofilm formation, and it is produced by most MRSA. 52 A variant of EbpS gene (ebpS-v) with internal 180-bp deletion was detected in MRSA (ST120) and MSSA (ST45) isolates in the present study. Phylogenetically, these ebpS-v genes clustered with those of ST121 strains, 38 which has been described as a hypervirulent clone disseminated globally. 53 However, the pathogenic role of the variant of EbpS still remains undetermined, because strains with ST121 lineage have exfoliative toxin A and B as main virulence factors. It was also noted that etd-edinB were first identified in MSSA from a healthy child. The presence of variant adhesin, and exfoliative toxin as well as enterotoxin genes and tsst-1, in the isolates in the present study may underscore the potential significance of S. aureus colonizing in oral cavity.
In a Japanese study, MR-CNS was isolated from nasal cavity far more frequently than MRSA, and its predominant species was S. epidermidis having mostly SCCmec IV, 18 as observed in the present study. Previously, we reported a higher detection rate of ACME (45.8%) in clinical isolates 33 than the present study (23.8%; S. epidermidis). Although the prevalence of ACME in MRSE and methicillin-susceptible S. epidermidis (MSSE) was the same, the frequent detection of ACME types I and I′ (previously referred to as ΔI) in MRSE was commonly observed, suggesting relevance of colonizing and clinical isolates of MRSE. ACME is a genomic island consisting of arcA/opp3 gene clusters and considered to be related to increased adaptability to the host. ACME-positive MRSE has been suggested to persist in host, having a potential to be an opportunistic pathogen, and act also as a reservoir of SCCmec and ACME for S. aureus.
The present study has two limitations. First, to obtain isolates of broad staphylococcal species, we used mannitol salt agar, which is less sensitive in detecting S. aureus than more sensitive media, for example, chromogenic media (CHROMagar Staph aureus). 54 Therefore, the number and rates of S. aureus might have been potentially underestimated. Second, we analyzed only oral cavity for colonization of staphylococcus, but other body sites were not studied. The present results do not represent colonization status of the whole body, including skin and nasal cavity. Further study may be required for isolates from different body sites to understand colonization status and genetic relatedness of isolates.
In conclusion, the present study revealed the prevalence of MRSA and MR-CNS in the oral cavity in Japanese children. The identification of emerging MRSA clones ST5/SCCmec IV and ST8/SCCmec IVl, and ACME-positive MRSE, highlights the need for continuous surveillance and molecular epidemiological analysis of staphylococci colonizing in oral cavity.
Footnotes
Acknowledgment
This study was supported in part by JSPS (Japan Society for the Promotion of Science) KAKENHI Grant No. 25463260, 17H04664, and 18K10054.
Disclosure Statement
The authors of this article have no commercial associations that might create a conflict of interest in connection with the submitted article.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
