Identification of Novel Mobile Genetic Elements Associated with Resistance to Macrolide and Lincosamide in Streptococcus dysgalactiae subsp. equisimilis

1. INTRODUCTION

The human β-hemolytic streptococci (BHS) include streptococcus pyogenes (group A streptococci, GAS), streptococcus agalactiae (group B streptococci, GBS), and Streptococcus dysgalactiae subsp. equisimilis (group C and G streptococci, SDSE). SDSE colonizes the skin or mucosal surfaces, such as the respiratory tract, gastrointestinal tract, or vagina and causes a variety of diseases including invasive diseases. During the recent years the incidence of invasive SDSE diseases has increased, and in some geographic regions the rate of SDSE diseases was higher than those of GAS and GBS [1-3].

At present, penicillin is considered as drug of choice for treatment of β-hemolytic streptococcal (BHS) infections. Additionally, macrolides and clindamycin can be used for treatment of the patients intolerant to β-lactam antibiotics. Importantly, combined antibiotic therapy can reduce mortality in case of severe disease manifestations [4], probably through the abrogation of toxin synthesis [5]. Significant variations in resistance to macrolides, lincosamides (such as clindamycin) and streptogramin B (MLSB) in BHS were reported. MLSB resistance in β-hemolytic streptococci is mostly provided by the target modification enzymes encoded by erm genes, resulting in the resistance to all three classes (MLSB), or mef genes encoding efflux pumps specific for macrolides [6]. Lincosamide resistance genes lsa and lnu also occur in streptococci [7]. In GAS and GBS, all these resistance genes are located within mobile genetic elements. Their spreading among streptococci occurs by either horizontal genetic transfer or clonal expansion [6-9], however, their dissemination has not been extensively studied. In vitro studies showed the existence of conjugal transfer of integrative conjugative elements (ICEs) harboring resistance genes between streptococcal species [8].

As a consequence, the goal of the present study was to analyze the current status of antimicrobial resistance in SDSE. In addition, genome sequencing was performed to identify antibiotic resistance genes and their possible association with mobile genetic elements.

2. MATERIALS AND METHODS

3. RESULTS

As previously published, a total of 152 β-hemolytic streptococci were isolated from 1359 children of 7-10 years old from different regions of Vietnam in the period 2012-2015 [10]. GAS were isolated from 49 of 1359 (3,6%) examined children, while group C and G streptococci were isolated from 8 (0,6%) and 75 (5,5%) children, respectively [10]. Using cpn60 gene based PCR approach [20] for differentiation of the species within groups C and G, a total of 9 SDSE strains (1 – group C, and 8 – group G) were identified. Other identified strains included group C S. anginosus (4 isolates), S. parasanguinis (1 isolate), S. сonstellatus (1 isolate), S. gardonii (1 isolate), and group G S. anginosus (54 isolates), S. сonstellatus (4 isolates), S. parasanguinis (3 isolates), S.sanguinis (3 isolates), S. mitis (2 isolates), S. australis (1 isolate).

3.4. Mobile Elements Involved in Macrolide and Lincosamide Resistance

3.4.1. Erythromycin-resistant SDSE Isolates Carrying the mef Genes

We identified 3 isolates harboring different alleles of mef gene (mefA/E/G). As expected, in tetracycline-susceptible isolate T122 PCR assay demonstrated that mef(E) gene was located within mega (macrolide efflux genetic assembly) element, which has been previously described for BHS and Streptococcus salivarius [21, 22].

In SDSE strain NT15 (GenBank accession number JAFELF000000000) the mef(G) gene was located within the prophage-associated genetic element (54,700 bp). This genetic element consisted of 54 open reading frames (ORFs) and was chimeric in nature; it appeared due to the insertion of a Tn1207.1-related transposon into streptococcal prophage. BLASTN analysis revealed the high similarity of this prophage to other prophages, especially Tn1207.3 (52,491 bp) or Ф10394.4 (58,761 bp) discovered in tetracycline-susceptible GAS isolates [23, 24]. In both GAS isolates and SDSE NT15 strain these phages were integrated into the same chromosomal gene, comEC (Fig. 1). A Ф10394.4-like element in NT15, entitled ФNT15, harbored mef(G) instead of mef(A) and the additional 5 kb region containing unique lsa(E) and lnu(B) genes. The lnu(B) gene is involved in lincosamide modification/inactivation; and the lsa(E) encoding ABC transporter is responsible for active efflux of lincosamides, streptogramins A, and pleuromutilins. BLASTN analysis revealed that this region in SDSE strain NT15 is similar to the lnu(B)-containing sequences of S. agalactiae (JQ861959), Staphylococcus aureus (JX560992), Enterococcus faecalis (AF408195) and swine Enterococcus faecium isolate (KF421157.1) [25]. In addition, in NT15 isolate tetracycline susceptibility is associated with the presence of silent tet(S) gene.

The complete sequence of SDSE V123 isolate (GenBank accession number JAFELD000000000) revealed that tetracycline resistance is provided by tet(O) gene, and tet(O) was linked with mef(A) within phage-like element. It is highly similar to well-known Ф46.1 of S. pyogenes, which resulted from the insertion of transposon Tn1207.1 into a prophage [26]. However, an addition of IS1595 element and gene lnu(C), which confer resistance to lincosamides, was found (Fig. 2). The Ф46.1-like phage was found to be integrated into a 23S rRNA uracil methyltransferase gene.

3.4.2. Erythromycin-resistant SDSE Isolates Carrying erm(B) Gene

Among the 6 macrolide resistant isolates, 3 isolates had the erm(B) gene and exhibited constitutive resistant phenotype. According to the results of PCR analysis, the erm(B) was carried by the Тn917 in isolate V63 [27]. In another erm(B)-positive isolate (HF112), which was also resistant to tetracycline, erm(B) was associated with tet(M) on Tn6002 [28]. The third erm(B)-positive isolate, B82, was chosen for genome sequencing. The following analysis of B82 (GenBank accession number JAFELE000000000) revealed the presence of a 45.4-kb element. At present, this element was not found in other SDSE strains, but it has a partial similarity with SDSE strain WCHSDSE-1, which caused the streptococcal outbreak in China in 2013 [29]. Similar element, which contains the tet(M) gene instead of erm(B), is present in Filifactor alocis ATCC35896. This bacterium can cause periodontal diseases. In place of the 1.5-kb fragment containing erm(B) in SDSE, the strain F. alocis has a 14.1-kb fragment with tet(M) gene.

Thе 45.4-kb element of isolate B82 appears to be a conjugative transposon because it contains some specific genes, i.e., site-specific recombinase gene, relaxase-encoding gene (nicK), an origin of transfer (oriT), conjugal transfer coupling protein gene.

Transposition of this element is characterized by the presence of 3-bp direct repeat suggesting that the putative transposons were truly transposable elements (Fig. 3). This transposon was inserted between the genes encoding hypothetical proteins (SDSE167_0576 and SDSE167_0577 in strain 167). Additionally, Tn916 was found with gene tet(T) instead of tet(M) gene.

4. DISCUSSION

Based on the genome comparison, the SDSE which belong to group C and G streptococci, are closely related to GAS [2, 30, 31]. Almost all SDSE used in this study belonged to GGS. It corresponds to results of the previous studies, wherein Lancefield group G was found to be the most predominant among the human-recovered SDSE [32-34].

The emm genotyping was successfully performed for all 9 SDSE isolates, and 6 emm types were discovered. Four of them (stG6, stG480, stC5345, stC36) have already been found in SDSE isolated in North America, Europe and Australia [35], reflecting the successful dissemination of certain emm types in human.

Previously the certain correlation between stG480 and stG6 types and invasive infections was demonstrated [36]. However, in our study stG480 and stG6 strains were non-invasive.

Results of this study demonstrated that beta-lactams are useful for treatment of SDSE infections that correlates with other studies [3, 4, 37]. However, during the recent years the MIC for penicillin was slightly increased for some GAS and SDSE (0.12 and of 0.25 μg/ml), respectively [38]. Tetracycline resistance was noted in 6 of 9 (67%) isolates. The macrolide and lincosamide resistance in SDSE were high, 67% and 45%, respectively. Inducible clindamycin-resistant phenotype (iMLSB) was not found among erythromycin-resistant SDSE. Macrolide resistance in SDSE occurs world-wide, e.g., in Hong Kong (24%), in the USA (19%), in Europe (16%) [12, 38]. Furthermore, the percentage of clindamycin-resistant SDSE is significantly higher than in GAS. It should be taken into account during the treatment of streptococcal toxic shock syndrome caused by SDSE. In case of tonsillopharyngitis, fluoroquinolones can be considered as a second choice because the prevalence of fluoroquinolone-resistance among β-hemolytic streptococci is still low (1%), and just a few publications reported fluoroquinolone-resistance in SDSE and GAS. Resistance rate to tetracycline is higher than 60%, and for this reason it can no longer be used for empiral treatment of SDSE infections [38].

The resistance genes of macrolides, lincosamides, tetracyclines, and fluoroquinolones were examined in this study. As result, the presence of erm(A) (0%), erm(B) (33%), mef(A/E/G) (33%) differed from those of human-recovered SDSE in China: erm(A) (0%), erm(B) (78.6%), mef(A/E) (5.4%), Korea: erm(A) (4.3%), erm(B) (20.3%), mef(A) (8.7%), and Japan: erm(A) (15.5%), erm(B) (11.3%), mef(A) (2.8%) [39, 40]. Among the tetracycline resistance genes, tet(M) was the most common in this study, but tet(O), tet(S) and tet(T) were discovered (each by one). Other publications confirmed that tet(M) was more predominant than tet(O), e.g., in China and Korea (tet(M): 73.2% and 29.0%, and tet(O): 5.4% and 1.4%, respectively) in human SDSE [12, 39, 40]. Mutations in the quinolone resistance determining regions (QRDR) of gyr(A) or par(C) are considered as a major mechanism of fluoroquinolone resistance [12]. In this study, resistance to fluoroquinolone antibiotics in 4 isolates was associated with the presence gene pat(B) mediating antibiotic efflux pump.

The association of macrolide resistant determinants with mobile genetic elements was previously demonstrated for major streptococcal pathogens that explains spreading of antibiotic resistance. The resistance phenotype has been reported to be transferable by conjugation [8, 41] and transduction [42, 43]. In our study most of erythromycin-resistant strains were additionally resistant to clindamycin (phenotype MLSB). In SDSE strains we detected a number of novel genetic elements, including the new Ф10394.4-like element characterized by the insertion of lnu(B)/lsa(E)-containing sequence; and the new phi-m46.1-like prophage with insertion of lnu(C)-containing sequence, which seems to be acquired from other bacterial species [44]. The novel putative transposon carrying erm(B) gene was identified in SDSE, which is likely to be acquired from the bacteria of oral microflora. It represents one more example of genetic exchange of antibiotic resistance between gram positive cocci.

CONCLUSION

Identification of the novel antibiotic resistance genes in SDSE indicates the necessity of monitoring of antibiotic resistance spreading and gene transfer in this bacterium.

REFERENCES