Molecular Characterisation of Antibiotic Resistance in Staphylococcus haemolyticus isolates from Chennai, South india

Staphylococcus haemolyticus is a highly resistant opportunistic pathogen having close genomic relatedness with other virulent species of staphylococci. However, compared to Staphylococcus aureus and Staphylococcus epidermidis , little is known about the resistance genes of S. haemolyticus . the purpose of this study was to characterise antibiotic resistance genes in S. haemolyticus isolates. Standard microbiological techniques were used to identify and confirm 104 S. haemolyticus isolates included in the study. Antibiotic susceptibility testing and D-test were performed, followed by PCR amplification of various resistance determinants (mecA, ermA, ermC, msrA, aac(6′)-Ie-aph(2″), ant(4′)-Ia,aph(3′)-IIIa, tetK, tetM, dfrA, fusB, fusC, fusD and mupA). Methicillin resistance was observed in 93.3% of study isolates. the maximum number of isolates showed resistance to erythromycin (n=79, 76%), followed by ciprofloxacin (n=66, 63.5%) and cotrimoxazole (n=58, 55.8%). in the D-test, 8 isolates showed inducible (iMlS B ) and 11 showed constitutive (cMlS B ) resistance. Among the resistance determinants, mecA gene (93.3%) was the most prevalent, followed by dfrA (50.5%). Furthermore, aac(6’)-Ie-aph(2’’) and aph(3’)-IIIa combination was observed in 26.9% of isolates, and aac(6’)-Ie-aph(2’’) alone was present in 3.8% of isolates. Among the study isolates, 17.3% exhibited tetK gene, whereas only 1% exhibited tetM; a combination of tetK and tetM was observed in one isolate. the fusB and fusC were present in 11.5% of isolates, and 12.5% of the isolates were positive for mupA. in conclusion, the present study underlines the concern of increasing antibiotic resistance among S. haemolyticus isolates. Avoiding misuse/overuse of antibiotics along with continuous surveillance programs can reduce the spread of antibiotic resistance.

iNtRODUCtiON Staphylococcus haemoly ticus i s an opportunistic pathogen and the second most frequently isolated coagulase-negative staphylococci (CoNS), with high degree of genetic relatedness to Staphylococcus aureus and Staphylococcus epidermidis. 1 It has an average nucleotide sequence similarity of 75% with S. aureus and S. epidermidis. Thus, there is a high probability that S. haemolyticus could act as a reservoir of resistance genes and disseminate them, thereby posing a threat of antibiotic resistance in hospital setup. 2 Another unique feature of S. haemolyticus genome is that it undergoes constant rearrangement due to the presence of various insertion sequences. 3 Empirical treatment with broad-spectrum antibiotics and genomic diversity due to frequent genomic rearrangements have led to the selection of multiresistant strains that slowly replace susceptible strains in hospitals. These findings are consistent with the fact that among CoNS, S. haemolyticus possesses the highest level of resistance to commonly used antibiotics. 4,5 Despite being the second most commonly isolated CoNS, there is a paucity of available data regarding its antimicrobial resistance. Hence, this study aimed to elucidate the antibiotic pattern and molecular characterisation of resistance genes in S. haemolyticus isolates from various clinical samples.

Study isolates
A total of 104 S. haemolyticus isolates were collected from a tertiary care centre in Chennai during March 2016-January 2017 and used for further research. Initial sampling and identification of CoNS using standard sampling methods were performed by technical experts from the Microbiology Laboratory of the tertiary care centre. Further identification and species confirmation were performed as described below. The sources of the collected isolates are given in Table 1.

identification and Confirmation of S. haemolyticus
Staphylococcus isolates were initially identified using standard microbiological techniques such as Gram staining, catalase test, oxidation-fermentation test, coagulase test (tube and slide coagulase), DNase test, and mannitol fermentation on mannitol salt agar. Further species confirmation was performed using alkaline phosphatase, ornithine decarboxylase, urease, novobiocin and polymyxin B susceptibility, and carbohydrate (maltose, mannose, trehalose and sucrose) fermentation tests. 6

(ii) Detection of Methicillin Resistance
The Kirby-Bauer disc diffusion method was performed to screen for resistance to methicillin using cefoxitin antibiotic disc (30 µg). 7 For S. haemolyticus, a zone diameter of ≤ 24 mm was considered methicillin resistant.

(iii) Detection of inducible and Constitutive Clindamycin Resistance
The D-test was performed to detect inducible clindamycin resistance of the isolates, and the results were interpreted according to the CLSI guidelines, 2015. 8 Briefly, erythromycin (15 μg) disc and clindamycin (2 μg) disc were placed 15 mm apart (measured from the edge of the disc) in a previously swabbed lawn culture of the isolates with growth matching the turbidity of 0.5 McFarland standard. The zone of inhibition was observed the following day after incubation at 37°C. Blunting of the clindamycin zone near the erythromycin antibiotic disc (D-shape) showed inducible macrolide-lincosamide-streptogramin B (iMLS B ) resistance phenotype, whereas resistance to both erythromycin and clindamycin showed constitutive resistance (cMLS B ).

Genotypic Methods DNA extraction and Polymerase Chain Reaction
DNA extraction from all the study isolates was performed using the boiling lysis method; the extracted DNA was amplified for each of the resistance genes by polymerase chain reaction (PCR) in Mastercycler® Gradient (Eppendorf, Hamburg, Germany). The PCR products were then subjected to agarose gel electrophoresis, and the respective bands were visualised using Gel Logic 212 PRO imaging system. Analysis was carried out using the Carestream Molecular Imaging Software (Carestream Health, Incorporated, USA).

(i) Molecular Confirmation of S. haemolyticus
PCR amplification of the mvaA gene was performed for molecular confirmation of S. haemolyticus isolates. 9

Statistics
GraphPad Prism version 9 was employed to perform Fischer's exact test. The association between antibiotic resistance and its respective resistance determinants was tested (p ≤ 0.05 was considered statistically significant).

ReSUltS
All the phenotypically identified S. haemolyticus isolates (n=104) were confirmed by the presence of mvaA gene.

(ii) Methicillin Resistance
The cefoxitin disc diffusion result revealed that majority of S. haemolyticus isolates were resistant to methicillin (n=97, 93.3%).

(iii) Detection of inducible and Constitutive Clindamycin Resistance
Nineteen of the 104 isolates were non-susceptible to clindamycin, of which 14 were resistant, and the remaining five showed intermediate susceptibility. Inducible clindamycin resistance (iMLS B ) was observed in eight isolates, and the remaining 11 isolates showed constitutive resistance (cMLS B ).

Genotypic Screening of Antibiotic Resistant Genes
S. haemolyticus isolates (n=97, 93.3%) expressed the mecA gene, indicating resistance to methicillin. A high number of isolates were non-   24%), of which both genes were present in one of the isolates (1%). The tetK gene alone was present in 18 isolates (17.3%), and the tetM gene alone was present in one isolate (1%). Fusidic acid resistance was phenotypically observed in 24 isolates, of which 12 (11.5%) were positive for the fusB gene, and the remaining 12 (11.5%) were positive for the fusC gene; however, the fusD gene was absent. High level of mupirocin resistance was observed in 13 isolates; all 13 isolates (12.5%) were positive for the mupA gene. The representative gel pictures of the resistance genes screened in the present study are given in Figure 3. The complete antibiotic resistance profiles of the study isolates with both resistant phenotypes and genotypes are listed in Table 3. The correlation between the isolate source and antibiotic resistance was also determined ( Figure 2). Strains isolated from the skin and soft tissue infections exhibited a comparatively high percentage of resistance to all antibiotics, followed by isolates from genital tract samples, such as high vaginal swab and semen, which showed increased antibiotic resistance. The sample-wise distribution of antibiotic resistance and its determinants is given in Figure 2 and Table 4 and 5.

Statistical Analysis
No significant difference was observed between the antibiotic resistance and its determinants.

DiSCUSSiON
S. haemolyticus has been well known for its resistance to multiple antibiotics, which is also evident from the fact that it acquired methicillin resistance much earlier than other species of staphylococci. 1 In this study, 93.3% of the isolates were methicillin resistant, and all the resistant isolates exhibited the mecA gene; however, in the study conducted by Barros et al., 4 among 64 methicillin-resistant S. haemolyticus isolates, 87% showed the mecA gene. In another study by Silva et al., 16 the mecA gene was present in 26 of 27 methicillin-resistant isolates. This proves that the phenotypic method cefoxitin disc diffusion is economical and can be reliably performed in a limited setup for surveillance of methicillin resistance.
Identification of a high percentage (85%) of multi-resistant strains was consistent with the results of other studies. Multidrug-resistant (MDR) strains in the study were defined as those "acquired non-susceptibility to at least one agent in three or more antimicrobial categories" (Magiorakos et al.). 17 S. haemolyticus genome undergoes constant rearrangements, which is attributable to its multidrug resistance.
Indiscriminate and inappropriate use of broad-spectrum antibiotics has significantly increased the incidence of antibiotic resistance. This was reflected in the study results. Erythromycin non-susceptibility was observed in the maximum number of isolates (76%), followed by ciprofloxacin (63.5%) and cotrimoxazole (55.8%), the three being the most commonly prescribed broadspectrum drugs in clinical setting. Surprisingly, last resort and the least prescribed drugs such as vancomycin and linezolid have shown 100% susceptibility. Similar results were reported by Krzyminska et al. 18 MLS antibiotics, though chemically different, have similar resistance mechanism of ribosomal modification encoded by the erythromycin ribosome methylation (erm) gene. 19,20 MLS antibiotics are clinically significant in the treatment of Gram-positive infections. Hence, cross-resistance between them is a clinical concern. 21 In this study, among erythromycinresistant isolates, the MS B phenotype was   Trimethoprim resistance is either chromosomally mediated that occurs due to mutations in the dfrG gene encoding dihydrofolate reductase (DHFR), the enzyme involved in the folate pathway, or plasmid mediated that occurs due to variants of DHFR having low affinity for trimethoprim. 22 These DHFR variants are encoded by the dfrA, dfrD and dfrK genes, of which the dfrA gene is the most common. In the present study, 54/58 cotrimoxazole-resistant isolates exhibited the dfrA gene. The results were in concordance with the study by Aggarwal et al., 23 ; they screened three trimethoprim resistance genes from S. aureus isolates, of which the majority of isolates (45/74) carried the dfrA gene. In contrast, Manoharan et al. 5 reported that among S. haemolyticus study isolates, 89.7% of cotrimoxazole-resistant isolates were dfrG-positive, and the dfrA gene in combination with other genes, including dfrD and dfrG, was present only in 2% of the isolates.
A m i n o g l y c o s i d e r e s i s t a n c e i n staphylococci is due to target site modification, leading to inactivation of the drug caused by aminoglycoside modifying enzymes. 24 Plasmid mediated genes {aac(6')-Ie-aph(2''), aph(3')-IIIa and ant(4')} encoding three commonly found aminoglycoside modifying enzymes {AAC(6')/ APH(2'), APH(3')-III, and ANT (4')-I, respectively} were screened in this study. A high number of aminoglycoside-resistant isolates (26.9%) exhibited a combination of the aac(6')-Ie-aph(2'') and aph(3')-IIIa genes in the present study. These findings disagree with those of other published studies that revealed the presence of aac(6')-Ieaph(2'') alone rather than in combination. Both studies showed the lowest prevalence of the ant(4') gene, whereas it was completely absent in the present study. 25,18 Tetracycline resistance in staphylococci is either due to active efflux by acquiring the plasmid-mediated genes tetK and tetL or the chromosomal resistance genes tetM and tetO. 26 In this study, the most frequently observed genes, tetK and tetM, were screened. Among the 25 tetracycline non-susceptible isolates, 72% were positive for the tetK gene. Unlike the findings of the present study, the tetM gene was predominant (67%) and the tetK gene was present in only 33% of the isolates in the study conducted by Duran et al., 27 However, the findings of the study by Manoharan et al., 5 were similar to those of the present study, with 91.5% prevalence of the tetK gene. Apart from the resistant isolates, susceptible isolates exhibiting resistance genes were also observed in both previous studies but was absent in the present study.
The most common resistance mechanism to fusidic acid is protecting the target site by the genes encoding the fusB family of proteins, thereby preventing the translocation of elongation factor G (EF-G) from the ribosome, leading to inhibition of protein synthesis. Casanteira et al., 18 compared the occurrence rates of fusidic acid resistance in Australia, Canada and the USA, and observed that the prevalence of fusidic acid resistance in CoNS was the highest in Canada (20%), followed by Australia (10.8%) and the USA (7.2%). In this study, the occurrence of fusidic acid resistance in Chennai, southern India, was 23% among S. haemolyticus study isolates. Half of the isolates exhibited prevalence of the fusB gene, and the remaining 50% exhibited the fusC gene. However, other studies have demonstrated a higher prevalence of the fusB gene than that of the fusC gene. 15,28 Mupirocin is a bacteriostatic antibiotic that inhibits protein synthesis. Among the two phenotypes, high level of resistance is mediated by plasmid carrying the iles2 or mupA gene that encodes a novel tRNA synthetase. Among the study isolates, 12.5% exhibited highlevel phenotypic resistance to mupirocin, all of which carried the mupA gene. These findings are consistent with those of other studies. 29,30 Universal methicillin-resistant Staphylococcus aureus (MRSA) decolonisation protocol followed in hospitals is an important reason for increased resistance to mupirocin. Thus, stabilising the use of mupirocin with proper surveillance and targetbased decolonisation may be of great help in controlling mupirocin resistance.
The isolate source was correlated with the resistance phenotypes and genotypes. It is well known that staphylococci normally inhabit the skin and mucous membranes in humans. Hence, the predominant S. haemolyticus isolates having the highest resistance to various antibiotics were from the skin and soft tissues. These findings were consistent with those of Palestine and Ethiopia. 31,32 Interestingly, all resistant genotypes and their combinations were observed in isolates from the skin and soft tissues in the present study. In addition to skin and soft tissue infection samples, genital tract samples, such as high vaginal swab and semen, also exhibited high level of antibiotic resistance. Other samples (urine, ascitic, and sputum) were low in number to draw conclusions.

CONClUSiON
A high percentage of antibiotic resistance in opportunistic pathogens such as S. haemolyticus is a concern, as it may lead to treatment failure, prolonged hospital stay and increased mortality rate. In addition, there is a greater risk of disseminating resistance genes to other virulent species of staphylococci, making them increasingly arduous in hospital setup.