Decreased expression of femXAB genes and fnbp mediated biofilm pathways in OS-MRSA clinical isolates

Methicillin-Resistant Staphylococcus aureus (MRSA) is a significant threat to human health. Additionally, biofilm forming bacteria becomes more tolerant to antibiotics and act as bacterial reservoir leading to chronic infection. In this study, we characterised the antibiotic susceptibility, biofilm production and sequence types (ST) of 74 randomly selected clinical isolates of S. aureus causing ocular infections. Antibiotic susceptibility revealed 74% of the isolates as resistant against one or two antibiotics, followed by 16% multidrug-resistant isolates (MDR), and 10% sensitive. The isolates were characterized as MRSA (n = 15), Methicillin-sensitive S. aureus (MSSA, n = 48) and oxacillin susceptible mecA positive S. aureus (OS-MRSA, n = 11) based on oxacillin susceptibility, mecA gene PCR and PBP2a agglutination test. All OS-MRSA would have been misclassified as MSSA on the basis of susceptibility test. Therefore, both phenotypic and genotypic tests should be included to prevent strain misrepresentation. In addition, in-depth studies for understanding the emerging OS-MRSA phenotype is required. The role of fem XAB gene family has been earlier reported in OS-MRSA phenotype. Sequence analysis of the fem XAB genes revealed mutations in fem × (K3R, H11N, N18H and I51V) and fem B (L410F) genes. The fem XAB genes were also found down-regulated in OS-MRSA isolates in comparison to MRSA. In OS-MRSA isolates, biofilm formation is regulated by fibronectin binding proteins A & B. Molecular typing of the isolates revealed genetic diversity. All the isolates produced biofilm, however, MRSA isolates with strong biofilm phenotype represent a worrisome situation and may even result in treatment failure.

the prevalence, virulence, resistance mechanism of OS-MRSA strains and their impact on the current treatment regimen. Another imperative should be the molecular characterisation of existing clinical isolates.
In addition to the intriguing phenotype of S. aureus, their capacity to form biofilms on biotic or abiotic surfaces especially, medical devices, catheter, ocular implants and corneal stroma further magnify the problem 13 . Biofilm formation is an innate component of the prokaryotic life cycle which ensures the survival of the bacterial cells by fending off host immune attacks as well as external therapeutic/antibiotics and also promote dispersal to other uninfected parts thereby maintaining a chronic infection 14 . We hence need to identify and understand the molecular characteristics of such bacterial populations to unarm them. In the current study, we are trying to understand the molecular characteristic, resistance and biofilm profiles of MRSA and MSSA strains isolated from ocular infections and to determine the prevalence of OS-MRSA which is not well understood.

Materials and Methods
Isolation of S. aureus from clinical samples. Clinical isolates of randomly selected S. aureus were received from L.V Prasad Eye Institute (LVPEI), Hyderabad, India. Further, species identification and characterisation of S. aureus cultures were performed using various biochemical and molecular tests as discussed earlier 3 . Following an overnight incubation in trypticase soy broth (TSB) at 37 °C, the cultures were streaked on Baird Parker agar plates (30% egg yolk emulsion and 1% potassium tellurite) and after that onto mannitol salt agar plates. Yellow colonies on the plates were considered as S. aureus and inoculated further in TSB for other biochemical tests such as catalase, Latex agglutination, Coagulase test and Gram staining. Species confirmation was also done by sequence analysis of the amplified PCR product of 16S rRNA gene as mentioned below. PCR amplification. Genomic DNA was isolated using the Wizard genomic kit (Promega) as per manufacturer's instructions. Briefly, the overnight grown culture in TSB media was centrifuged at 3200 × g for 10 minutes and washed twice in PBS. Cells were pelleted and resuspended in 500 µl lysis buffer (50 mM EDTA) and 50 µl of lysozyme (20 mg/ml) and incubated at 37 °C for 2 hrs. PCR was performed for 16S rRNA, mecA, mecC and SCCmec types and agr typing as described previously 3,15-18 . Multilocus sequence typing (MLST). Internal fragments of seven housekeeping genes, Carbamate kinase (arc), Shikimate dehydrogenase (aroE), Glycerol kinase (glpF), Guanylate kinase (gmk), Phosphate acetyltransferase (pta), Triosephosphate isomerase (tpi), and Acetyl coenzyme A acetyltransferase (yqiL) were amplified using the approach described on http://www.mlst.net/. The sequence types (STs) were designated by comparing the data of seven sequenced internal fragments of each strain with the S. aureus MLST database (http://www. mlst.net/).
Global optimal eBURST (goeBURST) of identified STs, was analysed using the PHYLOViZ online software 19 . S. aureus MLST public database was used to make the minimum spanning tree based on the goeBURST analysis for the identified STs. The nLocus Variant (nLV) was kept to 2 to keep all STs with a distance equal to or below 2 connected in the tree.
Antibiotic susceptibility profile. The antibiotic susceptibility profile was determined using the disk diffusion method and microbroth dilution assay as per CLSI guidelines with minor modifications as mentioned before 18 . Briefly, the overnight grown culture was adjusted to 0.5 McFarland standard (Himedia, Mumbai, India) and was streaked onto Mueller-Hinton Agar (Himedia, Mumbai, India) plates for disk diffusion while Mueller-Hinton broth was used for micro-broth dilution assay. The disk diffusion assay was done against tetracycline (30 µg), gentamicin (10 µg), rifampicin (5 µg), erythromycin (15 µg), ciprofloxacin (10 µg), clindamycin (2 µg), and cefoxitin (30 µg). Resazurin dye based microbroth dilution assay was performed against oxacillin, vancomycin and linezolid 18 . ATCC 29213 (methicillin sensitive control strain) and ATCC 700699 (methicillin-resistant control strain with reduced vancomycin susceptibility) were used as a control for these assays 3,18 . Clinical isolates showing resistance against three or more class of antibiotics were described as multidrug-resistant (MDR), resistance against one or two class of antibiotic was referred as resistant and sensitivity towards all antibiotics was mentioned as pan-sensitive in the result section.
ATCC 43300 and ATCC 29213 were used as controls.
Sequence analysis of fem XAB family. The fem XAB gene family was amplified as described earlier 20 . The primers used for amplification of fem genes were as follows: femX (For-GCCATGGAAAAGATGCATATCAC and Rev-CTCGAGTTTTCGTTTTAATTTACGAG), femA (For-GGATCCATGAAGTTTACAAATTTAACAGC Rev-CTCGAGAAAAAT TCTGTCT T TAAGT T T T T TAAG) and femB (GGATCCATGAAGT T T ACAAATTTAACAGC and Rev-CTCGAGTTTCTTTAATTTTTTACGTAATTTATCG) 20 . The amplified PCR product was sequenced using forward and reverse primer along with the internal primers for all the respective genes (Internal femA-GATCCGTGCTACAAATTCGT; internal femX-CACCAATTGATAAAAATGATG and internal femB-CAACTGAG TATGATACATCGA). The data was analysed for mutation using DNA star software Expression analysis of fem XAB and biofilm-related genes. RNA was isolated using the Macherey Nagel RNA isolation kit as per the manufacturer's instructions (NucleoSpin ® RNA, 740955.10). Approximately from 2.5 µg of RNA, cDNA was synthesised using the Clontech cDNA synthesis kit (PrimeScript ™ 1st strand cDNA Synthesis Kit, 6110B). SYBR green-based relative gene expression analysis was carried out using the comparative 2 −∆∆ C t method using the ABI real-time detection system (7500, ABI). Real-time PCR was performed for www.nature.com/scientificreports www.nature.com/scientificreports/ fem XAB genes and biofilm-related genes (icaADBC and fnbp A & B). The gyrB gene was used as internal control in the study. All the primers used in the study are mentioned in Table 1. All the experiments were repeated thrice.
Biofilm profiling by crystal violet assay. The biofilm forming capacity of each strain was determined using crystal violet (CV) assay as described previously, ATCC 29213 and ATCC 33592 were used as controls 21 . Briefly, the overnight culture was diluted in TSB (1:100), and dispensed in 96 well plate (200 µl/well). The wells with only broth were treated as the control. After incubation for 18-24 hrs at 37°C, the plate was washed 3 times with PBS to remove planktonic cells or media particles. Methanol was added to each well for 15 minutes and then air-dried for 30 minutes. 0.1% crystal violet solution was added to each well and incubated at room temperature for 20 minutes followed by washing using distilled water. After washing, 33% acetic acid was added to each well and absorbance was taken at 590 nm. Mean absorbance values of each sample was calculated and compared with the mean values of controls. All experiments were repeated thrice in quadruplicates. The data were analyzed as described before 22 .
The following conditions were used to interpret the results:
Agr type I and III were predominant. Agr typing of the 74 clinical isolates showed type I (n = 32) and III (n = 18) to be the major types, followed by types IV (n = 12) and II (n = 10). One isolate displayed mixed type (I + II), and one was non-typeable.
Clinical isolates of S. aureus exhibited strong, moderate and weak biofilm forming capacity. Biofilm forming capacity of the 74 clinical isolates was determined using CV assay. The majority (55%) of the isolates exhibited moderate biofilm forming capacity followed by 39% isolates which manifested strong biofilm production. Only 6% of the isolates formed weak biofilms. All the isolates formed black colonies on congo red agar plates.

Characterisation of clinical isolates of MRSA.
Antibiotic susceptibility assays against oxacillin and cefoxitin determined 15 isolates as methicillin-resistant (Table 2). Twelve MRSA clinical isolates were multi-drug resistant, and the remaining 3 were resistant to one or two classes of antibiotics. In 2 isolates, the mecA gene was not amplified, but they exhibited oxacillin resistance (MIC >4 mg/L). Therefore, the SCCmec typing of the 13 isolates revealed 7 as type III, followed by 3 isolates as type V, 2 isolates as IVc and 1 isolate was non-typeable (Table S1). Molecular typing of MRSA isolates revealed ST239 (33%) as the major sequence type followed by ST217 (26.66%). ST772 (13.33%), ST1 (6.66%), ST8 (6.66%), ST9 (6.66%), ST5 (6.66%) represent smaller proportions. Agr typing of these strains showed 66.67% isolates belonged to type I followed by 26.67% of type II and 6.67% of type III.
Majority of the MRSA isolates (60%) formed strong biofilm, followed by 33% moderate and only 6.6% of the isolates formed weak biofilm as determined by CV assay.  (Table S1). Eighty one percent of the isolates were found resistant to one or two classes of antibiotic, and 18.2% were pan-sensitive. Sequence analysis of fem XAB genes in OS-MRSA isolates showed mutations in fem × (K3R, H11N, N18H and I51V) and fem B (L410F) gene, however, no mutations were found in fem A. Gene expression analysis revealed decreased expression of all the fem genes in comparison to MRSA (Fig. 2).

Discussion
The present study represents the molecular variability and antibiotic susceptibility profiles of S. aureus clinical isolates causing ocular infections. Antibiotic susceptibility determined majority of the isolates as resistant to one or two classes of antibiotics, followed by MDR and pan sensitive clinical isolates. These clinical isolates were further characterized as methicillin-sensitive or resistant based on the phenotypic and genotypic tests. MRSA in ocular infections had been previously reported from countries such as Taiwan (57.6%), China (52.8%), India (43%), United States (41.6%) and Brazil (9.9%) 2,23-26 . We found 17.6% MRSA (n = 15) in our study and two MRSA without the mecA gene, which is in line with previous findings where 9.8% of MRSA isolates were mecA negative 27 .
The present finding also highlighted a 14.86% prevalence rate of OS-MRSA clinical isolates which were not reported earlier from ocular infections in India. OS-MRSA has surfaced across the globe and reported from human and animal infections 2,3 . We also found 9 oxacillin susceptible clinical isolates which did not produce PBP2a but were mecA PCR positive 28 . OS-MRSA can be misinterpreted if only either phenotypic or genotypic test is used and may have serious implications 29 . Earlier reports of OS-MRSA have shown mutations in the fem XAB genes 20,30 .   (K3R, H11N, and N18H) and the femB (L410F) gene. An additional new mutation in the femX gene (I51V) was identified. The downregulation of fem XAB genes was seen, in contrast to earlier observation where no change has been reported 20 . Methicillin resistance is caused by changes associated with genes involved in cell wall metabolism and the fem gene family are also involved in cell wall biosynthesis, hence, decreased expression of the fem genes may lead to oxacillin sensitivity in case of OS-MRSA isolates 31 . Increased expression of the femA gene has been reported in high-level MRSA isolates which were comparatively found to be reduced in low-level MRSA followed by MSSA isolates 32 . Therefore, our study indicates the decreased expression of the fem genes, may be responsible for hypersusceptibility towards oxacillin in OS-MRSA isolate 20 . However, the underlying mechanism operative in OS-MRSA is still obscure and need investigation.
Further, molecular typing of the clinical isolates showed diversity among them. We found ST239 is a predominant lineage in MRSA isolates followed by ST217 and ST772 [33][34][35] . The ST239 clinical isolates were mostly reported in hospital-associated infections from India and other Asian countries, while ST217 and ST772 were reported from community-acquired MRSA infections 36,37 . ST772 is mainly reported from the southern part of India and is being considered to be replacing ST239 in hospitals 37 . Recently, ST217 has been identified as a new emerging clone in both hospital and community-acquired infections from India 37 . Along with their resistant phenotype, the add-on capacity to form biofilms assist their prolonged survival by enhanced tolerance against host mechanisms or antibiotics 14,38 . Therefore, we also studied the biofilm forming capacity of the clinical isolates with variable methicillin phenotype in an attempt to establish any link with strain lineages and phenotype.
The majority of the MRSA formed strong biofilm (n = 9), and predominantly belonged to ST239 (n = 5) with SCCmec type III (n = 4) and V (n = 1). ST239 MRSA causing ocular infections have been previously reported to be from Taiwan (29.4%), although its biofilm phenotype was not evaluated 26 . Strong biofilm forming ST239 isolates were also reported from other countries like Korea, Brazil, and India [39][40][41] . We also came across ST772 MRSA with SCCmec type V, (n = 2) which formed a strong biofilm and has been considered as an emerging threat responsible for disease outbreaks 42 . One of the mecA negative MRSA belonging to ST5, also formed a strong biofilm. The present findings suggest that MRSA isolates with SCCmec types III and V may be dominant biofilm producers. A recent study shows that SCCmec types III & V display higher adhesion and are less hydrophilic in comparison to SCCmec type II 40 . However, there can be a difference in the observed phenomenon when compared in vivo. Biofilm formation have been known to be dependent on icaADBC operon in MSSA isolate, whereas in MRSA, fnb genes are involved 43 . The biofilm mechanism in OS-MRSA isolates has not been studied earlier.
Our data reveal that biofilm formation in OS-MRSA isolates is icaADBC independent and regulated via fnb genes.
Agr is studied mostly as virulence locus but recently shown to be involved in biofilm formation 44 . Genotyping and transcriptional analysis to establish a direct correlation between biofilm formation and agr have been done in recent times 44 . In contradiction to the expected decreased biofilm forming capacity of isolates with agr type I, we found isolates (5 MRSA; 2MSSA) with agr type I forming strong biofilms 44 . Therefore, it would be interesting to study such isolates and understand their mechanism of biofilm production. Isolates with type III agr are known to induce moderate biofilms, however, our type III agr isolates showed strong biofilm phenotype 44 . Similar to previous findings, MRSA isolates belonging to ST772/SCCmec type V and agr type II and a mecA negative, ST5 isolates produced strong biofilms 44 . Our findings interestingly support the recent reports where methicillin resistance effects icaABCD and agr phenotype thereby altering expected biofilm phenotype 43,45 . Also 12 out of 15 MRSA isolates were MDR out of which 8 were forming strong biofilms. A recent publication also stated that 87% of hospital-acquired ocular MRSA infection were MDR 46 . This is an alarming situation because treating a MDR bacteria forming strong biofilm makes the treatment extremely difficult.
OS-MRSA isolates with strong biofilm (n = 5) belonged to variable ST types, but all belonged to agr type III. There seems to be hardly any study on the biofilm phenotype of the newly emerging class of OS-MRSA isolates, and hence this could shed some light on our understanding of these isolates. OS-MRSA isolates which could be misinterpreted as sensitive isolates with an add-on ability to form strong biofilm may pose a dangerous threat to public health. Another fact to note was that MSSA isolates belonging to agr IV and III appeared to form a strong biofilm.
Our data suggest no association of biofilm intensity with their antibiotic susceptibility phenotype. However, the variable biofilm forming ability might be linked with the differences in their metabolic and biosynthetic processes as evident in the case of Enterococcus faecalis isolates 47 . Hence, it will be interesting to do comparative studies exploring the biofilm mechanism among strong, moderate and weak biofilm isolates which may help us in identifying drug targets against biofilm-forming bacteria. However, any resistant strain with strong biofilm forming capacity will be an immense threat to patients with ocular infections and in other diseases too. Further, with the increasing prevalence of OS-MRSA strains in various infections across the globe, studies are warranted to understand this peculiar phenotype.