Association between SCCmec types and antimicrobial resistance in clinical MRSA isolates

MRSA is the causative agent of serious infections. MRSA isolates carry mecA gene which confers resistance to all β-lactams, markedly limiting the therapeutic options. SCCmec typing enables strain-based MRSA identi�cation. This study aimed to identify the prevalent SCCmec types among clinical MRSA isolates in Alexandria, Egypt, and their association with antibiotic resistance. One hundred MRSA clinical isolates were identi�ed and tested for susceptibility to antibiotics. SCCmec typing was done by PCR using previously published primers. Results Various patterns were SCCmec types and III displayed the highest resistance, while SCCmec IV showed the least resistance. There was a signi�cant association between SCCmec types and antibiotic resistance (p = 0.02-0.001).


Background
MRSA has been recognized as a causative agent of a diversity of serious hospital and community-acquired infections, particularly pyogenic infections of the skin. It can also cause infections associated with medical instruments such as central-line-associated bloodstream infection. The rising threat of antibiotic resistance in Methicillin-resistant Staphylococcus aureus (MRSA) has been an impetus for research to unveil any associations that might have prognostic implications and may partially guide empirical therapeutic decisions. [1].
Clinically, resistance against many antibiotic classes is considered one of the characteristic features of MRSA infection, as it carries an altered form of penicillin-binding protein; PBP2a, which renders it less sensitive to most semisynthetic penicillin drugs. This protein is expressed via an acquired gene named mecA, which is carried within a highly conserved mobile genetic element called the staphylococcal cassette chromosome mec (SCCmec). [2] [3].
MRSA has been known as a healthcare-associated (HA) infectious agent with high predominance all over the world since its emergence in 1960 [4]. It was highly implicated in multidrug-resistant healthcare-associated infections [5], Nowadays, CA-MRSA healthcare-associated outbreaks have been recorded in several countries worldwide, causing remarkable changes in the epidemiological distribution of MRSA worldwide and implying an increasingly di cult distinction between CA-MRSA and HA-MRSA based on the aforementioned description [7]. Hence, the true prevalence of this community-dwelling organism may be underestimated or exaggerated [8]. Accordingly, it is now preferred to establish a strain-based de nition for CA-MRSA because of its distinct epidemiology, genetic pro le, antibiotic resistance pattern, and clinical presentation [6].
Bacterial typing is an indispensable epidemiologic tool that enables the identi cation of bacteria at the strain level, elaborating clonal relationships between them. It may be done phenotypically by methods such as antibiogram typing or serotyping. Alternatively, bacteria may be typed more precisely by genotypic methods, based on analyzing variations in the genetic elements [9].
Staphylococcal Chromosomal Cassette mec (SCCmec) typing is one of the well-recognized MRSA genotyping methods. It is based on the identi cation of the SCCmec element, which is carried on a genomic island that can easily transfer horizontally between strains by the site-speci c action of two recombinases. SCC consists of 3 components; (i) mec gene complex, (ii)Ccr (cassette chromosome recombinase) gene complex, and (iii) J regions [10]. mec gene complex encompasses the mec gene, insertion sequences (IS), and the regulatory components mecR1 (signal transducer protein) and mecI (repressor protein). There are ve known classes of the mec gene complex (A -E) based on variations in the assortments of insertion sequences and regulatory regions [10].
A uni ed nomenclature scheme for the cassette types has been established. SCCmec is the outcome of integrating the mec gene complex classes with the ccr gene complex types to categorize SCCmec components into types. There are thirteen different forms of SCCmec (I-XIII) found in MRSA strains so far [6].
SCCmec typing has recently become part of the well-recognized nomenclature of MRSA, which enables getting information about SCCmec-typed MRSA isolates. SCCmec typing can be performed by Whole-genome sequencing and subsequent data analysis using bioinformatics tools such as SCCmecFinder. However, the conventional method of SCCmec typing using conventional PCR remains to be more widely applied, particularly in low-income countries. [11] This study aimed to identify the SCCmec types of MRSA strains causing clinical infection and/or nasal colonization state and their associated antibiotic resistance. The study was performed on 100 non-repetitive MRSA strains randomly collected from different types of clinical specimens submitted for routine culture and antimicrobial susceptibility testing to the Microbiology laboratory of the Medical Research Institute, Alexandria University. MRSA strains were collected over the period from September 2019 to March 2021, disregarding the source of infection being healthcare-associated or community-acquired.

Results
The 100 MRSA isolates collected during the study period included 60 isolates from pyogenic skin infection including abscess aspirates and wound swabs, 14 from bloodstream infection, 9 from lower respiratory tract infection, and 5 from urinary tract infection, in addition to 12 colonizing isolates collected from nasal swabs.
The 100 MRSA isolates showed different antimicrobial resistance patterns, the two most prominent resistance patterns were resistance to gentamicin, doxycycline, and Tetracycline, as well as resistance to gentamicin only (17%) each. This was followed by resistance to gentamicin and Tetracycline (12%). On the other hand,14% of the isolates were sensitive to all tested antibiotics other than cefoxitin (Table 2). and only the typed isolates were con rmed by SYBR-Green real-time PCR to ensure the speci city of ampli cation by melting curve analysis. Out of the 100 MRSA isolates, only 75 (%) were successfully SCCmec-typed using previously published primers speci c to each of SCCmec-types I-XII (Table 3).

Statistical correlation between SCCmec types and clinical condition
A statistically signi cant association was found between SCCmec types and pyogenic skin infection ( MC p<0.001), as SCCmec type V MRSA was the most prominent among all isolates from pyogenic skin lesions, isolated from 24/45 (53%) of the lesions. Type V was also the most prominent among isolates from lower respiratory tract infection 3/7 (43%), as well as urinary tract infection 3/5 (60%). As for bloodstream infection, type II was the most prominent 4/12 (33%), followed by type V 3/12 (25%). No statistically signi cant association was found between SCCmec types and different types of clinical infection other than pyogenic skin lesions. In nasal colonization, however, type IV was the most prominent 3/6 (50%), with a statistically signi cant association, MC p = 0.049 (Table 4).

Statistical correlation between SCCmec types and antibiotic resistance
Concerning Antimicrobial resistance, SCCmec types II and III had the highest resistance. SCCmec type II was resistant mainly to gentamicin, macrolides (p = 0.002 − 0.001) followed by uoroquinolones (p < 0.001). SCCmec type III showed high resistance to uoroquinolones (p < 0.001) followed by gentamicin and Tetracyclines (p < 0.001). On the other hand, SCCmec type IV showed the least resistance to antibiotics followed by SCCmec type V and VI (Table 5). Intermediate susceptibility to Linezolid was detected in 3 isolates, that were of SCCmec types III, V, and VI. Most of the isolates with the same SCCmec type displayed the same pattern of resistance to antibiotics. For instance, simultaneous resistance to gentamicin and tetracycline was displayed by 8 isolates typed as SCCmec type V, also resistance to gentamicin, doxycycline, and tetracycline was displayed by 5 isolates of SCCmec type V and 7 isolates of SCCmec type VI (Table 2).

Discussion
MRSA infection is of global concern worldwide. Epidemiologic studies about MRSA rely on the use of standard nomenclature that identi es the prevailing strains at the chromosomal level [11]. SCCmec typing is one of the internationally recognized MRSA typing methods. [12,13].
Pyogenic skin infection is the most common clinical presentation of MRSA infection. Sixty percent of the isolates in this study were collected from pyogenic skin lesions, followed by bloodstream infection (14%), lower respiratory tract infection (9%), and urinary tract infection (5%). Another study about MRSA in Egyptian hospital laboratories also reported a similar proportion of isolates from pyogenic lesions (64.3%) and bloodstream infection (9.5%) [14]. Similarly, it was reported in Kuwait that the majority of MRSA isolates were from wounds and pus, followed by blood [15]. Also, in the United Arab Emirates, pyogenic lesions and bloodstream infection were the sources of 73.4% and 15.2% of MRSA isolates, respectively [16].
Seventy-ve percent of our isolates were SCCmec typeable by PCR. Several studies worldwide employed SCCmec typing by PCR for identi cation of the prevailing SCCmec types in their regions and reported varying degrees of typeability that were all less than 100%. For instance, a study in Denmark reported 98% typeability by multiplex PCR [17]. Another study in Portugal reported 97.4% typeability [18]. A more recent study in Palestine reported a typeability of 96.4% [19]. Also in Alexandria, Mansoura, and Cairo, Egypt, the reported typeability was 90%, 94% and 88.8%, respectively. [20]. [21]. [22]. A lower percentage of typeability (77%) was reported by a study in Rwanda [23], which was close to the ndings of the current study.
The high percentage of isolation of SCCmec type V (45.3%) followed by SCCmec type IV (16%) and types II and III (13.3% each) among the 75 typeable MRSA isolates in our study was in accordance with the ndings of several studies, worldwide. A recent study in a tertiary hospital in Cairo, Egypt, reported that half of their MRSA isolates were SCCmec type V (50%) followed by SCCmec type VI (17%) [22]. Also, a study carried out in four University Teaching Hospitals in Iran, reported that SCCmec type V was the most prevalent (66.7%) among their clinical MRSA isolates [24]. Moreover, other studies conducted in Armenia [25], and in Iran [26] stated that, SCCmec types V and VI were the most identi ed among MRSA isolated from hospitals.
Consistently, a study in Saudi Arabia reported the detection of SCCmec type IV in 77.3% of their isolates, followed by SCCmec type V (13.2%), and type III (9.4%) [12]. Similarly, a study in Kuwait reported that the majority of their isolates belonged to SCCmec type IV (39.5%) followed by SCCmec type III (34.4%) [15]. In Africa, a study assessed the SCCmec types in correlation with spa types and reported that isolates of the common spa types harbored SCCmec types IV followed by type V, with a minority harboring SCCmec type I. [27] Conversely, a study in Alexandria conducted on 72 MRSA isolates collected over 4 months in 2015, reported that 57% of their MRSA isolates harbored SCCmec type III and only 11% were of SCCmec type V [20]. The discrepancy between their most prevalent SCCmec type (type III) and our results (type V) may be attributed to the fact that the study was conducted 4 years earlier, and it focused mainly on typing of MRSA isolates collected from healthcare-associated infections which represented 80% of their typed isolates. On the other hand, our study intentionally disregarded the source of infection, and typing was performed on randomly selected isolates including nasal colonizers, to allow for a better representation of the SCCmec types prevalent in Alexandria, Egypt.
SCCmec type I was not detected in any of our isolates. Despite being undetected in Egypt and nearby regions, a study on a small scale in Rwanda, reported the detection of SCCmec type I in 56% of the 39 MRSA isolates included in their study. They also reported that SCCmec type IV was the second most common type among their isolates (17.9%), while SCCmec types II and V were undetectable [23].
Apart from that, a study in Hungary stated that SCCmec type IV accounted for the vast majority of their MRSA isolates (66.7%), followed by SCCmec type II (23.5%), and SCCmec type I (9.2%). They reported that SCCmec type V was detected in only one isolate, while SCCmec types III and VI were not found [28].
The discrepancy in the distribution of SCCmec types reported from different geographic regions, and even from the same region at different points of time, can be attributed to the high plasticity of this region, and the limited capabilities of the conventional PCR detection method, in addition to the differences in the sensitivity and speci city of the primers used, which may eventually result in missed identi cation of some SCCmec types.
In the present study, SCCmec type V isolates were the most predominantly isolated (53%) from pyogenic skin lesions, with a statistically signi cant correlation (p < 0.001). This was in accordance with the ndings reported by a study in Mansoura University Hospital which stated that SCCmec type V is signi cantly associated with burns and abscesses, and of a moderate association with wound sources [21].
SCCmec IV showed the least resistance to antibiotics, while SCCmec types II and III displayed the highest resistance to antibiotics and were signi cantly associated with resistance to uoroquinolones (p < 0.001). The association between SCCmec type III and uoroquinolones resistance was in accordance with the ndings of previous studies in Egypt and Iran [20,29].
Similarly, in Hungary, it was reported that SCCmec type II is associated with the highest level of resistance to antibiotics while SCCmec type IV is associated with low resistance [28]. Also, a Russian study reported that Isolates carrying SCCmec type III demonstrated higher antibiotic resistance than SCCmec type IV [30].
The most common resistance patterns among our isolates were; resistance to gentamicin only, and simultaneous resistance to gentamicin, doxycycline, and Tetracycline, each detected in 17% of the isolates. Contrary to our ndings, a study conducted in a Hungarian tertiary care hospital reported that the most prevalent phenotype of resistance was to erythromycin, clindamycin, and cipro oxacin [28]. On the other hand, a study in Kuwait reported that a high proportion of their isolates was resistant to tetracycline, erythromycin, cipro oxacin, and trimethoprim/sulfamethoxazole [15].
Our isolates displayed very high resistance to gentamicin (71%), with no statistical difference between different SCCmec types. This was followed by resistance to tetracycline (44%). Resistance to uoroquinolones and macrolides was less (23-25%), while resistance to trimethoprim/sulfamethoxazole (10%) and Rifampicin (5%) was low. All isolates were susceptible to vancomycin, however, 3 isolates displayed intermediate susceptibility to Linezolid. This could be probably due to the over-prescription of this drug by physicians in Egypt.
In Spain, it was reported that cipro oxacin resistance was the highest (85%) in MRSA, followed by erythromycin resistance (65%), gentamicin resistance (35%), and tetracycline resistance (30%). All MRSA strains were susceptible to trimethoprim/sulfamethoxazole and rifampicin, which was not far from our susceptibility results for these 2 antibiotics [31]. Also, a study in Palestine reported that resistance to erythromycin in MRSA was 63.4%, and to cipro oxacin was 39.3%, with 18.8% resistance to trimethoprim/sulfamethoxazole [19].
Constitutive clindamycin resistance was displayed by 8% of our isolates, while 4% showed inducible resistance with a positive D-test. The percentage of clindamycin resistance was slightly higher in a study conducted in Spain which reported that 11.7% of their MRSA isolates have inducible clindamycin resistance [31]. Even higher percentages were reported in Kuwait, where the authors reported that inducible and constitutive clindamycin resistance among their MRSA isolates were 14.4% and 37.8%, respectively [15].

Page 14/24
The only SCCmec types detected by PCR were SCCmec II-VI, with high resistance to gentamicin among all types. SCCmec type V was the most prevalent and was signi cantly associated with pyogenic lesions and of relatively low resistance to antibiotics. SCCmec type IV was the least prevalent and showed the least resistance to antibiotics. There was a signi cant association between SCCmec types II and III and resistance to uoroquinolones. Macrolides resistance was signi cantly associated with SCCmec type II. Tetracyclines resistance was signi cantly associated with SCCmec type III.

Bacterial isolates
Primary isolation of 100 Staphylococcus aureus strains from clinical specimens was done by culture on Blood agar plates. Identi cation was done by colony morphology, and the characteristic microscopic morphology of Gramstained lms. This was further con rmed by the positive reaction to biochemical tests; namely, catalase test, slide coagulase test, tube coagulase test and mannitol fermentation [32].
S. aureus colonies were tested for methicillin resistance by the Kirby-Bauer method using cefoxitin disc (30 ug). Only cefoxitin-resistant isolates (≤ 21mm) after 16-18 hours were identi ed as MRSA and included in this study [33]. Subculture on ORSAB (Oxacillin Resistance Screening Agar Base ) and observation of the characteristic blue colonies of MRSA was also performed as a further con rmatory step for phenotypic identi cation [34].
The Kirby-Bauer disc diffusion method was used for antimicrobial susceptibility testing of the isolates to 14  A 10 µmolar working solution of each primer was prepared using DNase-free water. PCR reaction (25 µl) contained: 12.5 µl of MyTaq™ HS Red Mix (2x), 1µl of F primers (10 picomoles/µl), 1µl of R primers (10 picomoles/µl), 3 µl of DNA extract, and 7.5 µl of PCR grade water. Negative control was prepared by the addition of the same contents to the tubes with water placed instead of the extract. Conventional PCR ampli cation was carried out on Veriti Thermal Cycler (Applied Biosystems), using gene-speci c thermal cycling conditions.
All thermal pro les included one cycle of initial denaturation at 95℃ for 3 minutes, followed by 40 cycles of denaturation at 95℃, annealing at primer-speci c temperatures, then extension at 72℃ for 45 seconds, in addition to one cycle of nal extension at 72℃ for 1 minute (Table 3).
Detection of the ampli ed target genes was done using gel electrophoresis with 1.7% (w/v) agarose, carried out on Mupid-exU System gel electrophoresis equipment. The size of the amplicons was determined using a 100 bp DNA ladder (Thermoscienti c GeneRuler, US).

Real-time time PCR con rmation of typing results
Further con rmation of PCR amplicon speci city was done for typed isolates by SYBR Green real-time PCR followed by melting curve analysis. Real-time PCR was carried out on Agilent Stratagene MX 3000P Quantitative PCR System using SensiFAST™ SYBR Lo-ROX® master mix, with gene-speci c thermal cycling conditions. All thermal pro les started with an initial denaturation step (one cycle) at 95℃ for 3 minutes, followed by 40 cycles of denaturation at 95℃ for 5 seconds, annealing at primer-speci c temperatures, extension at 72℃ for 20 seconds, and eventually, one cycle of melting curve analysis at temperatures speci c to each primer pair (Table3).   Table 5 Correlation between SCCmec types and antibiotic resistance