Staphylococcus aureus is a leading cause of diseases such as skin and soft tissue infections, pneumonia, bloodstream infections, osteomyelitis and endocarditis, as well as toxin-mediated syndromes like toxic shock and food poisoning 1 2 . It has developed resistance to a wide range of antimicrobial drugs, which complicates the treatment of infections. In particular, methicillin-resistant S. aureus (MRSA) has become a notorious etiologic agent for a wide variety of infections and it is one of the most important nosocomial pathogens worldwide 3 4 5 6 . Methicillin-susceptible S. aureus (MSSA) become MRSA through the acquisition and insertion into their genomes of a large DNA fragment known as staphylococcal chromosome cassette mec (SCCmec), which contains the methicillin resistance determinant, mecA 7 . Several variants of SCCmec have been described, which differ with respect to the composition of their recombinase genes and mec gene complex (containing the mecA gene) 8 9 .

In the developing world, mortality associated with severe S. aureus infections far exceeds that in developed countries 10 11 . Recent studies have identified S. aureus as the main etiological agent of many infections in sub-Saharan Africa 12 13 14 15 16 , and a number of investigations have reported that S. aureus is among the most frequently encountered bacterial species in microbiology laboratories in Nigeria 17 18 19 20 21 22 . However, data on the molecular epidemiology of this pathogen in Nigeria is very limited. Recent reports have indicated that the prevalence of hospital-associated MRSA varies in health care institutions 23 24 . A community-associated MRSA clone with a unique resistance profile has also been reported from South-West Nigeria 25 . To understand and potentially predict trends in antibiotic-resistance patterns and to establish adequate infection control programs, it is crucial to understand the local epidemiology of S. aureus in Nigeria. Knowledge of the local antimicrobial resistance patterns of bacterial pathogens is essential to guide empirical and pathogen specific therapy. The threat of antibiotic-resistant bacteria has initiated studies on the nature of genes encoding resistance and the mechanism by which these genes spread and evolve. Antibiotic susceptibility testing of S. aureus in Nigeria is based on phenotypic testing especially the disk diffusion technique but recent studies have relied on the PCR detection of the mecA gene for the identification and confirmation of MRSA 23 24 25 26 . However, no information is available on the nature of antibiotic resistance genes of S. aureus in Nigeria. Our present study provides baseline information on antibiotic resistance and molecular epidemiology of MSSA and MRSA in Nigeria.

There was excellent correlation between the broth microdilution method and detection of the genetic determinants by multiplex PCR for S. aureus resistance to erythromycin, gentamicin, methicillin and tetracycline (Tables 2 and 3). About 55% (11 MRSA, 27 MSSA) and 70% (10 MRSA, 39 MSSA) of the S. aureus isolates were resistant to tetracycline and cotrimoxazole, and as previous studies from South-West Nigeria had shown 23 25 , it appears that there is a high proportion of S. aureus isolates resistant to these antibiotics in Nigeria. Tetracycline and cotrimoxazole historically had wide clinical application, is inexpensive, orally administered and available from diverse sources where they are sold with or without prescription in Nigeria. Moreover, they are listed in many developing countries as among the antibacterial agents that have been rendered ineffective, or for which there are serious concerns regarding bacterial resistance 28 . It appears that misuse and overuse of these antibiotics could have contributed to this trend in Nigeria. Therefore, to prevent treatment failures in the absence of data on antibiotic susceptibility testing, public enlightenment on the ineffectiveness of these antibiotics against S. aureus infections, and the enactment of effective drug policies in Nigeria are urgently needed. The predominant mechanism of trimethoprim resistance in S. aureus appears to be mutation of the dihydrofolate reductase (DHFR), which is selected even when trimethoprim is used in combination with sulfamethoxazole 29 . In this study, all the trimethoprim-resistant S. aureus isolates were dfrA negative suggesting that mutation of the dihydrofolate reductase (DHFR) is responsible for resistance. Isolates resistant to tetracycline carried either one of the resistance genes tetK or tetM (Tables 2 and 3), which mediate resistance through active drug efflux or ribosomoal protection mechanisms, respectively. This is the first study that provides baseline information on the nature of the antibiotic resistance genes from S. aureus isolates in Nigeria. The multiplex PCR assay was easy to perform, cost-effective and assisted in the prompt characterization of the resistance genes stated above. It could be adapted for use by clinical scientists in the referral health care institutions regarding the antibiotics administered and the prevalent resistance determinants in Nigeria.

The proportion of PVL-positive isolates among MSSA was high (40%). Most of the PVL-positive MSSA isolates were obtained from wound infections and classified in clonal complexes CC1, CC30, CC121 and ST152. Moreover, the detection of the seh gene in CC1 isolates and the identification of the etd gene in ST25 and CC80 isolates is in agreement with previous reports 27 30 31 32 . PVL is frequently associated with severe and recurrent skin and soft-tissue infections (SSTIs) and has previously been found in S. aureus isolates from various complexes. In particular, PVL-producing MSSA affiliated to CC121 are known to be common in many countries on all continents 30 33 34 , including Nigeria, Togo and South Africa in sub-Saharan Africa 25 30 35 . PVL-positive ST152 was the predominant clone in a study recently conducted in North-Eastern Nigeria 24 and it was the second most prevalent clone in a carriage study from a West-African country (Mali) 36 . Furthermore, the high prevalence of PVL positive MSSA ST152 emerging in the community as well as in hospitals in West Africa has also been described 31 . Hence, ST152 seems to be widespread and frequent in West Africa, whereas it is comparatively rare elsewhere 33 37 , in contrast to many other clonal complexes that display worldwide occurrence. The luk-PV genes are carried on mobile genetic elements (prophages), which may be incorporated into S. aureus lineages through horizontal transfer, either before or after acquisition of the mecA gene 38 . The high proportion of PVL-positive MSSA observed in this study indicate that conditions that increase the risk of inter-individual transmission (e.g skin-to-skin and skin-to-fomite contacts) could represent important routes of spread in the various hospital settings. Contact with colonized and/or infected individuals as well as contaminated fomites in the spread of PVL positive S. aureus have been described as risk factors for community-associated MRSA 39 . Moreover, the detection of PVL-positive MSSA ST152 from members of one family and their relatives with skin infections at the Canary Island underscore the pathogenic and contagious nature of this clone 40 . More detailed investigations on the prevalence of PVL-positive S. aureus are needed in Africa with respect to (i) nasal carriage of S. aureus in the hospitals and community, (ii) cross-transmission from post-operative wound infections acquired during hospital stay, and (iii) cross-transmission from patients admitted to the health institutions for treatment of an SSTI acquired in the community. The detection of PVL-positive MSSA isolates from the various health institutions, indicating their wide geographical distribution, could pose serious problem in the future as potential reservoirs for resistance and virulence factors, and could lead to the emergence and spread of PVL-positive MRSA clones in Nigeria causing severe infections. This could have important implications for the enactment of effective infection control guidelines.

MRSA has become a major public health problem worldwide and recent reports have indicated that the prevalence of hospital-associated MRSA (based on the detection of the mecA gene) in health care institutions in Nigeria may vary from 1.5% to 20% 23 24 25 . All the MRSA isolates obtained from Maiduguri (North-East Nigeria) had the same spa type (t037) and MLST profile (ST241), identical to isolates from the same region that had been investigated in a previous study 24 . Another study 25 also reported that the clone was identified in a hospital in Ibadan (South-Western Nigeria). ST241 is a single locus variant (slv) of the ST239 clone, which is prevalent in South East Asia and has also been reported from Europe, the Americas 41 , and several countries in Africa 6 42 43 44 . The multi-resistant nature of this MRSA clone could be explained by the presence of several resistance genes in the SCCmec cassette and it was recently shown to have spread across several continents since the 1960s 41 . MRSA ST239 demonstrating low level resistance to glycopeptides have been reported recently in Russia 45 and New Zealand 46 . In contrast, in South-Western Nigeria, we identified more diversity among the MRSA isolates. In three different hospitals in this region, we observed several different clones of MRSA that can be distinguished on the basis of MLST, SCCmec typing and spa typing, and displayed distinct antimicrobial resistance profiles (Table 2).

This study showed that the combination of susceptibility testing and various molecular methods provided useful information on the antibiotic resistance and molecular diversity of S. aureus in Nigeria. Although the number of S. aureus isolates available for our investigation and epidemiological information was limited, the high proportion of PVL-positive MSSA observed in this study indicate that adequate measures are needed to curtail the spread and establishment of MRSA and PVL-positive MSSA clones in Nigerian health care institutions.

AOS, WW, BS, FL and UN conceived the study. KO, SA and OO participated in the preliminary identification of the isolates, AOS carried out the phenotypic and molecular characterization of the isolates. All authors read and approved the final version of the manuscript.