Clindamycin Resistance among Methicillin Resistant Staphylococcus aureus Isolated from Human and Respective Household Swine in Greater Kabale Region—South Western Uganda

Introduction: S. aureus is recognized as the common cause of nosocomial and community-acquired infections. Macrolide-Lincosamide-Streptogramin B (MLS B ) is thought to be alternative therapies against MRSA infections. Clindamycin is the most favored agent because of exceptional pharmacokinetic characteristics. However, increasing resistance to clindamycin among MRSA strains is a serious challenge. The current study investigated the profile of clindamycin resistance among MRSA isolates from Humans, and their respective livestock (in particular swine) using D-test in greater Kabale region. Materials and Methods: Three hundred phenotypic MRSA isolates previously isolated from Humans and swine were confirmed by mecA PCR. We performed D-test using erythromycin (15 μg) and clindamycin (2 μg) discs in accordance to Clinical and Laboratory Standards Institute (CLSI) protocol. Results: Of all 300 MRSA isolates, 6% (n = 18) were sensitive to Erythromycin and Clindamycin (S). The rate of inducible clindamycin resistance (iMLSB) was 42% (n = 125) and 38% (n = 115) was resistance to both Erythromycin and clindamycin (cMLSB). However, 14%


Introduction
S. aureus is recognized as one of the most common organisms causing nosocomial and community-acquired infections worldwide. The emergence of multidrug resistant S. aureus strains, especially methicillin resistant S. aureus (MRSA), is of a particular concern. In Uganda, this has been largely attributed by empirical antibiotic prescriptions [1]. The cumulative MRSA problem is an indicator for urgent need for new antibiotics.
Macrolide-Lincosamide-Streptogramin B (MLS B ) antibiotics have been thought about as alternative solution to treat MRSA infections [2]. The most commonly used antibiotic in the MLSB group are the macrolides (e.g. erythromycin and azithromycin) and clindamycin which is a Lincosamide [3].
Macrolides act through inhibition of protein synthesis by binding irreversibly to the 23S ribosomal RNA (rRNA) on the bacterial 50S ribosomal subunit and subsequent disruption of the growing peptide chain by blocking translocation [4]. Lincosamide [e.g. clindamycin and lincomycin] bind to the 50S ribosomal subunit and prevent peptide elongation by interfering with the peptidyl transfer during protein synthesis [5]. Clindamycin is the most preferred agent because of exceptional pharmacokinetic characteristics [6] and is regularly used in the management of severe infections, caused by macrolide resistant S. aureus infections including MRSA [7]. However, increasing resistance to clindamycin among MRSA strains and other Staphylococcus is a serious challenge [8]. The expression of clindamycin resistance in Staphylococcus species can be constitutive or inducible [9] [10] through erm genes which codes for ribosomal methylases [11]. In addition, the resistance to the lincosamides (clindamycin), macrolides (erythromycin), and streptogramins (quinupristin/dalfopristin) is facilitated by three related genes, ermA, ermB, and ermC, that encode for erythromycin resistance methylases [12]. Methylase enzymes binds on to the ribosome resulting in a conformational change or modification in the ribosomal target and consequently, decreasing the ability of these drugs to bind to the ribosome [5] [13].
Inducible Clindamycin resistance (iMLSB) cannot be identified by standard methods of antibiotic susceptibility testing and failure to detection may result into treatment failure with Clindamycin [6]. Erythromycin-resistant staphylococci are routinely considered to be resistant to Clindamycin by clinicians [14], a phenomenon that is wrong that shuns Clindamycin prescription to patients infected with macrolide-resistant isolates that may be sensitive to Clindamycin. It is rational to routinely test for presence of iMLSB strains and this can be achieved cheaply by use of Erythromycin and clindamycin discs placed adjacent

Materials and Methods
This was cross-sectional study, where 300 phenotypic MRSA isolates previously isolated from humans (n = 200) and swine (n = 100) during the period of January 2015 to June 2016 and stored in glycerol (20%v/v) at −80˚C.

DNA Extraction
These isolates, were subjected to DNA extraction following Queipo et al. and Teeraputon et al., techniques [8] [17] by boiling using 100 µl of the bacterial suspension in 1.5 ml cryogenic vials (Eppendorf, Germany) followed by centrifugation at 3000 rpm for 15 minutes. The supernatant was removed, and the pellet suspended using molecular biology-grade water (Eppendorf, Germany) and re centrifuged at 3000 rpm for 10 min. The supernatant was discarded, and the pellet suspended in 100 µl of molecular biology-grade water. The suspension was subjected to boiling at 100˚C for 10 min, cooled on ice, and centrifuged at 15,000 rpm for 10 seconds before it was stored at −20˚C.

PCR Amplification
Aliquots of 2 µl of template DNA was used for PCR to detect and amplify mecA gene with origonucleotide primer mecA F;

Detection of Phenotypic Clindamycin Resistance
To detect clindamycin resistance, a suspension of 0.

Results
Among the MRSA isolates (n = 300), 6% (n = 18) of the strains were sensitive to both erythromycin and clindamycin, designated as S phenotype. However, 38% (n = 115) MRSA strains showed resistance to both erythromycin (zone size ≤ 13 mm) and clindamycin (zone size ≤ 14 mm) and these strains were constitutively resistant to clindamycin and designated as "cMLSB" phenotype. In addition, 42% (n = 125) showed resistance to erythromycin (zone size ≤ 13 mm) and sensitive to clindamycin (zone size ≥ 21 mm) with a "D-shaped" zone of inhibition around. These were identified as inducible clindamycin resistant strains and they were designated as iMLSB. Of note, 14% (n = 42) showed resistance to erythromycin (zone size ≤ 13 mm) and sensitive to clindamycin (zone size ≥ 21 mm) without "D" zone of inhibition around clindamycin and were designated as MS phenotypes as shown in Figure 1.
Results of mecA gene PCR where an Amplicon of approximately 180 bp was expected using positive and negative controls are shown in Figure 4.

Discussion
The increasing frequency of the Staphylococcal infections and their respective     Human and swine 42%, which is in agreement with other reports elsewhere [21] [22] [23]. Higher rates of iMLS B have been reported in other studies conducted in Uganda [24] and Kenya [25].  [26]. Probable widespread empirical use of erythromycin and clindamycin today and increasing consumption are the drivers of selective pressure [27]. We suggest reduction in macrolide usage to reverse such resistance pattern as it has been described elsewhere [28].  [19]. Also, Mohammad, 2012 reported 32.5% of MRSA were cMLSB phenotypes and 10% were iMLSB phenotype. These variations could probably be due to differences in the circulating clones or due to the variations in infection prevention practices and trends of antibiotics prescriptions in the community and veterinary practice [32].
The current study also reveals 6% of MS phenotype (E-R, Cl-S) among MRSA isolates. In this case, clindamycin can be used as treatment option only for less number of MRSA which are erythromycin resistant. While treating Erythromycin MRSA infection with Clindamycin antibiotic, there is always minimum chance of clinical efficacy compared to vancomycin antibiotic therapy [33]. We therefore emphasize routine use of D-test in diagnostic laboratories to avoid clinical failure while using clindamycin as an alternative to anti-MRSA antibiotics like vancomycin and linezolid [34] [35].

Conclusion
The prevalence of inducible and constitutive clindamycin resistance among MRSA Isolates from both humans and swine is high. The D test is a simple and affordable technique that can be used in low resourced settings to define precisely MLSB, both inducible and constitutive resistance patterns in addition to MSB in Staphylococcus aureus in the routine clinical laboratories. This can be an important strategy for good antibiotic stewardship in under resourced settings.

Conflicts of Interest
The authors declare no conflicts of interest regarding the publication of this paper.