Prevalence and Potential Risk Factors for the Acquisition of Antibiotic-Resistant Staphylococcus spp. Bacteria Among Pastoralist Farmers in Kajiado Central Subcounty, Kenya

Antimicrobial resistance (AMR) is a growing health problem globally. To address this challenge, there is a need to generate baseline data on the prevalence and AMR profile of the main disease-causing bacteria. Here, we interrogated the prevalence of bacteria in the nasal cavity of healthy pastoralists in Kajiado Central Subcounty, Kenya, and the occurrence of AMR in Staphylococcus isolates among the study subjects. Nasal swabs from 176 pastoralists were cultured, and the bacteria isolates identified using standard phenotypic and biochemical bacteriological methods. Among the obtained 195 isolates, the most prevalent isolates were coagulase-negative Staphylococcus (CoNS) (44.9%), followed by Enterococci spp. (43.2%) while Staphylococcus aureus prevalence was 8%. Antimicrobial sensitivity of the Staphylococcus spp. isolates to 14 antibiotics representing six antibiotic groups was undertaken using the Kirby-Bauer disk diffusion method. Among the CoNS, the highest resistance was reported in amoxicillin (78.7%) and ceftazidime (76%), while the most resistance for S. aureus was reported in ceftazidime (100%), amoxicillin (71.4%), and streptomycin (71.4%). From an administered questionnaire looking at gender, animal contact frequency, history of hospital visitation and antibiotic usage, and habitual intake of raw milk, the study showed that male participants had a higher risk of carrying multiple drug resistant (MDR) bacteria than females (p = 0.02, OR = 1.3). Likewise, habitual intake of raw milk was significantly associated MDR acquisition (p = 0.02, OR = 1.82). This study reveals a high prevalence of AMR Staphylococcus isolates in the study area laying a foundation for further analysis of molecular characterization of the observed resistance as well as the development of interventions that can reduce the occurrence of AMR in the study area.


Introduction
The upper respiratory tract harbors commensal and potentially pathogenic bacteria, i.e., Staphylococcus spp., Moraxella spp., Corynebacterium spp., Streptococcus spp., and Haemophilus spp. [1]. The commensal bacteria function as competitors to pathogenic bacteria and continuously prime the immune system against the pathogenic bacteria [2]. The Staphylococcus spp. colonizing the nasal cavity are divided into coagulase-positive (S. aureus) and coagulasenegative (CoNS) and account for a variety of infections and diseases in both humans and livestock [3]. In African countries, Staphylococcal carriage ranges from 8.0 to 57% in both patients and healthy populations [4][5][6].
The first-line treatment for Staphylococcal infection is penicillinase-stable penicillin [7,8]. However, resistance to this class of antibiotics was first identified in1960s in Staphylococcus aureus [9]. AMR in Staphylococcus is highest in low-income countries and low middle-income countries [10][11][12]. Currently, methicillin-resistant Staphylococcus (MRS) pathogens are among the leading causes of death with a prevalence of 1-80% worldwide [13,14]. MRS is majorly caused by the acquisition of the staphylococcal cassette chromosome mec (SCCmec) comprised mainly of mecA gene or its homologues (mecB or mecC) which encodes for a novel penicillin-binding protein (PBP2a) with a lower affinity to the beta-lactam ring and the site-specific recombinase gene (ccrA/B/C). After its first recovery in the 1980s, the diversity of the SCCmec was noted first in the 1990s, and afterward, more diversity was detected between 2001 and 2008 [15]. To date, fourteen types of SCCmec (I-XIV) have been reported [16]. The first community-associated (CA-MRS) was detected in the 1980s [17]. There is a steady increase of CA-MRS in the 1990s and early 2000s [7,15,[18][19][20]. CA-MRSA has majorly been associated with SCCmec types IV, V, and VI [16,[21][22][23].
The integrons have been reported to foster the spread of AMR in Staphylococcus spp. given their ability to naturally clone and express resistant genes. Integrons contain up to 40 genes associated with resistance to beta-lactams, sulfonamides, aminoglycosides, chloramphenicol, and macrolides [24,25]. The occurrence of integrons in HA-MRS isolates was first reported in 2007 in South China [26]. Since then, integrons have been identified in antibiotic-resistant staphylococcal isolates largely in the Asian continent [27][28][29][30][31]. However, data on integron associated Staphylococcus AMR in other parts of the world especially in the developed countries and Africa are lacking. Coexistence of class 1/2 integron gene cassettes and SCCmec in MRS has been reported occasionally [28,[32][33][34]. The simultaneous existence of both mechanisms is anticipated to increase the adaptability of microbes to antibiotic selective pressure speeding up the spread and survival fitness of resistant Staphylococcus while fostering MDR evolution [32][33][34][35].
MRS infections are treated using a combination of penicillin and vancomycin [36,37]. However, suboptimal responses to vancomycin treatment, vancomycin intermediate susceptible Staphylococcus (VIS) minimum inhibitory concentration (MIC 4-8 μg/ml), were first detected in Japan in 1997 whereas total resistance to vancomycin, vancomycin-resistant Staphylococcus (VRSA) MIC ≥ 16 μg/ ml, was detected in the USA in 2002 [38,39], whereas the VIS is thought to be because, by cell wall thickening, VRS is caused by the acquisition of the Van operon from Enterococcus faecalis bacteria [40,41]. Since their emergence, VIS and VRS isolates have been detected in different parts of the world albeit at low percentages [42][43][44][45]. Penicillin-resistant Staphylococcus on the other hand is caused by a blaz gene carried in the R plasmids which encodes for the penicillinase enzyme that hydrolyses the β-lactam ring [46,47]. On the other hand, accumulation of mutations on the drug target genes has been identified as a main cause of resistance for protein and DNA replication inhibitors [48,49].
The emergence of antimicrobial resistance is mainly caused by the indiscriminate use of antibiotics for treatment/prophylaxis in both human and animal ecologies [50]. Globalization and lifestyle factors such as the intake of raw/undercooked animal products, low hygiene level, and high population density as well foster the spread of antimicrobial resistance [5,51]. In Africa, AMR in Staphylococcus spp. ranges from 12 to 80% [52][53][54]. In Kenya, Staphylococcus spp. AMR ranges between 20 and 84.1% [54][55][56][57] with a steady increase from <50% in the 2000s to >50% by 2020 in Staphylococcus antimicrobial resistance [55,[57][58][59][60][61][62][63], whereas most Staphylococcus spp. AMR studies in Kenya have been conducted in the urban hospital setting (Nairobi, Kiambu, Kilifi, Kisumu, and Kericho); majority of the studies [60,64,65] investigating antibiotic resistance in rural areas, especially among pastoralist communities such as the Maasai, who use large quantities of antibiotics for veterinary care with >95% being self-administered and up to 75% carried out without professional consultation, are lacking or poorly documented [51]. The current study is aimed at determining the prevalence, risk factors, and detection of mecA or mecC antibiotic-resistant marker in Staphylococcus isolates obtained from the anterior nares of livestock farmers in Kajiado Central Subcounty in Kenya.

Material and Methods
2.1. Study Area. The study was undertaken in the Kajiado Central Subcounty region in Kajiado County, Kenya, located in the southern part of Kenya, approximately 80 km from Nairobi City and borders Northern Tanzania: 1.83874 latitudes and 36.79135 longitudes. The subcounty is divided into five wards: Matapato South, Matapato North, Purko, Ilidamati, and Dalalekutuk, which were included in this study. Kajiado Central experiences bimodal rainfall ranging from 500 to 1200 mm, with short rains occurring from October to December and long seasons between March and May, and an average temperature of 25°C. The total human population of the subcounty is 161,862, most of which are of the Maasai community and are pastoralists. The total livestock population in the subcounty as per the 2019 census was 802,898 [66].

Study Design.
A descriptive cross-sectional study design was used. The study sample size was determined using the Thrusfield method [67], which showed that a minimum of 150 participants were sufficient for the current study. A total of 176 participants were sampled. The sampling unit of interest was household members, which were selected randomly from a sampling frame provided by the county extension and public health officers. Based on the human population in specific wards, the study participants were from Matapato South (n = 54), Matapato North (n = 51), Purko (n = 25), Ilidamati (n = 30), and Dalalekutuk (n = 16) wards (Table 1).

Nasal Swab Collection and Administration of the
Questionnaire. Nasal swabs were collected, as described previously [68]. Briefly, the swabs were collected using a sterile nasal swab (BD BBL™ CultureSwab™) dipped in normal saline by slightly swabbing both anterior nares of the participants and transferred to the transport media (Scharlab S.L, Spain). The sample was labelled and transported back to the microbiology laboratory based at the Institute of Primate 2.6. Detection of the mecA and mecC Genes Using Singleplex PCR. DNA from the resistant bacteria was extracted using the heat lysis method. Briefly, one bacterial colony was suspended in 500 μl of nuclease-free water, heated at 95°C for 10 minutes followed by freezing at -80°C for 5 minutes. The DNA was harvested by centrifugation at 1200 rpm for 10 min, the supernatant was picked, and pellets were discarded. The extracted DNA was subjected to PCR for detection of mecA and mecC genes using primers mecA 513 bp C05/C06 (F-AAA ATC GAT GGT AAA GGT TGG C) and D09/D10 (R-AGT TCT GCA GTA CCG GAT TTG C) and mecC 718 bp (GAAAAAAAGGCTTAGAACGCC TC CCTGAATCTGCTAATAATATTTC  Table 2.  Table 3). The Staphylococcus aureus prevalence was 8%. Other isolates obtained from the study are outlined in Table 3. Respondents from Matapato South and Matapato North had the highest (50%) prevalence of CoNS, while Enterococcus spp. were more common in respondents from Matapato South ward.

Antibiotic
Resistance. The obtained Staphylococcus spp. were further investigated for antimicrobial susceptibility. Among the CoNS, the highest resistance was amoxicillin (78.7%), followed by ceftazidime (76%) and oxacillin (49.3%). Majority of the CoNS isolates, however, were sensitive to Augmentin (88.0%) and ciprofloxacin (84%). Resistance to other tested antibiotics is shown in Table 4.
For S. aureus, all (100%) isolates were resistant to ceftazidime. The percentage of S. aureus resistant to amoxicillin (71.4%), streptomycin (71.4%), and vancomycin (64.3%) was also high. The S. aureus isolates resistant to oxacillin (28.6%) and cefotaxime (21.4%) resistance were lower compared to the CoNS isolates. In contrast, S. aureus isolates were highly (100%) susceptible to Augmentin. The summary of the sensitivities of S. aureus to individual antibiotics is shown in Table 4.

Relationship between Risk Factors and Prevalence of Multidrug Resistance in Bacteria Isolated from the Study
Participants. The antibiotic susceptibility results were used to evaluate the potential association between various factors: gender, age, consumption of raw milk, animal contact, antibiotic usage, hospital visits, and the occurrence of MDR bacteria in nares (Table 7. From the analysis, male participants had a higher risk of carrying MDR bacteria than females (p = 0:02, OR = 1:3), while habitual intake of raw milk was significantly associated with a higher prevalence of MDR bacteria (p = 0:02, OR = 1:82). Majority of the study participants (82.4%) were frequently in contact with livestock; however, this factor was not associated with the occurrence 5 BioMed Research International of drug resistance bacteria (p = 1, OR = 9:7). The relationship between the prevalence of various risk factors and the occurrence of MDR bacteria is outlined in Table 7.

Prevalence of mecA and mecC Genes among Methicillin-Resistant Staphylococcus
Isolates. The phenotypically methicillin-resistant isolates were further investigated for

Discussion
The current study was geared at determining the spectrum of bacteria colonizing the nasal cavity of a pastoralist community in Kajiado Central Subcounty in Kenya as well as determine the antibiotic sensitivity of the obtained Staphylococcus isolates. First, CoNS, Enterococci spp., and S. aureus were the dominant bacteria in the nasal cavity of the study subjects with CoNS contributing close to half of the isolates. The dominance of these bacteria is not unique to these settings. Different studies across the world (Brazil, Saudi Arabia, Kuwait, Middle East, and Egypt) have reported similar trends [71][72][73]. The dominance of these bacteria is largely due to their commensal nature in humans, mainly on the skin, nasal cavity, and gut [2,74]. The prevalence of S. aureus is however lower compared to that recorded elsewhere in hospital settings and among healthcare workers in Kenya [4,56] and across communities in East Africa and sub-Saharan regions in Africa [55,75,76]. The difference may have been influenced by the difference in study, laboratory, experimental setting, and climate factors where the high temperature in the study area could have reduced S. aureus colonization [77,78]. Moreover, the high rates of CoNS isolated in this study might have lowered S. aureus colonization by secreting molecules that negatively affect the virulence of S. aureus [79,80]. Secondly, we show that most of the circulating Staphylococcus isolates in the study population are resistant to at least one antibiotic tested in our study. The highest resistance was recorded in the penicillin which is in line with other studies across the world [54,56,60,[81][82][83][84][85]. High penicillin resistance is linked with overuse of these antibiotics in treatment of bacterial infections especially in livestock [84][85][86]. In our opinion, HGT and interspecies transfer within the livestock and human ecology might have caused the observed resistance in the human population. Methicillin resistance among the CONS isolates was close to that reported in Nigeria (65%) [87] and Ghana (60-93%) [88,89] but slightly lower than that recorded in some sub-Saharan countries (<90%) [90][91][92]. This resistance is however higher than that recorded in hospital settings in some parts of Kenya (21-30%) [93,94] suggesting a higher circulation of MRS isolates in the study population compared to the other places which is alarming given the weak association of the history of antibiotic usage and hospital visitations to emergence and spread of MDR in this community. However, methicillin resistance for S. aureus in our study is comparable with that recorded elsewhere in the country in hospital, livestock, and community settings (1-90%) [4,60,83,95].
Interestingly, resistance to vancomycin in the study area was higher than that recorded elsewhere in the world: 0-11% in Kenyan hospital setting [54,82,96] and 0-16% in the rest of the world [53,97,98]. The resistance is however lower than that identified among camel slaughterhouse workers in Ethiopia (54%) [99]. Vancomycin resistance has been linked with presence of vanA operon especially in enteric bacteria. Molecular investigation of the causes of the observed vancomycin resistance will be beneficial especially in establishing possibility of HGT of vancomycin resistance gene to the Staphylococcus isolates given the codominance of the two bacteria in the study area. Importantly, the high vancomycin resistance in the current population compared to other studies might be partly due to low sample size especially among the S. aureus which limits the statistical power of the study. Of note, resistance of the Staphylococcus isolates across all six antibiotic classes tested (glycopeptide, aminoglycosides, macrolides, tetracycline, and quinolones) was noted which is alarming. Given the high susceptibility of the Staphylococcus isolates to Augmentin compared to the rest of the tested antibiotics, this drug can serve as an alternative drug for the treatment of staphylococcal infections in this population.
The presence of mecA and mecC genes in some of the CoNS isolates confirms the presence of SCCmec-associated MRS in the study area. The mecA gene is the gold standard for MRS; thus, occurrence of mecA gene and mecC gene in our study was expected. Although the mecC gene has been detected in various clinical studies across the world [100][101][102][103], this is the first study to report mecC gene resistance in Kenya. The presence of mecA and mecC genes in the methicillin-resistant isolates forms a basis for further molecular work especially SCCmec typing in the current study population. The absence of both the mecA and mecC genes in some of the MRS isolates is not unique to this setting; some studies have reported similar findings or lack of one of these genes in methicillin-resistant isolates [101,[104][105][106]. Thus, the possibility of a different MRS mechanism in the study area is highly probable which needs further investigation. Some studies that have explored further mecA and mecC negative isolates have identified the presence of the mecB homologue, hyperproduction of β-lactamase, and production of modified PBPs [17,[107][108][109]. Further, class 1 and 2 integrons containing dfrA12-orfF-aadA2, dfrA17-aadA5, aadA2, and aadB, oxa2, aacA4, orfD-aacA4-catB8, aadB-catB3, orfD-aacA4 and aadB-aadA1-cmlA6, dhfrA1-sat2-aadA1, dhfrA11, dhfrA1-sat2 gene cassettes, respectively, have been detected in resistant Staphylococcus isolates in the past [32][33][34]. Given the ability of the integrons to carry resistant genes for β-lactams, further investigation of the presence of these classes of integrons in the current study population will be necessary.
Multidrug resistance (MDR) was recorded in both CoNS and S. aureus majorly to the beta-lactams, macrolides, and aminoglycosides. Although most studies in Kenya and East Africa have focused on multiple drug resistance in livestock, food, and nosocomial infections, our findings are comparable to these studies [84,[110][111][112]. MDR resistance has been attributed accumulation of multiple drug resistant genes or overexpression of resistance genes encoding for efflux pumps and acquisition of less susceptible genes [113][114][115][116][117]. Investigation of the risk factors associated with MDR linked livelihood factors to the acquisition of MDR Staphylococcus 7 BioMed Research International isolates. Habitual intake of raw milk had a significant impact on the acquisition of resistant MDR Staphylococcus bacteria which is in accord with other studies conducted elsewhere in the world [118][119][120]. Staphylococcus spp. bacteria have been one of the foodborne pathogens occurring in raw milk; thus, the link between raw milk intake and occurrence of MDR in the study participants is not surprising. Also, from the study, the males had higher odds of acquiring resistant bacteria compared to female which is in line with other studies [121][122][123]. The likelihood of the males to acquire MDR bacteria may be linked to the occupational differences between the two genders in the Maasai community where the males are more involved in animal husbandry compared to the females in this community, as well as differences between hygiene levels in the two genders.
Although antibiotic usage in the current study was not significantly associated with the acquisition of antibioticresistant strain which differs with other studies [51,124,125], the study result relied heavily on study participant honesty which was dependent on their memory, a factor that might have limited our study findings. Most pastoralists are known to extensively use antibiotics in the management of livestock diseases with lack of withdrawal period during the medication period. This may have led to consumption of antimicrobial drugs eventually through milk and meat products. In such scenarios, HGT between livestock ecology and human is possible. The link of livelihood factors to acquisition of MDR acquisition necessitates proper understanding of the diverse beliefs in this community, and how they affect AMR acquisition and spread that will be essential in designing a proper AMR intervention to mitigate antibiotic resistance spread.

Conclusions
We outline three major findings from this study. First, most of the circulating nasal bacteria in the Kajiado Central were CoNS with some containing the mecA and/or mecC antibiotic-resistant markers, suggesting the possibility of factors promoting high nasal colonization of Staphylococcal spp. in humans that need further investigation. Second, we record a high prevalence of MDR for most of the antibiotics currently used for staphylococcal infection treatment. Lastly, livelihood factors (intake of raw milk and gender) were the major contributors to the acquisition of multidrug resistance strains, while antibiotic pressure and history of hospital visitation < 3 months prior to this study did not significantly impact the acquisition of resistant bacteria strains in the human population. These findings provide a basis to integrate social studies into antimicrobial resistance in this community to understand the cultural, social, and economic factors influencing the acquisition of antimicrobial resistance in this study area. The study also gives insight on the presence of mecA/mecC methicillin-resistant Staphylococcus strains laying a foundation for further investigation of molecular mechanisms in the observed resistance as well as building up on efforts to mitigate antimicrobial resistance at both the national and global scales. This will aid in the development of appropriate antimicrobial resistant mitigation strategies.

Data Availability
The datasets generated are available from the corresponding author on reasonable request.

Disclosure
The views expressed herein are those of the authors and not necessary that of the funding agency.

Conflicts of Interest
The authors declare that there is no conflict of interest.