Staphylococcus aureus isolates from bovine subclinical mastitis in central Ethiopia revealed the presence of antimicrobial resistance and resistant genes

Background: Staphylococcus aureus is one of the predominant causative agents of mastitis disease in dairy herds. Mastitis disease has a negative impact in the economic losses in the dairy sector across the globe. The aims of this study were to determine the prevalence and detect antimicrobial resistance genes in the Staphylococcus aureus isolated from milk samples of subclinical bovine mastitis in Central Ethiopia. Methods: A total of 265 lactating dairy cows from various dairy farms in four different geographical locations were screened by California mastitis test (CMT) for bovine subclinical mastitis. One-hundred thirty CMT positive milk samples were collected and transported to laboratory. Different biochemical tests and polymerase chain reaction (PCR) were used for the identication of S. aureus isolates. Finally, phenotypic and genotypic methods were performed for detection of some antimicrobial resistance patterns and genes (mecA, ermA, ermC, and msrA), respectively. Results: From total of 265 lactating dairy cows screened, 49% (n=130) were positive for bovine subclinical mastitis. One-hundred thirty mastitic milk samples were subjected to bacterial culturing, one hundred (76%) S. aureus isolates were identied based on phenotypic characters. Sixty-eight conrmed S. aureus isolates were obtained using PCR. Of the sixty-eight isolates tested 12 samples were contained the methicillin resistance gene A (mecA). No amplication was observed for the erythromycin resistance genes (ermA, ermC, and msrA). Conclusion: The high resistance of Staphylococcus aureus to commonly used antimicrobials contribute in dairy farms may cause health problems in the community consuming raw milk purchased from these farms. using electrophoresis, stained with gel red at 120 volts for 45min and visualized under UV light using a BioDoc-it™ imaging system (Cambridge, UK). We use 100bp plus DNA marker as molecular marker.


Background
Mastitis is a common disease of dairy cows and a major concern for the dairy industry because of economic losses due to decreased animal health and increased antibiotics usage (Heikkilä et al., 2018;Gussmann et al., 2019). Due to multiple etiologies, it always remained a challenge to veterinarian worldwide. Approximately, 140 species of microorganisms have been identi ed as etiological agents of bovine mastitis (Radostits et al., 2007). Of these various etiological agents, Staphylococcus aureus (S. aureus) is one of the major agents of contagious mastitis, responsible for mainly bovine subclinical and clinical infections in cattle worldwide (Barkema et al., 2006). This pathogen, in combination with both the bovine host and environmental factors, is characterized by low cure rates compared with other mastitis pathogens because of its capability to acquire antibiotic resistance and produce a wide array of virulence factors (Gao et al., 2012).
Programs for mastitis prevention and treatment use wide range of antimicrobials (Wald et al., 2019). The indiscriminate use of antibiotics has led to the development of multiple antibiotic resistances thereby rendering the antibiotic treatment ineffective (Alian et al., 2012). The resistant bacteria present in environments are in contact with human beings and animals (Hu et al., 2017). Antimicrobial resistance is a major public health concern in many countries due to the persistent circulation of resistant strains of bacteria in the environment and the possible contamination of water and food (Gelbrand et al., 2015).
Therefore, investigation of a wide spectrum of antimicrobial resistance genes of S. aureus isolated from cows with mastitis is crucial not only for bovine mastitis control, but also for public health (Xavier et al., 2017;Kadlec et al., 2019).
Although S. aureus responds poorly to treatment with many different antimicrobial agents, antibiotic therapy still plays a signi cant role in the prevention and cure of bovine Staphylococcal mastitis (Wald et al., 2019). The infection of cows with increasingly antibiotic-resistant strains can cause several therapeutic problems and is one of the main reasons for monitoring drug resistance (Saini et al., 2013).
The measurement of S. aureus antimicrobial resistance using phenotypic susceptibility tests, such as disk diffusion, is essential in order to select the most appropriate and e cient therapy (Walker, 2006). These methods can be combined with molecular analysis, as phenotypic S. aureus resistance to the most commonly used antimicrobials is related to the expression of antibiotic-resistance genes (Cockerill, 1999).
Rapid methods for accurate detection and susceptibility determination of S. aureus isolates are necessary to minimize patient suffering by identifying the antimicrobial agents to which the isolated strains may be sensitive to and hence provide treatment options (Duarte et al., 2015). Owing to the poor discriminatory power of the phenotypic techniques, deoxyribonucleic acid (DNA) based identi cation and genotyping techniques are now considered the ideal methods for the detection of antimicrobial resistance genes of S. aureus (Perez-Roth et al., 2001;Song et al., 2015). In previous study that was conducted in Ethiopia, S. aureus isolates were identi ed using only the phenotypic antimicrobial susceptibility test and identi cation of S. aureus (Duguma et al., 2014;Marama et al., 2016). Presently, there are limited published studies conducted in central highland of Ethiopia concerning molecular characterization of antimicrobial resistance of S. aureus in dairy farms. Therefore, the aim of this study was to determine antimicrobial resistance (AMR) patterns of S. aureus isolated from bovine subclinical mastitis in central highland of Ethiopia.

Study areas
This study was conducted in selected areas of central Ethiopia including Adaberga, Ambo, Bishoftu and Holeta as indicated in Fig. 1. The areas were purposively selected based on the abundance of dairy farms that constituting the known milk sheds to the Addis Ababa, and based on their agro-ecological differences.

Study Design And Study Population
A cross sectional study design was employed to determine the prevalence, antimicrobial resistance and resistant genes of S. aureus from bovine subclinical mastitis in central highland of Ethiopia from October Page 4/20 2018 to May 2019. The study population were lactating exotic Zebu cross-breed dairy cows which had not treated for mastitis either intra-mammary or systemic route during the study period. All dairy farms with herd size ranging from 20-110 cows, which were managed under semi intensive or intensive management system, were included as study population inclusion criteria.

Sample Collection And Transportation
The mastitis milk samples were taken from CMT positive dairy cows and collected according to earlier protocol (Quinn et al., 2004). Brie y, quarters were washed with tap water and dried with clean towel. The teat ends were then cleaned with cotton soaked with 70% ethyl alcohol. Then, after discarding the rst three streams of milk, 10 ml milk was collected aseptically into a sterile screw-cupped, pre-labeled test tube, by holding it in inclined position, so that, the pathogen that going to be recovered come from mammary gland. Finally, milk sample was held in an ice box for transportation to respective laboratory (National Agricultural Biotechnology Research Center, Holeta, Ethiopia) for isolation and identi cation of bacteria from milk samples. The samples were immediately cultured or stored at 4°C for a maximum of 24hr until cultured on standard bacteriological media.

Bacterial Isolation And Identi cation
Isolation and identi cation of S. aureus isolates were performed using the technique described beforehand (Quinn et al., 2015). Brie y, all samples were rst inoculated onto freshly prepared nutrient agar (HiMedia, India) and incubated at 37°C for 24hours. Bacterial colonies were identi ed based on colonial morphology, cultural characteristics, Gram's staining and biochemical tests. Staphylococcus aureus produces golden colonies on nutrient agar; it is non-motile; coagulase positive, catalase positive, ferments mannitol and produces double pattern of haemolysis on sheep blood agar. The isolates that were suspected to be Staphylococcus species were again sub-cultured onto freshly prepared Mannitol salt agar (MSA) (HiMedia, India) and incubated at 37°C for 24 hours. Golden yellow colonies were presumptively identi ed as S. aureus. Furthermore, a single golden yellow colony from the culture plate was sub-cultured on freshly prepared nutrient agar slants at 37°C for 24hours, after which the slants were stored in the refrigerator for further analysis.

Bacterial DNA Extraction
Each isolate was inoculated into Brain Heart Infusion (BHI, Hampshire, UK) broth and incubated at 37°C for 24 hr. Aliquots of each culture were centrifuged and the supernatant was discarded. The pellet was used to extract DNA using Wizard® Genomic DNA extraction Kit (Promega Corporation, Madison, USA), according to manufacturer's instructions. Extracted DNA was quanti ed at 260/280 nm by Nanodrop (Thermo Fisher Scienti c, Germany) and stored at -20 •C for further molecular work.

Antimicrobial Susceptibility Testing
Antimicrobial susceptibility testing of the PCR con rmed S. aureus isolates were tested against nine commonly used antimicrobial agents (Abetek, Liverpool, UK) using the Kirby-Bauer disc diffusion method on Mueller Hinton agar (HiMedia, India) following the guidelines of Clinical Laboratory Standards Institute (CLSI) (CLSI, 2012). The isolates were classi ed in accordance with the guideline of CLSI (CLSI, 2012) as susceptible, intermediate or resistance for each antimicrobials tested according to the manufacturer's instructions by measuring the zone of inhibition around the antimicrobials disc (in millimeter). The details of the nine antimicrobial agents utilized used in this study was shown in Table 1 below.

Molecular Detection Of Antimicrobial Resistance Genes
Polymerase chain reaction ampli cations of four antimicrobial resistant genes, which included methicillin resistant gene (mecA), various erythromycin resistant genes (ermA, ermC and msrA) were carried out with a pair of speci c primers and using previously described protocol (Sawant et al., 2009;Melo et al., 2014) ( Table 2). Details of primer sequences, their speci c targets and expected amplicon sizes were summarized in Table 2. The reactions were performed in a nal volume of 25µl each made by 12.5µl of 2XTaq PCR Master Mix, 1µl of 10 µM primer (each forward and reverse), 3µl of DNA template and 7.5µl sterile nuclease free water. PCR conditions were described by the original designers of primers (Table 2). Following completion of reactions, PCR products were run on a 1.5% agarose gel using electrophoresis, stained with gel red at 120 volts for 45min and visualized under UV light using a BioDoc-it™ imaging system (Cambridge, UK). We use 100bp plus DNA marker as molecular marker.

Antibiotic Susceptibility Test
One-hundred S. aureus isolates were tested to evaluate their susceptibility patterns a panel of nine antimicrobials agents. In the present study, S. aureus isolates were found variably resistant to the antimicrobials tested. Data depicting the susceptibilities of the isolates were presented as percentages are shown in Fig. 3. A large proportion (50-94.6%) of the S. aureus isolates obtained from Adaberga, Ambo, Bishoftu and Holeta were resistant to ampicillin, cefoxitin, penicillin and streptomycin. Despite the fact that a relatively large proportion (75-83.3%) of the isolates from Ambo, Bishoftu and Holeta were resistant to streptomycin, on the contrary, only a small proportion (43.24%) of the isolates from Adaberga sampled were resistant to this antimicrobial agent. Also intermediate sensitivity of S. aureus isolates was highest towards Erythromycin (50%), Cipro oxacillin (35%) and followed by Gentamycin (24%) and Streptomycin (14%). Moreover, isolates obtained from Ambo were relatively highly susceptible to cipro oxacillin and gentamycin. Similarly, low level resistance was observed against chloramphenicol as summarized in Table 3.
Similarly, multi-drug resistance (MDR) patterns were generated from 100 S. aureus isolates showing resistance to three or more antibiotics according to (Coyle, 2005). The MDR pattern PEN-AMP-CXT-TET-GEN was observed in 45% of the isolates from Adaberga and in 42% of Ambo isolates. The MDR pattern PEN-AMP-CXT was dominant among 27%, 22% and 26% of Adaberga and Bishoftu and Holeta samples, respectively. The predominant MDR pattern for isolates from Bishoftu and Holeta were PEN-AMP-CXT-STR-TET and PEN-AMP-CXT-STR was obtained at 33 and 24%, respectively. These results indicate that in the present study, MDR S. aureus was isolated from milk samples.

Discussion
The overall cow-level apparent prevalence of bovine subclinical mastitis in the central highland of Ethiopia was 49.05% (130/265). This result was almost in agreement with ndings from similar studies by Ayana et al. (2017) from Bishoftu, Arga et al. (2012) from Ambo, Dego and Tareke (2003) from southern Ethiopia and Mungube et al. (2004) from Addis Ababa, which reported prevalence of 46.09%, 58.82%, 40.4% and 39.8%, respectively. However, the result of the present study was lower than the reports of Duguma et al. (2014) who reported a prevalence of bovine subclinical mastitis was 81% in Holeta, central highland of Ethiopia, Bishi (1998) who reported 69.8% in dairy farms of Addis Ababa and its vicinity and Mekibib et al. (2010) who reported 74.7% around Addis Ababa. Moreover, the present nding was higher than previous reports by Workineh et al. (2002) who reported a prevalence of 25.1% in Addis Ababa and Delesse (2010) who reported a 10.3% with prevalence of mastitis in dairy farms around Holeta town, Ethiopia. As mastitis is a complex disease, the difference of prevalence observed from the current study could be due to interactions of various factors such as dairy cow management system and husbandry practice, environmental conditions, animal risk factors and virulence factors of the circulating causative agents (Radositis et al., 2007) and also difference in study methods used, period of investigation and speci c farm-level intervention might contribute in the variation of the prevalence of mastitis.
According to the biochemical tests and PCR nding in this study, 79.92% (100/130) of S. aureus isolates were recovered from 130 CMT positive milk samples. The nding of the current study was higher than the previous ndings which were done around Sebeta (44.03%) by Sori et al. (2005), in Holeta agricultural research centre (43.3%) by Duguma et al. (2014), in and around Holeta town (47.1%) by Mekibib et al. (2010) and in Bishoftu area (39.5%) by Asrat et al. (2013). This high rate of S. aureus in this study might suggest the high rate transmission of S. aureus infection which might occurred because of poor hygienic conditions during milking process in which contamination with S. aureus might occurred from contaminated milker's hands and milking equipment (Radositis et al., 2007). Apart from Ethiopia, S. aureus has also been reported as the chief etiological agent of mastitis in cattle by many studies from African and Asian countries (Abebe et al., 2016). Though direct comparisons among studies might be di cult, but in general, the variation in the prevalence between the present and previous studies might be due to differences in detection methods, geographical location of the study sites and differences in farm management practices in each studied farms. .
Previous studies have revealed that an increasing trend towards the occurrence of S. aureus isolates that portray multiple antimicrobials resistance phenotypes (Normanno et al., 2007;Pesavento et al., 2007) ; hence, S. aureus isolates that harbour multiple antimicrobials resistant traits have been reported to negatively impact on the treatment of staphylococcal infections. The present study showed a higher level of resistance of S. aureus to penicillin (93%), ampicillin (87%), tetracycline (75%) and cefoxitin (87%). The current nding was in line with the ndings of Abebe et al. (2016) who reported resistance of S. aureus to penicillin (94%), tetracycline (73.8%) around Addis Ababa and (Abera et al., 2013) who recorded 94.4% to penicillin around Adama and Sori et al. (2005) who recorded 87.2% S. aureus isolates were found to be resistant to penicillin. In general, the current study able to show that susceptibility of S. aureus to commonly used antimicrobials, penicillin, ampicillin, cefoxtin and tetracycline in study area was very low. The possible justi cation for this could be the development of alarming level of resistance of S. aureus due to the regular use of these commonly used antibiotics for the treatment of cows that might resulted in the spread of resistant strain in study area. This result was in accordance with reports from earlier studies in other countries, suggesting a possible development of resistance from prolonged and indiscriminate usage of some antimicrobials (Enright et al., 2002). Previous report indicated that the prevalence of antimicrobials resistance in S. aureus isolates become a serious problem in a dairy herds (WANG et al., 2009).The over-use of antibiotics in dairy farms is one of the major factors responsible for the emergence of drug resistant bacteria. Furthermore, isolates that are resistant to ampicillin may cross select for resistance to other beta-lactams including penicillin, therefore, resistance to ampicillin is an indication of the resistance of the isolates to other beta-lactam antibiotics (WANG et al., 2009). Methicillin-resistant Staphylococcus aureus (MRSA) is currently a major burden in veterinary and human medicine (Tangka et al., 2002). This type of resistance is considered to be one of the most important and has been implicated in many animal and human illnesses that have resulted in high mortality. It is therefore, very important to implement a systemic application of an in vitro antibiotic susceptibility test prior to the use of antibiotics in both treatment and prevention of intra-mammary infections.
In the present study, multiple drug resistant (MDR) S. aureus strains de ned as isolates that were resistant to three or more antibiotics were obtained in 54.26% of some of the milk samples analyzed. Development of multiple antibiotic resistances among most of these isolates may be attributed to transmission of resistance (R-factor) which is a plasmid-mediated genetic determinant. S. aureus often contain multiple plasmids that may contain various numbers of antibiotic resistant genes (Yamamoto et al., 2013). Given the fact that antibiotic resistance traits in bacteria species including S. aureus may occur either spontaneously by mutation and selection or by acquisition of new genetic material from other resistant organisms through transformation, transduction and conjugation it is usually not surprising that the antibiotic resistance pro les of isolates from the same region may vary considerably.
In the present study, the isolates of S. aureus were tested for four antimicrobial resistance genes which included methicillin resistance gene (mecA), various erythromycin resistance genes (ermA, ermC) and macrolide resistance A gene (msrA). Among the four antimicrobial resistance genes screened, only mecA gene was ampli ed, 12% (12/100) of S. aureus isolates out of 100 S. aureus isolates and other antimicrobial resistance genes which included ermA, ermC and msrA were not detected in any of the isolates in the present study. In this study, mecA positive S. aureus was detected in milk samples collected from different sampling areas, including Ambo 16.67% (2/12), Bishoftu 25% (3/12) and Holeta 53.33% (7/12) based on the presence of the mecA gene amplicon. However, the present study was disagree with the report from the Central highlands of Ethiopia by Seyoum et al. (2016) and (Mekonnen et al., 2018) from North-West Ethiopia, who did not detect any mecA positive S. aureus in their study. Variation in the proportion of mecA positive S. aureus in comparison to other researcher might be due to the difference in sample size, antibiotic use in animal husbandry and hygiene practices among the dairy farms.
The mecA gene was detected in some isolates that resist cefoxitin. The existence of mecA positive cefoxitin resistance S. aureus in milk has been reported in many previous studies (UGWU et al., 2015). The presence of mecA-positive MRSA strains in bovine milk samples has been reported in many countries (Kreausukon et al., 2012). The presence of mecA negative MRSA strains in bovine milk samples has also been reported by (Kumar et al., 2011). However, mecA-negative cefoxitin resistance S. aureus has been also recovered from bovine milk; the resistance revealed by mecA-negative cefoxitin resistance S. aureus isolates might be attributed to the presences of other beta-lactam resistance mechanisms (Malik et al., 2007). This indicated the presence of incompatibility between the detection of methicillin resistance phenotypically using cefexitin discs and the absence of mecA gene in some MRSA isolates. This nding is in accordance with (García-Álvarez et al., 2011) who identi ed phenotypic MRSA isolates without mecA gene. This may be attributed to the presence of PCR inhibitors or other physical factors that may have compromised the sensitive of PCR in the detection of mecA gene.

Conclusion
The present study has revealed that bovine mastitis is a widely prevalent disease in the dairy farms of the central Ethiopia. The present investigation also explored that S. aureus is an important cause for bovine mastitis, which warns the higher public health risk due to consumption of raw milk and its products. The study also demonstrated that isolates are characterized by multiple drug resistance to commonly prescribed drugs in veterinary and human pharmacies. Dairy cows in the study area had high rates of infection by multi-drug resistant S. aureus isolates, especially methicillin resistance A (mecA) gene which may hold a serious threat to human and animal health. Though the development of antimicrobial resistant determinants in S. aureus is associated with the uncontrolled usage of antimicrobial agents in human and veterinary medicine, the incidence of drug-resistant S. aureus in bovine milk samples warrants closer monitoring. Therefore, careful monitoring for the resistance status is an utmost need since the transmission of this pathogen is dynamic and involves human, animals, and likely the farm production environment. This study was conducted after gaining full approval by the ethical review board of the College of Veterinary Medicine and Agriculture, Addis Ababa University, Ethiopia. Informed written consent was taken from all participants prior to participation in this study. Also, permission from dairy farm owners/managers was obtained before collection of milk samples

Consent for publication
Not applicable

Competing interests
The authors declare that they have no competing interests Funding This study was supported by Ethiopian Institute Agricultural Research and Addis Ababa University. The institutions had no role on design of the study and collection, analysis and interpretation of data and in writing the manuscript.  Agarose gel electrophoresis analysis for the mecA gene in S. aureus isolates, Lane M = l00bp DNA marker, Lanes 1-7 = test samples