Prevalence and Antibiogram of Escherichia coli and Staphylococcus spp. Isolated from Cattle Milk Products Sold in Juja Sub-County, Kenya

Dairy ruminant milk provides a conducive environment for bacterial proliferation. In animals, these bacteria are exposed to antibiotics, whose overuse has led to increased cases of drug resistance. A cross-sectional study was conducted on milk and milk products vended in Juja Sub-County, Kenya to determine the prevalence of bacteria and antibiogram of Staphylococcus spp. and Escherichia coli. A total of 169 milk samples were obtained from various outlets in the study area. Milk samples were cultured and isolated bacteria were identified using standard bacteriological procedures. Various bacteria (15 species) were isolated in different proportions. Staphylococcus spp. and E. coli were isolated from 25.4% and 11.8% of the collected samples, respectively. The highest number of Staphylococcus spp. were isolated from raw milk (n = 34) while the highest number of E. coli where isolated from fermented milk (n = 15). Staphylococcus spp. and E. coli isolates were subjected to antimicrobial susceptibility tests using CLSI guidelines. The Staphylococcus spp. isolates were highly resistant to penicillin G (93%) but susceptible to norfloxacin (100%), gentamicin (90.6%), and chloramphenicol (86%). The E. coli isolates were highly resistant to cephalexin (85%) and ceftazidime (60%) but susceptible to chloramphenicol (100%), norfloxacin (95%), gentamicin (95%), azithromycin (95%) and cefepime (80%). Furthermore, 44.3% of Staphylococcus spp. and 50% of E. coli isolates had a Multiple Antibiotic Resistance (MAR) Index greater than 0.2. This implies that these bacteria were high-risk bacteria whose treatment with current antibiotics would be challenging. The high prevalence and multidrug resistance patterns shown by the Staphylococcus spp. and E. coli isolated from milk products in Juja Sub-county highlights the importance of proper handling and processing of milk from the farm to consumers. This will in turn reduce the possibility of zoonotic transfer of multidrug-resistant bacteria.


Introduction
Te high nutritional content of dairy ruminant milk makes it conducive to bacterial growth. Bacteria, such as Escherichia coli and Staphylococcus spp., can colonize the udder of cattle leading to the occurrence of mastitis [1,2]. Tese bacteria are transmitted to the dairy animals from the environment, during the milking process when the milker's hands or milking equipment lack adequate hygiene or along the milk product production chain by contaminated water or utensils used by vendors [3]. Humans can subsequently consume these bacteria leading to disease.
Te bacterium E. coli is of great importance in milk safety and health and has been labeled as an indicator of fecal contamination, inadequate sanitary measures, and handling practices in milk [4]. Tis bacterium is one of the most common causes of intestinal tract infections and diarrheal diseases [5] making it the second most causative agent of child death globally [6]. Te emergence and spread of antibiotic resistance has compounded the morbidity caused by E. coli. Staphylococcus spp. is among the most important pathogens afecting human health with staphylococcal poisoning being the most common type of food-borne disease [7]. Although Staphylococcus spp. from dairy products are nonpathogenic they can become opportunistic and cause nosocomial infections [8].
Because of the aggressive use and misuse of antibiotics, we have established that bacteria such as Staphylococci spp. and E. coli are becoming increasingly antibiotic-resistant [9,10]. For example, studies undertaken in Kenya reported multidrug resistance ranging from 18% to 34% in Staphylococcus spp. isolated from milk from cows [11]. It also reported raw milk samples from Ethiopia to have multidrugresistant E. coli [12]. Tese resistant bacteria can be transferred to humans through milk products that have been contaminated, adulterated, or unpasteurized [13].
Africa may be taking a heavier toll of antimicrobial resistance (AMR) than the rest of the world because surveillance data are scarce and some reports only give a fraction of the real situation [14]. According to the African Union Framework for Antimicrobial Resistance Control (2020)(2021)(2022)(2023)(2024)(2025) in the eforts to combat antimicrobial resistance, one of the primary goals is to increase the number of foods tested for pathogens and their resistance patterns. Te current study sought to evaluate the bacteria present in milk and the antimicrobial patterns of Staphylococcus spp. and E. coli isolated from raw milk and milk products sold in Juja Sub-county, Kenya. Te area of study is characterized by urban and peri-urban keeping of dairy cattle.

Study Area
We carried this study out in 5 wards of Juja Sub-County (Juja, Withethie, Murera, Kalimoni, and Teta) of Kiambu County, Kenya. Juja Sub-County is located 30 kilometers north of Nairobi with geographical coordinates of 1°11′ 0″ south and 37°7′ 0″ east. Te study area has 300,948 inhabitants with an average household size of 2.9. It has a land area of 342 sq. km [15] and is 1519m above sea level with a climate classifed as warm and temperate. Furthermore, it has an average temperature of 18.8°C. Juja Sub-County has about 1,457 households rearing 5,084 exotic dairy cattle and 1,148 households rearing 9,417 indigenous cattle [16]. In Juja Sub-County, there is a rising human population whose milk needs are partly met by small-scale dairy peri-urban farming.

Collection of Milk Samples.
Te current study used the cross-sectional design to collect samples of milk products from the 5 wards of Juja Sub-County. Te sample size was determined using the formula adjusted for small populations [17].
Using these formulae, the minimum sample size was 150. Te study planned to collect equal number of samples but the availability of diferent milk types led to diferences. In the current study, we collected a total of 169 samples of cattle milk products from diferent outlets and manufacturers on the main roads and markets of the wards (Table 1). Te UHT-packed milk sampled included 2 diferent brands of milk most sold within the 5 wards.

Morphological Characterization of Bacteria.
We inoculated the milk samples onto Sheep Blood Agar (HiMedia, India) and MacConkey Agar (Oxoid, UK). Briefy, 100 μl of the milk samples were inoculated onto 5% Sheep Blood Agar and MacConkey Agar. For Staphylococcus spp. and E. coli, we cultured isolates as described previously [18,19]. For Staphylococcus spp. discrete colonies from Sheep Blood Agar were subcultured unto Mannitol Salt Agar (Oxoid, UK). Colonies that appeared pink or fermented mannitol producing a pH that changed the color of the Mannitol Salt Agar from pink to yellow were presumptively taken to be Staphylococcus spp. [20]. For E. coli, discrete bright pink colonies from MacConkey Agar were subcultured onto Eosin Methylene Blue Agar (Oxoid, UK). It presumptively took colonies producing a green metallic sheen with a dark nucleated center to be E. coli colonies (20). We incubated aerobically all plates at 37°C for 18-20 hours.

Identifcation of Bacteria Using MALDI-TOF.
All isolated bacteria and presumptive Staphylococcus spp. and E. coli bacteria were further identifed by Matrix-assisted Laser Desorption Ionization-time of Flight Mass Spectrometry (MALDI-TOF) using procedures described [21] with slight modifcations. Briefy, discrete overnight colonies of bacteria isolated from the milk samples were picked from petri dishes using sterile pipette tips and added to 0.5 μl 25% formic acid. Tis mixture was then smeared on a FlexiMass-DS TM disposable polymeric plate and 0.5 μl of the matrix, α-cyano-4-hydroxycinnamic acid (CHCA) was added, and the plate inserted into the MALDI-TOF equipment. Te bacteria colonies were analyzed using the iDplus Axima Confdence Refectron (SHIMADZU BIOTECH, Japan) in a positive linear mode (2,000-20,000 m/z range). Te spectra profles of each bacterium, generated by detection of proteinaceous parts by the MALDI-TOF equipment, serve as a unique fngerprint for each bacteria and were searched against the SuperSpectra in the Spectral ARchive And Microbial Identifcation System (SARAMIS) database for bacterial identifcation.

Antibiotic Susceptibility Test.
Te antibiotic susceptibility test for isolated Staphylococcus spp. and E. coli bacteria was conducted according to CLSI guidelines [22]. For Staphylococcus spp., their susceptibility to 7 antibiotics (penicillin G, gentamicin, azithromycin, erythromycin, tetracycline, chloramphenicol, and norfoxacin) was determined. For E. coli bacteria, their susceptibility to 12 antibiotics (ampicillin, amoxicillin-clavulanate, cephalexin, cefoxitin ceftazidime, cefepime, gentamicin, streptomycin, azithromycin, tetracycline, chloramphenicol, and norfoxacin) was determined. We purchased all the antibiotics from Oxoid (Oxoid, UK). Escherichia coli ATCC 25922 and S. aureus ATCC 25923 obtained from the Kenya Medical Research Institute were used as reference strains.

Determination of MAR Index. Te Multiple Antibiotic
Resistance (MAR) Index was calculated using the following formula: where "a" is the total number of antibiotics to which a particular bacterium is resistant and "b" is the total number of antibiotics against which the bacterium was tested [23].

Data Analysis.
We entered all data into MS Excel (Microsoft, USA) and using SPSS 28 (Microsoft, USA), computed all frequencies and percentages using descriptive statistics. We analyzed the diference in prevalence of bacteria in milk products and wards using Friedman test followed by Wilcoxon signed-rank tests with Bonferroni adjustment. We set statistical signifcance level at 0.05 (p < 0.05).

Prevalence of Bacteria in Various Milk and Milk Products.
Based on the various proteins, serving as structural and functional parts in the isolated bacteria, various unique spectra were generated by the MALDI-TOF equipment leading to the identifcation of 15 species of bacteria with varying prevalence from 130 (76.9%) of the 169 collected milk samples (Table 2). Te most common bacteria isolated were S. sciuri, E. coli, Enterobacter cloaceae, Serratia marcescens, S. saprophyticus, and Enterobacter asburiae. Te milk type with the lowest bacterial prevalence was UHTpacked milk (6.7%) followed by yogurt (8.3%). Tere was a statistically signifcant diference (χ 2 (4) � 45.916, p < 0.001) in the prevalence of bacteria among the milk products. Raw milk had a signifcantly higher (p < 0.001) proportion of bacteria compared to yogurt, UHT-packed milk, and pasteurized milk.

Prevalence of Bacteria in Milk and Milk Products
Obtained from Various Wards. Te prevalence of bacteria isolated from milk from the 5 wards of Juja Sub-County are shown in Table 3. In descending order, most of the bacterial isolates were obtained in milk products sampled from Murera, followed by Juja, Teta, and Kalimoni wards, and the least were from Withethie ward. Tere was no statistically signifcant diference in the prevalence of bacteria as related to ward of origin of milk from specifc wards (χ 2 (4) � 7.914, p < 0.095)

Resistance of Isolated Staphylococcus spp.
Te Staphylococcus spp. isolates showed high resistance to penicillin G (93%) but were highly susceptible to norfoxacin, gentamicin, and chloramphenicol (Table 4). Only 2 isolates (S. saprophyticus and S. warneri) from UHT-packed milk were susceptible to all antibiotics. Te resistance pattern of the isolated Staphylococcus spp., shown in Table 5 revealed that most of the isolates were resistant to penicillin G. Te resistant pattern further showed that 92.8% (n = 13) of the isolated S. saprophyticus were resistant only to penicillin. Te only isolated S. xylosus bacterium was resistant only to penicillin, gentamicin, erythromycin, azithromycin, and tetracycline. Furthermore, the MAR index of the Staphylococcus spp. revealed that 44.3% (n = 10) of the isolated Staphylococcus spp. had a MAR greater than 0.2.

Antibiogram of Isolated Escherichia coli.
Te antibiogram of E. coli isolates from milk and milk products in Juja Sub-County are shown in Table 6. Te E. coli isolates indicated the highest resistance to Cephalexin (85%) and the highest susceptibility to Chloramphenicol (100%). Te resistance pattern of the E. coli isolates (Table 7) showed that 95% (19/20) were resistant to at least one antibiotic. Te antibiotic against which the most resistance was demonstrated was cephalexin (85%), followed by cefoxitin (50%).
Only one E. coli isolate was susceptible to all 12 antibiotics. We observed that there was an overall decrease in resistance against cephalosporins as the generation of the cephalosporins increased. Furthermore, the MAR index of the E. coli isolates revealed that 50% (n � 10) had a MAR index greater than 0.2.

Discussion
We geared the current study toward investigating the prevalence of bacteria and the antibiogram of Staphylococcus spp. and E. coli bacteria isolated from milk and milk products sold in various outlets in Juja Sub-County, Kenya.
Te products are sold to an urban set-up with high human population. Various bodies such as the African Union and Food and Agricultural Organization (FAO) have called for surveillance studies on the occurrence of bacteria in milk, leading to contamination and emergence of AMR in different set-ups to develop intervention strategies [24,25]. 6 (6.3%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 6 (3.6%) Klebsiella pneumoniae   Te current study showed that high proportion of the milk samples was contaminated with diferent bacteria. It was seen that raw milk had the highest proportion of bacteria. Bacteria contaminated the raw milk either during the milking process, handling by retailers, or storage under poor conditions. Tis high percentage of raw milk samples contaminated with bacteria is alarming. In a recent study in urban and peri-urban centers in Kenya, 86% of milk was sold as raw [26]. Milk hawkers in this channel of marketing seldomly cool or preserve the milk after collection from dairy farmers. Since farmers rarely observe the antibiotic withdrawal periods, the milk could also have bacteria resistant to various antibiotics used to treat mastitis and other animal diseases. Apart from Nairobi county, raw milk is the most consumed milk type by both rural and urban consumers in Kenya [27]. One main reason for this preference is the low price of raw milk, which is 60-70% cheaper compared to pasteurized milk [28]. A study carried out in Tanzania and Kenya showed consumers prefer raw milk to UHT-packed milk, as they consider the latter to contain additives [29]. Others prefer raw milk as they can buy the amount of milk they need, as opposed to UHT-packed milk, which is sold relatively in small standard sizes. Te use of pasteurized milk is however recommended since it has reduced bacterial load levels [30] as seen in the present study.
Te current study showed that only a few pasteurized milk samples had E. coli, similar to a study from Iran that recorded E. coli in only 9% of pasteurized dairy milk samples [31]. Tis, however, is in contrast to a higher prevalence of milk sampled from similar environments in Brazil. In that study, an E. coli prevalence of 41.1% in pasteurized milk was reported. Furthermore, in Ghana, the E. coli prevalence in pasteurized milk was observed at 78% [32]. A reason for this diference in prevalence may be that in the present study, pasteurized milk was sold from milk dispensing machines, while in the latter study, pasteurized milk was sold in open market containers.
In the current study, most of the E. coli isolates were isolated from fermented milk (Mala). Mala milk is prepared by fermenting in local calabashes [33]. Tus, the high percentage of E. coli isolated from this milk type may be due to the poor handling techniques of the milk during the preparation process. Furthermore, E. coli bacteria serves as an indicator of poor hygiene practices or even fecal contamination [34]. People in Kenya, who tend to consume more fermented milk products such as Mala, may be at risk.  Both Staphylococcus spp. and E. coli were not isolated from sampled yogurt products. Similarly, in other studies in Iran and Burkina Faso, low or no E. coli were isolated from yogurt samples [35,36]. A reason for the low recovery from yogurt is that the lactic acid bacteria that are used as starter culture in the production of yogurt have antibacterial efects and properties against E. coli [37].
In the current study, Staphylococcus spp. were most resistant to penicillin G. Locally, 100% of the Staphylococcus spp. isolated from milk from dairy goats in Juja and Tika Sub-Counties were shown to be resistant to penicillin G [10]. Te high rate of resistance to penicillin G is consistent with a study of Staphylococcus spp. in artisanal dairy products in Benin where 100% of Staphylococcus spp. were resistant to penicillin G [38]. Furthermore, Staphylococcus spp. in the current study were highly susceptible to chloramphenicol, which agrees with a study of similar bacteria from unpasteurized milk in South Africa, which had a susceptibility to chloramphenicol at 93.7%. Te Staphylococcus spp. in the current study had low resistance to gentamicin (7%) and norfoxacin (0%). Similarly, in the only study describing mastitis in Juja Sub-County [39], the Staphylococcus spp. from cow milk had a 6% resistance against Gentamicin. In addition, the current study agrees with the study on raw milk from Kenya where Staphylococcus spp. had a 4% resistance to fuoroquinolones and 5.4% resistance to gentamicin [11]. With these low percentages, we can postulate that bacteria from the study area are susceptible to norfoxacin and gentamicin because these antibiotics are possibly not frequently used to treat infections in dairy cattle in this area. Te E. coli isolates in the current study were highly susceptible to chloramphenicol, also similar to the study by [39], norfoxacin, gentamicin, and azithromycin. Tese results are similar to a study of raw milk sold in markets in Egypt [40]. Te current study reports resistance of the E. coli isolates to the older generations of cephalosporins. Tese β-lactam antibiotics are among the antibiotics used to treat Mastitis in Kenya [41] since few efective alternatives of antibiotics exist for mastitis treatment.
MAR index has proven to be a low-cost method for determining a high-risk source of infection from bacteria where particular antibiotics are frequently used [42]. Te current study showed that 44.3% of the Staphylococcus spp. and half of the E. coli isolates had a MAR index greater than 0.2 indicating these isolates were having high multiple drug resistance making them high-risk bacteria. Resistance could have emerged from cattle that have been repeatedly treated with antibiotics and thus gained multidrug-resistant bacteria that could be transmitted to humans. Tis is a call for concern as 66.2% and 55% of Kenyans consume raw milk at least once a day in rural and urban households, respectively [33].
Te high burden of AMR is evident in humans, where a review by O'Niell [43] forecasted that there will be about 4 million human deaths in Africa attributed to antimicrobialresistant bacteria by the year 2050. Te current study showed that all the isolated staphylococci bacteria were CoNS, some of which were multidrug-resistant. Multidrug-resistantStaphylococcus spp. have also emerged and are becoming a public problem on a global scale [44], some reasons being the fact that they have a short incubation time and are capable of producing toxins [45]. Staphylococcus spp., especially those that have developed antimicrobial resistance, can be transferred to humans through milk products that have been contaminated, adulterated, or unpasteurized [13]. Staphylococcus spp. have been found to occur in the mammary glands of cattle and can easily pass on to milk and milk products used for human consumption [2]. In their study [39], Staphylococcus spp. was the most prevalent bacteria (56.7%) isolated from milk from cows with subclinical mastitis in the same study area as the current study. Te four species of staphylococci from the current study, S. saprophyticus, S. warneri, S. xylosus, and S. sciuri, have all been listed among causative agents of orthopedic infections in humans [46]. Furthermore, food poisoning due to dairy product contamination by CoNS has been reported in Morocco [47] with food poisoning outbreaks due to CoNS reported in Brazil [48]. Coliforms were also isolated from milk and milk products sold at various outlets in Juja Sub-County, Kenya. Tese included E. coli, C. freundii, K. pneumoniae, E. cloacae, E. asburiae, and Serratia marcescens. Te presence of these coliforms may point to low sanitary conditions of the vended milk and pose potential risks to consumers of the milk products from which these coliforms were isolated [49]. It has been observed that E. coli isolated from dairy raw milk can also afect the health of human beings through the production of toxins [18] as seen in Shiga-toxin producing E. coli which produces verocytotoxins that cause severe diseases in humans such as hemorrhagic colitis [50] or lifethreatening complications including kidney damage and renal failure [51].

Conclusion
Te present study showed that both Staphylococcus spp. and E. coli were among the bacteria prevalent in milk sold in Juja Sub-County, Kenya, with the raw milk being the most contaminated. Since these bacteria are highly prevalent where good hygiene or proper handling is lacking, street vendors and milk farmers alike should be trained on strategies to improve sanitary measures in milk handling. Similar to other studies in the locality, most of the Staphylococcus spp. isolates from the milk products were resistant to penicillin G but all were susceptible to norfoxacin. Tus, as observed in the study by others [10], penicillin G is currently an unsuitable drug for mastitis treatment in the area while norfoxacin or gentamicin could be better drugs for treatment of mastitis caused by Staphylococcus spp. as seen in the subclinical mastitis study by [39] in Juja Sub-County, Kenya. For the E. coli isolated from the dairy products, the most efcacious drugs included gentamicin, azithromycin, norfoxacin, and chloramphenicol. However, resistance against cephalosporins was quite notable making them not-so-good drug candidates against mastitis caused by E. coli in the study area. In the same study area as the current study, Kagira et al. [39] reported a 43.3% prevalence of subclinical mastitis in udders of dairy cows, meaning possible antibiotic intervention by farmers. More studies are needed to ascertain the antibiotic types used by farmers in the study area and Kenya at large so that AMR in Kenya can be easily mapped. Such information will inform policies regarding the control of AMR. Te current study highlights the importance of bacteria present in milk and milk products in the study area and emphasizes the need for proper handling and processing of milk and the possibility of transmission of these resistant bacteria to human beings.
Studies should be undertaken in the study area to determine the potential of antibiotic-resistant bacteria crossing the animal-human barrier, causing hard-to-treat infections in humans.

Data Availability
Te raw data used to support the fndings of this study are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that they have no conficts of interest.