Camel milk holds significant cultural, nutritional and economic importance in many African countries, including Tunisia. Owing to its unique composition and perceived health benefits, camel milk is frequently consumed raw or minimally processed, particularly in arid and semi-arid regions. However, the consumption of raw camel milk raises serious public health concerns, as it may harbour pathogenic and opportunistic microorganisms capable of causing disease and transmitting antimicrobial resistance (AMR) (Hirad et al., Reference Hirad, Ahmad, Alkhedhairy, Bahkali and Khan2018; Mohmed et al., Reference Mohmed, Barre, Mohamud, Gaciye and Hirsi2024).
Several pathogenic and opportunistic bacteria have been isolated from camel milk, including coagulase-positive staphylococci (CoPS) and coagulase-negative staphylococci (CoNS) (Goncalves, Reference Goncalves2017; Balemi et al., Reference Balemi, Gumi, Amenu, Girma, Gebru, Tekle, Ríus, D'Souza, Agga and Kerro Dego2021), Streptococcus agalactiae (Seligsohn et al., Reference Seligsohn, Nyman, Younan, Sake, Persson, Bornstein, Maichomo, de Verdier, Morrell and Chenais2020), Escherichia coli (Zuleka et al., Reference Zuleka, Abebe, Awot, Belayneh, Tehetna and Kumar2016), Listeria grayi (Lezzoum-Atek et al., Reference Lezzoum-Atek, Belhout, Bouchenafa and Bouayad2023) and Corynebacterium bovis (Aqib et al., Reference Aqib, Muzammil, Naseer, Shoaib, Bakht, Zaheer, Khan, Khan, Usman, Shafeeq, Tanveer, Hussain, Saleem and Prince2022). Infection with these pathogens may lead to clinical mastitis in dromedary camels, as recently reported in Kenya, where a prevalence of 17.6% was recorded (Mwangi et al., Reference Mwangi, Gitau, Ikiror, Kimeli, Gakuru, Machuchu and Kingori2022). Such infections not only affect animal health and productivity but also increase the risk of milk contamination. However, camel milk contains beneficial microorganisms, particularly lactic acid bacteria (LAB), which contribute to its probiotic, nutritional and fermentative properties. The LAB community in camel milk commonly includes Lactococcus lactis, Lactobacillus spp. and Enterococcus spp. (Sun et al., Reference Sun, Shao, Liu, Ma, Chen, Zheng and Zhoa2024). Additional potentially beneficial bacteria, such as Enterococcus faecalis, Enterococcus faecium, Propionibacterium spp., Streptococcus thermophilus, Solibacillus silvestris and Akkermansia muciniphila, have also been reported (Sheikh et al., 2024). The composition of LAB varies according to environmental conditions, animal diet and microbial ecosystems, which are important for optimizing traditional camel milk products (Fguiri et al., Reference Fguiri, Ziadi, Atigui, Ayeb, Arroum, Assadi and Khorchani2015).
Several Gram-positive bacteria present in camel milk may act as opportunistic pathogens and acquire AMR through genetic and ecological interactions (Qin et al., Reference Qin, Huang, Yang, Zhao, Zhao, Xu and Zhang2023). These bacteria, estimated to comprise 60–80% of the total bacterial isolates from camel milk, exhibit resistance to commonly used antibiotics (Hirad et al., Reference Hirad, Ahmad, Alkhedhairy, Bahkali and Khan2018). This resistance is largely driven by the heavy use of antibiotics in livestock production systems (Getahun et al., Reference Getahun, Abey, Beyene, Belete and Tessema2024). This poses a growing threat to food safety and public health, particularly in contexts characterized by inadequate hygiene practices, limited awareness of the risks associated with raw milk consumption and the overuse of antimicrobials (Shafii et al., Reference Shafii, Barre, Mohamud, Mohamed, Gaciye and Hirsi2024).
At the national level, Tunisia has made progress in standardizing the quality of raw camel milk intended for processing and export (NT 14.261:2009) (Codex, Reference Codex2024). Nevertheless, as camel milk production and consumption continue to increase, continuous monitoring of its microbial quality and AMR patterns remains essential. Therefore, this study aimed to investigate the prevalence and AMR patterns of multidrug-resistant (MDR) Gram-positive bacteria isolated from raw camel milk and assess associated risk factors in southern Tunisia. Findings are expected to support the national AMR surveillance efforts and contribute to improving the safety and sustainable use of camel milk.
Materials and methods
Study areas
The study was carried out from September 2022 to August 2023 in four governorates (Kebili, Gafsa, Medenine and Tataouine) from southern Tunisia (Fig. 1). These localities are characterized by arid climate with a low average of annual rainfall, high annual temperature means and predominance of thorny plants. Selected governorates harbour more than 60% of the national camel population which is managed mainly under an extensive system. The Tunisian annual camel milk production is estimated to be five million litres (Codex, Reference Codex2024).
Map of Tunisia showing the geographical location of the studied areas and examples of the sampled dromedary she-camels.
Figure 1 Long description
The map of Tunisia highlights four governorates: Kebili, Gafsa, Medenine and Tataouine, marked in orange. These regions are located in southern Tunisia. The map includes a scale bar indicating 100 km and a compass for orientation. To the right, there are two images of dromedary she-camels. The top image shows the udder of a camel, while the bottom image displays another camel's udder from a different angle. The images illustrate examples of the camels found in these regions.
Sample size determination
The required number of animals (n) was estimated for a 95% confidence interval as follows: N = [1.962 × P exp (1 − P exp)]/D 2, where P exp: expected prevalence which was estimated to 95.5% and D: the precision that was fixed to 5% (Thrusfield, Reference Thrusfield2018). Accordingly, based on the stated formula for random sampling, 66 samples were needed. After the selection of lactating females available in the visited farms, a total of 103 apparently healthy dairy camels were randomly selected and included in this study (Fig. 1).
Sample collection and risk factor analysis
About 10 ml of raw milk was taken from each female according to strict aseptic procedures to avoid contamination with microorganisms that can be found on the animal skin and/or the sampler's hands. Before sampling, hands and teat ends were washed and disinfected with 70% ethanol (Wattenburger et al., Reference Wattenburger, Schmidt, Placheta, Middleton and Adkins2020). We used a sterile bottle with tightly fitting caps. The first few drops of milk were discarded. After sampling, the bottles were labelled with a permanent marker and transported in an icebox to the Laboratory of Veterinary Microbiology for bacterial culture.
Risk factors were analysed to assess the distribution of MDR isolates from camel milk samples. The analysis focused on geographic distribution, camel age-classes, stage of lactation and history of contact with other ruminants. These data were recorded during the sampling step. These factors were evaluated to identify potential associations with the occurrence of MDR in camel milk and to better understand the epidemiological patterns that may contribute to the emergence and spread of AMR within camel populations.
Bacterial isolation and identification
A volume of 5 ml was taken from each sample, diluted in buffered peptone water and incubated for 12 h at 37°C. The resulting mixture was then cultured on blood agar and incubated aerobically overnight at 37°C. Distinct colonies were further identified at the species level using standard bacteriological tests including Gram staining, catalase and coagulase production, and conventional biochemical tests. The identified isolates were then confirmed by using the miniaturized biochemical galleries (API32 Staph and API20 Strep) (bioMérieux, France). Molecular confirmation based on 16S rRNA gene was applied but limited to Staphylococcus bacteria (Table 1). The obtained isolates were stored at −20°C in brain heart infusion broth supplemented with 20% glycerol.
Primers used in this study for the detection and the genetic characterization of the ten detected MDR bacteria
Table 1 Long description
The table lists PCR targets used to identify Staphylococcus species and to characterize antibiotic resistance by gene detection. For identification, a 16S rRNA primer pair yields a 370 base pair product using cycling with an annealing step at 60 degrees Celsius. Resistance targets include mecA for methicillin at 310 base pairs and blaZ for penicillin at 772 base pairs. Aminoglycoside resistance is covered by ant(6)-Ia at 850 base pairs and aph(3)-III-a at 500 base pairs. Quinolone-associated targets are grlA at 307 base pairs and gyrA at 251 base pairs, the smallest product listed. Sulfamethoxazole trimethoprim targets include dfrD at 582 base pairs and drfA at 374 base pairs. Macrolide targets are msr(B) at 595 base pairs and erm(A) at 642 base pairs, while tetracycline targets include tet(O) at 615 base pairs, tet(M) at 576 base pairs, and tet(K) at 697 base pairs; chloramphenicol is represented by cat(pC221) at 450 base pairs. Across assays, PCR programs generally start with 95 degrees Celsius for 5 minutes, run 35 cycles with 95 degrees Celsius denaturation and 72 degrees Celsius extension, and finish with 72 degrees Celsius for 10 minutes, with annealing temperatures varying by target from 49 to 65 degrees Celsius. The table provides primer sequences and expected product sizes for detection, but it does not report prevalence, expression, or phenotypic resistance outcomes.
Antimicrobial susceptibility and phenotypic profiles determination
Antimicrobial susceptibility analysis was systematically performed for all identified bacteria. The antibiotic panel included all drug classes prioritized for Gram-positive bacteria by the European Committee for Antibiotic Susceptibility Testing (EUCAST, v2024) based on their relevance to animal and public health therapy (Table 3). Antimicrobial susceptibility was determined by using the disc diffusion method on Mueller Hinton agar plates (Oxoid). Thus, isolates were classified as susceptible (S), intermediate (I) or resistant (R) based on the measured diameters of the inhibition zones compared to the critical interpretative zone diameter. The non-susceptible category, which corresponds to resistant and intermediate classes, was combined (R + I) to assess and monitor their global resistance levels. Bacteria exhibiting non-susceptibility patterns to more than three antimicrobial classes were considered as MDR (Magiorakos et al., Reference Magiorakos, Srinivasan, Carey, Carmeli, Falagas, Giske, Harbarth, Hindler, Kahlmeter, Olsson-Liljequist, Paterson, Rice, Stelling, Struelens, Vatopoulos, Weber and Monnet2012).
Bacterial DNA and detection of AMR genes
Genomic DNA was extracted from all identified MDR isolates using the boiling method. These MDR isolates, which exhibited phenotypic AMR, were subsequently subjected to polymerase chain reaction to detect genes encoding resistance to penicillin (blaZ), oxacillin (mecA), aminoglycosides (aph(3′)-IIIa, ant(6)-Ia), macrolides (erm(A), erm(B), msr(A)), tetracycline (tet(K), tet(O)), sulphonamide (drfA) and quinolones (grlA, gyrA). The primers used and the corresponding thermocycling conditions are provided in Table 1, following the protocol described by Ben Chehida et al. (Reference Ben Chehida, Gharsa, Tombari, Selmi, Khaldi, Daaloul, Ben Slama and Messadi2021).
Data analysis
Frequency analyses using contingency tables with Chi-square tests were conducted to assess the association between the presence of AMR and the following factors: strain, geographical zone, age and contact with ruminants. Statistical significance was set at P < 0.05.
Results
Identified bacterial species
Of the 103 sampled camels, 21 (20.4%) did not yield any bacteria, while from the remaining 82 samples, 86 distinct bacteria were isolated. However, more than one bacterial isolate was recovered from 11 milk samples (10.7%). Biochemical identification revealed the presence of different Gram-positive bacteria with a quite low prevalence rate. Based on molecular confirmation using 16S rRNA gene PCR, Staphylococcus was the most frequently isolated genus (54/86, 62.8%) (Table 2). Among these, 13% were CoPS including S. aureus (n = 6) and S. intermedius (n = 1). The remaining isolates (87%) were CoNS (n = 47) included S. simulans (n = 22), S. xylosus (n = 8), S. warneri (n = 5), S. hominus1 (n = 5), S. capitis (n = 3), S. sciuri (n = 2) and S. equorum (n = 2). Additionally, other Gram-positive bacteria were identified, including Cellulomonas spp. (16, 18.6%), Lactococcus lactis subsp. lactis (8, 9.3%), Aerococcus viridans (7, 8.5%) and Kocuria rosea (1, 1.1%) (Table 2).
List of the identified Gram-positive bacteria isolated from camel milk samples according to different risk factors
Table 2 Long description
Counts of Gram-positive bacteria isolated from 86 camel milk samples are broken down by bacterial category and species, then by geographic zone, lactation stage, camel age group, and whether camels had contact with ruminants. Coagulase-negative staphylococci were most common overall, especially Staphylococcus simulans with 22 isolates, followed by Cellulomonas species with 16 isolates. Coagulase-positive staphylococci were less frequent, with Staphylococcus aureus at 6 isolates and Staphylococcus intermedius at 1 isolate. By zone, Kebili contributed the most isolates overall with 40, and Staphylococcus simulans was concentrated there with 14 isolates; Benguerden had 17 total and included 11 of the 16 Cellulomonas isolates. Across lactation stages, totals were highest in the second stage with 31 isolates and lowest in the fourth stage with 16; Staphylococcus simulans appeared across all stages, while Cellulomonas was most frequent in the second stage. By age, more isolates came from camels under 10 years (48) than from those 10 years or older (38). Contact with ruminants was associated with most isolates overall (70 with contact versus 16 without), including 19 of 22 Staphylococcus simulans and 12 of 16 Cellulomonas isolates. Blank cells indicate no isolates reported for that species in that subgroup, and the table reports isolate counts rather than infection rates or statistical significance.
Notes:
a CoPS, coagulase-positive staphylococci;
b CoNS, coagulase-negative staphylococci.
Antimicrobial sensitivity rates
Significant antibiotic resistance rates were observed across all identified bacteria. Regarding Staphylococcus, the highest resistance rate was recorded for chloramphenicol (50%, 27/54), followed by clindamycin (42.6%, 23/54), neomycin (40.7%, 22/54), nalidixic acid (40.7%, 22/54), kanamycin (24%, 13/54) and gentamicin (22%, 12/54) (Table 3). Significant resistance rates were recorded to cefoxitin (16.6%, 9/54), penicillin G (13%, 7/54) and oxacillin (14.8%, 8/54).
Antimicrobial non-susceptibility rates of different species isolated from camel milk samples in the present study
Table 3 Long description
The table reports, for each bacterial species isolated from camel milk, the count and percent of isolates that were not susceptible to each antibiotic. Coagulase-positive S. aureus shows notable non-susceptibility to aminoglycosides and clindamycin, with two thirds not susceptible to gentamicin, kanamycin, neomycin, and clindamycin, and one third to penicillin G. Among coagulase-negative staphylococci, S. warneri is consistently high, reaching 100% not susceptible to neomycin and nalidixic acid and 80% to chloramphenicol, while S. hominis1 reaches 100% for chloramphenicol and nalidixic acid and 80% for oxacillin. S. capitis and S. sciuri show 100% not susceptible to several agents, including gentamicin and kanamycin for S. capitis and cefoxitin, chloramphenicol, and nalidixic acid for S. sciuri, but these species have very small sample sizes. S. simulans has moderate to high non-susceptibility for chloramphenicol and clindamycin and lower levels for beta-lactams such as penicillin G and oxacillin. Non-staphylococcal groups show selective resistance patterns: Cellulomonas spp. has about one quarter not susceptible to gentamicin and streptomycin and about one third to tetracycline, while Lactococcus lactis shows high non-susceptibility to kanamycin and gentamicin and moderate levels to neomycin and tetracycline. Across species, several antibiotics show no detected non-susceptibility in this dataset, including tylosin, lincomycin, and ciprofloxacin. Percentages should be interpreted cautiously because many species have very few isolates, so a single isolate can drive a very high percent.
Notes: N, number; %, percentage; CoPS, coagulase-positive staphylococci; CoNS, coagulase-negative staphylococci; K, Kocuria; not susceptible: R + I (I, intermediate; R, resistant); A, Aerococcus; L, Lactococcus; S, Staphylococcus.
Concerning Cellulomonas isolates, moderate resistance was observed against tetracycline (31.25%, 5/16), gentamicin (25%, 4/16), as well as both kanamycin and neomycin (18.7%, 3/16 each). For L. lactis subsp. lactis, relatively high resistance percentages were observed for aminoglycoside antibiotics, including gentamicin (62.5%), kanamycin (75%) and neomycin (50%). Similarly, identified Aerococcus viridans showed high resistance levels against gentamicin (57%), neomycin, tetracycline (42.5%) and nalidixic acid (28.5%). Finally, resistance to trimethoprim-sulfamethoxazole, ciprofloxacin, nitrofurantoin, tylosin and rifampicin was rarely evidenced for all tested bacteria (Table 3).
MDR isolates and associated risk factors
A total of 10 different isolates were categorized as MDR bacteria (11.7%) and varied according to bacterial isolates and the studied potential risk factors. A rate of 11% of the identified staphylococci isolates were considered as MDR with a significant difference between CoPS (28.5%) and CoNS (8.5%) (Table 4). Meaningful high rate of MDR (25%) was identified among Lactococcus isolates compared to 14.2% for A. viridans and 6.25% for Cellulomonas isolates. High rates of MDR bacteria were observed for aged female camels (16.2%, p = 0.042) and camels raised next to ruminants (37.5%, p = 0.030) compared to young ones and those reared in separate herds, respectively (Table 4). However, no significant associations were noted between the emergence of MDR isolates and other analysed risk factors.
Rates of multidrug-resistant strains isolated in this study from camel milk samples according to different risk factors
Table 4 Long description
The table reports how many multidrug-resistant isolates were found among 85 bacterial strains from camel milk, broken down by organism and by geographic zone, lactation stage, camel age, and contact with other ruminants. Overall, 10 of 85 isolates were multidrug resistant, about twelve percent. By zone, Gafsa had the highest multidrug-resistant proportion at about twenty three percent, compared with about ten percent in Kebili, about seven percent in Tataouine, and about twelve percent in Benguerden. By lactation stage, multidrug resistance ranged from about six percent in the fourth stage to about eighteen percent in the third stage, with no statistically significant difference reported. By age, camels ten years or older had a higher multidrug-resistant rate (about sixteen percent) than camels under ten years (about eight percent), and this difference was statistically significant. Contact with ruminants was associated with a much higher multidrug-resistant rate (about thirty eight percent with contact versus about six percent without), also statistically significant. Across organisms, multidrug resistance was observed in Staphylococcus aureus (six isolates total), several coagulase-negative staphylococci, Cellulomonas species, Lactococcus lactis subspecies lactis, and Aerococcus viridans, while several listed species had no multidrug-resistant isolates; some organism-by-factor cells are blank or sparse, so subgroup comparisons should be interpreted cautiously.
Notes:
a CoPS, coagulase-positive staphylococci;
b CoNS, coagulase-negative staphylococci;
c CI, 95% confidence interval.
Molecular characterization of revealed AMR profiles
Based on the observed phenotypic resistance profiles, molecular screening for antibiotic resistance genes was performed for the identified MDR bacteria. Given their epidemiological and clinical relevance, only MDR isolates were screened by PCR and are presented in Table 5. Notably, combinations of major resistance-encoding genes were detected in all tested MDR bacteria, including genes conferring resistance to aminoglycosides (aph 3-IIIa, ant 6-Ia), erythromycin and clindamycin (ermA and msrB), chloramphenicol (Catpc221), tetracycline (tet K) and trimethoprim-sulfamethoxazole (drfA) (Table 5).
Characteristics of the identified MDR isolates recovered from camel milk samples in the present study
Table 5 Long description
The table lists multidrug-resistant bacterial isolates from camel milk, giving each isolate’s species, sample code, antibiotic resistance pattern, detected resistance genes, and whether it occurred in each geographic zone, camel age group, and ruminant-contact category. Two coagulase-positive Staphylococcus aureus isolates show resistance to penicillin and multiple aminoglycosides and carry aminoglycoside genes aph(3)-IIIa and ant(6)-Ia; one is from Gafsa in older camels without ruminant contact, and the other is from Benguerden in younger camels with ruminant contact. Coagulase-negative staphylococci include S. simulans, S. xylosus, S. warneri, and S. capitis, mostly recorded in Kebili, with broad resistance that often includes cefoxitin, clindamycin, and nalidixic acid, and genes such as ermA, tet(K), aph(3)-IIIa, ant(6)-Ia, and Cat(pc221); S. xylosus also includes drfA. Non-staphylococcal isolates include Cellulomonas (Benguerden), two Lactococcus lactis subsp. lactis isolates (both Gafsa), and Aerococcus viridans (Tataouine), commonly showing aminoglycoside and tetracycline resistance with genes including tet(K), aph(3)-IIIa, and sometimes ermA, ant(6)-Ia, or drfA. Across isolates, aminoglycoside resistance genes and tet(K) are recurrent, and ermA appears in several staphylococci and Aerococcus, aligning with phenotypic resistance to macrolides or lincosamides where listed. Isolates occur in both younger and older camels and in both ruminant-contact categories, suggesting no single demographic group contains all MDR findings. Interpretation is limited because the table reports presence or absence by isolate rather than prevalence rates, and resistance phenotypes and gene panels may not capture all mechanisms.
Notes: CoPS, coagulase-positive staphylococci; CoNS, coagulase-negative staphylococci; OX, oxacillin; FX, cefoxitin; TS, trimethoprim/sulfamethoxazole; NA, nalidixic acid; EM, erythromycin; PG, penicillin G; TN, tetracycline; GM, gentamicin; SM, streptomycin; KM, kanamycin; NM, neomycin; CM, clindamycin; CL, chloramphenicol; NF, nitrofurantoin; ✓, presence of the encoding gene.
A variety of AMR genes was identified in the MDR isolates. The aminoglycoside resistance genes aph 3-IIIa and ant 6-Ia were detected in 10 (100%) and 8 (80%) samples, respectively. Six MDR isolates harboured ermA gene, while other resistance genes, including tetK, catpC221 and drfA were confirmed in 5, 4 and 2 MDR bacteria, respectively (Table 5). Although the observed resistant phenotypes to penicillin and oxacillin, encoding resistance genes blaZ and mecA were not detected in any isolate.
Discussion
Few studies were available for detailing and characterizing microorganisms in camel products, particularly in milk. For this, the present study highlighted the bacterial carriage and associated AMR in camel milk from Tunisia. California mastitis test was not performed in our study because the examined animals were apparently healthy and selected randomly to check the bacterial carriage in milk. The clinical and subclinical infections of the camel udders as well as incriminated bacteria will be addressed in further specific in-depth studies.
Based on conventional biochemical tests and API (Analytical Profile Index) galleries, only Gram-positive bacteria were identified with a quite low prevalence rate. However, previous investigations reported the detection of Gram-negative bacteria, yeast and mould in camel milk with low levels (Zuleka et al., Reference Zuleka, Abebe, Awot, Belayneh, Tehetna and Kumar2016; Karaman et al., Reference Karaman, Akgül, Öğüt, Canbay and Alvarez2022; Saeed et al., Reference Saeed, Amer, Keshta, Hafez, Sultan and Khalifa2022). While these methods are widely used in routine laboratory practices, they may have limited discriminatory power at the species level and carry a risk of misidentification. Molecular identification was limited to Staphylococci isolated in this study. This limitation is acknowledged, and future studies should incorporate molecular methods to confirm different bacterial identities.
In our study, Staphylococcus was the most frequently isolated genus (62.8%). This is in line with the findings shared in Saudi Arabia (Hirad et al., Reference Hirad, Ahmad, Alkhedhairy, Bahkali and Khan2018) and Algeria (Barka, Reference Barka2023; Lezzoum-Atek et al., Reference Lezzoum-Atek, Belhout, Bouchenafa and Bouayad2023). Within this genus, CoNS accounted for the majority (87%), compared to the CoPS, including S. aureus and S. intermedius. This has been earlier suggested that CoNS belongs to the normal commensal microbiota of camel milk although some species have been implicated in the pathogenic process of clinical and subclinical mastitis in dromedaries (Alebie et al., Reference Alebie, Molla, Adugna, Tesfaye and Ejo2021; Barka, Reference Barka2023). In our study, identified S. aureus showed intermediate phenotypic resistance to penicillin, and no resistance to oxacillin, and cefoxitin. However, resistance to β-lactam antibiotics, including penicillin, amoxicillin and oxacillin, was previously detected in camel milk samples (Njage et al., Reference Njage, Dolci, Jans, Wangoh, Lacroix and Meile2013). The blaZ gene responsible for penicillin resistance was not detected in the present study. This may be mediated by alternative genes or gene variants not targeted by the primers used in this study or by chromosomal mutations affecting antibiotic target sites (Jensen and Lyon, Reference Jensen and Lyon2009). In addition, the presence of silent or poorly expressed resistance genes, which are not transcriptionally active under the tested conditions, may result in phenotypic resistance without PCR amplification (Deekshit and Srikumar, Reference Deekshit and Srikumar2022). Moreover, non-enzymatic resistance mechanisms, such as altered membrane permeability, efflux pump overexpression or biofilm-associated tolerance, can confer resistance independently of known resistance genes (Galgano et al., Reference Galgano, Pellegrini, Catalano, Capozzi, Del Sambro, Sposato, Lucente, Vasinioti, Catella, Odigie, Tempesta, Pratelli and Capozza2025). These findings highlight the complexity of AMR and indicate that phenotypic resistance does not always correlate directly with the detection of resistance genes.
In contrast to our study, methicillin heat-resistant S. aureus were detected not only in pasteurized camel milk from Saudi Arabia with stability at 85–90°C for 60 s (Aljahani et al., Reference Aljahani, Alarjani, Hassan, Elkhadragy, Ismail, Al-Masoud and Yehia2020) but also in camel retail meats in Riyadh (Hirad et al., Reference Hirad, Ahmad, Alkhedhairy, Bahkali and Khan2018).
Staphylococcus aureus isolates exhibited resistance to gentamicin, kanamycin and clindamycin with the detection of the aph(3)-IIIa encoding gene. This aligns with previous reports reflecting global trends in aminoglycoside resistance within S. aureus isolates (Njage et al., Reference Njage, Dolci, Jans, Wangoh, Lacroix and Meile2013). CoNS species, commonly known as opportunistic and low-risk pathogens, are increasingly involved in a broad spectrum of diseases in both animals and humans (Hirad et al., Reference Hirad, Ahmad, Alkhedhairy, Bahkali and Khan2018; Kirwa et al., Reference Kirwa, Aboge, Maitho, M, Abuom and Mainga2021). S. simulans was the most predominant isolate (25% of the isolates). This is in line with previous studies suggesting its possible implications in mastitis occurrence (Njage et al., Reference Njage, Dolci, Jans, Wangoh, Lacroix and Meile2013; Goncalves, Reference Goncalves2017).
AMR associated with CoNS species may reflect farm biosecurity, animal management and improper antibiotic use which create selective pressure favouring pathogenic Staphylococcus species and facilitating their spread via horizontal gene transfer (Getahun et al., Reference Getahun, Abey, Beyene, Belete and Tessema2024).
CoNS exhibit higher rates of MDR compared to S. aureus. Genes encoding for resistance to erythromycin (ermA), aminoglycosides (aph(3)-IIIa), tetracycline (tet(K)) and chloramphenicol (cat(pC221)) were identified in some CoNS and explain in part the identified MDR profiles.
Although resistance to β-lactam antibiotics was phenotypically found, encoding genes were not detected in our study. This is in line with previous Tunisian studies reporting the absence of mecA-positive staphylococci isolated from nasal and rectal swabs from healthy camels (Ben Chehida et al., Reference Ben Chehida, Gharsa, Tombari, Selmi, Khaldi, Daaloul, Ben Slama and Messadi2021) as well as from dairy products of cow, goats and camel (Gharsa et al., Reference Gharsa, Chairat, Chaouachi, Ben Yahia, Boudabous and Ben Slama2019). However, mecA gene carrying S. aureus was reported in pasteurized camel milk from Saudi Arabia (Yehia et al., Reference Yehia, Al-Masoud, Alarjani and Alamri2020), in hospitalized patients from Tunisia (Ben Said et al., Reference Ben Said, Thabet, Cheriet, Messadi, Gómez, Ruiz-Ripa, Sghaier, Hassen, Hassen, Torres and Abbassi2023) and in healthy individuals of close contact with animals (Ben Slama et al., Reference Ben Slama, Gharsa, Klibi, Jouini, Lozano, Gómez-Sanz, Zarazaga, Boudabous and Torres2011). CoNS resistance to penicillin, oxacillin and reduced susceptibility to cefoxitin indicates varied β-lactam resistance mechanisms, highlighting the potential role of non-S. aureus isolates in spreading antibiotic resistance in camel raw milk.
Beyond Staphylococcus, other interesting Gram-positive bacteria were identified including Lactococcus lactis subsp. lactis, Cellulomonas spp., Aerococcus viridans and Kocuria rosea.
Lactococcus lactis is a key component of the natural microbiota in camel milk. It was isolated from all studies focusing on camel milk microbiology and is known for its medicinal and dietary properties (Akhmetsadykova et al., Reference Akhmetsadykova, Baubekova, Konuspayeva, Akhmetsadykov, Faye and Loiseau2015). LAB were previously identified from Tunisian camel milk including Lactococcus lactis, Lactobacillus pentosus, Lactobacillus plantarum, Lactobacillus brevis and Pediococcus pentosaceus (Fguiri et al., Reference Fguiri, Ziadi, Atigui, Ayeb, Arroum, Assadi and Khorchani2015). These bacteria are commonly implicated in traditional and industrial dairy production, ensuring safe, nutritious and high-quality products (Redruello et al., Reference Redruello, Saidi, Sampedro, Ladero, Del Rio and Alvarez2021; Sharma et al., Reference Sharma, Lavania, Singh and Lal2021).
The present study highlighted the MDR phenotype of L. lactis isolates (25%), particularly against aminoglycosides, tetracycline, chloramphenicol and sulphonamide. This result is in agreement with Hamdaoui et al. (Reference Hamdaoui, Benkirane, Bouaamali, Azghar, Mouncif, Maleb, Hammouti, Al-Anazi, Kumar, Yadav, Choi and Meziane2024)’s study. However, L. lactis was previously reported in Tunisian camel milk as sensitive to tetracycline and kanamycin (Fguiri et al., Reference Fguiri, Ziadi, Atigui, Ayeb, Arroum, Assadi and Khorchani2015). Accordingly, resistance profiles expressed by L. lactis isolates may be attributed to the production of distinct multidrug transporters (LmrA, LmrR and LmrP), which confer resistance to a wide range of clinically relevant antibiotics (Poelarends et al., Reference Poelarends, Mazurkiewicz and Konings2002). Our findings were interestingly supported by the identification of the encoding resistance genes for erythromycin (ermA), tetracycline (tet(K)), gentamicin (aph(3)-IIIa) and sulphonamide (drfA). Given the low resistance levels to chloramphenicol and nalidixic acid, no known resistance genes were identified, in agreement with earlier reports (Hummel et al., Reference Hummel, Hertel, Holzapfel and Franz2007).
Aerococcus viridans is a Gram-positive and opportunistic microorganism commonly found in the environment, skin and soil. It has been isolated from cow milk without clear pathogenic implications as it may contribute to subclinical mastitis (Saishu et al., Reference Saishu, Morimoto, Yamasato, Ozaki and Murase2015). While A. viridans has been identified from bovine milk, its role in camel milk microbiota remains largely unexplored, with no significant studies reporting its prevalence or potential implications. The low prevalence of A. viridans in Tunisian camel milk detected in our study (8.5%) is similar to those reported from Turkey (6.25%) (Kirkan et al., Reference Kirkan, Parin and Çapakçioglu2021) and Algeria (2.41%) (Boussouar, Reference Boussouar2017). This suggests that it may be part of the normal microbiota rather than directly associated with camel mastitis. Furthermore, one A. viridans isolate was recognized as MDR in agreement with resistance patterns reported from Algerian camels (Boussouar, Reference Boussouar2017). Although these resistance rates are very low and categorized as intermediate, they still raise a question about the origin of this resistance. Future research should focus on comprehensive microbiological and molecular investigations to clarify the role of A. viridans in camel milk and determine whether it has any significant impact on milk quality or public health concern.
Interestingly, Kocuria rosea was detected in the present study. This aligns with the findings reported in dromedaries from Saudi Arabia with a relatively high prevalence (ALKahtani, Reference ALKahtani2015; Hirad et al., Reference Hirad, Ahmad, Alkhedhairy, Bahkali and Khan2018). Although the identified K. rosea isolate exhibited no antibiotic resistance, serious multiresistance profiles have recently been reported against the second and third generation cephalosporins (39.6%), quinolones (36.6%) and vancomycin (47%) (Ziogou et al., Reference Ziogou, Giannakodimos, Giannakodimos, Baliou and Ioannou2023). Besides, this bacterium was identified as potential human pathogen associated with endocarditis, septicaemia, colitis and septic shock (Lai et al., Reference Lai, Wang, Lin, Tan, Wang, Liao, Chou, Huang, Lin and Hsueh2011) and even to human death (Fowora et al., Reference Fowora, Omolopo, Aiyedogbon, Abioye, Oladele, Tajudeen, Ogunniyan, Olanlege, Onyeaghasiri, Edu-Muyideen, Iwalokun, Adegbola and Salako2023; Ziogou et al., Reference Ziogou, Giannakodimos, Giannakodimos, Baliou and Ioannou2023). Thereby, ongoing surveillance is essential to safeguard public health, particularly among nomadic populations who commercialize and consume raw camel milk.
Previous studies have documented the emergence and the prevalence of antibiotic-resistant bacteria in numerous animal products. However, there is limited data on the occurrence of such bacteria in camel milk from Tunisia. In the current study, ten different Gram-positive bacteria were identified as MDR representing 11.7% of the total isolates. Notably, significant higher prevalences were observed in aged females (16.2%) and in camels raised in close contact with goats, sheep and/or cattle (37.5%). This observation is in line with previous reports indicating that udder abnormalities and mastitis are more prevalent in late lactation stages compared to early lactation, particularly in older females (Balemi et al., Reference Balemi, Gumi, Amenu, Girma, Gebru, Tekle, Ríus, D'Souza, Agga and Kerro Dego2021). Older animals may be more susceptible to infection due to cumulative exposure to pathogens, age-related declines of immune function and repeated antimicrobial treatments, which can select for resistant bacteria.
The higher MDR prevalence in camels raised alongside other ruminants suggests that interspecies contact may facilitate the transmission of resistant bacteria or resistance determinants, highlighting the potential role of mixed-species management in the spread of AMR. Similar patterns of increased MDR prevalence associated with animal age and co-housing with other species have been reported in dairy cattle and small ruminants, where management practices, environmental contamination and close physical contact contribute to cross-species dissemination of resistant strains.
In Tunisia, camels are commonly reared in extensive farming systems, often associated with ruminants, which expose them to significant microbial challenges. Mixed farms harbour a rich and heterogenous microbiota, encompassing bacteria from animals, human and environment. These initiate interactions, facilitate and promote the exchange and transfer of resistance genes. Raw milk is highly perishable mainly through its diverse bacterial charge and through poor hygiene practices during milking and handling (Qin et al., Reference Qin, Huang, Yang, Zhao, Zhao, Xu and Zhang2023). More concerningly, studies indicate that 88% of African camel farmers used antibiotics without veterinary supervision, and 79% consumed milk without adhering to the required withdrawal periods after antibiotic administration (Mohmed et al., Reference Mohmed, Barre, Mohamud, Gaciye and Hirsi2024).
Conclusion
Our study reveals diverse AMR patterns among Gram-positive bacteria isolated from camel milk in Tunisia. Although antibiotic use in camel farming is limited, resistant bacteria are increasingly detected in camel milk and meat, underscoring the expanding problem of AMR. Close interactions between rural and urban environments facilitate the spread of resistant bacteria, posing a public health risk. These findings emphasize the need for continuous surveillance of methicillin-resistant Staphylococcus in dairy products and for regular, integrated monitoring of Gram-positive bacteria across animal, human and environmental reservoirs to control AMR dissemination.
Funding statement
This work was supported by the research laboratory ‘Epidemiology of enzootic infections in herbivores in Tunisia: application to the control’ (LR16AGR01), which is funded by the Ministry of Higher Education and Scientific Research of Tunisia.
Competing interests
The authors declared that they have no potential conflicts of interest with respect to the research, authorship and publication of this article.
Author contributions
RS and LM conceived the idea. RS, GT and FBC performed the experiments. RS and HBY collected samples. RS, GT and LM performed the bacteriological and risk factors analysis. RS, GT and FBC performed the molecular protocol. RS, GT and LM wrote the manuscript and finalized it. All authors read and approved the final manuscript.
Ethics approval
Before collecting any information or samples, the study was explained to camel breeders, and their oral consent was obtained. Milk samples were collected by experienced veterinarians who ensured that the animals were handled according to the national ethical guidelines for animal welfare.
