First report of the occurrence, phenotypic and molecular characterization of ESBL-producing Escherichia coli and Salmonella spp. isolated from wildlife used as bushmeat in Cameroon

¹Department of Veterinary Public Health, School of Veterinary Medicine and Sciences, University of Ngaoundere, Ngaoundere, Cameroon, ²Laboratory of Microbiology, Infectious Diseases, and Immunology, Centre for Research on Health and Priority Pathologies, Institute of Medical Research and Medicinal Plant Studies (IMPM), Ministry of Scientific Research and Innovation, Yaoundé, Cameroon, ³Pharmacology and Drugs Discovery Laboratory, Centre for Research on Medicinal Plants and Traditional Medicine, Institute of Medical Research and Medicinal Plants Studies (IMPM), Ministry of Scientific Research and Innovation, Cameroon, ⁴Centre Pasteur du Cameroun, Annex of Garoua (CPC-AG), Garoua, Cameroon

^&Corresponding author
Karyom Djim-Adjim-Ngana, Department of Veterinary Public Health, School of Veterinary Medicine and Sciences, University of Ngaoundere, Ngaoundere, Cameroon

Introduction    

Antimicrobial resistance (AMR) is recognized as one of the most critical global health challenges due to the widespread dissemination of resistance genes across human, animal, and environmental reservoirs [1,2]. The most concerning resistance mechanisms are extended-spectrum β-lactamases (ESBLs), enzymes that hydrolyze a broad spectrum of β-lactam antibiotics, including third-generation cephalosporins, and are frequently associated with multidrug resistance [3]. Within the Enterobacteriaceae family, E. coli and Salmonella spp. are both significant commensals and pathogens. E. coli, a common inhabitant of the gastrointestinal tract, is a well-documented reservoir of antimicrobial resistance genes, including ESBLs [4,5]. Salmonella spp. are globally important foodborne pathogens that are increasingly found to harbor ESBL genes that complicate treatment and control [5,6]. Although the emergence of ESBL-producing Enterobacteriaceae has been extensively studied in humans, livestock, and companion animals [3,7], increasing evidence suggests that wildlife may also serve as important reservoirs and amplifiers of ESBL genes (blaCTX-M, blaTEM, blaSHV, and blaOXA).

Reports from Africa, Europe, Asia, and the Americas have documented ESBL-producing E. coli and Salmonella in a variety of wild species, including rodents, bats, birds, and nonhuman primates [8,9]. For example, high prevalence rates have been reported in wild rats in Gabon [10], bats and birds in Nigeria [11,12], birds in Turkey [13], nonhuman primates in the Peruvian Amazon [14], and red kangaroos in China [15]. These findings highlight the potential role of wildlife, particularly in ecosystems under anthropogenic pressure, as key players in the transmission of AMR at the human-animal-environment interface. In sub-Saharan Africa, bushmeat is a vital source of protein for both rural and urban populations [16]. However, hunting, handling, and consumption of wild animals may facilitate the transmission of zoonotic and antimicrobial-resistant pathogens [17]. Despite this risk, data on the prevalence and genetic diversity of ESBL-producing bacteria in bushmeat remain scarce, particularly in Central Africa. To date, no study has characterized this phenomenon in Cameroon, a country where bushmeat consumption and trade are widespread [16]. This study aimed to fill this knowledge gap by reporting, for the first time, the prevalence and molecular characteristics of ESBL-producing E. coli and Salmonella spp. isolated from wildlife intended for human consumption in Cameroon. By combining phenotypic and genotypic analyses, our findings provide essential baseline data for assessing the role of bushmeat in the dissemination of AMR, information which is critical for public health, food safety, and One Health surveillance strategies in the region.

Methods     

Study design and setting: this cross-sectional study was conducted between September 2023 and December 2024. This study was conducted in four regions of Cameroon, namely Centre, Far North, North, and South, which were selected for their ecological diversity, wildlife abundance, and varying levels of human-wildlife interaction. A total of 427 fecal samples were collected from 14 wild animal species across four regions of Cameroon, as previously described [18].

Bacterial isolates: among these, 282 E. coli isolates and 116 Salmonella spp. isolates were recovered via standard microbiological techniques. Following antibiotic susceptibility testing, 84 MDR isolates were identified, including 73 E. coli and 11 Salmonella spp., which were preserved for this study. Identification of MDR isolates was confirmed via conventional biochemical tests and the API 20E gallery system (bioMérieux, France), following the manufacturer’s instructions.

Antimicrobial susceptibility test: all bacterial isolates were tested against 18 antibiotics representing different antibiotic groups via the disk diffusion method in accordance with the guidelines of the Clinical Laboratory Standards Institute (CLSI) [5].

Phenotypic detection of extended‐spectrum β‐lactamases (ESBLs): phenotypic characterization of ESBL-producing isolates was performed via the double disc synergy test (DDST) in accordance with CLSI guidelines [5]. This involves placing an amoxicillin-clavulanic acid disc at the center of the plate, surrounded by discs of third- or fourth-generation cephalosporins (ceftriaxone, 30 μg; ceftazidime, 30 μg; cefotaxime, 30 μg; or aztreonam, 30 μg) [5]. A standardized inoculum of each isolate was used to inoculate a Mueller-Hinton agar plate. The plate was then incubated at 35 ± 2° C for 16 to 24 hours. A positive result was indicated by an enhanced zone of inhibition around the cephalosporin discs, which showed a characteristic "champagne cork" or "keyhole" effect, indicating the presence of ESBL.

DNA extraction: DNA extraction was carried out via a commercial kit (QIAGEN®, Germany) following the manufacturer’s instructions.

Molecular characterization and isolation: all phenotypically confirmed ESBL-producing E. coli and Salmonella spp. isolates were screened for the presence of β-lacta mase-encoding genes (blaCTX-M, blaTEM, blaSHV and blaOXA) via PCR, as described by Saeed et al. [19]. PCR was performed via the Qiagen Multiplex PCR Kit 1000 (Qiagen, Hilden, Germany) with the specific primers listed in Table 1.

Data analysis: statistical analysis was conducted via R 3.5.1 software (RStudio 1.1.442). The proportion of samples containing bacteria that were not susceptible (resistant or intermediate) to antibiotics was determined by dividing the number of samples with nonsusceptible bacteria by the total number of collected samples. The prevalence of ESBL in Salmonella spp. and E. coli was compared between categories with the chi-square test, and a p-value < 0.05 was considered statistically significant.

Ethical consideration: this study was approved by the Ethics Review Committee of the School of Veterinary Medicine and Sciences, University of Ngaoundere, Cameroon. The Ministry of Forestry and Wildlife gave authorization for samples to be collected (N°1783/L/MINFOF/SETAT/SG/DFAP/SDVEF/SC/ENJ). Verbal informed consent for sample collection was obtained from the animal owners after the researcher presented their credentials and provided a clear explanation of the study objectives and sampling procedures.

Limitations of the study: owing to limited funding, whole-genome sequencing and phylogenetic analyses could not be performed. This restricted the detection of potentially novel resistance mechanisms and the assessment of clonal relationships between isolates. Despite these limitations, this study provides valuable preliminary insights into ESBL gene carriage in bushmeat-associated wildlife in Cameroon and underscores the need for future genomic investigations.

Results     

Prevalence of digestive carriage of ESBL-producing E. coli and Salmonella spp. isolates recovered from wild animals intended for human consumption in Cameroon: overall, the prevalence of ESBL-producing isolates was 15.0% (64/427) among the wild animal population included in this study. Of these, 59 were E. coli and 5 were Salmonella spp., corresponding to carriage prevalences of 13.8% (59/427) and 1.2% (5/427), respectively. As shown in Figure 1, the overall prevalence of ESBL-producing isolates was 16.1% among bacterial isolates recovered from wild animals included in this study. The prevalence of ESBL production differed significantly between bacterial species (Figure 1i; χ² = 13.32, p = 0.000263), with E. coli exhibiting a higher proportion of ESBL-producing isolates (21%; n = 282) than Salmonella spp. (5%; n = 116). In contrast, no significant differences were observed across the four study regions (Figure 1ii; χ² = 2.20, p = 0.531), where ESBL prevalence ranged from 12% in the South Region (n = 107) to 22% in the Centre Region (n = 37), with similar proportions recorded in the Far North (17%; n = 136) and North (17%; n = 118) Regions. Likewise, ESBL prevalence did not vary significantly according to animal type (Figure 1iii; χ² = 0.69, p = 0.775), with ESBL-producing isolates detected in 15% of mammals (n = 142) and 17% of wild birds (n = 254), whereas no ESBL-producing isolates were identified among reptiles (0%; n = 2).

Species-specific prevalence of digestive carriage of ESBL-producing E. coli and Salmonella spp. among wildlife species used as bushmeat in Cameroon: a marked variation in the digestive carriage of ESBL-producing E. coli and Salmonella spp. was observed across the 14 wildlife species investigated, comprising 427 sampled animals (Table 2). The results show marked variation in ESBL carriage across species, with E. coli consistently more prevalent than Salmonella spp. Among the mammalian species, deer (n = 9) showed 2 ESBL-producing E. coli isolates, corresponding to a prevalence of 22.2%, while no ESBL-producing Salmonella spp. were detected. Tiger cats (n = 4) had 1 ESBL-producing E. coli isolate (25.0%), representing the highest prevalence of ESBL-producing E. coli among all sampled species, with no ESBL-producing Salmonella spp. detected. Squirrels (n = 14), hedgehogs (n = 8), and hares (n = 10) showed no ESBL-producing isolates of either E. coli or Salmonella spp. Porcupines (n = 36) and tortoises (n = 6) were also completely negative for both bacterial species. Among other mammals, rats (n = 60) showed a relatively high carriage, with 11 ESBL-producing E. coli isolates (18.3%) and 1 ESBL-producing Salmonella spp. isolate (1.7%). Wild boars (n = 10) harbored 2 ESBL-producing E. coli isolates (20.0%) but no ESBL-producing Salmonella. Monkeys (n = 24) carried 3 ESBL-producing E. coli isolates (12.5%) and 1 ESBL-producing Salmonella spp. isolate (4.2%), representing the highest prevalence of ESBL-producing Salmonella spp. among all wildlife species. Among the avian species, wild ducks (n = 51) yielded 6 ESBL-producing E. coli isolates (11.8%) and 1 ESBL-producing Salmonella spp. isolate (2.0%). Wild geese (n = 48) showed 8 ESBL-producing E. coli isolates (16.7%) and no ESBL-producing Salmonella spp.; Pigeons (n = 94) had the largest sample size and showed 18 ESBL-producing E. coli isolates (19.1%) and 2 ESBL-producing Salmonella spp. isolates (2.1%), making pigeons one of the major reservoirs identified in this study. Guinea fowls (n = 53) yielded 7 ESBL-producing E. coli isolates (13.2%) and 1 ESBL-producing Salmonella spp. isolate (1.9%).

Distribution of ESBL phenotypes in MDR bacterial isolates by wildlife host species and region: the results presented in Table 3 reveal marked differences in the prevalence of ESBL-producing isolates across wildlife species and regions. The distribution of the ESBL phenotype among multidrug-resistant bacteria varied considerably across wildlife species and study regions. Among multidrug-resistant E. coli isolates, the highest ESBL production rates were observed in wild boars and tiger cats from the South Region, where 100% of MDR isolates expressed this phenotype. High frequencies were also recorded in wild geese from the Far North (82%), wild ducks from the Far North (77%), pigeons from the North (70%), and wild ducks from the North (69%). Intermediate ESBL rates were observed in guinea fowl (56-62%), rats (59-64%), and monkeys (40-63%). Lower frequencies were found in porcupines (25-40%), squirrels (33-50%), and hares (33%), whereas no ESBL-producing MDR E. coli isolates were detected in turtles. For Salmonella spp. the proportions of ESBL-producing strains were generally lower than those observed for E. coli. The highest rates were recorded in wild boars (60%), followed by tiger cats and squirrels from the South Region (50% each). Moderate frequencies were observed in wild ducks from the Far North (40%), wild geese (38%), guinea fowl and rats from the Centre Region (33%), as well as rats from the South Region (31%). No ESBL-producing Salmonella spp. isolates were identified in deer, hedgehogs, hares, squirrels from the Centre Region, or porcupines from the Centre Region.

Distribution of β-lactamase genes and co-carriage among extended-spectrum beta-lactamase-producing Escherichia coli and Salmonella spp. as shown in Table 4, among the 84 bacterial isolates analysed, 64 (76.2%) exhibited an ESBL phenotype, including 58 E. coli and 6 Salmonella spp. The blaCTX-M gene was overwhelmingly predominant, being detected in 61 of the 64 ESBL-producing isolates (95.3%), including 94.8% (55/58) of E. coli and all Salmonella spp. isolates (100.0%, 6/6). In contrast, the blaSHV gene was identified in 45.3% (29/64) of isolates, whereas both blaTEM and blaOXA were detected in 20.3% (13/64) of isolates each. Regarding gene distribution, single β-lactamase gene carriage was observed in only 26 (40.6%) ESBL-producing isolates, corresponding to 39.7% (23/58) of E. coli and 50.0% (3/6) of Salmonella spp. By comparison, co-carriage of at least two β-lactamase genes was more common, occurring in 42 (65.6%) isolates overall, including 67.2% (39/58) of E. coli and 50.0% (3/6) of Salmonella spp. Among the different gene combinations, the blaCTX-M + blaSHV association was the most frequently identified, occurring in 29.7% (19/64) of ESBL isolates overall, including 29.3% (17/58) of E. coli and 33.3% (2/6) of Salmonella spp. The blaCTX-M + blaSHV + blaOXA combination represented the second most frequent pattern, accounting for 9.4% (6/64) of isolates, and was observed exclusively in E. coli. Other combinations were uncommon, including blaCTX-M + blaTEM (15.6%), blaCTX-M + blaOXA (4.7%), blaCTX-M + blaTEM + blaOXA (3.1%), and blaSHV + blaOXA (1.6%). Notably, the simultaneous carriage of all four β-lactamase genes (blaCTX-M + blaTEM + blaSHV + blaOXA) was detected in only one E. coli isolate (1.6%), indicating an exceptionally complex β-lactam resistance genotype. No isolates carried the blaTEM + blaSHV, blaTEM + blaOXA, blaCTX-M + blaTEM + blaSHV, or blaTEM + blaSHV + blaOXA combinations.

Distribution of ESBL genes according to bacterial species and wildlife hosts: the distribution of ESBL genes varied across wildlife species and bacterial species (Table 5). Overall, blaCTX-M was the predominant ESBL gene detected among both E. coli and Salmonella isolates, whereas blaTEM, blaSHV, and blaOXA occurred at lower frequencies. Among E. coli isolates, blaCTX-M was detected in all ESBL-producing isolates recovered from wild ducks (6/6, 75%), wild geese (8/8, 50%), guinea fowl (7/7, 50%), rats (11/11, 55%), monkeys (3/3, 60%), deer (2/2, 50%), tiger cats (1/1, 50%), and wild boars (2/2, 50%). Pigeons harbored the highest number of ESBL-producing E. coli isolates (18 isolates), of which 15 (40.5%) carried blaCTX-M. The blaSHV gene was the second most frequently detected ESBL determinant among E. coli, with the highest proportions observed in wild boars (2/2, 50%), tiger cats (1/1, 50%), monkeys (2/3, 40%), pigeons (11/18, 29.7%), guinea fowl (4/7, 28.6%), rats (5/11, 25.0%), and wild geese (3/8, 18.8%). In contrast, blaTEM was detected less frequently, occurring mainly in pigeons (4/18, 10.8%), wild geese (3/8, 18.8%), rats (3/11, 15.0%), wild ducks (1/6, 12.5%), and guinea fowl (1/7, 7.1%). The blaOXA gene was relatively uncommon and was identified primarily in pigeons (7/18, 18.9%), guinea fowl (2/7, 14.3%), wild geese (2/8, 12.5%), rats (2/11, 10.0%), wild ducks (1/6, 12.5%), and deer (1/2, 25.0%). Among Salmonella spp. isolates, the number of ESBL-producing isolates was limited. Nevertheless, blaCTX-M was consistently detected in all positive isolates recovered from wild ducks (1/1), pigeons (2/2), guinea fowl (1/1), rats (1/1), and monkeys (1/1). The blaSHV gene was detected only in guinea fowl and monkey isolates (50% each), whereas blaOXA was identified in pigeon and rat isolates (25% and 50%, respectively). No blaTEM gene was detected among Salmonella spp., isolates except in pigeons, where one isolate (25%) was positive. No ESBL-producing E. coli or Salmonella spp., carrying the investigated ESBL genes were detected in squirrels, hedgehogs, hares, porcupines, or tortoises.

Co-existence and distribution of ESBL genes by bacterial species: the distribution and co-occurrence patterns of ESBL-encoding genes among ESBL-positive E. coli and Salmonella spp. isolates recovered from wild animals are presented in Figure 2. As shown in Figure 2A, blaCTX-M was the most prevalent ESBL gene in both bacterial species, being detected in 94.8% of ESBL-positive E. coli isolates and in all ESBL-positive Salmonella spp. isolates (100%). Among E. coli, blaSHV was the second most frequently detected gene (46.6%), followed by blaTEM and blaOXA (20.7% each). A similar pattern was observed in Salmonella spp., where blaSHV was detected in 33.3% of isolates, whereas blaTEM and blaOXA were each identified in 16.7% of isolates.

Analysis of gene co-occurrence patterns (Figure 2B) revealed nine distinct ESBL gene combinations among the 64 ESBL-positive isolates. The most common profile consisted of blaCTX-M alone (31.2%; 20/64), followed by blaCTX-M + blaSHV (29.7%; 19/64) and blaCTX-M + blaTEM (15.6%; 10/64). Several multi-gene combinations were also identified, including blaCTX-M + blaSHV + blaOXA (9.4%; 6/64), blaCTX-M + blaOXA (4.7%; 3/64), blaSHV alone (3.1%; 2/64), and blaCTX-M + blaTEM + blaOXA (3.1%; 2/64). The quadruple combination blaCTX-M + blaTEM + blaSHV + blaOXA was detected in one isolate (1.6%), while one isolate (1.6%) carried blaSHV + blaOXA in the absence of blaCTX-M.

Extended-spectrum β-lactamase (ESBL) gene distribution in Salmonella spp. by animal species and region: the prevalence of ESBL genes according to host animal species and geographic region, among ESBL-positive Salmonella spp. isolates (n = 6), is presented in Figure 3A. The blaCTX-M gene was detected in 100% of all animal/region categories represented: wild ducks (Far North), rats (Centre), pigeons (North and Far North), monkeys (South), and guinea fowl (North). The pigeon from the North Region stood out by carrying three ESBL genes simultaneously (blaCTX-M, blaTEM, and blaOXA), each detected at 100% in this isolate. The frequency of ESBL gene co-existence patterns among Salmonella spp. is shown in Figure 3B. The blaCTX-M-alone profile was the most common (50%; n = 3/6), followed by the combination blaCTX-M + blaSHV (33.3%; n = 2/6) and the three-gene combination blaCTX-M + blaTEM + blaOXA (16.7%; n = 1/6).

Extended-spectrum β-lactamase (ESBL) gene distribution in E. coli by animal species and region: the prevalence of ESBL genes according to host animal species and geographic region, among ESBL-positive E. coli isolates (n = 58) across twelve categories, is presented in Figure 4A. The blaCTX-M gene remained nearly ubiquitous, reaching 100% in most groups, with the exception of pigeons from the North Region, where its prevalence was 70%, matched by blaSHV in the same group. Regional heterogeneity was observed: isolates from the North and Far North regions (wild geese, wild ducks, pigeons, guinea fowl) displayed more diverse gene profiles, frequently combining blaTEM, blaSHV, and blaOXA, whereas isolates from the South and Centre Regions (wild boar, tiger cats, monkeys, rats) showed more restricted profiles, generally dominated by the blaCTX-M + blaSHV combination. The frequency of ESBL gene co-existence patterns among E. coli is described in Figure 4B. The blaCTX-M-alone and blaCTX-M + blaSHV profiles were tied for the highest frequency (29.3% each; n = 17/58), followed by blaCTX-M + blaTEM (17.2%; n = 10/58) and blaCTX-M + blaSHV + blaOXA (10.3%; n = 6/58). Rarer profiles completed the distribution: blaCTX-M + blaOXA (5.2%; n = 3), blaSHV alone (3.4%; n = 2), and blaSHV + blaOXA, blaCTX-M + blaTEM + blaSHV + blaOXA, and blaCTX-M + blaTEM + blaOXA (1.7% each; n = 1).

Heatmap visualization and hierarchical clustering of antimicrobial resistance patterns in MDR Salmonella spp. and E. coli isolates: hierarchical clustering of antimicrobial resistance profiles among MDR bacterial isolates recovered from wildlife bushmeat in Cameroon is presented in Figure 5 for Salmonella spp. and in Figure 6 for E. coli. The column dendrogram for Salmonella spp. (Figure 5) identified two major antibiotic clusters. The first comprised IMP, NOR, ATM, SXT, CIP, FEP, and AZM, which exhibited relatively heterogeneous resistance patterns across isolates; IMP was the most distinct antibiotic within this cluster, with resistance detected in only one isolate. The second cluster grouped AMC, TZP, GEN, PIP, TIC, AMP, CTX, FOX, TET, CAZ, and CRO, consisting mainly of β-lactams and broad-spectrum antimicrobials that displayed highly similar resistance patterns across isolates. The row dendrogram revealed three principal isolate clusters. The upper cluster, composed mainly of mammalian isolates from the Centre and South Regions, exhibited limited and heterogeneous resistance profiles. The middle cluster included isolates from both wild birds and mammals originating primarily from the Far North and North Regions and was characterized by resistance to several β-lactam antibiotics with variable susceptibility to other antimicrobial classes. The lower cluster, composed predominantly of wild-bird isolates, displayed the broadest resistance profiles, including resistance to most β-lactams, cephalosporins, and several non-β-lactam agents.

ESBL-positive isolates were concentrated mainly within the cluster exhibiting the highest levels of multidrug resistance. No clear clustering according to geographic region was observed, as isolates from different regions were distributed across the dendrogram; however, isolates recovered from wild birds were more frequently represented within the highly resistant cluster. Resistance to imipenem remained uncommon among Salmonella spp. isolates, whereas resistance to β-lactams and extended-spectrum cephalosporins tended to occur concurrently, reflecting closely related resistance profiles among these antimicrobial classes. The column dendrogram for E. coli (Figure 6) revealed two major antibiotic clusters. The first grouped IMP, ATM, NOR, SXT, FEP, CIP, FOX, TET, GEN, and AZM, which exhibited relatively heterogeneous resistance patterns across isolates. The second comprised AMC, AMP, PIP, TZP, CTX, CAZ, TIC, and CRO, consisting mainly of β-lactams and extended-spectrum cephalosporins that displayed highly similar resistance profiles, suggesting a strong co-resistance pattern consistent with ESBL-associated resistance mechanisms. The row dendrogram identified three main isolate clusters. The upper cluster, comprising approximately 40 isolates, was characterized by extensive resistance to nearly all antibiotics tested and was predominantly composed of ESBL-positive isolates originating from different regions and animal hosts. The middle cluster displayed intermediate resistance profiles, with consistent resistance to β-lactams but variable susceptibility to IMP, ATM, NOR, and SXT. In contrast, the lower cluster, containing approximately 15 isolates, exhibited more restricted resistance patterns and consisted mainly of ESBL-negative isolates with resistance limited to a smaller number of antimicrobial classes. ESBL-positive isolates consistently clustered within the groups showing the highest levels of multidrug resistance, whereas ESBL-negative isolates were generally associated with less extensive resistance profiles. Resistance to imipenem remained uncommon across all clusters, indicating preserved susceptibility to carbapenems among the isolates studied. No distinct clustering was observed according to animal type or geographic origin, as isolates from mammals and wild birds, as well as from the Centre, Far North, North, and South Regions, were distributed throughout the dendrogram, suggesting a homogeneous distribution of resistance profiles across host groups and regions.

Association between ESBL genes and ciprofloxacin resistance by isolate, region and animal species: the hierarchical clustering of E. coli and Salmonella spp. isolates recovered from wildlife bushmeat in Cameroon, based on the combined prevalence of blaCTX-M, blaTEM, blaSHV, blaOXA, CIP, and NOR resistance, is presented in Figure 7. The resulting dendrogram identified three major clusters. The upper cluster was characterized by the highest levels of co-resistance and comprised predominantly E. coli isolates from tiger cats in the South Region, together with Salmonella spp. isolates from monkeys and guinea fowl. These groups exhibited complete carriage of blaCTX-M accompanied by 100% resistance to both ciprofloxacin and norfloxacin. In contrast, blaTEM, blaSHV, and blaOXA were infrequently detected within this cluster. The intermediate cluster displayed more heterogeneous resistance profiles, with ciprofloxacin and norfloxacin resistance ranging from 50% to 83% and blaCTX-M prevalence varying between 50% and 100%. This cluster included E. coli isolates recovered from rats in the Centre and South Regions, wild ducks from the North and Far North Regions, monkeys from the Centre Region, and wild geese from the Far North Region. Compared with the upper cluster, a higher frequency of blaSHV and blaOXA co-occurrence with blaCTX-M was observed. The lower cluster exhibited the greatest diversity of resistance profiles and included groups with high blaCTX-M prevalence but reduced or absent resistance to ciprofloxacin and norfloxacin. This pattern was particularly evident among Salmonella spp. isolates recovered from pigeons in the North and Far North Regions. These isolates clustered separately from the highly fluoroquinolone-resistant groups despite carrying ESBL determinants.

Discussion     

Antimicrobial resistance (AMR) is now recognized as one of the most pressing global health threats, and extended-spectrum β-lactamase(ESBL)-producing Enterobacteriaceae represent one of its most concerning manifestations, given their capacity to compromise first-line β-lactam therapy in both human and veterinary medicine [3,7]. Although ESBL surveillance has traditionally focused on human clinical settings, domestic livestock, and companion animals, growing evidence suggests that wildlife may represent an important, yet underexplored, reservoir of ESBL resistance determinants [8,20]. This study extends that evidence base to Cameroon, providing, to our knowledge, the first characterization of ESBL-producing E. coli and Salmonella spp. in wild animals used as bushmeat in Central Africa, and situates its findings within this broader, rapidly evolving literature.

The present study revealed substantial variation in the digestive carriage of ESBL-producing E. coli and Salmonella spp. across the wildlife species investigated. Overall, ESBL-producing E. coli was more frequently detected than ESBL-producing Salmonella spp., highlighting the potential importance of E. coli as a major reservoir and indicator of antimicrobial resistance (AMR) among wildlife used as bushmeat in Cameroon. The prevalence observed in this study is broadly consistent with, although somewhat lower than, those reported for ESBL-producing E. coli isolates from wild animals in other sub-Saharan African countries [10,11,21]. Our findings therefore place wildlife-associated ESBL carriage in Cameroon within the same order of magnitude as domestic-animal reservoirs described elsewhere on the continent, reinforcing the conclusion that ESBL dissemination in this region is not confined to clinical or peri-domestic settings but extends into wild animal populations directly exploited for human food [8,22,23]. These isolates were predominantly recovered from wild birds, non-human primates, and rodents, which are among the wildlife groups frequently reported as potential reservoirs and vectors of multidrug-resistant bacteria and a wide range of zoonotic pathogens [20]. These findings reveal an emerging and largely overlooked threat to food safety and public health in a context where bushmeat remains a major source of animal protein, particularly in rural and forest-dependent communities, where it may contribute up to 80% of the nutritional needs of Indigenous populations and rural communities [24].

The predominance of E. coli among ESBL-producing isolates is consistent with its ecological role as a ubiquitous intestinal commensal and a key reservoir of antimicrobial resistance genes [25]. Similar trends have been widely documented in human clinical settings, livestock, and animal-derived food products across sub-Saharan Africa, including Cameroon [5,8,12,26]. Salmonella spp. by contrast, occupies a fundamentally different epidemiological niche: rather than serving primarily as a resistance sentinel, it ranks among the most important foodborne pathogens worldwide, and its detection in wild animals destined for human consumption carries direct zoonotic and clinical implications [27]. E. coli as an indicator of the underlying AMR reservoir, and Salmonella as a direct agent of zoonotic foodborne disease underscore why both species must be considered jointly, rather than ranked by prevalence alone, when assessing the public health risk posed by the bushmeat value chain [27]. The transmission of ESBL-producing bacteria involves complex and interconnected direct and indirect pathways linking humans, animals, and the environment, thereby facilitating the bidirectional dissemination of antimicrobial resistance across these reservoirs [28]. ESBL-producing Enterobacteriaceae are associated with limited therapeutic options, increased morbidity and mortality, and substantial healthcare costs, underscoring the need for integrated surveillance [29,30]. Accordingly, the World Health Organization (WHO) has promoted a One Health approach, including the Tricycle Protocol, which uses ESBL-producing E. coli as a sentinel organism to monitor the circulation of antimicrobial resistance across human, animal, and environmental reservoirs [25,31].

At the molecular level, the overwhelming predominance of blaCTX-M (95.3% of ESBL-producing isolates overall, including 94.8% of E. coli and 100% of Salmonella spp.) mirrors the global and regional literature almost exactly. In Cameroon, increasing evidence suggests that ESBL-producing E. coli and their associated resistance determinants circulate across human, animal, and environmental reservoirs. Recent studies in Yaoundé have reported closely related ESBL-producing E. coli among pigs and slaughterhouse workers, with blaCTX-M being the predominant ESBL gene, suggesting potential exchange of resistant bacteria at the human-animal interface [32]. Similar evidence of blaCTX-M-mediated resistance has also been reported among human, animal, and environmental isolates in Cameroon, supporting the potential for interconnected transmission pathways within a One Health context [32]. In this context, the detection of blaCTX-M among ESBL-producing E. coli and Salmonella spp. in wildlife used as bushmeat in our study raises concern about the possible contribution of wildlife to the broader circulation of antimicrobial resistance in Cameroon. However, direct transmission between wildlife and humans cannot be established from the present data and would require genomic and epidemiological investigations to confirm such transmission pathways. The same sub-Saharan African meta-analysis identified blaCTX-M-15 as the most frequently reported ESBL gene in E. coli across the region [23,33]. Similarly, CTX-M-type ESBLs are widely reported as the predominant ESBL family among wild birds and other wildlife worldwide, where their frequent plasmid-mediated occurrence has facilitated horizontal transfer, persistence, and the global dissemination of antimicrobial resistance [34,35].

The relatively lower and more variable frequencies of blaSHV, blaTEM, and blaOXA observed in our isolates, together with the frequent co-carriage of two or more genes (65.6% of ESBL-positive isolates, dominated by the blaCTX-M + blaSHV combination), further support the hypothesis that these determinants circulate on shared, potentially conjugative plasmid platforms. The detection of a single isolate carrying all four investigated β-lactamase genes simultaneously, while numerically rare, illustrates the upper end of this genetic complexity and underscores the plasticity of β-lactam-resistance genotypes already circulating within this wildlife population.

Beyond Cameroon, the same molecular signature recurs at a broader scale: the high prevalence of blaCTX-M observed in E. coli and Salmonella isolates mirrors findings from wildlife studies conducted across Africa, Europe, Asia, and Latin America, supporting the role of wildlife as an important reservoir and interface for the maintenance and dissemination of clinically relevant resistance genes [5,12,36-39].

All multidrug-resistant E. coli and Salmonella isolates identified in this study were ESBL producers and were frequently co-resistant to fluoroquinolones and other antimicrobial classes. The significant association between fluoroquinolone resistance and resistance to extended-spectrum cephalosporins, including cefepime, is consistent with co-selection mediated by the same mobile genetic elements discussed above [40-42]. Although plasmid-mediated quinolone resistance genes were not investigated directly, the observed phenotypic patterns suggest their possible co-circulation on the same replicons, as reported in comparable settings [42-44].

Reduced susceptibility to carbapenems was detected in a subset of isolates. While carbapenems are not used in wildlife, comparable findings have been reported in livestock, wildlife, and community settings in Cameroon and other African regions [5,12,45]. The detection of OXA-type carbapenemase genes warrants caution in interpretation, as it does not by itself indicate widespread circulation but rather suggests environmental contamination and indirect selective pressures, consistent with the broader observation that carbapenem-resistance genes move across One Health sectors via plasmid platforms structurally similar to those carrying blaCTX-M [21]. Nevertheless, given that carbapenem-resistant enterobacterales are classified by the WHO as critical-priority pathogens [2], even sporadic detection in bushmeat species is of major public health significance.

From a risk-analysis perspective on foodborne antimicrobial resistance, these findings raise serious concerns. Hazard identification includes ESBL-producing and carbapenem-resistant bacteria and their associated resistance genes. Exposure is amplified by informal bushmeat value chains characterised by the absence of veterinary inspection, poor hygiene practices, and frequent direct contact during hunting and processing. Consequences are particularly severe in low-resource settings where access to diagnostics and second-line antimicrobials is limited, increasing the likelihood of difficult-to-treat infections and silent community-level dissemination.

Because the resistance determinants characterised here circulate on mobile genetic platforms that are not restricted to a single host species, control strategies confined to a single sector are unlikely to be sufficient. Cross-sectoral genomic surveillance sequencing isolates from wildlife, livestock, humans, and the environment within the same landscape and comparing plasmid and mobile-element content rather than gene presence alone has been proposed as the most informative way to trace the direction and drivers of AMR spread and to prioritise interventions where connectivity between sectors is highest [46,47]. For the bushmeat value chain specifically, this would mean pairing microbiological surveillance of hunted and traded species with monitoring at the points of greatest human contact hunting, slaughter, and market handling where exposure risk is concentrated and where targeted hygiene and inspection measures could plausibly interrupt transmission without requiring wildlife population-level intervention.

Conclusion     

This study provides the first evidence of ESBL-producing E. coli and Salmonella spp. in wild animals intended for human consumption in Cameroon, confirming that wildlife constitutes an important and previously underestimated reservoir of clinically relevant antimicrobial resistance. The predominance of blaCTX-M genes, together with the frequent co-occurrence of ESBL production, fluoroquinolone resistance, and reduced susceptibility to carbapenems, highlights a serious threat to food safety and public health. The asymptomatic carriage of multidrug-resistant bacteria in wildlife, particularly in high-risk taxa such as migratory birds (notably pigeons), rodents, and non-human primates, underscores their potential role in the maintenance and dissemination of foodborne antimicrobial resistance across ecosystems. In a context where bushmeat remains a critical source of animal protein, hunting, handling, and consumption of wild animals represent credible pathways for the transmission of difficult-to-treat infections to human populations, especially in resource-limited settings. From a foodborne antimicrobial resistance risk analysis perspective, these findings emphasize the combined impact of hazardous resistant bacteria, high exposure through informal bushmeat value chains, and severe health consequences due to limited therapeutic options. Integrating wildlife into national antimicrobial resistance surveillance and food safety frameworks is therefore imperative. Operational One Health strategies encompassing targeted surveillance of high-risk wildlife species, environmental monitoring, risk-based control measures along the bushmeat value chain, and community awareness are essential to mitigate the emergence and spread of antimicrobial resistance at the human-animal-environment interface in Cameroon and similar settings.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

All the authors read and approved the final version of this manuscript.

Tables and figures     

Table 1: distribution of children aged 0 to 59 months with severe acute malnutrition according to their sociodemographic characteristics, Kasaï-Central province, December 2024

Table 2: prevalence of intestinal carriage of extended-spectrum β-lactamase (ESBL)-producing E. coli and Salmonella spp. isolated from wild animals intended for human consumption in Cameroon

Table 3: distribution of extended-spectrum β-lactamase (ESBL)-producing multidrug-resistant (MDR) E. coli and Salmonella spp. according to wildlife species and region in Cameroon

Table 4: distribution of β-lactamase genes and co-carriage among extended-spectrum β-lactamase (ESBL)-producing E. coli and Salmonella spp. isolated from fecal samples of wild animals intended for human consumption in Cameroon

Table 5: distribution of extended-spectrum β-lactamase (ESBL) gene profiles among multidrug-resistant E. coli and Salmonella spp. recovered from wild animals intended for human consumption in Cameroon

Figure 1: prevalence of extended-spectrum β-lactamase (ESBL) producing E. coli and Salmonella spp. isolated from fecal samples of wild animals intended for human consumption in Cameroon

Figure 2: co-existence patterns of extended-spectrum β-lactamase (ESBL) genes (blaCTX-M, blaTEM, blaSHV, blaOXA) among bacterial isolates from wildlife bushmeat, Cameroon; (A): prevalence (%) of individual ESBL genes among ESBL-positive isolates of E. coli (n = 58) and Salmonella spp. (n = 6); (B): frequency of ESBL gene co-existence patterns among all ESBL-positive isolates (n = 64) (values represent the number of isolates carrying each gene combination and the corresponding percentage. ESBL = extended-spectrum β-lactamase)

Figure 3: distribution and co-occurrence patterns of extended-spectrum β-lactamase (ESBL)-encoding genes among ESBL-producing Salmonella spp. isolates recovered from wildlife bushmeat in Cameroon: (A) prevalence of blaCTX-M, blaTEM, blaSHV, and blaOXA genes according to wildlife host species and region of origin; (B) frequency of ESBL gene combinations detected among the six ESBL-producing Salmonella spp. isolates (values are expressed as percentages of the total number of ESBL-positive isolates (n = 6))

Figure 4: distribution and co-occurrence patterns of extended-spectrum β-lactamase (ESBL)-encoding genes among ESBL-producing E. coli isolates recovered from wildlife bushmeat in Cameroon; (A) prevalence of blaCTX-M, blaTEM, blaSHV, and blaOXA genes according to wildlife host species and geographic region; (B) frequency of ESBL gene combinations detected among the 58 ESBL-producing E. coli isolates (values are expressed as percentages of the total number of ESBL-positive isolates (n = 58))

Figure 5: co-resistance clustering of antibiotic resistance profiles among MDR Salmonella spp. isolates from wildlife bushmeat, Cameroon: heatmap with hierarchical clustering (Ward.D2 method, binary distance) showing antibiotic resistance profiles of Salmonella spp. isolates (rows) and co-resistance patterns among antibiotics (columns)

Figure 6: co-resistance clustering of antibiotic resistance profiles among MDR E. coli isolates from wildlife bushmeat, Cameroon; heatmap with hierarchical clustering (Ward.D2 method, binary distance) showing antibiotic resistance profiles of E. coli isolates (rows) and co-resistance patterns among antibiotics (columns) (blue =resistant; white = sensitive), row annotations indicate ESBL phenotype (ESBL+/ESBL-), geographic region, and animal type

Figure 7: association between extended-spectrum β-lactamase (ESBL) gene carriage and fluoroquinolone (NOR: norfloxacin and CIP: ciprofloxacin) resistance among E. coli and Salmonella spp. isolates recovered from wildlife bushmeat in Cameroon

References