The code of ethics was obtained from the Ethics Committee on Animal Experimentation of the Joseph KI-ZERBO University (permit no. CODE AGREEMENT: CEEA-UJKZ/2021-04).

The authorization to collect bats was obtained through the implementation of the framework cooperation agreement between Joseph Ki-ZERBO University and the Ministry of the Environment, Green Economy and Climate Change (MEECVCC) (permit no. 21-091 MEECVCC/SG/DGEF/DFRC).

This survey was conducted in six (06) cities in Burkina Faso and each city had several sampling points. The central region had seven sampling points as shown in Figure 1. These sites were chosen based on the presence of waterways, forest galleries, caves and abandoned houses, which are the bats’ preferred locations.

Bats were captured using mist nets from December 2020 to September 2021. Mist nets were placed in several places (e.g., dwellings, vegetation galleries, along water points). Care was taken to free the bats, and particular attention was paid to wing clearances because bats enter the nets with their wings spread and then fold them back up. The nets were opened at nightfall from 6 pm to 5 am depending on the activity of the bats.

Each individual was placed in a porous cotton bag and kept until the time of sampling. The risks of contamination were minimized as much as possible by using protective suits during the capture and handling of the bats. In order to identify bats, we determined weight, sex, and forearm length. Weight was measured with the pressure scale. Observation of the presence of male or female genitalia was used to determine the sex of the animal [22]. Bats were identified using the guide and identification key [23]-[25].

Figure 1. Site of capture of bats with the abundance of bats captured on each site.

Sterile swabs were used to collect oral and rectal samples from bats under aseptic conditions. Each swab was removed from its packaging without allowing the tip to come into contact with any surface, including gloves. Oral samples were collected by gently rotating the swab inside the oral cavity and behind the tongue. Rectal samples were obtained by carefully inserting the swab tip into the anus and gently rotating it against the mucosal surface for two to four circular motions. Excess fecal material was removed by gentle shaking before placing the swab into a cryovial containing transport medium.

Bacterial isolation was performed using selective and differential media. Swabs were inoculated onto URIselect, Hektoen, and Salmonella-Shigella (SS) agar plates and incubated at 37˚C for 24 h. Suspected colonies were subcultured for purification and identified using the API 20E system (bioMérieux, France).

Antimicrobial susceptibility testing was conducted using the disk diffusion method on Mueller-Hinton agar, in accordance with the guidelines of the Antibiogram Committee of the French Society of Microbiology (CA-SFM). The antibiotics tested included ceftriaxone (CRO; third-generation cephalosporin), amoxicillin-clavulanic acid (AMC; β-lactam), and imipenem (IMI; carbapenem).

A bacterial suspension was prepared from a pure colony in 5 mL of sterile physiological saline and adjusted to 0.5 McFarland turbidity. The suspension was evenly spread onto Mueller-Hinton agar plates (4 mm thickness). Antibiotic discs were aseptically placed on the agar surface at a minimum distance of 20 mm from each other. Plates were incubated at 37˚C for 18 - 24 h.

After incubation, inhibition zone diameters were measured using a calibrated ruler and interpreted according to CA-SFM breakpoints. Isolates were classified as susceptible (S), intermediate (I), or resistant (R). Detection of extended-spectrum β-lactamase (ESBL) production was performed using the double-disk synergy test, based on the interaction between clavulanic acid and third-generation cephalosporins [26].

Quality control for antimicrobial susceptibility testing was performed using reference strains, including Escherichia coli ATCC 25922 (susceptible control) and Klebsiella pneumoniae ATCC 700603 (ESBL-positive control), to validate test performance [13].

For bacterial nucleic acid extraction, genomic DNA was extracted using the Sacace^™ DNA-Sorb-B nucleic acid extraction kit (Version 2020, K-1-1/B/100) according to the manufacturer’s instructions. Real-time polymerase chain reaction (qPCR) was performed to detect two resistance-associated genes, bla_NDM and bla_IMP, which encode New Delhi metallo-β-lactamase and IMP-type metallo-β-lactamase, respectively.

Each PCR run included appropriate quality controls. DNA from known positive strains was used as a positive control (Escherichia coli ATCC 25922), while sterile distilled water was used as a negative control in place of template DNA to detect potential contamination and ensure the reliability of the amplification results (Table 1).

Table 1. Representation of the different primers for real-time PCR.

Statistical analyses were performed using R software (version 4.1.0). Data normality was assessed using the Shapiro-Wilk test and indicated that the variables were not normally distributed. Consequently, non-parametric statistical methods were applied.

Comparisons among groups were conducted using the Kruskal-Wallis rank-sum test for continuous variables. Associations between categorical variables were assessed using Pearson’s Chi-squared test when expected cell frequencies were sufficient. For contingency tables with small sample sizes or expected counts below five, Fisher’s Exact Test was applied to ensure statistical validity.

Proportions and statistical comparisons were calculated using the tbl_summary and add_p functions of the gtsummary package. Standardized Pearson residuals derived from Chi-squared tests were computed to identify cells contributing most strongly to significant associations and were visualized using the assocplot function from the reprex package.

All statistical tests were two-tailed, and a p-value < 0.05 was considered statistically significant.

Of a total of 204 bats captured, 84 bats were positive for the presence of at least one enterobacterium with a prevalence of 41.17%. The infected bats were composed of six genera and nine species, including one frugivore (Epomophorusgambianus (Ogilby, 1835)) and eight insectivores (Hyposideroscaffer (Sundevall, 1846)), Hyposiderosjonesi Hayman, 1947, Mopscondylurus (A. Smith, 1833), Mopsmidas (Sundevall, 1843), Mopspumilus (Cretzschmar, 1830), Pipistrellusnanulus (Peters, 1868), Rhinolophusalcyone Temminck, 1853, Scotophilusleucogaster (Cretzschmar, 1826). Of the nine bat species, 29 species of enterobacteria were isolated and tested with three families of antibiotics (Table 2).

Table 2. The different bacterial species with their resistance to CRO, AMC, IMI antibiotics isolated from bat species

Continued

Continued

Of all the bacteria isolated from bats in Burkina Faso, the majority are sensitive to third generation cephalosporins (CRO), betalactamines (AMC) and almost totally to carbapenems (IMI). There is no difference in resistance and susceptibility of the bacteria between AMC and CRO. The table below shows that 35% of the bacterial species are resistant to 3^rd generation cephalosporins, 46% are resistant to betalactam (AMC) and 2.4% are resistant to carbapenems (Figure 2).

Figure 2. Percentage of bacterial profiles according to the three families of antibiotics.

Resistance to CRO was observed in 11 species of bacteria and was 100% in Salmonella Typhi and Salmonellaparatyphi A. 13 species showed resistance with AMC and this resistance was 100% in six species, namely Proteus mirabilis, Salmonella Typhi, Salmonellaparatyphi, Salmonellaparatyphi A, Salmonellaspp., and Shigella spp. With IMI, resistance was only observed in two species, Salmonella spp. and Salmonellaparatyphi A and it was 100%. (Table 3). Thus, it is clear from Figure 3 that two species of bacteria are resistant to all three families of antibiotics and these are Salmonella spp. and Salmonellaparatyphi A. Two species are resistant only to CRO and 13 species only to AMC (Table 3).

Table 3.Abundance and percentage of resistance of bacterial species to the three antibiotic families.

Continued

Figure 3. Distribution of the number of bacterial species according to their resistance to the three families of antibiotics.

When evaluating the bacterial resistance profile of bat species to Ceftriaxone, the presence of variability within bat species (p < 0.05) was found. Indeed, the species R.alcyone, M. pumilus and H. jonesi are the bat species that harboured bacteria that were all resistant to Ceftriaxone and therefore to third-generation cephalosporins (Table 4).

Table 4. Bacterial resistance profile of bat species in relation to Ceftriaxone.

¹n (%); ²Kruskal-Wallis rank sum test.

Across all bat species, there was no difference in the resistance of their bacteria to amoxicillin plus clavulanic acid (p > 0.05). However, species such as Hyposideroscaffer, Hyposiderosjonesi,Mops pumilus and Rhinolophusalcyone harboured bacterial species that were all 100% resistant to amoxicillin and clavulanic acid (Table 5).

Table 5. Bacterial resistance profile of bat species in relation to amoxicillin and clavulanic acid.

¹n (%); ²Kruskal-Wallis rank sum test.

As for imipenem, the leading carbapenem, there was no variability between bat species in terms of resistance. However, it appears that only the bacterial species isolated from Mopscondylurus showed resistance to this family of antibiotics (Table 6).

Table 6. Bacterial resistance profile of bat species according to Imipenem.

¹n (%); ²Kruskal-Wallis rank sum test.

The analyses show that the resistance of bacteria to the three families of antibiotics is not related to the sex of the bats. Indeed, the Pearson’s Chi-squared test showed an absence of significant difference (p > 0.05). However, there was a strong tendency for bacteria isolated from female bats to be resistant to CRO (CRO residues = 1.12) and IMI (IMI residues = 0.13), whereas bacteria isolated from males showed a strong tendency for resistance to AMC (AMC residues = 1.38) (Figure 4).

Figure 4. Residues and number of observations of resistance to the three families of antibiotics of isolated bacteria according to the sex of the bats with Pearson’s Chi-squared test.

Among the 30 specimens of ROC-resistant bacteria, 43% and 57% were from female and male bats, respectively. Of the 40 specimens of AMC-resistant bacteria, 52% and 48% were from female and male bats, respectively. The two IMI-resistant bacteria were only isolated from female bats (Figure 5).

Figure 5. Percentage of bacterial resistance to the three families of antibiotics according to the sex of the bats.

Figure 6 shows the percentage of metallo-betalactamase (NDM) and Imipenemase (IMP) resistance genes in the population of bacterial strains. The NDM gene is abundant, with a rate of 97%, compared with 3% for the IMP gene.

Figure 6. Percentage of resistance genes according to their occurrence.