As part of this cross-sectional study investigating the bacterial profile and antibiotic bioresistance of residual bacteria colonizing plastics in Lake Nokoué at the Abomey-Calavi pier, sampling was carried out in areas characterized by a high concentration of plastic waste and intense anthropogenic activities. Three sampling points were selected based on their environmental relevance and human activity gradients. At each site, five samples were collected during three successive sampling campaigns conducted between August and October 2024, resulting in a total of 45 samples.

Figure 2 provides a detailed illustration of the sampling locations in Abomey-Calavi pier area.

For bacterial culture, the following media were used: Kligler Iron Agar (KIA), Simmon’s Citrate Agar, Mannitol Motility Medium, Urea Indole Medium, Mannitol Salt Agar (MSA), Eosin Methylene Blue Agar (EMB), Cystine-Lactose-Electrolyte-Deficient Agar (CLED), and Mueller-Hinton Agar (MH). The main laboratory equipment employed included a HETTICH EBA 20 centrifuge, an incubator, and a light microscope.

The bacterial isolation process consisted of three main stages:

1) Fresh State Examination: This preliminary step involved microscopic observation of bacterial mobility, shape, and arrangement, providing early indications of cell viability and morphology.

2) Inoculation: Microorganisms were cultured on selective and differential media to ensure optimal isolation and growth of distinct bacterial colonies.

3) Incubation: The inoculated media were incubated under controlled environmental conditions to promote bacterial growth, followed by colony enumeration and microscopic examination.

4) Biochemical identification tests were subsequently performed using the Leminor Gallery method.

The Kligler Iron Agar was used for the identification of Enterobacteriaceae, supplemented by Simmon’s Citrate Test to distinguish citrate-utilizing enterobacteria capable of alkalinizing the medium through pH variation. The Urea-Indole Test was applied for urease activity and indole production, while the Mannitol Motility Test was used to assess bacterial motility ( Figure 3 ).

Table 1 below presents the list of antibiotics tested to assess their effectiveness against bacterial bioresistance.

Antibiogram: The antibiotic bioresistance of bacterial isolates was assessed using the disk diffusion method on Mueller-Hinton agar.

Based on the diameter of the inhibition zones, bacterial strains were classified as resistant, sensitive, or intermediate (if the diameter fell between the two critical values).

Sensitive strains (S): Strains classified as S have a high probability of successful treatment, regardless of whether the antibiotic is administered orally or systemically.

Resistant strains (R): Strains classified as R have a high probability of treatment failure, irrespective of the route of antibiotic administration.

Intermediate strains (I): Strains classified as I exhibit inhibition zone diameters greater than the lower critical threshold but less than or equal to the upper critical threshold. These represent cases where the administered dose may be marginally effective or potentially suboptimal.

Figure 4 illustrates the results of the microbial flora enumeration.

After incubation on CLED agar, bacterial concentrations were observed to vary significantly, ranging from 10³ to 10⁶ CFU/mL. The first two sampling campaigns exhibited higher bacterial loads compared to the third campaign.

The results of the incubation of culture media, analyzed macroscopically, are presented in Figure 5 and Figure 6 .

Analysis of the samples revealed predominantly circular colonies exhibiting pink to yellow coloration, with sizes ranging from small to large. Most colonies were opaque and greasy, indicating a viscous texture, while a few were translucent and dry, particularly in the first sampling campaign. The overall odor was unpleasant, although some colonies from the first campaign emitted a mild, sweet scent.

The results revealed predominantly circular colonies, ranging from small to medium in size, with a dominant pink-purple coloration and pronounced opacity. Most colonies were greasy and emitted a strong, unpleasant odor. A few colonies were larger and exhibited a mucous-like texture.

Microscopic observation under a 100× oil immersion objective revealed the presence of Gram-negative bacilli and Gram-positive cocci. Subsequent analyses allowed the identification of the following bacterial species. Based on the results, a total of seven (7) bacterial species were enumerated ( Table 2 ).

With regard to the proportions of these seven identified bacterial species, Figure 7 shows some variation.

This Figure 7 shows that the most frequently encountered bacterial species on residual plastics in Lake Nokoué waters were Gram-positive cocci, particularly Staphylococcus aureus (48.3%). This was followed by Gram-negative bacilli, including Salmonella spp. (12.7%), Citrobacter freundii (12.7%), Escherichia coli (11.5%), Salmonella arizonae (9.2%), Salmonella typhi (3.3%), and Enterobacter cloacae (2.3%).

An antibiogram was performed on 50 bacterial isolates representing the different species identified on plastics ( Table 3 ) in Lake Nokoué waters, using seven antibiotics from different classes. Based on the measured inhibition zone diameters, bacterial strains were classified into three categories for each antibiotic: sensitive, resistant, or intermediate.

S: Sensitive, R: Resistant, I: Intermediate.

The results of the antibiotic resistance/sensitivity tests for the strains are presented in Figure 8 and Figure 9 respectively for Staphylococcus aureus and Enterobacteriaceae.

As shown in Figure 8 , the highest resistance rates (100%) were observed for oxacillin (OX) and colistin (CT), indicating that the tested Staphylococcus aureus strains were fully resistant to these antibiotics. A high resistance rate was also observed for amoxicillin (AML) (78.6%), indicating substantial resistance. Antibiotics with moderate resistance rates included ertapenem (ETP) (50%), erythromycin (E) (43%), and tobramycin (TOB) (21.4%), showing partial susceptibility of the bacterial strains. Finally, the lowest resistance rate was observed for gentamicin (CN) (0%), indicating that the strains were fully sensitive to this antibiotic.

Resistance to colistin was tested because, as observations have pointed out, it is a last-resort antibiotic for serious infections, and the emergence of transmissible resistance genes (such as the mcr gene) is a global threat. This resistance is particularly important for isolated enterobacteria (such as E. coli or Salmonella) because they are common, can be multi-resistant, and resistance genes can spread easily via transmissible genes. The resistance is particularly important for isolated enterobacteria (such as E. coli or Salmonella) because they are common, can be multi-resistant, and resistance genes can spread easily via transmissible genes. This resistance is particularly important for isolated enterobacteria (such as E. coli or Salmonella) because they are common, can be multi-resistant, and resistance genes can spread easily via plasmids, compromising the effectiveness of colistin in treating serious infections.

As shown in Figure 8 , the highest resistance rates among Enterobacteriaceae were observed for oxacillin (OX) (94.4%), colistin (CT) and amoxicillin (AML) (each 86.1%), as well as ertapenem (ETP) (72.2%). These high rates indicate strong resistance of Enterobacteriaceae to these antibiotics. Moderate resistance was observed for erythromycin (E) (55.6%), indicating partial susceptibility. Finally, the lowest resistance rates were recorded for tobramycin (TOB) (11.0%) and gentamicin (CN) (8.2%), suggesting that Enterobacteriaceae were largely sensitive to these two antibiotics.

After incubation, the discs containing the antibiotics are visible on the Mueller-Hinton medium and form inhibition zones around them, where bacterial growth is prevented. Figure 10 shows the antibiotic-impregnated disks placed on Mueller-Hinton agar after incubation.

The size of the inhibition zone, i.e., the diameter around the disc where no bacteria grow, is measured to determine the degree of sensitivity of the bacteria to the antibiotic ( Figure 10 ). A larger zone indicates greater sensitivity of the bacteria to the drug.

The study revealed a high prevalence of Staphylococcus aureus (48.3%) on plastics from Lake Nokoué, with bacterial concentrations ranging from 10³ to 10⁶ CFU/mL. This pathogen is well-known for causing skin infections, foodborne illnesses, and mucosal infections. Its elevated presence is associated with poor waste management and wastewater discharges. Other studies, such as [9] also report high bacterial contamination in Lake Nokoué, highlighting the link between organic pollution and pathogen prevalence. Our results are comparable to those of [10] who reported an average S. aureus load of 2511 CFU/100 mL in the Banco River, exceeding both the Quebec standard (30 CFU/100 mL) and the recommendation (20 CFU/100 mL) for surface waters intended for recreational use. These findings indicate a significant public health risk, particularly for populations living or working near the lake, who are exposed to these pathogenic bacteria.

The study also revealed a significant presence of thermotolerant coliforms, including Citrobacter freundii (12.7%), Escherichia coli (11.5%), and Enterobacter cloacae (2.3%), representing 26.5% of the isolated bacteria. These fecal indicator bacteria had concentrations ranging from 10³ to 10⁶ CFU/mL, substantially exceeding the WHO limit of 2000 CFU/100 mL for surface waters [11] . E. coli, in particular, is a reliable indicator of fecal contamination [12] , reflecting continuous exposure to human and animal fecal matter. This aligns with [13] who emphasizes the importance of E. coli in assessing microbiological water quality. Other studies, such as [10] reported high E. coli (1322 CFU/100 mL) and enterococci (2067 CFU/100 mL) counts in the Banco River, confirming significant fecal pollution. Our results also confirmed the presence of Salmonella spp. (12.7%), S. arizonae (9.2%), and S. typhi (3.3%), pathogens associated with severe infections, particularly in areas where untreated water is used for consumption or irrigation. These observations are consistent with [14] and [15] , who showed that surface water fecal contamination is often linked to domestic and industrial discharges. Additionally, [16] reported extremely high total coliform counts (7.4 × 10³ to 6.4 × 10¹⁰ CFU/100 mL) in the Oued Hassar, Casablanca, far exceeding our study’s levels. Our findings are similar to [17] , who found coliform concentrations ranging from 10² to 10⁴ CFU/100 mL in the Fès and Sebou rivers.

The S. aureus strains exhibited full resistance to oxacillin (OX) and colistin (CT), both at 100%. Oxacillin, a β-lactam, is used to detect methicillin-resistant strains, and resistance is mediated by the mecA gene encoding PBP2a, which prevents β-lactams from acting effectively [18] . This resistance is common in nosocomial infections, often requiring alternative treatments such as vancomycin [19] . Colistin’s intrinsic ineffectiveness against S. aureus is due to the Gram-positive cell wall structure, so its resistance has no direct clinical impact [20] [21] . High resistance to amoxicillin (AML) (78.6%) was also observed, primarily due to β-lactamase production that hydrolyzes the β-lactam ring [22] . Some strains may acquire additional resistance genes via plasmids, resulting in multidrug resistance, including macrolides, lincosamides, and fluoroquinolones, complicating treatment further [23] [24] . Moderate resistance was observed for ertapenem (ETP) (50%), erythromycin (E) (43%), and tobramycin (TOB) (21.4%), indicating partial antibiotic effectiveness. No resistance was detected against gentamicin (CN), confirming its continued efficacy for severe infections [25] .

Enterobacteriaceae showed particularly high resistance: 94% to oxacillin, 87.8% to colistin, and 84.8% to amoxicillin and ertapenem, reflecting a concerning multidrug-resistant trend. Oxacillin resistance is expected, as Gram-negative bacteria have an outer membrane limiting β-lactam activity [26] . Colistin resistance (87.8%) is alarming, often caused by lipopolysaccharide modifications associated with the mcr-1 gene [27] . Amoxicillin resistance (84.8%) is related to extended-spectrum β-lactamases (ESBLs) common in clinical settings [28] . Carbapenem resistance (ertapenem) is also concerning, mediated by enzymes such as KPC, NDM, or OXA-48, leaving few therapeutic options [29] . Lower resistance rates were observed for tobramycin (11.0%) and gentamicin (8.2%), highlighting their value as treatment options for severe infections [25] .

These results emphasize the adaptability of Enterobacteriaceae and S. aureus to antibiotics, complicating treatment. They underscore the urgent need for enhanced antibiotic stewardship and stricter strategies to limit the spread of resistant bacteria.

Based on the results, the following recommendations were made: