The study was carried out in the city of Yaoundé, capital of the Centre Region of Cameroon, located at 300 Km from the Atlantic coast between latitude 3˚5' North and Longitude 11˚31' East. After the prospection, seven social housing sites were selected based on specific criteria, including ease of access, population density, type of sanitation system, configuration of latrine pits, mode of effluent discharge into the environment, and spatial distribution. These criteria were applied to ensure the representativeness of various wastewater management practices and discharge conditions observed throughout the city’s social housing network. Wastewater samples were collected from seven designated points: MDG, RPE, BIY, Cverte, MSI, MSII, and MFD. Those geographic coordinates are given in Table 1.

Table 1. Geographic coordinates of the sampling stations, sampling point descriptions, and observations.

Figure 1.Map of the study area showing sampling stations.

Figure 1shows the geographical location of the study area and the sampling stations.

The samples for the bacteriological analysis were collected in sterile glass bottles of 500 mL. Samples for physicochemical analyses were collected bubble-free by a diver in 250 mL double-capped polyethylene bottles following the techniques recommended by some authors ([38]). At the level of each sampling station, the physico-chemical analysis focused on 12 variables. Physico-chemical parameters (water Temperature (˚C), pH, electrical conductivity of the water (μS/cm), salinity (ppm), suspended solids (mg/L), water colour (Pt. Co), turbidity (FAU), nitrates (mg/L), nitrates (mg/L), orthophosphates (mg/L PO 4 3 − ) and ammonium ions (mg/L NH 4 + )) were measured by colorimetry and/or spectrophotometry according to the techniques described by some authors ([38]; [41]). The various samples were then transported to the laboratory in a refrigerated chamber to undergo the various analyses and tests required. Bacteriological analysis begins within 4 hours of sampling.

The isolation of the target genera was carried out using the surface spread method, which consists of first taking 100 μL using a sterile micropipette, then depositing the volume of water taken on the surface of the Salmonella-Shigella (SS) agar previously poured into sterile 90 mm diameter Petri dishes. The inoculum was distributed uniformly across the surface using a sterile rake or spreader until complete absorption was achieved. This exercise was carried out under sterile conditions ensured by the Bunsen burner flame ([6]). Subsequently, the plates were subjected to an incubation process at a temperature of 37˚C for a duration of 24 hours. Bacterial strains with satisfactory characteristics were counted using the direct counting method ([23]) and the results were expressed in decimal logarithmic units (Log₁₀ (CFU/mL)) of water analysed to avoid large variations during the sampling period ([31]) and also to better represent the variation and limit of the significant differences between the densities of the bacteria sought.

The identification involved first conducting a macroscopic examination, followed by a microscopic examination, and finally biochemical tests. Macroscopic observation consisted of examining the cultural characteristics of bacterial colonies on the surface of different culture media. The cultural characteristics considered were colour, size, contours, colony surface configuration, and appearance ([14]). Microscopic observation of fresh presumptive colonies allowed their mobility or immobility to be observed. Microscopic observation after Gram staining revealed the nature of the bacterial cell wall, their shapes, and their modes of grouping ([14]). Bacterial species were identified using standard tests. These tests included testing for oxidase, catalase, glucose fermentation, gas production, oxygen affinity (facultative aerobic/anaerobic), motility, mannitol, lactose fermentation, H₂S production, urease, citrate, among others ([23]).

After the biochemical identification tests, the colonies were transferred to slanted PCA medium. After 24 hours of incubation, pure bacterial cells were harvested using a sterile platinum loop and placed in a test tube containing 10 mL of sterile physiological water. The homogenised bacterial suspension was vortexed and adjusted to a density of 0.5 MacFarland (BaCl₂ and H₂SO₄ 1%), corresponding to a bacterial concentration of approximately 1.5 × 10⁸ CFU/mL. This concentration was determined by reading the optical density (OD) on a DR/2800 spectrophotometer at 625 nm ([11]). Antimicrobial susceptibility testing was performed on pure strains in accordance with the recommendations of the Antibiogram Committee of the French Society of Microbiology ([11]). Bacterial suspensions were uniformly inoculated onto the entire surface of Mueller-Hinton agar plates (4 mm thickness) using sterile cotton swabs. The inoculated dishes were dried for 15 minutes at room temperature under aseptic conditions. Subsequently, four antibiotic discs were aseptically placed and gently pressed onto the surface of the agar to ensure complete contact. The plates were then incubated in an aerobic atmosphere at 37˚C for 18 - 24 h, after which inhibition zone diameters were measured and interpreted. The antibiotics used belonged to several families. They were in the form of a disc of blotting paper, 6 mm in diameter, impregnated with well-determined quantities of active substances and rigorously controlled for the diffusion method. A total of eight antibiotics representative of major antimicrobial classes routinely employed in the treatment of infections caused by pathogenic Enterobacteriaceae were selected. The selection was based on their widespread clinical use in hospital settings. These included Beta-lactamin, macrolides, quinolones, and aminoglycoside antibiotics. Table 2 lists the antibiotics used, their groups, and their characteristics.

Table 2. Antibiotics, disc loads, and critical reference diameters for Enterobacteriaceae ([10]).

After incubation, the reading was taken by measuring the diameters of the inhibition zones using a caliper in accordance with CA-SFM recommendations ([10]), to declare the germ sensitive, intermediate, or resistant in relation to the critical diameters of the antibiotic tested. The results yielded three clinical categories: sensitive (S), resistant (R), and intermediate (I). The sensitive (S) category means that the strain is susceptible to the antibiotic treatment given at the normal dose. In the intermediate category (I), the success of the therapy is not predictable. If the antibiotic is used at a higher dose than usual, the treatment will be effective. Intermediate means moderately susceptible or moderately resistant. The resistant (R) category means that there is a high probability of treatment failure regardless of the type of antibiotic treatment, or that the strain will not be affected regardless of the type of treatment.

Histograms plotted using Microsoft Excel 2016 were used to illustrate the temporal evolution of the density of the different bacterial species studied, as well as the percentages of antibiotic sensitivity. A double-axis histogram (or dual-axis chart) was used to represent two different physicochemical variables on the same graph. SPSS 25.0 software was used to calculate Spearman’s rank correlation coefficient “r”. Correlations between biological and abiotic variables were established using this coefficient. To characterize the sampling stations based on all the physicochemical parameters measured and the biological affinities at different stations throughout the study, principal component analysis (PCA) was performed using XL STAT 2016 software. The purpose of this descriptive factorial statistical method is to represent as much information as possible in a large data table in graphical form ([41]).

Table 3.Identification tests were carried out, and different species were isolated.

+: Positive for 90 to 100% of strains; −: Negative for 90 to 100% of strains; ±: variable.

The pathogenic enterobacterial germs sought were isolated in effluents from a few social housing units in the city of Yaoundé. Their characteristics on Salmonella-Shigella medium were as follows: opaque colonies with black centres, varying in size between 2 and 3 mm, presumed to be Salmonella sp., small pink colonies, presumed to be Shigellasp., and colourless and transparent colonies between 2 and 3 mm, presumed to be Proteussp. The biochemical tests carried out using the API 20E system from the colonies isolated made it possible to have the biochemical profile of the strains studied, as presented in Table 3.

Figure 2. Spatio-temporal variations in cell abundance of (A) Shigelladysenteria, (B) Proteus vulgaris, (C) Salmonella typhi.

The distribution of the abundances of the isolated bacterial species varied from one sampling point to another and from one month to another. During the study period, the highest density ofProteus vulgaris was recorded at the MFD point in April (6.48 Log₁₀ CFU/mL) and reached a value of zero during all sampling months at the MS1 point. The lowest value (0 Log₁₀ CFU/mL) of Shigelladysenteria was recorded at the RPE, Cverte, MS1, and MS2 points in April, May, and March, respectively. The highest value (6.59 Log₁₀ CFU/mL) was recorded at the Cverte point in March. The densities of Salmonella typhi varied between 0 Log₁₀ CFU/mL (at all sampling points) and 6.98 Log₁₀CFU/mL (at point MSII in February) (Figure 2).

Figure 3. Distribution of the percentages of sensitivity/resistance of bacteria to antibiotics during the dry season (AX20 = Amoxicillin20, CIP5 = Ciprofloxacin, GEN10 = Gentamicin10, AM10 = Ampicillin10, TE30 = Tetracycline30, CTX30 = Cefotaxime30, P10 = Penicillin10, AZM15 = Azithromycin15).

The antibiotic sensitivity/resistance of the three species from the sampling points was assessed. During the dry season, Proteus vulgaris and Salmonella typhi species showed 100% resistance to 5 of the 8 antibiotics tested (amoxicillin, ampicillin, cefotaxime, penicillin, and azithromycin for Proteus vulgaris and amoxicillin, ampicillin, ciprofloxacin, penicillin, cefotaxime for Salmonella typhi). Meanwhile, Shigelladysenteria species showed 100% resistance to 2 antibiotics (cefotaxime and penicillin). Figure 3 shows the sensitivity/resistance distributions of each bacterial species to each antibiotic during the dry season. During the rainy season, Proteus vulgaris showed 100% resistance to three antibiotics (ampicillin, tetracycline, and penicillin), while Shigelladysenteria showed 100% resistance to four antibiotics (amoxicillin, tetracycline, penicillin, and ampicillin) and Salmonella typhi to five antibiotics (amoxicillin, tetracycline, penicillin, ampicillin, and ciprofloxacin). In general, the isolated bacterial species were more resistant in the dry season than in the rainy season. Figure 4 shows the sensitivity/resistance distributions of each bacterial species to each antibiotic during the rainy season.

Figure 4. Distribution of percentages of bacterial sensitivity/resistance to antibiotics during the rainy season (AX20 = Amoxicillin20, CIP5 = Ciprofloxacin, GEN10 = Gentamicin10, AM10 = Ampicillin10, TE30 = Tetracycline30, CTX30 = Cefotaxime30, P10 = Penicillin10, AZM15 = Azithromycin15).

Temperature values fluctuated between 20.8˚C and 27.7˚C with an average value of 25.75˚C ± 1.65˚C. The highest value was recorded at point MS1 in April, and the lowest value at point RPE in February (Figure 5(A)). With regard to pH (basic), values fluctuated between 6.65 and 8.63 UC, with an average value of 7.74 UC ± 0.51 UC. The highest value was recorded at the MDG point in March, and the lowest value was obtained at the MFD point in April (Figure 5(A)). Turbidity values fluctuated between 2 and 9600 FAU, with an average value of 1018.25 FAU ± 1882.16 FAU. The highest value was recorded at point MFD in February, and the lowest value at point MDG in May. In addition, water color values ranged from 9 to 18,688 pt.co with an average value of 3100.2 pt.co ± 4138.7 pt.co. The highest value was recorded at the MFD point in February, and the lowest value at the MDG point in May (Figure 5(B)). The MES values fluctuated between 0 and 5928 mg/L with an average value of 579 mg/L ± 1139.6 mg/L. The highest value was recorded at the MFD point in February, and the lowest value at the MDG point in May. In addition, TDS values ranged from 129 to 1550 ppm with an average value of 594.1 ppm ± 321.6 ppm. The highest value was recorded at the BIY point in February, and the lowest value at the MDG point in May (Figure 5(C)).

Figure 5. Spatial variation of the mean values of (A) temperature & pH, (B) turbidity & water color, (C) suspended solids & total dissolved solids, (D) electrical conductivity & salinity, (E) nitrates & nitrites, and (F) ammonium ions & phosphates.

Electrical conductivity values ranged from 253 to 3100 µS/cm, with an average value of 1190.6 µS/cm ± 644.04 µS/cm. The highest value was recorded at the MDG point in May, and the lowest value was obtained at the BIY point in May. In addition, salinity values fluctuated between 0.01 and 0.15 ppm, with an average value of 0.05 ppm ± 0.03 ppm. The highest value was recorded at point BIY in February, and the lowest value was obtained at point MDG in May (Figure 5(D)). Nitrates ion values fluctuated between 0.6 and 216 mg/L, with an average value of 45.05 mg/L ± 53.2 mg/L. The highest value was recorded at point MFD in February, and the lowest value was obtained at point MDG in May. In addition, nitrite ion values ranged from 0.003 to 348 mg/L, with an average value of 40.09 mg/L ± 89.77 mg/L. The highest value was recorded at the BIY point in February, and the lowest value was obtained at the MDG point in May (Figure 5(E)). Ammoniacal nitrogen values fluctuated between 1.62 and 1408 mg/L, with an average value of 60.95 mg/L ± 264.13 mg/L. The highest value was recorded at the RPE point in February, and the lowest value was obtained at the MDG point in April. In addition, phosphate ion values ranged from 146 to 9888 mg/L, with an average value of 4475.07 mg/L ± 3036.55 mg/L. The highest value was recorded at the BIY point in February, and the lowest value was obtained at the MS1 point in May (Figure 5(F)).

Spearman’s correlations between the physicochemical and bacteriological parameters and between the bacteriological parameters revealed several significant positive or negative relationships at the 1% and 5% thresholds. Correlations between abiotic parameters and the densities of isolated bacteria were performed using Spearman’s “r” correlation test. This test revealed highly significant (p < 0.01) and positive correlations between electrical conductivity, TDS, water color, salinity and the species Shigelladysenteria (r = 0.589; 0.583; 0.499 and 0.545) on the one hand, and on the other hand, significant (p < 0.05) and positive correlations were recorded between pH, turbidity, suspended solids, nitrates, nitrites and Shigelladysenteria (r = 0.403; 0.419; 0.378; 0.457 and 0.382). In addition, an increase in nitrites leads to a significant increase (p < 0.05) in the cell densities of Salmonella typhi (r = 0.404). Table 4 summarises the various correlations.

The comparison between the abiotic and microbiological variables during the study period was performed using the Kruskal-Wallis H test. This test shows, on the one hand, that spatially, all species are identical at all sampling points with a significance level of 0.05. On the other hand, in temporal terms, Proteus vulgaris and Shigelladysenteria remained identical throughout the sampling study period, while Salmonella typhi varied significantly during the sampling months at a significance level of 0.05. The Principal Component Analysis (PCA) applied to the various biological and physicochemical variables shows a grouping of parameters into two clusters (Figure 6). In this factorial design, cluster 1 (N1) includes the MFD point, in which Salmonella typhi, Proteus vulgaris, and Shigelladysenteria have strong affinities with ammoniacal nitrogen, turbidity, suspended solids, TDS, water color, electrical conductivity, and nitrate ions. Cluster 2 (C2) contains the BIY and RPE points, where there are strong affinities between salinity, nitrite ions, pH, and phosphate ions. Figure 6 illustrates the affinities between bacterial abundances and abiotic variables.

Table 4. Correlations between bacteriological and physico-chemical variables in wastewater from SIC camps in Yaoundé.

^*: Significant correlationp < 0.05; ^**: Very significant correlation p < 0.01.

Figure 6. PCA grouping affinities between bacterial abundances and abiotic variables.