This District Municipality of Vhembe (Figure 1) is found in the province of Limpopo, further north of South Africa, and occupies a landmass of approximately twenty thousand square kilometres, according to Vhembe (2013). This area is blessed with fertile soil, and agricultural activities are taking place across the region. According to Kachienga et al. (2024), this area has both seasonal and permanent rivers. There are a few existing communal boreholes that are still found in some villages that lack access to municipal water supply. Permanent rivers (n = 11) and communal boreholes (n = 12) that were sampled for this study are indicated by the red dots in Figure 1.

Water samples were collected between the months of April and July 2024 in 2-L sterile plastic bottles using aseptic techniques from rivers at specific sites as described by Kachienga et al. (2024) and communal boreholes. The collection was done once off, and the samples were in duplicates from both river (down and downstream) and communal boreholes. The samples were transported to the laboratory at the University of Venda in a cooler box containing ice and analysed upon arrival (Rice et al., 2012).

Various parameters, such as total dissolved solids (TDS), temperature, electrical conductivity (EC), and pH, were determined on site using a pH probe (Hach Intellical™ PHC101). The dissolved oxygen (DO) was analysed using a DO probe (Hach LDO^® Model 2). In addition, a Lovibond TB 211 turbidimeter (Lovibond. IR, Germany) was used to analyse turbidity on site.

The filtration of the samples was done according to the protocol of the National Center for Infectious Diseases, 1994; Momtaz et al. (2013), where a 47 mm cellulose acetate filter with a 0.45 μm pore size was used to filter the samples. The enrichment of filtrate that was trapped in the filter paper after sample filtration was done in 100 mL of sterile alkaline peptone water (NutriSelect^® Plus, Sigma Aldrich, SA); thereafter, the incubation was done at 37°C for one full day.

The enriched samples (10 μL) were transferred into the sterile thiosulphate citrate bile salt (TCBS) (Mass Group Ltd. Reinfeld, Germany) agar plates and then incubated for 24 h at 37°C. The analysis was done in triplicate for each sample. The presumptive positive V. cholerae and their colonies forming units were determined according to Kachienga et al. (2024). The preservation of the presumptive Vibrio colonies was done according to Huq et al. (2012), and we further tested using biochemical tests for the confirmation of presumptive positive V. cholerae. The presumptive Vibrio colonies obtained were further used to extract the genomic DNA, which was used for molecular analysis for confirmation of toxigenic and serogroups of choleragenic V. cholerae.

The presumptive colonies were tested with the API 20E system, and the interpretation of the isolates was based on the API 20E analytical profile index (Version 5, Biomerieux Industries, France). This technique was used to identify the presumptive positive V. cholerae obtained from other closely related Vibrionaceae species that might be present.

DNA extraction from samples was carried out using the ZymoBIOMICSTM DNA Miniprep Kit (Zymo Research, CA, USA) per the protocol of the manufacturer from the purified preserved presumptive Vibrio colonies. The NanoDropTM 2000 spectrophotometer (Thermo Scientific, Johannesburg, South Africa) was used to quantify the extracted DNA. The PCR reactions were performed in a total volume of 25 μL, as previously reported (Kachienga et al., 2024). The PCR technique was necessary to confirm the presence of V. cholerae, toxigenic V. cholerae, and serogroups of choleragenic V. cholerae in the samples using specific primers reported in Table 1. Amplification reactions were done in the following order: heat denaturation at 94°C for 2 min, followed by 35 cycles of heat denaturation at 94°C for 1 min, annealing at 58°C for 1 min, and DNA extension at 72°C for 1 min (Momtaz et al., 2013) using a qPCR Eppendorf Master Cycler 5330 (Eppendorf-Nethel-Hinz GmbH, Hamburg, Germany). The specific primers used included OmpW, ctxA, and O1rfb/O139rfb for detecting V. cholerae, toxigenic V. cholerae, and serogroups of choleragenic V. cholerae (Nandi et al., 2000; Huang et al., 2009; Mehrabadi et al., 2012) (Table 1).

The correlation of physiochemical parameters based on different sampling sites (upstream and downstream rivers, communal boreholes) through principal component analysis (PCoA) was done using ClustVis (Metsalu and Vilo, 2015). The clustring together of these parameters visa vis their sampling sites based on the geographical areas and the surrounding activities.

The activities observed around these water sampling points were agricultural (subsistence and commercial farming), brickmaking, laundry, car washing, and open defecation by community members, amongst others. Various activities taking place around the water sources and occasional flooding trigger the presence of pathogenic microbes through runoff into the water sources. Yeung and Thorsen (2016) further reported that most of the urinal infections are also spread during recreational activities in and around rivers and communal boreholes.

Polluted water is associated with a heavy burden on the population’s overall health, and this has negatively impacted the life expectancy in developing countries such as South Africa (WHO, 2002). Cases of V. cholera have become a burden in poorer communities, especially in sub-Saharan countries where there is inadequate safe, clean water and sanitation infrastructure (Khouadja et al., 2014). According to Huq et al. (2012), assessment of vital parameters (temperature, turbidity, TDS, pH, DO, and EC) is necessary to establish the survival and adaptation of pathogenic and serogroups O1 and 139 and non-O1/O139. The results of the above parameters are captured in Table 2.

The pH counts of the rivers ranged from 7.03 to 8.21 for rivers and 6.88 to 8.21 for communal boreholes. They were both within the acceptable South African water quality guidelines of 6.5–8.5 for aquatic environments (rivers) and 6.0–9.0 for domestic usage (communal bore holes) (Department of Water Affairs and Forestry, 1996a). A pH between 7.5 and 9.0 normally influences the survival or adaptability of microorganisms; moreover, a pH > 8.5 is optimal for their survival in water (Grandjean et al., 2005). V. cholerae, especially serogroup O1, is extremely sensitive to acidic environments, and normachlorohydric people are not easily susceptible to cholera. This type of vibrio is also sensitive to desiccation and heat (ICMSF, 1996).

The average electrical conductivities (EC) for rivers and communal boreholes ranged between 18.78 and 154 μS/cm and 23.4 and 295 μS/cm (Table 2), which were above the acceptable South African water quality guidelines of 0–70 μS/cm for rivers and domestic use (Department of Water Affairs and Forestry, 1996a). The higher conductivity value recorded is directly proportional to an increase in ions and organic compounds that can conduct electric current (Holmes, 1996; Ferreira, 2011). Atekwana et al. (2004) further reported that the high levels of conductivity and TDS in water and the environment are due to the biodegradation of organic compounds by natural microorganisms. The presence of chitinase that is produced by V. cholera also assists them in adapting and surviving in extreme conditions such as saline and acidic environments, as reported by Huq et al. (2012).

The overall range for the TDS for the rivers in this study ranged between 10 and 90 mg/L, while for communal boreholes, it ranged between 70 and 320 mg/L. These data revealed that both rivers and communal boreholes were still within the acceptable South African water quality guidelines of 0–450 mg/L (Department of Water Affairs and Forestry, 1996a; Department of Water Affairs and Forestry, 1996b).

The average temperature obtained ranged between 18.05°C and 30.60°C for both rivers and communal boreholes. According to Franco et al. (1997), temperature is vital in triggering the formation of environmental biofilms and sediments that are supporting the survival of V. cholera, which will increase their counts and human illness. Huq et al. (2012) also state that a wider range of temperatures is critical for the heterogeneous nature of V. cholera. Temperature and other parameters, such as salinity, are the main drivers of the adaptability and survival of V. cholerae in any given environment, including contaminated water (Takahashi et al., 2023).

Most of the rivers ranged between 1.01 and 14.85 NTU, while the turbidity of communal boreholes ranged between 0.71 and 13.23 NTU. Despite all, most of the rivers were found to be within the acceptable set standards limit of <25 NTU for river water containing sediments and untreated aquatic water (Department of Water Affairs and Forestry, 1996a). However, only some communal boreholes (Val B1, Kwe B1, B3, and B4) were reported to be within the acceptable South African water quality guidelines of 0–1 NTU for domestic use (Department of Water Affairs and Forestry, 1996b). Turbidity is influenced by the formation of biofilms, sediments, and eutrophication by the presence of nitrates and phosphates because of the agricultural, bathing, laundry, and brick-making activities inside and along the water sources (Fong et al., 2010). Ntema et al. (2014) also stated that the presence of debris and biotic surfaces in the rivers also facilitates V. cholera attachment and mobility.

Based on the availability of DO, the study revealed that most of the rivers were above the acceptable South African water quality guidelines of 6.5–8.0 ppm, except for Madanzhe River, whose DO counts were below the acceptable water quality guidelines. This is driven by the high-water runoff, higher flow rates of waters within these rivers, and pollution of rivers.

The principal component analysis (PCoA) correlation of different physicochemical parameters also revealed that most of the rivers and boreholes are clustered based on the geographical areas and the surrounding activities. The rivers such as Mvudi, Livuvhu, Tshinane, and Nwedi have clustered together based on PCoA in terms of high levels of variance in physicochemical parameters because farming and laundry activities around these water catchments, while Madandze, Mutshundudi, Nzhelele, and Mukhase rivers were also clustering together due to anthropogenic activities such as brickmaking and farming around them (Figure 2). These activities play a role in the physicochemical parameters obtained from the samples from these river sources.

Figure 3 reported that the Mak B1, B2, and Val B1 boreholes were clustered due to high levels of variance in physicochemical parameters. This was also due to the similarity of geographical location and poor sanitation around these boreholes that influenced their physicochemical parameters. The Kwe B3, Kwe B4, Thi B1, and Tsh B1 boreholes highly clustered together, while the Kwe B2, B5, and B6 boreholes least clustered together based on the variance of parameters. Most of these boreholes that were highly clustered were also influenced by agricultural activities and salty water that was associated with them. The boreholes that were least clustered, such as Kwe B2, B5, and B6, were difficult to identify because they were from similar geographical settings yet they least clustered.

The API test is one of the biochemical tests that is used to identify various families of microorganisms, such as Enterobacteriaceae and V. cholerae. Most of these rivers reported presumptive Vibrio flavialis, Erwina spp., Raoutella planticolla or terrigena, Aeromonas spp., Klebsiella spp., and Serrati spp., except for Madanzhe, Sambandou, and Nzhelele, which also had presumptive V. cholerae after being identified on the downstream site by the API test (Biomerieux Industries, South Africa) (Table 3). There were no microorganisms identified in the Makhase River. In communal boreholes, most of them reported presumptive V. flavialis and/or Aeromonas spp., Erwina spp., and Raoutella planticolla or terrigena, and communal boreholes such as Mak B1, B2 and Val B1, B2 reported none of the microorganisms (Table 3). These Erwina spp. bacteria are known to be well adapted to growth on vegetables because they are capable of fermenting many of the sugars and alcohols that exist in certain vegetables, unlike other common bacteria (Warriner et al., 2003; Gustaw et al., 2021). They might find their way into rivers and boreholes because of runoff from the surrounding farms or backyards from floods or rainfall. Raoutella planticolla or terrigena, Aeromonas spp., Klebsiella spp., and Serrati spp. are some of the microorganisms that are Gram-positive and sugar fermenters that lead to yellow colonies (Abbott, 2011; Yeung and Thorsen, 2016). They also cause urinary tract infections.

Most of the rivers detected positive for V. cholerae using the OmpW primer, except for the Mukhase river, the downstream Livuvhu and Nwedi rivers, and the upstream Nzhelele river point (Table 4). Only Mak B1, B2, Kwe B3, 4 and 6 boreholes tested positive for V. cholerae, using the OmpW primer (Table 4), which specifically targets the outer membrane of V. cholerae, and according to Fu et al. (2018), this gene acts as a passage of iron and other organic molecules that are pumped across and regulate survival in extreme environments. The detection of V. cholerae was also directly connected to the higher physicochemical parameters such as EC, TDS, and turbidity that were reported in both river and communal boreholes. The higher values obtained trigger the survival and distribution of V. cholerae in these water sources.

From the rivers that tested positive for V. cholerae, the pathogenicity of these Vibrio isolates was assessed using the ctxA primer and Mutshundudi, Tshinane rivers, the upstream of Dzindi, Madanzhe, and Nwedi samples, and the downstream of Sambandou rivers reported positive for toxigenic V. cholerae (Table 4). Only Mak B1, B2, and Kwe 3 communal boreholes were found to have toxigenic V. cholerae (Table 4). Bina et al. (2003) reported that toxigenic V. cholerae produces a toxin-regulated pilus that facilitates colonisation of the intestine walls and is the main regulator of the ctxA gene, and this gene was targeted by the ctxA primer. According to Liang et al. (2007), colonisation results in the activation of the G protein, which stimulates the adenylate cyclase. This process leads to the loss of electrolytes from the cell membrane, resulting in vomiting and watery, loose stools into the large intestines (World Health Organization, 2017). It is imperative to understand that any mutation or mismatch in the ctxA gene sequence leads to either a positive ctxA or a non-toxin-producing strain (Hoshino et al., 1998; Takahashi et al., 2023). The presence of toxigenic V. cholerae in these water sources was driven by the conditions and activities around them, as shown in S1-3, where some of the water sources were closer to sewer lines, toilets, and backyard farming. For the contamination or pollution of communal boreholes, Barcelona et al. (1988) stipulated different routes such as infiltration, direct migration, inter-aquifer exchange, and recharge from surface water that are the major drivers. In the clinical context, pathogenic V. cholerae differs from the non-pathogenic ones based on toxin-co-regulated pilus (TCP) and cholera toxin (CTX), which allow colonisation of the gut encoded by the phage CTXφ, which uses the TCP to attach to the bacterium (Waldor and Mekalanos, 1996; Ja’afar et al., 2021). Toxigenic V. cholerae also utilises the commensal association, leading to the formation of biofilm with copepods on their chitinous surfaces, which further leads to the formation of algal (de Magny et al., 2011; Rita and Colwell, 2009).

Once the pathogenicity had been analysed using PCR, the serogroups of the detected toxigenic were determined using the O1 rfb and O139 rfb primers. Serogroup O1 was detected on the Mutshundudi and Tshinane rivers, while serogroup O139 was detected on the upstream of the Dzindi, Madanzhe, Mutshundudi, and Tshinane rivers (Table 4). There was detection of both the O1 and O139 serogroups in the communal boreholes Mak B1 and Kwe B3, while serogroup O139 was only detected in the communal borehole Mak B2 (Table 4). In a study done by Phetla (2021), there was a higher percentage of serogroup O1 from sewage sludge and the riverbed from the Limpopo region, and therefore, there was a potential for O1 circulation in water sources leading to sporadic cholera outbreaks. The study of Phetla (2021) also reported that the O139 V. cholerae serotype was not found, and it was similar to a study done by Ansaruzzaman et al. (2007), who reported that this serotype was not endemic to the African continent but more endemic to India, Bangladesh, and other parts of the Asian continent. However, in this study, serogroup O139 was identified in both rivers and communal boreholes. This might be due to cross-border migration from cholera-prone countries/areas into South Africa through seeking medical treatment, tourism, traders, relatives, and refugees from neighbouring cholera-endemic countries or communities with ongoing epidemics like Haiti, Malawi, or Mozambique and Asian countries (Chin et al., 2011; Zarocostas, 2017). The serogroups detected from rivers and communal boreholes in the study were also due to the discharge of raw sewage, especially on the upstream side of the Madanzhe River, and the discharge of partially or untreated wastewater effluents into the rivers. The communal boreholes were also closer to the septic pit, leaking sewer line, and latrine, as depicted in S1-3.

Keddy et al. (2013) and Alaoui et al. (2021) stated that non-O1 or non-O139 strains of V. cholerae have been identified in birds, which might act as carriers/hosts into the environment. Therefore, continuous surveillance of rivers and communal boreholes, an integrated approach, and robust policies are needed to monitor these cases in a timely and reliable manner to prevent further discharge of raw sewage into rivers and closer to communal boreholes.

S1-3 depict images of rivers (upstream and downstream) and communal boreholes where serogroups (O1 and O139) were discovered. The majority of the rivers were contaminated, and the communal boreholes were closer to the pit latrine while the sewer line also discharged dirty effluent directly into the river catchment.

The study involved analysing water samples from rivers and communal boreholes, which could also be reservoirs for other organisms such as fish, snails, and others that are involved in commensal association with V. cholerae, including feeding, but were not considered in this study (You et al., 2019; Cho et al., 2021). Further studies using appropriate techniques targeting ecosystems that were not covered by this study should be conducted to provide more informed and reliable information on the presence of choleragenic V. cholerae in both surface water sources and communal boreholes. This study was conducted for a period of 4 months only, and a longer study might be needed to establish the prevalence and distribution of V. cholerae O1 or O139 and the potential health risk to vulnerable communities in Vhembe district, which is predominantly in rural areas.