Incidence and characteristics of antibiotic resistant bacteria in raw and drinking water from Western Cape water production facilities

Abstract

South Africa is one of the driest countries in the world. There are some parts in the country which are classified as water stressed region like Western Cape Province. This province had been reliant on surface and groundwater for drinking water production. A recent drought and in some cases prolonged drought has led to the establishment of a direct portal reuse (DPR) plant in one of the municipalities. It is known that traditional water sources and particularly wastewater is contaminated with agricultural, pharmaceutical personal care products such as antibiotic residues, antibiotic resistant bacteria (ARB) and antibiotic resistant genes (ARGs), besides other contaminants, including pathogens. It is well documented that conventional drinking water processes are not designed to remove antibiotic residues, ARB and ARGs. Their presence in drinking water alters water quality which is an emerging public concern that persists in developing and developed countries. Antibiotic resistance in water sources has up to now often been overlooked. The aim of the present study was to determine the incidence and characteristics of ARB from selected water production facilities in the Western Cape Province. In addition, results obtained were used to determine the effectiveness of two DWPFs in removal of heterotrophic ARB, associated antibiotic resistant genes (ARGs) as well as antibiotic residues. Samples were collected from two water production facilities in the Western Cape (i) direct potable reuse (DPR) facility WC-A linked to a conventional drinking water production facility (DWPF) WC-A and (ii) a separate conventional drinking water production facility WC-F. Various physico-chemical parameters were measured in situ and in the laboratory. Selective medium was used to isolate heterotrophic plate count (HPC) bacteria, faecal coliforms and E. coli. Only HPC bacteria were further characterized for antibiotic resistance and virulence characteristics as well as presence of ARGs. Virulence was based on production of various associated extracellular enzymes. Water samples were subjected to isolation of environmental DNA, which was used to detect ARGs. Antibiotic residues were extracted directly from water environments using the SPE-DEX system and quantified using ultra-high performance liquid chromatography (UHPLC). Redundancy analysis (RDA) was used to plot the correlation of physico-chemical parameters and microbiological agents. Physico-chemical parameters of drinking water from WC-A and WC-F were within the limits set by South African National Standards (SANS 2015:241) of drinking water. Microbiological agents such as faecal coliforms, E. coli were not detected in drinking water. There was a reduction of HPC bacteria from raw to drinking water at WC-A and WC-F. Antibiotic resistance patterns among the isolates indicated that HPC bacteria were in general resistant to trimethoprim, ampicillin, cephalothin and chloramphenicol. An interesting finding was that the DPR plant produced drinking water that had lowest counts of ARB compared to the conventional system. Another interesting observation was that water immediately after treatment had fewer HPC bacteria compared to the distribution networks. An antibiotic resistant index that was used demonstrated that the multiple antibiotic resistant (MAR) index of reclaimed water at WC-A was lower than 0.2, indicating lower number of bacteria resistant to the range of antibiotics. Using Gram-staining and 16S rRNA gene sequencing, Pseudomonas spp. and Bacillus spp. were the dominant HPC bacteria isolated from water samples. Other bacteria identified included Massilia spp., Undibacterium spp., Acidovorax spp., Chromobacterium spp…Most HPC bacteria species were β and α haemolytic. Production of extracellular enzymes was as follows; 82.48% lecithinase, 77.99% DNase, 66.80% proteinase, 56.44% gelatinase and 41.67% lipase. Ciprofloxacin was detected in all water samples. Streptomycin was detected in most water samples, except for wastewater and drinking water from the DPR plant. As expected, most of the antibiotic residues were detected in wastewater samples. The ermB was the gene in most water samples. The ermF, intI1 and ampC were the most prevalent genes in drinking water samples. The blaTEM and tetM were not detected in water samples. The ermB and intI1 genes were only detected in eDNA from wastewater at WC-A and mixed raw water at WC-F. Statistical analysis showed that the growth of microbiological agents has a strong correlation with physico-chemical parameters of the water environments. This study shows that the DPR plant and DWPFs are effectively reducing physico-chemical parameters and microbiological agents. However, conventional treatment processes applied in the DWPFs are not designed to remove ARB, antibiotic residues and ARG in water. The advanced treatment processes that are used to treat wastewater and prepare it for potable use appear to be more effective