Environmental water quality issues have dominated global discourse and studies over the past five decades. Significant parameters of environmental water quality include changes in biological and physical parameters. Some of the biological parameters of significance include occurrence of enteric viruses. Enteric viruses can affect both human and animal’s health by causing diseases such as gastrointestinal and respiratory infections. In this study, the relationship between the occurrence of enteric viruses with reference to adenoviruses and enteroviruses and the physical water quality characteristics was assessed from water samples collected from Lake Victoria (LV) in Kenya. In order to understand the dynamics of season driven enteric viruses’ contamination of the lake waters, we additionally analysed seasonal behavior of the lake’s catchment area in terms of rainfall effects. Physical quality parameters were measured on-site while viral analysis was carried out by molecular methods using the nested polymerase chain reaction (nPCR). From 216 samples that were analysed for viral contamination, enteric viral genomes were discovered in 18 (8.3%) of the samples. Out of half of the samples (108) collected during the rainy season, enteric viral genomes were detected in 9.26% (10) while 8 (7.41%) samples tested positive from the other half of the samples (108) collected during the dry season. There was, however, no significant correlation noted between the physical water quality characteristics and the enteric viruses’ occurrence. Neither wet season nor dry season was significantly associated with the prevalence of the viruses. In Lake Victoria waters, most of the samples had an average of physical water quality parameters that were within the range accepted by the World Health Organization (WHO) for surface waters with exemption of turbidity which was above the recommended 5 NTU as recorded from some sampling sites. Continuous and long-term surveillance of the lake water to accurately monitor the contaminants and possible correlation between chemical, physical, and biological characteristics is recommended. This would be important in continuous understanding of the hydrological characteristics changes of the lake for proper management of its quality with reference to the WHO standards. A multiple varied-sampling approach in different geographical regions during different seasons is recommended to establish the geographical distribution and relatedness to seasonal distribution patterns of the viruses. The data generated from this study will be useful in providing a basis for assessment of seasonally driven fecal pollution load of the lake and enteric virus contamination for proper management of the sanitary situation around the lake.
Environmental water quality monitoring and research have been carried out around the world since 1970s [
Surface waters quality concerns are of great significance over the world because the surface waters are one of the main sources of drinking water after treatment [
As a result of pollution, the quality of the lake water is compromised, thus exposing the huge surrounding population of 1,131,950 [
The aims of the study were therefore as follows: (1) to determine whether the physical characteristics of water in Lake Victoria along Homa Bay town influence the occurrence of the enteric viruses; (2) to determine whether the lake water physical water quality parameters are within the World Health Organization (WHO) acceptable levels for domestic use; (3) to understand the dynamics of season driven enteric viruses’ contamination of surface waters through analysis of seasonal behavior of the surface waters’ catchment area in terms of variation in rainfall activities. Analyses of physical water quality dynamics in relation to the enteric viral contamination in environmental waters may be useful in proper diagnosis of environmental waters quality for proper remedial action. Six physical water quality characteristics, namely, temperature, dissolved oxygen, conductivity, pH, and turbidity, dissolved solids were analysed. For biological characteristics, enteric viruses were analysed as potential pathogens. Adenovirus (HAdV), a double-stranded DNA virus of family Adenoviridae and genus
The study will give an insight into the present status of the quality of the lake water and provide a reference point for future monitoring of the water quality. The future monitoring will inform the design of appropriate management of pollution issues of the lake waters. Data regarding effects of changes in seasons on the contamination of surface waters with these viruses will be important in forming a basis for understanding of the viruses’ epidemiology in the surrounding community and the potential for an outbreak. This intern would be useful in drawing of plans for control and prevention.
The study site was located between longitudes 34.30°E and 34.20°E and latitudes 0.30oS and 0.35oS in Homa Bay Town, Homa Bay County in Western region of Kenya (Figure
Map of Kenya showing Homa Bay County where the study was carried out.
Six points were selected from the study site for samples collection based on observed possible contamination impacts resulting from increased human activities of urbanization, including water transport, industrial and waste water treatment activities. The sampling sites were distributed along a strip in the nearshore of the lake in the town and were designated as sites S1 to S6 (Figure
Satellite image showing the distribution of the sampling points [
Ten litres of water samples were collected using a 10-litre sterilized clean plastic container from the surface of the water at a depth of about 50 cm from each of the sampling points. The sampling was carried out for a six-month period, from October 2011 to April 2012 with the period being dichotomised as rainy or dry season. Sampling months of October 2011, November 2011, and April 2012 were rainy months, while January, February, and March 2012 were all dry months. December 2012 was not included as a sampling month so as to balance the number of sampling trips for the wet/dry season dichotomy. The number of samples per every sampling trip/month from a single site was 6, making a total of 36 samples across all the six sampling sites per trip and 216 for the entire six-month sampling period. The total sampling volume per site was 60 litres, making a total volume of 2160 litres for the entire sampling period. According to a joint report from the Food and Agriculture Organization (FAO) of the United Nations and the Kenya Food Security Steering Group (KFSSG) from 2011 to 2012 when the sampling was carried out, the highest recording of rainfall was in April 2012 with the maxim recording being about 165 mm. On the other hand, the lowest rainfall amount was recorded in January 2012 which was only about 10 mm (Figure
2012 rainfall distributions, Homa Bay, Nyanza. Source: FAO and KFSSG [
The physical quality parameters were measured and recorded
Once the physical characteristics were recorded on-site, the samples were transported on ice to the Enteric Viruses Research Group-Institute of Primate Research Laboratory in Nairobi, Kenya, for enteric viruses’ analysis where they were temporarily stored at a temperature of 4°C until processing [
Being that virus particles are negatively charged, the viruses got adsorbed to the positively charged glass wool filters during the running of the sample through the perspex column [
Nucleic acids were extracted from the 2 ml supernatant using the automated commercially available extratcion kits. The MagNA pure total nucleic acid extraction kit (Roche Diagnostics) and RNeasy minikit (QIAGEN) were used for extraction of the DNA and RNA extratcion respectively, according to the manufacturer’s instructions. The DNA extracts were stored in Tris-EDTA (pH 8.0) at a temperature of −20°C until use following an elution process. The RNA extracts on the other hand were stored by freezing in RNase free water at −80°C. From 2 ml of the extracted nucleic acids solutions, nested PCR was used for amplification according to methods described by [
Two-way analysis of variance (ANOVA) test was used to analyse comparison of the variations between the physical parameters at different seasons and the sampling sites, while Pearson correlation was used to analyse the relationship between the physical parameters and viruses’ presence. SAS version 9.1(SAS Inst. INC., Car., NC) was used to carry out all the analyses in which the
From a total of 216 samples that were collected for viral analysis, the enteric viral genomes were detected in 18 (8.3%) of them. Of these 18 positive samples, 8 (44.4%) were recorded from site S5, while 5 (27.78%) were from site S6 apparently suggesting higher pollution levels in the surrounding areas. Sites S1, S2, and S4 recorded 1 viral genome each, jointly accounting for only 16.67% of the total number of positive samples while site S3 had two (11.11%) positive samples (Table
Summary of the bivariate analysis for the viral contamination by site.
Sampling sites | No. of samples ( | Contamination | |||
---|---|---|---|---|---|
Virus absent | Virus present | ||||
Frequency | (%) | Frequency | (%) | ||
S1 | 36 | 35 | 97.22 | 1 | 2.78 |
S2 | 36 | 35 | 97.22 | 1 | 2.78 |
S3 | 36 | 34 | 94.44 | 2 | 5.56 |
S4 | 36 | 35 | 97.22 | 1 | 2.78 |
S5 | 36 | 28 | 77.78 | 8 | 22.22 |
S6 | 36 | 31 | 86.11 | 5 | 13.89 |
As regards to seasonal distribution of the viruses, the genomes were detected at least once in each of the six sites sampled during the two seasons, although no site was virus-positive in every sampling month (Table
Summary of virus detection from each site during the wet and the dry seasons.
Month | Season | Enteric virus | L1 | L2 | L3 | L4 | L5 | L6 |
---|---|---|---|---|---|---|---|---|
October 2011 | Wet | Enterovirus | 0 | 0 | 0 | 0 | 0 | 0 |
Adenovirus | 0 | 0 | 0 | 0 | 1/11 | 0 | ||
November 2011 | Wet | Enterovirus | 0 | 0 | 0 | 0 | 2/7 | 0 |
Adenovirus | 1/11 | 0 | 0 | 0 | 2/11 | 0 | ||
January 2012 | Dry | Enterovirus | 0 | 0 | 0 | 0 | 1/7 | 0 |
Adenovirus | 0 | 0 | 2/11 | 0 | 0 | 0 | ||
February 2012 | Dry | Enterovirus | 0 | 0 | 0 | 1/7 | 0 | 0 |
Adenovirus | 0 | 0 | 0 | 0 | 0 | 2/11 | ||
March 2012 | Dry | Enterovirus | 0 | 1/7 | 0 | 0 | 0 | 0 |
Adenovirus | 0 | 0 | 0 | 0 | 0 | 1/11 | ||
April 2012 | Wet | Enterovirus | 0 | 0 | 0 | 0 | 0 | 2/7 |
Adenovirus | 0 | 0 | 0 | 0 | 2/11 | 0 |
The numerator represents the number of viruses detected in a given month, while the denominator represents the total number of the specific virus detected during the whole period. Zeroes represent nondetection.
As regards relationship between microbiological parameters in question (human adenoviruses and enteroviruses) and the seasons of samples collection, we observed that there was no significant difference in the number of adenovirus detection during the dry and wet season (
Two-way analysis of variance with post hoc analysis using Tukey’s HSD on physical parameters and viruses present.
Treatment | pH | Temp. (°C) | EC ( | TDS (mg/l) | DO (mg/l) | Turbidity (NTU) | HAdV detected | EV detected |
---|---|---|---|---|---|---|---|---|
Season | ||||||||
Dry | 7.00 ± 0.00a | 25.90 ± 0.03a | 70.73 ± 3.89a | 50.14 ± 3.24a | 8.81 ± 0.12a | 9.98 ± 0.25a | 0.05 ± 0.02a | 0.03 ± 0.02a |
Wet | 7.03 ± 0.011072b | 25.56 ± 0.07a | 0.28 ± 0.05b | 47.56 ± 2.08b | 8.58 ± 0.11b | 19.73 ± 0.85b | 0.056 ± 0.02a | 0.04 ± 0.02a |
Site | ||||||||
S1 | 7.00 ± 0.0b | 25.47 ± 0.13c | 56.97 ± 11.40a | 87.17 ± 6.13a | 9.55 ± 0.20a | 6.06 ± 0.25e | 0.03 ± 0.03a | 0.03 ± 0.03a |
S2 | 7.00 ± 0.00b | 25.78 ± 0.07abc | 60.02 ± 10.14a | 30.58 ± 1.66c | 9.46 ± 0.24a | 16.00 ± 1.05c | 0.00 ± 0.00b | 0.03 ± 0.03a |
S3 | 7.00 ± 0.00b | 25.83 ± 0.06ba | 34.63 ± 5.90b | 50.72 ± 1.58b | 8.46 ± 0.17bc | 11.00 ± 0.40d | 0.06 ± 0.04a | 0.00 ± 0.00a |
S4 | 7.00 ± 0.00b | 25.53 ± 0.13bc | 20.62 ± 3.38c | 25.47 ± 0.78d | 8.63 ± 0.15b | 15.50 ± 1.11c | 0.00 ± 0.00b | 0.00 ± 0.00a |
S5 | 7.06 ± 0.03a | 25.83 ± 0.06ab | 3.91 ± 0.00c | 50.00 ± 3.66b | 8.08 ± 0.11bc | 21.50 ± 1.56a | 0.14 ± 0.06a | 0.08 ± 0.05a |
S6 | 7.03 ± 0.02ab | 25.92 ± 0.05a | 4.25 ± 0.00c | 49.17 ± 3.52b | 7.98 ± 0.10c | 19.08 ± 1.34b | 0.08 ± 0.05a | 0.06 ± 0.04a |
WHO standards | 6.5–8.5 | 25°C ± 2 | 500–5000 | 500–1000 | 8–9 | |||
Season | 0.0090 | <0.0001 | <0.0001 | 0.0028 | 0.0447 | <.0001 | 0.7440 | 0.7010 |
Site | 0.0077 | 0.0002 | <0.0001 | <0.0001 | 0.0447 | <.0001 | 0.0373 | 0.3046 |
The result of the six physical water quality characteristics was analysed between the wet and the dry seasons and among the six sampling sites in relation to the World Health Organization (WHO) standards for environmental waters. Generally, the average values for most of all the physical parameters were found to be in compliance with WHO acceptable levels except in few cases (Table
Analysis of these parameters between the two seasons and among the different sampling sites revealed that there were generally significant differences with the probability values being
We evaluated the relationship of the occurrence of the two microbial parameters (HAdV and EV) in the water samples and the six physical water quality dynamics (temperature, electrical conductivity, total dissolved solids, dissolved oxygen, pH, and turbidity). The occurrence of the viruses did not show any significant relationship with any of the physical parameters (Table
Correlation of physical parameters and viruses present.
pH | Temp. (°C) | EC | TDS | DO | Turbidity | HAdV detection | EV detection | |
---|---|---|---|---|---|---|---|---|
Temp. (°C) | ||||||||
0.636 | ||||||||
EC | −0.131 | 0.267 | ||||||
0.055 | 0.0001 | |||||||
TDS | 0.125 | 0.147 | 0.331 | |||||
0.067 | 0.031 | 0.0001 | ||||||
DO | −0.108 | 0.012 | 0.245 | 0.214 | ||||
0.115 | 0.850 | 0.0001 | 0.002 | |||||
Turbidity | 0.247 | 0.011 | −0.542 | −0.132 | −0.157 | |||
0.0001 | 0.874 | 0.0001 | 0.053 | 0.021 | ||||
HAdV detection | 0.089 | 0.114 | −0.062 | 0.045 | −0.083 | 0.108 | ||
0.193 | 0.096 | 0.366 | 0.508 | 0.222 | 0.113 | |||
EV detection | 0.128 | 0.090 | −0.027 | 0.119 | −0.068 | 0.087 | −0.042 | |
0.060 | 0.188 | 0.693 | 0.082 | 0.317 | 0.202 | 0.535 |
Enteric viruses such as HAdV and EVs have been discovered in many of surface waters around the world [
Studies have found that virus can persist suspended in environmental waters for several days at both very low and high temperatures [
Studies have shown that there are chances of increase in ease of transportation of viruses with increase in water flow and this may increase chances of viral detection [
Higher detection of viruses from some of the sites such as sites S5 and S6 could be an indication to high level of contamination or discharge of pollutants around the sites. This could be attributed to intense human activities from pressure of urbanization leading to sewage and industrial and agricultural discharges to the lake. Lower detection from sites S1, S2, and S3 could as well be associated with factors that may have hindered PCR detection of the viruses. Certain conditions and chemicals have been reported to limit PCR amplification [
Most of the samples from all the sites recorded physical water quality characteristics that were within the recommended range of values according to the World Health Organization except for turbidity which was higher in most of the samples. The higher levels of turbidity however could be attributed to agricultural activities from the surrounding community especially during the rainy seasons as a result of runoff carrying silt [
Relationship between physical water quality parameters and viral contamination in surface waters has been reported in previous studies [
Most of the samples had an average of the physical water quality parameters that were within the range accepted by the WHO. However, the detection of turbidity above the levels recommended by the WHO and DO concentrations in some cases at levels below the recommended threshold is a justification that the level of pollution of the lake water is high and therefore the lake environment should be subjected to continuous monitoring for proper management of pollution challenges. Adenoviruses and enteroviruses occurrence in LV waters are relatively constant throughout the year without significant variations in their profile with changes in season. Changes in season dynamics may therefore be considered to be an unreliable factor for prediction of viral contamination peaks in the lake.
In addition to these findings, the physical and microbiological parameters were found not to be significantly correlated; the occurrence of the viruses was not affected by changes in the physical parameters. We recommend a more detailed continuous long-term surveillance of the lake waters to accurately monitor the contaminants and possible correlation between chemical parameters and other physical characteristics not covered in this study such as light penetration and the virus occurrence. There was no significant influence of season in detection of the viruses as per the present study; however, a multisampling approach in different regions and during different seasons is recommended to establish the viruses’ geographical distribution and relatedness to seasonal distribution patterns.
The metadata used to support the findings of this study have been deposited in the Kenyatta University Institutional repository at
The authors declare that they have no conflicts of interest.
WMO designed the study, carried out fieldwork, performed the experiments and the statistical data analysis, and wrote the first draft of the manuscript. MJ and OO managed the literature searches and analyses of the study. All the authors have read and approved the final manuscript.
The authors acknowledge a postgraduate bursary from Karachuonyo Constituency Development Fund for research. The authors acknowledge Mr. Nicholas Kiulia of the Institute of Primate Research in Kenya for many fruitful discussion and guidance during the whole period of the project. The authors thank the entire staff of the Institute of Primate Research and the National Museum of Kenya for providing laboratory space. The authors also like to thank Maxwell Adek for his assistance in field sampling.