Skip to main content
NCBI home page
GeoHealth logoLink to GeoHealth
. 2017 Aug 31;1(6):258–269. doi: 10.1002/2017GH000081

Quantification of multiple waterborne pathogens in drinking water, drainage channels, and surface water in Kampala, Uganda, during seasonal variation

Nora J Sadik 1, Sital Uprety 1, Amina Nalweyiso 2, Nicholas Kiggundu 2, Noble E Banadda 2, Joanna L Shisler 3, Thanh H Nguyen 1,✉
PMCID: PMC7007170  PMID: 32158991

Abstract

Longitudinal water quality monitoring is important for understanding seasonal variations in water quality, waterborne disease transmission, and future implications for climate change and public health. In this study, microfluidic quantitative polymerase chain reaction (MFQPCR) was used to quantify genes from pathogens commonly associated with human intestinal infections in water collected from protected springs, a public tap, drainage channels, and surface water in Kampala, Uganda, from November 2014 to May 2015. The differences in relative abundance of genes during the wet and dry seasons were also assessed. All water sources tested contained multiple genes from pathogenic microorganisms, with drainage channels and surface waters containing a higher abundance of genes as compared to protected spring and the public tap water. Genes detected represented the presence of enterohemorrhagic Escherichia coli, Shigella spp., Salmonella spp., Vibrio cholerae, and enterovirus. There was an increased presence of pathogenic genes in drainage channels during the wet season when compared to the dry season. In contrast, surface water and drinking water sources contained little seasonal variation in the quantity of microbes assayed. These results suggest that individual water source types respond uniquely to seasonal variability and that human interaction with contaminated drainage waters, rather than direct ingestion of contaminated water, may be a more important contributor to waterborne disease transmission. Furthermore, future work in monitoring seasonal variations in water quality should focus on understanding the baseline influences of any one particular water source given their unique complexities.

Keywords: pathogen, enteric, water, seasonality

Key Points

  • Pathogenic genes associated with human infections were quantified in water samples collected over seven months in Kampala, Uganda

  • An increased presence of pathogenic bacterial and viral genes in drainage channels in wet season was observed

  • Human interaction with contaminated drainage water could contribute to waterborne disease transmission

1. Introduction

Globally in 2010, diarrheal diseases were responsible for 810,000 deaths among children under the age of five, with about 90% of these deaths occurring in Sub‐Saharan Africa and South Asia [Johansson et al., 2012]. It is estimated that approximately 88% of all diarrhea attributable diseases are preventable through safe water, sanitation, and hygiene [Black et al., 2003]. While the Millennium Development Goal aiming to have 88% of the global population with access to improved drinking water was met in 2010 [World Health Organization, 2015], climate change introduces unique challenges in creating safe and resilient water sources necessary in maintaining and surpassing development goals. To provide robust climate change recommendations for public health, water, and sanitation sectors, the epidemiological relationship between local climate and waterborne disease must be well understood [Akanda et al., 2014; Armstrong et al., 2012; McIver et al., 2016], especially in developing countries where climate change is argued to have the greatest impact on water, food security, and public health [Bush et al., 2011; Field et al., 2014]. Previous research has emphasized the importance of empirical field‐based measurements for designing effective climate change interventions [Mellor et al., 2016; Olmstead, 2014]. We believe that field‐based water quality measurements would provide the greatest impact if assessed over time to capture seasonal variability in water quality of different source types.

Given that climate change is likely to exacerbate health risks in urban poor populations due to inadequate economic capacity, infrastructure resilience, and health services [Field et al., 2014; Lwasa, 2010], studying microbial reservoirs that harbor pathogenic microorganisms, such as those in community water sources, is imperative. The rapid urbanization of Kampala, Uganda, is exacerbated by poverty and inadequate physical planning. This has given rise to the expansion of informal settlements which are often subjected to overcrowding, poor water, and sanitation conditions, and limited access to basic health, energy, and security services [Dimanin, 2012; Dobson et al., 2015; Vermeiren et al., 2012]. In urban Uganda, unreliable and unsafe solid waste management and drinking water sources contribute to the transmission of waterborne diseases such as cholera, dysentery, cryptosporidiosis, and rotavirus [Bwire et al., 2013; Howard et al., 2002; J. S. Nakawesi et al., 2010b; Rugadya et al., 2008].

Long‐term monitoring of microbial water quality that is both efficient and accurate will have a greater global impact on improving water security in the context of climate change. Numerous publications have shown that fecal indicator bacteria do not consistently or precisely represent the presence of bacteria, viruses, and protozoa [Bonadonna et al., 2002; Harwood et al., 2005; Lund, 1996] particularly in tropical environments [Helena et al., 1999; Rochelle‐Newall et al., 2015]. High‐sensitivity and high‐throughput methods to monitor water quality over the long term are necessary to observe seasonal variation of water quality. Microfluidic quantitative polymerase chain reaction (MFQPCR) is a high‐throughput chip‐based PCR assay that is a promising monitoring tool for global biological water quality. MFQPCR utilizes microfluidics technology to increase sensitivity and specificity and reduce cost, reagent and sample consumption, and time compared to conventional singleplex and multiplex qPCR [Zhang et al., 2006]. Previous environmental water studies have shown that MFQPCR maintains and may surpass overall sensitivity compared to conventional qPCR [Byappanahalli et al., 2015; Ishii et al., 2014b]. Applying this method to water quality analysis results in the simultaneous detection of multiple waterborne pathogens across multiple water samples [Ishii et al., 2014a, 2014b, 2013].

The objectives of this study were to identify and quantify pathogenic genes commonly associated with human intestinal infections in drinking water, drainage channels, and surface water over seven months during seasonal variation in Kampala, Uganda. Water samples were collected from drinking water, drainage channels, and surface water in Kampala, Uganda, from November 2014 to May 2015. Target genes commonly representative of pathogenic organisms were quantified over time at multiple water sources. To the best of our knowledge, this is the first study to longitudinally measure waterborne pathogen presence in multiple water source types in a developing tropical region.

2. Methods

2.1. Locations and Sampling

The city of Kampala is composed of five administrative divisions: Central, Kawempe, Nakawa, Makindye, and Rubaga. Water samples from eight protected springs, one treated public tap, two drainage channels, and one lake were collected from November 2014 to May 2015. These water source types were selected to monitor the variation of water quality parameters as well as pathogenic gene presence among different water sources over time. Water sources in informal settlements and densely populated neighborhoods were selected over water sources in neighborhoods with lower population density. Sampling locations and water source types are indicated in Figure 1.

Figure 1.

Figure 1

Open in a new tab

Sampling map Google Maps indicates the location of all sampling sites and water source types. (a) Bwaise contained two protected springs (B1 and B2) and one drainage channel (B3). (b) Kalerwe contained one drainage channel (Ka1) and one public tap that is treated drinking water supplied by the National Water and Sewerage Corporation (Ka2). (c) Luzira contained two protected springs (L1, L2). (d) Mengo contained two protected springs (M1 and M2). (e) Namowongo contained two protected springs (N1 and N2). (f) Ggaba is on Lake Victoria with two sampling locations (G1 and G2).

The 2 L water samples were collected from protected springs and the public tap using reusable sterilized bags (Boli, Zhejiang, China). Bags were sterilized 24 h prior to sample collection according to Environmental Protection Agency (EPA) guidelines [Fout et al., 2001]. Drainage channel and Lake Victoria water samples were collected in 0.5 L aliquots in sterile Whirlpak (Nasco, Fort Atkinson, WI) bags. Upon collection, all samples were stored at 4°C and were processed within 24 h. The dates and seasons of collected samples are summarized in Table 1.

Table 1.

Summary of Collected Water Samplesa

Sample Set Dates Collected Season Number of Samples Collected
SS1 10 Nov to 20 Nov (2014) Wet 16
SS2 24 Nov to 5 Dec (2014) Wet 16
SS3 4 Dec to 18 Dec (2014) Wet 16
SS4 6 Jan to 7 Jan (2015) Dry 15
SS5 19 Jan to 23 Jan (2015) Dry 16
SS6 3 Feb to 10 Feb (2015) Dry 16
SS7 17 Feb to 2 Mar (2015) Dry 16
SS8 3 Mar to 17 Apr (2015) Wet 16
SS9 22 Apr to 11 May (2015) Wet 16
SS10 20 May to 27 May (2015) Wet 16
Open in a new tab
a

Season was determined based on supporting literature [Funk et al., 2012; Kigobe et al., 2011]. The Namuwongo channel water sample was not collected during SS4 due to low flows.

Within 24 h of collection, water samples were treated with sterile 2.5 M MgCl2‐6H20 (Sigma‐Aldrich, St. Louis, MO) for 30 min with periodic mixing to coagulate particles and microorganisms [Mattioli et al., 2013; Victoria et al., 2009]. Thereafter, water samples were sequentially vacuum filtered through a 1.6 μm pore glass fiber filter (Millipore, Ballerica, MA) followed by a 0.45 μm pore nitrocellulose filter (GVS Maine, Sanford, ME) placed in a 47 mm filtration funnel (Pall Corporation, New York, NY) [Ikner et al., 2012]. The 1.6 μm pore filter was used as a prefilter since the turbidity of the collected water samples, especially from the channels, was very high. Without the 1.6 μm pore filter to remove larger particles in the water sample, the 0.45 μm filter alone typically blocked filtration of the water sample. While this was an issue only for water samples collected from channels, the same methodology was applied for all water samples for consistency. The filtration housing and flasks to catch filtrate were sterilized prior to each sample filtration according to EPA guidelines [Fout et al., 2001]. Filters were treated with 500 μL RNAlater (Qiagen, Helden, Germany) to preserve microbial genomes and were stored in sterile Whirpak (Nasco) bags at −20°C until transport to the University of Illinois at Urbana‐Champaign (UIUC). At UIUC, filters were stored at −80°C until DNA/RNA extraction.

2.2. Genome Extraction and cDNA Synthesis

The 1.6 μm pore filters from protected springs in the same community (B1, B2; L1, L2; and N1, N2; M1, M2) and Lake Victoria water samples (G1 and G2) were combined during extraction to expedite sample processing and pathogen enumeration. The 0.45 μm pore nitrocellulose filter micrometer filters were similarly combined for protected springs and Lake Victoria water samples. The 1.6 μm pore and 0.45 μm pore filters from drainage channel water samples and public tap water samples (B3, Ka1, and Ka2) were extracted individually. Table 2 shows the manner in which samples were combined and the distances between the water sources of combined samples. Two extraction methodologies were applied due to the difference in the two filter materials. The FastDNA Spin kit for Soil (MP Biomedicals, Santa Ana, CA) was used to extract DNA from the 1.6 μm pore glass fiber filters according to the manufacturer's instructions with minor modifications. The PowerWater RNA Isolation Kit (Mo Bio Laboratories, Carlsbad, CA) was used to extract RNA and DNA from 0.45 μm pore nitrocellulose filter according to the manufacturer's instructions with minor modifications. All DNA/RNA extracts were stored at −80°C. DNA/RNA extracts eluted from the 0.45 μm pore nitrocellulose filter underwent reverse transcription (RT) prior to enumeration by MFQPCR. cDNA synthesis was performed in a MyCycler Thermal Cycler (Bio‐Rad, Hercules, CA) using the iScript cDNA Synthesis Kit (Bio‐Rad). The reaction mixture (20 μL) contained 5X iScript Reaction Mix, 1 μL iScript Reverse Transcriptase, and 2 μL template DNA/RNA and was performed under the following thermal cycle conditions: 25°C for 5 min, 42°C for 30 min, and 85°C for 5 min. cDNA/DNA samples were stored at −20°C.

Table 2.

Sampling Water Sourcesa

Water Source Type Community Name Sample Name (Pre) Sample Name (Post) Distance Between Two Water Sources (km)
Protected spring Bwaise B1 B12 0.21
Bwaise B2
Drainage channel Bwaise B3 Single Site
Drainage channel Kalerwe Ka1 Single Site
Public tap Kalerwe Ka2 Single Site
Protected spring Luzira L1 L12 0.014
Luzira L2
Lake Victoria Ggaba G1 G12 0.068
Ggaba G2
Protected spring Namuwongo N1 N12 0.14
Namuwongo N2
Protected spring Mengo M1 M12 0.052
Mengo M2
Open in a new tab
a

Membrane filters from protected spring and Lake Victoria water samples were combined during genome extraction. Membrane filters from drainage channels and the public tap were genome extracted individually.

2.3. Conventional qPCR and MFQPCR Assays

The objective of the MFQPCR assay was to detect 18 genes corresponding to 11 pathogens including 5 viruses and 6 bacteria and 1 nonpathogenic bacterium (used as an internal control). A spike in cholera infection was observed in Uganda before the rainy season [Bwire et al., 2013]. Epidemiological studies for other enteric infection were not available. We selected the pathogens based on previous studies, reporting the prevalence of pathogens in stool samples of Kampala residents. Specifically, rotavirus was present in up to 8% of stools from healthy children and adults [J. Nakawesi et al., 2010a]; Giardia intestinalis were present in 20% of healthy children [Ankarklev et al., 2012]; Cryptosporidium parvum were present in up to 25% of children with diarrhea [Tumwine et al., 2003]; Campylobacter spp. were present in up to 9% of diarrhea cases [Mshana et al., 2009]; Escherichia coli were 14% diarrheal cases, present in up to 88% water samples [Katukiza et al., 2014], and Salmonella sp. were present in up to 88% present in water samples [Katukiza et al., 2014]. In addition to these six pathogens, we added the following five pathogens: adenovirus, norovirus, enterovirus, Hepatitis E, and Shigella, because they are common waterborne pathogens and a common cause of outbreaks in Africa [Adefisoye et al., 2016; Holt et al., 2013; Kim et al., 2014; Mans et al., 2016; Walker et al., 2010].

Assay designs for all target genes were adapted from previous studies [Costafreda et al., 2006; Gonzalez‐Escalona et al., 2009; Ishii et al., 2014a, 2013; Jothikumar et al., 2005; Kitajima et al., 2009, 2010; Liu et al., 2012; Miura et al., 2013; Monpoeho et al., 2000; Tago et al., 2011] to detect adenovirus types 40 and 41, enterovirus, human norovirus types GI and GII, Group A, Hepatitis A virus (all genotypes), Hepatitis E virus (all genotypes), E. coli, enterohemorrhagic E. coli (EHEC), Shigella spp., Salmonella spp., Campylobacter jejuni, Vibrio cholerae, and Pseudogulbenkiania sp. NH8B. A list of target genes and primer and probe sequences are summarized in Table S1 in the supporting information.

Forward and reverse primers for all assays were obtained as Custom DNA Oligos (Integrated DNA Technologies, Coralville, IA). Three fluorogenic probe types were used in qPCR and MFQPCR assays. Double‐quenched hydrolysis probes were labeled with 6‐fluorescein (6‐FAM) at the 5′ end, Iowa Black FQ quencher at the 3′ end, and an internal ZEN quencher located 9 nucleotides from the 5′ end (Integrated DNA Technologies). Minor groove binder hydrolysis probes were labeled with 6‐FAM at the 5′ end and nonfluorescent quencher and minor groove binder (NFQ‐MGB) at the 3′ end (Thermo Fisher Scientific, Waltham, MA). Probes obtained from the Universal Probe Library (Roche, Basel, Switzerland) were labeled with 6‐FAM at the 5′ end and a dark quencher dye at the 3′ end and contained a short sequence (8–9 nucleotides) of locked nucleic acids [Mouritzen et al., 2005]. Plasmid standards were graciously obtained from Dr. Satoshi Ishii (University of Minnesota, St. Paul, MN) and Dr. Daisuke Sano (Hokkaido University, Sapporo, Japan) and were transformed into E. coli JM109 (Promega, Madison, WI) chemically competent cells or E. coli BL21 (DE3) (Promega) electrocompetent cells using an electroporator (Bio‐Rad). Plasmids were extracted and purified using the Plasmid Miniprep Kit (Qiagen) and were quantified using Qubit fluorometric quantitation (Thermo Fisher) prior to use in qPCR and MFQPCR. All other standards were obtained as gBlock Gene Fragments (Integrated DNA Technologies). Standard curves were generated by qPCR using serial dilutions (3 × 100 to 3 × 106 copies/μL) of a standard pool containing 14 plasmid DNA and 4 gBlock DNA standards to validate the assays prior to use in MFQPCR.

The average efficiency achieved by conventional qPCR for standard curves of plasmid standards and gBlock standards was 103% ± 12.3% (n = 14) and 96.8% ± 10.1% (n = 4), respectively. The average lower limit of detection for plasmid standards and gBlock standards was 46 ± 8.1 copies/μL and 47 ± 12 copies/μL, respectively. Conventional TaqMan real‐time qPCR was performed using a MiniOpticon Real‐Time PCR System (Bio‐Rad). The final reaction mixture (20 μL) contained 2X TaqMan Universal PCR Master Mix (Thermo Fisher), 500 nM of each forward/reverse primer, 250 nM hydrolysis probe, and 3 μL template DNA/cDNA. qPCR reactions were conducted under the following thermal conditions: 50°C for 2 min, 95°C for 10 min, followed by 40 cycles of 95°C for 15 s, and 60°C for 1 min.

Prior to enumeration by MFQPCR, all cDNA/DNA samples and standard pool dilutions underwent standard target amplification (STA) PCR to increase template DNA yields. Standard pool dilutions (3 × 100 to 3 × 106 copies/μL) amplified in the 14‐cycle STA were used to generate standard curves for MFQPCR. The 20X assays (18 μM of each primer and 5 μM probe) were pooled using 1 μL per assay and 179 μL of DNA Suspension Buffer (Teknova, Hollister, CA) to make a 0.2X TaqMan primer probe mix. The reaction (5 μL) contained 2.5 μL 2X TaqMan PreAmp Master Mix (Thermo Fisher), 0.5 μL 0.2X TaqMan primer probe mix, and 1.25 μL of template cDNA/DNA. The PCR plate was processed with the following thermal cycle on an MJ Research Tetrad thermal cycler (MJ Research, Waltham, MA): 95°C for 10 min and 14 cycles of 95°C for 15 s and 60°C for 4 min. The STA products were diluted fivefold with 20 μL of nuclease free water and were used for MFQPCR. The sample premix (5 μL) contained 2.5 μL 2X TaqMan Master Mix, 0.25 μL 20X Gene Expression Sample Loading Reagent (Fluidigm, South San Francisco, CA), and 2.25 μL fivefold‐diluted STA product. The assay mix (5 μL) contained 2.5 μL 2X Assay Loading Reagent (Fluidigm) and 2.25 μL 20X TaqMan primer probe mix. Aliquots (5 μL) of each sample and quadruplicates of each assay were loaded onto a 96.96 chip (Fluidigm). MFQPCR was performed in a Biomark HD Real‐Time PCR (Fluidigm) using the following thermal conditions: 70°C for 30 min, 25°C for 10 min, and 95°C for 1 min, followed by 35 cycles of 96°C for 5 s and 60°C for 20 s. ROX was used as a passive dye reference. Two MFQPCR chips were run for extracts from 0.45 μm pore nitrocellulose filters and from 1.6 μm pore glass fiber filters. MFQPCR label is indicated in Table 3.

Table 3.

MFQPCR Specifications for Standards Used in Final Quantification

Assay LLQa ULQb Slope r 2 Efficiency (%) MFQPCR Chip Label
EntV 3 3 × 106 −3.45 0.96 95 0.45 μm pore nitrocellulose filter
E. coli ftsZ 3 3 × 106 −3.32 0.98 100 0.45 μm pore nitrocellulose filter
EHEC/STEC eaeA c 30 3 × 106 −3.61 0.99 89 1.6 μm pore glass fiber filter
EHEC/STEC stx1 c 3 3 × 106 −3.36 0.99 99 1.6 μm pore glass fiber filter
EHEC/STEC stx1 c 3 3 × 106 −3.25 0.94 103 0.45 μm pore nitrocellulose filter
EHEC/STEC stx2 c 3 3 × 106 −3.42 0.96 96 1.6 μm pore glass fiber filter
Shigella spp. ipaH 7.8 3 3 × 106 −3.44 0.96 95 1.6 μm pore glass fiber filter
Shigella spp. ipaH all 3 3 × 106 −3.21 0.97 104 1.6 μm pore glass fiber filter
Salmonella spp. invA 3 3 × 106 −3.35 0.95 99 0.45 μm pore nitrocellulose filter
Salmonella spp. ttrC 3 3 × 106 −3.252 0.98 103 1.6 μm pore glass fiber filter
Salmonella spp. ttrC 3 3 × 106 −3.16 0.94 107 0.45 μm pore nitrocellulose filter
Vibrio cholerae ctxA 3 3 × 106 −3.37 0.94 98 0.45 μm pore nitrocellulose filter
Open in a new tab
a

Lower limit of quantification.

b

Upper limit of quantification.

c

Shiga toxin producing E. coli. Two MFQPCR chips were run, one was for the extracts from the 0.45 μm pore nitrocellulose filters and another from 1.6 μm pore fiber glass filters.

2.4. Genome Extraction, RT, and qPCR Inhibition Analysis

Control experiments were conducted to measure the efficiency of genome extraction and to determine whether potential inhibitors present in water sample extracts had an effect on cDNA synthesis (reverse transcription) and qPCR. Reverse transcription (RT) inhibition was evaluated by conducting RT and qPCR as well as comparing standard curves produced in extracted environmental samples and nuclease‐free water. Human rotavirus Wa was used as a control virus to measure the presence of inhibitors in the collected water samples. Human rotavirus Wa was obtained and propagated as described elsewhere [Romero‐Maraccini et al., 2013]. Virus particles were extracted using the E.Z.N.A. Total RNA Kit I (Omega Bio‐tek, Norcross, GA) and quantified using Qubit fluorometric quantitation (ThermoFisher). Human rotavirus Wa extracts (<0.2 ng/mL) were serially diluted and spiked into nuclease‐free water and collected water sample extracts. cDNA synthesis and conventional qPCR were performed as described previously. Following enumeration, quantification cycle (C q) values from environmental water samples versus nuclease‐free water were compared for statistical differences using a two‐way analysis of variance (ANOVA) by the statistical software Origin 2016. Significantly higher C q values in environmental samples were an indicator of inhibitors in collected water samples. Differences in standard curves were determined to be significant for p values less than 0.05.

PCR inhibition was evaluated for the STA and MFQPCR analysis by including Pseudogulbenkiania NH8B as an internal amplification control in environmental sample extracts and nuclease‐free water. Serial dilutions (106 copies/mL) of Pseudogulbenkiania NH8B standard were spiked into 1.6 μm pore filter DNA and RNA sample extracts, 0.45 μm pore filter RNA sample extracts, and nuclease‐free water prior to STA. Following enumeration, Pseudogulbenkiania NH8B standard C q values in collected water samples and nuclease‐free water were compared for statistical differences using a two‐way ANOVA. Significantly higher C q values in environmental samples were an indicator of inhibitors in collected water samples. Differences in standard curves were determined to be significant for p values less than 0.05.

The genome extraction efficiency test was conducted to quantify the relative fraction of genomic material that is recovered during the extraction procedure. Pseudogulbenkiania sp. NH8B bacteria were used as an internal control to measure genome extraction efficiency. Pseudogulbenkiania sp. NH8B bacteria were graciously obtained from Dr. Satoshi Ishii (University of Minnesota, St. Paul, MN) and were grown in R2A agar (Sigma‐Aldrich, St. Louis, MO) with 100 μg/L kanamycin antibiotic marker (Sigma‐Aldrich, St. Louis, MO). The final concentration of the Pseudogulbenkiania sp. NH8B bacteria was quantified by microscope. Pseudogulbenkiania bacteria were spiked directly onto the filters to achieve a final number of 2 × 107 cells on the filters, which were subsequently subjected to DNA extraction with 100 μL of extraction eluate. Extracted samples from the Pseudogulbenkiania sp. NH8B spiked filters were enumerated in triplicate using conventional qPCR. Extraction efficiency was calculated by dividing the fraction of quantified Pseudogulbenkiania bacteria by the concentration of the cells spiked.

2.5. qPCR and MFQPCR Data Analysis

Quantification cycle (C q) values and standard pool dilutions (log copies/μL) were used to generate standard curves for each assay. C q values were determined by Bio‐Rad CFX Manager software (Bio‐Rad) and Real‐Time PCR Analysis software (Fluidigm) for qPCR and MFQPCR, respectively. Linear regression analysis was performed to fit the standard curves and calculate the goodness of fit (R 2). Assay efficiencies were calculated based on the slopes of the standard curves for each qPCR and MFQPCR assay to validate adequate target amplification [Bustin et al., 2009]. Standard curves were accepted as quantifiable if the efficiency achieved was greater than or equal to 90% and if the lower limit of detection was less than or equal to 30 copies/μL. Data points were accepted if at least two of the four replicates were found to be positive and if C q values fell into the accepted standard curve range. Data points with C q values outside of the standard curve range (i.e., detectable but not quantifiable) were considered negative [Byappanahalli et al., 2015; Sinigalliano et al., 2013]. Positive data points detected by MFQPCR were evaluated for seasonal significance using a t test and chi‐square test using Excel. Data points from each water source that contained positive detection of target genes in both seasons were compared by grouping wet season samples versus dry season samples. Positive data points that were present in only one season were not evaluated for statistical significance. Seasonal variation was determined to be significant for p values less than 0.05.

3. Results and Discussion

3.1. Sensitivities of Genome Extraction and qPCR Assays

Efficiency of genome extraction methods and sensitivity of the applied RT and qPCR assays contribute to the effectiveness of detection of target sequences in environmental samples. Genome extraction efficiencies were found to be higher for the 1.6 μm pore glass fiber filters compared to the 0.45 μm pore nitrocellulose filters. The extraction efficiency achieved was 36.6% ± 14.9% (n = 6) for 0.45 μm pore nitrocellulose filter micrometer nitrocellulose filters and 71.4% ± 11.7% (n = 6) for 1.6 μm pore glass fiber filters. These results suggest that the extraction techniques used for the 1.6 μm pore glass fiber filters resulted in a greater retention of genomic material or the extracts from the 0.45 μm pore nitrocellulose filters contained more PCR inhibitors. Overall, the extraction efficiencies achieved were slightly higher and more consistent than other studies [Ishii et al., 2014a, 2013]. The higher extraction efficiency observed here is likely due to the addition of the control microorganism to the filter before the extraction instead of to the water prior to microfiltration.

Reverse transcription (RT) and qPCR inhibition were evaluated to measure the presence or absence of contaminants that may interfere with RT and qPCR performance. A two‐way ANOVA analysis showed that reverse transcription of RNA in environmental samples versus nuclease‐free water was not significantly different (p = 0.22), suggesting that sample extracts did not contain reverse transcription inhibitors. A negative control was also included to validate that the internal control (Human Rotavirus Wa) was not previously present in the sample. The PCR inhibition analysis conducted in parallel with MFQPCR revealed that some environmental samples contained PCR inhibition. A two‐way ANOVA analysis comparing standards versus environmental samples was not statistically different for the 1.6 μm pore MFQCPR plate (p =0.203) but were statistically different for the 0.45 μm pore nitrocellulose filter MFQPCR plate (p = 0.01). These findings suggest that PCR inhibitors were present in 0.45 μm pore nitrocellulose filter sample extracts. Sample extracts were not diluted in response to the presence of inhibition due to the anticipated low abundance of target genes in the extracted samples. The difference in PCR inhibitor presence between the 1.6 μm pore and 0.45 um pore nitrocellulose filter sample extracts is potentially due to the differences in genome extraction, which may result in excess PCR inhibitors, such as salts and phenol in sample extracts [Fleige and Pfaffl, 2006]. This is consistent with the MFQPCR results, which showed a greater number of positive data points in 1.6 μm pore sample extracts.

The average lower limit of quantification (LLQ) of detection for the final selected MFQPCR assays was 5 ± 8 copies/μL. This detection limit corresponds to an average LLQ of approximately 2.65–2.95 log copies/L in drinking water and 3.26–3.56 log copies/L in drainage channel waters and Lake Victoria water. These detection limits are comparable to similar studies [Ishii et al., 2014a, 2014b, 2013].

3.2. Detection and Quantification of Target Genes

The presence and quantity of specific genes from a microbe was detected by using MFQPCR evaluated for each individual water source. Data from different water sources were compared, as well as the seasonal variation between collection times. These factors were utilized to observe the trends among groups of samples. Standard curves that were accepted based on efficiency and linear dynamic range included assays detecting three genes from EHEC (eaeA, stx1, and stx2), two genes from Shigella spp. (ipaH 7.8 and ipaH all), Salmonella spp. (invA and ttrC), E. coli (ftsZ), V. cholerae (ctxA), and enterovirus (Table 3). The average efficiency achieved for standard curves of the final selected assays was 99.2% ± 4.96% (n = 12).

The standard curves and corresponding sample quantifications for enterovirus, V. cholerae, E. coli, and Salmonella spp. (invA), were based on data obtained from the MFQCPR chip run with 0.45 μm pore nitrocellulose filter sample extracts. The standard curves and corresponding sample quantifications for EHEC/STEC (eaeA and stx2) and Shigella spp. (ipaH 7.8 and ipaH all) were based on data obtained from the MFQPCR chip run with 1.6 μm pore sample extracts. The standard curves and corresponding sample quantification for Salmonella spp. (ttrC) and EHEC/STEC (stx1) were based on data obtained from each MFQPCR chip containing 1.6 μm pore or 0.45 μm pore nitrocellulose filter sample extracts. Data were analyzed by comparing relative concentration of the gene(s) of a target pathogen in water sources in the dry season versus the wet season. The presence of target genes in a water sample was assumed to indicate the presence of the corresponding microorganism. Positive data points were expressed as gene copies/L of water.

3.3. Occurrence of Target Genes Based On Water Source Type

Of the accepted positive detection data points (n = 441), 14% were positive for EHEC/STEC eaeA, 4% were positive for EHEC/STEC stx1, 29% were positive for EHEC/STEC stx2, 7% were positive for Shigella spp. ipaH 7.8, 10% were positive for Shigella spp. ipaH all, 0.5% were positive for Salmonella spp. invA, 2% were positive for Salmonella spp. ttrC, 0.2% were positive for V. cholerae ctxA, and 7% were positive for enterovirus.

We observed that the presence of certain pathogens' genes depended on water source type. There are 441 positive data points, indicating the detection of a target gene. Among these data points, 12%, 6%, 53%, and 29% were from protected springs, the public tap, drainage channel waters, and Lake Victoria, respectively. Drainage channel waters, B3 and Ka1, contained the highest abundance of target genes over the 7 month period, followed by Lake Victoria water samples, G12. Protected springs, B12, L12, N12, and M12, and the public tap, Ka2, contained the least abundance of target genes. About 36% and 38% of positive data points for protected springs and public tap, respectively, were only found in one season.

The variation in the presence and abundance of these genes from pathogens is most obvious among the different water source types (Figure 2a). Namely, drainage channels and surface waters generally possessed a higher abundance of all target genes as compared to drinking water sources. However, among drinking water sources, some sources showed greater presence of contamination than others, highlighting unique features of these water sources that may or may not influence their susceptibility to contamination. Among four protected spring, the one at Mengo (M12) showed the greatest abundance of pathogenic gene presence (EHEC/STEC eaeA, EHEC/STEC stx1, EHEC/STEC stx2, Shigella spp. ipah 7.8, and Shigella spp. ipah all), most of which occurred during the dry season. Luzira and Namuwongo protected springs (L12 and N12, respectively) contained the lowest number of gene copies, implying that these protected springs had the lowest concentrations of pathogens present in the water. In addition, sampling sites L12 and N12 were observed to have the best maintained infrastructure for water collection.

Figure 2.

Figure 2

Open in a new tab

(A) Abundance of target genes in water sources during the wet and dry seasons. EHEC/STEC stx2 and E. coli ftsZ genes were the most prevalent across all water source types. (B) Frequency of target gene presence in water sources during the wet and dry seasons. The frequency was calculated as the number of sampling events for which the target gene was present divided by the total number of sampling events.

3.4. Occurrence of Target Genes Based On Seasonal Variation

While seasonal variation of waterborne disease incidence is well accepted [Colwell, 1996], there has been little research examining the impact of seasonality on the presence and quantity of waterborne pathogens in environmental water in developing countries. High‐sensitivity and high‐throughput means of measuring longitudinal microbial water quality in limited resource settings is a knowledge gap that this study was designed to address. Data generated from MFQPCR assays were used to examine if seasonal variations including precipitation and runoff, among others, may be related to the observed increases or decreases in target pathogen concentrations in each water source. The presence of target genes in water samples was evaluated for frequency of detection in the wet and dry seasons and for the duration of the study period (Figure 2b). The EHEC/STEC stx2 gene was detected in all water sources in both wet and dry seasons, indicating the widespread prevalence of Shiga toxin producing E. coli (STEC) in diverse water sources. In contrast, the Salmonella invA and ttrC genes were detected only during the wet season in water sources containing these genes. The V. cholerae ctxA gene was detected only once during the study period (in the dry season), implying that pathogenic V. cholerae was not common in the studied water supplies (Figure 2b).

Overall, there was an increased abundance of genes from target pathogens in the wet season versus the dry season for water from drainage channels. The presence of Shigella spp. ipaH 7.8 and ipaH all, EHEC/STEC eaeA, E. coli ftsZ, and enterovirus genes was significantly higher in the wet season compared to the dry season for drainage channel B3 (t test, p<0.05). EHEC/STEC stx1 and Salmonella spp. invA and ttrC were also detected in B3 only during the wet season, although they occurred only once. As shown in Table 4, drainage channel Ka1 showed statistically significant seasonal trends for the presence of E. coli ftsZ gene (p = 0.0003), EHEC/STEC eaeA (p = 0.03) and EHEC/STEC stx2 (p = 0.03) genes. Similar to the drainage channel B3, single occurrences of EHEC/STEC stx1 and Salmonella spp. ttrC genes were observed for the drainage channel Ka1 in the wet season only. The abundance and seasonality of microbial pathogen presence in the drainage channels studied supports previous understanding of increased disease incidence during the wet season [Gubler et al., 2001]. It is likely that these genes were found to be more consistently at drainage due to greater direct human activity [Colwell, 1996], as people were observed to use the water for washing boots and tools.

Table 4.

Statistical Significance of Wet and Dry Season Resultsa

Site Pathogen Gene p Value
B3 Shigella ipaH 7.8 0.003
Shigella ipaH all 0.003
EHEC/STEC eaeA 0.0002
E. coli ftsZ 0.0003
Enterovirus 5′ NCR 0.0003
Ka1 EHEC/STEC eaeA 0.03
EHEC/STEC stx2 0.03
E. coli ftsZ 0.0003
G12 EHEC/STEC eaeA 0.06
E. coli ftsZ 0.003
EHEC/STEC stx2 0.0007
Ka2 EHEC/STEC stx2 0.05
B12 EHEC/STEC stx2 0.011
L12 EHEC/STEC stx2 0.17
E. coli ftsZ 0.003
Open in a new tab
a

Statistical analysis of target gene presence was evaluated by a chi‐square test, and seasonal variation was considered significant for p values < 0.05. Samples G1 and G2 were combined for analysis as G12, due to small numbers of detected genes in each set. Similarly, B1 and B2 and L1 and L2 were combined as B12 and L12, respectively.

For surface water samples collected at Lake Victoria, G12, target genes showed inconclusive seasonal variability. While EHEC/STEC eaeA showed no seasonal variation (Table 4), the increased presence of E. coli ftsZ (t test, p<0.01) was observed in the wet season. EHEC/STEC stx2 appeared more in dry season, but its concentrations were insignificantly different for samples collected in wet and dry seasons (t test, p = 0.3). Shigella spp. ipaH all gene in Lake Vitoria was detected only once during the wet season. The public tap (Ka2) was observed to have similar levels of the EHEC/STEC stx2 gene during the dry and wet seasons (Table 4, p = 0.05). For public tap Ka2, Salmonella spp. ttrC and E. coli ftsZ genes were detected only in the wet season, and the V. cholerae ctxA gene was observed only in the dry season (Figure 2b). The presence of EHEC stx2 was not statistically significant with seasonal variation for protected spring L12 (p = 0.17) but significant for protected spring B12 (p = 0.011, Table 4). For protected spring L12, the presence of E. coli ftsZ was also significant with seasonal variation (p = 0.003). These data suggest that seasonal variations play an insignificant role in altering the water quality of the protected springs, public tap water, and Lake Victoria water.

Water sources of the same type showed varying responses to seasonality, suggesting that while seasonal variation of pathogenic gene presence may depend on water source type, each water source is a complex environmental system. For example, water samples taken from the protected spring at Mengo, as compared to protected springs at Luzira and Namuwongo, showed higher abundance of target genes overall. The structural integrity of the protected spring at Mengo was also inferior to the protected springs at Luzira and Namuwongo, indicating that the site conditions, such as water infrastructure and surrounding sanitation infrastructure, may influence water contamination pathways.

This longitudinal study conducted for seven months in Kampala, Uganda, presents new knowledge on how different water sources respond to seasonal variability in respect to detection of human enteric pathogens. There were several limitations to this study. First, pathogens of interest were preselected based on previous studies which limits the opportunity to identify new microbes present in different water sources. Second, PCR results were underestimated as samples were not diluted to remove PCR inhibitors because of low DNA concentration in some samples. To overcome the limitations, future studies will use shotgun metagenomics sequencing of the samples collected longitudinally for identification of evolving human enteric pathogens along with transfer of these pathogens in aquatic environment. Future research should also focus on evaluating the unique features such as water flow patterns, human interactions, and other factors that may influence water sources and their susceptibility to microbial contamination, prior to generalizing seasonal trends. Despite these limitations, the results of this study contributed to the understanding of seasonal effects on microbial water quality of different water source types.

Conflict of Interest

The authors declare no conflicts of interest relevant to this study.

Supporting information

Supporting Information S1

Click here for additional data file. (199.5KB, pdf)

Data Set S1

Click here for additional data file. (33KB, xls)

Acknowledgments

We thank Dana Mugisa, Edison Sempiira, and Alfred Ahumuza, graduate students at Makerere University, for their help with water sampling and laboratory analysis. This study was funded in part through a grant from the Fulbright Institute of International Education and the Institute for Sustainability, Energy, and Environment (iSEE) at the University of Illinois Urbana‐Champaign and NSF IRES 1559530. Data are available in the files in the supporting information. Nora J. Sadik and Sital R. Uprety contributed equally to this work.

Sadik, N. J. , Uprety S., Nalweyiso A., Kiggundu N., Banadda N. E., Shisler J. L., and Nguyen T. H. (2017), Quantification of multiple waterborne pathogens in drinking water, drainage channels, and surface water in Kampala, Uganda, during seasonal variation, GeoHealth, 1, 258–269, doi: 10.1002/2017GH000081.

This article was corrected on 15 JUL 2019. The online version of this article has been modified to include a Conflict of Interest statement.

References

  1. Adefisoye, M. A. , Nwodo U. U., Green E., and Okoh A. I. (2016), Quantitative PCR detection and characterisation of human adenovirus, rotavirus and hepatitis A virus in discharged effluents of two wastewater treatment facilities in the Eastern Cape, South Africa, Food Environ. Virol., 8(4), 262–274, doi: 10.1007/s12560-016-9246-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Akanda, A. S. , Jutla A. S., and Colwell R. R. (2014), Global diarrhoea action plan needs integrated climate‐based surveillance, Lancet Glob. Health, 2(2), 69–70. [DOI] [PubMed] [Google Scholar]
  3. Ankarklev, J. , Hestvik E., Lebbad M., Lindh J., Kaddu‐Mulindwa D. H., Andersson J. O., Tylleskär T., Tumwine J. K., and Svärd S. G. (2012), Common Coinfections of Giardia intestinalis and Helicobacter pylori in non‐symptomatic Ugandan children, PLoS Negl. Trop. Dis., 6(8), e1780, doi: 10.1371/journal.pntd.0001780. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Armstrong, B. , Hajat S., Kovats S., Lloyd S., Scovronick N., and Wilkinson P. (2012), Climate change: How can epidemiology best inform policy?, Epidemiology, 23(6), 780–784, doi: 10.1097/EDE.0b013e31826d0d13. [DOI] [PubMed] [Google Scholar]
  5. Black, R. E. , Morris S. S., and Bryce J. (2003), Where and why are 10 million children dying every year?, Lancet, 361(9376), 2226–2234. [DOI] [PubMed] [Google Scholar]
  6. Bonadonna, L. , Briancesco R., Ottaviani M., and Veschetti E. (2002), Occurence of Cryptosporidium oocysts in sewage effluents and correlation with microbial, chemical and physical water variables, Environ. Monit. Assess., 75(3), 241–252. [DOI] [PubMed] [Google Scholar]
  7. Bush, K. F. , et al. (2011), Impacts of climate change on public health in India: Future research directions, Environ. Health Perspect., 119(6), 765–770, doi: 10.1289/ehp.1003000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bustin, S. A. , et al. (2009), The MIQE guidelines: Minimum information for publication of quantitative real‐time PCR experiments, Clin. Chem., 55(4), 611–622, doi: 10.1373/clinchem.2008.112797. [DOI] [PubMed] [Google Scholar]
  9. Bwire, G. , Malimbo M., Maskery B., Kim Y. E., Mogasale V., and Levin A. (2013), The burden of cholera in Uganda, PLoS Negl. Trop. Dis., 7(12), e2545, doi: 10.1371/journal.pntd.0002545. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Byappanahalli, M. N. , Nevers M. B., Whitman R. L., and Ishii S. (2015), Application of a microfluidic quantitative polymerase chain reaction technique to monitor bacterial pathogens in beach water and complex environmental matrices, Environ. Sci. Technol. Lett., 2(12), 347–351, doi: 10.1021/acs.estlett.5b00251. [DOI] [Google Scholar]
  11. Colwell, R. R. (1996), Global climate and infectious disease: The cholera paradigm, Science, 274(5295), 2025–2031. [DOI] [PubMed] [Google Scholar]
  12. Costafreda, M. I. , Bosch A., and Pinto R. M. (2006), Development, evaluation, and standardization of a real‐time TaqMan reverse transcription‐PCR assay for quantification of hepatitis A virus in clinical and shellfish samples, Appl. Environ. Microbiol., 72(6), 3846–3855, doi: 10.1128/AEM.02660-05. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dimanin, P. (2012), Exploring Livelihoods of the Urban Poor in Kampala, ACF Int., Uganda. [Google Scholar]
  14. Dobson, S. , Nyamweru H., and Dodman D. (2015), Local and participatory approaches to building resilience in informal settlements in Uganda, Environ. Urban., 27(2), 605–620, doi: 10.1177/0956247815598520. [DOI] [Google Scholar]
  15. Field, C. , Barros V., Mach K., and Mastrandrea M. (2014), Climate change 2014: Impacts, adaptation, and vulnerability; Technical Summary, Intergovernmental Panel on Climate Change.
  16. Fleige, S. , and Pfaffl M. W. (2006), RNA integrity and the effect on the real‐time qRT‐PCR performance, Mol. Asp. Med., 27(2–3), 126–139, doi: 10.1016/j.mam.2005.12.003. [DOI] [PubMed] [Google Scholar]
  17. Fout, G. S. , Dahling D. R., and Safferman R. S. (2001), USEPA Manual of Methods for Virology, U. S. E. P. Agency, Washington, D. C. [Google Scholar]
  18. Funk, C. , Rowland J., Eilerts G., and White L. (2012), A Climate Trend Analysis of Uganda, U. S. G. S. U.S. Dep. of the Inter., USAID, Rolla Serv. Cent. [Google Scholar]
  19. Gonzalez‐Escalona, N. , Hammack T. S., Russell M., Jacobson A. P., De Jesus A. J., Brown E. W., and Lampel K. A. (2009), Detection of live Salmonella sp. cells in produce by a TaqMan‐based quantitative reverse transcriptase real‐time PCR targeting invA mRNA, Appl. Environ. Microbiol., 75(11), 3714–3720, doi: 10.1128/AEM.02686-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gubler, D. J. , Reiter P., Ebi K. L., Yap W., Nasci R., and Patz J. A. (2001), Climate variability and change in the United States: Potential impacts on vector‐ and rodent‐borne diseases, Environ. Health Perspect., 109(Suppl. 2), 223–233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Harwood, V. J. , Levine A. D., Scott T. M., Chivukula V., Lukasik J., Farrah S. R., and Rose J. B. (2005), Validity of the indicator organism paradigm for pathogen reduction in reclaimed water and public health protection, Appl. Environ. Microbiol., 71(6), 3163–3170, doi: 10.1128/AEM.71.6.3163-3170.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Helena, S. M. , Wolfert M. A., Desmarais T. R., and Palmer C. J. (1999), Sources of Escherichia coli in a coastal subtropical environment, Appl. Environ. Microbiol., 66(1), 230–237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Holt, K. E. , et al. (2013), Tracking the establishment of local endemic populations of an emergent enteric pathogen, Proc. Natl. Acad. Sci. U.S.A., 110(43), 17,522–17,527, doi: 10.1073/pnas.1308632110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Howard, G. , Teuton J., Luyima P., and Odongo R. (2002), Water usage patterns in low‐income urban communities in Uganda: Implications for water supply surveillance, Int. J. Environ. Health Res., 12(1), 63–73, doi: 10.1080/09603120120110068. [DOI] [PubMed] [Google Scholar]
  25. Ikner, L. A. , Gerba C. P., and Bright K. R. (2012), Concentration and recovery of viruses from water: A comprehensive review, Food Environ. Virol., 4(2), 41–67, doi: 10.1007/s12560-012-9080-2. [DOI] [PubMed] [Google Scholar]
  26. Ishii, S. , Segawa T., and Okabe S. (2013), Simultaneous quantification of multiple food‐ and waterborne pathogens by use of microfluidic quantitative PCR, Appl. Environ. Microbiol., 79(9), 2891–2898. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Ishii, S. , Kitamura G., Segawa T., Kobayashi A., Miura T., Sano D., and Okabe S. (2014a), Microfluidic quantitative PCR for simultaneous quantification of multiple viruses in environmental water samples, Appl. Environ. Microbiol., 80(24), 7505–7511, doi: 10.1128/AEM.02578-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ishii, S. , Nakamura T., Ozawa S., Kobayashi A., Sano D., and Okabe S. (2014b), Water quality monitoring and risk assessment by simultaneous multipathogen quantification, Environ. Sci. Technol., 48(9), 4744–4749, doi: 10.1021/es500578s. [DOI] [PubMed] [Google Scholar]
  29. Johansson, E. W. , Carvajal L., Newby H., Young M., and Wardlaw T. (2012), Pneumonia and diarrhoea: Tackling the deadliest diseases for the world's poorest childrenRep., 86 pp, Unicef.
  30. Jothikumar, N. , Cromeans T. L., Hill V. R., Lu X., Sobsey M. D., and Erdman D. D. (2005), Quantitative real‐time PCR assays for detection of human adenoviruses and identification of serotypes 40 and 41, Appl. Environ. Microbiol., 71(6), 3131–3136, doi: 10.1128/AEM.71.6.3131-3136.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Katukiza, A. Y. , Ronteltap M., van der Steen P., Foppen J. W. A., and Lens P. N. L. (2014), Quantification of microbial risks to human health caused by waterborne viruses and bacteria in an urban slum, J. Appl. Microbiol., 116(2), 447–463, doi: 10.1111/jam.12368. [DOI] [PubMed] [Google Scholar]
  32. Kigobe, M. , Mcintyre N., Wheater H., and Chandler R. (2011), Multi‐site stochastic modelling of daily rainfall in Uganda, Hydrol. Sci. J., 56(1), 17–33, doi: 10.1080/02626667.2010.536548. [DOI] [Google Scholar]
  33. Kim, J.‐H. , Nelson K. E., Panzner U., Kasture Y., Labrique A. B., and Wierzba T. F. (2014), A systematic review of the epidemiology of hepatitis E virus in Africa, BMC Infect. Dis., 14, 308–308, doi: 10.1186/1471-2334-14-308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Kitajima, M. , Matsubara K., Sour S., Haramoto E., Katayama H., and Ohgaki S. (2009), First detection of genotype 3 hepatitis E virus RNA in river water in Cambodia, Trans. R. Soc. Trop. Med. Hyg., 103(9), 955–957, doi: 10.1016/j.trstmh.2009.04.004. [DOI] [PubMed] [Google Scholar]
  35. Kitajima, M. , Oka T., Takagi H., Tohya Y., Katayama H., Takeda N., and Katayama K. (2010), Development and application of a broadly reactive real‐time reverse transcription‐PCR assay for detection of murine noroviruses, J. Virol. Methods, 169(2), 269–273, doi: 10.1016/j.jviromet.2010.07.018. [DOI] [PubMed] [Google Scholar]
  36. Liu, J. , et al. (2012), A laboratory‐developed TaqMan array card for simultaneous detection of 19 enteropathogens, J. Clin. Microbiol., 51(2), 472–480. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Lund, V. (1996), Evaluation of E. coli as an indicator for the presence of Campylobacter jejuni and Yersinia enterocolitica in chlorinated and untreated oligotrophic lake water, Water Res., 30(6), 1528–1534. [Google Scholar]
  38. Lwasa, S. (2010), Adapting urban areas in Africa to climate change: The case of Kampala, Curr. Opin. Environ. Sustain., 2(3), 166–171, doi: 10.1016/j.cosust.2010.06.009. [DOI] [Google Scholar]
  39. Mans, J. , Armah G. E., Steele A. D., and Taylor M. B. (2016), Norovirus epidemiology in Africa: A review, PLoS One, 11(4), e0146280, doi: 10.1371/journal.pone.0146280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Mattioli, M. C. , Pickering A. J., Gilsdorf R. J., Davis J., and Boehm A. B. (2013), Hands and water as vectors of diarrheal pathogens in Bagamoyo, Tanzania, Environ. Sci. Technol., 47(1), 355–363, doi: 10.1021/es303878d. [DOI] [PubMed] [Google Scholar]
  41. McIver, L. , et al. (2016), Health impacts of climate change in Pacific Island countries: A regional assessment of vulnerabilities and adaptation priorities, Environ. Health Perspect., 124(11), 1707–1714, doi: 10.1289/ehp.1509756. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Mellor, J. E. , et al. (2016), Planning for climate change: The need for mechanistic systems‐based approaches to study climate change impacts on diarrheal diseases, Sci. Total Environ., 548–549, 82–90, doi: 10.1016/j.scitotenv.2015.12.087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Miura, T. , Parnaudeau S., Grodzki M., Okabe S., Atmar R. L., and Le Guyader F. S. (2013), Environmental detection of genogroup I, II, and IV noroviruses by using a generic real‐time reverse transcription‐PCR assay, Appl. Environ. Microbiol., 79(21), 6585–6592, doi: 10.1128/AEM.02112-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Monpoeho, S. , Dehee A., Mignotte B., Schwartzbrod L., Marechal V., Nicholas J. C., Billaudel S., and Ferre V. (2000), Quantification of Enterovirus RNA in sludge samples using single tube real‐time RT‐PCR, BioTech., 29(1), 88–93. [DOI] [PubMed] [Google Scholar]
  45. Mouritzen, P. , Noerholm M., Nielsen P., Jacobsen N., Lomholt C., Pfundheller H., and Tolstrup N. (2005), Probe library: A new method for faster design and execution of quantitative real‐time PCR, Nat. Methods, 2(4), 313–316. [Google Scholar]
  46. Mshana, S. , Joloba M., Kakooza A., and Kaddu‐Mulindwa D. (2009), Campylobacter spp among Children with acute diarrhea attending Mulago hospital in Kampala‐Uganda, Afr. Health Sci., 9(3), 201. [PMC free article] [PubMed] [Google Scholar]
  47. Nakawesi, J. , Wobudeya E., Ndeezi G., Mworozi E., and Tumwine J. (2010a), Prevalence and factors associated with rotavirus infection among children admitted with acute diarrhea in Uganda, BMC Pediatr., 10(1), 1–5, doi: 10.1186/1471-2431-10-69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Nakawesi, J. S. , Wobudeya E., Ndeezi G., Mworozi E. A., and Tumwine J. K. (2010b), Prevalence and factors associated with rotavirus infection among children admitted with acute diarrhea in Uganda, BMC Pediatr., 10, 69, doi: 10.1186/1471-2431-10-69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Olmstead, S. M. (2014), Climate change adaptation and water resource management: A review of the literature, Energy Econ., 46, 500–509, doi: 10.1016/j.eneco.2013.09.005. [DOI] [Google Scholar]
  50. Rochelle‐Newall, E. , Nguyen T. M., Le T. P., Sengtaheuanghoung O., and Ribolzi O. (2015), A short review of fecal indicator bacteria in tropical aquatic ecosystems: Knowledge gaps and future directions, Front. Microbiol., 6, 308, doi: 10.3389/fmicb.2015.00308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Romero‐Maraccini, O. C. , Sadik N. J., Rosado‐Lausell S. L., Pugh C. R., Niu X. Z., Croue J. P., and Nguyen T. H. (2013), Sunlight‐induced inactivation of human Wa and porcine OSU rotaviruses in the presence of exogenous photosensitizers, Environ. Sci. Technol., 47(19), 11004–11012, doi: 10.1021/es402285u. [DOI] [PubMed] [Google Scholar]
  52. Rugadya, M. A. , Kamusiime H., Nsamba‐Gayiiya E., and Koojo C. (2008), National Slum Upgrading Strategy and Action Plan, edited by, H. a. U. D. Minist. of Lands, Nairobi. [Google Scholar]
  53. Sinigalliano, C. D. , et al. (2013), Multi‐laboratory evaluations of the performance of Catellicoccus marimammalium PCR assays developed to target gull fecal sources, Water Res., 47(18), 6883–6896, doi: 10.1016/j.watres.2013.02.059. [DOI] [PubMed] [Google Scholar]
  54. Tago, K. , Ishii S., Nishizawa T., Otsuka S., and Senoo K. (2011), Phylogenetic and functional diversity of denitrifying bacteria isolated from various rice paddy and rice‐soybean rotation fields, Microbes Environ., 26(1), 30–35, doi: 10.1264/jsme2.ME10167. [DOI] [PubMed] [Google Scholar]
  55. Tumwine, J. K. , Kekitiinwa A., Nabukeera N., Akiyoshi D. E., Rich S. M., Widmer G., Feng X., and Tzipori S. (2003), Cryptosporodium parvum in children with diarrhea in Mulago hospital, Kampala, Uganda, Am. J. Trop. Med. Hygiene, 68(6), 710–715. [PubMed] [Google Scholar]
  56. Vermeiren, K. , Van Rompaey A., Loopmans M., Serwajja E., and Mukwaya P. (2012), Urban growth of Kampala, Uganda: Pattern analysis and scenario development, Landsc. Urban Plan., 106(2), 199–206, doi: 10.1016/j.landurbplan.2012.03.006. [DOI] [Google Scholar]
  57. Victoria, M. , Guimaraes F., Fumian T., Ferreira F., Vieira C., Leite J. P., and Miagostovich M. (2009), Evaluation of an adsorption‐elution method for detection of astrovirus and norovirus in environmental waters, J. Virol. Methods, 156(1‐2), 73–76, doi: 10.1016/j.jviromet.2008.11.003. [DOI] [PubMed] [Google Scholar]
  58. Walker, C. L. F. , Sack D., and Black R. E. (2010), Etiology of diarrhea in older children, adolescents and adults: A systematic review, PLoS Negl. Trop. Dis., 4(8), e768, doi: 10.1371/journal.pntd.0000768. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. World Health Organization (2015), Key Facts from JMP 2015 ReportRep ., World Health Organization.
  60. Zhang, C. , Xu J., Ma W., and Zheng W. (2006), PCR microfluidic devices for DNA amplification, Biotechnol. Adv., 24(3), 243–284, doi: 10.1016/j.biotechadv.2005.10.002. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information S1

Click here for additional data file. (199.5KB, pdf)

Data Set S1

Click here for additional data file. (33KB, xls)

Articles from GeoHealth are provided here courtesy of Wiley

RESOURCES