CHAPTER 1
Health
Evaluation of detection methods targeting host-specific Bacteroides Spp. as a microbial source tracking marker
T. Miura, Y.M. Chan, Y. Masago and T. Omura
Department of Civil and Environmental Engineering, Tohoku University. 6-6-06, Aoba, Sendai, 980-8579, Japan (E-mail: miura@water.civil.tohoku.ac.jp, stephycym@gmail.com, masago@water.civil.tohoku.ac.jp, omura@water.civil.tohoku.ac.jp)
Abstract Microbial source tracking is an increasingly used approach to determine host-specific contributions of fecal contamination to water bodies. Throughout the world, many studies have been done on fecal contamination in various kinds of water bodies; however, studies on fecal contamination in sediment are limited. In this study, detection methods for host-specific Bacteroides spp. previously developed were evaluated for their ability for applications to sediment, as well as water samples. In addition, fecal coliforms in sediment and its water column were quantified in order to compare their concentrations and to evaluate contribution of the sediment to the water in case of sediment's resuspension to the water. Bacteroides were detected in 22 out of 36 sediment samples (61%) and 1 out of 18 water samples (6%). The primer set for human-specific Bacteroides developed by Okabe et al. (2007) showed much higher positive ratio (47%, n = 17) in the sediment samples compared with that developed by Kildare et al. (14%, n = 5) (2007). The concentration of fecal coliforms in the sediment samples (CFU/100 g (dry weight)) was 12 to 663 times as high as that in the water samples (CFU/100 ml). These results imply that sediment is more appropriate target for microbial source tracking.
Keywords fecal coliforms, fecal contamination, host-specific Bacteroides, microbial source tracking, sediment
INTRODUCTION
In Southeast Asia, diarrheal diseases are responsible for as much as 18% of all deaths of children under 5 years of age (WHO, 2005). WHO (2004) also reported that 88% of diarrheal diseases is attributed to unsafe water supply and inadequate sanitation. Water bodies used as drinking water sources in developing countries are contaminated by human and animal feces (Moe et al., 1991; Vollaard et al., 2005) that may include waterborne pathogens partly due to the insufficient wastewater treatment systems. As many pathogens are host-specific, it is important to assess contribution of human feces to the fecal contamination in water.
Microbial source tracking is an increasingly used approach to determine host-specific contributions of fecal contamination to water bodies (Kildare et al., 2007) and appropriate measures against the fecal contamination can be discussed based on the result of the source tracking. Bacteroides, a group of anaerobic bacteria, has been applied for fecal source tracking in water (Reischer et al., 2006; Kildare et al., 2007; Savichtcheva et al., 2007) mainly due to its high abundance in the feces of warm-blooded animals (Fiksdal et al., 1985) and high host specificity (Bernhard and Field, 2000a and b).
Throughout the world, many studies have been done on fecal contamination in various kinds of water bodies, such as raw sewage, river water, seawater and tap water. However, although it is reported that sediment may provide more reliable information of long-term fecal contamination (Craig et al., 2002), studies on fecal contamination in sediment are limited, that warrants needs for the knowledge and techniques for detection and quantification of microbes in sediment.
In this study, detection methods for host-specific Bacteroides spp. previously developed were evaluated for their ability for applications to sediment samples, as well as water samples. In addition, fecal coliforms in sediment and its water column were quantified in order to compare their concentrations and to evaluate contribution of the sediment to the water in case the sediment is resuspended to the water. This evaluation experiment was done using sediment and water samples collected at the Takagi River estuary in Japan. Outcomes of this study would be applicable for fecal source tracking undertaken in other areas.
MATERIALS AND METHODS
Samples
Sediment and water samples were collected in the Takagi River estuary during the ebb tide monthly from November 2007 to May 2008. Locations of sample sites are shown in Figure 1: St.A, St.B and St.C were located at the bay, where oyster beds are placed (Figures 1 and 2); St. D was located at the river mouth; and St.E and St.F were located at the river downstream. There is a small dam to control the river flow at St.F (Figure 3).
About 100 mL of top layer sediment was collected using a Ekman-Birge type bottom sampler at St.A, St.B, St.C, St.D, St.E and St.F; about 20 L of surface seawater was collected at St.A, St.B, St.C and St.D, about 1 L of surface river water was collected at St.E and St.F, and about 1 L of treated wastewater were collected at the wastewater treatment plant (WWTP). All the samples were transported to the laboratory on ice in sterile containers and processed within a few hours of collection.
Sediment characterization
Water content and ignition loss of sediment samples were measured according to the standard protocols of Japan Sewage Works Association (1997). Briefly, sediment was dried in an oven at 105°C for 2 hr and water content (%) was calculated using the weight before and after the desiccation. Then the dried sediment was placed in an electric furnace at 600°C for 1 hr and weighed. The ignition loss (%) was calculated from these results. Particle size distribution of sediment samples was measured by Microtrac (9320HRA (X-100); NIKKISO, Tokyo).
DNA extraction and detection of Bacteroides
DNA in the sediment samples (from November 2007 to April 2008, 36 samples in total) was extracted from 0.5 g well-mixed wet sediment using ISOIL DNA extraction kit (NIPPON GENE, Tokyo), according to the protocols described by the manufacturer.
Bacteroides in the water samples (from February 2008 to April 2008, 18 samples in total) were concentrated and recovered, following the protocols of Okabe et al. (2007) with some modifications. Ten to 20 L of sea water, 1 L of river water and 1 L of treated waste water samples were passed through HA membrane filters (0.45 µm pore size; Millipore, Tokyo). The membranes were placed in phosphate buffered saline and captured cells were dispatched from the membranes by ultrasonication with Ultra S.Homogenizer (50 W, 20 kHz; Model VP-5S; TAITEC, Koshigaya). The solution containing Bacteroides was collected and centrifuged at 9000 · g at 4°C for 10 min. Supernatant was discarded and the pellet was resuspended in 1 mL of sterile Milli-Q water. Genome DNA was extracted using QIAamp DNA mini kit (QIAGEN, Tokyo), following the protocols described by the manufacturer.
The 16S rRNA gene of Bacteroides was amplified using host-specific primer sets as described in Table 1. PCR was carried out with thermal cycler (Applied Biosystems, Tokyo). Each 20 µL PCR mixture contained 10 µL of master mix (Roche, Tokyo), 4 µL of distilled deionized water, 250 nM of forward primer, 250 nM of reverse primer, and 5 µL of extracted DNA sample. The PCR conditions included a denaturing step at 95°C for 5 min, followed by 40 cycles of 95°C for 1 min, annealing temperature specified in Table 1 for 1 min, and 72°C for 20 sec, followed by a final extension step of 72°C for 30 sec. The PCR products were electrophoresed in 1.5% (w/v) agarose gel stained with ethidium bromide, and visualized by UV illumination.
Quantification of fecal coliforms
Fecal coliforms in the water samples (from December 2007 to May 2008, 42 samples in total) were quantified with membrane-filter method according to the standard protocols of Japan Sewage Works Association (1997). One, 10 or 100 mL of water sample was passed through a HA membrane filter (0.45 mm pore size and 47 mm diameter; Millipore, Tokyo); the membrane was placed on m-FC agar (MERCK, Tokyo) and incubated for 24 hr at 44.5°C. Results were recorded as colony-forming unit (CFU)/100 mL of water. Three membranes were used for each sample, and the average was used for subsequent analysis.
Fecal coliforms in the sediment samples (from December 2007 to May 2008, 36 samples in total) were recovered, following the protocols of Craig et al. (2002) with some modifications. One gram of well-mixed wet sediment sample was placed in 5 mL Milli-Q water. Samples were vortexed for 15 sec and left to settle for 10 min prior to aspirating the supernatant. Membrane-filter method mentioned above was conducted to the supernatant. Results were recorded as CFU/100 g (dry weight) of sediment following previous studies (Craig et al., 2002; Alm et al., 2003) for comparison. Three membranes were used for each sample, and the average was used for subsequent analysis.
RESULTS AND DISCUSSION
Properties of sediment samples
Figure 4 shows the composition, water content and ignition loss of the sediment samples. All top layer sediment samples in this study were mainly consisted of silt in the range of 67 ~ 85%, and water content and ignition loss of them were in the range of 62 ~ 81% and 2.9 ~ 3.8%, respectively (Figure 4).
Bacteroides in the sediment and water samples
Table 2 shows detection frequencies of host-specific Bacteroides in the sediment and water samples. Bacteroides were detected in 22 sediment samples (61%) and 1 water sample (6%) using universal primer set developed by Kildare et al. (2007); human-specific Bacteroides were detected in 17 sediment samples (47%) by using the primer set developed by Okabe et al. (2007), and detected in 5 sediment samples (14%) by using the primer set developed by Kildare et al. (2007); but human-specific Bacteroides were not detected in any water sample. No sediment or water samples were positive for ruminant- or cow-specific Bacteroides. It is possible that feces of other kinds of animals, such as seagulls, may be the fecal sources.
As for the universal Bacteroides in the sediment samples, the detection frequencies at the river downstream (St.E and St.F) were the highest (both were 83%, Table 2), followed by the river mouth (St.D) with the detection frequency of 67%; the average detection frequencies at the bay (St.A, St.B and St.C) was the lowest (44%). As for the water samples, universal Bacteroides were only detected at the river mouth (St.D).
The detection frequencies of human-specific Bacteroides in the sediment samples by using the primer set developed by Okabe et al. (2007) was much higher than that by using the primer set developed by Kildare et al. (2007); samples which were positive by using the primer set developed by Kildare et al. (2007) were all positive as well by using the primer set developed by Okabe et al. (2007). Hence, the primer set developed by Okabe et al. (2007) showed much higher sensitivity in the sediment samples in this study area. However, Okabe et al. (2007) reported that their human-specific primer set also detected Bacteroides in cow's and pig's feces thus specificity of the primer set may be lower, which resulted in higher detection frequency. Kildare et al. (2007) reported that their human-specific primer set infrequently detected Bacteroides in dog's feces (1 out of 8 samples) and that their cow-specific primer set occasionally detected Bacteroides in horse's feces (3 out of 8 samples).
Fecal coliforms in the sediment and water samples
Figure 5 shows geometric mean of concentration of fecal coliforms in the sediment and water samples. The detection limit in this study was 1 CFU/filtrated volume. As for the samples whose concentration was below the detection limit, the concentration was assumed to be 0.1 CFU/filtrated volume. In St.A through F, the concentration in the sediment samples (7.3 x 102 to 7.5 x 104 CFU/100 g (dry weight)) was 12 to 663 times higher than those in the water samples. Craig et al. (2002) reported the concentration of fecal coliforms in river sedimemt samples were 1000 times higher than those in water samples, which was comparable to our observation. The differences of the concentration in the sedimemt samples and water samples at the river downstream was smaller than that at the bay (Figure 5).
Figure 6 shows the relationship of concentration of fecal coliforms in between the sediment and water samples. On the whole, sample site that had higher concentration in the sediment samples also had higher concentration in the water samples. In another word, the concentration of fecal coliforms in the sediment and water samples was similar in regularity for geographical distribution: the concentration at the river downstream (St.E and St.F) was the highest followed by that at the river mouth (St.D), and that at the bay (St.A, St.B and St.C) was the lowest. Whereas, no significant correlation of the concentration between the sediment and water samples was observed, which has been demonstrated by Craig et al. (2002).
CONCLUSIONS
This study is the first to apply host-specific Bacteroides for fecal source tracking in sediment throughout the world. The primer set for human-specific Bacteroides developed by Okabe et al. (2007) showed much higher sensitivity in the sediment samples in this study compared with that developed by Kildare et al. (2007). The presence of host-specific Bacteroides and their higher detection frequencies in the sediment samples imply that they are suitable for fecal source tracking in sediment.
ACKNOWLEDGEMENT
This work was supported by MEXT through Special Coordination Funds for Promoting Science and Technology, as a part of the project for "Integrated Research System for Sustainability Science (IR3S)" undertaken by Tohoku University and through JSPS Fellows (19/5067, 2007).
CHAPTER 2
Groundwater quality problems and issues in the dry-zone of Sri Lanka with special reference to fluoride contamination and Chronic Kidney Disease
G. Herath and U. Ratnayake
Department of Civil Engineering, Faculty of Engineering, University of Peradeniya Peradeniya, 20400, Sri Lanka. (Email: gemunuh@pdn.ac.lk, udithar@pdn.ac.lk)
Abstract Year 2004 estimate shows that nearly 70% of people in Sri Lanka, 22.4% urban and 71.8% rural rely on groundwater for their domestic water requirements. Many groundwater sources though thought not safe enough for direct consumption, is still being used by many as no other alternative is available. Hence, as a solution, the Government of Sri Lanka a few years back took a policy decision to increase its investment on water supply with the aim of providing safe drinking water to all by year 2025. However, with many new healthrelated issues surfacing especially in drier parts of the country, realization of such a target seems to have reached far short progress than expected. Thus, in order to investigate this issue, over 1,000 wells in the dry zone of Sri Lanka used as water sources, were analyzed for water quality. Fluoride, iron and hardness were identified as the main quality concerns. In some regions, almost one fourth of wells had fluoride levels unsafe for human consumption. Additionally the initial investigations comparing the basic groundwater quality parameters in high CKD risk areas with safe areas did not show much difference. However it is too early to make any conclusions as investigations are continuing.
Keywords Chronic kidney disease, fluoride, groundwater, safe drinking water sources
INTRODUCTION
Groundwater is considered a very valuable water resource today as it provides reliable, high-quality and low-cost water for domestic, industrial and agricultural purposes. Therefore many cities and regions in the world rely on groundwater for their water supplies to take these inherited advantages. A 2004 estimate shows that nearly 70% of the population in Sri Lanka too rely on groundwater for their domestic water requirements. However, the rapid urbanization, industrialization and intensive agriculture taking place in many cities and regions have intensified the stress placed on this precious resource, because in response to the growing demand for water, groundwater is often being over extracted and the poor waste management and drainage control is contaminating aquifers with various types of pollutants.
According to WHO/UNICEF (2004) report on "Joint Monitoring Program for Water Supply and Sanitation", as of the year 2000, 76.1% of urban population was provided with a piped water supply compared to 11.4% in rural areas in Sri Lanka. In addition the population in urban and rural relying on underground well-water was estimated as 22.4% and 71.8% respectively. The percentage distribution of urban and rural population in the whole country in 1999 was estimated at 31% and 69% respectively.
