Evaluation of different analysis and identification methods for Salmonella detection in surface drinking water sources

Science of the Total Environment 409 (2011) 4435–4441

Contents lists available at ScienceDirect

Science of the Total Environment j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c i t o t e n v

Evaluation of different analysis and identification methods for Salmonella detection in surface drinking water sources Bing-Mu Hsu a,⁎, Kuan-Hao Huang a, Shih-Wei Huang a, Kuo-Chih Tseng b, Ming-Jen Su c, Wei-Chen Lin d, Dar-Der Ji d, Feng-Cheng Shih e, Jyh-Larng Chen e, Po-Min Kao a a

Department of Earth and Environmental Sciences, National Chung Cheng University, Chiayi, Taiwan, ROC Department of Internal Medicine, Buddhist Dalin Tzu Chi General Hospital, Chiayi, Taiwan, ROC Department of Clinical Pathology, Buddhist Dalin Tzu Chi General Hospital, Chiayi, Taiwan, ROC d Research and Diagnostic Center, Centers for Disease Control, Taipei, Taiwan, ROC e Department of Environmental Engineering and Health, Yuanpei University of Science and Technology, HsinChu, Taiwan, ROC b c

a r t i c l e

i n f o

Article history: Received 27 January 2011 Received in revised form 23 May 2011 Accepted 25 May 2011 Available online 22 July 2011 Keywords: Salmonella PFGE Salmonella isolation Water safety

a b s t r a c t The standard method for detecting Salmonella generally analyzes food or fecal samples. Salmonella often occur in relatively low concentrations in environmental waters. Therefore, some form of concentration and proliferation may be needed. This study compares three Salmonella analysis methods and develops a new Salmonella detection procedure for use in environmental water samples. The new procedure for Salmonella detection include water concentration, nutrient broth enrichment, selection of Salmonella containing broth by PCR, isolation of Salmonella strains by selective culture plates, detection of possible Salmonella isolate by PCR, and biochemical testing. Serological assay and pulsed-field gel electrophoresis (PFGE) can be used to identify Salmonella serotype and genotype, respectively. This study analyzed 116 raw water samples taken from 18 water plants and belonging to 5 watersheds. Of these 116, 10 water samples (8.6%) taken from 7 water plants and belonging to 4 watersheds were positive for a Salmonella-specific polymerase chain reaction targeting the invA gene. Guided by serological assay results, this study identified 7 cultured Salmonella isolates as Salmonella enterica serovar: Alnaby, Enteritidis, Houten, Montevideo, Newport, Paratyphi B var. Java, and Victoria. These seven Salmonella serovars were identified in clinical cases for the same geographical areas, but only one of them was 100% homologous with clinical cases in the PFGE pattern. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Salmonella species are Gram-negative, non-spore forming, flagellated zoonotic bacteria that can infect people, mammals, birds, reptiles, and other animals (Covert, 1999; Garcia del Portillo, 2000). The genus Salmonella is composed of two species, Salmonella enterica and Salmonella bongori, seven subgroups, and more than 2,500 serovars, all of which are believed to be capable of causing human illnesses, such as typhoid fever, paratyphoid fever, and other salmonelloses (Boyd et al., 1996; Kuo et al., 1997; Popoff et al., 2003; Porwollik et al., 2004; Bhan et al., 2005). Salmonellosis typically causes gastrointestinal infection by the genus Salmonella that can be divided into four syndromes: enteric fever (typhoid-like disease), gastroenteritis (food poisoning), bacteremia with or without gastroenteritis, and the asymptomatic carrier state. Typhoid fever is caused by Salmonella typhi, which invade beyond the

⁎ Corresponding author at: Department of Earth and Environmental Sciences, National Chung Cheng University, 168 University Road, Minhsiung Township, Chiayi County 62102, Taiwan, ROC. Tel.: +886 5 2720411x66218; fax: +886 5 2720807. E-mail address: [email protected] (B.-M. Hsu). 0048-9697/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2011.05.052

epithelium of the gastrointestinal tract and cause enteric fever (Hohmann, 2001; Nair et al., 2002; Percival et al., 2004). However, other forms of salmonellosis occur at high frequencies in both industrialized countries and developing countries (Bell and Kyriakides, 2002). Salmonella enterica serovar Typhimurium and Enteritidis are the most commonly identified Salmonella in salmonellosis cases in industrialized countries. In Taiwan, the most frequent human Salmonella cases are caused by Salmonella enterica serovar Enteritidis, followed by Typhimurium, Stanley, Agona, Albany, Schwarzengrund, Newport, Choleraesuis, Derby, and Weltevreden (Chiou, 2004). Every year, approximately 1.3 million human cases of non-typhoidal Salmonella infections occur in the United States, and other industrialized countries have similar rates (Hardnett et al., 2004). Because many milder cases are not diagnosed or reported, the actual rate of infection may be much greater. The majority of human cases of salmonellosis are caused by the consumption of contaminated foods of animal origin (Soumet et al., 1999). However, Salmonella have also caused waterborne outbreaks (Furtado et al., 1998; Covert, 1999). Waterborne infections of Salmonella can be acquired by drinking contaminated water, swimming in contaminated water, or eating food washed with contaminated water.

4436

B.-M. Hsu et al. / Science of the Total Environment 409 (2011) 4435–4441

In the USA, Salmonella causes 8% of all waterborne outbreaks caused by contaminated drinking-water (Covert, 1999). The major Salmonella serovars identified in waterborne outbreaks in the USA include Typhimurium, Enteritidis, Bareilly, Javiana, Newport, and Weltervreden (Covert, 1999). The current method of detecting Salmonella in food relies on a nonselective pre-enrichment step followed by a selective enrichment step, isolation on selective agar media, identification of the organism by biochemical testing, and serotyping (APHA, 1995; D'Aoust, 1989). Waterborne pathogens often occur in relatively low concentrations in environmental waters. Therefore, some form of filtration and proliferation of target cells are needed for Salmonella detection. Furthermore, the traditional method for detecting Salmonella is timeconsuming, expensive, and laborious (Fung, 2002). For example, serotyping offers a precise and reliable method for differentiating strains, but it requires the use of more than 150 specific serum samples (Kilger and Grimont, 1993). Therefore, researchers have developed and evaluated rapid alternative methods for Salmonella detection and genotype identification, such as molecular methods. For example, the genomic macrorestriction fragment length analysis method that uses pulsed-field gel electrophoresis (PFGE) has been recognized as a reliable and discriminating tool for the genotyping of Salmonella (Giovannacci et al., 2001; Refum et al., 2002). To date, guidelines have been established for Salmonella in drinking water supplies in many countries. However, the genus Salmonella is not generally monitored in watersheds; and thresholds may not exist for raw water. Information on the diversity and occurrence of Salmonella strains in aquatic environments is very scarce, and as a consequence, the ecology of these species remains unknown. This aim of this study is to evaluate the detection efficiency of three Salmonella analysis methods, assess the occurrence of Salmonella, and identify the serotypes of Salmonella in the aquatic environments of Taiwan. In pararell, we used PFGE to genotype and compare relatedness within serotypes. 2. Materials and methods 2.1. Study location and sampling procedure A total of 116 raw water samples were collected at the intakes of 18 water purification plants (Plants A through R) in southern Taiwan. These samples were taken from the five following watersheds (Fig. 1). Choshui River, the longest river in Taiwan (186.6 km), is the main source of surface drinking water for Nantou, Changhua, and Yunlin counties; Puzih River and Bajhang River, which both run from A-Li mountain, are the main drinking water source for Chiayi county; Zengwen River, the water source of Zengwen reservoir and Wushantou reservoir, is the main irrigation and drinking water source of the Chianan plain; Kaupin River is the main drinking water supply for the Kaushiung metropolitan area, the second large city in Taiwan. The sampling period was from January 2009 to August 2009. Two liters of water were collected from each site, placed in sterilized polypropylene bottles, stored at 4 °C, and subjected to analyzing processes within 24 h of sampling.

2.3. Concentration for Salmonella To survey for the presence of Salmonella in the water samples, 1 L of each water sample was filtered through 45-mm diameter GN-6 material membranes (Pall, Mexico City, Mexico) with a pore size of 0.22 μm. Membranes were first washed by someone wearing disposable rubber gloves, and rubbed in 100 ml of sterilized deionized water. The solution was then transferred into a 50 ml conical centrifuge tube and centrifuged at 2600×g for 30 min. After aspirating the top solution, the remaining concentrated eluate was only 5 ml. Two experimental procedures – direct DNA extraction (Process I) and concentrate enrichment – were then conducted. 2.4. Cultivation for Salmonella For the enrichment procedure, 1 ml of concentrate was added to 5 ml buffered peptone water (BPW; Neogen, Lansing, MI, USA) and incubated for 16 h at 37 °C with shaking. Rappaport-Vassiliadis (RV) broth (Neogen, Lansing, MI, USA) was inoculated with 100 μl of BPW and incubated at 42 °C for 16–18 h. Two procedures – DNA extraction (Process II) and selective culture inoculation – were then conducted. In Process III, following filtration, elution, concentration, and enrichment, the enriched concentrate was streaked on CHROMagar™ Salmonella plates (CHROMagar, Paris, France) and incubated overnight at 37 °C for 18–24 h. The light purple colonies were picked from the culture plates and placed in Xylose Lysine Deoxycholate (XLD) agar (Merck, Gernsheim, Germany) at 37 °C for further analysis and identification of possible Salmonella colonies. 2.5. PCR analysis for Salmonella For the molecular analysis of Process I and II, a DNA extraction kit (Viogene, Taipei, Taiwan) was used to extract total genomic DNA following the kit manual. The suspension was then analyzed for the presence of target genes by PCR. In Process III, Salmonella suspected colonies on XLD agar were first picked and then suspended in 20 μl sterile Milli-Q water. The bacteria were lysed by heating for 10 min at 100 °C. After the bacterial debris were removed by centrifugation, the supernatant was immediately used as a template in PCR assays. To detect Salmonella species in water samples, the PCR primer pair targeted invA gene was selected in PCR assays to differentiate Salmonella and nonSalmonella (Chiu and Ou, 1996). Table 1 lists the PCR cycling conditions and the primer sequences. The reaction solution for PCR was prepared with 3 μl of the DNA templates and the PCR mixture to create a total volume of 25 μl. The PCR mixture contained 2.5 μl 10 ×Qiagen PCR buffer, 0.5 μl dNTP mix (10 mM of each dNTP), 100 pmol each of the oligonucleotide primers, and 0.2 μl Qiagen Taq polymerase mix, as well as DNase-free deionized water. PCR products were analyzed with gel electrophoresis on a 2% agarose gel (Biobasic Inc., Markham, ON, Canada) with 5 μl of the reaction solution. DNA fragments were visualized using ethidium bromide staining (0.5 μg/ml, 10 min). A 100-bp DNA ladder was used as a DNA size marker. Negative DNA controls (template DNA replaced with distilled water), positive controls (Salmonella enterica ATCC 13076), and sample DNA were analyzed in triplicate during each PCR run.

2.2. Experimental design of three analyzing methods 2.6. Biochemical testing and serological assay Fig. 2 depicts the three Salmonella analysis methods examined in this study. In Process I, the water sample was filtered, eluted/ concentrated, and then analyzed for Salmonella using the PCR method. Process II included filtration, elution/concentration, pre-enrichment, selective enrichment, and then detection of Salmonella by the PCR method. Process III modified the existing conventional process, Salmonella detection of water samples, according to the method described by Draft ISO 19250 (2003), with a few modification: filtration, elution/concentration, pre-enrichment, selective enrichment, isolation by solid selective agar, and detection by the PCR method.

All the Salmonella candidate colonies confirmed by PCR amplification in Process III were identified using an eight-tube diagnostic kit (Shin Tang Co. Ltd., Taipei, Taiwan) in the first screening, followed by screening with an API 20E system (bioMérieux, Marcy L'Etoile, France). These tests were performed following the manufacturer's instructions. The eight-tube diagnostic kit includes the triple sugar iron agar test, citrate utilization test, urease test, sulfide-indole-motility test, semivoges proskauer test, ornithine decarboxylase test, arginine dehydrolase test, and lysine dehydrolase test. The API 20E system consists of a plastic

B.-M. Hsu et al. / Science of the Total Environment 409 (2011) 4435–4441

4437

Fig. 1. Sampling sites of water sources in southwestern Taiwan.

strip of 20 individual, miniaturized tests tubes, each containing a different reagent to determine the metabolic capabilities and the genus as well as species of enteric bacteria in the family Enterobacteraceae. The Salmonella serotype was determined after an overnight culture on Kligler's iron agar (Sanofi Diagnostics Pasteur., Chaska, MN, USA) by agglutination with Difco™ Salmonella O antisera, H antisera and antiserum Vi (Becton, Dickinson and Company, Franklin Lakes, N.J., USA) following the Kauffmann-White scheme. The final confirmation of serological assay for Salmonella serovars was performed by the Taiwan Centers for Disease Control (CDC) National Reference Laboratory (NRLSalm, Taichung, Taiwan). The protocol was described in the paper of Chiou et al. (2006).

2.7. Pulsed-field gel electrophoresis The PFGE of Salmonella spp. was performed according to the standardized laboratory protocol of Pulse-Net, CDC, USA (Ribot et al., 2006). Briefly, a single colony of each presumptive Salmonella isolate was streaked on tryptic soy agar and incubated overnight at 37 °C. The cells were scraped and directly transferred into a 2 ml cell suspension buffer (100 mM Tris, 100 mM EDTA [pH 8.0]). The cell concentration was adjusted from 0.48 to 0.52 using a Microscan turbidity meter (Dade Behring, Sacramento, CA, USA). Aliquots of 400 μl adjusted cell suspension were then transferred to 1.5-ml microcentrifuge tubes with 20 μl of proteinase K (20 mg/ml), and then mixed with 400 μl of

4438

B.-M. Hsu et al. / Science of the Total Environment 409 (2011) 4435–4441

Water Sample

Filtration

Elution and Concentration

Pre-enrichment

PCR Analysis [Process I]

Selective Enrichment

Solid Selective Agar

PCR Analysis [Process II]

PCR Analysis [Process III] Fig. 2. Flowchart of three analysis processes for Salmonella detection.

1% melted SeaKem Gold agarose gels (Cambrex, East Rutherford, NJ, USA), and allowed to set immediately into plug molds. Three plugs were transferred to 50-ml polypropylene screw-cap tubes with 5 ml of cell lysis buffer (50 mM Tris, 50 mM EDTA [pH 8.0], 1% sarcosyl) and 25 μl of proteinase K (20 mg/ml) and then incubated at 54 °C in a shaker water bath for 2 h with agitation. The lysis buffer was then discarded and the plugs were washed thoroughly twice with 15 ml of sterile water and three more times with TE buffer at 50 °C for 15 min. Chromosomal DNA was digested with 50 U of XbaI (Promega, Southampton, UK), and the DNA fragments were then separated in Seakem Gold agarose gels at 14 °C using a CHEF Mapper system (Bio-Rad, Hercules, CA, USA) in 0.5× Tris–Borate–EDTA (pH 8) at a 120° fixed angle, a fixed voltage (6 V/cm) and, pulse time intervals varying from 2.2 to 35 sec for 18 h. DNA from Salmonella Braenderup H9812 restricted with XbaI was used as a size marker. After electrophoresis, the gel was stained with 1 mg/L ethidium bromide for 30 min and destained for 60–90 min with reversed osmosis water with a water change every 20–30 min. The gel was exposed on a UV transilluminator and photographed using Kodak digital camera system (Kodak Electrophoresis Documentation and Analysis System 290, Kodak Rochester, NY, USA). The macrorestriction pattern images of PFGE were analyzed using BioNumerics software v4.61 (Applied Maths, Sint-Martens-Latem, Belgium).

collected in 300 ml sampling bags (Nasco Whirl-Pak, Fort Atkinson, WI, USA). The samples were kept in coolers during transportation to the lab. Heterotrophic bacteria were measured by the spread method. Total coliforms were measured by membrane filtration procedures with a differential medium described in the Standard Method for the Examination of Water and Wastewater (Methods 9222 B and D) (APHA, 1995). Calculations of difference in the presence/absence of Salmonella in terms of water quality parameters were conducted using STATISTICA software (StatSoft, Inc., Tulsa, OK, USA). 3. Results and discussion 3.1. Comparison of Salmonella detection results for three different analysis methods This study evaluates three analysis methods combined with PCR to detect Salmonella spp. in water samples. Because the invA gene is highly conserved in almost all Salmonella serotypes (Daum et al., 2002), a primer that amplifies the invA gene by PCR was chosen. Table 2 shows the presence of invA gene-positive samples for the three analysis methods. Process I, a protocol for direct DNA extraction from water concentrate followed by PCR analysis, has been reported in previous studies to detect pathogenic bacteria. This technique directly detects Salmonella in water samples without an enrichment step, and is preferable for viable but non-culturable cells (APHA, 1995; Hsu et al., 2006, 2008). The major obstacle of this detection method is the presence of PCR inhibitors, such as humic substances, matrices, and

2.8. Physical and microbiological parameter analysis Turbidity was measured in situ using a ratio turbidimeter (HACH Co., Loveland, CO, USA). Water samples for microorganism analysis were obtained in parallel with the samples used for Salmonella detection and Table 1 Description of the primers used in PCR of invA genes in Salmonella. Target gene

Primer sequences

Denaturation, annealing and extension temperature

Cycling no.

Amplicon size (bp)

Source

invA

invA1 5'-ACAGTGCTCGTTTACGACCTGAAT-3' invA2 5'-AGACGACTGGTACTGATCGATAAT-3'

95 °C

35

244

Chiu and Ou, 1996

62 °C

72 °C

B.-M. Hsu et al. / Science of the Total Environment 409 (2011) 4435–4441 Table 2 Presence of invA gene-positive samples from three analysis processes and identification of Salmonella serotypes in the waters of southwestern Taiwan. Sampling Water site source

Total invA gene- positive number sample of Process Process Process samples I II III

A

23

–

–

–

4

–

–

–

4

–

C-2

C-2

3

–

–

–

11

–

E-11

E-11

7

–

–

–

3

–

–

–

2

–

–

–

6

–

–

–

3

–

–

–

5

–

K-2

K-2

4

–

–

–

4

–

–

–

13

–

–

–

4

–

O-1

O-1

S. enterica Houten

4

–

P-1

P-1

P-3

P-3

S. enterica Montevideo S. enterica Paratyphi B var. Java

P-4 Q-8

–

R-4 R-7

R-4

B C D E F G H I J K L M N O P

Q R

Choshui River Choshui River Puzih River Bajhang River Bajhang River Bajhang River Puzih River Bajhang River Bajhang River Bajhang River Zengwen River Bajhang River Kaupin River Kaupin River Kaupin River Kaupin River

Kaupin River Kaupin River

8

–

8

–

Salmonella identification by serological test and PFGE

S. enterica Newport

S. enterica Victoria

4439

were also positive for Salmonella detection by Process II. This indicates that Process II is more sensitive than Process III in detecting Salmonella in water samples. According the above results, we developed a new procedure for detecting Salmonella from environmental water samples that is based on concentration by membrane filtration followed by pre-enrichment and selective-enrichment. The DNA in the selective-enrichment broth is then extracted and tested by PCR amplification. Next, the positive selective-enrichment broth is spread on selective culture plates to isolate presumptive Salmonella. Then, we used the PCR method and biochemical testing (the eight-tube diagnostic kit and the API 20E system) as a confirmation. Finally, we used serological assay to differentiate the serotypes of Salmonella or used PFGE to determine the genotypes of Salmonella. 3.2. Distribution of Salmonella occurrence in the drinking water sources of southern Taiwan

S. enterica Enteritidis

S. enterica Albany

insoluble debris (Kreader, 1996; Wilson, 1997; Rossen et al., 1992). Of the 116 water samples tested by Process I combined with PCR analysis, none was positive for the invA gene. This indicates that the PCR inhibitors in the water concentrate were not eliminated effectively. The other explanation is that the detection limit was too high to detect the presence of Salmonella in the concentrated environmental water without any enrichment. In the study of Chiu and Ou (1996), the detection limit was 200 cells by PCR reaction. In Process II, the water was filtered, eluted, concentrated, enriched, and then detected Salmonella by the PCR method. We expected this process to decrease the PCR inhibitory substances and increase the sensitivity compared to the Process I. In the present study, 10 of the 116 samples were positive for the invA gene of Salmonella when analyzed with Process II. This indicates that Process II is more sensitive than direct concentration (Process I) for the detection of Salmonella in water. In Process III, the concentrate of each sample was pre-enriched by BPW and selective-enriched by RV broth before the broth was spread on selective medium plates. Finally, the candidate colonies were used in PCR analysis. Even though Process III is more complicated than the other two, especially for detecting Salmonella in a large number of water samples, only the Salmonella strains separated from Process III can be identified by biochemical testing, serological assay, and PFGE. Therefore, Process III is a good way to identify Salmonella serovars. A total of seven isolates were screened from seven different samples in Process III (Table 2). All the samples that were positive by Process III

Table 2 shows the results of monitoring Salmonella in the drinking water sources of southern Taiwan. The combined results of three different analysis methods reveal that Salmonella was detected in 10 of 116 different water samples (8.6%). Salmonella was detected in 7 of the 18 water plants. Plant P exhibited the most frequent occurrence of Salmonella (3/4), followed by plant R (2/8). The other 5 water plants had one occurrence each. The water sources of the 18 water plants belonged to 5 watersheds. In addition to the Choshui River, Salmonella was detected in 4 other watersheds, and frequently reported in the watershed of Kaupin River. The Kaupin River is the major raw water source for the greater Kaohsiung area, which has a population of 2.5 million. Compared with the other 4 watersheds, this watershed is heavily polluted by domestic sewage and farm wastewater. Therefore, the local water plants have employed advanced treatment processes to remove contaminants. The survey of Salmonella in the study accurately reflects the situation of drinking water sources. This study also includes a statistical test (Table 3) comparing the independent means of the values of water turbidity, heterotrophic plate counts (HPC), and total coliforms to reveal any significant differences between the samples with and without Salmonella. The Salmonellapositive samples showed higher turbidity and HPC. Significant differences (Mann–Whitney U test, P b 0.05) were observed between the presence/absence of Salmonella and water turbidity, and between the presence/absence of Salmonella and HPC. However, large standard deviations were simultaneously observed in Table 3. Although correlations between the occurrences of some pathogens and total coliforms were found previously (Rompré et al., 2002), the results of this study suggest that these correlations may not be universally applicable. The primary source of drinking water in southern Taiwan is surface water. Therefore, water quality can be greatly affected by human and animal excreta from inadequate sewage systems in the upper regions of watersheds. Salmonella species are excreted enteric pathogens which can survive for a long time in natural waters (Wright, 1989). The results of this study indicate that the spread of Salmonella through the drinking water supply is highly probable if the source water contamination and inadequate water treatment were to happen simultaneously. The majority of salmonellosis outbreaks have been linked to the

Table 3 Mean ± SD and nonparametric differences (Mann–Whitney U test, P levels) in the presence/absence of Salmonella in terms of water quality parameters. Water quality parameter

Mean ± SD Salmonella-presence

Salmonella-absence

Turbidity (NTU) 139.64 ± 84.63⁎ 118.18 ± 214.95⁎ Heterotrophic bacteria (CFU/ml) 3.98 × 104 ± 5.42 × 104⁎ 1.22 × 104 ± 3.81 × 104⁎ Total coliforms (CFU/100 ml) 1.17 × 103 ± 1.55 × 103 6.79 × 102 ± 1.77 × 103 ⁎ P b 0.05.

4440

B.-M. Hsu et al. / Science of the Total Environment 409 (2011) 4435–4441

consumption of contaminated foods. However, the ubiquitous nature of Salmonella and its widespread occurrence in the aquatic environment has been reported in several countries (Baudart et al., 2000; Kramer et al., 1996; Polo et al., 1998; Touron et al., 2005). This suggests that environmental water monitoring for the presence of Salmonellae and its possible spread through drinking water supplies may provide new insights regarding the epidemiology of Salmonella infection. 3.3. Identification of Salmonella using biochemical testing and serological assays Of the seven isolates from Process III containing the invA gene sequences, all were identified as Salmonella spp. by the eight-tube diagnostic kit. When these seven isolates were further checked using the API 20E system, all of the isolates were again positively identified as Salmonella spp. When the seven isolates were further analyzed by serological assay, they were identified as different serotypes of Salmonella (Table 2). All the samples that were positive by Process III were also positive for Salmonella detection by Process II. This indicates that Process II is more sensitive than Process III in detecting Salmonella in water samples. However, the Salmonella detected in Process II could not be serotyped. This shows that there is a large diversity of Salmonella serotypes within drinking water sources. Four of the seven Salmonella serovars identified in the study belonged to the ten most common Salmonella serovars reported in diarrhea patients from 2004–2008, according to the Taiwan CDC (http://www.nhri.org.tw/NHRI_ADM/userfiles/file/ID/Molecular% 20epidemiology%20and%20disease%20surveillance%20of%20clinical% 20Salmonella%20in%20Taiwan.pdf). These included Salmonella enterica serovar Enteritidis (Top 1), Newport (Top 4), Alnaby (Top 5) and Paratyphi B var. Java (Top 7). Although the Salmonella enterica serovar Houten, Montevideo, and Victoria are not frequently reported in Taiwan, they have been isolated from the clinical cases in southern Taiwan. Salmonella enterica serovar Typhimurium, the second most frequently reported Salmonella serotype in Taiwan and the most frequently reported in western countries from the cases of salmonellosis, was not detected in this study. Similar results have been recorded in Salmonella research on aquatic environments (Haley et al., 2009). This could reflect the lower survival rates of Salmonella enterica serovar Typhimurium in natural environments. This study found 7 different Salmonella serovars, which were probably mobilized from widespread animal-rearing activities. Therefore, the monitoring of Salmonella serotypes in aquatic environments may be a good indicator of specific contamination sources from animal excrements. 3.4. Genotyping of Salmonella by PFGE In this study, the seven Salmonella isolates identified by biochemical testing and serological assay were subjected to PFGE. The genotypes of these seven isolates were differentiated by their PFGE patterns, and the results agree with the serological assay identification. In addition to the PFGE pattern of Salmonella enterica serovar Montevideo from drinking water source matched with one clinical isolate, the other patterns of the genotypes from drinking water source and specimens of diarrhea patients varied slightly, and were identified as different subtypes. This indicates that the most of reported infection sources of Salmonellosis in southern Taiwan seems not include drinking water. However, the existence of Salmonella in drinking water sources should be considered a potential public health threat. Overall, PFGE was shown to be very useful in delineating the genetic variability of Salmonella strains, and was an valuable epidemiological tool for investigating Salmonella. 4. Conclusions (1) When monitoring for the presence or absence of Salmonella in water samples, the procedures of pre-enrichment of water

concentrate, selective-enrichment by RV broth, DNA extraction from RV broth, and PCR analysis are recommended. (2) The identified Salmonella serovars include Alnaby, Enteritidis, Houten, Montevideo, Newport, Paratyphi B var. Java, and Victoria. (3) Significant differences were observed between the Salmonellapositive and negative samples for turbidity and HPC. (4) Because water in the studied regions is used for drinking, the existence of Salmonella should be considered a potential public health threat. Acknowledgments This work was supported by a research grant from National Science Council of Taiwan, ROC (NSC97-2221-E-194-006-MY3). We are grateful the editing of Wallace Academic Editing in Taiwan and grateful the experimental work by Dr. Chien-Shun Chiou and Mrs. Yo-Wen Wang from the third branch office, Centers for Disease Control, Taiwan, ROC. References APHA Standard Method for the Examination of Water and Wastewater. 15th ed. APHA, WEF and AWWA, Washington, DC; 1995. Baudart J, Lemarchand K, Brisabois A, Lebaron P. Diversity of Salmonella strains isolated from the aquatic environment as determined by serotyping and amplification of the ribosomal DNA spacer regions. Appl Environ Microbiol 2000;66:1544–52. Bell C, Kyriakides A. Salmonella. A practical approach to the organism and its control in foods. Practical Food Microbiology Series. Blackwell Science Ltd., Oxford; 2002. p.1–25. Bhan MK, Bahl R, Bhatnagar S. Typhoid and paratyphoid fever. Lancet 2005;366: 749–62. Boyd EF, Wang FS, Whittam TS, Selander RK. Molecular genetic relationships of the Salmonellae. Appl Environ Microbiol 1996;62:804–8. Chiou CS. Experiences on PFGE manipulation and current status of lab-based molecular subtyping in Taiwan. In the proceeding of 8th Annual PulseNet Update Meeting, San Diego, CA; 2004. Chiou CS, Huang JF, Tsai LH, Hsu KM, Liao CS, Chang HL. A simple and low-cost paperbridged method for Salmonella phase reversal. Diagn Microbiol Infect Dis 2006;54: 315–7. Chiu CH, Ou JT. Rapid identification of Salmonella serovars in feces by specific detection of virulence genes, invA and spvC, by an enrichment broth culture-multiplex PCR combination assay. J Clin Microbiol 1996;34:2619–22. Covert TC. Salmonella. In Waterborne Pathogens, AWWA Manual M48, American Water Works Association, Denver, CO; 1999. p. 107–10. D'Aoust JY. Salmonella. In: Doyle MP, editor. Foodborne Bacterial Pathogens. New York: Marcel Dekker; 1989. p. 327–445. Daum LT, Barnes WJ, McAvin JC, Neidert MS, Cooper LA, Huff WB, et al. Real-time PCR detection of Salmonella in suspect foods from a gastroenteritis outbreak in Kerr county, Texas. J Clin Microbiol 2002;40(8):3050–2. Draft ISO 19250. Water quality — Determination of Salmonella species (Revision of ISO 6340:1995). International Standards Organization; 2003. Fung DYC. Predictions for rapid methods and automation in food microbiology. J AOAC Int 2002;85:1000–2. Furtado C, Adak GK, Stuart JM, Wall PG, Evans HS, Casemore DP. Outbreaks of waterborne infectious intestinal disease in England and Wales 1992–1995. Epidemiol Infect 1998;121:109–19. Garcia del Portillo JA. Molecular and cellular biology of Salmonella pathogenesis. In: Cary JW, Linz JE, Bhatnagar D, editors. Microbial Foodborne Diseases. Mechanisms of Pathogenesis and Toxin Synthesis. Lancaster, PA: Technomic Publishing; 2000. p. 3–86. Giovannacci I, Queguiner S, Ragimbeau C, Salvat G, Vendeuvre JL, Carlier V, et al. Tracing of Salmonella spp. in two pork slaughter and cutting plants using serotyping and macrorestriction genotyping. J Appl Microbiol 2001;90:131–47. Haley BJ, Cole DJ, Lipp EK. Distribution, diversity, and seasonality of waterborne Salmonellae in a rural watershed. Appl Environ Microbiol 2009;75:1248–55. Hardnett FP, Hoekstra RM, Kennedy M, Charles L, Angulo FJ. Epidemiologic issues in study design and data analysis related to FoodNet activities. Clin Infect Dis 2004;38 (Suppl. 3):S121–6. Hohmann EL. Nontyphoidal salmonellosis. Clin Infect Dis 2001;32:263–9. Hsu BM, Chen CH, Wan MT, Cheng HW. Legionella prevalence in hot spring recreation areas of Taiwan. Water Res 2006;40(17):3267–73. Hsu BM, Huang YL, Hsu YF, Shih FC. Occurrence of protozoan and bacterial pathogens in water samples from small water systems in mountainous areas of Taiwan. J Environ Eng-ASCE 2008;134(5):389–94. Kilger G, Grimont PAD. Differentiation of Salmonella phase 1 flagellar antigen types by restriction of the amplified fliC gene. J Clin Microbiol 1993;31:1108–10. Kramer M, Herwaldt B, Craun G, Calderon R, Juranek D. Surveillance for waterbornedisease outbreaks — United States, 1993–1994. MMWR Surveill Summ 1996;12:1–33. Kreader CA. Relief on amplification inhibition in PCR with bovine serum albumin or T4 gene 32 protein. Appl Environ Microbiol 1996;62:1102–6.

B.-M. Hsu et al. / Science of the Total Environment 409 (2011) 4435–4441 Kuo F, Carey JB, Ricke SC. UV Irradiation of Shell Eggs: Effect on populations of aerobes, molds and inoculated Salmonella Typhimurium. J Food Prot 1997;60(6):639–43. Nair S, Lin TK, Pang T, Altwegg M. Characterization of Salmonella serovars by PCRsingle-strand conformation polymorphism analysis. J Clin Microbiol 2002;40: 2346–51. Percival S, Chalmers R, Embrey M, Hunter P, Sellwood J, Wyn-Jones P. Microbiology of waterborne diseases. In: Salmonella. Elsevier Academic Press, Amsterdam; 2004. p. 173–83. Polo F, Figueras M, Inza I, Sala J, Fleisher J, Guarro J. Relationship between presence of Salmonella and indicators of faecal pollution in aquatic habitats. FEMS Microbiol Lett 1998;160:253–6. Popoff MY, Bockemuhl J, Gheesling LL. Supplement 2001 (no. 45) to the Kauffmann– White scheme. Res Microbiol 2003;154:173–4. Porwollik S, Boyd EF, Choy C, Cheng P, Florea L, Proctor E, et al. Characterization of Salmonella enterica subspecies I genovars by use of microarrays. J Bacteriol 2004;186:5883–98. Refum T, Heir E, Kapperud G, Vardund T, Holstad G. Molecular epidemiology of Salmonella enterica serovar Typhimurium isolates determined by pulsed-field gel electrophoresis: comparison of isolates from avian wildlife, domestic animals, and the environment in Norway. Appl Environ Microbiol 2002;68:5600–6. Ribot EM, Fair MA, Gautom R, Cameron DN, Hunter SB, Swaminathan B, et al. Standardization of pulsed-field gel electrophoresis protocols for the subtyping of

4441

Escherichia coli O157:H7, Salmonella, and Shigella for PulseNet. Foodborne Pathog Dis 2006;3(1):59–67. Rompré A, Servais P, Baudart J, de-Roubin MR, Laurent P. Detection and enumeration of coliforms in drinking water: current methods and emerging approaches. J Microbiol Methods 2002;49:31–54. Rossen L, Norskov P, Holmstrom K, Rasmussen OF. Inhibition of PCR by components of food samples, microbial diagnostic assays and DNA-extraction solutions. Int J Food Microbiol 1992;17(1):37–45. Soumet C, Ermel G, Rose N, Rose V, Drouin P, Salvat G, et al. Evaluation of a multiplex assay for simultaneous identification of Salmonella sp., Salmonella Enteritidis, Salmonella Typhimurium from environmental swabs of poultry houses. Lett Appl Microbiol 1999;28:113–7. Touron A, Berthe T, Pawlak B, Petit F. Detection of Salmonella in environmental water and sediment by a nested-multiplex polymerase chain reaction assay. Res Microbiol 2005;156:541–53. Wilson IG. Inhibition and facilitation of nucleic acid amplification. Appl Environ Microbiol 1997;63:3741–51. Wright R. The survival patterns of selected faecal bacteria in tropical fresh waters. Epidemiol Infect 1989;103:603–11.