DSpace at VNU: Prevalence ofEscherichia coliin surface waters of Southeast Asian cities

DSpace at VNU: Prevalence ofEscherichia coliin surface waters of Southeast Asian cities tài liệu, giáo án, bài giảng , l...

O R I G I N A L P A P E R

Prevalence of Escherichia coli in surface waters

of Southeast Asian cities

Kenneth Widmer•Nguyen Thi Van Ha•Soydoa Vinitnantharat•

Suthipong Sthiannopkao•Setiawan Wangsaatmaja•Maria Angela Novi Prasetiati•

Nguyen Cong Thanh•Kasame Thepnoo•Arief Dhany Sutadian•

Huynh Thi Thanh Thao• Deby Fapyane•Vibol San•Pierangeli Vital•

Hor-Gil Hur

Received: 21 September 2012 / Accepted: 10 May 2013

Ó Springer Science+Business Media Dordrecht 2013

Abstract Surface water samples were collected from

rivers which fed into large urban areas within Vietnam,

Indonesia, Cambodia, and Thailand and were processed to

enumerate Escherichia coli Selected isolates were further

characterized using PCR to detect the presence of specific

virulence genes Analyzing the four countries together, the

approximate mean cfu/100 ml for E coli counts in the dry

season were log 4.3, while counts in the wet season were

log 2.8 Of the 564 E coli isolates screened for the

pres-ence of pathogenic genes, 3.9 % possessed at least one

virulence gene The most common pathogenic types found

were Shiga toxin-producing E coli isolates These results

reinforce the importance of monitoring urban surface

waters for fecal contamination, that E coli in these water

environments may serve as opportunistic pathogens, and

may help in determining the impact water usage from these

rivers have on the public health of urban populations in Southeast Asia

Keywords Escherichia coli Pathogens  Surface water  Urban water quality  Southeast Asia  PCR

Introduction Surface water not only serves as drinking water sources for metropolitan areas in Southeast Asia, they also serve to facilitate a great deal of economic activity and play an essential part in regional agriculture Because of the important role surface waters play in the economic and sustainable development of Southeast Asian countries, ensuring sufficient water quality in these water sources is

K Widmer ( &)  D Fapyane

International Environmental Analysis and Education Center,

Gwangju Institute of Science and Technology,

123 Cheomdan-Gwagiro, Bukgu, Gwangju 500-712,

Republic of Korea

e-mail: kwidmer@gist.ac.kr

N T Van Ha

Ministry of Natural Resources and Environment,

Ho Chi Minh City University for Natural Resources and

Environment, 236B Le Van Sy Street, Ward 1, Tan Binh

District, Ho Chi Minh City, Vietnam

S Vinitnantharat

Division of Environmental Technology, School of Energy,

Environment and Materials, King Mongkut’s University

of Technology Thonburi (KMUTT), 126 Prachauthit Road,

Thungkru, Bangkok, Thailand

S Sthiannopkao

Department of Environmental Engineering, College

of Engineering, Dong-A University, Pusan, Republic of Korea

S Wangsaatmaja  M A N Prasetiati  A D Sutadian West Java Environmental Protection Agency, Indonesia,

Jl, Naripan No 25, Bandung 40111, West Java, Indonesia

N C Thanh International Center for Education Development T.H.T,

15 Thien Y St., Ward 4, Dalat City, Vietnam

K Thepnoo Department of Drainage and Sewerage, Bangkok Metropolitan Administration, 123 Mitmaitri Road, Dindaeng District, Bangkok, Thailand

H T T Thao Faculty of Environment, Ho Chi Minh City University

of Technology, 268 Ly Thuong Kiet, District 10,

Ho Chi Minh City, Vietnam DOI 10.1007/s11274-013-1376-3

essential for many urban cities Fecal contamination of

surface waters is an issue under greater scrutiny from

regulatory agencies in both developed and developing

countries (US-EPA2002; Santo Domingo et al.2007) The

impact of poor surface water quality on public health is

compounded as contamination of irrigation water may also

introduce pathogens to the food supply through

contami-nated agricultural products, as exemplified by outbreaks of

pathogenic Escherichia coli in fresh produce in developed

countries (Pakalniskiene et al 2009; CDC 2006; Ackers

et al.1998; Wheeler et al.2005) As such, the monitoring

of fecal contamination in water sources is a key issue in

evaluating urban water quality and understanding the

potential risk to urban populations

Escherichia coli is a commensal organism in many

mammals and has a broad range of hosts Although this

organism is part of the normal gut flora of many mammals,

it can act as opportunistic pathogens and may serve as an

agent for enteric disease in humans (Nataro and Kaper

1998; Hunter 2003) Given these characteristics, fecal

contamination of surface waters is a common route of

spreading E coli within the environment (Field and

Sa-madpour2007; Simpson et al.2002) and as this bacterium

can persist in the environment (Byappanahalli and Fujioka

1998; Solo-Gabriele et al 2000), fecal contamination of

surface waters is of some interest for gauging overall water

quality and its potential impact on public health

Addi-tionally, previous studies have demonstrated that bacterial

pathogens may be found in surface waters (Ha et al.2008;

Kobayashi et al 2003; Phan et al 2003) and well water

(Vollaard et al.2005) within Southeast Asian countries

Of particular interest to public health are pathogenic

E coli, of which several types can induce diarrheagenic

infections in humans, with some capable of causing more

serious infections such as hemorrhagic colitis (Ohno et al

1997; Nataro and Kaper1998) Determining potential risk

to public health is problematic, however quantitative

microbial risk assessment is a tool that policy makers can

utilize to better manage fecal contaminated waters and

predict the potential impacts on populations (Haas et al

1999) Some approaches have utilized models based on reported cases of E coli infections in relation to relative levels of fecal contaminants in surface waters (Soller et al

2010) Having access to empirical count data, combined with surveys of pathogenic types from these samples, may provide a more substantive base of information to create better predictive models This information may also be critical in establishing proper water quality indices, to better predict the impact human pathogens have on water usage Additionally if these waters are utilized for agri-cultural use with limited treatment, they may have a greater impact on public health than expected due to potential outbreaks of food-borne diseases

The focus of this study was to enumerate E coli in urban surface waters within Southeast Asian countries, and further characterize isolates as either, enteroinvasive E coli (EIEC), Shiga toxin-producing E coli (STEC), enterotoxigenic

E coli (ETEC), enteropahogenic E coli (EPEC), or entero-hemorrhagic E coli (EHEC) (Yatsuyanagi et al.2003; Nat-aro et al.1987; Jerse and Kaper1991; Sears and Kaper1996) Further, enumeration data would be compared to determine

if seasonal or urban land-use would have an impact on rel-ative numbers of E coli in these surface waters

Materials and methods Bacterial cultures Both cultures of E coli NCCP 10004 (ETEC) and E coli NCCP 13719 (EIEC) were obtained from commercial stocks supplied by the National Culture Collection for Pathogens, Korea A cultural isolate of E coli O157:H7 was obtained as generous donation by the Korean Centers for Disease Control and Prevention which served as a control for EPEC, STEC, and EHEC as it contained carried all target virulence genes for these pathogenic types Sampling sites, land use classification, and sample collection

Surface water samples were collected in four main rivers that flow through different major Southeast Asian cities Urban metropolitan sampling sites were selected based primarily on their proximity to high density population areas or close association with industrial activity As comparative sampling sites, more rural locations where surface waters were used for irrigation, aquaculture, or were fed agricultural runoff were also sampled The Cita-rum River in Bandung (West Java, Indonesia), the lower Chao Phraya in Bangkok (Thailand), the Saigon river in Ho Chi Minh City (Vietnam), and the Tonle Sap-Bassac in

V San

Department of Environmental Science, Royal University

of Phnom Penh, Russian Federation Boulevard, Toul Kork,

Phnom Penh 12157, Cambodia

P Vital

Natural Sciences Research Institute,

University of the Philippines Diliman,

1101 Quezon City, Philippines

H.-G Hur

School of Environmental Science and Engineering,

Gwangju Institute of Science and Technology,

123 Cheomdan-Gwagiro, Bukgu, Gwangju 500-712,

Republic of Korea

Phnom Penh (Cambodia) were selected due to these rivers

being within respective city boundaries and having a close

association with areas of high metropolitan populations

(over 5 million, except for Phnom Penh which has a

pop-ulation of approximately 1.3 million) (Fig.1)

Land use types for potential collection sites were

char-acterized by runoff sources within 200 m of each location

Sites were either characterized as: (1) agricultural/rural,

indicating sparsely populated areas, or those with heavy

agricultural activity, (2) urban, indicting urban developed

locations with relatively high populations and domiciles

within the metropolitan area, (3) industrial, considered

sampling locations with sparse domestic populations and

greater intensity of mining or manufacturing activities which

produced substantial water runoff during operation, (4)

mixed, being a relatively proportional combination of at least

two of other land use types, and finally (5) water treatment

sites, which were locations where both the influent and

effluent from water treatment plants were sampled

Addi-tionally these treatment sites received minimal runoff from

sources of urban populations and agricultural operations and

were considered ideal locations for establishing a baseline of

treated surface waters within each region

Approximately 10–20 sites were sampled over a 2 month

period for both the dry and wet seasons of each respective

country in 2010 An additional 2 month seasonal sampling

event was conducted for Thailand during the dry season in

2011 for a total of 157 samples from all four countries Both

Vietnam and Thailand had up to two sampling events for

each location over the 2 month periods during each season,

while Indonesia and Cambodia had a single sampling event

during both the dry and wet seasons Similar sampling

locations (within 20 m) were maintained as sample points

throughout the study over the different seasons

Approximately 100 ml grab samples of surface water

were collected into sterilized polypropylene bottles at

30 cm depths from the center of the river channels either

by boat or from bridge structures Samples were taken

below the water surface to minimize floating debris and a

head space of roughly 2 cm was maintained in each sample

bottle During sample collection, data for basic water

quality based on physical (temperature and turbidity) and

chemical characteristics (pH and TDS) were collected on

site Samples were transported in an improvised ice box

(kept under 10°C), and processed within 6–8 h of

collec-tion in a local laboratory for each respective country

Water sample processing and E coli isolation

and enumeration

Water samples were sequentially filtered through sterile,

0.45 lm, 47 mm filters (Pall Korea Ltd., Seoul, Korea) in

10, 1, and 0.1 ml volumes If the sample volume was under

10 ml, it was mixed with 10 ml sterile DI water to ensure

an even sample distribution over the filter surface Filters were ascetically transferred to modified membrane Ther-motolerant E coli (mTEC) agar plates (BD Scientific, Maryland, USA) and were initially incubated at 35°C for

2 h, and further incubated at 44.5°C for 22–24 h (Yan

et al.2007; Unno et al.2009) Colony counts were recorded and adjusted to per 100 ml based on the volume sampled

Up to 10 atypical colonies (red to magenta) were trans-ferred to plates of MacConkey medium with lactose (BD Scientific, Maryland, USA), incubated at 35 ± 0.5 °C for

24 h, and presumptive identification of E coli isolates were determined by the observation of light pink to red colonies (Byappanahalli et al.2007) E coli isolates obtained from Thailand and Vietnam were then transferred to tryptic soy agar (TSA) slants and maintained at 4°C until shipping and further processing in Korea These countries were selected to ship isolates due to better local facilities for long term storage and access to transportation resources to facilitate rapid shipping of isolates

Shipped TSA slants from Vietnam and Thailand were further processed in Korea and E coli isolates were con-firmed by streaking onto Eosin Methylene Blue (EMB) agar plates (Lab M Limited, Lancashire, UK) which were incubated at 35°C for 24 h Plates which demonstrated typical colony morphology for E coli (blue to black col-onies with a metallic green sheen) were transferred into 0.1 ml Luria–Bertani freezing medium (Zimmer and Verrinder Gibbins 1997) and incubated with moderate shaking for 24 h at 35°C After sufficient incubation, isolates were then maintained at -70 °C

E coli DNA extraction and PCR Escherichia coli cultures in Luria Betaini freezing medium were thawed, with a 50 ll aliquot removed and mixed with

50 ll 0.05 M NaOH, and then finally boiled at 95°C for

15 min This resulting lysate was then used directly as template for PCR (Unno et al 2009)

PCR reactions were run as multiplex or single reactions

To reduce variation with the different PCR reactions, a commercial master pre- mix was used (AccuPower HF PCR Mix, Bioneer, Daejeon, Korea) The primers for each reaction are provided in Table1 Each reaction was pre-pared using 2 ll template and primers at concentrations of 0.5 lM for AL65/125 (ETEC), 0.25 lM for primer sets

LTL/R(ETEC) and ipa III/IV (EIEC) (Toma et al 2003), and 0.25 lM of primer sets stx1F/R, stx2F/R (STEC), ea-eAF/R (EPEC), and hlyAF/R (EHEC) (Paton and Paton

1998), with DI water added to bring each reaction volume

up to 20 ll In addition to the samples, non-template controls (DI water) and 1 ll template DNA extracted from the control strains were used as negative and positive

Fig 1 Sampling sites for the Saigon River, Vietnam (a), Tonle Sap-Bassac Rivers, Cambodia (b), Lower Chao Phraya River, Thailand (c), and the Citarum River, Indonesia (d)

controls, respectively Each reaction was run under

previ-ously published conditions (Toma et al 2003; Paton

and Paton1998) The exception being for the primer sets

AL65/125 where an annealing temperature of 58°C was

used, as it was determined that this higher annealing

tem-perature produced an improved yield of product using

template from the control strain (data not shown)

PCR products were analyzed by electrophoresis on 2 %

agarose gels stained with ethidium bromide Digital images

were obtained after UV transillumination and the products

were compared to both positive control strains and a

commercial molecular weight standard Amplicons of the

appropriate size were scored as positive identification of

the respective pathogenic E coli gene, and extracted DNA

from E coli isolate samples which demonstrated amplicons

indicative of pathogenic genes were subjected to PCR

analysis a second time to confirm the initial results E coli

isolates were pathogen-typed based on the profile of

amplicons, where the presence of est and/or elt were

considered ETEC types, ipaH was considered EIEC types, the sole presence of stx1 and/or stx2 considered STEC types, hlyA being deemed EHEC types, and eaeA being classified as EPEC types Selected isolates that demon-strated expected sized fragments for virulence genes were cultured again onto EMB agar plates Genomic DNA was extracted from the cultured plates by transferring an iso-lated colony into 100 ll TE buffer and boiling the cell suspension at 95°C for 15 min This resulting lysate was then used directly as template for PCR as previously described, except that only a single set of primers were used to amplify expected PCR products (where multiple single primer reactions were run in place of multiplex PCR reactions) Amplicons were purified using a commercial kit (PCR Purification Kit, ELPIS-Biotech, Korea) and result-ing purified DNA samples were sequenced by a commer-cial analysis service that employed an Applied Biosystems 3730xl DNA Analyzer instrument Further confirmation of these PCR sequences were determined by BLAST analysis Fig 1 continued

Data analysis

Means of log E coli counts per 100 ml were analyzed to

determine the normality of their distributions Based on

Kolmogorov–Smirnov tests, it was deemed that

non-para-metric statistical tests (Wilcoxon ranked sum) would be

more suitable to interpret differences in the observed

means Surface waters associated with particular land

use categories were compared through analysis of mean

cfu/100 ml values for the urban, mixed land use, and

industrial sites against the agricultural/rural sampling sites

Statistical analysis was conducted using a commercially

available program (SPSS 14.0, SPSS inc., Chicago, USA)

Results

E coli counts

Mean cfu/100 ml E coli counts based on seasonal data are

summarized in Table2 Sixty eight and 89 samples were

collected and processed for the wet and dry seasons, respec-tively, for all four countries, with an overall mean log 3.61 cfu/100 ml (±0.14 s.e.) for all 157 water samples The mean dry season counts were log 4.27 cfu/100 ml, roughly log 1.5 higher than the mean wet season counts of log 2.76 cfu/100 ml for all the land use types (including 12 samples from the treatment sites) Overall means of cfu/100 ml E coli counts ranged from log 2.66–4.58 for individual rivers, with both the Citarum (log 4.58) and Lower Chao Phraya (log 3.94) being approximately 1 log or greater than the overall means of the Tonle Sap-Bassac (log 3.05) and Saigon rivers (log 2.86) Seasonal variations in the sites were not apparently different, except for the Lower Chao Phraya which had an almost 3 log increase from average counts in the wet season (Table2) Physical and chemical characteristics also varied for many

of the rivers, especially TDS readings which had an average over 2,800 mg/l in the Lower Chao Phraya far exceeding the other averages within the other sampled rivers The Tonle Sap-Bassac Rivers had the highest mean turbidity of 106 NTUs, almost double that of the Saigon River and nearly four times that of the Lower Chao Phraya River (Table2)

Table 1 PCR primer sets

employed in this study

Sequence displayed as 50–30

a Toma et al ( 2003 )

b Paton and Paton ( 1998 )

size (bp)

Table 2 E coli counts based on season and general water quality characteristics (mean ± standard error)

Total Wet seasona Dry seasona Temp (°C) Turbidity (NTU) pH TDS Tonle Sap-Bassac (22) 3.05 ± 0.07 3.05 ± 0.07 (11) 3.04 ± 0.12 (11) 28.9 ± 0.31 106.5 ± 18.2 7.2 51.2 ± 3.02 Citarum (20) 4.58 ± 0.05 4.49 ± 0.07 (10) 4.62 ± 0.05 (10) 25.1 ± 0.25 NA 7.2 144.5 ± 7.08 Lower Chao Phraya (74) 3.94 ± 0.24 1.95 ± 0.43 (27) 5.08 ± 0.11 (47) 29.9 ± 0.4 26.7 ± 4.0 7.2 2877.7 ± 924.25 Saigon (41) 2.86 ± 0.2 2.83 ± 0.29 (20) 2.9 ± 0.27 (21) 29.8 ± 0.33 47.7 ± 10.4 6.4 2.5 ± 0.76 General water quality characteristics measured at time of sample collection

Numbers in parenthesis are total number of samples collected

NA not analyzed

a Overall mean cfu/100 ml were significantly higher (p = 0.001) in the dry season [log 4.27 ± 0.14 (s.e.)] compared to the wet season [log 2.76 ± 0.22 (s.e.)]

To determine if there were seasonal differences for the

sampling sites across all rivers, further statistical analysis

was conducted However due to the relative low numbers at

the water treatment sites compared to other land-use types

which would disproportionately skew data with excessive

variance, these water treatment sites were removed from

the seasonal data sets prior to analysis When comparing

the means for all the rivers combined based on season (with

water treatment site removal), the observed difference for

combined means of the land use sites was found to be

statistically significant (p = 0.001) (Table2)

Additional comparisons were made for mean E coli

counts based on land use classifications, to discern if river

surface waters within urban areas and runoff associated

with urban activity were significantly different from rivers

impacted by agriculture or more-rural land use activities

The overall mean cfu/100 ml values for the land use types

ranged from log 1.7 (water treatment sites) to log 4.1

(urban sites), with values for agricultural/rural, mixed land

use, and industrial sites being log 3.2, 3.9, and 3.8,

respectively Analysis based on land use types was made

comparing the agricultural/rural sample sites (41 samples)

individually to the urban (77 samples), industrial (21

samples), and mixed land use sites (6 samples) based on the

yearly collected data Comparisons to the water treatment

sites were omitted due to the low observed counts at these

sites and limited collected samples (12 samples) There

were statistically significant differences observed

compar-ing the agricultural/rural land use types to both urban sites

(p = 0.001) and industrial sites (p = 0.022), while such

statistical differences were not observed when comparing

mean E coli counts of agricultural/rural sites to that of

mixed land use locations (Table3)

Observed pathogenic E coli types

Of 564 isolates processed, 22 (3.9 %) were observed to

have virulence genes initially determined by the presence

of PCR products and further confirmed by sequencing and

BLAST analysis All isolates presented in this study which possessed similar pathogen-type profiles, were either from different locations/rivers, or collected on different sam-pling dates (a different samsam-pling month or season) for the

157 water samples collected and processed For some water samples, multiple colonies from the same sample were isolated and processed (up to 10) It is possible that some isolates which had similar pathogen profiles (based on the presence of similar PCR products) could be clones While studies have demonstrated that E coli diversity in river systems can fluctuate monthly even if collected from the same location (Jang et al 2011), to reduce such potential redundancy, additional isolates from the same sampling location and collection dates were removed from the data set if they presented similar virulence gene profiles Less than seven isolates were initially removed from the data set, due to having similar virulence gene profiles and because they derived from water samples that were col-lected during a single sampling event

The most common pathogenic E coli isolates recovered were STEC (n = 9) and the second most common patho-genic E coli isolates observed were EPEC strains (n = 7) ETEC type isolates were also observed (n = 6) It is interesting to note that three EPEC strains possessed both eaeA and stx1 As intimin is considered a key virulence factor for enteropathogenic E coli, these isolates were considered EPEC, however this gene can also be present in Shiga toxin-producing E coli (Aidar-Ugrinovich et al

2007) (Table4)

Interestingly, the predominant Shiga-toxin gene found was stx1 (12 isolates), and 6 isolates possessed elt The intimin factor gene, eaeA, was also relatively common being found in 7 of the 22 isolates No other isolates har-bored hlyA, invA, or est genes which could be confirmed through amplification and/or sequencing of PCR products Pathogen-type E coli were observed in both agricultural/ rural surface waters (11 isolates) and also from urban land use types (12 isolates) in roughly the same proportion (Table4) Additionally, isolates which had virulence genes

Table 3 Mean E coli counts based on land use and season

(mean ± standard error)

Land use type Overall Dry season Wet season

Agricultural/rural (41) 3.15 ± 0.21a 3.63 ± 0.22 2.59 ± 0.34

Urban (77) 4.1 ± 0.19a 4.77 ± 0.16 3.11 ± 0.33

Industrial (21) 3.79 ± 0.39a 4.62 ± 0.27 2.7 ± 0.7

Mixed use (6) 3.86 ± 0.32 3.96 ± 0.5 3.76 ± 0.5

Treatment (12) 1.67 ± 0.52 2.26 ± 0.76 1.07 ± 0.68

Counts expressed as log CFU/100 ml

Number in parenthesis is total number of location land types

a Mean log cfu/100 ml for agricultural/rural sites significantly lower

when compared to urban (p = 0.001) or industrial sites (p = 0.022)

Table 4 Observed E coli virulence genes and pathogen types based

on land use Land use type Virulence genes E coli pathogen types

elt eaeA stx1 ETEC EPECa STEC

Combined land use types total

Isolates were obtained from either Vietnam (12) or Thailand (10) Number in parenthesis is total number of location land types

a Isolates positive for both eaeA and stx1were considered as EPEC

were recovered from both river systems within Thailand

and Vietnam in similar numbers, totaling 10 and 11

iso-lates, respectively

Discussion

For all countries, overall means for E coli counts exceeded

a proposed US-EPA coastal and recreational waters

stan-dard threshold value of log 2.61 cfu/100 ml (US-EPA

2012) although Vietnam was very close to remaining within

this proposed limit with an average of log 2.86 cfu/100 ml

This trend for individual rivers was also seen when looking

at just seasonal averages as observed cfu/100 ml counts

were as high as log 5.08, however Thailand during the wet

season was the only exception being under this threshold

with an average mean of log 1.95 cfu/100 ml (Table2)

This trend was also observed when accounting all of the data

when the rivers were combined, with the average mean

values in the dry seasons exceeding this proposed threshold

value, while the wet season was much closer to acceptable

limits with a mean value of log 2.76 cfu/100 ml (Table2)

Closer examination based on land use types indicated that

for both seasons, most urban water sources exceeded this

proposed threshold value, with some seasonal means

exceeding this by more than 1 log (Table3) A notable

exception was the industrial land use types which the mean

was very close to this proposed threshold value (log 2.7 cfu/

100 ml) It is also important to note that the water treatment

land use types had an overall mean of log 1.7 cfu/100 ml

which was below this recommended limit demonstrating

that regional water treatment efforts were sufficiently

employed

There was an observed seasonal difference in the overall

mean values, with the dry season having approximately 1.5

log higher counts than the wet season (Table2) While it

might be expected that heavy precipitation may also

increase runoff events and in turn, increase overall numbers

of fecal indicator organisms, it is quite possible that the

substantial rainfall due to monsoon events during the wet

season in these regions achieved a dilution effect with the

microbial populations in these surface waters This is

supported by other studies which also observed seasonal

variation of E coli numbers in Southeast Asian surface

waters with reduced numbers observed during the wet

season (Isobe et al 2004) Additionally, surveys of

Southeast Asian agricultural surface waters have reported

log cfu/100 ml values similar to what was observed with

this work (Diallo et al.2008; Yajima and Kurokura2008)

It is important to note that there were significantly

higher levels of E coli in urban surface waters compared to

agricultural/rural waters Although microbial loads due to

fecal runoff of livestock operations is a likely source of

pollution, it is expected that higher density urban areas may have a greater fecal contaminant load in surface waters which receive urban runoff, especially if such wastewater

is minimally treated A survey of treated septage sludge in Vietnamese households reported a mean of 6 log cfu/g of dry weight for E coli indicating that even conventional waste treatment systems may have a potential impact on surface waters if not managed properly (Yen-Phi et al

2010) and a survey of urban canals in Thailand had even higher values ranging from 5.7 to 6.8 log CFU/100 ml (Giri

et al.2005) In addition, surveys of rivers associated with metropolitan areas in Indonesia also have demonstrated similar results to what has been reported here with counts ranging from 2.9 to 4.8 log cfu/100 ml (Kido et al.2009) Approximately 4 % of the E coli isolates analyzed demonstrated the presence of pathogenic genes It is sur-prising that many of the isolates harbored Shiga toxin-producing genes as human sources of this pathogen type are typically associated with E coli O157:H7, however it has been demonstrated that Shiga toxin genes are present in several E coli strains isolated from animal hosts (Nataro and Kaper1998) It is quite possible that runoff from small livestock operations within urban areas may be a potential source for these strains as studies investigating the inci-dence of pathogenic E coli in Vietnamese swine operations found STEC strains in irrigation water systems (Kobayashi

et al.2003) EPEC types were also detected at a relatively similar proportional number with the other pathogenic

E coli types This may not be uncommon as a recent survey within Taiwan of water treatment plants and surface waters of nearby rivers found that EPEC was a common diarrheagenic E coli type, with this pathogen type being detected in approximately 9 % of the 55 samples (Huang

et al 2012) ETEC was also a commonly observed path-ogenic E coli type, determined by the presence of the elt virulence gene (Table4) This pathogenic E coli type is typically associated with traveler’s diarrhea and fecal contamination of water has been known to be a major factor in its epidemiology (Nataro and Kaper1998) Also, a survey of young children suffering from diarrhea in Jakarta, Indonesia, found that approximately 20 % of rectal swab samples were positive for ETEC strains (Richie et al

1997) supporting the notion that this E coli pathogen type may not be uncommon Southeast Asian urban populations From this work, there is an indication that a fair per-centage of E coli found in urban surface waters may be pathogenic strains, as 3.9 % of the isolates sampled and tested possessed virulence genes However, only a limited number of isolates from each water sample were further processed for PCR analysis Additionally enteroaggrega-tive E coli, considered another divergent pathogenic group expressing aggregative adherence to gut epithelia tissue, was not investigated (Nataro et al.1987) This pathogenic

phenotype strain was not included in this study primarily

due to lacking access to an appropriate clinical isolate

E coli strain as a comparative positive control for PCR

analysis Due to these previously stated limitations, the

results of this study may provide an incomplete picture to

the relative risk of populations that utilize surface waters in

Southeast Asia

Nonetheless it is important to note that even with these

limitations pathogenic strains of E coli were observed

Given the mean counts in urban waters were observed in

some sampling events to exceed 4 log cfu/100 ml, it is not

unreasonable to predict that pathogenic E coli could be

present in these surface waters and the incidence of

path-ogenic stains may be rather high Although most

patho-genic E coli types require larger infectious doses (Kothary

and Babu2001), the observed high numbers of E coli in

this study combined with the continual exposure to these

surface waters could negatively impact public health,

especially if such waters are used by neighboring

com-munities for agricultural production (Lynch et al 2009)

Further, as E coli is an indicator organism for fecal

con-tamination, it is not unreasonable to consider other

patho-gens transmitted by the oral-fecal route may be present and

that there may potentially be higher risks to the health of

populations that utilize these surface waters for drinking,

recreational use, or agriculture

The results of this study demonstrate that in Southeast

Asia, urban surface waters and rivers associated with urban

activity have substantially high levels of E coli Further,

roughly 4 % of the isolates harbored pathogenic genes with

the most common pathogen types being either EPEC or

STEC These study results highlight the importance for

monitoring and treatment of urban waters in Southeast

Asia, especially if these waters are to be used for drinking

sources or for agricultural and aquaculture activity, as there

may be a negative impact on public health

Acknowledgments This work was supported by the UNU&GIST

Joint Programme on Science and Technology for Sustainability,

Gwangju Institute of Science and Technology, Korea and part of a

project funded by the Asia–Pacific Network for Global Change

Research (Project Reference Number:

ARCP2010-01CMY-Sthiannopkao).

References

Ackers ML, Mahon BE, Leahy E, Goode B, Damrow T, Hayes PS,

Bibb WF, Rice DH, Barrett TJ, Hutwagner L (1998) An outbreak

of Escherichia coli O157:H7 infections associated with leaf

lettuce consumption J Infect Dis 177(6):1588–1593

Aidar-Ugrinovich L, Blanco J, Blanco M, Blanco JE, Leomil L,

Dahbi G, Mora A, Onuma DL, Silveira WD, Pestana de Castro

AF (2007) Serotypes, virulence genes, and intimin types of Shiga

toxin-producing Escherichia coli (STEC) and enteropathogenic

E coli (EPEC) isolated from calves in Sao Paulo, Brazil Int J Food Microbiol 115(3):297–306

Byappanahalli M, Fujioka R (1998) Evidence that tropical soil environment can support the growth of Escherichia coli Water Sci Technol 38(12):171–174

Byappanahalli MN, Whitman RL, Shively DA, Ferguson J, Ishii S, Sadowsky MJ (2007) Population structure of Cladophora-borne Escherichia coli in nearshore water of Lake Michigan Water Res 41(16):3649–3654

CDC (2006) Ongoing multistate outbreak of Escherichia coli serotype O157:H7 infections associated with consumption of Fresh Spinach—United States, September 2006 Morb Mortal Wkly Rep 55(38):1045–1046

Diallo MBC, Anceno AJ, Tawatsupa B, Houpt ER, Wangsuphachart

V, Shipin OV (2008) Infection risk assessment of diarrhea-related pathogens in a tropical canal network Sci Total Environ 407(1):223–232

Field KG, Samadpour M (2007) Fecal source tracking, the indicator paradigm, and managing water quality Water Res 41(16): 3517–3538

Giri RR, Takeuchi J, Ozaki H (2005) Influence of night soil contamination on activated sludge microbial communities in Bangkok, Thailand Ecol Eng 25(4):395–404

Ha N, Kitajima M, Hang N, Matsubara K, Takizawa S, Katayama H, Oguma K, Ohgaki S (2008) Bacterial contamination of raw vegetables, vegetable-related water and river water in Ho Chi Minh City, Vietnam Water Sci Technol 58(12):2403–2411 Haas CN, Rose JB, Gerba CP (1999) Quantitative microbial risk assessment Wiley, New York

Huang S-W, Hsu B-M, Su Y-J, Ji D-D, Lin W-C, Chen J-L, Shih F-C, Kao P-M, Chiu Y-C (2012) Occurrence of diarrheagenic Escherichia coli genes in raw water of water treatment plants Environ Sci Pollut Res 19(7):2776–2783 doi: 10.1007/s11356-012-0777-4

Hunter PR (2003) Drinking water and diarrhoeal disease due to Escherichia coli J Water Health 1(2):65–72

Isobe KO, Tarao M, Chiem NH, Minh LY, Takada H (2004) Effect of environmental factors on the relationship between concentrations

of coprostanol and fecal indicator bacteria in tropical (Mekong Delta) and temperate (Tokyo) freshwaters Appl Environ Micro-biol 70(2):814–821

Jang J, Unno T, Lee SW, Cho KH, Sadowsky MJ, Ko G, Kim JH, Hur H-G (2011) Prevalence of season-specific Escherichia coli strains in the Yeongsan River Basin of South Korea Environ Microbiol 13(12):3103–3113 doi: 10.1111/j.1462-2920.2011 02541.x

Jerse AE, Kaper JB (1991) The eae gene of enteropathogenic Escherichia coli encodes a 94-kilodalton membrane protein, the expression of which is influenced by the EAF plasmid Infect Immun 59(12):4302–4309

Kido M, Yustiawati Y, Syawal M, Sulastri S, Hosokawa T, Tanaka S, Saito T, Iwakuma T, Kurasaki M (2009) Comparison of general water quality of rivers in Indonesia and Japan Environ Monit Assess 156(1):317–329 doi: 10.1007/s10661-008-0487-z

Kobayashi H, Khai LTL, Phan TT, Yamasaki S, Taniguchi T (2003) Prevalence of pathogenic Escherichia coli in a swine breeding environment in Can Tho province, Vietnam Jpn Agric Res Q 37(1):59–64

Kothary MH, Babu US (2001) Infective dose of foodborne pathogens

in volunteers: a review J Food Saf 21(1):49–68 doi: 10.1111/ j.1745-4565.2001.tb00307.x

Lynch MF, Tauxe RV, Hedberg CW (2009) The growing burden of foodborne outbreaks due to contaminated fresh produce: risks and opportunities Epidemiol Infect 137(3):307–315

Nataro JP, Kaper JB (1998) Diarrheagenic Escherichia coli Clin Microbiol Rev 11(1):142–201

Nataro JP, Kaper JB, Robins-Browne R, Prado V, Vial P, Levine MM

(1987) Patterns of adherence of diarrheagenic Escherichia coli to

HEp-2 cells Pediatr Infect Dis J 6(9):829–831

Ohno A, Marui A, Castro ES, Reyes AA, Elio-Calvo D, Kasitani H,

Ishii Y, Yamaguchi K (1997) Enteropathogenic bacteria in the

La Paz River of Bolivia Am J Trop Med Hyg 57(4):438–444

Pakalniskiene J, Falkenhorst G, Lisby M, Madsen SB, Olsen KE,

Nielsen EM, Mygh A, Boel J, Molbak K (2009) A foodborne

outbreak of enterotoxigenic E coli and Salmonella Anatum

infection after a high-school dinner in Denmark, November

2006 Epidemiol Infect 137(3):396–401

Paton AW, Paton JC (1998) Detection and characterization of Shiga

toxigenic Escherichia coli by using multiplex PCR assays for

stx1, stx2, eaeA, enterohemorrhagic E coli hlyA, rfbO111, and

rfbO157 J Clin Microbiol 36(2):598–602

Phan TT, Khai LTL, Loc CB, Hayashidani H, Sameshima T,

Watanabe T, Taniguchi T, Kobayashi H, Ito H, Akiba M (2003)

Isolation of Salmonella strains from the aquatic environment and

comparison with those of animal origin in Tan Phu Thanh

village, Mekong Delta, Vietnam Jpn Agric Res Q (JARQ)

37(4):237–241

Richie E, Punjabi NH, Corwin A, Lesmana M, Rogayah I, Lebron C,

Echeverria P, Simanjuntak CH (1997) Enterotoxigenic

Esche-richia coli diarrhea among young children in Jakarta, Indonesia.

Am J Trop Med Hyg 57(1):85–90

Santo Domingo JW, Bambic DG, Edge TA, Wuertz S (2007) Quo

vadis source tracking? Towards a strategic framework for

environmental monitoring of fecal pollution Water Res 41(16):

3539–3552

Sears CL, Kaper JB (1996) Enteric bacterial toxins: mechanisms of

action and linkage to intestinal secretion Microbiol Rev 60(1):

167–215

Simpson JM, Santo Domingo JW, Reasoner DJ (2002) Microbial

source tracking: state of the science Environ Sci Technol 36(24):

5279–5288

Soller J, Embrey M, Tuhela L, Ichida A, Rosen J (2010) Risk-based

evaluation of Escherichia coli monitoring data from

undisin-fected drinking water J Environ Manag 91(11):2329–2335

Solo-Gabriele HM, Wolfert MA, Desmarais TR, Palmer CJ (2000)

Sources of Escherichia coli in a coastal subtropical environment.

Appl Environ Microbiol 66(1):230–237

Toma C, Lu Y, Higa N, Nakasone N, Chinen I, Baschkier A, Rivas M, Iwanaga M (2003) Multiplex PCR assay for identification of human diarrheagenic Escherichia coli J Clin Microbiol 41(6):2669–2671

Unno T, Han D, Jang J, Lee SN, Ko G, Choi HY, Kim JH, Sadowsky

MJ, Hur HG (2009) Absence of Escherichia coli phylogenetic group B2 strains in humans and domesticated animals from Jeonnam Province, Republic of Korea Appl Environ Microbiol 75(17):5659–5666

US-EPA (2002) The 2000 national water quality inventory Office of Water EPA-841-R-02-001

US-EPA (2012) Recreational water quality criteria Office of Water EPA 820-F-12-058

Vollaard AM, Ali S, Smet J, Van Asten H, Widjaja S, Visser LG, Surjadi C, Van Dissel JT (2005) A survey of the supply and bacteriologic quality of drinking water and sanitation in Jakarta, Indonesia Southeast Asian J Trop Med Public Health 36(6):1552–1561

Wheeler C, Vogt TM, Armstrong GL, Vaughan G, Weltman A, Nainan OV, Dato V, Xia G, Waller K, Amon J, Lee TM, Highbaugh-Battle A, Hembree C, Evenson S, Ruta MA, Williams IT, Fiore AE, Bell BP (2005) An outbreak of hepatitis

A associated with green onions N Engl J Med 353(9):890–897 Yajima A, Kurokura H (2008) Microbial risk assessment of livestock-integrated aquaculture and fish handling in Vietnam Fish Sci 74(5):1062–1068

Yan T, Hamilton MJ, Sadowsky MJ (2007) High-throughput and quantitative procedure for determining sources of Escherichia coli in waterways by using host-specific DNA marker genes Appl Environ Microbiol 73(3):890–896

Yatsuyanagi J, Saito S, Miyajima Y, Amano K, Enomoto K (2003) Characterization of atypical enteropathogenic Escherichia coli strains harboring the astA gene that were associated with a waterborne outbreak of diarrhea in Japan J Clin Microbiol 41(5):2033–2039

Yen-Phi VT, Rechenburg A, Vinneras B, Clemens J, Kistemann T (2010) Pathogens in septage in Vietnam Sci Total Environ 408(9):2050–2053

Zimmer R, Verrinder Gibbins AM (1997) Construction and charac-terization of a large-fragment chicken bacterial artificial chro-mosome library Genomics 42(2):217–226