Occurrence and potential health risk of Cryptosporidiumand Giardia in different water catchments in Belgium

Amimul Ehsan & Thomas Geurden & Stijn Casaert & Jef Paulussen & Lut De Coster &

Toon Schoemaker & Rachel Chalmers & Grietje Grit & Jozef Vercruysse &

Edwin Claerebout

Received: 7 April 2014 /Accepted: 10 November 2014# Springer International Publishing Switzerland 2015

Abstract Humanwastewater and livestock can contrib-ute to contamination of surface water withCryptosporidium and Giardia. In countries where asubstantial proportion of drinking water is producedfrom surface water, e.g., Belgium, this poses a constantthreat on drinking water safety. Our objective was tomonitor the presence of Cryptosporidium and Giardiain different water catchment sites in Belgium and todiscriminate between (oo)cysts from human or animalorigin using genotyping. Monthly samples were collect-ed from raw water and purified drinking water at fourcatchment sites. Cryptosporidium and Giardia weredetected using USEPA method 1623 and positive

samples were genotyped. No contamination was foundin purified water at any site. In three catchments, onlylow numbers of (oo)cysts were recovered from rawwater samples (<1/liter), but raw water samples fromone catchment site were frequently contaminated withGiardia (92 %) and Cryptosporidium (96 %), especiallyin winter and spring. Genotyping ofGiardia in 38 watersamples identified the presence of Giardia duodenalisassemblage AI, AII, BIV, BIV-like, and E.Cryptosporidium andersoni, Cryptosporidium suis,Cryptosporidium horse genotype, Cryptosporidiumparvum, and Cryptosporidium hominis were detected.The genotyping results suggest that agriculture may be amore important source of surface water contaminationthan human waste in this catchment. In catchment siteswith contaminated surface water, such as the Blankaart,continuous monitoring of treated water for the presenceof Cryptosporidium and Giardia would be justified and(point) sources of surface water contamination shouldbe identified.

Keywords Giardia .Cryptosporidium . Surface water .

Drinkingwater . Belgium

Introduction

The protozoan parasites Cryptosporidium and Giardiaduodenalis are worldwide considered as an importantcause of gastrointestinal disease in human patients andin animals. In Belgium, the national reported incidencein 2010 was 2.5 and 10.8 cases per 100,000 inhabitants

Environ Monit Assess (2015) 187:6 DOI 10.1007/s10661-014-4157-z

A. Ehsan : T. Geurden : S. Casaert :G. Grit : J. Vercruysse :E. Claerebout (*)Laboratory for Parasitology, Ghent University,Merelbeke, Belgiume-mail: edwin.claerebout@ugent.be

J. Paulussen : L. De Coster : T. SchoemakerThe Water Group, Central Laboratory,Heverlee, Belgium

R. ChalmersCryptosporidium Reference Unit, Singleton Hospital,Swansea, UK

A. EhsanDepartment of Medicine, Faculty of Veterinary Science,Agricultural University,Mymensingh, Bangladesh

T. GeurdenZoetis, Veterinary Medicine and Development,Zaventem, Belgium

for Cryptosporidium and G. duodenalis, respectively(WIV-ISP 2010), and gastrointestinal symptoms inBelgian patients are frequently associated with theseprotozoa (Geurden et al. 2009).

Gastrointestinal infection is due to oral ingestion ofCryptosporidium oocysts or Giardia cysts. The highnumber of (oo)cysts excreted shortly after infection,together with the low infectious dose, results in an easyspread of infection (Smith et al. 2006a). Since the ex-creted (oo)cysts remain infectious for several weeks inwater, both treated tap water and recreational water areconsidered as excellent vehicles for infection, especiallyas Cryptosporidium in particular is highly resistant tochlorine disinfection. Since large scale drinking water-borne outbreaks of cryptosporidiosis such as the one inMilwaukee (1993) affecting more than 400,000 people,regula tory moni tor ing for the presence ofCryptosporidium has been a requirement in the UK,The Netherlands, and the USA and some provinces ofCanada and Australia. The EU/98/83/EC directive onlyrequires an investigation of pathogenic microorganisms,e.g., Cryptosporidium if there is a non-compliance forClostridium perfringens. In the UK, Cryptosporidiumhas been implicated in 69 % of waterborne outbreaks ofinfectious intestinal disease (Smith et al. 2006b). Anoutbreak of waterborne giardiosis in the town of SaintHubert in 2008 illustrates the public health threat fromintestinal protozoa in Belgium (Radio Télévision BelgeFrancophone 2008).

Surface water supplies are especially vulnerable tocontamination. Contamination with Cryptosporidiumand Giardia has been detected in water catchments inseveral countries in the EU, for example, inThe Netherlands (Ketelaars et al. 1995), Germany(Ajonina et al. 2012, 2013) Luxemburg (Helmi et al.2011), France (Mons et al. 2009), Italy (Briancesco andBonadonna 2005), Spain (Carmena et al. 2007), andPortugal (Júlio et al. 2012). A pilot study in four watercatchment areas in 2002 indicated that Cryptosporidiumand Giardia are also prevalent in surface water inBelgium (Aquaflanders, unpublished data). Since up to30 % of Belgian tap water is produced from surfacewater, this puts drinking water safety under pressure.Some of the surface water that is used for production ofdrinking water in Belgium is abstracted downstream ofwastewater discharges. Several studies have shown thathuman wastewater is often contaminated with high con-centrations of Cryptosporidium and Giardia and haveidentified wastewater treatment plants as point sources

of surface water contamination (Chalmers et al. 2010;Castro-Hermida et al. 2010). Next to human wastewater,livestock is considered to contribute to surface watercontamination, directly or through effluent from farmsteadings and hard standings or from manure-ladenfields (Sischo et al. 2000; Bodley-Tickell et al. 2002).Several studies indicated that zoonoticCryptosporidiumand Giardia genotypes are highly prevalent in livestockin Belgium (Geurden et al. 2007, 2008), suggesting thatwater contamination by infected farm animals may posea safety risk to drinking water in Belgium.

The objective of this study was to determine therelative contribution of human waste and livestock tocontamination of surface water with Cryptosporidiumand Giardia and to investigate whether the appliedtreatment procedures were efficient in eliminating theseparasites from drinking water. To this purpose, the pres-ence of Cryptosporidium and Giardia in surface waterand treated water was monitored in water catchmentsites in a rural area in Belgium and (oo)cysts fromhuman or animal origin were discriminated using(multilocus) genotyping.

Materials and methods

Sampling sites and water samples

Four water catchment sites were selected in the provinceof West-Flanders (Belgium), based on their location inan agricultural area or nearby a community. FromFebruary 2010 until April 2012, each month, threedifferent water matrices were sampled at the Blankaartwater catchment site, i.e., raw surface water (RW), waterfrom the storage basin (after 4 months of sedimentationand prior to treatment, SW), and treated tap water (TW).In the Blankaart water catchment, surface water is col-lected from the river Yzer downstream of the town ofYpres and two wastewater treatment sites (human wasteas possible contamination source) and from theBlankaart nature reserve (livestock grazing area)(Fig. 1).

Two matrices of water, RW and TW, were also sam-pled monthly from three other water catchment sites,i.e., Dikkebus, Gavers, and Zillebeke, from September2010 until August 2011. In the catchment sites ofDikkebus and Zillebeke, surface water is collected fromthe Ieperlee (a tributary of the river Yzer) upstream ofYpres, in an agricultural area (possible contamination by

6 Page 2 of 12 Environ Monit Assess (2015) 187:6

runoff from farms and fields). In the Gavers catchmentsite (Harelbeke), water is collected from the riverScheldt, downstream of the urban area of northernFrance and a wastewater treatment site (human wasteas possible contamination source). The volume of watersampled in each catchment site was 60, 15, and 15 l forTW, SW, and RW respectively. All water samples weretransported to the lab for filtration on the day ofsampling.

The water treatment in Blankaart, Dikkebus, andZillebeke consists of the following steps: storage (for aperiod of 4 months, in Blankaart only), oxidation, coagu-lation-sedimentation, rapid sand filtration, granular activat-ed carbon filtration, and disinfection. The water treatmentin the Gavers treatment plant consists of nitrification,followed by dephosphation in a flocculation step prior tostorage in theGavers reservoir. After storage, the followingtreatment steps are used: ultrafiltration, granular-activatedcarbon filtration, and disinfection.

Detection of Cryptosporidium oocysts and Giardiacysts

A protocol was optimized to detect Cryptosporidiumand G. duodenalis in water samples, based on theUSEPA method 1623 (USEPA 2005). To validate theprotocol, spiking experiments were performed with 15

water samples collected from the Blankaart in Februaryand March 2010 (5 RW samples, 5 SW samples, and 5TW samples) and with RWand TWwater samples fromeach water catchment site in December 2010 or January2011. A total of 9 RW samples (15 l), 6 SW samples(15 l), and 9 TW samples (60 l) were spiked with 100inactivated Cryptosporidium and 100 inactivatedGiardia (oo)cysts, permanently labeled with red fluo-rescent dye (ColorSeed™, BTF Pty Ltd., North Ryde,Australia). The ColorSeed™ (oo)cysts were added tothe water samples prior to filtering to estimate the per-cent recovery of (oo)cysts according to the manufac-turer’s instructions.

Both spiked and non-spiked water samples werefiltered through Filta-Max Xpress filters (IDEXXLaboratories, Inc., Westbrook, ME, USA) with the aidof a peristaltic pump with recommended flow rates of2 l/min. The Filta-Max Xpress filters were washed withthe Filta-Max Xpress automated washing station forelution of the filters following the manufacturer’s in-structions. The eluate was centrifuged and the volume ofsediment was measured. Between 0.5 ml (TW and SW)and 2 ml (RW) of sediment was used forimmunomagnetic separation (IMS) of the (oo)cysts,using Cryptosporidium and Giardia-specific antibody-coated magnetic beads according to the manufacturer’sprotocol (Dynabeads® GC-Combo, Invitrogen Dynal,

Fig. 1 Picture of the Blankaart catchment site. AYzer River, B water storage basin, C water supply channel from the Yzer river, and D theBlankaart nature reserve

Environ Monit Assess (2015) 187:6 Page 3 of 12 6

A.S., Oslo, Norway). IMS-purified cysts and oocystswere stained on well slides by fluorescein isothiocya-nate (FITC)-conjugated anti-Cryptosporidium and anti-GiardiaMAbs FITC-conjugated monoclonal antibodies(EasyStainTM) (BTF Pty Ltd. Australia). Slides wereexamined using a Leica Leitz DMRB fluorescence mi-croscope. The well surface was scanned at 200 or 400times magnification using a FITC fluorescence filter(450–590 nm Chroma Technology Corp.) and TexasRed fluorescence filter (530–585 nm, FT600, LP615)to distinguish ColorSeed™ (oo)cysts (which fluorescered with the Texas Red filter) from unspiked/natural(oo)cysts (which fluoresce bright green with the FITCfilter).Giardia cysts and Cryptosporidium oocysts wereidentified and counted based on their size, morphology,and fluorescence. Results were expressed as recoverypercentage for spiked (oo)cysts and as count per liter fornaturally occurring (oo)cysts. Slides containing natural(non-spiked) (oo)cysts were kept at 4 °C for DNAextraction.

Bacterial analysis

Fecal bacteria indicators were quantified in all watersamples. Coliform bacteria and Escherichia coli werequantified in 100 ml water samples by using the inter-national standard method ISO 9308-2, based on theColilert®-18 Idexx method with a detection limit of 1colony-forming unit (CFU)/100 ml. Enterococci weredetected by the Enterolert®-DW Idexx method, with adetection limit of 1 CFU/100 ml. C. perfringens wasquantified by using standard method ISO 6461/2-1986—Water quality—Detection and enumeration ofthe spores of sulfite-reducing anaerobes (Clostridia)—part 2: Method by membrane filtration and WAC/V/A/007, with a detection limit of 1 CFU/100 ml. Totalnumber of colony counts at 22/37 °C followed theinternational standard method ISO 6222 (1999) Waterquality—Enumeration of culturable microorganisms—Colony Count by inoculation in a nutrient agar culturemedium and conform WAC/V/A/001 with a detectionlimit of 1 CFU/ml. The bacteriological analyses weredone within 12 h after sampling.

Physicochemical analysis

Physicochemical characteristics of the water were deter-mined at each sampling event. The pH of water sampleswas determined according to the international standard

method ISO 10523:1994 Water quality: Determinationof pH. The turbidity of water samples was determinedfollowing the international standard method ISO7027:1999 Water quality: Determination of turbidity.The result was expressed as nephelometric turbidity unit(NTU).

DNA extraction and molecular analysis

DNAwas extracted from non-spiked water samples thatwere positive by microscopy for Cryptosporidium and/orGiardia. Genomic DNAwas extracted from (oo)cyststhat were scraped from the microscope slides by theQIAamp DNA Mini Kit (Qiagen GmbH, Hilden,Germany) according to the manufacturer’s instructions,incorporating an initial step of three freeze-thaw cycles(freezing in liquid nitrogen for 5 min and heating at95 °C for 5 min) in the protocol to maximize disruptionof (oo)cysts.

Previously described PCR protocols were used toamplify the 18S rDNA gene (Xiao et al. 2001;Johnson et al. 1995) and the hsp-70 gene (Morganet al. 2001) of Cryptosporidium. For subgenotyping ofCryptosporidium parvum and Cryptosporidiumhominis-positive samples, the 60 kDa glycoprotein(gp60) was targeted (Peng et al. 2001). For the identifi-cation of Giardia, the β-giardin gene (Lalle et al. 2005)was used in two-step nested PCR. For assemblage-specific amplification of Giardia, the triose phosphateisomerase (tpi) gene (Sulaiman et al. 2003) was used(Levecke et al. 2009; Geurden et al. 2008). For all PCRreactions, negative (PCR water) and positive controls(genomic DNA) were included. The PCR products werevisualized in agarose gel (1.5 %) stained with ethidiumbromide under UV light. PCR products were fully se-quenced by the BIG Dye Terminator V3.1 Cycle se-quencing Kit (Applied Biosystems). Sequencing reac-tions were analyzed on a 3100 Genetic Analyzer(Applied Biosystems) and assembled with the programSeqman II (DNASTAR, Madison WI, USA). To deter-mine the subgenotype, the fragments were aligned withhomologous sequences available in the GenBank data-base, using MegAlign (DNASTAR, Madison WI,USA).

Precipitation data

Daily rainfall data recorded at representative weatherstations from the Royal Meteorological Institute were

6 Page 4 of 12 Environ Monit Assess (2015) 187:6

used. The average of 3 days rainfall prior to the collec-tion date of water samples as well as rainfall on each ofthese individual days and on the day of water samplingwas used to investigate a potential association betweenprecipitation and concentration of Cryptosporidium andGiardia (oo)cysts in water samples.

Statistical analysis

The χ2 test was used to compare proportions of positiveRW samples for Cryptosporidium and Giardia betweendifferent seasons in the Blankaart water catchment.Giardia cyst counts and Cryptosporidium oocyst countsin Blankaart RW samples were compared between dif-ferent seasons using the Kruskal-Wallis test. The degreeof association between parasite concentrations, the con-centration of indicator bacteria, physicochemical param-eters, and rainfall was determined in all water samplesfrom the Blankaart catchment using the Pearson’s cor-relation test and the nonparametric Spearman rank cor-relation test. Values of p<0.05 were considered statisti-cally significant. All statistical tests were performedusing IBM SPSS statistics version 21.

Results

Prevalence and concentration of Cryptosporidiumand Giardia in water samples

Good recovery rates for both parasites were obtained,with higher recovery rates for Cryptosporidium (41–44 %) than Giardia (28–45 %, Table 1). Recovery ratesfor both parasites met the standards set by USEPA 2005(initial recovery rates of 24–100 %).

In Zillebeke, Dikkebus, and Gavers, only low num-bers of (oo)cysts were recovered occasionally from rawwater samples (<1 /l), mainly in winter (December–

March, Fig. 2). In contrast, raw water samples fromthe Blankaart catchment were frequently contaminatedwith Giardia (92 % posi t ive samples) andCryptosporidium (96 % positive samples). No signifi-cant difference was found between different seasons inthe proportion of RW samples positive forGiardia (χ2=5.738, df=3, p=0.125) and Cryptosporidium (χ2=2.898, df=3, p=0.408). However, Cryptosporidium oo-cyst counts (p<0.01) and Giardia cyst counts (p<0.01)were significantly different between different seasons,with higher counts in winter and spring (Fig. 3). Peakvalues of 35 Giardia cysts/l and 51 Cryptosporidiumoocysts/l were obtained in April 2010 and February2011, respectively. (Oo)cyst contamination of waterafter the storage basin followed a similar seasonal pat-tern, but (oo)cyst numbers were lower compared to rawwater, with peak values of 24 Giardia cysts/l (average 5cysts/l) and 4 Cryptosporidium oocysts/l (average 1oocysts/l) (Fig. 3). All samples from treated tap wateri n a l l c a t chmen t s i t e s we r e nega t i v e fo rCryptosporidium and Giardia.

Bacteriological indicators of fecal contamination

E. coli, total coliforms, enterococci, and C. perfringenswere detected in all samples of raw water and reservoirwater (Table 2). In treated water, these organisms wereonly found in very low numbers in 3 out of 63 samples.At the Blankaart catchment, water samples from thestorage basin contained higher concentrations of bacte-ria than raw water samples from the river (Table 2). Nosignificant correlation was observed between fecal bac-terial indicators and Giardia and Cryptosporidium(oo)cyst counts in any sample at any catchment sites(p>0.05, results not shown).

Physicochemical parameters

Physicochemical data of the water samples are shown inTable 2. Most of the (oo)cysts were detected in winterand spring, when the turbidity was lowest. There was nosignificant correlation between pH or turbidity and theoccurrence of Giardia and Cryptosporidium (p>0.05,results not shown).

Rainfall

Rainfall data of 3 days prior to the collection date ofwater samples (each individual day as well as

Table 1 The mean recovery rates±standard deviation for 100spiked Cryptosporidium oocysts and 100 Giardia cysts in rawsurface water (RW, n=10), water from the storage basin (SW,n=4), and treated tap water (TW, n=10)

Water matrices Cryptosporidium oocysts (%) Giardia cysts (%)

RW 44±10 28±22

SW 41±12 34±15

TW 45±16 45±20

Environ Monit Assess (2015) 187:6 Page 5 of 12 6

cumulative rainfall for 3 days) were plotted againstmonthly (oo)cyst counts for Cryptosporidium andGiardia. No significant correlation was observed be-tween rainfall data and the concentration of (oo)cystsin water samples (p>0.05, results not shown).

Molecular identification of parasites recoveredfrom water samples

Giardia sequences were obtained from 38 out of 53microscopy-positive water samples. Most sequences atthe Blankaart catchment site belonged to assemblage AI(n=7), AII (n=8), and E (n=10). Assemblages BIV (n=2) and BIV-like (n=1) were detected less frequently(Table 3). At the other catchment sites, assemblage AIwas predominant (n=6, NCBI accession numbersKF963556, KF963557, and KF963563-KF963566),while one assemblage AII sample was detected at theGavers catchment site (NCBI accession numberKF963577) and one assemblage E sample in Zillebeke(NCBI accession number KF963578).

In water samples from the Blankaart catchment,C. parvum (n=7), C. hominis (n=2), Cryptosporidiumandersoni (n=2), Cryptosporidium horse genotype (n=4), and Cryptosporidium suis (n=1) were detected(Table 3). OneC. parvum isolate fromBlankaart rawwaterwas identified as gp60 subtype IIdA20G3T5 (NCBIaccession number KF944374). No successful gp60sequence was obtained from C. hominis-positive samples.

Discussion

In the spiking experiment, higher recovery rates wereobtained for Cryptosporidium than Giardia, which wasin an agreement with previous observations (USEPA2005; McCuin and Clancy 2003; Carmena et al.2007). Recovery rates for both parasites met the stan-dards set by USEPA (2005).

There were large differences in contamination betweendifferent water catchment sites. Low levels of contamina-tion were found in Dikkebus, Zillebeke, and Gavers,whereas in the Blankaart site, high Cryptosporidium and

Fig. 2 Monthly occurrence of Giardia cysts (a) and Cryptosporidium oocysts (b) in the Zillebeke, Dikkebus, and Gavers catchment sites,expressed as total numbers of (oo)cysts seen in RW (15 l samples)

6 Page 6 of 12 Environ Monit Assess (2015) 187:6

Giardia counts were recorded. In the Dikkebus andZillebeke catchments, water is collected in the basin ofthe river Ieperlee, a tributary of the river Yzer upstream ofthe city of Ypres, whereas in the Blankaart site, water iscollected from the river Yzer downstream of Ypres andfrom the Blankaart nature reserve. At first sight, these datasuggest a major role of the city of Ypres as a source ofcontamination, and one would expect this contaminationto be of human nature. However, the genotyping resultssuggested that many of the recovered (oo)cysts were from

animal origin. Although two Cryptosporidium-positivesamples were identified as C. hominis, mostCryptosporidium species/genotypes suggested livestock,pigs and horses, as potential sources of water contamina-tion. Similarly, the majority of identified Giardia geno-types were associated with livestock (assemblage E) oranimals (assemblage AI). Although assemblage AI caninfect humans, the most common Giardia assemblages inhumans are AII and B (Sprong et al. 2009; Geurden et al.2009). Assemblages AII, BIV, and BIV-like were also

Fig. 3 Monthly occurrence ofGiardia (a) and Cryptosporidium (b) in the Blankaart catchment site, expressed as total numbers of (oo)cystsseen in RW (in red, 15 l samples) and SW (in green, 15 l samples)

Environ Monit Assess (2015) 187:6 Page 7 of 12 6

found in some water samples, suggesting that both humanand animal sources contribute to the water contaminationin the Blankaart catchment site.

Cryptosporidium spp. and G. duodenalis were foundthroughout the year, but the highest numbers of(oo)cysts were found in winter and spring. Higher con-tamination levels of Cryptosporidium and/or Giardia insurface water in winter and spring were also observed inprevious studies (Van Dyke et al. 2012; Keeley andFaulkner 2008; Isaac-Renton et al. 1996; Helmi et al.2011; Ajonina et al. 2012, 2013). In other studies, con-centrations of Cryptosporidium and/or Giardia werehigher in other seasons (Wilkes et al. 2009; Mons et al.2009; Horman et al. 2004; Castro-Hermida et al. 2009;Carmena et al. 2007; Ajonina et al. 2012, 2013; Agullo-Barcelo et al. 2013) or no seasonal differences wereobserved (Robertson and Gjerde 2001; Julio et al.2012; Carmena et al. 2007). The presence of high con-centrations of (oo)cysts in surface water in spring andsummer has been associated with the presence of infect-ed cattle (Castro-Hermida et al. 2009). Runoff frommanure-laden fields has been suggested as a source ofcontamination with Cryptosporidium oocysts andGiardia cysts (Slifko et al. 2000; Fayer 2004). Otherstudies have also reported an effect of land use on thecontamination of water with Cryptosporidium andGiardia (Ongerth et al. 1995; Ong et al. 1996; Hansenand Ongerth 1991; Burnet 2012). The Blankaart catch-ment site is situated in an agricultural area, and highprevalences of Cryptosporidium and Giardia have beenobserved in cattle in Belgium (Geurden et al. 2004,2006). However, we found no correlation between highconcentrations of Cryptosporidium and Giardia in thesurface water and rainfall prior to sampling. Moreover,according to Belgian legislation, spreading of manureonto the fields is only allowed from mid-February untilm i d -Oc t ob e r a nd h i gh conc en t r a t i o n s o fCryptosporidium andGiardiawere sometimes observedoutside this period. Therefore, there were no strongindications that runoff from agricultural land was amajor source of water contamination in the Blankaartcatchment.

Despite the high counts in raw water, the currentlyapplied methods for water treatment seem to be effi-cient. Sedimentation of the water for 4 months at theBlankaart catchment site strongly reduced the concen-tration of (oo)cysts in the water and no (oo)cysts weredetected in any of the treated water samples at anycatchment sites. Although it cannot be excluded thatT

able2

Descriptiv

estatisticsof

microbiologicalandphysicochemicalparametersof

thewater

samples,collected

monthly

infour

catchm

entsitesin

Belgium

Water

samples

nTurbidity

(NTU)

pHTo

talcoliform/100

ml

Enterococci/100

ml

E.coli/1

00ml

Clostridium

/100

ml

RM

SDR

MSD

RM

SDR

MSD

RM

SDR

MSD

BTW

270.09–0.36

0.23

0.08

6.82–7.44

7.05

0.17

0–179

7.46

36.54

0–1

0.04

0.20

0–0

00

0–2

0.2

0.50

BSW

270.86–20.00

4.03

4.05

7.32–8.65

8.22

0.30

250–141,400

22,506

35,064

0–5200

638

1177

0–11,200

2123

3006

10–3600

710

778

BRW

276.87–49.90

25.11

12.19

7.11–9.00

8.11

0.37

0–52,350

4354

10,983

0–120,000

6371

25,671

0–11,950

902

2785

10–1500

192

353

B7V

87.45–52.80

23.14

15.27

7.90–8.87

8.12

0.31

900–173,300

31724

57,701

0–8200

1311

2799

0–9400

1946

3179

0–3600

961

1102

B10

85.01–20.20

9.21

4.95

7.57–9.06

8.04

0.56

100–120,300

24974

45,093

100–400

164

120

0–12,000

1588

4208

20–1000

471

321

DTW

120.08–0.89

0.25

0.26

7.52–8.34

7.88

0.24

00

00

00

00

00

00

DRW

122.88–34.40

16.63

8.45

7.83–8.64

8.19

0.24

0–6350

2323

2240

0–1400

283

479

0–1800

400

783

30–960

290

257

GTW

120.07–0.40

0.19

0.12

7.68–7.95

7.80

0.08

00

00

00

00

00

00

GRW

120.34–2.86

1.20

0.69

7.93–8.47

8.21

0.15

400–77,700

8384

21,942

0–180

7360

50–1150

288

428

10–290

174

91

ZTW

120.09–0.30

0.19

0.06

7.65–8.13

7.81

0.14

00

00

00

00

00

00

ZRW

122.54–17.30

7.67

4.37

7.69–9.03

8.20

0.35

0–24,200

4003

7676

0–500

133

200

0–100

4747

10–500

185

187

nnumberof

samples,R

range,M

mean,SD

standard

deviation,BBlankaart,D

Dikkebus,GGavers,ZZillebeke,TW

treatedwater,R

Wrawwater,SW

storagewater

6 Page 8 of 12 Environ Monit Assess (2015) 187:6

low concentration of (oo)cysts in the treated waterremained undetected by the relatively insensitiveUSEPA method 1623 (Zukerman and Tzipori 2006;Shaw et al. 2008; Di Giorgio et al. 2002), the resultssuggest that the applied treatment methods are effective.However, the presence of high numbers of (oo)cysts insurface water in winter and springtime poses a threat incase of failure of the filtration system. Several drinkingwater-related outbreaks of giardiosis or cryptosporidio-sis occurred after failure of the treatment system(Webber 2002; Wallis et al. 1996; Stirling et al. 2001;Risebro et al. 2007; Pozio et al. 1997; MacKenzie et al.1994; Levy et al. 1998; Lee et al. 2002; Kramer et al.1996; Herwaldt et al. 1992; Barwick et al. 2000), in-cluding an outbreak of waterborne giardiosis in Belgiumin 2008 (Radio Télévision Belge Francophone, 2008).

Monitoring drinking water for Giardia andCryptosporidium is not compulsory in Belgium.According to the European Directive 98/83/CE, no bac-terial fecal indicator should be detected in 100 ml ofdrinking water. The same directive recommendschecking the presence of Cryptosporidium whenC. perfringens has been detected in drinking water.Although several studies showed that the use of fecalindicator bacteria provides relevant informationconcerning drinking or surface water quality (Xiaoet al. 2013; Touron et al. 2007; Tallon et al. 2006;Payment and Franco 1993; Obiri-Danso and Jones1999), others failed to establish correlations betweenthe occurrence of Cryptosporidium and/or Giardia andthe level of fecal contamination (Xiao et al. 2013;

Horman et al. 2004; Helmi et al. 2011; Hanninen et al.2005; Briancesco and Bonadonna 2005). Similarly, inour study, no correlation was found between bacterialfecal indicators or C. perfringens and Giardia andCryptosporidium counts, confirming that bacterialcounts are not always a reliable tool to predict thepresence/absence of these parasitic protozoa (WHO2011).

Conclusions

In conclusion, our results showed the presence ofCryptosporidium and Giardia, including human andzoonotic genotypes, in surface water at all investigatedcatchment sites. Although genotyping results suggestedthat most of the recovered parasites were from animalorigin, surface water contamination could not be asso-ciated with runoff from contaminated fields. Furtherresearch should aim at identifying (point) sources ofcontamination, to enable the implication of measuresto reduce surface water contamination. In addition, pub-lic health regulators should consider continuous moni-toring of the treated drinking water for the presence ofCryptosporidium and Giardia since the presence ofthese parasites in the surface water poses a constantthreat for drinking water safety. This is of particularimportance in many countries where a substantial pro-portion of drinkingwater is produced from surface waterand where a significant proportion of the water re-sources are inflows from upstream rivers, e.g.,

Table 3 Giardia duodenalis genotypes (assemblages) and Cryptosporidium species/genotypes detected in water samples from theBlankaart water catchment

G. duodenalis Species/genotype

n NCBI accession numbers

AI 9 KF963541, KF963543, KF963544, KF963547, KF963558-KF963562

AII 8 KF963552–KF963555, KF963567, KF963568, KF963573, KF963575

BIV 2 KF963569, KF963576

BIV-like 1 KF963549

E 10 KF963542, KF963545, KF963546, KF963548, KF963550, KF963551,KF963570-KF963572, KF963574

Cryptosporidium C. hominis 2 KF944361, KF944366

C. parvum 7 KF944362, KF944365, KF944367, KF944369, KF944371, KF944372, KF944375

C. andersoni 2 KF944360, KF944363

C. suis 1 KF944376

Horse genotype 4 KF944364, KF944368, KF944370, KF944373

Environ Monit Assess (2015) 187:6 Page 9 of 12 6

Belgium, The Netherlands, Hungary, Lithuania,Bulgaria, and Romania (Eurostat 2013).

Acknowledgments This work was supported by a doctoralscholarship for candidates from developing countries from GhentUniversity (Grant No. 01W05309). The authors would like tothank the Royal Meteorological Institute for providing daily rain-fall data.

Conflict of interest J. Paulussen, L. De Coster, and T.Schoemaker are (former) employees of The Water Group, aBelgian water supply company which owns the sampled catch-ment sites. The employees of the Watergroup were involved insampling and in writing of the manuscript. T. Geurden is currentlyemployed by Zoetis, a veterinary pharmaceutical company.

References

Agullo-Barcelo, M., Oliva, F., & Lucena, F. (2013). Alternativeindicators for monitoring Cryptosporidium oocysts inreclaimed water. Environmental Science and PollutionResearch International, 20(7), 4448–4454.

Ajonina, C., Buzie, C., Ajonina, I. U., Basner, A., Reinhardt, H.,Gulyas, H., et al. (2012). Occurrence of Cryptosporidium in awastewater treatment plant in north Germany. Journal ofToxicology and Environmental Health A, 75(22–23), 1351–1358.

Ajonina, C., Buzie, C., & Otterpohl, R. (2013). The detection ofGiardia cysts in a large-scale wastewater treatment plant inHamburg, Germany. Journal of Toxicology andEnvironmental Health A, 76(8), 509–514.

Barwick, R. S., Levy, D. A., Craun, G. F., Beach, M. J., &Calderon, R. L. (2000). Surveillance for waterborne-diseaseoutbreaks—United States, 1997–1998. MMWR CDCSurveillance Summaries, 49(4), 1–21.

Bodley-Tickell, A. T., Kitchen, S. E., & Sturdee, A. P. (2002).Occurrence of Cryptosporidium in agricultural surface watersduring an annual farming cycle in lowland UK. WaterResearch, 36(7), 1880–1886.

Briancesco, R., & Bonadonna, L. (2005). An Italian study onCryptosporidium and Giardia in wastewater, fresh waterand treated water. Environmental Monitoring andAssessment, 104(1–3), 445–457.

Burnet, J. B. (2012). PhD Thesis Université de Liège, Belgium.Dynamique spatio-temporelle de Cryptosporidium et deGiardia dans un réservoir d’eau potable et son bassin versant:cas des lacs de barrage de la Haute-Sure (Grand-Duché deLuxembourg) (partially in French).

Carmena, D., Aguinagalde, X., Zigorraga, C., Fernandez-Crespo,J. C., & Ocio, J. A. (2007). Presence of Giardia cysts andCryptosporidium oocysts in drinking water supplies in north-ern Spain. Journal of Applied Microbiology, 102(3), 619–629.

Castro-Hermida, J. A., Garcia-Presedo, I., Almeida, A., Gonzalez-Warleta, M., Da Costa, J. M., & Mezo, M. (2009). Detection

of Cryptosporidium spp. and Giardia duodenalis in surfacewater: a health risk for humans and animals.Water Research,43(17), 4133–4142.

Castro-Hermida, J. A., Garcia-Presedo, I., Gonzalez-Warleta, M., &Mezo, M. (2010). Cryptosporidium and Giardia detection inwater bodies of Galicia, Spain. Water Research, 44(20), 5887–5896.

Chalmers, R. M., Robinson, G., Elwin, K., Hadfield, S. J.,Thomas, E., Watkins, J., et al. (2010). Detection ofCryptosporidium species and sources of contaminationwith Cryptosporidium hominis during a waterborneoutbreak in north west Wales. Journal of WaterHealth, 8(2), 311–325.

DiGiorgio, C. L., Gonzalez, D. A., & Huitt, C. (2002).Cryptosporidium and Giardia recoveries in natural watersby using environmental protection agency method 1623.Applied Environmental Microbiology, 68, 5952–5955.

Eurostat. (2013). http://epp.eurostat.ec.europa.eu/statistics_explained/index.php/Water_statistics.

Fayer, R. (2004). Cryptosporidium: a water-borne zoonotic para-site. Veterinary Parasitology, 126(1–2), 37–56.

Geurden, T., Claerebout, E., Vercruysse, J., & Berkvens, D. (2004).Estimation of diagnostic test characteristics and prevalence ofGiardia duodenalis in dairy calves in Belgium using a Bayesianapproach. International Journal for Parasitology, 34, 1121–1127.

Geurden, T., Berkvens, D., Geldhof, P., Vercruysse, J., &Claerebout, E. (2006). A Bayesian approach for the evalua-tion of six diagnostic assays and the estimation ofCryptosporidium prevalence in dairy calves. VeterinaryResearch, 37(5), 671–682.

Geurden, T., Berkvens, D., Martens, C., Casaert, S., Vercruysse, J.,& Claerebout, E. (2007). Molecular epidemiology with sub-type analysis of Cryptosporidium in calves in Belgium.Parasitology, 134(Pt.14), 1981–1987.

Geurden, T., Geldhof, P., Levecke, B., Martens, C., Berkvens, D.,Casaert, S., et al. (2008). Mixed Giardia duodenalis assem-blage A and E infections in calves. International Journal forParasitology, 38(2), 259–264.

Geurden, T., Levecke, B., Caccio, S. M., Visser, A., DeGroote, G.,Casaert, S., et al. (2009). Multilocus genotyping ofCryptosporidium and Giardia in non-outbreak related casesof diarrhoea in human patients in Belgium. Parasitology,136(10), 1161–1168.

Hanninen, M. L., Horman, A., Rimhanen-Finne, R., Vahtera, H.,Malmberg, S., Herve, S., et al. (2005). Monitoring ofCryptosporidium and Giardia in the Vantaa River basin,southern Finland. International Journal of Hygiene andEnvironmental Health, 208(3), 163–171.

Hansen, J. S., & Ongerth, J. E. (1991). Effects of time and watershedcharacteristics on the concentration of Cryptosporidium oocystsin river water. Applied Environmental Microbiology, 57(10),2790–2795.

Helmi, K., Skraber, S., Burnet, J. B., Leblanc, L., Hoffmann, L., &Cauchie, H. M. (2011). Two-year monitoring ofCryptosporidium parvum and Giardia lamblia occurrence ina recreational and drinking water reservoir using standardmicroscopic and molecular biology techniques.Environmental Monitoring and Assessment, 179, 163–175.

Herwaldt, B. L., Craun, G. F., Stokes, S. L., & Juranek, D. D.(1992). Outbreaks of waterborne disease in the United States,

6 Page 10 of 12 Environ Monit Assess (2015) 187:6

1989–1990. Journal of American Water Works Association,84, 129–135.

Horman, A., Rimhanen-Finne, R., Maunula, L., von Bonsdorff, C.H., Torvela, N., Heikinheimo, A., et al. (2004).Campylobacter spp., Giardia spp., Cryptosporidium spp.,noroviruses, and indicator organisms in surface water insouthwestern Finland, 2000–2001. Applied EnvironmentalMicrobiology, 70(1), 87–95.

Isaac-Renton, J., Moorehead,W., & Ross, A. (1996). Longitudinalstudies of Giardia contamination in two community drinkingwater supplies: cysts levels, parasite viability, and healthimpact. Applied Environmental Microbiology, 62, 47–54.

Johnson, D. W., Pieniazek, N. J., Griffin, D. W., Misener, L., &Rose, J. B. (1995). Development of a PCR protocol forsensitive detection of Cryptosporidium oocysts in water sam-ples. Applied Environmental Microbiology, 61(11), 3849–3855.

Julio, C., Sa, C., Ferreira, I., Martins, S., Oleastro, M., Angelo, H.,et al. (2012). Waterborne transmission of Giardia andCryptosporidium at river beaches in southern Europe(Portugal). Journal of Water Health, 10, 484–496.

Keeley, A., & Faulkner, B. R. (2008). Influence of land use andwatershed characteristics on protozoa contamination in apotential drinking water resources reservoir. WaterResearch, 42, 2803–2813.

Ketelaars, H. A. M., Medema, G. J., van Breemen, L. W. C. A.,van der Kooij, D., Nobel, P. J., & Nuhn, P. (1995).Occurrence of Cryptosporidium oocysts and Giardia cystsin the River Meuse and removal in the Biesbosch reservoirs.Journal of Water Supply Research and Technology,44(Suppl. 1), 108–111.

Kramer, M. H., Herwaldt, B. L., Craun, G. F., Calderon, R. L., &Juranek, D. D. (1996). Surveillance for waterborne-diseaseoutbreaks—United States, 1993–1994. MMWR CDCSurveillance Summaries, 45(1), 1–33.

Lalle, M., Pozio, E., Capelli, G., Bruschi, F., Crotti, D., & Caccio,S. M. (2005). Genetic heterogeneity at the giardin locusamong human and animal isolates of Giardia duodenalisand identification of potentially zoonotic sub genotypes.International Journal for Parasitology, 35, 207–213.

Lee, S. H., Levy, D. A., Craun, G. F., Beach,M. J., & Calderon, R.L. (2002). Surveillance for waterborne-disease outbreaks—United States, 1999–2000.MMWR Surveillance Summaries,51(8), 1–47.

Levecke, B., Geldhof, P., Claerebout, E., Dorny, P., Vercammen,F., Caccio, S. M., et al. (2009). Molecular characterization ofGiardia duodenalis in captive nonhuman primates revealsmixed assemblage A and B infections and novel polymor-phisms. International Journal for Parasitology, 139, 1595–1601.

Levy, D. A., Bens, M. S., Craun, G. F., Calderon, R. L., &Herwaldt, B. L. (1998). Surveillance for waterborne-diseaseoutbreaks—United States, 1995–1996. MMWR CDCSurveillance Summaries, 47(5), 1–34.

MacKenzie, W. R., Hoxie, N. J., Proctor, M. E., Gradus, M. S.,Blair, K. A., Peterson, D. E., et al. (1994). A massive out-break in Milwaukee of Cryptosporidium infection transmit-ted through the public water supply. New England Journal ofMedicine, 331, 161–167.

McCuin, R. M., & Clancy, J. L. (2003). Modifications to UnitedStates environmental protection agency methods 1622 and

1623 for detection of Cryptosporidium oocysts and Giardiacysts in water. Applied Environmental Microbiology, 69(1),267–274.

Mons, C., Dumetre, A., Gosselin, S., Galliot, C., & Moulin, L.(2009). Monitoring of Cryptosporidium and Giardia rivercontamination in Paris area.Water Research, 43(1), 211–217.

Morgan, U. M., Monis, P. T., Xiao, L., Limor, J., Sulaiman, I.,Raidal, S., et al. (2001). Molecular and phylogenetic charac-terisation of Cryptosporidium from birds. InternationalJournal for Parasitology, 31(3), 289–296.

Obiri-Danso, K., & Jones, K. (1999). Distribution and seasonalityof microbial indicators and thermophilic campylobacters intwo freshwater bathing sites on the River Lune in northwestEngland. Journal of Applied Microbiology, 87(6), 822–832.

Ong, C., Moorehead, W., Ross, A., & Isaac-Renton, J. (1996).Studies of Giardia spp. and Cryptosporidium spp. in twoadjacent watersheds. Applied Environmental Microbiology,62(8), 2798–2805.

Ongerth, J. E., Hunter, G. D., & DeWalle, F. B. (1995). Watersheduse and Giardia cyst presence. Water Research, 29, 1295–1299.

Payment, P., & Franco, E. (1993). Clostridium perfringens andsomatic coliphages as indicators of the efficiency of drinkingwater treatment for viruses and protozoan cysts. AppliedEnvironmental Microbiology, 59, 2418–2424.

Peng, M.M., Matos, O., Gatei, W., Das, P., Stantic-Pavlinic, M.,Bern, C., et al. (2001). A comparison of cryptosporidiumsubgenotypes from several geographic regions. Journal ofEukaryotic Microbiology, (Suppl.), 28S–31S.

Pozio, E., Rezza, G., Boschini, A., Pezzotti, P., Tamburrini, A.,Rossi, P., et al. (1997). Clinical cryptosporidiosis and humanimmunodeficiency virus (HIV)-induced immunosuppres-sion: findings from a longitudinal study of HIV-positive andHIV-negative former injection drug users. Journal ofInfectious Disease, 176(4), 969–975.

Radio Télévision Belge Francophone (RTBF). (2008). http://www.rtbf.be/info/societe/detail_eau-impropre-a-la-consommation-a-saint-hubert?id=5225223.

Risebro, H. L., Doria, M. F., Andersson, Y., Medema, G., Osborn,K., Schlosser, O., et al. (2007). Fault tree analysis of thecauses of waterborne outbreaks. Journal of Water Health,5(Suppl 1), 1–18.

Robertson, L. J., & Gjerde, B. (2001). Occurrence ofCryptosporidium oocysts and Giardia cysts in raw waters inNorway. Scandinavian Journal of Public Health, 29(3), 200–207.

Scientific Institute for Public Health (WIV-ISP). (2010). https://www.wiv-isp.be/epidemio/epinl/plabnl/plabannl/index10.htm.

Shaw, N. J., Villegas, L. F., Eldred, B. J., Gaynor, D. H., Warden,P. S., & Pepich, B. V. (2008). Modification to EPA method1623 to address a unique seasonal matrix effect encounteredin some U.S. source waters. Journal of MicrobiologicalMethods, 75, 445–448.

Sischo, W. M., Atwill, E. R., Lanyon, L. E., & George, J. (2000).Cryptosporidia on dairy farms and the role these farms mayhave in contaminating surface water supplies in the north-eastern United States. Preventive Veterinary Medicine, 43,253–267.

Slifko, T. R., Smith, H. V., & Rose, J. B. (2000). Emerging parasitezoonoses associated with water and food. InternationalJournal for Parasitology, 30(12–13), 1379–1393.

Environ Monit Assess (2015) 187:6 Page 11 of 12 6

Smith, H. V., Caccio, S. M., Tait, A., McLauchlin, J., &Thompson, R. C. (2006a). Tools for investigating the envi-ronmental transmission of Cryptosporidium and Giardia in-fections in humans. Trends in Parasitology, 22(4), 160–167.

Smith, A., Reacher, M., Smerdon, W., Adak, G. K., Nichols, G., &Chalmers, R. M. (2006b). Outbreaks of waterborne infec-tious intestinal disease in England and Wales, 1992–2003.Epidemiology and Infection, 134(6), 1141–1149.

Sprong, H., Caccio, S. M., & van der Giessen, J. W. (2009).Identification of zoonotic genotypes of Giardia duodenalis.PLoS Neglected Tropical Diseases, 3(12), e558.

Stirling, R., Aramini, J., Ellis, A., Lim, G., Meyers, R., Fleury, M.,et al. (2001). Waterborne cryptosporidiosis outbreak, NorthBattleford, Saskatchewan, Spring 2001. CanadaCommunicable Disease Report, 27(22), 185–192.

Sulaiman, I. M., Fayer, R., Bern, C., Gilman, R. H., Trout, J. M.,Schantz, P. M., et al. (2003). Triosephosphate isomerase genecharacterization and potential zoonotic transmission ofGiardia duodenalis. Emerging Infectious Diseases, 9(11),1444–1452.

Tallon, P., Magajna, B., Lofranco, C., & Leung, K. T. (2006).Microbial indicators of faecal contamination in water: acurrent perspective. Water, Air, and Soil Pollution, 166,139–166.

Touron, A., Berthe, T., Gargala, G., Fournier, M., Ratajczak, M.,Servais, P., et al. (2007). Assessment of faecal contaminationand the relationship between pathogens and faecal bacterialindicators in an estuarine environment (Seine, France).Marine Pollution Bulletin, 54, 1441–1450.

U. S. Environmental Protection Agency (USEPA). (2005). http://www.epa.gov/microbes/documents/1623de05.pdf.

Van Dyke, M. I., Ong, C. S., Prystajecky, N. A., Isaac-Renton, J.L., & Huck, P. M. (2012). Identifying host sources, humanhealth risk and indicators of Cryptosporidium and Giardia in

a Canadian watershed influenced by urban and rural activi-ties. Journal of Water Health, 10(2), 311–323.

WAC_V_A_001: https://esites.vito.be/sites/reflabos/2013/Online%20documenten/WAC_V_A_001.pdf.

WAC_V_A_007: https://esites.vito.be/sites/reflabos/2014/Online%20documenten/WAC_V_A_007.pdf.

Wallis, P. M., Erlandsen, S. L., Isaac-Renton, J. L., Olson, M. E.,Robertson, W. J., & van Keulen, H. (1996). Prevalence ofGiardia cysts and Cryptosporidium oocysts and characteriza-tion of Giardia spp. isolated from drinking water in Canada.Applied Environmental Microbiology, 62(8), 2789–2797.

Webber, C. (2002). Outbreak of giardiasis in bay of plenty andmanawatu. In Annual Summary of Outbreaks in NewZealand 2001, Report for the Ministry of Health, pp. 38–39.

Wilkes, G., Edge, T., Gannon, V., Jokinen, C., Lyautey, E.,Medeiros, D., et al. (2009). Seasonal relationships amongindicator bacteria, pathogenic bacteria, Cryptosporidium oo-cysts, Giardia cysts, and hydrological indices for surfacewaters within an agricultural landscape. Water Research,43, 2209–2223.

World Health Organisation (WHO). (2011). Guidelines fordrinking-water quality, 4th edition.

Xiao, L., Singh, A., Limor, J., Graczyk, T. K., Gradus, S., & Lal,A. (2001). Molecular characterization of cryptosporidiumoocysts in samples of raw surface water and wastewater.Applied Environmental Microbiology, 67(3), 1097–1101.

Xiao, G., Qiu, Z., Qi, J., Chen, J. A., Liu, F., Liu, W., et al. (2013).Occurrence and potential health risk of Cryptosporidium andGiardia in the Three Gorges Reservoir, China. WaterResearch, 47(7), 2431–2445.

Zuckerman, U., & Tzipori, S. (2006). Portable continuous flowcentrifugation and method 1623 for monitoring of water-borne protozoa from large volumes of various water matrices.Journal of Applied Microbiology, 100, 1220–1227.

6 Page 12 of 12 Environ Monit Assess (2015) 187:6