Epidemiological surveillance of human enteric viruses
by monitoring of different environmental matrices
A. Carducci, M. Verani, R. Battistini, F. Pizzi, E. Rovini, E. Andreoli and B. Casini
Department of Experimental Pathology, Medical Biotechnologies, Infectology and Epidemiology – University
of Pisa, via S. Zeno 35, 56127 Pisa, Italy
Abstract In the aim of studying possible relations between viruses detected in clinical specimens and the
ones found in different environmental matrices, in the period May 2004 to April 2005, the collection of
faecal samples from gastroenteritis cases and the monthly monitoring of raw and treated wastewater, river
water, seawater and mussels were carried out. The viruses considered for environmental monitoring were
adenovirus, rotavirus, enterovirus, norovirus, hepatitis A virus (HAV) and Torque teno virus (TTV): they were
searched for with PCR and RT-PCR and confirmed by gene sequencing. Faecal coliforms and somatic
coliphages’ counts were also determined. The surveillance of case detected 45 positive faecal samples out
of 255 (17.6%) while 35 of 56 environmental samples (62.5%) resulted positive for at least one of the
considered viruses. The detection of the same viral strain in the faeces of gastroenteritis cases and in water
was possible for adenovirus and rotavirus, which were also predominant in environmental matrices; thus they
could be considered as a reference for risk assessment.
Keywords Adenovirus; enterovirus; epidemiological surveillance; norovirus; rotavirus; water
Enteric viruses can be transmitted not only by direct contact, but also by indirect routes,
such as the consumption of contaminated food and water. In particular, the different uses
of water, such as drinking, irrigating, bathing and growing food (i.e. shellfish) can fre-
quently expose people to enteric viral infections (Lopman et al., 2003). The evaluation of
environmental viral hazards is an emerging problem, which deserves particular attention
because of the peculiar characteristics that differentiate viruses from other microorgan-
isms. Viruses, in fact, require lower infectious doses than most bacteria, can withstand
environmental factors and purification treatment, even for very long periods of time, and
present multiple transmission routes (WHO, 2004). In recent years progress in virological
assays (mostly biomolecular tests) has increased knowledge, but many questions remain
still unsolved: the standardization of detection methods, the choice of the most significant
agents for risk assessment, and the selection of a reliable indicator for viral contamination
(Carducci, 2005). For this last problem, a new possible agent can be proposed: it is the
TTV (Torque teno virus) (Nishizawa et al., 1997), a virus without a specific associated
pathology, but causing a persistent infection in the large majority of healthy people and
present in blood, faeces, respiratory secretions and other body fluids.
Environmental monitoring through effective, standardised virus detection techniques
combined with clinical surveillance could promote continuous, rapid exchange of infor-
mation on the spread and distribution of the main enteric viral agents and the incidence
of correlated pathologies. For this reason we planned a study aimed to analyse the human
enteric viruses environmental spread and its relations with virological diagnosis of gastro-
enteritis, to identify the most frequent viral pathogens in different types of water and to
evaluate the possible correlations between pathogenic enteric viruses and commonly used
Water Science & Technology Vol 54 No 3 pp 239–244 Q IWA Publishing 2006
faecal indicators. The search for TTV was also performed to evaluate its diffusion in
Materials and methods
Since May 2004 an epidemiological surveillance of viral gastroenteritis diagnosed on
faecal samples in the Clinical Virology Laboratory of the Department of Experimental
Pathology of the University of Pisa was carried out, simultaneously with an environmen-
tal monthly monitoring of raw and treated wastewater by a treatment plant of the city of
Pisa (samples of 1 and 10L respectively), of water from the river receiving this effluent
(10L), and from the sea at this river outfall (10L). Samples of mussels (20–30 individ-
uals for each sample) were also collected in this area.
Faecal samples analysed with immunological techniques and resulted positive were
retested with PCR and the detected nucleic acids were sequenced (ABI PRISM 310 Gen-
etic Analyzer, Applied Biosystems) to compare strains coming from clinical specimens
with the ones found in the environment.
Water monitoring and mussel analysis concerned enteric viruses (enterovirus, adeno-
virus, rotavirus, norovirus genotypes I and II, HAV) bacterial (Escherichia coli) and viral
(somatic coliphages and TTV) indicators. E. coli was determined by membrane filtration
technique on TBX agar (ISO 9308-1: 2000). For virological examination water samples
were concentrated using two stage tangential-flow ultrafiltration (Carducci et al., 2003).
Mussels were eviscerated and hepatopancreas from the same sampling were pooled. The
pool was divided in aliquots of 0.03g, then nucleic acids directly extracted by Total
Quick RNA cells and tissues (Talent kit) and QIAamp DNA kit (Qiagen-Germany)
(Carducci et al., 2004).
The concentrated samples were tested for coliphage analysis by plaque assay accord-
ing to the double agar layer method (using E. coli C, ATTC 13706, as host strain) (ISO,
1999). For enteric viruses the concentrated samples were decontaminated with chloro-
form, the nucleic acids extracted using QIAmp Viral RNA and DNA kits (Qiagen-
Germany) and the biomolecular tests (PCR, RT-PCR) performed using primers reported
in Table 1. The positivity was confirmed and the strains identified with gene sequencing.
Sequence analysis was carried out with NCBI Genebank.
Statistical analysis aimed to search correlations between E. coli and coliphages, and
the association between viral presence and indicators counts was performed by Excel,
Microsoft (Pearson correlation).
Results and discussion
In the period from May 2004 to March 2005 the surveillance of cases revealed 45
(17.6%) positive faecal samples out of 255, of which 17/255 (6.6%) were for rotavirus,
5/255 (1.9%) for adenovirus, 6/255 (2.3%) for astrovirus, 7/246 (2.8%) for norovirus gen-
otype I and 10/246 (4.1%) for norovirus genotype II, without particular epidemic peaks,
but with only small clusters of rotavirus infections in May 2004 and March to April 2005
The study of E. coli and somatic coliphage counts showed a slight decrease due to the
wastewater treatment (on average from 1 £ 107to 2.4 £ 105CFU/100mL for E.coli and
from 2.9 £ 106to 2 £ 104PFU/100mL for the somatic coliphages), followed by a new
increase in the river (7.9 £ 105CFU/100mL and 3.4 £ 105PFU/100mL) probably due to
further pollution (Figure 2).
At the river mouth E. coli was found only once and coliphage counts were highly vari-
able, probably due to the seawater dilution. No correlation was found between E. coli
and somatic coliphage concentrations. The coliphage concentration in mussels followed
A. Carducci et al.
the ones found in the seawater confirming the role of mussels as bioindicators of water
microbial pollution (Carducci et al., 1998).
Enteric viruses were mostly detected in raw sewage (nine positive samples out of 12
examined, 75%): two of them were positive for norovirus (genotype I in July, genotype
II in June), three for enterovirus in July 2004 (poliovirus 1-AY017242), February (entero-
virus 74 AY556057.1) and March 2005 (enterovirus 74 AY556057.1). Seven samples
Figure 1 Monthly number of the observed gastroenteritis cases according to the detected virus, specified
by X and Y content in the Figure (X: Month; Y: Number gastroenteritis cases)
Table 1 Primers used in RT-PCR and PCR assays
Region Sequences 50–30
Norovirus gen. I
Norovirus gen. II
50UTRATT GTC ACC ATA AGC AGC CA
CAC GGA CAC CCA AAG TA
CAA GCA CTT CTG TTT CCC CGG
Gilgen et al., 1997
Capside CCA ACC CAR CCA TTR TAC AT
AAA TGA TGA TGG CGT CTA
AAA AYR TCA CCG GGK GTA T
Gilgen et al., 1997
RNA pol.CGC CAT CTT CAT TCA CAA A
TWC TCY TTY TAT GGT GAT GAT GA
TTW CCA AAC CAA CCW GCT G
Gilgen et al., 1997
Gene VP7GTC ACA TCA TAC AATTCT AAT CTA AG
CTT TAA AAG AGA GAA TTT CCG TCT G
TGT ATG GTA TTG AAT ATA CCA C
ACT GAT CCT GTT GGC CAW CC
Gilgen et al., 1997
2A-2BATG CTT GGA TTG TCT GGA GT
GAA CAA ATA TCT CTT AAC CA
ATG ATG TTT GGA TTT CAT CAT
CTG GAG TCC ATT TGC CAA TT
Divizia et al., 1989
E1-A/B GCC SCA RTG GKC WTA CAT GCA CAT C
CAG CAC SCC ICG RAT GTC AAA
GCC CGY GCM ACI GAI ACS TAC TTC
CCY ACR GCC AGI GTR WAI CGM RCY
Allard et al., 1992
UTRAGC CCG AAT TGC CCC TTG AC
GTA AGT GCA CTT CCG AAT GGC TGA G
AGT TTT CCA CGC CCG TCC GCA GC
GCC AGT CCC GAG CCC GAA TTG CC
Okamoto et al., 1999
A. Carducci et al.
resulted positive for adenovirus in June, July and August 2004 (uncultured adenovirus
AY747675), in September (human adenovirus type 31 AY220987), October and Novem-
ber 2004 (human adenovirus type 31 AF161576.1) and in April 2005 (human adenovirus
type 41 AY220986.1).
In the plant effluent positive samples were less frequent (6/12, 50%), one for rotavirus
in March 2005 (human rotavirus A G3-AY900173.1) and three for adenovirus in May
(uncultured adenovirus AY747675), June 2004 (uncultured adenovirus AY747672.1) and
April 2005 (human adenovirus type 41 AY220986.1). Moreover, three noroviruses geno-
type II were found in January, February and March.
River water resulted contaminated in 9/12 (75%) samples, four were positive for rota-
virus: in May (human rotavirus A G4 AF373896.1), June (human rotavirus A G1
AF260950.1), August 2004 (human rotavirus A G2 AY660563.1) and April 2005 (human
rotavirus A G1 AB081799.1). Two positive samples for enterovirus were found in Janu-
ary (enterovirus 90 AY773285.1) and in February 2005 (human Coxsackie virus A1,
AF499635.1). Moreover, one strain of norovirus genotype I was found in January 2005
and three strains of genotype 2 in December 2004, February and March 2005. TT virus
was found in June (strain TCHN-G2), September (strain TCHN-A) and March 2005
In seawater (two positive samples of 12, 16.7%) were found human adenovirus type 2
AY293903.1 in April and two strains of norovirus genotype I and genotype II, both in
December. Only one of the eight samples of shellfish analysed (12,5%) was positive in
December for HAV (AF485328.1). This was the only case of HAV detection, but the
virus presence in mussels indicates a possible food related risk. No significant association
was found between the enteric viral presence and E. coli or coliphage counts.
The typing by gene sequencing of the detected viruses resulted in identification of
some link between adenovirus and rotavirus strains isolated from faeces and the ones
Figure 2 Results of microbiological analysis in raw sewage, treated sewage, river water and seawater
A. Carducci et al.
found in raw and treated sewage, in river and seawater. Such comparisons also showed
the same strain in different environmental matrices and periods.
Human adenovirus type 2 AJ293902 was found both in gastroenteritis cases in Decem-
ber 2004 and January 2005 and in seawater in April 2005; in June 2004 the strain
AY747675 was detected in raw and treated wastewater, indicating a possible high load of
the virus in this month, but also an insufficient efficiency of the plant in virus removal. In
the raw sewage other adenovirus strains were also found, the same (type 31 AF161576.1)
in October and November 2004.
The phylogenetic analysis of adenovirus strains proved that they belong to the same
superfamily A,F,C1 according to the VA RNA gene tree (Ma and Mathews, 1996). The
presence of the same strain of rotavirus (human rotavirus G3 AY900173.1) was revealed
in faeces in December 2004 and in treated wastewater in March 2005 showing a continu-
ous circulation of this strain. The phylogenetic distance among all the rotavirus detected
strains was great on a gene tree built on a 500-nucleotide sequence (Kostouros et al.,
TT virus strains were all different, they were found only in the river, and not in the
raw sewage. This apparent discrepancy may be due to the possible inhibition of PCR by
the concentrated wastewater.
Clinical virological data did not demonstrate any epidemic peak, but a continuous circula-
tion of enteric viruses, mainly adenovirus, norovirus and rotavirus. Environmental virolo-
gical analysis frequently resulted positive for adenovirus, enterovirus, norovirus, rotavirus
and TTV. E. coli and coliphages counts indicated a scarce reduction of microbial pollution
in the wastewater plant, a new increase in the river and a strong dilution in the sea. No
correlation was found between indicators and enteric virus presence. The detection of the
same viral strain in faeces of gastroenteritis cases and in water was possible for adenovirus
and rotavirus that were also predominant in environmental matrices; thus they could be
considered as a reference for risk assessment. Regarding a possible indicator for viral
pollution, the lack of correlation between somatic coliphages and coliforms and the scarce
representativity of both for the viral presence confirm the uselessness of these parameters
for virological risk assessment. The number of positive samples for TT virus is still insuffi-
cient to allow conclusions about its use as an indicator.
This work was funded by the Italian Ministry of University and Research (Prin 2004–2005).
Allard, A., Albinsson, B. and Wadell, G. (1992). Detection of adenoviruses in stool from healthy persons and
patients with diarrhea by two-step polymerase chain reaction. J. Med. Virol., 37, 149–157.
Carducci, A. (2005). Searching for viruses in aerosol: meaning and issues. Act of European Workshop on
water microbiology: a genome/proteome approach. Stresa, 11–12 April 2005.
Carducci, A., Cantiani, L., Gemelli, C. and Saggini, S. (1998). Valutazione sperimentale di metodi per il
controllo virologico dei mitili. Biologi Italiani, anno XXVIII, 4, 29–31.
Carducci, A., Casini, B., Bani, A., Rovini, E., Verani, M., Mazzoni, F. and Giuntini, A. (2003). Virological
control of groundwater quality using biomolecular test. Wat. Sci. Technol., 47(3), 261–266.
Carducci, A., Lucchesi, N., Casini, B., Mazzoni, F., Liaci, D. and Verani, M. (2004). Virological analysis of
shellfish for food safety and control. Wat. Sci. Technol., 50(1), 137–139.
Divizia, M., De Filippis, P., Di Napoli, A., Venuti, A., Perez-Bercoff, R. and Pana `, A. (1989). Isolation of
wild-type hepatitis A virus from the environment. Wat. Res., 23(9), 1155–1160.
A. Carducci et al.
Gilgen, M., Germann, D., Luthy, J. and Hubner, P. (1997). Three step isolation methods for sensitive Download full-text
detection of enteroviruses, rotaviruses, hepatitis A virus, and small round structured viruses in water
samples. Int. J. Food Microbiol., 37, 189–199.
ISO 9308-1 (2000). Water quality – detection and enumeration of Escherichia coli and coliform bacteria –
part 1: membrane filtration method.
Kostouros, E., Siu, K., Ford-Jones, E.L., Petric, M. and Tellier, R. (2003). Molecular characterization of
rotavirus strains from children in Toronto, Canada. J. Clin. Virol., 28, 77–84.
Lopman, B.A., Reacher, M.H., Van Dujinhoven, Y., Hanon, F.X., Brown, D. and Koopmans, M. (2003).
Viral outbreaks in Europe, 1995–2000. Emerg. Infect. Dis., 9, 90–96.
Ma, Y. and Mathews, M.B. (1996). Structure, function and evolution of adenovirus-associated RNA: a
phylogenetic approach. J. Virol., 70, 5083–5099.
Nishizawa, T., Okamoto, H., Konishi, K., Yoshizawa, H., Miyakawa, Y. and Mayum, M. (1997). A novel
DNA virus (TTV) associated with elevated transaminase levels in post tranfusion hepatitis of unknown
etiology. Biochem. Biophys. Res. Commun., 241, 92–97.
Okamoto, H., Takahashi, M., Nishizawa, M., Ukita, M., Fukuda, M., Tsuda, F., Miyakawa, Y. and
Mayumi, M. (1999). Marked genomic heterogeneity and frequent mixed infection of TT virus
demonstrated by PCR with primers from coding and noncoding regions. Virology, 259, 428–438.
WHO (2004). Guidelines for Drinking Water Quality, Chapter 7, 3rd edn, World Health Organization,
Geneva, Switzerland, pp. 121–144.
A. Carducci et al.