United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S1-83-018 Mar. 1984
&EPA Project Summary
Viruses in Water and
Reclaimed Wastewater
John L Riggs and David P. Spath
A study was initiated to determine the
occurrence and concentration of viruses
in high quality wastewater treatment
plant effluents from two treatment
plants in California. Disinfected second-
ary effluent from a treatment plant
(designated SP) was compared to
disinfected and filtered secondary
effluents from a second treatment plant
(designated LM). Biweekly quantitative
data from the two wastewater treat-
ment plants were obtained over a 16-17
month period. The identity of the viral
isolates and the sensitivity of four
different cell lines in permitting the
growth and isolation of the various virus
types were also determined. Approxi-
mately the same number of viruses per
gallon were isolated from the influents
obtained from both treatment plants.
On only two occasions were viruses
isolated from the disinfected final
effluent water. Both of those isolations
were made from samples collected at
treatment plant SP.
Four cell systems were used for virus
assays: Buffalo green monkey kidney
(BGMK), human rhabdomyosarcoma
(RD), monkey fetal kidney (MFK) and
human fetal diploid kidney (HFDK)
cells. The BGMK, RD, and MFK cells
routinely isolated 2 to 4 logio plaque
forming units of virus per gallon of
influent. Although less sensitive than
the other cell systems, the human fetal
diploid kidney cells were valuable in
isolating adenoviruses. In this study,
reovirus and echovirus type 7 were the
most frequently isolated viruses. Other
viruses recovered were echovirus,
poliovirus, coxsackievirus group A,
coxsackievirus group B, reovirus and
adenovirus.
Studies were also conducted to
determine the viral etiology of water-
borne outbreaks of acute infectious
nonbacterial gastroenteritis. During
the period of this report, five outbreaks
of acute gastroenteritis were suspected
to be of waterborne origin. In three of
the five outbreaks, evidence of viral
etiology was obtained. Small round
viruses, 27 nm in diameter, were found
by immune electron microscopy in
stools from two of the outbreaks. The
Norwalk agent was found in the third
outbreak investigated.
Cell cultures from chimpanzees,
marmosets and monkeys were used in
attempts to identify the agents produc-
ing gastroenteritis in humans. Such
techniques as centrif ugation of inoculum
onto cells and treatment of cells with
dilute solutions of trypsin, were used
separately and in combination. Evidence
of growth was sought by immuno-
fluorescence staining of inoculated
cultures with convalescent-phase hu-
man serum from individuals with
gastroenteritis who were found by
immune electron microscopy to shed
27 nm virus. In no case was a virus
isolated or growth detected by these
techniques.
This Project Summary was developed
by EPA's Health Effects Research
Laboratory, Research Triangle Park.
NC, to announce key findings of the
research project that is fully documented
in a separate report of the same title (see
Project Report ordering information at
back).
Objectives of the Study
In the U.S. today, there is a growing
awareness of the need to conserve and
reuse water. Reuse of wastewater
effluents is now an integral part of water-
resource management. In addition, there
has been strong support at the federal
-------
level for use of land treatment and
disposal as viable alternatives todischarge
of wastewater effluents into oceans or
streams. No Federal regulations define
treatment requirements or water quality
criteria for wastewater reuse Although
several states have established reuse
regulations, they differ significantly from
state to state.
One of the main objectives of this study
was to obtain quantitative data on virus
concentrations in high quality effluents
such as well-disinfected secondary
effluents, rather than disinfected, filtered
secondary effluents. Increased emphasis
on the reuse of wastewater effluents for
irrigation and recreation presents varying
degrees of health risks which are
commensurate with varying levels of
wastewater treatment, water quality, and
public exposure. From a health standpoint,
the chief concern is the spread of
infectious diseases from public contact
with wastewater effluents. Most bacterial
pathogens can be adequately controlled
with good disinfection practices, however,
very little is known about such practices
to control viral pathogens
A second major objective was the
detection and identification of viruses in
waters suspected to be sources of
infection in waterborne disease outbreaks.
A field investigation group consisting of
sanitary engineers, virologists, micro-
biologists and medical epidemiologists
was established to correlate information
on the viral content of suspected water,
the effectiveness of water treatment
processes, epidemiologic data, and other
relevant data pertaining to any suspected
waterborne outbreak of disease The
study made provision for virus concen-
trating equipment and personnel for the
sampling of large volumes of suspected
water immediately after an outbreak, and
for an assay of the samples for specific
causative viruses
A third, lesser objective was the
development and evaluation of cell
culture techniques for isolation of viral
gastroenteritis agents from fecal speci-
mens m cases of suspected viral diarrhea.
Different methods of treatment were
used to enhance cell susceptibility for
growth of viral agents.
Wastewater Treatment Plant
Study Design
Permission was obtained from local
authorities to conduct the virus study at
two community wastewater treatment
plants. Virus samples were taken of raw
influent wastewater and at pre- and post-
chlormation points after treatment.
Plant effluents were characterized by
chlorine residual, BOD, turbidity, sus-
pended solids, total and fecal coliforms
and fecal streptococcus. One plant
(designated SP) produced a well-disin-
fected secondary effluent, while the
second plant (designated LM) produced a
filtered and disinfected secondary effluent.
Both plants neutralized the chlorine used
for disinfection before releasing the final
effluent.
The equipment, instruments and
personnel needed for the concentration
procedures were transported to the
collection site. The concentration proce-
dure consisted of pumping the sample
water (usually 380 liters) through an
orlon pre-filter or sand pre-filter and then
to a proportional pump which added AlCIs
to 0 0005 M and adjusted the pH to
approximately 3.5. The sample was then
pumped through 0.45 fjm Filtente* filters
at a rate of approximately 2 gal/min. A 1 -
4 liter sample of plant influent was also
collected (usually as a 24 hr composite
sample) on the day of concentration
sampling. The collected samples were
transported to the laboratory on wet ice
In the laboratory, the Filtente filters
were eluted with one liter of 0 05 M
glycme buffer, pH 9.0, containing 0 1%
beef extract. The pre-filters were eluted
with one liter of 1 % beef extract and the
composite sample adjusted to 1% beef
extract.
The eluate from the Filtente was
adjusted to a pH of approximately 7 Oand
further concentrated to approximately 30
ml using a Millipore Pellicon membrane
concentrator. The eluates from the pre-
filter and the composite sample were
adjusted to a pH of 3.5 and allowed to floe
The floe was recovered by centrifugation
and dissolved m 015 M Na2HPO4, the
resulting volume was approximately 30
ml.
The samples were inoculated onto
human fetal diploid kidney cells (HFDK)
grown m culture tubes (5 tubes per
dilution, 3 dilutions) and quantitated
following an MPN procedure. The samples
were also inoculated for plaqueformation
onto RD cells, buffalo green monkey
kidney cells (BGMK), and monkey fetal
kidney cells (MFK) grown m Petri plates (5
plates per dilution, 3 dilutions). The
inoculum consisted of a 0.5 ml volume of
the virus concentrate per culture dish and
tube. Thus, approximately one half of
each virus concentrate samples was
'Mention of tradenames or commercial products
does not constitute endorsement or recommenda-
tion for use by the U S Environmental Protection
Agency
inoculated onto the cell culture systems.
The cultures were incubated at 36°C for
10 to 14 days. The plates were scored as
in the MPN determination (for example, if
3 of the 5 plates each had one or more
plaques, the plates were scored as 3 of 5
infected, if one plate of the five had one or
more plaques with the remainder showing
no plaques, the plates were scored as 1 of
5 infected ) The plaques were also
counted for a total virus count. Thus an
MPN determination and total plaque
count can be carried out on the same
cultures. Plaques were chosen and
passed for confirmation, and certain of
the isolates were identified serologically
using the LBM serum pools or serum
pools available in our laboratories.
Wastewater Treatment Plant
Study Results
A total of 63 samples were assayed for
virus in samples from plant SP influent,
the number of viruses isolated increased
in the late summer/early fall of the fiscal
year of the study but not during the
second year. At plant LM, a later
summer/early fall peak of viruses m the
influent was evident only in the case of
the HFDK cell system. The other three cell
systems showed no consistent pattern.
The BGM, RD, and MFK cell systems
routinely picked up from 2 to 4 logio
plaque forming units of virus per 3.8 liters
(1 gallon) of influent. The HFDK cell
system was less sensitive, showing
usually 2 logio of virus by the MPN
technique. The HFDK cell system is
valuable, however, in isolating some
types of vi ruses that grow only in that cell.
Virus levels in the pre-chlorinated
effluent samples were significant^
reduced. The highest plaque count?
were obtained with BGM cells anc
ranged upward to 24 PFU/3 8 liters
Virus counts were typically in the 1 to E
PFU/3.8 liter range and m a few sample;
were undetectable. A late summer/earh
fall peak was only slightly detectable
during the first year of sampling at plan
SP, counts ranged between 5 and 2(
PFU/3.8 liters during this period.
All viruses in the pre-disinfected fina
effluent from the two plants wen
identified, but because of the number o
viruses isolated from the plant influen
water, only representative isolates when
identified. Reovirus was the most fre
quently isolated virus group during th
course of the study. Isolates were passe
m cell culture and were identified a
reoviruses by immunofluorescence
Reoviruses were isolated in 42 of the 6
different runs (67%). Echoviruses wer
-------
the predominant subgroup of entero-
viruses isolated with echovirus type 7
being the most frequently isolated
enterovirus at both the LM and SP
treatment plants.
On only two occasions were viruses
isolated from the disinfected final effluent
water. Both isolations were made from
samples collected at the SP treatment
plant, the first isolation was in August
1979, when 0.099 MPN (1/5 undilute
infected, 0/5 1:4 dilution infected, 0/5
M6 dilution infected, "Filtente" filter)
per 3.8 liters was isolated in the BGMK
cell system and was subsequently
identified as reovirus type 2 The second
isolation was m October 1 979; a total of
0.192 MPN per 3 8 liters was isolated in
the HFDK cell system (0/5 undilute
infected, 1/5 1.4 dilution infected, 0/5
1 16 dilution infected, Prefilter; 1/5
undilute infected, 0/5 1:4 dilution
infected, 0/5 1:16 dilution infected,
"Filtente" filter) These viruses were
identified as echovirus type 7
Investigation of Suspected
Waterborne Virus Disease
Outbreaks
The procedure followed in investigating
gastroenteritis outbreaks was as follows:
county health officials notify the Infectious
Disease Section of the State Department
of Health Services that such an outbreak
is occurring; the Infectious Disease
Section then provides epidemiologic
assistance in the field investigation
When waterborne disease is suspected,
the Department's Sanitary Engineering
staff is also immediately notified and a
joint investigation is conducted.
During suspected waterborne virus
disease outbreaks, the Stat viral and
Rittbettsial Disease laboratory would
provide for immediate use, large-volume
virus concentrators for sampling the
suspect waters, and for the assay of the
concentrated laboratory samples for the
presence of specific causative viruses,
utilizing the same procedures as those
used for the concentrated wastewater
samples. Concurrently, fecal specimens
and acute and convalescent phase serum
samples are obtained from infected
patients. The fecal specimens would be
utilized in enteric virus isolation proce-
dures while the serum samples would be
utilized in any required immune electron
microscopic procedures.
Cell cultures from primates (chim-
panzee, marmoset, monkeys) were used
in attempts to culture the agents producing
gastroenteritis in humans (parvovirus-
like particles, rotaviruses) Such tech-
niques as centrifugation of inoculum onto
cells and treatment of cells with dilute
solutions of trypsin were used separately
and in combination Evidence of growth
was determined by using convalescent
phase human serum from individuals
with gastroenteritis which were shown to
be positive by IBM.
Waterborne Viral Disease
Results
Virus testing of water associated with
the three gastroenteritis outbreaks
yielded positive results. In the first
outbreak, thirty members of a little league
football team who drank creek water from
spray irngators being used to water
portions of the football field experienced
symptoms of acute gastroenteritis (vomit-
ing, diarrhea); thirteen other team
members who did not drink the water
remained symptom-free.
Stool samples and acute and convales-
cent serum samples from several of the
infected team members were examined
by immune electron microscopy (IEM).
Two of six stool samples submitted
showed a parvovirus-like particle to be
present, while 11 of 16 convalescent
serum samples reacted with the particles.
None of the acute serum samples reacted
with the parvovirus-like particles to any
extent and it is assumed that the particles
present were the cause of the gastroen-
teritis.
Some time after the outbreak, a sample
of one gallon of the creek water was
submitted for viral assay. After organic
flocculation to concentrate any viruses
present, the resulting concentrate was
inoculated onto RD and BGMK cells in an
MPN type assay No viruses were isolated
in the RD cell cultures, but in the BGMK
cells approximately 34 viruses per 3.8
liters were obtained. Additional cultures
of RD and BGMK cells were inoculated
with aliquots of the concentrates from the
water by high speed centrifugation. The
RD cell system detected a total of
approximately 19 viruses per 3.8 liters of
water, while the BGMK cell system
detected a total of >1,566 viruses per 3.8
liters, again showing the different
sensitivity of the two cell systems to the
different viruses present. Representative
samples of the isolates which showed
reovirus CPE were identified as reoviruses
by IFA and some were typed, all proving to
be reovirus type 2. The isolated viruses
which showed enterovirus CPE were
subsequently typed as echovirus types
11, 13, and 25. The creek from which the
irrigation water was obtained was
subsequently shown to be contaminated
by a leaking sewer line.
The second gastroenteritis outbreak
occurred in vacationers staying in
cottages served by chlorinated water
obtained from a nearby lake. Over 100
individuals showed symptoms of acute
gastroenteritis (vomiting, diarrhea) after
staying at different cabins or cottages
within the district. After appropriate
epidemiological investigation had taken
place to assure a suspect waterborne
cause, samples of the water were obtained
for viral assay A 380 liter sample was
concentrated at the water treatment
facility and samples were obtained from a
cottage involved in the outbreak and from
a fire hydrant on the water system. No
viruses (<1 virus per 190 liters of sample)
were isolated from these samples. A later
sample of source water before chlormation
proved to have a virus concentration of
220 plaque forming units (PFU) of virus
per 380 liters of water. These viruses
were identified as reovirus type 2 and
echovirus type 14.
Stool samples and samples of acute
and convalescent serum were examined
by immune electron microscopy. Positive
reactions were obtained with several of
the convalescent sera to a 27 nm
parvovirus-like particle in the stool
specimens. Ten paired serum specimens
were tested at the National Institutes of
Health for radio-immunoassay against
Norwalk-like agents. Eight of the 10 pairs
of sera showed evidence of a >4-fold rise
in titer to Norwalk virus The Norwalk-like
agent was thought to be either partially
or wholly responsible for this second
outbreak of gastroenteritis The results,
particularly the isolation of echovirus and
reovirus, strongly support the contention
that the etiological agent was derived
from contaminated source water.
The third outbreak occurred on an
Indian reservation where the water
supply consists of an infiltration gallery
with a wet well next to a creek. The water
was chlorinated and subsequently filtered
before going into the distribution system.
Samples of 380-liters of water were
concentrated at two sites at the reserva-
tion: 1) source water before chlorination,
and 2) an external tap at a residence.
Results of these samples showed 0.8 PFU
of viruses per 3 8 liters in the source
water and 9 6 PFU of virus per 3.8 liters at
the residence All of the isolated viruses
were identified as coxsackie B3.
Results of immune electron microscopy
showed a 27 nm parvovirus-like particle
in a stool specimen which reacted with
convalescent serum samples collected
-------
from six individuals experiencing gas-
troenteritis. These results indicate that a
Norwalk-like virus agent was probably
the cause of the outbreak. The coxsackie
B3 virus probably did not contribute to the
gastroenteritis, but isolation of this agent
proves human fecal contamination of the
water supply. A faulty chlormator in the
water treatment system resulted in
unchlorinated water being distributed
within the system. Subsequent concen-
trating and assaying for viruses after the
chlorinator was repaired resulted in no
viruses being isolated.
Cohform analysis of grab samples
taken during the collection of large-
volume virus samples yielded positive
findings each time viruses were recovered.
Total coliform counts were 66 and 64,000
per 100 ml in two water samples that
yielded enteric viruses in concentrates of
380-liter. Fecal coliform counts were 27
and 30,000 in the same samples. No
viruses were isolated from water supplies
that would be considered potable based
on coliform levels or chlorination practice.
Discussion and Conclusions
Although it has been reported in the
literature that an average of 50% of
seeded virus added to treated sewage
effluent can be recovered using pleated
membrane filters, the results of this study
would tend to indicate either that wild
viruses cannot be recovered to the extent
described or that viruses were not
present in the disinfected final effluent of
the two treatment plants. The poor quality
of the effluent especially at plant SPfhigh
turbidity, suspended solids, BOD and
coliform count) would lead one to expect
to recover viruses. However, on only two
occasions were viruses isolated, and then
only in low numbers. Possibly, the
organic load of the effluent water
competes with viruses for adsorption
sites on the pleated membrane filters
when this technique of concentration is
used. The actual viral content of a typical
disinfected secondary effluent therefore
could probably not be ascertained from
the study because of the shortcomings of
the virus recovery method.
The recovery of viruses from the
influent and from the predisinfected final
effluent showed that on the average up to
a 300 log 10 reduction of viruses occurred
during the treatment process. Depending
upon the cell type used for viral isolation,
a late summer peak of the number of
viruses isolated occurred only during the
first year of the study.
It is evident from the number and types
of viruses recovered in the cell cultured
systems used in this study that any test
for viruses in water will have to be
qualified as to the virus types being
sought. This is due to the variation in the
sensitivities of different cell culture
systems to different viruses. It has been
reported that RD cells are somewhat
more sensitive to group Acoxsackieviruses
than other conventional cell culture
systems These results were confirmed in
these studies by the greater number of
isolations of these viruses in RD cells as
compared to the number of isolates in th
three other cell systems combined.
In the three waterborne outbreaks (
gastroenteritis investigated, huma
viruses were isolated from'the wate
indicating that the water supply wa
contaminated with human fecal materia
In each of these three outbreaks
parvovirus-like agent was implicated i
the causative agent by immune electro
microscopy of stools collected fro
individuals involved in the gastroenterit
outbreaks. Although laboratory techniqui
are not available for the isolation an
cultivation of these agents from wat
samples, the detection of other huma
enteric viruses from the implicated wat
supplies provides strong supportiv
evidence that water was the source of th
etiologic agent.
Although efforts to grow the a
present-uncultivable agents (parvoviru
rotavirus, hepatitis virus) were n
successful, the time and effort expende
on this phase of the work was n
extensive enough to detail or discu
thoroughly.
John L Riggs and David P. Spath are with the State of California Department of
Health Services, Berkeley, CA 94704.
Elmer W. Akin is the EPA Project Officer (see below).
The complete report, entitled "Viruses in Water and Reclaimed Wastewater,"
(Order No. PB 84-128 461; Cost: $10.00, subject to change) will be available
only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Health Effects Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
6US GOVERNMENT PRINTING OFFICE 1984-759-015/7
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use S300
------- |