f/EPA
                                 United States
                                 Environmental Protection
                                 Agency
                                 Health Effects Research
                                 Laboratory
                                 Cincinnati OH 45268
                                 Research and Development
                                 EPA-600/S1-81-061  Feb. 1982
Project  Summary
                                 Evaluation  of the Tentative
                                 Standard  Method  for  Enteric
                                 Virus  Concentration and
                                 Detection
                                 Mark D. Sobsey
                                  The  tentative  Standard  Method
                                 (TSM) for concentrating enteric viruses
                                 from large volumes of drinking water
                                 and other finished waters was evalu-
                                 ated for nine enteric viruses using
                                 100-gal (378-I) tap  water volumes
                                 experimentally contaminated with
                                 about 100 plaque-forming units (PFU)
                                 of  virus. The mean recoveries of
                                 poliovirus 1. coxsackieviruses A9 and
                                 B3, echovirus 7, bovine enterovirus 1,
                                 Minute Virus of Mice (parvovirus),
                                 reovirus 3, adenovirus 5 and simian
                                 adenovirus 11 were 22, 25, 2.5, 6.7,
                                 22, 5.0, <1, <1 and <1 %, respectively
                                 when  the viruses were evaluated
                                 individually in succession. Although
                                 the recovery differences  for the five
                                 enteroviruses suggested variable TSM
                                 effectiveness depending  upon  virus
                                 type, subsequent TSM experiments in
                                 which  these viruses  were simultan-
                                 eously tested showed no significant
                                 recovery differences, averaging 16%
                                 (±2.7, 95% Cl) overall. TSM recoveries
                                 for purified,  dispersed mixtures of
                                 enteroviruses obtained from raw
                                 sewage and seeded into 100-gal
                                 volumes of tap water averaged only
                                 3.9%  which was  approximately the
                                 same  as poliovirus recoveries during
                                 the same time period.
                                  Enterovirus recoveries showed marked
                                 seasonal variations that may  be
                                 related to water quality changes.
                                 Reduced virus recovery was partly due
                                 to interference with virus adsorption
                                 to filters caused by soluble or colloidal
                                 tap water constituents that were
                                 removable by granular activated
                                 carbon.  The effects  of these inter-
                                 ferences could be overcome by adding
                                 aluminum or magnesium chloride to
                                 the tap  water to enhance  virus ad-
                                 sorption.
                                  Low virus recoveries by  the TSM
                                 were also due to virus inactivation by
                                 the pH 11.5 glycine-NaOH used for
                                 eluting adsorbed viruses from filters.
                                 In this eluent the T-90 values (time for
                                 90% virus inactivation) for Minute
                                 Virus of Mice, poliovirus, reovirus and
                                 the simian adenovirus SV-11 were 20,
                                 1.8, <1 and <1 minutes, respectively.
                                  Virus recoveries with TSM and other
                                 negatively charged filters  could be
                                 markedly improved by various pro-
                                 cedural modifications: (1) the use of
                                 aluminum or magnesium chloride to
                                 enhance virus adsorption, (2) the use
                                 of less alkaline glycine or beef extract
                                 both  with and without EOTA for
                                 elution of  viruses from primary ad-
                                 sorbents, and (3) recirculation of the
                                 eluent through the adsorbent filters.
                                 With the exception of MVM, viruses in
                                 beef extract eluents could be efficiently
                                 reconcentrated by organic flocculation.
                                  Improved virus recoveries were also
                                 obtained with adsorbent filters that
                                 are more electropositive than TSM
                                 filters. Zeta Plus filters  that  are

-------
positively charged up to about pH 5 to
6 adsorbed >99% of various viruses
from tap water at ambient pH levels of
7.0-7.5 without added polyvalent
cation salts. Most viruses could be
effectively eluted with glycine-NaOH,
pH >9.5 or beef extract. Low levels of
poliovirus in 100-gal (378-I) volumes
of tap  water  were  more  effectively
recovered  with  Zeta Plus filters than
with the TSM using Cox filters.
  This Project Summary was devel-
oped by EPA's Health Effects Research
Laboratory, Cincinnati, OH, to an-
nounce 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).

Introduction and Background
  The potential transmission of human
enteric viruses via drinking water and
other waters is of considerable public
health concern.  The viruses of major
concern in water are those that infect
the alimentary tract of man and in some
cases animals, and are excreted in large
quantities  with the feces by infected
persons. These are the so-called enteric
viruses, and they include members of
the following taxonomic groups: (1) en-
teroviruses, (polio-, coxsackie- A and B
and echoviruses), (2) reo- and rotaviruses,
and  (3) adenoviruses as well as the
Norwalk-type agents and Hepatitis type
A, which are not yet taxonomized due to
limited information on morphology and
chemical composition.
  The continued occurrence of water-
borne  outbreaks of Hepatitis A and
perhaps other virus diseases such as
gastroenteritis as well as the apparent
isolation of enteric viruses from drinking
water supplies in the U.S. and elsewhere,
indicate the need for a more thorough
and  systematic  evaluation  of  the po-
tential for  virus disease transmission
via  drinking water. Evaluation of this
problem requires the development and
utilization of simple and reliable methods
to quantitatively  detect relatively small
amounts  of enteric viruses in  large
volumes of raw and finished drinking
water supplies.
  Research efforts on methods to detect
enteric viruses in water have produced
two-step microporous filter adsorption-
elution  methods that  are  sometimes
capable of concentrating small numbers
of enteroviruses  from  100 or  more
gallons of drinking water, and a tentative
method of this type was introduced into
the 14th edition  of Standard Methods.
  In this method the water is pressure
filtered through microporous (0.45/u to
8.0 /urn pore size) filters having large
surface areas to which viruses adsorb
by a reversible reaction. The adsorption
reaction is thought  to  be primarily
electrostatic in nature, although  weak
attractive forces,  such  as  hydrogen
bonding, dipole-dipole interactions and
hydrophobic  effects  may also be  in-
volved.  In order  to enhance  virus
adsorption to the filter sufaces, appro-
priate ionic conditions are created in the
flowing water by the controlled addition
of acid and sometimes polyvalent cation
salts  prior to passage  of the  water
through the adsorbent filter.  After
processing the desired volume of water,
the adsorbed viruses that have accumu-
lated  on the  filter  surfaces are  eluted
with  approximately one liter  of  an
alkaline buffer solution consisting of pH
11.5 glycine-NaOH. The eluate obtained
is readjusted by the addition of acid and
Aids  to ionic conditions  that enhance
virus  adsorption: pH  3.5  and  .0005  M
Aids. The viruses are then adsorbed to a
second,  smaller  (47-mm diameter)
microporous filter  series composed  of
fiberglass,  asbestos  and epoxy. The
adsorbed viruses are then eluted from
this filter with several  milliliters  of
alkaline  buffer solution. The  virus-
containing  eluate is adjusted to a
physiological pH  and isotonicity and
then assayed for viruses in cell cultures
using conventional  virological  tech-
niques.
  The development  of  flow-through
filter  adsorption-elution  systems  for
concentrating small quantities of viruses
from large volumes of water represents
a much needed technical achievement.
However, many  of the  key variables
involved  in  this  type of methodology
have  not been adequately evaluated,
and some  of the procedures and
equipment  in current use have  been
somewhat arbitrarily selected. A rational
and systematic   evaluation of the
methodology and materials is  needed.
  In particular, it has not been demon-
strated  that the   tentative  Standard
Method (TSM) is capable of efficiently
detecting viruses  of  all  major enteric
virus groups including Hepatitis A virus.
The method was  developed and eval-
uated primarily  for enteroviruses,
especially poliovirus,  and  there  is
information in the literature indicating
that the  method may be incapable  of
recovering some other enteric viruses.
For example, in the application of filter
adsorption-elution  methods to adeno-
viruses, one  study  found that adeno-
virus type 5 was  inactivated by  the
alkaline glycine-NaOH used for elution
of adsorbed viruses.
  Another deficiency concerning eval-
uation of the TSM is that the method has
not  been  evaluated over a  sufficient
time period with the same viruses and
water source  to determine if changes in
water quality influence its effectiveness.
Dissolved  and  colloidal  substances in
water may interfere with virus recovery,
and  the types  and  concentrations ol
such interfering substances  may
over time.
Objectives


General Objective
  To  rigorously and systematicall
evaluate the tentative Standard Methc
for concentrating and detecting enter
viruses  in  large volumes of drinkir
water and other finished waters.

Specific Objectives
  To evaluate the tentative Standar
Method for detecting and concentratin
enteric viruses from tap water using th
four different types of microporous f ilte
systems that are currently recommendei  (
as virus adsorbents and representative
viruses  from the  major enteric viru:
groups. The four primary virus adsorben
systems being evaluated are: (a)Balstor
fiberglass-epoxy filter tubes,  (b) Co>
fiberglass-asbestos-epoxy filter discs
(c) Mi III pore cellulose nitrate mem-
branes, and (d) Commercial Filters
fiberglass  cartridge filters  and Cox
fiberglass-asbestos-epoxy filter discs in
series. A total of nine  different viruses
are being tested: poliovirus, coxsackie-
viruses A9 and B3, echovirus 7, bovine
enterovirus 1, reovirus 3,  simian
adenovirus SV-11,  (human) adenovirus
5 and Minute Virus of Mice (an enteric
parvovirus).
  After evaluating the tentative Standard
Method with the nine different laboratory
strains of enteric viruses,  the  method
will then be evaluated using mixtures of
naturally occurring enteric viruses
obtained from sewage.
  If the present tentative  Standard
Method  is  found  to be  incapable  of
efficiently recovering certain  enteric
viruses, then  modifications of the
presently recommended methodology
will be  evaluated  for their ability  to
efficiently detect representatives of all
enteric virus groups.

-------
Conclusions
1.  The  current tentative Standard
   Method (TSM) for enteric virus
   concentration from  drinking water
   and  other finished  waters  has
   serious deficiencies causing it to
   give poor virus recoveries.
2.  One major deficiency of the TSM is
   that the pH 11.5 glycine-NaOH used
   to elute adsorbed viruses from filters
   rapidly inactivates adenoviruses and
   reoviruses.  It also inactivates en-
   teroviruses and parvoviruses, but
   not  as rapidly. Alternative  eluents
   which employ less alkaline (pH 9.5-
   10.5)  conditions  in  glycine, pro-
   teinaceous media such as  0.3-3%
   beef extract, chelating agents such
   as EDTA, or combinations of these
   conditions  show  considerable
   promise for all major enteric virus
   groups.  Virus elution efficiency is
   further improved when these alter-
   native eluents are  recirculated
   through the  adsorbents,  thereby
   increasing the contact time.
3.  Another serious deficiency of the
   TSM causing reduced virus recover-
   ies is that soluble and/or  colloidal
   substances in finished tap watercan
   interfere with virus adsorption to
   conventional, electronegative TSM
   primary virus adsorbents. There
   appear to be temporal changes in
   magnitude  of  these interfering
   effects  that may  be related to
   seasonal changes  in the nature and
   amount of interfering substances in
   tap  water. The effects  of these
   interfering substances can be over-
   come  by adding polyvalent cation
   salts, such as MgCI2 and AICI3, to the
   tap water to enhance virus adsorption
   to TSM primary virus adsorbents and
   the  Filterite pleated  cartridge fiber-
   glass-epoxy filter.
4.  The four alternative primary virus
   adsorbents of the TSM do not differ
   significantly in terms of virus recovery
   efficiency, although  recoveries with
   the  Balston  filter were frequently
   lower than those with the other three
   adsorbent systems. The primary virus
   adsorbent systems did differ  in terms
   of their susceptibilty to  clogging,
   with the Millipore and Carborundum-
   Cox systems being more susceptible
   than the Balston and Cox systems.
5.  The results of this study suggest that
   all enteroviruses are apparently
   recoverable with equal effectiveness
   by the TSM. This conclusion  is based
   on finding:  (1) that the recovery
   efficiencies for five different labor-
   atory-grown enteroviruses did not
   differ  significantly in simultaneous
   experiments and (2) that  recovery
   efficiencies for  natural enterovirus
   mixtures  obtained from  sewage
   were not significantly different from
   recovery efficiencies for exogenously
   added, laboratory-grown poliovirus
   during the same time period.
6. Microporous filters  that are  more
   electropositive than the current TSM
   filters and are composed of  dia-
   tomaceous earth-cellulose-"charge
   modified"  resin mixtures appear  to
   offer  distinct advantages  as  virus
   adsorbents because: (1) they  effi-
   ciently adsorb viruses from tap water
   at neutral  pH  without added  poly-
   valent cation salts, (2) they are less
   affected by soluble and/or colloidal
   interferences  in tap water  that
   reduce virus adsorption, and (3) with
  • the possible exception of adeno-
   viruses, adsorbed viruses can be
   readily eluted with either pH 9.5  to
   10.5 glycine-NaOH or 0.1 to 3% beef
   extract, pH 9 to 10.
7. Highly electropositive microporous
   filters composed of asbestos-cellu-
   lose mixtures may also be effective
   virus  adsorbents because they effi-
   ciently adsorb viruses over a wide pH
   range without added  salts and the
   adsorbed viruses can be eluted with
   pH 9 to  10 beef extract.


Recommendations
1. Because the current TSM  for con-
   centrating enteric viruses from
   drinking water  and other  finished
   waters  is  relatively ineffective,
   improved filter adsorption-elution
   methods that will be  suitable and
   effective in field  use should be
   developed  and evaluated.
2. Further  efforts should be  made  to
   develop and evaluate alternative
   eluents to pH 11.5 glycine-NaOH.  In
   particular,  the use of glycine-NaOH
   at lower pH levels and 0.1 to 3% beef
   extract with and without chelating
   agents  such  as  EDTA or  other
   additives should be investigated.
3. Further studies should be done with
   beef extract and other proteinaceous
   eluents to  improve  the  organic
   flocculation procedure for parvo-
   viruses.
4. Because polyvalent cation salts such
   as MgCI2 and AICI3 are capable  of
   overcoming the effects of the soluble
   and/or colloidal interferences in tap
   water that reduce virus adsorption
   efficiency, it may be prudent to
   routinely  use salt enhancement of
   virus adsorption, unless it  can be
   experimentally  demonstrated the
   efficient virus recoveries from large
   volumes of a particular  water are
   obtained without added salts.
5.  Electronegative fiberglass-epoxy
   pleated cartridge filters (Filterite)
   should be  considered as  a fifth
   alternative primary virus adsorbent
   because:  (1) they have  a large
   surface area and, hence,  a  large
   filtration-volume capacity for water;
   (2) they efficiently adsorb  viruses
   under the same conditions as TSM
   filters, and (3) the adsorbed viruses
   are  elutable  with pH 10.5 glycine-
   NaOH and beef extract.
6.  Additional studies should be done to
   further develop and evaluate im-
   proved and simplified filter  adsorp-
   tion-elution methods for virus con-
   centration from  water based on the
   use of electropositive adsorbent
   filters.
7.  The physico-chemical  mechanisms
   of virus adsorption to and elution
   from  filters  and  other  surfaces
   should be investigated in  order to
   provide  a  rational basis  for the
   further development and optimization
   of filter adsorption-elution methods
   to concentrate enteric viruses from
   water. In addition, the  chemical
   characteristics and behavior of the
   dissolved and/or colloidal substances
   in tap water that interfere with virus
   adsorption to filters should  also be
   fully investigated in order to develop
   optimum  methods for their elimina-
   tion or control.
  The results of this investigation have
appeared in  three  publications: "Con-
centration of Poliovirus from Tap Water
Using Positively Charged Microporous
Filters" by M.D. Sobsey and B.L. Jones,
Appl. and Environ. Microbiol. 37(3):588-
95 (Mar.  1979); "Evaluation of the
Tentative Standard  Method for Enteric
Virus Concentration from Large Volumes
of  Tap Water" by M.D. Sobsey  et al., J.
AWWA 72(5): 292-99 (May 1980); and
"Modifications of the Tentative Standard
Method for  Improved Virus  Recovery
Efficiency" by M.D. Sobsey et al., J.
AWWA 72(6):350-55 (June 1980).

-------