United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-82-025 May 1982
Project Summary
Increasing ARCAT® Test
en
Sesitivity for
Examination of Potable Waters
",\x vv
Jenef ir D. Isbister, Jeanne L Aim, Roberta Foutch, Alice DeSouza, Randall S.
Wentzel, and Judith F. Kitchens
The report summarized here des-
cribes the development of a method
for concentrating coliphages from
potable waters. The method is used in
conjunction with a rapid coliphage
analysis technique (ARCAT)* to
detect 1 coliphage/100 ml of water
sample. Seven concentration tech-
niques were evaluated in preliminary
feasibility studies, and three methods
were selected for in-depth study
based on recovery of coliphages, time
required for availability of results, and
ease of performing the techinque.
Two of the three methods selected for
in-depth studies were membrane fil-
tration methods using positively
charged and negatively charged fil-
ters. The third method involved ampli-
fication of coliphages by incubation to
allow for a lytic burst to occur.
After in-depth studies and optimiza-
tion of procedures, membrane filtra-
tion with the use of positively charged
filters was selected as the concentra-
tion method of choice. Positively
charged filters can be used success-
fully to concentrate coliphages pre-
sent in potable waters at a level of 1
plaque forming unit (pfu)/100 ml of
water sample using the ARCAT
procedure. The ARCAT amplification
technique can be used to obtain qual-
* Mention of trade names or commercial products
does not constitute endorsement or recom-
mendation for use.
itative indication of the presence of
coliphage when coliphage numbers
are expected to be < 5/100 ml and
can be run in parallel with the ARCAT/
concentration procedure. Test results
are available in 6.5 hours at an approx-
imate cost of $4.12 per test. ARCAT
results can be used to estimate the
number of fecal and total coliforms in
the water sample or may be used inde-
pendently as an indicator of water
quality.
Examination of the effects of physi-
cal treatment of water indicated that
settling and filtration are effective
means for reducing levels of coli-
phages and total and fecal coliforms in
natural waters. Coliphages appeared
to be less efficiently removed by these
processes than total and fecal
coliforms.
Disinfection studies indicated that
chlorine dioxide was effective as a vir-
icide and bactericide. At ozone levels
achieved in this study, reductions in
coliphages and coliforms were not
consistent. Ozone disinfection effec-
tiveness was both concentration and
exposure-time dependent.
This Project Summary was devel-
oped by EPA's Municipal Environ-
mental Research Laboratory.
Cincinnati, OH, to announce key find-
ings of the research project that is
fully documented in a separate report
of the same title (see Project Report
ordering information at back).
-------
Introduction
Waterborne diseases can be spread
rapidly by bacterial and/or viral con-
taminants in any waters used for
human consumption. Community
water supplies are treated to remove or
reduce these contaminants by such
methods as settling, filtration, and dis-
infection. Although the most probable
number (MPN) and the membrane fil-
tration (MF) procedures for detecting
total coliform bacteria (indicators of
fecal contamination) are standard tests
for determining the quality of treated
water, the minimum time for results
from these procedures is 24 ± 2 hours.
A rapid test to evaluate the sanitary
quality of water could be used to decide
whether or not additional treatment,
such as booster chlorination in the sys-
tem, would be necessary.
Under contracts with the U.S. Depart-
ment of the Army, Atlantic Research
Corporation (ARC) developed a rapid
test to determine the number of bacteri-
ophages of a given bacterial species in a
water sample within 4 to 6 hours. The
detection of coliphages was of particu-
lar interest, and a quantitative relation-
ship between coliform bacteria and
their coliphages (infective for Escheri-
chia coli C) was established.
Field tests of the coliphage method,
ARCAT, were conducted in 1977 and
1978. The quantitative relationship
between coliphages and conforms
established in the earlier studies was
confirmed and shown to be valid for nat-
ural waters sampled throughout the
United States.
Under a National Science Foundation
Grant, the test procedure was simpli-
fied and conditions for coliphage repli-
cation and plaque formation were
optimized. The sensitivity of ARCAT fol-
lowing these modifications was 5 coli-
phages/100 ml of water (based on
plating four 5-ml aliquots of the water
sample). At this sensitivity, the method
was not suitable for determination of
water quality in the potable water
range. The study reported here focused
on increasing the sensitivity of ARCAT
to the level of 1 coliphage/100 ml.
Results and Discussion
The concentration techniques evalu-
ated to increase the sensitivity of
ARCAT to detect 1 coliphage/100 ml of
water sample are listed below:
• Concentration by filtration through
positively charged filters
• Concentration by filtration through
negatively charged filters
• Osmotic filtration (dialysis)
• Controlled lytic bursts
• Adsorption on inorganic salts
• Polymer two-phase separation
• Magnetic filtration
• Electrophoresis
From Table 1, the percent of coli-
phage recoveries achieved by the con-
centration techniques, the relative
complexity of the method, and the time
factor increase over the ARCAT test
alone can be compared.
Three methods were selected for in-
depth study based on recovery of coli-
phages, time required to obtain test
results, and ease of performing the
technique. Concentration using posi-
tively charged and negatively charged
filters yielded good coliphage recover-
ies and added 0.5 hour and 1.5 hours,
respectively, to the time required for
ARCAT. A third method, amplification of
coliphages by incubation to allow a lytic
burst to occur, increased the time
required for availability of test results by
70 minutes.
Concentration of coliphages with
positively and negatively charged filter
methods was initially evaluated with
the use of sterilized water samples
seeded with known concentrations of
stock coliphages. Both filter types dem-
onstrated good adsorption, and coli-
phage recoveries of greater than 70%
were generally obtained. These filtra-
tion techniques were optimized with
respect to coliphage adsorption and elu-
tion conditions. Because of occasional
clogging of the negatively charged fil-
ters and the additional hour required to
concentrate the sample using the nega-
tively charged filter, the positively
charged filter was selected as the filtra-
tion method of choice. The AMF Cuno
Zeta Plus 60S positively charged filter
consistently had the best coliphage
adsorption (80% to 100%) and recover-
ies (>90%). The filter performed best
when the water sample was adjusted to
pH 6.0 before filtration. Recoveries of
the coliphages from the filter were con-
sistently high when 3% Trypticase Soy
Broth adjusted to pH 8.5 was used as
the eluent.
Concentration of coliphages from
natural waters using the Zeta Plus 60S
filter was evaluated. Coliphage recover-
ies from the positively charged filter
were compared with coliphages
detected by plating 100 ml of the water
sample in the ARCAT procedure. Table
2 shows coliphage recoveries from nat-
ural water samples; percentages reco-
vered were more variable than those
obtained using samples spiked with T2
and 0174 coliphages (93% to 100%).
The types and densities of coliphages
present in the water samples were a
factor in the variable recoveries. A
severe drought during much of the
Table 1. Comparison of Concentration Methods
Concentration
Technique
Additional"
Time Required
% Coliphage
Recovered
Complexity of
Technique b
Positive filters
SOS
70C
1MDS
Negative filters
Cox 2 fjm
Cox 0.45 fjm
Bio-Rad 0.2 fjm
Osmotic filtration
Lytic burst
Adsorption on
inorganic salts
Polymer two-phase
separation
Magnetic filtration
30 min
40 min
15 min
15 min
30-60 mm
>2hr
8hrc
1 hr
2hr
18-24 hr
1 hr
13-73
27-65
42-108
3-47
0-125
0-71
0-1 SO
0-9
100
0-37.5
2
2
2
3
3
3
2
2
4
4
* Time in addition to that required for ARCAT. (Includes the set-up and performance
of the concentration techniques.)
b Comparedwith ARCATat complexityof1'. Complexity increased by additional manip-
ulation of samples and materials preparation or increased length of time needed for
processing.
c At least 8 hours were required to achieve a 2-fold concentration of 100 ml.
" This technique involves amplification of bacteriophages present, not recovery by
concentration.
-------
Table 2. Correlation of Coliphage Results from ARC AT and Zeta
Plus 60S Concentration/ ARCAT Methods.
Date Sample
8/20 Occoquan raw
8/26 Accotink
Holmes run
8/27 Accotink
8/28 Potomac raw
8/29 Accotink
9/02 Cherokee
Cherokee
9/04 Potomac raw
Potomac raw
Potomac settled
9/08 Cherokee
9/09 Potomac raw
Potomac raw
Potomac raw
Potomac settled
Potomac settled
Potomac settled
9/15 Potomac raw
Potomac raw
Potomac raw
9/15 Potomac filtered
9/16 Potomac raw
Potomac raw
Potomac settled
Potomac settled
9/17 Potomac raw
Potomac raw
Potomac settled
Potomac settled
9/18 Potomac raw
Potomac raw
Potomac settled
Potomac settled
9/19 Potomac raw
Potomac settled
Potomac settled
9/22 Potomac settled
Potomac filtered
9/23 Potomac raw
Potomac settled
" Plaque forming units.
Volume
Filtered
(ml)
750
500
500
500
500
500
500
100
500
100
1000
500
10O
500
WOO
100
500
1000
100
500
WOO
500
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
200
200
Coliphage pfu
ARCAT "
3
6
1
4
60
6
8
8
41
41
14
3
27
27
27
8
8
8
21
21
21
4
15
15
2
2
;5
15
5
5
807
907
295
295
241
72
72
3
1
9
5
Eluatec '
7
39
12
19
337
16
58
15
143
22
29
3
26
111
162
1
15
141
77
258
335
22
10
15*
1
1*
21
10*
5
8*
620
391*
295
159*
239
19
9*
3
2
40
6
Coliphage
% Recovered d
31.1
130.0
240.0
95.0
112.0
53.3
145.0
187.5
69.8
53.6
20.7
20.0
96.3
82.2
60.0
12.5
37.5
176.3
366.7
245.7
159.9
110.0
66.7
100.0
50.0
50.0
140.0
66.7
100.0
160.0
76.8
48.4
100.0
53.9
99.2
26.4
12.5
100.0
200.0
222.2
120.0
b Coliphage pfu/100 ml determined by A RCA T.
c Coliphage pfu in eluate. "*" indicates 10ml eluate volume plated; all others were 20
ml.
" Eluate pfu
x 100.
ARCAT pfu (multiplication factor)
Multiplication factor is number of 100 ml volumes equal to
for concentration.
original volume filtered
initial concentration of coliphages fil-
tered. The accuracy of the Zeta Plus
concentration method in predicting col-
iphages present in the original water
sample is illustrated in Table 3; calcula-
tions were made using the regression
line equation: Ypred =1.2853 X - 3.61.
Detection of low level (1 to 5 pfu/100
ml) coliphage contamination in water
samples using ARCAT without concen-
tration was done by amplifying the coli-
phage(s) present in the sample.
Nutrient addition and addition of the £
coliC host to a final concentration of 1 x
108 cells/ml in the amplification flask
were required. Incubation of the ampli-
fication sample (100 ml, 200 ml) for 70
minutes at 35 °C before plating in
ARCAT provided optimum amplification
of coliphages for which E. coli C is a
suitable host. The amplification tech-
nique was most valuable when used in
conjunction with the standard ARCAT
procedure for waters expected to con-
tain less than 5 coliphages/ 100 ml (the
detection limit for the standard ARCAT
procedure). If coliphages were not
found by the standard ARCAT proce-
dure, the amplification technique
allowed the user to determine if any
coliphages were present. Thus, the
amplification technique yielded only
qualitative results at the 1 to 5 coli-
phage/ 100 ml level.
Physical treatment of water (settling,
• filtration) reduced coliphages, total coli-
forms, and fecal coliforms by 90% to
100%. In a brief study of the effects of
ultraviolet light, chlorine dioxide, and
ozone disinfectants, coliphages
appeared to be as sensitive to ultraviolet
light as were total coliforms and fecal
coliforms. At a concentration of 5 mg/L,
chlorine dioxide was effective as a vir-
icide and in reducingfecal and total coli-
forms. Reduction of coliphages, total
coliforms, and fecal coliforms by expo-
sure to ozone was time and ozone-
concentration dependent. Bacterio-
phage and coliform reductions were not
consistent at the ozone concentrations
achieved during this study.
The ARCAT concentration procedure
for detecting 1 coliphage/100 ml of
water is given below:
study period resulted in unusually low
or variable coliphage occurrences com-
pared to previous nondrought periods.
Least squares linear regression anal-
ysis of the data shown in Table 2 was
used to determine how well the Zeta
Plus concentration method predicted
the number of coliphages originally
present in the natural water sample.
The regression (Figure 1) was highly
significant with a coefficient of determi-
nation of 0.8516, indicating that more
than 85% of the variation in the coli-
phage determination was related to the
Adjust water sample (100 to 1000
ml) to pH 6.0 with 0.1 M or 1.0
M HCI
Filter water sample at pH 6.0
through the Zeta Plus 60S filter
with vacuum
Remove flask containing the filtrate
and replace with a sterile flask
-------
500 -
/? = 0.5225
R2 = 0.8516
Slope 1,2853
Intercept = -3.61
100 200 300
X ( recovered)
400
500
Figure 1.
Least squares regression line ofcoliphages recovered versus coliphages
filtered.
Table 3. Predicted Coliphage Versus Coliphage Applied.
Coliphage (pfu)
Detected by
ARCAT Concentration
m
1
2
3
5
6
8
10
12
15
16
21
22
Coliphage
(pfu)
Predicted a
(Y)
—
—
0.23
2.8
4.1
6.6
9.2
11.75
15.59
16.87
23.3
24.6
Coliphage
(Pfu)
Applied
2,2
1
3
5
10
5
15. 15
5
8. 15
30
15
20.41
a The sensitivity of the method is > 3 pfu applied to the filter.
b Coliphage applied was determined by plating 100 ml in the regular ARCAT
procedure.
• Apply 10 ml of trypticase soy broth
at pH 8.5 to the filter, and
allow a contact time of 5
minutes before vacuum is
applied
• Apply a second 10 ml aliquot of TSB
at pH 8.5 to the filter
• Wash the filter with three 1-rrtl
aliquots of sterile distilled water
• Divide the total eluate (20 ml)
between four tubes each contain-
ing 5.5 ml of molten modified
nutrient agar
• Add 1 ml of E. coli C host to each
tube
• Mix the contents of each tube and
decant into a sterile Petri dish;
allow agar to solidify and incu-
bate at 35° C for 6 hours
• Enumerate plaques at 6 hours
Results
In summary, the use of the Zeta Plus
60S positively charged filter to concen-
trate low levels of coliphages in a water
sample provided the increase in ARCAT
test sensitivity needed for routine
examination of potable waters for coli-
phages. Test results were obtained in
6.5 hours at an approximate cost of
$4.12 per test. The ARCAT results can
be used to estimate the number of fecal
and total coliforms in the water sample
or may be used independently as an
indicator of water quality.
Recommendations
1. ARCAT sensitivity has been
increased using the Zeta Plus 60S con-
centration method for determination of
water quality in the potable water
range. This method should be field
tested in water quality laboratories in
various geographical areas in the Uni-
ted States. Laboratories selected for
testing should examine a wide variety of
water samples, treatment processes,
and testing expertise. Each laboratory
should evaluate at least 50 samples
during a 1 -year period. Data for each
sample should include: 1) water pH and
turbidity, 2) sample volume, 3) number
of coliphages detected, 4) number of
fecal and total coliforms detected, and
5) virus data, if available. Data should be
collected and evaluated at a central
laboratory.
2. Amplification of coliphages in nat-
ural water samples can detect low level
contamination but does not allow for
quantification of the actual coliphages
present in the original sample. Further
study of coliphage morphology, plaque
size, burst size, and the significance of
-------
the presence of any one coliphage type
in a water sample is needed. The types
of coliphages present in natural waters
must first be isolated and identified,
however. Correlation of specific coli-
phages with epidemiological informa-
tion should yield a predictive test for
waterbone disease outbreak.
3. Coliphages survive physical water
treatment processes as well as or per-
haps better than coliform bacteria. A
study to correlate enteric virus and coli-
phage survival through water treatment
processes is recommended. One or two
enteric viruses should be selected for
initial studies to compare survival of the
viruses with that of the coliphages fol-
lowing water treatment processes
under controlled conditions. Survival of
the viruses and coliphages should then
be evaluated under actual treatment
conditions.
The full report was submitted in ful-
fillment of Contract No. 68-03-2914 by
the Atlantic Research Corporation,
under the sponsorship of the U.S. Envir-
onmental Protection Agency.
J. D. Isbister. J. L Aim. R. Foutch, A. DeSouza, R. S. Wentsel. andJ. F. Kitchens
are with the Atlantic Research Corporation, Alexandria. VA 22134.
Donald J. Reasoner is the EPA Project Officer (see below).
The complete report, entitled "Increasing ARC A T® Test Sensitivity for Examina-
tion of Potable Waters," (Order No. PB 82-196 163; Cost: $12.00. subjectto
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:
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
OUSGPO: 1982 — 559-092/3404
-------
Untied States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
RETURN POSTAGE GUARANTEED
PS 0000329
U S ENVIR PROTECTION AGENCY
REGION 5 LIBRARY
230 S DEARBORN STREET
CHICAGO IL 60604
------- |