VALIDATION
OF
METHODS
TO
DETECT
COLIPHAGES
IN
GROUNDWATER
Prepared
for
Health
and
Ecological
Criteria
Division
Office
of
Science
&
Technology
Nena
Nwachuku,
Ph.
D.,
Work
Assignment
Manager
Prepared
by
Mark
Sobsey1
Sagar
Goyal2,
Aaron
Margolin3,
and
Suresh
Pillai4
With
assistance
from:

Nicola
Ballester3,
Baldev
R.
Gulati2,
Sigrun
Haugerud2,
Gregory
Lovelace1
Sunil
Maherchandani2,
Yashpal
Malik2,
Dorothy
Thompson1,
and
Douglas
Wait1
1University
of
North
Carolina,
CB#
7431,
McGavran­
Greenberg
Hall,
Chapel
Hill,
NC
27599­
7431
2University
of
Minnesota,
Department
of
Veterinary
Diagnostic
Medicine,
College
of
Veterinary
Medicine,
1333
Gortner
Avenue,
St.
Paul,
MN
55108
3University
of
New
Hampshire,
Department
of
Microbiology,
Rudman
Hall.
46
College
Rd.
Durham,
NH
03824
4Texas
A&
M
University,
Poultry
Science
Department
418
D
Kleberg
Center,
MS
2472,
College
Station,
Texas
77843­
2472
2
Table
of
Contents
Cover
Page
1
Table
of
Contents
2
INTRODUCTION
AND
BACKGROUND
5
Background
5
Experimental
Approach
6
Coliphages
and
their
Detection
Methods
6
EPA
Methods
for
Coliphage
Detection
in
Ground
water
8
Method
1601
9
Method
1602
10
Confirmation
of
Positive
Results
by
Methods
1601
and
1602
12
Simultaneous
Detection
of
both
Somatic
and
Male­
specific
Coliphages
on
a
Single
Host
12
Survival
of
Coliphages
in
Groundwater
13
Field
Application
of
Methods
1601
and
1602
to
Detection
of
Coliphages
as
Indicators
of
Fecal
Contamination
in
Vulnerable
Groundwater
14
PHASE
I
STUDIES
15
PURPOSES,
GOALS,
AND
TASKS
OF
PHASE
I
STUDIES
15
PHASE
I
METHODS
AND
MATERIALS
18
Method
1601
18
3
Method
1602
19
PHASE
I
RESULTS
20
Coliphage
Recovery
by
Method
1602
20
Confirmation
of
Plaques
in
Method
1602
25
Summary
of
Method
1602
Results
29
Coliphage
Recovery
by
Method
1601
30
Summary
of
Method
1601
Results
36
PHASE
I
CONCLUSIONS
38
PHASE
II
40
Statement
of
Work:
Coliphage
Method
1601
and
1602
Validation
and
Field
Testing
40
Background
40
Purpose
and
Objectives
of
the
Study
41
Specific
Objectives
41
Schedule
of
Deliverables
45
PHASE
II
METHODS
AND
MATERIALS
48
Groundwater
Samples
and
Wells
48
Coliphage
Analysis
of
Groundwater
52
Coliphage
Isolate
Characterization
53
Bacteriological
Analysis
of
Groundwater
55
Analysis
of
Groundwater
for
Human
Enteric
Viruses
57
Primary
virus
concentration
from
groundwater
57
4
Virus
isolation
in
cell
cultures
58
Virus
detection
by
nucleic
acid
amplification
61
Nucleic
acid
amplification
by
(
RT­)
PCR
62
PHASE
II
RESULTS
AND
DISCUSSION
75
Introduction
75
Results
of
Field
Sample
Analysis
of
Coliphage
and
Bacterial
Indicators
in
Groundwater
75
Comparative
Detection
of
Two
Indicators
in
Groundwater
Samples
80
Statistical
Comparisons
of
Fecal
Indicators
in
Groundwater
Samples
89
Analysis
for
Enteric
Viruses
in
Groundwater
91
Survival
of
Coliphage
in
Seeded
Groundwater
92
Comparison
of
Coliphage,
Bacterial
Indicator
and
Enteric
Virus
Detection
in
This
Study
and
in
Previous
Studies
in
the
USA
95
Responses
to
Questions
and
Comments
of
the
April
2004
Coliphage
Workshop
98
SUMMARY
AND
CONCLUSIONS
102
REFERENCES
105
APPENDIX
I:
Summary
Report
of
the
Northeast
Region
108
APPENDIX
II:
Summary
Report
of
the
Southwest
Region
122
APPENDIX
III:
Summary
Report
of
the
Upper
Midwest
135
5
INTRODUCTION
AND
BACKGROUND
Background
This
report
summarizes
the
results
of
Phase
I
and
phase
studies
on
a
project
to
evaluate
and,
if
necessary
further
improve,
EPA
Methods
1601
and
1602
to
detect
coliphages
in
groundwater.
In
the
first
phase
of
the
study
samples
of
groundwater
were
seeded
with
known
quantities
of
naturally
occurring
coliphages
from
sewage
and
the
recovery
efficiency
of
the
methods
in
detecting
these
added
coliphages
was
determined
in
a
series
of
controlled
experiments
performed
concurrently
by
four
participating
laboratories
located
in
different
regions
of
the
country.
The
data
from
the
seeded
sample
recovery
experiments
were
used
to
further
establish
and
quantify
the
performance
characteristics
of
the
methods.

In
the
second
phase
of
the
study
EPA
Methods
1601
and
1602
were
applied
to
geographically
representative
samples
of
groundwater
potentially
vulnerable
to
fecal
contamination
in
order
to
compare
the
performance
of
the
different
coliphage
methods
and
to
compare
their
ability
to
detect
fecally
contaminated
groundwater
relative
to
the
detection
of
fecal
indicator
bacteria
and
the
detection
of
culturable
enteric
viruses.
Each
of
the
four
geographically
representative
laboratories
(
southeast,
northeast,
upper
Midwest
and
southwest)
was
to
analyze
at
least
16
groundwater
samples
for
coliphages,
indicator
bacteria
and
enteric
viruses,
for
a
total
target
number
of
64
samples
to
analyzed
for
second
and
final
phase
of
the
project.
Several
other
tasks
were
linked
to
this
effort
to
further
validate
and
improve
the
coliphage
methods
and
their
ability
to
detect
and
characterize
coliphages
in
groundwater
6
Experimental
Approach
Coliphages
and
their
detection
methods.
Coliphages
are
viruses
infecting
Escherichia
coli
bacteria.
Coliphages
are
present
at
high
concentrations
in
sewage
and
other
fecal
wastes
and
they
are
indicators
of
fecal
contamination
of
groundwater,
other
waters
and
other
environmental
media.
There
are
two
main
groups
of
coliphages:
somatic
and
male­
specific.
The
relationships
between
these
coliphages
and
their
host
bacteria,
showing
specific
bacterial
strains
as
examples,

are
summarized
in
Figure
1.
The
conventional
method
to
detect
coliphages
is
by
their
ability
to
infect
host
cells
in
which
they
replicate
(
proliferate),
producing
large
numbers
of
progeny
viruses
and
lysing
(
killing)
the
host
cells
in
the
process.
It
is
this
killing
and
lysis
of
host
cells
that
forms
the
basis
of
most
coliphage
infectivity
assay
methods,
including
those
employed
for
coliphage
analysis
by
the
EPA
methods.

Figure
1.
Somatic
and
Male­
specific
(
F+)
Coliphages
and
their
Relationship
to
Host
Bacteria
7
Somatic
coliphages
infect
host
bacteria
by
attaching
directly
to
the
outer
cell
wall
(
outer
cell
membrane).
The
male­
specific
coliphages
infect
only
male
F+
strains
of
bacteria
by
attaching
to
the
hair­
like
appendages
projecting
from
the
cell
surface,
called
F­
pili
or
fimbrae,
that
are
the
characteristic
male
trait.
Somatic
coliphage
hosts
lack
the
F­
pili
and
cannot
be
infected
by
F+

coliphages.
F+
coliphage
hosts
differ
in
their
ability
to
be
infected
by
somatic
coliphages.
Some
F+
coliphage
hosts
are
very
resistant
to
somatic
coliphage
infection
because
they
have
an
outer
cell
membrane
that
differs
from
those
of
E.
coli
(
such
as
the
Salmonella
typhimurium
strain
WG49)
and
E.
coli
Famp
(
which
was
experimentally
selected
as
a
somatic­
coliphage
resistant
mutant).
Other
F+
coliphage
hosts
such
as
E.
coli
C3000
have
not
been
subjected
to
selection
for
resistance
to
somatic
coliphages
and
are
susceptible
to
F+
coliphage
infection
as
well
as
somatic
coliphage
infection.
Therefore,
some
host
bacteria
are
infected
only
by
somatic
coliphages
(
E.

coli
C
and
CN13),
others
only
by
male­
specific
coliphages
(
E.
coli
Famp
and
Salmonella
typhimurium
WG49)
and
yet
others
by
both
groups
of
coliphages
(
E.
coli
C3000).

There
are
still
questions
about
which
groups
of
coliphages,
somatic,
male­
specific
or
both
groups
together,
are
the
appropriate
and
preferred
indicators
of
fecal
contamination.
There
is
evidence
in
support
of
both
somatic
and
male­
specific
coliphages
as
being
effective
and
useful
virus
indicators
of
fecal
contamination.
Some
have
suggested
that
both
somatic
and
male­
specific
coliphages
should
be
detected
as
fecal
indicator
viruses
of
contamination
of
groundwater
and
other
waters.

It
is
the
understanding
of
the
authors
that
EPA
has
so
far
not
made
any
final
decisions
about
which
of
the
coliphage
groups
to
target
for
detection
in
future
guidelines
or
regulations.
It
also
been
suggested
that
both
groups
of
coliphages,
somatic
and
male­
specific,
could
be
8
simultaneously
detected
on
a
single
coliphage
host,
thereby
giving
the
greatest
probability
and
highest
sensitivity
in
detecting
any
coliphage
indicative
of
fecal
contamination.

EPA
Methods
for
Coliphage
Detection
in
Ground
Water.
EPA
Methods
1601
and
1602
were
developed
to
detect
somatic
and
male
specific
coliphages
in
large
volumes
of
groundwater,
with
target
sample
volumes
of
up
to
1000
mL
in
Method
1601
(
an
enrichment
method)
and
100
mL
in
method
1602
(
a
Single
Agar
Layer
plaque
assay
method),
respectively.
The
methods
are
based
upon
the
ability
of
the
coliphages
to
infect
host
bacteria,
which
results
in
the
lysis
of
the
host
bacteria.
This
a
widely
used
approach
to
detect
coliphages.
In
plaque
assays
or
other
assays
on
solid
media,
such
as
those
containing
agar,
the
lysis
of
the
host
bacteria
is
visualized
as
zones
of
lysis
or
clearing
of
the
bacteria
as
discrete,
circular
areas
(
called
a
lysis
zones
or
plaques)
in
a
confluent
layer
(
or
"
lawn")
of
host
bacteria
in
a
solid
nutrient
medium.
In
liquid
enrichment
cultures
in
broth
media,
the
lysis
of
host
bacteria
can
in
principle
be
observed
as
the
clearing
of
turbidity
from
the
culture
as
the
bacteria
are
lysed
and
their
resulting
cell
debris
settles
out
of
suspension.
Because
such
clearing
of
broth
cultures
as
evidence
of
host
cell
lysis
can
be
hard
to
observe
due
to
interference
from
other
bacteria
that
may
grow
in
the
broth
culture,
other
ways
to
confirm
the
presence
of
phages
are
often
used.
One
of
the
most
common
ways
is
to
take
some
of
the
enrichment
culture
containing
phages,
apply
it
to
a
lawn
of
bacteria
in
an
agar
medium,
and
allow
the
phages
to
infect
and
lyse
the
host
cells
in
the
lawn
to
produce
a
clear
zone
of
lysis
that
can
be
readily
observed.
9
Method
1601.
Method
1601
is
a
so­
called
two­
step
"
enrichment"
method
and
the
steps
of
the
method
are
outlines
in
Figure
2.
In
the
first
step
of
this
method,
liquid
bacterial
media,

magnesium
chloride
(
to
promote
coliphage
attachment
to
the
host
bacteria),
and
the
E.
coli
host
are
added
to
the
water
sample,
making
a
liquid
(
broth)
culture
for
coliphage
infection
of
the
E.

coli
host
bacteria.
After
allowing
for
coliphage
infection
and
lysis
of
the
host
bacteria
during
overnight
incubation,
a
small
volume
(
several
microliters)
of
the
enrichment
culture
is
placed
on
the
surface
of
a
Petri
dish
of
agar
medium
containing
E.
coli
host
bacteria
(
a
spot).
This
is
the
second
step
of
the
method.
If
the
applied
sample
contains
coliphages
able
to
infect
the
host
bacteria,
a
circular
zone
of
host
cell
lysis
(
clearing)
develops
after
several
hours
of
incubation
in
the
spot
where
the
sample
was
applied.
Such
a
lysis
zone
in
the
spot
indicates
coliphage
presence
in
the
enrichment
broth
and
is
a
positive
result.
If
no
such
lysis
zone
develops
in
the
sample
spot
on
the
plate,
the
enrichment
culture
of
the
sample
is
considered
negative
for
coliphages.
10
Figure
2.

Method
1601
 
Two­
Step
Enrichment­
Spot
Plate
Method
for
Coliphage
Presence­
Absence
When
Method
1601
is
applied
to
a
single
sample
volume,
the
analysis
provides
a
determination
of
the
presence
or
absence
of
coliphages
in
the
sample
volume
analyzed.
If
the
method
is
applied
to
multiple
sample
volumes,
each
in
separate
enrichment
cultures,
the
method
is
capable
of
giving
an
estimation
of
the
concentration
of
coliphages
in
the
water
sample,
based
on
which
sample
enrichment
volumes
become
positive
and
negative
for
coliphages.

Method
1602.
EPA
Method
1602
is
a
so­
called
single
agar
layer
method
for
the
enumeration
of
coliphage
plaques
(
discrete
clear
zones
of
lysis
of
host
bacteria)
developing
in
a
culture
of
host
bacteria
in
an
agar
medium
in
a
Petri
dish.
As
shown
in
Figure
3,
a
100­
mL
sample
of
11
groundwater
is
supplemented
with
magnesium
chloride,
host
bacteria
and
then
combined
with
molten
agar
medium.
The
mixture
is
then
distributed
into
Petri
plates,
the
agar
medium
is
allowed
to
solidify
and
the
plates
are
incubated
overnight
for
the
development
of
coliphage
plaques,
which
are
clear,
circular
zones
of
lysis,
each
produced
by
a
separate
or
individual
coliphage.
The
plaques
are
then
counted
to
determine
the
total
number
of
number
of
coliphages
in
the
sample,

assuming
each
plaque
arose
from
an
individual
infectious
coliphage.

Figure
3.

Method
1602
 
Single
Agar
Layer
(
SAL)
Plaque
Assay
Method
for
Coliphage
Enumeration
12
Confirmation
of
Positive
Results
by
Methods
1601
and
1602.
For
both
Methods
1601
and
1602,
EPA
has
proposed
a
method
to
confirm
positive
results.
For
confirmation
of
positive
results,
material
is
removed
(
picked
or
aspirated
with
a
capillary
pipette
or
a
micropipettor)
from
the
lysis
zones
of
enrichment
spots
on
agar
medium­
host
cell
plates
(
Method
1601)
or
from
the
plaques
that
develop
in
agar
medium­
host
cell
plates
of
Method
1602.
The
recovered
material
is
transferred
to
a
small
volume
of
buffered
water,
mixed
briefly,
and
then
a
small
volume
(
several
microliters)
of
the
material
is
placed
("
spotted")
on
the
surface
of
a
Petri
dish
of
agar
medium
containing
E.
coli
host
bacteria.
If
the
applied
material
contains
coliphages
capable
of
infecting
the
host
bacteria,
a
circular
zone
of
host
cell
lysis
develops
after
incubation
(
for
several
hours
or
overnight)
in
the
spot
where
the
sample
was
applied.
Such
a
lysis
zone
in
the
spot
is
indicative
of
coliphage
presence
in
the
material
recovered
from
either
a
lysis
zone
on
the
spot
plate
of
an
enrichment
broth
(
Method
1601)
or
from
the
plaque
of
a
Single
Agar
Layer
plate
(
Method
1602).

If
no
such
lysis
zone
develops
in
the
sample
spot
on
the
confirmation
plate,
the
sample
(
presumptive
lysis
zone
from
an
enrichment
culture
or
presumptive
plaque
from
an
SAL
plate)
is
considered
negative
for
coliphages.

Simultaneous
Detection
of
both
Somatic
and
Male­
specific
Coliphages
on
a
Single
Host.

EPA
Methods
1601
and
1602
were
originally
developed
to
separately
detect
somatic
and
malespecific
coliphages
using
separate
E.
coli
hosts
able
to
support
the
growth
of
only
one
or
the
other
coliphage
group
(
somatic
or
male­
specific,
respectively).
E.
coli
CN13
is
used
to
detect
somatic
coliphages
and
E.
coli
Famp
is
used
to
detect
male­
specific
coliphages
(
Figure
1).
It
was
later
suggested
that
perhaps
a
single
E.
coli
host
could
be
used
to
simultaneously
detect
both
13
somatic
and
male­
specific
coliphages
present
in
a
groundwater
samples
rather
than
having
to
use
two
separate
E.
coli
hosts
to
separately
detect
each
coliphage
group
(
Figure
1).
If
the
presence
of
either
or
both
groups
of
coliphages
indicates
fecal
contamination,
simultaneous
detection
of
both
on
one
host
would
reduce
time,
effort,
materials
and
cost
and
provide
appropriate
data
about
coliphage
presence
in
a
sample.
As
previously
noted,
E.
coli
C3000
is
such
a
host.
However
use
of
a
single
E.
coli
host
bacterium
capable
of
detecting
both
somatic
and
male­
specific
coliphages
had
not
been
adequately
tested
for
its
performance
characteristics
in
previous
studies
on
the
development
and
evaluation
of
Methods
1601
and
1602
and
their
application
to
either
seeded
samples
or
field
samples
of
groundwater.

Survival
of
Coliphages
in
Groundwater.
In
the
development
and
evaluation
of
methods
for
coliphage
detection
in
groundwater,
the
question
has
been
raised
as
to
how
long
samples
can
be
held
before
being
subjected
to
analysis.
It
has
been
suggested
that
samples
may
have
to
be
collected
and
sent
to
a
distant
lab
capable
of
coliphage
analyses,
but
that
the
time
between
sample
collection
and
analysis
may
be
more
than
1
or
2
days.
If
the
sample
holding
time
is
2
or
more
days
will
the
coliphages
still
be
present
and
be
detectable?
To
address
this
question
additional
experiments
were
done
as
an
added
task
in
Phase
II
of
this
study
at
the
request
of
the
EPA
project
manager.
Groundwater
was
seeded
with
known,
low
level
amounts
of
mixed
populations
of
sewage­
derived
coliphages
and
aliquots
of
these
samples
were
subjected
to
coliphage
analysis
by
Methods
1601
and
1602
on
days
0,
2,
3
and
6.
These
assay
days
were
chosen
to
model
those
that
might
be
used
if
samples
were
shipped
to
a
lab
for
coliphage
analysis
and
even
held
overnight
before
analysis
once
received
by
the
lab.
The
resulting
data
on
coliphage
concentrations
were
14
analyzed
to
determine
if
the
coliphages
were
stable
and
still
detectable
for
periods
ranging
from
1
to
6
days.

Field
Application
of
Methods
1601
and
1602
to
Detection
of
Coliphages
as
Indicators
of
Fecal
Contamination
in
Vulnerable
Groundwater.
An
important
test
of
the
newly
developed
EPA
methods
to
detect
coliphages
in
groundwater,
Methods
1601
and
1602,
would
be
to
validate
their
performance
for
coliphage
detection
in
vulnerable
groundwater,
in
comparison
with
the
detection
of
fecal
indicator
bacteria
and
human
enteric
viruses
in
the
same
samples.
Preferably
such
studies
would
apply
the
methods
to
different,
geographically
representative
groundwater
in
order
to
make
sure
that
the
methods
were
not
adversely
affected
by
interfering
constituents
in
the
groundwater,
or
so­
called
"
matrix
effects".
Furthermore,
the
concurrent
detection
of
coliphages
by
Methods
1601
and
1602
in
the
same
groundwater
samples
would
provide
an
opportunity
to
compare
their
relative
detection
sensitivities
and
lower
limits
of
coliphage
detection.
In
addition,

the
concurrent
detection
of
coliphages
as
well
as
fecal
indicator
bacteria
and
enteric
viruses
in
the
same
groundwater
samples
would
make
it
possible
to
determine
if
coliphages
were
as
good
or
better
than
fecal
indicator
bacteria
or
enteric
viruses
in
identify
fecally
contaminated
ground
water.
Such
analysis
would
make
it
possible
to
determine
if
one
of
these
microbe
groups
was
a
superior
indicator
of
fecal
contamination
because
it
was
detected
more
frequently
and/
or
at
higher
concentrations.
Such
analyses
were
done
in
Phase
II
of
this
study.
15
PHASE
I
STUDIES
PURPOSES,
GOALS
AND
TASKS
OF
PHASE
I
STUDIES
The
overall
purposes
and
goals
of
Phase
I
studies
were
to
determine
the
performance
characteristics
of
Methods
1601
and
1602
in
detecting
and
quantifying
somatic
and
male­
specific
coliphages
in
ground
water
samples
seeded
with
known
quantities
of
natural,
mixed
populations
of
coliphages
obtained
from
municipal
sewage.
These
studies
were
done
using
certain
modifications
and
additions
to
Methods
1601
and
1602
in
order
to
address
recommendations
suggested
for
the
methods
after
their
original
development,
evaluation
and
multi­
laboratory
testing.
Specifically,
host
E.
coli
C3000
was
tested
for
simultaneous
detection
and
quantification
of
both
somatic
and
male­
specific
coliphages
in
addition
to
testing
the
methods
with
the
individual
hosts
previously
specified
for
separate
detection
of
somatic
(
E.
coli
C3000)
and
male­
specific
(
E.

coli
Famp)
coliphages.
In
addition,
Methods
1601
and
1602
were
tested
using
the
confirmation
procedure,
as
proposed
by
EPA
for
all
methods
to
detect
microbes
in
ground
water.

The
key
tasks
and
activities
of
the
Phase
I
studies
are
listed
below.

1.
Recruit
a
total
of
4
experienced
laboratories,
each
from
a
different
region
of
the
country,
to
test
Methods
1601
and
1602
using
the
standard
protocols
with
the
modifications
indicated:
a)
include
host
E.
coli
3000
for
simultaneous
detection
of
both
somatic
and
male­
specific
coliphages,
and
b)

include
confirmation
of
presumptive
positive
results
obtained
from
samples.
The
4
laboratories
16
are:

University
of
North
Carolina
(
UNC),
under
the
direction
of
Mark
D.
Sobsey
(
southeast)

University
of
New
Hampshire
(
UNH),
under
the
direction
of
Aaron
Margolin
(
northeast)

Texas
A&
M
University
(
TAMU),
under
the
direction
of
Suresh
Pillai
(
southwest)

Wisconsin
State
Hygiene
Lab
(
WSHL),
under
the
direction
of
David
Battigelli
(
upper
Midwest)

2.
Develop
bench
sheets
(
bench
laboratory
aids
or
protocols
in
easy­
to­
follow
format)
to
be
used
by
analysts
performing
the
methods
in
these
repeated,
weekly
experimental
trials.

3.
Perform
weekly
experimental
tests
(
trials)
of
the
methods
using
the
developed
bench
sheets.

4.
Test
each
method
(
1601
and
1602)
simultaneously
by
the
4
laboratories
on
a
weekly
basis,

using
locally
collected
ground
waters
seeded
with
the
same
stock
of
sewage­
derived
coliphages
prepared
and
distributed
weekly
by
the
lead
or
reference
laboratory
(
UNC)
and
all
three
E.
coli
hosts
(
CN13
for
somatic,
Famp
for
male­
specifics
and
C3000
for
both).

5.
Perform
repeated
trials
of
each
method
and
submit
the
results
to
the
lead
(
UNC)
laboratory
for
compilation
and
data
analysis
in
order
to
develop
and
evaluate
a
sufficient
database
to
characterize
the
performance
of
the
methods.
17
6.
Identify
any
deficiencies
or
limitations
encountered
with
the
methods.
If
possible
within
a
short
time
period
(
no
more
than
a
few
weeks),
devise
and
implement
modifications
or
corrective
measures
to
improve
the
performance
characteristics
of
the
methods.

7.
Based
on
the
compiled
data
from
the
4
laboratories,
determine
if
the
performance
characteristics
of
the
methods
are
of
sufficient
quality
to
recommend
the
use
of
the
methods
to
detect
coliphages
in
ground
water
samples.

8.
Save
(
archive)
representative
coliphages
detected
by
each
method
on
each
E.
coli
host
for
further
characterization
by
the
UNC
laboratory
to
determine
if
the
coliphage
isolates
have
properties
consistent
with
a
fecal
origin.
These
properties
include
bacterial
host
range,
growth
temperatures
and
taxonomic
group
(
sub­
set
of
representative
isolates
only).
18
PHASE
I
METHODS
AND
MATERIALS
The
methods
and
materials
used
in
this
project
are
those
specified
in
the
documents
for
US
EPA
Methods
1601
and
1602.
Stepwise
procedural
steps
in
the
application
of
these
methods
for
the
specific
purpose
of
this
study
are
also
given
in
the
laboratory
bench
sheets
(
laboratory
bench
protocols)
presented
in
the
Appendix
to
this
report.
The
only
departures
or
modifications
to
Methods
1601
and
1602
employed
in
this
study
are:
(
a)
the
addition
of
E.
coli
C3000
as
a
host
bacterium
for
the
simultaneous
detection
of
both
somatic
and
male­
specific
coliphages,
and
(
b)
the
addition
of
the
newly
proposed
confirmation
procedure
for
plaques
from
plates
of
Method
1602
and
from
lysis
zones
of
plates
from
Method
1601.

Method
1601
For
Method
1601,
the
two­
step
enrichment
method,
the
goal
was
for
each
of
the
4
participating
laboratories
to
seed
30+
liters
of
ground
water
with
a
quantity
of
coliphage
stock
(
filtered
sewage)
to
achieve
between
1
and
2
infectious
units
of
coliphages
per
liter
of
water.
The
seeded
water
was
then
aliquotted
into
30
1­
liter
volumes.
Groups
of
10
1­
liter
volumes
were
subjected
to
the
enrichment
assay
method
using
one
of
the
three
host
bacteria,
thereby
testing
each
host
bacterium
for
coliphage
detection
using
10
replicate
1­
liter
volumes
per
host
bacterium
per
weekly
experiment.
As
negative
control
samples,
three
additional
1­
liter
volumes
of
unseeded
ground
water
were
also
subjected
to
coliphage
analysis
by
the
two­
step
enrichment
method
using
each
of
the
three
different
E.
coli
host
bacteria.
As
negative
controls,
these
samples
were
intended
to
demonstrate
no
background
level
of
coliphages
were
present
in
the
ground
water
19
prior
to
seeding
with
sewage­
derived
coliphages.
A
total
of
8
replicate
experiments
were
conducted,
one
experiment
per
week,
between
May
and
July,
2001.

Method
1602
For
Method
1602,
the
single
agar
layer
(
SAL)
method,
the
goal
was
for
each
of
the
4
participating
laboratories
to
seed
replicate
300­
mL
volumes
of
water
with
a
quantity
of
coliphage
stock
(
filtered
sewage)
to
give
about
100
infectious
units
of
coliphages
per
100
mL
of
ground
water.
The
seeded
water
was
aliquotted
as
3
100­
mL
volumes
and
each
of
these
volumes
was
assayed
by
the
single
agar
layer
method
using
one
of
the
three
different
E.
coli
host
bacteria.
As
negative
controls,
3
100­
mL
volumes
of
unseeded
ground
water
were
subjected
to
coliphage
analysis
by
the
SAL
method
using
each
of
the
three
different
host
bacteria.
As
negative
controls,

these
samples
were
intended
to
demonstrate
no
background
level
of
coliphages
were
present
in
the
ground
water
prior
to
seeding
with
sewage­
derived
coliphages.
A
total
of
10
replicate
experiments
were
performed,
once
experiment
per
week,
during
February
and
April,
2001.
20
PHASE
I
RESULTS
Coliphage
Recovery
by
Method
1602
Table
1
shows
the
recovery
of
seeded
coliphages
by
method
1602
(
single
agar
layer
assay)
for
the
total
of
10
successive
trials
performed
weekly.
In
some
initial
weekly
trials
no
data
were
available
from
the
WSLH
laboratory.
This
was
due
to
other
obligations
that
precluded
their
participation.

In
the
interest
of
time,
the
initial
three
experiments
were
performed
among
the
other
three
laboratories
in
order
to
initiate
the
project
and
to
begin
addressing
potential
logistical
issues
of
coordination
among
laboratories.
No
serious
logistical
problems
arose
among
the
three
labs
participating
initially.
This
indicated
a
reliable
system
for
concurrent
method
performance
among
the
labs
using
the
same
coliphage
stocks
prepared
by
UNC
lab
to
seed
test
groundwater.

The
WSLH
also
was
unable
to
participate
another
week
due
to
a
state­
mandated
holiday.
21
Table
1:
Recovery
of
Seeded
Coliphages
in
100­
mL
Groundwater
Samples
by
Method
1602
(
Single
Agar
Layer
Assay)

Date
Host
UNC
TAMU
WSLH
UNH
21­
Feb­
01
C3000
2%
14%
no
data
22%
CN13
80%
90%
no
data
65%
Famp
246%
160%
no
data
24%
27­
Feb­
01
C3000
9%
79%
no
data
131%
CN13
43%
43%
no
data
127%
Famp
22%
15%
no
data
53%
6­
Mar­
01
C3000
9%
6%
no
data
45%
CN13
37%
19%
no
data
50%
Famp
101%
18%
no
data
35%
13­
Mar­
01
C3000
9%
0%
20%
26%
CN13
19%
0%
35%
23%
Famp
72%
0%
28%
33%
20­
Mar­
01
C3000
12%
8%
56%
37%
CN13
20%
9%
94%
39%
Famp
91%
68%
92%
45%
27­
Mar­
01
C3000
4%
8%
39%
81%
CN13
73%
30%
120%
118%
Famp
48%
20%
40%
70%
3­
Apr­
01
C3000
10%
16%
28%
88%
CN13
30%
36%
70%
77%
Famp
49%
47%
67%
72%
10­
Apr­
01
C3000
5%
17%
34%
93%
CN13
21%
63%
64%
96%
Famp
26%
10%
32%
77%
17­
Apr­
01
C3000
44%
no
data
49%
86%
CN13
139%
no
data
77%
84%
Famp
72%
no
data
94%
98%
24­
Apr­
01
C3000
33%
55%
32%
77%
CN13
117%
88%
76%
89%
Famp
36%
33%
26%
84%

The
percent
coliphage
recovery
data
were
subjected
to
Analysis
of
Variance
(
ANOVA)
to
discover
if
there
were
significant
recovery
differences
among
the
hosts
and/
or
the
laboratories
(
Table
2)
.
As
shown
in
Table
2,
there
were
significant
differences
in
recovery
among
the
3
hosts
(
p=
0.00008),
and
significant
differences
in
recovery
among
the
four
labs
(
p=
0.000002).
The
highest
recovery
(
73%)
was
obtained
using
E
coli
CN13,
the
lowest
(
39%)
was
obtained
using
E
coli
C3000
and
an
intermediate
recovery
of
46%
was
obtained
with
E
coli
Famp.
The
differences
among
host
bacteria
were
consistent
(
not
significantly
different)
within
each
laboratory
(
p=
0.42).
22
This
latter
result
suggests
that
recoveries
among
the
three
host
bacteria
are
generally
similar
within
a
lab
and
therefore,
the
hosts
are
equivalent
on
a
within­
lab
basis.
In
other
words,
the
three
different
E.
coli
hosts
will
give
similar
recovery
efficiencies
when
used
by
an
individual
lab
to
analyze
mixed
populations
of
coliphages
of
sewage
(
fecal)
origin
in
a
groundwater
matrix.

Table
2:
Descriptive
Statistics
for
Seeded
Coliphage
Recovery
by
Method
1602
OVERALL
C3000
UNC
TAMU
WSLH
UNH
UNC
TAMU
WSLH
UNH
Mean
49%
35%
56%
68%
14%
23%
37%
69%
Median
35%
19%
49%
75%
9%
14%
34%
79%
Mode
NONE
0%
NONE
NONE
NONE
NONE
NONE
NONE
Std.
Dev.
52%
37%
28%
31%
14%
26%
12%
35%
Var.
27%
14%
8%
10%
2%
7%
2%
12%
Minimum
2%
0%
20%
21%
2%
0%
20%
22%
Maximum
246%
160%
120%
131%
44%
79%
56%
131%
Count
30
27
21
30
10
9
7
10
95%
CI
19%
15%
13%
12%
10%
20%
11%
25%
CN13
Famp
UNC
TAMU
WSLH
UNH
UNC
TAMU
WSLH
UNH
Mean
58%
42%
77%
77%
76%
41%
54%
59%
Median
40%
36%
76%
81%
61%
20%
40%
62%
Mode
NONE
NONE
NONE
NONE
NONE
NONE
NONE
NONE
Std.
Dev.
43%
33%
26%
33%
65%
49%
30%
25%
Var.
18%
11%
7%
11%
43%
24%
9%
6%
Minimum
19%
0%
35%
23%
22%
0%
26%
24%
Maximum
139%
91%
120%
127%
246%
160%
94%
98%
Count
10
9
7
10
10
9
7
10
95%
CI
31%
25%
24%
24%
47%
38%
28%
18%

The
differences
in
coliphage
recovery
efficiency
among
the
laboratory
groups
led
us
to
question
whether
there
were
differences
in
the
groundwater
of
each
region
(
i.
e.,
a
"
matrix"
effect)
which
might
account
for
those
observed
differences
in
recovery.
Further
experiments
were
conducted
using
regional
groundwater
and
additionally
reagent
water
(
as
a
control
measure)
in
an
attempt
to
answer
this
question.
Four
replicate
experiments
were
conducted
and
these
data
are
summarized
in
Table
3,
with
descriptive
statistics
in
Tables
4a,
4b
and
4c
for
hosts
E.
coli
C3000,
CN13
and
23
Famp,
respectively..

Table
3:
Coliphages
Recovery
Efficiency
of
Method
1602
Concurrently
Applied
to
Seeded
Groundwater
and
Reagent
Water
Date
Matrix
Host
UNC
TAMU
WSLH
UNH
3­
Apr­
01
ground
C3000
10%
16%
28%
88%
reagent
C3000
51%
22%
26%
87%
ground
CN13
30%
36%
70%
77%
reagent
CN13
61%
49%
57%
74%
ground
Famp
49%
47%
67%
72%
reagent
Famp
64%
88%
48%
72%
10­
Apr­
01
ground
C3000
5%
17%
34%
93%
reagent
C3000
52%
13%
37%
92%
ground
CN13
21%
63%
64%
96%
reagent
CN13
76%
66%
63%
78%
ground
Famp
26%
10%
32%
77%
reagent
Famp
66%
18%
47%
75%
17­
Apr­
01
ground
C3000
44%
no
data
49%
86%
reagent
C3000
73%
no
data
67%
76%
ground
CN13
139%
no
data
76%
84%
reagent
CN13
135%
no
data
68%
82%
ground
Famp
72%
no
data
94%
98%
reagent
Famp
87%
no
data
138%
97%
24­
Apr­
01
ground
C3000
33%
55%
32%
77%
reagent
C3000
31%
13%
51%
79%
ground
CN13
117%
88%
76%
89%
reagent
CN13
103%
48%
65%
83%
ground
Famp
36%
33%
26%
84%
reagent
Famp
59%
49%
78%
79%
24
Table
4a:
Descriptive
Statistics:
Coliphage
Recovery
by
Method
1602
for
Seeded
Groundwater
and
Reagent
Water
on
Host
E.
coli
C3000
GROUNDWATER
REAGENT
WATER
UNC
TAMU
WSLH
UNH
UNC
TAMU
WSLH
UNH
Mean
23%
29%
36%
86%
52%
16%
45%
84%
Median
21%
17%
33%
87%
52%
13%
44%
83%
Std.
Dev.
18%
22%
9%
7%
17%
5%
18%
7%
Var.
3%
5%
1%
0%
3%
0%
3%
1%
Minimum
5%
16%
28%
77%
31%
13%
26%
76%
Maximum
44%
55%
49%
93%
73%
22%
67%
92%
Count
4
3
4
4
4
3
4
4
95%
CI
29%
55%
14%
11%
27%
13%
28%
12%

Table
4b:
Descriptive
Statistics:
Coliphage
Recovery
by
Method
1602
for
Seeded
Groundwater
and
Reagent
Water
on
Host
E.
coli
CN­
13
GROUNDWATER
REAGENT
WATER
UNC
TAMU
WSLH
UNH
UNC
TAMU
WSLH
UNH
Mean
77%
63%
72%
86%
94%
54%
63%
79%
Median
73%
63%
73%
86%
90%
49%
64%
80%
Std.
Dev.
60%
26%
6%
8%
32%
10%
5%
4%
Var.
36%
7%
0%
1%
11%
1%
0%
0%
Minimum
21%
36%
64%
77%
61%
48%
57%
74%
Maximum
139%
88%
77%
96%
135%
66%
68%
83%
Count
4
3
4
4
4
3
4
4
95%
CI
95%
65%
9%
12%
52%
25%
8%
6%

Table
4c:
Descriptive
Statistics:
Coliphage
Recovery
by
Method
1602
for
Seeded
Groundwater
and
Reagent
Water
on
Host
E.
coli
Famp
GROUND
REAGENT
UNC
TAMU
WSLH
UNH
UNC
TAMU
WSLH
UNH
Mean
46%
30%
55%
83%
69%
52%
78%
81%
Median
43%
33%
49%
81%
65%
49%
63%
77%
Std.
Dev.
20%
19%
32%
11%
12%
35%
42%
11%
Var.
4%
4%
10%
1%
2%
12%
18%
1%
Minimum
26%
10%
26%
72%
59%
18%
47%
72%
Maximum
72%
47%
94%
98%
87%
88%
138%
97%
Count
4
3
4
4
4
3
4
4
95%
CI
32%
47%
51%
18%
20%
87%
67%
18%
25
These
coliphage
recovery
data
were
subjected
to
ANOVA.
No
interaction
effects
were
indicated
(
p>
0.09
in
all
cases),
implying
that
any
observed
factor
differences
were
consistent
throughout
the
experiment.
As
expected,
there
were
significant
differences
in
recovery
among
the
host
bacteria
(
p=
0.00007),
with
recovery
for
E.
coli
C3000
(
45%)
being
lower
than
recoveries
for
E.
coli
CN13
(
73%)
and
E.
coli
Famp
(
63%).
There
were
also
significant
differences
in
recovery
among
the
laboratory
groups
(
p=
0.0000006),
with
UNH
having
the
highest
overall
recovery
(
83%),

followed
by
UNC
(
60%),
WSLH
(
58%),
and
TAMU
showing
the
lowest
overall
recovery
(
40%).

But
there
was
no
significant
difference
due
to
a
possible
matrix
effect
(
p=
0.17),
leaving
unexplained
the
previously
observed
differences
among
the
laboratory
groups.

Confirmation
of
Plaques
in
Method
1602
The
laboratories
also
applied
several
modified
versions
of
a
confirmation
procedure
for
plaques
isolated
from
the
SAL
plates
of
Method
1602
when
applied
to
the
detection
of
coliphages
in
seeded
water
samples.
Plaques
were
picked
from
the
SAL
plates
using
a
variety
of
methods
(
i.
e.,

with
Pasteur
pipettes,
with
Eppendorf
pipettes,
etc.),
resuspended
in
Tryptic
Soy
Broth,
and
spotted
onto
pre­
poured
gridded
plates
of
Tryptic
Soy
Agar
containing
host
bacteria
(
as
used
in
the
Two­
Step
Enrichment
procedure).
The
confirmation
percentages
are
presented
in
Table
5a,

which
summarizes
all
data
by
experiment
date,
lab
and
host
and
in
Table
5b,
and
which
summarizes
the
descriptive
statistics
for
the
plaque
confirmations.
The
results
of
these
attempts
at
coliphage
plaque
confirmation
ranged
from
excellent
(
average
99­
100%
at
UNH)
to
moderate
(
38­
68%
at
UNC).
Overall,
there
are
high
likelihoods
that
the
plaques
detected
on
assay
plates
for
Method
1602
are
indeed
produced
by
coliphages,
with
a
78%
average
or
a
nearly
4
out
of
5
26
plaque
confirmation
rate.
It
is
likely
that
confirmation
rates
can
be
further
improved
to
give
a
greater
confirmation
efficiency
by
minor
modifications
in
the
plaque
recovery
and
re­
spotting
procedure.
27
Table
5a:
Percent
Confirmation
of
Picked
Plaques
Isolated
using
Method
1602
Date
Matrix
Host
UNC
TAMU
WSLH
UNH
21­
Feb­
01
ground
C3000
33%
not
done
95%
100%
ground
CN13
85%
no
data
100%
100%
ground
Famp
15%
not
done
90%
100%
27­
Feb­
01
ground
C3000
13%
not
done
100%
100%
ground
CN13
75%
no
data
80%
100%
ground
Famp
15%
not
done
45%
100%
6­
Mar­
01
ground
C3000
60%
80%
100%
100%
ground
CN13
45%
100%
60%
100%
ground
Famp
30%
88%
67%
100%
13­
Mar­
01
ground
C3000
21%
not
done
100%
100%
ground
CN13
35%
no
data
75%
95%
ground
Famp
15%
not
done
80%
95%
20­
Mar­
01
ground
C3000
78%
no
data
95%
90%
ground
CN13
50%
25%
90%
100%
ground
Famp
50%
0%
80%
95%
27­
Mar­
01
ground
C3000
40%
no
data
100%
95%
ground
CN13
80%
0%
80%
100%
ground
Famp
35%
53%
95%
95%
3­
Apr­
01
ground
C3000
50%
70%
100%
100%
reagent
C3000
65%
90%
No
data
100%
ground
CN13
45%
80%
95%
100%
reagent
CN13
75%
90%
No
data
100%
ground
Famp
0%
20%
100%
100%
reagent
Famp
0%
10%
No
data
100%
10­
Apr­
01
ground
C3000
93%
80%
100%
100%
reagent
C3000
100%
80%
100%
100%
ground
CN13
15%
90%
85%
100%
reagent
CN13
90%
80%
95%
100%
ground
Famp
10%
50%
90%
100%
reagent
Famp
60%
30%
100%
100%
17­
Apr­
01
ground
C3000
80%
not
done
100%
100%
reagent
C3000
95%
not
done
100%
100%
ground
CN13
95%
no
data
85%
100%
reagent
CN13
70%
not
done
100%
100%
ground
Famp
70%
not
done
100%
100%
reagent
Famp
70%
not
done
95%
100%
24­
Apr­
01
ground
C3000
90%
100%
100%
100%
reagent
C3000
90%
100%
100%
100%
ground
CN13
100%
100%
100%
100%
reagent
CN13
95%
100%
100%
100%
ground
Famp
85%
80%
95%
100%
reagent
Famp
80%
90%
95%
100%
28
Table
5b:
Descriptive
Statistics
for
Plaque
Confirmation
of
Coliphages
Isolated
by
Method
1602
(
Percent
Confirmation
Overall
and
by
E.
coli
Host
for
each
Laboratory)

Overall
C3000
UNC
TAMU
WSLH
UNH
UNC
TAMU
WSLH
UNH
Mean
57%
67%
91%
99%
65%
86%
99%
99%
Median
63%
80%
95%
100%
71%
80%
100%
100%
Mode
15%
80%
100%
100%
90%
80%
100%
100%
Std.
Dev.
31%
34%
13%
2%
29%
11%
2%
3%
Variance
10%
11%
2%
0%
9%
1%
0%
0%
Minimum
0%
0%
45%
90%
13%
70%
95%
90%
Maximum
100%
100%
100%
100%
100%
100%
100%
100%
Count
42
25
39
42
14
7
13
14
95%
CI
10%
14%
4%
1%
17%
10%
1%
2%
CN13
Famp
UNC
TAMU
WSLH
UNH
UNC
TAMU
WSLH
UNH
Mean
68%
74%
88%
100%
38%
47%
87%
99%
Median
75%
90%
90%
100%
33%
50%
95%
100%
Mode
75%
100%
100%
100%
15%
NONE
95%
100%
Std.
Dev.
26%
36%
12%
1%
30%
34%
16%
2%
Variance
7%
13%
1%
0%
9%
12%
3%
0%
Minimum
15%
0%
60%
95%
0%
0%
45%
95%
Maximum
100%
100%
100%
100%
85%
90%
100%
100%
Count
14
9
13
14
14
9
13
14
95%
CI
15%
28%
7%
1%
18%
26%
10%
1%

Subsequent
efforts
to
improve
confirmation
rates
for
plaques
picked
from
SAL
plates
or
lysis
zones
picked
from
spot
plates
of
the
enrichment
method
compared
the
original
EPA
conformation
method
described
above
to
a
modified
method.
In
the
modified
method
the
picked
material
from
plaques
or
lysis
zones
was
re­
enriched
by
culturing
in
host
bacteria
again.
The
picked
plaque
or
lysis
zone
material
was
transferred
to
5
mL
of
Tryptic
Soy
Broth
to
which
had
been
added
host
bacteria
and
the
mixture
was
incubated
overnight
at
37
/

C.
Volumes
of
10
microliters
were
withdrawn
from
the
resulting
overnight
enrichments
and
spotted
onto
prepoured
lawns
of
host
cells
in
nutrient
agar
media.
The
spot
plates
were
incubate
at
37
/

C
for
a
minimum
for
4
hours
and
then
observed
for
lysis
zones
indicative
of
coliphage
positivity.
The
results
of
a
side­
by­
side
comparison
of
the
original
confirmation
method
with
the
modified
method
are
summarized
in
29
Table
5c.

Summary
of
Method
1602
Results
Results
from
a
series
of
10
replicate
experiments
by
all
four
participating
laboratories
on
the
performance
of
the
SAL
method
have
been
presented.
The
method
was
applied
to
replicate
100­

mL
volumes
of
groundwater
seeded
with
sewage
coliphages
and
detected
with
each
of
three
E.

coli
host
bacteria.
The
summarized
results
of
these
experiments
(
Table
6)
show
efficient
coliphage
detection
(
average
53%)
and
confirmation
(
average
78%)
in
100­
mL
volumes
of
ground
water.
There
were
differences
in
recoveries
based
on
the
host
used,
and
there
were
unexplained
differences
in
recovery
among
the
laboratories.
Confirmation
of
plaque
isolates
gave
success
rates
ranging
from
moderate
to
excellent
among
labs.
Individual
adaptations
or
modifications
of
confirmation
methods
somewhat
improved
low
confirmation
rates.
Method
1602
gave
generally
acceptable
detection
of
coliphages
in
seeded
ground
water
and
the
majority
of
plaques
detected
by
the
method
could
be
easily
confirmed
by
a
simple
procedure.
Overall,
the
results
of
these
studies
indicate
that
there
is
high
likelihood
of
detecting
even
low
levels
of
coliphages
in
100­
mL
volumes
of
ground
water
using
Method
1602.

Table
6.
Coliphage
Detection
in
100­
mL
Volumes
of
Seeded
Ground
Water
by
Method
1602
Coliphage
Group
Estimated
Phages/
100
mL
Coliphage
Recovery
(%)
Plaque
Confirmation
(%)

Somatic
(
E.
coli
CN13)
100
64
82
Male­
specific
(
E.
coli
Famp)
100
58
68
Both
(
E.
coli
C3000)
100
38
87
30
Coliphage
Recovery
by
Method
1601
The
Two­
Step
Enrichment
(
SAL)
validation
study
consisted
of
8
replicate
experiments
performed
by
the
four
participating
laboratories.
In
each
experiment
a
small
volume
of
the
assayed
sewage
was
added
to
a
30­
liter
volume
of
groundwater
and
mixed
well
to
disperse
the
inoculum
evenly.

This
inoculated
groundwater
was
then
dispensed
into
30
1­
liter
bottles
to
which
were
added
the
enrichment
media
and
the
proper
host
bacteria
(
10
bottles
per
host).
The
bottles
were
incubated
overnight,
and
small
portions
were
spotted
onto
gridded
TSA
plates
as
described
above.
After
incubation,
these
plates
were
examined
for
zones
of
lysis.
Positive
zones
of
lysis
were
considered
positive
for
coliphage,
and
these
were
counted
and
recorded
for
each
host.
Based
on
the
volume
and
the
titer
of
the
inoculated
sewage,
an
expected
coliphage
titer
per
bottle
was
calculated
for
each
host.
Based
on
the
number
of
positive
bottles,
the
MPN
(
Most
Probable
Number)
of
coliphages
per
bottle
was
calculated
using
Thomas's
MPN
equation
and
taken
as
the
number
of
coliphages
recovered
(
observed
number
of
coliphages).
Using
this
observed
MPN
and
the
expected
coliphage
titer
per
bottle,
percent
recoveries
were
calculated.

The
coliphage
recovery
rates
for
experiments
in
which
replicate
ten
1­
liter
volumes
of
groundwater
were
seeded
with
about
1.5
to
3
infectious
units
of
coliphages
are
presented
in
Table
7a
as
percentage
recoveries
based
on
the
observed
(
calculated)
MPN
coliphage
concentrations
per
liter
and
in
Table
7b
as
a
comparison
of
expected
and
observed
number
of
coliphage­
positive
1­

liter
samples
out
of
10.
Descriptive
statistics
presented
in
Tables
8a
and
8b.
31
Table
7a:
Recovery
of
Seeded
Coliphages
in
1­
Liter
Groundwater
Samples
by
Method
1601
(
Two­
step
enrichment)

Date
Host
UNC
TAMU
WSLH
UNH
9­
May­
01
C3000
4%
no
data
13%
>
63%
CN13
134%
no
data
>
122%
131%
Famp
26%
no
data
19%
>
70%
14­
May­
01
C3000
30%
281%
6%
62%
CN13
193%
92%
>
189%
9%
Famp
>
39%
348%
7%
>
21%
21­
May­
01
C3000
10%
no
data
134%
no
data
CN13
101%
no
data
148%
no
data
Famp
26%
no
data
35%
no
data
4­
Jun­
01
C3000
41%
no
data
23%
122%
CN13
372%
no
data
>
359%
22%
Famp
143%
no
data
62%
99%
11­
Jun­
01
C3000
20%
79%
24%
82%
CN13
215%
<=
5%
>
204%
304%
Famp
44%
<=
14%
15%
123%
18­
Jun­
01
C3000
15%
39%
28%
130%
CN13
>
137%
163%
>
124%
77%
Famp
13%
112%
10%
21%
25­
Jun­
01
C3000
no
data
125%
6%
131%
CN13
18%
122%
39%
220%
Famp
no
data
45%
56%
224%
9­
Jul­
01
C3000
261%
220%
24%
108%
CN13
61%
191%
187%
230%
Famp
<=
518%
57%
33%
2317%
32
Table
7b:
Descriptive
Statistics
for
Recovery
of
Seeded
Coliphages
by
Method
1601
C3000
UNC
TAMU
WSLH
UNH
Mean
54%
149%
33%
109%
Median
20%
125%
24%
122%
Std.
Dev.
92%
100%
42%
27%
Variance
84%
100%
18%
7%
Minimum
4%
39%
6%
62%
Maximum
261%
281%
134%
131%
Count
7
5
8
7
95%
CI
85%
124%
35%
25%

CN13
UNC
TAMU
WSLH
UNH
Mean
171%
114%
296%
142%
Median
163%
122%
247%
131%
Std.
Dev.
117%
74%
208%
113%
Variance
136%
55%
431%
127%
Minimum
18%
0%
39%
9%
Maximum
372%
191%
718%
304%
Count
8
5
8
7
95%
CI
97%
92%
174%
104%

Famp
UNC
TAMU
WSLH
UNH
Mean
47%
112%
30%
424%
Median
26%
57%
26%
123%
Std.
Dev.
49%
138%
21%
838%
Variance
24%
190%
4%
7018%
Minimum
0%
0%
7%
21%
Maximum
143%
348%
62%
2317%
Count
7
5
8
7
95%
CI
45%
171%
17%
775%

The
results
in
Tables
7a
and
7b
indicate
that
when
1­
liter
volumes
of
ground
water
seeded
with
1­

2
PFU
of
coliphages
are
analyzed
by
the
enrichment
method,
there
is
a
very
high
likelihood
that
coliphages
will
be
detected
with
relatively
high
efficiency.
Average
coliphage
recoveries
from
8
replicate
trials
per
coliphage
host
per
lab
ranged
were
86%
for
combined
coliphages
on
host
E.

coli
C3000,
181%
for
somatic
coliphages
on
host
E.
coli
CN­
13,
and
153%
for
male­
specific
coliphages
on
host
E.
coli
Famp.
The
results
in
Tables
7a
and
7b
indicate
variability
in
coliphage
recoveries
from
trial
to
trial.
However,
this
extent
of
variability
is
to
be
expected
because
33
coliphage
recoveries
are
based
on
MPN
estimates
of
the
number
of
positive
1­
liter
enrichment
culture
bottles
out
of
ten.
Based
on
calculated
95%
confidence
intervals
(
CIs),
the
observed
degree
of
variability
was
within
the
range
expected
for
a
10­
culture,
single
dilution
Most
Probable
Number
method.
Examination
of
the
10­
replicate
single
dilution
MPN
table
in
Standard
Methods
for
the
Examination
of
Water
and
Wastewater
indicates
that
MPN
estimates
can
have
95%
CIs
that
vary
by
nearly
6­
fold
(
600%)
at
low
rates
of
positivity
and
almost
always
3­
fold
(
300%)
or
more
at
intermediate
and
high
levels
of
positivity.
Probably
more
important
is
that
of
the
total
82
trials
in
which
10
1­
liter
bottles
of
seeded
ground
water
were
used
per
trial,
coliphages
were
not
detected
in
only
3
trials.
Therefore,
there
is
a
very
high
probability
of
detecting
low
levels
of
(
1
to
3)
coliphages
in
1­
liter
volumes
of
groundwater
when
using
this
method.

As
in
the
statistical
analyses
for
Method
1602
described
earlier,
the
data
for
Method
1601
were
subjected
to
ANOVA.
This
analysis
detected
no
significant
differences
in
recovery
among
the
laboratories
(
p=
0.38);
nor
did
it
detect
any
significant
differences
in
recovery
among
the
hosts
(
p=
0.41).

The
results
from
this
series
of
8
replicate
experiments
per
lab
are
also
summarized
in
Table
8
and
Figure
1
based
on
the
observed
and
expected
number
of
positive
1­
liter
enrichment
culture
bottles
out
of
10.
These
results
indicate
sensitive
coliphage
detection
that
is
close
to
the
theoretical
level
of
detection.
34
Table
8.
Comparison
of
Observed
and
Expected
Coliphage
Detection
in
1­
Liter
Volumes
of
Seeded
Ground
Water
by
Method
1601
Coliphage
Group
Estimated
Coliphages/
L
#
Positive
Bottles
out
of
10
Expected*
Observed
(
Average)
Somatic
(
E.
coli
CN13)
1.7
8.2
(~
8)
4.8
(~
5)
Male­
specific
(
E.
coli
Famp)
3.1
9.5
(
9­
10)
7.1
(~
7)
Both
(
E.
coli
C3000)
1.7
8.5
(
8­
9)
5.7
(~
6)

Figure
4.
Coliphage
Detection
in
1­
Liter
Volumes
of
Seeded
Ground
Water
by
Method
1601
35
Based
on
direct
analysis
(
plaque
assay)
of
the
sewage­
derived
coliphage
stocks
added
to
the
ground
water
samples,
the
average
concentrations
of
coliphages
per
1­
liter
enrichment
bottle
were:
1.7
for
somatic
coliphages
(
detected
on
host
E.
coli
CN13),
3.1
for
male­
specific
coliphages
(
detected
on
E.
coli
Famp)
and
1.9
for
both
groups
of
coliphages
(
detected
on
host
E.

coli
C3000).
According
to
these
estimated
coliphage
concentrations
per
liter
of
seeded
ground
water,
the
expected
numbers
of
positive
enrichment
bottles
out
of
a
total
10
enrichment
bottles
per
coliphage
host
are
computed.
These
estimates
of
expected
numbers
of
positive
enrichment
bottles
out
of
10
were
based
on
the
principles
of
Poisson
statistics
(
the
Poisson
Distribution
and
Poisson
probabilities).
The
estimates
were
computed
directly
from
the
Poisson
probability
equation
using
the
estimated
coliphage
concentrations
per
liter
(
based
on
direct
assay
of
the
sewage
coliphages
seeded
into
groundwater
as
an
estimate
of
the
mean
number
of
coliphages
per
bottle.
The
expected
numbers
of
positive
bottles
out
of
10
were:
about
8
for
somatic
coliphages,

9
to
10
for
male­
specific
coliphages
and
8
to
9
for
both
groups
of
coliphages.
The
experimentally
observed
numbers
of
positive
enrichment
bottles
out
of
10
for
each
group
of
coliphages
are
shown
in
Table
8
and
Figure
1.
Rounded
to
the
nearest
whole
bottle,
the
numbers
of
positive
enrichment
bottles
out
of
10
were:
5
for
somatic
coliphages,
7
for
male­
specific
coliphages
and
6
for
both
groups
of
coliphages.
These
actual
results
indicate
that
the
likelihood
of
detecting
coliphages
when
analyzing
1­
liter
volumes
of
ground
water
containing
only
about
1.5­
3
coliphages
per
liter
by
the
enrichment
method
are
very
high
and
very
close
to
the
theoretically
expected
results.
It
is
noteworthy
that
out
of
a
total
of
87
trials
with
10
1­
liter
enrichment
bottles
per
trial
there
were
only
4
occasions
when
all
10
bottles
were
negative.
This
is
well
within
the
expected
number
of
times
when
all
10
bottles
would
yield
negative
results,
based
on
statistical
36
considerations.

The
expected
probability
of
getting
10
negative
bottles
in
a
10­
bottle
enrichment
test
when
there
is
an
average
of
1.5
to
3
infectious
coliphages
per
bottle
was
actually
higher
than
the
observed
numbers
of
10
negative
enrichment
bottles
in
a
test.
It
should
be
noted
that
the
detection
of
both
somatic
and/
or
male­
specific
coliphages
together
on
host
E.
coli
C3000
was
similar
to
the
detection
of
either
somatic
or
male­
specific
coliphages
on
their
respective
E.
coli
hosts.
Although
the
numbers
of
positive
enrichment
bottles
out
of
10
were
no
higher
on
host
E.
coli
C3000
than
on
the
other
two
hosts,
the
ability
to
simultaneously
detect
both
groups
of
coliphages
using
this
host
was
not
appreciably
different
than
the
detection
of
each
coliphage
group
alone.
Based
on
the
number
of
times
that
all
10
bottles
in
an
enrichment
test
were
negative,
E.
coli
C3000
was
the
same
as
or
better
than
the
other
two
hosts.
The
negativity
rates
per
30
trials
of
the
method
were
1,
1
and
2
per
30
trials
for
E.
coli
C3000
(
both),
CN­
13(
somatic)
and
Famp
(
male
specific)

respectively.
These
results
indicate
that
E.
coli
C3000
can
be
successfully
and
reliably
used
to
simultaneously
detect
low
levels
of
both
somatic
and
male
specific
coliphages
in
1­
liter
volumes
of
ground
water
using
the
two­
step
enrichment
method
(
Method
1601).

Summary
of
Method
1601
Results
In
summary,
recoveries
of
somatic,
male­
specific
and
total
(
somatic
plus
male­
specific)
coliphages
from
1­
liter
volumes
of
ground
water
were
efficient
using
Method
1601.
Coliphage
recoveries
at
input
levels
of
about
1.5
to
3
infectious
units
per
liter
of
ground
water
were
somewhat
variable
but
close
to
those
expected
based
on
the
infectivity
titer
of
the
sewage
seed
and
to
the
expected
37
number
of
positive
1­
liter
enrichment
bottles
out
of
a
total
of
10.
The
observed
variability
of
coliphage
detection
was
no
more
variable
than
would
be
predicted
for
a
10­
sample
MPN
test.

Coliphage
recoveries
were
not
significantly
different
among
E.
coli
hosts
and
participating
labs
using
ANOVA.
Therefore,
there
is
a
high
likelihood
of
detecting
as
few
as
1­
3
coliphages
in
1­

liter
volumes
of
water
using
the
two­
step
enrichment
methods
of
Method
1601.
38
PHASE
I
CONCLUSIONS
In
comparing
the
two
coliphage
recovery
and
detection
methods,
the
participating
laboratory
groups
tended
to
favor
the
Two­
Step
Enrichment
over
the
Single
Agar
Layer
method.
This
is
because
the
former
test
was
considered
easier
to
perform,
sensitive
in
detecting
low
numbers
of
coliphages
and
more
consistent
in
its
results.
The
Single
Agar
Layer
(
SAL)
assay
proved
to
be
cumbersome
when
assaying
multiple
samples,
and
the
time
constraints
imposed
by
the
method
were
difficult
to
adhere
to.
The
Two­
Step
Enrichment
method
is
simpler
to
set
up
and
much
easier
to
carry
out.
The
statistical
analyses
showed
it
to
be
more
consistent
among
different
laboratory
groups.
In
addition,
the
results
for
the
enrichment
method
showed
that
somatic
and
male­
specific
coliphages
can
be
detected
simultaneously
on
a
single
host,
E.
coli
C3000,
at
a
sensitivity
comparable
to
detecting
either
somatic
or
male­
specific
coliphages
individually.
The
simultaneous
detection
of
both
somatic
and
male­
specific
coliphages
simplifies
the
method
as
well
as
lowers
costs.

Further
studies
were
done
to
characterize
the
performance
of
Methods
1601
and
1602
when
applied
to
the
detection
of
coliphages
in
unseeded
samples
of
fecally
contaminated
ground
water
in
Phase
II
of
this
study.
The
presence
and
concentrations
of
coliphages
in
field
samples
of
groundwater
were
compared
and
also
were
to
also
be
compared
to
the
presence
and
concentrations
of
human
enteric
viruses
in
these
fecally
contaminated
ground
water
samples.
This
information
was
to
be
used
to
determine
if
somatic,
male­
specific
and
total
coliphages
are
39
sensitive
and
reliable
indicators
of
fecal
contamination
and
the
presence
of
human
enteric
viruses
in
groundwater.
40
PHASE
II
STUDIES
Statement
of
Work:
Coliphage
Method
1601
and
1602
Validation
and
Field
Testing
Background.
The
US
EPA's
proposed
Ground
Water
Rule
may
propose
the
examination
of
ground
waters
for
coliphages.
Coliphages
have
been
found
to
be
reliable
indicators
of
fecal
contamination
of
ground
and
surface
water
and
of
the
fate
of
human
enteric
viruses
in
the
subsurface
environment.
Recently,
two
different
EPA
methods
were
developed
to
detect
somatic
and
male­
specific
coliphages
in
ground
water.
Method
1601
detects
and
quantifies
coliphages
by
liquid
enrichment
culture
method
and
Method
1602
detects
and
enumerates
coliphages
by
a
single
agar
layer
(
SAL)
plaque
assay.
The
original
methods
round
robin
studies
analyzed
for
somatic
and
male­
specific
coliphages
separately,
and
did
not
evaluate
E.
coli
C3000
as
host
bacterium
for
detection
and
quantification
of
both
somatic
and
male­
specific
coliphage.
The
SAL
method
for
coliphage
detection
did
not
require
a
confirmation
step
for
the
plaques
that
were
observed
in
the
agar­
host
cell
medium.
There
was
some
concern
about
the
detection
of
"
false
positives"
based
on
simply
scoring
plaques.
Therefore,
the
methods
needed
to
be
further
substantiated
and
validated
and
scientifically
supportable
to:
(
1)
show
that
coliphages
detected
by
these
methods
are
of
likely
fecal
origin
and
have
characteristics
and
properties
consistent
with
fecal
origin,
(
2)
reduce
the
cost
and
burden
of
using
two
different
hosts
to
separately
analyze
somatic
and
male­
specific
coliphage
by
measuring
both
simultaneously
in
a
single
E.
coli
host
bacterium,
(
3)
include
a
simple
confirmation
step.
41
Purpose
and
Objectives
of
the
Study.
The
purpose
of
this
study
was
to
further
validate
EPA
method
1601
and
1602
and
to
test
these
methods
in
groundwater
samples
in
four
geographically
representative
regional
laboratories
in
the
USA
(
North
Carolina,
Minnesota,
New
Hampshire
and
Texas).
The
study
was
to
determine
if
coliphages
detected
by
Methods
1601
and
1602
can
be
confirmed
and
show
properties
and
characteristics
consistent
with
a
fecal
origin.
The
labs
that
conducted
the
studies
were
equipped
and
experienced
to
conduct
the
tasks
required
in
the
SOW.

The
study
was
headed
by
a
coliphage
expert
who
has
at
least
20
years
experience
in
coliphage
virology,
who
has
experience
in
round
robin
testing,
in
developing
research
procedures
and
in
method
1601
and
1602.
The
expert
participated
in
the
original
EPA
round
robin
testing
for
method
1601
and
1602,
is
knowledgeable
about
EPA
programs,
and
the
proposed
Ground
Water
Rule.

Specific
Objectives.
The
specific
objectives
of
the
study
were
to:
(
1)
conduct
a
field
validation
of
Methods
1601
and
1602
for
coliphage
detection,
(
2)
determine
the
ability
of
coliphage
indicators
in
predicting
the
presence
of
human
enteric
viruses
in
the
same
ground
water
samples,

(
3)
confirm
the
presence
of
coliphages
detected
in
ground
water
samples,
(
4)
characterize
the
properties
of
these
coliphages
to
confirm
that
they
are
of
likely
fecal
origin,
(
5)
determine
the
correlation
and
the
reliability
of
such
correlation
in
detecting
fecal
contamination
and
the
presence
of
human
enteric
viruses
in
ground
water,
and
(
6)
address
some
key
questions
about
these
coliphage
methods
and
their
use
for
coliphage
detection
in
groundwater
that
arose
in
the
April
2004
EPA
International
Workshop
on
Coliphages
as
Indicators
of
Fecal
Contamination
in
Water
and
Other
Environmental
Media.
An
additional
task
was
added
to
the
study
at
the
request
of
the
42
EPA
project
manager.
This
task
was:
(
7)
to
determine
the
stability
or
survival
of
coliphages
in
groundwater
samples
that
held
for
up
to
several
days
before
analysis
by
Methods
1601
and
1602.

Task
1.
Characterization
and
determination
of
properties
of
confirmed
coliphage
isolates.

A
total
of
800
coliphages
(
200
from
each
of
the
four
participating
laboratories)
were
to
be
characterized
of
which
400
(
100
from
each
laboratory)
were
to
be
from
field
samples
and
400
(
100
from
each
laboratory)
were
to
be
from
sewage­
seeded
ground
water
samples
used
in
methods
validation
studies
in
Phase
I.
The
contractor
was
to
subject
confirmed
coliphage
isolates
to
the
analyses
described
below.

(
a)
bacterial
host
range
analyses.
Determine
the
ability
of
test
coliphage
isolates
to
be
grown
in
both
E.
coli
and
non­
E.
coli
coliform
hosts
and
other
bacteria
by
spotting
onto
pre­
poured
agar
medium­
host
cell
lawns
of
the
following
23
different
host
bacteria,
if
available:
E.
coli
strains
C,

CN13,
C3000,
K12F,
K12F
and
Famp,
S.
typhimurium
WG45
and
WG49,
Klebsiella
pneumoniae
ATCC
strains
23356
and
23357,
Enterobacter
cloacae
ATCC
strain
223355,

Citrobacter
braakii
(
formerly
Citrobacter
freundil)
ATTC
strain6570
ATCC
strain
12012,

Serratia
marcescens
ATCC
strain
14764,
Shigella
sp.
ATCC
23354,
Shigella
flexneri
ATCC
12661,
Yersinia
pseudotuberculosis
ATCC
strain
23207,
Proteus
mirabilis
ATCC
strain
9921,

Yersinia
enterocolitica
ATCC
strains
9610,
29913,
Pseudomonas
aeruginosa
ATCC
s
train
12175,
Aeromonas
hydrophila
ATCC
strain
23211.

(
b)
growth
temperature
range.
Test
the
coliphage
isolates
for
their
ability
to
grow
at
43
temperatures
of
25,
36,
42
and
44.5
/

C
on
E.
coli
hosts.

(
c
)
nucleic
acid
analyses.
Examine
coliphage
isolates
for
taxonomy.
Male
specific
coliphages
were
be
tested
to
determine
the
type
of
nucleic
acid
as
either
DNA
or
RNA.

The
contractor
shall
analyze
32
geographically
representative
ground
water
samples
(
8
per
laboratory
in
four
geographically
representative
laboratories)
for
coliphages
and
human
enteric
viruses
by
combined
cell
culture
and
nucleic
acid
amplification
methods.

Task
2.
Cell
culture
RT­
PCR
or
cell
culture­
PCR.
The
contractor
shall
analyze
32
geographically
representative
ground
water
samples
(
8
per
laboratory
in
four
geographically
representative
laboratories)
for
coliphages
and
human
enteric
viruses
by
combined
cell
culture
and
nucleic
acid
amplification
methods.

Task
3.
Coliphages
and
enteric
viruses
from
groundwater.
Each
of
the
four
participating
laboratories
shall
analyze
an
additional
8
samples
of
ground
water
for
culturable
human
enteric
viruses
and
for
coliphages.
Four
different
ground
waters
shall
be
analyzed
on
two
different
occasions.
A
total
of
36
ground
water
samples
shall
be
analyzed
for
coliphages
and
human
enteric
viruses.
The
extent
to
which
somatic
and
male­
specific
coliphages
detected
by
Methods
1601
(
in
1­
liter
sample
volumes)
and
1602
(
in
0.1­
liter
sample
volumes)
are
associated
with
the
occurrence
of
human
enteric
viruses
in
100­
1000­
liter
sample
volumes
of
fecally
contaminated
ground
water
will
be
determined.
The
contractor
shall
statistically
analyze
data
on
the
occurrence
of
coliphages
44
and
human
enteric
viruses
in
field
ground
water
samples
to
determine
if
there
is
a
co­
occurrence
and
the
extent
to
which
they
co­
occur.

Task
4.
Coliphage
and
bacterial
analyses.
Each
lab
will
sample,
process,
and
analyze
ground
water
by
Methods
1601
and
1602
using
hosts
E.
coli
Famp,
E.
coli
CN­
13
and
E.
coli
C3000
according
to
the
established
methods.
In
parallel
to
the
coliphage
analysis,
each
lab
will
sample
and
process
the
same
fecally
contaminated
ground
waters
by
the
EPA
ICR
methods
for
human
enteric
viruses.
Also,
in
parallel
each
lab
will
sample
and
process
the
same
fecally
contaminated
ground
waters
for
E.
coli
and
enterococci.

Task
5.
Enteric
virus
analyses.
Processed
ground
water
samples
will
be
analyzed
for
culturable
human
enteric
viruses
by
observation
for
cytopathogenic
effects
(
CPE)
in
BGMK
cells
according
to
EPA
ICR
method.
In
addition,
the
inoculated
cell
cultures
also
will
be
examined
for
noncytopathogenic
enteroviruses,
caliciviruses,
adenoviruses,
hepatitis
A,
rotaviruses,
reoviruses
by
combined
cell
culture
and
nucleic
acid
amplification
methods,
as
previously
described.
The
labs
also
will
analyze
the
samples
for
culturable
human
enteric
viruses
in
CaCo2
cells.
The
data
on
the
occurrence
and
concentrations
of
coliphages
and
concentrations
of
human
enteric
viruses
as
detected
by
CPE
and
by
PCR
will
be
statistically
analyzed
to
determine
co­
occurrence
with
coliphages.

Task
6.
Report.
The
contractor
shall
prepare
a
consolidated
draft
report
of
all
the
data
generated
in
all
the
tasks
in
all
4
laboratories
and
statistically
analyzed.
The
report
shall
include
an
45
interpretation
of
the
results
and
recommendations
to
EPA.
The
report
shall
be
submitted
in
3
double
spaced
hard
copies
and
a
3
½
diskette
in
WordPerfect,
version
9/
8.0
for
Windows.
A
summary
fact
sheet
of
the
study
and
results
shall
accompany
the
draft
report.

Task
7.
Peer
review.
The
contractor
shall
incorporate
internal
and
external
peer
review
comments
and
a
workshop
input
comments
in
a
final
report.

The
EPA
Work
Assignment
Manager
will
give
technical
direction
in
this
study.
The
contractor
shall
not
cite,
quote
or
distribute
the
results
of
this
EPA
study
until
EPA
publishes
it.
Publications
from
any
aspect
of
this
EPA
research
study
will
be
subjected
to
EPA
review
and
will
be
published
jointly
with
EPA.
A
monthly
conference
call
shall
be
scheduled
by
the
contractor
until
completion
of
the
study.
An
on
site
visit
will
be
conducted
by
the
EPAWAM
on
a
mutually
acceptable
date
with
the
technical
lead.

Schedule
of
Deliverables
The
deliverables
will
include
a
consolidated
1st
draft
report
with
data
from
all
the
completed
7
tasks
listed
above
and
a
final
peer
review
report.

NOTE:
These
project
objectives
were
not
fully
met
due
to
extenuating
circumstances.

Specifically,
Task
1
could
not
be
completed
due
to
circumstances
beyond
the
control
of
the
project
investigators.
Although
more
than
800
coliphage
isolates
were
obtained
during
this
study,

the
vast
majority
of
the
inventory
of
these
coliphages
were
lost
to
due
failures
of
ultracold
46
freezers
in
which
the
isolates
were
stored
at
the
UNC
laboratories
for
subsequent
characterization.
These
freezer
failures
were
due
to
both
freezer
malfunctions
and
to
power
outages
at
the
laboratories
that
were
due
to
natural
disasters
(
ice
storms
and
hurricanes)
beyond
the
control
of
the
project
investigators.
Furthermore,
these
freezer
failures
caused
the
loss
of
the
majority
of
bacterial
hosts
on
which
the
coliphages
isolates
were
to
be
characterized
as
to
host
range
in
order
to
fulfill
Task
1a.
Most
of
these
bacterial
strains
had
been
previously
purchased
from
the
American
Type
Culture
Collection
and
the
costs
of
replacing
them
were
prohibitive
and
had
not
been
included
in
the
project
budget.

To
compensate
for
the
loss
of
these
coliphage
isolate
samples
and
their
further
characterization,

the
project
labs
undertook
additional
work
in
support
of
meeting
other
project
objectives
and
tasks.
Specifically,
the
participating
labs
analyzed
more
samples
of
groundwater
in
the
Phase
II
studies
than
were
originally
specified.
Contract
specifications
called
for
the
analysis
of
a
total
of
64
samples
(
16
per
laboratory)
for
coliphages,
bacterial
indicators
and
human
enteric
viruses.
The
eventual
number
of
samples
analyzed
was
actually
106
samples
(
27
by
three
laboratories
and
25
by
the
fourth
laboratory).
It
was
believed
that
the
analysis
of
extra
field
samples
would
provide
more
representative
data
for
determining
if
coliphages
were
effective
indicators
of
fecal
contamination
of
groundwater
and
of
human
enteric
viruses.
Additionally,
the
analysis
of
human
enteric
viruses
in
groundwater
samples
was
expanded
to
include
astroviruses,
which
were
not
originally
included
in
the
specifications
for
human
enteric
viruses
to
be
analyzed.
Hence,
this
additional
task
was
taken
on
to
further
improve
the
opportunities
to
detect
human
enteric
viruses
in
groundwater
as
part
of
the
effort
to
obtain
more
definitive
data.
47
At
the
request
of
the
UNC
project
manager,
the
UNC
lab
also
took
an
additional
task
that
was
not
in
the
original
scope
of
work
or
its
budget.
This
additional
task
was
to
determine
the
survival
of
coliphages
in
groundwater
samples
to
be
analyzed
for
viruses
by
Methods
1601
(
two­
step
enrichment
spot
plate)
and
Method
1602
(
SAL
plaque
assay).
Groundwater
samples
seeded
with
mixed
populations
of
sewage­
derived
viruses
were
analyzed
for
coliphages
initially
(
on
day
zero)

and
also
after
2,
3
and
6
days
of
storage
at
4oC.
These
survival
experiments
were
done
to
determine
if
collected
samples
held
for
several
days
prior
to
analysis
due
shipping
and
storage
still
had
most
of
their
initial
coliphages
that
still
could
be
detected
by
Methods
1601
and
1602.
48
PHASE
II
METHODS
AND
MATERIALS
Groundwater
Samples
and
Wells
The
original
goal
of
this
study
was
for
each
of
the
four,
regionally
representative
laboratories
(
southeast,
northeast,
upper
midwest
and
southwest)
to
collect
and
analyze
27
ground
water
samples
from
public
water
supply
wells.
Efforts
were
made
to
identify
candidate
public
water
supplies
that
previously
had
coliform
bacteria
violations
or
other
evidence
of
vulnerability
to
fecal
contamination.
In
some
cases
candidate
wells
were
prescreened
by
bacteriological
and
coliphage
analyses
for
evidence
of
fecal
contamination.
Because
not
all
participating
labs
could
identify
and
get
access
to
27
public
water
supply
wells,
some
labs
also
included
non­
public
and
private
wells
in
their
sampling.
Three
labs
obtained
27
ground
water
samples
and
one
lab
obtained
a
total
of
25
samples
for
a
total
of
106
samples
overall.
The
characteristics
of
the
wells
that
were
sampled
are
presented
below,
by
region.

Southeast.
Of
the
27
wells
in
the
Southeast,
13
were
in
North
Carolina
and
4
were
in
Florida.

The
Florida
wells
were
all
public
water
supply
wells.
Florida
Well
UNC
#
1
is
in
Orange
County,

FL.
There
is
no
history
of
that
well
ever
being
disinfected.
Florida
Well
UNC
#
2
is
in
Orange
County,
Fl.
The
pump
was
taken
out
of
service
for
repairs
(
rebuilt
pump),
and
it
was
disinfected
in
January
­
February
of
2002.
Prior
to
placing
the
well
back
into
production
it
was
disinfected
with
chlorine.
Approximately
30
gallons
of
12%
liquid
bleach
was
placed
into
the
well
for
24
hours
(
100ppm).
Then
water
was
discharged
for
a
minimum
of
4
hours
and
bacteriological
samples
were
taken
to
confirm
their
absence.
Theses
Florida
wells
were
samples
in
June
and
49
September,
2002,
which
was
4
and
7
months
after
Well
#
2
had
been
chlorinated.

Two
other
Florida
wells,
designated
UNC
#
3
and
UNC
#
4
and
located
in
Ocala
County
also
were
sampled.
Both
wells
had
periodic
coliform
positivity
during
the
and
prior
to
the
study
period
(
2002).
They
serve
a
population
of
about
57,000
in
the
Ocala
area.
UNC
#
3
and
#
4
were
not
disinfected
prior
to
or
during
sampling
for
this
study.
However,
the
utility
currently
(
year
2004)
adds
calcium
hypochlorite
(
granular
chlorine)
to
control
the
total
coliforms
they
are
getting
(
and
have
been
getting
a
lot
more
since
the
hurricanes
this
year
­
2004).
The
chlorine
is
now
added
weekly
to
both
wells
to
achieve
the
CT
for
4
log
virus
removal
through
their
treatment
process,
including
chlorination.

There
were
13
wells
in
North
Carolina,
and
the
characteristics
of
these
wells
are
summarized
below.

Type
County
Well
Identification
Private
industrial
Cartaret
BF
Community
water
supply
Cartaret
BMHP
Community
water
supply
Carteret
SB
MHP
Non­
community
water
supply
(
private
campgrounds)
Pamlico
Camp
DL
Non­
community
water
supply
(
private
campgrounds)
Pamlico
Camp
SF
#
1
Non­
community
water
supply
(
private
campground)
Pamlico
Camp
SF
#
2
50
Private
Carteret
TGUMC
Private
Carteret
GL
Private
Pender
SC
Private
Pender
VE
Private
Duplin
RH
Private
Duplin
KC
Private
Duplin
TH
Southwest.
Only
PWS
wells
were
investigated
in
this
study,
and
a
total
of
eleven
different
PWS
wells
were
included.
The
sampling
sites
were
located
in
the
San
Antonio
region
of
Texas
(
wells
RS,
KK,
and
HCR)
and
along
the
US­
Mexico
border
in
southern
New
Mexico
(
wells
MHPa,

MHPb,
MHPc,
FVE,
AVC,
SME,
and
LME).
The
wells
in
the
San
Antonio
region
were
part
of
a
karst
aquifer
and
were
previously
implicated
in
a
documented
groundwater
contamination
event.

Also,
during
the
initial
pre­
screening
of
the
wells
some
of
the
samples
were
positive
for
somatic
and
male­
specific
coliphages.
The
wells
in
southern
New
Mexico
were
identified
as
being
vulnerable
to
groundwater
contamination
based
on
parameters
such
as
closeness
to
septic
tanks,

proximity
to
the
Rio
Grande
river
and
the
aquifer
in
question.
These
wells
were
part
of
a
previous
EPA­
funded
project
on
the
microbiological
quality
of
wells
in
the
shallow
aquifer
along
the
US­
Mexico
border
during
which
some
of
the
wells
in
the
sampling
area
were
positive
for
enterococci,
E
.
coli,
male­
specific
coliphages
and
somatic
coliphages.
The
wells
were
in
the
100­

150
feet
depth
range.
The
static
water
levels
were
around
10­
20
feet
and
in
terms
of
their
hydrogeologic
setting,
they
were
located
in
the
Rio
Grande
alluvium/
Hueco­
Tularosa
aquifers.
51
Groundwater
samples
were
collected
between
June
2002
and
January
2003.
Multiple
samples
were
collected
from
each
of
the
wells
to
be
representative
of
the
aquifer
and
the
sampling
location.
During
each
sampling
adequate
volumes
were
collected
for
the
coliphage
analysis
as
well
as
for
the
enteric
virus
analysis.
Grab
samples
were
collected
for
the
coliphage
and
bacterial
analysis
while
the
1MDS
filters
were
used
for
collecting
the
large
volume
enteric
virus
samples.

Upper
Midwest.
A
total
of
27
groundwater
samples
were
collected
from
25
wells.
Two
wells
were
tested
twice.
Details
of
these
wells
are
provided
in
the
report
of
the
Upper
Midwest
lab,

which
is
in
the
Appendix).
All
wells
except
6
private
ones
in
Minnesota
were
considered
noncommunity
public
water
supplies
by
the
State
of
Minnesota
and
none
were
disinfected.

Noncommunity
transient
public
water
supplies
(
i.
e.,
groundwater)
are
monitored
for
nitrate
and
total
coliform
bacteria
as
required
by
the
SDWA.
Private
water
systems
including
those
places
of
business
not
meeting
the
federal
definition
of
PWS
have
no
long
term
monitoring
requirements.

Northeast.
All
sample
sites
were
located
in
New
England.
Eight
well
sites
were
public
water
sources
and
17
were
private
wells.
A
total
of
25
wells
samples
were
collected
instead
of
27
due
to
a
very
severe
and
harsh
winter.
NH
had
its
first
snowfall
at
the
end
of
October
and
a
second
snowfall
at
the
beginning
of
November,
2002.
Plans
to
sample
two
additional
wells
as
soon
as
the
weather
permitted
could
not
be
carried
out
because
New
England
experienced
one
of
the
snowiest
winters
ever.
Therefore,
only
25
well
samples
were
collected
and
analyzed.
Of
the
25
wells,
there
were
12
sample
sites
in
New
Hampshire,
two
of
which
were
from
public
wells
that
were
approximately
500
and
700
ft
deep,
respectively.
None
of
these
wells
had
any
form
of
52
disinfection.
The
other
wells
from
NH
were
all
private
wells.
These
wells
also
were
not
disinfected.
One
well
from
NH
was
a
private,
very
shallow
well,
less
then
35
feet
deep
and
lined
with
stone.
This
was
not
considered
a
potable
well
but
was
used
for
farm
irrigation.
Four
sites
in
Maine
were
all
privately
owned
wells
and
not
disinfected.
Three
sites
were
in
Vermont,
and
they
were
all
privately
owned
wells
and
not
disinfected.
All
of
the
privately
owned
wells
were
drilled
wells,
excepted
for
the
one
in
NH
as
indicated
above,
and
they
were
of
varying
depths
that
were
unknown
to
the
homeowner
at
the
time
samples
were
collected.
There
were
6
samples
from
public
water
supply
wells
in
Massachusetts.
The
public
water
supplies
in
Massachusetts
were
chosen
due
to
positive
results
previously
found
for
total
and
fecal
coliforms,
enterococci,
and
male­
specific
coliphages.
Additionally
3
of
the
6
locations
had
positives
previously
reported
for
rotavirus
and
enterovirus,
by
molecular
methods.

Coliphage
Analysis
of
Groundwater
Groundwater
samples
were
analyzed
by
Method
1601,
the
two­
step
spot­
plate
enrichment
method
and
by
Method
1602,
the
Single
Agar
Layer
(
SAL)
plaque
assay
using
sample
volumes
of
1
liter
and
100
mL,
respectively,
for
each
target
group
of
coliphages
(
male­
specific,
somatic
and
"
total"
coliphages).
Host
bacteria
for
the
target
groups
of
coliphages
were
E.
coli
CN­
13
for
somatic
coliphages,
E.
coli
Famp
for
F+
coliphages
and
E.
coli
C3000
for
"
total"
(
somatic
plus
F+)
coliphages.
Coliphage
analyses
were
performed
according
to
the
EPA­
approved
methods,

except
lysis
zones
from
enrichment
spot
plates
and
plaques
from
SAL
plates
were
confirmed
using
the
proposed
EPA
confirmation
method.
In
this
method,
material
from
individual
SAL
plaques
or
lysis
zones
on
spot
plates
was
removed
(
aspirated)
with
a
Pasteur
pipette,
micropipette
tip,
or
53
other
device
and
the
recovered
material
was
resuspended
in
0.5
mL
of
tryptic
soy
broth.
These
suspensions
were
held
briefly
for
coliphages
to
diffuse
out
of
the
agar
and
then
the
samples
were
vortex
mixed
vigorously
to
disperse
the
coliphages.
Then,
10:
l
aliquots
were
removed
from
the
suspension
and
spotted
onto
pre­
poured
spot­
plates
of
the
appropriate
E.
coli
host
bacterium
as
in
the
enrichment
procedure.
The
spot­
plates
were
incubated
overnight
and
checked
for
zones
of
lysis.
Any
spots
showing
lysis
were
scored
as
confirmed
coliphages.

Coliphage
Isolate
Characterization
A
total
of
800
coliphages
(
200
from
each
of
the
four
participating
laboratories)
were
to
be
characterized
for
their
properties
to
determine
if
they
were
of
likely
fecal
origin.
For
each
of
the
four
participating
laboratories
100
coliphage
isolates
from
the
phase
I
studies
with
groundwater
samples
seeded
with
sewage­
derived
coliphages
and
another
100
isolates
from
the
unseeded
field
groundwater
samples
of
each
laboratory
were
to
be
subjected
to
characterization
by
bacterial
host
range
analysis,
growth
temperature
range
analysis
and
determination
of
type
of
nucleic
acid
(
for
F+
coliphages).

For
bacterial
host
range
analyses
coliphage
isolates
were
to
be
tested
for
their
ability
to
grow
in
both
E.
coli
and
non­
E.
coli
coliform
hosts
and
other
bacteria
by
spotting
onto
pre­
poured
agar
medium­
host
cell
lawns
of
the
following
23
different
host
bacteria
if
available:
E.
coli
strains
C,

CN13,
C3000,
K12F,
K12F
and
Famp,
S.
typhimurium
WG45
and
WG49,
Klebsiella
pneumoniae
ATCC
strains
23356
and
23357,
Enterobacter
cloacae
ATCC
strain
223355,

Citrobacter
braakii
(
formerly
Citrobacter
freundil)
ATTC
strain6570
ATCC
strain
12012,
54
Serratia
marcescens
ATCC
strain
14764,
Shigella
sp.
ATCC
23354,
Shigella
flexneri
ATCC
12661,
Yersinia
pseudotuberculosis
ATCC
strain
23207,
Proteus
mirabilis
ATCC
strain
9921,

Yersinia
enterocolitica
ATCC
strains
9610,
29913,
Pseudomonas
aeruginosa
ATCC
s
train
12175,
Aeromonas
hydrophila
ATCC
strain
23211.
Spotted
plates
are
incubated
at
370C
overnight
and
observed
for
evidence
of
lysis
of
the
host
bacteria
in
each
spot
as
evidence
of
growth
on
each
host
bacterium.

For
growth
temperature
range
characterization,
coliphage
isolates
were
to
be
tested
for
their
ability
to
grow
at
temperatures
of
25,
36,
42
and
44.5
/

C
on
E.
coli
hosts.
Coliphage
isolates
were
to
be
serially
diluted
10­
fold
and
several
dilution
were
to
be
spotted
in
10
uL
amounts
onto
replicate
pre­
poured
lawns
of
E.
coli
host
bacteria
in
agar
media
Petri
dishes.
Each
replicate
plate
was
to
be
incubated
at
the
aforementioned
temperatures
overnight
and
then
the
spots
n
these
plates
were
to
be
observed
and
quantified
for
coliphage
growth
at
each
of
the
4
test
temperatures.

Coliphage
growth
at
temperatures
of
not
only
36oC
but
also
growth
at
the
temperatures
of
42
and
or
44.5oC
was
considered
evidence
of
thermotolerance
and
of
a
likely
fecal
origin.

For
nucleic
acid
analyses
of
F+
coliphages,
isolates
were
to
be
examined
for
taxonomy
as
F+

DNA
or
F+
RNA
coliphages
using
previously
described
methods
(
Hsu
et
al.,
1995).
Briefly,
10
:
l
volumes
of
F+
coliphage
suspensions
were
to
be
spotted
onto
duplicate
pre­
poured
lawns
of
E.

coli
host
bacteria
in
agar
medium.
One
plate
contained
Rnase
at
100
:
g/
mL
and
the
other
plate
did
not.
Plates
were
to
be
incubated
overnight
at
37oC
and
then
they
were
to
be
observed
for
lysis
or
the
appearance
of
plaques
in
the
spots
of
applied
coliphage
suspensions.
Presence
of
a
lysis
55
zone
or
plaques
in
the
spot
on
the
plate
without
Rnase
and
the
absence
of
such
lysis
or
plaques
in
the
spot
on
the
plate
with
Rnase
were
considered
evidence
of
an
RNA
coliphage.
The
presence
of
lysis
or
plaques
in
the
spots
of
plates
with
and
without
Rnase
was
considered
evidence
of
an
F+

DNA
coliphage.

As
indicated
above,
these
coliphage
characterization
activities
were
not
completed
due
to
extenuating
circumstances
beyond
the
control
of
the
project
investigators.
Ultracold
freezer
failures
caused
the
loss
of
archived
coliphage
isolates
to
be
characterized
and
also
the
loss
of
most
of
the
bacterial
hosts
that
were
to
be
used
for
host
range
characterization
studies
of
these
coliphage
isolates.

Bacteriological
Analysis
of
Groundwater
Field
groundwater
samples
were
analyzed
for
E.
coli
and
enterococci
and
in
some
cases
for
fecal
coliforms
using
EPA­
approved
methods.
For
E.
coli,
some
labs
used
mFC
agar
for
fecal
coliforms,
with
incubation
at
44.5oC
for
20­
22
hours,
followed
by
transfer
or
membranes
to
nutrient
agar­
MUG
medium,
re­
incubation
for
several
hours,
and
observation
for
colonies
fluorescing
blue
under
long­
wavelength
UV
light
as
evidence
of
E.
coli
colonies.
Another
lab
used
mEC
medium
for
E.
coli
with
incubation
at
44.5
oC
(
APHA,
1998).
Another
lab
used
mColiBlue
agar
for
simultaneous
detection
of
total
coliforms
and
E.
coli,
according
to
USEPAapproved
methods.
For
enterococcus
analysis,
labs
used
standard
membrane
filter
methods
and
with
either
modified
ME
agar
or
MEI
agar
and
incubation
conditions
as
specified
in
the
EPA
method
(
APHA,
1995;
Levin
et
al.,
1975;
USEPA,
2002).
Samples
for
E.
coli
analysis
were
100
56
mL,
although
one
lab
also
analyzed
volumes
of
1000
mL.
Data
for
the
results
of
this
larger
1000
mL
volume
were
not
included
in
the
compilation
and
analysis
of
data
for
all
labs,
as
no
other
lab
analyzed
this
volume
and
it
is
not
a
standard
volume
used
for
bacteriological
analysis
of
water.

The
results
for
this
larger
volume
are
in
the
report
from
this
participating
laboratory,
which
is
in
the
Appendix
to
this
report.
57
Analysis
of
Groundwater
for
Human
Enteric
Viruses
Primary
virus
concentration
from
groundwater.
Ground
water
from
candidate
wells
was
sampled
using
the
EPA
ICR
method,
with
minor
modifications
(
USEPA,
1996).
Groundwater
sample
volumes
of
1,500
liters
(
397
gallons)
were
to
be
filtered
through
a
1
MDS
pleated
cartridge
filter
(
CUNO)
at
pH
6­
8.
The
filter
was
eluted
with
1.5%
beef
extract
(
Becton
Dickinson
#
212303)
buffered
with
0.05
M
glycine
at
pH
9.5.
Viruses
in
the
resulting
beef
extract
eluate
were
further
concentrated
by
organic
flocculation
(
acid
precipitation)
as
specified
by
the
ICR
methods.
The
only
significant
change
to
the
ICR
procedure
was
that
the
acid
precipitate
was
resuspended
in
20
about
mL
of
sodium
phosphate
rather
than
30
mL
in
order
to
reduce
the
concentrate
volume
and
the
number
of
cell
cultures
required
to
assay
the
concentrate.
The
entire
concentrate
was
filter­
sterilized
using
a
0.2
micrometer
pore
size
Gelman
Serum
Acrodisc
filter
(#
4525)
which
has
been
pretreated
with
a
small
volume
of
beef
extract
eluent
to
minimize
viral
adsorption.

The
filter­
sterilized
concentrate
was
subdivided
into
the
following
aliquots
prior
to
being
frozen
at
­
80
°
:

 
Equivalent
of
500
L
of
water
sample
for
assay
of
viruses
in
Caco­
2
cell
cultures
 
Equivalent
of
500
L
of
water
sample
for
assay
of
viruses
in
BGMK
cell
cultures,
further
subdivided
into
a
1.5
mL
subsample,
and
the
remainder.
The
1.5
mL
subsample
was
used
in
a
pre­
test
for
cytotoxicity
in
BGMK
cultures.

 
Equivalent
of
100
L
of
water
sample
for
assay
of
HAV
in
FRhK­
4
cell
cultures.
58
 
Equivalent
of
100
L
of
water
sample
for
assay
by
direct
RT­
PCR
for
caliciviruses
(
noroviruses).

The
remainder
of
the
sample
concentrate
(
20%,
equivalent
to
300
L)
was
archived
at
­
80
°
C.

Virus
isolation
in
cell
cultures.
Three
cell
lines
were
used
to
detect
a
range
of
infectious
enteric
viruses.
The
BGMK
cell
line
was
used
to
propagate
adenoviruses,
enteroviruses,
and
reoviruses
according
to
the
procedure
of
(
Chapron
et
al.
(
2000).
Caco­
2
cells
were
used
for
the
detection
of
astroviruses
and
rotaviruses
according
to
Chapron
et
al.
(
2000).
The
FRhK­
4
cell
line
was
used
to
detect
hepatitis
A
virus
(
HAV).

Cell
culture
infectivity
assays
were
performed
in
a
minimal
number
of
75
cm2
flasks
(
generally
4
or
5).
A
pretest
to
screen
each
sample
concentrate
for
cytotoxicity
was
performed
in
25
cm2
BGMK
culture
flasks,
with
one
flask
being
inoculated
with
1.0
mL
of
sample
concentrate
and
a
second
flask
being
inoculated
with
0.5
mL
of
sample
concentrate.
The
pretest
cultures
were
observed
microscopically
for
evidence
of
cytotoxicity
(
or
CPE)
for
one
week,
before
the
remainder
of
the
sample
was
inoculated
into
cell
cultures.

Sample
concentrates
inoculated
into
BGMK
and
Caco­
2
cultures
were
pre­
activated
by
treatment
with
the
proteolytic
enzyme
trypsin
prior
to
inoculation.
This
was
done
because
previous
studies
had
indicated
enhanced
enteric
viruses
detection
using
this
trypsin
pre­
treatment.
Each
59
concentrate
was
mixed
with
a
solution
of
type
IX
trypsin
(
Sigma
T­
0303)
yielding
a
final
10
:
g/
mL
concentration,
then
incubated
30
minutes
at
37
°
C.
Pre­
activation
was
not
necessary
and
therefore
not
employed
for
HAV
propagation
in
FRhK­
4
cultures.
Because
divalent
cations
enhance
attachment
of
HAV
and
many
other
enteric
viruses
to
cells,
sample
concentrates
were
diluted
with
an
equal
volume
of
Dulbecco
=

s
PBS
before
being
inoculated
into
FRhK­
4
cultures.

Cell
cultures
in
75
cm2
(
confluent
monolayers
for
BGMK
cells
and
90­
95%
confluency
for
Caco­
2
and
FRhK­
4
cell
cultures)
were
drained
and
rinsed
three
times
with
Dulbecco
=

s
phosphate
buffered
saline
(
Sigma
#
D­
8662,
Gibco
#
14040
or
equivalent)
to
remove
residual
serum.
FRhK­
4
cell
cultures
were
rinsed
once.
Replicate
cultures
were
inoculated
with
sample
concentrates,
and
incubated
at
37
°
C
for
90
minutes,
while
being
rocked
every
15­
20
minutes
to
re­
distribute
the
inoculum,
to
allow
for
virus
adsorption
to
cells.
One
negative
control
flask
was
inoculated
with
PBS
before
any
flasks
were
inoculated
with
sample
concentrates,
and
a
second
negative
control
culture
was
similarly
be
inoculated
at
the
end
of
the
sample
inoculation
step.
No
enteric
virus
positive
control
flasks
were
to
be
prepared
at
this
time
in
order
to
avoid
possible
laboratory
virus
contamination.
Maintenance
medium
was
then
added
to
each
flask,
and
the
cultures
were
incubated
at
37
°
C.
The
maintenance
medium
for
BGMK
and
Caco­
2
cultures
consisted
of
serum­
free
Eagle's
minimum
essential
medium
with
Earle's
salts
(
MEM)
supplemented
with
5
:
g/
mL
trypsin.
The
maintenance
medium
used
in
FRhK­
4
cultures
consisted
of
Eagle's
minimum
essential
medium
with
Earle's
salts
supplemented
with
2%
serum
and
30
mM
MgCl
2
.
60
Cultures
were
observed
microscopically
on
days
1
and
2,
then
at
least
every
other
day
following
inoculation.
The
occurrence
of
cytopathology
or
cytopathic
effects
(
CPE)
on
the
first
two
days
was
tentatively
assumed
to
be
evidence
of
sample
cytotoxicity
or
the
release
of
cells
from
the
bottom
surface
of
the
tissue
culture
flask
by
the
action
of
the
trypsin.
If
cytotoxicity
thought
to
be
associated
with
sample
inocula
was
not
too
far
advanced,
the
affected
cultures
were
given
a
change
of
maintenance
medium
to
saved
them
from
possible
destruction
by
sample
cytotoxicity.

Alternative
approaches
for
cytotoxicity
reduction
included
removing
the
inoculum
following
the
90
minute
incubation
period
and
then
rinsing
the
cell
layer
with
PBS,
diluting
the
sample
concentrate
in
Dulbecco
=

s
PBS,
or
inoculating
less
concentrate
into
each
cell
culture.
Every
reasonable
effort
was
made
to
reduce
sample
cytotoxicity
and
maximize
enteric
virus
detection.

BGMK
and
Caco­
2
cultures
were
incubated
for
7
at
370C
days
following
inoculation.
All
cultures
were
freeze­
thawed,
and
10%
of
the
lysate
from
each
flask
was
inoculated
into
fresh
cultures
for
a
second
7­
day
passage.
If
a
flask
exhibited
possible
viral
cytopathology
(
CPE),
the
lysate
was
passed
through
a
0.22:
m
pore
size,
sterilizing
filter
into
a
fresh
culture
to
confirm
the
presence
of
viruses
and
the
absence
of
bacterial
or
fungal
contamination.
FRhK­
4
cultures
were
incubated
for
two
14­
day
passages,
with
the
maintenance
medium
being
changed
after
seven
days,
to
maximize
the
detection
of
typically
slow­
growing
HAV.

At
the
end
of
the
final
cell
culture
passage,
flasks
were
frozen
and
thawed
twice.
A
1­
mL
aliquot
from
each
of
the
first
and
second
passage
BGMK
and
Caco­
2
flasks
inoculated
with
a
given
sample
was
pooled
in
a
centrifuge
tube.
A
half­
volume
of
chloroform
was
added,
and
the
tube
was
vortex
mixed
at
high
speed
for
two
minutes.
The
tube
was
centrifuged
at
1,200­
1,800
x
g
for
61
20
minutes,
then
the
supernatant
extract
was
removed
and
split
into
aliquots
for
viral
analysis
or
archiving.

Virus
detection
by
nucleic
acid
amplification.
Chloroform­
extracted
cell
culture
lysate
pools
were
examined
for
adenoviruses,
astroviruses,
enteroviruses,
HAV,
reoviruses
and
rotaviruses
by
RT­
PCR
or
PCR.
Viral
nucleic
acids
were
extracted
from
lysates
using
the
QIAamp
Viral
RNA
Mini
Kit
(
Qiagen
#
52904).
By
modifying
two
steps
of
the
QIAamp
protocol,
adenovirus
DNA
could
be
efficiently
recovered
without
compromising
extraction
of
viral
RNA:

In
step
#
8
of
the
Qiagen
protocol,
The
sample
column
was
incubated
for
one
minute
after
adding
buffer
AW1,
before
centrifuging
the
column.

In
step
#
9,
the
sample
column
was
incubated
for
one
minute
after
adding
buffer
AW2,

before
centrifuging
the
column.

Prior
to
RNA
extraction,
enteric
viruses
in
2­
milliliter
aliquots
of
pooled,
chloroform­
extracted
cell
culture
lysates
were
concentrated
using
polyethylene
glycol
(
PEG)
precipitation.

Polyethylene
glycol
(
Sigma
#
P2139,
molecular
weight
=
8,000)
was
added
to
a
final
8%

concentration.
Sodium
chloride
was
added
to
a
0.3
M
concentration,
and
the
sample
was
mixed
until
the
additives
dissolved.
The
solution
was
incubated
for
two
hours
at
room
temperature
and
then
centrifuged
at
6,700
x
g
for
20
minutes
at
4
°
C.
The
pellet
was
resuspended
in
300
:
L
of
Dulbecco
=

s
PBS,
then
extracted
with
300
:
L
of
chloroform.
62
An
aliquot
of
each
water
sample
concentrate
was
assayed
directly
for
human
caliciviruses
(
noroviruses)
by
nucleic
acid
amplification
using
RT­
PCR
because
these
viruses
cannot
be
propagated
in
cell
cultures.
Each
sample
was
reconcentrated
using
polyethylene
glycol
precipitation.
PEG
was
added
to
a
final
10%
w/
v
concentration.
Sodium
chloride
was
added
to
a
final
0.3
M
concentration.
Since
the
sample
has
previously
been
adjusted
to
pH
7.0­
7.5,
no
further
pH
adjustment
was
necessary.
The
mixture
was
shaken
until
the
additives
had
dissolved.

The
solution
was
incubated
at
room
temperature
for
two
hours
or
at
4
°
C
overnight.
The
solution
was
centrifuged
at
6,000
to
10,000
x
g
for
15
minutes,
and
the
supernatant
was
removed
by
aspiration
and
discarded.
The
pellet,
which
may
not
be
visible,
was
resuspended
in
a
maximum
of
140
:
L
of
Dulbecco
=

s
PBS
containing
magnesium
and
calcium
ions.

Prior
to
RNA
extraction,
the
resuspended
pellet
was
extracted
with
chloroform.
A
100­:
L
volume
of
chloroform
was
added,
and
the
sample
vortex
mixed
for
one
minute.
The
sample
was
centrifuged
at
about
3,000
x
g
for
5­
10
minutes.
The
supernatant
was
removed
by
aspiration
and
recovered.
Viral
RNA
was
extracted
from
the
recovered,
chloroform­
extracted
supernatant
using
the
standard
QIAamp
Viral
RNA
Mini
Kit
protocol.

Nucleic
acid
amplification
by
(
RT­)
PCR
Introduction.
Viruses
in
chloroform­
extracted
cell
culture
lysates
that
had
been
inoculated
with
water
sample
concentrates
and
human
caliciviruses
(
noroviruses)
in
aliquots
of
water
sample
concentrates
were
analyzed
by
either
PCR
for
adenoviruses
or
RT­
PCR
for
astroviruses,

caliciviruses,
enteroviruses,
hepatitis
A
virus
(
HAV),
reoviruses,
and
rotaviruses.
The
procedures
63
for
combined
cell
culture
and
(
RT­)
PCR
were
based
on
those
previously
used
by
Chapron
et
al.

(
2000)
with
minor
modifications.
The
nucleic
acid
extraction
procedures
and
the
(
RT­)
PCR
primers
and
amplification
procedures
applied
to
each
virus
or
virus
group
are
described
in
more
detail
in
the
sections
that
follow
and
in
the
individual
report
of
the
other
three
participating
laboratories
(
see
Appendix).
RT­
PCR
for
HAV
and
caliciviruses
was
done
by
the
University
of
North
Carolina
lab
(
Southeast),
RT­
PCR
for
enteroviruses
was
done
by
the
University
of
Minnesota
lab
(
Upper
Midwest),
RT­
PCR
for
reovirus
and
rotavirus
was
done
by
the
TAMU
lab
(
Southwest),
and
(
RT­)
PCR
for
adenoviruses
and
astroviruses
was
done
by
the
University
of
New
Hampshire
lab
(
Northeast).
The
details
of
the
virus
(
RT­)
PCR
procedures
of
the
participating
laboratories
are
given
below
and
also
in
more
detail
in
the
individual
project
reports
of
the
other
three
participating
labs,
which
appear
in
the
Appendix
of
this
report.

Enterovirus
RT­
PCR
The
primers
for
RT­
PCR
amplification
of
enteroviruses
were:

3'
pan­
enterovirus
primer:
5'­
ACC
GGA
TGG
CCA
ATC
CAA
5'
pan­
enterovirus
primer:
5'­
CCT
CCG
GCC
CCT
GAA
TG
Random
hexamers
may
also
be
used
as
the
primer
for
enteroviruses
for
reverse
transcription.

The
reaction
mixtures
for
a
3.5
:
L
sample
were
as
follows.
(
For
larger
sample
volumes
the
amounts
were
increased
proportionally).
(
Note:
These
mixtures
utilized
reagents
from
the
GeneAmp
RNA
PCR
Core
Kit,
Applied
Biosystems
#
N808­
0143.)
64
RT
master
Mix
Stock
conc.
reaction/
reaction
Final
conc.

MgCl2
25
mM
4
5
mM
10x
PCR
Buffer
II,
pH
8.3
each
dNTP
10x
10
mM
2
2
each
1x
1
mM
3'
primer
or
random
hexamers
50
:
M
0.5
1.26
:
M
MuLV
reverse
transcriptase
50
U/:
L
0.9
45
units
Rnase
inhibitor
20
U/
µ
L
0.9
18
units
The
RT
conditions
were:
95
°
for
5
minutes,
42
°
for
60
minutes,
and
95
°
for
5
minutes.

PCR
master
mix
Stock
conc.
reaction/
reaction
Final
conc.

MgCl2
25
mM
4
2
mM
10x
PCR
Buffer
II,
pH
8.3
10x
8
1x
Water
(
Sigma
#
W­
4502)
66
Ampli­
Taq
DNA
polymerase
5
U/:
L
0.5
2.5
units
5'
primer
(
and
3'
primer
if
used
random
hexamers)
50
:
M
0.5
0.25
:
M
The
PCR
conditions
were
per
cycle:
95
°
C
for
1.5
minutes,
55
°
C
for
1.5
minutes,
and
72
°
C
for
1.5
minutes,
for
a
total
of
40
cycles.
The
expected
product
(
amplicon)
size
was
197
bp.
The
internal
oligonucleotide
probe
for
hybridization
was:
5'­
TAC
TTT
GGG
TGT
CCG
TGT
TTC.

Hybridization
was
at
55
°
C.
65
Hepatitis
A
virus
RT­
PCR
The
primers
for
RT­
PCR
amplification
of
HAV
were:

3'
HAV
primer:
5'­
CTC
CAG
AAT
CAT
CTC
CAA
C
5'
HAV
primer:
5'­
CAG
CAC
ATC
AGA
AAG
GTG
AG
(
VP1­
VP3
capsid
protein
interface
region)

The
RT­
PVR
reaction
mixtures
and
reaction
conditions
were
the
same
as
for
enteroviruses,
and
the
expected
product
(
amplicon)
size
was192
bp.
The
internal
oligonucleotide
probe
for
HAV
was:
5'­
TGC
TCC
TCT
TTA
TCA
TGC
TAT
G.
and
the
hybridization
temperature
was
55
°
C
Rotavirus
RT­
PCR
The
primers
for
RT­
PCR
amplification
of
rotaviruses
were
those
for
Group
A,
gene
9:

3'
rotavirus
primer:
5'­
GGT
CAC
ATC
ATA
CAA
TTC
T
5'
rotavirus
primer:
5'­
GAT
ATA
ACA
GCT
GAT
CCA
ACA
AC
The
reaction
mixtures
and
reaction
conditions
were
the
same
as
for
enteroviruses,
and
the
expected
product
(
amplicon)
size
was
208
bp.
The
internal
probe
that
could
be
used
for
product
confirmation
by
hybridization
was
as
follows:
5'­
AAT
TGG
AAA
AAA
TGG
TGG
CAA
GT.

The
hybridization
temperature
was
55
°
C.
66
Adenovirus
and
Astrovirus
(
RT­)
PCR
Nested
PCR
was
performed
on
UNH,
UNC,
UMN
and
TAMU
samples
for
both
astrovirus
and
adenovirus
type
40
and
41.
The
equivalent
volume
of
original
water
sample
examined
for
each
virus
was
500
liters.
Positive
controls
were
at
the
level
of
(
RT­)
PCR.
Virus
was
added
to
cell
culture
lysate
to
act
as
a
positive
control
for
(
RT­
PCR)
PCR
Astrovirus.
All
molecular
techniques
were
done
as
specified
in
the
methods
and
materials
developed
by
the
project
team
in
communication
with
the
EPA
project
manager.
Astrovirus
RTPCR
was
done
according
to
Chapron
et
al.
(
2000).
The
primers
used
were
specific
for
human
astrovirus:

RT
primer
5'­
GTAAGATTCCCAGATTGGT­
3',
and
PCR
primer
5'­
CCTGCCCCGAGAACAACCAAG­
3'.

An
11­:
L
sample
of
the
combined
(
pooled)
cell
lysate
was
denatured
with
0.5
:
L
each
of
0.05
M
EDTA
and
downstream
primer
at
99
/

C
for
8
min.
Eighteen
:
L
of
the
RT
mixture
was
then
added
and
run
for
42
min.
at
42
/

C
to
reverse
transcribe,
followed
by
5
min.
at
99
°
C.
The
RT
mixture
per
sample
consisted
of
2.5
:
L
10X
buffer
II,
8.5
:
L
of
25mM
MgCl
2,
1.25
:
L
of
each
10mM
dNTP,
0.5
:
L
of
100mM
DTT
(
Promega),
10
units
of
Rnasin,
and
50
units
of
RT.

After
the
RT
step,
28.5
:
L
of
a
PCR
master
mix
was
added.
The
PCR
mixture
per
sample
consisted
of:
3
:
L
of
10X
buffer
II,
1
:
L
of
the
PCR
primer,
0.5
:
L
of
the
RT
primer,
24
:
L
of
67
molecular
grade
water,
and
2.5
units
of
Ampli­
Taq
DNA
polymerase.
The
PCR
amplification
parameters
were
95
°
C
for
5
minute
hot
start,
followed
by
35
cycles
of:
95
°
C
for
30
seconds,
56
/

C
for
30
seconds,
72
°
C
for
30
seconds,
with
a
final
extension
at
72
°
C
for
5
minutes.
These
primers
yielded
a
193
and/
or
243
bp
amplicon.

For
nested
PCR,
1
µ
L
from
each
RT­
PCR
reaction
was
added
to
a
new
tube
containing
90
µ
L
of
a
nested
PCR
reaction
mixture,
which
contained
8
mM
MgCl
2,
10
:
L
10x
buffer,
1mM
of
each
dNTP,
2.5
units
of
Ampli­
Taq
DNA
polymerase
and
1
:
M
of
each
primer.
The
primers
used
were:
5'­
CCTTGCCCCGAGCCAGAA­
3'
and
5'­
TTGTTGCCATAAGTTTGTGAATA­
3'.

These
primers
yield
a
143
and/
or
183­
bp
amplicon.
Twelve
µ
l
of
each
RT­
PCR
product
as
well
as
12
:
L
of
the
nested
PCR
product
was
resolved
and
sized
by
electrophoresis
on
an
1.8%

agarose
gel,
stained
with
ethidium
bromide.
Molecular
weights
were
determined
by
comparison
with
a
1
Kb
DNA
ladder
(
Life
Technologies).
Astrovirus
serotype
2
was
used
as
a
positive
control.

Adenovirus.
All
molecular
techniques
were
done
as
specified
in
the
methods
and
materials
developed
by
the
project
team
and
communicated
to
the
EPA
project
manager.
Adenovirus
Hexon
PCR
was
done
generally
according
to
the
procedures
of
Xu
et
al.
(
2000).

The
PCR
primers
used
were:

Ad1
5'­
CCCTGGTA(
G/
T)
CC(
A/
G)
AT(
A/
G)
TTGTA­
3'
and
Ad2
5'­
TTCCCCATGGC(
Inosine)
CA(
C/
T)
AACAC­
3'.
68
A
5:
L
sample
of
the
combined
cell
lysates
was
added
to
47.5
:
L
final
volume
PCR
master
mix.

Final
concentrations
in
the
PCR
master
mix
per
sample
were
1.5mM
MgCl
2
,
1x
(
10x
Buffer
II),

0.2mM
dNTP
mix,
0.6:
M
of
each
primer,
and
2.5
units
of
Ampli­
Taq
DNA
polymerase.
The
PCR
parameters
were
95
°
C
for
5
minutes,
followed
by
40
cycles
of:
94
°
C
for
1
minute,
55
°
C
for
1
minute,
and
72
°
C
for
2
minutes,
with
a
final
extension
at
74
°
C
for
5
minutes.
These
primers
yielded
a
482
bp
amplicon.

For
nested
PCR,
1
µ
l
from
each
PCR
reaction
was
added
to
a
new
tube
containing
90
µ
l
of
a
nested
PCR
reaction
mixture,
which
contained
8
mM
MgCl
2,
10
:
L
10x
buffer,
1mM
of
each
dNTP,
and
1
:
M
of
each
primer.
The
primers
used
were:

5'­
GCCACCGAGACGTACTTCAGCCTG­
3'
and
5'­
TTGTACGAGTACGCGGTATCCTCGCGGTC­
3.

These
nested
primers
were
specific
for
Adenovirus
type
40
and
41.
Samples
were
run
for
35
cycles
of:
95
°
C
for
30
seconds,
55
°
C
for
30
seconds,
72
°
C
for
30
seconds,
yielding
a
142
bp
amplicon.
Twelve
:
L
of
each
nested
PCR
product
was
resolved
and
sized
by
electrophoresis
on
1.8%
agarose
gels
and
stained
with
ethidium
bromide.
Molecular
weights
were
determined
by
comparison
with
a
1
Kb
DNA
ladder
(
Life
Technologies).
Adenovirus
40
and
41
were
used
as
positive
controls.
69
Reovirus
RT­
PCR
The
primers
for
RT­
PCR
amplification
of
reoviruses
and
kindly
provided
by
Shay
Fout
of
the
US
EPA,
Cincinnati,
were:

3'
pan­
reovirus
primer:
5'­
GTG
CTG
AGA
TTG
TTT
TGT
CCC
AT
5'
pan­
reovirus
primer:
5'­
ACG
TTG
TCG
CAA
TGG
AGG
TGT
Reaction
mixtures
for
5
:
L
samples
were:

RT
master
mix
Stock
conc.
reaction/
reaction
Final
conc.

MgCl
2
25
mM
1.8
1.5
mM
10X
PCR
Buffer
II
10X
3
1X
(
Applied
Biosystems)

dNTP
mix
10
mM
each
2
0.7
mM
3'
primer
10
:
M
5
1.7
:
M
Water
11.45
The
initial
RT­
PCR
reaction
conditions
were:
99
°
C
for
5
minutes;
then
tubes
were
placed
in
ice.

Enzyme
mix
Stock
conc.
reaction/
reaction
Final
conc.

RNasin
30
units/:
L
0.75
22
units
(
Promega
N2511)

MuLV
RT
50
units/:
L
1
50
units
(
Applied
Biosystems)

The
RT
reaction
conditions
were:
43
°
C
for
60
minutes,
95
°
C
for
5
minutes,
and
then
tubes
were
placed
in
ice.
70
PCR
master
mix
Stock
conc.
reaction/
reaction
Final
conc.

MgCl
2
25
mM
4.2
1.5
mM
10X
PCR
Buffer
II
10X
7
1X
5'
primer
10
:
M
5
0.5
:
M
AmpliTaq
Gold
1
(
pH
8.3
buffer
only)

The
PCR
reaction
conditions
were
per
cycle:
95
°
C
for
1
minute,
55
°
C
for
1.5
minutes,
and
72
°

C
for
1.5
minutes,
for
a
total
of
40
cycles.
The
expected
product
(
amplicon)
size
was125
bp.

Internal
oligonucleotide
probes
for
the
individual
reovirus
types
1,
2
and
3
were
kindly
provided
by
Shay
Fout,
and
the
hybridization
temperature
was
51
°
C.

Calicivirus
direct
RT­
PCR
Calicivirus
(
Norovirus)
RT­
PCR
analysis
of
concentrated
virus
samples
from
groundwater
was
done
with
the
modified
generic
primers
designated
JV12/
JV13
(
Vinje
et
al.,
2001;
Hamidjaja
et
al.

2004).

3'
RegA
primer:
5'­
CTC
(
A/
G)
TC
ATC
(
Inosine)
CC
ATA
(
A/
G)
AA
(
Inosine)
GA
5'
MJV12
primer:
5'­
TA(
C/
T)
CA(
C/
T)
TAT
GAT
GC(
A/
C/
T)
GA(
C/
T)
TA
The
RT­
PCR
reaction
mixture
for
5:
l
samples
had
the
following
composition:

Antisense
mix
Stock
conc.
reaction/
reaction
Final
conc.

RegA
primer
50
:
M
1.2
60
pM
Water
2.8
71
The
initial
RT
reaction
conditions
were
as
follows:
94
°
C
for
2
minute
and
then
chilling
in
ice.

RT
master
mix
Stock
conc.
reaction/
reaction
Final
conc.

MgCl2
25
mM
1.8
3
mM
10X
PCR
Buffer
II
10X
1.5
1X
(
pH
8.3)

dNTP
mix
10
mM
each
1.5
1
mM
Water
0.2
AMV­
RT
10
units/:
L
0.5
5
units
Rnase
inhibitor
40
units/:
L
0.5
20
units
The
subsequent
RT
reaction
conditions
were:
42
°
C
for
60
minutes
and
then
94
°
C
for
5
minutes,

followed
by
chilling
in
ice.

PCR
master
mix
Stock
conc.
reaction/
reaction
Final
conc.

MgCl2
25
mM
2.4
1.5
mM
10X
PCR
Buffer
(
pH
9.0)
10X
4.5
dNTPs
10
mM
each
0.5
0.2
mM
MJV12
primer
50
:
M
1
50
pM
RegA
primer
50
:
M
0.6
50
pM
Taq
polymerase
5
units/:
L
0.5
2.5
units
Water
35.5
72
The
PCR
reaction
conditions
were
an
initial
94
°
C
for
3
minutes,
followed
by
40
cycles
with
each
cycle
consisting
of:
94
°
C
for
1
minute,
50
°
C
for
1.5
minutes,
and
74
°
C
for
1
minute.
This
was
followed
by
74
°
C
for
7
minutes.
The
expected
product
(
amplicon)
size
was
327
bp.

The
internal
oligonucleotide
probe
for
human
caliciviruses
was
a
mixture
of
one
Group
1
probe
(
GGI)
and
three
Group
II
probes
as
follows:

GGI
probe:
5'­
ATG
GA(
CT)
GTT
GG(
CT)
GA(
C/
T)
TAT
GT
(
20
pM)

GGIId
probe:
5'­
TGG
AAC
TCC
ATC
GCC
CAC
TGG
(
40
pM)

GGIIe
probe:
5'­
TGG
AAC
TCC
ATC
ACA
CAT
TGG
(
80
pM)

GGLeeds
probe:
5'­
TCA
CCA
GAT
GTT
GTC
CAA
GC
The
hybridization
temperature
was
42
°
C.

MS­
2
RT­
PCR
Coliphage
MS2
was
used
as
the
positive
control
for
RT­
PCR
analyses
by
some
laboratories
to
avoid
introduction
of
potential
human
enteric
virus
contamination.
About
100
pfu
of
MS2
per
reaction
tube
was
to
be
used
and
RT­
PCR
was
done
according
to
Meschke
and
Sobsey
(
1998).

The
RT­
PCR
primers
for
MS2
were
as
follows:

3'
(
downstream)
MS2
primer:
5'­
CCC
TAC
AAC
GAG
CCT
AAA
TTC
5'
(
upstream)
MS2
primer:
5'­
GCA
ACC
TCC
TCT
CTG
GCT
AC
73
Random
hexamers
could
be
used
as
the
primer
MS2
reverse
transcription.
Reaction
mixtures
and
reaction
conditions
for
MS2
were
the
same
as
for
enteroviruses.
The
expected
PCR
product
size
was
220
bp.

RT­
PCR
and
PCR
controls
To
reduce
the
possibility
of
cross­
contamination
of
field
samples
during
nucleic
acid
amplification,

RNA
coliphage
MS­
2
was
selected
to
act
as
a
positive
control
for
RT­
PCR
procedures.
This
positive
control
with
its
own
pair
of
primers
was
included
in
each
set
of
reactions
done
by
some
participating
labs.
Other
labs
already
had
their
own
RT­
PCR
and
PCR
controls
and
they
used
those
existing
QA/
QC
control
procedures
and
reagents
that
were
already
in
place.
The
details
of
those
measures
can
be
found
in
the
reports
of
the
other
3
participating
laboratories,
which
are
in
the
Appendix
to
this
report.

The
minimal
negative
control
samples
that
were
to
be
run
as
part
of
each
set
of
RT­
PCR
and
PCR
samples
included
the
following:
(
1)
combined
master
mixes,
enzymes
and
water
done
twice,
one
tube
placed
at
the
beginning
of
the
set,
and
one
tube
placed
at
the
end
of
the
set,
and
(
2)
a
cell
culture
negative
control.
Because
the
same
pooled
cell
cultures
were
tested
for
multiple
groups
of
human
enteric
viruses,
cell
culture
negative
control
RNA
extracts
needed
to
be
assayed
using
all
of
the
appropriate
virus
primer
pairs
used
by
a
given
participating
laboratory.
RT­
PCR
for
HAV
and
Caliciviruses
was
done
by
the
UNC
lab
(
Southeast),
RT­
PCR
for
enteroviruses
was
done
by
the
University
of
Minnesota
lab
(
Upper
Midwest),
RT­
PCR
for
reovirus
and
rotavirus
was
done
by
the
TAMU
lab
(
Southwest),
and
(
RT­)
PCR
for
adenoviruses
and
astroviruses
was
done
by
the
University
of
New
Hampshire
lab
(
Northeast).
Further
details
of
the
virus
(
RT­)
PCR
74
methods
of
the
participating
laboratories
are
given
in
their
individual
project
reports,
which
appear
in
the
Appendix
of
this
report.

Confirmation
of
presumptive
(
RT­)
PCR
positive
samples
Amplified
PCR
products
from
non­
nested
protocols
were
to
be
examined
by
agarose
gel
electrophoresis.
If
cDNA
bands
of
the
appropriate
size
were
detected,
the
presence
of
enteric
virus
sequences
was
to
be
confirmed
using
a
labeled
oligonucleotide
probe
internal
to
the
original
amplicon,
as
specified
above.
If
enteric
virus
cDNA
was
detected,
it
was
to
be
preserved
for
possible
nucleotide
sequencing.
75
PHASE
II
RESULTS
AND
DISCUSSION
Introduction
The
Phase
II
studies
of
the
project
consisted
of
both
additional
lab
studies
as
well
as
field
studies.

Lab
studies
were
performed
to
further
characterize
and
improve
the
method
to
confirm
coliphage
isolates
from
plaques
on
SAL
plates
and
lysis
zones
on
spot
plates
from
the
enrichment
method.

Lab
studies
also
were
conducted
to
determine
the
survival
of
coliphages
in
groundwater
held
at
4oC
for
up
to
6
days
prior
to
coliphage
assay
by
Methods
1601
and
1602.
Field
studies
consisted
of
the
analysis
of
groundwater
samples
from
wells
for
F+,
somatic
and
"
total"
coliphages
(
by
Methods
1601
and
1602),
fecal
indicator
bacteria
(
E.
coli
and
enterococci),
and
human
enteric
viruses
by
each
of
the
four
regional
labs.
Each
lab
collected
and
analyzed
groundwater
samples
in
its
region.
The
samples
concentrated
for
recovery
of
human
enteric
viruses
by
each
of
the
4
labs
were
divided
into
aliquots
so
that
individual
aliquots
could
be
sent
to
other
participating
labs
for
centralized
analysis
of
one
or
two
the
different
target
groups
of
human
enteric
viruses.
The
data
for
coliphages
and
fecal
indicator
bacteria
in
groundwater
were
analyzed
to
determine
if
the
analysis
of
both
coliphages
and
fecal
indicator
together
in
the
same
sample
of
groundwater
gave
greater
detection
of
fecally
contaminated
groundwater
than
the
analysis
of
only
one
indicator,

either
a
bacterium
or
a
coliphage.

Results
of
Field
Sample
Analysis
of
Coliphage
and
Bacterial
Indicators
in
Groundwater
Table
9
contains
all
of
the
data
for
the
presence
and
concentrations
coliphages
and
fecal
indicator
bacteria
in
samples
of
groundwater
from
all
four
laboratories.
76
Table
9.
Coliphages
Detected
by
Methods
1601
and
1602
and
Indicator
Bacteria
in
Groundwater
Lab*
and
Samples
SAL
(#/
100
mL)
Enrichment
(
1
L)
Bacteria/
100
mL)
Famp
CN­
13
C3000
Famp
CN­
13
C3000
Fec.
Colif.
E.
coli
Ent.
TAMU­
RS
(
1)
0
0
0
0
1
0
No
Data
0
0
TAMU­
HCR
(
1)
0
0
0
0
0
0
No
Data
0
0
TAMU­
RS
(
2)
0
0
0
0
0
0
No
Data
0
0
TAMU­
BM
(
1)
0
0
0
0
0
0
No
Data
0
0
TAMU­
KK
(
1)
0
0
0
0
0
0
No
Data
0
0
TAMU­
RS
(
3)
0
0
0
0
0
0
No
Data
0
0
TAMU­
KK
(
2)
0
0
0
0
1
0
No
Data
0
0
TAMU­
HCR
(
2)
0
1
0
0
1
1
No
Data
0
1
TAMU­
RS
(
4)
0
0
0
0
0
0
No
Data
0
0
TAMU­
RS(
5)
0
0
0
0
0
0
No
Data
1
0
TAMU­
MHP1a
0
0
0
0
0
0
No
Data
1
5
TAMU­
MHP1b
0
0
0
0
0
0
No
Data
0
0
TAMU­
AVC1
0
0
0
0
1
0
No
Data
0
0
TAMU­
FVE1
0
0
0
0
0
0
No
Data
0
0
TAMU­
AVC2
0
0
0
0
0
0
No
Data
0
0
TAMU­
FVE2
0
0
0
0
0
0
No
Data
0
0
TAMU­
FVE3
0
0
0
0
0
0
No
Data
0
0
TAMU­
AVC3
0
0
0
1
1
1
No
Data
0
0
TAMU­
MHP1c
0
0
0
0
0
0
No
Data
0
0
TAMU­
MHP2a
0
0
0
0
0
0
No
Data
0
0
TAMU­
MHP2c
0
0
0
0
0
0
No
Data
0
0
TAMU­
MHP3a
0
0
0
0
0
0
No
Data
0
0
TAMU­
MHP2b
0
0
0
0
0
0
No
Data
0
0
TAMU­
SME1
0
0
0
0
0
0
No
Data
0
0
TAMU­
SME2
0
0
0
0
0
0
No
Data
0
0
TAMU­
LME1
0
0
0
0
0
0
No
Data
0
0
TAMU­
MHP3b
0
0
0
0
0
0
No
Data
0
0
UNH­
1
0
0
0
0
0
0
1
0
0
UNH­
2
0
0
0
0
0
0
0
0
0
UNH­
3
4
0
0
1
0
1
0
0
0
UNH­
4
0
0
0
0
0
0
0
0
0
UNH­
5
0
0
0
0
0
0
0
0
0
UNH­
6
0
0
0
0
0
0
0
0
0
UNH­
7
0
0
0
0
0
0
0
0
0
UNH­
8
0
0
0
0
0
0
0
0
0
UNH­
9
0
0
0
0
0
0
0
0
0
UNH­
10
0
0
0
0
0
0
0
0
0
UNH­
11
0
0
0
0
0
0
0
0
0
UNH­
12
0
0
0
0
0
0
0
0
0
UNH­
13
0
0
0
0
0
0
0
0
0
UNH­
14
0
0
0
0
0
0
0
0
0
UNH­
15
0
0
0
0
0
0
35
0
0
UNH­
16
0
0
0
0
0
0
0
0
0
UNH­
17
0
0
0
0
0
0
0
0
0
UNH­
18
0
0
0
0
0
0
0
0
1
UNH­
19
0
0
0
0
0
0
2
0
13
77
UNH­
20
0
0
0
0
0
0
0
0
0
UNH­
21
0
0
0
0
0
0
0
0
2
UNH­
22
0
0
0
0
0
0
0
0
0
UNH­
23
0
0
0
0
0
0
5
0
2
UNH­
24
0
0
0
0
0
0
0
0
0
UNH­
25
0
0
0
0
0
0
0
0
0
MN­
01
Amu
0
TNTC
0
0
0
No
data
0
0
0
MN­
02
Ger
0
4
5
0
0
0
0
0
0
MN­
03
Rou
4
2
4
0
0
0
0
0
0
MN­
04
Tur
2
2
3
0
1
1
0
0
0
MN­
05
Bro
0
0
0
0
0
0
0
0
0
MN­
06
OG
0
0
0
0
0
0
0
0
0
MN­
07
KM
1
58
4
0
0
0
0
0
0
MN­
08
KM
40
12
7
1
0
1
0
0
1
MN­
09
Ham
0
0
0
0
0
0
0
0
0
MN­
10
Nor
9
28
0
0
0
0
0
0
0
MN­
11
Pre
0
0
0
0
0
0
0
0
0
MN­
12
Imm
0
0
0
0
0
0
30
0
1
MN­
13
Cen
0
0
0
0
0
0
0
0
0
MN­
14
His
0
0
0
0
0
0
1
0
0
MN­
15
Nor
0
0
0
0
0
0
0
0
0
MN­
16
Lak
1
234
574
0
1
0
0
1
0
0
MN­
17
Lak
M
0
0
0
0
0
0
0
0
0
MN­
18
Lak
M
0
0
0
0
0
0
0
0
0
MN­
19
Al
0
0
0
0
0
0
0
0
0
MN­
20
Day
0
0
0
0
0
0
0
0
0
MN­
21
Lak
M
0
9
1
0
0
0
0
0
0
MN­
22
GF
0
2
2
0
0
0
0
0
1
MN­
23
TA
3
2
7
0
1
1
1
0
0
MN­
24
Mil
11
6
1
0
0
0
17
15
12
MN­
25
Jay
0
6
3
0
0
0
3
1
1
MN­
12
ChR
3
3
4
0
0
0
0
0
0
MN­
16
Cem
R
6
5
6
0
0
0
0
0
0
MN­
24
Mil
R
2
5
0
0
0
0
248
3
20
UNC­
1­
BMH
0
0
0
0
0
0
0
0
0
UNC­
2­
GL
0
0
0
0
0
0
0
0
0
UNC­
3­
VE
0
0
0.4
0
0
0
0
0
0
UNC­
4­
KC
0
0
0.4
0
1
0
0
0
0.5
UNC­
5­
OC­
FL#
1
0
0
0
0
0
0
0
0
0
UNC­
6­
OC­
FL#
2
0
0
0
0
0
0
0
0
1.5
UNC­
7­
KC
0
0.4
0
0
0
0
0
0
0
UNC­
8­
VE
0
0.4
0
0
0
0
0
0
0
UNC­
9­
BF
0
0
0
0
0
0
0
0
0
UNC­
10_
SB
MHP
0
0
0
0
0
0
0
0
0
UNC­
11­
BMH1
0
0
0
0
0
0
0
0
0
UNC­
12_
GL
0
0
0
0
0
0
0
0
0
UNC­
13_
OC­
FL#
1
0
0
0
0
0
0
0
0
0
UNC­
14_
OC­
FL
#
2
0
0
0
0
0
0
0
0
0
UNC­
15_
GL
0
0
0
0
0
0
0
0
0
UNC­
16­
CDL
0
0
0
0
0
0
0
0
0
78
UNC­
17­
BF
0
0
0
0
0
0
0
0
0
UNC­
18­
SB
MHP
0
0
0
0
0
0
0
0
0
UNC­
19­
CDL
0
0
0
0
0
0
0
0
0
UNC­
20­
CSF
#
1
0
0
0
0
0
0
0
0
0
UNC­
21­
CSF
#
2
0
0
0
0
0
0
0
0
0
UNC­
22­
Oca
FL#
1
0
0
0
0
0
0
0
0
0
UNC­
23­
Oca
FL#
2
0.4
0
0
0
0
0
0
0
0
UNC­
24­
CSF
#
1
0
0
0
0
0
0
0
0
0
UNC­
25­
CSF
#
2
0
0
0
0
0
0
0
0
0
UNC­
26­
Oca
FL
#
1
0
0
0
0
0
0
0
0
0
UNC­
27­
Oca
FL
#
2
0
0
0
0
0
0
0
0
0
*
TAMU
=
Texas
Agricultural
and
Mechanical
University,
UNH
=
University
of
New
Hampshire,

MN
=
University
of
Minnesota
and
UNC
=
University
of
North
Carolina
Table
10
summarizes
these
data
on
the
presence
of
fecal
indicator
microbes,
including
somatic
coliphages,
male­
specific
coliphages,
"
total"
coliphages
(
detected
on
host
E.
coli
C3000),
fecal
coliform
bacteria,
E.
coli
and
enterococci
in
groundwater
samples
in
this
study
on
the
basis
of
positive
samples,
regardless
of
microbe
concentration.
A
total
of
107
samples
were
analyzed
and
these
samples
correspond
to
the
samples
that
were
also
analyzed
for
human
enteric
viruses.

Additional
groundwater
samples
were
analyzed
by
some
laboratories
in
the
initial
screening
of
groundwater
wells
for
possible
inclusion
in
the
study.
However,
these
samples
are
not
included
in
the
table
because
not
all
microbial
indicators
were
measured
by
all
methods
during
this
prescreening
analysis
effort
and
there
was
no
concurrent
analysis
of
human
enteric
viruses
for
possible
comparison.
79
Table
10.
Frequency
of
Occurrence
of
Fecal
Indicator
Microbes
in
Field
Ground
Water
Samples
#.
Pos./#
Tested
at:
Total
#
Pos./
Total
#
Tested,
All
Labs
%
Positive
Indicator
TAMU
UMN
UNC
UNH
Somatic
Coliphage
­
SAL
1/
27
16/
28
2/
27
0/
25
19/
116
16.4%
F+
Coliphage
by
SAL
0/
27
11/
28
1/
27
1/
25
13/
116
11.2%
"
Total"
Coliphage"
­
SAL
0/
27
12/
28
2/
27
0/
25
14/
116
12%
Somatic
Coliphage
Enrichment
5/
27
2/
28
1/
27
0
/
25
8/
116
6.9%

F+
Coliphage
Enrichment
1/
27
2/
28
0/
27
1/
25
4/
116
3.4%
"
Total"
Coliphage
Enrichment
2/
27
3/
28
0/
27
1/
25
6/
116
5.2%

Fecal
Coliform
Not
done
7/
28
0/
27
4/
25
11/
80
13.8%
E.
coli
2/
7
3/
28
0/
27
0/
25
5/
116
4.3%
Fecal
Coliform
and/
or
E.
coli
2/
27
7/
28
0/
27
4/
25
13/
116
11.2%
Enterococci
2/
27
6/
28
2/
27
4/
25
14/
116
12.1%

As
shown
in
Table
10,
The
frequency
of
detection
of
any
single
fecal
indicator
microbe
was
highest
for
somatic
coliphages
as
measured
by
the
SAL
method
at
16.4%
and
second
highest
for
fecal
coliform
at
13.8%.
However,
the
frequency
of
detection
of
somatic
coliphage
by
the
SAL
method
and
of
fecal
coliform
was
not
significantly
different
(
P
=
0.768
by
Mann­
Whitney
U­
test).

Interpretation
of
these
statistical
results
for
comparative
detection
of
somatic
coliphages
by
SAL
and
fecal
coliforms
is
limited.
This
is
because
not
all
samples
were
analyzed
for
both
of
these
fecal
indicators
and
therefore
a
paired
statistical
analysis
of
the
results
was
not
possible.

Enterococci
and
"
total
coliphage"
by
the
SAL
method
were
tied
for
third
in
detection
frequency
at
12.1%.
Overall,
these
results
indicate
the
rate
of
detection
of
any
single
fecal
indicator
was
higher
for
coliphages,
specifically
somatic
coliphage
detected
by
SAL,
than
any
other
single
indicator
tested.
It
is
also
noteworthy
that
the
simultaneous
detection
of
both
somatic
and
male­
specific
coliphages
as
"
total
coliphages"
by
the
SAL
method
on
a
single
host
bacterium,
E.
coli
C3000,
80
gave
a
high
frequency
of
detecting
fecal
contamination
at
12.1%,
making
it
one
of
the
best
indicators
tested.

Examination
of
the
results
of
coliphage
analyses
in
Table
10
indicate
that
each
coliphage
group
was
detected
more
frequently
by
the
single
agar
layer
(
SAL)
method
than
by
the
two­
step
enrichment
spot
plate
method.
This
finding
is
striking
given
that
the
sample
volume
for
the
SAL
method
was
only
100
mL
and
for
the
enrichment
method
it
was
1
liter.
The
comparative
detection
of
coliphages
by
SAL
and
enrichment
methods
was
16.4%
versus
6.9%
for
somatic
coliphage,
11.2%
versus
3.4%
for
F+
coliphages
and
12.1%
versus
5.2%
for
"
total"
coliphages.

The
results
for
the
frequency
of
detection
of
the
different
coliphage
groups
by
the
SAL
or
enrichment
method
were
statistically
compared
by
a
non­
parametric,
paired
t­
test
(
Wilcoxon
matched­
pairs
signed­
ranks
test).
The
detections
frequencies
between
SAL
and
enrichment
methods
were
significantly
different
for
somatic
coliphages
(
P
=
0.137)
and
for
F+
coliphages
(
P
=
0.0351)
and
they
were
nearly
significant
for
"
total"
coliphages
(
P
=
0.580).
Overall,
these
results
indicate
that
the
SAL
method
gave
significantly
better
detection
coliphages
than
did
the
enrichment
method.
The
reasons
for
this
are
not
known
and
probably
deserve
further
investigation.
It
should
be
remembered
that
both
methods
were
highly
efficient
in
detecting
coliphages
when
tested
in
phase
I
studies
on
seeded
samples
of
groundwater.

Comparative
Detection
of
Two
Indicators
in
Groundwater
Samples
It
was
of
interest
to
consider
the
simultaneous
detection
of
two
indicators
in
groundwater
samples.
This
is
because
the
proposed
groundwater
rule
has
considered
the
possibility
of
measuring
only
one
indicator
in
a
sample
(
either
bacterium
or
a
coliphage
indicator)
versus
81
measuring
both
a
bacterial
and
a
coliphage
indicator.
Therefore,
a
fundamental
consideration
is
whether
or
not
the
dual
measurement
of
two
indicators
improves
the
detection
of
fecal
contamination
in
groundwater
samples
by
increasing
the
frequency
of
fecal
indicator
(
a
positive
sample).
The
results
for
selected
pairs
of
fecal
indicators
that
gave
the
highest
detection
frequencies
are
summarize
in
Table
11.

Table
11.
Frequency
of
Occurrence
of
Dual
Indicators
in
Field
Ground
Water
Samples
Indicator
Pair
#.
Pos./#
Tested
at:
Total
#
Pos./
Total
#
Tested,
All
Labs
%
Positive
Somatic
and/
or
F+
Coliphage
­
SAL
1/
27
16/
28
3/
27
1/
25
21/
116
18.1%
Somatic
and/
or
F+
Coliphage
­
Enrichment
5/
27
4/
28
1/
27
1/
25
11/
116
9.5%
Enterococci
and
Fecal
Coliform
and/
or
E.
coli
3/
27
9/
28
2/
27
6/
25
20/
116
17.2%
Somatic
Coliphage
­
SAL
and/
or
Enterococci
2/
27
16/
28
2/
27
4/
25
24/
116
20.7%
Somatic
Coliphage
­
SAL
and/
or
Fecal
Coliform
or
E.
coli
3/
27
17/
28
2/
27
4/
25
26/
116
22.4%

Fecal
Coliform
and/
or
E.
coli
2/
27
7/
28
0/
27
4/
25
13/
116
11.2%
Enterococci
2/
27
6/
28
2/
27
4/
25
14/
116
12.1%
Enterococci
and
Fecal
Coliform
and/
or
E.
coli
3/
27
9/
28
2/
27
6/
25
20/
116
17.2%
Somatic
Coliphage
­
SAL
and/
or
Enterococci
2/
27
16/
28
2/
27
4/
25
24/
116
20.7%
Somatic
Coliphage
­
SAL
and/
or
Fecal
Coliform
or
E.
coli
3/
27
17/
28
2/
27
4/
25
26/
116
22.4%

Dual
versus
individual
detection
of
somatic
and
F+
coliphages
by
SAL.
As
a
first
case
of
comparing
the
detection
of
positive
samples
with
pairs
of
indicators
versus
single
indicators
is
the
SAL
detection
of
somatic
and/
or
male­
specific
coliphages
in
groundwater
samples.
This
coliphage
indicator
pair
was
considered
because
the
measurement
of
both
of
these
two
groups
of
coliphages
in
an
option
in
the
proposed
groundwater
rule.
Currently,
there
is
no
clear
basis
for
choosing
one
coliphage
group
over
the
other
and
therefore,
the
measurement
of
both
coliphage
groups
in
a
sample
on
their
respective
E.
coli
hosts
is
an
option.
The
frequency
of
detecting
a
82
positive
sample
(
positive
for
one
or
the
other
or
both)
was
18.1%
(
Table
11).
For
each
group
alone,
the
SAL
detection
frequency
was
16.4%
for
somatic
coliphages
and
11.2%
for
F+

coliphages
(
Table
10).
The
SAL
detection
of
either
or
both
of
these
coliphage
indicators
in
a
sample
by
dual
analysis
(
18.1%)
was
compared
to
the
frequency
of
detection
of
each
of
them
alone
(
F+
SAL
=
11.2%
and
somatic
SAL
=
16.4%)
using
the
Friedman
test,
a
non­
parametric
Analysis
of
Variance
(
ANOVA).
The
P
value
was
very
significant
(
0.0055),
indicating
the
detection
frequencies
were
significantly
different
and
were
highest
for
the
detection
of
either
or
both
coliphage
groups
when
both
are
measured
in
a
sample.

When
the
individual
SAL
detection
frequencies
of
F+
coliphages
(
11.2%)
and
somatic
coliphages
(
16.4%)
were
compared
by
a
non­
parametric
t­
test
(
Wilcoxon
matched­
pairs
signed­
ranks
test),

there
was
no
significant
difference
(
P=
0.105),
indicating
equivalent
detection
of
either
of
these
two
coliphage
groups
alone.
Furthermore,
when
the
SAL
detection
frequencies
of
F+
coliphages
alone
(
11%)
were
compared
to
the
dual
detection
of
either
F+
coliphages
and/
or
somatic
coliphages
(
18.1%)
by
the
Wilcoxon
matched­
pairs
signed­
ranks
test,
the
difference
was
very
significant
(
P
=
0.0078).
This
indicates
that
SAL
detection
of
both
coliphage
groups
is
better
than
detecting
F+
coliphages
alone.
A
similar
comparison
for
SAL
detection
of
somatic
coliphages
alone
(
16%)
versus
the
SAL
detection
of
either
or
both
coliphages
when
both
are
measured
in
a
sample
(
18.1%)
indicated
no
significant
difference
because
the
sample
size
was
too
small.

Overall,
SAL
detection
of
both
groups
of
coliphages
(
F+
and
somatic)
is
better
than
SAL
detection
of
either
group
alone.
83
Dual
versus
individual
detection
of
bacterial
indicator
pairs.
When
the
detection
of
two
fecal
indicator
bacteria
such
as
fecal
coliforms
and/
or
E.
coli
versus
enterococci
is
considered
because
both
groups
are
measured
simultaneously,
the
frequency
of
detecting
a
positive
sample
(
positive
for
one
or
the
other
or
both)
was
also
high
at
17.2%
(
Table
11).
For
each
group
alone,
the
detection
frequency
was
12.1%
for
enterococci
and
11.2%
for
fecal
coliforms
and/
or
E.
coli
(
Table
10).
The
dual
detection
of
either
or
both
of
these
bacterial
indicator
groups
in
a
sample
(
18.1%)
was
compared
to
the
frequency
of
detection
of
each
of
them
alone
(
enterococci
=
12.1%

and
fecal
coliforms
and/
or
E.
coli
=
11.2%)
using
the
Friedman
test.
The
P
value
was
significant
(
0.0366),
indicating
the
detection
frequencies
were
significantly
different
and
were
highest
for
the
detection
of
either
or
both
indicator
bacteria
groups
when
both
are
measured
in
a
sample.
When
the
individual
detection
frequencies
of
enterococci
(
12.1%)
and
fecal
coliforms
and/
or
E.
coli
(
11.2%)
were
compared
by
the
non­
parametric
Wilcoxon
matched­
pairs
signed­
ranks
test,
there
was
no
significant
difference
(
P=
0.839),
indicating
equivalent
detection
of
either
of
these
bacterial
indicator
groups
alone.
Furthermore,
when
the
detection
frequencies
of
enterococcus
alone
(
12.1%)
was
compared
to
the
dual
detection
of
either
or
both
enterococcus
and/
or
fecal
coliforms
and/
or
E.
coli
(
17.2%)
by
the
Wilcoxon
matched­
pairs
signed­
ranks
test,
the
difference
was
significant
(
P
=
0.031).
This
indicates
that
dual
detection
of
both
bacterial
indicator
groups
is
better
than
detecting
enterococci
alone.
A
similar
comparison
for
detection
of
fecal
coliforms
and/
or
E.
coli
alone
(
11.2%)
versus
the
dual
detection
of
either
or
both
enterococcus
and/
or
fecal
coliforms
and/
or
E.
coli
(
17.2%)
by
the
Wilcoxon
matched­
pairs
signed­
ranks
test
also
was
significant
(
P
=
0.016).
This
indicates
that
detection
of
both
bacterial
indicator
groups
is
better
than
detecting
fecal
coliforms
and/
or
E.
coli
alone.
Overall,
detection
of
both
groups
of
bacteria
is
better
than
detection
of
either
group
of
bacteria
alone.
84
Dual
versus
individual
detection
of
bacterial
and
coliphage
indicator
pairs.
The
extent
to
which
dual
detection
of
a
coliphage
indicator
and
bacterial
indicator
versus
individual
detection
of
either
one
alone
was
considered.
This
is
because
the
detection
of
both
a
coliphage
indicator
and
a
bacterial
indicator
in
a
groundwater
sample
has
been
an
option
for
the
proposed
groundwater
rule.
The
first
pair
of
coliphage
and
bacteria
indicators
to
compare
was
somatic
coliphages
detected
by
the
SAL
method
(
16.4%)
and
enterococci
(
12.1%)
(
Table
10).
This
pair
was
chosen
because
these
were
the
individual
coliphage
and
bacterial
indicators
measured
in
all
samples
and
detected
most
frequently.
For
SAL
somatic
coliphages
and/
or
enterococcus
being
measured
together
in
samples,
the
frequency
of
detecting
a
positive
sample
(
positive
for
one
or
the
other
or
both)
was
20.7%
(
Table
11),
which
was
higher
than
either
indicator
alone
or
any
coliphage
pair
or
bacterial
pair.
Furthermore,
a
statistical
comparison
of
measuring
either
indicator
alone
or
both
indicators
together
showed
significant
improvement
in
detecting
fecal
contamination
of
groundwater.
The
detection
of
either
or
both
of
these
indicator
groups
in
a
sample
(
20.7%)
was
statistically
compared
to
the
frequency
of
detection
of
each
of
them
alone
(
enterococci
=
12.1%

and
SAL
somatic
coliphages
=
16.4%)
using
the
Friedman
test.
The
P
value
was
significant
(
0.028),
indicating
detection
frequencies
were
significantly
different
and
were
highest
for
the
detection
of
either
or
both
indicator
groups
(
enterococci
and/
or
SAL
somatic
coliphages)
when
both
are
measured
in
a
sample.
When
the
individual
detection
frequencies
of
enterococci
(
12.1%)

and
SAL
somatic
coliphages
(
16.4%)
were
compared
by
the
non­
parametric
Wilcoxon
matchedpairs
signed­
ranks
test,
there
was
no
significant
difference
(
P=
0.355),
indicating
equivalent
detection
of
either
of
these
indicator
groups
alone.
Furthermore,
when
the
detection
frequencies
of
enterococcus
alone
(
12.1%)
were
compared
to
the
detection
of
either
or
both
enterococcus
and/
or
SAL
somatic
coliphages
(
20.7%)
by
the
Wilcoxon
matched­
pairs
signed­
ranks
test,
the
85
difference
was
very
significant
(
P
=
0.002).
This
indicates
that
detection
of
both
indicator
groups
is
better
than
detecting
enterococci
alone.
A
similar
comparison
for
detection
of
SAL
somatic
coliphages
alone
(
16.4%)
versus
the
detection
of
either
or
both
enterococcus
and/
or
SAL
somatic
coliphages
(
20.7%)
by
the
Wilcoxon
matched­
pairs
signed­
ranks
test
was
not
significantly
different
(
P
=
0.206).
Overall,
the
detection
of
both
groups
of
indicators
(
SAL
somatic
coliphages
and
enterococci)
was
generally
better
than
detection
of
either
indicator
group
alone.

When
the
coliphage
with
the
highest
detection
frequency,
which
was
SAL
somatic
coliphages
(
16.4%),
and
the
bacterial
indicator
with
the
highest
detection
frequency
(
and
measured
in
any
project
samples),
which
was
fecal
coliforms
and/
or
E.
coli
(
11.2%),
were
being
measured
together
in
samples,
the
frequency
of
detecting
a
positive
sample
(
positive
for
one
or
the
other
or
both)

was
22.4%%
(
Table
11).
This
indicator
detection
frequency
was
even
higher
than
either
indicator
alone
or
any
coliphage
pair
or
any
bacterial
pair.
Furthermore,
a
statistical
comparison
of
measuring
either
indicator
alone
or
both
indicators
together
showed
significant
improvement
in
detecting
fecal
contamination
of
groundwater.
The
detection
of
either
or
both
of
these
indicator
groups
in
a
sample
(
SAL
somatic
coliphages
and/
or
fecal
coliforms
and/
or
E.
coli
=
22.4%)
was
statistically
compared
to
the
frequency
of
detection
of
each
of
them
alone
(
SAL
somatic
coliphages
=
16.4%
and
fecal
coliforms
and/
or
E.
coli
=
11.2%)
using
the
Friedman
test).
The
P
value
was
very
significant
(
0.003),
indicating
the
detection
frequencies
were
significantly
different.

When
the
individual
detection
frequencies
of
fecal
coliforms
and/
or
E.
coli
(
11.2%)
and
SAL
somatic
coliphages
(
16.4%)
were
compared
by
the
non­
parametric
Wilcoxon
matched­
pairs
signed­
ranks
test,
there
was
no
significant
difference
(
P=
0.276),
indicating
equivalent
detection
of
either
of
these
indicator
groups
alone.
Furthermore,
when
the
detection
frequency
of
SAL
86
somatic
coliphages
alone
(
16.4%)
was
compared
to
the
combined
detection
of
either
or
both
fecal
coliforms
and/
or
E.
coli
and/
or
SAL
somatic
coliphages
(
22.4%)
by
the
Wilcoxon
matched­
pairs
signed­
ranks
test,
the
difference
was
significant
(
P
=
0.039).
This
indicates
that
detection
of
both
indicator
groups
is
better
than
detecting
SAL
somatic
coliphages
alone.
A
similar
comparison
for
detection
of
fecal
coliforms
and/
or
E.
coli
alone
(
11.2%)
versus
the
combined
detection
of
either
or
both
fecal
coliforms
and/
or
E.
coli
and/
or
SAL
somatic
coliphages
(
22.4%)
by
the
Wilcoxon
matched­
pairs
signed­
ranks
test
was
extremely
significantly
different
(
P
=
0.0002).
Therefore,

the
dual
detection
of
both
a
coliphage
indicator
and
a
bacteria
indicator
in
groundwater
samples
is
better
than
detecting
either
indicator
group
alone.

Because
there
is
interest
in
using
a
single
coliphage
host
to
simultaneously
detect
both
somatic
and
male­
specific
(
or
"
total")
coliphages,
it
also
was
of
interest
to
examine
the
dual
detection
of
total
coliphages
and
a
bacterial
indicator.
When
"
total"
coliphages
detected
by
the
SAL
method
(
detection
frequency
=
12.1%)
and
the
bacterial
indicator
of
enterococci
(
detection
frequency
=

12.1%)
are
considered
together,
the
frequency
of
detecting
a
positive
sample
(
for
either
or
both
indicators)
was
19.8%
(
Table
11).
A
statistical
comparison
of
measuring
either
indicator
alone
or
both
indicators
together
showed
significant
improvement
in
detecting
fecal
contamination
of
groundwater.
The
detection
of
either
or
both
of
these
indicator
groups
in
a
sample
(
19.8%)
was
statistically
compared
to
the
frequency
of
detection
of
each
of
them
alone
(
SAL
total
coliphages
=

12.1%
and
enterococci
=
12.1%)
using
the
Friedman
test.
The
P
value
was
significant
(
0.011),

indicating
the
detection
frequencies
were
significantly
different.
When
the
individual
detection
frequencies
of
enterococci
(
12.1%)
and
SAL
total
coliphages
(
12.1%)
were
compared
by
a
nonparametric
t­
test
(
Wilcoxon
matched­
pairs
signed­
ranks
test),
there
was
no
significant
difference
87
(
P>
0.999),
indicating
equivalent
detection
of
either
of
these
indicator
groups
alone.
Furthermore,

when
the
detection
frequencies
of
SAL
total
coliphages
alone
(
12.1%)
was
compared
to
the
dual
detection
of
either
or
both
SAL
total
coliphages
and
enterococci
(
19.8%)
by
the
Wilcoxon
matched­
pairs
signed­
ranks
test,
the
difference
was
very
significant
(
P
=
0.0039).
This
indicates
that
detection
of
both
indicator
groups
is
better
than
detecting
SAL
total
coliphages
alone.
A
similar
comparison
for
detection
of
enterococci
alone
(
12.1%)
versus
the
dual
detection
of
either
or
both
enterococcus
and/
or
SAL
total
coliphages
(
19.8%)
by
the
Wilcoxon
matched­
pairs
signed­
ranks
test
the
difference
also
was
very
significant
(
P
=
0.0039).
Therefore,
dual
detection
of
both
a
coliphage
indicator
("
total"
coliphages
by
SAL)
and
a
bacteria
indicator
(
enterococci)
in
groundwater
samples
is
better
than
detecting
either
indicator
group
alone.
This
is
the
case
even
for
the
single
coliphage
indicator
capable
of
detecting
both
male­
specific
and
somatic
coliphages
("
total"
coliphages).

Detection
of
"
total"
coliphages
by
SAL
(
detection
frequency
=
12.1%)
and
the
bacterial
indicator
of
fecal
coliforms
and/
or
E.
coli
(
detection
frequency
=
11.2%)
also
were
considered
together,

and
the
frequency
of
detecting
a
positive
sample
(
for
either
or
both
indicators)
was
20.7%
(
Table
11).
A
statistical
comparison
of
measuring
either
indicator
alone
or
both
indicators
together
showed
significant
improvement
in
detecting
fecal
contamination
of
groundwater.
The
detection
of
either
or
both
of
these
indicator
groups
in
a
sample
(
20.7%)
was
statistically
compared
to
the
frequency
of
detection
of
each
of
them
alone
(
SAL
total
coliphages
=
12.1%
and
fecal
coliforms/
E.
coli
=
11.2%)
using
the
Friedman
test.
The
P
value
was
very
significant
(
0.005),

indicating
the
detection
frequencies
were
significantly
different
and
were
highest
for
the
detection
of
either
or
both
indicator
groups
(
fecal
coliforms/
E.
coli
and
SAL
total
coliphages)
when
are
88
both
measured
in
a
sample.
When
the
individual
detection
frequencies
of
the
bacterial
indicator
(
fecal
coliforms/
E.
coli
=
12.1%)
and
SAL
total
coliphages
(
11.2%)
were
compared
by
the
nonparametric
Wilcoxon
matched­
pairs
signed­
ranks
test,
there
was
no
significant
difference
(
P
=

0.865),
indicating
equivalent
detection
of
either
of
these
indicator
groups
alone.
Furthermore,

when
the
detection
frequencies
of
SAL
total
coliphages
alone
(
12.1%)
was
compared
to
the
detection
of
either
or
both
SAL
total
coliphages
and/
or
fecal
coliforms/
E.
coli
(
20.7%)
by
the
Wilcoxon
matched­
pairs
signed­
ranks
test,
the
difference
was
very
significant
(
P
=
0.002).
This
indicates
that
detection
of
both
indicator
groups
is
better
than
detecting
SAL
total
coliphages
alone.
A
similar
comparison
for
detection
of
fecal
coliforms/
E.
coli
alone
(
11.2%)
versus
the
dual
detection
of
either
or
both
fecal
coliforms/
E.
coli
and/
or
SAL
total
coliphages
(
20.7%)
by
the
Wilcoxon
matched­
pairs
signed­
ranks
test
also
was
extremely
significant
(
P
=
0.001).
Therefore,

these
results
again
show
that
detection
of
both
a
coliphage
indicator
("
total"
coliphages)
and
a
bacterial
indicator
(
fecal
coliforms
and/
or
E.
coli)
in
groundwater
samples
is
better
than
detecting
either
indicator
group
alone.

It
is
noteworthy
that
the
frequency
of
detecting
fecal
contamination
in
a
groundwater
sample
was
nearly
as
high
with
"
total"
coliphages
detected
by
SAL
and
a
bacteria
indicator
(
with
either
enterococci
or
fecal
coliforms/
E.
coli
at
19.8
and
20.7%,
respectively)
as
with
somatic
coliphages
detected
by
SAL
with
a
bacteria
indicator
(
with
either
enterococci
or
fecal
coliforms/
E.
coli;

20.7%
and
22.4%,
respectively).
Overall,
these
results
indicate
that
the
measurement
of
both
a
coliphage
indicator
and
a
bacterial
indicator
together
in
a
groundwater
sample
gives
a
higher
frequency
or
likelihood
of
detecting
fecal
contamination
than
measuring
any
single
indicator
alone
or
even
measuring
pairs
of
bacterial
indicators
or
pairs
of
coliphage
indicators.
89
Statistical
Comparisons
of
Fecal
Indicators
in
Groundwater
Samples
Additional
statistical
analyses
of
data
on
indicator
occurrence
in
groundwater
samples
were
performed
for
the
data
set
of
all
groundwater
samples
for
which
human
enteric
viral
analysis
was
done.
Initially,
statistical
analysis
was
done
on
samples
for
which
there
were
results
for
all
of
the
fecal
indicators
tested
by
all
methods.
Because
fecal
coliforms
were
not
analyzed
by
one
of
the
4
participating
labs,
they
were
excluded
from
some
analyses
that
required
complete
sample
and
indicator
pairing
or
matching
(
no
missing
data
for
any
indicator
in
any
sample).
An
analysis
of
all
indicator
data
was
done
by
a
repeated
measures
ANOVA
(
Friedman
Test),
which
assumes
that
the
data
in
each
row
(
which
represents
a
water
sample)
is
matched
(
a
reasonable
assumption
because
it
is
for
a
specific
sample).
The
analysis
gave
a
P­
value
of
<
0.0001,
which
is
highly
significant,
and
therefore,
indicates
that
variation
among
column
medians
(
for
microbial
indicators
and
coliphage
methods)
is
significantly
greater
than
expected
by
chance.
Hence,
the
different
indicators
and
coliphage
methods
gave
significantly
different
results
in
detecting
fecal
contamination.
Interestingly
all
post­
tests
for
significant
differences
between
each
possible
combination
of
indicator
pair
were
not
significant,
with
all
P­
values
>
0.05.

The
data
for
all
indicators,
including
fecal
coliforms,
also
were
analyzed
by
the
Kruskal­
Wallis
Test
(
Nonparametric
one­
way
ANOVA),
which
assumes
no
matching
and
does
not
require
a
complete
matrix,
thereby
allowing
for
missing
data
for
some
samples.
This
analysis
can
be
justified
on
the
basis
of
simply
considering
an
unknown
distribution
of
microbial
indicators
of
fecal
contamination
in
groundwater
with
unknown
variability
over
time
and
space.
The
P
value
was
significant
at
0.0086,
indicating
that
variation
among
column
medians
was
significantly
90
greater
than
expected
by
chance.
This
analysis
again
shows
that
different
indicators
and
coliphage
methods
gave
significantly
different
results
in
detecting
fecal
contamination.

These
indicator
data
for
groundwater
samples
were
re­
analyzed
based
on
sample
positivity
instead
of
microbe
concentrations
by
dichotomizing
the
data
as
positive
and
negative
samples
and
assigning
a
value
of
1
to
a
positive
sample
and
keeping
0
for
a
negative
sample.
This
analysis
was
again
done
by
a
repeated
measures
ANOVA
(
Friedman
Test),
which
assumes
that
the
data
in
each
row
(
which
represents
a
water
sample)
is
matched
(
a
reasonable
assumption
because
it
is
results
for
the
same
specific
sample).
The
analysis
gave
a
P­
value
of
<
0.0001,
which
is
highly
significant,

and
therefore,
indicates
variation
among
column
medians
(
for
microbial
indicators
and
coliphage
methods)
is
significantly
greater
than
expected
by
chance.
Hence,
the
different
indicators
and
methods
gave
significantly
different
results
in
detecting
fecal
contamination.
Interestingly
all
posttests
for
significant
differences
between
each
possible
combination
of
indicator
pair
were
not
significant,
with
all
P­
values
>
0.05.
When
these
data
were
re­
analyzed
by
the
Kruskal­
Wallis
Test
(
Nonparametric
one­
way
ANOVA),
which
assumes
no
matching
and
does
not
require
a
complete
matrix,
thereby
allowing
for
missing
data
in
some
samples,
the
P
value
also
was
significant
at
0.0086.
This
result
indicates
that
variation
among
column
medians
(
for
the
different
fecal
indicators
and
coliphage
methods)
was
significantly
greater
than
expected
by
chance.

These
indicator
data
for
sample
positivity
in
groundwater
samples
also
were
re­
analyzed
by
a
nonparametric
repeated
measures
Analysis
of
Variance
(
Friedman
Test)
by
excluding
the
fecal
coliform
data
and
including
only
those
samples
that
were
analyzed
for
human
enteric
viruses.
The
P
value
for
this
test
was
0.0001,
considered
extremely
significant,
and
indicating
that
variation
91
among
column
means
(
for
the
different
fecal
indicators
and
coliphage
methods)
was
significantly
greater
than
expected
by
chance.
As
a
post­
test
to
the
Friedman
Test,
the
Tukey­
Kramer
multiple
comparisons
test
was
used
to
determine
if
there
were
significant
differences
in
sample
positivity
between
pairs
of
indicators.
Significant
differences
in
sample
positivity
were
observed
for
the
indicator
pairs
of:
SAL
somatic
versus
Enrichment
F+
(
P<
0.01),
SAL
somatic
versus
Enrichment
"
total"
coliphage
(
E.
coli
C3000)
(
P<
0.05),
and
SAL
somatic
versus
E.
coli
(
P<
0.01).
None
of
the
other
indicator
pairs
were
significantly
different
(
P>
0.05).

Analysis
for
Enteric
Viruses
in
Groundwater
A
total
106
samples
of
groundwater
were
analyzed
for
enteric
viruses
and
no
enteric
viruses
were
detected
in
any
of
the
samples
analyzed.
All
but
three
(
3)
of
the
groundwater
samples
were
1500
liters
in
volume.
There
were
3
well
samples
from
the
Northeast
for
which
the
sample
volume
was
less
than
1500
liters
due
to
a
lack
of
water
in
the
well
or
to
clogging
of
the
filter
used
to
concentrate
viruses
from
water.
These
Northeast
wells
were:
Well
#
2
=
592
L,
Well
#
4
=
229
L,

and
Well
#
14
=
400
L.

None
of
the
106
groundwater
samples
collected
and
analyzed
were
positive
for
human
enteric
viruses
by
cell
culture
and
(
RT­)
PCR
for
adenoviruses,
astroviruses,
enteroviruses,
reoviruses,

rotaviruses
or
hepatitis
A
virus
or,
in
the
case
of
caliciviruses,
by
direct
RT­
PCR
analysis.

Presumptive
positive
results
observed
in
one
participating
lab
for
detachment
of
cells
from
flasks
of
all
samples
were
due
to
non­
viral
effects
such
as
the
action
of
the
trypsin
in
the
medium
or
to
cytotoxicity
from
the
sample
concentrate
inocula.
Trypsin
can
cause
the
cells
to
dislodge
from
the
surface
of
the
flask
and
appear
abnormal.
Trypsin
effects
are
the
most
likely
explanation
because
92
cell
detachment
was
also
observed
in
the
negative
control
cultures.
Regardless
of
the
cause
of
this
effect,
it
was
not
due
to
the
presence
of
any
enteric
viruses,
based
on
the
negative
results
from
sample
analysis
by
nucleic
acid
amplification
methods.
Therefore,
despite
well
developed
protocols
and
the
analysis
of
large
sample
volumes
for
a
range
of
human
enteric
viruses,
none
were
found
in
any
of
the
samples
analyzed.
It
is
also
noteworthy
that
no
laboratory
experienced
any
episode
of
false
positive
viral
contamination
in
negative
control
samples
or
other
types
of
virus­
free
control
samples.
Hence,
no
viral
contamination
occurred
that
could
have
compromised
the
interpretation
of
positive
results
had
there
been
any
virus­
positive
field
samples.

Survival
of
Coliphage
in
Seeded
Groundwater
Coliphage
survival
in
groundwater
after
storage
at
4oC
for
times
periods
of
0­
6
days
was
determined
for
samples
seeded
with
sufficient
coliphages
(
as
filtered
raw
sewage)
to
give
about
30­
70
PFU
per
100
mL
when
assayed
by
the
SAL
method.
Duplicate
experiments
were
done
and
the
average
results
of
these
experiments
are
summarized
in
Figure
5
below.
93
0
20
40
60
80
100
120
Time
(
Days)
Survival
(%)
F+
Somatic
"
Total"
Figure
5.
Survival
of
F+,
Somatic
and
"
Total"
Coliphages
in
Seeded
100­
mL
Samples
of
Groundwater
Held
at
4oC
and
Assayed
by
the
SAL
Method
on
Days
0,
2,
3
and
6.

As
shown
in
Figure
5,
F+,
somatic
and
"
total"
coliphages
survived
relatively
well
for
2
or
3
days,

with
average
survivals
of
>
70%
compared
to
day
zero
when
detected
by
the
SAL
method
(
Method
1602).
By
day
6,
SAL
coliphage
titers
were
somewhat
lower,
with
average
survivals
of
about
40­
60%
compared
to
day
zero.
Overall,
these
results
indicate
that
samples
of
groundwater
for
coliphage
analysis
by
SAL
can
be
held
for
periods
of
2
or
3
days
with
only
relatively
minor
losses
in
coliphage
titer
(<
30%)
and
with
high
probabilities
of
detecting
coliphages
that
were
initially
present
when
the
samples
were
collected.
These
coliphage
survival
data
were
subjected
to
statistical
analyses
by
both
parametric
and
non­
parametric
analysis
of
variance
(
ANOVA).
The
coliphage
titers
were
not
significantly
different
at
the
5%
level
(
p
>
0.05)
among
the
days
of
analysis
(
days
0,
2,
3
and
6).
These
results
indicate
that
coliphage
titers
in
groundwater
as
measured
by
SAL
did
not
significantly
decrease
over
the
6­
day
holding
period.
0
2
3
6
94
Coliphage
survival
in
groundwater
after
storage
at
4oC
for
times
periods
of
0­
6
days
was
also
determined
for
groundwater
samples
seeded
with
sufficient
coliphages
(
as
filtered
raw
sewage)
to
give
about
5
infectious
units
per
sample
bottle
mL
of
1000
mL
when
assayed
by
the
enrichment
method
(
Method
1601).
As
shown
in
Table
12,
F+,
somatic
and
"
total"
coliphages
survived
relatively
well
for
as
long
as
6
days,
with
the
number
of
positive
enrichment
bottles
out
of
10
remaining
at
high
levels
of
7
to
10.

Table
12.
Survival
of
Coliphages
in
Seeded
Groundwater
Held
at
4oC
and
Assayed
by
the
Two­

Step
Enrichment
Spot­
plate
Method
Time
(
Days)
No.
of
Positive
Enrichment
Bottles
of
10
for
Indicated
Coliphage
Group
F+
Coliphages
Somatic
Coliphages
"
Total"
Coliphages
0
8
10
10
2
10
8
10
3
9
8
10
6
9
8
7
Overall,
these
results
indicate
that
samples
of
groundwater
for
coliphage
analysis
by
the
enrichment
method
can
be
held
for
periods
of
as
long
as
6
days
with
no
appreciable
loss
losses
in
coliphage
titer
and
with
high
probabilities
of
detecting
coliphages
that
were
initially
present
when
the
samples
were
collected.
These
coliphage
survival
data
were
subjected
to
statistical
analyses
by
both
parametric
and
non­
parametric
analysis
of
variance
(
ANOVA).
The
coliphage
titers
were
not
significantly
different
at
the
5%
level
(
p
>
0.05)
among
the
days
of
analysis
(
days
0,
2,
3
and
6).
These
results
indicate
that
low
coliphage
titers
in
groundwater
(
about
5
infectious
units
per
95
liter)
as
measured
by
enrichment
did
not
significantly
decrease
over
the
6­
day
holding
period.

Comparison
of
Coliphage,
Bacterial
Indicator
and
Enteric
Virus
Detection
in
This
Study
and
in
Previous
Studies
in
the
USA
Only
a
few
previous
studies
have
examined
coliphages
and
bacteria
in
groundwater
of
the
USA.

In
one
study
by
Abbaszadegan
et
al.
(
1999),
coliphages
were
analyzed
in
the
equivalent
of
about
15­
liter
samples
of
water
using
eluates
from
adsorbent
filters
used
to
concentrate
enteric
viruses
from
groundwater
samples.
Coliphages
were
assayed
on
the
following
host
bacteria:
E.
coli
C
for
somatic
coliphages
(
this
host
is
similar
t
E.
coli
CN­
13),
Salmonella
WG49
for
F+
coliphages
and
E.
coli
C3000
for
both
somatic
and
male­
specific
("
total"
coliphages).
Of
the
444
samples
analyzed
the
percentages
of
positive
samples
were:
4.1%
on
E.
coli
C,
10.8%
on
E.
coli
C3000
and
9.5%
on
Salmonella
WG49.
The
rates
of
coliphage
positivity
in
this
previous
study
are
lower
than
the
rates
of
positivity
in
this
current
study.
In
the
current
study
16.4%
of
samples
were
positive
for
somatic
coliphages
detected
in
100­
mL
sample
volumes
by
the
SAL
method
on
E.

coli
CN­
13
(
Method
1602)
compared
to
4.1%
positive
for
somatic
coliphages
detected
on
E.
coli
C.
In
the
current
study
11.2%
of
samples
were
coliphage
positive
for
F+
coliphages
on
host
E.

coli
Famp
by
SAL
compared
to
9.5%
positive
for
F+
coliphages
on
Salmonella
WG49.
In
the
current
study
12.1%
of
samples
were
positive
for
"
total"
coliphages
on
E.
coli
C3000
compared
to
10.8%
positive
on
this
host
in
previous
studies.
The
percent
of
samples
positive
for
any
of
the
three
coliphage
hosts
was
20.7%
and
for
all
three
hosts
together
it
was
0.2%.
In
the
current
study
the
percent
of
samples
positive
for
the
coliphage
host
pair
of
E.
coli
CN­
13
(
somatic
coliphages)
and
E.
coli
Famp
(
F+
coliphages)
was
20.7%.
Thus
the
rate
of
positivity
of
two
hosts
in
the
current
study
was
the
same
as
the
rate
of
positivity
for
3
hosts
in
this
previous
study.
96
Furthermore,
these
results
for
the
current
study
employed
100­
mL
samples
assayed
by
SAL
compared
to
15­
liter
samples
assayed
as
filter
eluate
concentrates
by
the
double
agar
layer
plaque
assay
in
sample
concentrate
volumes
of
2.5
or
5
mL
per
plate.
The
recovery
efficiency
and
lower
detection
limit
of
the
coliphage
assay
method
used
in
the
previous
study
was
not
reported.

In
the
same
study
by
Abbaszadegan
et
al.
(
1999),
culturable
human
enteric
viruses
were
analyzed
in
the
equivalent
of
160­
gallon
(
605­
liter)
samples
of
water
by
CPE
in
BGM
cell
cultures.
In
442
samples,
4.8%
of
sample
sites
and
4.1%
of
total
samples
were
positive
for
culturable
viruses
by
CPE.
In
comparison,
no
culturable
human
enteric
viruses
were
detected
in
any
of
the
106
groundwater
samples,
each
of
1500­
liter
(
400­
gallon)
volume,
analyzed
in
the
current
study.

In
a
later
study
Karim
et
al.
(
2004)
sampled
20
groundwater
wells
monthly
for
12
months
from
11
states
for
coliphages,
bacterial
indicators
and
human
enteric
viruses.
Sixteen
of
the
wells
were
known
to
be
fecally
contaminated.
Wells
were
monitored
for
the
presence
of
culturable
viruses,

enteric
virus
nucleic
acid
(
enterovirus,
hepatitis
A,
norwalk
virus,
rotavirus,
and
adenovirus)
by
(
RT­)
PCR,
coliphages
using
USEPA
Methods
1601
and
1602,
double
agar
layer
method
(
DAL),

and
RT­
PCR,
and
indicator
bacteria
(
total
coliforms,
E.
coli,
enterococci,
and
Clostridium
perfringens
spores).
A
total
of
231
to
235
samples
were
analyzed
per
well.
The
percentages
of
(
RT­)
PCR­
positive
samples
for
enteric
viruses
were:
2.1%
for
enteroviruses,
0%
for
HAV,
5.6%

for
rotavirus,
4.3%
for
Norwalk
Virus,
and
0.4%
for
adenovirus.
For
culturable
viruses
by
CPE,

positivity
was
3.9%.
As
previously
indicated,
no
human
enteric
viruses
were
detected
in
this
current
study.
97
For
coliphage
indicators
detected
by
the
enrichment
method
(
1601)
the
percentage
of
positive
samples
was
0%
for
somatic
coliphages
in
100
mL
and
1000
mL
volumes,
and
0.4
and
2.2%
for
F+
coliphages
in
100
mL
and
1000
mL,
respectively.
These
are
lower
rates
of
sample
positivity
than
were
obtained
in
this
current
study,
which
were
6.9%
for
somatic
and
3.4%
for
F+

coliphages
in
1000
mL
volumes.
For
coliphage
indicators
detected
by
the
SAL
method
(
1602)
in
100­
mL
sample
volumes,
the
percentage
of
positive
samples
was
0.9%
for
somatic
coliphages
and
5.6%
for
F+
coliphages.
These
also
are
lower
rates
of
positivity
than
in
this
current
study,
which
were
16.4%
and
11.2%
for
somatic
and
F+
coliphages,
respectively.
For
enterococcus,
the
percentage
of
positive
samples
was
0.4
and
5.5%
for
100
mL
and
1000
mL
volumes
respectively.

In
this
current
study
enterococcus
positivity
in
100­
mL
samples
was
much
higher
at
12.1%.
For
E.
coli,
the
percentage
of
positive
samples
was
4.3%
and
11.1%
for
100
mL
and
1000
mL
samples,
respectively.
In
this
current
study,
the
frequency
of
E.
coli­
positive
100­
mL
samples
was
4.3%,
which
is
the
same
E.
coli­
positivity
the
as
the
in
previous
study.

The
results
of
the
previous
study
suggested
that
dual
monitoring
for
both
a
bacterial
indicator
and
coliphage
would
be
useful
for
detecting
fecal
contamination
of
groundwater.
As
in
this
current
study,
monitoring
coliphages
and
bacteria
together
detected
fecally
contaminated
wells
more
frequently
than
either
a
coliphage
or
a
fecal
bacterial
indicator
alone.
As
a
single
indicator,
total
coliforms
in
1­
L
sample
volumes
were
found
to
occur
most
frequently
(
80%
of
the
wells
and
38.3%
of
the
samples).
However,
total
coliforms
are
not
fecal
indicator
bacteria
and
in
our
opinion
would
not
seem
to
be
appropriate
or
useful
as
a
single
fecal
indicator
organism.
The
authors
of
the
previous
study
concluded
that
the
dual
measurement
of
both
a
coliphage
and
a
bacterial
indicator
would
increase
the
detection
of
fecally
contaminated
groundwater.
No
single
98
fecal
indicator
alone
was
as
effective
in
detecting
fecal
contamination
as
the
dual
use
of
a
coliphage
and
a
bacterial
indicator.
These
previous
findings
are
consistent
with
those
of
this
current
study,
which
also
found
that
the
dual
measurement
of
two
indicators
and
especially
a
coliphage
and
a
bacterial
indicator
increased
the
likelihood
of
detecting
fecally
contaminated
groundwater.

Responses
to
Questions
and
Comments
of
the
April
2004
Coliphage
Workshop
Several
questions
about
coliphage
methods
were
identified
by
participants
at
an
"
International
Workshop
on
Coliphages
as
Indicators
of
Fecal
Contamination
in
Water
and
Other
Environmental
Media,"
that
was
sponsored
by
US
EPA
and
held
in
Arlington,
VA,
April
20­
21,
2004.
These
questions
and
our
responses
to
them
are
given
below
in
this
section
of
the
report.

1.
Costs
of
the
coliphage
tests?

Response.
The
four
participating
laboratories
have
estimated
the
costs
of
coliphage
testing
and
these
costs
are
listed
in
the
Table
below.
99
Table
13.
Costs
of
Coliphage
Analysis
by
the
Four
Study
Laboratories
 
November
2004
Coliphage
Assay
Costs
UNC
SAL
100
mL
water
sample
assayed
for
each
host
C3000
Famp
CN13
Famp+
CN13
Labor
and
materials
$
54
$
55
$
55
$
84
46%
indirect
costs
$
25
$
25
$
25
$
38
Total
$
79
$
80
$
80
$
122
Enrichment
C3000
Famp
CN13
Famp
+
CN13
Labor
and
materials
$
67
$
68
$
68
$
92
46%
indirect
costs
$
31
$
31
$
31
$
42
Total
$
98
$
99
$
100
$
134
TAMU
Actual
Costs
of
doing
coliphage
analysis.
Single
Agar
Layer
(
per
host
bacterium)
per
sample
Total
Labor
time:
(
3.5
hours
@
$
20.00/
hour):
$
70.00
Material
costs:
$
10.00
Total
cost:
$
80.00
2­
step
Enrichment
Total
Labor
Time:
3
hours
@
$
20.00/
hour:
20.00
Material
costs:
$
10.00
Total
Cost:
$
70.00
These
are
the
costs
per
sample,
per
host
bacterium
and
does
not
include
"
overhead"
or
other
costs.

The
labor
includes
media
preparation,
analysis
time
and
data
recording.

UNH
­
All
Methods
C3000
Famp
CN13
Famp
+
CN13
$
68
$
68
$
68
$
103
$
32
$
32
$
32
$
47
$
100
$
100
$
100
$
150
U
of
Minn
SAL
100
mL
water
sample
assayed
for
each
host
C3000
Famp
CN13
Labor
and
materials
$
80
$
80
$
80
49%
indirect
costs
$
39
$
39
$
39
Total
$
119
$
119
$
119
Enrichment
C3000
Famp
CN13
Labor
and
materials
$
100
$
100
$
100
49%
indirect
costs
$
49
$
49
$
49
Total
$
149
$
149
$
149
2.
The
need
for
and
effectiveness
of
the
method
of
confirming
coliphage­
positives
in
the
tests?

In
this
study
the
plaque
conformation
procedure
was
carefully
studies.
I
was
found
that
the
plaque
confirmation
rate
based
on
the
development
of
lysis
or
plaques
on
spots
of
host
lawns
in
100
agar
medium
averaged
nearly
80%.
It
was
concluded
this
was
a
simple
and
sufficiently
reliable
confirmation
method
and
would
be
adequate
for
routine
use
by
labs
doing
coliphage
analysis
3.
The
issue
of
using
a
singe
indicator,
such
as
a
bacterial
indicator
or
a
coliphage
indicator,

versus
using
both
a
bacterial
indicator
and
a
coliphage
indicator
in
detecting
a
positive
ground
water
sample?
Specifically,
the
extent
to
which
there
is
increased
detection
of
positives
when
using
only
one
indicator
such
as
a
bacterial
indicator
versus
two
indicators
­
a
coliphage
indicator
and
a
bacterial
indicator?

The
results
of
the
phase
II
field
studies
of
this
project
show
quite
clearly
that
the
dual
measure
of
two
indicators,
especially
a
coliphage
indicator
and
a
bacterial
indicator
significantly
increases
the
frequency
of
getting
a
positive
sample.
Measuring
either
a
coliphage
or
a
bacterial
indicator
alone
gave
significantly
lower
detection
of
positive
samples.
Therefore,
the
results
of
this
study
support
the
use
of
both
coliphage
and
bacterial
indicators
together
in
the
analysis
of
groundwater
samples
for
evidence
of
fecal
contamination.

4.
The
choice
of
coliphages
to
detect:
somatic,
male­
specific
or
"
total"
coliphages?

The
results
of
the
current
study
provide
data
showing
that
the
frequency
of
detecting
a
coliphage
in
ground
water
is
highest
for
somatic
coliphages
and
nearly
as
high
for
"
total"
coliphages
using
either
the
SAL
(
Method
1602)
or
enrichment
(
Method
1601)
methods.
Therefore,
it
is
concluded
from
these
results
that
either
of
these
coliphage
groups
are
likely
to
give
greater
detection
of
coliphages
than
the
measurement
of
F+
coliphages.
However,
F+
coliphages
also
are
important
indicators
of
fecal
contamination.
Because
they
have
the
ability
to
distinguish
human
from
animal
fecal
contamination
F+
coliphages
and
especially
F+
RNA
coliphages
also
have
merit
as
coliphage
101
indicators
of
fecal
contamination
of
groundwater.

5.
Whether
or
not
study
wells
were
subjected
to
treatment
(
disinfection)?

The
wells
of
the
current
study
were
not
disinfected.
Most
were
non­
community
public
water
supplies,
some
were
private
wells
and
some
were
public
water
supplies.
These
wells
were
not
required
to
disinfect
or
otherwise
treat
in
the
States
where
the
wells
were
located.
Because
two
of
the
wells
in
one
state
had
periodic
coliform
violations,
they
are
now
routinely
chlorinated.

However,
at
the
time
of
the
study
they
were
not
being
chlorinated.
102
SUMMARY
AND
CONCLUSIONS
In
initial
studies
Methods
1601
and
1602
were
evaluated
for
their
ability
to
detect
somatic,

malespecific
(
F+)
and
total
(
somatic
plus
F+)
coliphages
in
groundwater
samples
seeded
with
mixed,

natural
populations
of
coliphages
from
sewage.
The
SAL
method
(
Method
1601)
was
applied
in
10
experiments
to
replicate
100­
mL
volumes
of
groundwater
seeded
with
sewage
coliphages
for
coliphage
detection
with
each
of
three
E.
coli
host
bacteria:
E.
coli
CN­
13
for
somatic
coliphages,

E.
coli
Famp
for
male­
specific
(
F+)
coliphages
and
E.
coli
C3000
for
somatic
plus
F+
("
total")

coliphages.
There
was
efficient
coliphage
detection
(
average
53%)
and
plaque
confirmation
(
average
78%)
in
100­
mL
volumes
of
ground
water.
Overall,
the
results
of
these
studies
indicate
that
there
is
high
likelihood
of
detecting
even
low
levels
of
coliphages
in
100­
mL
volumes
of
ground
water
using
Method
1602.

For
evaluation
of
the
enrichment
method
(
Method
1601),
recoveries
of
somatic,
F+
and
total
coliphages
from
10
replicate
1­
liter
volumes
of
seeded
ground
water
in
eight
replicate
experiments
were
efficient
at
coliphage
input
levels
of
about
1.5
to
3
infectious
units/
L.
Recoveries
were
somewhat
variable
but
close
to
those
expected
based
on
the
expected
number
of
positive
1­
liter
enrichment
bottles
out
of
a
total
of
10.
There
is
a
high
likelihood
of
detecting
as
few
as
1­
3
coliphages
in
1­
liter
volumes
of
water
using
the
two­
step
enrichment
methods
of
Method
1601.

For
both
Method
1601
and
1602,
the
results
of
studies
with
seeded
samples
of
groundwater
showed
that
holding
samples
at
4oC
for
up
to
3
days
did
not
significantly
reduce
the
ability
to
detect
low
levels
of
coliphages.
Hence
samples
can
be
collected,
shipped
and
stored
prior
to
assay.
103
These
improved
and
validated
versions
of
Methods
1601
and
1602
were
further
validated
and
evaluated
studies
of
coliphage
detection
and
occurrence
in
more
than
100
field
samples
of
groundwater
from
wells
located
in
4
different
geographic
regions
of
the
USA
(
Northeast,

Southeast,
Southwest
and
upper
Midwest).
Overall,
the
results
for
fecal
indicator
occurrence
in
the
field
groundwater
samples
analyzed
in
this
study
indicated
that
coliphages
are
reliable
indicators
for
detecting
fecal
contamination
and
can
detect
fecal
contamination
as
frequently
or
more
frequently
than
do
bacterial
indicators.
In
more
than
100
groundwater
samples
collected
from
wells
coliphages
were
detected
with
greater
or
similar
frequency
than
were
fecal
indicator
bacteria.
The
percentages
sample
positivity
for
coliphages
were
11
to16%
by
the
SAL
method
(
Method
1602)
and
6.9
to
3.4%
by
the
two­
step
enrichment
spot
plate
method
(
Method
1601).

By
comparison,
the
percentages
of
sample
positivity
for
bacteria
(
fecal
coliforms,
E.
coli
or
enterococci)
ranged
from
13.8
to
4.3%.

Coliphage
detection
in
groundwater
was
higher
using
the
SAL
assay
(
Method
1602)
on
100­
mL
sample
than
using
the
two­
step
enrichment
spot
plate
method
(
Method
1601)
on
1­
liter
samples.

Additionally,
coliphage
detection
by
either
method
was
highest
for
somatic
coliphages,
next
highest
for
"
total"
coliphages
and
lowest
for
F+
coliphages.
The
relatively
high
detection
of
"
total"
coliphages
by
the
SAL
method
indicates
that
a
single
host,
E.
coli
C3000,
can
be
used
to
detect
either
somatic
or
male­
specific
coliphage
or
both
with
a
high
degree
of
sensitivity.

The
results
from
the
analyses
of
these
groundwater
samples
indicate
that
there
is
a
significantly
greater
likelihood
of
detecting
fecal
contamination
if
two
indicators
are
analyzed
in
the
same
sample
than
if
only
one
indicator
is
analyzed.
Detection
of
two
indicators
was
higher
with
a
104
coliphage
and
a
bacteria
indicator
pair
(
as
high
as
20.7
and
22.4%
positivity)
than
with
either
pairs
of
coliphage
indicators
(
as
high
18%
positivity)
or
pairs
of
bacterial
indicators
(
17.2%
positivity).

Therefore,
rates
or
frequencies
of
detecting
fecal
contamination
in
groundwater
are
higher
when
using
two
fecal
indicators
than
using
a
single
fecal
indicators,
and
highest
when
using
a
coliphage
and
a
bacterial
indicator
together.
These
findings
clearly
support
the
position
of
determining
groundwater
vulnerability
to
fecal
contamination
by
measuring
both
a
coliphage
indicator
and
a
bacterial
indicator,
rather
than
measuring
either
one
alone.

Human
enteric
viruses,
including
adenoviruses,
astroviruses,
enteroviruses,
hepatitis
A
virus,

reoviruses
and
rotaviruses
were
not
detected
by
combined
cell
culture
and
(
RT­)
PCR
in
any
of
the
106
samples
analyzed.
Human
Caliciviruses
(
Noroviruses)
were
not
detected
by
direct
RT­
PCR
of
virus
concentrates
from
the
same
106
samples
groundwater.
Therefore,
it
was
not
possible
to
compare
or
look
for
associations
in
occurrence
of
coliphages
and/
or
bacteria
relative
to
human
enteric
viruses
in
groundwater
samples.

It
is
recommended
that
EPA
adopt
these
improved
methods
for
coliphage
detection
for
the
forthcoming
Groundwater
Rule.
It
is
also
recommended
that
the
Groundwater
Rule
require
the
analysis
of
both
coliphages
and
fecal
indicator
in
the
same
sample
of
groundwater
in
order
to
significantly
increase
the
likelihood
of
detecting
fecal
contamination.
Examination
of
groundwater
samples
for
a
single
indicator,
either
a
virus
or
a
bacterium,
will
significantly
reduce
the
chances
of
detecting
fecal
contamination.
105
REFERENCES
Abbaszadegan,
M.,
P.
W.
Stewart,
M.
W.
LeChevallier,
J.
S.
Rosen
and
C.
P.
Gerba
(
1999)

Occurrence
of
Viruses
in
Ground
Water
in
the
United
States.
American
Water
Works
Service
Company.
Belleville,
IL
and
Voorhees,
NJ.

American
Public
Health
Association.
(
1995)
Section
9230.
Standard
Methods
for
the
Examination
of
Water
and
Wastewater,
19th
ed.
American
Public
Health
Association,
Washington
DC.

Chapron,
C.
D.,
Ballester,
N.
A.,
Fontaine,
J.
H.,
Frades,
C.
N.
&
Margolin,
A.
B.
2000
The
Detection
of
Astrovirus,
Enterovirus
and
Adenovirus
Type
40
and
41
in
Surface
Waters
Collected
and
Evaluated
by
the
Information
Collection
Rule
and
Integrated
Cell
Culture/
Nested
PCR
Procedure.
Appl.
and
Environ.
Microbiol.,
60
(
6),
2520­
2525.

De
Leon,
R.,
Y.
S.
C.
Shieh,
R.
S.
Baric,
and
M.
D.
Sobsey.
1990.
Detection
of
enteroviruses
and
hepatitis
A
virus
in
environmental
samples
by
gene
probes
and
polymerase
chain
reaction.
Proc.

Water
Quality
Technology
Conference,
1989,
pp.
833­
853.
Amer.
Water
Works
Assoc.,
Denver,

Co.

Environmental
Protection
Agency.
1995
Virus
Monitoring
Protocol
for
the
Information
Collection
Requirements
Rule.
U.
S.
Environmental
Protection
Agency,
publication
EPA/
814­
B­

95­
002.
Government
Printing
Office,
Cincinnati,
Ohio.
106
Fout,
Shay
(
2001)
personal
communication.
Not
yet
published.

Gouvea,
V.,
R.
I.
Glass,
P.
Woods,
K.
Tanuguchi,
H.
F.
Clark,
B.
Forester,
and
Z.
Y.

Fang.
1990.
Polymerase
Chain
Reaction
Amplification
and
Typing
of
Rotavirus
Nucleic
Acid
From
Stool
Specimens.
J.
Clin.
Microbiol.
59:
276­
282.

Hsu
FC,
Shieh
YS,
van
Duin
J,
Beekwilder
MJ,
Sobsey
MD.
(
1995)
Genotyping
male­
specific
RNA
coliphages
by
hybridization
with
oligonucleotide
probes.
Appl
Environ
Microbiol.
1995
Nov;
61(
11):
3960­
6.

Karim,
M.
R.,
M.
W.
LeChevallier,
M.
Abbaszadegan,
A.
Alum,
J.
Sobrinho
and
J.
Rosen
(
2004)

Field
Testing
of
USEPA
Methods
1601
and
1602
for
Coliphage
in
Groundwater
Awwa
Research
Foundation,
American
Water
Works
Association,
Denver,
CO.

Levin,
M.
A.,
J.
R.
Fischer
and
V.
J.
Cabelli.
1975.
Membrane
Filtration
Technique
for
Enumeration
of
Enterococci
in
Marine
Waters.
Applied
Microbiology
30:
66­
71.

Meschke,
J.
S.,
and
M.
D.
Sobsey.
1998.
Comparative
adsorption
of
Norwalk
virus,
poliovirus
1
and
F+
RNA
coliphage
MS2
to
soils
suspended
in
treated
wastewater.
Water
Sci.
Tech.
38:
187­

189.

Schwab,
K.
J.,
R.
De
Leon,
and
M.
D.
Sobsey.
1995.
Concentration
and
purification
of
beef
extract
mock
eluates
from
water
samples
for
the
detection
of
enteroviruses,
hepatitis
A
virus
and
107
Norwalk
virus
by
reverse
transcription­
PCR.
Appl.
Environ.
Microbiol.
61:
531­
537.

Schwab,
K.
J.,
R.
De
Leon,
and
M.
D.
Sobsey.
1995.
Concentration
and
purification
of
beef
extract
mock
eluates
from
water
samples
for
the
detection
of
enteroviruses,
hepatitis
A
virus
and
Norwalk
virus
by
reverse
transcription­
PCR.
Appl.
Environ.
Microbiol.
61:
531­
537.

Sobsey,
M.
D.,
K.
S.
Schwab,
R.
De
Leon,
and
Y.
S.
C.
Shieh.
1996.
Enteric
Virus
Detection
by
Nucleic
Acid
Methods.
American
Water
Works
Association
Research
Foundation,
Denver.

USEPA
(
2002)
Method
1600:
Enterococci
in
Water
by
Membrane
Filtration
Using
membrane­

Enterococcus
Indoxyl­
 ­
D­
Glucoside
Agar
(
mEI),
15pp.
EPA­
821­
R­
02­
022.
U.
S.

Environmental
Protection
Agency,
Office
of
Water,
Washington,
DC
20460.

Vinje,
J.
et
al.
2001.
Evaluation
of
different
RT­
PCR
primer
pairs
for
the
detection
of
low
numbers
of
A
Norwalk­
like
viruses
@.
Poster
presented
at
American
Society
for
Microbiology
annual
meeting,
Orlando,
Florida,
May,
2001.

Vinjé
J,
Hamidjaja
RA,
and
Sobsey
MD.
2004
Detection
of
novel
genotypes
using
VP1
typing
for
genetic
classification
of
genogroup
I
and
II
noroviruses.
J.
Virological
Methods.
116(
2):
109­
117.

Xu,
W.,
M.
C.
McDonough,
and
D.
D.
Erdman.
2000.
Species­
specific
identification
of
human
adenoviruses
by
a
multiplex
PCR
assay.
J.
Clin.
Microbol.
38:
4114­
4120.

APPENDICES
108
APPENDIX
I
EPA
Coliphage
Groundwater
Study
Summary
Report
of
the
Northeast
Region
Mark
D.
Sobsey,
Study
Principal
Investigator,
University
of
North
Carolina
and
Aaron
Margolin,
Co­
Principal
Investigator,
and
Nicola
Ballester
University
of
New
Hampshire,
Durham,
NH
November,
2004
Introduction
The
purpose
of
this
study
was
to
validate
and
apply
US
EPA
Methods
1601
and
1602
for
detection
coliphages
in
water
by
applying
them
to
field
samples
of
groundwater.
The
goal
was
to
examine
27
samples
of
groundwater,
preferably
from
public
water
supply
wells,
for
somatic,
F+

and
total
coliphages,
fecal
indicator
bacteria
and
human
enteric
viruses
in
the
Northeast
United
States.
109
Methods
and
Materials
Sampling
sites
All
sample
sites
were
located
in
New
England.
Eight
well
sites
were
public
water
sources
and
17
were
private
wells.
A
total
of
25
wells
were
sampled
instead
of
27
due
to
a
very
severe
and
harsh
winter.
NH
had
its
first
snowfall
at
the
end
of
October
and
a
second
snowfall
at
the
beginning
of
November,
2002.
Plans
to
sample
two
additional
wells
as
soon
as
the
weather
permitted
could
not
be
carried
out
because
New
England
experienced
one
of
the
snowiest
winters
ever.
Therefore,

only
25
well
samples
were
collected
and
analyzed.
Of
the
25
wells,
there
were
12
sample
sites
in
New
Hampshire,
two
of
which
were
from
public
wells
that
were
approximately
500
and
700
ft
deep,
respectively.
None
of
these
wells
had
any
form
of
disinfection.
The
other
wells
from
NH
were
all
private
wells.
These
wells
also
were
not
disinfected.
One
well
from
NH
was
a
private,

very
shallow
well,
less
then
35
feet
deep
and
lined
with
stone.
This
was
not
considered
a
potable
well
but
was
used
for
farm
irrigation.
Four
sites
in
Maine
were
all
privately
owned
wells
and
not
disinfected.
Three
sites
were
in
Vermont,
and
they
were
all
privately
owned
wells
and
not
disinfected.
All
of
the
privately
owned
wells
were
drilled
wells,
excepted
for
the
one
in
NH
as
indicated
above,
and
they
were
of
varying
depths
that
were
unknown
to
the
homeowner
at
the
time
samples
were
collected.
There
were
6
samples
from
public
water
supply
wells
in
Massachusetts.
The
public
water
supplies
in
Massachusetts
were
chosen
due
to
positive
results
previously
found
for
total
and
fecal
coliforms,
enterococci,
and
male­
specific
coliphages.

Additionally
3
of
the
6
locations
had
positives
previously
reported
for
rotavirus
and
enterovirus,

by
molecular
methods.
110
Sampling
All
groundwater
samples
were
collected
using
the
EPA
ICR
method
between
June
and
November
2002
(
US
EPA,
1996).
At
each
well
site,
1500
L
of
well
water
was
collected
through
a
sterile
1MDS
filter
setup.
For
a
few
samples,
the
filter
clogged
or
the
well
ran
dry,
before
1500
liters
could
be
processed.
In
these
cases
the
successfully
filtered
volume
is
reported.
Additionally
a
10
L
grab
sample
was
collected
from
each
well
in
a
sterile
container
for
bacteriological
and
coliphage
analyses.
Enteric
virus
sampling
equipment
was
sterilized
between
well
sites
with
0.1%
Bleach
solution
followed
by
successive
2%
sodium
thiosulfate
and
distilled
water
rinses.
All
samples
were
kept
at
40C
and
analyzed
within
48
hours.

Bacteriological
analysis
All
bacterial
analysis
of
fecal
coliforms
and
enterococcus
was
done
as
specified
in
EPA­
approved
methods.
The
samples
were
analyzed
by
membrane
filtration
using
mFC
and
MEI
agars.

Coliphage
analysis
All
analysis
of
male­
specific
(
F+),
somatic
coliphages
and
total
coliphages
was
done
as
specified
in
EPA­
approved
methods.
The
US
EPA
Methods
1601
(
enrichment)
and
1602
(
single
agar
layer)

were
used
with
the
host
bacteria
E
.
coli
Famp
for
F+
coliphages,
E.
coli
CN­
13
for
somatic
coliphages,
and
E.
coli
C3000
for
"
total"
coliphages.

Enteric
virus
recovery
and
analysis
The
1MDS
filters
used
to
adsorb
viruses
from
samples
of
ground
were
eluted
and
concentrated
as
specified
in
the
EPA
ICR
method.
The
only
change
was
that
the
final
resuspended
acid
111
(
flocculated
beef
extract)
precipitate
was
suspended
in
20
mL
of
phosphate
buffered
saline
(
Sigma
D8662)
rather
than
30
mL
of
sodium
phosphate.
Samples
were
filtered
through
a
37
mm
diameter,
0.2
micrometer
pore
size,
beef
extract
pre­
treated
Gelman
Serum
Acrodisc
filter
(
4525).

The
filter­
sterilized
concentrate
was
divided
into
5
aliquots.
Aliquots
were:
2/
6.7
mL
portions
for
UNH
Caco­
2
and
BGMK
cell
cultures,
2/
1.3
mL
portions
for
UNC
FRhK­
4
cell
culture
and
Calicivirus
(
norovirus)
analysis,
and
4
mL
was
archived.
These
sample
concentrate
volumes
are
equivalent
to
500
liters
of
initial
water
for
inoculation
into
Caco­
2
and
BGMK
cells,
respectively,

the
equivalent
of
100
liters
of
initial
water
for
FRhK­
4
(
HAV)
and
Calicivirus
(
norovirus)

analysis,
and
300
liters
of
initial
water
for
archiving.
All
aliquots
were
frozen
at
­
800C
prior
to
shipment
and
analysis.

Tissue
culture
protocol
for
virus
isolation
in
BGMK
and
CaCo­
2
cells
UNH
screened
concentrates
for
BGMK
cytotoxicity
on
25
cm2
flasks
prior
to
inoculation
of
samples
onto
75
cm2
flasks.
Sample
concentrates
were
pre­
activated
for
30
minutes
at
370C
with
10
:
g/
mL
of
type
IX
trypsin
(
Sigma
T­
0303)
for
both
Caco­
2
and
BGMK
inoculates.
BGMK
and
Caco­
2
cell
cultures
were
rinsed
three
times
with
PBS
before
inoculation.
Inoculated
cultures
were
incubated
at
370C
for
90
minutes
with
rocking
every
15­
20
minutes.
Only
negative
controls
were
run.
Serum
free
maintenance
media
containing
5
:
g/
mL
trypsin
was
added
to
cultures
after
incubation.
Cultures
were
incubated
at
370C
for
7
days.
The
cultures
were
checked
microscopically
daily
for
the
first
two
days
and
then
every
other
day
thereafter.
After
7
days
all
cultures
were
freeze
thawed
and
10%
of
the
lysate
was
filtered
through
a
0.22
um
filter
and
inoculated
onto
new
cells
for
a
second
passage.
At
the
end
of
the
second
passage
cultures
were
freeze
thawed
twice.
Lysates
were
pooled
and
divided
into
aliquots
for
further
analysis
and
112
shipping.
Samples
were
not
chloroform
extracted.
Sample
aliquots
were
sent
to
UMN
for
enterovirus
analysis
and
TAMU
for
rotavirus
and
reovirus
analysis.

RT­
PCR
Analysis
for
Astrovirus
and
Adenovirus
Nested
PCR
was
performed
on
UNH,
UNC,
UMN
and
TAMU
samples
for
both
Astrovirus
and
Adenovirus
type
40
and
41.
The
equivalent
volume
of
original
water
sample
examined
for
each
virus
was
500
liters.
Positive
controls
we
did
were
at
the
level
of
(
RT­)
PCR.
Virus
was
added
to
cell
culture
lysate
to
act
as
a
positive
control
for
(
RT­
PCR)
PCRNested
PCR
was
performed
on
UNH,
UNC,
UMN
and
TAMU
samples
for
both
Astrovirus
and
Adenovirus
type
40
and
41.
The
equivalent
volume
of
original
water
sample
examined
for
each
virus
was
500
liters.
Positive
controls
we
did
were
at
the
level
of
(
RT­)
PCR.
Virus
was
added
to
cell
culture
lysate
to
act
as
a
positive
control
for
(
RT­
PCR)
PCRNested
PCR
was
performed
on
UNH,
UNC,
UMN
and
TAMU
samples
for
both
Astrovirus
and
Adenovirus
type
40
and
41.
The
equivalent
volume
of
original
water
sample
examined
for
each
virus
was
500
liters.
Positive
controls
we
did
were
at
the
level
of
(
RT­)
PCR.
Virus
was
added
to
cell
culture
lysate
to
act
as
a
positive
control
for
(

RTPCR
PCRNested
PCR
was
performed
on
UNH,
UNC,
UMN
and
TAMU
samples
for
both
Astrovirus
and
Adenovirus
type
40
and
41.
The
equivalent
volume
of
original
water
sample
examined
for
each
virus
was
500
liters.
Positive
controls
we
did
were
at
the
level
of
(
RT­)
PCR.

Virus
was
added
to
cell
culture
lysate
to
act
as
a
positive
control
for
(
RT­
PCR)
PCR
113
Astrovirus.
All
molecular
techniques
were
done
as
specified
in
the
methods
and
materials
developed
by
the
project
team
in
communication
with
the
EPA.
Astrovirus
RT­
PCR
was
done
according
to
Chapron
et
al.
2000.
The
primers
used
were
specific
for
human
astrovirus,
RT
primer
5'­
GTAAGATTCCCAGATTGGT­
3'
and
PCR
primer
5'­

CCTGCCCCGAGAACAACCAAG­
3'.
An
11­:
L
sample
of
the
combined
cell
lysates
was
denatured
with
0.5
:
L
each
of
0.05
M
EDTA
and
downstream
primer
at
99
/

C
for
8
min.

Eighteen
:
L
of
the
RT
mixture
was
then
added
and
run
for
42
min
at
42
/

C
to
reverse
transcribe
and
then
5
min
at
99
/

C.
The
RT
mixture
per
sample
consisted
of
2.5
:
L
10X
buffer
II,
8.5
:
L
of
25mM
MgCl
2,
1.25
:
L
of
each
10mM
dNTP,
0.5
:
L
of
100mM
DTT
(
Promega),
10
units
of
Rnasin,
and
50
units
of
RT.
After
the
RT
step
28.5
:
L
of
a
PCR
master
mix
was
added.
The
PCR
mixture
per
sample
consisted
of
3
:
L
of
10X
buffer
II,
1
:
L
of
the
PCR
primer,
0.5
:
L
of
the
RT
primer,
24
:
L
of
molecular
grade
water,
and
2.5
units
of
Ampli­
Taq
DNA
polymerase.

The
parameters
were
95
/

C,
5
minute
hot
start,
followed
by
35
cycles
of
95
/

C
for
30
seconds,

56
/

C
for
30
seconds,
72
/

C
for
30
seconds,
with
a
final
extension
at
72
/

C
for
5
minutes.
These
primers
yielded
a
193
and/
or
243
bp
amplicon.

For
nested
PCR,
1
:
L
from
each
RT­
PCR
reaction
was
added
to
a
new
tube
containing
90
:
L
of
a
nested
PCR
reaction
mixture,
which
contained
8
mM
MgCl
2,
10
:
L
10x
buffer,
1mM
of
each
dNTP,
2.5
units
of
Ampli­
Taq
DNA
polymerase
and
1
:
M
of
each
primer.
The
primers
used
were
5'­
CCTTGCCCCGAGCCAGAA­
3'
and
5'­
TTGTTGCCATAAGTTTGTGAATA­
3'.
These
primers
yield
a
143
and/
or
183­
bp
amplicon.
Twelve
:
L
of
each
RT­
PCR
product
as
well
as
12
:
L
of
the
nested
PCR
product
was
run
and
sized
by
electrophoresis
on
1.8%
agarose
gel,
stained
with
ethidium
bromide.
Molecular
weights
were
determined
by
comparison
with
a
1
Kb
DNA
114
ladder
(
Life
Technologies).
Astrovirus
serotype
2
was
used
as
a
positive
control.

Adenovirus.
All
molecular
techniques
were
done
as
specified
in
the
methods
and
materials
provided
by
EPA.
Adenovirus
Hexon
PCR
was
done
according
to
Xu
et
al.
2000.
The
primers
used
were
Ad1
5'­
CCCTGGTA(
G/
T)
CC(
A/
G)
AT(
A/
G)
TTGTA­
3'
and
Ad2
5'­

TTCCCCATGGC(
Inosine)
CA(
C/
T)
AACAC­
3'.
A
5:
L
sample
of
the
combined
cell
lysates
was
added
to
47.5:
L
final
volume
PCR
master
mix.
Final
concentrations
in
the
PCR
master
mix
per
sample
were
1.5mM
MgCl
2
,
1x
(
10x
Buffer
II),
0.2mM
dNTP
mix,
0.6:
M
of
each
primer,
and
2.5
units
of
Ampli­
Taq
DNA
polymerase.
The
PCR
parameters
were
95
/

C
for
5
minutes,

followed
by
40
cycles
of
94
/

C
for
1
minute,
55
/

C
for
1
minute,
72
/

C
for
2
minutes,
with
a
final
extension
at
74
/

C
for
5
minutes.
These
primers
yielded
a
482
bp
amplicon.

For
nested
PCR,
1
:
L
from
each
PCR
reaction
was
added
to
a
new
tube
containing
90
:
L
of
a
nested
PCR
reaction
mixture,
which
contained
8
mM
MgCl
2,
10
:
L
10x
buffer,
1mM
of
each
dNTP,
and
1
:
M
of
each
primer.
The
primers
used
were
5'­

GCCACCGAGACGTACTTCAGCCTG­
3'
and
5'­

TTGTACGAGTACGCGGTATCCTCGCGGTC­
3'.
These
nested
primers
were
specific
for
Adenovirus
type
40
&
41.
Samples
were
run
for
35
cycles
of
95
/

C
for
30
seconds,
55
/

C
for
30
seconds,
72
/

C
for
30
seconds
yielding
a
142
bp
amplicon.
Twelve
:
L
of
each
nested
PCR
product
was
run
and
sized
by
electrophoresis
on
1.8%
agarose
gels
and
stained
with
ethidium
bromide.
Molecular
weights
were
determined
by
comparison
with
a
1
Kb
DNA
ladder
(
Life
Technologies).
Adenovirus
40
&
41
were
used
as
positive
controls.
115
Data
management
Results
of
all
analyses
were
entered
into
the
attached
excel
spreadsheet
as
well
as
a
laboratory
notebook.

Results
and
Discussion
Summary
The
objectives
of
this
project
were:
1)
evaluate
EPA
methods
1601
and
1602
for
the
recovery
and
detection
of
male
specific
coliphage,
somatic
coliphage
and
total
)
somatic
and
male­
specific)

coliphage
from
well
water;
2)
Compare
the
efficiency
of
using
a
single
host,
C3000
for
the
detection
of
both
phages;
3)
Compare
EPA
methods
1601
and
1602
for
phage
detection
using
all
three
hosts;
4)
Determine
if
there
is
a
correlation
between
the
detection
of
coliphage
using
either
EPA
Method
1601
or
1602
and
indicator
bacteria
(
fecal
coliforms
and
Enterococcus)
and
5)

Determine
if
there
is
any
correlation
between
the
detection
of
indicator
bacteria,
coliphage
(
using
either
method)
and
certain
enteric
viruses
detected
by
the
Polymerase
Chain
Reaction
Assay
(
PCR).
For
this
work,
25
wells
in
the
Northeast,
some
from
New
Hampshire,
Vermont,
Maine
and
Massachusetts,
were
sampled
and
evaluated
for
each
organism.

An
entire
summary
of
the
results
can
be
found
in
the
accompanying
data
spreadsheet.
All
samples
were
negative
for
Adenovirus
and
Astrovirus
using
an
integrated
cell
culture
Polymerase
Chain
Reaction
Assay
followed
by
a
nested
PCR
assay.

Overall,
9
of
25
samples
or
36%
were
positive
for
one
or
more
fecal
indicator
microbe,
either
a
coliphage
or
a
bacterial
indicator.
Only
one
well
of
25
(
4%)
was
positive
for
coliphage
(
Table
1).
116
This
coliphage­
positive
was
by
the
enrichment
assay
on
hosts
E.
coli
Famp
and
C3000.
The
same
well
was
positive
for
4
plaques/
100
mL
using
Method
1602
on
F+
host
E.
coli
Famp,
but
it
was
negative
for
coliphage
plaques
on
E.
coli
C3000.
The
one
well
that
was
positive
for
coliphage
was
negative
for
both
indicator
bacteria
and
enteric
viruses.
No
well
was
positive
for
coliphage
by
either
method
with
any
other
microorganism
tested.

Eight
wells
of
25
(
32%)
were
positive
for
indicator
bacteria.
Two
wells
were
positive
for
both
fecal
coliforms
and
Enterococcus,
3
wells
were
positive
for
fecal
coliforms
only
and
negative
for
Enterococcus
and
3
wells
were
positive
for
Enterococcus
only
but
negative
for
fecal
coliforms.

Since
so
many
samples
from
community
wells
were
negative
for
all
microorganisms,
the
decision
was
made
to
include
private
drilled
wells
in
the
study.
It
was
hoped
that
because
these
wells,
on
average,
were
probably
less
deep
then
the
community
wells,
that
there
would
be
an
increased
chance
of
detecting
indicator
organisms
as
well
as
enteric
viruses.
While
some
of
the
wells
were
positive
for
indicator
bacteria
and
one
was
positive
for
coliphage
(
though
both
coliphage
and
bacteria
were
not
the
same
well),
no
wells
were
positive
for
enteric
viruses.
Two
of
these
wells
in
VT
were
less
then
100
ft
deep.
These
wells
were
included
on
the
study
in
further
efforts
to
increase
the
probability
of
detecting
enteric
viruses.
Both
of
these
wells
were
negative
for
viruses
and
coliphages
while
one
of
the
wells
was
positive
for
1
fecal
coliform
colony
in
the
1
L
volume
assayed.
To
further
increase
the
probability
of
detecting
enteric
viruses,
a
stoned
lined,

nonpotable
well,
which
was
only
approximately
35
feet
deep
was
also
sampled.
This
well
was
negative
for
all
indicator
bacteria,
coliphage
and
virus.
117
There
were
6
samples
from
public
wells
in
Massachusetts.
The
public
water
supplies
in
Massachusetts
were
chosen
due
to
positive
results
previously
found
for
total
and
fecal
coliforms,

enterococci,
and
male­
specific
coliphages.
Additionally
3
of
the
6
locations
had
positives
previously
reported
for
rotavirus
and
enterovirus
by
molecular
methods.
All
of
these
6
samples
were
negative
for
coliphage
and
enteric
virus.
One
of
these
samples
was
positive
for
both
fecal
coliforms
and
enterococcus,
2
of
the
remaining
5
wells
were
positive
for
enterococcus
but
for
no
other
microorganism.

One
of
the
key
study
objectives
was
to
evaluate
coliphage
occurrence,
by
either
method,
as
an
indicator
for
the
presence
of
enteric
virus.
The
results
of
this
study
indicate
that,
overall,
the
wells
were
not
contaminated
by
enteric
viruses
at
the
time
they
were
sampled.
However,
the
results
did
yield
positive
results
for
coliphage
presence
in
groundwater
in
the
absence
of
detectable
bacterial
indicators
in
that
sample.
However,
bacterial
indicators
were
found
more
frequently
than
coliphages
(
8
samples
versus
1
sample)
and
they
were
found
in
the
absence
of
detectable
coliphages
in
these
samples.
Therefore,
coliphages
and
bacteria
were
not
detected
together
in
any
of
the
samples
analyzed.
This
finding
of
a
lack
of
co­
occurrence
of
coliphages
and
bacteria
in
the
same
sample
supports
the
measurement
of
both
coliphages
and
bacteria
in
groundwater
samples
as
a
way
to
increase
the
likelihood
of
detecting
fecal
contamination.
118
Table
1.
UNH
Coliphage
Results
for
Northeast
Groundwater
Well
Sample
Number
and
Type
Samples
positive
for
phage
by
SAL
(
100
mL)
Samples
positive
for
phage
by
enrichment
(
1
L)

Famp
Cn­
13
C3000
Famp
Cn­
13
C3000
Well
#
3,
Private
0
0
0
1
0
1
Wells
1,
2
and
4­
25
0
0
0
0
0
0
Table
2.
UNH
Groundwater
Wells
Positive
for
Bacterial
Indicators
Positive
UNH
Bacterial
Results
for
Groundwater
Samples
Fecal
Coliforms
Enterococcus
Well
Type
and
Number
100
ML
1
L
100
ML
1
L
Community,
Well
#
1
1
200
0
0
Private;
shallow,
Well
#
4
0
1
0
0
Private
Well
#
9
0
0
0
3
Private,
Well
#
15
35
TNTC
0
0
Private,
Well
#
18
0
0
1
93
Private,
Well
#
19
2
10
13
89
Community,
#
23
0
0
2
87
Community,
#
23
5
69
2
32
119
Table
3.
Results
of
UNH
Groundwater
Samples
Analyzed
for
Adenovirus
and
Astrovirus
Well
#
Results
1
Negative
for
all
viral
analyses
2
Negative
for
all
viral
analyses
3
Negative
for
all
viral
analyses
4
Negative
for
all
viral
analyses
5
Negative
for
all
viral
analyses
6
Negative
for
all
viral
analyses
7
Negative
for
all
viral
analyses
8
Negative
for
all
viral
analyses
9
Negative
for
all
viral
analyses
10
Negative
for
all
viral
analyses
11
Negative
for
all
viral
analyses
12
Negative
for
all
viral
analyses
13
Negative
for
all
viral
analyses
14
Negative
for
all
viral
analyses
15
Negative
for
all
viral
analyses
16
Negative
for
all
viral
analyses
17
Negative
for
all
viral
analyses
18
Negative
for
all
viral
analyses
19
Negative
for
all
viral
analyses
20
Negative
for
all
viral
analyses
21
Negative
for
all
viral
analyses
22
Negative
for
all
viral
analyses
23
Negative
for
all
viral
analyses
24
Negative
for
all
viral
analyses
25
Negative
for
all
viral
analyses
120
Table
4.
Summary
of
UNH
Samples
Positive
for
Coliphages,
Bacterial
Indicators,
Adenoviruses
and/
or
Astroviruses
Well
Number
and
Type
Samples
positive
for:

Coliphage
Bacterial
Indicators
Phage
and
Bacterial
Indicators
Coliphage
and
Virus
Bacterial
Indicator
s
and
Virus
FC
Ent
Any
Well
#
3,
Private
1
0
0
0
0
Well
#
1,
Community
0
1
0
1
0
0
0
Well
#
9,
Private
0
0
1
1
0
0
0
Well
#
15,
Community
0
1
0
1
0
0
0
Well
#
18,
Community
0
0
1
1
0
0
0
Well
#
19,
Community
0
1
1
1
0
0
0
Well
#
21,
Private
0
0
1
1
0
0
0
Well
#
22,
Private
0
1
1
1
0
0
0
Well
#
23,
0
1
1
1
0
0
0
All
other
wells
0
0
0
0
0
0
0
References
Chapron,
C.
D.,
Ballester,
N.
A.,
Fontaine,
J.
H.,
Frades,
C.
N.
&
Margolin,
A.
B.
2000
The
Detection
of
Astrovirus,
Enterovirus
and
Adenovirus
Type
40
and
41
in
Surface
Waters
Collected
and
Evaluated
by
the
Information
Collection
Rule
and
Integrated
Cell
Culture/
Nested
PCR
Procedure.
Appl.
and
Environ.
Microbiol.,
60
(
6),
2520­
2525.

Environmental
Protection
Agency.
1995
Virus
Monitoring
Protocol
for
the
Information
Collection
Requirements
Rule.
U.
S.
Environmental
Protection
121
Agency,
publication
EPA/
814­
B­
95­
002.
Government
Printing
Office,
Cincinnati,

Ohio.

Environmental
Protection
Agency.
2000
Method
1601:
Male­
specific
(
F+)
and
Somatic
Coliphage
in
Water
by
Two­
step
Enrichment
Procedure.
Draft
April
2000.
Office
of
Water,
Washington,

D.
C.

Environmental
Protection
Agency.
2001
Method
1602:
Male­
specific
(
F+)
and
Somatic
Coliphage
in
Water
by
Single
Agar
Layer
(
SAL)
Procedure.
Draft
January
2001.
Office
of
Water,

Washington,
D.
C.

Xu,
W.,
McDonough,
M.
C.,
&
Erdman
D.
D.
2000.
Species­
specific
identification
of
human
adenoviruses
by
a
multiplex
PCR
assay.
J.
Clin.
Microbiol.
39,
4114­
4120.
122
APPENDIX
II
EPA
OST
Groundwater
Coliphage
Project
Coliphage,
Bacteria
and
Human
Enteric
Virus
Isolation
from
Ground
Water 

Southwest
Region
Laboratory
Texas
A&
M
University
 
Prof.
Suresh
Pillai,
Co­
PI
Introduction
and
Background
Ground
water
samples
for
microbial
analyses.

The
original
goal
of
this
study
was
for
each
of
the
four,
regionally
representative
laboratories
(
southeast,
northeast,
upper
Midwest
and
southwest)
to
collect
and
analyze
27
ground
water
samples
from
public
water
supply
wells.
Efforts
were
made
to
identify
candidate
public
water
supplies
that
previously
had
coliform
bacteria
violations
or
other
evidence
of
vulnerability
to
fecal
contamination.
In
some
cases
candidate
wells
were
prescreened
by
bacteriological
and
coliphage
analyses
for
evidence
of
fecal
contamination.
Because
not
all
participating
labs
could
identify
and
get
access
to
27
public
water
supply
wells,
some
labs
also
included
non­
public
and
private
wells
in
their
sampling.
Three
labs
obtained
27
ground
water
samples
and
one
lab
obtained
a
total
of
25
samples
for
a
total
of
106
samples
overall.
The
characteristics
of
the
wells
that
were
sampled
are
presented
in
data
tables
in
the
Results
section
of
this
report.
This
report
describes
the
methods,

materials,
coliphage
and
bacterial
indicator
and
enteric
virus
results
for
samples
from
the
southwest
region.
The
results
for
bacteria
and
coliphage
analyses
of
these
samples
are
also
presented
in
an
123
Excel
spreadsheet
that
accompanies
this
document
and
in
the
PowerPoint
presentation
that
was
delivered
at
the
April,
2004
EPA
Coliphage
Workshop.

Methods
and
Materials
Sampling
sites
and
wells
in
Texas
and
New
Mexico:
Only
PWS
wells
were
included
in
this
study,

and
a
total
of
eleven
different
PWS
wells
were
identified
for
this
study.
The
sampling
sites
were
located
in
the
San
Antonio
region
of
Texas
(
wells
RS,
KK,
and
HCR)
and
along
the
US­
Mexico
border
in
southern
New
Mexico
(
wells
MHPa,
MHPb,
MHPc,
FVE,
AVC,
SME,
and
LME).
The
wells
in
the
San
Antonio
region
were
part
of
a
karst
aquifer
and
were
previously
implicated
in
a
documented
groundwater
contamination
event.
Also,
during
the
initial
pre­
screening
of
the
wells
some
of
the
samples
were
positive
for
somatic
and
male­
specific
coliphages.
The
wells
in
southern
New
Mexico
were
identified
as
being
vulnerable
to
groundwater
contamination
based
on
parameters
such
as
closeness
to
septic
tanks,
proximity
to
the
Rio
Grande
river
and
the
aquifer
in
question.
These
wells
were
part
of
a
previous
EPA­
funded
project
on
the
microbiological
quality
of
wells
in
the
shallow
aquifer
along
the
US­
Mexico
border
during
which
some
of
the
wells
in
the
sampling
area
were
positive
for
enterococci,
E.
coli,
male­
specific
coliphages
and
somatic
coliphages.
The
wells
were
in
the
100­
150
feet
depth
range.
The
static
water
levels
were
around
10­
20
feet
and
in
terms
of
their
hydrogeologic
setting,
they
were
located
in
the
Rio
Grande
alluvium/
Hueco­
Tularosa
aquifers.

Sampling:
Groundwater
samples
were
collected
between
June
2002
and
January
2003.
Multiple
samples
were
collected
from
each
of
the
wells
to
be
representative
of
the
aquifer
and
the
sampling
124
location.
During
each
sampling
adequate
volumes
were
collected
for
the
coliphage
analysis
as
well
as
for
the
enteric
virus
analysis.
Grab
samples
were
collected
for
the
coliphage
and
bacterial
analysis
while
the
1MDS
filters
were
used
for
collecting
the
large
volume
enteric
virus
samples.

Microbiological
Analysis:
The
USEPA
methods
1601
and
1602
were
used
for
the
coliphage
analysis.
The
host
bacteria
used
in
these
analyses
included
E
.
coli
Famp
for
F+
coliphage,
E
.
coli
CN­
13
for
somatic
coliphages
and
E.
coli
C­
3000
for
"
total"
coliphages.
The
samples
were
also
analyzed
for
E
.
coli
and
Enterococcus
spp
using
the
membrane
filtration
protocol
and
M­
coli
Blue
and
MEI
agars,
respectively.

Virus
concentration
from
ground
water
samples.
Viruses
were
concentrated
from
1500­
liter
ground
water
samples
by
filtering
the
water
at
its
ambient
pH
using
standard
1­
MDS
electropositive
cartridge
filters
(
ZetaPor
Virosorb,
Cuno
Product
No.
45144­
01­
1MDS)
and
using
procedures
described
by
the
US
Environmental
Protection
Agency
(
2001).
In
cases
where
the
sample
size
is
not
1500
liters,
the
filtered
volume
is
specified
in
the
results
section
of
this
report.

Filter
elution
and
concentration.
Human
enteric
viruses
were
eluted
from
the
Cuno
1­
MDS
cartridge
filters
with
a
solution
of
1.5%
beef
extract
(
Becton
Dickinson)
plus
0.05
M
glycine
at
pH
9.5.
The
eluent
was
allowed
to
contact
the
filter
cartridge
for
a
minimum
of
six
minutes.
Viruses
in
the
beef
extract­
glycine
eluate
were
subsequently
concentrated
into
a
smaller
volume
by
acid
precipitation
(
organic
flocculation).
Briefly,
the
eluates
were
adjusted
to
pH
3.5,
stirred
slowly
for
30
minutes,
then
centrifuged
at
6,200
x
g
for
20
minutes.
The
resulting
pellets
were
resuspended
with
15
mL
of
0.15
M
Na
2
HPO
4
,
adjusted
to
pH
9.0­
9.5,
and
centrifuged
at
6,200
x
g
for
15
125
minutes
to
remove
residual
particulates.
The
resulting
supernatants
were
adjusted
to
pH
7.0­
7.5,

and
filtered
through
0.2:
m
pore
size,
serum
Acrodisc
syringe
filters
(
Pall)
to
remove
bacterial
and
fungal
contaminants.
Each
concentrated
sample
of
18­
20
mL
was
subdivided
into
aliquots
for
subsequent
detection
of
specific
enteric
virus
groups,
then
stored
at
 
80
/.
Aliquots
of
the
concentrate
were
shipped
to
participating
labs
for
their
respective
viral
analyses
by
integrated
cell
culture
and
(
RT­)
PCR
(
for
hepatitis
A
virus
[
HAV],
enteroviruses,
adenoviruses,
rotaviruses,

reoviruses
and
astroviruses)
and
direct
RT­
PCR
for
noroviruses
(
human,
Norwalk­
like
caliciviruses).

Infectivity
assays
in
BGMK
and
Caco­
2
cell
cultures.
One­
third
of
each
sample
concentrate,

equivalent
to
500
L
of
source
water,
was
inoculated
into
cultures
of
the
Buffalo
Green
Monkey
Kidney
(
BGMK)
continuous
cell
line,
and
another
third
was
inoculated
into
cultures
of
the
Caco­
2
continuous
cell
line.
Another
portion,
corresponding
to
100
liters
of
ground
water
,
was
used
for
cell
culture
plus
RT­
PCR
analysis
of
hepatitis
A
virus
(
HAV),
and
another
portion,
also
corresponding
to
100
liters
of
ground
water,
was
used
for
direct
RT­
PCR
of
human
caliciviruses
(
noroviruses).
The
remaining
one­
fifth
of
the
sample
was
archived
as
a
contingency
for
possible
future
analysis.
Sample
concentrates
for
cell
culture
inoculation
were
pre­
activated
by
adding
type
IX
trypsin
(
Sigma
T­
0303)
to
a
10
:
g/
mL
concentration,
and
incubating
at
37
/

for
30
minutes
prior
to
inoculation.
Newly
confluent
layers
of
each
cell
type
in
75
cm2
tissue
culture
flasks
were
rinsed
three
times
with
Dulbecco's
phosphate
buffered
saline
(
PBS)
supplemented
with
magnesium
and
calcium
(
Gibco)
to
remove
residual
calf
serum
associated
with
the
cell
growth
medium.
The
cultures
were
inoculated
with
trypsin
pre­
activated
concentrate,
and
incubated
at
37
/

for
80
minutes.
Serum­
free
maintenance
MEM
medium
with
Earle's
salts
supplemented
with
5:
g/
mL
126
type
IX
trypsin
was
added.
Inoculated
cultures
were
incubated
for
7
days
at
37
/

with
periodic
microscopic
examination
for
evidence
of
viral
cytopathology.

After
seven
days,
the
inoculated
cultures
were
frozen
and
thawed
twice.
Newly
confluent
PBSrinsed
layers
of
the
same
cell
line
were
inoculated
with
10%
of
the
cell
culture
lysate
from
each
first
passage
(
initial)
culture,
calf
serum­
free
medium
supplemented
with
trypsin
was
added,
and
the
cultures
were
incubated
at
37
/

for
a
second
passage
of
the
sample
material.
The
second
passage
cultures
were
periodically
observed
microscopically,
then
frozen
seven
days
after
inoculation.

All
first
and
second
passage
cell
cultures
were
frozen
and
thawed
twice.
A
single
lysate
pool
of
about
35
mL
was
prepared
for
each
ground
water
sample
by
combining
10%
of
the
lysate
from
both
first
and
second
passage
BGMK
and
Caco­
2
cultures
that
had
been
inoculated
with
a
specific
water
sample
concentrate.
A
10­
mL
portion
of
each
lysate
pool
was
extracted
with
5
mL
of
chloroform,
and
centrifuged
at
1,800
x
g
for
15
minutes.
Each
sample
extract
was
subdivided
into
aliquots
for
isolation
of
viral
nucleic
acid
and
viral
nucleic
detection
using
the
nucleic
acid
amplification
methods
of
either
polymerase
chain
reaction
(
PCR)
for
DNA
viruses
(
adenoviruses)

or
reverse
transcription
PCR
(
RT­
PCR)
by
other
participating
laboratories,
and
stored
at
 
80
/.

Tissue
culture
protocol
for
virus
isolation
in
BGMK
and
CaCo­
2
cells
at
Texas
A&
M
University.
The
groundwater
concentrates
(
equivalent
to
500L)
were
initially
pre­
tested
for
cytotoxicity
after
an
initial
pre­
activation.
(
No
cytotoxicity
tests
were
done
prior
to
the
CaCo­
2
cell
cultures
since
none
of
the
samples
were
positive
for
cytotoxicity
on
BGMK
cells).
127
Preactivation
was
done
using
0.5
mL
and
1.0
mL
of
the
groundwater
concentrate.
The
0.5ml
sample
was
added
to
5:
Lof
trypsin
and
the
1.0
mL
sample
was
added
to
10ul
of
trypsin.
The
samples
were
incubated
for
30
minutes
and
then
refrigerated
prior
to
the
cytotoxicity
tests.
The
T25
flasks
(
having
80%
confluency)
were
washed
twice
with
5
mL
of
Hanks
Balanced
Salt
Solution
(
HBSS).
The
cells
were
inoculated
with
0.5ml
of
the
preactivated
sample,
and
incubated
for
90
minutes
with
cells
being
rocked
every
15
minutes.
After
the
90­
minute
incubation,
5ml
of
MEM
complete
(
serum
free
with
0.25:
L/
mL
of
trypsin)
was
added
and
the
cells
were
observed
for
2
days.
Cytotoxicity
was
evaluated
using
a
sterile
HBSS
­
inoculated
"
negative
control."

Each
of
the
T75
flasks
were
washed
with
15
mL
of
HBSS
two
times.
The
HBSS
was
siphoned
off
and
the
flasks
were
inoculated
with
the
remainder
of
the
sample
across
3
flasks.
Two
negative
controls
(
1
before
inoculation
of
sample
and
1
after
inoculation
of
sample)
were
also
included.

The
flasks
were
incubated
for
90
minutes
at
37oC
with
5%
CO
2
and
rocking
every
15
min.
After
the
90­
minute
incubation,
15
mL
of
MEM
(
serum
free
with
0.25:
L/
mL
of
trypsin)
was
added.
The
flasks
were
incubated
at
37oC
for
7
days
and
observed
every
day
for
cytopathic
effects
(
CPE).
The
same
procedure
was
followed
for
CaCo­
2
cells
as
well
The
samples
were
passaged
a
second
time
by
freeze
thawing
once
and
removing
approximately
10%
of
the
lysate
from
the
original
flaks
and
placed
in
new
100%
confluent
flasks
that
were
washed
as
mentioned
previously.
The
samples
were
incubated
for
90
minutes,
rocking
every
15
minutes
and
15
mL
of
MEM
(
serum
free
containing
25:
L/
mL
of
trypsin)
was
subsequently
added.

The
samples
for
incubated
for
another
7
days
and
observed
by
microscopy
daily.
128
The
samples
were
passaged
a
third
time
by
removing
the
lysate
from
the
second
passage,
filter
sterilized
through
a
100
mm
diameter,
0.22
:
m
pore
size
cellulose
ester
filter
into
T75
flasks
that
were
prepared
as
before.
The
flasks
were
placed
in
the
incubator
at
37oC
for
5­
7
days
and
observed
for
cytopathic
effects.

Table
1.
Tissue
Culture
Results
for
Virus
Isolation
from
Ground
Water
Samples
Based
on
Microscopic
Observation
Only
Sample
ID
Sample
volume
CPE
Results
(
BGMK
and
CaCo­
2)

BGMK
cells
CaCo­
2
cells
Passage
#
1
Passage
#
2
Passage
#
3
RS
(
1)
4.75
5.5
+
+
+
HCR
(
1)
Groundwater
concentrate
sample
lost
due
to
centrifuge
tube
breakage
RS
(
2)
Groundwater
concentrate
sample
lost
due
to
centrifuge
tube
breakage
BM
(
1)
5.0
6.0
+
+
+
KK
(
1)
4.75
6.0
+
+
+
RS
(
3)
5.0
5.5
+
+
+
KK
(
2)
5.0
5.5
+
+
+
HCR
(
2)
6.0
6.5
+
+
+
RS
(
4)
5.0
5.5
+
+
+
RS(
5)
5.0
4.5
+
+
+
MHP1a
5.0
5.5
+
+
+
MHP1b
5.0
6.5
+
+
+
AVC1
5.0
6.0
+
+
+
FVE1
5.0
6.0
+
+
+
AVC2
4.0
4.5
+
+
+
FVE2
4.5
6.0
+
+
+
FVE3
6.5
7.75
+
+
+
AVC3
4.5
5.0
+
+
+
MHP1c
5.5
6.0
+
+
+
MHP2a
4.75
5.5
+
+
+
MHP2c
4.8
5.0
+
+
+
MHP3a
5.3
5.5
+
+
+
MHP2b
6.1
6.5
+
+
+
SME1
4.8
5.0
+
+
+
SME2
6.0
6.0
+
+
+
LME1
4.6
5.25
+
+
+
MHP3b
4.3
4.5
+
+
3
+
=
indicates
possible
cytopathic
effect.
129
Viral
RNA
Extraction
for
Rotavirus
and
Reovirus
detection
by
RT­
PCR
The
cell
culture
extracts
from
the
BGMK
cells
(
from
passage
#
1
and
passage
#
3)
(
1
mL
each)

were
combined
with
2
mL
from
CaCo­
2
cell
lysates
and
to
this
was
added
to
2
mL
of
chloroform.

The
mixture
was
vortexed
for
2
min
at
high
speed,
then
centrifuged
at
18K
rpm
for
20
minutes.

The
top
layer
was
pipetted
out
and
aliquotted
into
4
cryo­
tubes
(
1
mL
each).
(
Samples
1­
9
that
was
sent
from
Texas
A&
M
University
contained
only
extracts
from
CaCo­
2
cells
due
to
a
laboratory
error).
One
cryo­
tube
of
each
sample
was
shipped
to
UNC,
Univ.
of
Minnesota
and
UNH.

The
QiAmp
viral
RNA
extraction
kit
was
used
for
RNA
extraction
from
the
cell
culture
lysates
per
the
manufacturer's
recommended
protocols
(
Qiagen,
Valencia,
CA).
The
final
extract
was
resuspended
in
80
µ
L
of
buffer,
which
was
stored
at
­
80C
until
the
RT­
PCR
analyses.

RT­
PCR
Analysis
for
Rotaviruses
and
Reoviruses
Rotavirus
analysis.
For
Rotavirus,
3.5
:
L
of
the
RNA
extract
was
used.
Separate
RT
and
PCR
amplifications
were
performed
with
10
and
50
:
L
total
reaction
volumes,
respectively.
The
final
concentrations
in
the
RT
step
were
5mM
(
1X
PCR
Buffer
II),
5mM
MgCl
2
,
1mM
of
each
dent,

1.26
:
M
of
3'
rotavirus
primer,
45
units
of
Reverse
Transcriptase,
RNAse
inhibitor
(
18
units).
The
sample
was
"
hot­
started"
(
95oC
for
5
min)
and
when
the
temperature
reached
60oC,
reverse
transcriptase,
RNAse
inhibitor
and
dNTP
were
added.
The
RT
step
was
conducted
at
42oC
for
60
minutes.
A
wax
layer
was
used
to
prevent
accidental
aerosolization
of
samples
when
the
tubes
130
were
subsequently
opened.
The
samples
were
heated
at
95oC
for
5
minutes.
The
sample
was
maintained
at
80oC.
The
PCR
master
mix
was
then
added
to
this
sample.
The
final
concentration
in
the
sample
after
the
addition
of
the
PCR
master
mix
was
2mM
MgCl
2
,
1X
PCR
Buffer
II,
2.5
units
of
Taq
DNA
polymerase,
and
0.25:
M
of
5'
Rotavirus
primer.
The
cycling
conditions
were
95oC
for
1.5
min,
55C
for
1.5
min
and
72C
for
1.5
minutes.
Forty
PCR
cycles
were
performed.
The
PCR
products
were
run
on
a
pre­
made
(
6
%)
Novex
TBE
gels
(
Invitrogen,
Valencia,
CA)
for
detection
of
the
208
bp
product.

The
controls
included
a
Rotavirus
RNA­
spiked
positive
control
and
a
water
negative
control.

Additionally,
MS2
RNA
was
spiked
into
a
select
number
of
samples
to
detect
any
possible
sample
inhibition.
Primers
directed
to
the
capsid
gene
of
the
MS2
RNA
were
used
for
this
purpose
(
Valenzuela
and
Pillai,
1998).

Reovirus
Analysis.
A
volume
of
5ul
of
the
RNA
extract
was
used
for
RT­
PCR
analysis.
Separate
RT
and
PCR
amplifications
were
performed,
with
10
:
l
and
50
:
l
reaction
volumes
respectively.

The
final
concentration
of
the
RT
components
were
1.5
mM
Mgcl
2
,
1X
of
PCR
Buffer
II,
0.7
mM
of
each
dNTP
and
1.7:
M
of
the
3'
Reovirus
primer.
A
wax
layer
was
used
to
prevent
accidental
aerosolization
during
subsequent
handling.
The
samples
were
heated
at
99oC
for
5
minutes
and
then
placed
on
ice.
Once
the
samples
were
cooled,
RNAse
inhibitor
(
22
units),
and
50
units
of
Reverse
Transcriptase
were
added.
The
RT
conditions
were
43oC
for
60
minutes.
The
samples
were
subsequently
heated
for
5
minutes
at
95oC
and
placed
on
ice.
The
PCR
master
mix
was
then
added
to
this
sample,
giving
a
final
volume
of
50
microliters.
The
final
concentrations
of
the
PCR
mix
ingredients
were
1.5mM
of
MgCl
2
,
1X
PCR
Buffer
II,
05
:
M
of
the
5'
Reovirus
primer
and
131
5.0
units
of
the
Taq
DNA
polymerase.
The
PCR
amplification
conditions
were
95oC
for
1
minute,

55oC
for
1.5
minutes,
72oC
for
1.5
minutes.
Forty
PCR
cycles
were
performed.
The
PCR
products
were
resolved
on
a
6%
TBE
premade
Novex
gels.

RESULTS
Enteric
viruses
None
of
the
27
groundwater
samples
from
either
Texas
or
New
Mexico
were
positive
by
cell
culture
and
(
RT­)
PCR
for
adenoviruses,
astroviruses,
enteroviruses,
reoviruses,
rotaviruses
or
hepatitis
A
virus
or,
in
the
case
of
caliciviruses,
by
direct
RT­
PCR
analysis.
The
presumptive
positive
results
for
cytopathic
effects
shown
in
Table
1
must
have
been
due
to
non­
viral
effects
such
as
the
action
of
the
trypsin
in
the
medium.
Trypsin
can
cause
the
cells
to
dislodge
from
the
surface
of
the
flask
and
appear
abnormal,
or
to
cytotoxicity
from
the
sample
concentrate
inocula.

Regardless
of
the
cause
of
this
effect,
it
was
not
due
to
the
presence
of
any
of
the
viruses
for
which
samples
were
analyzed
by
nucleic
acid
methods.

Bacterial
and
Coliphage
Indicators
in
Groundwater.

The
results
for
bacterial
and
coliphage
indicators
in
positive
samples
are
summarized
in
Table
2.
In
all,
7
of
27
samples
(
26%)
were
positive
for
at
least
one
indicator
microbe.
132
Table
2:
Summarized
data
showing
groundwater
wells
that
were
positive
for
bacterial
and/
or
viral
(
coliphage)
indicators.

Sample
Enterococci
(
Number/
100
mL)
E.
coli
Number/
100
mL
Method
1602
Number/
100
mL
Method
1601
Positive
(+)
or
Negative
(­)
per
Indicated
Volume
Famp
CN­
13
C3000
Famp
CN­
13
C3000
100
mL
1000
mL
100
mL
1000
mL
100
mL
1000
mL
RS(
1)
0
0
0
0
0
­
­
+
­
­
­

KK
(
2)
0
0
0
0
0
­
­
­
+
­
­

HCR(
2)
1
0
0
1
0
­
­
+
+
+
­

MHPa(
1)
5
1
0
0
0
­
­
­
­
­
­

RS
(
5)
0
1
0
0
0
­
­
­
­
­
­

AVC(
1)
0
0
0
0
0
­
­
­
+
­
­

AVC(
3)
0
0
0
0
0
+
­
+
­
­
1000
Bacterial
Indicators.
Out
of
27
samples
that
were
analyzed,
only
2
sample
(
7.4%)
were
positive
for
E.
coli
and
2
samples
(
7.4%)
were
positive
for
Enterococci.
There
was
only
1
sample
that
was
positive
for
both
E.
coli
and
Enterococci.
The
maximum
density
of
E.
coli
in
a
sample
was
1
CFU/
100
mL
compared
to
Enterococci,
which
showed
a
maximum
density
of
5
CFU/
100
mL.

Viral
(
Coliphage)
Indicators.
Out
of
27
samples,
5
samples
were
positive
for
coliphages.
There
was
only
1
sample
that
was
positive
for
male­
specific
coliphages
(
based
on
detection
of
a
plaque
on
E.
coli
host
Famp).
This
is
in
contrast
to
5
samples
that
were
positive
for
somatic
coliphages
(
based
on
plaques
on
E.
coli
host
CN­
13
or
growth
in
enrichment
cultures)
while
2
samples
were
positive
for
"
all"
coliphages
based
on
E.
coli
host
C­
3000.
Four
samples
were
positive
for
coliphages
when
1000
mL
was
analyzed
compared
to
3
samples
that
were
positive
when
only
100
mL
samples
were
analyzed.
Two
of
the
samples
were
positive
when
100
mL
and
1000
mL
aliquots
133
of
the
sample
were
screened
for
coliphages.

Comparison
of
Bacterial
and
viral
Indicator
Results:
Table
2
shows
the
results
from
the
bacterial
and
viral
(
coliphage)
indicator
analyses
so
that
the
two
types
of
indicators
can
be
compared.
Out
of
27
samples
that
were
analyzed
for
bacterial
and
viral
indicators,
7
(
25.9%)
were
positive
for
either
bacterial
or
viral
indicators.
Only
3
of
the
samples
(
11.1%)
were
positive
for
either
of
the
bacterial
indicators
(
E.
coli
or
enterococci)
while
5
samples
(
18.5%)
were
positive
for
coliphages
(
either
by
Method
1601
or
Method
1602).
Four
samples
(
14.8%)
were
positive
for
coliphages
but
negative
for
bacterial
indicators.
This
is
in
comparison
to
only
2
samples
(
7.4%)
that
were
positive
for
bacterial
indicators
but
negative
for
viral
indicators.

These
results
suggest
that
coliphages
can
be
used
as
a
tool
for
screening
ground
water
samples
for
the
presence
of
fecal
contamination.
The
results
strongly
suggest
that
coliphage
analysis
should
be
conducted
along
with
or
in
addition
to
conventional
bacterial
indicator
analysis.
This
is
because
the
inclusion
of
coliphages
increases
the
likelihood
of
detecting
a
contaminated
samples,
based
on
the
presence
of
either
bacteria
or
coliphage
indicators.
The
total
absence
of
human
enteric
viruses
in
the
presence
of
the
selected
indicator
organisms
suggest
that
it
is
highly
unlikely
that
pathogens
would
be
detected
routinely.
It
is
possible
that
only
under
heavily
contaminated
conditions
would
there
be
a
direct
correlation
or
co­
occurrence
between
the
presence
of
viral
pathogens
and
fecal
indicator
organisms.
134
References
Chapron,
C.
D.,
Ballester,
N.
A.,
Fontaine,
J.
H.,
Frades,
C.
N.,
and
A.
B.
Margolin
(
2000)
Detection
of
astroviruses,
enteroviruses,
and
adenovirus
types
40
and
41
in
surface
waters
collected
and
evaluated
by
the
information
collection
rule
and
an
integrated
cell
culture­
nested
PCR
procedure.

Appl.
Environ.
Microbiol.,
66(
6):
2520­
5.

Schwab,
K.
J.,
R.
De
Leon,
and
M.
D.
Sobsey.
1995.
Concentration
and
purification
of
beef
extract
mock
eluates
from
water
samples
for
the
detection
of
enteroviruses,
hepatitis
A
virus
and
Norwalk
virus
by
reverse
transcription­
PCR.
Appl.
Environ.
Microbiol.
61:
531­
537.

US
EPA
(
2001)
USEPA
Manual
of
Methods
for
Virology,
Chapter
14.
EPA
600/
4­
84/
013
(
N14),

April
2001,
Office
of
Research
and
Development,
Washington
DC
45260
Xu,
W.,
McDonough,
M.
C.,
and
D.
D.
Erdman
(
2000)
Species­
specific
identification
of
human
adenoviruses
by
a
multiplex
PCR
assay.
J.
Clin.
Microbiol.,
38(
11):
4114­
20.
Erratum
in:
J.
Clin.

Microbiol.,
2001,
Apr.;
39(
4):
1686
Valenzuela,
R.
B.,
and
S.
D.
Pillai.
1998.
Persistence
of
naked
viral
RNA
molecules
in
groundwater.
8th
Intl
symposium
on
Microbial
Ecology.
Nova
Scotia.
August
Vinje
J,
Koopmans
MP.
(
1996)
Molecular
detection
and
epidemiology
of
small
round­
structured
viruses
in
outbreaks
of
gastroenteritis
in
the
Netherlands.
J
Infect
Dis.
1996
Sep;
174(
3):
610­
5.
135
APPENDIX
III
EPA
Coliphage
Method
Validation
Project
Report:
Detection
of
Coliphages,
Indicator
Bacteria
and
Enteric
Viruses
in
Groundwater
AUTHORS
Sagar
M.
Goyal,
DVM,
PhD
Yashpal
Malik,
DVM,
PhD
Baldev
R.
Gulati,
DVM,
PhD
Sunil
Maherchandani,
DVM,
PhD
Sigrun
Haugerud,
BS
University
of
Minnesota
And
Mark
D.
Sobsey
University
of
North
Carolina
STUDY
COMPLETED
ON
June
30,
2003
PERFORMING
LABORATORY
Department
of
Veterinary
Diagnostic
Medicine
College
of
Veterinary
Medicine,
University
of
Minnesota
1333
Gortner
Avenue,
St.
Paul,
MN
55108,
USA
Contact
Information
Sagar
M.
Goyal
Department
of
Veterinary
Diagnostic
Medicine
College
of
Veterinary
Medicine,
University
of
Minnesota,
1333
Gortner
Avenue,
St.
Paul,
MN
55108,
USA.
Phone:
612­
625­
2714;
Fax:
612­
624­
8707
Email:
goyal001@
umn.
edu
Mark
D.
Sobsey
University
of
North
Carolina
CB#
7431,
McGavran­
Greenberg
Hall,
Room
4114a,
Chapel
Hill,
NC
27599­
7431
Telephone
:
919­
966­
7303
Email:
Mark_
Sobsey@
unc.
edu
136
Purpose:
To
determine
if
FRNA
phages
are
useful
indicators
of
fecal
contamination
and
human
enteric
viruses
by
testing
well
water
samples
for
the
presence
of
fecal
coliforms,
Escherichia
coli,

Enterococcus,
somatic
coliphages,
FRNA
coliphages,
"
total"
coliphages
and
human
enteric
viruses.

Materials
and
Methods
Source
of
samples.
Ground
water
samples
were
collected
from
27
candidate
wells
(
address
with
contact
numbers
are
provided
in
Table
1).
All
wells
except
6
private
ones
in
Minnesota
are
considered
public
water
supplies
by
the
State
of
Minnesota
and
none
are
disinfected.

Recovery
of
Enteric
Viruses.
From
each
well
1,500
liters
of
water
was
pumped
through
a
1­

MDS
filter
cartridge
followed
by
virus
elution
in
1.5%
beef
extract­
0.05
M
glycine
solution.

Another
5
liter
sample
of
water
was
collected
from
each
well
in
a
sterile
container
for
bacteriological
and
coliphages
analysis.
These
samples
were
maintained
at
40C
until
analyzed,

usually
within
24
hrs
of
collection.
The
results
are
given
in
Table
2.

Bacteriological
evaluation.
Grab
samples
of
water
were
analyzed
for
fecal
coliforms,
E.
coli
and
Enterococcus
using
Membrane
Filter
(
MF)
technique
as
recommended
in
chapter
9
of
Standard
Methods
for
the
Examination
of
Water
and
Wastewater
(
American
Public
Health
Association,

1998).
Briefly,
a
100
mL
volume
of
a
water
sample
was
filtered
through
a
0.45
:
m
pore
size,
47
mm
diameter
membrane
filter.
These
filters
were
then
placed
on
plates
of
selective
mFC
agar
for
fecal
coliforms,
mEC
for
E.
coli
and
mE
media
for
Enterococcus.
For
fecal
coliforms,
the
plates
137
were
incubated
at
44.50C
for
fecal
coliforms
and
E.
coli
and
at
41.50C
for
Enterococcus..
The
number
of
characteristic
colonies
was
counted
following
incubation
for
24
hrs
and
concentrations
are
expressed
as
colony­
forming
units
per
100
mL.
All
media
were
obtained
from
Becton
Dickinson,
Cockeysville,
MD.

Coliphages
analyses.
All
27grab
samples
were
analyzed
for
the
presence
of
FRNA
(

malespecific
coliphages,
somatic
coliphages
and
"
total"
coliphages
using
single
agar
layer
procedure
and
enrichment
method
(
Methods
1601
and
1602;
Environmental
Protection
Agency,
2001a;

2001b).
The
host
bacteria
were
E.
coli
F
amp
(
ampicillin
and
streptomycin
resistant
mutant
of
E.
coli;

ATCC
700891)
for
FRNA
coliphages,
CN13
(
nalidixic
acid
resistant
mutant
of
E
coli;
ATCC
700609)
for
somatic
coliphages,
and
E.
coli
C3000
(
ATCC
15597)
for
"
total"
coliphages.
A
log
phase
culture
of
the
host
bacterium
was
prepared
by
inoculating
a
stock
of
the
bacteria
into
30
mL
of
trypticase
soy
broth
followed
by
incubation
for
4
hrs
at
370C
on
a
shaker
platform.
To
100
mL
aliquots
of
water
samples
were
added
0.5
mL
of
4
M
MgCl
2
,
10
mL
of
log
phase
culture
of
host
bacteria,
and
100
mL
of
molten
and
cooled
double
strength
tryptic
soy
agar.
The
sample
was
thoroughly
mixed
and
poured
into
four
150­
mm
Petri
plates
followed
by
incubation
at
370C
for
24
hrs.
Positive
results
were
indicated
by
circular
zones
of
lysis
in
contrast
to
opaque
lawn
of
host
bacterial
growth.
Plaques
from
all
four
plates
were
counted
for
each
sample.
Plaques
were
confirmed
by
picking
them,
resuspending
the
picked
material
in
100
ul
of
TSB,
spotting
onto
prepoured
lawns
of
the
respective
host
bacterium,
incubating
for
4
hours
at
370C,
and
observing
the
spots
for
evidence
of
coliphage
presence
as
lysis
zones
or
plaques.
138
In
the
other
method
of
coliphages
testing,
the
enrichment­
spot
plate
method,
12.5
mL
of
MgCl
2
,

50
mL
of
10X
TSB
and
10
mL
of
ampicillin/
streptomycin
or
nalidixic
acid
and
5
mL
of
host
culture
E.
coli
Famp,
CN13
or
C3000
were
added
to
1­
liter
aliquots
of
water.
After
incubation
at
370C
for
24
hrs,
10
:
L
of
the
culture
was
spotted
on
freshly
prepared
Spot
plates
of
the
respective
host
culture
(
E.
coli
Famp,
N13
or
C3000).
Positive
results
were
indicated
by
circular
zones
of
lysis
in
contrast
to
opaque
lawn
of
host
bacterial
growth.
.
((
Plaques
were
confirmed
by
picking
them,
resuspending
the
picked
material
in
100
:
L
of
TSB,
spotting
onto
prepoured
lawns
of
the
respective
host
bacterium,
incubating
for
4
hours
at
370C,
and
observing
the
spots
for
evidence
of
coliphage
presence
as
lysis
zones
or
plaques.
This
method
is
qualitative
in
nature
because
it
scores
sample
volumes
as
either
positive
or
negative
for
coliphages.

Enteric
virus
isolation.
Viruses
were
isolated
from
groundwater
samples
using
the
US
EPA
ICR
Method
with
minor
modifications
(
US
EPA,
1996).
After
filtering
1,500
liters
of
water
through
the
CUNO
1­
MDS
filter,
adsorbed
viruses
were
eluted
from
the
filter
with
1.5%
beef
extract­
0.05
M
glycine
solution
(
pH
9.5).
The
eluate
was
further
concentrated
using
the
acid
precipitation
method.
All
concentrates
were
suspended
in
the
same
volume
(
22
mL)
of
sodium
phosphate
buffer.

The
final
sample
was
filter
sterilized
using
a
25
mm
diameter
0.22
micrometer
pore
size
Gelman
Acrodisc
filter.
Two
aliquots
of
2
mL
each
were
sent
to
UNC
for
detection
of
human
caliciviruses
(
noroviruses)
and
Hepatitis
A
viruses.
Two
aliquots
of
7.5
mL
each,
corresponding
to
500
liters
of
groundwater,
were
used
for
culturable
virus
isolation
by
inoculation
of
BGM
and
Caco­
2
cell
lines.

All
samples
were
passaged
twice
in
BGM
and
Caco­
2
cells,
with
incubation
periods
of
one
week
per
passage.
The
culture
fluids
were
pooled
separately
(
one
sample
passaged
twice
in
BGMK
and
Caco­
2
was
pooled).
Pooled
lysates
were
chloroform
extracted
and
aliquots
of
2
mL
each,
139
corresponding
to
100
liters
of
groundwater,
were
sent
to
UNH
and
TAMU
for
detection
of
adenoviruses,
reovirus,
rotavirus
and
astrovirus.
Cell
culture
lysates
from
UNC,
UNH,
and
TAMU
were
also
received
for
detection
of
enteroviruses
by
the
cell
culture
and
RT­
PCR
methods
described
here.
A
volume
of
concentrated
sample
corresponding
to
100
liter
of
groundwater
was
also
examined
for
human
caliciviruses
(
noroviruses)
by
direct
RT­
PCR
at
UNC
RT­
PCR.
Approximately
5
mL
volumes
of
all
cell
culture
lysates
were
concentrated
to
300
µ
L
using
PEG
8000.
Of
this,
140
:
L
was
used
for
RNA
extraction
using
Qiagen
RNA
extraction
kit.

The
remaining
160
:
L
was
archived.
The
primers
used
for
amplification
of
enterovirus
nucleic
acid
are
shown
below
(
Schwab
et
al.,
1996).

3'
Primer:
5'
ACC
GGA
TGG
CCA
ATC
CAA
3'

5'
Primer:
5'
CCT
CCG
GCC
CCT
GAA
TG
3'

RT­
PCR
conditions
were
according
to
those
previously
used
and
were:
RT
­
420C
for
60
min,

followed
by
inactivation
of
RT
at
950C
for
15
min.
Denaturation
­
950C
for
90
sec;
Annealing
 

550C
for
1.5
min;
extension
­
720C
for
1.5
min;
final
extension
­
720C
for
10
min.
No.
of
cycles­
40
(
3).
The
RT­
PCR
products
were
analyzed
by
agarose
gel
electrophoresis
and
confirmed
by
ethidium
bromide
staining
for
observation
of
DNA
amplicons
of
the
correct
size.
For
positive
amplification,
an
amplicon
of
197
bp
was
expected.

Results
Summary.
Of
the
27
wells
tested,
fecal
coliforms
were
detected
in
7
(
26%),
E.
coli
in
3
(
11%)
and
Enterococcus
in
6
(
8
positive
samples)
(
22%).
Three
of
27
wells
contained
fecal
140
coliforms,
E.
coli
and
Enterococci
while
one
well
was
positive
for
both
fecal
coliforms
and
E.
coli.

Somatic
coliphages
were
detected
in
16
wells
(
59%)
male­
specific
FRNA
phages
in
11
wells
(
41%)
(
12
positive
samples),
and
"
total"
coliphages
in
12
samples.
None
of
the
samples
showed
cytopathological
effects
(
CPE)
characteristic
of
enteric
viruses
during
their
passages
in
BGM
and
Caco­
2
cell
lines.
None
of
the
samples
was
positive
for
enteric
viruses
by
RT­
PCR
or
PCR.

References
US
EPA
(
1996)
ICR
Microbial
Laboratory
Manual.
Office
of
Research
and
Development,
EPA
Number:
600R95178.
Pages:
233,
Washington,
DC
Environment
Protection
Agency
(
2001a).
Method
1601:
Male­
specific
(
F+)
and
somatic
coliphages
in
water
by
two­
step
enrichment
procedure.
United
States,
Environment
Protection
Agency,
Office
of
Water,
Washington,
D.
C.
2001.
http://
epa.
gov/
nerlcwww/
1601ap01.
pdf
Environment
Protection
Agency
(
2001b).
Method
1602:
Male­
specific
(
F+)
and
somatic
coliphages
in
water
by
single
agar
layer
procedure.
United
States,
Environment
Protection
Agency,

Office
of
Water,
Washington,
D.
C.
2001.
http://
epa.
gov/
nerlcwww/
1602ap01.
pdf
American
Public
Health
Association.
Standard
Methods
for
Examination
of
Water
and
Wastewater,
20th
Edition,
Washington,
DC.
1998.

Schwab,
K.
J,
De
Leon
R.,
and
M.
D.
Sobsey
(
1996)
Immunoaffinity
concentration
and
purification
141
of
waterborne
enteric
viruses
for
detection
by
reverse
transcriptase
PCR.
Appl
Environ
Microbiol.,

62(
6):
2086­
94.
142
Table
1.
Details
of
Groundwater
Wells
Selected/
Screened
During
the
Study
Date
on
Wells
for
EPA
groundwater
study
on
coliphage
methods
Sample
Well
Well
Name
Date
of
sampling
Well
Address
Contact
person
Contact
Phone
1
01
Amundson
Amundson
Farms
4/
11/
2002
Amundson
Farms,
RR1
Box
25,
Chattfield,
MN
55923
Brad
507­
867­
3396
2
02
Gervais
Lake
Gervais
4/
23/
2002
Lake
Gervais,
2500
Ederton
St.,
Maplewood,
MN
Richard
(
Dick)
Haus
651­
748­
2500
3
03
Round
Round
Lake
Park
4/
23/
2002
Round
Lake
Park,
910
Frost,
St.
Paul,
MN
Richard
(
Dick)
Haus
651­
748­
2500
4
04
Turtle
Turtle
Lake
4/
23/
2002
Turtle
Lake,
4079
Hodgson
Rd.,
Shoreview,
MN
Dick
651­
748­
2500
5
05
Brookdale
Brookdale
Park
4/
30/
2002
Brookdale
Park,
7650
June
Ave.
North,
Brooklyn
Park,
MN55443
Layne
763­
493­
8350
6
06
Oak
Groove
Oak
Groove
Park
4/
30/
2002
Oak
Grove
Park,
6941
102nd
Avenue
N.,
Brooklyn
Park,
MN55443
763­
493­
8350
7
07
Keller
Golf
Keller
Golf
Course
5/
3/
2002
2166
Maplewood
Drive,
Maplewood,
MN
55109
651­
766­
4173
8
08
Keller
Golf
Keller
Main
Park
5/
3/
2002
Keller
Main
Park,
Hwy.
61,
Maplewood,
MN
55109
Dick
651­
748­
2500
9
09
Hamilton
Hamilton
Park
5/
7/
2002
6101
Candlewood
drive,
Brooklyn
Park,
MN
763­
493­
8350
10
10
Norwood
Norwood
Park
5/
7/
2002
8100
Newton
Ave.
N.,
Brooklyn
Park,
MN
763­
493­
8350
11
11
Presbyterian
Presbyterian
Church
Maple
Plain
5/
9/
2002
558
County
Rd.
110,
Maple
Plain,
MN
55359
763­
479­
2158
12
12
Immanuel
Immanuel
United
Methodist
Church
5/
9/
2002
10095
County
Rd.
101,
Cocoran,
MN
763­
420­
2585
143
13
13
Central
Central
Park
5/
13/
2002
8440
Regent
Ave.
Brooklyn
Park,
MN
763­
493­
8350
14
14
Historical
Brooklyn
Park
Historical
Farm
5/
13/
2002
4345
101st
Avenue
N.,
Brooklyn
Park
763­
493­
8350
15
15
Northwood
Northwood
Park
5/
13/
2002
107th
Quebeck
Ave.
N.,
Brooklyn
Park,
MN
763­
493­
8350
16
16
Lakewood
Lakewood
5/
13/
2002
3600
Hennepin
Ave.
Ron
Gjerde
612­
822­
2171
17
17
Lakewood
Mausoleum
Lakewood
Cemetery
5/
16/
2002
3600
Hennepin
Ave.
Ron
Gjerde
612­
822­
2171
18
18
Lakewood
Maintenance
Lakewood
Cemetery
5/
16/
2002
3600
Hennepin
Ave.
Ron
Gjerde
612­
822­
2171
19
19
Al
Alan
Ducommun's
Father
5/
20/
2002
5435
152nd
Ave.,
Anoka,
MN
55303
Alan
Ducommu
n
20
20
Dayne
Dayne
Ducommun
5/
20/
2002
4841
Salish
Circle,
Ramsey,
MN55303
Dayne
Ducommu
n
763/
753/
5090
21
21
Lake
Maria
Lake
Maria
5/
29/
2002
11411
Clementa,
Monticello,
NN
55362
Tom/
Mark763­
878­
2325
22
22
Gibbs
Farm
Gibbs
Farm
Museum
5/
29/
2002
Larpentar
 
Cleveland
Av.

23
23
Tom
Arendt
Tom
Arendt
Home
6/
1/
2002
9871
John
Trail,
Chisago
City,
MN55193
Tom
Arendt
651­
257­
2295
24
24
Milena
Milena's
house
6/
3/
2002
58585
222nd
Street,
Litchfield,
MN55355
Milena
320­
693­
6754
25
25
Jay
Jay
Keil
home
6/
3/
2002
18076
68th
Ave,
Darwin,
MN
55355
Milena
320­
693­
6754
26
12
Immanuel
Immanuel
United
Methodist
Church
6/
14/
2002
3600
Hennepin
Ave.
Ron
Gjerde
612­
822­
2171
27
24
Milena
Milena's
house
6/
17/
2002
58585
222nd
Street,
Litchfield,
MN55355
Milena
320­
693­
6754
144
Table
2.
Bacteriological,
Coliphage
and
Virological
Analysis
of
27
Well
Water
Samples
from
Minnesota
EPA
PROJECT
 
COMPLETE
RESULTS
OF
WATER
SAMPLE
TESTING
Bacteriological
analysis
COLIPHAGES
(
Method
1601,
1602)
Virus
Isolation
in
cells
RT­
PCR
on
Pooled
cell
lysate
for
Enterovirus
Well#
Fec.
Col.
E
.
coli
Enterococci
SOMATIC
F+
TOTA
L
Method
BGMK
Caco­
2
01
Amundson
Neg.
Neg.
Neg.
Pos.,
(
tntc)
Neg.
NT
1601
&
1602
2
passages,
no
cpe
2
passage,
no
cpe
Negative
02
Gervais
Neg.
Neg.
Neg.
Pos.,
4/
100
Neg.
Pos.,
5/
100
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
03
Round
Neg.
Neg.
Neg.
Pos.,
2/
100
Pos.,
4/
100
Pos.,
4/
100
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
04
Turtle
Neg.
Neg.
Neg.
Pos.,
4/
100
Pos.,
2/
100
Pos.,
3/
100
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
05
Brookdale
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
06
Oak
Groove
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Method
1601
&
1602
3
passages,
no
cpe
2
passages,
no
cpe
Negative
07
Keller
Main
Neg.
Neg.
Neg.
Pos.,
1/
100
Pos.,
58/
100
Pos.,
4/
100
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
08
Keller
Main
Neg.
Neg.
Pos.,
1/
100
Pos.,
12/
100
Pos.,
40/
100
Pos.,
7/
100
Method
1601
&
1602
3
passages,
no
cpe
3
passages,
no
cpe
Negative
09
Hamilton
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
10
Norwood
Neg.
Neg.
Neg.
Pos.,
28/
100
Pos.,
9/
100
Neg,
0/
100
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
11
Presbyteria
n
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
12
Immanuel
Pos.,
30/
10
0
Neg.
Pos.,
1/
100
Neg.
Neg.
Neg.
Method
1601
&
1602
3
passages,
no
cpe
2
passages,
no
cpe
Negative
13
Central
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
14
Historical
Pos.
1/
100
Neg.
Pos.
1/
100
Neg.
Neg.
Neg.
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
145
15
Northwood
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
16
Lakewood
1
Pos.
1/
100
Neg.
Pos.,
1/
100
Pos.,
574/
100
Pos.
,
234/
10
0
Neg.
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
17
Lakewood
Mausoleum
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Method
1601
&
1602
3
passages,
no
cpe
3
passages,
no
cpe
Negative
18
Lakewood
Mausoleum
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
19
Al
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Method
1601
&
1602
2
passages,
no
cpe
3
passages,
no
cpe
Negative
20
Dayne
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
21
Lake
Maria
Neg.
Neg.
Neg.
Pos.,
9/
100
Neg.
Pos.,
1/
100
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
22
Gibbs
Farms
Neg.
Neg.
Pos.,
1/
100
Pos.,
2/
100
Neg.
Pos.,
2/
100
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
23
Tom
Arendt
Pos.
1/
100
Neg.
Neg.
Pos.,
574/
100
Pos.,
3/
100
Pos.,
7/
100
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
24
Milena
Pos.
17/
10
0
Pos.,
12/
1
00
Pos.,
2/
100
Pos.,
574/
100
Pos.,
11/
100
Pos.,
1/
100
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
25
Jay
Pos.
3/
100
Pos.,
1/
10
0
Pos.,
15/
100
Pos.,
574/
100
Neg.
Pos.,
3/
100
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
12
Church
Repeat
Neg.
Neg.
Neg.
Pos.,
574/
100
Pos.,
3/
100
Pos.,
4/
100
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
16
Cemetery
Repeat
Neg.
Neg.
Neg.
Pos.,
574/
100
Pos.,
6/
100
Pos.,
6/
100
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
24
Milena
Repeat
Pos.
248/
1
00
Pos.,
3/
10
0
Pos.,
20/
100
Pos.,
574/
100
Pos.,
2/
100
Neg.
Method
1601
&
1602
2
passages,
no
cpe
2
passages,
no
cpe
Negative
146
Positive
UNH
Bacterial
Results
for
Groundwater
Samples
Well
Type
and
Number
Fecal
Coliforms
Enterococcus
100
ML
1
L
100
ML
1
L
Community,
Well
#
1
1
200
0
0
Private;
shallow,
Well
#
4
0
1
0
0
Private
Well
#
9
0
0
0
3
Private,
Well
#
15
35
TNTC
0
0
Private,
Well
#
18
0
0
1
93
Private,
Well
#
19
2
10
13
89
Community,
#
23
0
0
2
87
Community,
#
23
5
69
2
32
Table
3.
Results
of
UNH
Groundwater
Samples
Analyzed
for
Adenovirus
and
Astrovirus
Well
#
Results
1
Negative
for
all
viral
analyses
2
Negative
for
all
viral
analyses
3
Negative
for
all
viral
analyses
4
Negative
for
all
viral
analyses
5
Negative
for
all
viral
analyses
6
Negative
for
all
viral
analyses
7
Negative
for
all
viral
analyses
8
Negative
for
all
viral
analyses
9
Negative
for
all
viral
analyses
10
Negative
for
all
viral
analyses
11
Negative
for
all
viral
analyses
12
Negative
for
all
viral
analyses
13
Negative
for
all
viral
analyses
14
Negative
for
all
viral
analyses
15
Negative
for
all
viral
analyses
16
Negative
for
all
viral
analyses
17
Negative
for
all
viral
analyses
18
Negative
for
all
viral
analyses
19
Negative
for
all
viral
analyses
20
Negative
for
all
viral
analyses
21
Negative
for
all
viral
analyses
22
Negative
for
all
viral
analyses
23
Negative
for
all
viral
analyses
24
Negative
for
all
viral
analyses
25
Negative
for
all
viral
analyses
147
Table
4.
Summary
of
UNH
Samples
Positive
for
Coliphages,
Bacterial
Indicators,
Adenoviruses
and/
or
Astroviruses
Well
Number
and
Type
Samples
positive
for:

Coliphage
Bacterial
Indicators
Phage
and
Bacterial
Indicators
Coliphage
and
Virus
Bacterial
Indicators
and
Virus
FC
Ent.
Any
Well
#
3,
Private
1
0
0
0
0
Well
#
1,
Community
0
1
0
1
0
0
0
Well
#
9,
Private
0
0
1
1
0
0
0
Well
#
15,
Community
0
1
0
1
0
0
0
Well
#
18,
Community
0
0
1
1
0
0
0
Well
#
19,
Community
0
1
1
1
0
0
0
Well
#
21,
Private
0
0
1
1
0
0
0
Well
#
22,
Private
0
1
1
1
0
0
0
Well
#
23,
0
1
1
1
0
0
0
All
other
wells
0
0
0
0
0
0
0
References
Chapron,
C.
D.,
Ballester,
N.
A.,
Fontaine,
J.
H.,
Frades,
C.
N.
&
Margolin,
A.
B.
2000
The
Detection
of
Astrovirus,
Enterovirus
and
Adenovirus
Type
40
and
41
in
Surface
Waters
Collected
and
Evaluated
by
the
Information
Collection
Rule
and
Integrated
Cell
Culture/
Nested
PCR
Procedure.
Appl.
and
Environ.
Microbiol.,
60
(
6),
2520­
2525.

Environmental
Protection
Agency.
1995
Virus
Monitoring
Protocol
for
the
Information
Collection
Requirements
Rule.
U.
S.
Environmental
Protection
Agency,
publication
EPA/
814­
B­
95­
002.

Government
Printing
Office,
Cincinnati,
Ohio.
148
Environmental
Protection
Agency.
2000
Method
1601:
Male­
specific
(
F+)
and
Somatic
Coliphage
in
Water
by
Two­
step
Enrichment
Procedure.
Draft
April
2000.
Office
of
Water,
Washington,

D.
C.

Environmental
Protection
Agency.
2001
Method
1602:
Male­
specific
(
F+)
and
Somatic
Coliphage
in
Water
by
Single
Agar
Layer
(
SAL)
Procedure.
Draft
January
2001.
Office
of
Water,

Washington,
D.
C.

Xu,
W.,
McDonough,
M.
C.,
&
Erdman
D.
D.
2000.
Species­
specific
identification
of
human
adenoviruses
by
a
multiplex
PCR
assay.
J.
Clin.
Microbiol.
39,
4114­
4120.
