Environmental and Molecular Mutagenesis 29:418 – 430 (1997)
Genotoxicity Monitoring of Small Bodies of Water
Using Two Species of Tadpoles and the Alkaline
Single Cell Gel (Comet) Assay
Steven Ralph and Michael Petras*
Department of Biological Sciences, University of Windsor, Windsor, Ontario, Canada
To monitor genotoxicity in small bodies of water
(e.g., creeks, ponds, and drainage ditches) we examined tadpole erythrocytes of two species: Rana
clamitans and Rana pipiens, using the alkaline single cell gel DNA electrophoresis (SCG) or ‘‘comet’’
assay. This approach involves detection, under alkaline conditions, of cell DNA fragments which on
electrophoresis migrate from the nuclear core, resulting in a ‘‘comet with tail’’ formation. Fifty-six
samples, a total of 606 tadpoles, from 18 sites in
southern Ontario, collected between 1993 and
1995, were examined. Samples of R. clamitans tadpoles collected in 1994 and 1995, from regions
with heavy agricultural activity, gave significantly
higher (P õ 0.001) DNA length to width ratios than
samples of R. clamitans tadpoles collected from sites
in the Bruce Peninsula and near the French River,
which have little or no agriculture. Samples of R.
pipiens tadpoles collected in 1994 from sites on the
outskirts of Windsor, Ontario, sites which receive
genotoxic inputs from nearby industries, gave significantly higher (P õ 0.001) DNA ratios than sam-
ples from agricultural areas and the Bruce Peninsula. R. clamitans tadpoles showed significant annual variation in DNA damage which was greater
in samples of tadpoles collected from agricultural
areas than from the Bruce Peninsula. The higher
levels of DNA damage in tadpoles collected from
agricultural areas may be due to the pesticides
used, and the increased variation in DNA damage
in the same areas is likely due to the impact of crop
rotation, including leaving fields fallow, the timing
of rainfall, and/or the application of pesticides. R.
clamitans tadpoles, especially those collected from
agricultural areas, also showed significant seasonal
variation in DNA damage. There was no significant
(P ú 0.05) seasonal or annual variation in the levels
of DNA damage in R. pipiens tadpoles collected
from the Tallgrass Prairie. This study indicates that
both species are suitable for use in the alkaline SCG
assay and as in situ sentinel organisms for environmental biomonitoring. Environ. Mol. Mutagen.
29:418 –430, 1997. q 1997 Wiley-Liss, Inc.
Key words: DNA damage; aquatic monitoring; alkaline comet assay; Rana clamitans; Rana
pipiens; tadpoles; pesticides; amphibians
INTRODUCTION
The continued deposition of genotoxic agents into the
environment has resulted in a need for sensitive assays
to monitor their accumulation and impact. Small bodies
of water (e.g., creeks, ponds and drainage ditches) are
often the first to receive run-off containing pollutants from
a number of sources including industrial effluents, sewage
contaminants, accidental spills, internal combustion engine emissions, landfills, and pesticide uses.
To address this problem, we have chosen to use the
alkaline single cell gel DNA electrophoresis (SCG) or
‘‘comet’’ assay to quantify genotoxicity of small bodies
of water in southwestern Ontario. This technique was
originally developed by Rydberg and Johanson [1978]
who used isolated cells in a microgel to determine DNA
damage. Modifications by Ostling and Johanson [1984],
involving electrophoresis under neutral conditions, permitted the detection of double-stranded DNA breaks. Subsequently, detection of single-stranded breaks and alkalilabile damage was made possible by Singh et al. [1988],
who performed electrophoresis under alkaline conditions
[see McKelvey-Martin et al., 1993; Fairbairn et al., 1995
for comprehensive reviews of this assay].
We chose to examine contamination in small bodies of
water, rather than larger bodies of water, for several reasons. As mentioned above, small bodies of water are often
the first to be impacted by contaminants, and at least some
ponds, and to a lesser degree the shallow ditches, tend to
be closed systems. Therefore, an accumulation of contaminants, primarily in the sediment, may result in multi-year
effects. In addition, there is the possibility of tracking
seasonal and annual fluctuations resulting from contaminants in the water. Finally, in these bodies of water, it may
be possible to localize the source of the contaminants.
To evaluate the significance of any changes in the environment and the effectiveness of remedial steps, we fo-
*Correspondence to: Dr. Michael Petras, Department of Biological Sciences, University of Windsor, Windsor, Ontario, N9B 3P4, Canada.
Received 2 July 1996; revised and accepted 18 February 1997.
q 1997 Wiley-Liss, Inc.
504
/ 8I17$$0504
06-09-97 10:51:23
wlema
W Liss: EM
8I17
Genotoxicity Monitoring Using Tadpoles
cused on genotoxicity as a measure of contamination because endpoints, in some genotoxicity monitoring assays,
are specific and the effects can be readily quantified.
Therefore, the level of genotoxicity can be used as an
indicator of general contaminant levels which affect various biological processes (e.g., physiology, development,
reproduction) in organisms.
Tadpoles were used for reasons which included: Availability in the study area, ease of collection, localization
to particular bodies of water, and the direct exposure to
contaminants in the water, sediment, and vegetation. Tadpoles used included those of the frog species, Rana clamitans (green frog) and Rana pipiens (leopard frog). R.
clamitans tadpoles range in length from 55 to 80 mm at
transformation, and appear in early spring and overwinter
before transforming [Walker, 1946]. These tadpoles generally inhabit permanent ponds. It is possible to distinguish between first year and second year tadpoles of this
species by their size and thus, the duration of exposure
to genotoxic agents can be estimated. R. pipiens tadpoles
appear about the first of May and transform about the
middle of July [Walker, 1946]. These tadpoles inhabit
both temporary and permanent ponds, and are about the
same size at transformation (65–83 mm [Walker, 1946]).
A number of laboratory studies have demonstrated that
amphibians are sensitive organisms, suitable for the detection of genotoxic agents [Hart and Armstrong, 1984; Siboulet et al., 1984; Jaylet et al., 1986, 1987; Zoll et al.,
1988; Krauter, 1992; LeCurieux et al., 1992; Djomo et
al., 1995; Ralph et al., 1996]. For many small bodies of
water, especially those in agricultural areas, a significant
contributing source of genotoxicity is pesticide usage.
Toxic effects (developmental abnormalities, altered
growth rates, and modifications in behavior and feeding
habits) have been observed in amphibians exposed to organochlorine pesticides such as DDT (dichlorodiphenyltrichloroethane) and its metabolites [Cooke, 1979; Osborn
et al., 1981], dieldrin [Kaplan and Overpeck, 1964;
Brooks, 1981], endrin, toxaphene [Kaplan and Overpeck,
1964; Hall and Swineford, 1980], aldrin, chlordane
[Kaplan and Overpeck, 1964], and lindane [Marchal-Segault and Ramande, 1981] that were used in the past and
still persist in the environment. The potential of tadpoles
to accumulate pesticides such as DDT and dieldrin has
been demonstrated in adult Pseudacris crucifer [Russell
et al., 1995]. Toxic effects in amphibians have also been
observed with some of the currently used pesticides such
as triclopyr [Berrill et al., 1994], permethrin, fenvalerate
[Berrill et al., 1993], paraquat [Bauer Dial and Dial,
1995], esfenvalerate [Materna et al., 1995], fenitrothion
[Elliott-Freely and Armstrong, 1982; Berrill et al., 1994],
and carbaryl [Elliott-Freely and Armstrong, 1982]. Also,
in a genotoxicity study by Rudek and Rozek [1992], a
concentration-dependent increase in the frequency of micronuclei in red blood cells was observed in Rana tem-
504
/ 8I17$$0504
06-09-97 10:51:23
419
poraria and Xenopus laevis tadpoles after a 14-day exposure to relatively low levels of the pesticide Fastac 10EC.
Overall, however, very little information is available concerning the genotoxic impact of pesticides on amphibians.
To determine the genotoxic impact of pesticides and
other contaminants on organisms inhabiting small bodies
of water, we have utilized a modified alkaline SCG test
[Ralph et al., 1996] for use in two species of tadpoles
collected from 18 sites in southern Ontario during the
summers of 1993–1995. Information is reported on geographic distributions of DNA damage, and annual and
seasonal variations in DNA damage.
MATERIALS AND METHODS
Chemical Reagents
The low melting agarose (electrophoresis purity quality) was obtained
from BioRad (Mississauga, Ontario, Canada). PBS (phosphate buffered
saline, calcium- and magnesium-free) was purchased from Gibco-BRL
(Grand Island, NY). Hank’s balanced salt solution, EDTA (disodium
ethylenediamine-tetra-acetate), Tris (tris(hydroxy-methyl)aminomethane
hydrochloride), DMSO (dimethylsulfoxide), EtBr (ethidium bromide),
Triton X-100, and Na sarcosinate came from Sigma (St. Louis, MO).
Fully frosted microscope slides and number one cover glasses were
supplied by Fisher (Toronto, Ontario, Canada).
Collection Sites
See Figure 1 for the location of the various sampling sites in southwestern Ontario. Collection sites were typically permanent ditches or
ponds, though some were only temporary. Sites were categorized as
industrial (site 1), agricultural (sites 2–10), or non-agricultural (sites
11–18) depending on land uses and possible sources of contaminants
(see Discussion).
Treatment of Freshly Caught Samples
Tadpoles were transported to the laboratory in water from the collection sites and housed in polypropylene containers (29 cm 1 19 cm 1
13 cm) containing this water until they were bled, which occurred within
9 days of their capture.
Alkaline SCG Assay
The procedure used is basically that described by Singh et al. [1988].
Modifications, due to the uniqueness of the biological material studied
and to the equipment available, were relatively minor. The procedure
is described fully in Ralph et al. [1996].
Blood samples were collected from the R. clamitans and R. pipiens
tadpoles by decapitation followed by immediately placing the animals
in a 10% solution of Hank’s balanced salt solution for 2 min. All animals
were treated individually. The vast majority of cells collected were
erythrocytes since they are much more numerous than any other cell
type in the circulatory system. Very few free cells from any other tissues
were present. Erythrocytes were chosen because they are nucleated in
amphibians. Serial dilutions were made so that three or four cells would
be seen without crowding in a single field at 4001 magnification. The
appropriate erythrocyte suspension was then mixed with 0.5% low melting agarose, and this suspension was pipetted onto fully frosted slides
and covered with coverslips. The slides were stored at 37C for 20 min
wlema
W Liss: EM
8I17
420
Ralph and Petras
Fig. 1. Map of tadpole collection sites in southwestern Ontario. Collection sites: (1) Tallgrass Prairie Heritage Park, (2) Gesto, (3) South
Woodslee, (4) Point Pelee National Park, (5) Highgate, (6) Duart,
(7) New Glasgow, (8) Churchville, (9) Thedford, (10) Walkerton,
(11) Southampton, (12) North Sauble Beach, (13) Boat Lake, (14) Spry
Lake, (15) Ferndale, (16) Stokes Bay, (17) Clarke’s Corner, (18) French
River. Counties and districts listed in italics, with value in brackets
representing the quantities of active ingredients of all pesticides applied
(kg per hectare of total land) per county or district in 1993 [modified
from Hunter and McGee, 1994].
to allow complete polymerization of the agarose. A single layer of
agarose was used as described by Pandrangi et al. [1995].
After the agarose polymerized, the coverslips were removed and the
slides were lowered into freshly made lysing solution (2.5 M NaCl, 100
mM Na2EDTA, 10 mM Tris, 10% DMSO, 1% Na sarcosinate, 1%
Triton X-100) at pH 10.0 and incubated at room temperature in the dark
for 2 hr. After lysing, the slides were drained and placed in the alkaline
electrophoresis buffer (1 mM EDTA, 300 mM NaOH) for 15 min. For
electrophoresis, the power supply was set at 25 V and the current adjusted to 265–270 mA by slowly changing the buffer level in the tray.
Slides were routinely exposed to this current in the dark at 37C for 20
min. After electrophoresis, the slides were placed in a staining tray and
covered with a neutralizing buffer (0.4 M Tris pH 7.5) in the dark for
5 min. This last step was repeated. The slides were then drained and
overlaid with ethidium bromide and covered with coverslips. The slides
were examined the next day at 4001 using a fluorescent microscope.
All slides were coded and examined blindly. Routinely, 25 cells were
examined per animal, unless otherwise noted. The length and width of
the DNA mass were measured using an ocular micrometer disk. The
length:width ratios were used in all comparisons. Under these conditions, a DNA pattern with a ratio of one has a DNA length of approximately 40 mm and with a ratio of three, a DNA length of approximately
120 mm. Cell viability was found to be greater than 95% using the
trypan blue exclusion technique.
504
/ 8I17$$0504
06-09-97 10:51:23
wlema
W Liss: EM
8I17
Genotoxicity Monitoring Using Tadpoles
Statistical Analyses
An alpha level of 0.05 was used to determine significance in all
statistical analyses. Different groups were compared using either a Student’s one-tailed t-test or a nested analysis of variance (Systat 5.0). The
extent of intercellular heterogeneity within each of the data sets was
determined from range of the DNA length:width ratios to the standard
deviation of these ratios. Values below 2 or above 6 indicated the
data to be extremely homogeneous or extremely heterogeneous [see
Vijayalaxmi et al., 1992]. An estimate of the normality of the distribution
of the DNA length:width ratios of tadpoles was determined using the
method of ranked normal deviates [see Sokal and Rohlf, 1995].
RESULTS
Fifty-six samples, totalling 606 freshly caught tadpoles,
collected from 18 sites in southern Ontario between 1993
and 1995 were examined using the alkaline SCG assay.
The DNA length:width ratios of 43 samples of R. clamitans tadpoles are given in Table I. Normally, 12 tadpoles
were scored from any one site. Less than 12 animals were
scored in those cases where either fewer tadpoles were
caught or the odd slide was not scorable because of technical difficulties. The ratios ranged from a high of 2.784 in
one sample collected from Highgate (8/27/93) to a low
of 1.244 in the sample collected near Clarke’s Corner (7/
18/95). The ratio of the range and standard deviation for
all samples suggests that the intercellular distribution of
the DNA length:width ratios is neither extremely homogeneous nor extremely heterogeneous (Table I). The DNA
length:width ratios among pooled samples of R. clamitans
tadpoles collected in 1994 from Churchville (n Å 40,
Table I) were shown to be normally distributed using the
method of ranked normal deviates.
Samples of R. clamitans tadpoles collected during the
summers of 1994 and 1995 were grouped by region to
examine the geographic pattern of DNA damage. The
DNA length:width ratios of samples of R. clamitans tadpoles collected in 1995 from agricultural regions were
significantly higher (P õ 0.001) than the ratios of samples
from non-agricultural regions (Table II). There was significant (P õ 0.001) variation among samples from each
region, but this was less than the variance between regions. Since only one sample was collected that year from
an industrial region, no comparisons were made with samples from the other two regions. Analysis of R. clamitans
tadpoles collected in 1994 gave similar results. Typical
cell profiles (Fig. 2) of freshly caught tadpoles (1995)
from non-agricultural regions showed less DNA damage
than those of samples collected from agricultural and industrial regions. The cell profile of R. clamitans tadpoles
that were collected from Highgate (8/27/93) and then
maintained in the laboratory, in dechlorinated water for
4 months, is given for comparison.
Analysis of 3 years of R. clamitans data (Table II),
based on collections in agricultural regions of southern
Ontario, indicated that the variance among years was
504
/ 8I17$$0504
06-09-97 10:51:23
421
greater than the variance among samples within years.
Both were, however, significant (P õ 0.001). The amount
of DNA damage in R. clamitans tadpole samples collected
from non-agricultural regions of southern Ontario was
significantly different (P õ 0.01) between 1994 and 1995,
and the variance among samples within these years was
about the same as the variance between years (Table II).
Comparisons of DNA damage levels in samples of R.
clamitans tadpoles collected during 1994 and 1995, from
agricultural and non-agricultural regions, indicated that
there was significant variation between years, between
regions within years, and among samples within regions
within years (P õ 0.001 in all cases, Table II). Variation
in the levels of DNA damage was greatest between regions during each year, followed by the variation between
years. Finally, in all cases, the variance among tadpoles
in any single sample was small.
To determine if the year to year variation in DNA
damage levels showed a specific pattern, the DNA
length:width ratios of samples of R. clamitans tadpoles
collected at approximately the same time each year (over
a 3-year period) were compared. At two sites (Highgate
and Thedford), there was a significant reduction in DNA
damage over the 3 years (P õ 0.001, Table I, Fig. 3).
Reductions in DNA damage levels were also seen at three
other sites (Duart, Southampton, and Spry Lake, Table I)
which had been studied over a 2-year period.
The DNA length:width ratios of 13 samples of R. pipiens tadpoles are given in Table III. The ratios ranged
from a high of 2.956, in one sample collected from the
Tallgrass Prairie ditch (5/31/94), to a low of 1.653 in the
sample collected from Boat Lake (7/19/94). The ratio of
the range and standard deviation for all samples suggests
that the intercellular distribution of the DNA length:width
ratios is neither extremely homogeneous nor extremely
heterogeneous (Table III). The DNA length:width ratios
among pooled samples of R. pipiens tadpoles collected
in 1994 from the Tallgrass Prairie ditch (n Å 33, Table
III) were shown to be normally distributed using the
method of ranked normal deviates.
Samples of R. pipiens tadpoles collected during the
summer of 1994 were grouped by region to examine
the geographic pattern of DNA damage. The DNA
length:width ratios of samples of R. pipiens tadpoles collected from the three regions were significantly different
(P õ 0.001) from one another (Table IV). The variance
among samples within these regions was significant (P õ
0.001) but, less than the variance among regions. Samples
from agricultural and non-agricultural regions had significantly lower DNA ratios (P õ 0.001 in both cases)
than samples from the industrial region. In these comparisons, the variance among samples within each region was
significant (P õ 0.01 in both cases) but, less than the
variance between regions. There was a significant difference (P õ 0.05) between the DNA ratios of samples taken
wlema
W Liss: EM
8I17
422
Ralph and Petras
TABLE I. Detection of DNA Damage in Erythrocytes of Freshly Caught R. clamitans Tadpoles (1993–1995) From Sites
in Southern Ontario
Year and location
1993 Gesto
Highgate
Thedford
1994 Highgate
Duart
Churchville
Churchville
Churchville
Churchville
Thedford
Thedford
Southampton
North Sauble Beach
Spry Lake
Spry Lake
French River
1995 Tallgrass Prairie pond #1
South Woodslee
Point Pelee pond #1
Point Pelee pond #2
Point Pelee pond #2
Highgate
Highgate
Highgate
Highgate
Duart
Duart
Duart
New Glasgow pond #1
New Glasgow pond #1
New Glasgow pond #1
Thedford
Walkerton
Southampton
Southampton
Boat Lake
Boat Lake
Spry Lake
Spry Lake
Ferndale
Stokes Bay
Stokes Bay
Clarke’s Corner
Regiona
Date caught
Sample size
A
A
A
A
A
A
A
A
A
A
A
N-A
N-A
N-A
N-A
N-A
I
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
N-A
N-A
N-A
N-A
N-A
N-A
N-A
N-A
N-A
N-A
8/24/93
8/27/93
8/17/93
9/20/94
6/1/94
5/2/94
5/10/94
5/17/94
6/1/94
6/22/94
7/19/94
6/22/94
7/19/94
6/22/94
7/19/94
9/25/94
6/20/95
8/23/95
6/15/95
6/15/95
8/23/95
5/3/95
5/31/95
6/21/95
8/8/95
5/31/95
6/21/95
8/8/95
5/31/95
6/21/95
8/8/95
7/18/95
7/18/95
5/9/95
7/18/95
5/9/95
7/18/95
5/9/95
7/18/95
5/9/95
5/9/95
7/18/95
7/18/95
12
10
6
12
11
10
10
8
12
12
11
12
9
12
11
12
11
11
12
11
11
11
12
12
11
12
12
12
12
12
12
12
12
12
12
12
10
12
12
6
11
12
12
DNA length:width
ratio { SEMb
Range:SD
ratio
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
3.42
3.41
2.73
3.21
2.95
2.76
3.28
3.22
3.00
2.78
3.55
3.87
2.68
3.08
3.20
3.16
3.04
2.90
3.63
2.78
3.38
3.35
4.20
2.95
3.28
3.20
3.28
2.73
3.30
2.76
2.90
3.14
2.99
2.90
2.74
3.42
3.26
2.85
2.55
2.56
3.65
3.57
3.40
2.150
2.784
2.481
2.114
2.525
2.082
2.145
2.187
2.223
2.034
2.738
2.177
1.582
1.724
2.113
1.562
2.175
1.679
2.038
1.857
2.621
2.478
2.112
2.026
1.911
1.989
1.703
2.182
1.549
1.470
2.117
2.087
1.965
1.623
1.538
1.921
1.494
1.787
1.963
1.648
2.030
1.439
1.244
0.064
0.112
0.196
0.069
0.108
0.082
0.060
0.105
0.055
0.089
0.078
0.109
0.112
0.073
0.101
0.097
0.126
0.102
0.124
0.114
0.100
0.044
0.094
0.160
0.092
0.074
0.106
0.161
0.101
0.089
0.127
0.108
0.123
0.078
0.089
0.081
0.079
0.092
0.136
0.116
0.065
0.085
0.061
a
I, industrial; A, agricultural; N-A, non-agricultural.
Ratios based on 25 cells/tadpole, except for samples collected in 1993 which were based on 50 cells/tadpole.
b
from agricultural and non-agricultural regions (Table IV).
The variance between regions was less than the variance
between samples from each region perhaps because the
number of samples from each region was small (2). In
all cases the variance among tadpoles per sample was
small. Figure 4 summarizes the distribution of DNA damage in individual cells of R. pipiens tadpoles collected at
typical sites during the summer of 1994.
Although the collecting procedure did not vary from
year to year, the number of R. pipiens tadpoles collected
504
/ 8I17$$0504
06-09-97 10:51:23
in 1995 was considerably lower than in 1994. Most of
the sites did not have any R. pipiens tadpoles in 1995.
As a result, only samples of freshly caught R. pipiens
tadpoles collected at the same time during 1994 and 1995
(6/20) from the Tallgrass Prairie ditch could be compared
for DNA damage. The levels of DNA damage did not
differ significantly (P ú 0.05, Table III).
Seasonal variation of DNA damage in freshly caught
R. pipiens and R. clamitans tadpoles is summarized in
Table V. Samples of R. pipiens tadpoles collected from
wlema
W Liss: EM
8I17
Genotoxicity Monitoring Using Tadpoles
423
TABLE II. Comparison of DNA Damage in Erythrocytes of R. clamitans Tadpoles
(1993–1995) Collected From Different Regions of Southern Ontario*
Source of variation
1. Agricultural vs. non-agricultural—1994
Between regions
Among samples within regions
Among tadpoles within samples
2. Agricultural vs. non-agricultural—1995
Between regions
Among samples within regions
Among tadpoles within samples
3. 1993 vs. 1994 vs. 1995—Agricultural
Among years
Among samples within years
Among tadpoles within samples
4. 1994 vs. 1995—Non-agricultural
Between years
Among samples within years
Among tadpoles within samples
5. 1994 vs. 1995—Agricultural—non-agricultural
Between years
Between regions within years
Among samples within regions within years
Among tadpoles within samples
DFa
MSb
Fsc
P
1
11
129
6.223
0.765
0.087
71.656
8.807
õ0.001
õ0.001
1
24
272
6.855
0.941
0.127
53.771
7.384
õ0.001
õ0.001
2
25
287
5.140
0.915
0.124
41.346
7.357
õ0.001
õ0.001
1
13
152
0.963
0.817
0.096
10.003
8.483
õ0.01
õ0.001
1
2
35
401
4.167
6.436
0.889
0.114
36.524
56.408
7.796
õ0.001
õ0.001
õ0.001
* Based on nested analysis of variance.
a
Degrees of freedom.
b
Mean square.
c
Sample variance ratio.
Fig. 2. Distribution of DNA damage (based on length:width ratios of
DNA patterns) observed at the cellular level in R. clamitans tadpoles
collected in 1995 at five select sites in southern Ontario. The distribution
of DNA damage observed at the cellular level in R. clamitans tadpoles
collected at Highgate (8/27/93) and maintained in the laboratory in
dechlorinated water for 4 months is given for comparison and should
reflect a baseline ratio. I, industrial; A, agricultural; N-A, non-agricultural.
the industrial region of the study area did not demonstrate
any significant (P ú 0.05) seasonal variation. Typically
there was a seasonal increase in DNA damage among
tadpoles collected from sites in agricultural regions (Table
V, only two of six sites shown), but one exception, Highgate (1995, Table I) did indicate a decrease in DNA damage over one season. Among samples from non-agricultural regions, there was typically either no change or a
504
/ 8I17$$0504
06-09-97 10:51:23
wlema
W Liss: EM
8I17
424
Ralph and Petras
Fig. 3.
Distribution of DNA damage (based on length:width ratios of DNA patterns) observed at the
cellular level in R. clamitans tadpoles collected at Highgate (8/27/93, 9/20/94, 8/8/95) and Thedford (8/17/
93, 7/19/94, 7/18/95) at approximately the same time each year between 1993 and 1995.
TABLE III. Detection of DNA Damage in Erythrocytes of Freshly Caught R. pipiens Tadpoles (1994–1995) From Eight
Sites in Southern Ontario
Year and location
1994 Tallgrass Prairie
Tallgrass Prairie
Tallgrass Prairie
Tallgrass Prairie
Tallgrass Prairie
Tallgrass Prairie
Tallgrass Prairie
Duart
Walkerton
Boat Lake
Spry Lake
1995 Tallgrass Prairie
Tallgrass Prairie
pond #1
pond #1
pond #2
ditch
ditch
ditch
ditch
pond #5
ditch
Regiona
Date caught
Sample size
I
I
I
I
I
I
I
A
A
N-A
N-A
I
I
5/31/94
6/20/94
5/31/94
5/31/94
6/20/94
7/8/94
8/8/94
6/1/94
6/22/94
7/19/94
6/22/94
6/20/95
6/20/95
11
10
10
4
8
12
9
10
11
10
9
11
12
DNA length:width
ratio { SEMb
Range:SD
ratio
{
{
{
{
{
{
{
{
{
{
{
{
{
3.66
2.64
2.57
2.35
2.75
3.51
2.95
3.12
3.48
3.41
2.75
2.96
3.07
2.487
2.405
2.769
2.956
2.788
2.782
2.650
2.305
1.857
1.653
2.157
2.940
2.941
0.050
0.135
0.072
0.170
0.154
0.104
0.129
0.087
0.045
0.046
0.087
0.118
0.036
a
I, industrial; A, agricultural; N-A, non-agricultural.
Ratios based on 25 cells/tadpole.
b
reduction in DNA damage (Table V, only two of five
sites shown). However, samples from Spry Lake in 1994
(Table I) indicated a seasonal increase in DNA damage
among tadpoles, but this may have been due to a localized
disturbance (see Discussion).
DISCUSSION
Of the numerous assays that have been developed for
monitoring genotoxicity, three have been especially useful for in vivo, in situ studies, chromosomal aberrations,
micronuclei induction, and sister chromatid exchanges.
All three have their limitations because they require mitotically active cells with relatively large chromosomes [see
review by Landolt and Kocan, 1983]. Other tests that
504
/ 8I17$$0504
06-09-97 10:51:23
have been frequently used under laboratory conditions
involve reversions in Salmonella (Ames test), DNA adducts, and unscheduled DNA synthesis. This last group
of tests are limited because they either do not provide
information on individual cells or cannot be used readily
in situ. Tice [1995] points out that any method used to
detect genotoxic damage in sentinel organisms should
detect many classes of DNA damage in a variety of tissues, provide information at the level of individual cells,
be sensitive to a broad array of mutagens, be rapid, and
be relatively inexpensive.
The alkaline SCG assay addresses some of these problems and is ideal for use in sentinel organisms. The technique permits the visualization of DNA damage in individual cells [Olive et al., 1990], relatively few cells are
wlema
W Liss: EM
8I17
Genotoxicity Monitoring Using Tadpoles
425
TABLE IV. Comparison of DNA Damage in Erythrocytes of R. pipiens Tadpoles (1994)
Collected From Different Regions of Southern Ontario*
Source of variation
1. Industrial vs. agricultural vs. non-agricultural
Among regions
Among samples within regions
Among tadpoles within samples
2. Industrial vs. agricultural
Between regions
Among samples within regions
Among tadpoles within samples
3. Industrial vs. non-agricultural
Between regions
Among samples within regions
Among tadpoles within samples
4. Agricultural vs. non-agricultural
Between regions
Between samples within regions
Among tadpoles within samples
DFa
MSb
Fsc
P
2
8
93
4.904
0.681
0.130
37.679
5.229
õ0.001
õ0.001
1
7
76
3.529
0.626
0.151
23.430
4.156
õ0.001
õ0.01
1
7
74
8.45
0.403
0.099
85.191
4.062
õ0.001
õ0.01
1
2
36
0.842
7.843
0.151
5.594
12.240
õ0.05
õ0.001
* Based on nested analysis of variance.
a
Degrees of freedom.
b
Mean square.
c
Sample variance ratio.
Fig. 4.
Distribution of DNA damage (based on length:width ratios of DNA patterns) observed at the
cellular level in R. pipiens tadpoles collected in 1994 at six select sites in southern Ontario. I, industrial;
A, agricultural; N-A, non-agricultural.
required (i.e., a few hundred), any cells that have a nucleus can be used, and the assay has been shown to be
sensitive to a number of mutagens using various tissues
[for reviews, see McKelvey-Martin et al., 1993; Fairbairn
et al., 1995]. The ability of this procedure to identify
‘‘sensitive’’ cells in an otherwise normal population of
cells permits analysis of low dose-responsive relationships [Tice, 1995]. The simplicity and relatively low cost
of the assay makes it suitable for large scale studies.
However, because the alkaline SCG measures only single
504
/ 8I17$$0504
06-09-97 10:51:23
strand breaks and alkali-labile damages, it can miss DNA
alterations involving processes such as cross-linkage induction, nucleotide substitution, and adduct formation.
Yet, to date, very little work has been done using this
procedure for environmental biomonitoring. One application of the assay has been testing for genotoxicity in
environmental samples in the laboratory. Fairbairn et al.
[1994] examined the levels of DNA damage in Raji cells
(a human transformed promyelocytic cell line) exposed
to different water samples collected from four sites in
wlema
W Liss: EM
8I17
426
Ralph and Petras
TABLE V. Seasonal Variations of Detectable DNA Damage in Erythrocytes of Freshly Caught Tadpoles
Year and location
1994 Tallgrass Prairie pond #1
Regiona
Species
I
R. pipiens
I
R. pipiens
A
R. clamitans
New Glasgow pond #1
A
R. clamitans
Southampton
N-A
R. clamitans
Boat Lake
N-A
R. clamitans
Tallgrass Prairie ditch
1995 Duart
Date caught
Sample
size
5/31/94
6/20/94
5/31/94
6/20/94
7/8/94
8/8/94
5/31/95
6/21/95
8/8/95
5/31/95
6/21/95
8/8/95
5/9/95
7/18/95
5/9/95
7/18/95
11
10
4
8
12
9
12
12
12
12
12
12
12
12
12
10
DNA length:width
ratio { SEMb
2.487
2.405
2.956
2.788
2.782
2.650
1.989
1.703
2.182
1.549
1.470
2.117
1.623
1.538
1.921
1.494
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
0.050a
0.135b
0.170
0.154
0.104
0.129
0.074c
0.106d
0.161e
0.101f
0.089g
0.127h
0.078i
0.08j
0.08k
0.079l
Pc
a vs. b ú 0.05
ú0.05 for all comparisons
c vs. d õ 0.05
d vs. e ú 0.05
c vs. e õ 0.05
f vs. g ú 0.05
g vs. h õ 0.01
f vs. h õ 0.001
i vs. j ú 0.05
k vs. 1 õ 0.01
a
I, industrial; A, agricultural; N-A, non-agricultural.
Ratios based on 25 cells/tadpole.
c
Based on Student’s one-tailed t-test.
b
Utah, USA. A significant increase in the levels of DNA
damage, relative to the deionized water control, was observed in cells exposed to two of the four water samples.
Another study by Verschaeve and Gilles [1995] examined
the levels of DNA damage in coelomocytes of earthworms (Eisenia fetida) kept in soil samples collected from
‘‘polluted’’ and ‘‘non-polluted’’ areas. Earthworms from
‘‘polluted’’ soils gave longer comet tails than those obtained from ‘‘non-polluted’’ areas.
The use of the SCG assay in situ using sentinel organisms has, likewise, been limited. Nascimbeni et al. [1991]
examined the extent of DNA damage in four tissues of
the golden mouse, Ochrotomys nuttalli, collected from a
hazardous waste site in North Carolina, USA. These results were compared with mice collected at three control
sites. An increase in the level of DNA damage was observed in all tissues, however, only the brain tissue
showed a significant increase. An SCG protocol has also
been developed to assess DNA damage in several tissues
of the fresh-water medaka fish (Oryzias latipes) [Tice,
1995].
In our laboratory, work has been conducted using several organisms including bullheads (Ameiurus nebulosus)
and carp (Cyprinus carpio). Freshly caught bullheads collected from seven different sites in southern Ontario
showed a wide range of DNA damage [Pandrangi et al.,
1995]. The highest levels of DNA damage were observed
in bullheads collected from western Lake Erie, western
Lake Ontario, and the Detroit River. Studies examining
the levels of polycyclic aromatic hydrocarbons and polychlorinated biphenyls indicated that sediments from these
areas were highly polluted. Significantly lower levels of
504
/ 8I17$$0504
06-09-97 10:51:23
DNA damage were observed in the remaining four sites
that were considered relatively ‘‘clean.’’
The procedure used for tadpoles [Ralph et al., 1996]
was arrived at by modifying the protocol developed and
reported by Singh et al. [1988, 1989].
In this study, a sample was categorized as coming from
an agricultural region if the body of water received runoff from adjacent cultivated fields and/or could be subjected to pesticide drift during spraying. Samples categorized as coming from non-agricultural regions were collected from the Bruce Peninsula in northern Bruce county,
or from a site adjacent to the French River south of Sudbury, Ontario. In both areas, there is little or no agriculture
(no field crops grown or orchards) with most of the land
consisting of pasture or wetland. Samples categorized as
coming from industrial regions were collected in the Tallgrass Prairie Heritage Park. This area is adjacent to several industries located in Windsor, Ontario, and is near
the industrial complexes in Detroit, Michigan, which may
contribute contaminants through atmospheric deposition.
R. clamitans tadpoles collected in 1995 from agricultural areas had significantly higher DNA length:width ratios (P õ 0.001) than tadpoles collected from non-agricultural regions. A similar pattern was seen in the previous
year (Table I).
The increased levels of DNA damage in R. clamitans
tadpoles collected from agricultural regions compared to
that observed in tadpoles collected from non-agricultural
regions is consistent with the types of land uses in these
areas. Within the agricultural regions, the crops grown
include cereal grains (e.g., corn, wheat, oats), fruits (e.g.,
apples, grapes, peaches, pears, plums), soya beans and a
wlema
W Liss: EM
8I17
Genotoxicity Monitoring Using Tadpoles
wide variety of garden vegetables (e.g., lettuce, radishes,
tomatoes, potatoes). Consequently, a broad range of herbicides, insecticides, and fertilizers are added to the chemical load of the area’s waters [Manny et al., 1988]. Starting from Essex county and heading north, there is a reduction in the quantity of active ingredients of pesticides
(herbicides, fungicides, insecticides, nematocides, and
growth regulators) applied (kg/ha of total land) by county
or district (1993) (Fig. 1) [Hunter and McGee, 1994].
Within the study area, the highest level of current pesticide usage is recorded for Elgin county on the north shore
of Lake Erie. This level is about nine times that used in
Bruce county, and more than 1,200 times greater than
that used in the Sudbury district. It should be noted that
the quantity of pesticides used in Bruce county is not
representative of the use of pesticides in the northern half
of this county, the Bruce Peninsula, where there is almost
no growing of crops. These values of pesticide application
should be considered only as crude indices, since use
varies among farms and from area to area within a county.
In addition to current applications, past pesticide usage
may also contribute to the observed levels of DNA damage in tadpoles. Organic pesticides have been in use in
Ontario since the end of the Second World War [Hunter
and McGee, 1994]. Though the use of many of these
compounds has been restricted or banned in recent years,
some still persist in the environment. These include: DDT,
DDD (dichlorodiphenyldichloroethane), DDE (dichlorodiphenyl chloroethane), dieldrin, hexachlorobenzene, heptachlor, and lindane [Socha et al., 1993].
In one study, Johnson and Kauss [1991] examined contaminant levels of water, suspended solids and bottom
sediments in 13 tributaries on the Canadian side of the
St. Clair River, the Detroit River, and Lake St. Clair
(1984–1985), and found a-BHC (alpha-1,2,3,4,5,6hexachlorocyclohexane), an impurity of the insecticide
lindane and p,p*-DDE, a breakdown product of DDT, to
be routinely detected. In addition, they reported that DDT
and dieldrin were detected in 1984–1985, even though
their use has been severely restricted since the early
1970s. Johnson and Kauss found that phenols, which can
arise from the degradation of phenolic pesticides
[CCREM, 1995], along with metals such as aluminum,
cadmium, copper, and iron, frequently exceeded Provincial Water Quality Objectives (PWQO). The presence of
cadmium in some tributaries, draining mainly agricultural
areas, may be attributed to high cadmium levels associated with fertilizers [Great Lakes Institute, 1986]. Also,
industrial organic compounds such as hexachlorobenzene
(HCB) and octachlorobenzene were frequently detected
and, in some cases, exceeded PWQOs. A number of sites
in the Lake Huron–Lake Erie corridor (e.g., St. Clair
River, Detroit River, and the Clinton River (Michigan)
draining into Lake St. Clair) have been classified as
504
/ 8I17$$0504
06-09-97 10:51:23
427
‘‘Areas of Concern’’ (1995) [International Joint Commission, 1995].
In comparison, examination of water quality data of
three rivers (Maitland, Saugeen, and Ausable, Fig. 1), that
are farther north and drain into Lake Huron, indicates that
the maximum detected levels of pesticides (p,p*-DDE, aBHC), industrial organic compounds (HCB, phenols) and
inorganics (cadmium, copper, iron, lead, mercury) are at
levels much lower than that observed in the 13 tributaries
mentioned above, with PWQOs being exceeded in a few
cases, but only marginally [OMEE, 1992]. No data are
available for suspended solids and sediment contaminants
in this region. Though no contaminant burden data are
available for streams or rivers draining the Bruce Peninsula or for the French River, these regions are not classified as ‘‘Areas of Concern’’ and, in general, are considered relatively ‘‘pristine.’’ The tributary contaminant information given should serve only as a broad index of
pollution levels.
Levels of DNA damage in R. clamitans tadpoles collected annually from agricultural regions over the three
years were significantly different (P õ 0.001, Table II).
There was significant variance (P õ 0.001) among samples within years in the agricultural regions, but this was
less than the variance among years. The variance in the
levels of DNA damage in R. clamitans tadpoles collected
in 1994 and 1995 from non-agricultural regions, though
significant (P õ 0.01), was about equal between years
and among samples within years (Table II). The impact
of varying weather conditions and agricultural practices
may account for the greater amount of annual variation
in DNA damage seen in samples of tadpoles from agricultural regions relative to non-agricultural regions. For instance, leaving a field fallow in a given year would greatly
reduce the input of genotoxic agents from pesticide runoff and may contribute to the annual fluctuations. Crop
rotation would also impact the levels of genotoxic agents
since pesticide usage and practices vary with the type of
crop being grown. Also, the timing of rainfall and pesticide application can greatly impact the concentration of
genotoxicants entering adjacent small bodies of water.
For instance, more than 90% of the total annual pesticide
run-off load of a given pesticide is the result of rainfall
shortly after application [Stover and Hamill, 1994]. Rainfall at other times serves to dilute genotoxicants in the
small bodies of water studied which often contain low
volumes of water. However, a specific cause for the variability in DNA damage cannot be provided from this
study.
A comparison among R. clamitans tadpoles collected
at approximately the same time each year, from the same
sites, indicated that overall there was a significant reduction in DNA damage. The cause of this reduction is not
obvious. Changes in rainfall patterns, especially at the
time the collections were made, could cause annual fluc-
wlema
W Liss: EM
8I17
428
Ralph and Petras
tuations. Baker [1993], for instance, reporting on water
quality in tributaries draining agricultural areas on the
southern shore of Lake Erie over an 18-year period, indicated that there is large day-to-day, season-to-season, and
year-to-year variability in pollutant concentrations and
loads. This appears to be characteristic of non-point
source pollution and is consistent with our findings. A
comparison of DNA damage levels in samples of R. clamitans tadpoles collected from agricultural and non-agricultural areas in 1994 and 1995 indicated that greater variation was seen between regions within years than between
years, or among samples of tadpoles within regions and
within years (Table II). However, all of these were significantly variable (P õ 0.001). The different levels of
pesticides used in each of these two regions appears to
be the most likely source of this variation, but other
sources of pollution cannot be ruled out from this study.
No comparison was made with samples from the industrial regions, since only one sample of R. clamitans tadpoles was collected over 1994 and 1995.
The overall geographic pattern of DNA damage in R.
pipiens tadpoles was similar to that observed in R. clamitans tadpoles (Table IV). In addition, in R. pipiens tadpoles there were high levels of DNA damage in samples
from the industrial region (Tallgrass Prairie, Table III).
This can be attributed to a number of factors. Atmospheric
fallout has been shown to be an important pathway for
the introduction of some toxic substances into the Great
Lakes [Eisenreich et al., 1981; Arimoto, 1989; Kelly et
al., 1991]. Deposition of atmospheric contaminants from
the industrial complexes in Detroit, Michigan (2 km to
the west) and Toledo, Ohio (70 km to the southwest),
may contribute genotoxicity to this area since the prevalent winds are from the west and southwest. Industries
located in Detroit include oil refineries, steel mills, and
auto manufacturers. Located in Windsor, within 2 km of
the study site are a sewage treatment plant, a petrochemical installation, and a recently closed sanitary landfill. In
addition, there are also several auto manufacturing plants
in the area. Finally, internal combustion engine emissions
associated with the large populations of Detroit and Windsor (approximately 4 million and 200,000 citizens, respectively), likewise contribute genotoxicity to the area. There
is, however, no direct evidence of the specific contribution
that any of these sources have made to the contaminant
load [see Edsall et al., 1988; Manny et al., 1988].
A comparison between R. pipiens tadpoles collected at
the same time in consecutive years (6/20/94 and 6/20/95)
from the Tallgrass Prairie ditch indicated that levels of
DNA damage did not differ significantly (P ú 0.05) between years (Table III). In addition, R. pipiens tadpoles
collected periodically during the summer of 1994 at the
Tallgrass Prairie ditch did not demonstrate any significant
(P ú 0.05) seasonal variation (Table V). Tadpoles collected from industrial regions would not be subject to the
504
/ 8I17$$0504
06-09-97 10:51:23
variation in seasonal agricultural practices and in fact,
based on the levels of DNA damage, it appears that the
input of contaminants was stable over the study period.
Seasonal increases in DNA damage in R. clamitans
tadpoles were typically observed at locations in agricultural regions (Table V). These may be associated with
the use of herbicides. In southwestern Ontario, application
of herbicides begins in May and continues into June.
Other pesticides (insecticides, nematocides, fungicides,
growth regulators) may be applied throughout the year.
The effect of herbicides does not appear to be immediate.
Laboratory studies on R. catesbeiana tadpoles, using the
alkaline SCG assay, suggested that the effects of the tested
herbicides (atrazine, metolachlor, glyphosate, metribuzin
and 2,4-D amine) can be seen after only a 24-hr exposure
[Clements et al., 1997]. The concentrations found to be
genotoxic were 10–75,000 times lower than the recommended application concentrations. Whether the apparent
delay in in situ studies is due to the contaminants making
their way through the ecosystem, or simply the gradual
bioaccumulation of low levels of contaminants, has yet
to be established. The lack of a seasonal increase in DNA
damage in tadpoles from some agricultural areas may, as
mentioned previously, be due to the impact of leaving
fields fallow, crop rotation, rainfall, and/or the timing of
pesticide application.
Among sites in non-agricultural regions, there was typically either a seasonal decrease or no change in DNA
damage in R. clamitans tadpoles (Table V). These
changes may, for the most part, be the result of fluctuations in the general environment. One exception may be
the 1994 samples from Spry Lake in which a seasonal
increase in DNA damage was observed. This increase
may be due to the dredging that took place in the time
between the two collections.
Two other species, Pseudacris nigrita triseriata
(chorus frog) and Bufo americanus (American toad) were
also collected, but in smaller numbers. Though these tadpoles are considerably smaller, 22–30 mm in length at
transformation [Walker, 1946], they were sufficiently
large to use on an individual basis in the SCG assay.
In conclusion, our results indicate that a combination
of the alkaline SCG assay and any of the four species of
tadpoles would provide a suitable monitoring system for
environmental genotoxicity. Based on R. clamitans and
R. pipiens tadpoles collected from the 18 sites in southern
Ontario, there appears to be a reduction in DNA damage
in the small bodies of water, as one goes from industrial
to agricultural and then to non-agricultural regions. This
pattern corresponds with the limited information available
concerning contaminant, especially herbicide, burdens in
streams and rivers in these areas. There was significant
annual, regional and seasonal variation in the levels of
DNA damage in samples of R. clamitans tadpoles from
both agricultural and non-agricultural regions. The annual
wlema
W Liss: EM
8I17
Genotoxicity Monitoring Using Tadpoles
and seasonal variation was greater in agricultural samples.
There were no annual or seasonal variations in the levels
of DNA damage in R. pipiens tadpoles collected from
one industrial region.
ACKNOWLEDGMENTS
We thank Dr. Raymond Tice, Integrated Laboratory
Systems, Research Triangle Park, NC, for introducing us
to this assay and for comments on an earlier version of
this article; M. Vrzoc for her technical assistance; C.
Clements and V.K. Perry for their laboratory/field assistance; S. Dilliot, P. and J. Dunker, G. Wiehle and R.
Wilcocks for allowing us to collect numerous field samples on their farms; T. Linke for serving as our liaison
with the Point Pelee National Park personnel; and the
many others who have contributed to the project. This
study was supported by grants from the Great Lakes and
Universities Research Fund to G.D. Haffner et al., from
the University of Windsor Research Board to M.L. Petras,
and from the Summer Employment/Experience Development program of Employment and Immigration Canada
to M.L. Petras.
REFERENCES
Arimoto R (1989): Atmospheric deposition of chemical contaminants
to the Great Lakes. J Great Lakes Res 15:339–356.
Baker DB (1993): The Lake Erie Agroecosystem program: Water quality
assessments. Agri Ecosys Environ 46:197–215.
Bauer Dial CA, Dial NA (1995): Lethal effects of consumption of field
levels of paraquat-contaminated plants on frog tadpoles. Bull
Environ Contam Toxicol 55:870–877.
Berrill M, Bertram S, Wilson A, Louis S, Brigham D, Stromberg C
(1993): Lethal and sublethal impacts of pyrethroid insecticides
on amphibian embryos and tadpoles. Environ Toxicol Chem
12:525–539.
Berrill M, Bertram S, McGillivray L, Kolohon M, Pauli B (1994):
Effects of low concentrations of forest-use pesticides on frog
embryos and tadpoles. Environ Toxicol Chem 13:657–664.
Berrill M, Bertram S, Pauli B, Coulson D, Kolohon M, Ostrander D
(1995): Comparative sensitivity of amphibian tadpoles to single
and pulsed exposures of the forest-use insecticide fenitrothion.
Environ Toxicol Contam 14:1011–1018.
Brooks JA (1981): Otolith abnormalities in Limnodynastes tasmaniensis
tadpoles after embryonic exposure to the pesticide dieldrin. Environ Pollut Ser A 25:19–25.
Canadian Council of Resource and Environment Ministers (CCREM)
(1995): ‘‘Canadian Water Quality Guidelines.’’ Ottawa: Health
Canada.
Clements C, Ralph S, Petras M (1997): Genotoxicity of select herbicides
in Rana catesbeiana tadpoles using the alkaline single cell gel
(comet) assay. Environ Mol Mutagen 29:277–288.
Cooke AS (1979): The influence of rearing density on the subsequent
response to DDT dosing for tadpoles of the frog Rana temporaria. Bull Environ Contam Toxicol 21:837–841.
Djomo JE, Ferrier V, Gauthier L, Zoll-Moreux C, Marty J (1995):
Amphibian micronucleus test in vivo: Evaluation of the genotoxicity of some major polycyclic aromatic hydrocarbons found in
crude oil. Mutagenesis 10:223–226.
504
/ 8I17$$0504
06-09-97 10:51:23
429
Edsall TA, Manny BA, Raphael CN (1988): ‘‘The St. Clair River and
Lake St. Clair, Michigan: An ecological profile.’’ U.S. Fish Wildl
Serv Washington, DC. Biol Rep 85.
Eisenreich S, Looney B, Thornton J (1981): Airborne organic contaminants in the Great Lakes ecosystem. Environ Sci Tech 15:30–
38.
Elliott-Freely E, Armstrong JB (1982): Effects of fenitrothion and carbaryl on Xenopus laevis development. Toxicology 22:319–335.
Fairbairn D, Meyers D, O’Neil K (1994): Detection of DNA damaging
agents in environmental water samples. Bull Environ Contam
Toxicol 52:687–690.
Fairbairn DW, Olive PL, O’Niel KL (1995): The comet assay: A comprehensive review. Mutat Res 339:37–59.
Great Lakes Institute (1986): ‘‘A Case Study of Selected Toxic Contaminants in the Essex Region, Vol. 1.’’ Windsor: Great Lakes Institute.
Hall RJ, Swineford DM (1980): Toxic effects of endrin and toxaphene
on the southern leopard frog (Rana sphenocephala). Environ
Pollut Ser A 23:53–65.
Hart D, Armstrong J (1984): Assessment of mutagenic damage by
monofunctional alkylating agents and radiation in haploid and
diploid frogs, Xenopus laevis. Environ Mutagen 6:719–735.
Hunter C, McGee B (1994): ‘‘Survey of Pesticide Use in Ontario,
1993.’’ Toronto: Ontario Ministry of Agriculture, Food and Rural
Affairs.
International Joint Commission (1995): ‘‘1993–1995 Priorities and
Progress Under the Great Lakes Water Quality Agreement.’’
Windsor: International Joint Commission.
Jaylet A, Deparis P, Ferrier V, Grinfeld S, Siboulet R (1986): A new
micronucleus test using peripheral blood erythrocytes of the newt
Pleurodeles waltl to detect mutagens in fresh water pollution.
Mutat Res 164:245–257.
Jaylet A, Gauthier L, Fernandez M (1987): Detection of mutagenicity
in drinking water using a micronucleus test in newt larvae (Pleurodeles waltl). Mutagenesis 2:211–214.
Johnson GD, Kauss PB (1991): ‘‘Tributary Contaminant Inputs to the
St. Clair and Detroit Rivers and Lake St. Clair, 1984–1985.’’
Toronto: Ontario Ministry of the Environment.
Kaplan HM, Overpeck JG (1964): Toxicity of halogenated hydrocarbon
insecticides for the frog, Rana pipiens. Herpetolology 20:164–
169.
Kelly T, Czuczwa J, Sticksel P, Sverdrup G (1991): Atmospheric and
tributary inputs of toxic substances to Lake Erie. J Great Lakes
Res 17:504–516.
Krauter P (1992): Micronucleus incidence and hematological effects in
bullfrog tadpoles (Rana catesbeiana) exposed to 2-acetylaminofluorene and 2-aminofluorene. Arch Environ Contam Toxicol 23:487–493.
Landolt ML, Kocan RM (1983): Fish cell cytogenetics: A measure of
the genotoxic effects of environmental pollutants. In Nriagu JO
(ed): ‘‘Advances in Environmental Science and Technology.’’
New York: Plenum Press, pp 335–353.
LeCurieux F, Marzin D, Erb F (1992): Genotoxic activity of three
carcinogens in peripheral blood erythrocytes of the newt Pleurodeles waltl. Mutat Res 283:157–160.
Manny BA, Edsall TA, Jawarski E (1988): ‘‘The Detroit River, Michigan: An ecological profile.’’ U.S. Fish Wildl Serv Washington,
D.C. Biol Rep 85.
Marchal-Segault D, Ramande R (1981): The effects of lindane, an insecticide, on hatching and post embryonic development of Xenopus
laevis (Daudin) anuran amphibians. Environ Res 24:250–258.
Materna EJ, Rabeni CF, LaPoint TW (1995): Effects of the synthetic
pyrethroid insecticide, esfenvalerate, on larval leopard frogs. Environ Toxicol Chem 14:613–622.
McKelvey-Martin VJ, Green MHL, Schmezer P, Pool-Zobel BL, De-
wlema
W Liss: EM
8I17
430
Ralph and Petras
Meo MP, Collins A (1993): The single cell gel electrophoresis
assay (comet assay): A European review. Mutat Res 288:47–63.
Nascimbeni B, Phillips MO, Croom DK, Andrews PW, Tice RR (1991):
Evaluation of DNA damage in golden mice (Ochrotomys nuttalli)
inhabiting a hazardous waste site. Environ Mol Mutagen 17
(Suppl 19):55.
Olive PL, Banath JP, Durand RE (1990): Heterogeneity in radiationinduced DNA damage and repair in tumour and normal cells
using the ‘‘comet assay.’’ Radiat Res 122:86–94.
Ontario Ministry of Environment and Energy (OMEE) (1992): ‘‘Water
Quality Data Ontario Lakes and Streams, 1990, Vol. XXVI,
Southwestern Region.’’ Toronto: Ontario Ministry of Environment and Energy.
Osborn D, Cooke AS, Freestone S (1981): Histology of a teratogenic
effect of DDT on Rana temporaria tadpoles. Environ Pollut Ser
A 25:305–319.
Ostling O, Johanson KJ (1984): Microelectrophoretic study of radiationinduced DNA damages in individual mammalian cells. Biochem
Biophys Res Commun 123:291–298.
Pandrangi R, Petras M, Ralph S, Vrzoc M (1995): Alkaline single cell
gel (comet) assay and genotoxicity monitoring using bullheads
and carp. Environ Mol Mutagen 26:345–356.
Ralph S, Petras M, Pandrangi R, Vrzoc M (1996): Alkaline single cell
gel (comet) assay and genotoxicity monitoring using two species
of tadpoles. Environ Mol Mutagen 28:112–120.
Rudek Z, Rozek M (1992): Induction of micronuclei in tadpoles of
Rana temporaria and Xenopus laevis by the pyrethroid Fastac
10 EC. Mutat Res 298:25–29.
Russell R, Hecnar S, Haffner D (1995): Organochlorine pesticide residues in southern Ontario spring peepers. Environ Toxicol Chem
14:815–817.
Rydberg B, Johanson KJ (1978): Estimation of DNA strand breaks in
single mammalian cells. In Hanawalt PC, Friedberg EC, Fox CF
(eds): ‘‘DNA Repair Mechanisms.’’ New York: Academic Press,
pp 465–468.
Siboulet R, Grinfeld S, Deparis P, Jaylet A (1984): Micronuclei in red
blood cells of the newt Pleurodeles waltl Michah: Induction with
X-rays and chemicals. Mutat Res 125:275–281.
504
/ 8I17$$0504
06-09-97 10:51:23
Singh NP, McCoy MT, Tice RR, Schneider EL (1988): A simple technique for quantitation of low levels of damage in individual cells.
Exp Cell Res 175:184–191.
Singh NP, Banner DB, McCoy MT, Collins GD, Schneider EL (1989):
Abundant alkali-sensitive sites in DNA of human and mouse
sperm. Exp Cell Res 184:461–470.
Socha AC, Aucoin R, Dickie T, Angelow R, Kauss P (1993): ‘‘Candidate Substances for Bans, Phase-outs or Reductions—Multimedia Revision.’’ Toronto: Ontario Ministry of Environment and
Energy.
Sokal RR, Rohlf JF (1995): ‘‘Biometry, 3rd ed.’’ New York: WH Freeman and Company.
Stover J, Hamill AS (1994): Pesticide contamination of surface waters
draining agricultural fields: Pesticide contamination classification
and abatement resources. In Wall GJ, Coote DR, Dekimse C,
Hamill AS, Marks F (eds): ‘‘Great Lakes Water Quality Program
Summary of Achievements, 1989–1994.’’ London: Agriculture
and Agri-Food Canada.
Tice RR (1995): Applications of the single cell gel assay to environmental biomonitoring for genotoxic pollutants. In Butterworth FM,
Corkum LD, Guzmán-Rincón J (eds): ‘‘Biomonitoring and Biomarkers as Indicators of Environmental Change.’’ New York:
Plenum Press, pp 69–79.
Verschaeve L, Gilles J (1995): Single cell gel electrophoresis assay in
the earthworm for the detection of genotoxic compounds in soils.
Bull Environ Contam Toxicol 54:112–119.
Vijayalaxmi, Tice RR, Strauss GHS (1992): Assessment of radiationinduced DNA damage in human blood lymphocytes using the
single-cell gel electrophoresis technique. Mutat Res 271:243–
252.
Walker C (1946): ‘‘The Amphibians of Ohio.’’ Columbus: Ohio State
Archaeological and Historical Society.
Zoll C, Saouter E, Boudou A, Ribeyre F, Jaylet A (1988): Genotoxicity
and bioaccumulation of methyl mercury and mercuric chloride
in vivo in the newt Pleurodeles waltl. Mutagenesis 3:337–343.
wlema
Accepted by—
B. Pool-Zobel
W Liss: EM
8I17