This article was downloaded by:
On: 28 October 2008
Access details: Access Details: Free Access
Publisher Informa Healthcare
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Critical Reviews in Toxicology
Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713401167
Effects of Atrazine on Fish, Amphibians, and Aquatic Reptiles: A Critical Review
Keith R. Solomon a; James A. Carr b; Louis H. Du Preez c; John P. Giesy def; Ronald J. Kendall g; Ernest E.
Smith g; Glen J. Van Der Kraak h
a
Department of Environmental Biology and Centre for Toxicology, University of Guelph, Guelph, Ontario,
Canada b Department of Biological Sciences, Texas Tech University, Lubbock, Texas c School of
Environmental Sciences and Development, North West University, Potchefstroom, South Africa d Toxicology
Centre, University of Saskatchewan, Saskatoon, Saskatchewan, Canada e Department of Biology and
Chemistry, City University of Hong Kong, Kowloon, Hong Kong, SAR, China f Zoology Department, Michigan
State University, East Lansing, Michigan g Institute of Environmental and Human Health and Department of
Environmental Toxicology, Texas Tech University, Lubbock, Texas h Department of Integrative Biology,
University of Guelph, Guelph, Ontario, Canada
Online Publication Date: 01 October 2008
To cite this Article Solomon, Keith R., Carr, James A., Du Preez, Louis H., Giesy, John P., Kendall, Ronald J., Smith, Ernest E. and
Van Der Kraak, Glen J.(2008)'Effects of Atrazine on Fish, Amphibians, and Aquatic Reptiles: A Critical Review',Critical Reviews in
Toxicology,38:9,721 — 772
To link to this Article: DOI: 10.1080/10408440802116496
URL: http://dx.doi.org/10.1080/10408440802116496
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf
This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.
Critical Reviews in Toxicology, 38:721–772, 2008
c 2008 Informa UK Ltd.
Copyright ISSN: 1040-8444 print / 1547-6898 online
DOI: 10.1080/10408440802116496
Effects of Atrazine on Fish, Amphibians, and Aquatic
Reptiles: A Critical Review
Keith R. Solomon
Department of Environmental Biology and Centre for Toxicology, University of Guelph, Guelph,
Ontario, Canada
James A. Carr
Department of Biological Sciences, Texas Tech University, Lubbock, Texas
Louis H. Du Preez
School of Environmental Sciences and Development, North West University, Potchefstroom, South Africa
John P. Giesy
Downloaded At: 15:43 28 October 2008
Toxicology Centre, University of Saskatchewan, Saskatoon, Saskatchewan, Canada, Department of
Biology and Chemistry, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, SAR,
China, and Zoology Department, Michigan State University, East Lansing, Michigan
Ronald J. Kendall and Ernest E. Smith
Institute of Environmental and Human Health and Department of Environmental Toxicology, Texas Tech
University, Lubbock, Texas
Glen J. Van Der Kraak
Department of Integrative Biology, University of Guelph, Guelph, Ontario, Canada
The herbicide atrazine is widely used in agriculture for the production of corn and other crops.
Because of its physical and chemical properties, atrazine is found in small concentrations in
surface waters—habitats for some species. A number of reports on the effects of atrazine on
aquatic vertebrates, mostly amphibians, have been published, yet there is inconsistency in the
effects reported, and inconsistency between studies in different laboratories. We have brought
the results and conclusions of all of the relevant laboratory and field studies together in this critical review and assessed causality using procedures for the identification of causative agents of
disease and ecoepidemiology derived from Koch’s postulates and the Bradford–Hill guidelines.
Based on a weight of evidence analysis of all of the data, the central theory that environmentally relevant concentrations of atrazine affect reproduction and/or reproductive development
in fish, amphibians, and reptiles is not supported by the vast majority of observations. The same
conclusions also hold for the supporting theories such as induction of aromatase, the enzyme
that converts testosterone to estradiol. For other responses, such as immune function, stress endocrinology, parasitism, or population-level effects, there are no indications of effects or there
is such a paucity of good data that definitive conclusions cannot be made.
Keywords adverse effects, amphibians, atrazine, fish, endocrine, reptiles
Table of Contents
I.
INTRODUCTION ............................................................................................................................................ 723
II.
CHEMICAL, PHYSICAL, AND BIOLOGICAL PROPERTIES ...................................................................... 723
A. Environmental Behavior of Atrazine ............................................................................................................. 723
Address correspondence to Keith Solomon, Centre for Toxicology, University of Guelph, Guelph, ON, N1G 2W1 Canada. E-mail:
ksolomon@uoguelph.ca
721
722
K. R. SOLOMON ET AL.
Downloaded At: 15:43 28 October 2008
B. Mechanism of Action .................................................................................................................................. 726
C. Bioconcentration/Bioaccumulation ............................................................................................................... 727
III.
ACUTE AND CHRONIC TOXICITY .............................................................................................................. 727
A. Lethality and Physiological Effects ............................................................................................................... 727
IV.
EXTERNAL DEVELOPMENTAL ABNORMALITIES ................................................................................... 731
A. Trematode Infections ................................................................................................................................... 731
B. Direct Effects of Atrazine on Limb Deformities ............................................................................................. 732
V.
SEXUAL DIFFERENTIATION AND DEVELOPMENT .................................................................................. 733
A. The Use of Xenopus laevis as a Model for Endocrine Responses ...................................................................... 733
B. Effects on Sex Ratio .................................................................................................................................... 734
C. Effects on Sexual Development .................................................................................................................... 734
VI.
MECHANISMS MEDIATING REPRODUCTIVE EFFECTS .......................................................................... 739
A. Mechanisms Mediated Through the HPG Axis ............................................................................................... 740
B. Mechanisms Mediated Through Aromatase ................................................................................................... 740
C. Effects of Atrazine on Plasma Sex Steroid Hormones in Amphibians and Fish .................................................. 742
VII.
EFFECTS ON LARYNGEAL DEVELOPMENT ............................................................................................. 745
A. Effects on the Laryngeal Dilator Muscle ........................................................................................................ 745
VIII. EFFECTS ON THYROID FUNCTION AND DEVELOPMENT ...................................................................... 745
A. Effects of Atrazine on Amphibian Metamorphosis .......................................................................................... 746
B. Effects of Atrazine on Smoltification ............................................................................................................. 746
IX.
EFFECTS ON STRESS PHYSIOLOGY .......................................................................................................... 747
A. Effects on Plasma Corticosteroids ................................................................................................................. 747
B. Effects on Adrenal Steroidogenesis and Secretion .......................................................................................... 747
C. Effects of Pesticide Mixtures on Corticosteroid Secretion ............................................................................... 748
X.
EFFECTS ON IMMUNE FUNCTION ............................................................................................................. 749
A. Effects of Atrazine on Immune Function in Fish ............................................................................................. 749
B. Effects of Atrazine on Immune Function in Amphibians ................................................................................. 750
XI.
EFFECTS OF ATRAZINE ON BEHAVIOR .................................................................................................... 754
A. Effects on Olfactory Neurons and Behavior in Fish ........................................................................................ 754
B. Effects on Behavior in Amphibians ............................................................................................................... 756
XII.
EFFECTS OF ATRAZINE AT THE POPULATION LEVEL ........................................................................... 756
A. Atrazine and Reptiles ................................................................................................................................... 756
B. Atrazine and Fish ........................................................................................................................................ 756
C. Atrazine and Amphibians ............................................................................................................................. 757
XIII. OVERALL CONCLUSIONS, AND RESEARCH DIRECTIONS ..................................................................... 762
A. Strengths and Uncertainties .......................................................................................................................... 762
B. Conclusions ................................................................................................................................................ 762
XIV. SUMMARY ..................................................................................................................................................... 764
ACKNOWLEDGMENTS ........................................................................................................................................... 764
REFERENCES .......................................................................................................................................................... 764
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
Downloaded At: 15:43 28 October 2008
I.
INTRODUCTION
The potential for ecological risks of atrazine (CAS number
1912-24-9) in fresh and estuarine waters has been extensively
reviewed with respect to its potential effects on a number of
endpoints and processes in ecological systems (Solomon et al.,
1996; Giddings et al., 2005). The primary focus was on plants,
arthropods, and fish exposed to typical concentrations. These
assessments concluded that atrazine did not present significant
acute or chronic ecological risks. Although some of the measures
of effect used in these assessments, such as those in full-lifecycle studies on several species of fish and those in microcosms,
include the aggregate responses of many possible mechanisms
of action, these have not specifically characterized reproduction, the endocrine system, or development as possible targets
for atrazine. In the 1990s, a number of studies reported that
atrazine affected the reproductive system of mammals at large
exposure concentrations (≥40 mg/kg). Specifically an increased
incidence of mammary tumors was observed in female SpragueDawley (SD) rats. This effect was strain-, sex-, and speciesspecific and judged to not be relevant to humans (U.S. EPA,
2000). Considerable research on the mechanism of this response
has been conducted and was recently reviewed (Gammon et al.,
2005, Cooper et al., 2007). The mammalian data are discussed
briefly in relevant sections of this review, which is focused on
the identified lack of data for sublethal endpoints related to development and reproduction in nontarget terrestrial, aquatic, and
semi-aquatic species, such as reptiles and amphibians (Solomon
et al., 1996, Giddings et al., 2005). Since that time, a number of
reports on the effects of atrazine on aquatic wildlife, mostly amphibians, have been published, yet there is inconsistency in the
effects reported, and inconsistency between studies in different
laboratories.
This review was written to provide a critical assessment of
current information on atrazine and the effects that it may have on
fish, amphibians, and aquatic reptiles. Much of the review is focused on amphibians because more studies have been conducted
in these organisms. In conducting this work we followed the general review guidelines as outlined by Weed (1997). We searched
the current and historical literature using PubMed, Scopus, Science Direct, and the Agricola databases. In addition, we used
books and other reviews to obtain information from the older
literature. We also included information presented at meetings
where the results were not described by the authors as preliminary and where a physical copy of the presentation was available
(such as a poster). In collecting this information, we attempted
to be inclusive and did not exclude papers or information based
on source. We did, however, critically assess the validity and
quality of the study and results based on the published description of the study design and the interpretation of the results by
the authors. In assessing all of this evidence, we used guidelines for causality developed from those of Koch (1942) and
Bradford-Hill (1965) as modified for assessment of causality
associated with adverse effects of substances that are mediated
723
through endocrine and developmental pathways (IPCS, 2002).
In organizing this review, we have summarized key information
related to the properties of atrazine and how these may affect
exposures and then focused the bulk of the review on the effects
of atrazine at the level of the individual through the population.
Effects on individuals were assessed from the point of view of
responses at various levels of biological organization as well as
information on mechanisms of action. This review is divided into
a number of sections, each with a specific focus. In developing
overall conclusions, we assessed the strength of the evidence on
the basis of the guidelines for causality and also identified the
relevance of these to the conclusions.
II.
CHEMICAL, PHYSICAL, AND BIOLOGICAL
PROPERTIES
Detailed descriptions of the chemical and physical characteristics of atrazine have been presented previously (Solomon
et al., 1996) and recently updated (Giddings et al., 2005), and
only the properties of atrazine most relevant to its environmental fate and effects are presented here. Because of its properties,
atrazine is most commonly found in surface waters and, to a
small extent, in groundwater. Atrazine is relatively persistent
in higher pH surface waters associated with its major uses in
agriculture. Thus, much of the focus on atrazine exposures in
the literature has been directed to the water route and a similar focus is followed here. In addition, several key points are
highlighted. These include the primary mechanism of action of
atrazine as a photosynthesis inhibitor—a mode of action that is
specific to plants and therefore confers selectivity to other organisms in terms of acute toxicity. The greater persistence of
atrazine increases the probability that aquatic organisms will be
exposed; however, there is little bioaccumulation in aquatic organisms such as fish and frogs (BCFs range from <1 to about
8), it is rapidly metabolized in fish and frogs, and there is no
evidence of concentration in specific tissues such as those implicated in gonadal effects. Lack of bioaccumulation is consistent
with water solubility, low K OW , and metabolic degradation and
excretion. Thus, although moderately persistent in the environment, atrazine does not bioconcentrate nor does it biomagnify in
the food chain. In organisms exposed through the water matrix,
concentrations in the organisms will be similar to those in the
water and, should these vary, will closely follow them. Transfer of atrazine from the F1 to F2 generation will be negligible.
Thus, from the point of view of toxicological and physiological
responses, water concentrations during the period of exposure
are the critical determinants of potential effects.
A. Environmental Behavior of Atrazine
Persistence in the Environment
The s-triazine ring makes the atrazine molecule somewhat
resistant to microbiological degradation in aquatic systems
(Howard, 1991). Chemical degradation occurs by hydrolysis and
N -dealkylation. Photolysis of atrazine does not occur in water
724
K. R. SOLOMON ET AL.
Downloaded At: 15:43 28 October 2008
FIG. 1. Structure of atrazine and its major metabolites.
at wavelengths greater than 300 nm. Half-lives in water for six
studies ranged from 41 to 237 d with a mean of 159 d (SD =
71 d) (Giddings et al., 2005), and those in anaerobic soil or sediment ranged from 58 to 547 d with a mean of 228 d (SD =
168 d), and field dissipation half lives ranged from 8 to 99 d
(Novartis, 2000). The relative persistence of atrazine in surface
waters increases the potential for exposure in aquatic organisms,
particularly those in static water systems.
Transformation Products
Known transformation products of atrazine are desethylatrazine (DEA), hydroxyatrazine (HA), desisopropylatrazine
(DIA), diaminochlorotriazine (DACT), and two dealkylated hydroxyatrazines, desethylhydroxyatrazine (DEHA), and desisopropylhydroxyatrazine (DIHA) (Fig. 1). Formation of these
transformation products has been measured in laboratory studies, and some data are available for transformation products
detected in the aquatic environment (summarized in Giddings
et al., 2005). Measurements of the persistence of atrazine transformation products are limited. Soil half-lives have been reported as 26 d for DEA, 17 d for DIA, 19 d for DAC, and 121 d
for HA (Winkelmann and Klaine, 1991); however, aqueous halflives of atrazine degradates and metabolites are not available.
Bioconcentration data for atrazine transformation products are
also limited, since studies reporting biological concentration focus on metabolite production in biota upon exposure to atrazine,
rather than uptake of these degradates from environmental media. Based on results of studies in mammals and other organisms
(Sanderson et al., 2001), these metabolites should be considered
in assessing potential risks.
Use Pattern and Geographic Distribution
Atrazine is usually applied preemergent as a water-dispersed
spray, although preplant and postemergent applications are also
used. Typically, a single soil/field application of 1 kg/ha (1.1
lbs a.i./acre) is made by ground equipment. The estimated total
for all uses of atrazine in 2001 was 35–36 million tonnes active
ingredient (a.i.) (U.S. EPA, 2004). Most atrazine use occurs on
corn (85% of total), mostly in the Midwestern United States,
which is also where the most acres are planted in corn. Applications of atrazine in Illinois, Iowa, Nebraska, Indiana, Ohio,
and Missouri alone accounted for approximately 60% of the
total atrazine applied to corn in the United States in 1998 (Giddings et al., 2005). Atrazine is used on sorghum throughout the
United States, which accounts for approximately 10% of total
atrazine use. Atrazine use on sugarcane in parts of Florida and
Louisiana represents approximately 2.5% of the total annual use
of atrazine. Giddings et al. (2005) combined data on rainfall and
atrazine use in a geographic information system (GIS) to define climate-use areas of greatest likelihood of atrazine runoff
and showed that measurements from surface water monitoring
stations were centered near these greater risk areas. Freshwater
ecosystems in these regions are at greatest risk from exposures
to atrazine (Fig. 2).
Pathways for Exposure in Wildlife
Several exposure pathways for aquatic wildlife are possible,
but some are more important than others (Fig. 3). Terrestrial
wildlife can be exposed to atrazine via consumption of contaminated food or water. Estimates of exposure of terrestrial wildlife
range from 60 to 960 mg/kg for a small herbivore (15 g) exposed
at the rate of 4 kg/ha used in sugar cane to 16.5–264 mg/kg for
a small herbivore exposed to the more typical rate of 1.2 kg/ha
used in corn (U.S. EPA, 2003c).
Runoff and erosion are the major routes of atrazine entry into surface waters, while leaching and lateral movement
through the soil or tile drains are a secondary route of entry
(Giddings et al., 2005). Due to typical methods of atrazine application, spray drift is a minor route of exposure (Giddings
et al., 2005; Solomon et al., 1996), and volatilization and
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
725
FIG. 2. Atrazine climate-use regions showing areas with higher use and rainfall and greater potential risk (redrawn from data of
Giddings et al., 2005).
co-distillation with subsequent rainout are other relatively minor routes of exposure. Habitats where there is possible exposure
of aquatic wildlife include natural ponds and ephemeral pools
to farm ponds, streams, rivers, reservoirs, lakes, and eventually
saltwater. Estuarine and marine environments are not important
habitats for amphibians and are only relevant to fish and a few
reptiles. Because of dilution in these environments, exposures
are likely to be very small and do not present a direct or indirect risk (Solomon et al., 1996; Giddings et al., 2005). Significant differences in exposure regimens exist for the remaining
FIG. 3. Possible routes of exposure of wildlife to atrazine. The width of the arrow indicates the relative importance of the exposure
route.
Downloaded At: 15:43 28 October 2008
726
K. R. SOLOMON ET AL.
habitats that are most relevant to fish, amphibians, and possibly
reptiles.
Analysis of extensive data sets of atrazine concentration measurements in running water showed that exposures are pulses
highly correlated with rainfall-driven runoff (Giddings et al.,
2005). Thus, fish, amphibians, and reptiles in these environments
are likely exposed to atrazine for short durations with intervals
of lesser or no exposure between applications. Assessment of
these exposures requires consideration of their acute nature and
the coincidence of periods of development in the organisms that
can confer greater sensitivity.
Exposures in lakes, reservoirs, ponds, and pools are likely
different from exposures in flowing waters. The relatively long
aqueous persistence of atrazine and the lack of flow in lentic
systems result in longer exposures more consistent with the
type used in chronic laboratory toxicity tests. In these situations, relative inputs will be similar to flowing water but dilutions will be less, and evaporation of water can even increase
residue concentrations to values greater than initially present.
Thus, exposures in pools and ephemeral ponds near agricultural lands where atrazine is used probably represent greatest
exposure scenarios for fish, amphibians, and reptiles. In lakes
and large reservoirs, dilution with uncontaminated waters from
areas where atrazine is not used will likely result in a much
smaller and narrower range of concentrations than that experienced in ponds and pools. However, these exposures will also
be chronic in nature.
Although ponds and wetland areas are important habitats
for amphibians and some reptiles, few studies have reported
atrazine concentrations in these types of surface waters. Two
ponds receiving runoff from cornfields treated with 2.6 or 4.4
kg atrazine/ha (2–4 times greater than currently recommended
label rates) have been studied (Klaasen and Kadoum, 1979).
Concentrations from the field receiving the lesser rate ranged
from 2 to 282 μg/L, and from 1 to 309 μg/L for the greater rate
of application. Only nine sampling times were used in the study,
and raw data were not available for distributional analysis.
Concentrations of pesticides in a number of rural and farm
ponds in southwestern Ontario were investigated between 1975
and 1981 (Frank et al., 1990). In total, 211 water samples from
different ponds were analyzed. Some of these samples were
from ponds known to be contaminated via spills or other accidents. Triazines were detected in 82 of 124 ponds studied; 73
were atrazine. From ponds where runoff and drift were the major
routes of exposure, concentrations ranged from 0.1 to 57 μg/L.
Concentrations ranged from 1.1 to 681 μg/L in the eight ponds
where spills were known to have occurred. Again, raw data were
not available for distributional analysis, and many of the pond
water samples were likely submitted for analysis because of
suspected contamination (Richard Frank, personal communication). This suggests that some samples may have been biased
and not representative of typical exposure.
Concentrations of atrazine have been measured in shallow
depressions in fields treated with 2.24 kg/ha atrazine (Edwards
et al., 1997). Rainfall was simulated with an irrigation device
(between 1 and 32 d after application) and water was collected
in surface depressions at 5 and 10 min intervals for the first
30 min after irrigation. Initial concentrations (t = 0) of atrazine
measured in the no-till field were between 2000 μg/L and 10,000
μg/L. Concentrations decreased with time after irrigation as well
as with time after application. The depressions in these fields
were ephemeral (the water moved into the soil within hours)
and are not representative of suitable habitats for fish or amphibians. These results were similar to those obtained by Baker
and Laflen (1979) in studies on dissipation of pesticides from
shallow depressions such as tractor ruts. Atrazine concentrations
in wheel tracks were 9000 μg/L immediately after application.
However, concentrations declined rapidly and were less than 50
μg/L within 100 min of application, presumably as a result of
percolation into the soil. Risks to amphibians in these systems
should be assessed in the context of their ephemeral nature,
likely use by amphibians, and the importance of these habitats
in relation to the entire landscape.
An extensive characterization of exposure concentrations for
atrazine in surface waters of North America has recently been
undertaken (Giddings et al., 2005). These authors utilized four
tiers of exposure characterization, including several levels of
sophistication of models, as well as an extensive analysis of data
sets of measured concentrations in surface waters. The models
focused on ponds, flowing waters, and reservoirs, while most
of the measured data sets were from small and large flowing
waters, the Great Lakes, and reservoirs. For example, as reported
in Giddings et al. (2005), cumulative probability distributions
of annual maximum concentrations based on results of Tier-4
Monte Carlo modeling of 14,000 pond systems × 36 years of
meteorology in Ohio (504,000 data points) yielded 90th centile
concentrations of less than 10 μg/L. Similarly, the distribution
of 30-d maximum concentrations for 14,000 pond systems ×
36 years of meteorology × 12 months per year (6,048,000 data
points) yielded smaller values (<2 μg/L).
There are relatively few data on measured concentrations of
atrazine in small water bodies such as farm ponds and almost
no data on the temporal trends of these exposures. A study on
temporal trends and exposures in an area of intensive atrazine
use in South Africa revealed concentrations as great as 9 μg/L
and considerable fluctuations in concentration in response to
heavier than normal rainfall (Du Preez et al., 2005a). These
data emphasize the need to characterize field exposures with
due consideration for variation caused by rainfall-driven runoff
events and loss through outflow.
B.
Mechanism of Action
Atrazine is an herbicide developed specifically as a phytotoxin through a mechanism of action unique to plants; thus, the
toxic potency of atrazine is greater in plants than animals. In target plants, atrazine inhibits photosynthesis via competition with
plastoquinone II at its binding site in the process of electron
transport in photosystem II (Devine et al., 1993). This inhibition
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
results in the cessation of carbohydrate synthesis, leading to a
subsequent reduction in the carbon pool and a buildup of CO2
within the plant cell (Giddings et al., 2005). In plants, the binding
of atrazine to the active site is reversible. Removal of the stressor
from the site of action results in recovery (Jensen et al., 1977;
Hoberg, 2007). When atrazine-exposed plants are removed to
uncontaminated media, therefore, levels of photosynthetic activity increase (Brockway et al., 1984; Hamala and Kollig, 1985;
Hoagland et al., 1993). If atrazine has effects in nonplant species,
these effects must be mediated by other mechanisms; however,
if used to control plants, the removal of habitat or food sources
can have indirect effects on other organisms.
Downloaded At: 15:43 28 October 2008
C.
Bioconcentration/Bioaccumulation
Bioconcentration and bioaccumulation of atrazine directly
affect exposures and can result in exposures of organisms via
the food chain. Based on bioconcentration factors (BCFs) and
uptake data in the literature, atrazine bioconcentration and food
chain biomagnification are negligible (Giddings et al., 2005).
Its small octanol–water partition coefficient (log K OW = 2.68
at 25◦ C), relatively large water solubility, and susceptibility to
biological metabolism and rapid elimination combine to produce small BCFs (generally less than 10) in most species tested
(Giddings et al., 2005; Nikkilä et al., 2001). Consequently, exposure via the food chain is of lesser importance than via the
water column.
Several studies on toxicokinetics have been conducted in
fish. Bioconcentration factors for atrazine in fish were summarized in Giddings et al. (2005) and are generally small,
ranging from 12 for bluegill sunfish to <0.27 for brook trout.
Some studies have observed uptake, metabolism, and excretion in fish. Rapid absorption of 14 C-atrazine was observed in
whitefish (Gunkel and Streit, 1980). Absorption of atrazine into
the embryos of zebrafish (Danio rerio) was also rapid and a
BCF of 19 was observed (Wiegand et al., 2000). Metabolism
in zebrafish was to the glutathione conjugate (Wiegand et al.,
2001). The BCFs in various tissues (including gonads) of banded
tilapia (Tilapia sparrmanii) were reported to be as great as 8.2
and appeared to be greatest in tissues with greater amounts
of lipids. BCFs also increased with increasing exposure concentration (Du Preez and van Vuren, 1992). A terminal excretion half-life of whole-body radioactive residues of atrazine
(and metabolites) in juvenile zebrafish of 21 h was reported
(Görge and Nagel, 1990). In summary, atrazine is not greatly
bioconcentrated in fish, is rapidly excreted and/or metabolized,
and does not appear to be significantly accumulated in specific
tissues.
Three bioconcentration values have been reported for amphibians, 0 for the bull frog (Rana catesbeiana) (Klaasen and
Kadoum, 1979), 6 for the leopard frogs (R. pipiens) larvae (Allran and Karasov, 2000), and 1.6 for atrazine and 4.4 for total
body residues including atrazine and all its metabolites in metamorphs of the African clawed frog (Xenopus laevis) (Edginton
and Rouleau, 2005). These are in the same range as fish. In an ar-
727
ticle, Hayes (2004) stated that the body doses of atrazine in frogs
in his studies were 8-fold greater than those in other studies (such
as Carr et al., 2003). This argument was based on the volume of
the exposure tanks and the total amount of atrazine available for
uptake by the frogs. Presumably, the estimate was based on the
assumption that atrazine is greatly bioconcentrated in the frogs,
a phenomenon that is inconsistent with the known large water
solubility of atrazine and its measured small bioconcentration
factors (BCF < 12) in several aquatic species, including frogs.
In fact, contrary to what has been suggested (Hayes, 2004), the
volume of the tanks and the density of larvae would have had
little effect on the overall exposure of the individual larvae to
atrazine (Edginton and Rouleau, 2005).
As observed in fish, atrazine is rapidly metabolized in frogs
(Edginton and Rouleau, 2005). In addition, autoradiography
studies in X. laevis have not shown tissue-specific accumulation
in Nieuwkoop & Faber (NF) stage 66 metamorphs (Nieuwkoop
and Faber, 1967) in organs other than those associated with
metabolism and excretion—the liver and the gut (Edginton and
Rouleau, 2005). This shows that, within the body, large concentrations do not accumulate in gonadal tissues.
III. ACUTE AND CHRONIC TOXICITY
Based on traditional endpoints such as survival and growth,
the toxicity of atrazine to aquatic organisms has been well studied and reported in the literature. These data were reviewed in
Giddings et al. (2005), and this section summarizes some representative values and the more recent data for reptiles, fish, and
amphibians. Because effect concentrations in the literature are
reported in a variety of units, all concentrations in this review
have been converted to micrograms per liter for ease of comparison. The potential mechanisms of acute toxic action of atrazine
in organisms other than plants are not well understood but are
not mediated through the same receptors as in plants. Atrazine
is not very acutely toxic to aquatic animals. For amphibians,
the smallest LC50 was reported to be 410 μg atrazine/L. Based
on other nonlethal endpoints, the most sensitive endpoint reported in the literature was time to development in X. laevis
with a lowest-observed-effects concentration (LOEC) of 100
μg/L. The relevance of developmental and other endpoints are
discussed in greater detail in other sections of this review since
they do not necessarily relate directly to adverse responses at the
population level. In the context of risk assessment, the likelihood
of acute and chronic toxicity values being exceeded under field
conditions is small.
A.
Lethality and Physiological Effects
The mechanism of action of atrazine in fully aquatic organisms is likely related to nonspecific narcosis (Lipnick, 1993),
and, consistent with this mechanism, in chronic exposure studies with fish, the responses were reversible when exposure was
removed (Whale et al., 1994). The toxicity of atrazine to fish
has been presented in detail in two reviews (Solomon et al.,
Downloaded At: 15:43 28 October 2008
728
K. R. SOLOMON ET AL.
1996; Giddings et al., 2005). Acute toxicity values for freshwater fish ranged from a 96-h LC50 of 4300 μg/L for the guppy
(Poecilia reticulata) to a 96-h LC50 of >100,000 μg/L for the
carp (Carassius carassius) (Giddings et al., 2005). Chronic noobservable-effect concentration (NOEC) values for fish ranged
from 65 μg/L for brook trout (Salvelinus fontinalis) to 4300
μg/L for channel catfish (Ictalurus punctatus) (Giddings et al.,
2005). Other than the information in these reviews, no additional
data on lethal effects of atrazine on fish have been reported in
the literature.
A number of studies have been conducted on the effects of
atrazine on the survival of frogs. These report effects only at
large concentrations (Allran and Karasov, 2000, 2001; Detenbeck et al., 1996; Howe et al., 1998; Hayes et al., 2002; and
several others summarized in Giddings et al., 2005). Most of
these studies were toxicity tests with lethality or growth as an
endpoint and none have specifically addressed reproductive or
endocrine endpoints. These results are summarized (Table 1) and
the results of several studies pertinent to frogs are discussed in
more detail later. The overall conclusion is that atrazine concentrations found to have adverse effects in amphibian embryos and
adults were considerably greater than exposure concentrations
currently found or predicted in surface waters in North America (Giddings et al., 2005). It is evident that direct toxicity of
atrazine is probably not a significant factor in recent amphibian
declines (also see Allran and Karasov, 2001).
Studies using the Frog Embryo Teratogenic Assay Xenopus
(FETAX) assay suggest that atrazine has effects on embryonic
and early postembryonic development only at large concentrations. Embryotoxicity and teratogenicity to X. laevis embryos
occur only at concentrations approaching maximum solubility
in water (30,000 μg/L, Morgan et al., 1996). In studies on wildcollected R. pipiens and the American toad (B. americanus)
larvae raised in the laboratory to Gosner stage 29 and 40, the
96-h LC50s for atrazine in the two stages of tadpoles of R. pipiens were found to be 47,500 and 14,500 μg/L, respectively,
while those for B. americanus were 26,500 and 10,700 μg/L,
respectively (Howe et al., 1998). The authors used a commercial
formulation of atrazine (4L) but confirmed exposures through
immunoassay to be within 10% of nominal concentrations. The
authors also reported that alachlor (CAS number 15972-60-8)
and atrazine appeared to act synergistically (more than additive
toxicity) when present in a 50:50 mixture; however, they used
formulated product in their assays and the concentrations where
effects were observed were greater than those that would be
expected in surface waters. The authors also estimated chronic
no-observable-effect concentrations (NOECs) from their acute
data. They suggested that these concentrations (690 to 5100
μg/L) can occur in ponds and pools exposed to runoff from
atrazine-treated fields; however, from the modeling conducted
by Giddings et al. (2005) these concentrations would be extremely rare in farm ponds and have not been observed in field
studies (Du Preez et al., 2005a; McDaniel et al., 2008; Smith,
2007 personal communication).
It has been reported that atrazine exposures of 20 and
200 μg/L had no effect on development rate, percent metamorphosis, time to metamorphosis, percent survival, mass at metamorphosis, or hematocrit in R. pipiens larvae when exposed
from first feeding (Gosner stage 25) through metamorphosis in
a laboratory study (Allran and Karasov, 2000). Measured concentrations in the atrazine exposure solutions were 19.2 ± 0.3
and 192 ± 4.2 μg/L. In the same study, the authors assessed the
effects of nitrate at 5, and 30 mg NO3 (as nitrogen)/L and found
a statistically significant decrease in growth rate of larvae; however, they suggested that this effect of nitrate was probably not
biologically important when compared with natural variation in
the environment. Their conclusion was that concentrations of
atrazine and nitrate commonly found in the environment do not
appear to pose a significant threat to R. pipiens larvae through
direct toxicity. In a subsequent study, the same authors reported
that concentrations of atrazine as large as 20,000 μg/L did not affect hatchability of embryos or 96-h posthatch mortality of larvae
of R. pipiens, the wood frog (R. sylvatica), or B. americanus (Allran and Karasov, 2001). Atrazine also had no effect on swimming
speed in R. pipiens. However, the authors reported that there
was a concentration-dependent increase in deformed larvae of
all three species with increasing atrazine concentration (NOEC
= 2590 μg/L). In adult R. pipiens exposed to atrazine, buccal and thoracic ventilation rates were greater than in untreated
frogs, which indicated respiratory distress. The NOECs for these
responses were 4320 and 12,000 μg/L, respectively. Frogs exposed to the greatest atrazine concentrations (>12,000 μg/L)
stopped eating immediately after treatment began and did not
eat during the 14-d experiment. Disruption of organ development was reported in larvae of X. laevis (NF stage 47) exposed
to concentrations ≥10,000 μg/L (Lenkowski et al., 2008).
A study of the effects of atrazine on the development of the
streamside salamander (Ambystoma barbouri) found effects on
survival, size at metamorphosis, and behavior (Rohr et al., 2004).
Animals from field-collected eggs were exposed to atrazine in
two studies conducted over 2 years in the laboratory. In the first
year, treatments of atrazine were applied by dissolving 80%
pure technical material in dimethyl sulfoxide (DMSO), while in
the second year, solutions of 99% pure technical material from
a different source were applied in acetone. Exposure solutions
were changed every week; however, from the methods (“The
aquaria . . . . contained 9.5 L of constantly aerated, charcoalfiltered, dechlorinated municipal water (pH 8, 158C),” it appeared that charcoal filters were used in the tanks. Although the
authors described analyses of exposure solutions only in the first
year of the study, these data were not reported in the paper. Since
charcoal will absorb atrazine (U.S. EPA, 1989), this may have
affected exposure concentrations; however, the extent of this is
not known and may seriously have compromised the study. In
analyzing the data, the authors combined the data from the 2
years and presented the results as year-standardized means using Z -scores (Rohr et al., 2004). Although this may have been
a statistically valid approach, the facts that different sources of
729
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
TABLE 1
Acute and chronic toxicity data for atrazine in amphibians
Effect measure (μg/L)
Species
NOEC
LOEC
LC50
Downloaded At: 15:43 28 October 2008
Bufo americanus
B. americanus (early
life stage)
B. americanus (later
life stage)
Comment
86 d
>400
A. barbouri larvae
A. barbouri
Pleurodeles walti
Duration
250 μg/L reduced
growth (1)
37 d
No effects on survival
of embryos at 4, 40,
and 400 μg/L but a
greater response to
noise/vibration in
animals exposed to
400 μg/L atrazine
(2).
47 and 56 d Decreased survival in
one year but not
another.
Reduced size at
metamorphosis.
Ambystoma tigrinum
A. barbouri (embryos)
EC50
400
>300
≥4
12 d
300 μg/L did not alter
erythrocytes (1, 2)
(1)
(3)
1,900
>48,000
26,500
8d
96 h
690
10,700
30 d
96 h
Reference
(Larson et al.; 1998)
(Rohr et al.; 2003)
(Rohr et al.; 2004)
(Fernandez et al.;
1993, L’Haridon
et al.; 1993)
(Birge et al.; 1983)
(Howe et al.; 1998)
(Howe et al.; 1998)
30 d
B. americanus
Hyla versicolor
20,000
Hatching
2,590
20 to 200 200 to 2000
H. versicolor
Rana catesbeiana
R. catesbeiana
2000
410
28 d
8d
24 h
Deformities
Based on mass and
length at
metamorphosis
Mortality
4,810 μg/L damaged
DNA (1, 2)
(3)
R. pipiens (early life
stage)
5,100
47,000
96 h
R. pipiens (later life
stage)
650
14,500
30 d
96 h
R. pipiens
200
30 d
138 d
Growth
20,000
10 d
Hatching
R. pipiens, R.
sylvatica
2,590
(Allran and Karasov
2001)
(Diana et al.; 2000)
(Mazanti et al.; 2003)
(Birge et al.; 1983)
(Clements et al.; 1997)
(Howe et al.; 1998)
(Howe et al.; 1998)
(Allran and Karasov
2000)
(Allran and Karasov
2001)
Deformities
(Continued on next page)
730
K. R. SOLOMON ET AL.
TABLE 1
Acute and chronic toxicity data for atrazine in amphibians (Continued)
Effect measure (μg/L)
Species
NOEC
LOEC
LC50
Duration
Comment
R. pipiens adults
4,320
14 d
Buccal ventilation
Xenopus laevis
12,000
12,000
800
14 d
14 d
35 d
Thoracic ventilation
Feeding
Lethality
100
21 d
Uptake of propidium
iodide in nuclei from
exposed larvae.
Time to development
Malformations
reported at low water
hardness (4)
Malformations
reported at high
water hardness
Buffer solution (5)
Natural water
Buffer solution
Natural water
Buffer solution
Natural water
800
X. laevis
Downloaded At: 15:43 28 October 2008
EC50
X. laevis
30
28 d
96 h
100
96 h
33,000
<8,000
96 h
100,000
126,000
3,030
1,100
Reference
(Allran and Karasov
2001)
(Freeman and Rayburn
2005)
(Napier et al.; 1998)
(Morgan et al.; 1996)
Note. (1) Standard toxicity value not reported (i.e., LC50 and EC50 values for acute studies generally 96 h or less in duration and NOEC,
LOEC, MATC/chronic values for chronic studies exceeding 10 d in duration). (2) Nonstandard measurement endpoint not clearly related to
survival, growth, and reproduction was used (i.e., enzyme activity, blood parameters). (3) Toxicity value from a more sensitive life stage was
used. (4) May have been affected by lack of calcium in the medium. (5) A smaller toxicity value from a low water hardness condition was used.
atrazine were used in the two studies and that concentrations
were not reported raise serious questions about the analysis of
the results and whether the reported effects were related to exposure or not. Absolute differences in the percent hatch, survival until d 16, day of hatching, percent larvae in refuge, mean
day of metamorphosis, and snout–vent length at metamorphosis
(author Table 1) were not large and are of questionable biological significance in view of the potential problems in the study
design.
In a follow-up study, newly metamorphosed A. barbouri, exposed to atrazine during larval development as described in the
second year above, were followed for a further 410–433 d after
metamorphosis (Rohr et al., 2006). Animals were transferred
to terreria at metamorphosis and received no further exposures
to atrazine. Results, presented as standardized means and Z scores, indicated statistically significant reductions in survival
(increased mortality) at all exposure concentrations and with an
apparent relationship to exposure concentrations (author Figure
2D). Actual mortality values were not presented so are difficult
to interpret in terms of population-level effects. The interpreta-
tion of the responses by regression of transformed data (log x+
2) was also in error since the control, with zero nominal concentration, cannot be made proportional to the other nominal
concentrations and should only have been used as a reference
point. In addition, this study is based on a previous study (Rohr
et al., 2004) where no measurements of exposure were made
and where the presence of activated charcoal in the system may
have affected exposures. Because of errors in design, these results cannot be interpreted and the conclusions are speculative
at best.
EC50 values based on malformations in the FETAX assay
were reported to occur 30 μg/L (meeting poster Napier et al.,
1998) in low-hardness water and may have been confounded by
lack of calcium in the medium, a necessary element for normal
development in X. laevis (ASTM, 1992). Since detailed descriptions of these data have not been published, these effects cannot
be interpreted.
An assay was developed with X. laevis in which flow cytometric histograms of developing larvae were observed to be representative of developmental stage when native (unfixed) nuclei
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
were analyzed (Freeman and Rayburn, 2004). Using these techniques, studies of the effects of atrazine (technical) on nuclei of
X. laevis exposed for 35 d showed no statistical differences at
measured concentrations ≤600 μg/L but the authors did observe
effects at 800 μg/L (Freeman and Rayburn, 2005). The biological significance of these observations is difficult to determine
in terms of survival or development but the authors did observe
longer development times in larvae exposed for 28 d to 100 μg
atrazine/L.
A study of the effects of exposures to a commercial formulation of atrazine reported on the development of tadpoles of spring
peepers (Pseudacris crucifer), American toad (B. americanus),
green frogs (Rana clamitans), and wood frogs (R. sylvatica)
(Storrs and Kiesecker, 2004). Tadpoles were exposed to concentrations of atrazine prepared from AAtrex Nine-O (85.5%
atrazine) at nominal concentrations of 3, 30, and 100 μg/L for
30 d. Based on concentrations measured in two samples, exposures were 2.84 ± 0.05 μg/L, 25.20 ± 1.82 μg/L, and 64.80
± 2.88 μg/L (mean ± SD). The authors reported “counterintuitive” responses in that survivorship (across species and stage)
in frogs was less (20%) at the nominal concentration of 3 μg/L,
compared to 50% at 30 and 100 μg/L and 15% in the control.
Although endocrine-mediated responses were not specifically
measured in this study, the authors invoked the concept of the
nonmonotonic concentration-response curve reported by others for some endocrine-mediated effects to explain their results.
Given the lack of concordance with observations in other studies,
other factors such as the presence of unknown stressors and/or
confounders are more likely explanations. Preliminary results
of a laboratory study on the effects of exposures of R. pipiens larvae (Gosner stage 25 to 42) to concentrations of 0.1 and
1.8 μg atrazine/L (as AAtrex Liquid 480) reported an exposuredependent reduction in rate of development and size of the ovary
(poster by Fridgen et al., 2005); however, this study may have
been confounded by the use of formulated product containing
surfactants and the results are not yet published. A study on R.
pipiens reported a significant decrease in weight and snout-vent
length in animals exposed to 0.1 μg atrazine/L from Gosner
stage 21 to metamorphosis (Hayes et al., 2006a). However, only
one concentration was tested, and the only control used was
for solvent. In addition, the differences in size and weight were
small (<10%).
Other laboratory studies investigating endocrine responses in
frogs at concentrations ranging from 0.1 to 100 μg atrazine/L
have not reported adverse effects on larval growth, developmental rate, mortality, time to metamorphosis, or size at metamorphosis in female or male frogs (Hayes et al., 2002; Carr et al.,
2003; Coady et al., 2004, 2005; Kloas et al., 2008). All of these
data suggest that, at environmentally realistic exposures, atrazine
does not cause lethality during larval development and/or metamorphosis and that it does not affect size at metamorphosis.
Specific studies on the toxicity and non-reproductive developmental effects of atrazine in reptiles were not found in the
literature. The effects reported in alligators in Lake Apopka,
731
Florida (Guillette et al., 1994, 1996) have not been linked to
atrazine exposures (Crain et al., 1997, 1999).
IV.
EXTERNAL DEVELOPMENTAL ABNORMALITIES
A number of studies have investigated the possible causes
of deformities in frogs and there is no evidence that synthetic
chemicals were directly responsible. While a number of other
possible causes have been suggested, parasitism has been identified as a major factor (Blaustein and Johnson, 2003; Sessions
et al., 1999; Johnson et al., 1999). Some studies on the effects
of pesticides on limb deformities in frogs have been reported in
the literature but there is a need to clarify the contribution of
changes in habitat structure and function, parasite density, and
the frequency of dysmorphogenesis in amphibians under field
conditions.
A.
Trematode Infections
Cercariae of trematode parasites induced severe limb abnormalities in the Pacific tree frog (Hyla regilla) in laboratory studies (Johnson et al., 1999). These abnormalities were reported
to closely match those observed in the field. In addition, increased parasite density caused an increase in the frequency of
abnormalities with an associated decline in tadpole survivorship
(Johnson et al., 1999). More recently, a study on trematode infections in R. sylvatica was conducted to assess the potential
effects of pesticides, including atrazine on the induction of limb
deformities caused by trematode infection (Kiesecker, 2002).
The author concluded that some pesticides, including atrazine,
could increase the susceptibility of R. sylvatica to parasite infection, which could lead to a greater incidence of limb deformities.
However, several key deficiencies in the experimental design of
this study make its interpretation difficult.
In the field component of the study, tadpoles of R. sylvatica
were exposed to the cercaria (the mobile infective stage) of the
trematode parasite Ribeiroia sp. in mesh enclosures placed in
ponds close to agriculture and ponds remote from agriculture.
More frogs from the agricultural ponds developed limb deformities than did frogs from nonagricultural sites. Snail and parasite
densities in the ponds were reported in the paper but no measures of cercarial density inside the enclosures were reported.
Thus, it is not certain what the exposures to cercaria inside the
enclosures were and if they were the same between sites. Unfortunately, the infection rates in the field-exposed frogs were not
reported. Since not all meta-cercarial cysts result in limb abnormalities, the number of cysts may have been a better index of
interactions between exposure to agricultural runoff and infectivity. Although more frogs exposed to cercaria at the agricultural sites developed limb deformities, the lack of information
on number of meta-cercarial cysts makes it difficult to know
whether, in general, this was the result of greater infection rates
or more infections in the limb buds of tadpoles at these sites.
Water samples from the ponds were analyzed for pesticide residues; however, the exact identity of the pesticides
was not stated except that “both organochlorine pesticides and
Downloaded At: 15:43 28 October 2008
732
K. R. SOLOMON ET AL.
organophosphorus compounds (e.g., atrazine and malathion)
were detected” (Kiesecker, 2002). The detection limits of the
method, and measured concentrations were not reported. It also
is uncertain what changes in concentration occurred over the
course of the exposure period. Thus, it is impossible to use the
results of these studies to make any conclusions about the potential effects of atrazine on immune function or the effect on the
rate of infection with trematodes. Analyses for other potential
confounding factors such as other agricultural chemicals, including other classes of insecticides, herbicides, and fungicides,
heavy metals, nutrients such as nitrate/nitrite and phosphate, or
ultraviolet (UV) light that are known to affect frog development
(Ankley et al., 2002; Peterson et al., 2002; Diamond et al., 2002)
were not reported. The omission of these particular measurements is unfortunate since the author cites each of these factors
as possible co-factors that are related to population declines of
amphibians (Kiesecker, 2002). The temperatures of the ponds
were not reported. Since temperature has an effect on amphibian
development, this also may have confounded the comparisons
between ponds.
The ability to make conclusions about the possible interactions between atrazine and trematode infections based on the laboratory component of this study (Kiesecker, 2002) was also limited. The laboratory study used three model pesticides, atrazine,
malathion, and es-fenvalerate. The nominal exposure concentrations for atrazine were 3 and 30 μg/L. Tadpoles (presumably
Gosner stage 25) were exposed to the pesticides for 4 weeks.
Exposure solutions were replaced every 2 d and tadpoles were
fed ad libitum. Measurements of atrazine concentrations in the
exposure solutions were not made. After 4 weeks, the tadpoles
were individually exposed to 50 cercaria in a test chamber for
4 h. Tadpoles were then placed in uncontaminated water for
1 week, after which a blood sample was taken to enumerate
eosinophils (the type of white blood cell responsible for cellmediated immune response) and the frogs were then prepared
for determination of the number of cercaria that had successfully
encysted to form meta-cercaria.
The author reported that meta-cercarial infection rates for
two trematodes (Telorchis sp. and Riberiroia sp.) in the atrazineexposed frogs were greater than in the solvent controls and that
number of eosinophils were less in the exposed frogs. The implications of these observations on the possible immunotoxic
effects of atrazine are discussed later, in the section on immunotoxicology. The number of eosinophils appeared to be inversely
correlated with the proportion of cercaria that formed metacercarial cysts and also with pesticide exposure. But, since only
two concentrations were tested, a concentration-response relationship could not be developed, thus limiting the interpretation
of the results.
More recent reports on the effects of atrazine (AAtrex, 40.8%
active atrazine) on the infectivity of cercaria in frogs have reported negative or no effects. In a study of the effects of atrazine
on the infectivity of cercaria of the trematode parasite Echinostoma trivolvis in R. clamitans, it was reported that 200 μg
atrazine/L caused increased mortality of cercaria and that concentrations of 20 or 200 μg/L decreased infectivity in tadpoles of
R. clamitans (Koprivnikar et al., 2006). The exposures were not
reported to have adverse effects on the tadpoles (Gosner stage
25/26; 20 tadpoles per treatment). Toxicity values for cercaria of
several other species of trematode were also reported and 12-h
LC50s ranged from <20 μg/L for Haematoloechus sp., 92 μg/L
for Alaria sp., 110 μg/L for E. trivolvis, to >850 μg/L for Megalodiscus sp. (Koprivnikar et al., 2006). Unfortunately, concentrations in the exposure solutions were not verified by analysis.
In a subsequent paper by the same authors on the infectivity
of cercaria of E. trivolvis in tadpoles of R. sylvatica, different responses from those in R. clamitans were reported (Koprivnikar
et al., 2007). Using the same formulated product also without
verification of exposure concentration, there was no effect on intensity of parasitism when both cercaria and Gosner stage 28/29
tadpoles were exposed to 3 or 30 μg/L atrazine. However, if
only the tadpoles were exposed to 30 μg/L (but not 3 μg/L),
intensity of infection increased by a factor of approximately
2. Given the differences in response reported in these studies
and the relatively small number of tadpoles per replicate (21
per treatment), these results are difficult to interpret but neither
study showed increased parasitism when both tadpoles and cercaria were exposed to atrazine, which is what would occur under
field conditions. The responses observed may have been due to
formulants in the commercial product used (AAtrex, 40.8% active ingredient) and thus may only be representative of direct
contamination of frog habitats, such as from spray drift. Under
most conditions of exposure in the field, atrazine and its formulants would be expected to not co-occur. Regardless of the
reasons, the results of the study do not appear to be internally
consistent or reproducible and are not ecologically relevant.
Several authors have pointed out that anthropogenic activities
can result in changes in the physical and biological properties
of water bodies. Some of these changes, such as eutrophication,
can increase the prevalence of both the snail hosts and trematode
parasites that they carry (Johnson et al., 2002; Kiesecker, 2002;
Blaustein and Johnson, 2003). Eutrophication itself has not been
shown to affect survival of tadpoles of R. sylvatica infected with
cercaria of E. trivolvis (Belden, 2006), but infectivity of cercaria
of the trematode parasite Ribeiroia ondatrae in tadpoles of the
green frog (R. clamitans) was increased by eutrophication. The
mechanism by which eutrophication promoted parasitism was
via increasing the density of infected snail hosts and enhancing per-snail production of cercaria (Johnson et al., 2007). This
potentially confounding variable was not addressed in the field
study by Kiesecker (2002).
B.
Direct Effects of Atrazine on Limb Deformities
In another study on limb deformities, exposures to a mixture
of atrazine and carbaryl were included as treatments in a study
of R. pipiens; however, no responses were observed (Bridges et
al., 2004). A mixture of 5 μg/L atrazine and 5 μg/L carbaryl
did not affect the incidence of embryonic deformities, hatching
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
success, survival to metamorphosis, tadpole mass, bony triangles, skin webbing, or multiple deformities. Although only one
concentration of atrazine (in a mixture) was tested with this
species, the results are consistent with the lack of effect seen in
the FETAX embryonic frog development assay at greater concentrations (1100 to 26,000 μg/L) (Morgan et al., 1996). In
contrast, extracts of water from a reference pond in Minnesota
were associated with a small incidence (5%) of deformities, indicating that other environmental contaminants or factors are
potentially responsible for the observed abnormalities noted under field conditions (Bridges et al., 2004).
Overall, there is no credible evidence to suggest that atrazine
directly causes or contributes to limb deformities caused by parasites in frogs. However, it has been shown that activities that
increase populations of the snail vectors of trematode parasites
increase parasite pressure and increase the frequency of deformities in frogs by increasing populations of snail intermediate
hosts and increasing production of cercaria. Concentrations of
atrazine in ponds from runoff in agricultural areas are unlikely
to exceed thresholds for effects in macrophytes (Giddings et al.,
2005) and thus cause eutrophication. However, deliberate (offlabel) use to control weeds in ponds could affect eutrophication.
This would also occur if other herbicides were used and is not a
specific response to atrazine. Changes in habitat have the potential to induce dysmorphogenesis in amphibians. Loss of habitat
has impacted amphibians for decades. In addition, UV-B irradiation, emerging diseases, the introduction of alien species, direct
exploitation, changes in land use, and climate change (Beebee
and Griffiths, 2005) can potentially interact and contribute to
abnormalities that are observed under field conditions. Agricultural activity, including the use of pesticides, can contribute to
eutrophication of surface waters; however, this is not specific to
particular pesticides and would only occur at concentrations that
directly affect aquatic plants. The likelihood of this occurring is
judged to be small unless the products are specifically used for
weed control in ponds.
SEXUAL DIFFERENTIATION AND DEVELOPMENT
Much of the current research investigating the risks posed to
wildlife by exposure to atrazine has focused on the potential for
adverse effects on reproduction and sexual development. This
concern is justified given the paramount role of reproduction in
population sustainability and reports that atrazine, at relatively
high concentrations, negatively affects reproduction in the laboratory rodents and may disrupt reproductive endocrine function
in mammals (Cooper et al., 2000, 2007; IPCS, 2002). Investigation of the reproductive responses is a broad topic encompassing sexual differentiation, gonadal development, and studies
of mechanisms of reproductive dysfunction, through to actions
on sex-dependent processes. Investigations have at times been
hampered by a limited knowledge of reproductive and developmental processes in wildlife. While this is most evident in the
case of species in the wild, it has become apparent that there are
significant gaps in our knowledge of these processes in species
733
such as X. laevis despite it having being used a laboratory model
species for over 70 years (OECD, 2004).
A.
The Use of Xenopus laevis as a Model
for Endocrine Responses
While X. laevis is a popular and well-studied model for examining the effects of chemicals on sexual differentiation, recent
tests have shown that husbandry conditions may well have a
marked effect on the sensitivity of amphibian species to chemicals (Kloas et al., 2008). It is also used to study several endocrine
responses, such as the control of metamorphosis by thyroxine.
In this regard, it has been suggested as the test organism for
assessing sensitivity to chemicals that interact with the thyroid
axis (OECD, 2004). While some of its popularity stems from its
ease of breeding and maintenance in the laboratory, that fact that
it is exclusively aquatic makes exposures more easily controlled
than other amphibians that are terrestrial in the adult stage. Ease
of use does not mean that the species is very sensitive—a necessary attribute for a sentinel species—and this question has been
raised with respect to responses to steroid hormones.
Compared to other amphibians, X. laevis is relatively sensitive to estradiol. Using the least concentration of water-borne
exposures to estradiol that caused 100% feminization of developing frog larvae (Hayes, 1998) and urodeles (Wallace et al.,
1999) that had been reported in the literature, a species sensitivity
distribution was constructed using published methods (Solomon
and Takacs, 2002). From these data (Fig. 4) it is apparent that,
although not the most sensitive species to estradiol, X. laevis is
close to the 20th centile for sensitivity in static exposures and
V.
FIG. 4. Distributional analysis of sensitivity of sexual differentiation in larval frogs (•) and urodeles (◦) exposed to estradiol
in static water-borne test systems. Data for urodeles were excluded from the regression. Data from Hayes (1998), Wallace
et al. (1999), Kloas et al. (2008), and Lutz et al. (2008).
734
K. R. SOLOMON ET AL.
is only two-fold less sensitive than the most sensitive species,
R. sylvatica. Sensitivity of X. laevis in flow-through exposures
was greater (Kloas et al., 2008), probably because of the greater
body dose resulting from this exposure method, but this is the
only species to be tested in this way. This observation is important since it points out that flow-through protocols should be
used when assessing possible effects of hydrophobic endocrineactive substances, such as estradiol. Thus, we conclude that X.
laevis is a useful and sensitive model for affects of estradiol and
that ease of exposure via the matrix makes it a useful sentinel
animal for assessment of steroid and thyroid-mediated effects,
provided that this is done under flow-through conditions.
Downloaded At: 15:43 28 October 2008
B.
Effects on Sex Ratio
Effects of atrazine exposures on determination of gender have
not been studied in fish or terrestrial wildlife species; however,
atrazine does not appear to affect sex ratios in mammals, reptiles,
or frogs. No effects on sex ratios in mammals were reported in
multigeneration studies (U.S. EPA, 2000, Gammon et al., 2005).
It has been reported that atrazine exposures did not result in any
adverse effects on developing alligators (Alligator mississippiensis) either in the laboratory or under field conditions (Crain et
al., 1997, 1999). In fact, the authors of those studies concluded
that atrazine did not affect sexual development in A. mississippiensis hatchlings and was not responsible for effects on sexual
development observed in wild A. mississippiensis. Treatment of
A. mississippiensis eggs with atrazine at concentrations as great
as 14,000 μg/kg applied to the surface of the egg in ethanol did
not alter the sex ratio of hatchlings, nor did it alter the height
of epithelial cells in the Müllerian duct, affect degeneration of
the ovarian medulla, or the diameter of the sex-cord (Crain et
al., 1999). All of these parameters responded to estradiol exposure as a positive control. A study on the effect of several
pesticides on the eggs of the broad-snouted caiman (Caiman
latirostris) reported no effects on sex ratio in eggs treated topically with 15 μg/egg dissolved in 50 μl ethanol (Beldomenico et
al., 2007). This is equivalent to ∼178 μg/kg, given an egg weight
of ∼84 g as has been reported in the literature (Groombridge,
1982). Treatment with estradiol (105 μg/egg; ∼1250 μg/kg)
resulted in 100% females. Treatment of eggs with atrazine or
several other pesticides resulted in greater loss of weight during incubation when compared to the vehicle control (11.3%
for atrazine vs. 8.6%) and decreased weight of hatchling (as
percent of initial weight 65.1% vs. 67%). Given the fact that
weights were determined to the nearest gram, the small differences were close to the limits of detection and raise questions
about possible measurement errors. In addition, the physiological significance of these differences in weight was not tested in
the study (Beldomenico et al., 2007). Other studies on eggs of
red-eared slider turtle (Pseudemys elegans) and A. mississippiensis showed no responses in terms of sex ratio to nominal aqueous exposures of as great as 500 μg atrazine/L used to drench
the eggs (Gross, 1999a, 1999b). Although topical treatment of
eggs has been shown to give inconsistent penetration into the
egg (Muller et al., 2007a, 2007b), this is a more realistic route
of environmental exposure for a water soluble substance, such
as atrazine. The lack of response confirms the lack of effect of
atrazine in the development and sexual differentiation of reptiles
such as turtles and alligators under realistic exposure conditions.
A study by Suzawa and Ingraham (2008) reported that zebrafish (Danio rerio) exposed to atrazine for 60 d at concentrations of 0.1, 1, or 10 μM (21.7, 217, 2167 μg/L) beginning on
posthatch d 17 displayed a dose-dependent increase in the percentage of females and a decrease in the percentage of males.
The data for sex ratios are unclear in author Figure 2A (Suzawa
and Ingraham, 2008). As the fish were not sexually differentiated
at the beginning of the study, the authors could not know how
many males and females there were in each treatment. However,
given a sex ratio of 50:50, 7 or 8 of the 15 fish used per concentration would be expected to be females. Author Figure 2A
shows that females increased in proportion by 400%, a 4-fold
increase relative to the control. If there were 7 or 8 females at the
start of the exposure, a 4-fold increase would mean there were
28–32 females in each beaker by the end of the exposure. This
is impossible and is inconsistent with the decrease in number
of males (50% ≈ 4 fish) reported at 2167 μg/L, which could
only have produced 4 sex-changed fish. These data are uninterpretable and, in addition, the experimental design was flawed as
the treatment unit was the tank and there was only one replicate.
Few effects on sex ratio have been reported in frogs. Laboratory studies on R. clamitans exposed to atrazine at concentrations
as great as 25 μg/L (Coady et al., 2004) and X. laevis at concentrations as great as 100 μg/L did not reveal any effects on
sex ratio (Coady et al., 2005; Kloas et al., 2008). Studies in wild
populations of X. laevis in areas of corn production and atrazine
use and in reference areas did not reveal effects on sex ratios of
adults or metamorphs (Du Preez et al., 2005b); sex ratios were
near 50:50. Similar results were observed in a field study on R.
catesbeiana in Iowa (Smith, 2007 personal communication). No
effects on sex ratio were observed in NF stage 66 metamorphs in
studies on X. laevis larvae exposed to atrazine at concentrations
as great as 30 μg/L in outdoor microcosms, (Jooste et al., 2005).
No transgenerational effects on sex ratio were observed in F2 larvae of F1 X. laevis exposed to concentrations of atrazine as great
as 25 μg/L from 96 h of age to breeding (Du Preez et al., 2008b).
One study reported an effect on sex ratio in X. laevis at concentrations with significantly more females at 10 and 100 μg/L but
not at 0.1 and 1 μg/L (Oka et al., 2008). Because of high mortality at100 μg/L, the analysis was based on only one replicate of
30 animals. The reason for this is not clear. Aromatase mRNA
was not induced at any exposure concentration, nor was there
any evidence of estrogenicity based a hepatic vitellogenin assay.
Sex ratio in all ZZ males was unaffected at exposures of 0.1 and 1
μg atrazine/L; however, greater concentrations were not tested.
C.
Effects on Sexual Development
If atrazine were to affect maturation of the gonads, this
would be manifested in terms of a quantitative difference in the
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
FIG. 5. Proportion of different sex cell types in the testis of C.
auratus exposed to 100 and 1000 μg/L atrazine at 0, 7, 14 and
21 d. Redrawn from data of Spanó et al. (2004).
Downloaded At: 15:43 28 October 2008
distribution of cell types within the testis and ovary, through effects on sperm production, or fecundity. There is no evidence to
support the contention that atrazine affects sexual development
in fish or amphibians.
Testicular Cell Types and Gonadal Development
In goldfish exposed to relatively large concentrations of
atrazine (100 and 1000 μg/L) for 21 d, there were no differences
in the relative size, number of sperm, or the relative proportions
of each cell type cell type (Fig. 5) (Spanó et al., 2004). A number
of measures of testicular development were made in a study of
fathead minnows (Pimephales promelas) exposed to measured
concentrations of 224 and 25 μg atrazine/L for 21 d (U.S. EPA,
2005). Males in the control group and the two atrazine-treated
groups had well-developed testes and there was no difference in
the mean stage of development between treatments. There were
slight differences in the proportion of testis cells in stage 2A (primary spermatogonia) between treatments, varying from 0% in
controls to 0.3 and 0.5% of the testis cells in the 25 and 224 μg
atrazine/L groups, respectively. Mean seminiferous tubule diameter in males from the greatest atrazine exposure group was
smaller than those of males in the control group (123.8 versus
153.8 μm). However, the biological relevance of these histological changes has been questioned considering that no other
abnormalities were reported (U.S. EPA, 2005). The lack of effect
on vitellogenin, which is a sensitive indicator of exposure to estrogen, suggests that atrazine was neither directly nor indirectly
estrogenic. There were no differences in vitellogenin concentrations between fish exposed to either atrazine concentration or
that of control fish for either males or females (U.S. EPA, 2005).
Overall, this study suggests that atrazine at exposures as great as
224 μg/L had no significant effects on important reproductive
endpoints in P. promelas. It is important to place these results
in the context that this was a robust and well-characterized bioassay procedure that has been used to identify the adverse effects
of a variety of endocrine active compounds.
735
No differences in the absolute or relative numbers of testicular
cell types were observed in X. laevis from corn and non-corngrowing areas in South Africa where atrazine concentrations
ranged from 0 to 9 μg/L (Smith et al., 2005). This included an
assessment of the fractional volume of the testis occupied by
spermatagonia, spermatocytes, sperm, connective tissues, and
blood cells. All frogs appeared normal when evaluated histologically and fully developed sperm were observed in all frogs
(Smith et al., 2005). Similar results were obtained in a field
study on bull frog (Rana catesbeiana) in agricultural and nonagricultural areas in Iowa (Smith, 2007 personal communication). Atrazine concentrations ranged from less than the limit
of detection (LOD) to 40 μg/L in the agricultural area but were
consistently less than the LOD in the nonagricultural areas. In
a study on R. pipiens from areas associated with row crop agriculture (and where atrazine and other pesticides were found),
no differences were detected in the gonadosomatic indices or
stage of spermatogenesis between frogs from agricultural and
nonagricultural regions ( p > .05) (McDaniel et al., 2008). In
a study of R. pipiens metamorphs exposed to 15 μg atrazine/L
throughout metamorphosis, Orton et al. (2006) reported no difference in the total number of spermatogenic cells. However,
they did report an increase in the percentage of testicular cells in
the latter stages of spermatogenesis relative to controls (38% vs.
20% in the controls). The biological significance of this effect is
unclear since there were no other effects of atrazine on testicular
development, and only one atrazine concentration was tested.
A series of laboratory studies on the effects of atrazine on
gonadal and kidney development in male and female X. laevis tadpoles were reported in a thesis and two published papers
(Tavera-Mendoza 2001; Tavera-Mendoza et al., 2002a, 2002b).
After quantitatively assessing testicular volume, primary spermatogonial cell nests, and nursing cells from histological sections, they reported a 57% reduction in testicular volume in tadpoles exposed to atrazine at 18 μg/L relative to controls (TaveraMendoza et al., 2002a). Similarly, primary spermatogonial cell
nests were reported to be 70% fewer than in unexposed individuals, while the nursing cells that provide the nutritive support
for development of germ cells were reported to be 74% fewer
(Tavera-Mendoza et al., 2002a). Unfortunately, the data in the
published papers and in the thesis are inconsistent. For example, the numbers of animals used in the study were not clearly
reported and are different in the description of the methods and
the figure captions. In the Tavera-Mendoza thesis (2001), it is
stated that six tanks were used with 15 tadpoles each; however, in
the published paper (Tavera-Mendoza et al., 2002a), it is stated
that two control tanks and two exposed tanks were used, each
with 24 tadpoles. In another experiment, more tadpoles were
sampled than were initially stated to be present in the tanks.
Responses at greater concentrations may have been confounded
by general necrosis observed in several tissues of exposed tadpoles (Tavera-Mendoza, 2001). The cause of the necrosis was not
clear. The authors reported that testicular volume was reduced in
atrazine-exposed tadpoles but they did not make measurements
736
K. R. SOLOMON ET AL.
Downloaded At: 15:43 28 October 2008
of testicular volume in a subsample of animals before exposure to
atrazine. Furthermore, the responses reported in the testes were
inconsistent with those reported elsewhere (Hayes et al., 2001;
Coady et al., 2005; Carr et al., 2003). Because of incomplete
descriptions in the methods and inconsistencies in the data, the
results in these papers (Tavera-Mendoza et al., 2002a, 2002b)
and the thesis (Tavera-Mendoza, 2001) are essentially uninterpretable and cannot be cited as supporting any adverse effects
of atrazine on gonadal development in frogs.
Intersex Testis and Testicular Ovarian Follicles
It has been known for many yrs that exposure of developing X. laevis to estradiol causes sex reversal in genetic males,
with males developing ovaries instead of testes. In some cases,
estradiol-exposed animals can exhibit gonads that are intermediate in appearance between testes and ovaries (Chang and
Witschi, 1956), a condition that has been variously referred to as
hermaphroditic or intersex gonads (see Hecker et al., 2006, for
a review of the terminology). Given that gonadal differentiation
in X. laevis has been reported to be sensitive to several types
of aquatic contaminants ranging from PCBs (Qin et al., 2003)
to bisphenol A (Levy et al., 2004), and alkylphenols (Bögi et
al., 2003), it is logical that several studies have focused on this
model to examine the potential effects of atrazine on gonadal
development.
Laboratory Studies. The first study to report on this response stated that atrazine concentrations ranging from 0.1 to
200 μg atrazine/L produced gonadal abnormalities in developing X. laevis exposed from hatching until completion of metamorphosis (Hayes et al., 2002). These authors reported that
16–20% of the exposed animals had multiple gonads or were
“hermaphrodites” with multiple testes and ovaries. The incidence of gonadal abnormalities at each test concentration was
not reported, making it impossible to determine if these effects
were concentration-related, although the authors claim that similar gonadal abnormalities were never observed in over 10,000
control animals examined over a 6-year period in their laboratory
(Hayes, 2004). This claim is also not clear since it was later stated
that, “By definition, ‘gonadal malformations” were defined initially as morphologies observed in atrazine-exposed larvae but
not in controls” (Hayes et al., 2006b, p. 135). This begs the question as to how gonadal malformations that did occur in control
animals were dealt with in terms of data reporting and statistical analysis. In another paper, Hayes (2004) plotted the data
from the 2002 study as a function of atrazine test concentration.
Based on this report, Hayes observed hermaphroditism and/or
“single sex polygonadism” in as many as 15% of the animals
exposed to atrazine. There was no apparent association between
atrazine concentration and the incidence of hermaphroditism or
single-sex polygonadism.
A concentration-related increase in the incidence of segmented and anomalous gonads was observed in X. laevis exposed
to atrazine from <24 h after fertilization for 70 d (Carr et al.,
2003). The incidence of abnormal gonads based on gross morphology and categorized as intersex individuals differed significantly from controls only in animals exposed to 25 μg atrazine/L
(4.7% relative to 0.6% in controls) (Figure 3 in Carr et al., 2003).
In their study, Carr et al. (2003) defined intersex gonads as those
gonads that could not be unambiguously identified as testis or
ovary because of shared or undifferentiated traits in size, shape,
and physical appearance. Phenotype-specific (male versus female) characteristics of differentiated gonads in developing X.
laevis include size, shape, and pigmentation differences (Carr
et al., 2003), as well as more obvious differences in germ-cell
type (oogonia, spermatogonia) and presence/absence of cortex
and medulla in differentiated gonads. While gonad size, shape,
and pigmentation can be qualitatively assessed based on gross
appearance, germ cells, and the organization of cortex, medulla
and ovarian cavity can only be observed at the histological level.
Subsequent histological evaluation of the gonads from the 25μg/L group, as described in Carr et al. (2003) and more recently
by independent analysis of the same slides by pathologists from
EPL Laboratories (Wolf, 2007), confirmed (as originally reported by Carr et al., 2003) that gonads identified as intersex
based upon their outward appearance showed no evidence of
mixed ovarian and testicular tissue. It is important to note that
the definition of intersex as it appeared in Carr et al. (2003) was
based on the ambiguous physical appearance of gonads at the
gross morphological level and differs from more recent definitions of the term, which are based solely on the simultaneous
presence of male and female germ cells within the same gonadal
tissue. While many papers have cited the Carr et al. (2003) paper as providing evidence that atrazine causes a mixture of male
and female germ cells within differentiating gonadal tissue in X.
laevis, this was never reported by Carr et al. (2003) nor in the
more recent analysis by Wolf (2007). In fact, to our knowledge,
the only circumstances in which mixed ovarian and testicular
tissue have been consistently observed in developing X. laevis
is when the animals are exposed to estradiol (Carr et al., 2003;
Kloas et al., 2008; Hu et al., 2008) or other estrogenic pollutants
such as nonyl-phenol (Mosconi et al., 2002) or polychlorinated
biphenyls (PCBs) (Qin et al., 2003). The biological significance
of the effects reported by Carr et al. (2003) remains unclear
since mixed ovarian/testicular tissue was not observed at the
histological level.
In another study, Coady et al. (2005) expanded on these observations by examining gonadal development in X. laevis exposed
to atrazine or sex steroids (dihydrotestosterone or estradiol)
throughout larval development and continuing for 2–3 months
postmetamorphosis. These authors reported a small incidence
of rudimentary hermaphroditism (van Tienhoven, 1983) based
on gross gonadal morphology in animals exposed to sex steroid
from completion of metamorphosis and 2–3 months postmetamorphosis (Coady et al., 2005). Detailed histological analysis
revealed that 8% of the control animals possessed testicular ovarian follicles (TOFs), defined as oocytes with an intact nucleus,
nucleoli, and a surrounding squamous epithelial layer. Testicular
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
ovarian follicles have been described as testicular oocytes (TOs)
in a number of papers, but the more correct term—testicular
ovarian follicles—is used in this review. There was no apparent
relationship between atrazine concentration and the incidence
of TOFs, although they were more common in animals treated
with estradiol (32%) or ethanol (20%) relative to the FETAX
controls or atrazine treatment groups (Coady et al., 2005). The
TOFs reported by Coady et al. (2005) were generally small,
nonvitellogenic, and represented a very small percentage of the
total tissue. The observation of TOFs by Coady et al. (2005) is
interesting, given previous reports of this phenomenon by Gallien (1974) and recent data from microcosm studies in which
male X. laevis exposed to atrazine or control medium during
larval development exhibited TOFs regardless of whether they
were exposed to atrazine or not (Jooste et al., 2005). In this latter
study, the incidence of TOFs declined in 10-month-old animals
from all treatment groups and were virtually absent in 2-yearold animals (See Fig. 6). The maximum number of TOFs per
testis furthermore reduced from 58 at NF stage 66 to 5 after a
10-month period to 1 follicle in 1 of 4 frogs at 2 years of age
(Du Preez et al., 2008b). In addition, there was no indication
of any transgenerational response to atrazine exposures in adult
F1 frogs of a number of parameters including of frequency of
F2 metamorphs with TOFs and number of oocytes per frog (Du
Preez et al., 2008b).
In recently reported studies conducted under the requirements
and guidance of the U.S. EPA and recommendations of its 2003
Scientific Advisory Panel (U.S. EPA, 2003a), X. laevis were
exposed to concentrations of atrazine ranging 4 orders of magnitude from 0.01 μg/L to 100 μg/L from NF stage 47–48 through
to metamorphosis (NF stage 66). Two similar studies were conducted under Good Laboratory Practices (GLP) with full quality
FIG. 6. Mean number of TOFs found per specimen in the reference and atrazine-exposed NF stage 66, 10-mo grow-out juvenile male frogs and 2-year-old adult frogs. ∗ A single TOF was
observed in 1 of 4 adult frogs in the 25 μg atrazine/L treated
group. Redrawn from data of Jooste et al. (2005) and Du Preez
et al. (2008b).
737
assurance/quality control (QA/QC). No effects of atrazine on
metamorphosis, sexual differentiation, frequency of males, or
mixed sex were observed. Testicular ovarian follicles were not
observed in negative controls or in larvae exposed to atrazine but
one TOF was observed in the positive control larvae exposed to
17β estradiol at 0.2 μg/L (Kloas et al., 2008). No TOFs or other
indications of intersex were observed in a study of X. laevis exposed to concentrations ranging from 0.1 to 100 μg atrazine/L
(Oka et al., 2008). Exposures were static with renewal every 3 d.
Analysis by GC-MS showed exposure concentrations from 100
to 200% of nominal, a relatively large range.
Only a few studies have examined effects of atrazine on gonadal development in frogs native to the United States, where
most atrazine is used. In a laboratory study, Hayes et al. (2003)
exposed R. pipiens to 0.1 μg atrazine/L or 25 μg atrazine/L
throughout larval development. They reported the presence of
TOFs in males from the 0.1- and 25-μg atrazine/L treatment
groups but not in the controls. This observation is confusing
because it is also reported that two control animals with TOFs
were observed (Hayes et al., 2003, p. 570).
Two other studies examining the effects of atrazine on ranid
frogs have reported the presence of TOFs in males irrespective of treatment and no effect of atrazine on the incidence of
hermaphroditism. Based on laboratory exposures of R. clamitans, Coady et al. (2004) reported TOFs in 12% of the control
frogs and 0% of the frogs exposed to 10 or 25 μg atrazine/L.
There were no atrazine-related effects on incidence of intersex or
phenotypic sex ratio (based upon gonadal appearance), although
exposure to dihydrotestosterone resulted in a shift toward more
phenotypic males (Coady et al., 2004). In a similar study, Orton
et al. (2006) tested only one concentration of atrazine (10 μg
atrazine/L nominal) and found no significant difference in the
incidence of males with TOFs between controls (4 of 40) and
atrazine treated (2 of 68) R. pipiens exposed from shortly after
hatching through metamorphosis.
Field Studies. In general, field studies have not observed a
link between exposure to atrazine and the presence of TOFs in
frogs. In a field study conducted on R. pipiens in 2002, Hayes et
al. (2003) reported that only frogs collected from sites with measurable atrazine concentrations exhibited TOFs that were similar
to those observed in the laboratory after exposure to atrazine.
This was interpreted as suggesting that hermaphroditism or “sex
reversal” never occurs in the absence of atrazine and that the
phenomenon is solely the result of exposure to atrazine. As has
been pointed out (Gammon et al., 2005), there was no consistent
response of gonadal abnormalities to concentration of atrazine.
This does not support a causal relationship. In addition, the study
design (Hayes et al., 2003) was flawed because concentrations of
atrazine were only measured at the time of collection of sexually
differentiated frogs and there is no knowledge of their exposure
when they were undergoing sexual differentiation prior to the
time of collection. Since concentrations were not known during critical developmental stages, any conclusions reported by
Hayes et al. (2003) are speculative. In addition, for some sites,
Downloaded At: 15:43 28 October 2008
738
K. R. SOLOMON ET AL.
there was no likely source of atrazine identified. In one site in
Wyoming, atrazine was detected only in 2001, at which time
92% of the male R. pipiens had one or more TOFs. These results
are inconsistent and cannot be interpreted as atrazine having a
causative relationship with TOFs. As far as we are aware, this
phenomenon has not been observed again. It is also instructive
to note that there seems to have been a robust population of frogs
in the subsequent year, which indicates that the occurrence of
TOFs the previous year did not have a measurable impact on the
population in the following year.
A field study on frogs in Ontario, Canada reported the incidence of TOFs in R. pipiens from areas associated with row-crop
agriculture as well as in reference areas (McDaniel et al., 2008).
Mean proportion of R. pipiens with one or more TOFs were
25 and 42% in the two agricultural sites where atrazine was
detected (0.068 to 3.13 μg atrazine/L), 25% in an agricultural
reference site (0.044 to 0.39 μg atrazine/L), to 7% in the nonagricultural site (atrazine 0.015 to 0.090 μg atrazine/L). Residues of
pesticides were measured only during the growing season from
August 2003 to July 2005. There was no correlation between
the proportion of R. pipiens with TOFs and measured atrazine
concentrations and the authors speculated that there may be a
natural background incidence of this phenomenon. Atrazine was
present in greater concentrations in the agricultural sites than in
the reference sites; however, several other pesticides such as the
phenoxy herbicides were also detected in agricultural locations.
Thus, causality could not be assigned.
A review of the available literature demonstrates that the appearance of rudimentary hermaphroditism and small incidence
of TOFs is a widespread phenomenon particularly in Ranids (see
Hecker et al., 2006, for a review of the terminology), with many
cases having been reported decades prior to the introduction of
atrazine (in 1959). As early as 1929, TOFs were reported in
R. clamitans (Witschi, 1929). Also, historical studies conducted
on museum specimens of cricket frog (Acris crepitans) collected
before and after the introduction of atrazine have found TOFs
(Reeder et al., 2005). These studies reported a historical increase
and subsequent decrease in the incidences of TOFs, which does
not match the temporal trends in the use of atrazine. In fact, these
authors speculate on the potential effects of organochlorine compounds such as PCBs as being a possible cause and conclude
that it is unlikely that the incidence was related to exposure to
atrazine. In addition, studies of frog populations conducted after the introduction of atrazine have not found a relationship
between the exposures of frogs to atrazine and the incidence of
TOFs (Smith et al., 2005; Coady et al., 2005; Jooste et al., 2005;
McDaniel et al., 2008; Smith, 2007 personal communication) or
a statistically significant association to the presence of atrazine
(Reeder et al., 1998).
Testicular Ovarian Follicles in Frogs as a Natural Phenomenon. Whether TOFs are a natural or induced phenomenon
has been extensively debated. A recent study of X. laevis from
a number of locations in South Africa reported that TOFs were
observed in frogs from a number of locations NE of the Cape-
fold Mountains (Du Preez et al., 2008a). Some of these locations
were remote from atrazine use and had no detectable concentrations of atrazine in the water (MDL 0.025 μg atrazine/L).
No TOFs were observed in frogs southwest of the Cape-fold
Mountains where no atrazine is used (see later discussion). Phylogenetic analysis of mtDNA and two nuclear DNA sequences in
these frogs showed that they belonged to two distinct haplotypes
and that the presence of TOFs was likely haplotype-specific.
This has serious implications for the use of TOFs as a marker
of endocrine-modulated responses in this species. Unless the
genetic background of the frogs being used and the homogeneity of the culture are known, results may be confounded. This is
particularly true for X. laevis where mating in cultured animals is
artificially stimulated and interhaplotype (or interspecies) mating can occur (Blackler et al., 1965; Blackler and Fischberg,
1968).
This haplotype-specificity must be considered in the interpretation of laboratory studies with X. laevis. In frogs from northeast of the Cape-fold Mountains, TOFs have been consistently
observed in controls as well as atrazine exposed animals but
there was no concentration-response to atrazine (Smith et al.,
2005; Jooste et al., 2005; Du Preez et al., 2008b). In frogs from
southwest of the Cape-fold Mountains, the source of most of
the frogs exported to other locations, no TOFs were observed in
animals captured in the wild (Du Preez et al., 2008a) or in laboratory studies where frogs, identified by haplotype to be from
this area, were exposed to concentrations from 0.1 to 100 μg
atrazine/L (Kloas et al., 2008). TOFs were not observed in control or treated animals in a study on X. laevis tadpoles exposed
from NF stage 49 to 66 to concentrations of atrazine between 0.1
to 100 μg/L (Oka et al., 2008). Although not specifically identified to genotype, these frogs were from a commercial source and
were most probably collected from southwest of the Cape-fold
mountains. In other studies of X. laevis, such as those conducted
by Hayes et al., the provenance of the culture and the genetic
homogeneity is unknown or unclear, which may explain some
of the anomalous results. Similar phenomena may apply to other
species of frogs and there may be a need to consider genotype
in these frogs as well.
Testicular Ovarian Follicles in Fish and Reptiles
Although there have been no reported studies on the effects of
atrazine on the incidence of TOFs in fish, this endpoint has been
used in assays for endocrine-disrupting substances (reviewed in
Grim et al., 2007). As in the case of frogs, it is important to
consider background incidence of this response in controls. In a
review of unexposed control Japanese medaka (Oryzias latipes)
used in 41 studies on a number of chemicals, TOFs (TOs) were
observed in 30% of all studies but with large variation between
the four laboratories in the study (0 to 100%) (Grim et al., 2007).
These observations suggest that, as in frogs, the presence of
TOFs in O. latipes is a natural phenomenon and that there may
be variations between strains of fish.
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
Downloaded At: 15:43 28 October 2008
A study where eggs of snapping turtles (Chelydra serpentina)
were exposed to atrazine via soil revealed no effects on thyroid
gland morphology or on the number of TOFs (TOs in De Solla
et al., 2006). Eggs from a nonagricultural site were incubated
in soil treated with 1× and 10× the concentration of atrazine
(as Atrazine 480 formulated product) applied in the field (3.1
and 31 L/ha) at a temperature of 25◦ C. Although the testes of
some hatchlings from the atrazine soil had 1 or more TOFs,
there were no statistically significant differences between these
and the unexposed control. It is also possible that TOFs, which
had a diameter of about 50 μm, were missed because sections
were taken at 175-μm intervals. These authors suggest that small
numbers of males with TOFs may be a natural phenomenon, as
has been reported by other authors, and does not appear to affect
fertility (Pieau et al., 1999).
Effects on the Ovary
There is little evidence to suggest that atrazine affects ovarian
development in Xenopus. Tavera-Mendoza et al. (2002b) evaluated the frequency of occurrence of primary and secondary
oogonia, which are the only stages of oogenesis present in the
ovary during sexual differentiation. They also evaluated the frequency of occurrence of atresia among primary and secondary
oogonia. Following exposure of stage-56 tadpoles to 21 μg/L
atrazine for 48 h, the frequency of occurrence of primary oogonia
was reduced but the frequency of secondary oogonia and atresia
was significantly greater (Tavera-Mendoza et al., 2002b). These
observations were inconsistent with the lack of effects reported
in the ovary of amphibians by others (Hayes et al., 2001; Carr
et al., 2003; Orton et al., 2006) and may have been compromised by the experimental design and other factors (see earlier
discussion). A study with P. promelas showed that exposure of
adults to measured concentrations 25 and 224 μg atrazine/L for
21 d had no effect on a range of parameters in females, including
body weight, gonadal somatic index (ratio of gonad mass to body
mass; GSI), stage of ovarian development, proportion of atretic
follicles or postovulatory follicles, number of eggs produced, or
the number of eggs hatched (U.S. EPA, 2005).
In other studies, Hayes reported that nonpigmented ovaries,
which occurred at relatively high frequencies in atrazine-treated
X. laevis larvae, were found in 4 individuals out of more than
400 control examined (1%) (Hayes et al., 2006b). Exposure of X.
laevis to the androgen receptor antagonist (cyproterone acetate,
CPA) from NF stages 50 to 66 resulted in a high proportion of animals (36%) with nonpigmented ovaries (Hayes et al., 2006b).
By comparison, exposure to estradiol had no effect on ovary
pigmentation. Hayes (2006b) speculated that the induction of
unpigmented ovaries by CPA suggests that this malformation
is the result of androgen depletion in atrazine-treated larvae,
potentially as a result of the induction of aromatase. This interpretation should be considered with some caution due to the lack
of evidence that atrazine induces aromatase in vivo and the high
mortality (42%) in the CPA-treated groups in the study. Differ-
739
ences in numbers of ovarian melanophores (pigmented cells)
between control and treated X. laevis were observed in only one
of two laboratories in the recent study reported by Kloas et al.
(2008). At the one laboratory, by gross observation and by histological examination, the proportion of NF stage 66 females
with fewer melanophores in the ovaries was statistically significant by trend test only and then only when all treatments were
included. The actual differences were not great (10% incidence
of frogs with fewer melanophores in the control and 15% in the
100 μg/L atrazine treatment) and a similar trend was not observed in the study at the other laboratory (Kloas et al., 2008).
Observations such as these have not been reported in other studies and the biological significance of changes in numbers of
ovarian melanophores is unknown.
Based on the available literature, there is little evidence to
suggest an effect of atrazine on sex differentiation or gonadal
development in reptiles, fish, or amphibians. Studies in the laboratory and field have failed to demonstrate an effect of atrazine
on sex ratio of reptiles or amphibians. Hayes has reported that
small concentrations of atrazine (≤0.1 μg/L) affect sexual development through effects on gross anatomy (multiple gonads) and
functional morphology (hermaphroditism) in X. laevis. However, these responses were not confirmed in a series of investigations working with X. laevis in the laboratory and in its native
habitat in South Africa. Hayes also reported that atrazine, at concentrations as little as 0.1 μg/L, contributed to the development
of TOFs in R. pipiens. However, subsequent studies showed that
the presence of TOFs is widespread among many species of
frogs and that these have been reported in museum specimens
collected before the introduction of atrazine. It is likely that the
presence of TOFs is a result of the general plasticity of gonadal
development in some amphibian species. While effects on reproduction would represent a significant concern, the weight of the
available evidence does not substantiate claims that atrazine is
a reproductive toxicant that feminizes and demasculinizes male
frogs.
VI.
MECHANISMS MEDIATING REPRODUCTIVE
EFFECTS
Atrazine has been proposed to exert adverse effects on the reproductive fitness of animals including mammals, fish, and amphibians. Some of these effects are well substantiated while others are not. Furthermore, mechanisms observed in one species
are often uncritically cited as support of proposed mechanisms
in other species and responses to relatively large exposures are
cited as support of theories to explain purported effects caused
by smaller exposures. Here we present each of the proposed
mechanisms of toxic action and discuss the species-specific responses and dose/concentration-response relationships and put
them into perspective relative to potential effects in frogs and
other aquatic vertebrates. Based on the available information,
the only evidence for effects of atrazine on concentrations of
steroid hormones in blood plasma are those reported by Hayes
et al. (2002). While Hayes and coworkers have hypothesized
740
K. R. SOLOMON ET AL.
that upregulation of CYP19 is a cause for adverse effects in
frogs, they have never reported any measurements of CYP19
mRNA or aromatase enzyme activity. In the one study where
they reported a decrease in concentration of testosterone, they
failed to measure estrogen and a mass balance, which would
have been supportive of this hypothesis, could not be calculated. When other workers have measured both CYP19 mRNA
expression and aromatase activity, no effects of atrazine have
ever been observed. There is no evidence to support the hypothesis that atrazine modulates CYP19 expression or the associated
aromatase activity in vivo. Atrazine has been found to upregulate CYP19 expression in some transformed cell systems, but not
others. The concentrations of atrazine required to affect CYP19
expression in vitro are much greater than those observed in the
environment or tissues of animals exposed to environmentally
relevant concentrations. Thus, there is essentially no support for
the aromatase hypothesis.
Downloaded At: 15:43 28 October 2008
A.
Mechanisms Mediated Through the HPG Axis
It has been hypothesized that atrazine or its degradation products can modulate the endocrine system through the central
nervous system (CNS). Atrazine has been found to suppress
the amplitude of the luteinizing hormone (LH) surge and prolactin concentrations in Sprague-Dawley and Long-Evans female rats by altering the hypothalamic control of these hormones (Cooper et al., 2000). Similar effects were observed recently when atrazine degradation products were found to affect
the onset of puberty and thyroid function in male Wistar rats via
actions on the CNS and its subsequent control of the pituitarygonadal axis (Stoker et al., 2002). These effects have only been
reported for rodents, and it is unknown whether amphibians
would respond in a similar manner. Regardless, because of the
small bioconcentration factor of atrazine for aquatic organisms
(see earlier description), the doses at which atrazine caused the
observed effects in rodents (100–200 mg/kg bw, via oral administration) are much greater than those to which frogs or other
aquatic organisms are likely to be exposed. For example, based
on a BCF of 1.5, these are equivalent to water exposures of
66,000–133,000 μg/L, well above the maximum water solubility of atrazine (33,000 μg/L) and several orders of magnitude
greater than typical environmental concentrations. Therefore, it
can be concluded that these types of effects are not ecologically
relevant in aquatic organisms. This is a good example of the need
to consider the fundamental difference between body dose and
matrix exposure concentration when citing papers as supporting
a particular mechanism of action.
B.
Mechanisms Mediated Through Aromatase
It has been proposed that atrazine can increase estradiol availability by upregulating expression of CYP19 (aromatase gene)
mRNA and thereby increasing aromatase activity (Fig. 7). This
could result in an increase in local availability of estradiol, depending upon where CYP19 is expressed. Since testosterone is
FIG. 7. Illustration of the function of aromatase in the synthesis of E2. Depending on the action of aromatase, the ratios of
testosterone and estradiol will change with resulting changes in
the expression of sexual characteristics that are determined by
these hormones.
the endogenous substrate for aromatase, this could also result
in increased plasma estradiol concentrations with a concomitant
decrease in plasma concentrations of testosterone. It has been
further proposed that this change in the ratio of testosterone to
estradiol can result in changes in the expression of those sexual characteristics that are under the control of these hormones.
While atrazine and several other triazine herbicides have been
shown to upregulate CYP19 expression and aromatase activity
in certain cell lines in vitro (Sanderson et al., 2000, 2001), other
cell lines have been found to be unresponsive (Heneweer et al.,
2004). These studies were conducted, in part, to explain anomalous results of in vitro studies that indicated that atrazine was
estrogenic (Davis et al., 1993). Atrazine has never been shown to
affect aromatase activity in vivo, although one study reported increased Cyp19A1 gene expression in ovaries of zebrafish (Danio
rerio) exposed to atrazine for 3 d at concentrations as small as
2.17 μg/L (Suzawa and Ingraham, 2008). Later in this paper we
review and critique the results of the studies that have examined the potential for effects of atrazine on aromatase in aquatic
organisms.
Aromatase activity in juvenile zebrafish was not affected
by exposure to atrazine (Kazeto et al., 2004). Transcription of
aromatase (CYP19 A1 and A2) in juvenile zebrafish was not
significantly affected by exposure to concentrations as great as
1000 μg atrazine/L (Fig. 8). The report of increased expression
of Cyp19A1 in ovaries of D. rerio (Suzawa and Ingraham, 2008)
is in contrast to the findings of Kazeto et al. (2004) but may have
been the result experimental design. There was no replication
of treatments (one tank of 15 fish only per concentration), concentrations were not measured, and pooling of samples (5 fish
each) for analysis may have obscured interindividual variability.
In the study by Kazeto et al. (2004), exposure to ethinyl estradiol
as a positive control caused reduced transcription. Fish were not
exposed to estradiol as a negative control in the concentrationresponse study by Suzawa and Ingraham (2008). The results
of Kazeto et al. (2004) are similar to those in frogs (discussed
below) and are not consistent with the theory the atrazine induces
aromatase activity during in vivo exposures.
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
FIG. 8. Effects of ethinyl estradiol (positive control) and
atrazine on CYP19A1 transcript abundance. Zebrafish juveniles
at 17 d post fertilization were exposed for 3 d. The results represent the mean ± S.E.M. of six samples: ∗ statistically different
from the control ( p < .05), ∗∗ data not actually given in original
figure but were described in text as not being different from 1000
μg/L. Redrawn from data of Kazeto et al. (2004).
A study on frogs exposed to atrazine under field conditions
in South Africa, where X. laevis is native, found no effects on
aromatase activity (Hecker et al., 2004) as did a similar field
study on R. clamitans in Michigan (Murphy et al., 2006a). Aromatase was not induced in X. laevis tadpoles (NF stage 49 to
66) exposed to concentrations of 0.1 to 100 μg atrazine/L (Oka
et al., 2008). In the same study, atrazine did not induce the production of vitellogenin in tadpoles at concentrations up to 100
μg/L (Oka et al., 2008). These studies included concentrations
in excess of the 0.1 μg/L that has been suggested by Hayes et al.
(2002) as the threshold above which this effect causes adverse
effects in frogs.
It has been suggested that induction of aromatase activity
results in the estrogen-like effects that are responsible for the
hypothesized feminization and demasculinization of male frogs
(Hayes et al., 2002). This theory was based on experiments
with adult maleX. laevis that reportedly exhibited lesser plasma
testosterone concentrations when treated with atrazine (Hayes et
al., 2002). Unfortunately, neither CYP19 mRNA expression nor
aromatase activity was measured in this study. Furthermore, concentrations of estradiol, the product of aromatase action, which
could have confirmed the hypothesis, were not measured by
Hayes et al. (2002). However, a different study with adult male
X. laevis exposed to atrazine concentrations ranging from 1 to
250 μg/L showed no significant effects on aromatase activity
or CYP19 mRNA expression (Hecker et al., 2005a). Because
aromatase activity can be small in testes, it is often difficult to
detect. However, the amplification methods developed by Park
et al. (2006) allow detection of as little as a single copy of CYP19
mRNA. Furthermore, it needs to be recognized that there are several forms of CYP19, with the “A” type occurring in gonad and
the “B” form occurring in brain.
741
The mechanism of induction of aromatase activity in cancer cell lines (KGN—human ovarian granulosa-like tumor cell
line, H295R—adrenal carcinoma cells, and NIH/3T3—mouse
fibroblasts) has been investigated (Fan et al., 2007). Aromatase
activity had previously been shown to be induced by atrazine and
simazine in H295R and JEG-3 cells in vitro (Sanderson et al.,
2000, 2001), but only at relatively large concentrations and not
in the rat R2C cell line (Heneweer et al., 2004). In this study, Fan
et al. (2007) provided evidence that mammalian cell lines cells
expressing the transcription factor steroidogenic factor-1 (SF-1)
and the aromatase promoter (ArPII), which is activated by SF-1,
were responsive to atrazine and simazine in terms of upregulated
mRNA expression and aromatase activity. However, this was a
highly artificial situation whereby cell lines were transfected
with multiple copies of the ArPII promoter, high copy levels
of SF-1, and exposed to very large concentrations of atrazine
or simazine (often >2000 μg/L). Even in these cases, CYP19
mRNA expression was often only about 1.5-fold greater than the
control levels. Given the complexity of this artificial cell system
and the very large concentrations of the triazines required to
mediate these effects, the significance of these observations in
whole organisms and tissues other than these cell lines is questionable. As discussed elsewhere in this section, environmentally
relevant concentrations of atrazine do not induce aromatase in
vivo in frogs or fish, which suggests that the responses observed
in mutated cells in vitro should not be extrapolated to whole
organisms in the field. This also suggests that the extrapolations
to human cancers discussed in the paper (Fan et al., 2007) are
highly speculative at best and are not supported by the greater
weight of evidence in the literature.
Because they were unable to make measurements of E2 or
CYP19 mRNA or aromatase activity and were thus unable to
test the aromatase hypothesis directly, Hayes et al. attempted to
investigate the mechanism of action of atrazine in frogs using
two model chemicals with different endocrine modes of action,
the anti-androgen cyproterone acetate (CPA) and the estrogen
17β-estradiol (Hayes et al., 2006b). They compared the effects
of a range concentrations of atrazine (0.1, 0.4, 0.8, 1.0, and 25
μg/L) on gonad morphology of early X. laevis life stages from
a previous study (Hayes et al., 2002) with those caused by exposure to CPA or estradiol. Larvae of X. laevis were exposed to
atrazine and CPA from NF stage 50 through 66. In the estradiol
experiments, larvae were treated for three different time periods: 7 d (NF stages 50–53), 14 d (NF stages 50–55), or 49 d (NF
stages 50–66). Each of the positive control chemicals was tested
at a single concentration (estradiol = 100 μg/L; CPA = 5 g/L).
Based on gross morphological and histological analyses, several malformations were reported to occur in juvenile X. laevis
exposed to atrazine (Hayes et al., 2006b). These included lobed
testes, unpigmented ovaries, the occurrence of TOFs, and, in
some rare occasions, multiple combinations of testes and ovaries
in the same individual. To support their findings, Hayes et al.
(2006b) presented a photograph of gross morphology and micrographs, which was already published in an earlier manuscript.
Author Figure 5 in Hayes et al. (2006b) states that the exposure
Downloaded At: 15:43 28 October 2008
742
K. R. SOLOMON ET AL.
concentration was 0.1 μg/L (0.1 ppb), but the identical photographs also appeared in author Figure 2 in Hayes et al. (2002)
where the text and caption state that the exposure was 1 μg/L
(1 ppb). The reason for the use of this figure and difference in
treatment concentration is unclear. No concentration-response
relationships were observed for any of the malformations described. The descriptions of the procedures used in the study
were unclear and difficult to follow and there was inconsistency
between the methods and the results in terms of the number
of animals reportedly used in the study. Not surprisingly, given
the large concentration (5 g/L = 5,000,000 μg/L), mortality in
the only concentration of CPA tested was 35%, which suggests
that any responses observed may have been artifacts of general toxicity. The authors stated that the results presented in this
study “suggest that atrazine-induced gonadal malformations result from the depletion of androgens and production of estrogens,
perhaps subsequent to the induction of aromatase by atrazine a
mechanism established in fish, amphibians, reptiles, and mammals (rodents and humans)” (2006b). However, neither the study
design nor the results they reported allow such conclusions to be
drawn. In fact, the authors (Hayes et al., 2006b) did not analyze
concentrations of testosterone or dihydrotestosterone in plasma
or tissue or the activity of steroidogenic enzymes such as17βHSD or CYP19. The authors were thus unable to test whether a
depletion of androgens or an increased production of estrogens
had occurred. The analysis of gross morphological and histological endpoints, as presented in this study, does not provide
sufficient information to extrapolate to a specific mode of action.
As for the statement that the aromatase mechanism has been established in fish, amphibians, reptiles, and mammals, there is
no information given in the cited papers to support this statement. As discussed earlier, the phenomenon of upregulation of
CYP19 activity has been observed only in transformed cell lines
but not in other cell lines. The current information suggests that
while atrazine at relatively great concentrations can upregulate
in vitro expression of CYP19 in some cell lines, the phenomenon
has never been demonstrated in amphibians in vivo, in the laboratory, or in the field.
C.
Effects of Atrazine on Plasma Sex Steroid Hormones
in Amphibians and Fish
Plasma Sex Steroids in Frogs
Interpreting the effects of chemicals, such as atrazine, on
plasma hormones in frogs is difficult due to the relatively great
degree of variation among individuals. Changes in plasma hormone concentrations may reflect alterations in synthesis, secretion, binding to plasma binding proteins, or changes in
metabolism of the hormone. This variation is exacerbated by seasonal effects and the duration of the breeding season (Licht et al.,
1983; Fasano et al., 1989; Mosconi et al., 1994). This is an important consideration since some frogs can breed once whereas
others can breed repeatedly or skip a season. In addition to the
factors just listed, the role of testosterone, 11-ketotestosterone
(KT), and dihydrotestosterone (DHT) in sexual development and
reproduction of amphibians is not clear.
Effects of Atrazine on Plasma Hormones in Frogs
If the putative aromatase-mediated mechanism of feminization and/or demasculinization of male frogs as proposed by
Hayes et al. (2002) were correct, one would expect to observe a
decrease in plasma testosterone concentrations and an increase
in plasma estradiol concentrations. Few studies have reported
effects of atrazine on plasma testosterone or estradiol concentrations in frogs. Only one study has reported a change in testosterone in response to exposure to atrazine at environmentally
realistic concentrations (Hayes et al., 2002). In that study, it was
reported that exposure of adult, male X. laevis to 25 μg atrazine/L
for 46 d resulted in a statistically significant decrease in plasma
testosterone concentrations to values that were the same as those
observed in unexposed, adult females. Concentrations of estradiol were not measured in the study by Hayes et al. (2002) and
only four animals were used. As discussed earlier, it was postulated that upregulation of the CYP19 gene was the cause of
this effect; however neither aromatase activity nor CYP19 gene
expression were measured in the study. If this effect was indeed
a result of induction of aromatase activity, then a decrease in
testosterone should be accompanied with an increase in estradiol. A study in which juvenile X. laevis were exposed to a range
of atrazine concentrations (0.1–25 μg/L) from 72 h posthatch
until 2 to 3 months postmetamorphosis found no statistically significant differences in plasma testosterone concentrations among
atrazine treatments and between treatments and the controls
(Coady et al., 2005). There were also no statistically significant
effects of exposure to waterborne atrazine on plasma concentrations of estradiol in females. However, in males exposed to 1.0
μg/L atrazine, plasma concentrations of estradiol were significantly less than those of controls, but not at greater or lesser exposures, such that there was no consistent concentration-response
relationship (Coady et al., 2005). Furthermore, the decrease in
concentrations of estradiol was opposite to the effects that would
have been caused by induction of aromatase.
No studies, conducted under controlled conditions, have been
able to repeat the observation of a decrease in plasma concentration of testosterone in male X. laevis at environmentally realistic
exposure concentrations, in either the laboratory (Coady et al.,
2005; Hecker et al., 2005a) or the field (Hecker et al., 2004).
It was reported that atrazine exposure reduced the concentration of testosterone in plasma of adult male X. laevis (Hecker
et al., 2005a), but the lowest-observed-effect concentration for
this response was 250 μg atrazine/L.
The concentrations of testosterone in plasma shown in (Fig. 9)
illustrate an error made in the article by Hayes (2004) when he
compared concentrations from his study to those of Hecker et
al. (2003). Because the frogs used in the Hecker et al. study
were juveniles, the results should be compared to other studies
on juveniles such as the Kang et al. (1995, data shown for
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
743
FIG. 9. Testosterone concentrations in plasma of Xenopus laevis in a number of studies. Mean and SE data except for field study
in South Africa (Field SA) where the median of medians is shown. PM4 and PM5 are post-metamorphic juveniles not exposed to
atrazine. Estradiol is a positive control. Redrawn from data from Kang et al. (1995), Hecker et al. (2003), Hecker et al. (2005b),
Hecker et al. (2005a), Hecker et al. (2004), and Hayes et al. (2002), as referenced in the figure.
postmetamorphic stages 4 and 5), not to the adults tested by
Hayes et al. (2002) or the laboratory and field studies on adults
by Hecker et al. (2005a). As is clear from the observations in
juveniles and adults (Fig. 9), the only study that claims an effect
of atrazine at environmentally realistic concentrations is that of
Hayes. Given the lack of effect seen on aromatase (discussed
earlier) and the fact that concentrations of estradiol were not
reported by Hayes et al. (2002), the significance and physiological mechanisms underlying this response remain unclear. Since
the results reported by Hecker et al. (2005a) found that atrazine
reduced plasma testosterone only at the greatest concentration
tested (250 μg/L) and did not result in an concomitant increase
in plasma estradiol concentrations, CYP19 gene expression, or
aromatase activity, to date there is no published in vivo information supporting the proposed aromatase mode of atrazine action
in frogs.
Several studies have examined the relationship between concentrations of atrazine in the field and plasma hormone con-
centrations in frogs. A negative correlation was found between
concentrations of atrazine in water and concentrations of testosterone in plasma of adult female X. laevis inhabiting atrazineexposed ponds in South Africa; however, due to the presence
of other confounding factors, such as other agricultural chemicals, it was impossible to establish a direct cause–effect relationship between atrazine and plasma testosterone concentrations. Concentrations of testosterone and 11-ketotestosterone in
plasma of R. clamitans and R. pipiens from wetland areas in
Ontario showed no correlation with concentrations of atrazine
(McDaniel et al., 2008).
Concentrations of the hormones, testosterone, estradiol, and
11-ketotestosterone in plasma were measured in R. clamitans inhabiting Michigan (Murphy et al., 2006b). Estradiol, testosterone, and 11-ketotestosterone concentrations,
as well as the ratios of estradiol/testosterone and 11ketotestosterone/testosterone, of adult male frogs were significantly different among all of the locations. Atrazine
744
K. R. SOLOMON ET AL.
Downloaded At: 15:43 28 October 2008
concentrations were significantly and positively correlated with
11-ketotestosterone in adult females in 2002 and with testosterone in juvenile males in 2003, but were not significantly
correlated with any other parameter. Estradiol/testosterone,
11-ketotestosterone, and 11-ketotestosterone/testosterone ratios were significantly greater at agricultural sites than nonagricultural sites (Murphy et al., 2006b), with a power (1 –
β) to detect differences in GSI, testosterone, and estradiol in
2002 that was less than 0.20. The conclusion of Murphy et al.
(2006b) was that there were no consistent effects on plasma hormone concentrations along an atrazine exposure concentration
gradient.
Overall, there is no evidence from either laboratory or field
studies that exposure to atrazine at realistic environmental concentrations leads to changes in the concentrations of estradiol
or testosterone in the blood plasma of male frogs or that this
subsequently results in feminization.
Effects of Atrazine on Plasma Hormones in Fish
Several studies have evaluated the effects of atrazine on sex
steroids in fish. Most of these studies have been on adults and
have determined levels of sex steroids in the plasma, while some
studies on small-bodied fish have included measurement of levels of steroids within the whole body or the gonads. Overall,
changes in sex steroid levels have been minimal, and when
the effects have been seen this has occurred at atrazine concentrations much greater than those typically observed in the
environment.
No effects on the concentration of testosterone, estradiol, or
11-ketotestosterone in the testes were observed when sexually
mature goldfish (Carassius auratus) were exposed to atrazine
(nominal concentration of100 or 1000 μg/L) for 21 d (Spanó
et al., 2004). Exposure to atrazine at 1000 μg/L, but not at 100
μg/L, caused a decrease in testosterone and 11-keto-testosterone
concentrations in plasma and increased plasma estradiol concentrations in males. There were no changes in the concentration of
vitellogenin in plasma, which is an estradiol-dependent response
(Spanó et al., 2004). This result indicates that the changes observed in plasma concentrations of the steroid hormones were
either transitory, in error, or not sufficiently great to cause adverse effects on reproductive function of C. auratus.
Largemouth bass (Micropterus salmoides), approximately 2
y of age, were exposed to technical atrazine at nominal concentrations of 0, 25, 35, 50, 75, or 100 μg/L in the water column for 20 d during the nonreproductive season. An additional
treatment of 100 μg/L commercial formulation of atrazine was
also utilized, which contained surfactants and other inert ingredients (Gross et al., 1997). Both studies showed small effects on concentrations of some steroid hormones and vitellogenin (Fig. 10), but there was no concentration-dependent
relationship and the statistical differences were not consistent
so no overall conclusions could be made. Results for female
bass indicated an inconsistent, non-concentration-dependent re-
FIG. 10. Plasma concentration of steroid hormones associated
with reproduction and vitellogenin in male and female bass exposed to technical and formulated atrazine (100 μg/L). The star
symbol indicates statistically significant differences from the
control. Redrawn from data of Gross et al. (1997).
sponse for testosterone in plasma with no statistically significant differences between control and treatments. Concentrations of 11-ketotestosterone in plasma of female M. salmoides
were the same, regardless of atrazine exposure concentration.
Results for female M. salmoides did, however, indicate significantly greater estradiol concentrations in plasma of fish exposed to 100 μg/L commercial formulated atrazine and almost significantly greater concentrations in plasma of fish exposed to 100 μg/L technical atrazine. No statistically significant responses in plasma vitellogenin concentrations, which are
a sensitive estradiol-dependent response in fish, were observed,
which suggests that there were no functional changes due to the
small, inconsistent, and transient changes in plasma estradiol
concentrations.
Results for male M. salmoides did not indicate any significant effects of atrazine on plasma concentration of estradiol,
testosterone, or plasma vitellogenin, regardless of atrazine exposure concentration or whether formulated or technical atrazine
was used. Concentrations of 11-ketotestosterone in plasma were,
however, significantly less at exposure concentrations greater
than or equal to 50 μg atrazine/L, regardless of formulation.
The 11-ketotestosterone response in male fish appeared to be a
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
threshold response. Previous, preliminary studies with greater
atrazine exposures did not report significant changes in plasma
11-ketotestosterone concentrations (Gross et al., 1997; Grady
et al., 1998).
A study of Atlantic salmon (Salmo salar) reported that exposure to atrazine at nominal concentrations of 2 or 20 μg/L
resulted in changes in response to the priming effect of ovulated female salmon urine, changes in androgen secretion, and
changes in steroid concentrations in the bile (Moore and Waring,
1998). The authors hypothesized that atrazine can affect testosterone concentrations by increasing enzymatic transformation of
the hormone. However, no estradiol measurements were made in
that study. Increased metabolic activity is a common response to
exposure to environmental pollutants as well as being a natural
phenomenon in the recrudescence cycle. Thus, it cannot be excluded as a possible cause for the observed response. Results of
a 21-d reproduction bioassay where P. promelas were exposed to
atrazine at measured concentrations of 25 and 224 μg/L showed
no treatment-related effects on estradiol or testosterone in females or testosterone and 11-ketotestosterone in males (U.S.
EPA, 2005).
Based on the available information, the only evidence for effects of atrazine on concentrations of steroid hormones in blood
plasma are those reported by Hayes et al. (2002). No other researchers have been able to reproduce these results, and the study
by Hayes et al. (2002) does not report concentrations of estradiol so it is impossible to make inferences about the potential
mechanisms of action.
VII. EFFECTS ON LARYNGEAL DEVELOPMENT
Laryngeal development in frogs is a sexually dimorphic process, and the formation of a larynx capable of male-calling behavior is androgen dependent. It has been hypothesized that
atrazine could act as an endocrine disruptor by decreasing
plasma concentrations of testosterone in X. laevis, which results
in the development of a laryngeal dilator muscle with smaller
volume (Hayes et al., 2002). Theoretically, this endpoint could
serve as an integrating measure of androgen-dependent processes that would respond to subtle changes in androgen status
during critical periods of development.
A.
Effects on the Laryngeal Dilator Muscle
Under normal conditions, the laryngeal dilator muscle of
male X. laevis is larger than that of females (Sassoon and Kelley
1986; Tobias et al., 1993). It has been reported that exposure
of X. laevis to concentrations of atrazine as little as 0.1 μg/L
resulted in smaller laryngeal adductor muscles at NF stage 66
than in unexposed males (Hayes et al., 2002). In contrast, Carr
et al. (2003) and Coady et al. (2005) found no evidence based on
laboratory studies to suggest that atrazine at concentrations as
great as 25 μg/L reduced laryngeal muscle size in male X. laevis.
In the field, there were no differences in larynx weight in either
male or female X. laevis from the corn-growing regions and the
non-corn-growing regions in South Africa (Smith et al., 2005).
745
The differences between the findings of Carr et al. (2003)
and Coady et al. (2005) and those of Hayes et al. (2002) may be
explained by differences in sampling, both at the level at which
animals were selected for analysis as well as how laryngeal size
was determined. Xenopus laevis that completed metamorphosis earlier were found to always be larger than their siblings
that completed metamorphosis a few weeks later (Carr et al.,
2003). For this reason, Carr et al. (2003) randomly selected animals that represented the entire range of body sizes from each
tank. In contrast, Hayes et al. (2002) systematically selected
animals that completed metamorphosis early. Whether differences in body size confounded their analysis is unknown. An
additional difference is the method used to determine dilator
muscle size. The dilator muscles are not uniform in shape, and
Carr et al. (2003) employed a method for determining dilator
muscle volume that previously has been used by other investigators to determine laryngeal muscle volume in frogs (McClelland et al., 1996, 1998). This analysis utilized measurements
from evenly spaced sections through the rostral–caudal extent
of the muscle, rather than sampling from one point in the muscle. In contrast, Hayes et al. (2002) determined dilator muscle
size by subjectively selecting the “largest section” through the
muscle and determining the cross-sectional area of this section.
Even though the Carr et al. (2003) subsequent analyses have
determined that largest crosssectional area through the muscle
is the best descriptor of sex differences in muscle size, their
method of choosing the largest section was quite different from
that of Hayes et al. (2002).
The studies on the laryngeal muscle provide important insights into the endocrine physiology of frogs and their susceptibility to atrazine. Measurable differences in muscle size at NF
stage 66 between sexes suggests that androgen secretion begins
prior to completion of metamorphosis in X. laevis. It is evident
that the laryngeal muscle is responsive to androgens since exposure to dihydrotestosterone caused the laryngeal dilator muscle
cross-sectional area to be greater in exposed than unexposed
X. laevis, an expected response for the positive control. Based
on differences observed in laryngeal dilator muscle size between sexes, the data suggest that larval androgen secretion proceeds normally in frogs exposed to atrazine (Carr et al., 2003).
Overall, there is little evidence to support the theory that exposure to atrazine affects laryngeal development, either directly or
indirectly.
VIII.
EFFECTS ON THYROID FUNCTION AND
DEVELOPMENT
The thyroid hormones (TH: triiodothyronine, T3;
tetraiodothyronine, T4) have a constellation of effects on
wildlife, ranging from the control of postembryonic growth
and tissue differentiation to effects on reproduction (Carr
and Norris, 2006). Given the particular importance of TH in
developing organisms, it is not surprising that several studies
have considered the thyroid as a possible target for the effects of
atrazine. In theory, chemicals can affect TH levels in the blood
746
K. R. SOLOMON ET AL.
via a number of pathways including (1) disruption of thyroidal
iodide uptake and TH synthesis, (2) alteration in plasma TH
binding protein levels, (3) effects on deiodination of T4 to
T3, (4) effects on TH metabolism, or (5) direct action on TH
receptors. Two model systems in particular have been studied
with respect to atrazine effects on the thyroid axis: amphibian
metamorphosis and smoltification in juvenile salmonids. Most
studies have reported no consistent effects of atrazine on
amphibian metamorphosis. Some effects of atrazine have been
reported in smolting salmon, although there appear to be no
consistent patterns of effect, with some studies demonstrating
elevated plasma TH while others report decreased levels of TH
after atrazine exposure.
Downloaded At: 15:43 28 October 2008
A.
Effects of Atrazine on Amphibian Metamorphosis
Since most amphibian species require normal TH synthesis to complete metamorphosis, metamorphosis is a particularly
sensitive endpoint for assessing effects on thyroid function. Disruption of TH synthesis during metamorphosis can lead to reduced hind limb growth, a delay in metamorphosis, and feminization of the gonads (in anuran amphibians, which require
TH for androgen receptor expression, Robertson and Kelley,
1996).
In general, studies examining effects of atrazine on metamorphosis suggest no consistent concentration-related effect of
atrazine on thyroid function (Allran and Karasov, 2001; Hayes
et al., 2002, 2003; Sullivan and Spence, 2003; Jooste et al., 2005;
Carr et al., 2003; Orton et al., 2006; Kloas et al., 2008). Several
studies have reported no effect of atrazine on the time to metamorphosis (Hayes et al., 2002, 2003; Carr et al., 2003; Kloas et
al., 2008) in X. laevis or R. pipiens (Allran and Karasov, 2000;
Orton et al., 2006), while Coady et al (2004) reported that 10
μg atrazine/L, but not 25 μg atrazine/L, inhibited metamorphosis relative to controls in larval R. clamitans. When X. laevis
tadpoles were exposed to concentrations from 20 to 320 μg
atrazine/L, Sullivan and Spence (2003) reported a significant
positive relationship between atrazine concentrations and time
to metamorphosis. These effects are difficult to interpret considering that individual concentrations of atrazine both accelerated
(20 μg atrazine/L) and slowed (320 μg atrazine/L) time to metamorphosis to a small degree, whereas concentrations between 40
μg atrazine/L and 160 μg atrazine/L had no effect on metamorphosis (Sullivan and Spence, 2003). Metamorphosis was slightly
delayed by approximately 5 d relative to controls in larval tiger
salamanders (Ambystoma tigrinum) exposed to 75 μg atrazine/L
(Larson et al., 1998). The same authors found that salamanders
exposed to 250 μg atrazine/L actually reached later stages of
metamorphosis more quickly (by 1–2 d) than controls (Larson
et al., 1998). They also reported that exposures to 75 and 250
μg atrazine/L resulted in elevation of thyroxine in plasma of
stage-IV larvae (from 0.9 μg/ml in control to 1.5 μg/ml) but
with no concentration response, possibly explaining the shorter
time to metamorphosis. This effect was not observed in stage-II
larvae (Larson et al., 1998). The mechanism by which elevation
of thyroxin occurred was not investigated but could have been
related to toxic stress.
B.
Effects of Atrazine on Smoltification
Thyroid hormones, prolactin, corticosteroids, and growth
hormones (GHs) are all required for normal smoltification, a
developmental process that accompanies migration from fresh
water (FW) to salt water (SW) in anadromous salmon (Barron, 1986). Waring and Moore (2004) reported that exposure
to atrazine concentrations between 1 and 10 μg/L resulted in
less Na+ /K+ -ATPase activity in gills in Atlantic salmon (Salmo
salar) smolts exposed for ≤7 d in fresh water. There were no
effects of atrazine on plasma T3 or T4 in smolts exposed in
fresh water (Waring and Moore, 2004). The authors reported a
concentration-related increase in plasma T4 but not T3 in smolts
exposed to atrazine ranging from 1.1 μg atrazine/L to 22.7 μg
atrazine/L in fresh water and then transferred to seawater (Waring and Moore, 2004). The mechanism underlying the effect
of atrazine on plasma T3 in smolts after transfer to seawater is
unknown.
In another study (Nieves-Puigdoller et al., 2007), S. salar
smolts were exposed to atrazine at measured concentrations of
8.5 ± 1.1 (SEM) and 84.3 ± 1.3 μg/L in fresh water (FW) for
21 d, then exposed to a saltwater challenge for 24 h, and then
returned to FW for a further 3-month period of observation. At
8.5 μg/L, atrazine had no effect on plasma levels of cortisol,
growth hormone (GH), insulin growth factor I (IGF-I), T4 and
T3, Cl− , Mg2+ , Na+ , or Ca2+ in FW or after SW challenge. No
effect on plasma levels of GH, IGF-I, T4, or T3 was found in
FW smolts exposed to atrazine at 84 μg/L, but, following SW
challenge, fish had significant increases in hematocrit, plasma
cortisol, Cl− , Mg2+ , Na+ , and Ca2+ and a decrease in plasma
levels of T4 and T3. These data are not consistent with an effect on deiodinase activity, since plasma T4 and T3 were both
decreased in fish exposed to 84 μg atrazine/L. Whether these
effects are related to changes in plasma TH binding proteins is
unclear since differences in free versus bound T4 and T3 were
not reported. The reported lessening of plasma T3 is different
from the effects of atrazine on plasma T3 reported by Waring
and Moore (2004) and was only observed at the greater test concentration. The authors did report mortality (9%) in the smolts
exposed to atrazine at 84 μg/L but not at 8.5 μg/L. In addition,
feeding was reduced by close to 75% after 10 d of exposure
and by 100% after 15 d of exposure to 84 μg atrazine/L but
not in fish exposed to 8.5 μg/L (Nieves-Puigdoller et al., 2007).
Reduced food consumption was accompanied by a significant
loss of weight at the end of the exposure period when compared
to all other exposures. Fish exposed to 84 μg atrazine/L were
stressed based on the elevated blood glucose and cortisol levels and anorexia observed in these animals. Dietary restriction
has been reported to affect concentrations of testosterone in rats
(Trentacoste et al., 2001), but reports of similar effects in fish
were not found in the literature. The reason why animals exposed
to 84 μg atrazine/L exhibited a stress response is unclear, since
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
atrazine would not be expected to be acutely toxic to S. salar,
although it has not yet been tested in this species. The 10th centile for the 96-h LC50 species sensitivity data in SW and FW
fish was 3840 μg/L and the 10th centile for chronic exposures in
all aquatic animal studies was 40 μg/L (Giddings et al., 2005).
The most sensitive fish to chronic exposures was S. fontinalis
with a NOEC of 65 μg/L for a 308-d exposure. Thus, the effects reported in this study may be caused by general toxicity
and are not a specific hormone-mediated response. The authors
of the study suggest that atrazine can have adverse effects on
salinity tolerance in anadromous fish, but they also point out
(correctly) that the concentration at which responses were observed would be rarely observed in the environment, especially
in flowing water.
Overall, there are no consistent reported effects of atrazine
on metamorphosis in fish or amphibians. Given the fact that
amphibian metamorphosis is a very sensitive indicator of thyroid
function in frogs, and has been recommended as a Tier I test
for EDCs by the U.S. EPA Endocrine Disruptor Screening and
Testing Advisory Committee (EDSTAC, U.S. EPA, 1998), these
results suggest that atrazine does not consistently affect thyroid
function in developing amphibians. Atrazine has inconsistent
effects on plasma TH levels in salmon smolts, with no clear
mechanism of action.
IX. EFFECTS ON STRESS PHYSIOLOGY
Exposure to a wide range of environmental or physiological
stressors can increase the activity of the hypothalamus-pituitaryadrenal axis, leading to greater concentrations of corticotrophin
(ACTH) and adrenal corticosteroids in plasma. There are reports of atrazine causing greater plasma corticosteroid concentrations in salmon. Some effects of atrazine on in vitro adrenal
steroidogenesis have been reported at high atrazine concentrations, although the effects appear to be species specific. Overall
these studies suggest that atrazine at environmentally relevant
concentrations has limited effects on the adrenal axis.
A.
Effects on Plasma Corticosteroids
The major corticosteroids secreted by adrenal steroidogenic
tissue can differ depending upon the species; in most fish species
cortisol is produced, whereas corticosterone is the primary corticosteroid in amphibians and reptiles. There are only a few studies examining the effects of atrazine on cortisol in plasma of
aquatic vertebrates. Exposure of carp (Cyprinus carpio) to 100
μg atrazine/L for 72 h or less resulted in greater plasma cortisol
concentrations (Gluth and Hanke, 1985). This effect was dependent upon exposure temperature, since carp exposed to 100 μg
atrazine/L exhibited a 5-fold increase in plasma cortisol relative
to controls at 17o C but only a 3-fold increase when exposed at
22o C (Gluth and Hanke, 1985) for 72 h. Exposure to atrazine
resulted in greater plasma glucose concentrations as well, possibly an effect that was secondary to elevated cortisol concentrations. Interestingly, similar cortisol responses were observed
after 72 h of exposure to a wide variety of contaminants (100
747
μg/L aldrin, 50 μg/L DDT, 20 μg/L dieldrin, 2 μg/L endrin,
l00 μg/L lindane) and industrial chemicals (100 μg/L toluene;
1 mL/L methanol), which suggests that this was a nonspecific
stress response. Exposure to atrazine, at concentrations between
6.5 and 22.7 μg/L, was reported to cause a concentration-related
increase in plasma cortisol concentrations in S. salar smolts exposed in fresh water (Waring and Moore, 2004). Similarly, concentrations of cortisol in plasma of S. salar smolts were greater
when they were exposed to 84 μg atrazine/L (Nieves-Puigdoller
et al., 2007), although these authors did not find an effect of
lesser (8.4 μg/L) atrazine concentrations. Salmo salar exposed
to 84 μg atrazine/L also exhibited hyperglycemia and anorexia
(Nieves-Puigdoller et al., 2007), suggestive of a stress response.
The mechanisms underlying these effects on plasma cortisol is
unclear since atrazine is not overtly toxic at these concentrations
and plasma ACTH and corticosteroid binding globulin were not
measured in any of these studies.
The effects of atrazine on plasma corticosterone concentrations were studied in larval A. tigrinum reared in the laboratory
(Larson et al., 1998). Animals were exposed to atrazine at concentrations as great as 250 μg/L in a static renewal protocol for
86 d. The authors reported differences in plasma concentrations
of corticosterone between stage II and stage IV but no concentration response to atrazine; however, only two concentrations
were tested. They concluded that atrazine does not directly affect
corticosterone at the concentrations tested.
B.
Effects on Adrenal Steroidogenesis and Secretion
Adrenocortical tissue in fish and amphibians is intermingled with kidney tissue, making it impossible to directly isolate adrenal tissues for in vitro studies in these species. Thus,
the few studies that have examined a direct effect of atrazine
on adrenal steroidogenesis have used mixed kidney and adrenocortical tissue. The direct effects of atrazine on the responsiveness of adrenal cortical tissue to an ACTH challenge have been
examined in trout head-kidney cells in vitro (Bisson and Hontela, 2002). Atrazine had no effect on viability of cells, even
at the greatest concentration tested (500 μM ≈ 100,000 μg/L,
Fig. 11). A significant increase in cortisol secretion was observed
at 500 μM atrazine (≈ 100,000 μg/L, Fig. 11). When cells
were stimulated with ACTH but not with a membrane-permeable
form of cAMP, dibutyryl adenosine-cyclic monophosphate (dbcAMP). This concentration of atrazine is very unrealistic, even
if one assumes some bioconcentration of atrazine in fish—the
median value for the bioconcentration factor in fish reported
in the literature is 2 and the maximum, 12 (Giddings et al.,
2005). A reduction in ACTH-stimulated cortisol secretion was
observed at concentrations of atrazine between 0.005 and 5 μM
(≈ 1 to 1000 μg/L, Fig. 11). Stimulation of cortisol secretion
with dbcAMP resulted in an increase in cortisol secretion but no
concentration response to atrazine exposures was observed.
Using two different types of suspended cell preparations
from X. laevis and R. catesbeiana, Goulet and Hontela (2003)
examined the effects of atrazine on adrenal corticosterone
Downloaded At: 15:43 28 October 2008
748
K. R. SOLOMON ET AL.
FIG. 11. ACTH- and dbcAMP-stimulated cortisol secretion and viability (± SEM) of trout head-kidney cells following in vitro
exposure to atrazine. Statistical significance was evaluated by Dunnett’s test ( p < .05) and Student’s t-test ( p < .05). The number
of replicates was five to eight for ACTH and dbcAMP and three to six for viability. Redrawn from data of Bisson and Hontela
(2002) to show concentrations in μg/L.
secretion. Kidney cells from X. laevis and “adrenal” cells from R.
catesbeiana were used, although the degree to which the adrenal
cell preparation was contaminated with kidney cells was not reported. Atrazine had no effect on cell viability or corticosterone
secretion induced by ACTH or dbcAMP at concentrations ranging from 10–8 M to 10–4 M atrazine (2.16 to 21,600 μg
atrazine/L for X. laevis kidney cells). Atrazine significantly decreased ACTH- and dbcAMP-evoked corticosterone secretion
from R. catesbeiana adrenal cells at concentrations of 10 μM
and greater (2160 μg atrazine/L). The ecological significance
of these results is unclear given that atrazine was only effective at concentrations (>1000 μg/L) unlikely to be found under natural conditions. The authors claim that R. catesbeiana
is a more sensitive model for studying contaminant effects on
adrenal function. However, the basis for such a direct comparison of sensitivities to atrazine is weak since it is based upon
different cell preparations (kidney cells for X. laevis vs. adrenal
cells for R. catesbeiana).
C.
Effects of Pesticide Mixtures on Corticosteroid
Secretion
A study was conducted to evaluate the effects of nine different pesticides, alone and in combination, on early development
in R. pipiens and corticosteroid homeostasis in adult X. laevis
(Hayes et al., 2006a). Pesticides used in the exposure experiments were four herbicides (atrazine, metolachlor, alachlor, and
nicosulfuron), three insecticides (cyfluthrin, cyhalothrin, and
tebupirimphos), and two fungicides (metalaxyl and propiconazole). Effects were assessed either for each pesticide alone or for
the mixture of all nine compounds. In addition, a binary mixture
of atrazine and S-metolachlor and the commercial formulation
Bicep II Magnum, which contains both of these herbicides, was
investigated. Responses to individual pesticides were assessed
at a concentration of 0.1 μg/L only. The binary mixtures of
atrazine and metolachlor were tested at 0.1 and 10 μg/L. In addition to the individual studies, a nine-chemical mixture was also
assessed with each pesticide at a concentration of 0.1 μg/L. Not
surprisingly, the authors reported that the pesticide mixture had
a much greater effect on larval growth and development than
did the individual chemicals when tested alone. Furthermore,
the authors reported that exposure to the nine-pesticide mixture
resulted in damage to the thymus and theorized that this resulted
in subsequent immuno-suppression and greater incidence of infection with Flavobacterium menigosepticum in tadpoles and
young frogs (Hayes et al., 2006a). They also reported increased
plasma corticosterone levels in adult X. laevis exposed to the
nine-pesticide mixture. Gonadal development was not assessed
in this study. The experimental design in this study was seriously
flawed. It is possible to test interactions between two or three
substances using the isobologram approach but more complex
mixtures require a ray or multifactorial design (McConkey et
al., 2000). Furthermore, tests for additivity or interactions must
be based on potency, rather than amount of the chemical. This
requires that the concentration response be separately characterized for all components of the mixture before any combinations
are assessed. That the nine-component mixture caused a greater
response is because the mixture was nine times more concentrated than the individual components. Because of problems in
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
the experimental design, it is not possible to determine whether
the observed effects were the result of potency addition, response
addition, synergism, antagonism, or a combination of all of these
(LeBlanc and Wang, 2006). Thus, this study was neither informative with respect to the potential effects of mixtures, nor did
it offer any elucidation on the potential effects of atrazine.
Thus, although atrazine exposure has been reported to elevate plasma cortisol in salmon smolts, the mechanism underlying these effects is not clear. Large concentrations of atrazine
have been reported to alter adrenal steroid secretion in vitro in
frogs (>1000 μg/L) and fishes (84 μg/L), but the responses
are inconsistent between species and may be a general stress
response to large exposures that would be rarely found in the
environment.
Downloaded At: 15:43 28 October 2008
X.
EFFECTS ON IMMUNE FUNCTION
Few studies of the immunotoxic potential of atrazine are
available. The majority of studies on the immunotoxic properties of atrazine that have been published so far were conducted
using mammalian in vitro or in vivo test systems. The effects
observed for the exposure of mammalian systems to a range
of concentrations of atrazine included alterations of both the
humoral and/or cellular immune response. Responses to subacutely toxic exposures in mammals were decreased numbers
of natural killer cells (Whalen et al., 2003), decreased numbers
of T and B lymphocytes (Filipov et al., 2005; Fournier et al.,
1992; Karrow et al., 2005), increased macrophage phagocytotic
activity (Fournier et al., 1992), decreased cytokine production
(Hooghe et al., 2000), decreased lymphocyte activation capacities by PHA (Pistl et al., 2002), and decreased B16F10 tumor
challenge capacity in mice (Karrow et al., 2005). In a review
of the regulatory and published studies conducted to evaluate
the potential effect of atrazine and/or its chlorometabolites on
immune system parameters in mammals, Pastoor et al., (2008)
reported that there were few studies that reported an observed
association between exposure to atrazine and effects on the immune system at realistic exposures. Two sensitization studies
conducted on technical atrazine in guinea pig were positive,
while a third was negative and a sensitization study conducted
in humans was negative. There was no evidence of effects of
atrazine or its chlorometabolites on the immune system of rats,
rabbits, or dogs at chronic doses that resulted in low to moderate reductions in body weight gain (5% to 15%—below the
maximum tolerated dose—MTD). Doses in excess of the MTD
(i.e., body weight gain reductions greater that 15%) were associated with reduced spleen and/or thymus weight along with
reductions in other organ weights. In addition, such high doses
appeared to delay the development of hematopoiesis in bone
marrow. This was primarily reflected as reduced erythroid parameters in such animals and, in extreme cases, by the activation
of extramedullary hematopoiesis in the liver. Multigeneration
reproduction studies conducted in rats did not indicate any increased sensitivity of the immune system resulting from exposure to atrazine during development. There was no evidence
749
of increased incidence or onset of cancers originating in the
immune system and no evidence of increased disease susceptibility resulting from lifetime treatment of rats or mice with
MTDs of atrazine. Some effects in immune parameters were
reported in short-term, high-dose studies conducted either in
vivo or in vitro with cell models. Many of these were conducted
at concentrations that were near the solubility limit of atrazine
(30,000 μg/L) or at acute and repeat doses that greatly exceeded
the MTD (Pastoor et al., 2008). These effects were judged to be
secondary to a generalize stress response as been observed by
others for atrazine and other substances (Pruett et al., 2003).
Effects observed for the exposure to atrazine using artificial
or in vivo exposure systems raised the question whether atrazine
may cause similar responses in aquatic species such as fish and
amphibians. Compared to the literature on the effects of atrazine
on the mammalian immune system, however, little information
is available on similar effects in fish or amphibians.
A.
Effects of Atrazine on Immune Function in Fish
Effects of atrazine exposure on the immune system of fish
included several aspects of the cellular immune response such
as degeneration of macrophages, an increase in the number
and size of hepatic melano-macrophage centers in the euryhaline muglid fish species Liza aurata and L. ramada (BiagiantiRisbourg, 1990), and leucopenia and atrophy of lymphoid organs in salmonids (Walsh and Ribelin, 1975). Other effects
on salmonid species (Oncorhynchus kisutch, Salvelinus namaycush) included reduction of spleen weight and decreased number
of lymphocytes (reviewed in Zeeman and Brindley, 1981). A series of studies on the effects of atrazine on immune function of C.
carpio did not find any significant alterations at concentrations
as great as 28,000 μg atrazine/L (Cossarinidunier and Hattenberger, 1988; Cossarinidunier, 1987) or with a dose of 10,000
μg atrazine/kg body weight (b.w.) (Cossarinidunier et al., 1988)
in vitro or in vivo, respectively. A summary of the effects of a
range of atrazine concentrations on the immune system of teleost
fish is given in Table 2.
It appears that there are distinct differences in the sensitivity
of fish species in terms of their immune response to exposure
with atrazine. While L. aurata, L. ramada, and some salmonid
species exhibited the first signs of an alteration in their cellular immune response at concentrations as small as 25 and 100
μg atrazine/L, respectively (Walsh and Ribelin, 1975; BiagiantiRisbourg, 1990), no effects on either the humoral or the cellular
level, were observed in C. carpio at concentrations as great as
28,000 μg atrazine/L (Cossarinidunier and Hattenberger, 1988;
Cossarinidunier, 1987). Concentrations of atrazine have been
reported to exceed 20 μg atrazine/L only in rare occasions,
even directly after application (Battaglin et al., 2000; Solomon
et al., 1996; Giddings et al., 2005). These concentrations typically occur only for a short time in environments that are contiguous to agricultural lands, and seldom exceed the threshold
concentration of 25 μg atrazine/L for which immunological effects have been observed in a single study (Biagianti-Risbourg,
750
K. R. SOLOMON ET AL.
TABLE 2
Effects of atrazine on the immune system of teleost fish
Species
Cyprinus carpio
Downloaded At: 15:43 28 October 2008
Liza auratus, Liza
ramada.
Salmonidae (species
not specified)
Oncorhynchus kisutch,
Salvelinus
namaycush
Endpoint
Humoral immune
response
Phagocytosis
Head kidney
macrophages
chemiluminesense
response
Replication of spring
viraemia of carp
virus
Macrophages (liver)
Effect
System
Exposure
No
In vivo
100–10,000
μg/kg BW
No
No
In vivo
In vitro
No
In vitro
Degeneration
In vitro/ex vivo ≥ 25μg/L
Atrophy
No effect/decrease
In vivo
In vivo
Number of
lymphocytes
No effec/decrease
In vivo
(Cossarinidunier
et al.; 1988)
7000–28,000
μg/L
28,000 μg/L
Number of endocytes Increase
In vitro/ex vivo ≥ 25μg/L
(liver)
Increase in
Increase
In vitro/ex vivo ≥ 25μg/L
melano-macrophage
centres (liver)
White blood cells
Decrease (leucopenia) In vivo
100–1000 μg/L
Lymphoid organs
Spleen weight
Reference
1,500–13,500
μg/L
(BiagiantiRisbourg;
1990)
(Walsh and
Ribelin; 1975)
(Zeeman and
Brindley; 1981)
BW = bodyweight, (Data from Dunier and Siwicki 1993).
1990). Furthermore, maximum exposures to atrazine occur in
environments such as small ponds or lowland streams that are
typically not inhabited by the more sensitive species such as
some salmonids or the euryhaline muglid species L. aurata and
L. ramada. Considering factors such as exposure likelihood and
maximum environmental concentrations, the risk of exposure to
atrazine compromising the immune system of fish in the wild appears to be small. To our knowledge, no ecotoxicological studies
are available that tried to link exposure to atrazine in the wild
with effects on the immune system, and therefore, the environmental relevance of possible immunological consequences of
the exposure of fish to atrazine remains unclear. In general, the
overall body of information regarding the effects of atrazine
on the immune system of fish is very scarce, and to be able
to conduct an appropriate risk assessment of the immunotoxic
properties of atrazine, additional studies are needed. These studies should specifically address the complexity and multiplicity
of immune responses as well as differences in species sensitivity, as currently only patches of information are available for
individual species, making it difficult or impossible to compare
results among studies.
B.
Effects of Atrazine on Immune Function
in Amphibians
There have been few original research papers on amphibians
that report on the potential effects of atrazine on the immune system of anurans (Table 3). Three of these studies tested effects of
mixtures of pesticides, making it difficult to characterize the contribution of atrazine to the observed results. In one study, exposure of R. pipiens to mixture of pesticides (metribuzin, aldicarb,
dieldrin, endosulfan, lindane, and atrazine, Table 4) resulted in
a decrease in magnitude of a number of responses (Christin et
al., 2003). Cellularity of frog splenocytes was determined before and after infection. A significant limitation with this data
set is that it was normalized and reported as percent of control
where 100% is arbitrarily assigned to both of the water control groups (i.e., infected and uninfected). Although the authors
state that the infected water control frogs exhibited a mean value
of 58.15 × 104 cells/ml, they failed to provide the mean value
of cells/ml for the uninfected water-control group. Therefore,
it is not possible to determine the change in spleen cellularity
due to parasitic infection. In addition, by reporting the results
as cells/ml, it is difficult, if not impossible, to determine the
751
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
TABLE 3
Effects of atrazine on the immune system of anurans
Species
Endpoint
Downloaded At: 15:43 28 October 2008
Rana pipiens
Cellularity of splenocytes(1)
Viability of splenocytes (1)
Number of phagocytic
splenocytes(1)
T-cell proliferation (1)
Phagocytic activity of
splenocytes(1)
R. pipiens
Thymic plaques
Thymic lymphocytes
R. pipiens
Formation of thymic plaques(1)
R. pipiens
Thioglycollate-stimulated
recruitment of white blood
cells to the peritoneal cavity
R. sylvatica
Number of white blood cells(3)
Xenopus laevis Cellularity of splenocytes(1)
Viability of splenocytes(1)
Number of phagocytic
splenocytes(1)
T-cell proliferation(1)
Phagocytic activity of
splenocytes (1)
Effect
System
Exposure
No
In vivo (2–2,100 μg/L) (4)
No
Decrease
In vivo (≥2 μg/L)(4)
Decrease
No
In vivo (≥2 μg/L) (4)
In vivo
Decrease
No effect
Increase
Decrease
In vivo
In vivo
In vivo
In vivo
Decrease (2)
Decrease
Decrease
Decrease
In vivo 3 and 30 μg/L
In vivo (≥210 μg/L) (4)
In vivo
In vivo
No
Increase
In vivo (2–2,100 μg/L) (4)
In vivo (≥21 μg/L) (4)
≥ 1μg/L (30 d)
≤ 10μg/L (60 d)
(0.1 μg/L) (5)
10 to 0.01 μg/L
Reference
(Christin et al.; 2003,
Christin et al.; 2004)
(Houck and Sessions; 2006)
(Hayes et al.; 2006a)
(Brodkin et al.; 2007)
(Kiesecker; 2002)
(Christin et al.; 2004)
Note. (1) Mixture study. Effect cannot be assigned to atrazine because other pesticides were tested in the mixture. (2) No information on
significance of effect given. (3) Not clear if effect is due to exposure to atrazine or increased infection with trematode cercaria. (4) Number
in brackets = concentrations of atrazine in mixture with 5 other pesticides. (5) Number in brackets = concentration of atrazine alone or in
mixture with nine other pesticides.
total number of spleen cells per frog. No significant pesticideassociated effects on spleen cellularity were observed in any of
the treatment groups.
The effect of the pesticide mixtures on T cell proliferative
responses induced by two separate mitogens, concanavalin A
(ConA) and phytohemaglutinin (PHA) were measured pre- and
postinfection. A significant confounding factor to this set of experiments is that both Con A or PHA produced a very weak
induction of T cell proliferation, which, at best, was 2-fold in
spleen cells from uninfected frogs. The modest proliferative responses suggest that both ConA and PHA have weak mitogenic
activity in frog T cells or that the conditions for this assay have
not been optimized. Regardless of the reasons, due to the extremely poor level of stimulation with each of the mitogens, it
is difficult to evaluate the biological relevance of the decreased
proliferation in the presence of pesticide exposure.
The authors also observed that the overall magnitude of proliferation was greater in mitogen-activated spleen cells from
infected frogs (author Figure 2B). However, even under these
conditions, only a twofold increase in proliferation (mitogen
stimulated versus no mitogen stimulation) was observed. In addition, that the spleen cells from infected frogs were modestly
TABLE 4
Nominal concentrations of pesticides used in studies on frogs
Mixture
0.1×
1×
10×
100×
Atrazine μg/L
Metribuzin (μg/L)
Aldicarb (μg/L)
Dieldrin (ng/L)
Endosulfan (ng/L)
2.1
21
210
2, 100
0.056
0.56
5.6
56
1.7
17
170
1, 700
0.015
0.15
1.5
15
0.002
0.02
0.2
2
Note. Data from Christin et al. (2004) and Gendron et al. (2003).
Lindane (ng/L)
0.033
0.33
3.3
33
Downloaded At: 15:43 28 October 2008
752
K. R. SOLOMON ET AL.
more responsive to mitogen-induced proliferation versus spleen
cells from uninfected frogs is not too surprising. Within the environment of an active immune response, the lymphocytes from
the infected frogs were likely primed as well as exposed to a milieu of growth-promoting cytokines. No statistically significant
effect on either PHA- or Con A-induced proliferation by spleen
cells isolated from infected frogs was observed due to pesticide
treatment at any of the dose groups. In addition, no statistically
significant pesticide-associated treatment effects were observed
on phagocytosis by spleen-derived phagocytic cells and no statistically significant pesticide-associated treatment effects were
observed on the prevalence of lung infection by R. ranae. However, since no tests were conducted during this study with individual compounds, it is impossible to separate possible effects
of atrazine from those that might have been caused by the other
chemicals.
In a second study of the same mixture on X. laevis and R.
pipiens, a significant decrease in total spleen cell number was
observed in X. laevis in the 10× and/or 100× treatment groups
(Christin et al., 2004). However, there was mortality (15 of 30
of frogs in the 100× and 2 of 30 in the 10× group). It is quite
common to observe a decrease in lymphoid organ cellularity
when overt toxicity is chemically induced. This can be attributed
to a variety of nonspecific effects, including stress. Therefore,
a major shortcoming of this study was that body and spleen
weights were not measured and reported in light of the indications that overt toxicity was being produced in some of the treatment groups. There was also a decrease in phagocytic activity in
spleen cells from R. pipiens, which occurred in the absence of a
decrease in spleen cellularity. This is contradictory to the results
from an identical study in 2003 (Christin et al., 2003) where no
effect on phagocytic activity was observed at any of the concentrations of the pesticide mixture. As with the previous study, no
tests were conducted with individual compounds. Therefore, it
is impossible to separate possible effects of atrazine from those
that might have been caused by the other chemicals.
A study on R. pipiens reported on the effects of exposure
to atrazine on the formation of hemolytic plaques and lymphocytes in the spleens of frogs challenged with sheep red blood
cells (Houck and Sessions, 2006). Frogs of unreported age and
size and from an unknown source were exposed to nominal concentrations of 0, 1, and 10 μg atrazine/L in glass containers of
unreported volume and kept at room temperature (unreported)
for 30 and 60 d. Four frogs were exposed to each concentration, except for the control in the 60-d study where only three
frogs were used. Exposure solutions were replaced every 3 d
but concentrations were not measured. At the end of the exposures, the frogs were injected intraperitoneally (ip) with the
equivalent of 0.5 ml of washed sheep red blood cells for 5 consecutive days. Frogs were then sacrificed, the spleen removed,
and flushed to collect cells for counting and for the hemolytic
plaque assay, performed with sheep red blood cells and guinea
pig complement (GPC). The results of the 30-d exposure (only
three of the four exposed frogs were used) showed a statistically
significant decrease in the formation of the hemolytic plaques
at 1 and 10 μg atrazine/L. In the 60-d exposure, treatments did
not produce scorable plaques (reasons not reported), counts of
lymphocytes and lymphocytes and red blood cells were not statistically different between control and treatments, and there was
no concentration response. The methods used in this study were
poorly reported, small numbers of animals were used, and the
same assay endpoints were not reported for the two exposure
periods. Thus, the results are essentially uninterpretable and do
not justify the very speculative and over-extrapolated discussion
in the article. At best, the results reported in this article are preliminary and a better designed study should be conducted to test
this hypothesis.
Another study reported effects of atrazine in combination
with exposure to trematode cercariae (Ribeiroia sp. and Telorchis
sp.) on eosinophil counts and susceptibility to successful encystation of cercariae in R. sylvatica (Kiesecker, 2002). The
only immunological parameter investigated was the number of
eosinophils in circulating blood. In light of the fact that the author’s goal was to draw a linkage between chemical exposure
and decreased immune competence, it is puzzling that information concerning immune status of the frogs was not included in
the paper. It appeared that the number of eosinophils inversely
correlated with the formation of meta-cercarial cysts, and to
increasing concentrations of atrazine. However, no statistical
comparisons were reported between the controls and atrazineexposed frogs, leaving the question open whether there was a
significant decrease in eosinophil counts due to atrazine exposure. Furthermore, blood samples for immunological analyses
were taken after infestation of frogs with cercariae, making it
impossible to determine whether the observed effect on the immune response was due to the infection or herbicide exposure.
The author’s conclusion that the pesticide exposure resulted in
a “dramatic effect” on the immune system would perhaps be
justifiable if eosinophils had been observed to be reduced in
pesticide-exposed frogs before exposure to cercaria, and this
reduction showed a consistent concentration-response. In the
absence of these data, it could be argued that the reduction in
eosinophils in the blood was a result of the cercarial infection.
Eosinophils are known to migrate from the blood vessels to sites
of parasite infection (Guyton and Hall 1996; Dhabhar et al.,
1993, 1994), and this may have been the cause of the apparent
reduction in numbers circulating in the blood.
The study on the effects of pesticide mixtures on frogs discussed earlier (Hayes et al., 2006a) also reported effects on the
thymus gland in R. pipiens. This effect was “discovered” because
of unexpected infections in the frogs with F. menigosepticum and
was not related to a specific hypothesis test, nor was the study designed to test this hypothesis. The authors reported an increased
incidence of “thymic plaques” in animals exposed to atrazine,
metolachlor (0.1 μg/L), the formulated mixture of atrazine and
metolachlor (0.1 μg/L atrazine), and the nine-pesticide mixture
(total concentration of 0.9 μg/L). Incidence was greatest (26%)
in the latter exposure. There is no reference to the literature that
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
would causally link the presence of “thymic plaques” to reduced
immunocompetence and it appeared to be a nonspecific response
related only to exposure to pesticides in general. There is no indication of whether this was a response to infection, the cause
of the infection, or whether the incidence of thymic plaques
increased at greater exposure concentrations, nor was the experiment designed to elucidate this. The extrapolation of this effect
to suggest that declines of amphibians caused by an “inability
to mount proper immune responses as a result of pesticide exposure” is overinterpretation of chance observations and needs
to be properly tested in a well-designed study.
A recent study reported the effects of atrazine on the innate
immune system in adult R. pipiens at exposure concentrations as
small as 0.1 μg/L (Brodkin et al., 2007). The frogs employed in
this study were caught in the wild and there was no knowledge of
prior exposure to environmental contaminants, a fact acknowledged by the authors. No analysis of tissues was performed at
the end of the study to assess the potential impact of exposure to
other agents may have had on the study results prior to captivity.
The experimental model consisted of maintaining 6 individual
frogs for 8 d at each exposure concentration in separate containers with 500 ml of aged tap water or atrazine-supplemented water, about 1 cm deep. Because the frogs were not fully immersed,
there may have been unquantified differences in exposures between replicates as well as between treatments. The sex of the
frogs was not reported. In measurements of innate immune responses, thioglycollate-induced peritoneal cells were employed.
Historically, thioglycollate has been used in many studies to increase the total number of phagocytes that can be isolated from
the peritoneal cavity. However, this approach is now not commonly employed, especially for assessment of immunotoxicity.
This is due to the well-known fact that, in addition to elicitation
of cells into the peritoneal cavity, thioglycollate elicitation also
activates phagocytic cells. The concern is that the thioglycollate
activation may confound the assessment of functional responses
by macrophages and other phagocytic cells. This is especially a
concern in the present study since phagocytosis is used to both
identify phagocytic cells as well as serve as a functional measure
of immune status.
The authors reported an almost complete inhibition of thioglycollate elicitation of leukocytes into the peritoneum by exposure of frogs to 21 μg/L atrazine. The same effect was observed
in a concentration response study (0.01–10 μg/L atrazine in
thioglycollate-stimulated frogs). Although 8 d of atrazine exposure markedly suppressed leukocyte recruitment into the peritoneal cavity of thioglycollate-stimulated frogs, there was no
assessment of the response of frogs to a range of concentrations
of atrazine alone, a logical question to ask since it had been
shown to be stimulatory at 21 μg/L. In addition, no attempt
was made to phenotype the composition of the thioglycollate
elicited cell exudates. Both of these omissions are major flaws
of the study.
The authors also characterized the percentage of phagocytes present in the cells retrieved from the peritoneal cavity.
753
These studies showed that after thioglycollate elicitation, approximately 23% of the cells in the exudate were phagocytes,
as assessed by their ability to take up fluorescent spheres, which
decreased to approximately 3% in frogs exposed to 21 μg/L
atrazine, approximately an 80% difference between the two
treatment groups. In contrast, frogs injected ip with Ringers
(control) possessed approximately 2% phagocytic cells in their
peritoneum, which increased to about 10% when treated with
atrazine. These results, at least without further investigation,
are somewhat paradoxical since atrazine in the presence of
thioglycollate-elicitation suppressed the percentage of phagocytic cells yet in the absence of thioglycollate (i.e., no elicitation)
increased the percentage of phagocytic cells. In fact, these results
serve as a good illustration of why thioglycollate-elicitation is
no longer commonly employed when assessing the effects of an
agent on innate immune function. The authors noted these paradoxical effects of atrazine in the discussion section and reiterate
that atrazine appears to stimulate the phagocytic activity of resident peritoneal cells while inhibiting phagocytic activity in the
thioglycollate-induced peritoneal cells. This may or not be the
case. Since it is unclear whether atrazine actually alters phagocytosis, whether atrazine alters the profile of cell types recruited
to the peritoneum in the presence of thioglycollate, or whether
thioglycollate-activated peritoneal cells are more susceptible to
modulation of phagocytic activity by atrazine than are resident
peritoneal cells, drawing conclusions from these results is highly
speculative at best.
The authors also presented data indicating that atrazine treatment increased the number of nonphagocytic cells after an ip
injection of thioglycollate. As already discussed, it is difficult to
discern the meaning of these results since “nonphagocytic” cells
are defined functionally (i.e., do not possess fluorescent spheres).
Therefore it is unclear whether the increase in nonphagocyic
cells following atrazine treatment is due to a change in the profile of cell types recruited to the peritoneum or whether there is no
change in the profile of cell types recruited but merely a decrease
in the phagocytic activity of the phagocytic peritoneal cells.
The authors’ primary conclusion from this study was that
atrazine treatment acts as an “innate immune response disruptor” (Brodkin et al., 2007). The conclusion is based on the observation that atrazine treatment suppressed thioglycollate-induced
recruitment of leukocytes into the peritoneum as assessed by cell
counts. In the context of the experimental conditions and the concentrations of atrazine employed, the effects appear to be real.
However, it is unclear whether the effects are directly mediated
by atrazine, which cell types are being affected, and/or whether
the effects are unique to thioglycollate elicitation. Moreover,
although the authors state in the discussion that thioglycollateinduced peritonitis is a common model to study inflammation,
it is not a common approach to assess the immunotoxicity of an
agent on innate immune responses, which was the stated objective for this study.
Many of the published studies on the effects of atrazine on
immune function in amphibians have suffered from flaws in the
754
K. R. SOLOMON ET AL.
Downloaded At: 15:43 28 October 2008
design, poor descriptions of methods, and the use of inappropriate techniques. Given the very limited information on the immunological effects of atrazine in aquatic organisms as a whole,
it is impossible at the current state of research to evaluate the
possible status of atrazine as an immunotoxic or suppressive
compound for amphibians. As for fish, there have been no efforts so far to determine possible links between the exposure
to atrazine and effects on the immune system in frogs from the
wild. Further studies using better designed protocols will be necessary to address the potential of atrazine to interfere with the
immune system of amphibians before any definitive conclusions
can be drawn.
XI. EFFECTS OF ATRAZINE ON BEHAVIOR
The ability of organisms to survive to reproductive age
requires that they make appropriate behavioral responses to
changes in their environment. Behavior is an important determinant of reproductive success and, as such, is an important
target for evolutionary selective forces, which can act to mold
and shape species-specific behaviors. As a result, it can be difficult to generalize about a contaminant affecting “behavior,” as
an organism’s response to changes in the environment may be
species-specific. Ultimately, any behavior requires that an organism detect changes in its environment, integrate this information
within the central nervous system, and then elicit appropriate
motor commands to carry out the behavior. Major challenges
remain in extrapolating from experimental data to populationlevel effects (Beebee and Griffiths, 2005), particularly utilizing
wildlife populations (Kendall and Lacher, 1994). In response
to potential behavioral effects, there are data in mammals suggesting that atrazine can act within the CNS by interacting with
receptors for the inhibitory neurotransmitter gamma aminobutyric acid (Shafer et al., 1999) or by altering monoamine turnover
(Das et al., 2000). There are only limited data on the potential
effects of atrazine on behavior in non-mammalian vertebrates.
The behaviors studied were not accompanied by neuroanatomical or pharmacological studies to identify mode of action and
what neurotransmitter pathways may be affected; thus the responses reported are difficult to characterize in relation to those
reported in mammals.
A.
Effects on Olfactory Neurons and Behavior in Fish
Goldfish exhibited altered burst swimming, grouping and surfacing behavior in response to small concentrations of atrazine
(Saglio and Trijasse, 1998). It is unclear whether the data were
reported as a percentage or animals responding to the treatments.
Moreover, the standard deviation measurements were large for
the measured endpoints reported to be significantly different
from controls, bringing into question whether appropriate statistical tests were used to analyze the data. For example, the
authors reported that “control” fish exhibited no burst swimming (0.00 ± 0.00, SD) whereas fish exposed to 0.5 or 50 μg
atrazine/L, but not 1 μg atrazine/L, exhibited greater burst swimming (3.38 ± 3.38 and 2.25 ± 3.15 for the 0.5- and 50-μg
atrazine/L groups, respectively). Although test concentrations
of atrazine in the exposure water were not confirmed analytically, the authors reported that baseline levels of atrazine in the
tap water reached levels as great as 0.235 μg atrazine/L. Thus,
there were no negative control groups employed in the study.
Another issue is the fact that some behaviors in “control” animals differed nearly 15-fold between experiments: a difference
that was greater than differences between atrazine treated and
control animals in individual experiments. For example, sheltering behavior in “control” animals ranged from 4.63 ± 3.46
to 1.50 ± 1.91 to 15.50 ± 10.28 in three different experiments.
Collectively, these issues make the Saglio and Trijasse (1998)
paper impossible to interpret.
A study by Moore and Waring (1998) on S. salar evaluated
the effects of atrazine on pheromonally induced olfactory responses including the measurement of plasma steroid levels.
Fish were exposed to graded concentrations of atrazine (0–20
μg/L) for 5 d and then were challenged with urine collected
from ovulated female salmon. Blood samples were collected
after 5 h, which is when the urine stimulated a pheromonallyinduced increase in plasma T, 11-ketotestosterone, or 17,20βdihydroxy-4-pregnen-3-one concentrations. Consistent with the
suppression of olfactory responses to pheromones by atrazine,
an exposure-dependent decrease in concentrations of plasma
steroids (17,20β-dihydroxy-4-pregnen-3-one, testosterone, and
11-ketotestosterone) was observed (testosterone shown only,
Fig. 12). The statistical analysis of the data was incorrect but
a concentration response was evident. In the same study, Moore
and Waring evaluated the in vitro release of free and conjugated
steroids from fish primed with ovulated female urine alone or in
combination with atrazine. These results failed to show a consistent effect of atrazine, although again the statistical analyses
were inappropriate. The mechanism for this was most likely
through a direct action of atrazine on olfactory neurons and not
a direct action on the endocrine system. As discussed later, the
explanation in the conclusion of the paper is not clear as it states
that “atrazine is a known inhibitor of acetylcholinesterase,” a
fact that is completely incorrect.
A second experiment (Moore and Lower, 2001) showed that
neither atrazine nor simazine, alone or in combination, appeared
to alter concentrations of testosterone, K11-ketotestosterone, or
17,20β-dihydroxy-4-pregnen-3-one, compared to control fish
that were also exposed to female priming pheromone (Fig. 13,
testosterone shown only). Although this was a key question to
ask, the authors did not actually test these differences for statistical significance. The authors reported statistical comparisons of
treatment groups relative to the ethanol-treated control when the
more appropriate comparisons are relative to fish treated with
PGF2a (which was common to all treatment groups). The previously reported effects of atrazine on the priming pheromone response of plasma testosterone, 11-ketotestosterone, and 17,20βdihydroxy-4-pregnen-3-one concentrations (Moore and Waring,
1998) were not observed with either atrazine or simazine alone
or in combination; however, smaller concentrations were tested
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
FIG. 12. Effect of atrazine on testosterone in mature male
salmon parr. Data represent the mean + SEM of five fish per
group (redrawn from data of Moore and Waring, 1998).
than in the previous study (Moore and Waring, 1998). It is interesting that the apparent effect reported is an increase in concentration of plasma testosterone induced by atrazine and simazine:
not a response consistent with the induction of aromatase. Inconsistency of responses and incorrect comparisons of treatments make these results difficult to interpret. Although effects
on olfactory responses were reported, these were not reflected
in changes in concentrations of the hormones measured in the
studies. Further, the inconsistency of these results makes their
significance at the population-level difficult to interpret.
The study by Moore and Waring (1998) on S. salar reported concentrationdependent effects of atrazine on electrophysiology of the olfactory epithelium at nominal concentrations from 2 to 20 μg/L, but not at 0.5 μg/L. However, no
analyses of exposure concentrations were conducted, no detail
of how these solutions were prepared, and no information was
given on whether solvents were used or not. Responses of the
endocrine system showed inconsistent concentration responses
but the authors did suggest that atrazine was affecting androgen metabolism. The authors reported that the priming effect on
milt and plasma 17,20β-dihydroxy-4-pregnen-3-one concentrations were reduced at water atrazine concentrations at and above
0.04 μg/L. The mechanism for this was not clear and some of the
statements in discussion are inconsistent with the known prop-
755
FIG. 13. The effects of the priming pheromone PGF2 alone and
in combination graded concentrations of simazine and atrazine
on plasma testosterone concentrations in mature male salmon
parr. Control parr were treated ethanol alone. Data represent the
mean + SEM of 7 fish/group (redrawn from Moore and Lower,
2001).
erties of atrazine. The authors suggest that atrazine is a highly
lipophilic substance that would bioaccumulate in the lipid-rich
testes. But atrazine is not highly lipophilic, and the data for
bioconcentration and bioaccumulation in fish (Giddings et al.,
2005) and discussed earlier do not support this suggestion.
It was reported that atrazine affected odor (amino acid Lhistidine)-evoked behavioral and neurophysiological responses
in rainbow trout (Oncorhynchus mykiss) (Tierney et al., 2007).
However, the actual exposure concentrations of atrazine in the
tank water were not measured in this study. In these experiments,
O. mykiss were allowed to acclimate to test troughs for 20 min
prior to a 30-min exposure to the pesticide. After the 30-min pesticide exposure, fish were exposed to L-histidine dissolved in the
water for 10 min on one side of the test trough (test side) and
the percent time spent on the test side of the trough measured.
Locomotor activity also was monitored for 10 min after the pesticide exposure. Exposure for 30 min to 100 but not 1 or 10 μg
atrazine/L reduced the preference/avoidance response ratio to Lhistidine, whereas 1 and 10 μg atrazine/L increased locomotor
activity (measured for 10 min after 30 min of atrazine exposure).
Electro-olfactograms (EOG) also were measured in response to
L-histidine as an indicator of olfactory neuron activity. Exposure
756
K. R. SOLOMON ET AL.
to 10 and 100 μg atrazine, but not 1 μg atrazine/L, reduced the
L-histidine-induced EOG. The population and reproductive significance of the reported behavioral effects are not clear. Three
reports indicate that atrazine may affect locomotor behavior.
Exposure for 24 h to 0.5 μg atrazine/L reportedly caused a significant increase in burst swimming reactions in goldfish (Saglio
and Trijasse, 1998).
Downloaded At: 15:43 28 October 2008
B.
Effects on Behavior in Amphibians
Although exposures of A. barbouri to atrazine in the range
of 4–400 μg/L for 37 d had no effect on embryo survival or
growth, Rohr et al. (2003) reported that exposure to the greatest
concentration (400 μg/L) adversely affected antipredator behavior, which consisted of seeking refuge in response to a potential
threat. Exposure of A. barbouri to >40 μg atrazine/L during
the larval period resulted in postmetamorphic salamanders that
displayed greater locomotor activity, fewer water-conserving behaviors, and greater desiccation (water loss) 4 and 8 months later
(Rohr and Palmer, 2005). Although the control of motor patterns
in urodeles is entirely dependent upon subcortical neuronal pathways and is, arguably, relatively primitive compared to mammals, the number of factors that can modulate motor command
neurons and premotor areas of the brainstem and spinal cord are
numerous and complex. The possible mechanisms underlying
these reported effects and whether a direct effect of atrazine on
nervous system activity is even plausible in these species are
unknown. Because of our relative lack of understanding of role
of behavior in amphibian field biology it is difficult to interpret
the potential effects of atrazine, if any, on amphibian behavior.
Overall, although several studies have reported effects of
atrazine on the olfactory system in salmon, in many cases the responses are inconsistent between studies and do not show clear
concentration-related responses. In at least one study, the behavioral responses are difficult to interpret based upon inadequacies
in the experimental design (lack of a negative control). Effects
on behavior of salamanders have been reported at concentrations ranging from 40 to 400 μg atrazine/L. The population and
reproductive significance of the reported behavioral effects are
not clear.
XII.
EFFECTS OF ATRAZINE AT THE POPULATION
LEVEL
Any contaminant that adversely alters biochemistry, physiology, development, reproduction, or behavior may alter reproductive success and subsequently affect population health. However,
many of these responses do not result in ecologically relevant
changes that affect the fitness of a population or its sustainability. Specifically, if the response does not affect the survival,
growth, or reproduction of the individuals in a population, there
would be no adverse effects on populations. In many cases, the
changes at these levels of organization are adaptive responses
that allow the individuals to adapt to exposures to stressors in a
manner that does not translate into population-level effects. In
addition, there are a number of density-dependent interactions
between individuals in populations, between populations, and
with the environment that they occupy. For this reason, subtle
effects of a substance or effects on a small number of individuals
would not necessarily be expected to result in a negative effect
on the population. Finally, populations of animals are dynamic
and fluctuate in response to the most important determinants
of population sustainability—food supply, weather conditions,
and habitat quality and quantity. This makes identification of
clear effects difficult and, without clear and consistent effects,
causality is very difficult to assign.
There are no published population studies to suggest that
atrazine is associated with declines in populations of amphibians, reptiles, and fish. Indeed, frogs from every continent where
amphibians occur have suffered high mortality rates in recent
times. It is generally agreed that no single cause can be invoked to explain the loss of diverse amphibian populations from
several continents. The effect of habitat change on amphibian
populations has been known for years but our understanding
of the effects of chemicals on amphibians is comparatively recent. Chemical pollution does not totally explain the global amphibian declines in often montane areas or where agricultural
chemicals are not used, and in recent years attention has shifted
to five other possible stressors: increased exposure to UV radiation, direct exploitation, the spread of alien species, climate
change, and emerging diseases. Chytridiomycosis, an emerging
infectious disease, was identified as the most likely cause of
massive population declines of frogs on multiple continents and
is spreading globally, and rapidly.
A.
Atrazine and Reptiles
There are no published population studies or other data to
suggest that atrazine is associated with declines in the populations of reptiles. Even though atrazine has been reported to affect
a steroidogenic enzyme in alligators in Lake Apopka, Florida,
the researchers who reported these observations concluded that
atrazine did not have any endocrine-disrupting effects on populations of alligators in that lake (Crain et al., 1997, 1999, and
discussed earlier). The declines in populations of alligators in
some states of the United States have been linked to excessive
hunting and/or changes in habitat (Rhodes, 1998) and not to
exposures to atrazine or other substances in the environment.
Alligator populations in North Carolina have, in fact, recovered
where hunting has been restricted (Rhodes, 1998). Based on this
observed recovery, hunting in North Carolina was most likely
responsible for the declines.
B.
Atrazine and Fish
While some populations of fishes are threatened or in decline,
most of these are the direct result of over-harvesting but some,
such as salmon, are because of habitat alteration. As was the case
for reptiles, there are no published studies of populations of fish
that suggest that atrazine is associated with declines of fish. No
adverse effects were observed in a number of microcosm studies
of the effects of atrazine, except in situations where indirect
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
effects occurred through reduction in food supply for grazing fish
or alterations in habitat through loss of macrophytes (reviewed
in Solomon et al., 1996; Giddings et al., 2005).
Downloaded At: 15:43 28 October 2008
C.
Atrazine and Amphibians
Frogs from every continent where amphibians occur have
suffered high mortality rates in recent times, resulting in various species undergoing severe population reductions and some
even becoming extinct (U.S. Fish and Wildlife Service, 1984;
Fellers and Drost, 1993, Ingram and McDonald, 1993; Hines
et al., 1999; Houlahan et al., 2000; Stuart et al., 2004; Mendelson
et al., 2006). Before considering the limited data on atrazine and
amphibian populations, it is valuable to briefly summarize what
is known about the causes underlying amphibian declines and
the evidence supporting a role of pesticides in these declines.
Historical data indicate that amphibian declines began in the
1970s in northern Australia (Czechura and Ingram, 1990), the
western United States (Sherman and Morton, 1993; Drost and
Fellers, 1996), and Puerto Rico (Burrowes et al., 2004). Many
declines took place in seemingly pristine and often montane areas (Pounds et al., 1997; Pounds and Crump, 1994; Young et al.,
2001). The first reports of amphibian declines were received with
skepticism since amphibian populations often fluctuate widely
(Pechmann and Wilbur, 1994). A recent report from the IUCN’s
Global Amphibian Assessment suggests that as many as a third
of amphibian species (>5700) have undergone severe declines
or extinction with many species on the brink of extinction (Stuart et al., 2004). Today it is generally agreed that no single cause
can be invoked to explain the loss of diverse amphibian populations from several continents. Various amphibian species have
specific habitat requirements and habitat loss has driven various
species to the brink of extinction. For example, loss of migration
routes has been implicated as a major cause for the reductions
in the populations of two North American Bufonids to single
populations (Bufo baxteri and Bufo houstonensis; U.S. Fish and
Wildlife Service, 1984, 2001).
Although the effects of, for example, habitat change have
been known for many years, our understanding of the effects of
chemicals on amphibians is comparatively recent (Collins and
Storfer, 2003). Various studies have been conducted at the laboratory, microcosm, mesocosm, and field enclosure level, but
proving deleterious effects of pesticides at the population level
is a difficult problem (Beebee and Griffiths, 2005). Although the
effects of atrazine treatment on aquatic communities have been
well studied in a number of microcosm experiments (summarized in Giddings et al., 2005), few experimental studies of the
effects of atrazine on amphibians at the population-level have
been reported. One microcosm study reported on the effects of
atrazine on community and food-web structure in small microcosms (11.3 L) containing Rana sylvatica tadpoles (Rohr and
Crumrine, 2005). The treatment concentrations for atrazine in
this study were two applications of 25 μg/L each spaced 2 weeks
apart. Although exposure concentrations were not measured, this
likely produced a nominal final concentration of 50 μg/L. The
757
authors reported direct effects of the atrazine treatment on the
abundance of periphyton with indirect effects on chironomids,
snails, and tadpoles. The effects on algae are not surprising at the
stated nominal concentration—they are sensitive and the overall NOECcommunity reported in the review of over 20 microcosm
studies was of the order of 20 μg/L. That grazers of phytoplankton may have been affected is also not unexpected but the
relevance of this larger systems and the field should be tempered by the lack of realism in the size of the microcosms and
the lack of power inherent in small systems with few replicates
(Sanderson, 2002).
Sophisticated analyses taking historical pesticide application
data into account have strongly linked organophosphate and carbamate pesticides in agricultural use with the declines of four
Californian anurans (Davidson, 2004). Another potential hazard that has been debated is pH shifts due to acid rain; however,
there is no evidence to link acidification to amphibian declines
(Vertucci and Corn, 1997).
Chytridiomycosis, an emerging infectious disease, was identified as the most likely cause of massive population declines
of frogs in Australia, New Zealand, Spain, Tanzania, and Meso
America. Furthermore, recent evidence suggests that chytridiomycosis is spreading rapidly, sometimes resulting in the decline or disappearance of rare and endemic amphibians. This
fungal disease was first described from moribund and dead amphibians that were collected at sites of mass deaths of frogs in
Australia and Panama from 1993 to 1998 (Berger et al., 1998).
The chytrid that infects Australian and Central American amphibians is similar in morphology, while analysis of zoospore ultrastucture and 18s rDNA sequence data placed the fungus in the
order Chytridiales (Berger et al., 1998). The chytrid was subsequently described as a new genus and species, Batrachochytrium
dendrobatidis (Longcore et al., 1999). Batrachochytrium dendrobatidis has low host specificity and is likely to infect any
species of amphibian as infections have been detected globally
in 15 amphibian families that include 94 species (Speare, 2001).
Amphibian chytridiomycosis is an emerging infectious disease
of amphibians and has been recognized as such on a global
scale (Daszak et al., 1999, 2003; Mendelson et al., 2006) and
was nominated for listing as a key threatening process under
the Environment Protection and Biodiversity Conservation Act
1999 of New Zealand (Speare, 2001).
Given the diverse habitats in which amphibian declines have
been observed, a great deal of attention has been directed at
prioritizing and classifying potential causes for declines. According to Collins and Storfer (2003), Class I hypotheses include factors such as habitat loss, introduction of nonindigenous
species, and overexploitation and collection of amphibians. Normal variation in breeding success, which can be linked to the
stochastic nature of rainfall and moisture availability, coupled
with isolation of certain populations by reduction in ecological
corridors, may be a primary cause of extinctions at the population level (Richter et al., 2003). Given the rapid urbanization
of southern California and the limited natural water sources in
758
K. R. SOLOMON ET AL.
TABLE 5
Relative sensitivity of reproductive endpoints to atrazine in fish and amphibians–laboratory studies
Downloaded At: 15:43 28 October 2008
Endpoint
Species
Gonadal effects
Testicular ovarian R. clamitans
follicles
R. pipiens
R. pipiens
X. laevis
X. laevis
X. laevis
X. laevis
X. laevis
X. laevis
X. laevis
Intersex
C. auratus
P. promelas
R. clamitans
R. pipiens
R. pipiens
X. laevis
X. laevis
X. laevis
X. laevis
X. laevis
X. laevis
X. laevis
Gonadal
R. pipiens
dysgenesis4
X. laevis
Spermatogenesis C. auratus
P. promelas
P. promelas
R. pipiens
Oogenesis
Seminiferous
tubule diameter
Sex ratio
GSI
Atrazine-specific effect Concentration response LOEC (μg/L)1
Reference
No
No
>10
(Coady et al.; 2004)
Yes
No
No
No2
No
No
No
No
No
No
No
No
Yes
No
Yes
Yes
No3
No
No
No
No
Yes
No
No
No
No
No
No
No
No
No
No
No
No
No
Not tested
No
No
No3
No
No
No
No
No
<0.1
>15
>25
>25
>25
>100
>31
>25
>100
>859
50
>28
<0.1
>15
<0.1
<0.1
25
>25
>100
>31
>100
<0.1
(Hayes et al.; 2003)
(Orton et al.; 2006)
(Hayes et al.; 2002)
(Carr et al.; 2003)
(Coady et al.; 2005)
(Kloas et al.; 2008)
(Jooste et al.; 2005)
(Du Preez et al.; 2008b)
(Oka et al.; 2008)
(Spanó et al.; 2004)
(U.S. EPA 2005)
(Coady et al.; 2004)
(Hayes et al.; 2003)
(Orton et al.; 2006)
(Hayes et al.; 2002)
(Hayes et al.; 2006b)
(Carr et al.; 2003)
(Coady et al.; 2005)
(Kloas et al.; 2008)
(Jooste et al.; 2005)
(Oka et al.; 2008)
(Hayes et al.; 2003)
Yes
No
No
No
No
25
>859
>44
>224
15
(Carr et al.; 2003)
(Spanó et al.; 2004)
(Bringolf et al.; 2004)
(U.S. EPA 2005)
(Orton et al.; 2006)
Yes
103
(Spanó et al.; 2004)
P. promelas
P. promelas
Yes
No
No
No
Yes, slightly
accelerated
spermatogenesis
Yes, follicular
atresia
No
Yes, decrease
No
Yes
>224
224
(U.S. EPA 2005)
(U.S. EPA 2005)
R. clamitans
R. pipiens
R. pipiens
X. laevis
X. laevis
X. laevis
X. laevis
X. laevis
X. laevis
X. laevis
C. auratus
P. promelas
No
No
No
No
No
No
No
No
No
Yes
No
No
No
No
No
No
No
No
No
No
No
Yes
No
No
>28
>25
>15
>200
>19
>25
>25
>100
>31
10
>859
>44
(Coady et al.; 2004)
(Hayes et al.; 2003)
(Orton et al.; 2006)
(Hayes et al.; 2002)
(Carr et al.; 2003)
(Coady et al.; 2005)
(Du Preez et al.; 2008b)
(Kloas et al.; 2008)
(Jooste et al.; 2005)
(Oka et al.; 2008)
(Spanó et al.; 2004)
(Bringolf et al.; 2004)
C. auratus
759
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
TABLE 5
Relative sensitivity of reproductive endpoints to atrazine in fish and amphibians–laboratory studies (Continued)
Endpoint
Species
P. promelas
X. laevis
Downloaded At: 15:43 28 October 2008
Plasma sex steroids
Plasma estradiol
Atrazine-specific effect Concentration response LOEC (μg/L)1
No
Yes, increase
C. auratus
Yes, increase
M. salmoides Yes (formulated
product only)
P. promelas No
X. laevis
No
X. laevis
No
Plasma testosterone C. auratus
Yes, decrease
M. salmoides No
P. promelas No
X. laevis
Yes, decrease
X. laevis
No
X. laevis
Yes, decrease
Plasma
C. auratus
Yes, decreased
11-ketotestosterone
M. salmoides Yes, decrease
P. promelas No
Induction of
X. laevis
No
aromatase CYP19
mRNA expression
X. laevis
No
X. laevis
No
X. laevis
No
Secondary sex effects
Laryngeal dilator
X. laevis
Yes, reduced
muscle
diameter
X. laevis
No
X. laevis
No
Fecundity
P. promelas No
Fertilization
P. promelas No
Hatching success
P. promelas No
X. laevis
No
Plasma vitellogenin C. auratus
No
P. promelas No, when
compared to
vehicle controls
P. promelas No
M. salmoides No
X. laevis
No
Transgenerational X. laevis
No
effects
1
Reference
No
No
>224
12
(U.S. EPA 2005)
(Hecker et al.; 2005a)
Yes
No
1000
100
(Spanó et al.; 2004)
(Gross et al.; 1997)
No
No
No
Yes
No
No
Not tested
No
Yes
Yes
>224
>259
>100
1000
>100
>224
25
>100
259
1000
(U.S. EPA 2005)
(Hecker et al.; 2005b)
(Hecker et al.; 2005a)
(Spanó et al.; 2004)
(Gross et al.; 1997)
(U.S. EPA 2005)
(Hayes et al.; 2002)
(Hecker et al.; 2005a)
(Hecker et al.; 2005b)
(Spanó et al.; 2004)
No
No
No
50
>224
>259
(Gross et al.; 1997)
(U.S. EPA 2005)
(Hecker et al.; 2005b)
No
No
No
>100
>259
>100
(Hecker et al.; 2005a)
(Hecker et al.; 2005b)
(Oka et al.; 2008)
No
1
No
No
No
No
No
No
No
No
>20
>25
>44
>44
>44
>25
>859
>44
(Carr et al.; 2003)
(Coady et al.; 2005)
(Bringolf et al.; 2004)
(Bringolf et al.; 2004)
(Bringolf et al.; 2004)
(Du Preez et al.; 2008b)
(Spanó et al.; 2004)
(Bringolf et al.; 2004)
No
No
No
No
>224
>100
>1000
>25
(U.S. EPA 2005)
(Gross et al.; 1997)
(Oka et al.; 2008)
(Du Preez et al.; 2008b)
(Hayes et al.; 2002)
As reported by the author, nominal or measured.
No TOFs were observed in the atrazine-exposed frogs; however, they were observed in estradiol-exposed positive control frogs. In Table 2
of Hecker et al. (2006), the presence of TOFs in atrazine-exposed frogs is incorrect.
3
Based upon inability to determine phenotypic sex from the physical appearance of gonads in 4.7% of the animals. Histological evaluation
revealed no mixture of testicular and ovarian tissue in these animals.
4
For the purposes of this review, gonadal dysgenesis also includes the term “discontinuous testes.”
2
760
K. R. SOLOMON ET AL.
TABLE 6
Relative sensitivity of reproductive endpoints to atrazine in amphibians—Field studies
Endpoint
Larynx weight
Downloaded At: 15:43 28 October 2008
Gonadal
anomalies
Testicular cell
types
GSI
Sex ratio
Species
Response
X. laevis
Correlation with Concentration
atrazine
-response
Reference
No relationship between atrazine
use and relative larynx mass in
males and females.
R. pipiens
Hermaphroditism in frogs from
sites with atrazine based on a
single measurement in surface
water at time of frog collection.
X. laevis
Gonadal anomalies in males and
females.
R. clamitans
Gonadal anomalies in males and
females.
R. catesbeiana Gonadal anomalies in males and
females.
A. crepitans
Intersex. No temporal relationship
between incidence and historical
use of atrazine.
X. laevis
No difference between exposed and
reference sites.
R. catesbeiana
No
No
(Smith et al.;
2005)
No
No
(Hayes et al.;
2003)
No
No
No
No
No
No
No
No
No
No
No
No
X. laevis
No
No
R. catesbeiana GSI or testicular cell types in males.
No
No
X. laevis
Sex ratio.
No
No
R. catesbeiana Sex ratio.
No
No
No differences between exposed
and reference.
Naturally present in one haplotype
and absent in another.
R. clamitans
No differences between exposed
and reference.
Reported to occur well before the
introduction of atrazine.
R. catesbeiana No differences between exposed
and reference.
No
No
No
NA
No
No
No
No
(Smith et al.;
2005)
(Murphy et al.;
2006a)
(Smith, 2007 pers.
com.)
(Reeder et al.;
1998, Reeder et
al.; 2005)
(Smith et al.;
2005)
(Smith, 2007 pers.
com.)
(Hecker et al.;
2004)
(Smith, 2007 pers.
com.)
(Du Preez et al.;
2005b)
(Smith, 2007 pers.
com.)
(Smith et al.;
2005)
(Du Preez et al.;
2008a)
(Murphy et al.;
2006a)
(Witschi 1929)
No
No
R. pipiens
No
No
No
No
No
No
GSI in males and females.
Testicular ovarian X. laevis
follicles
No differences between exposed
and reference.
Different between agricultural and
non agricultural sites
Found in males in sites where
atrazine exposure may have
occurred during development1 .
(Murphy et al.;
2006a)
(Smith, 2007 pers.
com.)
(Murphy et al.;
2006a)
(McDaniel et al.;
2008)
(Hayes et al.;
2003)
(Continued on next page)
761
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
TABLE 6
Relative sensitivity of reproductive endpoints to atrazine in amphibians—Field studies
Endpoint
Aromatase
Plasma
testosterone
Downloaded At: 15:43 28 October 2008
Plasma estradiol
Species
Correlation with Concentration
atrazine
-response
Response
Acris crepitans No temporal response, found before
and after introduction of atrazine.
X. laevis
No differences between exposed
and reference sites
R. clamitans
X. laevis
R. clamitans
No
NA
No
No
No
No
Testosterone in plasma of females.
Yes
Negative
Testosterone in plasma of males
and concentrations of the atrazine
metabolite (DACT) in water.
Plasma testosterone.
Yes
Negative
No
No
R. catesbeiana Plasma testosterone.
No
No
X. laevis
Plasma estradiol and atrazine.
No
No
Yes
Negative
R. clamitans
Plasma estradiol and DEA
concentrations.
Plasma estradiol and atrazine.
No
No
Yes
Negative
No
No
Plasma estradiol and DEA
concentrations.
R. catesbeiana Plasma estradiol and atrazine use.
Reference
(Reeder et al.;
2005)
(Hecker et al.;
2004)
(Murphy et al.;
2006a)
(Hecker et al.;
2004)
(Hecker et al.;
2004)
(Murphy et al.;
2006b)
(Smith, 2007 pers.
com.)
(Hecker et al.;
2004)
(Murphy et al.;
2006b)
(Smith, 2007 pers.
com.)
Note. Pers. com., personal communication.
1
Atrazine concentrations were measured when frogs were collected and exposures during development (if any) were not measured.
the western United States, it is not surprising that 3 of the 10
anuran species currently listed as endangered or threatened in
North America inhabit California (U.S. Fish and Wildlife Service, 2003). The introduction of aggressive anuran species such
as bullfrogs also has been linked empirically to adverse effects
on survival in the California Red-legged frog (Rana aurora draytonii) (Lawler et al., 1999; Department of the Interior Fish and
Wildlife Service, 2000). In a long-term study, Vredenburg et al.
(2004) demonstrated that the introduction of trout, which prey
on larval anurans, into mountain ponds in the Sierras was responsible for the decline of mountain yellow-legged frogs (Rana
mucosa).
Class II hypotheses for amphibian declines include global
changes in climate (UV radiation, global warming), emerging diseases (such as the chytrid fungus), and contaminants
such as pesticides and industrial waste products (Berger et al.,
1998). Amphibians have a thin and permeable integument and
undergo embryonic development in eggs with relatively little
protection from chemicals in the aquatic environment. Some
have speculated that amphibians may be sensitive indicators
of contaminant exposure (see Blaustein and Johnson, 2003)
because of their unique morphology and life history patterns.
However, the role of contaminants in amphibian declines has
been hotly debated, especially since many reports of declining
species have occurred in areas that should be protected, at least
in theory, from widespread agricultural or industrial contamination (USGS, 2004). There are some data linking aerial drift of
organophosphorus and other pesticides with amphibian declines
in California and Costa Rica. Sparling et al. (2001) reported
that surface waters in Sequoia National Park at an elevation
(2000+ m) that had been associated with declining frog populations contained greater than 100 ng/L chlorpyrifos and greater
than 65 ng/L diazinon. Furthermore, tree frog (Hyla regilla)
tadpoles collected from populations in Sequoia and Yosemite
National Parks, and located downwind from agricultural areas
in the Sacramento and San Joaquin valleys, had body burdens
of chlorinated pesticide residues 2–3 times greater than tadpoles
from coastal areas of California (Sparling et al., 2001). Although
chlorpyrifos, malathion, and diazinon, with 24-h LC50s of 2140
to 7490 μg/L, are not highly toxic to frogs at the measured
Downloaded At: 15:43 28 October 2008
762
K. R. SOLOMON ET AL.
environmental exposures to Rana boylii, their oxon metabolites
were reported to be 10 to 100-fold more toxic (Sparling and
Feller, 2007). Atmospheric transportation of current-use pesticides into montane regions of Costa Rica has been suggested
as potential risk to wildlife (Daly et al., 2007), but no specific
effects of these pesticides have been identified.
There are only limited data on the potential relationship between atrazine and amphibian population dynamics. Amphibians inhabiting ponds on agricultural land in Minnesota and
exposed to atrazine (0.1–0.5 μg/L) and de-ethyl atrazine (0.1–
0.3 μg/L) concentrations 5-fold greater than those reported to
produce gonadal effects (Hayes et al., 2003) exhibited no differences in species richness or reproductive success (Knutson
et al., 2004). This suggests that, if any gonadal anomalies exist, they do not appear to have effects at the population level.
Du Preez et al. (2005b) examined populations of X. laevis inhabiting maize-growing areas with atrazine application versus
non-maize-growing areas in South Africa and found no differences in several aspects of population structure including age
and size classes. A study in R. catesbeiana in Iowa revealed no
relationship between a number of population and reproductive
parameters and concentrations of atrazine in the ponds (Smith,
2007 personal communication). Although the data are limited,
the studies that are available to date do not make a compelling
case for population-level effects of atrazine in amphibians. Additional well-controlled studies are needed before any conclusions regarding potential population-level effects of atrazine on
amphibians can be reached.
XIII.
OVERALL CONCLUSIONS, AND RESEARCH
DIRECTIONS
A. Strengths and Uncertainties
We have identified significant strengths as well as uncertainties in the studies we have reviewed. Strengths relate to experimental designs where a range of concentrations were tested and
where exposure concentrations were verified. Strengths are also
apparent in large numbers of animals used in some studies and in
the consistency of observations of response across several laboratories. Several studies were carried out under full good laboratory practice (GLP) guidelines with quality assurance and quality control (QA/QC). Several others were conducted in the spirit
of GLP with QA/QC. Two studies were conducted to specifically address issues raised by a U.S. EPA Science Advisory
Panel (U.S. EPA, 2003b) and were conducted under full GLP
QA/QC in separate laboratories, one in the United States, the
other in the European Union (EU) (Kloas et al., 2008).
Additional strengths are in the comprehensiveness of the totality of the studies. These include a good understanding of the
exposure scenarios resulting from the use of atrazine, concentrations present in surface waters, and knowledge of the pharmacokinetics of uptake and depuration of atrazine in fish and amphibians. In terms of effects on the individual organism, there
were data on acute lethality, physiological effects, and induc-
tion of developmental abnormalities. In terms of reproductive
endpoints, there were data for effects on reproduction and sexdependent processes, such as sex differentiation, sexual and gonadal development, and secondary sexual characteristics, as well
as hormone titers, in exposed animals. There were also data related to putative mechanisms mediated through aromatase. In addition to effects mediated through the reproductive system and
associated hormones, there have been some studies on stress
physiology, immune function, and behavior. There were also
some observations on the effects of atrazine at the population
level that serve to integrate effects at all other levels.
The strengths of the available data were tempered by uncertainty at several levels. Uncertainties in exposures were evident
in several laboratory studies where nominal concentrations were
used or exposures were not characterized analytically. In other
studies, there was contamination of controls with low concentrations of atrazine. The same uncertainty was evident in some
field studies where there was no temporal characterization of exposure concentrations. There were uncertainties resulting from
poor experimental designs, such as the testing of too few concentrations, use of inappropriate methods for mixture studies,
lack of clarity and/or complete description of methods, and, in a
few cases, inconsistencies in the data between published studies
on the same experiments. In some cases, incorrect or inappropriate statistical methods and comparisons were used. An important
potential uncertainty was identified relating to variation between
haplotypes within a species in terms of background incidence of
responses. In addition to these, there were uncertainties in several overarching issues that include a lack of understanding of
the relevance of some physiological responses at higher levels of
organization, such as on populations. These relate in particular
to responses occurring at low frequency, which do not show a
monotonic concentration response, and also occur naturally.
B.
Conclusions
The primary focus of these conclusions is on the putative effects of atrazine on reproduction in aquatic vertebrates. Effects
on stress, behavior, and effects at the population level were discussed earlier. In formulating our conclusions, we used a subset
of the Bradford–Hill guidelines (Hill, 1965) as modified to assess causality of endocrine-modulated effects (IPCS, 2002) and
reproductive responses to atrazine. These guidelines are based
on temporality; strength of association; consistency; biological
plausibility; and recovery. The effects of atrazine on aquatic
species tested under laboratory conditions summarized in
Table 5 and those for field studies in Table 6. Some studies
(such as those of Tavera-Mendoza, discussed earlier) have been
omitted from these tables because of significant concerns about
the quality of the data.
Temporality
In terms of temporality, the study on A. crepitans in Illinois
showed the presence of indicators of effects in sexual development (TOFs) prior to the introduction of atrazine to the market
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
Downloaded At: 15:43 28 October 2008
in 1957 (Reeder et al., 2005). In addition, TOFs were observed
decades prior to the introduction of atrazine in other species
of frogs from other locations (Witschi, 1929). The number and
frequency of occurrence of TOFs decrease in X. laevis as frogs
mature, whether in the presence of atrazine or not (Jooste et al.,
2005; Du Preez et al., 2008a). This same phenomenon had been
observed earlier (Gallien, 1974). Thus, there is no temporal evidence of any association between atrazine and reproductive effects as indicated by the presence of TOFs in frogs.
Strength of Association
Strength of association is best assessed by concentration response. This is particularly relevant and one of the key theories as to the mechanism of action of atrazine in that it causes
changes in concentration of estradiol and testosterone by causing
induction of aromatase (Hayes et al., 2003). In work on tissue
cultures, this induction has been observed to show a consistent, monotonic concentration response (Sanderson et al., 2000).
Thus, if the aromatase theory is correct, responses mediated via
this mechanism should increase with increasing atrazine concentration. As summarized earlier, very few laboratory studies
have reported a monotonic concentration response to atrazine.
In terms of frequency and number of TOFs, zero of nine studies
reported monotonic concentration responses. Using gross morphology of testes as an endpoint, 1 of 10 studies reported a concentration response and this was only evident at the greatest concentration tested (25 μg/L). For sex ratio and gonado-somatic
index, 1 of 10 and 0 of 4 studies reported a concentration response, respectively. Of four studies on spermatogenesis, only
one reported effects related to atrazine exposure—an acceleration of the process. This effect did not show a concentration
response. One study reported a concentration-related decrease
in seminiferous tubule diameter in the fish, P. promelas. Oogenesis, which is apparently unaffected by atrazine in frogs,
responded to atrazine in C. auratus but not in P. promelas. No
effects were observed on fecundity, fertilization, and hatching
success in P. promelas exposed to 44 μg/L and no effects on
these processes were observed in the F2 generation of X. laevis
exposed throughout their lifespan to atrazine at concentrations
as great as 25 μg/L. Similarly, larynx size in male frogs, a developmental process mediated by androgens, was reported to show
an atrazine-related effect in only one of three laboratory studies but with no concentration-response and was not observed in
the field (Table 6). Overall, these studies show poor strength of
association between atrazine exposure and concentration for a
number of reproductive and developmental endpoints. This is
further evidenced by the presence of robust populations of amphibians in areas where atrazine is widely used and is present
in surface waters. Given the number of studies done, there is no
evidence of any atrazine-related effect.
Biochemical endpoints related directly to aromatase responded equivocally. In the fish, C. auratus, plasma testosterone,
and 11-ketotestosterone decreased while plasma estradiol in-
763
creased. However, this was only observed at large concentrations (>859 μg/L), and no effects were observed at the highest
concentration tested (224 μg/L) in another fish, P. promelas.
Atrazine was reported to cause a decrease in plasma testosterone in male X. laevis in one study (based on 4 animals) at
25 μg/L and also at 259 μg/L in a more robust study (15 males)
but not at 100 μg/L (42 males). The change in plasma testosterone was not correlated with changes in aromatase activity,
which remained the same as the control at 100 and 259 μg/L
and plasma estradiol was not increased (as would be expected if
the aromatase theory was correct) at 100 or 259 μg atrazine/L.
Although testosterone concentrations in plasma appeared to be
affected only at large exposure concentrations (259 μg/L), these
were not accompanied by changes in aromatase or increases in
plasma estradiol. This suggests that another mechanism may
have been responsible, but only at large exposures. Atrazine is
rapidly metabolized in frogs (Edginton and Rouleau, 2005) and
also induces mixed-function oxidases in cell lines (Oh et al.,
2003) and in frogs (Murphy et al., 2006c). Thus, the decrease in
testosterone and some other steroids at large exposure concentrations may be the result of increased rates of degradation but
is certainly not consistent with the aromatase-induction theory.
Consistency
There is consistency in the studies that have reported on the
effects of atrazine on reproductive development in amphibians.
The common and consistent theme in most of studies (Table 5
and Table 6) is that atrazine has no effects on reproduction or reproductive development. With rare exceptions, the only studies
that report adverse effects on amphibian development and reproduction are those from the Hayes laboratory. Even these are
not internally consistent; for example, the initial observation of
effects on larynx size has not been reported from any other study.
Some effects have been reported in fish, but the consistency here
is that the effects are only observed at large concentrations that
are not relevant to those measured in the environment.
Biological Plausibility
There is no evidence that atrazine itself (or its metabolites)
act directly at hormone receptors such as those for estrogen or
thyroid hormones. The theory of aromatase induction would
result in increases in titers of endogenous estradiol and is biologically plausible. It has been shown to apply at large concentrations in cancer cell lines tested in vitro. However, in amphibians and fish, it is not supported by experimental observations,
either of increased transcription of mRNA (some using very
sensitive methods), induction of aromatase activity, or changes
in the ratios of testosterone and estradiol in exposed animals.
Thus, this is not a mechanism by which atrazine could affect
reproductive development or reproduction in amphibians and
fish. There are other theories where biological plausibility is absent. If atrazine changes the relationship between estradiol and
testosterone, this would result in clear downstream effects on larynx size. Likewise, the nonmonotonic concentration response,
764
K. R. SOLOMON ET AL.
Downloaded At: 15:43 28 October 2008
which is inconsistent with the observed concentration-related
induction of aromatase in cancer cell lines, has been invoked to
explain anomalous results (Hayes et al., 2003) when parsimony
would suggest that there are other explanations for the observed
effects. The implausible theory that atrazine is bioconcentrated
to a large extent in frogs is similarly invoked to explain differences between studies in different laboratories (Hayes, 2004)
when this is not supported by the physicochemical properties
of atrazine or prior observations of bioconcentration in fish and
amphibians, including X. laevis.
Recovery
The guideline of recovery is applicable where a stressor has
been removed from the affected environment and recovery of
the affected organisms occurs. Recovery has been observed in
experimental systems, such as microcosms, and is used as a criterion for classification of responses in communities and ecosystems (Brock et al., 2006). For recovery to be used as a guideline
for causality in assessing the effects of atrazine on aquatic vertebrates, there must be an effect from which recovery can be
observed. This effect also must be consistent, reproducible, and
robust so that it can be studied in this context. No such direct
(or indirect) effects of atrazine on reproductive development or
reproduction in aquatic vertebrates have been observed; thus,
the guideline of recovery cannot be tested.
Overall, the central theory that environmentally relevant concentrations of atrazine affect reproduction and/or reproductive
development in amphibians is not supported by the vast majority of observations. The same conclusion also holds for the
supporting theories such as induction of aromatase.
XIV. SUMMARY
The herbicide atrazine is widely used in agriculture for the
production of corn and other crops. Atrazine is found in surface
waters and several reports on the effects of atrazine on aquatic
organisms have been published in the literature. However, there
is inconsistency in the effects reported and inconsistency between studies in different laboratories. Some studies reported
adverse effects on sexual development of atrazine in frogs and
other amphibians. To assess whether atrazine causes adverse effects in frogs through mechanisms mediated by endocrine and
other pathways, several hypotheses were tested in laboratory and
field studies, using guidelines for the identification of causative
agents of disease and ecoepidemiology derived from Koch’s
postulates and the Bradford–Hill guidelines. The hypotheses
were that atrazine used in crop protection causes adverse effects
in amphibians through: (1) estrogen-mediated mechanisms, (2)
androgen-mediated mechanisms, (3) thyroid-mediated mechanisms, (4) adverse effects on gonadal development in amphibians, or (5) adverse effects at the population level in exposed
amphibians. The biological plausibility of the proposed mechanisms of endocrine disruption was critically assessed in relation
to results of controlled laboratory and microcosm studies as well
as field observations. These data include DNA genotyping in relation to the haplotype specificity of a developmental response
based on the presence of testicular ovarian follicles in male frogs
and the potential for transgenerational effects resulting from exposure to atrazine in frogs. Based on a weight-of-evidence analysis of all of the data, the central theory that environmentally
relevant concentrations of atrazine affect reproduction and/or
reproductive development in amphibians is not supported by the
vast majority of observations. The same conclusion also holds
for the supporting theories such as induction of aromatase, the
enzyme that converts testosterone to estradiol. We conclude that
environmentally relevant concentrations of atrazine do not affect
amphibian growth, sexual development, reproduction, and survival. Although fewer data are available, the same conclusions
apply to fish and reptiles.
ACKNOWLEDGMENTS
This review was developed with a grant from Syngenta Crop
Protection, Inc. The authors specifically thank the following individuals for their help in preparing this review: Dr. Norbert
Kaminski of the Department of Pharmacology and Toxicology,
Michigan State University, for his contributions to reviewing
the immunotoxicity papers; the late Sir Richard Doll, Green
College, Oxford, for his helpful discussions of the BradfordHill guidelines; Cathy Bens for advice on quality assurance and
report data; Susanne Williamson for logistics, coordination, and
meeting arrangements; and Robert Bruce of Ecorisk for his management support.
REFERENCES
Allran, J.W., and Karasov, W.H. (2000). Effects of atrazine and nitrate
on northern leopard frog (Rana pipiens) larvae exposed in the laboratory from posthatch through metamorphosis. Environ. Toxicol.
Chem. 19:2850–2855.
Allran, J.W., and Karasov, W.H. (2001). Effects of atrazine on embryos,
larvae, and adults of anuran amphibians. Environ. Toxicol. Chem.
20:769–775.
Ankley, G.T., Diamond, S.A., Tietge, J., Holcombe, G.W., Jensen,
K.M., Defoe, D.L., and Peterson, R. (2002). Assessment of the risk
of solar ultraviolet radiation to amphibians. I. Dose-dependent induction of hindlimb malformations in the northern leopard frog (Rana
pipiens). Environ. Sci. Technol. 36:2853–2858.
ASTM. (1992). Standard guide for conducting the Frog Embryo Teratogenesis Assay—Xenopus (FETAX). E1439-91. In Annual Book
of ASTM Standards, Vol 11.04. Philadelphia, PA, pp. 1199–1209.
Baker, J.L., and Laflen, J.M. (1979). Runoff losses of surface-applied
herbicides as affected by wheel tracks and incorporation. J. Environ.
Qual. 8:602–607.
Barron, M.G. (1986). Endocrine control of smoltification in anadromous salmonids. J. Endocrinol. 108:313–319.
Battaglin, W.A., Furlong, E.T., Burkhardt, M.R., and Peter, C.J. (2000).
Occurrence of sulfonylurea, sulfonamide, imidazolinone, and other
herbicides in rivers, reservoirs and ground water in the Midwestern
United States, 1998. Sci. Tot. Environ. 248:123–133.
Beebee, T.J.C., and Griffiths, R. (2005). The amphibian decline crisis:
A watershed for conservation biology? Biol. Conserv. 132:136–142.
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
Belden, L.K. (2006). Impact of eutrophication on wood frog, Rana sylvatica, tadpoles indected with Echinostoma tricolvis cercaria. Can.
J. Zool. 84:1315–1321.
Beldomenico, P.M., Rey, F., Prado, W.S., Villarreal, J.C., Munõz-deToro, M., and Luque, E.H. (2007). In ovum exposure to pesticides
increases the egg weight loss and decreases hatchlings weight of
Caiman latirostris (Crocodylia: Alligatoridae). Ecotoxicol. Environ.
Safety. 68:246–251.
Berger, L., Speare, R., Daszak, P., Green, D.E., Cunningham, A.A.,
Goggin, C.L., Slocombe, R., Ragan, M.A., Hyatt, A.D., McDonald,
K.R., Hines, H.B., Lips, K.R., Marantelli, G., and Parkes, H.
(1998). Chytridiomycosis causes amphibian mortality associated
with population declines in the rain forests of Australia and Central America. Proc. Natl. Acad. Sci. USA. 95:9031–9036.
Biagianti-Risbourg S. (1990). Contribution a l’etude du foie de juveniles de muges Teleosteens, (Mugilides) contamines experimentalment par l’atrazine (s-triazine, herbicide): interet en ecotoxicologie
[PhD thesis]. Perpignan, France: Academie de Montpellier, University of Perpignan, France.
Birge, W.J., Black, J.A., Westerman, A.G., and Ramey, B.A. (1983).
Fish and amphibian embryos—A model system for teratogenicity.
Fundam. Appl. Toxicol. 3:237–242.
Bisson, M., and Hontela, A. (2002). Cytotoxic and endocrine-disrupting
potential of atrazine, diazinon, endosulfan, and mancozeb in adrenocortical steroidogenic cells of rainbow trout exposed in vitro. Toxicol.
Appl. Pharmacol. 180:110–117.
Blackler, A.W., Fischberg, M., and Newth, D.R. (1965). Hybridization of two subspecies of Xenopus laevis (Daudin). Rev Suisse Zool.
72:841–857.
Blackler, A.W., and Fischberg, M. (1968). Hybridization of Xenopus
laevis petersi/poweri and X. l. laevis. Rev. Suisse Zool. 75:1023–
1103.
Blaustein, A.R., and Johnson, P.T.J. (2003). The complexity of deformed amphibians. Front. Ecol. Environ. 1:87–94.
Bögi, C., Schwaiger, J., Ferling, J.H., Mallow, U., Steineck, C.,
Sinowatz, F., Kalbfus, W., Negele, R.D., Lutz, I., and Kloas, W.
(2003). Endocrine effects of environmental pollution on Xenopus
laevis and Rana temporaria. Environ. Res. 93:195–201.
Bridges, C., Little, E., Gardiner, D., Petty, J., and Huckins, J. (2004).
Assessing the toxicity and teratogenicity of pond water in northcentral Minnesota to amphibians. Environ. Sci. Res. Int. 11:233–239.
Bringolf, R.B., Belden, J.B., and Summerfelt, R.C. (2004). Effects of
atrazine on fathead minnow in a short-term reproduction assay. Environ. Toxicol. Chem. 23:1019–1025.
Brock, T.C.M., Arts, G.H.P., Maltby, L., and van den Brink, P.J. (2006).
Aquatic risks of pesticides, ecological protection goals and common
aims in EU legislation. Integr. Environ. Assess. 2:e20–e46.
Brockway D.L., Smith P.D., and Stancil F.E. (1984). Fate and effects of
atrazine in small aquatic microcosms. Bull Environ Contam Toxicol.
32:345–353.
Brodkin, M.A., Madhoun, H., Rameswaran, M., and Itzicj, V. (2007).
Atrazine is an immune disruptor in adult nothern leopard frogs (Rana
pipiens). Environ. Toxicol. Chem. 26:80–84.
Burrowes, P.A., Joglar, R.L., and Green, D.E. (2004). Potential causes
for amphibian declines in Puerto Rico. Herpetologica 60:141–154.
Carr, J.A., Gentles, A., Smith, E.E., Goleman, W.L., Urquidi, L.J.,
Thuett, K., Kendall, R.J., Giesy, J.P., Gross, T.S., Solomon, K.R.,
and Van Der Kraak, G.J. (2003). Response of larval Xenopus lae-
765
vis to atrazine: Assessment of gonadal and laryngeal morphology.
Environ. Toxicol. Chem. 22:396–405.
Carr, J.A., and Norris, D.O. (2006). The hypothalamus–pituitary axis.
In: Norris, D.O., and Carr, J.A., ed. Endocrine Disruption: The Biological Basis for Health Effects in Wildlife and Humans, pp. 87–110.
Oxford University Press, New York.
Chang, C.Y., and Witschi, E. (1956). Genetic control and hormonal
reversal of sex differentiation in Xenopus. Proc. Soc. Exp. Biol. Med.
93:140–144.
Christin, M.-S., Gendron, A.D., Brousseau, P., Ménard, L.,
Marcogliese, D.J., Cyr, D., Ruby, S., and Fournier, M. (2003). Effects of agricultural pesticides on the immune system of Rana pipiens
and on its resistance to parasitic infection. Environ. Toxicol. Chem.
22:1127–1133.
Christin, M.-S., Ménard, L., Gendron, A.D., Ruby, S., Cyr, D.,
Marcogliese, D.J., Rollins-Smith, L., and Fournier, M. (2004). Effects of agricultural pesticides on the immune system of Xenopus
laevis and Rana pipiens. Aquat. Toxicol. 67:33–43.
Clements, C., Ralph, S., and Petras, M. (1997). Genotoxicity of select
herbicides in Rana catesbeiana tadpoles using the alkaline singlecell gel DNA electrophoresis (Comet) assay. Environ. Mol. Mutagen.
29:277–288.
Coady, K.K., Murphy, M.B., Villeneuve, D.L., Hecker, M., Jones,
P.D., Carr, J.A., Solomon, K.R., Smith, E.E., Van Der Kraak,
G.J., Kendall, R.J., and Giesy, J.P. (2004). Effects of atrazine on
metamorphosis, growth, and gonadal development in the green
frog (Rana clamitans). J. Toxicol. Environ. Health A 67:941–
957.
Coady, K.K., Murphy, M.B., Villeneuve, D.L., Hecker, M., Carr, J.A.,
Solomon, K.R., Van Der Kraak, G.J., Smith, E.E., Kendall, R.J., and
Giesy, J.P. (2005). Effects of atrazine on metamorphosis, growth,
laryngeal and gonadal development, aromatase activity, and plasma
sex steroid concentrations in Xenopus laevis. Ecotoxicol. Environ.
Safety. 62:160–173.
Collins, J.P., and Storfer, A. (2003). Global amphibian declines: Sorting
the hypotheses. Div. Distrib. 9:89–98.
Cooper, R.L., Stoker, T.E., Tyrey, L., Goldman, J.M., and McElroy,
W.K. (2000). Atrazine disrupts the hypothalamic control of pituitary–
ovarian function. Toxicol. Sci. 53:297–307.
Cooper, R.L., Laws, S.C., Das, P.C., Narotsky, M.G., Goldman, J.M.,
Tyrey, E.L., and Stoker, T.E. (2007). Atrazine and reproductive function: Mode and mechanism of action studies. Brth Defects Res. (B)
80:98–112.
Cossarinidunier, M. (1987). Effects of the pesticides atrazine and lindane and of manganese ions on cellular-immunity of carp, Cyprinus
carpio. J. Fish. Biol. 31:67–73.
Cossarinidunier, M., Demael, A., Riviere, J.L., and Lepot, D. (1988).
Effects of oral doses of the herbicide atrazine on carp (Cyprinus
carpio). Ambio 17:401–405.
Cossarinidunier, M., and Hattenberger, A.M. (1988). Effect of pesticides (atrazine and lindane) on the replication of spring viremia of
carp virus in vitro. Ann. Res. Vet. 19:209–211.
Crain, D.A., Guillette, L.J., Rooney, A.A., and Pickford, D.B. (1997).
Alterations in steroidogenesis in alligators (Alligator mississippiensis) exposed naturally and experimentally to environmental contaminants. Environ. Health Perspect. 105:528–553.
Crain, D.A., Spiteri, I.D., and Guillette, L.J., Jr. (1999). The functional
and structural observations of the neonatal reproductive system of
Downloaded At: 15:43 28 October 2008
766
K. R. SOLOMON ET AL.
alligators exposed in ovo to atrazine, 2,4-D, or estradiol. Toxicol.
Ind. Health 15:180–185.
Czechura, G.V., and Ingram, G. (1990). Taudactylus diurnus and the
case of the disappearing frogs. Mem. Queenl. Mus. 29:361–365.
Daly, G.L., Lei, Y.D., Teixeira, C., Muir, D.C.G., Castillo, L.E., and
Wania, F. (2007). Accumulation of current-use pesticides in neotropical montane forests. Environ. Sci. Technol. 41:1118–1123.
Das, P.C., McElroy, W.K., and Cooper, R.L. (2000). Differential modulation of catecholamines by chlorotriazine herbicides in pheochromocytoma (PC12) cells in vitro. Toxicol. Sci. 56:324–331.
Daszak, P., Berger, L., Cunningham, A.A., Hyatt, A.D., Green, D.E.,
and Speare, R. (1999). Emerging infectious diseases and amphibian
population declines. Emerg. Infect. Dis. 5:745–748.
Daszak, P., Cunningham, A.A., and Hyatt, A.D. (2003). Infectious disease and amphibian population declines. Div. Distrib. 9:141–150.
Davidson, C. (2004). Declining downwind: Amphibian population declines in California and historical pesticide use. Ecol. Appl. 14:1892–
1902.
Davis, D.L., Bradlow, H.L., Wolff, M., Woodruff, T., Hoel, D.G., and
Anton-Culver, H. (1993). Medical hypothesis: Xenoestrogens as preventable causes of breast cancer. Environ. Health Perspect. 101:372–
377.
De Solla, S.R., Martin, P.A., Fernie, K.J., Park, B.J., and Mayne,
G. (2006). Effects of environmentally relevant concentrations of
atrazine on gonadal development of snapping turtles (Chelydra serpentina). Environ. Toxicol. Chem. 25:520–526.
Department of the Interior Fish and Wildlife Service. (2000). 50 CFR
Part 17. Endangered and threatened wildlife and plants; Proposed
designation of critical habitat for the California Red-legged Frog
(Rana aurora draytonii). Fed. Reg. 65:54892.
Detenbeck, N.E., Hermanutz, R., Allen, K., and Swift, M.C. (1996).
Fate and effects of the herbicide atrazine in flow-through wetland
mesocosms. Environ. Toxicol. Chem. 15:937–946.
Devine, M.D., Duke, S.O., and Fedtke, C. (1993). Physiology of Herbicide Action. Englewood Cliffs, NJ: Prentice Hall.
Dhabhar, F.S., McEwen, B.S., and Spencer, R.L. (1993). Stress response, adrenal steroid receptor levels and corticosteroid-binding
globulin levels—A comparison between Sprague-Dawley, Fischer
344 and Lewis rats. Brain Res. 616:89–98.
Dhabhar, F.S., Miller, A.H., Stein, M., McEwen, B.S., and Spencer, R.L.
(1994). Diurnal and acute stress-induced changes in distribution of
peripheral blood leukocyte subpopulations. Brain Behav. Immunol.
8:66–79.
Diamond, S.A., Peterson, G.S., Tietge, J., and Ankley, G.T. (2002).
Assessment of the risk of solar ultraviolet radiation to amphibians.
III. Prediction of impacts in selected Northern Midwestern wetlands.
Environ. Sci. Technol. 36:2866–2874.
Diana, S.G., Resetarits, W.J., Jr., Schaeffer, D.J., Beckman, K.B., and
Beasley, V.R. (2000). Effects of atrazine on amphibian growth and
survival in artificial aquatic communities. Environ. Toxicol. Chem.
19:2961–2967.
Drost, C.A., and Fellers, G.M. (1996). Collapse of a regional frog fauna
in the Yosemite area of the California Sierra Nevada. Conserv. Biol.
10:410–425.
Du Preez, H.H., and van Vuren, H.J. (1992). Bioconcentration of
atrazine in the banded tilapia, Tilapia sparrmanii. Comp. Biochem.
Physiol. C 101:651–655.
Du Preez, H.H., Kunene, N., Hanner, R., Evans, B.J., Giesy, J.P., Hosmer, A.J., Solomon, K.R., and Van Der Kraak, G.J. (2008a). Clade-
specific occurrence of testicular oocytes in Xenopus laevis from
South Africa. Aquat. Toxicol., in press.
Du Preez, L.H., Jansen van Rensburg, P.J., Jooste, A.M., Carr, J.A.,
Giesy, J.P., Gross, T.S., Kendall, R.J., Smith, E.E., Van Der Kraak, G.,
and Solomon, K.R. (2005a). Seasonal exposures to triazine and other
pesticides in surface waters in the western Highveld corn-production
region in South Africa. Environ. Pollut. 135:131–141.
Du Preez, L.H., Solomon, K.R., Carr, J.A., Giesy, J.P., Gross, T.S.,
Kendall, R.J., Smith, E.E., Van Der Kraak, G.J., and Weldon, C.
(2005b). Population structure of the African clawed frog (Xenopus
laevis) in maize-growing areas with atrazine application versus nonmaize-growing areas in South Africa. Afr. J. Herpatol. 54:61–68.
Du Preez, L.H., Kunene, N., Everson, G.J., Carr, J.A., Giesy, J.P., Gross,
T.S., Kendall, R.J., Smith, E.E., Solomon, K.R., and Van Der Kraak,
G.J. (2008b). Reproduction, larval growth, and reproductive development in African clawed frogs (Xenopus laevis) exposed to atrazine.
Chemosphere 71:546–552.
Dunier, M., and Siwicki, A.K. (1993). Effects of pesticides and other
organic pollutants in the aquatic environment on immunity of fish—
A review. Fish Shellfish Immunol. 3:423–438.
Edginton, A.N., and Rouleau, C. (2005). Toxicokinetics of 14 C-atrazine
and its metabolites in stage-66 Xenopus laevis. Environ. Sci. Technol.
39:8083–8089.
Edwards, W.M., Shipitalo, M.J., Lal, R., and Owens, L.B. (1997). Rapid
changes in concentration of herbicides in corn field surface depressions. J. Soil Water Conserv. 52:227–281.
Fan, W., Yanase, T., Morinaga, H., Gondo, S., Okabe, T., Nomura,
M., Komatsu, T., Morohashi, K.-I., Hayes, T.B., Takayanagi, R., and
Nawata, H. (2007). Atrazine-induced aromatase expression is SF1 dependent: Implications for endocrine disruption in wildlife and
reproductive cancers in humans. Environ. Health Perspect. 115:720–
727.
Fasano, S., Minucci, S., Di Matteo, L., D’Antonio, M., and Pierantoni,
R. (1989). Intratesticular feedback mechanisms in the regulation of
steroid profiles in the frog, Rana esculenta. Gen. Comp. Endocrinol.
75:335–342.
Fellers, G.M., and Drost, C.A. (1993). Disappearence of the Cascades
frog Rana cascadae at the southern end of its range, California. Biol.
Conserv. 65:177–181.
Fernandez, M., L’Hardidon, J., Gautnier, L., and Zoll-Moreux, C.
(1993). Amphibian micronucleus test(s): A simple and reliable
method for evaluating in vivo genotoxic effects of freshwater pollutants and radiation. Initial assessment. Mutat. Res. 292:83–99.
Filipov, N.M., Pinchuk, L.M., Boyd, B.L., and Crittenden, P.L. (2005).
Immunotoxic effects of short-term atrazine exposure in young male
C57BL/6 mice. Toxicol. Sci. 86:324–332.
Fournier, M., Friborg, J., Girard, D., Mansour, S., and Krzystyniak, K.
(1992). Limited immunotoxic potential of technical formulation of
the herbicide atrazine (Aatrex) in mice. Toxicol. Lett. 60:263–274.
Frank, R., Braun, H.E., Ripley, B.D., and Clegg, B.S. (1990). Contamination of rural ponds with pesticide, 1971–1985, Ontario, Canada.
Bull. Environ. Contam. Toxicol. 44:401–409.
Freeman, J.L., and Rayburn, A.L. (2004). In vivo genotoxicity of
atrazine to anuran larvae. Mutat. Res. 560:69–78.
Freeman, J.L., and Rayburn, A.L. (2005). Developmental impact of
atrazine on metamorphing Xenopus laevis as revealed by nuclear
analysis and morphology. Environ. Toxicol. Chem. 24:1648–1653.
Fridgen, C.M., Pauli, B.D., Berrill, M., Doe, K., and Jackman, P.
(2005). Chronic effects of atrazine herbicide on the development of
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
Northern leopard frog (Rana pipiens) tadpoles. SETAC National
Meeting, Baltimore, MD. SETAC, Pensacola, FL.
Gallien, L. (1974). Intersexuality. In: Lofts, B., ed. Physiology of the
Amphibia. pp. 523–549. Academic Press, New York.
Gammon, D.W., Aldous, C.N., Carr, W.C., Jr., Sanborn, J.R., and
Pfeifer, K.F. (2005). A risk assessment of atrazine use in California:
Human health and ecological aspects. Pestic. Manage. Sci. 61:331–
355.
Gendron, A.D., Marcogliese, D.J., Barbeau, S., Christin, M.S.,
Brousseau, P., Ruby, S., Cyr, D., and Fournier, M. (2003). Exposure of
leopard frogs to a pesticide mixture affects life history characteristics
of the lungworm Rhabdias ranae. Oecologia 135:469–476.
Giddings, J.M., Anderson, T.A., Hall, L.W., Jr, Kendall, R.J., Richards,
R.P., Solomon, K.R., and Williams, W.M. (2005). A Probabilistic
Aquatic Ecological Risk Assessment of Atrazine in North American
Surface Waters. Pensacola, FL: SETAC Press.
Gluth, G., and Hanke, W.A. (1985). Comparison of physiological
changes in carp, Cyprinus carpio, induced by several pollutants at
sublethal concentrations. I. The dependency on exposure time. Ecotoxicol. Environ. Safety. 9:179–188.
Görge, G., and Nagel, R. (1990). Kinetics and metabolism of 14 Clindane and 14 C-atrazine in early life stages of zebrafish (Brachydanio
rerio). Chemosphere 21:1125–1137.
Goulet, B.N., and Hontela, A. (2003). Toxicity of cadmium, endosulfan, and atrazine in adrenal steroidogenic cells of two amphibian
species, Xenopus laevis and Rana catesbeiana. Environ. Toxicol.
Chem. 22:2106–2113.
Grady, J., Wieser, C., Wiebe, J., and Gross, T.S. (1998). An evaluation
of atrazine as a potential endocrine disruptor in largemouth bass.
SETAC Annual Meeting, November 15–19, 1998; Charlotte, NC.
Grim, K.C., Wolfe, M., Hawkins, W., Johnson, R.D., and Wolf, J.
(2007). Intersex in Japanese medaka (Oryzias latipes) used as negative controls in toxicological bioassays: A review of 54 vases from
41 studies. Environ. Toxicol. Chem. 26:1636–1643.
Groombridge, B. (1982). The IUCN Amphibia–Reptilia Red Data Book.
Gland, Switzerland: International Union for Conservation of Nature
and Natural Resources.
Gross, T.S., Shrestha, S., Wieser, C.M., Wiebe, J.J., Denslow, N.D.,
Chow, C., Johnson, W.E., and Stout, R. (1997). Evaluation of potential endocrine-disrupting effects of water-soluble herbicides in
largemouth bass. SETAC Annual Meeting, November 16–20, 1997;
San Francisco, CA.
Gross, T.S. (1999a). Determination of Potential Effects of 10 Day
Neonatal Exposure of Atrazine on Histological and Hormonal Sex
Determination in Incubated American Alligator (Alligator mississippiensis) Eggs. Gainesville, FL: University of Florida, Wildlife
Reproductive Toxicology Laboratory. No. NOVA98.02a. 22 p.
Gross, T.S. (1999b). Determination of potential effects of 10 day neonatal exposure of atrazine on histological and hormonal sex determination in incubated red-eared slider (Psuedemys elegans) eggs.
Gainesville: University of Florida, Wildlife Reproductive Toxicology Laboratory. No. NOVA98.02b.
Guillette, L.J., Jr, Gross, T.S., Masson, G.R., Matter, J.M., and
Woodward, A.R. (1994). Developmental abnormalities of the gonad of juvenile alligators from contaminated and control lakes in
Florida. Environ. Health Perspect. 102:680–688.
Guillette, L.J., Jr, Pickford, D.B., Crain, D.A., Rooney, A.A., and
Percival, H.F. (1996). Reduction in penis size and plasma testos-
767
terone concentrations in juvenile alligators living in a contaminated
environment. Gen. Comp. Endocrinol. 101:32–42.
Gunkel, G., and Streit, B. (1980). Mechanisms of bioaccumulation of
a herbicide (atrazine, s-triazine) in a freshwater mollusc (Ancylus
fluviatilis Muell.) and a fish (Coregonus fera Jurine.). Water Res.
14:1573–1584.
Guyton, A.C., and Hall, J.E. (1996). Human Physiology and Mechanisms of Disease. Philadelphia, PA: W.B. Saunders.
Hamala, J.A., and Kollig, H.P. (1985). The effects of atrazine on periphyton communities in controlled laboratory ecosystems. Chemosphere 14:1391–1408.
Hayes, T.B. (1998). Sex determination and primary sex differentiation
in amphibians: Genetic and developmental mechanisms. J. Exp. Zool.
281:373–399.
Hayes, T.B., Stuart, A.A., Vonk, A., and Liu, R. (2001). Atrazine disrupts sex differentiation in the African clawed frog (Xenopus laevis) at ecologically relevant doses. SETAC 22nd Annual Meeting,
November 12–15 2001; Baltimore, MD. SETAC. Abstract 421 p.
Hayes, T.B., Collins, A., Mendoza, M., Noriega, N., Stuart, A.A., and
Vonk, A. (2002). Hermaphroditic, demasculinized frogs exposure to
the herbicide atrazine at low ecologically relevant doses. Proc. Natl.
Acad. Sci. USA 99:5476–5480.
Hayes, T.B., Haston K., Tsui, M., Hoang, A., Haeffele, C., and Vonk, A.
(2003). Atrazine-induced hermaphroditism at 0.1 ppb in American
leopard frogs (Rana pipiens): Laboratory and field evidence. Environ.
Health Perspect. 111:568–575.
Hayes, T.B. (2004). This is no denying this: Defusing the confusion
about atrazine. Bioscience 54:1138–1149.
Hayes, T.B., Case, P., Chui, S., Chung, D., Haefele, C., Haston, K.,
Lee, M., Mai, V.P., Marjuoa, Y., Parker, J., and Tsui, M. (2006a).
Pesticide mixtures, endocrine disruption, and amphibian declines:
Are we underestimating the impact? Environ. Health Perspect. 114
Sup. 1:40–50.
Hayes, T.B., Stuart, A.A., Mendoza, M., Collins, A., Noriega, N., Vonk,
A., Johnston, G., Liu, R., and Kpodzo, D. (2006b). Characterization
of atrazine-induced gonadal malformations in African clawed frogs
(Xenopus laevis) and comparisons with effects of an androgen antagonist (cyproterone acetate) and exogenous estrogen (17ß-estradiol):
Support for the demasculinization/feminization hypothesis. Environ.
Health Perspect. 114 Sup. 1:134–141.
Hecker, M., Coady, K.K., Villeneuve, D.L., Murphy, M.B., Jones, P.D.,
and Giesy, J.P. (2003). Response of Xenopus laevis to atrazine exposure: assessment of the mechanism of action of atrazine. East
Lansing: Michigan State University–Aquatic Toxicology Laboratory. Syngenta. No. 1165–02. 143 p.
Hecker, M., Giesy, J.P., Jones, P.D., Jooste, A.M., Carr, J.A., Solomon,
K.R., Smith, E.E., Van Der Kraak, G.J., Kendall, R.J., and Du Preez,
L.H. (2004). Plasma sex steroid concentrations and gonadal aromatase activities in African clawed frogs (Xenopus laevis) from
the corn-growing region of South Africa. Environ. Toxicol. Chem.
23:1996–2007.
Hecker, M., Kim, W.J., Park, J.-W., Murphy, M.B., Villeneuve, D.,
Coady, K.K., Jones, P.D., Solomon, K.R., Van Der Kraak, G.J.,
Carr, J.A., Smith, E.E., du Preez, L.H., Kendall, R.J., and Giesy,
J.P. (2005a). Plasma concentrations of estradiol and testosterone,
gonadal aromatase activity, and ultrastructure of the testis in Xenopus laevis exposed to estradiol and atrazine. Aquat. Toxicol. 72:383–
396.
Downloaded At: 15:43 28 October 2008
768
K. R. SOLOMON ET AL.
Hecker, M., Park, J.-W., Murphy, M.B., Jones, P.D., Solomon, K.R., Van
Der Kraak, G.J., Carr, J.A., Smith, E.E., Du Preez, L.H., Kendall,
R.J., and Giesy, J.P. (2005b). Effects of atrazine on CYP19 gene
expression and aromatase activity in testes and on sex steroid concentrations in plasma of male African clawed frogs (Xenopus laevis).
Toxicol. Sci. 86:273–280.
Hecker, M., Murphy, M.B., Coady, K.K., Villeneuve, D.L., Jones,
P.D., Carr, J.A., Solomon, K.R., Smith, E.E., Van Der Kraak, G.L.,
Gross, T.S., du Preez, L.H., Kendall, R.J., and Giesy, J.P. (2006).
Terminology of gonadal anomalies in fish and amphibians resulting
from chemical exposures. Rev. Environ. Contam. Toxicol. 187:103–
132.
Heneweer, M., van den Berg, M., and Sanderson, J.T. (2004). A comparison of human H295R and rat R2C cell lines as in vitro screeing
tools for effects on aromatase. Toxicol. Lett. 146:183–194.
Hill, A.B. (1965). The environment and disease: association or causation? Proc. R. Soc. Med. 58:295–300.
Hines, H., Mahony, M., and McDonald, K. (1999). An assessment of
frog declines in wet subtropical Australia. In: Campbell A., ed. Declines and Disappearances of Australian Frogs, pp. 44–63. Environment Australia, Canberra.
Hoagland, K.D., Drenner, R.W., Smith, J.D., and Cross, D.R. (1993).
Freshwater community responses to mixtures of agricultural pesticides: Effects of atrazine and bifenthrin. Environ. Toxicol. Chem.
12:627–637.
Hoberg, J.R. (2007). The Toxicity of Atrazine to the Freshwater Macrophyte Elodea canadensis at Three Light Intensities for 14 Days.
Greensboro, NC: Syngenta Crop Protection, Inc. No. 1781.6691. 96
p.
Hooghe, R.J., Devos, S., and Hooghe-Peters, E.L. (2000). Effects of selected herbicides on cytokine production in vitro. Life Sci. 66:2519–
2525.
Houck, A., and Sessions, S.K. (2006). Could atrazine affect the immune
system of the frog, Rana pipiens? Bios 77:107–112.
Houlahan, J.E., Findlay, C.S., Schmidt, B.R., Meyer, A.H., and Kuzmin,
S.L. (2000). Quantitative evidence for global amphibian population
declines. Nature 404:752–755.
Howard, P.H. (1991). Handbook of Environmental Fate and Exposure
Data for Organic Chemicals. Chelsea, MI: Lewis.
Howe, G.E., Gillis, R., and Mowbray, R.C. (1998). Effects of chemical
synergy and larval stage on the toxicity of atrazine and alachlor to
amphibian larvae. Environ. Toxicol. Chem. 17:519–525.
Hu, F., Smith, E.E., and Carr, J.A. (2008). Effects of larval exposure to
estradiol on spermatogenesis and in vitro gonadal steroid secretion
in African clawed frogs, Xenopus laevis. Gen. Comp. Endocrinol.
155:190–200.
Ingram, G.J., and McDonald, K.R. (1993). An update on the decline of
Queenslands frogs. In: Lunney D., and Ayers D., eds. Herpetology
in Australia: A Diverse Discipline, pp. 297–303. Royal Zoological
Society of NSW, Mosman.
IPCS. (2002). Global Assessment of the State-of-the-Science of
Endocrine Disruptors. Geneva, Switzerland: International Programme on Chemical Safety of the World Health Organization. No.
WHO/PCS/EDC/02.2. http://www.who.int/pcs
Jensen, K.I.N., Stephenson, G.R., and Hunt, L.A. (1977). Detoxification
of atrazine in three gramineae subfamilies. Weed Sci. 25:212–220.
Johnson, P.T.J., Lunde, K.B., Richie, E.G., and Launer, A.E. (1999).
The effect of tematode infection on amphibian limb development
and survivorship. Science 284:802–804.
Johnson, P.T.J., Lunde, K.B., Thurman, E.M., Ritchie, E.G., Wray,
S.N., Sutherland, D.R., Kapper, J.M., Frest, T.J., Bowerman, J.,
and Blaustein, A.R. (2002). Parasite (Ribeiroia ondatrae) infection
linked to amphibian malformation in the western United States. Ecol.
Monogr. 72:151–168.
Johnson, P.T.J., Chase, J.M., Dosch, K.L., Hartson, R.B., Gross, J.A.,
Larson, D.J., Sutherland, D.R., and Carpenter, S.R. (2007). Aquatic
eutrophication promotes pathogenic infection in amphibians. Proc.
Natl. Acad. Sci. USA 104:15781–15786.
Jooste, A.M., Du Preez, L.H., Carr, J.A., Giesy, J.P., Gross, T.S.,
Kendall, R.J., Smith, E.E., Van Der Kraak, G.J., and Solomon,
K.R. (2005). Gonadal development of Xenopus laevis larvae exposed
through larval development to atrazine in outdoor microcosms. Environ. Sci. Technol. 39:5255–5261.
Kang, L., Marin, M., and Kelley, D. (1995). Androgen biosynthesis
and secretion in developing Xenopus laevis. Gen. Comp. Endocrinol.
100:293–307.
Karrow, N.A., Mccay, J.A., Brown, R.D., Musgrove, D.L., Guo, T.L.,
Germolec, D.R., and White, K.L. (2005). Oral exposure to atrazine
modulates cell-mediated immune function and decreases host resistance to the B16F10 tumor model in female B6C3F1 mice. Toxicology 209:15–28.
Kazeto, Y., Place, A.R., and Trant, J.M. (2004). Effects of endocrine
disrupting chemicals on the expression of CYP19 genes in zebrafish
(Danio rerio) juveniles. Aquat. Toxicol. 69:25–34.
Kendall, R.J., and Lacher, T.E., Jr, eds. (1994). Wildlife Toxicology and
Population Modeling: Integrated Studies of Agroecosystems. Lewis,
Chelsea, MI.
Kiesecker, J.M. (2002). Synergism between trematode infection and
pesticide exposure: A link to amphibian limb deformities in nature?
Proc. Natl. Acad. Sci. USA 99:9900–9904.
Klaasen, H.E., and Kadoum, A.H. (1979). Distribution and retention
of atrazine and carbofuran in farm pond ecosystems. Arch. Environ.
Contam. Toxicol. 8:345–353.
Kloas, W., Lutz, I., Springer, T., Krueger, H., Wolf, J., Holden, L., and
Hosmer, A. (2008). Atrazine does not induce gonadal feminization
in Xenopus laevis. Toxicol. Sci., accepted.
Knutson, M.G., Richardson, W.B., Reineke, D.M., Gray, B.R.,
Parmelee, J.R., and Weick, S.E. (2004). Agricultural ponds support
amphibian populations. Ecol. Appl. 14:669–684.
Koch, R. (1942). The aetiology of tuberculosis (translation of Die Aetiologie der Tuberculose [1882]). In: Clark, D.H., ed. Source Book
of Medical History, pp. 392–406. Dover, New York.
Koprivnikar, J., Forbes, M.R., and Baker, R.L. (2006). Effects of
atrazine on cercarial longevity, activity, and infectivity. J. Parasitol.
92:306–311.
Koprivnikar, J., Forbes, M.R., and Baker, R.L. (2007). Contaminant
effects on host–parasite interactions: Atrazine, frogs, and trematodes.
Environ. Toxicol. Chem. 26:2166–2170.
Larson, D.L., McDonald, S., Fivizzani, A.J., Newton, W.E., and
Hamilton, S.J. (1998). Effect of the herbicide atrazine on Ambystoma
tigrinum metamorphosis: Duration, larval growth and hormonal response. Physiol. Zool. 71:671–679.
Lawler, S.P., Dritz, D., Strange, T., and Holyoak, M. (1999). Effects of
introduced mosquitofish and bullfrogs on the threatened California
red-legged frog. Conserv. Biol. 13:613–622.
LeBlanc, G.A., and Wang, G. (2006). Chemical mixtures: Greaterthan-additive effects? Environ. Health Perspect. 114:A517–
A518.
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
Lenkowski, J.R., Reed, J.M., Deininger, L., and McLaughlin, K.A.
(2008). Perturbation of organogenesis by the herbicide atrazine in
the amphibian Xenopus laevis. Environ. Health Perspect. 116:223–
230.
Levy, G., Lutz, I., Kruger, A., and Kloas, W. (2004). Bisphenol A induces feminization in Xenopus laevis tadpoles. Environ. Res. 94:102–
111.
L’Haridon, J., Fernandez, M., Ferrier, V., and Bellan, J. (1993).
Evaluation of the genotoxicity of N -nitrosoatrazine, N nitrosodiethanolamine and their precursors in vivo using the
newt micronucleus test. Water Res. 28:855–862.
Licht, P., McCreery, B.R., Barnes, R., and Pang, R. (1983). Seasonal
and stress related changes in plasma gonadotropins, sex steroids,
and corticosterone in the bullfrog, Rana catesbeiana. Gen. Comp.
Endocrinol. 50:124–145.
Lipnick, R.L. (1993). Baseline toxicity QSAR models: A means to assess mechanism of toxicity for aquatic organisms and mammals. In:
Gorsuch, J., Dwyer, F., Ingersoll, C., and La Point, T.W., eds. Environmental Toxicology and Risk Assessment STP 1216, Volume 2,
pp. 610–619. American Society for Testing and Materials, Philadelphia, PA.
Longcore, J.E., Pessier, A.P., and Nichols, D.K. (1999). Batrachochytrium dendrobatidis gen. et sp. nov., a chytrid pathogenic to
amphibians. Mycologia 91:219–227.
Lutz, I., Kloas, W., Springer, T., Holden, L., Wolf, J., Kruger, H., and
Hosmer, A. (2008). Development, standardization and refinement
of procedures for evaluating effects of endocrine active compounds
on development and sexual differentiation of Xenopus laevis. Anal.
Bioanal. Chem. 390:2031–2048.
Mazanti, L., Sparling, D.W., Rice, C., Bialeck, K., Stevenson, C., and
Teels, B. (2003). Synergistic effects of a combined exposure to herbicides and an insecticide in Hyla versicolor. In: Linder E., Krest
S.K., Sparling D.W., and Little E.E., eds. Multiple Stressor Effects
in Relation to Declining Amphibian Populations, ASTM STP 1443,
pp. 111–129. ASTM, West Conshohocken, PA.
McClelland B.E., Wilczynski W., and Ryan M.J. (1996). Correlations
between call characteristics and morphology in male cricket frogs
(Acris crepitans). J. Exp. Biol. 199:1907–1919.
McClelland, B.E., Wilczynski, W., and Ryan, M.J. (1998). Intraspecific
variation in laryngeal and ear morphology in male cricket frogs (Acris
crepitans). Biol. J. Linn. Soc. Lond. 63:51–67.
McConkey, B.J., Dixon, G.G., and Greenburg, B.M. (2000). Fractional
simplex designs for mixture interaction screening in complex mixtures. Biometrics 56:824–833.
McDaniel, T., Martin, P.A., Struger, J., Sherry, J., Marvin, C.H.,
McMaster, M.E., Clarence, S., and Tetreault, G. (2008). Potential
endocrine disruption of sexual development in free ranging male
northern leopard frogs (Rana pipiens) and green frogs (Rana clamitans) from areas of intensive row crop agriculture. Aquat. Toxicol.
88:230–242.
Mendelson, J.R. III, Lips, K.R., Gagliardo, R.W., Rabb, G.B., Collins,
J.P., Diffendorfer, J.E., Daszak, P., Ibáñez, D R., Zippel, K.C.,
Lawson, D.P., Wright, K.M., Stuart, S.N., Gascon, C., da Silva,
H.R., Burrowes, P.A., Joglar, R.L., La Marca, E., Lötters, S., du
Preez, L.H., Weldon, C., Hyatt, A., Rodriguez-Mahecha, J.V., Hunt,
S., Robertson, H., Lock, B., Raxworthy, C.J., Frost, D.R., Lacy, R.C.,
Alford, R.A., Campbell, J.A., Parra-Olea, G., Bolaños, F., Domingo,
J.J.C., Halliday, T., Murphy, J.B., Wake, M.H., Coloma, L.A.,
Kuzmin, S.L., Price, M.S., Howell, K.M., Lau, M., Pethiyagoda,
769
R., Boone, M., Lannoo, M.J., Blaustein, A.R., Dobson, A., Griffiths,
R.A., Crump, M.L., Wake, D.B., and Brodie, E.D., Jr. (2006). Confronting amphibian declines and extinctions. Science 313:48.
Moore, A., and Waring, C.P. (1998). Mechanistic effects of a triazine
pesticide on reproductive endocrine function in mature male Atlantic
salmon (Salmo salar L.) parr. Pestic. Biochem. Physiol. 62:41–50.
Moore, A., and Lower, N. (2001). The impact of two pesticides
on olfactory-mediated endocrine function in mature male Atlantic
salmon (Salmo salar L.) parr. Comp. Biochem. Physiol. B 129:269–
276.
Morgan, M.K., Scheuerman, P.R., Bishop, C.S., and Pyles, R.A. (1996).
Teratogenic potential of atrazine and 2,4-D using FETAX. J. Toxicol.
Environ. Health 48:151–168.
Mosconi, G., Yamamoto, K., Carnevali, O., Nabissi, M., PolzonettiMagni, A., and Kikuyama, S. (1994). Seasonal changes in plasma
growth hormone and prolactin concentrations of the frog Rana esculenta. Gen. Comp. Endocrinol. 93:380–387.
Mosconi, G., Crnevali, O., Franzoni, M.F., Cottone, E., Lutz, I., Kloas,
W., Yamamoto, K., Kikuyama, S., and Polzonetti-Magni, A.M.
(2002). Environmental estrogens and reproductive biology in amphibians. Gen. Comp. Endocrinol. 126:125–129.
Muller, J.K., Gross, T.S., and Borgert, C.J. (2007a). Topical dose delivery in the reptilian egg treatment model. Environ. Toxicol. Chem.
26:914–919.
Muller, J.K., Scarborough, J.E., Sepúveda, M.S., Casella, G., Gross,
T.S., and Borgert, C.J. (2007b). Dose verification after topical treatment of alligator (Alligator mississippiensis) eggs. Environ. Toxicol.
Chem. 26:908–913.
Murphy, M.B., Hecker, M., Coady, K.K., Tompsett, A.R., Du Preez,
L.H., Everson, G.J., Solomon, K.R., Carr, J.A., Smith, E.E., Kendall,
R.J., Van Der Kraak, G.J., and Giesy, J.P. (2006a). Atrazine concentrations, gonadal gross morphology and histology in ranid frogs collected in Michigan agricultural areas. Aquat. Toxicol. 76:230–245.
Murphy, M.B., Hecker, M., Coady, K.K., Tompsett, A.R., Higley, E.B.,
Jones, P.D., Du Preez, L.H., Solomon, K.R., Carr, J.A., Smith, E.E.,
Kendall, R.J., van der Kraak, G.J., and Giesy J.P. (2006b). Plasma
steroid hormone concentrations, aromatase activities and GSI in
ranid frogs collected from agricultural and non-agricultural sites in
Michigan (USA). Aquat. Toxicol. 77:153–166.
Murphy, M.B., Hecker, M., Coady, K.K., Tompsett, A.R., Jones, P.D.,
Solomon, K.R., Van Der Kraak, G., Carr, J.A., Smith, E.E., Kendall,
R.J., and Giesy, J.P. (2006c). Sediment TCDD-EQs and EROD and
MROD activities in Ranid frogs from agricultural and nonagricultural
sites in Michigan (USA). Arch. Environ. Contam. Toxicol. 51:467–
477.
Napier, J.D., Scheuerman, P.R., and Pyles, R.A. (1998). The effect of
water hardness and humic acid on the teratogenicity and toxicity
of atrazine using FETAX. SETAC Annual Meeting, November 18,
1988; Charlotte, NC.
Nieuwkoop, P.O., and Faber, J. (1967). Normal table of Xenopus laevis(Daudin). North Holland, Amsterdam, the Netherlands: North
Holland.
Nieves-Puigdoller, K., Björnsson, B.T., and McCormick, S.D. (2007).
Effects of hexazinone and atrazine on the physiology and endocrinology of smolt development in Atlantic salmon. Aquat. Toxicol. 84:27–
37.
Nikkilä, A., Paulsson, M., Almgren, K., Blanck, H., and Kukkonen,
J.V.K. (2001). Atrazine uptake, elimination, and bioconcentration by periphyton communities and Daphnia magna: Effects
Downloaded At: 15:43 28 October 2008
770
K. R. SOLOMON ET AL.
of dissolved organic carbon. Environ. Toxicol. Chem. 20:1003–
1011.
Novartis. (2000). Summary of Environmental Fate of Atrazine. Greensboro, NC: Novartis Crop Protection, Inc. No. Study No. 1213-99.
OECD. (2004). Detailed Review Paper on Amphibian Metamorphosis
Assay for the Detection of Thyroid Active Substances. Paris, France:
OECD. OECD Environment Health and Safety Publications Series
on Testing and Assessment No. 46. No. ENV/JM/MONO(2004)17.
http://www.epa.gov/scipoly/oscpendo/pubs/edmvac/oecd
amphibian drp.pdf
Oh, S.M., Shim, S.H., and Chung, K.H. (2003). Antiestrogenic action
of atrazine and its major metabolites in vitro. J. Health Sci. 49:65–71.
Oka, T., Tooi, O., Mitsui, N., Miyahara, M., Ohnishi, Y., Takase, M.,
Kashiwagi, A., Santo, N., and Iguchi, T. (2008). Effect of atrazine on
metamorphosis and sexual differentiation in Xenopus laevis. Aquat.
Toxicol. 87:215–226.
Orton, F., Carr, J.A., and Handy, R.D. (2006). Effects of nitrate and
atrazine on larval development and sexual differentiation in the
Northern leopard frog Rana pipiens. Environ. Toxicol. Chem. 25:65–
71.
Park, J.-W., Hecker, M., Murphy, M.B., Jones, P.D., Solomon, K.R.,
van der Kraak, G.J., Carr, J.A., Smith, E.E., Du Preez, L.H., Kendall,
R.J., and Giesy, J.P. (2006). Development and optimization of a Q-RT
PCR method to quantify CYP19 mRNA expression in testis of male
adult Xenopus laevis: Comparisons with aromatase enzyme activity.
Comp. Biochem. Physiol. B 144:18–28.
Pastoor, T., Stevens, J.T., and Breckenridge, C.B. (2008). Evaluation of
the potential effects of atrazine and its metabolites on the immune
system: Routine toxicity studies conducted in rat, mouse and dog.
Society of Toxicology; Seattle, WA, USA. Society of Toxicology,
Poster 2140.
Pechmann, J.E., and Wilbur, H.M. (1994). Putting declining amphibian
populations in perspective: Natural fluctuations and human impacts.
Herpetologica 50:65–84.
Peterson, G., Johnson, L., Axler, R.P., and Diamond, S.A. (2002). Assessment of the risk of solar ultraviolet radiation to amphibians. II.
In situ characterization of exposure in amphibian habitats. Environ.
Sci. Technol. 36:2859–2865.
Pieau, C., Dorizzi, M., and Richard-Mercier, N. (1999). Temperaturedependent sex determination and gonadal differentiation in reptiles.
Cell Mol. Life Sci. 55:887–900.
Pistl, J., Kovalkovicova, N., Legath, J., Novotny, J., Holovska, V.,
and Mikula, I. (2002). Metabolic activity of sheep peripheral blood
phagocytes after exposure to selected pesticides in vitro. Bull. Vet.
Inst. Pulawy 46:247–253.
Pounds, J.A., and Crump, M.L. (1994). Amphibian declines and climate
disturbance: The case of the golden toad and the harlequin frog.
Conserv. Biol. 8:72–85.
Pounds, J.A., Fogden, M.P.L., Savage, J.M., and Gorman, G.C. (1997).
Tests of null models for amphibian declines on a tropical mountain.
Conserv. Biol. 11:1307–1322.
Pruett, S.B., Fan, R., Zheng, Q., Myers, L.P., and Hebert, P. (2003).
Modeling and predicting immunological effects of chemical stressors: Characterization of a quantitative biomarker for immunological
changes caused by atrazine and ethanol. Toxicol. Sci. 75:343–354.
Qin, Z.F., Zhou, J.M., Chu, S.G., and Xu, X.B. (2003). Effects of Chinese domestic polychlorinated biphenyls (PCBs) on gonadal differentiation in Xenopus laevis. Environ. Health Perspect. 111:553–556.
Reeder, A.L., Foley, G.L., Nichols, D.K., Hansen, L.G., Wikoff, B.,
Faeh, S., Eisold, J., Wheeler, M.B., Warner, R., Murphy, J.E., and
Beasley, V.R. (1998). Forms and prevalence of intersexuality and
effects of environmental contaminants on sexuality in cricket frogs
(Acris crepitans). Environ. Health Perspect. 106:261–266.
Reeder, A.L., Ruiz, M.O., Pessier, A., Brown, L.E., Levengood, J.M.,
Phillips, C.A., Wheeler, M.B., Warner, R.E., and Beasley, V.R.
(2005). Intersexuality and the cricket frog decline: Historic and geographic trends. Environ. Health Perspect. 113:261–265.
Rhodes, W.E. (1998). Health of alligator populations in South Carolina.
In: Kendall R.J., Dickerson R., Giesy J.P., and Suk W., eds. Principles
and Processes for Evaluating Endocrine Disruption in Wildlife, pp.
301–310. SETAC Press, Pensacola, FL.
Richter, S.C., Young, J.E., Johnson, G.N., and Seigel, R.A. (2003).
Stochastic variation in reproductive success of a rare frog, Rana
sevosa: Implications for conservation and for monitoring amphibian
populations. Biol. Conserv. 111:171–177.
Robertson, J.C., and Kelley, D.B. (1996). Thyroid hormone controls
the onset of androgen sensitivity in the developing larynx of Xenopus
laevis. Dev. Biol. 176:108–123.
Rohr, J.R., Elskus, A., A, Shepherd, B.S., Crowley, P.H., McCarthy,
T.M., Neidzwiecki, J.H., Sager, T., Shi, A., and Palmer, B.D. (2003).
Lethal and sublethal effects of atrazine, carbaryl, endosulfan, and
octylphenol on the streamside salamander (Ambystoma barbouri).
Environ. Toxicol. Chem. 22:2385–2392.
Rohr, J.R., Elskus, A.A., Shepherd, B.S., Crowley, P.H., McCarthy,
T.M., Niedzwiecki, J.H., Sager, T., Sih, A., and Palmer, B.D. (2004).
Multiple stressors and salamanders: Effects of an herbicide, food
limitation, and hydroperiod. Ecol. Appl. 14:1028–1040.
Rohr, J.R., and Crumrine, P.W. (2005). Effects of an herbicide and an
insecticide on pond community structure and processes. Ecol. Appl.
15:1135–1147.
Rohr, J.R., and Palmer, B.D. (2005). Aquatic herbicide exposure increases salamander desiccation risk eight months later in a terrestrial
environment. Environ. Toxicol. Chem. 24:1253–1258.
Rohr, J.R., Sager, T., Sesterhenn, T.M., and Palmer, B.D. (2006). Exposure, postexposure, and density-mediated effects of atrazine on amphibians: Breaking down net effects into their parts. Environ. Health
Perspect. 114:46–50.
Saglio, P., and Trijasse, S. (1998). Behavioral responses to atrazine and
diuron in goldfish. Arch. Environ. Contam. Toxicol. 35:484–491.
Sanderson, H. (2002). Pesticide studies: Replicability of micro/mesocosms. Environ. Sci. Pollut. Res. 9:429–435.
Sanderson, J.T., Seinen, W., Giesy, J.P., and van den Berg, M. (2000).
2-Chloro-s-triazine herbicides induce aromatase (CYP19) activity in
H295R human adrenocortical carcinoma cells: A novel mechanism
for estrogenicity? Toxicol. Sci. 54:121–127.
Sanderson, J.T., Letcher, R.J., Heneweer, M., Giesy, J.P., and van den
Berg, M. (2001). Effects of chloro-s-triazine herbicides and metabolites on aromatase activity in various human cell lines and on vitelllogenin production in male carp hepatocytes. Environ. Health Perspect.
109:1027–1031.
Sassoon, D., and Kelley, D.B. (1986). Androgen-induced myogenesis
and chondorgenesis in the larynx of Xenopus laevis. Am. J. Anat.
177:457–472.
Sessions, S.K., Franssen, R.A., and Horner, V.L. (1999). Morphological clues from multilegged frogs: Are retinoids to blame? Science
284:800–801.
Downloaded At: 15:43 28 October 2008
ATRAZINE EFFECTS ON FISH, AMPHIBIANS, REPTILES
Shafer, T.J., Ward, T.R., Meacham, C.A., and Cooper, R.L. (1999).
Effects of the chlorotriazine herbicide, cyanazine, on GABA(A) receptors in cortical tissue from rat brain. Toxicology 142:57–68.
Sherman, C.K., and Morton, M.L. (1993). Population declines of
Yosemite toads in the eastern Sierra Nevada of California. J. Herpetol. 27:186–198.
Smith, E.E., du Preez, L.H., Gentles, B.A., Solomon, K.R., Tandler,
B., Carr, J.A., Van Der Kraak, G.J., Kendall, R.J., Giesy, J.P., and
Gross, T.S. (2005). Assessment of laryngeal muscle and testicular
cell types in Xenopus laevis (Anura Pipidae) inhabiting maize and
non-maize growing areas of South Africa. Afr. J. Herpatol. 54:69–
76.
Solomon, K.R., Baker, D.B., Richards, P., Dixon, K.R., Klaine, S.J., La
Point, T.W., Kendall, R.J., Giddings, J.M., Giesy, J.P., Hall, L.W.J.,
Weisskopf, C., and Williams, M. (1996). Ecological risk assessment of atrazine in North American surface waters. Environ. Toxicol.
Chem. 15:31–76.
Solomon, K.R., and Takacs, P. (2002). Probabilistic risk assessment
using species sensitivity distributions. In: Posthuma L., Suter G.W.,
and Traas T., eds. Species Sensitivity Distributions in Ecotoxicology,
pp. 285–313. CRC Press, Boca Raton, FL.
Spanó, L., Tyler, C.R., Van Aerle, R., Devos, P., Mandiki, S.N.M.,
Thomé, J.-P., and Kestemont, P. (2004). Effects of atrazine on sex
steroid dynamics, plasma vitellogenin concentration and gonad development in adult goldfish (Carasius auratus). Aquat. Toxicol.
66:369–379.
Sparling, D.W., Feller, G.M., and McConnell, L.L. (2001). Pesticides and amphibian declines in California. Environ. Toxicol. Chem.
20:1591–1595.
Sparling, D.W., and Feller, G.M. (2007). Comparative toxicity of chlorpyrifos, diazinon, malathion and their oxon derivatives to larval Rana
boylii. Environ. Pollut. 147:535–539.
Speare, R. (2001). Recommendations from Workshop in Getting the
Jump on Amphibian Disease. In: Speare R., ed. Getting the Jump
on Amphibian Disease, Developing Management Strategies to Control Amphibian Diseases: Decreasing the Risks Due to Communicable Diseases, pp. 131–147. School of Public Health and Tropical
Medicine, James Cook University, Townsville, NZ.
Stoker, T.E., Guidici, D.L., Laws, S.C., and Cooper, R.L. (2002). The
effects of atrazine metabolites on puberty and thyroid function in the
male Wistar rat. Toxicol. Sci. 67:198–206.
Storrs, S.I., and Kiesecker, J.M. (2004). Survivorship patterns of larval amphibians exposed to low concentrations of atrazine. Environ.
Health Perspect. 112:1054–1057.
Stuart, S.N., Chanson, J.S., Cox, N.A., Young, B.E., Rodrigues, A.S.L.,
Fischman, D.L., and Waller, R.W. (2004). Status and trends of
amphibian declines and extinctions worldwide. Science 306:1783–
1786.
Sullivan, K.B., and Spence, K.M. (2003). Effects of sublethal concentrations of atrazine and nitrate on metamorphosis of the African
clawed frog. Environ. Toxicol. Chem. 22:627–635.
Suzaw,a M., and Ingraham, H.A. (2008). The herbicide atrazine
activates endocrine gene networks via non-steroidal NR5A nuclear receptors in fish and mammalian cells. PLoS ONE. 3:e2117.
doi:2110.1371/journal.pone.0002117.
Tavera-Mendoza, L., Ruby, S., Brousseau, P., Fourier, M., Cyr, D.,
and Marcogliese, D. (2002a). Response of the amphibian tadpole
(Xenopus laevis) to atrazine during sexual differentiation of the testis.
Environ. Toxicol. Chem. 21:527–531.
771
Tavera-Mendoza, L., Ruby, S., Brousseau, P., Fourier, M., Cyr, D.,
and Marcogliese, D. (2002b). Response of the amphibian tadpole
Xenopus laevis to atrazine during sexual differentiation of the ovary.
Environ. Toxicol. Chem. 21:1264–1267.
Tavera-Mendoza, L.E. (2001). Influences of atrazine on gonadal
differentiation in Xenopus laevis tadpoles during metamorphosis [MSc thesis]. Montreal, PQ, Canada: Concordia University.
73 p.
Tierney, K.B., Singh, C.R., Ross, P.S., and Kennedy, C.J. (2007). Relating olfactory neurotoxicity to altered olfactory-mediated behaviors
in rainbow trout exposed to three currently-used pesticides. Aquat.
Toxicol. 81:55–64.
Tobias, M.L., Marin, M.L., and Kelley, D.B. (1993). The roles of sex,
innervation, and androgen in laryngeal muscle of Xenopus laevis. J.
Neurosci. 13:324–333.
Trentacoste S.V., Friedman A.S., Youker R.T., Breckenridge C.B.,
and Zirkin B.R. (2001). Atrazine effects on testosterone levels and
androgen-dependent reproductive organs in peripubertal male rats. J
Androl. 22:142–148.
U.S. Fish and Wildlife Service. (1984). Houston Toad Recovery Plan.
Albuquerque, NM: United States Fish & Wildlife.
U.S. Fish and Wildlife Service. (2001). Wyoming Toad (Bufo baxteri)Population and Habitat Viability Assessment (PHVA). Laramie,
WY: U.S. Fish and Wildlife Service.
U.S. Fish and Wildlife Service. (2003). Threatened and Endangered Species System (TESS). United States Fish and
Wildlife Service web site http://ecos.fws.gov/tess public/servlet/
gov.doi.tess public.servlets.VipListed?code=V&listings=0#D, accessed July 31 2005.
U.S. Environmental Protection Agency. (1989). Methods for the Determination of Organic Compounds in Drinking Water. Cincinnati, OH:
U.S. Environmental Protection Agency. No. EPA-600/4-88/039.
U.S. Environmental Protection Agency. (1998). Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC) Final Report. Washington, DC: Endocrine Disruptor Screening
and Testing Advisory Committee. http://www.epa.gov/scipoly/
oscpendo/edspoverview/finalrpt.htm
U.S. Environmental Protection Agency. (2000). Atrazine: Evaluation of Carcinogenic Potential. Washington, DC: U.S. Environmental Protection Agency. Memorandum, Office of Prevention,
Pesticides, and Toxic Substances. No. HED DOC. NO 014431.
http://www.epa.gov/pesticides/reregistration/atrazine/carc-rpt.pdf
U.S. Environmental Protection Agency. (2003a). White Paper on
Potential Developmental Effects of Atrazine in Amphibians.
Washington, DC: U.S. Environmental Protection Agency, Office of Prevention, Pesticides and Toxic Substances, Office of
Pesticide Programs, Environmental Fate and Effects Division.
www.epa.gov/scipoly/sap/index.htm
U.S. Environmental Protection Agency. (2003b). A Set of Scientific Issues Being Considered by the Environmental Protection
Agency Regarding: Potential Developmental Effects of Atrazine on
Amphibians. Washington, DC: United States Environmental Protection Agency. Report of the FIFRA Scientific Advisory Panel
Meeting, June 17–20, 2003,. No. SAP Report No. 2003-01.
www.epa.gov/scipoly/sap/index.htm
U.S. Environmental Protection Agency. (2003c). Office of Prevention, Pesticides and Toxic Substances. Interim Reregistration Eligibility Decision for Atrazine. U.S. EPA web site
http://www.epa.gov/REDs/atrazine ired.pdf, accessed July 28 2005.
Downloaded At: 15:43 28 October 2008
772
K. R. SOLOMON ET AL.
U.S. Environmental Protection Agency, Biological and Economic Analysis Division Office of Pesticide Programs Office of Prevention,
Pesticides, and Toxic Substances. (2004). Pesticides Industry Sales
and Usage 2000 and 2001: Market Estimates. U.S. EPA web site
http://www.epa.gov/oppbead1/pestsales/index.htm, accessed July 28
2005.
U.S. Environmental Protection Agency. (2005). Draft Final Report
on Multi-Chemical Evaluation of the Short-Term Reproduction Assay with the Fathead Minnow. Washington, DC: U.S. Environmental Protection Agency. No. 68-W-01-023. http://www.epa.gov/
scipoly/oscpendo/docs/edmvac/draft fish repro multichemical.pdf
U.S. Geological Survey. (2004). Where Have All the Frogs Gone?
USGS website http://www.usgs.gov/amphibian faq.html, accessed
October 5 2004.
van Tienhoven, A. (1983). Reproductive Physiology of Vertebrates.
Ithaca, NY: Cornell University Press.
Vertucci, F.A., and Corn, P.S. (1997). Evaluation of episodic acidification and amphibian declines in the Rocky Mountains. Ecol. Appl.
6:449–457.
Vredenburg, V.T. (2004). Reversing introduced species effects: Experimental removal of introduced fish leads to rapid recovery of a declining frog. Proc. Natl. Acad. Sci. USA 101:7646–7650.
Wallace, H., Badawy, G.M.I., and Wallace, B. (1999). Amphibian sex
determination and sex reversal. Cell Mol. Life Sci. 55:901–909.
Walsh, A.H., and Ribelin, W.E. (1975). The pathology of pesticide
poisoning. In: Ribelin W.E., and Migaki E., eds.The Pathology of
Fish, pp. 515–557. Madison: University of Wisconsin Press.
Waring, C.P., and Moore, A. (2004). The effect of atrazine on Atlantic
salmon (Salmo salar) smolts in fresh water and after sea water transfer. Aquat. Toxicol. 66:93–104.
Weed, D.L. (1997). Methodologic guidelines for review papers. JNCI
89:6–7.
Whale, G.F., Sheahan, D.A., and Kirby, M.F. (1994). Assessment of the
value of including recovery periods in chronic toxicity test guidelines
for rainbow trout (Oncorhynchus mykiss). In: Sublethal and Chronic
Effects of Pollutants on Freshwater Fish, pp. 175–187. Fishing News
(Books) Ltd, London.
Whale, M.M., Loganathan, G., Yamashita, N., and Saito, T. (2003).
Immunomodulation of human natural killer cell cytotoxic function
by triazine and carbamate pesticides. Chem. Biol. Interact. 145:311–
319.
Wiegand, C., Pflugmacher, S., Giese, M., Frank, H., and Steinberg,
C. (2000). Uptake, toxicity, and effects on detoxication enzymes
of atrazine and trifluoroacetate in embryos of zebrafish. Ecotoxicol.
Environ. Safety. 45:122–131.
Wiegand, C., Krause, E., Steinberg, C., and Pflugmacher, S. (2001).
Toxicokinetics of atrazine in embryos of the zebrafish (Danio rerio).
Ecotoxicol. Environ. Safety. 49:199–205.
Winkelmann, D.A., and Klaine, S.J. (1991). Degradation and
bound residue formation of four atrazine metabolites, deethylatrazine, deisopropylatrazine, dealkylatrazine and hydroxyatrazine,
in a western Tennessee soil. Environ. Toxicol. Chem. 10:347–
354.
Witschi, E. (1929). Studies on sex differentiation and sex determination
in amphibians. III. Rudimentary hermaphrodites and Y chromosome
in Rana temporaria. J. Exp. Zool. 54:157–222.
Wolf, J.C. (2007). Supplementary analysis of X. laevis gonadal histology from Carr et al., 2003 and Coady et al., 2005 investigations. Greensboro, NC: Syngenta Crop Protection, Inc. Draft. No.
T000799-07.
Young, B.E., Lips, K.R., Reaser, J.K., Ibáñez, R., Salas, A.W., Cedeño,
J.R., Coloma, L.A., Ron, S., La Marca, E., Meyer, J.R., Muñoz, A.,
Bolaños, F., Chaves, G., and Romo, D. (2001). Population declines
and priorities for amphibian conservation in Latin America. Conserv.
Biol. 15:1213–1223.
Zeeman, M.G., and Brindley, W.A. (1981). Effects of toxic agents upon
fish immune systems: A review. In: Sharma R.P., ed. Immunologic
Consideration in Toxicology, pp. 1–47. CRC Press, Boca Raton, FL.