Patuxent Wildlife Research Center

NAAMP III Archive - deformed frogs
Home | Archive by Alphabetical Order | Archive by Category

Discussion of Findings
Relative to Meeting Objectives

Joseph E. Tietge, Michael Lannoo, and Val Beasley

U.S. Environmental Protection Agency
Duluth, MN

Michael Lannoo
Ball State University, Muncie, IN

Val Beasley
Univ. of Illinois, Urbana, IL Mid-Continent Ecology Division

Disclaimer: The summary information presented for the workshop, to the best of our knowledge, is accurate. The opinions or policies which may appear in the presentations from the workshop are those of the respective authors and do not represent the policy or opinion of the U.S. Environmental Protection Agency.

For summary of presentaions and introduction, please see:
Summary of Workshop on Central North American Amphibian Deformities

Review existing data to establish whether or not the apparent increase in amphibian deformities is an environmental problem.

Establishing whether or not the current reports of amphibian deformities represent an environmental problem is mostly dependent upon a comparison of current deformity prevalences with historical deformity prevalences. Although there have been a few reports of amphibian deformities in isolated populations, workshop participants were unaware of any published information on the historical prevalence of deformities in feral amphibian populations. This data gap creates uncertainty about any further conclusions as to whether current deformity prevalences are different than historical deformity prevalences. In the absence of comparative data, it is reasonable and necessary to rely on the opinions of experts. There were several very experienced herpetologists at the workshop who had each logged in decades of anuran field work. It was the consensus of these herpetologists that they would have noted deformities in the past had they been as prevalent as they are reported to be now. Therefore, it was their opinion that the current prevalence of deformities is abnormally high.

Our knowledge of the geographical distribution of the apparent increase in amphibian deformities has been changing since the workshop was held. In fact, since the workshop, reports have appeared in the press indicating the presence of amphibian deformities in several states that previously had no reports. There is a great deal of uncertainty over the actual geographic distribution due to a lack of systematic field sampling of the deformity endpoint, the unconfirmed nature of press reports, and the lack of a central reporting center to collect and analyze data. The states or provinces reporting deformities include: Minnesota, Wisconsin, Quebec, Ohio, California, Kansas, Missouri, and Vermont. However, this geographic distribution is viewed as incomplete at this time.

Abnormal increases in deformities seems to be occurring in several species. This observation is key to discussions of cause and effect since it appears, to some degree, that the phenomenon occurs across several species. Indeed, differential sensitivity of different species may provide useful information for cause and effect studies. The species which have been reported with deformities include: northern leopard frog (Rana pipiens), mink frog (R. septentrionalis), bull frog (R. catesbiana), wood frog (R. sylvatica), green frog (R. clamitans), American toad (Bufo americanus), gray tree frog (Hyla versicolor), spring peeper (Pseudacris crucifer), Pacific tree frog (H. regilla), and long-toed salamander (Ambystoma macrodactylum).

It is completely unclear at this time if the increase in deformities is responsible for any population changes. It stands to reason that the deformities are maladaptive and would lead to death of the individual due to locomotor deficiencies. Although no studies have been conducted to assess this hypothesis directly, it has been supported by the observation that essentially all of the deformed individuals have been part of the year class that just metamorphosed. Few, if any, deformed individuals have been observed as adults, implying that survival to maturity is unlikely. In addition, the grossly observable deformities may not represent the total numbers or types of developmental abnormalities in any one population. Indeed, the frogs with limb deformities may be a segment of an affected population that is able to survive, while other abnormalities in other individuals may be lethal. If one assumes that there are some additional lethal deformities, then estimates of population effects based on gross limb deformities may not be accurate. Clearly, there is a need to study other deformities in populations where gross limb deformities are apparent.

Review and evaluate possible mechanisms by which deformities could be manifest.

Since the process of development is complex, it is likely that there are several different mechanisms which could be adversely affected in a way that results in developmental abnormalities. There were three classes of developmental mechanisms considered: genetic, regulatory, and other. Genetic mechanisms, both heritable gene defects and somatic cell DNA damage, were considered unlikely to be relevant to the current increase in amphibian deformities, although both were considered as possible mechanisms. Heritable gene defects were eliminated from consideration given the multiple species affected. Somatic cell DNA damage was considered unlikely because it would require similar target genes to be effected in multiple species.

The regulatory signaling necessary to control limb development is quite complex and generally requires both spacial and temporal concentration gradients of signaling molecules. Even though this process is complex, it is well conserved across many classes of organisms. The normal endogenous control of these molecules is achieved through regulation of gene transcription and translation, and through other metabolic pathways. The signaling molecules could have xenobiotic analogs which have either agonist or antagonistic effects. Retinoic acid, for example, is an important endogenous chemical necessary for proper limb differentiation (retinoic acid acts as the ligand to a receptor which is in the nuclear hormone receptor superfamily).

There are undoubtedly other developmentally related mechanisms not mentioned at this workshop which could be relevant. For example, physical disruption of the developing limb field can result in a discontinuity of existing cellular gradients which are important to establishing cellular location and identity. It is hypothesized that this disrupted field attempts to re-establish the cellular level gradients which results in a process termed intercalation. The intercalation can thereby cause a whole host of developmental anomalies, including supernumerary limbs.

Review and evaluate possible environmental causes of deformities.

Several possible environmental causes of amphibian deformities were presented by the speakers and were reviewed and discussed by the participants. None of the possible causes were supported by evidence from the field, so they all remain hypothetical at this time. The least likely cause was considered to be acidification. This conclusion was based on the fact that the deformities were occurring in areas where the waters are sufficiently buffered to minimize low pH excursions during acidifying events such as precipitation and spring run off. Low pH is known to be toxic to amphibian larvae and would likely have adverse effects under chronic exposure conditions.

UV-B radiation was considered as a possible cause since it is know to elicit adverse biological effects and has been demonstrated to be on the increase in northern latitudes due to ozone depletion. Biological studies demonstrate that UV-B alone is toxic and could be contributing to amphibian population declines. UV-B has also been shown to have adverse effects on developmental patterning in amphibians when the exposure occurs before or near fertilization. One criticism of the hypothesis that UV-B can cause deformities, is that the cumulative incidence is higher overall in the southern latitudes. However, total cumulative incidence is not a relevant measurement of UV-B in the context of amphibian development. The incidence that is relevant is that which occurs during spawning and early development. In northern latitudes, the longer day lengths in late spring and early summer do correspond to the early developmental stages of several amphibians. This relationship warrants further analysis. Another aspect of UV-B radiation is its ability to photoactivate chemicals from a non-toxic form to a toxic form. This characteristic of UV-B is well studied and suggests that chemical contamination in combination with UV-B could have developmental effects. This combination could be consistent with the apparent recent increase in deformities and with the broad geographic distribution of deformities.

Xenobiotic chemicals are plausible causative agents based on the fact that specific chemicals have been used for decades to modulate the process of organ and limb patterning in developmental biology studies. Clearly, there is a whole body of teratological data which supports this plausibility as well. Although there are many possible pathways in which xenobiotic chemicals could hypothetically interfere with pattern development, one of the best studied is the retinoic acid pathway. Exogenously applied retinoic acid, for example, has the ability to drastically modify normal development of limbs, resulting in hypermorphic or hypomorphic development depending on the time of application. Recent evidence demonstrates that there are xenobiotic chemicals which are able to bind to at least one of the retinoid receptors, suggesting that these chemicals could induce developmental abnormalities if they are present at the right time and concentration.

Viral, bacterial, fungal, and parasitic diseases are common in feral amphibian populations and are possible causative agents. However, not much is known about the role of viral, bacterial, or fungal infections on limb development. Trematode parasites, on the other hand, are particularly plausible causes, since evidence from studies in California demonstrate that the metacercarial cyst of a trematode parasite was sufficient to cause a range of limb deformities in frogs and salamanders. It was hypothesized that the cyst caused a solid state obstruction in the developing limb field and that the normal patterning was disrupted and the process of intercalation ensued. This phenomenon was demonstrated in the laboratory by the insertion of beads into the limb field of developing organisms. The implanted beads caused similar deformities as the encysted trematodes, suggesting that the phenomenon is indeed a physical disruption of the developing limb field.

There is a distinct possibility that causation is due to a complex interaction of multiple stressors. That is, there may be several factors acting in combination which result in abnormal development. The combinations could include additive and synergistic factors, such as the hypothesis that photoactivation of xenobiotic chemicals by UV-B converts the chemical from a relatively non-toxic state to a highly toxic state which affects developmental processes. Furthermore, the cause of the deformities could be secondary to another phenomenon. A plausible hypothesis is that environmental changes which affect non-amphibian host status result in an abnormal increase and distribution of a parasitic trematode which is associated with deformities. In turn, higher infestation rates in amphibians could result and lead to an increase in deformities.

Identify research needs to address this issue.

It is important at this time to establish the distribution and prevalence of deformities in anuran populations. Current reports and methods are insufficient to quantify the phenomenon and more emphasis needs to be placed on properly designed sampling to avoid bias. Formation of a workgroup is recommended to deal with: sampling methods, specimen handling, data objectives, and standardized data reporting and analysis. Abnormalities need to be described in proper technical terms to facilitate inter-species and inter-site comparisons and to help in making inferences regarding the possible mechanisms involved. Deformity distribution maps should be generated by deformity type, species, and year. Field surveys should focus on metamorphosing frogs. Voucher specimens may be needed depending on the objectives of each study. Additional effort needs to be put into literature review, especially to include "grey" literature and international literature.

If the increased prevalence of deformities in anurans is real, then research needs to be conducted to determine the cause. And, while there are no data to exclude the broad categorical causes mentioned above, it was suggested that initial research should focus on the parasitic and chemical etiologies. Both of these causes have precedence in the scientific literature. Parasites have been shown to be the likely cause of an increase of amphibian deformities at a California site. This association was supported by laboratory studies using implanted beads. While this study seems reasonably conclusive for the species involved at that site, uncertainty about this cause for all sites arises from the fact that multiple species are involved over an extensive geographical area. Some chemicals, on the other hand, are well documented teratogens, suggesting that chemical teratogenesis may be involved. This argument is strengthened by the developmental literature which demonstrates that strikingly similar deformities can be caused by exogenous application of chemicals in limb development models. Even with the prioritization of parasites and chemicals as plausible causes warranting research, the participants recognized that there is so little information at this time, the research agenda should not be too narrow. Cause and effect field studies should focus on sites with high deformity rates and deformities in more than one species, if possible. Both laboratory and field work should be framed with Koch's postulates and epidemiological procedures in mind.

There is a great deal of uncertainty about the effect of amphibian deformities on population status. The reduced fitness of the deformed individuals seems obvious and they are unlikely to be recruited into the adult population. High incidences of deformities are likely to result in local population declines, if one assumes that all deformed individuals are unable to reproduce due to mortality or dysfunction. The effects of deformities on population status should be further investigated.

It was obvious at the time of the workshop that a national reporting center needed to be established to collect and analyze data. This idea has been reinforced further by numerous reports of deformed amphibians since the workshop and the subsequent publicity. States are in the best position to receive reports and establish their validity, but they are generally not in a position to handle information on a national scale. A national reporting center is recommended to facilitate data standardization, establish a national database, and to conduct regional or national analyses.

A considerable amount of discussion focussed on the use of citizens in data collection. Obviously, citizen reports are crucial to recognizing some environmental problems and were important in the events which led to the convening of the workshop. However, when it comes providing the type of data required by hypothesis driven research, it is unlikely that citizen efforts will always be effective. Therefore, in general, expansion of citizen-based sampling was not advocated, unless their efforts can be channeled through research programs which establish standards and training.

SUMMARY STATEMENTS

1. Deformities are uncommon in natural populations of anurans, though not unknown. There are a few reports of amphibian deformities in the published literature.

2. Recent evidence suggests that the current incidence of deformities in amphibian populations is abnormally high and that there may be an environmental factor affecting it. This conclusion is based on:

-Expert opinions of experienced herpetologists suggest that limb abnormalities are extremely rare in natural populations.

-The incidence of limb abnormalities in laboratory raised animals is very low.

It should be stressed, however, that due to the lack of historical or baseline data, it is difficult to estimate the change in deformity incidence rates with confidence

3. Systematic regional sampling needs to be conducted to develop adequate data to calculate prevalence of deformities in anuran populations and ultimately to determine if the apparent developmental problems are relevant to amphibian population declines.

-A working group should be established to detail sampling methods etc.

4. More than one phenomenon may be occurring simultaneously.

-Different types of limb malformations could be unrelated.

5. Limb malformations may represent only one of several developmental anomalies. There are no data on other developmental endpoints at this time.

6. Since a parasitic mechanism has been documented as the likely cause of limb abnormalities in a California site, parasite evaluation should be conducted on individuals and perhaps on specific populations.

7. Although no causes have been completely eliminated, the role of parasites and the role of xenobiotic chemicals (perhaps in concert with UV-B) should have a higher priority than other possible causes.

-Parasites have been shown to be sufficient to cause specific types of malformations.

-There are well documented developmental mechanisms that could be affected by xenobiotic chemicals, such as those that affect the retinoic acid system.

8. Citizen reports have been crucial to identifying amphibian deformities as a possible environmental problem. Citizen efforts need to be directed toward appropriate research programs that provide oversight and training. However, at this time, further use of citizen-based sampling should be discouraged based on the following concerns:

-Unnecessary sampling could be adverse to amphibian populations.

-Widespread sampling by the public could result in unnecessary habitat disturbances in wetlands.

-Data collected by untrained individuals may not meet the objectives of future studies. Furthermore, it is difficult or impossible to enforce data quality assurance.

9. Data are being collected by various state agencies; however, it is currently difficult to conduct regional or national analyses of these isolated databases. A national reporting center should be established to support the analysis of data collected by states, universities, museums, etc.

U.S. Department of the Interior
U.S. Geological Survey
Patuxent Wildlife Research Center
Laurel, MD, USA 20708-4038
http://www.pwrc.usgs.gov/naamp3/naamp3.html
Contact: Sam Droege, email: Sam_Droege@usgs.gov
Last Modified: June 2002