Abstract

Reports of amphibians with extra limbs in natural populations have been a puzzle for decades. When such animals occur as single individuals they are usually dismissed as isolated developmental anomalies. However, when populations show a high frequency of individuals with extra limbs this suggests that a major biological perturbation has spread through the population.

In this paper we report the discovery of a pond in which limb abnormalities are caused by parasitic larvae of a digenetic trematode that uses the amphibians as intermediate hosts. The cercarial stage of the clinostomid trematode attack the amphibians, penetrating the skin to form cysts (metacercaria). The cysts are preferentially localized in the cloacal and gill region, including the front and hind limbs of both salamanders (Ambystoma macrodactylum) and frogs (Hyla regilla). When the parasites attack young amphibian larvae in which the hind limbs and digits have not fully developed, the cysts cause morphological disruptions of the developing limbs. A wide range of limb abnormalities are seen, including ectopic limb structures ranging from extra digits to several extra whole limbs. Experiments with implanted inert, 200 micron beads, approximately the same size as the metacercarial cysts, were sufficient to induce supernumerary structures without postulating any kind of growth stimulating substance being secreted. However, this does not rule out that possibility.

Introduction

Reports of amphibians with extra limbs in natural populations have been in the scientific literature for centuries (1). In this paper we report the occurrence of a high frequency of hind limb abnormalities, including extra limbs and extra digits, in a population of Pacific Tree Frogs (Hyla regilla), and Long Toed Salamanders (Ambystom macrodactylum), in which we have identified the causative factor as a parasitic flatworm (Trermtoda:Clinostormtidae). The multilegged amphibians reported here were discovered by one of us (SR) in a pond during the spring of 1986 while conducting a survey of endangered populations of A. macrodactylurn in northern California.

Associated with these ectopic hind limb structures was a massive infestation of an encysted larval stage of a digenetic trematode. In an effort to identify the parasite involved and its possible life cycle in the pond, samples of snails, frogs and salamanders were collected and returned to the laboratory at CSU Long Beach.

Materials and Methods

Tadpoles and newly metamorphosed frogs were fixed, stained and cleared in glycerin for study. Metacercaria were removed from infested areas of both frogs and salamanders. The metacercaria were stained, cleared and mounted for identification.

Living animals (snails, frogs, salamanders) from the pond were established in aquaria for experimental infections and the isolation of sporocyst, redial and cercarial stages of the clinostomatid trematode.

The experiment to test the possibility that supernumerary amphibian limbs could be caused by the mechanical effects of the cysts was conducted by one of us (SKS). Inert, 200 micron beads were implanted into developing limb buds of salamanders (axolotls) and frogs (Xenopus).

Results and Conclusions

The frogs were in various stages of development when they were collected, from tadpoles at pre-limb bud stages to fully metamorphased froglets. It was found that approximately 70% of the collected frogs had gross hind limb abnormalities, including variable numbers of ectopic limb structures. A wide range of hind limbs were seen in these frogs as well as salamanders, and ectopic structures (digits, long bones, or both) sprout from anywhere along the length of a normal limb, or from their own extra pelvic structures. As in most previous reports of frogs with extra limbs (1,2), none of the frogs or tadpoles that we examined show abnormalities in the structure of their front limbs Salamanders collected from the same pond showed a much lower frequency (30%) of limb abnormality than frogs, but did have altered anterior limb structures.

Associated with these ectopic limb structures were massive infestation of an encysted larval digenetic trematode. These stages varied in size from 0.3 mm to 6.5 mm in length depending on the size of the host animal. The older salamanders harbored the mature, large, metacercarial stage that were identified as a species of the genus Clinostonum. The implanted inert, 200 micron beads (approximately the same size as the metacercarial cysts) did indicate that the presence of inert foreign bodies in developing lirth buds could induce supernumary structures.

Our conclusion from these observations is that the limb abnormalities in these amphibians are caused by the disruption of developing tissues by the trermtode metacercarial cysts of the genus Clinostonum. This conclusion is supported not only by the association between limb abnormalities and the presence and localized concentrations of metacercariae in two unrelated species of amphibians in the same pond, but also by the visable disruption of the hind limb forming regions by cysts in young tadpoles. Further convincing evidence comes from a single salamander specimen which showed a one-to-one association between cysts and extra digits. It is our feeling that the cysts act as solid obstacles that passively interfere with normal cell interactions during limb development. At early stages, the cysts mechanically partition the developing limb field and morphological abnormalities result from regulatory responses of the affected tissues within the subdivisions. At later stages the cysts cause localized injuries to limb structures triggering a regenerative response. In frogs, which generally lose the ability to regenerate limb structures during ontogeny (3), metacercarial infestation results in ectopic limb structures only during the larval stage. In salamanders, which can regenerate throughout life, infestation can result in ectopic limb structures at any time during ontogeny. The morphological effects of the infestation most likely depends on the stage of limb development at which the amphibian larvae are exposed to the cercariae. At early stages major disruptions of the limb pattern are expected, including extra whole limbs. At later stages the consequences would be correspondingly diminished.

There are two explanations for why the hind limbs are preferentially affected in the frogs while the throat and anterior region are more heavily infested in the salamander. The life cycle of Clinostomum attentuatum Cort, 1913, utilizes the mouth of aquatic birds for the-adult worms. The eggs are shed while the bird feeds on tadpoles, frogs, salamanders, etc. The eggs are washed into the pond where they hatch into the miracidial stage which in turn burrows into the rams horn snails of the genus Helisoma. Snails of this genus were the most common type collected from the study site. The fluke undergoes development through the sporocyst, redial and cercarial stages in the snail. The cercaria, when shed are attracted to the natural openings in the intermediate hosts. In this case, the cloacal opening of the frogs and gill chambers of the larval salamanders. In frogs the front limbs develop protected within the gill pouch.

In this study it appears the extra limbs seen in these intermediate hosts constitute a negative survival advantage in the frogs and endangered salamanders. The extra legs are apparently poorly innervated since they are nearly incapable of movement and are dragged passively during swimming. Furthermore, most of the captive multilegged frogs in this study died soon after metamorphosis. However, we have evidence that frogs with extra limb structures can survive to reproductive maturity if the abnormality is sufficiently limited. Of the adult frogs collected in 1987 at the same pond that yielded multilegged froglets in 1986, approximately 2% had limb abnormalities and these were relatively minor. If there is a selective advantage in this system it belongs to the trematode, since the handicapped frogs, acting as intermediate hosts in the parasites life cycle, make easy prey for the primary host, in this case the heron or bittern. Increased levels of trematode infestations, however, would risk local extinction of the amphibian hosts that it requires for completion of the cycle. We think that the occurrence of multilegged frogs and salamanders reported in this paper reflect simply a sporadic and localized population explosion of trematodes and/or one of the hosts (e.g., snails), and that it is highly likely that many previous reports of multilegged frogs may have also involved parasites but went undetected.

Summary

The observed phenotypic effects reported here, including extra limbs, probably reflect intrinsic regulatory properties of the cells of developing tissues reacting to an extrinsic perturbation (infestation of Clinostomum attenuatum cysts). Similar effects on limb morphology can be obtained experimentally by disrupting positional relationships between cells in the developing or regenerating limbs of amphibians. These phenotypically plastic effects underscore the resiliency of normal epigenetic patterning mechanisms in the vertebrate limb; no genetic mutations are required.

References

1.  Van Valen, L. 1974. J. Herpetol. 8:109.

2.  T.D. Reynolds and T.D. Stephens. 1984. Great Basin Naturalist. 44:166.

3.  K. Muneoka, G. Haller - Dinsmore and S.V. Bryant. 1986. J. Exp. Zool. 240, 47.