This study demonstrated that several anuran genera (Xenopus, Lithobates [Rana], Hyla, and Anaxyrus [Bufo]), as well as Nile monitor lizards, green anoles, and featherfin catfish, are susceptible to infection with D. insignis and/or D. medinensis L3s. We also found that D. insignis and D. medinensis larvae can persist in anuran tissues for at least eight and two months, respectively, although the number of L3s recovered from each infected animal was generally low. Regardless, these data show that these animals could serve as paratenic hosts if they ingest infected copepods in nature and are subsequently ingested by an appropriate definitive host.
We exposed animals using two different methods (group batch or by mouth [PO]), but aimed to primarily batch expose animals as that better mimics natural exposure. A few anuran species (i.e., American toads, Cope’s gray treefrogs, and adult African clawed frogs) were exposed to D. medinensis-infected copepods PO, as they had metamorphosed into adults before D. medinensis larvae became available for use and would be unlikely to ingest all copepods autonomously. Our primary goal in this study was to determine susceptibility to Dracunculus infection.
Six anurans that were exposed as tadpoles underwent metamorphosis to froglets before being necropsied. Dracunculus L3s were recovered from two of these animals, supporting previous findings that D. insignis larvae can persist in anuran tissues through metamorphosis14. The persistence of larvae in the tissues through metamorphosis may facilitate Dracunculus transmission from aquatic to terrestrial food chains. This could be an important factor in transmission, as the majority of definitive hosts of Dracunculus nematodes are terrestrial. This study found that, in addition to X. laevis and Lithobates spp. (which have previously been infected with Dracunculus spp. larvae), Anaxyrus sp. and Hyla sp. can also become infected with Dracunculus L3s14,21. The infection of Anaxyrus sp. and Hyla sp. is particularly interesting, as members of these genera transition to a terrestrial or arboreal existence as adults, compared to Xenopus sp. and Lithobates spp. which remain completely or predominantly aquatic, even as adults. This transition to a terrestrial habitat could carry infectious larvae further from water sources, making them available to definitive hosts more widely across the landscape. However, the role of these animals in Dracunculus transmission would still depend on many other factors, including the natural history of these amphibian species, diets of definitive hosts, and how long Dracunculus L3s persist in paratenic hosts, as terrestrial anurans would be unlikely to acquire new infections after metamorphosis.
During a previous experimental study, D. insignis L3s persisted in amphibian paratenic hosts for up to 37 DPI, at which time the animals were necropsied16. In this long-term infection trial, we found that D. insignis larvae persisted for at least 244 days (approximately eight months), while D. medinensis larvae persisted for at least 58 days (approximately two months). These results demonstrate that infection of a paratenic host can extend the time that L3s may persist in the environment well beyond the lifespan of a copepod21. As we had a limited supply of D. medinensis L3s, we were unable to conduct sufficient trials to determine whether D. insignis may persist longer in paratenic hosts than D. medinensis. If this difference was found to exist, it could contribute to the higher proportion of wild-caught adult frogs found to be infected with D. insignis than with D. medinensis during field surveys18,19. Further testing with an increased sample size would be required to determine whether the persistence of larvae actually differs between Dracunculus species or paratenic host species.
No Dracunculus larvae were recovered from the two adult African clawed frogs that were fed D. medinensis L3s that had been recovered from other paratenic hosts. It is likely that our very small sample size (two animals) and the prolonged period before necropsy (4 months) explain these negative results. In our persistence trials, there was attrition over time so these animals should have been examined earlier after exposure. Future efforts to investigate transmission of Dracunculus between different paratenic hosts should use larger sample sizes and shorter infection periods. It would also be interesting to know if predatory animals, such as Nile monitor lizards, which can experimentally become infected with Dracunculus sp. larvae could become infected by ingesting other paratenic hosts.
Fish were investigated for their potential role in Dracunculus transmission, as many fish species consume copepods as part of a natural diet25,26. Despite this, Dracunculus larvae have not been recovered during multiple studies screening wild-caught fish17,19. Dracunculus insignis L3s have rarely been recovered from previous experimental trials with fish16. When larvae were recovered from fish, larval recovery rates were very low (0.6–2.0% recovery; 1–2 larvae per fish) and only 3/43 (7.0%) of the fish harbored Dracunculus larvae upon necropsy16. In a separate trial, fish experimentally functioned as short-term transport hosts of D. medinensis and D. insignis to infect domestic ferrets7. Our findings from this trial were surprising, as we recovered up to 6 D. medinensis L3s from the tissues of three out of four (75%) exposed featherfin catfish. This fish species is common in the Chari River Basin area in Chad, Africa where high numbers of D. medinensis infections are reported in domestic dogs living in fishing villages, and is consumed by both people and dogs17. Dogs in these villages often eat discarded small fish or fish viscera4. Although our sample size was small, our current findings are evidence that some fish species may be more capable of serving as paratenic hosts for Dracunculus than those that have been previously tested. This finding further supports the continuation of the screening of wild fish muscle tissues for Dracunculus larvae.
Lizards were included in this study because large, subcutaneous nematodes (believed to be Dracunculus sp.) were historically reported from Nile monitor lizards and these lizards are consumed by people20,22. However, a lack of contemporary reports and recent work in Chad, Africa, determining that large, subcutaneous nematodes recovered from wild Nile monitor lizards were not Dracunculus sp. but actually most similar to Ochoterenella sp., suggest that monitor lizards in this region are not definitive hosts for D. medinensis17. This current study confirms that Nile monitor and green anole lizards could become infected with Dracunculus larvae. As the diet of Nile monitor lizards can include amphibians and fish, were those prey to contain Dracunculus larvae, it is possible that monitors could serve as paratenic hosts, either by ingestion of larvae in fish intestines or in tissues of amphibians or fish, although these modes of transmission to paratenic hosts have not been confirmed23,24. It is unlikely that green anoles would become naturally infected with Dracunculus spp. due to their diet and primarily arboreal habitat; however, their infection demonstrates that multiple, distantly related lizard species are susceptible to experimental infection.
Although anoles were exposed to both D. insignis and D. medinensis larvae, it is most likely that the recovered larvae were D. insignis, as only two D. medinensis-infected copepods were administered (in addition to 23 D. insignis-infected copepods). Species identity of these larvae could not be confirmed, however, as Dracunculus larvae can only be identified to species using molecular diagnostic techniques, which would destroy the sample, and these larvae were used in an experimental infection trial after recovery. Exposure of a ferret PO to the four larvae recovered from this anole (as part of a separate study) did not yield an infection, which is unsurprising given the low dose of larvae used. A previous study has shown that as few as 10 Dracunculus larvae may lead to infection of a ferret when administered interperitoneally (IP) (which was a more effective infection route than PO inoculation), therefore, four larvae administered PO would be unlikely to yield infection of a ferret27,28.
In all trials, infection occurred only in those animals that were inoculated with or exposed to at least 20 copepods per individual, suggesting an impact of parasite dose-dependent infection probability for Dracunculus infection in paratenic hosts. As copepod infection rate during this study was estimated to be (ge) 25%, it is likely that animals ingesting 20 copepods would consume at least 5 Dracunculus sp. larvae. Previous studies demonstrated that 10 larvae (administered IP) were sufficient to infect a ferret, but that percent recovery was higher with IP infection than PO27,28. It is likely that a similar minimum infectious dose also exists for paratenic hosts and may differ by paratenic host species and mode of infection. Parasite dose-dependent infection probability of Dracunculus spp. merits further investigation, as understanding this relationship could help researchers to more effectively study transmission in the laboratory by performing experimental infection trials with greater reliability.
Despite the variable sample sizes and exposure routes in this study, we demonstrated that a wide range of animals (anurans, fish, and lizards) were susceptible to infection with D. insignis and/or D. medinensis L3s. Importantly, one exposed fish species (Synodontis eupterus) was susceptible, opening up further concerns that certain fish species could serve as transport and paratenic hosts of Dracunculus species. Nile monitor lizards and anoles were successfully infected with L3s, demonstrating the first experimental infection of lizards with Dracunculus larvae. Dracunculus larvae remained L3s in the tissues of tested anurans for up to 244 days, extending the known persistence time of infectious larvae. Although no larvae were recovered from frogs that were fed L3s recovered from other paratenic hosts, continued investigation into the possibility of paratenic host to paratenic host transmission would be particularly interesting in determining if some predatory frogs (tadpoles or adults), fish, or lizards may concentrate higher numbers of L3s over time through predation of other infected paratenic hosts. Despite this study not determining how infectious larvae recovered from each of these paratenic hosts would be to another host, our findings contribute to a better understanding of the ability of these paratenic hosts to harbor Dracunculus L3s. This information is valuable to understanding how transmission to animal definitive hosts may be occurring, in addition to informing GWEP management decisions aiming to decrease transmission of D. medinensis to humans and animals.
Source: Ecology - nature.com
