Research Article: Different but overlapping populations of Strongyloides stercoralis in dogs and humans—Dogs as a possible source for zoonotic strongyloidiasis

Date Published: August 9, 2017

Publisher: Public Library of Science

Author(s): Tegegn G. Jaleta, Siyu Zhou, Felix M. Bemm, Fabian Schär, Virak Khieu, Sinuon Muth, Peter Odermatt, James B. Lok, Adrian Streit, Hans-Peter Fuehrer.

Abstract: Strongyloidiasis is a much-neglected soil born helminthiasis caused by the nematode Strongyloides stercoralis. Human derived S. stercoralis can be maintained in dogs in the laboratory and this parasite has been reported to also occur in dogs in the wild. Some authors have considered strongyloidiasis a zoonotic disease while others have argued that the two hosts carry host specialized populations of S. stercoralis and that dogs play a minor role, if any, as a reservoir for zoonotic S. stercoralis infections of humans. We isolated S. stercoralis from humans and their dogs in rural villages in northern Cambodia, a region with a high incidence of strongyloidiasis, and compared the worms derived from these two host species using nuclear and mitochondrial DNA sequence polymorphisms. We found that in dogs there exist two populations of S. stercoralis, which are clearly separated from each other genetically based on the nuclear 18S rDNA, the mitochondrial cox1 locus and whole genome sequence. One population, to which the majority of the worms belong, appears to be restricted to dogs. The other population is indistinguishable from the population of S. stercoralis isolated from humans. Consistent with earlier studies, we found multiple sequence variants of the hypervariable region I of the 18 S rDNA in S. stercoralis from humans. However, comparison of mitochondrial sequences and whole genome analysis suggest that these different 18S variants do not represent multiple genetically isolated subpopulations among the worms isolated from humans. We also investigated the mode of reproduction of the free-living generations of laboratory and wild isolates of S. stercoralis. Contrary to earlier literature on S. stercoralis but similar to other species of Strongyloides, we found clear evidence of sexual reproduction. Overall, our results show that dogs carry two populations, possibly different species of Strongyloides. One population appears to be dog specific but the other one is shared with humans. This argues for the strong potential of dogs as reservoirs for zoonotic transmission of S. stercoralis to humans and suggests that in order to reduce the exposure of humans to infective S. stercoralis larvae, dogs should be treated for the infection along with their owners.

Partial Text: Soil-transmitted helminthiasis (STH) affects up to one in four individuals in the world, disproportionately impacting impoverished populations with less access to clean water, sanitation, and opportunities for socioeconomic development [1]. Strongyloidiasis is one of the most neglected tropical diseases [2,3]. Estimates of the number of people infected with the causative agent Strongyloides stercoralis vary and go up to 370 million worldwide [2,4,5]. The local prevalence can reach more than 40% in some tropical and subtropical countries [3,6]. Factors such as high temperature, high moisture, poor sanitation and sharing premises with domestic animals may contribute to high prevalence of S. stercoralis [3,7,8]. S. stercoralis is the major causative agent of human strongyloidiasis [9] but there are also reports of people infected with Strongyloides fuelleborni and Strongyloides fuelleborni kellyi, in Africa and in Papua New Guinea [9]. Based on molecular data, S. fuelleborni kellyi should probably be considered an independent species rather than a subspecies of S. fuelleborni, [10]. Although S. stercoralis infection frequently remains asymptomatic, immuno-compromised patients can develop a systemic infection, which may lead to fatal forms of strongyloidiasis. The medical relevance of this parasite has probably been grossly underestimated due to difficulty of diagnosis [4,5,11]. Also, it should be noted that Strongyloides is not limited to tropical and underdeveloped areas, and the presence of S. stercoralis and fatal cases caused by it have also been reported from well-developed regions with temperate climates such as the European Union and North America [12–18]. S. stercoralis has a complex, rather unique life cycle (Fig 1) consisting of parasitic and free-living generations [19–21]. In brief: infective third stage larvae (L3i), which are all females, invade a new host by skin penetration and, after migrating through the blood and the lungs, are coughed up and swallowed and eventually establish in the small intestine of the host. The parasitic adult females reproduce by parthenogenesis. The progeny of the parasitic females have four developmental options: 1) Firstly, they may become female, and develop into infective third stage larvae (iL3) within the host and re-infect the same host individual (autoinfective cycle); 2) Secondly, they may become female, but this time leave the host as first-stage larvae, develop into iL3 and search for a new host (direct/homogonic development); 3) Thirdly, they may become female and leave the host, but this time develop into free-living, non-infective third stage larvae and subsequently into adult females (indirect/heterogonic development); 4) Or fourthly, they become male and leave the host and develop into free-living adult males (indirect/heterogonic cycle). The free-living adults mate and reproduce in the environment and all their progeny are females and develop to iL3s. No male iL3s have been reported in any Strongyloides species. For two species of Strongyloides (Strongyloides ratti and Strongyloides papillosus), it has been shown that the reproduction in the free-living generation is sexual, in spite of some earlier literature that had described it as pseudogamic (by sperm dependent parthenogenesis) [22–24]. For S. stercoralis prior to this report no genetic analysis of the mode of reproduction had been conducted and non-sexual (pseudogamic) reproduction as proposed based on cytological observations remained an option [25,26]. Whilst all species of Strongyloides may undergo homogonic or heterogonic development, the autoinfective cycle (option 1) appears to be specific for S. stercoralis and maybe a few other less well-investigated species [19]. This autoinfective cycle allows the parasite to persist in a particular host individual for many years, much longer than the life expectancy of an individual worm. Usually, healthy individuals tolerate such long lasting infections well and control them at very low levels [5]. These people have no clinical symptoms and the infection is unlikely to be detected. However, if such a chronically infected person becomes immunodeficient due to disease or immunosuppressive treatment (i.e. cancer chemotherapy or organ transplantation), this may lead to failure to control the infection and consequentially to self-enhancing progression of strongyloidiasis (hyperinfection syndrome and disseminated strongyloidiasis), which is lethal if not treated [5].

In rural communities in Cambodia, many people share their premises with domestic animals and the general hygenic, water and sanitation infrastructures are precarious [60,61]. Therefore, the conditions appear very favorable for human to animal and animal to human transmission of STH including S. stercoralis [3]. In order to find evidence for or against zoonotic transmission of S. stercoralis under such circumstances, we isolated large numbers of S. stercoralis from humans and dogs at the same time and in the same households and analyzed individual worms using molecular genetic markers. To our knowledge there had been no such study of S. stercoralis of comparable scale undertaken anywhere. It should be noted that our experimental strategy aimed to sample individuals with a large potential for transfer of Strongyloides spp. between the two hosts (e.g., only dogs found close to households with positive people were sampled). Therefore our study was not designed to yield accurate estimates of haplotype frequencies in the entire population. We also point out that we did not directly demonstrate transmission from dogs to humans and therefore cannot exclude that the transmission is mostly or exclusively from human to dog. Nevertheless, our results strongly suggest that there is a considerable risk for dog to human transmission. This would not be in agreement with conclusions by Takano and colleagues [62] who found that humans in households with Strongyloides-infected dogs were not more likely to be parasitized by S. stercoralis than those with parasite free dogs and concluded that natural transmission does not occur between humans and dogs. However, this study was conducted in Japan, in areas with presumably much better sanitary conditions than in the Cambodian villages where the present study was conducted. Consequently, only five Strongyloides positive dogs were found and none of their owners was infected. Likewise, a study conducted in Southern China, in a setting probably more comparable to our study area [47], did not identify the presence of animals as a statistically significant risk factor for human strongyloidiasis. However, this conclusion was based on only 21 infected individuals (11.7% of the tested), and no details about the exposure to dogs are given. In rural settings dogs are usually semi-domesticated and roam freely such that the risk of exposure to contamination by canine feces among people who do not own a dog themselves might be approximately equal to the risk among people who do. Therefore, the lack of statistical significance cannot be taken as evidence against zoonotic transmission. Interestingly, a later study in a similar setting [63] revealed that anthelmintic treatment of people alone was not sufficient to significantly reduce the prevalence of S. stercoralis.

Our results provide a compelling solution for the long-standing controversy about whether the Strongyloides sp. of dogs is identical to the S. stercoralis of humans or not. In fact, both scenarios appear to be true. Dogs, at least in our study area, host two different populations. These either represent separate species or well-separated sub-species of Strongyloides spp, and only one of them is shared with humans. It remains to be determined if the different types of Strongyloides we observed in humans and dogs also occur in other regions of the world. Among S. stercoralis in humans there is variability in the rDNA sequence. While we did not find further genomic evidence supporting multiple genetically separate populations in humans the absence of hybrids between the different SSU HVR I haplotypes is striking. It will be most interesting to ascertain whether different SSU HVR I types indeed interbreed and, even more importantly, if they might be associated with different clinical outcomes. Therefore, we suggest using molecular diagnostics for Strongyloides spp. wherever possible. In order to generate comparable data, we propose following the lead of Hasegawa and colleagues [35,38], and using the SSU HVRs I and IV and the mitochondrial cox1 locus as primary markers as was done in this study. With respect to strongyloidiasis control and prevention, this study suggests that dogs should be seriously considered as a source for human S. stercoralis infection at least in settings similar to our study area. Prevention of human contact with dog feces and of dog contact with human excrement as well as anthelmintic treatment of dogs are likely to reduce the exposure of humans to infective S. stercoralis larvae.