Molecular characterization of nodule worm in a community of Bornean primates

Abstract Strongyles are commonly reported parasites in studies of primate parasite biodiversity. Among them, nodule worm species are often overlooked as a serious concern despite having been observed to cause serious disease in nonhuman primates and humans. In this study, we investigated whether strongyles found in Bornean primates are the nodule worm Oesophagostomum spp., and to what extent these parasites are shared among members of the community. To test this, we propose two hypotheses that use the parasite genetic structure to infer transmission processes within the community. In the first scenario, the absence of parasite genetic substructuring would reflect high levels of parasite transmission among primate hosts, as primates’ home ranges overlap in the study area. In the second scenario, the presence of parasite substructuring would suggest cryptic diversity within the parasite genus and the existence of phylogenetic barriers to cross‐species transmission. By using molecular markers, we identify strongyles infecting this primate community as O. aculeatum, the only species of nodule worm currently known to infect Asian nonhuman primates. Furthermore, the little to no genetic substructuring supports a scenario with no phylogenetic barriers to transmission and where host movements across the landscape would enable gene flow between host populations. This work shows that the parasite's high adaptability could act as a buffer against local parasite extinctions. Surveys targeting human populations living in close proximity to nonhuman primates could help clarify whether this species of nodule worm presents the zoonotic potential found in the other two species infecting African nonhuman primates.


| INTRODUC TI ON
Parasites infecting closely related host species, or host species with overlapping home ranges, are often assumed to be generalist parasites (Davies & Pedersen, 2008). Given their ability to infect multiple host species, they are also considered to be important threats to biodiversity conservation and public health (Cleaveland, Laurenson, & Taylor, 2001;Jones et al., 2008). This perspective, however, has been challenged in recent years, as many alleged generalist parasites have turned out to be specialists in an ecological sense despite using a phylogenetically widespread set of resources, that is, specialization not for host species but for specific resources shared among them (Agosta, Janz, & Brooks, 2010;Nyman, 2009). This suggests that host specificity is more of a dynamic trait moving along a continuum than a black and white delineation (Janz, Nyblom, & Nylin, 2001;Nosil, 2002).
Gastrointestinal nematodes are an example of such a group, being geographically widespread and having the potential to infect multiple host species. Those nematode species with direct life cycles pass infective stages into the external environment, where they can remain for extended periods of time and transmit through shared habitat use, without the need for direct contact between hosts. Strongylid nematodes, for example, are among the most commonly reported gastrointestinal parasites found in wild primates.
Despite being common, strongylids such as Oesophagostomum spp.
As is the case with other soil-transmitted helminths of public health concern, the first challenge for diagnosis is species identification, as strongyle eggs have almost no diagnostically useful morphological features. Similarities in morphology and morphometric measurements of certain eggs make species-specific diagnosis particularly difficult. Primate species can be infected with several nematodes of the suborder Strongylida, such as hookworms (Ancylostoma, Necator), Trichostrongylus, Ternidens, and Oesophagostomum, whose eggs cannot be distinguished with certainty by microscopy alone (Polderman & Blotkamp, 1995). For example, there is considerable overlap in egg morphology and size for Necator and Oesophagostomum, but the former is significantly more pathogenic than the latter. Although morphology of infective-stage L 3 larvae is suitable for genus-level identification in strongylid nematodes (Blotkamp et al., 1993 shared across the primate host community. We tested two competing but not necessarily mutually exclusive hypotheses with our data set. First, overlapping ranges among primate hosts are suspected to be high in the area, which can facilitate panmixia (i.e., random mating or homogenization among parasite populations) in environmentally transmitted parasites such as nodule worm. We therefore predicted that there would be little or no genetic substructuring in the parasite. However, we also hypothesized that, if substructuring is evident, phylogenetic barriers may be the most likely explanation. In the latter case, we predicted that substructuring in the parasite population would occur between rather than within primate lineages.

| Study site and sample collection
The Lower Kinabatangan Wildlife Sanctuary is located in the east- From the collected fecal samples, we set up coprocultures (N = 24, silvered langurs (7), proboscis monkeys (7), long-tailed macaques (4), and orangutans (6)) using a modified Harada-Mori filter-paper technique following Hasegawa (2009) to develop L 3 larvae and extract DNA directly from them instead. Developed larvae migrating into the water were fixed in 70% ethanol and later morphologically identified to genus level using standard keys (Anderson, Chabaud, & Willmott, 2009).

| DNA extraction and phylogenetic analyses
As many samples were not collected after observing defecation, and most primate species have overlapping ranges in the area, we conducted host species identification for all samples used in coprocultures by extracting whole DNA from each collected fecal sample and amplifying a small fragment of the cytochrome b (cytb) gene, following the protocol described in Frias et al. (2018). Then, 64 strongylid larvae were individually selected at random for DNA extraction (

| Ethics statement
Authorization to conduct research in Sabah, collect samples, and export them to Japan was provided by the Sabah Biodiversity Centre (SaBC) and the Sabah Wildlife Department. Our field protocols also adhered to the guidelines set by these agencies.

| RE SULTS
The overall prevalence of strongyles in the studied community was low to moderate (38%), and although most fecal samples used for coprocultures turned out to be strongyle-negative by microscopy, we were still able to successfully develop L 3 larvae in 10 of them (Table 2) Samples from orangutans and long-tailed macaques are spread throughout the Bornean group, while samples from colobines are F I G U R E 1 Phylogenetic relationships among Oesophagostomum species infecting primates inferred from cox1 gene sequences (only ML tree shown). Branch support is represented by ML/NJ bootstrap values, respectively only present in one of them. It should be noted here that the colobine data comprised only three individual larvae from three fecal samples, so these results should be interpreted with caution until further data are available.

| D ISCUSS I ON
Studies on sympatric host species offer a good opportunity to understand the role of species heterogeneity in a transmission process involving parasites acquired through shared habitat use, as certain species will be more susceptible than others to infection or more likely to contribute to parasite transmission (Morgan, Milner-Gulland, Torgerson, & Medley, 2004). In contrast to the advances in understanding transmission dynamics of Oesophagostomum species infecting human and nonhuman primates in Africa, we still know little about O. aculeatum, the only species presumed to infect nonhuman primates and humans in Asia (Stewart & Gasbarre, 1989).
Our study thus investigated its patterns of diversity in a multi-host, multi-parasite system in Malaysian Borneo, in an area where several primate species are known to live sympatrically and to be infected by strongylid nematodes across their range. Through coprocultures and genetic characterization of L 3 larvae, we showed that O. aculeatum is widely distributed in the study area and that it infects all studied members of the community with little clear genetic differentiation among them.
As parasitological surveys of wild primates tend to provide coarse information at best, we generally lack information regarding not only which parasites are infecting primates but also how they may be impacting their host populations. In humans, the clinical manifestation of oesophagostomiasis is generally associated with the presence of nodules (encysted L 4 larvae) in the intestinal mucosa (Polderman & Blotkamp, 1995). Oesophagostomum spp. infection in communities of African primates at Mahale, Kibale, and Taï has revealed that not only is the parasite detectable in feces Kooriyama et al., 2013;Kouassi et al., 2015), but it may have consequences for individual health. Clinical signs associated with the onset of disease have been observed in both chimpanzees and baboons (Papio anubis), and chimpanzees have been observed swallowing whole leaves to purge intestinal worms (Huffman & Caton, 2001;Huffman et al., 1996;Terio et al., 2011Terio et al., , 2018. Given that Oesophagostomum spp. and other strongylid nematodes are known to impact health and fitness in a broad range of nonprimate hosts as well (Roepstorff, Bjørn, Nansen, Barnes, & Christensen, 1996;Waghorn, Bouchet, Bekelaar, & Leathwick, 2019), efforts to better understand its distribution and diversity are warranted.
In general, although Oesophagostomum spp. seem to have the potential to infect all or at least many members of primate communities, the parasite's ability to infect each host species and its fitness in each one of them remains to be explored. For example, Oesophagostomum spp. infection in colobines suggests different host-parasite dynamics than that observed in chimpanzees. A recent study looking at Oesophagostomum-associated pathology in sympatric primates at Gombe reported that, in contrast to chimpanzees and baboons that showed high numbers of nodules, none were found in red colobus (Procolobus rufomitratus) also inhabiting the area (Terio et al., 2018).
The effect of diet on the establishment and persistence of the parasite could be a venue for future exploration in this regard. It has been suggested that diets with high levels of insoluble digestible fiber, such as wheat bran, whole grains, and cereals, favor O. dentatum survival, growth, and reproduction in farm pigs, leading to a significant increase in the parasite's infection intensity (Petkevičius, Nansen, Knudsen, & Skjøth, 1999). Colobines' low-digestible diet, consisting mostly of leaves, seeds, and unripe fruit (Chivers, 1994) may have a regulatory function in the parasite's ecology.
Oesophagostomum aculeatum was not the only strongylid nematode infecting Bornean primates observed in this study. We were also able to genetically identify a single isolate of Ternidens deminutus (Nematoda, Strongylida) from a Bornean orangutan.
T. deminutus is an understudied nematode infecting primates and has been regarded as a zoonotic parasite able to affect human health (Hemsrichart, 2005 ), but they are often not assigned to parasite species (but see Ngui, Lim, Traub, Mahmud, & Mistam, 2012, Sahimin et al., 2017.
Surveys targeting human populations living in close proximity to nonhuman primates would go a long way toward exploring that possibility.

ACK N OWLED G M ENTS
The authors are grateful to the Sabah Wildlife Department and Sabah Biodiversity Centre (SaBC) for allowing us to conduct this research. We also thank Hirohisa Hirai for his support of this project.
We owe a large debt to the staff and students/volunteers at the

CO N FLI C T O F I NTE R E S T
None Declared.

AUTH O R CO NTR I B UTI O N S
LF, MO, and AJJM designed the research. LF and AJJM collected the data. LF conducted the laboratory work and data analysis. LF wrote the manuscript with input from MO and AJJM. DJS, MSL, SKSSN, and BG provided logistical support and comments to the manuscript.

DATA ACCE SS I B I LIT Y
All DNA sequences obtained in this study were deposited in DDBJ/ GenBank, under Accession Numbers LC428761-LC428824.