The Schistosoma indicum species group in Nepal: presence of a new lineage of schistosome and use of the Indoplanorbis exustus species complex of snail hosts

Abstract

From 2007–2014, 19,360 freshwater snails from the Terai and hilly regions of Nepal were screened for cercariae of mammalian schistosomes. Based on analysis of mitochondrial cytochrome oxidase I (cox1), 12S, 16S and 28S sequences (3,675 bp) of the cercariae recovered, we provide, to our knowledge, the first report of the Schistosoma indicum species group in Nepal. Five samples of Schistosoma nasale, nine of Schistosoma spindale and 17 of Schistosoma sp. were recovered, all from the snail Indoplanorbis exustus. The last-mentioned lineage failed to group in any of our analyses with S. nasale, S. spindale or S. indicum. It diverged in cox1 sequence from them by 16%, 13% and 13%, respectively, levels of difference comparable to well-studied species pairs of Schistosoma. Analysis of cox1, 16S and internal transcribed spacer 1 (ITS1) sequences (1,874 bp) for Nepalese specimens of I. exustus was also surprising in revealing the presence of four genetically distinct clades. They diverged from one another at levels comparable to those noted for species pairs in the sister genus Bulinus. There was no obvious pattern of use by Nepalese Schistosoma of the Indoplanorbis clades. We found high support for a close relationship between S. indicum and Schistosoma haematobium groups, but failed to retrieve support for a clean separation of the two, with a tendency for S. nasale to fall as the most basal representative. If this pattern holds, hypotheses for the origin of the Asian Indoplanorbis-transmitted S. indicum group from the Bulinus-transmitted S. haematobium group may require modification, including consideration of more contemporaneous origins of the two groups. The Indian subcontinent is under-studied with respect to schistosome diversity and our current knowledge of the S. indicum and I. exustus species groups is inadequate. Further study is warranted given the ability of indicum group species to cause veterinary problems and cercarial dermatitis, with a worrisome potential in the future to establish infections in humans.

Graphical Abstract

1. Introduction

Within the medically important digenetic trematode genus Schistosoma, four prominent species groups are customarily recognized. These are the japonicum species group with five species, the mansoni group with two species, the haematobium species group with nine species and the indicum species group with three species. In addition, other species groups can be considered including the schistosomes hosted by the hippopotamus (two or three species), and what can be called the turkestanicum species group with four recognized species, although verification of their relatedness is required. The species groups are defined by their tendency to consistently separate as distinct molecularly-defined lineages, the identities and relatedness of their snail hosts, general geographic distributions and more loosely by the morphology of the eggs produced by the females (Rollinson and Southgate, 1987; Lockyer et al., 2003; Morgan et al., 2003; Attwood et al., 2007).

The Schistosoma indicum species group as currently recognized is comprised of three species: S. indicum, Schistosoma nasale and Schistsosoma spindale (Attwood et al., 2007). Although Schistosoma incognitum was formerly included in this group, it has been shown that it is a more distant relative (Agatsuma et al., 2002; Attwood et al., 2002; Morgan et al., 2003; Webster et al., 2006: Lawton et al., 2011). Among the four customarily recognized and prominent species groups, our overall knowledge of the S. indicum group lags far behind the other three. The primary reason for this is that the members of this group, although they can and do cause dermatitis in people (Anantaraman, 1958; Narain et al., 1998; Agrawal et al., 2000), rarely if ever establish patent human infections (Agrawal et al., 2000; Agrawal and Rao, 2011), so consequently they have attracted less attention. They are however of considerable veterinary significance and they pose a number of interesting questions regarding the evolution and biogeography of schistosomes.

Overviews of the biology of the three indicum group species can be found in Kumar and deBurbure (1986), Rollinson and Southgate (1987) and Agrawal (2012). Briefly, all three species are hosted by the bulinine planorbid snail, Indoplanorbis exustus. This snail is widely distributed across southern Asia (see map range from Budha, P.B., Dutta, J., Daniel, B.A., 2012. Indoplanorbis exustus. The IUCN red list of threatened species. Version 2014.3. Available from www.iucnredlist.org.), and has since been introduced into exotic locales such as West Africa and the Caribbean islands by human activities. As noted by Liu et al. (2010), and as we discuss further below, I. exustus actually represents a complex of cryptic species. The extent of genetic diversification within what at least superficially looks like a fairly morphologically homogeneous gastropod species was surprising, and will surely improve our overall understanding of the evolution and diversification of the S. indicum group and of other dependent trematodes. Some literature also implicates lymnaeid snails in transmission of members of the indicum group, but when careful experimental infections have been performed, only I. exustus has been experimentally infected (Dutt and Srivastava, 1968; De Bont et al., 1991). In our view, it is likely that the Asian mammalian schistosomes recovered from lymnaeids and attributed to the S. indicum group are most probably representatives of the lymnaeid-transmitted Schistosoma turkestanicum species group or of S. incognitum.

The most widely distributed member of the S. indicum group is S. spindale which causes intestinal schistosomiasis with its most prominent hosts being water buffaloes, cattle, sheep, and goats. It can also infect equines and wild rodents. It is known from India, Bangladesh, Sri Lanka, Malaysia, Vietnam, Laos, Indonesia and Thailand. It produces characteristic elongated, spindle-shaped eggs with a terminal spine. The males of this species are atuberculate (Rollinson and Southgate, 1987; Gupta and Agrawal, 2002).

Schistosoma indicum also causes intestinal schistosomiasis in buffaloes, cattle, sheep, goats, camels and horses (Srivastava and Dutta, 1951). This species, as is best known, is confined in its distribution to the “Indian subcontinent” including India and Bangladesh. It is said to be present in every Indian state (Agrawal, 2012). The males are atuberculate and the eggs produced are oval and possess a terminal spine (Rollinson and Southgate, 1987).

Lastly, S. nasale is distinctive among mammalian schistosomes in occupying the veins of the nasal mucosa of ruminants, giving rise to nasal schistosomiasis, or “snoring disease,” as affected animals, especially cattle, breathe with a distinctive snoring sound. The large, sinuous, terminally-spined eggs are passed from the body with nasal discharges. Sheep do not reliably support patent infections of S. nasale (Agrawal and Rao, 2011). Adult males of this species have prominent tubercles, most of which lack spines (Southgate et al., 1990). This species is known from India, Bangladesh, Myanmar and Sri Lanka (Dutt, 1967; Rollinson and Southgate, 1987; Agatsuma et al., 2002; Lockyer et al., 2003; Attwood et al., 2007).

A number of molecular phylogenetic studies have consistently shown that the lymnaeid-transmitted S. incognitum is not part of the indicum group even though its distribution (India, Thailand, Indonesia) overlaps that of members of the indicum group (Agatsuma et al., 2002; Lockyer et al., 2003; Attwood et al., 2007). Furthermore, S. indicum and S. spindale are consistently retrieved as sister species with a high degree of support. This is interesting because the egg morphology of the two species is so divergent. The position of S. nasale relative to the other two species has been more problematic, and has been presented as either grouping with weak or equivocal support as basal to the other two species (Lockyer et al., 2003; Webster and Littlewood, 2012), or as being in a more distant position, with African species of the Bulinus-transmitted Schistosoma haematobium group intercalated, suggestive that the S. indicum group is paraphyletic (Attwood et al., 2007). Several lines of evidence, including mitochondrial gene order data (Agatsuma et al., 2002; Sato et al., 2008; Webster and Littlewood, 2012), consistently indicate that the S. indicum group is more closely related to the African schistosomes, particularly to the S. haematobium group, than to the Asian Schistosoma japonicum group.

The number of indicum group specimens available for molecular analysis has to date been limited. They consist of S. indicum from Bangladesh, S. spindale from Bangladesh, Sri Lanka and Thailand, and S. nasale from Bangaladesh and Sri Lanka. Specimens of S. incognitum from Bangladesh, Indonesia and Thailand have thus far been examined. Mitochondrial gene order has been determined for all four of the above species and found to be of the “derived” type (Lockyer et al., 2003; Littlewood et al., 2006; Sato et al., 2008; Webster and Littlewood, 2012). Remarkably, to date no specimens of the S. indicum group from India have been subjected to molecular phylogenetic analysis in a published study.

No specimens from the S. indicum group have previously been reported from Nepal, although the presence of mammalian schistosome cercariae from I. exustus has been reported (Devkota et al., 2011). The elephant schistosome Bivitellobilharzia nairi occurs in Nepal in both elephants and rhinoceros (Devkota et al., 2014a), and at least two species of avian schistosomes also occur there (Devkota et al., 2014b), but our knowledge of schistosomes present in Nepal is otherwise rudimentary. Laterally-spined schistosome eggs closely resembling the eggs of Schistosoma mansoni have been reported from human fecal samples from the Terai region of southern Nepal (Sherchand et al., 1999), but no additional reports of a schistosome with laterally-spined eggs from Nepal have since come to light.

The same freshwater snails discussed by Devkota et al. (2014b) were also screened for cercariae of mammalian schistosomes. These collections took place over a 7-year time interval, mostly in the Terai region of southern Nepal, in and around Chitwan National Park. Here we present the first known report from Nepal of the presence of cercariae representing the S. indicum group, all of which were recovered from I. exustus. Sequence data for the mitochondrial cytochrome oxidase I (cox1), 12S, 16S and nuclear 28S gene regions were acquired from most of the specimens and were subjected to molecular phylogenetic analyses. Additionally, for both uninfected I. exustus, and for some of the schistosome-positive I. exustus specimens, we also acquired cox1, 16S and internal transcribed spacer 1 (ITS1) (including parts of adjacent 18S and 5.8S genes) sequence data. This information has been used to examine patterns of diversification within I. exustus, and to relate those to both other studies of this snail and phylogenetic results obtained for the S. indicum group.

2. Materials and methods

2.1. Collection and morphological identification of freshwater snails

Snails were collected from 40 freshwater habitats in different areas of Nepal (Table 1, Supplementary Table S1) using kitchen sieves and triangular scoops mounted on long bamboo handles. The collected snails were kept moist and shaded prior to separation and cleaning. Identification of snails was made using conchological and morphological features (Subba Rao, 1989) and some of the snails were preserved in 96% ethanol for molecular identification.

Table 1.

List of localities in Nepal sampled for freshwater snails from 2007 to 2014. The number of snails in parentheses indicates the number of Indoplanorbis exustus positive for Schistosoma indicum group schistosomes. The identifications for the schistosomes provided are derived from the phylogenetic analyses that follow (see Fig. 3), and the designation “Schistosoma sp.” reflects the genetic distinctiveness of these worms from the few available specimens of S. indicum in GenBank. We were unable to obtain sequence data from three of our samples and these are referred to as “unidentified” in the table.

Locations	Number of snails examined	Co-ordinates	Number of Indoplanorbis collected	Species identity and W numbera
1. Amreni, Tanahu	65	N 27°59′15.9″, E 84°16′58.2 ″	65	-
2. Baghmara Community forest, Chitwan	219	N 27°35′22.0 ″, E 84°28′52.4 ″	156	-
3. Baruwa, Tamasariya-9, Nawalparasi	93	N 27°34′54.51 ″, E 84°01′17.41 ″	51 (1)	Schistosoma sp. (W379)
4. Begnas Lake, Kaski	43	N 28°09′58.00 ″, E 84°05′34.50 ″	14	-
5. Budhi Rapti River near Elephant Breeding Center, Chitwan	844	N 27°34′57.8 ″, E 84°27′56.6 ″	104 (2)	2 Schistosoma spindale (W525, W546)
6. Chisapani Village, Godar-2/3, Dhanusa	153	N 26°55′51.5 ″, E 86°08′45.8 ″	1	-
7. Chitwan National Park, Chitwan	95	N 27°33′15.8 ″, E 84°21′19.7 ″	59	-
8. Chitwan National Park, Chitwan	147	N 27°33′22.6 ″, E 84°29′5.40 ″	22 (1)	Schistosoma sp. (W532)
9. Chitwan National Park, Chitwan	51	N 27°32′43.1 ″, E 84°30′08.1 ″	29	-
10. Chitwan National Park, Chitwan	162	N 27°33′59.3 ″, E 84°30′14.9 ″	49	-
11. Chitwan National Park, Chitwan	422	N 27°33′37.5 ″, E 84°29′24.5 ″	231 (2)	2 Schistosoma sp. (W463, W528)
12. Chitwan National Park, Chitwan	359	N 27°33′37.0 ″, E 84°30′09.1 ″	244	-
13. Dhad Khola, Tulsi Chauda-2/3, Dhanusa	390	N 27°00′57.75 ″, E 85°55′27.00 ″	14	-
14. Dhalkebar near Basai bridge, Dhanusa	75	N 26°55′39.0 ″, E 85°57′58.2 ″	-	-
15. Dhumre river, Kumrose, Chitwan	1,753	N 27°34′34.11 ″, E 84°31′02.21 ″	658 (1)	Schistosoma nasale (W545)
16. Fish Pond in Panchakanya Community Forest	339	N 27°39′28.1 ″, E 84°29′18.7 ″	27	-
17. Ghansikuwa, Tanahu	183	N 28°00′33.4 ″, E 84°08′04.6 ″	15	-
18. Jagdishpur reservoir, Niglihawa VDC, Kapilvastu	802	N 27°36′59.67 ″ E 83°05′49.36 ″	24	-
19. Jamunapur, jutpani-5, Chitwan	417	N 27°39′42.48 ″, E 84°30′57.43 ″	218	-
20. Jankauli, Bachhauli-7, Chitwan	419	N 27°34′29.28 ″, E 84°30′48.64 ″	248 (1)	S. spindale (W378)
21. Khageri river, Near Panchakanya Community Forest	837	N 27°39′34.9 ″, E 84°29′00.2 ″	411	-
22. Kuchkuche Community forest, Kathar, Chitwan	183	N 27°34′26.85 ″, E 84°36′28.88 ″	48 (1)	Schistosoma sp. (W557)
23. Kuchkuche Community forest, near Rapti dam Kathar, Chitwan	340	N 27°34′02.28 ″, E 84°37′16.55 ″	154 (1)	S. spindale (W558)
24. Kudauli, Pithauli-7, Nawalparasi	155	N 27°39′12.21 ″, E 84°10′24.70 ″	101	-
25. Kumaraura, Dhanusa	135	N 26°46′36.76 ″, E 85°56′05.50 ″	49 (1)	S. nasale (W439)
26. Kumrose, Chitwan	337	N 27°34′03.42 ″, E 84°32′22.87 ″	28	-
27. Nawalpur, Hetauda, Makwanpur	325	N 27°25′55.06 ″, E 89°58′57.02 ″	66	-
28. Phewa Lake, Pokhara	59	N 28°12′31.27, E 83°57′19.42 ″	5	-
29. Pragatinagar-2, Nawalparasi	63	N 27°40′04.72 ″, E 84°11′03.59 ″	63	-
30. Ramaidaiya Bhawadi village, Dhanusa	26	N 26°49′26.31 ″, E 85°57′05.80 ″	-	-
31. Rapti river, Ghailari, Jagatpur-1	27	N 27°33′25.03 ″, E 84°20′02.48 ″	21	--
32. Rapti river, Sauraha, Chitwan	296	N 27°34′52.24 ″, E 84°28′56.12 ″	106	-
33. Rato river, Gauribash, west of Tulsi Chauda village, Dhanusa	181	N 27°01′43.56 ″, E 85°55′21.30 ″	1	-
34. Rice fields in Chisapani Village, Godar-2/3, Dhanusa	257	N 26°56′26.09, E 86°08′51.19 ″	21	-
35. Shishuwar bagar, Bachhauli-3, Chitwan	3,023	N 27°35′31.76 ″, E 84°30′00.31′ ′	1,321 (11)	6 Schistosoma sp. (W380, W381, W383, W531, W534, W541), 2 S. spindale (W535, W556), 3 unidentified (W384, W385, W533)
36. Small canal in Sauraha, Chitwan	373	N 27°34′59.55 ″, E 84°29′39.44′ ′	170	-
37. Tikauli marshy land, Ratnanagar- 7, Chitwan	95	N 27°37′12.66 ″, E 84°28′13.87 ″	5	-
38. Tikauli, Ratnanagar-7, Chitwan	4, 674	N 27°37′45.52 ″, E 84°29′21.57 ″	1,890 (12)	6 Schistosoma sp. (W517, W518, W519, W550, W798, W799); 3 S. spindale (W464, W538, W804); 3 S. nasale (W542, W549, W551)
39. Tulsi Chauda village, Dhanusa	357	N 27°00′50.37 ″, E 85°55′31.80 ″	6	-
40. Twenty thousand lake, Chitwan	586	N 27°36′54.0 ″, E 84°26′19.9 ″	564	-

^aW number indicates schistosome cercariae sample number; the cercariae of different sample numbers come from different individual Indoplanorbis exustus snails

2.2. Screening of infected snails and morphological identification of cercariae

Collected snails were isolated in individual wells of a 24-well tissue culture plate or in Petri dishes containing clean water. The isolated snails were exposed to window light or artificial illumination to stimulate cercarial shedding. Approximately 1 h later, snails were individually screened for shed cercariae using a dissecting microscope. If cercariae were observed, a few were transferred to a microscope slide and observed with the aid of a compound microscope. The cercariae were identified morphologically by means of cercarial keys (Frandsen and Christensen, 1984). Cercariae were ethanol-preserved, measured and photographed using a digital camera fitted to the compound microscope. Snails that did not shed cercariae in the first hour were re-examined for shedding of cercariae at least twice within the following 24 h.

Cercariae were preserved in RNAlater (Ambion The RNA Company, Life Technologies, Grand Island, New York, USA) or in 96% ethanol. All preserved samples were hand-carried with the permission of the Nepal Health Research Council (ref. number 44, 25 July 2011) to the Department of Biology at the University of New Mexico, Albuquerque, New Mexico, USA for molecular analysis and further morphological analysis.

2.3. Molecular and phylogenetic analyses

DNA from schistosome cercariae was extracted from alcohol- or RNAlater©-preserved cercarial samples either by using the Qiagen DNeasy Blood and Tissue Kit or the Qiagen QIAamp DNA Micro Kit (Valencia, California, USA). Cercariae were digested for 2 – 3 h or overnight. The nuclear ribosomal 28S and mitochondrial cox1, 16S and 12S genes were amplified by PCR by using a Takara Ex Taq kit (Takara Biomedicals, Otsu, Japan) and previously published primers (U178; 5′-GCA CCC GCT GAA YTT AAG-3′ and L1642; 5′-CCA GCG CCA TCC ATT TTC A -3′ for 28S sequences, cox1F6; 5′-TTT GTY TCT TTR GAT CAT AAG CG-3′ and cox1 3; 5′-TAA TGC ATM GGA AAA AAA CA- 3′ for cox1 sequences and P12SF; 5′ – TTT GTC CAC AGT TAT AAC TGA AAG G -3′ and P12SR; 5′ – GAT TCT TCA AGC ACT ACC ATG TTA CGA C -3′) (Attwood et al., 2002; Lockyer et al., 2003; Morgan et al., 2003). New primers were designed to amplify the 16S gene (R16SF 5′-TGT TTT TTT CCK ATG CAT TA - 3′ and R16SR 5′ - GGC TTA CAC CGG TCT TAA CT - 3′). PCR products were purified either with an Omega E.Z.N.A Cycle-Pure Kit (Omega Bio-Tek, Norcross, GA 30071, USA) or USB ExoSAP-IT PCR Product Cleanup (Affymetrix, Inc., Cleveland Ohio 44128, USA) according to the manufacturers’ guidelines. Sequencing reactions were performed with Applied Biosystems BigDye direct sequencing kit, version 3.1 (Applied Biosystems, Foster City, California, USA).

DNA from snails was extracted from alcohol-preserved samples either by using the Qiagen DNeasy Blood and Tissue Kit or the Omega E.Z.N.A Mollusc DNA Kit following the manufacturers’ guidelines. The partial sequences of a nuclear ITS1 and two mitochondrial genes; cox1 and 16S genes were amplified by PCR by using a Takara Ex Taq kit and previously published primers (ITS1-S; 5′ CCA TGA ACG AGG AAT TCC CAG 3′ and 5.8S-AS 5′ TTA GCA AAC CGA CCC TCA GAC 3′ for ITS1; LCO1490; 5′ GGT CAA CAA ATC ATA AAG ATA TTG G 3′ and HCO2198; 5′ TAA ACT TCA GGG TGA CCA AAA AAT CA 3′ for cox1 sequences and 16Sar; 5′ CGC CTG TTT ATC AAA AAC AT 3′ and 16Sbr; 5′ CCG GTC TGA ACT CAG ATC ACG T 3′ for 16S sequences) (Palumbi et al., 1991; Folmer et al., 1994; DeJong et al., 2001). PCR products were purified either with an Omega E.Z.N.A Cycle-Pure Kit or USB ExoSAP-IT PCR Product Cleanup according to the manufacturers’ guidelines. Sequencing reactions were performed with Applied Biosystems BigDye direct sequencing kit, version 3.1.

The schistosome cercariae 28S, 16S, 12S and cox1 gene fragments and the snail host I. exustus ITS1, 16S and cox1 gene fragments were used in phylogenetic analyses using Bayesian inferences (BI) with the use of MrBayes, v 3.1.2 (Huelsenbeck and Ronquist, 2001). NEXUS files used with MrBayes to create alignments have been made public as Supplementary Data S1 – S11. Model selection was estimated using ModelTest (Posada and Crandall, 1998). The BI analyses were as follows: the 28S and ITS1 dataset, Nst = 6 rates = invgamma ngammacat = 4; the cox1 dataset, for codons one and two Nst = 2 and for codon three Nst = 6 rates = gamma, ngammacat = 4; for 12S and 16S combined Nst = 6 rates = invgamma ngammacat = 4 (parameters were unlinked among partitions). Four chains were run simultaneously for 5 × 10⁵ generations, with four incrementally heated chains sampled at intervals of 100 generations. The first 5000 trees with pre-asymptotic likelihood scores were discarded as burn-in using Tracer v1.6 (available from http://beast.bio.ed.ac.uk/Tracer), and the retained trees were used to generate 50% majority-rule consensus trees and posterior probabilities. Outgroups for the schistosome cox1, 12S, 16S and 28S datasets included members of the S. mansoni species group (sister to the S. indicum + S. haematobium groups), and in some cases also included members of the S. turkestanicum and S. japonicum species groups (Supplementary Table S2). The remaining taxa used in the tree were obtained from the published literature (Attwood et al., 2002, 2007; Lockyer at al., 2003). Outgroups for the phylogenetic analysis of Indoplanorbis were selected as the sister genus Bulinus (GenBank accession numbers AM921851, AM921835, AM921842, GU176747, KJ157497, EF489313, AY029546, EU076730, EU076727, EU076726), Planorbarius (GenBank accession numbers AY282590, AY577473) and Biomphalaria (GenBank accession numbers DQ084823, AY126606, AY198103, JQ886409) snails as outgroups. The character variation in the ingroup for the alignments used to create each phylogenetic tree is given in Supplementary Table S3. All DNA sequence data were deposited in GenBank (Tables 2, 3).

Table 2.

List of the Schistsoma indicum group schistosome specimens and GenBank accession numbers for which DNA sequence data was obtained for phylogenetic analysis.

Taxon or Sample number	GenBank Accession number
	28S sequences	12S sequences	cox1a sequences	16S sequences
W378	KR423856	KR607250	KR607222	KR423839
W379	KR423845	KR607233	KR607213	KR423832
W380	-	KR607234	-	-
W381	-	KR607235	-	-
W383	KR423846	KR607236	-	-
W439	KR423863	-	-	-
W463	KR423847	KR607237	KR607214	-
W464	KR423857	KR607251	KR607223	-
W517	-	KR607238	-	-
W518	-	KR607239	KR607215	-
W519	-	KR607240	-	-
W525	KR423858	KR607252	KR607224	-
W528	KR423848	KR607241	KR607216	KR423833
W531	KR423849	KR607242	KR607217	KR423834
W532	KR423850	KR607243	KR607218	KR423835
W534	-	KR607244	-	-
W535	-	KR607253	-	-
W538	-	KR607254	-	-
W541	KR423851	KR607245	-	-
W542	KR423864	KR607259	KR607229	-
W545	KR423865	KR607260	KR607230	KR423843
W546	KR423859	KR607255	KR607225	-
W549	KR423866	KR607261	KR607231	-
W550	KR423852	KR607246	-	-
W551	KR423867	KR607262	KR607232	KR423844
W556	KR423860	KR607256	KR607226	KR423840
W557	KR423853	KR607247	KR607219	KR423836
W558	KR423861	KR607257	KR607227	KR423841
W798	KR423854	KR607248	KR607220	KR423837
W799	KR423855	KR607249	KR607221	KR423838
W804	KR423862	KR607258	KR607228	KR423842
S. indicum	AY157258 (Bangladesh)	EF534276 (Bangladesh)	AY157204 (Bangladesh)	EF534284 (Bangladesh)
Schistosoma spindale	Z46505 (Sri Lanka); AY157257 (Sri Lanka); AF465925 (Thailand); AF465926 (Thailand)	EF534283 (Thailand); AF465920 (Thailand); AF465919 (Thailand); AF534282 (Sri Lanka); EF534281 (Bangladesh); DQ157223 (Lab strain originally from Sri Lanka)	AY157203 (Sri Lanka); DQ157223 (Lab strain originally from Sri Lanka)	EF534290 (Thailand); EF534289 (Sri Lanka); EF534288 (Bangladesh); DQ157223 (Lab strain originally from Sri Lanka)
Schistosoma nasale	AY157259 (Sri Lanka)	EF534280 (Bangladesh)	AY157205 (Sri Lanka)	-

^acox1, mitochondrial cytochrome oxidase I

Table 3.

List of the sources of Indoplanorbis exustus DNA sequence data used in this study. The SW number indicates the snail from which a particular schistosome (designated by a W followed by the same number) was obtained.

Taxon	Co-ordinates	Collection date	GenBank accession number
			cox1a	16S	ITS1b
Snail infected with Schistosoma nasale (SW439)	N 26°46′36.76 ″, E 85°56′05.50 ″	9 th January 2007	KR811332	KR607278	KR811325
Snail infected with Schistosoma spindale (SW464)	N 27°37′45.52 ″, E 84°29′21.57 ″	4th July 2010	KR811347	KR607268	KR811315
Snail infected with S. spindale (SW525)	N 27°34′57.8 ″, E 84°27′56.6 ″	2nd July 2010	KR811348	KR607263	KR811316
Snail infected with Schistosoma sp. (SW528)	N 27°33′37.5 ″, E 84°29′24.5 ″	9th July 2010	KR811338	KR607284	-
Snail infected with S. spindale (SW538)	N 27°37′45.52 ″, E 84°29′21.57 ″	4th July 2010	KR811349	KR607272	KR811321
Snail infected with S. nasale (SW545)	N 27°34′34.11 ″, E 84°31′02.21 ″	12th May 2011	KR811350	KR607273	KR811319
Snail infected with S. nasale (SW551)	N 27°37′45.52 ″, E 84°29′21.57 ″	22nd July 2010	KR811351	KR607274	KR811318
Snail infected with Schistosoma sp. (SW557)	N 27°34′26.85 ″, E 84°36′28.88 ″	3rd July 2012	KR811352	KR607276	KR811320
Snail infected with Schistosoma sp. (SW798)	N 27°37′45.52 ″, E 84°29′21.57 ″	1st August 2014	KR811339	KR607285	-
Snail infected with S. spindale (SW804)	N 27°37′45.52 ″, E 84°29′21.57 ″	29th July 2014	KR811354	KR607271	-
Snail infected with S. spindale (SW558)	N 27°34′02.28 ″, E 84°37′16.55 ″	3rd July 2012	KR811353	KR607269	KR811317
Snail infected with strigeids and xiphidiocercariae Khageri river, Near Panchakanya Community Forest - location 21	N 27°39′34.9 ″, E 84°29′00.2 ″	9th June 2011	KR811335	KR607281	-
Snail infected with strigeids and xiphidiocercariae Tikauli, Ratnanagar-7, Chitwan – location 38	N 27°33′59.3 ″, E 84°30′14.9 ″	6th July 2010	KR811336	KR607282	KR811328
Snail infected with sangunicolids and xiphidiocercariae Chitwan National Park, Chitwan - location 9	N 27°32′43.1 ″, E 84°30′08.1 ″	13 May 2011	KR811360	KR607270	KR811322
Snail infected with sangunicolids and xiphidiocercariae Chitwan National Park, Chitwan - location 7	N 27°33′15.8 ″, E 84°21′19.7 ″	13 May 2011	KR811359	KR607267	-
Snail infected with sangunicolids Chitwan National Park, Chitwan - location 8	N 27°33′22.6 ″, E 84°29′5.40 ″	8th July 2010	KR811337	KR607283	KR811327
Two snails infected with echistosome Baghmara Community forest, Chitwan - location 2	N 27°35′22.0 ″, E 84°28′52.4 ″	22 July 2012	KR811341, KR811342	KR607288, KR607290	-
Snail infected with hooked-bodied strigeids Rapti river, Sauraha, Chitwan - location 32	N 27°34′52.24 ″, E 84°28′56.12 ″	11 May 2011	KR811355	KR607277	KR811324
Two uninfected snails Kumaraura, Dhanusa - location 25	N 26°46′36.76 ″, E 85°56′05.50 ″	9th January 2007	KR811333, KR811334	KR607279, KR607280	KR811326
Three uninfected snails Budhi Rapti River near Elephant Breeding Center, Chitwan - location 5	N 27°34′57.8 ″, E 84°27′56.6 ″	11th May 2011	KR811356-KR811358	KR607264-KR607266	KR811323
Two uninfected snails Tulsi Chauda village, Dhanusa - location 39	N 27°00′50.37 ″, E 85°55′31.80 ″	7th January 2007	KR811344, KR811345	KR607286, KR607292	KR811331
Two uninfected snail Rice fields in Chisapani Village, Godar-2/3, Dhanusa - location 34	N 26°56′26.09, E 86°08′51.19 ″	8th January 2007	KR811346	KR607287, KR607293	KR811329
Snail infected with sangunicolid Dhumre river, Kumrose, Chitwan - location 15	N 27°34′34.11 ″, E 84°31′02.21 ″	12 May 2011	KR811361	KR607275	-
Uninfected snail Shishuwar bagar, Bachhauli-3, Chitwan - location 35	N 27°35′31.76 ″, E 84°30′00.31′ ′	2 July 2010	KR811340	KR607291	KR811330
Uninfected snail Chitwan National Park, Chitwan - location 12	N 27°33′37.0 ″, E 84°30′09.1 ″	23rd July 2010	KR811343	KR607289	-
I. exustus (Janakpur, Nepal)c	-	-	GU451739	GU451732	-
I. exustus (Assam, India)c	-	-	GU451744	GU451726	-
I. exustus (Mymensingh, Bangladesh)c	-	-	GU451745	GU451727	-
I. exustus (Kekirawa, Sri Lanka)c	-	-	GU451742	GU451735	-
I. exustus (Bintulu, Borneo, Malaysia 1)c	-	-	GU451746	GU451728	-
I. exustus (Kampang Pelegong, Malaysia 2)c	-	-	GU451738	GU451731	-
I. exustus (Bogor, Indonesia)c	-	-	GU451747	GU451729	-
I. exustus (Xang, Laos)c	-	-	GU451750	GU451751	-
I. exustus (Wadi Bani Khaled, Oman 1)c	-	-	GU451740	GU451733	-
I. exustus (Wadi Qab, Oman 2)c	-	-	GU451741	GU451734	-
I. exustus (Bulan, Philippines)c	-	-	GU451748	GU451730	-
I. exustus (Khon Kaen, Thailand 1)c	-	-	GU451743	GU451736	-
I. exustus (Phitsanulok, Thailand 2)c	-	-	HM104223	HM104222	-
I. exustus (Maenam Loei, Thailand 3)d	-	-	AY282587	-	-

^acox1, mitochondrial cytochrome oxidase I;

^bITS1, internal transcribed spacer 1;

^centries in the table are from Liu et al. (2010);

^dentry in the table is from Albrecht et al. (2004)

3. Results

3.1. Snail survey results

From 2007–2014, in the 40 locations listed in Table 1 and Supplementary Table S1, we collected and screened 19,360 freshwater snails for trematodes. The nine families represented 14 different species and their numbers were: Bellamya bengalensis (429), Brotia costula (15), Gabbia orcula (1,374), Gyraulus spp. (4,454), Indoplanorbis exustus (7,286), Lymnaea acuminata (1,380), Melanoides pyramis (194), Melanoides tuberculata (254), Physa sp. (16), Pila globosa (74), Radix luteola (2,588), Segmentina spp. (529), Succinea sp. (14), and Thiara spp. (753). Only specimens of I. exustus shed schistosome cercariae subsequently shown to be consistent with affiliations to the S. indicum group. We found 34 (0.47%) I. exustus snails from 11 different localities (Fig. 1, Table 1, Supplementary Table S1) to be infected. As indicated in the phylogenetic analyses that follow (Fig. 3), one location (#35) yielded two schistosome lineages (S. spindale and Schistosoma sp.) collected at the same time point from two different snails. Another location (#38) yielded schistosomes from all three lineages (S. spindale, S. nasale and Schistosoma sp.), but at different times within the same year or in different years (see Supplementary Table S1 for details).

Fig. 1.

Map of the study area showing sites positive for Indoplanorbis exustus-transmitted mammalian schistosomes, with numbers corresponding to locality data in Table 1 and Supplementary Table S1. (A) Map of Nepal and (B) expanded map of the Chitwan district within Nepal where many of the sampling sites were located.

Fig. 3.

Bayesian phylogenetic trees for Schistosoma. (A) Tree based on mitochondrial cytochrome oxidase I (cox1) sequences (1125 bp). This tree includes the only Schistosoma indicum sequence of this length available in GenBank. (B) Tree based on cox1 sequences (372 bp). This tree includes two S. indicum sequences available in GenBank. (C) Tree based on 28S sequences (1515 bp). (D) Tree based on combined analysis of cox1, 16S and 12S sequences (1740 bp). Samples in bold are those collected in this study. Node support is indicated by Bayesian posterior probabilities.

3.2. Morphological features of the schistosome cercariae recovered

Measurements of representatives of cercariae from each of the three lineages from Nepalese I. exustus were compared with those available from other studies of the S. indicum group, and for one species (S. incognitum) formerly included in the group (Supplementary Table S4). Although differences in preservation techniques make rigorous comparisons among studies difficult, we found all three kinds of cercariae that were recovered to be somewhat smaller than the cercariae of S. incognitum. With respect to tail stem length, which in our experience is useful for facilitating discrimination among species (Morgan et al., 2003), the relative order (shortest to longest) from other studies of the S. indicum group was: S. indicum, S. spindale and S. nasale. From shortest to longest tail stem length, the Nepalese specimens were ranked Schistosoma sp. (Fig. 2), S. spindale and S. nasale.

Fig. 2.

Cercaria of a Schistosoma sp. obtained from an Indoplanorbis exustus snail.

3.3. Phylogenetic analyses of the schistosomes recovered

With respect to our phylogenetic studies of the schistosome cercariae, we attempted to obtain cox1, 12S, 16S and 28S sequences from all specimens. At least one of the above sequence regions (Table 2) for all but three of our specimens was obtained. We also included as much pertinent corresponding published sequence that could be found from closely-related schistosomes from elsewhere in Asia or Africa.

First, regarding the Bayesian analysis of cox1 schistosome sequences (Figs. 3AB), several features are noteworthy, many of them recapitulated in the analyses of other genes. The two trees differed in that Fig. 3B incorporated additional sequences from GenBank, but these sequences were shorter (372 bp) compared with sequences used in Fig. 3A (1,125 bp). In Fig. 3A, for which better overall resolution was obtained, sequences representing seven members of the S. haematobium group consistently grouped together with strong support, and this group was sister (0.91 posterior probability) to a group containing S. indicum from Bangladesh, S. spindale from Sri Lanka and Nepal, and Schistosoma sp. from Nepal. Lying sister to these two lineages with strong support in the analysis of the longer cox1 sequences, was a tightly grouped clade consisting solely of specimens of S. nasale from both Sri Lanka and Nepal.

Interestingly, relative to the amount of diversity among the specimens of S. indicum, Schistosoma sp. and S. spindale, the amount of sequence diversity amongst the five representatives of S. nasale was minimal. The appearance of S. nasale as a lineage basal to both the S. haematobium group and to the remaining members of the S. indicum group was recovered in several of our analyses, with variable degrees of support. None of our analyses could significantly delineate the S. haematobium group from the entirety of the S. indicum group as customarily defined.

Two specimens of S. spindale from Sri Lanka grouped with high support with seven Nepalese isolates and although further study is warranted, particularly from intervening geographic regions, all are likely to be S. spindale. However, there is an appreciable degree of genetic distance (approximately 8%, Table 4, see Supplementary Table S3 as well for a summary of variation found for particular genes) between the specimens from the two countries. Also included in the analysis are nine closely-related Nepalese isolates that consistently formed a distinct, well-supported group that diverged in cox1 sequence from S. nasale or S. spindale by 16% and 13%, respectively. We initially assumed this third lineage would correspond by sequence similarity to S. indicum. We were surprised to note (as in Fig. 3A), however, that the few available sequences of S. indicum from GenBank consistently failed to group with the nine Nepalese isolates in this third lineage. Furthermore, the genetic distance separating S. indicum Bangladesh and our third lineage was 13%. This suggested to us that this third Nepalese lineage was distinct from all three recognized members of the S. indicum group, and it is referred to here as Schistosoma sp.

Table 4.

p -distances for the mitochondrial cytochrome oxidase I (cox1), 12S, 16S and 28S sequences among representative cercariae samples which we collected, and compared with other mammalian schistosomes.

Taxa	cox1	12S	16S	28S
W798 (Schistosoma sp.) – Schistosoma indicum (Bangladesh)	13%	7%	15%	0%
W804 (Schistosoma spindale) – S. spindale (Sri Lanka)	8%	2.5%	6%	0%
W551 (Schistosoma nasale) – S. nasale (cox1and 28S-Sri Lanka, 12S- Bangladesh)	0%	0%	-	0%
W798 (Schistosoma sp.) - W804 (S. spindale)	13%	6.7%	13%	0%
W551 (S. nasale) - W798 (Schistosoma sp.)	16%	8.6%	17%	2%
W551 (S. nasale) - Our sample W804 (S. spindale)	16%	1%	17%	2%
W528 (Schistosoma sp.) – W531 (Schistosoma sp.)	0%	0%	0%	0%
W558 (S. spindale) – W804 (S. spindale)	0%	0%	1%	0%
W545 (S. nasale) - W551 (S. nasale)	0%	0%	0%	0%
S. indicum – S. spindale	14% (S. indicum Bangladesh, S. spindale SL)	7% (Both from Bangladesh)	17% (Both from Bangladesh)	0% (S. indicum Bangladesh, S. spindale SL)
S. spindale – S. nasale	16% (Both from SL)	9% (Both from Bangladesh)	-	2% (Both from SL)
S. indicum – S. nasale	16% (S. indicum Bangladesh, S. nasale SL)	9% (Both from Bangladesh)	-	2% (S. indicum Bangladesh, S. nasale SL)
Schistosoma mansoni- Schistosoma rodhaini	11.5%	-	-	0%
Schistosoma mattheei – Schistosoma haematobium	13%	4%	-	1%
Schistosoma intercalatum – S. haematobium	11%	4%	-	0%

Note: W number indicates the schistosome cercariae sample number; the cercariae of different sample numbers come from different individual Indoplanorbis exustus snails

The tree in Fig. 3B was made to allow inclusion of a second available short cox1 sequence for S. indicum from Bangladesh, and although there was some tendency for Nepalese Schistosoma sp. to group with the two specimens of S. indicum from Bangladesh in this tree, support was not strong. Many nodes in this tree were not strongly supported and no support was found for uniting presumptive S. spindale from Nepal with S. spindale from elsewhere.

The nodal support provided by the analyses of 28S, 12S and 16S sequences (Fig. 3C, Supplementary Fig. S1 – S3) was generally not as strong as noted for Fig. 3A, but this may be expected as these genes are less variable than cox1. However, many patterns similar to those noted for the cox1 analyses were again observed using these other gene regions. The 28S analysis shown in Fig. 3C again resolved S. nasale as separate from either the remainder of the S. indicum group or the S. haematobium group, but was unable to resolve relative branch orders among these three groups. A Bayesian analysis based on combined cox1, 16S and 12S sequences (Fig. 3D) revealed the following patterns in the data: i) weakly supported basal position of S. nasale; ii) strong support (0.94 posterior probability) for a clade containing two major lineages, the S. haematobium group and the remainder of the S. indicum group consisting of the Bangladesh samples of S. indicum, S. spindale, and Schistosoma sp. from Nepal; iii) substantial variation among lineages ascribed to S. spindale from Sri Lanka and Nepal; and iv) the Schistosoma sp. lineage from Nepal was divergent from the one set of sequences available for S. indicum from Bangladesh.

Of the 20 specimens for which we obtained both a 28S nuclear sequence and at least one mitochondrial sequence (cox1, 16S or 12S), none revealed a nuclear sequence characteristic of one species and a mitochondrial sequence characteristic of another.

3.4 Phylogenetic analyses of the Indoplanorbis exustus snails recovered

Regarding the phylogenetic analyses for Indoplanorbis exustus, 30 cox1, 31 16S and 17 ITS1 (the latter also including parts of 18S and 5.8S genes) sequences were obtained from specimens of this snail, mostly from the Terai region (Table 3). For two of these samples, snail sequences were actually amplified from extracts of schistosome cercariae (W798 and W804) originally derived from I. exustus but for which the snails were no longer available. In addition, we compared our sequences with those from 14 additional specimens acquired from GenBank (Table 3, Fig. 4). Included in the analyses are sequences for other planorbid snails including species of Bulinus, the genus believed to be the most closely related to Indoplanorbis.

Fig. 4.

Map showing localities from which specimens of Indoplanorbis exustus were collected, and for which 16S and mitochondrial cytochrome oxidase I (cox1) sequences are available in GenBank from Albrecht et al. (2004) and Liu et al. (2010).

Results of the cox1 analysis (Fig. 5A) provide two especially noteworthy features. The first, and most surprising, is the presence of four differentiated clades of I. exustus (designated I – IV), two of which (Clades I and IV), including the basal group in our cox1 analysis (Clade I), are comprised exclusively of snails from Nepal. Representatives of each of the four groups are found in Nepal, and three of the four groups (Clades I, III and IV) included specimens infected with members of the S. indicum group. Strong nodal support was obtained for a clade uniting all identified I. exustus specimens. Notably lacking in this analysis are snail specimens from India or Pakistan, or Iran in the western part of the natural range of I. exustus.

Fig. 5.

Bayesian phylogenetic trees for Indoplanorbis. (A) Tree based on mitochondrial cytochrome oxidase I (cox1) sequences (620 bp). (B) Tree based on Indoplanorbis cox1 and 16S combined sequences (1036 bp). In both trees samples in bold are those collected in this study. Node support is indicated by Bayesian posterior probabilities. The “L” followed by a number designates the location number indicated in Table 1.

In the 16S analyses (Supplementary Fig. S4), all specimens of I. exustus were again shown to group together with high support. Once again four major clades of I. exustus were discerned (each containing specimens from Nepal) but with more variable support, and in this case there was not a strongly supported basal clade. The pattern of relationships among the three remaining clades was often poorly resolved, although the members within each clade grouped together with strong support (0.92–1 posterior probability). Regarding the analysis of 1,036 bp of combined cox1 and 16S sequence (Fig. 5B), strong support for the monophyly of the Indoplanorbis lineages was again retrieved. Strong support at most nodes was retrieved for the four major clades, although the topology differed from that noted in Fig. 5A. This analysis also showed strong support for the sister relationship between Bulinus and Indoplanorbis.

The p-distance values for both cox1 and 16S genes among the four different lineages of I. exustus (Table 5, Supplementary Table S3) are in the same range (cox1 0–14% and 16S 0–8%) as reported for species pairs in Bulinus. Due to a lack of ITS1 data for Indoplanorbis in GenBank we were unable to compare our data with I. exustus group specimens from other localities. In our ITS1 analysis (Supplementary Fig. S5), we were unable to resolve the four major clades shown in our other analyses.

Table 5.

Uncorrected p-distances for the mitochondrial cytochrome oxidase I (cox1) and 16S sequences among representative snail samples we collected, and with other snails. The SW number indicates the snail from which a particular schistosome (designated by a W followed by the same number) was obtained.

Taxa	cox1	16S
Within Indoplanorbis	0–14%	0–8%
Within Bulinus	2–15%	1–13%
Indoplanorbis infected with sanguinicolid Chitwan National Park – Indoplanorbis exustus Nepal (Liu et al., 2010)	2%	4%
Indoplanorbis uninfected Chitwan National Park (location 12) – I. exustus Nepal (Liu et al., 2010)	14%	7%
Indoplanorbis Chisapani, Dhanusa – Indoplanorbis infected with sanguinicolid Chitwan National Park	13%	7%
Indoplanorbis Chisapani, Dhanusa – Indoplanorbis infected with Schistosoma spindale (SW538)	12%	8%
Indoplanorbis infected with S. spindale (SW538) – Indoplanorbis infected with Schistosoma nasale (SW439)	9%	5%
Indoplanorbis infected with Schistosoma sp. (SW528) – Indoplanorbis infected with S. nasale (SW439)	13%	7%
Indoplanorbis infected with Schistosoma sp. (SW528) - Indoplanorbis infected with S. spindale (SW538)	12%	8%
Indoplanorbis infected with S. spindale (SW464) - Indoplanorbis infected with S. spindale (SW525)	0%	0%
Bulinus tropicus – Bulinus nasutus	15%	9%
Bulinus globosus – Bulinus. forskali	12%	13%
Biomphalaria glabrata – Biomphalaria peregrine	10%	7%

We observed multiple clades (I, II and IV) of Indoplanorbis in one location (# 38) at different collection times. Based on our limited number of snail sequences, we did not find any coexisting clades.

4. Discussion

This study was first stimulated by a report of S. mansoni-like eggs from human stool samples from the Terai region of Nepal (Sherchand et al, 1999). Although we could not provide corroborating evidence for the presence of either S. mansoni or Biomphalaria, the survey work that followed revealed four lineages of Schistosoma in Nepal: the three lineages of the S. indicum group discussed here, all of which were recovered from I. exustus, and a fourth lineage of lymnaeid-transmitted Schistosoma we will discuss in a separate paper.

We were unable to examine adult schistosomes as part of this study because buffaloes and especially cattle are maintained for milk production and are not commonly slaughtered in the Terai region. Government sanctioned abattoirs are not present in the area and we found owners of private abattoirs were unwilling to grant permission to examine viscera of slaughtered animals. Also, mice or hamsters are not readily available for experimental infections and sanctioned animal care committees and procedures are either not in place or lack the ability to articulate with animal care committees in the USA. The lack of adult worms and associated eggs, the usual benchmarks whereby species of schistosomes are described and delineated, means that what is said about the specific identity of the S. indicum group cercariae that were recovered, has to be qualified.

One consistent feature of our analyses was that S. nasale was retrieved as separate from, and often basal to, members of the combined S. haematobium-S. indicum groups. Furthermore, relative to the other lineages we found, the amount of genetic diversity present in this species was minimal, both for specimens retrieved from Nepal and when specimens from Sri Lanka or Bangladesh were included. With respect to the remaining two S. indicum group lineages we found in Nepal, one is almost certainly S. spindale, although the Nepalese specimens diverge genetically from the few S. spindale specimens available for comparison from elsewhere in Asia. Thus, whereas S. nasale from Nepal, Sri Lanka and Bangladesh exhibit 0% p-distance values for their cox1 sequences, specimens of S. spindale from Nepal and Sri Lanka differ by 8% in p-distance values for the same target gene (Table 4). Nonetheless, strong posterior probability support is obtained for grouping Nepalese specimens of presumptive S. spindale with S. spindale from other locations.

We refer to the third S. indicum group lineage in Nepal as “Schistosoma sp.” Although we originally expected this third lineage to be S. indicum, it never grouped closely in any of our trees with available sequences for S. indicum from GenBank. Nepalese Schistosoma sp. differs substantially from known S. indicum with respect to p-distance values: 13% for the cox1 gene, a value commensurate with typical species differences within both the S. indicum and S. haematobium groups (Table 4). A note of caution is in order here as very few specimens of S. indicum for which sequence data exist (potentially only two worms represented) are available. Furthermore, they are from the same locality in Bangladesh. Only small fragments of 16S (60 bp) and 28S (346 bp) of S. indicum from India were available in GenBank. The Bangladesh and Indian specimens may not be representative of the genetic variation within S. indicum, or it is possible S. indicum is broadly distributed and encompasses a particularly wide range of genetic variation. It is certainly a possibility that Nepalese Schistosoma sp. is actually a distinct species, possibly one confined to the northern part of the collective range of the S indicum group. It is hoped that this possibility will stimulate additional study including verification with specimens of adult worms and eggs, and experimental infection studies.

Although Schistosoma sp. from Nepal does not cluster with previously identified S. indicum, it persistently groups with S. spindale as a close but distinct relative in our trees (p-distance value 13% for cox1, Table 4). Even though worms of both lineages likely inhabit the mesenteric veins of domestic ruminants in Nepal, we found no evidence in our samples suggestive of hybridization of Schistosoma sp. with S. spindale. There was also no indication of hybridization of either lineage with S. nasale. Does hybridization occur in the S. indicum group as now seems to be commonly the case with the S. haematobium group (Rollinson et al., 1990; Webster et al., 2005, 2013; Huyse et al., 2009)? Although we saw no evidence for similar admixture in the S. indicum group, this was not the main objective of our work, so further study is needed to settle this point.

Regarding I. exustus, our results are in agreement with Liu et al. (2010) in showing a surprising amount of structure within what has traditionally been considered a single widespread snail species. Liu et al. (2010) identified three major clades, one of which includes specimens from localities as diverse as Oman, Malaysia, Indonesia, Thailand, the Philippines and Nepal, and corresponds to clade II identified in our trees. The other two clades from the study of Liu et al. (2010) were more closely related to one another, were from Bangladesh and northern India, and corresponded to clade III in our cox1 analysis. The sample from Bangladesh was moved to clade I in our 16S and combined cox1 and 16S analyses. Our clade IV only includes specimens from Nepal. The specimen from Bangladesh noted by Liu et al. (2010) was relatively divergent from those we collected from Nepal, suggesting it might be appropriate to recognize even a fifth lineage. Our results indicate multiple clades can inhabit the same specific habitat (#38) from which clades I, II and IV of Indoplanorbis were all found.

The p-distance values for both cox1 and 16S genes among the four different lineages of I. exustus we identified (Table 5) are not dissimilar from values recorded for species pairs within the probable sister genus Bulinus. Samples of this snail from many other parts of its known range have yet to be included in molecular phylogenetic studies, so the overall diversity inherent in the I. exustus species complex is bound to increase. In particular, specimens from north of the Himalayas, most of the western portion of the range including much of India, Pakistan, Afghanistan and Iran, island populations from the Indian Ocean, and specimens from Africa are bound to provide more surprises.

Caution is required in interpreting both the relative abundance of the various I. exustus lineages in Nepal and the role each plays in schistosome transmission. Snails from lineage I, for which representatives are mostly known from the current study, were commonly collected and two echinostome infections and two Schistosoma sp. infections were recovered from them. Lineage II, which includes specimens from several countries (including one specimen from Nepal) collected by others, was also found by us but not commonly, and of the three specimens we sequenced, two had double trematode infections (xiphidiocercariae and strigeids) and one had a sanguinicolid infection. Lineage III, which also includes specimens collected by others from Assam, or Assam and Bangladesh depending on the tree, was less common in Nepal but included one snail positive for S. nasale. Finally, Lineage IV, known from only our collections to date, was probably the most common lineage in our collections, and in addition to infections with strigeids, sanguinicolids and xiphidiocercariae, samples of all three of the S. indicum group lineages were recovered: five S. spindale, two S. nasale, and one Schistosoma sp. In Nepal, three of the four major I. exustus lineages supported at least one member of the S. indicum group, Lineage IV supported all three schistosome lineages, and Lineage II was not found to support any schistosomes although it is likely snails of this lineage support the S. indicum group elsewhere in Asia where they seem to be common (Liu et al., 2010). Lineage II may also transmit schistosomes in Nepal where, based on admittedly incomplete sampling, it appears to be uncommon.

As noted in several other studies (Lockyer et al, 2003; Morgan et al., 2003; Webster et al., 2006; Lawton et al, 2011; Webster and Littlewood, 2012), we found clear support for a close relationship between the S. haematobium and S. indicum species groups. As was also noted in the most recent study of the S. indicum group by Attwood et al. (2007), we found the two groups did not separate cleanly and there was a persistent tendency for the S. haematobium group to nest within the S. indicum group, with S. nasale basal, albeit often without strong support, e.g. the S. indicum group as traditionally considered appears to be paraphyletic (see also Attwood et al., 2007). The results may be dependent on the particular sequences selected for study as other investigations (Morgan et al., 2003; Webster et al., 2006; Webster and Littlewood, 2012) have indicated separation of the S. indicum and S. haematobium groups, but bootstrap support for the basal position of S. nasale within the S. indicum group is always reported as relatively weak, again raising questions as to the delineation of these two groups. Provision of more sequence data may resolve what looks like an obvious anomaly in schistosome systematics, given that members of the recognized S. haematobium groups all parasitize Bulinus snails and are predominantly African and southwest Asian in distribution, whereas members of the recognized S. indicum group are transmitted by the I. exustus species group and are Asian in distribution. The close relationship between the two species groups of schistosomes is supported by the close relationship of their snail hosts. Bulinus and Indoplanorbis are united by the presence of a distinctive synapomorphy, the ultrapenis, and by molecular systematic studies that place them in a bulinine clade distinct from other planorbid snails (Fig. 5B, Morgan et al., 2002; Albrecht et al., 2007), but note the difficulties in documenting a sister group relationship mentioned by Jørgensen et al. (2011).

Scenarios for the origins of the S. indicum group (see discussions in Barker and Blair 1996; Attwood et al., 2002, 2007; Lockyer et al., 2003; Lawton et al., 2011) involve an origin in Africa in the Plio-Pleistocene from an artiodactylid-inhabiting schistosome with a bulinid snail host that colonized Asia via the Sinai and Levant. Land connections between Africa and Asia opened during the mid-Miocene with prominent faunal exchanges occurring thereafter. Among the animal groups in transit were bovids, which were well established by that time. Also, a host shift into Indoplanorbis is a likely possibility given the relatedness of the two snail genera.

This scenario may need to change somewhat, however, especially if further study shows that S. nasale retains its basal position relative to the remainder of the combined S. indicum-S. haematobium group. If so, this is noteworthy for at least two reasons. One is that the original snail host for this combined lineage may have been Indoplanorbis, not Bulinus. Second, given the present-day distribution of Indoplanorbis in southern Asia (African representatives of I. exustus appear to be relatively recent colonists), it raises a question as to where the origins of this larger group of schistosomes might actually lie. Today, aside from recent human-influenced range changes, the ranges of Indoplanorbis and Bulinus are contiguous in regions of Iraq and Iran, and may overlap in some areas there. In the past, especially considering the occurrence of several pluvial periods in northern Africa and southwest Asia at various periods ranging from 3.2 million years ago (mya) to 12,500 ya (Lawler, 2014), opportunities for intermingling of both definitive and intermediate hosts may have been much more extensive in the Levant, the Arabian peninsula, Mesopotamia and present-day Iran than they are today.

That I. exustus is actually a species complex with much more genetic structure than previously considered is particularly germane with respect to the evolution of the S. indicum group. Particular lineages within the I. exustus complex, assuming they were already in existence, may have played a critical role in the transition of Schistosoma into Indoplanorbis. Additional study is needed to better characterize the full diversity within both the S. indicum group and the I. exustus species complex, as is a thorough analysis of how the two species groups interact. For example, are particular species in the S. indicum group confined to certain lineages of the Indoplanorbis group, or do some lineages of Indoplanorbis host all, whereas others host none, of the indicum group species? Our results are not suggestive of congruent patterns of cospeciation as all three schistosome species from Nepal were found in I. exustus Lineage IV, and two schistosome species were found in two different I. exustus lineages. Our results suggest that Lineage IV is important in transmission in Nepal, but more extensive sampling of both worms and snails across a broader geographic area may well change this story.

In considering the timing involved in the emergence of the S. indicum group, it is noteworthy and fortuitous that members of the S. indicum group do not establish patent infections in humans, unlike at least two members of the S. haematobium group in Africa. This tends to suggest that the time available for members of the S. indicum group to adapt to and infect humans has been limited relative to members of the S. haematobium group, which, by virtue of their origin in Africa, would have had opportunity for continual contact with hominins. Also, the dependency of indicum group worms on Indoplanorbis may similarly have prevented them from gaining any longterm access to hominins as they evolved and diversified, because Indoplanorbis is an Asian genus. Unlike Bulinus-transmitted schistosomes, indicum group worms did not evolve in Africa, the acknowledged location of origin and diversification for hominins.

We note our survey focused on a relatively small part of Nepal. The presence there of three genetically distinct lineages of the S. indicum group (including one that is apparently new), and at least four well-differentiated clades of I. exustus, indicate that Nepal and other parts of the northern Indian subcontinent are areas in much need of further study. The triculine snails in hilly regions of Nepal (Nesemann et al., 2007) warrant further study with respect to recovery of new schistosomes, given the role of triculines elsewhere in transmission of S. japonicum group parasites. It will be of particular interest to learn whether members of the S. indicum group extend to areas north of the Himalayas as I. exustus occurs in Tibet and China. The Himalayas have been shown to have a strong effect in isolating Gyraulus snails north of the mountains from those to the south (Oheimb et al., 2013) and might also have similar effects on Indoplanorbis and associated parasites. Once the I. exustus species complex has been better fleshed out, it will be interesting to determine whether any of its lineages are susceptible to infection with any members of the S. haematobium group, or if S. indicum group parasites ever infect any (particularly Asian representatives of) bulinid snails. To our knowledge, such cross-infections have rarely been attempted (Agrawal and Rao, 2011). Wright (1971) exposed I. exustus from Socotra to S. haematobium, Schistosoma bovis and Schistosoma mattheei, without success.

As members of the S. indicum group are frequently implicated in causing cercarial dermatitis in humans in southern Asia (Anantaraman, 1958; Narain et al., 1998; Agrawal et al., 2000), better characterization of the worms and snails involved in outbreaks may eventually help to alleviate this problem and highlight risk areas for possible emergence of human-adapted worms. With respect to where we began our studies of schistosomiasis in Nepal, it seems prudent to remain vigilant to the possibility of some of these worms adapting to, and establishing patent infections in, human hosts (Agrawal and Rao, 2011). Certainly there is no shortage of human contact with waters containing schistosome-infected snails.

Supplementary Material

10. Supplementary Data S1.

NEXUS file used to create Fig. 3A

11. Supplementary Data S2.

NEXUS file used to create Fig. 3B

12. Supplementary Data S3.

NEXUS file used to create Fig. 3C

13. Supplementary Data S4.

NEXUS file used to create Fig. 3D

14. Supplementary Data S5.

NEXUS file used to create Fig. 5A

15. Supplementary Data S6.

NEXUS file used to create Fig. 5B

16. Supplementary Data S7.

NEXUS file used to create Supplementary Fig. S1

17. Supplementary Data S8.

NEXUS file used to create Supplementary Fig. S2

2. Supplementary Fig. S1.

Bayesian phylogenetic tree based on Schistosoma 28S sequences (477 bp). Samples in bold are those collected in this study. Node support is indicated by Bayesian posterior probabilities.

3. Supplementary Fig. S2.

Bayesian phylogenetic tree based on Schistosoma 12S sequences (386 bp). Samples in bold are those collected in this study. Node support is indicated by Bayesian posterior probabilities.

4. Supplementary Fig. S3.

Bayesian phylogenetic tree based on Schistosoma 16S sequences (649 bp). Samples in bold are those collected in this study. Node support is indicated by Bayesian posterior probabilities.

5. Supplementary Fig. S4.

Bayesian phylogenetic tree based on Indoplanorbis 16S sequences (420 bp). Samples in bold are those collected in this study. Node support is indicated by Bayesian posterior probabilities. The “L” followed by a number designates the location number indicated in Table 1.

6. Supplementary Fig. S5.

Bayesian phylogenetic tree based on an ingroup analysis of Indoplanorbis ITS1 sequences (834 bp; includes parts of adjacent 18S and 5.8S genes). All the samples in this analysis were collected as part of this study, there were no samples available in GenBank for comparison. Node support is indicated by Bayesian posterior probabilities. The “L” followed by a number designates the location number indicated in Table 1.

7. Supplementary Data S9.

NEXUS file used to create Supplementary Fig. S3

8. Supplementary Data S10.

NEXUS file used to create Supplementary Fig. S4

9. Supplementary Data S11.

NEXUS file used to create Supplementary Fig. S5

Highlights.

• The Schistosoma indicum group is reported for the first known time from Nepal

• Included are Schistosoma spindale, Schistosoma nasale and a previously uncharacterized lineage

• Four distinct lineages of Indoplanorbis were represented in our Nepalese collections

• The S. indicum group is hosted by a complex of cryptic species of Indoplanorbis

• Origins of the S. indicum and Schistosoma haematobium groups require additional consideration