INTRODUCTION
Wild murid populations worldwide are commonly examined for their helminth fauna. The most commonly reported gastrointestinal digeneans found in the UK wood mouse Apodemus sylvaticus tend to be Corrigia vitta and Brachylaemus recurvum (Elton et al. Reference Elton, Ford and Baker1931; Behnke et al. Reference Behnke, Lewis, Mohd Zain and Gilbert1999; Abu-Madi et al. Reference Abu-Madi, Behnke, Lewis and Gilbert2000). In 2007, however, Rogan et al. (Reference Rogan, Craig, Hide, Heath, Pickles and Storey2007) reported a rare UK occurrence of the intestinal digenean Plagiorchis muris from the small intestine of A. sylvaticus at the Malham Tarn Nature Reserve and which had been previously recorded over a 13-year period. Reports of digeneans from UK vole populations appear to be limited; however, here we report the occurrence of another species of the genus Notocotylus Diesing, 1839, from the caecum of 2 vole populations at Malham Tarn. Previously, Notocotylus noyeri has been recorded from the water-vole Arvicola amphibius (syn. A terrestris) and the short-tailed field vole Microtus hirtus (syn. M. agrestis) in Cambridgeshire, UK (Baylis, Reference Baylis1928b, Reference Baylis1939). A further 6 Notocotylus spp. recorded in the UK have all involved waterfowl as definitive hosts. The current report identifies a new Notocotylus species that can be morphologically differentiated from all previously reported notocotylids.
The genus Notocotylus Diesing, 1839 is cosmopolitan, parasitizing waterfowl, and small mammals with an affinity for water. The genus has a complicated taxonomic history and misidentification has been commonplace leading to synonymization and much confusion. Morphological differences between species are small, and there is ongoing dispute over reliable diagnostic morphological criteria (Nath and Pande, Reference Nath and Pande1963). As a result many newly described Notocotylus species have been suppressed as synonyms and the genus has been continuously revised (Lal, Reference Lal1935a; Harwood, Reference Harwood1939; Dubois, Reference Dubois1951; Stunkard, Reference Stunkard1966; Simon-Vicente et al. Reference Simon-Vicente, Mas-Coma, Lopez-Roman, Tenora and Gallego1985a). Even the number of valid species within the genus Notocotylus can be problematic, a fact that has been reflected in the variable number of species included in differential keys (Dubois, Reference Dubois1951; Skrjabin, Reference Skrjabin1953; Yamaguti, Reference Yamaguti1958). Following an extensive study into the morphology of Notocotylus species infecting rodents in Europe, Simon-Vicente et al. (Reference Simon-Vicente, Mas-Coma, Lopez-Roman, Tenora and Gallego1985a) identified the existence of 4 stable intraspecific morphological features that are of differential systematic value. Using these criteria, in addition to those previously established by Lal (Reference Lal1935a) and Dubois (Reference Dubois1951), the digeneans from bank voles and field voles at Malham Tarn have been designated as a new species in the genus Notocotylus.
MATERIALS AND METHODS
The study was carried out at Malham Tarn Nature Reserve located in North Yorkshire, Northwest England at an altitude of 375 m above sea level. Malham Tarn is a ‘site of special scientific interest’ (SSSI) and is the only upland marl lake in Britain (Rogan et al. Reference Rogan, Craig, Hide, Heath, Pickles and Storey2007). This area has previously been investigated for a range of host parasite systems (Allan et al. Reference Allan, Craig, Sherington, Rogan, Storey, Heath and Iball1999; Hughes et al. Reference Hughes, Williams, Morley, Cook, Terry, Murphy, Smith and Hide2006, Reference Hughes, Thomasson, Craig, Georgin, Pickles and Hide2008; Rogan et al. Reference Rogan, Craig, Hide, Heath, Pickles and Storey2007; Behnke et al. Reference Behnke, Eira, Rogan, Gilbert, Torres, Miquel and Lewis2009; Thomasson et al. Reference Thomasson, Wright, Hughes, Dodd, Cox, Boyce, Gerwash, Abushahma, Lun, Murphy, Rogan and Hide2011).
Rodents were trapped under permit from the National Trust, using Longworth small mammal traps for 4 days each season between January 2010 and October 2011. The trapping point of each infected rodent was recorded using a Garmin GPS 60 set to the co-ordinate framework WGS84. Four sites were examined throughout the sampling period including 3 woodland sites (Tarn Wood 54°06′03.3″N, 002°09′44.9″W, Spiggot Hill 54°05′72.9″N, 002°10′43.1″W; Ha Mire Plantation 54°05′64.5″N, 002°09′53.7″W) and 1 wetland site, (Tarn Fen 54°06′00.0″N, 002°10′43.4″W). In total, 126 wood mice (Apodemus sylvaticus) and 63 voles (54 Myodes glareolus and 9 Microtus agrestis) were trapped from all 4 sampling sites. All rodents trapped were examined for a range of helminth parasites using a consistent detailed post-mortem examination. In the collection of M. glareolus and M. agrestis, Capillaria spp., Hymenolepis diminuta and Heligmosomoides glareoli were also found in addition to Notocotylus malhamensis. Helminth species richness varied from 0 to 3 in individual animals.
Rodents were euthanized and examined according to Rogan et al. (Reference Rogan, Craig, Hide, Heath, Pickles and Storey2007) and morphologically identified to species level using the criteria of Sargent and Morris (Reference Sargent and Morris2003). Morphological identification of vole species was verified using molecular analysis. A 1 cm2 section of thigh muscle was aseptically removed from each of the vole species and was placed into an Eppendorf tube containing 400 μl of lysis buffer (100 mM NaCl, 25 mM ethylene diamine tetraacetic acid, 0·5% sodium dodecyl sulphate, 20 mM Tris, pH 8·0). DNA extraction was performed according to Thomasson et al. (Reference Thomasson, Wright, Hughes, Dodd, Cox, Boyce, Gerwash, Abushahma, Lun, Murphy, Rogan and Hide2011).
The cytochrome oxidase subunit I gene was amplified from vole DNA by PCR using the forward primer RonM (5′GGMGCMCCMGATATRGCATTCCC3′) and reverse primer NancyM (5′CCTGGGAGRATAAGAATATAWACTTC3′) according to Pfunder et al. (Reference Pfunder, Holzgang and Frey2004). Each 25 μl reaction contained 2·5 μl of 10X DreamTaq buffer including 2 mM MgCl2 (Fermentas, Life Sciences), 0·025 μmol deoxynucleotide triphosphate (dNTPs; 100 mM, Bioline), 1 μM forward primer, 1 μM reverse primer, 2·5 U DreamTaq DNA polymerase and 1 μl gDNA template (50 μg/μl). All PCR reactions were performed using a Robocycler 96 PCR machine (Stratagene, CA, USA) and visualized on a 1% (w/v) Tris-acetate-EDTA (TAE) agarose gel stained with gel red using a G: Box gel imaging system (Syngene, UK). The cycling profile consisted of an initial denaturation of 1 cycle at 95 °C for 15 min, 45 cycles of denaturation at 95 °C for 40 sec, annealing at 50 °C for 40 sec and elongation at 72 °C for 1 min, and 1 final cycle at 72 °C for 7 min. The target bands were excised from the gel using a UV transilluminator and purified using a PCR purification kit (Geneflow) according to the manufacturer's instructions. Samples were commercially sequenced in both directions (Source Bioscience, Nottingham, UK).
Notocotylus was not detected from any of the A. sylvaticus despite careful observation. All specimens of N. malhamensis from the caecum of the bank voles (Myodes glareolus) and the field voles (Microtus agrestis) were relaxed in distilled water for morphological examination. Some digeneans were fixed in 5% formal saline and flattened under light cover-slip pressure for morphological analysis. All measurements were taken using a Leica DM500 microscope and eyepiece graticule. Eggs were recovered from the feces of infected rodents and examined under phosphate-buffered saline (PBS) to measure the length of the egg filaments. Twenty specimens were stained in borax carmine and mounted in Canada balsam according to Gurr (Reference Gurr1963) with alteration of the staining time to 3 h. The ventral glands and cirrus composition were examined on unstained unflattened specimens. Drawings were made from photographs taken with a Leica ICC50 digital camera attached to a Leica DM500 microscope. The difference in morpho-anatomic measurements between the two specimens of N. malhamensis recovered from M. glareolus and M. agrestis were statistically analysed using a one-way parametric ANOVA with equal replicates.
The remainder of the flukes were fixed in 70% ethanol suitable for molecular analysis. DNA was extracted from individual worms using a phenol: chloroform method modified from Thomasson et al. (Reference Thomasson, Wright, Hughes, Dodd, Cox, Boyce, Gerwash, Abushahma, Lun, Murphy, Rogan and Hide2011) by halving the amount of reagent at each stage of the protocol. The Internal transcribed spacer (ITS), including the ITS1, 5.8S, ITS2 and flanking regions of the 3′ end of the 18S and 5′ end of the 28S were amplified using the forward universal primer BR (5′GTAGGTGAACCTGCGGA3′) and reverse digenean specific primer dig11 (5′GTGATATGCTTAAGTTCAGC3′) according to Tkach et al. (Reference Tkach, Pawlowski and Sharpilo2000a). The partial 28S rDNA gene region was amplified using the forward digenean specific primer dig12 (5′AAGCATATCACTAAGCGG3′) and the reverse universal primer Lo (5′GCTATCCTGAGRGAAACTTCG3′) according to Tkach et al. (Reference Tkach, Pawlowski and Mariaux2000b).
Each 50 μl PCR reaction contained 5 μl of 10X DreamTaq buffer including 2 mM MgCl2 (Fermentas, Life Sciences), 0·05 μmol dNTPS (100 mM, Bioline), 2·5 μM forward primer, 2·5 μM reverse primer, 5 U DreamTaq DNA polymerase and 2 μl of gDNA template (50 μg/μl). The amplification profile consisted of 1 cycle at 94 °C for 10 min, followed by 35 cycles of 1 min at 94 °C, 1 min at 54 °C and 1 min at 72 °C and 1 final cycle at 72 °C for 10 min. Excision of the target bands, PCR purification and DNA sequencing was performed as previously described. This same procedure was conducted on 3 independent occasions. In total 9 sequences for each gene were aligned.
RESULTS
Confirmation of vole species
Identification of the 2 vole species was confirmed by molecular barcoding of Cytochrome Oxidase 1 as well as by overall morphology. In both cases, animals identified by morphology as M. glareolus and M. agrestis were confirmed by 99% sequence homology with NCBI reference sequences (M. glareolus AY332679 and M. agrestis AY332684). Sequences from Malham rodents have been deposited into GenBank under Accession numbers: (M. glareolus JQ794805 and M. agrestis JQ794806).
Notocotylus malhamensis n. sp. ( Figs 1–4)
Type-locality: Tarn Fen wetlands (54°06′00.0″N, 002°10′43.4″W, WGS84 framework), Malham Tarn Nature Reserve, North Yorkshire, UK.
Type-hosts:Myodes glareolus, Microtus agrestis.
Site in host: Caecum.
Prevalence and intensity: 568 specimens were collected from M. glareolus (overall prevalence 66·7%, 6/9) and 13 from M. agrestis (overall prevalence 50%, 3/6) from the Tarn Fen area. Infected M. glareolus had infection intensities ranging from 1 to 294 flukes (median = 48) and M. agrestis had intensities ranging from 1 to 6 flukes (median = 6).
Type specimens: 45 specimens have been deposited at the Department of Zoology, Natural History Museum, Cromwell Road, London, UK (holotype NHMUK 2012.3.14.1, and paratypes NHMUK 2012.3.14.2-30 (10 borax carmine-stained specimens and 35 specimens stored in 70% molecular grade ethanol). A further 10 borax carmine-stained specimens and 120 specimens stored in 70% molecular grade ethanol are held at The School of Environment and Life Science, The University of Salford, Salford Crescent, Manchester, UK.
Morphological description
Figure 1 illustrates the ventral and dorsal morphology of N. malhamensis recovered from M. glareolus. Measurements were based on 10 specimens. In live specimens the fluke is dorso-ventrally flattened with lateral margins that fold dorsally to provide a curvature to the body and create a ventral concavity. Tegument is unspined. In flattened specimens the anterior end of the body attenuates and is bluntly pointed in comparison to the posterior end, which is generally rotund. The body length ranges from 2·47 to 4·86 mm (mean 4·04 mm) and the maximum body width (midway across the uterus) from 1·17 to 1·53 mm (mean 1·39 mm). The ventral surface possesses 3 rows of protrusible glands, 42–47 in total; lateral rows consisting of 14–17 glands (most commonly 16) and a median row of 14–15 glands (most commonly 15). The first median gland is positioned half an interval behind the first lateral glands.
Fig. 1. Notocotylus malhamensis n. sp. from Myodes glareolus. (A) Dorsal surface. (B) Ventral surface. OS, oral sucker; GP, genital pore; C, caecum; M, metraterm; CS, cirrus sac; U, uterus; V, vitelline glands; VG, ventral glands; O, ovary; T, testis.
The oral sucker ranges from 170 to 288 μm in length (mean 234 μm) by 180 to 300 μm (mean 258 μm) in width. The oesophagus measures 150–170 μm (mean 164 μm) in length and bifurcates in front of the genital pore extending into very long blindly ending caeca. The caeca extend posteriorly underlying the uterine loops and curving in-between the laterally positioned testes and a single centrally located ovary. The caeca appear red in live specimens.
Fig. 2. Notocotylus malhamensis n. sp. from Microtus agrestis. (A) Dorsal surface. (B) Ventral surface. OS, oral sucker; GP, genital pore; C, caecum; M, metraterm; CS, cirrus sac; U, uterus; V, vitelline glands; VG, ventral glands; O, ovary; T, testes.
The laterally positioned testes measure 650 and 800 μm in length are lobed in form and extracaecal. The claviform cirrus sac extends from 840 to 1380 μm (mean 1170 μm) in length with the proximal extremity positioned at 40–45% the length from the anterior. The length of the cirrus was difficult to measure. The cirrus is coiled in relaxed specimens when everted (Fig. 3). The width of the cirrus measured from 84 to 86 μm in diameter at the distal end, and is densely spinulated over its entire surface.
Fig. 3. Spinulated cirrus of Notocotylus malhamensis.
The genital pore lies behind the intestinal bifurcation. The uterus is transversely coiled tightly between the base of the cirrus sac and the anterior border of the vitelline reservoir. There are 12–14 uterine loops that overflow the caecal field and 3–4 uterine coils lie ahead of the vitelline glands. The anterior border of the vitelline glands is positioned at 48–50% from the anterior of the body length. Two lateral groups of 14–18 follicles extend to the posterior border of the uterus. All principal uterine loops are posterior to the base of the cirrus sac. Secondary uterine loops can be observed on the lateral side of the cirrus sac, which adjoin the metraterm. The metraterm is long and rectilineal in form and measures from 1:1·5 to 1:1·2 the length of the cirrus sac. Metratermic glands can be observed along the full length of the metraterm. The egg measures 20 μm × 10 μm and bears 2 polar filaments, one on either side (Fig. 4). The filaments range in length from 60 μm to 180 μm and are often unequal in length.
Fig. 4. Egg recovered from the feces of an infected bank vole indicating the extended egg filaments.
Variation in specimens recovered from Microtus agrestis
The following description of a variant type is based on 9 specimens. An illustration of the morphology can be seen in Fig. 2. The body appears identical in form to those recovered from M. glareolus but is more elongated with a body length ranging from 3·73 to 5·05 mm (mean 4·52 mm) and a maximum body width (midway across the uterus) from 1·10 to 1·47 mm (mean 1·36 mm). This elongation of the worms appears to create a more bluntly pointed posterior. Three protrusible rows of ventral glands are present conforming to the pattern of the glands observed in the specimens recovered from M. glareolus, differing only by the last lateral glands extending slightly further towards the posterior and bypassing the last median gland.
The oral sucker measures 210–300 μm (mean 286 μm) in width and 170–230 μm (mean 213 μm) in length. The oesophagus is more elongated ranging from 220 to 290 μm (mean 260 μm) in length. The lobed testes measure 700–900 μm (mean 800 μm) in length. The cirrus sac is more elongated extending from 1050 to 1900 μm (mean 1610 μm). The majority of uterine loops lie intracaecal with only a few loops slightly overlapping the caecal field and there appears to be more space separating the uterine loops. Three to four uterine coils lie ahead of the vitelline glands although in one specimen the vitelline glands extended as far as the anterior border of the uterine coils. The metraterm is more elongated with a range of 770 to 1460 μm (mean 1165 μm); however, the metraterm to cirrus sac ratio remained stable measuring 1:1·5 to 1:1·2 the length of the cirrus sac. Despite size variation seen in N. malhamensis between the two host species, these measurements were not found to be significantly different using a one-way ANOVA with equal replicates (P = 0·5233, d.f. = 1, F = 0·41). DNA sequencing of the ITS region furthermore indicates a 100% sequence homology between N. malhamensis recovered from both M. glareolus and M. agrestis.
Differential comparison of N. malhamensis with other Notocotylus spp
Following an extensive literature search, 68 Notocotylus species names were identified worldwide. Of these, 63 taxa appear to be valid. Cribb (Reference Cribb1991) listed 41 species in his comparison with Notocotylus johnstoni. Kinsella and Tkach (Reference Kinsella and Tkach2005) added a further 8 species. From these 49 species, both Notocotylus gippyensis and N. tadornae (see Bisset, Reference Bisset1977), which possess a single row of ventral glands, have been re-located into Uniserialis Beverley-Burton, 1958 (Barton and Blair, Reference Barton, Blair, Jones, Bray and Gibson2005). Two further species have been described since 2005 N. biomphalaria (Flores and Brugni, Reference Flores and Brugni2005) and N. loeiensis (Chaisiri et al. Reference Chaisiri, Morand and Ribas2011), and 15 names not mentioned, including N. linearis (Rudolphi, Reference Rudolphi1819), N. triserialis Diesing, 1839, N. urbanensis Cort, 1914, N. chionis (Baylis, Reference Baylis1928a), N. intestinalis (Tubangui, Reference Tubangui1932), N. babai (Bhalerao, Reference Bhalerao1935), N. lucknowensis (Lal, Reference Lal1935a), N. anatis and N. orientalis (Ku, Reference Ku1937), N. dafilae and N. porzanae (Harwood, Reference Harwood1939), N. stagnicolae (Herber, Reference Herber1942), N. solitaria (Singh, Reference Singh1954), N. wetlugensis (Shaldybin, Reference Shaldybin1965), and N. lianhuensis (Qiongzhang, Reference Qiongzhang1988), include species that are more controversially valid.
Cribb (Reference Cribb1991) regarded N. anatis, N. babai, N. lucknowensis and N. solitaria as synonyms of N. imbricatus Looss, 1893, and while N. triserialis and N. attenuatus may be synonyms, the relative precedence and validity of the two forms remains disputed (Dubois, Reference Dubois1951; Beverley-Burton, Reference Beverley-Burton1961, Reference Beverley-Burton1972; Pike, Reference Pike1969). In the current context, the most important controversial taxon is N. wetlugensis, described from voles in European Russia (Shaldybin, Reference Shaldybin1965). Frequently considered a synonym of N. noyeri, the taxon was re-established by Tenora et al. (Reference Tenora, Henttonen and Haukisalmi1983), a decision supported by Simone-Vicente et al. (Reference Simon-Vicente, Mas-Coma, Lopez-Roman, Tenora and Gallego1985a).
Notocotylus malhamensis n. sp. could be differentiated from all of these species using morphological characteristics, including the arrangement and composition of the ventral glands, the positioning of the genital pore, the structure of the caecum, the metraterm to cirrus sac ratio, cirrus surface composition and the number and positioning of the uterine coils.
The group of species that include N. gibbus Mehlis, 1846 (Beverley-Burton, Reference Beverley-Burton1961), N. pacifera (Noble, Reference Noble.1933), N. porzanae (Harwood, Reference Harwood1939), and N. regis (Harwood, Reference Harwood1939), have short oval bodies and only 4–5 ventral glands in the central row that are flat and non-protrusible. This group of species is easily distinguishable from Notocotylus sensu stricto.
N. malhamensis could be differentiated from a further 19 Notocotylus species by the arrangement and composition of the ventral glands (Table 1).
Table 1. The number of ventral glands found in each of the three rows of previously described Notocotylus species that can be differentiated from N. malhamensis on this basis
N. malhamensis has a post-bifurcal genital pore and can therefore be differentiated from N. naviformis (Tubangui, Reference Tubangui1932), N. vinodae (Gupta and Singh, Reference Gupta and Singh1983), N. johnstoni (Cribb, Reference Cribb1991), N. fosteri Kinsella and Tkach (Reference Kinsella and Tkach2005), and N. loeiensis (Chaisiri et al. Reference Chaisiri, Morand and Ribas2011), all of which possess a prebifurcal genital pore. N. tachyeretis (Duthoit, Reference Duthoit1931), N. lucknowensis (Lal, Reference Lal1935a), N. mamii (Hsu, Reference Hsu1954), N. nathipandei (Odening, Reference Odening1964), and N. poecilorhynchai (Gupta and Jehan, Reference Gupta and Jehan1977), all have a genital pore that is located ventral to the intestinal bifurcation. Additionally, N. tachyeretis possesses from 21 to 26 uterine coils and a cirrus sac length from 2500–2800 μm, N. lucknowensis has a metraterm to cirrus sac ratio of 1:3, N. mamii has a metraterm that is slightly longer than the cirrus sac and N. nathipandei possesses more than 20 uterine coils, 7 of which are positioned ahead of the vitelline glands.
N. malhamensis has a spinulated cirrus and can additionally be distinguished from N. triserialis Diesing, 1839 (Pike, Reference Pike1969), N. filamentis (Barker, Reference Barker1915), and N. solitaria (Singh, Reference Singh1954), which have a papillated cirrus, and N. noyeri (Joyeux, Reference Joyeux1922), N. intestinalis (Tubangui, Reference Tubangui1932), and N. gonzalezi Simon-Vicente et al. 1985a, all of which possess a cirrus smooth in composition.
The metraterm to cirrus sac ratio of 1:1·5–1:1·2 observed in N. malhamensis can furthermore be used for differentiation from the following described species. The ratios have been indicated in parentheses: N. linearis (Rudolphi, Reference Rudolphi1819) (1:3), N. aegyptiacus Odhner, 1905 (<1:2) (Dubois, Reference Dubois1951), N. urbanensis Cort, 1914 (1:2) (Harrah, Reference Harrah1922), N. magniovatus Yamaguti, Reference Yamaguti1934 (1:4–1:2), N. minutus (Stunkard, Reference Stunkard1960) (1:2), N. atlanticus (Stunkard, Reference Stunkard1966) (<1:2), N. neyrai Gonzalez Castro, 1945 (<1:3·3) (Simon-Vicente et al. Reference Simon-Vicente, Mas-Coma, Lopez-Roman, Tenora and Gallego1985a) and N. zduni Chiaberachvili and Djavelidze 1968 (1:2) (Vassilev and Kanev, Reference Vassilev, Kanev and Vassilev1984). The eggs of N. zduni furthermore possess a long filament at one end and a tuft of short and thin filaments at the opposite end (Vassilev and Kanev, Reference Vassilev, Kanev and Vassilev1984). All eggs observed for N. malhamensis possess one long filament at each end of the egg (Fig. 4).
Furthermore, the length of the cirrus of N. marinus (Ginetsinskaya and Naumov, Reference Ginetsinskaya and Naumov1958), equals twice the length of the uterine coils and the longest length recorded for N. indicus (Lal, Reference Lal1935b) was much less than the smallest record for N. malhamensis. The cirrus composition of N. indicus is furthermore long, thin and feeble with no mention of spines on the surface. In addition, N. urbanensis possesses 10 uterine coils that lie ahead of the vitelline glands.
The species which N. malhamensis most closely resembles, in terms of possessing a long and rectilineal metraterm in combination with spinulation of the cirrus, is N. wetlugensis (Shaldybin, Reference Shaldybin1965). This species, however, also possesses caecal diverticulations, which are absent in N. malhamensis.
The number and arrangement of the uterine coils can also be used for differentiation. N. malhamensis possesses 12–14 uterine coils that overflow the caecal field. The uterine coils of N. attenuatus (Rudolphi, Reference Rudolphi1809) are strictly confined to the caecal field. This species was furthermore ruled out using molecular analysis. The uterine coils of N. imbricatus Looss, 1893 are also confined to the caecal field and this species has a metraterm that is only 1:1·8 the length of the cirrus sac (Beverley-Burton, Reference Beverley-Burton1961). The following species can furthermore be ruled out on having a greater number of uterine coils, as indicated in parenthesis: N. chionis (Baylis, Reference Baylis1928a) (16–26), N. parviovatus (Yamaguti, Reference Yamaguti1934) (>21), N. ralli (Baylis, Reference Baylis1936) (37–34) and N. anatis (Ku, Reference Ku1937) (22–27). Additionally, N. babai (Bhalerao, Reference Bhalerao1935) has 10–11 uterine loops that lie ahead of the vitelline glands and a metraterm to cirrus sac ratio of 1:4–1:2, N. micropalmae (Harwood, Reference Harwood1939), has 9 coils and N. dafilae (Harwood, Reference Harwood1939), has 6–8 coils ahead of the vitelline glands. In addition, N. stagnicolae (Herber, Reference Herber1942), can be ruled out by the anterior half of the body being covered with spines, the caeca possessing dilatations and indentations and although not mentioned in the text, the figure indicates 18–19 uterine coils that are all confined to the caecal field.
Molecular characterization
Amplification of the internal transcribed spacer (ITS) including the 3′ end of the 18S, ITS1, 5.8S, ITS2 and the 5′ end of the 28S generated a sequence of 1236 bp (GenBank Accession number: JQ766940). Sequences obtained from specimens recovered from the caecum of M. glareolus and M. agrestis were 100% identical.
Amplification of the 28S gene generated a partial sequence of 1260 bp. The 28S sequence of N. malhamensis was compared against 3 available DNA sequences that were available from the NCBI database: N. attenuatus (AF184259) collected from an athyid duck in southern Ukraine (Tkach et al. Reference Tkach, Pawlowski, Mariaux, Swiderski, Littlewood and Bray2001), Notocotylus sp. UK-PO-2003 (AY222219), based on sporocyst material from Lymnaea in the UK, and Notocotylus BH-2008 (EU712725) based on larval material from Physa from Nebraska, USA (Hanelt, Reference Hanelt2009). The sequence for N. malhamensis appeared most closely related to Notocotylus BH-2008 sharing 99% sequence homology, followed by 98% with Notocotylus sp. UK-PO-2003 and only 96% with N. attenuatus. The 28S sequence of N. malhamensis has been deposited into GenBank under Accession number: JQ766939.
DISCUSSION
Members of the genus Notocotylus predominantly infect waterfowl; however, a limited number of species infect rodents. From the 63 described species only 10 have previously been recorded as infecting rodents naturally of which 4 have been recorded within Europe: N. noyeri Joyeux, Reference Joyeux1922, N. neyrai Gonzalez Castro, 1945, N. wetlugensis Shaldybin, Reference Shaldybin1965, and N. gonzalezi Simon-Vicente et al. 1985 (Tenora et al. Reference Tenora, Henttonen and Haukisalmi1983; Simon-Vicente et al. Reference Simon-Vicente, Mas-Coma, Lopez-Roman, Tenora and Gallego1985a). Notocotylus malhamensis n. sp appears to be most closely related to these other Notocotylus species from European rodents, sharing similarities in the number and arrangement of ventral glands, the number and positioning of the uterine coils and the positioning of the genital pore. As shown by Simon-Vicente et al. (Reference Simon-Vicente, Mas-Coma, Lopez-Roman, Tenora and Gallego1985a), the metraterm to cirrus sac ratio in combination with cirrus composition can be used as a basis for differentiating these species. Those criteria in combination with the structure of the caecum can furthermore be used to distinguish N. malhamensis from the other European forms. The presence of a long and rectilineal metraterm, dense spinulation of the cirrus and simple caeca are a combination seen in N. malhamensis that has not been observed in any of the other European forms.
Notocotylus malhamensis from the field vole Microtus agrestis were morphologically comparable to the specimens from Myodes glareolus when using the above key taxonomic features. The M. agrestis specimens were, however, more elongated with a greater range in length, smaller width and elongation of internal structures, although these differences were not statistically significant. A difference in the width and spatial separation of the uterine loops was also observed. In the specimens from M. glareolus the uterine loops consistently extend beyond the caecal field, whereas in specimens from M. agrestis the uterine loops appear to only slightly extend beyond the caeca. The specimens from M. agrestis may represent juvenile adults that had not fully developed, as Kinsella (Reference Kinsella1971) demonstrated that in Quinqueserialis quinqueserialis, a species closely related to Notocotylus, at 12 days post-infection the uterine coils were confined within the caecal field, while by day 15 they had begun to overlap the gut caeca.
The detailed life cycles of most Notocotylus species remain unknown. Typically, a brackish or freshwater lymnaeid or hydrobiid becomes infected by eating eggs released into water with feces of the definitive host (Murrils et al. Reference Murrills, Southgate and Reader1988). Following development in the snail, actively swimming eyed cercariae are released which quickly encyst on nearby solid objects such as aquatic vegetation (Simon-Vicente et al. Reference Simon-Vicente, Mas-Coma, Lopez-Roman, Tenora and Gallego1985b) or even the snail shell. The definitive host becomes infected by ingestion of the encysted metacercariae together with vegetation during foraging.
Notocotylus malhamensis was recovered from M. glareolus and M. agrestis captured from the wetland site (Tarn Fen) only. No infection was observed in either host captured from any of the 3 woodland sites (0/48), suggesting that flooding of the wetland area is potentially important in terms of transmission of N. malhamensis to the vole populations. Typically, voles may be restricted to feeding at the water edge and as such will be susceptible to infection by metacercariae present only around the water margin (Webber et al. Reference Webber, Rau and Lewis1987). During periods of flooding, invertebrate species inhabiting the water column are likely to be conveyed into the typical foraging range of the vole communities along with floating aquatic vegetation and detritus. As the water recedes, both snails and plants infected with metacercariae are likely to be left on the ground. Aquatic plants have been observed littering this area on several occasions, a factor that could contribute to exposure of voles to infective metacercariae while foraging.
Prior to sampling, the last bout of flooding took place on 13 October 2011 at which point the water level at Tarn Fen rose to 200 mm above normal levels (Hodgson, personal communication). Microtus agrestis were trapped on the 28 October 2011. If flooding is an important criterion for infection to occur then it might be speculated that the estimated age of the N. malhamensis samples recovered from M. agrestis would be less than 15 days old, indicating that the uterine coils may not yet be fully developed.
The elongate appearance and increased range of internal measurements of N. malhamensis from M. agrestis may, however, represent their ability to grow to a much larger size in this host than in M. glareolus. Variation in morpho-anatomic measurements of digeneans in different host species has been previously reported to be host-dependent (Kinsella, Reference Kinsella1971). This intraspecific variation observed in morpho-anatomic measurements between the two host species at Tarn Fen reinforces the fact that no reliance can be placed upon the size of the worm or its internal organs for species differentiation and that key taxonomic features which remain constant across a species should instead be used (Lal, Reference Lal1935a).
A maximum of 6 N. malhamensis were recovered from any individual M. agrestis infection, while up to 294 worms were recovered from the caecum of a single M. glareolus. The gross anatomy of the caecum differs between the two rodent species, being only 13 cm in M. glareolus compared to 24 cm in M. agrestis (see Lange and Staaland, Reference Lange and Staaland1970). This difference alone may contribute to ‘crowding’ during heavy infections (Read, Reference Read1951). N. malhamensis may therefore appear more stunted in M. glareolus due to adaptation in response to limitations in physical space and may undergo morpho-anatomic variation in size according to host species.
The origin of N. malhamensis at Malham Tarn could be avian. Malham Tarn boasts a diverse population of birds of which as many as 200 species have been recorded (Sutton and Shorrock, Reference Sutton and Shorrock2008). Other species of Notocotylus such as N. imbricatus Looss, 1893 (Cribb, Reference Cribb1991) and N. ephemera Nitzsch 1817 (Gibson et al. Reference Gibson, Bray and Harris2005) have been recorded from both bird and mammalian hosts. Ablasov and Iksanov (1958) identified N. noyeri Joyeux, Reference Joyeux1922, from the caeca of a piscivorous bird and commented that this species, that is frequently recovered from the caecum of murids, might actually be a parasite of birds (Simon-Vicente et al. Reference Simon-Vicente, Mas-Coma, Lopez-Roman, Tenora and Gallego1985a). Further research into the life cycle of N. malhamensis at Tarn Fen is required. Five species of freshwater snail (Lymnaea peregra, Lymnaea truncatula, Anisus leucostoma, Psidium sp. and Potamopygrus antipodarum) have been repeatedly sampled from Tarn Fen throughout the study period, but to date no snails infected with N. malhamensis have been identified, although it is likely that one of these snail species is the intermediate host at Tarn Fen.
Confirmation of a new species can be simplified by combining the use of modern molecular tools with the classical approach of species identification. Currently, DNA sequences for the genus Notocotylus are poorly represented with only 5 sequences (3 for 28S rDNA, 2 for 18S rDNA) available in NCBI. Furthermore, 2 of the 3 28S sequences can only be matched with Notocotylus at the genus level, as they were derived from incompletely identified larval material. This genus fully illustrates the value of molecular differentiation in conjunction with classical identification for genera where differences in morphology are minute and many life cycles are incompletely understood. Reports that include the DNA sequence of a common gene in addition to a full morphological description will be beneficial for future workers, in particular if we are to record new parasite occurrences.
ACKNOWLEDGEMENTS
This project was funded by the University of Salford Ph.D. Studentship scheme. We would like to thank The National Trust for granting the licence which enabled us to carry out this work and all staff at Malham Tarn FSC field centre, in particular the current warden Mike Cawthorn and Kate Martin for support and assistance during the sampling period. We would also like to thank Professor Richard Birtles and Professor Jerzy Behnke for advice provided during species identification and Dr Belgees Boufana and Mr Tony Bodell for technical advice. We would furthermore like to thank David Hodgson for providing important information and finally Jaroslav Bajnok and Michelle Dasic for assistance during the summer sampling and the many undergraduate and post-graduate students who contributed to the fieldwork during the autumn sampling period.
