Introduction
Serpinema Yeh, 1960 (Nematoda: Camallanidae) currently includes 12 valid species (Silva et al., Reference Silva, Jesus, Santos, Nandyara, Willkens, Santos and Melo2023, Reference Silva, Santos, Trindade and Melo2026) whose adult stages parasitize the gastrointestinal tract of freshwater turtles and occurring in the Neotropical, Oriental and Palearctic regions (Ivashkin et al., Reference Ivashkin, Soboloev and Khromova1977; Harnoster et al., Reference Harnoster, Svittin and Du Preez2019; Silva et al., Reference Silva, Jesus, Santos, Nandyara, Willkens, Santos and Melo2023). Members of this genus are characterized by a well-sclerotized buccal capsule formed by two lateral, shell-like valves, with internal longitudinal ridges arranged into dorsal and ventral groups, separated by a distinct gap (Yeh, Reference Yeh1960; Silva et al., Reference Silva, Santos, Trindade and Melo2026).
Serpinema was erected based on 10 species of the genus Camallanus to found parasitizing freshwater turtles (Yeh, Reference Yeh1960), namely: Se. amazonicum (Ribeiro, Reference Ribeiro1941), Se. intermedius (Hsu and Hoeppli, 1931), Se. kachugae (Baylis and Daubney, Reference Baylis and Daubney1922), Se. magathi (Sprehn, 1932), Se. magnorugosus (Caballero, 1939), Se. microcephalus (Dujardin, Reference Dujardin1845), Se. octorugatum (Baylis, 1933), Se. ptychozondis (MacCallum, 1918), Se. trispinosum (Leydi, 1851) and Se. undulatus (Railliet and Henry, 1915). Subsequently, Se. magnorugosus was regarded as a junior synonym of Se. trispinosum (Moravec and Vargas-Vázquez, (Reference Moravec and Vargas-Vázquez1998)), whereas Se. ptychozondis and Se. undulatus were considered invalid (Sharma et al., Reference Sharma, Rigby, Sumita, Sanira, Vidyadaran, Jasni and Dailey2002). In addition, Se. lissemysus (Gupta and Singh, 1959) was not included by Yeh (Reference Yeh1960) and four further species were later described: Se. cayennensis Harnoster, Svitin and Du Preez, 2019, Se. emydidium (Mascarenhas and Müller, Reference Mascarenhas and Müller2017), Se. monospiculatum Freitas and Dobbin Jr., 1962, and Se. pelliculatum Silva, Jesus and Melo, 2023 (Harnoster et al., Reference Harnoster, Svittin and Du Preez2019; Silva et al., Reference Silva, Santos, Trindade and Melo2026).
Serpinema spp. are morphologically very similar to those of the genera Camallanus Railliet and Henry, 1915 and Zeylanema Yeh, 1960, which has led several authors to question the validity of Serpinema and Zeylanema as independent taxa. Consequently, both have been treated either as synonyms (Stromberg and Crites, Reference Stromberg and Crites1974; Rigby and Rigby, Reference Rigby, Rigby and Schmidt-rhaesa2014) or as subgenera (Baker, Reference Baker1983; Moravec and Scholz, Reference Moravec and Scholz1991; Moravec and Van As, Reference Moravec and Van As2015) of Camallanus.
The taxonomic status of Serpinema was previously addressed by Harnoster et al., (Reference Harnoster, Svittin and Du Preez2019), who conducted phylogenetic analyses of species of Camallanus and Serpinema using three molecular markers. Analyses of the 28S gene showed that, although Serpinema spp. did not form a fully distinct clade, Se. octorugatum and Se. cayennensis did not cluster with other Camallanus species, and the 18S and COI analyses indicated more distant relationships between Se. cayennensis and Camallanus spp. parasitizing fishes and amphibians.
Based on the evidence supporting the validity of the genus Serpinema provided by Harnoster et al., (Reference Harnoster, Svittin and Du Preez2019) and Ailán-Choke and Pereira (Reference Ailán-Choke and Pereira2021), Silva et al. (Reference Silva, Santos, Trindade and Melo2026) proposed an identification key for the genera of Camallanidae. In this key, Serpinema and Zeylanema are recognized as valid genera, and the authors emphasize that the current family classification does not support the recognition of subgenera.
In this context, seven species of Serpinema have been recorded in Brazil, 6 of which parasitize freshwater chelonians: Se. amazonicum, Se emydidium, Se. magathi, Se. microcephalus, Se. monospiculatum and Se. pelliculatum. Additionally, larval stages of Se. trispinosum have been reported to infect fish (Falkenberg et al., Reference Falkenberg, De Lima, Yamada, Ramos and Lacerda2024; Silva et al., Reference Silva, Santos, Trindade and Melo2026).
During a helminthological survey of parasites of freshwater turtles, nematodes were recovered from the gastrointestinal tract of Rhinoclemmys punctularia (Daudin, 1801) in the Brazilian Amazon region. The specimens were assigned to the genus Serpinema, with morphological features that clearly differentiate them from their congeners. Thus, the present study aims to describe a new species of Serpinema based on morphological and molecular data.
Materials and methods
Nematode collection and morphological analysis
We examined the gastrointestinal tracts of four specimens of Rhinoclemmys punctularia collected in July 2022 in the municipality of Abaetetuba, Pará State, Brazil (SISBIO license no. 53527–4). Hosts were anaesthetized by injection of 2% ketamine and subsequently euthanized by ketamine overdose. All internal organs were removed and examined under a Leica EZ4 stereomicroscope (Leica Microsystems, Wetzlar, Germany). Nematodes were washed in 0.9% saline solution, fixed in heated 70% ethanol and preserved in microtubes with room-temperature 70% ethanol. Some specimens were stored at 4 °C in 100% ethanol for molecular analyses.
For morphological and morphometric analyses, specimens were cleared in Amann’s lactophenol, mounted on temporary slides and examined under an Olympus BX41 microscope (Olympus, Tokyo, Japan) equipped with a drawing tube. For scanning electron microscopy (SEM), 6 specimens (3 males and 3 females) were post-fixed in 1% OsO₄ (osmium tetroxide), dehydrated through an ascending graded ethanol series, dried at the CO₂ critical-point, mounted on metallic stubs, coated with gold–palladium (Silva et al., Reference Silva, Jesus, Nandyara, Willkens, Tavares-Costa, Santos and Melo2024) and analysed under a Vega3 microscope (TESCAN, Brno, Czech Republic) with an accelerating voltage of 10–20 kV, at the Laboratory of Structural Biology (LBE), Federal University of Pará (UFPA).
All measurements are given in micrometres unless otherwise stated and are presented as Holotype/allotype, followed by the mean value and minimum and maximum measurements of the type series indicated in parentheses. Prevalence, mean abundance and mean intensity were calculated using the methods of Bush et al., (Reference Bush, Lafferty, Lotz and Shostak1997). The specimens were deposited in the Other Invertebrate Collection of Museu Paraense Emílio Goeldi (MPEG), Pará, Brazil.
Molecular and phylogenetic analyses
For molecular analyses, nematode specimens were preserved in microtubes with absolute ethanol and stored at −20 °C. Specimens used for DNA extraction were sectioned, and the anterior and posterior regions of a single female specimen were preserved and stored separately in ethanol. The mid-body region was used for DNA extraction. The hologenophore, as defined by Pleijel et al., (Reference Pleijel, Jondelius, Norlinder, Nygren, Oxelman, Schander, Sundberg and Thollesson2008), was preserved and deposited as a voucher specimen in a helminthological reference collection (MPEG).
Genomic DNA was extracted using the DNeasy Blood and Tissue Kit (Qiagen) according to the manufacturer’s standard protocol, and we amplified the ribosomal 18S gene and the mitochondrial cytochrome c oxidase subunit I (COI) gene. The 18S gene was amplified using the primers 18S_F (forward) (5′-CGC GAA TRG CTC ATT ACA ACA GC-3′) and 18S_R (reverse) (5′-GGG CGG TAT CTG ATC GCC-3′), as described by Ailán-Choke et al. (Reference Ailán-Choke, Tavares, Luque and Pereira2020). The thermocycling conditions consisted of an initial denaturation at 94 °C for 5 min, followed by 36 cycles of denaturation at 94 °C for 30 s, annealing at 52.7 °C for 30 s and extension at 72 °C for 1 min, with a final extension at 72 °C for 5 min (Ailán-Choke et al., Reference Ailán-Choke, Tavares, Luque and Pereira2020).
The COI gene was amplified using the primers LCO (forward) (5′-GGT CAA CAA ATC ATA AAG ATA TTG G-3′) and HCO (reverse) (5′-TAA ACT TCA GGG TGA CCA AAA AAT CA-3′), as described by Harnoster et al. (Reference Harnoster, Svittin and Du Preez2019). The thermocycling conditions were as follows: initial denaturation at 95 °C for 5 min; 36 cycles of denaturation at 95 °C for 1 min, annealing at 40 °C for 1 min and extension at 72 °C for 1 min 30 s; followed by a final extension at 72 °C for 7 min (Folmer et al., Reference Folmer, Black, Hoeh, Lutz and Vrijenhoek1994).
PCR reactions for both genes were performed in a final volume of 25 μL, containing 12.5 μL of OneTaq® 2X Master Mix with Standard Buffer, 1.0 μL (10 μM) of each primer, 2.0 μL of genomic DNA (≥20 ng) and 8.5 μL of nuclease-free water. PCR products were quantified, enzymatically purified using ExoProStar™ (GE Healthcare) and submitted for sequencing at ACTGene (Ludwig Biotec, Rio Grande do Sul, Brazil) with the same primers used for PCR amplification.
Contiguous sequences were assembled using Geneious v. 9.1.5 (Biomatters, available from http://www.geneious.com/) and deposited in GenBank. A preliminary BLASTn search was conducted to assess genetic similarity between the newly generated sequences and those of Camallanidae available in the NCBI database (https://www.ncbi.nlm.nih.gov/genbank/).
For phylogenetic analysis, we selected sequences with a minimum length of 500 bp for 18S and 300 bp for COI. We also excluded sequences whose amplified regions did not align with those generated in the present study. In cases where multiple highly similar sequences of the same gene represented a single taxon, often resulting from clones or isolates of a single species within the same study, we retained only one representative sequence.
Spirocerca lupi (Rudolphi, 1809) (AY751497) was selected as the outgroup for the 18S dataset, while Ancyracanthus electrophoris Virgilio and Pereira, 2023 (OP765585) and Spiroxys ankarafantsika Roca and Garcia, 2008 (MW545829) were outgroups for the COI dataset. Table 1 lists the taxa included in the analysis. Although we updated the species names to follow the classification proposed by Silva et al. (Reference Silva, Santos, Trindade and Melo2026), the original names recorded in the NCBI database are also provided.
Camallanid species included in the study, showing updated taxonomic names, corresponding names as listed in the NCBI database, host species, country of record, habitat and gene sequences used in the phylogenetic analyses
Table 1 Long description
Camallanid nematode species are listed by subfamily, with updated species names alongside the corresponding NCBI database name, host, country of record, habitat, and available gene sequences used for phylogenetic analyses. Most taxa fall under Camallaninae and Procamallaninae, plus three outgroup species. Records cover a wide geographic range including South Africa, China, Mexico, Argentina, Brazil, Peru, New Caledonia, India, and other locations, and include both freshwater and marine habitats, with freshwater appearing more frequently. Many species have multiple rows reflecting different hosts, countries, or sequence submissions, such as Camallanus cotti, Procamallanus laeviconchus, and several Spirocamallanus species. Sequence availability is uneven: 18S accessions are common, while COI is missing for many entries; a smaller subset provides both markers, for example Paracamallanus cyathhopharynx, Procamallanus pseudolaeviconchus, Procamallanus spiculogubernaculus, Zeylanema sp., and some Spirocamallanus records. Several entries note unpublished sequences or incomplete metadata such as missing country, habitat, or host identification, which limits direct comparisons across all rows. The table also highlights naming mismatches where the study’s updated taxonomy differs from the NCBI name used for deposited sequences, especially among Procamallanus and Spirocamallanus listings.
Notes: aUsed as in Ailán-Choke and Pereira (Reference Ailán-Choke and Pereira2021).
Our sequences were aligned using MUSCLE (Edgar, Reference Edgar2004) with default parameters implemented in Geneious v. 7.1.3. After trimming the terminal regions of the alignments, we assessed the presence of stop codons using the second translation frame under the invertebrate mitochondrial genetic code in Geneious (Kearse et al., Reference Kearse, Moir, Wilson, Stones-havas, Cheung, Sturrock and Drummond2012).
Substitution saturation in the aligned dataset was assessed using the ISS index in DAMBE 7 (Xia, Reference Xia2018). Genetic divergence estimates, including pairwise base substitution rates per site, were calculated in MEGA11 (Kimura, Reference Kimura1980; Tamura et al., Reference Tamura, Stecher and Kumar2021), with standard errors estimated using a bootstrap procedure with 1000 replicates. Model selection based on the Akaike Information Criterion identified GTR + I + G as the best-fitting nucleotide substitution model, which was subsequently applied in phylogenetic analyses.
We performed phylogenetic reconstructions using maximum likelihood (ML) in RAxML v. 8.2.12 and Bayesian inference (BI) in MrBayes v. 3.2.7a (Guindon and Gascuel, Reference Guindon and Gascuel2003; Ronquist et al., Reference Ronquist, Huelsenbeck, Teslenko, Zhang and Nylander2003). Both analyses were performed in the CIPRES Science Gateway (Miller et al., Reference Miller, Pfeiffer and Schwartz2010). ML analyses included 1000 bootstrap replicates, and we considered only nodes with bootstrap support values greater than 70% as well-supported clades.
BI employed the following settings: lset nst = 6, rates = invgamma, ngammacat = 4, nucmodel = 4by4, code = universal and prset statefreqpr = Dirichlet (1,1,1,1). Markov Chain Monte Carlo analyses were run for 10 000 000 generations, sampling one tree every 1500 generations. The initial 25 000 generations were discarded as burn-in. A majority-rule consensus tree was generated from the remaining topologies, using the commands sumt relburnin = yes and sump relburnin = yes. Only nodes with Bayesian posterior probabilities greater than 90% were considered well supported. Trees were visualized and edited in FigTree v. 1.3.3 (Rambaut, Reference Rambaut2009).
Results
Taxonomic summary
Family Camallanidae Railliet and Henry, 1915
Subfamily Camallaninae Yeh, 1960
Genus Serpinema Yeh, 1960
Serpinema yehi n. sp. Silva, Tavares-Costa, and Melo (Figs. 1, 2, 3).
Line drawings of the anterior region of Serpinema yehi n. sp. Silva, Tavares-Costa and Melo, parasite of Rhinoclemmys punctularia from the Brazilian Amazon. (A) Anterior region of male, ventral view. (B) Anterior region of female, ventral view. (C) Valve of buccal capsule, ventral view. (D) Cephalic extremity, apical view. (E) Trident, ventral view. Scale bars: (A and B) 350 µm; (C and D) 100 µm; (E) 50 µm.
Line drawings of the posterior region of Serpinema yehi n. sp. Silva, Tavares-Costa, and Melo, parasite of Rhinoclemmys punctularia from the Brazilian Amazon. (A) Posterior region of male, showing caudal papillae and spicules, lateral view. (B) Vulva, lateral view. (C) Posterior region of female, ventral view. (D) Posterior extremity of female tail, showing the four mucrons, ventrolateral view. Scale bars: (A and B) 200 µm; (C) 100 µm; (D) 15 µm.
Scanning electron microscopy of Serpinema yehi n. sp. Silva, Tavares-Costa, and Melo, parasite of Rhinoclemmys punctularia from the Brazilian Amazon. (A) Anterior region, ventral view, showing deirid (arrow) and excretory pore (arrowhead). Inserts: Lateroventral view of the anterior region and detail of excretory pore. (B) Posterior region of male, lateroventral view, showing precloacal papillae (asterisk). (C) Vulva, ventral view, showing the anterior lip and the small lateral projection (arrow). (D) Posterior region of female, subventral view. Insert: detail of tip of female showing four mucrons (asterisk). Abbreviations: Al, anterior lip; An, anus; Tr, trident. Scale bars: (A) 100 µm, Insert, 50 µm (anterior region), 10 µm (excretory pore); (B) 100 µm; (C) 200 µm; (D) 50 µm, Insert, 10 µm.
Type host: Rhinoclemmys punctularia (Daudin, 1801).
Type locality: Abaetetuba (01°43’46” S 48°52’27” W), Pará, Brazil.
Site of infection: Stomach, small intestine and large intestine.
Prevalence: 100% (4 hosts infected out of 4 examined).
Mean abundance and mean intensity: both 35.75 (143 specimens recovered, including 71 males and 72 females).
Specimens deposited: Holotype (MPEG.NEM 000423), allotype (MPEG.NEM 000426), 9 male paratypes (MPEG.NEM 000424), 9 female paratypes (MPEG.NEM 000425) and voucher hologenophore (MPEG.NEM 000427) deposited in the Other Invertebrate Collection of Museu Paraense Emílio Goeldi, Pará, Brazil.
ZooBank registration: To comply with the regulations set out in article 8.5 of the amended 2012 version of the International Code of Zoological Nomenclature (ICZN, International Commission on Zoological Nomenclature, 2012), details of the new species have been submitted to ZooBank. The Life Science Identifier for Serpinema yehi n. sp. is urn:lsid:zoobank.org:act:C92AF89C-C3E2-4D72-A349-A3F94DF44316.
Representative DNA sequences: GenBank [PZ113237] (18S); [PZ113265] (COI).
Etymology: The specific epithet honors Yeh Liang-Sheng in recognition of his significant contributions to parasitology, particularly to the systematics of the family Camallanidae.
Description
General: Large nematodes. Reddish in life. Cuticle bearing fine transverse striations along entire body length. Anterior extremity truncated. Lips absent. Oral opening as transverse slit; one pair of lateral amphids; two pairs of small cephalic papillae close to oral opening; two pairs of cephalic papillae more distant from oral opening. Two apical pairs of curved sclerotized plates. Buccal capsule composed of two laterally positioned, internally concave valves (Figure 1D). Valves internally with longitudinal complete ridges (extending from anterior to posterior margin of buccal capsule) and incomplete ridges (extending from anterior margin to before midlength of buccal capsule) (Figure 1C). Two tridents, dorsal and ventral (Figure 1D), slightly rounded basal portion, 3 divergent branches; median branch shorter, less robust; lateral branches similar (Figures 1E, 3A). Chitinous basal ring surrounding junction between buccal capsule and oesophagus. Oesophagus elongate, divided; anterior muscular region short; posterior glandular region long. Nerve ring at midlength of muscular oesophagus. Excretory pore at level of posterior half of muscular oesophagus. Deirids posterior to excretory pore (Figure 3A).
Males (based on holotype and 9 paratypes): Total length 10.92 (10.70, 9.72–11.73) mm. Body width at midbody 307 (332, 267–387). Sclerotized plates 9 (14, 9–18) × 40 (42, 34–47). Buccal capsule 122 (124, 118–129) × 153 (152, 143–162). Buccal valves with 9 (10, 8–11) complete ridges, 6 (4, 3–7) incomplete ridges. Median trident branch 69 (81, 68–100) long; lateral trident branches 98 (101, 94–113) long. Basal ring 16 (17, 13–21) × 103 (105, 99–116). Oesophagus 1.283 (1.246, 1.157–1.328) mm long. Muscular oesophagus 512 (480, 435–525) × 152 (168, 144–181). Glandular oesophagus 771 (767, 717–813) × 168 (195, 160–237). Nerve ring, excretory pore, and deirids 245 (242, 221–284), 464 (417, 309–485), and 507 (501, 453–555), respectively, from anterior end (Figure 1A). Tail ventrally curved, ending with two tips, 160 (192, 157–227) long; strong transverse musculature. Caudal alae well developed, beginning at level of globose, vesicle-like swelling, ending near tail tip. Twelve pairs of caudal papillae. Seven pedunculate precloacal papillae, all lateral (Figure 3B). Two adcloacal, one pair on anterior cloacal lip, one on posterior cloacal lip. Three pedunculate postcloacal, all lateral, clustered near cloaca. Phasmids present, near posterior end. Gubernaculum absent. Spicules unequal in length and shape; right spicule 869 (800–891) long, posterior end spatulate or leaf-like in shape; left spicule 389 (373–453) long, with tapered posterior end (Figure 2A).
Females (based on allotype and 9 paratypes): Total length 17.16 (16.65, 15.24–18.00) mm. Body width at midbody 440 (455, 413–507). Sclerotized plates 16 (17, 14–20) × 52 (52, 48–57). Buccal capsule 151 (151, 132–160) × 197 (191, 182–199). Buccal valves with 10 (10, 9–11) complete ridges, 4 (4, 1–6) incomplete ridges. Median trident branch 110 (102, 82–127) long; lateral trident branches 121 (126, 110–141) long. Basal ring 19 (20, 17–22) × 129 (130, 119–147). Oesophagus 1.435 (1.391, 0.94–1.53) mm long. Muscular oesophagus 517 (560, 517–592) × 200 (197, 184–213). Glandular oesophagus 864 (875, 803–939) × 245 (225, 181–253). Nerve ring, excretory pore and deirids 280 (286, 261–304), 552 (472, 371–563) and 563 (567, 509–677), respectively, from anterior end (Figure 1B). Vulva near midbody; located 8.16 (8.37, 7.53–8.77) mm from posterior extremity; anterior lip well developed, bearing small lateral projection near vulvar opening (Figures 2B, 3C); posterior lip indistinct. Ovoviviparous. Tail 301 (337, 237–544) long (Figure 2C); ending in 4 tips or mucrons, 2 ventrolateral, 1 dorsal and 1 small dorsolateral (Figures 2D, 3D). Phasmidial pores 184 (181, 157–213) from posterior end (Figure 2C).
Molecular analyses results
We obtained a 583 bp sequence of the mitochondrial COI gene of Se. yehi n. sp., whereas the ribosomal 18S gene yielded a sequence of 833 bp. BLASTn similarity searches revealed no identical matches (100% identity) within Camallanidae sequences currently available in the NCBI database. After aligning our sequences with those from GenBank, we generated three distinct matrices: a 505 bp matrix for the 18S gene and a 340 bp matrix for the COI gene, both of which included representative taxa from the family Camallanidae. Additionally, we constructed a 427 bp COI gene matrix comprising only species from the subfamily Camallaninae.
We did not find any 18S gene sequences of Serpinema species available in the NCBI database. Thus, Se. yehi n. sp. appeared more closely related to Camallanus pallidipectoris, a parasite of the freshwater turtle Acanthochelys pallidipectoris from Argentina, and to Zeylanema sp., a parasite of the freshwater fish Mastacembelus erythrotaenia from Japan (Figure 4). The new species exhibited genetic divergence of up to 10% compared with Camallanidae sequences available in GenBank. The lowest divergence (1%) was observed between Se. yehi n. sp. and sequences MN936176 (C. pallidipectoris) and LC813289 (Zeylanema sp.), whereas the highest (10%) occurred relative to sequence PP581381 (Procamallanus laeviconchus). Interestingly, divergence between Se. yehi n. sp. and Pr. laeviconchus was greater than that observed in relation to the outgroup Spirocerca lupi (AY751497) (7%). Genetic divergence values among sequences are presented in Supplementary Material 1.
Maximum likelihood phylogenetic tree of Camallanidae species based on the 18S gene, using Spirocerca lupi as outgroup. The position of Serpinema yehi n. sp. is indicated in bold italics. GenBank accession numbers are provided after each taxon. Node support values are shown above or below branches; posterior probabilities <0.90 and bootstrap values <70 are either not shown or indicated by a dash. The branch-length scale bar represents the number of substitutions per site.
Figure 4 Long description
The diagram is a maximum likelihood phylogenetic tree of Camallanidae species based on the 18S gene. The tree is oriented vertically with branches extending from top to bottom. The outgroup is Spirocerca lupi, labeled at the bottom. The tree is divided into two main clades, Camallaninae and Camallaninae, highlighted with blue shading. Each branch is labeled with species names and GenBank accession numbers. Node support values are shown above or below branches, with some values omitted or indicated by a dash if below thresholds. The position of Serpinema yehi n. sp. is marked in bold italics. The branch-length scale bar at the bottom left represents the number of substitutions per site, set at 0.03. The tree illustrates the evolutionary relationships among the species, with specific nodes indicating divergence points.
Our phylogenetic analysis of the 18S gene recovered four well-supported subclades comprising more than one species. The first subclade includes Spirocamallanus hilarii, from Psalidodon endyi (freshwater) from Argentina, Sp. pintoi, recorded from Corydoras atropersonatus (Peru) and Corydoras micracanthus (Argentina) (both freshwater), together with Sp. inopinatus, parasitizing Anostomoides passionis and Megaleporinus elongatus (freshwater) from Brazil and Parupeneus forssalis from Egypt (marine).
The second subclade includes Denticamallanus rarus, a parasite of Trichomycterus spegazzinni (freshwater) from Argentina and Sp. huacraensis, a parasite of Calophysus macropterus (freshwater) from Peru. The third subclade comprises Sp. fulvidraconis, a parasite of Pelteobagrus fulvidraconis (freshwater) from China; Pr. pseudolaeviconchus, from Clarias gariepinus from South Africa, and Pr. laeviconchus, recorded from Clarias gariepinus in Egypt and from Synodontis schall in Sudan (all freshwater). Finally, the fourth subclade groups D. rarus, parasitizing Aguarunichthys cf. tocantinensis (freshwater) from Peru, and Pr. pacificus, from Anguilla obscura (freshwater) from New Caledonia (Figure 4).
Nine well-supported subclades composed of a single species were recovered: Sp. istblenni, a parasite of Lutjanus kasmira from Hawaii and Lutjanus fulvus (both marine); C. oxycephalus, parasitizing marine fishes from India; Sp. pintoi, parasitizing Corydoras atropersonatus from Peru and Corydoras micracanthus from Argentina (both freshwater); Sp. inopinatus, parasitizing Anostomoides passionis and Megaleporinus elongatus (freshwater) from Brazil and Parupeneus forsskali from Egypt (marine); Paracamallanus cyathopharynx, from Clarias gariepinus in South Africa (freshwater); C. lacustris, parasitizing Sander lucioperca (freshwater) and an unidentified freshwater fish from China; C. cotti, parasitizing Lentipes concolor (freshwater); Pr. pseudolaeviconchus, also from Clarias gariepinus in South Africa (freshwater); Pr. laeviconchus, recorded from Clarias gariepinus in South Africa and from an unidentified freshwater fish in Egypt. The subclades corresponding to Sp. inopinatus and Sp. pintoi are nested within a larger clade composed of Sp. inopinatus, Sp. hilarii and Sp. pintoi. Likewise, the subclades of Pr. pseudolaeviconchus and Pr. laeviconchus are included within the clade formed by Sp. fulvidraconis, Pr. pseudolaeviconchus and Pr. laeviconchus (Figure 4).
In the phylogenetic analysis of the COI gene, Se. yehi n. sp. clustered with Se. cayennensis, a parasite of Rhinoclemmys punctularia from French Guiana. Together, these taxa formed a well-supported clade representing the genus Serpinema (Figure 5). The new species exhibited high genetic divergence compared with other Camallanidae taxa available in GenBank. The lowest divergence (6%) was observed with the only other congener, Se. cayennensis (MN104841), whereas the highest divergence (21%) occurred between Se. yehi n. sp. and Pr. pseudolaeviconchus (OP602967 and OP602969). In comparison with the outgroups, divergence values were 38% and 43% relative to Spiroxys ankarafantsika (MW54582) and Ancyracanthus electrophoris (OP765585), respectively. Genetic divergence values among sequences are presented in Supplementary Material 2.
Maximum likelihood phylogenetic tree of Camallanidae species based on the COI gene, using Ancyracanthus electrophoris and Spiroxys ankarafantsika as outgroups. The position of Serpinema yehi n. sp. is indicated in bold italics. GenBank accession numbers are provided after each taxon. Node support values are shown above or below branches; posterior probabilities <0.90 and bootstrap values <70 are either not shown or indicated by a dash. The branch-length scale bar represents the number of substitutions per site.
Figure 5 Long description
A horizontal phylogenetic tree is arranged left to right, with branching lines extending from a left trunk toward taxon labels on the right. Two wide horizontal shaded bands on the left are each labeled, Camallaninae. At the bottom, a separate baseline is labeled, Outgroup. A scale bar at the lower left is labeled, 0.4. The two outgroup tips at the lower right read, Ancyracanthus electrophoris OP765585 and Spiroxys ankarafantsika MW545829. Within the main tree, many tip labels show a genus and species name followed by an accession code. Near the middle right, the label, Serpinema yehi n. sp. ZJ113265 appears in bold italics and is paired as a two tip split with, Serpinema cayennensis MN104841. A node label next to the branch leading to this Serpinema pair reads, 0.90 slash 98. Other node labels appear at multiple branch points as numbers or as a dash, including 1 slash 100 near the left trunk and 0.99 slash 98 near the upper right. Several nearby tips include Camallanus cotti with multiple accession codes, and other genera including Spirocamallanus, Paracamallanus, and Procamallanus, each shown as clusters of branching tips.
Based on the COI matrix for Camallanidae, we obtained four well-supported subclades comprising multiple species. The first subclade includes Sp. barlowi, a parasite of Amphilophus citrinellus from Nicaragua, and Sp. rebecae, which parasitizes cichlids, including Cibroheros robertsoni, in Mexico (freshwater). The second subclade comprises C. cotti from China, specimens infecting Raiamas bola in India, and C. hypophthalmichtys, also from China. The third subclade corresponds to the clade formed by Serpinema yehi n. sp and Se. cayennensis. The fourth subclade grouped Sp. xenopodis, parasitic in Xenopus muelleri and Pr. slomei, from Xenopus laevis, both recorded in South Africa. This subclade is nested within a larger, well-supported clade comprising Sp. xenopodis, Pr. slomei, Pr. spiculogubernaculus (a parasite of Heteropneustes fossilis from India) and Pr. pseudolaeviconchus (from Clarias gariepinus in South Africa) (Figure 5).
We also obtained nine single-species subclades with strong support. These included: Sp. neocaballeroi, parasitizing Astyanax aeneus and Astyanax caballeroi in Mexico; Sp. barlowi, from Amphilophus citrinellus in Nicaragua; Sp. rebecae, infecting cichlids, including Cibroheros robertsoni, in Mexico; Sp. istblenni, a parasite of Lutjanus kasmira from Hawaii; Sp. gobiomori, recorded from Eleotris sp. and Gobiomorus maculatus in Mexico; Pa. cyathopharynx, from Clarias gariepinus in South Africa; C. cotti, from China; Pr. pseudolaeviconchus, also parasitizing Clarias gariepinus in South Africa; and Sp. inopinatus, infecting Megaleporinus elongatus and Anostomoides passionis in Brazil (Figure 5).
To more accurately evaluate relationships within the subfamily Camallaninae, we conducted an analysis using representative COI sequences from this group. Three well-supported subclades were recovered: the Pa. cyathopharynx subclade, the C. cotti subclade and the Serpinema clade, comprising Se. yehi n. sp. and Se. cayennensis (Figure 6). In this dataset, Se. yehi n. sp. exhibited high genetic divergence relative to the remaining sequences, ranging from 7% compared with Se. cayennensis (MN104841) to 18% compared with C. xenopodis (MG948462) and Pa. cyathopharynx (MN523683, OP602980 and OP602981). Divergence from the outgroups reached 32% and 38% relative to Spiroxys ankarafantsika (MW545829) and Ancyracanthus electrophoris (OP765585), respectively. Genetic divergence values among all sequences are presented in Supplementary Material 3.
Maximum likelihood phylogenetic tree of species of the subfamily Camallaninae based on the COI gene, using Ancyracanthus electrophoris and Spiroxys ankarafantsika as outgroups. The position of Serpinema yehi n. sp. is indicated in bold italics. GenBank accession numbers are provided after each taxon. Node support values are shown above or below branches; posterior probabilities <0.90 and bootstrap values <70 are either not shown or indicated by a dash. The branch-length scale bar represents the number of substitutions per site.
Figure 6 Long description
The diagram is a maximum likelihood phylogenetic tree of species within the subfamily Camallaninae, based on the COI gene. The tree uses Spiroxys ankarafantsika and Ancyracanthus electrophoris as outgroups. The main branch labeled 'Camallaninae' splits into several sub-branches representing different species, each with GenBank accession numbers. Camallanus kaapstaadi MG948461, Zeylanema sp. LC815080, and Camallanus xenopodis MG948462 form one clade. Another clade includes Paracamallanus cyathopharynx with accession numbers OP602981, OP602980, and MN523683. Serpinema cayennensis MN104841 and Serpinema yehi n. sp. PZ113265 form a separate clade. Camallanus cotti EU598890 and EU598886, along with Camallanus hypophthalmichthys EU598816, are grouped together. Node support values are shown above or below branches, with posterior probabilities and bootstrap values indicated. The branch-length scale bar represents the number of substitutions per site, marked as 0.2. The position of Serpinema yehi n. sp. is highlighted in bold italics.
Remarks
The specimens were assigned to Serpinema based on the presence of a buccal capsule composed of two lateral valves with internal longitudinal ridges arranged into two groups, dorsal and ventral, separated by a distinct gap. These features correspond to the diagnostic characters of the genus as defined by Yeh (Reference Yeh1960) and Silva et al. (Reference Silva, Santos, Trindade and Melo2026).
Following Silva et al. (Reference Silva, Jesus, Santos, Nandyara, Willkens, Santos and Melo2023), Serpinema spp. can be divided into two morphogroups based on the number of spicules. Morphogroup I includes species with a single spicule (Se. lissemysus, Se. magathi, Se. monospiculatum and Se. pelliculatum), whereas morphogroup II comprises species with two unequal spicules (Se. amazonicum, Se. cayennensis, Se. emydidium, Se. intermedius, Se. kachugae, Se. microcephalus, Se. trispinosum and Se. octorugatum). Thus, Se. yehi n. sp. belongs to morphogroup II. Accordingly, the new species was compared with other congeners assigned to this morphogroup, regardless of geographic distribution.
Based on the original description of Se. amazonicum by Ribeiro (Reference Ribeiro1941) of specimens collected from Podocnemis expansa (Schweigger, 1812) (Podocnemididae) in Brazil, Se. yehi n. sp differs by having fewer buccal capsule ridges (11–18 in males, 10–17 in females in Se. yehi n. sp. vs. 17–20 in both sexes) and a longer tail in both sexes (157–227 and 237–544 vs. 130 and 210 for males and females, respectively). Males of Se. yehi n. sp. further differ by having longer spicules (800–891 and 373–453 vs. 570 and 240 for right and left spicules, respectively), and a different number and arrangement of pairs of caudal papillae (7:2:3 vs. 7:0:2). Females can additionally be distinguished by the morphology of vulvar lips; in the new species, the anterior lip is well developed, and the posterior vulvar lip is indistinct (vs. either a prominent anterior or posterior lip in Se. amazonicum), as well as by the presence of 4 mucrons at the tail tip (vs. 3 in Se. amazonicum).
Serpinema yehi n. sp. can be distinguished from Se. cayennensis, a parasite of Rhinoclemmys punctularia (same type host for both species) (Geoemydidae) from French Guiana (Harnoster et al., Reference Harnoster, Svittin and Du Preez2019), by the greater total length (800–891 vs. 360–806) and distinct morphology of the right spicule (leaf-like posterior end vs. bearing a short process near the distal extremity), as well as by the number of caudal papillae (7:2:3 vs. 6:2:5). Females also differ by the number of the mucrons at tail tip (4 vs. 3).
Based on the description of Se. emydidium by Mascarenhas and Müller (Reference Mascarenhas and Müller2017) from Trachemys dorbigni (Duméril and Bibron, 1835) (Emydidae) in Brazil, the new species differs by the number of caudal papillae (7:2:3 vs. 7:2:4), right spicule morphology (flattened extremity vs. simple distal extremity) and the number of mucrons at the female tail tip (4 vs. 3). Although specimens of Se. emydidium are generally smaller, they may reach body lengths comparable to those of the new species (9.72–11.73 and 15.24–18.00 mm in Se. yehi n. sp. vs. 6.55–11.4 and 11.28–18.71 mm for males and females, respectively) and exhibit a slightly longer right spicule (800–950 vs. 800–891 in the new species).
Serpinema yehi n. sp. can be easily distinguished from Se. intermedius, a parasite of Mauremys reevesii (Gray, 1831) (Geoemydidae) from China (Ivashkin et al., Reference Ivashkin, Soboloev and Khromova1977), by the number and arrangement of caudal papillae (7:2:3 vs. 7:2:5) and morphology of the right spicule. In Se. intermedius, the right spicule ends in a bifid tip, whereas in the new species terminate in leaf-like extremity, with one tip. Furthermore, females differ by the presence of phasmids in Se. yehi n. sp. (vs. not observed) and by vulva morphology, with a well-developed anterior lip and indistinct posterior lip in the new species, whereas both lips are indistinct in Se. intermedius.
Based on the original description of specimens collected from Pangshura smithii (Gray, 1863) (Geoemydidae) in India (Baylis and Daubney, Reference Baylis and Daubney1922), Se. kachugae can be distinguished from Se. yehi n. sp. by the smaller number of buccal capsule ridges in both sexes (8–10 in both sexes in Se. kachugae vs. 11–18 in males and 10–17 in females), a greater number of caudal papillae pairs (7:2:6 vs. 7:2:3 in the new species), and a longer right spicule (970 vs. 800–891 in Se. yehi n. sp.). Females of the new species also differ in their smaller body length (15.24–18.00 mm vs. 20.8–22.0 mm in Se. kachugae), vulva lip morphology, characterized by a well-developed anterior lip and indistinct posterior lip (vs. prominent vulvar lips, with the anterior considerably larger and overlapping the posterior) and by the presence of 4 mucrons at the tail tip (vs. 2).
Compared with Se. microcephalusfrom the original description by Dujardin (Reference Dujardin1845) of specimens collected from Emys orbicularis (Linnaeus, 1758) (Emydidae) in France, Serpinema yehi n. sp. differs by having a higher number of buccal capsule ridges (11–18 in males and 10–17 in females vs. 10 in both sexes in Se. microcephalus) and a different number and arrangement of caudal papillae pairs (7:2:3 in Se. yehi n. sp. vs. 7:0:5). In addition, the new species exhibits larger females (15.24–18.00 vs. 11.68–12.73 mm) and a longer tail in both sexes (157–227 and 237–544 in the new species vs. 62 and 172 for males and females, respectively).
Based on the redescription by Sharma et al. (Reference Sharma, Rigby, Sumita, Sanira, Vidyadaran, Jasni and Dailey2002) of specimens collected from Cuora amboinensis (Daudin, 1801) (Geoemydidae) in Malaysia, Se. octorugatum differs from the new species by the arrangement of caudal papillae (7:2:3 in Se. yehi n. sp. vs. 7:2:5) and right spicule morphology. In Se. octorugatum, the right spicule is U-shaped in cross-section, ventrally flexed, bearing a deep ventral groove and a complex lamella-like formation near the distal end, followed by abrupt flattening. In contrast, that of Se. yehi n. sp. is simple, with a leaf-like posterior end. Females differ by the number of mucrons at the tail tip (4 vs. 3 in Se. octorugatum).
Serpinema yehi n. sp. further differs from Se. trispinosum, described by Moravec and Vargas-Vázquez (Reference Moravec and Vargas-Vázquez1998) based on specimens collected from Trachemys scripta (Thunberg, 1792) (Emydidae) in Mexico, by the number and arrangement of caudal papillae (7:2:3 in the new species vs. 7:0:6) and by its larger body size in both males (9.72–11.73 mm vs. 4.43–7.53 mm in Se. trispinosum) and females (15.24–18.00 mm vs. 7.47–10.35 mm). Females also differ in the number of mucrons at the tail tip (4 in the new species vs. 3 in Se. trispinosum).
Although the new species belongs to morphogroup II (with two spicules), we also compared it with three species assigned to morphogroup I (Se. magathi, Se. monospiculatum and Se. pelliculatum). We included these species in our analysis because they also parasitize freshwater turtles in Brazil, mainly Kinosternon scorpioides (Linnaeus, 1766) (Izidro de Brito and Lacerda, Reference Izidro de Brito and Lacerda2025).
Thus, Se. yehi n. sp. can be distinguished from Se. magathi, reported from Kinosternon scorpioides (Kinosternidae) in the state of Pernambuco, northeastern Brazil (Freitas and Dobbin Jr, Reference Freitas and Dobbin1971), by spicule morphology (leaf-like posterior end in Se. yehi n. sp. vs. bifid distal tip bearing unequal branches), number and arrangement of caudal papillae (7:2:3 in new species vs. 7:0:7), vulvar lip morphology (well-developed anterior lip and indistinct posterior lip in new species vs. 2 prominent lips) and number of mucrons on female tail (4 vs. 3 in Se. magathi).
The new species also differs from Se. monospiculatum, described from specimens obtainded from Mesoclemmys tuberculata (Luederwaldt, 1926) (Chelidae) in the state of Pernambuco, northeastern Brazil (Freitas and Dobbin Jr, Reference Freitas and Dobbin1971), by spicule morphology (leaf-like posterior extremity in Se. yehi n. sp. vs. a filiform tip after curvature), number and arrangement of caudal papillae (7:2:3 in new species vs. 6:1:3), vulvar lip morphology (well-developed anterior lip and indistinct posterior lip vs. two non-prominent lips in Se. monospiculatum) and number of mucrons on female tail (4 vs. 3).
Serpinema yehi n. sp. can be distinguished from Se. pelliculatum, described from K. scorpioides (Kinosternidae) in Pará state, northern Brazil, Amazon region (Silva et al., Reference Silva, Jesus, Santos, Nandyara, Willkens, Santos and Melo2023), by spicule morphology (leaf-like posterior end, without ornamentations vs. covered by a sheath in Se. pelliculatum), number of caudal papillae (7:2:3 in Se. yehi n. sp. vs. 7:2:6), vulvar lip morphology (well-developed anterior lip and indistinct posterior lip vs. both lips prominent, with the posterior lip more developed than the anterior in Se. pelliculatum) and number of mucrons on female tail (4 in the new species vs. 3).
Thus, Se. yehi n. sp. can be distinguished from its congeners by the number of buccal capsule ridges, the number and arrangement of caudal papillae, the morphology and length of the spicules, the morphology of the vulvar lip and the number of mucrons on the female tail. Additional morphometric differences between Se. yehi n. sp. and its congeners are provided in Table 2.
Host species, country of record and main morphological and morphometric characteristics of the 13 valid species of Serpinema, organized by morphogroup
Table 2 Long description
Thirteen Serpinema nematode species are grouped into two morphogroups and compared by host turtle, country of record, literature reference, and key morphology and measurements for males and females. For each species, the table lists ridge counts, caudal papillae pattern, total body length, tail length, spicule length (single spicule in morphogroup I; right and left spicules in morphogroup II), vulva position, vulvar lip shape, phasmid presence, and number of mucrons on the female tail. Morphogroup I includes four species from India and Brazil, mostly with about nine to sixteen ridges and female mucrons typically three when reported; one species lacks spicule and vulva values in the source. Morphogroup II includes nine species spanning Brazil, French Guiana, China, India, France, Malaysia, and Mexico, with ridge counts ranging from about eight up to the low twenties depending on species. Across morphogroup II, right spicules are consistently longer than left spicules, and several species report phasmids present in males and females while others have no report. Body size varies widely: some species have males under eight millimeters while others exceed ten millimeters, and females commonly exceed males in total length. Vulvar lip descriptions differ by species, ranging from indistinct lips to prominent anterior lips that may overlap the posterior lip, which can help separate otherwise similar taxa. Values are compiled from multiple references and some cells are missing or marked as not reported, so comparisons should be interpreted as literature-based rather than standardized measurements.
Notes: aBased on Ivashkin et al. (Reference Ivashkin, Soboloev and Khromova1977).
Discussion
Following the classification proposed by Silva et al. (Reference Silva, Santos, Trindade and Melo2026), the specimens examined in the present study were assigned to Serpinema. The unique morphological features of Serpinema yehi n. sp., together with the molecular evidence, support its recognition as a new species and further corroborate the validity of Serpinema as a distinct genus.
Our phylogenetic analyses revealed that Se. yehi n. sp. forms a well-supported clade with Se. cayennensis described by Harnoster et al. (Reference Harnoster, Svittin and Du Preez2019). Notably, both species parasitize Rhynoclemmys punctularia in the Neotropical Region. However, the limited availability of COI sequences for Serpinema spp. prevents a more robust assessment of this relationship. Thus, it remains unclear whether this clustering reflects genuine phylogenetic affinity at the genus level or is additionally influenced by shared host association and common occurrence within the Neotropical biogeographical region.
Phylogenetic analyses based on the 18S marker recovered well-supported clades composed of individual species. This pattern aligns with the typically low genetic divergence observed among conspecific individuals, even across populations separated by broad geographic distances. Notably, Sp. inopinatus formed a well-supported subclade containing three sequences from different regions: two derived from specimens parasitizing freshwater fishes in Brazil and one from a marine fish from Egypt.
However, this subclade nested within a broader, well-supported clade comprising Sp. hilarii and Sp. pintoi, both of which parasitize South American freshwater fishes. This grouping may reflect the influence of host habitat and biogeographical distribution on parasite phylogeny, as suggested by Ailán-Choke and Pereira (Reference Ailán-Choke and Pereira2021).
The well-supported clade containing Pr. laeviconchus and Pr. pseudolaeviconchus, both from the African continent, further supports the hypothesis that host geographical distribution may influence parasite phylogeny. Nofal et al. (Reference Nofal, Santos, Jirsa and Avenant-oldewage2022) proposed that sequence KP274849 (attributed to Pr. laeviconchus) actually represents Pr. pseudolaeviconchus, based on its substantial genetic divergence from sequence JF803934 (3% genetic divergence in our study), the only other sequence available at the time for Pr. laeviconchus. However, KP274849 (Pr. laeviconchus) also exhibited marked divergence from sequences attributed to Pr. pseudolaeviconchus in their analyses (5% genetic divergence in our study), leading the authors to conclude that the relationship between these two nominal species requires further investigation.
In our phylogenetic reconstruction, each nominal species formed distinct subclades. However, the three sequences assigned to Pr. laeviconchus (KP274849, PP581381 and JF803934) showed high intraspecific genetic divergence (2% between KP274849 and PP581381; 3% between KP274849 and JF803934; and 5% between PP581381 and JF803934). Our findings suggest that these sequences may not represent a single taxon, or alternatively, that Pr. laeviconchus constitutes a cryptic species, distinguishable only through molecular approaches.
Denticamallanus rarus (DQ494195) and Pr. pacificus (DQ442665) clustered in a well-supported clade with 0% genetic divergence, suggesting that they represent the same taxon. Conversely, the two currently available sequences for D. rarus (DQ494195 and JF803912) showed substantial genetic divergence (6%) and did not cluster together, indicating a probable case of morphological misidentification, as previously suggested by Nofal et al. (Reference Nofal, Santos, Jirsa and Avenant-oldewage2022).
This discrepancy may result from errors during sequence submission or from experimental contamination, particularly because sequences DQ494195 and DQ442665 originated from the same study (Wijová et al., Reference Wijová, Moravec, Horák and Lukes2006). We conclude that sequence DQ494195, originally attributed to D. rarus, corresponds to the species Pr. pacificus.
The remaining sequence of D. rarus (JF803912) clustered in a well-supported clade with Sp. huacraensis. Notably, both taxa share the presence of tooth-like structures in the buccal capsule (Ailán-Choke et al., Reference Ailán-Choke, Davies, Tavares and Pereira2019; Ailán-Choke and Pereira, Reference Ailán-Choke and Pereira2021). Based on this shared morphology and the monophyletic clade recovered both here and by Ailán-Choke and Pereira (Reference Ailán-Choke and Pereira2021), this provides additional evidence to the conclusions made by Silva et al. (Reference Silva, Santos, Trindade and Melo2026), who proposed that Sp. huacraensis should be reassigned to Denticamallanus. Our findings also support the recognition of Denticamallanus as a valid and distinct genus. We emphasize that tooth-like structures represent a reliable morphological synapomorphy for the group.
In the 18S phylogenetic reconstruction, Se. yehi n. sp. did not cluster with any other available sequence. Instead, it formed a well-supported clade with C. pallidipectoris (1% of genetic divergence) and Zeylanema sp. (1% of genetic divergence). The proximity between the new species and C. pallidipectoris highlights the influence of host phylogeny in the evolution of these nematodes, as both species parasitize freshwater turtles.
Furthermore, C. pallidipectoris did not group with other Camallanus species. This suggests that chelonian-parasitizing Camallanus are more closely related to Serpinema than among Camallanus spp. infecting other host groups. Although Zeylanema sp. parasitizes a fish host and resembles Camallanus morphologically, its sequence did not group with fish-parasitizing Camallanus species.
Based on 28S ribosomal DNA, Saito et al. (Reference Saito, Hayashi, Ono, Okada and Takashima2025) also recovered a clade comprising Camallanus, Serpinema and Zeylanema, leading them to propose that Serpinema and Zeylanema might be treated as junior synonyms or subgenera of Camallanus. However, Ailán-Choke and Pereira (Reference Ailán-Choke and Pereira2021) and Silva et al. (Reference Silva, Santos, Trindade and Melo2026) rejected the use of subgeneric classification within Camallanidae. Therefore, a more robust assessment of the taxonomic status of Zeylanema will require the inclusion of additional sequences from its representatives.
Consistent with 18S-based analyses, the COI phylogeny recovered several well-supported single-species clades. Serpinema yehi n. sp. clustered with Se. cayennensis (6% of genetic divergence) in a strongly supported clade, which nested outside the subclade comprising Camallanus spp. Although Camallanus, Serpinema and Zeylanema formed a single major clade, with Camallanus and Serpinema forming distinct subclades, Zeylanema grouped with Serpinema.
This topology reinforces the close phylogenetic relationships among these three genera, suggesting relatively recent divergences. The observed evolutionary proximity aligns with their marked morphological similarity, as, according to Silva et al. (Reference Silva, Santos, Trindade and Melo2026), these taxa are distinguished by the morphology and arrangement of the internal ridges of the buccal capsule. Additionally, C. kaapstaadi and C. xenopodis, both parasites of Xenopus laevis from South Africa, formed a distinct subclade separate from other Camallanus species, further confirming that the genus Camallanus is not monophyletic.
Our results support the classification by Silva et al. (Reference Silva, Santos, Trindade and Melo2026), which rejects the subgeneric classification within Camallanidae and recognizes Camallanus, Serpinema and Zeylanema as distinct genera. However, a comprehensive understanding of the phylogenetic relationships among these three taxa will require additional molecular data, particularly from underrepresented taxa of Zeylanema and Serpinema.
Spirocamallanus xenopodis (MN523681) and Pr. slomei (MG948463), both parasitizing anurans of the genus Xenopus in South Africa and previously classified in Batrachocamallanus, clustered in a fully supported clade. Notably, the two COI sequences are nearly identical (99.85% identical) and show 0% genetic divergence, suggesting that they may represent the same taxon, as Nofal et al. (Reference Nofal, Santos, Jirsa and Avenant-oldewage2022) previously observed. However, Svitin et al. (Reference Svitin, Truter, Kudlai, Smit and Du Preez2019) reported a 2.9% divergence between Sp. xenopodis and Pr. slomei based on 28S sequences. Given this discrepancy, we agree with Nofal et al. (Reference Nofal, Santos, Jirsa and Avenant-oldewage2022) that a sequence submission error likely explains the absence of genetic divergence among COI sequences, rather than biological conspecificity.
Traditional taxonomy uses buccal capsule morphology as a key diagnostic character for camallanid subfamilies and genera; however, molecular evidence suggests that this trait does not necessarily correspond to monophyletic lineages (Ailán-Choke and Pereira, Reference Ailán-Choke and Pereira2021). Both our COI and 18S phylogenetic analyses clearly recovered this pattern. Consequently, the subdivision of Camallanidae into subfamilies based on buccal capsule morphology appears to be entirely artificial, as our results failed to recover exclusive clades for either Procamallaninae or Camallaninae.
Our analyses provide evidence for the paraphyly of Camallanidae, demonstrating that distinct buccal capsule morphologies arose independently throughout the family’s evolutionary history. Both the smooth internal-surface buccal capsule, typical of Procamallanus, and the spiral ridges, characteristic of Spirocamallanus, appear to have originated multiple times during camallanid diversification. These results align with 28S-based phylogenetic analyses by Svitin et al. (Reference Svitin, Truter, Kudlai, Smit and Du Preez2019), which recovered species with smooth internal capsules and spiral ridges within the same clade.
The lack of well-supported clades restricted to Procamallanus and Spirocamallanus demonstrates the limited taxonomic utility of buccal capsule internal ridges for generic delimitation within the subfamily Procamallaninae. While these characters remain useful for morphological identification and systematic arrangement, they do not accurately reflect the phylogenetic relationships or the evolutionary processes that occurred during the diversification of these nematodes. Nevertheless, for the practical purposes of systematic organization, we follow the classification proposed by Silva et al. (Reference Silva, Santos, Trindade and Melo2026).
Our phylogenetic study corroborates the three factors proposed by Ailán-Choke and Pereira (Reference Ailán-Choke and Pereira2021) as primary drivers of camallanid diversification and evolution: host phylogeny, habitat and biogeography. For instance, the clade comprising Sp. barlowi and Sp. rebecae, both parasites of freshwater fishes from the Nearctic and Neotropical regions (geographically close areas), illustrates this pattern. Similarly, in COI phylogeny, the clade formed by C. cotti and C. hypophthalmichthys, both originating from Asia, reinforces the roles of geographic distribution and habitat in structuring the phylogeny of Camallanidae.
However, this pattern was not consistent across all taxa. In our COI analyses, the Sp. inopinatus clade (Brazilian freshwater parasites) did not cluster with either Sp. barlowi/Sp. rebecae group or the Sp. neocaballeroi clade, despite their shared freshwater habitats and Nearctic/Neotropical distributions. This discrepancy suggests that, while biogeography and habitat are important, additional factors, such as host evolutionary history or species-specific ecological traits, also play a significant role in shaping the phylogeny of Camallanidae.
Interestingly, although C. kaapstaadi, C. xenopodis (Camallaninae), Pr. slomei and Sp. xenopodis (Procamallaninae) all parasitize South African amphibians of the genus Xenopus (Svitin et al., Reference Svitin, Schoeman and Du Preez2018, Reference Svitin, Truter, Kudlai, Smit and Du Preez2019), they did not group within the same subclade in our COI analyses. Instead, the two Camallanus species clustered together, whereas Pr. slomei and Sp. xenopodis formed a separate, distinct clade. This topology suggests a clear phylogenetic separation between the Camallaninae and Procamallaninae nematode lineages in this host group, despite their shared host and biogeographic affinities.
In the 18S phylogenetic tree, Sp. inopinatus, Sp. hilarii and Sp. pintoi formed a well-supported clade with parasites of freshwater fishes from the Neotropics. However, a single sequence of Sp. inopinatus from a marine host in Egypt clustered within this group, showing 0% genetic divergence from three sequences of Sp. inopinatus. This finding suggests that, although biogeography shapes camallanids evolution, conspecific sequences still tend to cluster together, regardless of whether hosts are from geographically distant regions or occupying distinct habitats.
The COI-based phylogenetic tree for the subfamily Camallaninae resulted in three well-supported clades: one corresponding to Pa. cyathopharynx (also recovered in the other two phylogenetic analyses); a second clade representing the genus Serpinema, including Se. yehi n. sp. and Se. cayennensis; and a third clade corresponding to C. cotti. However, this reconstruction placed Serpinema and Paracamallanus in the same subclade, and the C. cotti clade did not cluster with C. hypophthalmichthys, contradicting the topology of 18S and the broader COI tree. These inconsistencies further confirm that Camallanus is a non-monophyletic group. Additionally, the unstable position of this taxon reinforces the need for additional Zeylanema sequences to establish a robust taxonomic status of this genus.
Serpinema yehi n. sp. represents the eighth species of the genus recorded in Brazil and the second species described parasitizing the freshwater turtle R. punctularia, the spot-legged turtle. This turtle species, commonly known as ‘perema’, is distributed throughout the Brazilian Amazon and is semiaquatic and diurnal. It is omnivorous, feeding primarily on fruit but also consuming animal-derived items (Ferrara et al., Reference Ferrara, Fagundes, Morcatty and Vogt2017).
Silva et al. (Reference Silva, Jesus, Santos, Nandyara, Willkens, Santos and Melo2023) suggested that turtle feeding habits strongly influence exposure to different nematode larvae in aquatic environments, through the consumption of small prey, such as crustaceans, amphibians and fishes, which may harbour camallanid larvae. This dietary behaviour may contribute to the high diversity of Serpinema species reported from a single host, such as K. scorpioides (Mascarenhas and Müller, Reference Mascarenhas and Müller2021; Silva et al., Reference Silva, Jesus, Santos, Nandyara, Willkens, Santos and Melo2023; Conga et al., Reference Conga, Helen Costa, Bejarano Alegre, Magalhães Bezerra and Maciel de Castro Cardoso Jaques2026). In contrast, the more restricted diet of R. punctularia may be associated with the comparatively lower diversity of Serpinema spp. recorded from this host, particularly relative to other freshwater turtles.
Our results support the description of a new species of Serpinema based on both morphological and molecular evidence, as well as the recognition of Serpinema as a valid genus distinct from Camallanus. Additionally, they reinforce the polyphyly of Camallanus and the artificial nature of traditional subfamily divisions. Thus, the independent origin of internal buccal ridges limits their phylogenetic signal and delimitation of monophyletic lineages. However, despite the evolutionary homoplasy in buccal structures, these structures remain indispensable for practical systematic organization and morphological identification of Camallanidae.
Host phylogeny, habitat and biogeography significantly shape camallanid diversification and evolution. However, the genetic stability of Sp. inopinatus across different regions suggests that species identity remains conserved despite extreme geographic or environmental shifts. Additionally, the unstable position of certain clades and the identification of likely database errors underscore the urgent need for expanded molecular sampling, particularly for the genus Zeylanema, to establish robust evidence to understand the evolutionary processes for the family.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0031182026102509.
Acknowledgements
We are grateful to Dra. Edilene Oliveira da Silva, from the Laboratory of Cellular Structural Biology at the Federal University of Pará, Belém, for her technical support with the SEM analyses. This study is part of the PhD thesis of Leandro Mauricio Oliveira Silva in the Postgraduate Program in Biology of Infectious and Parasitic Agents at the Federal University of Pará (PPGBAIP/UFPA).
Author contributions
L.M.O.S.: Conceptualization, formal analysis, methodology, validation, visualization, and writing – original draft, review and editing. L.F.S.T.-C.: Formal analysis, methodology, and writing – review and editing. R.F.J.: Formal analysis, methodology, supervision, and writing – review and editing. M.L.: Formal analysis, methodology, and writing – review and editing. J.N.S.: Resources, funding acquisition, and writing – review and editing. F.T.V.M.: Conceptualization, formal analysis, funding acquisition methodology, project administration, validation, resources, supervision, and writing – review and editing. All authors have read and agreed to the manuscript that has been submitted.
Financial support
This study was supported by Coordination for the Improvement of Higher Education Personnel, Brazil (CAPES); and the National Council for Scientific and Technological Development (CNPq) Research productivity scholarship of CNPq to F.T.V.M. (Process number 310825/2025-3) and J.N.S. (Process number 310616/2025-5).
Competing interests
The authors declare that there are no conflicts of interest.
Ethical standards
All procedures contributing to this work comply with all applicable institutional, national, and international guidelines for animal care and use, and were approved by the Animal Research Ethics Committee of the Federal University of Pará (CEUA/UFPA), under license CEUA Nº 8241310322. The present study was authorized by Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio), Brazil, and host specimens were collected under license number SISBIO: 53527-4. This study was also registered in the National System for the Management of Genetic Heritage and Associated Traditional Knowledge (SisGen) under record number A8626D1, in compliance with Brazilian Law No. 13,123/2015.
Open access
