Introduction
The diversification of parasites is tightly linked to the ecology and dispersal of their hosts. Unlike free-living organisms, parasites with direct life cycles rely entirely on host dispersal and population connectivity to expand their ranges (Blasco-Costa and Poulin, Reference Blasco-Costa and Poulin2013). Viviparous monogeneans of the genus Gyrodactylus von Nordmann, 1832, exemplify this dependence; they reproduce rapidly, disperse strictly with their hosts and exhibit fine-scale morphological variability that makes them excellent models for studying host-driven adaptation and geographic structuring (Bakke et al. Reference Bakke, Cable and Harris2007).
Gyrodactylus is among the most species-rich monogenean lineages, with >400 described taxa infecting marine and freshwater fishes worldwide (Malmberg, Reference Malmberg1970; Harris et al. Reference Harris, Shinn, Cable and Bakke2004; Pugachev et al. Reference Pugachev, Gerasev, Gussev, Ergens and Khotenowsky2010). Their small size and morphological conservatism, however, have long complicated species identification, and numerous cryptic species complexes have been reported across diverse fish lineages (Malmberg, Reference Malmberg1970; Kuusela et al. Reference Kuusela, Ziȩtara and Lumme2008; Razo-Mendivil et al. Reference Razo-Mendivil, García-vásquez and Rubio-godoy2016; Ondračková et al. Reference Ondračková, Seifertová, Bryjová, Leis and Jurajda2020; Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2025). Taxonomy in Gyrodactylus traditionally relies on the haptor, the attachment organ, formed by sclerotized hamuli (anchors), bars and marginal hooks, which provides key diagnostic characters (Malmberg, Reference Malmberg1970; Pugachev et al. Reference Pugachev, Gerasev, Gussev, Ergens and Khotenowsky2010). However, intraspecific variability in these structures across hosts and regions can blur diagnostic boundaries, particularly when original descriptions were based on limited sample sizes or when host identifications were uncertain (Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2023b, Reference Rahmouni, Seifertová and Šimková2025; Benovics et al. Reference Benovics, Rahmouni, Rehulková, Nejat and Šimková2024). Consequently, integrating comparative morphometrics with molecular data has become essential for resolving taxonomic questions and evaluating patterns of host–parasite associations.
Haptoral sclerites are not merely taxonomic markers; they are functional structures shaped by ecological and evolutionary pressures. Variability in sclerite dimensions may reflect anchor efficiency in different gill microhabitats, host-related traits (e.g. body size, architecture of scales and gills) and local environmental conditions (e.g. temperature, salinity, flow regime), while still retaining a phylogenetic signal on broader scales (Olstad et al. Reference Olstad, Bachmann and Bakke2009; Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2023b). Comparative work in related monogeneans (e.g. Ligophorus) shows that dorsal and ventral anchors differ in movement ability and functional roles, with morphology reflecting a balance between shared evolutionary history and functional convergence (Rodríguez-González et al. Reference Rodríguez-González, Míguez-lozano, Llopis-belenguer and Balbuena2015). This nature of the haptor, as both an adaptive apparatus and a repository of phylogenetic information, provides a tractable system for testing how geography and hosts shape parasite phenotypes.
Because parasite morphology is molded by both host traits and historical biogeography (Rahmouni et al. Reference Rahmouni, Seifertová, Benovics and Šimková2023a, Reference Rahmouni, Seifertová and Šimková2025), studying Gyrodactylus across diverse hosts and regions can reveal how microevolutionary pressures and large-scale drainage histories interact. Morphological plasticity may reflect short-term ecological fitting, whereas conserved patterns indicate deeper phylogenetic and biogeographic signals. In North America, freshwater fish distributions have been affected by historical drainage connections and climatic oscillations, particularly Pleistocene glaciations, which have repeatedly fragmented and reconnected river systems (Mayden, Reference Mayden1988). These processes have influenced host dispersal and contact zones, thereby shaping opportunities for parasite colonization, host switching and diversification (Kearn, Reference Kearn1994; Ziętara and Lumme, Reference Ziętara and Lumme2002; Meinilä et al. Reference Meinilä, Kuusela, Ziȩtara and Lumme2004; Xavier et al. Reference Xavier, Faria, Paladini, van Oosterhout, Johnson and Cable2015; Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2023b).
Despite the ecological dominance of Leuciscidae in the region, the diversity and geographic variability of their Gyrodactylus parasites remain incompletely characterized, even though these same processes likely facilitated both host diversification and the co-diversification and spatial structuring of their parasites. Within the biogeographic context, Gyrodactylus atratuli Putz and Hoffman, Reference Putz and Hoffman1963, and Gyrodactylus mediotorus King, Marcogliese, Forest, McLaughlin and Bentzen, Reference King, Marcogliese, Forest, McLaughlin and Bentzen2013, are widespread ectoparasites of leuciscids in northern part of North America. The former species was originally described from Rhinichthys atratulus (Hermann, 1804) in West Virginia and has since been reported from several leuciscid hosts in the USA and Canada (Putz and Hoffman, Reference Putz and Hoffman1963; Hanek and Fernando, Reference Hanek and Fernando1971; Kirby, Reference Kirby1981; Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2023b). Gyrodactylus mediotorus was described from Notropis hudsonius (Clinton, 1824) in Quebec and has been documented on other Notropis and Cyprinella hosts across North America (King et al. Reference King, Marcogliese, Forest, McLaughlin and Bentzen2013; Leis et al. Reference Leis, Easy and Cone2021; Rahmouni et al. Reference Rahmouni, Seifertová, Bean and Šimková2024). While both species are well recorded in the northern part of North America, their occurrence and morphological variability in southern populations remain unknown. There are no previous records of these species from Mexico. The potential influences of host species, host size and latitude on haptoral sclerite dimensions have not been investigated. Understanding these patterns is critical for disentangling the relative roles of host-driven adaptation, phenotypic plasticity and historical biogeography in shaping Gyrodactylus morphology.
In this study, we collected Gyrodactylus specimens from diverse leuciscid hosts in Mexico and identified them, on the basis of morphological features and ITS sequences, as the 2 known species G. atratuli and G. mediotorus, representing their first records in the country. Given the limited sample sizes, we aimed to: (i) document these first records from Mexico, (ii) explore potential trends in intraspecific variability of attachment organ morphology across hosts and geographical distribution and (iii) examine whether host size or latitudinal gradients might be associated with the distributional patterns observed in G. atratuli and G. mediotorus. From a biogeographic perspective, these preliminary observations provide insights into the intraspecific variability of the attachment organ and its possible evolutionary context, offering a basis for future, more comprehensive studies on North American Gyrodactylus diversity, host specificity and factors shaping attachment organs.
Material and methods
Fish collection and identification
Fish hosts were collected in Mexico in May 2023; 3 specimens of Cyprinella lutrensis (Baird and Girard, 1853) and 2 specimens of Cyprinella cf. proserpina (Girard, 1856) from Las Lajas, Candela, Coahuila (26°47′33″N, 100°41′34″W), and 7 specimens of Rhinichthys cataractae (Valenciennes, 1842) from Arroyo de Covadonga, Peñón Blanco, Durango (24°50′48″N, 104°02′27″W). Fish identification was performed by local collaborators (listed in the Acknowledgements section), with molecular characterization performed via cyt-b sequencing following the protocols of Rahmouni et al. (Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2023b, Reference Rahmouni, Seifertová, Bean and Šimková2024). The resulting sequences were deposited in the GenBank database under the accession numbers PZ784772 and PZ784773. In the present study, the fish host nomenclature follows FishBase (Froese and Pauly, Reference Froese and Pauly2026).
Parasite screening, mounting and morphogenetic identification
Monogeneans were screened from gills under a stereomicroscope, removed and mounted on slides with a mixture of glycerine and ammonium picrate (GAP) (Malmberg, Reference Malmberg1957). Haptoral sclerites were examined and measured under phase contrast at ×1000 magnification, following known terminology and protocols (Malmberg, Reference Malmberg1957; Pugachev et al. Reference Pugachev, Gerasev, Gussev, Ergens and Khotenowsky2010; Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2023b, Reference Rahmouni, Seifertová, Bean and Šimková2024). Photographs were taken using an Olympus BX51 phase-contrast microscope and processed with Olympus Stream Image Analysis v. 1.9.3 software. Species identification was based on comparisons with original descriptions (see results section). Measurements (in μm) are presented as ranges to facilitate comparison with previous studies, which often reported only a limited set of dimensions for certain haptoral elements, such as bars and marginal hooks. Not all haptoral structural components were included in the comparison, as meristic data were unavailable for all elements. Infection indices followed Bush et al. (Reference Bush, Lafferty, Lotz and Shostak1997). Voucher specimens of identified Gyrodactylus species were deposited in the Helminthological collection of the Institute of Parasitology, Biology Centre of Academy of Sciences of the Czech Republic, České Budějovice (IPCAS), under the accession numbers IPCAS M-794 and IPCAS M-852. For molecular confirmation, genomic DNA of Gyrodactylus specimens was extracted from ethanol-preserved samples with the DNeasy® kit (Qiagen). The ITS1–5.8S–ITS2 region was amplified and sequenced following Rahmouni et al. (Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2023b, Reference Rahmouni, Seifertová, Bean and Šimková2024). Sequences of gyrodactylids and their corresponding hosts were aligned in MEGA v. 11 (Tamura et al. Reference Tamura, Stecher and Kumar2021) and compared withnBLAST (accessed 20 September 2025) to confirm species identity. Genetic divergences among parasite populations were estimated as uncorrected p-distances in MEGA v. 11 (Tamura et al. Reference Tamura, Stecher and Kumar2021). The newly obtained ITS sequences were deposited in the GenBank database under the accession numbers PZ770417 and PZ770418.
Results
Gyrodactylus atratuli Putz and Hoffman, Reference Putz and Hoffman1963 (Figure 1A and B, Figure 2A)
Genetic identification
Three gyrodactylid specimens collected from R. cataractae in Mexico yielded identical ITS sequences (raw size: 898–908 bp). nBLAST search identified G. atratuli (OR270008) from Canadian R. cataractae (Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2023b) as the closest match (99% similarity, 99% coverage). Pairwise distance analysis of 868 bp revealed a divergence of 0.7% (6 bp difference) between Mexican specimens and Canadian G. atratuli, remaining within the accepted intra-specific range for Gyrodactylus spp. (Ziȩtara and Lumme, Reference Ziȩtara and Lumme2003; Huyse et al. Reference Huyse, Malmberg and Volckaert2004; Rahmouni et al. Reference Rahmouni, Seifertová, Benovics and Šimková2023a, Reference Rahmouni, Seifertová and Šimková2023b, Reference Rahmouni, Seifertová, Bean and Šimková2024, Reference Rahmouni, Seifertová and Šimková2025).
Taxonomic summary
Type-host and locality: Rhinichthis atratulus (Hermann, 1804) (Leuscicidae), South branch of the Leetown Run, West Virginia, USA (Putz and Hoffman, Reference Putz and Hoffman1963).
Previous records: Margariscus margarita (Cope, 1867) (Leusciscidae), South branch of the Leetown Run, West Virginia, USA (Putz and Hoffman, Reference Putz and Hoffman1963); R. atratulus and Rhinichthis cataractae (Valenciennes, 1842) (Leusciscidae), Shelter Valley stream, Canada (Hanek and Fernando, Reference Hanek and Fernando1971); Cyprinella spiloptera (Cope, 1867) (Leusciscidae), Stehman’s Run and the west branch of Little Conestoga Creek, Pennsylvania, USA (Kirby, Reference Kirby1981); and R. cataractae, Leatherstocking Creek, Otsego, New York, USA, Cap-Rouge River, Quebec, Canada, and R. atratulus (Hermann, 1804), Morrys Creek, Door County, Wisconsin, USA (Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2023b).
Present record: on R. cataractae, Arroyo de Covadonga (24°50′48″N, 104°02′27″W), Peñón Blanco, Durango, Mexico.
Site of infection: fins.
Parasitological indices: 9 of 11 hosts infected, prevalence = 82%, mean infection intensity = 0.8.
Voucher specimens: IPCAS M-852.
Host GenBank accession number: PZ784772.
Parasite GenBank accession number: PZ770418.
Morphological characterization
Based on morphology, 9 gyrodactylid specimens were determined as G. atratuli (Figure 1A and B, Figure 2A). Taken from flattened gyrodactylid specimens, measurements of the haptoral parts of G. atratuli collected from Mexican R. cataractae specimens, as well as from all other hosts known to harbour this species to date, are presented in Table 1. Hamuli with bases slightly curved inward, often with folds, shafts bowed tapering towards elongate, curved points. Ventral bar broad with prominent blunt lateral processes extending out of bar, membrane trapezoid, rounded posteriorly with longitudinal striations. Dorsal bar straight, constricted at midpoint, with projections near each end, attenuated ends inserted into terminal plates. Marginal hooks possess well-developed, rounded heel, short stubby toe and inconspicuous shelf. Relatively short sickle proper with robust blade curving to sharp point; filament loop U-shaped, extending to about one-third of handle length. The male copulatory organ was not found.
Light photomicrograph of haptoral sclerotized structures of Gyrodactylus atratuli Putz and Hoffman, Reference Putz and Hoffman1963 ex. Rhinichthys cataractae (Valenciennes, 1842) (A, B) and Gyrodactylus mediotorus King, Marcogliese, Forest, McLaughlin and Bentzen, Reference King, Marcogliese, Forest, McLaughlin and Bentzen2013 ex. Cyprinella lutrensis (Baird and Girard, 1853) (C, D) and ex. (E, F) Cyprinella cf. Proserpina (Girard, 1856) from Mexico.
Drawing of haptoral sclerotized structures of (A) Gyrodactylus atratuli Putz and Hoffman, Reference Putz and Hoffman1963 ex. Rhinichthys cataractae (Valenciennes, 1842) and (B) Gyrodactylus mediotorus King, Marcogliese, Forest, McLaughlin and Bentzen, Reference King, Marcogliese, Forest, McLaughlin and Bentzen2013 ex. Cyprinella lutrensis (Baird and Girard, 1853) from Mexico.
Morphometric comparisons of Gyrodactylus atratuli Putz and Hoffman, 1963 and Gyrodactylus mediotorus King, Marcogliese, Forest, McLaughlin and Bentzen, Reference King, Marcogliese, Forest, McLaughlin and Bentzen2013 from various leuciscid hosts and geographic localities in North America
Table 1 Long description
Morphometric ranges in micrometers are listed for haptoral structures labeled H, VB, DB, and MH for Gyrodactylus atratuli and Gyrodactylus mediotorus, alongside leuciscid fish hosts, locality of record, and literature. Gyrodactylus atratuli is reported from Rhinichthys atratulus and Margariscus margarita in the South branch of Leetown Run, West Virginia, with H total length 59 to 63, VB total length 26 to 29, DB total length 17 to 19, and MH total length 28 to 31. Gyrodactylus mediotorus is reported from multiple hosts and sites, including Rhinichthys atratulus at Shelter Valley Stream, Canada with H total length 66 to 68 and MH total length 35 to 36, and Rhinichthys cataractae at Shelter Valley Stream with H total length 58 to 61 and MH total length 29 to 30. Additional G. mediotorus records include mixed Rhinichthys atratulus and Rhinichthys cataractae localities in New York, Wisconsin, and Quebec with H total length 58.2 to 64.4 and MH total length 28.8 to 35.5, and Rhinichthys cataractae in Durango, Mexico with H total length 54.1 to 57.5 and MH total length 27.3 to 30.3. For G. mediotorus on Notropis hudsonius in the St. Lawrence River, Quebec, VB total length is 36.4 to 41.3 and MH total length is 35 to 37.8; on Notropis texanus in the Upper Mississippi River, VB total length is 20 to 22.7 and MH total length is 33.9 to 38.7; and on Notropis cf. stramineus in the Guadalupe River, Texas, VB total length is 30.7 to 36.9 and MH total length is 31 to 38.8. For G. mediotorus on Cyprinella lutrensis and Cyprinella cf. proserpina at Las Lajas in Coahuila, Mexico, H total length ranges are 58.4 to 58.7 and 54 to 55.7, and MH total length ranges are 33.6 to 35.3 and 30.9 to 32.6. Some cells use a dash to indicate no value reported, including H inner root length for G. atratuli and for G. mediotorus on Notropis texanus, and DB median width for several G. mediotorus columns.
Type hosts and type localities for both species are highlighted in bold.
Morphometric values for key haptoral structures are presented as ranges (µm). For G. mediotorus, the separate subpopulation ranges originally reported by Leis et al. (Reference Leis, Easy and Cone2021) were consolidated into a single overall range per structure to facilitate regional comparisons. Host species, locality records and corresponding literature sources are provided.
H, hamulus; VB, ventral bar; DB, dorsal bar; MH, marginal hook.
Remarks
Morphometric comparison of the haptoral sclerites of G. atratuli (Table 1) revealed distinct patterns of variation correlated with both geographic distribution (north–south gradient) and fish host species (R. atratulus vs R. cataractae). The anchoring apparatus was consistently larger in northern populations and when parasitizing R. atratulus. Hamuli total length was greatest in Canadian specimens from R. atratulus (66–68 µm), followed by intermediate lengths in US and Canada populations (59–63 µm from West Virginia; 58.2–64.4 µm from New York, Wisconsin and Quebec). Conversely, hamuli were shortest in the southern Mexican population from R. cataractae (54.1–57.5 µm). Notably, even within the same northern locality (Shelter Valley Stream, Canada), hamuli were longer in specimens from R. atratulus (66–68 µm) than in those from R. cataractae (58–61 µm). A similar dual trend was observed for the hamulus inner root, which reached maximum length in Canadian populations parasitizing R. atratulus (21–22 µm) and dropped to 16–18 µm on Canadian R. cataractae, while ranging broadly across other northern localities (14.2–21.4 µm) before reaching its minimum dimensions in Mexico (14–17.7 µm). Hamulus point length followed this geographic decline, measuring longest in northern populations (Canada from R. atratulus: 29–31 µm; USA: 27–29 µm) and shortest in the Mexican specimens (23.2–26.5 µm). The ventral bar dimensions exhibited varying degrees of differentiation. Ventral bar total length remained relatively uniform across most populations, generally ranging from 26 to 30 µm, though a broader range was observed in the combined northern datasets (23.1–29.9 µm). However, the ventral bar membrane length showed a clearer geographic and host split. It was the longest in the US specimens from West Virginia (23–27 µm) and in Canadian specimens from R. atratulus (22–24 µm) but dropped sharply in specimens from R. cataractae in both Canada (14–16 µm) and Mexico (13.7–15.6 µm). For the ventral bar lateral processes, measurements were longer and highly consistent between the US (23–27 µm) and Mexican specimens (23.2–26.5 µm), whereas both Canadian populations displayed noticeably shorter processes (11–14 µm). The dorsal bar total length exhibited substantial overlap across all studied regions, showing no clear geographic pattern (Canada: 19–22 µm; USA: 17–27.5 µm; Mexico: 19.1–22 µm). Marginal hook dimensions strongly reinforced north–south and host-dependent variation. Total marginal hook length was greatest in Canadian specimens from R. atratulus (35–36 µm), intermediate in USA and other northern localities (28–35.5 µm) and shortest in the Mexican population (27.3–30.3 µm). Similarly, the marginal hook sickle proper was slightly larger in the Canadian R. atratulus population (5–6 µm) compared to the Canadian R. cataractae (4–5 µm) and Mexican (4.2–4.8 µm) populations. The shape of haptoral sclerites was not considered for comparison because previous data, particularly from the earliest publications, lacked detailed or morphometrically accurate drawings, rendering reliable visual comparisons unfeasible.
Gyrodactylus mediotorus King, Marcogliese, Forest, McLaughlin and Bentzen, 2013 (Figure 1C–F, Figure 2B)
Genetic identification
Two gyrodactylid specimens collected from C. cf. proserpina had identical ITS sequences (raw size: 1045 bp). The sequencing of gyrodactylid specimens from C. lutrensis was unsuccessful. nBLAST search showed that our ITS sequences were identical (100% similarity, 100% coverage) to G. mediotorus (PP309999) from Cyprinella venusta Girard, 1856 and Notropis cf. stramineus (Cope, 1865) from Texas, USA (Rahmouni et al. Reference Rahmouni, Seifertová, Bean and Šimková2024). The closest additional hits included G. mediotorus (MW666182) from Notropis texanus (Girard, 1856) in Wisconsin, USA (Leis et al. Reference Leis, Easy and Cone2021) and G. mediotorus (KF178301) from the type-host N. hudsonius (Clinton, 1824) in Quebec, Canada (King et al. Reference King, Marcogliese, Forest, McLaughlin and Bentzen2013), showing 99.2% and 98.7% similarity, respectively (100% coverage in both cases). Pairwise distance analysis based on 963 bp of the newly obtained sequences from Mexican specimens and published sequences of G. mediotorus revealed divergences ranging from 0.2% with Texas specimens to 0.8% with the Canadian (type-locality) specimens. These values fall within the threshold typically considered for species discrimination in Gyrodactylus (Ziętara and Lumme, Reference Ziȩtara and Lumme2003; Huyse et al. Reference Huyse, Malmberg and Volckaert2004; Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2023b), supporting the identification of the Mexican specimens as G. mediotorus.
Taxonomic summary
Type-host and locality: N. hudsonius, Îles de la Paix, St. Louis Lake, Quebec, Canada (King et al. Reference King, Marcogliese, Forest, McLaughlin and Bentzen2013).
Previous records: N. hudsonius, Îles Vert, St. Lawrence Lake, Quebec, Canada (King et al. Reference King, Marcogliese, Forest, McLaughlin and Bentzen2013); N. texanus, Lake Onalaska, Upper Mississippi River, Wisconsin, USA (Leis et al. Reference Leis, Easy and Cone2021); and C. venusta, West Mud Creek, Neches River, and N. cf. stramineus, Guadalupe River, both in Texas, USA (Rahmouni et al. Reference Rahmouni, Seifertová, Bean and Šimková2024).
Present record: C. lutrensis and C. cf. proserpina, Las Lajas (26°47′33″N, 100°41′34″W), Candela, Coahuila, Mexico.
Site of infection: fins.
Parasitological indices: all examined hosts infected, prevalence = 100%, infection intensity = 1.5.
Voucher specimens: IPCAS M-794.
Host GenBank accession number: PZ784773.
Parasite GenBank accession number: PZ770417.
Morphological characterization
A total of 5 Gyrodactylus specimens were collected and attributed to G. mediotorus (Figure 1C–F, Figure 2B) based on morphological characters of the haptoral sclerites. These included 2 specimens from C. lutrensis (Figure 1C and D) and 3 from C. cf. proserpina (Figure 1E and F). Measurements (Table 1) were taken from 3 and 2 flattened gyrodactylid specimens found on C. cf. proserpina and C. lutrensis, respectively. Hamuli with bases slightly curved outward, often with folds, shafts bowed and tapering towards elongate, curved points. Ventral bar broad with prominent blunt lateral processes extending out of bar, median part with prominent knob, membrane rectangular with fine longitudinal striations, extending almost 1/2 of length of hamuli shaft. Dorsal bar straight, swollen at midpoint, with projections near each end, attenuated ends inserted into terminal plates. Marginal hooks possess downward globose heel, prominent triangular, straightforward toe and conspicuous shelf, sickle proper straight with robust short shaft curving to sharp point, filament loop U-shaped, extending to about one-third of handle length, handle ending in prominent filament. Male copulatory organ not found.
Remarks
Morphometric evaluation of the haptoral sclerites of G. mediotorus (Table 1) demonstrated notable variations linked to both latitude (north–south axis) and specific leuciscid species associations (Notropis vs Cyprinella spp.). A clinal reduction in the primary anchoring hooks was evident from North to South. The maximum values for hamulus total length were restricted to Canadian individuals collected from N. hudsonius (65.7–69.7 µm). Mid-latitude populations from USA exhibited intermediate, highly overlapping ranges, spanning 55.3–62.5 µm in the Upper Mississippi River (ex N. texanus) and 55.9–62.6 µm in the Guadalupe River (ex N. cf. stramineus). The lowest limits for this metric were recorded in Mexico, where a clear host-induced split occurred within the same locality, according to which specimens from C. lutrensis maintained slightly larger anchors (58.4–58.7 µm) than those infesting C. cf. proserpina, which possessed the smallest hamuli overall (54–55.7 µm). This geographic and host-dependent variation was mirrored in the deeper structural elements of the hamuli. The length of the hamulus shaft contracted significantly outside of Canada (46.6–49.4 µm), dropping to 34–37.4 µm in the Upper Mississippi River and settling between 37.3 and 41.3 µm across both Mexican cohorts. Hamulus point length tracked this downward latitudinal trajectory, decreasing from a peak of 25.4–28.4 µm in Canada to compressed ranges in Texas (20.3–22.6 µm) and Mexican C. cf. proserpina specimens (21.9–22.2 µm). Conversely, the hamulus inner root departed from a strict latitudinal gradient. While Canadian (24–25.8 µm) and Texan (18.3–21.6 µm) parameters remained intermediate, the Mexican population of C. lutrensis displayed an elongated root (26.6–27.2 µm) that stood in sharp contrast to the brief inner root found on its sympatric counterpart, C. cf. proserpina (17.4–18.8 µm). Dimensions of the ventral bar structure reflected diverse patterns of differentiation. Total width of the ventral bar was heavily developed in Canadian (36.4–41.3 µm) and Texan (30.7–36.9 µm) gyrodactylids specimens but appeared uniquely stunted in the Upper Mississippi River sample (20.0–22.7 µm); the Mexican specimens occupied a middle tier, showing greater width on C. lutrensis (31.8–32.2 µm) than on C. cf. proserpina (27.9–28.5 µm). The ventral bar membrane followed a cleaner latitudinal decline, shrinking from its maximum in Canada (19.8–22.3 µm) down to its minimum size in Mexico (15.5–16.8 µm on C. lutrensis). The lateral processes of the ventral bar remained consistently minute across all regions, hitting their lowest thresholds in the US Upper Mississippi population (5.4–7.4 µm) and the Mexican C. cf. proserpina population (5.5–5.9 µm). The dorsal bar total length demonstrated significant structural consistency across most regions, with a slight compression evident in the far southern populations. Specimens from Canada (19.6–22.8 µm) and the USA (18.2–23 µm in the Upper Mississippi River; 21–26.2 µm in the Guadalupe River) overlapped substantially in total length. Meanwhile, the Mexican populations exhibited marginally smaller dimensions, measuring 15.4–16.2 µm on C. lutrensis and 17.7–19.6 µm on C. cf. proserpina. The dorsal bar median width was rarely reported across the literature, but remained stable where recorded, measuring 1.7–2.5 µm in the Canadian population and 1.8–2.9 µm in the Texan population. Total marginal hook length was longest in the far north (Canada: 35.0–37.8 µm; Upper Mississippi River: 33.9–38.7 µm), became moderately restricted in Texas (31–38.8 µm), and reached its shortest manifestation in Mexican parasites from C. cf. proserpina (30.9–32.6 µm). This regional shrinking was driven by the marginal hook handle, which measured 30.1–33.1 µm in Canada but only 27.4–28.1 µm in the Mexican C. cf. proserpina cohort. Similarly, the marginal hook sickle proper was largest in Canadian specimens (4.8–5.1 µm) and smallest in Mexican specimens from C. cf. proserpina (3.9–4.1 µm). Sclerite shape was omitted from this analysis due to the lack of morphological precision or diagnostic quality in older published illustrations, preventing reliable visual comparisons.
Discussion
Understanding geographic, host and population-level patterns in Gyrodactylus species requires integrating molecular and morphological variability. In the present study, 2 already known and apparently widespread species in the Nearctic, G. atratuli parasitizing R. cataractae and G. mediotorus parasitizing both C. lutrensis and C. cf. proserpina, were found in Mexican freshwaters. Rhinichthys cataractae and C. lutrensis are ecologically versatile leuciscid hosts with wide distributions, ranging from cold Canadian streams to warmer-temperate systems in the USA and northern Mexico. In contrast, C. cf. proserpina has a more restricted distribution in northeastern Mexico (Page and Burr, Reference Page and Burr2011; Froese and Pauly, Reference Froese and Pauly2026). These distributional patterns may contribute to population-level variation in parasite morphology and genetics.
Sequences of the ITS regions corresponding to G. atratuli revealed intraspecific variability, which likely reflects natural genetic structure rather than host-species level differentiation. Such variability is not uncommon in Gyrodactylus, where small but consistent differences in ribosomal DNA markers have been reported across distinct populations (Ziȩtara and Lumme, Reference Ziȩtara and Lumme2003; Huyse et al. Reference Huyse, Malmberg and Volckaert2004; Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2023b). Given the vast geographic range of R. cataractae and its associated cypriniform hosts, local parasite populations are expected to accumulate subtle genetic differences resulting from isolation-by-distance, host-associated divergence, or historical demographic events. The observed divergence in the ITS region (0.7%) likely represents the early stages of allopatric differentiation. Restricted gene flow across distant host populations allows genetic drift and localized selection pressures to shape the parasite’s genetic profile, an evolutionary pattern previously noted in gyrodactylids from the northwest Mediterranean (Rahmouni et al. Reference Rahmouni, Seifertová and Šimková2025).
With respect to host specificity and on the basis of this study and previous ones (Putz and Hoffman, Reference Putz and Hoffman1963; Hanek and Fernando, Reference Hanek and Fernando1971; Kirby, Reference Kirby1981), the occurrence of G. atratuli on multiple leuciscid hosts across the North American continent indicates that this parasite is not strictly host-specific but exhibits a wide host range within the family Leuciscidae. Such host use across different fish genera suggests ecological host specificity at the family level, consistent with patterns observed in other Gyrodactylus species that parasitize a range of closely-related cypriniform fishes (Rahmouni et al. Reference Rahmouni, Seifertová, Benovics and Šimková2023a, Reference Rahmouni, Seifertová and Šimková2023b, Reference Rahmouni, Seifertová and Šimková2025).
Parallel to the findings for G. atratuli, the minor genetic variations observed among G. mediotorus populations (up to 8 bp over 963 bp in the ITS region) reflect a natural geographic structure. This genetic cohesion across diverse leuciscid hosts and major river basins in North America suggests a relatively recent common ancestry and historical gene flow facilitated by the migratory capacity or post-glacial dispersal of their hosts. Similar levels of intraspecific variability without accompanying speciation events have been well-documented in other broadly distributed monogenean (Rahmouni et al. Reference Rahmouni, Seifertová, Benovics and Šimková2023a, Reference Rahmouni, Seifertová and Šimková2023b, Reference Rahmouni, Seifertová and Šimková2025).
Regarding host specificity, G. mediotorus has previously been documented infecting various leuciscid hosts in northern North America, including species of Notropis and Cyprinella (King et al. Reference King, Marcogliese, Forest, McLaughlin and Bentzen2013; Leis et al. Reference Leis, Easy and Cone2021; Rahmouni et al. Reference Rahmouni, Seifertová, Bean and Šimková2024). In Mexico, our findings restrict its known hosts to Cyprinella species, demonstrating a low degree of host specificity constrained primarily at the host family level (Leuciscidae). Further parasitological surveys of co-occurring Notropis species in southern Nearctic drainages are necessary to fully resolve the geographic host boundaries and host-parasite association dynamics of this monogenean.
Morphologically, both G. atratuli and G. mediotorus exhibit a conspicuous latitudinal (north-to-south) decrease in haptoral sclerite dimensions. This reduction is primarily manifested in the hamulus total length, inner root and point, as well as the marginal hook total length, handle and sickle proper. This clinal variation likely reflects an adaptive ecomorphological response to ambient temperature regimes. In colder northern waters, larger, more robust haptoral structures may optimize attachment by compensating for increased water viscosity and reduced host metabolic activity, whereas smaller anchoring structures suffice in warmer southern environments. Similar environmentally driven morphometric plasticity has been reported in other gyrodactylids, such as Gyrodactylus salaris Malmberg, Reference Malmberg1957, Gyrodactylus rhodei Zitnan, 1964, and several Black Sea species (Gyrodactylus alviga Dmitrieva and Gerasev, 2000; Gyrodactylus crenilabri Zaika, 1966; Gyrodactylus flesi Malmberg, Reference Malmberg1957; and Gyrodactylus sphinx Dmitrieva and Gerasev, 2000), where seasonal temperature drops and salinity fluctuations regularly induce larger sclerite phenotypes (Mo, Reference Mo1991; Dmitrieva and Dimitrov, Reference Dmitrieva and Dimitrov2002; Dávidová et al. Reference Dávidová, Jarkovský, Matějusová and Gelnar2005). The ventral bar and dorsal bar complexes in both species revealed complex morphometric dynamics. While the ventral bar total length remained relatively stable across regions, components like the ventral bar membrane length showed a sharp, clinal reduction from North to South. Interestingly, the dorsal bar total length of G. mediotorus remained highly conserved and uniform between Canada and the USA, experiencing only a minor structural compression in the far southern Mexican populations (see Table 1). This regional stability of the dorsal bar indicates it is under tighter evolutionary constraints than more plastic components of the attachment apparatus. The differential variability among haptoral structures aligns with observations from Ligophorus cephali Rubtsova, Balbuena, Sarabeev, Blasco-Costa and Euzet, 2006 (Dactylogyridae), where specific anchoring elements exhibit strict morphological control, while connecting bars show greater phenotypic plasticity to accommodate local physical forces (Rodríguez-González et al. Reference Rodríguez-González, Míguez-lozano, Llopis-belenguer and Balbuena2015). In Gyrodactylus, the hamuli and marginal hooks appear to be under strong stabilizing selection due to their primary role in penetrating the host epidermis, whereas parts of the connecting bars and membranes display higher flexibility, allowing for subtle adjustments to local environmental variables or host-specific skin traits.
Furthermore, host species and their relative microhabitats contribute significantly to the morphological architecture of these parasites. Specimens from R. atratulus and C. lutrensis generally possessed larger, more robust hamuli, inner roots and ventral bars compared to those collected from R. cataractae or C. cf. proserpina. This variation corresponds to a cumulative gradient where northern populations parasitizing wider-bodied or less hydrodynamically stressed hosts display maximum sclerite sizes. In contrast, populations from high-velocity specialists like R. cataractae, or restricted southern species like C. cf. proserpina, present significantly shorter attachment elements. These trends demonstrate that host body size, scale structure and microhabitat preferences interact with latitudinal gradients to influence parasite attachment mechanics, this situation consistent with macroevolutionary models showing that monogeneans optimize their attachment organelle size to match the specific colonization constraints and surface features of their hosts (Sasal et al. Reference Sasal, Trouve, Muller-graf and Morand2001; Šimková et al. Reference Šimková, Desdevises, Gelnar and Morand2001, Reference Šimková, Verneau, Gelnar and Morand2006). Ultimately, the geographic and host-dependent patterns observed in the haptor morphology of G. atratuli and G. mediotorus reflect a balance between historical biogeography and contemporary ecological pressures. Conserved anchoring frameworks, such as the hamulus points, marginal hooks and the total length of the dorsal bar, primarily track a shared evolutionary history and genetic ancestry, whereas more plastic structures, including the ventral bar membrane and process extensions, are dynamically modulated by local environmental temperatures and host-specific conditions (Rodríguez-González et al. Reference Rodríguez-González, Sarabeev and Balbuena2017).
Conclusion
Although sample sizes were limited, the observed phenotypic patterns in G. atratuli and G. mediotorus align with documented host- and geographic-related variability. The evident latitudinal trends, characterized by a north-to-south reduction in haptoral sclerite size, likely reflect compensatory morphological responses to ambient temperature and host-specific traits. While highly conserved elements like the hamulus points, marginal hooks and dorsal bar total length track a shared evolutionary history, more plastic structures like the ventral bar membrane adjust dynamically to environmental and microhabitat pressures. Ultimately, these findings underscore a dual-gradient matrix of adaptation within the Nearctic, though future studies with larger sample sizes remain essential to confirm these ecomorphological patterns.
Data availability statement
The data supporting the conclusions of this paper are included within the article. Accession numbers for both voucher specimens and genetic sequences deposited in GenBank are provided in the taxonomic summary for each identified species.
Acknowledgements
We would like to express our sincere gratitude to Eva Řehulková from Masaryk University (Brno, Czech Republic), and to Roman Kuchta, Tomáš Scholz and Carlos Mendoza-Palmero from the Institute of Parasitology, Biology Centre of the Czech Academy of Sciences (České Budějovice, Czech Republic) for their assistance with fish dissection, parasite isolation and fixation, as well as for their valuable support, collaboration and scientific contributions to this work. We are also deeply indebted to Rodolfo Pérez-Rodríguez from Universidad Michoacana de San Nicolás de Hidalgo (Morelia, Michoacán, Mexico) for his collaboration and assistance throughout this study. Special thanks are extended to Gabriel Fernando Cardoza-Martínez and his colleagues from Universidad Juárez del Estado de Durango (Durango, Mexico) for their collaboration, support and contribution to the collection and study of the material used in this research. The authors express deep gratitude to Blanka Škoríková from the Institute of Parasitology, Biology Centre of the Czech Academy of Sciences (České Budějovice, Czech Republic) for her kind help with the deposition of the type material at IPCAS. Special thanks are also extended to Matthew Nicholls for English language editing, as well as to the anonymous reviewers for their valuable feedback on the manuscript.
Author contributions
A.Š. supervised the study and provided scientific background in the field of monogenean fauna of North African cyprinid fish. C.R. designed the study. C.R. and M.Z. performed microscopical observations, identified the new species and photographed the hard parts. M.S. and M.Z. isolated, amplified and sequenced the DNA samples. C.R. and A.Š. discussed the results. C.R. wrote the manuscript. C.R., M.S. and A.Š. revised the manuscript. All authors have read and approved the final version of the manuscript.
Financial support
This study was funded by the Ministry of Education, Youth and Sports of the Czech Republic, project no. LUAUS23080.
Competing interests
The authors declare that they have no competing interests.
Ethical standards
Fish were collected under personal permits issued to researchers from the Mexican institutions; permit holders are listed in the Acknowledgements section.
Open access
