Introduction
Phenotypic plasticity is the ability of a single species to produce multiple distinct phenotypes in response to the environmental conditions (Miner et al., Reference Miner, Sultan, Morgan, Padilla and Relyea2005). In parasites, with complex life cycles, different environmental conditions include (1) the host's immune system, (2) different host species and (3) the geographical distribution of the definitive hosts. All of these factors are correlated with the phenotypic plasticity, mainly in the body size or fecundity of parasites (Poulin, Reference Poulin2007).
Trematodes species are characterized mainly using morphological features, such as body size, proportion, shape and location of internal organs. Understanding the phenotypic plasticity in populations is essential and key to defining, recognizing and delineating species (Hildebrand et al., Reference Hildebrand, Adamcyk, Laskowski and Zaleśny2015). Parasite taxonomy has heeded the call to shift toward ‘integrative taxonomy’, i.e. the use of multiple and complementary sources (Dayrat, Reference Dayrat2005). The recent application of molecular markers in species identification has uncovered an extraordinary genetic richness in parasites, revealing many times more species than presently described (Poulin, Reference Poulin2011; Pérez-Ponce de León and Nadler, Reference Pérez-Ponce de León and Nadler2011). In addition, integrative taxonomy has helped to define, recognize, delineate and better understand the intraspecific variation that can be attributed to differences in the development and phenotypic plasticity of parasites (Hildebrand et al., Reference Hildebrand, Adamcyk, Laskowski and Zaleśny2015; Poulin and Presswell, Reference Poulin and Presswell2016).
Saccocoelioides Szidat, 1954 is the most diverse genus of trematodes belonging to the subfamily Chalcinotrematinae, and includes 24 recognized species, of which 14 are distributed in South America, seven in Middle America, two in North America and one in Puerto Rico; all these species are associated with freshwater, brackish and marine fishes from 10 families (Curran et al., Reference Curran, Pulis, Andres and Overstreet2018; Andrade-Gómez et al., Reference Andrade-Gómez, Sereno-Uribe and García-Varela2019; Gallas and Utz, Reference Gallas and Utz2019). The genetic library of species of Saccocoelioides has increased significantly in the last few years. Curran et al. (Reference Curran, Pulis, Andres and Overstreet2018) and Andrade-Gómez et al. (Reference Andrade-Gómez, Sereno-Uribe and García-Varela2019) evaluated the systematics of the genus Saccocoelioides by combining nuclear and mitochondrial molecular markers, and ecological and morphological characteristics, detecting an extraordinary diversity in the Americas. Currently, Middle America harbours seven species of Saccocoelioides, S. macrospinosus Andrade-Gómez et al., Reference Andrade-Gómez, Sereno-Uribe and García-Varela2019; S. orosiensis Curran et al., Reference Curran, Pulis, Andres and Overstreet2018; S. tkachi Curran et al., Reference Curran, Pulis, Andres and Overstreet2018; S. olmecae, Andrade-Gómez et al., Reference Andrade-Gómez, Pinacho-Pinacho, Hernández-Orts, Sereno-Uribe and García-Varela2016; S. cichlidorum (Aguirre-Macedo and Scholz, 2005) Andrade-Gómez et al., Reference Andrade-Gómez, Pinacho-Pinacho and García-Varela2017; S. chauhani Lamothe-Argumedo, 1974; and S. lamothei Aguirre-Macedo and Violante-González, Reference Aguirre-Macedo and Violante-González2008.
Saccocoelioides lamothei was described from the Pacific fat sleeper fish, Dormitator latrifons Richardson, 1844, from coastal lagoons of Guerrero state, Mexico (Aguirre-Macedo and Violante-González, Reference Aguirre-Macedo and Violante-González2008), and was subsequently reported as being associated with four fish species from three families (Poeciliidae, Profundulidae and Gobiidae) along the Pacific coasts of Middle America (Aguirre-Macedo and Violante-González, Reference Aguirre-Macedo and Violante-González2008; Pinacho-Pinacho et al., Reference Pinacho-Pinacho, García-Varela, Hernández-Orts, Mendoza-Palmero, Sereno-Uribe, Martínez-Ramírez, Andrade-Gómez, Hernández-Cruz, López-Jiménez and Pérez-Ponce de León2015; Andrade-Gómez et al., Reference Andrade-Gómez, Pinacho-Pinacho, Hernández-Orts, Sereno-Uribe and García-Varela2016, Reference Andrade-Gómez, Pinacho-Pinacho and García-Varela2017, Reference Andrade-Gómez, Sereno-Uribe and García-Varela2019; Curran et al., Reference Curran, Pulis, Andres and Overstreet2018). Curran et al. (Reference Curran, Pulis, Andres and Overstreet2018) noted that the biodiversity of Saccocoelioides in Middle America is far from well-known and that parasitological studies that combine morphological and molecular data are necessary to documenting its diversity in this biogeographical region.
The aim of the present study was to combine morphological and molecular characteristics to investigate the specific status of Saccocoelioides spp. in association with five fish families: Eleotridae, Mugilidae, Gobiidae, Poeciliidae and Profundulidae distributed along the Pacific coasts of Mexico, Guatemala, El Salvador, Honduras, Nicaragua and Costa Rica in Middle America. Sequences of three molecular markers were generated: the domains D1–D3 of the large subunit (LSU) from nuclear ribosomal DNA, the cytochrome c oxidase subunit 1 (cox1) and nicotinamide adenine dinucleotide dehydrogenase subunit 1 (nad1) from mitochondrial DNA.
Materials and methods
Specimen collection
Digeneans were collected from 2010 through 2019 from the intestines of their definitive hosts in 29 localities from Middle America, 19 from Mexico, three from Guatemala, three from El Salvador, two from Nicaragua, one from Honduras and one from Costa Rica (see Fig. 1; Table 1). Fishes were collected with seine nets, cast nets and electrofishing, kept alive and transported to the laboratory. Each fish was euthanized and immediately examined. The collected digeneans were preserved either in 100% ethanol for DNA extraction or in hot (steaming) 4% formalin for morphological examination. Fishes were identified following the keys of Miller et al. (Reference Miller, Minckley and Norris2005).
Fig. 1. Map indicating the 29 localities for Saccocoelioides spp., in Middle America; localities correspond with Table 1.
Table 1. Specimens analysed in this study; localities (in parenthesis are localities sampled in this study and correspond with Fig. 1); host species; host family and GenBank accession numbers of each molecular marker. Sequences in bold were generated in this study
Amplification, sequencing of DNA, phylogenetic analyses and haplotype network
Each specimen of Saccocoelioides spp. was placed individually in tubes and digested overnight at 56°C in a solution containing 10 mm Tris–HCl (pH 7.6), 20 mm NaCl, 100 mm Na2 EDTA (pH 8.0), 1% Sarkosyl and 0.1 mg/ml proteinase K. Following digestion, DNA was extracted from the supernatant using the DNAzol reagent (Molecular Research Center, Cincinnati, OH, USA) according to the manufacturer's instructions. The domains D1–D3 of the LSU of nuclear ribosomal DNA plus two partial regions of mitochondrial DNA, cytochrome c oxidase subunit 1 (cox1) and nicotinamide adenine dinucleotide dehydrogenase subunit 1 (nad1) were amplified using polymerase chain reaction (PCR). Domains D1–D3 from LSU were amplified using forward 5′–AGCGGAGGAAAAGAAACTAA–3′ (Nadler et al., Reference Nadler, D´Amelio, Fagerholm, Berland and Paggi2000) and reverse 5′–CAGCTATCCTGAGGGAAAC–3′ primers (García-Varela and Nadler, Reference García-Varela and Nadler2005). Two new primers were designed for the fragment of the cox1, forward primer SaccoF, 5′–TGTAAAACGACGGCCAGTTTWCITTRGATCATAAG–3′ and reverse primer SaccoR, 5′–TAAAGAAAGAACATAATGAAAATG–3′. Finally, the gene nad1 was amplified using forward 5′–AGATTCGTAAGGGGCCTAATA–3′ (Morgan and Blair, Reference Morgan and Blair1998) and reverse 5′–CTTCAGCCTCAGCATAAT–3′ primers (Kostadinova et al., Reference Kostadinova, Herniou, Barrett and Littlewood2003). PCR reactions (25 μL) consisted of 1 μL of each primer (10 μ m), 2.5 μL of 10× buffer, 1.5 μL of 2 mm MgCl2, 0.5 μL of dNTPs (10 mm), 16.375 of water, 2 μL of genomic DNA and 1 U of Taq DNA polymerase (Platinum Taq, Invitrogen Corporation, São Paulo, Brazil). PCR cycling condition amplifications included denaturation at 94°C for 3 min, followed by 35 cycles of 94°C for 1 min, annealing at 40°C for cox1 and 50°C for nad1 and LSU for 1 min, and extension at 72°C for 1 min, followed by postamplification incubation at 72°C for 10 min. The sequencing reactions were performed using the initial primers cox1, nad1 and LSU plus two internal primers: forward 5′–CCTTGGTCCGTGTTTCAAGACG–3′ and reverse 5′–CGTCTTGAAACACGGACTAAGG–3′ primers (García-Varela and Nadler, Reference García-Varela and Nadler2005), for LSU with ABI Big Dye (Applied Biosystems, Boston, MA, USA) terminator sequencing chemistry, and reaction products were separated and detected using an ABI 3730 capillary DNA sequencer. Contigs were assembled and base-calling differences resolved using CodonCode Aligner v 9.0.1 (CodonCode Corporation, Dedham, MA, USA). Sequences obtained in the current research for LSU and cox1 were aligned with sequences of Saccocoelioides spp., downloaded from GenBank (Table 1). In addition, sequences obtained for nad1 of Saccocoelioides spp., and S. lamothei from type host and type locality, were aligned.
Sequences of each molecular marker were aligned separately using the software SeaView v.4 (Gouy et al., Reference Gouy, Guindon and Gascuel2010) and adjusted with the Mesquite program (Maddison and Maddison, Reference Maddison and Maddison2011). The nucleotide substitution model was selected for each molecular marker using jModelTest v0.1.1 (Posada, Reference Posada2008) and applying the Akaike information criterion; for the LSU dataset, the selected model was TVM + I, and for cox1 TrN + I + G. The phylogenetic analyses were performed with LSU and cox1 using maximum likelihood (ML) and Bayesian inference (BI) methods, using the online interface Cyberinfrastructure for Phylogenetic Research (CIPRES) Science Gateway v3.3 (Miller et al., Reference Miller, Pfeiffer and Schwartz2010). The ML analyses were carried out with RAxML v.7.0.4 (Silvestro and Michalak, Reference Silvestro and Michalak2011), and 10 000 bootstrap replicates were run to assess nodal support. The BI analyses were inferred with MrBayes v.3.2.7 (Ronquist et al., Reference Ronquist, Teslenko, Van der Mark, Ayres, Darling, Höhna, Larget, Liu, Suchard and Huelsenbeck2012), with two simultaneous runs of the Markov Chain Monte Carlo (MCMC) for 10 million generations, sampled every 1000 generations, a heating parameter value of 0.2 and a ‘burn-in’ of 25%. Trees were drawn using FigTree v.1.3.1 (Rambaut, Reference Rambaut2012). The genetic divergence among taxa was estimated using uncorrected ‘P’ distances with the program MEGA version 6 (Tamura et al., Reference Tamura, Stecher, Peterson, Filipski and Kumar2013).
In order to examine the relationships among Saccocoelioides spp., the nad1 haplotype frequency was estimated, an unrooted statistical network was constructed using the program NETWORK version 5.0 (www.fluxus-engineering.com) keeping the ɛ = 0. This method starts with minimum spanning trees combined within a single network and then, to reduce tree length, median vectors (consensus sequences) are added. Such vectors can be interpreted as possibly extant unsampled sequences or extinct ancestral sequence (Bandelt et al., Reference Bandelt, Forster and Röhl1999). In addition, the median-joining algorithm was employed to build the network.
Morphometrics analysis
Unflattened specimens preserved in formalin were stained with Mayer's paracarmine, dehydrated in a graded ethanol series, cleared with methyl salicylate and mounted on permanent slides with Canada balsam. Specimens collected in the present study were compared with the specimens of S. lamothei deposited at the Colección Nacional de Helmintos (CNHE), Instituto de Biología, Universidad Nacional Autónoma de México, Ciudad de México, México. All the specimens were examined using a bright-field Leica DM1000 LED compound microscope (Leica, Wetzlar, Germany) and new specimens were deposited in CNHE.
Drawings were made using a drawing tube attached to the microscope. A total of 17 morphological features (BL, body length; BW, maximum body width; OSL, oral sucker length; OSW, oral sucker width; VSL, ventral sucker length; VSW, ventral sucker width; P, prepharynx; PHL, pharynx length; PHW, pharynx width; HSL, hermaphroditic sac length; HSW, maximum hermaphroditic sac width; TL, testis length; TW, testis width; OL, ovary length; OW, ovary width; EL, egg length; EW, egg width) were measured of 53 individuals with the Leica Application Suite microscope software. All measurements are in micrometres (μm). A principal component analysis (PCA) was implemented to explore and describe the patterns of morphological variation of the specimens. PCA was conducted using R-packages ggplot2, ggfortify, cluster and alpha implemented in R (R Core Team, 2016). For the analysis, 21 specimens identified as S. lamothei from CNHE (4349, 4350, 4906, 6671, 9370–9373), and 32 newly collected individuals from Saccocoelioides spp. were selected. PCA included the 17 characteristics; BL, BW, OSL, OSW, VSL, VSW, P, PHL, PHW, HSL, HSW, TL, TW, OL, OW, EL, EW.
Results
Phylogenetic trees
The LSU dataset included 1271 characters and the best model obtained was TVM + I. The alignment included a total of 91 sequences, 74 new sequences of Saccocoelioides spp., plus six sequences of S. lamothei (KU061120–KU061124, MG925110) and one sequence each of S. chauhani (KU061119), S. macrospinosus (MK749169), S. olmecae (KU061136), S. orosiensis (MG925108), S. tkachi (MG925122), S. cichlidorum (MG925106), S. sogandaresi (MG925120), S. magnus Szidat, 1954 (MG925112), S. elongatus Szidat, 1954 (MG925108), S. nanii Szidat, 1954 (MG925114) and S. beauforti Hunter and Thomas, 1961 (MG925104) (see Table 1). The phylogenetic analyses inferred with ML and BI yielded several clades. The base of the tree was formed by a polytomy that contained two species from South America, S. magnus and S. elongatus. Another clade was formed by S. orosiensis and S. nanii and an unresolved branch belonging to S. macrospinosus. Other clade was formed by four species (S. sogandaresi, S. chauhani, S. olmecae and S. beauforti). A clade was formed by S. tkachi and S. cichlidorum which was supported with a good bootstrap value and Bayesian posterior probabilities (Fig. 2). The remaining new sequences of Saccocoelioides spp. formed a clade that included six sequences previously identified as S. lamothei, two (KU061120–KU061121) from the type host and type locality; others from Chacahua, Oaxaca, Mexico (KU01122–KU01124) and one (MG925110) from Río Tempisque, Costa Rica. This clade received strong support from the bootstrap and Bayesian posterior probabilities (100/1) (Fig. 2).
Fig. 2. Maximum likelihood tree and consensus Bayesian inference trees inferred from the LSU dataset. Numbers near internal nodes show ML bootstrap clade frequencies and posterior probabilities (BI). n, number of identical sequences; H, host name. In parenthesis, number of locality (Table 1). Sequences in bold of Saccocoelioides lamothei were downloaded from GenBank
The genetic divergence estimated using the LSU dataset from rDNA among the new sequences of Saccocoelioides spp. and S. lamothei ranged from 0 to 0.31%. The monophyly of the new sequences and the six sequences previously identified as S. lamothei, in combination with the low genetic divergence among the sequences, suggests that all new sequences belong to S. lamothei, expanding its distribution range and host spectrum along the Pacific coasts of Middle America.
A second dataset for the cox1 gene included 588 characters, and the best selected model was TrN + I + G. The alignment included 57 terminals, 47 new sequences of S. lamothei, plus 10 sequences of the genus Saccocoelioides, two of S. lamothei (MK749571–72) from the type host and type locality, two of S. orosiensis (MK749590 and MK749602), two of S. tkachi (MK749577–78) and one of S. cichlidorum (MK749574), S. olmecae (MK749584), S. macrospinosus (MK749566) and S. chauhani (MK749589) (Table 1). The phylogenetic analyses inferred with ML and BI from the cox1 dataset showed that the specimens collected in this study formed a clade with specimens previously identified as S. lamothei with strong support from the bootstrap and Bayesian posterior probabilities (100/1) (Fig. 3), suggesting that all of the sequences belonged to the same lineage. The genetic intraspecific divergence estimated with the cox1 dataset among the new sequences and two sequences of S. lamothei from the type host and type locality ranged from 0 to 6.62%.
Fig. 3. Maximum likelihood tree and consensus Bayesian inference trees inferred with cox1 dataset; numbers near internal nodes show ML bootstrap clade frequencies and posterior probabilities (BI). H, host. In parenthesis, number of locality (Table 1). Sequences in bold of Saccocoelioides lamothei were downloaded from GenBank.
Haplotype network
A haplotype network was built using the nad1 gene. This dataset was formed by 119 specimens with 485 characters and 57 haplotypes were detected. The haplotype network yielded three haplogroups separated by a few mutational steps (no more than 12 steps) (Fig. 4). The first haplogroup contained 17 distinct haplotypes from 32 specimens from eleotrid and mugilid fishes. The second haplogroup contained three haplotypes (BS, BU and BT) from mugilid fishes. Only the third haplogroup contained 37 distinct haplotypes from 84 individuals from five host families (see Table S1). The most frequent haplotype was CB, which was formed by 13 specimens from four localities (4, 5, 9, 25 see Fig. 5; Table S1) in two countries, Mexico and El Salvador, from eleotrid and mugilid fishes. Most of the haplotypes had a restricted geographic distribution. For instance, five localities (3, 7, 10, 23, 29) showed a unique haplotype (a single haplotype per locality; AD, BJ, BO, CD, CF), and 15 localities (2, 4, 6, 9, 11, 13–15, 18–20, 24–27) had 45 exclusive haplotypes. The remaining nine localities (1, 5, 8, 12, 16, 17, 21, 22, 28) shared haplotypes (Fig. 5; Table S1).
Fig. 4. Haplotype network of samples of Saccocoelioides lamothei, build with the gene nicotinamide adenine dinucleotide dehydrogenase subunit 1 (nad1). Each circle represents a haplotype, with size proportional to the haplotype's frequency in the populations. The haplotype name is indicated in letters and corresponds with Table S1.
Fig. 5. Distribution of haplotypes in the localities sampled of Saccocoelioides lamothei. Size corresponds to the number of specimens sampled (see Table S1).
Morphometric analyses
Morphometric analysis was conducted to corroborate that the morphological differences among the isolates of S. lamothei were associated with their different hosts. A total of 17 variables were considered from 53 specimens (Table 2). PCA was used to classify host species (Fig. 6A). The measurements of the specimens from D. latifrons formed a separate polygon. However, the measurements of the other specimens from different host species overlapped with each other (Fig. 6A). Later, the measurements of specimens from each host species were clustered by host families (Eleotridae, Mugilidae and Gobiidae). PCA showed three polygons representing each definitive host family (Fig. 6B). Those three polygons relate to each morphotype found (Fig. 7).
Fig. 6. Principal component analysis conducted with 17 morphometric variables from 53 specimens of Saccocoelioides lamothei. (A) Specimens analysed by host species. (B) Specimens analysed by host families.
Fig. 7. Morphotypes of Saccocoelioides lamothei: (A) from Dormitator latifrons, Tres Palos, Guerrero, México; (B) from Mugil curema, Puerto San José, Guatemala; (C) from Sycidium sp., Río Tamarindo, Nicaragua. Scale bars = 100 μm
Table 2. Comparative morphometric data for the three morphotypes of Saccocoelioides lamothei.
Morphological description
Saccocoelioides lamothei Aguirre-Macedo and Violante-Gónzalez, Reference Aguirre-Macedo and Violante-González2008.
Based on 53 specimens studied from Dormitator latifrons, Mugil curema, Mugil sp., Sicydium multipunctatum, Sicydium sp. and Awaous banana from 10 localities distributed in Mexico, Guatemala, El Salvador, Nicaragua and Costa Rica (Table 2).
Tegument entirely covered by minute spines, being scatter in the posterior end in mugilid and gobiid fishes (Fig. 7A–C). Eye-spot remnants present in the anterior of the body reaching half of the pharynx (Fig. 7). Oral sucker subterminal. Ventral sucker slightly anterior to middle of the body or at the middle body in gobiid fishes (Fig. 7C). Prepharynx present or absent in gobiid fishes (Fig. 7C). Pharynx oval to spherical. Oesophagus long. Caeca sac-shaped elongated, terminating in hindbody. Testis oval to subspherical, in the middle of hindbody or the posterior end of hindbody in eleotrid fishes (Fig. 7A). External seminal vesicle small, contiguous to hermaphroditic sac. Hermaphroditic sac oval to spherical, at the level of ventral sucker, or anterior to ventral sucker in eleotrid fishes (Fig. 7A). Internal seminal vesicle elongated to spherical. Genital pore opening anterior to ventral sucker. Ovary elongated at the middle of the body. Laurer's canal not observed, Mehlis' gland not observed. Uterus confined between hermaphroditic sac and testis (Fig. 7B–C) or filling the entire body in eleotrid fishes (Fig. 7A), with the metraterm entering the posterior end of hermaphroditic sac. Vitelline follicles elongated, irregular, distributed in lateral fields from the level of the hermaphroditic sac to the posterior of the testis, anterior in gobiid fishes (Fig. 7C). Eggs operculate. Miracidia observed in eleotrid fishes. Excretory vesicle Y-shaped. Excretory pore terminal (Fig. 7; see Table 2).
Taxonomic summary
Saccocoelioides lamothei Aguirre-Macedo and Violante-González, Reference Aguirre-Macedo and Violante-González2008.
Type-host: Dormitator latifrons (Eleotridae).
Other hosts: Dajaus monticola, Mugil cephalus, M. curema, Mugil sp. (Mugilidae), Poecilia gillii, Poecilia mexicana, Poecilia sphenops, Poeciliopsis gracilis (Poeciliidae), A. banana, S. multipunctatum, Sicydium salvini, Sicydium sp. (Gobiidae) y Profundulus sp. (Profundulidae).
Type-locality: Tres Palos, Guerrero, México.
Other localities: México: El Huizache, Sinaloa; La Tovara and Nuevo Vallarta, Nayarit; Quémaro and Playa Punta Pérula, Jalisco; Cuyutlán and Estero Tecuanillo, Colima; Barra de Nexpa, Michoacán; Playa las Peñitas and Marquelia, Guerrero; Río Salado, San José de las Flores, Chacahua, Barra Navidad, Matías Romero and Ensenada la Ventosa, Oaxaca; Pijijiapán and Puerto Chiapas, Chiapas. Guatemala: Río Nahualate, Puerto San José and Las Lisas. El Salvador: Río Sunza, Río Banderas and Bahía de San Antonio. Honduras: Río Choluteca. Nicaragua: Río Tamarindo. Costa Rica: Río Ciruelas.
Site in host: Intestine.
Discussion
The phylogenetic analyses inferred with the LSU unequivocally placed all the new sequences from the Pacific coasts of Middle America into a monophyletic clade together with six sequences previously identified as S. lamothei, including specimens from the type host and type locality (see Fig. 2). The genetic divergence estimated among the 12 species of the genus Saccocoelioides ranged from 0.2 to 5.7% and its range was similar than found previously by Andrade-Gómez et al. (Reference Andrade-Gómez, Sereno-Uribe and García-Varela2019), who reported a range of genetic divergence from 0 to 4.8%. The intraspecific genetic divergence among the isolates of S. lamothei ranged from 0 to 0.31% for LSU. The intraspecific divergence found herein is similar to the LSU reported previously for S. tkachi (0–0.2%) and S. orosiensis (0–0.4%) (Andrade-Gómez et al., Reference Andrade-Gómez, Sereno-Uribe and García-Varela2019). The phylogenetic analysis inferred with the cox1 clearly distinguished species previously recognized within Saccocoelioides (see Fig. 3). The genetic divergence estimated with cox1 dataset among the seven species of Saccocoelioides ranged from 9.7 to 17% and its range (from 8.3 to 17%) was similar than reported previously (Andrade-Gómez et al., Reference Andrade-Gómez, Sereno-Uribe and García-Varela2019). The cox1 tree placed all the isolates of S. lamothei in a monophyletic subclade. From the 49 isolates of S. lamothei, 35 were recorded on four fish families (Eleotridae, Poeciliidae, Gobiidae and Mugilidae), with a sympatrical distribution. The remaining 14 isolates were found in six localities (3, 4, 6, 10, 13 and 15; Fig. 3; Table 1) associated with the Pacific fat sleeper (D. latifrons) and a mullet fish (Mugil sp.). The fat sleeper is an amphidromous species distributed from Northern Mexico to Ecuador (Galván-Quesada et al., Reference Galvan-Quesada, Doadrio, Alda, Perdices, Reina, Garcia Varela, Hernandez, Campos Mendoza, Bermingham and Dominguez-Dominguez2016). Meanwhile, mullets are distributed in freshwater, brackish and marine habitats in the Pacific coasts (Colín et al., Reference Colín, Hernández-Pérez, Guevara-Chumacero, Castañeda-Rico, Serrato-Díaz and Ibáñez2020). The intraspecific genetic divergence among the isolates of S. lamothei ranged from 0 to 6.62% for cox1 [as observed between one specimen from Mugil sp. in Barra de Navidad, Oaxaca, Mexico (MW283208), and one from P. mexicana Steindachner, 1863, in Río Choluteca, Honduras (MW283223); localities 15 and 26, respectively; Fig. 1; Table 1]. The intraspecific divergence of cox1 found herein is higher than reported previously; for example, the intraspecific divergence of S. tkachi ranged from 0 to 3.1% and for S. macrospinosus ranged from 0 to 3.3% (Andrade-Gómez et al., Reference Andrade-Gómez, Sereno-Uribe and García-Varela2019). The current research confirmed that the cox1 gene is a good molecular marker that allows delineating species and populations within Saccocoelioides.
The haplotype network analysis of nad1 detected 57 distinct haplotypes obtained from 119 individual sequences, which divided into three haplogroups separated by a few mutational steps (fewer than 12 steps) (Fig. 4). The three haplogroups were found in mugilid fishes from 18 localities from four countries (Mexico, Guatemala, El Salvador and Nicaragua). The distribution pattern of S. lamothei along the Pacific coasts of Middle America may have been formed by a combination of environmental factors and those related to the biology of the intermediate and definitive hosts, it is well known that adult mullets (M. cephalus and M. curema) have been found in sympatry (Ibañez et al., Reference Ibáñez, Chang, Hsu, Wang, Iizuka and Tzeng2012; Nirchio et al., Reference Nirchio, Oliveira, Siccha-Ramirez, de Sene, Sola, Milana and Rossi2017; Colín et al., Reference Colín, Hernández-Pérez, Guevara-Chumacero, Castañeda-Rico, Serrato-Díaz and Ibáñez2020). Both mullet species live and mature sexually in the open sea, where they migrate to different regions following marine currents (Funicelli et al., Reference Funicelli, Meineke, Bryant, Dewey, Ludwig and Mengel1989; Thomson, Reference Thomson1997). Mullets spawn offshore, and larval stages migrate from the open sea to the estuaries and lagoons near their nursery grounds (De Silva, Reference De Silva1980; Thomson, Reference Thomson1997). The life cycle of three species of the genus Saccocoelioides (S. tilapiae Nasir and Gómez, 1976; S. carolae Lunaschi, 1984; S. tarpazensis Díaz and González, Reference Díaz and González1990) are well known (see Martorelli, Reference Martorelli1986; Díaz and González, Reference Díaz and González1990; Díaz et al., Reference Díaz, Bashirullah, Hernández and Gómez2009).
Adult worms live and reproduce sexually in the digestive tracts of freshwater fishes, which serve as definitive hosts. Eggs are expelled into the environment in the feces of their host. Then, the eggs develop into miracidia, ciliate free-swimming larval forms that search for and penetrate snails of the genus Pyrgophorus Ancey, 1888, which serves as the intermediate host and in which the parasites develop into cercariae. Cercariae emerge from snails and are encysted on the water surface where they develop into metacercariae. Metacercariae are frequently found on aquatic vegetation that is ingested by their definitive hosts (Martorelli, Reference Martorelli1986; Díaz and González, Reference Díaz and González1990; Díaz et al., Reference Díaz, Bashirullah, Hernández and Gómez2009). Mullets feed on aquatic vegetation, and their life cycle is completed in the open sea, estuaries and lagoons along the Pacific coasts of Middle America (Ibañez et al., Reference Ibáñez, Chang, Hsu, Wang, Iizuka and Tzeng2012). Andres et al. (Reference Andres, Pulis, Curran and Overstreet2018) noted that mugilid fishes act as ‘ecological bridges’ between marine, estuarine and freshwater habitats and can disperse parasites along their range of distribution.
Morphometric analyses of the 53 specimens of S. lamothei recovered from three host families, exhibited remarkable morphological differences (Fig. 6B). PCA considered 17 variables (Table 2) and clearly showed three polygons corresponding to specimens recovered from the families Eleotridae, Mugilidae and Gobiidae, which sympatrically inhabit the Pacific coasts of Middle America, suggesting host-induced phenotypic plasticity. For example, S. lamothei associated with eleotrids has the widest body (240–510 μm); those associated with mugilids have the longest body (533–906 μm); those associated with gobiids have the smallest testis (52–164 μm) (see Table 2; Fig. 7). The phenotypic plasticity found in S. lamothei along its distribution range is consistent with a previous study of Saccocoelium tensum Looss, 1902 (a haploporid that parasitizes two mugilids, Liza ramada Risso, 1827 and Liza aurata Risso, 1810, from the Mediterranean Sea coasts of Spain) which has four morphotypes (Blasco-Costa et al., Reference Blasco-Costa, Balbuena, Raga, Kostadinova and Olson2010). Morphological plasticity has also been documented in numerous trematodes species, and its variation has been linked to its definitive hosts (Blankespoor, Reference Blankespoor1974; Pérez-Ponce de León, Reference Pérez-Ponce de León1995; Blasco-Costa et al., Reference Blasco-Costa, Balbuena, Raga, Kostadinova and Olson2010). Many morphologically distinct taxa of trematodes are assumed to represent several species or complexes of species, but they have been resolved into a single species under molecular and morphometric analyses, showing that parasites can alter their morphology depending on their host. This trait allows them to utilize a wide variety of definitive hosts.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0031182020002334
Acknowledgements
We thank Luis García Prieto for providing material from the CNHE and Laura Márquez and Nelly López Ortiz from LaNabio for their help during the sequencing of the DNA fragments. LAG thanks the support of the Programa de Posgrado en Ciencias Biológicas, UNAM and CONACYT (LAG CVU. No. 640068), for granting a scholarship to complete his PhD program.
Financial support
This research was supported by grants from the Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica (PAPIIT-UNAM) IN207219. Specimens were collected under the Cartilla Nacional de Colector Científico (FAUT 0202) issued by the Secretaría del Medio Ambiente y Recursos Naturales (SEMARNAT) to M.G.V.
Conflict of interest
None.
Ethical standards
Not applicable.
