Contrasting the population genetic structure of a specialist (Hexaglandula corynosoma: Acanthocephala: Polymorphidae) and a generalist parasite (Southwellina hispida) distributed sympatrically in Mexico

Polymorphidae is a monophyletic group of acanthocephalans distributed worldwide. Within this family, Hexaglandula corynosoma is a specialist species that uses a single bird species as a definitive host. Southwellina hispida is a generalist species that uses a broad spectrum of definitive hosts to complete its life cycle. In the current research, sequences of cytochrome c oxidase subunit 1 (cox1) from mitochondrial DNA were generated from 44 specimens of H. corynosoma and 76 of S. hispida distributed sympatrically in 6 biogeographic provinces of Mexico with the objective of characterizing and comparing the population genetic structure of 2 acanthocephalan species with opposing life strategies. The phylogeographic studies indicated that the populations of both species lacked a phylogeographic structure and exhibited high haplotype diversity, low nucleotide diversity and low Fst values among the biogeographic provinces; in combination with negative values on the neutrality test, this suggests that the populations of acanthocephalans are expanding. Paratenic hosts are key for the transmission from intermediate to definitive hosts in the generalist species. However, the inclusion of paratenic hosts does not play a principal role in the population genetic structure of S. hispida within its distribution along the coasts of Mexico.


Introduction
Parasitism is a highly successful lifestyle and has evolved independently at least 60 times in different groups of metazoans worldwide (Price, 1980;Poulin and Morand, 2004). Parasites have been traditionally divided into 2 major groups depending on their life cycle: generalists and specialists (Thompson, 1994(Thompson, , 2005. The generalist parasites use a wide range of definitive hosts, whereas the specialist parasites use a single definitive host to complete their life cycle. Under these 2 opposing strategies, generalist parasites infect a broad spectrum of hosts resulting in an optimal or suboptimal level of fitness, whereas specialist parasites prioritize a single optimal host in which fitness is maximized (Rigaud et al., 2010;Lievens et al., 2018). Some studies have suggested that parasite life cycle complexity (generalists vs specialists) could influence population genetic structure (Nadler, 1995;Criscione and Blouin, 2004;Barrett et al., 2008;Archie and Ezenwa, 2011). According to Li et al. (2014), a specialist parasite shows significantly less genetic flow; therefore, populations are less connected and are subdivided into smaller populations, leading to strong genetic differentiation. In some cases, the populations might experience bottlenecks, decreasing the effective population size. Moreover, specialist species are more sensitive to stochastic fluctuations that can cause local extinction. In contrast, a generalist parasite shows a high effective population size, high genetic flow and a population that is structured or panmictic. Moreover, a generalist parasite may show greater persistence of populations over the long term because the generalist may be less sensitive to stochastic fluctuations in any given resource as it is able to replace a scarce resource with another (see Sehgal et al., 2001;Brant and Ortí, 2003;Archie and Ezenwa, 2011;Li et al., 2014).
In the current study, we examined the sequences of cox1 from mitochondrial DNA of 2 closely related polymorphid species, a specialist species (H. corynosoma) and a generalist species (S. hispida), distributed sympatrically in 6 biogeographic provinces (California, Baja California, Pacific Lowlands, Tamaulipas, Veracruzan and the Yucatán Peninsula) of Mexico with the objective of characterizing and comparing the population genetic structure of 2 acanthocephalan species with opposing life history strategies.

Specimen collection
A total of 75 birds, and 36 fishes, were collected between October 2006 and December 2021 in 23 localities across 6 biogeographic provinces (California, Baja California, Pacific Lowlands, Tamaulipas, Veracruzan and the Yucatán Peninsula) of Mexico (Table 1; Fig. 1). Birds and fishes were dissected within the following 4 h, and their viscera were placed in separate Petri dishes containing a 0.75% saline solution and examined under a dissecting microscope. The acanthocephalans recovered were washed in 0.75% saline solution and placed overnight in distilled water at 4°C and subsequently preserved in 100% ethanol. Birds and fishes were identified using the field guide of Howell and Webb (1995) and Miller et al. (2005), respectively.

Morphological analyses
Selected adult acanthocephalans were gently punctured with a fine needle in the trunk, stained with Mayer's paracarmine, destained in 70% acid ethanol, dehydrated in a graded ethanol series, cleared in methyl salicylate and mounted in Canada balsam. Specimens were examined using a compound microscope Leica DM 1000 LED equipped with bright field (Leica, Wetzlar, Germany). The acanthocephalans were identified by conventional morphological criteria following Petrochenko (1958). In addition, descriptions of H. corynosoma and S. hispida were consulted as needed (Schmidt, 1973;Nickol et al., 2002;Amin et al., 2022). For scanning electron microscopy, 2 adult specimens of each species were dehydrated with an ethanol series, critical point dried, sputter coated with gold and examined with a Hitachi Stereoscan Model S-2469N scanning electron microscope operating at 15 kV from the Instituto de Biología, Universidad Nacional Autónoma de México (UNAM). Adult specimens were deposited in the Colección Nacional de Helmintos, under numbers CNHE: 11823 and 11824, Instituto de Biología, Universidad Nacional Autónoma de México, Mexico City

DNA isolation, amplification and sequencing
A total of 97 specimens, 41 identified as H. corynosoma and 56 as S. hispida were 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 Na 2 -EDTA (pH 8.0), 1% Sarkosyl and 0.1 mg mL −1 proteinase K. Following digestion, DNA was extracted from the supernatant using the DNAzol reagent (Molecular Research Center, Cincinnati, Ohio, USA) according to the manufacturer's instructions. The cox1 of the mitochondrial DNA was amplified using the forward primer 5 ′ -AGTTCTAATCATAA(R)GATAT(Y)GG-3 ′ and reverse primer 5 ′ -TAAACTTCAGGGTGACCAAAAAATCA-3 ′ (Folmer et al., 1994). Polymerase chain reactions (PCRs) (25 μL) consisted of 10 μL of each primer, 2.5 μL of 10× buffer, 2 mM MgCl 2 and 1 U of Taq DNA polymerase (Platinum Taq, Invitrogen Corporation, São Paulo, Brazil). PCR cycling parameters for the molecular marker consisted of denaturation at 94°C for 1 min, 35 cycles of 94°C for 1 min, 40°C for 1 min and 72°C for 1 min, followed by a post-amplification incubation at 72°C for 10 min. Sequencing reactions were performed using ABI Big Dye (Applied Biosystems, Boston, Massachusetts, 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 version 9.0.1 (Codoncode Corporation, Dedham, Massachusetts, USA) and submitted to the GenBank dataset (Table 1).

Alignments, population genetic structure and historical demographic
Newly obtained sequences in the current research of S. hispida were aligned with 22 other sequences of S. hispida (EF467867-868, HM636468-469, FJ824172-189) downloaded from GenBank (Table 1), forming a dataset of 78 sequences with 646 characters, and the new sequences of H. corynosoma were aligned with other sequences of H. corynosoma (EU189485-486, EU189488 and EF467869), downloaded from GenBank (Table 1), forming a dataset of 46 sequences with 644 characters. Sequences of each dataset were aligned separately using software ClustalW with default parameters implemented in MEGA version 7.0 (Kumar et al., 2016).
To analyse the molecular information in the framework of population genetics, we grouped individuals of H. corynosoma and S. hispida into populations considering the biogeographic Table 1. Specimens' information, collection sites (CS), host name; number of host examined/infected (prevalence of infection); number of specimens analysed (n); locality, geographical coordinates, GenBank accession number for specimens studied in this study  (Rozas et al., 2003). To examine haplotype frequency among the populations of H. corynosoma and S. hispida a statistical network was constructed independently, using the program PopART with the median joining algorithm (Bandelt et al., 1999). The degree of genetic differentiation among the populations was estimated using the fixation indices F st (Hudson et al., 1992), with the program Arlequin v.3.5 (Excoffier and Lischer, 2010).
To investigate the genetic variation among populations or within populations, the analysis of molecular variance was performed, considering genetic distance among the haplotypes using Arlequin v.3.5. To investigate the population history and demography, Tajima's D (Tajima, 1989) and Fu's F s (Fu, 1997) were calculated using DnaSP v.5.10 (Rozas et al., 2003). The values were considered significant when the P values were less than 0.05. The sample number for each locality corresponds with the same number in Fig. 1. Sequences in bold were generated in the current study.

Morphological identification
The acanthocephalans recovered from diverse definitive hosts such as herons, gulls, cormorants, pelicans and hawks on both coasts of Mexico show similar morphological characteristics compared with those assigned to S. hispida by García-Varela et al.

Population genetic structure and demographic analysis
The mitochondrial marker was successfully amplified for 56 S. hispida individuals and 41 H. corynosoma individuals. The complete alignment of the cox1 dataset contained 78 S. hispida individuals with a total length of 646 bp, whereas the cox1 dataset of H. corynosoma contained 46 individuals with a total length of 644 bp. No insertions or deletions were detected in any of the sequences, and when the sequences were translated into proteins, no stop codons were found. The haplotype network built for the generalist species S. hispida did not show a phylogeographic structure; in fact, of the  Table 1.
55 mtDNA haplotypes detected, 50 were unique (singlets), and 5 others were shared haplotypes (H4, H11, H17, H29 and H30). The most frequent haplotype (H4, n = 12) corresponded to specimens from 4 biogeographic provinces (Pacific Lowlands, Baja California, Veracruzan and the Yucatán Peninsula). Most of the identified haplotypes were separated from one another by 1, 2, 3 or up to 5 substitutions (see Fig. 4). The haplotype diversity was very high (H d = 0.970) and nucleotide diversity was low ( p i = 0.00555) among the populations from the 4 biogeographic provinces sampled (Pacific Lowlands, Baja California, Veracruzan and the Yucatán Peninsula) of Mexico. Neutrality tests (Tajima's D, −2.473 and Fu's F s , −34.252) were negative for all regions (see Table 2), indicating an excess of rare alleles greater than what would be expected under neutrality and suggesting a recent population expansion of S. hispida. The F st values were estimated to assess genetic differentiation among populations from the 4 biogeographic provinces analysed (Pacific Lowlands, Baja California, Veracruzan and the Yucatán Peninsula). Despite the large geographic distances, the F st values were low, ranging from −0.007 to 0.058 (Table 3), which indicates that the populations were poorly genetically differentiated from one another. The sampled definitive and paratenic host species of S. hispida were grouped into 8 families (Fig. 5). The host haplotype network did not show a pattern; 30 haplotypes were found in herons scattered throughout the network. In the paratenic hosts sampled (cichlid fishes), 11 haplotypes were detected (H3, H6, H7, H8, H10, H17, H30, H32, H33, H34 and H35), suggesting that the paratenic hosts are able to harbour and transmit diverse haplotypes to definitive hosts (Fig. 5).
The haplotype network of the specialist species H. corynosoma did not show a phylogeographic structure for the 42 mtDNA haplotypes detected; 41 were unique (singlets), and only 1 was shared (H2, n = 4) between 2 biogeographic provinces (Veracruzan and the Yucatán Peninsula). Most of the identified haplotypes were separated by several substitutions (see Fig. 6). The haplotype diversity (H d = 0.993) was very high and nucleotide diversity was low ( p i = 0.01166) among the populations from the 3 biogeographic provinces sampled (Pacific Lowlands, Veracruzan and the Yucatán Peninsula). Neutrality tests (Tajima's D, were negative for all regions (see Table 4), which indicates an excess of rare alleles greater than what would be expected under neutrality, suggesting a recent population expansion of H. corynosoma. The F st values were estimated to assess genetic differentiation among the populations from the 3 biogeographic provinces analysed (Pacific Lowlands, Veracruzan and the Yucatán Peninsula). Despite the large geographic distances, the F st values were low, ranging from −0.019 to 0.059 (Table 5), which indicates that the populations were poorly genetically differentiated from one another.

Discussion
To the best of our knowledge, S. hispida and H. corynosoma are 2 species that use piscivorous birds as definitive hosts and decapods as intermediate hosts, share a common ancestor within Polymorphidae and are distributed sympatrically along the coastline of Mexico (Guillén-Hernández et al., 2008;García-Prieto et al., 2010;García-Varela et al., 2013b). The generalist species S. hispida was described in Japan by Van Cleave in 1925, and since then it has been recorded in the body cavities of fishes and reptiles as a cystacanth (larval stage) and as an adult form in the intestines of diverse piscivorous birds in Asia, Europe and the Americas (see Amin et al., 2022). In Mexico, adults of S. hispida have been recorded in 15 piscivorous bird species  Fig. 1 and Table 1): adult male, whole worm (A); proboscis (B); male anterior region (C); adult male, whole worm (D) and adult female, whole worm (E). (García-Prieto et al., 2010;García-Varela et al., 2012). It is well known that helminths with a broad spectrum of definitive hosts show phenotypic plasticity in their morphological traits. Schmidt (1973), García-Varela et al. (2012) and Amin et al. (2022) documented the phenotypic plasticity of S. hispida recovered from diverse definitive hosts including in traits such as body size, leminisci, proboscis receptacle, testes, cement glands and the reproductive system in females. However, diagnostic characters such as proboscis shape, number of proboscis hooks and the presence of 2 fields of spines on the anterior trunk region in both sexes did not vary among specimens, including our specimens collected from the coasts of the Gulf of Mexico and the Pacific Ocean slopes (see Fig. 2A-E). In Brazil, the specialist species H. corynosoma was described as parasite of the yellowcrowned night heron (N. violacea) by Travassos in 1915. Since then, H. corynosoma has been recorded in Puerto Rico, the USA and Mexico in the same definitive host species (Cable and Quick, 1954;Nickol et al., 2002;Guillén-Hernández et al., 2008). Our specimens identified as H. corynosoma agree morphologically with those previously described by Nickol et al. (2002) and Guillén-Hernández et al. (2008). For example, an elongated trunk, swollen anteriorly; covered with spines on the anterior region of the trunk; a cylindrical proboscis with 16 longitudinal rows of 11-12 hooks each; a conical neck; ovoid testes located in the swollen portion of the trunk and long, tubular cement glands (see Fig. 3A-E).
The intraspecific genetic divergence estimated in the current study among the 78 isolates of S. hispida and the 46 isolates of H. corynosoma ranged from 0.00 to 1.5% and from 0.00 to 2.6%, respectively. These values of intraspecific genetic divergence are similar to those previously reported for isolates of polymorphid species such as Andracantha sigma (Presswell et al., 2017)     The haplotype network genealogy generated in this study based on cox1 sequences from S. hispida (generalist) and H. corynosoma (specialist) did not show a phylogeographic structure; therefore, the haplotypes could not be grouped into their own geographic clusters. In fact, the specimens of S. hispida were not correlated with their definitive host family (see Figs 4 and 5). The populations analysed for both species of acanthocephalans were classified into biogeographical provinces separated by geographical barriers as follows: mountains, the dry lowlands of the Isthmus of Tehuantepec, the Balsas Depression and the central Trans-Mexicana Volcanic Belt (Barrier et al., 1998;Ferrari et al., 2012;Morrone et al., 2017). Despite the large geographic distances, the F st values estimated among the populations of both species were very low (Tables 3 and 5), indicating that the populations were poorly genetically differentiated from each other; this can be explained by the migration patterns of birds along the coasts of Mexico. Historical events leave signatures in the DNA, and neutrality tests can infer the demographic history of populations. In both species analysed (generalist and specialist), the estimated values of Fu's F s and Tajima's D among the populations were negative (see Tables 2 and 4). In addition, high haplotype diversity and an excess of low-frequency haplotypes were detected in both networks (Figs 4 and 6), although the haplotypes differed from one another by fewer than 5 nucleotide substitutions. These findings confirmed that both populations of acanthocephalans had experienced rapid population growth in the past.
In this study, we found that 2 species of acanthocephalans with opposing life history strategies (generalist and specialist) showed similar population genetics patterns. This pattern was not consistent with the specialist-generalist variation hypothesis (SGVH), which predicts that: (i) populations of specialists may be less connected and more subdivided into smaller populations than generalists; (ii) specialists are expected to have lower effective population sizes than generalists and to be composed of populations with less gene flow and (iii) generalists will show high effective population sizes, high genetic flow and highly structured populations (see Dennis et al., 2011;Li et al., 2014). According to the SGVH, the populations of S. hispida (generalist species) resemble the specialist model. To date, only a few phylogeographic studies of generalist acanthocephalans have been conducted using cox1 as a molecular marker. For example, Profilicollis altmani (Perry, 1942), which has a broad distribution across North and South America and parasitizes multiple species of intermediate hosts and diverse species of marine birds that act as definitive hosts (gulls, ducks, sanderling and common tern), showed a lack of population genetic structure with high haplotype diversity and low nucleotide diversity, suggesting that P. altmani have experienced a period of rapid population growth in the past (see Goulding and Cohen, 2014). Similarly, Profilicollis novaezelandensis Brockerhoff and Smales (2002) is a parasite that has been recorded as an adult in gulls (Larus spp.) and oystercatchers (Haematopus spp.) and as cystacanths in the shore crab Hemigrapsus crenulatus (Milne-Edwards) distributed along the east coast of New Zealand's South Island (Brockerhoff and Smales, 2002); population genetic analyses from 50 P. novaezelandensis individuals from 8 localities showed a lack of population genetic structure with high haplotype diversity and low nucleotide diversity (Hay et al., 2018).
The 2 species of acanthocephalans analysed here are considered to be typical components of the helminth fauna of piscivorous birds in the Americas. Of the 19 piscivorous birds recorded as definitive hosts, 11 belong to the family Ardeidae, which represent 58% of the host diversity, suggesting that the ardeids could be the ancestral hosts of S. hispida and H. corynosoma with secondary and independent colonization events to other piscivorous birds. In particular, S. hispida was recorded in diverse piscivorous birds such as pelicans, cormorants, eagles, hawks, anhingas and royal terns with a lower level of infection (prevalence values <20%) than those of ardeid hosts (prevalence values from 80 to 100%), suggesting that these piscivorous birds may act as suboptimal definitive hosts.
The population genetic structures of S. hispida and H. corynosoma showed similar patterns. However, these species use different strategies to complete their life cycles. The main ecological difference between S. hispida and H. corynosoma is the inclusion of paratenic hosts in S. hispida (fishes from the families Cichlidae, Eleotridae, Characidae, Lutjanidae, Centropomidae, Sparidae, Scianidae and Paralichthyidae) (García-Prieto et al., 2010). The paratenic hosts act as a trophic bridge and facilitate the transmission between the intermediate host and definitive host, but apparently do not play a central role in the population genetic structure of S. hispida within its distribution along the coasts of Mexico.