Host relationships and geographic distribution of species of Acanthobothrium Blanchard, 1848 (Onchoproteocephalidea, Onchobothriidae) in elasmobranchs: a metadata analysis

Abstract Species of Acanthobothrium have been documented as parasites of the spiral intestine of elasmobranchs. Results of a metadata analysis indicate that 114 species of elasmobranchs have been reported as hosts of 200 species of Acanthobothrium. The metadata analysis revealed that 3.7% of species of sharks and 14.9% of species of rays that have been reported as hosts to date; some species are parasitized by more than one species of Acanthobothrium. This work provides a Category designation, as proposed by Ghoshroy and Caira (2001), for each species of Acanthobothrium. These Category designations are a tool to facilitate comparisons among members of Acanthobothrium for descriptions of new species in the future.


Introduction
According to Last et al. (2016b), there are 34 families comprised of 516 valid species of sharks and 26 families that include 633 valid species of rays. Since that publication, six new species of sharks and rays were described by: Yokota and Carvalho (2017) (two species of rays), Vaz and Carvalho (2018) (one species of shark), Rutledge (2019) (one species of ray), Grace et al. (2019) (one species of shark) and Concha et al. (2019) (one species of ray). This brought the current number of recognized species to 517 species of sharks and 637 species of rays.
Elasmobranchs (sharks, skates and rays) are host to a great variety of parasites in nature, particularly helminths. Acanthobothrium Blanchard, 1848 (Onchoproteocephalidea) is the most diverse genus that has been reported as parasite of the spiral intestine of elasmobranchs (Caira and Jensen 2017). At the present time, 201 species of Acanthobothrium are considered to be valid (Maleki et al. 2013;Caira and Jensen 2017;Rodríguez-Ibarra et al. 2018;Franzese and Ivanov 2018;Maleki et al. 2019;Zaragoza-Tapia et al. 2019, 2020. The genus consists of species that exclusively parasitize elasmobranchs as adults and, in many cases, individual species are thought to parasitize only a single species of elasmobranch (Caira 2011;Caira and Jensen 2017). Therefore, the genus Acanthobothrium is an excellent model for future studies of hostparasite co-speciation.
The main goal of this work is to provide a revised checklist based on a metadata analysis of the host relationships of members of Acanthobothrium and their geographic distribution based on records that have been generated from different parts of the world. The checklist focuses on the 201 valid species of Acanthobothrium and reports correlated with the genera and species of elasmobranchs, and includes the geographical distribution of each.
The number of species of Acanthobothrium continues to grow and there are still regions of the world without a single report of this genus (see Figure 1). For some time, the process of distinguishing new species of Acanthobothrium from existing species has become more and more unwieldy because of the large number of species. As an identification tool, Ghoshroy and Caira (2001) developed a categorical method for identifying species for initial comparisons. Therefore, in order to provide an update to this method, categorical designations are provided in the present checklist for each species of Acanthobothrium in the manner proposed by Ghoshroy and Caira (2001). The categories are based on and obtained from the combination of four quantitative characters: total length of the worm; the number of proglottids comprising the strobila; the number of testes per proglottid; and symmetry of the ovarian lobes. This categorical designation allows parasitologists working with this genus to postulate a group of similar species, those of the same category designation, for comparison of a new species or to aid in the preliminary identification of known species. As an additional aid, in the checklist the accession number, if known, of type specimens of each species is provided.

Materials and methods
The checklist, updated until March 2020, was based on bibliographical information from two sources of information: 1. a compilation of the records of species of Acanthobothrium as originally described, complemented by information gathered from Global Cestode Database  and from recent compilation studies (e.g., Ghoshroy and Caira 2001;Campbell and Beveridge 2002;Fyler and Caira 2006;Caira and Jensen 2017); and 2. information for the distribution and taxonomy of elasmobranchs that integrated a bibliographical search using different databases of literature published to date (e.g., Del Moral-Flores et al. 2015;Last et al. 2016b;Merlo-Serna and García-Prieto 2016;Alves et al. 2017) and data from FishBase (Froese and Pauly 2019).
In the checklist, the species of Acanthobothrium are arranged in alphabetical order. The scientific names and geographic distribution of elasmobranchs have been updated based on Last et al. (2016aLast et al. ( , 2016b, Amaral et al. (2018) and Froese and Pauly (2019). The regional classification scheme of the geographic distribution of the hosts is according to Last et al. (2016b) with additional information from Froese and Pauly (2019). The following abbreviations are used for biogeographic regions (see Figure 1): Information for each species of Acanthobothrium presented herein includes the name of the species, authority (original description referenced in the literature cited), abbreviation of the name of the collection where specimens are deposited and the accession numbers of the specimens, followed by the status of the specimens (holotype, paratype, neotype, syntype or voucher). The acronym "NR" was used for data that are not reported in the original source. Localities (type or/and additional localities) were given and referenced in the literature cited. A Category designation was supplied for all species using the categorical method proposed by Ghoshroy and Caira (2001).
The categorical method was developed as a system of grouping species of Acanthobothrium based on the combination of four qualitative characters: the total length of worms-≤ 15 mm = S (short) or > 15 mm = L (long); the number of proglottids comprising the strobila-≤ 50 = F (few) or > 50 = M (many); the number of testes per proglottid-≤ 80 = F (few) or > 80 = M (many); and symmetry of the ovarian lobessymmetrical = S or asymmetrical = A. Of the possible combinations the following 10 Figure 1. Type localities of species of Acanthobothrium reported worldwide and the biogeographic regions (Last et al. 2016b) of the geographic distribution of their hosts (see Table 1). categories currently are recognized and coded as follows: 1 = SFFS; 2 = SFFA; 3 = LMMA; 4 = LMMS; 5 = LMFS; 6 = LMFA; 7 = LFFA; 8 = SMFS; 9 = LFFS; 10 = SMMS. This method limited the number of necessary comparisons required in the description between known species with new species assigned to the same Category. For this work, the categories and characteristics were used as in Ghoshroy and Caira (2001) and Fyler and Caira (2006) but the character values are as given in the original descriptions or as supplemented by the most recent taxonomic publications. In the Category designation, the type species is identified by number for this classification; the symbol "-" was used for the additional reports of species with additional hosts and/or localities.

Results
The information obtained from the metadata analysis (Table 1) is comprised of 336 reports of the 201 valid species of Acanthobothrium. The list includes the type host of each species, type locality, and additional hosts and/or localities. Five of the elasmobranchs that were reported as hosts of Acanthobothrium were only identified to genus and four others are reported as "cf." (= similar to) (see Table 1). The type localities where species of Acanthobothrium have been reported is shown in Figure 1. The currently known diversity of sharks comprises 517 species (34 families); of these, 19 species of sharks (eight families) have been reported to be parasitized by species of Acanthobothrium (Fig. 2). Eighteen of the 201 valid species have been described from sharks. The families of sharks that have the highest number of reports are Orectolobidae (three different species of Acanthobothrium), Heterodontidae (five species) and Triakidae (six species) (Fig. 2B). In contrast, currently known diversity of rays comprises 637 species (26 families); of these, 95 species (18 families) have been reported to be parasitized by species of Acanthobothrium (Fig. 3). Of the 201 valid species of Acanthobothrium, 182 have been described from rays. The families of rays that have the highest number of reports are Rajidae (20 species of Acanthobothrium) and Dasyatidae (70 species) (Fig. 3B).
Species of Acanthobothrium are not evenly grouped in the different categories. In Category 1 there are 55 species, 44 in Category 2, 19 in Category 3, 37 in Category 4, 17 in Category 5, 14 in Category 6, four in Category 7, four in Category 8, and three in Category 9. Although there is a Category 10, species in that category also are in grouped with those in Category 8 because their characteristics are thought to fall into both categories (Table 1). The categories of four species of Acanthobothrium were classified as unknown ("?") because the original descriptions do not have sufficient information for assignment in one of the ten categories (Table 1).

Discussion
Currently, 517 species of sharks have been described worldwide with 3.7% (19 of the 517 species) have been reported as hosts for species of Acanthobothrium (Fig. 2C). In contrast, 637 species of rays have been described with 14.9% (95 of the 637 species) have been reported as hosts for species of Acanthobothrium (Fig. 3C). Estimates of cestode diversity in elasmobranchs discussed by Caira (2011) assumes that the fauna of cestodes of a species of elasmobranchs does not vary substantially across in its distribution. Knowledge of life cycles are essential in understanding the distribution of species of Acanthobothrium; however, for this study it is assumed that the distribution of adults of these parasites normally is limited to that of its normal definitive host. Thus, it is hypothesized that the limits of the distribution of the host limits the species of its parasites to the same biogeographic regions proposed for the distribution of elasmobranchs by Last et al. (2016b). It is recognized that an infected elasmobranch Figure 2. Families of sharks: A number of species of sharks per family B number of species of sharks parasitized by species of Acanthobothrium. Note: The first number within parentheses corresponds to the number of species of shark that have been reported as hosts of Acanthobothrium and the second is the number of species that have been described from that Family C percentage of species of shark reported to be parasitized within the total number of families of sharks-note: Red color = parasitized; Blue color = not parasitized. could move outside of the region where it has been designated, but until an extension to its distribution has been reported, it must be assumed that the normal distribution for each species of parasite also is that same designated region. The information in the table will be subject to future research, not forgetting that there is a lack of knowledge of the life cycle of the species of Acanthobothrium; a partial life cycle of a single species Figure 3. Families of rays: A number of species of rays per family B number of species of rays parasitized by species of Acanthobothrium. Note: The first number within parentheses corresponds to the number of species of ray that have been reported as hosts of Acanthobothrium and the second is the number of species that have been described from that Family C percentage of species of rays reported to be parasitized within the total number of families of rays-note: Red color = parasitized; Blue color = not parasitized. has been reported (Holland and Wilson 2009). Publication of molecular sequences for more species will provide new discoveries in this subject.
The information in the Figures 1 and 4 indicates that there is an absence of reports from several regions of the world, such as ECA, ESA, WIO, ARC, and SOC. According to the percentages of species of elasmobranchs that have been reported as hosts of species of Acanthobothrium, we can infer that there are still many new species of Acanthobothrium to be discovered. In the GenBank database records, molecular sequences of only 16 of the 201 species of Acanthobothrium have been reported. However, more molecular information about species of Acanthobothrium is required for future analyzes, both for identification and life cycle studies; these would provide more solid information for delimiting distributions.
In Table 1, Acanthobothrium chilensis Rêgo, Vicente & Herrera, 1968, was included for reference, although it was described from a fish, Sarda chiliensis (Cuvier, 1832) (Perciformes: Scombridae) (see Rêgo et al. 1968). Extensive recent studies of this species of fish Luque et al. 2016) failed to report A. chilensis; there is only the report by Rêgo et al. (1968). The report of the host for this species of Acanthobothrium likely is an accidental infection and not a normal host.
According to Fyler et al. (2009) and Franzese and Ivanov (2018), species of Acanthobothrium appear to exhibit oioxenous specificity for their elasmobranch hosts. In the present metadata analysis, for species exclusively in elasmobranchs, 83% of the species of Acanthobothrium show remarkable host specificity for their definitive host, and thus, should be considered to be an oioxenous species. In contrast, 34 of the 200 species (17%) of Acanthobothrium have been reported in more than one species of elasmobranch (Table 1). However, with the metadata analysis of the distribution of the hosts and the reports of the species of Acanthobothrium, 45 of the type specimens of Acanthobothrium require confirmation of the host (Table 1) because some appear to be problematic identifications and other hosts were reported as "cf." or only as an unidentified member of a particular genus In addition, there are reports of species of Acanthobothrium that suggest misidentification of the parasites; these should reevaluated in future studies. To mention obvious cases, A. batailloni has been reported from the Mediterranean Sea and from the Pacific coast of Peru and Chile and A. brevissime has been reported from the Gulf of Mexico and the Pacific coast of Peru.
The categorical method developed by Ghoshroy and Caira (2001) was proposed in order to delimit the number of taxonomic comparisons when describing new species. Using the method of Ghoshroy and Caira (2001), which focused only on species from the Americas, Fyler and Caira (2006) later applied the same methodology to biodiversity data for species from other regions; those works are augmented by this study. Of the 201 known species of Acanthobothrium, 13 have been classified in more than one category (see category designations in Table 1) because some characteristics of those species overlap with those of more than one category (see descriptions found in Zschokke 1888 ;Linton 1890;Baer 1948;Alexander 1953;Euzet 1955;Riser 1955;Yamaguti 1959;Goldstein 1964;Williams 1969;Goldstein et al. 1969;Appy and Dailey 1973;Severino and Sarmiento 1979;Marques et al. 1997;Reyda 2008). This does not decrease the usefulness of the categorical method as a tool for the initial stages in identification.
Having more information, such as molecular sequences, could solve some problems in identification, such as the two cases mentioned above. A species of Acanthobothrium that has been assigned to more than one category suggests that the categories still need some refining, or it is an example of cryptic species that cannot be distinguished without molecular information. However, molecular information cannot replace morphological descriptions of species. One reason is the lack of material for sequencing of the vast majority of already-known species. Morphology also augments molecular data in studies of the phylogeny of platyhelminths (Zamparo et al. 2001;Littlewood 2008). A complete phylogenetic hypothesis based on total evidence (morphological and molecular data) such as that of Littlewood (2008) for any major group of cestodes is still distant. Until that time, a categorical method provides the easiest and most direct method for selection of a group of species similar to a new species of Acanthobothrium. This updated database includes the category designation for each species described to date will be an important tool for the future taxonomic studies.