Synopsis of the Species of Coccidians Reported in Marine Fish

Simple Summary Apicomplexa (Levine, 1970) are obligate parasites, of both vertebrate and invertebrate hosts, and may be responsible for important diseases. The majority of piscine apicomplexans are coccidians, which infect the cells of the alimentary tract and other extra-intestinal tissues. The goal of this work is to provide up-to-date information about the diversity of these parasites in marine fish. For this purpose, available information concerning Coccidia morphology and molecular profile, the hosts and the organs they infect, as well as their geographic distribution, was assembled. It is remarkable that from the 100 morphologically described species, only 6 have available genetic data. Abstract Eimeriid coccidians represent one of the largest groups of parasitic unicellular organisms and comprise many species of veterinary and medical importance. The aim of this work is to provide information, as complete as possible, concerning the oocyst phase of the species of coccidians (Apicomplexa: Eimeriorina) with four sporocysts, which were reported in marine fish. For each species, the accepted scientific name and respective authorities, its synonyms, reported hosts, site of infection (organ), and geographic distribution have been assembled. Available information on morphology (oocyst, sporocyst, and sporozoite) and GenBank accession numbers were also compiled. A total of 100 species of coccidians were described and reported from 60 families of marine fishes. Most species have been described from marine teleosts, with only 14 of the species described from marine elasmobranchs. Most of the species reported in marine fish belong to the genera Eimeria and Goussia, and only a handful belong to the genera Epieimeria, Calyptospora, and Crystallospora. Although marine coccidians have began to be described for more than one century, the collection of genetic data on marine fish coccidians only started in the 2010s and remains largely disconnected from the morphological analysis of specimens, which is available for only six formally described species.


Introduction
Eimeriid coccidians represent one of the largest groups of parasitic unicellular organisms and comprise many species of veterinary and medical importance [1]. Although the first fish coccidian studies were conducted in the last decade of the nineteenth century, still, little is known about the life cycles, morphology, phylogenetic, and infection patterns of these parasites, especially those that occur in marine fish.
Coccidians infecting marine fish belong to the genera Eimeria Schneider, 1875; Goussia Labbé, 1896; Epieimeria Dyková and Lom, 1981; Calyptospora Overstreet, Hawkins and Fournie, 1984; and Crystallospora Labbé, 1896. It should be stressed that the genus Goussia was erected to accommodate piscine coccidians with oocysts possessing four sporocysts, with two valves joined by a longitudinal suture. This genus has a controversial history, but the studies conducted in the last decades indicate that Goussia is a valid genus, which seems to represent the ancestral state of the Sarcocystidae, Eimeriidae, and Calyptosporidae [2][3][4]. Importantly, Benajiba et al. [5] suppressed the genus Epieimeria Dyková and Lom, 1981, on the basis of the intracellular gamogony, merogony, and sporogony of the type species E. anguillae. Observations conducted by Sitjà-Bobadilla et al. [6] further support the suppression of this genus. If these features are confirmed in all described species of the genus Epieimeria, they should be moved to the genus Eimeria.
Adding to a troubled taxonomy, the information on coccidians that infect marine fish is currently widely dispersed across both older and more recent scientific publications and, apart from the annotated list provided in the work of Dyková and Lom [7] from the early 1980s, no other attempt has been made to review all the information on fish coccidians. The aim of this work is to compile, as complete as possible and for the first time, the available information concerning the oocyst phase of the species of coccidians (Apicomplexa: Eimeriorina) with four sporocysts which has been reported from marine fish.
The specific objectives of this review were to provide: (i) an up-to-date list of species of marine piscine coccidians; (ii) a complete overview of their known hosts, geographic range, and infected organs; (iii) morphological characterization of each species and, whenever possible, drawings of their oocysts; (iv) a summary of their phylogenetic affinities; and (v) a discussion of the state of the art, identifying knowledge gaps and future research areas.

Materials and Methods
A bibliographic search was performed using all available databases and fields included on Web of Science (https://www.webofscience.com/wos/woscc/basic-search) with the following search string: (Coccid* or Goussia or Eimeria or Epieimeria or Crystallospora or Calyptospora) and fish.
For the purpose of constructing a list of species infecting marine fish, species that were given a generic but not specific assignations were not considered. Additionally, species for which a Stieda body was not referred, but that were still reported in genus Eimeria, were maintained in this genus until new data are available.
For each coccidian species, we assembled the accepted scientific name and respective authorities, eventual synonyms (syn), hosts, site of infection (organ), and geographic distribution. Regarding the geographic distribution, the precision of the available information in the reviewed manuscripts varied significantly (in some cases, a country was mentioned; in others, an ocean; and in others, a very specific location on the coast of a country). In cases for which sea(s) or ocean(s) were not referenced in the original publication, this information was added.
Available information on morphology and dimensions (oocyst, sporocyst, and sporozoite) is detailed according to what was reported in the reviewed publications. GenBank accession numbers, whenever available, were also compiled. Additionally, any other information that was deemed important was also supplied. Hosts' family and names (both vernacular and scientific names) were reported according to FishBase [8].
Measurements, unless otherwise stated, are provided in µm. Oocysts were redrawn whenever good drawings were available in the literature (Figures 1-4). In all presented drawings, scale bars are adapted to equal to 10 µm. Some drawings are not to scale.
Additionally, genetic data available on NCBI's GenBank (https://www.ncbi.nlm.nih. gov/genbank/) for the 18S rRNA region of piscine coccidians, including those of lineages recovered from marine fish which were not identified to the species level, were compiled for phylogenetic analysis. Based on previously published phylogenies and the recognized existence of different genetic groups [4], as well as results from BLAST (Basic Local Alignment Tool) analysis, five different phylogenetic trees were reconstructed using Bayesian inference, implemented in MrBayes v3.2.6 [9]. Sequence alignment was performed using MAFFT software [10], and sequence models of evolution were selected using jModelTest 2.1.6 [10]. Based on the BLAST analysis of the sequences from ingroups, either Isospora belli (DQ060661) or Intranuclear coccidium (AY728896) were selected as outgroups. Three available sequences were not considered due to their short length, all belonging to Goussia clupearum (MF468314, MF468308 and MT463285).

Phylogenetic Analysis
Phylogenetic affinities of marine piscine coccidians are depicted in Supplementary  Figures S1-S5. Most of the available sequences belong to Goussia clupearum ( Figure S1), forming the clupearum group described by Xavier et al. [62]. Many of the unidentified sequences of coccidians infecting marine fish clustered with piscine epicellular Goussia and Eimeria (Figures S2 and S3). As reported By Xavier et al. [85], most of the coccidians so far sequenced from elasmobranchs (both rays and sharks) were basal to many Eimeriorina species infecting other non-piscine vertebrates ( Figure S4). Additionally, two of such sequences were instead basal to piscine Eimeria ( Figure S3). Finally, two sequences of unidentified Coccidia clustered with nodular Goussia mostly sequenced by Rosenthal et al. [4] (Figure S5).

Discussion
A total of 100 species of coccidians were described and reported from 60 families of marine fishes. Most species have been described from marine teleosts, with only 14 species described from marine elasmonbrachs. While roughly only a handful of marine Coccidia belong to genus Epieimeria, Calyptospora, and Crystallospora, species of the genus Goussia and Eimeria seem to typically infect marine hosts.
Most of the Goussia and Eimeria infecting marine fish species were either only reported in the works where they were described, or have been subsequently reported in only one other study (ca. 65% of the species in each genus). Therefore, it is not surprising that the current knowledge on the geographic distribution of marine Coccidia is so limited. For example, the majority of species of Goussia (60%) have only been reported from the Atlantic region, and an important fraction (ca. 16%) have also only been reported from the Pacific. Likewise, the vast majority of species of Eimeria have either been reported from the Atlantic region (25%), the Mediterranean (26%), or the Pacfic Ocean (25%). Additionally, species that were more frequently reported, besides having a wider known geographic distribution, usually have more known hosts. For example, Goussia cruciata, Goussia clupearum, and Eimeria sardinae have been reported from the Mediterranean Sea and both the Atlantic and Pacific oceans, thus appearing to be fairly cosmopolitan. The three species also appear to be host generalists, with at least Goussia clupearum and Eimeria sardinae being known to infect hosts belonging to more than one family, and Goussia cruciata being known to infect fish from at least two genera within Carangidae. Goussia thelohani was also reported from the Mediterranean Sea and the Atlantic and Pacific oceans from two different fish families, despite only being reported in four studies. Eimeria southwelli infecting rays from at least two different families has also been reported from the Mediterranean Sea and the Atlantic and Pacific oceans. These observations lead us to conclude that neither geographic range nor host range are sufficiently studied in the vast majority of marine coccidians. However, it is important to note that the existence of cryptic diversity could be common in widespread coccidians, as was found for Goussia clupearum [62,63], for which several closely related lineages were found, with most specific to each host species. It is noteworthy that G. cluperaum also exhibits high morphological variability, indicating that it may be a species complex (see overview by Xavier and Saraiva [86] regarding the "clupearum" group). A similar scenario has also been proposed for G. cruciata, although genetic data are not available to test this hypothesis [75].
Although marine coccidians began to be described more than one century ago, genetic data on marine fish coccidians have only started to be produced since 2010s, with various genetic sequences presently available in GenBank. However, genetic analysis of marine fish Coccidia has been largely disconnected from the morphological analysis of specimens, and genetic data are only available for six formally described marine coccidian species. Though generally, genetic data are still largely missing for piscine Coccidia (both freshwater and marine), phylogenetic reconstructions based on 18S rRNA permitted raising the hypothesis that Coccidia infecting aquatic organisms are basal to those infecting terrestrial taxa [87], and that Calyptospora might have diverged earlier than Goussia and Eimeria [88]. Current knowledge shows that neither Eimeria nor Goussia are monophyletic. Specifically, piscine Goussia spp. were shown to be polyphyletic with different evolutionary lineages, with the species infecting the gut clustering mostly according to their development (i.e., nodular, dispersed, and epicellular) [4]. Goussia infecting other organs (Goussia leucisci isolated from the kidney of the roach and Goussia siliculiformis sequenced from the gut mesentery of the freshwater bream) were considered to have evolved independently [4]. Furthermore, several studies genetically characterized Goussia clupearum (the type species from the genus) from different marine fishes and tissues, indicating that it is possibly a complex of species which are sister taxa to the Calyptospora [62,63]. Interestingly, G. clupearum and Calyptospora exhibit high levels of intraspecific genetic variability [62,63,88].
Our phylogenetic reconstruction, combining all available sequences of marine Coccidia, showed that most unidentified genetic lineages likely belong to Goussia (n = 11) and have phylogenetic affinities with epicellular and nodular Goussia. Those clustering with epicellular Goussia included some Coccidia sequenced from other organs than the gut, with such lineages clustering together. Three sequences retrieved from several tissues of Pagrus caeruleostictus formed a basal clade to Calyptospora; however, they could correspond to a different Goussia. For example, Goussia clupearum seems to be closely related to this group. Only four unidentified coccidian sequences clustered within Eimeria, one likely being E. variabilis (see [89]). Finally, the lineages sequenced from elasmobranchs are the most enigmatic, some being basal to Eimeria (n = 2) and others being basal to other Eimeriorina Coccidia (including Eimeria and Schellackia), infecting marine/freshwater turtles, terrestrial reptiles, birds, and mammals [85]. Unfortunately, none of the Coccidia sequenced from elasmobranchs have been formally identified. However, these patterns suggest that Coccidia infecting this host group comprise older lineages than the ones infecting other vertebrates.

Conclusions
Our synopsis shows that the current knowledge regarding marine fish Coccidia is still limited. It seems that a taxonomic gap does exist, exaggerated by a shortage of taxonomists [90]. This is not unexpected due to the innacessible nature of the marine environement which hinders the access to fresh fish tissues which would enable species descriptions and accurate identifications. On the contrary, genetic data are starting to be available for many coccidians infecting marine fish; however, it is necessary to merge the gap between the genetic data being produced and appropriate morphological characterization. It is also worth noting that phylogenetic inferences are currently restricted to information from a single gene (18S rRNA) and that developing new genetic markers that amplify across Coccidia is therefore essential.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ani13132119/s1, Figure S1. Bayesian phylogenetic reconstruction of available sequences of Goussia clupearum, Calyptospora and other unidentified lineages based on 558 bp of the 18S rRNA gene and HKY+G sequence model of evolution. Figure S2. Bayesian phylogenetic reconstruction of available sequences of epicellular Goussia and closely related unidentified sequences, based on 508 bp of the 18S rRNA gene and GTR+G sequence model of evolution. Figure S3. Bayesian phylogenetic reconstruction of available sequences of Eimeria and closely related unidentified sequences, based on 569 bp of the 18S rRNA gene and GTR+G sequence model of evolution. Figure S4. Bayesian phylogenetic reconstruction of available sequences of Coccidia infecting elasmobranchs based on 515 bp of the 18S rRNA gene, using GTR+G sequence model of evolution. Figure S5. Bayesian phylogenetic reconstruction of available sequences of nodular and dispersed Goussia and closely related unidentified sequences, based on 565 bp of the 18S rRNA gene and GTR+G sequence model of evolution.