Introduction
The family Enenteridae Yamaguti, 1958 comprises five genera, and most of its species have been reported from intestines of herbivorous marine teleost, mainly Kyphosidae (Bray & Cribb, Reference Bray and Cribb2001, Houston et al., Reference Huston, Cutmore and Cribb2019, Reference Huston, Cutmore and Cribb2022). The genus Enenterum was established by Linton in 1910 with Enenterum aureum from Kyphosus sectatrix (Linnaeus) from Florida as type species. Currently, the genus Enenterum includes a total of 11 species with worldwide distribution (Huston et al., Reference Huston, Cutmore and Cribb2022).
Gomes et al. (Reference Gomes, Fabio and Rolas1974) collected specimens of Enenterum from the intestine of Kyphosus sp. off Rio de Janeiro, Brazil, and identified them as Enenterum pimelopteri Nagaty, Reference Nagaty1942 a species originally described from Kyphosus cinerascens (Forsskål) (=Pimelopterus tahmel) from the Red Sea. A similar species, E. pseudaureum from K. sectatrix from Dakar, Africa, was described by Dollfus (Reference Dollfus1946). Manter (Reference Manter1947) redescribed E. aureum Linton, Reference Linton1910 from K. sectatrix (type host) and K. incisor (Cuvier) from Florida suggesting that probably E. pseudaureum was a synonym of E. pimelopteri. Afterward Bray (Reference Bray1978) considered E. pimelopteri reported from South Atlantic by Fischthal & Thomas (Reference Fischthal and Thomas1972) and Gomes et al. (Reference Gomes, Fabio and Rolas1974), and E. pseudaureum, as synonyms of E. aureum.
Collection of new specimens of Enenterum from Kyphosus incisor (Cuvier) from off Rio de Janeiro, revealed it to be a new species now described with molecular data.
Material and methods
Sample collection and morphological analysis
Two kyphosid fish acquired from fishermen were examined: one from Jurujuba Beach, Niterói (22°55′35″S, 43°06′00″W) and the other from Copacabana Beach, Rio de Janeiro (22°59′08″S, 43°11′18″W). The fish were measured and weighed, and the intestine was removed and examined in a saline medium under a stereomicroscope. The trematodes were collected alive, washed in saline solution at room temperature, and fixed in alcohol 70% or hot 4% formalin under slight coverslip pressure. Specimens were stained in Mayer’s paracarmine and Gomori’s trichrome and mounted in Canada balsam. Measurements are presented in micrometres, with the range followed by the mean in parentheses. Drawings were made with the aid of a drawing tube. Representative specimens were deposited in the Helminthological Collection of Instituto Oswaldo Cruz (CHIOC), Rio de Janeiro, Brazil. For comparative purposes, specimens of E. pimelopteri previously reported by Gomes et al. (Reference Gomes, Fabio and Rolas1974) and reassigned to E. aureum by Bray (Reference Bray1978), deposited at CHIOC 31012a-h, were reexamined. Comparative measurements are presented in Table 1.
* Considered E. aureum by Bray.
** Measured in the original drawing.
Genetic analysis
DNA extraction was performed using a QIAamp DNA Mini Kit (QIAGEN) according to the manufacturer’s instructions, and a set of primers were used to amplify different regions of the DNA. The rDNA region 28S was amplified by polymerase chain reaction (PCR) using the primers LSU5 (5′- TAGGTCGACCCGCTGAAYTTAAGCA- 3′) and 1500R (5′- GCTATCCTGAGGGAAACTTCG- 3′) (Tkach et al., Reference Tkach, Timothy, Littlewood, Olson, Kinsella and Swiderski2003). For partial 18S rDNA the primers SB3a (5’-GGAGGGCAAGTCTGGTGC-3’) and A27a (5’-CCATACAAATGCCCCCGTCTG-3’) (Hall et al., Reference Hall, Cribb and Barker1999) were used. For partial ITS1-5.8S-ITS2 region of the rDNA the BD1 (5′-GTCGTAACAAGGTTTCCGTA-3′) and BD2 (5′-TATGCTTAARTTCAGCGGGT-3′) primers were used (Luton et al., Reference Luton, Walker and Blair1992). Forward and reverse primers were used for all regions. PCRs were carried out using cycling parameters as previously described by these authors.
PCR products were analyzed by electrophoresis in 1.5% agarose in Tris-borate EDTA gels, stained with SyberGreen (Invitrogen, Eugene, Oregon, USA) and photographed under ultraviolet transillumination. Amplified PCR products were purified using ExoSap-IT (USB Products Affymetrix Inc., Cleveland, Ohio, USA). DNA cycle sequencing reactions were performed using the BigDye Terminator v.3.1 (Applied Biosystems, Foster City, CA, USA) and automated Sanger sequencing was done using the Sequencing Platform at Fundação Oswaldo Cruz (PDTIS/Fiocruz) in Brazil. Sequences of both strands generated (.ab1 files) were oriented in the same direction, aligned (CLUSTAL W) and edited by using the MEGA11 software (Tamura et al., Reference Tamura, Stecher and Kumar2021). The low-quality trailing ends were removed. Sequences were compared to others available in the GenBank database using the Basic Local Alignment Search Tool (BLAST) program from the National Center for Biotechnology Information server (http://www.ncbi.nlm.nih.gov/BLAST) (Altschul et al., Reference Altschul, Gish, Miller, Myers and Lipman1990). Evolutionary divergence estimates between sequences were conducted in MEGA11 using the Kimura 2-parameter (K2p) model (Kimura, Reference Kimura1980).
To examine phylogenetic relationships, nucleotide sequences were aligned using MEGA11. Bayesian inference phylogenetic trees were conducted using Monte Carlo Markov Chain analysis available in the BEAST v2.6.3 software (Bouckaert et al., Reference Bouckaert, Vaughan, Barido-Sottan, Duchêne, Fourment, Gavryushkina, Heled, Jones, Kuhnert and de Maio2019). Likelihood parameters set for the BI analysis were based on the Akaike Information Criteria test in jModelTest2 (Nylander, Reference Nylander2004). The selected model was the General Time-Reversible for 28S, and the Hasegawa-Kishino-Yano for the 18S and ITS1-5.8S-ITS2, employing the birth-death model. Posterior probabilities were calculated via 10,000,000 generations, sampling every 1000th tree. Tracer v1.7.2 (Rambaut et al., Reference Rambaut, Drummond, Xie, Baele and Suchard2018) was used to validate the convergence and mixing to ensure all effective sample size values greater than 200. Trees were presented as Maximum-Clade Credibility trees using the TreeAnnotator v2.6.3 software after discarding the first 10% as burn-in and visualized using the FigTree v1.4.4 (Rambaut et al., Reference Rambaut, Drummond, Xie, Baele and Suchard2018). For tree rooting, the best sequences used as outgroups were Pygidiopsis macrostomum Travassos, 1928 (KT877408) and Haplosplanchnus pachysoma (Eysenhardt, 1829) (LK932143) for 18S, Affecauda annulata Hall & Chambers, 1999 (FJ788501) and Endochortophagus protoporus Huston, Miller, Cutmore & Cribb, 2019 (MK396257) for 28S, and H. pachysomus (KY852459) and Schikhobalotrema acutum (Linton, 1910) (KY852465) for 5.8S-ITS2. The ITS1 region was not included in the phylogenetic analysis because the GenBank sequences for comparison only had 5.8S-ITS2. Sequences from GenBank that were used for the phylogenetic analysCis are listed in Table 2.
Results
Lepocreadioidea Odhner, 1905
Enenteridae Yamaguti, 1958
Enenterum Linton, 1910
Enenterum gomesae n. sp.
Syns: Enenterum pimelopteri of Gomes, Fabio & Rolas (1974),
re-identified as E. aureum by Bray (1978)
http://zoobank.org/urn:lsid:zoobank.org:pub:21E575C6-F226-4CD8-94EC-94EF015CB233
Description based on eight specimens: Body elongate, tapering at each end, 9.04–13.90 (11.99) long by 0.76–1.88 (1.38) wide (Fig. 1a). Tegument spinous. Eyespot pigment sparce in forebody. Oral sucker terminal, infundibuliform, bordered by 10 lobes disposed as: two pairs of dorsal lobes with anterior notches separated by a sagittal cleft; two lateral pairs, one at each side of the sucker, with anterior notches; one pair of strong ventral lobes, with pointed tip, separated by a deep central cleft (Figs. 1a, 2a). At the level of this central cleft a longitudinal groove, as an inverted “Y”, runs down to the base of the oral sucker (Figs. 1a, 2a). Oral sucker longer than wide, 0.64–0.99 (0.81) long by 0.33–0.88 (0.56) wide. Ventral sucker round 0.55–0.92 (0.72) long by 0.56–0.95 (0.74) wide, in anterior third of body. Prepharynx 0.45–0.80 (0.60) long. Pharynx well-developed 0.24–0.44 (0.32) long by 0.26–0.36 (0.32) wide. Sucker width ratio 1: 0.98-1.20 (1:1.12). Oesophagus short 0.46–0.62 (0.52) long. Intestine bifurcates anterior to cirrus-sac forming broad caeca with irregular contour that unite posterior to testes forming a single cecum, which opens through a rectum with funnel-shaped muscular sphincter into the anus (Figs. 1b, 3b). Excretory pore opens posterior to anus. The anus and excretory pore open in a terminal common cavity. Testes slightly lobed, posterior to midbody, intercaecal. Anterior testis 0.93–1.60 (1.06) long by 0.60–0.82 (0.68) wide; posterior one 0.87–1.38 (1.15) by 0.44–0.76 (0.63). Distance from posterior testis to posterior end of body 1.60–2.74 (2.23). Cirrus sac preacetabular, between the ventral sucker and caecal bifurcation contains tubular and coiled seminal vesicle, large pars prostatica and muscular ejaculatory duct (Figs. 1c, 2d). Ovary round, pretesticular, postequatorial, 0.36–0.52 (0.43) long by 0.33–0.52 (0.41) wide. Seminal receptacle round, large 0.22–0.40 (0.33) long by 0.30–0.35 (0.32) wide, posterior to ovary. Laurer’s canal present. Mehlis’ gland anterior to ovary (Figs. 2c, 3a). Uterus pre-ovarian; metraterm long, muscular, passes dorsal to acetabulum to open in the genital pore at the genital atrium. Vitellarium follicular extends from posterior end of ventral sucker to almost the body end (ventral and lateral to caeca in uterine region; ventral, lateral and dorsal to caeca posteriorly) (Figs. 3b-c). Eggs numerous 0.06-0.07 × 0.03-0.04 (0.06 × 0.03). In total, six new sequences were generated for this study: two partial 18S rDNA, two partial 28S rDNA and two ITS1-5.8S-ITS2 sequences. The 18S rDNA sequences of Enenterum gomesae n. sp. were 406 bp long in both sequences (GenBank OP829047 and OP829048), the 28S sequences were 1076 and 1075 bp long (GenBank OP829051 and OP829052) and the ITS1-5.8S-ITS2 sequences were 985 and 1002 bp long (GenBank OP829053 and OP829054). There was no genetic variation between the new sequences generated in all regions.
The partial 18S rDNA sequence of E. gomesae n. sp. indicated similarity of 99.02% with E. aureum (AY222124). The K2p distance between these species was 0.74%, with four divergent nucleotides in a 406 bp. The partial 28S rDNA sequence of E. gomesae n. sp. indicated 99.72% similarity with E. aureum (AY222232); the K2p distance was 0.28% with three divergent nucleotides in a 1075 bp. The 5.8S-ITS2 sequence of E. gomesae n.sp. indicated 95.87% similarity with Enenterum kyphosi Yamaguti, 1970 (ON228452) with K2p distance of 4.27%, with 17 divergent nucleotides in a 388 bp.
The Bayesian phylogenetic 18S rDNA tree showed that E. aureum was the closest species to the new sequence, with a node support of one, and in the same clade as Koseiria xishaensis Gu Shen, 1983. Both genera, belonging to Enenteridae, were separated from species of Lepocreadiidae Odhner, 1905 and Lepidapedidae Yamaguti, 1958 (Fig. 4). The topologies of the 28S rDNA and 5.8S-ITS2 trees were similar and the new sequences formed a clade with E. aureum, E. kyphosi and Enenterum petrae Huston, Cutmore & Cribb, 2022, with the exception of E. aureum for 5.8S-ITS2, for which there is no sequence available to date. The Enenterum clade was separated from other genera of the Enenteridae, including Koseiria Nagaty, 1942, Proenenterum Manter, 1954 and Enenterageitus Huston, Cutmore & Cribb, 2019 (Figs. 5-6).
Type host. Kyphosus incisor
Type locality. Jurujuba Beach, Niterói, RJ.
Site of infection. Intestine.
Intensity. Two fish with eight and 18 specimens each.
Additional material studied. CHIOC 31012a-h
Deposition of types. CHIOC 40451 a (holotype) and 40451b-h…. (paratypes) and 40452 a-j (voucher).
Etymology. The specific name of this species is in honour of Dr Delir Correa Gomes Maués da Serra Freire for her contribution to the study of Helminthology in Brazil.
Remarks
The main diagnostic characters of the new species include 10 lobes on the oral sucker, prepharynx two times longer than pharynx, testes slightly lobed, ovary round, and presence of a rectum with muscular funnel-shaped sphincter. E. pimelopteri previously reported by Gomes et al. (Reference Gomes, Fabio and Rolas1974) from Rio de Janeiro and reassigned to E. aureum by Bray (Reference Bray1978) is now considered E. gomesae n. sp. The species of Enenterum which have 10 lobes on the oral sucker include E. aureum, E. pimelopteri, Enenterum elongatum Yamaguti, 1970, E. kyphosi and Enenterum ghardaguensis Saoud & Ramadan, 1985. The new species is closer to E. aureum from K. sectatrix from Florida, which can be distinguished by having a longer prepharynx, while in E. aureum the prepharynx is about the same length of pharynx when extended. Enenterum gomesae n. sp. presents an oesophagus (absent in E. aureum), the testes are slightly lobed and ovary is round, while in E. aureum the testes are lobed and the ovary is slightly lobed. Besides this, E. gomesae n. sp. has a rectum with muscular funnel-shaped sphincter connected to the anus and the ventral sucker is longer than in E. aureum. Enenterum gomesae n. sp. can be distinguished from E. pimelopteri from K. cinerascens, from Red Sea by the prepharynx longer than oesophagus (in E. pimelopteri oesophagus is longer than prepharynx) and the testes of E. pimelopteri are entire (not lobed). The new species differs from E. elongatum, a parasite from K. cinerascens off Hawaii and Kyphosus sydneyanus (Günther) from Australia, by the size of oesophagus (300-507 vs. 460-620) and position of ovary near midbody, while in E. gomesae n. sp. it is postequatorial. Enenterum gomesae n. sp. differs from E. kyphosi originally found in K. cinerascens off Hawaii, by presenting a larger length of body (10.290–14.800 vs. 9.04–13.90), a larger prepharynx (0.50–0.70 vs. 0.12–0.45) and larger oesophagus (0.30–0.51 vs. 0.10–0.20), and by presenting testes slightly lobed (vs. deeply lobed in E. kyphosi). Enenterum ghardaguensis from K. cinerascens (=Pimelopterus tahmel) off the Red Sea differs from E. gomesae n. sp. by the smaller length of body, by presenting a small prepharynx (0.03–0.04 vs. 0.45–0.80 in E. gomesae n. sp.) and oesophagus (0.05 vs. 0.30–0.51 in E. gomesae n. sp.). The testes are deeply lobed, while in E. gomesae n. sp. testes are slightly lobed.
Discussion
The genus Enenterum was erected by Linton in 1910 with E. aureum from Kyphosus sectatrix from Florida as type species. Now, this genus comprises 12 valid species: E. pimelopteri, E. elongatum, E. kyphosi, Enenterum elsti Bray, 1978, Enenterum prudhoei Bray, 1978, Enenterum mannarense Hafeezullah, 1980, E. ghardaguensis, Enenterum stinkvis Bray, 1986, Enenterum tongaatensis Bray, 1986, E. petrae and E. gomesae n. sp. (Nagaty, Reference Nagaty1942; Yamaguti, Reference Yamaguti1970; Zaidi & Khan, Reference Zaidi and Khan1977; Bray, Reference Bray1978; Hafeezullah, Reference Hafeezullah1980; Saoud & Ramadan, Reference Saoud and Ramadan1985; Bray, Reference Bray1986; Huston et al., Reference Huston, Cutmore and Cribb2022). Two other species, Enenterum minutum Yadav, 1977 and E. theraponii have been considered with a doubtful status (Bray, Reference Bray1986).
The Bayesian-inference analysis of the 28S rDNA dataset resulted in a phylogram in which the four Enenterid genera (Enenterum, Koseiria, Proenenterum and Enenterageitus) are in different clades, similar to that observed by Huston et al. (Reference Huston, Cutmore and Cribb2019). In our 5.8S-ITS2 Bayesian-inference analysis, this pattern was also observed. However, the genera closest to Enenterum was Koseiria in the 28S phylogenetic tree, while for the 5.8S-ITS2 analysis, it was Enenterageitus. Enenterogeitus huxleyi was transferred from Koseiria by Huston et al. (Reference Huston, Cutmore and Cribb2019). These four genera can be easily differentiated by the morphology of oral sucker, presence/absence of anus and presence/absence of muscular post-oral ring. In our phylogenetic analyses of 28S rDNA, E. huxleyi was found sister to, but distinct from, Proenenterum, similar to that demonstrated by Huston et al. (Reference Huston, Cutmore and Cribb2019, Reference Huston, Cutmore and Cribb2022). However, our ITS analysis showed E. huxleyi as sister to the Enenterum clade. Future studies are suggested to better understand the relationship between these genera.
One of the main characteristics of the genus Enenterum is the shape of the oral sucker with variable number of lobes in the anterior margin. Some species have 10 lobes (E. aureum, E. pimelopteri, E. elongatum, E. kyphosi, E. ghardaguensis and E. gomesae n. sp.), others have eight (E. stinkvis and E. prudhoei), seven (E. elsti), six (E. petrae and E. mannarense), and two lobes (E. tongaatensis).
Species with eight oral lobes like E. prudhoei were reported from southwestern Indian Ocean and E. stinkvis from Neoscorpis lithophilus (Gilchrist & Thompson) from South Africa. Enenterum elsti with about seven irregularly conical projections and E. tongaatensis with four lobes were also described from N. lithophilus from South Africa. Enenterum petrae which appears to have three-lobed when protracted and six-lobed oral sucker when retracted was described from K. vaigiensis from Australia. Enenterum mannarense also described with six pointed oral lobes was found in Kyphosidae from Australia and India. Enenterum theraponii described by Zaidi & Khan (Reference Zaidi and Khan1977) from the intestine of Terapon jarbua (Forsskål, 1775) from the Arabian was considered incertae sedis by Gibson & Bray, Reference Gibson and Bray1982.
Considering the closest species with 10 oral lobes, E. aureum was considered a cosmopolitan species based on their attributed wild geographical distribution ranging from Gulf of Mexico, Caribbean Sea to French Polynesia, Great Barrier Reef, Indian Ocean and Tropical Eastern Pacific (Bray & Cribb Reference Bray and Cribb2001, Reference Bray and Cribb2002). The specimens from Australia and French Polynesia presented prepharynx smaller than pharynx, smaller bodies length, ventral suckers, testes and ovary, apart from their geographical distribution. The full picture of the entire concept of E. aureum would include some populations from the Western Indo-Pacific but also from the Eastern Indo-Pacific realms that we believe need to be revised.
The concept of species delineation over geographic range was discussed by Huston et al. (Reference Huston, Cutmore, Miller, Sasal, Smit and Cribb2021) considering Gorgocephalus yaaji Bray & Cribb, 2005, which parasitises kyphosid fish in an expansive marine ecoregion stretching from the east coast of Africa to Australia and French Polynesia (see figure 1 in Huston et al., Reference Huston, Cutmore, Miller, Sasal, Smit and Cribb2021). They reported molecular variation suggesting the possibility of multiple species with specimens morphologically indistinguishable from G. yaaji. They concluded that additional specimens collected between Australia and South Africa would be necessary to split G. yaaji into multiple morphologically cryptic species. The marine ecoregion reported for G. yaaji shares not only the Western, Central Indo-Pacific and Eastern Indo-Pacific realms, after the Marine Ecoregions of the world (Spalding et al., Reference Spalding, Fox, Allen, Davidson, Ferdaña, Finlayson, Halpern, Jorge, Lombana, Lourie, Martin, Mcmanus, Molnar, Recchia and Robertson2007), giving additional evidence for future new species to be described.
Enenterum gomesae n. sp., for instance, is described from the lowest level of the Tropical Southwestern Atlantic realm, far distant from the type locality of E. aureum from Florida, which is in the Tropical Northwestern Atlantic realm, both well-separated by the North Brazil Shelf (see maps in Spalding et al., Reference Spalding, Fox, Allen, Davidson, Ferdaña, Finlayson, Halpern, Jorge, Lombana, Lourie, Martin, Mcmanus, Molnar, Recchia and Robertson2007). We believe that the E. aureum population described from Western Indo-Pacific and the Eastern Indo-Pacific must contain several cryptic species, as they do not present monophyly and have a variety of hosts. That’s why we focused on the North Atlantic original description of E. aureum which is probably distributed only in the Atlantic and the other E. aureum–type worms can constitute a group of cryptic species, with few morphological differences.
Bray (Reference Bray1978) considered E. pimelopteri, with 10 oral lobes, reported by Fischthal & Thomas (Reference Fischthal and Thomas1972) and Gomes et al. (Reference Gomes, Fabio and Rolas1974) from Senegal and Brazil, respectively, a synonym from E. aureum. In the Pacific and Atlantic Oceans, additional references included Winter (Reference Winter1957), Sogandares-Bernal (Reference Sogandares-Bernal1959), Overstreet (Reference Overstreet1969) and Pérez-Ponce de León et al. (Reference de León G, Garcia-Prieto and Mendoza-Garfias2007).
A review of the specimens reported by Gomes et al. from Rio de Janeiro showed that although the oesophagus was contracted in their specimens, in the fresh material now collected from K. incisor the oesophagus ranged from 0.46 to 0.62 (0.52). The testes of E. gomesae n. sp. are much larger than in the type of E. aureum from Florida based on its original figure (0.93–1.6 vs. 0.70). The ratio between the length of the ventral and oral suckers also differed, being 1:1.25–1.66 in E. aureum and 1:0.98–1.20 (1.12) in E. gomesae n. sp. Additionally, the results of our molecular phylogenetic analyses of 18S and 28S rDNA place Enenterum gomesae n. sp. as sister of E. aureum from K. vaigiensis from French Polynesia (Table 2). Therefore, we describe here that Enenterum gomesae n. sp. representing the only species described so far in South America. Other species of Enenterum observed in 28S rDNA and the 5.8S-ITS2 trees were in the same clade as Enenterum gomesae n. sp. but on different and well-supported branches.
Enenterum gomesae n. sp. is described based on morphological differences, genetic data and distribution on well separated marine ecoregion realms. New sequences of E. aureum from type–host and locality are necessary for future comparison with E. gomesae n. sp. and entire concept of E. aureum. To date, there is scarce molecular data available for Enenterum species, limiting the understanding of the phylogenetic relationship of this family. The new sequences generated, the partial 28S and 18S rDNA genes and the ITS1-5.8S-ITS2 of E. gomesae n. sp., contribute to further comprehension of this group.
Acknowledgements
We are grateful to the RPT01A/FIOCRUZ sequencing facility for their technical support.
Financial support
The present study was supported financially by the Oswaldo Cruz Foundation (PAEF no. IOC-023-FIO-18-2-4 and IOC-008-FIO-22-2-42). AGLO is fellow from “Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ)”.
Competing interest
The author(s) declare none.
Ethics
Collection of fish was authorized by the Brazilian Institute of Environment and Renewable Natural Resources (IBAMA, license no. 15898-1).