Hexabothriid monogeneans from the gills of deep-sea sharks off Algeria, with the description of Squalonchocotyle euzeti n. sp. (Hexabothriidae) from the kitefin shark Dalatias licha (Euselachii, Dalatiidae)

Sharks were obtained from fishermen in Dellys (36° 55’ N; 3° 53’ E), Cap Djenet (36° 43’ N; 3° 36’ E), Bou Haroun (36° 40’ N; 4° 40’ E), and Cherchell (36° 37’ N; 2° 11’ E). All four localities are on the Mediterranean coast within 100 km near Algiers, Algeria and thus results are not detailed according to the localities. The following deep-sea shark species were examined for gill monogeneans from 2009 to 2015: gulper shark, Centrophorus granulosus (Bloch et Schneider, 1801): 27 specimens; little gulper shark, Centrophorus uyato (Rafinesque, 1810): 39; velvet belly, Etmopterus spinax (Linnaeus, 1758): 67; little sleeper shark, Somniosus rostratus (Risso, 1827): 19; blackmouth catshark, Galeus melanostomus Rafinesque, 1810: 189; lesser spotted dogfish, Scyliorhinus canicula (Linnaeus, 1758): 261; bluntnose sixgill shark, Hexanchus griseus (Bonnaterre, 1788): 3; and kitefin shark, Dalatias licha (Bonnaterre, 1788): 100. In addition, two pelagic shark species were examined: thresher, Alopias vulpinus (Bonnaterre, 1788), 7 specimens, and blue shark, Prionace glauca (Linnaeus, 1758), 53. Sharks were collected as fresh as possible, photographed and immediately brought back to the laboratory for examination. Identification was done according to usual keys (Fischer et al., 1987). The parasitological survey and fish identifications were done by HK.

Fig. 1

The gills were removed and observed in filtered seawater for mono-geneans. Monogeneans, located using a stereo-microscope were removed alive (dead for the few specimens from H. griseus) from between the gill lamellae and were studied either directly or fixed, slightly flattened, between a slide and cover slip. Monogeneans were fixed either with ethanol or Bouin’s fixative. Specimens were stained with carmine, cleared in clove oil and mounted in Canada balsam. Specimens for molecular analysis were collected in 95 % ethanol. All drawings were made with the help of an Olympus BH-2 microscope drawing tube. Drawings were scanned and redrawn on a computer with Adobe Illustrator. Measurements are in micrometres.

We used a QIAmp DNA Micro Kit (Qiagen) to extract DNA. Eution was performed in 60μL. The specific primers JB3 (=COI-ASmit1) (forward 5'-TTTTTTGGGCATCCTGAGGTTTAT-3') and JB4.5 (=COI-ASmit2) (reverse 5'-TAAAGAAAGAACATAATGAAAATG-3') were used to amplify a fragment of the COI gene (Bowles et al., 1995; Littlewood et al., 1997). The PCR reaction was performed in 20 μl, containing 1 ng of DNA, 1 × CoralLoad PCR buffer, 3 mM MgCl2, 66 μM of each dNTP, 0.15 μM of each primer, and 0.5 units of Taq DNA polymerase (Qiagen). The amplification protocol was: 4’ at 94 °C, followed by 40 cycles of 94 °C for 30’’, 48 °C for 40’’, 72 °C for 50’’, with a final extension at 72 °C for 7’. Sequences were edited with CodonCode Aligner software version 3.7.1 (CodonCode Corporation, Dedham, MA, USA), compared to the GenBank database content with BLAST, and deposited in GenBank under accession numbers KX389260 – KX389262. Trials to obtain 28S partial sequences with the routine method previously used for other polyopisthocotylean monogeneans (Justine et al., 2013) were unsuccessful.

Among the 765 sharks examined over six years, belonging to ten species, only two species had monogeneans on their gills. Hexanchus griseus had Protocotyle grisea, and Dalatias licha had a new species of Squalonchocotyle.

Our specimens were not in optimal state of conservation because these sharks were not fresh; however, the sclerotised parts could be observed and measured. Measurements (in parenthesis, measurements in Justine, 2011 for comparison): anterior sclerites 1,480 – 2,054 (1,680 – 1,720); median sclerites 1,850 – 2,498 (1,950 – 2,550); posterior sclerites 1,795 – 2,331 (1,820 – 2,330); hamulus outer length 89 – 96 (66 – 88); hamulus inner length 74 – 85 (70 – 85).

Fig. 2

Type host: Hexanchus griseus Bonnaterre, 1788 Type locality: Naples, Italy (Cerfontaine, 1899)

Additional localities: Trieste (Italy) (Cerfontaine, 1899); Sète (France) (Maillard & Oliver, 1966; Euzet & Maillard, 1974); near Algiers (Algeria) (this paper).

Specimens examined: 7 specimens from 3 host fish.

Prevalence in Algeria: 100 % (3/3).

Material deposited: MNHN, slides HEL558.

Measurements of our specimens from Algeria are consistent with an identification with P. grisea and allow to differentiate the specimens from the two only other species in the genus, namely Protocotyle taschenbergi (Maillard et Oliver, 1966) Euzet et Maillard, 1974 and Protocotyle euzetmaillardiJustine, 2011 (Maillard & Oliver, 1966; Euzet & Maillard, 1974; Justine, 2011). Algeria is a new geographical record for the species.

Based on 32 specimens; measurements in Table 2, including separate measurements for holotype and means for all specimens. Body elongate, slender, haptor wider than body. Haptor symmetrical, armed with six suckers, each provided with hook-shaped sclerite, and appendix bearing single pair of terminal suckers and single pair of hamuli, each with one sclerite. Haptoral sclerites in 3 pairs arranged symmetrically, each with same shape and with point at right-angles to distal end of sclerite shaft; median sclerites slightly longer than those of anterior and posterior pairs. Appendix elongate, directed anteriorly in flattened specimens. Pair of hamuli with V-shaped root situated near distal end of appendix. Pair of terminal suckers oblong.

Anterior sucker terminal. Pharynx subspherical. Oesophagus short. Caeca internally moderately diverticulate, confluent in posterior part of body, end as two short caeca, one which extends into haptor and one into appendix.

Testes numerous, occupy intercaecal area of posterior part of body, end posteriorly before confluence of caeca. Single sperm duct (vasa efferentia) well visible from testes to seminal vesicle. Seminal vesicle, begins just anterior to oötype, convoluted, thinwalled, contains spermatozoa, continues anteriorly and connects with cirrus; no posterior lobe. Cirrus elongate, unarmed, connects with genital atrium. Prostatic glands not seen. Genital atrium ventral, median, just posterior to bifurcation of caeca.

Ovary located at mid-length of body proper; proximal part of ovary slightly branched; descending and ascending ovarian parts straight; ovary terminates as slender canal superposed to seminal receptacle. Connections of terminal ovary, anterior part of seminal receptacle, posterior part of ovovitelloduct, posterior part of median vitelloduct and genitointestinal canal apparently all located just anteriorly to seminal vesicle. Ovovitelloduct convoluted, without diverticulum, connects anteriorly with oötype. Seminal receptacle cylindrical, oblique with anterior connection. Two lateral vitelloducts unite to form posteriorly directed median vitelloduct, with coil. Oötype wall with longitudinal rows of large cells (‘oötype côtelé’ of Euzet and Maillard 1974). Mehlis’ glands surround oötype. Oötype anteriorly joins uterus. Uterus straight, contains few eggs, ends anteriorly in genital atrium. Two vaginal openings, located just posteriorly to genital atrium or at the same level; anterior portion of vaginae often widened, filled with spermatozoa; posterior portion not well visible.

Eggs fusiform, elongate, operculum not seen, with two polar filaments.

We obtained COI sequences, 396 bp in length, for three specimens; the sequences differed between them by 4 and 6 nucleotide (1 – 1.5 %). A GenBank BLAST of the sequences showed that the closest species were polystomatid polyopisthocotylean monogeneans. These sequences were widely different (20 – 30 %) as polystomatids and hexabothriids are not closely related family. COI sequences are generally appropriate for distinguishing species; in the absence of any other sequence of hexabothriid monogeneans in databases, further comments are useless. Our sequences of S. euzeti n. sp. might be useful only when other hexabothriid sequences are available.

Type-host: Dalatias licha (Bonnaterre, 1788) (Dalatiidae). Type-Locality: Off Dellys (36° 55’ N; 3° 53’ E), Algeria.

Additional localities: Off Cap Djenet (36° 43’ N; 3° 36’ E), off Bou Haroun (36° 40’ N; 4° 40’ E), off Cherchell (36° 37’ N; 2° 11’ E), Algeria; all these localities are within 100 km of Algiers.

Site of infection: gills

Comparative material observed: One slide of Squalonchocotyle cerfontaini collected by Claude Maillard and deposited in the MNHN collections, MNHN 711H-Ti 52 (measurements in Table 1). Prevalence: 85/100 (85 %).

Etymology: named in honour of Professor Louis Euzet, famous parasitologist and author of major works on hexabothriids, who examined the specimens and confirmed their interest.

Species included in Squalonchocotyle

Species attributed to Squalonchocotyle Cerfontaine, 1899 include: S. borealis (Van Beneden, 1853), the type-species, and S. abbreviata (Olsson, 1876) Cerfontaine, 1899, S. cerfontainiMaillard, 1970, S. centrophoriMaillard, 1970, S. laymaniYamaguti, 1958, S. mitsukuriiKitamura, Ogawa, Taniuchi et Hirose, 2006, S. rajaeBrinkmann, 1971, S. spinacis (Goto, 1894), S. squaliMacCallum, 1931, and S. tropai (Tendeiro et Valdez, 1955) (Van Beneden, 1853; Olsson, 1876; Goto, 1894; Cerfontaine, 1899; MacCallum, 1931; Tendeiro & Valdez, 1955; Yamaguti, 1958; Maillard, 1970; Brinkmann, 1971; Kitamura et al., 2006)

Boeger & Kritsky (1989) included only four species in the genus: S. borealis, S. cerfontaini, S. centrophori, and S. squali. They considered that S. somniosi (Causey, 1926) was a synonym of S. borealis, but did not comment on the other species they considered as “unconfiired”.

Kitamura et al. (2006) apparently followed Boeger & Kritsky (1989) when they considered their new species S. mitsukurii as the fifth species of the genus. They commented that the taxonomic position of S. spinacis was uncertain because the type-specimens were lost.

The list of species of Squalonchocotyle in WoRMS (Bray, 2004) includes nine species, i.e. the ten listed above minus S. rajae.

We provide here a few remarks about S. tropai. The species was described as Erpocotyle tropai by Tendeiro and Valdez in 1955, from Squalus acanthias (designated as S. fernandinus, now considered a synonym (Froese & Pauly, 2016)) off Luanda, Angola, and never mentioned or redescribed again in the scientific literature. However, we found that Maillard (1966) in his unpublished thesis, described new specimens from the same host, collected off Sète, Mediterranean Coast, France (Maillard, 1966); his measurements are included in Table 1. Maillard did not examine the type-specimens and wrote that he could only compare with photographs (the origin and whereabouts of these photographs is unknown; the original description by Tendeiro and Valdez includes only drawings). Euzet and Maillard (1974) claimed that the types of species described by Tendeiro and Valdez were lost. Unfortunately, the slides prepared and described by Claude Maillard were not located in the Euzet collection (13,000 slides, now in MNHN, Paris) and should probably be considered lost. Maillarďs thesis (1966) should be considered unpublished for nomenclatural purposes. Euzet and Maillard (1974) used the binomial Squalonchocotyle tropai but did not formally indicate that they made a new combination for this species, but it is likely that they were the authors of the current combination, as S. tropai (Tendeiro et Valdez, 1955) Euzet & Maillard, 1974; we did not find it in earlier published works.

The characteristic oötype with longitudinal rows of cells (‘oötype côtelé’ of Euzet and Maillard, 1974) is found only in three hexabothriid genera, including Protocotyle Euzet et Maillard, 1974, Rajonchocotyle Cerfontaine, 1899 and Squalonchocotyle. This was considered a synapomorphy uniting these three genera (Boeger & Kritsky, 1989). Our specimens have the characteristic oötype and all characters listed for Squalonchocotyle (Boeger & Kritsky, 1989, and in Justine, 2011) i.e. distal cirrus unarmed, ovary branched in its proximal part, two egg filaments, vaginal ducts parallel, seminal receptacle present, and thus belong to the genus.

In the following discussion, we do not consider S. rajae; whether the species is valid and is really a member of Squalonchocotyle is an interesting question, since it is the only member of the genus described from rays (Raja smirnovi, R. rosispinis and Breviraja isotrachys) (Brinkmann, 1971); this is outside of the scope of this paper, but we are confident that the new species described here is distinct from S. rajae, on the basis of very different hosts (Rays vs Sharks) and widely separate localities (North Western Pacific vs Mediterranean).

We found that species of Squalonchocotyle can be separated into two groups according to body length: a group of seven relatively small species includes S. abbreviata, S. centrophori, S. cerfontaini, S. mitsukurii, S. spinacis, S. squali, and S. tropai (Table 1); a group of relatively large species includes S. borealis, S. laymani and S. euzeti n. sp. (Table 2). It cannot be excluded, however, that some of the small species were described from immature specimens, as it was the case for Mobulicola dubium (Euzet & Maillard, 1974) Patella & Bullard, 2013 (Euzet & Maillard, 1967; Patella & Bullard, 2013).

Squalonchocotyle euzeti is differentiated from S. laymani by much longer sclerites (ca 2,000 vs ca 600), different hosts (Dalatias licha vs Mustelus manazo) and widely separated localities (Mediterranean vs Japan). We measured sclerites on the figures of S. borealis by Cerfontaine, and found that they were of similar size to S. euzeti. Differential characters include body length (7 – 21 mm vs 25 – 30 in S. borealis) and, more importantly, body width (777 – 1,813 vs 3,000 – 4,000 in S. borealis) which gives to S. euzeti n. sp. a characteristic slender body. Since our specimens were flattened, we consider that their slender body is a genuine condition and not a consequence of insufficient flattening. Therefore, we consider that the slender body separates S. euzeti from S. borealis. In addition, the hosts are different (D. licha vs Somniosus microcephalus) and the localities are separate (Mediterranean vs Northern Atlantic).