Monogeneans

Molecular barcoding of fish

Total genomic DNA was isolated using a QIAamp DNA Mini Kit (Qiagen, Courtaboeuf, France), as per the manufacturer’s instructions. The 5′ region of the mitochondrial cytochrome c oxidase subunit I (COI) gene was amplified with the primers FishF1 (5′ – TCAACCAACCACAAAGACATTGGCAC – 3′) and FishR1 (5′ – TAGACTTCTGGGTGGCCAAAGAATCA – 3′) [75]. PCR reactions were 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 min at 94 °C, followed by 40 cycles at 94 °C for 30 s, 48 °C for 40 s, and 72 °C for 50 s, with a final extension at 72 °C for 7 min. PCR products were purified (Ampure XP Kit, Beckman Coulter) and sequenced in both directions on a 3730 × l DNA Analyzer 96-capillary sequencer (Applied Biosystems, Foster City, CA, USA). We used CodonCode Aligner version 3.7.1 software (Codon Code Corporation, Dedham, MA, USA) to edit sequences, compared them to the GenBank database content with BLAST, and deposited them in GenBank under accession numbers MT666082–MT666086. Species identification was confirmed with the BOLD identification engine [68].

COI sequences of monogeneans

Total genomic DNA was isolated using a QIAmp DNA Micro Kit (Qiagen). The specific primers JB3 (=COIASmit1) (forward 5′ – TTTTTTGGGCATCCTGAGGTTTAT – 3′) and JB4.5 (=COI-ASmit2) (reverse 5′ – TAAAGAAAGAACATAATGAAAATG – 3′) were used to amplify a fragment of 402 bp of the COI gene [11, 55]. PCR reaction was performed in 20 μL of a mixture containing 1 ng of DNA, 1× CoralLoad PCR buffer, 3 mM MgCl₂, 0.25 mM dNTP, 0.15 μM of each primer, and 0.5 units of Taq DNA polymerase (Qiagen). Thermocycles consisted of an initial denaturation step at 94 °C for 2 min, followed by 37 cycles of denaturation at 94 °C for 30 s, annealing at 48 °C for 40 s, and extension at 72 °C for 50 s. The final extension was conducted at 72 °C for 5 min. PCR products were purified (Ampure XP Kit, Beckman Coulter) and sequenced in both directions on a 3730 × l DNA Analyzer 96-capillary sequencer (Applied Biosystems, Foster City, CA, USA). We used CodonCode Aligner version 3.7.1 software (Codon Code Corporation, Dedham, MA, USA) to edit sequences, compared them to the GenBank database content with BLAST, and deposited them in GenBank under accession numbers MT666075–MT666081 (Table 1).

Molecular characterization of monogeneans

A preliminary dataset involved 20 COI sequences, including 9 sequences of P. trachuri and sequences from GenBank. After alignment and trimming, the dataset was 391 bp in length with 211 informative sites. The main result for this preliminary analysis was that our sequences of P. trachuri from off the Algerian coast formed a robust monophylum with the single sequence of P. trachuri available from the type-host T. trachurus off Sète, France (AY009168) [40]; this sequence is therefore confirmed as P. trachuri. However, we noted that sequence AY009167, also from T. trachurus off Sète, France, grouped with P. trachuri, although it was identified as Gastrocotyle trachuri Van Beneden & Hesse, 1863 and labelled as such in GenBank by its authors, Jovelin & Justine [40]; we interpret this as a misidentification, and sequence AY009167 should be considered P. trachuri.

We noticed that several GenBank sequences were unconfirmed and/or unpublished, and there was no certainty in the systematic assignment for some of them (in addition to the obvious error mentioned above); also, some had apparent insertions-deletions, or ambiguous bases, and the alignment included gaps. Therefore, from the preliminary dataset, we selected only the sequences that had the same span as our sequences and no ambiguous bases. The restricted dataset included only 12 sequences: 9 sequences of P. trachuri (including 7 new), 2 sequences of Allogastrocotyle bivaginalis, and the COI part of the complete mitogenome (NC016950) of the chauhaneid Pseudochauhanea macrorchis Lin, Liu & Zhang in Zhang, Yang & Liu, 2001 [81]. We obtained perfect alignment, 389 bp in length, without any gap or missing data. The tree (Fig. 1) showed a robust monophylum that includes all sequences of P. trachuri. The P. trachuri clade contained sequences from the type-host and from the sparid host B. boops, confirming the occurrence of P. trachuri on a sparid. Moreover, all specimens of P. trachuri with apparent “multiple ventro-lateral vaginae” and specimens without vaginae clustered in the same clade, without having a distinct branch.

Open in a separate window

Figure 1

Molecular phylogenetic analysis of monogenean COI sequences using the Maximum Likelihood method. Bootstraps percentages (1000 replicates) are indicated next to the branches when higher than 70. There were a total of 389 positions in the final dataset. The Neighbour-Joining tree had a similar topology and is not represented. All specimens of Pseudaxine trachuri, whether with no vagina or with “multiple vaginae”, and whether from the type-host Trachurus trachurus or from Boops boops, were grouped into a single monophylum and showed little variation (<2%).

Distances were computed using p-distance and the Kimura-2 parameter. All sequences of P. trachuri, whatever host or apparent number of vaginae, were very similar, with less than 2% variation. This suggests that P. trachuri also occurs on B. boops and that specimens with apparent multiple vaginae are conspecific with typical P. trachuri.

General structure of the genital atrium of Pseudaxine trachuri

The original description of P. trachuri is apparently very detailed and beautifully illustrated, especially for a paper of the 19th Century [63]. It has been supplemented by redescriptions and more morphometric data by Yamaguti [76, 77], Sproston [70], Parukhin [64], Lebedev [53] and Radujkovic & Euzet [67]. However, the illustrations were sometimes poor or redrawn from the original 1890 account of Parona & Perugia [63].

Parona and Perugia indicated that the genital atrium was armed with two crowns of hooks [63]: “L’apertura cloacale […] coronata da un cerchio di 24 uncini minutissimi (Fig. 11), con base allargata e punta arcuata rivolta verso il lume del canale. Inferiormente e nell’interno della cloaca si osserva altro cerchio di aculei (Fig. 12) molto più lunghi dei superiori (0,017) che soprastanno all’apertura del deferente e che si può rigardare quale armatura maschile” which translates as “The cloacal opening […] is crowned by a circle of 24 very fine hooks (Fig. 11), with an enlarged base and the point turned towards the lumen of the canal. Beneath and inside the canal, another circle of spines is observed (Fig. 12), which are much longer than those above (0.017) and are present distal to the aperture of the vas deferens; these can be considered the male armature”.

Yamaguti redescribed the species twice, on the basis of two (1938) and one (1942) specimens, from the gills of the type-host off Japan [76, 77]. In both descriptions, Yamaguti clearly indicated that the genital atrium was surrounded by a single crown of hooks, but did not comment on the differences in the genital atrium.

Sproston (1946) redescribed P. trachuri from its type-host, based on a single specimen collected from off Plymouth, England, and indicated the presence of 16 atrial hooks. However, she reproduced in her paper one of the drawings by Parona & Perugia and commented on the original description: “these authors give a curious figure of the genital corona […] and describe it as having an outer ring of 24 small hooks and 17 larger hooks below at the end of the vas deferens”; she stated “neither the Japanese nor Plymouth specimens from the same host reveal anything like this structure”. In this sentence, Sproston mentioned 17 larger hooks, but Parona & Perugia did not provide the number of the larger hooks; a possible explanation is that Sproston confused the length (17 μm) and the number of the larger hooks in the original description in Italian, although the length was provided as “0.017”.

Unnithan [73], probably using Sproston’s text, wrote in his description of Pseudaxine kurra Unnithan, 1968 “no structure identical to the genital corona having an outer rim of 24 small hooks and 17 larger hooks below it at the end of the vas deferens is observed in the new species”.

Lebedev commented “The discussion of Sproston [70] and Unnithan [73] about the allegedly incorrect description of these structures by Parona and Perugia [63] seems unreasonable to us: if we accept that the outer ring of spines in the picture of the copulatory apparatus given in the first description is the apical parts of the hooks, and the inner ring is a stylet, then everything falls into place [53].

Radujkovic & Euzet, on the basis of specimens from off Montenegro, described briefly the genital atrium as having a crown of 12 spines [67].

Our careful examination of new specimens of Pseudaxine trachuri from different hosts and localities showed the presence of a single crown (corona) of 12 hooks. Clearly, the two circles described by Parona & Perugia [63] have never been mentioned in all subsequent descriptions of this species (Table 4). Although Lebedev [53] suggested that this could be considered a minor misinterpretation of the structure, we believe that it is important to clarify this point.

The number of atrial hooks in the material examined in this study and other accounts (Table 2) varies according to locality. Yamaguti suggested that it was due to intraspecific variability [76, 77]. However, the number of hooks given in this study is the same as for the specimens from the Mediterranean described by Radujkovic and Euzet [67]. It may be that specimens from seas other than the Mediterranean correspond to different species; this needs to be investigated using molecular methods. We provisionally consider that all records worldwide correspond to one and the same species.

In none of the previous description was a stylet mentioned. Although Lebedev used this term to explain the drawing by Parona & Perugia, he did not explicitly describe a stylet in his specimens.

The central stylet of the genital atrium

We describe in our specimens of P. trachuri the presence of a conical structure in the centre of the crown of spines of the genital atrium; this is the first time that such as structure has been described in this species. Llewellyn (1983) provided a detailed description of the genital atrium of Gastrocotyle trachuri; he named the hooks forming the crown “ribs” and the central stylet a “penis tube” [56]; he also proposed a hypothesis for sperm transfer in which the genital atrium would work as a sucker and the penis tube would penetrate the tegument to allow the deposition of sperm in the subtegumental tissue of the partner. What could be a stylet-like structure was also described by George (1960) in another gastrocotylid, Engraulicola forcipopenis George, 1960, as a “forceps-like spine at the end of the penis” [31]. In the diclidophorid Diclidophora merlangi (Kuhn, 1829) Krøyer, 1838, which has no vagina, MacDonald & Caley showed that sperm were injected into the tissue of the partner [57]; however, a stylet was not described in this species.

Our observations of the structure of the stylet show that it is hollow and thus could allow sperm to be injected into the tissue of the partner, and the absence of a vagina (see below) suggests that P. trachuri uses traumatic insemination, a process which is not uncommon in polyopisthocotylean monogeneans [41].

Vaginae in specimens of the genus Pseudaxine and allied genera

Parona & Perugia, in their original description of P. trachuri, which is interspersed with comparison with another species, clearly stated that there was no vagina “Infatti non travammo traccia di canale e di apertura vaginale”, i.e. “In fact we did not find a trace of a canal or of a vaginal opening” [63].

Yamaguti redescribed the species twice [76, 77]. In neither description did he mention the presence or absence of a vagina. However, in his “Systema Helminthum” he wrote for the diagnosis of Pseudaxine “vagina opening middorsally” [78].

Dillon & Hargis described Pseudaxine bivaginalis Dillon & Hargis, 1965, and emended Yamaguti’s (1963) diagnosis of the genus to include the presence of “a single vagina opening middorsally or paired vaginae opening near lateral margins at or near level of genital atrium” [23].

Unnithan described Pseudaxine kurra Unnithan, 1968, with “vaginal pore unarmed, dorsal, with few gland cells and situated in front of the ovarian loop, slightly shifted to the right side”; apparently unaware of Dillon & Hargis’s work (1965), he emended the diagnosis of the genus as “vagina single median dorsal and unarmed” [73]. Lebedev commented on the position of the vagina in Pseudaxine kurra: “Most likely a mistake was made, since the vagina cannot be in this place. In our specimen (not in very good condition) no vagina was found”; and he questioned the validity of the species [53].

Radujkovic & Euzet, for P. trachuri from off Montenegro, wrote “the vagina seems absent” [67].

Lebedev recorded Pseudaxine trachuri from Trachurus novaezelandiae (which is the type-host of Pseudaxine bivaginalis Dillon & Hargis, 1965) off Australia in 1968, and mentioned the presence of two lateral vaginal openings [49]. Since two vaginae have never been observed in P. trachuri (Table 5), we believe it likely that the gastrocotylids recorded by Lebedev on T. novaezelandiae were in fact P. bivaginalis (see the taxonomic decision made regarding this taxon below).

Yamaguti described Pseudaxine decapteri Yamaguti, 1968 with a “vagina on each side divided into longitudinal row of 5–13 transversely elongated cuticular pits, beginning just behind level of genital pore” [80]. He also claimed “The peculiar structure of the vagina must have been overlooked by Parona and Perugia and subsequent authors because it is recognizable in well preserved specimens alone”. Subsequently, Lebedev erected the genus Pseudaxinoides Lebedev, 1968 for monogeneans resembling Pseudaxine spp. but differing from it by the presence of multiple ventro-lateral vaginal papillae [49]. Lebedev (1977) studied descriptions and figures of Pseudaxine decapteri and assigned this species to Pseudaxinoides [51].

Taxonomic decisions

As explained above, three genera of gastrocotylids are very similar in all aspects, except for their (alleged) number of vaginae. Figure 6 shows drawings of, side by side, specimens of Pseudaxine trachuri (no vagina), Pseudaxine trachuri (“multiple vaginae”), and Allogastrocotyle bivaginalis (two vaginae). Figure 5 shows micrographs of the anterior region of similar specimens. Specimens with multiple vaginae would have been attributed to Pseudaxinoides in its traditional concept.

Open in a separate window

Figure 6

Illustrations of worms with the apparent vaginal arrangement agreeing with the diagnoses of three different genera: Pseudaxine, Pseudaxinoides, and Allogastrocotyle. (A and B) Pseudaxine trachuri, typical aspect, no vagina, MNHN HEL1290, HEL1291; (C and D) Pseudaxine trachuri, specimen with apparent “multiple vaginae”, MNHN HEL1283, HEL1292; (E and F), Allogastrocotyle bivaginalis, two real vaginae, MNHN HEL1293 (habitus redrawn from Bouguerche et al. [10]).

In the present study, we have demonstrated, on the basis of morphology and molecules, that specimens of P. trachuri without a vagina and specimens with apparent multiple lateral vaginae are conspecific. We thus consider that all similar species described with apparent multiple vaginae are based on an erroneous interpretation of the excretory system; this makes the distinction between Pseudaxine and Pseudaxinoides void, and we propose the synonymy of these two genera. Since Pseudaxine has priority over Pseudaxinoides (1890 vs. 1968), we propose to transfer most Pseudaxinoides spp. to Pseudaxine, with the following new combinations: Pseudaxine australis (Lebedev, 1968) n. comb.; Pseudaxine bychowskyi (Lebedev, 1977) n. comb.; Pseudaxine caballeroi (Lebedev, 1977) n. comb.; Pseudaxine cariacoensis (Nasir & Fuentes-Zambrano, 1983) n. comb.; and Pseudaxine vietnamensis (Lebedev, Parukhin & Roitman, 1970) n. comb. [49–52, 54, 60, 80]. Pseudaxine decapteri Yamaguti, 1968 is kept within Pseudaxine. The status of P. kurra is pending [53]. We noticed that Pseudaxinoides pusanowi Lebedev, 1984 has two of its testes that are paraovarian [52], in contrast to species of Pseudaxine in which they are all post-ovarian; Lebedev noted later [53] that is was close to species of Allopseudaxinoides Yamaguti, 1968; a reappraisal of this species is probably necessary and thus we provisionally refrain from transferring it to Pseudaxine.

A comparison of Pseudaxine bivaginalis Dillon & Hargis, 1965 with the single species of Allogastrocotyle, A. bivaginalis Nasir & Fuentes Zambrano, 1984 [10, 60], suggests that the former species should be transferred to Allogastrocotyle, as indicated by the presence of two ventrolateral vaginal openings surrounded by muscular tissue near the genital atrium [23, 53]. Since the name Allogastrocotyle bivaginalis is already occupied, we propose here the new name Allogastrocotyle dillonhargisorum nom. nov. to avoid homonymy. The new name has been registered in ZooBank as: urn:lsid:zoobank.org:act:1ECC6CF2-4BC4-45E8-9AE7-049C00FC7735.

This research was supported by Direction Générale de la Recherche Scientifique et du Développement Technologique (DGRSDT, Algiers, Algeria) and Institut de Systématique, Évolution, Biodiversité (ISYEB, MNHN, Paris, France). We thank Drs. Gerardo Pérez-Ponce de León, David Gibson, Delane Kritsky, Ken MacKenzie and Mario Santoro who kindly provided literature. We are grateful to Drs. Rodney Bray and José Amet Rivaz Hérnandez for valuable discussions concerning nomenclature. Special thanks go to Nawel Mekarma and Doria Bouabache for their help in collecting biological material.