Redescription and Molecular Characterisation of Derogenes ruber Lühe, 1900 (Hemiuroidea: Derogenidae) from Chelidonichthys lastoviza (Scorpaeniformes: Triglidae) in the Western Mediterranean

Purpose Derogenes ruber Lühe, 1900, the type-species of the genus Derogenes Lühe, 1900, is a poorly known derogenid digenean. The original description of this species was not illustrated and aspects of the morphology of the parasite from the type-host remain scarce. Available records of this species were brief and/or lacked illustrations and were based on morphology alone. Additionally, molecular data for Derogenes spp. are warranted to untangle species complexes as they provide a better assessment of interspecific genetic divergence. Methods Derogenes ruber is redescribed based on newly collected specimens from the gall bladder of its type-host Chelidonichthys lastoviza (Bonnaterre, 1788) collected in the Western Mediterranean off the Algerian coast during 2017–2019 and molecular data are provided using a partial fragment of the nuclear 28S ribosomal RNA gene (28S rRNA), the internal transcribed spacer 2 (ITS2) and a fragment of the mitochondrial cytochrome c oxidase subunit 1 (cox1) gene. Results We herein provide a detailed illustrated redescription and morphometric data of D. ruber from its type-host C. lastoviza. We report a new geographical record (off Algeria) for it. Derogenes ruber is also genetically characterised for the first time. Species/lineages of Derogenes were recovered in five strongly supported reciprocally monophyletic clades: (i) D. ruber from C. lastoviza off Algeria; (ii) D. lacustris from Galaxias maculatus (Jenyns) off Argentina; (iii) Lineage “D. varicus DV1” (D. varicus sensu stricto) from fish hosts in the White and Barents seas and the North Sea; (iv) Lineage “D. varicus DV2” from mollusc hosts in the White Sea; and (v) Lineage “D. varicus DV3” from Eumicrotremus fedorovi Mandrytsa. in the Pacific Ocean. Hence, comparison of the newly generated sequences with other available data for Derogenes species supports the distinction of D. ruber confirming its taxonomic status and helping assess interspecific variation. Comparison of D. ruber with the closely related species Derogenes latus revealed overlaps in morphometric data and the validity of the latter species is questioned. Conclusion The combination of morphological and molecular data provided for D. ruber provides a firm foundation for further investigations of Derogenes spp. Although we do describe herein material of D. ruber from the type-host, given that the occurrence of a single Derogenes species in various hosts has been challenged by molecular data, and both D. lacustris and D. varicus sensu stricto had been genetically proven to occur in various hosts, D. ruber and D. latus may be indeed synonymous. Additional sequencing effort on Derogenes spp. will strengthen systematic comparative studies and evolutionary relationships within the Derogenidae in general.


Introduction
Derogenids are hemiuroid digenean gut parasites, occurring in fishes.Throughout most of their taxonomic history, they were accommodated within a broad concept of the family Hemiuridae Looss, 1899 [1].The Derogenidae Nicoll, 1910 was first used at full family rank by Dollfus [2] but was Chahinez Bouguerche and Fadila Tazerouti have equally contributed to this work and are both senior authors.
Extended author information available on the last page of the article off Algeria, we collected representatives of D. ruber from the gall bladder of its type-host, C. lastoviza.The aim of the present study is to provide a formal redescription of D. ruber and to characterise the species genetically based on partial 28S ribosomal RNA gene (28S rRNA), internal transcribed spacer ITS2, and a fragment of the mitochondrial cytochrome c oxidase subunit 1 (cox1) gene sequences.

Morphological Methods
Live digeneans were killed and fixed in near-boiling water.Specimens for morphological analysis were fixed under cover-glass pressure in Bouin's fluid [10], then preserved in 70% ethanol, stained with acetic carmine, dehydrated through a graded alcohol series, cleared in clove oil, and mounted in Canada balsam as permanent mounts.Five specimens were preserved immediately in 96% ethanol for molecular characterisation and were processed as hologenophores (sensu Pleijel et al. [22]).
Permanent mounts of the hologenophores, consisting of 2/3 of the body (posterior third excised and used for sequencing), stained and mounted in Canada balsam.Drawings were made using a Zeiss microscope (Université des Sciences et de la Technologie Houari Boumediene, USTHB) and a Nikon Eclipse i80 microscope with DIC (differential interference contrast) (Swedish Museum of Natural History, SMNH) equipped with a drawing tube, and scanned and redrawn with Adobe Illustrator 2023, version 28.0.
Measurements are in given in micrometres and presented as the range followed by the mean in parentheses.Voucher material was deposited at the Swedish Museum of Natural

Molecular Methods
Genomic DNA was extracted from a total of five hologenophores, and genetic sequence data were generated for three genetic markers: a partial region of the mitochondrial cytochrome c oxidase subunit 1 gene (cox1), the second internal transcribed spacer region (ITS2 rDNA), and the large (28S) ribosomal RNA gene.A small fragment of each hologenophore (posterior third) was placed in a 1.5 ml microcentrifuge tube containing 20 μL buffer ATL (Qiagen, Hilden, Germany).For extraction of genomic DNA (gDNA), 20 μL buffer ATL and 20 μL proteinase K were added to each sample, followed by vortexing and incubation in an incubating microplate shaker at 56 °C and 300 rpm overnight.The lysed samples were processed to obtain gDNA following the manufacturer's instructions for gDNA extraction using the Qiagen QiAmp DNA Microkit.Polymerase chain reaction (PCR) amplification was performed in 25 µl reaction mix using Illustra Hot Start Mix RTG (0.2 µl) reaction kit (GE Healthcare Life Sciences, Uppsala, Sweden).The reaction mix consisted of 1 µl (0.4 µM) of each primer, 2 µl template DNA, and 21 µl nuclease-free water.The primer set JB3 (5′-TTT TTT GGG CAT CCT GAG GTT TAT-3′) and COI R-Trema (5′-CAA CAA ATC ATG ATG CAA AAG G-3′) were used to amplify a fragment the cox1 gene [23].The thermocycling profile consisted of an initial denaturation step at 94 °C for 5 min, followed by 35 cycles of denaturation at 94 °C for 30 s, annealing at 45 °C for 30 s, and extension at 72 °C for 1 min, with a final extension step at 72 °C for 10 min [14].Primers, amplification, and sequencing protocols for the 28S rDNA region followed Pérez-Ponce de León et al. [24] and García-Varela and Nadler (2005) [25].The thermocycling profile consisted of an initial denaturation step at 94 °C for 3 min, followed by 35 cycles of denaturation at 94 °C for 60 s, annealing at 54 °C for 60 s, and extension at 72 °C for 1 min, with a final extension step at 72 °C for 7 min.ITS2 rDNA spacer was amplified using the primers 3S [26] and ITS2.2 [27] and the following thermocycling profile: an initial denaturation step at 94 °C for 3 min, followed by 35 cycles of denaturation at 94 °C for 1 min, annealing at 54 °C for 1 min, and extension at 72 °C for 1 min, with a final extension step at 72 °C for 7 min.PCR products were purified (Ampure XP Kit, Beckman Coulter, Indianapolis, USA) 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 and compared them to the GenBank database content using BLAST.The newly generated sequences are deposited in the GenBank database under the accession numbers OQ919798-OQ919804, OQ919806, OR245546, and OR245386.
Phylogenetic analyses were performed using the newly generated sequences of D. ruber and those for Derogenidae species available in GenBank (Table 1).Alignments for each gene region were constructed in AliView [28] and trimmed to the length of the shortest sequence.Nucleotide substitution models for phylogenetic analyses using the maximumlikelihood method were estimated using MEGA11 [29].The best-fit models selected were the Kimura 2-parameter model with gamma distributed amongst-site rate variation (K2 + G) for the 28S rDNA alignment, Kimura 2-parameter (K2) model for the ITS2 alignment, and Tamura-Nei model (TN93) with estimates of invariant sites and gamma distributed amongst-site rate variation (HKY + I + G) for cox1.All trees were constructed in MEGA11, with 500 replications.Genetic distances [uncorrected p-distance model (Kimura 1980)] were computed with MEGA11.
Site in host: Gall bladder.
Five ITS2 sequences (∼566 bp) were obtained for D. ruber.The tree built using the newly generated sequences aligned with 12 sequences for Derogenes spp.and Prosogonotrema bilabiatum as the outgroup is shown in Fig. 3A.Derogenes ruber and the lineages "D.varicus DV1" from various fish hosts in the White and Barents seas and "D.varicus DV2" from the molluscs B. scalariforme, Amauropsis islandica (Gmelin) and Euspira pallida (Broderip & Sowerby) from the White and Barents seas clustered in reciprocally monophyletic groups with a maximum nodal support.
The five newly generated ITS2 sequences for D. ruber were also identical and differed from those for the lineage "D.varicus DV2" by 4% (16 substitutions) and from those for the lineage "D.varicus DV1" by 5% (21 substitutions).None of the taxa included in the analysis showed intraspecific/intralineage variation.

Discussion
Derogenes ruber was described from the gall bladder of the streaked gurnard C. lastoviza off Rovinj, Croatia, Adriatic Sea [16].Although the original description of D. ruber was detailed, it lacked illustrations.The only subsequent illustration of this species is that of Sey [17], which barely shows any internal organs and omits any details of the terminal genitalia.Sey (1968) examined three specimens of a distinct host, T. lyra, and redescribed briefly D. ruber based on two specimens.Although the geographical distribution of the type-host, C. lastoviza, is wide, D. ruber has been reported only from the Central Mediterranean (Adriatic Sea off Croatia, type-locality in the original description [16] and later, from a different host [17] and recently from the type-host off Italy, based on A. Looss's material) [15].The latest record despite providing few morphometrical data and illustration did not include any genetic data.Derogenes ruber was reported from the type-host in the North-East Atlantic, off Azores, Canary and Cape Verde islands [18] and off Spain [19].Consequently, this paper provides a detailed illustrated description of D. ruber and Algeria as a new locality for this digenean.Additionally, we genetically characterised for the first time D. ruber using the partial fragments of the   [14] and Tsuchida et al. [31] who provided abundant data, corresponding to the "candidade" D. varicus species complex and D. lacustris, respectively.Krupenko et al. [14] have shown the existence of four groups (labelled as DV1-DV4) within the "candidade" D. varicus species complex; of these, they considered that two (DV1 and DV2) may belong to distinct species [14].Recently, Bouguerche et al. [15] demonstrated that DV1 is in fact D. varicus sensu stricto.
Herein, the 28S rDNA analysis recovered D. ruber in a clade distinct from lineages "D.varicus DV1, DV2, and DV3" and the well-established species D. lacustris.The ITS2 analysis supported the monophyly of D. ruber, and lineages "D.varicus DV1" and "D.varicus DV2" and the cox1 tree yielded a similar topology.Although the sequences obtained herein were short affecting thus the alignment's length, the analysis led to results similar to those of Krupenko et al. [14].More importantly, the genetic distance for the cox1 gene between D. ruber and lineages "D.varicus DV1" and "D.varicus DV2" was 19% and 17%, respectively; D. ruber also differed from D. lacustris by 23%.These levels of genetic divergence agree well with previously reported interspecific divergence based on cox1 within the closely related halipegine derogenids ranging between 10.5-15.1% for Genarchopsis spp.[23] and 16.9-20.4%for Genarchopsis Ozaki, 1925 and Allogenarchopsis Urabe & Shimazu, 2013 [33].Furthermore, the levels of interspecific genetic divergence are more than ten times greater than those for the intraspecific divergence for the mitochondrial "barcode" marker, thus supporting the recognition of D. ruber as a valid distinct species.The molecular data generated herein advance our knowledge on interspecific genetic variations within Derogenes and will help further efforts to untangle the D. varicus species complex and delimit the potentially cryptic species hidden under the single name "D. varicus".Additionally, the morphometrical data of D. ruber from the type-host (Table 2) will help accessing interspecific morphological differences.
A problem arises when comparing D. ruber to a closely related species, D. latus Janiszewska, 1953, first described based on a single specimen in the intestine of Mullus barbatus Linnaeus from the same Adriatic locality as that of D. ruber, off Split, Croatia [34].Derogenes latus was redescribed from the intestine of M. barbatus and Trisopterus capelanus (Lacépède) in the North Adriatic Sea [35] and 2 Tree inferred using the maximum-likelihood method based on the 28S rDNA sequence data; only bootstrap values higher than 70 are indicated.The newly generated sequences are indicated in red.Lineages "Derogenes varicus DV1, DV2, DV3" and Derogenes lacustris are highlighted in differently colored boxes from the gall bladder of M. surmuletus off Corsica (France), Western Mediterranean [10].The redescription provided by Bartoli and Gibson [10] (based on accessible voucher material and serial sections) should undoubtfully be referred to as the most detailed modern redescription of D. latus.Derogenes latus has been frequently reported from its type-host in the Western Mediterranean, off Spain [36] and off France [37], and from a closely related host, M. surmuletus, in the Western Mediterranean (off France and Algeria) [37][38][39].
The taxonomic status of D. latus is uncertain.The distinction D. ruber and D. latus has been questioned [10], and the two species share a stout body, post-testicular vitellarium composed of two multi-lobed masses and a uterus occupying almost the entire body [10,35,42].The typehosts are, however, different: C. lastoviza for D. ruber [16] and M. barbatus for D. latus [34].Overall, all morphometric data for D. ruber and D. latus overlapped (Tables 2, 3) except for specimens of D. latus from M. surmuletus and S. scrofa from the Western Mediterranean having larger eggs (see Table 3) and the two species clearly share the deeply loped shape of the vitelline masses.It is worth noting that a comparison of the present specimens of D. ruber with those of D. latus provided by Bartoli and Gibson [10]  Bartoli and Gibson [10] convincingly highlighted the striking morphological similarity between D. latus and D. ruber and indicated that the egg size given by Sey [17] for D. ruber is probably an inaccuracy.They refrained from synonymising the two species formally until further studies of material from the type-hosts and localities are available.Although we found morphometric differences between the present material of D. ruber from the type-host and the material of D. latus described by Bartoli and Gibson [10], and given that the occurrence of a single Derogenes species in various hosts has been challenged by molecular data [14,15,31], and both D. lacustris and D. varicus sensu stricto (D. varicus lineage DV1 of Krupenko et al. [14]) had been genetically proven to occur in various hosts (see Fig. 3A), it is possible that D. ruber and D. latus are indeed synonymous, thus transforming D. ruber to a euryxenic species.However, since molecular data for D. latus are still lacking, we also refrained from synonymising the two species.The genetic data generated herein for D. ruber from its type-host will be certainly valuable for a future investigation of the synonymy of these two species.Funding Open access funding provided by Swedish Museum of Natural History.Chahinez Bouguerche was supported individually by a framework agreement projects: 1. the DeepBlue Project: Distance Crossborder Traineeship Programme co-financed by "The European Maritime and Fisheries Fund (EMFF)" for the analysis, interpretation of data, and the writing of the manuscript, 2. "The Ocean Fellowship" (edition 2021), offered by TBA21-Academy and held at Ocean Space in Venice and 3. the Swedish Taxonomy Initiative, Artdatabanken, Swedish University of Agricultural Sciences within the scope of the project "Taxonomy and systematics of digenetic trematodes parasitising fishes of Sweden" (dha 2019.4.3-48).The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

B
Derogenes and the subfamily Halipeginae and Prosogonotrema bilabiatum Vigueras, 1940 as the outgroup yielded the topology shown in Fig.2.There were a total of 688 positions in the final dataset.The general topology of the ML tree agreed with the taxonomic classification of the included species and distinct lineages.Species/lineages of Derogenes were recovered in five strongly supported reciprocally monophyletic clades: (i) D. ruber from C. lastoviza off Algeria; (ii) D. lacustris Tsuchida, Flores, Viozzi, Rauque et Urabe, 2021from Galaxias maculatus (Jenyns) off Argentina[31];

Fig. 3
Fig.3Trees inferred using the maximum-likelihood method based on the ITS2 rDNA and cox1 sequence data.A, ITS2 rDNA tree; only bootstrap values > 70 are indicated.The newly generated sequences are indicated in red.Lineages "D.varicus DV1 and "D.varicus DV2" are highlighted in differently colored boxes.There were no ITS2 sequences available for D. lacustris.B, cox1 tree; only bootstrap values > 70 are indicated.The newly generated sequences are indicated in red.Derogenes lacustris and lineages "D.varicus DV1", "D.varicus DV2" are in different colors.There were no cox1 sequences available for the lineage "D.varicus DV3" ◂

Table 1
Hosts, locality, and GenBank accession data for the sequences of Derogenes spp.and halipegine derogenids analysed in this study

Table 2
Metrical data for Derogenes ruber from Chelidonichthys lastoviza and Trigla lyra Abbreviations: FO/BL (%) forebody length as a percentage of body length, RT/BL (%) right testis length as a percentage of body length, LT/BL (%) left testis length as a percentage of body length, RPT/BL (%) right post-testicular region length as a percentage of body length, LPT/BL (%) left post-testicular region length as a percentage of body length, OV/BL (%) post-ovarian field length as a percentage of body length.