A monogenean fish parasite, Gyrodactylus chileani n. sp., belonging to a novel marine species lineage found in the South-Eastern Pacific and the Mediterranean and North Seas

Gyrodactylus chileani n. sp. is the first Gyrodactylus species reported from Chile. It is an ectoparasite living on fins and skin of a small fish, the Chilean tidal pond dweller Helcogrammoides chilensis (Cancino) (Perciformes: Tripterygiidae). A phylogenetic analysis based on 5.8S+ITS2 of rDNA placed the new species close to marine Gyrodactylus species found in Europe: G. orecchiae Paladini, Cable, Fioravanti, Faria, Cave & Shinn, 2009 on gilthead seabream Sparus aurata L. from the Adriatic and Tyrrhenian Sea fish farms (Perciformes: Sparidae), and an undescribed species on the black goby Gobius niger L. from the North Sea (Perciformes: Gobiidae). A morphological description of the latter species is unavailable. These geographically distant parasite samples on different host families form a new well supported Gyrodactylus orecchiae lineage. Using molecular phylogenetics, it is shown that the marine species groups of Gyrodactylus may have a worldwide distribution.

The mechanism resulting in such a phylogeny is the basal radiation of Gyrodactylus (sensu lato) generating new species groups, some marine, some freshwater, and some inhabiting both environments. The latter are perhaps the most interesting, and include examples where the host lineage has moved from marine to freshwater, and the parasite has followed. The best known examples are two species, G. lotae Gusev, 1953 andG. alexgusevi Ziętara &Lumme, 2003, on the freshwater burbot Lota lota (L.), a gadid, and two species, G. hrabei Ergens, 1957 and G. mariannae Winger, Hansen, Bachmann & Bakke, 2008, on an inland cottid, Cottus poecilopus Heckel, in Europe. The nearest relatives of all four of these species are parasites of marine hosts (Ziętara & Lumme, 2003;Rokicka et al., 2009).
To enlarge the geographical range of the Gyrodactylus spp., we describe the first Gyrodactylus species from Chile, which was among 215 fish ectoparasitic species listed by Muñoz & Olmos (2007). Two other records of Gyrodactylus spp. on small intertidal fishes found in rocky areas, Scartichthys viridis (Valenciennes) (see Muñoz & Randhawa, 2011) and Sicyases sanguineus Müller & Troschel (see Muñoz & Zamora, 2011), await molecular analyses and further description. The species described below is also the first molecularly defined Gyrodactylus reported from the Southern Pacific. Together with two other species of Gyrodactylus from Europe, which are related by ITS rDNA, it forms a new marine species group, which has crossed the equator during its evolutionary history.

Materials and methods
Fish were collected from rocky pools during low tides in July, 2009. Nineteen of 32 specimens were infected (prevalence 59.4%). Five ethanol-preserved worms were picked from the fins and skin of different hosts. The haptor was cut off, softened and cleared in 120 lg/ ml proteinase K. It was then prepared for a microscopy on a slide with a saturated concentration of ammonium picrate in glycerine (Malmberg 1970). The remainder of the body was used for molecular analysis.
The phylogenetic placement of the new species was estimated by aligning the 5.8S?ITS2 segment of the ribosomal DNA with selected sequences published in GenBank. Only predominantly marine lineages of gyrodactylid species were included. The alignment was made using MUSCLE (Edgar, 2004) as implemented in MEGA5 (Tamura et al., 2011) and corrected manually. The phylogenetic tree was estimated from Maximum Composite Likelihood distances (Tamura et al., 2004) using the Neighbor Joining algorithm (Saitou & Nei, 1987) with the pairwise deletion option. The validity of the branches was evaluated by bootstrapping 500 replicates (Felsenstein, 1985). Bootstrap values lower than 70% were hidden.
For the species description, the preserved haptors were photographed and measured under a microscope using the measurements of Gusev et al. (1985). Measurements of G. orecchiae Paladini, Cable, Fioravanti, Faria, Cave & Shinn, 2009by Paladini et al. (2009 were included for comparison. The North Sea species on Gobius niger remains undescribed (Huyse et al., 2003).
Phylogenetic characterisation using the 5.8S1ITS2 rDNA fragment As this report focuses on the new species from Chile and its nearest relatives, we present a phylogeny based on 5.8S?ITS2 sequences of marine representatives of Gyrodactylus ( Fig. 1), omitting all exclusively freshwater subgenera and species groups, which have already been well analysed by Vanhove et al. (2011). The new species from Chile (JQ045347) clustered with high (98%) bootstrap support with two geographically distant species. These were G. orecchiae (FJ013097), described from Mediterranean cultured gilthead seabream Sparus aurata (Perciformes: Sparidae) by Paladini et al. (2009), and an undescribed species sequenced and discussed by Huyse et al. (2003) from the black goby Gobius niger (Perciformes: Gobiidae, AY338452) in the North Sea. The relatedness of these two European species was noted by Paladini et al. (2009), but they did not place the species into a wider phylogenetic context. The phylogeographic coverage of this lineage is now extended significantly to include the Southern Hemisphere and the Pacific Ocean.
The complete ITS rDNA (including ITS1?5.8S rDNA?ITS2) sequence is not available for Gyrodactylus sp. from Gobius niger (see Huyse et al., 2003) and therefore the complete ITS segment of the rDNA was aligned only for Gyrodactylus orecchiae and the new species from Chile. The Maximum Composite Likelihood distance was estimated to be 15.4% between these two species, but it must be stressed that the ITS1?5.8S?ITS2 segments of the rDNA are not optimal for distance estimation, due to problems with the reliable alignment of ITS rDNA, from less related Gyrodactylus species with distances above 10%.

Remarks on the systematic relationships of the new taxon
Subgeneric position. On the basis of the molecular sequence of the ITS region of the ribosomal DNA, the new species cannot be placed in any of the subgenera previously suggested by Malmberg (1970), which are already rendered poly-and paraphyletic by the inclusion of molecular data (Fig. 1). The Chilean Gyrodactylus species is placed within the mixed basal group of ''short ITS'' (Cable et al., 1999) species. The ''short ITS'' clade receives 100% bootstrap support in Fig. 1 as the sister group of the G. rugiensis, G. micropsi and G. eyipayipi species groups, which are the only marine groups (100% bootstrap support) among the ''long ITS clade'', and a sister clade of the subgenus G. (Limnonephrotus) (not demonstrated in Fig. 1, but see the phylogeny of Vanhove et al., 2011).
New species group. The new species clusters with G. orecchiae from the Mediterranean Sea (100% bootstrap support) and with an undescribed species on black goby from the North Sea. The three species cluster together with 98% bootstrap support and are relatively basal in the badly resolved clade (91% bootstrap support) of mostly marine, but also some derived freshwater, species. These species have previously been assigned to G. (Mesonephrotus), G. (Metanephrotus) and G. (Paranephrotus), or are not assigned at all. The considerable geographical spread of the three species assigned to this new G. orecchiae species lineage and the systematic diversity of their hosts suggest that a more comprehensive sampling may add numerous new species to this group.

Gyrodactylus chileani n. sp.
Type-host: Helcogrammoides chilensis (Cancino) (Perciformes: Tripterygiidae); local name 'Trombollito de tres aletas' (three-finned trombollito). Type-locality: Tidal ponds at Las Cruces, Valparaiso (33°30 0 S, 71°37 0 W), El Tabo, Chile. Sampling date July 22nd, 2009. Site: Fins and skin. Type-material: Slides of 5 isolated opisthaptors and 1 complete specimen of G. chileani n. sp. were deposited in Finnish National History Museum in Helsinki University. Holotype: MZH 118095; paratypes: MZH 118096 (complete specimen); MZH 118097-MZH 118100 (isolated haptors). Molecular data: Definitive identification is based on the nucleotide sequence of the internal transcribed spacers of the nuclear ribosomal DNA. The accession number for the ITS1-5.8S-ITS2 and short flanking segments of 18S and 28S rDNA from the holotype specimen is JQ045347. The holotype was cut in two, the haptor was mounted on a slide, and the remainder was used for DNA analysis. The ITS rDNA sequence was repeated from all of the five specimens that were measured. Etymology: The name of the species is derived from the country name, as it is the first Gyrodactylus species described from Chile.   (Fig. 2A). Marginal hooks clearly more delicate than in G. orecchiae; toe triangular in shape, not rhomboid as in G. orecchiae. Shaft of marginal hook points downwards and extends beyond toe. Heel pronounced, but smaller than in G. orecchiae (Figs. 2B, C, 3C).

Discussion
An overt phylogenetic revision of monogenean parasites of fishes was published by Perkins et al. (2009), with an informative title ''Looks can deceive…''. This seems to be a general rule among this class of parasites, certainly extending to the specific level among Gyrodactylus spp. With respect to the geographical coverage, the few molecular phylogenetic studies which have been attempted during the last decade are encouraging but far from satisfactory in terms of the strategic planning of a systematic revision (Cable et al., 1999(Cable et al., , 2005Harris & Cable, 2000;Ziętara et al., 2000Ziętara et al., , 2008, 2004Boeger et al., 2003;Huyse et al., 2003Huyse et al., , 2006Matȇjusová et al., 2003;LeBlanc et al., 2006;Malmberg et al., 2007;Kuusela et al., 2008;Rokicka et al., 2009;Vanhove et al., 2011). Due to the small fraction of the genus analysed in the above works, little can be said about its systematics. Additionally, a global systematic revision is certainly premature and unwarranted when less than 2% of the suspected species have been found and described; nevertheless, as the genus appears not to be monophyletic , Vanhove et al., 2011, some kind of revision is needed. An alternative strategy towards the global systematics of the family Gyrodactylidae could be an extraction of Fig. 1 Hypothetical phylogenetic position of Gyrodactylus chileani n. sp. based on 5.8S?ITS2 rDNA sequences of marine (or of marine origin) gyrodactylids. Bootstrap support less than 70% is omitted. Gyrodactylus sp. FJ040182 from Liza richardsonii (Smith); Gyrodactylus sp. AY338452 from Gobius niger (see Huyse et al., 2003); Gyrodactylus sp. AY338449 from Pomatoschistus norvegicus (Collett), a species provisionally called G. cf. longicactylus (see Huyse et al., 2003); Gyrodactylus sp. AJ427221 from Pomatoschistus lozanoi (de Buen), a species provisionally called G. cf. micropsi (see ; and Gyrodactylus sp. AY338447 from P. lozanoi, a species provisionally called G. cf. micropsi (see Huyse et al. 2003 . The parasites on S. aurata were recorded in fish farms from Croatia and Albania (Paladini et al., 2009) or Italy (Paladini et al., 2011b), so it is not known whether the fish is the natural host of Gyrodactylus orecchiae. Considering the species richness of these host families (e.g. Hickey et al., 2009), we may expect that the three parasite species mentioned here are just a small subsample. They are all marine monogeneans and the geographical distance indicates that it is likely that at least the coastal waters of all seas contain related parasite species.
It has been already reported that the two Antarctic gyrodactylids, G. coriicepsi Rokicka, Lumme &Ziętara, 2009 andG. nudifronsi Rokicka, Lumme &Ziętara, 2009, are related to the European Gyrodactylus fauna, indicating the evolutionary continuum of the marine species in the Northern and Southern Hemispheres (Rokicka et al., 2009). This lineage also accommodates two recently described species, i.e. the European G. longipes Paladini, Hansen, Fioravanti & Shinn, 2011, reported as a mixed infection with G. orecchiae from Sparus aurata in the waters of Bosnia-Herzegovina and Italy (Paladini et al., 2011b), and the North American G. aideni Mullen, Cone, Easy & Burt, 2010 along with G. pleuronecti Cone, 1981, a species recently tagged by ITS rDNA (Mullen et al., 2010).
So far, the few Gyrodactylus species sampled from the coasts of the Southern Pacific and Southern Atlantic have been found to have quite close relatives in the species groups studied in the Northern Hemisphere. Connections between the Northern Pacific and Northern Atlantic are even tighter, including common species such as G. arcuatus. This finding supports the hypothesis of Boeger et al. (2003) suggesting the origin of viviparous gyrodactylids in South American waters and a further expansion from that continent via the marine environment. Such a scenario requires a mixing of the marine gyrodactylid fauna as demonstrated in the present paper.
Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.