A new species of Gyrodactylus ( Monogenea , Gyrodactylidae ) , an ectoparasite from the endemic Iranocichla hormuzensis ( Teleostei , Cichlidae ) , the only Iranian cichlid

Iranocichla hormuzensis occupies a biogeographically peculiar position. This endemic of southern Iran is the only Iranian cichlid. While it is phylogenetically related to African oreochromine members of the cichlid family, it remains unclear how it has dispersed into its current range. It is one of the many lasting enigmas of cichlid biogeography. Monogenean fish parasites may provide useful additional information in such cases. Therefore, I. hormuzensis was examined for these flatworms. A gyrodactylid parasite is reported and compared to congeners from the Palearctic and from cichlids. In this way, we verify whether it shows affinities to parasites from fishes that are either biogeographically or phylogenetically close to Iranocichla hormuzensis. The species is new to science and is described as Gyrodactylus jalalii sp. nov. This is the first description of a parasite infecting I. hormuzensis. Because of the fixation method or age of the material, DNA could not be isolated. Due to the lack of genetic data, no conclusions can be drawn on its phylogenetic positioning. Indeed, Gyrodactylus phylogeny cannot be European Journal of Taxonomy 30: 1-10 ISSN 2118-9773 http://dx.doi.org/10.5852/ejt.2012.30 www.europeanjournaloftaxonomy.eu 2012 · Vanhove M.P.M. et al.. This work is licensed under a Creative Commons Attribution 3.0 License. R e s e a r c h a r t i c l e urn:lsid:zoobank.org:pub:489A6BC2-795C-4945-A642-440D4F92A126


Introduction
The fascination and scientific interest that cichlids raise in many scientists and hobbyists is well reviewed by Barlow (2000). This author aptly shows the importance of cichlid fishes in a wide range of fields in evolutionary biology, from ecology to behavior and from developmental biology to speciation. Numbering about 1350 described species, with hundreds remaining to be discovered (Nelson 2006), these "model" fishes are the most species-rich vertebrate family worldwide (Kocher 2004). Counting only described species, Cichlidae is the largest non-ostaryophysan freshwater fish family (Kullander 2003). Its species richness reaches its peak in the more than 900 African representatives. Some relatives to African cichlids occur in the Middle East (belonging to Astatotilapia Pellegrin, 1904;Oreochromis Günther, 1889;Sarotherodon Rüppell, 1852;Tilapia Smith, 1840 andTristramella Trewavas, 1942) (Werner & Mokady 2004) and one in Iran. The latter is Iranocichla hormuzensis Coad, 1982, the only representative of the monotypic Iranocichla Coad, 1982. The species was first reported by Behnke (1975) and Saadati (1977) before being formally described (Coad 1982). This southern Iranian endemic cichlid is found in mostly saline riverine systems draining into the Strait of Hormuz (Persian Gulf). The question rises how it arrived in this biogeographically isolated range.
Most cichlid species in the Middle East have congeners among the African "tilapiine" (Oreochromis, Tilapia and Sarotherodon: Schwarzer et al. 2009) or haplochromine (Astatotilapia: Werner & Mokady 2004) cichlids. Tristramella and Iranocichla, however, are endemic genera. Phylogenetically, they are also placed in Oreochromini, one of the tribes into which "tilapiine" cichlids were split by Schwarzer et al. (2009). Thus, they are closely affiliated to African cichlids. Middle-Eastern cichlids at present only occur in the Levant and these Iranian systems. This fact might indicate a relict-type consequence of a recent (Pleistocene to Recent) aridification event and decrease in temperature (overview in Coad 1982). Furthermore, the Persian Gulf was a river valley between 90,000 and 10,000 years ago. See also Arndt et al. (2003) for a discussion on how the low level of the Mediterranean during Pleistocene episodes is suggested to have allowed freshwater dispersal from the Nile to the Levant. Although those conditions would have allowed dispersal of cichlids, it is also possible that the marine environment would not represent a barrier to the spread of these animals, anyway (see references below). Indeed, occupying mostly saline streams shows that I. hormuzensis is salt-tolerant (Coad 1982). Moreover, several "tilapiines" are known to disperse between rivers through brackish coastal waters (Nelson 2006). Conversely, Murray (2001) favors marine dispersal through the Tethys Sea/Indian Ocean to coastal dispersal in view of the absence of cichlids along the rest of the region's coastlines.
Either way, both freshwater and marine dispersal pathways could explain the presence of this cichlid in waters draining into the Strait of Hormuz. Be it on a local or a global scale, the history behind current cichlid distribution across continents has not been unequivocally unraveled. The roles of intercontinental European Journal of Taxonomy 30: 1-10 (2012) dispersal, on the one hand, and of vicariance (following the break-up of Gondwana) and intra-continental freshwater dispersal, on the other hand, are still highly debated. A potential source of information that remains largely unexplored is the (monogenean) parasite fauna of cichlid species (reviewed in Pariselle et al. 2011). Many parasites have a close association with their host and a shorter generation time than their host, hence providing an alternative perspective on host evolution. Ectoparasitic monogenean flatworms often do not tolerate substantial changes in salinity, as indicated by their distribution and phylogenetic patterns worldwide (although there are exceptions, e.g., this is less the case in Gyrodactylidae, certainly when infecting euryhaline hosts, see below). One might therefore hypothesize that certain freshwater monogeneans rarely survive dispersal through barriers represented by marine environments. Hence, the phylogenetic relationships within a lineage of monogenean ectoparasites reflect both continental patterns and host phylogeny. Freshwater fish hosts most likely retain their monogenean assemblage when dispersing through continental surface waters. Conversely, it was suggested that most freshwater fishes might lose their autochtonous parasites during marine dispersal events. This would allow subsequent recolonisation by "local" parasite species once arriving in freshwater at the other end of a marine barrier (Pariselle et al. 2011).
Here, a Gyrodactylus species parasitizing I. hormuzensis is described. We analyze its affinity to congeners infecting other cichlids and to species known from fishes of the same region. Because of the ability to switch hosts, gyrodactylids are expected to mainly mirror continental affinities rather than the relationships between its host species (Boeger et al. 2003), depending of course on the time-scale and the dispersal routes used. The result might hence shed light on the pathway cichlids used to reach southern Iran.

Material and Methods
Eight host fishes (fixed and stored in ethanol or in formaldehyde) were retrieved from collections of the Royal Museum for Central Africa (Tervuren, Belgium) (MRAC B2-28-P-1-8). Branchial arches, body, fins and recipient were inspected for parasites under a Wild M8 stereomicroscope. Monogeneans were removed with a dissection needle. They were treated with 10% sodium dodecyl sulphate for a couple of minutes, before being fixed using ammonium picrate-glycerine (Malmberg 1957) and mounted on a slide under a cover-slip. Measurements were adapted from Shinn et al. (2004) using a Zeiss Axio Imager Z1 microscope at a magnification of 100 x (oil immersion, 10 x ocular) under differential interference contrast, with an AxioCamMR3 camera and AxioVision v.4.2.8 software. For drawings, some specimens were stained with Gomori's trichrome and mounted in Canada balsam for study of their soft anatomy; other specimens were cleared and mounted in Hoyer's or Gray and Wess' media for study of their sclerotized structures (all solutions prepared as in Humason, 1979). Illustrations were prepared with the aid of a digital camera (5 megapixels) and a projector attached to an Olympus BX51 microscope equipped with phase contrast. Taxon and author names in this study follow Eschmeyer (2012) for hosts and MonoDb (http://www.monodb.org) for Monogenea.

Results
A single (incomplete and therefore unfortunately unidentified) ancyrocephaline monogenean (Dactylogyridae, Ancyrocephalinae) was recovered. All other monogeneans found on the I. hormuzensis specimens belong to Gyrodactylus.
He was a researcher in aquatic animal health and fish parasitology at the Veterinary Department of the Islamic Azad University (Iran) and a keen student of monogeneans. The authors express the hope that this patronym might serve as an indication for the respect and appreciation this kind man enjoyed from his colleagues.

Type material examined
Thirty-one specimens, twenty of which (ethanol-preserved) were used for measurements. The holotype (

Description
(measurements in micrometres (µm) and angles in degrees (°); average ± standard deviation, followed by range and number of measurements in parentheses).
Comparison to Palearctic congeners followed Pugachev et al. (2009). The rather large ventral bar processes, in combination with the length of the marginal hooks, and MCO armed with one large apical spine and one row of smaller spines of similar size, resemble the morphology of G. ophiocephali Gussev, 1955 ). However, in G. ophiocephali and G. tokobaevi, the processes are longer than the ventral bar proper (median length, i.e., without the shield), which is not the case in G. jalalii sp. nov. Elongate antero-lateral processes, albeit not longer than the ventral bar proper, are also found in G. hrabei Ergens, 1957and G. mariannae Winger, Hansen, Bachmann & Bakke, 2008, parasites of Cottus Linnaeus, 1758. These species, however, have a comparatively shorter anchor root than G. jalalii sp. nov. The longitudinal striae on the ventral bar shield as well as an overlap in size of anchor and marginal hook are reminiscent of G. lotae Gussev, 1953 from Lota lota (Linnaeus, 1758) (Gadiformes, Lotidae). This species can be distinguished from G. jalalii sp. nov. because the new species has blunter and larger ventral bar processes, and a marginal hook sickle toe which joins smoothly into the platform, whereas this transition leaves a sharp "bump" in the platform in G. lotae.

Discussion
Just as Barlow (2000) praises cichlids, early students of Gyrodactylus expressed their high expectations for the scientific interest of these flatworms, as they recognized the species richness of this genus and the amount of work and skill it would take to adequately study it (von Nordmann 1832; Van Beneden & Hesse 1863). Given this scientifically challenging position of both cichlids and gyrodactylids, G. jalalii sp. nov. was described, a monogenean ectoparasite of the only Iranian cichlid I. hormuzensis. To the best of our knowledge, it is the first parasite to be formally described from this fish species. It brings the total number of Gyrodactylus species described from cichlid hosts to 15 (Paperna 1979;Boeger & Popazoglo 1995;Christison et al. 2005;Přikrylová et al. 2009Přikrylová et al. , 2012García-Vásquez et al. 2011;Vanhove et al. 2011). Jalali et al. (2001) observed that the highly distinct zoogeographical regions of Iran and the high degree of endemicity of its freshwater fishes gave rise to a diverse monogenean fauna. These authors did not mention Gyrodactylus in their overview. Indeed, Jalali et al. (2005) noted that representatives of this genus from the endemic freshwater fishes of the country are basically unknown and undescribed. They did not report gyrodactylids from Iranocichla. This cichlid seems distributed mostly in the Mesopotamian faunal region of Iran, from which Jalali et al. (2001) reported a rather unexplored, endemic and specific monogenean fauna, containing both Palearctic and African elements. For this reason, G. jalalii sp. nov. was compared to Palearctic and African congeners. Morphological similarities are unsurprisingly apparent (see above). Haptoral morphology alone does not allow reliable phylogenetic inferences for species of Gyrodactylus (Ziętara & Lumme 2004). Unfortunately, we were unsuccessful in amplifying DNA from G. jalalii sp. nov. Hence, genetic data are at present unavailable for the new species, as is the case for several congeners used in this comparison. However, it is noteworthy that some of the aforementioned species show affinities in molecular phylogenetic reconstructions. Indeed, G. mariannae is close to G. hrabei, which is genetically similar to G. flesi Malmberg, 1957(Winger et al. 2008. This species clusters with G. lotae and G. zimbae, among other species, in a Gyrodactylus clade containing representatives from various subgenera as defined by Malmberg (1970) and from many continents and host taxa (Ziętara & Lumme 2004;Vanhove et al. 2011). Despite the need for genetic data, assuming, that G. jalalii sp. nov. is indeed related to this lineage, this fact alone does not make it a very suitable candidate to provide information on biogeographical dispersal pathways of cichlid fishes. Indeed, several species also belonging to this diverse clade have been observed to show a broad salinity tolerance (G. branchialis Huyse, Malmberg & Volckaert, 2004 and G. ostendicus Huyse & Malmberg, 2004: Huyse et al. 2006 or the ability to switch infection sites (G. arcuatus Bychowsky, 1933: Raeymaekers et al. 2008 or hosts (G. arcuatus: Huyse et al. 2003Huyse et al. , 2006. However, while there is hence no proof that G. jalalii sp. nov. belongs to an exclusively African or cichlid-infecting lineage, this broad ecological spectrum enhances the potential of these parasites as biogeographical markers, conditional to the availability of molecular data.