DAPHNIOLA RADOMAN, 1973 (CAENOGASTROPODA: TRUNCATELLOIDEA) AT EAST AEGEAN ISLANDS

a bStract : Shell habitus and COI mitochondrial gene sequences of one freshwater snail from Khios and three from Rhodes islands were analysed. Both methods confirmed assignment of these specimens to the genus Daphniola Radoman, 1973. Genetic distance between individuals from these two islands is surprisingly low, strongly suggesting that they belong to the same species, still undescribed. Comparison of COI sequences with other known species of this genus shows that the closest relative of the Khios and Rhodes populations is D. louisi Falniowski et Szarowska, 2000 from Attica. The results are discussed in the context of geological and climatic history of the Mediterranean.


INTRODUCTION
A number of freshwater valvatiform hydrobiid gastropods that occur in European habitats were recently reviewed by bodon et al. (2001). radoMan (1973) described a new genus Daphniola, with its new type species D. graeca Radoman, 1973, from Daphne Spring in Témbi Valley, northern Greece ( Fig. 1). In his review of the Greek Hydrobiidae, ScHütt (1980) assigned the valavatiform hydrobiids, which had been reported from that territory, to Horatia Bourguignat, 1887. He not only expanded the range of this genus far to the south without any justification (radoMan 1983), but also he included the species of Daphniola in Horatia unreasonably. Our molecular data did not confirm any closer relationship between Horatia and Daphniola (SzarowSka & falniowSki 2014).
There are three species assigned to Daphniola currently. ScHütt (1980) synonymised Daphniola graeca with D. exigua (A. Schmidt, 1856). gittenberger (1982) described a new species of Horatia (Horatia hadei Gittenberger, 1982), found 5 km SW of Yíthion (Gythion), Peloponnese (Fig. 1) In this work we describe Daphniola individuals from two East Aegean islands. In June 2012 we found a single specimen of Daphniola at Khios Island, and three specimens at Rhodes Island. The material was too scarce for any morphological study, but we obtained cytochrome oxidase subunit I (COI) mtDNA partial sequences to confirm the genus-level assignment of those specimens, and to infer their phylogenetic relationships with the other, previously described continental representatives of Daphniola.

MATERIAL AND METHODS
The snails were collected from two islands, situated in the eastern Aegean Sea (Fig. 1): -Khios Island, a small spring (Fig. 2) near the road, 4 km E of Leptópodha, 38°34'23.6"N, 25°59'12.3"E, 390 m a.s.l., together with Pseudamnicola and Bythinella, only a single specimen of Daphniola was found; -Rhodes Island, a spring ( Fig. 3) 5 km E of Sálakos, 36°17'0.4.1"N, 28°00'30.6"E, 210 m a.s.l.; together with Pseudamnicola, only three juvenile specimens of Daphniola were found. Snails were washed twice in 80% ethanol and left to stand in it for around 12 hours. The 80% ethanol was then changed twice more within 24 hours and finally, after a few days, replaced with 96% ethanol, in which the samples were then stored at -20°C. The shells were photographed with a CANON EOS 50D digital camera mounted on a NIKON SMZ-18 microscope.
The PCR conditions were as follows: initial denaturation step of 4 min at 94°C, followed by 35 cycles of 1 min at 94°C, 1 min at 55°C, 2 min at 72°C, and a final extension of 4 min at 72°C. The total volume of each PCR reaction mixture was 50 μl. To check the quality of the PCR products, 10 μl of the PCR product were run on a 1% agarose gel. The PCR products were purified using Clean-Up columns (A&A Biotechnology) and were then amplified in both directions (HilliS et al. 1996) with the primers described above, using BigDye Terminator v3.1 (Applied Biosystems) according to the manufacturer's protocol. The sequencing reaction products were purified using ExTerminator Columns (A&A Biotechnology); DNA sequences then underwent electrophoresis on an ABI Prism sequencer. The sequences KM887914 and KM887915 were deposited in GenBank for Daphniola from Khios Island and Rhodes Island, respectively. They were aligned manually in Bioedit 7.1.3.0 (Hall 1999

RESULTS
The shells of the single specimen from Khios (Fig.  4), as well as of the specimens from Rhodes (Figs 5-6) were typical of Daphniola.
The pairwise p-distances (Table 1) confirmed the assignment of the specimens from both Khios and Rhodes islands to Daphniola. The distance between the populations from the two islands was 0.005, and the distances between those populations and the other Daphniola species were similar to those between the latter species. For ML analysis, the Hasegawa, Kishino & Yano (HKY) model (nei & kuMar 2000) with Г distribution was found. The ML tree (Fig. 7) Figs 4-6. Shells of Daphniola from Aegean Islands: 4 -Khios Island, 5-6 -Rhodes Island. Bar represents 0.5 mm

DISCUSSION
Both p-distance and the inferred ML tree indicate that the populations of Daphniola from Khios and Rhodes islands belong to the same species, despite the long distance and salt water isolation between those islands. Interestingly, the distance is more than three times smaller than between D. exigua and D. graeca living in the same valley and considered by falniowSki et al. (2007) as a single species. Perhaps either founder effect or strong selection may be an explanation. On the other hand, those populations represent a distinct species of Daphniola, at least as distinct molecularly as the three usually recognised species. Similarly high distances were found between many truncatelloidean species , falniowSki et al. 2009, falniowSki & SzarowSka 2011a. The sister group relationship of the island Daphniola with D. louisi from Attica is noteworthy as well.
The geographical history of Daphniola seems complex. According to our results, the splitting between the two main Daphniola clades occurred about 5.99 Mya, but with confidence intervals it is comparable with earlier estimates ) and coincides with the Messinian Salinity Crisis, during which at the place of the Recent Mediterranean there was a desert with vast canyons of big rivers and with some water bodies, characterised by too low or too high salinity. Splitting time intervals occurring between the Khios/Rhodes/D. lousi clade and D. hadei from the Peloponnese could also indicate isolation of the Peloponnese populations of Daphniola at the end of the Messinian Crisis. The rapid changes in hydrographic conditions caused by the Zanclean flooding, restored the sea in the Mediterranean and thus formed barriers for freshwater fauna and changed climatic conditions. The geological and climatic events The next event involved splitting of D. exigua and D. graeca, at present found in the same valley in Thesally. This is probably a relic of recent differentiation of the Balkans' Daphniola, but more sampling is necessary for phylogeographic inferences. On the other hand, genetic distances between these two clades are of the same magnitude as those within these two populations, thus D. exigua and D. graeca are conspecific .
The low genetic divergence between D. louisi from Attica and the two island populations is surprising. The split between these two clades occurred about 2.8 Mya, long after the Messinian Crisis, so other events may have played a role in the divergence of these groups. It has to be noted that the present geography of the Aegean is no more than 8-6 Kya old (kougiouMoutziS et al. 2014: references therein). Since the peak of Wisconsin-Würm glaciation (24-10 Kya), global sea level has risen by 120-130 m, and the Aegean islands replaced dry land; in the late Pleistocene the eastern Aegean islands were connected with Anatolia, and the strait between Europe and the Cyclade-mega-island was rather narrow (kougiouMoutziS et al. 2014).
High similarity between the Khios and Rhodes populations is unexpected, since the distance between these islands is longer than 250 km. The most likely explanation may involve recent dispersal of Daphniola into one of these islands. Recent over-sea dispersal may be possible by stepping stones through the neighboring islands, or through the Anatolia peninsula. There are also other possible ways of colonisation, including introduction by man or by birds, which has been previously reported for other snails (e.g. Gittenberger et al. 2006). There are current data indicating that small snail species may survive the passage through the digestive tract of birds (Wada et al. 2012), which makes this way of long distance colonisation a probable hypothesis.