Herpetology Notes, volume 8: 589-598 (2015) (published online on 06 December 2015)

High incidence of dorsal dark spots in north-western Greek populations of alpine newts Ichthyosaura alpestris (Salamandridae, Caudata)

Haritakis Papaioannou1, Anna Batistatou2, Maria-Christina Mavrogiorgou3, Constantinos Konidaris4, Stathis Frillingos5, Evangelos Briasoulis3, Katerina Vareli3,6 and Ioannis Sainis3,*

Abstract. Large dorsal dark spots were observed in a number of alpine newts Ichthyosaura alpestris inhabiting Lake Gistova, the highest altitude alpine lake in the Balkans ever studied. The trait is mainly exhibited by female paedomorphs. Histopathological examination excluded any pathological cause of pigmented skin and revealed that even in paedomorphic skin sections, the altered pigmentation pattern was located only in regions with metamorphic characteristics. Phylogenetic analysis of the previously unstudied Gistova’s newt population based on mtDNA sequences revealed that it belongs to the D2 subclade, comprising only two other newt populations of the wider geographic region that interestingly, also exhibit the trait. The putative genetic or environmental basis for dark spot appearance in these populations is discussed.

Keywords: alpine lakes, Balkan Peninsula, black spots, 16S mitochondrial rRNA, mitochondrial cytb gene.

Introduction small animal with a total length of about 80 to 100 mm for males and up to 120mm for females (Sotiropoulos, The Balkan Peninsula is considered to be a major et al., 2008; Vukov et al., 2011). Phylogenetic analysis hotspot for European biodiversity (Griffiths et al., revealed the existence of a putative ancestor relict 2004; Strong et al., 2008). Balkan low-altitude lakes have been intensively studied in terms of biodiversity, as many of them are among the most ancient European lakes and thus old enough to feature endemic faunas and floras (Frogley et al., 2001; Frogley and Preece, 2007; 1 Department of Environmental and Natural Resources Albrecht and Wilke, 2008; Vareli et al., 2009; Wilke Management, University of Patras, 30100 Agrinio, Greece et al., 2010; Albrecht et al., 2012). In contrast, Balkan 2 Laboratory of Pathological Anatomy, Department of alpine lakes are less well studied, due to their tiny size Medicine, School of Health Sciences, University of Ioannina, and difficult access. Originated in the end of the last 45110 Ioannina, Greece 3 glacial period, only a few thousand years ago, their Interscience Molecular Oncology Laboratory (iMol), Cancer Biobank Center (UICBC), University of Ioannina, 45110 species assemblages probably derive from a process Ioannina, Greece of recent colonization from neighbouring, geologically 4 Laboratory of Biochemistry, Department of Biological older habitats (Shehu et al., 2009). Greek alpine lakes Applications and Technology, University of Ioannina, 45110, are small, often fishless and located at high-altitude Ioannina Greece areas above 1900m, near the summit of mountains. 5 Laboratory of Biological Chemistry, Department of Medicine, They are called “dragon-lakes” (Drakolimnes in Greek) School of Health Sciences, University of Ioannina, 45110 because their main vertebrate inhabitant, the alpine newt Ioannina, Greece 6 Laboratory of Biology, Department of Biological Applications (Ichthyosaura alpestris), is reminiscent of the shape and and Technology, University of Ioannina, 45110, Ioannina appearance of a small dragon. Greece Ichthyosaura alpestris (initially referred to as Triturus * Corresponding author e-mail: alpestris and more recently as Mesotriton alpestris) is a [email protected], [email protected] 590 Haritakis Papaioannou et al. lineage (Clade A) originating from south-eastern 1988) newt populations from lakes of the northern Serbia and two other lineages designated Western Pindus area were characterized as potentially endemic and Eastern further divided in Clades B, C and D, E and it has been reported that newts from lakes Tymphi respectively (Sotiropoulos et al., 2007). Clades D and and Smolikas exhibit colour variants (e.g. specimens E representing southern and central-northern Balkan with large black spots). However, no effort was made populations were further subdivided, suggesting greater in this study, either to examine histopathologically isolation into multiple refugia (Sotiropoulos et al., these colour variants or to reveal any correlation to 2007). More recently, Recuero and colleagues, by using population’s demographics. a combined analysis of mtDNA, nuclear (nDNA) and Thermoregulation, UV protection and avoiding allozyme data tried to resolve the evolutionary history detection from predators have been proposed as putative of the species I.alpestris (Recuero et al., 2014). The adaptive functions of colour patterns in amphibians phylogenetic hypotheses based on mtDNA from this (Rudh and Qvarnström, 2013). In common frogs the study recovered the same lineages and phylogenetic degree of melanism increases as a function of latitude structure as suggested earlier (Sotiropoulos et al., and altitude (Alho et al., 2010; Vences et al., 2002). 2007), but nDNA and allozyme analyses revealed more Dark markings in montane common frogs were found to complex phylogenetic relationships. Nevertheless it is be valuable to attain faster heating rates of higher body of interest that both nDNA and allozyme data support temperatures, and thereby performance benefits (Vences also that southern Balkan populations (clade D) is a et al., 2002). In larvae of newts and salamanders, UV discrete phylogenetic lineage (Recuero et al., 2014). induces skin darkening implying that melanization may Paedomorphosis is common in a vast number of newt have a protective role against UV radiation. However, populations living in the south margin of the species direct experimental evidence for such a role is missing geographical range, mainly in the Italian and Balkan (Rudh and Qvarnström, 2013). Peninsulas (Denoël et al., 2001). It is a polymorphism On the other hand, in the spotted salamander widespread in newts and salamanders and by which Ambystoma maculatum, UV induces DNA damage larval characters, such as gills and gill slits, are despite the increased melanin production in this species retained in adults (paedomorphs) (Whiteman, 1994). (Lesser et al., 2001). Furthermore, dark spots in some The adults characterized by absence of gill slits are Urodele amphibian species (e.g. Calotriton arnoldi, called metamorphs. In the five Greek paedomorphic Triturus cristatus) were found to be of cancerous origin populations examined so far, the percentage of (Koussoulakos, et al., 1994; Martínez-Silvestre et al., paedomorphs in the total population ranged between 2011). 23% and 77% (Denoël et al., 2001). Paedomorphosis is Based on these facts, we set the following specific under genetic control in several species of salamanders objectives for our study: (a) to investigate the histological (Voss and Shaffer, 1997). A genetic basis favouring nature of the dark spots, (b) to study the distribution of paedomorphosis in I. alpestris is also suspected (Denoël the trait by sex and life history type, (c) to investigate et al., 2001). Moreover, resource partitioning between the presence/absence of the trait in other geographically morphs is in favour of the maintenance of polymorphism neighbouring populations and (d) to resolve the in this species (Denoël and Joly, 2001a; Denoël and phylogenetic relationship of Gistova’s unstudied newt Joly, 2001b). population to other newt populations exhibiting the The Greek alpine newt was classified as a subspecies same colour trait. named Ichthyosaura alpestris veluchiensis (Breuil and Parent, 1988; Denoël, 2004). Although I. alpestris Materials and Methods specimens in northwestern Greece are found in various Study Area water bodies above 1190m such as ponds, drinking troughs and watering basins (Denoël, 2004), large Our study area comprises three alpine lakes (Figure. populations inhabit only the alpine lakes of the region 1). Lake Gistova, located on Mt. Grammos (40o21’N, (Denoël and Schabetsberger, 2003). 20o47’E), is the highest lake in Greece (2360m a.s.l) In our survey conducted in October 2012, large dorsal (Figure 1). It is a fishless elliptical lake (150m x 70m), dark spots were observed in a number of newts from with a maximum depth of about 4m, characterized by Lake Gistova, the highest altitude lake of Greece (2360m a muddy bottom, rocky shores, surrounded by alpine a.s.l) and one of the highest in the Balkans (Sehu et al., pasture. Aquatic vegetation is absent along the shoreline 2009). Interestingly, in a 1988 study (Breuil and Parent, and in the lake. High incidence of dorsal dark spots in Ichthyosaura alpestris, north-western Greece 591

Figure 1. A) Distribution of genetically, already studied I. alpestris populations in SW Balkans. B) A satellite view of the study area in the mountains of NW Greece. The studied populations are indicated with square dots. Known, but genetically unstudied populations are indicated with red dots. (a: Drakolimni Gistova, b: Drakolimni Smolika, c: Drakolimni Tymphis). (Satellite photos by Google-earth) [1. Jankovac (Papuk Mt. Croatia), 2. Kutarevo (Lika , Croatia), 3. Zegar (Usljerbka, Croatia), 4. Joseva (Vajero Mt. Serbia), 5. Prokosko Lake (Vranica Mt. Bosnia Herzegovina), 6. Ljubina bara (Juzni Kucaj Mt. Serbia), 7. Seljani (Rogatica, Bosnia Herzegovina), 8. Zminicko Lake (Sinjavina Mt. Montenegro), 9. Kapetanovo Lake (Lukavica Mt. Montenegro), 10. Bukumirsko Lake (Zijovo Mt. Montenegro), 11. Vlasina Lake, Serbia, 12. Vranje jezero (Sveti Ilija Mt. Serbia), 13. Zli Dol (Bosilegrand, Serbia), 14. Donje ravne mlake (Sara Mt. Serbia), 15. Satovica (W. Rodope Mts. Bulgaria), 16. Elatia (C. Rodope Mts. Greece), 17. Podgorecko Lakre (Jablanica Mt. FYROM), 18. Drakolimni Gistova (Grammos Mt. Greece), 19. Drakolimni Smolika (Smolikas Mt. Greece), 20. Drakolimni Tymphis (Tymphi Mt. Greece), 21. Zygos (Zygos Mt. Greece), 22. Elati (Kerketio Mt. Greece), 23. Velouchi (Velouchi Mt. Greece), 24. Katavothra (Oeta Mt. Greece), 25. Rakita (Panachaiko Mt. Greece), 26. Kyllini (Kyllini Mt. Greece)]

Two other lakes in the wider region are also included Carex sp. (Denoël and Schabetsberger, 2003). (Figure 1). The first one, located on Mt. Smolikas Lake Gistova differs from the other two lakes in being (40°05’N, 20°54’E, 2140m a.s.l), has a rectangular more homogeneous (lacking of shelters) with Lake shape (122m long, 61m wide) and a maximum depth of Tymphi being the most heterogeneous compared to the 3.7 m. Vegetation is limited to small shallow patches. other two. Gistova Lake lays 41.51 km away from Lake The bottom is muddy and rocks are rare (Denoël and Tymphi and 32.46 km away from Lake Smolikas. The Schabetsberger, 2003). The other one is located on distance between Lake Tymphi and Lake Smolikas is Mt. Tymphi (39°59’N, 20°47’E, 2000m a.s.l), has a 14.85 km. No exchange exists between newt populations somewhat quadratic appearance (max. diameter: 100m) inhabiting the three lakes due to impassable deep valleys and a maximum depth of 4.95m. It is characterized by that separate them (Figure 1). rich vegetation along the shoreline, consisting mostly of 592 Haritakis Papaioannou et al.

Sampling, gender ratios and proportion of life history mm diameter) on a Mini Beat-beater (Biospec products, phases-preservation of samples Bartlesville, OK, USA). Subsequently, genomic DNA was extracted with the nucleospin tissue kit (Macherey- We conducted three visits to sample newt populations, Nagel) according to the manufacturer’s protocol. during 14 and 15 October of 2012 at Gistova Lake, on For molecular phylogenetic analysis two target 23 October 2013 at Tymphi Dragon Lake and on 26 genes were used: a partial sequence (309bp) of the October 2013 at Smolikas Dragon Lake. mitochondrial protein-coding cytochrome b gene (cytb), Newt specimens (250 from each lake) were caught by and a partial sequence (596bp) of the mitochondrial dip-netting from the shores of each lake. Gender ratios gene coding for the 16S rRNA (16S). Primers and PCR and proportion of life history phases were measured in conditions have been described earlier (Sotiropoulos et the field. al., 2007; Korcher et al., 1989) Two larval, two paedomorphic, two metamorphic and PCR products were purified with a Macherey- seven spotted (both paedomorphic and metamorphic) Nagel DNA clean-up kit (Nucleospin Extract), newts from each lake were anaesthetized (0.5% and subsequently cloned with a TOPO TA cloning phenoxyethanol solution) and then fixed in 10% Kit (Invitrogen) according to the manufacturer’s buffered formalin to be histologically examined. instructions. Inserts were fully determined by sequencing both strands. Sequencing was performed by Histological evaluation of skin sections VBC- BIOTECH Service GmbH. Dorsal skin sections from both spotted and unspotted The cytb and mt16S rDNA sequences from Lake areas were taken from formalin fixed specimens and Gistova’s alpine newts were deposited in GenBank embedded in paraffin blocks. Tissue sections 5μm thick with accession numbers KF941301 and KF941302 were cut by using a microtome and applied on microscope respectively. slides. Subsequently, they were deparaffinized, hydrated and exposed to Mayer’s haematoxylin for 15 min, Phylogenetic analysis washed in tap water for 20 minutes and counterstained Cloned sequences KF941301 and KF941302 were with eosin for 1 min. Finally, sections were dehydrated compared to GenBank entries using BLAST in order to with alcohol/xylene baths and stabilized with mounting obtain a preliminary phylogenetic affiliation. medium. Cytb and mt16S rDNA sequences of Icthyosaura alpestris specimens already characterized as members Molecular analysis of the clades D, E and A (Sotiropoulos et al., 2007) were At Lake Gistova tail tips were obtained from five manually concatenated and along with the KF941301/ alpine newts both spotted and unspotted (metamorphs KF941302 concatenated sequence were imported into and paedomorphs) and homogenized in lysis buffer after MEGA 5.1. All sequences were automatically aligned adding three stainless steel beads per sample (Qiagen 5 using the integrated aligner tool.

Figure 2. Female paedomorph with a large dorsal dark spot. High incidence of dorsal dark spots in Ichthyosaura alpestris, north-western Greece 593

Phylogenetic analyses were performed with MEGA5.1 and two unspotted metamorphic newts were (Molecular Evolutionary Genetics Analysis) (Tamura histologically examined. In all cases only dorsal skin et al., 2011). Trees were constructed using either the sections, either spotted or unspotted, were studied. Neighbour-Joining (NJ) method with Jukes–Cantor In larval skin the epidermis consists of an inner layer distance correction or the Maximum Likelihood (ML) of basal cells, intermediate layers of Leydig cells and method. Both methods gave rise to identical phylogenetic a top layer of pavement cells (Figure 3A) (Brunelli et trees (Data not shown). Haplotype networks were built al., 2007). Small pale dark granules were dispersed using the minimum spanning algorithm (Bandelt et al., throughout the cytoplasm of the top layer of epidermal 1999) in PopART (Population Analysis with Reticulate cells giving the dark body appearance in these Trees) software (http://popart.otago.ac.nz). specimens. Epidermis in metamorphic newts consisted of stratified and flattened epidermal cells (Brunelli et Results al., 2007). Dark granules were absent (Figure 3B). The skin of paedomorphic individuals exhibited some Dorsal dark spot pattern of appearance in lake Gistova’s features of the larvae (such as Leydig cells) and some of alpine newts the metamorphic stage (such as stratified and flattened

A total of 250 individuals (Ichthyosaura alpestris) epidermal differentiating cells) (Figure 3C1, C2, C3). were collected from Lake Gistova and large dorsal dark In spotted areas, numerous large dark granules spots were found on 35 of them (14%) (Figure 2). were noted in the upper part of the cytoplasm in In half of the spotted newts, the spots were located on the top cellular layer, as well as between epithelial the head. The spots were circular or slightly elliptical cells throughout the epidermis (Figure 3D). It is (0.4-1cm in diameter). Both paedomorphic and noteworthy that in paedomorphic individuals the altered metamorphic newts were found with dark spots but this pigmentation pattern was observed only in skin regions trait was more prevalent in the paedomorphs as they with metamorphic characteristics. As proliferation accounted for the 66% of the spotted newts (Table 1). abnormalities, ulcerations or necrotizing lesions The percentage of paedomorphs in the total population were not detected, the possibility of a pathological was found to be 44%. The majority of spotted newts causative agent for black spot appearance was ruled out were females (95%) while females accounted only for (Koussoulakos et al., 1994; Martínez-Silvestre et al., the 61% of the total population (Table 1). 2011; Duffus, and Cunningham 2010).

Histological evaluation of spotted and unspotted skin Retention of the dorsal trait in other newt populations sections We studied the possible appearance of the same trait in Four spotted paedomorphic, three spotted two other newt populations inhabiting lakes in the wider metamorphic, two larval, two unspotted paedomorphic region. The same colour variant was also observed in

Table 1. Comparison of gender and life history phases proportions in both unspotted and spotted alpine newt populations in the three alpine lakes.

unspotted newts spotted newts

In the total ♀♂ paedomorphs In the total ♀♂ paedomorphs population population

Gistova 215/250 131/215 84/215 95/215 35/250 33/35 2/35 23/35 (86 %) (61 %) (39 %) (44 %) (14 %) (95 %) (5 %) (66 %)

Smolikas 195/250 144/195 51/195 137/195 55/250 54/55 1/55 44/55 (78 %) (74 %) (26 %) (70 %) (22 %) (98 %) (2 %) (80 %)

Tymphi 225/250 79/225 146/225 86/225 25/250 23/25 2/25 16/25 (90 %) (35 %) (65 %) (38 %) (10 %) (92 %) (8 %) (63 %) 594 Haritakis Papaioannou et al.

Figure 3. Histological examination of Haematoxylin-Eosin (H-E) stained skin sections of Ichthyosaura alpestris specimens from Lake Gistova. A: skin section from a larval specimen (400X). Small pale dark granules are indicated by arrows, B: metamorph’s skin section (400X), C: paedomorph’s skin section (40X) revealed its dual nature with either metamorphic (C2, 400X) or larval

(C3, 400X) characteristics, D: skin section of a spotted area taken from a paedomorphic female (400X). Note the presence of numerous large dark granules in the upper part of the cytoplasm in the top cellular layer, as well as between epithelial cells throughout the epidermis indicated by arrows. High incidence of dorsal dark spots in Ichthyosaura alpestris, north-western Greece 595

Figure 4. Haplotype network of mtDNA markers for the clades E and D of I. alpestris populations. Lake Gistova’s newt population is a distinct population of the subclade D2.

these geographically neighbouring populations. The Genetic relationship of Gistova’s newt population to frequency of appearance of these black spots was similar other newt populations in Lake Tymphi (10%) but higher in Lake Smolikas Cytochrome b and 16S rDNA gene fragments (22%) compared to Lake Gistova (14%) (Table 1). The amplified from five specimens (three spotted and two majority of affected newts were also females (92% in unspotted) were used to reveal the genetic relationship Lake Tymphi, 98% in Lake Smolikas) (Table 1). Both of Lake Gistova’s newt population to other previously paedomorphs and metamorphs were found to be affected studied newt populations. An haplotype network was with the paedomorphs being more susceptible than constructed based on previously published phylogenetic metamorphs (63% in Lake Tymphi and 80% in Lake results (Sotiropoulos K, et al., 2007) on Ichthyosaura Smolikas) (Table 1). Paedomorphs accounted only for alpestris (Figure 4). Based on the constructed haplotype 38% of the total population in Lake Tymphi while in network it is clear that Gistova’s newt population is a Lake Smolikas they accounted for 70% (Table 1). phylogenetically distinct population of the D2 subclade (Figure 4). 596 Haritakis Papaioannou et al.

Discussion According to a previously published study, eastern European Ichthyosaura alpestris paedomorphic The Ichthyosaura alpestris population from Lake lineages (such as subclade D2) seem to have Gistova represents the highest altitude population evolved at the early Pleistocene around 1.4-1.2 Mya ever studied in the Balkans (Figure 1). In this remote as a consequence of the oscillating glacial cycles population, dorsal dark spots were found to be present (Sotiropoulos et al., 2007). Moreover, the almost in 14% of the specimens collected along the shoreline. complete fixation of different haplotypes in many single Moreover, the observed colour variant was detected populations within each of the phylogroups identified, mainly on paedomorphic females. Histopathologic suggests that the relevant populations have survived examination ruled out the possibility that the pigmented the adverse climatic conditions of the Pleistocene in skin of our samples is of cancerous origin as it has been allopatry with limited postglacial expansion and contact suggested earlier for other Urodele amphibian species (Sotiropoulos et al., 2007). In contrast, according to a (Koussoulakos, et al., 1994; Martínez-Silvestre, et al., more recent study (Recuero et al., 2014), the split of the 2011). major eastern lineages (clades D-E) took place during Most interestingly, it revealed that even in the Miocene, in the Tortonian (7.3 Mya). paedomorphs, the altered pigmentation pattern It would be of interest to study the dorsal dark spot was located only in skin regions with metamorphic trait in populations from other phylogenetically related characteristics and not in larval-like skin regions subclades (such as subclade D1) (Figure. 4) as well (Figure 3). as in other already known newt populations from the In a previous study (Breuil and Parent, 1988) newt broader region of Northern Pindus, inhabiting ponds populations from lakes of the northern Pindus area were or swamps at lower altitudes for which systematic characterized as potentially endemic. The same study phylogenetic studies are currently missing (Figure 1). mentioned that newts from lakes Tymphi and Smolikas To date, I. alpestris populations of Northern Pindus exhibit colour variants (e.g. specimens with large black have been found at 15 sites all above 1,190m a.s.l. spots). However, no effort was made, in that study either (14 sites described in previous studies (Denoël, 2004) to examine those colour variants histopathologically, and one described in this study). Only three of these or with respect to any correlations with gender and 15 already known populations, have been studied life history type. We studied also those geographically phylogenetically mainly due to the fact that these neighbouring populations of the Northern Pindus lakes populations are large compared to the others which to confirm the presence and the histological nature of inhabiting ponds or swamps. The glacial sequence in the trait. Indeed, in Lake Smolikas 22% of the newts the Pindus Mountains, especially for Mt.Tymphi and were found to be affected while in Lake Tymphi only Mt. Smolikas constitutes the best-dated glacial record 10%. The histological appearance was the same and in the Mediterranean area (Hughes, et al., 2006). Based once again, paedomorphic females were found to on this glacial sequence, three separate glacial cycles prevail in the total spotted population as was the case have been identified (Hughes, et al., 2007). The oldest with Lake Gistova newts. one dates back to > 350,000 years BP with an ELA Based on our observations, a number of new questions (Equilibrium Line Altitude - “snowline”) of 1741m and arise. Since this trait has only been observed in the 1680m respectively, the middle one dates to 127,000 above three populations, is this indicative that it is under years BP with an ELA of 1862m and 1997m and the genetic control? In other words, is this a colour variant most recent one occurred during the last cold stage with characterizing potential endemic populations in this an ELA of 2174m and 2241m (Hughes, et al., 2006; area? Hughes, et al., 2007). Thus, it is obvious that all three By using mtDNA phylogenies, Gistova’s newt newt populations in our study have migrated to these population found to be a member of the D2 subclade high altitude lakes after the last glacial period. Given (Figure 4). Moreover, Gistova’s newt population is that ELAs for the oldest and the middle glacial periods distinct from the other two populations of the same are well above the altitude of the numerous small lakes, subclade described thus far and has to be referred to as ponds or swamps where other newt populations inhabit D2-3. In conclusion, the appearance of dorsal dark spots nowadays, further research efforts might be important is a common trait in populations of the D2 subclade, to focus on the lower altitude populations in search of a suggesting a possible genetic basis for its origin. putative parental population. High incidence of dorsal dark spots in Ichthyosaura alpestris, north-western Greece 597

Nevertheless, the already studied populations of necessary to allow a paedomorphic ontogenetic the D2 subclade are the highest altitude populations pathway in I. alpestris (Denoël et al., 2001). Based on examined thus far (Figure 4) and it is tempting to our demographic measurements the high percentage of speculate that their dorsal dark spots could be related affected newts in Lake Smolikas compared to the other to adaptations to high altitudes. Melanism is known two lakes is related to the fact that 74% of the lake’s to occur in amphibians from high altitudes and life total population are females and paedomorphs account history patterns are known to vary along altitudinal and for 70% of the total population. Moreover, although latitudinal gradients (Alho et al., 2010; Vences et al., paedomorphic populations are comparable in Lakes 2002). For instance, comparison of Rana temporaria and Tymphi and Gistova, the higher frequency of black I.alpestris populations showed that both species share spot appearance in the latter is probably related to the similar life history trait responses to increasing altitude higher percentage of females in the total population and latitude (Miaud and Merilä, 2001). Interestingly, (Table 1). Rana temporaria alpine populations consist of individuals with extended black dorsal pattern and Acknowlegments. We would like to acknowledge Prof. J. M. individuals with few dark spots (Vences et al., 2002). A Halley, Ms A. Makris and Mr V. Varelis for critically reading the potential thermoregulating function has been suspected manuscript and for helpful discussions. We also acknowledge Mr but no significant difference was found in the body Georgios-Arsenios Sainis for his help during sampling in Lake Gistova. temperature between dark and light individuals in the field (Vences et al., 2002). An experimental procedure led to the conclusion that the potential benefit of the dark References colouration may be that it allows a shift in time allocation Albrecht, C., Wilke, T. (2008): Ancient Lake Ohrid: biodiversity due to faster heating, an effect that is negligible under and evolution. Hydrobiologia 615: 103-140. the conditions prevailing in the environment (Vences et Albrecht, C., Hauffe, T., Schreiber, K., Wilke, T. (2012): Mollusc al., 2002). In our case, we did not perform experiments biodiversity in a European ancient lake system: lakes Prespa and Mikri Prespa in the Balkans. Hydrobiologia 47-59. that could provide evidence for the primary function of 682: Alho, J.S., Herczeg, G., Söderman, F., Laurila, A., Ingemar- the dark spots in I. alpestris specimens from the Greek Jönsson, K., Merilä, J. (2010): Increasing melanism along a alpine lakes but it seems unlikely that such spots serve latitudinal gradient in a widespread amphibian: local adaptation, thermoregulatory purposes in view of the fact that we ontogenic or environmental plasticity? BMC Evolutionary. did not find individuals with extended dark pattern but Biology 10: 317. only with one or two black spots. Bandelt, H., Forster, P., Röhl, A. (1999): Median-joining networks In larvae of newts and salamanders, UV induces for inferring intraspecific phylogenies. Molecular Biology and Evolution : 37–48. skin darkening, implying that melanization may have 16 Breuil, M., Parent, G. H. (1988): Essai de caractérisation du Triton a protective role against UV radiation. However, alpestre hellénique Triturus direct tests of the fitness significance of colouration in alpestris veluchiensis. II. Relations entre le triton alpestre hellénique protecting amphibians from UV-radiation are limited et la sous-espèce nominative. Alytes 7: 19-43. (Rudh, and Qvarnström 2013). With respect to a Brunelli, E., Perrotta, I., Bonacci, A., Tripepi, S. (2007): putative protective function of the black spots from Differential expression of aquaporin 3 in Triturus italicus UV-radiation, the higher frequency of the black spotted from larval to adult epidermal conversion. European Journal of newts observed in Lake Smolikas population compared Histochemistry 51: 25-32. Denoël, M. (2004): Distribution and characteristics of aquatic to Lake Tymphi could be attributed to the fact that Lake habitats of newts and Yellow bellied Toads in the district of Smolikas is higher in altitude and more homogeneous - Ioannina (Epirus, Greece). Herpetozoa 17: 49-64. with lack of shelters - than Lake Tymphi. Similarly, the Denoël, M., Joly, P. (2001a): Size-related predation reduces expected frequencies in Lake Gistova might be higher intramorph competition in paedomorphic Alpine newts. than those measured, since Gistova is higher in altitude Canadian Journal of Zoology 79: 943-948. and more homogeneous than the other two. Denoël, M., Joly, P. (2001b): Adaptive significance of facultative A possible genetic basis is also not to be excluded paedomorphosis in Triturus alpestris (Amphibia, Caudata): resource partitioning in an Alpine lake. Freshwater Biology because in all cases the trait seems to be prevalent in 46:1387-1396. female paedomorphs. Paedomorphosis is under genetic Denoël, M., Schabetsberger, R. (2003): Resource partitioning in control in several species of salamanders (Voss and two heterochronic populations of Greek Alpine newts, Triturus Shaffer 1997) and a genetic underpinning is probably alpestris veluchiensis. Acta Oecologica 24: 55–64. 598 Haritakis Papaioannou et al.

Denoël, M., Duguet, G., Dzukic, M., Kalezic, M., Mazzoti, Greece border) Biodiversity of an isolated microcosm. Thalassia S. (2001): Biogeography and ecology of paedomorphosis Salentina 32: 53-62. in Triturus alpestris (Amphibia, Caudata). Journal of Sotiropoulos, K., Eleftherakos, K., Džukić, G., Kalezić, Biogeography 28: 1271-1280. M.L., Legakis, A., Polymeni, R.M. (2007): Phylogeny Duffus, L.J.A., Cunningham, A.A. (2010): Major disease threats to and biogeography of the alpine newt Mesotriton alpestris European amphibians. Herpetology Journal 20: 117-127. (Salamandridae, Caudata), inferred from mtDNA sequences. Frogley, M.R., Preece, R.C. (2007): A review of the aquatic Molecular Phylogenetics and Evolution 45: 211–226. mollusca from Lake Pamvotis, Ioannina, an ancient Lake in NW Sotiropoulos, K., Legakis, A., Polymeni, R.M. (2008): Patterns of Greece. Journal of Conchology 39(3): 1-24. morphometric variation in the alpine newt (Mesotriton alpestris) Frogley, M.R, Griffiths, H.I, Heaton. T.H.E. (2001): Historical at the southern limit of its distribution: environmental correlates. biogeography and Late Quaternary environmental change of Integrative Zoology. 3: 123–133. Lake Pamvotis, Ioannina (north-western Greece): evidence Sotiropoulos, K., Tomović, L., Džukić, G., Kalezić, M.L. (2001): from ostracods. Journal of Biogeography 28: 745-756. Morphological differentiation of the alpine newt (Triturus Giacoma, C., Picariello, O., Puntillo, D., Rossi, F., Triperi, S. alpestris) in the Balkans: taxonomic implications. British (1988): The distribution and habitat of the newt (Triturus, Journal of Herpetology 11: 1-8. Amphibia) in Calabria (Southern Italy). Monitore Zoologico Strong, E.E., Gargominy, O., Ponder, W.F., Bouchet, P. (2008): Italiano 22: 449-464. Global diversity of gastropods (Gastropoda: Mollusca) in Griffiths, H.I., Krýstufek, B., Reed, J.M. (2004): Balkan freshwater. Hydrobiologia 595: 149-166 biodiversity, Pattern and process in the European hotspot. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Dordrecht, Boston, London, Kluwer Academic Publishers Kumar, S. (2011): MEGA5: molecular evolutionary genetics Hughes, P.D., Gibbard, P.L., Woodward, J.C. (2007): Geological analysis using maximum likelihood, evolutionary distance, controls on Pleistocene glaciation and cirque form in Greece. and maximum parsimony methods. Molecular Biology and Geomorphology 88: 242-253. Evolution 28: 2731-2739. Hughes, P.D., Woodward, J.C., Gibbard, P.L. (2006): Quaternary Vareli, K., Pilidis, G., Mavrogiorgou, M.Ch., Briasoulis, E., Sainis, glacial history of the Mediterranean mountains. Progress in I. (2009): Molecular characterization of cyanobacterial diversity Physical Geography 30: 334-364. and yearly fluctuations of Microcystin loads in a suburban Korcher, T.D, Thomas, W.K., Meyer, A., Edwards, S.V., Pääbo, Mediterranean Lake (Lake Pamvotis, Greece). Journal of S., Villablanca, F.X., Wilson, A.C. (1989): Dynamics of Environmental Monitoring 11: 1506-1512. mitochondrial DNA evolution in animals: Amplification and Vences, M., Galán, P., Vieites, R.D., Puente, M., Oetter, K., sequencing with conserved primers. Proceedings of the National Wanke, S. (2002): Field body temperatures and heating rates Academy of Sciences USA 86: 6196-6200. in a montane frog population: the importance of black dorsal Koussoulakos, S., Lelouda, M., Anton, H. J. (1994): Allografting pattern for thermoregulation. Annales Zoologici Fennici 39: of a nontransmissible, spontaneous dermal melanoma in the 209-220. newt Triturus cristatus. Tumor Biology 15: 141-146. Voss, S.R., Shaffer, H.B. (1997): Adaptive evolution via a Lesser, M.P., Turtle, S.I., Farrell, J.H., Walker, C.W. (2001): major gene effect: paedomorphosis in the Mexican Axolotl. Exposure to ultraviolet radiation (260-400 nm) causes oxidative Proceedings of the National Academy of Sciences USA 94: stress, DNA damage and expression of p53/p73 in laboratory 14185-14189. experiments on embryos of the spotted salamander, Ambystoma Vukov, T.D., Sotiropoulos, K., Kalezić, M.L., Džukić, G. (2011): maculatum. Physiological and Biochemical Zoology 74: 733- Morphing of the phylogeographic lineages of the Balkan alpine 741. newts (Ichthyosaura alpestris, Caudata, Salamandridae): In situ Martínez-Silvestre, A., Amat, F., Bargalló, F., Carranza, S. (2011): morphological Diversification. Comptes Rendus Biologies 334: Incidence of pigmented skin tumors in a population of wild 896–905. Montseny brook newt (Calotriton arnoldi). Journal of Wildlife Whiteman, H.H. (1994): Evolution of facultative paedomorphosis Diseases 47: 410–414. in salamanders. Quarterly Review Of Biology 69: 205-221. Miaud, C., Merilä, J. (2001): Local adaptation or environmental Wilke, T., Schultheiß, R., Albrecht, C., Bornmann, N., Trajanovski, induction? Causes of population differentiation in alpine S., Kevrekidis, T. (2010): Native Dreissena freshwater mussels amphibians. Biota 2: 31- 50. in the Balkans: in and out of ancient lakes. Biogeosciences 7: Recuero, E., Buckley, D., García-París, M., Arntzen, W. J., 3051-3065. Cogălniceanu, D.,Martínez-Solano, I. (2014): Evolutionary history of Ichthyosaura alpestris (Caudata, Salamandridae) inferred from the combined analysis of nuclear and mitochondrial markers. Molecular Phylogenetics and Evolution 81: 207-220. Rudh, A., Qvarnström, A. (2013): Adaptive colouration in amphibians. Seminars in Cell & Developmental Biology 24: 553-561. Shehu, M., Serravalle, F., Alfonso, G., Moscatello, S., Belmonte, Accepted by Arie van der Meijden G. (2009): The alpine Lake Gistova (Mount Gramos, Albania-