Pelophylax cf. bedriagae

In the last three to four decades, Anatolian water frogs (Pelophylax cf. bedriagae) have been introduced into several western European countries mainly for culinary purposes but also as ornamental animals for garden ponds. P. cf. bedriagae comprises several distinct evolutionary lineages that exhibit a high genotypic and phenotypic variability. A clear differentiation of P. cf. bedriagae from closely related water frog species, for example P. ridibundus and P. bedriagae, requires molecular tools. P. cf. bedriagae may influence native water frog populations ecologically as predators and disease vectors, or genetically by transmitting allochthonous genes into the endemic gene pool. Rapid dispersion of P. cf. bedriagae is documented for Belgium, France, Switzerland, and south-western Germany. As a consequence fundamental genetic changes of the native gene pools and a decrease of autochthonous water frogs can be expected.

The lineage-specificity of mtDNA provides an opportunity to localize the area of origin of allochthonous water frogs. P. cf. bedriagae specific haplotypes (MHG4-6) are distributed from western Turkey to central Russia; haplotypes of this group have also been found in Iran and Syria (Plötner et al. 2001; Akin et al., 2010b; Ç. Akin et al., Middle East Technical University, Ankara, Turkey, unpublished data). In north-eastern Greece and west of the Caspian Sea, MHG6 occurs syntopically with haplotypes characteristic of European P. ridibundus (Akin et al. 2010b; Plötner et al. 2010; J. Plötner (Museum für Naturkunde Berlin, Germany) and S. N. Litvinchuk, Institute of Cytology, Russian Academy of Sciences, St. Petersburg, Russia), unpublished data). Haplotypes of MHG4 are found in the Cilician plain west of the Amanos Mountains and haplotypes of MHG5 east of it (Akin et al., 2010b; Plötner et al., 2010).

MHG6 can be divided into four subgroups (6a-d). Haplotypes of subgroup 6a (cf. caralitanus) are mainly found in southwestern Turkey. They occur syntopically with haplotypes of subgroup 6c (cf. bedriagaesensu stricto), for example, in southwestern and central Anatolia, and with haplotypes of subgroup 6b (cf. cerigensis) in southern Anatolia (Akin, 2007; Akinet al., 2010a, b). In the Gökova plain (Silifke) MHG6a overlaps with MHG4. MHG6b is distributed west of Antalya; haplotypes of this group have been found in several populations of the coastal area and on the nearby Greek islands of Rhodos (Rhodes) and Karpathos. Haplotypes of MHG6c are distributed from western Anatolia to the northern shore of the Caspian Sea; this group occurs in the whole of Anatolia except the costal region east of Antalya, and west and east of the Amanos Mountains. Haplotypes of subgroup 6c were also detected on several Mediterranean islands along the Anatolian coast, for example Chios, Lesvos (Lesbos) and Samos. Haplotypes of subgroup 6d (Euphrates) have been found in the Tigris and Euphrates catchments of north-eastern Syria, eastern Anatolia, and western Iran (Akin et al., 2010b; J. Plötner, Museum für Naturkunde Berlin, Germany, unpublished data).

Allochthonous water frogs are still being imported and cultivated for culinary purposes, especially in France (Pagano et al., 2001; 2003; Schmeller et al., 2007) and Italy (C. C. Bilgin and Ç. Akin (Middle East Technical University, Ankara, Turkey), and J. Plötner (Museum für Naturkunde Berlin, Germany), unpublished data), and for pet shops and garden ponds, for example in Belgium (Holsbeek et al., 2008, 2010). In this country a growing demand for water frogs by ornamental pond owners is recognized. Although intended for garden ponds, these frogs often escape into natural biotopes. A successful establishment of Anatolian water frogs (P. cf. bedriagae) in western and Central Europe is probably linked with global warming; frogs are now able to hibernate in a more temperate climate. Further spread of P. cf. bedriagae is favoured by running waters (rivers, channels, ditches, etc.). Although there is no evidence for introductions with fish spawn, this possibility cannot be excluded completely. To prevent further spread of P. cf. bedriagae to "unpolluted" areas such as eastern Europe, the world-wide trade in living water frogs including eggs and tadpoles must be stopped by legislation.

P. cf. bedriagae are semi-aquatic frogs, inhabiting (and breeding in) a wide variety of flowing and stagnant water habitats such as reservoirs, small ponds, banks of slowly flowing streams and rivers, irrigation channels, ditches, and even temporary water bodies (e.g. Beerli, 1994; Schneider and Sinsch, 2001; Akin et al., 2010a; J. Plötner, Museum für Naturkunde Berlin, Germany, unpublished data). P. cf. bedriagae, like closely related species of the P. ridibundus group, is probably tolerant of stagnant and fairly brackish water (e.g. Günther, 1990; Baier et al., 2009). Typical breeding waters are shallow (50-150 cm) and have a high sun exposure and a rich vegetation comprising plants such as Eleocharis palustris, Carex elata, Potamogeton spp., and Ceratophyllum submersum (e.g. Ayaz et al., 2007). On the other hand, water frogs have also been observed in small dirty ponds, streams, and rivers polluted with agricultural and industrial sewage, especially in western Anatolia (Ç. Akin, Middle East Technical University, Ankara, Turkey, personal communication, 2009). Because of their relatively broad ecological plasticity, P. cf. bedriagae can adapt easily to the ecological conditions of western Europe; here, they occur mainly along rivers and floodplains (e.g. Ohst, 2008; Holsbeek et al., 2008) but also in garden ponds (Holsbeek et al., 2008, 2010).

It is assumed that P. cf. bedriagae, like its close relative the European lake frog P. ridibundus, hibernates in the beds of water bodies (e.g. Berger, 1982). Hibernation under water potentially poses physiological problems because of the scanty amount of oxygen in the water (e.g. Lutschinger, 1988). Oxygen deficiency may occur during cold winters when the water is covered by ice; such conditions often cause a high mortality.

Genetics

The diploid genome of Anatolian water frogs (P. cf. bedriagae) consists of 2N=26 chromosomes, five larger and eight smaller pairs (e.g. Alpagut and Falakali, 1995). Population-specific differences in the morphology of some chromosomes were described by Alpagut and Falakali (1995) and Alpagut Keskin and Falakali Mutaf (2006), these probably reflect lineage-specific characters of caralitanus and non-caralitanus frogs.

P. cf. bedriagae possesses several segregating protein-coding alleles (Beerli et al., 1996; Jdeidi, 2000; reviewed by Plötner and Ohst, 2001) that can be used as diagnostic markers to differentiate between allochthonous and autochthonous (European) water frog lineages (Pagano et al., 2003). Likewise, microsatellite markers can be used to assign individuals to parental species or to identify them as putative hybrids (Holsbeek et al., 2008). To date, however, no fixed  alleles have been found in Anatolian frogs. Among 20 Turkish populations represented by 378 specimens, five out of twelve enzyme-coding loci were polymorphic within or between populations; the highest genetic variability was detected in Lake District populations (cf. caralitanus), with an average heterozygosity of 0.12 (Jdeidi, 2000).

In their mitochondrial DNA, Turkish water frogs exhibit huge variability; among 612 sequences, 116 mitochondrial haplotypes were detected on the basis of the 340-bp NADH dehydrogenase subunit 3 (ND3) gene alone; 119 sequences of the ND2 gene yielded 80 haplotypes (Akin et al., 2010b). Because there are many lineage-specific substitutions, both genes are appropriate markers for reconstructing mitochondrial phylogenies (Plötner 2005, Akin et al. 2010b, Plötner et al. 2010). The non-coding nuclear internal transcribed spacer 2 (ITS-2) and the serum albumin intron 1 (SAI-1) possess sites specific for Anatolian water frogs (P. cf. bedriagae) and European water frogs (P. ridibundus), respectively (Ohst, 2008; Plötner et al., 2009). In all western Palaearctic water frogs investigated, the SAI-1 contains a 5' truncated chicken-repeat-like retroelement named RanaCR1 (Plötner et al., 2009). The 3’ end of RanaCR1 is defined by a perfect octameric repeat. Length variation in water frog SAI-1 sequences is caused by deletions at both the 5’ and 3’ ends of RanaCR1. Unlike other CR1 elements, RanaCR1 contains a CA microsatellite upstream of the octameric repeats in its 3’ UTR. Several sites of RanaCR1 are species-specific, so this retroelement can be used as a diagnostic marker. Length polymorphisms and specific nucleotide substitutions of RanaCR1 can be used as diagnostic characters for differentiation between distinct water frog lineages (Plötner et al., 2009), but introgression means that other markers must also be used -- a single marker is insufficient for species identification (J. Mergeay, Instituut voor Natuur- en Bosonderzoek, Geraardsbergen, Belgium, personal communication, 2014).

Reproductive biology

Little is known about reproductive biology of P. cf. bedriagae; it can be assumed, however, that they reproduce in a similar way to European lake frogs (P. ridibundus). Mating starts with the onset of spring, the particular date depending on the geographic region and the specific weather conditions. Generally, most matings occur in May and June (Basoglu and Özeti, 1973). In coastal areas of the Aegean region, the breeding season starts in early March and continues until the end of April (Çaydam, 1973 cited in Ayaz et al., 2007) while in regions with a continental climate or at higher altitudes the reproductive period can last through to the end of June.

Lake frog females (P. ridibundus) usually lay between 2,000 and 10,000 eggs in a few clutches among water plants (e.g., Günther 1990); the egg number correlates with body size (e.g. Berger and Uzzell, 1980). According to Basoglu and Özeti (1973), P. cf. bedriagae females lay 5000-10000 eggs in a few clutches in open water or among aquatic plants.

Larvae hatch approximately 7-10 days after eggs have been laid and complete metamorphosis two to three months later, depending on water temperature and food supply. Water temperatures below 11 °C and above 33 °C are unfavourable for normal embryonic development of central European lake frogs (Günther, 1974); embryogenesis of P. cf. bedriagae is thought to proceed in a similar range of water temperature values (13.3 °C to 33.5 °C) to that in which the adult males call (Schneider and Sinsch, 2001).

If the environmental conditions during larval development are not optimal, a few tadpoles may hibernate and complete metamorphosis in the year after hatching, especially in continental regions.

Nutrition

P. cf. bedriagae feeds mainly on terrestrial arthropods (especially insects and spiders), molluscs, nematodes, and annelids, but has also been observed to prey on small vertebrates, for example fish, bird nestlings, small mammals, and other amphibians including juveniles of their own species (e.g. Yilmaz and Kutrup, 2006, Çiçek and Mermer, 2006, 2007). Adult P. cf. bedriagae are generalist and opportunistic predators; their diet is strongly influenced by prey availability. In contrast to adults, tadpoles are herbivores and detritivores; they eat algae, dead organisms, and plants.

Associations

In its natural range, P. cf. bedriagae is commonly associated with the southern crested newt (Triturus karelini), the fire bellied toad (Bombina bombina), the common tree frog (Hyla arborea), and the green toad (Bufo viridis) (Ayaz et al., 2007). In Central Europe it co-occurs in the same habitats as the endemic water frog forms P. ridibundus and P. kl. esculentus (Ohst, 2008; Holsbeek et al., 2008). In Anatolia, it is typically associated with certain reptile species, for example grass snakes (Natrix natrix), dice snakes (Natrix tessellata), and the European pond turtle (Emys orbicularis), which all prey on water frogs. Larger water bodies, such as channels and lakes, are usually inhabited by different fish communities and aquatic birds that may prey on water frogs.

Potential predators for adult water frogs include several mammals (e.g. racoon (Procyon lotor) and polecat (Mustela putorius)), bird species (especially egrets (Ardeidae) and storks (Ciconiidae)), water snakes (genus Natrix) and turtles (Emys spp., Mauremys spp.). Carnivorous water insects (e.g. dragonfly larvae (Odonata) and great diving beetles (Dytiscidae)), many fish species, and different aquatic birds (e.g. ducks (Anatidae)) feed on eggs and larvae; some carnivorous fish species (e.g., pike, Esox lucius) even prey on juvenile and adult frogs (Ayaz et al., 2007; reviewed by Günther, 1990).

General

• Pet/aquarium trade

Human food and beverage

• Meat/fat/offal/blood/bone (whole, cut, fresh, frozen, canned, cured, processed or smoked)

A clear identification of allochthonous water frogs and hybrids between allochthonous and autochthonous forms requires molecular tools. A combination of mitochondrial and nuclear markers is the best diagnostic procedure, as mitichondrial DNA sequences do not distinguish between species and species hybrids -- they only provide evidence for a maternal genetic impact of an allochthonous species or lineage on a native population. Appropriate mitochondrial markers are, for example, the ND1 gene (Holsbeek et al. 2008), the ND2 and ND3 genes (e.g. Plötner and Ohst, 2001, Akin et al., 2010b), and the Cytb gene (Lymberakis et al., 2007). Nucleotide sequences of the nuclear serum albumin intron-1 (SAI-1) containing a 5’ truncated non-long terminal repeat retrotransposon (RanaCR1) allow a clear differentiation between pure lineages of Central European P. ridibundus and Anatolian P. cf. bedriagae (Plötner et al., 2009; J. Plötner (Museum für Naturkunde Berlin, Germany) and Ç. Akin (Middle East Technical University, Ankara, Turkey), unpublished data); introgressed individuals or hybrids cannot be identified in this way, but Holsbeek et al. (2008) could distinguish among both pure matrilines and intrigressed individuals using clustering approaches on the basis of microsatellite markers. As well as the aforementioned publications, Plötner et al. (2001, 2008) provide details on PCR and DNA sequencing.

Single morphological characters or character combinations do not allow a clear differentiation between Anatolian P. cf. bedriagae and European P. ridibundus (e.g., Jdeidi, 2000); there is a great overlap in many morphometric parameters, and hybrids between the two forms may reduce the discriminative power of statistical methods because they show intermediate values in many morphometric characters. Multivariate approaches can be used to distinguish allochthonous from autochthonous forms; for example, based on discriminant analysis, Sinsch and Schneider (1999) were able to assign about 89% of individuals correctly to P. cf. bedriagae or P. ridibundus.

In the field, specialists may identify P. cf. bedriagae on the basis of their mating calls. Compared to Central European P. ridibundus, the number of pulse groups per call is greater and the pulses per pulse group are fewer in P. cf. bedriagae (Schneider and Sinsch, 1999; reviewed by Plötner, 2005). For an exact species diagnosis, however, direct comparisons of mating calls can be done only if the calls are recorded at the same water temperature, because many call parameters are temperature-dependent (e.g. Schneider and Sinsch, 2001).

Precise identification of allochthonous water frogs and their differentiation from autochthonous forms requires molecular markers. Besides specific enzymes (Beerli et al., 1996; Jdeidi, 2000; reviewed by Plötner and Ohst, 2001), a combination of mitochondrial and nuclear DNA sequences allows a clear differentiation between allochthonous and autochthonous water frogs and their hybrids (Ohst, 2008). Serum albumin intron 1 (SAI-1) and the embedded retroelement RanaCR1, for example, are specific markers that can be used to distinguish between the nuclear genomes of Anatolian P. cf. bedriagae, European P. ridibundus, and Balkan P. kurtmuelleri (Plötner et al., 2009).

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Prevention

There is no practical way to prevent the further expansion of allochthonous water frogs, among them P. cf. bedriagae, in western Europe. Eradication of non-native water frogs seems to be impossible. To prevent additional introductions of allochthonous water frogs, the international trade and sale of water frogs has to be prohibited by legislation. In small areas with known recent introductions water frogs could be tested genetically and allochthonous forms or hybrids between allochthonous and autochthonous lineages could be removed.

Control

The occurrence of lake frogs, considered as P. ridibundus, in areas where such forms were never detected before, is a first indication that allochthonous water frogs may be present. In such cases, frogs should be screened genetically and individuals with allochthonous genes should be removed from the population. A complete eradication of allochthonous water frogs seems possible only in small populations that live in clearly structured ecosystems and at a very early stage of invasion. If introgression of allochthonous genes into the indigenous gene pool has already occurred, the allochthonous form should be accepted as an established (naturalized) element of the ecosystem. Populations with allochthonous forms should be monitored to gain more insights into the population dynamics of the invader and to detect factors linked with its invasiveness. The general public has to be informed via mass media of the detrimental consequences that invasive water frog species may have for ecosystems and native water frog populations.

There is still a big gap in the ecological and physiological data necessary to estimate the fitness (and thus the invasive potential) of P. cf. bedriagae outside its natural distribution area. Additional genetic data from natural populations are required to evaluate the extent of genetic pollution caused by P. cf. bedriagae lineages and other allochthonous forms.

Europe: SEH (Societas Europaea Herpetologica), Museum Natural History and Territory, University of Pisa, via Roma 79, 56011 Calci (Pisa), Italy, http://www.seh-herpetology.org/

Germany: DGHT (Deutsche Gesellschaft für Herpetologie u. Terrarienkunde), Postfach 14 21, 53351 Rheinbach, http://www.dght.de/

Switzerland: KARCH (Koordinationsstelle für Amphibien- und Reptilienschutz in der Schweiz), Passage Maximilien-de-Meuron 6, 2000 Neuchâtel, http://www.karch.ch

03/11/09 Original text by:

Jörg Plötner, Museum für Naturkunde Berlin, Leibniz-Institut für, Evolutions- und Biodiversitätsforschung, an der Humboldt-Universität zu Berlin, Invalidenstrasse 43, D-10115 Berlin, Germany