Five-toed jerboas (genus Allactaga) from the subgenus Paralactaga (sensu Shenbrot et al. 1995) have been a long-lasting source of disagreement over the number of species they encompass. The prevailing opinion distinguishes between two species, the smaller Allactaga euphratica Thomas, 1881, and the larger Allactaga williamsi Thomas, 1897 (Ellerman 1948, Ellerman and Morrison-Scott 1951, Holden and Musser 2005, Kryštufek and Vohralík 2005). The competing view claims that williamsi and euphratica are merely the extremes of a continuous variation and are, therefore, conspecific (Atallah and Harrison 1968, Corbet 1978, Harrison and Bates 1991, Holden 1993, Shenbrot et al. 1995). A meticulous revision of Allactaga in Turkey yielded a set of morphological characteristics between euphratica and williamsi, therefore, providing conclusive evidence that Paralactaga contains two species (Çolak et al. 1994). Another taxonomic uncertainty in Paralactaga concerns an isolated population in Afghanistan. Although originally described as a subspecies A. williamsi caprimulga (Ellerman 1948) and mainly considered in the scope of williamsi (Holden and Musser 2005), a recent study placed caprimulga closer to A. euphratica (Shenbrot 2009).
The chromosomal evidence is of no help in disentangling the relationships within Allactaga because of a stable chromosomal complement (Arslan et al. 2012), and the molecular assessments of the phylogenetic relationships among the species are in their very initial state (Dianat et al. 2012). The traditional taxonomy is based on the species delimitation on the color, size, and shape of the skull and glans penis. All these morphological traits are subjected to intraspecific variation (Shenbrot 2009), which diminishes their taxonomic value and compromises their power in detecting cryptic species diversity (Dianat et al. 2012). In this paper, we address the species taxonomy of the five-toed jerboas in the Middle East by applying molecular markers. Although our results are still preliminary, they clearly demonstrate the serious deficiencies in the current taxonomic arrangement of the subgenus Paralactaga.
Materials and methods
This study consisted of 15 Allactaga specimens from seven localities in Turkey, Syria, and Lebanon (Figure 1, Table 1). The samples were classified into two species on the basis of their geographical origins (Çolak et al. 1994, Abi-Said 2004, Holden and Musser 2005) and length of hindfoot as the only morphometric characteristic with non-overlapping ranges between the two species (Kryštufek and Vohralík 2005).
DNA extraction, PCR amplification, and sequencing
A 2×2-mm sample was dissected from the ethanol-preserved tissue and air dried under sterile conditions. The DNA was extracted using a QIAamp DNA Mini Kit (Qiagen, Valencia, CA, USA). The cytochrome b gene (cyt b) was amplified partially with primers H15319Marv, L15162Marv, L15408Marv (Haynes et al. 2003), H15752 (Ohdachi et al. 2006), L15702 (Harrison et al. 2003), L14727-SP, and H15915-SP (Jaarola and Searle 2002). The alignment of the three fragments yielded the sequence data for a partial cyt b gene (1104 bp). A polymerase chain reaction (PCR) was performed in a total volume of 15 μl containing 2.5 mm MgCl2, 0.3 μm of forward and reverse primers, 0.2 mm dNTPs, and 1 unit Fermentas Taq polymerase (Fermentas, Thermo Fisher Sci. Inc., Waltham, MA, USA) supplied with buffer containing (NH4)2SO4. The cycling conditions included an initial denaturation step at 95°C for 7 min, followed by 45 cycles of denaturation (1 min at 94°C), primer annealing (1 min at 48°C), and extension (2 min at 72°C). The final extension at 72°C was run for 10 min.
A partial sequence (∼326 bp) of the gene encoding 16S rRNA (16S) was amplified using the primers MT1-L and MT2-H (Bibb et al. 1981). Amplification was performed in a 15-μl reaction volume containing 2.5 mm MgCl2, 0.3 μm of forward and reverse primers, 0.2 mm of dNTPs, and 1 unit of Fermentas Taq polymerase supplied with buffer containing (NH4)2SO4. The PCR conditions consisted of an initial step at 94°C for 2 min, followed by 40 cycles of denaturation at 94°C for 60 s, annealing at 50°C for 60 s, and extension at 60°C for 60 s. The final extension at 72°C was run for 5 min. The sequencing was performed on an ABI PRISM 3130 Genetic Analyzer using BigDye Terminators (Applied Biosystems, Foster City, CA, USA).
The CodonCode Aligner 1.63 (Ewing et al. 1998) was used to align the forward and reverse sequences. The resulting consensus sequences for each individual were aligned using ClustalW 4.0, implemented in the Molecular Evolutionary Genetics Analysis (MEGA) package 5 (Tamura et al. 2011). The genetic distances were analyzed using the Kimura two-parameter (K2P) sequence evolution with 1000 bootstraps in the MEGA program. The nucleotide and haplotype diversities were assessed with the program DNASp (Rozas et al. 2003).
The best-fitting models of sequence evolution were determined using the MrModeltest 2.3 (Nylander 2004) for the Bayesian inference (BI). Both the Akaike information criterion (AIC) and the hierarchical likelihood ratio test (hLRT) were used.
The phylogenetic analysis of the concatenated sequences was performed using the MrBayes 3.1.2 program (Huelsenbeck and Ronquist 2001, Ronquist and Huelsenbeck 2003) for the Bayesian inference. Four Monte Carlo Markov chains were run simultaneously for 106 generations, with the resulting trees sampled every 10 generations. The Bayesian posterior probabilities (BPP) were used to assess the branch support of the BI tree. We considered BPP>0.95 as ‘good’ and PP=0.90–0.95 as ‘moderate’ support, in line with the other authors. The ML analyses were conducted using the RAxML program (version 7.3.0; Stamatakis 2006). The branch support in the ML tree was estimated by 1000 bootstrap replicates, and we accepted BP=90% as the cutoff for ‘good’ support. The trees were rooted with Jaculus jaculus (GenBank accession no. NC005314; Reyes et al., unpublished) for both cyt b and 16S, and Jaculus orientalis (JN652663; Ben Faleh et al. 2012a) for cyt b.
The data sets for the analysis were comprised of 14 sequences for cyt b and 11 sequences for 16S (see Table 1 for GenBank accession numbers). We included further two cyt b sequences for the Iranian jerboas (Dianat et al. 2012): Allactaga willimasi (GenBank accession no. JQ954952) from Ardebil (37°56′ N, 46°59′E; pt. 7 in Figure 2) and Allactaga euphratica (JQ954953) from Ilam (37°57′ N, 47°01′E; pt. 8). Of the 1104 bp sequenced for cyt b, 259 (23.5%) polymorphic sites were found, 221 of which were parsimony informative. No stop-codon insertions or deletions were observed in the alignment. For 16S, the 320-bp sequenced yielded 51 (15.9%) polymorphic sites, 44 of which were parsimony informative.
The GTR+I+G model of the DNA substitution was selected by MrModeltest for cyt b (with a proportion of invariable sites of 0.6084 and gamma distribution parameter of 4.2434) under both the hLRT and AIC criteria. For the 16S, two models were selected in MrModeltest: GTR+G model (with a zero proportion of invariable sites and gamma distribution parameter of 0.1373) under the hLRT criterion and HKY+G model (with a zero proportion of invariable sites and a gamma distribution parameter of 0.1446) under AIC. For the Bayesian analysis, the model GTR+G was used. The GTRGAMMAI model (with a proportion of invariable sites of 0.5805 and a gamma distribution parameter of 3.4227) was used for both the gene partitions in the ML analysis employed in the RAxML software.
Both, the BI and ML trees yielded highly similar results; consequently, only the BI tree is shown (Figure 2). Both the reconstructions recovered three divergent lineages, which failed to conform to the recent taxonomy of Paralactaga. The first lineage (williamsi lineage) encompassed all the samples of Allactaga williamsi from Turkey, both specimens of Allactaga euphratica from Lebanon, and a reference sample of A. williamsi from Iran. The haplotypes of A. euphratica sorted into two lineages, which showed strong geographic associations. One lineage contained the haplotypes from Harran in Turkey and the only sample from Iran (hereafter euphratica Turkey), while all the samples from two localities in Syria clustered together as an independent lineage (euphratica Syria). The supports for monophyly of the three major lineages were strong. There was also a strong support for a further subdivision of the major lineages. In the williamsi lineage, the Iranian sample was in a supported sister position against the clade containing the Turkish and Lebanese sequences. Similarly, in the euphratica Turkey lineage, the sample from Iran held the basal position. Pairwise K2P values suggested similar divergences between the three lineages, both for mean and net values (Table 2).
Our study retrieved three lineages within the Paralactaga samples considered here, and similar genetic distances between the lineages allow straightforward taxonomic interpretations. Namely, all the major K2P divergences in our results (≥11.5) are above the intraspecific divergences for rodents (≤6.5) and within the range reported for a sister species of rodents (Baker and Bradley 2006). Although numerous studies confirmed the utility of the cyt b sequences in discerning the cryptic species in rodents (Avise 2000), the mitochondrial evidence needs further corroboration from the nuclear markers and from the morphology (Baker and Bradley 2006). In the case of the williamsi and the euphratica Turkey lineages, the supplementary morphological evidence clearly points to two distinct species (Çolak et al. 1994). Specifically, the larger Allactaga williamsi has a shorter glans penis, which is sparsely covered with a low (30–40) number of horny spines. Contrary to this, Allactaga euphratica from Turkey is smaller but with a longer glans penis, which is densely covered with numerous (140–150) small spines.
The morphological details of the Syrian lineage of euphratica are not known. Although the genetic evidence on its own strongly suggests that this lineage represents another cryptic species in the Paralactaga, we refrain from altering the established taxonomic arrangement of the group. Such a step would also require proper nomenclatural solutions, which are still premature for several reasons. The first serious drawback is our lack of knowledge regarding the genetic identity and the morphological details of Allactaga euphratica from ‘Mesopotamia’ (Thomas 1881), which is the type locality for the name (restricted to Iraq; cf. Holden and Musser 2005). This ignorance prevents us from unambiguously linking the name A. euphratica with any of the two euphratica lineages, which were retrieved in our phylogenetic reconstruction.
Of the remaining species-group names so far proposed for the Paralactaga in the Middle East (Figure 1), Allactaga euphratica kivanci is topotypical with the euphratica Turkey lineage. This taxon is, therefore, well defined. The taxonomic and nomenclatural scope of Allactaga williamsi is also rather unequivocal. Two subspecies of A. williamsi are currently recognized, in addition to the nominotypical one (Çolak et al. 1997), but the differences between them are slight, and evidence on the discontinuous spatial variation is lacking (Kryštufek and Vohralík 2005). However, this paper shows that the geographic scope of A. williamsi is only imperfectly known. Two individuals from Lebanon, which were identified as A. euphratica on morphological grounds (Abi-Said 2004) clustered in our study along the Turkish A. williamsi. This surprising finding, therefore, points to an isolated population of A. williamsi in northern Lebanon, i.e., the far south of the main range of a species, which is stretching continuously from the Anatolian plateau to the highlands of the western Iran. In Lebanon, the five-toed jerboas are known only from the high-altitude pastures (2327–2660 m above sea level) in the north of the country (Abi-Said 2004). Such an environment more closely resembles the habitat requirements of A. williamsi (continental steppes at 360–2500 m a.s.l.) than the plain semi-deserts at low elevations (<600 m), which is the typical habitat of A. euphratica in Turkey (Kryštufek and Vohralík 2005). The lineage euphratica Syria is in urgent need of a detailed morphological study, and the topotypical material of A. euphratica requires both, genetic and morphological, definitions.
Shenbrot (2009) stressed the importance of taxonomic revisions ‘for the relatively poor known rodent groups, such as the five-toed jerboas’ genus Allactaga of the Middle East’, and our results fully corroborate this view. Furthermore, the cryptic species richness in Paralactaga is evidently not an exception among the jerboas (family Dipodidae). The recent molecular assessments have demonstrated the existence of a cryptic species in the desert jerboas Jaculus jaculus in Sahara (Ben Faleh et al. 2012b, Boratynski et al. 2012) and in Allactaga elater in Iran (Dianat et al. 2012). Considering that only a tiny fraction of the 51 jerboas species (Holden and Musser 2005) have so far been screened for the molecular markers, one can only speculate on the amount of the cryptic species diversity in the group. We presume that this number may be substantial. Namely, there is only a limited number of ways for a small mammal to adapt to a rigorous desert environment (Mares 1993). A strong directional selection results in convergent morphotypes, which in a traditional taxonomic analysis, conceal the interspecific differences. The resulting underestimate in the species diversity can only be overcome by the utilization of the neutral molecular markers. Clearly, much more work on the genome of the jerboas from the entire Great Palaearctic Desert Belt is needed before the species richness in this group will be properly assessed.
We would like to thank Ms. Karolyn Close for the English editing. Two anonymous referees and the associate editor provided valuable comments on an earlier draft.
Abi-Said, M.R. 2004. First record of the five-toed jerboa, Allactaga euphratica, Thomas, 1881 in Lebanon. Zool. Middle East 33: 149–152.Google Scholar
Arslan, A., T. Yorulmaz, K. Toyran, I. Albayrak and J. Zima. 2012. C-banding and Ag-NOR distribution patterns in Euphrates jerboa, Allactaga euphratica (Mammalia: Rodentia), from Turkey. Mammalia 76: 435–439.Web of ScienceGoogle Scholar
Atallah, S.I. and D.L. Harrison. 1968. On the conspecificity of Allactaga euphratica Thomas, 1881 and Allactaga williamsi Thomas, 1897 (Rodentia: Dipodidae) with a complete list of subspecies. Mammalia 32: 628–638.Google Scholar
Avise, J.C. 2000. Phylogeography: the history of formation of species. Harvard University Press, Cambridge, MA. pp. 447.Google Scholar
Ben Faleh, A., L. Granjon, C. Tatard, A. Ben Othmen, K. Said and J.-F. Cosson. 2012a. Phylogeography of the Greater Egyptian jerboa Jaculus orientalis (Rodentia: Dipodidae) in Mediterranean North Africa. J. Zool. 286: 208–220.Google Scholar
Ben Faleh, A., L. Granjon, C. Tatard, Z. Boratyński, J.F. Cosson and K. Said. 2012b. Phylogeography of two cryptic species of African desert jerboas (Dipodidae: Jaculus). Biol. J. Linn. Soc. 107: 27–38.CrossrefGoogle Scholar
Çolak, E., E. Kıvanc and N. Yiğit. 1994. A study on taxonomic status of Allactaga euphratica Thomas, 1881 and Allactaga williamsi Thomas, 1897 (Rodentia: Dipodidae) in Turkey. Mammalia 58: 591–600.Google Scholar
Çolak, E., E. Kıvanc and N. Yiğit. 1997. Taxonomic status of Allactaga williamsi Thomas, 1897 (Rodentia: Dipodidae) in Turkey. Tr. J. Zool. 21: 127–133.Google Scholar
Corbet, G.B. 1978. The mammals of the Palaearctic region: a taxonomic review. British Museum (Natural History), London. pp. 314.Google Scholar
Dianat, M., M. Aliabadian, J. Darvish and S. Akbarirad. 2012. Molecular phylogeny of the Iranian Plateau five-toed jerboa, Allactaga (Dipodidae: Rodentia), inferred from mtDNA. Mammalia 77: 95–103.Web of ScienceGoogle Scholar
Eken, G., M. Bozdogan and S. Molur. 2008. Allactaga williamsi. In: IUCN 2012. IUCN Red List of Threatened Species. Version 2012.1. www.iucnredlist.org. Downloaded on 10 September 2012.
Ellerman, J.R. 1948. A key to the rodents of Southwest Asia in the British Museum collection. Proc. Zool. Soc. Lond. 118: 765–816.Google Scholar
Ellerman, J.R. and T.C.S. Morrison-Scott. 1951. Checklist of Palaearctic and Indian mammals 1758 to 1946. British Museum (Natural History), London. pp. 810.Google Scholar
Harrison, D.L. and P.J.J. Bates. 1991. The mammals of Arabia. Harrison Zoological Museum Publ., Sevenoaks. pp. 354.Google Scholar
Harrison, R.G., S.M. Bogdanowicz, R.S. Hoffmann, E. Yensen and P.W. Sherman. 2003. Phylogeny and evolutionary history of the ground squirrels (Rodentia: Marmotinae). J. Mamm. Evol. 10: 249–276.CrossrefGoogle Scholar
Haynes, S., M. Jaarola and J.B. Searle. 2003. Phylogeography of the common vole (Microtus arvalis) with particular emphasis on the colonization of the Orkney archipelago. Mol. Ecol. 12: 951–956.PubMedCrossrefGoogle Scholar
Holden, M.E. 1993. Family Dipodidae. In: (D.E. Wilson and DA.M. Reeder, eds.) Mammal species of the world: a taxonomic and geographic reference. 2nd ed. Smithsonian Institution Press, Washington D.C. pp. 487–499.Google Scholar
Holden, M.E. and G.G. Musser. 2005. Family Dipodidae. In: (D.E. Wilson and DA.M. Reeder, eds.) Mammal species of the world: a taxonomic and geographic reference, 3rd ed., Vol. 2. John Hopkins Univ. Press, Baltimore. pp. 871–893.Google Scholar
Kryštufek, B. 2008. Allactaga euphratica. In: IUCN 2012. IUCN Red List of Threatened Species. Version 2012.1. www.iucnredlist.org. Downloaded on 10 September 2012.
Kryštufek, B. and V. Vohralík. 2005. Mammals of Turkey and Cyprus. Rodentia I: Sciuridae, Dipodidae, Gliridae, Arvicolinae. Annales Majora, Koper. pp. 292.Google Scholar
Mares, M.A. 1993. Heteromyids and their ecological counterparts: a pandesertic view of rodent ecology and evolution. In: (H.H. Genoways and J.H. Brown, eds.) Biology of Heteromyidae. Special publication No. 10. American Society of Mammalogists. pp. 652–714.Google Scholar
Nylander, J.A.A. 2004. Mrmodeltest, Version 2.2.: Uppsala University, Department of Systematic Zoology, Uppsala.Google Scholar
Ohdachi, S.D., M. Hasegawa, M.A. Iwasa, P. Vogel, T. Oshida, L.-K. Lin and H. Abe. 2006. Molecular phylogenetics of soricid shrews (Mammalia) based on mitochondrial cytochrome b gene sequences: with special reference to the Soricinae. J. Zool. (Lond.) 270: 177–191.Google Scholar
Shenbrot, G. 2009. On the conspecifity of Allactaga hotsoni Thomas, 1902 and Allactaga firouzi Womochel, 1978 (Rodentia: Dipodidae). Mammalia 75: 231–237.Google Scholar
Shenbrot, G.I., V.E. Sokolov, V.G. Heptner and Y.M. Koval’skaya. 1995. Mlekopitayushchye Rossii i sopredel’nykh regionov: Tushkanchikoobraznye. Nauka Publishers, Moscow. pp. 576.Google Scholar
Tamura, K., D. Peterson, N. Peterson, G. Stecher, M. Nei and S. Kumar. 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28: 2731–2739.Web of ScienceCrossrefGoogle Scholar
Thomas, O. 1881. Description of a new species of Allactaga from Mesopotamia. Ann. Mag. Nat. Hist. ser. 5, 8: 14–16.Google Scholar