Turkish Journal of Zoology Turk J Zool (2015) 39: 447-451 http://journals.tubitak.gov.tr/zoology/ © TÜBİTAK Research Article doi:10.3906/zoo-1403-6

Phylogenetic relationships among populations of rupestris Blanford, 1874 (Sauria: ) in southern

1 2, 1 Sugol YOUSOFI , Eskandar RASTEGAR-POUYANI *, Vida HOJATI 1 Department of Biology, Damghan Branch, Islamic Azad University, Damghan, Iran 2 Department of Biology, Faculty of Science, Hakim Sabzevari University, Sabzevar, Iran

Received: 05.03.2014 Accepted: 23.09.2014 Published Online: 04.05.2015 Printed: 29.05.2015

Abstract: We examined intraspecific relationships of the subspecies iranicus from the northern Persian Gulf area (Hormozgan, Bushehr, and Sistan and Baluchestan provinces). Phylogenetic relationships among these samples were estimated based on the mitochondrial cytochrome b gene. We used three methods of phylogenetic tree reconstruction (maximum likelihood, maximum parsimony, and Bayesian inference). The sampled populations were divided into 5 clades but exhibit little genetic divergence between populations, suggesting that these all represent a single and subspecies. The low genetic divergences suggest that the species may have expanded its range within Iran since the last drying period of the Persian Gulf.

Key words: Molecular phylogeny, cytochrome b, intraspecific relationship, genetic distance, Pristurus rupestris

1. Introduction P. r. rupestris Blanford, 1874; P. r. iranicus Schmidt, 1952; The Persian Gulf is located between the Arabian plate and and P. r. guweirensis Haas, 1943. The first subspecies is the Iranian continental block in the . distributed in the Arabian Peninsula and ; the About 18,000 years ago, at the end of latest glaciation second inhabits southern Iran, the Arabian Peninsula, period, sea levels dropped and the Persian Gulf was and West Pakistan; and third is found in Jordan (Uetz and no longer inundated by the ocean, and the sea level Hošek, 2013). This grouping seems to be superficial and has changed several times since (Lambeck, 1996). The needs further attention, particularly considering the fact regional terrestrial fauna, including , is thought to that P. flavipunctatus, which had previously been thought have dispersed across the Persian Gulf during periods of to be closely related to P. rupestris, turned out to be quite far aridity and lack of inundation. There are several examples from it (Arnold, 2009; Papenfuss et al., 2009). Considering of reptiles that have come into Iran from the Arabian the significant divergence, both morphologically and Peninsula, including the geckoes Pseudoceramodactylus genetically, among the populations of Pristurus across the khobarensis (Dakhteh et al., 2007) and Pristurus rupestris. Arabian Peninsula (Arnold, 2009; Papenfuss et al., 2009), The Pristurus Rüppell, 1835 currently we decided to investigate the intraspecific divergence includes approximately 26 species (Uetz and Hošek, among the populations of Pristurus rupestris across its entire 2013), most of which occur in the Arabian Peninsula distribution range in Iran (from Bushehr to Chabahar) using and the , with only one species occurring a molecular marker, the mitochondrial gene cytochrome b. in the coastal regions of Iran. A recent study evaluating the phylogenetic relationships among species of Pristurus 2. Materials and methods in the Arabian Peninsula and Horn of Africa found that The 60 specimens used in this study were collected from formation of the Red Sea and the isolation of Socotra southern Iran in April 2012. Figure 1 shows the distribution were important in driving vicariance within the genus area of this taxon in Iran and the localities in which the (Papenfuss et al., 2009). sampling for this study took place. All voucher specimens Blandfor’s semaphore gecko, Pristurus rupestris, is are catalogued in the Sabzevar University Herpetological a species that is widely distributed across the Arabian Collection (SUHC). Details on localities and specimens Peninsula and in the southern coastal regions of Iran are available from the herpetological collection of the (Anderson, 1999). This species comprises three subspecies: university (Appendix; on the journal’s website). * Correspondence: [email protected] 447 YOUSOFI et al. / Turk J Zool

Figure 1. Map of southern Iran and coastal regions of Persian Gulf showing localities of samples used in this study. 1- Busheher; 2- Dayyer; 3- Siraf ; 4- Nayband; 5- Charak; 6- Gheshm; 7- Bandar Abas; 8- Minab; 9- Jask; 10- Konarak; 11- Guater.

Total genomic DNA was extracted from tail tips using through jModelTest 2.1.1 (Posada, 2008), under corrected a salt-based method (Kabir et al., 2006). A fragment of Akaike information (AICc) and Bayesian information mtDNA that encodes cytochrome b was amplified using criterion (BIC) criteria. Phylogenetic analyses were two processes of amplification, standard and nested PCR. performed using different methods of reconstruction: Standard PCR was performed using primers L14724 maximum likelihood (ML), maximum parsimony (MP), (Palumbi et al., 1991) and H16064 (Burbrink et al., 2000), and Bayesian inference (BI). ML and MP were performed following the conditions and program described by Feroze using PAUP* 4.0 (Swofford, 2003) and Bayesian analysis et al. (2010). The nested primers used in this study were was carried out using Mr.Bayes 3.1.2 (Huelsenbeck and Mt-anew(5’-CTCCCAGCCCCATCCAACATCTCAG Ronquist, 2001). MP analyses were based on heuristic GATGATGAAAC -3’) and Mt-Fsh AGTTGGCCAAT searches with the tree bisection reconnection (TBR) GATGATGAATGGGTGTTCTACTGG-3’) (de Queiroz branch swapping algorithm with 100 replicates and a et al., 2002). The nested PCR conditions involved random taxon addition with all codon positions weighted initial denaturing at 95 °C for 1 min, followed by 36 equally . Confidence in the nodes was assessed by 2000 cycles (denaturing at 95 °C for 40 s, annealing at 49° bootstrap replicates (Felsenstein, 1985) with random for 40 s, and elongation at 72 °C for 1 min) and a final addition of taxa. Heuristic ML searches were done with elongation at 72 °C for 5 min. Standard PCR was 10 replicates of random sequence addition and TBR performed for most specimens but nested PCR was branch swapping using the search strategy of successive carried out on some samples because of the low quality approximations, and significance was estimated by 100 of the genomic DNA. The amplified fragments were then repartitions. For BI analysis, a portioning by codon analysis sequenced using automated sequencer ABI Prism 3700 was performed with four runs and four chains for each run (Macrogen, Korea) following the manufacturer’s protocol. for 6 million generations and the current tree was saved Sequences were aligned using Clustal W as every 100 generations. A 50% majority rule consensus implemented in the program BioEdit Sequence Alignment tree was produced from the posterior distribution of Editor ver. 7.0 (Hall, 1999) with default parameters. As the trees, and the posterior probabilities were calculated the gene is protein-coding, sequences were translated into as the percentage of samples recovering any particular amino acids to check for the presence of stop codons using clade, with posterior probabilities P ≥ 95% indicating MEGA 6.0 (Tamura et al., 2013) (none were detected). To significant support. Average genetic uncorrected distances perform maximum likelihood and Bayesian analyses, the (p-distance) among clades of P. rupestris were calculated best-fitting evolutionary model was chosen for our dataset with MEGA 6.0 (Tamura et al., 2013).

448 YOUSOFI et al. / Turk J Zool

3. Results the dataset using AICc and BIC criteria was the (GTR+I+G) In the final alignment of the sequences, 1018 bp were model as the best-fitting model. All phylogenetic methods unambiguously aligned in 62 individuals, including two used in this study (ML, MP, BI) supported the same samples of Cyrtopodion scabrum as the outgroup taxon. Of tree topology with regard to the major clades; therefore, these, 809 characters were conserved, 212 were variable, only the BI tree is shown in Figure 2. Relatively high and 203 were parsimony informative. The selected homogeneity among the populations of P. rupestris in evolutionary model by jModelTest 2.1.1 (Posada, 2008) for southern Iran was revealed in all analyses. Uncorrected

Figure 2. Phylogenetic relationships among the Pristurus rupestris populations included in the analysis using BI. Individuals of Tenuidactylus caspius were used as the outgroup taxon. Numbers close to the branches are posterior probabilities of BI followed by MP and ML bootstrap supports (2000 replicates). Numbers in front of the tree refer to the sample localities as follows: 1- Busheher; 2- Dayyer; 3- Siraf ; 4- Nayband; 5- Charak; 6- Gheshm; 7- Bandar Abas; 8- Minab; 9- Jask; 10- Konarak; 11- Guater. Four clades are shown by numbers.

449 YOUSOFI et al. / Turk J Zool p-distance among the populations for cytochrome b is or of recent range expansion in Iran (Kuhn et al., 2011). presented in the Table. It ranges from 0.04% to 3.8%. The It is also notable that P. rupestris is a house gecko in the genetic divergences among populations of P. rupestris are area; this in turn can lead to severe genetic introgression approximately 3% or lower (Table). The P. rupestris clade between geographically distant populations due to is well supported (BI/MP/ML: 1.00/90/85) relative to the anthropogenic activities and commuting. However, on outgroup taxon. The tree suffers from severe polytomy the Arabian Peninsula, the amounts of genetic distances phenomena in its deep branches, most likely because of among species within the genus Pristurus are considerable lack of sufficient information in the dataset. Although four (Popenfuss et al., 2009). As our study area covered all the major clades in the phylogenetic tree can be identified, distribution range of this taxon in Iran, we can assume these are all polytomous and genetic divergence among the that this species may have invaded Iran quite recently; clades is quite low. A clade in the top of the tree contains 23 when the Persian Gulf was dried out about 18,000 years samples from Bushehr, Dayyer, Siraf, Nayband, and Minab ago (Lambeck, 1996), it caused the Persian Gulf to serve with 0.93/77/76 support values. The second clade contains as a corridor for the exchange of many terrestrial samples from Guater and Konarak with 1.00/88/80 support between Iran and Arabia. values. Another clade contains the six specimens sampled Following the last glaciation event about 10,000 years from Jask, located in the middle of the distribution of the ago, the Persian Gulf was inundated, inhibiting dispersal species in Iran, with 0.99/90/88 support vales. The last of of reptiles from peninsular Arabia and coastal Iran. these four clades contains twelve samples from Bandarabas Another species that was isolated during this period and Charak with 0.98/80/86 support values. These results was Pseudoceramodactylus khobarensis, which occurs indicate that the populations of P. rupestris in Iran are not on Gheshm Island, the northernmost record of its range genetically significantly diverged but do contain some (Dakhteh et al., 2007). This species probably represents a localized genetic differentiation across the range of this population isolated on this island following inundation species in Iran. of the Persian Gulf. After colonizing Iran,P. rupestris dispersed along the coastal regions toward the west and 4. Discussion east (to Bushehr and Guater, respectively), and most Our samples from eleven localities across southern Iran likely its dispersal towards the inland was inhibited by the did not reveal any significant differences among these Zagros Mountains, which can act as a severe barrier. An populations. Most likely the polytomies observed in the alternative assumption for the observed situation is that phylogenetic tree are soft polytomies related to recent the species was brought to Iran by man several decades diversification or dispersion in a species that did not have ago; if this is true, then the ancestral populations of this sufficient time to accumulate variations and mutations, species arrived in Iran via several waves of anthropogenic while hard polytomy signifies rapid evolution in a species introduction at different times as the present-day complex where all populations of a taxon evolved at populations exhibit some degree of divergence. Because the same time for reasons like climate change or others the Arabian populations of P. rupestris were not available (Maddison, 1989; Hoelzer and Meinick, 1994a, 1994b). to us, our data are insufficient for testing either of these Our results indicate that the poor resolution among alternative hypotheses. Inclusion of another mitochondrial populations is most likely the result of insufficient data gene, ND2, in the study would allow us to make a direct

Table. Mean uncorrected genetic distances (p-distances) between groups of P. rupestris from the southern coastal regions of Iran based on 1023-bp fragment of cytochrome b. Numbers are as follows: 1- outgroup; 2- Bushehr_Dayyer_Siraf; 3- Nayband; 4- Bandar Abas_Gheshm_Minab_Charak; 5- Jask; 6- Guater_Konarak.

1 2 3 4 5 6 1 0.365 0.365 0.363 0.368 0.392 2 0.365 0.015 0.012 0.021 0.038 3 0.365 0.015 0.004 0.013 0.030 4 0.363 0.012 0.004 0.010 0.027 5 0.368 0.021 0.013 0.010 0.036 6 0.392 0.038 0.030 0.027 0.036

450 YOUSOFI et al. / Turk J Zool comparison of the Iranian populations with the Arabian Acknowledgments populations using the data presented by Papenfuss et al. We are grateful to Hamze Oraei and Azar Khosravani for (2009). This may provide further information, including their help in providing molecular data. We thank Dr David the amounts of genetic divergence from other subspecies Blackburn of the California Academy of Sciences, USA, for of P. rupestris. This comparison may also allow us to assistance with improving the English of this manuscript. estimate the timing of divergence across the Persian Gulf The project was partially supported by the Iranian National between the Iranian and Arabian populations. Science Foundation under project number 89001493.

References

Anderson SC (1999). The Lizards of Iran. Contributions to Kabir S, Shahriar M, Hamidul Kabir ANM, Uddin MG (2006). High Herpetology, Vol. 15. St. Louis, MO, USA: Society for the Study salt SDS-based method for the direct extraction of genomic of Amphibians and Reptiles. DNA from three different gram-negative organisms. CDR Journal 1: 57–64. Arnold EN (2009). Relationships, evolution and biogeography of Semaphore , Pristurus (: Sphaerodactylidae) Kuhn TS, Mooers AQ, Thomas GH (2011). A simple polytomy based on morphology. Zootaxa 2060: 1–21. resolver for dated phylogenies. Method Ecol Evol 2: 427–436. Burbrink FT, Lawson R, Slowinski JB (2000). Mitochondrial DNA Lambeck K (1996). Shoreline reconstruction for the Persian Gulf phylogeography of the North American rat snake (Elaphe since the last glacial maximum. Earth Planet Sc Lett 142: 43–57. obsoleta): a critique of the subspecies concept. Evolution 54: Maddison W (1989). Reconstructing character evolution on 2107–2114. polytomous cladograms. Cladistics 5: 365–377. Dakhteh SMH, Anderson SC, Kami HG (2007). Stenodactylus Palumbi SA, Martin A, Romano S, McMillan WO, Stice L, Grabowski khobarensis (Haas, 1957): an addition to the Iranian G (1991). The Simple Fool’s Guide to PCR Version 2.0. herpetofauna. Russ J Herpet 14: 229–231. Honolulu, HI, USA: University of Hawaii. de Queiroz A, Lawson R, Lemos-Espinal JA (2002). Phylogenetic Papenfuss TJ, Jackman T, Bauer A, Stuart BL, Robinson MD, Parham relationships of North American garter snakes (Thamnophis) JF (2009). Phylogenetic relationships among species in the based on four mitochondrial genes: how much DNA sequence Sphaerodactylidae lizard genus Pristurus. P Calif Acad Sci 60: is enough? Mol Phylogenet Evol 22: 315–329. 675–681. Felsenstein J (1985). Confidence limits on phylogenies: an approach Posada D (2008). jModelTest: Phylogenetic model averaging. Mol using the bootstrap. Evolution 39: 783–791. Biol Evol 25: 1253–1256. Feroze A, Malik SA, Kilpatrick WC (2010). Phylogenetic relationships Swofford DL (2003). PAUP*. Phylogenetic Analysis Using Parsimony of medically important vipers of Pakistan inferred from (* And Other Methods), Version 4.0b10. Sunderland, MA, cytochrome b sequences. J Anim Plant Sci 20: 147–157. USA: Sinauer Associates. Hall TA (1999). BioEdit: a user-friendly biological sequence Tamura K, Stecher G, Peterson D, Filipski A, Kumar S (2013). alignment editor and analysis program for Windows 95 ⁄ 98 ⁄ MEGA6: Molecular Evolutionary Genetics Analysis version NT. Nucl Acid S 41: 95–98. 6.0. Mol Biol Evol 30: 2725–2729. Hoelzer G, Meinick D (1994a). Patterns of speciation and limits to Uetz P, Hošek J (2013). The Database. Available at http:// phylogenetic resolution. Trends Ecol Evol 9: 104–107. www.reptile-database.org. Hoelzer G, Meinick D (1994b). Reply from G.A. Hoelzer and D.J. Melnick. Trends Ecol Evol 9: 298–299. Huelsenbeck JP, Ronquist F (2001). MR BAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754–755.

451 YOUSOFI et al. / Turk J Zool

Appendix. List of Pristurus rupestris samples including locality name and GenBank accession numbers for each sequence of cytochrome b. Sabzevar University Herpetological Collection: SUHC.

Species Museum number Locality Latitude Longitude GenBank accession number P. rupestris SUHC 1191 Bushehr 27°17′N 50°08′E KF982716 P. rupestris SUHC 1192 Bushehr 27°17′N 50°08′E KF982715 P. rupestris SUHC 1193 Bushehr 27°17′N 50°08′E KF982717 P. rupestris SUHC 1194 Bushehr 27°17′N 50°08′E KF982718 P. rupestris SUHC 1195 Bushehr 27°17′N 50°08′E KF982719 P. rupestris SUHC 1198 Bushehr 27°17′N 50°08′E KF982720 P. rupestris SUHC 1236 Dayyer 27°51′N 51°59′E KF982722 P. rupestris SUHC 1238 Dayyer 27°51′N 51°59′E KF982721 P. rupestris SUHC 1240 Siraf 27°40′N 52°21′E KF982723 P. rupestris SUHC 1241 Siraf 27°40′N 52°21′E KF982724 P. rupestris SUHC 1242 Nayband 27°18′N 52°42′E KF982725 P. rupestris SUHC 1243 Nayband 27°18′N 52°42′E KF982726 P. rupestris SUHC 1244 Nayband 27°18′N 52°42′E KF982727 P. rupestris SUHC 1245 Nayband 27°18′N 52°42′E KF982728 P. rupestris SUHC 1246 Nayband 27°18′N 52°42′E KF982729 P. rupestris SUHC 1247 Nayband 27°18′N 52°42′E KF982730 P. rupestris SUHC 1248 Nayband 27°18′N 52°42′E KF982731 P. rupestris SUHC 1302 Minab 27°09′N 57°04′E KF982751 P. rupestris SUHC 1303 Minab 27°09′N 57°04′E KF982752 P. rupestris SUHC 1304 Minab 27°09′N 57°04′E KF982753 P. rupestris SUHC 24 Minab 27°09′N 57°04′E KF982754 P. rupestris SUHC 25 Minab 27°09′N 57°04′E KF982755 P. rupestris SUHC 26 Minab 27°09′N 57°04′E KF982756 P. rupestris SUHC 18 Guater 25°07′N 61°32′E KF982773 P. rupestris SUHC 20 Guater 25°07′N 61°32′E KF982774 P. rupestris SUHC 1324 Konarak 25°35′N 60°39′E KF982766 P. rupestris SUHC 1325 Konarak 25°35′N 60°39′E KF982767 P. rupestris SUHC 1326 Konarak 25°35′N 60°39′E KF982768 P. rupestris SUHC 1327 Konarak 25°35′N 60°39′E KF982769 P. rupestris SUHC 1328 Konarak 25°35′N 60°39′E KF982770 P. rupestris SUHC 1329 Konarak 25°35′N 60°39′E KF982771 P. rupestris SUHC 1330 Konarak 25°35′N 60°39′E KF982772 P. rupestris SUHC 1311 Jask 25°38′N 57°52′E KF982760 P. rupestris SUHC 1312 Jask 25°38′N 57°52′E KF982761 P. rupestris SUHC 1313 Jask 25°38′N 57°52′E KF982762 P. rupestris SUHC 1314 Jask 25°38′N 57°52′E KF982763 P. rupestris SUHC 1315 Jask 25°38′N 57°52′E KF982764 P. rupestris SUHC 1316 Jask 25°38′N 57°52′E KF982765 P. rupestris SUHC 1288 Bandar Abas 27°08′N 56°15′E KF982743 P. rupestris SUHC 1289 Bandar Abas 27°08′N 56°15′E KF982744 P. rupestris SUHC 1290 Bandar Abas 27°08′N 56°15′E KF982745 P. rupestris SUHC 1291 Bandar Abas 27°08′N 56°15′E KF982746 P. rupestris SUHC 1292 Bandar Abas 27°08′N 56°15′E KF982747 P. rupestris SUHC 1293 Bandar Abas 27°08′N 56°15′E KF982748 P. rupestris SUHC 1294 Bandar Abas 27°08′N 56°15′E KF982749 P. rupestris SUHC 1295 Bandar Abas 27°08′N 56°15′E KF982750 P. rupestris SUHC 1277 Charak 26°45′N 54°19′E KF982732 P. rupestris SUHC 1278 Charak 26°45′N 54°19′E KF982733 P. rupestris SUHC 1279 Charak 26°45′N 54°19′E KF982734 P. rupestris SUHC 1280 Charak 26°45′N 54°19′E KF982735 P. rupestris SUHC 1281 Charak 26°45′N 54°19′E KF982736 P. rupestris SUHC 1282 Charak 26°45′N 54°19′E KF982737 P. rupestris SUHC 1283 Charak 26°45′N 54°19′E KF982738 P. rupestris SUHC 1284 Charak 26°45′N 54°19′E KF982739 P. rupestris SUHC 1285 Charak 26°45′N 54°19′E KF982740 P. rupestris SUHC 1286 Charak 26°45′N 54°19′E KF982741 P. rupestris SUHC 1287 Charak 26°45′N 54°19′E KF982742 P. rupestris SUHC 845 Gheshm 26°56′N 56°17′E KF982757 P. rupestris SUHC 846 Gheshm 26°56′N 56°17′E KF982758 P. rupestris SUHC 847 Gheshm 26°56′N 56°17′E KF982759 Tenuidactylus caspius SUHC 142 Tehran 35°42′N 51°25′E XXXXXXX Tenuidactylus caspius SUHC 143 Tehran 35°42′N 51°25′E XXXXXXX

1