Mitochondrial DNA Variability within Uromastyx ornata philbyi (Agamidae: Squamata) from Southwestern Saudi Arabia
Approximately 2.4 kbp of mitochondrial DNA was sequenced from 9 individuals of Uromastyx ornata philbyi originating from Taif, Namas, Al-Baha, and Jazan in southwestern Saudi Arabia. The sequenced regions cover eight tRNA genes (, , , , , , , and ) and two protein-coding genes (NADH dehydrogenase subunit 2 and cytochrome b). U. ornata philbyi had an insertion of 170 bp length between and genes. The first 128 bp of this insertion was similar to the one identified earlier in U. ornata ornata and can be folded into a stem-and-loop structure, which was less stable in U. ornata philbyi than in U. ornata ornata, or the second gene. The next 42 bp of the insertion was unique in U. ornata philbyi and additionally retained a stable stem-and-loop structure. Most base substitutions found in the sequenced genes were synonymous transitions rather than transversions. Tree analyses supported the sister group relationship between the two U. ornata subspecies and divided U. ornata philbyi into two groups: Taif+Namas group in the east of Sarawat and Al-Baha+Jazan group in the west of Sarawat. These molecular data are in agreement with current classification of U. ornata.
Two subspecies of Uromastyx ornata have been recognized: Uromastyx ornata ornata (Heyden 1827) and Uromastyx ornata philbyi Parker, 1938 . The distribution of U. ornata is confined to the eastern vicinity of the Red Sea. The range includes the Sinai Peninsula in the north and extends to the south east to Aqaba and southward through western Saudi Arabia to the northern border of Yemen. The presumed distribution of Uromastyx ornata philbyi extends along the mountainous region of the western Arabian Peninsula from Aqaba in the north down to Yemen. This area extends for about 1700 km through the arid and semiarid mountain range close to the coast of the Red Sea. The known geographical distribution includes the Median Mountains extending to the north, the Hijaz Mountains extending to the west of Makkah, and the Sarawat Mountains extending from west of Makkah down to the Yemen borders. Uromastyx o. ornata occurs on the Sinai Peninsula and northwestern Saudi Arabia as well as at the extreme southern tip of Palestine.
Wilms and Böhme  discussed the overlapping distribution of the two subspecies. The authors found that occurrence of both phenotypes broadly overlaps in the mountains of western Saudi Arabia, based on specimens of U. o. ornata from inside the range of U. o. philbyi at the vicinity of Jeddah. These localities are 800–1300 km further south from the type locality of U. o. ornata. According to Arnold , the northernmost locality of U. o. philbyi is not far from the type locality of U. o. ornata. In Jeddah and Jazan both phenotypes are obviously sympatric .
Based on the specimens available, a final decision on the taxonomic status of U. o. philbyi is not yet possible. Wilms and Böhme  suggested three hypotheses to deal with this taxonomic problem. (1) Both taxa (U. o. ornata and U. o. philbyi) are part of an intraspecific cline: U. o. philbyi had to be synonymised with U. o. ornata. (2) There are no intermediate specimens in the region in question (no gene flow between the two sympatric phenotypes): U. o. philbyi had to be lifted to specific rank. (3) There is a secondary contact zone with gene flow between the two types: U. o. philbyi must be treated as a subspecies of Uromastyx ornata. The authors concluded that none of these hypotheses can be excluded with certainty and suggested that until more specimens andor other data are available, U. o. philbyi had to be treated as a subspecies of U. ornata. The present study therefore aimed to highlight the molecular variability within U. o. philbyi and its relation to U. o. ornata to address its taxonomic status.
2. Material and Methods
Nine specimens of U. o. philbyi were collected from four localities on either side of the Sarawat Mountains in southwestern Saudi Arabia. These localities are Taif, Namas, Al-Baha, and Jazan (Figure 1). The collected animals were 3 specimens from Namas and 2 from each of the other localities. Individuals were taken to the lab, euthanized, and dissected. Samples of blood and muscle tissues were taken and immediately frozen at −80°C until use.
DNA was extracted from 0.5 mL blood samples with EZ-10 Spin Column Genomic DNA MiniPreps Kit according to the manufacturer’s instruction. Extracted DNA was spectrophotometrically quantified at 260/280 nm and was used for polymerase chain reaction (PCR).
Polymerase chain reaction was set up in a total volume of 10 μL that contained 1.0 μL of 10X Fast buffer, 0.8 μL of 2.5 μM dNTPs, 1.0 μL each of 10 μM primers, 0.05 μL of SpeedStar HS (5 unit/uL) DNA polymerase (Takara Shuzo Co., Japan), and approximately 1 ng/mL of total genomic DNA. PCR was carried out with a Takara thermal cycler using the following cycling program: 32 cycles of denaturation (98°C, 5 s), annealing (55°C, 15 s), and extension (72°C, 20 s). After treatment with ExoSAP-It (Amersham, code US78200) at 37°C, the PCR products were sequenced on an automated DNA capillary sequencer (ABI 3500 Genetic Analyzer with 8 capillaries; Applied Biosystems) with amplification primers, using the BigDye Terminator v3.1 Cycle Sequencing Kit (Life Technologies). The sequencing reactions consisted of 25 cycles of 96°C for 10 s, 50°C for 5 s, and 60°C for 4 min. The positions and sequences of the primers that have been used in both amplifications and sequencing are listed in Table 1.
Nucleotide sequences of the mitochondrial ND2 region (the complete NADH dehydrogenase subunit 2 gene plus 8 flanking tRNA genes) were aligned with those obtained from the DDBJ database (3 U. o. ornata, 3 U. ocellata, and 1 U. benti). Two additional outgroup taxa (U. a. aegyptia and U. a. microlepis) were also included in the alignment. The alignment was carried out by using the DNASIS 3.5 (Hitachi) and MacClade 4.03 (Sinauer Associates, Inc.) with manual adjustments. All unalignable and gap-containing sites were excluded from analyses so that 1504 sites were left for constructing the relationship. We primarily conducted the tree analyses by maximum-parsimony (MP), neighbor-joining (NJ), and maximum-likelihood (ML) methods. These analyses were done in PAUP* 4.0b10  by heuristic searches with the TBR branch swapping, 10 random taxon additions, and 1000 bootstrap replications for each. We also run Bayesian (MrBayes version 3.0) analysis  with conditions similar to those published by Amer and Kumazawa  and Kumazawa .
3. Results and Discussion
Approximately 1750 bp from ND2 and 8 tRNA genes and 700 bp from cytb gene were sequenced for U. o. philbyi in this study. The genes sequenced in this study were deposited in the GenBank database (accession numbers AB641364-AB641383). The 8 tRNA genes represent two clusters which flank the ND2 gene from either side. These clusters are QIM cluster (the first three genes) and WANCY cluster (the next 5 genes).
All available individuals from U. o. philbyi possessed an insertion of 170 bp between and genes, and the first 128 bp of this insertion was similar to the one that was previously found for U. o. ornata  (Figure 2). This insertion assumed a less stable stem-and-loop secondary structure compared to that of U. o. ornata (Figure 3(a)). However, the remaining 42 bp of the insertion were only found in U. o. philbyi and exhibited a very strong stem-and-loop structure (Figures 3(a) and 3(b)). The 128 bp insertion showed 78% sequence identity between U. o. ornata and U. o. philbyi (base difference = 25 and 24 sites between U. o. ornata and both U. o. philbyi-East and U. o. philbyi-West, resp.) and 90% identity between the subpopulations of U. o. philbyi on either side of the Sarawat Mountains. The remaining part of the insertion (42 bp) showed 81% similarity between the haplogroups of east and west of the Sarawat Mountains. We searched for any similarity of the 42 bp in the vicinity of the sequenced genes and we did not find it. As revealed for U. o. ornata, the inserted sequence (from base 65 to base 124) could be folded into a clover-leaf structure to produce another copy of the gene or pseudogene (Figure 3(b)). This clover-leaf structure was more stable in U. o. ornata than in U. o. philbyi as the baseparing in the anticodon stem, D-stem and TC-stem was very weak in the latter by acquiring more wobble parings.
Including substitutions and insertion-deletion events (indels), the 2.4 kbp region sequenced herein exhibited 84 differences between the individuals on either side of the Sarawat Mountains, of which 22 were characteristic to Jazan samples. The substitutions between the haplogroups of east and west Sarawat were at 25 sites, of which 6 were transversions. Indels are only present in the inserted sequence where only one base deletion was recorded at position 145. Six out of genes acquired 18 substitutions, of which only 2 transversions were found in T-loop of and () genes (Table 2). Thirty-four polymorphic sites in the 1035 base pairs representing the complete sequence of the ND2 gene were detected in U. o. philbyi, defined the two groups east and west of the Sarawat Mountains. These sites included 29 transitions and 5 transversions. These mutations accounted for 23 synonymous and 11 nonsynonymous substitutions, of which only 1 was characteristic for Jazan (Figure 4). Cytb gene exhibited 6 nonsynonymous substitutions which were characteristic to the west Sarawat haplogroups (Figure 5).
The overall base compositions of the L-strand of these sequences as accounted by the ML analysis were A: 34.8%; T: 21.4%; C: 31.6%; G: 13.2%. Figure 6 depicts an ML tree based on the nucleotide sequences of 1504 unambiguous sites aligned in 2 specimens of U. aegyptia, 3 of U. ocellata, 3 of U. o. ornata, and 8 of U. o. philbyi. According to the parsimony criterion, the analysis indicated that 1058 sites are constant, 329 are informative, and 62 are parsimony-uninformative. A single optimal ML tree was found with a negative log likelihood of 4195.16. The same tree topology was obtained by the maximum parsimony and neighbor-joining methods with strong bootstrap probabilities for each nodal relationship as shown on the tree (Figure 6). Bayesian method also supported the tree topology with very strong posterior probabilities at each node. The topological relationship at the basal node of U. o. philbyi was only statistically supported by Bayesian analysis (posterior probability = 0.81). The other three analytical methods (MP, NJ, and ML) showed clustering of the west Sarawat group (Jazan and Al-Baha) with the haplogroup of U. o. ornata (data not shown); however, this clustering was not trustable because of the weak statistical supports and most probably due to the small sampling size of both subspecies.
As noted above, both subpopulations of U. ornata philbyi share the extra 42 bp insertion that U. ornata ornata does not have. This qualitative trait provides strong support for the closer relationships of eastern and western U. ornata philbyi subpopulations to each other than to U. ornata ornata. Moreover, we estimated uncorrected pairwise distances among the different haplotypes (Table 3). Average pairwise distances between eastern and western U. ornata philbyi subpopulations were 0.035, which was smaller than that between eastern subpopulations and U. ornata ornata (0.038) and that between western subpopulations and U. ornata ornata (0.043). These distance data support the conclusion described above.
Individuals of U. o. philbyi were shown to be divided into two groups on either side of Sarawat Mountains in spite of the grouping of one Jazan individual with the east Sarawat cluster. A study on the polymorphism of isoenzymes and proteins supported this separation of different haplotypes on either side of the Sarawat Mountains . We believe that collecting more samples and data for U. o. philbyi at the subpopulation level could be of great interest to address the status of this subspecies.
Regarding the taxonomic status of the two taxa involved (ornata and philbyi) we are inclined to interpret the results presented in the present paper as evidence of a differentiation of both taxa on subspecific level. Moreover, the insertion pattern clearly suggests that U. o. ornata and U. o. philbyi are closely related to each other in relation to U. ocellata. There were no obvious differences in the base substitution pattern at the intraspecific and intrasubspecific levels, and the sequence alignment of all studied genes indicated small differences between these two taxa as shown in Figure 5 and Table 3. This is in accordance with the results of Wilms et al. , who found the genetic distance between ornata and philbyi to be 0.7% in the mitochondrial 16S ribosomal RNA gene. This could support the subspecific status of U. o. philbyi in relation to U. o. ornata.
The mountains from Taif in the middle to Abha in the south parallel to the Red Sea with 1800 km length and over 2 km elevation are known as Sarawat, Aseer, or Hejaz mountains, while the north eastern Jazan area has separate mountains with different names [14–16]. These mountains were found recently to suffer habitat fragmentation and exhibited extinctions for some critically endangered mammals . Such habitat diversity could explain the genetic variation among different haplogroups of U. o. philbyi inhabiting these mountains. Further molecular study for more samples from different localities and for samples of U. o. ornata from the localities where both subspecies are sympatric could be of great interest in resolving the taxonomic status of both subspecies. We believe that such study is necessary for this endangered and endemic animal in order to be a clue in building a strategy of its conservation.
The authors are grateful to Taif University of Saudi Arabia (Project no. 1-431-670) and Nagoya City University of Japan for their financial support.
T. Wilms et al., “On the phylogeny and taxonomy of the genus Uromastyx Merrem, 1820 (Reptilia: Squamata: Agamidae: Uromastycinae)—resurrection of the genus Saara Gray, 1845,” Bonner Zoologische Beiträge, vol. 56, pp. 55–99, 2009.View at: Google Scholar
T. Wilms and W. Böhme, “The taxonomy and distribution of Uromastyx ocellata group (Sauria: Agamidae),” Zoology in the Middle East, vol. 21, pp. 55–76, 2000.View at: Google Scholar
E. N. Arnold, “A key and annotated check list to the lizards and amphisbaenians of Arabia,” Fauna of Saudi Arabia, vol. 8, pp. 385–435, 1986.View at: Google Scholar
T. D. Kocher, W. K. Thomas, A. Meyer et al., “Dynamics of mitochondrial DNA evolution in animals: amplification and sequencing with conserved primers,” Proceedings of the National Academy of Sciences of the United States of America, vol. 86, no. 16, pp. 6196–6200, 1989.View at: Google Scholar
T. Wilms and W. Böhme, “Review of the taxonomy of the spiny-tailed lizards of Arabia (Reptilia: Agamidae: Leiolepidinae: Uromastyx),” Fauna Arabia, vol. 23, pp. 435–468, 2007.View at: Google Scholar
D. L. Swofford, PAUP∗. Phylogenetic Analysis Using Parsimony (∗and Other Methods), vol. 4, Sinauer Associates, Sunderland, Mass, USA, 2003.
J. P. Huelsenbeck and F. Ronquist, “MRBAYES: Bayesian inference of phylogenetic trees,” Bioinformatics, vol. 17, no. 8, pp. 754–755, 2001.View at: Google Scholar
S. A. M. Amer and Y. Kumazawa, “Mitochondrial DNA sequences of the Afro-Arabian spiny-tailed lizards (genus Uromastyx; family Agamidae): phylogenetic analyses and evolution of gene arrangements,” Biological Journal of the Linnean Society, vol. 85, no. 2, pp. 247–260, 2005.View at: Publisher Site | Google Scholar
S. A. M. Amer, “Biochemical genetics of Uromastyx ornata philbyi inhabiting South Western Saudi Arabia,” Egyptian Journal of Experimental Biology. Zoology, vol. 6, no. 2, pp. 331–336, 2010.View at: Google Scholar
S. A. Chaudhary, Ed., Flora of the Kingdom of Saudi Arabia, vol. 1, National Agriculture and Water Research Centre, Ministry of Agriculture, Saudi Arabia, 1997.
NCWCD and JICA, “The Management Plan for Conservation of Juniper Woodlands,” The final report of the jointed study between National Commission for Wildlife, Conservation and Development (NCWCD) and Japan International Co-operation Agency (JICA), 2006.View at: Google Scholar
L. I. El-Juhany, “Forestland degradation and potential rehabilitation in southwest Saudi Arabia,” Australian Journal of Basic and Applied Sciences, vol. 3, no. 3, pp. 2677–2696, 2009.View at: Google Scholar