A new karst-dwelling bent-toed gecko (Squamata: Gekkonidae: Cyrtodactylus) from Xiangkhoang Province, northeastern Laos

We describe a new karst-dwelling Cyrtodactylus from Ban Thathom, Xiangkhoang Province, northeastern Laos. The new species can be distinguished from other congeners by having four dark dorsal bands between limb insertions, a discontinuous nuchal loop, 10 precloacal pores in males or 10–12 precloacal pits (females) separated by a diastema from a series of enlarged femoral scales bearing 18 or 19 pores (male) or 8–10 pits (females) along each femur, 14–18 dorsal tubercle rows at midbody, no precloacal groove, 30–36 midbody scale rows across belly between ventrolateral skin folds, transversely enlarged subcaudal plates, and a maximal known snout-vent length of 75.5 mm. Our description brings to 22 the number of Cyrtodactylus species recorded from Laos.


INTRODUCTION
The diversity of the gecko genus Cyrtodactylus has increased at a phenomenal rate. Of the more than 250 species recognized within the genus, about 80% of the diversity has been described in the twenty-first century (Uetz et al., 2018). Numbers of Cyrtodactylus species in Thailand and Vietnam have shown considerable expansion, including at least 33 and 39 recognized species in each country, respectively (Pauwels et al., 2016). Moreover, several species remain to be named in Vietnam, in particular within the Cyrtodactylus irregularis species complex (Nguyen et al., 2017). Although Laos is located between these two species-rich countries and harbors a varied geological relief propitious to micro-endemism, its Cyrtodactylus fauna is currently known to include only 21 species (Luu et al., 2016a;Nazarov et al., 2014). Such low species number may be due to the fewer herpetological surveys conducted in Laos compared to Vietnam and Thailand, particularly on the karst reliefs (Teynié & David, 2014).
Our team co-described five of the Cyrtodactylus species currently recorded from Laos (Nazarov et al., 2014;Ngo & Pauwels, 2010), and we are pursuing our efforts to better inventory the diversity of the genus in this country.
During fieldwork in the forested karst massifs in the Laotian province of Xiangkhoang, we (A.S.C. and E.L.K.) encountered three geckos that differed in color pattern and scalation from known species, and which we show hereafter to belong to an undescribed species. Morphological characters, such as scalation and dorsal color and pattern, indicate that this newly discovered Cyrtodactylus population clearly belongs to the C. phongnhakebangensis species group sensu Luu et al. (2016a), which includes a dozen species from Laos and Vietnam. In agreement with the definition of this group provided by Luu et al. (2016a: 132), this new species exhibits a maximum adult SVL comprised between 73.0 and 100.6 mm, 0 or 1 supranasal, a DorTub (14-18) between 10 and 24, no webbing between fingers or toes, tubercles nearly absent on forelimbs but present on hind limbs, precloacal and femoral pores in males (in total 47) between 20 and 60, number of postcloacal tubercles (5 or 6 in male) between 3 and 8, enlarged subcaudals, and dorsum displaying well-defined dark bands.

Sampling
Field surveys were conducted in Xiangkhoang Province, Laos, from May to July 2008 within the frameworks of the Memorandum of Understanding (MOU) on the Academic and Scientific Cooperation between the National University of Laos and Kaluga State University, Russia (MOU No. 72 20-11-2012). Specimens were collected by hand from 1900 h to 2300 h. Specimens were photographed in life and subsequently euthanized in a closed vessel with a piece of cotton wool containing ethyl acetate (Simmons, 2002). Tissue samples were preserved separately in 95% ethanol and specimens were fixed in 85% ethanol, after which they were transferred to 70% ethanol for permanent storage. Specimens were subsequently deposited in the herpetological collection of the Zoological Institute of the Russian Academy of Science (ZISP) and in the herpetological collection of the Zoological Museum of Lomonosov Moscow State University (ZMMU), Moscow, Russia.
The specimens accessed for molecular analyses are listed in Table 1.
Our sampling is to date the most comprehensive sampling covering all known lineages of the C. phongnhakebangensis species complex. A map showing the distribution of the C. phongnhakebangensis species complex in Laos and adjacent territories and the location of the sampling site for the present work is provided in Figure  1. Museum

Morphological descriptions
Measurements and meristic counts follow Pauwels et al. (2016). Measurements were taken on the right side. Paired meristic characters are given in left/right order. Numbers of supralabial and infralabial scales were counted from the largest scale immediately posterior to the dorsal inflection of the posterior portion of the upper jaw to the rostral and mental scales, respectively; the number of longitudinal rows of body tubercles was counted transversely across the center of the dorsum from one ventrolateral skin fold to the other; the number of longitudinal rows of ventral scales was counted transversely across the center of the abdomen from one ventrolateral skin fold to the other; subdigital lamellae beneath the toes were counted from the base of the first phalanx to the claw; dorsal dark bands between limb insertions are those strictly contained on the dorsum between the posterior insertion of the anterior limbs and the anterior insertion of the posterior limbs. In previous literature, dorsal bands often include those contained between the anterior insertion point of the anterior limbs and the posterior point of insertion of the posterior limbs, thus interspecific comparisons on band numbers were performed with caution.
The following measurements were taken with a digital caliper to the nearest 0.1 mm: AG: axilla to groin length, taken from the posterior margin of the forelimb at its insertion point on the body to the anterior margin of the hind limb at its insertion point on the body; EarL: ear length, the greatest horizontal distance of the ear opening; ForeaL: forearm length, taken on the dorsal surface from the posterior margin of the elbow while flexed 90 • to the inflection of the flexed wrist; HeadH: head height, the maximum height of head from the occiput to the throat; HeadL: head length, from the posterior margin of the retroarticular process of the lower jaw to the tip of the snout; HeadW: head width, measured at the angle of the jaws; Internar: internarial distance, measured between the nares across the rostrum; Interorb: interorbital distance, measured between the anterior edges of the orbits; ML: mental length, the maximum length of mental shield; MW: mental width, the maximum width of mental shield; NosOrb: nostril to orbit distance, from the posterior margin of the external nares to the anterior margin of the orbit; OrbD: orbit diameter, the greatest horizontal diameter of the orbit; OrbEar: orbit to ear distance, from the anterior edge of the ear opening to the posterior edge of the orbit; RH: rostral height, the maximum height of the rostral shield; RW: rostral width, the distance between border of rostral shield and the first supralabial scales on right and left sides; SnOrb: snout to orbit distance, from the tip of the snout to the anteriormost margin of the orbit; SVL: snout-vent length, taken from the tip of snout to the vent; TailL: tail length, taken from the vent to the tip of the tail, original or regenerated; TailW: tail width, taken at the base of the tail immediately posterior to the postcloacal swelling; TibiaL: tibia length, taken on the ventral surface from the posterior surface of the knee while flexed 90 • to the base of heel. Meristic characters abbreviations: DorTub: number of longitudinal rows of dorsal tubercles at midbody; EnlFemSc: enlarged femoral scales; FemPi: femoral pits; FemPo: femoral pores; FemPreclPo: number of femoral and precloacal pores in continuous series; IL: infralabial scales; InterorbSc: interorbital scales; ParaTub: number of paravertebral tubercles between the limbs insertions, counted in a straight line immediately left of the vertebral column; PreclPi: precloacal pits (shallow depressions without waxy exudates); PreclPo: precloacal pores (deeper than pits, and with waxy exudates); SL: supralabial scales; SLF4: subdigital lamellae beneath 4 th finger (basal and distal lamellae combined); SLT4: subdigital lamellae beneath 4 th toe (basal and distal lamellae combined); Ven: number of ventral scale rows.

DNA isolation, PCR, and sequencing
For molecular phylogenetic analyses, total genomic DNA was extracted from ethanol-preserved femoral muscle tissue (Table 1) using standard phenol-chloroform-proteinase K (final concentration 1 mg/mL) extraction procedures with consequent isopropanol precipitation (protocols followed Hillis et al., 1996 andSambrook et al., 1989). The isolated total genomic DNA was visualized using agarose electrophoresis in the presence of ethidium bromide. The concentration of total DNA was measured in 1 µL using NanoDrop 2000 (Thermo Scientific, USA), and consequently adjusted to ca. 100 ng DNA/µL.
The mitochondrial cytochrome oxidase subunit I (COI) was selected as a genetic marker to clarify the taxonomic position of the newly discovered population of Cyrtodactylus.
A total of 655 bp of COI was amplified using a mitochondrial marker widely used as a barcoding marker for vertebrates, including both reptiles and amphibians (Murphy et al., 2013;Nagy et al., 2012;Smith et al., 2008), and which proved to be useful for species identification and assessment of cryptic diversity in various groups of lizards, including the genus Cyrtodactylus (Brennan et al., 2017;Hartmann et al., 2013;Luu et al., 2016a;Nazarov et al., 2012Nazarov et al., , 2014Nguyen et al., 2013Nguyen et al., , 2014Nguyen et al., , 2017Solovyeva et al., 2011). Primers used both for PCR and sequencing were the VF1-d (5 -TTCTCAACCAACCACAARGAYATYGG-3 ) and the VR1-d (5 -TAGACTTCTGGGTGGCCRAARAAYCA-3') (following Ivanova et al., 2006). The obtained fragments were sequenced in both directions for each sample, and a consensus sequence was generated. PCR analyses were performed in 25 µL reactions using ca. 50 ng genomic DNA, 10 pmol of each primer, 15 nmol of each dNTP, 50 nmol additional MgCl 2 , Taq PCR buffer (10 mmol/L Tris-HCl, pH 8.3, 50 mmol/L KCl, 1.1 mmol/L MgCl 2 , and 0.01% gelatin), and 1 U of Taq DNA polymerase. The PCR conditions for the COI gene fragment followed Nazarov et al. (2012) and included an initial denaturation step at 95 • C for 3 min; 5 cycles at 95 • C for 30 s, annealing at 45 • C for 1 min, extension at 72 • C for 2 min, followed with 35 cycles at 95 • C for 30 s, annealing at 51 • C for 1 min, extension at 72 • C for 2 min, and a final extension at 72 • C for 5 min.
The PCR products were loaded onto 1.5% agarose gels in the presence of ethidium bromide and visualized using agarose electrophoresis. If distinct bands were produced, products were purified using 2 µL, from a 1:4 dilution of ExoSapIt (Amersham, UK), per 5 µL of PCR product prior to cycle sequencing. A 10 µL sequencing reaction included 2 µL of template, 2.5 µL of sequencing buffer, 0.8 µL of 10 pmol primer, 0.4 µL of BigDye Terminator version 3.1 Sequencing Standard (Applied Biosystems), and 4.2 µL of water. The cycle sequencing reaction included 35 cycles of 10 s at 96 • C, 10 s at 50 • C, and 4 min at 60 • C. Cycle sequencing products were purified by ethanol precipitation. Sequence data collection and visualization were performed on an ABI 3730xl automated sequencer (Applied Biosystems, USA). The obtained sequences were deposited in GenBank under the accession numbers MG791873-MG791875 (Table 1).

Phylogenetic analyses
A total of 66 COI sequences of congeners were obtained from GenBank for phylogenetic analyses, including the three sequences obtained in this study (Table 1).
Nucleotide sequences were initially aligned using ClustalX 1.81 (Thompson et al., 1997) with default parameters, and then optimized manually in BioEdit 7.0.5.2 (Hall, 1999) and MEGA 6.0 (Tamura et al., 2013). Mean uncorrected genetic distances (P-distances) between sequences were determined with MEGA 6.0. MODELTEST v.3.06 (Posada & Crandall, 1998) was used to estimate the optimal models of DNA evolution. The best-fitting model selected for COI dataset was K80+I+G for the first and the second codon positions and GTR+G for the third codon position, as suggested by the Akaike Information Criterion (AIC).
Phylogenetic trees were inferred using two different methods: Bayesian inference (BI) and maximum likelihood (ML). BI was conducted in MrBayes 3.1.2 (Huelsenbeck & Ronquist, 2001;Ronquist & Huelsenbeck, 2003). Metropolis-coupled Markov chain Monte Carlo (MCMCMC) analyses were run with one cold chain and three heated chains for four million generations and sampled every 1 000 generations. Five independent MCMCMC runs were performed and 1 000 trees were discarded as burn-in. Confidence in tree topology was assessed by posterior probability (BI PP) (Huelsenbeck & Ronquist, 2001). The ML analyses were conducted using Treefinder (Jobb et al., 2004). The COI sequence of Cyrtodactylus elok Dring was used as an outgroup following Luu et al. (2016a). Confidence in tree topology was tested by non-parametric bootstrap analysis (ML BS) with 1 000 replicates (Felsenstein, 2004). We a priori regarded tree nodes with bootstrap (ML BS) values of 70% or greater and posterior probabilities (BI PP) values over 0.95 as sufficiently resolved, ML BS values between 70% and 50% (and BI PP between 0.95 and 0.90) were regarded as tendencies, and ML BS values below 50% (BI PP below 0.90) were considered unresolved (Felsenstein, 2004;Huelsenbeck & Hillis, 1993).

Genetic differentiation Sequence data
The final alignment of the examined mtDNA COI gene fragment consisted of 672 sites: 386 sites were conserved and 286 sites were variable, of which 270 were found to be potentially parsimony-informative. The transition-transversion bias (R) was estimated as 3.66. Nucleotide frequencies were A=23.14%, T=26.77%, C=30.53%, and G=19.95%.
The newly discovered Cyrtodactylus population from Ban Thathom, Xiangkhoang Province, was placed within the first subclade of the C. phongnhakebangensis species group and was clustered in one group with C. multiporus, C. teyniei, and C. cf. jarujini (1.0/99). Among these species, the Ban Thathom population appeared to be a sister lineage of C. cf. jarujini from the Bolikhamxay Province of Laos (1.0/100).

Genetic distances
The uncorrected genetic P-distances among and within the COI gene fragment of the studied members of the Cyrtodactylus phongnhakebangensis species group are summarized in Table 2. The observed interspecific distances in the COI gene within the C. phongnhakebangensis species group varied from P=4.2% (between C. darevskii and C. hinnamnoensis) to P=19.4% (between C. soudthichaki and C. hinnamnoensis) ( Table 2). These values slightly overlapped with interspecific comparisons between the C. phongnhakebangensis species group members and the outgroup Cyrtodactylus species (18.1%<P<24.4%, data not shown in Table 2). The observed intraspecific distances in our analysis varied from P=0% to P=3.8%, with the last value corresponding to the genetic differentiation between mtDNA lineages of C. pageli (Table 2).   The newly discovered Cyrtodactylus population from Ban Thathom, Xiangkhoang Province, Laos, is genetically divergent from all other members of the C. phongnhakebangensis species group, and it is most closely related to C. multiporus (P=9.5%), C. teyniei (P=9.3%), and C. cf. jarujini (P=4.5%). The last population from Bolikhamxay Province of Laos was tentatively identified as C. cf. jarujini by Luu et al. (2016a); however, its taxonomic status requires further confirmation.

Systematics
Based on the morphological, chromatical, and genetic distinctiveness of the newly discovered Cyrtodactylus population from all other populations in northern Laos and neighboring areas (see Diagnosis and Comparisons), and also on the results of phylogenetic analyses of the partial COI gene fragment, indicating that this population represents a clearly distinct mtDNA lineage, different from all congeners for which homologous sequences are available (see Results), we conclude that the Ban Thathom population represents a previously undescribed new species, which we describe and name herein.
Rostral contacting first supralabial on each side, nostrils, two supranasals and one internasal. Nostrils rounded, more or less laterally directed, each surrounded by supranasal, rostral, first supralabial and two postnasals. Three or four rows of small scales separate orbit from supralabials. Mental triangular, wider (3.4 mm) than deep (2.2 mm). A single pair of greatly enlarged postmentals in broad contact behind mental, each bordered anteromedially by mental, anterolaterally by first infralabial, posterolaterally by an enlarged lateral chinshield, and posteriorly by two granules (in total three granules contact the postmentals) ( Figure 3E). Supralabials to mid-orbital position 9/9, enlarged supralabials to angle of jaws 10/11. Infralabials 9/9 ( Figure 3D). Interorbital scale rows across narrowest point of frontal bone 30. Body moderately slender, relatively short (AG/SVL 0.43) with poorly defined, non-denticulate, ventrolateral skin folds. Dorsal scales weakly heterogeneous, domed, with tubercles about four times size of adjacent dorsal scales extending from neck onto tail base, smaller tubercles on postocular region, crown and occiput; tubercles smooth or bearing a very small keel, tubercles on posterior trunk and sacral region most prominent. Dorsal tubercles in 18 rows at midbody, typically separated from one another by two dorsal granules ( Figure 3F). Paravertebral tubercles 26. Ventral scales larger than dorsal scales, smooth, oval and subimbricate, largest in precloacal region. Midbody scale rows across belly between ventrolateral folds 30. Gular region with homogeneous, smooth, juxtaposed granular scales. A patch of enlarged precloacal scales on top of which lies a continuous series of ten pore-bearing scales, separated by a diastema of 3/4 enlarged poreless scales from a continuous series of 19/18 pore-bearing femoral scales. Enlarged femoral scales nearly two times the size of the scales of the adjacent anterior scale row ( Figure 3B). No precloacal groove. Hemipeneal bulges evident. Postcloacal spurs bearing 6/5 enlarged conical scales ( Figure 3C).
Scales on palm and sole smooth, rounded to oval or hexagonal, slightly domed. Scalation on dorsal surfaces of hind limbs similar to body dorsum with enlarged tubercles interspersed among smaller scales; tubercles smaller and rare on forelimbs. Forelimbs and hind limbs moderately long (ForeaL/SVL 0.16, TibiaL/SVL 0.19), moderately slender. Digits long, slender, inflected at interphalangeal joints, all bearing robust, slightly recurved claws. Basal subdigital lamellae broad, oval to rectangular, without scansorial surfaces; lamellae distal to digital inflection narrow; SLF4 17/17; SLT4 20/20. Subcaudals scales larger than supracaudal scales, forming a row of strongly enlarged transverse plates.

Coloration in preservative:
Dorsal ground color of head, neck, body, limbs and tail light brown. Dorsal surface of head with irregular dark brown markings. Rostral, supralabials and infralabials dark brown, posterior ones heavily maculated with beige. On each side a postocular stripe reaching the nape but not meeting the one of the opposite side as it breaks into spots (i.e., a discontinuous nuchal collar). Upper surface of limbs showing irregular dark brown bands. Dorsum showing four dark brown bands between limb insertions. Each of the four bands on dorsum posteriorly limited by a discontinuous series of whitish tubercles, similarly to the band above shoulders, the one on the neck and the one above sacrum. Original part of the tail showing three dark brown bands. Regenerated part of the tail light brown. Undersurfaces of the head, throat, venter and members uniformly beige. Coloration of the holotype in life is shown in Figure 3A. In life its dorsal ground color is darker than in preservative. The tubercles posteriorly bordering the four bands on dorsum are lighter and more contrasting than after preservation. The yellow color of the outer extremities of the supraciliaries disappears in preservative.
Besides differentiating Cyrtodactylus thathomensis sp. nov. from all congeneric species found within a 500-km radius, its combination of characters presented in the Diagnosis allows to unambiguously separate it from all species found in Bangladesh, Cambodia, Myanmar, Thailand and Vietnam (see, among other references in the literature cited, Connette et al., 2017;Grismer et al., 2012;Le et al., 2016;Mahony et al., 2009;Mahony, 2009;Panitvong et al., 2014;Pauwels et al., , 2016Sumontha et al., 2014Sumontha et al., , 2015. Distribution and natural history: Cyrtodactylus thathomensis sp. nov. is so far known only from its type-locality. The types were collected on karst boulders on a steep forested limestone hill (Figure 7).

Phylogenetic position:
The new species is a member of the C. phongnhakebangensis species group sensu Luu et al. (2016a) within which it is most closely related to C. multiporus, C. teyniei and C. cf. jarujini (see Results).

DISCUSSION
Mitochondrial genealogy indicates Cyrtodactylus thathomensis sp. nov. as a member of the C. phongnhakebangensis species group, within which it appears to be most closely related to C. multiporus and C. teyniei. Table 4 compares the main diagnostic characters of the species included in the C. phongnhakebangensis species group sensu Luu et al. (2016a) and our new species. All species in this group, including the one newly described here, live in the Cammon Plateau and the southern edge of the Xiangkhoang Plateau in the Annam Cordillera. This species group currently represents 15 out of the 22 Cyrtodactylus species presently recorded from Laos (i.e., 68%), with the remaining seven species belonging to the C. interdigitalis, C. irregularis and C. wayakonei species groups sensu Luu et al. (2016a).  As far as we know, Cyrtodactylus thathomensis sp. nov. is not found in the pet trade nor used in traditional medicine. The type-locality not being located within a protected area, the main potential threats to this new gecko species are habitat destruction through deforestation and limestone exploitation. To date, however, there appears to be no immediate concern as to the conservation status of this species, despite its limited distribution. Our new discovery stresses again the necessity to systematically survey karst massifs to inventory their unique, often micro-endemic and fragile, biodiversity.