INTRODUCTION

Parahelicops annamensis Bourret, 1934, has a history of entanglement with Amphiesma (e.g., Stuart, 2006; Teynie et al., 2013; David et al., 2015) and Opisthotropis (e.g., Bourret, 1934b; Smith, 1943; Stuart, 2006; Stuart and Chuaynkern, 2007; Murphy et al., 2008; Teynie et al., 2013; David et al., 2015). Citing unpublished data, Teynie et al. (2013) thought Parahelicops annamensis seemed “referable to the genus Amphiesma” but ultimately decided to retain the original name (despite using “Amphiesma annamense” in their key to species). David et al. (2015) cited unpublished molecular data that supported the monophyly of Amphiesma with respect to Parahelicops and, hence, retained the latter. At the same time, however, they listed specific morphological character states in Parahelicops also found in the A. venningi complex. Stuart (2006) recognized the validity of the genus Parahelicops, at least for P. annamensis, and suggested a close relationship of Parahelicops with Opisthotropis. The poorly known Pararhabdophis chapaensis Bourret (1934a) also exhibits morphological similarity to Parahelicops annamensis (David et al., 2015).

More inclusive phylogenetic studies have resulted in rearrangements of taxa relevant to the positions of Parahelicops and Pararhabdophis. After finding Amphiesma to be polyphyletic, Guo et al. (2014) resurrected the genus Hebius Thompson, 1913, for all species except Amphiesma stolatum. Guo et al. (2012) and Figueroa et al. (2016) found Opisthotropis to be outside the clade that includes Hebius and Amphiesma and closer to Sinonatrix and New World Natricinae. None of the previous molecular studies, however, addressed Parahelicops or Pararhabdophis. Herein, we reevaluate the phylogenetic relationships of Parahelicops and Pararhabdophis with respect to other natricines using nucleotide sequence data.

Materials and Methods

Molecular Data: Tissue samples (appendix) were obtained from the American Museum of Natural History, New York (AMNH), the Field Museum, Chicago (FMNH), and the North Carolina Museum of Natural Sciences, Raleigh (NCSM). In total, we included six new samples (fig. 1), from two genera Pararhabdophis and Parahelicops, in the matrix published by Guo et al. (2014). In addition, we included Opisthotropis cheni and O. lateralis in phylogenetic analyses to root phylogenetic trees. Extracted DNA from the fresh tissue was amplified by PCR Master Mix (Fermentas, Burlington, ON, Canada) using the same primers and conditions employed by Guo et al. 2014. PCR products were subjected to electrophoresis through a 1% agarose gel (UltraPure™, Invitrogen, La Jolla, CA). Gels were stained for 10 min in 1 X TBE buffer with 2 pg/ml ethidium-bromide, and visualized under UV light. Successful amplifications were purified to eliminate PCR components using a GeneJET™ PCR Purification kit (Fermentas). Purified PCR products were sent to FirstBase Malaysia for sequencing.

Phylogenetic Analyses: The sequences were aligned in Clustal X v2 (Thompson et al., 1997) with default settings. Data were analyzed using maximum parsimony (MP) and maximum likelihood (ML) as implemented in PAUP 4.0b 10 (Swofford, 2001), and Bayesian analysis in MrBayes 3.2 (Ronquist et al., 2012). For MP analysis, heuristic analysis was conducted with 100 random taxon-addition replicates using tree-bisection and reconnection (TBR) branchswapping algorithm, with no upper limit set for the maximum number of trees saved. Bootstrap support (BP) (Felsenstein, 1985) was calculated using 1,000 pseudorepli cates and 100 random taxon-addition replicates. All characters were equally weighted and unordered. For ML analysis, we used the optimal evolution model as selected by ModelTest v3.7 (Posada and Crandall, 1998). To estimate BP in the ML analysis, a simple taxon-addition option and 100 pseudoreplicates were employed. We arbitrarily assumed bootstrap values of >70 % to represent strong support and values of <70 % as weak support (Hillis and Bull, 1993).

FIGURE 1.

Combined Bayesian phylogram based on all concatenated data. Numbers above and below branches are MP/ML bootstrap values and combined/partitioned Bayesian posterior probabilities (>50%), respectively. Hyphen and asterisk denote <50% and 100% values, respectively. Bold text indicates samples sequenced for this study.

For Bayesian analyses, we used the optimal model determined by Modeltest with parameters estimated by Mr Bayes 3.2.1. Two simultaneous analyses with four Markov chains (one cold and three heated) were run for 10 million generations with a random starting tree and sampled every 1000 generations. Log-likelihood scores of sample points were plotted against generation time to determine stationarity of Markov chains. Trees generated before log-likelihood scores reached stationarity were discarded from the final analyses using the burn-in function. The posterior probability (PP) values for all clades in the final majority-rule consensus tree are provided. We ran analyses using both combined and partitioned datasets to examine the robustness of the tree topology (Nylander et al., 2004; Brandley et al., 2005). In the mixed-model analysis, we partitioned the data into 12 sets based on gene codon positions (first, second, and third) of cytochrome b, CMOS, NT3, and RAG1. Optimal models of molecular evolution for the partitions were calculated using Modeltest, and then assigned to these partitions in MrBayes 3.2 using the command APPLYTO. Model parameters were inferred independently for each data partition using the UNLINK command. All models employed in Bayesian analyses are shown in table 1.

Table 1.

Models used in Bayesian analyses

RESULTS

The final matrix consisted of 3162 aligned characters, of which 614 were parsimony informative. The alignment contained no gap. MP analysis of the dataset recovered nine most parsimonious trees with 3259 steps (CI = 0.38; RI = 0.59). In the ML analysis, the -Ln likelihood score of the single best tree found was 18,729.82. The cutoff point for the burn-in function was set to 20 and 21 in combined and partitioned Bayesian analyses as -InL scores reached stationarity after 20,000 and 21,000 generations, respectively. The topologies derived from our study are similar to those in Guo et al. (2014). Most relevant here, we found Parahelicops and Pararhabdophis nested within Hebius with strong support in all analyses, and within the smallest clade including H. deschauenseei, H. modestus, and some H. venningi with high statistical values from all, but the MP analysis (fig. 1).

DISCUSSION

Our analysis provides a phylogenetic explanation for the reported similarity between Parahelicops, Hebius venningi, and H. deschauenseei noted by David et al. (2015: 216). Specifically, we find Parahelicops and Pararhabdophis to be imbedded within Hebius, in a clade including H. deschauenseei, H. modestus, and some of the specimens identified as H. venningi. To align taxonomy with the recovered phylogeny of this group, we synonymize Parahelicops Bourret, 1934, and Pararhabdophis Bourret, 1934, with Hebius Thompson (1913), yielding the new combinations, Hebius chapaensis (Bourret, 1934) and Hebius annamensis (Bourret, 1934). We corroborated (not presented) Guo et al. (2012) and Figueroa et al. (2016) who found Opisthotropis to be outside Hebius and, therefore, not closely related to Parahelicops (contra Stuart, 2006).

Sequence divergence (approximately 6%) between Hebius annamensis from the Ca (Vietnam samples) and Mekong (Lao samples) drainages suggests that multiple species might exist under that binomial. Divergence between northern and southern H. annamensis might reflect isolation by low-elevation habitats characterized by a mixture of evergreen, semievergreen, and dry forest types between northern and central Annamite ranges discussed by Bain and Hurley (2011).

We also corroborate the findings of David et al. (2013) and Guo et al. (2014), who reported variation suggestive of additional unrecognized species diversity under the names H. boulengeri and H. venningi, the latter of which is polyphyletic in Guo et al. (2014) and herein. Stuart et al. (2006) suggested that there are no “geographically widespread, forest-dwelling frog species in Southeast Asia.” Such a pattern may also exist in snakes such as Hebius, some of which are known to be connected to anurans through trophic relations (e.g., Moriguchi and Naito, 1982; David et al., 2007).