Phylogeny and Specific Determination of Gloydius halys-intermedius Complex Based on Complete Mitochondrial Genes

Abstract

Background: The phylogenetic resolution within the Gloydius halys-intermedius Complex remains debatable due to the following reasons: loci selection in previous studies varied between authors; limited dataset (1−5 mitochondrial or nuclear gene fragments); lack of sampling density; and nodal supports at specific nodes remain weak, specifically within Gloydius cognatus, G. halys, and G. stejnegeri. Objectives: To revise the taxonomic and phylogenetic relationships within the G. halys-intermedius Complex, we reconstructed the molecular phylogeny and performed species delimitation based on the complete mitochondrial genomes. Methods: In this study, twelve nomenclatural groups of Gloydius species were involved in the computation of Bayesian phylogenomic inference, five of the twelve nomenclature groups were newly sequenced, while the rest were acquired from the National Center for Biotechnology Information (NCBI). The Bayesian phylogenomic inference was constructed based on 13 mitochondrial protein-coding genes. Species delimitation was performed by two distance-based methods (ABGD and ASAP) and two tree-based methods (GMYC and bPTP). Results: This research resolved the systematic relationship within the G. intermedius Complex with the support of mitogenome-based phylogenomics, while indicating cryptic diversity within the Gloydius halys-intermedius Complex: G. intermedius samples from South Korea show as paraphyletic to the cluster of the samples from northeastern China. Species delimitation results based on four models resemble each other, supporting Gloydius caucasicus, G. cognatus, G. halys, and G. stejnegeri, each representing full species. The species delimitation results of this research also resemble the nomenclatural species based on previous morphometrical results. This research indicates that species delimitation efforts based on the phylogenomic approach would likely resolve complex evolutionary relationships.

1. Introduction

The Asian pit vipers, genus Gloydius Hoge & Ramano-Hoge, 1978, a group of small-bodied anterior-fanged snakes, widely distributed from northeastern to central Asia. This genus contains more than 23 specific-leveled clades spread throughout diverse habitats from temperate forests at lower elevations to alpine meadows at higher elevations, such as the Qinghai–Tibetan Plateau [1,2,3,4]. Morphometrics and previous molecular phylogenetics conducted in this genus indicate three intragenic lineages: the G. blomhoffii Complex, G. intermedius-halys Complex, and G. strauchi Complex [5,6].

The current molecular phylogenetics inference has resolved the systematic relationship at specific levels under both the G. blomhoffii Complex and G. strauchi Complex with strong nodal support [2,3,4,7,8,9,10]. However, due to complex overlapping distributions, possible interspecific gene flows, and less genomic data involvement within the G. intermedius-halys Complex, the evolutionary relationships and the species delimitation under this species complex remain debatable [5,11,12,13]. In the molecular phylogeny of Sino-distributed Gloydius species constructed by Xu et al. [14], the two taxa labeled as “G. intermedius” did not display monophyletic topology [14], and were subsequently clarified as misidentifications of specimens of the two different subspecies under G. halys. Shi et al. [13] initially reconstructed the molecular phylogenetics of the G. intermedius-halys Complex with combined mtDNA ND4 and Cyt b [13], indicating G. halys and G. stejnegeri each represents a valid taxon at specific level. This taxonomic conception has been accepted by most studies [1,2,3,11,15,16]. However, in several recent publications, the diversity of the G. intermedius-halys Complex is still underestimated, these species are still conflated as G. halys without illustrations or reliable data support [12,17,18].

In previous studies, little genomic data were used to construct molecular phylogeny compared to the accumulating sampling and sequencing strength in recent years. Previous data usually contains 2–5 mtDNA gene sequences or a combination of both mtDNA and nucDNA [1,2,3,4,13,14,16,19]. As a result, some nodal supports remain at questionable levels.

Recent studies on molecular species delimitation in reptiles and insects [20,21,22] have provided new approaches for resolving taxonomic problems at specific level. However, these efforts have not been applicated to Asian pit vipers yet. To provide higher-resolution molecular phylogenomic inference within the G. intermedius-halys Complex, with the support of accumulated genomic data and strong sequencing strength, multiple mtDNA genomes of the Sino-distributed Gloydius species were acquired. Hence, to further investigate the inter-specific evolutionary relationship within the G. intermedius-halys Complex, an initial phylogenomic inference and molecular species delimitation of Gloydius species were performed in this study.

2. Materials and Methods

2.1. Samples and DNA Extraction

In this study, five taxa of Gloydius were sampled. The detailed specimen information is listed in

Table 1. Samples used in this study.

Vouchers	Species	Locations
AB175670	Ovophis okinavensis *	Japan, Okinawa Islands
B474	G. caucasicus	Azerbaijan, Lankaran Region
B6	Gloydius brevicaudus	Liaoning, China
IA	G. cognatus	Bayan Obo, Inner Mongolia, China
Q7	Gloydius shedaoensis	Wafangdian, Liaoning, China
S6	Gloydius stejnegeri	Lingshi, Shanxi, China
KM186844	Gloydius halys	Inner Mongolia, China
KM434236	Gloydius intermedius 2	Heilongjiang, China
KP262412	Gloydius ussuriensis	Heilongjiang, China
KT726956	Gloydius shedaoensis	Lvshun, Liaoning, China
MK559438	Gloydius himalayanus	Himachal Pradesh, India
MT731652	Gloydius changdaoensis	Changdao, Shandong, China
MW143075	Gloydius intermedius 1	Samcheok-si, Gangwon-do, South Korea

Note: * Outgroup.

. In addition, the mitochondrial genome of seven Gloydius species were downloaded from GenBank for phylogenomic analysis. The distribution map of Gloydius in this study was drawn with ArcGis (Figure 1). Ophis okinavensis, a close relative of Gloydius, was selected to be the outgroup. Liver tissues were dissected to extract the whole genome using a TGuide Smart Universal DNA Kit (TIANGEN, Beijing, China) with the TGuide S16 Nucleic Acid Extractor. And the whole genome sample was deposited in a refrigerator at −20 °C at the Institute of Zoology, Chinese Academy of Sciences (IOZ, CAS).

2.2. Genome Sequencing, Assembly, and Annotation

The genomes were sequenced by the Illumina HiSeq 6000 platform at BerryGenomics (Beijing, China) with a 400 bp insert size and a pair-end 150 bp sequencing strategy. The sequence reads were first filtered with MitoZ 3.6 [23] at default parameters. Then, the remaining clean paired reads were assembled using GetOrganelle-1.7.7.1 [24]. The annotation of genes was performed by MitoZ and manually double-checked with Geneious 8.0.5 [25]. The composition of the mitochondrial genome was calculated with MEGA7 [26].

2.3. Phylogenetic Tree Construction and Pairwise Distance Estimation

The phylogenetic tree was constructed based on 13 mitochondrial protein-coding genes (PCGs, Appendix A). All the PCG sequences were extracted by the script, extract_genes.py (https://github.com/tjcreedy/biotools, accessed on 8 July 2024). After aligning each gene sequence with MAFFT v7.526 [26], all alignments were concatenated with PhyloSuite v1.2.2 [27]. Model determination for MrBayes (Bayesian inference, BI) was generated with ModelFinder [28]. The phylogenetic tree was reconstructed using MrBayes version 3.2.7a [1]. Two Markov chains in the Bayes phylogenetic tree ran simultaneously, totaling 600,000 generations. Samples were collected every 5000 generations, and the first 25% was discarded as burn-in. The phylogenetic trees were visualized with iDOL (https://itol.embl.de/, accessed on 15 July 2024). Pairwise distances of each two species were computed using the bootstrap method of 1000 replications, calculating the p-distance with MEGA7 [26], other detailed parameters and the results can be seen in Appendix B.

2.4. Molecular Species Delimitation

This research focused on the molecular species delimitation for the genus Gloydius (Appendix C). To conduct a molecular species delimitation, two distance-based methods (ABGD [20] and ASAP [27]) and two tree-based methods (GMYC [28] and bPTP [22]) were performed. The ABGD (Automatic Barcode Gap Discovery) is a convenient method for alignment-based species delimitation, it enables rapid classification of species, performing under the JC69 Jukes-Cantor model with relative gap width (X = 0.015). ASAP (Assemble Species by Automatic Partitioning) can automatically delineate species, reducing intervention and enhancing the objectivity and accuracy of species delimitation, performed at default settings. GMYC (Generalized Mixed Yule Coalescent) takes into account evolutionary processes, such as speciation and gene flow, to provide more precise species delimitation, with an ultrametric tree generated from MrBayes using multiple sequences per species. bPTP (Bayesian Poisson Tree Processes) delineate species by identifying temporal shifts between interspecific and intraspecific branches, offering high accuracy and resolution, executing 100,000 Markov chain Monte Carlo generations with a thinning of 100 and with 20% discarded as burn-in.

3. Results

3.1. The Composition of the Mitochondrial Genome

In this study, all Gloydius species contained a typical 37 genes (22 tRNA and 2 rRNA genes, and 13 PCGs, Appendix A). The five newly sequenced mitogenomes resembled the order in previous sequenced samples [1,2,3,4,13,16]. The gene rearrangement phenomenon is not present in this genus. The nucleotide compositions of these mitogenomes are shown in

Table 2. Nucleotide compositions of the whole mitogenomes of five newly sequenced Gloydius species. Both the AT skew and GC skew values from five species were positive, and the GC skew amplitude was greater than that of AT skew.

Samples	Species	Length /bp	T%	C%	A%	G%	AT Skew	CG Skew
B474	G. caucasicus	17,224	26.0	28.8	32.2	13.1	0.11	0.37
B6	Gloydius brevicaudus	16,655	26.1	28.5	32.2	13.2	0.10	0.37
IA	G. cognatus	17,228	25.9	28.8	32.4	12.9	0.11	0.38
Q7	Gloydius shedaoensis	17,216	25.8	28.9	32.4	12.9	0.11	0.38
S6	Gloydius stejnegeri	17,225	26.0	28.7	32.4	12.9	0.11	0.38

. These Gloydius species exhibited the same AT nucleotide bias: 58%. Moreover, these mitogenomes had both a positive AT skew (0.10–0.11) and a CG skew (0.37 to–0.38).

3.2. Phylogenetic Relationships

The topology of the Bayesian inference (BI) tree displayed an identical cladogram with those in previous studies [1,2,3,4,7]. Gloydius species were clustered in a strongly supported monophyletic group with 100/100 posterior probabilities on all of the nodes (Figure 2). Figure 2 illustrates the Bayesian phylogenetic inference based on 13 mitochondrial protein-coding genes. The topology reveals a well-supported monophyletic lineage for the Gloydius halys-intermedius Complex, with strong Bayesian posterior probabilities at essential nodes. Notably, G. caucasicus and G. stejnegeri form distinct branches, supporting their divergence at specific level.

The phylogenetic position of Gloydius himalayanus from the southern slopes of the Himalayan ranges, is basal to, and considerably distant from, other species of Gloydius (p-distance: 11.7−13.2%). Nine samples that represent G. halys-intermedius are clustered in a monophyletic group sister to another monophyletic group comprising G. brevicaudus and G. ussuriensis (G. blomhoffii Complex). However, the samples of G. intermedius from South Korea did not form a monophyletic group with the samples from northeast China, as is mentioned by Lee et al. (2022) [19]. The four samples of four species, G. caucasicus, G. cognatus, G. halys, and G. stejnegeri, display paraphyly even though they were treated as subspecies of G. halys [5] or one single species in previous studies [10,12,18]. The taxonomic relationship between those clades will be discussed in the species delimitation section.

3.3. Species Delimitation

The results of specific delimitation by two distance-based methods (ABGD [20] and ASAP [27]) and two tree-based methods (GMYC [28] and bPTP [22]) are shown as vertical black bars (Figure 3). The summary of molecularly delimited species of all four approaches was identical: all four approaches revealed 11 molecular clades of Gloydius within the samples included in this study, and resemble the morphological species delimitation opinion [5,13].

The G. halys-intermedius Complex, G. cognatus, G. caraganus, G. caucasicus, G. halys, and G. stejnegeri each represent a valid specific taxon based on the species delimitation techniques. Note that the two populations of G. intermedius from northeast China and South Korea are determined as two distinct species (p-distance 2.1%), indicating cryptic diversity that requires further investigation.

4. Discussion

This study provides a higher-resolution molecular phylogenomic inference within the G. intermedius-halys Complex, based on the complete mtDNA genomes of the Sino-distributed species of genus Gloydius which face prolonged debate from different scientific publications. The specific-level taxonomic relationships within the Gloydius halys-intermedius Complex are clarified by both molecular phylogeny and specific delimitation models. The results correspond with previous morphological studies [5,13]. The complex situation of the taxonomy and phylogeny of the G. halys-intermedius Complex may be caused by the interspecific or intraspecies gene flows between different adjacent habitats. A further phylogenomic inference of Gloydius species with more samples included is required in order to investigate the origin, evolution, and migration of Asian pit vipers. The results indicate the specific-leveled genetic differentiation between the populations of G. intermedius from South Korea and northeast China. Further advanced species delimitation is encouraged to investigate this cryptic diversity.

Increasing the sample density and sequencing strength have provided a solid platform for species delimitation reviews, and phylogenomic inference construction. n globally widespread and complex species systems, the use of nuclear loci in phylogenomics is gaining increasing popularity. At the same time, sequencing known regions, combinations, and newly developed regions has become a trend in advanced species delimitation projects. The utility of mtDNA loci in molecular phylogenetics remains advantageous when it comes to accessibility, and many studies combine nucDNA and mtDNA when performing computations. Although nucDNA loci have been actively used in recent projects, mtDNA phylogenetic analysis remains essential for assessing systematics in complex biogeographical regions. While the trend encourages the combined use of mtDNA and nucDNA in constructing molecular phylogenomics, inferences based on mitochondrial genomes still provide a reference topology for subsequent topology comparisons.