Abstract
Background
Dorcus stag beetles in broad sense are one of the most diverse group in Lucanidae and important saproxylic insects playing a crucial role in nutrient recycling and forest biomonitoring. However, the dazzling morphological differentiations have caused numerous systematic confusion within the big genus, especially the puzzlingly generic taxonomy. So far, there is lack of molecular phylogenetic study to address the chaotic situation. In this study, we undertook mitochondrial genome sequencing of 42 representative species including 18 newly-sequenced ones from Eastern Asia and reconstructed the phylogenetic framework of stag beetles in Dorcus sensu lato for the first time.
Results
The mitogenome datasets of Dorcus species have indicated the variable mitogenomic lengths ranged from 15,785 to 19,813 bp. Each mitogenome contained 13 PCGs, 2 rRNAs, 22 tRNAs, and a control region, and all PCGs were under strong purifying selection (Ka/Ks < 1). Notably, we have identified the presence of a substantial intergenic spacer (IGS) between the trnAser (UCN) and NAD1 genes, with varying lengths ranging from 129 bp (in D. hansi) to 158 bp (in D. tityus). The mitogenomic phylogenetic analysis of 42 species showed that Eastern Asia Dorcus was monophyletic, and divided into eight clades with significant genetic distance. Four of them, Clade VIII, VII, VI and I are clustered by the representative species of Serrognathus Motschulsky, Kirchnerius Schenk, Falcicornis Séguy and Dorcus s.s. respectively, which supported their fully generic positions as the previous morphological study presented. The topology also showed the remaining clades were distinctly separated from the species of Dorcus sensu lato, which implied that each of them might demonstrate independent generic status. The Linnaeus nomenclatures were suggested as Eurydorcus Didier stat. res., Eurytrachellelus Didier stat. res., Hemisodorcus Thomson stat. res. and Velutinodorcus Maes stat. res. For Clade V, IV, III and II respectively.
Conclusion
This study recognized the monophyly of Dorcus stag beetles and provided a framework for the molecular phylogeny of this group for the first time. The newly generated mitogenomic data serves as a valuable resource for future investigations on lucanid beetles. The generic relationship would facilitate the systematics of Dorcus stag beetles and thus be useful for exploring their evolutionary, ecological, and conservation aspects.
Similar content being viewed by others
Background
Dorcus stag beetles are a diverse group in Lucanidae with over 150 species that have been described worldwide, and about 80 taxa of them are found in East Asia [1,2,3,4,5,6]. Like most stag beetles, the Dorcus members are well-known for their robust body shape, and exaggerated mandibles of males that resemble sword or knife shapes [3, 7, 8]. Some of them have been recognized as male trimorphism and thus as a good model for studying the evolution of sexually selected traits and behavior [9, 10]. Ecologically, stag beetles can serve as important bioindicators of forest health and ecosystem quality, as they are saproxylic in nature and their larvae feed on dead and decaying wood thus adding the organic matter back to the soil. Moreover, the diversity and population of stag beetles can provide information about the environmental conditions of an ecosystem [11,12,13]. Kuranouchi et al. [14] reported that during feeding, larvae of the Dorcus rectus reduced the acetylene into ethylene, thus playing a vital role in nitrogen fixation. A few species, such as Dorcus binodulosus in Japan and Dorcus antaeus in China, have been on the local conservational list due to their sensitivity to environmental changes [15].
Despite the peculiarity of these beetles, the taxonomy and phylogeny of Dorcus have long been unclear or chaotic situation. MacLeay (1819) established the genus Dorcus based on the male morphological traits and hence named Dorcus sensu stricto (Dorcus s.s. MacLeay) [16, 17]. Successively, some scholars added the species in Dorcus s.s. MacLeay. Later, Arrow (1950) packed 27 genera in Lucanidae into Dorcus MacLeay and formed the “Dorcus Arrow; also called Dorcus sensu lato (Dorcus s.l.)” with the opinion that male morphological characters are more dynamic (greatly varied) while female morphology is relatively stable and suitable for classification [18]. After that, different classification was presented in different catalogues or monographs [7, 19,20,21,22,23]. Fujita [2] largely accepted the reinstatements in the part work of Arrow, Dorcus s.l. contains Dorcus s.s. MacLeay, Serrognathus Motschulsky, Macrodorcas and Hemisodorcus. Later, Huang and Chen [3] based on male morphology and genital characteristics, indicated that Serrognathus Motschulsky, Falcicornis Séguy, and Kirchnerius Schenk were independent genera. So far, the systematics of this genus remained debatable and might be influenced by coevolution and phenotypic plasticity [24, 25]. Although, many attempts have been made to discuss the phylogenetic relationships of the genus Dorcus by using monogenic and polygenic genes as molecular data sets. Hosoya et al. (2001) investigated the genetic variation of 16 S rRNA gene in Ceruchus lignarius and Dorcus rectus rectus, intraspecific, intraspecific and interspecific relationships were discussed [26]. Hosoya and Araya (2005) supported the monophyly of Dorcus velutinus group using mitochondrial 16 S rDNA sequences as evidence [27]. However, their research only explored the complex species of Dorcus, and not involve the classification of Dorcus at the genus level. Hosoya (2003) carried out a phylogenetic analysis of Dorucs (MacLeay, 1819) and its two related genera (Prosopocoilus (Hope, 1845) and Prismognathus (Motschulsky, 1860)) based on COI gene [28]. Although their results supported the monophyly of Dorcus, however, its phylogenetic analysis found that the species in the genus Dorcus formed multiple lineages, but it still defined it as Dorcus, and their findings were questioned. Moreover, the genetic attempt has not been carried out to resolve the controversies in Dorcus taxonomy at generic level.
Recently, mitogenomics has revolutionized the field of taxonomy that uses the mitochondrial genome data sets. Because the mitochondrial genome exhibits a range of advantageous features including rapid evolutionary dynamics, maternal inheritance, limited recombination incidence, low molecular weight, and conserved gene order [34]. Such attributes facilitate broad comparisons for many animals, making it a valuable tool for phylogenetic reconstruction and as a model for genome evolution [34,35,36,37]. Moreover, the mitogenomes of insects generally contain 13 protein-coding genes (PCGs), two ribosomal RNAs (rRNAs), 22 transfer RNAs (tRNAs), and a non-codding region also known as the control region [29, 38]. The arrangement of genes within the insect mitochondrial genome is highly conserved across different species. This conserved gene order allows for the identification and comparison of homologous genes, aiding in the alignment and analysis of mitogenomic sequences. Various biologists analyzed the mitogenome data set through bayesian inferences (BI) and maximum likelihood (ML) methods and evaluated as a powerful tool to reconstruct the phylogenetic relationship among insects [39, 40]. IB analysis provides a statically robust framework to estimate evolutionary relationships. Its flexibility and ability to integrate phylogenetic uncertainty into downstream analysis make it dispensable for studying evolutionary relationships and processes in diverse taxa [41]. Both BI and ML analysis are essential for phylogenetic studies as they are complement each other. BI allows for the incorporation of prior knowledge and provides an estimation of uncertainty [41], while ML analysis offers a computationally efficient method to infer phylogenetic trees and evaluate alternative evolutionary models, collectively enhancing the accuracy and reliability of phylogenetic reconstructions [42]. Recently, similar methods have been employed for the establishment of phylogenetic relationships among various lucanid taxa, and completely resolved topology characterized by substantial nodal support have been assessed [31, 33, 43].
Therefore, the mitogenomic-based phylogenetic investigation has been conducted to reconstruct the phylogenetic relationship among different species of the genus Dorcus using BI and ML analysis. This attempt has successfully resolved the controversies in Dorcus because this genus has long been considered relatively debatable. Additionally, genomic organization, composition, and evolutionary rates in the mitogenome of 18 newly identified Dorcus species have been documented.
Results
Mitogenome Composition and Organization
In current study, mitochondrial genomes of 18 Dorcus s.l. specimens (D. curvidens, D. davidis, D. linwenhsini, D. rectus, D. tityus, D. tanakai, D. hansi, D. hopei, F. taibaishanensis, H. arrowi, H. donkeiri, H. derelictus, H. macleayii, H. rubrofemoratus, H. sinensis, S. castanicolor, D. cervulus and D. hirticornis) sequenced by Illumina HiSeq 2000 sequencer. The sequencing results represented that all the specimens have circular mitogenomes with the size range 15,785 to 19,813 bp (Fig. 1). Out of all, 12 genomes were composed of 37 genes including 13 PCGs, 2 rRNAs and 22 tRNAs, and a control region (Fig. 1, Fig. S1-Fig. S3). Perhaps due to probability in the practical error, 6 Dorcus s.l. (D. davidis, F. taibaishanensis, H. macleayii, H. sinensis, S. castanicolor, and D. hirticornis) gave partially complete mitogenome sequences i.e., sequences contained 37 genes like 13 PCGs, 2 rRNAs, and 22 tRNA and unfortunately didn’t have control region (Fig, S4, Fig. S5). Among these, 23 genes [9 PCGs (COI, COII, COIII, ATP6, ATP8, NAD2, NAD3, NAD6, and Cytb) and 14 tRNAs (trnA, trnD, trnE, trnG, trnI, trnK, trnL (uaa), trnM, trnN, trnR, trnS(uga), trnS (ucu), trnT and trnW)] were present on the J-strand (Also known as major/majority strand) and remaining 14 genes [4 PCGs (NAD1, NAD4, NAD4L, and NAD5), 2 rRNAs (rrnl and rrns) and 8 tRNAs (trnC, trnF, trnH, trnL(uag), trnP, trnQ, trnV, and trnY)] were present on the N-strand (Also known as minus/minority/light strand).
A + T content and codon usage
The nucleotide compositions of 18 Dorcus s.l. mitogenomes have displayed a higher “A + T” content with overall “A + T” contents from 66.29% (D. linwenhsini) to 72.85% (H. donckieri). Throughout the whole genome, the “AT” skews are positive, but the “GC” skews are negative (Table 1). The data indicates that “A” occurs more frequently than “T”, and “C” occurs more frequently than “G”. In protein-coding regions, there is a negative skew in “AT” values (ranging from − 0.18 to -0.14), and a negative skew in “GC” values (ranging from − 0.07 to -0.03) across all species. This means that there is a bias towards “T” and “C” in protein-coding regions. When analyzing tRNA, there is a strong preference for “G” over “C” (with GC skews ranging from 0.09 to 0.12), and a preference for “A” over “T” (with AT skews ranging from 0.02 to 0.05). Lastly, the rRNA genes demonstrate negative AT-skew values (ranging from − 0.12 to -0.07) and positive GC-skew values (ranging from 0.37 to 0.41), indicating a significant preference for “T” and “G”. This information is summarized in Table 1.
The 12 PCGs in 18 Dorcus s.l. mitogenomes are initiated with the standard start codon “ATN” (“ATA”, “ATG”, “ATC” and “ATT”), and the start codon of COI is “AAT” or “AAC” (Table S1). All PCGs in 18 newly sequenced mitogenomes of Dorcus s.l. terminated with “TAA”, “TAG”, “TA”, or “T” codons. Among the PCGs, COII, COIII, ND4 and ND5 in majority of the Dorcus s.l. terminated with incomplete stop codons (Table S2). The codons “ATT” (Iie (Isoleucine)), “TTA” (Leu (Leucine)), “TTT” (Phe (Phenylalanine)), and “ATA” (Met (Methionine)) are the four most commonly used codons in the mitotic genome of Dorcus s.l. in Table S1 and Table S2. The relative synonymous codon usage (RSCU) patterns of these 18 Dorcus are roughly the same, with RSCU values shown in Fig. 2.
Evolutionary rates of PCGs
The evolutionary rates of PCGs were examined in all newly sequenced mitogenomes of Dorcus s.l. The ratio of nonsynonymous substitutions (Ka) to synonymous substitutions (Ks) were calculated for each PCG (Fig. 3). The Ka/Ks value of the 13PCGs among all new mitogenomes of Dorcus s.l. is less than 1.0, indicating that they are all under strong purifying selection. This means, synonymous substitutions occurs at a faster rate than the nonsynonymous substitutions (under strongest selection pressure). The cytochrome oxidase subunits (COI, COII, and COIII) and cytochrome b (Cytb) exhibited lower Ka/Ks ratios compared to ATP synthase subunits (ATP8 and ATP6) and NADH dehydrogenase subunits (ND1-6 and 4 L). The order of Ka/Ks of 13 PCGs is ATP8 > NAD6 > NAD5 > NAD4L > NAD2 > NAD3 > NAD4 > NAD1 > ATP6 > Cytb > COIII > COII > COI. The fastest evolutionary rate was observed in ATP8 while the slowest rate was noted COI gene in all Dorcus mitogenomes (Fig. 3).
Intergenic spacers
Among the 18 newly sequenced mitogenomes, large intergenic spacers (IGS) are only found in Dorcus hansi and Dorcus tityus. IGSs more than 30 bp are only observed between trnAser(UCN) and NAD1. A short sequence (TACTAAATT) repeatedly occurred in large IGSs, while the locations and time of repetition have variability in its existence in D. hansi and D. tityus. Comparison of D. hansi and D. tityus (Fig. 4) reveals that the IGS of D. hansi is 129 bp with two discontinuous short sequences (TACTAAATT), while the IGS of D. tityus is 532 bp and there are seven discontinuous short sequences (TACTAAATT) in this sequence, with 47 bp–, 55 bp–, 60 bp–, 93 bp–, 61 bp–, 61 bp–, 52 bp–, 40 bp– long intergenic region among of the seven repeats from the 5´ to 3´, respectively (Fig. 4).
Composition of the large intergenic spacer between trnAser(UCN) and NAD1 among the two mitochondrial genomes in the present study. The light grey-shaded region is the short sequence repeat (TACTAAATT). The light green-shaded region indicates the length of the spacers between the short sequence repeats
Phylogenetic relationships
Including newly sequenced 18 Dorcus s.l. mitogenomes, a total of 42 Lucanidae mitogenomes as ingroups and five Scarabaeidea genomes as outgroups considered for reconstruction of phylogenetic tree based on different genomic datasets like 13 PCGs and 13 PCGs + 2 rRNAs (rrnl and rrns). Phylogenetic analysis was performed with “Maximum Likelihood” and “Bayesian Inferences”. The trees constructed with IQtree and PhyloBayes (Fig. 5, Fig. S6) have similar topologies for two data sets, thereby strongly supporting the monophyly of Dorcus and formed a sister group relationship with Prosopocoilus and Rhaetus. The representative species in this genus are clustered into the following eight clades (Fig. 5, Fig. S6). Clade VIII is the Serrognathus clade (MLB = 100%, BPP = 1.00), comprising Serrognathus platymelus, Serrognathus castanicolor and Epidorcus gracilis; Clade VII is the Kirchnerius clade (MLB = 55%, BPP = 1.00 in Fig. 5; MLB = 47%, BPP = 0.96 in Fig. S6), comprising Kirchnerius mandibularis and Kirchnerius guangxii; Clade VI is the Falcicornis clade (MLB = 64%, BPP = 1.00 in Fig. 5; MLB = 59%, BPP = 1.00 in Fig. S6), comprising Falcicornis taibaishanensis and Falcicornis seguyi; Clade V is the Eurydorcus clade (MLB = 92%, BPP = 1.00 in Fig. 5; MLB = 92%, BPP = 0.98 in Fig. S6), comprising Dorcus tanakai, Dorcus cervulus, Dorcus hirticornis and Dorcus linwenhsini; Clade IV is the Eurytrachellelus clade (MLB = 78%, BPP = 0.89 in Fig. 5; MLB = 92%, BPP = 0.98 in Fig. S6), comprising Dorcus tityus, Dorcus hansi and Dorcus davidis; Clade III is the Hemisodorcus clade (MLB = 54%, BPP = 1.00 in Fig. 5; MLB = 69%, BPP = 0.96 in Fig. S6), comprising Hemisodorcus rubrofemoratus, Hemisodorcus derelictus, Hemisodorcus arrowi, Hemisodorcus sinensis, Hemisodorcus macleayii and Hemisodorcus donckieri; Clade II is the Velutinodorcus clade (MLB = 79%, BPP = 1.00 in Fig. 5; MLB = 86%, BPP = 1.00 in Fig. S6), formed by Dorcus velutinus, Dorcus ursulus and Dorcus tenuihirsutus; Clade I is the Dorcus s. s. clade (MLB = 79%, BPP = 1.00 in Fig. 5; MLB = 86%, BPP = 1.00 in Fig. S6), formed by Dorcus hopei, Dorcus hopei, Dorcus rectus, Dorcus parallelipipedus and Dorcus curvidens. Additionally, the genetic distances (K2P-distances) have been calculated among different clades using COI genes and significant genetic distance among the different clades has been noticed ranging from 16.1 to 19.6%. The largest genetic distance (19.6%) has been recorded between Clade I and Clade III, Clade V and Clade VIII while the lowest genetic distance (16.1%) has been recorded between Clade III and Clade IV (Table 2).
Discussion
In the current study genome length of all the newly identified Dorcus species was in the range of 15,785 to 19,813 bp that are consistent with already identified Dorcus species like D. tenuihirsutus (18,505 bp), D. ursulus (18,820 bp), D. velutinus complex (14,949 bp without control region i.e., partially complete mitogenomes) [31], and D. hopei (16,026 bp) and D. seguyi (17,950 bp) [4]. Also, the reported mitogenome size of lucanid members ranges from 15,261 bp (Lucanus mazama) to 21,628 bp (Prosopocoilus astacoides) [40, 44]. Researchers have suggested that the difference in mitochondrial genome size could be attributed to the variability in tandemly repeated elements within the potential control region, intergenic space, and the repetition of large fragments within both the coding and non-coding regions of the mtDNA [31, 38, 45]. Chen et al. [31] reported that although the genomic length doesn’t have any role in systematics because the overlapping genomic length of beetles belongs to different families and genera the mitochondrial genome of beetles retains the typical gene bases contents and gene organization of the ancestor and the evolutionary rates of all protein-coding genes (PCGs) that denote their evolution is according to purifying selection [31, 46]. All our Dorcus genomes have 13 PCG, 2 rRNAs 22 tRNAs, and a control region. Among these, 23 genes (9 PCGs and 14 tRNAs) are present on the majority strand while the minority strand contains the remaining 14 genes. Similar findings have been reported for other Lucanidae species [4, 29,30,31,32]. The arrangements of all the newly sequenced Dorcus mitogenome share the ancestral type of Lucanidae without rearrangement [4, 30,31,32,33].
Although the overall genome of all our Dorcus species projected the positive AT skew and negative GC skews for all PCGs, both skews were negative in rRNAs, and in tRNAs, both were positive (Table 1). This trend is recognized as the common ancestral genome character of Lucanidae members [4, 32, 47]. Moreover, the 12 protein-coding genes (PCGs) in these species predominantly initiated with the standard start codons ATN (ATA, ATG, ATC, and ATT), while the COI gene specifically started with AAT or AAC as commonly observed in other Dorcus genome [4, 31]. Additionally, the PCGs terminated with TAA, TAG, TA, or T codons. Ojala et al. [48] reported that the presence of both complete and incomplete stop codons indicates that post-transcriptional RNA processing mechanisms, such as polyadenylation, may be involved in generating the mature mRNA transcripts. RSCU analysis of different studies revealed the four most frequent used codons [ATT (Ile), TTA (Leu), TTT (Phe), and ATA (Met)] in the Dorcus mitogenomes [4, 31, 32, 43]. These codons exhibit a high usage frequency across the 18 mitogenomes, indicating a preference for specific codons during translation. The observation of similar RSCU patterns among the Dorcus species suggests a conserved codon usage bias within this genus. Such conservation may be attributed to functional constraints, selective pressures, or shared evolutionary history [31, 32, 40, 49].
The Ka/Ks value of the 13PCGs among all new mitogenomes of Dorcus is less than 1.0, indicating that all the PCGs are under strong purifying selection. The evolutionary rates of PCGs in the mitochondrial genome of Dorcus showed that their evolution is based upon purifying selection [31, 33, 40, 50]. Additionally, IGS studies have helped to resolve phylogenetic uncertainties, clarify evolutionary lineages, and provide insights into the diversification and biogeography of Lucanidae [27, 36, 40]. The previously reported Dorcus velutinus complex also has large IGSs and a short sequence (TACTAAATT), which could provide a unique phylogenetic signal in the genus Dorcus [31, 40]. Similarly, D. tityus and D. hansi also contained large IGS and formed an independent clade with other genera in phylogenetic analysis (Fig. 5, Fig. S6), and the large IGSs once again has played a key role in the classification of Dorcus.
This research unveils the first report phylogeny of Eastern Asian Dorcus stag beetles. Utilizing two distinct datasets encompassing 13 PCGs and 13 PCGs + rRNA (rrnl and rrns), the study elucidates robust phylogenetic relationships within Eastern Asian Dorcus stag beetles. The overall phylogenetic tree signifies that the taxa included in this study effectively capture the genetic diversity of East Asian Dorcus stag beetles, with no apparent impact of long-branch attraction within the ingroup. Both of our methods (ML & BI) illustrated congruent phylogenetic trees. However, the basal relationships, particularly under ML, exhibit less conclusive resolution, potentially attributed to inadequate sampling across diverse lineages. Tree topology in the current study illustrated that Sinodendron yunnanense is the earliest branch of Lucanidae while the genus Figulus and Prismognathus have very close relationships genus Lucanus within the Lucanidae family (Fig. 5) and similar findings have been published by different Biologists [30, 32, 33]. Similarly, tree topology indicated that Dorcus s.l. and the genus Prosopocoilus have their common ancestors (Fig. 5) which is supported by the finding of Huang who constructed the best-rooted tree by TNT under equal weights with 36 morphometric characters [16]. Alike results have also been documented based on complete mitochondrial genome datasets [30]. Our research divided the Dorcus s.l into eight distinct clades. Serrognathus, E. gracilis was recovered as sister to S. castanicolor and S. platymelus, this clade was hence named as Serrognathus. Saunder (1854) first assigned E. gracilis to Cladognathus [19], but Van Roon (1910) divided it to the genus Hemisodorcus [51]. Séguy (1954) established Epidorcus and assigned it as Epidorcus [52]. Benesh (1960) then assigned E. gacilis to Prosopocoilus [5], and Huang & Chen (2013) subsequently assigned E. gracilis to Epidorcus [3]. However, E. gracilis was similar to the typical Serrognathus in morphology, most of them were medium to large in size, and the large male had multiple fine small teeth in the upper jaw, while the male phallus valgus sacs in the genitalia were thick and short without bifurcating. E. gracilis was distinguished from the typical genus Prosopocoilus by the long trifurcation of the male phallus valgus bursa (Fig. S7). Meanwhile, Our result strongly supports the E. gracilis sistering to S. castanicolor + S. platymelus with mitogenomic data, herein, in line with Wan [53, 61]. Within the genus Kirchnerius (Fig. S8), K. guangxii shares a close affinity with K. mandibularis as verified by the comparative mt genome evidence and consistent with Maes classification [54, 55]. Falcicornis comprises two species (F. taibaishanensis and F. seguyi) which are characterised by ventral plate of basal piece triangular and body blackish brown or dark brown colour (Fig. S9). Huang & Chen (2013) divided D. linwenhsini into Dorcus s. s., however, the current phylogenetic analysis reveals that D. linwenhsini is distantly related to Dorcus s. s [3].. Most of the body appears flat wide, black or maroon (Fig. S10). The male is characterized by two separate inner teeth in the maxilla, the frontal inner teeth are strongly protruding, and the end teeth of the maxilla are far away from the top of maxilla. Hence, D. linwenhsini is classified as Eurydorcus with D. hirticornis, D. tanakai and D. cervulus based on their aforementioned morphological trait similarity. Similarly, D. tanakai has resemblance with other members including D. linwenhsini and D. cervulus with internal teeth of the mandible and maxilla (Fig. S10). According to morphology (Fig. S11) and combined with phylogenetic tree analysis, D. davidis, D. hansi and D. tityus belong to Eurytrachellelus. So Eurydorcus and Eurytrachellelus formed the sister group. H. donckieri, H. macleayii, H. sinensis, H. arrowi, H. derelictus and H. rubrofemoratus clustered into a branch with a high support rate. In addition, their male genital valgus pouches are stronger than those of typical Dorcus s. s., and the sac is longer. These features are similar to the genital characteristics of Hemisodorcus (Fig. S12). Therefore, these species are assigned to the genus Hemisodorcus, which is consistent with the study of Benesh (1960) [5]. The current analysis hinted at the D. velutinus complex as being an independent genus of the genus Dorcus, as proposed by Chen et al. [31], and the sister group of Dorcus s. s. with high node support. In terms of external morphology, the prothorax, back plate and elytra of these three species are extremely rough, covered with brown bristles (Fig. S13) and differ from those of Dorcus s. s., therefore, these belonged to the genus Velutinodorcus. As the type species of Dorcus, D. parallelipipedus clustered into a branch with D. curvidens, D. rectus, D. hopei and D. hopei, and their node support is highly supported. Morphologically, the upper jaw is not particularly large, with only one large unbranched tooth in the middle (Fig. S14). So D. curvidens, D. rectus, D. hopei and D. hopei should be assigned to Dorcus s. s. This result supported the studies of Mizunuma and Nagai (1994) [23] and Huang and Chen (2013) [3]. The genetic distances of 13 PCGs and 2 rRNA genes between Lucanidae species were examined to gain further insights into the phylogenetic relationships, and the results are provided in Table S8. The K2P genetic distances of 13 PCGs and 2 rRNA genes between ingroup were all higher than 0.20 (Table S8), thereby further confirming that these genera should be considered as a distinct clade. The Ka/Ks values of the 13PCGs among the 18 Dorcus mitogenomes were all less than 1.0, indicating that they are all under strong purifying selection (Table 3; Fig. 2). This finding deepens the existing knowledge on the adaptation of Dorcus to the complicated changing environment.
Moreover, strong genetic evidence supports the notion that species belonging to different clades may also belong to different taxa. Additionally, the K2P distance between different clades provides valuable information for assessing their generic relationships. A study conducted by Wu [58] reported an average inter-genetic K2P distance of 0.220 (range: 0.174–0.259) in Lucanidae. Within closely related lucanid genera, such as Falcicornis Planet (1894) and Dorcus MacLeay (1819), as well as Rhaetus Parry (1864) and Rhaetulus Westwood (1871), the K2P distance values were found to be 0.173 and 0.174, respectively [4, 56, 58]. The same difference exists between clades A and B (0.176) of Cyclommatus, which are considered different genera [59]. Our study reveals K2P-distance values of 0.161 to 0.196 among different clades, suggesting that these distinct clades could potentially represent different genera.
Conclusively, the current study has provided sufficient information for the identification and classification of 18 newly sequenced Dorcus species by mitogenomic information, especially 13 PCGs, rRNAs, and LIGSs. Moreover, phylogenetic analysis based on these genes has classified the genus Dorcus into 8 distinct clades/ genera (Serrognathus Motschulsky, Kirchnerius Schenk, Falcicornis Séguy, Eurydorcus, Eurytrachellelus, Hemisodorcus, Velutinodorcus, and Dorcus s.s. MacLeay). Subsequently, Large IGSs are identified as another key character for the understanding of Dorcus systematics especially D. hansi and D. tityus. Due to the strong purification selection in Dorcus, this study could be helpful to enhance our understanding regarding evolution within the genus with the passage of species inclusion from other regions.
Conclusion
In conclusion, our research successfully unraveled the mitogenomic phylogeny of Eastern Asian Dorcus stag beetles (Coleoptera: Lucanidae) and provided a generic taxonomy within the big genus. Through the integration of previously published data with newly sequenced mitochondrial genomes from 42 species, we established the monophyly of Dorcus and divided the genus into eight distinct lineages with strong nodal support. Our findings confirm the existence of four recognized genera and reinstate four other genera within Dorcus, enhancing our understanding of their evolutionary relationships. Notably, the identification of a unique intergenic spacer (IGS) and specific sequence fragment in D. tityus and D. hansi offers valuable insights for future phylogenomic reconstruction. The newly generated mitogenomic data provide a valuable resource for further investigations on the ecological and evolutionary aspects of these fascinating beetles, facilitating conservation efforts and sustainable management of their forest habitats.
Materials and methods
Samples collection and isolation of genomic DNA
The adult specimens of the genus Dorus were collected from the East Mountains of China (Table 3). A total of 18 new Dorcus specimens were identified based on their morphological characters [2, 5, 18, 21] and were stored at -20 °C for genetic investigation. All taxa voucher specimens were placed in the museum of Anhui University Hefei, China. For mitogenome investigation, the total genomic DNA was isolated from the muscle tissues of the collected Dorcus specimens using DNAeasy Blood & Tissue Kit (Qiagen, Germany). The isolated DNAs were quantified via UV-visible nano-spectrophotometer (Model: Nano-100; ALLSHENG, China) and sequenced. The recently acquired sequence data has been deposited to the database of the National Center for Biotechnology Information (https://www.ncbi.nlm.nih.gov/genbank/). The corresponding accession numbers for these sequences are provided in Table 4.
PCR amplification and sequencing
The three mitochondrial genes (COI, Cytb, and 16 S) were used for the amplification of genomic DNA (Table S3). The polymerase chain reactions (PCR) were performed by following the primer’s manufacturer protocol. Briefly, each reaction mixture was prepared in a total volume of 25µL, containing template DNA: 2 µL (with at least 50 ng), 2 × EasyTaqSuperMix (+ dye): 12.5 µL, 1 µM of each primer (forward and reverse): 1 µL, and sterilized double-distilled water (ddH2O): 8.5 µL. The PCR amplification was carried out in a thermocycler (Model; company) using the PCR conditions as led temperature: 104 ºC, initial denaturation: 94 ºC/2 min. Subsequently, PCR was run about 35 cycles by following the initial denaturation phase: 94 ºC/40 s, annealing phase: 54–58 ºC/50 s, elongation phase: 70–72 ºC/70 s, and a final extension phase: 72 ºC/7 min. Table S3 provides a list of all primers utilized for DNA amplification. Finally, the amplified PCR products were sequenced using the Illumina HiSeq 2000 platform (Berry Genomics, Neijing, China) using the TruSeq nano DNA Kit [60].
Sequence assembly, annotation, and composition analysis
We employed IDBA-UD, a de novo assembler known for reconstructing longer contigs with high accuracy [61], to assemble high-quality mitogenome reads. It was configured with the K values in the range of 80 to 240 bp. Our approach involved selectively identifying mitogenome assemblies from the assembled contigs by employing BLAST using Sanger sequence data from three anchor loci (COI, Cytb & 16 S) with a minimum similarity threshold of 98%. To assess the accuracy of the mitogenome assemblies, we employed Geneious v6.1.7 (Biomatters Ltd., New Zealand) to map the clean reads back onto the assembled mitogenomes. The mapping was conducted with a tolerance of up to 2% mismatches, 3 bp gap size, and 100 bp minimum overlap. Initial annotations were performed using the invertebrate mitochondrial code on the MITOS Web Server (http://mitos.bioinf.uni-leipzig.de/index.py). Protein-coding genes were identified by aligning them with previously published genome sequences in Geneious v6.1.7. Moreover, the rRNAs (rrnl and rrns) were computed based on sequence similarity with closely related species [62]. Additionally, to gain insights into the genome composition, we determined nucleotide composition, codon usage, and relative synonymous codon usage (RSCU) using MEGA-X [63]. Subsequently, composition skew analysis was performed using different formulas such as AT skew = [A-T]/[A + T] and GC skew = [G-C]/[G + C] [32]. Finally, we calculated the evolutionary rates (Ka/Ks ratios) for each protein-coding gene using DnaSP v5.0 [64].
Computation of Dorcus phylogeny and genetic distance
Phylogenetic analyses of 18 newly sequenced Dorcus mitogenomes were carried out along with 24 Lucanidae mitogenomes available in the GenBank as ingroups (Table 3). Five mitochondrial genomes from Scarabaeidae genomes were also retrieved from the GenBank for the outgroup (Table 3). Individually, we extracted the sequences of each coding gene from the annotated genomes using Geneious Prime v2019.1.1 and aligned using the MAFFT v7.263 [65, 66]. Gaps and sites of undefined alignment were filtered from the data using Gblocks v0.91b [67]. Phylogenetic analyses were assembled based on 2 datasets of the mitochondrial genome: [1] the “PCG matrix” (including 13 PCGs); [2] the “PCGR matrix” (including 13 PCGs and 2rRNA). The selection of the optimal model for each dataset was performed using PartitionFinder 2 in Geneious Prime [68]. An input configuration file was generated, which included 37 predefined partitions based on genes. Unlinked branch lengths and a greedy search algorithm were employed to estimate the best-fitting schemes, while the Akaike Information Criterion (AIC) was used to search for the most suitable scheme (76). Two distinct algorithms, maximum likelihood (ML) and Bayesian inference (BI), were employed for conducting phylogenetic analyses.
Maximum likelihood analysis was performed by uploading a splicing file to the IQ-TREE Web Server (IQ-TREE: Efficient phylogenomic software by maximum likelihood (iqtree.org)). Set the “automatic” option under the optimal evolutionary model and build a phylogenetic tree using an ultra-fast bootstrapping approximation method with 10,000 replicates using SH-aLRT branch test, 0.5 perturbation strength and IQ-TREE stopping rule set as 100 in IQ-TREE search parameters [69]. BI analysis was performed using MrBayes 3.2.6 [70] and a selected site-heterogeneous mix model (GTR + CAT) [30]. Two independent chains started with a random tree and simulated 20,000 generations, where the tree was sampled every 10 generations. Each Markov chain Monte Carlo (MCMC) run in which the first 25% of the tree is excluded as aging. To achieve consensus, a total of 1500 trees obtained from both runs were combined, ensuring that the two runs converged with a maximum difference (maxdiff) below 0.1. to visualize and root the phylogenetic trees, Figtree v1.4.4 [31] was utilized, with the five species in Scarabaeidae serving as outgroups. Moreover, we estimated the average genetic distance among different lineages of taxa using MEGA 11 via K2P distance.
Data availability
The datasets generated and/or analyzed during the current study are included in this published article [and its supplementary information files].
References
Wan X, Yang X. Chinese names of Lucanidae and Genera from China. Chin B Entomol. 2006;43:418–22.
Fujita H. The lucanid beetles of the world. Mushi-sha’s iconographic series of Insects. 2010;6.
Huang H, Chen C. Phylogeny and systematics of Lucanidae (Sensu Holloway, 1960 & 1968) from China. Stag Beetles China II Formosa Ecol Co Taipei. 2013;xviii:41–114.
Chen Y, Liu J, Cao Y, Zhou S, Wan X. Two new complete mitochondrial genomes of Dorcus stag beetles (Coleoptera, Lucanidae). Genes Genomics. 2018;40:873–80.
Benesh B, Lucanidae. Coleopterorum Catalogus. In: W. Junk TH, editor. 2nd ed. ed1960. p. pp. 178.
Krajčík M. Lucanidae of the World, Catalogue Part I, Checklist of the stag beetles of the world (Coleoptera: Lucanidae) Plzen, Czech Republic2003.
Wan X. Study on the systematics of Lucanidae from China (Coleoptera: Scarabaeoidea): Institute of Zoology. Chinese Academy of Sciences Beijing, China; 2007.
Hu XY. Study on the Taxonomy and Phylogeny of Dorcus (s. l.) from China (Coleoptera:Scarabaeoidea:Lucanidae) [Master dissertation]. Open Access - Slavic Studies: Anhui University, Hefei, China; 2012.
Iguchi Y. Male mandible trimorphism in the stag beetle Dorcus rectus (Coleoptera: Lucanidae). Eur J Entomol. 2013;110:159–63.
Bing-Xing W, Ya-Nan W, Feng Z, Xia W. Male trimorphism in Dorcus hopei: evaluating the superiority of different morphs in mate competition. Chin J Appl Entomol. 2023;60(1).
Huang T-I. Diversity and ecology of stag beetles (Lucanidae). Saproxylic Insects: Diversity, Ecology and Conservation. 2018:149– 65.
Kuźmiński R, Chrzanowski A, Mazur A, Rutkowski P, Gwiazdowicz DJ. Distribution and habitat preferences of the stag beetle Lucanuscervus (L.) in forested areas of Poland. Sci Rep. 2020;10(1):1043.
Oxbrough A, Ziesche T. Spiders in forest ecosystems. 2013. p. 284.
Kuranouchi T, Nakamura T, Shimamura S, Kojima H, Goka K, Okabe K, et al. Nitrogen fixation in the stag beetle, Dorcus (Macrodorcus) rectus (Motschulsky)(Col., Lucanidae). J Appl Entomol. 2006;130(9–10):471–2.
Wei C, Tong Y, Zeng Y, Bai M, Wan X. Vigilance towards species endangerment and biological invasion risks: investigating beetle trade on Chinese e-commerce platforms. Biodivers Sci. 2023;31(6):22541.
Huang H. and C. C. Chen. Stag beetles of China II. Taiwan: Formosa Ecological Company,Xinbei; 2013. p. 716.
Macleay WS. Horae entomologicae: or, essays on the annulose animals. S. Bagster; 1819.
Arrow G, Coleoptera. Lamellicornia, Lucanidae and Passalidae, Vol. IV. Fauna of India including Pakistan, Ceylon, Burma and Malaya. 1950.
Saunders WW. VIII. Characters of undescribed Lucanidæ, collected in China, by R. Fortune, Esq. Trans Royal Entomol Soc Lond. 1854;8(2):45–55.
Thomson J. Catalogue des Lucanides de la collection de M. James Thomson, suivi d’un appendix renfermant la description des coupes génériques et spécifiques nouvelles: editor no identificado; 1862.
Didier R, Séguy E. Catalogue illustré des Lucanides du Globe. 1952.
Kikuta T. On the higher taxa of the stag beetle family Lucanidae. Spec Bull Jpn Soc Coleopterol Tokyo. 1986;2:131–8.
Mizunuma TaSN. The lucanid beetles of the world. Mushi-Sha’s Iconographic Ser Insects. 1994;1.
Abe H, Goji K. Chromosomes of three species of Dorcus (Coleoptera, Lucanidae), with a note on their sex chromosome evolution. Elytra. 2005;33(2):513–21.
Kang TH, Han SH, Park SJ. Development of 12 microsatellite markers in Dorcus titanus Castanicolor (Motschulsky, 1861)(Lucanidae, Coleoptera) from Korea using next-generation sequencing. Int J Mol Sci. 2016;17(10):1621.
Hosoya T, Honda M, Araya K. Genetic variation of 16S rRNA gene observed in Ceruchus lignarius and Dorcus rectus rectus (Coleoptera: Lucanidae). Entomol Sci. 2001;4(3):335–44.
Hosoya T, Araya K. Phylogeny of Japanese stag beetles (Coleoptera: Lucanidae) inferred from 16S mtrRNA gene sequences, with reference to the evolution of sexual dimorphism of mandibles. Zoolog Sci. 2005;22(12):1305–18.
Hosoya T, Araya K, Shirota Y. Molecular phylogeny of Japanese stag beetles of the genus Dorcus (Coleoptera, Lucanidae) and its allied genera inferred from mitochondrial COI gene sequences. Elytra. 2003;31(1):127–42.
Cao Y, Liu J, Zhou S, Chen Y, Wan X. Complete mitochondrial genome of Prismognathus prossi (Coleoptera: Lucanidae) with phylogenetic implications. Entomologica Fennica. 2019;30(2):90–6.
Chen D, Liu J, Bartolozzi L, Wan X. The complete mitochondrial genome of stag beetle Lucanuscervus (Coleoptera: Lucanidae) and phylogenetic analysis. PeerJ. 2019;7:e8274.
Chen Y-J, Kim S, Wan X. Mitochondrial genomes of the Dorcus velutinus complex (Coleoptera: Lucanidae) with the large intergenic spacer showing unique short sequence repeats and their implications for systematics. J Asia Pac Entomol. 2021;24(1):493–501.
Lin Z-Q, Song F, Li T, Wu Y-Y, Wan X. New mitogenomes of two Chinese stag beetles (Coleoptera, Lucanidae) and their implications for systematics. J Insect Sci. 2017;17(2):63.
Kim SI, Farrell BD. Phylogeny of world stag beetles (Coleoptera: Lucanidae) reveals a Gondwanan origin of Darwin’s stag beetle. Mol Phylogenet Evol. 2015;86:35–48.
Magoulas A, Mitochondrial DNA. Stock identification methods: Elsevier; 2005. p. 311– 30.
Boore JL, Macey JR, Medina M. Sequencing and comparing whole mitochondrial genomes of animals. Methods in enzymology. Volume 395. Elsevier; 2005. pp. 311–48.
Wang L-J, Huang J-P, Shiao S-F, Ko H-P, Sung C-H. Characterisation of the complete mitochondrial genome of Lucanus chengyuani (Coleoptera: Lucanidae). Mitochondrial DNA Part B. 2019;4(2):3460–2.
Zhang L, Sun K, Csorba G, Hughes AC, Jin L, Xiao Y, et al. Complete mitochondrial genomes reveal robust phylogenetic signals and evidence of positive selection in horseshoe bats. BMC Ecol Evol. 2021;21:1–15.
Kim K-G, Hong MY, Kim MJ, Im HH, Kim MI, Bae CH, et al. Complete mitochondrial genome sequence of the yellow-spotted long-horned beetle Psacothea hilaris (Coleoptera: Cerambycidae) and phylogenetic analysis among coleopteran insects. Mol Cells. 2009;27:429–41.
Zhou Z, Zhao L, Liu N, Guo H, Guan B, Di J, et al. Towards a higher-level Ensifera phylogeny inferred from mitogenome sequences. Mol Phylogenet Evol. 2017;108:22–33.
Xu M, Zhou S, Wan X. Phylogenetic implication of large intergenic spacers: insights from a mitogenomic comparison of Prosopocoilus Stag beetles (Coleoptera: Lucanidae). Animals. 2022;12(13):1595.
Nascimento FF, Md R, Yang Z. A biologist’s guide to bayesian phylogenetic analysis. Nat Ecol Evol. 2017;1(10):1446–54.
Lin Y, Hu F, Tang J, Moret BM. Maximum likelihood phylogenetic reconstruction from high-resolution whole-genome data and a tree of 68 eukaryotes. Pac Symp Biocomput. 2013:285–96.
Kim E, Kim P, An SL. The complete mitochondrial genome of Korean indigenous stag beetle, Dorcus koreanus. Mitochondrial DNA Part B. 2020;5(3):3741–2.
Sheffield NC, Song H, Cameron SL, Whiting MF. Nonstationary evolution and compositional heterogeneity in beetle mitochondrial phylogenomics. Syst Biol. 2009;58(4):381–94.
McKnight ML, Shaffer HB. Large, rapidly evolving intergenic spacers in the mitochondrial DNA of the salamander family Ambystomatidae (Amphibia: Caudata). Mol Biol Evol. 1997;14(11):1167–76.
Du C, Zhang L, Lu T, Ma J, Zeng C, Yue B, et al. Mitochondrial genomes of blister beetles (Coleoptera, Meloidae) and two large intergenic spacers in Hycleus genera. BMC Genomics. 2017;18(1):1–15.
Lee J, Park J, Xi H, Park J. Comprehensive analyses of the complete mitochondrial genome of Figulus binodulus (Coleoptera: Lucanidae). J Insect Sci. 2020;20(5):10.
Ojala D, Montoya J, Attardi G. tRNA punctuation model of RNA processing in human mitochondria. Nature. 1981;290(5806):470–4.
Sharp PM, Li W-H. The codon adaptation index-a measure of directional synonymous codon usage bias, and its potential applications. Nucleic Acids Res. 1987;15(3):1281–95.
Ueki G, Tojo K. The phylogeography of the stag beetle Dorcus montivagus (Coleoptera, Lucanidae): comparison with the phylogeography of its specific host tree, the Japanese beech Fagus crenata. Entomol Sci. 2023;26(1):e12535.
Van Roon G. Coleopterorum Catalogus. Lucanidae Berlin: Junk Ed; 1910.
Séguy E. Les hemisodorcites du Museum de Paris (Col. Lucanidae).: Revue Francaise d’Entomologie 1954. 184–94 p.
Wan X. Study on the systematics of Lucanidae. Coleoptera: Scarabaeoidea); 2007.
Zhai YN, Zhou SJ, Chen YJ, Wan X. The mitochondrial genome of a rare Chinese stag beetle Kirchnerius guangxii (Coleoptera: Lucanidae). Mitochondrial DNA Part B. 2020;5(2):1633–5.
Maes J-M. Lista De Los Lucanidae (Coleoptera) Del mundo. Rev Nicaraguense Ent. 1992;22:1–121.
Planet L. Description d’une nouvelle espèce de Lucanidae: Le Falcicornis groulti. Le Naturaliste. 1894;2(8):44–5.
Zhong F, Bai M, Ge Y, Wan X. Taxonomic revision of Prosopocoilus gracilis (Saunders, 1854) and its allied species from China (Coleoptera: Lucanidae). 动物分类学报. 2014;39(1):136–48.
Wu Y-Y, Cao Y-Y, Fang J, Wan X. The first complete mitochondrial genome of stag beetle from China, Prosopocoilus gracilis (Coleoptera, Lucanidae). Mitochondrial DNA Part A. 2016;27(4):2633–4.
Zhu XL, Yuan JJ, Zhou LY, Bartolozzi L, Wan X. Molecular phylogeny and historical biogeography of Cyclommatus stag beetles (Coleoptera: Lucanidae): insights into their evolution and diversification in tropical and subtropical Asia. Front Ecol Evol. 2023;11:974315.
Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–20.
Peng Y, Leung HC, Yiu S-M, Chin FY. IDBA-UD: a de novo assembler for single-cell and metagenomic sequencing data with highly uneven depth. Bioinformatics. 2012;28(11):1420–8.
Lowe TM, Chan PP. tRNAscan-SE On-line: integrating search and context for analysis of transfer RNA genes. Nucleic Acids Res. 2016;44(W1):W54–7.
Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: molecular evolutionary genetics analysis across computing platforms. Mol Biol Evol. 2018;35(6):1547.
Librado P, Rozas J. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics. 2009;25(11):1451–2.
Kuraku S, Zmasek CM, Nishimura O, Katoh K. aLeaves facilitates on-demand exploration of metazoan gene family trees on MAFFT sequence alignment server with enhanced interactivity. Nucleic Acids Res. 2013;41(W1):W22–8.
Katoh K, Rozewicki J, Yamada KD. MAFFT online service: multiple sequence alignment, interactive sequence choice and visualization. Brief Bioinform. 2019;20(4):1160–6.
Castresana J. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Biol Evol. 2000;17(4):540–52.
Lanfear R, Frandsen PB, Wright AM, Senfeld T, Calcott B. PartitionFinder 2: new methods for selecting partitioned models of evolution for molecular and morphological phylogenetic analyses. Mol Biol Evol. 2017;34(3):772–3.
Lartillot N, Lepage T, Blanquart S. PhyloBayes 3: a bayesian software package for phylogenetic reconstruction and molecular dating. Bioinformatics. 2009;25(17):2286–8.
Ronquist F, Huelsenbeck J, Teslenko M, Nylander J. Draft MrBayes version 3.2 manual: Tutorials and model summaries. 2011. Google Scholar. 2014:103.
Acknowledgements
We acknowledge this work to the efforts and hard work of our research team and “National Natural Science Foundation of China” for the provision of research funds to support this study.
Funding
This study was supported by the National Natural Science Foundation of China, 31872276 and 31572311.
Author information
Authors and Affiliations
Contributions
Contributions from all the authors were considered in the research and manuscript write-up. Y.C. & X.W. planned the research, M.J., Y.C. and L.Z. performed analysis, M.J. & L.Z. designed the figures and tables, M.J. & Y.C. prepared the original draft. L.Z. reviewed and edited the manuscript, M.J. & X.W. revised the proofread. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
About this article
Cite this article
Jafir, M., Zhou, L., Chen, Y. et al. The first mitogenomic phylogenetic framework of Dorcus sensu lato (Coleoptera: Lucanidae), with an emphasis on generic taxonomy in Eastern Asia. BMC Ecol Evo 24, 66 (2024). https://doi.org/10.1186/s12862-024-02225-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s12862-024-02225-2