DEAR EDITOR,

Muntjac deer (Cervidae:Muntiacus) are often cited as an excellent model for the study of vertebrate evolution due to their fast rate of change in chromosome number among vertebrates.However, the phylogenetic relationships withinMuntiacusgenerally, and the taxonomic status ofMuntiacus gongshanensisspecifically, remain unclear.Here, the phylogenetic relationships withinMuntiacuswere studied using mitochondrial genome (mitogenome) and cytochromeb(cytb) segments.Our results recognize 12 species withinMuntiacusand support the controversial speciesM.gongshanensis,M.putaoensis, andM.malabaricus.Furthermore, Bayesian inference (BI) and maximum-likelihood(ML) approaches revealedM.gongshanensisandM.crinifronsto be closely related species, withM.feaeas their sister species, andM.putaoensisandM.truongsonensisto be closely related, withM.rooseveltorumas their sister species.The distribution range ofM.gongshanensiswas also confirmed in southwest China (Namdapha, Modong, Zayu and Gongshan) and northern Myanmar (Putao).The results of this study provide insight into the evolution ofMuntiacusand further provide a molecular basis for the taxonomic evaluation of the genus in the future and fundamental data for the conservation ofM.gongshanensis.

Species of the genusMuntiacusRafinesque, 1815(Artiodactyla: Cervidae: Muntiacinae) are distributed throughout Southeast Asia and southern China(Supplementary Figure S1).Muntiacusspecies are of great interest in evolutionary studies because of their chromosomal variations and recent classification of several new species(Wang & Lan, 2000; Yang et al., 1997).Indeed, the number of known muntjac species has increased in the last 30 years,includingM.gongshanensis, Ma, Wang & Shi, 1990;M.vuquangensisTuoc, Dung, Dawson, Arctander & Mackinnon,1994;M.truongsonensisGiao, Tuoc, Dung, Wikramanayake,Amato, Arctander & Mackinnon, 1998; andM.putaoensisAmato, Egan & Rabinowitz, 1999.Although much attention has been paid toMuntiacus, the taxonomy of this genus remains controversial (Supplementary Table S1).There is dispute regarding the number of classified species withinMuntiacus.Indeed, only five species were reported in 1986(Ma et al., 1986), but 16 species were listed in 2011 (Groves &Grubb, 2011).Thus, much debate remains regarding the taxonomic status of reported species (Supplementary Table S1).Presently, 13 species ofMuntiacusare recorded in the IUCN list, 54% of which are considered Data Deficient.

Species represent the basic unit of biodiversity.Ambiguous definitions and diagnoses can impact biodiversity conservation, and potentially result in common species being classified as endangered while truly endangered species are neglected (Hong, 2016).While several extant species of muntjac differ very generally in body size, color, and antlers,many share similar basic morphology that can make them difficult to distinguish.Furthermore, as the distribution ranges of muntjac species continue to be updated, the taxonomic status of certain groups as independent species is questionable (Groves & Grubb, 1982; James et al., 2008; Le et al., 2014; Ma et al., 1990; Smith et al., 2010).

The significance of historical distributions of muntjac species must be interpreted with caution.For example,identification of archaeological materials has indicated that the Holocene distributions of some muntjac species were much more extensive than historical records indicate (Turvey et al.,2016).Muntiacus gongshanensis, described in GongshanCounty, Yunnan Province, southwestern China (Ma et al.,1990), is one of the least-known ungulate species in China(Liu & Wu, 2019), including its taxonomic status and distribution range.Previously, it was considered a northern subspecies ofM.feaeor western subspecies ofM.crinifrons(Smith et al., 2010).Furthermore, recent reports ofM.crinifronsin Tibet, China, far from its generally accepted range(eastern China), may result from its misidentification withM.gongshanensis described at Mount Gaoligong (Chen et al.,2008; Timmins & Duckworth, 2016).Muntiacus gongshanensis is described as Data Deficient by the IUCN but is listed as Critically Endangered in the latest red list of China’s vertebrates (Jiang et al., 2016).Muntiacus crinifronsis listed as Vulnerable in the latest red list of the IUCN but is categorized as Endangered in the latest red list of China’s vertebrates.The current population trends for bothM.gongshanensis andM.crinifronsshow a decline (Timmins &Duckworth, 2016).Thus, there is an urgent need to reconstruct the taxonomic classification and phylogeny of the genusMuntiacusto establish appropriate levels of protection for muntjac species.

The phylogeny of the genusMuntiacushas long been debated (James et al., 2008; Timmins & Duckworth, 2016).Comparative phylogenomic approaches can solve the problem of deep branches resulting from rapid radiations (Chen et al.,2019; Jiang et al., 2019).Therefore, mitogenomes have been used to explore the interspecific relationships withinMuntiacus.Previous phylogenies ofMuntiacushave largely been based on morphological characters (Ma et al., 1986) and partial fragments of mitochondrial DNA (Cao et al., 2002;Wang & Lan, 2000; Yi et al., 2002).In recent years,mitogenomes have been used for phylogenetic reconstruction ofMuntiacus(Li et al., 2017; Martins et al., 2017; Singh et al.,2019; Srisodsuk et al., 2018; Zhang et al., 2019a), but this approach has been limited to one or only a few species.Complete mitogenomes contain increased phylogenetic signals compared to partial fragments and therefore provide more comprehensive insight into the phylogeny of taxa.In the classification of species, phylogenetic relationships based on complete mitochondrial sequences can be used as a reference (Zhang et al., 2019b).In this study, we sequenced and annotated the mitogenome ofM.gongshanensis, then reconstructed the phylogenetic relationships within the genus by combining our data with 30 complete mitogenomes and an additional 18 cytbsequences representing species ofMuntiacusfrom GenBank.We also confirmed the distribution range ofM.gongshanensis.

Both ML and BI analyses recovered similar tree topologies,and all nodes had significant support, except forM.putaoensiswith other muntjacs (bootstrap support for ML/posterior probability in BI=73/0.54).Twelve monophyletic clusters were contained within two major clades on the tree (Figure 1).The first major clade includedM.reevesi,M.vuquangensis,M.rooseveltorum,M.truongsonensis, andM.putaoensis.The second major clade includedM.atherodes,M.malabaricus,M.vaginalis,M.muntjak,M.feae,M.crinifrons, andM.gongshanensis.In the first major clade,M.reevesidifferentiated first, followed byM.vuquangensis.In the posterior subclade,M.rooseveltorumdifferentiated first, andM.putaoensiswas most closely related toM.truongsonensis.In the second major clade,M.atherodesdifferentiated first,and the remaining species split into two subclades.The ‘M.muntjakclade’ included the Sri Lankan red muntjac (M.malabaricus), southern red muntjac (M.muntjak), and northern red muntjac (M.vaginalis).Muntiacus malabaricusdifferentiated first, withM.muntjakandM.vaginalisdifferentiating subsequently.Muntiacus gongshanensiswas in the same clade asM.crinifronsandM.feae.Muntiacus feaedifferentiated first, followed byM.crinifronsandM.gongshanensis.

We used network analysis of theM.gongshanensis,M.crinifrons, andM.feaehaplotypes to provide further resolution of the closely relatedM.gongshanensishaplotypes and reveal the interspecific relationships of the three species (Figure 1C).Results were similar to those obtained from the phylogenetic tree; the haplotype network supported the recognition of three clusters:M.crinifronsandM.feaewere segregated into two clusters and distinct haplotypes 1 and 2, and also separated fromM.gongshanensisin 25 and 12 mutational steps,respectively; eleven haplotypes (haplotypes 3 to 13) ofM.gongshanensis(n=13) were shared among three different regions.Network analysis of theM.gongshanensishaplotypes showed a star-like appearance.Central haplotype 11 was shared byM.gongshanensisindividuals from the Namdapha region, China, and Putao, Myanmar.Haplotype 7 was shared byM.gongshanensisfrom Yunnan (China) and Putao(Myanmar).The other haplotypes were shared by the original collection region of each individual (Figure 1C).

The complete mitogenomes ofMuntiacusspecies provide a deeper understanding of the phylogenetic relationships within the genus.To date, our results represent the most comprehensive analysis of molecular data forMuntiacusand will help to elucidate the evolutionary relationships within the genus.There is no question of the validity ofM.reevesias an independent species (Figure 1).Our results supportM.putaoensis,M.vuquangensis,M.truongsonensis, andM.rooseveltorumbeing of the same lineage.Muntiacus putaoensisis the most recently discovered species of muntjac,as confirmed based on partial fragments of mtDNA (Amato et al., 1999).Subsequent studies suggest thatM.putaoensis,M.truongsonensis, andM.rooseveltorumlikely belong to theM.rooseveltorumspecies complex (James et al., 2008; Li et al.,2017).Our results supportM.putaoensis,M.truongsonensis,andM.rooseveltorumas independent species, withM.putaoensismost closely related toM.truongsonensisandM.rooseveltorumas their sister species.Notably, the geographical range ofM.vuquangensisoverlaps with that ofM.rooseveltorumandM.truongsonensisin Laos and Vietnam, butM.putaoensisdoes not overlap withM.rooseveltorumorM.truongsonensis, despite being more closely related to these two species (Figure 1B).

Figure 1 Bayesian inference (BI)- and maximum-likelihood (ML)-based phylogenetic trees for 12 muntjac species based on mitochondrial DNA

At present, the classification of red muntjacs is still controversial.Although up to five species have been described, the most commonly accepted are the northern red muntjac (M.vaginalis) and southern red muntjac (M.muntjak)(Groves & Grubb, 2011).Recent research supports the delineation of either three monotypic species or three subspecies of red muntjac (Martins et al., 2017).Our research supports the classification of red muntjac into three monotypic species.We proposeM.malabaricusas a valid monotypic species ofMuntiacusbecause our phylogenetic trees strongly suggest thatM.malabaricusis monophyletic; notably,M.malabaricusalso differs morphologically from its most closely related species,M.vaginalis, being smaller in size with shorter antlers and pelage color differentiation (Groves & Grubb,2011).In addition,M.malabaricusandM.vaginalisare distributed relatively independently.Our phylogenetic analyses indicate that subspecies or multiple populations may exist within theM.muntjakandM.vaginalisspecies.

Our results also suggest thatM.gongshanensisis a valid species.We base this conclusion on the following evidence:(a) Our phylogenetic trees strongly supportM.gongshanensisandM.crinifronsas monophyletic; (b) The haplotype networks strongly support the separation ofM.gongshanensisandM.crinifronsby 12 mutational steps and three missing vectors;(c) The geographical distributions ofM.gongshanensisandM.crinifronsdo not overlap at all (Figure 1A), being in the west and east of China, respectively, and separated by more than 2 000 km and many mountain and river barriers; (d)Muntiacus gongshanensisshows different morphology fromM.crinifrons,e.g.,M.gongshanensisis small, light in color, long in hoof,and short in tail, with no crown (Groves & Grubb 2011; Ma et al., 1990).In addition, althoughM.gongshanensispossesses the same number of karyotypes asM.crinifrons, they are different in structure (Shi & Ma, 1988).However, results showed thatM.gongshanensisis most closely related toM.crinifrons, followed byM.feae, similar to previous studies (Ma et al., 1986; Wang & Lan, 2000; Zhang et al., 2019a).

Some researchers have argued thatM.gongshanensisis endemic to China (Jiang et al., 2016; Wang, 2003).However,other studies have shown that the distribution ofM.gongshanensismay have extended to northern Myanmar and India (Smith et al., 2010; Timmins & Duckworth, 2016).Our study provides molecular evidence ofM.gongshanensisdistribution in southwest China (including Namdapha region),northern Myanmar.Eight sequences published asM.crinifronsare in factM.gongshanensis(GenBank accession Nos.: DQ445732-DQ445735, EF523661-EF523664).These samples were obtained from Tibet and the Namdapha region,respectively (Chen et al., 2008; James et al., 2008).In addition, images ofM.gongshanensiswere acquired in the Tengchong Section of Gaoligongshan National Nature Reserve, Yunnan Province in May 2018 (Huang et al., 2019).In summary,M.gongshanensisis distributed in southwest China (Namdapha, Modong, Zeyu, and Gongshan), northern Myanmar (Putao).The expansion of theM.gongshanensisdistribution provides new evidence for its assessment.

AlthoughMuntiacusis a model genus for studying evolution,research remains scarce.At present, 54% ofMuntiacusspecies listed in the IUCN have not been assessed due to insufficient data, although evidence indicates that their populations are declining.Therefore, we strongly recommend the planning of effective conservation strategies forMuntiacusand for these distant regions to maintain genetic diversity and protect these precious species.This study improves our understanding of the phylogenetic relationships within theMuntiacusgenus and will assist in future research and the protection of genetic diversity of muntjacs.

SUPPLEMENTARY DATA

Supplementary data to this article can be found online.

COMPETING INTERESTS

The authors declare that they have no competing interests.

AUTHORS’ CONTRIBUTIONS

Y.C.Z., G.G.L., and R.C.Q.conceived, designed, and performed the experiments and analyzed the data.Y.H.L.,R.L., and K.M.performed field and laboratory work.Y.C.Z.analyzed the data.Y.C.Z.and G.G.L.participated in writing the manuscript.All authors read and approved the final version of the manuscript.

ACKNOWLEDGEMENTS

We thank the Central Laboratory, Public Technology Service Center, Xishuangbanna Tropical Botanical Garden (XTBG),Chinese Academy of Sciences (CAS) for laboratory space.We thank the Kunming Cell Bank for providing theM.gongshanensissample.We thank Qiang Liu for helping with sample collection in the field.