Introduction

The long-tailed goral (Naemorhedus caudatus) is a member of the tribe Rupicaprini in the subfamily Caprinae (An, 2006). The distribution of this species ranges from the northern to central Baekdoodaegan mountain range in South Korea (Choi & Park, 2004; 2005; Yang, 2002). The goral has been listed as an endangered species (Ministry of Environment, 2004) and a natural monument (No. 217; Cultural Heritage Administration of Korea, 1999) by the South Korean government. Internationally, this species is also considered endangered (Baillie & Groombridge, 1996). Commercial trade of this species is prohibited in countries signatory to the Convention on International Trade in Endangered Species (Hutton & Dickson, 2000). The population size of the goral has gradually decreased to fewer than 800 individuals because of illegal hunting, overexploitation, habitat destruction, and habitat fragmentation in South Korea (Ministry of Environment, 2002; Yang, 2002).

In particular, habitat fragmentation is a major issue in the conservation of the goral. For wild populations, habitat fragmentation has caused decreased connectivity, i.e., the degree to which the landscape facilitates or impedes movement between resource patches (Coulon et al., 2004; Taylor et al., 1993). Heterogeneity caused by fragmentation can create natural barriers against population movement, because unfavorable habitats do not provide cover against predators or because distances between suitable patches are greater than can be rapidly crossed (Arnold et al., 1993). Therefore, the movement ability of wild animals may be altered by habitat fragmentation. Such changes can have negative consequences for wildlife populations, partially due to the reduction of gene flow between populations, which leads to greater inbreeding and loss of genetic diversity within fragments (Coulon et al., 2004; Frankham et al., 2002). Fragmentation can even lead to the extinction of species, and many studies on habitat fragmentation have focused on fragmentation created by barriers, such as roads, islands, and settlements (Burkey, 1989; Coulon et al., 2004; Soulé et al., 1992). In the case of wild goral in South Korea, few studies have hitherto been conducted on the ecology and genetics of this species (Choi & Park, 2004; 2005). Suitable goral habitats are expected to significantly decline and/or disappear from South Korea in this century (Lee et al., 2019).

Collecting elusive animal samples (e.g., tissue, blood, etc.) in the wild is difficult, but non-invasive sampling methods could overcome such constraints (Kohn & Wayne, 1997). Fecal sampling is the most frequently used method of obtaining wildlife DNA (Wehausen et al., 2004). As a useful genetic marker in non-invasive wildlife research, the mitochondrial cytochrome b gene has been used for phylogenetic, population, and forensic investigations (Parson et al., 2000). The nucleotide sequence of this gene contains species-specific information. Furthermore, the cytochrome b gene is located on the mitochondrial genome, making it suitable for forensic PCR-based mitochondrial DNA typing. Taxonomic researches—using invasive samples—based on mitochondrial genes, such as cytochrome b (An, 2006; Min et al., 2004) gene or D-loop (An, 2006) region have been conducted on the goral in South Korea. A population study on the goral was recently performed using invasive samples based on 12 microsatellite loci (Choi et al., 2015). For the first time to my knowledge, the present study assessed the phylogenic status and genetic diversity of the long-tailed goral inhabiting South Korea via a simple method using non-invasive fecal samples.

Materials|Methods

Results

Three goral haplotypes defined by only two polymorphic sites were found (Table 2). The polymorphic sites were a synonymous transversion (Ala, GCA/GCC) and a synonymous transition (Asp, GAC/GAT), and no insertion or deletion existed (Fig. 2). The haplotype (H) frequencies of H1, H2, and H3 were 0.056, 0.917, and 0.028, respectively, in the total Korean goral population. Among the three haplotypes, H2 was found in the non-Soraksan (1), Soraksan (0.882), and unknown (0.889) populations, but H1 was restricted to the Soraksan population (0.118), and H3 only occurred in the unknown population (0.111; Fig. 3). The frequency of H2 was similar in the non-Soraksan (1) and Soraksan (0.882) populations, excluding the unknown population (Fig. 3). I also tested the neutral evolution hypothesis for the Korean goral, and neither Tajima’s D value nor Fu and Li’s D value showed strong selective sweeps in this species (Table 2).

The consensus neighbor-joining tree was reconstructed with the three haplotypes and other reference sequences and rooted by two outgroup species, such as goat and sheep (Fig. 4). The three haplotypes formed a separate cluster from those of Chinese and Himalayan gorals but not from the Russian goral (Fig. 4). The latter showed one of the three Korean goral haplotypes (H2). Furthermore, TCS generated a 95% parsimony haplotype network for the Korean, Chinese, and Himalayan goral haplotypes (Fig. 5). The network showed a clearly separated distribution pattern between the Korean goral and Chinese and Himalayan gorals (Fig. 5). The haplotype diversity was 0.160 in the total population and ranged from 0 in the non-Soraksan population to 0.221 in the Soraksan population (Table 2). The nucleotide diversity was 0.053% in the total population and differed slightly between the non-Soraksan (0%) and Soraksan (0.071%) populations (Table 2).

Discussion

The Korean long-tailed goral is a species in the tribe Rupicaprini, subfamily Caprinae, and family Bovidae (An, 2006). Within the tribe Rupicaprini, gorals and serows are classified as Naemorhedus and Capricornis, respectively (Mead, 1989; Nowak, 1999; von Dolan, 1963). In contrast, Groves and Grubb (1985) suggested that serows should also be classified as Naemorhedus (Corbet & Hill, 1992; Grubb, 1993). Few studies have hitherto reported genetic differences between the two genera (An, 2006; Min et al., 2004). The present study also showed similarities in the phylogenetic status of the two genera based on the neighbor-joining tree (Fig. 4) and maximum spanning network (Fig. 5). The Korean goral should be classified as a genus distinct from Capricornis. In addition, the Korean and Russian gorals have the same haplotype and clustered within the same clade (Fig. 4). However, the Chinese goral (N. caudatus) has relatively higher genetic variation than the Korean and Russian gorals do and therefore clustered with the Himalayan goral (N. goral; Fig. 4). This result agreed with those of previous studies (An, 2006; Min et al., 2004). For the reasonable solution of this taxonomical issue, more samples of Chinese and Russian gorals in eastern China and Russia adjacent to Korean habitats are necessary.

Based on the partial mitochondrial cytochrome b gene sequences (310 bp), only three haplotypes were found for the Korean goral (Table 2). The sequence divergence between these haplotypes was 0.4% for the total population (Table 2). Min et al. (2004) found only two haplotypes (Korean goral types a and b) in partial cytochrome b sequences (646 bp) of the species, with a 0.16% sequence divergence. According to a more recent study on the Korean goral (An, 2006), seven haplotypes (A) of complete mitochondrial cytochrome b gene sequences were identified (N. caudatus KG-a, b, c, d, e, f, and g), but the sequence divergence (ps) ranged from 0 to 0.4%. Furthermore, the haplotype (h=0.323) and nucleotide (π=0.053%) diversity of the species according to the present results were low (Table 2). In contrast, other ungulate species, such as moose (Alces alces; 403 bp; A=8; ps=0.2-1.8%; h=0.56-0.60; π=0-0.3%; Hundertmark et al., 2002), western red deer (Cervus elaphus; complete 1,140 bp; A=21; ps=1.86±0.22%; h=0.98±0.02; Ludt et al., 2004), and eastern red deer (C. elaphus; complete 1,140 bp; A=13; ps=1.27±0.020%; h=0.98±0.03; Ludt et al., 2004), showed relatively high genetic diversity in their mitochondrial cytochrome b gene sequences.

The population structure of the long-tailed goral has been rarely studied despite it being one of the most endangered species in South Korea. Yang (2002) reported that the total population consisted of many small, fragmented populations. Choi and Park (2004; 2005) indicated that Korean goral populations are isolated by local highways or forest trails in one of its main habitats. Using invasive sampling based on 12 microsatellite loci (n=68 gorals), Choi et al. (2015) recently reported that the goral population in the lower northeastern regions of South Korea was distinct from the upper northeastern population. The genetic integrity and individual identification of goral in Soraksan National Park were studied using non-invasive sampling based on the mitochondrial D-loop region (n=38 fecal samples) and nine microsatellite loci (Jang et al., 2020). This study is the first to report results of the phylogenic relationship and genetic diversity of the Korean goral compared with those of other species using non-invasive fecal samples. In particular, non-invasive molecular genetic studies using fecal samples could provide more information to support the conservation of South Korean goral populations.