The phylogeny and taxonomic status of the red goral (Naemorhedus baileyi) is still unclear. We sequenced the complete cyt b (1140bp) gene extracted from hair samples from three animals from the Shanghai Zoo, and compared them with thirty sequences of Naemorhedus downloaded from GenBank. Our results show three distinct lineages within Naemorhedus. Two closely related N. baileyi haplotypes were found. They belong to the most basal group within Naemorhedus together with N. griseus haplotype 2 from Thailand, which is a sequence likely to be misidentified.
The red goral (Naemorhedus baileyi, Caprinae, Bovidae) is under first class protection in China, classified as Vulnerable on the IUCN Red List (Duckworth & MacKinnon 2011), and is listed in Appendix I of CITES (2012). The red goral inhabits the southeastern Tibet and the northwestern Yunnan Province, and is also found in northern Myanmar and northeastern Arunachal Pradesh in India (Smith 2009). However, the red goral is threatened by hunting and habitat loss (Wang 1998). Wilson & Mittermeier (2011) suggest that fewer than 10000 red goral are believed to survive today, and fewer than 1500 red goral are thought to live in China. Guo (2004) estimated that almost 500 red goral had been hunted every year in Medog, Tibet.
Because of its small population size and distribution range, research into the red goral is sparse. We sequenced the red goral mtDNA cyt b genes, and compared them with cyt b sequences from other species including long-tailed goral (N. caudatus), Himalaya goral (N. goral) and Chinese goral (N. griseus), which are generally considered as separate species in Naemorhedus (Wilson & Reeder 2005, Smith 2009), to investigate the relationships among them.
Material and Methods
We collected hair samples from three red gorals using a non-invasive sampling method from the Shanghai Zoo. The three red gorals were descendants of the red goral captured in the wild from Milin in 1981, and they were bred in captivity (Wang 1998). All the samples were preserved at -20 °C. The other thirty sequences analyzed in this study were downloaded from GenBank (Table 1).
DNA extraction and PCR amplification
5′-GGAATTCATCTCTCCGGTTTACAAGAC-3′ Thermal cycling was performed on a PTC-200 thermocycler (MJ Research, Inc. USA). The PCR amplifications were carried out in 25 μl volumes using 7 μl of DNA extract, 2.5 μl of 10 × PCR buffer, 2.5 μl of MgCl2 (25 mM), 2 μl of dNTPs (25 mM), 0.7 μl of each primer at 10 μM, and 0.2 μl of 5U rTaq (Takara, Dalian, China). The PCR program had 35 cycles with 95 °C for 30 s, 54 °C for 30 s and 73 °C for 60 s. All reactions were started with a denaturation step at 95 °C for 2 min and the last cycle was followed by a 5 min extension at 72 °C.
The PCR products were visualized using 2 % agarose gel electrophoresis, then purified and sequenced at Shanghai Map Biotech Co., Ltd.
The sequences of cyt b genes were aligned in ClustalX 2.0 (Larkin et al. 2007) and then manually collated. The total genetic divergence and its standard error (uncorrected p-distances) was calculated in MEGA 4.0 (Kumar et al. 2008). Haplotype diversity and nucleotide diversity were explored using DnaSP 5 (Librado & Rozas 2009). Maximum likelihood (ML) phylogenetic tree was constructed using Phyml 3.0 (Guindon & Gascuel 2003). We used Modeltest 3.7 (Posada & Crandall 1998) to test the optimal substitution model. The optimal model selected by the Akaike Information criterion (AIC) is GTR + I (lnL = -3621.1016, K = 9, AIC = 7260.2031). One hundred bootstrap replications were used for the ML analysis. We used Ovis aries (accession No. FR873153), Oreamnos americanus (accession No. AF190632) and Budorcas taxicolor (accession No. FJ207524) as outgroup species.
The sequence fragment was 1140bp long, no insertions, deletions or stop codons were observed in the coding sequence. The fragment contained the entire mitochondrial cyt b gene because all the sequences began with ATG and terminated with the stop codon AGA (Hassanin et al. 1998).
Comparison of all sequences in Naemorhedus is listed in Table 2. Genetic divergence between the red goral and long-tailed goral, Himalaya goral and Chinese goral was 0.090, 0.092 and 0.072, respectively. The haplotype diversity and nucleotide diversity of the red goral was comparable to that found in the long-tailed goral, which is the only species represented by more than two individuals (Table 3).
Genetic divergence (p-distance) (below diagonal) and standard error (above diagonal) for cytochrome b nucleotide sequences of animals examined in this study.
Haplotype diversity and nucleotide diversity of all sequences.
The Maximum likelihood phylogenetic tree is shown in Fig. 1. The ML phylogeny showed polyphyletic relationships of Naemorhedus species. The most basal supported group contained the red goral and Chinese goral from Thailand (Hassanin et al. 2012). Next diverged group containing most long-tailed goral sequences and a group that included Himalaya goral, Chinese goral from an unlisted locality (Hassanin et al. 2009) and long-tailed goral from San Diego Zoo, USA (Groves & Shields 1996).
This study provided insights into the taxonomic status of Naemorhedus. The results indicate that the red goral is an independent species of Naemorhedus. The four sequences are closely related and they belong to two haplotypes. The red goral sequence downloaded from GenBank was collected from the Rotterdam Zoo (Hassanin et al. 2012). Because of high similarity, the red goral in Shanghai Zoo and Rotterdam Zoo may have the same origin. The known distribution range of the red goral includes the southeastern Tibet, northwestern Yunnan Province, northern Myanmar and northeastern Arunachal Pradesh in India. Although Groves & Grubb (1985) distinguished two subspecies of the red goral (Tibetan red goral N. b. baileyi and Burmese red goral N. b. cranbrooki), our sampling does not allow assessing their status. More data from multiple regions in the distribution area of the red goral are needed to clarify the subspecies status of the red goral, but our results refute Rabinowitz's (1999) view recognizing the red goral as long-tailed goral's subspecies.
It is generally considered that there are four separate species in Naemorhedus: red goral (N. baileyi), longtailed goral (N. caudatus), Himalaya goral (N. goral) and Chinese goral (N. griseus) (Wilson & Reeder 2005, Smith 2009). However, their phylogenetic relationships were not resolved based on cyt b. The phylogenetic analyses showed three major evolutionary clades, whose monophyly was supported by high bootstrap values. The long-tailed goral and the Chinese goral were polyphyletic in the tree. This might indicate either misidentification of some samples or lack of lineage sorting of mitochondrial genomic lineages in Naemorhedus.
The lineage that is well defined in our data included long-tailed goral individuals sampled from the wild. They are found in eastern Russia, northeastern China and Korean Peninsula (Smith 2009). The Korean goral (N. c. raddeanus) is considered a separate subspecies, but cyt b data do not indicate diversification within the lineage that would merit the subspecies classification. The sister lineage to the long-tailed goral from the wild is enigmatic. It groups three species with little divergence. Here, the long-tailed goral individual with haplotype N.c 3 from San Diego Zoo seems to carry different mtDNA, which might be an introgression. Sequences of the Himalaya goral and the Chinese goral are very similar indicating conspecificity, introgression or sample misidentification.
The most basal lineage of Naemorhedus includes the red goral and Chinese goral from Thailand. Hassanin et al. (2012) admitted that the haplotype N.gr 2 from the sample collected in Thailand by local hunters probably suffers from species misidentification. Their reason was the high divergence between the N.gr 2 and the haplotype N.gr 1 reported by Hassanin et al. (2009). The Chinese goral from Thailand was described as a subspecies of N. griseus by Hassanin et al. (2012), and the authors suggested that Chinese goral in Thailand could be elevated to a full species. Detailed systematic evaluation of the Chinese goral from Thailand is necessary as in this study, we show that they are closely related to the red goral.
We thank Richard Fautley and Tian Xinxin for comment on English writing, we thank Zhou Lichen and He Ya for assistance in data analyses, we thank He Xin and Yu Xiaojun for assistance in experiment, and we also thank veterinarian Gui Jianfeng from Shanghai Zoo for assistance in sample collection.