The tiger Panthera tigris is the most threatened large felid, with an estimated global population of c. 3,500 individuals (Morell, Reference Morell2007). Three of nine subspecies have been extirpated by anthropogenic pressures and associated habitat loss (Luo et al., Reference Luo, Kim, Johnson, Walt, Martenson and Karanth2004; Dinerstein et al., Reference Dinerstein, Loucks, Wikramanayake, Ginsberg, Sanderson and Seidensticker2007). Three of the extant tiger subspecies, one being the Amur tiger P. tigris altaica, are found in China (Luo et al., Reference Luo, Kim, Johnson, Walt, Martenson and Karanth2004). The Amur tiger, categorized as Endangered on the IUCN Red List (Miquelle et al., Reference Miquelle, Darman and Seryodkin2011), has experienced significant reduction of its extensive distribution across northern China as a result of rapid human population growth, habitat loss and poaching (Ma et al., Reference Menotti-Raymond, David, Lyons, Schäffer, Tomlin, Hutton and O'Brien2005), declining from c. 150 individuals in the 1970s to < 20 in 2000 (Yang et al., Reference Yang, Jiang, Wu, Li, Yang and Han1998; Sun et al., Reference Sun and Dong1999; State Forestry Administration, 2009). Intensive hunting and the capture of cubs for zoos led to a similar decline of the Amur tiger in the Russian Far East during the early 20th century (Heptner & Sludskii, Reference Höglund, Johansson, Beintema and Schekkerman1972).
Since 1947 the Amur tiger population in the Russian Far East has increased as a result of a prohibition on tiger hunting, reduction in capture of cubs and implementation of management plans for prey. It is estimated there were 428–502 tigers in 2005 (Miquelle et al., Reference Miquelle, Goodrich, Kerley, Pikunov, Dunishenko, Aramiliev, Tilson and Nyhus2005), mostly in the Sikhote-Alin Mountains, with a subpopulation of 16–21 in the Russian south-west Primorye (Pikunov et al., Reference Pikunov, Miquelle, Abramov, Nikolaev, Seryodkin, Murzin and Korkiskho2003) potentially serving as a source population for nearby Jilin and Heilongjiang provinces in China (Sun & Dong, Reference Sun, Miquelle, Xiaochen, Zhang, Hiyai and Goshen2005; Chen et al., Reference Chen, Na, Shun, Zhang, Tang and Lang2011). Previous studies have focused on population variation and genetic structure of tigers in Russia (Russello et al., Reference Russello, Gladyshev, Miquelle and Caccone2004; Henry et al., Reference Heptner and Sludskii2009; Miquelle et al., Reference Mondol, Karanth, Kumar, Gopalaswamy, Andheria and Ramakrishnan2010; Sugimoto et al., Reference Sugimoto, Nagata, Aramilev and McCullough2012). The demographic and genetic status of tigers in China is unclear (State Forestry Administration, 2009). In this study we therefore analysed scat samples to assess genotypic variation for identification of individuals of the Chinese tiger population in Jilin Province, and provide baseline estimates of genetic diversity to inform recovery plans for this population.
The field survey was conducted in the 1,087 km2 Hunchun Nature Reserve in Jilin Province, China, which occupies a long, narrow strip adjacent to south-west Primorye, Russia (Fig. 1). Eleven scat samples were collected within and adjacent to the reserve during tiger monitoring surveys. The surveys covered 1,182 km2 and were conducted during 14–23 November 2009, 22 January–2 February 2010, and for 5 days during the winter/spring of 2011. Genomic DNA was extracted using the QIAamp DNA Stool Mini kit (Qiagen, Valencia, USA) following the manufacturer's protocol, with modifications to improve DNA yield and quality (Caragiulo et al., Reference Caragiulo, Dias-Freedman, Clark, Rabinowitz and Amato2014). Extractions were prepared within a polymerase chain reaction (PCR) product-free laboratory reserved exclusively for scat extractions.
Fig. 1 Location of Hunchun Nature Reserve, showing sampling localities where tiger scats were and were not identified. Some sampling localities were outside the Reserve because there was anecdotal evidence of tigers in these areas. The black circle indicates the nearest tiger population in Russia (Reference Russello, Gladyshev, Miquelle and CacconeRussello et al., Reference Russello, Gladyshev, Miquelle and Caccone2004) and the inset the location of the main map on the China–Russia border. The three shades of grey indicate different zones within the Reserve.
Samples were identified to species via PCR and Sanger sequencing using six mitochondrial DNA primer sets spanning four mitochondrial gene regions (Caragiulo et al., Reference Caragiulo, Dias-Freedman, Clark, Rabinowitz and Amato2014). Sequences were edited using Sequencher v. 4.2 (Gene Codes Corporation, Ann Arbor, USA), and species were identified via neighbour-joining trees (1,000 bootstrap replicates) implemented in Geneious Pro v. 5.6 (Biomatters, Auckland, New Zealand), constructed using > 2,000 in-house reference sequences spanning multiple animal species. Each identified tiger sequence was accessioned into the BOLD System (Ratnasingham & Hebert, Reference Ratnasingham and Hebert2007) and GenBank (Benson et al., Reference Benson, Karsch-Mizrachi, Lipman, Ostell and Wheeler2005) when exceeding 200 bp. Generated sequences were compared to references on GenBank for additional confirmation of species identity.
Fluorescently labelled nuclear microsatellite loci were used to identify individuals, and amplified using the QIAgen Multiplex Kit (Qiagen, Valencia, USA) in 20 μl multiplex reactions using 5 μl of template DNA, 10 μl of MasterMix, 2 μl of Q-solution, 0.1–0.8 μM of each 10 μM primer, brought to volume using ultrapure water. A total of 11 microsatellite loci developed from the domestic cat (Menotti-Raymond et al., Reference Miquelle, Darman and Seryodkin1999) were used (Supplementary Table S1). Microsatellite amplifications were analysed using GeneMapper v. 4.0 (Applied Biosystems, Carlsbad, USA). Genotypes were validated using the quality index (Mondol et al., Reference Miquelle, Pikunov, Dunishenko, Aramilev, Nikolaev and Abramov2009) to ensure reliable genotyping. Samples amplifying fewer than five loci were removed from further analysis.
The gender of each tiger sample was determined via amplification of the zinc finger regions on the X and Y chromosomes (Wei et al., Reference Wei, Zhang, Zhang, Xu, Liang and He2008). Gender-typing PCRs consisted of 25 μl reactions with 0.60 μl of each 0.25 μM primer, 5 μl of template DNA, one illustra puReTaq ready-to-go PCR bead (GE Healthcare, Piscataway, USA), brought to volume with ultrapure water. The two forward primers (ZF-1F and ZFY-2F) were fluorescently labelled for analysis in an ABI 3730xl DNA Analyser (Applied Biosystems) and scored using GeneMapper.
Each sample underwent at least four independent PCR reactions per locus and gender-typing reaction using a modified multiple-tubes approach to identify allelic dropout or false alleles, and to confirm allele calls and gender-typing results (Taberlet et al., Reference Taberlet, Griffin, Goossens, Questiau, Manceau and Escaravage1996; Bhagavatula & Singh, Reference Bhagavatula and Singh2006; Mondol et al., Reference Miquelle, Pikunov, Dunishenko, Aramilev, Nikolaev and Abramov2009). GIMLET v. 1.3.2 (Valière, Reference Valière2002) was used to determine error rates, CERVUS v. 3.0 (Kalinowski et al., Reference Luo, Johnson, Martenson, Antunes, Martelli and Uphyrkina2007) was used to identify individual genotypes, and genetic diversity indices were estimated using both CERVUS and GENEPOP v. 4.0 (Rousset, Reference Rousset2008).
Nine scat samples were identified as tiger, and the remaining two as canids. Mitochondrial DNA sequences were accessioned into BOLD (GFG001–GFG009) and GenBank (JQ040982–JQ040983, JQ040929–JQ040937). Recapture samples were found (samples CH004/CH005 and CH007/CH010/CH011 were from the same individual tiger). Samples amplified all loci, except for CH008 and CH009, which amplified 9 loci; CH003 only amplified four loci and was excluded from individual identification analysis. We discerned a total of five individuals, representing four prospective males (CH001, CH004/005, CH007/010/011, CH008), and one prospective female (CH009). The P ID = 0.0001 and the P ID(sib) = 0.013 indicated our loci panel had sufficient statistical power for discerning individuals. Overall, the mean PCR success for all loci was 0.76 (range 0.52−0.88), the mean allelic dropout rate across loci was 0.290 (0.000−0.838), and the mean number of false alleles across loci was 0.008 (0.000−0.025). The mean number of alleles per locus (N A ) was 2.55, the mean expected heterozygosity (H e ) was 0.438, and mean polymorphic information content (PIC) was 0.288 (Supplementary Table S2). Tests for deviation from Hardy–Weinberg equilibrium and linkage disequilibrium could not be performed because of the small sample size. The F IS value was 0.125.
Molecular techniques are widely used to investigate the genetics of threatened species because they provide important insights into taxonomic status, evolutionary history and population genetic structure (Bhagavatula & Singh, Reference Bhagavatula and Singh2006). Such information is especially important for small populations, as they are more susceptible to loss of genetic variation through inbreeding (Frankham et al., Reference Henry, Miquelle, Sugimoto, McCullough, Caccone and Russello2010). To our knowledge this is the first study to provide population genetic information on wild tigers in China using non-invasive techniques. Based on track surveys, it was estimated previously that there were 3–5 tigers in Hunchun Nature Reserve (State Forestry Administration, 2009). Our analysis identified at least five individual tigers in this area. The presence of at least one female indicates breeding potential but the male-biased sex ratio is not ideal for this. Our genetic diversity indices are not statistically powerful because of the small sample size but are nonetheless valuable because of the scarcity of information on wild Amur tigers in China.
The Amur tiger may be the most genetically impoverished extant tiger subspecies (Luo et al., Reference Luo, Kim, Johnson, Walt, Martenson and Karanth2004, Reference Ma, Zhang, Miquelle, Wang, Kang, Zhang, Miquelle, Wang and Kang2008; Henry et al., Reference Heptner and Sludskii2009). The sex ratio of Amur tigers in the Russian Far East appears to be female-biased (Sugimoto et al., Reference Sugimoto, Nagata, Aramilev and McCullough2012); the male-bias of the population in Hunchun Nature Reserve could be a result of the dispersal of male tigers from nearby Russian reserves. The protected areas on both sides of the border should therefore be considered a single geographical unit for tiger conservation.
This study provides valuable baseline genetic diversity estimates through non-invasive monitoring for a rare tiger population warranting management. Although we found relatively few tiger scats, our results indicate there is a potentially viable population in Hunchun Nature Reserve, albeit with low levels of observed genetic diversity. Previous studies have shown that small, threatened populations can be managed in the wild to maintain viability (Höglund et al., Reference Kalinowski, Taper and Marshall2009), and we believe this applies to the Hunchun Amur tiger population. It could be queried whether conservation resources should be invested in a small, peripheral population, but we believe this is appropriate if it contributes to the genetic diversity of the species, especially one as geographically disparate as the tiger. Our findings will aid the development of conservation management strategies for the tiger in China.
Acknowledgements
We are grateful to the staff of Hunchun Nature Reserve, Institute of Forestry of Jilin Province, the WWF Changchun Office, and the Wildlife Conservation Society Hunchun Office, who supported our field surveys and sample collection. We thank Dr Xie Yan from the Institute of Zoology, Chinese Academy of Sciences for providing some of the scat samples. The National Science Foundation of China funded this work for the National Biological Sciences Base (J1103516) and the Forestry Industry Research Special Funds for Public Welfare Projects (201104029). We also thank Panthera, and Ashley Yang, Ellen Trimarco, Eric Dougherty, Mohammad Faiz and Stephen Gaughran of the Sackler Institute for Comparative Genomics of the American Museum of Natural History, New York.
Biographical sketches
Anthony Caragiulo’s research interests include conservation genetics and using non-invasive molecular techniques to monitor wildlife, especially cougars. Yang Kang is interested in monitoring threatened animal habitat. Salisa Rabinowitz is interested in global conservation of large felids, with the help of non-invasive molecular techniques. Isabela Dias-Freedman’s research interests include animal and human genetics, using molecular techniques for wildlife conservation, genetic resistance to emergent diseases, and forensic science. Simone Loss is interested in molecular techniques for wildlife conservation. Xu-Wei Zhou focuses on the molecular ecology of rare mammals. Wei-Dong Bao’s research interests include wildlife ecology and felid conservation. George Amato’s research interests include non-invasive sampling for threatened species and monitoring the trade in threatened species products, using DNA-based forensic science.
