Individualization of tiger by using microsatellites

Forensic Science International 151 (2005) 45 51 www.elsevier.com/locate/forsciint Individualization of tiger by using microsatellites Yan Chun Xu a,b, *,BoLi a, Wan Shui Li c, Su Ying Bai a, Yu Jin a, Xiao Ping Li d, Ming Bo Gu d, Song Yan Jing a, Wei Zhang a a State Forestry Administration Detecting Center of Wild Fauna and Flora, Northeast Forestry University, No. 26, Hexing Road, Harbin 150040, PR China b State Conservation Center for Gene Resources of Endangered Wildlife, College of Life Science, Zhejing University, No. 268, Kaixuan Road, Hangzhou 310029, PR China c Institute of Forensic Sciences, Ministry of the Public Security, No. 18, Muxidi Nanli, Xicheng District, Beijing 100038, PR China d Department of Heilongjiang Province, Forensic Laboratory of the Public Security, No. 45, Shizi Street, Harbin 150008, PR China Received 20 March 2003; received in revised form 1 July 2004; accepted 7 July 2004 Available online 25 August 2004 Abstract In investigating criminal cases of poaching and smuggling involving tigers (Panthera tigris), the number of tiger individuals involved is critical for determining the penalty. Morphological methodologies do not often work because tiger parts do not possess the distinctive characteristics of the individual. Microsatellite DNAs have been proved a reliable marker for the individualization of animals. Seven microsatellite loci derived from domestic cat (Felis catus) were selected to individualize tigers, namely F41, F42, F146, Fca304, Fca391, Fca441 and Fca453. A reference population containing 37 unrelated tigers were used to investigate alleles, allelic frequencies, genotypes and genotype frequencies of each locus. Consequently, the data was used to assess the validity of the combination of seven loci for tiger individualization. All loci were polymorphic and easy to amplify. Three out of the seven loci were significantly departure from the Hardy Weinberg Equilibrium (P < 0.05). Cumulative discrimination power (DP) calculated with observed genotype frequencies was 0.99999789. Match probability of an individual in the reference population with a random individual in seven loci ranged from 7.34 10 9 to 2.77 10 5. This suggests that combining the seven microsatellite loci provides desirable power to individualize tigers. The combination of seven loci was applied to a case of tiger bone smuggling. Genotypes of all samples were identical in all seven loci, and the P M of the evidence samples in the seven loci hit 5.63 10 7, provided evidence that the bones belong to a single tiger. # 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Microsatellite; Tiger; Panthera tigris; Individualization 1. Introduction * Corresponding author. Present address: College of Life Sciences, Zhejiang University, No. 268, Kaixuan Road, Hangzhou, 310029 Zhejiang, PR China. E-mail address: xu_daniel@yahoo.com (Y.C. Xu). Tiger parts, like bone, penis, eyeball, blood, muscle, heart, etc., are considered precious medicines that have been used for many centuries within traditional Chinese medicine (TCM). The impression of tiger parts in the mind of oriental people has not yet disappeared though many countries have 0379-0738/$ see front matter # 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2004.07.003

46 Y.C. Xu et al. / Forensic Science International 151 (2005) 45 51 announced a cessation in the trade and use of tiger parts. The continuing black-market trade fuelled by these persistent strong beliefs in the benefits of such derived medicines in turn drives a number of individuals to continue to venture upon the speculation of tiger parts [1]. Therefore, species identification and individualization of tigers from seized evidence become an important issue for forensic detection. It is difficult to identify or individualize tigers based on the morphological characteristics except for in the cases where the tiger parts with entire taxonomic characteristics are obtained such as a complete mounted specimen. However, molecular biological strategies are often reliable to deal with samples that lack such complete characteristics. Molecular biological methods for the species identification of tigers have been previously reported [2,3]. Nevertheless, complete individualization of tigers has not yet been instigated. Microsatellites have been proved to be a reliable marker for the individualization of animals [4,5]. But very few microsatellite loci had been isolated from the tiger genome except for the 11 loci deposited in the GenBank. The flanking regions of microsatellite DNAs are very conservative across taxa, so transferring PCR primers to related species becomes possible [6,7]. Thus tiger individualization can be based on the feline microsatellites. Cases where the application of such a technique that would be a great benefit are not hard to find. For instance on 21 October 2002, the officers of Suifenhe Custom, China were informed that somebody would be attempting to smuggle tiger bones into China across the Russian border. The police subsequently arrested four suspects who were in possession of some tiger bones. It was required to find out how many tigers were involved in this case. Therefore, the technique firstly developed and described in this paper was utilized for this case and the results also included here. 2. Materials and methods 2.1. Tigers for the population investigation A population consisting of 37 unrelated tigers were taken as a reference population to investigate the allele number, allelic frequency and genotype frequency for validating these loci prior to the case analysis. The tigers were from the Feline Breeding Base of Guangxi Endangered and Rare Wildlife Refuge, China, which purchased three tiger subspecies from various zoos in China and Italy, France, and USA. Most individuals of the population were unrelated. The reference population used here includes 21 Amur tigers (Panthera tigris altaica), six south China tigers (P. t. amoyensis) and 10 Bengal tigers (P. t. tigris). Venous blood was sampled from each tiger, anticoagulated by adding with 10 ml 0.5 M EDTA (ph 8.0) per microlitre blood and kept at 4 8C for hours before treatment. 2.2. DNA isolation, PCR amplification and genotyping DNA was isolated with routine phenol:chloroform method [8] and quantified with DU-640 Nucleic Acid Protein Analysis System (Beckman Coulter) according to the user s manual. Primer pairs of seven microsatellite loci, namely F41, F42, F146, Fca304, Fca391, Fca441 and Fca453, derived from domestic cat (Felis catus) were selected to amplify tiger microsatellite DNAs [9]. PCR was set up in a 10 ml system containing 1 PCR buffer containing 10 mm Tris HCl (ph 8.3), 50 mm KCl; 2.5 mm MgCl 2, 250 mm each of four dntp (Takara), 4.0 pmol each of forward and reverse primer, 0.5 units of Taq DNA polymerase (Takara) and 50 ng of genomic DNA. The 5 0 end of each forward primer was labeled with FAM. PCR amplification was performed in a Model 9700 Thermocycler (Perkin- Elmer) using the following program: 1 cycle of 3 min at 94 8C, 10 cycles of 94 8C for 15 s, 55 for 15 s, 72 8C for 30 s, 20 cycles of 89 8C for 15 s, 55 8C for 15 s, 72 8C for 30 s, and 1 cycle of 72 8C for 10 min. An 1 ml PCR product of each locus was mixed with 23 ml loading dye containing 95% formamide and 1 ml GeneScan-500 [ROX] internal size standard, then denatured at 95 8C for 5 min and chilled on ice immediately. PCR products were isolated and sized with 5% denaturing polyacrylamide gel (Acr:Bis = 19:1) on an ABI PRISM 377 Automated DNA Sequencer (Applied Biosystems) and genotyped with GeneScan 3.1 and Geno- Typer 3.1 (Applied Biosystems). Microsoft Excel 2000 was used to sort fragments of PCR products by size and the alleles of each locus were deter- Table 1 The characteristics of the sequenced alleles of the seven microsatellite loci Locus Allele Size (bp) Observed sized (bp) Motif F41 16 130 130.70 (AAGG) 3 (AAAG) 13 F42 22 232 229.40 (TTCT) 3 (CTTT) 4 (CCTT) 3 TTTATTTCCTCTTTCCCTTCCTCCT(CTTT) 12 Fca146 12 160 159.83 (AAC) 12 Fca304 24 131 133.04 (GT) 17 (GG) 1 (GT) 6 Fca391 24.3 210 211.26 (ATGG) 10 (GATA) 11 (TAGA) 2 TGA(TAGA) 1 Fca391 22.3 202 203.11 (ATGG) 7 (GATA) 12 (TAGA) 2 TGA(TAGA) 1 Fca441 13.2 149 150.37 (ATAG) 9 (GTAG) 1 (ATAG) 2 AG(ATAG) 1 Fca453 11 195 196.99 (TAGA) 11

Table 2 Alleles and their size (bp) and frequency of the seven microsatellite loci in the reference population (n =37 a ) Locus Alleles and their size and frequency F41 Allele 33 32 31 29 28 27 26 25 21 20 17 16 15 12 Observed size 200.20 195.74 192.13 184.44 180.61 176.54 172.84 168.76 151.55 147.32 134.78 130.70 126.80 115.19 Actual size 198 194 190 182 178 174 170 166 150 146 134 130 126 114 Frequency 0.0270 0.0135 0.0811 0.0135 0.2297 0.0135 0.0676 0.0811 0.0405 0.0270 0.0270 0.2838 0.0811 0.0135 F42 Allele 24 23 22 19 18 17 16 Observed size 237.50 233.03 229.40 217.20 213.77 209.74 205.91 Actual size 240 236 232 220 216 212 208 Frequency 0.0405 0.2973 0.1216 0.0135 0.0135 0.1486 0.3649 F146 Allele 13 12 11 10 Observed size 162.84 159.83 158.06 156.77 Actual size 163 160 157 154 Frequency 0.0135 0.6351 0.0405 0.3108 Fca304 Allele 25 24 22 21 20.1 19.1 18.1 Observed size 135.09 133.04 128.73 127.27 126.11 124.37 122.39 Actual size 133 131 127 125 124 122 120 Frequency 0.2059 0.4118 0.0735 0.0441 0.2353 0.0147 0.0147 Fca391 Allele 27.3 26.3 25.3 24.3 23.3 22.3 21.3 Observed size 223.23 219.43 215.15 211.26 207.42 203.11 199.12 Actual size 222 218 214 210 206 202 198 Frequency 0.0135 0.0135 0.0676 0.2973 0.2838 0.2027 0.1216 Fca441 Allele 16.2 15.2 14.2 13.2 12.2 11.2 Observed size 162.18 157.86 154.10 150.37 145.66 141.44 Actual size 161 157 153 149 145 141 Frequency 0.0946 0.0405 0.4324 0.1486 0.2432 0.0405 Fca453 Allele 13 12 11 10 Observed size 205.16 201.16 196.99 192.39 Actual size 203 199 195 191 Frequency 0.0676 0.6892 0.2027 0.0405 a For the locus Fca304, n = 34. Y.C. Xu et al. / Forensic Science International 151 (2005) 45 51 47

48 Table 3 Genotypes and their frequencies of the seven microsatellite loci in the reference population (n =37 a ) Locus Genotype and frequency b F41 31/32 16/28 15/28 28/31 26/31 15/16 25/27 25/28 15/21 16/31 16/29 16/16 28/28 16/21 16/26 12/15 25/33 17/25 26/26 16/33 20/20 0.0270 0.2162 0.0541 0.0270 0.0270 0.0270 0.0270 0.0541 0.0541 0.0811 0.0270 0.0541 0.0541 0.0270 0.0541 0.0270 0.0270 0.0541 0.0270 0.0270 0.0270 F42 23/24 23/23 22/22 19/22 17/24 17/23 17/22 17/18 16/23 16/17 16/16 0.0541 0.1351 0.0541 0.0270 0.0270 0.0541 0.1081 0.0270 0.2162 0.0811 0.2162 F146 12/13 12/12 11/11 10/12 10/11 10/10 0.0270 0.4865 0.0270 0.2703 0.0270 0.1622 Fca304 25/25 24/25 24/24 22/25 22/24 18.1/25 21/24 20.1/25 20.1/24 20.1/20.1 19.1/24 0.0294 0.1471 0.1471 0.0588 0.0882 0.0294 0.0882 0.1176 0.1765 0.0882 0.0294 Fca391 23.2/24.3 22.3/24.3 24.3/24.3 21.3/23.3 23.3/25.3 23.3/23.3 21.3/22.3 22.3/22.3 22.3/23.3 22.3/25.3 21.3/24.3 25.3/26.3 21.3/21.3 25.3/27.3 0.1081 0.1081 0.1622 0.0811 0.0541 0.1081 0.0811 0.0270 0.1351 0.0270 0.0270 0.0270 0.0270 0.0270 Fca441 14.2/16.2 14.2/15.2 14.2/14.2 13.2/16.2 13.2/14.2 13.2/13.2 12.2/16.2 12.2/15.2 12.2/14.2 12.2/13.2 12.2/12.2 11.2/16.2 11.2/15.2 11.2/13.2 0.1081 0.0270 0.2432 0.0270 0.0541 0.0541 0.0270 0.0270 0.1892 0.0811 0.0811 0.0270 0.0270 0.0270 Fca453 11/12 11/11 10/11 10/12 9/11 9/10 0.0541 0.5135 0.2703 0.0811 0.0270 0.0541 a For the locus Fca304, n = 34. b The upper data with slash are the genotypes and the lower data are their corresponding frequencies. Y.C. Xu et al. / Forensic Science International 151 (2005) 45 51

Y.C. Xu et al. / Forensic Science International 151 (2005) 45 51 49 mined with the method as reported [10]. One or more alleles in each locus amplified from homozygotes were cloned with pmd18-t vector (TaKara) and sequenced with BigDye 1 Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) on ABI PRISM 1 3100 Genetic analyzer (Applied Biosystems). The sequenced alleles were named according to the nomenclature of STRs [11,12]. The repeat number and name of all alleles remaining unsequenced were then deduced in reference to their observed size and the sequenced alleles. The allelic frequency and genotype frequency in each locus was calculated. The coincidence with Hardy Weinberg expectation (HWE) in each locus was tested by x 2 -test [13]. The discrimination power (DP) of each locus and cumulative DP of seven loci and the probability of the match with a random individual (P M ) were calculated with observed genotype frequency of each locus as mentioned [13]. 2.3. Case analysis We found that some limb bones in the evidence samples could be paired up and likely from a single individual. Therefore, we sampled one piece of bone of each pair and some rachis bones, samples totally numbering 8, representing all the bones. Bone samples were cleaned with 75% ethanol and drilled without cross contamination. The drill dusts of bones were collected to extract DNA with the routine phenol:chloroform method as above. DNA was purified with silica [14]. The amplification and genotyping of the evidence DNAs were performed with the same methods as mentioned above. The genotype of each locus of each bone was used to determine the identity. 3. Results 3.1. Alleles of each locus Various numbers of alleles were defined in the seven loci. The characteristics of the alleles in each locus are listed in Table 1. Except for the locus Fca304 that had a 2-base repeat motif and Fca146 that had a 3-base motif, all other five loci had a 4-base motif. All loci generated clear signals. F41 was the most polymorphic locus that had 14 alleles. F146 and Fca453 were the least polymorphic, four alleles being discovered in each. The other four loci were medium in polymorphism, possessing five to seven alleles. The alleles and their frequencies of all the loci are shown in Table 2. 3.2. Genotype frequency and discrimination power The observed genotype frequency was used to assess the validity of these loci as the genetic markers to individualize tigers (Table 3). The distributions of genotype frequencies of Table 4 Discrimination power (DP) of each microsatellite locus and cumulative DP of the seven loci in tiger individualization Locus DP Cumulative DP F41 0.9174 0.9174 F42 0.8590 0.9884 F146 0.6618 0.9961 Fca304 0.8224 0.99954 Fca391 0.8999 0.999954 Fca441 0.8693 0.9999939 Fca453 0.6501 0.99999789 these loci were not even. A few genotypes had a significantly higher frequency than others. Locus Fca304, Fca441, Fca391 and Fca453 were in accordance with the Hardy Weinberg equilibrium, while the other three, F41, F42 and F146 showed a departure from the HWE (P < 0.05). The discrimination power of each locus ranged from 0.6513 to 0.9135 and the cumulative DP of seven loci hit 0.99999789 (Table 4). Match probability of an individual of the reference population with a random individual in seven loci ranged from 7.34 10 9 to 2.77 10 5. 3.3. Individualization of evidence samples The amplified fragments of evidence in seven microsatellite loci were assigned to alleles discovered in the reference population by size (Table 5). No new allele was found in the evidence samples. All eight samples were identical in genotype through seven loci. The P M calculated with corresponding genotype frequencies of seven loci hit 4.52 10 6, which suggested these bone pieces were from one individual. 4. Discussion The tiger is a species consisting of five surviving subspecies, existing as some scattered captive populations and segmented wild populations. The history of the populations differs greatly from each other and the genetic background also seemed very complex. The individualization by using molecular genetic markers requires a database representing the species and detailing the statistical genetic parameters. However, the actual genetic background of the all tiger populations remained unclear up to the present. We sampled three subspecies of tigers in this experiment, basically representing the whole population in China, to create a molecular genetic database to support the statistical interpretation of the genotyping results in criminal cases involving the three subspecies. Nevertheless, the sample size was limited in each subspecies and thus the diversity of the microsatellites was probably underestimated. There are two subspecies, the Sumatran tiger (P. t. sumatrae) and Indochinese tiger (P. t. corbetii) also not yet included in this reference population. The individualization of these two

50 Y.C. Xu et al. / Forensic Science International 151 (2005) 45 51 Table 5 Fragment size and genotype of evidence samples in the seven microsatellite loci a Sample F41 F42 F146 Fca304 FcA391 Fca441 Fca453 1 147.69/147.69 205.57/205.57 156.26/156.26 125.71/133.94 207.87/207.87 154.51/154.51 201.59/201.59 2 147.82/147.82 205.66/205.66 156.06/156.06 125.61/133.93 207.53/207.53 154.31/154.31 201.34/201.34 3 147.56/147.56 205.64/205.64 156.31/156.31 125.64/133.98 207.52/207.52 154.64/154.64 201.39/201.39 4 147.60/147.60 205.91/205.91 156.16/156/16 125.68/133.97 207.66/207.66 154.55/154.55 201.48/201.48 5 147.56/147.56 205.67/205.67 156.40/156.40 125.58/133.91 207.73/207.74 154.42/154.42 201.53/201.53 6 147.74/147.74 205.68/205.68 156.19/156.19 125.56/133.99 207.47/207.47 154.59/154.59 201.63/201.63 20/20 16/16 10/10 20/24 23.3/23.3 15.2.2/15.2.2 11/11 7 147.61/147.61 205.59/205.59 156.32/156.32 125.67/133.95 207.39/207.39 154.61/154.61 201.64/201.64 20/20 16/16 10/10 20/24 23.3/23.3 15.2.2/15.2.2 11/11 8 147.58/147.58 205.59/205.59 156.32/156.32 125.72/133.97 207.61/207.61 154.59/154.59 201.58/201.58 20/20 16/16 10/10 20/24 23.3/23.3 15.2.2/15.2.2 11/11 a The upper data with slash of each sample are the observed allele size (bp), the lower data is the corresponding genotypes. subspecies or an unknown subspecies has therefore a possibility of inaccuracy unless their population genetic data were added to the database. We found that the DP of each locus varied. Two of them (F146 and Fca453) were lower than 0.7000, due to the small number of alleles. However, the cumulative DP of the seven loci calculated in this reference population exceeded 0.99999789, suggesting that combining the seven loci is able to give a conclusive result sufficient to individualize any tiger from multiple samples. In the case analysis, the genotypes of the eight samples in all loci were identical. The probability that they were from different individuals, e.g. P M, hits 4.52 10 6, being extremely low, strongly supported the validity of the conclusion that they are from one individual. Acknowledgements We would like to thank the donor of tiger samples, Mr. Wei Sen Zhou and the veterinarian Mr. Jian Bin Zhang, Mr. Hong Wen Guo and staff of the Feline Breeding Base of Guangxi Endangered and Rare Wildlife Refuge, China, who contributed time in cooperation with us during the sampling. We thank Dr. Lan Hu and Dr. Song Chen of the Institute of Forensic Sciences, Ministry of the Public Security, PR China who opened their laboratory to us and gave us a lot of valuable advice. We are also very grateful to those who gave us valuable suggestions for the experiments including Ms. Marilyn A. Menotti-Raymond of the Laboratory of Genomic Diversity, National Cancer Institute USA, Prof. Kai Ya Zhou of the Nanjing Normal University and Prof. Xian Hua Jiang of the Forensic Laboratory of the Department of the Public Security of Liaoning Province. Finally we are very grateful to Mr. Christopher Wood in his help in editing this paper. References [1] K. Nowell, Far from a Cure: The Tiger Trade Revisited, TRAFFIC International, Cambridge, UK, 2000. [2] J.H. Wetton, C.S.F. Tsang, C.A. Roney, A.C. Spriggs, An extremely sensitive species-specific ARMS PCR test for the presence of tiger bone DNA, Forensic Sci. Int. (2002) 126. [3] Q.H. Wan, S.G. Fang, Application of species-specific polymerase chain reaction in the forensic identification of tiger species, Forensic Sci. Int. 131 (2003) 75 78. [4] M. Menotti-Raymond, V.A. David, J.C. Stephens, L.A. Lyons, S.J. O Brien, Genetic individualization of domestic cats using feline STR loci for forensic applications, J. Forensic Sci. 42 (1997) 1039 1051. [5] M. Menotti-Raymond, V.A. David, S.J. O Brien, Pet cat hair implicates murder suspect, Nature 386 (1998) 774. [6] S.R. Engel, R.A. Linn, J.E. Taylor, S.K. Davis, Conservation of microsatellite loci across species of Arctiodactyls: implications for population studies, J. Mam. 77 (1999) 504 518. [7] M.A. Menotti-Raymond, S.J. O Brien, Evolutionary conservation of ten microsatellite loci in four species of Felidae, J. Hered. 86 (1995) 319 322. [8] J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular cloning: a laboratory manual, 2nd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1989. [9] M. Menotti-Raymond, V.A. David, L.A. Lyons, A.A. Schäffer, J.F. Tomlin, M.K. Hutton, S.J. O Brien, A genetic linkage map of microsatellites in the domestic cat (Felis catus), Genomics 57 (1999) 9 23. [10] V.A. David, M. Menotti-Raymond, Automated DNA detection with fluorescent-based technologies, IRL Press, Oxford, 1998. [11] P.J. Lincoln, DNA recommendations further report of the DNA Commission of the ISFH regarding the use of short tandem repeat systems, Forensic Sci. Int. 87 (3) (1997) 181 184. [12] W. Bär, B. Brinkmann, B. Budowle, A. Carracedo, P. Gill, P. Lincoln, W. Mayr, B. Olaisen, DNA recommendations. Further report of the DNA Commission of the ISFH regarding the use

Y.C. Xu et al. / Forensic Science International 151 (2005) 45 51 51 of short tandem repeat systems, Int. J. Legal. Med. 110 (1997) 175 176. [13] X-F. Zheng, Forensic DNA Analysis, The People s Public Security University Press, 2002. [14] P. Hoff-Olsen, B. Mevåg, E. Staalstrøm, B. Hovde, T. Egeland, B. Olaisen, Extraction of DNA from decomposed human tissue, an evaluation of five extraction methods for short tandem repeat typing, Forensic Sci. Int. 105 (1999) 171 183.