Conserv Genet (2007) 8:1199–1207 DOI 10.1007/s10592-006-9276-9

ORIGINAL PAPER

Sustainability of the South China tiger: implications of inbreeding depression and introgression Y. C. Xu Æ S. G. Fang Æ Z. K. Li

Received: 9 August 2006 / Accepted: 11 December 2006 / Published online: 5 January 2007
Springer Science+Business Media B.V. 2007

Abstract The South China tiger (Panther tigris amoyensis) is critically endangered with 73 remaining individuals living in captivity, all derived from six wild founders since 1963. The population shows a low level of juvenile survivorship and reproductive difficulties, and faces a huge conservation challenge. In this study, inbreeding depression and genetic diversity decline were examined by using pedigree data and 17 microsatellites. The constant B, which is related to the number of lethal equivalents, was estimated to be 0 for the offspring of noninbred parents, but was >0 for the offspring of inbred parents and for all offspring. Percentage of successfully breeding tigers inversely correlated with inbreeding level (r = -0.626, a = 0.05). Taken together, these findings suggest the population is suffering from inbreeding depression in juvenile survivorship and fecundity. No significant correlation was detectable for the mean litter size with f of either dams (r = –0.305, a = 0.46) or kittens (r = 0.105, Y. C. Xu
S. G. Fang (&) College of Life Sciences, State Conservation Center for Gene Resources of Endangered Wildlife, and the Key Laboratory of Conservation Genetics and Reproductive Biology for Endangered Wild Animals of the Ministry of Education, Zhejiang University, No. 388 Yu Hang Tang Road, Hangzhou 310058, People’s Republic of China e-mail: [email protected] Present Address: Y. C. Xu College of Wildlife Resources, Northeast Forestry University, No. 26 Hexing Road, Harbin 150040, People’s Republic of China Z. K. Li Shanghai Zoological Park, No. 2381 Hongqiao Road, Shanghai 200335, People’s Republic of China

a = 0.71), indicating litter size was not strongly subject to inbreeding depression. The average number of alleles per locus was 4.24 ± 1.03 (SE), but effective number of alleles was only 2.53 ± 0.91. Twenty-one alleles carried by early breeders at 13 loci were absent in the present breeders and potential breeders. Multilocus heterozygosity was inversely correlated with inbreeding levels (r = -0.601, a = 0.004). These findings suggest rapid allelic diversity loss is occurring in this small captive population and that heterozygosity is being lost as it becomes more inbred. Our phylogenetic analysis supports past work indicating introgression from northern Indochinese tigers in the population. As no wild representatives of the South China tiger can be added to the captive population, we may consider the alternate scenario of further introgression in the interest of countering inbreeding depression and declining genetic diversity. Keywords South China tiger
Panthera tigris amoyensis
Inbreeding depression
Genetic diversity
Introgression

Introduction The South China tiger (Panthera tigris amoyensis) is an endemic subspecies that used to be distributed widely across China (Liu and Yuan 1983). Factors including habitat loss, hunting and poaching (Ma and Jin 2004) resulted in the decline in number from an estimated 4,000 in the late 1940s to 150–200 in the early 1980s (Lu and Sheng 1986; Tan 1987). The decline continued into the mid 1990s when the number of free-ranging tigers was estimated to be fewer than 20 individuals (Tilson

123

1200

et al. 1997). The latest field surveys in 2001 and 2002 presented no evidence of viable populations remaining within its historical range (Tilson et al. 2004), though there is speculation that a few tigers may remain in the wild (Wan et al. 2003; Ma and Jin 2004). Nevertheless, fragmentation, habitat destruction, and prey shortages call the sustainability for any such remnant wild populations into question, if they do indeed exist (Tilson et al. 1997, 2004; Ma and Jin 2004). Four females and two males were captured from the wild during the 1950s and 1960s (Fig. 1), and were used to found a captive population of this most endangered subspecies of tiger. Up to November 2004, 373 tigers were recorded as born in captivity; this number does not include 29 dead juveniles not recorded in the studbook (Fang and Yang 1998; Yin 2004). It was not until the early 1990s that inbreeding depression became evident in the lowreproductive rate (about 35.3%) and the high mortality (about 50%) of juveniles (Liu et al. 1997; Tilson et al. 1997; Chen et al. 2001, 2003; Wang et al. 2003). Up to November 2004, the highest inbreeding coefficient was 0.59. The current population has 73 living individuals, including 40 males and 33 females, with an inbreeding coefficient ranging from 0 to 0.50. Only six males and seven females have reproduced within the current captive population (Yin 2004). No wild individuals are likely to be added to the population, and genetic management based solely on careful pairing schemes cannot avoid close inbreeding. The captive population fell into two independent bloodlines, referred to as the Guiyang and Shanghai Lines, before the mid 1980s. The Guiyang Line went

Conserv Genet (2007) 8:1199–1207

extinct after four generations when its last members died in 2002. The Shanghai Line was more successful and became the dominant population. Inbreeding occurred in the early 1970s and continues to the present because only a small number of tigers are able to breed in each generation (Yin 2004). In the mid 1980s, interbreeding occurred between the two lines. One male and one female from the Guiyang Line were mated with tigers from the Shanghai Line and produced a few outbred tigers (Fig. 2). A recent phylogenetic study found evidence of introgression from of the northern Indochinese tiger (P. t. corbetti) into the captive population (Luo et al. 2004). Five individuals from the captive population appeared to represent two phylogenetic lineages. One lineage [including two individuals (Pti219, studbook No. #129 and Pti220, #141) from Chongqing Zoo representing the Guiyang Line] was the lineage of the South China tiger. The other lineage [including 1 individual (Pti217, #242) from Chongqing Zoo, and two individuals (Pti218, #256 and Pti222, #258) from Suzhou Zoo representing the Shanghai Line] was indistinguishable from P. t. corbetti. If a more extensive sampling of the population finds further evidence of introgression by P. t. Corbetti, this knowledge needs to be incorporated into the conservation strategy for the South China tiger. The small number of founders, inbreeding depression, and potential introgression all contribute to the challenge of conserving the South China tiger. In order to fully understand the genetic status of the population and create effective conservation measures, juvenile

Fig. 1 Geographic origin of six founders of the captive South China tiger population

#12

#8

#26

#3

#6

#7

123

Conserv Genet (2007) 8:1199–1207 Fig. 2 Pedigree of the captive South China tigers. Squares and circles stand for males and females, respectively; broken bordered squares stand for individuals of unknown sex; numbers in squares and circles in the top block stand for the founders, and numbers in lower blocks stand for the studbook number of individuals subjected to microsatellite analysis. Individuals within region A belong to the Shanghai Line, within region B belong to the Guiyang Line, and within region C are hybrids of the two bloodlines

1201

26

211

3

12

8

6

7

228

111

236

238

237

208

229 172 141

B 280

157

227

215

235

226

C

241

251

226 250

252

A

survivorship and reproduction were studied throughout the history of the captive population based on pedigree data and the records of zoos. Microsatellite loci were used to quantify loss of genetic variation and assess the degree to which introgression has influenced the population.

Materials and methods Inbreeding depression Original breeding records were collected for all the captive South China tigers from zoos where they are bred. These data were combined with the records of the studbook (Yin 2004), including information on birth, reproduction, longevity, causes of death. Following Kalinowski and Hedrick (1998), juveniles were sorted into two categories: births with noninbred parents and births with at least one inbred parents (Table 1). We defined Bnoninbred as the half number of lethal equivalents in offspring of noninbred parents and Binbred as the half number of equivalents in the offspring of inbred parents. If purging of deleterious recessives had occurred due to past inbreeding Bnoninbred > Binbred. In order to isolate the effects of husbandry on juvenile survivorship, juveniles were also sorted on year of birth and B was estimated for 1963– 1980 (B63–80) when husbandry was considered at low

level and 1980 to present (B81–present) when husbandry was improved steadily (Table 1). The causes of juvenile mortality (defined as juvenile death within 3 months after birth, i.e., from birth to weaning) and reproduction failure were categorized. Inbreeding coefficients (f) of all individuals were calculated using software SPARKS Version 1.4 based on the pedigree data. Inbreeding coefficients for all potential offspring were calculated using software PM2000 based on all possible pairing combinations among current and potential breeders. Data of juvenile mortality was used to estimate the numbers of lethal equivalents using the model proposed by Morton et al. (1956): Si ¼ S0 eBf i ; where Si is the probability of an individual with an inbreeding coefficient of fi surviving to 3 months, S0 is the survivorship of noninbred individuals, and B is a constant describing the rate of decline of survivorship with inbreeding for the population. The number of lethal equivalents in a population is approximately 2 · B (Kalinowski and Hedrick 1998). S0 and B were estimated using the maximum likelihood method as proposed by Kalinowski and Hedrick (1998), Kalinowski et al. (2000). Percentage of breeding tigers (PBT) was calculated at different inbreeding levels for all males and females

123

1202

Conserv Genet (2007) 8:1199–1207

Table 1 Demographic data of the captive South China tiger F

0.0000 0.1250 0.1875 0.2187 0.2265 0.2500 0.2656 0.3203 0.3281 0.3437 0.3593 0.3750 0.4140 0.4218 0.4921 0.5898

Offspring of noninbred parents

Offspring of inbred parents

1963–1980

1981–present

Mean litter size (LS)

Born

Survived

Born

Survived

Born

Survived

Born

Survived

By dam’s f

By kitten’s f

48 30 12 – – 30 – – 12 – – – – – – –

27 13 11 – – 25 – – 9 – – – – – – –

19 4 20 7 16 30 2 3 – 2 32 45 6 4 14 2

12 0 8 7 7 18 2 3 – 0 1 30 2 1 4 0

69 23 4 – – 28 – – – – – 3 – – – –

35 12 0 – – 23 – – – – – 3 – – – –

22 7 28 7 16 27 2 3 12 2 32 42 6 4 14 2

13 1 19 7 7 20 2 3 9 0 1 27 2 1 2 0

2.37 2.20 2.67 3.50 – 2.28 1.50 – 2.00 – – 1.91 – – – –

2.35 2.58 2.15 2.33 3.00 2.40 2.29 1.50 – 2.00 2.27 3.46 3.00 4.00 2.33 2.00

that had been paired. Litter size (LS) defined as mean number of kittens per litter was calculated for all female breeders. The relationship between f and PBT and LS were tested using the least square linear regression module of SYSTAT 11 (Systat Software Inc., Richmond, CA, USA). Two-tailed tests were performed to evaluate statistical significance. Microsatellite analysis Samples and DNA isolation Twenty skin samples were donated by the Chinese Association of Zoological Gardens South China Tiger Genome Resource Bank. These included ten breeding and ten nonbreeding individuals (five were potential breeders) covering seven inbreeding levels from 0 to 0.42. The South China Tiger Genome Resource Bank was established in 1996 under the guidelines as proposed by Wildt et al. (1995). Tissues were cryopreserved and stored in liquid nitrogen until they were subjected to DNA isolation. In addition to the 20 skin samples, a muscle sample of a cryopreserved neonate was donated by Suzhou Zoo in 2003. Information from sampled individuals is shown in Fig. 2. Whole genomic DNAs were isolated with the phenol:chloroform method (Sambrook and Russell 2001). Genotyping and statistics Primer pairs of 17 microsatellite loci as shown in Table 2 (Menotti-Raymond et al. 1999), derived from

123

± ± ± ±

1.01 0.63 1.11 0.69

± 0.96 ± 0.71 ± 0.00

± 1.04

± ± ± ± ± ± ± ±

1.20 0.90 0.80 1.53 1.26 0.89 0.91 0.71

± ± ± ± ± ± ±

0.00 1.01 0.66 1.41 0.00 1.21 0.00

Percentage of breeding tigers (PBT)

82.35 100.00 60.00 75.00 33.33 38.46 50.00 – 50.00 – 0.00 40.00 – – 50.00 –

domestic cat (Felis catus), were selected to amplify tiger microsatellites. All loci were polymorphic, with 4–11 alleles, and have been used in previous studies of the tiger (Xu 2001; Xu et al. 2005). The PCR amplification conditions and genotype determination followed Xu et al. (2005). The number of alleles per locus (A), observed heterozygosity (Ho), expected heterozygosity (He) and number of effective alleles (ne) were calculated following Nei and Roychoudhury (1974) and Frankham et al. (2002). Tests for deviations from Hardy–Weinberg equilibrium for each locus were performed using GENEPOP web Version of 3.4 (http:// www.wbiomed.curtin.edu.au/genepop/). The individual multilocus heterozygosity (MLH) across all genotyped loci was calculated following Coulson et al. (1998). Correlation between individual MLH and f and its significance was test with the same method as mentioned above. Individual phylogenetic trees were constructed with the software Populations 1.2.28 (Free Software Foundation Inc. 2002) using Nei’s minimum genetic distance and Neighbor joining methods.

Results Inbreeding depression From the birth of the first litter in captivity in 1963 up to Nov 2004, 171 kittens (46% of 375) died within 3 months after birth. Juvenile mortality fluctuated markedly throughout the history. Expertise and husbandry were at a lower level during the 1960s and

Conserv Genet (2007) 8:1199–1207

1203

Table 2 Characteristics of the 17 microsatellites in the South China tiger population (n = 21) Locus

F164 F37 F41 F53 Fca043 Fca066 Fca077 Fca090 Fca107 F146 Fca221 Fca304 Fca391 Fca441 Fca453 F42 Fca074 Means SE

Allele frequency 1

2

3

4

0.05 0.14 0.05 0.02 0.19 0.26 0.95 0.26 0.38 0.60 0.86 0.14 0.38 0.07 0.19 0.07 0.02 –

0.86 0.14 0.05 0.07 0.43 0.10 0.02 0.40 0.48 0.12 0.02 0.17 0.36 0.86 0.05 0.07 0.02 –

0.05 0.40 0.07 0.10 0.07 0.64 0.02 0.33 0.14 0.26 0.12 0.64 0.17 0.02 0.31 0.12 0.40 –

0.05 0.31 0.64 0.48 0.02

5

Ho

He

ne

PHWE

4 4 5 6 5 3 3 3 3 4 3 4 4 5 5 6 5 4.24 1.03

0.00 0.81 0.29 0.76 0.52 0.53 0.10 0.67 0.53 0.52 0.24 0.43 0.74 0.19 0.33 0.81 0.58 0.47 0.25

0.26 0.72 0.55 0.72 0.71 0.52 0.09 0.67 0.62 0.58 0.26 0.55 0.43 0.26 0.72 0.78 0.69 0.54 0.21

1.35 3.33 2.18 3.37 3.25 2.04 1.10 2.91 2.55 2.29 1.33 2.16 3.23 1.35 3.33 4.24 3.04 2.53 0.91

0.0000* 0.0027* 0.0033* 0.0651 0.1160 0.7204 1.0000 0.0922 0.4622 0.7094 0.1302 0.0351* 0.0038* 0.0324* 0.0000* 0.0539 0.2911 – –

6

0.19 0.17 0.29

0.17

0.02 0.05 0.14 0.36 –

0.21

0.02 0.05 0.10 0.02 0.40 0.38 0.19 –

A

–

A is the number of alleles per locus; Ho is observed heterozygosity; He is expected heterozygosity; ne is number of effective alleles; HWE stands for the Hardy–Weinberg equilibrium Asterisk stands for significant departure from the Hardy–Weinberg equilibrium (PHWE<0.05)

1970s when captive breeding of tiger was just started. Most zoos had no experience in raising tiger and were short of financial, infrastructural and medical support. Husbandry improved steadily after the 1980s with the accumulation of experience and upgrades of infrastructure. Inbreeding first occurred in 1972 and significantly increased over time. Up to November 2004, the highest inbreeding coefficient reached 0.59. In the current population, according to the studbook data, all possible pairing combinations will result in further elevation of inbreeding coefficient, and 58% of pairing options may generate offspring with inbreeding coefficients greater than 0.25. Considering the difficulties of transportation and the health of candidate breeders, only a few possible pairings are practical. Thus, the rate of increase of inbreeding will occur faster than might be expected given the number of individuals in the population. For the 166 dead juveniles for which we have complete information, eight categories of lethal factors were identified: cold shock, accidental injury, stillbirth, fetal malformation, mother rejection, inadequate lactation, disease, and general weakness. Accidental injury is not a genetic linked factor, so eight kittens that died of injury were removed from further analyses of inbreeding depression. In order to analyze juvenile mortality, we estimated Bnoninbred = 0 [95% CI = 0–0.73] and Binbred = 1.16 [0.47–2.01]. Estimates

of S0noninbred was 0.64 [0.52–0.76] and S0inbred was 0.64 [0.50–0.77]. When estimated for the combined sample of individuals, B = 0.62 [0.09–1.28] and S0 = 0.61 [0.52– 0.70]. Furthermore, we estimated that B63–80 = 0 [0–0.97], and S0 63–80 = 0.57 [0.44–0.70]; more recently B81–present = 1.20 [0.58–1.98], and S0 81–present = 0.70 [0.56–0.82]. A total of 141 tigers lived to reach reproductive maturity at 4 years of age. Reproductive data were available for 134 of these tigers. Eighty-four tigers (43 males and 41 females) had been involved in pairing and 52 (24 males and 28 females) either produced offspring or experienced pregnancy and abortion. The total PBT was 61.9%. PBT at each class of inbreeding coefficient (Table 1) was significantly and inversely correlated with level of inbreeding (r = -0.626, a = 0.05). Reasons for mating failure fell into four categories: (1) lack of libido (n = 7) that males manifested little sexual desire when they were paired; (2) low-grade estrus (n = 7) that females manifested slight estrus and inadequate sexual desire; (3) barren of mating (n = 9) that mating did not result in pregnancy (we were unable to investigate exact causes); and (4) abnormal behavior (n = 11) where either the male or female escaped from or attacked its mate during pairing, or they performed improperly during copulation. All males that lacked libido had poor sperm quality inferred from previous examinations (Byers et al. 1996; Liu et al. 1998).

123

1204

Mean litter size for 28 female breeders was calculated based on a total of 131 litters (Table 1). Mean litter size did not correlate significantly with f of either dams (r = –0.305, a = 0.46) or themselves (r = 0.105, a = 0.71). This implies litter size is not strongly subject to inbreeding depression in the South China tiger. Microsatellite analysis Characteristics of the 17 microsatellites in the population are shown in Table 2. Ten loci showed agreement with Hardy–Weinberg equilibrium and seven deviated significantly from it (a = 0.05). Our samples included ten living breeder and potential breeders in the current population. Compared to the data above, some alleles were absent in the ten individuals at 13 loci: allele 1 and 3 at Fca164, allele 1 and 2 at F41, allele 1 at F53, allele 4 at Fca043, allele 2 and 3 at Fca077, allele 4 at F146, allele 2 at Fca221, allele 4 at Fca304, allele 4 at Fca391, allele 1, 3, 4, 5 at Fca441, allele 2, 5, 6 at Fca453, allele 2 at F42, allele 2 at Fca074. These absent alleles were mainly carried by earlier breeders that have died: #111, #141, #157, and #172. Inferred from the pedigree, the only group that might carry these absent alleles is the set of hybrid individuals of the two blood lines (region C of Fig. 2). Genotyped individuals had inbreeding coefficients between 0 and 0.42, and MLH between 0.29 and 0.71; these quantities were negatively correlated (r = –0.601, a = 0.004). Seven alleles unique to the Guiyang Line were identified at five loci: two at F42, one at Fca391, one at F146, one at Fca441 and two at F41. The neighbor joining tree of individuals revealed significant genetic differentiation between the Guiyang Line represented by tiger 141 and tiger 172 and the Shanghai Line represented by other individuals (Fig. 3). Tiger 280 is a hybrid of the two lines (Fig. 1) and was positioned between the clusters representing the two lines.

Discussion Inbreeding depression The occurrence of inbreeding depression is supported by numerous breeding experiments showing a decline in fitness-associated traits in individuals with highinbreeding coefficients (Charlesworth and Charlesworth 1987; Frankham et al. 2002). However, it has been suggested that mammalian carnivores suffer no or minor deleterious effects from inbreeding due to the

123

Conserv Genet (2007) 8:1199–1207

elimination of genetic load by regular inbreeding (as reviewed by Laikre and Ryman 1991). Ralls et al. (1988) observed a Sumatran tiger population founded by a mix of wild-caught tigers and captive-born tigers had a lethal equivalent as small as 0.02 per individual. Kalinowski and Hedrick (1998) re-estimated lethal equivalent of three leopard subspecies, Amur leopard (Panthera pardus orientalis), Ceylon leopard (P. p. kotiya) and Chinese leopard (P. p. japonensis) using maximum likelihood method. Their Bs ranged from 0 to 0.09. The level of Bs in Sumatran tiger and leopard were nearly negligible compared to 2.75 in Speke’s gazelle (Kalinowski and Hedrick 1998) and 4.66, an average level of 40 mammalian species (Ralls et al. 1988). However, B was 0.62 in the South China tiger population, much higher than that in the Sumatran tiger and leopard but lower than at of other species. These results agree with Laikre et al. (1996) that mammalian carnivores do suffer deleterious effects on viability from inbreeding, but the magnitude of inbreeding depression may be small relative to other mammals (Ralls et al. 1988). Inbreeding in the captive population of South China tiger started in the early 1970s. B was 0 during 1963– 1980, indicating that the effect of inbreeding depression on the viability of kittens was small relative to other sources of mortality. In contrast, B was 1.20 after 1981 suggesting a negative effect of inbreeding on viability of kittens. We suspect that inconsistencies in the quality of husbandry prior to 1981 obscured the effects of inbreeding depression on viability. It is also possible that the greater range of levels of inbreeding in the population after 1980 (Table 1), enhanced our ability to detect inbreeding depression. Previous reports demonstrated the susceptibility to inbreeding varied with species, subspecies and population, although the variation partly resulted from heterogeneous data used for the analysis (Ralls et al. 1988; Kalinowski and Hedrick 1998; Laikre and Ryman 1991). The difference of B between South China tiger and Sumatran tiger (Ralls et al. 1988) presents another example of inter-subspecies variation of inbreeding depression. Interestingly, our data showed B of offspring of nonibred parents was 0, but was 1.16 for the offspring of inbred parents, in contrast with the data of captive Speke’s gazelle, whose survivorship of inbred animals that had inbred parents (Bnoninbred = 2.60,) was higher than that of inbred animals with no history of inbreeding (Binbred = 0.57) (Kalinowski and Hedrick 1998). The survivorship for the two categories of captive bighorn sheep (Ovis canadensis) was nearly identical (0.71 vs. 0.72), indicating that survivorship in this sample was not influenced by previous inbreeding

Conserv Genet (2007) 8:1199–1207

1205

Fig. 3 Individual phylogenetic tree inferred from 17 microsatellites showing distinction between Guiyang Line and Shanghai Line

#141

Guiyang Line

#172

#280

Hybrid #215 #236

#237

#250 #226

#229

#252

#157 #235 #216 #211 #111

#208

#241

#227

Shanghai Line #251

#238 #228

0.1

(Kalinowski and Hedrick 2001). The case of captive Speke’s gazelle was explained as genetic load purging (Kalinowski and Hedrick 1998). However, the South China tiger, as well as the bighorn sheep, demonstrated no purging. The possible reason is that reduction of survivorship of these species might be resulted from different genetic mechanism, videlicet, genes involved in reduction of survivorship in Speke’s gazelle are severely deleterious or lethal and are subjected to purging process, while genes involved in reduction of survivorship in South China tiger and bighorn sheep are mildly deleterious and thus not purged (As reviewed in Keller and Waller 2002). Low fertility and fecundity is also an indicator of inbreeding depression in feline species, such as the cheetah (Acinonyx jubatus), and Florida panther (Felis concolor coryi), both of which had poor sperm quality (Lindburg et al. 1993; Roelke et al. 1993; Barone et al. 1994). Our investigation showed PBT was significantly inversely correlated with level of inbreeding, suggesting inbreeding reduced capability of reproduction in this tiger subspecies. All males that lacked libido had poor sperm quality inferred from previous examinations (Byers et al. 1996; Liu et al. 1998). They were inferior in either one or all sperm properties, such as sperm density, viability and frequency of abnormal sperm. The occurrence of these problems is not sufficient to exclude the potential influence of husbandry on fertility. However, a comparison of fertility in South China tiger and Siberian tiger, which were both raised in all zoos that we investigated, suggests husbandry was

not a major factor associated with reproductive difficulties. The husbandry of the South China tiger was the same as or relatively better than that of Siberian tiger, a captive population that had multiple origins and large number of founders. The Siberian tiger did not manifest the reproductive difficulties observed in the South China tiger, in spite of intensive efforts to breed the endangered subspecies at both the Shanghai Zoological Garden and Suzhou Zoological Garden. Therefore, we believe husbandry was probably not a key factor influencing the fertility of tigers. We observed no evidence that litter size was associated with the inbreeding coefficient of either kittens or their dams; in at least one carnivore, the brown bear (Ursus arctos) litter sized is affected by inbreeding (Laikre et al. 1996). Genetic diversity decline The genetic effect of inbreeding is the incremental loss of heterozygosity; therefore, inbred populations, especially those suffering inbreeding depression, often have low levels of genetic variation. The DNA fingerprint heterozygosity of the Florida panther, that suffered severe inbreeding depression, was only 0.104, while the western subspecies has an estimated heterozgyosity of 0.469 (Frankham et al. 2002). The mean heterozygosity of 17 microsatellites was 0.54 for the captive South China tiger. In contrast, the mean heterozygosity was 0.701 in our previous studied captive Siberian tiger population using 15 of 17 microsatellites (Xu 2001). The Siberian tiger population consisted of 383

123

1206

individuals derived from 20 unrelated founders and showed a relatively high-survival rate and fecundity. Alleles present at low frequencies have a low probability of being present in the next generation, when only a small proportion of available individuals form the breeding population. We observed 21 alleles present in past breeding individuals that were absent in the currently living breeders and potential breeders. Alleles will continue to be lost if breeding is limited to only a few individuals and inbreeding continues. MLH of individual is an indicator that reflects inbreeding consequence on genome wide heterozygosity. Our investigation revealed a declining trend of MLH with elevation of inbreeding level. In the current population, 100% of possible pairing options will generate inbred offspring, and 58% may generate offspring with inbreeding coefficients greater than 0.25 (Table 2). However, these estimates assure that all potentially breeding individuals will breed. Actually, at least 50% of the pairings would not be practical because of behavioral problems, unfamiliarity, infecundity, disharmony between female estrus and male availability, and transportation problems around zoos. This suggests that the South China tiger will continue to lose alleles and become more inbred if current management practices continue. Introgression and implications for conservation Introducing wild individuals into this population is the only way to slow or stop the decline of genetic diversity and counter inbreeding depression. No wild individual has been able to be added to the population in the past 40 years. However, a recent phylogenetic analysis of tiger subspecies discovered that two individuals representing the Shanghai Line were indistinguishable from northern Indochinese tigers (Luo et al. 2004). Our study included #141 (Guiyang Line) and full sib of #242, #256, and #258 (Shanghai Line) used in Luo’s analysis. The phylogenetic tree inferred from 17 microsatellites confirmed significant differentiation between the Guiyang Line (#141 and #172) and the Shanghai Line. We also identified seven alleles unique to the Guiyang Line at five microsatellite loci. Five (#3, #6, #7, #8, #12) of the six founders were captured from the wild of Guizhou Province and one (#26) was from Fujian Province (Fang and Yang 1998, Fig. 1). Female #26, made a large contribution to the Shanghai Line. The South China tiger was formerly distributed in Guizhou, Guangxi, and Guangdong provinces (Tan 1996). There is no geographic barrier

123

Conserv Genet (2007) 8:1199–1207

between the distributions of the northern Indochinese tiger and South China tiger; instead, Fujian area might be a transitional zone shared by two subspecies (Kitchener and Dugmore 2000). As proposed by Luo et al., the ambiguity of geographic distribution of the South China tiger and mainland Indochinese tiger likely made the subspecies origin of the founders indecipherable. Our study supports the suggestion of Luo et al. (2004) that the present South China tiger population is a mixture of two different subspecies. Given that introgression has already occurred in the Shanghai line and that there may not be enough of South China tigers without Indochinese genes to sustain the linage, future inbreeding and loss of alleles might be reduced by allowing further limited introgression. The introgression of Florida panther from the Texas subspecies achieved hybrid kittens has resulted in nearly a three times greater chance of becoming adults as do purebred ones, and hybrid adult females survive better than purebred females (Pimm et al. 2006). The success of rescuing Florida panther by introgression provides a positive example for the management of South China tiger. Acknowledgments This work was supported by grants from the State Key Basic Research and Development Plan of China (G2000046906), the National Natural Science Foundation of China (30300039) and the Key Scientific Research Project of the Ministry of Education of China (02094). Special thanks go to two anonymous reviewers. We are very grateful to Yu Zhong Yin, Gong Qing Huang, Wen Yang Gu, Ai Shan Wang, Wen Li Guo, Qing Yong Shen, and Jian Rong Wang for their help in collecting demographic data. We thank Paul L. Leberg, Barry K. Hartup and Mark Morgan for editing the manuscript.

Reference Barone MA, Roelke ME, Howard J, Brown JL, Anderson AE, Wildt DE (1994) Reproductive characteristics of male Florida panthers: comparative studies from Florida, Texas, Colorado, Latin America and North American zoos. J Mammal 75:150–162 Byers A, Armstrong D, Miller E, Traylor-Holzer K (1996) Reproductive evaluation and genome resource banking. In: Tilson R, Armstrong D, Miller E, Byers A, TraylorHolzer K, Brady G, Wang H, Xie Z (eds) Medical, reproductive and management evaluation of South China tiger in China. Minnesota Zoo, Apple Valley, MN, USA, pp 47–50 Charlesworth D, Charlesworth B (1987) Inbreeding depression and its evolutionary consequences. Annu Rev Ecol Syst 18:237–268 Chen GL, Li ZK, Jin K, Shen QY (2003) The impact of inbreeding on the South China tiger. Zool Gard 9: 315–317

Conserv Genet (2007) 8:1199–1207 Chen GL, Li ZK, Song PL, Jin K, Sheng QY (2001) The current situation and the pedigree analysis of South China tiger. Chin J Zool 36:45–48 Coulson TN, Pemberton JM, Albon SD, Guinness FE, CluttonBrock TH (1998) Microsatellites measure inbreeding depression and heterosis in red deer. Proc R Soc Lond B 265:489–495 Fang X, Yang JL (1998) The South China tiger in Guizhou. In: Shi XQ, Cui LP (eds) South China tiger, the raising, breeding, management, disease provention and conservation. Chinese Association of Zoological Gardens South China Tiger Conservation and Coordination Committee, Shanghai, China, pp 20–23 Frankham R, Ballou JD, Briscoe DA (2002) Introduction to conservation genetics. Cambridge University Press, Cambridge, UK Kalinowski ST, Hedrick PW (1998) An improved method for estimating inbreeding depression in pedigrees. Zoo Biol 17:481–497 Kalinowski ST, Hedrick PW (2001) Inbreeding depression in captive bighorn sheep. Anim Conserv 4:319–324 Kalinowski ST, Hedrick PW, Miller PS (2000) Inbreeding depression in the Speke’s gazelle captive breeding program. Conserv Biol 14:1375–1384 Keller LF, Waller DM (2002) Inbreeding effects in wild populations. Trends Ecol Evol 17:230–241 Kitchener AC, Dugmore AJ (2000) Biogeographical change in the tiger, Panthera tigris. Anim Conserv 3:113–124 Laikre L, Andre´n R, Larsson H-O, Ryman N (1996) Inbreeding depression in brown bear. Biol Conserv 76:69–72 Laikre L, Ryman N (1991) Inbreeding depression in a captive wolf (Canis lupus) population. Conserv Biol 5:33–40 Lindburg DG, Durrant BS, Millard SE, Oosterhuis JE (1993) Fertility assessment of cheetah males with poor quality semen. Zoo Biol 12:97–103 Liu J, Li YH, Shen QY (1997) The preservation of the South China tiger in China. Chin Wildl 3:24–26 Liu J, Zhao GL, Wu DH, Bai ZM, Zhang TX, Sheng QY, Hu Y, Zheng XC, Xie YX, Deng Y, Wang AP, Luo L, Chen YX, Fu MQ, Chen GZ, Chen SJ (1998) Development of techniques for sperm collection in the South China tiger. In: Shi XQ, Cui LP (eds) South China Tiger, the raising, breeding, management, disease prevention and conservation. Chinese Association of Zoological Gardens South China Tiger Conservation and Coordination Committee, Shanghai, China, pp 146–150 Liu ZH, Yuan YC (1983) The resources of the South China tiger in China. Chin Wildl 4:20–22 Lu HJ, Sheng HL (1986) Distribution and status of the Chinese tiger. In: Miller SD, Everett DD (eds) Cats of the world: biology, conservation and management. National Wildlife Federation, Washington DC, USA, pp 51–58 Luo S-J, Kim J-H, Johnson WE, van der Walt J, Martenson J, Yuhki N, Miquelle DG, Uphyrkina O, Goodrich JM, Quigley HB, Tilson R, Brady G, Martelli P, Subramaniam V, McDougal C, Hean S, Huang S-Q, Pan W-S, Karanth UK, Sunquist M, Smith JLD, O’Brien SJ (2004) Phylogeography and genetic ancestry of tigers (Panthera tigris). PLoS Biol 2:2275–2293

1207 Ma JZ, Jin K (2004) Tiger research. Shanghai Scientific and Technical Publishers, Shanghai, China Menotti-Raymond M, David VA, Lyons LA, Scha¨ffer AA, Tomlin JF, Hutton MK, O’Brien SJ (1999) A genetic linkage map of microsatellites in the domestic cat (Felis catus). Genomics 57:9–23 Morton NE, Crow JF, Muller HJ (1956) An estimate of the mutational damage in man from data on consanguineous marriages. Proc Natl Acad Sci USA 42:855–863 Nei M, Roychoudhury AK (1974) Sampling variances of heterozygosity and genetic distance. Genetics 76:379–390 Pimm SL, Dollar L, Bass OL (2006) The genetic rescue of the Florida panther. Anim Conserv 9(2):115–122 Ralls K, Ballou JD, Templeton A (1988) Estimates of lethal equivalents and the cost of inbreeding in mammals. Conserv Biol 2:185–193 Roelke ME, Martenson J, O’Brien SJ (1993) The consequences of demographic reduction and genetic depletion in the endangered Florida panther. Curr Biol 3:340–350 Sambrook J, Russell DW (2001) Molecular cloning: a laboratory manual 3. Cold Spring Harbor Laboratory Press, NY, USA Tan BJ (1987) Status and problems of captive tigers in China. In: Tilson R, Seal US (eds) Tigers of the world, the biology, biopolitics, management, and conservation of an endangered species. Noyes Publications, Park Ridge, NJ, USA, pp 134–148 Tan BJ (1996) Distribution of the tiger in China. In: Shi XQ, Cui LP (eds) South China tiger, the raising, breeding, management, disease provention and conservation. Chinese Association of Zoological Gardens South China Tiger Conservation and Coordination Committee, Shanghai, China, pp 1–5 Tilson R, Hu DF, Muntifering J, Nyhus PJ (2004) Dramatic decline of wild South China tigers Panthera tigris amoyensis: field survey of priority tiger reserves. Oryx 38:40–47 Tilson R, Traylor-Holzer K, Qiu MJ (1997) The decline and impending extinction of the South China tiger. Oryx 31:243–252 Wan QH, Fang SG, Chen GF, Wang ZM, Ding P, Zhu MY, Chen KS, Yu JH, Zhao YP (2003) Use of oligonucleotide fingerprinting and faecal DNA in identifying the distribution of the Chinese tiger (Panthera tigris amoyensis Hilzheimer). Biodivers Conserv 12:1641–1648 Wang W, Shen QY, Yin YZ (2003) Demographic studies on captive South China tigers. Acta Theriol Sin 23:6–9 Wildt D, Byers A, Johnston L, Howard J, Willis K, O’Brien S, Tilson R, Rall W, Seal U (1995) Tiger genome resource banking (GRB) action plan: global need and a plan for North America and Indonesia. IUCN/SSC conservation breeding specialist group and the American zoo and aquarium association, Apple Valley, MN, USA Xu YC (2001) Tiger (Panthera tigris) microsatellite polymorphisms and their application to the management of captive population. Ph.D thesis, Northeast Forestry University, Harbin, China Xu YC, Li B, Li WS, Bai SY, Jin Y, Li XP, Gu MB, Jing SY, Zhang W (2005) Individualization of tiger by using microsatellites. Forensic Sci Int 151:45–51 Yin YZ (2004) Studbook of the South China tiger. Chongqing Zoological Garden, China

123