Int J Primatol (2007) 28:791–799 DOI 10.1007/s10764-007-9157-1 Mitochondrial Control Region and Population Genetic Patterns of Nycticebus bengalensis and N pygmaeus Deng Pan & Jing-Hua Chen & Colin Groves & Ying-Xiang Wang & Etsuo Narushima & Helena Fitch-Snyder & Paul Crow & Xiangyu Jinggong & Vu Ngoc Thanh & Oliver Ryder & Leona Chemnick & Hong-Wei Zhang & Yun-Xin Fu & Ya-Ping Zhang Received: 27 October 2005 / Accepted: 22 May 2006 / Published online: August 2007 # Springer Science + Business Media, LLC 2007 Abstract Bengal slow lorises (Nycticebus bengalensis) and pygmy slow lorises (Nycticebus pygmaeus) are nocturnal which creates difficulties to study them in the field There is a scarcity of data on them and their population genetics are poorly understood We sequenced and analyzed a partial fragment in the first hypervariable region of the mitochondrial control region or D-loop HVRI of 21 Nycticebus bengalensis and 119 N pygmaeus from the boundary between China and Vietnam where they are sympatric Though the sample size for Nycticebus pygmaeus is much larger, the polymorphism level is much lower than that of N bengalensis, possibly Deng Pan and Jing-Hua Chen made equal contributions to the work D Pan : J.-H Chen : X Jinggong : Y.-P Zhang (*) Laboratory of Cellular and Molecular Evolution, Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming 650223, China e-mail: zhangyp@mail.kiz.ac.cn D Pan : J.-H Chen : H.-W Zhang : Y.-X Fu : Y.-P Zhang Laboratory for Conservation and Utilization of Bio-resources, Yunnan University, Kunming, China D Pan The Graduate School, Chinese Academy of Sciences, Beijing 100039, China J.-H Chen School of Life Science, Shandong University, Jinan 250000, China C Groves School of Archaeology & Anthropology, Australian National University, Canberra, Australia Y.-X Wang Department of Phylogensis and Evolution, Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming 650223, China 792 D Pan, et al because of (1) external gene flow from other habitats of N bengalensis, (2) gene ingression from Sunda slow lorises (N coucang coucang) to N bengalensis, (3) a skewed birth sex ratio in N pygmaeus, and (4) a possible low survival rate of infant N pygmaeus Based on genetic comparisons to Nycticebus bengalensis, we propose that N pygmaeus in southern China and northern Vietnam might have migrated from middle or southern Vietnam recently Keywords Bengal slow loris pygmy slow loris population genetics mitochondrial control region Introduction Bengal slow lorises (Nycticebus bengalensis) and pygmy slow lorises (N pygmaeus) are of recognized nocturnal strepsirrhines in Nycticebus (Oxnard et al 1990; Tan and Drake 2001) Because of recent speciation events in Nycticebus (Lu et al 2001), and because distinct species occupy overlapping niches, taxa of Nycticebus share many similar features, including behavior, diet, and morphology (Groves 2001) However, Nycticebis bengalensis and N pygmaeus differ in main aspects: 1) range: the areas where N pygmaeus live are a subset of the regions where N bengalensis live In addition to some sympatric areas, notably southeast of China, Vietnam, and Laos, N bengalensis also lives in India, Bangladesh, and Thailand (Groves 2001) 2) Appearance: N bengalensis is large-bodied, with shorter ears and strong frosting neck and forelimbs, whereas the smaller-bodied N pygmaeus has longer ears and a coat with dark brown pelage (Groves 2001), so they are visually distinct 3) Reproduction: Twinning is common in N pygmaeus (50% of births, in Fitch-Snyder’s 2000 study; 100% in 27 litters as described by Feng et al unpublished data) but rare (3%) in N bengalensis (Fitch-Snyder 2000) In addition, pygmy slow lorises are seasonal breeders while the Bengal lorises are not (Fitch-Snyder 2000) Because of the difficulties in conducting field studies and the previous scarcity of samples, the population patterns of the species are still poorly understood and genetic studies were rare Our previous study (Chen et al 2005) on the molecular E Narushima Ueno Zoological Gardens, Taito-ku, Tokyo, Japan H Fitch-Snyder Loris Conservation International, Hanoi, Vietnam P Crow Kadoorie Farm & Botanic Garden Corporation, Hongkong, China V N Thanh Department of Vertebrate Zoology, Vietnam National University, Hanoi, Vietnam H Fitch-Snyder : O Ryder : L Chemnick Zoological Society of San Diego, San Diego, CA 92112, USA Y.-X Fu Human Genetics Center, University of Texas at Houston, Houston, TX 77030, USA Population Genetic Patterns of Slow Lorises 793 phylogeny of Nycticebus showed that the genetic distance between N bengalensis and N pygmaeus is the farthest among all the taxa in Nycticebus, and some N bengalensis mix with Sunda slow lorises (N coucang coucang) After years of collecting, we have acquired samples from 21 Nycticebus bengalensis and 119 N pygmaeus from the boundary areas between China and Vietnam, where they are sympatric The availability of so many samples provided us with the unique opportunity to study and report on their population genetic patterns We used the first hypervariable region (HVRI) of the mitochondrial control region or D-loop as the marker because HVRI in animals evolves very quickly (Avise 1991), and researchers have used it widely in population genetic studies In addition to the unknown genetic patterns, another intriguing question is how the lorises, which share overlapping niches, compete with each other in the sympatric area Because our samples were collected from this area, it should be possible for us to touch on some points regarding sympatry However, to explore the question fully requires extensive additional knowledge from both field and genetic studies; hence, with limited information, our preliminary conclusions are understandably speculative Materials and Methods Samples, DNA Extraction, Polymerase Chain Reaction (PCR), and Sequencing We collected samples from 119 Nycticebus pygmaeus and 21 N bengalensis in the border areas between China and Vietnam Within the sampled group, Chen et al (2005) had studied Nycticebus pygmaeus (P1-4) and N bengalensis (B1-9) previously We isolated total genomic DNA from hair, blood, or tissue, and the procedures for DNA extraction, PCR amplification, and sequencing of the D-loop region are per Chen et al (2005).We sequenced both strands for accuracy Data Analysis We manually checked DNA sequences for accuracy and aligned them via MegaAlign in DNAStar (DNAStar Inc.) We determined the number of variable sites via Mega2 version 2.1 (Kumar et al 2001) We constructed median-joining networks to show the relationship of haplotypes (Bandelt et al 1995) We obtained the percentage of nucleotide diversity (Nei 1987) via DNASP version 4.0 (Rozas et al 2003) We estimated population parameters such as mutation parameter θ (2 Nμ, wherein N is the effective population size and μ is the mutation rate per gene per generation), population growth parameter β (Nl, wherein l is the exponential growth rate per gene per generation) and the rescaled time of most recent common ancestor (TMRCA, scaled via N) via Griffiths’ maximum likelihood method (ML; Griffiths and Tavare 1994) via GeneTree version 9.0 (http://www.stats.ox.ac.uk/ simgriff/software.html) We used million simulations for each parameter set to ensure accurate results We also calculated other θ estimates such as Watterson’s estimate (θw; Watterson 1975) and Tajima’s estimate (θπ; Tajima 1983) and Tajima’s neutrality test (Tajima 1989) via DNASP version 4.0 (Rozas et al 2003) 794 D Pan, et al Table I Variable sites of Bengal slow loris Haplotype code B1 B2 B3 B4 B5 B6 B7 B8 B9 Variable sites 111111122222333 228234588901689333 795747445664220128 Number of individuals Accession no 1 3 AY875941 AY875942 AY875943 AY875944 AY875945 AY875946 AY875947 AY875948 AY875949 ATCT–GATTAAAGCTGTA G C G C .G T C GG A C.A.C .G A C GG AT C AGC G T A T A CC G TC AG B1-B9 are published in Chen et al (2005) but without frequency Results Sequence Variation We deposited the haplotype sequences in GenBank except those having been published in Chen et al (2005) (see Tables I and II for detail) We obtained fragments of ca 390 base pairs from the HVRI region of D-loop We found haplotypes in the 21 Bengal slow loris samples and 10 haplotypes in the 119 pygmy slow loris samples We previously reported all Bengal slow loris haplotypes (B1-9) and of the pygmy slow loris haplotypes (P1-4) in our previous phylogenic study of Nycticebus (Chen et al 2005) There were 18 variable sites (including one indel) in Nycticebus bengalensis but only 9, also including indel, in N pygmaeus We found no transversion in Nycticebus pygmaeus and only in N bengalensis The nucleotide diversities (Nei 1987) for Nycticebus bengalensis and N pygmaeus are Table II Variable sites of pygmy slow lorises Haplotype code P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 Variable sites 1222333 448334017 163689167 A C C TA C G C G T C T ‐ C T T T GT G C T C TA C T A C T T A P1-P4 are published in Chen et al (2005) Number of individuals Accession no 48 55 2 1 AY875959 AY875960 AY875961 AY875962 AY878369 AY878370 AY878371 AY878372 AY878373 AY878374 Population Genetic Patterns of Slow Lorises 795 1.378% and 0.192%, respectively Our data show that though the sample size of pygmy slow lorises is much larger than that of Bengal slow lorises, the polymorphism level of the former, conversely, is much lower Network With a limited number of variable sites, we did not build treelike genealogies, but instead an MJ network, which incorporates both parsimonious and minispan criteria (Bandelt et al 1995) The networks for the species showed different topologies (Fig 1) In the network of pygmy slow lorises (Fig 1b), there are dominant haplotypes, P2 and P3 All the other haplotypes are only 1- or 2-character or -site changes away from P2 or P3 Moreover, there is only a 1-character change between P2 and P3, which indicates that the current pygmy slow loris population may have originated from a single major haplotype or ancestor , which split only recently into P2 and P3 In the network of Bengal slow lorises (Fig 1a), we found no dominant haplotype The concurrent substitutions (G147A and G214A) and the haplotype B4 formed a square torso All the other haplotypes except B8 and B9 are from 1- to 3- Fig Median-joining network (a) Bengal slow loris (b) Pygmy slow loris Circles indicate the sampled haplotypes TRIANGLE = indel sites The networks are plotted by taking indels as an additional character If not, such characters should be removed Particularly for pygmy slow loris haplotype P4, if the indel is removed, P4 should be joined into haplotype P3 DIAMOND = intermediate inferred haplotypes 796 D Pan, et al character changes from it, respectively Both B8 and B9 are 6-character changes farther The topologies of the networks show that the Bengal slow loris population has more complex structures than that of pygmy slow lorises Parameter Estimation The Tajima estimate of θ (Tajima 1983) for the Bengal slow loris population (5.038) is greater than its Watterson counterpart (4.725) (Watterson 1975) But for pygmy slow lorises, the Tajima estimate (0.746) is smaller than that of Watterson (1.495) One can test the differences in the estimates quantitatively via Tajima’s neutrality test, the statistic of which—Tajima’s D—is commonly used as an indicator for recent population growth or structure when the marker is nearly neutral (Tajima 1989) Though Tajima’s Ds are not significant in the Bengal slow loris (positive) or in the pygmy slow loris population (negative), the opposite sign implies that the Bengal slow loris population was more structured while the pygmy slow loris showed the signature of population growth Without recombination, Griffiths’ ML method (Griffiths and Tavaré 1994) could estimate both θ and β simultaneously For the pygmy slow loris population, θ is 2.3 and β is 1.6, and for that of Bengal slow lorises, θ is 5.0 and β is 1.575 We plotted the likelihood distributions of the rescaled TMRCA for the populations (Fig 2) For the pygmy slow loris population, the true scaled TMRCA should be in the range of 0.63– 2.49 (p=95%), while that for Bengal slow lorises should be 0.525–1.65 (p=95%) To obtain the estimation of effective population size and absolute TMRCA, a mutation rate of mitochondrial DNA in Nycticebus is needed, which unfortunately is not available now Easteal (1991) showed that the mutation rate of mitochondrial DNA (mtDNA) is approximately the same in all animals; however, Raaum et al (2005) have questioned this inference For a rough estimation, we took the average mutation rate of the human HVRI D-loop as 5×10−7 (ranging from 1×10−7 to 1×10−6 ) per site per year (Excoffier and Yang 1999; Jazin et al 1998; Parsons and Holland 1998; Parsons et al 1997) and set the generation time as yr (Wang 1998) We obtained N from θ via Griffiths’ ML estimate Accordingly, the effective population sizes for the pygmy and Bengal slow loris population are about 2950 and 6400, respectively, and the absolute TMRCA for the pygmy slow loris population is 1860–7350 yr (p=95%) and for the Bengal form, it is ca 3360–10,560 yr (p=95%) Discussion We obtained the sequences of the partial HVRI D-loop region in both Nycticebus pygmaeus and N bengalensis Because nuclear pseudogenes (numt) are prone to be amplified when doing mtDNA PCR using total genomic DNA (Lu et al 2002), Chen et al (2005) sequenced additional cytochrome b mtDNA genes in some individuals to prove that numts had not been amplified We did not have to that because with limited intrapopulational divergences, numts were not likely to be in our data It is not likely that all of our sequences are numts In fact, Chen et al (2005) proved some of the sequences were not numts Population Genetic Patterns of Slow Lorises 797 Fig Distributions of TMRCA (a) Bengal slow loris (b) Pygmy slow loris In a, the true TMRCA would locate in 0.525–1.65 (p=95%) In b, the true TMRCA would locate in 0.63–2.49 (p=95%) The time unit is N generation, wherein N is the effective population size The polymorphism level of Nycticebus pygmaeus is much lower than that of N bengalensis, even though the sample size of the pygmy slow loris is almost times larger Given similar mutation rates between the species, the increase of polymorphism level of Nycticebus bengalensis might have resulted from the signature of population structure, presumably caused by gene flow from other habitats Another possible explanation is introgression from Nycticebus coucang coucang Groves (2001) reported and recently confirmed the presence of hybrids between Nycticebus bengalensis and N c coucang in southern peninsular Thailand In fact, in our previous study (Chen et al 2005), some Nycticebus coucang coucang were closer to N bengalensis than other subspecies of N coucang (Fig in Chen et al 2005) Unfortunately, without Nycticebus bengalensis samples from other habitats, we were unable to evaluate the strength of external gene flow As a consequence, with possible 798 D Pan, et al gene flow or gene ingression or both, the parameters estimated for the local population of Nycticebus bengalensis were biased such that θ and the absolute TMRCA could be overestimated and β could be underestimated The low level of polymorphism in Nycticebus pygmaeus generally suggests that a recent bottleneck has occurred, especially when the sample size is large and the marker evolves quickly However, the possibility of a bottleneck having occurred is low Sharing overlapping niches, the species should exhibit similar population patterns of historical dynamics, but there is no evidence for a recent bottleneck in Nycticebus bengalensis Moreover, if a bottleneck had occurred, the polymorphism level of Nycticebus bengalensis would also be low because low-frequency external haplotypes should be swept out by a bottleneck Therefore, we propose that the low polymorphism of pygmy lorises might be the result of a founder effect That is to say, the Nycticebus pygmaeus we examined from southern China and northern Vietnam might have originated from middle or southern Vietnam recently, possibly ca 1860–7350 yr (p=95%) ago The estimate of effective population size of Nycticebus pygmaeus is only about one half that for N bengalensis, a much lower ratio than expected Though it could be lessened by the overestimation of θ for Nycticebus bengalensis, there are likely other factors affecting the population sizes, of which the birth sex ratio and the infant survival rate are the most probable candidates A 7-yr study of captive breeding in Nycticebus pygmaeus (Feng et al unpublished data) showed that the birth sex ratio was 1:1.68 (female:male, totaling 52 infants from 27 litters, of which 19 were female, 32 were male, and were unknown) The skewed sex ratio would be expected to decrease the effective population size (Li and Graur 1991) In addition, the pygmy slow lorises spent less time caring for individuals in twin litters, probably because mothers need to divide their time between them, which may lead to a lower infant survival rate (confirmed in Feng’s breeding project, but not observed in Fitch-Snyder’s data, both unpublished) Though Nycticebus bengalensis generally produces only offspring per litter, they not exhibit seasonal breeding restrictions those of pygmy lorises A mother Bengal loris can become pregnant again by the time her first infant is ca mo, which means she could then produce offspring within a 1-yr period, and this, to some extent, may account for the higher effective population size of Nycticebus bengalensis Acknowledgment The Bureau of Science and Technology of Yunnan Province, and National Natural Science Foundation of China (grants 30460026, 30621092) provided funding for this work References Avise, J C (1991) Ten unorthodox perspectives on evolution prompted by comparative population genetic findings on mitochondrial DNA Annual Review of Genetics, 25, 45–69 Bandelt, H J., Forster, P., Sykes, B C., and Richards, M B (1995) Mitochondrial portraits of human populations using median networks Genetics, 141, 743–753 Chen, J H., Pan, D., Groves, C., Wang, Y X., Narushima, E., Fitch-Snyder, H., et al (2005) Molecular phylogeny of Nycticebus inferred from mitochondrial genes International Journal of Primatolology, In press Population Genetic Patterns of Slow Lorises 799 Easteal, S (1991) The relative rate of DNA evolution in primates Molecular Biology and Evolution, 8, 115– 127 Excoffier, L., and Yang, Z (1999) Substitution rate variation among sites in mitochondrial hypervariable region of humans and chimpanzees Molecular Biology and Evolution, 16, 1357–1368 Fitch-Snyder, H (2000) Reproductive patterns in a breeding colony of pygmy lorises (Nycticebus pygmaeus) Measured by behavioral and physiological correlates of gonadal activity San Diego: San Diego State University Griffiths, R C., and Tavare, S (1994) Sampling theory for neutral alleles in a varying environment Philosophical Transactions of the Royal Society of London Series B-Biological Science, 344, 403–410 Groves, C P (2001) Primate taxonomy Washington, D.C: Smithsonian Institution Press Jazin, E., Soodyall, H., Jalonen, P., Lindholm, E., Stoneking, M., and Gyllenstein, U (1998) Mitochondrial mutation rate revisited: Hot spots and polymorphism Nature Genetics, 18, 109–110 Kumar, S., Tamura, K., Jakobsen, I B., and Nei, M (2001) MEGA2: molecular evolutionary genetics analysis software Bioinformatics, 17, 1244–1245 Li, W H., and Graur, D (1991) Fundamentals of molecular evolution Sunderland, MA: Sinauer Associates Lu, X M., Fu, Y X., and Zhang, Y P (2002) Evolution of mitochondrial cytochrome b pseudogene in genus Nycticebus Molecular Biology and Evolution, 19, 2337–2341 Lu, X M., Wang, Y X., and Zhang, Y P (2001) Divergence and phylogeny of mitochondrial cytochrome b gene from species in genus Nycticebus Zoology Research, 22, 93–98 Nei, M (1987) Molecular evolutionary genetics New York: Columbia University Press Oxnard, C E., Crompton, R H., and Lieberman, S S (1990) The case of the prosimian primates Seattle: University of Washington Press Parsons, T J., and Holland, M M (1998) Reply to Jazin et al Nature Genetics, 18, 110 Parsons, T J., Muniec, D S., Sullivan, K., Woodyatt, N., Alliston-Greiner, R., Wilson, M R., et al (1997) A high observed substitution rate in the human mitochondrial DNA control region Nature Genetics, 15, 363–368 Raaum, R L., Sterner, K N., Noviello, C M., Stewart, C B., and Disotell, T R (2005) Catarrhine primate divergence dates estimated from complete mitochondrial genomes: Concordance with fossil and nuclear DNA evidence Journal of Human Evolution, 48, 237–257 Rozas, J., Sanchez-Delbarrio, X M., and Rozas, R (2003) DNASP, DNA polymorphism analyses by the coalescent and other methods Bioinformatics, 19, 2496–2497 Tajima, F (1983) Evolutionary relationship of DNA sequences in finite populations Genetics, 105, 437–460 Tajima, F (1989) Statistical method for testing the neutral mutation hypothesis by DNA polymorphism Genetics, 123, 585–595 Tan, C L., and Drake, J H (2001) Evidence of tree gouging and exudate eating in pygmy slow lorises (Nycticebus pygmaeus) Folia Primatologica, 72, 37–39 Wang, S (1998) China red data book of endangered animals Beijing, Hong Kong, New York: Science Press Watterson, G A (1975) On the number of segregating sites in genetical models without recombination Theoretical Population Biology, 7, 256–276 ... inferred from mitochondrial genes International Journal of Primatolology, In press Population Genetic Patterns of Slow Lorises 799 Easteal, S (1991) The relative rate of DNA evolution in primates Molecular... the higher effective population size of Nycticebus bengalensis Acknowledgment The Bureau of Science and Technology of Yunnan Province, and National Natural Science Foundation of China (grants 30460026,... estimated for the local population of Nycticebus bengalensis were biased such that θ and the absolute TMRCA could be overestimated and β could be underestimated The low level of polymorphism in Nycticebus