Gonghua LIN1,2, Haixin CI1,2, Simon J. THIRGOOD3, Tongzuo ZHANG1*, Jianping SU1**

1 Key Laboratory of Adaptation and Evolution of Plateau Biota, Northwest Institute of Plateau Biology, Chinese Academy of Sciences, Xining, 810008, China

e-mails: *zhangtz@nwipb.ac.cn **nwipb@hotmail.com (corresponding authors)2 Graduate University of Chinese Academy of Sciences, Beijing 100049, China

3 Macaulay Institute, Craigiebuckler, Aberdeen AB15 8QH, UK

GENETIC VARIATION AND MOLECULAR EVOLUTION OF ENDANGERED KOZLOV’S PIKA (OchOtOna kOslOwi

BüCHNER) BASED ON MITOCHONDRIAL CYTOCHROME B GENE

POLISH JOURNAL OF ECOLOGY(Pol. J. Ecol.)

58 3 563-568 2010

Regular research paper

ABSTRACT: Kozlov’s pika is a rare and en-dangered lagomorph species with a limited dis-tribution in the southern Kunlun Mountains in western China. Because of its endangered status, Kozlov’s pika is considered a priority species for research and conservation action. Genetic varia-tion and molecular evolution of the Kozlov’s pika were studied based on a total of 14 individuals from four locations along the eastern boundary between Xinjiang and Tibet province (35.20–36.48°N, 86.08–83.04°E) on extremely high eleva-tion (usually over 4800 m a.s.l.). The density of local populations was about 3–4 per ha, living in a typical alpine desert grassland habitat. The com-plete mitochondrial cytochrome b (cytb) gene was amplified and sequenced. Based on the cytb gene sequences the genetic variation and molecular evolution were analyzed. Unexpected high haplo-type diversity (0.956 ± 0.045) but low nucleotide diversity (0.00537 ± 0.00126) was found, indicat-ing past demographic expansion. Significant par-titioning of variance (P <0.01) among populations (46.7%), and within populations (53.3%), indicat-ing low level of genetic differentiations among local populations. Our results gave an optimistic survival status of Kozlov’s pika at the genetic level. Bayes Empirical Bayes analysis with model M2a and M8 detected three positively selected amino acid sites at the significance level of 0.05. The mu-tant types with either or both of the mutations aspartic acid to asparagine and glutamic acid to lysine had higher isoelectric point values. We sug-

gested these mutant types might have biological significance to help individuals to adapt to the ex-tremely high elevation habitats.

KEY WORDS: Kozlov’s pika, genetic diver-sity, genetic structure, molecular evolution

1. INTRODUCTION

Kozlov’s pika is a rare and endangered lagomorph species with a limited distribution in the southern Kunlun Mountain in western China (IUCN 2007). After initial description in 1884 it was not recorded by scientists for 100 years until it was rediscovered in 1984 (Zheng 1986). Kozlov’s pika is considered a priority species for research and conservation action because of its endangered status (L i et al. 2006). However, with the exception of the brief description of basic biology by L i et al. (2006), little is known about the survival sta-tus and ecology of Kozlov’s pika.

Molecular approaches have been devel-oped to evaluate the survival status of a spe-cies at the genetic level. The mitochondrial cytochrome b (cytb) gene is one of the most commonly used gene markers. Genetic di-versity and genetic structure are important indices that reflect a species’ genetic status. There is now a compelling evidence that loss

Gonghua Lin et al.

of genetic diversity can increase the extinc-tion risk for a species (Frankham 2005). In addition, high level genetic structure between rudimental populations can cause further de-ficiency of total diversity due to the ‘Wahlund effect’ (Hedr ick 2000). Therefore studies of pika’s genetic diversity and genetic structure are crucial for the conservation of this endan-gered species.

Moreover, Kozlov’s pika has the ex-tremely high elevation habitats in Qinghai-Tibetan Plateau. Recent study on the plateau pika (Ochotona curzoniae Hodgson), which also lives at high elevation habitats, detected positively selected sites in the nuclear ob gene (Zhao et al. 2008). They have been interpret-ed as adaptations to extremely high elevation. The mitochondrion is the ‘energy factory’ of eukaryote cell, and the product of cytb gene constitutes the cytochrome bc1 complex which plays an important role in mitochon-drial respiratory chain (Iwata et al. 1998). We suggested that the study on molecular evolution of cytb gene in Kozlov’s pika may help in understanding evolution of adapta-tions to high elevation habitats.

The aim of this paper is to provide esti-mates of genetic variation in the pika and to

discuss selection of molecular adaptations to life at high elevation in mammals.

2. MATERIALS AND METHODS

In October and November 2007, we in-vestigated the distribution of the Kozlov’s pika in the Kunlun Mountains along the eastern boundary between Xinjiang and Ti-bet province during an expedition organized by the Chinese Academy of Sciences. The study area ranged from 35.20~36.48°N and 86.08~83.04°E and the average elevation was over 4800 m a.s.l. (Fig.1). The pika lived in a typical alpine desert grassland habitat, with a small number of plant species including car-ex moorcroftii, ceratoides compacta, as well as some Leguminosae and Cruciferae species. The density of local populations was about 3–4 per ha. The body weight of adults was about 300 g and each had 7–8 shallow bur-rows with a depth of only 30–40 cm. With the permission of local government, we sampled some individuals to analyze the genetic varia-tion and molecular evolution of this species based on the mitochondrial cytb gene.

Pikas were trapped on 1–2 ha plots and each plot was sampled in the day time with

Fig. 1. Study region in the west of China. P1~P4, the four local populations in this study; M1, Arjin Mountains; M2, Kunlun Mountains; M3, Kekexili Mountains.

564

Genetic variation and molecular evolution of Kozlov’s pika

15 traps randomly placed at burrows. Pikas were decollated immediately after trapping and muscle tissue samples for DNA extrac-tion were collected and preserved in 95% al-cohol. The longitude, latitude and elevation at the centre of each sampling plot were mea-sured using a GPS unit (Garmin Etrex, Tai-wan). Due to the limited time available and the low activity of pikas only a small number of individuals were trapped (a total of 13 pi-kas were trapped from four populations (P1-P4) (Fig.1).

Total genomic DNA was extracted from approximately 0.3 g muscle following a modi-fication of the Sambrook and Russell method (Sambrook and Russel l 2001). Cytb gene was amplified using the primers:

L 1 4 7 2 4 ( 5 ’ - C G A A G C T T G ATAT-GAAAAACCATCGTTG-3’) and

H15915 (5’-CGGAATTCCATTTTTG-GTTTACAAGAC-3’) with the reference of Kocher et al. (1989). Thermocycling was conducted in a Biometra T-Gradient Ther-moblock PCR machine (Biometra, Got-tingen, Germany). The PCR products were purified using a CASpure PCR Purification Kit (Casarray, Shanghai, China), and directly sequenced using the same primers used for amplification mentioned above. Sequencing reactions were conducted in a Biometra ther-mocycler using a DYEnamic Dye Terminator Cycle Sequencing Kit (Amersham Pharmacia Biotech Inc., USA.) following the manufac-turer’s protocol. Sequencing products were separated and analyzed on a MegaBACE 1000 DNA Analysis System (Amersham Pharma-cia Biotech Inc., USA.). The sequences were checked by eye and aligned using CLUSTAL W (Thompson et al. 1997) and refined manually. Due to the small sample size from the Kunlun Mountains, the sequence from a single individual collected in 2006 near Jinyuhu Lake was also used (P4, see Fig 1). In total, 14 sequences from four populations were used in this study.

Genetic diversity, both haplotype diver-sity (Hd) and nucleotide Diversity (Pi) were calculated with DnaSP v5 software (DNA Se-quence Polymorphism; L ibrado and Rozas 2009) based on the whole samples. For ge-netic structure, standard analysis of molecu-lar variance (AMOVA) among and within the four local populations (one-group set) was

calculated with Arlequin 3.0 software based on Kimura 2P algorithm (Excof f ier et al. 2005). There was a high haplotype diversity but a low nucleotide diversity level (see Chap-ter 3 – Results and Discussion), indicating a past population expansion. In order to test the hypothesis on past population growth, we used Fu’s Fs neutral test (Arlequin 3.0. soft-ware), which is one of the most powerful tests for detecting population growth (R amos-Onsins and Rozas 2002).

Positive selection sites were detected us-ing codeml program in PAML (Phylogenetic Analysis by Maximum Likelihood) program package version 4.1 (Yang 2007). The tree used was an unrooted Neighbor Joining tree generated using MEGA 4 (Molecular Evo-lutionary Genetic Analysis; Kumar et al. 2008) based on the haplotypes. Following the suggestions in User Guide of PAML pro-gram package, the M2a (selection) and M8 (beta&ω) models with the Bayes Empirical Bayes algorithm were used. The methods use nonsynonymous/synonymous substitution rate ratio (dN/dS, or synonymically ω) as a measure of selective pressure at the protein level, with ω> 1 indicating positive selection.

3. RESULTS AND DISCUSSION

3.1. Genetic variation

A total of 24 variable sites and 11 haplo-types were identified (Table 1). Unexpected high haplotype diversity (Hd ± SD = 0.956 ± 0.045) was found, while the nucleotide diver-sity is low (Pi ± SD = 0.00537 ± 0.00126). The Fu’s Fs neutral test showed a nearly signifi-cant (P = 0.059) negative Fs value (–3.055), indicating that there was at least once past population expansion events (Excof f ier et al. 2005). Standard AMOVA analysis re-vealed significant partitioning of variance (P <0.01) among populations (46.7%), and within populations (53.3%), indicating a low level of genetic differentiations among local populations.

It is generally thought that a rare species will exhibit low genetic diversity, high inter-population genetic differentiations and de-creasing population size (Jacquemyn et al. 2007). Our results to some degree contradict this common view, although we must be cau-5

565

Gonghua Lin et al.

tious given the small sample sizes obtained during our expedition. We found evidence of high haplotype diversity and low nucleo-tide diversity which may represent a signal of rapid demographic expansion from a small effective population size (Avise 2000). The negative value in Fu’s Fs test also supports this hypothesis. Moreover, field surveys in 1999 (Li et al. 2006) showed that density of Ko-zlov’s pika in suitable habitats can be high, to some degree supporting our view. Based on our survey, the distribution of Kozlov’s pika is continuous in the study region and there is no obvious fragmentation between the four lo-cal populations. We suggest that the low level inter-population genetic differentiations have resulted from the habitat connectivity among local populations.

Our results gave an optimistic prognosis for the survival of Kozlov’s pika. However, this does not imply that there are no current conser-vation concerns. The distribution of Kozlov’s pika is so limited that previous expeditions to the Kunlun Mountains in 1989 and 1992 were unable to find the species (L i et al. 2006). Moreover, the distribution of Kozlov’s pika appears to be at risk of invasion by the plateau pika which is the dominant and keystone spe-cies on the Qinghai-Tibetan Plateau (Smith and Foggin 1999). L i et al. (2006) suggest that the plateau pika competes with Kozlov’s pika and plateau pika individuals were also found in several Kozlov’s pika populations during our own survey. However, whether and how Ko-zlov’s pika is affected by plateau pika remains unknown and requires further investigation.

Table 1. The geographic records and haplotype constitution in local populations (P1-P4) (see Fig.1) and the isoelectric point as well as variable site information of each haplotype (H1 – H11). The dot in each nucleotide position of relevant haplotype means the same as in H1.

566

Genetic variation and molecular evolution of Kozlov’s pika

3.2. Molecular evolution

The Bayes Empirical Bayes analysis with model M2a and M8 detected 3 positively selected amino acid sites at the significance level of 0.05 (Table 2). One (111 E&K) locates in transmembrane region (next to the mi-tochondria matrix side) and two (58 D&N, 171 D&N) locate in the mitochondria matrix side, according to Ir win et al. (1991) (The capital letters D, E, K and N in the context represents aspartic acid, glutamic acid, lysine and asparagine, respectively.).

Previous study (Ir win et al. 1991) on evolution of the cytb gene in mammals found that these three sites were very conservative, i.e. most of the sequences used in their study had D, E and D in No. 58, 111 and 171 Aa site, respectively. Interestingly however, these sites in some of our haplotypes (mutant types) mutated to other Aa types with significantly higher isoelectric point values (D to N, 2.77 to 5.41; E to K, 3.22 to 9.74) than the original products detected prior to mutation (Haig and Hurst 1991). Using the program Edit-Seq in DNAStar program package (version 6.0; Dr. Steve ShearDown, Madison, USA), we calculated the rough isoelectric point val-ues of the products of these haplotypes. The results indicated that either of the mutations or both mutations (mutation from D to N and from E to K) results in haplotypes (H5, H6, H9, H10 and H11) with products of higher isoelectric point values (Table 1). Accord-ing to L lopis et al. (1998), the pH value in mitochondria matrix is about 7.98 in HeLa cells and 7.91 in cardiomyocytes. Because proteins’ solubility in the matrix generally increases in response increasing deviation of pH value from the protein’s isoelectric point, we suggest that the products of slightly higher

isoelectric point values could have higher bi-ological activity. In other words, we hypoth-esize that the haplotypes with either or both of the mutations (D to N and E to K) in their Aa sequences might help individuals to adapt to the extremely high elevation habitats.

ACKNOWLEDGEMENTS: This study was supported by the Training Qualified People Plan “Hope of the West China” of The Chinese Acad-emy of Sciences and the Ministry of Personnel of China (No. O954021211) and General Programs of the National Natural Science Foundation of China (No. 30970366). Simon J. Thirgood was funded by the International Exchange Programme of the Royal Society of Edinburgh.

4. REFERENCES

Avise J.C. 2000 – Phylogeography: The history and formation of species – Harvard University Press, Cambridge (USA), 447 pp.

Excof f ier L . , L ava l L .G. , S chneider S . 2005 – Arlequin ver. 3.0: an integrated soft-ware package for population genetics data analysis – Evol. Bioinformat. Online, 1: 47–50.

Frankham R . 2005 – Genetics and extinction – Biol. Conserv. 126: 131–140.

Haig D. , Hurst L .D. 1991 – A quantitative measure of error minimization in the genetic code – J. Mol. Evol. 33: 412–417.

Hedrick P.W. 2000 – Genetics of populations – Jones and Bartlett Publishers, Massachusetts, 553 pp.

Ir win D.M. , Kocher T.D. , Wi lson A.C. 1991 – Evolution of the cytochrome b gene of mammals – J. Mol. Evol. 32: 128–144.

IUCN 2007 – IUCN red list of threatened species – www.iucnredlist.org.

Iwata S . , Lee J.W. , Okada K. , Lee J.K. , Iwata M. , R asmussen B. , L ink T.A. , R amaswamy S . , Jap B.K. 1998 – Com-plete structure of the 11-subunit bovine mito-

Table 2. Positively selected amino acid (Aa) sites (refer to conserved & mutated Aa type according to Ir win et al. (1991) detected by Bayes Empirical Bayes analysis). ω, the ratio of nonsynonymous/syn-onymous substitution rates; Pr (ω >1), probability of the case that ω >1; M2a, the selection model and M8, the beta&ω model according to Yang (2007); D, E, K and N represents aspartic acid, glutamic acid, lysine and asparagine, respectively; *, P >95%; **, P >99%.

Aa sitePr (ω >1) ω (mean ±SE)

M2a M8 M2a M858 D&K 0.995** 0.998** 8.446 ± 1.859 7.681 ± 2.041111 E&K 0.959* 0.974* 8.195 ± 2.298 7.536 ± 2.278171 D&N 0.969* 0.980* 8.270 ± 2.185 7.576 ± 2.220

567

Gonghua Lin et al.

chondrial cytochrome bc1 complex – Science, 281: 64–71.

Jacquemyn H. , Vandepitte K. , Br ys R . , Honnay O. , Roldan-Ruiz I . 2007 – Fit-ness variation and genetic diversity in small, remnant populations of the food deceptive orchid Orchis purpurea – Biol. Conserv. 139: 203–210.

Kocher T.D. , Thomas W.K. , Meyer A. , Edwards S .V. , Paabo S . , Vi l lablanca F.X. , Wi lson A.C. 1989 – Dynamics of mitochondrial DNA evolution in animals: Amplification and sequencing with conserved primers – Proc. Nat. Acad. Sci. USA, 86: 6192–6200.

Kumar S . , Nei M. , Dudley J. , Tamura K. 2008 – MEGA: A biologist-centric software for evolutionary analysis of DNA and protein sequences – Brief. Bioinform. 9: 299–306.

Li W.D. , Zhang H.B. , L iu Z.H. 2006 – Brief report on the status of Kozlov’s pika, Ochotona koslowi (Büchner), in the east Kunlun Moun-tains of China – Integr. Zool. 1: 22–24.

L lopis J. , McCaf fer y M. , Miyawaki A. , Farquhar M.G. , Tsien R .Y. 1998 – Mea-surement of cytosolic, mitochondrial, and Golgi pH in single living cells with green fluo-rescent proteins – Proc. Nat. Acad. Sci. USA, 95: 6803–6808.

Librado P. , Rozas J. 2009 – DnaSP v5: a software for comprehensive analysis of DNA

polymorphism data – Bioinformatics, 25: 1451–1452.

R amos-Onsins S .E. , Rozas J. 2002 – Statis-tical properties of new neutrality tests against population growth – Mol. Biol. Evol. 19: 2092–2100.

Sambrook J. , Russel l D. 2001 – Molecular cloning: a laboratory manual – Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 2344 pp.

Smith A.T. , Foggin J.M. 1999 – The plateau pika (Ochotona curzoniae) is a keystone spe-cies for biodiversity on the Tibetan Plateau – Anim. Conserv. 2: 235–240.

Thompson J.D. , Gibson T.J. , Plewniak F. , Jeanmougin F. , Higgins D.G. 1997 – The CLUSTAL_X windows interface: flex-ible strategies for multiple sequence alignment aided by quality analysis tools – Nucl. Acid. Res. 25: 4876–4882.

Yang Z.H. 2007 – PAML 4: Phylogenetic analy-sis by maximum likelihood – Mol. Biol. Evol. 24: 1586–1591.

Zhao X.Q. , Qi D.L. , Yang J. 2008 – Molecu-lar evolution and adaptation of local species in Qinghai-Tibetan Plateau – Science Press, Bei-jing, pp. 118–150.

Zheng C.L. 1986 – Recovery of Kozlov’s pika (Ochotona koslowi Büchner) in the Kunlun Mountains of Xinjiang Uygur Autonomous region, China – Acta. Their. Sin. 5: 285.

Received after revision september 2009

568