The Phylogenetic Position and Speciation Dynamics of the Genus Perdix (Phasianidae, Galliformes)

ARTICLE IN PRESS

Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier.com/locate/ympev

The phylogenetic position and speciation dynamics of the genus Perdix (Phasianidae, Galliformes)

Xin-kang Bao a, Nai-fa Liu a,*, Jiang-yong Qu b, Xiao-li Wang a, Bei An a, Long-ying Wen c, Sen Song a a School of Life Science, Lanzhou University, Lanzhou 730000, China b Department of Biological Science and Technology, Qiongzhou College, Wuzhishan, Hainan 572200, China c Department of Chemistry & Life Sciences, Leshan Teachers College, Leshan, Sichuan 614004, China article info abstract

Article history: The nuclear gene (c-mos) and mitochondrial genes (CYT B and ND2) sequences, representing 44 phasia- Received 30 December 2009 nid species and 26 genera (mainly distributed in China), were used to study the phylogeny of the genus Revised 24 March 2010 Perdix, which comprises three partridge species. Maximum parsimony and Bayesian methods were Accepted 29 March 2010 employed, and the analysis of mitochondrial sequence data and the combined dataset showed that Perdix Available online xxxx is speciﬁcally related either to typical pheasants or to Ithaginis. Phylogenetic trees also indicated that: (1) Perdix is monophyletic; (2) the Tibetan partridge (Perdix hodgsoniae) has been consistently placed as basal Keywords: to all other Perdix, and the Daurian partridge (Perdix dauuricae) is placed sister to gray partridge (Perdix Phasianidae perdix); (3) the Daurian partridge subspecies przewalskii and Tibetan partridge subspecies hodgsoniae are Perdix c-mos basal to other subspecies in their species clade, respectively. Speciation in Perdix was likely promoted by Cytochrome b the late Pliocene/early Pleistocene intensive uplift of the Tibetan Plateau and by Pleistocene glaciations. ND2 Ó 2010 Published by Elsevier Inc. Phylogenetics Speciation

1. Introduction evidence, Johnsgard (1988) constructed a highly speculative phyletic dendrogram of perdicine genera, in which Perdix was placed The typical partridges (Perdix) contain three grassland-adapted sister to Francolinus. Another point of view is that Perdix belongs species: gray, Daurian and Tibetan partridge (Perdix perdix, Perdix to a clade of small ‘‘pheasants” (Hudson et al., 1966), and has a clo- dauuricae and Perdix hodgsoniae). They have a number of typical ser relationship with true pheasant species than with other genera partridge traits such as small build, little sexual dimorphism in traditionally considered ‘‘partridges”, such as Alectoris (Randi et al., weight and plumage and little or no plumage iridescence. They 1991; Wen et al., 2005). Some DNA-based evidences have sug- were traditionally placed in the tribe Perdicini, subfamily Phasiani- gested a novel hypothesis that gray partridge, wild turkey (Melea- nae and family Phasianidae (Johnsgard, 1988). The gray partridge is gris gallopavo) and grouse (Tetraoninae) might share a closer distributed across a wide range of Eurasian grasslands in eight sub- relationship (Kimball et al., 1999; Lucchini and Randi, 1999; Dim- species, which range from Scandinavia south to northern Spain and cheff et al., 2002; Pereira and Baker, 2006; Crowe et al., 2006). Re- Italy, and east to the western Siberia and Xinjiang (see Fig. 1). The cent molecular study (Kimball and Braun, 2008) supported the gray partridge was recently introduced into North America. Dauri- placement of the gray partridge at the base of a large clade contain- an partridge has three subspecies (Cheng et al., 1978; Zheng, 2005) ing typical pheasants, rather than the weakly supported with wild living across Asian grasslands from the Russian Altai east to Amur- turkey as indicated by Crowe et al. (2006). land and Ussuriland. There is limited overlap with the gray par- Johnsgard (1988) used fossil and zoogeographic data to postu- tridge in western Xinjiang and with Tibetan partridge in the late that Perdix species probably originated in northern Asia, and northern part of Qinghai. The four subspecies of Tibetan partridge glaciations may have isolated their ancestral form into a westerly are residents of montane grasslands around the Tibetan Plateau component that gave rise to gray partridge, an easterly form that (Fig. 1). produced Daurian partridge and a southern and more montane The phylogenetic position of Perdix within Phasianidae has been adapted population in the Himalayas that evolved into Tibetan par- controversial. Based on morphological and zoogeographic tridge. Based on morphologic and zoogeographic informations, Cheng et al. (1978) conjectured that Daurian and Tibetan par-

* Corresponding author. Fax: +86 0931 8912561. tridges have a common origin, and gray partridge separated earlier E-mail address: [email protected] (N.-f. Liu). from the proto-Perdix stock.

1055-7903/$ - see front matter Ó 2010 Published by Elsevier Inc. doi:10.1016/j.ympev.2010.03.038

Please cite this article in press as: Bao, X.-k., et al. The phylogenetic position and speciation dynamics of the genus Perdix (Phasianidae, Galliformes). Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.03.038 ARTICLE IN PRESS

2 X.-k. Bao et al. / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx

Fig. 1. Distribution of the Daurian (D), gray (G), Tibetan (T) partridges and the speciational course (as arrows indicate) inferred from the combined data (c-mos + CYT B + ND2) analyses.

In this paper, nucleotide sequences of nuclear proto-oncogene Lucchini (1998); primers used for PCR amplification and sequenc- (c-mos) and two mitochondrial genes (cytochrome b and NADH ing of CYT B, ND2 and c-mos were gathered from the literature dehydrogenase subunit 2) representing all three extant species of (Table 2). Perdix and multiple genera of Phasianidae were analyzed to (1) To amplify DNA, 50 lL PCR reactions were performed using ascertain the evolutionary relationship between Perdix and other standard buffer and MgCl2 concentrations, 2.5 mM dNTP 4 lL, related phasianid genera, (2) produce a molecular phylogeny of 10 lM each primer 3 lL, 2 units of BIOTAQ DNA polymerase (Bio- the extant species of Perdix, and (3) correlate the inferred molecu- line, Randolph, MA), and 100 ng of genomic DNA. The thermal pro- lar phylogeny and extent of interspecific genetic divergence with file comprised an initial denaturation step at 94 °C for 5 min, Pliocene/Pleistocene biogeographical scenarios in the Qinghai–Ti- followed by 35 cycles of 94 °C for 1 min, 48–55 °C for 1 min and bet Plateau region to suggest speciational patterns in Perdix. 72 °C for 2 min, with a final extension step of 72 °C for 8 min. PCR products were gel-purified using 1.5% low melting point aga- 2. Materials and methods rose, and were sequenced using an ABI 377 DNA sequencer. The nucleotide sequences were double-stranded and aligned 2.1. Taxon sampling using CLUSTAL X with the default options (Thompson et al., 1997) and refined manually. Gene sequence boundaries were DNA samples were collected with a focus on the taxa distributed determined by comparison with published sequences of other gen- in China because the country is rich in gamebirds, especially in era of Galliformes downloaded from Genbank. Sequences collected phasianid species, and most of the distribution of Daurian and Tibe- for this study have been deposited in GenBank (accession numbers tan partridge is restricted to China. Multiple representatives of all are shown in Table 1). the phasianine genera and almost all perdicine genera (except the genus Lerwa) throughout China have been sampled. Taxa studied herein (Table 1, Classification of Galliformes cited from Johnsgard 2.3. Phylogenetic analysis (1988)) include 44 phasianid species and 26 of 37 phasianine genera. The choice of outgroup on which to root cladograms are based on the Considering that all mtDNA gene sequences are virtually inher- assumptions that the Anseriformes (ducks, geese and screamers) are ited as one linkage group, and CYT B and ND2 showed similar rates sister to the Galliformes (Sibley and Ahlquist, 1990; Groth and Bar- and types of both nucleotide and amino acid substitutions (John- rowclough, 1999; Cracraft et al., 2004; Hackett et al., 2008). son and Sorenson, 1998), the two mtDNA gene segments were concatenated into a single partition and analyzed simultaneously. 2.2. Laboratory methods Congruence among the different DNA data sets was evaluated using the incongruence-length-difference (ILD) test (Farris et al., DNA was extracted from blood, pin feather or muscle tissue by 1995), which was conducted using only parsimony-informative the ethanol sedimentation procedure as described by Randi and characters. A total of 1000 replicates were conducted for the ILD

X.-k. Bao et al. / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx 3

Table 1 Studied taxa and source of DNA sequence data (classiﬁcation of Galliformes cited from Johnsgard (1988)).

Avian species Sample locality/source GenBank Accession Nos. CYT B ND2 c-mos Order: Galliformes Family: Phasianidae Subfamily: Phasianinae Tribe: Perdicini Tetraogallus tibetanus tibetanus Artux, Xinjiang Uygur Autonomous Region EU839457 GU214303 GU214274 Tetraogallus tibetanus przewalskii Tianzhu county, Gansu province EU839456 EU845746 GU214273 Tanggula, Qinghai province GU214288 EU845747 Tetraogallus himalayensis himalayensis Zhaosu county, Xinjiang U.A.R. EU839459 EU845748 GU214275 Tetraogallus himalayensis koslowi Delingha, Qinghai province GU214290 EU845750 GU214272 Tetraogallus himalayensis grombczewskii Taxkorgan, Xinjiang U.A.R. GU214289 GU214304 Tetraophasis obscurus Tongren county, Qinghai province EU839482 EU845756 GU214262 EU839483 EU845757 GU214260 Tetraophasis szechenyii Yajiang county, Sichuan province EU83944 EU845759 GU214261 GU214259 Alectoris chukar potanini Helan Mountain, Ningxia GU214291 GU214295 GU214235 Alectoris chukar pubescens Hebei province GU214292 GU214299 GU214237 Alectoris chukar falki Zhaosu county, Xinjiang U.A.R. GU214293 GU214298 GU214234 Alectoris chukar dzungarica Changji, Xinjiang U.A.R. EU839475 GU214296 Alectoris magna Qinghai province EU839473 EU845744 GU214236 Francolinus swainsonii Crowe et al. (2006) AM236907 DQ768287 Francolinus levaillantii Nadeau et al. (2007) and Crowe et al. (2006) EF571184 DQ768291 Francolinus lathami Crowe et al. (2006) AM236893 DQ768257 Francolinus sephaena Crowe et al. (2006) AM236894 DQ768274 Perdix perdix Italy EU839466 EU845763 GU214242 GU214276 GU214308 GU214241 Perdix dauuricae dauuricae Ku’erle, Xinjiang U.A.R. GU214278 GU214307 GU214239 Perdix dauuricae przewalskii Qinghai province GU214277 EU845762 GU214243 Perdix dauuricae suschkini Hulunbeier, Inner Mongolia EU839467 EU845761 GU214240 EU839469 GU214306 GU214238 Perdix hodgsoniae hodgsoniae Maizhokunggar County, Tibet GU214279 GU214309 GU214244 Perdix hodgsoniae koslowi Sunan county, Gansu province EU839472 GU214310 GU214246 Perdix hodgsoniae sifanica Xunhua county, Qinghai province EU839470 EU845764 GU214245 Coturnix coturnix Lanzhou zoo EU839461 EU845745 GU214268 Arborophila javanica Crowe et al. (2006) and Bowie and Fjeldsa (2005) AM236890 DQ093804 Arborophila ruﬁpectus He et al. (2009) NC_012453 Xenoperdix udzungwensis Crowe et al. (2006) and Bowie and Fjeldsa (2005) AM236887 DQ093798 Bambusicola thoracica Guizhou province EU839452 EU845753 GU214271 GU214286 GU214305 GU214270 Tribe: Phasianini Ithaginis cruentus berezowskii Diebu county, Gansu province EU839462 EU845776 GU214258 GU214280 GU214311 GU214257 Ithaginis cruentus geoffroyi Yajiang county, Sichuan province EU839465 GU214312 GU214256 Tragopan temminckii Diebu county, Gansu province EU839485 EU845774 GU214265 EU839486 EU845775 GU214264 Tragopan blythii Randi et al. (2000) and Kimball and Braun (2008) AF200722 DQ307013 Pucrasia macrolopha Kimball et al. (1999) and Crowe et al. (2006) AF028800 DQ768269 GU214267 Diebu county, Gansu province GU214287 GU214313 GU214266 Lophophorus lhuysii Diebu county, Gansu province EU839487 EU845760 GU214263 Lophophorus impejanus Kimball et al. (1999) and Kimball and Braun (2008) AF028796 DQ307007 Gallus gallus Hainan province EU839454 EU845754 GU214269 Crossoptilon auritum Ruoergai, Sichuan province EU839478 EU845771 Crossoptilon crossoptilon Yajiang county, Sichuan province GU214282 GU214314 GU214247 Crossoptilon harmani Biru county, Tibet GU214283 GU214314 GU214248 Lophura nycthemera Kornegay et al. (1993) and Kimball and Braun (2008) L08380 DQ307009 GU214252 Lophura swinhoii Bush and Strobeck (2003) and Kimball and Braun (2008) AF534558 DQ307010 Lophura inornata Randi et al. (2001) and Kimball and Braun (2008) AF314642 DQ307008 Catreus wallichii Kimball et al. (1999) and Kimball and Braun (2008) AF028792 DQ307003 Syrmaticus ellioti Guizhou province GU214284 GU214318 GU214253 GU214251 Syrmaticus humiae Zhan and Zhang (2003) Unpublished AF534706 NC_010774 Kato et al. (2008) Unpublished Syrmaticus reevesii Kato et al. (2008) Unpublished NC_010770 Phasianus colchicus Qingshui county, Gansu province EU839480 EU845770 GU214255 GU214285 GU214316 GU214254 Chrysolophus pictus Qingshui county, Gansu province EU839476 EU845767 GU214250 EU839477 EU845768 GU214249 Chrysolophus amherstiae Zhan and Zhang (2005) and Crowe et al. (2006) AY368052 DQ768277 Polyplectron bicalcaratum Davison et al. (2007) Unpublished EU036227 EF569479 Kimball and Braun (2008) Pavo muticus Zhu (2005) Unpublished, Kimball and Braun (2008) DQ010650 EF569478 Afropavo congensis Kimball et al. (1999) and Kimball and Braun (2008) AF013760 DQ307000

(continued on next page)

4 X.-k. Bao et al. / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx

Table 1 (continued)

Avian species Sample locality/source GenBank Accession Nos. CYT B ND2 c-mos

Subfamily: Tetraoninae Tetrao urogallus Shibusawa et al. (2004) and Dimcheff et al. (2002) AB120132 AF222565 Lagopus lagopus Nadeau et al. (2007) and Dimcheff et al. (2002) EF571187 AF222552 Subfamily: Meleagridinae Meleagris gallopavo Kornegay et al. (1993) and Dimcheff et al. (2002) L08381 AF222556 Order: Anseriformes Anas platyrhynchos Tu et al. (2007) Unpublished NC_009684 EU585672 AF478185 Gonzalez et al. (2009) Unpublished Garcia-Moreno et al. (2003)

Table 2 We used an internal fossil-based anchor point to estimate Primers used for DNA amplification and sequencing. divergence times within Perdix. The earliest known fossil record Gene Primer Primer sequence (50 ? 30) Source of Daurian partridge is approximately 2.0 million years old from region namea Chou-kou-tien in northeastern China (Hou, 1982). CYT B L14578 CTAGGAATCATCCTAGCCCTAGA Crowe et al. (2006) L15311 CTCCCATGAGGCCAAATATC Kimball et al. (1999) 3. Results H15670 GGGTTACTAGTGGGTTTGC H16065 TTCAGTTTTTGGTTTACAAGAC 3.1. The analyses of the nuclear c-mos data set ND2 L5216 GCCCATACCCCRAAAATG Sorenson et al. (1999) H6313 CTCTTATTTAAGGCTTTGAAGGC The aligned c-mos matrix is 613 bp long (54 ingroup taxa), 69 bp c-mos F944 GCCTGGTGCTCCATCGACTGG Cooper and Penny (1997) R1550 GCAAATGAGTAGATGTCTGCT of which were variable but phylogenetically uninformative, and 93 bp were variable and potentially phylogenetically informative. a 0 Primer names indicate light (L) or heavy (H) strand and the position of the 3 Parsimony analysis identified 168 trees with a length of 281 steps, end of the oligonucleotide numbered according to the complete chicken mitochondrial DNA genome (Desjardins and Morais, 1990). a consistency index (CI) of 0.658 and a retention index (RI) of 0.866. The consensus MP tree was congruent with the 50% majority rule consensus tree from Bayesian analysis, and the topology was test and the searches used TBR branch swapping with the number similar with the combined dataset analyses (not showed here). of trees retained for each replicate limited to 1000. Each data set was subjected to maximum parsimony (MP) using 3.2. Combined analyses of mitochondrial DNA sequences (CYT B + PAUP*4.0b10 (Swofford, 2003) and Bayesian analyses using MrBa- ND2) yes 3.0 (Huelsenbeck and Ronquist, 2001). MP analyses involved a heuristic search strategy with 100 replicates of random addition of The partition homogeneity test with equal weighting indicated sequences, in combination with ACCTRAN character optimization, that cyt b and ND2 did not represent significantly different parti- MULPARS + TBR branch swapping and STEEPEST DESCENT options tions of the data (p = 0.06 > 0.05), so combining these two gene re- on. Bootstrap proportions (BP; see Felsenstein, 1985) were calcu- gions is justified. This combined data set was composed of 61 lated from 1000 replicates using a heuristic search with simple accessions from all 26 genera. The aligned matrix consisted of addition with TBR and MULPARS options on. The Bayesian analyses 2166 positions (CYT B complete 1143 bp and ND2 partial 1023 bp), were partitioned, which to be found advantageous for the coding including 135 parsimony uninformative variable sites and 970 regions in the same group of birds (Kimball et al., 2006), and four potentially parsimony-informative variable sites. The MP analyses simultaneous Monte Carlo Markov chains (MCMCs) were run for retained 1 tree (length = 6261steps; CI = 0.286; RI = 0.603). 5000,000 generations saving a tree every 100 generations. The For combined mitochondrial DNA sequences, the topology of best-fit model GTR + I + G (for CYT B + ND2 dataset and combined the MP tree was consistent with the 50% majority consensus tree dataset) and SYM + I + G (for c-mos dataset) selected by hLRT in from Bayesian analysis (Fig. 2), the only difference being the place- MrModeltest 2.3 (Nylander, 2008) was used in the Bayesian analy- ment of the blood pheasant (Ithaginis) clade. In MP analysis, the ses. A majority rule consensus tree was calculated with PAUP* from genus Ithaginis was sister to Perdix with strong support (BP = 94%, the last 37,501 out of the 50,001 trees sampled. The first 12,500 Fig. 2), but in the Bayesian consensus tree, it was placed in basal trees (burn-in) were excluded to avoid trees that might have been position to the pheasants (as indicated in Fig. 3). In this cladogram, sampled prior to convergence of the Markov chains. Posterior three lineages received relatively high support: (1) the true par- probability (PP) of each topological bipartition was estimated by tridge clade (including Tetraogallus/Alectoris/Coturnix/Francolinus its frequency across all the 37,501 trees sampled. swainsonii, BP = 77%, PP = 100%), (2) the junglefowl-partridge clade A Bayesian analysis of combined data (c-mos + CYT B + ND2) (including Bambusicola/Gallus/Francolinus, BP = 92%, PP = 100%), was also used to estimate the divergence times for clades of Perdix and (3) the large pheasant-partridge clade (Lophura/Catreus/Phasi- by software BEAST 1.4.8 (Drummond and Rambaut, 2007). The anus/Chrysolophus/Crossoptilon/Syrmaticus, and Perdix/Ithaginis, more parameter-rich and the best-fit model GTR + I + G was used Tetraophasis/Lophophorus/Tragopan, Pucrasia, also including the in BEAST. In the Markov chain Monte Carlo analysis, samples from Tetraoninae and Meleagridinae genera, BP = 79%, PP = 99%). the posterior were drawn at every 5000 steps over a total of 50,000,000 steps, following a discarded burn-in of 5000,000 steps. 3.3. Combined analyses of nuclear c-mos and mitochondrial genes The analysis was repeated and the samples from the two runs were fragments combined. Convergence was assessed in Tracer v1.3 (Rambaut and Drummond, 2004) and the effective sample sizes (ESSs) of param- The ILD test (P = 0.16) showed that the nuclear and mitochon- eters sampled from MCMC in our run were more than 200. drial datasets were fully congruent, and it is therefore justifiable

X.-k. Bao et al. / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx 5

Fig. 2. Maximum-parsimony tree constructed from the combined CYT B + ND2 data set. Numbers at nodes represent percentage of MP bootstrap values (above) and proportion of posterior probability values from Bayesian analyses (below).

to combine them into a single data set for further analysis. The 3.4. Phylogeny of the typical partridges (genus Perdix) ﬁnal combined sequence matrix comprised 2779 characters after alignment, of which 182 sites were variable but uninformative, All of our analyses grouped all Perdix haplotypes into a mono- and 954 sites were variable and potentially parsimony- phyletic cluster (supported by BP = 100%, PP = 100%). Tibetan par- informative. tridges were consistently placed basal to other Perdix species. All A heuristic MP search resulted in 21 trees (length = 3712 steps; three subspecies of Daurian partridge congregated together and CI = 0.441; RI = 0.743). The MP analyses of the concatenated data- were sister to the gray partridge samples from Italy. Within the set produced a similar topology to the majority consensus tree of Tibetan partridge clade, the subspecies hodgsoniae was basal to the Bayesian inferences (Fig. 3) except for the position of the Ithag- the other two subspecies, and within Daurian partridge, subspecies inis clade, as described for the mitochondrial DNA analysis. Only 43 przewalskii was in a basal position. haplotypes and 16 genera (excluding the outgroup) were included in the dendrogram because of missing data of the c-mos partition. 3.5. Estimates of evolutionary divergence of Perdix The ﬁnal combined data generally provided a higher resolution than separate analyses. Similar to the analyses of the combined The minimum for the divergence of Daurian partridge and gray mitochondrial dataset, Perdix was placed consistently within the partridge was set to 2.0 myr (95%HPD: 1.67–2.33 myr). The result tribe Phasianini (PP = 91%, BP = 90%). The true partridge clade (Tet- of divergence time estimated by software BEAST indicated that raogallus/Alectoris/Coturnix) was grouped with the junglefowl-par- the bifurcation of Tibetan partridges from proto-perdix stock tridges clade (Bambusicola/Gallus) into one lineage in the Bayesian probably occurred around 3.63 million years ago (95%HPD: 1.71– tree (PP = 68%) but not in the MP tree. 7.01 myr).

6 X.-k. Bao et al. / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx

Fig. 3. The 50% majority rule consensus tree derived from Bayesian analysis of the combined dataset (c-mos + CYT B + ND2). Numbers at nodes represent percentage of posterior probability values from Bayesian analyses and proportion of MP bootstrap values.

X.-k. Bao et al. / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx 7

4. Discussion to any other species of Perdicini (Johnsgard, 1988), and it now ap- pears certain that the genus must be reassigned to the pheasants. 4.1. Phylogeny of Perdix partridges 4.3. Speciation of Perdix Some previous studies suggested that Perdicini and Phasianini cannot be monophyletic (Kimball et al., 1999; Lucchini and Randi, Divergence time estimates indicated that the Tibetan partridge 1999; Armstrong et al., 2001; Bush and Strobeck, 2003; Pereira and split from the ancestor of Daurian partridge and gray partridge Baker, 2006; Crowe et al., 2006; Kimball and Braun, 2008). Our about 3.63 myr ago (corresponding to middle Pliocene, geologic analyses also support this and show that not only the partridge time scale conforming to GTS2004 (Gradstein et al., 2004)). This genera Bambusicola, Francolinus and the pheasant genus Gallus date approximately corresponds to the beginning of the intensive should be placed within one clade, but also other partridge genera uplift of the Tibetan Plateau, known as the Qingzang Movement (Perdix and Tetraophasis) were consistently placed within that occurred in three phases: 3.4 myr, 2.5 myr, and 1.7 myr (Li pheasants. et al., 1979, 1996, 2001; Liu et al., 1998; Shi et al., 1999). The Late The genus Perdix has long been of ‘‘uncertain” placement Miocene Hipparion fauna and the Late Pliocene Equus fauna, the (Johnsgard, 1988; Kimball et al., 1999; Crowe et al., 2006; Kimball most typical representatives of mammalian fauna in the northeast- and Braun, 2008). The sexes are somewhat dimorphic, with males ern part of Qinghai–Xizang Plateau, indicate that the plateau used having brown to blackish breast patches, and their tails have 16–18 to be characterized by tropical semiarid savanna with strong sea- rectrices, which are congruent with the key ‘‘characters” (Johns- sonal variation in the Late Miocene and converted into a cold gard, 1988) of pheasants used to distinguish from partridges. The and arid climate with its elevation in the Late Pliocene (Deng, seemingly distinctive courtship behavior of Perdix also suggests 2004). Therefore, we consider that the grassland-adapted ancestor that it is not very close to Alectoris or other genera typically consid- of the typical partridges spread to the plateau during the middle ered to form the perdicine lineage (Johnsgard, 1988). These traits Pliocene (4.52–2.75 myr ago). Throughout the course of the Qingz- make it difficult to define an entirely satisfactory morphological ang Movement and the Pleistocene glaciations, some populations or behavioral classification for Perdix. According to past nucleotide of the ancestor stayed in some basins such as the Linxia Basin, sequences analyses, Perdix has been suggested to be sister to tur- where there was no glacier effect because of its lower altitude keys and grouse (Dimcheff et al., 2002; Pereira and Baker, 2006; and precipitation (Li and Li, 1991), and stepwise acclimatized pla- Crowe et al., 2006). In contrast, our investigation of gene sequences teau environments, survived the glaciations and gave rise to Tibe- employing all three species of the genus Perdix confirm that Perdix tan partridge. Other groups were forced to move north by rapid should either sister to the Gallopheasant clade (in Bayesian analy- ascension of the Plateau, whereafter the big second glacier sis) or sister to Ithaginis (in MP analyses), corroborating both non- (approximately 2.05–1.28 myr ago) in central China (Cao, 1996) DNA evidence (Hudson et al., 1966; Randi et al., 1991) and the con- compelled them to split into west and east branches, which now clusion of multigene study (Kimball and Braun, 2008). form the prey partridge in the west and the Daurian partridge in Within the typical partridges, the morpho-behavioral attributes the east (as arrows indicate in Fig. 1). Liukkonen-Anttila et al. of gray partridge and Daurian partridge, such as their abdominal (2002)’s findings showed the possible post-glacial dispersal events hooflike patch, non-white chin, lack of a black patch on the cheek and recolonization pattern of European gray partridge from wes- and calls, are more alike and distinguish them from the Tibetan tern refugia. partridge. In addition, the sexes of gray partridge and Daurian par- In summary, we showed that Perdix is monophyletic and should tridge are more dimorphic than Tibetan partridge. Tibetan par- be placed into the pheasant group. It possibly has a sister relation- tridges have no abdominal hooflike patch and the female differs ship to the blood pheasants. Speciation in Perdix was likely pro- from the male only in being slightly smaller. With regard to the moted by the sudden rise of the Tibetan Plateau and the climatic phylogeny of these three species, the gene sequence analyses fluctuations during the late Pliocene/Pleistocene, which were con- showed that gray partridges are sister to Daurian partridges and sidered to be one of the most important impact on phylogeograph- that the Tibetan partridge separated earlier. We sampled all three ic patterns and distributions of other phasianid species inhabiting subspecies of Daurian partridge and Tibetan partridge and found central Asia (e.g. Randi et al., 2000). that the subspecies P. dauuricae przewalskii and P. hodgsoniae hodgsoniae were basal to the other two subspecies in their own clade, Acknowledgments respectively.

This work was supported by National Natural Science Founda- 4.2. Phylogenetic position of Ithaginis and Tetraophasis tion of China (No. 30530130). We thank Haijun Gu (Forestry Pro- tection Department Ofﬁce of Sichuan province), Lin Zuo There is still no good basis for classifying the blood pheasants, a (Ruoergai Nature Reserve) and Huaming Zhou (Gongga National small partridge-like montane species which occurs in upper Nature Reserve) for their help with specimens collecting, as well reaches of the coniferous forest around Tibetan Plateau, in either as Dr. Dongshi Wan, Dr. Dongrui Jia and Dr. Zhaofeng Wang for the Perdicini or the Phasianini (Cheng et al., 1978; Johnsgard, help with data analysis. We are especially grateful to the anony- 1999). Delacour (1977) suggested that the lanceolate feathers, mous reviewers for helpful comments on the manuscript, and Dr. crest, and short beak of the blood pheasants suggest a possible Jicheng Liao and Emily H. Gray for help in improving the English. relationship to koklass pheasant (genus Pucrasia). We prefer to consider it as an ancestral stock within the pheasant clade based References on their unusual characteristics and the molecular phylogenetic basal position to other pheasants in our Bayesian analyses (Fig. 3). Armstrong, M.H., Braun, E.L., Kimball, R.T., 2001. Phylogenetic utility of avian ovomucoid intron G: a comparison of nuclear and mitochondrial phylogenetics The monal-partridges (genus Tetraophasis), which are endemic in Galliformes. Auk 118, 799–804. to China, were clearly monophyletic and present in a clade with Bowie, R.C.K., Fjeldsa, J., 2005. Genetic and morphological evidence for two species monal pheasant (Genus Lophophorus)(Figs. 2 and 3; support value in the Udzungwa forest partridge Xenoperdix udzungwensis. J. East Afr. Nat. Hist. 94 (1), 191–201. P91%), as previously suggested by Meng et al. (2008). The adult Bush, K.L., Strobeck, C., 2003. Phylogenetic relationships of the Phasianidae reveals plumage of this large alpine species shows no great similarities possible non-pheasant taxa. J. Hered. 94, 472–489.

8 X.-k. Bao et al. / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx

Cao, X.S., 1996. The division of quaternary climate in Gansu province. Arid Zone Res. Li, J.J., Wen, S.X., Zhang, Q.S., 1979. A discussion on the period, amplitude and type 13, 28–40. of the uplift of the Qinghai–Xizang plateau. Scientia Sinica 22, 1314–1328. Cheng, T.S., Tan, Y.K., Lu, T.CH., Tang, CH.ZH., Bao, G.J., Li, F.L., 1978. Fauna Sinica, Li, B.Y., Li, J.J., 1991. Quaternary Glacial Distribution of the Tibet plateau. Science series Vertebrata Aves. Galliformes, vol. 4. Science, Press, Peking. Press, Beijing. Cooper, A., Penny, D., 1997. Mass survival of birds across the K-T boundary: Li, J., Fang, X., Ma, H., Zhu, J.J., Pan, B.T., Chen, H.L., 1996. Geomorphological and molecular evidence. Science 275, 1109–1113. environmental evolution in the upper reaches of the Yellow River during the Cracraft, J., Barker, F.K., Braun, M., Harshman, J., Dyke, G., Feinstein, J., Stanley, S., late. Cenozoic. Sci. China D 39, 380–390. Cibois, A., Schikler, P., Beresford, P., García-Moreno, J., Sorenson, M.D., Yuri, T., Li, J.J., Fang, X.M., Pan, B.T., Zhao, Z.J., Song, Y.G., 2001. Late Cenozoic intensive uplift Mindell, D.P., 2004. Phylogenetic relationships among modern birds: of Qinghai–Xizang plateau and its impacts on environments in surrounding (Neornithes): towards an avian tree of life. In: Cracraft, J., Donoghue, M.J. area. Quaternary Sci. 21, 381–391. (Eds.), Assembling the Tree of Life. Oxford University Press, New York, pp. 468– Lucchini, V., Randi, E., 1999. Molecular evolution of the mtDNA control-region in 489. galliform birds. In: Adams, N.J., Slotow, R.H. (Eds.), Proceedings of the 22 Crowe, T.M., Bowie, R.C.K., Bloomer, P., Mandiwana, T.G., Hedderson, T.A.J., Randi, E., International Ornithological Congress. Durban. Birdlife South Africa, Pereira, S.L., Wakeling, J., 2006. Phylogenetics, biogeography and classification Johannesburg, pp. 732–739. of, and character evolution in, gamebirds (Aves: Galliformes): effects of Liukkonen-Anttila, T., Uimaniemi, L., Orell, M., Lumme, J., 2002. Mitochondrial DNA character exclusion, data partitioning and missing data. Cladistics 22, 495–532. variation and the phylogeography of the grey partridge (Perdix perdix)in Delacour, J., 1977. The Pheasants of the World, second ed. World Pheasant Europe: from Pleistocene history to present day populations. J. Evol. Biol. 15, Association and Spur Publications, Hindhead, UK. 971–982. Deng, T., 2004. Evolution of the late Cenozoic mammalian faunas in the Linxia Basin Liu, D.S., Zhang, X.S., Yuan, B.Y., 1998. The impact of plateau uplifting on and its background relevant to the uplift of the Qinghai–Xizang plateau. surrounding areas. In: Sun, H.L., Zheng, D. (Eds.), Formation, Evolution and Quaternary Sci. 24, 413–420. Development of Qinghai–Xizang (Tibetan) Plateau. Guangdong Science and Desjardins, P., Morais, R., 1990. Sequence and gene organization of the chicken Technology Press, Guangdong, China. mitochondrial genome. A novel gene order in higher vertebrates. J. Mol. Biol. Meng, Y., Dai, B., Ran, J.H., Li, J., Yue, B.S., 2008. Phylogenetic position of the genus 212, 599–634. Tetraophasis (Aves, Galliformes, Phasianidae) as inferred from mitochondrial Dimcheff, D.E., Drovetski, S.V., Mindell, D.P., 2002. Phylogeny of Tetraoninae and and nuclear sequences. Biochem. Syst. Ecol. 36, 626–637. other galliform birds using mitochondrial 12S and ND2 genes. Mol. Phylogenet. Nadeau, N.J., Burke, T., Mundy, N.I., 2007. Evolution of an avian pigmentation gene Evol. 24, 203–215. correlates with a measure of sexual selection. Proc. Biol. Sci. 274 (1620), 1807– Drummond, A.J., Rambaut, A., 2007. BEAST: Bayesian evolutionary analysis by 1813. sampling trees. MC Evol. Biol. 7, 214. Nylander, J.A.A., 2008. MRMODELTEST, Version 2.3. Program Distributed by the Farris, J.S., Kallersjo, M., Kluge, A.G., Bult, C., 1995. Constructing a significance test Author. Available from: . Dept. for incongruence. Syst. Biol. 44, 570–572. Systematic Zoology, Evolutionary Biology Centre, Uppsala University, Uppsala, Felsenstein, J., 1985. Confidence limits on phylogenies: an approach using the Sweden. bootstrap. Evolution 39, 783–791. Pereira, S.L., Baker, A.J., 2006. A molecular timescale for Galliform birds accounting Garcia-Moreno, J., Sorenson, M.D., Mindell, D.P., 2003. Congruent avian phylogenies for uncertainty in time estimates and heterogeneity of rates of DNA inferred from mitochondrial and nuclear DNA sequences. J. Mol. Evol. 57 (1), substitutions across lineages and sites. Mol. Phylogenet. Evol. 38, 499–509. 27–37. Rambaut, A., Drummond, A.J., 2004. Tracer v1.3. Available from: . time scale, with special reference to Precambrian and Neogene. Episodes 27, Randi, E., Fusco, G., Lorenzini, R., Timothy, M.C., 1991. Phylogenetic relationships 83–100. and rates of allozyme evolution with the Phasianidae. Biochem. Syst. Ecol. 19, Groth, J.G., Barrowclough, G.J., 1999. Basal divergences in birds and the 213–221. phylogenetic utility of the nuclear RAG-1 gene. Mol. Phylogenet. Evol. 12, Randi, E., Lucchini, V., 1998. Organization and evolution of the mitochondrial DNA 115–123. control-region in the avian genus Alectoris. J. Mol. Evol. 47, 449–462. Hackett, S.J., Kimball, R.T., Reddy, S., et al., 2008. A phylogenomic study of birds Randi, E., Lucchini, V., Armijo-Prewitt, T., Kimball, R.T., Braun, E.L., Ligon, J.D., 2000. reveals their evolutionary history. Science 320, 1763–1767. Mitochondrial DNA phylogeny and speciation in the tragopans. Auk 117, 1003– He, L., Dai, B., Zeng, B., Zhang, X., Chen, B., Yue, B., Li, J., 2009. The complete 1015. mitochondrial genome of the Sichuan hill partridge (Arborophila rufipectus) and Randi, E., Lucchini, V., Hennache, A., Kimball, R.T., Braun, E.L., Ligon, J.D., 2001. a phylogenetic analysis with related species. Gene 435 (1–2), 23–28. Evolution of the mitochondrial DNA control region and cytochrome b genes and Hou, L.H., 1982. Fossil birds of pleistocene from Chou-kou-tien. Vert. Palasiatica 20, the inference of phylogenetic relationships in the avian genus Lophura 366–367. (Galliformes). Mol. Phylogenet. Evol. 19 (2), 187–201. Hudson, G.E., Parker, R.A., Van den Berge, Lanzillotti, P.J., 1966. A numerical analysis Shi, Y.F., Li, J.J., Li, B.Y., 1999. Uplift of the Qinghai–Xizang (Tibetan) plateau and east of the modifications of the appendicular muscles in the various genera of Asia environmental changes during late Cenozoic. Acta Geograph. Sinica 54 (1), gallinaceous birds. Am. Midl. Nat. 76, 1–73. 10–21 (in Chinese). Huelsenbeck, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogenetic Shibusawa, M., Nishibori, M., Nishida-Umehara, C., Tsudzuki, M., Masabanda, J., trees. Bioinformatics 17, 754–755. Griffin, D.K., Matsuda, Y., 2004. Karyotypic evolution in the Galliformes: an Johnsgard, P.A., 1988. The Quails, Partridges, and Francolins of the World. Oxford examination of the process of karyotypic evolution by comparison of the University Press, Oxford. molecular cytogenetic findings with the molecular phylogeny. Cytogenet. Johnsgard, P.A., 1999. The Pheasants of the World, second ed. Smithsonian Genome Res. 106 (1), 111–119. Institution Press, Washington, DC. Sibley, C.G., Ahlquist, J.E., 1990. Phylogeny and Classification of Birds. Yale Johnson, K.P., Sorenson, M.D., 1998. Comparing molecular evolution in two University Press, New Haven. mitochondrial protein coding genes (cytochrome b and ND2) in the dabbling Sorenson, M.D., Ast, J.C., Dimcheff, D.E., Yuri, T., Mindell, D.P., 1999. Primers for a ducks (tribe: Anatini). Mol. Phylogenet. Evol. 10, 82–94. PCR-based approach to mitochondrial genome sequencing in birds and other Kimball, R.C., Braun, E.L., Zwartjies, P.W., Crowe, T.M., Ligon, J.D., 1999. A molecular vertebrates. Mol. Phylogenet. Evol. 12, 105–114. phylogeny of the pheasants and partridges suggests that these lineages are not Swofford, D.L., 2003. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other monophyletic. Mol. Phylogenet. Evol. 11, 38–54. Methods), Version 4. Sinauer Associates, Sunderland, MA. Kimball, R.C., Braun, E.L., Ligon, J.D., Randi, E., Lucchini, V., 2006. Using molecular Thompson, J.D., Jeanmougin, F., Gouy, M., Higgins, D.G., Gibson, T.J., 1997. Multiple phylogenetics to interpret evolutionary changes in morphology and behavior in sequence alignment with Clustal X. Trends Biochem. Sci. 23, 403–405. the Phasianidae. Acta Zool. Sinica 52 (Suppl.), 362–365. Wen, L.Y., Zhang, L.X., Liu, N.F., 2005. Phylogenetic relationship of Perdix dauuricae Kimball, R.T., Braun, E.L., 2008. A multigene phylogeny of Galliformes supports a inferred from mitochondrial cytochrome b gene. Zool. Res. 26, 69–75. single origin of erectile ability in non-feathered facial traits. J. Avian Biol. 39, Zhan, X.J., Zhang, Z.W., 2005. Molecular phylogeny of avian genus Syrmaticus based 438–445. on the mitochondrial cytochrome B gene and control region. Zool. Sci. 22 (4), Kornegay, J.R., Kocher, T.D., Williams, L.A., Wilson, A.C., 1993. Pathways of lysozyme 427–435. evolution inferred from the sequences of cytochrome b in birds. J. Mol. Evol. 37, Zheng, G.M., 2005. A Checklist on the Classification and Distribution of the Birds of 367–379. China. Science Press, Beijing.