Molecular Phylogenetics and Evolution 66 (2013) 43–53 Contents lists available at SciVerse ScienceDirect Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev Molecular phylogeny of Cissus L. of Vitaceae (the grape family) and evolution of its pantropical intercontinental disjunctions Xiu-Qun Liu a, Stefanie M. Ickert-Bond b, Long-Qing Chen a, Jun Wen c,⇑ a Key Laboratory of Horticultural Plant Biology (Ministry of Education), College of Horticulture and Forestry Science, Huazhong Agricultural University, Wuhan 430070, PR China UA Museum of the North Herbarium and Department of Biology and Wildlife, University of Alaska Fairbanks, Fairbanks, AK 99775-6960, USA c Department of Botany, National Museum of Natural History, MRC166, Smithsonian Institution, Washington, DC 20013-7012, USA b a r t i c l e i n f o Article history: Received 17 June 2012 Revised 28 August 2012 Accepted 4 September 2012 Available online 18 September 2012 Keywords: Cissus Vitaceae Plastid DNA Molecular phylogeny Biogeographic diversification Pantropical intercontinental disjunction a b s t r a c t Pantropical intercontinental disjunct distribution is a major biogeographic pattern in plants, and has been explained mainly by boreotropical migration via the North Atlantic land bridges (NALB) and transoceanic long-distance dispersal (LDD), and sometimes by vicariance. However, well-resolved phylogenies of pantropical clades are still relatively few. Cissus is the largest genus of the grape family Vitaceae and shows a pantropical intercontinental disjunction with its 300 species distributed in all major tropical regions. This study constructed the phylogenetic relationships and biogeographic diversification history of Cissus, employing five plastid markers (rps16, trnL-F, atpB-rbcL, trnH-psbA and trnC-petN). The results confirmed that Cissus polyphyletic, consisting of three main clades: the core Cissus, the Cissus striata complex, and the Australian–Neotropical disjunct Cissus antarctica – C. trianae clade. The latter two clades need to be removed from Cissus to maintain the monophyly of the genus. The core Cissus is inferred to have originated in Africa and is estimated to have diverged from its relatives in Vitaceae in the late Cretaceous. It diversified in Africa into several main lineages in the late Paleocene to the early Eocene, colonized Asia at least three times in the Miocene, and the Neotropics in the middle Eocene. The NALB seems the most plausible route for the core Cissus migration from Africa to the Neotropics in the middle Eocene. Three African–Asian and two Neotropical–Australian disjunctions in Cissus s.l. are estimated to have originated in the Miocene and may be best explained by LDD. Published by Elsevier Inc. 1. Introduction Pantropical intercontinental disjunct distribution is a major biogeographic pattern in plants (Thorne, 1972; Givnish and Renner, 2004; Bartish et al., 2011). Thorne (1972) documented that 334 genera and 59 families of seed plants show pantropical distributions in all major tropical regions of the world. Compared with the disjunct temperate floras in the Northern Hemisphere (Wen, 1998; Manos and Donoghue, 2001; Wen and Ickert-Bond, 2009; Wen et al., 2010), tropical intercontinental disjunctions remain poorly understood perhaps due to the greater species richness, the inaccessibility of the study material (Renner et al., 2001), the greater ocean gulfs, and the less dynamic latitudinal distributions of taxa (Givnish and Renner, 2004). Recent molecular phylogenetic studies combined with molecular clocks now allow a more precise understanding of the evolution of pantropical distributions in several vascular plant families and genera such as in Melastomataceae (Renner et al., 2001; Renner, 2004), Malpighiaceae (Davis et al., 2002, 2004), Annonaceae (Doyle et al., 2004; Richardson et al., ⇑ Corresponding author. Fax: +1 202 786 2563. E-mail address: wenj@si.edu (J. Wen). 1055-7903/$ - see front matter Published by Elsevier Inc. http://dx.doi.org/10.1016/j.ympev.2012.09.003 2004; Erkens et al., 2009), Myristicaceae (Doyle et al., 2004), Burseraceae (Weeks et al., 2005), Moraceae (Zerega et al., 2005), Meliaceae (Muellner et al., 2006), Campanulaceae (Antonelli, 2009), Rubiaceae (Razafimandimbison et al., 2010; Smedmark et al., 2010), Simaroubaceae (Clayton et al., 2009), Sapotaceae (Bartish et al., 2011), and Diospyros (Duangjai et al., 2009). Three hypotheses have been invoked to explain how the tropical/pantropical lineages evolved to occupy such a wide disjunct geographic range. With the acceptance of plate tectonics theory, vicariance was used to explain the wide distribution of lineages on the ancient Gondwana continent (Schönenberger and Conti, 2003; Thorne, 2004; de Queiroz, 2005). Vicariance as the biogeographic mechanism has been proposed in Annonaceae (Doyle and Le Thomas, 1997; Doyle et al., 2004) and Meliaceae (Muellner et al., 2006). The boreotropical migration hypothesis supports the migration of some tropical lineages with intercontinental disjunctions between the Old and the New World via the North Atlantic land bridges (NALB) during the early Tertiary, when climate conditions in the Northern Hemisphere accommodated a tropical vegetation (Davis et al., 2002, 2004). This hypothesis has been proposed to explain many lineages with a classical western Gondwanan disjunct pattern such as in Burseraceae (Weeks et al., 2005), 44 X.-Q. Liu et al. / Molecular Phylogenetics and Evolution 66 (2013) 43–53 Malpighiaceae (Davis et al., 2002, 2004), and Guatteria (Annonaceae, Erkens et al., 2009). The transoceanic long distance dispersal (LDD) has been commonly proposed, especially when divergence times of many lineages are far too young to implicate vicariance via tectonic plate movement (Davis et al., 2004; Thorne, 2004) or the NALB. Examples of LDD have been reported in Melastomataceae (Renner, 2004), Simaroubaceae (Clayton et al., 2009), and Chrysophylloideae (Sapotaceae, Bartish et al., 2011). Nevertheless, well-resolved phylogenies of pantropical clades are still relatively few (Clayton et al., 2009). Cissus L. of Vitaceae (the grape family) shows a pantropical intercontinental disjunct pattern. Cissus contains approximately 300 species (Wen, 2007a) and represents the largest of the 14 genera of Vitaceae (Lombardi, 1997, 2007; Wen, 2007a; Wen et al., 2007). The genus has about 135 species in Africa, 85 species in Asia, 12 species in Australia, and 65 species in the Neotropics (Wen, 2007a). Cissus shows remarkable morphological diversity (Jackes, 1988; Lombardi, 2007), and is generally characterized by welldeveloped thick and undivided floral disks, four-merous flowers, one-seeded fruits, and seeds with a long and linear chalaza (Descoings, 1960; Wen, 2007a; Chen and Manchester, 2011). Jackes (1988) classified the Australian Cissus species into three groups. Lombardi (2007) placed the Neotropical Cissus species into 15 informal groups. The phylogenetic relationships of Cissus have been discussed in the context of Vitaceae phylogeny in several recent studies (Rossetto et al., 2001, 2002, 2007; Ingrouille et al., 2002; Soejima and Wen, 2006; Wen et al., 2007; Ren et al., 2011). These studies have shown that Cissus is polyphyletic, with most Cissus species belonging to a core clade; while four Neotropical Cissus (the Cissus striata complex) and four Australian Cissus (C. antarctica Vent., and its close relatives) did not form a clade with the core Cissus. We herein expand the sampling scheme in Cissus and conduct phylogenetic, molecular dating, and biogeographic analyses to reconstruct the evolutionary diversification history of Cissus based on five plastid markers (rps16, trnL-F, atpB-rbcL, trnH-psbA and trnC-petN). Our sampling included taxa from all major tropical regions. The aims of this study are to: (1) reconstruct the Cissus phylogeny; (2) infer the ancestral area of Cissus; and (3) test competing hypotheses on the evolution of pantropical disjunctions. 2. Materials and methods 2.1. Sampling, DNA isolation and sequencing The study sampled 174 accessions representing 117 accessions of Cisuss (including 74 species) and 57 accessions (55 species) of related taxa of Vitaceae and Leeaceae (the sister family of Vitaceae; Wen, 2007b) and generated sequences for trnL-F, the rps16 intron, atpB-rbcL, trnH-psbA and trnC-petN (Appendix Table A1). The sampling covered the geographic and morphological diversity of Cissus, with 16 species from Asia, 34 from Africa including Madagascar, 22 from the Neotropics, and two from Australia. We also included representatives of twelve other genera of Vitaceae in this study. Three Leea species (Leeaceae) were selected as outgroups. Total DNAs were extracted from silica-gel-dried leaves or herbarium material by using a modified CTAB method (Doyle and Doyle, 1987) or the DNeasy Plant Mini Kit (Qiagen, Mississauga, Ontario, Canada) following the manufacturer’s protocol. The trnLF region was amplified and sequenced using primers c and f (Taberlet et al., 1991). When amplification of the trnL-F region was unsuccessful, we used primer combinations of c and d, and e and f0 (Taberlet et al., 1991; Soejima and Wen, 2006; Chen et al., 2011a). The rps16 intron was amplified and sequenced using primers F and R2 (Oxelman et al., 1997; Andersons and Rova, 1999). When primers F and R2 failed to sequence a few Cissus species, we used primer combinations of P3F (50 -TGC TCT TGG CTC GAC ATC G-30 ) and P2R (50 -GCG TTT CCT TGT TCC GGG-30 ), and V1F (Chen et al., 2011a) and R2. The atpB-rbcL, trnH-psbA and trnC-petN regions were amplified and sequenced following Manen et al. (1994), Lee and Wen (2004), and Shaw et al. (2005), respectively. PCR products were purified with the polyethylene glycol (PEG) precipitation method (Wen et al., 2007). DNA sequences were assembled using the program Sequencher version 4.1.4 (Gene Codes Corp., Ann Arbor, Michigan, USA). 2.2. Sequence alignment and phylogenetic analysis Sequence alignment was initially performed using the program MUSCLE 3.8.31 (Edgar, 2004) in multiple alignment routine, followed by manual adjustment with the program Se-Al version 2.0a11 (Rambaut, 2002). Phylogenetic trees of the combined matrix of the five plastid DNA markers were reconstructed using maximum parsimony (MP, Fitch, 1971), maximum likelihood (ML) and Bayesian inference (BI) (Rannala and Yang, 1996; Mau et al., 1999). MP analyses were conducted under the heuristic search option using 10 random stepwise additions and tree-bisection-reconnection (TBR) branch swapping in PAUP⁄ version 4.0 b10 (Swofford, 2003). Zero-length branches were collapsed and gaps were treated as missing data or coded as simple indels (Simmons and Ochoterena, 2000) using the program SeqState (Müller, 2005). Parsimony bootstrap analyses (Felsenstein, 1985) with 1000 replicates were subsequently performed under the option fast and stepwise addition to evaluate the robustness of the MP trees. MrModeltest (Nylander, 2004) was used to determine the best available model for nucleotide substitutions. The generalized time reversible model (GTR + I + G model) was suggested as the best-fit model of sequence evolution for the combined plastid dataset. In the ML and BI analyses, we used the substitution models and parameters as estimated from MrModeltest. Bayesian inference was used to estimate the posterior probabilities of phylogenetic trees by employing an analysis of five million generations Metropolis-coupled Markov chain Monte Carlo (MCMC) with MrBayes version 3.1.2 (Huelsenbeck and Ronquist, 2001). Different sequences were partitioned with unlinked substitution models as estimated before. The sampling rate of the trees was 1000 generations. The Bayesian trees sampled for the last four million generations were used to construct a 50%-majority rule consensus tree after discarding the first 10% samples as burn-in. The proportion of bifurcations found in this consensus tree was given as posterior clade probabilities (PP) as an estimator of the robustness of the BI trees. 2.3. Bayesian dating, fossil calibration The combined rps16, trnL-F, atpB-rbcL, trnH-psbA and trnC-petN matrix was used to estimate the divergence times of clades. Representatives of Cissus from all main clades were sampled for the dating analyses with a fossil calibration. We largely followed the dating strategies in analyzing diversification of Parthenocissus (Nie et al., 2010) and Ampelopsis (Nie et al., 2012) of Vitaceae. The Program BEAST version 1.6.1 (Drummond and Rambaut, 2007) was used to date divergence times and employed a Bayesian relaxed clock model. After optimal operator adjustment as suggested by the output diagnostics from several preliminary BEAST runs, two final independent runs (each 50 million generations) were performed on a cluster of Mac XServes used for analysis of biological data at the Smithsonian Institution (http://topazweb.si.edu). Convergence between runs was assessed with MrBayes using Tracer version 1.5. After discarding the first 10% X.-Q. Liu et al. / Molecular Phylogenetics and Evolution 66 (2013) 43–53 samples as burn-in, the trees and parameter estimates from the two runs were combined by LogCombiner 1.6.1 (Drummond and Rambaut, 2007). Results were considered reliable once the effective sample size (ESS) for all parameters exceeded 200 as suggested by Drummond et al. (2007). The samples from the posterior were summarized on the posterior probabilities on its internal nodes (Drummond et al., 2007) using the program TreeAnnotator version 1.6.1 (Drummond and Rambaut, 2007) with posterior probability limit set to 0.5 and summarizing mean node heights. These were visualized using the program FigTree version 1.2.2 (Drummond et al., 2007). Mean and 95% highest posterior density (HPD) of age estimates were obtained from the combined outputs using Tracer version 1.5. Possible vitaceous fossils were reported from as early as the early Cenomanian (99.6–93.5 Ma) (the Paleobiological Database, 2010; Zecca et al., 2012), but the taxonomic attributions of the oldest fossils are very controversial (Chen and Manchester, 2007; Chen, 2009). Many Vitaceae fossils are available from sediments from the early Eocene to the Pleistocene and include leaves, pollen, stems, and seeds (Greguss, 1969; Tiffney and Barghoorn, 1976; Wheeler and Lapasha, 1994; Chen and Manchester, 2007). Nevertheless, the seed record of the family is potentially more informative for addressing questions of evolutionary and phytogeographic divergences, because these fossils can be differentiated at the generic level (Chen and Manchester, 2007; Chen, 2009). The oldest confirmed vitaceous seed fossil is undoubtedly assigned to Ampelocissus s.l. (A. parvisemina) and dates back to the late Paleocene in North Dakota of North America (Chen and Manchester, 2007; Wang et al., 2009). Nie et al. (2012) conservatively considered the A. parvisemina fossil as representing an early member of the Ampelocissus – Nothocissus – Pterisanthes – Vitis – Parthenocissus – Yua clade because of the poorly supported deep relationships within the clade. In the current study, the Ampelocissus – Nothocissus – Pterisanthes – Vitis – Parthenocissus – Yua clade was resolved to two strongly supported clades: (1) the Ampelocissus – Nothocissus-Pterisanthes – Vitis clade (clade I) and (2) the Parthenocissus – Yua clade (clade II). Yet the relationships among Ampelocissus species were poorly resolved. Because the seed fossil A. parvisemina was assigned with certainty to the genus Ampelocissus (Chen and Manchester, 2007), it seems reasonably conservative to assign the fossil as representing an early member of the former clade (clade I). The stem of the clade was thus fixed at 58.5 ± 5.0 million yeas ago (Ma) (Nie et al., 2010, 2012). Recently, Gong et al. (2010) reported several fossil Vitis seeds including three morphotaxa from the Gray fossil site in Washington County, northeastern Tennessee (7–4.5 Ma, latest Miocene to earliest Pliocene). With the very good preservation, the fossil seeds were assigned to subg. Vitis. One of the three morphotaxa, V. latisulcata, is closely comparable to V. labrusca from eastern North America (Gong et al., 2010). Hence, we adopted a second calibration point of V. labrusca and its closely related North American relatives in subg. Vitis at 5.75 ± 0.5 Ma (nearly matching 7–4.5 Ma). Herrera et al. (2012) described a new genus Saxuva Herrera, Manchester et Jaramillo based on fossil seeds in Vitaceae from the late Eocene of Panama. Its type species S. draculoidea Herrera, Manchester et Jaramillo seems to be closely related to taxa of Vitaceae with four-merous flowers. Although the taxon exhibited characters seen in the modern genera Cissus, Cayratia, and Cyphostemma (Herrera et al., 2012), it seems unsuitable for our calibration strategy due to its uncertain phylogenetic position. For the root age of Vitaceae, Nie et al. (2010) and Zecca et al. (2012) fixed the split between Vitaceae and Leea as 85 ± 4.0 Ma based on the estimated age of 78–92 Ma by Wikström et al. (2001). Magallón and Castillo (2009) reported a pre-Tertiary origin at 90.65–90.82 Ma for Vitaceae. The estimated ages from Magallón and Castillo (2009) and Wikström et al. (2001) are close, but the latter was criticized for using nonparametric rate smooth- 45 ing and for calibrating the tree using only a single calibration point (Nie et al., 2012). We herein use the estimate from Magallón and Castillo (2009) and set the normal prior distribution of 90.7 ± 1.0 Ma for the stem age of the family. 2.4. Ancestral area reconstruction To reconstruct the geographical diversification of Cissus, we used maximum likelihood (ML) optimization in Lagrange 2.0.1 (Ree and Smith, 2008). The program not only finds the most likely ancestral area at a node and the split of the areas in the two descendant lineages, it also calculates the probabilities of these most-likely areas at each node (Ree and Smith, 2008), using an ultrametric tree combining the ML topology with internal node age estimates from a BEAST analysis based on the combined data set. All combinations of areas were allowed in the adjacency matrix, and baseline rates of dispersal and local extinction were estimated. The maximum number of areas in ancestral ranges was set to two in Lagrange, as no species of Vitaceae is distributed in more than two areas of endemism (Chen et al., 2011b). Four areas of endemism were defined according to the pantropical distribution of Cissus: A = Asia, B = Neotropics (Central and South America including the southern border of North America), C = Africa (including Madagascar), and D = Australia. 3. Results 3.1. Phylogenetic analysis of Cissus The aligned lengths of the rps16 intron, trnL-F, atpB-rbcL, trnHpsbA and trnC-petN datasets were 989, 1283, 1017, 831, and 1351 bp, respectively. The combined plastid dataset was 5471 bp in length, containing 3488 constant sites, 725 variable sites but parsimony-uninformative, and 1258 parsimony-informative sites. Trees generated with different analytical methods (MP, ML, and BI) were consistent with respect to various clades in Cissus. Therefore, only the BI strict consensus cladogram with bootstrap and Bayesian support values is shown (Fig. 1A and B). The parsimony search of the combined dataset yielded more than 100,000 most parsimonious trees (consistency index CI = 0.52, retention index RI = 0.89). The strict consensus tree of the combined dataset corresponded to the majority-rule consensus of 7500 trees (10,000 trees minus 2500 as burn-in derived from the BI analysis (Fig. 1A and B). Within Vitaceae, six strongly supported clades were recognized: the Ampelocissus – Nothocissus – Pterisanthes – Vitis clade (BS = 95%, PP = 1.00; clade I in Fig. 1A), the Parthenocissus – Yua clade (BS = 99%, PP = 1.00; clade II in Fig. 1A), the Ampelopsis – Cissus striata complex – Clematicissus – Rhoicissus clade (BS = 99%, PP = 1.00; clade III in Fig. 1A), the core Cissus clade (BS = 100%, PP = 1.00; clade IV in Fig. 1A and B), the Australian–Neotropical Cissus antarctica – C. trianae clade (BS = 100%, PP = 1.00; clade V in Fig. 1A), and the Cayratia – Tetrastigma – Cyphostemma clade (BS = 100%, PP = 1.00; clade VI in Fig. 1A). The core Cissus was also unambiguously supported as monophyletic (BS = 100%; PP = 1.00). The three African Cissus species (C. sagittifera Desc., C. floribunda (Baker) Planch. and C. integrifolia (Baker) Planch) were placed at the basal positions of the core Cissus clade. Besides the three species at basal positions, the core Cissus clade contained two subclades: subclade I (BS = 92%; PP = 1.00; Fig. 1B) and subclade II (BS = 78%; PP = 1.00; Fig. 1B). While subclade I is composed of taxa from the Old World only (Africa and Asia), subclade II contains species from both the Old and New Worlds. Within the core Cissus clade, species from the Neotropics formed a monophyletic group (BS = 86%; PP = 1.00; Fig. 1B). 46 X.-Q. Liu et al. / Molecular Phylogenetics and Evolution 66 (2013) 43–53 1.00 99 1.00 1.00 98 1.00 88 1.00 95 99 1.00 100 1.00 100 1.00 100 0.97 1.00 79 99 1.00 0.98 100 1.00 1.00 72 86 1.00 96 1.00 1.00 89 100 99 1.00 100 1.00 Parthenocissus chinensis NM455 Parthenocissus henryana NM359 Parthenocissus suberosa NM358 Parthenocissus tricuspidata NM355 Parthenocissus quinquefolia W8684 Yua austro-orientalis SIB1313 Yua thomsoni NM469 Ampelopsis bodinieri R55193 Ampelopsis cordata W7141 Ampelopsis cantoniensis W10242 Ampelopsis rubifolia W9285 Cissus granulosa W8573 Cissus granulosa W8596 Cissus granulosa W8611 Cissus striata NW53854 D1 Cissus striata NW53883 Cissus striata NW53887 Cissus striata W7355 A348 Cissus striata W7424 A352 Cissus simsiana NW53805 Clematicissus opaca S11005 Rhoicissus tomentosa 19656252 Rhoicissus tridentata LL11453 Core Cissus (see Fig. 1B) 100 1.00 1.00 0.55 100 76 1.00 100 1.00 100 1.00 100 1.00 100 1.00 97 1.00 100 1.00 100 1.00 1.00 1.00 100 100 100 1.00 100 0.86 50 1.00 98 Cissus antarctica W6684 Cissus antarctica W6685 Cissus trianae W8585 Cissus hypoglauca W12185 D2 Cayratia cordifolia W10548 Cayratia mollissima W8403 Cayratia imerinensis W9571 Cyphostemma adenocaule LL11459 Cyphostemma jiguu LL11551 Cyphostemma duparquetii LU11534 Cyphostemma simulans Gernath s.n. Cyphostemma maranguense LL11468 Cyphostemma montagnacii W6672 Cayratia japonica SH81847 Cayratia trifolia W10167 Tetrastigma hemsleyanum NM451 Tetrastigma lanyuense W9404 Tetrastigma pachyphyllum W8319 Tetrastigma obtectum NM454 Tetrastigma triphyllum NM342 Leea guineensis W9408 Leea indica W10910 Leea indica W8341 Leea macrophylla R55105 Clade II 77 Clade III 1.00 Clade V Clade IV 1.00 1.00 68 95 1.00 99 1.00 58 100 0.99 1.00 83 1.00 Ampelocissus acapulcensis W8696 Ampelocissus elephantina W9583 Ampelocissus erdwendbergii W8702 Vitis betulifolia W9308 Vitis flexuosa W10647 Vitis mengziensis NM415 Vitis heyneana W9378 Vitis popenoei W8724 Vitis rotundifolia W9972 Nothocissus spicifera W8384 Pterisanthes eriopoda W8336 Pterisanthes heterantha W8415 Pterisanthes stonei W8346 Clade VI Neotropics Australia Outgroup A total of 100 million generations (two runs of 50 million generations each) were necessary to reach sufficient ESS. Bayesian estimation of divergence times of Cissus is presented in Fig. 2. When the stem of the Ampelocissus – Pterisanthes – Nothocissus – Vitis subclade was constrained at 58.5 ± 5.0 Ma (Nie et al., 2010, 2012), the core Cissus clade was estimated to have diverged from its closest relative in Vitaceae at 72.6 (95% HPD: 63.2–84.0) Ma in the late Cretaceous (node 1 in Fig. 2). The core Cissus diverged initially at 61.6 (95% HPD: 50.5–72.4) Ma in the Paleocene in Africa (node 2 in Fig. 2). Subsequent diversification in Africa was at 55.8 Clade I (95% HPD: 44.7–66.3) Ma in the early Eocene (node 3 in Fig. 2), which showed that subclade I (strictly Old World taxa) and subclade II (Old and New World taxa) diverged from their close relatives in Africa (C. sagittifera, C. floribunda and C. integrifolia) during that time. The divergence of subclades I and II was estimated to have occurred at 47.2 (95% HPD: 37.1–57.2) Ma (node 4 in Fig. 2). The monophyletic Neotropical core Cissus group and the African C. diffusiflora (Baker) Planch. split at 36.1 (95% HPD: 27.2–44.5) Ma in the late Eocene (node 5 in Fig. 2). The Neotropical Cissus striata complex was nested within the Ampelopsis clade and it diverged from its closest relative at 21.8 (95% HPD: 10.7–35.0) Ma in the early middle Miocene (node 6 in Fig. 2). The Australian 3.2. Bayesian estimation of divergence times of Cissus Fig. 1. Phylogenetic relationships of Cissus and its relatives from Vitaceae using the combined plastid dataset (rps16, trnL-F, atpB-rbcL, trnH-psbA and trnC-petN) based on the BI strict consensus cladogram with bootstrap and Bayesian support. A, Cissus species were scattered in three clades (III–V) among the six major clades (Clades I–VI) in Vitaceae. D1 and D2 represent two Neotropical–Australian disjunctions. B, the core Cissus clade (clade IV). D3, D4, and D5 represent three Asian–African intercontinental disjunctions, and D6 represents the African–Neotropical disjunction in the core Cissus clade. Numbers above branches are Bayesian posterior probability values (PP), and numbers below branches indicate Bootstrap values (BS). Bold branches represent PP = 1.00 and BS > 90%. X.-Q. Liu et al. / Molecular Phylogenetics and Evolution 66 (2013) 43–53 47 Fig. 1. (continued) Cissus clade diverged from its relative of Vitaceae at 74.4 (95% HPD: 63.2–84.0) Ma in the late Cretaceous (node 7 in Fig. 2) and diversified at 19.8 (95% HPD: 7.1–34.8) Ma in the Miocene (node 8 in Fig. 2). The disjunction between the Neotropical C. trianae and the Australian C. antarctica was estimated at 16.1 (95% HPD: 6.5– 29.1) Ma in the Miocene (node 9 in Fig. 2). 3.3. Ancestral area reconstruction of core Cissus Lagrange reconstructed the ancestral area of the core Cissus in Africa (C|C with 0.89 relative probability, node a in Fig. 3). The ancestral area of the C. striata complex was inferred in the Neotropics (D|B with 1.00 relative probability, node b in Fig. 3). The ances- 48 X.-Q. Liu et al. / Molecular Phylogenetics and Evolution 66 (2013) 43–53 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0Ma 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0Ma Fig. 2. Chronogram of Cissus and its relatives from Vitaceae based on the combined plastid dataset (rps16, trnL-F, atpB-rbcL, trnH-psbA and trnC-petN) inferred from BEAST. Gray bars represent the 95% highest posterior density credibility interval for node ages. Calibration points are indicated with stars. Estimated divergence times of Cissus and its relatives from Vitaceae are indicated at the nodes (nodes 1–13) using circles and the estimated ages are shown at the left. 49 X.-Q. Liu et al. / Molecular Phylogenetics and Evolution 66 (2013) 43–53 Outgroup Clade VI Clade II Clade I c C|C(0.92) C|C(0.99) C|C(0.99) C|C(0.80) C|BC(0.12) C|A(1.00) A|C(1.00) AC|C(0.96) AC|A(0.90) A|A(1.00) C|C(0.48) AC|C(0.26) C|C(0.88) d B|B(1.00) C|C(0.89) a e B|C(1.00) C. hypoglauca W12185 C. antarctica W6684 C. antarctica W6685 C. trianae W8585 C. sciaphila LL11477 C. rhodotrichus W9652 C. rostrata W7501 C. adnata W10268 C. quarrei LL11581 C. leucophleus W9676 C. lanea W9536 C. auricoma W9549 C. madecassa W9597 C. microdonta W9635 C. aralioides RE2048 C. polita W9577 C. welwitschii KI3 C. rotundifolia LL11478 C. oliveri F2237 C. pseudoguerkeana C. cactiformis LL11480 C. quadrangularis W7368 C. phymatocopa LL11474 C. quadrangularis W9538 C. subtetragona W10921 C. hastata W10993 C. javana SH81970 C. elongata W6639 C. repanda W9027 C. repanda W7396 C. cornifolia LL1625 C. cornifolia A1613 C. producta LU11528 C. pileatus W9582 C. rubiginosa K2242 C. assamica N362 C. repens R55094 C. hastata W7509 C. nodosa W10713 C. discolor W7468 C. wenshanensis SH81897 C. gongylodes NW53777 C. rhombifolia W8591 C. microcarpa W11954 C. ulmifolia NW53899 C. obliqua NW53861 C. erosa W8586 C. tuberosa W2010-091 C. verticillata NW53879 C. anisophylla W6999 C. pseudoverticillata C. amazonica W2010-099 C. verticillata W10165 C. trifoliata W9727 C. trifoliata W7287 C. diffusiflora J1813 Core Cissus Clade D|BD(0.96) C. integrifolia LL11475 C|BC(0.55) C|C(0.20) D|B(1.00) b B|B(1.00) C. floribunda W9463 C. sagittifera W9605 Clematicissus opaca C. granulosa W8611 C. striata NW53887 C. striata W7424 C. simsiana NW53805 Ampelopsis cantoniensis Ampelopsis rubifolia Ampelopsis cordata Ampelopsis bodinieri Rhoicissus tridentata Rhoicissus tomentosa A C B D Fig. 3. Results of the Lagrange analyses of Cissus and its close relatives in Vitaceae. The tree was based on a 50% majority-rule consensus tree of a Bayesian Markov chain Monte Carlo (MCMC) analysis of the combined plastid dataset. The four areas of endemism are: A, Asia (purple); B, Neotropics (orange); C, Africa (blue); and D, Australia (green). Colored circles at the tip of the nodes indicate species distributions as seen in the map below. For the Lagrange results, a slash indicates the split of areas into two daughter lineages, i.e., left/right, where ‘‘up’’ and ‘‘down’’ are the ranges inherited by each descendant branch. The values in brackets represent relative probabilities. Nodes a, b, c, d and e show the ancestral areas of the core Cissus clade, C. striata complex, the C. antarctica – C. trianae clade, the Neotropical core Cissus – C. diffusiflora clade, and Neotropical core Cissus. 50 X.-Q. Liu et al. / Molecular Phylogenetics and Evolution 66 (2013) 43–53 tral area of the Australian–Neotropical Cissus antarctica – C. trianae clade was inferred to be in Australia (D|BD with 0.96 relative probability, node c in Fig. 3). 4. Discussion 4.1. Polyphyly and major clades of Cissus The phylogenetic results presented here are the most comprehensive for Cissus to date. This study confirmed the findings by Rossetto et al. (2002, 2007) that Cissus is polyphyletic. Our results with an expanded sampling scheme also support three major clades of Cissus (Fig. 1A and B): the core Cissus clade (clade IV in Fig. 1A and B), the Cissus striata complex nested within clade III (Fig. 1A) and an Australian–Neotropical Cissus antarctica – C. trianae clade (clade V in Fig. 1A). The core Cissus clade included all species from the Old World (Asia and Africa), and most species from the Neotropics (Fig. 1B). It was unambiguously supported to be monophyletic (BS = 100%, PP = 1.00; Fig. 1B), which corroborates previous studies (Rossetto et al., 2002, 2007; Wen et al., 2007; Ren et al., 2011). Three African Cissus species (C. sagittifera and C. floribunda from Madagascar, and C. integrifolia from Kenya) were found to occupy the basalmost positions within the core Cissus clade (Fig. 1B). As previously suggested by Rossetto et al. (2007), the core Cissus clade most likely represents the ‘‘true’’ Cissus clade, because taxa in this clade share the morphological characteristics associated with the type species C. vitiginea L. of the genus Cissus, such as leaf-opposed umbellate cymes, four-merous flowers, cupular floral disk raised above the ovary, and monospermic fruits (Wen et al., 2007). Seeds of core Cissus are characterized by long and linear chalazas visible from the ventral side and terminated very near the beak at the dorsal side, a condition termed ‘‘perichalaza’’ (Chen and Manchester, 2011). The Cissus striata complex from the Neotropics (Fig. 1A) was confirmed to be a distinct clade from the core Cissus as suggested in previous analyses (Rossetto et al., 2007; Wen et al., 2007; Ren et al., 2011; Nie et al., 2012). We sampled three species from this complex: C. striata, C. simsiana Roem. and Schult., and C. granulosa Ruiz. and Pav. Based on morphology (Lombardi, 2000, 2007), C. tweediana (Baker) Planch. is part of this complex, but our samples of this taxon failed to amplify for the markers used. The C. striata complex from the Neotropics is shown to be closely related to the Australian Clematicissus opaca (F. Muell.) Jackes & Rossetto. This sister relationship was also supported in our analyses (BS = 99%, PP = 1.00; Fig. 1A). The C. striata complex in turn was nested within the paraphyletic Ampelopsis Michx. and showed a close relationships with the Australian Clematicissus and African Rhoicissus Planch., which were confirmed to form a well-supported clade of Ampelopsis within the phylogeny of Vitaceae (BS = 99%, PP = 1.00; clade III in Fig. 1A), as reported by Soejima and Wen (2006) and Wen et al. (2007). The results suggest that the C. striata complex should be removed from Cissus, which was also supported by previous analyses of molecular data (Rossetto et al., 2007; Wen et al., 2007). Members of the C. striata complex are morphologically distinct from core Cissus and united by the presence of short unbranched trichomes, 1–4 seeded purple fruits, revolute inflorescence branches and adnate stipules (Lombardi, 2007). Their seeds are very similar to those of Ampelopsis, which are typically small and have an oval chalaza and short ventral infolds, and lack of perichalaza (Chen and Manchester, 2011). The two Australian Cissus species (C. antarctica and C. hypoglauca A. Gray) also form a clade distinct from the core Cissus, but group with the Neotropical C. trianae (Fig. 1A). Two additional Australian Cissus species (C. oblonga (Benth.) Planch. and C. sterculiifolia (F. Muell. ex Benth.) Planch.) and a few taxa in New Guinea should also be included in this clade according to the analysis of Rossetto et al. (2002). Several morphological characters are shared among this group of Cissus species, such as four-merous flowers (Gerrath and Posluszny, 1994), the similar shape of the two cotyledons (rather than one reniform and the other cordiform as in most other Australian core Cissus) and their nearly entire leaflet margins (Rossetto et al., 2007). Their seeds have linear chalazas, long ventral infolds, rugose surfaces, and lack of perichalaza (Chen and Manchester, 2011). The four Australian Cissus species were proposed to be segregated from the traditional Cissus and interpreted to belong to a distinct genus based on chromosome numbers, DNA variation, presence of supernumerary buds, degree of stipule connectivity, uncommitted primordia, and inflorescence branching type (Rossetto et al., 2002, 2007; Timmons et al., 2007). Our results support a new intercontinental disjunct genus between Australia and the Neotropics in Vitaceae. 4.2. African origin and diversification of the core Cissus Based on our divergence time estimates, the core Cissus clade diverged from its close relative in Vitaceae in the late Cretaceous estimated at 72.6 (95% HPD: 63.2–84.0) Ma (node 1 in Fig. 2). But we cannot exclude a possibility that the core Cissus clade may have diverged earlier than our inference from its close relatives in Vitaceae, because the vitaceous fossil record dates back to the early Cenomanian (93.5–99.6 Ma) (The Paleobiological Database, 2010), although the taxonomic affinity of the possible oldest fossils to Vitaceae is controversial (Chen and Manchester, 2007; Chen, 2009; Zecca et al., 2012). Lagrange suggests the ancestral area of the core Cissus in Africa (C|C with 0.89 relative probability, node a in Fig. 3). The core Cissus crown group (node 2 in Fig. 2) is estimated at 61.6 (95% HPD: 50.5–72.4) Ma in the Paleocene. Then the African ancestors diversified into two main lineages at 47.2 (95% HPD: 37.1– 57.2) Ma in the Eocene (node 4 in Fig. 2). The core Cissus then colonized Asia at least three times at 7.8 Ma (95% HPD: 3.0–15.1 Ma; node 10 in Fig. 2), 25.6 Ma (95% HPD: 17.4–34.1 Ma; node 11 in Fig. 2), and 21.7 Ma (95% HPD: 12.1–33.0 Ma; node 12 in Fig. 2) in the Miocene. The biogeographic scenario that the core Cissus originated and first diversified in Africa is consistent with the greatest extant diversity and the highest endemism in Africa. Lagrange suggests the Neotropical-Cissus diffusiflora clade originated in Africa (C|C and AC|C with 0.48 and 0.26 relative probabilities, respectively, node d in Fig. 3). The monophyletic Neotropical core Cissus group formed a clade with the African C. diffusiflora (Fig. 1B, D6; node e in Fig. 3), but the support for this intercontinental disjunct clade was low (PP < 0.85, BS < 50%). Thus, it may be unreliable that the core Cissus group was inferred to have colonized the Neotropics based on the split time of the NeotropicalAfrican clade at 36.1 (95% HPD: 27.2–44.5) Ma in the late Eocene (node 5 in Fig. 2). Nevertheless, the next node (node d in Fig. 3) was estimated to be at 40.4 (95% HPD: 30.8–50.2) Ma in the middle Eocene (node 13 in Fig. 2). Although vicariance is a plausible scenario for some amphi-Atlantic tropical disjunct lineages like calamoid palms (Baker and Dransfield, 2000), it is apparent that many such plant lineages originated and diversified well after the last direct connection between Africa and the Neotropics at ca. 105 Ma (Magallón and Sanderson, 2001; McLoughlin, 2001; Wikström et al., 2001; Davis et al., 2002, 2004; Richardson et al., 2004). Since our dating analysis estimates core Cissus diversified in the Paleocene, vicariance cannot account for its distribution. Rather, the scenario of the core Cissus migration from Africa to the Neotropics through the NALB seems most plausible. The climate during the time span from the late Paleocene to early Eocene was the warmest globally of the Tertiary. The most drastic cooling after the thermal maximum did not occur until the beginning of the Oligocene (Wolfe, 1975; Zachos et al., 2001). X.-Q. Liu et al. / Molecular Phylogenetics and Evolution 66 (2013) 43–53 Boreotropical vegetation existed at much higher latitudes (50– 60°N) and floristic exchanges may have occurred frequently between the Old and the New World in the Northern Hemisphere during that time (Tiffney and Manchester, 2001). Several studies have supported the importance of the NALB in explaining the global distribution of tropical plant taxa (Annonaceae, Doyle and Le Thomas, 1997; Lauraceae, Chanderbali et al., 2001; Malpighiaceae, Davis et al., 2002, 2004; Burseraceae, Weeks et al., 2005; Meliaceae, Muellner et al., 2006; Guatteria, Erkens et al., 2009). Nevertheless, we cannot exclude the hypothesis of LDD between Africa and the Neotropics. LDD has been viewed as a dominant mechanism of distribution of many relatively young tropical plant lineages such as in Melastomataceae (Renner, 2004), Simaroubaceae (Clayton et al., 2009), and Chrysophylloideae (Bartish et al., 2011). We have detected at least three African–Asian disjunctions within the core Cissus clade (Fig. 1B, D3, D4, D5; Fig. 3), with one of them (between the African C. cornifolia (Baker) Planch. and the Asian C. repanda Vahl.) only poorly supported (PP < 0.50, BS < 50%). The African–Asian disjunctions are common among paleotropical taxa, which have been explained using four hypotheses (Zhou et al., 2012). The first is the rafting of the Indian subcontinent carrying with it tropical plant elements from Africa. The Indian plate separated from Gondwana in the early Cretaceous, drifted northward in the middle Cretaceous, and collided with the Asian plate in the middle Eocene (Morley, 2003; Smedmark et al., 2010). Conti et al. (2002) supported this out-of-India hypothesis in Crypteroniaceae. At the time of the collision, a corridor of tropical climate existed from India to Southeast Asia, allowing the dispersal of many rain forest taxa (Morley, 2003). However, the divergence times in the Miocene of the African–Asian disjuncts in Cissus are too young to be explained by the hypothesis. There were few land connections in the region between Africa and Asia during the Cenozoic (Smedmark et al., 2010). Secondly, the observed African–Asian disjunction in core Cissus may be explained by dispersal from previously widespread boreotropical forests in southern Europe and Central Asia (Zachos et al., 2001; Zhou et al., 2012). This hypothesis has been proposed to explain the divergence of African–Asian taxa during the late Paleocene to early Eocene associated with the thermal maximum (Zachos et al., 2001) in several angiosperm families: Burseraceae (Weeks et al., 2005), Malpighiaceae (Davis et al., 2002), and Meliaceae (Muellner et al., 2006). We inferred the African–Asian migration from 25.6 to 7.8 Ma (nodes 10, 11 and 12 in Figs. 2 and 3), which is also too young to be explained by this scenario. Thirdly, Zhou et al. (2012) suggested an alternative scenario, which involved dispersal across Arabia and Central Asia via the tropical forests that developed during the late Middle Miocene thermal maximum (17–15 Ma) as seen in Uvaria (Annonaceae), which has fleshy fruits that are consumed by primates. Primates dispersed from Africa to Eurasia in the early to middle Miocene (Fleagle and Gilbert, 2006). However, the Miocene tropical forests were less extensive than those of the Eocene, and were unlikely to have formed a continuous pathway that would have facilitated widespread migration of tropical taxa (Zhou et al., 2012). This scenario seems quite unlikely to explain the Asian-African disjunction in Cissus. Finally, LDD has also been invoked as an explanation for dispersal from Africa to Asia or in the opposite direction (Gaertnera (Rubiaceae, Malcomber, 2002); Osbeckia (Melastomataceae, Renner 2004); Exacum (Gentianaceae, Yuan et al., 2005); Macaranga and Mallotus (Euphorbiaceae, Kulju et al., 2007); Brucea, Eurycoma and Soulamea (Simaroubaceae, Clayton et al., 2009); Bridelia (Phyllanthaceae, Li et al., 2009); and Ranunculus (Ranunculaceae, Emadzade et al., 2011)). Cissus species bear blackish purple, fleshy and bird-dispersed fruits that may have facilitated long distance dispersal. 51 4.3. Two Neotropical–Australian disjunctions in non-core Cissus Our results support two Neotropical–Australian disjunctions in the non-core Cissus groups (Fig. 1A, D1 and D2). One disjunction is between the Neotropical Cissus striata complex and the Australian Clematicissus opaca (formerly assigned to Cissus, namely ‘Cissus opaca’; Jackes and Rossetto, 2006). This Cissus striata – Clematicissus alliance is nested within Ampelopsis (Fig. 1A, D1). This disjunction was inferred to have involved migration from North America to the Neotropics and finally to Australia (Nie et al., 2012). A second Australian–Neotropical disjunction is exhibited by C. trianae (Neotropics) and C. antarctica (Australia) and the close allies (Fig. 1A, D2). Both of these Neotropical–Australian splits (the C. striata complex – Clematicissus group, as well as the C. trianae – C. antarctica group) were estimated to have occurred in the Miocene with the former at 21.8 (95% HPD, 10.7–35.0) Ma (node 6 in Fig. 2) and the latter at 16.1 (95% HPD, 6.5–29.1) Ma (node 9 in Fig. 2). Trans-Antarctica exchange (Sanmartín and Ronquist, 2004) has been used to explain disjunctions in tropical plant groups of Annonaceae (Richardson et al., 2004) and Sapotaceae (Bartish et al., 2011). Yet the Antarctic route existed during the late Cretaceous-early Tertiary and was interrupted in the late Eocene (30– 35 Ma). The relatively young age estimates of the Neotropical–Australian disjunctions in Cissus corroborates the hypothesis by Nie et al. (2012) that the Cissus striata complex arrived in Australia from the Neotropics via LDD. Acknowledgments We thank Z.-L. Nie, Y. Meng, X. Kan, Deden Girmansyah and Y.M. Shui for collecting leaf material and laboratory assistance. We also acknowledge support of grants from the National Science Foundation (DEB 0743474 to S.R. Manchester and J. Wen), the Smithsonian Endowment Grant Program, the Small Grant Program of the National Museum of Natural History of the Smithsonian Institution, and the John D. and Catherine T. MacArthur Foundation, and a scholarship for X.-Q. Liu from the China Scholarship Council. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ympev.2012. 09.003. References Andersons, L., Rova, J.H.E., 1999. The rps16 intron and the phylogeny of the Rubioideae (Rubiaceae). Pl. Syst. Evol. 214, 161–186. Antonelli, A., 2009. Have giant lobelias evolved several times independently? Life form shifts and historical biogeography of the cosmopolitan and highly diverse subfamily Lobelioideae (Campanulaceae). BMC Biol. 7, 82. Baker, W.J., Dransfield, J., 2000. Towards a biogeographic explanation of the calamoid palms. In: Wilson, K.L., Morrison, D.A. (Eds.), Monocots: Systematics and Evolution. CSIRO, Melbourne, pp. 545–553. Bartish, I.V., Antonelli, A., Richardson, J.E., Swenson, U., 2011. Vicariance or longdistance dispersal: historical biogeography of the pantropical subfamily Chrysophylloideae (Sapotaceae). J. Biogeogr. 38, 177–190. Chanderbali, A.S.H., Van der Werff, H., Renner, S.S., 2001. The relationships and historical biogeography of Lauraceae: evidence from the chloroplast and nuclear genomes. Ann. Mo. Bot. Gard. 88, 104–134. Chen, P.T., Chen, L.Q., Wen, J., 2011a. The first phylogenetic analysis of Tetrastigma (Miq.) Planch., the host of Rafflesiaceae. Taxon 60, 499–512. Chen, P.T., Wen, J., Chen, L.Q., 2011b. Spatial and temporal diversification of Tetrastigma (Vitaceae). Gard. Bull. Singapore 63, 307–327. Chen, I., Manchester, S.R., 2007. Seed morphology of modern and fossil Ampelocissus (Vitaceae) and implications for phytogeography. Am. J. Bot. 94, 1534–1553. Chen, I., Manchester, S.R., 2011. Seed morphology of Vitaceae. Int. J. Plant Sci. 172, 1–35. 52 X.-Q. Liu et al. / Molecular Phylogenetics and Evolution 66 (2013) 43–53 Chen, I., 2009. History of Vitaceae Inferred from Morphology-Based Phylogeny and the Fossil Record of Seeds. Ph.D. Dissertation, University of Florida, Gainesville, USA. Clayton, J.W., Soltis, P.S., Soltis, D.E., 2009. Recent long-distance dispersal overshadows ancient biogeographical patterns in a pantropical angiosperm family (Simaroubaceae, Sapindales). Syst. Biol. 58, 395–410. Conti, E., Eriksson, T., Schönenberger, J., Sytsma, K.J., Baum, D.A., 2002. Early Tertiary out-of-India dispersal of Crypteroniaceae: evidence from phylogeny and molecular dating. Evolution 56, 1931–1942. Davis, C.C., Bell, C.D., Mathews, S., Donoghue, M.J., 2002. Laurasian migration explains Gondwanan disjunctions: evidence from Malpighiaceae. Proc. Natl. Acad. Sci. USA 99, 6833–6837. Davis, C.C., Fritsch, P.W., Bell, C.D., Mathews, S., 2004. High-latitude Tertiary migrations of an exclusively tropical clade: evidence from Malpighiaceae. Int. J. Plant Sci. 165 (Suppl.), S107–S121. De Queiroz, A., 2005. The resurrection of oceanic dispersal in historical biogeography. Trends Ecol. Evol. 20, 68–73. Descoings, B., 1960. Un genre méconnu de Vitacées: compréhension et distinction des genres Cissus L. et Cyphostemma (Planch.) Alston. Notul. Syst. 16, 113–125. Doyle, J.A., Sauquet, H., Scharaschkin, T., LeThomas, A., 2004. Phylogeny, molecular and fossil dating, and biogeographic history of Annonaceae and Myristicaceae (Magnoliales). Int. J. Plant Sci. 165 (Suppl.), S55–S67. Doyle, J.A., Le Thomas, A., 1997. Phylogeny and geographic history of Annonaceae. Géogr. Physiq. Quatern. 51, 353–361. Doyle, J.J., Doyle, J.L., 1987. A rapid isolation procedure from small quantities of fresh leaf tissue. Phytochem. Bull. 19, 11–15. Drummond, A. J., Ho, S.Y.W., Rawlence, N., Rambaut, A., 2007. A rough guide to BEAST 1.4. University of Auckland, Auckland, New Zealand. <http:// code.google.com/p/beast-mcmc/downloads/list>. Drummond, A.J., Rambaut, A., 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol. Biol. 7, 214. Duangjai, S., Samuel, R., Munzinger, J., Forest, F., Wallnöfer, B., Barfuss, M.H.J., Fischer, G., Chase, M.W., 2009. A multi-locus plastid phylogenetic analysis of the pantropical genus Diospyros (Ebenaceae), with an emphasis on the radiation and biogeographic origins of the New Caledonian endemic species. Mol. Phylogenet. Evol. 52, 602–620. Edgar, R.C., 2004. MUSCLE, multiple sequence alignment with high accuracy and high throughput. Nucl. Acids Res. 32, 1792–1797. Emadzade, K., Gehrke, B., Linder, H.P., Hörandl, E., 2011. The biogeographical history of the cosmopolitan genus Ranunculus L. (Ranunculaceae) in the temperate to meridional zones. Mol. Phylogenet. Evol. 58, 4–21. Erkens, R.H.J., Maas, J.W., Couvreur, T.L.P., 2009. From Africa via Europe to South America: migrational route of a species-rich genus of Neotropical lowland rain forest trees (Guatteria, Annonaceae). J. Biogeogr. 36, 2338–2352. Felsenstein, J., 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, 783–791. Fitch, W.M., 1971. Toward defining the course of evolution: minimum change for a specific tree topology. Syst. Zool. 20, 406–416. Fleagle, J.G., Gilbert, C.C., 2006. The biogeography of primate evolution: the role of plate tectonics, climate and chance. In: Lehman, S.M., Fleagle, J.G. (Eds.), Primate Biogeography: Progress and Prospects. Springer, New York, pp. 374–417. Gerrath, J.M., Posluszny, U., 1994. Morphological and anatomical development in the Vitaceae. VI. Cissus antarctica. Can. J. Bot. 72, 635–643. Gong, F., Karsai, I., Liu, Y.-S., 2010. Vitis seeds (Vitaceae) from the late Neogene Gray fossil site, northeastern Tennessee, USA. Rev. Paleobot. Palynol. 162, 71–83. Givnish, T., Renner, S., 2004. Tropical intercontinental disjunctions: Gondwana breakup, immigration from the boreotropics, and transoceanic dispersal. Int. J. Plant Sci. 165 (Suppl.), S1–S6. Greguss, P. 1969. Tertiary Angiosperm Woods in Hungary. Akádemiai Kiádo, Budapest, pp. 1–152. Herrera, F., Manchester, S.R., Jaramillo, C., 2012. Permineralized fruits from the late Eocene of Panama give clues of the composition of forests established early in the uplift of Central America. Rev. Paleobot. Palynol. 175, 10–24. Huelsenbeck, J.P., Ronquist, F.R., 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17, 754–755. Ingrouille, M.J., Chase, M.W., Fay, M.F., Bowman, D., Van der Bank, M., Bruijin, A.D.E., 2002. Systematics of Vitaceae from the viewpoint of plastid rbcL sequence data. Bot. J. Linn. Soc. 138, 421–432. Jackes, B.R., Rossetto, M., 2006. A new combination in Clematicissus Planch. (Vitaceae). Telopea 11, 390–391. Jackes, B.R., 1988. Revision of the Australian Vitaceae, 3. Cissus L. Austrobaileya 2, 481–505. Kulju, K.K., Sierra, S.E., Draisma, S.G., Samuel, R., Welzen, P.C., 2007. Molecular phylogeny of Macaranga, Mallotus, and related genera (Euphorbiaceae s.s.): insights from plastid and nuclear DNA sequence data. Am. J. Bot. 94, 1726– 1743. Lee, C., Wen, J., 2004. Phylogeny of Panax using chloroplast trnC–trnD intergenic region and the utility of trnC–trnD in interspecific studies of plants. Mol. Phylogent. Evol. 31, 894–903. Li, Y.Q., Dressler, S., Zhang, D.X., Renner, S.S., 2009. More Miocene dispersal between Africa and Asia—the Case of Bridelia (Phyllanthaceae). Syst. Bot. 34, 521–529. Lombardi, J.A., 1997. Types of names in Ampelopsis and Cissus (Vitaceae) referring to taxa in the Caribbean, Central and N. America. Taxon 46, 423–432. Lombardi, J.A., 2007. Systematics of Vitaceae in South America. Can. J. Bot. 85, 712– 721. Lombardi, J.A., 2000. Vitaceae—Gêneros Ampelocissus. Ampelopsis e Cissus. Flora Neotrop. Monogr. 80, 1–250. Magallón, S.A., Castillo, A., 2009. Angiosperm diversification through time. Am. J. Bot. 96, 349–365. Magallón, S.A., Sanderson, M.J., 2001. Absolute diversification rates in angiosperm clades. Evolution 55, 1762–1780. Malcomber, S.T., 2002. Phylogeny of Gaetnera Lam. (Rubiaceae) based on multiple DNA markers: evidence of a rapid radiation in a widespread, morphologically diverse genus. Evolution 56, 42–57. Manen, J.F., Natali, A., Ehrendorfer, F., 1994. Phylogeny of Rubiaceae–Rubieae inferred from the sequence of a cpDNA intergene region. Plant Syst. Evol. 190, 195–211. Manos, P.S., Donoghue, M.J., 2001. Progress in Northern Hemisphere phytogeography. Int. J. Plant Sci. 162 (Suppl.), S1–S2. Mau, B., Newton, M., Larget, B., 1999. Bayesian phylogenetic inference via Markov chain Monte Carlo methods. Biometrics 55, 1–12. McLoughlin, S., 2001. The breakup history of Gondwana and its impact on preCenozoic floristic provincialism. Aust. J. Bot. 49, 271–300. Morley, R.J., 2003. Interplate dispersal paths for megathermal angiosperms. Perspect. Plant Ecol. Evol. Syst. 6, 5–20. Muellner, A.N., Savolainen, V., Samuel, R., Chase, M.W., 2006. The mahogany family ‘‘out-of-Africa’’: divergence time estimation, global biogeographic patterns inferred from plastid rbcL DNA sequences, extant, and fossil distribution of diversity. Mol. Phylogenet. Evol. 40, 236–250. Müller, K., 2005. SeqState—primer design and sequence statistics for phylogenetic DNA data sets. Appl. Bioinform. 4, 65–69. Nie, Z.L., Sun, H., Manchester, S.R., Meng, Y., Luke, Q., Wen, J., 2012. Evolution of the intercontinental disjunctions in six continents in the Ampelopsis clade of the grape family (Vitaceae). BMC Evol. Biol. 12, 17. Nie, Z.L., Sun, H., Chen, Z.D., Meng, Y., Manchester, S.R., Wen, J., 2010. Molecular phylogeny and biogeographic diversification of Parthenocissus (Vitaceae) disjunct between Asia and North America. Am. J. Bot. 97, 1342–1353. Nylander, J.A.A., 2004. MrModeltest, Version 2 (Computer program). Evolutionary Biology Centre, Uppsala University, Uppsala, Sweden. Oxelman, B., Liden, M., Berglund, D., 1997. Chloroplast rps16 intron phylogeny of the tribe Sileneae (Caryophyllaceae). Plant Syst. Evol. 206, 393–410. Rambaut, A., 2002. Se-Al: Sequence Alignment Editor, Version 2.0 a11, University of Oxford, Oxford. <http://tree.bio.ed.ac.uk/software/seal/>. Rannala, B., Yang, Z.H., 1996. Probability distribution of molecular evolutionary trees: a new method of phylogenetic inference. J. Mol. Evol. 43, 304–311. Razafimandimbison, S.G., McDowell, T.D., Halford, D.A., Bremer, B., 2010. Origin of the pantropical and nutriceutical Morinda citrifolia L. (Rubiaceae): comments on its distribution range and circumscription. J. Biogeogr. 37, 520–529. Ree, R.H., Smith, S.A., 2008. Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis. Syst. Bio. 57, 4– 14. Ren, H., Lu, L.M., Soejima, A., Luke, Q., Zhang, D.X., Chen, Z.D., Wen, J., 2011. Phylogenetic analysis of the grape family (Vitaceae) based on the noncoding plastid trnC-petN, trnH-psbA, and trnL-F sequences. Taxon 60, 629–637. Renner, S.S., 2004. Multiple Miocene Melastomataceae dispersal between Madagascar, Africa and India. Philos. Trans. Roy. Soc. B (Biol. Sci.) 359, 1485– 1494. Renner, S.S., Clausing, G., Meyer, K., 2001. Historical biogeography of Melastomataceae: the roles of Tertiary migration and long-distance dispersal. Am. J. Bot. 88, 1290–1300. Richardson, J.E., Chatrou, L.W., Mols, J.B., Erkens, R.H.J., Pirie, M.D., 2004. Historical biogeography of two cosmopolitan families of flowering plants: Annonaceae and Rhamnaceae. Philos. Trans. Roy. Soc. B (Biol. Sci.) 359, 1495–1508. Rossetto, M., Crayn, D.M., Jackes, B.R., Porter, C., 2007. An updated estimate of intergeneric phylogenetic relationships in the Australian Vitaceae. Can. J. Bot. 85, 722–730. Rossetto, M., Jackes, B.R., Scott, K.D., Henry, R.J., 2001. Intergeneric relationships in the Australian Vitaceae: new evidence from cpDNA analysis. Genet. Res. Crop Evol. 48, 307–341. Rossetto, M., Jackes, B.R., Scott, K.D., Henry, R.J., 2002. Is the genus Cissus (Vitaceae) monophyletic? Evidence from plastid and nuclear ribosomal DNA. Syst. Bot. 7, 522–533. Sanmartín, I., Ronquist, F., 2004. Southern Hemisphere biogeography inferred by event-based models: plant versus animal patterns. Syst. Biol. 53, 216–243. Schönenberger, J., Conti, E., 2003. Molecular phylogeny and floral evolution of Penaeacee, Oliniaceae, Rhynchocalycaceae, and Alsateaceae (Myrtales). Am. J. Bot. 90, 293–309. Shaw, J., Lickey, E.B., Beck, J.T., Farmer, S.S., Liu, W., Miller, J., Chaw, S.K., Winder, C.T., Schilling, E.E., Small, R.L., 2005. The tortoise and the hare: relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. Am. J. Bot. 92, 142–166. Simmons, M.P., Ochoterena, H., 2000. Gaps as characters in sequence-based phylogenetic analyses. Syst. Biol. 49, 36–381. Smedmark, J.E.E., Eriksson, T., Bremer, B., 2010. Divergence time uncertainty and historical biogeography reconstruction—an example from Urophylleae (Rubiaceae). J. Biogeogr. 37, 2260–2274. Soejima, A., Wen, J., 2006. Phylogenetic analysis of the grape family (Vitaceae) based on three chloroplast markers. Am. J. Bot. 93, 278–287. Swofford, D.L., 2003. PAUP⁄: Phylogenetic Analysis Using Parsimony (and Other Methods), Version 4.0b10. Sinauer, Sunderland, Massachusetts, USA. X.-Q. Liu et al. / Molecular Phylogenetics and Evolution 66 (2013) 43–53 Taberlet, P., Gielly, L., Pautou, G., Bouvet, J., 1991. Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Mol. Biol. 17, 1105–1109. The Paleobiological Database. 2010. <http://paleodb.org/cgi-bin/bridge.pl> (accessed 11.12.10). Thorne, R., 2004. Tropical plant disjunctions: a personal reflection. Int. J. Plant. Sci. 165 (Suppl.), S137–S138. Thorne, R.F., 1972. Major disjunctions in the geographic ranges of seed plants. Quart. Rev. Biol. 47, 365–411. Tiffney, B.H., Barghoorn, E.S., 1976. Fruits and seeds of Brandon Lignite. I. Vitaceae. Rev. Paleobot. Palynol. 22, 169–191. Tiffney, B.H., Manchester, S.R., 2001. The use of geological and paleontological evidence in evaluating plant phylogeographic hypotheses in the Northern Hemisphere Tertiary. Int. J. Plant Sci. 162 (Suppl.), S3–S17. Timmons, S.A., Posluszny, U., Gerrath, J.M., 2007. Morphological and anatomical development in the Vitaceae. X. Comparative ontogeny and phylogenetic implications of Cissus quadrangularis L. Can. J. Bot. 85, 860–872. Wang, H., Moore, M.J., Soltis, P.S., Bell, C.D., Brockington, S.F., Alexandre, R., Davis, C.C., Latvis, M., Manchester, S.R., Soltis, D.E., 2009. Rosid radiation and the rapid rise of angiosperm-dominated forests. Proc. Natl. Acad. Sci. USA 106, 3853– 3858. Weeks, A., Daly, D.C., Simpson, B.B., 2005. The phylogenetic history and biogeography of the frankincense and myrrh family (Burseraceae) based on nuclear and chloroplast sequence data. Mol. Phylogenet. Evol. 35, 85–101. Wen, J., 1998. Evolution of the eastern Asian and eastern North American disjunct pattern: insights from phylogenetic studies. Korean J. Plant Taxon 28, 63–81. Wen, J., 2007a. Vitaceae. In: Kubitzki, K. (Ed.), The Families and Genera of Vascular Plants, Springer-Verlag, vol. 9. Berlin, Germany, pp. 466–478. Wen, J., 2007b. Leeaceae. In: Kubitzki, K. (Ed.), The Families and Genera of Vascular Plants, Springer-Verlag, vol. 9. Berlin, Germany, pp. 221–225. 53 Wen, J., Nie, Z.L., Soejima, A., Meng, Y., 2007. Phylogeny of Vitaceae based on the nuclear GAI1 gene sequences. Can. J. Bot. 85, 731–745. Wen, J., Ickert-Bond, S.M., 2009. Evolution of the Madrean–Tethyan disjunctions and the North and South American amphitropical disjunctions in plants. J. Syst. Evol. 47, 331–348. Wen, J., Ickert-Bond, S.M., Nie, Z.L., Li, R., 2010. Timing and modes of evolution of eastern Asian–North American biogeographic disjunctions in seed plants. In: Long, M., Gu, H., Zhou, Z. (Eds.), Darwin’s Heritage Today: Proceedings of the Darwin 200 Beijing International Conference. Higher Education Press, Beijing, pp. 252–269. Wheeler, E.A., Lapasha, C.A., 1994. Woods of the Vitaceae—fossil and modern. Rev. Paleobot. Palynol. 80, 175–207. Wikström, N., Savolainen, V., Chase, M.W., 2001. Evolution of the angiosperms: calibrating the family tree. Proc. Roy. Soc. Lond. B 268, 2211–2220. Wolfe, J.A., 1975. Some aspects of plant geography of the Northern Hemisphere during the late Cretaceous and Tertiary. Ann. Mo. Bot. Gard. 62, 264–279. Yuan, Y.M., Wohlhauser, S., Möller, M., Klackenberg, J., Callmander, M., Küpfer, P., 2005. Phylogeny and biogeography of Exacum (Gentianaceae): a disjunctive distribution in the Indian Ocean basin resulted from long-distance dispersal and extensive radiation. Syst. Biol. 54, 21–34. Zachos, J., Pagani, M., Sloan, L., Thomas, E., Billups, K., 2001. Trends, rhythms, and aberration in global climate 65 Ma to present. Science 292, 686–693. Zecca, G., Abbott, J.R., Sun, W.B., Spada, A., Sala, F., Grassi, F., 2012. The timing and the mode of evolution of wild grapes (Vitis). Mol. Phylogenet. Evol. 62, 736–747. Zerega, N.J.C., Clement, W.L., Datwyler, S.L., Weiblen, G.D., 2005. Biogeography and divergence times in the mulberry family (Moraceae). Mol. Phylogenet. Evol. 37, 402–416. Zhou, L.L., Su, Y.C.F., Thomas, D.C., Saunders, R.M.K., 2012. ‘Out-of-Africa’ dispersal of tropical floras during the Miocene climatic optimum: evidence from Uvaria (Annonaceae). J. Biogeogr. 39, 322–335.