Evolution of the intercontinental disjunctions in six continents in the Ampelopsis clade of the grape family (Vitaceae)


Material Information

Evolution of the intercontinental disjunctions in six continents in the Ampelopsis clade of the grape family (Vitaceae)
Series Title:
BMC Evolutionary Biology
Physical Description:
Mixed Material
Nie, Ze-Long
Sun, Hang
Manchester, Steven R.
Meng, Ying
Luke, Quentin
Wen, Jun
BioMed Central
Publication Date:


Background: The Ampelopsis clade (Ampelopsis and its close allies) of the grape family Vitaceae contains ca. 43 species disjunctly distributed in Asia, Europe, North America, South America, Africa, and Australia, and is a rare example to study both the Northern and the Southern Hemisphere intercontinental disjunctions. We reconstruct the temporal and spatial diversification of the Ampelopsis clade to explore the evolutionary processes that have resulted in their intercontinental disjunctions in six continents. Results: The Bayesian molecular clock dating and the likelihood ancestral area analyses suggest that the Ampelopsis clade most likely originated in North America with its crown group dated at 41.2 Ma (95% HPD 23.4 - 61.0 Ma) in the middle Eocene. Two independent Laurasian migrations into Eurasia are inferred to have occurred in the early Miocene via the North Atlantic land bridges. The ancestor of the Southern Hemisphere lineage migrated from North America to South America in the early Oligocene. The Gondwanan-like pattern of intercontinental disjunction is best explained by two long-distance dispersals: once from South America to Africa estimated at 30.5 Ma (95% HPD 16.9 - 45.9 Ma), and the other from South America to Australia dated to 19.2 Ma (95% HPD 6.7 - 22.3 Ma). Conclusions: The global disjunctions in the Ampelopsis clade are best explained by a diversification model of North American origin, two Laurasian migrations, one migration into South America, and two post-Gondwanan long-distance dispersals. These findings highlight the importance of both vicariance and long distance dispersal in shaping intercontinental disjunctions of flowering plants.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution.
Resource Identifier:
doi - 10.1186/1471-2148-12-17
System ID:

This item is only available as the following downloads:

Full Text
!DOCTYPE art SYSTEM 'http:www.biomedcentral.comxmlarticle.dtd'
ui 1471-2148-12-17
ji 1471-2148
dochead Research article
title p Evolution of the intercontinental disjunctions in six continents in the it Ampelopsis clade of the grape family (Vitaceae)
au id A1 snm Niefnm Ze-Longinsr iid I1 I5 email niezl@mail.kib.ac.cn
A2 SunHanghsun@mail.kib.ac.cn
A3 Manchestermi RStevenI2 stevsteven@flmnh.ufl.edu
A4 MengYingI3 mengying@mail.kib.ac.cn
A5 LukeQuentinI4 quentin.luke@swiftkenya.com
A6 ca yes WenJunwenj@si.edu
ins Key Laboratory of Biodiversity and Biogeography, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, Yunnan 650204, China
Florida Museum of Natural History, University of Florida, Gainesville, FL 32611, USA
Institute of Tibetan Plateau Research at Kunming, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, Yunnan 650204, China
East African Herbarium, National Museums of Kenya, Nairobi 00502, Kenya
Department of Botany, National Museum of Natural History, MRC 166, Smithsonian Institution, Washington, DC 20013-7012, USA
source BMC Evolutionary Biology
issn 1471-2148
pubdate 2012
volume 12
issue 1
fpage 17
url http://www.biomedcentral.com/1471-2148/12/17
xrefbib pubidlist pubid idtype doi 10.1186/1471-2148-12-17pmpid 22316163
history rec date day 4month 10year 2011acc 822012pub 822012cpyrt 2012collab Nie et al; licensee BioMed Central Ltd.note This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
sec st Abstract
The Ampelopsis clade (Ampelopsis and its close allies) of the grape family Vitaceae contains ca. 43 species disjunctly distributed in Asia, Europe, North America, South America, Africa, and Australia, and is a rare example to study both the Northern and the Southern Hemisphere intercontinental disjunctions. We reconstruct the temporal and spatial diversification of the Ampelopsis clade to explore the evolutionary processes that have resulted in their intercontinental disjunctions in six continents.
The Bayesian molecular clock dating and the likelihood ancestral area analyses suggest that the Ampelopsis clade most likely originated in North America with its crown group dated at 41.2 Ma (95% HPD 23.4 61.0 Ma) in the middle Eocene. Two independent Laurasian migrations into Eurasia are inferred to have occurred in the early Miocene via the North Atlantic land bridges. The ancestor of the Southern Hemisphere lineage migrated from North America to South America in the early Oligocene. The Gondwanan-like pattern of intercontinental disjunction is best explained by two long-distance dispersals: once from South America to Africa estimated at 30.5 Ma (95% HPD 16.9 45.9 Ma), and the other from South America to Australia dated to 19.2 Ma (95% HPD 6.7 22.3 Ma).
The global disjunctions in the Ampelopsis clade are best explained by a diversification model of North American origin, two Laurasian migrations, one migration into South America, and two post-Gondwanan long-distance dispersals. These findings highlight the importance of both vicariance and long distance dispersal in shaping intercontinental disjunctions of flowering plants.
Understanding the underlying mechanisms for the evolution of wide-ranging disjunct patterns has long been a major focus of biogeography abbrgrp abbr bid B1 1B2 2B3 3B4 4B5 5. Taxa disjunct at the global level involving both Northern and Southern Hemisphere are particularly informative because their histories may have general implications for other groups. Biogeographic history in the Northern Hemisphere is complicated, but has usually been explained by the widespread distribution of the Boreotropical flora in the Eocene and followed by appearance of more temperate forest elements during the mid-Tertiary extirpations of thermophilic elements in response to climatic cooling episodes of the late Eocene and the Plio-Pleistocene B6 6B7 7B8 8B9 9B10 10B11 11. The Southern Hemisphere is often interpreted to show a vicariance pattern attributed to the sequential breakup of Gondwanan landmasses B12 12. Recent studies on Nothofagus have demonstrated the relevance of long distance dispersal rather than vicariance in shaping Gondwanan distributional patterns B13 13B14 14B15 15B16 16.
Commonly three main routes for the migration of taxa between the Northern and the Southern Hemisphere have been recognized (Figure figr fid F1 1). The first is the opening of biotic exchanges between North and South America at various times in the Tertiary 1B17 17. The second hypothetical migration route is the trans-Tethyan dispersal between Europe and Africa B18 18. The third is less common concerning the possible route between Asia and Australia in the Miocene and later 18B19 19. These three routes can be viewed as alternative hypotheses for the ex situ origin of elements of global diversity.
fig Figure 1caption Delimitation of the six areas for taxa in the Ampelopsis clade and the schematic models used in the Lagrange analysestext
b Delimitation of the six areas for taxa in the Ampelopsis clade and the schematic models used in the Lagrange analyses. N = North America; A = eastern Asia; E = Europe to central Asia; F = Africa; S = South America; and U = Australia.
graphic file 1471-2148-12-17-1 hint_layout double
There are many examples of global distribution between the Southern and the Northern Hemisphere, particularly in pantropical families (e.g., Rubiaceae, Annonaceae, Lauraceae) or in many large cosmopolitan genera, such as Ranunculus B20 20, Senecio B21 21, and Lobelia B22 22B23 23. Nevertheless, all of them have continuous distribution from the tropics to the temperate zones. Molecular studies usually show a very complex disjunct history of large taxa with relatively low resolution B24 24B25 25. There are also some taxa with relatively few species that exhibit an intercontinental disjunction involving both the Northern and the Southern Hemisphere. However, the disjunction in such taxa usually involves only one or two northern or southern landmasses. Examples include Caltha (Ranunculaceae) with a global disjunction but absent from Africa B26 26 and Thamnosma (Rutaceae) disjunct between North America and Africa B27 27. Of great interest, we have recently found a truly global disjunct pattern in a small group including Ampelopsis Michx. and its relative taxa from the grape family, Vitaceae B28 28B29 29B30 30.
Vitaceae is a well-known group of flowering plants having a largely pantropical distribution in Asia, Africa, Australia, the neotropics, and the Pacific islands, with only a few genera in temperate regions B31 31B32 32B33 33. As currently circumscribed, Ampelopsis is one of the few genera mostly restricted to the north temperate zone. It has approximately 25 species disjunctly distributed in Eurasia (c. 22 spp.) and North and Central America (3 spp.). Recent phylogenetic analyses based on plastid or nuclear sequences revealed that there are at least two disjunctions between the New and the Old World in the genus 2829. More interestingly, both chloroplast and nuclear data clearly suggested that the African Rhoicissus Planch. and the South American Cissus striata Ruiz & Pav. complex form a clade nested within the paraphyletic Ampelopsis 2829. Rhoicissus consists of about 12 species endemic to tropical and southern Africa. The Cissus striata complex contains four species from South America 29B34 34. Furthermore, the Australian genus Clematicissus Planch. seemed to be closely related to the Cissus striata Rhoicissus clade based on plastid data 30B35 35. There are only two Clematicissus species known from Australia: C. opaca from Australia's eastern region and C. angustissima from the west coast 30. Therefore, the Northern Hemisphere Ampelopsis and its three close relatives (Rhoicissus, Clematicissus and the Cissus striata complex) from the Southern Hemisphere (hereafter referred to as the Ampelopsis clade) demonstrate an unusual global intercontinental disjunct pattern involving six continents (Figure 1). Yet morphologically, the Ampelopsis clade seems to be heterogeneous, such as in having 4-7-merous flowers and fleshy to dry fruits 29.
The Ampelopsis clade offers a good opportunity to explore the origin and evolution of the global intercontinental disjunct pattern in flowering plants, especially concerning both Northern and Southern Hemisphere intercontinental disjunctions. A hierarchical global distribution was predicted by our previous studies, but with limited sampling 2829. We herein employ phylogenetic, molecular dating, and biogeographic methods to reconstruct the evolutionary history of the Ampelopsis clade based on a comprehensive sampling scheme using four plastid regions (trnL-F, rps16, psbA-trnH, and atpB-rbcL).
Taxon sampling, DNA sequencing, and phylogenetic analyses
We sample 28 of the 43 species (65%) of the Ampelopsis clade including all three North American Ampelopsis species, 15 of the 22 Eurasian Ampelopsis, five of the 12 Rhoicissus species from Africa, three of the four species from the Cissus striata complex from South America, and the two Australian Clematicissus species (Additional file supplr sid S1 1, Table S1). The sampling covers the entire extant geographic range of Ampelopsis and its close relatives from both the Northern and Southern Hemisphere. But we still have some missing taxa in our sampling scheme, such as Rhoicissus with only 41% species sampled. However, each group is supposed to be monophyletic based on morphological, biogeographic, and molecular evidence 28293034B36 36. The missing taxa in our sampling should have little effect in the present study that focused on phylogenetic relationships among genera and biogeographic evolution at the intercontinental level. In order to place our analyses of the Ampelopsis clade in a broad framework on the family level, we sampled 66 additional taxa from the other major groups in Vitaceae (i.e., the Vitis Parthenocissus Ampelocissus clade, the core Cissus clade, and the Cyphostemma Caryatia Tetrastigma clade) plus three Leea species of Leeaceae based on previous investigations 2829.
Additional file 1
Table S1. Voucher information and GenBank accession numbers of the Ampelopsis clade and representative taxa in Vitaceae. Abbreviations of herbaria are as follows: KUN, Kunming Institute of Botany, Chinese Academy of Sciences; and US, the United States National Herbarium. Accession numbers beginning with JQ indicate sequences generated for this study and the others were obtained from GenBank. A dash means sequences missing.
name 1471-2148-12-17-S1.DOC
Click here for file
Total DNAs were extracted from silica gel dried leaves using the Dneasy Plant Mini Kit (QIAGEN, Crawley, UK). Amplification and sequencing followed Soejima and Wen (2006) for trnL-F, rps16, and atpB-rbcL, and Meng et al. B37 37 for psbA-trnH. DNA sequences were assembled using Sequencher v4.1.4 (Gene Codes Corp., Ann Arbor, Michigan, USA). Sequence alignment was initially performed using MUSCLE 3.8.31 B38 38 in the multiple alignment routine followed by manual adjustment in Se-Al v2.0a11 (http://tree.bio.ed.ac.uk/software/seal/). The chloroplast genome is generally considered as one unit without recombination although there have been reports of recombination in the chloroplast genome B39 39. Therefore, we combined all the plastid data (trnL-F, rps16, psbA-trnH, and atpB-rbcL) in our analysis. The combined plastid data were analyzed using Bayesian inference as implemented in MrBayes 3.1.2 B40 40. The best-fit model of nucleotide substitution (GTR + I + G) was determined by MrModelTest 2.3 B41 41 using the Akaike Information Criterion (AIC). Variation of gaps in our sequences is not complicate. A total of 31 binary characters were coded for gaps according to Simmons and Ochoterena B42 42 and separated into independent partition in all analyses. Bayesian tree topology and posterior probabilities (PP) were determined from two independent runs of four incrementally heated chains. Runs were performed for 5 million generations with sampling of trees every 500sup th generation. When the log-likelihood scores were found to have stabilized, a consensus tree was calculated after omitting the first 10% of trees as burn-in.
Divergence time estimation
For molecular dating analyses, the strict molecular clock model was rejected from our dataset based on a likelihood ratio test performed in PAUP* B43 43. Therefore, we estimated node ages within the Ampelopsis group using a Bayesian relaxed clock model as implemented in BEAST v1.6.1 B44 44. We largely followed the dating strategies in Nie et al. (2010), which analyzed diversification in Parthenocissus of Vitaceae. 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). Tracer version 1.5 was used to check for convergence between the runs 44. After discarding the first 10% samples as burn-in, the trees and parameter estimates from the two runs were combined using LogCombiner 1.6.1 44. Results were considered reliable once the effective sampling size (ESS) for all parameters exceeded 200 as suggested by the program manual B45 45. The samples from the posterior were summarized on the maximum clade credibility tree using the program TreeAnnotator 1.6.1 44 with posterior probability limit of 0.5 and mean node heights summarized.
Fossil seeds of Vitaceae can be differentiated to the generic level B46 46B47 47. The oldest best preserved seed fossil of the family is from the late Paleocene of the Beicegel Creek locality in North Dakota. This fossil is undoubtedly assigned to Ampelocissus s.l. (as A. parvisemina Chen & Manchester) and is easily distinguished from all other vitaceous genera by its long, parallel ventral infolds and a centrally positioned oval chalazal scar 46. Since the Ampelocissus s.l. is not monophyletic with Vitis nested within it 28, the A. parvisemina fossil thus may represent an early member of the Ampelocissus clade retaining some characters shared with its common ancestor to Vitis 46. The stem age of the Ampelocissus Vitis clade was thus fixed at 58.5 ± 5.0 million years ago (Ma).
For the root age of Vitaceae, Nie et al. (2010) fixed the split between Vitaceae and Leea as 85 ± 4 based on the estimated age of 78-92 Ma by Wikström et al. (2001). Recently, Bell et al. (2010) reported an estimate ranging from 65 (45 81) to 48 (21 79) Ma for the crown age Vitis Leea clade, which is roughly consistent with the earliest fossil evidences of Vitaceae in the Palaeocene 46. However, their results may have underestimated for Vitaceae because the oldest fossil of A. parvisemina is undoubtedly assigned to the Ampelocissus s.l. within Vitaceae and the family is predicted to have a Cretaceous history in view of its basal position in the rosids and the presence of Cretaceous rosid fossils B48 48. The time estimates of angiosperms by Magallón and Castillo (2009) also suggested a pre-Tertiary origin as 90.65 (90.47 90.84) to 90.82 (90.64 91) Ma for Vitaceae. The inferences from Magallón and Castillo (2009) and Wikström et al. (2001) are close, although the later was criticized for the nonparametric rate smoothing method and for calibrating the tree using only a single calibration point. Therefore, we used 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. A low standard error was used because of the narrow 95% confidence from Magallón and Castillo (2009).
Ancestral area reconstruction
Several methods have been recently proposed that take into account of genetic branch lengths, phylogenetic uncertainty, and branch length uncertainty for reconstructing distributional change through evolutionary time, using either maximum likelihood B49 49 or Bayesian inference B50 50. The ancestral area of the Ampelopsis clade was reconstructed with the likelihood analysis using the program Lagrange version 20110117 49B51 51. Unlike the DIVA method B52 52, this likelihood approach incorporates an explicit dispersal-extinction-cladogenesis (DEC) model of dispersal routes available at historical intervals correlating stochastic events with lineage persistence 49. The likelihood analysis is prone to estimate wide ancestral ranges for early-branching lineages B53 53B54 54B55 55. In our case, ancestral ranges were assumed to include no more than two areas since all extant species in the Ampelopsis clade are restricted into only one area. Moreover, spatial and temporal constraints (e.g., area distances, continent connections, dates of geological origin) may be imposed in the DEC model estimation, providing a more accurate estimation of the ancestral ranges and hypothesis testing of different geographic scenarios. We did not conduct the Bayesian calculation of ancestral geographic distributions with standard continuous-time Markov chains (CTMCs), because geologic information (e.g., the presence and dissolution of land bridges and island chains) is not explicitly incorporated into the analyses.
We herein used the likelihood method to test a null model and six alternative biogeographic scenarios (Figure 1) based on the hypothesized dispersal or migration routes between the Northern and the Southern Hemisphere. Six areas were delimited by continental divisions and the extant distributions of the Ampelopsis clade: 1) N North America including Central America; 2) S South America; 3) F Africa; 4) A eastern Asia; 5) E Europe to central Asia; and 6) U Australia (Figure 1). The unconstrained null model (M0) assumes that spatial and temporal distribution has no effect on biogeographic patterns of evolution and allows geographic ranges to include any possible combination of continents and permits direct dispersal between any area pairs. The M1 model favors a migration route from North to South America (N S) with the biogeographic connections between Europe and Africa (E F) and eastern Asia and Australia (A U) excluded from our analyses. Similarly, the M2 model considers the connection between E F and did not allow other possibilities. The migration route between Asia and Australia (M3) seems less likely, but we still considered it in our analyses as a comparison. We also test models that allow two connections between the Northern and Southern hemispheres (M4 M6 in Figure 1). Following Ree et al. (2005), the results between models were assessed by directly comparing their log-likelihood scores. The conventional cut-off value over two log-likelihood units was considered statistically significant, and models with lower likelihood score were rejected B56 56B57 57.
The total length of the aligned data matrix is 3933 bp. The Bayesian consensus tree is highly congruent with the maximum clade credibility tree obtained from BEAST and the later is shown in Figure F2 2 with PP support values > 0.50. Our results support the monophyly of the Ampelopsis clade with three major lineages resolved within the Ampelopsis clade (Figure 2). Two distinct lineages (hereafter named as North I and II) correspond to the two sections of Ampelopsis B58 58. North I includes all species of section Leeaceifoliae with pinnately to bipinnately compound leaves. North II consists of taxa of section Ampelopsis with simple or palmately-divided or palmately-compound leaves. The Southern Hemisphere taxa (the African Rhoicissus, the South American Cissus striata complex, and the Australian Clematicissus) form a clade (the South group in Figure 2).
Figure 2Maximum clade credibility tree inferred with BEAST, with the 95% highest posterior density indicated by gray bars
Maximum clade credibility tree inferred with BEAST, with the 95% highest posterior density indicated by gray bars. Nodes of interests were marked as 0 to 5 as in Table 1; and calibrations are indicated with black stars. Values above branches represent Bayesian posterior probabilities.
A total of 100 million generations (2 runs of 50 million generations each) are necessary to reach sufficient ESS. The Ampelopsis clade is estimated to have diverged from its close relatives in Vitaceae at 41.2 (23.4 61.0) Ma in the Eocene. The Bayesian estimates also suggest that all the other major clades of Vitaceae (e.g., Vitis Parthenocissus Ampelocissus clade, core Cissus clade, and Cyphostemma Cayratia Tetrastigma) had already diversified in the Eocene (Figure 2). Ages of major groups within the Ampelopsis clade obtained in our study are summarized in Table tblr tid T1 1.
tbl Table 1Results of molecular dating and ancestral range reconstruction for major nodes within the Ampelopsis clade.tblbdy cols 10
Node 0:
Crown Ampelopsis clade
Node 1:
Crown North I
Node 2:
Stem South group
Node 3:
Crown South group
Node 4:
Crown Cissus striata complex Clematicissus
Node 5:
Molecular dating
with BEAST (Ma)
41.2 (23.4-61.0)
N|NF (0.25)
A|N (0.81)
F|N (0.32)
S|F (0.22),
U|F (0.21)
U|S (0.63)
N|A (0.47)
N|NS (0.61)
A|N (0.85)
S|N (0.92)
S|F (0.54)
U|S (0.79)
N|A (0.46),
N|E (0.36)
A|E (0.25),
N|E (0.20)
A|N (0.51)
F|E (0.55)
F|F (0.39)
U|S (0.32)
N|E (0.51)
A|AU (0.60)
A|N (0.68)
U|A (0.90)
U|F (0.42),
U|U (0.41)
U|S (0.77)
N|A (0.72)
N|NS (0.61)
A|N (0.84)
S|N (0.91)
S|F (0.51)
U|S (0.78)
N|A (0.46),
N|E (0.35)
N|NS (0.58)
A|N (0.84)
S|N (0.86)
S|F (0.51)
U|S (0.79)
N|A (0.48),
N|E (0.35)
A|AU (0.55)
A|N (0.66)
U|A (0.80)
U|U (0.40),
U|F (0.37)
U|S (0.73)
N|A (0.65)
The nodes of interest are shown in Figure 2, and the likelihood scores (-lnL), and estimates of dispersal (D) and extinction (E) rates (events per million years) are given. M0 is a null model without constraints and M1-M6 are alternative models in the Lagrange analyses. Only the highest relative probability is shown. Bold text represents the model with a significantly better likelihood when compared with the other models tested (more than two log-likelihoods better)
Patterns of temporal and spatial distribution of the Ampelopsis clade are inferred using the maximum likelihood DEC method. We compare seven models (i.e., a null and six alternatives) for the six areas (Figure 1) and the effects of different models on likelihood reconstructions are shown in Table 1. Analyses based on M1, M4, and M5 typically have lower likelihood scores than other models and produced nearly identical results (Table 1). For example, all of them suggested that the ancestral range split at the stem South lineage is between North and South America (Node 2 in Table 1). Our results also suggest that the three models (M1, M4, and M5) are significantly different from the others (M2, M3, and M6) with scores over two log-likelihood units. The M1 model is suggested as the best one with the highest likelihood score, and this model is shown in Figure F3 3.
Figure 3Biogeographic scenario for the global disjunction of the Ampelopsis clade based on molecular dating and the best M1 model with the highest likelihood score in the ancestral range analyses
Biogeographic scenario for the global disjunction of the Ampelopsis clade based on molecular dating and the best M1 model with the highest likelihood score in the ancestral range analyses. Long distance dispersal is indicated as dash lines and migration as solid lines. The ancestral area of the Ampelopsis clade is shown with black stars on the maps. The tree branches and ranges on the tree are coded as follows: blue = North America (N); grey = eastern Asia (A); orange = Europe and central Asia (E); green = South America (S); yellow = Africa (F); and red = Australia (U).
North American origin
The Ampelopsis clade is composed of two distinct Laurasian lineages each disjunct between the Old and the New World and one South group with a Gondwana-like intercontinental disjunction: (Africa (Australia, and South America)) (Figure 3). The most likely model M1 in the DEC analyses suggested that the Ampelopsis clade had an early diversification in North America with a geographic split between N (North America) and NS (North and South America) (Table 1; Figure 3). The fact that most fossil records of the family including the oldest record in the Paleocene are found from North America 46 is consistent with this "out of North America" hypothesis. Although South America is inferred as part of the ancestral areas (Table 1), it seems less likely to be the ancestral area of the Ampelopsis clade than North America because there are very few fossils known before the Eocene of South America 47. Furthermore, phylogenetic results also contradict the possibility of South American origin because the South American group is well embedded within the Ampelopsis clade (Figure 2). Asia also seems less likely than North America to be the ancestral area of the Ampelopsis clade, in spite of its highest extant species richness of the lineage. No seed records with ages younger than Oligocene are known from Asia 47.
The ancestral area for a taxon is usually expected to be correlated with high extant species richness. For example, eastern Asia usually has a higher level of species diversity and endemism, and has been suggested to be the ancestral area for many eastern Asian eastern North American disjunct groups B59 59B60 60B61 61B62 62B63 63. Donoghue and Smith (2004) found a predominance of directionality from Asia to the New World. Of the 29 lineages they analyzed with an eastern Asian and eastern North American disjunction, only one lineage showed directionality from eastern North America to eastern Asia. However, Wen et al. (2010) reported many more lineages with North American origins and subsequent migrations into eastern Asia, with 29 of the total 98 examined (30%) lineages migrated/dispersed from the New World to the Old World. It seems that eastern Asia has been over-emphasized as an ancestral area for Laurasian taxa due to its retention of the greater number of species 62B64 64. North America is supported to have played an important role in the early evolution of the two Ampelopsis lineages in spite of the lower species diversity today in North America compared with eastern Asia. The lower species richness in North America is often explained by the hypothesis that both North America and eastern Asia were occupied by Boreotropical elements in the early Tertiary but North America suffered more severe extinctions with global cooling beginning in the late Eocene or Oligocene 1063. The high level of species diversity and endemism in Asia can also be attributed to secondary diversification due to habitat heterogeneity as well as a lower rate of extinctions in the late Tertiary 6061.
Diversification pattern in the Southern Hemisphere
The likelihood analyses using Lagrange based on M1, M4, and M5 suggest that the most likely route between the Northern and the Southern Hemisphere is from North to South America, although more options of connections are permitted in M4 (between Europe and Africa) and M5 (between eastern Asia and Australia) (Table 1 and Figure 3). Ampelopsis may have dispersed into South America via scattered continental and/or volcanic islands that connected North and South America at various times in the Tertiary 117, such as via the proto Greater Antilles (ca. 50 Ma) or via GAARlandia that existed around 33 35 Ma 17B65 65. The separation of the southern lineage from its Laurasian ancestor at 35.5 (20.7 52.3) Ma in the late Eocene broadly coincides with a possible biological connection between North and South America around the Eocene-Oligocene boundary B66 66. Fossil seeds found from the Eocene of South America are closely related to those of the Central American Ampelocissus, indicating the possible floristic connection between North and South America at that time 46.
Another possible migration route is that the Laurasian ancestors of the Ampelopsis clade reached Africa using the Boreotropical connection via the North Atlantic and Europe in the late Eocene to early Oligocene. However, this hypothesis is rejected by the Lagrange analysis (see M2 in Table 1). The model M4 with two possible connections between the Northern and Southern Hemisphere (N S and E F, Figure 1) did not support the European-African (E F) route. Although the separation of the southern lineage from its Laurasian ancestor in the early Oligocene broadly coincides with the disruption of the Boreotropical flora around the Eocene-Oligocene boundary 66, both the shallow seas that separate Africa from Eurasia and the dry belt in northern Africa were barriers to biotic exchange between the two continents in the early to mid Tertiary 1B67 67B68 68. The third hypothesis (the M3 model, Table 1) is that Ampelopsis entered the Southern Hemisphere via the Asian Australian connection. This model seems quite unlikely based on our analyses (Table 1). The model M5 that permits two connections between Northern and Southern Hemisphere (N S, A U, Figure 1) also prefers the connection between North and South America (N S) rather than the Asian Australian connection (A U). The availability of biotic interchange between Australia and Asia beginning at the Miocene 18 is too recent to support this scenario.
The divergence time 30.5 (95% HPD: 16.9 45.9) Ma in the early Oligocene was estimated for the first split between South America and Africa (node 3 in Figure 2 and Table 1). This time is well after the last possible connection of Africa and other southern landmasses at around 96-105 Ma 1819. We thus argue that long distance dispersal (LDD) is the most plausible mechanism for their southern intercontinental disjunction. Vitaceae taxa are usually dispersed by animals, especially birds 32B69 69B70 70B71 71. All taxa in the Ampelopsis clade except the Australian Clematicissus angustissima bear fleshy berries that may have facilitated LDD. In particular, LDD has been recently accepted by a number of studies as the driving force for plant disjunctions in the Southern Hemisphere, especially for those with relatively recent divergence times 520B72 72B73 73B74 74. Biogeographic studies on Vochysiaceae B75 75 suggested a LDD from South America to Africa in the Oligocene. Givnish et al. B76 76 showed that the single African genus Maschocephalus of Rapateaceae is of recent origin in the late Miocene and reached Africa from South America via LDD. Dispersals between Africa and South America have also been suggested for a number of well-studied taxa, such as in Melastomataceae B77 77, and Simaroubaceae 73.
Our results support that the southern lineage of Ampelopsis arrived in Australia from South America in the early Miocene (node 4 in Figure 2). Migration between Australia and South America may be alternatively explained by a trans-Antarctic exchange 12. This Antarctic route existed during the late Cretaceous-early Tertiary and was interrupted only in the late Eocene (30-35 Ma) when the South Tasman Sea opened up between Australia and eastern Antarctica 19B78 78. This route is supported by evidence from several plant groups, such as Annonaceae B79 79, and Sapotaceae B80 80. Yet the split between the South American Cissus striata complex and the Australian Clematicissus in the early Miocene is too young to be explained by an Antarctic migration. LDD is the most plausible explanation for this disjunction. Finally, the disjunction of the two Clematicissus species in eastern and western Australia 30B81 81 may represent a relict distribution and their divergence time in the Pliocene is consistent with the aridification in central Australia at that time B82 82B83 83.
Laurasian migrations
The DEC reconstruction suggests a North American eastern Asian split (N|A) for the North I disjunction and a North American eastern Asian or North American Europe split (N|A or N|E) for the disjunction in the North II lineage (Table 1; Figure 3). The two Northern Hemisphere disjunctions may have involved the North Atlantic land bridges or the Bering land bridge from North America to eastern Asia 10B84 84. We prefer to use the North Atlantic route because it is well supported in sect. Ampelopsis (North II in Figure 3). The southern North American Ampelopsis denudata diverged first, followed by the southeastern US A. cordata. The western Asian/southern European A. orientalis is then sister to the large Asian clade. The relative position of these areas in the cladogram is congruent with the migration of the lineage from North America to Europe across the North Atlantic land bridges, and the lineage then reached Asia subsequently (Figure 3).
The Bayesian molecular clock dating with fossil calibration suggests an early Miocene split of the two disjunct groups in Ampelopsis (nodes 1 and 5 in Figure 2). The divergence times are also consistent with the possibility of the North Atlantic migration route. Based on paleogeological, zoological, and botanical fossil evidence, Tiffney (1985b) argued for the importance of the North Atlantic land bridges to tropical or warm temperate species in the early Eocene to middle Miocene. A continuous belt of Boreotropical elements covered much of the southern part of North America, southern Eurasia, and northwestern Africa in the Eocene 966. At that time, plant migrations through direct land connection or across limited water gaps were possible through the North Atlantic land bridges. Significant cooling during the Oligocene resulted in southward retreats and the extirpation of some lineages comprising this flora 64B85 85B86 86. A gradual warming period occurred into the early Miocene, resulting in the expansion of some evergreen and thermophilic lineages in Europe and North America B87 87B88 88. Dispersal of Boreotropical or warm temperate thermophilic elements, such as Ampelopsis, is therefore considered likely across the North Atlantic land bridges during this period. There are very few extant Vitaceae species in Europe, but many vitaceous seeds were reported in the early Tertiary of Europe 47B89 89. Together, these lines of evidence strongly suggest that the Ampelopsis clade may have used the corridors via the North Atlantic land bridges as a pathway to reach Eurasia in the early Miocene.
Our results suggest a complex history of diversification in the Ampelopsis clade to explain the global disjunctions that includes a North American origin, two Laurasian migrations, one migration into South America, and two post-Gondwanan LDDs. These findings may have general implications for the origin and diversification of plants with global disjunctions. Asia, Africa, or South America has often been suggested as the ancestral area for many intercontinental disjunct groups 4B90 90B91 91. Evidence from the Ampelopsis clade suggests that North America may have played an important role in the origin of some modern flowering plants in spite of its often lower species diversity when compared with other areas, such as eastern Asia B92 92 or Africa B93 93. Recent biogeographic analyses of several other groups appear to provide additional examples of North American origins (e.g., Phryma B94 94; Simaroubaceae 73). This study also highlights the importance of the North and South American route in the global diversification between the Northern and the Southern Hemisphere B95 95. This route apparently played an important role in the wide distribution of many pantropical plants in the early Tertiary, such as Annonaceae B96 96B97 97, Malpighiaceae 91, and Rubiaceae B98 98.
Recent biogeographic analyses have underscored the relative importance of LDD to intercontinental disjunctions in the Southern Hemisphere than traditionally assumed 313. The fit between area cladograms and the history of tectonic fragmentation might have been overstated 5. If we accepted the results of our calibration analyses as absolute, rather than minimum ages, then the Southern Hemisphere clade is too young to have been achieved by the Gondwanan breakup. The Gondwana-like disjunction in the Ampelopsis clade was reconstructed to have a North American origin with an initial migration into South America and then dispersed from South America into Africa and Australia independently via LDD. A similar example from Lycium (85 spp., Solanaceae) was suggested a New World origin of Lycium with recent dispersal from the Americas to Africa, and then to eastern Asia B99 99B100 100B101 101. Dispersal has been hypothesized to be the dominant pattern in this genus that has red, fleshy, bird-dispersed fruits 99. On the other hand, however, the Laurasian lineages in Ampelopsis clade favor a vicariance migration pattern from North America via the North Atlantic land bridges to Eurasia. Our results thus support both the Laurasian migrations and the post-Gondwanan LDD to explain the global disjunction of the Ampelopsis clade.
Authors' contributions
JW, ZLN, HS, and SRM conceived the ideas; JW, YM, QL and ZLN collected the materials; ZLN and YM analyzed the data; and ZLN and JW led the writing. All authors read and approved the final submission.
This study was supported by grants from the National Science Foundation (DEB 0743474 to SR Manchester and J Wen), the Natural Science Foundation of China (NSFC 31061160184 to H Sun and 31129001 to J Wen), the NSFC-Yunnan Joint project (U1136601 to H Sun), and the John D and Catherine T MacArthur Foundation (to J Wen, R Ree, and G Mueller). Laboratory work was conducted in and partially supported by the Laboratory of Analytical Biology of the National Museum of Natural History, the Smithsonian Institution. Fieldwork in North America was supported by the Small Grants Program of the National Museum of Natural History, the Smithsonian Institution. Michael Nee, Maurizio Rossetto, Ihsan Al-Shehbaz, and Daigui Zhang kindly helped with the sample collection.
refgrp Angiosperm biogeography and past continental movementsRavenPHAxelrodDIAnnals of the Missouri Botanical Garden1974613539lpage 67310.2307/2395021Plant species disjunctions: a summaryRavenPHAnnals of the Missouri Botanical Garden197259223424610.2307/2394756Gondwanan break-up: legacies of a lost world?UpchurchPTrends in Ecology & Evolution200823422923610.1016/j.tree.2007.11.00622331889Phylogenetic patterns in Northern Hemisphere plant geographyDonoghueMJBellCDLiJHInternational Journal of Plant Sciences2001162S41S5210.1086/323278The resurrection of oceanic dispersal in historical biogeographyde QueirozATrends in Ecology & Evolution2005202687310.1016/j.tree.2004.11.00622331889Outline of the origin and historical recognition of floristic affinities between Asia and eastern North AmericaGrahamAFloristics and Paleofloristics of Asia and Eastern North Americapublisher Amsterdam: Elseviereditor Graham A1972118The origin and evolution of tertiary relict florasMilneRIAbbottRJAdvances in Botanical Research200238281314Perspectives on the origin of the floristic similarity between eastern Asia and eastern North AmericaTiffneyBHJournal of the Arnold Arboretum19856617394The Eocene North Atlantic land bridge: its importance in tertiary and modern phytogeography of the Northern hemisphereTiffneyBHJournal of the Arnold Arboretum1985662243273Evolution of eastern Asian and eastern North American disjunct distributions in flowering plantsWenJAnnual Review of Ecology and Systematics19993042145510.1146/annurev.ecolsys.30.1.421Patterns of animal dispersal, vicariance and diversification in the HolarcticSanmartínIEnghoffHRonquistFBiological Journal of the Linnean Society2001734345390Southern Hemisphere biogeography inferred by event-based models: Plant versus animal patternsSanmartínIRonquistFSystematic Biology200453221624310.1080/10635150490423430link fulltext 15205050Relaxed molecular clock provides evidence for long-distance dispersal of Nothofagus (southern beech)KnappMStocklerKHavellDDelsucFSebastianiFLockhartPJPLos Biology2005313843Not so ancient: the extant crown group of Nothofagus represents a post-Gondwanan radiationCookLGCrispMDProceedings of the Royal Society B-Biological Sciences200527215802535254410.1098/rspb.2005.3219Nothofagus and Pacific biogeographyLinderHPCrispMDCladistics199511153210.1111/j.1096-0031.1995.tb00002.xBiogeography of Nothofagus supports the sequence of Gondwana break-upSwensonUHillRSMcLoughlinSTaxon20015041025104110.2307/1224719Paleogeography of the Caribbean region: Implications for cenozoic biogeographyIturralde-VinentMAMacPheeRDEBulletin of the American Museum of Natural History1999238195Interplate dispersal paths for megathermal angiospermsMorleyRJPerspectives in Plant Ecology, Evolution and Systematics200361-252010.1078/1433-8319-00039The breakup history of Gondwana and its impact on pre-Cenozoic floristic provincialismMcLoughlinSAustralian Journal of Botany200149327130010.1071/BT00023The biogeographical history of the cosmopolitan genus Ranunculus L. (Ranunculaceae) in the temperate to meridional zonesEmadzadeKGehrkeBPeter LinderHHörandlEMolecular Phylogenetics and Evolution201158142110.1016/j.ympev.2010.11.00221078403An ITS phylogeny of tribe Senecioneae (Asteraceae) and a new delimitation of Senecio LPelserPBNordenstamBKadereitJWWatsonLETaxon2007561077110410.2307/25065905Higher level phylogeny and evolutionary trends in Campanulaceae subfam. Lobelioideae: Molecular signal overshadows morphologyAntonelliAMolecular Phylogenetics and Evolution200846111810.1016/j.ympev.2007.06.01518042405Have giant lobelias evolved several times independently? Life form shifts and historical biogeography of the cosmopolitan and highly diverse subfamily Lobelioideae (Campanulaceae)AntonelliABMC Biology200978210.1186/1741-7007-7-82pmcid 278905519941635Multiple origins of Southern Hemisphere Anemone (Ranunculaceae) based on plastid and nuclear sequence dataSchuettpelzEHootSBSamuelREhrendorferFPlant Systematics and Evolution2002231114315110.1007/s006060200016Phylogenetic affinities of South American Anemone (Ranunculaceae), including the endemic segregate genera, Barneoudia and OreithalesMeyerKMHootSBArroyoMTKInternational Journal of Plant Sciences2010171332333110.1086/650153Phylogeny and biogeography of Caltha (Ranunculaceae) based on chloroplast and nuclear DNA sequencesSchuettpelzEHootSBAmerican Journal of Botany200491224725310.3732/ajb.91.2.24721653380Old-New World and trans-African disjunctions of Thamnosma (Rutaceae): Intercontinental long-distance dispersal and local differentiation in the succulent biomeThivMvan der NietTRutschmannFThulinMBruneTLinderHPAmerican Journal of Botany2010981768721613086Phylogenetic analysis of the grape family (Vitaceae) based on three chloroplast markersSoejimaAWenJAmerican Journal of Botany200693227828710.3732/ajb.93.2.27821646189Phylogeny of Vitaceae based on the nuclear GAI1 gene sequencesWenJNieZ-LSoejimaAMengYCanadian Journal of Botany200785873174510.1139/B07-071An updated estimate of intergeneric phylogenetic relationships in the Australian VitaceaeRossettoMCraynDMJackesBRPorterCCanadian Journal of Botany20078572273010.1139/B07-022VitaceaeChenZDRenHWenJFlora of ChinaBeijing and St. Louis, Mo.: Science Press and Missouri Botanical Garden PressWu Z-Y, Hong D-Y, Raven PH200712VitaceaeWenJThe families and genera of vascular plants, volume 9Berlin: Springer-VerlagKubitzki K20079466478LiCLVitaceae, Flora Reipublicae Popularis SinicaBeijing: Science Press1998482Systematics of Vitaceae in South AmericaLombardiJACanadian Journal of Botany20078571272110.1139/B07-021Is the genus Cissus (Vitaceae) monophyletic? Evidence from plastid and nuclear ribosomal DNARossettoMJackesBRScottKDHenryRJSystematic Botany2002273522533Morphological and anatomical development in the Vitaceae. VII. Floral development in Rhoicissus digitata with respect to other genera in the familyGerrathJMWilsonTPoslusznyUCanadian Journal of Botany20048219820610.1139/b03-120Phylogeny and biogeographic diversification of Maianthemum (Ruscaceae: Polygonatae)MengYWenJNieZ-LSunHYangYPMolecular Phylogenetic and Evolution200849242443410.1016/j.ympev.2008.07.017MUSCLE: multiple sequence alignment with high accuracy and high throughputEdgarRCNucleic Acids Research20043251792179710.1093/nar/gkh34039033715034147Structural evolution and flip-flop recombination of chloroplast DNA in the fern genus OsmundaSteinDBPalmerJDThompsonWFCurrent Genetics1986101183584110.1007/BF00418530MRBAYES: Bayesian inference of phylogenetic treesHuelsenbeckJPRonquistRBioinformatics20011775475510.1093/bioinformatics/17.8.75411524383MrModeltest v2. Program distributed by the author. Evolutionary Biology Centre, Uppsala UniversityNylanderJAA2004Gaps as characters in sequence-based phylogenetic analysesSimmonsMPOchoterenaHSystematic Biology200049236938112118412PAUP*: Phylogenetic analysis using parsimony (*and other methods), version 4.0b10SwoffordDLSunderland, Massachusetts: Sinauer Associates2003BEAST: Bayesian evolutionary analysis by sampling treesDrummondAJRambautABMC Evolutionary Biology2007721410.1186/1471-2148-7-214224747617996036DrummondAJHoSYWRawlenceNRambautAA rough guide to BEAST 1.42007http://code.google.com/p/beast-mcmc/Seed morphology of modern and fossil Ampelocissus (Vitaceae) and implications for phytogeographyChenIManchesterSRAmerican Journal of Botany20079491534155310.3732/ajb.94.9.153421636520History of Vitaceae inferred from morphology-based phylogeny and the fossil record of seedsChenIPhD thesisGainesville: University of Florida2009Rosid radiation and the rapid rise of angiosperm-dominated forestsWangHMooreMJSoltisPSBellCDBrockingtonSFAlexandreRDavisCCLatvisMManchesterSRSoltisDEProceedings of the National Academy of Sciences of the United States of America2009106103853385810.1073/pnas.0813376106264425719223592Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesisReeRHSmithSASystematic Biology200857141410.1080/1063515070188388118253896Bayesian phylogeography finds its rootsLemeyPRambautADrummondAJSuchardMAPLoS Computational Biology200959e100052010.1371/journal.pcbi.1000520274083519779555A likelihood framework for inferring the evolution of geographic range on phylogenetic treesReeRHMooreBRWebbCODonoghueMJEvolution200559112299231116396171Dispersal-vicariance analysis: A new approach to the quantification of historical biogeographyRonquistFSystematic Biology19974619520310.1093/sysbio/46.1.195Reconstructing ancestral ranges in historical biogeography: properties and prospectsLammKSRedelingsBDJournal of Systematics and Evolution200947536938210.1111/j.1759-6831.2009.00042.xProspects and challenges for parametric models in historical biogeographical inferenceReeRHSanmartinIJournal of Biogeography20093671211122010.1111/j.1365-2699.2008.02068.xAmazonian amphibian diversity is primarily derived from late Miocene Andean lineagesSantosJCColomaLASummersKCaldwellJPReeRCannatellaDCPLOS Biology200973e1000056The maximum likelihood approach to reconstructing ancestral character states of discrete characters on phylogeniesPagelMSystematic Biology19994861262210.1080/106351599260184LikelihoodEdwardsAWFBaltimore: Johns Hopkins University Press1992Recherches sur les methods d'identification et de classification des Vitacées temperées. II Thèse, présentée a la Faculté des SciencesGaletPUniversité de Montpellier1967Continental comparisons of temperate-zone tree species diversityLathamRERicklefsRESpecies Diversity in Ecological Communities: Historical and Geographical PerspectivesChicago: Univ. Chicago PressRicklefs RE, Schluter D1993294314Regional differences in rates of plant speciation and molecular evolution: A comparison between eastern Asia and eastern North AmericaXiangQ-YZhangWHRicklefsREQianHChenZDWenJLiJHEvolution200458102175218415568220Timing and modes of evolution of eastern Asian North American biogeographic disjunctions in seed plantsWenJIckert-BondSNieZ-LLiRDarwin's Heritage Today: Proceedings of the Darwin 2010 Beijing International ConferenceBeijing: Higher Education PressLong M, Gu H, Zhou Z201025226922330132Patterns in the assembly of temperate forests around the Northern HemisphereDonoghueMJSmithSAPhilosophical Transactions of the Royal Society of London Series B-Biological Sciences200435914501633164410.1098/rstb.2004.1538Northern hemisphere plant disjunctions: a window on tertiary land bridges and climate change?MilneRIAnnals of Botany200698346547210.1093/aob/mcl148280357616845136Biogeographical relationships of North American Tertiary florasManchesterSRAnnals of the Missouri Botanical Garden199986247252210.2307/2666183Meso-Cenozoic Caribbean paleogeography: Implications for the historical biogeography of the regionIturralde-VinentMAInternational Geology Review200648979182710.2747/0020-6814.48.9.791Some aspects of plant geography of the Northern Hemisphere during the late Cretaceous and TertiaryWolfeJAAnnals of the Missouri Botanical Garden197562226427910.2307/2395198Plate Tectonics and the Evolution of the Alpine SystemDeweyJFPitmanWCIRyanWBFBonninJGeological Society of America Bulletin197384103137318010.1130/0016-7606(1973)84<3137:PTATEO>2.0.CO;2Plate tectonics, seaways and climate in the historical biogeography of mammalsCoxCBMemórias do Instituto Oswaldo Cruz200095450951622331609Fruits and seeds of Brandon lignite. 1. VitaceaeTiffneyBHBarghoornESReview of Palaeobotany and Palynology19762216919110.1016/0034-6667(76)90001-4Virginia creeper as a winter food for birdsMcAteeWLAuk190623346347The dispersal of plants throughout the worldRidleyHNKent: L. Reeve1930Tropical intercontinental disjunctions: Gondwana breakup, immigration from the boreotropics, and transoceanic dispersalGivnishTJRennerSSInternational Journal of Plant Sciences20041654S1S610.1086/424022Recent long-distance dispersal overshadows ancient biogeographical patterns in a pantropical angiosperm family (Simaroubaceae, Sapindales)ClaytonJWSoltisPSSoltisDESystematic Biology200958439541010.1093/sysbio/syp04120525593Northern Hemisphere origin, transoceanic dispersal, and diversification of Ranunculeae DC. (Ranunculaceae) in the CenozoicEmadzadeKHörandlEJournal of Biogeography201138351753010.1111/j.1365-2699.2010.02404.xClades, clocks, and continents: Historical and biogeographical analysis of Myrtaceae, Vochysiaceae, and relatives in the Southern HemisphereSytsmaKJLittAZjhraMLPiresJCNepokroeffMContiEWalkerJWilsonPGInternational Journal of Plant Sciences20041654S85S10510.1086/421066Ancient vicariance or recent long-distance dispersal? Inferences about phylogeny and South American-African disjunctions in Rapateaceae and Bromeliaceae based on ndhF sequence dataGivnishTJMillamKCEvansTMHallJCPiresJCBerryPESytsmaKJInternational Journal of Plant Sciences20041654 suppls35s54Multiple Miocene Melastomataceae dispersal between Madagascar, Africa and IndiaRennerSSPhilosophical Transactions of the Royal Society of London Series B-Biological Sciences200435914501485149410.1098/rstb.2004.1530Evolution of Cenozoic seaways in the circum-Antarctic regionLawverLAGahaganLMPalaeogeography, Palaeoclimatology, Palaeoecology20031981-2113710.1016/S0031-0182(03)00392-4Historical biogeography of two cosmopolitan families of flowering plants: Annonaceae and RhamnaceaeRichardsonJEChatrouLWMolsJBErkensRHJPirieMDPhilosophical Transactions of the Royal Society of London Series B-Biological Sciences200435914501495150810.1098/rstb.2004.1537Vicariance or long-distance dispersal: historical biogeography of the pantropical subfamily Chrysophylloideae (Sapotaceae)BartishIVAntonelliARichardsonJESwensonUJournal of Biogeography201138117719010.1111/j.1365-2699.2010.02389.xA new combination in Clematicissus Planch. (Vitaceae)JackesBRRossettoMTelopea2006113390391Radiation of the Australian flora: what can comparisons of molecular phylogenies across multiple taxa tell us about the evolution of diversity in present-day communities?CrispMCookLSteaneDPhilosophical Transactions of the Royal Society of London Series B-Biological Sciences200435914501551157110.1098/rstb.2004.1528Origins of the southeastern Australian vegetationHillRSPhilosophical Transactions of the Royal Society B: Biological Sciences20043591537154910.1098/rstb.2004.1526Evolution of eastern Asian-Eastern North American biogeographic disjunctions: A few additional issuesWenJInternational Journal of Plant Sciences2001162S117S12210.1086/322940Climatic, floristic, and vegetational changes near the Eocene/Oligocene boundary in North AmericaWolfeJAEocene-Oligocene Biotic and Climatic EvolutionPrinceton: Princeton University PressProthero DR, Berggren WA1992421436Eocene-Oligocene climatic and biotic evolution: an overviewBerggrenWAProtheroDPEocene-Oligocene Climatic and Biotic EvolutionPrinceton: Princeton University PressProthero DP, Berggren WA1992128Tertiäre Vegetationsgeschichte EuropasMaiDHNew York: Fischer, Jena, Stuttgart1995Late Cretaceous and Cenozoic History of North American VegetationGrahamAOxford: Oxford University Press1999Neue Früchte und Samen aus paläozänen Ablagerungen MitteleuropasMaiDHFeddes Repertorium198798197229The mahogany family "out-of-Africa": divergence time estimation, global biogeographic patterns inferred from plastid rbcL DNA sequences, extant, and fossil distribution of diversityMuellnerANSavolainenVSamuelRChaseMWMolecular Phylogenetics and Evolution200640123625010.1016/j.ympev.2006.03.00116624592Laurasian migration explains Gondwanan disjunctions: evidence from MalpighiaceaeDavisCCBellCDMathewsSDonoghueMJProceedings of the National Academy of Sciences of the United States of America200299106833683710.1073/pnas.10217589912448911983870Floristic relationships between eastern Asia and North America: Test of Gray's hypothesisQianHAmerican Naturalist2002160331733210.1086/34152318707442Africa, the odd man out: Molecular biogeography of dalbergioid legumes (Fabaceae) suggests otherwiseLavinMThulinMLabatJNPenningtonRTSystematic Botany200025344946710.2307/2666689Evolution of biogeographic disjunction between eastern Asia and eastern North America in Phryma (Phrymaceae)NieZ-LSunHBeardsleyPMOlmsteadRGWenJAmerican Journal of Botany20069391343135610.3732/ajb.93.9.134321642199Evolution of the Madrean-Tethyan disjunctions and the North and South American amphitropical disjunctions in plantsWenJIckert-BondSMJournal of Systematics and Evolution200947533134810.1111/j.1759-6831.2009.00054.xPhylogeny, molecular and fossil dating, and biogeographic history of Annonaceae and Myristicaceae (Magnoliales)DoyleJASauquetHScharaschkinTLe ThomasAInternational Journal of Plant Sciences2004165suppl 4s35s67From Africa via Europe to South America: migrational route of a species-rich genus of Neotropical lowland rain forest trees (Guatteria, Annonaceae)ErkensRHJMaasJWCouvreurTLPJournal of Biogeography200936122338235210.1111/j.1365-2699.2009.02162.xTracing the impact of the Andean uplift on Neotropical plant evolutionAntonelliANylanderJAAPerssonCSanmartinIProceedings of the National Academy of Sciences of the United States of America2009106249749975410.1073/pnas.0811421106268573819470489Relationships within tribe Lycieae (Solanaceae): paraphyly of Lycium and multiple origins of gender dimorphismLevinRAMillerJSAmerican Journal of Botany200592122044205310.3732/ajb.92.12.204421646122Out of America to Africa or Asia: Inference of dispersal histories using nuclear and plastid DNA and the S-RNase self-incompatibility locusMillerJSKamathADamashekJLevinRAMolecular Biology and Evolution201128179380110.1093/molbev/msq25320855430Phylogeny and biogeography of the genus Lycium (Solanaceae): inferences from chloroplast DNA sequencesFukudaTYokoyamaJOhashiHMolecular Phylogenetics and Evolution20011924625810.1006/mpev.2001.092111341807

xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EF86T1MCB_DNUY1J INGEST_TIME 2012-04-04T13:43:12Z PACKAGE AA00010466_00001

xml version 1.0 encoding utf-8 standalone no
mets ID sort-mets_mets OBJID sword-mets LABEL DSpace SWORD Item PROFILE METS SIP Profile xmlns http:www.loc.govMETS
xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
xsi:schemaLocation http:www.loc.govstandardsmetsmets.xsd
metsHdr CREATEDATE 2012-03-12T20:19:35
name BioMed Central
dmdSec sword-mets-dmd-1 GROUPID sword-mets-dmd-1_group-1
epdcx:descriptionSet xmlns:epdcx http:purl.orgeprintepdcx2006-11-16 xmlns:MIOJAVI
epdcx:description epdcx:resourceId sword-mets-epdcx-1
epdcx:statement epdcx:propertyURI http:purl.orgdcelements1.1type epdcx:valueURI http:purl.orgeprintentityTypeScholarlyWork
epdcx:valueString Evolution of the intercontinental disjunctions in six continents in the Ampelopsis clade of the grape family (Vitaceae)
The Ampelopsis clade (Ampelopsis and its close allies) of the grape family Vitaceae contains ca. 43 species disjunctly distributed in Asia, Europe, North America, South America, Africa, and Australia, and is a rare example to study both the Northern and the Southern Hemisphere intercontinental disjunctions. We reconstruct the temporal and spatial diversification of the Ampelopsis clade to explore the evolutionary processes that have resulted in their intercontinental disjunctions in six continents.
The Bayesian molecular clock dating and the likelihood ancestral area analyses suggest that the Ampelopsis clade most likely originated in North America with its crown group dated at 41.2 Ma (95% HPD 23.4 61.0 Ma) in the middle Eocene. Two independent Laurasian migrations into Eurasia are inferred to have occurred in the early Miocene via the North Atlantic land bridges. The ancestor of the Southern Hemisphere lineage migrated from North America to South America in the early Oligocene. The Gondwanan-like pattern of intercontinental disjunction is best explained by two long-distance dispersals: once from South America to Africa estimated at 30.5 Ma (95% HPD 16.9 45.9 Ma), and the other from South America to Australia dated to 19.2 Ma (95% HPD 6.7 22.3 Ma).
The global disjunctions in the Ampelopsis clade are best explained by a diversification model of North American origin, two Laurasian migrations, one migration into South America, and two post-Gondwanan long-distance dispersals. These findings highlight the importance of both vicariance and long distance dispersal in shaping intercontinental disjunctions of flowering plants.
Nie, Ze-Long
Sun, Hang
Manchester, Steven R
Meng, Ying
Luke, Quentin
Wen, Jun
http:purl.orgeprinttermsisExpressedAs epdcx:valueRef sword-mets-expr-1
http:purl.orgdcelements1.1language epdcx:vesURI http:purl.orgdctermsRFC3066
epdcx:sesURI http:purl.orgdctermsW3CDTF 2012-02-08
BioMed Central Ltd
http:purl.orgeprinttermsstatus http:purl.orgeprinttermsStatus
Nie et al.; licensee BioMed Central Ltd.
http:purl.orgdctermsaccessRights http:purl.orgeprinttermsAccessRights
BMC Evolutionary Biology. 2012 Feb 08;12(1):17
http:purl.orgdctermsURI http://dx.doi.org/10.1186/1471-2148-12-17
fileGrp sword-mets-fgrp-1 USE CONTENT
file sword-mets-fgid-0 sword-mets-file-1
FLocat LOCTYPE URL xlink:href 1471-2148-12-17.xml
sword-mets-fgid-1 sword-mets-file-2 applicationpdf
sword-mets-fgid-3 sword-mets-file-3 applicationmsword
structMap sword-mets-struct-1 structure LOGICAL
div sword-mets-div-1 DMDID Object
sword-mets-div-2 File


RESEARCHARTICLE OpenAccessEvolutionoftheintercontinentaldisjunctionsin sixcontinentsinthe Ampelopsis cladeofthe grapefamily(Vitaceae)Ze-LongNie1,5,HangSun1,StevenRManchester2,YingMeng1,3,QuentinLuke4andJunWen5*AbstractBackground: The Ampelopsis clade( Ampelopsis anditscloseallies)ofthegrapefamilyVitaceaecontainsca.43 speciesdisjunctlydistributedinAsia,Europe,NorthAmerica,SouthAmerica,Africa,andAustralia,andisarare exampletostudyboththeNorthernandtheSouthernHemisphereintercontinentaldisjunctions.Wereconstruct thetemporalandspatialdiversificationofthe Ampelopsis cladetoexploretheevolutionaryprocessesthathave resultedintheirintercontinentaldisjunctionsinsixcontinents. Results: TheBayesianmolecularclockdatingandthelikelihoodancestralareaanalysessuggestthatthe Ampelopsis clademostlikelyoriginatedinNorthAmericawithitscrowngroupdatedat41.2Ma(95%HPD23.4-61.0Ma)inthe middleEocene.TwoindependentLaurasianmigrationsintoEurasiaareinferredtohaveoccurredintheearlyMiocene viatheNorthAtlanticlandbridges.TheancestoroftheSouthernHemispherelineagemigratedfromNorthAmericato SouthAmericaintheearlyOligocene.TheGondwanan-likepatternofintercontinentaldisjunctionisbestexplainedby twolong-distancedispersals:oncefromSouthAmericatoAfricaestimatedat30.5Ma(95%HPD16.9-45.9Ma),and theotherfromSouthAmericatoAustraliadatedto19.2Ma(95%HPD6.7-22.3Ma). Conclusions: Theglobaldisjunctionsinthe Ampelopsis cladearebestexplainedbyadiversificationmodelof NorthAmericanorigin,twoLaurasianmigrations,onemigrationintoSouthAmerica,andtwopost-Gondwanan long-distancedispersals.Thesefindingshighlighttheimportanceofbothvicarianceandlongdistancedispersalin shapingintercontinentaldisjunctionsoffloweringplants.BackgroundUnderstandingtheunderlyingmechanismsfortheevolutionofwide-rangingdisjunctpatternshaslongbeena majorfocusofbiogeography[1-5].Taxadisjunctatthe globallevelinvolvingbothNorthernandSouthernHemisphereareparticularlyinformativebecausetheirhistories mayhavegeneralimplicatio nsforothergroups.BiogeographichistoryintheNorthernHemisphereiscomplicated,buthasusuallybeenexplainedbythewidespread distributionoftheBoreotropicalfloraintheEoceneand followedbyappearanceofmoretemperateforestelementsduringthemid-Tertiaryextirpationsofthermophilicelementsinresponsetoclimaticcoolingepisodes ofthelateEoceneandthePlio-Pleistocene[6-11].The SouthernHemisphereisofteninterpretedtoshowa vicariancepatternattributedtothesequentialbreakupof Gondwananlandmasses[12].Recentstudieson Nothofagus havedemonstratedtherelevanceoflongdistancedispersalratherthanvicarianceinshapingGondwanan distributionalpatterns[13-16]. Commonlythreemainroutesforthemigrationoftaxa betweentheNorthernandtheSouthernHemisphere havebeenrecognized(Figure1).Thefirstistheopening ofbioticexchangesbetweenNorthandSouthAmericaat varioustimesintheTertiary[1,17].Thesecondhypotheticalmigrationrouteisthetrans-Tethyandispersal betweenEuropeandAfrica[18].ThethirdislesscommonconcerningthepossibleroutebetweenAsiaand AustraliaintheMioceneandlater[18,19].Thesethree routescanbeviewedasalternativehypothesesfortheex situoriginofelementsofglobaldiversity. *Correspondence:wenj@si.edu5DepartmentofBotany,NationalMuseumofNaturalHistory,MRC166, SmithsonianInstitution,Washington,DC20013-7012,USA FulllistofauthorinformationisavailableattheendofthearticleNie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 2012Nieetal;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommons AttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,andreproductionin anymedium,providedtheoriginalworkisproperlycited.


Therearemanyexamplesofglobaldistributionbetween theSouthernandtheNorthernHemisphere,particularly inpantropicalfamilies(e.g.,Rubiaceae,Annonaceae,Lauraceae)orinmanylargecosmopolitangenera,suchas Ranunculus [20], Senecio [21],and Lobelia [22,23].Nevertheless,allofthemhavecontinuousdistributionfromthe tropicstothetemperatezones.Molecularstudiesusually showaverycomplexdisjuncthistoryoflargetaxawith relativelylowresolution[24,25].Therearealsosometaxa withrelativelyfewspeciesthatexhibitanintercontinental disjunctioninvolvingboththeNorthernandtheSouthern Hemisphere.However,thedisjunctioninsuchtaxausually involvesonlyoneortwonorthernorsouthernlandmasses. Examplesinclude Caltha (Ranunculaceae)withaglobal disjunctionbutabsentfromAfrica[26]and Thamnosma (Rutaceae)disjunctbetweenNorthAmericaandAfrica [27].Ofgreatinterest,wehaverecentlyfoundatrulyglobaldisjunctpatterninasmallgroupincluding Ampelopsis Michx.anditsrelativetaxafromthegrapefamily,Vitaceae [28-30]. Figure1 Delimitationofthesixareasfortaxainthe Ampelopsis cladeandtheschematicmodelsusedintheLagrangeanalyses .N= NorthAmerica;A=easternAsia;E=EuropetocentralAsia;F=Africa;S=SouthAmerica;andU=Australia. Nie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page2of13


Vitaceaeisawell-knowngroupoffloweringplantshavingalargelypantropicaldistributioninAsia,Africa,Australia,theneotropics,andthePacificislands,withonlya fewgeneraintemperateregions[31-33].Ascurrentlycircumscribed, Ampelopsis isoneofthefewgeneramostly restrictedtothenorthtempe ratezone.Ithasapproximately25speciesdisjunctlydistributedinEurasia(c.22 spp.)andNorthandCentralAmerica(3spp.).Recent phylogeneticanalysesbasedonplastidornuclear sequencesrevealedthatthereareatleasttwodisjunctionsbetweentheNewandtheOldWorldinthegenus [28,29].Moreinterestingly,bothchloroplastandnuclear dataclearlysuggestedthattheAfrican Rhoicissus Planch. andtheSouthAmerican Cissusstriata Ruiz&Pav.complexformacladenestedwithintheparaphyletic Ampelopsis [28,29]. Rhoicissus consistsofabout12species endemictotropicalands outhernAfrica.The Cissus striata complexcontainsfourspeciesfromSouthAmerica[29,34].Furthermore,theAustraliangenus Clematicissus Planch.seemedtobecloselyrelatedtothe Cissus striata Rhoicissus cladebasedonplastiddata[30,35]. Thereareonlytwo Clematicissus speciesknownfrom Australia: C.opaca fromAustralia seasternregionand C.angustissima fromthewestcoast[30].Therefore,the NorthernHemisphere Ampelopsis anditsthreeclose relatives(Rhoicissus Clematicissus andthe Cissusstriata complex)fromtheSouthernHemisphere(hereafter referredtoasthe Ampelopsis clade)demonstrateanunusualglobalintercontinentaldisjunctpatterninvolvingsix continents(Figure1).Yetmorphologically,the Ampelopsis cladeseemstobeheterogeneous,suchasinhaving 4-7-merousflowersandfleshytodryfruits[29]. The Ampelopsis cladeoffersagoodopportunityto exploretheoriginandevolutionoftheglobalintercontinentaldisjunctpatterninfl oweringplants,especially concerningbothNorthernandSouthernHemisphere intercontinentaldisjunctions.Ahierarchicalglobaldistributionwaspredictedbyo urpreviousstudies,but withlimitedsampling[28,29].Wehereinemployphylogenetic,moleculardating,andbiogeographicmethodsto reconstructtheevolutionaryhistoryofthe Ampelopsis cladebasedonacomprehensivesamplingschemeusing fourplastidregions( trnL-F rps16 psbA-trnH ,and atpBrbcL ).MethodsTaxonsampling,DNAsequencing,andphylogenetic analysesWesample28ofthe43species(65%)ofthe Ampelopsis cladeincludingallthreeNorthAmerican Ampelopsis species,15ofthe22Eurasian Ampelopsis ,fiveofthe12 Rhoicissus speciesfromAfrica,threeofthefourspecies fromthe Cissusstriata complexfromSouthAmerica, andthetwoAustralian Clematicissus species(Additional file1,TableS1).Thesamplingcoverstheentireextant geographicrangeof Ampelopsis anditscloserelatives fromboththeNorthernandSouthernHemisphere.But westillhavesomemissingtaxainoursamplingscheme, suchas Rhoicissus withonly41%speciessampled.However,eachgroupissupposedtobemonophyleticbased onmorphological,biogeographic,andmolecularevidence[28-30,34,36].Themis singtaxainoursampling shouldhavelittleeffectinthepresentstudythatfocused onphylogeneticrelationshipsamonggeneraandbiogeographicevolutionattheintercontinentallevel.Inorder toplaceouranalysesofthe Ampelopsis cladeinabroad frameworkonthefamilylevel,wesampled66additional taxafromtheothermajorgroupsinVitaceae(i.e.,the Vitis Parthenocissus Ampelocissus clade,thecore Cissus clade,andthe Cyphostemma Caryatia Tetrastigma clade)plusthree Leea speciesofLeeaceaebased onpreviousinvestigations[28,29]. TotalDNAswereextractedfromsilicageldriedleaves usingtheDneasyPlantMiniKit(QIAGEN,Crawley, UK).AmplificationandsequencingfollowedSoejimaand Wen(2006)for trnL-F rps16 ,and atpB-rbcL,andMeng etal .[37]for psbA-trnH.DNAsequenceswere assembledusingSequencher v4.1.4(GeneCodesCorp., AnnArbor,Michigan,USA).Sequencealignmentwas initiallyperformedusingMUSCLE3.8.31[38]inthe multiplealignmentroutinefollowedbymanualadjustmentinSe-Alv2.0a11(http://tree.bio.ed.ac.uk/software/ seal/).Thechloroplastgenomeisgenerallyconsideredas oneunitwithoutrecombinationalthoughtherehave beenreportsofrecombinationinthechloroplastgenome [39].Therefore,wecombinedalltheplastiddata( trnL-F rps16 psbA-trnH ,and atpB-rbcL )inouranalysis.The combinedplastiddatawereanalyzedusingBayesian inferenceasimplementedinMrBayes3.1.2[40].The best-fitmodelofnucleotidesubstitution(GTR+I+G) wasdeterminedbyMrModelTest2.3[41]usingthe AkaikeInformationCriterion(AIC).Variationofgapsin oursequencesisnotcomplicate.Atotalof31binary characterswerecodedforgapsaccordingtoSimmons andOchoterena[42]andseparatedintoindependent partitioninallanalyses.Bayesiantreetopologyandposteriorprobabilities(PP)weredeterminedfromtwoindependentrunsoffourincrementallyheatedchains.Runs wereperformedfor5milliongenerationswithsampling oftreesevery500thgeneration.Whenthelog-likelihood scoreswerefoundtohavesta bilized,aconsensustree wascalculatedafteromittingthefirst10%oftreesas burn-in.DivergencetimeestimationFormoleculardatinganalyses,thestrictmolecularclock modelwasrejectedfromourdatasetbasedonalikelihoodratiotestperformedinPAUP*[43].Therefore,weNie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page3of13


estimatednodeageswithinthe Ampelopsis groupusing aBayesianrelaxedclockmodelasimplementedin BEASTv1.6.1[44].WelargelyfollowedthedatingstrategiesinNie etal .(2010),whichanalyzeddiversification in Parthenocissus ofVitaceae.Afteroptimaloperator adjustmentassuggestedbytheoutputdiagnosticsfrom severalpreliminaryBEASTru ns,twofinalindependent runs(each50milliongenerations)wereperformedona clusterofMacXServesusedf oranalysisofbiological dataattheSmithsonianInstitution(http://topazweb.si. edu).Tracerversion1.5wasusedtocheckforconvergencebetweentheruns[44]. Afterdiscardingthefirst 10%samplesasburn-in,thetreesandparameterestimatesfromthetworunswerecombinedusingLogCombiner1.6.1[44].Results wereconsideredreliable oncetheeffectivesamplingsize(ESS)forallparameters exceeded200assuggestedbytheprogrammanual[45]. Thesamplesfromtheposteriorweresummarizedon themaximumcladecredibili tytreeusingtheprogram TreeAnnotator1.6.1[44]withposteriorprobabilitylimit of0.5andmeannodeheightssummarized. FossilseedsofVitaceaecanbedifferentiatedtothe genericlevel[46,47].Theoldestbestpreservedseedfossil ofthefamilyisfromthelatePaleoceneoftheBeicegel CreeklocalityinNorthDakota.Thisfossilisundoubtedly assignedto Ampelocissus s.l.(as A.parvisemina Chen& Manchester)andiseasilydistinguishedfromallother vitaceousgenerabyitslong,parallelventralinfoldsanda centrallypositionedovalchalazalscar[46].Sincethe Ampelocissus s.l.isnotmonophyleticwith Vitis nested withinit[28],the A.parvisemina fossilthusmayrepresentanearlymemberofthe Ampelocissus claderetaining somecharacterssharedwit hitscommonancestorto Vitis [46].Thestemageofthe Ampelocissus-Vitis clade wasthusfixedat58.55.0millionyearsago(Ma). FortherootageofVitaceae,Nie etal .(2010)fixedthe splitbetweenVitaceaeand Leea as854basedonthe estimatedageof78-92MabyWikstrm etal .(2001). Recently,Bell etal .(2010)reportedanestimateranging from65(45 81)to48(21 79)Maforthecrownage Vitis Leea clade,whichisroughlyconsistentwiththe earliestfossilevidencesofVitaceaeinthePalaeocene [46].However,theirresultsmayhaveunderestimatedfor Vitaceaebecausetheoldestfossilof A.parvisemina is undoubtedlyassignedtothe Ampelocissus s.l.within VitaceaeandthefamilyispredictedtohaveaCretaceous historyinviewofitsbasalpositionintherosidsandthe presenceofCretaceousrosidfossils[48].ThetimeestimatesofangiospermsbyMa gallnandCastillo(2009) alsosuggestedapre-Tertiaryoriginas90.65(90.47 90.84)to90.82(90.64 91)MaforVitaceae.TheinferencesfromMagallnandCastillo(2009)andWikstrm etal.(2001)areclose,althoughthelaterwascriticized forthenonparametricratesmoothingmethodandfor calibratingthetreeusingonlyasinglecalibrationpoint. Therefore,weusedtheestimatefromMagallnandCastillo(2009)andsetthenormalpriordistributionof90.7 1.0Maforthestemageofthefamily.Alowstandard errorwasusedbecauseofthenarrow95%confidence fromMagallnandCastillo(2009).AncestralareareconstructionSeveralmethodshavebeenrecentlyproposedthattake intoaccountofgeneticbranchlengths,phylogenetic uncertainty,andbranchlengthuncertaintyforreconstructingdistributionalc hangethrough evolutionary time,usingeithermaximumlikelihood[49]orBayesian inference[50].Theancestralareaofthe Ampelopsis cladewasreconstructedwiththelikelihoodanalysis usingtheprogramLagrangeversion20110117[49,51]. UnliketheDIVAmethod[52],thislikelihoodapproach incorporatesanexplicitdispersal-extinction-cladogenesis (DEC)modelofdispersalrou tesavailableathistorical intervalscorrelatingstochasticeventswithlineagepersistence[49].Thelikelihoodanalysisispronetoestimatewideancestralrangesforearly-branchinglineages [53-55].Inourcase,ancestralrangeswereassumedto includenomorethantwoareassinceallextantspecies inthe Ampelopsis cladearerestrictedintoonlyone area.Moreover,spatialandtemporalconstraints(e.g., areadistances,continentconnections,datesofgeological origin)maybeimposedintheDECmodelestimation, providingamoreaccurateestimationoftheancestral rangesandhypothesistestingofdifferentgeographic scenarios.WedidnotconducttheBayesiancalculation ofancestralgeographicdistributionswithstandardcontinuous-timeMarkovchains(CTMCs),becausegeologic information(e.g.,thepresenceanddissolutionofland bridgesandislandchains)isn otexplicitlyincorporated intotheanalyses. Wehereinusedthelikelihoodmethodtotestanull modelandsixalternativebiogeographicscenarios(Figure1) basedonthehypothesizeddispersalormigrationroutes betweentheNorthernandtheSouthernHemisphere.Six areasweredelimitedbycontinentaldivisionsandtheextant distributionsofthe Ampelopsis clade:1)N NorthAmerica includingCentralAmerica;2)S SouthAmerica;3)F Africa;4)A easternAsia;5)E EuropetocentralAsia; and6)U Australia(Figure1).Th eunconstrainednull model(M0)assumesthatspatialandtemporaldistribution hasnoeffectonbiogeographicpatternsofevolutionand allowsgeographicrangestoincludeanypossiblecombinationofcontinentsandpermitsdirectdispersalbetweenany areapairs.TheM1modelfavorsamigrationroutefrom NorthtoSouthAmerica(N S)withthebiogeographic connectionsbetweenEuropeandAfrica(E F)andeastern AsiaandAustralia(A U)excludedfromouranalyses. Similarly,theM2modelconsiderstheconnectionbetweenNie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page4of13


E Fanddidnotallowotherpossibilities.Themigration routebetweenAsiaandAustralia(M3)seemslesslikely, butwestillconsidereditinouranalysesasacomparison. Wealsotestmodelsthatallowtwoconnectionsbetween theNorthernandSouthernhemispheres(M4-M6in Figure1).FollowingRee etal .(2005),theresultsbetween modelswereassessedbydirectlycomparingtheirlog-likelihoodscores.Theconventionalcut-offvalueovertwologlikelihoodunitswasconsideredstatisticallysignificant,and modelswithlowerlikelihood scorewererejected[56,57].ResultsThetotallengthofthealigneddatamatrixis3933bp. TheBayesianconsensustreeishighlycongruentwiththe maximumcladecredibilitytreeobtainedfromBEAST andthelaterisshowninFigure2withPPsupportvalues >0.50.Ourresultssupportthemonophylyofthe Ampelopsis cladewiththreemajorlineagesresolvedwithinthe Ampelopsis clade(Figure2).Twodistinctlineages(hereafternamedasNorthIandII)correspondtothetwosectionsof Ampelopsis [58].NorthIincludesallspeciesof section Leeaceifoliae withpinnatelytobipinnatelycompoundleaves.NorthIIconsistsoftaxaofsection Ampelopsis withsimpleorpalmately-dividedorpalmatelycompoundleaves.TheSouthernHemispheretaxa(the African Rhoicissus,theSouthAmerican Cissusstriata complex,andtheAustralian Clematicissus )formaclade (theSouthgroupinFigure2). Atotalof100milliongenerations(2runsof50milliongenerationseach)arenecessarytoreachsufficient ESS.The Ampelopsis cladeisestimatedtohavediverged fromitscloserelativesinVitaceaeat41.2(23.4-61.0) MaintheEocene.TheBayesianestimatesalsosuggest thatalltheothermajorcladesofVitaceae(e.g., Vitis Parthenocissus Ampelocissus clade,core Cissus clade, and Cyphostemma Cayratia Tetrastigma )hadalready diversifiedintheEocene(Figure2).Agesofmajor groupswithinthe Ampelopsis cladeobtainedinour studyaresummarizedinTable1. Patternsoftemporalandspatialdistributionofthe Ampelopsis cladeareinferredusingthemaximumlikelihoodDECmethod.Wecomparesevenmodels(i.e.,anull andsixalternatives)forthesixareas(Figure1)andthe effectsofdifferentmodelsonlikelihoodreconstructions areshowninTable1.AnalysesbasedonM1,M4,andM5 typicallyhavelowerlikelihoodscoresthanothermodels andproducednearlyidenticalresults(Table1).Forexample,allofthemsuggestedthattheancestralrangesplitat thestemSouthlineageisbetweenNorthandSouthAmerica(Node2inTable1).Ourresultsalsosuggestthatthe threemodels(M1,M4,andM5)aresignificantlydifferent fromtheothers(M2,M3,andM6)withscoresovertwo log-likelihoodunits.TheM1modelissuggestedasthe bestonewiththehighestlikelihoodscore,andthismodel isshowninFigure3.DiscussionNorthAmericanoriginThe Ampelopsis cladeiscomposedoftwodistinctLaurasianlineageseachdisjunctbetweentheOldandtheNew WorldandoneSouthgroupwithaGondwana-likeintercontinentaldisjunction:(Africa(Australia,andSouth America))(Figure3).ThemostlikelymodelM1inthe DECanalysessuggestedthatthe Ampelopsis cladehadan earlydiversificationinNorthAmericawithageographic splitbetweenN(NorthAmerica)andNS(Northand SouthAmerica)(Table1;Figure3).Thefactthatmost fossilrecordsofthefamilyincludingtheoldestrecordin thePaleocenearefoundfromNorthAmerica[46]isconsistentwiththis outofNorthAmerica hypothesis. AlthoughSouthAmericaisinferredaspartoftheancestralareas(Table1),itseemslesslikelytobetheancestral areaofthe Ampelopsis cladethanNorthAmericabecause thereareveryfewfossilsknownbeforetheEoceneof SouthAmerica[47].Furthermore,phylogeneticresults alsocontradictthepossibi lityofSouthAmericanorigin becausetheSouthAmericangroupiswellembedded withinthe Ampelopsis clade(Figure2).Asiaalsoseems lesslikelythanNorthAmericatobetheancestralareaof the Ampelopsis clade,inspiteofitshighestextantspecies richnessofthelineage.Noseedrecordswithagesyounger thanOligoceneareknownfromAsia[47]. Theancestralareaforataxonisusuallyexpectedtobe correlatedwithhighextantspeciesrichness.Forexample, easternAsiausuallyhasahigherlevelofspeciesdiversity andendemism,andhasbeensuggestedtobetheancestral areaformanyeasternAsian-easternNorthAmericandisjunctgroups[59-63].DonoghueandSmith(2004)founda predominanceofdirectionalityfromAsiatotheNew World.Ofthe29lineagestheyanalyzedwithaneastern AsianandeasternNorthAmericandisjunction,onlyone lineageshoweddirectionalityfromeasternNorthAmerica toeasternAsia.However,Wen etal .(2010)reported manymorelineageswithNorthAmericanoriginsand subsequentmigrationsintoeasternAsia,with29ofthe total98examined(30%)lineagesmigrated/dispersedfrom theNewWorldtotheOldWorld.Itseemsthateastern Asiahasbeenover-emphasizedasanancestralareafor Laurasiantaxaduetoitsretentionofthegreaternumber ofspecies[62,64].NorthAmericaissupportedtohave playedanimportantroleintheearlyevolutionofthetwo Ampelopsis lineagesinspiteofthelowerspeciesdiversity todayinNorthAmericacomparedwitheasternAsia.The lowerspeciesrichnessinNorthAmericaisoftenexplained bythehypothesisthatbothNorthAmericaandeastern AsiawereoccupiedbyBoreotropicalelementsintheearlyNie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page5of13


TertiarybutNorthAmericasufferedmoresevereextinctionswithglobalcoolingbeginninginthelateEoceneor Oligocene[10,63].Thehighlevelofspeciesdiversityand endemisminAsiacanalsobeattributedtosecondary diversificationduetohabitatheterogeneityaswellasa lowerrateofextinctionsinthelateTertiary[60,61]. 1.00 0.95 1.00 0.94 1.00 1 1.00 1 0.90 1 0.98 1.00 1.00 0.99 1.00 0.99 1.00 1 1.00 0.96 1.00 1 1 0.83 1 1.00 0.96 0.86 1 1 1.00 0.85 1 1 1 1.00 1.00 0.96 1.00 1.00 1.00 1 1 1.00 0.94 North I North II1.00 1.00 1.00 1.00 0.98 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.93 0.71 1.00 1.00 0.98 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00South group1.00 1.00 1.00 1.00 1 00 096 1.00 100 00 100 00 098 1.00 1.0 0 1.00 100 1.00 100 0.95 09 094 0.83 1.00 00 0.93 100 1.0 0 1.00 1.00 100 0 3 0.93 100 1.0 0 0 1.00 90 086 1.00 00 100 00 100 .98 098 1.00 0 1.0 0 00 00 0 00 1.00 10 0 1 00 00 085 00 1 00 0.98 095 1.00 1.00 0.99 09 1.00 1.0 0 100 1.00 1 2 3 4 5 0 Eocene Oligocene Paleocene Cretaceous Miocene Plio 0.0 Ma 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 Ampelocissus ascendiora Ampelopsis grossedentata Parthenocissus quinquefolia Tetrastigma sichouense Ampelocissus javalensis Cissus erosa Ampelocissus africana Cissus simsiana Ampelopsis aconitifolia Cissus repens Vitis mengziensis Ampelocissus obtusata Cissus trianae Leea indica Cyphostemma maranguense Cissus striata ssp. striata Ampelopsis brevipedunculata Parthenocissus semicordata Yua thomsonii Leea guineensis Ampelopsis cordata Cyphostemma thomasii Cyphostemma serpens Ampelopsis glandulosa var. heterophylla Ampelocissus acapulcensis Ampelopsis cantoniensis Cissus granulosa Cissus pileata Cissus nodosa Ampelopsis glandulosa var. kulingensis Parthenocissus suberosa Caryatia sp. Parthenocissus vitacea Tetrastigma serrulatum Ampelocissus thyrsiora Cissus integrifolia Vitis riparia Vitis rotundifolia Cissus albiporcata Ampelopsis arborea Tetrastigma caudatum Ampelopsis megalophylla Cayratia geniculata Cissus striata ssp. argentina Rhoicissus revoilii Parthenocissus feddei Rhoicissus tomentosa Vitis vulpina Cissus aralioides Ampelopsis hypoglauca Ampelopsis humulifolia Cissus javana Vitis poponoii Cissus verticillata Tetrastigma obovatum Ampelopsis orientalis Cyphostemma kilimandscharicum Cayratia wrayi Cayratia mollissima Rhoicissus digitata Ampelopsis delavayana Ampelocissus elegans Cissus phymatocarpa Cissus adnata Ampelopsis denudata Parthenocissus chinensis Tetrastigma ceratopetalum Leea gonioptera Cayratia japonica Ampelopsis chaffanjonii Ampelopsis bodinieri Vitis davidiana Ampelocissus africana Ampelopsis rubifolia Ampelocissus erdwendbergii Clematicissus opaca Rhoicissus rhomboidea Ampelopsis acutidentata Clematicissus angustissima Tetrastigma erubescens Ampelocissus elephantina Vitis aestivalis Vitis exuosa Cissus producta Rhoicissus tridentata Tetrastigma henryi Cayratia maritima Rhoicissus tomentosa Ampelopsis japonica Cayratia cordifolia Ampelopsis clade Figure2 MaximumcladecredibilitytreeinferredwithBEAST,withthe95%highestposteriordensityindicatedbygraybars .Nodesof interestsweremarkedas0to5asinTable1;andcalibrationsareindicatedwithblackstars.ValuesabovebranchesrepresentBayesianposterior probabilities. Nie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page6of13


Table1Resultsofmoleculardatingandancestralrangereconstructionformajornodeswithinthe Ampelopsis clade.-lnLDE Node0: Crown Ampelopsis clade Node1: CrownNorthI Node2: StemSouthgroup Node3: CrownSouthgroup Node4: Crown Cissusstriata complexClematicissus Node5: Disjunct Moleculardating withBEAST(Ma) 41.2(23.4-61.0)21.8(6.2-26.3)35.5(20.7-52.3)30.5(16.9-45.9) 17.4(7.4-29.0) 19.2(6.7-22.3) M032.420.0034010.004318N|NF(0.25) A|N(0.81)F|N(0.32) S|F(0.22), U|F(0.21) U|S(0.63) N|A(0.47) M130.190.0070850.003586N|NS(0.61)A|N(0.85)S|N(0.92) S|F(0.54) U|S(0.79) N|A(0.46), N|E(0.36) M234.470.009860.008992 A|E(0.25), N|E(0.20) A|N(0.51)F|E(0.55) F|F(0.39) U|S(0.32) N|E(0.51) M332.940.0075390.005598A|AU(0.60) A|N(0.68)U|A(0.90) U|F(0.42), U|U(0.41) U|S(0.77) N|A(0.72) M430.650.0064860.003784N|NS(0.61) A|N(0.84)S|N(0.91) S|F(0.51) U|S(0.78) N|A(0.46), N|E(0.35) M530.890.0060960.003555N|NS(0.58) A|N(0.84)S|N(0.86) S|F(0.51) U|S(0.79) N|A(0.48), N|E(0.35) M633.30.0070280.005918A|AU(0.55) A|N(0.66)U|A(0.80) U|U(0.40), U|F(0.37) U|S(0.73) N|A(0.65)ThenodesofinterestareshowninFigure2,andthelikelihoodscores(-lnL),andestimatesofdispersal(D)andextinction(E)rates(eventspermillionyears)aregiven.M0isanullmodelwithoutconstraintsandM1M6arealternativemodelsintheLagrangeanalyses.Onlythehighestrelativeprobabilityisshown.Boldtextrepresentsthemodelwithasignificantlybetterlikelihoodwhencomparedwiththeothermodelstested (morethantwolog-likelihoodsbetter)Nie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page7of13


DiversificationpatternintheSouthernHemisphereThelikelihoodanalysesusingLagrangebasedonM1, M4,andM5suggestthatthemostlikelyroutebetween theNorthernandtheSouthernHemisphereisfrom NorthtoSouthAmerica,althoughmoreoptionsofconnectionsarepermittedinM4(betweenEuropeand Africa)andM5(betweeneasternAsiaandAustralia) (Table1andFigure3). Ampelopsis mayhavedispersed intoSouthAmericaviascatteredcontinentaland/orvolcanicislandsthatconnectedNorthandSouthAmerica atvarioustimesintheTertiary[1,17],suchasviathe protoGreaterAntilles(ca.50Ma)orviaGAARlandia thatexistedaround33-35Ma[17,65].Theseparation ofthesouthernlineagefromitsLaurasianancestorat 35.5(20.7-52.3)MainthelateEocenebroadlycoincideswithapossiblebiolo gicalconnectionbetween Oligocene Eocene early Miocene 0 Ma 10 20 30 40 North I North IISouth group U S S S S U F F A F A A A A N A A A A A A A A E N N A F F F AAmpelopsis grossedentata Cissus simsiana Ampelopsis aconitifolia Cissus striata ssp. striata Ampelopsis brevipedunculata Ampelopsis cordata Ampelopsis glandulosa var. heterophylla Ampelopsis cantoniensis Cissus granulosa Ampelopsis glandulosa var. kulingensis Ampelopsis arborea Ampelopsis megalophylla Cissus striata ssp. argentina Rhoicissus revoilii Rhoicissus tomentosa Ampelopsis hypoglauca Ampelopsis humulifolia Ampelopsis orientalis Rhoicissus digitata Ampelopsis delavayana Ampelopsis denudata Ampelopsis chaffanjonii Ampelopsis bodinieri Ampelopsis rubifolia Clematicissus opaca Rhoicissus rhomboidea Ampelopsis acutidentata Clematicissus angustissima Rhoicissus tridentata Rhoicissus tomentosa Ampelopsis japonica N S N N S N A A S N S F U N N A E Eor N N N S S S A A A E E E F F F U U U Figure3 Biogeographicscenariofortheglobaldisjunctionofthe Ampelopsis cladebasedonmoleculardatingandthebestM1 modelwiththehighestlikelihoodscoreintheancestralrangeanalyses .Longdistancedispersalisindicatedasdashlinesandmigration assolidlines.Theancestralareaofthe Ampelopsis cladeisshownwithblackstarsonthemaps.Thetreebranchesandrangesonthetreeare codedasfollows:blue=NorthAmerica(N);grey=easternAsia(A);orange=EuropeandcentralAsia(E);green=SouthAmerica(S);yellow= Africa(F);andred=Australia(U). Nie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page8of13


NorthandSouthAmericaaroundtheEocene-Oligocene boundary[66].FossilseedsfoundfromtheEoceneof SouthAmericaarecloselyrelatedtothoseoftheCentralAmerican Ampelocissus ,indicatingthepossiblefloristicconnectionbetweenNorthandSouthAmericaat thattime[46]. AnotherpossiblemigrationrouteisthattheLaurasian ancestorsofthe Ampelopsis cladereachedAfricausing theBoreotropicalconnectionviatheNorthAtlanticand EuropeinthelateEocenetoearlyOligocene.However, thishypothesisisrejectedbytheLagrangeanalysis(see M2inTable1).ThemodelM4withtwopossibleconnectionsbetweentheNorthernandSouthernHemisphere(N-SandE-F,Figure1)didnotsupportthe European-African(E-F)route.Althoughtheseparation ofthesouthernlineagefromitsLaurasianancestorin theearlyOligocenebroadlycoincideswiththedisruptionoftheBoreotropicalfloraaroundtheEocene-Oligoceneboundary[66],boththeshallowseasthatseparate AfricafromEurasiaandthedrybeltinnorthernAfrica werebarrierstobioticexchangebetweenthetwocontinentsintheearlytomidTertiary[1,67,68].Thethird hypothesis(theM3model,Table1)isthat Ampelopsis enteredtheSouthernHemi sphereviatheAsian-Australianconnection.Thismodelseemsquiteunlikely basedonouranalyses(Table1).ThemodelM5that permitstwoconnectionsbetweenNorthernandSouthernHemisphere(N-S,A-U,Figure1)alsoprefersthe connectionbetweenNorthandSouthAmerica(N-S) ratherthantheAsian-Australianconnection(A-U). TheavailabilityofbioticinterchangebetweenAustralia andAsiabeginningattheMiocene[18]istoorecentto supportthisscenario. Thedivergencetime30.5(95%HPD:16.9-45.9)Ma intheearlyOligocenewasestimatedforthefirstsplit betweenSouthAmericaandAfrica(node3inFigure2 andTable1).Thistimeiswellafterthelastpossible connectionofAfricaandothersouthernlandmassesat around96-105Ma[18,19].Wethusarguethatlongdistancedispersal(LDD)isthem ostplausiblemechanism fortheirsouthernintercontinentaldisjunction.Vitaceae taxaareusuallydispersedbyanimals,especiallybirds [32,69-71].Alltaxainthe Ampelopsis cladeexceptthe Australian Clematicissusangustissima bearfleshyberries thatmayhavefacilitatedLDD.Inparticular,LDDhas beenrecentlyacceptedbyanumberofstudiesasthe drivingforceforplantdisjunctionsintheSouthern Hemisphere,especiallyforthosewithrelativelyrecent divergencetimes[5,20,72-74].Biogeographicstudieson Vochysiaceae[75]suggestedaLDDfromSouthAmerica toAfricaintheOligocene.Givnishetal.[76]showed thatthesingleAfricangenus Maschocephalus ofRapateaceaeisofrecentorigininthelateMioceneand reachedAfricafromSouthAmericaviaLDD.Dispersals betweenAfricaandSouthAmericahavealsobeensuggestedforanumberofwell-studiedtaxa,suchasin Melastomataceae[77],andSimaroubaceae[73]. Ourresultssupportthatthesouthernlineageof Ampelopsis arrivedinAustraliafromSouthAmericain theearlyMiocene(node4inFigure2).Migration betweenAustraliaandSouthAmericamaybealternativelyexplainedbyatrans-Antarcticexchange[12].This AntarcticrouteexistedduringthelateCretaceous-early Tertiaryandwasinterrup tedonlyinthelateEocene (30-35Ma)whentheSouthTasmanSeaopenedup betweenAustraliaandeasternAntarctica[19,78].This routeissupportedbyevidencefromseveralplant groups,suchasAnnonaceae[79],andSapotaceae[80]. YetthesplitbetweentheSouthAmerican Cissusstriata complexandtheAustralian Clematicissus intheearly MioceneistooyoungtobeexplainedbyanAntarctic migration.LDDisthemostplausibleexplanationfor thisdisjunction.Finally,thedisjunctionofthetwo Clematicissus speciesineasternandwesternAustralia [30,81]mayrepresentarelictdistributionandtheir divergencetimeinthePlioceneisconsistentwiththe aridificationincentralAustraliaatthattime[82,83].LaurasianmigrationsTheDECreconstructionsuggestsaNorthAmericaneasternAsiansplit(N|A)fortheNorthIdisjunctionand aNorthAmerican-easternAsianorNorthAmericanEuropesplit(N|AorN|E)forthedisjunctioninthe NorthIIlineage(Table1;Figure3).ThetwoNorthern Hemispheredisjunctionsma yhaveinvolvedtheNorth AtlanticlandbridgesortheBeringlandbridgefrom NorthAmericatoeasternAsia[10,84].Weprefertouse theNorthAtlanticroutebecauseitiswellsupportedin sect. Ampelopsis (NorthIIinFigure3).Thesouthern NorthAmerican Ampelopsisdenudata divergedfirst,followedbythesoutheasternUS A.cordata .Thewestern Asian/southernEuropean A.orientalis isthensisterto thelargeAsianclade.Therelativepositionoftheseareas inthecladogramiscongruentwiththemigrationofthe lineagefromNorthAmericatoEuropeacrosstheNorth Atlanticlandbridges,andthelineagethenreachedAsia subsequently(Figure3). TheBayesianmolecularclockdatingwithfossilcalibrationsuggestsanearlyMiocenesplitofthetwodisjunct groupsin Ampelopsis (nodes1and5inFigure2).The divergencetimesarealsoconsistentwiththepossibility oftheNorthAtlanticmigrationroute.Basedonpaleogeological,zoological,andbotanicalfossilevidence,Tiffney(1985b)arguedfortheimportanceoftheNorth AtlanticlandbridgestotropicalorwarmtemperatespeciesintheearlyEocenetomiddleMiocene.Acontinuous beltofBoreotropicalelementscoveredmuchofthe southernpartofNorthAmerica,southernEurasia,andNie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page9of13


northwesternAfricaintheEocene[9,66].Atthattime, plantmigrationsthroughdirectlandconnectionor acrosslimitedwatergapswerepossiblethroughthe NorthAtlanticlandbridges.Significantcoolingduring theOligoceneresultedinsouthwardretreatsandthe extirpationofsomelineagescomprisingthisflora [64,85,86].Agradualwarmingperiodoccurredintothe earlyMiocene,resultingintheexpansionofsomeevergreenandthermophiliclineagesinEuropeandNorth America[87,88].DispersalofBoreotropicalorwarm temperatethermophilicelements,suchas Ampelopsis ,is thereforeconsideredlikelyacrosstheNorthAtlanticland bridgesduringthisperiod. Thereareveryfewextant VitaceaespeciesinEurope,butmanyvitaceousseeds werereportedintheearlyTertiaryofEurope[47,89]. Together,theselinesofevidencestronglysuggestthat the Ampelopsis clademayhaveusedthecorridorsviathe NorthAtlanticlandbridgesasapathwaytoreachEurasia intheearlyMiocene.ConclusionsOurresultssuggestacomplexhistoryofdiversification inthe Ampelopsis cladetoexplaintheglobaldisjunctionsthatincludesaNorthAmericanorigin,twoLaurasianmigrations,onemigrationintoSouthAmerica,and twopost-GondwananLDDs.Thesefindingsmayhave generalimplicationsfortheoriginanddiversificationof plantswithglobaldisjunctions.Asia,Africa,orSouth Americahasoftenbeensuggestedastheancestralarea formanyintercontinentaldisjunctgroups[4,90,91].Evidencefromthe Ampelopsis cladesuggeststhatNorth Americamayhaveplayedanimportantroleintheoriginofsomemodernfloweringplantsinspiteofitsoften lowerspeciesdiversitywhencomparedwithotherareas, suchaseasternAsia[92]orAfrica[93].RecentbiogeographicanalysesofseveralothergroupsappeartoprovideadditionalexamplesofNorthAmericanorigins (e.g., Phryma [94];Simaroubaceae[73]).Thisstudyalso highlightstheimportanceoftheNorthandSouth Americanrouteintheglobaldiversificationbetweenthe NorthernandtheSouthernHemisphere[95].Thisroute apparentlyplayedanimportantroleinthewidedistributionofmanypantropicalplantsintheearlyTertiary, suchasAnnonaceae[96,97],Malpighiaceae[91],and Rubiaceae[98]. Recentbiogeographicanalyseshaveunderscoredthe relativeimportanceofLDDtointercontinentaldisjunctionsintheSouthernHemispherethantraditionally assumed[3,13].Thefitbetweenareacladogramsand thehistoryoftectonicfragmentationmighthavebeen overstated[5].Ifweacceptedtheresultsofourcalibrationanalysesasabsolute,ratherthanminimumages, thentheSouthernHemispherecladeistooyoungto havebeenachievedbytheGondwananbreakup.The Gondwana-likedisjunctioninthe Ampelopsis cladewas reconstructedtohaveaNor thAmericanoriginwithan initialmigrationintoSouthAmericaandthendispersed fromSouthAmericaintoAfricaandAustraliaindependentlyviaLDD.Asimilarexamplefrom Lycium (85 spp.,Solanaceae)wassuggestedaNewWorldoriginof Lycium withrecentdispersalfromtheAmericasto Africa,andthentoeasternAsia[99-101].Dispersalhas beenhypothesizedtobethedominantpatterninthis genusthathasred,fleshy,bird-dispersedfruits[99].On theotherhand,however,theLaurasianlineagesin Ampelopsis cladefavoravicariancemigrationpattern fromNorthAmericaviatheNorthAtlanticlandbridges toEurasia.OurresultsthussupportboththeLaurasian migrationsandthepost-GondwananLDDtoexplainthe globaldisjunctionofthe Ampelopsis clade.AdditionalmaterialAdditionalfile1:TableS1.VoucherinformationandGenBank accessionnumbersoftheAmpelopsiscladeandrepresentative taxainVitaceae .Abbreviationsofherbariaareasfollows:KUN,Kunming InstituteofBotany,ChineseAcademyofSciences;andUS,theUnited StatesNationalHerbarium.AccessionnumbersbeginningwithJQ indicatesequencesgeneratedforthisstudyandtheotherswere obtainedfromGenBank.Adashmeanssequencesmissing. Acknowledgements ThisstudywassupportedbygrantsfromtheNationalScienceFoundation (DEB0743474toSRManchesterandJWen),theNaturalScienceFoundation ofChina(NSFC31061160184toHSunand31129001toJWen),theNSFCYunnanJointproject(U1136601toHSun),andtheJohnDandCatherine TMacArthurFoundation(toJWen,RRee,andGMueller).Laboratorywork wasconductedinandpartiallysupportedbytheLaboratoryofAnalytical BiologyoftheNationalMuseumofNaturalHistory,theSmithsonian Institution.FieldworkinNorthAmericawassupportedbytheSmallGrants ProgramoftheNationalMuseumofNaturalHistory,theSmithsonian Institution.MichaelNee,MaurizioRossetto,IhsanAl-Shehbaz,andDaigui Zhangkindlyhelpedwiththesamplecollection. Authordetails1KeyLaboratoryofBiodiversityandBiogeography,KunmingInstituteof Botany,ChineseAcademyofSciences,Kunming,Yunnan650204,China.2FloridaMuseumofNaturalHistory,UniversityofFlorida,Gainesville,FL 32611,USA.3InstituteofTibetanPlateauResearchatKunming,Kunming InstituteofBotany,ChineseAcademyofSciences,Kunming,Yunnan650204, China.4EastAfricanHerbarium,NationalMuseumsofKenya,Nairobi00502, Kenya.5DepartmentofBotany,NationalMuseumofNaturalHistory,MRC 166,SmithsonianInstitution,Washington,DC20013-7012,USA. Authors contributions JW,ZLN,HS,andSRMconceivedtheideas;JW,YM,QLandZLNcollected thematerials;ZLNandYManalyzedthedata;andZLNandJWledthe writing.Allauthorsreadandapprovedthefinalsubmission. Received:4October2011Accepted:8February2012 Published:8February2012 References1.RavenPH,AxelrodDI: Angiospermbiogeographyandpastcontinental movements. AnnalsoftheMissouriBotanicalGarden 1974, 61(3) :539-673.Nie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page10of13


2.RavenPH: Plantspeciesdisjunctions:asummary. AnnalsoftheMissouri BotanicalGarden 1972, 59(2) :234-246. 3.UpchurchP: Gondwananbreak-up:legaciesofalostworld? Trendsin Ecology&Evolution 2008, 23(4) :229-236. 4.DonoghueMJ,BellCD,LiJH: PhylogeneticpatternsinNorthern Hemisphereplantgeography. InternationalJournalofPlantSciences 2001, 162 :S41-S52. 5.deQueirozA: Theresurrectionofoceanicdispersalinhistorical biogeography. TrendsinEcology&Evolution 2005, 20(2) :68-73. 6.GrahamA: Outlineoftheoriginandhistoricalrecognitionoffloristic affinitiesbetweenAsiaandeasternNorthAmerica. In Floristicsand PaleofloristicsofAsiaandEasternNorthAmerica. Editedby:GrahamA. Amsterdam:Elsevier;1972:1-18. 7.MilneRI,AbbottRJ: Theoriginandevolutionoftertiaryrelictfloras. AdvancesinBotanicalResearch 2002, 38 :281-314. 8.TiffneyBH: Perspectivesontheoriginofthefloristicsimilaritybetween easternAsiaandeasternNorthAmerica. JournaloftheArnoldArboretum 1985, 66(1) :73-94. 9.TiffneyBH: TheEoceneNorthAtlanticlandbridge:itsimportancein tertiaryandmodernphytogeographyoftheNorthernhemisphere. JournaloftheArnoldArboretum 1985, 66(2) :243-273. 10.WenJ: EvolutionofeasternAsianandeasternNorthAmericandisjunct distributionsinfloweringplants. AnnualReviewofEcologyandSystematics 1999, 30 :421-455. 11.SanmartnI,EnghoffH,RonquistF: Patternsofanimaldispersal,vicariance anddiversificationintheHolarctic. BiologicalJournaloftheLinneanSociety 2001, 73(4) :345-390. 12.SanmartnI,RonquistF: SouthernHemispherebiogeographyinferredby event-basedmodels:Plantversusanimalpatterns. SystematicBiology 2004, 53(2) :216-243. 13.KnappM,StocklerK,HavellD,DelsucF,SebastianiF,LockhartPJ: Relaxed molecularclockprovidesevidenceforlong-distancedispersalof Nothofagus (southernbeech). PLosBiology 2005, 3(1) :38-43. 14.CookLG,CrispMD: Notsoancient:theextantcrowngroupof Nothofagus representsapost-Gondwananradiation. Proceedingsofthe RoyalSocietyB-BiologicalSciences 2005, 272(1580) :2535-2544. 15.LinderHP,CrispMD: Nothofagus andPacificbiogeography. Cladistics 1995, 11(1) :5-32. 16.SwensonU,HillRS,McLoughlinS: Biogeographyof Nothofagus supports the sequenceofGondwanabreak-up. Taxon 2001, 50(4) :1025-1041. 17.Iturralde-VinentMA,MacPheeRDE: PaleogeographyoftheCaribbean region:Implicationsforcenozoicbiogeography. BulletinoftheAmerican MuseumofNaturalHistory 1999,, 238: 1-95. 18.MorleyRJ: Interplatedispersalpathsformegathermalangiosperms. PerspectivesinPlantEcology,EvolutionandSystematics 2003, 6(1-2) :5-20. 19.McLoughlinS: ThebreakuphistoryofGondwanaanditsimpactonpreCenozoicfloristicprovincialism. AustralianJournalofBotany 2001, 49(3) :271-300. 20.EmadzadeK,GehrkeB,PeterLinderH,HrandlE: Thebiogeographical historyofthecosmopolitangenus Ranunculus L.(Ranunculaceae)inthe temperatetomeridionalzones. MolecularPhylogeneticsandEvolution 2011, 58(1) :4-21. 21.PelserPB,NordenstamB,KadereitJW,WatsonLE: AnITSphylogenyof tribeSenecioneae(Asteraceae)andanewdelimitationof Senecio L. Taxon 2007, 56 :1077-1104. 22.AntonelliA: Higherlevelphylogenyandevolutionarytrendsin Campanulaceaesubfam.Lobelioideae:Molecularsignalovershadows morphology. MolecularPhylogeneticsandEvolution 2008, 46(1) :1-18. 23.AntonelliA: Havegiantlobeliasevolvedseveraltimesindependently? Lifeformshiftsandhistoricalbiogeographyofthecosmopolitanand highlydiversesubfamilyLobelioideae(Campanulaceae). BMCBiology 2009, 7 :82. 24.SchuettpelzE,HootSB,SamuelR,EhrendorferF: Multipleoriginsof SouthernHemisphere Anemone (Ranunculaceae)basedonplastidand nuclearsequencedata. PlantSystematicsandEvolution 2002, 231(1) :143-151. 25.MeyerKM,HootSB,ArroyoMTK: PhylogeneticaffinitiesofSouth American Anemone (Ranunculaceae),includingtheendemicsegregate genera, Barneoudia and Oreithales InternationalJournalofPlantSciences 2010, 171(3) :323-331. 26.SchuettpelzE,HootSB: Phylogenyandbiogeographyof Caltha (Ranunculaceae)basedonchloroplastandnuclearDNAsequences. AmericanJournalofBotany 2004, 91(2) :247-253. 27.ThivM,vanderNietT,RutschmannF,ThulinM,BruneT,LinderHP: OldNewWorldandtrans-Africandisjunctionsof Thamnosma (Rutaceae): Intercontinentallong-distancedispersalandlocaldifferentiationinthe succulentbiome. AmericanJournalofBotany 2010, 98(1) :76-87. 28. SoejimaA,WenJ: Phylogeneticanalysisofthegrapefamily(Vitaceae) basedonthreechloroplastmarkers. AmericanJournalofBotany 2006, 93(2) :278-287. 29.WenJ,NieZ-L,SoejimaA,MengY: PhylogenyofVitaceaebasedonthe nuclearGAI1genesequences. CanadianJournalofBotany 2007, 85(8) :731-745. 30.RossettoM,CraynDM,JackesBR,PorterC: Anupdatedestimateof intergenericphylogeneticrelationshipsintheAustralianVitaceae. CanadianJournalofBotany 2007, 85 :722-730. 31.ChenZD,RenH,WenJ: Vitaceae. In FloraofChina.Volume12. Editedby: WuZ-Y,HongD-Y,RavenPH.BeijingandSt.Louis,Mo.:SciencePressand MissouriBotanicalGardenPress;2007. 32.WenJ: Vitaceae. In Thefamiliesandgeneraofvascularplants,volume9. Volume9. Editedby:KubitzkiK.Berlin:Springer-Verlag;2007:466-478. 33.LiCL:In Vitaceae,FloraReipublicaePopularisSinica.Volume48. Beijing: SciencePress;1998(2). 34.LombardiJA: SystematicsofVitaceaeinSouthAmerica. CanadianJournal ofBotany 2007, 85 :712-721. 35.RossettoM,JackesBR,ScottKD,HenryRJ: Isthegenus Cissus (Vitaceae) monophyletic?EvidencefromplastidandnuclearribosomalDNA. SystematicBotany 2002, 27(3) :522-533. 36.GerrathJM,WilsonT,PoslusznyU: Morphologicalandanatomical developmentintheVitaceae.VII.FloraldevelopmentinRhoicissus digitatawithrespecttoothergenerainthefamily. CanadianJournalof Botany 2004, 82 :198-206. 37.MengY,WenJ,NieZ-L,SunH,YangYP: Phylogenyandbiogeographic diversificationof Maianthemum (Ruscaceae:Polygonatae). Molecular PhylogeneticandEvolution 2008, 49(2) :424-434. 38.EdgarRC: MUSCLE:multiplesequencealignmentwithhighaccuracyand highthroughput. NucleicAcidsResearch 2004, 32(5) :1792-1797. 39.SteinDB,PalmerJD,ThompsonWF: Structuralevolutionandflip-flop recombinationofchloroplastDNAintheferngenus Osmunda Current Genetics 1986, 10(11) :835-841. 40.HuelsenbeckJP,RonquistR: MRBAYES:Bayesianinferenceof phylogenetictrees. Bioinformatics 2001, 17 :754-755. 41.NylanderJAA: MrModeltestv2.Programdistributedbytheauthor. EvolutionaryBiologyCentre,UppsalaUniversity. 2004. 42. SimmonsMP,OchoterenaH: Gapsascharactersinsequence-based phylogeneticanalyses. SystematicBiology 2000, 49(2) :369-381. 43.SwoffordDL: PAUP*:Phylogeneticanalysisusingparsimony(*andother methods),version4.0b10. Sunderland,Massachusetts:SinauerAssociates; 2003. 44.DrummondAJ,RambautA: BEAST:Bayesianevolutionaryanalysisby samplingtrees. BMCEvolutionaryBiology 2007, 7 :214. 45.DrummondAJ,HoSYW,RawlenceN,RambautA: AroughguidetoBEAST 1.4 2007[http://code.google.com/p/beast-mcmc/]. 46.ChenI,ManchesterSR: Seedmorphologyofmodernandfossil Ampelocissus (Vitaceae)andimplicationsforphytogeography. American JournalofBotany 2007, 94(9) :1534-1553. 47.ChenI: HistoryofVitaceaeinferredfrommorphology-basedphylogeny andthefossilrecordofseeds. PhDthesis Gainesville:UniversityofFlorida; 2009. 48.WangH,MooreMJ,SoltisPS,BellCD,BrockingtonSF,AlexandreR, DavisCC,LatvisM,ManchesterSR,SoltisDE: Rosidradiationandtherapid riseofangiosperm-dominatedforests. ProceedingsoftheNational AcademyofSciencesoftheUnitedStatesofAmerica 2009, 106(10) :3853-3858. 49.ReeRH,SmithSA: Maximumlikelihoodinferenceofgeographicrange evolutionbydispersal,localextinction,andcladogenesis. Systematic Biology 2008, 57(1) :4-14. 50.LemeyP,RambautA,DrummondAJ,SuchardMA: Bayesian phylogeographyfindsitsroots. PLoSComputationalBiology 2009, 5(9) : e1000520.Nie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page11of13


51.ReeRH,MooreBR,WebbCO,DonoghueMJ: Alikelihoodframeworkfor inferringtheevolutionofgeographicrangeonphylogenetictrees. Evolution 2005, 59(11) :2299-2311. 52.RonquistF: Dispersal-vicarianceanalysis:Anewapproachtothe quantificationofhistoricalbiogeography. SystematicBiology 1997, 46 :195-203. 53.LammKS,RedelingsBD: Reconstructingancestralrangesinhistorical biogeography:propertiesandprospects. JournalofSystematicsand Evolution 2009, 47(5) :369-382. 54.ReeRH,SanmartinI: Prospectsandchallengesforparametricmodelsin historicalbiogeographicalinference. JournalofBiogeography 2009, 36(7) :1211-1220. 55.SantosJC,ColomaLA,SummersK,CaldwellJP,ReeR,CannatellaDC: AmazonianamphibiandiversityisprimarilyderivedfromlateMiocene Andeanlineages. PLOSBiology 2009, 7(3) :e1000056. 56.PagelM: Themaximumlikelihoodapproachtoreconstructingancestral characterstatesofdiscretecharactersonphylogenies. SystematicBiology 1999, 48 :612-622. 57.EdwardsAWF: Likelihood. Baltimore:JohnsHopkinsUniversityPress;1992. 58.GaletP: Recherchessurlesmethodsd identificationetdeclassification desVitacestemperes.IIThse,prsentealaFacultdesSciences. UniversitdeMontpellier;1967. 59.LathamRE,RicklefsRE: Continentalcomparisonsoftemperate-zonetree speciesdiversity. In SpeciesDiversityinEcologicalCommunities:Historical andGeographicalPerspectives. Editedby:RicklefsRE,SchluterD.Chicago: Univ.ChicagoPress;1993:294-314. 60.XiangQ-Y,ZhangWH,RicklefsRE,QianH,ChenZD,WenJ,LiJH: Regional differencesinratesofplantspeciationandmolecularevolution:A comparisonbetweeneasternAsiaandeasternNorthAmerica. Evolution 2004, 58(10) :2175-2184. 61.WenJ,Ickert-BondS,NieZ-L,LiR: Timingandmodesofevolutionof easternAsian-NorthAmericanbiogeographicdisjunctionsinseed plants. In Darwin sHeritageToday:ProceedingsoftheDarwin2010Beijing InternationalConference. Editedby:LongM,GuH,ZhouZ.Beijing:Higher EducationPress;2010:252-269. 62.DonoghueMJ,SmithSA: Patternsintheassemblyoftemperateforests aroundtheNorthernHemisphere. PhilosophicalTransactionsoftheRoyal SocietyofLondonSeriesB-BiologicalSciences 2004, 359(1450) :1633-1644. 63.MilneRI: Northernhemisphereplantdisjunctions:awindowontertiary landbridgesandclimatechange? AnnalsofBotany 2006, 98(3) :465-472. 64.ManchesterSR: BiogeographicalrelationshipsofNorthAmericanTertiary floras. AnnalsoftheMissouriBotanicalGarden 1999, 86(2) :472-522. 65.Iturralde-VinentMA: Meso-CenozoicCaribbeanpaleogeography: Implicationsforthehistoricalbiogeographyoftheregion. International GeologyReview 2006, 48(9) :791-827. 66. WolfeJA: SomeaspectsofplantgeographyoftheNorthernHemisphere duringthelateCretaceousandTertiary. AnnalsoftheMissouriBotanical Garden 1975, 62(2) :264-279. 67.DeweyJF,PitmanWCI,RyanWBF,BonninJ: PlateTectonicsandthe EvolutionoftheAlpineSystem. GeologicalSocietyofAmericaBulletin 1973, 84(10) :3137-3180. 68.CoxCB: Platetectonics,seawaysandclimateinthehistorical biogeographyofmammals. MemriasdoInstitutoOswaldoCruz 2000, 95(4) :509-516. 69.TiffneyBH,BarghoornES: FruitsandseedsofBrandonlignite.1.Vitaceae. ReviewofPalaeobotanyandPalynology 1976, 22 :169-191. 70.McAteeWL: Virginiacreeperasawinterfoodforbirds. Auk 1906, 23 :346-347. 71.RidleyHN: Thedispersalofplantsthroughouttheworld. Editedby:Kent: L.Reeve1930. 72.GivnishTJ,RennerSS: Tropicalintercontinentaldisjunctions:Gondwana breakup,immigrationfromtheboreotropics,andtransoceanicdispersal. InternationalJournalofPlantSciences 2004, 165(4) :S1-S6. 73.ClaytonJW,SoltisPS,SoltisDE: Recentlong-distancedispersal overshadowsancientbiogeographicalpatternsinapantropical angiospermfamily(Simaroubaceae,Sapindales). SystematicBiology 2009, 58(4) :395-410. 74.EmadzadeK,HrandlE: NorthernHemisphereorigin,transoceanic dispersal,anddiversificationofRanunculeaeDC.(Ranunculaceae)inthe Cenozoic. JournalofBiogeography 2011, 38(3) :517-530. 75.SytsmaKJ,LittA,ZjhraML,PiresJC,NepokroeffM,ContiE,WalkerJ, WilsonPG: Clades,clocks,andcontinents:Historicalandbiogeographical analysisofMyrtaceae,Vochysiaceae,andrelativesintheSouthern Hemisphere. InternationalJournalofPlantSciences 2004, 165(4) :S85-S105. 76.GivnishTJ,MillamKC,EvansTM,HallJC,PiresJC,BerryPE,SytsmaKJ: Ancientvicarianceorrecentlong-distancedispersal?Inferencesabout phylogenyandSouthAmerican-AfricandisjunctionsinRapateaceaeand BromeliaceaebasedonndhFsequencedata. InternationalJournalofPlant Sciences 2004, 165(4suppl) :s35-s54. 77.RennerSS: MultipleMioceneMelastomataceaedispersalbetween Madagascar,AfricaandIndia. PhilosophicalTransactionsoftheRoyal SocietyofLondonSeriesB-BiologicalSciences 2004, 359(1450) :1485-1494. 78.LawverLA,GahaganLM: EvolutionofCenozoicseawaysinthecircumAntarcticregion. Palaeogeography,Palaeoclimatology,Palaeoecology 2003, 198(1-2) :11-37. 79.RichardsonJE,ChatrouLW,MolsJB,ErkensRHJ,PirieMD: Historical biogeographyoftwocosmopolitanfamiliesoffloweringplants: AnnonaceaeandRhamnaceae. PhilosophicalTransactionsoftheRoyal SocietyofLondonSeriesB-BiologicalSciences 2004, 359(1450) :1495-1508. 80. BartishIV,AntonelliA,RichardsonJE,SwensonU: Vicarianceorlongdistancedispersal:historicalbiogeographyofthepantropicalsubfamily Chrysophylloideae(Sapotaceae). JournalofBiogeography 2011, 38(1) :177-190. 81.JackesBR,RossettoM: Anewcombinationin Clematicissus Planch. (Vitaceae). Telopea 2006, 11(3) :390-391. 82.CrispM,CookL,SteaneD: RadiationoftheAustralianflora:whatcan comparisonsofmolecularphylogeniesacrossmultipletaxatellusabout theevolutionofdiversityinpresent-daycommunities? Philosophical TransactionsoftheRoyalSocietyofLondonSeriesB-BiologicalSciences 2004, 359(1450) :1551-1571. 83.HillRS: OriginsofthesoutheasternAustralianvegetation. Philosophical TransactionsoftheRoyalSocietyB:BiologicalSciences 2004, 359 :1537-1549. 84.WenJ: EvolutionofeasternAsian-EasternNorthAmericanbiogeographic disjunctions:Afewadditionalissues. InternationalJournalofPlantSciences 2001, 162 :S117-S122. 85.WolfeJA: Climatic,floristic,andvegetationalchangesneartheEocene/ OligoceneboundaryinNorthAmerica. In Eocene-OligoceneBioticand ClimaticEvolution. Editedby:ProtheroDR,BerggrenWA.Princeton: PrincetonUniversityPress;1992:421-436. 86.BerggrenWA,ProtheroDP: Eocene-Oligoceneclimaticandbiotic evolution:anoverview. In Eocene-OligoceneClimaticandBioticEvolution. Editedby:ProtheroDP,BerggrenWA.Princeton:PrincetonUniversityPress; 1992:1-28. 87.MaiDH: TertireVegetationsgeschichteEuropas. NewYork:Fischer,Jena, Stuttgart;1995. 88.GrahamA: LateCretaceousandCenozoicHistoryofNorthAmerican Vegetation. Oxford:OxfordUniversityPress;1999. 89.MaiDH: NeueFrchteundSamenauspaloznenAblagerungen Mitteleuropas. FeddesRepertorium 1987, 98 :197-229. 90.MuellnerAN,SavolainenV,SamuelR,ChaseMW: Themahoganyfamily out-of-Africa":divergencetimeestimation,globalbiogeographic patternsinferredfromplastid rbcL DNAsequences,extant,andfossil distributionofdiversity. MolecularPhylogeneticsandEvolution 2006, 40(1) :236-250. 91.DavisCC,BellCD,MathewsS,DonoghueMJ: Laurasianmigrationexplains Gondwanandisjunctions:evidencefromMalpighiaceae. Proceedingsof theNationalAcademyofSciencesoftheUnitedStatesofAmerica 2002, 99(10) :6833-6837. 92.QianH: FloristicrelationshipsbetweeneasternAsiaandNorthAmerica: TestofGray shypothesis. AmericanNaturalist 2002, 160(3) :317-332. 93.LavinM,ThulinM,LabatJN,PenningtonRT: Africa,theoddmanout: Molecularbiogeographyofdalbergioidlegumes(Fabaceae)suggests otherwise. SystematicBotany 2000, 25(3) :449-467. 94. NieZ-L,SunH,BeardsleyPM,OlmsteadRG,WenJ: Evolutionof biogeographicdisjunctionbetweeneasternAsiaandeasternNorth Americain Phryma (Phrymaceae). AmericanJournalofBotany 2006, 93(9) :1343-1356. 95.WenJ,Ickert-BondSM: EvolutionoftheMadrean-Tethyandisjunctions andtheNorthandSouthAmericanamphitropicaldisjunctionsinplants. JournalofSystematicsandEvolution 2009, 47(5) :331-348.Nie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page12of13


96.DoyleJA,SauquetH,ScharaschkinT,LeThomasA: Phylogeny,molecular andfossildating,andbiogeographichistoryofAnnonaceaeand Myristicaceae(Magnoliales). InternationalJournalofPlantSciences 2004, 165(suppl4):s35-s67. 97.ErkensRHJ,MaasJW,CouvreurTLP: FromAfricaviaEuropetoSouth America:migrationalrouteofaspecies-richgenusofNeotropical lowlandrainforesttrees( Guatteria ,Annonaceae). JournalofBiogeography 2009, 36(12) :2338-2352. 98.AntonelliA,NylanderJAA,PerssonC,SanmartinI: Tracingtheimpactof theAndeanupliftonNeotropicalplantevolution. Proceedingsofthe NationalAcademyofSciencesoftheUnitedStatesofAmerica 2009, 106(24) :9749-9754. 99.LevinRA,MillerJS: RelationshipswithintribeLycieae(Solanaceae): paraphylyof Lycium andmultipleoriginsofgenderdimorphism. AmericanJournalofBotany 2005, 92(12) :2044-2053. 100.MillerJS,KamathA,DamashekJ,LevinRA: OutofAmericatoAfricaor Asia:InferenceofdispersalhistoriesusingnuclearandplastidDNAand theS-RNaseself-incompatibilitylocus. MolecularBiologyandEvolution 2011, 28(1) :793-801. 101.FukudaT,YokoyamaJ,OhashiH: Phylogenyandbiogeographyofthe genus Lycium (Solanaceae):inferencesfromchloroplastDNAsequences. MolecularPhylogeneticsandEvolution 2001, 19 :246-258.doi:10.1186/1471-2148-12-17 Citethisarticleas: Nie etal .: Evolutionoftheintercontinental disjunctionsinsixcontinentsinthe Ampelopsis cladeofthegrape family(Vitaceae). BMCEvolutionaryBiology 2012 12 :17. Submit your next manuscript to BioMed Central and take full advantage of: Convenient online submission Thorough peer review No space constraints or color gure charges Immediate publication on acceptance Inclusion in PubMed, CAS, Scopus and Google Scholar Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Nie etal BMCEvolutionaryBiology 2012, 12 :17 http://www.biomedcentral.com/1471-2148/12/17 Page13of13


Taxa Voucher/Source Locality trnL F atpB rbcL rps16 psbA trnH Ampelocissus acapulcensis (Kunth) Planch. Wen 8696 (US) Mexico JF437281 JQ182472 JQ182627 JF437058 Ampelocissus africana (Lour.) Merr. Luke & Luke 11449 (US) Kenya, Makindu JQ182553 JQ18244 8 JQ182603 JQ182507 Ampelocissus africana (Lour.) Merr. Luke & Luke 11536 (US) Tanzania, Udzungwa Mountain JQ182550 JQ182444 JQ182599 JQ182504 Ampelocissus ascendiflora Latiff Wen 8422 (US) Malaysia, Selangor AB234982 JQ182430 JQ182583 Ampelocissus elegans Gagnep. Wen 7507 (US) Singapore, Bukit Timah Nature Reserve AB234981 JQ182427 JQ182581 Ampelocissus elephantina Planch. Wen 9583 (US) Madagascar HM585516 HM585792 HM585659 Ampelocissus erdwendbergii Planch. Wen 8697 (US) Mexico JQ182569 JQ 182471 JQ182626 JQ182527 Ampelocissus javalensis (Seem.) W.D. Stevens & A. Pool Wen 6920 (US) Costa Rica AB234984 AB234911 AB234943 Ampelocissus obtusata (Welw. ex Baker) Planch. Luke & Luke 11590 (US) Tanzania, Inyonga JQ182556 JQ182457 JQ182612 JQ1 82510 Ampelocissus thyrsiflora (Blume) Planch. Deden 870 (US) Indonesia, SE Sulawesi JQ182546 JQ182438 JQ182593 JQ182499 Ampelopsis aconitifolia Bunge Wen 8518 (US) China, Beijing JQ182560 JQ182461 JQ182616 JQ182517


Ampelopsis acutidentata W.T. Wang Wen et al. (Tibet MacArthur) 3479 (KUN, US) China, Yunnan JQ182561 JQ182462 JQ182617 JQ182518 Ampelopsis arborea Kochne Wen 7164 (US) USA, Alabama AB234990 AB234946 JQ182487 Ampelopsis bodinieri (H. Lv. & Vaniot) Rehder Wen 9003 (US) China, Shaanxi JQ182562 JQ182463 JQ182618 JQ182519 Ampelopsis cantoniensis K. Koch Wen 10810 (US) Vietnam JQ182545 JQ182437 JQ182592 JQ182498 Ampelopsis chaffanjonii (H. Lev.) Rehder Wen 9382 (US) China, Guangxi JQ182570 JQ182475 JQ182630 JQ182528 Ampelopsis cordat a Michx. Wen 9700 (US) USA, Texas JQ182558 JQ182459 JQ182614 JQ182513 Ampelopsis delavayana Planch. ex Franch. Wen 9377 (US) China, Guangxi JF437287 JQ182476 JQ182631 JF437065 Ampelopsis denudata Planch. Wen 8695 (US) Mexico JQ182577 JQ182483 JQ182638 JQ182534 Ampelopsis glandulosa var. hancei (Planch.) Momiy. Wen 8289 (US) Philippines JQ182571 JQ182477 JQ182632 JQ182529 Ampelopsis glandulosa var. heterophylla (Thunb.) Mom. Wen 9380 (US) China, Guangxi JF437289 JQ182474 JQ182629 JF437067 Ampelopsi s glandulosa var. kulingensis (Rehd.) Mom. Wen 9361 (US) China, Hunan JQ182572 JQ182478 JQ182633 JQ182530 Ampelopsis grossedentata (Hand. Mazz.) W. T. Wang Wen 9336 (US) China, Hunan JQ182573 JQ182479 JQ182634 JQ182531


Ampelopsis humulifolia Bge. Wen 8 519 (US) China, Beijing JQ182563 JQ182464 JQ182619 JQ182520 Ampelopsis hypoglauca (Hance) C.L. Li Wen 8195 (US) China, Chongqing AB235000 JQ182431 JQ182584 JQ182490 Ampelopsis japonica (Thunb.) Makino Wen s.n. (US) Cult. i n Washington D.C. JQ182538 J Q182585 JQ182491 Ampelopsis megalophylla Diels & Gilg Wen 9035 (US) China, Shaanxi JQ182564 JQ182465 JQ182620 JQ182521 Ampelopsis orientalis Planch. Al Shehbaz 0687 (US) Turkey JQ182565 JQ182466 JQ182621 JQ182522 Ampelopsis rubifolia (Wall.) Planch. Wen 9285 (US) China, Hunan JF437293 JQ182473 JQ182628 JF437072 Cayratia cordifolia C.Y. Wu Wen 10548 (US) China, Yunnan HM585934 HM585518 HM585794 HM585661 Cayratia geniculata (Blume) Gagnep. Wen 10680 (US) Indonesia, West Java JQ182575 JQ182481 JQ182 636 JQ182532 Cayratia japonica (Thunb.) Gagnep Shui et al. 81847 (KUN, US) China, SE Yunnan JQ182578 JQ182484 JQ182639 JQ182535 Cayratia maritima Jackes Wen 10701 (US) Indonesia, Papua JQ182576 JQ182482 JQ182637 JQ182533 Cayratia mollissima (Wall.) Gagn ep. Wen 8403 (US) Malaysia, Pahang HM585938 HM585522 HM585798 HM585665 Cayratia sp. Wen 10301 (US) Indonesia, SE Sulawesi JQ182547 JQ182439 JQ182594 JQ182500 Cayratia wrayi (King) Gagnep. Wen 10913 (US) Vietnam JQ182544 JQ182436 JQ182591 JQ182497 Ci ssus adnata Roxb. Wen 10519 (US) China, Yunnan JQ182579 JQ182485 JQ182640 JQ182536 Cissus albiporcata Masinde & L. E. Newton Luke & Luke 11456 (US) Kenya, Chyulu Plains JF437304 JQ182442 JQ182597 JF437087 Cissus aralioides Planch. Aplin s.n. (US) Cult i n Belgium National Bot Garden JQ182554 JQ182455 JQ182610 JQ182508 Cissus erosa Rich. Wen 8586 (US) Peru, Pasco HM585942 HM585526 HM585802 HM585668


Cissus granulosa Ruiz & Pav. Wen 8611 (US) Peru, Pasco JQ182543 JQ182435 JQ182590 JQ182496 Cissus int egrifolia (Baker) Planch. Luke & Luke 11475 (US) Kenya, Shimba Hills JQ182551 JQ182445 JQ182600 JQ182505 Cissus javana DC. Shui et al. 81970 (KUN, US) China, SE Yunnan JQ182580 JQ182486 JQ182641 JQ182537 Cissus nodosa Blume Wen 10713 (US) Indonesia, P apua HM585945 HM585529 HM585805 HM585671 Cissus phymatocarpa Masinde & L.E. Newton Luke & Luke 11474 (US) Kenya, Diani Forest JF437311 JQ182452 JQ182607 JF437095 Cissus pileata Desc. Wen 9662 (US) Madagascar JQ182557 JQ182458 JQ182613 JQ182512 Cissus producta Afzel. Luke & Luke 11528 (US) Tanzania, Udzungwa Mountain JF437312 JQ182447 JQ182602 JF437096 Cissus repens Lam. Shui et al. 81807 (KUN, US) China, Yunnan HM585946 HM585530 HM585806 HM585672 Cissus simsiana Schult. & Schult. f. Nee & Wen 5380 5 (US) Bolivia, Santa Cruz JQ182539 JQ182586 JQ182492 Cissus striata ssp. argentina (Suess.) Lombardi Nee & Wen 53854 (US) Bolivia, Santa Cruz JQ182540 JQ182432 JQ182587 JQ182493 Cissus striata ssp. striata Ruiz & Pav. Wen 7355 (US) Chile, Concepcion AB235018 JQ182428 JQ182582 JF437104 Cissus trianae Planch. Nee & Wen 53942 (US) Bolivia, Santa Cruz JQ182541 JQ182433 JQ182588 JQ182494


Cissus verticillata (L .) Nicolson & C.E. Jarvis Wen 10165 (US) Indonesia JQ182542 JQ182434 JQ182589 JQ182495 Clema ticissus angustissima (F. Muell.) Planch. Rossetto et al., 2002 Australia, Western Australia JQ182574 JQ182480 JQ182635 Clematicissus opaca (F. Muell.) Jackes & Rossetto Rossetto et al., 2002 Australia, Queensland JQ182548 JQ182440 JQ182595 JQ182501 Cyphostemma kilimandscharicum (Gilg) Wild & R.B. Drumm. Luke & Luke 11469 (US) Kenya, Chyulu Hills JF437327 JQ182451 JQ182606 JF437112 Cyphostemma maranguense (Gilg) Desc. Luke & Luke 11468 (US) Kenya, Chyulu Hills JF437329 JQ182449 JQ182604 JF437114 C yphostemma serpens (Hochst. ex A. Rich.) Desc. Luke & Luke 11447 (US) Kenya, Kiboko JQ182552 JQ182446 JQ182601 JQ182506 Cyphostemma thomasii ( Gilg & Brandt) Descoings Luke & Luke 11448 (US) Kenya, Makindu JF437331 JQ182450 JQ182605 JF437117 Leea goniop tera Lauterb. Wen 10711 (US) Indonesia, Papua JQ182559 JQ182460 JQ182615 Leea guineensis G. Don Wen 80_84 (US) Cult. i n Hawaii, originally from Africa JQ182549 JQ182441 JQ182596 JQ182503 Leea indica (Burm. f.) Merr. Wen 10910 (US) Vietnam HM585953 HM585537 HM585813 HM585679 Parthenocissus chinensis C.L. Li Nie & Meng 470 (KUN, US) China, Sichuan HM223270 HM223381 HM223327 JQ182502


Parthenocissus feddei (Lvl.) C. L. Li Zhang 319 (US) China, Hunnan HM223307 HM223416 HM223359 JQ182526 Parthenocis sus quinquefolia (L.) Planch. Wen 9711 (US) USA, Texas HM223258 HM223368 HM223315 JQ182514 Parthenocissus semicordata (Wall.) Planch. Wen et al. (Tibet MacArthur) 887 (KUN, US) China, Xizang HM223271 HM223382 HM223328 JQ182511 Parthenocissus suberosa H and. Mazz. Nie & Meng 358 (KUN, US) China, Guizhou HM223273 HM223384 HM223330 JF437134 Parthenocissus vitacea (Knerr) Hitchc. Wen 10488 (US) Canada, Quebc HM223295 HM223407 HM223352 HM585681 Rhoicissus digitata Gilg & Brandt Gerrath s.n. (US) cult. i n Iowa USA AB235047 JQ182429 AB234966 JQ182489 Rhoicissus revoilii Planch. Luke & Luke 11698 (US) Kenya, Magadi JQ182555 JQ182456 JQ182611 JQ182509 Rhoicissus rhomboidea Planch. Wen 6673 (US) Cult. in Missouri Bot. Gard., USA AB235048 AB234931 AB2349 67 JQ182488 Rhoicissus tomentosa (Lam.) Wild & R.B. Drumm. Aplin 19656252 (US) Cult. i n Belgium National Bot Garden JF437342 JQ182454 JQ182609 JF437139 Rhoicissus tomentosa (Lam.) Wild & R.B. Drumm. Wen 10076 (US) South Africa, KwaZulu Natal HM223252 HM223362 HM223309 JQ182516 Rhoicissus tridentata (L. f.) Wild & R.B. Drumm. Luke & Luke 11453 (US) Kenya, Chyulu Hills JF437341 JQ182443 JQ182598 JF437138 Tetrastigma caudatum Merr. & Chun Wen 10812 (US) Vietnam, Vinh Phuc HM585967 HM585551 HM585827 HM 585691


Tetrastigma ceratopetalum C.Y. Wu Shui et al. 81836 (KUN, US) China, SE Yunnan HM585937 HM585521 HM585797 HM585664 Tetrastigma erubescens Planch. Wen 10968 (US) Vietnam, Guangnam HM585987 HM585571 HM585847 HM585711 Tetrastigma henryi Gagnep. W en 10518 (US) China, Yunnan HM586002 HM585586 HM585862 HM585724 Tetrastigma obovatum (M.A. Lawson) Gagnep. Wen 10567 (US) China, Yunnan HM586024 HM585608 HM585883 HM585746 Tetrastigma serrulatum (Roxb.) Planch. Wen 10532 (US) China, Yunnan HM586003 HM 585587 HM585863 HM585725 Tetrastigma sichouense C.L. Li Wen 10547 (US) China, Yunnan HM586046 HM585631 HM585905 HM585768 Vitis aestivalis Michx. Wen 10428 (US) USA, Delaware HM223286 HM223398 HM223343 JQ182515 Vitis davidiana (Carrire) G. Nichol son Wen 9060 (US) China, Shaanxi JQ182568 JQ182470 JQ182625 JQ182525 Vitis flexuosa Thunb. Wen 8540 (US) Japan, Chiba Ken JQ182567 JQ182469 JQ182624 JQ182524 Vitis mengziensis C.L. Li Nie & Meng 415 (KUN, US) China, Yunnan HM223276 HM223387 HM22333 3 JF437158 Vitis pope noei Fennell Wen 8724 (US) Mexico HM586072 HM585657 HM585930 HM585790 Vitis riparia Michx. Wen 8658 (US) USA, Virginia JF437357 JQ182453 JQ182608 JF437165 Vitis rotundifolia Michx. Wen 11087 (US) USA, Arkansas JF437358 JQ1 82468 JQ182623 JF437166 Vitis vulpina L. Wen 11082 (US) USA, Arkansas JQ182566 JQ182467 JQ182622 JQ182523 Yua thomsonii (M.A. Lawson) C.L. Li Nie & Meng 469 (KUN, US) China, Sichuan HM223277 HM223389 HM223335 JF437171