| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Systematics |
Forest and Evolutionary Genomics Laboratory, and the Centre for Structural and Functional Genomics, Biology Department, Concordia University, 7141 Sherbrooke St. West, Montreal, Quebec, Canada H4B 1R6
Received for publication October 14, 2003. Accepted for publication April 29, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: cpDNA • phylogeny • Populus • rDNA • reticulate evolution • Salicaceae
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Because of their fast growth rates, profuse vegetative propagation, adaptability to a variety of ecological sites, and the numerous uses of their wood (i.e., timber, paper pulp), Populus species have become one of the most economically important groups of forest trees (Stettler et al., 1996b
). They have also become a subject of many interdisciplinary studies, including biotechnology and genetic engineering, and are becoming a model organism for the study of tree biology (Stettler et al., 1996a
).
The extensive interspecific hybridization and the high levels of morphological variation among poplars have posed great difficulties in species delimitation for systematic and comparative evolutionary studies. The number of Populus species currently described in the literature ranges from 22 to 85 plus hundreds of hybrids, varieties, and cultivars (Eckenwalder, 1977b
, 1996
; Dickmann and Stuart, 1983
). Discrepancies in the number of species could be attributed to the misinterpretation of some hybrids and to difficulties involved in delineating species boundaries. According to a recent classification (Eckenwalder, 1996
), the genus Populus is classified into 29 species in six sections (Abaso, Aigeiros, Leucoides, Populus, Tacamahaca, Turanga). These sections are considered to be natural in most cases because they are delineated by the occurrence of major hybridization barriers (Zsuffa, 1975
; Eckenwalder, 1996
). However, the placement of several taxa within these sections remains controversial. For instance, P. nigra of section Aigeiros has a genetic affinity to species of section Tacamahaca. In P. nigra, cpDNA RFLP analysis showed similarity to species of the section Populus, but RFLP patterns of nuclear rDNA were distinct from the section Populus, suggesting a possible hybrid origin of P. nigra (Smith and Sytsma, 1990
).
Recent phylogenetic analyses of the family Salicaceae using DNA sequence data from chloroplast rbcL (Azuma et al., 2000
) and ITS of nuclear rDNA (Leskinen and Alstrom-Rapaport, 1999
) strongly suggest that Populus is a monophyletic group sister to Salix. However, phylogenetic relationships within the genus remain controversial. A phylogenetic analysis of poplars using 76 morphological traits of buds, leaves, inflorescences, flowers, and fruits supported the monophyly of all sections except Tacamahaca, which resolved into two paraphyletic groups (Eckenwalder, 1996
). However, relationships between sections as well as relationships among taxa within sections were only partially resolved.
Based on phylogenetic analysis of restriction fragment length polymorphism of chloroplast DNA, the section Tacamahaca has been suggested to be polyphyletic and the section Populus is considered as the terminal clade (Smith, 1988
). On the other hand, a phylogenetic tree based on DNA sequences from the ITS region of the nuclear rDNA of four species of Populus showed an opposite trend with P. alba of section Populus as basal, followed by P. lasiocarpa of the section Leucoides and species of the sections Aigeiros and Tacamahaca as the terminal clade (Leskinen and Alstrom-Rapaport, 1999
). Mitochondrial and chloroplast restriction site analysis of four Populus species suggests polyphyletic relationships for species in the section Aigeiros (Barrett et al., 1993
; Rajora and Dancik, 1995
).
In phylogenetic studies of groups in which hybridization between lineages has played a substantial role in their evolution (Smith, 1988
), the combined use of cpDNA and nuclear DNA data is crucial to gain a comprehensive understanding of the evolutionary history. A striking feature of poplars that has received considerable attention from many researchers is the occurrence of interspecific hybrids (Eckenwalder, 1982
, 1996
; Dickmann and Stuart, 1983
; Barnes and Pregitzer, 1985
; Whitham et al., 1996
). Hybrids are regularly found in regions where species of sections Aigeiros and Tacamahaca are sympatric, such as in the contact zones of P. angustifolia, P. trichocarpa, and P. balsamifera (Brayshaw, 1965
). Similarly, species of section Populus are known to hybridize naturally with other members of the section, such as P. alba with P. tremula (Stettler et al., 1996b
) and P. grandidentata with P. tremuloides (Barnes, 1961
). Moreover, RFLP analysis of cpDNA and rDNA (Smith, 1988
) has raised the possibility of ancient hybridization by which P. nigra appeared to be an introgressant of the P. alba (cpDNA) lineage and some other presently unknown paternal lineage of section Populus. Similarly, P. tristis (a central Asian member of section Tacamahaca) appeared to be an introgressant or a hybrid of the P. nigra (cpDNA) lineage and the lineage with Asian species of section Tacamahaca. Thus, it is not unreasonable to assume that there have been abundant opportunities for gene exchange among sympatric species, even between taxa of different sections (Stettler et al., 1996b
).
Many Populus species are likely to have undergone complex reticulate evolution. In a hybrid lineage resulting from reticulate evolution, parental nuclear genes may become fixed for one of the parental types through homogenization by concerted evolution of gene families (Page and Holmes, 1998
), segregation during sexual reproduction, or lineage sorting (Doyle, 1992
). These events will generate discrepancies among phylogenetic trees based on uniparentally inherited organeller sequences (Sears, 1980
; Neale and Wheeler, 1986
) and biparentally inherited nuclear sequences or their phenotypic expression in morphology. Therefore, data from both nuclear and chloroplast genes are required to parse out reticulate evolutionary events (Page, 2000
) and reconstruct robust phylogenetic trees.
The objective of our present study is to reconstruct the phylogeny of genus Populus based on DNA sequences of chloroplast and nuclear genomes to improve understanding of the evolutionary history of the genus and to provide a framework for biosystematic classification of the genus Populus.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
, b
, 1996
) and Dickmann and Stuart (1983)
. Fresh leaves were obtained from 21 taxa including 17 species representing three sections of Populus [sections Tacamahaca Spach, Aigeiros Duby, and Populus (Leuce) Duby]. This sampling includes two varieties of P. deltoides and two putative species not included in Eckenwalder's (1977a
, b)
and Dickmann and Stuart's (1983)
taxonomic treatments. DNA sequences from two species of Salix were used as outgroups. The list of species used in this study and the accession numbers are given in Appendix 1 (see Supplemental Data accompanying the online version of this article). The collected leaf samples were stored at –80°C prior to extraction of DNA. Total genomic DNA was extracted from frozen leaf tissue using the methods of Doyle and Doyle (1987)
and Dayanandan et al. (1997)
.
The three noncoding regions of cpDNA: trnT-trnL intergenic spacer (IGS1), trnL intron, and trnL-trnF intergenic spacer (IGS2) (hereafter referred to as cpDNA) and ITS1, ITS2, and part of the 5.8S and the 5' region of the 28S subunit of the rDNA (hereafter referred as rDNA) were amplified by the polymerase chain reaction (PCR). The oligonucleotide primers a, c, d, and f designed by Taberlet et al. (1991)
for cpDNA, primers a and b modified from Leskinan and Alstrom-Rapaport (1999)
for ITS1, and primers ITS3 (Becerra and Venable, 1999
) and ITS 28kj modified from Culling (1992)
for ITS2 were used for PCR amplification and sequencing. Two additional primers, trnCR and trnDF were designed for sequencing to cover some gaps in the cpDNA sequences (Fig. 1a, b; Tables 1, 2). Amplification reactions contained 230 µmol/L dNTP, 2.5 mmol/L MgCl2, 5 µmol/L of each primer, 1 unit of Taq DNA polymerase, and 2.5 µL buffer (0.2 mol/L Tris pH 9.5; 0.25 mol/L KCl; 1 mg/mL BSA, 5 µL/mL Tween 20) in a total volume of 25 µL. The PCR amplification was performed in an Mastercycler gradient thermal cycler (Eppendorf, Westbury, New York, USA) at 94°C for 60 s, 55°C for 30 s, and 72°C for 60 s for 35 cycles. Amplified DNA was purified using a PCR purification kit (QIAGEN, Mississauga, Ontario, Canada) , electrophoresed on 1% agarose gel with ethidium bromide (0.33 µg/mL) at 3.5 V/cm for 90 min. Amplified fragments were visualized and documented using a GeneSnap 4.00-Gene Genius Bio Imaging System (Syngene, Frederick, Maryland, USA). The digital image files were analyzed using Gene Tools software from Syngene. The quantity of DNA was estimated using a Mass Ruler DNA Ladder Mix (Fermentas, Burlington, Ontario, Canada). The purified, amplified DNA was directly sequenced using ABI Big Dye Terminator version 3.0 and 3.1 Cycle Sequencing Ready Reaction kit and an ABI310 automated genetic analyzer (Applied Biosystems, Foster City, California, USA). Each region was sequenced between two and seven times. Same primers were used for both PCR amplification and sequencing. The thermal cycling profile of sequencing reactions were: 96°C for 10 s, 50°C for 5 s, and 60°C for 4 min for 25 cycles. The chromatograms of the DNA sequencing results were processed and analyzed using the Staden software package (Staden, 1979
; http://staden.sourceforge.net). The assembled contigs of cpDNA and rDNA of each species were aligned using ClustalW (Thompson et al., 1994
) multiple sequence alignment software. Aligned DNA sequences were imported to MacClade 4.0 software (Maddison and Maddison, 2001
) for verification and manual editing of the sequence alignments. The 5'/3' trnL exon (identified by comparing with the Nicotiana chloroplast sequence; Genebank accession number = NC001879). The final data set included three noncoding regions of the trnT-trnF of cpDNA (intergenic region of trnT-trnL, trnL intron, and intergenic region of TrnL-trnF) and ITS I, partial 5.8S rRNA, ITS 2, and part of 28S subunit of the rDNA. The two data sets (cpDNA and rDNA) were analyzed separately.
|
|
|
). For each data set, 106 random trees were analyzed to obtain the frequency distribution of tree lengths to assess the phylogenetic signal of the data matrix by calculating the skewness (g1) and kurtosis (g2). The nucleotide frequency distribution was investigated with a chi-square test of homogeneity of base distribution across sequences.
Parsimony analysis
Maximum parsimonious (MP) phylogenetic trees were reconstructed through heuristic search with equal character weights, gaps treated as missing, multistate taxa interpreted as uncertainty, starting tree obtained via stepwise addition, and sequence addition was as-is for the cpDNA data set (random addition was not possible because of computing limitations) and random addition of sequences with 1000 replicates for rDNA data set. Tree bisection-reconstruction (TBR) was used as the branch-swapping algorithm. Strict and 50% majority rule consensus trees were obtained. The phylogenetic trees were rooted using Salix species as outgroups. Bootstrap analysis with a fast heuristic search based on 1000 replicates was performed to assess the robustness of branches.
Maximum likelihood analysis
The software program Modeltest version 3.06 (Posada and Crandall, 1998
) was used to find the best nucleotide substitution model that fits the data set using hierarchical likelihood ratio test. The maximum likelihood (ML) analyses of both cpDNA and rDNA were performed through heuristic search with TBR branch swapping, addition of sequences as-is, and the Tamura-Nei + I model (Tamura and Nei, 1993
) as the nucleotide substitution model. The bootstrap analysis with fast heuristic search based on 100 replicates was performed to assess the robustness of branches.
Comparison of cpDNA with rDNA tree
The two consensus trees obtained through maximum parsimony analysis of two data sets were compared using nonparametric Templeton (Wilcoxon signed-ranks) and winning-sites tests (Templeton, 1993
). To compare two maximum likelihood trees, we used the Kishino-Hasegawa (1989)
test using bootstrap with full optimization, two-tailed test, and the Shimodaira-Hasegawa (1999)
test using bootstrap with full optimization, one-tailed test. Each bootstrap analysis was performed with 1000 replicates.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The frequency distribution of the length of 106 random trees yielded g1 = –1.726, g2 = 4.396 for cpDNA and g1 = –1.313, g2 = 2.718 for the rDNA suggesting a strong phylogenetic signal in these data matrices. The chi-square test of homogeneity of base frequencies across taxa showed P = 1.0 in both cases, and the null hypothesis of homogeneous base distribution across sequences was accepted.
The length of ITS 1 in all Populus species included in our study was 224 base pairs (bp) except for P. fremontii, which was 225 bp. The length of ITS 2 was 212 bp for all Populus species examined, except for P. davidiana and P. tremula (214 bp) and P. maximowiczii and P. simonii (203 bp), which shared a deletion of 11 bp. The cpDNA data matrix comprised IGS1 (473 bp), trnT-trnL intron (628 bp), and IGS 2 (312 bp).
cpDNA trees
Maximum parsimony analysis
Maximum parsimony searches of chloroplast DNA yielded 30 939 equally parsimonious trees (tree length 118; consistency index [CI] = 0.924; retention index [RI] = 0.927; rescaled consistency index [RC] = 0.857). The 50% majority rule consensus tree (Fig. 2) and strict consensus tree differed only in the placement of P. tremula and P. tremuloides (with 53% occurrence). They appeared as sister taxa in a basal position to the other species of section Populus in the 50% majority rule tree, while in the strict consensus tree they grouped with other members of the section with an unresolved polytomy. The bootstrap value also did not support a basal position for these two aspens.
|
Besides P. nigra, which did not group with the other Aigeiros group members, the cpDNA data of P. tristis and P. szechuanica (Asiatic balsam poplars) suggested that these were more closely related to North American cottonwoods of the Aigeiros group than to species of section Tacamahaca. The MP analysis placed the lineage consisting of North American cottonwoods (P. fremontii, P. deltoides, and its two varieties) and P. tristis of Asian balsam poplars (section Tacamahaca) in a terminal clade. However, based on the MP tree, the basal lineage could not be determined. Section Populus was monophyletic and sections Tacamahaca and Aigeiros were polyphyletic groups (Fig. 2).
Maximum likelihood analysis
The results of the Modeltest analysis showed that TrN + I (Tamura and Nei, 1993
: equal rate for all transversions and different transition rates with unequal base frequencies) nucleotide substitution model was the most suitable model for the cpDNA. The parameters of the model were: base frequencies: A = 0.4175, C = 0.1336, G = 0.1467, T = 0.3023; rate matrix: (A–C) = (A–T) = (C– G) = (G–T) = 1.0; (A–G) = 1.7645; (C–T) = 2.5848; among-site rate variation: proportion of invariable sites, I, = 0.6953; equal rates of substitution for all variable sites.
The maximum likelihood analysis of cpDNA with the described model parameters retained a single tree (Fig. 3) with a topology identical to the 50% majority rule consensus tree obtained from parsimony analysis. Moreover, the bootstrap values for branch robustness under the maximum likelihood criterion were similar to the values obtained from the same test under the parsimony criterion.
|
rDNA trees
Maximum parsimony analysis
The maximum parsimony analysis based on nuclear rDNA yielded 497 equally parsimonious trees (tree length = 94; CI = 0.851; RI = 0.888; RC = 0.756). In the 50% majority rule consensus tree (Fig. 4), two North American aspens, P. tremuloides and P. grandidentata, grouped as sister taxa in the lineage consisting of other Populus species, but their placement in the strict consensus tree remained unresolved. In the 50% majority rule consensus tree, a group of balsam poplars of section Tacamahaca, namely P. angustifolia, P. cathayana, P. trichocarpa, P. balsamifera, P. tristis, and P. szechuanica clustered as a sister group to the lineage comprising members of section Aigeiros. Although the branch representing Tacamahaca occurred in 75% and Aigeiros 100% of 497 most parsimonious trees, these branches were not supported by bootstrap analysis. Moreover, in the 50% majority rule consensus tree, within the balsam poplar lineage, P. trichocarpa, P. balsamifera, P. tristis, and P. szechuanica clustered together as a sister group to P. angustifolia and P. cathayana. However, none of these internal nodes and relationships was supported by the bootstrap analysis, even though they occurred with a high percentage in the 497 retained equally parsimonious trees (Fig. 4). Moreover, a sister relationship of P. maximowiczii and P. laurifolia to the clade comprising two lineages of the Aigeiros and group of Tacamahaca species (as mentioned earlier) was weakly supported. The basal position of P. simonii was not supported by the bootstrap analysis.
|
The relationships among species within section Tacamahaca were unresolved, and they grouped as a polytomy. However, this section could be divided into two distinct groups of taxa: P. maximowiczii, P. simonii, and P. laurifolia in one group and the other members of the section in another group. The relationships among species within section Aigeiros were resolved, and the MP analysis of rDNA suggested a monophyletic origin for this section. However, this was not supported by bootstrap analysis (<50%). A close relationship among P. nigra, P. deltoides var. angulata, and P. roegneriana was evident. In contrast to the cpDNA-based MP tree, the rDNA-based tree did not have a close affinity between P. nigra and members of section Populus. Populus tristis and P. szechuanica clustered as an unresolved polytomy with the remaining species of sections Tacamahaca and Aigeiros.
Maximum likelihood analysis
The result of the Modeltest analysis showed that TrN + I (Tamura and Nei, 1993
: equal transversion rate with variable transition substitution rate and unequal base frequencies) was the most suitable model for the rDNA. The parameters of the model were: base frequencies: A = 0.1841, C = 0.3205, G = 0.3242, T = 0.1711; rate matrix: (A–C) = (A–T) = (C–G) = (G–T) = 1.0, (A–G) = 3.5870, (C–T) = 5.6386; and among-site rate variation: proportion of invariable sites, I, = 0.8085, equal rates of substitution for all variable sites.
The ML analysis of rDNA sequences with the described model parameters yielded three trees. The 50% majority rule consensus tree (Fig. 5) and the strict consensus tree differed only at the placement of P. deltoids var. angulata. In the 50% majority rule consensus tree, it occupied the basal position to the clade, comprising P. nigra and P. roegneriana, whereas in strict consensus they all were sister to one another.
|
Topology comparison test results
The Templeton (Wilcoxon signed-ranks) test statistic Z = –5.9244 was smaller than the absolute values of the rank sums. The Winning-site test with P < 0.0001 indicates a significant difference at P < 0.05 under the null hypothesis of no difference between two trees. Both the Kishino-Hasegawa and Shimodaira-Hasegawa tests rejected the null hypothesis (P = 0.000), suggesting that there was no difference between the two trees.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Compared to the willows (Salix spp.), there are relatively few species of poplars, and they fall into a number of groups that are traditionally recognized as sections. With few exceptions, there is a reasonable agreement in the literature on the characteristics and species composition of these sections, and major barriers to hybridization are known to lie between the sections (Zsuffa, 1975
). However, the relationships between sections and among species within each section and the placement of some controversial species or taxa of purported hybrid origin have been subjects of controversy for a long time.
The species of section Populus are monophyletic with respect to their maternal lineage (cpDNA), with the exception of P. nigra, a taxon considered to be a member of section Aigeiros, which clusters with species in section Populus. In terms of rDNA (nuclear) lineage section Populus is monophyletic, and P. nigra clusters with the other species of section Aigeiros. The phylogenetic relationships among the species within section Populus based on cpDNA and rDNA sequence data are not in agreement with one another. Based on maternal chloroplast sequence data, European aspen (P. tremula) and the American trembling aspen (P. tremuloides) cluster together as sister taxa and occupy the terminal position in the Populus clade. However, this relationship is weak and has less than 50% bootstrap support. In the rDNA-based phylogenetic tree, white poplar (P. alba) and Korean aspen (P. davidiana) cluster together as sister species and occupy the terminal position. These two species along with European aspen (P. tremula) form a monophyletic group within section Populus with sister relationships to the North American aspens, P. tremuloides and P. grandidentata.
The relationships between sections Aigeiros and Tacamahaca have been controversial. The species of these two sections are known to be freely interfertile (Zsuffa, 1975
; Eckenwalder, 1984a
). Both cpDNA and rDNA sequence data suggest a polyphyletic origin of section Tacamahaca. The polyphyly of section Tacamahaca was also suggested by morphology-based phylogenenetic analysis (Eckenwalder, 1996
). The rDNA-based phylogenetic tree suggests a monophyletic origin of section Aigeiros. Similarly, the cpDNA-based phylogeny suggests a monophyletic origin, with the exception of P. nigra, which shares its chloroplast ancestor with the section Populus and P. tristis from section Tacamahaca, which clustered with North American cottonwoods of section Aigeiros. In other words, North American cottonwoods of section Aigeiros are monophyletic with respect to their chloroplast lineage with the exception of P. tristis of section Tacamahaca, which clusters within this group.
The phylogenetic relationships among species within sections, based on both cpDNA and rDNA, are not well resolved. However, a few congruencies between the phylogenetic trees based on cpDNA and rDNA (with a low confidence level in the case of rDNA) can be recognized. The three Asian balsam poplars, P. simonii, P. maxomowiczii, and P. laurifolia, form a clade distinct from other members of the section Tacamahaca. Section Aigeiros has a closer affinity to the lineages of balsam poplars that includes P. tristis and P. szechuanica (Figs. 2, 4). Interspecific hybridization within Populus is generally limited to intrasectional crosses or intersectional crosses between species of Aigeiros and Tacamahaca (Ronald, 1982
). Intersectional crosses between Tacamahaca and Aigeiros are generally compatible in most combinations (Zsuffa, 1975
) and have given rise to many vigorous clones used in plantations (Dickmann and Stuart, 1983
).
The polyphyletic origin of section Tacamahaca and Aigeiros and the unresolved relationships among species within and between sections suggest relatively close evolutionary relationships among the species. This finding is consistent with observations of spontaneous hybridization among these species in nature and in cultivation. For instance, spontaneous hybridization between European P. nigra and introduced North American P. deltoides resulted in a dominant and widespread hybrid P. x euramericana (Lefevre et al., 2001
).
Inter- and intra-sectional hybridization is common in regions of sympatry (Eckenwalder, 1996
). Only section Populus is strongly reproductively isolated from the others and has strong intersectional incompatibility barriers. However, the cellular and molecular bases of this have not yet been characterized (Villar et al., 1986
). Nevertheless, this significant reproductive isolation reflects the evolutionary divergence of species of section Populus from those in other sections including the species of two closest sections, Aigeiros and Tacamahaca (Eckenwalder, 1996
; Figs. 2, 4). This is consistent with the manipulation required to succeed in the artificial crossing of aspens with other poplars (Guries and Stettler, 1976
; Ronald, 1982
).
It is generally assumed in the taxonomic literature that the relative interfertility of poplars (especially where these crosses are spontaneous) is a reflection of their genetic similarity (Stettler et al., 1996b
; Dickmann et al., 2001
) and therefore can form a basis on which to infer taxonomic affinities. According to Eckenwalder (1977b)
, P. trichocarpa and P. balsamifera of section Tacamahaca have clear signs of introgression (Viereck and Foote, 1970
) in their regions of overlap (e.g., the Rocky Mountains and Alaska), probably because of the affinity of their genomes, which permits natural hybridization and repeated back-crossing with parental species. Sections Tacamahaca and Aigeiros are broadly sympatric in North America (Little, 1971
), with overlapping ecological preferences (Fowells, 1965
). The North American representatives of sections Aigeiros and Tacamahaca have such strong similarities in floral traits that they can hardly be distinguished on the basis of floral morphology (Eckenwalder, 1977b
, 1984b
). In fact, based on an extensive study of New World poplars, Eckenwalder (1977b)
predicts the eventual merger of these two sections. The sister relationship of section Aigeiros (with North American cottonwoods in the basal position) with the rDNA-based clade containing three North American balsam poplars of section Tacamahaca (P. balsamifera, P. trichocarpa, and P. angustifolia; Fig. 4) and the close affinities of P. deltoides and P. fremontii with North American balsam poplars in the cpDNA-based phylogenetic tree (Fig. 2) suggest a close evolutionary relatedness between these species and support the merger proposed by Eckenwalder (1977b)
.
This trend of extensive hybridization in Populus is a major cause of disagreements on the total number of poplar species and their classification. In fact, some poplars such as P. Balm-of-Gilead (a variety of P. x jackii) and P. x tomentosa, which were originally described as species and cultivated for centuries, have later been identified as natural or near-natural hybrids (Eckenwalder, 1996
). In addition to the long-term commercial benefit of hybridization in Populus including exploitation for increased variability and novel gene combinations, hybridization may also have played a significant role in the evolution of sections and their rapid allopatric speciation (Eckenwalder, 1996
).
Discrepancies between phylogenetic trees based on chloroplast and nuclear nucleotide sequence data
The major discrepancies between the nuclear rDNA and cpDNA phylogenetic trees involve the placement of P. nigra, P. tristis, and P. szechuanica and may suggest an ancient hybridization event between ancestors of paternal and maternal lineages of these extant species.
Similar to the results of chloroplast RFLP analysis (Smith and Sytsma, 1990
), the cpDNA sequence data indicate that P. nigra has a chloroplast genome derived from the clade of section Populus and divergent from the American cottonwoods of section Aigeiros. However, based on the number of nucleotide substitutions mapped on the maximum parsimonious tree (Fig. 2), P. nigra, P. tremula, and P. davidiana, with a single substitution from the common ancestral node, are more closely related to each other than to P. alba, which has four base substitutions. This indicates that the chloroplasts of P. nigra may have originated from ancestors of P. tremula or P. davidiana rather than from P. alba, as suggested by Smith and Sytsma (1990)
. Alternatively, we can conclude that either an extinct ancestor of section Populus or any extant species within this section but not included in this study may have played the maternal role in this hybridization event. Moreover, similarity between data from rDNA sequences and rDNA restriction site variation (Smith and Sytsma, 1990
) indicate that the nuclear genome of black poplar (P. nigra) is distinct from species in section Populus and, contrary to Smith and Sytsma (1990)
, very closely related to P. deltoides var. angulata from section Aigeiros (Fig. 4). It also shows a sister relationship with P. fremontii. Populus deltoides and P. sargentii (P. deltoides var. occidentalis) occupied a basal position. However, these relationships were not strongly supported by bootstrap analysis (<50%). Thus, the extant P. nigra may have derived from an ancient hybridization event involving an ancestor or extant species of section Populus as the maternal (cpDNA) donor and the ancestor of the cottonwoods (probably the immediate ancestor of P. deltoides) of section Aigeiros as the paternal (rDNA) donor. The possible geographic location of this ancient hybridization event and its likely geographic isolation from the paternal species are discussed later.
The evidence from classical morphological, chemical, crossing, and pathogenic studies do not consistently favor the placement of P. nigra in either section Aigeiros or section Populus (Smith, 1988
), both of which have been proposed as the origin of P. nigra. In the cladistic analysis of 76 morphological characters of buds, leaves, inflorescences, and male and female flowers and fruits performed by Eckenwalder (1996)
, P. nigra had a sister relationship with the clade comprising P. fremontii and P. deltoides from section Aigeiros. Thus, the placement of P. nigra within section Populus may necessitate invoking a considerable amount of morphological convergence of P. nigra to the cottonwoods. In addition, of all possible crosses involving P. nigra as one parent (Dickmann et al., 2001
), the most problematic crosses are among members of its own maternal lineage, section Populus. Successful crosses of section Populus with P. nigra apparently are only rarely achieved. Moreover, section Populus is also distinguished from P. nigra by its relative immunity to certain leaf rust organisms. Melampsora medusae (American) and M. populina (European) infect members of section Aigeiros including P. nigra and Tacamahaca, but they rarely attack the aspens and white poplars of section Populus (Smith, 1988
; Newcombe, 1996
; Lefevre et al., 2001
). Based on this evidence, the placement of P. nigra in either section Populus or section Aigeiros remains controversial.
Another major incongruence between the two data sets is the status of P. tristis, which has a chloroplast affinity to section Aigeiros (cottonwoods) and a nuclear genome related to Tacamahaca (balsam poplars; Figs. 2, 4). This suggests that P. tristis (Himalayan balsam poplar) may have derived from an ancient hybridization event with an ancestor of North American cottonwoods as the maternal (cpDNA) donor and probably the immediate ancestor of the lineage comprising species of section Tacamahaca (North American P. angustifolia, P. balsamifera, P. trichocarpa, or Chinese P. cathayana and P. szechuanica) as a paternal donor. The introgressant status of P. tristis was also suggested earlier by Smith (1988)
. Based on the results of chloroplast RFLP studies and a preliminary analysis of nuclear rDNA RFLP, he concluded that P. tristis is an introgressant or hybrid of the P. nigra (cpDNA) lineage and the Asian portion of section Tacamahaca. Again, however, the cpDNA sequence data in our study links P. tristis with the North American cottonwood (P. deltoides and P. fremontii) lineage. The nuclear sequence data suggest an affinity to species of the section Tacamahaca, especially to North American P. balsamifera and P. trichocarpa. The high morphological similarity of P. tristis to North American P. balsamifera and the recent development of the highly valued hybrid clone P. tristis x P. balsamifera (Dickmann and Stuart, 1983
) strengthen the proposed model for the evolution of the Himalayan poplar.
Populus szechuanica, a species native to China, is another source for discrepancies in the Populus taxonomy, and its placement in section Tacamahaca is controversial. The cpDNA RFLP study clustered this species with species of section Populus and P. nigra. Conversely, rDNA RFLP analysis showed a close affinity between balsam poplars (Section Tacamahaca) and P. szechuanica (Smith and Sytsma, 1990
). Our nuclear rDNA data confirmed the rDNA-based RFLP analysis results and clustered P. szechuanica in the monophyletic clade that consisted of certain Tacamahaca species, including P. tristis (Fig. 4). The cpDNA, however, had a different pattern. In the cpDNA-based most parsimonious tree, P. szechuanica occupied a basal position to the lineage comprising members of section Aigeiros and P. tristis (Fig. 2). However, the cladistic analysis of 76 morphological characters of buds, leaves, inflorescences, and male and female flowers and fruits put P. szechuanica in the same clade as other Tacamahaca species, which is paraphyletic to the Aigeiros lineage (Eckenwalder, 1996
). Therefore, the evolutionary history of P. szechuanica may be similar to that of P. tristis as discussed earlier.
Eckenwalder (1996)
proposed that natural hybridization, while common in genus Populus over at least the last several million years, has had little effect on speciation in this group, relative to factors promoting divergence. Eckenwalder's model of evolution for Populus (Eckenwalder, 1996
) is based on cladistic analysis of morphological characters. He proposed an evolutionary pattern involving phases of ecological radiation and geographical vicariance. Nevertheless, our data and phylogenetic trees suggest ancient intersectional hybridization or reticulate evolution for at least three of the taxa, namely P. nigra, P. tristis, and P. szechuanica. Hence, natural hybridization in poplars, particularly intersectional hybridization, merits further investigation at the molecular level for a better understanding of the evolutionary history of genus Populus.
It could be speculated that the putative hybridization and introgression events, mentioned earlier, must have predated the beginning of the Miocene about 23.5 million years ago (Parrish, 1987
), when the northern land mass, Laurasia, had broken apart, opening up the Atlantic Ocean and consequently separating Eurasia from North America. In the putative hybrids studied, at least one of the parental lineages (Aigeiros cottonwoods) is native to North America and the other putative parental species is native to Eurasia with no geographical sympatry. Because of the short viability periods of poplar pollen and seeds, as well as unfavorable chemical and physical conditions, cross-oceanic hybridization is unlikely (but cannot be ruled out). Thus, hybridization likely occurred while North America and Europe/Asia were contiguous within the Laurasian land mass. Following hybridization, introgression to one of the parental lineage or geographical isolation of the hybrid population through long-distance seed dispersal followed by local adaptation could have led to significant phenotypic divergence.
Finally, for the two taxa collected from the Montreal Botanical Garden, identified as P. songarica and P. roegneriana, we were unable to find any information about their history or the geographic source of the trees. The botanical garden had obtained them from the Sheridan Nursery in Montreal in 1973 and the Boyce Thompson Arboretum, USA, in 1937, respectively, and no further information was available. However, the cpDNA data of P. songarica had a close affinity to P. laurifolia from section Tacamahaca, and its rDNA data placed the species with P. maximowiczii and P. laurifolia. Thus, even though the relationships among these species are not resolved, they occupied the basal position to a clade comprising species of section Aigeiros and other members of Tacamahaca (with less than 50% bootstrap confidence value; Figs. 2, 4). Populus songarica, therefore, could be a variety or cultivar of P. laurifolia. Both cpDNA and rDNA sequence data showed close relationships between P. roegneriana and species in section Aigeiros.
The overall conclusions of this study are that, although Populus is a readily defined genus consisting of well-marked sections, at least two such sections, Tacamahaca and Aigeiros, are not monophyletic in origin. Species of section Aigeiros are monophyletic based on rDNA data, but polyphyletic with respect to their chloroplast lineage. Both cpDNA and rDNA sequences data suggest a polyphyletic origin for section Tacamahaca. The lineage comprising species of section Populus is distinct from the two other sections and the lineage of Asiatic balsam poplars (P. simonii, P. laurifolia, and P. maximowiczii) of section Tacamahaca is diverged from other members of the section. The incongruence between phylogenetic trees based on nuclear- and chloroplast-DNA sequence data suggests a reticulate evolution in the genus Populus. The fossil evidence and current biogeographic distribution patterns of poplars suggest that the probable hybridization underlying the origin of P. nigra, P. tristis, and P. szechuanica likely predated the break up of the Laurasian land mass in the Miocene.
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Barnes B. V. 1961 Hybrid aspens in the lower Peninsula of Michigan. Rhodora 63: 311-324
Barnes B. V. K. S. Pregitzer 1985 Occurrence of hybrids between bigtooth and trembling aspen in Michigan. Canadian Journal of Botany 63: 1888-1890[Web of Science]
Barrett J. W. O. P. Rajora F. C. H. Yeh B. P. Dancik C. Strobeck 1993 Mitochondrial DNA variation and genetic relationships of Populus species. Genome 36: 87-93
Beccerra J. D. L. Venable 1999 Nuclear ribosomal DNA phylogeny and its implications for evolutionary trends in Mexican Bursera. American Journal of Botany 86: 1047-1057
Braatne J. H. T. M. Hinckly R. F. Stettler 1992 Influence of soil water supply on the physiological and morphological components of plant water balance in Populus trichocarpa, Populus deltoides and their F1 hybrids. Tree Physiology 11: 325-340[Abstract]
Brayshaw T. C. 1965 Native poplars of southern Alberta and their hybrids. Canadian Forest Service Publication 1109, Ottawa, Ontario, Canada
Culling K. W. 1992 Design and testing of plant-specific PCR primer for ecological and evolutionary studies. Molecular Ecology 1: 223-240
Dayanandan S. K. S. Bawa R. V. Kesseli 1997 Conservation of microsatellite among tropical trees (Leguminosae). American Journal of Botany 84: 1658-1663[Abstract]
Dickmann D. I. J. G. Isebrands J. E. Eckenwalder J. Richardson 2001 Poplar culture in North America. NRC Research Press, National Research Council of Canada, Ottawa, Ontario, Canada
Dickmann D. I. K. Stuart 1983 The culture of poplars in Eastern North America. Department of Forestry, Michigan State University, East Lansing, Michigan, USA
Doyle J. J. 1992 Gene trees and species trees: molecular systematics as one-character taxonomy. Systematic Botany 17: 144-163[CrossRef][Web of Science]
Doyle J. J. J. L. Doyle 1987 A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11-15
Eckenwalder J. E. 1977a North American cottonwoods (Populus, Salicaceae) of sections Abaso and Aigeiros. Journal of the Arnold Arboretum 58: 193-208[Web of Science]
Eckenwalder J. E. 1977b Systematics of Populus L. in southwestern North America with special reference to sect. Aigeiros Duby. Ph.D. dissertation, University of California, Berkeley, California, USA
Eckenwalder J. E. 1982 Populus xinopia hybr. nov. (Salicaceae), a natural hybrid between the native North American P. fremontii S. Watts and the introduced Eurasian P. nigra L. Madroño 29: 67-78
Eckenwalder J. E. 1984a Natural intersectional hybridization between North American species of Populus in sections Aigeiros and Tacamahaca. I. Population studies of P. parryi. Canadian Journal of Botany 62: 317-324[CrossRef]
Eckenwalder J. E. 1984b Natural intersectional hybridization between North American species of Populus in sections Aigeiros and Tacamahaca. II. Taxonomy. Canadian Journal of Botany 62: 325-335[CrossRef][Web of Science]
Eckenwalder J. E. 1996 Systematics and evolution of Populus. In R. F. Stettler, H. D. Bradshaw, Jr., P. E. Heilman, and T. M. Hinckley [eds.], Biology of Populus and its implications for management and conservation, 7–32. NRC Research Press, National Research Council of Canada, Ottawa, Ontario, Canada
Fowells H. A. 1965 Silvics of forest trees of the United States. U.S. Forestry Service Agricultural Handbook 271. U.S. Department of Agriculture, Washington, D.C., USA
Guries R. P. R. F. Stettler 1976 Pre-fertilization barriers to hybridization in the poplars. Silvae Genetics 25: 37-44
Kishino H. M. Hasegawa 1989 Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data, and the branching order in Homonoidea. Journal of Molecular Evolution 29: 170-179[CrossRef][Web of Science][Medline]
Lefevre F. D. Kajba B. Heinze J. Turok 2001 Black poplar: a model for gene resource conservation in forest ecosystems. Forestry Chronicle 77: 239-244[Web of Science]
Leskinen E. C. Alstrom-Rapaport 1999 Molecular phylogeny of Salicaceae and closely related Flacourtiaceae: evidence from 5.8S, ITS1 and ITS2 of the rDNA. Plant Systematics and Evolution 215: 209-227[CrossRef][Web of Science]
Little E. L. 1971 Atlas of United States trees, vol. 1, Conifers and important hardwoods. U.S. Department of Agriculture Miscellaneous Publication 1146. U.S. Department of Agriculture, Washington, D.C., USA
Maddison D. R. W. P. Maddison 2001 MacClade 4. Sinauer, Sunderland, Massachusetts, USA
Neale D. B. N. C. Wheeler 1986 Paternal inheritance of chloroplast DNA in Douglas fir. Canadian Journal of Forest Research 16: 1152-1154[CrossRef]
Newcombe G. 1996 The specificity of fungal pathogens of Populus. In R. F. Stettler, H. D. Bradshaw, Jr., P. E. Heilman, and T. M. Hinckley [eds.], Biology of Populus and its implications for management and conservation, 223–246. NRC Research Press, National Research Council of Canada, Ottawa, Ontario Canada
Page R. D. 2000 Extracting species trees from complex gene trees: reconciled trees and vertebrate phylogeny. Molecular Phylogenetics and Evolution 14: 89-160[CrossRef][Web of Science][Medline]
Page R. D. E. C. Holmes 1998 Molecular evolution, a phylogenetic approach. Blackwell, Oxford, UK
Parrish J. T. 1987 Global palaeogeography and palaeoclimate of the Late Cretaceous and Early Tertiary. In E. M. Friis, W. G. Chaloner, and P. R. Crane [eds.], The origins of angiosperms and their biological consequences, 51–73. Cambridge University Press, Cambridge, UK
Posada D. K. A. Crandall 1998 Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817-818
Rajora O. P. B. P. Dancik 1995 Chloroplast DNA variation in Populus. Interspecific restriction fragment polymorphisms and genetic relationships among P. deltoides, P. nigra, P. maximowiczii, P. x canadensis. Theoretical and Applied Genetics 90: 324-330[Web of Science]
Ronald W. G. 1982 Intersectional hybridization of Populus sections, Leuce-Aigeiros and Leuce-Tacamahaca. Silvae Genetica 31: 94-99[Web of Science]
Sears B. B. 1980 The elimination of plastids during spermatogenesis and fertilization in the plant kingdom. Plasmid 4: 233-255[CrossRef][Web of Science][Medline]
Shimodaira H. M. Hasegawa 1999 Multiple comparison of log-likelihoods with applications to the phylogenetic inference. Molecular Biology and Evolution 16: 1114-1116[Web of Science]
Smith R. L. 1988 Phylogenetics of Populus L. (Salicaceae) based on restriction site fragment analysis of cpDNA. M.S. thesis, University of Wisconsin, Madison, Wisconsin, USA
Smith R. L. K. J. Sytsma 1990 Evolution of Populus nigra (sect. Aigeiros): introgressive hybridization and the chloroplast contribution of Populus alba (sect. Populus). American Journal of Botany 77: 1176-1187[CrossRef][Web of Science]
Staden R. 1979 A strategy of DNA sequencing, employing computer programs. Nucleic Acids Research 7: 2601-2610
Stettler R. F. H. D. Bradshaw Jr P. E. Heilman T. M. Hinckley 1996a Biology of Populus and its implications for management and conservation. NRC Research Press, National Research Council of Canada, Ottawa, Ontario, Canada
Stettler R. F. L. Zsuffa R. Wu 1996b The role of hybridization in the genetic manipulation of Populus. In R. F. Stettler, H. D. Bradshaw, Jr., P. E. Heilman, and T. M. Hinckley [eds.], Biology of Populus and its implications for management and conservation, 87–112. NRC Research Press, National Research Council of Canada, Ottawa, Ontario, Canada
Swofford D. L. 2001 PAUP*: Phylogenetic analysis using parsimony (* and other methods). Version 4.0b8 for Macintosh. Sinauer, Sunderland, Massachusetts, USA
Taberlet P. L. Gielly G. Pautou J. Bouvet 1991 Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105-1109[CrossRef][Web of Science][Medline]
Tamura K. M. Nei 1993 Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Molecular Biology and Evolution 10: 512-526[Abstract]
Templeton A. R. 1993 Convergent evolution and non–parametric inferences from restriction fragment and DNA sequence data. In B. Weir [eds.], Statistical analysis of DNA sequence data. Marcel Dekker, New York, New York, USA
Thompson J. D. D. G. Higgins T. J. Gibson 1994 CLUSTAL W, improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position, specific gap penalties and matrix choice. Nucleic Acids Research 22: 4673-4680
Viereck L. A. J. M. Foote 1970 The status of Populus balsamifera and P. trichocarpa in Alaska. Canadian Field Naturalist 84: 169-173
Villar M. M. Gaget C. Dumas 1986 Sexual reproduction biology in Populus, compatibility and incompatibility. In D. L. Mulcahy, G. B. Mulcahy, and G. Ottaviano [eds.], Biotechnology and ecology of pollen. Springer-Verlag, New York, New York, USA
Whitham T. G. K. D. Floate G. D. Martinsen E. M. Driebe P. Keim 1996 Ecological and evolutionary implications of hybridization: Populus–herbivore interactions. In R. F. Stettler, H. D. Bradshaw, Jr., P. E. Heilman, and T. M. Hinckley [eds.], Biology of Populus and its implications for management and conservation, 247–275. NRC Research Press, National Research Council of Canada, Ottawa, Ontario, Canada
Zsuffa L. 1975 A summary review of interspecific breeding in the genus Populus. In Proceedings of the 14th annual meeting of the Canadian Tree Improvement Association, part 2, 107–23. Canadian Forest Service, Ottawa, Ontario, Canada
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
