Am. J. Bot. Li-Cor Advertisement
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Semerikov, V. L.
Right arrow Articles by Lascoux, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Semerikov, V. L.
Right arrow Articles by Lascoux, M.
Agricola
Right arrow Articles by Semerikov, V. L.
Right arrow Articles by Lascoux, M.
(American Journal of Botany. 2003;90:1113-1123.)
© 2003 Botanical Society of America, Inc.

Genetics and Molecular Biology

Nuclear and cytoplasmic variation within and between Eurasian Larix (Pinaceae) species1

Vladimir L. Semerikov2 and Martin Lascoux3

Department of Conservation Biology and Genetics, Evolutionary Biology Center, Uppsala University Norbyvägen 18 D, 752 36 Uppsala, Sweden

Received for publication October 10, 2002. Accepted for publication January 30, 2003.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
The genetic variation in nuclear and cytoplasmic markers was investigated in 28 populations belonging to Eurasian Larix species (L. decidua, L. sibirica, L. gmelinii, L. olgensis, and L. kaempferi). Nuclear genetic variation was assessed at 214 AFLP loci, and both PCR-RFLP and four microsatellite loci were used to estimate variation of the chloroplast DNA. Variation of the mitochondrial genome was measured using RFLPs. Although population differentiation at both nuclear and chloroplast markers was much weaker than at mitochondrial DNA, it nonetheless corroborated the grouping observed with mitochondrial DNA. The AFLPs led to the same population grouping as mtDNA. Notably, the presence of two ancient western and eastern groups within L. sibirica was confirmed and possible postglacial routes inferred. The genetic composition of the northernmost L. sibirica population in our sample established that it is located at the confluence of the eastern and western recolonization routes. The joint use of the three markers also indicated that populations around Lake Baikal are hybrids between L. gmelinii and L. sibirica, with L. gmelinii primarily acting as the pollinator. Finally, AFLP-based estimates of nucleotide variation were an order of magnitude larger than the strikingly low estimates of nucleotide variation recently reported in Pinus sylvestris.

Key Words: AFLP • cpDNA • hybridization • Larix • mtDNA • phylogeography • Pinaceae


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Sixty years after Húlten's path-breaking treatise on the history of arctic and boreal biota during the Quaternary period (Hultén, 1937 ), the Quaternary history of most major plant species in the area still remains an enigma (but see Abbott et al., 2000 ). Húlten strongly believed that a large proportion of the arctic and boreal flora radiated from the Bering Sea area. The few available paleontological and phylogeographic studies tend to lend support to Húlten's conclusions (e.g., Prentice et al., 2000 ), though it might still be too early to draw any final conclusions, because recent comparative studies of postglacial recolonization clearly showed that recolonization patterns were extremely species-dependent (Taberlet et al., 1998 ; Hewitt, 2000 ). This does not, however, necessarily imply that species can be studied separately. In plants in general, and in Larix in particular, hybridization is widespread, and consequently, the history of species belonging to the same complex should, as far as possible, be studied simultaneously. For instance, asymmetrical hybridization between Quercus robur and Q. petraea followed by backcrosses apparently played an important role in the postglacial recolonization of western Europe by these two species (Petit et al., 2001 ).

In a previous study, we investigated allozyme diversity within and between Eurasian Larix species of section Pauciseriales (Semerikov et al., 1999 ). Variation at isozyme loci supported the separation of L. sibirica into a western and an eastern group (Dylis, 1961 ), which apparently originated from at least two last glacial maximum (LGM) Southern refugia, one in the Southern Urals and/or Northern Kazakstan, the other in the Southern Siberian mountain ranges. In addition, individuals in populations of the Siberian northwest originated from both refugia. Reconstructed biomes at LGM are in agreement with such a scenario (Tarasov et al., 2000 ). Surprisingly, given the present geographical distribution of Eurasian Larix species, L. sibirica was genetically closer to L. olgensis than to L. gmelinii (Fig. 1). This may not, however, conflict with the fossil record. It is generally inferred that after its emergence L. gmelinii expanded south and southwest, forcing out both L. sibirica and L. olgensis, which were less adapted to the new, more continental, climatic conditions of the Pleistocene (Dylis, 1961 ). As a result, the common L. sibirica–L. olgensis range was disrupted. The genetic closeness of these two species for isozyme frequencies might therefore be a consequence of the relatively short period of time that elapsed since that event. Finally, there was a vast transitional area, with populations with intermediate isozyme frequencies, between L. gmelinii and L. olgensis. The nature of this transitional zone is unclear: it might be the result of ongoing speciation or of secondary gene flow. Hence, climatic reconstructions, fossils, and genetic data suggest that the Eurasian Larix complex of species went through a rather dynamic and complex history.



View larger version (66K):
[in this window]
[in a new window]
  		Fig. 1. (a) Location of the populations used in the present study. The full names and coordinates of all populations are given in the Appendix (see Supplementary Data accompanying the online version of this paper). (b) Distribution of the most informative mitochondrial haplotypes. Black, white, and shaded squares correspond to haplotypes 3, 7, and 8, respectively. White, black, and shaded circles correspond to haplotypes 15 and 18, 23, and 30, respectively. The white stars correspond to haplotypes 1 and 2

 
In conifers, mtDNA and cpDNA are maternally and paternally inherited, respectively, hence providing an exceptional opportunity for studying maternal and paternal lineages within a single species (Palmer, 1992 ). The number of large-scale population genetic studies based on plant mitochondrial DNA has been rather limited so far. Soltis and Soltis (1999) listed up to eight reasons to explain the lack of emphasis on plant mitochondrial DNA. Briefly, plant mitochondrial DNA are large, unstable, and dynamic genomes that undergo frequent rearrangements, with additionally a very low rate of nucleotide substitution compared to the mitochondrial genome of other organisms or the chloroplast or nuclear genomes of plants. In spite of these limitations, the availability of new PCR-based markers, together with its maternal inheritance in conifers, has led to a renewed interest lately (e.g., Latta and Mitton, 1997 , 1999 ; Latta et al., 1998 ; Tsumura and Suyama, 1998 ; Wu et al., 1998 ; Mitton et al., 2000 ). In contrast, chloroplast DNA has been the marker of choice for phylogenetic and phylogeographic studies in plants, especially angiosperms, in which it is maternally inherited (Taberlet et al., 1998 ; Abbott et al., 2000 ; Hewitt, 2000 ). Two techniques were primarily used, PCR-RFLP, and more recently, microsatellites. In conifers, chloroplast DNA is paternally inherited (Neale et al., 1986 ; Wagner et al., 1987 ), and therefore one would expect a lesser phylogeographic resolution than in angiosperms. While this is certainly correct, the variation and geographical structure at three chloroplastic microsatellite loci was nonetheless largely sufficient to recover the main post-glacial recolonization routes in Picea abies (Vendramin et al., 2000 ).

Because of their different inheritance mode, effective population size, and mutation rates, cytoplasmic and nuclear markers will capture demographic processes acting on different time scales, a property that will ultimately be needed to assess Hultén's hypothesis. The effective population size as assessed by cytoplasmic markers in a monoecious species is half that derived from nuclear markers, and for maternally inherited markers, dispersal is more restricted geographically because gene flow occurs only through seeds. Hence ceteris paribus, cytoplasmic markers, especially those that are maternally inherited, should (1) allow the detection of stronger genetic population differentiation than nuclear ones and (2) be more strongly affected by recent demographic events than nuclear markers. For example, the effect of a bottleneck will be more pronounced and last longer for mtDNA than for nuclear DNA (Fay and Wu, 1999 ). In general, because of their smaller effective population size, cytoplasmic markers will reflect more recent history than nuclear ones. There is also another incentive to the joint use of cytoplasmic and nuclear markers. Both chloroplast and mitochondrial DNA experience little or no recombination and are therefore equivalent to a single locus. They will consequently allow us to retrieve a single outcome of the genealogical process or coalescent, a process that is notoriously variable (e.g., Rosenberg and Nordborg, 2002 ). Therefore, if one wants to go beyond the identification of the main postglacial recolonization routes, nuclear genes also need to be considered.

As nuclear markers we used amplified fragment length polymorphism (AFLPs). Until recently, the use of AFLPs (Vos et al., 1995 ) was mainly limited to molecular plant and animal breeding and to fingerprinting in natural populations. The AFLPs have been less popular in phylogeny or population genetics studies, but this is slowly changing, and AFLPs have been successfully used in both phylogenetics and phylogeographic studies (Sharbel et al., 2000 ; de Kniff et al., 2001 ; Giannasi et al., 2001 ). The AFLPs should also allow us to get a first, accordingly rough estimate of the nuclear nucleotide variation (Innan et al., 1999 ). Dvornyk et al. (2002) recently showed that nucleotide variation in Pinus sylvestris at the pal1 locus and 11 other loci (albeit with a much more limited sample size) was unexpectedly low given the high heterozygosity generally observed at isozyme loci in conifers.

The aim of the present study is to test some simple hypotheses on genetic variation and past demographics in Eurasian Larix species. Are northern populations of L. sibirica admixed ones as suggested by isozymes (Semerikov et al., 1999 )? If that is the case, what is the nature of the admixture? What is the extent of hybridization between L. sibirica and L. gmelinii? When hybridization takes place, is the pollen flow bidirectional? Can populations of L. gmelinii and L. olgensis be separated using AFLPs? Is the nucleotide variation at nuclear genes as limited in Larix as it was in Pinus sylvestris (Dvornyk et al., 2002 )? We used mitochondrial (mt), chloroplast (cp), and nuclear (n) DNA polymorphisms on the same set of individuals in populations of L. decidua, L. sibirica, L. gmelinii, L. olgensis, L. kaempferi, and a putative hybrid zone between L. sibirica and L. gmelinii. Because the numerous Larix varieties traditionally recognized in the Russian Far East were part of the L. olgensis–L. gmelinii transitional zone in our previous study (Semerikov et al., 1999 ) we chose to regard them as L. olgensis in the present study.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 LITERATURE CITED
 
Plant material and DNA extraction
Leaves or seeds were sampled from natural populations across most of the range of Eurasian Larix species (Table 1 in Supplementary Data in the online version of this paper; Fig. 1). Freshly collected needles were kept in liquid nitrogen and then at 70°C. DNA was isolated from 0.2–0.5 g of needles according to Devey et al.'s (1996) protocol (cited in Ostrowska et al., 1998 ), with small modifications: the concentration of Tris-HCl was 100 mmol/L instead of 50 mmol/L, and 0.1% ascorbic acid was added to the extraction buffer. Two chloroform-isoamyl alcohol extractions were used. After precipitation, the pellets were washed by 70% ethanol and dissolved in TE buffer pH 7.5. After RNAse (400 µg/mL) treatment and two extractions, phenol/chlorophorm/IAA (25 : 24 : 1) and chloroform/IAA (24 : 1), DNA was precipitated with isopropanol, washed with 70% ethanol, and dissolved in TE. The DNA concentration was measured using fluorometry. Seeds were first germinated and DNA extracted from leaves. The same DNA samples were used in all analyses.


View this table:
[in this window]
[in a new window]
  		Table 1. Mitochondrial haplotypes in North Baikal populations of Larix

 
Restriction, electrophoresis, and Southern blotting
DNA (3–5 µg) was digested separately with EcoRV, BamHI, HindIII, and DraI enzymes. The resulting fragments were separated at 1.5 V/cm on 0.8% agarose gel in 1x TAE buffer. After electrophoresis, DNA was transferred to a nylon HybondN + membrane (Amersham Pharmacia Biotech, Freiburg, Germany) using the alkali transfer method.

Mitochondrial DNA probe preparation
The DNA fragments for the hybridization experiments were PCR-amplified from Larix genomic DNA using pairs of universal primers based on the conservative sequences of the following mitochondrial genes: nad1/B-C intron (Demesure et al., 1995 ), nad4/3-4r, nad5/1-2r, and nad5/4-5r introns (Dumolin-Lapègue et al., 1997 ), coxI gene and coxIII gene (Tsumura and Suyama, 1998 ), atp1 gene, atp6 gene, cob gene, nad3 gene, and rps14 gene (Wu et al., 1998 ). The PCR reaction mixture contained 1x Taq polymerase buffer, 1.8 mmol/L MgCl2, 0.1 mmol/L dNTPs each, 0.2 µmol/L primer 1 and 0.2 µmol/L primer 2, 0.016 unit/µL Taq polymerase, and ~20 ng of genomic DNA. Temperature profile was as following: one cycle of 3 min at 94°C, 35 cycles of 30 s at 94°C, 30 s at annealing temperature, 2 min at 72°C, and one cycle of 6 min at 72°C. The annealing temperature was recommended by the authors who designed the primers or was optimized using a temperature-gradient PCR. PCR product amplified from Larix genomic DNA was purified after separation on an agarose gel using a QIAEX II gel extraction kit (QIAGEN, Hilden, Germany). A radioactive probe was prepared from 100–200 ng of PCR product using a random labeling kit (Pharmacia, Uppsala, Sweden). The labeled mixture was then passed through Sepharose CL-6B spin column to purify the labeling probe.

Hybridization
Prehybridization (Sambrook et al., 1989 ) was conducted for 4 h at 65°C in a solution containing 1% dextran sulfate, 4x SETS (3 mol/L NaCl, 20 mmol/L EDTA pH 8.0, 0.6 mol/L Tris-HCl pH 8.0, 11 mmol/L tetra-sodium pyrophosphate), 10x Denhardt's solution, 0.1% sodium dodecyl sulfate (SDS). Then 100 µL of freshly boiled herring testes DNA (10 mg/µL) were added to 30 mL volume of this solution before the prehybridization. Hybridization was conducted overnight at 65°C. The hybridization solution was the same as the prehybridization one, but 10% dextran sulfate was used instead of 1%. Hybridized filters were then washed twice in 2x SSC (NaCl, sodium citrate), 0.1% SDS for 15 min at 65°C and 3–4 times in 0.2x SSC, 0.1% SDS for 15 min at 65°C. Filters were then exposed to BIOMAX Kodak film with an intensifying screen 1–7 d at –70°C.

Initial screening for variation
Eight individuals on average were taken in four populations of western L. sibirica, eastern L. sibirica, L. gmelinii, and L. olgensis for initial screening of genetic variation. Only probe-enzyme combinations giving a variable and clear pattern were retained for subsequent analysis.

Chloroplast and mitochondrial PCR-RFLP
In a companion study, chloroplast PCR-RFLP was used for phylogenetic reconstruction in the genus Larix (Semerikov et al., 2003 ) with polymorphic chloroplast fragment/enzyme combinations that were polymorphic for L. decidua, L. sibirica, L. gmelinii, L. olgensis, and L. kaempferii: TF/HinfI; rp120trnW/HpaII; trnLV/HaeIII; CS/HaeIII; and psbD16S/MboI. Two mitochondrial fragment/enzyme combinations were also used: nad5(1/2)/HinfI and nad4(3c4r)/HinfI. The five chloroplast and two mitochondrial fragments were amplified using published universal primers (Taberlet et al., 1991 ; Demesure et al., 1995 ; Dumolin-Lapégue, 1997 ; Parducci and Szmidt, 1999 ) following the authors' recommendations. Five-microliter portions of the PCR product were digested by one of the restriction enzymes HaeIII, HinfI, HpaII, or MboI, according to the manufacturer's instructions, and resolved in a 6% polyacrylamide sequencing gel with subsequent silver staining. Special attention was paid to the temperature during electrophoresis. At the start of the electrophoresis the temperature was around 29–33°C. During the gel run the power was kept to 75 W or at 43°C. This temperature regime gives the possibility of obtaining two bands corresponding to a single restriction fragment: the first one corresponds to a single-stranded DNA band and the second one, migrating faster, is a double-stranded DNA band. The polymorphism of the single-stranded bands reveals variation in restriction sites or length mutations. The polymorphism of the double strand bands usually appears either as two distinct variants with different electrophoretic mobilities or simply as morphological variants, the stains being of different shapes. This variation is likely the result of sequence differences inside the fragment or of a very short indel that is not detectable by a single-stranded band. This approach allows the discovery of mutations that were not observed solely by restriction fragment size. We scored these double-stranded variants whenever possible. In most cases, there were two clearly distinguishable variants, and the results were reproducible (Fig. 2). It should be noted that in the case of the chloroplast rpl20-trnW/HpaII combination, we verified the polymorphism separating L. sibirica, American Larix species, and Pseudotsuga menziesii from the other Larix species by sequencing most of this fragment (Semerikov et al., 2003 ; GenBank accessions AY131239–AY131260). The sequences confirmed that the polymorphism corresponded to a single nucleotide difference. The restriction fragment patterns produced by each enzyme were defined as restriction site or indel presence (1) or absence (0). The same code was also used when scoring double stranded polymorphisms.



View larger version (85K):
[in this window]
[in a new window]
  		Fig. 2. Chloroplast fragments amplified from Larix species using the rp120-trW pair of primers. The PCR product was cut with the HpaII resptriction enzyme. The top line gives the species identity: s, L. sibirica; g, L. gmelinii; d, L. decidua; k, L. kaempferi; l, L. Laricina; and p, L. potaninii. The polymorphism associated with the 170-bp double-stranded fragment is indicated at the bottom. The corresponding single-stranded fragment is indicated by an asterisk. See text for additional details. The 5-bp insert mutation specific to L. decidua is also visible at 305 bp

 
Chloroplast microsatellite procedure
The following loci (Vendramin et al., 1996 ) were assayed: Pt9383, Pt15169, Pt26081, Pt30204, Pt36480, Pt41093, Pt45002, Pt48210, Pt51873, Pt63718, Pt71936, Pt87268, and Pt110048 on 16 individuals sampled in different larch populations across Eurasia and America. Only Pt9383, Pt26081, and Pt30204 were used subsequently. The PCR conditions for Pt26081 and Pt30204 were as follows: 1x PCR buffer, 1.8 mmol/L MgCl2, 0.1 mmol/L dNTPs, 0.4 µmol/L each primer, 0.016 unit/µL Taq polymerase, 4 ng/µL genomic DNA. Thermal profile was 5 min at 94°C, 35 cycles of 30 s at 94°C, 30 s at 55°C, and 30 s at 72°C, 6 min at 72°C and then 10°C. For Pt9383, the conditions were different: MgCl2 concentration was 2.5 mmol/L and annealing temperature was 48°C. The PCR products were resolved in 6% polyacrilamide sequencing gel and visualized using a silver staining kit (Stratagene, La Jolla, California, USA). A fourth microsatellite locus was obtained by resolving in sequencing gel the trnLV/HpaII product. One of the restriction fragments contained the microsatellite that was scored.

AFLPs
The AFLP technique (Vos et al., 1995 ) was used for analyzing species differentiation at the nuclear level. DNA was digested with EcoRI and MseI. Three selective nucleotides were used in the case of the EcoRI primer and four for the MseI primer. The EcoRI primer was labeled by {gamma}33P-ATP. A total of three primer combinations were used: Eco+ACGxMse+CCCA, Mse+CCAC, Mse+CCAG.

Data analysis
Because, as we shall show later, hybridization does occur among species and because we are primarily interested in population inferences, classical phylogenetic methods are not adequate (Rosenberg and Nordborg, 2002 ). We have therefore refrained from producing any phylogenetic tree because such a tree may simply not exist.

Mitochondrial RFLP
The relationship among haplotypes was described by a median joining (MJ) network drawn using the program Network 3.1.1.1 (http://www.fluxus-engineering.com; Bandelt et al., 1999 ). Because MJ networks can be unreliable for long branches, we also drew a reduced median (RM) network (Bandelt et al., 1995 ), which is more robust. Because both methods led to the same haplotypes grouping we only report the MJ network.

Chloroplast DNA
Only two major haplotypes were found with PCR-RFLP, defining two main haplogroups. The haplotypes within the haplogroups could be defined based on the variation at the four microsatellite loci. Principal components analysis (PCA) was used to visualize the relationship among microsatellite haplotypes with the program NTSYS-pc v. 2.02h (Rohlf, 1998 ).

AFLP
To analyze the relationship among populations, we have applied a model-based genetic structure analysis, recently developed by Pritchard et al. (2000) and implemented in the program structure. Because AFLPs are dominant markers, the model assumed no admixture. Multilocus genotypes are considered, and no particular mutational model is assumed. Each individual is assumed to originate in one of K populations. The allelic frequencies at each locus within each population follow a Dirichlet distribution. Under the assumption that each of the populations is in Hardy-Weinberg equilibrium, the number of populations, K, is estimated by a Monte Carlo-Markov Chain algorithm. Nucleotide variation in all species was estimated following Innan et al. (1999) .

Wright fixation indices
The hierarchical components of mtDNA, cpDNA, and AFLP variation were computed under the AMOVA framework using ARLEQUIN software, version 2.0 (Schneider et al., 2000 ).

Coalescent
The present genetic composition of a population is intimately related to its genealogical history. A major recent advance in the understanding of gene genealogies is the so-called coalescent approach. Loosely speaking, the basic idea of the coalescent is to draw a sample of genes from a population and trace back the genes' ancestry, focusing on when two or more genes in the sample derived from a common ancestor. Kingman (1982) showed that this backward process is well described for a classical Wright-Fisher model by a simple time-homogeneous Markov chain, called the coalescent. Since then, the coalescent has been extended to a wide range of demographic and mutation models (see Donnelly and Tavaré, 1995 ; Donnelly, 1999 ). Genealogical inferences based on the coalescent remain difficult, but computationally tractable methods are starting to be available. In the present study we used the program MICSAT developed by Wilson and Balding (1998) for linked microsatellite loci. The program is based on a Markov Chain Monte Carlo (MCMC) algorithm and estimates the posterior distributions of the mutation rate, µ; the effective population size, N; the product {theta} = Nµ, the height and the length of the genealogy. The time to the most recent common ancestor, TMRCA, can be derived from the latter as height x N x generation length. The program was run with 50 000 warm-ups and 50 000 samples.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 LITERATURE CITED
 
Mitochondrial DNA
Figure 3 gives the network relating the haplotypes. Roughly, haplotypes can be grouped into three main subnetworks radiating from a group of haplotypes that includes the two main L. olgensis haplotypes. A first subnetwork, to which L. decidua haplotypes are connected, corresponds to eastern Larix sibirica haplotypes. A second subnetwork includes the two western L. sibirica haplotypes as well as a few eastern L. sibirica ones. Finally, a third subnetwork consists of L. gmelinii haplotypes and a few haplotypes present in L. olgensis.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 3. Median-joining network based on mitochondrial RFLPs. The size of the circles is proportional to the haplotype frequencies. Light gray, dark gray, shaded gray, and white circles correspond to haplotypes present only or predominantly in western Larix sibirica, eastern L. sibirica, L. gmelinii, and L. olgensis, respectively. Haplotypes 1 and 2, 41, and 43, all given in black, correspond to L. decidua, L. kaempferi, and L. laricina, respectively. The dotted lines circumscribe the three main clusters described in the text

 
The number of haplotypes varies strongly among species, ranging from one (or two but with a predominant one) in "marginal" species (L. decidua, L. kaempferi) to up to 13 in eastern L. sibirica (Table 2 in Supplementary Data in the online version of this paper; Figs. 1b, 3). Somewhat surprisingly, only two haplotypes were found in western L. sibirica, a species with a large distribution range, one of them being almost fixed (haplotype 3). Few haplotypes were shared by western and eastern L. sibirica: only two copies of haplotype 3 were found in population Kahem. In agreement with our previous allozyme study, Pangodi, a population close to the Ob River Delta (Pan on the map), which otherwise had mostly haplotype 3, also had a few copies of haplotypes 6 and 7 that are specific to the eastern populations of L. sibirica. Haplotypes 6 and 7 were not common in the southernmost mountain populations (Ong, Kiz, Chd, Tan, Kah, Ere), and it is noteworthy that both Pangodi and Krasnoiarsk lack haplotype 8, which was the most frequent in these southern mountain populations. The fact that Krasnoiarsk differed from other eastern Larix sibirica populations was also clear from AFLP data (see later). Finally, some haplotypes are shared by L. gmelinii and L. olgensis, though this is in large part due to the Urgal population (Urg) that was classified as a L. gmelinii but only shares haplotypes with other L. olgensis populations and has few haplotypes in common with other L. gmelinii populations. One can also observe a geographical separation of the most common ones (Fig. 1).


View this table:
[in this window]
[in a new window]
  		Table 2. Nucleotide diversity in nuclear DNA of some Larix species estimated from AFLP using the method of Innan et al. (1999). N is the number of individuals, m is the average number of bands scored, F is the expected proportion of shared bands, and {pi} is the nucleotide diversity

 
The hybrid nature of the North Baikal population is also confirmed by the distribution of its haplotypes: haplotype 7 is shared with eastern L. sibirica, haplotypes 15 and 18 with L. gmelinii, while haplotypes 9 and 16, which occur at much lower frequency, were specific to the population. Table 1 shows the five haplotypes observed in the hybrid L. sibirica x L. gmelinii population. In haplotype 16 all probe/enzyme combinations that separate L. sibirica and L. gmelinii show the pattern specific for L. gmelinii, and a single one (atp1/DraI) shows the pattern specific for L. sibirica.

Chloroplast DNA
Variation at PCR-RFLPs was limited and essentially defined two major strongly differentiated haplogroups, one including L. sibirica and the other including all other species. Within the L. sibirica haplogroup a rare variant could also be observed in the eastern group. The PCA analysis based on four microsatellite loci delineates two major clusters that correspond, with a few exceptions, to the two haplogroups defined by PCR-RFLP (Table 4 in Supplementary Data in the online version of this paper; Fig. 4). The rightmost one consists of haplotypes found in western and eastern L. sibirica. Few of these were shared by both subspecies. The leftmost cluster contains haplotypes from all other species including L. decidua as well as a group of haplotypes characteristic of western L. sibirica. As for mitochondrial DNA, the current relative geographical positions are not entirely reflected by the relative position of the haplotypes specific of the different species. Larix olgensis and L. gmelinii share most of their chloroplast haplotypes, whereas other species tend to have their own haplotypes with little or no common ones. One would have expected haplotypes of eastern L. sibirica to be closest to those of the L. gmelinii-olgensis cluster. Instead, haplotypes of western L. sibirica occupy this position. This, however, likely reflects homoplasy because these microsatellite haplotypes all belong to the L. sibirica PCR-RFLP haplogroup. The North Baikal population, which is presumably of hybrid origin, comprises only L. gmelinii haplotypes.


View this table:
[in this window]
[in a new window]
  		Table 4. Allele-frequency divergence among populations (Kullback-Leibler distance; notice that DA,B and DB,A are not, in general, the same)

 


View larger version (29K):
[in this window]
[in a new window]
  		Fig. 4. Principal components analysis of chloroplast microsatellite haplotypes. Haplotypes belonging to the Larix gmelinii–L. olgensis–L. decidua–L. kaempferi group are circled by a dashed line. Haplotypes belonging to western and eastern L. sibirica are circled by a continuous line. Larix laricina is represented by a pentagon. The circles are approximately proportional to the importance of the haplotypes

 
While mitochondrial variation was much less in western L. sibirica than in eastern L. sibirica, no such pattern clearly emerged at the chloroplast level, the number of haplotypes were roughly the same in the two subspecies. Eastern L. sibirica haplotypes form a single cluster, while western L. sibirica haplotypes split into two clusters: a first one roughly overlapped with eastern L. sibirica and a second one clearly separated from the first. This subdivision depends entirely on the Pt30 microsatellite and haplotypes in the second cluster differing from those in the first one by 4–5 nucleotides. Presumably, a single 4–5 bp deletion is the cause of this subdivision.

Coalescent
The distribution of estimates of the mutation rate, the effective population size, and the height of the genealogy in coalescent units are given in Fig. 5. We assumed a single, constant population size. The mean effective population size and the mean height of the coalescent tree were 4213 and 0.702, respectively. To get the equivalent in years, multiply by N and the generation time. In larch, a generation time of 80–100 yr seems reasonable. We then obtain an estimate for the time to the most recent common ancestor (TMRCA) around 240 000 yr. When population splitting is assumed (data not shown), the TMRCA was a bit lower (200 000 yr). Interestingly, values of the same order of magnitude were obtained in lodgepole pine (Marshall et al., 2002 ).



View larger version (13K):
[in this window]
[in a new window]
  		Fig. 5. Distribution of the mutation rate, µ (top panel), the effective population size, N (middle panel), and the time to the most recent common ancestor in coalescent units, T (bottom panel), obtained through coalescent analysis of Larix sibirica chloroplast microsatellites

 
Nuclear DNA
A total of 214 fragments was scored when the three AFLP primer combinations were pooled. Fifteen bands were monomorphic and 199 were polymorphic. Estimates of nucleotide diversity are given in Table 2. There were basically no differences among taxa. The number of clusters with the highest probability was six. The proportion of membership of each population in each of the six clusters is given in Tables 3 and 4. This shows that the six clusters correspond to eastern L. sibirica, western L. sibirica, L. olgensis, L. gmelinii, L. decidua, and L. kaempferi. It should be noted that the allele-frequency divergence between the L. olgensis and L. gmelinii cluster was very weak; consequently, these two clusters were not very stable. The structural analysis also clarifies the status of some populations. Both the Pangodi and Krasnoiarsk populations were admixed, and the South Sakalhin population appears to be L. gmelinii rather than L. olgensis. As for mtDNA, the situation of the Pangodi population probably reflects its mixed origin at the confluence of the eastern and western postglacial recolonization routes. Similarly, individuals from populations Urgal, Vanino, and Kavalero assign to both the L. olgensis and L. gmelinii clusters, although in all three cases most individuals assign primarily to one of the two species.


View this table:
[in this window]
[in a new window]
  		Table 3. Summary of the clustering results based on AFLPs

 
Wright fixation indices
The AMOVA analysis for mtDNA, cpDNA, and AFLP indicates that differentiation among species as well as among populations within species was much more pronounced for mtDNA than for either cpDNA or nuclear AFLPs (Table 5).


View this table:
[in this window]
[in a new window]
  		Table 5. Wright fixation indices. FCT is the differentiation among species, FSC measures the differentiation among populations within species, and FST measures the differentiation among populations. All values are highly significantly different from zero

 
Hybridization and admixture
The allozyme pattern of the North Baikal population (Semerikov et al., 1999 ) as well as its geographical location at the L. sibirica–L. gmelinii border (Kruklis and Milyutin, 1977 ) suggested that this population consists of hybrids between L. sibirica and L. gmelinii. This is confirmed by the presence of mitochondrial haplotypes characteristic of either species in almost equal proportion. The structure of haplotype 16, which is specific for the North Baikal population, might have originated through recombination among L. sibirica and L. gmelinii lineages. The pattern for chloroplast microsatellites and nuclear polymorphism (allozymes and AFLP) was different. Only L. gmelinii chloroplast haplotypes were found in this population and, although the population was genetically closer to L. gmelinii, there was a significant proportion of alleles common in L. sibirica for both allozymes and AFLPs.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
LITERATURE CITED
 
Phylogeography
The joint analysis of variation at chloroplast, mitochondrial, and nuclear DNA in Eurasian Larix species has confirmed and extended the conclusions reached in our previous study on the phylogeography of Eurasian Larix species (Semerikov et al., 1999 ). Based on the variation in mitochondrial DNA and AFLPs, Eurasian Larix species can clearly be divided into six groups, from west to east: L. decidua, western L. sibirica, eastern L. sibirica, L. gmelinii, L. olgensis, and L. kaempferi with hybrid or transition zones between L. sibirica and L. gmelinii and between L. gmelinii and L. olgensis. Chloroplast DNA microsatellites basically retrieved the same grouping, but the division between western and eastern L. sibirica was less pronounced than that between L. gmelinii and L. olgensis. Present populations of L. sibirica originate from at least two different refugia, one in the southern Urals and northern Kazakhstan, the other in southern Siberian mountain ranges. The recolonization of the western Siberian plain was associated with both of these refugia as indicated by the presence of cytoplasmic haplotypes specific both to western and eastern L. sibirica in the Pangodi population. These two putative refugia are consistent with recent paleoecological data that indicates that taiga was present during the Last Glacial Maximum in southern Urals-Northern Kazakstan on the one hand and in the mountain ranges of south Siberia on the other hand (Tarasov et al., 2000 ), while the vegetation in central and northern Siberia was either a steppe or a tundra. The absence in northern populations of L. sibirica of mtDNA haplotype 8, which is the most common in the Altaï and Tuva regions, suggests that the central part of the mountain area of Altaï and Tuva was unlikely to be the direct source of the last recolonization. However, more extensive sampling is necessary to determine the location of the refugia in East Siberia. The large rivers flowing northwards in the area, the Ob and the Ienissei, could have been major recolonization routes. Savile (1972) considered dispersal by wind in winter to be the most important dispersal mechanism in arctic plants. In the present case, the rivers when frozen would provide smooth surfaces on which the seeds could be dispersed over long distances before being deposited. Water flow, especially during the spring breakup of ice, is also a likely powerful dispersing agent: no data are available for Larix, but Picea abies seeds, for instance, can float for a few days and survive a few additional days once they sink (S. N. Sannikov, Botanical Garden, Yekaterinburg, personal communication). In any case, additional sampling along the main rivers of the area will be needed to ascertain their role as major dispersal corridors.

We can envision two possible scenarios regarding the origin of western and eastern L. sibirica. First, western L. sibirica could be derived from eastern L. sibirica. This scenario is in part supported by the structure of the mitochondrial network, with the most common haplotype in western L. sibirica derived from a haplotype in eastern L. sibirica. Alternatively, western and eastern L. sibirica could have stemmed out of a common ancestor, most likely L. olgensis, at approximately the same time, the few shared mitochondrial haplotypes between the two species being the consequence of secondary contact. If these two scenarios are difficult to tell apart, the data nonetheless unravel two aspects of L. sibirica history. First, the coalescent estimates of the time to the most recent common ancestor based on cpDNA, the structure of the mtDNA network, and the marked differentiation between western and eastern L. sibirica at AFLP loci (FCT between eastern and western L. sibirica is 0.14, P < 0.001) all support an ancient separation into eastern and western L. sibirica, likely predating the last glacial, but more recent than the separation of L. decidua from other species. The latter is suggested by the fact that L. decidua mtDNA haplotypes are closer to eastern L. sibirica haplotypes than they are to western L. sibirica ones. Generally, L. decidua differs from other species at all three classes of markers. This, however, could be an artifact from the origin and limited size of our L. decidua sample. Second, the almost complete lack of mitochondrial variation in western L. sibirica, but similar levels of variation at AFLP loci between eastern and western L. sibirica, suggests that western L. sibirica went through an ancient bottleneck, with sufficient time for nuclear DNA but not for mitochondrial DNA to recover the lost variation (Fay and Wu, 1999 ). Third, it is noteworthy that a few copies of the mtDNA haplotypes specific to western L. sibirica are found in eastern L. sibirica, but the converse is not true.

The L. olgensis–L. gmelinii complex experiences recurrent gene flow among pure L. olgensis, L. gmelinii, and putative hybrid populations, leading to the homogenizing of the chloroplast gene pool and the formation of intermediate mitochondrial haplotypes. In the mitochondrial network the presence of shared haplotypes between L. olgensis and L. gmelinii ("21," "23," "22") and of haplotypes located closely to both L. olgensis and L. gmelinii ("26," "33," "38," "20") suggests that these species diverged recently or that gene flow between them is extensive. At the same time, it is clear that gene flow has not been sufficient to completely homogenize the two species: mitochondrial haplotypes such as "30" are only found in L. olgensis and haplotypes "15" and "18" were not found in any other species than L. gmelinii. Generally, the L. olgensis group appears to be the most ancient: its mitochondrial haplotypes are central in the median joining network and the two "outgroups" (L. kaempferi and L. laricina) derive from them. Larix olgensis also has, though one should admit marginally so, the highest AFLP nucleotide diversity. The latter can of course simply be explained by a larger effective size but the fact that diversity is of the same order of magnitude in species covering very different range sizes makes this unlikely. The MJ network based on mitochondrial DNA further suggests that L. gmelinii could derive from populations harboring haplotype 23 (Urgal, Vanino, and Kavalero), while eastern L. sibirica would have originated from populations located southeast of those (Olga Bay, Ussuriisk, Korea). This last hypothesis could be tested by sampling populations along a transect going from Korea to the Tuva and Altai Mountains.

Hybridization
Differences in hybridization patterns among mitochondrial, chloroplast, and nuclear genomes indicate that L. gmelinii is the main pollen source. This could be due to biological barriers preventing L. sibirica from pollinating L. gmelinii or simply be a consequence of directional pollen dispersal by winds blowing from the L. gmelinii region to the areas occupied by L. sibirica. Such pollen dispersal will eventually lead to the gradual replacement of the L. sibirica nuclear genome and to the fast disappearance of the L. sibirica cpDNA. However, this mechanism would not affect the mitochondrion that is only transmitted through seeds. According to pollen data (Horiuchi et al., 2000 ), forest vegetation started to expand around Lake Baikal only about 15 000 yr BP. Hence, at most 150 generations have passed since L. sibirica and L. gmelinii occupied the area around Lake Baikal. This relatively short time could explain the fact that only a single mitochondrial haplotype specific to the hybrid population was observed. This haplotype could be the result of mutations or recombination among hybridizing lineages. While DeVerno et al. (1993) showed that mitochondria are maternally inherited in Larix, recent cytological studies of fertilization in conifer species suggest that paternal leakage of mitochondria occurs during embryo formation (e.g., Bruns and Owens, 2000 ; Guo et al., 2000 ). Heteroplasmic individuals could therefore occur and homologous recombination among different mtDNA could, in principle, take place, especially in hybrids. Heteroplasmy was indeed observed in Picea abies (mitochondrial nad1(B/C) intron, Sperisen et al., 2001 ), and Laser et al. (1997) found inter-lineage mtDNA recombination to be frequent in rye–wheat hybrids. In Larix, the available evidence is too limited to conclude that recombination occurred, but this hypothesis cannot be ruled out.

Nuclear and cpDNA diversity
With a mean value close to 0.02 the nuclear nucleotide diversity, {pi}, across Eurasian Larix species was an order of magnitude higher than the nucleotide diversity reported in P. sylvestris (Dvornyk et al., 2002 ). These figures should, however, be interpreted with caution. As already pointed out by Innan et al. (1999) , nucleotide diversity estimated from AFLP tends to be larger than nucleotide diversity from DNA sequences. First, AFLPs provide an estimate of {pi} for the total genome, which contains regions of both low and high selective constraints. Second, Innan's et al. (1999) method also assumes that insertions and deletions are very rare. However, if the latter are frequent then an overestimate will be obtained. On the other hand, size homoplasy among AFLP fragments should lead to a downward bias. Innan et al. (1999) takes the latter into account but, as shown by Vekemans et al. (2002) , their overall correction factor may not be sufficient because the downward bias strongly varies with fragment size. In summary, keeping all these caveats in mind, it appears that, contrary to the situation in P. sylvestris, Larix species do not have a particularly low level of nucleotide variation.

Similarly, the mutation rate of cpDNA microsatellites was much higher than the value reported by Provan et al. (1999) in P. torreyana (µ {approx} 10–5) but was of the same order of magnitude as the estimates obtained in P. contorta (Marshall et al., 2002 ) or Picea abies (M. Lascoux, unpublished data). The latter were also obtained through coalescent analysis, whereas the former was obtained by assuming that P. torreyana went through a severe bottleneck and that the gene genealogy is perfectly star-shaped. Because no mutations were observed in P. torreyana the resulting estimate therefore only reflects the assumed demographic model (Brookfield, 1997 ). Though coalescent-based estimates of mutation rate should be taken with caution as they too reflect the assumed demographic model and are extremely variable, the fact that most analyses so far have led to estimates closer to 10–3 than 10–5 suggests that the latter estimate should not be taken at face value. In any case, these indirect estimates are only a surrogate for direct estimates that will be required in the long run.


   FOOTNOTES
 
1 The authors thank the Royal Swedish Academy of Sciences for support to VLS. Back

2 Present address: Institute of Plant and Animal Ecology, Laboratory of Dendrochronology and Population Biology, Russian Academy of Sciences, Yekaterinburg, Russia Back

3 Author for reprint requests (e-mail: Martin.Lascoux{at}ebc.uu.se ; Tel: (46) 18 471 64 16; Fax: (46) 18 471 64 24) Back


   LITERATURE CITED
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Abbott R. J. L. C. Smith R. I. Milne R. M. M. Crawford K. Wolff J. Balfour 2000 Molecular analysis of plant migration and refugia in the Arctic. Science 289: 1343-1346[Abstract/Free Full Text]

Bandelt H.-J. P. Forster A. Röhl 1999 Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution 16: 37-48[Abstract]

Bandelt H.-J. P. Forster B. C. Sykes M. B. Richards 1995 Mitochondrial portraits of human populations. Genetics 141: 743-753[Abstract]

Brookfield J. F. Y. 1997 Importance of ancestral DNA ages. Nature 388: 134.[Medline]

Bruns D. J. N. Owens 2000 Western white pine (Pinus monticola Dougl.) reproduction: II. Fertilisation and cytoplasmic inheritance. Sexual Plant Reproduction 13: 75-84[CrossRef][ISI]

de Kniff P. F. Denkers N. D. van Swelm M. Kuiper 2001 Genetic affinities within the herring gull Larus argentatus assemblage revealed by AFLP genotyping. Journal of Molecular Evolution 52: 85-93[ISI][Medline]

Demesure B. N. Sodzi R. J. Petit 1995 A set of universal primers for amplification of polymorphic noncoding regions of mitochondrial and chloroplast DNA in plants. Molecular Ecology 4: 129-131[Medline]

DeVerno L. L. P. J. Charest L. Bonen 1993 Inheritance of mitochondrial DNA in the conifer Larix. Theoretical and Applied Genetics 86: 383-388[ISI]

Donnelly P. 1999 The coalescent and microsatellite variability. In D. B. Goldstein and C. Schlötterer [eds.], Microsatellites: evolution and applications, 116–128. Oxford University Press, Oxford, UK

Donnelly P. S. Tavaré 1995 Coalescent and genealogical structure under neutrality. Annual Review of Genetics 29: 401-421[CrossRef][ISI][Medline]

Dumolin-Lapegue S. M. H. Pemonge R. J. Petit 1997 An enlarged set of consensus primers for the study of organelle DNA in plants. Molecular Ecology 6: 393-397[CrossRef][Medline]

Dvornyk V. A. Sirvio M. Mikkonen O. Savolainen 2002 Low nucleotide diversity at the pal1 locus in the widely distributed Pinus sylvestris. Molecular Biology and Evolution 19: 179-188[Abstract/Free Full Text]

Dylis N. W. 1961 Larch of Eastern Siberia and Far East. Academy of Sciences of USSR, Moscow, Soviet Union (in Russian)

Fay J. C. C. I. Wu 1999 A human population bottleneck can account for the discordance between patterns of mitochondrial versus nuclear DNA variation. Molecular Biology and Evolution 16: 1003-1005[ISI][Medline]

Giannasi N. R. S. Thorpe A. Malhotra 2001 The use of amplified fragment length polymorphism in determining species trees at fine taxonomic levels: analysis of a medically important snake, Trimeresurus albolabris. Molecular Ecology 10: 419-426[CrossRef][Medline]

Guo F. L. S. Y. Hu S. X. Xu Z. F. Yuan 2000 Cytological mechanism of cytoplasmic inheritance in Pinus tabulaeformis. II. Transmission of male and female organelles during fertilization and proembryo development. Acta Botanica Sinica 42: 341-352[ISI]

Hewitt G. 2000 The genetic legacy of the Quaternary ice ages. Nature 405: 907-913

Horiuchi K., K. Minoura K. Hoshino T. Oda T. Nakamura T. Kawai 2000 Palaeoenvironmental history of Lake Baikal during the last 23000 years. Palaeogeography, Palaeoclimatology, Palaeoecology 157: 95-108[CrossRef]

Hultén E. 1937 Outline of the history of arctic and boreal biota during the Quaternary Period. Bokförlags Aktiebolaget Thule, Stockholm, Sweden

Innan H. R. Terauchi G. Kahl F. Tajima 1999 A method for estimating nucleotide diversity from AFLP data. Genetics 151: 1157-1164[Abstract/Free Full Text]

Kingman J. F. C. 1982 The coalescent. Stochastic Processes and Applications 13: 235-248[CrossRef]

Kruklis M. V. L. I. Milyutin 1977 Larix czekanovskii. Nauka, Moscow, Soviet Union (in Russian)

Laser B. S. Mohr W. Odenbach G. Oettler U. Kuck 1997 Parental and novel copies of the mitochondrial of 25 gene in the hybrid crop-plant Triticale: predominant transcriptional expression of the maternal gene copy. Current Genetics 32: 337-347[CrossRef][ISI][Medline]

Latta R. G. Y. B. Linhart D. Fleck M. Elliot 1998 Direct and indirect estimates of seed versus pollen movement within a population of ponderosa pine. Evolution 52: 61-67[CrossRef][ISI]

Latta R. G. J. B. Mitton 1997 A comparison of population differentiation across four classes of gene marker in limber pine (Pinus flexilis James). Genetics 146: 1153-1163[Abstract]

Latta R. G. J. B. Mitton 1999 Historical separation and present gene flow through a zone of secondary contact in ponderosa pine. Evolution 53: 769-776[CrossRef][ISI]

Marshall H. D. C. Newton K. Ritland 2002 Chloroplast phylogeography and evolution of highly polymorphic microsatellites in lodgepole pine (Pinus contorta). Theoretical and Applied Genetics 104: 367-378[CrossRef][ISI][Medline]

Mitton J. B. B. R. Kreiser R. G. Latta 2000 Glacial refugia of limber pine (Pinus flexilis James) inferred from the population structure of mitochondrial DNA. Molecular Ecology 9: 91-97[CrossRef][Medline]

Neale D. B. N. C. Wheeler R. W. Allard 1986 Paternal inheritance of chloroplast DNA in Douglas-fir. Canadian Journal of Forest Research 16: 1152-1154[CrossRef]

Ostrowska E. M. Muralitharan S. Chandler P. Volker S. Hetherington F. Dunshea 1998 Optimizing conditions for DNA isolation from Pinus radiata. In Vitro Cellular and Developmental Biology—Plant 34: 108-111

Palmer J. D. 1992 Mitochondrial DNA in plant systematics: applications and limitations. In P. S. Soltis, D. E. Soltis, and J. J. Doyle [eds.], Molecular systematics of plants, 36–49. Chapman & Hall, New York, New York, USA

Parducci L. A. E. Szmidt 1999 PCR-RFLP analysis of cpDNA in the genus Abies. Theoretical and Applied Genetics 98: 802-808[CrossRef][ISI]

Petit R. J. R. Bialozyt S. Brewer R. Cheddadi B. Comps 2001 From spatial patterns of genetic diversity to postglacial migration processes in forest trees. In J. Silvertown and J. Antonovics [eds.], Integrating ecology and evolution in a spatial context, 295–318. Blackwell Science, Oxford, UK

Prentice I. C. D. Jolly and BIOME 6000 participants 2000 Mid-Holocene and glacial-maximum vegetation geography of the northern continents and Africa. Journal of Biogeography 27: 507-519[CrossRef][ISI]

Pritchard J. K. M. Stephens P. Donnelly 2000 Inference of population structure using multilocus genotype data. Genetics 155: 945-959[Abstract/Free Full Text]

Provan J. N. Soranzo N. J. Wilson D. B. Goldstein W. Powell 1999 A low mutation rate for chloroplast microsatellites. Genetics 153: 943-947[Abstract/Free Full Text]

Rohlf F. J. 1998 NTSYS-pc: numerical taxonomy and multivariate analysis system. Exeter, New York, New York, USA

Rosenberg N. A. M. Nordborg 2002 Genealogical trees, coalescent theory and the analysis of genetic polymorphisms. Nature Reviews in Genetics 3: 380-390[CrossRef][ISI][Medline]

Sambrook J. E. F. Fritsch T. Maniatis 1989 Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, USA

Savile D. B. O. 1972 Arctic adaptations in plants. Canadian Department of Agriculture Monographs 6: 1-81

Schneider S. D. Roessli L. Excoffier 2000 Arlequin v.2.000. A software for population genetic data analysis. Genetics and Biometry Laboratory, Department of Ecology and Anthropology, University of Geneva, Geneva, Switzerland

Semerikov V. L. M. Lascoux 1999 Genetic relationships among Eurasian and American Larix species based on allozymes. Heredity 83: 62-70

Semerikov V. L. H. Zhang M. Sun M. Lascoux 2003 Differentiation of the Larix genus: conflicting phylogenies from nuclear ITS sequences and chloroplast DNA. Molecular Phylogenetics and Evolution 27: 173-184.[CrossRef][ISI][Medline]

Sharbel T. F. B. Haubold T. Mitchell-Olds 2000 Genetic isolation by distance in Arabidopsis thaliana: biogeography and postglacial colonization of Europe. Molecular Ecology 9: 2109-2118[CrossRef][Medline]

Soltis D. E. P. S. Soltis 1999 Choosing an approach and an appropriate gene for phylogenetic analysis. In D. E. Soltis, P. S. Soltis, and J. J. Doyle [eds.], Molecular systematics of plants II, 1–42. Kluwer Academic, Dordrecht, Netherlands

Sperisen C. U. Buchler F. Gugerli G. Matyas T. Geburek G. G. Vendramin 2001 Tandem repeats in plant mitochondrial genomes: application to the analysis of population differentiation in the conifer Norway spruce. Molecular Ecology 10: 257-263[CrossRef][Medline]

Taberlet P. L. Geilly G. Pautou J. Bouvet 1991 Universal primers for amplification of 3 noncoding regions of chloroplast DNA. Plant Molecular Biology 17: 1105-1109[CrossRef][ISI][Medline]

Taberlet P. L. Fumagalli A. G. Wust-Saucy J. F. Cosson 1998 Comparative phylogeography and postglacial colonization routes in Europe. Molecular Ecology 7: 453-464[CrossRef][Medline]

Tarasov P. E. et al 2000 Last glacial maximum biomes reconstructed from pollen and plant macrofossil data from northern Eurasia. Journal of Biogeography 27: 609-620[CrossRef][ISI]

Tsumura Y. Y. Suyama 1998 Differentiation of mitochondrial DNA polymorphisms in populations of five Japanese Abies species. Evolution 52: 1031-1042[CrossRef][ISI]

Vekemans X., T. Beauwens M. Lemaire I. Roldán-Ruiz 2002 Data from amplified fragment polymorphism (AFLP) markers show indication of size homoplasy and of relationship between degree of homoplasy and fragment size. Molecular Ecology 11: 139-151[CrossRef][Medline]

Vendramin G. G. M. Anzidei A. Madaghiele C. Sperisen G. Bucci 2000 Chloroplast microsatellite analysis reveals the presence of population subdivision in Norway spruce (Picea abies K). Genome 43: 68-78[Medline]

Vendramin G. G. L. Lelli P. Rossi M. Morgante 1996 A set of primers for the amplification of 20 chloroplast microsatellites in Pinacease. Molecular Ecology 5: 595-598[CrossRef][Medline]

Vos P. R. Hogers M. Bleeker M. Reijans T. Van der Lee M. Hornes A. Frijters J. Pot J. Peleman M. Kuiper M. Zabeau 1995 AFLP: a new technique for DNA fingerprinting. Nucleic Acids Research 23: 4407-4414[Abstract/Free Full Text]

Wagner D. B. G. R. Furnier M. A. Saghaimaroof S. M. Williams B. P. Dancick R. W. Allard 1987 Chloroplast DNA polymorphisms in lodgepole and Jack pines and their hybrids. Proceedings of the National Academy of Sciences, USA 84: 2097-2100[Abstract/Free Full Text]

Wilson I. J. D. J. Balding 1998 Genealogical inference from microsatellite data. Genetics 150: 499-510[Abstract/Free Full Text]

Wu J. K. V. Krutovskii S. H. Strauss 1998 Abundant mitochondrial genome diversity, population differentiation and convergent evolution in pines. Genetics 150: 1605-1614[Abstract/Free Full Text]




This article has been cited by other articles:


Home page
 GeneticsHome page
J. P. Jaramillo-Correa and J. Bousquet
Mitochondrial Genome Recombination in the Zone of Contact Between Two Hybridizing Conifers
Genetics, December 1, 2005; 171(4): 1951 - 1962.
[Abstract] [Full Text] [PDF]

Home page
 Genetics