The Canadian side of the Pacific Northwest was almost entirely covered by ice during the last glacial maximum, which has induced vicariance and genetic population structure for several plant and animal taxa. Lodgepole pine (Pinus contorta Dougl. ex. Loud.) has a wide latitudinal and longitudinal distribution in the Pacific Northwest. Our main objective was to identify relictual signatures of glacial vicariance in the population structure of the species and search for evidence of distinct glacial refugia in the Pacific Northwest. A maternally inherited mitochondrial DNA minisatellite-like marker was used to decipher haplotype diversity in 91 populations of lodgepole pine located across the natural range. Overall population differentiation was sizeable (G(ST) = 0.365 and R(ST) = 0.568). Four relatively homogeneous groups of populations, possibly representative of as many genetically distinct glacial populations, were identified for the two main subspecies, ssp. latifolia and ssp. contorta. For ssp. contorta, one glacial lineage is suggested to have been located at high latitudes and possibly off the coast of mainland British Columbia (BC), while the other is considered to have been located south of the ice sheet along the Pacific coast. For ssp. latifolia, two genetically distinct glacial populations probably occurred south of the ice sheet: in the area bounded by the Cascades and Rocky Mountains ranges, and on the eastern side of the Rockies. A possible fifth refugium located in the Yukon may have also been present for ssp. latifolia. Zones of contact between these ancestral lineages were also apparent in interior and northern BC. These results indicate the role of the Queen Charlotte Islands and the Alexander Archipelago as a refugial zone for some Pacific Northwest species and the vicariant role played by the Cascades and the American Rocky Mountains during glaciation.

This article described the population structure of trembling aspen (Populus tremuloides Michx.) in Alberta, a dioecious tree with continuous and wide distribution, and with a primary mode of reproduction through suckering. We studied random amplified polymorphic DNA (RAPD) variation in 249 trees from eight natural populations. Trees within a population were a minimum of 200 m apart to decrease the risk of sampling ramets of a single ortet. Of a total of 28 amplified RAPD products (bands) from five random oligonucleotide primers, the frequencies at seven (25%) were heterogeneous across populations and the percentage of polymorphism averaged 90.2 per population. Estimates of Shannon's phenotypic diversity index ranged between 0.58 and 0.69 among populations, averaging 0.65. There were 246 multiband phenotypes among the 249 trees; three were each shared by two trees from different populations and the remaining 243 were unique. Thus, trees within populations probably were different clones. Analysis of molecular variance partitioned the RAPD variation into the among- and within-population components. The within-population component accounted for 97.4% of the variation and was significantly different from zero at the 2% level of probability. The among-population component, although accounting for only 2.6% of the variation, was significantly different from zero at the 1% level of probability. Pairwise tests for the homogeneity of the RAPD variance between populations suggested significant divergences among 18 of the 28 (64%) population pairs.

A survey was conducted of genetic variation at 25 loci in extracts of individual megagametophytes of lodgepole pine. Collections were made in nine widely separated localities representing four marginal, two intermediate and three central populations. Single populations of lodgepole pine were, on the average, polymorphic at 58.67% of their loci, and had 1.90 alleles per locus. Both expected and observed heterozygosity averaged 0.16. There was a definite trend towards decreased genetic variability at the margins. The measures of gene diversity for the 25 loci showed a 4% but significant effect of interpopulation differentiation; 96% of the total gene diversity resided within populations. Estimated outcrossing rates [Formula: see text] for the nine populations ranged between 0.92–1.29. Comparisons among populations with different levels of outcrossing revealed no clear relationship between [Formula: see text] and amount of genetic variability. The overall pattern of genetic differentiation agrees with expectations based on the neutral mutation theory. However, two loci demonstrated conspicuous clinal variation patterns which may be incompatible with this stochastic model.

We have characterized poplar aquaporins (AQPs) to investigate their possible functions in differential drought responses of Populus balsamifera and Populus simonii×balsamifera leaves. Plants were exposed to mild and severe levels of drought stress and to drought stress recovery treatment, and their responses were compared with well-watered controls. Compared with P. balsamifera, P. simonii×balsamifera used drought avoidance as the main drought resistance strategy, and rapidly reduced stomatal conductance in response to stress. This strategy is correlated with growth rate reductions. Eleven AQPs were transcriptionally profiled in leaves from these experiments and five were functionally characterized for water channel activity. PIP1;3 and PIP2;5 were among the most highly expressed leaf AQPs that were responsive to drought. Expression of PIP1;3 and five other AQPs increased in response to drought in the leaves of P. simonii×balsamifera but not in P. balsamifera, suggesting a possible role of these AQPs in water redistribution in the leaf tissues. PIP2;5 was upregulated in P. balsamifera, but not in P. simonii×balsamifera, suggesting that this AQP supports the transpiration-driven water flow. Functional characterization of five drought-responsive plasma membrane intrinsic proteins (PIPs) demonstrated that three PIP2 AQPs (PIP2;2, PIP2;5, PIP2;7) functioned as water transporters in Xenopus laevis oocytes, while the two PIP1 AQPs (PIP1;2 and PIP1;3) did not, consistent with the notion that they may be functional only as heterotetramers.

not available

Jack pine (Pinus banksiana) and lodgepole pine (Pinus contorta var. latifolia) are two North American boreal hard pines that hybridize in their zone of contact in western Canada. The main objective of this study was to characterize their patterns of introgression resulting from past and recent gene flow, using cytoplasmic markers having maternal or paternal inheritance. Mitochondrial DNA (mtDNA) and chloroplast DNA (cpDNA) diversity was assessed in allopatric populations of each species and in stands from the current zone of contact containing morphological hybrids. Cluster analyses were used to identify genetic discontinuities among groups of populations. A canonical analysis was also conducted to detect putative associations among cytoplasmic DNA variation, tree morphology, and site ecological features. MtDNA introgression was extensive and asymmetric: it was detected in P. banksiana populations from the hybrid zone and from allopatric areas, but not in P. contorta populations. Very weak cpDNA introgression was observed, and only in P. banksiana populations. The mtDNA introgression pattern indicated that central Canada was first colonized by migrants from a P. contorta glacial population located west of the Rocky Mountains, before being replaced by P. banksiana migrating westward during the Holocene. In contrast, extensive pollen gene flow would have erased the cpDNA traces of this ancient presence of P. contorta. Additional evidence for this process was provided by the results of canonical analysis, which indicated that the current cpDNA background of trees reflected recent pollen gene flow from the surrounding dominant species rather than historical events that took place during the postglacial colonization.

We studied isozyme variation at 21 loci in 66 populations from three subspecies of Pinus contorta Dougl.; 35 in spp. latifolia, 20 in spp contorta and 11 in spp. murrayana. The objectives were to assess gametic disequilibria and multilocus structure. There was considerable differentiation of allele frequencies at 19 polymorphic loci across the 66 populations and within the subspecies. Allele frequencies at many loci correlated with geographic variables. Genetic variability varied considerably among populations within subspecies but the subspecies means were similar. The mean number of polymorphic loci and the mean heterozygosity over 19 polymorphic loci were, respectively, 13 and 0.194 in latifolia, 12 and 0.196 in murrayana, and 12 and 0.180 in contorta. The mean heterozygosity correlated with longitude and altitude across the 66 populations and with latitude in latifolia. Gametic disequilibria were evident in 40 populations; 29 in latifolia, eight in murrayana and three in contorta. Gametic disequilibria correlated with latitude across the 66 populations and with longitude in latifolia. The single-locus F ST averaged 0.0339 in latifolia, 0.0567 in murrayana, and 0.0764 in contorta. The multilocus F STM was 0.1227 in latifolia, 0.2926 in murrayana, and 0.3328 in contorta. Multilocus Wahlund and founder effects, migration patterns, and natural selection, probably played significant roles in generating and maintaining the multilocus genetic structure in P. contorta in general and the subspecies latifolia in particular.

Genetic variation at 24 enzyme loci was studied for 10 IUFRO populations of Sitka spruce (Piceasitchensis (Bong.) Carr.). The average proportion of polymorphic loci per population was 0.51 ± 0.02 and the average proportion of heterozygous loci per individual was 0.15 ± 0.03. The following generalizations were advanced: (1) the amount of genetic polymorphism varied considerably from locus to locus; (2) many populations were similar in the amount and pattern of genetic variation for most loci; (3) but some loci had large differences among populations.Eight percent of the gene diversity was attributed to interpopulation differentiation; 92% of the total gene diversity resided within populations. Pairwise comparisons of genetic distances averaged 0.014, a value which reflected very little differentiation between geographic sources.