Analyses of Gene Diversity in Some Species of Conifers

Genetic variation at 21 to 25 loci in extracts of individual megagametophytes was surveyed in Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco), Sitka spruce (Picea sitchensis [Bong.] Carr.) and lodgepole pine (Pinus contorta ssp. latifolia [Engelm.] Critchfield). The overall mean proportion of polymorphic loci was 61.19 percent, and the overall mean heterozygosity per individual was 15.81 percent. Sitka spruce was on the low side and interior Douglas-fir on the high side of overall mean genetic variation. Distribution of loci relative to frequency of heterozygotes was rather even for heterozygosities between 0.05 and 0.60: however, between 38 to 56 percent of the loci had heterozygosities lower than 0.05. More than 90 percent of the total gene diversity resided within populations. Although subpopulations were differentiated by only 2.6 to 7.9 percent, level of population subdivision was considered significant for the species tested. The overall pattern of genetic differentiation agreed with the expected on the basis of the neutral-mutation theory. Some loci, however, demonstrated conspicuous clinal variation patterns that are not readily compatible with this stochastic model. he recent use of gel electrophoresis in isozyme Tstudies of genic variability in natural populations of conifers (Rudin 1976) has permitted researchers to investigate many basic questions of evolutionary biology. These questions concern levels of heterozygosity within populations, distribution of genic variation within and between local populations, and relative amounts of genetic variation in central, as opposed to, marginal populations. Until recently, most investigators addressing these questions on conifers were restricted to a small sample of loci. A striking feature of the more recent and extensive isozyme surveys, however, has been the demonstration of a high degree of interlocus variation in heterozygosity within populations (O'Malley and others 1979, Yeh and ElKassaby 1979, Yeh and Layton 1979). It is imperative, therefore, that a large sample of loci be surveyed when assessing isozyme variation in conifers. For the past 3 years, the major thrust of our research in British Columbia has concerned several conifers of commercial importance. Up to 30 loci were identified and used as genetic markers in population surveys to quantify the amount and organization of genetic variation in coastal and interior Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco), Sitka spruce (Picea sitchensis [Bong.] Carr.), and lodgepole pine (Pinus contorta ssp. latifolia [Engelm.] Critchfield). Our research has not been restricted to the theoretical issues of evolutionary relationships between populations, but has focused on problems that complePresented at the Symposium on Isozymes of North American Forest Trees and Forest Insects, July 27, 1979, Berkeley, Calif. Technical Advisor in Genetics, Research Branch, British Columbia Ministry of Forests, Victoria, B.C., Canada. Yeh, F. C., and D. M. O'Malley. Enzyme variations in natural populations of Douglas-fir (Pseudotsuga menziesii [Mirb] Franco.)from British Columbia. I. Genetic variation patterns in coastal populations. Silvae Genetica (in press). 48 ment our applied tree-breeding program. This latter aspect includes defining subpopulations, delineating seed transfer rules, and investigating associations between allozyme frequencies and quantitative traits for indirect selection. This paper summarizes results of our study on the amount and organization of genetic variation in Douglasfir, Sitka spruce, and lodgepole pine. GENETIC VARIATION IN SEVERAL SPECIES OF CONIFERS Nineteen enzymes were surveyed by one of five buffer systems (table 1). Data were collected in our laboratory on the basis of electrophoretic surveys of protein extracts from individual megagametophytes (table 2). These data show the proportion of loci polymorphic, defined as the proportion of loci in which the most common allele does not exceed a frequency of 0.99, and the proportion of loci at which an individual can be expected to be heterozygous (Nei 1973). Because the criterion defining polymorphic loci is somewhat arbitrary and has a high variance as a result of the relatively small number of loci surveyed, the heterozygosity per individual is the more informative figure. For the species shown (table 2), from 51 to 59 percent of the loci were segregating within a population, and the heterozygosity per individual fell in a relatively narrow range for all species, between 14.67 and 17.47 percent. The overall mean proportion of polymorphic loci was 61.19 percent and the overall mean heterozygosity per individual was 15.81 percent. Values for Sitka spruce were on the low side, and those for interior Douglas-fir were on the high side of the overall mean genetic variation. It is not incorrect, therefore, to characterize these conifers as being polymorphic for 60 percent of their genes, and individuals within the species as being heterozygous for about 16 percent of their loci, It must be emphasized that these surveys were concerned only with alleles that are detected by using conventional starch gel techniques. Only a proportion—probably no more than 33 percent of all possible mutant alleles that produce structural changes in enzyme proteins—are detected in conventional electro­ phoretic surveys. The true level of gene diversity in these conifers, therefore, is likely to be greater than these estimates suggest. Probably many isozyme differences occur that are not detected by the procedures currently used in our laboratory (Coyne 1976). The rich gene pool in conifers is not surprising because they are exceedingly variable in morphology, both across their native range and from tree-to-tree within stands. This high level of gene diversity probably results from a number of variables. Most conifer species grow in large continuous stands over wide geographic ranges. Divergent selection for macrogeographical adaptation (Allard and others 1972), balancing selection for microgeographical differentiation (Hamrick and Allard 1972, Milton and others 1977), combined with an open breeding system that facilitates gene flow within and between subpopulations, tend to maintain a rich gene pool. Heterosis would promote further the maintenance of genetic variation. The notable exception to this pattern is red pine (Pinus resinosa Ait.) (Fowler and Morris 1977). The lack of genetic variation within this species, however, has been hypothesized as being the result of a severe bottleneck, probably during the Pleistocene, when red pine was reduced to a small refugial population. Estimates of mean heterozygosity (table 2) are comparable to that of 10 percent obtained by outbreeding organisms (Nei 1975) but are considerably lower than those reported previously for coastal Douglas-fir (Yang and others 1977) and Norway spruce (Lundkvist and Rudin 1977). This difference results from the bias introduced by the small sample of loci analysed in the above two studies. The standard errors of average heterozygosity empha­ size the importance of examining a large number of loci Table 2—Survey of genic heterozygosity in some species of conifers (table 2). These standard errors are calculated from the variance among loci after the heterozygosities are averaged for populations within a species. They average about 25 percent as large as the mean heterozygosities. Such large Table 1 —Enzymes assayed and procedures used in the survey of genic heterozygosity in some species of conifers