How conifers adapt to the cold.

I n convergent evolution, different species independently acquire similar traits. Knowing whether divergent species use the same or different genetic solutions to adapt to similar selection pressures can illuminate innate constraints on underlying molecular networks. Most cases of convergence that are understood at the genetic level are relatively simple, involving one or a few genes. On page 1431 of this issue, Yeaman et al. (1) report mapping of locally adaptive genes in lodgepole pine (Pinus contorta) and interior spruce (the species complex that includes Picea engelmannii and Picea glauca) (see the photo). The authors show that many genes implicated in local adaptation are shared among the two species. Many examples of convergent evolution at the morphological level have been documented, and in some cases the genetic basis for convergence is known. Examples include loss of pigmentation in mice, lizards, and humans via mutations in melanocortin-1 receptor (MC1R) (2), adjustment of reproductive timing in plants via flowering locus C (FLC) and its homologs (3, 4), loss of trichomes in Drosophila via shavenbaby (5), and resistance to toxic cardenalides in insects (6, 7). However, these well-studied cases tend to involve genetically simple traits, in which one or two genes are credited with the repeated phenotypic changes (8). In these cases, pathway constraints have been invoked to explain cases where different species have adapted via changes in the same genes (9). The idea is that a pathway can only be perturbed in a limited number of ways; thus, the possible genetic modifications that can occur to enable adaptation are also limited. However, many adaptive traits are likely to be genetically complex, involving many small-effect variants (10). In these cases, adaptation could act in more varied ways, making extensive overlap in the genes underlying adaptive responses across diverged species improbable. Plant species are particularly powerful models for examining local adaptation, because they cannot hide from environmental insults and must face them head-on. Yeaman et al. sequenced the protein-coding genome regions of more than 250 populations from two diverged conifer species, lodgepole pine and interior spruce. For each individual, they collected climatic data from the site of origin as well as phenotypic data, which were measured under simulated natural conditions. They then identified regions of the genome that were correlated with variation in phenotypes or climate to pinpoint loci implicated in local adaptation. In both species, the majority of traits and climatic factors with the strongest signals underlie the ability to deal with cold stress. Given the observed convergence in cold tolerance adaptation, the authors next explored whether the associated loci were shared or different between the two conifer species. They expected the traits and the adaptive signals examined to involve many genes and overlap of the loci identified in the two species to be limited. Yet they found extensive overlap, with 47 genes showing signals in both species (see the figure). Moreover, many shared signals were in genes that were duplicated in one or both species. This is consistent with the proposition that duplicated genes may be more likely than nonduplicated genes to be used in adaptive bouts because of reduced constraint on function. The reasoning is EVOLUTION