One of the major challenges in evolutionary biology is to determine the genetic basis of adaptive variation. In Norway spruce (Picea abies) the timing of bud set shows a very strong latitudinal cline despite a very low genetic differentiation between populations.
The timing of bud set in Norway spruce is under strong genetic control and triggered by changes in photoperiod, but no genes controlling this response have so far been described. In this thesis we used a combination of functional studies to identify candidate genes, and analyses of DNA sequence polymorphism to infer demographic history and test candidate genes for signs of selection.
By monitoring gene expression during bud set in two populations with divergent bud set response, more than 3300 genes were differentially expressed. The response was very similar in the two populations and no significant expression differences were found between populations. The majority of genes showed a gradual change in expression over time, but a small group of around 300 genes showed a strong increase in gene expression already after a single long night. Among these we found genes that are similar to photoperiod pathway genes in Arabidopsis and genes that have been assigned to stress- and cold-response in flowering plants. To further investigate the role of photoperiodic related genes in bud set response detailed expression pattern of four genes, PaFT1, PaFT2, PaFT3 and PaFT4, homologous to Arabidopsis FLOWERING LOCUS T (FT) and TERMINAL FLOWER 1 (TFL1) were monitored under more elaborate experimental conditions. PaFT4 showed an expression pattern correlating with bud set under several different light and photoperiod treatments, indicating that it has a role in the induction of bud set in response to changes in photoperiod.
To investigate the role of selection, demography and hybridization in the evolution of spruce species multilocus data sets of DNA sequence data were collected from Norway spruce and three North American spruce species (Picea breweriana, Picea glauca, Picea mariana). In general, few fixed and a large number of shared polymorphic sites were found among the four species. By employing an "Isolation with migration" model to the data, it was clear that most of the shared polymorphisms were retained ancestral variation and evidence of gene flow was only found between Norway spruce and Picea glauca. Despite the large number of shared polymorphisms, the actual level of diversity in Norway spruce was lower compared to expectations from a species with continental wide distribution range. The low diversity was coupled with a skewed allele frequency distribution with far more singletons than expected under neutrality, suggesting that fluctuating population sizes have had a large impact on the diversity observed today. The observed pattern at most genes seems to be consistent with an ancient and sever bottleneck.
To examine the role of selection at three photoperiodic related genes, an approach where the effect on genetic variation due to demographic events was taken into account. One of the genes, PaPRR3 deviated significantly from expectations based on the inferred demographic scenarios and hence might have been affected by selection.
In summary, this thesis shows that, by using several approaches, we might be able to identify genes involved in local adaptation even in non-model systems, like conifers.
Uppsala: Acta Universitatis Upsaliensis , 2009. , 36 p.