Research conducted: With the ongoing global warming, there are serious concerns about the persistence of locally adapted populations. Indeed, with the raising of temperature, the phenological cycle of tree species may be strongly affected since higher winter temperatures may have a negative impact on endodormancy release if chilling requirements are not fulfilled during winter and late frost in spring may expose trees if buds flush too early. Thus, Environmental gradients (showing continuous variations of environmental conditions) constitute a design of choice to analyze the effect of winter dormancy in locally adapted population. Methods: In the present study, we used an elevation gradient in the Pyrenees to explore the gene expression network involved in dormancy regulation in natural populations of sessile oak locally adapted to temperature. Terminal buds were harvested during dormancy induction and release at different elevations. Then, gene expression was quantified using RNAseq and we used a likelihood ratio test to identify genes displaying significant dormancy, elevation or dormancy-by-elevation interaction effects. Key results: Our results highlight molecular processes in locally adapted populations along this elevation cline, and made it possible to identify key dormancy-by-elevation responsive genes revealing that locally adapted populations have evolved distinct molecular strategies to adapt their bud phenology in response to environmental variation (i.e. temperature).