Functional genomics of diapause in two temperate butterflies

Natural selection will act on a given phenotype to maximize fitness in a particular environment, even if this would result in reduced fitness in other environments. In insects some of the strongest selection pressures act on timing life cycles to seasonal variation in environmental conditions, in order to maximize growth, reproduction, and to anticipate the onset of winter. Many temperate insects survive winter by entering a pre-programmed state of developmental arrest, called diapause. The decision to induce diapause is predominantly based on measuring day length. Populations have adapted to latitudinal variation in photoperiod, thereby synchronizing with local seasonal variation. However, there is no general understanding of the genetic basis for controlling diapause induction, maintenance and termination. In this thesis I aimed to gain a better understanding of the genetic basis underlying variation in the induction decision, as well as to gain insights into gene expression changes during diapause in temperate butterflies. I started by revealing local adaptation in the photoperiodic response of two divergent populations of Pieris napi (Paper I). I found that variation in diapause induction among populations of both P. napi and Pararge aegeria showed strong sex-linked inheritance in inter-population crosses (Paper I and II). The genome-wide variation across populations was relatively low in both species. However, there was strong divergence in genomic regions containing the circadian clock genes timeless and period in P. aegeria, and period, cycle, and clock in P. napi. The genetic variation in these specific regions segregated between diapausing and direct developing individuals of inter-population crosses, showing that allelic variation at few genes with known functions in the circadian clock correlated to variation in diapause induction (Paper II and III). Furthermore, I investigated the transcriptional dynamics in two tissues (head and abdomen) during diapause (Paper IV). Already at the first day of pupal development there are on average 409 differentially expressed genes (DEG) each up and down regulated between the direct development and diapause pathways, and this increases dramatically across these formative stages to an average of 2695. Moreover, gene expression is highly dynamic during diapause, showing more than 2600 DEG’s in the first month of diapause development, but only 20 DEG’s in the third month. Moreover, gene expression is independent of environmental conditions, revealing a pre-programmed transcriptional landscape that is active during the winter. Still, adults emerging from either the direct or diapause pathways do not show large transcriptomic differences, suggesting the adult phenotype is strongly canalized. Thus, by integrating whole-genome scans with targeted genotyping and bulk-segregant analyses in population crosses, I demonstrate that adaptive variation in seasonal life cycle regulation in the two butterflies P. napi and P. aegeria both converge on genes of the circadian clock, suggesting convergent evolution in these distantly related butterflies. Moreover, the diapause program is a dynamic process with a distinct transcriptional profile in comparison to direct development, showing that on a transcriptome level diapause development and direct development are two distinct developmental strategies.At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 3: Manuscript. Paper 4: Manuscript.