Association between circadian clock genes and diapause incidence in Drosophila triauraria

Abstract

Diapause is an adaptive response triggered by seasonal photoperiodicity to overcome unfavorable seasons. The photoperiodic clock is a system that controls seasonal physiological processes, but our knowledge about its physiological mechanisms and genetic architecture remains incomplete. The circadian clock is another system that controls daily rhythmic physiological phenomena. It has been argued that there is a connection between the two clocks. To examine the genetic connection between them, we analyzed the associations of five circadian clock genes (period, timeless, Clock, cycle and cryptochrome) with the occurrence of diapause in Drosophila triauraria, which shows a robust reproductive diapause with clear photoperiodicity. Non-diapause strains found in low latitudes were compared in genetic crosses with the diapause strain, in which the diapause trait is clearly dominant. Single nucleotide polymorphism and deletion analyses of the five circadian clock genes in backcross progeny revealed that allelic differences in timeless and cryptochrome between the strains were additively associated with the differences in the incidence of diapause. This suggests that there is a molecular link between certain circadian clock genes and the occurrence of diapause.

Figure 1. Diapause incidences in the original strains (A), F₁ hybrids (B), and BC progeny (C) of Drosophila triauraria.

LD, long daylength conditions (15 h light:9 h dark); SD, short daylength conditions (10 h light:14 h dark). Under both conditions the temperature was 15°C from egg to adult. Diapause incidences under LD are shown as gray bars, and those under SD are shown as black bars. Crosses from which F₁ and BC females were obtained are shown under each bar. F₁(ONM×OEB12) and F₁(OEB12×ONM) indicate F₁ flies obtained from ONM♀×OEB12♂ and OEB12♀×ONM♂, respectively. Numbers of dissected females are shown in parentheses: within the parentheses the values at left are those under LD conditions and those at right are under SD conditions.

Figure 2. Pairs of ovaries dissected from 16-day-old female Drosophila triauraria reared under short daylength conditions at 15°C.

Top: non-diapause strain (OEB12). Bottom: diapause strain (ONM).

Figure 3. Diapause incidences for each genotype of BC progeny of Drosophila triauraria under short daylength conditions, and their differences between heterozygous and homozygous females.

Red and blue bars show diapause incidences (%) of flies homozygous and heterozygous for the genes shown as “Combinations of genes” in the middle of the figure, respectively. Numbers of dissected flies are shown at the bottom of the figure for homozygous and heterozygous, respectively. Plotted circles show differences in diapause incidences between heterozygous and homozygous flies of each gene set. The differences indicate the strength of the genetic effects of each gene set on the phenotype: a gene set with a strong genetic effect will show a higher difference. Blue, differences involving tim; green, differences involving cry; red, differences involving both tim and cry; white, differences involving neither tim nor cry. Scale for diapause incidence is shown on the left side, and that for difference is on the right side.

Figure 4. Time-series proportions of Drosophila triauraria flies eclosing under 12-h light:12-h dark cycles at 23°C.

Eclosing flies were counted every hour over several days. Counted numbers of flies were pooled and the proportions calculated for each data point. Error bars represent ± SEM. R numbers are Winfree's R values.

Figure 5. Diapause incidences in allelic combinations of two of the five genes of BC females of Drosophila triauraria.

“homo,” homozygous for alleles from the non-diapause strain; “hetero,” heterozygous for those from the diapause and non-diapause strains. In the cases of per, Clk, and cyc (rows 1, 3, and 4), the blue and red lines almost overlap in each graph, indicating that there was no allelic effect on diapause incidence. In the cases of tim and cry (rows 2 and 5), the blue lines are always higher than the red lines in each graph, indicating that there is an effect of allelic differences on diapause incidence; furthermore, the blue and red lines are completely parallel in the combination of tim and cry, indicating that there is no interaction effect between them. These effects were confirmed by the statistical analysis.