Figure 1
Dynamics and manipulation of internal levels of ecdysone. (A) Experimental design for hormone manipulations. Hydroxyecdysone (20E) injections were done on female pupae reared at 19°C, 23°C or 27°C at two developmental stages corresponding to different phases of the natural 20E dynamics (compare with [21]): ‘early’, before ecdysone concentration starts to increase (at 3% of the total time it takes to complete pupal development at each of the temperatures), and ‘late’, corresponding to the ascending phase of the ecdysone level (at 16% of the total pupal development time). (B) Effect of early hormone injections on hormone titers. Internal levels of 20E at 3.5% and 8.5% of total pupal development time after early injection of hormone and control solutions at 19°C and 27°C. The bar represents the median value of four individuals per treatment, temperature, and time point (see Material and methods). We tested for the effect of temperature and injection treatment on the levels of 20E at two time points using the model 20E ~ time point + temperature * injection, for which the residuals showed no significant departure from normality (Shapiro-Wilk W test: W = 0.950, P = 0.146) or from homogeneity of variances (Fligner-Killeen test: Median Chi Square = 1.176, df = 1, P = 0.185). The analysis of variance revealed a statistically significant effect of temperature (F (1,32) = 13.848, P = 0.0009) and injection (F (1,32) = 114.501, P = 3.25e-11), but not of time point (F (1,32) = 0.026, P = 0.874) or temperature*injection (F (1,32) = 3.670, P = 0.066). Results of the post-hoc pairwise comparisons using the Tukey honest significance test are indicated in the figure: ns for P >0.01, ** for P <0.01, *** for P <0.001 [see Additional file 4 for more on this analysis].