Genome-wide screening for developental genes in Drosophila reveals important genotype by temperature interction.

MENSCH JULIÁN; LAVAGNINO NICOLÁS; CARREIRA VALERIA; FANARA JUAN JOSÉ

Stony Brook, New York,USA

Congreso; Joint Meeting of the Society for the Study of Evolution, the American Society of Naturalists and the Society of Systematic Biologists.; 2006

Society of Naturalists and the Society of Systematic Bilogists.

Genome-wide screening for developmental genes in Drosophila reveals important genotype by temperature interaction.   Mensch, Julián; Lavagnino, Nicolás; Carreira, Valeria & Fanara, Juan José. Laboratorio de Evolución. Departamento de Ecología, Genética & Evolución. Facultad de Ciencias Exactas y Naturales. Universidad de Buenos Aires. Argentina.   INTRODUCTION   Temperature during ontogenesis is one of the most important environmental keys affecting the evolution of developmental traits. Moreover, in organisms with metamorphosis characterized by stage-specific morphological, physiological and behavioural patterns, the effect of  developmental temperature is undoubtedly express in their plastic genomes. In this context, the time to reach the reproductive age is a quantitative complex trait with adaptive significance known as developmental time (DT). Phenotypic variation of DT results from the segregation of alleles at multiple quantitative trait loci (QTL) with effects that are sensitive to the genetic, sexual and external environments. The aim of this work is to quantify the different phenotypic components of DT variation among co-isogenic single P-element insertional lines screened at different developmental temperatures in order to understand the genetic architecture of developmental time in the model organism Drosophila melanogaster.     MATERIALS AND METHODS   We screened 179 independent co-isogenic single P-elements inserted lines  -p [GT1]- to identify candidate genes affecting DT at 25ºC. A subset of 45 lines selected at random those mentioned above were also screened at 17ºC.  For each line, batches of 30 first instar larvae were seeded in vials containing standard culture medium (4 replicates per line). Emerged flies from each vial were collected every 12 hours and sorted by sex. We estimated DT as the time elapsed since the transfer of first instar larvae to the vials until adult emergence .  Phenotypic differences among P-elements inserted lines and between P-elements inserted lines and a control (a co-isogenic P-element insertion free line) were tested by means of ANOVA and Dunnet test. Those lines that exhibited significant differences relative to  the control were considered as lines carrying a mutation in a candidate  gene affecting DT. In order to identify the mutated genes, nucleotide sequences flanking of each P-element insertion line that showed a mutant phenotype were employed as “in silico” probes to search for homologous regions in the D. melanogaster genome. All experiments were run under a  12:12 light:dark photoperiod and at 60-70% of humidity.     RESULTS & DISCUSSION   Developmental time at 25ºC. We observed a highly significant line effect indicating mutational variance among single P-element inserted lines (Table 1); and significant differences between sexes (on average males developed faster than females). Furthermore, the line by sex interaction was also highly significant (Pline x sex < 0.0001), indicating sex-specific mutational effects on developmental time. We also analyzed whether the significant line by sex interaction was due to differences in among-line variance components between males and females (change in magnitude) or to a cross-sex genetic correlation lower than one (change in rank order). Our results showed that the change in the rank order was the principal component (98,6 %) of the significant interaction suggesting that some mutations have a sex specific effect. This pattern has been previously reported for P-element insertions affecting traits such as olfactory behavior, sensory bristle number and starvation resistance. The line factor and the line by sex interaction account for 62.6% and 1.2% of  total phenotypic variance  respectively (Figure 1A). One hundred and twelve out of 179 (63%) lines differed significantly from the control. However, we were able to identified 85 genes due to methodological limitations (the nucleotide sequences flanking the P-element were too short to accurately identify homologous regions in the entire genome). In sixty-one genes (72%) the loss-of-function mutation showed a bias towards increasing the DT indicating that this trait is optimized for  shorter DT.  However, DT differences relative to the control were quite similar in both sexes. These results suggest that the genetic architecture of DT is highly conserved between sexes.             Developmental time at 17ºC. As it was observed in the 25ºC assay, we found a highly significant line effect (Table 1). The line by sex interaction was also highly significant (Pline x sex < 0.0001), indicating sex-specific mutational effects on developmental time at 17ºC. No differences in developmental time were observed between sexes at this temperature. The line by sex interaction was attributable solely to departures of the cross-genetic correlation from 1.The analysis of variance components reveals a similar pattern than at 25ºC: the line factor and the line by sex interaction accounted for 47.1% and 1.3% of  total phenotypic variance  respectively (Figure 1B).             Developmental time at 17ºC and 25ºC. Both line and temperature terms showed significant effects for developmental time (Table 2). The line by sex and the line by temperature factors were also highly significant, indicating that the line effect was not independent of the environment (both sex and temperature). Both the line factor and the line by temperature interaction accounted for 14.3% and 31.6% of  total phenotypic variance of DT, respectively (Figure 2). We also analyzed whether the significant line by temperature interaction can be explained by changes in variance magnitudes between treatments or to a cross-temperature genetic correlation lower than one (change in rank order). In this sense, our results showed that the change in variance and the change in  rank order represent 45% and 55% of the interaction, respectively, in both sexes, suggesting that half of the interaction variance is explained by temperature-specific mutation effects on DT and the other half is explained by the greater among-line variance at 17ºC.   CONCLUSIONS   The significant line by temperature interaction observed opens an excellent opportunity for the study of the genetics basis of phenotypic plasticity for an adaptive relevant trait as DT. In this context, and just as examples of contrasting patterns, we observed  that the loss-of-function mutation line for CG14478 gene showed significant DT differences relative to the control only at 25ºC in both sexes (no differences were observed at 17ºC). In contrast, the loss-of-function mutation line for Hsp27 gene showed high significant DT differences at 17ºC in both sexes but no differences were observed at 25ºC.  In this sense, our results provides strong evidences for candidate temperature plasticity genes as important components of the genetic architecture of DT.       PERSPECTIVES One of our futures goals is to investigate the patterns of genetic variation in natural populations for those genes that showed the strongest effects. Also, we are interested in the study of the temporal and spatial patterns expression of the genes showing the largest effects.