Genes of the ecdysone biosynthesis pathway are regulated by the dATAC histone acetyltransferase complex in Drosophila

Abstract

Uncovering mechanisms that regulate ecdysone production is an important step toward understanding the regulation of insect metamorphosis and processes in steroid-related pathologies. We report here the transcriptome analysis of Drosophila melanogaster dAda2a and dAda3 mutants, in which subunits of the ATAC acetyltransferase complex are affected. In agreement with the fact that these mutations lead to lethality at the start of metamorphosis, both the ecdysone levels and the ecdysone receptor binding to polytene chromosomes are reduced in these flies. The cytochrome genes (spookier, phantom, disembodied, and shadow) involved in steroid conversion in the ring gland are downregulated, while the gene shade, which is involved in converting ecdysone into its active form in the periphery, is upregulated in these dATAC subunit mutants. Moreover, driven expression of dAda3 at the site of ecdysone synthesis partially rescues dAda3 mutants. Mutants of dAda2b, a subunit of the dSAGA histone acetyltransferase complex, do not share phenotype characteristics and RNA profile alterations with dAda2a mutants, indicating that the ecdysone biosynthesis genes are regulated by dATAC, but not by dSAGA. Thus, we provide one of the first examples of the coordinated regulation of a functionally linked set of genes by the metazoan-specific ATAC complex.

FIG. 1.

dATAC mutants have reduced ecdysteroid levels. (A) Ecdysteroid titers measured for different mutant genotypes (dAda3, dAda2a, and dAda2b) and their corresponding sex-matched sibling controls (FM7i, and TM6 and Tb) at 112 h AEL, as determined by ELISA. The values are expressed as the means of 20E equivalents per mg of larvae. The error bars indicate the standard errors of the mean (SEM) (n = 3 samples of 15 larvae each). The asterisks indicate statistically significant differences at P ≤ 0.006 (t test). (B) EcR immunostaining of polytene chromosomes from dAda3² and control larvae. Note the reduction of EcR signal in the mutant (similarly reduced EcR binding to chromosomes can be observed in dAda2a mutants [7]).

FIG. 2.

Transcriptional changes in dADA mutants. (A) Venn diagrams showing the numbers of activated or repressed genes in dAda2b, dAda3, or dAda2a mutant larva. (B) Distribution of genes with different expression patterns compared to the control in the absence of dAda2a and dAda3. The genes were categorized based on the level of changes (in log₂ scale) in their expression.

FIG. 3.

Gene expression changes in ATAC mutants show tight coordination. The scatter plots show gene expression changes in dAda2a- and dAda3-null animals as detected by microarray hybridization. The mRNA levels (log₂) of genes involved in cuticle formation (A), regulated by ecdysone (B), and encoding cytochrome enzymes (C) and proteasome subunits (D) are plotted.

FIG. 4.

mRNA levels of ecdysone (ecd)- related genes in dAda3 mutants. The mRNA levels were determined by QRT-PCR using TaqMan probes. The results of three independent experiments are shown as ratios between mutant (dAda3²) and control (w¹¹¹⁸). In addition to the EcR isoforms and USP (A), representative early ecdysone-induced genes (Eig) (B) were included in these sets of assays. The error bars indicate the SEM.

FIG. 5.

Expression of Halloween genes in dATAC mutants. The mRNA levels were determined by QRT-PCR in three independent experiments and are shown as ratios between mutant and control (w¹¹¹⁸). (A) The four Halloween genes show consistent downregulation in dAda3² and dAda2a¹⁸⁹, in contrast to dAda2b⁸⁴², mutants. (B) The gene responsible for the E-to-20E conversion, shade, exhibits effects opposite to those of the genes in panel A. (C) The expression levels of two relatively uncharacterized genes of the ecdysone-synthesizing pathway, molting defective and neverland, in dATAC mutants are similar to those of the prothoracic gland-specific Halloween genes. The error bars indicate the SEM.

FIG. 6.

Effects of 20E treatment on gene expression and pupariation in dATAC mutants. Dissected salivary glands of dAda2a¹⁸⁹, dAda3², or heterozygous control L3 larvae were separated, and the two parts were incubated with 20E or vehicle control, respectively. The transcript levels of the ecdysone response genes Eig74 (A) and Eig75 (B) were measured by QRT-PCR, and the average and standard error of the induction in 20E- versus mock-treated matched samples were plotted. At least four matched samples were measured per genotype. (C) The chart shows the ratio of dAda2a¹⁸⁹, dAda3², or heterozygous control larvae in which pupariation was initiated after feeding on 20E or vehicle control containing food in mid-L3 stage. The averages and standard errors of four feeding experiments with a sum of 30 larvae in each category are shown.

FIG. 7.

dADA3 expression rescues the mutant when driven in the ring gland. (A) Expression of the ADA3 transgene in the phantom domain at the larval stage visualized with GFP. Note the selective expression in the ring gland. (B) Lethality phase of dAda3 mutants. The dAda3² allele fails to pupariate or forms abnormal pupae. (C) Normal pupae at the P4 and P5 stages shown for comparison. (D) Rescue of dAda3² mutants by driving transgene expression in the phantom domain. (E) Phenotype quantifications. The fractions of animals that perished in the indicated developmental stages are shown. The error bars indicate the SEM.