The Drosophila nuclear receptors DHR3 and betaFTZ-F1 control overlapping developmental responses in late embryos

Abstract

Studies of the onset of metamorphosis have identified an ecdysone-triggered transcriptional cascade that consists of the sequential expression of the transcription-factor-encoding genes DHR3, betaFTZ-F1, E74A and E75A. Although the regulatory interactions between these genes have been well characterized by genetic and molecular studies over the past 20 years, their developmental functions have remained more poorly understood. In addition, a transcriptional sequence similar to that observed in prepupae is repeated before each developmental transition in the life cycle, including mid-embryogenesis and the larval molts. Whether the regulatory interactions between DHR3, betaFTZ-F1, E74A and E75A at these earlier stages are similar to those defined at the onset of metamorphosis, however, is unknown. In this study, we turn to embryonic development to address these two issues. We show that mid-embryonic expression of DHR3 and betaFTZ-F1 is part of a 20-hydroxyecdysone (20E)-triggered transcriptional cascade similar to that seen in mid-prepupae, directing maximal expression of E74A and E75A during late embryogenesis. In addition, DHR3 and betaFTZ-F1 exert overlapping developmental functions at the end of embryogenesis. Both genes are required for tracheal air filling, whereas DHR3 is required for ventral nerve cord condensation and betaFTZ-F1 is required for proper maturation of the cuticular denticles. Rescue experiments support these observations, indicating that DHR3 has essential functions independent from those of betaFTZ-F1. DHR3 and betaFTZ-F1 also contribute to overlapping transcriptional responses during embryogenesis. Taken together, these studies define the lethal phenotypes of DHR3 and betaFTZ-F1 mutants, and provide evidence for functional bifurcation in the 20E-responsive transcriptional cascade.

Fig. 1.

A stereotypic transcriptional cascade occurs during metamorphosis and embryogenesis. (A) A schematic representation of the 20E-triggered regulatory interactions at the onset of Drosophila metamorphosis is depicted, adapted from Thummel (Thummel, 2001). The green ovals represent the EcR/USP 20E receptor heterodimer, blue boxes represent genes that encode 20E-regulated transcription factors and orange boxes represent secondary-response target genes. Green arrows represent inductive effects and red lines represent repressive effects. (B) Temporal profiles of transcription factor gene expression during embryogenesis. Total RNA from staged w¹¹¹⁸ embryos was analyzed by northern blot hybridization to detect expression of the 20E-regulated genes DHR3, βFTZ-F1, E75A and E74A. rp49 was used as a control for loading and transfer.

Fig. 2.

Regulatory interactions in the 20E-triggered embryonic transcriptional cascade. (A,B) Stage 14 control (dib^F8/TM3) and dib^F8 mutant embryos were immunostained with anti-DHR3 antibodies. Scale bars: 100 μm. (C) Total RNA isolated from staged control embryos, DHR3 mutants and βFTZ-F1 mutants was analyzed by northern blot hybridization to detect expression of the 20E-regulated genes βFTZ-F1, E75A and E74A. rp49 was used as a control for loading and transfer. Most control animals have hatched by 20-22 hours AEL, whereas mutant embryos can still be collected.

Fig. 3.

Ectopic βFTZ-F1 expression fails to rescue the lethality of DHR3 mutants. (A) FTZ-F1^ex7 and DHR3^G60S mutant embryos were heat treated (+HS) or not heat treated (−HS), in either the absence or presence of a hs-βFTZ-F1 transgene, and scored for hatching into viable first-instar larvae. The transgene had no effect on DHR3 mutants, but rescued βFTZ-F1 mutants in a heat-dependent manner. (B) DHR3^G60S mutant embryos were heat treated (+HS) or not heat treated (−HS), in either the absence or presence of a hs-DHR3 transgene, and scored for hatching into viable first-instar larvae. DHR3 mutants are rescued by the transgene, independent of its heat-induced expression. Error bars represent s.e.m., n≥3 independent experiments, N≥150 embryos per condition. *P<0.05; N.S., not significant when compared with corresponding mutant embryos, Student's t-test.

Fig. 4.

βFTZ-F1 mutants have abnormally small and unpigmented denticles. (A-C) Dark-field images of cuticle preparations of control, DHR3 mutant and βFTZ-F1 mutant embryos. The overall morphology of DHR3 and βFTZ-F1 mutant cuticles appears normal. Scale bars: 100 μm. (D-F) DIC images of the A3 denticle belt of embryos of the same genotype. βFTZ-F1 mutants have small and unpigmented denticles (F) relative to the control (D), whereas DHR3 mutant denticles appear normal (E). Scale bars: 10 μm.

Fig. 5.

DHR3 and βFTZ-F1 mutants fail to air fill their trachea. (A-C) Bright-field images of late-stage control, DHR3 mutant and βFTZ-F1 mutant embryos. Control embryos have completed liquid clearance and gas filling of their tracheal branches at 19 hours AEL (A), whereas DHR3 and βFTZ-F1 mutant embryos 24-28 hours AEL are defective for this process (B,C). White arrowheads indicate dorsal tracheal trunks and yellow arrowheads indicate Malpighian tubules. In B, the red arrowhead indicates a gap between the cuticle and epidermis as a result of muscle movement. Scale bars: 100 μm. (D-F) Staining of stage 16 embryos with rhodamine-conjugated CBP shows normal patterning, expansion and chitin accumulation in the tracheal systems of DHR3 and βFTZ-F1 mutant embryos. Scale bars: 100 μm. (G-L) At higher magnification, CBP staining (red) coincides with the lumen of the dorsal tracheal trunk and lateral branches marked with breathless>GFP (green) in both control and βFTZ-F1 mutant embryos. Scale bars: 10 μm.

Fig. 6.

DHR3 mutants display defective VNC condensation. (A) Late stage 17 control embryo carrying an elav-GAL4 transgene driving GFP expression, showing normal VNC condensation. The arrows indicate the measurements that were used to calculate VNC condensation in the anteroposterior axis (Olofsson and Page, 2005). (B) DHR3 mutant embryo showing an uncondensed VNC. The arrowhead marks where the VNC would terminate in a wild-type embryo (43% embryo length). Scale bars: 100 μm. (C) Quantification of VNC condensation phenotypes. DHR3 mutants show a bimodal distribution, with 25% (DHR3^G60S/Df) and 11% (DHR3^22-35/Df) of the embryos exhibiting an uncondensed VNC. In contrast, VNC condensation is essentially normal in βFTZ-F1 mutant embryos. Penetrance of mutant phenotypes was assayed by determining the percentage of embryos in which the VNC occupies ≥53% of the embryo length (which corresponds to the control average plus10%) (N≥28 individuals per genotype). ***P<0.001; N.S., not significant when compared to control, chi-square test.

Fig. 7.

Overlapping patterns of gene regulation in DHR3 and βFTZ-F1 mutants. Total RNA isolated from control, DHR3^G60S/Df, DHR3^22-35/Df and FTZ-F1^ex7 mutant embryos 16-20 hours AEL was analyzed by northern blot hybridization to detect retn, E93, kkv, E74A, E75A and Idgf5 expression. Blots were hybridized with rp49 as a control for loading and transfer.