Transcriptional regulation of xenobiotic detoxification in Drosophila

Abstract

Living organisms, from bacteria to humans, display a coordinated transcriptional response to xenobiotic exposure, inducing enzymes and transporters that facilitate detoxification. Several transcription factors have been identified in vertebrates that contribute to this regulatory response. In contrast, little is known about this pathway in insects. Here we show that the Drosophila Nrf2 (NF-E2-related factor 2) ortholog CncC (cap 'n' collar isoform-C) is a central regulator of xenobiotic detoxification responses. A binding site for CncC and its heterodimer partner Maf (muscle aponeurosis fibromatosis) is sufficient and necessary for robust transcriptional responses to three xenobiotic compounds: phenobarbital (PB), chlorpromazine, and caffeine. Genetic manipulations that alter the levels of CncC or its negative regulator, Keap1 (Kelch-like ECH-associated protein 1), lead to predictable changes in xenobiotic-inducible gene expression. Transcriptional profiling studies reveal that more than half of the genes regulated by PB are also controlled by CncC. Consistent with these effects on detoxification gene expression, activation of the CncC/Keap1 pathway in Drosophila is sufficient to confer resistance to the lethal effects of the pesticide malathion. These studies establish a molecular mechanism for the regulation of xenobiotic detoxification in Drosophila and have implications for controlling insect populations and the spread of insect-borne human diseases.

Figure 1.

Xenobiotics induce a coordinated transcriptional response in Drosophila. (A) Wild-type (CanS) flies were treated with either no PB (−) or different concentrations of PB, as indicated, for 6 h, after which RNA was extracted and analyzed by Northern blot hybridization to detect the transcription of PB-inducible genes, as shown. (B) Wild-type flies were treated with either no PB (−) or 0.3% PB for the indicated time periods, after which RNA was extracted and analyzed by Northern blot hybridization to detect the transcription of PB-inducible genes. (C) Adult CanS flies were treated with no PB (−) or either 0.3% PB, 0.3% chlorpromazine (CP), or 1.5 mg/mL caffeine (Caff), as indicated (+), for 6 h, after which RNA was extracted and analyzed by Northern blot hybridization to detect the transcription of PB-inducible genes. Hybridization to detect rp49 mRNA was used as a control for loading and transfer in all panels.

Figure 2.

A 15-bp element in the Cyp6a2 promoter is necessary for transcription. (A) Adult CanS flies or Met¹-null mutants were treated with either no PB (−) or PB for the indicated time periods, after which RNA was extracted and analyzed by Northern blot hybridization to detect the transcription of PB-inducible genes. Although Cyp6a2 does not appear to be expressed in Met¹ mutants, full-length mRNA is detectable upon longer exposure. (B) The Met¹ flies were crossed to a stock carrying a second chromosome CyO balancer that carries a wild-type Cyp6a2 locus (left panel), or a stock carrying the Df(2R)1612 deficiency that removes the Cyp6a2 locus (right panel), and the progeny from these crosses were treated with either no PB (−) or PB (+) for 4 h, after which RNA was extracted and analyzed by Northern blot hybridization to detect Cyp6a2 mRNA. In all panels, hybridization to detect rp49 mRNA was used as a control for loading and transfer. (C) A schematic representation of the Cyp6a2 5′ region is shown, with the transcribed sequences and ATG codon marked. Shown below is the sequence of a portion of the Cyp6a2 5′ flanking region located 84 bp upstream of the start site of transcription (5′ arrow), which was determined by 5′ RACE. The 15-bp sequence that is deleted in the Met¹ mutants and the canonical Nrf2/Maf-binding site are marked by bracketed lines above and below the sequence, respectively.

Figure 3.

The Nrf2/Maf-binding site in the Cyp6a2 promoter is necessary and sufficient for its xenobiotic-induced transcription. (A) Transgenic animals carrying a lacZ reporter gene fused to either a wild-type 313-bp Cyp6a2 promoter fragment (WT-lacZ) or the same promoter fragment carrying the 15-bp deletion identified in the Met¹ mutants (Δ15-lacZ) were fed with either no drug (−) or PB, chlorpromazine (CP), or caffeine (Caff) for 6 h (+), after which RNA was extracted and analyzed by Northern blot hybridization to detect lacZ and Cyp6a2 transcription. (B) Transgenic animals carrying a lacZ reporter gene fused to either five tandem copies of a 25-bp sequence that encompasses the Nrf2/Maf-binding site in the Cyp6a2 promoter (5XWT-lacZ) or a mutant version of this promoter fragment lacking the Nrf2/Maf-binding site (5XMUT-lacZ) were fed with either no PB (−) or PB (+) for 6 h, after which RNA was extracted and analyzed by Northern blot hybridization to detect Cyp6a2 and lacZ transcription. (C) RNA was extracted from animals carrying either the WT-lacZ or the Δ15-lacZ transgenes, following either no heat treatment (−hs) or heat treatment (+hs), in the absence or presence of the hsp70-CncC transgene (hs-CncC). This RNA was analyzed by Northern blot hybridization to detect lacZ and Cyp6a2 transcription. In all panels, hybridization to detect rp49 mRNA was used as a control for loading and transfer. (D) CantonS (CanS) control and Met¹ mutant flies were fed with PB for 6 h, after which they were homogenized and subjected to formaldehyde cross-linking, followed by sonication and ChIP. The presence of Cyp6a2 or rp49 (as a negative control) sequences was assayed by quantitative PCR (qPCR). Results from a representative experiment are shown. The Y-axis depicts the fold enrichment over a mock immunoprecipitation control that lacks Cnc antibody. (*) P < 0.015.

Figure 4.

The CncC/Keap1 pathway is necessary and sufficient for xenobiotic-induced transcription. (A) Flies carrying the Tub-Gal80ts;Tub-GAL4 driver and/or UAS-CncC-RNAi transgene were shifted for 5 d to 29°C, after which they were fed either no PB (−PB) or PB (+PB) for 6 h. RNA extracted from these animals was analyzed by Northern blot hybridization to detect the transcription of PB-inducible genes, as shown. (B) RNA was extracted from either control w¹¹¹⁸ animals (−hs-CncC) or transgenic animals carrying a heat-inducible copy of CncC (+hs-CncC) following either no heat treatment (−hs) or heat treatment (+hs) and analyzed by Northern blot hybridization to detect the transcription of PB-inducible genes. (C) Flies carrying the Tub-Gal80ts;Tub-GAL4 driver and/or UAS-Keap1 transgene were shifted for 4 d to 29°C, after which they were fed either no PB (−PB) or PB (+PB) for 4 h. RNA extracted from these animals was analyzed by Northern blot hybridization to detect the transcription of PB-inducible genes. (D) Flies carrying the Tub-Gal80ts;Tub-GAL4 driver and/or UAS-Keap1-RNAi transgene were shifted for 4 d to 29°C, after which RNA was extracted and analyzed by Northern blot hybridization to detect the transcription of PB-inducible genes. Hybridization to detect rp49 mRNA was used in all panels as a control for loading and transfer.

Figure 5.

Most PB-regulated genes are also controlled by CncC. Venn diagrams are depicted that compare the genes that change their expression in wild-type flies treated with PB with genes that change their expression in response to ectopic CncC expression (A), or the genes that are up-regulated in wild-type flies treated with PB (PB-up) with genes that are up-regulated in response to ectopic CncC expression (CncC-up) (B). The P-value for overlap of the gene sets is shown for each diagram. (C) GOstat analysis of the genes that change expression in response to ectopic CncC expression. The top GO categories for each gene set are listed in order of significance along with the number of genes affected in that category, the total number of genes in that category (in parentheses), and the statistical significance of the match.

Figure 6.

Activation of the CncC/Keap1 pathway confers pesticide resistance. The GAL4/UAS system was used to activate the CncC/Keap1 pathway in different tissues by inducing RNAi against Keap1, after which the animals were tested for resistance to the pesticide malathion. All experiments were done in the absence or presence of a UAS-Keap1-RNAi transgene in combination with different tissue-specific GAL4 drivers: CG-GAL4 and r4-GAL4 fat body drivers, UO-GAL4 and c42-GAL4 principal cell-specific Malpighian tubule drivers, C724-GAL4 stellate cell-specific Malpighian tubule driver, and Mex-GAL4 midgut driver. w¹¹¹⁸ animals were used as a control. The Y-axis represents the number of animals surviving after a 36-h exposure to 10 μM malathion (+). Ten replicates were used per genotype. White bars represent control animals, and gray bars represent animals in which the CncC/Keap1 pathway has been activated.