The Oxidative Stress Response in Caenorhabditis elegans Requires the GATA Transcription Factor ELT-3 and SKN-1/Nrf2

Abstract

Cellular damage caused by reactive oxygen species is believed to be a major contributor to age-associated diseases. Previously, we characterized the Caenorhabditis elegans Brap2 ortholog (BRAP-2) and found that it is required to prevent larval arrest in response to elevated levels of oxidative stress. Here, we report that C. elegans brap-2 mutants display increased expression of SKN-1-dependent, phase II detoxification enzymes that is dependent on PMK-1 (a p38 MAPK C. elegans ortholog). An RNA-interference screen was conducted using a transcription factor library to identify genes required for increased expression of the SKN-1 target gst-4 in brap-2 mutants. We identified ELT-3, a member of the GATA transcription factor family, as a positive regulator of gst-4p::gfp expression. We found that ELT-3 interacts with SKN-1 to activate gst-4 transcription in vitro and that elt-3 is required for enhanced gst-4 expression in the brap-2(ok1492) mutant in vivo Furthermore, nematodes overexpressing SKN-1 required ELT-3 for life-span extension. Taken together, these results suggest a model where BRAP-2 acts as negative regulator of SKN-1 through inhibition of p38 MAPK activity, and that the GATA transcription factor ELT-3 is required along with SKN-1 for the phase II detoxification response in C. elegans.

Keywords: BRAP-2; ELT-3; GATA transcription factors; SKN-1; oxidative stress.

Figure 1

gst-4 expression levels are higher in brap-2(ok1492) worms. (A) brap-2(ok1492) mutants display increased gst-4p::GFP in hypodermis and the intestine, but the sod-3p::GFP expression is not affected. (B) GFP levels of brap-2(ok1492) worms expressing gst-4p::GFP were ∼2.5 times that of wild type. Quantification was done using ImageJ (n = 18 for each sample). (C) The mRNA levels of 11 phase II detoxification genes (log₂ scale) were compared between wild type and brap-2(ok1492) mutants. A significant increase in mRNA levels was found in 9 of 11 phase II genes in the mutant strain. (D) The mRNA levels of nonphase II stress-related genes were measured by qPCR. Of the seven genes tested, only one showed a significant increase in the brap-2(ok1492) mutant. *** P < 0.001, ** P < 0.01, * P < 0.05.

Figure 2

Enhanced gst-4 expression in brap-2(ok1492) requires SKN-1. (A) Synchronized gst-4p::gfp and brap-2(ok1492);gst-4p::gfp worms were grown in control (L4440) or skn-1 RNAi. brap-2(ok1492) shows decreased gst-4 expression when skn-1 is knocked down in L4 worms. (B) RNA was extracted and the gst-4 mRNA levels (log₂ scale) were quantified using qRT-PCR. The brap-2(ok1492);skn-1(zu67) double mutant showed a reduction of gst-4 mRNA expression. ** P < 0.01.

Figure 3

Increased SKN-1 intestinal nuclear localization and binding to the gst-4 promoter in brap-2(ok1492). (A) Localization of skn-1::gfp in N2 (top) and brap-2(ok1492) (bottom) worms. Synchronized worms were grown on NGM plates and L4 worms were picked to visualize GFP expression using confocal microscopy. The white arrows indicate the locations of intestinal nuclei. Results indicated an accumulation of SKN-1::GFP in intestinal nuclei in brap-2(ok1492) animals. (B) Percentage of SKN-1::GFP nuclear localization in brap-2(ok1492) and N2 worms were categorically scored and quantified as described in Materials and Methods. The percentage of nuclear localization increased 2.6-fold in brap-2 mutant worms as compared to wild type. (C) RNA was isolated from N2 and brap-2(ok1492) worms followed by quantification of skn-1a, skn-1b, and skn-1c mRNA transcript levels using qRT-PCR. Results demonstrated an enhanced skn-1b and skn-1c expression in brap-2(ok1492). *** P < 0.001, ** P < 0.01 vs. N2. (D) Isoform-specific RNAi of skn-1a, skn-1b, and skn-1c isoforms of skn-1 was performed and gst-4 mRNA levels quantified by qRT-PCR. Knockdown of skn-1c showed a significant reduction of gst-4 expression (log₂ scale) in in wild-type and brap-2 mutant animals. *** P < 0.001. (E) ChIP–qPCR analysis of SKN-1 binding at two gst-4 promoter regions in skn-1::gfp::3xFlag- and brap-2(ok1492); skn-1::gfp::3xFlag-expressing worms. Relative percentage input is normalized to N2 (absence of GFP antigen). Data represented the mean of four independent trials.

Figure 4

PMK-1 is activated in brap-2(ok1492) mutants and is partially required for enhanced gst-4 expression. (A) Phosphorylation level of PMK-1 is increased in brap-2 mutants. (Left) A Western blot of lysates from wild-type, brap-2(ok1492), pmk-1(km25), and wild-type worms treated with arsenite probed with an α-phospho-p38 antibody. (Right) Relative intensity of bands corresponding to phospho-PMK levels were compared to α-tubulin levels using ImageJ and normalized to wild type. The experiment was carried out three times and the relative intensity of the phospho-PMK bands were pooled and graphed. (B) As a measure of PMK-1 activity, transcript levels of two pmk-1 targets (left, F08G5.6; and right, F35E12.5) in brap-2, pmk-1, and brap-2;pmk-1 mutants were measured using qRT-PCR (log₁₀ scale). brap-2(ok1492) animals showed higher levels of transcripts that were dependent on functional pmk-1. (C) PMK-1 is required to promote phase II detoxification gene expression in brap-2(ok1492) worms. The mRNA levels of (top left) gst-4, (top right) gst-10, (bottom left) gsto-2, and (bottom right) sdz-8 were quantified using qRT-PCR. The brap-2;pmk-1 double mutant showed a reduction of mRNA expression for each gene tested. *** P < 0.001, ** P < 0.01, * P < 0.05.

Figure 5

Loss of elt-3 reduced gst-4 expression in brap-2(ok1492) mutants. (A) Synchronized gst-4p::gfp and brap-2(ok1492); gst-4p::gfp worms were grown on control (L4440) or elt-3 RNAi plates and GFP expression was examined at L4 stage of development. brap-2(ok1492) worms show a reduced gst-4p::gfp expression when elt-3 is knocked down. (B) Synchronized worms were grown on NGM plates followed by RNA extraction and qRT-PCR. The double mutant brap-2(ok1492);elt-3(vp1) shows a 70% reduction in gst-4 mRNA expression in comparison to brap-2(ok1492). *** P < 0.001, ** P < 0.01.

Figure 6

ELT-3 is required for gst-4 expression during oxidative stress and synergistically activates gst-4 promoter expression through interaction with SKN-1. (A) Synchronized worms were collected and treated with (left) 5 mM sodium arsenite (As) or (right) 100 mM paraquat (PQ) for 2 hr followed by RNA isolation. qRT-PCR was performed to measure the gst-4 mRNA transcript levels. (B) Survival assays of N2 and elt-3(vp1) strains were determined in (left) arsenite and (right) paraquat. Results indicated ELT-3 is essential for gst-4 expression during oxidative stress and that elt-3 mutant worms are more susceptible to arsenite exposure compared to wild type. (C) Luciferase assay was performed and the transcriptional activity was determined by the activation of the gst-4 promoter/luciferase reporter vector. Maximum activation occurs upon coexpression of both ELT-3 and SKN-1 constructs. (D) A pull down of GST-ELT-3 fusion protein using glutathione sepharose and Western blot using anti-GFP antibody to detect GFP-SKN-1 fusion protein. This assay reveals that EGFP::SKN-1 and GST::ELT-3 physically interact in vitro. * P < 0.05, ** P < 0.01, *** P < 0.001.

Figure 7

Extended life span in overexpressing SKN-1 requires ELT-3. Survival curves of N2 and worms containing the SKN-1 transgene [strain LD1250 (ldEx7) expressing skn-1b/c::gfp] fed on control bacteria (L4440) or bacteria expressing elt-3 RNAi from L1 worms at 20°. While skn-1b/c::gfp has a significantly longer life span than N2 (mean life spans are 16.7 ± 0.79 days and 13.6 ± 0.62 days, respectively; n = 40), less extension of life span was observed in overexpressing SKN-1B/C in elt-3 RNAi (skn-1b/c::gfp mean = 11.1 ± 0.45 days compared to N2 mean = 11.1 ± 0.38 days; n = 40). Two independent experiments were performed, data represents one trial.