8-oxoguanine DNA glycosylase-1 augments proinflammatory gene expression by facilitating the recruitment of site-specific transcription factors

Abstract

Among the insidious DNA base lesions, 8-oxo-7,8-dihydroguanine (8-oxoG) is one of the most abundant, a lesion that arises through the attack by reactive oxygen species on guanine, especially when located in cis-regulatory elements. 8-oxoG is repaired by the 8-oxoguanine glycosylase 1 (OGG1)-initiated DNA base excision repair pathway. In this study, we investigated whether 8-oxoG repair by OGG1 in promoter regions is compatible with a prompt gene expression and a host innate immune response. For this purpose, we used a mouse model of airway inflammation, supplemented with cell cultures, chromatin immunoprecipitation, small interfering RNA knockdown, real-time PCR, and comet and reporter transcription assays. Our data show that exposure of cells to TNF-α altered cellular redox, increased the 8-oxoG level in DNA, recruited OGG1 to promoter sequences, and transiently inhibited base excision repair of 8-oxoG. Promoter-associated OGG1 then enhanced NF-κB/RelA binding to cis-elements and facilitated recruitment of specificity protein 1, transcription initiation factor II-D, and p-RNA polymerase II, resulting in the rapid expression of chemokines/cytokines and inflammatory cell accumulation in mouse airways. Small interfering RNA depletion of OGG1 or prevention of guanine oxidation significantly decreased TNF-α-induced inflammatory responses. Taken together, these results show that nonproductive binding of OGG1 to 8-oxoG in promoter sequences could be an epigenetic mechanism to modulate gene expression for a prompt innate immune response.

FIGURE 1

Effect of OGG1 on expression of cytokines and chemokines induced by TNF-á. A) TNF-á-induced expression of mRNAs for pro-inflammatory mediators. B) OGG1 deficiency decreases TNF-á-induced expression of pro-inflammatory chemokine and cytokine mRNAs. C) Effect of AO pre-treatment on TNF-á-induced expression of pro-inflammatory mediators. In A, B and C, groups of mice proficient or deficient in expressing OGG1 in the airway epithelium were challenged intranasally with TNF-á. One hr later, lungs were excised and RNA was extracted. Pooled cDNA from each group (n = 5) was used as a template to perform plate-based inflammation-related PCR arrays. D) Changes in Cxcl-1 (left panel) and Cxcl-2 (right panel) mRNA levels as determined by real-time PCR in individual lung RNA extracts. OGG1 depletion and AO pre-treatment are described in Materials and Methods. E) Effect of OGG1 depletion and AO on the number of neutrophils in BALF of TNF-α-challenged mice. Mice were TNF-α-treated and lavaged as in the Materials and Methods, at 16 h. The percentage of neutrophils was determined as in Materials and Methods. n= 5; *, p<0.05; **, p<0.01; ***, p<0.001.

FIGURE 2

TNF-α-induced expression of Cxcl-2 mRNA is decreased in OGG1-depleted cells. A,B) Time course of Cxcl-2 mRNA expression upon TNF-α exposure. MLE-12 (A) and HEK 293 (B) cells were TNF-α-treated for various lengths of time, RNA was extracted and real-time PCR performed. C,D) OGG1 depletion by siRNA lowers the increase in Cxcl-2 mRNA levels upon TNF-α treatment. OGG1-depleted and control cells (MLE-12, C; HEK 293, D) were exposed to TNF-α for 30 min, RNA was extracted and real-time PCR was performed. E) OGG1 protein levels after siRNA silencing of Ogg1. MLE-12 (upper panels) and HEK 293 (lower panels) cells were transfected with siRNA to Ogg1, lysed and Western blot analyses performed. F) Mouse embryonic fibroblast (MEF) cells lacking OGG1 activity express low levels of Cxcl-2 mRNA after TNF-α exposure. Ogg1^+/+ and Ogg1^−/− MEF cells were treated with TNF-α for 30 min, RNA was extracted and real-time PCR was performed. G,H) The antioxidant NAC decreases TNF-α-induced expression of the Cxcl-2 gene. MLE-12 (G) and HEK 293 (H) cells were challenged with TNF-α for 30 min ± NAC, RNA was extracted and real-time PCR performed. I) Kinetic changes in cellular ROS levels in TNF-α-exposed cells as determined by Amplex UltraRed assays. J) NAC pre-treatment decreases ROS levels in TNF-α-exposed cells. n = 3–5 **, p<0.01; ***, p<0.001

FIGURE 3

OGG1 expression enhances TNF-α-induced activation of the CXCL-2 promoter. A) Diagram of the proximal region of the mouse Cxcl-2 promoter (containing the TATA box, two NF-κB and one Sp1 binding sites). B) TNF-α exposure activates the Cxcl2 promoter. HEK 293 cells were transfected with a reporter plasmid (Cxcl-2-Luc) or control vector (pGL4.2) and then challenged with TNF-α for the time intervals indicated. Dual reporter assays were performed as in Materials and Methods. C) Activation of the Cxcl-2 promoter by TNF-α is inhibited by an NF-κB superrepressor (IκBα-SR). HEK 293 cells were transfected with the reporter plasmid Cxcl-2-Luc with or without overexpression of IκBα-SR (pcDNA3-IκBα-SR), then challenged ± TNF-α for 6 h. D) Deletion of the Sp-1 consensus sequence decreases TNF-α-induced Cxcl-2 promoter activation. HEK 293 cells were transfected with report plasmids Cxcl-2-Luc or Cxcl-2-Sp1-del-Luc, and then mock or ±TNF-α-challenged for 6 h. Luciferase activity in cells without TNF-α challenge was taken as 1. E) OGG1 depletion inhibits TNF-α-induced Cxcl-2 promoter activation. HEK 293 cells were transfected with siRNA to OGG1 (or control), and transfected with reporter plasmid. After a 12 h recovery, cells were treated ± TNF-α for 6 h. Luciferase activity in cells without TNF-α challenge was taken as 1. F) Antioxidant pre-treatment decreased TNF-α-induced Cxcl-2 promoter activation. HEK 293 cells were transfected with the reporter plasmid Cxcl-2-Luc, treated with NAC as in Materials and Methods and then mock- or TNF-α-challenged for 6 h. AV: absolute value, RV: relative value. n = 4–6 **, p<0.01; ***, p<0.001.

FIGURE 4

OGG1 binds to the CXCL-2 promoter and facilitates NF-κB/RelA recruitment. A) TNF-α increases binding of OGG1 and NF-κB/RelA to the CXCL-2 promoter. HEK 293 cells were transfected with Flag-OGG1 plasmid, then treated ± TNF-α for 30 min. ChIP assays were performed using Abs against Flag and NF-κB/RelA. The pulled-down CXCL-2 promoter was detected by PCR amplification and agarose electrophoresis (amplification from samples without Ab incubation served as negative controls). A representative set of experiment is shown out of three. B) OGG1 depletion decreases the association of NF-κB/RelA with the CXCL-2 promoter. HEK 293 cells were transfected with siRNA (or control siRNA) to OGG1, then treated ± TNF-α for 30 min. C) Lack of OGG1 activity in MEFs hampers the binding of NF-κB/RelA to Cxcl-2 promoter sequences. Ogg1^−/− and Ogg1^+/+ MEFs were treated ± TNF-α for 30 min. In B and C, ChIP assay was performed using Ab to NF-κB/RelA. Quantitative amplification of the CXCL-2 promoter from the ChIP products of different cells was compared by real-time PCR. Amplification from the ChIP products was normalized to that from input genomic DNA, and the value for untreated cells was taken as 1. n = 3–4 **, p<0.01

FIGURE 5

TNF-α treatment increases the integrity of the Cxcl-2 promoter. A) Accumulation of 8-oxoG in genomic DNA in TNF-α treated cells. MLE-12 cells were challenged ± TNF-α for 0, 15, 30, 60 and 90 min. OGG1 FLARE™ Comet Assays were performed as in Materials and Methods. B) Representative images show comet moments of DNA with or without rOGG1 digestion (after 30 min of TNF-α exposure). C) 8-oxoG excision activity of rOGG1. OGG1’s activity was determined using Cy5-labeled 8-oxoG probe (Materials and Methods). D) The amplifiable amount of Cxcl-2 promoter is increased in response to TNF-α challenge. MLE-12 cells were exposed to TNF-α for time intervals indicated. Genomic DNA was extracted and real-time PCR was performed to determine amount of proximal 240 bp long region of Cxcl-2 promoter. E) 8-oxoG accumulates in cxcl-2 promoter. MLE-12 cells were exposed to TNF-α as in legend to A and the extracted DNA was digested with rOGG1 then 240 bp long region of Cxcl-2 promoter was amplified by real-time PCR. F) Oxidative modifications at cysteines of OGG1 in TNF-α-exposed cells as shown by DCP-Bio1 a sulfenic acid reacting reagent (left upper panel). Right panel shows percentage of oxidatively modified OGG1 at cysteine(s). n = 3–4, *p<0.05; **, p<0.01; ***, p<0.001.

FIGURE 6

OGG1 interacts with general and site-specific transcription factors. A) Interactions of OGG1 with NF-κB/RelA, p-RNA Pol II, Sp1 and TFIID in response to TNF-α challenge. HEK 293 cells were transfected with a Flag-OGG1-expressing plasmid and 24 h later cells were treated with TNF-α for the indicated time intervals. Co-IP was performed using an Ab against Flag. B) Anti-oxidant decreased the interaction of OGG1 with NF-κB/RelA and p-Pol II. HEK 293 cells were transfected with the Flag-OGG1-expressing plasmid, and subjected to TNF-α challenge for 30 min. Co-IP was performed to analyze the interaction of OGG1 with NF-κB/RelA and RNA p-Pol II. Shown are representative results of three independent experiments. C) Physical interaction between NF-κB/RelA and OGG1. His-OGG1 was immobilized to NTA-agarose beads, washed and then incubated with equimolar non-tagged RelA in interaction buffer for 30 min (Materials and Methods). Bound proteins were eluted and analyzed by Western blotting. D) Protein-protein interaction between OGG1 and Sp1. Assays were carried out as in legend to C except His-Sp1 was NTA-agarose-immobilized. E) NF-κB/RelA interacts with oxidatively modified OGG1 at cysteine. Flag-OGG1 expressing cells were TNF-α exposed and lysed in buffer containing DCP-Bio1. Co-IP was performed using Ab to NF-κB/RelA. OGG1 associated with RelA was analyzed for cysteine oxidation (Materials and Methods). F) Oxidatively modified OGG1 at cysteine(s) interacts with Sp1. Flag-OGG1 expressing cells were TNF-α exposed and lysed in buffer containing DCP-Bio1. Co-IP was performed using Ab to Sp1. OGG1 associated with Sp1 was analyzed for cysteine oxidation (Materials and Methods). n = 3–4.

FIGURE 7

A model of OGG1-driven transcriptional initiation of pro-inflammatory mediators. A) Oxidative modification to guanine and OGG1 as well as activation of trans-acting factors by ROS. B) Non-productive binding of OGG1 to 8-oxoG in promoter region of pro-inflammatory gene(s). C) Assembly of transcriptional machinery facilitated by OGG1.