Transcriptional suppression of CYP2A13 expression by lipopolysaccharide in cultured human lung cells and the lungs of a CYP2A13-humanized mouse model

Abstract

CYP2A13, a human P450 enzyme preferentially expressed in the respiratory tract, is highly efficient in the metabolic activation of tobacco-specific nitrosamines. The aim of this study was to test the hypothesis that inflammation suppresses CYP2A13 expression in the lung, thus explaining the large interindividual differences in CYP2A13 levels previously found in human lung biopsy samples. We first demonstrated that the bacterial endotoxin lipopolysaccharide (LPS) and the proinflammatory cytokine IL-6 can suppress CYP2A13 messenger RNA (mRNA) expression in the NCI-H441 human lung cell line. We then report that an ip injection of LPS (1mg/kg), which induces systemic and lung inflammation, caused substantial reductions in CYP2A13 mRNA (~50%) and protein levels (~80%) in the lungs of a newly generated CYP2A13-humanized mouse model. We further identified two critical CYP2A13 promoter regions, one (major) between -484 and -1008bp and the other (minor) between -134 and -216bp, for the response to LPS, through reporter gene assays in H441 cells. The potential involvement of the nuclear factor NF-κB in LPS-induced CYP2A13 downregulation was suggested by identification of putative NF-κB binding sites within the LPS response regions and effects of an NF-κB inhibitor (pyrrolidine dithiocarbamate) on CYP2A13 expression in H441 cells. Results from gel shift assays further confirmed binding of NF-κB-like nuclear proteins of H441 cells to the major LPS response region of the CYP2A13 promoter. Thus, our findings strongly support the hypothesis that CYP2A13 levels in human lung can be suppressed by inflammation associated with disease status in tissue donors, causing underestimation of CYP2A13 levels in healthy lung.

Keywords: CYP2A; LPS; chemical carcinogenesis.; inflammation; lung.

Fig. 1.

Effects of LPS and IL-6 on CYP2A13 expression in AzaC- and TSA-treated NCI-H441 cells. Where indicated, NCI-H441 cells were treated with dimethyl sulfoxide (0.02%) and ethanol (0.05%) as vehicle control, AzaC (2μM, 72h), and TSA (0.5μM, 24h) only (added to make the cells permissive of constitutive CYP2A13 expression), or AzaC/TSA plus LPS (10 μg/ml, 24h) or IL-6 (100ng/ml, 24h), as described under Materials and Methods section. RNA was isolated at 24h after LPS or IL-6 addition, for determination of CYP2A13 mRNA level using real-time PCR. Data represent means ± SD (n = 3) and were normalized by levels of β-actin. *p < 0.05, compared with cells treated with AzaC/TSA only; one-way ANOVA followed by Dunnett’s test.

Fig. 2.

Effects of LPS treatment on lung CYP2A13 expression in vivo in a CYP2A13-humanized mouse model. (A, B, and C) LPS effect on hepatic SAP mRNA level (A) and serum (B) and lung IL-6 protein level (C) in CYP2A13-humanized mice. Mice (2-month-old female) were injected ip with 1 (A and B) or 0.5 (C) mg/kg LPS, and serum and tissues were obtained at various time points as indicated for analysis. SAP mRNA levels were normalized by levels of GAPDH. IL-6 was detected by ELISA. Data represent means ± SD (n = 4–6). *p < 0.05 and ***p < 0.001, respectively, compared with 0h; one-way ANOVA followed by Dunnett's test (A and B); or with saline group (C); Student’s t-test. (D, E, and F) LPS effect on lung CYP2A13 mRNA (D) and protein (E and F) levels in CYP2A13-humanized mice. Mice (2-month-old female) were treated with 1mg/kg LPS. Lungs were obtained at various time points for analysis. Relative CYP2A13 mRNA levels (D) (normalized by levels of GAPDH) represent means ± SD (n = 4–6); *p < 0.05, compared with 0h group; one-way ANOVA followed by Dunnett's test. For immunoblot analysis, lung microsomal proteins (10 μg/lane) were prepared from tissues pooled from four mice per group and were analyzed in duplicate using a rabbit anti-CYP2A5 polyclonal antibody. In the representative blots shown (E), recombinant CYP2A13 protein was used as a positive control and lung microsome from a Cyp2a5-knockout mouse was used as a negative control, for identification of the CYP2A13 band. The position of a nonspecific band (NS) is indicated. Calnexin was detected as a loading control. For relative CYP2A13 protein levels (F), data represent means ± SD for results from three separate analyses, normalized by the amounts of calnexin detected in the same samples. **p < 0.01 and ***p < 0.001, respectively, compared with 0h; one-way ANOVA followed by Dunnett's test.

Fig. 3.

Identification of LPS response regions within CYP2A13 promoter. (A) Effects of LPS on CYP2A13 promoter activity in NCI-H441 cells. NCI-H441 cells were transiently transfected with one of four CYP2A13 promoter constructs, at 24h prior to LPS treatment of the cells (at 10 μg/ml), and luciferase activities were determined 24h after the LPS treatment. Activities of firefly luciferase in cell lysates were normalized to those of cotransfected Renilla reniformis luciferase, and the results are shown as relative activities (LPS treated/untreated × 100) for each promoter construct. The amounts of CYP2A13 promoter sequence included in each construct are indicated. Data (means ± SD) represent three independent experiments. *p < 0.05, indicating significant decreases in promoter activity in LPS-treated relative to untreated cells (Student’s t-test). (B and C) Dose-response of the effects of LPS on CYP2A13 promoter activity. NCI-H441 cells were transfected with p2A13_1008 (B) or p2A13_216 (C) at 24h prior to LPS treatment at various doses, and luciferase activities were determined 24h after the LPS treatment. Data represent means ± SD (n = 3). *p < 0.05; **p < 0.01, one-way ANOVA followed by Dunnett’s multiple comparison test, compared with corresponding untreated cells.

Fig. 4.

Effects of PDTC, an NF-κB inhibitor, on LPS-induced suppression of CYP2A13 expression and CYP2A13 promoter activity. (A) Effects on CYP2A13 expression. NCI-H441 cells were treated with AzaC (72h)/TSA (24h) and LPS (10 μg/ml, 24h) as described in Figure 1, with or without cotreatment with PDTC (100μM, 24h). Cells were harvested 24h after LPS or LPS/PDTC treatment. Data represent means ± SD (n = 3) and were normalized by levels of β-actin. *p < 0.05; compared with cells treated with AzaC/TSA only; ^# p < 0.05; compared with cells treated with AzaC/TSA/LPS; one-way ANOVA, followed by Tukey’s test. (B) Effects on CYP2A13 promoter activity. NCI-H441 cells were transiently transfected with the p2A13_1008 vector at 24h prior to LPS (10 μg/ml), PDTC (100μM), or LPS plus PDTC treatment, and luciferase activities were determined 24h after the LPS or PDTC treatment. Data represent means ± SD (n = 3). **p < 0.01, compared with p2A13_1008 alone; ^## p < 0.01, compared with p2A13_1008 plus LPS; one-way ANOVA, followed by Tukey’s test.

Fig. 5.

Potential role of NF-κB in LPS-induced suppression of CYP2A13 expression and CYP2A13 promoter activity. (A) Predicted NF-κB binding sites in the proximal promoter region of CYP2A13. The two predicted binding sites are boxed, with the similarity scores shown in parentheses. The distal NF-κB binding site overlaps with a putative C/EBP binding site (underlined); the common sequence between the two is italicized. The sequence of the oligonucleotide probe used for gel shift assays is indicated by double-headed arrows. (B and C) Gel shift assays. H441 cells were either untreated or treated with LPS at 10 μg/ml and harvested 3h later for preparation of nuclear fractions. Nuclear proteins (5 μg) were incubated with ³²P-labeled oligonucleotide probe at room temperature for 30min. The incubation mixtures were resolved by electrophoresis, and bands were visualized by autoradiography. Various unlabeled oligonucleotide probes were added in excess amounts in order to test their ability to compete with the radiolabeled probe, including the gel shift probe (i.e., self-competitor; 100-fold molar excess) (B), NF-κB consensus probe, and C/EBP consensus probe (100-fold molar excess) (C). The position of specific shift bands is indicated by arrows; nonspecific (NS) bands and free probe are also indicated.