Nrf2/cyclooxygenase 2 signaling in Cr(VI)-induced carcinogenesis

Abstract

Long-term exposure to hexavalent chromium [Cr(VI)] has been linked to lung cancer, and cyclooxygenase-2 (COX-2) is a well-known inflammatory factor. However, the role and mechanism of COX-2 in Cr(VI)-induced carcinogenesis are not clear yet. To address this question, we employed a mouse model exposed to Cr(VI) through intranasal instillation of particulate zinc chromate (ZnCrO₄) for 12 weeks. Metabolomics and RNA-seq assays revealed enhanced activity of the arachidonic acid (AA)/eicosanoid metabolism pathway in lung tissues from mice exposed to Cr(VI). COX-2, the key enzyme of the AA/eicosanoid pathway, was significantly upregulated in Cr(VI)-exposed lung tissues, as well as in the Cr(VI)-induced transformed (Cr-T) cells compared to parental BEAS-2B (B2B) cells. We then employed multidisciplinary in vitro and in vivo functional assays to characterize the role of COX-2 in Cr(VI)-induced lung cancer. The results indicated that COX-2 functioned as an oncogene to promote the malignant transformation of B2B cells and enhance the proliferation, migration, tumor growth, and angiogenesis of Cr-T cells. Nuclear factor E2-related factor-2 (Nrf2) was identified as a transcription factor for COX-2. Nrf2 was upregulated in response to Cr(VI) exposure and contributed to Cr(VI)-induced lung cancers, in part by upregulating COX-2 expression. Moreover, microRNA-379 (miR-379) was found to target COX-2 to inhibit its expression posttranscriptionally. MiR-379 was downregulated in Cr(VI)-exposed lung tissues and Cr-T cells, and ectopic miR-379 expression reduced Cr-T cell viability and migration, with partial reversal upon COX-2 restoration. In summary, our study revealed the oncogenic role of COX-2 and identified two novel regulatory mechanisms for COX-2 overexpression in Cr(VI)-induced carcinogenesis.

Keywords: COX-2; Carcinogenesis; Cr(VI); Nrf2; miR-379.

Fig. 1. Activation of the AA/eicosanoid pathway and induction of COX-2 after Cr(VI) exposure.

BALB/cJ mice were exposed to zinc chromate (ZnCrO₄) or normal saline by intranasal instillation once per week for 12 weeks. (A) Lung tissues were snap-frozen and subjected to a metabolomics assay. The heat map displays the levels of AA/eicosanoid pathway intermediate metabolites in the lung tissues of the Cr(VI)-treated mice and the control mice. (B) GSEA analysis for the RNA-seq data reveals significant enrichment of the AA/eicosanoid metabolism pathway-related genes among the differentially expressed genes between B2B and Cr-T cells. (C) IHC staining was performed to detect COX-2 in formalin-fixed lung tissues of mice in the Cr(VI) exposure and control groups. Scale bar, 25 μm. (D) COX-2 mRNA expression in the lung tissues was measured using RT-qPCR assay. Ctrl: n=15, Cr(VI) instillation: n=20. (E) The plasma of the mice was separated through density gradient centrifugation, and the COX-2 protein level in the plasma was determined using an ELISA assay. Ctrl: n=15, Cr(VI) instillation: n=20. (F) COX-2 protein expression in B2B and Cr-T cells was assessed using an immunoblotting assay. Data are presented as the mean ± SE of each group of animals. ***, P < 0.001.

Fig. 1. Activation of the AA/eicosanoid pathway and induction of COX-2 after Cr(VI) exposure.

BALB/cJ mice were exposed to zinc chromate (ZnCrO₄) or normal saline by intranasal instillation once per week for 12 weeks. (A) Lung tissues were snap-frozen and subjected to a metabolomics assay. The heat map displays the levels of AA/eicosanoid pathway intermediate metabolites in the lung tissues of the Cr(VI)-treated mice and the control mice. (B) GSEA analysis for the RNA-seq data reveals significant enrichment of the AA/eicosanoid metabolism pathway-related genes among the differentially expressed genes between B2B and Cr-T cells. (C) IHC staining was performed to detect COX-2 in formalin-fixed lung tissues of mice in the Cr(VI) exposure and control groups. Scale bar, 25 μm. (D) COX-2 mRNA expression in the lung tissues was measured using RT-qPCR assay. Ctrl: n=15, Cr(VI) instillation: n=20. (E) The plasma of the mice was separated through density gradient centrifugation, and the COX-2 protein level in the plasma was determined using an ELISA assay. Ctrl: n=15, Cr(VI) instillation: n=20. (F) COX-2 protein expression in B2B and Cr-T cells was assessed using an immunoblotting assay. Data are presented as the mean ± SE of each group of animals. ***, P < 0.001.

Fig. 1. Activation of the AA/eicosanoid pathway and induction of COX-2 after Cr(VI) exposure.

BALB/cJ mice were exposed to zinc chromate (ZnCrO₄) or normal saline by intranasal instillation once per week for 12 weeks. (A) Lung tissues were snap-frozen and subjected to a metabolomics assay. The heat map displays the levels of AA/eicosanoid pathway intermediate metabolites in the lung tissues of the Cr(VI)-treated mice and the control mice. (B) GSEA analysis for the RNA-seq data reveals significant enrichment of the AA/eicosanoid metabolism pathway-related genes among the differentially expressed genes between B2B and Cr-T cells. (C) IHC staining was performed to detect COX-2 in formalin-fixed lung tissues of mice in the Cr(VI) exposure and control groups. Scale bar, 25 μm. (D) COX-2 mRNA expression in the lung tissues was measured using RT-qPCR assay. Ctrl: n=15, Cr(VI) instillation: n=20. (E) The plasma of the mice was separated through density gradient centrifugation, and the COX-2 protein level in the plasma was determined using an ELISA assay. Ctrl: n=15, Cr(VI) instillation: n=20. (F) COX-2 protein expression in B2B and Cr-T cells was assessed using an immunoblotting assay. Data are presented as the mean ± SE of each group of animals. ***, P < 0.001.

Fig. 1. Activation of the AA/eicosanoid pathway and induction of COX-2 after Cr(VI) exposure.

BALB/cJ mice were exposed to zinc chromate (ZnCrO₄) or normal saline by intranasal instillation once per week for 12 weeks. (A) Lung tissues were snap-frozen and subjected to a metabolomics assay. The heat map displays the levels of AA/eicosanoid pathway intermediate metabolites in the lung tissues of the Cr(VI)-treated mice and the control mice. (B) GSEA analysis for the RNA-seq data reveals significant enrichment of the AA/eicosanoid metabolism pathway-related genes among the differentially expressed genes between B2B and Cr-T cells. (C) IHC staining was performed to detect COX-2 in formalin-fixed lung tissues of mice in the Cr(VI) exposure and control groups. Scale bar, 25 μm. (D) COX-2 mRNA expression in the lung tissues was measured using RT-qPCR assay. Ctrl: n=15, Cr(VI) instillation: n=20. (E) The plasma of the mice was separated through density gradient centrifugation, and the COX-2 protein level in the plasma was determined using an ELISA assay. Ctrl: n=15, Cr(VI) instillation: n=20. (F) COX-2 protein expression in B2B and Cr-T cells was assessed using an immunoblotting assay. Data are presented as the mean ± SE of each group of animals. ***, P < 0.001.

Fig. 1. Activation of the AA/eicosanoid pathway and induction of COX-2 after Cr(VI) exposure.

BALB/cJ mice were exposed to zinc chromate (ZnCrO₄) or normal saline by intranasal instillation once per week for 12 weeks. (A) Lung tissues were snap-frozen and subjected to a metabolomics assay. The heat map displays the levels of AA/eicosanoid pathway intermediate metabolites in the lung tissues of the Cr(VI)-treated mice and the control mice. (B) GSEA analysis for the RNA-seq data reveals significant enrichment of the AA/eicosanoid metabolism pathway-related genes among the differentially expressed genes between B2B and Cr-T cells. (C) IHC staining was performed to detect COX-2 in formalin-fixed lung tissues of mice in the Cr(VI) exposure and control groups. Scale bar, 25 μm. (D) COX-2 mRNA expression in the lung tissues was measured using RT-qPCR assay. Ctrl: n=15, Cr(VI) instillation: n=20. (E) The plasma of the mice was separated through density gradient centrifugation, and the COX-2 protein level in the plasma was determined using an ELISA assay. Ctrl: n=15, Cr(VI) instillation: n=20. (F) COX-2 protein expression in B2B and Cr-T cells was assessed using an immunoblotting assay. Data are presented as the mean ± SE of each group of animals. ***, P < 0.001.

Fig. 1. Activation of the AA/eicosanoid pathway and induction of COX-2 after Cr(VI) exposure.

BALB/cJ mice were exposed to zinc chromate (ZnCrO₄) or normal saline by intranasal instillation once per week for 12 weeks. (A) Lung tissues were snap-frozen and subjected to a metabolomics assay. The heat map displays the levels of AA/eicosanoid pathway intermediate metabolites in the lung tissues of the Cr(VI)-treated mice and the control mice. (B) GSEA analysis for the RNA-seq data reveals significant enrichment of the AA/eicosanoid metabolism pathway-related genes among the differentially expressed genes between B2B and Cr-T cells. (C) IHC staining was performed to detect COX-2 in formalin-fixed lung tissues of mice in the Cr(VI) exposure and control groups. Scale bar, 25 μm. (D) COX-2 mRNA expression in the lung tissues was measured using RT-qPCR assay. Ctrl: n=15, Cr(VI) instillation: n=20. (E) The plasma of the mice was separated through density gradient centrifugation, and the COX-2 protein level in the plasma was determined using an ELISA assay. Ctrl: n=15, Cr(VI) instillation: n=20. (F) COX-2 protein expression in B2B and Cr-T cells was assessed using an immunoblotting assay. Data are presented as the mean ± SE of each group of animals. ***, P < 0.001.

Fig. 2. COX-2 acts as an oncogene to promote Cr(VI)-induced lung cancer.

(A) B2B cells with stable COX-2 overexpression (COX-2 OE) or vector control were generated by lentiviral transduction and puromycin selection. COX-2 protein levels were assessed using an immunoblotting assay. (B) COX-2 overexpressing and control B2B cells were treated with 0.5 μM Cr(VI) for up to 4 months. The colony formation ability of these cells was evaluated at 2 months, 3 months, and 4 months using the soft agar assay. Left: representative images of the soft agar assay; right: quantification of the soft agar assay. Data represent the mean ± SE of six replicate wells at each time point. (C and D) Lentivirus-mediated overexpression of COX-2 was achieved in Cr-T cells. (C) Cell viability was determined by counting cell numbers. (D) A transwell migration assay was conducted to evaluate cell migratory ability. Left panel: representative image of migrated cells; right panel: quantification of migrated cells. Scale bar, 100 μm. (E-H) Cr-T cells stably overexpressing COX-2 or control Cr-T cells were subcutaneously injected into posterior flanks of 6-week-old nude mice at a dose of 2x10⁶ cells per injection. The xenografts were collected and analyzed 18 days after injection. n=8 for each group. (E) Tumor sizes were measured, and the volumes were calculated using the formula: tumor volume = (length × width²) × 0.5. The comparison of the two curves was performed using two-way ANOVA followed by post hoc multiple comparisons. (F) The representative images of the tumors are shown. (G) Measurement of tumor weights. (H) IHC staining for CD31-positive microvessels in xenografts. Left panel: representative images of IHC staining; right panel: quantification of CD31-positive microvessels, which were normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments unless otherwise specified. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 2. COX-2 acts as an oncogene to promote Cr(VI)-induced lung cancer.

(A) B2B cells with stable COX-2 overexpression (COX-2 OE) or vector control were generated by lentiviral transduction and puromycin selection. COX-2 protein levels were assessed using an immunoblotting assay. (B) COX-2 overexpressing and control B2B cells were treated with 0.5 μM Cr(VI) for up to 4 months. The colony formation ability of these cells was evaluated at 2 months, 3 months, and 4 months using the soft agar assay. Left: representative images of the soft agar assay; right: quantification of the soft agar assay. Data represent the mean ± SE of six replicate wells at each time point. (C and D) Lentivirus-mediated overexpression of COX-2 was achieved in Cr-T cells. (C) Cell viability was determined by counting cell numbers. (D) A transwell migration assay was conducted to evaluate cell migratory ability. Left panel: representative image of migrated cells; right panel: quantification of migrated cells. Scale bar, 100 μm. (E-H) Cr-T cells stably overexpressing COX-2 or control Cr-T cells were subcutaneously injected into posterior flanks of 6-week-old nude mice at a dose of 2x10⁶ cells per injection. The xenografts were collected and analyzed 18 days after injection. n=8 for each group. (E) Tumor sizes were measured, and the volumes were calculated using the formula: tumor volume = (length × width²) × 0.5. The comparison of the two curves was performed using two-way ANOVA followed by post hoc multiple comparisons. (F) The representative images of the tumors are shown. (G) Measurement of tumor weights. (H) IHC staining for CD31-positive microvessels in xenografts. Left panel: representative images of IHC staining; right panel: quantification of CD31-positive microvessels, which were normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments unless otherwise specified. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 2. COX-2 acts as an oncogene to promote Cr(VI)-induced lung cancer.

(A) B2B cells with stable COX-2 overexpression (COX-2 OE) or vector control were generated by lentiviral transduction and puromycin selection. COX-2 protein levels were assessed using an immunoblotting assay. (B) COX-2 overexpressing and control B2B cells were treated with 0.5 μM Cr(VI) for up to 4 months. The colony formation ability of these cells was evaluated at 2 months, 3 months, and 4 months using the soft agar assay. Left: representative images of the soft agar assay; right: quantification of the soft agar assay. Data represent the mean ± SE of six replicate wells at each time point. (C and D) Lentivirus-mediated overexpression of COX-2 was achieved in Cr-T cells. (C) Cell viability was determined by counting cell numbers. (D) A transwell migration assay was conducted to evaluate cell migratory ability. Left panel: representative image of migrated cells; right panel: quantification of migrated cells. Scale bar, 100 μm. (E-H) Cr-T cells stably overexpressing COX-2 or control Cr-T cells were subcutaneously injected into posterior flanks of 6-week-old nude mice at a dose of 2x10⁶ cells per injection. The xenografts were collected and analyzed 18 days after injection. n=8 for each group. (E) Tumor sizes were measured, and the volumes were calculated using the formula: tumor volume = (length × width²) × 0.5. The comparison of the two curves was performed using two-way ANOVA followed by post hoc multiple comparisons. (F) The representative images of the tumors are shown. (G) Measurement of tumor weights. (H) IHC staining for CD31-positive microvessels in xenografts. Left panel: representative images of IHC staining; right panel: quantification of CD31-positive microvessels, which were normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments unless otherwise specified. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 2. COX-2 acts as an oncogene to promote Cr(VI)-induced lung cancer.

(A) B2B cells with stable COX-2 overexpression (COX-2 OE) or vector control were generated by lentiviral transduction and puromycin selection. COX-2 protein levels were assessed using an immunoblotting assay. (B) COX-2 overexpressing and control B2B cells were treated with 0.5 μM Cr(VI) for up to 4 months. The colony formation ability of these cells was evaluated at 2 months, 3 months, and 4 months using the soft agar assay. Left: representative images of the soft agar assay; right: quantification of the soft agar assay. Data represent the mean ± SE of six replicate wells at each time point. (C and D) Lentivirus-mediated overexpression of COX-2 was achieved in Cr-T cells. (C) Cell viability was determined by counting cell numbers. (D) A transwell migration assay was conducted to evaluate cell migratory ability. Left panel: representative image of migrated cells; right panel: quantification of migrated cells. Scale bar, 100 μm. (E-H) Cr-T cells stably overexpressing COX-2 or control Cr-T cells were subcutaneously injected into posterior flanks of 6-week-old nude mice at a dose of 2x10⁶ cells per injection. The xenografts were collected and analyzed 18 days after injection. n=8 for each group. (E) Tumor sizes were measured, and the volumes were calculated using the formula: tumor volume = (length × width²) × 0.5. The comparison of the two curves was performed using two-way ANOVA followed by post hoc multiple comparisons. (F) The representative images of the tumors are shown. (G) Measurement of tumor weights. (H) IHC staining for CD31-positive microvessels in xenografts. Left panel: representative images of IHC staining; right panel: quantification of CD31-positive microvessels, which were normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments unless otherwise specified. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 2. COX-2 acts as an oncogene to promote Cr(VI)-induced lung cancer.

(A) B2B cells with stable COX-2 overexpression (COX-2 OE) or vector control were generated by lentiviral transduction and puromycin selection. COX-2 protein levels were assessed using an immunoblotting assay. (B) COX-2 overexpressing and control B2B cells were treated with 0.5 μM Cr(VI) for up to 4 months. The colony formation ability of these cells was evaluated at 2 months, 3 months, and 4 months using the soft agar assay. Left: representative images of the soft agar assay; right: quantification of the soft agar assay. Data represent the mean ± SE of six replicate wells at each time point. (C and D) Lentivirus-mediated overexpression of COX-2 was achieved in Cr-T cells. (C) Cell viability was determined by counting cell numbers. (D) A transwell migration assay was conducted to evaluate cell migratory ability. Left panel: representative image of migrated cells; right panel: quantification of migrated cells. Scale bar, 100 μm. (E-H) Cr-T cells stably overexpressing COX-2 or control Cr-T cells were subcutaneously injected into posterior flanks of 6-week-old nude mice at a dose of 2x10⁶ cells per injection. The xenografts were collected and analyzed 18 days after injection. n=8 for each group. (E) Tumor sizes were measured, and the volumes were calculated using the formula: tumor volume = (length × width²) × 0.5. The comparison of the two curves was performed using two-way ANOVA followed by post hoc multiple comparisons. (F) The representative images of the tumors are shown. (G) Measurement of tumor weights. (H) IHC staining for CD31-positive microvessels in xenografts. Left panel: representative images of IHC staining; right panel: quantification of CD31-positive microvessels, which were normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments unless otherwise specified. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 2. COX-2 acts as an oncogene to promote Cr(VI)-induced lung cancer.

(A) B2B cells with stable COX-2 overexpression (COX-2 OE) or vector control were generated by lentiviral transduction and puromycin selection. COX-2 protein levels were assessed using an immunoblotting assay. (B) COX-2 overexpressing and control B2B cells were treated with 0.5 μM Cr(VI) for up to 4 months. The colony formation ability of these cells was evaluated at 2 months, 3 months, and 4 months using the soft agar assay. Left: representative images of the soft agar assay; right: quantification of the soft agar assay. Data represent the mean ± SE of six replicate wells at each time point. (C and D) Lentivirus-mediated overexpression of COX-2 was achieved in Cr-T cells. (C) Cell viability was determined by counting cell numbers. (D) A transwell migration assay was conducted to evaluate cell migratory ability. Left panel: representative image of migrated cells; right panel: quantification of migrated cells. Scale bar, 100 μm. (E-H) Cr-T cells stably overexpressing COX-2 or control Cr-T cells were subcutaneously injected into posterior flanks of 6-week-old nude mice at a dose of 2x10⁶ cells per injection. The xenografts were collected and analyzed 18 days after injection. n=8 for each group. (E) Tumor sizes were measured, and the volumes were calculated using the formula: tumor volume = (length × width²) × 0.5. The comparison of the two curves was performed using two-way ANOVA followed by post hoc multiple comparisons. (F) The representative images of the tumors are shown. (G) Measurement of tumor weights. (H) IHC staining for CD31-positive microvessels in xenografts. Left panel: representative images of IHC staining; right panel: quantification of CD31-positive microvessels, which were normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments unless otherwise specified. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 2. COX-2 acts as an oncogene to promote Cr(VI)-induced lung cancer.

(A) B2B cells with stable COX-2 overexpression (COX-2 OE) or vector control were generated by lentiviral transduction and puromycin selection. COX-2 protein levels were assessed using an immunoblotting assay. (B) COX-2 overexpressing and control B2B cells were treated with 0.5 μM Cr(VI) for up to 4 months. The colony formation ability of these cells was evaluated at 2 months, 3 months, and 4 months using the soft agar assay. Left: representative images of the soft agar assay; right: quantification of the soft agar assay. Data represent the mean ± SE of six replicate wells at each time point. (C and D) Lentivirus-mediated overexpression of COX-2 was achieved in Cr-T cells. (C) Cell viability was determined by counting cell numbers. (D) A transwell migration assay was conducted to evaluate cell migratory ability. Left panel: representative image of migrated cells; right panel: quantification of migrated cells. Scale bar, 100 μm. (E-H) Cr-T cells stably overexpressing COX-2 or control Cr-T cells were subcutaneously injected into posterior flanks of 6-week-old nude mice at a dose of 2x10⁶ cells per injection. The xenografts were collected and analyzed 18 days after injection. n=8 for each group. (E) Tumor sizes were measured, and the volumes were calculated using the formula: tumor volume = (length × width²) × 0.5. The comparison of the two curves was performed using two-way ANOVA followed by post hoc multiple comparisons. (F) The representative images of the tumors are shown. (G) Measurement of tumor weights. (H) IHC staining for CD31-positive microvessels in xenografts. Left panel: representative images of IHC staining; right panel: quantification of CD31-positive microvessels, which were normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments unless otherwise specified. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 2. COX-2 acts as an oncogene to promote Cr(VI)-induced lung cancer.

(A) B2B cells with stable COX-2 overexpression (COX-2 OE) or vector control were generated by lentiviral transduction and puromycin selection. COX-2 protein levels were assessed using an immunoblotting assay. (B) COX-2 overexpressing and control B2B cells were treated with 0.5 μM Cr(VI) for up to 4 months. The colony formation ability of these cells was evaluated at 2 months, 3 months, and 4 months using the soft agar assay. Left: representative images of the soft agar assay; right: quantification of the soft agar assay. Data represent the mean ± SE of six replicate wells at each time point. (C and D) Lentivirus-mediated overexpression of COX-2 was achieved in Cr-T cells. (C) Cell viability was determined by counting cell numbers. (D) A transwell migration assay was conducted to evaluate cell migratory ability. Left panel: representative image of migrated cells; right panel: quantification of migrated cells. Scale bar, 100 μm. (E-H) Cr-T cells stably overexpressing COX-2 or control Cr-T cells were subcutaneously injected into posterior flanks of 6-week-old nude mice at a dose of 2x10⁶ cells per injection. The xenografts were collected and analyzed 18 days after injection. n=8 for each group. (E) Tumor sizes were measured, and the volumes were calculated using the formula: tumor volume = (length × width²) × 0.5. The comparison of the two curves was performed using two-way ANOVA followed by post hoc multiple comparisons. (F) The representative images of the tumors are shown. (G) Measurement of tumor weights. (H) IHC staining for CD31-positive microvessels in xenografts. Left panel: representative images of IHC staining; right panel: quantification of CD31-positive microvessels, which were normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments unless otherwise specified. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 3. COX-2 depletion inhibits the cancerous phenotypes of Cr-T cells.

The lentivirus-mediated CRISPR/Cas9 technique was used to achieve COX-2 knockout (KO) in Cr-T cells, while negative control cells were transduced with lentivirus expressing a non-targeting sgRNA. (A) COX-2 protein levels were assessed via immunoblotting assay to validate the knockout. (B) Cell viability was determined by cell counting. (C) The migratory ability of the cells was measured using a transwell migration assay. Left: representative images of migrated cells; right: quantification of the migrated cells. Scale bar, 100 μm. (D-G) Tumor growth assay in nude mice was performed using COX-2-null and control Cr-T cells as above, and the xenografts were harvested 21 days after implantation. n=10 for each group. (D) Tumor sizes were measured at indicated time points, and tumor volumes were calculated and analyzed. The growth curve was plotted and compared with two-way ANOVA followed by post hoc multiple comparisons. (E) The representative images of the tumors are shown. (F) Tumor weights were measured. (G) IHC staining for CD31-positive microvessels in the xenografts. Left panel: representative images of the IHC staining; right panel: quantification of CD31-positive microvessels. The number was normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 3. COX-2 depletion inhibits the cancerous phenotypes of Cr-T cells.

The lentivirus-mediated CRISPR/Cas9 technique was used to achieve COX-2 knockout (KO) in Cr-T cells, while negative control cells were transduced with lentivirus expressing a non-targeting sgRNA. (A) COX-2 protein levels were assessed via immunoblotting assay to validate the knockout. (B) Cell viability was determined by cell counting. (C) The migratory ability of the cells was measured using a transwell migration assay. Left: representative images of migrated cells; right: quantification of the migrated cells. Scale bar, 100 μm. (D-G) Tumor growth assay in nude mice was performed using COX-2-null and control Cr-T cells as above, and the xenografts were harvested 21 days after implantation. n=10 for each group. (D) Tumor sizes were measured at indicated time points, and tumor volumes were calculated and analyzed. The growth curve was plotted and compared with two-way ANOVA followed by post hoc multiple comparisons. (E) The representative images of the tumors are shown. (F) Tumor weights were measured. (G) IHC staining for CD31-positive microvessels in the xenografts. Left panel: representative images of the IHC staining; right panel: quantification of CD31-positive microvessels. The number was normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 3. COX-2 depletion inhibits the cancerous phenotypes of Cr-T cells.

The lentivirus-mediated CRISPR/Cas9 technique was used to achieve COX-2 knockout (KO) in Cr-T cells, while negative control cells were transduced with lentivirus expressing a non-targeting sgRNA. (A) COX-2 protein levels were assessed via immunoblotting assay to validate the knockout. (B) Cell viability was determined by cell counting. (C) The migratory ability of the cells was measured using a transwell migration assay. Left: representative images of migrated cells; right: quantification of the migrated cells. Scale bar, 100 μm. (D-G) Tumor growth assay in nude mice was performed using COX-2-null and control Cr-T cells as above, and the xenografts were harvested 21 days after implantation. n=10 for each group. (D) Tumor sizes were measured at indicated time points, and tumor volumes were calculated and analyzed. The growth curve was plotted and compared with two-way ANOVA followed by post hoc multiple comparisons. (E) The representative images of the tumors are shown. (F) Tumor weights were measured. (G) IHC staining for CD31-positive microvessels in the xenografts. Left panel: representative images of the IHC staining; right panel: quantification of CD31-positive microvessels. The number was normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 3. COX-2 depletion inhibits the cancerous phenotypes of Cr-T cells.

The lentivirus-mediated CRISPR/Cas9 technique was used to achieve COX-2 knockout (KO) in Cr-T cells, while negative control cells were transduced with lentivirus expressing a non-targeting sgRNA. (A) COX-2 protein levels were assessed via immunoblotting assay to validate the knockout. (B) Cell viability was determined by cell counting. (C) The migratory ability of the cells was measured using a transwell migration assay. Left: representative images of migrated cells; right: quantification of the migrated cells. Scale bar, 100 μm. (D-G) Tumor growth assay in nude mice was performed using COX-2-null and control Cr-T cells as above, and the xenografts were harvested 21 days after implantation. n=10 for each group. (D) Tumor sizes were measured at indicated time points, and tumor volumes were calculated and analyzed. The growth curve was plotted and compared with two-way ANOVA followed by post hoc multiple comparisons. (E) The representative images of the tumors are shown. (F) Tumor weights were measured. (G) IHC staining for CD31-positive microvessels in the xenografts. Left panel: representative images of the IHC staining; right panel: quantification of CD31-positive microvessels. The number was normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 3. COX-2 depletion inhibits the cancerous phenotypes of Cr-T cells.

The lentivirus-mediated CRISPR/Cas9 technique was used to achieve COX-2 knockout (KO) in Cr-T cells, while negative control cells were transduced with lentivirus expressing a non-targeting sgRNA. (A) COX-2 protein levels were assessed via immunoblotting assay to validate the knockout. (B) Cell viability was determined by cell counting. (C) The migratory ability of the cells was measured using a transwell migration assay. Left: representative images of migrated cells; right: quantification of the migrated cells. Scale bar, 100 μm. (D-G) Tumor growth assay in nude mice was performed using COX-2-null and control Cr-T cells as above, and the xenografts were harvested 21 days after implantation. n=10 for each group. (D) Tumor sizes were measured at indicated time points, and tumor volumes were calculated and analyzed. The growth curve was plotted and compared with two-way ANOVA followed by post hoc multiple comparisons. (E) The representative images of the tumors are shown. (F) Tumor weights were measured. (G) IHC staining for CD31-positive microvessels in the xenografts. Left panel: representative images of the IHC staining; right panel: quantification of CD31-positive microvessels. The number was normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 3. COX-2 depletion inhibits the cancerous phenotypes of Cr-T cells.

The lentivirus-mediated CRISPR/Cas9 technique was used to achieve COX-2 knockout (KO) in Cr-T cells, while negative control cells were transduced with lentivirus expressing a non-targeting sgRNA. (A) COX-2 protein levels were assessed via immunoblotting assay to validate the knockout. (B) Cell viability was determined by cell counting. (C) The migratory ability of the cells was measured using a transwell migration assay. Left: representative images of migrated cells; right: quantification of the migrated cells. Scale bar, 100 μm. (D-G) Tumor growth assay in nude mice was performed using COX-2-null and control Cr-T cells as above, and the xenografts were harvested 21 days after implantation. n=10 for each group. (D) Tumor sizes were measured at indicated time points, and tumor volumes were calculated and analyzed. The growth curve was plotted and compared with two-way ANOVA followed by post hoc multiple comparisons. (E) The representative images of the tumors are shown. (F) Tumor weights were measured. (G) IHC staining for CD31-positive microvessels in the xenografts. Left panel: representative images of the IHC staining; right panel: quantification of CD31-positive microvessels. The number was normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 3. COX-2 depletion inhibits the cancerous phenotypes of Cr-T cells.

The lentivirus-mediated CRISPR/Cas9 technique was used to achieve COX-2 knockout (KO) in Cr-T cells, while negative control cells were transduced with lentivirus expressing a non-targeting sgRNA. (A) COX-2 protein levels were assessed via immunoblotting assay to validate the knockout. (B) Cell viability was determined by cell counting. (C) The migratory ability of the cells was measured using a transwell migration assay. Left: representative images of migrated cells; right: quantification of the migrated cells. Scale bar, 100 μm. (D-G) Tumor growth assay in nude mice was performed using COX-2-null and control Cr-T cells as above, and the xenografts were harvested 21 days after implantation. n=10 for each group. (D) Tumor sizes were measured at indicated time points, and tumor volumes were calculated and analyzed. The growth curve was plotted and compared with two-way ANOVA followed by post hoc multiple comparisons. (E) The representative images of the tumors are shown. (F) Tumor weights were measured. (G) IHC staining for CD31-positive microvessels in the xenografts. Left panel: representative images of the IHC staining; right panel: quantification of CD31-positive microvessels. The number was normalized to the control group. Scale bar, 50 μm. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 4. Nrf2 regulates COX-2 expression by directly binding to its promoter region.

(A) BALB/cJ mice were treated as shown in Fig. 1. IHC staining was performed to evaluate Nrf2 expression in the formalin-fixed lung tissues. Scale bar, 25 μm. (B) The correlation between COX-2 and Nrf2 expression in the lungs was evaluated using the Pearson correlation analysis. (C and D) Nrf2 was knocked out in Cr-T cells using the lentiviral CRISPR/Cas9-mediated genome editing technique. The control cells were transduced with lentivirus carrying a non-targeting sgRNA. (C) Nrf2 and COX-2 protein was measured using immunoblotting assay. Quantification of the COX-2 signal was performed using the Image Lab software. The data are presented as the mean ± SE of three independent experiments. (D) COX-2 mRNA in the control and Nrf2-null cells was determined using RT-qPCR assay. The data are presented as the mean ± SE of three independent experiments. (E) Nrf2 expression in B2B cells was induced using a lentivirus-mediated Tet-on system harboring the Nrf2 coding sequence. Control cells were transduced with lentivirus generated from the control vector. Both types of cells were incubated with 1.0 mg/mL tetracycline for 3 days. Nrf2 and COX-2 protein expression was measured by immunoblotting assay. (F) A schematic graph showing the potential binding site of Nrf2 in the promoter region of COX-2. There is an ARE consensus sequence between nucleotides −562 and −572. (G) The ChIP-PCR assay was performed to investigate the binding of Nrf2 to the COX-2 promoter in Cr-T cells, as described in the Materials and Methods section. Left: a representative image of the agarose gel displaying PCR products. COX-2 NT represents the non-target region in the COX-2 promoter without the ARE consensus sequence. Right: quantification of the bands on agarose gel. The data are presented as the mean ± SE of three independent experiments. (H) The interaction between Nrf2 and the COX-2 promoter was analyzed in both control and Nrf2-depleted Cr-T cells using the ChIP-PCR assay. Left: a representative image of PCR products was shown. Right: quantification of the agarose gel bands. The data are presented as the mean ± SE of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 4. Nrf2 regulates COX-2 expression by directly binding to its promoter region.

(A) BALB/cJ mice were treated as shown in Fig. 1. IHC staining was performed to evaluate Nrf2 expression in the formalin-fixed lung tissues. Scale bar, 25 μm. (B) The correlation between COX-2 and Nrf2 expression in the lungs was evaluated using the Pearson correlation analysis. (C and D) Nrf2 was knocked out in Cr-T cells using the lentiviral CRISPR/Cas9-mediated genome editing technique. The control cells were transduced with lentivirus carrying a non-targeting sgRNA. (C) Nrf2 and COX-2 protein was measured using immunoblotting assay. Quantification of the COX-2 signal was performed using the Image Lab software. The data are presented as the mean ± SE of three independent experiments. (D) COX-2 mRNA in the control and Nrf2-null cells was determined using RT-qPCR assay. The data are presented as the mean ± SE of three independent experiments. (E) Nrf2 expression in B2B cells was induced using a lentivirus-mediated Tet-on system harboring the Nrf2 coding sequence. Control cells were transduced with lentivirus generated from the control vector. Both types of cells were incubated with 1.0 mg/mL tetracycline for 3 days. Nrf2 and COX-2 protein expression was measured by immunoblotting assay. (F) A schematic graph showing the potential binding site of Nrf2 in the promoter region of COX-2. There is an ARE consensus sequence between nucleotides −562 and −572. (G) The ChIP-PCR assay was performed to investigate the binding of Nrf2 to the COX-2 promoter in Cr-T cells, as described in the Materials and Methods section. Left: a representative image of the agarose gel displaying PCR products. COX-2 NT represents the non-target region in the COX-2 promoter without the ARE consensus sequence. Right: quantification of the bands on agarose gel. The data are presented as the mean ± SE of three independent experiments. (H) The interaction between Nrf2 and the COX-2 promoter was analyzed in both control and Nrf2-depleted Cr-T cells using the ChIP-PCR assay. Left: a representative image of PCR products was shown. Right: quantification of the agarose gel bands. The data are presented as the mean ± SE of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 4. Nrf2 regulates COX-2 expression by directly binding to its promoter region.

(A) BALB/cJ mice were treated as shown in Fig. 1. IHC staining was performed to evaluate Nrf2 expression in the formalin-fixed lung tissues. Scale bar, 25 μm. (B) The correlation between COX-2 and Nrf2 expression in the lungs was evaluated using the Pearson correlation analysis. (C and D) Nrf2 was knocked out in Cr-T cells using the lentiviral CRISPR/Cas9-mediated genome editing technique. The control cells were transduced with lentivirus carrying a non-targeting sgRNA. (C) Nrf2 and COX-2 protein was measured using immunoblotting assay. Quantification of the COX-2 signal was performed using the Image Lab software. The data are presented as the mean ± SE of three independent experiments. (D) COX-2 mRNA in the control and Nrf2-null cells was determined using RT-qPCR assay. The data are presented as the mean ± SE of three independent experiments. (E) Nrf2 expression in B2B cells was induced using a lentivirus-mediated Tet-on system harboring the Nrf2 coding sequence. Control cells were transduced with lentivirus generated from the control vector. Both types of cells were incubated with 1.0 mg/mL tetracycline for 3 days. Nrf2 and COX-2 protein expression was measured by immunoblotting assay. (F) A schematic graph showing the potential binding site of Nrf2 in the promoter region of COX-2. There is an ARE consensus sequence between nucleotides −562 and −572. (G) The ChIP-PCR assay was performed to investigate the binding of Nrf2 to the COX-2 promoter in Cr-T cells, as described in the Materials and Methods section. Left: a representative image of the agarose gel displaying PCR products. COX-2 NT represents the non-target region in the COX-2 promoter without the ARE consensus sequence. Right: quantification of the bands on agarose gel. The data are presented as the mean ± SE of three independent experiments. (H) The interaction between Nrf2 and the COX-2 promoter was analyzed in both control and Nrf2-depleted Cr-T cells using the ChIP-PCR assay. Left: a representative image of PCR products was shown. Right: quantification of the agarose gel bands. The data are presented as the mean ± SE of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 4. Nrf2 regulates COX-2 expression by directly binding to its promoter region.

(A) BALB/cJ mice were treated as shown in Fig. 1. IHC staining was performed to evaluate Nrf2 expression in the formalin-fixed lung tissues. Scale bar, 25 μm. (B) The correlation between COX-2 and Nrf2 expression in the lungs was evaluated using the Pearson correlation analysis. (C and D) Nrf2 was knocked out in Cr-T cells using the lentiviral CRISPR/Cas9-mediated genome editing technique. The control cells were transduced with lentivirus carrying a non-targeting sgRNA. (C) Nrf2 and COX-2 protein was measured using immunoblotting assay. Quantification of the COX-2 signal was performed using the Image Lab software. The data are presented as the mean ± SE of three independent experiments. (D) COX-2 mRNA in the control and Nrf2-null cells was determined using RT-qPCR assay. The data are presented as the mean ± SE of three independent experiments. (E) Nrf2 expression in B2B cells was induced using a lentivirus-mediated Tet-on system harboring the Nrf2 coding sequence. Control cells were transduced with lentivirus generated from the control vector. Both types of cells were incubated with 1.0 mg/mL tetracycline for 3 days. Nrf2 and COX-2 protein expression was measured by immunoblotting assay. (F) A schematic graph showing the potential binding site of Nrf2 in the promoter region of COX-2. There is an ARE consensus sequence between nucleotides −562 and −572. (G) The ChIP-PCR assay was performed to investigate the binding of Nrf2 to the COX-2 promoter in Cr-T cells, as described in the Materials and Methods section. Left: a representative image of the agarose gel displaying PCR products. COX-2 NT represents the non-target region in the COX-2 promoter without the ARE consensus sequence. Right: quantification of the bands on agarose gel. The data are presented as the mean ± SE of three independent experiments. (H) The interaction between Nrf2 and the COX-2 promoter was analyzed in both control and Nrf2-depleted Cr-T cells using the ChIP-PCR assay. Left: a representative image of PCR products was shown. Right: quantification of the agarose gel bands. The data are presented as the mean ± SE of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 4. Nrf2 regulates COX-2 expression by directly binding to its promoter region.

(A) BALB/cJ mice were treated as shown in Fig. 1. IHC staining was performed to evaluate Nrf2 expression in the formalin-fixed lung tissues. Scale bar, 25 μm. (B) The correlation between COX-2 and Nrf2 expression in the lungs was evaluated using the Pearson correlation analysis. (C and D) Nrf2 was knocked out in Cr-T cells using the lentiviral CRISPR/Cas9-mediated genome editing technique. The control cells were transduced with lentivirus carrying a non-targeting sgRNA. (C) Nrf2 and COX-2 protein was measured using immunoblotting assay. Quantification of the COX-2 signal was performed using the Image Lab software. The data are presented as the mean ± SE of three independent experiments. (D) COX-2 mRNA in the control and Nrf2-null cells was determined using RT-qPCR assay. The data are presented as the mean ± SE of three independent experiments. (E) Nrf2 expression in B2B cells was induced using a lentivirus-mediated Tet-on system harboring the Nrf2 coding sequence. Control cells were transduced with lentivirus generated from the control vector. Both types of cells were incubated with 1.0 mg/mL tetracycline for 3 days. Nrf2 and COX-2 protein expression was measured by immunoblotting assay. (F) A schematic graph showing the potential binding site of Nrf2 in the promoter region of COX-2. There is an ARE consensus sequence between nucleotides −562 and −572. (G) The ChIP-PCR assay was performed to investigate the binding of Nrf2 to the COX-2 promoter in Cr-T cells, as described in the Materials and Methods section. Left: a representative image of the agarose gel displaying PCR products. COX-2 NT represents the non-target region in the COX-2 promoter without the ARE consensus sequence. Right: quantification of the bands on agarose gel. The data are presented as the mean ± SE of three independent experiments. (H) The interaction between Nrf2 and the COX-2 promoter was analyzed in both control and Nrf2-depleted Cr-T cells using the ChIP-PCR assay. Left: a representative image of PCR products was shown. Right: quantification of the agarose gel bands. The data are presented as the mean ± SE of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 4. Nrf2 regulates COX-2 expression by directly binding to its promoter region.

(A) BALB/cJ mice were treated as shown in Fig. 1. IHC staining was performed to evaluate Nrf2 expression in the formalin-fixed lung tissues. Scale bar, 25 μm. (B) The correlation between COX-2 and Nrf2 expression in the lungs was evaluated using the Pearson correlation analysis. (C and D) Nrf2 was knocked out in Cr-T cells using the lentiviral CRISPR/Cas9-mediated genome editing technique. The control cells were transduced with lentivirus carrying a non-targeting sgRNA. (C) Nrf2 and COX-2 protein was measured using immunoblotting assay. Quantification of the COX-2 signal was performed using the Image Lab software. The data are presented as the mean ± SE of three independent experiments. (D) COX-2 mRNA in the control and Nrf2-null cells was determined using RT-qPCR assay. The data are presented as the mean ± SE of three independent experiments. (E) Nrf2 expression in B2B cells was induced using a lentivirus-mediated Tet-on system harboring the Nrf2 coding sequence. Control cells were transduced with lentivirus generated from the control vector. Both types of cells were incubated with 1.0 mg/mL tetracycline for 3 days. Nrf2 and COX-2 protein expression was measured by immunoblotting assay. (F) A schematic graph showing the potential binding site of Nrf2 in the promoter region of COX-2. There is an ARE consensus sequence between nucleotides −562 and −572. (G) The ChIP-PCR assay was performed to investigate the binding of Nrf2 to the COX-2 promoter in Cr-T cells, as described in the Materials and Methods section. Left: a representative image of the agarose gel displaying PCR products. COX-2 NT represents the non-target region in the COX-2 promoter without the ARE consensus sequence. Right: quantification of the bands on agarose gel. The data are presented as the mean ± SE of three independent experiments. (H) The interaction between Nrf2 and the COX-2 promoter was analyzed in both control and Nrf2-depleted Cr-T cells using the ChIP-PCR assay. Left: a representative image of PCR products was shown. Right: quantification of the agarose gel bands. The data are presented as the mean ± SE of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 4. Nrf2 regulates COX-2 expression by directly binding to its promoter region.

(A) BALB/cJ mice were treated as shown in Fig. 1. IHC staining was performed to evaluate Nrf2 expression in the formalin-fixed lung tissues. Scale bar, 25 μm. (B) The correlation between COX-2 and Nrf2 expression in the lungs was evaluated using the Pearson correlation analysis. (C and D) Nrf2 was knocked out in Cr-T cells using the lentiviral CRISPR/Cas9-mediated genome editing technique. The control cells were transduced with lentivirus carrying a non-targeting sgRNA. (C) Nrf2 and COX-2 protein was measured using immunoblotting assay. Quantification of the COX-2 signal was performed using the Image Lab software. The data are presented as the mean ± SE of three independent experiments. (D) COX-2 mRNA in the control and Nrf2-null cells was determined using RT-qPCR assay. The data are presented as the mean ± SE of three independent experiments. (E) Nrf2 expression in B2B cells was induced using a lentivirus-mediated Tet-on system harboring the Nrf2 coding sequence. Control cells were transduced with lentivirus generated from the control vector. Both types of cells were incubated with 1.0 mg/mL tetracycline for 3 days. Nrf2 and COX-2 protein expression was measured by immunoblotting assay. (F) A schematic graph showing the potential binding site of Nrf2 in the promoter region of COX-2. There is an ARE consensus sequence between nucleotides −562 and −572. (G) The ChIP-PCR assay was performed to investigate the binding of Nrf2 to the COX-2 promoter in Cr-T cells, as described in the Materials and Methods section. Left: a representative image of the agarose gel displaying PCR products. COX-2 NT represents the non-target region in the COX-2 promoter without the ARE consensus sequence. Right: quantification of the bands on agarose gel. The data are presented as the mean ± SE of three independent experiments. (H) The interaction between Nrf2 and the COX-2 promoter was analyzed in both control and Nrf2-depleted Cr-T cells using the ChIP-PCR assay. Left: a representative image of PCR products was shown. Right: quantification of the agarose gel bands. The data are presented as the mean ± SE of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 4. Nrf2 regulates COX-2 expression by directly binding to its promoter region.

(A) BALB/cJ mice were treated as shown in Fig. 1. IHC staining was performed to evaluate Nrf2 expression in the formalin-fixed lung tissues. Scale bar, 25 μm. (B) The correlation between COX-2 and Nrf2 expression in the lungs was evaluated using the Pearson correlation analysis. (C and D) Nrf2 was knocked out in Cr-T cells using the lentiviral CRISPR/Cas9-mediated genome editing technique. The control cells were transduced with lentivirus carrying a non-targeting sgRNA. (C) Nrf2 and COX-2 protein was measured using immunoblotting assay. Quantification of the COX-2 signal was performed using the Image Lab software. The data are presented as the mean ± SE of three independent experiments. (D) COX-2 mRNA in the control and Nrf2-null cells was determined using RT-qPCR assay. The data are presented as the mean ± SE of three independent experiments. (E) Nrf2 expression in B2B cells was induced using a lentivirus-mediated Tet-on system harboring the Nrf2 coding sequence. Control cells were transduced with lentivirus generated from the control vector. Both types of cells were incubated with 1.0 mg/mL tetracycline for 3 days. Nrf2 and COX-2 protein expression was measured by immunoblotting assay. (F) A schematic graph showing the potential binding site of Nrf2 in the promoter region of COX-2. There is an ARE consensus sequence between nucleotides −562 and −572. (G) The ChIP-PCR assay was performed to investigate the binding of Nrf2 to the COX-2 promoter in Cr-T cells, as described in the Materials and Methods section. Left: a representative image of the agarose gel displaying PCR products. COX-2 NT represents the non-target region in the COX-2 promoter without the ARE consensus sequence. Right: quantification of the bands on agarose gel. The data are presented as the mean ± SE of three independent experiments. (H) The interaction between Nrf2 and the COX-2 promoter was analyzed in both control and Nrf2-depleted Cr-T cells using the ChIP-PCR assay. Left: a representative image of PCR products was shown. Right: quantification of the agarose gel bands. The data are presented as the mean ± SE of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig. 5. Nrf2 promotes Cr(VI)-induced lung cancer through regulating COX-2 expression.

(A) Nrf2 was knocked out in Cr-T cells, as depicted in Fig. 4C. Nrf2 knockout cells (KO#1 and KO#2) were achieved using Nrf2 sgRNA #1 and #2, respectively. Cell viability of control and Nrf2-null Cr-T cells was assessed by cell counting. (B) The migration ability of control and Nrf2-null Cr-T cells was evaluated using a transwell migration assay. Left: representative images of migrated cells; right: quantification of migrated cells. Scale bar, 100 μm. (C-E) Nrf2 KO Cr-T cells were transduced with lentivirus expressing a Tet-on inducible COX-2 coding sequence, while control cells were transduced with a negative control Tet-on lentivirus. Cells were treated with 1.0 μg/mL tetracycline to induce COX-2 expression. (C) COX-2 protein expression in control and COX-2-overexpressing cells was evaluated using an immunoblotting assay. (D) Cell viability was assessed by cell counting. (E) Cell migration ability was evaluated using a transwell migration assay. Left: representative images of migrated cells; right: quantification of migrated cells. Scale bar, 100 μm. Data are presented as the mean ± SE of three independent experiments. **, P < 0.01; ***, P < 0.001.

Fig. 5. Nrf2 promotes Cr(VI)-induced lung cancer through regulating COX-2 expression.

(A) Nrf2 was knocked out in Cr-T cells, as depicted in Fig. 4C. Nrf2 knockout cells (KO#1 and KO#2) were achieved using Nrf2 sgRNA #1 and #2, respectively. Cell viability of control and Nrf2-null Cr-T cells was assessed by cell counting. (B) The migration ability of control and Nrf2-null Cr-T cells was evaluated using a transwell migration assay. Left: representative images of migrated cells; right: quantification of migrated cells. Scale bar, 100 μm. (C-E) Nrf2 KO Cr-T cells were transduced with lentivirus expressing a Tet-on inducible COX-2 coding sequence, while control cells were transduced with a negative control Tet-on lentivirus. Cells were treated with 1.0 μg/mL tetracycline to induce COX-2 expression. (C) COX-2 protein expression in control and COX-2-overexpressing cells was evaluated using an immunoblotting assay. (D) Cell viability was assessed by cell counting. (E) Cell migration ability was evaluated using a transwell migration assay. Left: representative images of migrated cells; right: quantification of migrated cells. Scale bar, 100 μm. Data are presented as the mean ± SE of three independent experiments. **, P < 0.01; ***, P < 0.001.

Fig. 5. Nrf2 promotes Cr(VI)-induced lung cancer through regulating COX-2 expression.

(A) Nrf2 was knocked out in Cr-T cells, as depicted in Fig. 4C. Nrf2 knockout cells (KO#1 and KO#2) were achieved using Nrf2 sgRNA #1 and #2, respectively. Cell viability of control and Nrf2-null Cr-T cells was assessed by cell counting. (B) The migration ability of control and Nrf2-null Cr-T cells was evaluated using a transwell migration assay. Left: representative images of migrated cells; right: quantification of migrated cells. Scale bar, 100 μm. (C-E) Nrf2 KO Cr-T cells were transduced with lentivirus expressing a Tet-on inducible COX-2 coding sequence, while control cells were transduced with a negative control Tet-on lentivirus. Cells were treated with 1.0 μg/mL tetracycline to induce COX-2 expression. (C) COX-2 protein expression in control and COX-2-overexpressing cells was evaluated using an immunoblotting assay. (D) Cell viability was assessed by cell counting. (E) Cell migration ability was evaluated using a transwell migration assay. Left: representative images of migrated cells; right: quantification of migrated cells. Scale bar, 100 μm. Data are presented as the mean ± SE of three independent experiments. **, P < 0.01; ***, P < 0.001.

Fig. 5. Nrf2 promotes Cr(VI)-induced lung cancer through regulating COX-2 expression.

(A) Nrf2 was knocked out in Cr-T cells, as depicted in Fig. 4C. Nrf2 knockout cells (KO#1 and KO#2) were achieved using Nrf2 sgRNA #1 and #2, respectively. Cell viability of control and Nrf2-null Cr-T cells was assessed by cell counting. (B) The migration ability of control and Nrf2-null Cr-T cells was evaluated using a transwell migration assay. Left: representative images of migrated cells; right: quantification of migrated cells. Scale bar, 100 μm. (C-E) Nrf2 KO Cr-T cells were transduced with lentivirus expressing a Tet-on inducible COX-2 coding sequence, while control cells were transduced with a negative control Tet-on lentivirus. Cells were treated with 1.0 μg/mL tetracycline to induce COX-2 expression. (C) COX-2 protein expression in control and COX-2-overexpressing cells was evaluated using an immunoblotting assay. (D) Cell viability was assessed by cell counting. (E) Cell migration ability was evaluated using a transwell migration assay. Left: representative images of migrated cells; right: quantification of migrated cells. Scale bar, 100 μm. Data are presented as the mean ± SE of three independent experiments. **, P < 0.01; ***, P < 0.001.

Fig. 5. Nrf2 promotes Cr(VI)-induced lung cancer through regulating COX-2 expression.

(A) Nrf2 was knocked out in Cr-T cells, as depicted in Fig. 4C. Nrf2 knockout cells (KO#1 and KO#2) were achieved using Nrf2 sgRNA #1 and #2, respectively. Cell viability of control and Nrf2-null Cr-T cells was assessed by cell counting. (B) The migration ability of control and Nrf2-null Cr-T cells was evaluated using a transwell migration assay. Left: representative images of migrated cells; right: quantification of migrated cells. Scale bar, 100 μm. (C-E) Nrf2 KO Cr-T cells were transduced with lentivirus expressing a Tet-on inducible COX-2 coding sequence, while control cells were transduced with a negative control Tet-on lentivirus. Cells were treated with 1.0 μg/mL tetracycline to induce COX-2 expression. (C) COX-2 protein expression in control and COX-2-overexpressing cells was evaluated using an immunoblotting assay. (D) Cell viability was assessed by cell counting. (E) Cell migration ability was evaluated using a transwell migration assay. Left: representative images of migrated cells; right: quantification of migrated cells. Scale bar, 100 μm. Data are presented as the mean ± SE of three independent experiments. **, P < 0.01; ***, P < 0.001.

Fig. 6. miR-379 inhibits Cr-T cell viability and migration via directly targeting COX-2.

(A) miR-379 expression was measured in B2B and Cr-T cells using TaqMan real-time PCR assay. (B) Mice were treated with intranasal instillation of saline or Cr(VI) for 12 weeks. miR-379 level in the lung tissues of the mice was measured using the TaqMan assay. Ctrl: n=15, Cr(VI) instillation: n=20. (C) The correlation between COX-2 mRNA expression and miR-379 level in the mouse lung tissues was analyzed using Pearson correlation analysis. (D) A diagram shows the putative seed sequence between miR-379 and 3’-UTR of COX-2 and the wild-type (WT) and mutant (Mut) luciferase reporter constructs. (E) Luciferase reporter assay was performed using HEK293 cells to detect the relative luciferase activities of WT and Mut COX-2 reporters with or without miR-379 overexpression. The pRL-TK Renilla luciferase vector was used as an internal control. The miR-NC was used as a negative control for miR-379 transfection. (F) Cr-T cells were transfected with 50 nM non-targeting control oligos (miR-NC) or miR-379 mimics and incubated for 72 h. COX-2 protein levels were determined using an immunoblotting assay. (G-H) Cr-T cells stably expressing negative control or an inducible COX-2 were transfected with 50 nM miR-NC or miR-379 mimics; the cells were then treated with 1 μg/mL doxycycline for 3 days to restore COX-2 expression. (G) Cell viability was evaluated by counting cell numbers. (H) The migratory ability of cells was evaluated using a transwell migration assay. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. Scale bar, 100 μm. n.s., not significant; **, p < 0.01; ***, p < 0.001.

Fig. 6. miR-379 inhibits Cr-T cell viability and migration via directly targeting COX-2.

(A) miR-379 expression was measured in B2B and Cr-T cells using TaqMan real-time PCR assay. (B) Mice were treated with intranasal instillation of saline or Cr(VI) for 12 weeks. miR-379 level in the lung tissues of the mice was measured using the TaqMan assay. Ctrl: n=15, Cr(VI) instillation: n=20. (C) The correlation between COX-2 mRNA expression and miR-379 level in the mouse lung tissues was analyzed using Pearson correlation analysis. (D) A diagram shows the putative seed sequence between miR-379 and 3’-UTR of COX-2 and the wild-type (WT) and mutant (Mut) luciferase reporter constructs. (E) Luciferase reporter assay was performed using HEK293 cells to detect the relative luciferase activities of WT and Mut COX-2 reporters with or without miR-379 overexpression. The pRL-TK Renilla luciferase vector was used as an internal control. The miR-NC was used as a negative control for miR-379 transfection. (F) Cr-T cells were transfected with 50 nM non-targeting control oligos (miR-NC) or miR-379 mimics and incubated for 72 h. COX-2 protein levels were determined using an immunoblotting assay. (G-H) Cr-T cells stably expressing negative control or an inducible COX-2 were transfected with 50 nM miR-NC or miR-379 mimics; the cells were then treated with 1 μg/mL doxycycline for 3 days to restore COX-2 expression. (G) Cell viability was evaluated by counting cell numbers. (H) The migratory ability of cells was evaluated using a transwell migration assay. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. Scale bar, 100 μm. n.s., not significant; **, p < 0.01; ***, p < 0.001.

Fig. 6. miR-379 inhibits Cr-T cell viability and migration via directly targeting COX-2.

(A) miR-379 expression was measured in B2B and Cr-T cells using TaqMan real-time PCR assay. (B) Mice were treated with intranasal instillation of saline or Cr(VI) for 12 weeks. miR-379 level in the lung tissues of the mice was measured using the TaqMan assay. Ctrl: n=15, Cr(VI) instillation: n=20. (C) The correlation between COX-2 mRNA expression and miR-379 level in the mouse lung tissues was analyzed using Pearson correlation analysis. (D) A diagram shows the putative seed sequence between miR-379 and 3’-UTR of COX-2 and the wild-type (WT) and mutant (Mut) luciferase reporter constructs. (E) Luciferase reporter assay was performed using HEK293 cells to detect the relative luciferase activities of WT and Mut COX-2 reporters with or without miR-379 overexpression. The pRL-TK Renilla luciferase vector was used as an internal control. The miR-NC was used as a negative control for miR-379 transfection. (F) Cr-T cells were transfected with 50 nM non-targeting control oligos (miR-NC) or miR-379 mimics and incubated for 72 h. COX-2 protein levels were determined using an immunoblotting assay. (G-H) Cr-T cells stably expressing negative control or an inducible COX-2 were transfected with 50 nM miR-NC or miR-379 mimics; the cells were then treated with 1 μg/mL doxycycline for 3 days to restore COX-2 expression. (G) Cell viability was evaluated by counting cell numbers. (H) The migratory ability of cells was evaluated using a transwell migration assay. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. Scale bar, 100 μm. n.s., not significant; **, p < 0.01; ***, p < 0.001.

Fig. 6. miR-379 inhibits Cr-T cell viability and migration via directly targeting COX-2.

(A) miR-379 expression was measured in B2B and Cr-T cells using TaqMan real-time PCR assay. (B) Mice were treated with intranasal instillation of saline or Cr(VI) for 12 weeks. miR-379 level in the lung tissues of the mice was measured using the TaqMan assay. Ctrl: n=15, Cr(VI) instillation: n=20. (C) The correlation between COX-2 mRNA expression and miR-379 level in the mouse lung tissues was analyzed using Pearson correlation analysis. (D) A diagram shows the putative seed sequence between miR-379 and 3’-UTR of COX-2 and the wild-type (WT) and mutant (Mut) luciferase reporter constructs. (E) Luciferase reporter assay was performed using HEK293 cells to detect the relative luciferase activities of WT and Mut COX-2 reporters with or without miR-379 overexpression. The pRL-TK Renilla luciferase vector was used as an internal control. The miR-NC was used as a negative control for miR-379 transfection. (F) Cr-T cells were transfected with 50 nM non-targeting control oligos (miR-NC) or miR-379 mimics and incubated for 72 h. COX-2 protein levels were determined using an immunoblotting assay. (G-H) Cr-T cells stably expressing negative control or an inducible COX-2 were transfected with 50 nM miR-NC or miR-379 mimics; the cells were then treated with 1 μg/mL doxycycline for 3 days to restore COX-2 expression. (G) Cell viability was evaluated by counting cell numbers. (H) The migratory ability of cells was evaluated using a transwell migration assay. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. Scale bar, 100 μm. n.s., not significant; **, p < 0.01; ***, p < 0.001.

Fig. 6. miR-379 inhibits Cr-T cell viability and migration via directly targeting COX-2.

(A) miR-379 expression was measured in B2B and Cr-T cells using TaqMan real-time PCR assay. (B) Mice were treated with intranasal instillation of saline or Cr(VI) for 12 weeks. miR-379 level in the lung tissues of the mice was measured using the TaqMan assay. Ctrl: n=15, Cr(VI) instillation: n=20. (C) The correlation between COX-2 mRNA expression and miR-379 level in the mouse lung tissues was analyzed using Pearson correlation analysis. (D) A diagram shows the putative seed sequence between miR-379 and 3’-UTR of COX-2 and the wild-type (WT) and mutant (Mut) luciferase reporter constructs. (E) Luciferase reporter assay was performed using HEK293 cells to detect the relative luciferase activities of WT and Mut COX-2 reporters with or without miR-379 overexpression. The pRL-TK Renilla luciferase vector was used as an internal control. The miR-NC was used as a negative control for miR-379 transfection. (F) Cr-T cells were transfected with 50 nM non-targeting control oligos (miR-NC) or miR-379 mimics and incubated for 72 h. COX-2 protein levels were determined using an immunoblotting assay. (G-H) Cr-T cells stably expressing negative control or an inducible COX-2 were transfected with 50 nM miR-NC or miR-379 mimics; the cells were then treated with 1 μg/mL doxycycline for 3 days to restore COX-2 expression. (G) Cell viability was evaluated by counting cell numbers. (H) The migratory ability of cells was evaluated using a transwell migration assay. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. Scale bar, 100 μm. n.s., not significant; **, p < 0.01; ***, p < 0.001.

Fig. 6. miR-379 inhibits Cr-T cell viability and migration via directly targeting COX-2.

(A) miR-379 expression was measured in B2B and Cr-T cells using TaqMan real-time PCR assay. (B) Mice were treated with intranasal instillation of saline or Cr(VI) for 12 weeks. miR-379 level in the lung tissues of the mice was measured using the TaqMan assay. Ctrl: n=15, Cr(VI) instillation: n=20. (C) The correlation between COX-2 mRNA expression and miR-379 level in the mouse lung tissues was analyzed using Pearson correlation analysis. (D) A diagram shows the putative seed sequence between miR-379 and 3’-UTR of COX-2 and the wild-type (WT) and mutant (Mut) luciferase reporter constructs. (E) Luciferase reporter assay was performed using HEK293 cells to detect the relative luciferase activities of WT and Mut COX-2 reporters with or without miR-379 overexpression. The pRL-TK Renilla luciferase vector was used as an internal control. The miR-NC was used as a negative control for miR-379 transfection. (F) Cr-T cells were transfected with 50 nM non-targeting control oligos (miR-NC) or miR-379 mimics and incubated for 72 h. COX-2 protein levels were determined using an immunoblotting assay. (G-H) Cr-T cells stably expressing negative control or an inducible COX-2 were transfected with 50 nM miR-NC or miR-379 mimics; the cells were then treated with 1 μg/mL doxycycline for 3 days to restore COX-2 expression. (G) Cell viability was evaluated by counting cell numbers. (H) The migratory ability of cells was evaluated using a transwell migration assay. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. Scale bar, 100 μm. n.s., not significant; **, p < 0.01; ***, p < 0.001.

Fig. 6. miR-379 inhibits Cr-T cell viability and migration via directly targeting COX-2.

(A) miR-379 expression was measured in B2B and Cr-T cells using TaqMan real-time PCR assay. (B) Mice were treated with intranasal instillation of saline or Cr(VI) for 12 weeks. miR-379 level in the lung tissues of the mice was measured using the TaqMan assay. Ctrl: n=15, Cr(VI) instillation: n=20. (C) The correlation between COX-2 mRNA expression and miR-379 level in the mouse lung tissues was analyzed using Pearson correlation analysis. (D) A diagram shows the putative seed sequence between miR-379 and 3’-UTR of COX-2 and the wild-type (WT) and mutant (Mut) luciferase reporter constructs. (E) Luciferase reporter assay was performed using HEK293 cells to detect the relative luciferase activities of WT and Mut COX-2 reporters with or without miR-379 overexpression. The pRL-TK Renilla luciferase vector was used as an internal control. The miR-NC was used as a negative control for miR-379 transfection. (F) Cr-T cells were transfected with 50 nM non-targeting control oligos (miR-NC) or miR-379 mimics and incubated for 72 h. COX-2 protein levels were determined using an immunoblotting assay. (G-H) Cr-T cells stably expressing negative control or an inducible COX-2 were transfected with 50 nM miR-NC or miR-379 mimics; the cells were then treated with 1 μg/mL doxycycline for 3 days to restore COX-2 expression. (G) Cell viability was evaluated by counting cell numbers. (H) The migratory ability of cells was evaluated using a transwell migration assay. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. Scale bar, 100 μm. n.s., not significant; **, p < 0.01; ***, p < 0.001.

Fig. 6. miR-379 inhibits Cr-T cell viability and migration via directly targeting COX-2.

(A) miR-379 expression was measured in B2B and Cr-T cells using TaqMan real-time PCR assay. (B) Mice were treated with intranasal instillation of saline or Cr(VI) for 12 weeks. miR-379 level in the lung tissues of the mice was measured using the TaqMan assay. Ctrl: n=15, Cr(VI) instillation: n=20. (C) The correlation between COX-2 mRNA expression and miR-379 level in the mouse lung tissues was analyzed using Pearson correlation analysis. (D) A diagram shows the putative seed sequence between miR-379 and 3’-UTR of COX-2 and the wild-type (WT) and mutant (Mut) luciferase reporter constructs. (E) Luciferase reporter assay was performed using HEK293 cells to detect the relative luciferase activities of WT and Mut COX-2 reporters with or without miR-379 overexpression. The pRL-TK Renilla luciferase vector was used as an internal control. The miR-NC was used as a negative control for miR-379 transfection. (F) Cr-T cells were transfected with 50 nM non-targeting control oligos (miR-NC) or miR-379 mimics and incubated for 72 h. COX-2 protein levels were determined using an immunoblotting assay. (G-H) Cr-T cells stably expressing negative control or an inducible COX-2 were transfected with 50 nM miR-NC or miR-379 mimics; the cells were then treated with 1 μg/mL doxycycline for 3 days to restore COX-2 expression. (G) Cell viability was evaluated by counting cell numbers. (H) The migratory ability of cells was evaluated using a transwell migration assay. Data are presented as the mean ± SE of each group of animals or three independent cell line-based experiments. Scale bar, 100 μm. n.s., not significant; **, p < 0.01; ***, p < 0.001.