Indomethacin Induces Spermidine/Spermine-N1-Acetyltransferase-1 via the Nucleolin-CDK1 Axis and Synergizes with the Polyamine Oxidase Inhibitor Methoctramine in Lung Cancer Cells

Abstract

Indomethacin is a non-selective NSAID used against pain and inflammation. Although cyclooxygenase (COX) inhibition is considered indomethacin's primary action mechanism, COX-independent ways are associated with beneficial effects in cancer. In colon cancer cells, the activation of the peroxisome proliferator-activated receptor-γ (PPAR-γ) is related to the increase in spermidine/spermine-N¹-acetyltransferase-1 (SSAT-1), a key enzyme for polyamine degradation, and related to cell cycle arrest. Indomethacin increases the SSAT-1 levels in lung cancer cells; however, the mechanism relying on the SSAT-1 increase is unclear. Thus, we asked for the influence of the PPAR-γ on the SSAT-1 expression in two lung cancer cell lines: H1299 and A549. We found that the inhibition of PPAR-γ with GW9662 did not revert the increase in SSAT-1 induced by indomethacin. Because the mRNA of SSAT-1 suffers a pre-translation retention step by nucleolin, a nucleolar protein, we explored the relationship between indomethacin and the upstream translation regulators of SSAT-1. We found that indomethacin decreases the nucleolin levels and the cyclin-dependent kinase 1 (CDK1) levels, which phosphorylates nucleolin in mitosis. Overexpression of nucleolin partially reverts the effect of indomethacin over cell viability and SSAT-1 levels. On the other hand, Casein Kinase, known for phosphorylating nucleolin during interphase, is not modified by indomethacin. SSAT-1 exerts its antiproliferative effect by acetylating polyamines, a process reverted by the polyamine oxidase (PAOX). Recently, methoctramine was described as the most specific inhibitor of PAOX. Thus, we asked if methoctramine could increase the effect of indomethacin. We found that, when combined, indomethacin and methoctramine have a synergistic effect against NSCLC cells in vitro. These results suggest that indomethacin increases the SSAT-1 levels by reducing the CDK1-nucleolin regulatory axis, and the PAOX inhibition with methoctramine could improve the antiproliferative effect of indomethacin.

Keywords: cancer; indomethacin; non-steroidal anti-inflammatory drugs; polyamines.

Figure 6

Proposed model of the induction of SSAT-1 by indomethacin and the synergistic effect with methoctramine. Indomethacin can induce the PPAR-γ receptor but also decrease the levels of CDK1, allowing the increase in SSAT-1 synthesis. Methoctramine also inhibits the polyamine oxidase (PAOX) enzyme, increasing the effect of SSAT-1 and having a synergistic effect with indomethacin. The increased acetylated polyamines can be further exported from the cell by the solute carrier 3 family member 2 (SLC3A2) transporter [56]. Figure designed with Biorender.

Figure 1

Effect of indomethacin on the spermidine/spermine-N¹-acetyltransferase-1 (SSAT-1) levels and its relationship with the inhibition of the peroxisome proliferator-activated receptor-γ (PPAR-γ). H1299 were exposed to indomethacin 0.5 and 1 mM with or without the PPAR-γ inhibitor GW9662 (10 µM) for 24 h. The levels of SSAT-1 were evaluated by immunoblotting. (A) Representative blot of the SSAT-1 protein levels in H1299 cells exposed to indomethacin (0.5 and 1 mM). (B) Quantitation of the SSAT-1 levels in H1299 cells exposed to indomethacin (1 mM). (C) Representative blot of the SSAT-1 protein levels in H1299 cells exposed to indomethacin 1 mM and the PPAR-γ inhibitor GW9662 (10 µM). (D) Quantitation of the SSAT-1 levels in H1299 cells exposed to indomethacin (1 mM). (E) Cell viability of H1299 after 96 hrs. of exposition to indomethacin alone, or in presence of GW9662 (10 µM). Panel (B,D) show relative expression levels that were normalized using the control value from each individual replicate. *: p < 0.05, compared with control, calculated using one-way ANOVA and Dunnet’s post-test. Data summarize the results of three independent experiments. Original Images can be found in Figure S1.

Figure 2

Effect of indomethacin on the spermidine/spermine-N¹-acetyltransferase-1 (SSAT-1) and nucleolin levels in lung cancer cell lines. H1299 and A549 cells (non-small cell lung cancer) were exposed to indomethacin 0.5 and 1 mM for 24 h. (A) SSAT-1 detection by immunofluorescence microscopy in H1299 cells exposed to 0.5 mM indomethacin. DAPI was used for staining the nuclei. (B) Representative blot of the SSAT-1 and nucleolin protein levels in H1299 cells exposed to indomethacin. (C) Quantitation of the SSAT-1 and nucleolin levels in H1299 cells exposed to indomethacin. (D) SSAT-1 detection by immunofluorescence microscopy in A549 cells exposed to 0.5 mM indomethacin. DAPI was used for staining the nuclei. (E) Representative blot of the SSAT-1 and nucleolin protein levels in A549 cells exposed to indomethacin (F) Quantitation of the SSAT-1 and nucleolin levels in A549 cells exposed to indomethacin. For (A,D) panels, white arrows show nuclear spots of SSAT-1 (SSAT-1 images) and cytosolic expression of SSAT-1 (merge panels). Panel (C,F) show relative expression levels that were normalized using the control value from each individual replicate. *: p < 0.05; **: p < 0.01; ****: p < 0.0001, compared with control, calculated using one-way ANOVA and Dunnet’s post-test. Data summarize the results of three independent experiments. Original Images can be found in Figure S1.

Figure 3

Effect of indomethacin on the Cdk1 and CK2 levels in lung cancer cell lines. H1299 and A549 cells (non-small cell lung cancer) were exposed to indomethacin 0.5 and 1 mM for 24 h. CDK1 and CK2 were detected by immunoblotting. (A) Representative blots of the CDK1 and CK2 protein levels in H1299 cells exposed to indomethacin. (B) Quantitation of the CDK1 levels in H1299 cells exposed to indomethacin. (C) Quantitation of the CK2 levels in H1299 cells exposed to indomethacin. (D) Representative blot of the SSAT-1-1 and nucleolin protein levels in A549 cells exposed to indomethacin. (E) Quantitation of the CDK1 levels in A549 cells exposed to indomethacin. (F) Quantitation of the CK2 levels in H1299 cells exposed to indomethacin. Panel (B,C,E,F) show relative expression levels that were normalized using the control value from each individual replicate. **: p < 0.01; ****: p < 0.0001, compared with control, calculated using one-way ANOVA and Dunnet’s post-test. Data summarize the results of three independent experiments. Original Images can be found in Figure S1.

Figure 4

Overexpression of nucleolin reverses the effect of indomethacin over spermidine/spermine-N1-acetyltransferase-1 (SSAT-1) levels and cell viability. H1299 cells were transfected with a Nucleolin-CFP (Cyan Fluorescent Protein) expressing vector and exposed to indomethacin. (A) Immunoblot of nucleolin in H1299 cells transfected with nucleolin-CFP vector. White arrow shows the nucleolin protein that carry the CFP. (B) Fluorescence microscopy showing the CFP fluorescence in transfected H1299 cells. (C) Representative blot of the SSAT-1 protein levels in H1299 cells transfected with the nucleolin-CFP vector and exposed to indomethacin 1 mM for 24 h. (D) Quantitation of the SSAT-1 levels in H1299 cells ells transfected with the nucleolin-CFP vector and exposed to indomethacin 1 mM for 24 h. Relative expression levels that were normalized using the control value from each individual replicate. *: p < 0.05; compared with untreated control, calculated using one-way ANOVA and Dunnet’s post-test. (E) Cell viability of H1299 cells transfected with the nucleolin-CFP vector and exposed to indomethacin 1 mM for 96 h. ***: p < 0.001 and ****: p < 0.0001, compared with the control at the same indomethacin concentration, calculated using two-way ANOVA. Data summarize the results of three independent experiments. Original Images can be found in Figure S1.

Figure 5

Effect of Indomethacin combined with methoctramine on the viability of NSCLC cells. The effect of the combination of indomethacin with methoctramine on cell viability was measured by resazurin reduction assay. H1299 and A549 cells were exposed to the different drugs for 96 h. For drug combinations, two-fold dilution series, comprising eight concentrations, were mixed in every possible combination. Effects matrices were plotted using COMBENEFIT software, which builds an XYZ model of the combination, using the effect of each drug alone and the Bliss model of drug additivity. (A,B) Concentration-response curves for indomethacin and methoctramine in H1299 cells. (C,D) Heatmaps in 3D (C) and the statistical analysis (D) showing the synergistic effect between indomethacin and methoctramine in H1299 cells. (E,F) Concentration-response curves for indomethacin and methoctramine in H1299 cells. (G,H) Heatmaps in 3D (G) and the statistical analysis (H) showing the synergistic effect between indomethacin and methoctramine in H1299 cells. Differences between the theoretical combinations and empirical data are represented by a number generated for every combination point. Positive numbers represent synergistic combinations, while negative numbers indicate antagonistic interactions. The color code shows combination points that are significantly different from the theoretical model, calculated by the t-test. * p < 0.05; ** p < 0.01 and *** p < 0.001. The graphs show the mean ± standard deviation of four independent experiments.