The therapeutic effect of DX2 inhibition in nicotine-induced lung cancer progression

Abstract

Alternative splicing products of AIMP2 and AIMP2-DX2 (DX2) have been reported to be associated with human lung cancer. In fact, DX2 expression is elevated in human lung cancers, and DX2 transgenic mice also develop lung cancer, in particular small cell lung cancer (SCLC). However, the mechanism by which DX2 is induced during cancer progression has not been clearly elucidated. Here, we show that DX2 is induced by nicotine, the main component of smoking-related chemicals, which can stabilize the human epidermal growth factor receptor 2 (HER2) protein and transcriptionally increase sonic hedgehog (Shh). Indeed, nicotine showed tumorigenicity via DX2 by promoting spheroid formation and in vivo lung and kidney cancer progression. Moreover, the elimination of DX2 using small interfering RNA (siRNA) or an optimized inhibitor (SNU-14) blocked the induction of HER2 and Shh and completely suppressed tumor sphere formation in response to nicotine. These results indicate that DX2 is critical for lung cancer progression, and a specific DX2 inhibitor would be useful for the treatment of human cancers, including SCLC and non-SCLC (NSCLC).

Keywords: DX2; HER2; Sonic hedgehog; anti-cancer drugs; lung cancer; nicotine; oncogenesis; small cell lung cancer; small-molecule drugs; smoking.

Graphical abstract

Figure 1

Induction of DX2 expression by nicotine in pulmonary cells (A) Nicotine induces DX2 expression in a dose-dependent manner in human diploid lung fibroblasts (WI-26). Cells were treated with nicotine at the indicated concentrations for 24 h, and DX2 expression was measured by western blotting. Actin was used as a loading control. (B and C) Nicotine induced DX2 expression only at the protein level. NSCLC (H460) and SCLC (H209) cells were treated with nicotine at 30 μM for the indicated times, and immunoblot or RT-PCR was performed. Actin and GAPDH were used for loading. (D) Nicotine induced proliferation of WI-26 cells. Using live-cell imaging analysis, the proliferation of WI-26 cells was measured every 4 h. Error bars indicate the standard deviations (SD). ∗∗p < 0.005. (E) Nicotine elevated DX2 expression in a time-dependent manner in human diploid lung fibroblasts. Cells were treated with nicotine (30 μM) for the indicated time, and immunoblotting was performed. Actin was used as a loading control. L.E and S.E indicate long exposure and short exposure, respectively. (F) Cycloheximide (CHX) chase assay showed prolonged nicotine-induced DX2 protein stability. After incubation with nicotine at 30 μM for 12 h, cells were treated with CHX for the indicated time. Immunoblotting was performed. Actin was used for the loading. (G) The half-life of DX2 Y47F is shorter than that of wild-type DX2. Myc-tagged vectors encoding wild-type DX2 and DX2 Y47F were transfected into the WI-26 cells. After 24 h, cells were treated for the indicated times. Immunoblot analysis was performed. Actin was used for the loading. (H) Src kinase inhibitor (PP2) blocks nicotine-mediated DX2 induction and does not occur in the DX2 Y47F mutant. Myc-tagged vectors encoding wild-type DX2 and DX2 Y47F were transfected into the WI-26 cells. Subsequently, cells were co-treated with PP2 (10 μM) and nicotine (30 μM) for 12 h. Then immunoblot analysis was performed. Actin was used as a loading control. L.E and S.E indicate long exposure and short exposure, respectively.

Figure 2

Induction of HER2 by nicotine in a DX2-dependent manner (A) Expression of HER2 is induced by nicotine. WI-26 cells were treated with nicotine at 30 μM for the indicated time, and immunoblot was performed. Actin was used as a loading control. (B) Cell surface HER2 expression is elevated by nicotine. WI-26 cells were treated with nicotine at 30 μM for 6 h. Subsequently, PFA fixing was conducted without permeabilization. Then, cell surface HER2 expression was measured using flow cytometry. (C) Nicotine induces HER2 expression in SK-N-MC cells. Cells were treated with nicotine at 30 μM for the indicated time, and immunoblot was performed. Actin was used as a loading control. L.E and S.E indicate long exposure and short exposure, respectively. (D) Cell surface HER2 expression is elevated by nicotine. SK-N-MC cells were treated with nicotine at 30 μM for 6 h. (E) Knockdown of DX2 blocks nicotine-mediated HER2 induction. SK-N-MC cells were transiently transfected with control siRNA or DX2-specific siRNA. After 48 h, nicotine was treated. Expression of DX2 and HER2 was measured by western blot, and actin was used as a loading control. (F) Nicotine induces cell surface HER2 expression. Cells were treated with nicotine at 30 μM for 6 h. Subsequently, PFA fixing was conducted without permeabilization. For immunofluorescence (IF) assay, cells were stained with HER2 antibody, and DAPI was used for nuclei staining. Scale bar, 20 μm. (G) Half-life of HER2 is prolonged by nicotine. After being incubated with nicotine at 30 μM for 6 h, CHX was treated for the indicated time. Then, immunoblot analysis was performed. Actin was used as a loading control. L.E and S.E indicate long exposure and short exposure, respectively. (H) β-TrCP reduces HER2 expression. WI-26 cells were transiently transfected with β-TrCP and Siah1. Expression of HER2 was measured by western blot, and actin was used as a loading control. (I) Nicotine inhibits the interaction of HER2 and β-TrCP. To detect the interaction HER2 and β-TrCP, an immunoprecipitation (IP) assay was conducted using HER2 antibody. PPT indicates proteins that co-precipitate with HER2, and SUP indicates the supernatant.

Figure 3

Activation of hedgehog signaling via induction of Shh expression (A) In p14/ARF knockdown conditions, DX2 translocates to cytoplasm. WI-26 cells were transiently transfected with control siRNA or p14/ARF-specific siRNA. After 24 h, Myc-tagged vector encoding wild-type DX2 was transfected to WI-26 cells. For immunofluorescence (IF) assay, cells were stained with anti-Myc antibody, and DAPI was used for nuclei staining. Scale bar, 20 μm. (B) The intensity of rhodamine signal in nucleus or cytoplasm was measured and visualized. Error bars indicate the standard deviations (SD). (C) Nicotine induces Shh expression in WI-26 cells. Cells were treated with nicotine of the indicated concentration for 24 h. Then, immunoblot or RT-PCT analysis was performed. Actin and GAPDH were used as loading controls. (D and E) Nicotine induces Shh expression in SCLC cells (H146 and H69). Cells were treated with nicotine at 30 μM for the indicated time. Then, immunoblot or RT-PCT analysis was performed. Actin and GAPDH were used as loading controls. (F) Exogenous expression of DX2 induces Shh. SK-N-SH cells were transiently transfected by Myc-tagged vector encoding DX2. After 24 h, immunoblot was conducted. Actin was used as a loading control. (G) Nicotine-mediated Shh induction is blocked by elimination of DX2 using siRNA. WI-26 cells were transiently transfected with control siRNA or DX2-specific siRNA. After 48 h, nicotine was treated for 24 h. Then, immunoblot was conducted, and actin was used as a loading control. (H) DX2 inhibitor reduces Shh expression. H446 cells were treated with SLC-B050 for 24 h. Subsequently, immunoblot was conducted. Actin was used as a loading control. (I) Diagram summarizing the experiment. Nicotine treatment was given to neuronal cells (SK-N-MC). After incubation, the conditioned medium was collected and given to WI-26 cells as treatment. Then, WI-26 cells were incubated with conditioned medium for 24 h. (J) In nicotine-treated SK-N-MC cells, expression of Shh is induced. In conditioned-medium-treated cells, expression of cyclin D1 was elevated. SK-N-MC cells were treated with nicotine at 30 μM for the indicated time. Then, conditioned medium was collected and given to WI-26 cells as a treatment. After 24 h, immunoblot was conducted. Actin was used as a loading control.

Figure 4

Enhancement of tumor spheroid formation by nicotine (A) Nicotine facilitates tumor spheroid formation. WI-26 cells were seeded in a non-coated cell plate and treated with nicotine for 7 days. Then, an IF assay was conducted to observe the spheroid. Phalloidin and DAPI were used for actin filaments and nuclei staining. Scale bar, 100 μm. (B) Based on Figure S6A, the diameter of spheroid was measured. ∗∗p < 0.005. Error bars indicate the standard deviations (SD). (C) In nicotine-treated tumor spheroids, DX2, HER2, and Shh expression is elevated. Nicotine treatment was given to WI-26 spheroids for 3 weeks. Subsequently, immunoblot was performed. Actin was used as a loading control. L.E and S.E indicate long exposure and short exposure, respectively. (D and E) Nicotine-mediated tumor spheroid formation is blocked by elimination of DX2. WI-26 cells were transiently transfected with control siRNA or DX2-specific siRNA. After 48 h, cells were transported to a non-coated cell plate and treated with nicotine. After incubation for 3 days, tumor spheroid formation was monitored using a microscope. Scale bar, 1 mm. (F and G) Knockdown of DX2 blocks nicotine-mediated tumor spheroid formation. Based on Figure 4E, the number of spheroids and diameter of spheroids was measured. ∗∗p < 0.005. N.S. indicates no significance. Error bars indicate the standard deviations (SD). (H) Immunohistochemical staining of Ki-67 expression in lung. Compared to vehicle-treated mice, tumor regions were detected in nicotine-treated mice (n = 3). 20×. Scale bar, 100 μm.

Figure 5

Improved effect of SLC-B050 derivatives (A) Some SLC-B050 derivatives inhibit the expression of DX2 and HER2 more effectively at the same concentration. WI-26 cells were transiently transfected by Myc-tagged vector encoding DX2. After 24 h, SLC-B050 and its derivatives were treated at 10 or 20 μM for 24 h. Subsequently, immunoblot was performed. Actin was used as a loading control. (B) SNU-14 more effectively inhibits H446 cell viability than SLC-B050. H446 cells were treated with SLC-B050 or SNU-14 of the indicated concentration for 48 h. Then, cell viability was measured by MTT assay. ∗0.005 < p < 0.05 and ∗∗p < 0.005. Error bars indicate the standard deviations (SD). (C) SNU-14 reduces DX2 expression more effectively than SLC0B050 in H446 cells. Cells were treated with SLC-B050 and SNU-14 of the indicated concentration for 12 h. Subsequently, immunoblot was performed. Actin was used as a loading control. (D) Cell viability measured by MTT assay indicates that SNU-14 was more effective than SLC-B050 for SCLC treatment. After treatment with chemicals of the indicated concentration for 96 h in serum-containing conditions, cell viability was measured by MTT assay (n = 4 independent experiment; two-tailed Student’s t test). ∗0.005 < p < 0.05 and ∗∗p < 0.005. Error bars indicate the standard deviations (SD). (E) Cell death of H146 occurred more excessively with SNU-14. After treated with chemicals of the indicated concentration for 96 h, using MTT solution, formazan crystal was observed (n = 4 independent experiment; two-tailed Student’s t test). (F) Both SLC-B050 and SNU-14 do not show cytotoxicity to normal fibroblasts. Cells were treated with SLC-B050 and SNU-14 of the indicated concentration for 48 h. Then, cell viability was measured by MTT assay (n = 3 independent experiment; two-tailed Student’s t test). ∗∗p < 0.005. N.S. indicates no significance.

Figure 6

Improved anti-cancer effect of DX2 inhibitor (A) Tumor spheroid formation is blocked by treatment of SLC-B050 and SNU-14. WI-26 cells were seeded to non-coated cell plate, and SLC-B050 or SNU-14 treatment at 10 μM was given to these cells. After incubation for 7 days, tumor spheroid formation was monitored using a microscope. (B) DX2 inhibitors effectively block tumor spheroid formation. Based on Figure 6A, the number of spheroids was counted. ∗0.005 < p < 0.05 and ∗∗p < 0.005. Error bars indicate the standard deviations (SD). (C) SNU-14 generates smaller tumor spheroids than SLC-B050. Based on Figure 31A, the diameter of the spheroid was measured. Error bars indicate the standard deviations (SD). (D) SNU-14 blocks nicotine-mediated HER2 induction. Cells were treated with SNU-14 at 2 μM and nicotine at 30 μM for 24 h. Then, immunoblot was conducted. Actin was used as a loading control. Scale bar, 20 μm. (E) SNU-14 reduces HER2 cell surface expression induced by nicotine. Cells were treated with SNU-14 at 2 μM and nicotine at 30 μM for 24 h. Cells were fixed with 4% PFA and incubated with HER2 antibody. DAPI was used for nuclei staining. (F) Summary diagram. DX2 is stabilized by nicotine, which is triggered by Src kinase through nAChR-mediated Ca²⁺ signaling. In the p14/ARF-intact state, DX2, stabilized by nicotine, transports to the nucleus. DX2 in the nucleus inhibits p14/ARF and activates Hh signaling. It results in cell death or proliferation of its own or surrounding cells and, consequently, promotes SCLC. In p14/ARF-null cells, DX2 exists in the cytoplasm and blocks the interaction of β-TrCP/HER2. Increased HER2 localizes on the cell surface or transports to the nucleus, probably to act as a transcription factor. Consequently, it causes NSCLC. The optimized DX2 inhibitor SNU-14 may inhibits the progression of lung cancer, particularly SCLC.