γ-secretase inhibitors, DAPT and RO4929097, promote the migration of Human Glioma Cells via Smad5-downregulated E-cadherin Expression

Abstract

Malignant gliomas are a type of central nervous system cancer with extremely high mortality rates in humans. γ-secretase has been becoming a potential target for cancer therapy, including glioma, because of the involvement of its enzymatic activity in regulating the proliferation and metastasis of cancer cells. In this study, we attempted to determine whether γ-secretase activity regulates E-cadherin to affect glioma cell migration. The human glioma cell lines, including LN18 and LN229, and the γ-secretase inhibitors, including N-[N-(3,5-difluorophenacetyl)-L-alanyl]-S-phenylglycine t-butyl ester (DAPT) and RO4929097, were used in this study. It was shown that γ-secretase activity inhibition by DAPT and RO4929097 could promote LN18 and LN229 glioma cell migration via downregulating E-cadherin mRNA and protein expressions, but not via affecting E-cadherin protein processing. In addition, γ-secretase activity inhibition was regulated by bone morphogenetic proteins-independent Smad5 activation in glioma cells. Moreover, endogenous Smad1 in glioma cells was found to play an important role in regulating E-cadherin expression and subsequent cell migration but did not affect DAPT-stimulated effects. These results help further elucidate the molecular mechanisms of γ-secretase activity regulation involved in controlling glioma cell malignancy. Information about a potential role for Smad1/5 activity upregulation and subsequent E-cadherin downregulation during inhibition of γ-secretase activity in the development of gliomas is therefore relevant for future research.

Keywords: E-cadherin; Smad1/5; bone morphogenetic proteins; glioma; γ-secretase.

Figure 1

DAPT and RO4929097, γ-secretase activity inhibitors, inhibits E-cadherin expression in LN18 and LN229 glioma cells. (A-B) LN18 glioma cells were kept as controls (CL) or treated with DAPT (20, 40, or 80 µM) for 24 h or 48 h and then the protein (A) and/or mRNA (B) levels of N-/E-cadherin were examined by western blot and real-time PCR, respectively. (C-D) LN18 (C-left panels and D) and LN229 (C-right panels) glioma cells were kept as controls (CL) or treated with 1 µM RO4929097 (RO) or 80 µM DAPT (D) for 48 h and then the E-cadherin protein levels (C) and cellular localization (D) were examined by western blot and cell fractionation, respectively. Results in (A, C-D) are representative of three independent experiments. Data in (A-D) are shown as mean ± SD from three independent experiments. *, P < 0.05 vs. LN18-untreated control (CL) cells. &, P < 0.05 vs. LN229-untreated control (CL) cells.

Figure 2

DAPT and RO4929097 promotes LN18 and LN229 glioma cell migration by chemotaxis method. (A) LN18 glioma cells were kept as controls (CL) or treated with 80 µM DAPT for 24, 48, 72, or 96 h and (B) LN18 (upper panels) and LN229 (down panels) glioma cells were kept as controls (CL) or treated with 1 µM RO4929097 (RO) or 80 µM DAPT for 96 h and cell migration were examined by chemotaxis method (transwell). Data are shown as mean ± SD from three independent experiments. *, P < 0.05 vs. LN18-untreated control (CL) cells. &, P < 0.05 vs. LN229-untreated control (CL) cells.

Figure 3

DAPT and RO4929097 promotes LN18 and LN229 glioma cell migration by wound-healing method. (A-B) LN18 (A) and LN229 (B) glioma cells were kept as controls (CL) or treated with 80 µM DAPT or 1 µM RO4929097 for 48 h and cell migration were examined by wound-healing method. Data are shown as mean ± SD from three independent experiments. *, P < 0.05 vs. LN18-untreated control (CL) cells. &, P < 0.05 vs. LN229-untreated control (CL) cells.

Figure 4

DAPT-inhibited E-cadherin expression is regulated by BMP-independent Smad5 activation. (A) LN18 glioma cells were kept as controls (CL) or treated with 80 µM DAPT for 0.5 h, 1 h, 4 h, 8 h, or 24 h and then Smad1/5 phosphorylation was examined by western blot. (B-C) LN18 glioma cells were pretreated with control (siCL), Smad1 (siSmad1)-, or Smad5 (siSmad5)-specific siRNA and then were kept as controls (CL) or treated with 80 µM DAPT (D) for 48 h. (B) E-cadherin/Smad1/5 protein and (C) E-cadherin mRNA levels were examined by western blot and real-time PCR, respectively. (D) LN18 glioma cells were pretreated with DMSO or Noggin (BMPs-specific antagonist) for 1 h and then were kept as controls (CL) or treated with 80 μM DAPT (D) for 48 h. The E-cadherin expression was examined by Western blot. Results in (A-B, D) are representative of three independent experiments. Data in (A-D) represent mean ± SD from three independent experiments. *, P < 0.05 vs. untreated control (CL) cells (A), siCL-CL-treated cells (B-C), or DMSO-CL-treated cells (D). **, P < 0.05 vs. siSmad1-CL-treated cells (B-C). #, P < 0.05 vs. siCL-DAPT (D)-treated cells (B-C).

Figure 5

Smad5 signaling is involved in DAPT-promoted LN18 glioma cell migration. (A-B) LN18 glioma cells were pretreated with control (siCL)-, Smad1 (siSmad1)- or Smad5 (siSmad5)-specific siRNA and then were kept as controls (CL) or treated with 80 μM DAPT (D) for 48 h (wound-healing) or 96 h (chemotaxis). Cell migration was examined by chemotaxis (transwell) (A) and wound-healing (B) methods. Data in (A-B) are mean ± SEM from three independent experiments. *, P < 0.05 vs. siCL-CL-treated cells. **, P < 0.05 vs. siSmad1-CL-treated cells. #, P < 0.05 vs. siCL-DAPT (D)-treated cells.

Figure 6

Endogenous E-cadherin level affects LN18 and LN229 glioma cell migration. (A-B) LN18 and LN229 glioma cells were treated with control (siCL) or E-cadherin (siE-cadherin)-specific siRNA for 96 h and then cell migration and endogenos E-cadherin mRNA expression were examined by chemotaxis methods and real-time PCR, respectively. Data in (A-B) are mean ± SEM from three independent experiments. *, P < 0.05 vs. LN18-siCL-treated cells. &, P < 0.05 vs. LN229-siCL-treated cells.

Figure 7

Schematic representation of signaling pathways regulating DAPT/RO4929097-promoted cell migration in human LN18/LN229 glioma cells.