Overexpression of ceramide synthase 1 increases C18-ceramide and leads to lethal autophagy in human glioma

Abstract

Ceramide synthase 1 (CERS1) is the most highly expressed CERS in the central nervous system, and ceramide with an 18-carbon-containing fatty acid chain (C18-ceramide) in the brain plays important roles in signaling and sphingolipid development. However, the roles of CERS1 and C18-ceramide in glioma are largely unknown. In the present study, measured by electrospray ionization linear ion trap mass spectrometry, C18-ceramide was significantly lower in glioma tumor tissues compared with controls (P < 0.001), indicating that C18-ceramide might have a role in glioma. These roles were examined by reconstitution of C18-ceramide in U251 and A172 glioma cells via addition of exogenous C18-ceramide or overexpression of CERS1, which has been shown to specifically induce the generation of C18-ceramide. Overexpression of CERS1 or adding exogenous C18-ceramide inhibited cell viability and induced cell death by activating endoplasmic reticulum stress, which induced lethal autophagy and inhibited PI3K/AKT signal pathway in U251 and A172 glioma cells. Moreover, overexpression of CERS1 or adding exogenous C18-ceramide increased the sensitivity of U251 and A172 glioma cells to teniposide (VM-26). Thus, the combined therapy of CERS1/C18-ceramide and VM-26 may be a novel therapeutic strategy for the treatment of human glioma.

Keywords: C18-ceramide; autophagy; ceramide synthase 1; glioma; mass spectrometry.

Figure 1. Qualitative and quantitative analysis of C18-ceramide in human glioma tissue samples

(A) MS2 spectrum of m/z 630, characteristic fragmentation products (m/z 278.3) for permethylated C18-ceramide in the human glioma tissue samples in positive-mode MS2. (B) Fragmentation pathway and characteristic decomposition products for permethylated C18-ceramide in the positive mode. (C) MS1 profile of C18-ceramide isolated from a control tissue sample; the expression of C18-ceramide (m/z 630) was high. (D) MS1 profile of C18-ceramide isolated from a glioma tissue sample; the expression of C18-ceramide (m/z 630) was low. (E) Relative quantification of C18-ceramide (m/z 630) in the tissue samples of controls and glioma. Data represent the tissue samples from controls (n = 5) and glioma (n = 14). Statistical significance between glioma and controls was analyzed using the two-tailed Student’s t-test of means. Compared with control, *** P < 0.001.

Figure 2. Inhibition of cell viability and promotion of cell death induced by overexpression of CERS1 and exogenous C18-ceramide in U251 and A172 glioma cells

(A) Effect of catalytically inactive CERS1 (H138A) and CERS1 overexpression on the cell viability of U251 and A172 cells for 48h. (B) Effect of exogenous C18-ceramide (20 μM) on the cell viability of U251 and A172 cells for 48h. (C) Effect of catalytically inactive CERS1 (H138A) and CERS1 overexpression on the cell death of U251 and A172 cells for 48h. (D) Effect of exogenous C18-ceramide (20 μM) on the cell death of U251 and A172 cells for 48h. (E) Quantitative analysis of catalytically inactive CERS1 (H138A) and CERS1 overexpression on the cell death of U251 and A172 cells for 48h. (F) Quantitative analysis of exogenous C18-ceramide (20 μM) on the cell death of U251 and A172 cells for 48h. Statistical significance between CERS1/C18-ceramide and controls was analyzed using the two-tailed Student’s t-test of means. Values represent the means ± SD, n = 3 independent experiments. Compared with control, *P < 0.05, **P < 0.01.

Figure 3. Activation of ER stress induced by overexpression of CERS1 in U251 and A172 glioma cells

(A) The fold change of DDIT4, PIK3CA, MAP1LC3β, ATF-6, ATF-3, CHOP, XBP-1, and ATF-4 in CERS1 overexpression in U251 cells compared with controls. (B) The qRT-PCR results of ATF-4, XBP-1 (s), and CHOP mRNA levels in CERS1 overexpression in U251 cells compared with controls. (C) The qRT-PCR results of ATF-4, XBP-1 (s), and CHOP mRNA levels in CERS1 overexpression in A172 cells compared with controls. (D) The qRT-PCR results of ATF-4, XBP-1 (s), and CHOP mRNA levels in exogenous C18-ceramide (20 μM) in U251 cells compared with controls. (E) The qRT-PCR results of ATF-4, XBP-1 (s), and CHOP mRNA levels in exogenous C18-ceramide (20 μM) in A172 cells compared with controls. Statistical significance between CERS1/C18-ceramide and controls was analyzed using the two-tailed Student’s t-test of means. Values represent the means ± SD, n = 3 independent experiments. Compared with control, *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 4. Activation of ER stress induced by overexpression of CERS1

(A) Western blot analysis was performed to detect the expression levels of ATF-4, XBP-1 (s), and CHOP and β-actin in U251 and A172 cells after overexpression of CERS1 compared with controls. (B) ATF-4 signals were normalized to β-actin for relative quantification in U251 and A172 cells after overexpression of CERS1 compared with controls. (C) XBP-1 (s) signals were normalized to β-actin for relative quantification in U251 and A172 cells after overexpression of CERS1 compared with controls. (D) CHOP signals were normalized to β-actin for relative quantification in U251 and A172 cells after overexpression of CERS1 compared with controls. (E) Western blot analysis was performed to detect the expression levels of p-PERK, PERK, p-eIF2α, eIF2α, and β-actin in U251 and A172 cells after overexpression of CERS1 compared with controls. (F) p-PERK signals were normalized to PERK for relative quantification in U251 and A172 cells after overexpression of CERS1 compared with controls. (G) p-eIF2α signals were normalized to eIF2α for relative quantification in U251 and A172 cells after overexpression of CERS1 compared with controls. (H) Effect of loss of CHOP function (using CHOP siRNA) on cell viability of overexpression of CERS1 in U251 and A172 cells for 48h. Statistical significance was analyzed using the two-tailed Student’s t-test of means. Values represent the means ± SD, n = 3 independent experiments. *P < 0.05, **P < 0.01.

Figure 5. Lethal autophagy induced by overexpression of CERS1 in U251 and A172 cells

(A) Western blot analysis was performed to detect the expression levels of LC3B, p62, and β-actin in U251 and A172 cells after overexpression of CERS1. (B) LC3B-II signals were normalized to LC3B-I for relative quantification in U251 and A172 cells after overexpression of CERS1 compared with controls. (C) p62 signals were normalized to β-actin for relative quantification in U251 and A172 cells after overexpression of CERS1 compared with controls. (D) Overexpression of CERS1 or exogenous C18-ceramide (20 μM) increased the number of GFP–LC3 puncta (arrows) were visualized using confocal microscopy in U251 and A172 cells. Scale bar = 10 μm. (E) Quantification of GFP puncta per cell were performed in U251 and A172 cells. (F) Formation of double-membrane autophagosomal vesicles (black arrows) in the control or overexpression of CERS1 was visualized by TEM (left and right, respectively) in U251 cells. Higher magnification of TEM visualization is shown in lower panels. Scale bars, 1 μm (top) and 0.5 μm (bottom). (G) Effect of loss of LC3 function (using LC3 siRNA) on cell viability of overexpression of CERS1 in U251 and A172 cells for 48h. (H) 3-MA (5 mM) effect on cell viability of overexpression of CERS1 in U251 and A172 cells for 48h. Statistical significance was analyzed using the two-tailed Student’s t-test of means. Values represent the means ± SD, n = 3 independent experiments. *P < 0.05, **P < 0.01.

Figure 6. CERS1 inhibits PI3K/AKT pathway in U251 and A172 cells

(A) Western blot analysis was performed to detect the expression levels of PI3K, p-AKT (S473), AKT, and β-actin in U251 and A172 cells after overexpression of CERS1. (B) PI3K signals were normalized to β-actin for relative quantification in U251 and A172 cells after overexpression of CERS1 compared with controls. (C) p-AKT (S473) signals were normalized to AKT for relative quantification in U251 and A172 cells after overexpression of CERS1 compared with controls. (D) Effect of IGF-1 (20 ng/mL) on cell viability of overexpression of CERS1 in U251 and A172 cells for 48h. Statistical significance was analyzed using the two-tailed Student’s t-test of means. Values represent the means ± SD, n = 3 independent experiments. *P < 0.05, **P < 0.01.

Figure 7. Overexpression of CERS1 and exogenous C18-ceramide increases the chemosensitivity to VM-26 in U251 and A172 glioma cells

(A) Synergistic interactions of overexpression of CERS1 and VM-26 in the inhibition of cell viability were examined by quantitative isobologram studies. The IC50 concentrations of CERS1 plasmid in the presence of increasing concentrations of VM-26 were determined by CCK-8 assays, and the data were plotted in isobolograms. A straight line joining points on x- and y-axes represents the IC50 concentrations of transfection CERS1 plasmid and VM-26 alone. The points on the isobologram representing the IC50 values of VM-26 obtained in the transfection of 100, 200, and 400 ng/μl CERS1 plasmid fell within the left of the straight line, which indicates synergism. (B) Synergistic interactions of exogenous C18-ceramide and VM-26 in the inhibition of cell viability were examined by quantitative isobologram studies. The IC50 concentrations of exogenous C18-ceramide in the presence of increasing concentrations of VM-26 were determined by CCK-8 assays, and the data were plotted in isobolograms. A straight line joining points on x- and y-axes represents the IC50 concentrations of exogenous C18-ceramide and VM-26 alone. The points on the isobologram representing the IC50 values of VM-26 obtained in the present of 4, 8, and 16 μM exogenous C18-ceramide fell within the left of the straight line, which indicates synergism. (C) Effect of CERS1 overexpression and VM-26 treatment on the cell death of U251 (0.25 μM VM-26) and A172 (10 μM VM-26) cells for 48h. (D) Effect of exogenous C18-ceramide (20 μM) and VM-26 treatment on the cell death of U251 (0.25 μM VM-26) and A172 (10 μM VM-26) cells for 48h. (E) Quantitative analysis of CERS1 overexpression and VM-26 treatment on cell death of U251 (0.25 μM VM-26) and A172 (10 μM VM-26) cells for 48h. (F) Quantitative analysis of exogenous C18-ceramide (20 μM) on cell death of U251 (0.25 μM VM-26) and A172 (10 μM VM-26) cells for 48h. (G) Cell growth of control, VM-26 group, VM-26/CERS1 group and VM-26/C18-ceramide (20 μM) group in U251 (0.25 μM VM-26) and A172 (10 μM VM-26) cells for 48h detected by BrdU incorporation assay. Statistical significance between VM-26 CERS1/C18-ceramide group and VM-26 group was analyzed using the two-tailed Student’s t-test of means. Values represent the means ± SD, n = 3 independent experiments. Compared with VM-26, *P < 0.05, **P < 0.01, ***P < 0.001.