Valproic acid upregulates the expression of the p75NTR/sortilin receptor complex to induce neuronal apoptosis

Abstract

The antiepileptic and mood stabilizer agent valproic acid (VPA) has been shown to exert anti-tumour effects and to cause neuronal damage in the developing brain through mechanisms not completely understood. In the present study we show that prolonged exposure of SH-SY5Y and LAN-1 human neuroblastoma cells to clinically relevant concentrations of VPA caused a marked induction of the protein and transcript levels of the common neurotrophin receptor p75NTR and its co-receptor sortilin, two promoters of apoptotic cell death in response to proneurotrophins. VPA induction of p75NTR and sortilin was associated with an increase in plasma membrane expression of the receptor proteins and was mimicked by cell treatment with several histone deacetylase (HDAC) inhibitors. VPA and HDAC1 knockdown decreased the level of EZH2, a core component of the polycomb repressive complex 2, and upregulated the transcription factor CASZ1, a positive regulator of p75NTR. CASZ1 knockdown attenuated VPA-induced p75NTR overexpression. Cell treatment with VPA favoured proNGF-induced p75NTR/sortilin interaction and the exposure to proNGF enhanced JNK activation and apoptotic cell death elicited by VPA. Depletion of p75NTR or addition of the sortilin agonist neurotensin to block proNGF/sortilin interaction reduced the apoptotic response to VPA and proNGF. Exposure of mouse cerebellar granule cells to VPA upregulated p75NTR and sortilin and induced apoptosis which was enhanced by proNGF. These results indicate that VPA upregulates p75NTR apoptotic cell signalling through an epigenetic mechanism involving HDAC inhibition and suggest that this effect may contribute to the anti-neuroblastoma and neurotoxic effects of VPA.

Keywords: Apoptosis; Histone deacetylase inhibitors; Human neuroblastoma cells; Mouse cerebellar granule cells; Sortilin; p75NTR.

Fig. 1

VPA upregulates p75NTR and sortilin expression in human neuroblastoma cells. SH-SY5Y and LAN-1 cells were treated for 24 h with either vehicle or 1 mM VPA and the expression of p75NTR (a and d) and sortilin (b and e) was analysed by Western blot and normalised to actin. Values are the mean ± SD of five (a and d) and four (b and e) independent experiments. c and f quantitative real-time RT-PCR analysis of p75NTR and sortilin mRNA levels in SH-SY5Y and LAN-1 cells, treated for 24 h with either vehicle or 1 mM VPA. Values are the mean ± SD of four independent determinations. *p < 0.05, **p < 0.01 vs. control (vehicle-treated cells)

Fig. 2

VPA enhances the plasma membrane expression of p75NTR and sortilin. a SH-SH5Y and LAN-1 cells were treated with either vehicle or VPA (1 mM) for 24 h and then exposed to the cell impermeant biotinylating agent sulpho-NHS-LC-biotin. The isolated surface proteins were analysed for p75NTR and sortilin by Western blot. The levels of p75NTR and sortilin in the cell surface protein preparation were normalised to the corresponding levels of pan-cadherin, a plasma membrane marker. Densitometric values are expressed as fold increase with respect to vehicle and are the mean ± SD of four independent experiments. b The expression of p75NTR was analysed by immunofluorescence (green color) in non-permeabilised SH-SY5Y and LAN-1 cells with an antibody directed against an extracellular domain of the receptor. Nuclei were stained in blue with DAPI. Values are the mean ± SD of four separate experiments. Bar = 25 µm. *p < 0.05 vs. control (vehicle) (Color figure online)

Fig. 3

Time- and concentration-dependent upregulation of p75NTR and sortilin by VPA. a, b: SH-SY5Y and LAN-1 cells were incubated in the presence of 1 mM VPA for the indicated periods of time. Zero time samples were incubated with vehicle and used as control. c-f: SH-SY5Y (c, e) and LAN-1 (d, f) were incubated for 24 h in the presence of either vehicle or VPA at the indicated concentrations. Cell lysates were analysed for p75NTR and sortilin expression by Western blot. Values are the mean ± SD of four independent experiments. *p < 0.05 vs. control

Fig. 4

Effects of HDAC inhibitors on p75NTR and sortilin expression. SH-SY5Y (a, b, e and f ) and LAN-1 (c, d, g and h) cells were incubated for 24 h with either vehicle, 1 mM VPA, 1 µM entinostat, 1 mM sodium butyrate (NaButyr), 300 nM trichostatin A (TSA), 10 µM MC1568, 30 nM romidepsin, 5 µM tubacin, or 5 µM PCI-34,051. Cell lysates were analysed for p75NTR and sortilin expression. Values are the mean ± SD of four (a, b, e and f) and six (c, d, g and h) independent experiments. *p < 0.05, **p < 0.01 vs. control (vehicle-treated cells)

Fig. 5

VPA-induced p75NTR upregulation is mimicked by HDAC1 knockdown and involves CASZ1 derepression. a SH-SY5Y cells were transfected with either control siRNA or HDAC1 siRNA duplexes and cell lysates were analysed for HDAC1 and p75NTR expression 48 h post-transfection. Values are the mean ± SD of four separate experiments. *p < 0.05 vs. the corresponding sample treated with control siRNA. b, c Cells treated for 24 h with either vehicle or 1 mM VPA (b) or transfected with either control or HDAC1 siRNAs (c) were analysed for the levels of EZH2 and CASZ1 by Western blot. Values are the mean ± SD of four separate experiments. *p < 0.05 vs. control (vehicle) or the corresponding sample treated with control siRNA. d Cells were treated for 72 h with either vehicle, deazaneplanocin (a) (DZNep) (1 µM) or tazemetostat (TZM) (1 µM). The levels of p75NTR were measured by Western blot and normalised to actin. Values are the mean ± SD of four independent experiments. *p < 0.05 vs. vehicle. e Cells were transfected with either control siRNA or CASZ1 siRNA and 45 h post-transfection treated with either vehicle or 1 mM VPA for 24 h. Cell lysates were analysed for the expression of CASZ1 and p75NTR by Western blot. Values are the mean ± SD of four experiments. *p < 0.05 vs. vehicle in control siRNA-treated cells; ^#p < 0.05 vs. VPA in control siRNA-treated cells

Fig. 6

Exposure to VPA promotes proNGF-induced p75NTR/sortilin interaction and JNK activation. a SH-SY5Y cells were incubated for 24 h with either vehicle or 1 mM VPA and then treated for 2 h with either vehicle or proNGF (1 µg/ml). Thereafter, cell lysates were subjected to immunoprecipitation with an anti-p75NTR antibody. Immunoprecipitates and cell lysates (input) were analysed for sortilin and p75NTR immunoreactivities. The immunoblot is representative of three independent experiments. b SH-SY5Y cells were incubated for 24 h with either vehicle or 1 mM VPA and then exposed to either vehicle or proNGF (5 ng/ml) for 3 and 6 h. Cell lysates were analysed for phospho-JNK (pJNK) and JNK levels. Values are the mean ± SD of four independent experiments. *p < 0.05 vs. control (vehicle + vehicle); ^#p < 0.05. c SH-SY5Y cells were incubated for 24 h with either vehicle or 1 mM VPA and then treated for 6 h with either vehicle or 5 ng/ml proNGF. Cell lysates were analysed for the expression of phospho-c-Jun (p-c-Jun) and c-Jun. d SH-SY5Y cells were treated as indicated in C and analysed for phospho-c-Jun expression (green color) by immunofluorescence microscopy. Cell nuclei were stained in blue with DAPI. Values indicate the percent of phospho-c-Jun positive cells and are the mean ± SD of four independent experiments. Bar = 50 µm. *p < 0.05 vs. control (vehicle + vehicle); ^#p < 0.05 (Color figure online)

Fig. 7

Cell treatment with proNGF potentiates VPA-induced apoptosis. A SH-SY5Y cells grown onto glass coverslips were incubated for 24 h with either vehicle or 1 mM VPA and then exposed for additional 24 h to either vehicle or 5 ng/ml proNGF. Dead cells were identified by propidium iodide fluorescence (red color), whereas cell nuclei were stained with DAPI (blue color). Values are the mean ± SD of four experiments. Bar = 100 µm. B LAN-1 cells were grown, treated and analysed as indicated in A. C SH-SY5Y and LAN-1 cells were incubated for 24 h with either vehicle or 1 mM VPA and then exposed for 24 h to either vehicle or 5 ng/ml proNGF. Cell lysates were analysed for cleaved caspase (cleav casp) 9, procaspase (procasp) 9, cleaved caspase 3, procaspase 3, cleaved PARP and total PARP. Values are the mean ± SD of five independent experiments. D, E SH-SY5Y (D) and LAN-1 (E) cells grown onto glass coverslips were treated as indicated in C and then processed for cleaved caspase 3 immunofluorescence (green color) analysis. a = vehicle; b = proNGF; c = VPA; d = VPA + proNGF. Positive cells are expressed as percent of total cells. Values are the mean ± SD of four independent experiments. Bar = 50 µm. *p < 0.05 vs. vehicle. ^#p < 0.05 (Color figure online)

Fig. 8

Involvement of p75NTR/sortilin receptor complex in the apoptosis induced by VPA and pro-NGF. a SH-SY5Y cells were transfected with either control siRNA or p75NTR siRNA duplexes. Forty-eight h post-transfection cells were incubated for 24 h with either vehicle or 1 mM VPA and then exposed to either vehicle or 5 ng/ml proNGF for additional 24 h. Cell lysates were analysed for p75NTR, phospho-JNK and cleaved PARP expression. Values are the mean ± SD of four independent experiments. *p < 0.05 vs. vehicle in control siRNA-treated cells. ^a p < 0.05; ^#p < 0.05 vs. the corresponding sample in control siRNA-treated cells. b SH-SY5Y cells were pretreated for 24 with either vehicle or 1 mM VPA. Thereafter, the cells were first exposed to either vehicle or 5 µM neurotensin (NT) for 2 h and then to vehicle or 5 ng/ml proNGF for additional 22 h. Cell lysates were analysed for cleaved and uncleaved PARP. Densitometric ratios are expressed as percent of PARP cleavage induced by VPA in the absence of NT and are the mean ± SD of four independent experiments. *p < 0.05 vs. control (vehicle + vehicle). ^a p < 0.05; ^#p < 0.05 vs. the corresponding sample in vehicle-treated cells

Fig. 9

VPA upregulates p75NTR and sortilin expression and promotes proNGF-induced apoptosis in mouse cerebellar granule cells. a cells were incubated for 24 h with either vehicle, 1 mM VPA or 1 µM entinostat and cell lysates were analysed for p75NTR and sortilin expression. Values are the mean ± SD of four independent experiments. b cells were incubated for 24 h with either vehicle or 1 mM VPA and then exposed for additional 24 h to either vehicle or 5 ng/ml proNGF. Cells were analysed for cleaved caspase 3 (green color) and neurofilament 160/200 (red color) by immunofluorescence microscopy. Nuclei were stained with DAPI (blue color). Bar = 50 µm. Values are the mean ± SD of four experiments. *p < 0.05 vs. vehicle. ^#p < 0.05 (Color figure online)

Fig. 10

Schematic diagram illustrating the induction of p75NTR and sortilin by VPA and the consequent promotion of proNGF-induced neuronal apoptosis. The inhibition of HDACs is shown to induce the transcription of NGFR and SORT1 genes, leading to an enhanced p75NTR and sortilin expression at the plasma membrane. These receptor changes sensitise neuronal cells to the proapototic action of proNGF through the activation of JNK signalling