α-Synuclein upregulates bim-mediated apoptosis by negatively regulating endogenous GCN5

Overexpression of αS reduces cell viability

Several studies indicated that ectopic αS is a major hallmark of neuronal cell death in PD pathology [23] and that αS overexpression promotes cell death via changing sub-cellular localization [12]. Therefore, we investigated whether our αS (human) plasmid overexpression (Figure 1A) affected the viability of SH-SY5Y neuroblastoma (Figure 1B, 1C). We used DNA fragmentation (DNA ladder) and DAPI staining assay of transiently transfected vectors or αS plasmids. After 48 hours of transfection, αS showed significant chromatin aggregation and increased DNA fragmentation. αS at the nucleus masks H3 histone and inhibits H3 activities at the chromatin level [12]. Also, overexpression of sporadic or familial (A53T and A30P) αS phosphorylates at ser129 reduces cell viability via aggregation [24]. Corroborating previous findings, our data indicate that ectopic αS regulates neuronal cell fate.

Figure 1. αS overexpression and toxicity. (A) SH-SY5Y neuroblastoma cells were transfected by human αS plasmid. (B) SH-SY5Y cells were transiently transfected by αS, and post-transfection incubated for an additional 24 h and subjected to test cell viability by DNA laddering and nucleic staining by DAPI (C) where the whitish inclusions (cropped) indicated apoptotic cells with nuclear pyknosis; the pyknosis rate (% chromatin aggregation) was calculated by the scoring percent of cells with nuclear pyknosis in total DAPI-stained cells (scale bar = 40 μm). ^###p < 0.001 compared between vector (vec) vs. αS^+/+.

Overexpression of αS activates bim-mediated apoptotic signaling

The BH3-only protein Bim is a pro-apoptotic protein that has been previously shown to upregulate during apoptosis. Bim induces apoptosis by binding to Bcl-2 or Bcl-xl and antagonizes their anti-apoptotic functions. In the αS^+/+ model, both the mRNA and protein levels of Bim increased substantially (Figure 2A, 2B). A previous study demonstrated that Bim genes are transcriptionally regulated by Egr-1 and E2F1 [17, 25]. Therefore, we tested the impact of αS overexpression on Bim transcriptional factors E2F1 and Egr-1 (Figure 2A, 2B). We found that, in both mRNA and protein levels, overexpression of αS significantly upregulated E2F1 but not Egr-1. This was further corroborated by co-immunoprecipitating both transcription factors in non-transfected SH-SY5Y cells (Figure 2C). Immunoprecipitation showed that anti-E2F1 endogenously precipitated anti-αS more intensely than anti-Egr-1. Moreover, Bim is an upstream regulator of the caspase-3-mediated apoptotic signaling pathway. αS overexpression increases the Bim level that inhibits Bcl-2, frees Bax, and activates caspase-3 (Figure 2D).

Figure 2. Overexpression of αS promotes Bim-mediated apoptotic pathway. (A) mRNA and (B) protein expression of Bim and its transcription factors in non-transfected cells (NC) and transfected (αS^+/+) cells. (C) SH-SY5Y cells were co-immunoprecipitated by anti-E2F1 and anti-Egr1 or A/G beads only (No 1⁰ ab) as negative control and probed for anti-αS. (D) Protein expression of apoptotic markers.

αS Inhibits endogenous GCN5 expression

A recent study reported that GCN5 transcriptionally regulates Bim and downstream caspase-3 activation [17]. Though αS interacts and inhibits neuronally expressed acetyltransferase enzymes CBP, p300, and P/CAF [12, 26], it is generally unknown whether αS and GCN5 endogenously interact. To confirm endogenous binding, we co-immunoprecipitated GCN5 in non-transfected SH-SY5Y cells and probed anti-αS (Figure 3A). Furthermore, we evaluated the impact of αS overexpression on endogenous GCN5 expression in neuronal SH-SY5Y (Figure 3B) and non-neuronal HEK293T cells (Figure 3C). A significant decrease, compared to usual, in GCN5 expression in αS^+/+ indicates that GCN5 may be involved in αS-mediated Bim upregulation.

Figure 3. αS^+/+ inhibiting endogenous GCN5 and inhibiting GCN5 expression. (A) co-immunoprecipitation of GCN5 in wildtype SH-SY5Y and probed for anti-αS. Endogenous protein expression of GCN5 in neuronal SH-SY5Y cells (B) and non-neuronal HEK293T cells (C) transfected (αS) or NC.

Inhibition of GCN5 triggers αS overexpression-mediated bim upregulation

As we found αS^+/+ inhibits GCN5 and regulates pro-apoptotic Bim, we tested the hypothesis that GCN5 is involved in Bim’s transcriptional regulation. To test this hypothesis, we co-immunoprecipitated GCN5 and E2F1 and probed anti-αS (Figure 4A). Since the loss of GCN5 upregulates both transcription factors of Bim [17], our results indicate that ectopic αS interacts and inhibits GCN5 and that this upregulates the E2F1 level. Next, we checked whether the knockdown of GCN5 by siRNA affects endogenous αS expression (Figure 4B); however, no significant changes were observed. siGCN5 substantially upregulates Bim, Egr-1, and E2F1 levels but decreases Bcl2 expression in neuronal cells (Figure 4C). These data suggest that GCN5 could be a potential target to reduce neuronal apoptosis in PD and other synucleinopathy.

Figure 4. Inhibition of GCN5 by αS triggering Bim expression. (A) Co-immunoprecipitation of GCN5 and E2F1 in neuronal SH-SY5Y cells and probed for anti-αS. (B) GCN5 siRNA was used to knockdown GCN5 in SH-SY5Y cells; protein lysates were collected 24 h after siGCN5 induction and probed for anti-GCN5 and anti-αS. (C) Protein expression of apoptotic and anti-apoptotic markers.

Targeting GCN5 ameliorates αS-mediated apoptosis

Until now, this study has indicated that αS overexpression dismantles GCN5 homeostasis, which triggers apoptosis in neuronal cells. Therefore, we hypothesized that GCN5 homeostasis by co-activation could benefit PD patients by ameliorating apoptosis. First, we tested whether GCN5 overexpression (oeGCN5) modulates αS endogenous levels (Figure 5A) and found that oeGCN5 downregulated endogenous αS expression. Later, we co-overexpressed αS and GCN5 (αS+GCN5) in SH-SY5Y cells (Figure 5B) and tested for an apoptotic marker expression (Figure 5C). The co-overexpression of αS and GCN5 were found to substantially reduce transcriptional upregulation of Bim and improve Bcl2 levels, which reduced caspase-3 activation. Subsequently, co-overexpression downregulated neuronal cell death (Figure 5D).

Figure 5. Co-overexpression of αS-GCN5 inhibits apoptotic neuronal cell death. (A) Endogenous protein expression of αS in GCN5 overexpressed (oeGCN5) or NC cells. (B) Protein was isolated from NC, αS^+/+ and αS+GCN5 cells and probed for specific proteins. (C) Protein expression of apoptotic markers using lysates from NC, αS^+/+, oeGCN5 and αS+GCN5 cells. (D) Nucleic staining by DAPI where the whitish inclusions (cropped) indicated apoptotic cells with nuclear pyknosis; the pyknosis rate (% chromatin aggregation) was calculated by the scoring percent of cells with nuclear pyknosis in total DAPI-stained cells (scale bar = 40 μm). ^###p < 0.001 compared between vector (vec) vs. αS^+/+ and ^**p < 0.01 compared between αS^+/+ vs. αS+GCN5.

Cell culture and treatment

Human neuroblastoma SH-SY5Y cells were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). The neuroblastoma cells were cultured in DMEM/F12 (1:1) supplemented with 100 U/mL penicillin/streptomycin and 10% (v/v) inactivated FBS in an incubator with the conditions maintained at 37°C and 5% CO₂. The cells were trypsinized (0.05% trypsin-EDTA) after reaching 80–90% confluence for sub-culture, and the media was replaced every 2 days. Each experiment was conducted at least three times from three consecutive passages for statistical analysis.

Lentiviral transfection

The lentiviral plasmid SNCA (pLenti-Glll-CMV-C-term-HA; cat# LV315044), KAT2A (pLenti-Glll-CMV-C-term-HA; cat# LV194350) and empty vector (pLenti-lll-Blank; cat# LV587) were obtained from Applied Biological Materials Inc. (Crestwood Place, BC, Canada). After transforming into E. coli (DH5α), the large colonies were picked for further amplification and purification using the PureLink plasmid filter kit (Thermo Fisher Scientific, V.A., Lithuania), as per the manufacturer’s instructions. The empty vector served as a control. Amplified and purified plasmids were infected into 70–80% confluent SH-SY5Y cells using a Lipofectamine 3000 (cat# L3000015; Thermo Fisher Scientific) reagent. Post-transfection, cells were kept for 48 hours and then treated with 0.75–1 μg/ml puromycin to select stably transfected cells. The stably transfected αS, GCN5, and Vec were subjected to WB and reverse transcription-PCR (RT-PCR) for further analysis.

Reverse transcription-PCR (RT-PCR)

The total RNA was extracted and isolated using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA) as described by the supplier. We used the following primers specific for:

• Bim (forward 5′-CTACCAGATCCCCACTTTTC-3′, reverse 5′-GCCCTCCTCGTGTAAGTCTC-3′);

• E2F1 (forward 5′-GACTGTGACTTTGGGGACT-3′, reverse 5′-TGTTCACCTTCATTCCC-3′);

• Egr-1 (forward 5′-CGAGCGAACAACCCTACGAGC-3′, reverse 5′-GAGGCAGAGGAAGACGATGAAGC-3′);

• αS (forward 5′-AGTGGCTGAGAAGACCAAAG-3′, reverse 5′-GTCAGGATCCACAGGCATATC-3′);

• GCN5 (forward 5′-AGCGGTTGTTCCCAGCACCC-3′, reverse 5′-GACAGCACAGAAGACAATCT-3′);

• GAPDH (forward 5′-GCAAAGTGGAGATTGTTGCCATC-3′ and reverse 5′-CATATTTCTCGTGGTTCACACCC-3′).

Photos were taken in a Davinch-K Gel Imaging System (MC2000/CG550, Youngin Lab Plus Co., the Republic of Korea).

Western blot analysis

The cells were washed twice with cold PBS and subjected to lyse using a lysis buffer (1× RIPA lysis buffer containing a protease and phosphatase inhibitor (1:1) cocktail). Whole mixtures were centrifuged at 14,000 rpm at 4°C for 15 minutes, and the supernatants were collected carefully without disturbing the pellets. The total protein obtained from cell lysates was quantified using a DC™ protein assay kit (Bio-Rad, USA) according to manufacturer instructions. Equal amounts of protein (20 μg) were separated electrophoretically using 8–15% sodium dodecyl sulfate-polyacrylamide electrophoresis (SDS-PAGE) gel and then transferred onto polyvinylidene-difluoride (PVDF) membranes (Millipore, Bedford, MA, USA). The membranes were incubated at room temperature for 1 hour with 3% bovine serum albumin in Tris-buffered saline (containing 0.1% Tween 20 buffer) to prevent nonspecific binding. The blots were then incubated overnight at 4°C with specific primary antibodies, including anti-Caspase3 (#H-277; Santa Cruz Biotechnology), anti-clvCaspase3 (#9661; cell signaling technology), anti-alpha-synuclein (#ab212184; Abcam), anti-alpha-synuclein LB509 (#ab27766; Abcam), anti-alpha-synuclein (211) (#sc-12767; Santa Cruz Biotechnology), anti-EGR-1 (#4154; cell signaling technology), anti-E2F1 (#3742; cell signaling technology), anti-Bim (#2819; cell signaling technology), anti-GCN5 (#sc-365321; Santa Cruz Biotechnology), anti-GCN5 (#MA45-14886; Invitrogen) anti-Bcl-2 (#N-19; Santa Cruz Biotechnology), anti-Lamin B1 (#12586; cell signaling technology), anti-Ac Histone H3 (#sc-56616; Santa Cruz Biotechnology), anti-Histone H3 (#sc-517576; Santa Cruz Biotechnology), anti-Ac Lys (#AB3879; Sigma-Aldrich) at 1:1000, anti-β-actin (#C4; Santa Cruz Biotechnology), and anti-β-tubulin (#2146; cell signaling technology) at 1:5000 concentration. Next, each blot was incubated at room temperature with either an anti-mouse or anti-rabbit (1:10,000) secondary antibody. The blots were visualized with an enhanced chemiluminescence detection system (LAS 500; GE Healthcare Bio-Sciences AB, Sweden) per the recommended protocol.

DAPI staining

SH-SY5Y or transfected cells were grown to a confluency of 70–80% then washed with cold PBS once, followed by fixation with 4% cold PFA for 15 minutes at room temperature. After discarding PFA, cells were washed twice and permeabilized with 0.1% Triton X-100 for 10 minutes at room temperature. Cells were stained with DAPI (2 μg/ml) for 10 minutes at room temperature and images were immediately captured using a Nikon Eclipse Ts2R fluorescence microscopy system and were processed by NIS-Elements BR-2.01.00 software (which came with the instrument).

RNA interference

siRNA of GCN5 (#sc-37946; Santa Cruz Biotechnology) was used. The interference efficiency was detected by comparing siRNAs with the normal control (NC). Each siRNA was transfected into SH-SY5Y cells by RNAiMax (Invitrogen) according to the manufacturer’s protocol. Forty-eight hours after transfection, cell lysates were harvested and processed for WB to detect the expression of these proteins.

Co-immunoprecipitation

The SH-SY5Y cells were seeded at 5 × 10⁵ density for 48 hours in a P100 plate and lysed in a non-denaturing RIPA buffer for 1 hour of incubation at 4°C on a wheel. After centrifugation, the supernatant was quantified with a DC™ protein assay kit (Bio-Rad, USA), and 0.6 mg of total protein extract was added to 50 μL of A/G agarose containing anti-GCN5, anti-E2F1, and anti-Egr-1 O/N at 4°C (rotation). The next day, the same binding buffer was used to perform four washes by centrifugation at 11,000 × g for 1 minute. The precipitated agarose was finally boiled and used for a western blot with the respective antibodies.

Statistical analysis

Statistical analyses were performed using GraphPad Prism (version 8.0.1; La Jolla, CA, USA) software. Data are represented as means ± SEM (standard error mean) of the three independent experiments. Student’s two-tailed t-test was used for comparison of the two different groups, and a one-way ANOVA, followed by Sidak’s multiple comparisons, was used for analysis more than two groups. A p-value of <0.05 was considered statistically significant.

Data availability

The datasets generated and/or analyzed during the current study are publicly available upon acceptance of this manuscript.