Aβ inhibits SREBP-2 activation through Akt inhibition

Abstract

We previously demonstrated that oligomeric amyloid β₄₂ (oAβ₄₂) inhibits the mevalonate pathway impairing cholesterol synthesis and protein prenylation. Enzymes of the mevalonate pathway are regulated by the transcription factor SREBP-2. Here, we show that in several neuronal types challenged with oAβ₄₂, SREBP-2 activation is reduced. Moreover, SREBP-2 activation is also decreased in the brain cortex of the Alzheimer's disease (AD) mouse model, TgCRND8, suggesting that SREBP-2 may be affected in vivo early in the disease. We demonstrate that oAβ₄₂ does not affect enzymatic cleavage of SREBP-2 per se, but may impair SREBP-2 transport from the endoplasmic reticulum (ER) to the Golgi. Trafficking of SREBP-2 from the ER to the Golgi requires protein kinase B (Akt) activation. oAβ₄₂ significantly reduces Akt phosphorylation and this decrease is responsible for the decline in SREBP-2 activation. Overexpression of constitutively active Akt prevents the effect of oAβ₄₂ on SREBP-2 and the downstream inhibition of cholesterol synthesis and protein prenylation. Our work provides a novel mechanistic link between Aβ and the mevalonate pathway, which will impact the views on issues related to cholesterol, isoprenoids, and statins in AD. We also identify SREBP-2 as an indirect target of Akt in neurons, which may play a role in the cross-talk between AD and diabetes.

Keywords: Alzheimer’s disease; amyloid β; cholesterol/biosynthesis; cholesterol/trafficking; isoprenoids; neurons; nuclear receptors/sterol regulatory element-binding protein; oligomeric amyloid β42; prenylation; protein kinase B; sterol regulatory element-binding protein-2.

Fig. 1.

oAβ₄₂ reduces neuronal SREBP-2 activation. A–G: Neurons were incubated for 24 h with different concentrations of oAβ₄₂ or reverse peptide (rAβ₄₂), as indicated at the top of the blots, at which time they were harvested and lysed. H, I: Frontal cortices of 10-week-old wild-type mice and TgCRND8 mice were collected and homogenized and an equal amount of protein from each homogenate was analyzed. A, C, E, H: Proteins were separated by SDS-PAGE and SREBP-2 was detected by immunoblot analysis. C_SREBP is a control generated by overexpression of human-(M)SREBP-2 in St14A cells and used to confirm the band of (M)SREBP-2 in immunoblots. B, D, F, I: Representations of densitometric analysis of mature forms of SREBP-2 combining at least five experiments and normalized to neurons grown in the presence of serum (B, D) or to neurons cultured in the absence of Aβ (F). G: Analysis of gene expression by real-time PCR. Data are expressed as mean ± SEM. For all experiments, **P < 0.01 and ***P < 0.005 one-way ANOVA.

Fig. 2.

oAβ₄₂ reduces (M)SREBP-2 independently of apoptosis or SREBP-2 degradation. A–E: Cortical neurons were challenged with oAβ (20 μM), ZVAD (50 μM), or a combination of both for 24 h. A: Nuclear fragmentation was evaluated by Hoechst 33258 staining using fluorescence microscopy. White arrowheads point to normal nuclei and white arrows indicate fragmented nuclei. B: Data are expressed as mean ± SEM of three experiments. Each experiment was performed in quintuplicate and 500–1,000 neurons per treatment were counted. Statistically significant differences from cultures given no Aβ (P < 0.01) are indicated by the symbol ** and were evaluated by the Kruskal-Wallis test with Dunn’s multiple post hoc comparison test. C: Proteins were separated by SDS-PAGE and cleaved caspase 3 was detected by immunoblot analysis. D: Proteins were separated by SDS-PAGE and SREBP-2 was detected by immunoblot analysis. C_SREBP is a control generated by overexpression of human-(M)SREBP-2 in St14A cells and used to confirm the band of (M)SREBP-2 in immunoblots. E: Representation of densitometric analysis of mature forms of SREBP-2 combining at least five experiments. Values have been normalized to neurons that have not received oAβ₄₂ or Z-VAD. F–H: Cortical neurons were treated with oAβ₄₂ (20 μM) for 24 h in the absence or presence of the proteasome inhibitor, MG132 (5 μM), for the last 2 h of treatment. F: Proteins were separated by SDS-PAGE and SREBP-2 was detected by immunoblot analysis. G: Representation of densitometric analysis of mature forms of SREBP-2 combining at least five experiments. Values have been normalized to neurons that have not received oAβ₄₂ or MG132. H: Detection of ubiquitinated proteins with anti-ubiquitin antibody. Data in E and G are expressed as mean ± SEM *P < 0.05, **P < 0.01 one-way ANOVA.

Fig. 3.

oAβ₄₂ does not affect SREBP-2 proteolytic cleavage at the Golgi. A: BFA causes disassembly of the Golgi complex in cortical neurons. Notice the dense perinuclear Golgi staining in untreated neurons compared with the diffused red fluorescence in neurons that have received 1 μg/ml BFA for 5 h. B–E: BFA prevents inhibition of SREBP-2 cleavage by the PI3K inhibitor, LY294002, and by oAβ₄₂. Cortical neurons were incubated with LY294002 (20 μM) (B, C) or oAβ₄₂ (20 μM) (D, E) for 24 h. In the last 5 h, BFA was added to the medium at different final concentrations (1, 5, and 10 μg/ml). Neurons were harvested and lysed. Proteins were separated by SDS-PAGE and SREBP-2 was detected by immunoblot analysis. C_SREBP is a control generated by overexpression of human-(M)SREBP-2 in St14A cells and used to confirm the band of (M)SREBP-2 in immunoblots. C, E: The densitometric analysis of SREBP-2 cleavage combining four and five experiments, respectively. Data are expressed as mean ± SE. **P < 0.01 one-way ANOVA compared with untreated neurons.

Fig. 4.

oAβ₄₂ inhibits Akt signaling in cortical neurons. Cortical neurons were exposed to oAβ₄₂, reverse Aβ_42-1, or LY294002, as indicated on the top of the blots, for 24 h after which neurons were harvested and analyzed for p^Ser473Akt and total Akt (A, B), p-p70S6K and total p70S6K (C, D), and Sec24D (E, F). Densitometric analysis (B, D, G) combines four to six experiments. Data are expressed as mean ± SEM. **P < 0.01, ****P < 0.01 one-way ANOVA with Dunnett’s multiple comparisons test.

Fig. 5.

Inhibition of SREBP-2 maturation by oAβ₄₂ requires inhibition of Akt. A–F: Cortical neurons uninfected or infected with an empty virus or a virus containing myr-Akt were exposed to oAβ₄₂ (20 μM) or LY294002 (20 μM) for 24 h. A–C: The cells were harvested and lysed. Proteins were analyzed by SDS-PAGE. SREBP-2, pAkt, Akt, and actin were detected by immunoblot analysis. B, C: Densitometric analysis combining five experiments. Data are expressed as mean ± SEM. D: [³H]acetate (100 μCi/ml) was added for the last 2 h of treatment. Lipids were extracted and separated by TLC. Incorporation of [³H]acetate into unesterified cholesterol was calculated as disintegrations per minute and per microgram of protein and expressed as a percentage of values obtained for untreated neurons. The results are the mean ± SEM of three experiments performed in three to five replicates. E: Analysis of Rab7 prenylation by extraction with Triton X-114: neuronal lysates (equal amount of protein) were extracted with Triton X-114. Proteins in detergent (D, prenylated proteins) and aqueous (A, unprenylated proteins) phases were analyzed by SDS-PAGE and immunoblotting. Ratios between total pixels of Rab7 in detergent/total pixels of Rab 7 in detergent plus aqueous were calculated. The data are presented as percentage of control (untreated neurons in each group). Data are expressed as the mean ± SEM of three experiments. F: Analysis of Rab prenylation by Rab-GDI capture. Neuronal lysates were prepared. Half of the sample was used to assess total protein input (I). The other half was used to extract prenylated proteins with recombinant GST-GDI (E). Proteins were analyzed by immunoblot analysis. Ratios between total pixels of Rab7 in GDI-bound (E)/total pixels of Rab7 in lysate before extraction (I) were calculated. Densitometric analysis of I included both bands. Data are presented as percentage of control (untreated neurons). Data are expressed as the mean ± SEM of three experiments. G, H: Neurons were incubated for 24 h with different concentrations of oAβ₄₂ or reverse peptide (rAβ_42-1), as indicated at the top of the blots, at which time they were harvested and lysed. Proteins were separated by SDS-PAGE and SREBP-1 was detected by immunoblot analysis. H: Densitometric analysis of mature forms of SREBP-1 combining three experiments and normalized to neurons cultured in the absence of Aβ. Data are expressed as mean ± SEM. For all experiments, **P < 0.01 and ***P < 0.005 one-way ANOVA.