Inhibition of zDHHC7-driven protein S-palmitoylation prevents cognitive deficits in an experimental model of Alzheimer's disease

Abstract

Protein post-translational modifications (PTM) play a crucial role in the modulation of synaptic function and their alterations are involved in the onset and progression of neurodegenerative disorders. S-palmitoylation is a PTM catalyzed by zinc finger DHHC domain containing (zDHHC) S-acyltransferases that affects both localization and activity of proteins regulating synaptic plasticity and amyloid-β (Aβ) metabolism. Here, we found significant increases of both zDHHC7 expression and protein S-palmitoylation in hippocampi of both 3×Tg-AD mice and post-mortem Alzheimer's disease (AD) patients. Chronic intranasal administration of the S-palmitoylation inhibitor 2-bromopalmitate counteracted synaptic plasticity and cognitive deficits, reduced the Aβ deposition in the hippocampus and extended the lifespan of both male and female 3×Tg-AD mice. Moreover, hippocampal silencing of zDHHC7 prevented the onset of cognitive deficits in the same experimental model. We also identified a FoxO1-mediated epigenetic mechanism inducing zDHHC7 expression, which was triggered by brain insulin resistance in 3×Tg-AD mice. Finally, in hippocampi of AD patients S-palmitoylation levels of Beta-Secretase 1 were associated with Aβ 1 to 42 load and they inversely correlated with Mini Mental State Examination scores. Our data reveal a key role of both zDHHC7 overexpression and protein hyperpalmitoylation in the onset and progression of AD-related alterations of synaptic plasticity and memory.

Keywords: Alzheimer’s disease; BACE1; brain insulin resistance; protein S-palmitoylation; zDHHC.

Fig. 1.

Hippocampus of 3×Tg-AD mice shows elevated levels of protein S-palmitoylation. (A) Silver staining assay and bar graph showing total protein S-palmitoylation in the hippocampus of WT and 3×Tg-AD mice (n = 8 mice per group composed by males and females, statistics by unpaired Student’s t test). (B) Immunoblots and bar graphs showing increase of S-palmitoylated fraction of GABAγ2A, NSF, BACE1, CaMKIIα, PICK1, GRIK2, GRIN2B, and SNAP25 (n = 6 mice per group composed by males and females, statistics by unpaired Student’s t test). Palmitoylated fraction of the proteins has been calculated by the ratio palm-protein X/total protein X (i.e., HAM⁺ protein X/protein X in the input). (C) Immunoblots and bar graphs showing elevated levels of both palmitoylated and total APP, GluA1, GluA2, and nNOS (n = 6 mice per group composed by males and females, statistics by unpaired Student’s t test). Palmitoylated protein levels have been calculated by the ratio palm-protein X/total housekeeping protein (i.e., HAM⁺ protein X/housekeeping protein in the input). Protein S-palmitoylation has been examined using the ABE assay (see Acyl-Biotin Exchange Assay section in Methods). Immunoblots show palmitoylated (acyl-biotin exchanged and affinity purified by streptavidin) proteins and total proteins (Total Input, TI). Samples without HAM (NH₂OH) are negative controls (−). Data are expressed as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001; n.s. not significant.

Fig. 2.

Intranasal administration of 2-BP relieves AD phenotype in 3×Tg-AD mice. (A) Timeline of experimental design (pharmacological treatment, molecular, electrophysiological, and behavioral analyses). Preference index of (B) male and (C) female 3×Tg-AD mice intranasally treated with either saline (veh) or 2-BP for 3, 6, or 9 mo in both NOR and OPR tests (n = 8 to 12 males and 8 to 10 females per group; statistics by Welch’s t test). (D) Time course (Left) and bar graph (Right) showing slope of LTP at CA3-CA1 synapses in hippocampal slices obtained from vehicle- and 2-BP- treated female 3×Tg-AD mice (3×Tg-AD_veh and 3×Tg-AD_2BP, respectively). Results are expressed as percentages of baseline (n = 12 slices from n = 5 mice for 3×Tg-AD_veh and n = 13 slices from n = 7 mice for 3×Tg-AD_2BP; statistics by unpaired student’s t test). (E) Kaplan–Meier survival curve comparing survival probability of vehicle (veh)- and 2-BP-treated male (Left) and female (Right) 3×Tg-AD from 10 wk of age until death (n = 17 to 18 males and 20 to 21 females for each group; statistics by log rank test). Data are expressed as mean ± SEM. **P < 0.01; ***P < 0.001; n.s. not significant.

Fig. 3.

2-BP reduces protein S-palmitoylation and Aβ levels in 3×Tg-AD hippocampi. (A) Silver staining assay and bar graph showing total protein S-palmitoylation in the hippocampus of WT, 3×Tg-AD_veh, and 3×Tg-AD_2BP mice (n = 8 mice per group composed by males and females, statistics by unpaired Student’s t test). (B) Aβ 1 to 42 peptide levels (pg of Aβ per mg of total proteins) in the hippocampus of male (Left) and female (Right) 3×Tg-AD_veh and 3×Tg-AD_2BP mice. ELISA was performed in triplicate (n = 6 mice; statistics by unpaired Student’s t test). (C) Immunostaining (Left) and bar graph (Right) showing Aβ load in the hippocampus of female 3×Tg-AD_veh and 3×Tg-AD_2BP animals (n = 4 mice, five slices have been analyzed for each animal; statistics by unpaired Student’s t test; (Scale bar: 250 µm.) Immunoblots (Left) and bar graphs (Right) showing (D) Aβ load and (E) tau phosphorylation levels in the hippocampus of female 3×Tg-AD_veh and 3×Tg-AD_2BP mice (n = 6 mice; statistics by unpaired Student’s t test). Data are expressed as mean ± SEM. **P < 0.01; ***P < 0.001; n.s. not significant.

Fig. 4.

zDHHC7 silencing counteracts Aβ deposition and the onset of cognitive deficits in 3×Tg-AD mice. (A) mRNA expression of different zDHHC enzymes (–5, 7, 8, 12, 13, 15, 17, 20, 21) in both WT and 3×Tg-AD mice at 3 and 9 mo of ages (n = 6 mice for each group composed by males and females; statistics by two-way ANOVA and Bonferroni post hoc). (B) Immunoblot (Left) and bar graph (Right) showing protein expression of zDHHCs 3, 7, and 21 in WT vs. 3×Tg-AD young mice (n = 6 mice for each group composed by males and females; statistics by unpaired Student’s t test). (C) Preference index in NOR (Left) and OPR (Right) tests of 3×Tg-AD mice stereotaxically injected with lentiviral particles encoding a shRNA targeting zDHHC7 (LV-sh zDHHC7), zDHHC21 (LV-sh zDHHC21), or control lentiviral particles (LV-sh scrambled, mock) (n = 10 to 13 female mice per group; statistics by Welch’s t test). (D) Immunoblots (Left) and bar graphs (Right) showing expression of zDHHCs 7 and 21 in LV-shzDHHC7 and LV-sh zDHHC21 injected mice (n = 6 female mice for each group; statistics by unpaired Student’s t test). (E) Immunoblots and bar graphs showing S-palmitoylation of SNAP25, nNOS, GABAγ2A, and BACE1 (n = 6 female mice for each group, statistics by unpaired Student’s t test). (F) Aβ 1 to 42 levels (pg of Aβ per mg of total proteins) of in the hippocampus of LV-sh zDHHC7 and mock 3×Tg-AD mice. ELISA was performed in triplicate (n = 6 female mice for each group; statistics by unpaired Student’s t test). Data are expressed as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001; n.s. not significant.

Fig. 5.

FoxO1 inhibition epigenetically regulates zDHHC7 expression. (A) Immunoblots (Top) and bar graphs (Bottom) showing the responsivity of insulin signaling pathway (GSK3βSer9, AKTSer473, IRS-1Ser612, and FoxO1 Ser256 phosphorylation) in 3-, 6-, and 9-mo-old 3×Tg-AD mice (n = 6 mice per group composed by males and females; statistics by unpaired Student’s t test). (B) Scheme displaying several putative FRE elements within the mouse zDHHC7 gene. The numbers indicate the distance from the ATG starting codon. (C) ChIP assays showing the binding of FoxO1 and histone 3 lysine 9 acetylation (H3K9ac) on the regulatory sequences (FRE) of the zDHHC7 gene in the hippocampus of WT and 3×Tg-AD mice at 6 mo of age. Experiments were performed in triplicate (n = 6 mice per group composed by males and females; statistics by unpaired Student’s t test). Data are expressed as mean ± SEM. *P < 0.05; ***P < 0.001; n.s. not significant.

Fig. 6.

Hippocampi of AD patients show increased levels of both zDHHC7 expression and protein S-palmitoylation. (A) Immunoblots and bar graph showing zDHHC7 expression in postmortem hippocampal tissue of ND and AD patients (n = 9 hippocampi; statistics by unpaired student’s t test). (B) Immunoblots and bar graphs showing S-palmitoylation levels of zDHHC7 in the hippocampus of ND and AD patients (n = 9 for each group, statistics by unpaired Student’s t test). Immunoblots show palmitoylated (acyl-biotin exchanged and detected by streptavidin) proteins and total proteins (Total Input, TI). Palmitoylated protein levels have been calculated by the ratio palm-protein X/total housekeeping protein (i.e. HAM⁺ protein X/housekeeping protein in the input). (C) Immunoblots and bar graphs showing S-palmitoylation levels of BACE1, SNAP25, and APP in the hippocampus of ND and AD patients (n = 9 for each group, statistics by unpaired Student’s t test). Immunoblots show palmitoylated (acyl-biotin exchanged and affinity purified by streptavidin) proteins and total proteins (TI). (D) Aβ 1 to 42 levels in the hippocampus of ND and AD patients. ELISA was performed in triplicate (n = 9; statistics by unpaired Student’s t test). (E) Scatter plots reporting the distribution for BACE1 S-palmitoylation and Aβ 1 to 42 levels (on Left, n = 9 per group; statistics by Pearson correlation coefficient) or for BACE1 S-palmitoylation and MMSE values (on Right, n = 9 per group; statistics by Pearson correlation coefficient). Data are expressed as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.005.