Involvement of peroxisome proliferator-activated receptor β/δ (PPAR β/δ) in BDNF signaling during aging and in Alzheimer disease: possible role of 4-hydroxynonenal (4-HNE)

Abstract

Aging and many neurological disorders, such as AD, are linked to oxidative stress, which is considered the common effector of the cascade of degenerative events. In this phenomenon, reactive oxygen species play a fundamental role in the oxidative decomposition of polyunsaturated fatty acids, resulting in the formation of a complex mixture of aldehydic end products, such as malondialdehyde, 4-hydroxynonenal, and other alkenals. Interestingly, 4-HNE has been indicated as an intracellular agonist of peroxisome proliferator-activated receptor β/δ. In this study, we examined, at early and advanced AD stages (3, 9, and 18 months), the pattern of 4-HNE and its catabolic enzyme glutathione S-transferase P1 in relation to the expression of PPARβ/δ, BDNF signaling, as mRNA and protein, as well as on their pathological forms (i.e., precursors or truncated forms). The data obtained indicate a novel detrimental age-dependent role of PPAR β/δ in AD by increasing pro-BDNF and decreasing BDNF/TrkB survival pathways, thus pointing toward the possibility that a specific PPARβ/δ antagonist may be used to counteract the disease progression.

Keywords: BDNF; JNK; TrkB; aging; neurodegenerative disease; oxidative stress; p75; transcription factors.

Figure 1. HNE adducts in Wt and Tg cortices at different ages. (A) HNE immunolocalization in 3-, 9- and 18-mo Wt and Tg cortex sections. Three Tg mice and 3 Wt littermates for each age considered were deeply anesthetized before rapid killing by transcardial perfusion. Mice were perfusion-fixed, and brains were paraffin-embedded. Sagittal brain sections from Tg and Wt animals were deparaffinized and incubated with PBS containing 0.01% trypsin, for 10 min at 37° and processed for the antigen-retrieval procedure, using a microwave oven operated at 720 W. After cooling, slides were transferred to PBS containing 4% BSA for 2 h at RT, then incubated overnight at 4 °C with mouse monoclonal anti-HNE. Sections were thoroughly rinsed with PBS, then incubated for 2 h at RT with goat anti-mouse IgG Alexafluor 488/546 conjugated diluted 1:2000 in blocking solution. Controls were performed in parallel by omitting the primary antibody. Slides were observed at florescence microscope AXIOPHOT Zeiss microscope equipped with Leica DFC 350 FX camera. Image acquisition was performed with Leica IM500 program. Bar = 50 μm. (B) Western blotting analysis for HNE protein adducts in Wt and Tg animals at the indicated ages. SDS-PAGE was performed running samples (20 μg protein) on 7.5–15% polyacrylamide denaturing gels. Protein bands were blotted onto polyvinylidene fluoride sheets. Non-specific binding sites were blocked for 1 h at RT with 5% (w/v) non-fat dry milk in TBS-T. Membranes were incubated overnight at 4°C with mouse monoclonal anti-HNE (1:400), followed by incubation with 1:2000 HRP-conjugated anti-mouse IgG secondary antibody in blocking solution, for 1 h at 4°C. After rinsing, immunoreactive bands were visualized by ECL. The relative densities of the immunoreactive bands were determined and normalized with respect to β-actin, using a semiquantitative densitometric analysis. Data are mean ± SD of 4 different experiments. **, P < 0.001, AD vs. the respective Wt. (C) Western blotting analysis for gluthatione S-transferase (GSTP1) in Wt and Tg animals at the indicated ages. SDS-PAGE was performed running samples (20 μg protein) on 7.5–15% polyacrylamide denaturing gels. Protein bands were blotted onto polyvinylidene fluoride sheets. Non-specific binding sites were blocked for 1 h at RT with 5% (w/v) non-fat dry milk in TBS-T. Membranes were incubated overnight at 4°C with rabbit polyclonal anti-GSTP1 (1:1000), followed by incubation with 1:2000 HRP-conjugated anti-rabbit IgG secondary antibody in blocking solution, for 1 h at 4°C. After rinsing, immunoreactive bands were visualized by ECL. The relative densities of the immunoreactive bands were determined and normalized with respect to β-actin, using a semiquantitative densitometric analysis. Data are mean ± SD of 4 different experiments. **P < 0.001, AD vs. the respective Wt .

Figure 2. PPARβ/δ expression in Wt and Tg animals at different ages. (A) Real-time PCR in 3-, 9-, and 18-mo cortices for peroxisome proliferator-activated receptor (PPARβ/δ). The gene expressions were quantified in a 2-step RT-PCR. Complimentary DNA was reverse transcribed from total RNA samples using High-Capacity cDNA Reverse Trancription Kit. PCR products were synthesized from cDNA using the TaqMan universal PCR master mix and Assays on Demand gene expression reagents for mouse PPARβ/δ (Assay ID: Mm01305432-m1). Measurements were made using the ABI Prism 7300HT sequence detection system. As reference, TBP gene expression assay was used. Results represent normalized PPARβ/δ mRNA amounts relative to 3 mo healthy tissue using the 2^-ΔΔCt method. Data are mean ± SD of 4 different experiments. **P < 0.001, AD vs. the respective Wt. (B) Western blotting analysis for PPARβ/δ. SDS-PAGE was performed running samples (20 μg protein) on 7.5–15% polyacrylamide denaturing gels. Protein bands were blotted onto polyvinylidene fluoride sheets. Non-specific binding sites were blocked for 1 h at RT with 5% (w/v) non-fat dry milk in TBS-T. Membranes were incubated overnight at 4°C with rabbit polyclonal anti-PPARβ/δ (1:1000), followed by incubation with 1:2000 HRP-conjugated anti-rabbit IgG secondary antibody in blocking solution, for 1 h at 4°C. After rinsing, immunoreactive bands were visualized by ECL. The relative densities of the immunoreactive bands were determined and normalized with respect to β-actin, using a semiquantitative densitometric analysis. Data are mean ± SD of 4 different experiments. **P < 0.001, AD vs. the respective Wt. (C) PPARβ/δ immunolocalization in 18-mo Wt and Tg brain sections. Three Tg mice and 3 Wt littermates for each age considered were deeply anesthetized before rapid killing by transcardial perfusion. Mice were perfusion-fixed, and brains were paraffin-embedded. Sagittal brain sections from Tg and Wt animals were deparaffinized and incubated with PBS containing 0.01% trypsin, for 10 min at 37° and processed for the antigen-retrieval procedure, using a microwave oven operated at 720 W. After cooling, slides were transferred to PBS containing 4% BSA for 2 h at RT, then incubated overnight at 4°C with rabbit polyclonal anti-PPARβ/δ (1:100). Sections were thoroughly rinsed with PBS, then incubated for 2 h at RT with goat anti-rabbit IgG Alexafluor 488/546 conjugated diluted 1:2000 in blocking solution. Controls were performed in parallel by omitting the primary antibody. Slides were observed at florescence microscope AXIOPHOT Zeiss microscope equipped with Leica DFC 350 FX camera. Image acquisition was performed with Leica IM500 program. Bar = 15 μm.

Figure 3. TrkB expression in Wt and Tg cortices at different ages. (A) Real-time PCR in 3-, 9-, and 18-mo cortices for TrkBfl. The gene expressions were quantified in a 2-step RT-PCR. Complimentary DNA was reverse transcribed from total RNA samples using High-Capacity cDNA Reverse Trancription Kit. PCR products were synthesized from cDNA using the TaqMan universal PCR master mix and Assays on Demand gene expression reagents for mouse TrkB (Assay ID: Mm00435422-m1). Measurements were made using the ABI Prism 7300HT sequence detection system. As reference TBP gene expression assay was used. Results represent normalized TrkB mRNA amounts relative to 3-mo healthy tissue using the 2^-ΔΔCt method. Data are mean ± SD of 4 different experiments. ⁺P < 0.01, 9 mo vs. 18 mo, both Wt and Tg. (B) Western blotting analysis for TrkBfl SDS-PAGE was performed running samples (20 μg protein) on 7.5–15% polyacrylamide denaturing gels. Protein bands were blotted onto polyvinylidene fluoride sheets. Non-specific binding sites were blocked for 1 h at RT with 5% (w/v) non-fat dry milk in TBS-T. Membranes were incubated overnight at 4°C with rabbit polyclonal anti-TrkBfl (1:200), followed by incubation with 1:2000 HRP-conjugated anti-rabbit IgG secondary antibody in blocking solution, for 1 h at 4°C. After rinsing, immunoreactive bands were visualized by ECL. The relative densities of the immunoreactive bands were determined and normalized with respect to β-actin, using a semiquantitative densitometric analysis. Data are mean ± SD of 4 different experiments. **P < 0.001, AD vs. the respective Wt. (C) Western blotting analysis for TrkBt. Membranes were incubated overnight at 4°C with rabbit polyclonal anti-TrkB-t (1:200), followed by incubation with 1:2000 HRP-conjugated anti-rabbit IgG secondary antibody in blocking solution for 1 h at 4°C. After rinsing, immunoreactive bands were visualized by ECL. The relative densities of the immunoreactive bands were determined and normalized with respect to β-actin, using a semiquantitative densitometric analysis. Data are mean ± SD of 4 different experiments. **P < 0.001, AD vs. the respective Wt. (D) Immunolocalization in 9 mo Wt and Tg brain sections. Three 9 Tg mice and 3 Wt littermates, 9 mo of age were deeply anesthetized before rapid killing by transcardial perfusion. Mice were perfusion-fixed, and brains were paraffin-embedded. Sagittal brain sections from Tg and Wt animals were deparaffinized and incubated with PBS containing 0.01% trypsin, for 10 min at 37° and processed for the antigen-retrieval procedure, using a microwave oven operated at 720 W. After cooling, slides were transferred to PBS containing 4% BSA for 2 h at RT, then incubated overnight at 4°C with rabbit polyclonal anti-TrkB (1:50). Sections were thoroughly rinsed with PBS, then incubated for 2 h at RT with goat anti-rabbit IgG Alexafluor 488/546 conjugated diluted 1:2000 in blocking solution. Controls were performed in parallel by omitting the primary antibody. Slides were observed at florescence microscope AXIOPHOT Zeiss microscope equipped with Leica DFC 350 FX camera. Image acquisition was performed with Leica IM500 program. Bar = 50 μm.

Figure 4. BDNF levels in Wt and Tg cortices at different ages. (A and B) pro-BDNF and mBDNF expression, respectively, in Wt and Tg cortices at different ages. SDS-PAGE was performed running samples (20 μg protein) on 7.5–15% polyacrylamide denaturing gels. Protein bands were blotted onto polyvinylidene fluoride sheets. Non-specific binding sites were blocked for 1 h at RT with 5% (w/v) non-fat dry milk in TBS-T. Membranes were incubated overnight at 4°C with rabbit polyclonal anti-BDNF (1:100) (able to recognize both pro and mature BDNF forms), followed by incubation with 1:2000 HRP-conjugated anti-rabbit IgG secondary antibody in blocking solution for 1 h at 4°C. After rinsing, immunoreactive bands were visualized by ECL. The relative densities of the immunoreactive bands were determined and normalized with respect to β-actin, using a semiquantitative densitometric analysis. Data are mean ± SD of 4 different experiments. **P < 0.001, AD vs. the respective Wt. (C) Western blotting analysis for the astrocyte marker GFAP in Wt and Tg cortices at different ages. Membranes were incubated overnight at 4°C with rabbit polyclonal anti-GFAP (1:200) followed by incubation with 1:2000 HRP-conjugated anti-rabbit IgG secondary antibody in blocking solution, for 1 h at 4°C. After rinsing, immunoreactive bands were visualized by ECL. The relative densities of the immunoreactive bands were determined and normalized with respect to β-actin, using a semiquantitative densitometric analysis. Data are mean ± SD of 4 different experiments. **P < 0.001, AD vs. the respective Wt. (D) Western blotting analysis for p-ERK5 in Wt and Tg cortices at different ages. Membranes were incubated overnight at 4°C with rabbit polyclonal anti-p-ERK5 (1:200),followed by incubation with 1:2000 HRP-conjugated anti-rabbit IgG secondary antibody in blocking solution, for 1 h at 4°C. After rinsing, immunoreactive bands were visualized by ECL. The relative densities of the immunoreactive bands were determined and normalized with respect to β-actin, using a semiquantitative densitometric analysis. Data are mean ± SD of 4 different experiments. **P < 0.001, AD vs. the respective Wt.

Figure 5. p75^NTR levels in Wt and Tg cortices at different ages. (A) Western blotting analysis for p75.^NTR SDS-PAGE was performed running samples (20 μg protein) on 7.5–15% polyacrylamide denaturing gels. Protein bands were blotted onto polyvinylidene fluoride sheets. Non-specific binding sites were blocked for 1 h at RT with 5% (w/v) non-fat dry milk in TBS-T. Membranes were incubated overnight at 4 °C with rabbit polyclonal anti- p75^NTR (1:500), followed by incubation with 1:2000 HRP-conjugated anti-rabbit IgG secondary antibody in blocking solution, for 1 h at 4 °C. After rinsing, immunoreactive bands were visualized by ECL. The relative densities of the immunoreactive bands were determined and normalized with respect to β-actin using a semiquantitative densitometric analysis. Data are mean ± SD of 4 different experiments. **P < 0.001, AD vs. the respective Wt. (B) Western blotting analysis for ICD. Membranes were incubated overnight at 4°C with rabbit polyclonal anti- p75.^NTR (1:500), (able to recognize also the cleaved p75^NTR form), followed by incubation with 1:2000 HRP-conjugated anti-rabbit IgG secondary antibody in blocking solution, for 1 h at 4°C. After rinsing, immunoreactive bands were visualized by ECL. The relative densities of the immunoreactive bands were determined and normalized with respect to β-actin, using a semiquantitative densitometric analysis. Data are mean ± SD of 4 different experiments. **P < 0.001, AD vs. the respective Wt. In (C) the graphic representation of the ratio ICD/p75^NTR. In (D) Tunel assay on 9- and 18-mo Wt and Tg cortices paraffin-embedded sections were de-waxed in xylene, rehydrated in descending concentration of ethanol, and rinsed in PBS; sections were then subjected to 350 W microwave irradiation in 0.1 M citrate buffer pH 6.0 for 5 min for permeabilization and washed in PBS. Sections were incubated with TdT and fluorescein-labeled dUTP in TdT buffer in a dark humid chamber at 37 °C for 60 min. Negative controls were performed omitting TdT. Slides were mounted with Vectashield Mounting Medium and were observed at florescence microscope AXIOPHOT Zeiss microscope equipped with Leica DFC 350 FX camera. Image acquisition was performed with Leica IM500 program. Bar = 40 μm.

Figure 6. Schematic representation depicting the interactions between the different molecular intermediates studied. The scheme proposes that the increase of oxidative stress activates, by 4-HNE, PPARβ/δ, which, in turn, triggers the decrease of TrkBfl. On the other hand, the oxidative stress, through a not well-identified pathway (PPARβ/δ-dependent?), increases pro-BDNF and the ratio ICD/p75^NTR with concomitant increase of neuronal death.