Neuroprotective and Anti-Inflammatory Effects of Linoleic Acid in Models of Parkinson's Disease: The Implication of Lipid Droplets and Lipophagy

Abstract

Parkinson's disease (PD) is the second most prevalent neurodegenerative disease after Alzheimer's disease. The principal pathological feature of PD is the progressive loss of dopaminergic neurons in the ventral midbrain. This pathology involves several cellular alterations: oxidative stress, mitochondrial dysfunction, loss of proteostasis, and autophagy impairment. Moreover, in recent years, lipid metabolism alterations have become relevant in PD pathogeny. The modification of lipid metabolism has become a possible way to treat the disease. Because of this, we analyzed the effect and possible mechanism of action of linoleic acid (LA) on an SH-SY5Y PD cell line model and a PD mouse model, both induced by 6-hydroxydopamine (6-OHDA) treatment. The results show that LA acts as a potent neuroprotective and anti-inflammatory agent in these PD models. We also observed that LA stimulates the biogenesis of lipid droplets and improves the autophagy/lipophagy flux, which resulted in an antioxidant effect in the in vitro PD model. In summary, we confirmed the neuroprotective effect of LA in vitro and in vivo against PD. We also obtained some clues about the novel neuroprotective mechanism of LA against PD through the regulation of lipid droplet dynamics.

Keywords: autophagy; lipid droplets; lipophagy; neurodegeneration; oxidative stress.

Figure A1

Neuroprotective and anti-inflammatory effect of LA on a hemiparkinsonian animal model. Representative images of coronal sections of the mouse SNpc showing the presence of dopaminergic neurons (TH, blue); astroglial cells labeled with glial fibrillary acidic protein (GFAP, green) and Tomato lectin (red) as a marker of activated microglial cells, both in the contralateral (not injected) and ipsilateral (injected) hemisphere. Scale bar, 200 μm.

Figure A2

Cytotoxic effect of LA on SH-SY5Y cell line. Determination of cell viability by MTT assay and nitrites concentration by Griess assay of the human cell line SH-SY5Y treated or not with LA. Each graph corresponds to different LA concentrations and preincubation times: 1 (A), 6 (B), 18 (C), 24 (D) and 48 (E) hours. The mean of the normalized cell viability percentage of the different treatment groups to the control group is shown. The mean of nitrites concentration of the different treatment groups is shown. Results were obtained from three independent experiments with six repetitions each per experimental group, and the error bars correspond to ± SD. *** p < 0.001; ** p < 0.01; * p < 0.05 versus control group.

Figure A3

(A,B) The human cell line SH-SY5Y was exposed to 35 µM 6-OHDA in the presence or not of LA, PF, and CQ. The cell viability was measured by MTT assay and determination of nitrites concentration was evaluated by Griess reaction. Each bar represents the mean ± SD of three independent experiments with 6 replications each per experimental group. (C) BODIPY 493/503 flow cytometry experiments on SH-SY5Y PD model. Representative histograms of flow cytometry experiments with the SH-SY5Y cells marked with BODIPY 493/503 (FITC-A signal) and treated or not with LA, 6-OHDA, PF, and/or CQ. The results correspond to 6 independent experiments per group.

Figure A4

Study of the LA effect on lipophagic flux and LDs dynamics on SH-SY5Y PD model. Representative maximum intensity projection of confocal images with the BODIPY 493/503 marker (green), ADRP (red), and p62 (magenta) from the SH-SY5Y cell line treated or not with LA, 6-OHDA, PF (A), and/or CQ (B) are shown.

Figure A5

(A) BODIPY 581/591 C11 flow cytometry experiments on SH-SY5Y PD model. Representative histograms of flow cytometry experiments with the SH-SY5Y cells marked with BODIPY 581/591 C11 and treated or not with LA, 6-OHDA, PF, and/or CQ. The left column corresponds to the FITC-A signal (peroxidized lipids), and the right column corresponds to the PE-A signal (non-peroxidized lipids). The results correspond to 6 independent experiments per group. (B) Representative maximum intensity projection of confocal images showing the immunocytochemical assay with BODIPY 581/591 C11 marker. It shows peroxidized lipids in green and non-peroxidized lipids in red on cells treated or not with LA, PF, CQ, and/or 6-OHDA.

Figure 1

Neuroprotective effect of LA on a hemi-Parkinsonian animal model. Animals were injected unilaterally with 6-OHDA in the SNpc in combination or not with LA. The epifluorescence microscope images are ipsilateral coronal sections containing SNpc samples from the different experimental groups. (A) Representative images of the SNpc showing the presence of dopaminergic neurons labeled with tyrosine hydroxylase (TH, blue); astroglial cells stained with glial fibrillary acidic protein (GFAP, green) and Tomato lectin (red) as a marker for activated microglial cells. Scale bar, 200 μm. The right columns show greater magnification of the area delimited in the images in the left column (100 μm). (B) Quantification of the number of dopaminergic, astrocyte, and microglial cells shown in (A). (C) Analysis of the presence of degenerating neurons (green) in the SNpc measured using Fluoro-Jade B staining (DAPI was used as a nuclear stain). Scale bar, 100 μm. (D) Quantification of the number of Fluoro-Jade B (FJ-B) cells shown in (C). (E) Apoptotic neurons in the SNpc measured by the TUNEL assay. Scale bar, 100 μm. (F) Quantification of the number of TUNEL-positive cells in (E). The values in (B,D,F) represent the mean ± SD, expressed as the percentage of positive cells in the SNpc, given a particular marker, from three different experiments. There were four animals/experiment/experimental group and five independent sections per animal. *** p < 0.001.

Figure 2

Anti-inflammatory effect of LA on a hemi-Parkinsonian animal model. Animals were injected unilaterally with 6-OHDA in the SNpc in combination or not with LA. The epifluorescence microscope images are ipsilateral coronal sections containing the SNpc from the different experimental groups. (A) Immunohistochemical identification of proinflammatory molecules (green) in the SNpc: cyclooxygenase 2 (COX-2) and tumor necrosis factor-alpha (TNFα). Dopaminergic neurons are labeled with a tyrosine hydroxylase antibody (TH, red), astroglial cells are marked with glial fibrillary acidic protein (GFAP, red), and nuclei are marked with DAPI. Scale bar, 100 μm. (B) The quantification of COX-2 and TNFα positive cells through the SNpc is shown. The values represent the mean ± SD, expressed as percentages, from three different experiments. There were four animals/experiment/experimental group and five independent sections per animal. *** p < 0.001.

Figure 3

In vitro neuroprotective and anti-inflammatory effects of LA on the SH-SY5Y PD model. The human cell line SH-SY5Y was pre-exposed to different concentrations of LA (15, 25, 35, 45, 55, and 65 μM) for 1 (A), 6 (B), 18 (C), 24 (D), or 48 h (E) before incubation with 35 µM 6-OHDA for 18 h. The cell viability was measured by MTT assay, and determination of the nitrite concentration was evaluated by Griess reaction. Each bar represents the mean ± SD of three independent experiments with 6 replicates per experimental group. Data were analyzed using one-way analysis of variance (ANOVA) followed by the Games–Howell correction in some cases and the Kruskal–Wallis test followed by Dunn’s multiple comparisons test in other cases. *** p < 0.001, ** p < 0.01, * p < 0.05 versus control group; ##: p < 0.01 and ###: p < 0.001 versus 6-OHDA-treated cultures. (F) Immunofluorescence images (epifluorescence microscope) of the apoptotic protein active caspase-3 (red) and the proinflammatory factor cyclooxygenase-2 (COX-2, green). Dapi was used as a nuclear marker. Representative results of three independent experiments are shown. Calibration bar: 100 µm. (G) Representative Western blot and quantification showing the expression of cleaved caspase-3 normalized by GAPDH. Data are the results of three independent experiments, and the error bars correspond to ± SD. Data were analyzed using a one-way analysis of variance (ANOVA) followed by a Tukey correction. ** p <0.01 in comparison with the basal group and ## p < 0.01 in comparison with the 6-OHDA group.

Figure 4

Regulation of the autophagic flux and lipid droplet levels by LA in the SH-SY5Y PD model. The human cell line SH-SY5Y was exposed to 35 µM 6-OHDA in the presence or not of LA, PF, and CQ. (A,B) Cell viability was measured by MTT assay, with LA, 6-OHDA, CQ, or PF added or not. The nitrite concentration was evaluated by Griess reaction under same conditions. Each bar represents the mean ± SD of three independent experiments with 6 replicates per experimental group. Data were analyzed using one-way analysis of variance (ANOVA) followed by the Games–Howell correction in some cases and the Kruskal–Wallis test followed by Dunn’s multiple comparisons test in other cases. *** p < 0.001, ** p < 0.01, * p < 0.05. (C,D) Representative Western blot and quantification showing the expression of LC3-II normalized by tubulin and by the control levels (%). Data represent the results of three independent experiments, and the error bars correspond to ± SD. Data were analyzed using one-way analysis of variance (ANOVA) followed by the Games–Howell correction. * p < 0.05 and ** p < 0.01 (if not indicated, it is with respect to the control). (E) The graph shows the autophagic flux rate (arbitrary units) of the different experimental groups. These measurements represent the subtraction of the LC3-II levels (arbitrary units normalized by tubulin) in the non-CQ group from those in the CQ group. The mean ± SD correspond to four independent experiments. Data were analyzed using the ANOVA test followed by Tukey’s multiple comparisons test. * p < 0.05 in comparison with 6-OHDA. (F) Determination of the BODIPY 493/503 signal was done by flow cytometry in the SH-SY5Y cell line treated or not with LA, 6-OHDA, and/or PF. The graph shows the mean value of the BODIPY 493/503 signal for the different treatment groups normalized by the control group. The results correspond to 6 independent experiments per group, and the error bars show ±SD. Data were analyzed using the Kruskal–Wallis test followed by Dunn’s multiple comparisons test. *** p < 0.001 (if not indicated, it is with respect to the control), ** p < 0.01; * p < 0.05.

Figure 5

Effect of LA on lipophagic flux and LD dynamics in the SH-SY5Y PD model. The human cell line SH-SY5Y was exposed to 35 µM 6-OHDA in the presence or not of LA, PF, and CQ. (A) Representative maximum intensity projection of confocal images showing the immunocytochemical analysis with the BODIPY 493/503 marker (green), ADRP (red), and p62 (magenta) on SH-SY5Y cells. (B) Colocalization analysis of BODIPY 493/503 on ADRP with the Manders overlap coefficient. (C) Analysis of BODIPY 492/503 particles per cell. (D) Colocalization analysis of BODIPY 493/503 on p62 with the Manders overlap coefficient. (E) Analysis of BODIPY 492/503 particles per cell. (F) Analysis of the size of BODIPY 492/503 particles per cell. (G) Analysis of p62 punctae per cell. (H) Colocalization analysis of p62 on BODIPY 493/503 with the Manders overlap coefficient. Graphs show the mean ± SD of 6 independent experiments per group. Data were analyzed by one-way analysis of variance (ANOVA), followed by a multiple comparison test with Bonferroni in some cases and one-way analysis of variance (ANOVA) followed by the Games–Howell correction in other cases. *** p < 0.001; * p < 0.05 vs. control or between two specific groups when indicated). (I) Determination of the BODIPY 493/503 signal by flow cytometry. The mean BODIPY 493/503 signal normalized by the control for the different treatment groups is shown. The results correspond to 6 independent experiments per group. Data were analyzed using the Kruskal–Wallis test followed by Dunn’s multiple comparisons test. * p < 0.05 vs. control or between two specific groups when indicated.

Figure 6

Regulation of lipid peroxidation by LA in the SH-SY5Y PD model. (A) Representative maximum intensity projection of confocal images showing the immunocytochemical assay with the BODIPY 581/591 C11 marker. Peroxidized lipids are shown in green, and nonperoxidized lipids are shown in red on cells treated or not with LA and 6-OHDA. (B) Quantification of the peroxidized lipids/nonperoxidized ratio in the LDs of the different experimental groups. The mean ± SD of 6 experiments per group is shown. Data were analyzed using one-way analysis of variance (ANOVA) followed by a multiple comparison test with Bonferroni. *** p < 0.001; * p < 0.05 (if not indicated, it is for the control). (C) Determination of the BODIPY 581/591 C11 signal by flow cytometry on the SH-SY5Y cell line treated or not with LA, PF, CQ, and/or 6-OHDA. The graph shows the mean of the peroxidized/nonperoxidized lipids ratio obtained from the BODIPY 581/591 c11 signals of the different treatment groups. The results correspond to the mean ± SD of 4 independent experiments per group. Data were analyzed by one-way analysis of variance (ANOVA), followed by the Games–Howell correction.