Cell-penetrating, antioxidant SELENOT mimetic protects dopaminergic neurons and ameliorates motor dysfunction in Parkinson's disease animal models

Abstract

Parkinson's disease (PD) is a neurodegenerative disorder characterized by motor dysfunction for which there is an unmet need for better treatment options. Although oxidative stress is a common feature of neurodegenerative diseases, notably PD, there is currently no efficient therapeutic strategy able to tackle this multi-target pathophysiological process. Based on our previous observations of the potent antioxidant and neuroprotective activity of SELENOT, a vital thioredoxin-like selenoprotein, we designed the small peptide PSELT from its redox active site to evaluate its antioxidant properties in vivo, and its potential polyfunctional activity in PD models. PSELT protects neurotoxin-treated dopaminergic neurons against oxidative stress and cell death, and their fibers against neurotoxic degeneration. PSELT is cell-permeable and acts in multiple subcellular compartments of dopaminergic neurons that are vulnerable to oxidative stress. In rodent models of PD, this protective activity prevented neurodegeneration, restored phosphorylated tyrosine hydroxylase levels, and led to improved motor skills. Transcriptomic analysis revealed that gene regulation by PSELT after MPP⁺ treatment negatively correlates with that occurring in PD, and positively correlates with that occurring after resveratrol treatment. Mechanistically, a major impact of PSELT is via nuclear stimulation of the transcription factor EZH2, leading to neuroprotection. Overall, these findings demonstrate the potential of PSELT as a therapeutic candidate for treatment of PD, targeting oxidative stress at multiple intracellular levels.

Keywords: Neurodegenerative disease; Neuron; Oxidative stress; Peptide therapy; Selenoprotein.

Fig. 1

PSELT, and not a control peptide, protects SH-SY5Y cells. A: Photomicrographs of SH-SY5Y cells treated or not with MPP⁺ (1 mM) in the presence or absence of PSELT (1 μM) for 36 h. Scale bar: 100 μm B: Quantification of cell numbers in the different conditions indicated in (A). Data are expressed as mean ± SEM and are compared using Student t-test, **p < 0.01 (n = 3 per condition). C: PSELT dose-response effect on cell survival. Cell number was determined after treatment with MPP⁺ (1 mM) in the presence or absence of the indicated concentrations of PSELT for 36 h. Data are expressed as mean ± SEM (n = 4 per condition). D: Cell number was quantified after treatment with or without PSELT or a control PSELT (PSELTc) where the Sec was replaced by a Ser residue (1 μM), in the presence or absence of MPP⁺ (1 mM) for 36 h. Data are expressed as mean ± SEM and are compared using Student t-test, **p < 0.01 (n = 3 per condition). E: Caspase 3/7 activity was assessed as described in Materials and Methods, in the different conditions indicated in (A). Data are expressed as mean ± SEM and are compared using Student t-test, **p < 0.01 (n = 3 per condition). F: Free radicals were quantified through DCF fluorescence as described in Materials and Methods in the different conditions indicated in (A), and the data were presented as mean ± SEM and compared using Student t-test, **p < 0.01 (n = 3 per condition).

Fig. 2

PSELT effects in SY5Y cells. A: Cell number was quantified after treatment with or without PSELT, PSELT-TAT or TAT alone (1 μM), in the presence or absence of MPP⁺ (1 mM) for 36 h. Data are expressed as mean ± SEM and are compared using ANOVA and Tukey's test, **p < 0.01 (n = 3 per condition). B: Alpha-synuclein (α-Syn) was examined by Western blot in control or PSELT-treated cells and normalized to GAPDH levels. C: Quantification of α-Syn levels from (B). Data are expressed as mean ± SEM and are compared using Student t-test, **p < 0.01 (n = 3 per condition). D: Glucose-regulated protein (GRP78) was examined by Western blot in control or PSELT-treated cells and normalized to GAPDH levels. E: Quantification of GRP78/BIP levels from (D). Data are expressed as mean ± SEM and are compared using Student t-test, *p < 0.05 (n = 3 per condition) F: The methylation mark H3K27me3 was examined by Western blot in control or PSELT-treated cells and normalized to histone H3 levels. G: Quantification of H3K27me3 levels from (F). Data are expressed as mean ± SEM and are compared using Student t-test, *p < 0.05 (n = 3 per condition). The data shown are representative of at least 2–3 experiments with similar results.

Fig. 3

PSELT reaches intracellular compartments. A: Confocal microscopic images of SH-SY5Y cells that were incubated for 5 min, 15 min and 6 h in the presence of PSELT labeled by the fluorochrome dansyl. The nuclei were stained by the TO-PRO3 dye. Scale bars: 10 and 20 μm. B: Mass spectrometric profile of PSELT alone showing several pics due to the existence of different selenium isotopes. C: Mass spectrometry analysis of SH-SY5Y cell extracts after incubation of the cells with PSELT for 30 min. D: Mass spectrometry analysis of SH-SY5Y cell extracts after incubation of the cells with PSELT for 1 h E: Mass spectrometry analysis of SH-SY5Y cell extracts after incubation of the cells with PSELT for 48 h. F: Mass spectrometry analysis of a control SH-SY5Y cell extract without PSELT incubation. No peak was seen in the PSELT mass position.

Fig. 4

Intranasal delivery of PSELT to the nigrostriatal pathway. A: TH and TAT immunostaining was performed in the SNc after intranasal administration of PSELT-TAT, at 2 and 4 h post-treatment. Note that there is no PSELT-TAT signal in control nigral tissue while a strong signal was detected at 4 h in this area (dashed). Nuclei were stained in blue with DAPI. Scale bar: 100 μm. B: TH and TAT immunostaining was performed in the Str after intranasal administration of PSELT-TAT, at 2 and 4 h post-treatment. Note that there is no PSELT-TAT signal in control striatal tissue while a strong signal was detected at 2 and 4 h in this area. Nuclei were stained in blue with DAPI. Scale bar: 100 μm. C: Higher magnification in the SNc and Str areas 4 h after intranasal administration of PSELT showing that the peptide is present in dopaminergic neurons and fibers and reaches the nuclei. Scale bar: 20 μm. D: Mass spectrometry analysis of a striatal extract after intranasal administration of PSELT for 5 consecutive days as described in the present study in PD models (see below). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5

PSELT protects the nigrostriatal dopaminergic pathway against MPTP insult. A: Exposure to MPTP provoked the degeneration of dopaminergic neurons in the SNc as revealed by TH and caspase 3 immunolabelings of tissue sections at 7 days post-treatment. Scale bar: 50 μm B: Exposure to MPTP provoked a degeneration of dopaminergic fibers in the Str as revealed by TH and caspase 3 immunolabelings of tissue sections at 7 days post-treatment. Scale bar: 50 μm. C: Western blot analysis of TH, caspase 3 and DJ-1 levels in the striatum of control and MPTP-treated mice in the presence or absence of PSELT. Alpha-tubulin was used as a protein loading control to normalize the data. D: Quantification of TH level after Western blot analysis of striatal samples from control and MPTP-treated animals in the presence or absence of PSELT (*p < 0.05; n = 10). E: Quantification of caspase 3 level after Western blot analysis of striatal samples from control and MPTP-treated animals in the presence or absence of PSELT (*p < 0.05; n = 10) F: Quantification of DJ-1 level after Western blot analysis of striatal samples from control and MPTP-treated animals in the presence or absence of PSELT (**p < 0.01; n = 10). G: Western blot analysis of TH-Ser 40, TH-Ser 31 and TH levels levels in the striatum of control and MPTP-treated mice in the presence or absence of PSELT. Alpha-tubulin was used as a protein loading control to normalize the data. H: Quantification of TH-Ser 40 level after Western blot analysis of striatal samples from control and MPTP-treated animals in the presence or absence of PSELT (*p < 0.05; n = 10) I: Quantification of TH-Ser 31 level after Western blot analysis of striatal samples from control and MPTP-treated animals in the presence or absence of PSELT (*p < 0.01; n = 10). J: Striatal dopamine level of control and MPTP-treated animals in the presence or absence of PSELT (n = 5). This experiment was repeated 3 times with similar results.

Fig. 6

PSELT protects the nigrostriatal dopaminergic pathway in 6-OHDA-treated rats. A: Stereotactic administration of 6-OHDA provoked the degeneration of dopaminergic neurons in the SNc as revealed by TH and caspase 3 immunolabelings of tissue sections 2 weeks after 6-OHDA treatment. PSELT (1 μM) administration during the first week attenuated the neurodegeneration. Scale bar: 50 μm. B: Stereotaxic administration of 6-OHDA also provoked the degeneration of dopaminergic fibers in the Str as revealed by TH and caspase 3 immunolabelings of tissue sections 2 weeks after 6-OHDA treatment. PSELT (1 μM) administration during the first week attenuated the neurodegeneration. Scale bar: 50 μm. C: Quantification of TH-positive neurons after 6-OHDA treatment in the presence or absence of PSELT. Data are expressed as mean ± SEM and are compared using Student t-test, *p < 0.05 (n = 5) D: Quantification of caspase 3 immunoreactivity in the SNc from control and 6-OHDA-treated animals in the presence or absence of PSELT (*p < 0.05; n = 5). E: Western blot analysis of TH levels in the str of control and 6-OHDA-treated rats in the presence or absence of PSELT. Alpha-tubulin was used as a protein loading control to normalize the data. F: Quantification of TH level after Western blot analysis of striatal samples from control and 6-OHDA-treated animals in the presence or absence of PSELT (*p < 0.05; n = 5).

Fig. 7

PSELT effect on gene expression levels in SH-SY5Y cells correlates with different gene sets. A: Unsupervised clustering based on gene expression profiling of the different experimental conditions indicated under the dendrogram (n = 3 per condition). Genes with a relatively higher level of expression are shown in red, and those with a lower level are shown in blue according to the color scale at the bottom. B: Gene set enrichment analysis of a ranked list of all genes comparing MPP⁺ and PSELT treatment versus MPP⁺ with those up-regulated in Parkinson's disease (p value = 0, enrichment score = 1.15). Genes up-regulated in Parkinson's disease positively correlate with genes up-regulated by MPP⁺ and down-regulated by PSELT. Data were taken from GSE68719_Parkinson disease [36] or from GSE78757_Gaucher disease, GSE20292_Parkinson substantia nigra and GSE9397_Parkinson MSN [[36], [61], [62]] in supplementary Fig.S1A. C: Gene set enrichment analysis of a ranked list of all genes comparing MPP⁺ and PSELT treatment versus MPP⁺ with those down-regulated in cortical neurogenesis (p value = 0, enrichment score = −1.29). Genes down-regulated during impaired corticogenesis correlate with genes up-regulated by PSELT. Data were taken from GSE74683_Cortex_Neurogenesis (ELP3 knockout) [63] or from GSE 42904_HIPPO [64], GSE8425_RASGRF1 KO [65] and the orphan nuclear receptor TLX targets [66] in supplementary Fig.S1B. D: Gene set enrichment analysis of a ranked list of all genes comparing MPP⁺ and PSELT treatment versus MPP⁺ with those up-regulated during caloric restriction (p value = 0, FDR = 0, enrichment score = 1.30). Genes that are up-regulated by caloric restriction are also up-regulated by PSELT. Data were taken from GSE75569 [67]. E: Gene set enrichment analysis of a ranked list of all genes comparing MPP⁺ and PSELT treatment versus MPP⁺ with those up-regulated by resveratrol (p value = 0, enrichment score = 1.23). Genes that are up-regulated by resveratrol are also up-regulated by PSELT and inversely, genes down-regulated by resveratrol are also down-regulated by PSELT. Data were taken from GSE36930 [68] or from GSE36930 [68] and GSE42432 [69] in supplementary Fig. S1C. F: Gene set enrichment analysis of a ranked list of all genes comparing MPP⁺ and PSELT treatment versus MPP⁺ with those associated/interacting with the transcription regulator EZH2 (p value = 0, enrichment score = −1.25). The EZH2-interacting genes are mostly correlated with PSELT treatment. Data were taken from the NCBI database (https://www.ncbi.nlm.nih.gov). G: Gene set enrichment analysis of a ranked list of all genes comparing MPP⁺ and PSELT treatment versus MPP⁺ with those up-regulated by EZH2 knockout (p value = 0, enrichment score = 1.14). Genes that are up-regulated by EZH2 knockout and have a H3K27me3 mark correlate with genes up-regulated by MPP⁺ and are down-regulated by PSELT. Data were taken from GSE80222 [38] and from several other studies [[70], [71], [72]] in supplementary Fig. S1D. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 8

PSELT regulates EZH2/H3K27me3 levels to promote cell survival. A: SH-SY5Y cells were incubated with PSELT-dansyl for 15 min or 6 h and immunolabeled for EZH2. The nuclei were labeled by the TO-PRO dye. Scale bars: 10 and 20 μm B: Quantification of EZH2 immunoreactivity after exposure to PSELT-dansyl. Data are expressed as mean ± SEM and are compared using Student t-test, ***p < 0.001 (n = 3 per condition). C: Western blot analysis of H3K27me3 levels in SH-SY5Y cells treated or not with PSELT in the presence or absence of EPZ6438. The histone H3 levels were used as a protein loading control to normalize the data. D: Quantification of H3K27me3 level after Western blot analysis of SH-SY5Y cells treated or not with PSELT in the presence or absence of EPZ6438. Data are expressed as mean ± SEM and are compared using ANOVA and Tukey's test, *p < 0.05, **p < 0.01, ***p < 0.001 (n = 3 per condition). E: SH-SY5Y cells were treated by PSELT and MPP⁺ in the presence or absence of EPZ6438. Scale bar: 100 μm. F: Quantification of cell survival after PSELT and MPP⁺ treatment in the presence or absence of EPZ6438. Data are expressed as mean ± SEM and are compared using ANOVA and Tukey's test, *p < 0.05, **P < 0.01 (n = 3 per condition). Experiments were repeated at least twice with similar results.

Fig. 9

PSELT ameliorates motor activity after MPTP treatment. A: Schematic representation of the protocol used for MPTP and PSELT treatments. Animals received 1 dose of MPTP (30 mg/kg) or NaCl 0.9% (controls) intraperitoneally each day for 5 days. PSELT was administered (30 μg/animal) intranasally each day for 6 days. Behavioral tests were performed at days 0, 2 and 7. Results of behavioral tests recorded at day 7 are shown as supplemental data (Fig. S2) and only data from days 0 and 2 are shown here. B: The coordination capacity of the animal groups was monitored using a rotarod. The latency at which the mice were able to maintain their balance on the bar was recorded. The average latency of the last two trials was used for analyses. Data are expressed as mean ± SEM and are compared using Student t-test, **p < 0.01 (n = 12). C and D: Horizontal activity of all animal groups was recorded during 3 consecutive periods of 10 min at days 0 (C) and 2 (D). Total horizontal activity for 30 min is shown for each animal group. Data are expressed as mean ± SEM and are compared using Student t-test, *p < 0.05, **p < 0.01 (n = 12). E and F: Vertical activity of all animal groups was monitored during 3 consecutive periods of 10 min at days 0 (E) and 2 (F). Data are expressed as mean ± SEM and are compared using Student t-test, *p < 0.05, **p < 0.01 (n = 12) G and H: Total distance traveled by all animals was recorded in 3 consecutive periods of 10 min at days 0 (G) and 2 (H). Data are expressed as mean ± SEM and are compared using Student t-test, *p < 0.05, **p < 0.01 (n = 12).