Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Aug 8;9(1):120.
doi: 10.1038/s41531-023-00564-3.

Parkinson's disease neurons exhibit alterations in mitochondrial quality control proteins

Chun Chen  1 , David McDonald  2   3 , Alasdair Blain  4 , Emily Mossman  4 , Kiera Atkin  4 , Michael F Marusich  5 , Roderick Capaldi  6 , Laura Bone  4 , Anna Smith  4 , Andrew Filby  2   3 , Daniel Erskine  4 , Oliver Russell  4 , Gavin Hudson  7 , Amy E Vincent  4 , Amy K Reeve  8
Affiliations

Parkinson's disease neurons exhibit alterations in mitochondrial quality control proteins

Chun Chen et al. NPJ Parkinsons Dis. .

Abstract

Mitochondrial dysfunction has been suggested to contribute to Parkinson's disease pathogenesis, though an understanding of the extent or exact mechanism of this contribution remains elusive. This has been complicated by challenging nature of pathway-based analysis and an inability simultaneously study multiple related proteins within human brain tissue. We used imaging mass cytometry (IMC) to overcome these challenges, measuring multiple protein targets, whilst retaining the spatial relationship between targets in post-mortem midbrain sections. We used IMC to simultaneously interrogate subunits of the mitochondrial oxidative phosphorylation complexes, and several key signalling pathways important for mitochondrial homoeostasis, in a large cohort of PD patient and control cases. We revealed a generalised and synergistic reduction in mitochondrial quality control proteins in dopaminergic neurons from Parkinson's patients. Further, protein-protein abundance relationships appeared significantly different between PD and disease control tissue. Our data showed a significant reduction in the abundance of PINK1, Parkin and phosphorylated ubiquitinSer65, integral to the mitophagy machinery; two mitochondrial chaperones, HSP60 and PHB1; and regulators of mitochondrial protein synthesis and the unfolded protein response, SIRT3 and TFAM. Further, SIRT3 and PINK1 did not show an adaptive response to an ATP synthase defect in the Parkinson's neurons. We also observed intraneuronal aggregates of phosphorylated ubiquitinSer65, alongside increased abundance of mitochondrial proteases, LONP1 and HTRA2, within the Parkinson's neurons with Lewy body pathology, compared to those without. Taken together, these findings suggest an inability to turnover mitochondria and maintain mitochondrial proteostasis in Parkinson's neurons. This may exacerbate the impact of oxidative phosphorylation defects and ageing related oxidative stress, leading to neuronal degeneration. Our data also suggest that that Lewy pathology may affect mitochondrial quality control regulation through the disturbance of mitophagy and intramitochondrial proteostasis.

PubMed Disclaimer

Conflict of interest statement

M.F.M. declares receipt of licensing revenues from Abcam and EMD/Millipore on sales of anti-mtDNA-encoded protein antibodies. The remaining authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Mitochondrial adaptability in Parkinson’s neurons.
Mitochondrial quality control (MQC) operates through the intricate cooperation of three processes: synthesis and import of new building blocks for mitochondria; correct protein folding and degradation to maintain protein homoeostasis within the organelle; alongside mitochondrial dynamics and mitophagy to turnover dysfunctional mitochondria. This MQC system allows neuronal mitochondria to monitor and adapt themselves to a certain degree of damage, such as OxPhos defects, oxidative stress and protein aggregation, without causing interference to neuron homoeostasis. It is hypothesised that the MQC system is impaired in PD; the weakening mitochondrial adaptability is unable to compensate for the increasing burden of oxidative damage and protein degradation that advancing with age, leads to premature neuronal death (a). In this study, we investigated the expression change of key MQC signalling proteins on PD neurons using imaging mass cytometry (IMC) (b). These tested proteins were categorised into three major groups based on their most well-known function for easy interpretation.
Fig. 2
Fig. 2. IMC analysis of mitochondrial MQC proteins expression in human SN neurons.
a An outline of the IMC workflow on FFPE midbrain sections. Immunostaining of metal conjugated antibodies (Table 1) for IMC detection were performed in two experiments, using EDTA (pH = 8) and sodium citrate (pH = 6) buffer respectively for heat-mediated antigen retrieval. Individual dopaminergic neurons within the SN were segmented based on positive tyrosine hydroxylase (TH) staining, a clear nucleus (anti-Histone H3) and intracellular neuromelanin signals (visualised via Ir-intercalator binding); single pixel intensities were extracted, allowing measurement of the mean pixel intensity of each cytoplasmic area for statistical analysis. These neurons were further categorised into those with complex I/IV/V deficiency and those with normal protein levels respectively; the level of OxPhos protein deficiency was statistically defined based on the lower 10% limit of the prediction interval of the control dataset. Differences of the signalling protein expression between the two categories (deficient/normal) of neurons were further analysed to understand their changes in association with OxPhos deficiency. Detailed information of the tissue cohort included in the study were summarised in Supplementary Table 1. Scale bar, 20 µm. Example IMC images were selected from PD08 (male, 90 yrs), POLG01 (female, 90 yrs) and CON04 (female, 54 yrs) (b); PD01(male, 70 yrs), POLG02 (male, 59 yrs) and CON09 (male, 88 yrs) (c), demonstrating differential signal intensities from test proteins between neurons within selected individuals, and between groups. Scale bar, 20 µm.
Fig. 3
Fig. 3. Profiling of MQC protein abundance in disease and control cases.
a Heatmap demonstration of the average levels of signalling protein abundance for individual cases in groups, ranked in diseases (Supplementary Table 1) and protein groups (Fig. 1b). The mean log-transformed intensity data for each tested protein were normalised across tested individuals (scaled by column). b–l Using the linear mixed regression modelling (LMM), difference in the abundance signalling proteins between PD neurons and neurons from mitochondrial disease and healthy controls were analysed. The slopes and p values for each tested target from the LMM analysis were summarised into a volcano plot for an integral visualisation (b), with the detailed comparison of proteins that showed a significant change, along with the mitochondrial mass marker (Porin/VDAC1) demonstrated (LMM; p, *≤0.05; ns > 0.05) (c–l). Mitochondrial cases with mtDNA point mutation and POLG mutations as disease controls, alongside healthy aged controls were compared to PD group. Each data point represents each paired group (b) or single neuron (c–l); Boxplots show median and mean (triangle) 25th and 75th percentile, with whiskers indicating 95th upper/lower interquartile range.
Fig. 4
Fig. 4. Single-neuron correlation analysis of mitochondrial MQC and OxPhos proteins.
a–h Correlation matrices showing the strength of the association between tested proteins, basing on Spearman correlation analysis using single neuronal data. The colour and shape of the ellipse represent the Spearman’s rank correlation coefficient r of each paired protein (thinner ellipses represent stronger correlations); paired proteins showing no significant correlation were left blank (adjusted p > 0.0003, a–d; adjusted p > 0.0006, e–h).
Fig. 5
Fig. 5. Investigation of the changes in signalling protein expressions in association with OxPhos deficiency.
a–c SN neurons showing deficiency (that fell below the lower 10% limit of the prediction interval of the control) in complex I (NDUFA13), IV (MTCO1) or V (ATP5B) and those with normal expression were categorised as two subgroups within each individual case. Within each individuals, differences in the expression level of tested proteins between the two subgroups were analysed using Bayesian Estimation modelling (µ1, PD; µ2, Control). The Difference of Means (µ1–µ2) generated from the modelling for each tested signalling protein were summarised and ranked in descending order based on the control, for the three types of OxPhos deficiencies respectively. Comparison of these differences in the responsive proteins between PD, mitochondrial disease and control group were also performed (ANOVA, p, *≤0.05, **≤0.01), no significant difference was identified in the complex I and IV responses. Each dot represents an individual case; bars and lines represent the mean and SD; dot size shows the effect size.
Fig. 6
Fig. 6. Changes in MQC proteins in PD neurons with α-Synuclein aggregation.
a Example IMC images of anti-phosphorylated α-Synuclein Ser129(α-Syn) antibody labelling in a TH-positive neuron, showing Lewy body within the neuronal soma and Lewy neurites. Scale bar, 20 µm. b To investigate the response of tested signalling proteins to α-Syn pathology, intra-individual analysis of changes in the protein abundance was performed between PD neurons with and without α-Syn aggregates (signal pixel intensity>5, the number of adjacent positive pixel > 3), using Bayesian Estimation modelling (µ1, α-Syn positive; µ2, α-Syn negative). Each dot represents an individual case (Number of neurons analysed, PD03, 04 &07, n = 44, 39 &11); bars and lines represent the mean and SD; dot size shows the effect size. c Corresponding to the response analysis, three TH-positive neurons selected from PD03 (male, 80 yrs) with varying degree of α-Syn aggregation were demonstrated, alongside signals from the three signalling proteins, phosphorylated ubiquitinSer65 (pUb), LonP and HTRA2 which showed the highest level of increases in α-Syn positive neurons, compared to the α-Syn negative ones among tested proteins. Scale bar, 20 µm. d Proportion of neurons with α-Synuclein or/and pUb aggregates were calculated to support the observation of their co-existence within the PD neurons. Each dot represents an individual case; boxplots show median and mean (triangle) 25th and 75th percentile, with whiskers indicating 95th upper/lower interquartile range. e–g LMM analysis further confirmed the changes of pUb, LonP and HTRA2 to α-Syn pathology at a group level (6/8 PD cases, n = 121 neurons analysed; p < 0.05).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Schapira AHV, et al. Mitochondrial complex I deficiency In Parkinson’s disease. Lancet. 1989;333:1269. - PubMed
    1. Chen, C., Turnbull, D. M. & Reeve, A. K. Mitochondrial dysfunction in Parkinson’s disease-cause or consequence? Biology8. 10.3390/biology8020038 (2019). - PMC - PubMed
    1. D’Amato RJ, Lipman ZP, Snyder SH. Selectivity of the parkinsonian neurotoxin MPTP: toxic metabolite MPP+ binds to neuromelanin. Science. 1986;231:987–989. - PubMed
    1. Simola N, Morelli M, Carta AR. The 6-hydroxydopamine model of Parkinson’s disease. Neurotox. Res. 2007;11:151–167. - PubMed
    1. Martinez TN, Greenamyre JT. Toxin models of mitochondrial dysfunction in Parkinson’s disease. Antioxid. Redox Signal. 2012;16:920–934. - PMC - PubMed