This study presents a method to analyze the morphology of mitochondria based on immunostaining and image analysis in mouse brain tissue in situ. It also describes how this allows one to detect changes in mitochondrial morphology induced by protein aggregation in Parkinson’s disease models.
Mitochondria play a central role in the energy metabolism of cells, and their function is especially important for neurons due to their high energy demand. Therefore, mitochondrial dysfunction is a pathological hallmark of various neurological disorders, including Parkinson’s disease. The shape and organization of the mitochondrial network is highly plastic, which allows the cell to respond to environmental cues and needs, and the structure of mitochondria is also tightly linked to their health. Here, we present a protocol to study mitochondrial morphology in situ based on immunostaining of the mitochondrial protein VDAC1 and subsequent image analysis. This tool could be particularly useful for the study of neurodegenerative disorders because it can detect subtle differences in mitochondrial counts and shape induced by aggregates of α-synuclein, an aggregation-prone protein heavily involved in the pathology of Parkinson’s disease. This method allows one to report that substantia nigra pars compacta dopaminergic neurons harboring pS129 lesions show mitochondrial fragmentation (as suggested by their reduced Aspect Ratio, AR) compared to their healthy neighboring neurons in a pre-formed fibril intracranial injection Parkinson model.
The central nervous system has an intense demand for ATP: neurons use ATP to support ionic gradients, neurotransmitter synthesis, synaptic vesicle mobilization, release, and recycling, and to enable local protein translation and degradation. More than 95% of the ATP used by the brain is produced by the mitochondria1. Therefore, it is not surprising that mitochondrial dysfunction is particularly harmful to neurons. In fact, mitochondrial function impairments play an important role in several neurological diseases, including neurodegenerative conditions, such as Parkinson's Disease (PD) and Alzheimer's Disease (AD)2,3.
Multiple genes are unequivocally linked to PD-encoding proteins that are relevant for mitochondrial function and homeostasis, such as Parkin4,5,6, PTEN-induced kinase 1 (PINK1)7,8 and DJ-19. Further evidence for a role for mitochondrial dysfunction in PD is that treatments with inhibitors of Complex I of the mitochondrial electron transport chain (such as Rotenone and MPTP) recapitulate several aspects of PD in vitro and in vivo10. However, it is important to state that many pathological processes may drive neuronal loss in PD, together with mitochondrial deficits: oxidative stress, altered calcium homeostasis, failure of the ubiquitin-proteasome and of autophagy-lysosomal systems, and protein aggregation are among the most studied (reviewed in11,12,13 and).
Mitochondria are heterogeneous in shape: in addition to individual units, they are commonly found as extended reticular and tubular networks. The structure and the cellular location of mitochondria are critical for their function14; in fact, mitochondrial networks are extremely dynamic, undergoing frequent processes of fission, fusion, and mitophagy in order to meet the needs of the cells and to respond to environmental cues15,16. In addition, the morphology of mitochondria is intimately linked to their health status. For example, in human optic atrophy, genetic mutations that reduce mitochondrial activity lead to abnormal, slender and hyperfused mitochondria17. On the other hand, a variety of human diseases present aberrant mitochondrial morphology, including mitochondrial fragmentation or excessive mitochondrial fusion, which have harmful effects on mitochondrial function (reviewed in18). In the context of PD, we and others have previously shown that abnormal mitochondrial shape correlates with dysfunction in response to α-synuclein aggregates19. While mitochondrial morphology has been extensively studied in vitro both in the context of PD and other diseases20,21,22, protocols for the evaluation of mitochondrial morphology from in vivo sections are lacking. This makes the in vivo study of mitochondria in the context of diseases such as PD highly dependent on transgenic animals23 or the evaluation of midbrain extracts that cannot provide cellular resolution.
Here, a protocol is presented to study the mitochondrial morphology in situ as an indicator of their functional status and health, based on immunostaining of the mitochondrial protein VDAC124 followed by image analysis in paraffin-embedded tissue sections. We also show the results of this protocol in in vitro and in vivo PD models: neuroblastoma cells overexpressing SNCA (Synuclein Alpha) and brain tissue from mice subjected to intracranial injection of α-synuclein Pre-Formed Fibrils (PFFs). Co-immunostaining with an antibody against α-synuclein (in cells) or phosphoSer129- α-synuclein pS129 (in mouse brains) allowed us to identify cells with aggregate protein pathology (overexpressed α-synuclein and α-synuclein fibrils, respectively) in the samples, while negative cells served as a non-pathological control within the same samples. Through this analysis and the data described here, a reduced aspect ratio was observed, indicating the fragmentation of mitochondria in cells overexpressing SNCA or presenting pS129 lesions.
Overall, this study shows that immunostaining combined with image analysis is a reliable method for analyzing mitochondrial morphology. In fact, it allows quantifying the number of mitochondria as well as some morphological parameters such as aspect ratio in both cell culture and tissue. The number of mitochondria is directly linked to the functional status of the fission and fusion mechanisms of the samples, whereas the AR value relies on the elongation of the organelle. This method may be particularly valuable for the quick evaluation of the mitochondrial abnormalities in models of PD in which altered mitochondrial morphology, dynamics and functions are well-known pathological mechanisms28,29. α-synuclein also plays a relevant role in PD: indeed, α-synuclein is one of the components of Lewy Bodies, the cytoplasmatic fibrillary aggregates that are used for post-mortem diagnosis of PD patients30. Moreover, mutations in the SNCA gene were found in patients with both familiar and sporadic PD (reviewed in31). Phosphorylation of α-synuclein at Ser129 has extensively been shown to label Lewy-Body-like pathology, which emerges after PFF insult and elicits various toxic effects32,26.
Using the tool presented here, we were able to detect a reduction in mitochondrial number and AR values in the presence of both overexpressed and aggregated α-synuclein (cells with α-synuclein-staining and neurons bearing phosphoSer129α-synuclein-positive lesions, respectively) compared to cells lacking such lesions (α-synuclein- and phosphoS129α-synuclein-negative cells). These results agree with previous reports showing how direct α-synuclein-mitochondria interactions produce toxic effects on mitochondrial function and homeostasis in PD26,3334. Indeed, it was reported that mice with α-synuclein mutations exhibit increased mitochondrial DNA damage35 and mitophagy36,37. Moreover, it was described that increased α-synuclein levels promote mitochondrial fission/fragmentation, induce reactive oxygen species within mitochondria, and dysregulate mitochondrial protein expression in cell lines and mouse models overexpressing α-synuclein26,38,39.
It is important to highlight that this tool highly depends on the antibodies used for the study; careful morphological evaluation of the antibody stain used is imperative to detect the appropriate subcellular compartment. As this technique is based on 5 µm sections and therefore requires single focal planes for the analysis of the mitochondrial structures, the absence of a phenotype will not rule out the existence of a phenotype, as it is possible that subtle differences in mitochondrial morphology may not be detected by this method.
While this work and others have previously used similar approaches to evaluate mitochondrial morphology in vivo40, there is a need for a detailed protocol to be made accessible to the research community for this assessment. The significance of this study is that it is possible to apply this method to various in vivo disease models to assess mitochondrial morphological abnormalities and identify potential pathology, which may eventually facilitate the screening of lead compounds for the treatment of such disorders. While this analysis is currently limited to paraffin-embedded tissue, the advantage of the method is that it can be applied to any disease model after terminal tissue collection, making it a very versatile tool.
The authors have nothing to disclose.
We wish to acknowledge the funders of this study, specifically Ikerbasque, the Spanish Ministry for Science and Innovation, the Michael J Fox Foundation, IBRO, and the Achucarro Basque Center for Neuroscience.
32 G Hamilton syringe | Hamilton | 7632-01 | |
4',6-diamidino-2-fenilindol, dihidrocloruro (DAPI) | Invitrogen | D1306 | |
4/0 USP 45 cm suture | SSa90 pga | 32345n-36u | |
Alexa fluor 488/594-Donkey anti-Mouse | Invitrogen | A21202; A21203 | green/red dye-Donkey anti-Mouse |
Alexa fluor 594/647-Donkey anti-Rabbit | Invitrogen | A21207 A31573 | red/far red dye-Donkey anti-Rabbit |
AlexaFluor 488-Donkey anti-Chicken | Jackson ImmunoResearch | 703-545-155 | green dye-Donkey anti-Chicken |
Anti-PSer129 α-synuclein EP1536Y (Rabbit) antibody | Abcam | ab51253 | |
Anti-TOM 20 (Mouse) antibody | Santa Cruz | sc-17764 | |
Anti-Tyrosine Hydroxylase (Chicken) antibody | Abcam | ab76442 | |
Anti-VDAC1 (Mouse) antibody | Santa Cruz | sc-390996 | |
Anti-α-synuclein antibody MJFR1 (Rabbit) | Abcam | ab138501 | |
Citrate buffer 100X stock: 120mM citrate buffer, 5% Tween in water (pH 6) | Home-made | ||
Disposable base mold for tissue embedding | Fisher | 22-363-553 | Plastic embedding boxes |
D-MEM F12 | Gibco | A321331020 | |
EVOS M7000 Imaging System | ThermoFisher Scientific | High-content automated fluorescence microscope | |
Fetal Bovine Serum | Gibco | 10270106 | |
Flat optical bottom 96 well plates | Greiner | 675090 | |
FluorSave Reagent | Millipore | 345789-20ML | Mounting reagent |
Glutamine 200 mM | Gibco | 25030-024 | |
ImmEdge Hydrophobic Barrier Pen | Vector Laboratories | H-4000 | PAP-pen |
Lipofectamine and Plus Reagent | Invitrogen | 11668-019; 11514-015 | Transfection reagent and transfection adjuvant |
Matrigel | Corning | 354230 | Coating matrix |
Microtome | ThermoFisher Scientific | ||
Normal Donkey Serum | Gibco | PCN5000 | |
Opti-MEM | Gibco | 31985070 | Transfection medium |
PCDNA4 plasmid (backbone) | Addgene | 41036 | |
Penicillin/Streptomycin solution | Gibco | 15140-122 | |
SH-SY5Y cells/well | ATCC | HTB-11 | |
Xylene substitute | Labbox | 22L36504 | |
Zeiss Axio Imager Apotome 2 | Carl Zeiss | Structured illumination fluorescence imaging system | |
α-synuclein peptide | rpeptide | S-1010-2 |