Assays for the Degradation of Misfolded Proteins in Cells

Podsumowanie

This report describes protocols for measuring degradation rates of misfolded proteins by either western blot or fluorescence-based assays. The methods can be applied to analysis of other misfolded proteins and for high throughput screening.

Streszczenie

Protein misfolding and aggregation are associated with various neurodegenerative diseases. Cellular mechanisms that recognize and degrade misfolded proteins may serve as potential therapeutic targets. To distinguish degradation of misfolding-prone proteins from other mechanisms that regulate their levels, one important method is to measure protein half-life in cells. However, this can be challenging because misfolding-prone proteins may exist in different forms, including the native form and misfolded forms of distinct characteristics. Here we describe assays to examine the half-life of misfolded proteins in mammalian cells using a highly aggregation-prone protein, Ataxin-1 with an extended polyglutamine (polyQ) stretch, and a conformationally unstable luciferase mutant as models. Cycloheximide chase is combined with cell fractionation to examine the turnover rate of misfolding-prone proteins in various cellular fractions. We further depict a fluorescence-based assay using an enhanced green fluorescence protein (EGFP)-fusion of the luciferase mutant, which can be adapted for high throughput screening on a microplate-reader.

Wprowadzenie

Proteins are the most abundant macromolecules in cells, and they play an essential role in virtually all biological processes. The biological activity of most proteins requires their folding into, and maintaining, the native three-dimensional structures. Proteins with aberrant conformations not only lose their normal functions, but also frequently form soluble oligomeric species or aggregates that impair the functions of other proteins and are toxic to cells^1,2. To counteract protein misfolding, cells employ both molecular chaperones, which assist unfolded or partially folded polypeptides to reach their native conformation, and degradation pathways, which eliminate misfolded proteins³. Given the complexity and stochastic nature of the folding process, protein misfolding is inevitable, and it may not be reversed in the case of mutations, biosynthetic errors, and post-translational damages¹. Hence, cells ultimately rely on degradation pathways to maintain their protein quality.

The importance of cellular protein quality control (PQC) systems is underscored by the prevalence of protein-misfolding diseases, including cancer, diabetes, and many neurodegenerative disorders such as Alzheimer's disease, Parkinson's disease, amyotrophic lateral sclerosis, Huntington's disease (HD), and spinocerebellar ataxias (SCA)^4,5. For example, mutations in the tumor suppressor p53 is the single most frequently genetic lesion in tumors, associated with ~50-70% of all cases⁶. A substantial fraction of p53 mutations are missense mutations that alter the conformation of p53, leading to the formation of aggregates⁷. Furthermore, proteins with expanded polyQ stretches are genetically and pathologically associated with HD and SCA. These progressive and often fatal diseases manifest when the length of the polyQ stretch in the affected proteins exceeds a certain threshold, and becomes increasingly severe as the length of the polyQ stretch prolongs⁸.

An attractive approach for treating these diseases is to therapeutically bolster cellular PQC systems, especially the degradation pathways. However, the pathways involved in the degradation of defective proteins, particularly those in mammalian cells, remain poorly understood. Although it has been recognized that the proteasome is critically important for the degradation of misfolded proteins, a vital issue remains undefined: how misfolded proteins are specifically recognized and targeted for degradation. Moreover, although PQC systems have been identified in cellular compartments including the cytoplasm, the endoplasmic reticulum, and the mitochondrion, the PQC systems in the nucleus remain unclear².

Recent studies by our lab have identified a system that recognizes and degrades a variety of misfolded proteins in the nucleus of mammalian cells⁹. This system is comprised of the promyelocytic protein (PML), a nuclear protein and a member of the tripartite motif-containing (TRIM) protein family, and RNF4, a RING-domain containing protein. PML, and several other TRIM proteins, possess SUMO (small ubiquitin-like modifier) E3 ligase activity, which facilitates the specificity and efficiency of protein SUMOylation¹⁰. RNF4 belongs to a small group of SUMO-targeted ubiquitin ligases (STUbL), which contain one or more sumo-interacting motifs (SIMs) in addition to the RING domain that affords them the ubiquitin ligase activity¹¹. We found that PML specifically recognizes misfolded proteins through discrete substrate recognition sites that can discern distinct features on misfolded proteins. Upon binding, PML tags misfolded proteins with poly-chains of SUMO2/3, two nearly identical mammalian SUMO proteins that can form poly-chains due to the existence of an internal SUMOylation site. SUMOylated misfolded proteins are then recognized by RNF4, which leads to their ubiquitination and proteasomal degradation. We further demonstrated that the PML-RNF4 system is important for protection against neurodegeneration, as deficiency in PML exacerbates the behavioral and neuropathological defects of a mouse model of SCA type 1 (SCA1)⁹.

To distinguish protein degradation from other cellular mechanisms that may regulate protein levels, the rates of protein turnover was measured⁹. Among the most frequently used methods to determine protein turnover are pulse chase and cycloheximide (CHX) chase. These two methods examine over time, respectively, the radioisotope-labeled proteins of interest in translation-proficient cells and total preexisting proteins of interest in translation-inhibited cells. However, a major challenge for studying pathogenic and misfolding-prone proteins is that the half-life of these proteins can be extremely long. For example, Ataxin-1, Ataxin-7, Huntingtin, α-synuclein, and TDP-43 all have half-lives of more than 12 to 24 hr^9,12-16. The slow turnover rates of these proteins preclude the use of the CHX chase analysis because the cells harboring these proteins may not survive prolonged translation inhibition, especially because the misfolded proteins themselves can be highly toxic to cells. The pulse-chase analysis with isotopic labeling may also be challenging for proteins that are highly aggregation-prone. Most pulse-chase assays rely on immunoprecipitation to separate the protein of interest from all the other proteins that are also radioactively labeled. This procedure normally includes lengthy immunoprecipitation and wash procedures, during which SDS-insoluble aggregates can form, making the analysis with SDS-PAGE electrophoresis inaccurate.

Here a protocol to analyze nuclear misfolded proteins with a slow turnover rate is described⁹. A pathogenic form of Ataxin-1 (Atxn1) that contains a stretch of 82 glutamines (Atxn1 82Q) is used for this purpose⁸. When expressed in cells, an enhanced green fluorescent protein (GFP) fusion of Atxn1 82Q forms microscopically visible inclusions in the nucleus (Figure 1A). Pulse chase analysis reveals that the half-life of Ataxn1 82Q is over 18 hr⁹. Atxn1 82Q is composed of species with misfolded conformations of different characteristics, as well as species with native conformation. It is likely that these species are degraded at different rates, and thus should be analyzed separately. Lysates from Atxn1 82Q-GFP-expressing cells are fractionated into NP-40-soluble (soluble or NS, likely representing native proteins or misfolded monomeric/oligomeric proteins) and NP-40-insoluble (NI, aggregated/misfolded) portions. The latter can be further divided into SDS-soluble (SS; likely disordered aggregates) or SDS-resistant (SR; likely amyloid fibrils) fractions (Figure 1B). NS and SS fractions can be analyzed by SDS-PAGE followed by western blotting, whereas the SR fraction can be detected by filter retardation assays followed by immunoblotting. CHX chase is combined with the detergent fractionation method, and discovered that the half-life of SS Atxn1 82Q is much shorter than that of NS Atxn1 82Q and total Atxn1 82Q (Figure 2A), indicating that the SS fraction can be readily recognized and degraded in cells⁹. Thus, this method provides a powerful tool to study the dynamics of misfolded proteins and to compare their degradation pattern.

We also describe a method that is suitable for high throughput screening for identifying macromolecules or small compounds that can modulate the degradation of misfolded proteins. This method is based on a conformationally destabilized mutant of firefly luciferase (LucDM)¹⁷, a model chaperone substrate. We have fused LucDM to a nuclear localization signal (NLS) to probe the PQC systems in this cellular compartment, and GFP (NLS-LucDM-GFP) for convenience of detection. NLS-LucDM-GFP forms microscopically visible nuclear aggregates in a small percentage of cells (Figure 3A). Similar to Atxn1 82Q, NLS-LucDM-GFP-but not its wild-type counterpart NLS-LucWT-GFP-is modified by SUMO2/3 and regulated by the PML-RNF4 pathway⁹. NLS-LucDM-GFP also forms SS and SR species, although the relative amounts of SS and SR species are minimal compared to NS, using the conditions described in protocol 3 below. To simplify the assay, we only analyze SDS soluble LucDM (including both the NS and SS fractions) by SDS-PAGE and western blot. Importantly, CHX chase assay showed that the half-life of SDS-soluble NLS-LucDM-GFP is much shorter than that of its wild-type counterpart (Figure 3B), suggesting that LucDM is a specific substrate for the system that recognizes and degrades misfolded proteins.

The degradation of LucDM-GFP causes a significant drop in overall fluorescence signal. Therefore, we have also developed a protocol for real-time detection of cellular LucDM-GFP using microplate fluorescence-based assay. Many high-throughput screen (HTS) systems are developed for drugs or genes modifying cellular aggregates and cellular viability caused by aberrant proteins^18-20. However, very few HTSs are specifically designed for targeting degradation in mammalian cells. This protocol serves as a robust system for rapid and large-scale analysis of the effects of protein expression, knockdown, and drug treatment on cellular misfolded protein degradation. Using HeLa cells as example, below we describe the protocols for the analysis of these two misfolded proteins. The assays can also be applied to other cell lines, although transfection conditions and time course may need to be optimized for individual cell lines.

Protokół

1. Preparation of Reagent

1. Prepare cell lysis buffer (50 mM Tris, pH 8.8, 100 mM NaCl, 5 mM MgCl₂, 0.5% NP-40). Supplement 2 mM DTT, 1x complete protease cocktail, and 250 IU/ml benzonase before use.
2. Prepare pellet buffer (20 mM Tris, pH 8.0, 15 mM MgCl₂). Supplement 2 mM DTT, 1x complete protease cocktail and 250 IU/ml benzonase before use.
3. Prepare 3x boiling buffer (6% SDS, 20 mM Tris, pH 8.0). Supplement 150 mM DTT before use.
4. Prepare low-fluorescence DMEM medium for assays using microplate fluorescence reader. Mix 25 mM glucose, 0.4 mM glycine, 0.4 mM arginine, 0.2 mM cysteine, 4.0 mM glutamine, 0.2 mM histidine, 0.8 mM isoleucine, 0.8 mM leucine, 0.8 mM lysine, 0.2 mM methionine, 0.4 mM phenylalanine, 0.4 mM serine, 0.8 mM threonine, 0.078 mM tryptophan, 0.4 mM tyrosine, 0.8 mM valine, 1.8 mM CaCl₂, 0.81 mM MgSO₄, 5.33 mM KCl, 44.0 mM NaHCO₃, 110 mM NaCl, 0.9 mM NaH₂PO₄. Adjust pH of solution using HCl or NaOH to pH 7.4. Sterilize the medium through filtration.
   NOTE: This medium contains components of the standard DMEM medium with high glucose except that Fe(NO₃)₃, vitamins and Phenol Red are omitted. Phenol Red, riboflavin and pyridoxal in regular DMEM culture medium significantly interfere with the detection of fluorescence signal²¹. Culture medium without those components is crucial for successful live cell GFP imaging. The medium is stable for 12 months when stored refrigerated.

2. Degradation Assay of Atxn1 82Q GFP

1. Plate approximately 3 x 10⁵ HeLa cells into 35 mm plates with DMEM medium supplemented with 10% fetal bovine serum (FBS), so that after O/N culturing cells grow to a confluence of 40-60% at the time of transfection.
   NOTE: The number of plates needed is based on the number of time points and treatments described below.
2. Transfect HeLa cells with 0.3 µg of Atxn1 82Q-GFP/pRK5 plasmid into each well using transfection reagent according to manufacturer's instruction⁹. Make a master transfection mix containing DNA and transfection reagent and aliquot it for each well.
3. 4-5 hr post transfection, examine the live cells under an inverted fluorescent microscope for GFP expression with excitation wavelength 450-490 nm. Return cells back to incubator after imaging.
   NOTE: Observe both diffused GFP signals and small speckles of GFP signals in nucleus at this time.
4. Remove the medium by vacuum aspiration and add 2 ml fresh DMEM containing 50 µg/ml CHX. Treat cells for 0, 4, 8, 12 and 16 hr with CHX before harvesting. To examine proteasomal degradation, include proteasome inhibitor MG132 (10 µM) in one plate treated for 16 hr as a control.
5. Harvest one plate of cells at each time point. Remove the medium by vacuum aspiration and wash the plates twice with 3 ml ice-cold 1x phosphate buffered saline (PBS). Snap-freeze the plate on dry ice.
6. After the last time point (16 hr), scrape the frozen cells (from all time points) into 150 µl of ice-cold cell lysis buffer and incubated on ice for 30 min.
7. Centrifuge the cell lysates in a benchtop centrifuge at 17,000 x g for 15 min at 4 °C.
8. Transfer the supernatant, which contains NP-40-soluble (NS) proteins, to another tube using a pipette.
   NOTE: Use supernatants for measuring protein concentrations by Bradford assay if needed.
9. Rinse the pellets by gently adding approximately 200 µl 1x PBS to the tubes without disturbing the pellets. Carefully remove PBS by vacuum aspiration or pipette. Re-suspend them in 150 µl ice-cold cell pellet buffer, followed by 15-30 min incubation on ice.
   NOTE: The resuspended pellet fraction contains NP-40 insoluble (NI) proteins. See Figure 1B for a diagram of detergent fractionation.
10. Add 75 µl of 3x boiling buffer into NS fractions and NP-40 insoluble (NI) fractions resuspended from the pellets. Heat the samples at 95 °C on a heat block for 5 min.
    NOTE: Clumps in the NP-40 insoluble fractions are dissolved after heating and samples will become clear.
11. Add SDS-gel loading buffer to an aliquot of boiled NS and NI. Load equal volume of samples collected from all time points to SDS-PAGE gel. The volume loaded corresponds to approximately 20 μg of NS from sample collected at Time 0. NOTE: NS, as well as SDS-soluble (SS) protein from NI faction, can be resolved by SDS separating gel. In contrast, SDS-resistant (SR) aggregates from NI fraction are stuck in the gel loading wells (Figure 1B). To improve western blot detection, double volume of NI can be loaded.
12. Detect NS and SS Atxn1 82Q-GFP by western blot using anti-GFP antibody and enhanced chemiluminescence (ECL).
    NOTE: Atxn1 82Q in the SS fraction is generally less compared to the NS fraction when using this protocol. Longer ECL exposure is needed for optimal signals.
13. Examine SR Atxn1 82Q from the pellet fraction by filter retardation assays using a dot-blot apparatus (Figure 1B). Briefly, set up a dot-blot apparatus holding a 0.2 µm cellulose acetate membrane. Load 80-120 µl of boiled NI into each well of the dot-blot apparatus.
    NOTE: As only small amounts of SR aggregates are formed in cells, for the dot blot assay, load a volume of the SR fraction that is approximately 10-15 times of the volume of the NS fraction used for western blot analysis.
    1. After filtering the samples through the membrane by vacuum, Atxn1 82Q-GFP aggregates stuck on the membrane can be detected by anti-GFP immunoblotting^9,12,22.
       NOTE: See previous reports^9,12,22 for detailed description of filter retardation assay. This step is optional because SR Atxn1 82Q is minimal compared to SS and NS fractions following the short transfection described in this protocol. In addition, the levels of SR Atxn1 82Q largely remain the same over CHX chase (Figure 2B). However, it is still crucial to examine whether the amounts of SS Atxn1 82Q are affected over time for any drug treatment or gene expression in initial experiments. SS protein species stay in SDS-PAGE gel loading wells can be detected by western blot. However, this method is less sensitive and generates more variable results (Figure 1B).

3. Degradation Assay of NLS-luciferase-GFP Using SDS-PAGE and Western Blot

1. Plate approximately 3 x 10⁵ HeLa cells into 35 mm plates with DMEM medium supplemented with 10% fetal bovine serum (FBS). After O/N culturing, a confluence of 40-60% is reached at the time of transfection. The number of plates needed is based on the number of time points and treatments described below.
2. Transfect HeLa cells with 1.0 µg NLS-luciferase-GFP/pRK5 plasmid into each well using transfection reagent according to manufacturer's instruction⁹. Make a master transfection mix containing DNA and transfection reagent for aliquot of each well.
3. After O/N transfection (around 15 hr), examine the live cells under an inverted fluorescent microscope for GFP expression with excitation wavelength 450-490 nm. Return cells back to incubator after imaging.
   NOTE: Diffused nuclear GFP signal can be observed in the majority of the cells (70%). A small percentage (5%) of cells have nuclear aggregates.
4. Remove the medium by vacuum aspiration and add 2 ml fresh DMEM containing 50 µg/ml CHX. Treat cells for 0, 1.5, 3, 4.5 and 6 hr with CHX before harvesting. To examine proteasomal degradation, include additional proteasome inhibitor MG132 (10 µM) in one plate treated for 6 hr as control.
5. Harvest one plate of cells at each time point. Remove the medium by vacuum aspiration and wash the plates twice with 3 ml ice-cold phosphate buffered saline (PBS). Snap-freeze the plate on dry ice.
6. After the last time point (6 hr) is finished, cells frozen at all-time points are scraped into 150 µl of ice-cold cell lysis buffer and incubated on ice for 30 min.
7. Add SDS gel-loading buffer to whole cell lysates for a final concentration of 2% SDS and 50 mM DTT. Incubate samples on a heat block at 95 °C for 5 min.
8. Analyze NLS-luciferase-GFP by SDS-PAGE and western blot using anti-GFP antibody. Load equal volume of samples collected from all time points to SDS-PAGE gel. The volume loaded corresponds to approximately 20 µg of whole cell lysates from sample collected at Time 0.

4. Real-time Degradation Assay of NLS-luciferase-GFP Using Fluorescence Microplate Reader

1. Seed approximately 1 x 10⁴ HeLa cells into black 96-well tissue culture plates with transparent bottom. After O/N culturing, a confluence of 50-70% is reached at the time of transfection.
   NOTE: 60 µl of medium containing completely suspended HeLa cells are seeded directly into each well. Do not add additional medium or rock plate back and forth after seeding, otherwise cells may be distributed unevenly.
2. Transfect HeLa cells with 0.05-0.1 µg of NLS-luciferase-GFP/pRK5 plasmid⁹ into each well using transfection reagent according to manufacturer's instruction. Make a master transfection mix containing DNA and transfection reagent and aliquot it for each well. Three wells with cells are not transfected with DNA, which serves as control for background fluorescence signal for each treatment condition.
   NOTE: An alternative way to reduce transfection variation is to transfect cells before seeding. However, a large portion of cells transfected with misfolded proteins fail to attach or show reduced viability using this method. It is likely that some cells may not stand the stress caused by trypsin digestion when expressing toxic misfolded proteins. As a result, the overall fluorescent signal is significantly reduced.
3. 20-24 hr after transfection, examine the live cells under an inverted fluorescent microscope for GFP expression with excitation wavelength 450-490 nm.
4. Remove the medium by vacuum aspiration. Add approximately 200 µl 1x PBS to each well and then aspirate it to remove residual amount of DMEM medium.
5. Add 60 µl of low-fluorescence DMEM medium with 5% FBS and 50 µg/ml CHX. To examine proteasomal degradation, include additional proteasome inhibitor MG132 (10 µM) in one set of samples. Set triplicate of wells for each treatment/condition.
   NOTE: MG132 treatment is included as controls for proteasomal degradation because it is crucial to rule out other factors that can cause drop of fluorescence signal, including cell death and fluorescence quenching.
6. Measure the fluorescence signal of GFP immediately on a fluorescent plate reader after adding CHX.
   NOTE: The software measurement settings are shown in Table 1.
7. Read the plate every hour for up to 8-10 hr. After reading, return the plate back to cell culture incubator.
8. Export the data as spreadsheet file. Use mean value of multi-reads for each well as fluorescence intensity. Normalize values of each data point to the mean values of Time 0 in the respective groups. Plot the normalized fluorescence intensity over time as shown in Figure 4A and 4B, and Figure 5, right panels.
9. Perform statistical analysis of degradation rates between two conditions using two-way ANOVA with repeated measures²³.
   NOTE: In this analysis, "fluorescence intensity" is the dependent variable whereas "treatment conditions" and "time" are the two factors. Instead of comparing data at individual time points, two-way ANOVA with repeated measures analyzes the difference between the two treatment groups over the entire time course.

Wyniki

In a steady state analysis, microscopically visible Atxn1 82Q-GFP nuclear aggregates can be observed in 30 - 50% of HeLa cells 20 hr after transfection (Figure 1A). Western blot analysis of NS and SS fractions using anti-GFP antibody shows a distinct band of Atxn1 82Q-GFP between 100 kDa and 150 kDa markers, corresponding to the protein's molecular weight (Figure 1B). Atxn1 82Q-GFP in the SR fraction can be detected either by filter retardation assay,...

Dyskusje

Mechanisms that regulate the degradation of misfolded proteins are essential for maintaining the homeostasis of cellular proteins, and they likely represent valuable drug targets for treating neurodegenerative disorders and other protein-misfolding diseases. Here, assays that examine the degradation of misfolded proteins are described, using a pathogenic Atxn1 protein (Atxn1 82Q) and a nuclear localized luciferase mutant (NLS-LucDM) as examples.

To examine the degradation Atxn1 82Q, which has ...

Materiały

Name	Company	Catalog Number	Comments
Dulbecco's Modified Eagle Medium	Life Technologies	11995-092
Fetal Bovine Serum	Life Technologies	10082147
Lipofectamine 2000	Life Technologies	11668019
NP-40	Sigma-Aldrich	NP40S-500ML
SDS	Sigma-Aldrich	L3771
MG132	Sigma-Aldrich	M8699
Cycloheximide	Sigma-Aldrich	C7698
Dithiothreitol (DTT)	Fisher Scientific	45000232
Complete Protease Inhibitor Cocktail Tablets	Roche Boehringer Mannheim	4693159001
Bio-Dot Apparatus	Bio-Rad	1706545
Living Colors GFP Monoclonal Antibody	Clonetech	632375
Anti-Actin mAb Rabbit, IgG	Sigma-Aldrich	A45060-200UL
Amino acids	Sigma-Aldrich		Amino acids are used for making low fluorecence culturing medium
Olympus IX-81  Inverted Fluorescence Microscope	Olympus	IX71/IX81
96 Well Black TC Plate w/ Transluscent Clear Bottom	Sigma-Greiner	89135-048
Fluorescence Bottom Plate Reader Infinite 200® PRO	TECAN	Infinite 200® PRO
Cellulose acetate membrane 0.2 µm	Sterlitech	CA023001
Prism 5	GraphPad		Statistical analysis software