CHIP Targets Toxic α-Synuclein Oligomers for Degradation

Construct Generation

The two constructs utilized in the present study were GN-αSyn and αSynGC (previously referred to as GN-link-αSyn and αSynGC) (19). αSyn bimolecular fluorescence complementation (BiFC) constructs were generated by PCR, cloned into pcDNA3.1, and verified by DNA sequencing as described previously (19). For CHIP expression, pcDNA3-CHIP, pcDNA3-CHIPΔU (residues 196 –303 deleted), and pcDNA3-CHIPΔTPR (residues 32–144 deleted) constructs were utilized and were kind gifts from Dr. Cam Patterson, University of North Carolina School of Medicine (20).

Cell Culture and Transient Transfections

Human H4 neuroglioma cells (HTB-148; ATCC) were maintained in OPTIMEM medium supplemented with 10% fetal bovine serum (both from Invitrogen) and incubated at 37 °C, 5%CO₂. H4 cells were plated 24 h prior to transfection and grown to 80–90% confluency. Cells were transfected with an equimolar ratio of DNA constructs using Superfect (Qiagen) according to the manufacturer’s instructions. Co-transfection of pcDNA3.1+ vector was used as a control.

GN-αSyn/αSynGC Fluorescence Complementation Assay

For optimal fluorophore maturation, transiently transfected cells were incubated overnight at 30 °C (21) after an initial 4-h incubation at 37 °C. 24 h after transfection cells were either observed using a Zeiss LSM510 confocal microscope or harvested for cell lysate preparation.

Live Cell Imaging

Cells were plated on poly-d-lysine-coated coverslipped 35-mm dishes (MatTek Cultureware, Ashland, MA) and transfected the following day. Cells were imaged 24 h post-transfection on a Zeiss LSM510 confocal microscope system. Cells were observed at ×63 for quantification of pixel intensity. For each experimental group the average fluorescence intensity from 50 cells, across three independent experiments, was measured using Adobe Photoshop® software. Adobe Photoshop® translated the average GFP fluorescence to average pixel intensity. Therefore, the average pixel intensity was recorded for each cell. Data were normalized to empty vector (pcDNA3.1+) controls.

Fluorescent-activated Cell Sorting

Cells were plated and transfected in 10-cm dishes. 24 h post transfection, cells were trypsinized, centrifuged, and the pellet reconstituted in phosphate-buffered saline. The resulting suspension was filtered with cell strainer caps into polypropylene tubes (both from BD Biosciences). Fluorescence was measured on a FACSCanto (BD Biosciences) flow cytometer.

Toxicity Assay

24 h post-transfection levels of toxicity were measured utilizing the ToxiLight™ non-destructive cytotoxicity assay kit (Cambrex, Rockland, ME). This assay quantitatively measures the release of adenylate kinase from damaged cells into the culture medium and was performed according to the manufacturer’s protocol.

Detergent Solubility Fractionation and Gel Electrophoresis

After 24 h, cells were washed with cold phosphate-buffered saline, harvested by scraping in cold lysis buffer with or without detergent for SDS-PAGE and native gels, respectively, (50 mm Tris/HCl, pH 7.4, 175 mm NaCl, 5 mm EDTA, pH 8.0, and protease inhibitor mixture), and sheared once through a 28-gauge needle followed by sonication for 10 s (total cell lysates). Fractionation was performed by adding Triton X-100 (Tx) to total cell lysates (final concentration of 1%) and incubating for 30 min on ice followed by centrifugation (15,000 × g, 60 min, 4 °C). The supernatant was designated as the Tx-soluble fraction, and the pellet was redissolved in 2% SDS-containing lysis buffer and sonicated for 10 s (Tx-insoluble fraction). Protein concentration was determined using a BCA protein assay. 15–40 μg of protein of each cell lysate was loaded onto a 10–20% Tris/glycine gel (Invitrogen) or a Native PAGE Novex 4–16% Bis-Tris gel (Invitrogen) for Western blot analysis. SDS-PAGE was performed with SDS-containing running and sample loading buffer (Invitrogen), whereas native electrophoresis was performed using detergent-free running and sample buffers (Invitrogen). Proteins resolved on the gels were transferred onto polyvinylidene difluoride membrane (Millipore, Billerica, MA) and blocked in 5% milk in Tris-buffered saline (TBS) containing 0.05% Tween 20 (TBS-T) for 2 h prior to the addition of primary antibody, anti-synuclein (610787, 1:1000; BD Biosciences), or anti-GFP (ab6556, 1:3000; Abcam, Cambridge, MA) at room temperature for 2 h or 4 °C overnight. Horseradish peroxidase-labeled anti-mouse or anti-rabbit antibodies (1:2000; Jackson ImmunoResearch, West Grove, PA) were incubated at room temperature for 1 h, processed, and quantified. The results were visualized with ECL (GE Healthcare) or SuperSignal Femto (Pierce) chemiluminescent reagents. Denatured blots were probed with anti-glyceraldehyde-3-phosphate dehydrogenase (ab9485, 1:3000; Abcam) and anti-rabbit horseradish peroxidase (1:2000) as a loading control. A nonspecific protein band at ~150–200 kDa across all samples in the series (Coomassie Blue staining of the gel) served as a loading control for the native blots. ImageJ and ImageQuant were used to quantify optical densities of total, soluble, and insoluble blots and native blots, respectively.

Coimmunoprecipitation

For immunoprecipitation, H4 cells were harvested 24 h after transfection and lysed in lysis buffer: 50 mm Tris/HCl, pH 7.4, 150 mm NaCl, 0.3% CHAPS, 10 mm NaF, 100 μm sodium orthovanadate, 1 mm phenylmethylsulfonyl fluoride, and protease inhibitor mixture (Roche Applied Science) for 1 h on ice. Cell lysates were centrifuged at 10,000 rpm for 10 min at 4 °C, and supernatant was used for immunoprecipitation using protein G-agarose and rat anti-Myc antibody (JAC6; Serotec, Oxford, UK) overnight at 4 °C. Immunoprecipitates were washed with lysate buffer three times and loaded onto 10–20% Tris/glycine gels (Invitrogen). Protein was transferred to polyvinylidene difluoride membrane (Millipore) and blocked in 5% milk in TBS containing 0.05% TBS-T for 2 h prior to the addition of primary antibody, anti-Hsp70 (SPA-812, 1:10000; StressGen, Ann Arbor, MI), or anti-GFP (ab6556, 1:3000; Abcam) at room temperature for 2 h or 4 °C overnight. Horseradish peroxidase-labeled anti-mouse or anti-rabbit antibody (1:2000; Jackson ImmunoResearch) was incubated at room temperature for 1 h, processed, and quantified. The results were visualized with ECL (GE Healthcare).

CHIP Reduces αSyn Oligomerization

We have previously shown that CHIP modifies and prevents α-Syn inclusion formation by interacting (either directly or indirectly) with αSyn (22). To further investigate the specific αSyn species that is targeted by CHIP, we employed a novel BiFC assay that we have previously demonstrated to be a specific and effective method by which to visualize and monitor αSyn oligomerization in vitro (19). BiFC involves the reconstitution of non-fluorescent fragments of GFP when fused to proteins of interest. Upon protein-protein interaction the two fragments are able to come together and reconstitute the fluorophore. We investigated the ability of CHIP to reduce αSyn oligomerization by transfecting H4 human neuroglioma cells with the αSyn BiFC constructs αSynGC and GN-αSyn in the presence or absence of CHIP (Fig. 1) and found that CHIP dramatically reduced αSyn oligomerization as measured by intensity of GFP fluorescence (Fig. 2, A and B). As a control, we confirmed that CHIP had no effect on fluorescence intensity of enhanced green fluorescent protein (EGFP) (Fig. 2, A and B). To confirm that CHIP’s prevention of αSyn oligomerization is specific to αSyn, and not EGFP, we conducted co-immunoprecipitation experiments and determined that CHIP did not co-immunoprecipitates with EGFP alone (data not shown).

To confirm that the effect of CHIP on αSyn oligomerization is specific, we overexpressed domain mutant constructs of CHIP (CHIPΔU and CHIPΔTPR) with αSynGC and GN-αSyn (Fig. 2C) and determined that CHIPΔU, but not CHIPΔTPR, was able to significantly reduce the fluorescence produced by αSyn oligomer formation (Fig. 2D). Because the TPR domain of CHIP is essential for Hsp70 interactions, these data suggest that the CHIP-mediated reduction in αSyn oligomer formation occurs in an Hsp70-associated manner.

To quantify the CHIP domain mutant effects, we co-transfected H4 human neuroglioma cells with αSynGC, GN-αSyn, and either empty vector, CHIP, CHIPΔU, or CHIPΔTPR and measured GFP fluorescence intensity with fluorescent-activated cell sorting. In agreement with our observations of GFP pixel intensity (Fig. 2), fluorescent-activated cell sorting analysis demonstrated that CHIP and CHIPΔU mediate a reduction in αSyn oligomerization (Fig. 3, A and B). As a control, we verified that CHIP had no effect on αSyn-EGFP fluorescence, confirming that CHIP reduces αSyn oligomer formation in a TPR domain-dependent manner.

CHIP Reduces αSyn Oligomer-induced Toxicity

We have shown previously that overexpression of the BiFC pairs αSynGC and GN-αSyn results in increased αSyn oligomerization formation and cytotoxicity (19). To determine whether CHIP can rescue cytotoxicity conferred by oligomeric αSyn, H4 cells were cotransfected with αSynGC, GN-αSyn, and αSynGC and GN-αSyn together, in the presence or absence of CHIP or its domain mutants (CHIPΔU and CHIPΔTPR), and toxicity was monitored. As expected, CHIP and CHIPΔU rescued toxicity conferred by αSynGC and GN-αSyn coexpression (Fig. 4). By contrast, CHIP had no effect on αSyn toxicity induced by overexpression of individual αSyn constructs, i.e. αSynGC expressed alone (Fig. 4). These data suggest that CHIP preferentially reduces cytotoxicity conferred by a stabilized αSyn oligomeric species and not that of the less toxic, transient αSyn constructs. This experiment also confirmed that the TPR domain is critical for the toxicity-rescuing effects of CHIP because CHIPΔTPR was unable to rescue toxicity. These data are in agreement with CHIP preventing the formation of αSyn oligomers (Figs. 2 and 3).

CHIP Preferentially Degrades αSyn Oligomeric Species

To determine the mechanism of CHIP-mediated effects on αSyn, we examined the expression levels of αSyn in the presence or absence of CHIP under two conditions in which oligomeric species are to determine the mechanism of CHIP-mediated effects on αSyn, we examined the expression levels of αSyn in the presence or absence of CHIP under two conditions in which oligomeric species are, 1) less stabilized: αSynGC or GN-αSyn expressed individually or, 2) more stabilized: αSynGC and GN-αSyn expressed together. Total cell lysate and Tx-soluble and Tx-insoluble lysates were prepared. Interestingly, in the total cell lysate fraction, CHIP had only a modest effect on αSyn levels when αSynGC and GN-αSyn were expressed individually (Fig. 5A). However, when the αSyn BiFC proteins were coexpressed, CHIP significantly reduced the protein levels (Fig. 5A).

Previous studies have demonstrated that CHIP can shuttle proteins such as mutant tau, huntingtin, and ataxin to different soluble and insoluble cellular compartments (15–17, 23–25). To determine the effect of CHIP on αSyn solubility we extracted Tx-soluble and -insoluble fractions. We found that CHIP had little or no effect on levels of αSyn in the soluble or insoluble fraction when αSynGC and GN-αSyn were expressed individually (Fig. 5, B and C). However, when coexpressed, CHIP degrades αSyn in both the soluble and insoluble fractions (Figs. 5, B and C).

CHIP Reduces High Molecular Weight αSyn Oligomeric Species

The co-expression of αSynGC and GN-αSyn results in high molecular weight (HMW) oligomeric species when examined by native Western blot. To determine whether CHIP could degrade HMW αSyn oligomers we co-transfected H4 cells with αSynGC and GN-αSyn with an empty vector (pcDNA), CHIP, CHIPΔU, or CHIPΔTPR. Interestingly, CHIP efficiently degrades the HMW αSyn population. Furthermore, consistent with our previous data, the TPR domain, but not the U-box domain, is critical for this oligomeric degradation because CHIPΔTPR is ineffective (Fig. 6, A and B).

CHIP, but Not CHIPΔTPR, Coimmunoprecipitates with Hsp70

CHIP and Hsp70 form a complex between the amino-terminal TPR domain of CHIP and the carboxyl-terminal EEVD motif of Hsp70. Our data indicate that the TPR domain of CHIP is critical for the CHIP-mediated reduction of αSyn oligomers and associated toxicity. To determine whether the association of Hsp70 with CHIP, αSynGC, and GN-αSyn is critical, we co-transfected H4 cells with Hsp70, αSynGC, GN-αSyn, and CHIP or CHIPΔTPR. Both CHIP and CHIPΔTPR coimmunoprecipitate with GN-αSyn and αSynGC; however, only fulllength CHIP, and not CHIPΔTPR, coimmunoprecipitates with Hsp70 (Fig. 7). These data imply that Hsp70 is required for the CHIP-mediated reduction of αSyn oligomers and their associated toxicity.