Inactivation of Parkin by Oxidative Stress and C-terminal Truncations

Loss of parkin function is linked to autosomal recessive juvenile parkinsonism. Here we show that proteotoxic stress and short C-terminal truncations induce misfolding of parkin. As a consequence, wild-type parkin was depleted from a high molecular weight complex and inactivated by aggregation. Similarly, the pathogenic parkin mutant W453Stop, characterized by a C-terminal deletion of 13 amino acids, spontaneously adopted a misfolded conformation. Mutational analysis indicated that C-terminal truncations exceeding 3 amino acids abolished formation of detergent-soluble parkin. In the cytosol scattered aggregates of misfolded parkin contained the molecular chaperone Hsp70. Moreover, increased expression of chaperones prevented aggregation of wild-type parkin and promoted folding of the W453Stop mutant. Analyzing parkin folding in vitro indicated that parkin is aggregation-prone and that its folding is dependent on chaperones. Our study demonstrates that C-terminal truncations impede parkin folding and reveal a new mechanism for inactivation of parkin. Loss of parkin function is linked to autosomal recessive juvenile parkinsonism. Here we show that proteotoxic stress and short C-terminal truncations induce misfolding of parkin. As a consequence, wild-type parkin was depleted from a high molecular weight complex and inactivated by aggregation. Similarly, the pathogenic parkin mutant W453Stop, characterized by a C-terminal deletion of 13 amino acids, spontaneously adopted a misfolded conformation. Mutational analysis indicated that C-terminal truncations exceeding 3 amino acids abolished formation of detergent-soluble parkin. In the cytosol scattered aggregates of misfolded parkin contained the molecular chaperone Hsp70. Moreover, increased expression of chaperones prevented aggregation of wild-type parkin and promoted folding of the W453Stop mutant. Analyzing parkin folding in vitro indicated that parkin is aggregation-prone and that its folding is dependent on chaperones. Our study demonstrates that C-terminal truncations impede parkin folding and reveal a new mechanism for inactivation of parkin. Autosomal recessive juvenile parkinsonism (AR-JP), 1The abbreviations used are: AR-JPautosomal recessive juvenile parkinsonismwtwild typeBSAbovine serum albuminPBSphosphate-buffered salineERendoplasmic reticulumPDParkinson's diseaseYFPyellow fluorescent protein. the major cause of early onset parkinsonism, is characterized by mutations within the parkin gene. Parkin, a 465-amino acid protein, shows homology to ubiquitin at the N terminus and harbors a RING box near the C terminus, consisting of two RING finger motifs that flank a cysteine-rich domain (in-between RING fingers domain) (1Kitada, T., Asakawa, S., Hattori, N., Matsumine, H., Yamamura, Y., Minoshima, S., Yokochi, M., Mizuno, Y., and Shimizu, N. (1998) Nature, 392, 605-608Google Scholar, 2Morett E. Bork P. Trends Biochem. Sci. 1999; 24: 229-231Abstract Full Text Full Text PDF PubMed Scopus (129) Google Scholar). Functional studies established that parkin acts as a ubiquitin-protein isopeptide ligase and that pathogenic mutations compromise this activity (3Imai Y. Soda M. Takahashi R. J. Biol. Chem. 2000; 275: 35661-35664Abstract Full Text Full Text PDF PubMed Scopus (663) Google Scholar, 4Imai Y. Soda M. Inoue H. Hattori N. Mizuno Y. Takahashi R. Cell. 2001; 105: 891-902Abstract Full Text Full Text PDF PubMed Scopus (942) Google Scholar, 5Shimura H. Schlossmacher M.G. Hattori N. Frosch M.P. Trockenbacher A. Schneider R. Mizuno Y. Kosik K.S. Selkoe D.J. Science. 2001; 293: 263-269Crossref PubMed Scopus (973) Google Scholar, 6Zhang Y. Gao J. Chung K.K. Huang H. Dawson V.L. Dawson T.M. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 13354-13359Crossref PubMed Scopus (849) Google Scholar). As a consequence, substrates destined for proteasomal degradation via parkin might accumulate in parkin-deficient cells. Indeed, recent studies with cell culture models provide experimental evidence for such a scenario. It was shown that disease-related mutations in the parkin gene impair protein interactions of parkin with either a parkin substrate or another component of the ubiquitin ligase complex. Accumulation of Pael-R, one of the identified parkin substrates, causes endoplasmic reticulum (ER) stress, indicating that parkin has the potential to suppress unfolded protein stress-induced cell death (3Imai Y. Soda M. Takahashi R. J. Biol. Chem. 2000; 275: 35661-35664Abstract Full Text Full Text PDF PubMed Scopus (663) Google Scholar, 4Imai Y. Soda M. Inoue H. Hattori N. Mizuno Y. Takahashi R. Cell. 2001; 105: 891-902Abstract Full Text Full Text PDF PubMed Scopus (942) Google Scholar). Recent studies (7Staropoli J.F. McDermott C. Martinat C. Schulman B. Demireva E. Abeliovich A. Neuron. 2003; 37: 735-749Abstract Full Text Full Text PDF PubMed Scopus (351) Google Scholar) revealed that parkin deficiency potentiates the accumulation of cyclin E and promotes apoptosis in neuronal cells exposed to excitotoxic stress. Interestingly, parkin is a significant component of Lewy bodies, the histopathologic hallmark of PD (5Shimura H. Schlossmacher M.G. Hattori N. Frosch M.P. Trockenbacher A. Schneider R. Mizuno Y. Kosik K.S. Selkoe D.J. Science. 2001; 293: 263-269Crossref PubMed Scopus (973) Google Scholar, 8Chung K.K. Zhang Y. Lim K.L. Tanaka Y. Huang H. Gao J. Ross C.A. Dawson V.L. Dawson T.M. Nat. Med. 2001; 7: 1144-1150Crossref PubMed Scopus (676) Google Scholar, 9Schlossmacher M.G. Frosch M.P. Gai W.P. Medina M. Sharma N. Forno L. Ochiishi T. Shimura H. Sharon R. Hattori N. Langston J.W. Mizuno Y. Hyman B.T. Selkoe D.J. Kosik K.S. Am. J. Pathol. 2002; 160: 1655-1667Abstract Full Text Full Text PDF PubMed Scopus (279) Google Scholar). Furthermore, it has been shown that parkin is protective against the toxic effects of proteasomal dysfunction and mutant α-synuclein (10Petrucelli L. O'Farrell C. Lockhart P.J. Baptista M. Kehoe K. Vink L. Choi P. Wolozin B. Farrer M. Hardy J. Cookson M.R. Neuron. 2002; 36: 1007-1019Abstract Full Text Full Text PDF PubMed Scopus (512) Google Scholar), implying that the impact of parkin function and dysfunction might not be restricted to the entity of AR-JP. autosomal recessive juvenile parkinsonism wild type bovine serum albumin phosphate-buffered saline endoplasmic reticulum Parkinson's disease yellow fluorescent protein. It still remains enigmatic why dopaminergic neurons in the substantia nigra are highly vulnerable in AR-JP, as parkin as well as its substrates identified so far are not selectively expressed in these cells. However, an inherent feature of dopaminergic neurons is an elevated level of reactive oxygen and nitrogen species due to the metabolism of dopamine. In this study we show that oxidative and thermal stress causes misfolding and aggregation of wild-type parkin. In addition, a C-terminal deletion mutant associated with AR-JP, W453Stop, is constitutively misfolded, revealing that the C-terminal amino acids are essential for the native folding of parkin. Moreover, molecular chaperones interfere with the misfolding and aggregation of parkin in vivo and vitro, indicating a potential prophylactic or therapeutic strategy. DNA Constructs—The gene encoding human parkin was amplified from a human brain cDNA library and inserted into the pcDNA3.1 vector (Invitrogen). The following primers were used: forward 5′-gtgcatatgatagtgtttgtcaggttcaactccagc-3′ and reverse 5′-tggctacacgtcgaaccagtggtccc-3′. Amino acid deletions and substitutions were introduced by PCR cloning techniques. The construct pEYFP-N1-Hsp70 encodes human Hsp70 fused to yellow fluorescent protein (YFP) and was kindly provided by Richard I. Morimoto (11Kim S. Nollen E.A. Kitagawa K. Bindokas V.P. Morimoto R.I. Nat. Cell Biol. 2002; 4: 826-831Crossref PubMed Scopus (253) Google Scholar). The construct pHsp40 was a gift of William J. Welch; it was generated by inserting the gene encoding human Hsp40 into the pcDNA3.1 vector (Invitrogen). Antibodies—Polyclonal anti-parkin antiserum hP1 was raised against a recombinant N-terminal fragment of parkin (amino acids 1–210) that was expressed in and purified from bacteria. The mouse monoclonal antibody specific for inducible Hsp72 (C92) and the mouse monoclonal antibody N27, which identifies both the inducible (Hsp72) and the constitutive form (Hsp73) of Hsp70, have been described previously (12Welch W.J. Suhan J.P. J. Cell Biol. 1986; 103: 2035-2052Crossref PubMed Scopus (458) Google Scholar). Cell Culture and Transfections—N2a cells and SH-SY5Y cells were cultivated as described (13Winklhofer K.F. Heske J. Heller U. Reintjes A. Muranji W. Moarefi I. Tatzelt J. J. Biol. Chem. 2003; 278: 14961-14970Abstract Full Text Full Text PDF PubMed Scopus (56) Google Scholar). Cells were transfected by a liposome-mediated method using LipofectAMINE Plus Reagent (Invitrogen) according to the manufacturer's instructions. Detergent Solubility Assay and Proteolysis Experiments—Cells were washed twice with cold PBS, scraped off the plate, pelleted by centrifugation, and lysed in cold detergent buffer (0.1% Triton X-100 in PBS). The lysate was centrifuged at 15,000 × g for 20 min at 4 °C. The supernatant was then separated from the pellet. After adding Laemmli sample buffer to both fractions the samples were boiled for 10 min. To compare the relative distribution of parkin, equal percentages of the detergent-soluble and -insoluble fractions were analyzed by immunoblotting. For proteolysis experiments, cells were lysed as described above and incubated with proteinase K (Roche Applied Science) at the concentrations indicated at 4 °C for 30 min. The reaction was terminated by the addition of Pefabloc SC (Roche Applied Science) and boiling in Laemmli sample buffer. Residual parkin was detected by Western blotting. Stress Treatment and Proteasomal Inhibition—For the heat shock, cell culture dishes were placed into a water bath for the time and temperature indicated. For oxidative stress, cells were incubated in PBS containing 10 or 20 mm H2O2 for 30 min. ER stress was induced by exposing the cells to 10 μg/ml tunicamycin (Sigma) for 30 min. For inhibition of the proteasome, cells were incubated in the presence of 10 μm MG132 (Calbiochem) for 16 h. Western Blot Analysis and Indirect Immunofluorescence—Proteins were size-fractionated by SDS-PAGE (10% polyacrylamide) and transferred to nitrocellulose (Protran BA 85, Schleicher & Schüll) by electroblotting. The nitrocellulose membranes were blocked with 5% nonfat dry milk in PBST (PBS containing 0.1% Tween 20) for 30 min at room temperature and subsequently incubated with the primary antibody in PBST for 16 h at 4 °C. After extensive washing with PBST, the membranes were incubated with horseradish peroxidase-conjugated secondary antibody for 40 min at room temperature. Following washing with PBST, the antigen was detected with the enhanced chemiluminescence detection system (Amersham Biosciences) as specified by the manufacturer. Quantification was performed using AIDA 3.26 image analysis software (Raytest). Indirect immunofluorescence experiments were performed as described previously (14Winklhofer K.F. Reintjes A. Hoener M.C. Voellmy R. Tatzelt J. J. Biol. Chem. 2001; 276: 45160-45167Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar). Briefly, SH-SY5Y cells were grown on glass cover slides and fixed by immersion in cold methanol for 5 min. The fixed cells were incubated with the anti-parkin antiserum hP1 or the anti-Hsp72/73 antibody N27 (dilution 1:100) in PBS containing 1% bovine serum albumin for 45 min at 37 °C. After washing with PBS, incubation with Cy2- or Cy3-conjugated secondary antibodies (Dianova) (dilution 1:200) followed at 37 °C for 30 min. The washed coverslips were mounted onto glass slides and examined by phase contrast and fluorescence microscopy using a Zeiss Axiovert 200M microscope. Metabolic Labeling and Immunoprecipitation—Cells were starved for 30 min in methionine-free minimum Eagle's medium (Invitrogen) and subsequently labeled for 30 min with 300 μCi/ml Pro-mix l-35S in vitro cell label mix (Amersham Biosciences, >37 TBq/mmol) in methionine-free minimum Eagle's medium (pulse). For the chase, the labeling medium was removed, and the cells were washed twice and then incubated in complete medium for the time indicated. Radiolabeled cells were lysed in detergent buffer and fractionated into detergent-soluble and -insoluble fractions as described above. The supernatants were adjusted to 0.5% Sarkosyl. After pre-clearing with protein A beads (Pierce) for 30 min, the primary antibody hP1 was added, and the samples were incubated overnight at 4 °C. The antigen-antibody complexes were captured by the addition of immobilized protein A and then washed 3 times with detergent buffer. Proteins present in the immunoprecipitates were released from the protein A-Sepharose by the addition of Laemmli sample buffer containing 1% SDS and were heated at 100 °C for 5 min. Expression of Recombinant Proteins—Wild-type parkin and the W453Stop mutant were cloned into the pET-22b(+) vector (Novagen) by amplifying the open reading frame using the following primer pairs: forward 5′-ccg ccc tgg aag ctt cta cac gtc gaa cca gtg gtc cc-3′, reverse 5′-ccg ccc tgg aag ctt cac gtc gaa cca gtg gtc cc-3′ (wild-type), and forward 5′-ccg ccc tgg aag ctt cta cac gtc gaa cca gtg gtc cc-3′, reverse 5′-ccg ccc tgg aag ctt ctc gac gcc aca gtt cca gc-3′ (W453Stop), digesting the products with NdeI and HindIII, and ligating them into these sites in the vector. This produces the proteins without their terminal Stop codons and with C-terminal His6 epitope tags. The His-tagged proteins were expressed in the BL21 pLysS strain of Escherichia coli by induction with 1 mm isopropyl-1-thio-β-d-galactopyranoside at 37 °C for 4 h. The proteins were present in inclusion bodies and were purified under denaturing conditions in 6 m guanidinium chloride by affinity chromatography using nickel-nitrilotriacetic acid beads (Qiagen). The proteins were eluted from the beads at pH 4.0, again in 6 m guanidinium chloride. The concentrations of the eluted proteins were calculated by measuring absorbance at 280 nm and dividing these values by the molar extinction coefficients. The eluted proteins were then diluted to 8 μm in elution buffer. Preparation of Mammalian Chaperones—Mammalian Hsp40 and Hsp70 were prepared as described previously. Briefly, human HDJ-1 (Hsp40) was purified as a recombinant protein from E. coli (15Minami Y. Hoehfeld J. Ohtsuka K. Hartl F.U. J. Biol. Chem. 1996; 271: 19617-19624Abstract Full Text Full Text PDF PubMed Scopus (285) Google Scholar) and bovine Hsc70 (Hsp73) was purified from bovine brain (16Schlossman D.M. Schmid S.L. Braell W.A. Rothman J.E. J. Cell Biol. 1984; 99: 723-733Crossref PubMed Scopus (315) Google Scholar). In Vitro Aggregation Assay—8 μm parkin was diluted 1:100 with slow shaking into aggregation buffer (20 mm HEPES-KOH, pH 7.4, 150 mm KCl, 3 mm MgCl2, 2 mm ATP, 20 mm phosphoenolpyruvate, 0.2 mg/ml pyruvate kinase) with 0.2 μm Hsp40 and 0.4 μm Hsp70, 0.6 μm BSA or no additions. The reaction was incubated at room temperature for 3 h and then ultracentrifuged at 50,000 × g for1hat4 °C. The supernatant was removed and precipitated with trichloroacetic acid, and the pellet was resuspended in 2× Laemmli sample buffer with 0.1 m dithiothreitol. Supernatant and pellet fractions were separated by SDS-PAGE (10% polyacrylamide), and the proteins were transferred to nitrocellulose membranes. Parkin was then detected by immunoblotting using the hP1 antibody. Gel Filtration—Cells were lysed in 1% Triton X-100, and cell lysates were centrifuged at 27 × g for 5 min to remove cell nuclei. A 200-μl sample (∼500 μg of protein) was loaded onto a Superdex 200 HR 10/30 column (Amersham Biosciences) and eluted with 0.1% Triton X-100 at a flow rate of 0.4 ml/min. Proteins present in the eluted fractions were precipitated by trichloroacetic acid and analyzed by Western blotting. The gel filtration system was calibrated using the following standard proteins: thyroglobulin (669 kDa), ferritin (440 kDa), aldolase (168 kDa), bovine serum albumin (67 kDa), and ovalbumin (43 kDa). Thermal and Oxidative Stress Induces Misfolding of Wildtype Parkin—The metabolism of dopamine gives rise to various reactive oxygen and nitrogen species; consequently, nigral dopaminergic neurons are particularly exposed to oxidative stress. To analyze the impact of cellular stress on the folding state of parkin, murine N2a and human SH-SY5Y neuroblastoma cells transiently transfected with wild-type (wt) parkin were subjected to thermal or oxidative stress. To monitor a possible conformational change of parkin, the cells were lysed in buffer containing non-ionic detergent (0.1% Triton X-100) and fractionated by centrifugation into the detergent-soluble and -insoluble fraction. Under physiological conditions parkin was found almost exclusively in the detergent-soluble fraction (Fig. 1, A and B, S). However, under stressful conditions, be it thermal stress (46 °C for 10 min) or oxidative stress (10 and 20 mm H2O2 for 30 min), the majority of parkin adopted a detergent-insoluble conformation (Fig. 1, A and B, P). Of note, the parkin aggregates formed after proteotoxic stress were also insoluble in buffer containing the ionic detergent deoxycholate (data not shown). A quantitative analysis revealed that about 90% of parkin was soluble under physiological conditions. Depending on the severity of the cellular stress, up to 86% of parkin was converted into an insoluble form (Fig. 1D). We then examined the solubility of parkin in the presence of tunicamycin. This drug interferes with N-linked glycosylation leading to the accumulation of misfolded proteins specifically in the ER. Under ER stress parkin remained in the detergent-soluble phase, indicating that the folding state of parkin was unaffected (Fig. 1C, TM). Of note, the stress conditions applied in our study did not cause protein aggregation in general; analysis of cellular proteins by SDS-PAGE and Coomassie Blue staining did not show an obvious increase in the pool of detergent-insoluble proteins (Fig. 1E). Long-term effects of stress-induced parkin aggregation were investigated by a time course (Fig. 1F). Cells subjected to oxidative stress were either harvested directly afterward (0 h) or placed in fresh medium and analyzed after an additional 4 or 24 h. 4 h after the stress the pool of detergent-soluble parkin remained depleted, indicating that misfolded parkin was not resolubilized after the stress ceased. After 24 h, detergent-soluble parkin reappeared, most likely due to de novo synthesis. Moreover, detergent-insoluble parkin disappeared 24 h after H2O2 treatment, indicating cellular degradation. Indeed, inhibition of the proteasome by MG132 or lactacystin interfered with the clearance of parkin aggregates (data not shown). These experiments established that thermal and oxidative stress induces the formation of detergent-insoluble parkin, whereas ER stress has no impact on the folding state of parkin. Parkin aggregates do not accumulate, and 24 h after the stressful event insoluble parkin is no longer detectable. Spontaneous Misfolding of a Parkin Mutant Linked to AR-JP—Pathogenic parkin mutations linked to AR-JP are distributed throughout the parkin gene and include missense, nonsense, and frameshift mutations as well as exon deletions and multiplications (1Kitada, T., Asakawa, S., Hattori, N., Matsumine, H., Yamamura, Y., Minoshima, S., Yokochi, M., Mizuno, Y., and Shimizu, N. (1998) Nature, 392, 605-608Google Scholar, 17West A. Periquet M. Lincoln S. Lucking C.B. Nicholl D. Bonifati V. Rawal N. Gasser T. Lohmann E. Deleuze J.F. Maraganore D. Levey A. Wood N. Durr A. Hardy J. Brice A. Farrer M. Am. J. Med. Genet. 2002; 114: 584-591Crossref PubMed Scopus (192) Google Scholar, 18Periquet M. Lucking C. Vaughan J. Bonifati V. Durr A. De Michele G. Horstink M. Farrer M. Illarioshkin S.N. Pollak P. Borg M. Brefel-Courbon C. Denefle P. Meco G. Gasser T. Breteler M.M. Wood N. Agid Y. Brice A. Am. J. Hum. Genet. 2001; 68: 617-626Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar, 19Lucking C.B. Durr A. Bonifati V. Vaughan J. De Michele G. Gasser T. Harhangi B.S. Meco G. Denefle P. Wood N.W. Agid Y. Brice A. N. Engl. J. Med. 2000; 342: 1560-1567Crossref PubMed Scopus (1288) Google Scholar, 20Hedrich K. Kann M. Lanthaler A.J. Dalski A. Eskelson C. Landt O. Schwinger E. Vieregge P. Lang A.E. Breakefield X.O. Ozelius L.J. Pramstaller P.P. Klein C. Hum. Mol. Genet. 2001; 10: 1649-1656Crossref PubMed Scopus (138) Google Scholar, 21Abbas N. Lucking C.B. Ricard S. Durr A. Bonifati V. De Michele G. Bouley S. Vaughan J.R. Gasser T. Marconi R. Broussolle E. Brefel-Courbon C. Harhangi B.S. Oostra B.A. Fabrizio E. Bohme G.A. Pradier L. Wood N.W. Filla A. Meco G. Denefle P. Agid Y. Brice A. Hum. Mol. Genet. 1999; 8: 567-574Crossref PubMed Scopus (496) Google Scholar). Some of the parkin mutants have been shown to interfere with the ubiquitin-protein isopeptide ligase activity of parkin due to impaired protein interactions. This might be an interaction of parkin with a ubiquitin carrier protein with another component of the ubiquitin ligase complex, or with a parkin substrate destined for degradation (3Imai Y. Soda M. Takahashi R. J. Biol. Chem. 2000; 275: 35661-35664Abstract Full Text Full Text PDF PubMed Scopus (663) Google Scholar, 4Imai Y. Soda M. Inoue H. Hattori N. Mizuno Y. Takahashi R. Cell. 2001; 105: 891-902Abstract Full Text Full Text PDF PubMed Scopus (942) Google Scholar, 5Shimura H. Schlossmacher M.G. Hattori N. Frosch M.P. Trockenbacher A. Schneider R. Mizuno Y. Kosik K.S. Selkoe D.J. Science. 2001; 293: 263-269Crossref PubMed Scopus (973) Google Scholar, 6Zhang Y. Gao J. Chung K.K. Huang H. Dawson V.L. Dawson T.M. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 13354-13359Crossref PubMed Scopus (849) Google Scholar, 7Staropoli J.F. McDermott C. Martinat C. Schulman B. Demireva E. Abeliovich A. Neuron. 2003; 37: 735-749Abstract Full Text Full Text PDF PubMed Scopus (351) Google Scholar, 8Chung K.K. Zhang Y. Lim K.L. Tanaka Y. Huang H. Gao J. Ross C.A. Dawson V.L. Dawson T.M. Nat. Med. 2001; 7: 1144-1150Crossref PubMed Scopus (676) Google Scholar, 22Shimura H. Hattori N. Kubo S. Mizuno Y. Asakawa S. Minoshima S. Shimizu N. Iwai K. Chiba T. Tanaka K. Suzuki T. Nat. Genet. 2000; 25: 302-305Crossref PubMed Scopus (1741) Google Scholar). We included the parkin mutant W453Stop in our analysis, a pathogenic mutant with a small C-terminal truncation (Fig. 2A). We found this mutant particularly interesting because of the fact that the C-terminal truncation of 13 amino acids does not extend into the second RING finger motif. The expression analysis in transiently transfected N2a and SH-SY5Y cells revealed that the W453Stop mutant spontaneously adopted a misfolded conformation. Although the majority of wt-parkin was soluble, over 90% of the W453Stop mutant was found in the detergent-insoluble fraction (Fig. 2, B and C). Thus, the biochemical phenotype of the W453Stop mutant under physiological conditions was similar to that of wt-parkin subjected to thermal or oxidative stress. Notably, the W453Stop mutant was not only insoluble in Triton X-100 but also in the ionic detergent deoxycholate (Fig. 2D). Experimental evidence for the assumption that variant solubility profiles were the consequence of conformational differences between wt-parkin and the W453Stop mutant was provided by a limited proteolytic digestion. In contrast to wt-parkin, which was almost completely degraded at the highest proteinase K concentration, a significant resistance against proteinase K was observed for the W453Stop mutant (Fig. 2E). To answer the question if misfolding of parkin is a specific feature of the W453Stop mutant, we analyzed two pathogenic parkin mutants with a point mutation within (G430D) or close to (T415N) the second RING finger motif. As shown in Fig. 2F, the solubility in detergents of the two point mutants was comparable with that of wt-parkin, indicating that differences in the folding state might not generally account for a loss of parkin function. We next examined the stability of the W453Stop mutant in comparison to wt-parkin. Transiently transfected N2a cells were metabolically labeled with [35S]methionine, and parkin was analyzed by immunoprecipitation. Wt-parkin turned out to be a relatively stable protein; after a chase of 24 h about 60% of the protein was still detectable (Fig. 2, G and H). In contrast, the stability of the soluble fraction of the W453Stop mutant was significantly reduced; after 6 h less than 20% of the protein was left. We were not able to analyze the pellet fraction from metabolically labeled cells, because the conditions required to solubilize the pellet fraction did not permit antibody binding and hence immunoprecipitation. However, the soluble fraction of the W453Stop mutant was not increased in the presence of the proteasomal inhibitor MG132 after a chase of 2 h, indicating that the short lifetime of detergent-soluble W453Stop was not due to proteasomal degradation but rather to a shift into the detergent-insoluble fraction (Fig. 2I). To analyze the impact of proteasomal degradation on the detergent-insoluble fraction of mutant parkin, we analyzed the pellet fraction by Western blotting. After proteasomal inhibition the amount of detergent-insoluble W453Stop was significantly increased (Fig. 2J). The biochemical characterization of the W453Stop parkin mutant revealed that in contrast to wt-parkin this mutant rapidly adopts a misfolded conformation under physiological conditions. Detergent-insoluble W453Stop does not accumulate in the cell but is degraded by the proteasome, similarly to the stress-induced misfolded wt-parkin. Misfolding of Parkin Interferes with Its Assembly into a Higher Molecular Weight Complex—Previous studies indicated that parkin functions in a multiprotein ubiquitin ligase complex; furthermore, parkin was identified as a component of a high molecular weight complex isolated from brain (7Staropoli J.F. McDermott C. Martinat C. Schulman B. Demireva E. Abeliovich A. Neuron. 2003; 37: 735-749Abstract Full Text Full Text PDF PubMed Scopus (351) Google Scholar, 23Imai Y. Soda M. Hatakeyama S. Akagi T. Hashikawa T. Nakayama K.I. Takahashi R. Mol. Cell. 2002; 10: 55-67Abstract Full Text Full Text PDF PubMed Scopus (435) Google Scholar). To analyze the presence of parkin complexes in cell culture, we performed gel filtration experiments with transiently transfected N2a cells. Cells were lysed in 1% Triton X-100, and the lysates were centrifuged at 27 × g for 5 min to remove cell nuclei before loading onto the column. The majority of wt-parkin eluted at the molecular mass expected for the monomer (52 kDa) (Fig. 3A, wt-parkin); in addition, wt-parkin was found in higher molecular weight fractions corresponding to the large parkin-containing complex derived from cerebral cortex (7Staropoli J.F. McDermott C. Martinat C. Schulman B. Demireva E. Abeliovich A. Neuron. 2003; 37: 735-749Abstract Full Text Full Text PDF PubMed Scopus (351) Google Scholar). When we analyzed wt-parkin derived from cells that had been subjected to a heat shock (46 °C, 10 min), wt-parkin was partly found in the pellet after the low speed centrifugation, indicating that extremely large parkin aggregates must have been formed after the heat shock (Fig. 3B). Moreover, wt-parkin from heat-shocked cells was no longer found in the high molecular weight fraction (Fig. 3A, wt-parkin + hs). Remarkably, the low speed centrifugation was sufficient to quantitatively deplete the cell lysate from the W453Stop mutant (Fig. 3B). Only a small amount of protein eluted at the expected molecular weight fraction (Fig. 3A, W453Stop, fraction 16). Although the two C-terminal parkin point mutants T415N and G430D did not display overt conformational differences compared with wt-parkin, we speculated that mutations within or close to the second RING finger motif might impair complex formation. Indeed, we were not able to detect high molecular weight complexes in lysates of T415N-parkin- and G430D-parkin-expressing cells (Fig. 3A, T415N). Determination of C-terminal Amino Acids Essential for Correct Folding of Parkin—The data presented above indicated that the C-terminal 13 amino acids are critical for the correct folding of wt-parkin. To define the maximal C-terminal deletion tolerated in the native folding pathway of parkin, we constructed a series of successively smaller deletion mutants (Fig. 4). Deletion of as little as 4 C-terminal amino acids completely interfered with parkin folding; the W462Stop mutant was exclusively found in the detergent-insoluble fraction. Shorter deletions, truncating parkin only by one or two C-terminal amino acids, did not interfere significantly with the folding of parkin; the majority of the D464Stop and the V465Stop mutant was detergent-soluble. To analyze specifically a possible role of the last three amino acids in parkin folding, we fused the FDV motif to the C terminus of the W453Stop mutant. This motif did not render the W453FDV mutant detergent-soluble, emphasizing that the presence of the last three C-terminal amino acids was not sufficient to restore parkin folding (Fig. 4). Thus, parkin requires an intact C-terminal region, except for the last 2 amino acids, to adopt its native conformation. The W453Stop Parkin Mutant Forms Aggregates in the Cytosol—To investigate the cellular localization of the misfolded W453Stop mutant, we performed indirect immunofluorescence studies with transiently transfected SH-SY5Y cells. For wt-parkin, a homogenous distribution throughout the cytosol was observed (Fig. 5, wt-parkin). In contrast, the W453Stop mutant was found in scattered aggregates (Fig. 5, W453St