Available online at www.sciencedirect.com R
Experimental Cell Research 288 (2003) 60 – 69
www.elsevier.com/locate/yexcr
Proteasome inhibition induces inclusion bodies associated with intermediate filaments and fragmentation of the Golgi apparatus Masaru Harada,a,* Hiroto Kumemura,a M. Bishr Omary,b Takumi Kawaguchi,a Noriko Maeyama,a Shinichiro Hanada,a Eitaro Taniguchi,a Hironori Koga,a Tatsuo Suganuma,c Takato Ueno,a and Michio Sataa a
Second Department of Medicine and Research Center for Innovative Cancer Therapy, Kurume University School of Medicine, Kurume, Japan b Digestive Disease Center, Stanford University School of Medicine, Palo Alto, CA 94304, USA c Department of Anatomy, Miyazaki Medical College, Miyazaki, Japan Received 29 November 2002, revised version received 25 February 2003
Abstract The ubiquitin-proteasome system is involved in a variety of biological processes. Inclusion bodies associated with intermediate filaments (IFs) and ubiquitin are observed in various diseases; however, the precise mechanisms of formation and the pathological significance of inclusion bodies have not been fully understood. We examined the effect of proteasome inhibitors on the structure of IF using anti-cytokeratin antibodies or transfection of green fluorescent protein-fused cytokeratin 18 in a hepatoma cell line, Huh7. Intracellular organelles were visualized by immunofluorescent and electron microscopies. Proteasome inhibitors induced IF inclusions associated with ubiquitin. Electron microscopic examination revealed inclusion bodies surrounded by filamentous structures. Autophagic vacuoles and lysosomes were frequently observed, and the organization of the Golgi apparatus was disrupted in these cells. After the removal of the proteasome inhibitors, the IF network and organization of the Golgi apparatus were restored. The IF inclusions could be induced by inhibition of the proteasome function. IF inclusions induced fragmentation of the Golgi apparatus and might inhibit the function of this important station of membrane traffic. The IF inclusions disappeared by restoring proteasome function, and autophagy and lysosomal degradation might be, at least in part, associated with the elimination of inclusion bodies. © 2003 Elsevier Science (USA). All rights reserved. Keywords: Cytokeratin; Mallory body; Ubiquitin
Introduction Mammalian cells possess mainly two distinct proteolytic pathways. The lysosomal pathway is the general mechanism involved in the degradation of long-lived proteins and proteins entered from the extracellular milieu. The ubiquitin–proteasome system (UPS)1 is involved in a wide variety of biological processes including the cell cycle, apoptosis, metabolism, sig* Corresponding author. Second Department of Medicine, Kurume University School of Medicine, 67 Asahi-Machi, Kurume 830-0011, Japan. Fax: ⫹81-942-34-2623. E-mail address:
[email protected] (M. Harada). 1 Abbreviations used: ALLN, acetyl-leucyl-leucyl-norleucinal; ALS, amyotrophic lateral sclerosis; CK, cytokeratin; DMSO, dimethyl sulfoxide; FITC, fluorescein isothiocyanate; GalT, 1, 4-galactosyltransferase; GFP, green fluorescent protein; IF, intermediate filament; MB, Mallory body;
nal transduction, immune response, and quality control for misfolded proteins which could form protein aggregates. The large assembled proteasome acts as a protein-destroying machine in an ATP-dependent manner. Degradation of proteins by this pathway involves two processes: covalent attachment of multiple ubiquitin molecules to the target protein and degradation of the protein by 26S proteasome [1,2]. Several lines of evidence suggest a role of UPS in the pathogenesis of various diseases. Accumulation of protein aggregates associated with ubiquitin and certain disease-characteristic proteins has been reported in various neurodegenerative diseases including familial amyotrophic lateral sclerosis (ALS), Alzhei-
MT, microtubule; MTOC, microtubule-organizing center; TRITC, tetramethylrhodamine isothiocyanate; UPS, ubiquitin-proteasome system.
0014-4827/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0014-4827(03)00162-9
M. Harada et al. / Experimental Cell Research 288 (2003) 60 – 69
mer disease, Parkinson disease, Huntington disease, spinobulbar muscular atrophy, spinocerebellar ataxias types 1, 3, and 7, and prion encephalopathies [3,4]. These diseases are caused by modifications in the protein structures (mainly misfolding), and this new entity of disorders has been proposed as conformational disease [5]. In ordinary circumstances, cells refold the abnormal misfolded proteins in chaperon-dependent processes or degrade them via the UPS. However, when the capacity of the proteasome is exceeded by the production of misfolded proteins, the formation of intracellular protein aggregates occurs. This structure is associated with intermediate filaments (IFs) and ubiquitin, and termed aggresomes [5]. The formation of aggresomes was induced by various disease-characteristic proteins such as cystic fibrosis transmembrane regulator, huntingtin fragment, multidrug resistance protein 2, and Wilson disease protein (ATP7B) [5–9]. The existence of IF inclusions associated with ubiquitin is a common feature of inclusion bodies observed in various diseases as well, and they were reported as Lewy bodies in Parkinson disease, Pick bodies in Pick disease, cytoplasmic bodies in a specific myopathy, and Mallory bodies (MBs) in alcoholic liver diseases [10]. MBs are cytoplasmic inclusions in hepatocytes, and are observed in alcoholic and nonalcoholic steatohepatitis, copper toxicity, drug toxicity, and hepatocellular carcinoma. They are also associated with IFs and ubiquitin [11]. Therefore, aggresomes and IF inclusion bodies including MBs are similar, and elucidation of the formation of inclusion bodies associated with IFs and ubiquitin is very important in understanding the pathogenesis of various diseases. However, the precise mechanisms of the formation of IF inclusions and their pathological significance have not been fully understood. Furthermore, it is not clear whether the formation of inclusions causes cellular dysfunction or whether it is merely a defense mechanism of cells to various stresses [5]. In the present study, we examined the effect of proteasome inhibitors on the formation of inclusion bodies associated with IFs in cultured cells. In addition, we further examined the features of the intracellular organelles in these cells. Materials and methods Cells A human hepatoma cell line established from a hepatocellular carcinoma [12], Huh7 cell, and a kidney-derived
61
immortalized cell line, HEK293 cell, were cultured in Dulbecco’s modified Eagle’s medium (Sigma, St. Louis, MO) supplemented with 10% fetal calf serum (Wako Pure Chemical Industries, Ltd., Osaka, Japan), penicillin (100 U/ml, crystalline penicillin G Meiji, Meiji Seika Kaisya, Tokyo, Japan), and streptomycin (0.1 mg/ml, Meiji Seika Kaisya) at 37°C in 5% CO2. The original components of IFs of Huh7 cells are cytokeratins (CKs); however, the cell possesses both CKs and vimentin (data not shown). HEK293 cells possess only vimentin as a component of IFs (data not shown). cDNAs were transiently transfected into cultured cells using Effecten Transfection Reagent (Qiagen GmbH, Hilden, Germany) according to the manufacturer’s recommendations at 24 h after plating. Construction of cDNAs Green fluorescent protein (GFP)-tagged CK18 cDNA construct (GFP–CK18) was prepared by fusing enhanced GFP to the NH2 termini of human CK18 cDNAs. The cDNAs for CK18 were subcloned into a pBluescript SK(⫹) plasmid. To produce GFP–CK18 fusion proteins, the EcoRI fragment, including the entire coding region of CK18, was ligated into the pEGFP-C2 vector (Clontech Laboratories Japan, Tokyo, Japan) and digested with EcoRI; the site was located at the 3⬘ end of enhanced GFP cDNA. The insert was verified by sequencing using flanking primers. pcDNA3.1flag-ubiquitin was a kind gift from Dr. F. Tokunaga. Antibodies and reagents The following antibodies were used: mouse monoclonal anti-␥-adaptin antibody (Sigma); rabbit polyclonal anti-calnexin antibody (StressGen Biotechnologies Corp., Victoria, Canada); rabbit polyclonal anti-cathepsin D antibody (Upstate Biotechnology Incorporated, New York, NY); mouse monoclonal anti-CK18 antibody (Sigma); mouse monoclonal anti-CKs type I and II antibodies (Progen Biotechnik GMBH, Heidelberg, Germany); mouse monoclonal antiflag antibody (Sigma); mouse monoclonal anti-1, 4-galactosyltransferase (GalT) antibody [13]; mouse monoclonal anti-lysosome-associated membrane protein 1 antibody (a kind gift from Professor J.T. August, Johns Hopkins University) [14]; mouse monoclonal anti-␣-antibody, anti-␥-
Fig. 1. Confocal laser scanning microscopic images of intermediate filaments in nontransfected (A and B) or GFP–CK18-transfected Huh7 cells (C–E). IFs were labeled with anti-pancytokeratin antibodies (A, B, D, and E). GFP signals and TRITC signals were completely colocalized and showed a filamentous network of IF (C–E). Green: (A and B) pancytokeratin; (C and E) GFP-CK18. Red: (D and E) pancytokeratin. Abbreviations: CK, cytokeratin; GFP, green fluorescent protein. Fig. 2. Confocal laser scanning microscopic images of intermediate filaments in nontransfected (A) or GFP–CK18-transfected (B, C, D, and E) Huh7 and HEK293 (F and G) cells. Cells were treated with proteasome inhibitors for 24 h (A, B, and G, ALLN; C, lactacystin; D, MG132). Cells in E were treated with ALLN for 24 h and further incubated with ALLN-free medium for 48 h. Green: (A) pancytokeratin; (B–E) GFP–CK18; (F and G) vimentin. Abbreviations: ALLN, acetyl-leucyl-leucyl-norleucinal; CK, cytokeratin; GFP, green fluorescent protein. Fig. 3. Confocal laser scanning microscopic images of ␥-tubulin and intermediate filaments of GFP–CK18-transfected cells. Cells in D–F were treated with ALLN. Green: GFP-CK18; Red: ␥-tubulin. Abbreviations: ALLN, acetyl-leucyl-leucyl-norleucinal; CK, cytokeratin; GFP, green fluorescent protein.
62
M. Harada et al. / Experimental Cell Research 288 (2003) 60 – 69
M. Harada et al. / Experimental Cell Research 288 (2003) 60 – 69
63
Fig. 4. Confocal laser scanning microscopic images of ubiquitin and intermediate filaments in GFP–CK18-transfected Huh7. Cell in G–L were transfected with flag-tagged ubiquitin and cells in D–F and J–L were treated with ALLN. Green: GFP–CK18. Red: (B, C, E, and F) ubiquitin; (H, I, K, and L) flag. Abbreviations: ALLN, acetyl-leucyl-leucyl-norleucinal; CK, cytokeratin; GFP, green fluorescent protein. Fig. 5. Confocal laser scanning microscopic images of the Golgi apparatus in GFP–CK18-transfected Huh7 (A–L) cells stained with anti-GalT antibody. Cells were treated with proteasome inhibitors. ALLN (D–I), for 24 h. Cells in J–L were treated with ALLN for 24 h, and chased with ALLN-free medium for 48 h. Green: GFP-CK18; Red: GalT. Abbreviations: ALLN, acetyl-leucyl-leucyl-norleucinal; CK, cytokeratin; GalT, 1, 4-galactosyltransferase; GFP, green fluorescent protein.
tubulin, and anti-vimentin antibodies (Sigma); mouse monoclonal anti-ubiquitin antibody (Santa Cruz Biotechnology, Inc., CA); fluorescein isothiocyanate (FITC)-conjugated and tetramethylrhodamine isothiocyanate (TRITC)conjugated rabbit anti-mouse immunoglobulins (Dako); and TRITC-conjugated swine anti-rabbit immunoglobulins (Dako). Acetyl-leucyl-leucyl-norleucinal (ALLN) (7 g/ml, Calbiochem, La Jolla, CA), lactacystine (20 M, Kyowa, Tokyo, Japan), and MG 132 (10 M, Calbiochem) were
used as proteasome inhibitors. Stock solutions were prepared as follows and stored in small aliquots at ⫺20°C: ALLN, 350 mM in dimethyl sulfoxide (DMSO); lactacystine, 2.65 mM in water; MG132, 10 mM in DMSO. Immunofluorescence Cells on glass coverslips were fixed in freshly prepared 3% paraformaldehyde in phosphate-buffered saline (PBS)
64
M. Harada et al. / Experimental Cell Research 288 (2003) 60 – 69
for 30 min, and permeabilized in 0.2% Triton X-100 (Wako) in PBS for 10 min or 0.1% saponin (Wako) in PBS for 30 min. Transfection of cDNAs was performed at 48 h before fixation, when the constructs were transiently transfected. Nonspecific binding was blocked by incubation with Protein Block Serum Free (Dako) for 30 min followed by incubation with the primary antibodies for 1 h and the secondary antibodies for 1 h. Confocal laser scanning microscopes (Fluoview FV 300; Olympus, Tokyo, Japan, and LSM5 PASCAL; Carl Zeiss, Jena, Germany) equipped with an Argon/Krypton laser and an Argon/He-Ne laser were used for observation, respectively. For double labeling analyses, images were acquired sequentially using separate excitation wavelengths of 488 nm for GFP or FITC and 568 nm for TRITC, and then merged. Image analysis of the IF inclusions IF network of the cultured cells was examined by immunofluorescence, and the percentage of IF inclusion-containing cells was counted. Data were derived from three independent experiments. Electron microscopy Cells were fixed with 1.0% glutaraldehyde, postfixed in 1% osmium tetroxide, dehydrated in a graded ethanol series, and embedded in Quetol 812 (Nisshin EM., Tokyo, Japan). Ultrathin sections were stained with uranyl acetate and lead citrate and examined with a transmission electron microscope (JEM 2000-EX, JEOL, Tokyo, Japan).
Results The intracellular distribution of endogeneous IFs was examined by immunofluorescence in Huh7 cells, and fine cytoplasmic networks were observed (Figs. 1A and B). In GFP–CK18-transfected cells, typical cytoplasmic IF networks composed of GFP–CK18 were observed (Fig. 1C). When GFP–CK18-transfected cells were stained with antiCK18 or anti-pan-CKs antibodies, GFP signals and immuoreactive signals for CK18 or pan-CKs (Figs. 1C–E) coincided completely (Figs. 1C–E). GFP alone was observed throughout the cell, when pEGFP-C2 was transfected (data not shown). DMSO alone did not affect the distribution of CK18 or GFP–CK18 (data not shown). The effect of ALLN, a proteasome inhibitor, on the IFs was examined in Huh7 cells by immunofluorescence for pan-CKs or GFP–ATP7B. At 24 h after the treatment with ALLN, fine cytoplasmic networks of IF disappeared and endogenenous CKs aggregated as IF inclusions at the perinuclear site (Fig. 2A, Table 1). In GFP–CK18-transfected and ALLN-treated cells, GFP–CK18 aggregated (Fig. 2B). The effects of the other proteasome inhibitors were also examined, and GFP–CK18 aggregated in lactacystin-treated
Table 1 Percentage of cells with IF inclusions after the treatment with ALLN (%) 0h 12 h 24 h
1.8 ⫾ 0.4 61.4 ⫾ 6.2 91.0 ⫾ 2.8
Note. Cells were treated with ALLN for indicated hours. The values are expressed as means ⫾ SD. Abbreviations: ALLN, acetyl-leucyl-leucylnorleucinal; IF, intermediate filament.
(Fig. 2C) or MG132-treated (Fig. 2D) cells. In these cells, endogeneous CKs also aggregated (data not shown). The results coincided completely among these three agents. This phenomenon was not specific for CKs, because the vimentin filaments were aggregated by the treatment with ALLN in HEK293 cells (Figs. 2F and G). Previous reports showed that IF inclusions associated with disease-characteristic proteins, termed aggresomes, were formed at the microtubule-organizing center (MTOC) [5]. Therefore, we examined the relation between the IF inclusions and ␥-tubulin, a MTOC marker (Figs. 3A–C) [15,16]. When ALLN-treated GFP–CK18-transfected Huh7 cells were stained with anti-␥-tubulin antibody, IF inclusions were localized near ␥-tubulin (Figs. 3D–F), indicating that the IF inclusions were localized near the MTOC. Therefore, IF inclusions induced by proteasome inhibitors were similar to the aggresomes previously reported in diseasecharacteristic protein-expressing cells [5]. The relation between IF inclusions and ubiquitin was examined by two methods: an anti-ubiquitin antibody and transfection of flag-tagged ubiquitin. Ubiquitin detected by immunofluorescence was diffusely observed in the cytoplasm in GFP–CK18-transfected Huh7 cells (Figs. 4A–C); however, it was observed as inclusions and localized near the IF inclusions in ALLN-treated cells (Figs. 4D–F). In GFP–CK18 and flag-tagged ubiquitin-transfected Huh7 cells, immunofluorescent signals for flag were observed in the cytoplasm and nucleus (Figs. 4G–I). When these cells were treated with ALLN, the signals for flag were observed in the nucleus and IF inclusions composed of GFP–CK18 (Figs. 4J–L). These results indicated that IF inclusions produced by ALLN treatment were associated with ubiquitin. The relationship between IF inclusions induced by ALLN and intracellular organelles was examined using anti-calnexin antibody as an endoplasmic reticulum marker, anti-GalT antibody as a Golgi marker, anti-lysosome-associated membrane protein 1 antibody as a late endosomelysosome marker, and anti-cathepsin D antibody as a lysosome marker, in GFP–CK18-transfected and ALLN-treated Huh7 cells (data not shown). IF inclusions were not colocalized with these organelle markers. Although the Golgi apparatus was observed as a Golgi ribbon in normal cells (Figs. 5A–C and 6A–C), it was observed as a scattered pattern in ALLN-treated cells (Figs. 5D–F). Therefore, we further examined the effect of proteasome inhibitors on the shape of the Golgi apparatus using two Golgi markers, GalT
M. Harada et al. / Experimental Cell Research 288 (2003) 60 – 69
and ␥-adaptin. Exposure to proteasome inhibitors ALLN (Figs. 5D–I and 6D–F), lactacystin (Fig. 6G–I), and MG132 (Figs. 6J–L) for 24 h resulted in fragmentation of the Golgi apparatus. The effect of ALLN on the shape of the Golgi apparatus was also examined in HEK293 cells, and ALLN induced fragmentation of the Golgi apparatus (data not shown). These results indicate that proteasome inhibition affects the organization of the Golgi apparatus, and this phenomenon does not depend on the cell type. The change in the Golgi structure was linked to a change in IF, because a small number of cells with a normal IF network had normal Golgi structures even after the treatment with ALLN (Figs. 5G–I). Because microtubule is important for the organization of the Golgi apparatus [5], we examined the effect of ALLN on the distribution of microtubules using anti-␣-tubulin antibody. The distribution of the microtubules was not affected by the treatment with ALLN (Figs. 7A and B). The formation of IF inclusions after the treatment with ALLN and the disappearance of IF inclusions after the treatment and removal of ALLN were analyzed morphometrically by immunofluorescence. After the treatment with ALLN for 24 h, IF inclusions were observed in most cells (Figs. 2A and B). When the cells were incubated with ALLN-free medium for 48 h after the treatment with ALLN for 24 h, the ratio of IF inclusion-containing cells decreased over time (Table 2 , Fig. 2E). Forty-eight hours after 24-h treatment and removal of ALLN, the structure of the Golgi apparatus was restored and Golgi ribbons were observed in most Huh7 cells (Figs. 5M–O). The ultrastructure of the cytoplasmic inclusions induced by ALLN was examined by transmission electron microscopy. In ALLN-treated Huh7 cells, membrane-free, spherical, or irregular-shaped structures composed of electrondense materials surrounded by filamentous structures were observed in the cytoplasm (Fig. 8A). Membranous organelles, such as mitochondria and vesicles, were sometimes observed in these structures. At higher magnification, filamentous structures were clearly recognized with a diameter of about 8 –10 nm (Fig. 8B). In these cells, autophagic vacuoles and electron-dense lysosomes were sometimes observed. When cells were treated with ALLN for 24 h and chased in ALLN-free medium for 24 h, many autophagic vacuoles and electron-dense lysosomes were observed (Fig. 8C).
Discussion Intracellular inclusions composed of abnormal misfolded protein aggregates are a prominent pathological feature of most neurodegenerative disorders, and these protein aggregates are related to the redistribution of neurofilament type IV neuronal IF [6,7,18 –21]. In ordinary circumstances, cells refold misfolded proteins in chaperon-dependent processes or degrade them via UPS. When the capacity of the protea-
65
some is exceeded by the production of misfolded proteins, the formation of intracellular protein aggregates occurs. These structures are designated as aggresomes, and are associated with IFs and ubiquitin [5]. Aggresomes are pericentriolar, membrane-free cytoplasmic inclusions containing misfolded and ubiquitinated protein ensheathed in cages composed of IFs [5]. Many disease-characteristic proteins have been reported to form aggresome cores. This evidence was mainly obtained from experiments in which abnormal aggregation-prone proteins were transfected to cultured cells [5,8,9,22–24]. In the present study, we observed IF inclusions after the inhibition of proteasomes in cultured cells without transfection of any abnormal proteins. These structures were localized at the MTOC and associated with IFs and ubiquitin. The ultrastructure of the inclusion bodies was similar to that of the previously reported aggresomes, because they included electron-dense particles and IFs [5,23]. Although the previously described aggresomes were round structures [5,23], we observed both round-shaped and irregular-shaped inclusion bodies in the present study. The reason for this may be due to differences in the distribution of ␥-tubulin in the cells examined, because the aggresomes and the inclusion bodies in the present study were associated with ␥-tubulin and the distribution of ␥-tubulin was different among cells, as we previously reported [9]. Therefore, proteasome inhibition alone without overexpression of abnormal proteins is an inducing factor in the formation of IF inclusions, similar to aggresomes and MBs. Recently, it was reported that 30% of newly synthesized proteins were defective ribosomal products and were degraded by proteasomes [25]. Therefore, such defective proteins may have become the cores of the IF inclusions observed in the present study. This evidence indicates that the formation of IF inclusions or aggresomes depends on the balance between the production of misfolded proteins and chaperondependent refolding and proteasomal degradation of misfolded proteins. Some evidence supports this hypothesis. The causative gene product of autosomal recessive juvenile parkinsonism, parkin, is an E3 ubiquitin protein ligase. Disruption of parkin’s function in UPS by mutation contributes to the development of autosomal recessive juvenile parkinsonism [26,27]. Furthermore, a frame shift mutant of ubiquitin, ubiquitin⫹1 protein, which inhibits the function of UPS, is associated with Alzheimer disease [28] and MBs in hepatocytes [29,30]. Chronic ethanol administration by continuous intragastric infusion inhibited proteasome activity in an animal experiment and MBs are associated with alcoholic liver disease [31]. Perinuclear aggregates rich in ubiquitin and proteasomal antigens were reported in cells treated with an inhibitor of the chymotrypsin-like activity of proteasomes, although the association of IFs was not described in the study [32]. Therefore, it is suggested that decreased proteasomal activity may be a factor involved in the formation of IF inclusions. The inhibition of UPS activity, for whatever reason, may result in the accumulation of misfolded proteins. Recent studies using cultured cells
66
M. Harada et al. / Experimental Cell Research 288 (2003) 60 – 69
Fig. 7. Confocal laser scanning microscopic images of the microtubules in GFP–CK18-transfected Huh7 stained with anti-␣-tubulin antibody. Cells in D–F were treated with ALLN (D–F) for 24 h. Green: GFP-CK18; Red: ␣-tubulin. Abbreviations: ALLN, acetyl-leucyl-leucyl-norleucinal; CK, cytokeratin; GFP, green fluorescent protein.
Fig. 6. Confocal laser scanning microscopic images of the Golgi apparatus in GFP–CK18-transfected Huh7 stained with anti-␥-adaptin antibody. Cells were treated with proteasome inhibitors, ALLN (D–F), lactacystin (G–I), or MG132 (J–L) for 24 h. Green: GFP-CK18, Red: ␥-adaptin. Abbreviations: ALLN, acetyl-leucyl-leucyl-norleucinal; CK, cytokeratin; GFP, green fluorescent protein.
indicated that the expression of misfolded proteins and formation of aggresomes decreased proteasome activity [33,34]. Therefore, the accumulation of misfolded proteins induces IF inclusions and further decreases the UPS activity. This accelerated mechanism may explain the progressive clinical course of various diseases including neurodegenerative diseases and MB-associated diseases. A recent study reported that inhibitors of the mitochondrial electrontransport chain induced cytoplasmic inclusions associated with vimentin, a component of IF [35]. In this model, impaired energy production may be a possible factor in the
formation of the inclusions, because some of the steps in UPS and chaperon-mediated refolding are dependent on ATP. Therefore, disruption of the balance among production, refolding, and degradation of the misfolded protein, for whatever reason, may induce IF inclusions. The IF inclusions were located at the MTOC, similar to aggresomes [5], but were not colocalized with calnexin, GalT, lamp 1, or cathepsin D. Therefore, it was a distinct structure from the endoplasmic reticulum, Golgi apparatus, late endosome, or lysosome. Fragmentation of the Golgi apparatus was observed in proteasome inhibitor-treated cells and was reported in neurons of patients with ALS [36] and its animal model [37]. A significant correlation between the fragmentation of the Golgi apparatus and ubiquitinpositive inclusions was demonstrated [36]. Our recent study indicated that production of IF inclusions by a mutant CK inhibited protein transport from the ER to the Golgi apparatus and the intact IF network was necessary for the organization of the Golgi apparatus (data not shown). Furthermore, a recent study demonstrated the direct interaction of vimentin, a component of IF, with a Golgi protein, formim-
Table 2 Percentage of cells with IF inclusions after the 24-h incubation and removal of ALLN (%) 0h 24 h 48 h 72 h 96 h 120 h
91.0 ⫾ 2.8 54.1 ⫾ 9.6 24.1 ⫾ 3.0 8.5 ⫾ 2.0 7.7 ⫾ 1.9 7.2 ⫾ 1.0
Note. Cells were treated with ALLN for 24 h, washed with ALLN-free medium, and chased for indicated hours. The values are expressed as means ⫾ SD. Abbreviations: ALLN, acetyl-leucyl-leucyl-norleucinal; IF, intermediate filament.
M. Harada et al. / Experimental Cell Research 288 (2003) 60 – 69
67
Fig. 8. Transmission electron micrographs of Huh7 cells. Cells were treated with ALLN for 24 h. (A) Bar, 2 m; (B) bar, 1 m. Membranous free structures (A, surrounded by arrowheads) composed of electron-dense materials (B, arrows) surrounded by filamentous structures (B, arrowheads) are observed near the nucleus. The diameter of the filaments is about 8 –10 nm. Some cells were treated with ALLN for 24 h and chased in ALLN-free medium for 24 h. (C) Bar, 2 m. Autophagic vacuoles and electron-dense lysosomes (C, arrows) are observed. AV, autophagic vacuole; N, nucleus. Abbreviations: ALLN, acetyl-leucyl-leucyl-norleucinal.
inotransferase cyclodeaminase [38]. Therefore, proteasome inhibition induces IF inclusions accompanied by a loss of the intact IF network, which inhibits the organization of the Golgi apparatus. The change in the structure of the Golgi apparatus is attributed to the change in IF, because a small number of cells with a normal IF network showed a normal Golgi appearance even after treatment with ALLN (Figs. 5J–L) and the microtubules were not affected by ALLN (Fig. 7). Because the Golgi apparatus is a crucial organelle in membrane traffic and sorting of proteins and lipids [39,40], disturbance of this organelle may affect the intracellular transport of various proteins and lipids, and may affect various important cellular functions. It is an important question whether the IF inclusions reversibly disappear or not. In the present study, the IF inclusions disappeared over time after the treatment and removal of ALLN. Yamamoto and Lucus [41] showed that the expression of misfolded proteins resulted in protein
aggregates and progressive motor dysfunction using a mouse model of Huntington disease. Blockage of the expression of the protein in symptomatic mice led to the disappearance of the aggregates and the phenotype. This evidence indicates that the Huntington disease phenotype is dependent on the continuous expression of the abnormal protein and the existence of protein aggregates, which reversibly disappear [41]. Lee et al. also reported that the formation of ␣-synuclein-containing inclusions after mitochondrial inhibition was reversed by restoring the mitochondrial function [35]. Therefore, when the factor inducing IF inclusions was removed, cells could remove the IF inclusions by some mechanism. In such cases, the organization of the Golgi apparatus was also restored. In the electron microscopic examination, many autophagic vacuoles and electron-dense lysosomes were observed after the removal of ALLN. Recently, it was reported that mutant huntingtin expression stimulated endosome-lysosome activity and au-
68
M. Harada et al. / Experimental Cell Research 288 (2003) 60 – 69
tophagy [42]. Furthermore, accumulation of unfolded proteins induces unfolded protein response, which stimulates the expression of genes associated with vacuolar targeting [43]. Therefore, it is likely that cytoplasmic IF inclusions are, at least in part, sequestered into autophagosomes, and degraded in the lysosomes, thus enabling cells to remove the IF inclusions. In the case of genetic disorders, the expression of abnormal proteins continues; therefore, inclusions may not easily disappear. In conclusion, inhibition of proteasome function alone induced IF inclusions and fragmentation of the Golgi apparatus and these changes are restored after the removal of the proteasome inhibitor. Disturbance of the balance between the production and degradation of unfolded proteins induces the formation of IF inclusions, inhibits the organization of the Golgi apparatus, and may affect the function of this important organelle.
[11] [12]
[13]
[14]
[15]
[16]
Acknowledgments [17]
The research was supported in part by grants-in-aid (12670535, 14570528) from the Ministry of Education of Japan and a grant from the Foundation for Advancement of International Science (FAIS) (to M.H.). The authors thank Dr. F. Tokunaga for providing pcDNA3.1 flag-ubiquitin and Ms. R. Inayoshi for expert technical assistance.
[18]
References
[20]
[1] K. Tanaka, H. Chiba, The proteasome: a protein-destroying machine, Genes Cells 3 (1998) 499 –510. [2] A. Ciechanover, The ubiquitin-proteasome pathway: on protein death and cell life, EMBO J. 17 (1998) 7151–7160. [3] R.W. Carrell, D.A. Lomas, Conformational disease, Lancet 350 (1997) 134 –138. [4] R.R. Kopito, Aggresomes, inclusion bodies and protein aggregation, Trends Cell Biol. 10 (2000) 524 –530. [5] J.A. Johnston, C.L. Ward, R.R. Kopito, Aggresomes: a cellular response to misfolded proteins, J. Cell Biol. 143 (1998) 1883–1898. [6] S.W. Davies, M. Turmaine, B.A. Cozens, M. DiFiglia, A.H. Sharp, C.A. Ross, E. Scherzinger, E.E. Wanker, L. Mangiarini, C.P. Bates, Formation of neuronal intranuclear inclusions underlies the neurological dysfunction in mice transgenic for the HD mutation, Cell 90 (1997) 537–548. [7] M. DiFiglia, E. Sapp, K.O. Chase, S.W. Davies, G.P. Bates, J.P. Vonsattel, N. Aronin, Aggregation of huntingtin in neuronal intranuclear inclusions and dystrophic neuritis in brain, Science 277 (1997) 1990 –1993. [8] V. Keitel, J. Kartenbeck, A.T. Nies, H. Spring, M. Brom, D. Keppler, Impaired protein maturation of the conjugate export pump multidrug resistance protein 2 as a consequence of a deletion mutation in Dubin–Johnson syndrome, Hepatology 32 (2000) 1317–1328. [9] M. Harada, S. Sakisaka, K. Terada, R. Kimura, T. Kawaguchi, H. Koga, M. Kim, E. Taniguchi, S. Hanada, T. Suganuma, K. Furuta, T. Sugiyama, M. Sata, A mutation of Wilson disease protein, ATP7B, is degraded in the proteasomes and forms protein aggregates, Gastroenterology 120 (2001) 967–973. [10] J. Lowe, A. Blanchard, K. Morrell, G. Lennox, L. Reynolds, M. Billett, M. Landon, R.J. Mayer, Ubiquitin is a common factor in
[19]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
intermediate filament inclusion bodies of diverse type in man, including those of Parkinson’s disease, Pick’s disease, and Alzheimer’s disease, as well as Rosenthal fibres in cerebellar astrocytomas, cytoplasmic bodies in muscle, and Mallory bodies in alcoholic liver disease, J. Pathol. 155 (1998) 9 –15. H. Denk, C. Stumptner, K. Zatloukal, Mallory bodies revisited, J. Hepatol. 32 (2000) 689 –702. H. Nakabayashi, K. Taketa, K. Miyano, T. Yamane, J. Sato, Growth of human hepatoma cells lines with differentiated functions in chemically defined medium, Cancer Res. 42 (1982) 3858 –3863. J. Kawano, S. Ide, T. Oinuma, T. Suganuma, A protein-specific monoclonal antibody to rat liver 134 galactolsyltransferase and its application to immunohistochemistry, J. Histochem. Cytochem. 42 (1994) 363–369. J.W. Chen, T.L. Murphy, M.C. Willingham, I. Pastan, J.T. August, Identification of two lysosomal membrane glycoproteins, J. Cell Biol. 101 (1985) 85–95. J.B. Dictenberg, W. Zimmerman, C.A. Spark, A. Young, C. Vidair, Y. Zheng, W. Carrington, F.S. Fay, S.J. Doxsey, Percentrin and ␥-tubulin form a protein complex and are organized into a novel lattice at the centrosome, J. Cell Biol. 141 (1998) 163–174. P.J.I. Salas, Insoluble ␥-tubulin-containing structures are anchored to the apical network of intermediate filaments in polarized Caco-2 epithelial cells, J. Cell Biol. 146 (1999) 645– 657. B. Storrie, J. White, S. Ro¨ttger, E.H.K. Stelzer, T. Suganuma, T. Nilsson, Recycling of Golgi-resident glycosyltransferases through the ER reveals a novel pathway and provides an explanation for nocodazole-induced Golgi scattering, J. Cell Biol. 143 (1998) 1505–1521. H.L. Paulson, Protein fate in neurodegenerative proteinopathies: polyglutamine disease join the (mis) fold, Am. J. Hum. Genet. 64 (1999) 339 –345. R.J. Mayer, J. Lowe, G. Lennox, F. Doherty, M. Landon, Intermediate filaments and ubiquitin: a new thread in the understanding of chronic neurodegenerative diseases, Prog. Clin. Biol. Res. 317 (1989) 809 – 818. M.L. Schmidt, V.M.Y. Lee, J.Q. Trojanowski, Analysis of epitopes shared by Hirano bodies and neurofilament proteins in normal and Alzheimer’s disease hippocampus, Lab. Invest. 60 (1989) 513–522. T. Ishihara, M. Hong, B. Zhang, Y. Nakagawa, M.K. Lee, J.Q. Trojanowski, V.M.Y. Lee, Age-dependent emergence and progression of a tauopathy in transgenic mice overexpressing the shortest human tau isoform, Neuron 24 (1999) 751–762. J.A. Johnston, M.J. Dalton, M.E. Gurney, R.R. Kopito, Formation of high molecular weight complexes of mutant Cu, Zn-superoxide dismutase in a mouse model for familial amyotrophic lateral sclerosis, Proc. Natl. Acad. Sci. USA 97 (2000) 12571–12576. R. Garcia-Mata, Z. Bebok, E.J. Sorscher, E.S. Sztul, Characterization and dynamics of aggresome formation by a cytosolic GFP-chimera, J. Cell Biol. 146 (1999) 1239 –1254. S. Waelter, A. Boeddrich, R. Lurz, E. Scherzinger, G. Lueder, H. Lehrach, E.E. Wanker, Accumulation of mutant huntingtin fragments in aggresome-like inclusion bodies as a result of insufficient protein degradation, Mol. Biol. Cell 12 (2001) 1393–1407. U. Schubert, L.C. Anton, J. Gibbs, C.C. Norbury, J.W. Yewdell, J.R. Bennink, Rapid degradation of a large fraction of newly synthesized proteins by proteasomes, Nature 404 (2000) 770 –774. Y. Imai, M. Soda, H. Inoue, N. Hattori, Y. Mizuno, R. Takahashi, An unfolded putative transmembrane polypeptide, which can lead to endoplasmic reticulum stress, is a substrate of Parkin, Cell 105 (2001) 891–902. H. Shimura, N. Hattori, S. Kubo, Y. Mizuno, S. Asakawa, S. Minoshima, N. Shimizu, K. Iwai, T. Chiba, K. Tanaka, T. Suzuki, Familial Parkinson disease gene product, parkin, is a ubiquitin-protein ligase, Nature Genet. 25 (2000) 302–305. F.W. van Leeuwen, D.P.V. de Kleijn, H.H. van den Hurk, A. Neubauer, M.A.F. Sonnemans, J.A. Sluijs, S. Ko¨ycu¨, R.D.J. Ramdjielal, A. Salchi, G.J.M. Marten, F.G. Grosveld, J. Peter, H. Burbach,
M. Harada et al. / Experimental Cell Research 288 (2003) 60 – 69
[29]
[30]
[31]
[32]
[33] [34]
[35]
E.M. Hol, Frameshift mutants of  amyloid precursor protein and ubiquitin-B in Alzheimer’s disease and Down patients, Science 279 (1998) 242–247. N.E. Riley, J. Li, S. Worrall, J.A. Rothnagel, C. Swagell, F.W. van Leeuwen, S.W. French, The Mallory body as an aggresome: in vitro studies, Exp. Mol. Pathol. 72 (2002) 17–23. L.W. Mcphaul, J. Wang, EM. Hol, M.A.F. Sonnemans, N. Riley, V. Nguyen, Q.X. Yuan, Y.H. Lue, F.W.V. Leeuwen, S.W. French, Molecular misreading of the ubiquitin B gene and hepatic Mallory body formation, Gastroenterology 122 (2002) 1878 –1885. V. Fataccioli, E. Andraud, M. Gentil, S.W. French, H. Rouach, Effect of chronic ethanol administration on rat liver proteasome activities: relationship with oxidant stress, Hepatology 29 (1999) 14 –20. C. Wojcik, D. Schroeter, S. Wilk, J. Lamprecht, N. Paweletz, Ubiquitin-mediated proteolysis centers in HeLa cells: indication from studies of an inhibitor of chymotrypsin-like activity of the proteasome, Eur. J. Cell Biol. 71 (1996) 311–318. N.F. Bence, R.M. Sampat, R.R. Kopito, Impairment of the ubiquitin-proteasome system by protein aggregation, Science 292 (2001) 1552–1555. N.R. Jana, E.A. Zemskov, G. Wang, N. Nukina, Altered proteasomal function due to expression of polyglutamine-expanded truncated Nterminal huntingtin induces apoptosis by caspase activation through mitochondrial cytochrome c release, Hum. Mol. Genet. 10 (2001) 1049 –1059. J.H. Lee, S.Y. Shin, C. Choi, Y.H. Lee, S.J. Lee, Formation and removal of ␣-synuclein aggregates in cells exposed to mitochondrial inhibitors, J. Biol. Chem. 277 (2002) 5411–5417.
69
[36] Z. Mourelatos, A. Hirano, C.R. Alan, K.G. Nicholas, Fragmentation of Golgi apparatus of motor neurons in amyotrophic lateral sclerosis (ALS), Am. J. Pathol. 144 (1994) 1288 –1300. [37] Z. Mourelatos, N.K. Gonatas, A. Stieber, M.E. Gurney, M.C.D. Canto, The Golgi apparatus of spinal cord motor neurons in transgenic mice expressing mutant Cu, Zn superoxide dismutase becomes fragmented in early, preclinical stages of the disease, Proc. Natl. Acad. Sci. USA 93 (1996) 5472–5477. [38] Y.S. Gao, E. Sztul, A novel interaction of the Golgi complex with the vimentin intermediate filament cytoskeleton, J. Cell Biol. 152 (2001) 877– 893. [39] P Kellers, K. Simons, Post-Golgi biosynthetic trafficking, J. Cell Sci. 110 (1997) 3001–3009. [40] W.M. Rohn, Y. Rouille, S. Waguri, B. Hoflack, Bi-directional trafficking between the trans-Golgi network and the endosome/lysosomal system, J. Cell Sci. 113 (2000) 2093–2101. [41] A. Yamamoto, J. Lucus, Reversal of neuropathology and motor dysfunction in a conditional model of Huntington’s disease, Cell 101 (2000) 57– 66. [42] K.B. Kegel, M. Kim, E. Sapp, C. McIntyre, J.G. Castano, N. Aronin, M. DiFiglia, Huntingtin expression stimulates endosomal-lysosomal activity, endosome tubulation, and autophagy, J. Neurosci. 20 (2000) 7268 –7278. [43] K.J. Travers, C.K. Patil, L. Wodicka, D.J. Lockhart, J.S. Weissman, P. Walter, Functional and genomic analysis reveal an essential coordination between the unfolded protein response and ER-associated degradation, Cell 101 (2000) 249 –258.