Proteasome-mediated cleavage of the Y-box-binding protein 1 is linked to DNA-damage stress response
2005; Springer Nature; Volume: 24; Issue: 20 Linguagem: Inglês
10.1038/sj.emboj.7600830
ISSN1460-2075
AutoresAlexey V. Sorokin, Anastasia A Selyutina, Maxim A. Skabkin, Sergey Guryanov, I. V. Nazimov, Christina Richard, John P.H. Th'ng, J. Yau, Poul H. Sorensen, Lev P. Ovchinnikov, Valentina Evdokimova,
Tópico(s)DNA Repair Mechanisms
ResumoArticle29 September 2005free access Proteasome-mediated cleavage of the Y-box-binding protein 1 is linked to DNA-damage stress response Alexey V Sorokin Alexey V Sorokin Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Search for more papers by this author Anastasia A Selyutina Anastasia A Selyutina Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Search for more papers by this author Maxim A Skabkin Maxim A Skabkin Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Search for more papers by this author Sergey G Guryanov Sergey G Guryanov Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Search for more papers by this author Igor V Nazimov Igor V Nazimov Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, Moscow, Russian Federation Search for more papers by this author Christina Richard Christina Richard Thunder Bay Regional Health Sciences Centre, Medical Sciences Division, Northern Ontario School of Medicine, Thunder Bay, Ontario, Canada Search for more papers by this author John Th'ng John Th'ng Thunder Bay Regional Health Sciences Centre, Medical Sciences Division, Northern Ontario School of Medicine, Thunder Bay, Ontario, Canada Search for more papers by this author Jonathan Yau Jonathan Yau Thunder Bay Regional Health Sciences Centre, Medical Sciences Division, Northern Ontario School of Medicine, Thunder Bay, Ontario, Canada Search for more papers by this author Poul HB Sorensen Poul HB Sorensen Department of Pathology, British Columbia Research Institute for Children's and Women's Health, and the University of British Columbia, Vancouver, British Columbia, Canada Department of Pediatrics, British Columbia Research Institute for Children's and Women's Health, and the University of British Columbia, Vancouver, British Columbia, Canada Search for more papers by this author Lev P Ovchinnikov Lev P Ovchinnikov Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Search for more papers by this author Valentina Evdokimova Corresponding Author Valentina Evdokimova Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Department of Pediatrics, British Columbia Research Institute for Children's and Women's Health, and the University of British Columbia, Vancouver, British Columbia, Canada Search for more papers by this author Alexey V Sorokin Alexey V Sorokin Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Search for more papers by this author Anastasia A Selyutina Anastasia A Selyutina Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Search for more papers by this author Maxim A Skabkin Maxim A Skabkin Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Search for more papers by this author Sergey G Guryanov Sergey G Guryanov Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Search for more papers by this author Igor V Nazimov Igor V Nazimov Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, Moscow, Russian Federation Search for more papers by this author Christina Richard Christina Richard Thunder Bay Regional Health Sciences Centre, Medical Sciences Division, Northern Ontario School of Medicine, Thunder Bay, Ontario, Canada Search for more papers by this author John Th'ng John Th'ng Thunder Bay Regional Health Sciences Centre, Medical Sciences Division, Northern Ontario School of Medicine, Thunder Bay, Ontario, Canada Search for more papers by this author Jonathan Yau Jonathan Yau Thunder Bay Regional Health Sciences Centre, Medical Sciences Division, Northern Ontario School of Medicine, Thunder Bay, Ontario, Canada Search for more papers by this author Poul HB Sorensen Poul HB Sorensen Department of Pathology, British Columbia Research Institute for Children's and Women's Health, and the University of British Columbia, Vancouver, British Columbia, Canada Department of Pediatrics, British Columbia Research Institute for Children's and Women's Health, and the University of British Columbia, Vancouver, British Columbia, Canada Search for more papers by this author Lev P Ovchinnikov Lev P Ovchinnikov Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Search for more papers by this author Valentina Evdokimova Corresponding Author Valentina Evdokimova Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation Department of Pediatrics, British Columbia Research Institute for Children's and Women's Health, and the University of British Columbia, Vancouver, British Columbia, Canada Search for more papers by this author Author Information Alexey V Sorokin1, Anastasia A Selyutina1, Maxim A Skabkin1, Sergey G Guryanov1, Igor V Nazimov2, Christina Richard3, John Th'ng3, Jonathan Yau3, Poul HB Sorensen4,5, Lev P Ovchinnikov1 and Valentina Evdokimova 1,5 1Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation 2Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, Moscow, Russian Federation 3Thunder Bay Regional Health Sciences Centre, Medical Sciences Division, Northern Ontario School of Medicine, Thunder Bay, Ontario, Canada 4Department of Pathology, British Columbia Research Institute for Children's and Women's Health, and the University of British Columbia, Vancouver, British Columbia, Canada 5Department of Pediatrics, British Columbia Research Institute for Children's and Women's Health, and the University of British Columbia, Vancouver, British Columbia, Canada *Corresponding author. Department of Pediatrics, University of British Columbia, 3064-950 West 28th Avenue, Vancouver, British Columbia, Canada BC V5Z 4H4. Tel.: +1 604 822 2211; Fax: +1 604 875 3417; E-mail: [email protected] The EMBO Journal (2005)24:3602-3612https://doi.org/10.1038/sj.emboj.7600830 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info YB-1 is a DNA/RNA-binding nucleocytoplasmic shuttling protein whose regulatory effect on many DNA- and RNA-dependent events is determined by its localization in the cell. Distribution of YB-1 between the nucleus and the cytoplasm is known to be dependent on nuclear targeting and cytoplasmic retention signals located within the C-terminal portion of YB-1. Here, we report that YB-1 undergoes a specific proteolytic cleavage by the 20S proteasome, which splits off the C-terminal 105-amino-acid-long YB-1 fragment containing a cytoplasmic retention signal. Cleavage of YB-1 by the 20S proteasome in vitro appears to be ubiquitin- and ATP-independent, and is abolished by the association of YB-1 with messenger RNA. We also found that genotoxic stress triggers a proteasome-mediated cleavage of YB-1 in vivo and leads to accumulation of the truncated protein in nuclei of stressed cells. Endoproteolytic activity of the proteasome may therefore play an important role in regulating YB-1 functioning, especially under certain stress conditions. Introduction Y-box protein 1, also known as dbpB, has been initially cloned as a transcription factor that specifically recognizes the Y-box promoter element in a variety of different genes and designated YB-1, accordingly (Didier et al, 1988). Independently, we have cloned and characterized this 50 kDa protein (p50) as a major protein bound to messenger RNAs (mRNAs) in mammalian cells (Minich and Ovchinnikov, 1992; Evdokimova et al, 1995). At present, it is clear that YB-1/p50 is a nucleocytoplasmic shuttling protein involved in many DNA- and RNA-dependent events. When in the nucleus, this protein regulates transcription of many genes including those involved in cell proliferation and differentiation. It has also been implicated in repair, replication, recombination of DNA and alternative mRNA splicing (reviewed in Matsumoto and Wolffe, 1998; Kohno et al, 2003). When in the cytoplasm, it plays a key role in packaging of mRNAs into messenger ribonucleoprotein particles (mRNPs), and also regulates translational activity and stability of mRNA (Evdokimova et al, 2001; Skabkin et al, 2004). Distribution of YB-1 between nuclear and cytoplasmic compartments must be therefore stringently regulated. In accordance with this idea, it has been recently shown that although YB-1 is localized predominantly in the cytoplasm throughout the cell cycle, it moves to the nucleus at the G1 to S phase transition in a pattern similar to cyclin E (Jurchott et al, 2003). Nuclear accumulation of YB-1 is accompanied by transcriptional activation of cyclin A and B1 genes, which possess the YB-1 recognition element in their promoters. Translocation of YB-1 to the nucleus has also been shown to be activated by various insults including adenovirus infection, UV irradiation, hyperthermia or association with certain proteins such as tumor suppressor p53 or splicing factor SRp30c (reviewed in Kohno et al, 2003). Predominant nuclear localization of YB-1 is thought to be a characteristic of transformed cells, and was also associated with development of multiple drug resistance of cancer cells (Kuwano et al, 2004). Experiments on a set of deletion mutants showed that the C-terminus of YB-1 contains one or two potential noncanonical nuclear localization signals (NLS) that are required for nuclear import of YB-1 (Stenina et al, 2001; Jurchott et al, 2003; Bader and Vogt, 2005). In addition, the C-terminal part of YB-1 possesses a cytoplasmic retention signal (CRS; see Figure 6F) that prevails over the NLS and ensures predominant cytoplasmic localization of YB-1 (Jurchott et al, 2003; Bader and Vogt, 2005). Recent data have suggested a novel mechanism for nuclear targeting of YB-1 that involves cleavage of its CRS-containing C-terminal part in response to thrombin stimulation of endothelial cells (Stenina et al, 2001). Truncated YB-1 exhibited nuclear localization and preserved its activity as a transcription factor facilitating transcription of the platelet-derived growth factor (PDGF) gene (Stenina et al, 2000), although the mechanism of proteolytic processing of YB-1 remained unclear. Figure 1.Identification of the YB-1-specific protease as 20S proteasome. (A) Purification of protease that cleaves YB-1. (B) Selective cleavage of YB-1 by purified protease. YB-1, poly(A)-binding protein (PABP), glycogen phosphorylase (GP), glutathione S-transferase (GST), casein, tubulin or bovine serum albumin (BSA) were incubated with purified protease for 1 h and analyzed by SDS–15% PAGE and Coomassie staining. (C) Purified protease preparation was analyzed by Coomassie staining or Western using antibodies against α5 or α6 20S proteasomal subunits. (D) List of polypeptides identified in the purified protease complex by mass spectrometry. BN, band numbers as in (C); Protein name, NCBI annotation; Score, generated by MASCOT tools; Proteasome subunit name is given in systemic nomenclature (Groll et al, 1997). (E) Sedimentation distribution of the purified protease. Purified protease (∼30 μg) was fractionated by centrifugation at 45 000 r.p.m. (SW 60 rotor; Beckman Instruments) for 4 h through a linear 5–20% (w/v) sucrose gradient. Fractions (275 μl) were collected from the bottom and analyzed for the ability to cleave YB-1. Coomassie staining of YB-1 after incubation with the corresponding fractions is shown. 28S, 18S rRNAs and 9S α-globin RNA served as sedimentation markers. (F) Effect of proteasomal and protease inhibitors on YB-1 cleavage. YB-1 was incubated with purified protease preparation in the presence of the indicated inhibitors. After 60 min incubation, YB-1 and degradation products were resolved by SDS–15% PAGE and visualized by Coomassie staining. Download figure Download PowerPoint Here, we report that cleavage of a short C-terminal YB-1 fragment containing the CRS is mediated by the 20S proteasome. Of physiological relevance, we found that association with mRNA prevents YB-1 cleavage and that only unbound, mRNA-free YB-1 may be targeted. Interestingly, ubiquitin and ATP were not required for proteolytic processing of YB-1 in vitro, albeit polyubiquitylated YB-1 species were observed in cells treated with proteasomal inhibitors. Limited proteolysis of YB-1 was triggered by DNA-damaging drugs and resulted in accumulation of the truncated YB-1 form in nuclear punctuate structures resembling nuclear speckles, zones of accumulation of transcriptional and splicing factors. We propose that specific cleavage of YB-1 mediated by the 20S proteasome may generate a polypeptide with a potentially altered biological function. Results Identification of the YB-1-specific protease as 20S proteasome YB-1 purified from various cell extracts including rabbit reticulocyte lysate (RRL) often contains an additional 32 kDa band immunoreactive with anti-YB-1 antibodies (data not shown). This observation together with literature reports indicating that YB-1/dbpB can be specifically cleaved in endothelial cells (Stenina et al, 2000) prompted us to search for proteins that may be involved in YB-1 cleavage. To identify these proteins, we performed a multistep purification procedure using RRL (Figure 1A). The resulting protease preparation selectively cleaved YB-1 to give two fragments with relative molecular masses of about 32 and 22 kDa, while having no effect on other proteins tested (Figure 1B). Figure 2.YB-1 cleavage is mediated by the 20S proteasome in a ubiquitin- and ATP-independent manner. (A) Comparison of proteolytic activities exhibited by various 20S and 26S proteasomal preparations toward YB-1. YB-1 (1.5 μg) was incubated with purified 20S proteasome (0.5 μg), commercial 20S proteasome (0.5 μg), purified 26S proteasome (1.5 μg) or commercial 26S proteasome (1 μg). Incubations were performed for 1 h in a total volume of 20 μl of buffer G. Coomassie staining of the gel is shown. (B) Proteolytic activities of proteasomes used in (A) were analyzed using fluorogenic peptide Suc-LLVY-AMC (100 μM). Incubations were carried out for 1 h in a total volume of 200 μl of buffer G. Fluorescence of released AMC was measured at 360 nm excitation and 430 nm emission. AMC extinction intensity upon cleavage of Suc-LLVY-AMC by purified 20S proteasome was taken as 100%. (C) Time course of protein cleavage by the 26S proteasome. For each time point, YB-1 (1 μg) or casein (1 μg) was incubated with commercial 26S proteasome (1 μg) in a total volume of 20 μl of buffer G. Degradation products were analyzed by SDS–15% PAGE and Coomassie staining. The bottom numbers indicate percentage of the remaining protein as measured from two independent experiments by densitometry, with values obtained for time 0 set as 100%. (D) Effect of ATP on YB-1 cleavage. YB-1 was incubated with 20S proteasome in buffer G in the presence or absence of ATP, an ATP-depleting system (10 mM glucose, 1 μg/ml of hexokinase (HXK)) or an ATP-regenerating system (10 mM creatine phosphate (CP), 10 μg/ml of creatine kinase (CK)), as indicated. Coomassie staining of the gel is shown. (E) Effect of the mRNA on YB-1 cleavage. YB-1 (1 μg; 28 pmol) was incubated with α-globin mRNA (1 μg; 5 pmol) at 30°C for 15 min in a total volume of 20 μl of buffer G. YB-1–mRNA complexes were then either directly incubated with 20S proteasome for 60 min or pretreated with RNase A (0.15 μg/μl) and micrococcal nuclease (0.05 U/μl) at 37°C for 30 min. Coomassie staining of the gel is shown. (F) Effect of the mRNA on proteolytic activity of 20S proteasome. 20S proteasome (0.5 μg) was incubated with Suc-LLVY-AMC (100 μM) in the presence or absence of α-globin mRNA (7.5 μg). AMC extinction intensity upon cleavage of Suc-LLVY-AMC without mRNA was taken as 100%. Download figure Download PowerPoint YB-1-specific protease was eluted from Superose 12 in the region corresponding to a molecular weight of about 700 kDa and remained intact even at 2 M NaCl, suggesting that it represents a highly stable multiprotein complex. Electrophoretic analysis of the purified protease preparation revealed that it consists of eight polypeptides within the size range of 22–32 kDa, plus one of ∼90 kDa (Figure 1C, lane 1). As a similar subunit composition is characteristic of the 20S proteasome, we employed antibodies directed against human proteasomal subunits to test if these might be present in our protease preparation. Indeed, two of them were identified as α5 and α6 subunits of the 20S proteasome (Figure 1C, lanes 2 and 3). Furthermore, eight other major polypeptides were revealed by mass spectrometry as subunits of the 20S proteasome (Figure 1D). The 90 kDa polypeptide was identified as a heat-shock protein Hsp90α (Hsp86), whose copurification with the 20S proteasome was previously reported (Tsubuki et al, 1994). The sedimentation coefficient of the purified protease complex was approximately 20S (Figure 1E), which is also characteristic of the 20S proteasome. The cleavage of YB-1 was suppressed by specific proteasome inhibitors including MG132, YU102, epoxomicin, lactacystin and bactenecin-5 but not by common protease inhibitor phenylmethanesulfonyl fluoride (PMSF) (Figure 1F), further indicating involvement of the 20S proteasome. Of further note, YU102, which is known to inhibit post-acidic activity of the 20S proteasome, markedly reduced YB-1 cleavage when added at low concentrations (0.2–5 μM). Similarly, MG132, which inhibits both the post-acidic and chymotrypsin-like proteasomal activities, prevented YB-1 cleavage at a concentration of 10 μM. In contrast, inhibitors preferably suppressing chymotrypsin-like activity of the proteasome (epoxomicin, lactacystin and bactenecin-5) were significantly less efficient, suggesting that the post-acidic activity of the 20S proteasome is mainly required for YB-1 cleavage. Taken together, these results indicate that the 20S proteasome mediates selective cleavage of YB-1. YB-1 cleavage is mediated by the 20S proteasome in a ubiquitin- and ATP-independent manner To date, the majority of studies have been focused on the role of proteasome in the ubiquitin-dependent degradation pathway. This ATP-dependent pathway implies conjugation of multiple ubiquitin moieties to target proteins, which are then degraded by the 26S proteasomal complex (Hershko and Ciechanover, 1998). However, recent data indicate that the 26S proteasome and its catalytic core, the 20S proteasome, can also degrade some proteins in a ubiquitin- and ATP-independent manner (Orlowski and Wilk, 2003; Hoyt and Coffino, 2004). To address whether YB-1 might be cleaved by the 26S proteasome as well, we utilized a commercial human 26S proteasome or a preparation purified from RRL in parallel with preparations of the 20S proteasome. All proteasome preparations showed comparable activity in degrading a model fluorogenic substrate Suc-LLVY-AMC (Figure 2B); however, neither of the two preparations of the 26S proteasome stimulated YB-1 cleavage (Figure 2A). In contrast to casein, a known substrate of the 26S proteasome (Kisselev et al, 1999), YB-1 remained stable even after a prolonged 6 h incubation (Figure 2C). These results suggest that although a role for the 26S proteasome in vivo cannot be ruled out, cleavage of YB-1 in vitro is mediated exclusively by the 20S proteasome. Figure 3.Identification of the cleavage site on YB-1. (A) Time course of YB-1 cleavage by the 20S proteasome. For each time point, YB-1 (1 μg) was incubated with purified 20S proteasome (0.5 μg) in a total volume of 20 μl of buffer G. Coomassie staining of the gel is shown. (B) Schematic representation of YB-1 and its truncated fragments. The N-terminal AP, CSD and CTD are indicated. (C) Cleavage of YB-1 and its derivatives by the 20S proteasome. Full-length YB-1 and its fragments (1–1.5 μg) were incubated with purified 20S proteasome (0.5 μg) for 60 min. Peptides generated from full-length YB-1 are designated as peptides I and II. (D) Alignment of the full-length YB-1 sequence with those of peptides I and II, which were determined by N-terminal amino-acid sequencing. Download figure Download PowerPoint To analyze whether cleavage of YB-1 by the 20S proteasome is ATP-dependent, cleavage reactions were carried out in the presence or absence of ATP. The ATP-containing reaction mixture was supplemented with an ATP-regenerating system consisting of creatine kinase and creatine phosphate. The reaction mixture lacking ATP contained hexokinase and glucose to eliminate any trace of ATP. As seen in Figure 2D, YB-1 was cleaved by the 20S proteasome independently of the presence or absence of ATP. It is also highly unlikely that YB-1 cleavage is ubiquitin-dependent, since both YB-1 and proteasome preparations used in this study showed electrophoretic homogeneity; no ubiquitin contamination was detected. Also, utilization of recombinant YB-1 purified from Escherichia coli excludes a possibility that YB-1 was originally ubiquitylated. These data indicate that YB-1 is cleaved by the 20S proteasome in vitro in an ATP- and ubiquitin-independent manner. As the vast majority of cytosolic YB-1 is found exclusively in complexes with mRNAs (Minich and Ovchinnikov, 1992; Davydova et al, 1997), we next tested whether or not YB-1 binding to the mRNA affects its susceptibility to proteasomal cleavage. As seen in Figure 2E, preincubation of YB-1 with α-globin RNA abolished its cleavage by the 20S proteasome (lane 4), whereas treatment of YB-1–mRNA complexes with RNases fully restored sensitivity of YB-1 (lane 5). Inhibition of YB-1 cleavage in the presence of RNA was not due to the block of enzymatic activity of the 20S proteasome by RNA (Figure 2F). Therefore, binding to the mRNA protects YB-1, and only unbound, mRNA-free YB-1 appears to be a target for proteasomal cleavage. 20S proteasome cleaves YB-1 before Gly-220 Cleavage of YB-1 by the 20S proteasome was rapid, with a half-life of about 20 min (Figure 3A). Importantly, no intermediate products between full-length YB-1 and the 32 kDa YB-1 fragment were detected, suggesting that the 20S proteasome may endoproteolytically cleave YB-1 rather than processively degrade it from its terminal ends. However, prolonged incubation with proteasome caused additional cleavage of p32, as some minor degradation products became visible after 90 min incubation (Figure 3A, lane 9). To define the cleavage site on YB-1, we utilized various fragments that represent separate domains of YB-1 (Figure 3B). YB-1 is known to be composed of three domains including a short N-terminal Ala/Pro-rich domain (AP), an evolutionarily conserved cold-shock domain (CSD) in its central part and a C-terminal domain (CTD) with alternating negatively and positively charged amino-acid clusters (Wolffe, 1994). As seen in Figure 3C, the 20S proteasome specifically cleaved CTD to generate two subfragments of similar size and had no effect on the integrity of the N-terminal AP-CSD or YB-1 (1–204) fragments. Of further note, the larger fragment of the CTD (fragment II) showed the same mobility as the smaller fragment produced by cleavage of full-length YB-1 (Figure 3C, compare lanes 2 and 4). Because only two major fragments were generated by cleavage of both full-length YB-1 and the CTD, it is likely that the 20S proteasome cleaves YB-1 at a specific internal site that is located approximately in the middle of the YB-1 CTD. Figure 4.DNA-damaging drugs induce YB-1 cleavage in vivo in a proteasome-dependent manner. (A) K-Ras-NIH3T3 cells stably expressing HA-YB-1 were treated for 14 h with doxorubicin (0.6 μg/ml), 17-AAG (1.7 μM), nocodazole (0.2 μg/ml) or dexamethasone (10 μg/ml). Whole-cell extracts were analyzed by Western using anti-HA or anti-YB-1 antibodies. Unrelated RNA-binding protein TIAR served as a loading control. (B) K-Ras-NIH3T3 cells expressing HA-YB-1 were treated for 14 h with doxorubicin (0.6 μg/ml), cisplatin (10 μM), etoposide (10 μM), camptothecin (3 μg/ml), tunicamycin (5 μg/ml), thapsigargin (5 μg/ml) or brefeldin A (2 μg/ml) and analyzed as in (A). (C) K-Ras-NIH3T3 cells treated as above were analyzed by flow cytometry. Both detached and adherent cells were collected for the analysis. (D) K-Ras-NIH3T3 cells were treated for 24 h with 0.6 μg/ml of doxorubicin in the presence of ALLN (30 μg/ml), MG132 (20 μM), Inhibitor I (50 μM), PMSF (1 mM), leupeptin (2 μg/ml), pepstatin (2 μg/ml) or aprotinin (2 μg/ml). Whole-cell extracts were analyzed by Western using anti-YB-1 antibodies. (E) K-Ras-NIH3T3 cells expressing HA-YB-1 were treated with doxorubicin and MG132 (20 μM) for 18 h, and corresponding cytosolic cell extracts were subjected to immunoprecipitation using rabbit anti-ubiquitin or preimmune antibodies. Immunoprecipitated proteins were then analyzed by Western using mouse anti-HA antibodies. Smear in the immunoprecipitates from cells treated with both doxorubicin and MG132 supposedly corresponds to polyubiquitylated HA-YB-1. Download figure Download PowerPoint To localize precisely the cleavage site, peptides I and II obtained by cleavage of full-length YB-1 were sequenced. Alignment of N-terminal amino-acid sequences of these peptides with the sequence of full-length YB-1 revealed that the larger YB-1 fragment is derived from its N-terminus and starts from Ser-2 (Figure 3D). Lack of the first Met can be explained by its co- or post-translational removal. The other peptide begins with Gly-220, suggesting that cleavage may occur immediately after glutamic acid (Glu-219). This is also consistent with the notion that a post-acidic activity of the 20S proteasome was responsible for YB-1 cleavage (Figure 1F). These results demonstrate that the 20S proteasome cleaves YB-1 before Gly-220, removing a 105-amino-acid fragment from its C-terminus. DNA-damaging drugs facilitate YB-1 cleavage in vivo in a proteasome-dependent manner We next asked whether YB-1 cleavage may occur in the cell. Because YB-1 has been shown to regulate transcription of various stress-responsive genes, we considered the possibility that cleavage of YB-1 may be triggered by stress and required for nuclear translocation of YB-1. In the experiments described below, we utilized NIH3T3 cell lines, nontransformed or transformed with K-Ras. These cell lines were engineered to ectopically express N-terminal hemagglutinin (HA)-tagged YB-1, which allowed detection of both the full-length and truncated N-terminal YB-1 forms using highly sensitive and monospecific anti-HA antibodies. Screening of a broad spectrum of therapeutic agents revealed a substantial reduction of full-length YB-1 and an appearance of the 32 kDa fragment in K-Ras-NIH3T3 cells treated with doxorubicin, a DNA-damaging drug that inhibits topoisomerase II activity (Figure 4A, lane 2). Noteworthy, the p32 polypeptide was detected by both anti-YB-1 and anti-HA antibodies, indicating that it was generated from the N-terminal part of YB-1. Neither 17-AAG, which interferes with the action of Hsp90, nor nocodazole, which disrupts microtubules, nor glucocorticoid dexamethasone, all displaying potent anticancer activity and causing various stress responses, affected YB-1 integrity (Figure 4A). We next tested whether cleavage of full-length YB-1 is specific to doxorubicin or other drugs of this class may exert a similar effect. Indeed, in cells treated with compounds causing DNA crosslinking (cisplatin) or inhibiting topoisomerase II (etoposide) or topoisomerase I (camptothecin) activities, full-length YB-1 was substantially cleaved to produce the 32 kDa truncated fragment (Figure 4B, lanes 3–5). In contrast, treatment with other drugs including those activating endoplasmic reticulum stress response (tunicamycin, thapsigargin and brefeldin A) was not sufficient to stimulate YB-1 cleavage (Figure 4B, lanes 6–8). The cleavage of YB-1 was not caused by apoptosis, as no significant elevation in a proportion of dead cells was observed by flow cytometry under these experimental conditions (Figure 4C). Also, despite a significant accumulation of mitotic cells in response to doxorubicin or cisplatin, YB-1 cleavage is unlikely to be a result of a cell cycle arrest, since nocodazole had no effect on YB-1 integrity, while causing accumulation of cells at G2/M (Figure 4C). It should be noted, however, that increased concentrations or prolonged incubation with the above drugs caused complete degradation of full-length YB-1 without generation of a truncated product (data not shown), suggesting that under conditions of a severe stress YB-1 may be degraded by proteasomes or due to activation of apoptotic programs. Figure 5.Truncated YB-1 accumulates in the nucleus following DNA-damaging stress. (A) NIH3T3 cells stably expressing HA-YB-1 were treated with doxorubicin (0.6 μg/ml), etoposide (10 μM) or cisplatin (10 μM) for 14 h. Western of the corresponding cytosolic (C) and nuclear (N) fractions is shown. (B) NIH3T3-HA-YB-1 cells were treated with doxorubicin for the time indicated and analyzed by immunofluorescence microscopy using mouse anti-HA and rabbit anti-PABP antibodies, followed by secondary Alexa Fluor 488 anti-mouse and Alexa Fluor 594 anti-rabbit antibodies. Nuclei were visualized with DAPI. Micrographs were taken at × 100 magnification. (C) NIH3T3-HA-YB-1 cells were
Referência(s)