QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Vol. 19, Issue 4, 1706-1716, April 2008

This Article Abstract Full Text (PDF) Supplemental Materials All Versions of this Article: E07-07-0637v1 19/4/1706    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Baldin, V. Articles by Coux, O. PubMed PubMed Citation Articles by Baldin, V. Articles by Coux, O.

A Novel Role for PA28-Proteasome in Nuclear Speckle Organization and SR Protein Trafficking

Véronique Baldin^*, Muriel Militello^*, Yann Thomas^*, Christine Doucet^*, Weronika Fic, Stephanie Boireau, Isabelle Jariel-Encontre, Marc Piechaczyk, Edouard Bertrand, Jamal Tazi, and Olivier Coux^*

*Centre de Recherche de Biochimie Macromoléculaire (CRBM-CNRS UMR 5237) and Institut de Génétique Moléculaire de Montpellier (IGMM-CNRS UMR 5535), IFR122, Universités Montpellier 1 et 2, Montpellier, France

Submitted July 5, 2007; Revised January 22, 2008; Accepted January 30, 2008
Monitoring Editor: A. Gregory Matera

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In eukaryotic cells, proteasomes play an essential role in intracellular proteolysis and are involved in the control of most biological processes through regulated degradation of key proteins. Analysis of 20S proteasome localization in human cell lines, using ectopic expression of its CFP-tagged 7 subunit, revealed the presence in nuclear foci of a specific and proteolytically active complex made by association of the 20S proteasome with its PA28 regulator. Identification of these foci as the nuclear speckles (NS), which are dynamic subnuclear structures enriched in splicing factors (including the SR protein family), prompted us to analyze the role(s) of proteasome-PA28 complexes in the NS. Here, we show that knockdown of these complexes by small interfering RNAs directed against PA28 strongly impacts the organization of the NS. Further analysis of PA28-depleted cells demonstrated an alteration of intranuclear trafficking of SR proteins. Thus, our data identify proteasome-PA28 complexes as a novel regulator of NS organization and function, acting most likely through selective proteolysis. These results constitute the first demonstration of a role of a specific proteasome complex in a defined subnuclear compartment and suggest that proteolysis plays important functions in the precise control of splicing factors trafficking within the nucleus.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Proteasomes are large proteolytic complexes that play a central role in intracellular proteolysis and are particularly important for the tightly controlled degradation of many critical regulatory proteins (Ciechanover and Schwartz, 2004). The term proteasomes defines a family of complexes that share a common proteolytic core (the 20S proteasome) but differ in the regulators associated with the complex (Demartino and Slaughter, 1993; Rechsteiner and Hill, 2005). This diversity of proteasome complexes might reflect the necessity for the cell to degrade hundreds of different proteins at the same time, often in a highly selective and regulated manner and in certain cases only at precise locations in the cell (Pines and Lindon, 2005). However, whether the activity of each of these complexes is restricted to a particular set of substrates or to particular cellular domains is presently unclear.

The 20S proteasome is a cylindrical-shaped protein complex, comprising 28 subunits arranged in four stacked rings: two outer rings each made of seven alpha-type subunits (1–7) and two inner rings each made of seven beta-type subunits (β1–β7). The catalytic sites are enclosed in the inner catalytic chamber made by the β-type subunits (Groll et al., 1997). In cells, the 20S proteasome is found associated with a variety of regulators that modulate its functions. The best understood 20S regulator is the 19S complex, which binds in an ATP-dependent manner to both ends of the 20S cylinder to form the 26S proteasome that is, in particular, involved in the degradation of ubiquitylated proteins (Coux et al., 1996; Voges et al., 1999). The 20S proteasome can also associate with a variety of other regulators (DeMartino and Slaughter, 1999), including PA200 (Ortega et al., 2005) and the complexes of the PA28 family, PA28/β and PA28 (Rechsteiner and Hill, 2005), to form other proteolytic complexes that likely carry out specific functions. PA28/β is a heptamer of the two closely related PA28 and PA28β subunits (Whitby et al., 2000). It can bind to both ends of 20S proteasome (Gray et al., 1994) and also can be integrated into "hybrid proteasomes" in which one molecule of 19S complex and one molecule of PA28/β are bound at each end of 20S proteasome (Cascio et al., 2002). PA28/β is essentially cytoplasmic and, in vitro, has been shown to stimulate the peptidase, but not the protease, activities of the 20S proteasome (Ma et al., 1992). It is thought to play an important role in the production of MHC class I antigenic peptides (Groettrup et al., 1996). PA28, initially called "Ki antigen," is similar to PA28 and β (Kandil et al., 1997; Tanahashi et al., 1997), but forms only homoheptamers and is mostly nuclear. Its functions are poorly characterized, but studies in mice (Murata et al., 1999; Barton et al., 2004) and flies (Masson et al., 2003) suggest a role for this complex in cell cycle progression and apoptosis. Furthermore, PA28 has been recently identified as a key player in the degradation of the oncogene SRC-3 (Li et al., 2006) and of the CKI p21 (Chen et al., 2007; Li et al., 2007) and was shown to act as a novel regulator of Cajal Bodies integrity under genotoxic stress (Cioce et al., 2006).

To better understand the specific functions of the various proteasome complexes, it is important to know where in the cells each type of complex is present and active. However, in most previous studies of proteasome localization, only immunofluorescence analyses of the 20S proteasome have been conducted, and it was not known whether the fluorescence corresponded to free 20S or 20S bound to the 19S complex (i.e., the 26S proteasome), or to other regulators. In general, the 20S proteasome appears distributed in both cytoplasmic and nuclear compartments (Reits et al., 1997; Wojcik and DeMartino, 2003), and most of it is found in the soluble fraction after cell lysis. A small percentage, however, seems to be associated with defined cellular structures. In the cytoplasm, the 20S proteasome is found associated with the cytoskeleton (Grossi de Sa et al., 1988), the endoplasmic reticulum (Brooks et al., 2000), and a perinuclear structure that coincides with the centrosome (Wigley et al., 1999). In the nucleus, the 20S proteasome is found in distinct nuclear regions, including PML bodies (Rockel and von Mikecz, 2002; Wojcik and DeMartino, 2003) and nuclear speckles (Rockel and von Mikecz, 2002; Rockel et al., 2005).

Nuclear speckles (NS), also called interchromatin granule clusters (IGCs), correspond to subnuclear domains enriched in components of the pre-mRNA splicing machinery such as the SR protein family (Misteli et al., 1997; Rockel and von Mikecz, 2002; Lamond and Spector, 2003; Rockel et al., 2005). They are highly dynamic domains with a constant in-out shuttling of splicing factors that undergo posttranslational modifications (particularly phosphorylation/dephosphorylation) necessary for their subsequent recruitment at transcription sites (Misteli et al., 1998). Indeed, disassembly of NS leads to dramatic reduction of accumulation of splicing factors at transcription sites and perturbs coordination between transcription and splicing (Sacco-Bubulya and Spector, 2002). The presence of proteasomes in NS, together with the fact that proteasome inhibitors promote the accumulation of SR protein SC35 in NS (Rockel and von Mikecz, 2002), suggests that, as with phosphorylation, proteolysis might be intrinsically involved in the control of the intranuclear trafficking of splicing factors and their repartition between NS and transcription sites.

In this study, we show that a specific form of proteasomes (i.e., 20S proteasome-PA28 complexes) is present in the NS in an active form. Also, we demonstrate that a reduction of PA28 expression by siRNA triggers a profound alteration of nuclear speckles organization and affects the recruitment of SR proteins to transcription sites.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Antibodies and Compounds
Antibodies against green fluorescent protein (GFP) were obtained from Roche Diagnostic (Meylan, France). Antibodies against 3 (PW8115), 4 (PW8120), 6 (PW8100), 7 (PW8110), Mss1/Rpt1 (PW8825), Sug1/Rpt6 (PW9265), PA28 (PW8190), and PA28β (PW8280) were purchased from Affiniti Research Products (Nottingham, UK), and antibodies against SC35 (S4045), SF2/ASF (32–4500), and HA (12CA5) were purchased from Sigma (L'isle d'Abeau, France), Zymed (South San Francisco, CA), and Roche Diagnostic, respectively. Alexa- and HRP-conjugated secondary antibodies were purchased respectively from Interchim (Montluçon, France) and Bio-Rad SA (Hercules, CA). Suc-Leu-Leu-Val-Tyr-7amido-4-methylcoumarin (suc-LLVY-AMC, Bachem Biochimie SARL, Voisins Le Bretonneux, France), MG132 (Boston Biochem, Cambridge, MA), DRB (Sigma), and actinomycin D (Sigma) were dissolved in DMSO. DQ-Ovalbumin (DQ-OVA) was purchased from Calbiochem (La Jolla, CA).

Plasmid Constructs and Small Interfering RNA
Human cDNA (768 base pairs) encoding the full-length 7 (PSMA3 or C8, Accession no. P25788), with or without the final stop codon, was amplified from a human fibroblast cDNA library by PCR and inserted into pcDNA₃ or pML1-ECFP vectors and verified by sequencing. The cDNA coding for 7-ECFP and enhanced cyan fluorescent protein (ECFP) were then inserted into pTRE2 vector (Clontech, Saint-Germain-en-Laye, France). The pEGFP-hSC35, pcDNA₃-Tat, and pEGFP-SF2/ASF constructs were generated in the laboratory of J. Tazi and E. Bertrand, respectively. To construct mammalian expression vectors for bimolecular fluorescence complementation (BiFC) analyses, sequences encoding amino acids 1–154 of enhanced yellow fluorescent protein (EYFP; N-terminal fragment [YN]) and amino acids 155–239 of EYFP (C-terminal fragment of YFP [YC]) were cloned into pcDNA₃. Full-length 7 cDNA (without the final stop) was inserted in pcDNA₃-YN and full-length PA28 cDNA in pcDNA₃-YC to create 7-NY and CY-PA28 fusion proteins, respectively. Small interfering RNA (siRNA) duplexes against human PA28 were purchased from Ambion (Austin, TX; PSME3, ID: 138669; siRNA 1) or MWG-Biotech (Ebersberg, Germany) (siRNA 2; Cioce et al., 2006). Scrambled duplexes (Silencer negative control 1) were purchased from Ambion. A PA28 construct resistant to the siRNA 2 was made by mutation of six nucleotides in PA28 cDNA subcloned in pcDNA₃-3HA vector.

Cell Culture and Transfection
HeLa cells (ATCC, Manassas, VA), U₂OS-tTA (Theis-Febvre et al., 2003), 2E11 cell line (Boireau et al., 2007; du Chéné et al., 2007), and MelJuso cell line expressing constitutively the LMP2-GFP fusion protein (Reits et al., 1997) were cultured as described in the respective articles. U₂OS-tTA cells expressing stably 7-CFP or CFP were established as described (Theis-Febvre et al., 2003). Plasmid transfections in cells were carried out using the JetPEI reagent (Qbiogene, Carlsbad, CA). Twenty-five nanomoles per milliliter of each siRNA were transfected into cells with the oligofectamine reagent (Invitrogen, Carlsbad, CA), following the manufacturer's instructions.

Immunoprecipitation and Immunoblot Analysis
Cells were lysed for 15 min at 4°C in lysis buffer (50 mM Tris-HCl, pH 8.0, 5 mM MgCl₂, 0.5 mM EDTA, 1 mM ATP, 1 mM DTT, 10% glycerol, 0.5% IGEPAL CA630). This treatment does not solubilize NS components (not shown), which can be later solubilized using a buffer containing 0.3% Triton X-100. Lysates were clarified by centrifugation for 10 min at 10,000 x g, and the protein concentration of the supernatant was determined using BSA as a standard (Bradford reagent assay, Pierce, Rockford, IL). Total lysate (200 µg) was precleared for 30 min, and immunoprecipitations were performed using 3 µl of indicated antibodies and protein A or G magnetic beads (Dynal, Lake Success, NY) in lysis buffer containing 100 mM NaCl. After several washes, bead pellets were boiled in SDS sample buffer, separated by SDS-PAGE and transferred to nitrocellulose membranes. Immunoblotting was performed as described previously (Baldin et al., 1993). Visualization was performed by enhanced chemiluminescence (Perkin Elmer-Cetus, Norwalk, CT).

Proteasome Activity Assay
Proteasome activity was measured on beads of immunopurified proteasome by incubation at 37°C for 20 min in 100 µl of activity buffer (20 mM Tris-HCl, pH 7.5, 5 mM MgCl₂, 1 mM ATP, 1 mM DTT, 10% glycerol) containing 100 µM suc-LLVY-AMC. The fluorescence of the free AMC generated by proteasome activity (_ex = 380 nm, _em = 440 nm) was measured on a spectrofluorimeter (Perkin Elmer-Cetus). Proteasome activity in living cells was analyzed in U₂OS cells as described (Rockel et al., 2005).

Indirect Immunofluorescence
Cells were transfected or not with expression vectors encoding the mentioned cDNAs. At indicated times after the transfection and/or induction, cells were fixed using 3.7% formaldehyde and then permeabilized with 0.25% Triton X-100 and cold methanol. Proteasome subunits were detected by incubating the cells with the first antibody, followed by a second incubation with Alexa-594– or 488–conjugated goat anti-mouse (or anti-rabbit) antibody (1:1000; Molecular Probes, Eugene, OR). In all cases, DNA was stained with DAPI dye (Sigma). In situ hybridization was performed as described (Verheggen et al., 2002). Microscopic examinations were performed with a 63x/1.32 NA or 100x/1.4 NA oil immersion objective lens under a Leica DMRA microscope and a 12-bit Micromax YHS 1300 camera (Deerfield, IL). Images were acquired as TIF files using the MetaMorph 6.2 imaging software (Universal Imaging, West Chester, PA).

Fluorescence Recovery after Photobleaching
Fluorescence recovery after photobleaching (FRAP) analyses were performed at 37°C on HeLa cells transiently expressing 7-GFP with a laser scanning microscope (LSM Meta 510; Zeiss, Jena, Germany). The argon laser spectral line (wavelength of 488 nm) was set to an intensity not greater than 0.3% of its total power (458, 477, 488, 514, 30 mW) for image collection. A 40x/1.2 NA water immersion objective was used. After 12 prebleach scans (1 scan/s), a region of interest (ROI) was photobleached using 15 iterations at 3 mW laser power. Twelve-bit images were collected before, immediately after, and at defined intervals after bleaching (once every 0.8 s). Our total experiment time was set to 130 s, which was sufficient for complete recovery from photobleaching. The mobile fraction (R) was determined using the formula: R = (F – F0)/(Fi – F0). _1/2 was directly read off of the graph as the time when fluorescence had recovered by 50%.

RT-PCR
Total RNA was extracted using the Mini RNA isolation II kit (Zymo Research, Orange, CA). One microgram of RNA was reverse-transcribed using the first-strand cDNA synthesis kit (GE Healthcare, Waukesha, WI) according to the supplier's specifications. cDNAs were amplified with the Taq polymerase (Invitrogen) using specific primers corresponding to S26 and artificial gene mRNA (100 and 193 nucleotides, respectively).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Ectopically Expressed Proteasome 7 Subunit Accumulates into Nuclear Foci
To study the subcellular localization of the 20S proteasome in living cells and to avoid overproduction of the protein by transient expression, we established stable U₂OS-tTA cell lines expressing 20S -subunits in fusion with CFP, under an inducible promotor (tet-off). In the tested cell lines, exogenous -CFPs were expressed at a level similar to that of endogenous proteins (Figure 1A; 7-CFP as an example). Surprisingly, in addition to the relatively even cytoplasmic and nuclear distribution of 7 detected by indirect immunofluorescence in parental U₂OS-tTA cells, distinct nuclear foci were apparent in 7-CFP expressing cells (Figure 1B; compare panel endogenous 7 to panel ectopic 7-CFP), which contrasted with the images obtained with cell lines expressing other tagged subunits such as 3-CFP (not shown) or LMP2 (Reits et al., 1997). Importantly, despite the apparent brightness of the 7-CFP containing foci, quantification of the fluorescence using the Metamorph software showed that 7-CFP containing foci represent only 7.7% of the total cellular 7-CFP and 22% of the nuclear 7-CFP.

View larger version (42K): [in this window] [in a new window]   Figure 1. Ectopic 7-CFP and 7 accumulate into nuclear foci. (A) Expression of the 7-CFP protein in cells. Total cellular extracts (30 µg) of human osteosarcoma U2OS-tTA (U) or U2OS-tTA cells stably transfected with vectors conditionally expressing 7-CFP (7) cultured in the absence of tetracycline for 48 h were separated by electrophoresis and analyzed by immunoblotting using anti-7 antibodies. (B) The cellular distribution of 7 in U2OS-tTA cells either untransfected (endogenous 7), stably transfected with pTRE2-7-CFP (ectopic 7-CFP), or transiently transfected with pcDNA3-7 (ectopic 7) was analyzed by indirect immunofluorescence using anti-7 antibodies (red) or by CFP fluorescence (cyan) on fixed cells. Arrows indicate transiently transfected cells. Observation with a 63x objective. Bar, 10 µm.

Biochemical experiments indicated that 7-CFP was readily incorporated into proteasomal complexes (Figure 2). Immunoprecipitation on whole cell extracts demonstrated the presence of other 20S proteasome subunits (6 subunit) as well as 19S complex subunits (Sug1 protein, an ATPase of this complex) in anti-GFP immunoprecipitates and showed that 7-CFP is also present in anti-Sug1 immunoprecipitates (Figure 2A). The immunoprecipitated CFP-containing complexes were proteolytically active when tested against the proteasome substrate suc-LLVY-AMC (Figure 2B, right panel). Morever, using an in situ proteasome-dependent proteolysis assay (Rockel et al., 2005), based on the microinjection of a fluorogenic substrate (DQ-OVA) into the cell nucleus, we demonstrated that 7-CFP complexes were active in the nuclear foci. After microinjection of DQ-OVA in parental U₂OS cell nuclei, we confirmed that its degradation (monitored by the appearance of fluorescence) occurs throughout the nucleoplasm but also in discrete foci, as previously published (Rockel et al., 2005; Figure 2C, top panel). However, in U₂OS cells expressing ectopic 7, the fluorescent foci resulting from DQ-OVA degradation colocalized completely, in number and shape, with 7 positive nuclear foci (Figure 2C, bottom panel). Based on these observations, it can be concluded that the proteasomes complexes containing 7-CFP protein are active in protein degradation.

View larger version (34K): [in this window] [in a new window]   Figure 2. 7-CFP is incorporated into active proteasome complexes that accumulate in nuclear foci. (A) Total cell lysate from human osteosarcoma U2OS-tTA cells stably transfected with vectors conditionally expressing 7-CFP (7) or CFP (CFP) cultured in the absence of tetracycline for 48 h and from MelJuso cells stably expressing LMP2-GFP (LMP2) were subjected to immunoprecipitation (IP) with anti-GFP or anti-Sug1 antibodies. Immunoprecipitated proteins as well as 30 µg of U2OS-tTA total extract (T) were separated by electrophoresis and analyzed by immunoblotting using the indicated antibodies. (B) Proteins immunoprecipitated with either anti-Sug1 or anti-GFP antibodies from total cell extracts were tested for proteasomal activity using the peptide suc-LLVY-AMC as a substrate, in the presence (+) or not (–) of 50 µM MG132 (first bar: no IP in the assay). (C) Proteasomes are active in nuclear foci. U2OS-tTA cells (top panel) or U2OS-tTA cells transiently expressing 7 (bottom panel) were microinjected (24 h after transfection) with the fluorogenic substrate protein DQ-OVA (0.5 mg/ml). The substrate was injected into the nuclei. Cells were incubated 10 min at 37°C and then fixed, and the fluorescent DQ-OVA degradation products were visualized by the appearance of green fluorescence. 7 was detected by indirect immunofluorescence with anti-7 (red; 63x objective). Bar, 10 µm. U2OS-tTA-7-CFP cells were not used in this experiment because of the overlap of the fluorescence spectrums of CFP and of DQ-OVA degradation products.

View larger version (80K): [in this window] [in a new window]   Figure 3. Ectopic expression of 7-CFP promotes recruitment of whole 20S proteasomes into the nuclear foci. Parental U2OS-tTA cells (A) and U2OS-tTA-7-CFP cells induced for 48 h for 7-CFP expression (B) were subjected to indirect immunofluorescence using antibodies directed against the 4 and 6 subunits of the 20S proteasome (red). 7-CFP fusion protein was detected by CFP fluorescence. Wide-field overlay images and enlargements of marked area are shown. All observations were done with a 63x objective. Bar, 10 µm. (C) Expression of 7-CFP does not alter expression level of other 20S proteasome subunits. Thirty micrograms of total protein extract from U2OS-tTA (U) and induced U2OS-tTA-7-CFP (7) cells were separated by electrophoresis and analyzed by immunoblot (IB) using the indicated antibodies.

View larger version (63K): [in this window] [in a new window]   Figure 4. The nuclear foci enriched in 20S proteasome-7-CFP complexes correspond to the NS. (A) U2OS-tTA-7-CFP were fixed 48 h after 7-CFP induction, permeabilized, and stained with a mAb directed against the phosphorylated NS protein SC35 (red) and with DAPI dye (blue). 7-CFP fluorescence is in green for easier codetection (7-CFP/SC35 image). Wide-field of individual detection and overlay images are presented (63x objective). Bar, 10 µm. (B) Forty-eight-hour–induced U2OS-tTA-7-CFP cells, untreated or treated with DRB (100 µM) or actinomycin D (10 µg/ml) for 2 h before fixation, were stained with anti-SC35 (red); 7-CFP expression was directly visualized by CFP fluorescence (cyan; 100x objective). Bar, 10 µm.

20S Proteasome-PA28 Complexes, But Not 26S Proteasome Complexes, Are Recruited to NS upon 7 Expression

View larger version (77K): [in this window] [in a new window]   Figure 5. PA28, but not other 20S proteasome regulatory complexes, is recruited together with 20S proteasome into the nuclear foci. (A) U2OS-tTA-7-CFP cells induced for 48 h were analyzed by indirect immunofluorescence using antibodies directed against the Mss1/Rpt1 and Sug1/Rpt6 ATPase subunits of the 19S complex (red) or PA28β. 7-CFP fusion protein was detected by CFP fluorescence. (B) Subcellular localization of PA28 (indirect immunofluorescence, red) in parental U2OS-tTA cells and in induced U2OS-tTA-7-CFP cells. 7-CFP was detected by direct fluorescence. Wide field of individual areas of detection, overlay images, and enlargements of marked area are presented. All observations were done with a 63x objective. Bar, 10 µm.

We confirmed the interaction between 7-CFP–containing 20S proteasome and PA28 in cell extracts by coimmunoprecipitation from mildly solubilized (S1 fraction, no extraction of 7-CFP nuclear foci) or triton-solubilized cell extracts (S2 fraction, obtained after solubilization of 7-CFP nuclear foci) using antibodies directed against either 7, GFP, or PA28 (Figure 6A). Even though the majority of 20S proteasome-PA28 complexes was present in the S1 fraction, the complexes were also present in the S2 fraction enriched in NS (Figure 6A) and were active when tested for degradation of the proteasomal peptide substrate Suc-LLVY-AMC (data not shown). Notably, the distribution of the complexes between the S1 and S2 fractions is in agreement with the low percentage of 7-CFP present in the nuclear foci, estimated by fluorescence quantification (see above). We then used the BiFC technique to evaluate directly, in cellulo, the ability of 7 to interact with PA28 in the NS. This technique is based on the fact that the YFP chromophore can be reconstituted in cells when two chimeric proteins, each harboring half of the chromophore, interact (Kerppola, 2006). When 7 fused to the C-terminal half of YFP (7-CY) was coexpressed in cells with PA28 fused to the N-terminal half of YFP (NY-PA28), the fluorescence could be reconstituted throughout the nuclei, with a marked accumulation in nuclear foci, which most likely correspond to NS (Figure 6B). Interaction was also observed in the nucleus between molecules of PA28, as expected because PA28 forms heptamers and curiously between molecules of 7 as well (Figure 6B; see Discussion). Together, these results confirm the in cellulo interaction between 20S proteasome and its PA28 regulator and strongly suggest that 20S proteasome-PA28 complexes are indeed present in the NS.

View larger version (61K): [in this window] [in a new window]   Figure 6. 7 and PA28 are physically associated in total cell extracts, in a nuclear foci-enriched fraction and in cellulo. (A) After lysis of induced U2OS-tTA-7-CFP cells, the cell extract was clarified by centrifugation, and the supernatant was collected (S1). The insoluble material was then resuspended in lysis buffer containing 0.3% Triton X-100 (15 min at 4°C) and clarified by centrifugation, yielding the S2 supernatant. Proteins immunoprecipitated from the S1 and S2 supernatants, using the indicated antibodies, were then separated by SDS-PAGE and analyzed by immunoblotting using antibodies directed against the proteasome subunits 7 (left panel), β2 (right top panel) or 4 (right bottom panel). T: 30 µg of total proteins. (B) The interaction between 7-CFP and PA28 was visualized in cellulo by BiFC. Proteins indicated in each panel were coexpressed in HeLa cells. Twenty-four hours after transfection of expression vectors, cells were incubated at 32°C for a further 24 h to promote chromophore maturation and then fixed to allow YFP fluorescence monitoring. c-Fos (118–210) and c-Jun (257–318) constructs (Hu et al., 2002) were used as positive (Jun-NY/Fos-CY) or negative (Jun-NY/7-CY and NY-PA28/Fos-CY) BiFC controls (63x objective). Bar, 10 µm.

PA28 Is Present in the Nuclear Speckles without Expression of 7-CFP

View larger version (49K): [in this window] [in a new window]   Figure 7. Presence of PA28 in the NS without 7-CFP expression and association of 20S proteasome with SF2/ASF protein. (A) Parental U2OS-tTA cells were permeabilized and fixed simultaneously to remove most of soluble PA28 complexes. Cells were then stained with anti-SC35 (green) and anti-PA28 (red), and an overlay image is shown (100x objective). (B) U2OS-tTA cells, cotransfected with pcDNA3-7 and pEGFP-SF2/ASF vectors, were lysed 24 h after transfection in lysis buffer containing 0.3% Triton X-100. Two hundred micrograms of total cell extract were subjected to immunoprecipitation using either anti-4, anti-PA28 or anti-GFP antibodies. Immunoprecipitated proteins were then separated by SDS-PAGE (10%) and visualized by immunoblotting using anti-GFP antibody. T, 30 µg of total cells extract; S, supernatant of the IP (1/10); IP, immunoprecipitation.

Roles of 20S Proteasome-PA28 Complexes in the Structural Integrity of the NS

View larger version (44K): [in this window] [in a new window]   Figure 8. Proteasome-PA28 complexes are required for the integrity of the NS. (A) U2OS-tTA-7-CFP cells were induced for 7-CFP expression and concomitantly transfected with control-siRNA or PA28-siRNA (siRNA 1) duplexes. Cells were recovered 48 h after transfection, and the expression level of PA28 was analyzed either by immunoblotting or by indirect immunofluorescence using an anti-PA28 antibody. Bar, 10 µm. (B) NS were visualized by the fluorescence of 7-CFP and by indirect immunofluorescence using anti-SC35 antibodies in induced U2OS-tTA-7-CFP cells untransfected or transfected with PA28- or control-siRNA duplexes. Enlargements of the marked areas are presented. Bar, 10 µm. (C) Quantification of the number of 7-CFP–labeled NS in cells treated with PA28- or control-siRNA duplexes. Quantification was performed visually, by counting on the pictures the number of compact fluorescent foci in each cell. The values correspond to the means of five independent experiments (n = 100 cells, ±SD). (D) 3D reconstruction. SC35 localization was detected by indirect immunofluorescence in induced U2OS-tTA-7-CFP cells treated with control- or PA28 (1)-siRNA. Fixed cells were observed with a Leica DMRA microscope equipped with a 63x PL APO (NA = 1.32) oil immersion objective and a N2.1 (Leica) filter set. Stacks of images were acquired using a piezo stepper (Physik Instruments, Waldbronn, Germany), Metamorph 7.1 (Molecular Devices, Menlo Park, CA) and a Micromax 1300YHS CCD camera (Princeton Research Instruments, Princeton, NJ). Stacks were further deconvolved using a MLE algorithm and the Huygens 2.9 software (Scientific Volume Imaging, Hilversrum, The Netherlands) and analyzed in 3D using Imaris 5.3 (Bitplane, Zurich, Switzerland). (E) NS were visualized by the fluorescence of 7-CFP and by indirect immunofluorescence using anti-SF2/ASF antibodies in induced U2OS-tTA-7-CFP cells untransfected or transfected with PA28- or control-siRNA duplexes. Bar, 10 µm. (F) In parental U2OS-tTA cells, untreated or treated with PA28-siRNA duplexes, NS were visualized by indirect immunofluorescence using anti-SC35 antibodies. Bar, 10 µm.

View larger version (35K): [in this window] [in a new window]   Figure 9. Specificity of the effect of the siRNA directed against PA28. U2OS-tTA cells were treated or not with PA28-siRNA duplexes #2, PA28-siRNA duplexes #2, and pcDNA3-3HA-PA28 (allowing the expression of a PA28 construct refractory to siRNA #2), and pcDNA3-3HA-PA28 alone, as indicated. (A) Expression levels of endogenous or exogenous PA28 were analyzed by immunoblotting using anti-PA28 or -HA antibodies. (B) Analysis of SC35 and PA28 localization by indirect immunofluorescence as described in Figure 8. Bar, 10 µm.

View larger version (71K): [in this window] [in a new window]   Figure 10. Effect of PA28 siRNA on subnuclear domains. Localization of proteins marker of Cajal (A) and PML bodies (B) was analyzed in induced U2OS-tTA-7-CFP cells treated with control- or PA28-siRNA, as described in Figure 8. Cells were stained with antibodies raised against three components of CB: TGS1, SMN, and coilin (A), or with antibodies raised against the protein PML (B). Bar, 10 µm.

View larger version (47K): [in this window] [in a new window]   Figure 11. PA28-siRNA reduces SF2/ASF accumulation at transcription sites. 2E11 cells were cotransfected with pcDNA3-Tat, pEGFP-SF2/ASF, and control- or PA28-siRNA (1) duplexes. After 48 h, cells were fixed and subjected to in situ hybridization using a MS2-Cy3 probe. (A) Right, merged images of the intracellular distribution of SF2/ASF-GFP (green) and MS2 TS and mRNA (red) under the indicated conditions. Left, magnification of the TS (white square): MS2 detection (red), SF2/ASF-GFP (green), merge (100x objective). Bar, 10 µm. (B) Quantification of the level of SF2/ASF-GFP recruited at the TS. The relative levels of SF2/ASF-GFP at the TS and at the NS was estimated by quantification of the green fluorescence using the Metamorph software, in cells expressing similar levels of transcripts (estimated by quantification of MS2-Cy3 at TS) and SF2/ASF-GFP in the NS. The values correspond to the means of four independent experiments (n = 40 cells, ±SD). See Figure S3 for illustration of the quantification procedure. (C) PA28-siRNA does not affect the level of transcription. Relative mRNA levels for the ribosomal S26 subunit (control) and for the artificial gene in 2E11 cells untreated or treated with the indicated siRNA were analyzed by RT-PCR and agarose gel electrophoresis.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
We show here that active 20S proteasome-PA28 complexes are present in the NS and that reduction of PA28 levels, using siRNA, dramatically alters NS organization and nuclear trafficking of splicing factors. Our data thus demonstrate a new role of 20S-PA28 complexes in nuclear organization and provide novel important information for our understanding of both the specific functions and localization of the various forms of proteasomes and the regulation of NS structure and dynamics.

Our discovery originates from the serendipitous observation that ectopic expression of the proteasomal 7 subunit triggers accumulation of 20S proteasome-PA28 complexes in the NS. This observation raises the intriguing and puzzling problem of how this effect is mediated. One possibility was that overexpression of 7 led to an increase of the cellular concentration of 20S proteasome, but we found no evidence to support this supposition. One alternative hypothesis is that 7 ectopic expression entails incorporation of more than two 7 subunits into individual 20S proteasome complexes. Indeed, some plasticity in proteasome assembly has been documented in yeast, in which 3 could be replaced by 4 (Velichutina et al., 2004). In this respect, it is remarkable that 7 seems to possess a greater propensity to interact to PA28 than other subunits. When we used the proteasome subunit 3, no interaction with PA28 could be detected by BiFC, contrary to the result obtained for 7, suggesting that 1) PA28 does not interact identically with all subunits of the proteasome and 2) within the 20S proteasome, the 7 subunit possesses certain sequence and/or structural features that makes it a favored partner for PA28. Thus, a proteasome with more than one 7 subunit per -ring could either have a higher affinity for PA28 or impose a particular conformation to the complex formed with PA28, two mechanisms that in turn could impact on the steady-state nuclear distribution of the 20S-PA28 complex.

In any case, even if the exact reasons for accumulation of 20S-PA28 in NS upon 7 ectopic expression remain to be elucidated, several arguments show that the presence of 20S-PA28 complexes in the NS is physiological, albeit not easily detectable normally. In particular, the presence of active 20S proteasome in the NS, as previously documented by others (Rockel et al., 2005), together with our results demonstrating that 1) PA28 can be detected in the NS without ectopic 7 expression, 2) knockdown of endogenous PA28 alters NS morphology, and 3) 20S proteasomes can interact with PA28 in the NS (as illustrated by colocalization, affinity purification, and in vivo interaction; BiFC), demonstrate that these complexes constitute an important active form of the proteasome in the NS.

Regarding the function(s) of 20S-PA28 complexes in the NS, it is most likely that these complexes mediate the local degradation of specific proteins, even though they certainly have other substrates in other nuclear regions. We hypothesized that substrates of 20S-PA28 complexes in the NS could be the SR proteins themselves because it has been reported that the intrinsic NS component SC35 is stabilized by proteasome inhibitors (Rockel and von Mikecz, 2002). However, in our cells, expression of 7-CFP or knockdown of PA28 did not alter the half-life of SC35 (data not shown). Other attractive substrates for 20S-PA28 complexes are kinases or phosphatases present in the NS (Colwill et al., 1996; Misteli and Spector, 1996; Wang et al., 1998; Ko et al., 2001). Indeed, these enzymes control the phosphorylation status of SR proteins and thereby their recruitment at transcription sites (Misteli et al., 1998). Thus, alteration of the expression level of one or several of these enzymes could explain the alteration of both the morphology of the NS and the recruitment of SR proteins to transcription sites seen upon PA28 knockdown. Obviously, to understand the role(s) of 20S-PA28 complexes in the NS, an important development for the future will be to identify their nuclear substrates, particularly those localized in the NS.

Nevertheless, even without this information, our observations provide additional demonstration that, within the family of proteasome complexes, 20S proteasome-PA28 complexes have specific functions in the cell nucleus. Until recently, the functions of PA28 have remained quite elusive. Genetic manipulations in mice and Drosophila have suggested that it might be involved in cell cycle progression and have anti-apoptotic functions (Rechsteiner and Hill, 2005). However, its exact role in these processes is not understood. Biochemical analyses have shown that PA28 can activate peptide degradation by the 20S proteasome in vitro (Realini et al., 1997), but it has been demonstrated only recently that PA28 is indeed involved in the degradation of proteins, namely the HCV (hepatitis C virus) core protein (Moriishi et al., 2003), the SRC-3 oncogene, and the CKI p21 (Li et al., 2006, 2007; Chen et al., 2007). PA28 has also been identified as a novel regulator of Cajal bodies integrity (Cioce et al., 2006), but it was unclear in this case whether 20S proteasome was required for this process. We now additionally show that 20S-PA28 complexes are involved in the control of nuclear trafficking of splicing factors, because reduction of PA28 levels by siRNA strongly perturbs NS morphology and impairs recruitment of SR proteins (SF2/ASF) at transcription sites. The physiological importance of this role is presently unclear, because knockdown of PA28 did not appear to affect transcription—as evaluated by RT-PCR of a specific gene (Figure 11C) or globally by 5-fluorouracil (5-FU) incorporation (Figure S4A), nor to impair globally the specificity of the splicing machinery, because no variations were detected on various endogenous and transfected genes subjected to alternative splicing (Figure S4, B and C). Our results are similar to previously described data (Caceres et al., 1997; Misteli et al., 1998) showing that an SF2/ASF mutant (-RS) was able to perform alternative splicing in vivo, in a manner very similar to the wild type, in absence of its recruitment to transcription sites, and suggesting that recruitment to the TS does not always reflect the functional properties of splicing factors. However, one possibility in our case is that the low level of SF2/ASF still present at TS is in fact sufficient for normal splicing.

Another question arising from our data are how 20S-PA28 complexes are targeted to the NS. Our FRAP analyses demonstrated that 7-GFP (and therefore the proteasome) mobility is high in the NS, indicating that the complexes shuttle into and out of the NS. Currently, the mechanism of the targeting of the complex to the NS is unclear, because neither proteasome subunits nor PA28 display signature motifs of NS proteins, such as RS domains, for example. Possibly, protein(s) binding to 20S proteasome and/or PA28 could act as adaptor(s) facilitating the docking of the complex to the NS. Interestingly, because we detected a physical interaction between the 20S proteasome and SF2/ASF, SR proteins themselves could be responsible for targeting proteasomes to the NS.

In conclusion, the data presented here demonstrate that 20S proteasome associated with its PA28 regulator plays a critical role in the NS because it appears necessary for their structural organization and for intranuclear trafficking of splicing factors. Our data also suggest that the specificity of proteasome-dependent proteolysis, which is often considered as being regulated at the level of the ubiquitylation, can be controlled by complex mechanisms operating at the level of the proteasome and its intracellular localization. Further analysis of the role of 20S proteasome-PA28 complexes in the NS should provide new information on the mechanisms involved in the control of nuclear organization. Our results highlight a novel layer of complexity in the regulation of NS architecture and function, as well as SR protein trafficking, and suggest that intracellular proteolysis is intrinsically involved in the regulation of nuclear plasticity.

	ACKNOWLEDGMENTS

 
We thank E. Reits (University of Amsterdam, The Netherlands) for MelJuso cells, U. Seifert (Charité University, Berlin, Germany) for PA28 cDNA, and R. Bordonné (IGMM-CNRS, Montpellier, France) for antibodies (TGS1 and SMN). We also thank our colleagues for their help, suggestions, and criticisms and the cell imaging facility (Montpellier RIO Imaging) for the quality of its service. This work was supported by a grant from the Association pour la Recherche sur le Cancer to O.C. (ARC no. 3243), an ACI-BCMS from the French Research Ministry (O.C.), the European contract no. QLG1-CT-2001-02026 (O.C.), and a grant to J.T. from the Agence Nationale de la Recherche (ANR-05-BLAN-0261-01). C.D. was a recipient of a fellowship from the Fondation de la Recherche Médicale (FRM). E.B. and M.P. laboratories are "Equipes labelisées de la Ligue Nationale contre le Cancer."

	Footnotes

 
This was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E07-07-0637) on February 6, 2008.

Address correspondence to: Olivier Coux (olivier.coux{at}crbm.cnrs.fr)

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Baldin, V., Lukas, J., Marcote, M. J., Pagano, M., and Draetta, G. (1993). Cyclin D1 is a nuclear protein required for cell cycle progression in G1. Genes Dev 7, 812–821.[Abstract/Free Full Text]

Barton, L. F., Runnels, H. A., Schell, T. D., Cho, Y., Gibbons, R., Tevethia, S. S., Deepe, G. S., Jr, and Monaco, J. J. (2004). Immune defects in 28-kDa proteasome activator gamma-deficient mice. J. Immunol 172, 3948–3954.[Abstract/Free Full Text]

Boireau, S. et al. (2007). The transcriptional cycle of HIV-1 in real-time and live cells. J. Cell Biol 179, 291–304.[Abstract/Free Full Text]

Brooks, P., Murray, R. Z., Mason, G. G., Hendil, K. B., and Rivett, A. J. (2000). Association of immunoproteasomes with the endoplasmic reticulum. Biochem. J 352, 611–615.[CrossRef][Medline]

Caceres, J. F., Misteli, T., Screaton, G. R., Spector, D. L., and Krainer, A. R. (1997). Role of the modular domains of SR proteins in subnuclear localization and alternative splicing specificity. J. Cell Biol 138, 225–238.[Abstract/Free Full Text]

Cascio, P., Call, M., Petre, B. M., Walz, T., and Goldberg, A. L. (2002). Properties of the hybrid form of the 26S proteasome containing both 19S and PA28 complexes. EMBO J 21, 2636–2645.[CrossRef][Medline]

Chen, M., Rockel, T., Steinweger, G., Hemmerich, P., Risch, J., and Von Mikecz, A. (2002). Subcellular recruitment of fibrillarin to nucleoplasmic proteasomes: implications for processing of a nucleolar autoantigen. Mol. Biol. Cell 13, 3576–3587.[Abstract/Free Full Text]

Chen, X., Barton, L. F., Chi, Y., Clurman, B. E., and Roberts, J. M. (2007). Ubiquitin-independent degradation of cell-cycle inhibitors by the REGgamma proteasome. Mol. Cell 26, 843–852.[CrossRef][Medline]

Ciechanover, A., and Schwartz, A. L. (2004). The ubiquitin system: pathogenesis of human diseases and drug targeting. Biochim. Biophys. Acta 1695, 3–17.[Medline]

Cioce, M., Boulon, S., Matera, A. G., and Lamond, A. I. (2006). UV-induced fragmentation of Cajal bodies. J. Cell Biol 175, 401–413.[Abstract/Free Full Text]

Colwill, K., Pawson, T., Andrews, B., Prasad, J., Manley, J. L., Bell, J. C., and Duncan, P. I. (1996). The Clk/Sty protein kinase phosphorylates SR splicing factors and regulates their intranuclear distribution. EMBO J 15, 265–275.[Medline]

Coux, O., Tanaka, K., and Goldberg, A. L. (1996). Structure and functions of the 20S and 26S proteasomes. Annu. Rev. Biochem 65, 801–847.[CrossRef][Medline]

Demartino, G. N., and Slaughter, C. A. (1993). Regulatory proteins of the proteasome. Enzyme Protein 47, 314–324.[Medline]

DeMartino, G. N., and Slaughter, C. A. (1999). The proteasome, a novel protease regulated by multiple mechanisms. J. Biol. Chem 274, 22123–22126.[Free Full Text]

du