• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Mar 20, 2007; 104(12): 5073–5078.
Published online Mar 14, 2007. doi:  10.1073/pnas.0608773104
PMCID: PMC1829266
Medical Sciences

Modification of nuclear PML protein by SUMO-1 regulates Fas-induced apoptosis in rheumatoid arthritis synovial fibroblasts

Abstract

The small ubiquitin-like modifier (SUMO)-1 is an important posttranslational regulator of different signaling pathways and involved in the formation of promyelocytic leukemia (PML) protein nuclear bodies (NBs). Overexpression of SUMO-1 has been associated with alterations in apoptosis, but the underlying mechanisms and their relevance for human diseases are not clear. Here, we show that the increased expression of SUMO-1 in rheumatoid arthritis (RA) synovial fibroblasts (SFs) contributes to the resistance of these cells against Fas-induced apoptosis through increased SUMOylation of nuclear PML protein and increased recruitment of the transcriptional repressor DAXX to PML NBs. We also show that the nuclear SUMO-protease SENP1, which is found at lower levels in RA SFs, can revert the apoptosis-inhibiting effects of SUMO-1 by releasing DAXX from PML NBs. Our findings indicate that in RA SFs overexpression of SENP1 can alter the SUMO-1-mediated recruitment of DAXX to PML NBs, thus influencing the proapoptotic effects of DAXX. Accumulation of DAXX in PML NBs by SUMO-1 may, therefore, contribute to the pathogenesis of inflammatory disorders.

Keywords: inflammation, autoimmunity, DAXX, SENP

Rheumatoid arthritis (RA) is the most common inflammatory disorder of the joints. It is characterized by chronic inflammation, autoimmune phenomena, and synovial hyperplasia and leads to the progressive destruction of articular structures (1). Although the role of inflammatory cells and their mediators in promoting synovial hyperplasia and disease progression is well established, the specific contribution of resident fibroblast-like cells has emerged only recently. It has been understood that in RA synovial fibroblasts (SFs) exhibit features of stable activation that facilitate the switch from acute to chronic inflammation and result in a tumor-like, aggressive behavior (2, 3). As in various malignancies, resistance against Fas-induced cell death is a characteristic feature of these cells, although in vitro experiments have shown considerable variability in the sensitivity of RA SFs to apoptosis (4, 5).

The small ubiquitin-like modifier (SUMO)-1 is of interest for this process, because it can associate with the Fas-associated death domain (FADD) (6), and overexpression of SUMO-1 was shown to protect BJAB cells from Fas-induced cell death (7). SUMO-1 also binds and modifies the FADD-interacting protein DAXX that, when overexpressed transiently, increases Fas-mediated apoptosis (8, 9). Previous studies have demonstrated increased levels of SUMO-1 in SFs from patients with RA but not in patients with degenerative joint diseases (osteoarthritis, OA) (10), and it has been speculated that increased expression of SUMO-1 may contribute to the activated phenotype of RA SFs (4). However, the lack of data showing in vivo interactions of SUMO-1 with FADD, together with the observation that overexpression of SUMO-1 can induce apoptosis, have challenged the antiapoptotic properties of SUMO-1.

Here, we show that SUMO-1 is involved in the resistance of RA SFs against Fas-induced apoptosis and that SUMOylation of the nuclear promyelocytic leukemia (PML) protein with subsequent “trapping” of the proapoptotic adaptor molecule DAXX in nuclear bodies (NBs) constitutes the mechanism by which SUMO-1 exerts its antiapoptotic effects in RA SFs. We also show that the nuclear SUMO- specific protease SENP1 is not increased in these cells but that overexpression of SENP1 in RA SFs can modulate the apoptosis-inhibitory effects of SUMO-1 by releasing the transcriptional repressor DAXX from PML NBs.

Results

SUMO-1 Confers the Resistance of RA SFs to Fas-Mediated Cell Death and Is Found at Elevated Levels in PML NBs.

We first investigated whether increased expression of SUMO-1 contributes to the resistance of RA SFs against Fas-induced cell death. Confirming our previous observations (10), primary RA SFs expressed significantly higher levels of SUMO-1 than SFs from OA patients (data not shown). To analyze the susceptibility of RA and OA SFs to Fas-induced apoptosis, we plated the cells at identical numbers and stimulated them with recombinant human (rh)FasL as described (11). As a positive control staurosporin was used to induce apoptosis. Apoptosis was determined by a histone fragmentation assay and confirmed by FACS analysis with propidium iodide labeling of intracellular DNA. As shown in Fig. 1a, RA SFs showed similar rates of spontaneous cell death but were significantly less susceptible to Fas-induced apoptosis than OA SFs. To establish a functional relationship with the expression of SUMO-1 and determine whether reduced expression of SUMO-1 restores the normal apoptotic response of RA SFs, we used siRNA to ablate the expression of SUMO-1. RA SFs were transfected with siRNA to SUMO-1 by nucleofection (Amaxa, Cologne, Germany), and apoptosis was induced with rhFasL 48 h after transfection. As shown in Fig. 1 b and c, transfer of siRNA to SUMO-1 inhibited the expression of SUMO-1 both at the mRNA (Fig. 1b) and the protein (Fig. 1c) level. siRNA to SUMO-1 affected all major SUMOylated proteins, including SUMO-1/RanGAP1, which is one of the most prominently expressed SUMOylated proteins and has been used in other expression studies on SUMO-1. Knockdown of SUMO-1 by siRNA clearly sensitized RA SFs to Fas-mediated apoptosis (Fig. 1 d and e). These effects were SUMO-1-specific, because mock siRNA was unable to alter the apoptotic response of the cells (Fig. 1 d and e). Staurosporin induced apoptosis in >90% of the cells after 12 h with no differences between SUMO-1-expressing and silenced cells (data not shown). To determine whether long-term overexpression or inhibition of SUMO-1 in a similar way alters the susceptibility of RA SFs to Fas-mediated apoptosis, we generated retroviral constructs that expressed either WT SUMO-1 or a full-length antisense construct to SUMO-1. RA SFs were infected with the retroviruses and selected with G418 over 60 days. As shown in Fig. 1f, WT SUMO-1 decreased the ability of RA SFs to undergo apoptosis after stimulation with rhFasL when compared with the mock-transfected controls. In contrast, gene transfer of the antisense construct sensitized the cells to Fas-induced cell death to a similar extent as RNAi (Fig. 1d).

Fig. 1.
SUMO-1 protects RA SFs against FasL-induced apoptosis. (a) As determined by a histone fragmentation assay, RA SFs were significantly less susceptible to FasL-induced apoptosis than were OA SFs. The data show the means (± SEM) of 10 RA-SF and 10 ...

Next, we performed immunostainings to study the distribution of SUMO-1 in RA and OA synovial cells. Immunohistochemical double-labeling of RA and OA synovial tissues with antibodies against SUMO-1 and PML revealed strong staining for SUMO-1 in RA tissues with SUMO-1 being prominently localized in the nucleus of the synoviocytes. There it was found in distinct foci costaining with PML. In contrast, far less such prominent focal colocalization was seen in OA synovial cells (Fig. 2a), indicating that increased expression of SUMO-1 in RA synoviocytes strongly affects nuclear association with PML. To further analyze the subcellular distribution of SUMO-1 in RA and OA SFs, immunocytochemistry and confocal laser scanning microscopy was used. Of interest, we were unable to locate SUMO-1 to the cell membrane and particularly to FADD even after stimulation of RA SFs with rhFasL (data not shown). Rather, as can be seen from a representative picture of multiple stainings in Fig. 2a, SUMO-1 was found predominantly in the nucleus, and a considerable proportion of SUMO-1 was found in association with PML. When compared with OA SFs, RA cells exhibited a markedly enhanced nuclear staining for SUMO-1 and a stronger association of SUMO-1 with PML (Fig. 2 b and c).

Fig. 2.
Increased expression of SUMO-1 in the nucleus of RA SFs versus OA SFs results in increased SUMOylation of PML in RA. (a) Immunohistochemical double-staining of RA and OA synovial tissues with antibodies against SUMO-1 (blue) and PML (red) showed strong ...

Expression of SENP1 Is Decreased in RA and Overexpression of SENP1 Sensitizes RA SFs to Fas-Mediated Cell Death.

Next, we hypothesized that if SUMOylation of PML mediates the antiapoptotic effects of SUMO-1, deconjugation of SUMO-1 from PML through overexpression of the nuclear SUMO-specific protease SENP1 would sensitize RA SFs to FasL-induced apoptosis. Analysis of the endogenous levels of SENP1 in RA and OA SFs revealed that there was no compensatory up-regulation of SENP1 in RA fibroblasts along with SUMO-1. Rather RA SFs showed less SENP1 expression at the mRNA level (Fig. 3a) and when nuclear protein extracts were examined by immunoblot (Fig. 3b). We then generated two expression constructs of SENP1, in which WT SENP1 (SENP1wt) or a catalytically inactive C603A mutant SENP1 (SENP1mt) were fused to GFP at their C terminus. RA SFs were transfected with SENP1wt or SENP1mt by nucleofection, and the expression of the transgenes was monitored by fluorescence microscopy and quantitative real-time PCR. As shown in Fig. 3, transfection of the RA SFs with these constructs resulted in a significant overexpression of SENP1wt and SENP1mt (Fig. 3c), and 48 h after transfection, SENP1 was expressed prominently in the nucleus of these cells, and there were no obvious differences in the expression of SENP1wt and SENP1mt with respect to the intracellular localization (Fig. 3d). Transient overexpression of SENP1 did not alter the rate of spontaneous apoptosis but most significantly increased the susceptibility of RA SFs to Fas-induced cell death (Fig. 3 e and f). The effects required a functional protease domain, because the mutant SENP1mt was unable to sensitize the RA SFs to Fas-mediated apoptosis.

Fig. 3.
The nuclear SUMO- protease SENP1 is expressed at decreased levels in RA SFs compared with OA SFs and facilitates FasL-induced apoptosis. (a and b) Quantitative real-time PCR revealed significantly higher expression of SENP1 mRNA in OA SFs vs. RA SFs (means ...

Deconjugation of SUMO-1 from PML Results in the Release of DAXX from PML NBs.

To determine whether the effects of SENP1 on Fas-induced apoptosis are accompanied by alterations in the association of DAXX with PML, we transiently transfected RA SFs with the GFP-tagged constructs of SENP1 and analyzed the subcellular distribution of DAXX, SENP1, SUMO-1, and PML by immunofluorescence staining and confocal laser scanning microscopy. As transfection efficacy ranged between 30% and 50% in our experiments (data not shown), we were able to study the distribution of these proteins in SENP1-transfected and untransfected cells at the same time and under identical conditions. Confirming previous data, we found that SUMO-1 colocalizes with DAXX (Fig. 4a) and PML (Fig. 4b) and that SENP1 is targeted to PML-NBs. This was seen particularly from the experiments where the nonfunctional mutant SENP1mt was used (Fig. 4). Overexpression of SENP1wt removed SUMO-1 from its nuclear depots (Fig. 4a) but did not decrease nuclear accumulation of PML (Fig. 4b). Rather, PML was found in fewer but larger aggregates in SENP1wt-transfected fibroblasts. However, de-SUMOylation of PML by SENP1wt clearly promoted the loss of DAXX from the PML NBs (Fig. 4 a Lower and b Lower). This loss of DAXX from the nucleus depended on the catalytic activity of SENP1 as SENP1mt was unable to demonstrate these effects (Fig. 4 a Upper and b Upper).

Fig. 4.
SENP1 removes SUMO-1 from NBs and results in the release and increased FADD association of the transcriptional repressor DAXX. (a) Double fluorescence staining of RA SFs overexpressing SENP1mt (Upper) or SENP1wt (Lower) with antibodies against SUMO-1 ...

Sequestration of DAXX Is Increased in the Nuclei of RA SFs, and DAXX Is Required for the Proapoptotic Effects of SENP1 in RA SFs.

Finally, we analyzed whether increased SUMOylation of PML in RA SFs results in higher levels of PML-bound DAXX and whether DAXX is required functionally for the proapoptotic effects of SENP1. First, we used immunohistochemical double staining of RA and OA synovial tissues with antibodies against DAXX and PML to study the colocalization of both proteins in vivo. As can be seen in Fig. 5a, there was a strong nuclear localization of DAXX in RA synoviocytes, and DAXX colocalized with PML in the majority of these cells. In contrast, significantly less such focal colocalization was seen in OA synovial cells. To confirm this observation in vitro, we performed immunoprecipitation of nuclear extracts with anti-PML antibodies and subsequent immunoblot with antibodies against DAXX to determine the levels of PML NB- bound DAXX in RA SFs and OA SFs. As shown in Fig. 5 b and c, there was a markedly enhanced sequestration of DAXX in PML NBs of RA SFs. Using siRNA, we next ablated the expression of DAXX in RA SFs. As demonstrated in Fig. 5, inhibition of DAXX expression was 70% at the mRNA level (Fig. 5d), and DAXX protein expression was reduced in siRNA-transfected cells (Fig. 5e). We then cotransfected RA SFs with SENP1wt and siRNA to DAXX and found that SENP1wt failed to sensitize RA SFs to FasL-induced apoptosis when DAXX expression was reduced (Fig. 5f). The effects were specific for DAXX, as mock siRNA cotransfection with SENP1wt facilitated Fas-induced cell death in a similar way as SENP1wt alone (Fig. 5f). To study whether the release of DAXX from PML NBs by SENP1 results in altered binding to the FADD, we performed immunoprecipitation of cell lysates of SENP1wt and mock-transfected RA SFs with antibodies against FADD followed by immunoblot using antibodies against DAXX. As shown in Fig. 5 g and h, these experiments clearly demonstrated an increased association of DAXX with FADD after liberation from the PML NBs in RA SFs.

Fig. 5.
The effects of SENP1 on Fas-induced apoptosis of RA SFs are mediated by DAXX, which is at elevated levels in PML NBs of RA SFs. (a) Immunohistochemical double-staining of RA and OA synovial tissues with antibodies against DAXX (blue) and PML (red) showed ...

Discussion

SUMO-1 acts as a posttranslational regulator of different signaling pathways and is involved in the formation of PML NBs (12, 13). Overexpression of SUMO-1 has been associated with alterations in apoptosis (7), but the mechanisms that mediate these effects and particularly their relevance for disease conditions are unclear. Based on previous data that have demonstrated increased expression of SUMO-1 in SFs from patients with RA (10), we investigated the relevance of SUMO-1 for the resistance of these fibroblasts to Fas-mediated cell death and studied the mechanisms by which SUMO-1 is involved in the regulation of apoptosis in these diseased cells. We demonstrate that, rather than by interacting directly with the FADD, SUMO-1 inhibits apoptosis through recruiting proapoptotic molecules such as DAXX into PML NBs, where they are trapped and cannot exert their proapoptotic effects.

In our studies, we first investigated whether increased expression of SUMO-1 contributes to the resistance of RA SFs against Fas-induced cell death. Knockdown of SUMO-1 by siRNA and long-term expression of antisense RNA sensitized RA SFs to Fas-mediated apoptosis. By contrast, overexpression of WT SUMO-1 further decreased the ability of RA SFs to undergo cell death through Fas. These data indicated to us that the increased levels of SUMO-1 as found in RA SFs contribute to the resistance of these cells against Fas-induced apoptosis. Because it has been hypothesized that SUMOylation of FADD itself or FADD-associated molecules mediates the antiapoptotic effects of SUMO-1 (7, 14), we analyzed the subcellular distribution of SUMO-1 in RA and OA synovial membranes and in fibroblasts isolated from these tissues but we were unable to locate SUMO-1 to the cell membrane and particularly to FADD even after stimulation with FasL. Rather, both in vivo and in vitro SUMO-1 was found predominantly in the nucleus and showed a pattern of distribution that has been described in a number of other cell types and where a considerable proportion of SUMO-1 was found in association with PML. This result is also in line with previous work of other groups demonstrating that PML is modified by SUMO-1, and that SUMOylation of PML is essential for the recruitment of other proteins to PML NBs (1517). Although recent data suggest that in addition to SUMO-1 other members of the SUMO family such as SUMO-3 are also capable of modifying PML, the role of these SUMOs appears to be distinct in that SUMO-3 is responsible primarily for stabilizing the nuclear localization of PML (18). Of interest, increased levels of SUMO-1 in RA SFs resulted in a more pronounced formation of PML NBs, suggesting that the differences in the expression of SUMO-1 between RA and OA SFs have functional consequences.

Based on these findings, we hypothesized that the association of SUMO-1 with nuclear PML is responsible for the inhibitory effects of SUMO-1 on Fas-mediated apoptosis through altering the composition of PML NBs. This notion was supported further by our observation that in RA SFs higher expression of SUMO-1 than in OA fibroblasts resulted in increased levels of PML-bound DAXX, a transcriptional repressor with proapoptotic properties. In this context the SUMO-specific protease SENP1 is of interest, because it is targeted to the nucleus through a single nonconsensus nuclear localization signal (19), and expression of SENP1 in COS cells removes PML NB conjugates from the nucleus (20). Conversely, generation of dominant negative mutants of SENP1 was shown to result in the accumulation of high-molecular-weight SUMO-1 conjugates in the nuclei of these cells (19). In our studies, overexpression of SENP1 did not alter the rate of spontaneous apoptosis but most significantly increased the susceptibility of RA SFs to Fas-induced cell death. These data prompted us to ask whether overexpression of SENP1 in RA SFs would not only restore the susceptibility of these cells to Fas-induced apoptosis but also alter the composition of PML NBs. Indeed, we were able to show that SENP1 removed SUMO-1 from its nuclear depots but did not decreased nuclear accumulation of PML. Expression of SENP1 resulted in fewer but larger PML aggregates, yet at the same time, de-SUMOylation of PML by SENP1 promoted the loss of DAXX from the PML NBs. These findings demonstrated that SENP1 is targeted to the nucleus of RA SFs and mediates the release of DAXX from the PML NBs. Data from Pml−/− cells have suggested that the ability of DAXX to potentiate Fas-induced apoptosis is linked to the presence of PML (21), and it has been proposed that even when released from PML NBs, DAXX needs to associate with PML to enhance Fas signaling (22). Our data do not necessarily support this hypothesis, despite clear evidence for the interaction of PML and DAXX (23). Rather, our confocal studies failed to demonstrate the presence of PML–DAXX complexes after overexpression of SENP1wt. This failure may be caused by the size and, thus, detectability of such complexes, but it may be hypothesized also that the stability of PML NBs, which in our and other studies was not affected by SENP1 (20), is a general prerequisite for the execution of the apoptotic cascade.

To analyze whether DAXX is involved functionally in the proapoptotic effects of SENP1 or whether these effects are mediated by other proteins that in a similar way as DAXX are released from the PML NBs upon de-SUMOylation of PML, we knocked down the expression of DAXX in RA SFs by siRNA and could demonstrate that SENP1 failed to sensitize RA SFs to Fas-induced apoptosis when DAXX was lacking. These data indicate that by deconjugating SUMO-1 from PML SENP1 can modulate the inhibitory effects of SUMO-1 on Fas-induced apoptosis and that these antiapoptotic effects are mediated through accumulation of DAXX in PML NBs. Our data further demonstrate that the nuclear release of DAXX increases its association with FADD, which may serve as an explanation for the proapoptotic effects of SENP1. The question of whether this is the only mechanism by which DAXX exerts its effects in RA SFs, however, remains to be answered, because it has been shown that DAXX may also function as an adaptor molecule that recruits other proteins to PML NBs (24) and more recent data have demonstrated a role for DAXX as a transcriptional repressor of key transcriptional factors. In this context, the observation that DAXX interacts with Smad4 and represses its transcriptional activity (25) as well as data showing that the interaction of SUMO-1-modified CREB-binding protein (CBP) with DAXX mediates SUMO-dependent transcriptional regulation of CBP (26) are certainly of importance. It also cannot be excluded completely that other proteins, which may get released from PML NBs after de-SUMOylation of PML, are involved at least in part in the effects of SENP1. However, our data most strongly suggest that the altered accumulation of DAXX substantially influences the susceptibility of fibroblast-like cells to Fas-induced apoptosis and that this accumulation of DAXX in PML NBs reduced the association of DAXX with FADD.

Collectively, we have shown that SUMO-1 contributes to the resistance of RA SFs against Fas-induced apoptosis by creating nuclear depots of proteins that under physiological conditions are involved in the dynamic balancing of death signals. The data suggest that, in addition to the altered expression of apoptosis-regulating proteins, increased SUMOylation of PML and subsequent trapping of proapoptotic regulators may be an important pathological factor for the resistance of cells against apoptotic death.

Methods

Cells and Apoptosis.

SFs were obtained from RA and OA patients at joint replacement surgery and grown in DMEM with 10% FCS until passages 4–6 as described (27). For the experiments, they were plated at identical densities (104 cells) into 96-well plates and allowed to adhere overnight. Apoptosis was then induced by incubation in 100 ng/ml rhFasL for 16 h as described (11), and cell death was determined by using a histone fragmentation assay (Cell Death Detection ELISAPlus; Roche, Minneapolis, MN). For confirmation of the data, cells seeded into six-well plates (2.5 × 105 cells per well) were treated accordingly but then harvested and stained with propidium iodide (5 μg/ml) for analysis by FACS.

Expression Vectors, siRNA, and Transfection.

Replication-deficient retroviruses carrying the SUMO-1 expression and antisense constructs were generated by PCR, subcloning the SUMO-1 coding sequence in sense and antisense orientation into the retroviral vector pLXIN (Clontech, Heidelberg, Germany) and transfection into the amphotropic packaging cell line PT67 (Clontech) as described (27). Nonviral, GFP-tagged SENP1wt, and SENP1mt expression constructs were generated by PCR, subcloning SENP1 into pEGFP-C1 (Clontech). SUMO-1 siRNA (5′-CUG GGA AUG GAG GAA GAA G-3′) and DAXX siRNA (5′-GGA UGU UGC AGA ACU CCG C-3′) oligonucleotides with 3′ dTdT overhangs were synthesized by Eurogentec, Brussels. Mismatch oligonucleotides (Eurogentec) were used as controls. Transfection of SENP1wt and SENP1mt as well as siRNA was performed by nucleofection. Fibroblasts were transfected once 12 h (SENP1) or 72 h (SUMO-1 and SENP1 plus siDAXX) before further analysis. For viral gene transfer, RA SFs were transduced at 60–70% confluence in the presence of 8 μg/ml polybrene and selected with 0.8 mg/ml G418.

Immunolocalization of SUMO, PML, and DAXX in RA and OA Synovial Tissues.

Paraffin sections of synovial tissue specimens from RA and OA patients were pretreated with trypsin (PAA, Pasching, Austria), blocked with 15% horse serum, and incubated with primary antibodies against SUMO-1 (Zymed Laboratories, Karlsruhe, Germany) or DAXX (Exbio, Praha, Czech Republic) diluted 1:100 in TBS containing 10% horse serum for 2 h at 37°C and subsequently with anti-PML antibodies (Sigma/Aldrich, St. Louis, MO) diluted 1:200 in Tris-buffered saline containing 10% horse serum overnight at 4°C. Immunostaining of the tissues was performed with secondary biotinylated antibodies using the Vectastain ABC Elite Kit with the Vector Red and Blue Substrate Kit (Vector Laboratories).

Expression of SUMO-1, SENP1, DAXX, and PML in SFs.

Expression of SUMO-1, SENP1 and DAXX mRNA was determined by quantitative PCR (TaqMan, Applied Biosystems, Foster City, CA) using predeveloped assays (PE Biosystems, Foster City, CA) and β-actin as endogenous control (27).

For the detection of SUMO-1 and DAXX protein, cells were lysed in single detergent buffer (50 mM Tris·HCl, pH 8.0/150 mM NaCl/1% Nonidet P-40/0.02% NaN3) with the protease inhibitor mixture SetIII (1:100; Calbiochem) and then separated on a denaturing SDS polyacrylamid gel (12% running, 5% stacking gel). Proteins were transferred onto a PVDF membrane using semidry blot. Membranes were blocked for 1 h with 5% dry defatted milk in PBS including 0.1% Tween 20. Membranes were then incubated with monoclonal anti-SUMO-1 (clone 21C7; Zymed Laboratories) or anti-DAXX (clone DAXX-01; Exbio) antibodies in PBS containing 5% milk and then with HRP-labeled anti-mouse antibodies (Amersham, Piscataway, NJ). Bound HRP was visualized with a ECL kit (Amersham). Blots were stripped and reprobed with anti-β-actin antibodies (Sigma/Aldrich) as loading controls.

For the detection of SENP1 protein, nuclear extracts were generated by using a two-step buffer system. A total of 4 × 106 cells was first lysed in a hypotonic buffer (10 mM Hepes, pH 7.9/10 mM KCl/0.1 mM EDTA/0.1 mM EGTA) and then nuclear proteins were extracted by a hypertonic buffer (20 mM Hepes, pH 7.9/400 mM KCl/1 mM EDTA/1 mM EGTA). Both buffers were supplemented with a protease inhibitor mixture (SetIII, 1:100; Calbiochem). Immunoblotting was performed as described above using polyclonal anti-SENP1 antibodies (Abgent), monoclonal anti-Lamin B1 antibodies as an internal control (clone L-5; Zymed Laboratories), and HRP-labeled secondary antibodies (DAKO).

Subcellular distribution of SUMO-1, SENP1, DAXX, and PML was studied by fluorescence microscopy. Cells were fixed in ice-cold methanol (5 min) and acetone (30 s), permeabilized (0.1% Triton X-100; Sigma/Aldrich), treated with blocking solution (PBS, 2% BSA; Sigma/Aldrich), and incubated for 60 min with rabbit polyclonal anti-SUMO-1 antibody (1:250, BSA; Alexis Biochemicals) and then monoclonal anti-DAXX antibodies (1:250; BD Pharmingen) or first rabbit polyclonal anti-PML antibodies (1:200; Santa Cruz Biotechnology, Santa Cruz, CA) and then monoclonal anti-human DAXX antibodies (1:250, BD Pharmingen). TRITC-conjugated anti-rabbit IgG (1:500; Dianova, Hamburg, Germany) and Cy5-conjugated anti-mouse IgG (1:250, Dianova) were used as corresponding secondary antibodies. Confocal microscopy was done on a DMIRE2 inverted microscope (Leica, Deerfield, IL) equipped with a TCS-SP2 confocal unit.

Analysis of DAXX Sequestration in PML NBs.

For endogenous DAXX-PML coimmunoprecipitation, nuclear extracts from 1 × 107 cells were prepared and incubated at 4°C for 16 h with 4 μg of monoclonal anti-PML antibodies (clone PML-97; Sigma/Aldrich). Lysates were then incubated with 40 μl of Dynabeads Protein G (Dynal Biotech) for 1 h, and antigen-antibody complexes were captured with a magnetic device. PML complexes were recovered by lowering pH using 0.1 M citrate buffer (pH 2.0), separated on a 10% SDS polyacrylamide gel, and electrotransferred onto a PVDF membrane. PML-associated DAXX was detected by monoclonal anti-DAXX antibodies (clone DAXX-01; Exbio) and visualized with a ECL kit (Amersham).

Analysis of DAXX Association with FADD.

For endogenous FADD–DAXX coimmunprecipitation, total cell extract from 2.5 × 106 cells was prepared and incubated at 4°C for 4 h with 5 μg of polyclonal anti-FADD (Upstate Biotechnology, Lake Placid, NY). Lysates were then incubated with 50 μl of Dynabeads Protein G (Dynal Biotech) overnight, and antigen–antibody complexes were captured with a magnetic device. Complexes were washed three times with PBS and then boiled in SDS-sample buffer and loaded on a 10% SDS polyacryamide gel and electrotransferred onto a nitrocellulose membrane. FADD-associated DAXX was detected as described above.

Acknowledgments

We thank S. Weinholz, D. Weber, S. Pietzke, and B. Truckenbrod for technical assistance. This work was supported by Deutsche Forschungsgemeinschaft Grant Pa689-2, Swiss National Science Foundation Grant 3200B0-103691, and the European Community Autocure FP6 program.

Abbreviations

SUMO
small ubiquitin-like modifier
PML
promyelocytic leukemia
NB
nuclear body
RA
rheumatoid arthritis
SF
synovial fibroblast
FADD
Fas-associated death domain
OA
osteoarthritis
rh
recombinant human
SENP1wt
WT SENP1
SENP1mt
mutant SENP1.

Footnotes

The authors declare no conflict of interest.

This article is a PNAS direct submission.

References

1. Firestein GS. Nature. 2003;423:356–361. [PubMed]
2. Pap T, Muller-Ladner U, Gay RE, Gay S. Arthritis Res. 2000;2:361–367. [PMC free article] [PubMed]
3. Buckley CD, Pilling D, Lord JM, Akbar AN, Scheel-Toellner D, Salmon M. Trends Immunol. 2001;22:199–204. [PubMed]
4. Baier A, Meineckel I, Gay S, Pap T. Curr Opin Rheumatol. 2003;15:274–279. [PubMed]
5. Pope RM. Nat Rev Immunol. 2002;2:527–535. [PubMed]
6. Jang MS, Ryu SW, Kim E. Biochem Biophys Res Commun. 2002;295:495–500. [PubMed]
7. Okura T, Gong L, Kamitani T, Wada T, Okura I, Wei CF, Chang HM, Yeh ET. J Immunol. 1996;157:4277–4281. [PubMed]
8. Yang X, Khosravi-Far R, Chang HY, Baltimore D. Cell. 1997;89:1067–1076. [PMC free article] [PubMed]
9. Torii S, Egan DA, Evans RA, Reed JC. EMBO J. 1999;18:6037–6049. [PMC free article] [PubMed]
10. Franz JK, Pap T, Hummel KM, Nawrath M, Aicher WK, Shigeyama Y, Muller-Ladner U, Gay RE, Gay S. Arthritis Rheum. 2000;43:599–607. [PubMed]
11. Peli J, Schroter M, Rudaz C, Hahne M, Meyer C, Reichmann E, Tschopp J. EMBO J. 1999;18:1824–1831. [PMC free article] [PubMed]
12. Muller S, Hoege C, Pyrowolakis G, Jentsch S. Nat Rev Mol Cell Biol. 2001;2:202–210. [PubMed]
13. Hofmann TG, Will H. Cell Death Differ. 2003;10:1290–1299. [PubMed]
14. Ryu SW, Chae SK, Kim E. Biochem Biophys Res Commun. 2000;279:6–10. [PubMed]
15. Zhong S, Muller S, Ronchetti S, Freemont PS, Dejean A, Pandolfi PP. Blood. 2000;95:2748–2752. [PubMed]
16. Lehembre F, Muller S, Pandolfi PP, Dejean A. Oncogene. 2001;20:1–9. [PubMed]
17. Ishov AM, Sotnikov AG, Negorev D, Vladimirova OV, Neff N, Kamitani T, Yeh ET, Strauss JF, 3rd, Maul GG. J Cell Biol. 1999;147:221–234. [PMC free article] [PubMed]
18. Fu C, Ahmed K, Ding H, Ding X, Lan J, Yang Z, Miao Y, Zhu Y, Shi Y, Zhu J, Huang H, Yao X. Oncogene. 2005;24:5401–5413. [PubMed]
19. Bailey D, O'Hare P. J Biol Chem. 2004;279:692–703. [PubMed]
20. Gong L, Millas S, Maul GG, Yeh ET. J Biol Chem. 2000;275:3355–3359. [PubMed]
21. Zhong S, Salomoni P, Pandolfi PP. Nat Cell Biol. 2000;2:E85–E90. [PubMed]
22. Bernardi R, Pandolfi PP. Oncogene. 2003;22:9048–9057. [PubMed]
23. Li H, Leo C, Zhu J, Wu X, O'Neil J, Park EJ, Chen JD. Mol Cell Biol. 2000;20:1784–1796. [PMC free article] [PubMed]
24. Ishov AM, Vladimirova OV, Maul GG. J Virol. 2002;76:7705–7712. [PMC free article] [PubMed]
25. Chang CC, Lin DY, Fang HI, Chen RH, Shih HM. J Biol Chem. 2005;280:10164–10173. [PubMed]
26. Kuo HY, Chang CC, Jeng JC, Hu HM, Lin DY, Maul GG, Kwok RP, Shih HM. Proc Natl Acad Sci USA. 2005;102:16973–16978. [PMC free article] [PubMed]
27. Rutkauskaite E, Zacharias W, Schedel J, Muller-Ladner U, Mawrin C, Seemayer CA, Alexander D, Gay RE, Aicher WK, Michel BA, et al. Arthritis Rheum. 2004;50:1448–1456. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links