(Received for publication, December 19, 1996, and in revised form, March 21, 1997)
From the Dana-Farber Cancer Institute and the Department of Pathology, Harvard Medical School, Boston, Massachusetts 02115
Proteolysis by the ubiquitin/proteasome pathway controls the intracellular levels of a number of proteins that regulate cell proliferation and cell cycle progression. To determine whether this pathway of protein turnover was also linked to apoptosis, we treated Rat-1 and PC12 cells with specific proteasome inhibitors. The peptide aldehydes PSI and MG115, which specifically inhibit the chymotrypsin-like activity of the proteasome, induced apoptosis of both cell types. In contrast, apoptosis was not induced by inhibitors of lysosomal proteases or by an alcohol analog of PSI. The tumor suppressor p53 rapidly accumulated in cells treated with proteasome inhibitors, as did the p53-inducible gene products p21 and Mdm-2. In addition, apoptosis induced by proteasome inhibitors was inhibited by expression of dominant-negative p53, whereas overexpression of wild-type p53 was sufficient to induce apoptosis of Rat-1 cells in transient transfection assays. Although other molecules may also be involved, these results suggest that stabilization and accumulation of p53 plays a key role in apoptosis induced by proteasome inhibitors.
Many types of mammalian cells undergo apoptosis during normal
development or in response to a variety of stimuli, including DNA
damage, growth factor deprivation, and abnormal expression of oncogenes
or tumor suppressor genes (1-3). Apoptosis induced by these various
agents appears to be mediated by a common set of downstream elements
that act as regulators and effectors of apoptotic cell death. In many
cases, apoptosis requires the p53 tumor suppressor protein (4).
Overexpression of p53 is not only induced by several cell death stimuli
but is itself sufficient to induce apoptosis in gene transfer assays
(5). Apoptosis is also regulated by members of the Bcl-2 family (2, 6), which act upstream of a family of cysteine proteases known as the
interleukin-1 converting enzyme (ICE)1
protease family (3, 7). Members of the ICE family, which are activated
by proteolytic cleavage of proenzymes, appear to act as common
executioners of apoptosis induced by a variety of cell death
stimuli.
The ubiquitin/proteasome pathway is a major pathway of proteolysis in eukaryotic cells and may contribute to controlling the intracellular levels of a variety of short-lived proteins (8-10). The proteasome is a large multicatalytic complex that catalyzes the degradation of ubiquitinated cellular proteins. Substrates of the ubiquitin/proteasome pathway include a number of cell regulatory molecules, such as cyclins, the Myc oncogene protein, and p53, and the regulated degradation of these molecules has been linked to the control of cell proliferation and cell cycle progression (8-10).
By controlling the intracellular levels of such proteins, the activity of the ubiquitin/proteasome pathway might also be linked to apoptosis. In the present study, we have tested this possibility and demonstrate that inhibition of the chymotrypsin-related activity of the proteasome induces apoptosis of proliferating Rat-1 cells and of PC12 cells. This induction of apoptosis is associated with increased intracellular levels of p53 and is blocked by dominant negative p53 mutants, suggesting that it results from inhibition of p53 degradation.
Rat-1 cells were cultured in Dulbecco's
modified Eagle's medium supplemented with 10% calf serum. PC12 cells
were maintained in Dulbecco's modified Eagle's medium supplemented
with 10% fetal bovine serum and 5% horse serum. Cells overexpressing
Bcl-2 and CrmA were kindly provided by J. Yuan (Massachusetts General
Hospital). Cells overexpressing dominant-negative p53 were obtained by
transfection with plasmids expressing the p53 mutant Val143
Ala from a cytomegalovirus promoter (11).
N-Benzyloxycarbonyl-Ile-Glu(O-t-butyl)-Ala-leucinal (PSI) and carbobenzoxyl-leucinyl-leucinyl-norvalinal-II (MG115) were purchased from the Peptide Institute Inc. and diluted in Me2SO. The alcohol analog of PSI, N-benzyloxycarbonyl-Ile-Glu(O-t-butyl)-Ala-leucinol, was kindly supplied by S. Wilk (Mount Sinai School of Medicine, New York, NY). Calpain inhibitor II was purchased from Boehringer Mannheim. The peptidyl aldehydes were added to cultures, whereas control plates were treated with the diluent. Treated cells were collected by centrifugation, and cytoplasmic DNA was isolated, electrophoresed in 1.8% agarose gels, and visualized by ethidium bromide staining (12). Cytological characterization of apoptotic cells was performed by staining with propidium iodide and TUNEL assay (Boehringer Mannheim).
Immunoblot AnalysisCell lysates (20 µg) were electrophoresed in SDS-polyacrylamide gels and transferred to nitrocellulose membranes. Primary monoclonal antibodies against p53, p21, and Mdm2 were purchased from Santa Cruz Biotechnology Inc. The secondary antibody was horseradish peroxidase-conjugated anti-mouse IgG and proteins were detected using the ECL system (Amersham Corp.).
Induction of Apoptosis by p53Rat-1 cells growing on
coverslips were transfected with 5 µg of human wild-type or
Val143 Ala mutant p53 plasmids (11) using the
liptofectamine method. 40 h after transfection, cells were fixed
with 2% paraformaldehyde for 10 min and permeabilized with 0.5%
Triton X-100 for 5 min. Blocking and incubation with antibodies were
carried out in phosphate-buffered saline containing 1% bovine serum
albumin and 5% goat serum. Anti-p53 DO-1 primary antibody (Santa Cruz
Biotechnology) and fluorescein isothiocyanate-conjugated secondary
antibody (Sigma) were applied for 1 h in 1:1000 and 1:500
dilutions, respectively. Cells were simultaneously stained with
propidium iodide and then examined with a fluorescence microscope.
The
proteasome contains three distinct proteolytic activities: a
chymotrypsin-related activity, a trypsin-related activity, and a
peptidylglutamyl-peptide hydrolyzing activity (13). The chymotrypsin-related activity is specifically inhibited by the peptide
aldehydes PSI and MG115 (14), so we initially tested the possibility
that these inhibitors would induce apoptosis. Actively proliferating
Rat-1 fibroblasts and PC12 pheochromocytoma cells were treated with PSI
and MG115 at concentrations selected based on the IC50
described for inhibition of proteasome degradation of IB-
and
processing of major histocompatability complex class I molecules
(15-18). After 4 h of treatment, apoptosis was assayed by DNA
fragmentation (Fig. 1).
Both Rat-1 and PC12 cells underwent apoptosis following treatment with PSI at concentrations of 15 or 25 µM (Fig. 1A). Apoptosis was similarly induced by treatment of Rat-1 cells with 30 µM MG115 (Fig. 1B, lane 4). These data were confirmed by the TUNEL assay, which provides an in situ assay for DNA fragmentation. In a typical experiment, 10-15% of cells were TUNEL positive following 5 h of treatment with PSI compared with 1-2% of untreated control cells (data not shown).
In contrast to the activity of PSI and MG115, apoptosis was not induced by treatment with NH4Cl, which nonspecifically inhibits lysosomal proteolysis, or with a specific inhibitor of calpain II (Fig. 1B, lanes 5 and 6). Thus, inhibition of lysosomal proteases did not lead to apoptosis. As a further specificity control, cells were treated with an alcohol analog of PSI, which only weakly inhibits proteasome function (14). The alcohol analog failed to induce apoptosis of Rat-1 cells (Fig. 1, B, lane 3, and C, lane 2), indicating that apoptosis induced by PSI and MG115 resulted from specific inhibition of the proteasome.
Apoptosis induced by a variety of agents can be inhibited by overexpression of Bcl-2 or by inhibition of ICE family proteases. We therefore tested the effects of Bcl-2 and CrmA, a cowpox virus-encoded ICE inhibitor, on apoptosis induced by proteasome inhibitors. Transfected Rat-1 cells overexpressing either Bcl-2 or CrmA were resistant to treatment with PSI and MG115 (Fig. 1C), indicating that apoptosis induced by these proteasome inhibitors was regulated by Bcl-2 and mediated by ICE family proteases.
Induction of apoptosis by proteasome inhibitors contrasted with recent results, indicating that inhibition of the proteasome prevents apoptosis of thymocytes and sympathetic neurons (24, 25). It appeared possible that this difference was due to the fact that the cells used in the present study were actively proliferating, in contrast to terminally differentiated thymocytes and neurons. We therefore tested the effects of proteasome inhibitors on nonproliferating Rat-1 cells, which had been rendered quiescent by serum deprivation, and on PC12 cells, which had been induced to differentiate by treatment with nerve growth factor. Treatment with MG115 failed to induce apoptosis of quiescent Rat-1 cells but was still able to induce apoptosis of nonproliferating differentiated PC12 cells (Fig. 1D). Thus, cell proliferation was not the sole determinant of cellular sensitivity to proteasome inhibitors.
Accumulation of p53 in Cells Treated with PSI and MG115p53 is required for apoptosis in a number of models and stabilization of p53 leads cells to undergo apoptosis (4, 5). Degradation of p53 occurs through the ubiquitin/proteasome pathway (19, 20). If this involved the chymotrypsin-related function of the proteasome, PSI and MG115 might induce apoptosis by inhibiting p53 degradation.
We therefore tested the levels of p53 in Rat-1 and PC12 cells following
treatment with proteasome inhibitors (Fig. 2). In both
cell types, PSI and MG115 lead to the accumulation of p53, which could
be detected as early as 2 h after initiation of treatment (Fig.
2A, lane 2). In contrast, treatment with the
lysosomal protease inhibitor calpain II did not result in p53
accumulation (Fig. 2A, lanes 6 and 7),
nor did treatment with the alcohol analog of PSI (Fig. 2B,
lanes 2, 3, 7, and 8). Most
of the p53 accumulated following treatment with the proteasome
inhibitors migrated at its normal electrophoretic mobility,
corresponding to 53 kDa, and therefore appeared to be nonubiquitinated.
However, higher molecular mass bands that reacted with anti-p53
monoclonal antibody were also observed after treatment with proteasome
inhibitors (for example, Fig. 2A, lanes 2-4,
8, and 9). These bands might correspond to
accumulation of ubiquinated p53 complexes.
p53 is a transcriptional activator that stimulates expression of
several genes, including the cyclin-dependent kinase
inhibitor p21 and the p53 negative regulator Mdm2 (21). To determine
whether the p53 accumulated in response to treatment with proteasome
inhibitors was functional, we assayed expression of these genes in
cells treated with MG115. Treatment with MG115-induced expression of both p21 (Fig. 3A) and Mdm-2 (Fig.
3B). Inhibition of the chymotrypsin-related function of the
proteasome thus led not only to stabilization of p53 but also to
transactivation of at least two p53 target genes.
Apoptosis Induced by Proteasome Inhibitors Is Inhibited by Dominant-Negative p53
To determine whether p53 was functionally
required for apoptosis induced by proteasome inhibitors, we transfected
Rat-1 and PC12 cells with a plasmid expressing a dominant-negative p53
mutant (Val143 Ala), which is defective in DNA binding
and transactivation (11). Transfected cell lines were screened by
immunoblotting to identify cells that expressed high levels of the
mutant p53 (data not shown). These cells were then tested for apoptosis
following treatment with MG115. Both PC12 and Rat-1 subclones
expressing the mutant p53 (143.32 and 143.3 cells, respectively) were
resistant to induction of apoptosis (Fig. 4), indicating
that p53 is required for apoptosis induced by inhibition of proteasome
function.
Overexpression of Wild-type p53 Induces Apoptosis of Rat-1 Cells
To determine whether accumulation of p53 was sufficient to
induce apoptosis, Rat-1 cells were transiently transfected with wild-type or mutant human p53 expression plasmids. Cells expressing human p53 were identified by immunofluorescence and nuclear morphology was assessed by staining with propidium iodide. Transfected cells expressing wild-type p53, but not the Val143 Ala
mutant, displayed the fragmented nuclei typical of apoptotic cells
(Fig. 5). Overexpression of wild-type p53 is thus
sufficient to induce apoptosis of proliferating Rat-1 cells, consistent
with the involvement of p53 accumulation in apoptosis induced by
proteasome inhibitors.
The results of the present study demonstrate that treatment of proliferating Rat-1 cells and of PC12 cells with inhibitors of the chymotrypsin-like function of the proteasome induces apoptosis. A number of previous investigations have demonstrated that the peptidyl aldehydes employed in this study are specific inhibitors of the proteasome chymotrypsin-related activity (14-18). In contrast, neither inhibitors of lysosomal proteases nor a peptide alcohol analog that is ineffective as a proteasome inhibitor was able to induce apoptosis. It thus appears that inhibition of this function of the proteasome specifically induces apoptosis, linking the ubiquitin/proteasome pathway of protein degradation to regulation of apoptotic cell death.
Induction of apoptosis by proteasome inhibitors implies that cell death is induced as a result of increased intracellular concentrations of a regulatory molecule(s) normally degraded by the ubiquitin/proteasome pathway. A prime candidate for such a regulator of apoptosis appeared to be p53, which is known to be ubiquitinated and degraded by the proteasome (19, 20). Not only is p53 induced by a variety of apoptotic stimuli, but overexpression of p53 has been demonstrate to induce apoptosis in a variety of cell types (4, 5). Consistent with this hypothesis, we found that treatment with proteasome inhibitors resulted in stabilization and rapid accumulation of p53 in both Rat-1 and PC12 cells. Moreover, the accumulated p53 was shown to be biologically active, based on transactivation of the p53 target genes encoding p21 and Mdm2. Finally, we have shown that apoptosis induced by proteasome inhibitors is blocked by expression of dominant-negative p53, and that overexpression of wild-type p53 is sufficient to induce apoptosis of Rat-1 cells in transient transfection assays. Although other molecules may also be involved in cell death, these data suggest that modulation of p53 turnover is a key event in apoptosis induced by proteasome inhibitors.
Our finding that apoptosis is induced by inhibitors of the chymotrypsin-like activity of the proteasome is consistent with a previous report showing that lactacystin, which inhibits all three proteasome activities (22), induces apoptosis of U937 myeloid leukemia cells (23). On the other hand, it has recently been reported that inhibition of the proteasome by either lactacystin or peptide aldehydes prevents apoptosis of thymocytes induced by radiation, glucocorticoids, or phorbol esters and of sympathetic neurons induced by nerve growth factor deprivation (24, 25). In these systems, proteasome activity appears to be required for induction of apoptosis upstream of the ICE family proteases. It appeared possible that this apparent discrepancy was due to the fact that U937 cells and the cells used in the present study were actively proliferating, in contrast to the nonproliferating thymocytes and sympathetic neurons in which proteasome activity was required for induction of apoptosis by other agents. Indeed, Grimm et al. (24) noted that prolonged exposure to proteasome inhibitors actually increased cell death in thymocytes that were not treated with other inducers of apoptosis. Consistent with this possibility, we found that nonproliferating Rat-1 cells were no longer susceptible to apoptosis induced by proteasome inhibitors. However, inhibition of the proteasome still induced apoptosis of nonproliferating PC12 cells that had been induced to differentiate by treatment with nerve growth factor. These results indicate that cell proliferation is one but not the only determinant of cellular response to inhibition of the proteasome. It thus appears that the proteasome can function to promote either cell survival or cell death, depending on both proliferative state and other factors that may be cell type-specific.