Inhibition of Proteasome Function Prevents Thymocyte Apoptosis: Involvement of Ornithine Decarboxylase

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.

250, 293–297 (1998)

RC989291

Inhibition of Proteasome Function Prevents Thymocyte Apoptosis: Involvement of Ornithine Decarboxylase Emanuela Grassilli,* Francesca Benatti,* Paola Dansi,† Anna Maria Giammarioli,‡ Walter Malorni,‡ Claudio Franceschi,*,§ and Maria Alfonsina Desiderio†,1 *Department of Biomedical Sciences, Section of General Pathology, University of Modena; §Italian National Research Centres on Aging, INRCA, Ancona, Italy; †Institute of General Pathology, University of Milan and CNR, Center for Research on Cell Pathology, Milan; and ‡Laboratory of Ultrastructures, Istituto Superiore di Sanita`, Rome—Italy

Received August 4, 1998

We have previously shown that polyamine levels rapidly decrease in thymocytes undergoing apoptosis, and that ornithine decarboxylase increases early but too transiently to maintain elevated polyamine levels. These data led us to suppose that a precocious ornithine decarboxylase degradation might be responsible for the imbalance of polyamine metabolism. Ornithine decarboxylase is known to be degraded by the cytosolic 26S proteasome that plays an essential role in thymocyte apoptosis. In this paper we demonstrate that the inhibition of proteasome function preserves ornithine decarboxylase activity and prevents thymocytes from undergoing apoptosis after dexamethasone treatment. Since intracellular polyamine levels are also preserved, ornithine decarboxylase seems to be functionally active in maintaining polyamine homeostasis after proteasome inhibition in thymocytes. Our proposed role for the proteasome in quiescent cells upon an apoptotic stimulus is to degrade proteins like ornithine decarboxylase that are involved in the control of the cell cycle and cell survival. © 1998 Academic Press

The critical role of polyamines in cell proliferation and differentiation is well recognised (1–3). Only recently it has been shown that polyamines play an important role also in apoptosis. In rat thymocytes undergoing cell death after treatment with dexamethasone (dex), heat shock or g rays, we have shown that polyamine intracellular levels progressively diminish starting before the onset of DNA fragmentation (4, 5). Putrescine and spermidine decreases are less pronounced than that of spermine, which in turn under1 To whom requests for reprints should be addressed. Maria Alfonsina Desiderio, Institute of General Pathology, University of Milan, via L. Mangiagalli, 31, 20133, Milan, Italy. Fax: 139-226681092. E-mail: [email protected].

goes total depletion. Similarly, polyamine levels are shown to decrease during apoptosis of RINm5F cells after oxidative stress (6), of CTLL-2 cells after IL-2 deprivation or inhibition of protein phosphorylation (7), and of human lung and colon carcinoma cells after seleniomethionine treatment (8). Ornithine decarboxylase (ODC), the first and ratelimiting enzyme of polyamine biosynthesis (3), shows a rapid and strong induction after triggering of apoptosis in thymocytes (4, 5). However, ODC enhancement seems to be too transient for maintaining polyamine levels in an up-regulated state and this enzymatic pattern might explain the unbalance of polyamine metabolism (5). The present paper was undertaken to investigate wheter it is possible to preserve ODC activity and to prevent the decrease of intracellular polyamine levels in thymocytes undergoing apoptosis after dex treatment by blocking proteasome function. Antizyme, a unique regulatory protein, forms a complex with ODC and renders the enzyme susceptible to proteolysis by the 26S proteasome in an ATP-dependent but ubiquitin-independent manner (9). The 26S proteolytic complex is widely distributed in eucaryotic cells, and is highly regulated and selective. Proteasome degrades most of the abnormal and short-lived proteins (10), and it is also critical for many important cellular processes such as cell cycle progression and NF-kB activation (11, 12). Very recently proteasome has been shown to play a central role also in apoptosis triggered in primary thymocytes by different stimuli (13), and in sympathetic neurons by nerve growth factor deprivation (14). These authors suggest that the proteasome may act either by activating the apoptotic proteolytic cascade or by degrading regulatory proteins that normally inhibit the apoptotic pathway. Here we report that proteasome inhibition not only prevents thymocyte apoptosis after dex treatment but

293

0006-291X/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.

Vol. 250, No. 2, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

also maintain ODC activity and intracellular polyamines at elevated levels. Our findings suggest that the preservation of ODC activity and, therefore, of polyamine biosynthesis and levels is critical for cell survival probably also through an antiapototic effect. MATERIALS AND METHODS Thymocyte cultures. Suckling rats of the Sprague-Dawley strain were sacrificed by decapitation, and the thymus was rapidly excised. Thymocytes were separated as previously described (15), and resuspended at a concentration of 5 3 106 cells/ml in RPMI 1640 medium supplemented with 10% heat-inactivated amino-oxidase free horse serum (4). Apoptosis was induced by addition of 0.1 mM dex, and cultures were harvested at the time points indicated in the Figures. When proteasome inhibition was carried out, thymocytes were preincubated with 20 mM N-acetyl-L-leucinyl-L-leucinal-L-norleucinal (LLnL; Sigma Aldrich, Milan, Italy) for 1 h after seeding, and then dex was added. LLnL was dissolved in DMSO at 0.008% final concentration in cell culture, that was shown to be non-toxic. DNA recovery and electrophoresis. Crude cell lysates were used to examine genomic integrity as described elsewhere (15). Hoechst staining of apoptotic nuclei. Samples of 0.5 3 106 cells were washed once with phosphate-buffered saline (PBS), and fixed in 4 % paraformaldehyde for 15 min. Fixed cells were then washed twice with PBS and smeared onto poly-lysine precoated slides. After 10 min permeabilization with 0.5 % Triton-X, staining of apoptotic cells was performed by incubation of the slides with Hoechst 33258 dye for 20 min at 37° C. Slides were then briefly soaked in PBS and finally observed with a Nikon Microphot fluorescence microscope. Quantitative evaluation of apoptotic cells was performed by counting at least 300 cells at a 100 3 magnification. Measurement of ODC activity. For ODC activity assay, 25 3 106 cells were sonicated in 200 ml of 20 mM Tris-HCl (pH 7.1) containing 0.25 M sucrose, and were centrifuged at 12,000 rpm for 15 min at 4° C with an Eppendorf microcentrifuge. The supernatants were frozen at 280° C until further use. The enzyme activity was measured in the supernatants as release of labeled CO2 from L-[114 C]ornithine by the method of Ja¨nne and Williams-Ashman (16), with some modifications. The reaction mixture (final volume 250 ml) contained cytosol supernatant (about 200 mg of protein), 50 mM Tris-HCl (pH 7.1), and 0.14 mM L-[1-14C]ornithine (57 Ci/mol) (Amersham Corp., Buchs, U.K.). Blanks contained 4 mM a-difluoromethylornithine (generously given by Marion Merrell Dow Research Institute, Strasbourg, France) to inhibit ODC activity (17). Western blot analysis of antizyme. The cells were solubilized in RIPA buffer by two cycles of freezing and thawing, and were centrifuged at 14,000 rpm for 15 min at 4° C with an Eppendorf microcentrifuge. The protein concentration of the cell extracts was measured by Bio-Rad protein assay kit using bovine serum albumin as standard, and 40 mg of protein were run on 12.5 % SDS-polyacrilamide gel and transferred to a nitrocellulose membrane Hybond ECL (Amersham). Thereafter, the membrane was blocked overnight at 4° C with the blocking reagent (5 % nonfat dry milk in PBS-0.1 % Tween 20). The incubation with rabbit polyclonal antibody for recombinant Z1 antizyme (1:1000) was performed for 1 h at room temperature (18). After washing, the membrane was incubated with goat antirabbit antibody (1:5000) conjugated with horseradish peroxidase, and detection was achieved with ECL chemioluminescence detection system according to manufacturer’s instructions (Amersham). Determination of polyamines by high performance liquid chromatography. Extracts from 107 cell samples were prepared in 250 ml of 0.2 N perchloric acid by ultrasonication, and were centrifuged at 5,000 rpm for 20 min with an Eppendorf microcentrifuge. The analysis of

polyamines in the supernatants was performed by high performance liquid chromatography (19, 20). A C18 reverse-phase Nova-Pack Column (4 mm particle size, 150 3 3.9 mm; Waters, Milford, MA, U.S.A.) was used for the chromatographic separation of the polyamines, which were derivatized post-column with o-phthalaldehyde. Protein content measurement. The protein contents of cell extracts used for ODC activity assay, and of the acid-precipitable fraction of the samples used for high performance liquid chromatography were measured by the method of Lowry (21). Statistical analysis. The data were analyzed with ANOVA, and P , 0.05 was considered significant.

RESULTS AND DISCUSSION Thymocytes are predominantly quiescent cells readily undergoing apoptosis in response to various stimuli, such as glucocorticoids, heat shock and g radiations, but the biochemical events involved in the apoptotic process are still largely unknown. A common mechanism proposed for these models of thymocyte apoptosis consists in the activation of the 26S proteasome (13). In the present work, the experimental model of dextreated thymocytes was used to extend the knowledge about proteasome function by evaluating its possible involvement in polyamine metabolism during apoptosis. ODC is known to be a growth-controlled gene (22), and polyamines synthesized starting from this ratelimiting reaction play an important role in the regulation of cell cycle specifically by promoting S-phase progression (23). In a first series of experiments, we examined the morphological features of thymocytes after dex addition to cultures pretreated with the selective proteasome inhibitor LLnL (13, 14). Peptide aldehydes such as LLnL readily enter cells and suppress 26S proteasome activity. This may explain the antiapoptotic effect of LLnL in thymocytes, since calpain or cathepsin inhibitors tested for comparison do not block cell death under various conditions (13). In Fig. 1, morphological evaluation after Hoechst staining (on the left of each panel) and analysis of DNA fragmentation by gel electrophoresis (on the right of each panel) are shown. The experiments were performed 8 h after dex treatment of thymocytes, i.e. when 40 % of apoptotic cells was observed by cytofluorimetric analysis in agreement with our previous data (4). The preincubation with LLnL prevented the appearence of typical apoptotic features (panel C) usually induced by dex treatment (panel B), i.e. both nuclear condensation and DNA fragmentation. The quantitative evaluation of apoptosis inhibition by LLnL in Hoechst stained cells was performed, and the results show a 65 % inhibition both at 4 and 8 h (Fig. 2). These data confirm the requirement of functional proteasome for apoptosis induction in quiescent cells, such as thymocytes and sympatethic neurons (13, 14). At variance, in proliferating cells proteasome-mediated proteolysis is known to be essential for cell survival

294

Vol. 250, No. 2, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

sis. In the present experiments ODC activity was studied in the presence or the absence of LLnL (Fig. 3). We observed an early induction of ODC activity 1 h after dex treatment with a peak at 2 h (3-fold increase relative to control values), followed by a rapid return to control values at 4 h according to our previous data (4). The induction of ODC activity was concomitant with an increase in antizyme levels, as assessed by Western blot analysis (Fig. 4). The mass of dex-induced band corresponding to antizyme was about 28 kDa (28). Antizyme is known to exert its negative regulatory function on ODC activity by forming an ODC-antizyme complex, that controls ODC degradation via 26S proteasome. The rate of ODC degradation is, therefore, accelerated by increasing the amount of antizyme (9). Due to the very low levels of ODC protein in thymocytes, we were not able to perform a satisfactory Western blot analysis of ODC under our experimental conditions. It cannot be excluded that the increase in antizyme level after dex treatment might also negatively regulate polyamine uptake and might be involved in the decrease of polyamine levels observed in apoptotic thymocytes. It is noteworthy that spermine replenishment prevents thymocyte apoptosis in our conditions (4). Figure 3 also shows that proteasome function blockade by thymocyte incubation with LLnL before dex treatment, prevented ODC activity down-regulation throughout the entire observation period. It is worth noting that in LLnL/dex-treated cells at 8 h, i.e. when cell death was reduced from 75% to 24%, ODC activity

FIG. 1. Morphological features of thymocytes 8 h after dex treatment in the presence or the absence of LLnL. On the left of each panel: morphological analysis by fluorescence microscopy of Hoechststained cells. On the right of each panel: agarose gel electrophoresis of DNA. Thymocytes were either untreated (A), or were treated with dex alone (B) or with dex in the presence of LLnL (C). The arrowheads indicate the apoptotic cells.

and cell cycle progression, since proteasome inhibition induces apoptosis (24 –27). Thus it appears that proteasome performs opposite functions when cells are in diverse growth conditions, even if a difference in proteasome targets depending on the growth status and/or the cell type cannot be excluded. Taking into account the essential role of ODC in the cell cycle and cell survival through the regulation of polyamine metabolism, as well as its proteasomemediated degradation, as a turn-over control mechanism, ODC may be considered a good candidate as key protein that the cell must get rid of to undergo apopto-

FIG. 2. Quantitative evaluation of apoptotic cells after Hoechst staining. The percentage values of apoptotic nuclei in control cells (h) and in cells treated with dex (■) or LLnL 1 dex (u) were obtained by averaging the results of 3 independent experiments in which at least 300 cells per field were counted. The data are reported as the means 6 S.E.

295

Vol. 250, No. 2, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 3. ODC activity in thymocytes treated with dex in the presence or in the absence of LLnL. ODC activity was measured as 14 CO2 release, using labeled ornithine as substrate. (h), control; (■), dex; (u), LLnL 1 dex. The data are the means 6 S.E. of 5 separate experiments. * P , 0.05 vs control cells; ** P , 0.01 vs control cells; ° P , 0.05 vs dex-treated cells; °° P , 0.01 vs dex-treated cells.

showed values 8-fold higher than those of dex-treated cells. Proteasome function inhibition in dex-treated cells reproduced, therefore, the situation previously observed in growing thymocytes, that show a longlasting and sustained ODC activity (4). The present data indicate for the first time that ODC might be degraded by proteasome after triggering of the apoptotic process, suggesting a possible mechanism for thymocyte apoptosis. Consistent with the pattern of ODC activity, polyamine levels were more elevated in LLnL/dex-treated cells than in cells treated with dex alone (Fig. 5). We found that intracellular polyamine pool was significantly preserved 8 and 12 h after the apoptotic stimu-

FIG. 4. Western blot analysis of antizyme in cell lysates. Thymocytes were treated with dex and harvested at indicated times. Equal amount of protein (40 mg) were fractionated by SDS/ polyacrilamide gel electrophoresis and blotted to nitrocellulose membrane. Blot was hybridized with rabbit polyclonal antibody for recombinant Z1 antizyme. The arrow indicates the band corresponding to antizyme.

FIG. 5. Polyamine levels in thymocytes treated with dex in the presence or in the absence of LLnL. Cells were harvested at various times after the treatments, and perchloric acid cell extracts were prepared. Polyamines were measured by high performance liquid chromatography. (■), dex; (u), LLnL 1 dex. The results are reported as percentage of control values (cosidered as 100%) at each time examined. The data represent the means 6 S.E. of 6 separate experiments. The statistical significance was calculated by using the absolute values of cellular polyamine levels after treatment with LLnL 1 dex relative to those after dex alone. For putrescine at 8 h, P , 0.05; for putrescine and spermine at 12 h, P , 0.01.

lus. At that times, putrescine levels were 30% and 100% higher in LLnL/dex-treated than in dex-treated cells. At 12 h also spermine and spermidine levels increased by 65% and 20%, respectively, in thymocytes treated with dex in the presence of LLnL relative to the dex-treated cells alone. In agreement with our previous data, intracellular spermidine pool seems to be the less affected by dex treatment (4). Two main conclusions can be drown from this work. First, the proteasome plays a dual role in the apoptotic process(es). Previous data from Osborne’s lab and Martinou’s group showed that the proteasome action is upstream of caspase cascade, since its inhibition pre-

296

Vol. 250, No. 2, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

vents interleukin-1 converting enzyme (ICE) processing and poly(ADP-ribose)polymerase (PARP) cleavage. Therefore, their proposed role for the proteasome in apoptosis is the activation of the proteolytic cascade. An alternative possibility is that the proteasome pathway degrades an inhibitor of apoptosis normally present to keep the cell death pathway in check (13, 14). Now we make a further hypothesis. The proteasome might exert its function in apotosis also by degrading proteins necessary for cell survival, and we propose ODC as one of these target proteins. Further studies are in progress to give a more direct proof of this hypothesis. Second, our results suggest that the antiapoptotic role of ODC is related to the control of polyamine metabolism and the homeostasis of intracellular polyamines. This is not surprising since polyamines are involved in many essential cellular functions, such as stabilization of nucleic acid structures, control of macromolecular synthesis, regulation of gene expression and nuclear enzymatic activities (29 –33). A variety of different factors have been proposed to influence the decision of the cell to undertake the apoptotic program upon activating signal(s). Among the others, the ratio of Bcl-2 family members, the redox balance, the p53 expression and function (34). On the basis of our findings, we suggest that polyamine metabolic pathway might be included in the group of the cellular “sensors” whose unbalance probably create a permissive condition for the cell to progress along the execution phase of apoptosis. ACKNOWLEDGMENTS This work was supported by grants of Ministero Universita` e Ricerca Scientifica e Tecnologica (MURST) to M.A.D. and C.F., and by grants from Associazione Italiana per la Ricerca sul Cancro (AIRC) to C.F. We thank Drs. S. Hayashi and S. Matsufuji for the generous gift of antibody to antizyme.

REFERENCES 1. Ja¨nne, J., Po¨so¨, H., and Raina, A. (1978) Biochim. Biophys. Acta 473, 241–293. 2. Heby, O. (1981) Differentiation 19, 1–20. 3. Pegg, A. E. (1988) Cancer Res. 49, 759 –774. 4. Desiderio, M. A., Grassilli, E., Bellesia, E., Salomoni, P., and Franceschi, C. (1995) Cell Growth Differ. 6, 505–513. 5. Grassilli, E., Desiderio, M. A., Bellesia, E., Salomoni, P., Benatti, F., and Franceschi, C. (1995) Biochem. Biophys. Res. Commun. 216, 708 –714. 6. Dypbukt, J. M., Ankarcrona, M., Burkitt, M., Sjo¨holm, A., Stro¨m, K., Orrenius, S., and Nicotera, P. (1994) J. Biol. Chem. 269, 30553–30560. 7. Min, A., Hasuma, T., Yano, Y., Matsui-Yuasa, I., and Otani, S. (1995) J. Cell. Physiol. 165, 615– 623. 8. Redman, C., Xu, M. J., Peng, Y.-M., Scott, J. A., Payne, C., Clark, L. C., and Nelson, M. A. (1997) Carcinogenesis 18, 1195–1202.

9. Hayashi, S., Murakami, Y., and Matsufuji, S. (1996) Trends Biochem. Sci. 21, 27–30. 10. Goldberg, A. L., Ross, L., and Adams, J. (1995) Chem. Biol. 2, 503–508. 11. Pagano, M., Tam, S. W., Theodoras, A. M., Beer-Romero, P., Del Sal, G., Chau, V., Yew, P. R., Draetta, G. F., and Rolfe, M. (1995) Science 269, 682– 685. 12. Palombella, V. J., Rando, O. J., Goldberg, A. L., and Maniatis, T. (1994) Cell 78, 773–785. 13. Grimm, L. M., Goldberg, A. L., Poirier, G. G., Schwartz, L. M., and Osborne, B. A. (1996) EMBO J. 15, 3835–3844. 14. Sadoul, R., Fernandez, P.-A., Quiquerez, A.-L., Martinou, I., Maki, M., Schroter, M., Becherer, J. D., Irmler, M., Tschopp, J., and Martinou, J.-C. (1996) EMBO J. 15, 3845–3852. 15. Grassilli, E., Carcereri de’Prati, A., Monti, D., Troiano, L., Menegazzi, M., Barbieri, D., Franceschi, C., and Suzuki, H. (1992) Biochem. Biophys. Res. Commun. 188, 1261–1266. 16. Ja¨nne, J., and Williams-Ashman, H. G. (1971) J. Biol. Chem. 246, 1725–1732. 17. Desiderio, M. A., Mattei, S., Biondi, G., and Colombo, M. P. (1993) Biochem. J. 293, 475– 479. 18. Matsufuji, S., Miyazaki, Y., Kanamoto, R., Kameji, T., Murakami, Y., Baby, T. G., Fujita, K., Ohno, T., and Haiashi, S. (1990) J. Biochem. 108, 365–371. 19. Lo¨ser, C., Wunderlich, U., and Fo¨lsch, U. (1988) J. Chromatogr. 430, 249 –262. 20. Desiderio, M. A. (1992) Hepatology 15, 928 –933. 21. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, L. J. (1951) J. Biol. Chem. 193, 265–275. 22. Stimac, E., and Morris, D. R. (1987) J. Cell. Physiol. 133, 590 – 594. 23. Laitinen, J., Stenius, K., Eloranta, T. O., and Ho¨ltta¨, E. (1998) J. Cell. Biochem. 68, 200 –212. 24. Imajoh-Ohmi, S., Kawaguchi, T., Sugiyama, S., Tanaka, K., Omura, S., and Kikuchi, H. (1995) Biochem. Biophys. Res. Commun. 217, 1070 –1077. 25. Fujta, E., Musaka, T., Tsukahara, T., Arahata, K., Omura, S., and Momoi, T. (1996) Biochem. Biophys. Res. Commun. 224, 74 –79. 26. Shinohara, K., Tomioka, M., Nakano, H., Tone, S., Ito, H., and Kawashima, S. (1996) Biochem. J. 317, 385–388. 27. Drexler, H. C. A. (1997) Proc. Natl. Acad. Sci. USA 94, 855– 860. 28. Mitchell, J. L. A., Choe, C.-y., Judd, G. G., Daghfall, D. J., Kurzeja, R. J., and Leyser, A. (1996) Biochem. J. 317, 811– 816. 29. Celano, P., Baylin, S. B., and Casero, A. C., Jr. (1989) J. Biol. Chem. 264, 8922– 8927. 30. Basu, H. S., Feuerstein, B. G., and Marton, L. J. (1991) in Polyamines in the Gastrointestinal Tract (Dowling, R. H., Fo¨lsch, U. R., and Lo¨ser, C., Eds.), pp. 35– 47, Kluwer Academic Publishers, Dordrecht. 31. Basu, H. S., Sturkenboom, M. C. J. M., Delcros, J.-G., Csokan, P. P., Szollosi, J., Feuerstein, B. G., and Marton, L. J. (1992) Biochem. J. 282, 723–727. 32. Auvinen, M., Paasinen, A., Andersson, L. C., and Ho¨ltta¨, E. (1992) Nature 360, 355–358. 33. Desiderio, M. A., Tacchini, L., Anzon, E., Pogliaghi, G., Radice, L., and Bernelli-Zazzera, A. (1996) Hepatology 24, 150 –156. 34. Green, D. R., and Martin, S. J. (1995) Curr. Opin. Immunol. 7, 694 –703.

297