Induction of G 1 Arrest and Apoptosis in Human Jurkat T Cells by Pentagalloylglucose through Inhibiting Proteasome Activity and Elevating p27 Kip1 , p21 , and Bax Proteins*

Pentagalloylglucose, which is found in many medicinal plants, can arrest the cell cycle at G 1 phase through down-regulation of cyclin-dependent kinases 2 and 4 and up-regulation of the cyclin-dependent kinase inhibitors p27 Kip1 and p21 Cip1/WAF1 in human breast cancer cells. Pentagalloylglucose also induces apoptosis in human leukemic cells. However, the mechanisms by which pentagalloylglucose induces these effects is unclear. We now show that pentagalloylglucose inhibits the activities of purified 20 and 26 S proteasomes in vitro , the 26 S proteasome in Jurkat T cell lysates, and chymotrypsin-like activity of the 26 S proteasome in intact Jurkat T cells. The turnover of p27 Kip1 and p21 Cip1/WAF1 , which is necessary for cell cycle progression mediated by proteasome degradation, was disrupted by treatment of human Jurkat T cells with pentagalloylglucose. This was shown by cycloheximide treatment and in vivo pulse-chase labeling experiments, and this effect correlated with the arrest of proliferation of Jurkat T cells at G 1 . Inhibition of the proteasome by pentagalloylglucose and by the proteasome inhibitor MG132 caused accumulation of ubiquitin-tagged proteins in Jurkat T cells. The addition of pentagalloylglucose

In eukaryotic cells, there are two distinct proteolytic mechanisms essential for regulating levels of cellular proteins. One is lysosomal mediated degradation, which is involved in the breakdown of intracellular proteins under conditions of cellular stress, membrane-associated proteins, and proteins that have entered the cell by endocytosis. The ubiquitin-proteasome pathway is the other mechanism responsible for the turnover of intracellular proteins. This second system works by marking specific substrates with ubiquitins and degrading the marked proteins by the 26 S proteasome in an ATP-dependent manner.
Inhibition of proteasome-mediated degradation of proteins can interfere with the ordered and temporal degradation of cyclindependent kinase inhibitors (CKIs) p27 Kip1 , p21 Cip1/WAF1 , and p19 INK4d (8) and retard cell cycle progression in tumor cells. The key regulators of cell survival and apoptosis, including members of the Bcl-2 family, some caspases, and inhibitor of apoptosis proteins, have all been recognized as substrates of the proteasome (8). Inhibition of proteasome activity can induce apoptosis in cancer cells (17). Therefore, the proteasome acts as a regulator of cell growth and apoptosis and appears to be a possible target for anti-cancer therapy (8, 18 -20).
Recent reports indicate that ester bond-containing phytopolyphenols, such as (Ϫ)-epigallocatechin-3-gallate (EGCG) and tannic acids, potently and specifically inhibit the chymotrypsin-like activity of proteasomes in vitro and in vivo (36,37). The inhibition of chymotrypsin-like proteasome activity resulted in cell cycle arrest at G 1 phase by elevating the levels of p27 Kip1 , a natural protein substrate of the 26 S proteasome. Because 5GG consists of five gallic acids, which are esterified to a glucose core, we presumed that the mechanism by which 5GG induces cell cycle arrest at G 1 in the human breast cancer cell line MCF7 (32) involves not only the up-regulation of the expression of p27 Kip1 and p21 Cip1/WAF1 but also the down-regulation of the activity of the 26 S proteasome, leading to increasing protein stabilities of p27 Kip1 and p21 Cip1/WAF1 .
In the current study, we demonstrated that 5GG can selectively and specifically inhibit chymotrypsin-like activity of the 20 and 26 S proteasome in vitro and in vivo. Furthermore, this inhibition results in cell cycle arrest at G 1 phase and apoptosis, accompanied by the accumulation of ubiquitin-tagged proteins. It also results in the induction of proteasome-related proteins such as p27 Kip1 , p21 Cip1/WAF1 , and the proapoptotic protein Bax in human Jurkat T cells.
Cell Culture-Human Jurkat T cells obtained from American Type Culture Collection were maintained in RPMI 1640, supplemented with 10% fetal calf serum (Invitrogen), 100 units/ml penicillin, 100 g/ml streptomycin, 2 mM L-glutamine (Invitrogen) and kept at 37°C in a humidified atmosphere of 5% CO 2 in air.
Assay for Proteasome Activity-For measurement of the activities of the purified 20 and 26 S proteasomes (36,37,39), 2 or 1 g of purified 20 S proteasome (from M. thermophila or rabbit, respectively) or 6 g of purified 26 S proteasome (from rabbit) was incubated with 20 M fluorogenic peptide substrates, Suc-Leu-Leu-Val-Tyr-AMC (for chymotrypsin-like activity) or Z-Leu-Leu-Glu-AMC (for PGPH activity) in 200 l of assay buffer (20 mM Tris-HCl, pH 8.0) with or without 5GG for 30 min at 37°C. After incubation, the reaction mixture was diluted to 400 l with the assay buffer followed by measurement of hydrolyzed AMC groups (using a fluorescence spectrophotometer with excitation at 380 nm and emission at 460 nm).
For measurement of the proteasome activity of whole cell extracts, 6 g of Jurkat whole cell extract was incubated with 20 M fluorogenic peptide substrates for chymotrypsin-like (Suc-Leu-Leu-Val-Tyr-AMC) or trypsin-like (Z-Gly-Gly-Arg-AMC) or PGPH (Z-Leu-Leu-Glu-AMC) proteasome activities in 200 l of the assay buffer with or without 5GG or MG132 or lactacystin for 90 min at 37°C. After the reactions, the assay buffer was diluted to 400 l, and the release of AMCs was quantified as described earlier.
For proteasome activity in intact Jurkat T Cells, Jurkat T cells (2 ϫ 10 5 cells/ml) in 24-well tissue culture plates were incubated with various concentrations of 5GG or MG132 or lactacystin for 12 h and then mixed with fluorogenic peptide substrate Suc-Leu-Leu-Val-Tyr-AMC for chymotrypsin-like proteasome activity followed by an additional 2-h incubation. Afterward, 200 l of cell medium was harvested and used for the measurement of free AMCs as described earlier.
Assay for Calpain I Activity-The assay was performed as described previously (36,37). Briefly, 6 g of purified calpain I was incubated with 40 M fluorogenic peptide calpain substrate, Suc-Leu-Tyr-AMC, for 30 min at 37°C in 200 l of assay buffer containing 50 mM Tris-HCl (pH 7.5), 50 mM NaCl, 1 mM EDTA, 1 mM EGTA, 5 mM ␤-mercaptoethanol, 5 mM CaCl 2 , and 0.1% CHAPS with or without phytopolyphenols or the specific calpain inhibitor calpeptin (40). After incubation, the reaction mixture was diluted to 400 l with the assay buffer, and a fluorescence spectrophotometer was used to detect free AMCs as described earlier.
Western Blot Analysis-Human Jurkat T cells were treated with various concentrations of 5GG for 12 h or the indicated times. To rule out the effect of 5GG on protein synthesis, Jurkat T cells were pretreated with 10 g/ml cycloheximide for 2 h followed by coincubation with 5GG. Cells were harvested and homogenized in a lysis buffer (50 mM Tris-HCl, pH 8.0, 5 mM EDTA, 150 mM NaCl, 0.5% Nonidet P-40, 0.5 mM phenylmethylsulfonyl fluoride, and 0.5 mM dithiothreitol) for 30 min at 4°C. Equal amounts of total cellular proteins (50 g) were resolved by SDS-PAGE (8% for Rb and pRb; 10% for ␤-actin, Akt, and Ser(P) 473 -Akt; 12.5% for p27, Thr(P) 187 -p27, and ubiquitin-tagged proteins; 15% for p21, Bax, and cytochrome c), transferred onto polyvinylidene difluoride membranes (Amersham Biosciences), and then probed with primary antibody followed by secondary antibody conjugated with horseradish peroxidase. The immunocomplexes were visualized with enhanced chemiluminescence kits (Amersham Biosciences).
Metabolic Labeling and Immunoprecipitation-Jurkat T cells were methionine-starved for 20 min, pulse-labeled with 75 Ci/ml [ 35 S]methionine (Amersham Biosciences; 1,000 Ci/mmol) in methionine-free medium for 30 min, washed with PBS, and chased in complete medium containing 2 mM unlabeled methionine for 0, 0.5, 1, 3, and 6 h. 5GG (10 M) was added at the beginning of the chase time. After incubation, cell lysates were prepared as described above and first precleared by incubation with protein A/G-agarose for 15 min, microcentrifuged, and then transferred to new tubes. Equal amounts of proteins were incubated with anti-p27 Kip1 antibody for 2 h at 4°C. Protein A/G-agarose was added to absorb the immunocomplexes at 4°C overnight. Immunoprecipitated proteins were resolved by SDS-PAGE, and radiolabeled p27 Kip1 was visualized by autoradiography.
Flow Cytometric Cell Analysis-Cell cycle distribution was analyzed by flow cytometry as follows. At each time point, cells were harvested, washed twice with phosphate-buffered saline (PBS), and fixed in 70% ethanol for at least 2 h at Ϫ20°C. Fixed cells were washed with PBS, incubated with 1 ml of PBS containing 0.5 g/ml RNase A and 0.5% Triton X-100 for 30 min at 37°C, and then stained with 50 g/ml propidium iodide. The stained cells were analyzed using a FAScan laser flow cytometer (Becton Dickinson, San Jose, CA) and ModFit LT cell cycle analysis software (Verity Software, Topsham, ME).
Subcellular Fractionation-Mitochondrial and cytosolic (S100) fractions were prepared using differential centrifugation (33,(41)(42)(43). Briefly, Jurkat T cells were treated with 5GG or vehicle (Me 2 SO) for the indicated times. At the end of the treatment, cells were harvested, washed twice in ice-cold PBS, and resuspended in homogenizing buffer (250 mM sucrose, 20 mM HEPES, 10 mM KCl, 1.5 mM MgCl 2 , 1 mM EDTA, 1 mM EGTA, 1 mM dithiolthione, 17 g/ml phenylmethylsulfonyl fluoride, 8 g/ml aprotinin, 2 g/ml leupeptin (pH 7.4)) and incubated on ice for 30 min. Cells were passed through a needle 10 times. Unlysed cells, large plasma membrane pieces, and nuclei were pelleted by centrifugation at 750 ϫ g for 10 min. The supernatant was spun at 10,000 ϫ g for 15 min. This pellet was resuspended in homogenizing buffer and represents the mitochondrial fraction. The remaining supernatant was spun at 100,000 ϫ g for 1 h. The supernatant from this final centrifugation represents the cytosolic (S100) fraction. The determination of cytochrome c release was performed by Western blot as described above.
DNA Extraction and Electrophoresis Analysis-Jurkat T cells (2 ϫ 10 5 cells/ml) were harvested, washed with PBS, and then lysed with digestion buffer containing 0.5% sarkosyl, 0.5 mg/ml proteinase K, 50 mM Tris (hydroxymethyl) aminomethane (pH 8.0), 10 mM EDTA at 56°C for 3 h and treated with RNase A (0.5 g/ml) for another 2 h at 56°C. The DNA was extracted by phenol/chloroform/isoamyl alcohol (25/24/1) before loading and analyzed by 2% agarose gel electrophoresis. The agarose gels were run at 50 V for 120 min in TBE buffer (Tris borate/EDTA electrophoresis buffer). Approximately 20 g of DNA was loaded in each well, visualized under UV light, and photographed.

Inhibition of the Purified 20 and 26 S Proteasome
Activities by 5GG-Because the polyphenol-containing ester bond is considered to have the ability to inhibit chymotrypsin-like activity of the proteasome (36, 37), we hypothesized that 5GG, which contains five ester bonds, might act as a proteasome inhibitor. In order to assess the inhibitory efficacy of 5GG on proteasome activity, we first performed the proteasome inhibition assay in a cell-free system. The chymotryptic activity of the 20 S proteasome was measured in vitro by monitoring the release of the fluorophore, AMC, from the synthetic peptide substrate, LLVY-AMC, in the presence of 5GG. As shown in Fig. 2, 5GG strongly inhibited chymotrypsin-like activity of purified M. thermophila ( Fig. 2A), rabbit (Fig. 2B) 20 S proteasome, and purified rabbit 26 S proteasome (Fig. 2C). IC 50 values were ϳ73.2 nM, ϳ4.64 M, and ϳ15.18 M, respectively. This result ( Fig. 2A) was similar to the inhibition curve of the ester bond-containing phytopolyphenols EGCG (36) and tannic acids (37). We also examined the effects of 5GG on the PGPH activity of the purified M. thermophila 20 S proteasome. The PGPH activity was significantly inhibited by 5GG with an IC 50 value of ϳ122.5 nM ( Fig. 2A). To ascertain that the inhibitory effect of 5GG on proteasome activity did not result from contaminated peptidase activities, the purities of proteasomes were assessed by SDS-PAGE and native PAGE. The purified M. thermophila 20 S proteasome was shown to be essentially homogeneous on native PAGE at a molecular mass of ϳ645 kDa and gave two major subunits, ␣ (24 kDa) and ␤ (22 kDa), when analyzed by SDS-PAGE ( Fig. 2A, right panel). The purified rabbit 20 or 26 S proteasome on native PAGE also revealed a single major holoenzyme band at a high molecular mass of ϳ700 or ϳ1000 kDa, respectively, although the subunit compositions of the purified rabbit 20 or 26S proteasome appeared more heterogeneous (Fig. 2, B and C, right panels). Taken together, these data imply that 5GG is a proteasome inhibitor.
Inhibition of 26 S Proteasome Activity in Jurkat T Cell Extracts by 5GG-Proteasome inhibitors make malignant cells more vulnerable to cell growth arrest and/or apoptosis by changing the stability or activity of numerous proteins that regulate cell cycle progression and apoptosis. Therefore, we next examined the inhibitory effects of 5GG on proteasome activity in tumor cell extracts. Whole cell lysates were prepared from exponentially growing cultures of human Jurkat T cells and then subjected to proteasome inhibition assays determined by the hydrolysis of fluorogenic substrates in the presence of 5GG as described under "Experimental Procedures." In the inhibition assay for chymotrypsin-like activity, 5GG showed a dose-dependent inhibitory effect at concentrations from 1 to 5 M. Similar to the inhibitory effects of well established proteasome inhibitors MG132 and lactacystin (IC 50 of ϳ0.53 and ϳ1.64 M, respectively), 5GG (IC 50 of ϳ1.38 M) was potent against the chymotrypsin-like activity in Jurkat cell lysates (Fig. 3A).
The Selectivity and Specificity of the 26 S Proteasome Inhibition Mediated by 5GG-To elucidate the effects of 5GG on PGPH and trypsin-like proteasome activities in Jurkat cell lysates, 5GG at various concentrations was added to cell extracts of Jurkat T cells, and the inhibition of PGPH and trypsin-like proteasome activities was determined by proteasome assays with specific substrates for PGPH and trypsin-like activities. The results show that 5GG inhibits PGPH proteasome activity in a dose-dependent manner with an IC 50 of 24.94 M, whereas 5GG only inhibited ϳ33% of trypsin-like proteasome activity at 40 M (Fig. 3, B and C). In contrast, the proteasome inhibitor MG132 led to the marked inhibition of PGPH and trypsin-like activities (IC 50 of ϳ2.83 and ϳ13.48 M, Fig. 3, B and C, respectively). Like MG132, lactacystin inhibited PGPH and trypsin-like activities with IC 50 values of ϳ9.02 and 34.96 M, respectively (Fig. 3, B and C).
To further determine the specificity of proteasome inhibition that is mediated by 5GG, we evaluated the effects of 5GG on the activity of purified calpain I (calcium-activated natural protease I), a cytosolic cysteine protease regulated by Ca 2ϩ and ubiquitously distributed in various cells (44). Calpain I activity was determined by measuring the fluorescent intensity of AMCs hydrolyzed from calpain I-specific substrate in the presence of 5GG or EGCG or calpeptin (a specific calpain I inhibitor) in a cell-free system. At a concentration of 10 M, 5GG had no influence on calpain I protease activity; EGCG at the same concentration was slightly inhibitory (ϳ27%); and calpeptin at 1 M completely abolished calpain I activity (Fig. 4). Taken together, these results indicate that 5GG can specifically inhibit chymotrypsin-like and PGPH activities of the 26 S pro-teasome in tumor cell lysates, and 5GG is a much more potent inhibitor of chymotrypsin-like activity than of PGPH activity.
Inhibition of the 26 S Proteasomal Chymotrypsin-like Activity in Intact Jurkat T Cells by 5GG-To determine whether 5GG is capable of penetrating inside cells and subsequently inhibiting chymotrypsin-like activity of the 26 S proteasome in vivo, intact Jurkat T cells were incubated with 5GG at various concentrations for 12 h followed by an additional 2-h incubation with the specific fluorogenic substrate for chymotrypsin-like activity. The catalytic activity of chymotrypsin-like proteasome was measured by monitoring the release of the fluorophore, AMCs, hydrolyzed from specific substrates. Fig. 5 shows that 5GG inhibited chymotrypsin-like activity in a dose-related manner in intact Jurkat T cells, but the inhibitory potency for chymotrypsin-like activity was lower compared with inhibition of the purified 20 S proteasome and in whole cell lysates (Figs. 2 and 3A). We supposed that this difference might come from the delay during which 5GG penetrated into the inside of cells. Consistent with our prediction, the cell-permeable proteasome inhibitors MG132 and lactacystin inhibited chymotrypsin-like activity with much higher concentrations in intact Jurkat T cells than in Jurkat T cell lysates (Fig. 5 versus Fig. 3A). Taken together, these findings clearly indicate the cell permeability of 5GG and the inhibition by 5GG of chymotrypsin-like activity of the 26 S proteasome on the inside of intact Jurkat T cells.
Induction of Proteasome Inhibition Marker Proteins p27 Kip1 , p21 Cip1/WAF1 , and Bax by 5GG-In our previous study, we showed that 5GG can increase levels of the CKIs p27 Kip1 and p21 Cip1/WAF1 and cause cell cycle arrest at G 1 phase in MCF7 cells (32). These CKIs are known substrates of the 26 S proteasome, and the proteasomal degradation of CKIs is responsible for cell cycle progression (8,19). Thus, we suggested that 5GG might increase the accumulation of p27 Kip1 and p21 Cip1/WAF1 proteins in parallel with its inhibition of the 26 S proteasome in Jurkat T cells. To examine this possibility, Jurkat T cells were treated with various concentrations of 5GG for 12 h, and the levels of p27 Kip1 and p21 Cip1/WAF1 proteins were determined by Western blotting. As shown in Fig. 6A, 5GG substantially upregulated levels of p27 Kip1 and p21 Cip1/WAF1 in a concentrationdependent manner. To further confirm that the enhanced stabilities of proteasome-related substrates such as p27 Kip in the presence of 5GG was due to proteasome inhibition, we examined the ability of proteasome inhibitors MG132 and lactacystin to induce up-regulation of p27 Kip in Jurkat T cells. Consistent with 5GG treatment, exposure of Jurkat T cells to proteasome inhibitors MG132 and lactacystin for 12 h caused a dose-dependent increase in p27 Kip1 (Fig. 6C). We further studied the effect of 5GG on the temporal expression of p27 Kip1 and p21 Cip1/WAF1 proteins. Levels of p27 Kip1 and p21 Cip1/WAF1 proteins accumulated in Jurkat T cells as early as 3-6 h and remained stable until 24 h after the initiation of treatment with 10 M 5GG (Fig. 6B). Threonine 187 is one of the phosphorylated sites of p27 Kip1 , and its phosphorylation contributes

FIG. 5. Inhibitory effect of 5GG on chymotrypsin-like activity of the 26 S proteasome in intact Jurkat T cells.
Jurkat T cells (2 ϫ 10 5 cells/ml) were incubated with various concentrations of 5GG, MG132, or lactacystin for 12 h and then coincubated for 2 h with fluorogenic peptide substrate for chymotrypsin-like proteasome activity. After incubation, the medium was collected and subsequently subjected to a proteasome activity assay as described under "Experimental Procedures." The chymotrypsin-like activity was expressed as the percentage of the control (defined as 100%). Shown are means Ϯ S.E. for three separate experiments. significantly to enhanced ubiquitination and degradation of p27 Kip1 by the proteasome (45). To further elucidate the mechanism by which 5GG induces p27 Kip1 accumulation, we inves-tigated dose-response and time course curves for 5GG-induced phosphorylation of p27 Kip1 at Thr 187 . The increased levels of p27 Kip1 phosphorylation at Thr 187 were consistent with protein levels of p27 Kip1 with regard to both dose and time (Fig. 6, A  and B). Similar findings were observed in Fig. 6C. We observed increases in the phospho form of p27 Kip1 (at Thr 187 ) caused by MG132 and lactacystin consistent with changes in the native form of p27 Kip1 . These results indicate that the accumulation of p27 Kip1 induced by 5GG is unlikely to have resulted from the down-regulation of p27 Kip1 phosphorylation at Thr 187 . In addition, since protein kinase B/Akt has been reported to phosphorylate p27 Kip1 at Thr 157 directly and contributes to tumor cell proliferation by cytosolic retention of p27 Kip1 (46), we further elucidated the effects of 5GG on the phosphorylation status of Akt as a marker representative of the activation of Akt in Jurkat T cells. Phosphorylation content of Akt was not substantially changed in response to 5GG treatment (data not shown). Taken together, this suggests that the increased levels of p27 Kip1 after 5GG treatment in Jurkat T cells are not due to alterations in the phosphorylation of p27 Kip1 but result from proteasomal inhibition by 5GG.
The proapoptotic protein Bax has a pivotal role in controlling programmed cell death by regulating mitochondrial integrity and mitochondria-initiated caspase activation and is known to be regulated by proteasomal degradation (47,48). To study whether the degradation of Bax was reduced by 5GG, the induction of Bax was determined by Western blotting under the same conditions as above. The results were in agreement with our expectation that Bax expression is increased in response to proteasomal inhibition by 5GG (Fig. 6, A and B).
The Expression of p27 Kip1 , p21 Cip1/WAF1 , and Bax in Jurkat T Cells Is Controlled at the Post-translational Level by 5GG-The expression of p27 Kip1 , p21 Cip1/WAF1 , and Bax is frequently regulated at the transcriptional and translational levels. To examine whether translation of p27 Kip1 , p21 Cip1/WAF1 , and Bax transcripts contributed to the 5GG-induced increase in these protein levels, Jurkat T cells were precultured with 10 g/ml cycloheximide, an inhibitor of protein synthesis, and p27 Kip1 , phospho-p27 Kip1 , p21 Cip1/WAF1 , and Bax were measured in the presence and absence of 5GG. The protein levels of p27 Kip1 , phospho-p27 Kip1 , p21 Cip1/WAF1 , and Bax were significantly decreased in cycloheximide-pretreated cells compared with cycloheximide-untreated cells (Fig. 7A, lane 2 and lane 1, respectively). However, dose-dependent increases in these proteins were clearly observed in the presence of 5GG with cycloheximide pretreatment (Fig. 7A, lanes 3-7). We next examined the half-life of p27 Kip1 protein by carrying out an in vivo pulsechase labeling experiment to compare the stability of p27 Kip1 protein between control and 5GG-treated Jurkat T cells. Jurkat T cells were pulse-labeled for 30 min and later chased with or without 5GG for the indicated times. As shown in Fig. 7B, the increase in half-life of radiolabeled p27 Kip1 protein in 5GGtreated Jurkat T cells when compared with control cells demonstrated that 5GG up-regulated p27 Kip1 protein levels via inhibition of its degradation. Collectively, these data provide additional support that p27 Kip1 , p21 Cip1/WAF1 , and Bax are up-regulated at the level of protein stability, which is modulated by 5GG-induced proteasomal inhibition in Jurkat T cells.
Accumulation of Ubiquitin-tagged Proteins in Jurkat T Cells Treated with 5GG-Inhibition of proteasome activity should prevent the rapid degradation of ubiquitin-tagged proteins. Our results (Fig. 8A) show that incubation of Jurkat T cells with 10 M of 5GG causes an accumulation of ubiquitin-tagged proteins in a time-dependent manner, as determined by Western blotting using anti-ubiquitin antibody. The increase was observed particularly in high molecular weight ubiquitin-pro-FIG. 6. Effects of 5GG treatment on the accumulation of proteasomal degradation-related proteins p27 Kip1 , p21 Cip1/WAF1 , and Bax in Jurkat T cells. A, Jurkat T cells were exposed for 12 h to various concentrations of 5GG as indicated, and then Western blotting analysis of p27 Kip1 , phosphorylated p27 Kip1 , p21 Cip1/WAF1 , d Bax proteins was performed as described under "Experimental Procedures." B, Jurkat T cells were incubated with 10 M 5GG for the indicated times, and then the induction of p27 Kip1 , phosphorylated p27 Kip1 , p21 Cip1/WAF1 , and Bax proteins were determined by Western blotting analysis. C, Jurkat T cells were exposed for 12 h to various concentrations of MG132 or lactacystin as indicated; p27 Kip1 and phosphorylated p27 Kip1 were detected by Western blotting. Phospho-p27 showed that p27 Kip1 protein was phosphorylated at Thr 187 . Simultaneous immunoblotting of ␤-actin was used as an internal control for equivalent protein loading. tein conjugates, which correspond to the preferred substrates for the 26 S proteasome (1,49). Additionally, a similar observation is shown in Fig. 8B, where MG132 resulted in a timedependent increase in ubiquitin-tagged proteins. This implies that 5GG may disrupt the function of the 26 S proteasome and prolong the half-life of ubiquitinated substrates.
Induction of p27 Kip1 and p21 Cip1/WAF1 in Response to 5GG Treatment Followed by G 1 Cell Cycle Arrest-Induction of CKIs such as p27 Kip1 and p21 Cip1/WAF1 , which are caused by proteasomal inhibition, is associated with cell growth arrest at G 1 phase in tumor cells. In order to determine whether 5GG causes cell cycle arrest following the induction of p27 Kip1 and p21 Cip1/WAF1 in Jurkat T cells, cells treated with Me 2 SO or 5GG were subjected to flow cytometric analysis after staining their DNA with propidium iodide. Histograms of flow cytometric data are shown in Fig. 9A. Control cells (Me 2 SO-treated) progressed smoothly through the cell cycle. In contrast, 5GGtreated Jurkat T cells were arrested at G 1 (87.12%) after 24 h of treatment. To further demonstrate that 5GG can cause cell cycle arrest at G 1 , the phosphorylation status of Rb, a hallmark of G 1 -to-S phase progression, was investigated by Western blotting. As shown in Fig. 9B, the degree of phosphorylation of Rb was decreased after 12 h of 10 M 5GG treatment compared with total Rb protein. These results indicate that the accumulation of CKIs p27 Kip1 and p21 Cip1/WAF1 in response to the inhibition of proteasome by 5GG is accompanied by cell cycle arrest at G 1 phase in Jurkat T cells.
Induction of Bax by 5GG Accompanied by Cytochrome c Release and Apoptosis-Bax, a Bcl-2 family protein, translocates to mitochondria in response to apoptotic (death) signal in cells and leads to loss of membrane potential (⌬ m ) and cytochrome c release during apoptosis (51)(52)(53). To determine whether the increased Bax induced by 5GG can cause the release of cytochrome c, we next evaluated the effect of 5GG on mitochondrial cytochrome c release into the cytosol. Subcellular fractions of Jurkat T cell lysates harvested for the indicated times after treatment with 10 M of 5GG were separated, and then the release of cytochrome c was determined by Western blotting. Fig. 10A shows that cytochrome c release into the cytosol is detected at 24 h after 5GG treatment relative to the decrease in mitochondrial cytochrome c. The expression of cytochrome c oxidase subunit IV and lactate dehydrogenase subunit H, markers for the mitochondrial and cytosolic fractions, respectively, were measured both in cytosolic and mitochondria-enriched fractions by immunoblot analysis. Fig. 10B shows that FIG. 7. 5GG-induced stabilization of p27 Kip1 , p21 Cip1/WAF1 and Bax protein via post-translation regulation. A, measurement of p27 Kip1 , p21 Cip1/WAF1 , and Bax protein degradation after inhibition of protein synthesis by cycloheximide. Jurkat T cells were pretreated with 10 g/ml cycloheximide for 2 h and subsequently coincubated with various concentrations of 5GG as indicated. The expression of p27 Kip1 , the phosphorylated form of p27 Kip1 , p21 Cip1/WAF1 , and Bax was determined by Western blotting analysis. ␤-Actin was used as an internal control for equivalent protein loading. As a control, cells were exposed for 12 h to 5 l/ml Me 2 SO after cycloheximide pretreatment or no treatment. B, measurement of p27 Kip1 protein stability by an in vivo pulse-chase labeling method. Jurkat T cells were incubated in methionine-free medium for 20 min and then metabolically labeled with [ 35 S]methionine for 30 min. Afterward, cells were chased in complete medium containing unlabeled Met with or without 10 M 5GG for the indicated times. Radiolabeled p27 Kip1 from lysis buffer cell extracts was immunoprecipitated with anti-p27 Kip1 antibody, and the immunocomplexes were precipitated with protein A/G-agarose beads. The immunocomplexes were separated by SDS-PAGE, and radioactive protein products were visualized by autoradiography. the cytochrome c oxidase subunit IV and lactate dehydrogenase subunit H are markedly enriched in mitochondrial and cytoplasmic fractions, respectively, indicating that isolation of mitochondria and cytosol from Jurkat T cells by differential centrifugation is workable.
To further show that the accumulation of Bax and cytochrome c release caused by 5GG treatment induces apoptosis in Jurkat T cells, DNA fragmentation as an index of apoptosis was measured at various time points after treatment with 10 M 5GG. Consistent with our prediction, oligonucleosomal DNA fragmentation sequentially occurred in Jurkat T cells exposed for 36 h to 10 M of 5GG (Fig. 10C) following G 1 cell cycle arrest (Fig. 9A) and the release of cytochrome c (Fig. 10A). DISCUSSION Proteasome-mediated degradation of cell proteins plays a pivotal role in the regulation of several basic cellular processes, including differentiation, proliferation, cell cycling, apoptosis, gene expression, and signal transduction. Imbalances in proteasome-mediated protein degradation contribute to various human diseases such as cancer and neurodegenerative and myodegenerative diseases, suggesting that the proteasome may be a novel target for anti-cancer therapy (8, 18 -20). Further, malignant cells are more susceptible than normal cells to the loss of proteasome activity (54 -56). Although the mechanism of this difference is not yet clear, inhibition of proteasome activity might down-regulate pathways that permit neoplastic proliferation and suppress apoptosis in malignant cells. In our current study, we demonstrated that 5GG can potently and specifically inhibit chymotrypsin-like proteasome activity in vitro (Figs. 2 and 3A), and these results are in agreement with the hypothesis that ester bond-containing polyphenols can act as chymotrypsin-like proteasome inhibitors due to nucleophilic attack between the threonine active site of chymotrypsin-like subunits and the ester bond carbon of ester bond-containing polyphenols (36,37). 5GG comprises five gallic acids individually linked with a glucose core by ester linkage (Fig. 1), and that is the possible reason why there was a more potent inhibitory effect on chymotrypsin-like activity by 5GG than by EGCG (Fig. 3A). However, 5GG had less potency in inhibiting chymotrypsin-like activity than EGCG in intact Jurkat T cells, showing IC 50 values of ϳ38.56 and 18 M, respectively (Fig. 5) (36). We suggest that the difference might result from the larger molecular weight of 5GG in comparison with EGCG (M r Ϸ 1024 versus M r Ϸ 456, respectively), where this molecular feature makes it difficult for 5GG to penetrate into cells.
Proteasome inhibitors induce cell cycle arrest at G 1 phase, and this has been shown to depend on intracellular accumulation and stabilization of cell cycle regulatory proteins including CKIs p27 Kip1 and p21 Cip1/WAF1 (8,19). Also, human leukemic cells display abnormally high levels of proteasome (57), implying that proteasome activity contributes to the malignant proliferation of leukemic cells such as human Jurkat T cells (acute T cell leukemia). Thus, we suggest that proteasome inhibition induced by 5GG might arrest the cell cycle at G 1 phase through the induction of p27 Kip1 and p21 Cip1/WAF1 . To test this possibility, we examined the effect of 5GG on cell cycle progression and the expression of CKIs p27 Kip1 and p21 Cip1/WAF1 . The results show that 5GG treatment leads to cell cycle arrest at G 1 phase in Jurkat T cells (Fig. 9A), which is probably attributable to proteasome inhibition and accumulation of p27 Kip1 and p21 Cip1/WAF1 (Figs. 6 and 7). However, blockade of cell cycle progression and accumulation of p27 Kip1 and p21 Cip1/WAF1 is unlikely to be completely dependent on the inhibition of proteasome activity by 5GG for the following reason: the apparent increases in p27 Kip1 and p21 Cip1/WAF1 proteins were observed after treatment with 10 M 5GG after 12-h incubation in Fig.  6B compared with Fig. 5, in which only ϳ27.76% inhibition of chymotrypsin-like activity in intact Jurkat T cells occurred under the same incubation condition. Therefore, we inferred that 5GG down-regulates growth factor receptor signaling responsible for cell survival. In fact, we had found that 5GG can inhibit EGFR signaling in lung epithelial carcinoma A549, 2 and it implies the possibility that 5GG might interfere with signal transduction of cytokine receptors that are essential for leukemic cell survival. Furthermore, 5GG can arrest cell cycle progression at G 1 phase through down-regulation of CDK4 and CDK2 in the breast cancer cell line MCF7 (32), and, as shown in Fig. 9B, 5GG can decrease the phosphorylation of Rb in Jurkat T cells. This raises the possibility that 5GG might directly or indirectly mediate the activity of cyclin D/CDK4 or cyclin E/CDK2 or both in Jurkat T cells, which is required for the cells to progress from G 1 to S phase in those cells possessing a functional pRb (58). It remains to be addressed whether or not 5GG blocks growth factor receptor signaling and has an inhibitory effect on CDK4 and CDK2 activities resulting in G 1 cell cycle arrest in Jurkat T cells.
5GG increased p27 Kip1 and p21 Cip1/WAF1 expression, an effect that was associated with G 1 arrest in Jurkat T cells and with concomitant induction of apoptosis (Fig. 10C). However, cell cycle status is not the sole determinant the cell response to proteasome inhibition induced by 5GG. In our current study, we found that the accumulation of Bax induced by 5GG was simultaneously accompanied by p27 Kip1 and p21 Cip1/WAF1 (Figs. 6 and 7). Because Bax is a member of the Bcl-2 family and abnormal accumulation of Bax, which binds to the voltagedependent anion channel in the outer mitochondrial membrane (52), may lead to mitochondrial perturbation and allow cytochrome c release, we inferred that proteasome-mediated apoptosis driven by 5GG treatment partly resulted from the accumulation of Bax and the release of cytochrome c (Fig. 10A).
Other members of the proapoptotic Bcl-2 family, such as Bik (59) and tBid (60), also act as proteasome substrates that induce apoptosis following an accumulation caused by proteasome inhibition. We cannot rule out the possibility that the accumulation of Bik and tBid might occur in Jurkat T cells after 5GG treatment. This possibility will require further study.
Because the proteasome acts as a regulator of cell growth and apoptosis by controlling levels of many key cellular proteins, perturbation of its function by proteasome inhibitors may have a cytotoxic effect on cancer cells. These characteristics clearly highlight the potential of proteasome inhibitors to act as novel anticancer agents. In fact, the proteasome inhibitor PS-341, a dipeptidyl boronic acid, has been evaluated under phase I and phase II clinical trials as an anticancer agent (17,21,50,61,62). In conclusion, the results presented in this study reveal that 5GG can specifically and potently inhibit the chymotrypsin-like activity of purified 20 and 26 S proteasome and 26 S proteasome in Jurkat T cell lysates in a cell-free system and exert an inhibitory effect on chymotrypsin-like proteasome activity in intact Jurkat T cells. This inhibitory effect might contribute to the accumulation of ubiquitin-tagged proteins and up-regulation of CKIs p27 Kip1 and p21 Cip1/WAF1 and the proapoptotic protein Bax and induce G 1 cell cycle arrest and sequential apoptosis in Jurkat T cells. The antiproliferative and proapoptotic effects induced by 5GG are consistent with those of current proteasome inhibitors and provide strong evidence that 5GG is a proteasome inhibitor.