Antiproliferative and Proapoptotic Activities of Triptolide

2666 Vol. 8, 2666–2674, August 2002 Clinical Cancer Research

Antiproliferative and Proapoptotic Activities of Triptolide (PG490), a Natural Product Entering Clinical Trials, on Primary Cultures of Human Prostatic Epithelial Cells1

Taija M. Kiviharju, Philip S. Lecane,2 INTRODUCTION Robert G. Sellers, and Donna M. Peehl3 Extracts of the herb Tripterygium wilfordii hook F have Department of Urology, Stanford University School of Medicine, been used for more than two centuries in traditional Chinese Stanford, California 94305 medicine to treat a variety of autoimmune and inflammatory diseases including rheumatoid arthritis (1). Triptolide (PG490), a diterpene triepoxide, is a purified compound from Triptery- ABSTRACT gium that has been identified as one of the major components Interest in exploiting traditional medicines for preven- responsible for the immunosuppressive and anti-inflammatory tion or treatment of cancer is increasing. Extracts from the effects of this herb (2). Recent explorations of the mechanisms herb Tripterygium wilfordii hook F have been used in China of action of triptolide revealed many properties relevant not only for centuries to treat immune-related disorders. Recently it to anti-inflammatory activity but also to anticancer activity. was reported that triptolide (PG490), a purified compound Antiproliferative and proapoptotic activity of triptolide has been from Tripterygium, possessed antitumor properties and in- shown with many different types of cancer cells in vitro and in duced apoptosis by p53-independent mechanisms in a vari- vivo (3, 4). Triptolide also potentiates the activities of other ety of malignant cell lines. This property of triptolide at- agents and therefore may be useful not only as a single com- tracted our attention because we have found that primary pound but also in combination with other cytotoxic drugs for cultures of human prostatic epithelial cells derived from cancer treatment. For example, triptolide sensitized tumor ne- normal tissues and adenocarcinomas are in general ex- crosis factor ␣-resistant tumor cells to tumor necrosis factor tremely resistant to apoptosis. Furthermore, the function of ␣-induced apoptosis (5) and showed cooperative proapoptotic wild-type p53 is impaired in these cells such that drugs that effects with doxorubicin. In the latter studies, triptolide was require p53 activity to induce cell death are ineffective. found to block doxorubicin-mediated induction of p21WAF1/CIP1

Therefore, the properties of triptolide and the recent ap- and accumulation of cells in G2-M (6). The investigators proval of its water-soluble form (PG490-88) for entry into proposed that triptolide, by blocking p21WAF1/CIP1-mediated Phase I clinical trials suggested that this drug was a prom- growth arrest, created conflicting cell cycle checkpoints that ising candidate to test for antitumor activity against prostate enhanced apoptosis in tumor cells. PG490-88, a succinate salt cells. Experiments presented here demonstrated that trip- water-soluble derivative of PG490 that is converted to triptolide tolide had dose-dependent effects on both normal and in the serum (7), has recently been approved for Phase I clinical cancer-derived primary cultures of human prostatic epithe- trials. lial cells. Low concentrations of triptolide inhibited cell pro- The exact targets and function (or molecular mechanism of liferation and induced a senescence-like phenotype. Higher action) of triptolide are still unknown. An attempt to discern the concentrations of triptolide induced apoptosis that was un- molecular signaling pathways triggered by triptolide was made expectedly associated with nuclear accumulation of p53. by cDNA microarray analysis of triptolide-treated normal and Paradoxically, levels of the p53 target genes, p21WAF1/CIP1 transformed bronchial epithelial cells. Triptolide reduced ex- and hdm-2, were reduced, as was bcl-2. Our preclinical pression of cell cycle regulators and survival genes such as studies suggest that triptolide might be an effective preven- cyclins D1, B1, and A1; cdc-25; bcl-x; and c-jun (8). Anti- tive as well as therapeutic agent against prostate cancer and inflammatory, antiproliferative, and proapoptotic properties of that triptolide may activate a functional p53 pathway in triptolide have also been associated with inhibition of nuclear prostate cells. factor-␬B (5, 8, 9). We became interested in triptolide because it was reported to direct cancer cells to undergo apoptosis independent of p53 activity (4). The tumor suppressor protein p53 responds to Received 1/28/02; revised 5/8/02; accepted 5/15/02. different forms of cellular stress such as DNA damage caused The costs of publication of this article were defrayed in part by the by ionizing radiation or chemotherapeutic drugs by targeting the payment of page charges. This article must therefore be hereby marked checkpoint genes that inhibit cell cycle progression (10) and/or advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. trigger apoptosis (review in Refs. 11 and 12). Previous studies 1 Supported in part by Department of Army Grant DAMD 17-99-1- conducted by us showed that activation of p53 in response to 9004. ␥-irradiation or other stresses such as hypoxia or DNA-damag- 2 Present address: Pharmacyclics, 995 East Arques Avenue, Sunnyvale, ing drugs is attenuated in primary cultures of normal and ma- CA 94086. 3 To whom requests for reprints should be addressed, at Department of lignant prostatic epithelial cells, despite the presence of the Urology, Stanford Medical Center, Stanford, CA 94305-5118. Phone: wild-type p53 gene (13). Because many chemotherapeutic drugs (650) 725-5531; Fax: (650) 723-0765; E-mail: [email protected]. use p53-mediated pathways to induce growth arrest or apopto-

sis, we have suggested that dysfunction of p53 may partially 60-mm dish containing 5 ml of Complete MCDB 105 with explain the resistance of prostate cancer to drug treatment. The experimental factors. Cells were incubated for 10 days without prevalence of mutated p53 in advanced prostate cancer (14) may feeding and then fixed with 10% formalin and stained with also impact response to chemotherapy. We therefore hypothe- crystal violet (17). Growth was quantitated with an Artek image size that drugs using p53-independent pathways would be most analyzer (Dynatech, Chantilly, VA), which measures the total efficacious against prostate cancer. area of the dish covered by cells. This relative value has been In our study, we found that triptolide exerted two qualita- shown to be directly proportional to cell number (22). tively different types of effects on primary cultures of prostatic High Density Growth Assay. Cells were inoculated at 5 epithelial cells. Low concentrations of triptolide inhibited 10 cells/dish into collagen-coated, 60-mm dishes containing growth and induced senescence, an irreversible growth-arrested Complete MCDB 105. One day later (day 0), various concen- state that is associated with tumor suppression. Higher concentrations (1–100 ng/ml) of triptolide were added. Cells treated trations of triptolide triggered apoptosis. Interestingly, unlike with diluent (0.001% DMSO) were included as controls. After 3 our results with any previously tested compound, triptolide- days, fresh media containing diluent or triptolide were replaced. induced apoptosis of primary cultures was accompanied by Cells in replicate dishes were counted by hemocytometer after nuclear accumulation of p53. This finding was unexpected, trypsinization on days 0, 3, and 6. given the original reports of p53-independent activity of triptol- Cell Viability. Loss of cell viability was assessed by the ide, but in fact more recent studies indicate that triptolide- trypan blue exclusion method. Cells treated with or without induced apoptosis may be at least partly mediated through triptolide were harvested by trypsinization. After incubation in activation of p53 in cells with wild-type p53 (6, 15). Down- 0.04% trypan blue (Sigma-Aldrich) for 4 min, cells were regulation of p53 target genes hdm-2 and p21WAF1/CIP1, as well counted under a hemocytometer. The number of cells that re- as bcl-2, was also observed in prostatic epithelial cells. Our tained the dye (nonviable) and the total cell number were noted. 5 preclinical findings support further investigation of chemopre- Apoptosis. Cells were inoculated at 10 cells/dish into ventive and chemotherapeutic activity of triptolide against pros- collagen-coated, 60-mm dishes containing Complete MCDB tate cancer. 105. One day later, the medium was replaced, and cells were treated with various concentrations (1–100 ng/ml) of triptolide. Cells treated with diluent (0.001% DMSO) were included as MATERIALS AND METHODS controls. After 24, 48, and 72 h, Hoechst 33342 and propidium Cell Culture and Reagents. Tissue samples were dis- iodide (Sigma-Aldrich) were added to medium at 10 and 20 sected from radical prostatectomy specimens. None of the pa- ␮g/ml, respectively. After incubation for 15 min at 37°C, 400 tients had received prior chemical, hormonal, or radiation ther- cells from each dish were counted (from 10 randomly selected apy. Histological assessment was performed as described fields) under fluorescence to determine the proportion of viable previously (16). Epithelial cells were cultured and characterized and apoptotic cells (23). as described previously (17). Culture medium was MCDB 105 Cell Cycle Analyses. Semiconfluent cell cultures were (Sigma-Aldrich, St. Louis, MO) supplemented with 10 ng/ml fed fresh medium containing triptolide (1 or 50 ng/ml). Cells cholera toxin, 10 ng/ml epidermal growth factor, 10 ␮g/ml treated with diluent (0.001% DMSO) were included as controls. bovine pituitary extract, 4 ␮g/ml insulin, 1 ␮g/ml hydrocorti- At 24 and 48 h, cells were harvested by trypsinization, washed sone, 0.1 mM phosphoethanolamine, 30 nM selenium, 0.03 nM with cold PBS, and fixed by dropwise addition of ice-cold 70% all-trans-retinoic acid, 2.3 ␮M ␣-tocopherol, and 100 ␮g/ml ethanol. After1hoffixation at 4°C, cells were pelleted and then gentamicin (Complete MCDB 105). The sources and prepara- incubated with RNase A (50 ␮g/ml; Sigma-Aldrich) and stained tion of these supplements were described previously (17). with propidium iodide (20 ␮g/ml). Analyses of DNA content PG490 (97% pure triptolide) was provided by Pharmagenesis were carried out on a FACScan flow cytometer, and cell cycle (Palo Alto, CA) and stored in a stock solution of 1 mg/ml in phase distribution was quantified using Cellfit software. DMSO at –20°C. Staining for SA-␤-gal.4 Forty thousand cells were inoc- Four cells strains used in this study were derived from ulated into each collagen-coated, 60-mm dish containing 5 ml of prostatic adenocarcinomas of Gleason grade 3/3 (E-CA-11), Complete MCDB 105. The next day, cells were treated with or 30% intraductal carcinoma/70% Gleason grade 4 (E-CA-12), without 1 ng/ml triptolide. SA-␤-gal-positive cells were de- and Gleason grade 3/4 (E-CA-13 and E-CA-14). An additional tected by the method of Dimri et al. (24) after 3, 5, and 10 days cell strain (E-PZ-10) was derived from histologically normal of treatment with triptolide. Cells were fed with fresh media and tissue of the peripheral zone. The characteristic traits of primary diluent or triptolide on day 5. The presence of positive staining cultures of human prostatic epithelial cells have been described was observed microscopically and photographed. previously (18). Cells cultured from adenocarcinomas are dis- Immunoblot Analyses. Cells were trypsinized and cen- tinguished from those derived from normal tissues on the basis trifuged. The cell pellet was washed in ice-cold PBS, resus- of cytogenetic abnormalities (19) and differential expression of pended in urea-Tris buffer [9 M urea, 75 mM Tris-HCl, (pH 7.5), metabolic enzymes (20, 21). Clonal Growth Assay. Cell cultures at ϳ30% conflu- ence were trypsinized, suspended in medium, and centrifuged.

The cell pellet was suspended in buffered saline at a concentra- 4 3 The abbreviations used are: GAPDH, glyceraldehyde-3-phosphate tion of 2 ϫ 10 cells/ml. Five hundred cells in 100 ␮lof dehydrogenase; mAb, monoclonal antibody; SA-␤-gal, senescence- buffered saline were inoculated into each collagen-coated, associated ␤-galactosidase.

Fig. 1 Clonal growth response to triptolide. Five prostatic epithelial cell strains (four derived from cancer and one derived from normal tissue) were inoculated at 500 cells/dish and grown in the presence of varying concentrations of triptolide for 10 days. The growth of each cell strain without triptolide was set as 100%. Each point represents the average value from two separate experiments, with triplicate dishes in each experiment, ϮSE.

and 0.15 M 2-mercaptoethanol], and sonicated briefly. The protein concentration was determined by a Bio-Rad assay (Bio- Rad, Hercules, CA). Typically, 50 or 80 ␮g of protein were separated by SDS-PAGE, transferred to polyvinylidene fluoride membranes (Osmonics, Westborough, MA), and blocked in PBS with 5% nonfat milk. Proteins were detected with the Fig. 2 Growth response to triptolide in high cell density cultures. A, 5 following antibodies: mouse anti-p53 mAb (DO-1), mouse anti- E-PZ-10 cells were inoculated at 10 cells/dish. The next day (day 0), WAF1/CIP1 INK4a cell number was determined in triplicate dishes, and replicate dishes p21 mAb, rabbit anti-p16 polyclonal antibody, were treated with varying concentrations of triptolide. Fresh medium rabbit anti-p27Kip1 polyclonal antibody, mouse anti-bax mAb, with or without triptolide was replaced on day 3. The number of attached and mouse anti-hdm-2 mAb (SMP14; Santa Cruz Biotechnol- cells/dish was determined at the indicated times. Each point represents Ϯ ogy, Santa Cruz, CA); mouse anti-bcl-2 mAb (Chemicon, Te- the average value from two to three replicate dishes SE. B, loss of viability of E-PZ-10 cells in the assay described in A as determined by mecula, CA); mouse anti-GAPDH mAb (Research Diagnostics, trypan blue exclusion. Each bar represents the average value from two Inc., Flanders, NJ); and mouse anti-hdm-2 mAb (2A10; a gift to three replicate dishes ϮSE. from Dr. Arnold Levine; Princeton University, NJ; Ref. 25). Anti-species horseradish peroxidase-conjugated secondary antibodies were obtained from Dako (Carpinteria, CA), and visual detection was performed using the enhanced chemilumines- nary tool to test the responses of prostatic epithelial cell strains cence method (Amersham, Piscataway, NJ). to triptolide. However, because our experimental plans called for the use of higher cell density cultures, and density can affect RESULTS cellular response, we further tested the effect of triptolide on Effects of Triptolide on Growth of Prostatic Epithelial growth of cells at higher density. Cells were inoculated into Cells. Clonal assays were used to test the effect of triptolide culture medium at 105 cells/dish and treated 1 day later with or on the growth of primary cultures of prostatic epithelial cells without triptolide (1–100 ng/ml). Cell numbers were determined derived from normal and malignant tissues. Cells were inocu- on days 3 and 6. Two cell strains, one derived from normal lated at 500 cells/dish into medium with or without triptolide tissue (E-PZ-10) and one derived from a primary adenocarci- (0.01–1 ng/ml), and clonal growth was evaluated after 10 days noma of the prostate (E-CA-12), were assayed. As shown in Fig. of incubation. Five cell strains (one derived from normal pe- 2A, treatment with triptolide inhibited cell proliferation in a ripheral zone tissue and four derived from adenocarcinomas) concentration-dependent manner. Triptolide at 1 ng/ml, which were assayed. Fig. 1 shows that triptolide was growth inhibitory. completely inhibited clonal growth, was less inhibitory in high Complete growth inhibition of all cell strains occurred with 1 density cultures. After 6 days, the number of E-PZ-10 cells ng/ml triptolide, with half-maximal growth inhibition at ϳ0.1 treated with 1 ng/ml triptolide was 38% of that of untreated ng/ml triptolide. cells. Slightly higher doses of triptolide (10–15 ng/ml) inhibited The clonal growth assay served as a convenient prelimi- growth by 65% at day 3 and by ϳ90% at day 6. The highest

Table 1 Effects of triptolide on cell cycle distribution Subconfluent cultures of E-PZ-10 and E-CA-12 cells were treated with indicated concentrations of triptolide and harvested for cell cycle analysis after 24 and 72 h of treatment with or without triptolide. Each entry indicates the proportion of cells (%) in the respective cell cycle compartments, as determined by flow cytometric analysis after propidium iodide staining. Triptolide (ng/ml) Triptolide (ng/ml) at 24 h at 72 h Cell phase 01500150 E-PZ-10

G1 67.8 57.5 55.5 74.3 73.4 46.5 S phase 20.3 26.4 32.3 15.4 12.8 40.4

G2-M 11.7 16.3 12.2 10.3 13.8 13.1 E-CA-12

G1 59.5 55.8 59.2 70.5 75.0 52.6 S phase 22.8 28.7 30.5 14.1 9.4 35.7

G2-M 17.7 15.5 10.3 15.5 15.6 11.7

Fig. 3 Induction of apoptosis by triptolide. E-CA-12 cells were treated with various concentrations of triptolide (1–100 ng/ml) for 24–72 h. The amount of apoptosis was quantified by Hoechst 33342/propidium iodide staining of nuclear DNA in conjunction with nuclear morphology. Four see higher apoptotic rates with E-CA-12 compared with E- hundred cells were randomly selected, and the number of cells with PZ-10 cells (data not shown). Possibly cells became nonviable normal and abnormal nuclei was noted for each treatment. Data repre- sent the mean (ϮSE) of two separate experiments. by other mechanisms in addition to apoptosis. Cell Cycle Distribution with Low and High Concentra- tions of Triptolide. Progression of cells through the cell cycle was evaluated by flow cytometric determination of cellular concentrations of triptolide (50 and 100 ng/ml) caused loss of DNA content. Cell cycle analyses were performed on E-PZ-10 cells over time. The response of E-CA-12 cells in high density and E-CA-12 cells that had been treated with 1 or 50 ng/ml assays to triptolide was comparable with that of E-PZ-10 cells. triptolide for 24–72 h (Table 1). The lower concentration of On day 6, growth was inhibited by 1 ng/ml triptolide to 58% of triptolide (1 ng/ml) slightly increased the accumulation of the control. With 50 ng/ml triptolide, the cell number on day 6 cells in S phase after 24 h. However, this effect was transient declined by 66% from day 0. and was not present at 72 h. Triptolide at 50 ng/ml for 24 h E-PZ-10 and E-CA-12 cells in this assay were also incu- increased the proportion of E-PZ-10 cells in S phase from 20.3% bated with trypan blue to evaluate the proportion of nonviable in untreated cells to 32.3% in treated cells and increased the cells in treated versus untreated populations. Concentrations of proportion of E-CA-12 cells in S phase from 22.8% in untreated triptolide Ն 50 ng/ml created Ͼ90% nonviable normal cells by cells to 30.5% in treated cells. Whereas the percentage of cells day 3 (Fig. 2B). By day 6, Ͼ90% of the normal cells treated with in S phase in untreated populations declined with time, the triptolide concentrations of Ն5 ng/ml were nonviable. Cell percentage of cells in S phase in treated populations remained viability declined by 47% after 6 days of treatment with 1 ng/ml elevated. triptolide. With E-CA-12 cells, Ͼ90% were already nonviable at Senescence of Cells Treated with Low Concentrations day 3 with concentrations of Ն10 ng/ml triptolide. Cell viability of Triptolide. We found that 1 ng/ml triptolide completely declined by 78% with 1 ng/ml triptolide at day 6 (data not inhibited clonal growth of normal and cancer-derived prostatic shown). epithelial cells (Fig. 1). However, inhibition was cell density Induction of Apoptosis by Triptolide. After observing dependent, and higher concentrations of triptolide were required the decline in cell viability caused by triptolide, particularly at to completely inhibit the growth of higher density cultures. We the higher concentrations, we evaluated the induction of apo- also observed that 1 ng/ml triptolide did not induce apoptosis in ptosis by triptolide (Fig. 3). For that purpose, E-CA-12 cells high density cultures (see previous sections and Fig. 3). We (inoculated at 105 cells/dish) were cultured for 24, 48, and 72 h considered the possibility that senescence might be involved in with or without triptolide, and cells with apoptotic morphology growth inhibition of prostatic epithelial cells by low doses of were visualized with Hoechst 33342 dye and propidium iodide. triptolide. SA-␤-gal assays were conducted to address this pos- Triptolide at 1 ng/ml did not cause apoptosis during the 3-day sibility. Semiconfluent cultures of normal and cancer-derived period of exposure. After 24 h of exposure to higher concentra- cells (E-PZ-10 and E-CA-13) were treated with or without 1 tions of triptolide (Ͼ50 ng/ml), a modest increase in the apo- ng/ml triptolide for up to 10 days. SA-␤-gal activity was found ptotic rate was detected (at maximum, 19% apoptosis with 100 to be minimal in untreated cells throughout the course of the ng/ml triptolide compared with 7.5% in control populations). experiment. In contrast, significant SA-␤-gal activity, as evi- However, at 48–72 h, the number of apoptotic cells in the denced by blue cytoplasmic staining, was noted in normal cells treated populations increased substantially, especially with 50 after 5 days of triptolide treatment (Fig. 4). SA-␤-gal activity and 100 ng/ml triptolide. Despite the preferential effect of became somewhat more pronounced after 10 days (data not triptolide on the decline of E-CA-10 cell viability, we did not shown). Cancer-derived cells exhibited relatively faint levels of

Fig. 4 SA-␤-gal expression in triptolide-treated cells. E-PZ-10 cells were treated with or without 1 ng/ml triptolide for 5 days and then stained for SA-␤-gal activity. Staining was minimal in untreated cultures but sub- stantial in treated populations. Cells with blue cytoplasmic SA-␤-gal staining appear in- tensely black in these black and white photos.

SA-␤-gal after 5 days of treatment with triptolide compared PZ-10 cells below the basal expression level of this protein at with normal cells. Blue staining was more evident after 10 days 24 h (Fig. 6B). The non-p53-regulated cell cycle inhibitor of treatment of cancer-derived cells (data not shown). p27Kip1 declined in triptolide-treated cells in a manner similar to Modulation of Signaling Pathways by Low Concentra- that of p21WAF1/CIP1 (Fig. 6B). No changes in the non-p53- tion of Triptolide. Triptolide at 1 ng/ml inhibited clonal regulated cell cycle inhibitor p16INK4a were found after triptol- growth as well as high-density cell proliferation after 6 days of ide treatment (Fig. 6B). Triptolide reduced levels of the anti- incubation. This concentration of triptolide also induced SA-␤- apoptotic protein bcl-2 after 48–72 h (data for the 72 time point gal, a marker of senescence. We wanted to further investigate not shown). The intensities of the bcl-2 bands normalized to the possible molecular pathways associated with the observed GAPDH were quantitated by scanning densitometry, and bcl-2 inhibition of cell proliferation and induction of senescence by was found to be reduced 5-fold in treated versus untreated cells low concentrations of triptolide. Expression of p53 protein and at 48 h. The decline in bcl-2 levels corresponded with 43% and INK4a WAF1/CIP1 Kip1 cell cycle regulators p16 , p21 , and p27 was 79% of cells undergoing apoptotic cell death at 48 and 72 h, investigated in E-CA-12 and E-PZ-10 cells after 3, 5, and 10 respectively, in response to 50 ng/ml triptolide (Figs. 3 and 6B). days of exposure to 1 ng/ml triptolide. These time points cor- No change in expression of the p53 target and proapoptotic responded to those at which expression of SA-␤-gal became protein bax was found after triptolide treatment (Fig. 6B). Ef- apparent. None of these proteins were altered in E-CA-12 (Fig. fects on signaling pathways (p21WAF1/CIP1, p27Kip1, p16INK4a, 5) or E-PZ-10 (data not shown) cells by 1 ng/ml triptolide. and bax) by triptolide in E-CA-12 cells were similar in E-PZ-10 Protein levels of p53 were also analyzed at earlier time points (6 cells (data not shown). The decline in bcl-2 levels was not so and 24 h after exposure to triptolide) in E-PZ-10 and E-CA-11 obvious in E-CA-12 cells as compared with E-PZ-10 cells cells. No changes in p53 protein levels were observed at these because the basal level of expression of bcl-2 in E-CA-12 cells time points (data not shown). was relatively low compared with that in E-PZ-10 cells. Modulation of Signaling Pathways by High Concentra- tion of Triptolide. Higher concentrations of triptolide were shown to induce apoptotic cell death. Because p53 may trigger DISCUSSION apoptosis, we investigated the expression of p53 in E-CA-12 Prostate cancer is the most common noncutaneous malig- and E-PZ-10 cells after 6, 24, 48, and 72 h of treatment with 50 nancy in American men and was responsible for an estimated ng/ml triptolide. Triptolide increased p53 levels in E-CA-12 200,000 new cases and over 31,000 deaths in the year 2001 (26). cells at 6 h, reaching a maximum after 24 and 48 h (Fig. 6A). Traditional approaches to localized disease are either radical Levels of p53 declined toward basal level after 72 h of exposure prostatectomy or radiation therapy. Chemotherapy for advanced to triptolide. Similar patterns of p53 expression were observed prostate cancer is still of limited efficacy, although the devel- in E-PZ-10 cells (Fig. 6B, 72 h time point not shown). Immu- opment of several promising new modalities is under way (27). nocytochemical analyses verified the nuclear localization of p53 Our previous studies on the role of p53 in mediating growth (data not shown). inhibition or apoptosis of prostate cancer cells by DNA damage- The levels of p53 target gene products (hdm-2, p21WAF1/CIP1, inducing drugs or ␥-irradiation led us to suggest that drugs that bax, and bcl-2) and other p53-independent cell cycle regulators work through p53-independent mechanisms might be most use- (p16INK4a and p27Kip1) were also analyzed in immunoblots. Trip- ful for treating prostate cancer, even those cancers that retain

tolide decreased levels of the intact, Mr 90,000 hdm-2 protein in wild-type p53. Previously, investigators reported that triptolide E-CA-12 and E-PZ-10 (data not shown) cells significantly after inhibited growth and induced apoptosis of HL-60 (promyelo- 24 h as detected by the hmd2-specific 2A10 (Fig. 6A) and SMP14 cytic leukemia) cancer cell lines in a p53-independent manner

mAbs (data not shown). Cleaved forms of hdm-2 (Mr 60,000) (4). We therefore decided to test the effects of triptolide on detected by the same hdm-2-specific antibodies increased steadily primary cultures of prostatic epithelial cells to see if this com- after 24 h of treatment with triptolide in E-CA-12 (Fig. 6A; results pound would show antitumor activity. A previous study of

with mAb 2A10 against Mr 60,000 hdm-2 not shown) and E-PZ-10 triptolide and the prostate cancer cell line LNCaP was limited in cells (data not shown). extent but indicated that triptolide was growth inhibitory, with WAF1/CIP1 Triptolide (50 ng/ml) reduced p21 levels in E- an ED50 of about 10 ng/ml (3).

Fig. 6 Molecular changes with high concentration of triptolide. A, p53 and hdm-2 in E-CA-12 cells in response to 50 ng/ml triptolide (ϩ)or diluent (Ϫ) were evaluated at 6–72 h by Western blot. Anti-hmd-2

2A10 mAb was used to detect the intact, Mr 90,000 hdm-2 protein (note that the intense band denoted with an asterisk is nonspecific). Anti-

hdm-2 SMP14 mAb was used to detect the Mr 60,000 cleaved form of hdm-2. Cell extracts were prepared, and the equivalent of 50 ␮g protein/ lane was subjected to SDS-PAGE. B, p53, p21WAF1/CIP1, p27Kip1, p16INK4a, bcl-2, and bax were evaluated by Western blot after incubation of E-PZ-10 cells for 6–48 h with 50 ng/ml triptolide (ϩ) or diluent (Ϫ). Cell extracts were prepared, and the equivalent of 50 ␮g protein/ lane was subjected to SDS-PAGE. The cell extract from diluent-treated Fig. 5 Western blot analysis of p53 and cell cycle regulators after E-PZ-10 cells was included as a control at each indicated time point. incubation of E-CA-12 cells for 3, 5, and 10 days with 1 ng/ml triptolide. Cells were fed fresh media with triptolide (ϩ) or with diluent (Ϫ) on days 0, 2, 4, 7, and 9. Cell extracts were prepared, and the equivalent of 50 or 80 ␮g (p21WAF1/CIP1) protein/lane was subjected to SDS-PAGE. Blots were prepared and labeled with antibodies against p53, p21WAF1/CIP1, p27Kip1, and p16INK4a (A). Each band was measured replicative senescence (30, 31). However, increased expression INK4a by scanning densitometry and normalized to GAPDH. In B, all relative of p16 has been associated with replicative senescence of values are presented as the ratio of the target protein in untreated to prostatic epithelial cells (30), which was not the case in the treated cells, normalized to t0 values. senescent-like cells induced by triptolide. A number of factors have been reported to induce a senescence-like state in cells, but it is not known whether this phenotype is exactly equivalent to Our results revealed that triptolide indeed exhibited many replicative senescence (28). The molecular signaling pathway antitumor activities against prostate cells, in an interesting dose- mediating triptolide-induced premature senescence of prostatic dependent manner. Low concentrations of triptolide (1 ng/ml) epithelial cells remains to be determined but does not appear to inhibited growth completely in clonal assays and partially in involve p53, p21WAF1/CIP1, p27Kip1,orp16INK4a. high cell density assays. Growth inhibition with 1 ng/ml trip- In normal cells, irreversible proliferative arrest may result tolide was accompanied by the induction of SA-␤-gal activity, a from terminal differentiation or replicative senescence. Pheno- widely used surrogate marker of senescence. In addition to typic alterations that resemble replicative senescence can be expression of SA-␤-gal, cells adopted several morphological induced in normal as well as in tumor cells by treatment with changes that have been associated with a senescence-like phe- different anticancer agents (29), as we saw with triptolide. notype (28, 29). These changes included an enlarged and flat- Induction of accelerated senescence that results in terminal tened shape and the development of vacuoles. growth arrest is proposed to be a programmed protective re- We did not observe any regulation of p53, p21WAF1/CIP1, sponse of normal cells to potentially carcinogenic insults (32). p27Kip1,orp16INK4a in prostatic epithelial cells in response to 1 The phenomenon of senescence has attracted increasing atten- ng/ml triptolide, despite the induction of a senescence-like phe- tion recently and is believed to be relevant to cancer prevention notype. In contrast to fibroblasts, prostatic epithelial cells do not and tumor suppression. The induction of senescence in primary have elevated levels of p53 or p21WAF1/CIP1 after undergoing cultures of prostatic epithelial cells suggests that the chemopre-

ventive activity of low doses of triptolide be further evaluated in reported to be unable to promote p53 degradation and may additional experimental models. function in a dominant negative fashion to stabilize p53 (36, 38). Higher concentrations of triptolide (15–100 ng/ml) effec- The relationship of cleaved hdm-2 to accumulation of p53 and tively inhibited high density cell growth, and 50–100 ng/ml apoptosis in prostatic epithelial cells in response to triptolide is triptolide induced apoptosis in the majority of cells after 3 days under investigation. of treatment. High doses of triptolide caused a number of Another interesting feature associated with the induction of molecular changes in prostatic epithelial cells in conjunction p53 by triptolide in prostatic epithelial cells and in other cells (6) with the induction of apoptosis. Interestingly, protein levels of was decreased expression of p21WAF1/CIP1. Reduction of p53 were significantly increased and predominantly accumu- p21WAF1/CIP1 protein levels by triptolide in the HT1080 (fibro- lated in the nuclei of prostatic epithelial cells. Although the sarcoma) cell line was reported to be due to transcriptional activity of triptolide was originally reported to be p53-independ- inhibition of p21WAF1/CIP1 (6). These same investigators at- ent, two more recent studies showed that triptolide-induced tempted to study the effect of triptolide on p21WAF1/CIP1 apoptosis may be at least partly mediated through p53 activation expression in p53 wild-type and p53-null mouse embryonic in cells with wild-type p53. Chang et al. (6) showed that fibroblasts to determine whether decreased expression of triptolide-mediated enhancement of chemotherapy-induced ap- p21WAF1/CIP1 was directly mediated by p53. However, basal optosis was accompanied by enhanced translation and accumu- expression of p21WAF1/CIP1 was almost absent in the p53-null lation of wild-type p53 protein in HT1080 (fibrosarcoma) cell cells, making this experiment impossible. However, they found lines. Jiang et al. (15) demonstrated that p53 expression was that triptolide did not affect p21WAF1/CIP1 expression in the up-regulated in AGS (gastric cancer) cell lines after triptolide p53-mutant HT29 colon cancer cell line, leading them to pro- treatment, and suppression of p53 expression by p53 antisense pose that down-regulation of p21WAF1/CIP1 was a p53-mediated oligonucleotides significantly abolished triptolide-induced ap- event (6). It has been reported that the p21WAF1/CIP1 protein can optosis. These investigators also found that MKN-28 and SGC- antagonize p53-mediated apoptosis because the induction of 7901 (gastric cancer) cell lines with mutant p53 did not show endogenous p21WAF1/CIP1 or overexpression of a p21WAF1/CIP1 any significant changes in growth or apoptotic rates after treat- transgene prevented apoptosis (39, 40). Therefore, down-regu- ment with triptolide. Taken together, these results suggest that lation of p21WAF1/CIP1 may contribute to the ability of triptolide triptolide may use multiple signaling pathways, some of which to induce apoptosis. However, it must be noted that apoptosis involve p53 in cells with wild-type p53, to arrest growth or and accumulation of p53 in AGS (gastric cancer) cells in re- induce apoptosis. sponse to triptolide were accompanied by up-regulation of It is of interest that up-regulation of p53 downstream target p21WAF1/CIP1 (15). This discrepancy suggests that the mechanism genes (such as hdm-2, p21WAF1/CIP1, or bax) was not seen in by which p53 induces apoptosis and the role of p21WAF1/CIP1 in conjunction with up-regulation of p53 by triptolide in our pri- this process are dependent on multiple different factors and are mary cultures of prostatic cells. This is in contrast to the up- cell type specific (23, 41). regulation of these genes that generally accompanies p53- Apoptosis promoted by p53 has also been associated with mediated growth arrest and/or apoptosis in response to DNA regulation of bcl-2 protein family members, including induction damage (33). Rather, protein levels of several p53 target genes of proapoptotic bax and repression of antiapoptotic bcl-2 (42). A were reduced after triptolide treatment. For example, our results p53 response element has been mapped to the bcl-2 P2 minimal

showed reduction of hdm-2 (the intact Mr 90,000 form) after promoter. The exact mechanism of p53-mediated repression is triptolide treatment. Given that hdm-2 targets p53 for degrada- not clear, but it has been proposed that the interaction between tion (34), the reduction in hdm-2 protein could partially provide p53 and the TATA-binding protein is most likely responsible for an explanation for the observed sustained elevation of p53. the repression of the bcl-2 promoter (43). Our results showed

Reduction of Mr 90,000 hdm-2 was detected by two antibodies, that apoptosis was associated with reduction of antiapoptotic 2A10 and SMP14. The former antibody does not bind hdm-2 bcl-2 protein in prostatic epithelial cells, whereas protein levels protein that is phosphorylated in the middle portion of the of the proapoptotic protein bax remained unchanged. The in- molecule (35). However, the fact that lower levels of hdm-2 creased ratio of bax:bcl-2 may have a role in the induction of were also detected by SMP14, whose reactivity with hdm-2 is apoptosis by triptolide. not altered by phosphorylation, suggests that total protein was Triptolide also reduced expression of p27Kip1, which is not decreased. a direct target of p53. Reduction in p27Kip1 protein can be due SKP2 As the intact, Mr 90,000 form of hdm-2 protein decreased to S-phase block and activity of ubiquitin protein ligase p45 Kip1 in response to triptolide, a Mr 60,000 form of hdm-2 accumu- that is known to promote p27 degradation and induction of lated. It has been reported that mdm-2, the mouse homologue of S phase (44, 45). Therefore, the reduced levels of p27Kip1 that hdm-2, is a substrate for proteases involved in apoptosis that we observed may be a consequence of the S-phase arrest in- share specificity with CPP32 (caspase-3). These proteases duced in prostatic cells by triptolide. Alternatively, reduction of Kip1 cleave mdm-2 at residue 361, generating a Mr 60,000 fragment p27 protein levels could reflect caspase activity in conjunc- (36, 37). More recently, work from the same investigators tion with apoptosis (46). showed that a distinct caspase activity for hdm-2 was induced It is tempting to speculate that p53 plays a role in induction by p53 before the onset of apoptosis in H1299 cells expressing of apoptosis in prostatic epithelial cells after triptolide treatment, a temperature-sensitive human p53 (38). The p53 binding and but further investigation is required to determine whether up- inhibitory functions of hdm-2 have not been reported to be regulation of p53 is coincidental or instrumental in this process. affected by the cleavage. However, cleaved hdm-2 has been Certainly p53 is not essential for the activity of triptolide in

prostate cells because regression of the prostate cancer cell line cell death is reflective of cancer in vivo, then any compound PC-3 with mutated p53 (47) occurred in mice treated with capable of inducing apoptosis in primary cultures may be par- PG490-88 (48). However, it is remarkable that triptolide is one ticularly worthy of further investigation. If Phase I trials of of the few agents that we have found to induce p53 in primary triptolide that are currently being initiated show safety, as might cultures of prostatic epithelial cells. DNA-damaging drugs and be anticipated because of the long-time use of the parental herb ␥-irradiation do not induce p53 in these cells, and the signaling in traditional medicine, then Phase II/III trials of this drug pathway used by these agents to stabilize and activate p53 against prostate cancer should be considered. appears nonfunctional in prostatic cells. However, different types of stresses initiate divergent signaling pathways to activate ACKNOWLEDGMENTS p53 (41), so we may have identified an alternative functional We thank Dr. Glenn Rosen for helpful discussions and critical pathway in prostatic epithelial cells that leads to induction reading of the manuscript. of p53. Regulation of expression of the p53 target genes hdm-2, p21WAF1/CIP1, and bcl-2 in conjunction with up-regulation of p53 in prostatic cells suggests but does not prove p53 REFERENCES activity. We must keep in mind that accumulation of nuclear p53 1. Ramgolam, V., Ang, S. G., Lai, Y. H., Loh, C. S., and Yap, H. K. protein, as we noted in prostatic cells treated with triptolide, Traditional Chinese medicines as immunosuppressive agents. Ann. Acad. Med. Singapore, 29: 11–16, 2000. may not be sufficient to initiate downstream signaling by p53. 2. Kupchan, S. M., Court, W. A., Dailey, R. G., Jr., Gilmore, C. J., and Generally, p53 must undergo specific posttranslational modifi- Bryan, R. F. Triptolide and tripdiolide, novel antileukemic diterpenoid cations to enable it to bind DNA and transactivate downstream triepoxides from Tripterygium wilfordii. J. Am. Chem. Soc., 94: 7194– gene targets (41). On the other hand, certain downstream effects 7195, 1972. of p53, particularly those leading to apoptosis, are known to 3. Shamon, L. A., Pezzuto, J. M., Graves, J. M., Mehta, R. R., Wang- occur in the absence of transcriptional activity of p53 (23). charoentrakul, S., Sangsuwan, R., Chaichana, S., Tuchinda, P., Cleason, It is of note that accumulation of p53 protein and S-phase P., and Reutrakul, V. Evaluation of the mutagenic, cytotoxic, and antitumor potential of triptolide, a highly oxygenated diterpene isolated arrest, as we observed in prostate cells treated with triptolide, from Tripterygium wilfordii. Cancer Lett., 112: 113–117, 1997. have been described in other types of cells in which DNA 4. Wei, Y. S., and Adachi, I. Inhibitory effect of triptolide on colony synthesis has been blocked. Agents such as hydroxyurea, which formation of breast and stomach cancer cell lines. Zhongguo Yaoli inhibits the activity of ribonucleotide reductase, or aphidicolin, Xuebao, 12: 406–410, 1991. which blocks DNA polymerase ␣, cause an early S-phase arrest 5. Lee, K. Y., Chang, W., Qiu, D., Kao, P. N., and Rosen, G. D. PG490 independent of p53. However, p53 protein levels increase in (triptolide) cooperates with tumor necrosis factor-␣ to induce apoptosis in tumor cells. J. Biol. Chem., 274: 13451–13455, 1999. cells treated with these compounds. Upon closer scrutiny, it has 6. Chang, W. T., Kang, J. J., Lee, K. Y., Wei, K., Anderson, E., been discovered that p53 is functionally impaired, and only Gotmare, S., Ross, J. A., and Rosen, G. D. Triptolide and chemotherapy some p53 targets are efficiently induced when cells are arrested cooperate in tumor cell apoptosis. A role for the p53 pathway. J. Biol. in S phase (49). The similarities between the effects of triptolide Chem., 276: 2221–2227, 2001. on prostate cells and hydroxyurea or aphidicolin on other types 7. Krishna, G., Liu, K., Shigemitsu, H., Gao, M., Raffin, T. A., and of cells suggest that it would be worthwhile to further investi- Rosen, G. D. PG490-88, a derivative of triptolide, blocks bleomycin- gate the possibility that triptolide blocks DNA replication. induced lung fibrosis. Am. J. Pathol., 158: 997–1004, 2001. The dose-dependent effects of triptolide that we noted are 8. Zhao, G., Vaszar, L. T., Qiu, D., Shi, L., and Kao, P. N. Anti- inflammatory effects of triptolide in human bronchial epithelial cells. comparable with those of certain other compounds with antitu- Am. J. Physiol. Lung Cell. Mol. Physiol., 279: L958–L966, 2000. mor activity. At low concentrations, the microtubule inhibitor 9. Qiu, D., Zhao, G., Aoki, Y., Shi, L., Uyei, A., Nazarian, S., Ng, J. C., paclitaxel was reported to induce mitotic arrest in several types and Kao, P. N. Immunosuppressant PG490 (triptolide) inhibits T-cell of cancer cells, but at high concentrations, paclitaxel triggered interleukin-2 expression at the level of purine-box/nuclear factor of rapid apoptosis (50). Another example is the induction of se- activated T-cells and NF-␬B transcriptional activation. J. Biol. Chem., 274: 13443–13450, 1999. nescence in normal human fibroblasts by low concentrations of 10. el-Deiry, W. S., Tokino, T., Velculescu, V. E., Levy, D. B., Parsons, hydrogen peroxide and the induction of apoptosis by high con- R., Trent, J. M., Lin, D., Mercer, W. E., Kinzler, K. W., and Vogelstein, centrations of hydrogen peroxide (51). The effects of different B. WAF1, a potential mediator of p53 tumor suppression. Cell, 75: doses of triptolide on prostatic epithelial cells might be ex- 817–825, 1993. ploited for different applications. The moderate growth-inhibi- 11. Amundson, S. A., Myers, T. G., and Fornace, A. J., Jr. Roles for p53 tory activity and induction of senescence by low doses of in growth arrest and apoptosis: putting on the brakes after genotoxic triptolide might be appropriate for chemopreventive strategies stress. Oncogene, 17: 3287–3299, 1998. against prostate cancer. The fact that normal as well as cancer- 12. Jimenez, G. S., Khan, S. H., Stommel, J. M., and Wahl, G. M. p53 regulation by post-translational modification and nuclear retention in derived cells responded to triptolide is supportive of this possi- response to diverse stresses. Oncogene, 18: 7656–7665, 1999. bility. Induction of senescence is especially relevant to slowly 13. Girinsky, T., Koumenis, C., Graeber, T. G., Peehl, D. M., and proliferating prostate cells that do not respond well to agents Giaccia, A. J. Attenuated response of p53 and p21 in primary cultures of that directly target replication. The proapoptotic activity of human prostatic epithelial cells exposed to DNA-damaging agents. higher doses of triptolide might be more suitable for chemother- Cancer Res., 55: 3726–3731, 1995. apeutic applications. It is notable that primary cultures of pros- 14. Navone, N. M., Troncoso, P., Pisters, L. L., Goodrow, T. L., Palmer, J. L., Nichols, W. W., von Eschenbach, A. C., and Conti, C. J. tatic cancer cells, in contrast to established cell lines, are very p53 protein accumulation and gene mutation in the progression of resistant to apoptosis. We have identified few agents capable of human prostate carcinoma. J. Natl. Cancer Inst. (Bethesda), 85: 1657– causing apoptosis in these cells; therefore, if this resistance to 1669, 1993.

15. Jiang, X. H., Wong, B. C., Lin, M. C., Zhu, G. H., Kung, H. F., 33. el-Deiry, W. S. Regulation of p53 downstream genes. Semin. Can- Jiang, S. H., Yang, D., and Lam, S. K. Functional p53 is required for cer Biol., 8: 345–357, 1998. ␬ triptolide-induced apoptosis and AP-1 and nuclear factor- B activation 34. Haupt, Y., Maya, R., Kazaz, A., and Oren, M. Mdm2 promotes the in gastric cancer cells. Oncogene, 20: 8009–18, 2001. rapid degradation of p53. Nature (Lond.), 387: 296–299, 1997. 16. Schmid, H. P., and McNeal, J. E. An abbreviated standard proce- 35. Khosravi, R., Maya, R., Gottlieb, T., Oren, M., Shiloh, Y., and dure for accurate tumor volume estimation in prostate cancer. Am. J. Shkedy, D. Rapid ATM-dependent phosphorylation of MDM2 precedes Surg. Pathol., 16: 184–191, 1992. p53 accumulation in response to DNA damage. Proc. Natl. Acad. Sci. 17. Peehl, D. M. Culture of human prostatic epithelial cells. In: R. I. USA, 96: 14973–14977, 1999. Freshney (ed.), Culture of Epithelial Cells, pp. 159–180. New York, 36. Chen, L., Marechal, V., Moreau, J., Levine, A. J., and Chen, J. NY: Wiley-Liss, Inc., 1992. Proteolytic cleavage of the mdm2 oncoprotein during apoptosis. J. Biol. 18. Peehl, D. M. Epithelial cell culture: prostate. In: A. Atala and R. Chem., 272: 22966–22973, 1997. Lanza (eds.), Methods of Tissue Engineering, pp. 247–261. San Diego, 37. Erhardt, P., Tomaselli, K. J., and Cooper, G. M. Identification of the CA: Academic Press, 2002. MDM2 oncoprotein as a substrate for CPP32-like apoptotic proteases. 19. Brothman, A. R., Patel, A. M., Peehl, D. M., and Schellhammer, J. Biol. Chem., 272: 15049–15052, 1997. P. F. Analysis of prostatic tumor cultures using fluorescence in-situ hybridization (FISH). Cancer Genet. Cytogenet., 62: 180–5, 1992. 38. Pochampally, R., Fodera, B., Chen, L., Lu, W., and Chen, J. Activation of an MDM2-specific caspase by p53 in the absence of 20. Hsu, J. Y., Feldman, D., McNeal, J. E., and Peehl, D. M. Reduced apoptosis. J. Biol. Chem., 274: 15271–15277, 1999. 1␣-hydroxylase activity in human prostate cancer cells correlates with 39. Bissonnette, N., and Hunting, D. J. p21-induced cycle arrest in G1 decreased susceptibility to 25-hydroxyvitamin D3-induced growth inhibition. Cancer Res, 61: 2852–6, 2001. protects cells from apoptosis induced by UV-irradiation or RNA polymerase II blockage. Oncogene, 16: 3461–3469, 1998. 21. Guo, X., Knudsen, B. S., Peehl, D. M., Ruiz, A., Bok, D., Rando, R. R., Rhim, J. S., Nanus, D. M., and Gudas, L. J. Retinol metabolism 40. Gorospe, M., Cirielli, C., Wang, X., Seth, P., Capogrossi, M. C., and lecithin:retinol acyltransferase levels are reduced in cultured human and Holbrook, N. J. p21(Waf1/Cip1) protects against p53-mediated prostate cancer cells and tissue specimens. Cancer Res, 62: 1654–61, apoptosis of human melanoma cells. Oncogene, 14: 929–935, 1997. 2002. 41. Giaccia, A. J., and Kastan, M. B. The complexity of p53 modula- 22. Tsao, M. C., Walthall, B. J., and Ham, R. G. Clonal growth of tion: emerging patterns from divergent signals. Genes Dev., 12: 2973– normal human epidermal keratinocytes in a defined medium. J. Cell. 2983, 1998. Physiol., 110: 219–229, 1982. 42. Miyashita, T., Krajewski, S., Krajewska, M., Wang, H. G., Lin, 23. Koumenis, C., Alarcon, R., Hammond, E., Sutphin, P., Hoffman, H. K., Liebermann, D. A., Hoffman, B., and Reed, J. C. Tumor sup- W., Murphy, M., Derr, J., Taya, Y., Lowe, S. W., Kastan, M., and pressor p53 is a regulator of bcl-2 and bax gene expression in vitro and Giaccia, A. Regulation of p53 by hypoxia: dissociation of transcriptional in vivo. Oncogene, 9: 1799–805, 1994. repression and apoptosis from p53-dependent transactivation. Mol. Cell. 43. Wu, Y., Mehew, J. W., Heckman, C. A., Arcinas, M., and Boxer, Biol., 21: 1297–1310, 2001. L. M. Negative regulation of bcl-2 expression by p53 in hematopoietic 24. Dimri, G. P., Lee, X., Basile, G., Acosta, M., Scott, G., Roskelley, cells. Oncogene, 20: 240–51, 2001. C., Medrano, E. E., Linskens, M., Rubelj, I., Pereira-Smith, O., et al.A 44. Sutterluty, H., Chatelain, E., Marti, A., Wirbelauer, C., Senften, M., biomarker that identifies senescent human cells in culture and in aging Muller, U., and Krek, W. p45SKP2 promotes p27Kip1 degradation and skin in vivo. Proc. Natl. Acad. Sci. USA, 92: 9363–9367, 1995. induces S phase in quiescent cells. Nat. Cell. Biol., 1: 207–214, 1999. 25. Chen, J., Marechal, V., and Levine, A. J. Mapping of the p53 and 45. Carrano, A. C., Eytan, E., Hershko, A., and Pagano, M. SKP2 is mdm-2 interaction domains. Mol. Cell. Biol., 13: 4107–4114, 1993. required for ubiquitin-mediated degradation of the CDK inhibitor p27. 26. Greenlee, R. T., Hill-Harmon, M. B., Murray, T., and Thun, M. Nat. Cell. Biol., 1: 193–199, 1999. Cancer statistics, 2001. CA Cancer J. Clin., 51: 15–36, 2001. 46. Levkau, B., Koyama, H., Raines, E. W., Clurman, B. E., Herren, B., 27. Wang, J., and Waxman, J. Chemotherapy for prostate cancer. Urol. Orth, K., Roberts, J. M., and Ross, R. Cleavage of p21Cip1/Waf1 and Oncol., 5: 93–96, 2000. p27Kip1 mediates apoptosis in endothelial cells through activation of 28. Young, J., and Smith, J. R. Epigenetic aspects of cellular senes- Cdk2: role of a caspase cascade. Mol. Cell, 1: 553–563, 1998. cence. Exp. Gerontol., 35: 23–32, 2000. 47. Carroll, A. G., Voeller, H. J., Sugars, L., and Gelmann, E. P. p53 29. Chang, B. D., Broude, E. V., Dokmanovic, M., Zhu, H., Ruth, A., oncogene mutations in three human prostate cancer cell lines. Prostate, Xuan, Y., Kandel, E. S., Lausch, E., Christov, K., and Roninson, I. B. 23: 123–34, 1993. A senescence-like phenotype distinguishes tumor cells that undergo 48. Li, K. K., and Filder, J. M. PG490–88 exerts potent anticancer terminal proliferation arrest after exposure to anticancer agents. Cancer activity alone and in combination therapy in a nude mouse xenograft Res., 59: 3761–3767, 1999. model. Proc. Am. Assoc. Cancer Res., 42: 73, 2001. 30. Sandhu, C., Peehl, D. M., and Slingerland, J. p16INK4A mediates 49. Gottifredi, V., Shieh, S., Taya, Y., and Prives, C. From the Cover: cyclin dependent kinase 4 and 6 inhibition in senescent prostatic epi- p53 accumulates but is functionally impaired when DNA synthesis is thelial cells. Cancer Res., 60: 2616–2622, 2000. blocked. Proc Natl Acad Sci USA, 98: 1036–41, 2001. 31. Jarrard, D. F., Sarkar, S., Shi, Y., Yeager, T. R., Magrane, G., 50. Wang, T. H., Wang, H. S., and Soong, Y. K. Paclitaxel-induced cell Kinoshita, H., Nassif, N., Meisner, L., Newton, M. A., Waldman, F. M., death: where the cell cycle and apoptosis come together. Cancer (Phila.), and Reznikoff, C. A. p16/pRb pathway alterations are required for 88: 2619–2628, 2000. bypassing senescence in human prostate epithelial cells. Cancer Res., 51. Chen, Q. M., Liu, J., and Merrett, J. B. Apoptosis or senescence-like 59: 2957–2964, 1999. growth arrest: influence of cell-cycle position, p53, p21 and bax in H2O2 32. Weinberg, R. A. The cat and mouse games that genes, viruses, and response of normal human fibroblasts. Biochem. J., 347: 543–551, cells play. Cell, 88: 573–5, 1997. 2000.

Downloaded from clincancerres.aacrjournals.org on October 1, 2021. © 2002 American Association for Cancer Research. Antiproliferative and Proapoptotic Activities of Triptolide (PG490), a Natural Product Entering Clinical Trials, on Primary Cultures of Human Prostatic Epithelial Cells

Taija M. Kiviharju, Philip S. Lecane, Robert G. Sellers, et al.

Clin Cancer Res 2002;8:2666-2674.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/8/8/2666

Cited articles This article cites 48 articles, 18 of which you can access for free at: http://clincancerres.aacrjournals.org/content/8/8/2666.full#ref-list-1

Citing articles This article has been cited by 12 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/8/8/2666.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/8/8/2666. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.