Apoptosis induction through proteasome inhibitory activity of cucurbitacin D in human T-cell leukemia
Abstract
BACKGROUND:
Human T-cell leukemia is an aggressive malignancy of T lymphocytes. T-cell leukemia has a very poor prognosis, even with intensive chemotherapy, indicating the need for development of new drugs to treat the disease. Triterpenoid cucurbitacins have been shown to have antitumor activity, but the mechanism of this activity is not fully understood.
METHODS:
The effects of cucurbitacin D on the proliferation and apoptotic induction of T-cell leukemia cells using the Cell viability assay and Annexin V staining were evaluated. To investigate the mechanisms of apoptosis, antiapoptotic protein, NF-κB, and the proteasome activity of leukemia cells treated with cucurbitacin D were evaluated by Western blotting both in vitro and in vivo.
RESULTS:
In this study, cucurbitacin D was found to inhibit proliferation and to induce apoptosis of T-cell leukemia cells. Constitutively activated NF-κB was inhibited by cucurbitacin D in the nucleus, which resulted in accumulation of NF-κB in the cytoplasm, leading to down-regulation of the expression of antiapoptotic proteins Bcl-xL and Bcl-2. Furthermore, cucurbitacin D induced the accumulation of inhibitor of NF-κB (IκB)α by inhibition of proteasome activity. Low doses of cucurbitacin D synergistically potentiated the antiproliferative effects of the histone deacetylase inhibitor VPA. Finally, the proapoptotic and proteasome inhibitory activities of cucurbitacin D also were demonstrated using SCID mice in an in vivo study.
CONCLUSIONS:
Cucurbitacin D induced apoptosis through suppression of proteasome activity both in vitro and in vivo, making cucurbitacin D a promising candidate for clinical applications in the treatment of T-cell leukemia. Cancer 2011;. © 2010 American Cancer Society.
Human T-cell leukemia is a highly aggressive disease with a poor prognosis; the median survival of patients in Japan is 6 months.1 The high mortality of patients with this disease, even after intensive chemotherapy, results from many factors such as rapid growth of large tumor burden and frequent infectious complication.1 Therefore, development of new drugs for this aggressive disease is urgently needed.
Cucurbitacins are a group of triterpenoids isolated from plant families such as Cucurbitaceae and Cruciferae, which have been shown to have anti-inflammatory activity and anticancer effects on various tumor types.2 Chemically, the cucurbitacins are highly diverse, characterized by a tetracyclic cucurbitane nucleus skeleton with a variety of oxygenation functionalities at different positions. Traditionally, cucurbitacin family members are divided into 12 categories such as cucurbitacin D, I, and Q.3 Cucurbitacin I and Q suppress proliferation of tumor cells through inhibition of STAT3 phosphorylation.4, 5 Recently, we isolated cucurbitacin D from Trichosanthes kirillowii and showed that this compound can induce apoptosis in human hepatocellular carcinoma cells in vitro, although the precise mechanism still remains to be elucidated.6
Induction of tumor cell apoptosis by chemotherapy is important for treatment of cancer. Activated caspase-3 is an essential mediator of apoptosis.7 The Bcl-2 family is composed of both pro- and antiapoptotic proteins. The suppression of antiapoptotic proteins such as Bcl-xL and Bcl-2 can induce apoptosis.8, 9
NF-κB has an important effect on cell growth and inhibition of apoptosis, and constitutive activation of NF-κB promotes proliferation and tumorigenesis in adult T-cell leukemia (ATL) cells.10 NF-κB activation is regulated by its upstream kinase, the IKK complex, which is composed of 2 catalytic subunits, IKK1/α and IKK2/β, and a scaffolding protein, NF-κB essential modulator (NEMO). In resting T cells, NF-κB forms a tight complex with inhibitor of NF-κB (IκBα, IκBβ, IκBε), which inhibits the nuclear translocation of NF-κB. Activated IKK phosphorylates IκB, which triggers ubiquitination and proteasome-dependent degradation of IκB. Active NF-κB is then translocated to the nucleus, where it regulates an array of genes responsible for cell survival and growth.11 Inhibitors of NF-κB activity have been reported to effectively induce apoptosis in ATL cell lines and primary ATL cells.12
The proteasome is an abundant catalytic complex found in both the nucleus and cytoplasm of eukaryotic cells. Proteasome-mediated degradation plays an essential role in the regulation of most intracellular proteins such as NF-κB.13 Recently, proteasome inhibitors have been used as new anticancer therapies. Bortezomib is the first proteasome inhibitor to enter clinical development as a new therapy for treatment of cancer patients.14 Bortezomib induces cell-cycle blockage and cell apoptosis in several cancer cell lines through suppression of NF-κB activity.15 Furthermore, bortezomib mediates anti-ATL activity by inducing caspase activation, suppressing the expression of the antiapoptotic protein XIAP.16
Combination with other inhibitor is important for the chemotherapy effect toward malignancy.17 Histone deacetylase (HDAC) inhibitors block the activity of histone deacetylase, which is involved in the expression of a variety of genes responsible for cell-cycle progression or induction of apoptosis.18 Numerous HDAC inhibitors have been found to inhibit cell proliferation and induce apoptosis. Recently, many studies demonstrated proteasome inhibitor to interact synergistically with HDAC inhibitors to induce apoptosis of leukemia cells.19, 20
In this article, we show that cucurbitacin D inhibits proliferation and induces apoptosis in human T-cell leukemia cell lines with reduced expression of Bcl-xL and Bcl-2. Furthermore, cucurbitacin D inhibits proteasome activity, resulting in attenuation of NF-κB translocation to the nucleus. In addition, cucurbitacin D correlates with HDAC inhibitor to induce synergistic apoptosis in leukemia cells.
MATERIALS AND METHODS
Cell Lines and Culture Conditions
The human T-cell leukemia cell lines MT-1, MT-2, MT-4, and Hut102 were kindly given by Dr. Mori (Ryukyu University, Naha, Japan). Hut78 cells were from the Cell Resource Center of Tohoku University (Sendai, Japan). Acute lymphoblastic leukemia Jurkat cells were obtained from RIKEN BioResource Center (Tsukuba, Japan). All cells were grown in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (Bio Whittaker, Walkersville, MD), l-glutamine, and penicillin–streptomycin in humidified 5% CO2 at 37°C.
Reagents and Antibodies
Cucurbitacin D was purchased from Nacalai Tesque (Kyoto, Japan) and dissolved in 10% ethanol to a stock concentration of 1 mg/mL. Antibodies for NF-κB p65 (sc-109, sc-372), Bcl-xL (sc-7195), Bcl-2 (sc-492), IκBα (sc-371), ubiquitin (sc-9133), Oct1 (sc-232), and Protein A/G PLUS-Agarose (sc-2003) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies for phospho-IKKα/IKKβ (#2681) were obtained from Cell Signaling Technology (Danvers, MA). Anti-β-actin antibody (A5441) and valproic acid sodium (P4543) were generated by Sigma (St. Louis, MO). HRP-conjugated antirabbit IgG (#458), HRP-conjugated antimouse IgG (NA931VS), Alexa Fluor–labeled rabbit antigoat antibody (A11078), and MG132 were purchased from MBL (Nagoya, Japan), Amersham Pharmacia Bioscience (Buckinghamshire, UK), Invitrogen (Carlsbad, CA), and Calbiochem (San Diego, CA), respectively.
Preparation of Human Peripheral Blood Lymphocytes
Peripheral blood lymphocytes were freshly prepared from 20 mL of EDTA-blood from healthy volunteers. Cells were isolated using Lymphoprep (Axis Shield, Oslo, Norway) according to the protocol from the manufacturer.
Cell Viability Assay
Cell proliferation was measured using a Cell Titer-Glo luminescent cell viability assay system (Promega, Madison, WI). Leukemia cell lines were plated in 96-well plates at a density of 1 × 104/well and incubated with various concentrations of cucurbitacin D for the indicated periods. After culture, 20 μL of the assay reagent was added to each well, and cell lysates were incubated on an orbital shaker for 30 seconds. Fluorescence was measured by luminescencer-JNR-II (ATTO, Tokyo, Japan).
Annexin V Staining and Detection of Apoptotic Cells
Apoptotic cells were measured by annexin V-FITC staining. Briefly, 1 × 106 cells were treated with cucurbitacin D for the indicated periods, washed twice in cold PBS, and incubated with 10 μL of Annexin V-FITC for 30 minutes. Samples were analyzed by an EPICS XL flow cytometer (Beckman Coulter, Fullerton, CA). Apoptotic cells were also detected using an APOPercentage Apoptosis Assay Kit (Biocolor, Belfast, Northern Ireland) via microscopic observation according to the manufacturer's instructions.
Western Blotting and Immunoprecipitation
MT-4 and Hut102 cells were lysed with RIPA lysis buffer21 for preparation of whole-cell extracts. Equivalent amounts of protein (10 μg) were resolved on SDS-PAGE gels, transferred and immobilized on nitrocellulose membranes (Amersham, Buckinghamshire, UK), and probed with appropriate primary and secondary antibodies. Immunodetection was performed using a chemiluminescence detection system (Alpha Innotech, San Leandro, CA). For immunoprecipitation, samples were incubated with immunoprecipitation antibody and Protein A/G PLUS-Agarose overnight. After washing 3 times with cold PBS, the precipitated proteins were eluted with sample buffer (0.5M Tris, 10% SDS, and bromophenol blue in 50% glycerol).
Transfection and Luciferase Assay
MT-4 and Hut102 cells were transfected with the transfection reagent Lipofectamine 2000 Reagent (Invitrogen, Carlsbad, CA) at 3 μL of reagent per microgram of DNA, as previously reported.22 Briefly, the Bcl-xL or NF-κB luciferase reporter construct and expression vectors were added to 90% confluent cells in 24-well culture plates. Eighteen hours after transfection, cells were stimulated with cucurbitacin D. After an additional 6 hours of incubation, the cells were lysed with Passive Lysis Buffer (Promega, Madison, WI). Luciferase activity was measured using a luminescencer-JNR-II.
Preparation of Nuclear Extracts and Cytoplasmic Fractions
After treatment with cucurbitacin D for the indicated periods, nuclear extracts and cytoplasm fractions were prepared from Hut102 and MT-4 cells using a NucBuster protein Extraction Kit according to the manufacturer's instructions (Novagen, Darmstadt, Germany).
Immunocytochemical Analysis
Hut102 cells were inoculated at 1 × 105 cells per well in 4-well chamber slides (BD Falcon, Bedford, MA). After 6 hours of stimulation with cucurbitacin D, cells were washed and fixed with 4% paraformaldehyde. After washing, cells were permeabilized with 0.5% saponin in 1% BSA in PBS for 15 minutes. Cells were stained with anti-NF-κB antibody (1:1000) for 1 hour at room temperature. After washing with PBS, the slides were incubated for 1 hour with Alexa Fluor 488 rabbit antigoat antibody (1:1000) at room temperature. Thereafter, slides were washed with PBS and mounted with coverslips. The immunostained slides were observed using a confocal fluorescence microscope (Carl Zeiss, Jena, Germany).
Proteasome Activity Assay
A proteasome activity assay was performed following cucurbitacin D treatment of tumor cells for the indicated period. To measure proteasome activity, cultured cells were lysed in Passive Lysis Buffer (Promega, Madison, WI), and 50 μL of cell extract was incubated with 50 μL of fluorogenic substrate (Proteasome-Glo 3-Substrate System; Promega, Madison, WI) on an orbital shaker for 30 minutes at room temperature. Proteasome activity was measured using a luminescencer-JNR-II.
In Vivo Studies
Eight- to 12-week-old female CB.17/severe combined immunodeficiency disease (SCID) mice were purchased from CLEA Japan (Tokyo, Japan). The SCID mice were injected intraperitoneally with 4 × 106 MT-4 or Hut102 cells in 0.2 mL of PBS. Viability of the injected cells was greater than 90%, as assessed by trypan blue exclusion. After injection of tumor cells for 1 week, mice were treated intraperitoneally with vehicle or cucurbitacin D (1 or 4 mg/kg). Twenty-four hours after cucurbitacin D injection, tumor cells were collected from the peritoneal cavity. All animal experiments were performed according to the guidelines for the care and use of animals, which was approved by the University of Occupational and Environmental Health, Japan.
Statistical Analysis
All experiments were repeated more than 3 times, and representative results are shown in the figures. Results are expressed as means ± SDs. Statistical analysis was performed using the Student t test. A confidence level of P < .05 was considered significant.
RESULTS
Cucurbitacin D Inhibits Proliferation and Induces Apoptosis of Human T-Cell Leukemia Cells
We first investigated the effects of cucurbitacin D on the proliferation of T-cell leukemia cell lines MT-1, MT-2, MT-4, Hut102, Hut78, and Jurkat. Cell proliferation was inhibited by cucurbitacin D in a dose-dependent manner, as shown by a cell viability assay (Fig. 1A). MT-4 showed the highest sensitivity to cucurbitacin D treatment in 6 kinds of leukemia cell lines. Viable cell counts using Trypan Blue staining also showed that cucurbitacin D inhibited the cell growth of leukemia cells (data not shown). Because each cell line was sensitive to cucurbitacin D at a concentration of 1 μg/mL, this concentration was used for subsequent experiments. Importantly, the viability of human peripheral blood lymphocytes (PBLs) was not affected by cucurbitacin D (data not shown). Next, we examined the time course effect of cucurbitacin D. As shown in Figure 1B, cucurbitacin D showed no effect within 6 hours but effectively inhibited the growth of tumor cells from 12 hours in a time-dependent manner. MT-4 and Hut102 cells were used for the following experiments to explore the mechanism.
Cucurbitacin D inhibited proliferation and induced apoptosis of human T-cell leukemia cells. (A) MT-1, MT-2, MT-4, Hut102, Hut78, and Jurkat cells (1 × 104) were treated with the indicated concentrations of cucurbitacin D or ethanol for 24 hours. (B) T-cell leukemia cell lines were treated with cucurbitacin D (1 μg/mL) or ethanol (0.01%) for the indicated periods. The viability of cells was determined using the Cell Titer-Glo luminescent cell viability assay. Results are expressed as the mean relative activity and SD from triplicate cultures. (C) MT-4 and Hut102 cells (1 × 106) were cultured with cucurbitacin D (1 μg/mL) or ethanol (0.01%) for the indicated periods, and apoptotic cells were analyzed by flow cytometry. (D) Cells undergoing apoptosis were detected by uptake of a purple dye (APOPercentage Apoptotic Assay). Ethanol served as the negative control, and MG132 (0.5 μg/mL) served as the positive control. The percentage of apoptotic cells was calculated by counting stained cells in selected areas (*significantly increased compared with control at P < .05). Results are representative of at least 3 similar experiments.
To clarify the mechanisms by which cucurbitacin D induced cell death, annexin V levels were analyzed by flow cytometry because annexin V is known as a marker for apoptosis. As shown in Figure 1C, the leukemia cells expressed annexin V after the presence of cucurbitacin D for 12 hours. Forty-eight hours after cucurbitacin D treatment, about 60% of the cells were positive. To further examine apoptosis induced by cucurbitacin D, apoptotic cells stained with purple dye were counted by microscopy. As shown in Figure 1D, about 35% of the MT-4 cells and 25% of the Hut102 cells treated with cucurbitacin D for 24 hours were apoptotic. The proteasome inhibitor MG132 was used as a positive control. This result was consistent with that shown in Figure 1A,B, that MT-4 was more sensitive to cucurbitacin D than were Hut102 cells. Treatment with cucurbitacin D also led to the appearance of a significant population of cells in the sub-G1 phase (data not shown). Apoptosis was totally induced by cucurbitacin D.
Expressions of Bcl-2 Family Proteins Are Reduced by Cucurbitacin D Through Inhibition of NF-κB
It is well known that Bcl-2 family proteins play important roles in protecting tumor cells from apoptosis. We therefore investigated the effects of cucurbitacin D on Bcl-xL and Bcl-2 by Western blotting. Although the expression of these proteins was not attenuated in PBLs (Fig. 2B), cucurbitacin D markedly reduced Bcl-xL and Bcl-2 protein expression after 24 hours both in MT-4 and in Hut102 cells (Fig. 2A). To confirm this finding, a Bcl-xL luciferase reporter construct was introduced into leukemia cells, and luciferase activity was measured. Treatment with cucurbitacin D resulted in attenuation of Bcl-xL reporter activity (Fig. 2C, left). We also examined NF-κB activity because constitutively activated NF-κB has been observed in T-cell leukemia cells and regulates Bcl-xL transcription.12 NF-κB reporter activity was also reduced after incubation with cucurbitacin D (Fig. 2C, middle).
Expression of Bcl-2 family proteins was reduced by cucurbitacin D via inhibition of NF-κB. (A) MT-4 cells (left) and Hut102 cell (right) were treated with cucurbitacin D (1 μg/mL) or ethanol (0.01%) for the indicated periods. (B) Human PBLs from a healthy donor were treated with cucurbitacin D (1 μg/mL) or ethanol (0.01%) for the indicated periods. Whole-cell extracts were probed by Western blot for Bcl-xL and Bcl-2. β-Actin is shown as a loading control. (C) MT-4 cells (upper) and Hut102 cells (lower) were transfected with Bcl-xL (left) or NF-κB (middle) luciferase reporter constructs and stimulated with cucurbitacin D (1 μg/mL) or ethanol (0.01%) for 6 hours prior to the measurement of luciferase. MT-4 cells (upper right) and Hut102 cells (lower right) were cotransfected with the expression vector for p65 or empty vector pcDNA3.1 and Bcl-xL or NF-κB luciferase reporter constructs. Cells were stimulated with cucurbitacin D (1 μg/mL) or ethanol (0.01%) for 6 hours to determine the luciferase reporter activity of Bcl-xL or NF-κB. The activity is represented by relative luciferase activity. The white column indicates 0.01% ethanol treatment, and the black column indicates cucurbitacin D treatment (pc3.1, empty vector pcDNA3.1; p65, pc3.1 p65 expression vector; *significantly decreased from the control at P<.05). Results are representative of at least 3 similar experiments.
To further characterize the relationship between NF-κB activity and Bcl-xL expression, MT-4 and Hut102 cells were cotransfected with p65, a component of NF-κB, and the Bcl-xL or NF-κB reporter constructs. It is known that overexpression of p65 is dominant-positive for NF-κB activity in the reporter assay, indicating the role that overexpressed p65 played in the nucleus.22 As shown in Figure 2C (right), overexpression of p65 enhanced NF-κB and Bcl-xL reporter activity in Hut102 cells and was not inhibited by cucurbitacin D treatment. These results demonstrate that cucurbitacin D may inhibit Bcl-2 family protein expression through suppression of NF-κB activity at the cytoplasmic level, rather than in the nucleus.
Treatment of Cucurbitacin D Induces Decreased NF-κB in the Nucleus and NF-κB Accumulation in Cytoplasm
To investigate the mechanism by which cucurbitacin D suppresses NF-κB activity, the fractions of NF-κB in the nucleus and cytoplasm were analyzed by Western blot. The amount of NF-κB in the nucleus decreased after incubation with cucurbitacin D beginning after 6 hours without affecting a control protein, Oct1, both in MT-4 and Hut102 cells. NF-κB accumulated in the cytoplasm beginning after 6 hours of treatment with cucurbitacin D (Fig. 3A). Accumulation of NF-κB was further observed by immunohistochemistry. Cucurbitacin D induced the accumulation of NF-κB in the cytoplasm and reduced its levels in the nuclei in Hut102 cells (Fig. 3B).
Treatment of cucurbitacin D induced a decrease of NF-κB in the nucleus and NF-κB accumulation in the cytoplasm. (A) MT-4 cells (upper) and Hut102 cells (lower) were cultured in the presence of cucurbitacin D (1 μg/mL) or ethanol (0.01%) for the indicated periods, and then nuclear extracts and cytoplasmic fractions of cultured cells were prepared and probed by Western blot analysis to measure NF-κB levels. Oct1 and β-actin served to demonstrate equal protein loading. (B) Hut102 cells, pretreated with cucurbitacin D (1 μg/mL) or ethanol (0.01%) for 6 hours, were fixed, permeabilized, and stained with anti-NF-κB antibody and Alexa Fluor–labeled 488 rabbit antigoat antibody. The fluorescence image (upper left), the corresponding phase-contrast image (upper right), and the merged image (lower) are shown. Results are representative of at least 2 similar experiments.
Ubiquitinated Proteins Are Accumulated by Cucurbitacin D in Leukemia Cells
To investigate the mechanism of NF-κB cytosolic accumulation, the expression of IκBα was examined by Western blot. IκBα forms a tight complex with NF-κB in the cytoplasm and inhibits the nuclear translocation signal of NF-κB. Western blot results demonstrated that the level of IκBα was increased by cucurbitacin D, both in MT-4 and in Hut102 cells (Fig. 4A). Cucurbitacin D also increased phosphorylation of IKKα/β, an upstream kinase for NF-κB. Furthermore, increased ubiquitination of IκBα was detected after stimulation with cucurbitacin D. Consequently, Western blotting analysis was performed to analyze the total amount of ubiquitinated proteins after cucurbitacin D treatment. As shown in Figure 4B, the increased amount of ubiquitinated protein was seen beginning after 6 hours of stimulation, both in MT-4 and in Hut102 cells, suggesting that cucurbitacin D exerts an inhibitory effect on the proteasome, blocking the degradation of ubiquitinated proteins in leukemia cells.
Ubiquitinated proteins were accumulated by cucurbitacin D in leukemia cells. (A) Whole-cell lysates of MT-4 and Hut102 cells treated with cucurbitacin D (1 μg/mL) or ethanol (0.01%) for the indicated periods were immunoblotted with specific antibodies (left). β-Actin demonstrated equal protein loading. Arrows indicate IKKα/β. Whole-cell lysates were then subjected to immunoprecipitation with anti-IκBα antibody, and subsequently evaluated by Western blot using anti-IκBα antibody and antiubiquitin antibody. β-Actin served as input (right). (B) Whole-cell lysates from MT-4 and Hut102 cells were probed by Western blot of ubiquitin. β-Actin served to demonstrate equal protein loading. Ethanol served as the control.
Cucurbitacin D Inhibits Proteasome Activity in Leukemia Cells
To determine whether cucurbitacin D influences proteasome activity, we performed an in vitro assay using cell lysates from MT-4 and Hut102 cells cultured with the indicated concentrations of cucurbitacin D for 6 or 12 hours. As shown in Figure 5, cucurbitacin D significantly suppressed proteasome activity of leukemia cells in a dose- and time-dependent manner. Similar inhibition was observed in MT-1, MT-2, Hut78, and Jurkat cells (data not shown).
Cucurbitacin D inhibited proteasome activity in leukemia cells. MT-4 and Hut102 cells were treated with indicated concentrations of cucurbitacin D or ethanol for 6 or 12 hours. Proteasome activity was determined using specific substrates as described in the Materials and Methods section. Results are represented as the reduction of relative luminescence activity. Results are representative of at least 3 similar experiments.
Low Doses of Cucurbitacin D Synergistically Potentiate Antiproliferative Effects of VPA
HDAC inhibitors have been previously reported to synergistically interact with the proteasome inhibitor to induce apoptosis. Therefore, we finally analyzed synergistic effect of cucurbitacin D with HDAC inhibitor VPA in leukemia cells. MT-4 and Hut102 cells were treated with VPA and low doses of cucurbitacin D (0.1 μg/mL). The combination of VPA and cucurbitacin D significantly inhibited the proliferation of leukemia cells (Fig. 6). This finding suggests that cucurbitacin D may be a useful drug with other inhibitors in the chemotherapy of leukemia cells.
Low doses of cucurbitacin D synergistically potentiated the antiproliferative effects of VPA. MT-4 cells were exposed to cucurbitacin D (0.1 μg/mL) in combination with 5 mM of VPA for 18 hours. Hut102 cells were coincubated with cucurbitacin D (0.1 μg/mL) and 1 mM of VPA for 24 hours. Proliferation was monitored with the Cell Titer-Glo luminescent cell viability assay. Results are represented as reduction of relative luminescence activity (*significantly increased from VPA treatment at P <.05). Results are representative of at least 3 similar experiments.
Cucurbitacin D Induces Apoptosis of Leukemia Cells In Vivo
The ability of cucurbitacin D to exert effects in vivo was examined using SCID mice with MT-4 or Hut102 cells. According to the dose of bortezomib widely used in the inhibition of tumor growth in mouse experiments, we used 1 or 4 mg/kg to demonstrate the antitumor activity of cucurbitacin D in vivo.23 Twenty-four hours after the administration of cucurbitacin D, strong apoptosis was induced in Hut102 cells (Fig. 7A). Furthermore, ubiquitinated protein and IκBα were increased in MT-4 and Hut102 cells, suggesting that cucurbitacin D could inhibit the proteasome system in vivo (Fig. 7B).
Cucurbitacin D induced apoptosis of leukemia cells in vivo. (A) CB.17/SCID mice were inoculated intraperitoneally with 4 × 106 MT-4 or Hut102 cells. Mice were administrated 4 mg/kg of cucurbitacin D or vehicle alone after injection of tumor cells for 1 week. Twenty-four hours after cucurbitacin D injection, tumor cells were collected from the peritoneal cavity. Cells undergoing apoptosis were detected by uptake of a purple dye. Ethanol treatment served as the control. (B) Whole-cell lysates of peritoneal tumor cells treated with cucurbitacin D (1 or 4 mg/kg) or ethanol alone were used for immunoblotting with specific antibodies (left). β-Actin demonstrated equal protein loading. Results are representative of at least 2 similar experiments.
DISCUSSION
In this study, we focused on the proapoptotic effect of cucurbitacin D on T-cell leukemia cells. Cucurbitacin D inhibited proteasome activity, resulting in the accumulation of ubiquitinated IκBα. Furthermore, increased ubiquitinated IκBα inhibited NF-κB activity, which resulted in decreased expression of antiapoptotic proteins such as Bcl-xL and Bcl-2, and thus promoted apoptosis of leukemia cells.
Bcl-2 family proteins are important regulators of apoptosis.24 We showed that Bcl-xL expression and NF-κB activity were suppressed by cucurbitacin D. It is well known that the expression of Bcl-xL is regulated by NF-κB at the transcriptional level, supporting the interpretation that cucurbitacin D suppresses Bcl-xL expression via inhibition of NF-κB.25, 26 Bcl-2 expression, which has been shown to be regulated by STAT327 in addition to NF-κB,28 was also suppressed by cucurbitacin D. Interestingly, cucurbitacin I, which has been used as a STAT3 inhibitor, has a chemical structure similar to that of cucurbitacin D.3 Therefore, cucurbitacin D may inhibit Bcl-2 expression partially through suppression of STAT3 activation.
Constitutively activated NF-κB sustains proliferation of tumor cells and regulates expression of antiapoptotic proteins to protect leukemia cells from apoptosis.12, 29 Therefore, chemicals that inhibit NF-κB can efficiently induce apoptosis.30, 31 Cucurbitacin D decreased the amount of NF-κB in the nucleus and increased the accumulation of NF-κB in the cytoplasm (Fig. 3), which resulted in the induction of apoptosis, suggesting that cucurbitacin D will be a useful chemotherapeutic agent for human T-cell leukemia. In addition, we observed that cucurbitacin D inhibited NF-κB activity, mainly in the cytoplasm, because cucurbitacin D inhibited IκBα degradation and did not suppress NF-κB activity after transfection of a dominant-positive p65 expression vector into cells (Fig. 2C). However, it is still unclear whether cucurbitacin D provokes the translocation of constitutively active NF-κB from the nucleus to the cytoplasm.
Cucurbitacin D inhibited proteasome activity and blocked the degradation of ubiquitinated IκBα. Because it is well known that the key regulator of activation of NF-κB is IκB,32 the accumulation of ubiquitinated IκB in cells provides a clear mechanism for the suppression of NF-κB activity by cucurbitacin D. Importantly, proapoptotic and proteasome inhibitory activities of cucurbitacin D were also demonstrated in vivo. Therefore, cucurbitacin D is potentially a very useful drug for cancer therapy.
Recently, proteasome inhibitors have also been developed as new anticancer agents in the clinic. One proteasome inhibitor, bortezomib, has cytotoxic effects on various kinds of tumor cells in vitro. Bortezomib can inhibit NF-κB by blocking the degradation of IκBα.33 In addition, bortezomib was reported to alter the levels of Bcl-2, leading to cell-cycle arrest and apoptosis in several tumor types.34 This indicates that bortezomib has effects on proteasome inhibition and regulation of NF-κB. Therefore, cucurbitacin D also may be a new alternative proteasome inhibitor for the treatment of cancer.
The proteasome is a multisubunit protease that consists of 1 catalytic 20S proteasome and 2 19S regulatory complexes. The assembly of the proteasome is precisely regulated by assembly factors at each step.35 Several kinds of proteasome chaperones have been reported to be dedicated to different steps in proteasome assembly. For example, PAC1-4 promotes α-ring formation by binding to intermediates of α-subunits. Ump1 mediates the assembly of the 15S complex to 20S, which inhibits stable dimerization of 15S until the completion of the β-ring assembly.36 Unexpectedly, cucurbitacin D did not directly suppress the activities of purified 20S or 26S (data not shown). One possibility is that cucurbitacin D might obstruct the assembly of the proteasome through interference with proteasome chaperones, although further research is required to test this hypothesis. Importantly, cucurbitacin D and bortezomib inhibit proteasome activity through different mechanisms.
There has been a growing interest in the use of herbs as a source of new drugs for cancer. The triterpenoids, including cucurbitacin, which are ubiquitous constituents in plants and fruits, have been found to possess anticancer effects. One of the main targets of cucurbitacins is activated STAT3.4 Blaskovich et al reported that cucurbitacin I suppressed phosphorylation of STAT3 in cancer cells. In contrast, the mechanism of cucurbitacin D's anticancer effects is different, although its structure is very similar, suggesting that different cucurbitacins operate via different mechanisms. An important and novel finding presented here is that cucurbitacin D induced apoptosis in T-cell leukemia cells via inhibition of the proteasome system. This finding is the first instance of proteasome inhibition in the cucurbitacin family. Although cucurbitacins may be useful anticancer drugs, they have not yet been developed because of their general toxicity to normal cells.37, 38 Park et al reported that 23,24-dihydrocucurbitacin D did not show apparent toxicity toward normal cells.39 We also found that cucurbitacin D did not affect human PBLs from healthy donors. Thus, cucurbitacin D might be a useful candidate for an anticancer drug.
Our results show that combination chemotherapy of low doses of cucurbitacin D and HDAC inhibitor synergistically induced apoptosis in T-cell leukemia cells, similarly to previous reports that HDAC inhibitor synergizes with other proteasome inhibitors such as bortezomib or MG132.19, 40 Previous studies have shown that exposure of tumor cells to HDAC inhibitors such as phenylbutyrate induced inactivation of NF-κB.41 It is still unclear whether cucurbitacin D affects HDAC activity. However, previous studies have documented that the potential mechanisms of the synergistic effect between the proteasome inhibitor bortezomib and HDAC inhibitors include cointerruption of NF-κB activation and aggresome induction.42, 43 Further experimentation is needed to explore the precise mechanism of this synergistic interaction.
Collectively, these data demonstrate that the proteasome inhibitor cucurbitacin D is a potential new therapeutic drug for treatment of T-cell leukemia.
CONFLICT OF INTEREST DISCLOSURES
Supported in part by Grant-in-Aid for Scientific Research (C)—19591147 and 21591253—from the Japan Society for the Promotion of Science (to Y.Y.).
