Mollugin induces apoptosis in human Jurkat T cells through endoplasmic reticulum stress-mediated activation of JNK and caspase-12 and subsequent activation of mitochondria-dependent caspase cascade regulated by Bcl-xL

Sun Mi Kim a,1, Hae Sun Park a,1, Do Youn Jun a, Hyun Ju Woo a, Mi Hee Woo b, Chae Ha Yang c, Young Ho Kim a,⁎

Abstract

T cells and mitochondria-dependent activation of caspase cascade were completely prevented by overexpression of Bcl-xL, the activation of JNK and caspase-12 was prevented to much lesser extent. Pretreatment of the cells with the pan-caspase inhibitor (z-VAD-fmk), the caspase-9 inhibitor (z-LEHD-fmk), the caspase-3 inhibitor (z-DEVD-fmk) or the caspase-12 inhibitor (z-ATAD-fmk) at the minimal concentration to prevent mollugininduced apoptosis appeared to completely block the activation of caspase-7 and -8, and PARP degradation, but failed to block the activation of caspase-9 and -3 with allowing a slight enhancement in the level of JNK phosphorylation. Both FADD-positive wild-type Jurkat clone A3 and FADD-deficient Jurkat clone I2.1 exhibited a similar susceptibility to the cytotoxicity of mollugin, excluding involvement of Fas/FasL system in triggering mollugin-induced apoptosis. Normal peripheral T cells were more refractory to the cytotoxicity of mollugin than were Jurkat T cells. These results demonstrated that mollugin-induced cytotoxicity in Jurkat T cells was mainly attributable to apoptosis provoked via endoplasmic reticulum (ER) stress-mediated activation of JNK and caspase-12, and subsequent mitochondria-dependent activation of caspase-9 and -3, leading to activation of caspase-7 and -8, which could be regulated by Bcl-xL.

Keywords:
Rubia cordifolia L.
Mollugin
Cytotoxicity
Apoptosis
Mitochondrial cytochrome c
ER stress
Caspase cascade
Bcl-xL
Leukemia Jurkat

Introduction

Chemotherapy toward tumor cells is known to induce apoptosis, a programmed cell death. Since the induction of apoptosis in tumor cells can lead to their own destruction into apoptotic bodies which can be cleared by surrounding cells without accompanying a local damaging inflammatory response, apoptosis has been proposed as an efficient mechanism by which malignant tumor cells can be removed upon treatment with chemotherapeutic drugs (Hannun, 1997). In chemotherapeutic drug-induced apoptosis of tumor cells, three different death signaling pathways leading to apoptosis, such as the extrinsic death receptor-dependent pathway (Wallach et al., 1997), the intrinsic mitochondria-dependent pathway (Desagher and Martinou, 2000), and the intrinsic endoplasmic reticulum (ER) stressing in tumor cells, following chemotherapy, can be provoked by the upregulation of FasL and/or Fas expressions with the subsequent induction of Fas signaling (Friesen et al., 1996; Muller et al., 1997; Nagarkatti and Davis, 2003), the mitochondria-dependent death signaling begins with the mitochondrial cytochrome c release into cytoplasm, which together with the apoptotic protease activating factor-1 (Apaf-1) activates caspase-9 in the presence of dATP, and then activates an effector caspase, caspase-3, leading to cell death (Herr and Debatin, 2001; Kaufmann and Earnshaw, 2000). The ER stress-mediated apoptotic pathway is initiated by the activation of caspase-12, which can directly activate caspase-9 independently of mitochondrial cytochrome c and Apaf-1 (Morishima et al., 2002; Rao et al., 2002). In addition to the caspase-12 activation, ER stress also mediated pathway (Nakagawa et al., 2000), are likely to be implicated. Whereas the death receptor-dependent apoptotic signal-triggers the activation of caspase-8 and c-Jun N-terminal kinase (JNK), both of which are known to cause mitochondrial cytochrome c release (Jimbo et al., 2003; Urano et al., 2000). The mechanism in chemotherapeutic drug-induced apoptosis needs to be studied further in order to evaluate the efficiency of an antitumor agent and to clarify whether the apoptogenic effect of the drug is confined to tumor cells rather than normal cells.
The roots of Rubia cordifolia L. have been used as traditional herbal medicine in Korea to treat cough, bladder and kidney stones, inflammation of the joints, uterine hemorrhage, and uteritis (Son et al., 2008). In addition, this plant has been used for traditional Chinese medicine for treatment of arthritis, dysmenorrheal, hematorrhea, hemostasis, and psoriasis (Chang et al., 2000; Tse et al., 2007). Pharmacological studies have demonstrated that the root of Rubia cordifolia L. possesses antibacterial (Basu et al., 2005), antiviral (Ho et al., 1996), antioxidant (Cai et al., 2004), anti-inflammatory (Tezuka et al., 2001), antitumor (Itokawa et al., 1993), and hepatoprotective activities (Gilani and Janbaz, 1995). A number of the bioactive components have been reported from Rubia cordifolia L. including anthraquinones, anthraquinone glycosides, naphthoquinones, naphthoquinone glycoside, furomollugin, mollugin, alizarin, lucidin pimeveroside, ruberythric acid, bicyclic hexapeptides, triterpenoids, oleanolic acid, and epoxymollugin (Itokawa et al., 1983; Chung et al., 1994; Han et al., 1990; Itokawa et al., 1993; Lee et al., 2008; Son et al., 2008). Among these, mollugin (C17H16O4; methyl 2,2-dimethyl-6hydroxy-2H-naphtho[1,2-b]pyran-5-carboxylate) has been reported for its cytotoxic and antiproliferative activity (Chang et al., 2000; Lu et al., 2007) as well as several pharmacological effects such as antiplatelet aggregation activity (Chung et al., 1994), and antiviral activity against hepatitis B virus (Ho et al., 1996). In relation to the antitumor activity, mollugin has been shown to exert cytotoxic effect on human colon cancer Col2 cells with IC50 value of 12.3 μM (Chang et al., 2000), and on human liver carcinoma HepG2 cells with IC50 value of ∼60.2 μM, although it has no detectable topoisomerase inhibitory activity (Son et al., 2008). However, the mechanism responsible for the cytotoxic effect of mollugin on tumor cells remains largely unknown.
In the present study, it has been investigated whether cytotoxicity of the mollugin extracted from the roots of Rubia cordifolia L. on human acute leukemia Jurkat T cells is attributable to apoptotic cell death. To understand the implication of mitochondrial cytochrome c release and endoplasmic reticulum (ER) stress in the molluginmediated apoptotic signaling pathway, mollugin-induced apoptotic events of Jurkat T cells transfected with the vector (J/Neo) have been compared with those of Jurkat T cells transfected with the Bcl-xL gene (J/Bcl-xL). In addition, the suppressive effects of individual caspase inhibitors on mollugin-mediated cytotoxicity and caspase cascade have been analyzed. The results show that the cytotoxicity of mollugin toward Jurkat T cells is mainly due to induced apoptosis, which is negatively regulated by overexpression of Bcl-xL. The results also indicate that mollugin-induced apoptosis is provoked by mitochondrial membrane potential disruption and cytochrome c release and resultant activation of caspase cascade including caspase-9, -3, -7, and -8, in which ER stress-mediated activation of JNK and caspase-12 is involved.

Materials and methods

Reagents, chemicals, antibodies, cells, and culture medium. Mollugin was extracted from the roots of Rubia cordifolia L. as previously described (Son et al., 2008). ECL Western blotting kit was purchased from Amersham (Arlington Heights, IL, USA), and Immobilon-P membrane was obtained from Milipore Corporation (Bedford, MA, USA). The broad-range caspase inhibitor z-VAD-fmk, the caspase-9 inhibitor z-LEHD-fmk, and the caspase-3 inhibitor z-DEVD-fmk were obtained from BD Sciences (Chicago, IL, USA), and the caspase-12 inhibitor z-ATAD-fmk and the caspase-4 inhibitor z-LEVD-fmk were obtained from Biovision (Mountain View, CA, USA). Anti-phosphoJNK, anti-JNK1, anti-caspase-3, anti-Bid, anti-FLICE inhibitory protein (FLIP), anti-poly (ADP-ribose) polymerase (PARP), and anti-β-actin were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-caspase-8, and anti-caspase-9 were from Cell signaling Technology (Beverly, MA, USA) and anti-caspase-12 was purchased from BD Sciences (Chicago, IL, USA). Human acute leukemia Jurkat T cell line E6.1, FADD-positive wild-type Jurkat T cell clone A3, and FADDdeficient Jurkat T cell clone I2.1 were purchased from ATCC (Manassas, VA). Stable transfectant of Jurkat T cells with the vector (clone J/Neo) and stable transfectant of Jurkat T cells with the antiapoptotic protein Bcl-xL gene (clone J/Bcl-xL) were kindly provided by Dr. Dennis Taub (Gerontology Research Center, NIA/ NIH, Baltimore, MD, USA). Jurkat T cells were maintained in RPMI 1640 (Life Technologies, Gaithersburg, MD, USA) containing 10% FBS, 20 mM HEPES (pH 7.0), 5×10–5 M β-mercaptoethanol, and 100 μg/ml gentamycin. For the culture of both J/Neo cells and J/Bcl-xL cells, G418 (A.G. Scientific Inc., San Diego, CA, USA) was added to RPMI 1640 medium at a concentration of 400 μg/ml.
HPLC system and conditions. The HPLC apparatus consisted of a Agilent 1100 HPLC System (Agilent Technologies, Palo Alto, CA, USA), equipped with a quaternary pump with a vacuum degasser, an autosampler, and UV/Vis detector. The separation was conducted on an Agilent TC-C18(2) column (150 mm×4.6 mm i.d., 5 μm particle size). The mobile phase was 0.1% trifluoroacetic acid (solvent A) and acetonitrile (solvent B). Chromatographic conditions were as follows: isocratic 5% B for 5 min, increasing to 95% B over 20 min, and then holding at 95% B for 5 min at the flow rate of 0.4 ml/min. The column temperature was set at 60 °C and the effluent was monitored at 254 nm.
Isolation and activation of human peripheral T cells. To prepare human peripheral blood mononuclear cells (PBMC), heparinized blood obtained from healthy laboratory personnel by venipuncture was centrifuged at 800×g for 20 min over HISTOPAQUE-1077 (Sigma Chemical, St. Louis, MO, USA), according to the manufacturer’s instructions. This protocol was approved by the Ethics Committee of the University of Kyungpook National University, Daegu, Korea. Informed written consent was obtained from the participants. Isolation of T cells from PBMC was performed using a human T cell enrichment column kit (R and D Systems, Minneapolis, MN, USA). For activation of the peripheral T cells, the isolated peripheral T cells at a density of 2×106/ml were incubated with phytohemagglutinin A (PHA) at a concentration of 1.0 μg/ml for 60 h. To induce the interleukin-2 (IL-2)-dependent proliferation of T cells, the PHAactivated T cells (1×105/well) were cultured with 50 units (U) of recombinant IL-2 in 96-well plates.
Cytotoxicity assay. The cytotoxic effect of mollugin on Jurkat T cell was analyzed by MTT assay reflecting the cell viability as previously described (Jun et al., 2007). Briefly, Jurkat T cells (5×104/well), human peripheral T cells (2×105/well), or PHA-activated human T cells (1×105/well) were added to serial dilutions of mollugin in 96-well plates. At 20 h after incubation, 50 μg of MTT solution (1.1 mg/ml) was added to each well and incubated for an additional 4 h. After centrifugation, the supernatant was removed from each well and then 150 μl of DMSO was added to dissolve the colored formazan crystal produced from MTT. OD values of the solutions were measured at 540 nm by a plate reader.
DNA fragmentation analysis. Apoptotic DNA fragmentation induced in Jurkat T cells following mollugin treatment was determined by Triton X-100 lysis methods using a 1.2% agarose gel electrophoresis as previously described elsewhere (Jun et al., 2008).
Flow cytometric analysis. The cell cycle progression of Jurkat T cells transfected with vector (JT/Neo) or Bcl-xL gene (JT/Bcl-xL) following mollugin treatment were analyzed by Flow cytometry as described elsewhere (Kim et al., 1992). The extent of necrosis was detected with Annexin V-FITC Apoptotic Kit (Clontech, Takara Bio, Shiga, Japan). After washing with 1× binding buffer, the cells (0.3×106) were suspended in 200 μl 1× binding buffer, to which 2 μl Annexin V-FITC (100 μg/ml) and 10 μl PI (400 μl/ml) were added. Incubation was performed at room temperature for 15 min before being analyzed by flow cytometry according to the manufacturer’s instructions.
Measurement of mitochondrialmembrane potentialdisruption. Changes in the mitochondrial membrane potential following treatment with mollugin was measured after staining with DePsipher™ (R and D Systems, Inc., Minneapolis, MN, USA). DePsipher™ is a lipophilic cation (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolocarbocyanine iodide) which possesses the property of aggregating upon membrane polarizationforminganorange-redfluorescentcompound(Smileyet al., 1991). If the potential is disturbed, the dye cannot access the membrane space and remains or reverts to its green monomeric form. In this context, the red and green fluorescences of DePsipher™ reflect changes in the mitochondrial membrane potential. After treatment with mollugin, the cells were harvested and were then resuspended in 1× Reaction buffer containing 5 μg/ml DePsipher™ for 20 min at 37 °C. The percentage of red and green fluorescence was estimated by flow cytometry according to the manufacturer’s instructions.
Detection of mitochondrial cytochrome c in cytosolic protein extracts. To assess mitochondrial cytochrome c release in Jurkat T cells following mollugin treatment, cytosolic protein extracts were obtained as described elsewhere (Jun et al., 2007). The cytosolic extracts free of mitochondria were analyzed for cytochrome c by Western blotting.
Preparation of cell lysates and Western blot analysis. Cellular lysates were prepared by suspending 5×106 Jurkat T cells in 250 μl of lysis buffer (137 mM NaCl, 15 mM EGTA, 1 mM sodium orthovanadate, 15 mM MgCl2, 0.1% Triton X-100, 25 mM MOPS, 2.5 μg/ml proteinase inhibitor E-64, and pH 7.2). The cells were disrupted by sonication and extracted at 4 °C for 30 min. An equivalent amount of protein lysate (15–20 μg) was subjected to electrophoresis on a 4–12% SDS gradient polyacrylamide gel with a MOPS buffer. The proteins were electrotransferred to Immobilon-P membranes and then probed with individual antibodies. Detection of each protein was performed using an ECL Western blotting kit according to the manufacturer’s instructions.
Determination of caspase activity. Caspase-12 activity was assayed by using the Caspase-12 Fluorometric Assay Kit (Biovision, Mountain View, CA, USA), and caspase-3 activity was assayed by using the Caspase-3 Colorimetric Activity Assay Kits (Chemicon International Inc., Temecula, CA, USA) according to the manufacturer’s protocols. Equal number of cells (5×106) from each sample were treated with Cell Lysis Buffer on ice for 10 min, and centrifuged at 10,000×g for 10 min. The supernatant (150 μg of protein) was incubated with each caspase substrate (ATAD-FMC for caspase-12, and DEVD-pNA for caspase-3) at 37 °C for 1 h.
For in vitro caspase-12 inhibition assay, the cell lysate (150 μg of protein) prepared from J/Neo cells treated with 30 μM mollugin for 24 h was added to various concentrations (0.1, 0.5, 1, and 4 μM) of the caspase-12 inhibitor z-ATAD-fmk. After these mixtures were incubated at room temperature for 30 min to allow z-ATAD-fmk to react with caspase-12, the substrate ATAD-FMC for caspase-12 was added to determine residual caspase-12 activity. Under the same conditions, to test for cross-reactivity of the caspase-12 inhibitor z-ATAD-fmk toward caspase-3 activity, the substrate DEVD-pNA for caspase-3 was added. Following addition of the substrates, the reaction mixture was incubated at 37 °C for 1 h. The caspase-12 activity was measured by a fluorometer equipped with a 400-nm excitation filter and a 510-nm emission filter. The caspase-3 activity was measured by a microplate reader at 405 nm.
Statistical analysis. Unless otherwise indicated, each result in this paper is representative of at least three separate experiments. Values represent the mean±standard deviation (SD) of these experiments. The statistical significance was calculated with Student’s t-test. P values less than 0.05 were considered significant.

Results

Identification of mollugin as an apoptogenic component in the roots of Rubia cordifolia L.

Roots of Rubia cordifolia L. (10 kg) were chopped into small pieces and extracted four times with methanol at 60 °C. When the methanol extract (1048.4 g) was dissolved in water (2 l) and then fractionated by a series of solvent extractions using methylene chloride, ethyl acetate, and n-butanol, the methylene chloride extract (360.6 g) appeared to contain the most cytotoxic activity toward Jurkat T cells whereas the other fractions showed no significant cytotoxic effects.
To purify the cytotoxic component further, the methylene chloride extract (180 g) was applied to silica gel (4 kg) column chromatography and fractionated into 34 fractions (RA-MC-1 to RA-MC-34) by twostep elution using 100% hexane to 100% methylene chloride in 2%→15% gradient and then 99% methylene chloride/1% methanol to 100% methanol in 1%→25% gradient. When the fraction RA-MC-2 (5 g) was chromatographed on a silica gel column by elution with 100% hexane, a substance (250 mg) exerting cytotoxic effect on Jurkat T cells, which was identified as mollugin (C17H16O4; methyl 2,2dimethyl-6-hydroxyl-2H-naphtho[1,2-b]pyran-5-carboxylate), was yielded. From 10 kg of the roots of Rubia cordifolia L., approximately 500 mg of mollugin was recovered as previously described (Son et al., 2008). When the purity of mollugin sample recovered was analyzed by HPLC as compared with a commercially available mollugin standard (Chromadex, Irvine, CA; purity 98.8%) by high performance liquid chromatography (HPLC), it appeared to be 97.0% (Figs. 1A and B).
To understand the mechanisms underlying the cytotoxicity of mollugin, which was extracted from the roots of Rubia cordifolia, its effect on Jurkat T cells transfected with the vector (J/Neo) and Jurkat T cells transfected with the Bcl-xL gene (J/Bcl-xL) was investigated. When J/Neo cells were treated with mollugin at concentrations of 15 μM and 30 μM for 24 h, the cell viability of J/Neo, which was determined by MTT assay, was reduced to the level of 75.5% and 40.2%, respectively (Fig. 2A). Under these conditions, the commercially available mollugin exhibited a similar level of cytotoxicity toward J/ Neo cells (data not shown). In accordance of the cytotoxicity of mollugin, apoptotic DNA fragmentation as well as sub-G1 peak representing apoptotic cells was enhanced by mollugin in a dosedependent manner (Figs. 2B and C). However, the mollugin-induced cytotoxicity along with apoptotic DNA fragmentation and apoptotic sub-G1 peak was abrogated in J/Bcl-xL cells overexpressing Bcl-xL. Previously, antiapoptotic protein Bcl-xL has been known to be one of the antiapoptotic members of Bcl-2 family, which can protect cells from apoptosis by blocking mitochondrial cytochrome c release into cytosol, resulting in the prevention of mitochondria-dependent apoptotic pathway (Kluck et al., 1997; Jun et al., 2007). Consequently, these results indicated that the cytotoxicity of mollugin toward Jurkat T cells was mainly due to induced apoptotic DNA fragmentation, which could be regulated by antiapoptotic protein Bcl-xL.

Involvement of mitochondrial membrane potential disruption, mitochondrial cytochrome c release, caspase cascade activation, and ER stress in mollugin-mediated apoptosis in Jurkat T cells

To understand the death-signaling pathway underlying the mollugin-induced apoptosis in Jurkat T cells, it was investigated whether mitochondrial membrane potential was disrupted in J/Neo cells after treatment with mollugin. The change in the mitochondrial membrane potential of the cells following exposure to mollugin was measured by flow cytometry using a DePsipher™ kit. As shown in Fig. 3, although there was nearly detectable disruption of mitochondrial membrane potential in continuously growing J/Neo cells, 15.7% and 51.6% of the cells exhibited mitochondrial membrane potential disruption in the presence of 15 μM and 30 μM mollugin, respectively. This indicated that mollugin (15–30 μM) was able to disrupt mitochondrial membrane potential in a dose-dependent manner. At the same time, however, the mollugin failed to disrupt mitochondrialmembrane potential in J/Bcl-xL cells. Since mitochondrial membrane potential disruption is known to be one of the initial intracellular changes that are accompanied by apoptotic cell death (Carosio et al., 2007), these results demonstrated that the disruption of mitochondrial membrane potential was involved in mollugin-induced apoptosis in J/Neo cells. These results also indicated that the disruption of mitochondrial membrane potential was caused by a conserved apoptogenic mechanism, which could be targeted by the antiapoptotic role of Bcl-xL protein.
To elucidate the mollugin-induced death signaling pathway, mitochondrial cytochrome c release into cytosol and activation of caspase cascade including caspase-9, -3, and -7, leading to poly (ADPribose) polymerase (PARP) degradation, were investigated by Western blot analysis. As shown in Fig. 4A, although there was barely detectable cytochrome c in the cytosolic fraction of continuously growing J/Neo cells, the level of mitochondrial cytochrome c release was enhanced after treatment with mollugin (15–30 μM). Along with cytochrome c release, the caspase-9 activation that proceeded through proteolytic cleavage of inactive proenzyme (47 kDa) to active forms (37/35 kDa) was detected (Fig. 4B). The cleavage of procaspase-3 (32 kDa) into active form (17 kDa) as well as the cleavage of procaspase-7 (35 kDa) into active form (20 kDa) was also detected in a dose-dependent manner. As a downstream target of the active caspase-3 and -7, the degradation of PARP was also detected along with the caspase-3 activation. In order to examine the involvement of endoplasmic reticulum (ER) stress-mediated apoptotic events as the upstream signals in the mollugin-induced mitochondrial cytochrome c release and activation of caspase cascade, the activation of JNK, caspase-12 and -8 was also investigated by Western blot analysis. In the presence of mollugin (15–30 μM), the phosphorylation of JNK increased significantly without a change in the level of total JNK1 protein, whereas the level of procaspase-12 (55 kDa) appeared to decline slightly, and the activation of caspase-8 through proteolytic cleavage of proenzyme (57 kDa) into active forms (43/41 kDa) was significantly enhanced. In addition, the level of Bid protein (22 kDa), which was previously degraded by active caspase-8 to generate the truncated Bid (tBid, 15 kDa) causing mitochondrial membrane transition pore opening and subsequent release of cytochrome c (Li et al., 1998; Desagher et al., 1999), appeared to decline in accordance with mollugin-induced caspase-8 activation in J/Neo cells. Like Bid protein, the FLIP, which was known to inhibit caspase-8 activation, was previously degraded by active caspase-8 (Boehrer et al., 2006). Along with mollugin-induced activation of caspase-8 as well as resultant reduction in the level of the Bid protein, the level of FLIP was also reduced. However, these mollugin-induced apoptotic changes were abrogated in J/Bcl-xL cells, except mollugin-induced increase in the level of phosphorylated JNK, which was largely unaffected by overexpressed Bcl-xL. Since the anti-caspase-12 antibody employed for Western blot analysis in this study was known to recognize the procaspase-12 but not the cleaved form of caspase-12, we further evaluated in vitro caspase-12 activity to confirm mollugin-induced caspase-12 activation in J/Neo and J/BclxL cells. Although the caspase-12 activity appeared to increase in a dose-dependent manner in J/Neo cells, the increase of caspase-12 activity was markedly, but not completely, prevented in J/Bcl-xL cells (Fig. 4C). Under these conditions, the caspase-3 activity was also enhanced in J/Neo cells in accordance with the results of Western blot analysis of mollugin-induced caspase-3 activation, and again the mollugin-induced enhancement of the caspase-3 activity was mostly prevented in J/Bcl-xL cells (Fig. 4D). These in vitro caspase activity assays confirmed that mollugin-induced apoptosis of Jurkat T cells was accompanied by caspase-12 activation which was negatively regulated by Bcl-xL. Since procaspase-12 and procaspase-8 were known to be activated in response to ER stress (Jimbo et al., 2003; Jun et al., 2007; Rao et al., 2001), and since the phosphorylated JNK, representing the active enzyme which could be translocated to mitochondria to provoke cytochrome c release into cytoplasm, was previously generated by ER stress (Aoki et al., 2002; Urano et al., 2000), these previous and current results raised the possibility that ER stressprovoked apoptotic signals such as the activation of JNK, caspase-12, and caspase-8 might be involved in mollugin-induced apoptosis as the upstream events for the mitochondrial cytochrome c release and subsequent activation of caspase-9 and -3.

Effect of caspase inhibitor on mollugin-induced death signaling in J/Neo cells

To elucidate further the death signaling pathway for mollugininduced apoptosis, we investigated the effect of the caspase-9 inhibitor (z-LEHD-fmk) (Ozoren et al., 2000), the caspase-3 inhibitor (z-DEVDfmk) (Barut et al., 2005), the pan-caspase inhibitor (z-VAD-fmk) (Slee et al., 1996), the caspase-4 inhibitor (z-LEVD-fmk) (Jiang et al., 2007), or the caspase-12 inhibitor (z-ATAD-fmk) (Yang et al., 2008) on mollugin-induced apoptotic events in J/Neo cells. After J/Neo cells were pretreated with individual caspase inhibitors for 2 h, the cells were exposed to 30 μM mollugin for 24 h. Although there was a barely detectable apoptotic sub-G1 peak in continuously growing J/Neo cells, it increased to the level of 44.6% in the presence of 30 μM mollugin for 24 h (Fig. 5A). The mollugin-induced sub-G1 peak was diminished to the basal level by pretreatment with z-LEHD-fmk, z-DEVD-fmk, z-VAD-fmk, z-LEVD-fmk, or z-ATAD-fmk, whereas the sub-G1 peak was partially reduced to the level of 22.7% by pretreatment with z-LEVD-fmk. These results suggested that the individual activities of caspase-9, -3, and -12 were crucial for mollugin-induced apoptosis in J/Neo cells, but the caspase-4 activity was required to a lesser extent.
As shown in Fig. 5B, Western blot analysis revealed that in the presence of the pan-caspase inhibitor z-VAD-fmk, mollugin-induced apoptotic events such as the cleavage of procaspase-3 into 17 kDa active form, activation of caspase-7 and -8, Bid cleavage, and degradation of PARP in J/Neo cells were completely blocked, with allowing the cleavage of 47 kDa procaspase-9 into 35 kDa active form at a significantly reduced level. Under the same conditions, the generation of 37 kDa active caspase-9 and 19 kDa active caspase-3 was barely detected. These results excluded the possible involvement of caspase-8 activation as an initial signal provoking the mitochondrial cytochrome c release in mollugin-induced apoptosis. In addition, mollugin-induced phosphorylation of JNK appeared to be sustained at a slightly enhanced level in the presence of z-VAD-fmk, suggesting that the JNK activation was an upstream event of the caspase cascade required for the induced apoptosis. The presence of either z-LEHDfmk or z-DEVD-fmk caused not only a complete prevention of mollugin-induced activation of caspase-7 and -8 and degradation of PARP, but also a significant reduction in the level of 37 kDa active caspase-9 along with no generation of 17 kDa active caspase-3. At the same time, 35 kDa active caspase-9 was detected at the similar level to that of the mollugin-treated control cells, and 19 kDa active caspase-3, which was not detected in the mollugin-treated control cells, was detected. Recently, it has been reported that the proteolytic cleavage of procaspase-9 (47 kDa) within the apoptosome yields 35/12 kDa active caspase-9 in order to cleave procaspase-3 (32 kDa) into active caspase-3 (19 kDa), and subsequent feedback cleavage of procaspase9 by 19 kDa active caspase-3 generates 37/10 kDa active caspase-9, which can cleave not only 19 kDa active caspase-3 into 17 kDa active caspase-3 but also 35 kDa procaspase-7 into 20 kDa active caspase-7 (Twiddy and Cain, 2007; Zou et al., 2003). These previous and current results demonstrated that the activation of caspase-9 and -3, which was a prerequisite for mollugin-induced apoptosis, was upstream of the activation of caspase-7 and -8. On the other hand, the caspase-12 inhibitor z-ATAD-fmk completely blocked mollugin-induced activation of caspase-7 and -8 and degradation of PARP with a significant reduction in the cleavage of 47 kDa procaspase-9 into 37/35 kDa active forms. The caspase-4 inhibitor z-LEVD-fmk partially suppressed mollugin-induced caspase-8 activation, but exerted no suppressive effect on the activation of caspase-9 and -7 and degradation of PARP. In particular, only 19 kDa active caspase-3 was produced from 32 kDa procaspase-3 in presence of z-ATAD-fmk, whereas both 19 kDa active form and much smaller amount of 17 kDa active form of caspase-3 were concurrently produced in the presence of z-LEVDfmk. Like the pan-caspase inhibitor z-VAD-fmk, none of these individual caspase inhibitors tested could suppress mollugin-induced JNK phosphorylation. Recently, it has been reported that some commonly used caspase inhibitors lack the specificity required to monitor the roles of specific caspases in the apoptotic cells (Pereira and Song, 2008). In order to examine the inhibitory activity and specificity of z-ATAD-fmk toward the caspase-12, we investigated the inhibitory effect of various concentrations (0.1, 0.5, 1, and 4 μM) of z-ATAD-fmk on the caspase-12 activity or the caspase-3 activity using the lysate of J/Neo cells treated with 30 μM mollugin as the enzyme solution. As shown in Fig. 5C, the caspase-12 activity was inhibited by z-ATAD-fmk in a dose-dependent manner with an inhibition of ∼50% at concentrations of 1–4 μM, whereas the caspase3 activity exhibited a suppression of 12.5%, indicating the specificity of z-ATAD-fmk (1–4 μM) toward the caspase-12. Consequently, current results indicated that the mollugin-induced apoptotic signaling pathway was mediated by mitochondria-dependent activation of caspase-9 and -3, where ER stress-mediated caspase-12 activation was required for its proper progression, leading to the activation of caspase-7 and caspase-8. These results also indicated that mollugin-induced JNK activation, which could be mediated by ER stress, was upstream of the mitochondria-dependent activation of caspase cascade.

Flow cytometric analysis of mollugin-induced apoptotic cell by FITC-conjugated Annexin V staining

In order to examine whether necrosis was accompanied by mollugin-mediated apoptotic cell death in J/Neo cells, the cells treated with 15–30 µM mollugin for 20 h were analyzed by Annexin V staining. As shown in Fig. 6, the treatment of J/Neo cells with 15 μM mollugin caused a slight enhancement in the levels of early apoptotic cells stained only with Annexin V-FITC, and late apoptotic cells stained with both Annexin V-FITC and propidium iodide (PI). While these apoptotic changes appeared to be more apparent when the cells were treated with 30 μM than with 15 μM mollugin, the necrotic cells stained only with PI were barely detected. Under these conditions, however, the levels of neither apoptotic nor necrotic cells were enhanced in J/Bcl-xL cells. These results demonstrated that mollugin could induce apoptotic cell death of Jurkat T cells in a dose-dependent manner, and confirmed that the cytotoxic effect exerted by mollugin on Jurkat T cells was mainly attributable to induced apoptosis, but not to necrosis.

Comparison of cytotoxic effect of mollugin on FADD-positive wild-type Jurkat T cell clone A3 and FADD-deficient Jurkat T cell clone I2.1

As a potential mechanism underlying the apoptosis induced by antineoplastic drugs, upregulation of FasL and/or Fas expression has been implicated (Friesen et al., 1996; Muller et al., 1997; Nagarkatti and Davis, 2003). In order to further examine an involvement of Fas/ FasL system in mollugin-mediated apoptosis, we compared cytotoxic effect of mollugin on FADD-positive wild-type Jurkat T cells (clone A3) with that on FADD-deficient Jurkat T cells (clone I2.1), which was previously refractory to Fas-mediated apoptosis (Juo et al., 1999). As shown in Fig. 7, irrespective of the FADD deficiency, both Jurkat clones showed similar sensitivity to the cytotoxicity of mollugin. These results confirmed that the mollugin-mediated apoptosis of Jurkat T cells was not provoked by the interaction of Fas with FasL.

Cytotoxic effect of mollugin on human peripheral T cells

Since mollugin appeared to possess cytotoxicity against human acute leukemia Jurkat T cells due to its capability of inducing apoptotic cell death, it was of interest to examine whether mollugin can exert the same cytotoxic effect on normal T cells. In this context, we have investigated the cytotoxic effects of mollugin on the viability of human resting peripheral T cells or the interleukin-2 (IL-2)-dependent proliferation of activated T cells, which were obtained by the stimulation of human peripheral T cells with 1.0 μg/ml phytohemagglutinin A (PHA) for 72 h. When the individual cells were incubated with various concentrations of mollugin in a 96-well plate for 20 h and then cell viability was measured by the MTT assay, the viability of unstimulated peripheral T cells was not markedly affected in the presence of 15–22.5 μM mollugin, and remained at the level of 70.5% at a concentration of 30 μM, whereas the IL-2-dependent proliferation of activated T cells was more sensitive to the cytotoxicity of mollugin than resting T cells and exhibited a viability of 52.6% at a concentration of 30 μM (Fig. 8). Under these conditions, the viability of malignant Jurkat T cells (J/Neo) was reduced to the level of 77.9%, 50.8%, and 42.6% at a concentration of 15 μM, 22.5 μM, and 30 μM mollugin, respectively. These results indicated that leukemia Jurkat T cells, as compared to normal T cells, were more sensitive to the apoptogenic activity of mollugin.

Discussion

Since antitumor agent-induced apoptosis of malignant tumor cells can be directly associated with its chemopreventive and chemotherapeutic activities (Kaufmann and Earnshaw, 2000), the apoptogenic activity and underlying mechanism of mollugin, extracted from the roots of Rubia cordifolia L. have been investigated in human acute leukemia Jurkat T cells. This is the first report to demonstrate that mollugin exerts cytotoxic effect on human acute leukemia Jurkat T cells through induction of apoptosis without necrosis. No contribution of necrosis to the cytotoxic effect was evidenced by flow cytometric analysis of Jurkat T cells (J/Neo) stained with Annexin V-FITC and PI following mollugin treatment. In the mollugin-mediated apoptosis of J/Neo cells, we could exclude an involvement of the extrinsic apoptotic pathway that is triggered by Fas/FasL system, because the sensitivity of FADD-positive wild-type Jurkat clone A3 to the cytotoxicity of molluin was similar to that of FADD-deficient Jurkat clone I2.1. Since mollugin-caused cytotoxicity and apoptotic DNA fragmentation in J/Neo cells were completely prevented by overexpression of Bcl-xL that was known to suppress mitochondrial cytochrome c release (Kluck et al., 1997) and endoplasmic reticulum (ER) stress-mediated activation of caspase-12 and -8 (Jun et al., 2007; Morishima et al., 2004), we decided to examine whether the mitochondria-dependent death signaling and ER stress-mediated death signaling were associated with mollugin-induced apoptosis. When the disruption of mitochondrial membrane potential and mitochondrial cytochrome c release into cytosol were investigated in J/Neo cells following treatment with mollugin (15–30 μM), the level of mitochondrial membrane potential disruption as well as cytochrome c release was enhanced by mollugin in a dose-dependent manner. In addition, the activation of caspase-9 and -3, and the degradation of PARP were enhanced in accordance with the mitochondrial cytochrome c release. In contrast, these molugininduced apoptotic events were completely abrogated in J/Bcl-xL cells overexpressing Bcl-xL, indicating that mollugin-induced activation of mitochondria-dependent caspase cascade, which could be prevented by Bcl-xL, was crucial for the induced apoptosis. In the antitumor agent-induced apoptosis of tumor cells, ER stress-mediated activation of caspase-8 has coupled to mitochondria-dependent death signaling as an upstream event of mitochondrial cytochrome c release (Jun et al., 2007; Jimbo et al., 2003). A proposed mechanism underlying contribution of ER stress-mediated activation of caspase-8 to mitochondrial cytochrome c release is that the active caspase-8 cleaves the Bid protein (26 kDa) into a truncated form, tBid (15 kDa) which is known to be translocated to mitochondria in order to mediate cytochrome c release into cytosol (Desagher et al., 1999; Li et al., 1998). In this regards, we investigated whether mollugin-induced apoptosis in J/Neo cells was accompanied by the activation of three pro-apoptotic regulators such as caspase-8, JNK, and caspase-12, which was previously induced by ER stress (Aoki et al., 2002; Jun et al., 2007; Rao et al., 2001; Urano et al., 2000). Although the generation of tBid was not observed by Western blot analysis in J/Neo cells following exposure to 15–30 μM mollugin, presumably due to the short half-life of tBid, a decrease in the level of Bid protein was detected, in accordance with the mollugin-induced caspase-8 activation as well as mitochondrial cytochrome c release. Like the Bid protein, the FLICE inhibitory protein (FLIP) was known to be the substrate of caspase-8 (Boehrer et al., 2006). Along with the molluginmediated activation of caspase-8 as well as resultant reduction in the level of the Bid protein, the FLIP level was also reduced, supporting that the active form of caspase-8, which was detected by Western blot analysis in Jurkat T cells following exposure to mollugin, was enzymatically active enough to cleave the substrates, Bid and FLIP. In addition to activating caspase-8, mollugin appeared to activate JNK, which was previously translocated to the mitochondrial membrane in order to stimulate the phosphorylation of Bim, leading to mitochondrial cytochrome c release (Urano et al., 2000). The IRE1α localized to the ER membrane was known to play a critical role in ER stressmediated activation of JNK (Urano et al., 2000; Kyriakis et al., 1994). These previous and current results suggested that mollugin-mediated cytochrome c release might be initiated through the Bid cleavage by caspase-8 and/or through the JNK activation. However, since it was also reported that caspase-8 was activated downstream of caspase-3 to comprise a positive feedback loop involving tBid-mediated mitochondrial cytochrome c release in the chemical agent-induced apoptosis of tumor cells (Tang et al., 2000), current results could not exclude the possibility that the caspase-8 activation was not the initial signal provoking mitochondrial cytochrome c release but was downstream of the caspase-3 activation in J/Neo cells treated with mollugin. Besides the activation of caspase-8 and JNK, caspase-12 was previously activated in response to ER stress (Nakagawa and Yuan, 2000; Nakagawa et al., 2000). In this process, Ca2+released from the ER in response to ER stress activates the m-calpain, which is then translocated from cytosol to ER to cleave off the CARD pro-domain of caspase-12, resulting in caspase-12 activation. In the presence of mollugin (15–30 μM), a slight decrease in the level of procaspase-12 (55 kDa) as well as an enhancement in the level of in vitro caspase-12 activity was detected in J/Neo cells, demonstrating the activation of caspase-12 following exposure to mollugin. Since the active caspase12 could directly activate procaspase-9, independently of both the mitochondrial cytochrome c and Apaf-1 (Morishima et al., 2002; Rao et al., 2002), and since the activation of caspase-9 within apoptosome and subsequent activation of caspase-3 were known to occur through reciprocal activation of caspase-9 and caspase-3 (Twiddy and Cain, 2007; Zou et al., 2003), these results indicated that the procaspase-12 activation occurred in parallel with mitochondrial cytochrome c release in order to synergize the caspase-3 activation targeted by the apoptosome in J/Neo cells following exposure to mollugin. While mollugin-induced caspase-8 activation, as well as mitochondrial cytochrome c release and subsequent activation of caspase cascade were completely prevented in J/Bcl-xL cells, the activation of JNK and caspase-12 was prevented to a much lesser extent. This indicated that the mollugin-induced activation of JNK and caspase-12, which was mediated via ER stress, was more refractory to the antiapoptotic role of Bcl-xL as compared with the mitochondria-dependent activation of caspase cascade in Jurkat T cells.
Although the presence of the pan-caspase inhibitor z-VAD-fmk at a concentration of 30 μM completely blocked mollugin-induced subG1 peak and most apoptotic events including the cleavage of procaspase-3 into 17 kDa active form as well as the activation of caspase-7 and -8 in J/Neo cells, it failed to completely prevent the cleavage of procaspase-3 into 19 kDa active form and the caspase-9 activation, in particular generation of 35 kDa active capase-9. The presence of zVAD-fmk also failed to exert suppressive effect on mollugin-induced phosphorylation of JNK. Since the active JNK can trigger mitochondrial cytochrome c release (Aoki et al., 2002; Urano et al., 2000), and since the proteolytic cleavage of 47 kDa procaspase-9 within the apoptosome appears to yield mainly 35/12 kDa active forms unless the feedback cleavage of 47 kDa procaspase-9 by 19 kDa active caspase-3 occurs (Twiddy and Cain, 2007; Zou et al., 2003), it was likely that mollugin-induced mitochondrial cytochrome c release was initiated by JNK rather than tBid generated from the caspase-8-dependent degradation of Bid. The notion that caspase-8 activation driven by 17 kDa active caspase-3 was a positive feedback mechanism, promoting mitochondrial cytochrome c release through the action of tBid, became more evident by examining the caspase-8 activation in the presence of the caspase-9 inhibitor z-LEHD-fmk (30 μM) or the caspase-3 inhibitor z-DEVD-fmk (30 μM), because either the inhibition of caspase-9 activity by z-LEHD-fmk or the inhibition of caspase-3 activity by z-DEVD-fmk could completely block mollugininduced activation of caspase-8 as well as generation of active caspase-3 (17 kDa) in J/Neo cells. While 37 kDa active caspase-9 was detected at a significantly reduced level in the presence of z-LEHDfmk or z-DEVD-fmk, 35 kDa active caspase-9 was detected at a comparable level to that of the mollugin-treated control cells. Under these conditions, only 19 kDa active caspase-3 was generated without inducing caspase-7 activation and PARP degradation. These results also confirmed that reciprocal activation of caspase-9 and caspase-3 downstream of mitochondrial cytochrome c release, which could generate two forms (37/35 kDa) of active caspase-9 and active form (17 kDa) of caspase-3, was crucial for mollugin-induced activation of caspase-7 and degradation of PARP. It is noteworthy that along with human caspase-12, human caspase-4 and -5, both of which have a CARD pro-domain at the N-terminal and show a high similarity to mouse caspase-12, have been proposed to play roles in the ER stressmediated apoptosis of human cells (Momoi, 2004). In this regard, to examine the involvement of ER stress-induced activation of caspase12 or caspase-4 in mollugin-mediated apoptosis of J/Neo cells, the effect of caspase-12 inhibitor z-ATAD-fmk or caspase-4 inhibitor zLEVD-fmk on mollugin-mediated apoptotic events was also investigated. In the presence of z-ATAD-fmk (4 μM), mollugin-mediated apoptotic sub-G1 peak, activation of caspase-8 and -7, and degradation of PARP were completely abrogated along with a significant decrease in the level of active forms (37/35 kDa) of caspase-9 and no generation of active caspase-3 (17 kDa), whereas the proteolytic cleavage of procaspase-3 into active form (19 kDa) was observed. In contrast, although the presence of z-LEVD-fmk (4 μM) suppressed mollugin-induced sub-G1 peak and caspase-8 activation by ∼50% with allowing generation of both 19 kDa active caspase-3 and much less amount of 17 kDa active caspase-3, there was no detectable suppressive effect on mollugin-induced activation of caspase-9 and -7, and degradation of PARP. Since 17 kDa active form was known to be efficient rather than 19 kDa active form of caspase-3 in exerting the proapoptotic effects including activation of caspase-6 and caspase-8, and degradation of PARP (Denault et al., 2007; Twiddy and Cain, 2007; Zou et al., 2003), current results indicated that when the caspase-12 activity was inhibited by z-ATAD-fmk, the mitochondria-dependent activation of caspase-9 and -3 was not provoked to a sufficient level required for subsequent activation of caspase-8 and -7, and degradation of PARP in J/Neo cells treated with mollugin. Current results also suggested that the inhibition of caspase-4 activity by zLEVD-fmk did not interfere with mitochondria-dependent activation of caspase-9 but did suppress in part the proteolytic cleavage of procaspase-3 into 17 kDa active caspase-3 required for the activation of caspase-8. Consequently, these results suggested that ER stressinduced activation of caspase-12 rather than caspase-4 was critical for the mitochondria-dependent activation of caspase-9 and -3, leading to activation of caspase-8 and -7 and degradation of PARP during mollugin-induced apoptosis of J/Neo cells. Recently, by in vitro caspase activity assay using recombinant human caspases, Pereira and Song have demonstrated that the inhibitory modes of six z-peptidefmk inhibitors (z-YVAD-fmk for caspase-1, z-VDVAD-fmk for caspase2, z-DEVD-fmk for caspase-3, z-VEID-fmk for caspase-6, z-IETD-fmk for caspase-8, and z-LEHD for caspase-9) are not specific for their designated caspases (Pereira and Song, 2008). Although these caspase inhibitors are widely used for cell-based assay at concentrations of up to 20–100 μM, the in vitro caspase activity assay has shown that the caspase inhibitors possess cross-reactivity toward non-targeted caspases at a concentration of 1 μM, and each of them cause complete inhibition of caspase-3, -7, and -8 activities at a higher concentration of 10 μM. In our hands, however, the minimal concentration of z-VAD-fmk, z-LEHD-fmk, or z-DEVD-fmk to completely prevent mollugin-induced apoptosis of Jurkat T cells was ∼30 μM, because mollugin-induced apoptotic sub-G1 peak was not abrogated by pretreatment with z-VADfmk, z-LEHD-fmk, or z-LEVD-fmk at concentrations of up to 20 μM (data not shown). On the other hand, the minimal concentration of the caspase-12 inhibitor z-ATAD-fmk to prevent the mollugin-induced apoptosis appeared to be ∼4 μM, because mollugin-induced apoptotic sub-G1 peak was completely abrogated at a concentration of 4 μM, but not at concentrations of up to 2 μM (data not shown). Since the in vitro caspase-12 activity assay using the cell lysate of J/Neo cells exposed to mollugin revealed that while z-ATAD-fmk (1–4 μM) could specifically inhibit the caspase-12 activity by ∼50%, it was likely that the inhibitory effect of z-ATAD-fmk (4 μM) on the mollugin-induced apoptotic signaling pathway was caused by its specific inhibition toward the induced caspase-12 activity.
Toexamineifthereisadifferenceinthe apoptogeniceffect of mollugin on tumor cells and normal cells, we have compared the cytotoxicity of mollugin against leukemia Jurkat T cells (J/Neo) with that against normal human T cells. The IC50 value for resting human T cells, activated T cells, andJurkatTcellsappearedtobeN30μM,≥30μM,and23μM,respectively. This indicated that normal human T cells, in particular, in the resting state were more refractory to the cytotoxicity of mollugin as compared with Jurkat T cells, which might permit the better application of mollugin to chemotherapeutic treatments. Since the cytotoxicity of mollugin in T cells was attributable to apoptotic cell death provoked by ER stress-mediated activation of JNK and caspase-12, and subsequent mitochondria-dependent activation of caspase cascade, it was likely that the best resistance of unstimulated peripheral T cells, as compared with other cells tested against mollugin, might be due to a poorly developed endoplasmic reticulum and mitochondria, and a low level of death signaling mediators in unstimulated peripheral T cells. However, the precise molecular mechanism underlying the best resistance of unstimulated T cells to mollugin-induced apoptosis still remains to be elucidated.
In summary, these results demonstrated that the cytotoxicity of mollugin (15–30 μM) toward human acute leukemia Jurkat T cells was attributable to apoptotic cell death. The mollugin-induced apoptotic DNA fragmentation was mediated by ER stress-mediated activation of JNK and caspase-12 and subsequent activation of mitochondriadependent caspase cascade including caspase-9, -3, -8, and -7, leading toPARP degradation,whichwas negatively regulated byoverexpression of Bcl-xL. Normal human T cells, in the unstimulated state, were more refractory to the cytotoxicity of mollugin as compared to Jurkat T cells. These results have provided the cellular mechanism underlying cytotoxic effect of mollugin on human acute leukemia Jurkat T cells, and will be useful for evaluating the potency of mollugin, a purified substance from the roots of Rubia cordifolia L., as an antitumor agent.

References

Aoki, H., Kang, P.M., Hampe, J., Yoshimura, K., Noma, T., Matsuzaki, M., Izumo, S., 2002. Direct activation of mitochondrial apoptosis machinery by c-Jun N-terminal kinase in adult cardiac myocytes. J. Biol. Chem. 277, 10244–10250.
Basu, S., Ghosh, A., Hazra, B., 2005. Evaluation of the antibacterial activity of Ventilago madraspatana Gaertn., Rubia cordifolia Linn. and Lantana camara Linn.: isolation
of emodin and physcion as active antibacterial agents. Phytother. Res. 19, 888–894.
Barut, S., Unlu, Y.A., Karaoglan, A., Tunçdemir, M., Dagistanli, F.K., Oztürk, M., Colak, A., 2005. The neuroprotective effects of z-DEVD.fmk, a caspase-3 inhibitor, on traumatic spinal cord injury in rats. Surg. Neurol. 64, 213–220.
Boehrer, S., Nowak, D., Puccetti, E., Ruthardt, M., Sattler, N., Trepohl, B., Schneider, B., Hoelzer, D., Mitrou, P.S., Chow, K.U., 2006. Prostate-apoptosis-response-gene-4 increases sensitivity to TRAIL-induced apoptosis. Leuk. Res. 30, 597–605.
Cai, Y., Sun, M., Xing, J., Corke, H., 2004. Antioxidant phenolic constituents in roots of Rheum officinale and Rubia cordifolia: structure-radical scavenging activity relationships. J. Agric. Food Chem. 52, 7884–7890.
Carosio, R., Zuccari, G., Orienti, I., Mangraviti, S., Montaldo, P.G., 2007. Sodium ascorbate induces apoptosis in neuroblastoma cell lines by interfering with iron uptake. Mol. Cancer 6, 55.
Chang, L.C., Chavez, D., Gills, J.J., Fong, H.H.S., Pezzuto, H.M., Kinghorn, A.D., 2000. Rubiasins A-C, new anthracene derivatives from the roots and stems of Rubia cordifolia. Tertahedron Lett. 41, 7157–7162.
Chung, M.I., Jou, S.J., Cheng, T.H., Lin, C.N., Ko, F.N., Teng, C.M., 1994. Antiplatelet constituents of formosan Rubia akane. J. Nat. Prod. 57, 313–316.
Denault, J.B., Eckelman, B.P., Shin, H., Pop, C., Salvesen, G.S., 2007. Caspase 3 attenuates XIAP (X-linked inhibitor of apoptosis protein)-mediated inhibition of caspase 9. Biochem. J. 405, 11–19.
Desagher, S., Martinou, J.C., 2000. Mitochondria as the central control point of apoptosis. Trends Cell Biol. 10, 369–377.
Desagher, S., Osen-Sand, A., Nichols, A., Eskes, R., Montessuit, S., Lauper, S., Maundrell, K., Antonsson, B., Martinou, J.C., 1999. Bid-induced conformational change of Bax is responsible for mitochondrial cytochrome c release during apoptosis. J. Cell Biol. 144, 891–901.
Friesen, C., Herr, I., Krammer, P.H., Debatin, K.M., 1996. Involvement of the CD95 (APO1/FAS) receptor/ligand system in drug-induced apoptosis in leukemia cells. Nat. Med. 2, 574–578.
Gilani, A.H., Janbaz, K.H., 1995. Effect of Rubia cordifolia extract on acetaminophen and CCl4-induced hepatotoxicity. Phytother. Res. 9, 372–375.
Han, B.H., Park, M.K., Park, Y.H., 1990. A lignan from Rubia akane. Arch. Pharm. Res. 13, 289–291.
Hannun, Y.A., 1997. Apoptosis and dilemma of cancer chemotherapy. Blood 89, 1845–1853.
Herr, I., Debatin, K.M., 2001. Cellular stress response and apoptosis in cancer therapy. Blood 98, 2603–2614.
Ho, L.-K., Don, M.-J., Chen, H.-C., Yeh, S.-F., Chen, J.-M., 1996. Inhibition of hepatitis B surface antigen secretion on human hepatoma cells; components from Rubia cordifolia. J. Nat. Prod. 59, 330–333.
Itokawa, H., Mihara, K., Takeya, K., 1983. Studies on a novel anthraquinone and its glycosides isolated from Rubia cordifolia and R. akane. Chem. Pharm. Bull. 31, 2353–2358.
Itokawa, H., Ibraheim, Z.Z., Qiao, Y.F., Takeya, K., 1993. Anthraquinones, naphthohydroquinones and naphthohydroquinone dimers from Rubia cordifolia and their cytotoxic activity. Chem. Pharm. Bull. 41, 1869–1872.
Jiang, C.C., Chen, L.H., Gillespie, S., Wang, Y.F., Kiejda, K.A., Zhang, X.D., Hersey, P., 2007. Inhibition of MEK sensitizes human melanoma cells to endoplasmic reticulum stress-induced apoptosis. Cancer Res 67, 9750–9761.
Jimbo, A., Fujita, E., Kouroku, Y., Ohnishi, J., Inohara, N., Kuida, K., Sakamaki, K., Yonehara, S., Momoi, T., 2003. ER stress induces caspase-8 activation, stimulating cytochrome c release and caspase-9 activation. Exp. Cell Res. 283, 156–166.
Jun, D.Y., Kim, J.S., Park, H.S., Han, C.R., Fang, Z., Woo, M.H., Rhee, I.K., Kim, Y.H., 2007. Apoptogenic activity of Zanthoxylum schinifolium toward human acute leukemia Jurkat T cells is associated with ER stress-mediated caspase-8 activation that stimulates mitochondria-dependent or-independent caspase cascade. Carcinogenesis 28, 1303–1313.
Jun, D.Y., Park, H.S., Kim, J.S., Kim, J.S., Park, W., Song, B.H., Kim, H.S., Taub, D., Kim, Y.H., 2008. 17α-estradiol arrests cell cycle progression at G2/M and induces apoptotic cell death in human acute leukemia Jurkat T cells. Toxicol. Appl. Pharmacol. 231, 401–412.
Juo, P., Woo, M.S., Kuo, C.J., Signorelli, P., Biemann, H.P., Hannun, Y.A., Blenis, J., 1999. FADD is required for multiple signaling events downstream of the receptor Fas. Cell Growth Differ. 10, 797–804.
Kaufmann, S.H., Earnshaw, W.C., 2000. Induction of apoptosis by cancer chemotherapy. Exp. Cell Res. 256, 42–49.
Kim, Y.H., Proust, J.J., Buchholz, M.J., Chrest, F.J., Nordin, A.A., 1992. Expression of the murine homologue of the cell cycle control protein p34cdc2 in T lymphocytes. J. Immunol. 149, 17–23.
Kluck, R.M., Bossy-Wetzel, E., Green, D.R., Newmeyer, D.D., 1997. The release of cytochrome c from mitochondria: a primary site for Bcl-2 regulation of apoptosis. Science 275, 1132–1136.
Kyriakis, J.M., Banerjee, P., Nikolakaki, E., Dai, T., Rubie, E.A., Ahmad, M.F., Avruch, J., Woodgett, J.R., 1994. The stress-activated protein kinase subfamily of c-Jun kinases. Nature 369, 156–160.
Lee, J-E., Hirrotsuyanagi, Y., Fukuya, H., Kondo, K., Takeya, K., 2008. New cytotoxic bicyclic hexapeptides, RA-XXII and RA-XXIV, from Rubia cordifolia L. Chem. Pharm. Bull 56, 730–733.
Li, H., Zhu, H., Xu, C., Yuan, J., 1998. Cleavage of Bid by caspase 8 mediates the mitochondrial damage in the Fas pathway of apoptosis. Cell 94, 491–501.
Lu, Y., Liu, R., Sun, C., Pan, Y., 2007. An effective high-speed countercurrent chromatographic method for preparative isolation and purification of mollugin directly from the ethanol extract of the Chinese medicinal plant Rubia cordifolia. J. Sep. Sci. 30, 1313–1317.
Momoi, T., 2004. Caspases involved in ER stress-mediated cell death. J. Chem. Neuroanat. 28, 101–105.
Morishima, N., Nakanishi, K., Takenouchi, H., Shibata, T., Yasuhiko, Y., 2002. An endoplasmic reticulum stress-specific caspase cascade in apoptosis. Cytochrome c-independent activation of caspase-9 by caspase-12. J. Biol. Chem. 277, 34287–34294.
Morishima, N., Nakanishi, K., Tsuchiya, K., Shibata, T., Seiwa, E., 2004. Translocation of Bim to the endoplasmic reticulum (ER) mediates ER stress signaling for activation of caspase-12 during ER stress-induced apoptosis. J. Biol. Chem. 279, 50375–50381.
Muller, M., Strand, S., Hug, H., Heinemann, E.M., Walczak, H., Hofmann, W.J., Stremmel, W., Krammer, P.H., Galle, P.R., 1997. Drug-induced apoptosis in hepatoma cells is mediated by the CD95 (APO-1/Fas) receptor/ligand system and involves activation of wild-type p53. J. Clin. Invest. 99, 403–413.
Nagarkatti, N., Davis, B.A., 2003. Tamoxifen induces apoptosis in Fas+ tumor cells by upregulating the expression of Fas ligand. Cancer Chemother Phamarcol. 51, 284–290.
Nakagawa, T., Yuan, J., 2000. Cross-talk between two cysteine protease families; activation of caspase-12 by calpain in apoptosis. J. Cell Biol. 150, 887–894.
Nakagawa, T., Zhu, H., Morishima, N., Li, E., Xu, J., Yankner, B.A., Yuan, J., 2000. Caspase12 mediates endoplasmic-reticulum-specific apoptosis and cytotoxicity by amyloid-beta. Nature 403, 98–103.
Ozoren, N., Kim, K.H., Burns, T.F., Dicker, D.T., Moscioni, A.D., El-Deiry, W.S., 2000. The caspase 9 inhibitor Z-LEHD-FMK protects human liver cells while permitting death of cancer cells exposed to tumor necrosis factor-related apoptosis-inducing ligand. Cancer Res. 60, 6259–6265.
Pereira, N.A., Song, Z., 2008. Some commonly used caspase substrates and inhibitors lack the specificity required to monitor individual caspase activity. Biochem. Biophys. Res. Commun. 377, 873–877.
Rao, R.V., Hermel, E., Castro-Obregon, S., del Rio, G., Lisa, M., Ellerby, L.M., Ellerby, H.M., Bredesen, D.E., 2001. Coupling endoplasmic reticulum stress to the cell death program; mechanism of caspase activation. J. Biol Chem. 276, 33869–33874.
Rao,R.V.,Castro-Obregon,S.,Frankowski,H.,Schuler,M.,Stoka,V.,delRio,G., Bredesen,D.E., Ellerby, H.M., 2002. Coupling endoplasmic reticulum stress to the cell death program; an Apaf-1-independent intrinsic pathway. J. Biol. Chem. 277, 21836–21842.
Slee, E.A., Zhu, H., Chow, S.C., MacFarlane, M., Nicholson, D.W., Cohen, G.M., 1996. Benzyloxycarbonyl-Val-Ala-Asp (OMe) fluoromethylketone (z-VAD-fmk) inhibits apoptosis by blocking the processing CPP32. Biochem J. 315, 21–24.
Smiley, S.T., Reers, M., Mottola-Hartshorn, C., Lin, M., Chen, A., Smith, T.W., Steele Jr., G.D., Chen, L.B., 1991. Intracellular heterogeneity in mitochondrial membrane potentials revealed by a J-aggregate-forming lipophilic cation JC-1. Proc. Natl. Acad. Sci. U.S.A. 88, 3671–3675.
Son, J.K., Jung, S.J., Jung, J.H., Fang, Z., Lee, C.S., Seo, C.S., Moon, D.C., Min, B.S., Kim, M.R., Woo, M.H., 2008. Anticancer constituents from the roots of Rubia cordifolia L. Chem. Pharm. Bull. 56, 213–216.
Tang, D., Lahti, J.M., Kidd, V.J., 2000. Caspase-8 activation and bid cleavage contribute to MCF7 cellular execution in a caspase-3-dependent manner during staurosporinemediated apoptosis. J. Biol. Chem. 275, 9303–9307.
Tezuka, Y., Irikawa, S., Kaneko, T., Banskota, A.H., Nagaoka, T., Xiong, Q., Hase, K., Kadota, S., 2001. Screening of Chinese herbal drug extracts for inhibitory activity on nitric oxide production and identification of an active compound of Zanthoxylum bungeanum. J. Ethnopharmacol. 77, 209–217.
Tse, W.P., Cheng, C.H., Che, C.T., Zhao, M., Lin, Z.X., 2007. Induction of apoptosis underlies the Radix Rubiae-mediated anti-proliferative action on human epidermal keratinocytes: implications for psoriasis treatment. Int. J. Mol. Med. 20, 663–672.
Twiddy, D., Cain, K., 2007. Caspase-9 cleavage, do you need it? Biochem. J. 405, e1–e2. Urano, F., Wang, X., Bertolotti, A., Zhang, Y., Chung, P., Harding, H.P., Ron, D., 2000. Coupling of stress in the ER to activation of JNK protein kinases by transmembrane protein kinase IRE1. Science 287, 664–666.
Wallach, D., Boldin, M., Varfolomeev, E., Beyaert, R., Vandenabeele, P., Fiers, W., 1997. Cell death induction by receptors of the THF family: towards a molecular understanding. FEBS Lett 410, 96–106.
Yang, S.H., Chien, C.M., Chang, LS., Lin, S.R., 2008. Cardiotoxin III-induced apoptosis is mediated by Ca2+-dependent caspase-12 activation in K562 cells. J. Biochem. Mol. Toxicol. 22, 209–218.
Zou, H., Yang, R., Hao, J., Wang, J., Sun, C., Fesik, S.W., Wu, J.C., Tomaselli, K.J., Armstrong,R.C., 2003. Regulation of the Apaf-1/caspase-9 apoptosome by caspase-3 and XIAP.J. Biol. Chem. 278, 8091–8098.