Constitutive androstane receptor agonist CITCO inhibits growth and expansion of brain tumour stem cells

Background: Brain tumours present unique challenges to conventional therapies and pose major health problems around the world. Brain tumour stem cells (BTSCs) represent a small fraction of tumour cells that maintain growth, drug resistance and recurrence properties. Constitutive androstane receptor (CAR) is a nuclear receptor transcription factor that regulates drug metabolism and homoeostasis. In this study, we examined the effect of CAR agonist, 6-(4-chlorophenyl)imidazo[2,1-b][1,3]thiazole-5-carbaldehydeO-(3,4-dichlorobenzyl)oxime (CITCO) on BTSCs. Methods: The expression of CAR in BTSCs was detected by quantitative RT–PCR and western blot. The antiproliferative effect of CITCO on BTSCs was determined by WST-1 and 3H thymidine uptake assays. The effect of CITCO on CD133 expression, cell cycle progression and apoptosis in BTSCs was analysed by immunostaining and flow cytometry. The in vivo effect of CITCO was studied using subcutaneous (s.c.) BTSC xenograft in nude mice. Results: We show for the first time that BTSCs express altered levels of nuclear receptors compared with glioma cells. The expression of CAR mRNA and protein was low in BTSCs and that increased following treatment with CITCO in culture. CITCO induced a dose-dependent decrease in growth and expansion of CD133+ BTSCs as gliospheres in culture. Cell cycle arrest and apoptosis in BTSCs were induced by CITCO, but not in normal astrocytes. Growth of s.c BTSC xenograft in nude mice was also inhibited by CITCO. Conclusion: These findings indicate that CITCO inhibits the growth and expansion of BTSCs, suggesting the use of CAR agonists for the treatment of brain tumour.

Brain tumours are among the most devastating cancers that present unique challenges to therapy and pose major health problems around the world. Among many different types of tumours, glioblastoma is the most frequent primary malignant brain tumour in adults. Standard therapy includes surgical resection to the extent that is safely feasible, followed by radiation and chemotherapy, which have significant side effects and limited efficacy (Deorah et al, 2006), (Peacock and Lesser, 2006). Despite recent advances in surgery, radiation and chemotherapy, median survival is less than 1 year and a cure for brain tumour remains elusive. Multidrug resistance and fast recurrence are some of the challenges in combating brain tumours. Cancer stem cells (CSCs) are a small population of cells in cancer tissues with asymmetric division, self-renewal and tumour initiation capabilities. In acute myeloid leukaemia, CSCs were first identified (Bonnet and Dick, 1997) and subsequently in breast (Al-Hajj et al, 2003), prostate (Patrawala et al, 2006), liver (Yang et al, 2008), colon O'Brien et al, 2007), pancreas (Li et al, 2007) and skin cancers (Schatton et al, 2008). Brain tumour stem cells (BTSCs) have also been isolated from gliomas that are positive for CD133 and posses tumour initiation potential in NOD/SCID mice (Singh et al, 2004). The BTSCs are resistant to standard therapies and are considered responsible for the recurrence of brain tumours after radiation and chemotherapy in patients (Singh et al, 2004;Bao et al, 2006). The failure to cure brain tumour has been attributed to the fact that typical therapies target rapidly proliferating tumour cells, which respond transiently, while sparing the highly tumourigenic BTSCs (Bao et al, 2006;Stupp and Hegi, 2007).
Nuclear hormone receptors are a family of transcriptional factors that regulate cell growth, differentiation and homoeostasis. Genomic studies have cloned 48 and 50 nuclear receptors in human and rodents, respectively, and many of them have therapeutic values in human diseases (Blumberg and Evans, 1998). The effect of nuclear receptor agonists on brain tumours has been shown recently (Naveilhan et al, 1994;Grommes et al, 2006;Papi et al, 2009), but their use in targeting BTSCs is not known. Constitutive androstane receptor (CAR, NR1I3) is an orphan nuclear receptor that contains a DNA-binding domain but uniquely lacking an activation domain (Baes et al, 1994). The CAR is retained in the cytoplasm by forming a complex with phosphatase 2A, HSP90 and cytosolic CAR retention protein (Kobayashi et al, 2003). Phenobarbital, 5b-pregnane-3,20-dione, and 5-androstan-3-ol are known endogenous CAR ligands . The hepatomitogen 1,4-Bis[2-(3,5-dichloropyridyloxy)]benzene (TCPOBOP) is a synthetic agonist for murine CAR (Tzameli et al, 2000) and 6-(4-chlorophenyl)imidazo [2,1-b] [1,3]thiazole-5-carbaldehydeO-(3,4-dichlorobenzyl)oxime (CITCO) is an imidazothiazole derivative that functions as a selective agonist for human CAR (Maglich et al, 2003). Upon activation with specific agonist, CAR translocates into the nucleus and binds to the response elements as monomers or CAR/RXR heterodimers (Baes et al, 1994). The CAR functions as a xenobiotic receptor that regulates detoxification and clearance of toxic substances from the liver ; however, its role in cancer is not known.
In this study, we show that CAR agonist CITCO inhibits growth and expansion of BTSCs in culture and xenograft model. Our findings highlight that BTSCs can be targeted through CAR for the treatment of brain tumour patients.

Tissues, cells and reagents
The T98G and U87MG human glioma cells were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA). The DB29 and DB33 human glioma cells were established in the laboratory by culturing brain tumour samples obtained from the tissue repository at Methodist Research Institute (Indianapolis, IN, USA) with an IRB approved protocol. The glioma cells were cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% FBS, 1 mM sodium pyruvate, 100 U ml À1 penicillin G, 100 mg ml À1 streptomycin, 2 mM glutamine, 1 mM MEM nonessential amino acids and 50 mM 2b-mercaptoethanol in 5% CO 2 incubator at 371C. The glioma cells were dissociated using 0.25% trypsin with 0.05 mM EDTA solution and subcultured once in 3 -5 days. Primary human astrocytes (HAs) were obtained from ScienCell (Carlsbad, CA, USA) and cultured in astrocyte medium. The human CAR agonist CITCO was obtained from BioMol (Plymouth Meeting, PA, USA). The CITCO is an imidazothiazole derivative (MW 436.7) with an EC 50 of 49 nm and 450-fold selectivity to CAR over pregnane X receptor (PXR), and no activity on other nuclear receptors.

Gliosphere culture
To generate gliospheres, we have adopted a culture condition as standardised in our laboratory . Briefly, the glioma cells cultured as monolayer in DMEM were dissociated using trypsin -EDTA and cultured in neurobasal medium (NBM) supplemented with B27 (Invitrogen, Madison, WI, USA) in the presence of 10 ng ml À1 EGF and bFGF. The cells were cultured in 12-well plates (5 Â 10 4 per ml per well) with different doses of CITCO in 5% CO 2 incubator at 371C. The gliospheres generated in primary cultures were photographed after 5 -10 days using BX40 Olympus microscope (Olympus America Inc., Center Valley, PA, USA).

Isolation of BTSCs and CD133 þ BTSCs
To isolate BTSCs, the gliospheres were harvested after 5-7 days and dissociated by incubating (10 Â 10 6 ml À1 ) in Accutase (Sigma, St Louis, MO, USA) at 371C for 30 min. The cells were washed and resuspended in fresh NBM, and used as BTSCs for the experiments. To isolate CD133 þ BTSCs, the dissociated gliosphere cells (BTSCs) were resuspended (10 Â 10 6 ) in 200 ml of PBE buffer (PBS, pH7.2, 0.5% BSA, 2 mM EDTA). Biotin conjugated anti-CD133 antibody (20 ml) and FcR blocking reagent (30 ml) (Miltenyi Biotech, Auburn, CA, USA) were added and incubated at 41C for 30 min. The cells were washed and incubated in 200 ml PBE buffer with 50 ml streptavidin-conjugated magnetic microbeads (Miltenyi Biotech, Auburn, CA, USA) at 41C for 30 min. The cells were passed through a MACS-LS column equilibrated with PBE buffer placed on a Midi-Macs Magnet (Myltenyi Biotech Inc., Auburn, CA, USA), and the CD133 negative cells in the flow-through were collected. The column was removed from the Magnet, and the CD133-positive cells were flushed out with the buffer using a plunger. The cells were washed and resuspended in fresh NBM for experiments.

Immunostaining and flow cytometry
The gliospheres were cultured in fresh NBM with B27 and 10 ng ml À1 EGF þ bFGF in the presence of different concentrations of CITCO in 12-well tissue culture plates in 5% CO 2 incubator at 371C. After 48 h, the gliospheres were fixed in 1% paraformaldehyde and stained with anti-Ki67 Ab (1 : 100, Santa Cruz, Santa Cruz, CA, USA) in PBS with 1% BSA at 41C for 2 h. Alexa 488conjugated 2nd Ab (1 : 1500, Cell Signaling, Danvers, MA, USA) diluted in PBS with 1% BSA was added for 1 h and photographed using a fluorescent microscope (DMRB, Leica, Bannockburn, IL, USA). To determine the percentage of CD133 þ BTSCs, gliospheres cultured with different doses of CITCO for 48 h were dissociated using Accutase, followed by resuspension in PBS containing 0.1% BSA and incubated with anti-CD133 Ab (1 : 100, Cell Signaling) at 41C for 2 h. The cells were washed and incubated with Alexa 488conjugated secondary Ab at 41C for 1 h, and acquired using FACS Calibur Flow Cytometer (BD Biosciences, San Jose, CA, USA) and analysed using FlowJo 8.2.6 software (Ashland, OR, USA).

Quantitative real-time RT -PCR
Total RNA was extracted from glioma and BTSCs using RNesy kit (Qiagen, Valencia, CA, USA). Equal amount of RNA was reverse transcribed into cDNA, and quantitative RT-PCR (qRT-PCR) was performed using 384-well TaqMan Low Density Human Nuclear Receptor Array Card (384 well) in 7900 HT Fast Real time PCR system (Applied Biosystems, Foster City, CA, USA). To detect CAR transcription, 5 mg total RNA was reverse transcribed into cDNA by incubating in 10 ml reaction of random hexamer primers and master mix from TaqMan reverse transcription kit (Applied Biosystems, Branchburg, NJ, USA). For qRT-PCR, 2 ml of the cDNA was amplified using TaqMan Universal Master Mix with optimised concentrations of CAR primer sets and probes in a standard optical 96-well reaction plate. The results were analysed using the Prism 7900 (Applied Biosystems, Carlsbad, CA, USA) relative quantification (delta delta Ct) study software (Mo et al, 2008). The level of CAR transcription was normalised to 18S and expressed as fold change compared with control.

Cell cycle analysis
To determine the effect of CITCO on cell cycle progression, gliospheres were cultured in NBM with B27 and 10 ng ml À1 EGF þ bFGF in the presence of different concentrations of CITCO. After 48 h, the cells were dissociated with Accutase and incubated in PBS containing 100 mg ml À1 propidium iodide, 0.6% NP-40 and 20 mg ml À1 RNase (Sigma Chemicals, St Louis, MO, USA) at 41C for 1 h. The percentages of BTSCs at different cell cycle stages (G0/G1, G2/M and S phases) were determined based on DNA content by flow cytometry using FACS Calibur Flow Cytometer (BD Biosciences), and analysed using ModFit LT2.0 software (Verity software house, Topsham, ME, USA).

Apoptosis assay
To determine the effect of CITCO on apoptosis, BTSCs were cultured in NBM with B27 and 10 ng ml À1 EGF þ bFGF in the presence of CITCO. After 48 h, the cells were harvested, washed in PBS and stained with Annexin V-FITC (Roche, Indianapolis, IN, USA) in binding buffer (0.1 M Hepes/NaOH, pH 7.4, 1.4 M NaCl, 0.2 mM) containing 100 mg ml À1 propidium iodide according to the manufacturer's instruction (Roche). The cells were incubated at room temperature for 30 min in dark, acquired using FACS Calibur Flow Cytometer (BD Biosciences) and analysed using FlowJo 8.2.6 software.

Induction, treatment and evaluation of BTSC xenograft
To determine the in vivo effect of CAR agonists on BTSCs, we induced BTSC xenograft in nude mice. All the animal protocols used for in vivo experiments were reviewed and approved by the institutional animal care and use committees at Methodist Research Institute, and performed accordingly. Six-to eight-week-old male athymic nude mice were obtained from Harlan (Indianapolis, IN, USA) and maintained under specific pathogen-free conditions in the animal care facility at Methodist Research Institute. The BTSCs were isolated by dissociating U87MG gliospheres and transplanted (25 Â 10 4 ) subcutaneously in 100 ml NBM in the dorsum of 8-weekold nude mice. The mice were treated intraperitoneally with 25 and 100 mg CITCO in 25 ml DMSO on days 22, 24, 26, 30 and 36 following xenograft. The control mice received only 25 ml DMSO. Tumours were measured once a week till day 50 using a digital Vernier Calipers (Marathon, Ontario, Canada), and the tumour volume (T V ) was calculated as follows: (Euhus et al, 1986;Jiang et al, 2010;Johns et al, 2010). Each group contained at least four mice, and the experiments were repeated twice. On day 50, the BTSC xenografts were dissected, fixed in buffered formalin, embedded in paraffin and sliced into 6 mm thick sections. The sections were stained with haematoxylin and eosin (H&E), and photographed using a phase-contrast microscope (DMRB, Leica) (Mo et al, 2008). The tissue sections were deparafinised, followed by incubation with 10% goat serum in PBS to block non-specific binding sites and stained overnight with anti-Ki67 Ab (1 : 100, Santa Cruz) in PBS with 1% BSA at 41C. Alexa 488-conjugated 2nd Ab (Cell Signaling) diluted in PBS with 1% BSA was added for 1 h ,and the sections were photographed using a fluorescent microscope (DMRB, Leica).

Altered expression of nuclear receptors in BTSCs
To identify novel therapeutic targets for BTSCs, we examined the expression of nuclear receptors by qRT -PCR using TaqMan Low Density Human Nuclear Receptor Array Card (384 well). We found that T98G -BTSCs express altered levels of many nuclear receptors compared with the glioma cells (Table 1). Among the 48 nuclear receptors examined, BTSCs expressed detectable levels of ERRb and RXRg that were not detected in glioma cells. The BTSCs also showed X100-fold increase in the transcription of TLX and HNF4a, and between 10-and 100-fold increase in PXR, RORa, Abbreviations: bFGF ¼ basic fibroblast growth factor; BTSC ¼ brain tumour stem cell; cDNA ¼ complementary DNA; DMEM ¼ Dulbecco's modified Eagle's medium; EGF ¼ epidermal growth factor; FBS ¼ fetal bovine serum; NBM ¼ neurobasal medium. The T98G glioma cells were cultured as monolayer in DMEM with 10% FBS or as gliospheres in NBM with B27 and EGF+bFGF. After 5 days, the glioma cells were harvested using trypsin+EDTA, and gliospheres were dissociated using Accutase. Total RNA was extracted from glioma and BTSCs, and reverse transcribed into cDNA using random hexamer primers and TaqMan reverse transcription kit. The gene expression profile was determined using TaqMan Universal Master Mix with nuclear receptor primer sets and probes in an optical 384-well card using the 7900 Fast Sequence Detection Realtime PCR System (Applied Biosystems). The results were analysed using the Prism 7900 relative quantification (delta delta Ct) study software. The levels of nuclear receptor gene expression are normalised to 18 s, and the values are presented as fold change in BTSCs compared with glioma cells. RORg, NORI and LRH compared with glioma cells. Moreover, 15 nuclear receptors showed 0-to 10-fold increase in BTSCs (Table 1). On the other hand, 10 genes showed 0-to 2-fold decrease, 6 genes showed 2-to 10-fold decrease and RORb, SF1 and PNR showed more than 10-fold decrease in BTSCs compared with glioma. Five nuclear receptors tested were not detected either in glioma or BTSCs (Table 1). These results suggest that BTSCs express altered levels of nuclear receptors, which may serve as novel therapeutic targets for the treatment of brain tumour.

The CITCO induces the expression of CAR in BTSCs
Although many altered nuclear receptors are currently under investigation in our laboratory, in this manuscript, we present our findings on the use and mechanism of action of CAR and its agonist CITCO in targeting BTSCs for the treatment of glioma. As shown in Figure 1, qRT-PCR analyses confirmed that T98G-BTSC (A) and U87MG-BTSC (B) express significantly lower levels of CAR mRNA compared with the glioma cells. The CD133 þ BTSCs purified from T98G (A) and U87MG (B) gliospheres showed further decrease in CAR transcription. Interestingly, treatment with 2.5 mM CITCO resulted in a significant increase in the transcription of CAR in both T98G-BTSC and U87MG-BTSCs in 48 h. Western blot analyses showed that the T98G and U87MG glioma, and BTSCs express very low levels of CAR protein that increased significantly following treatment with 2.5 mM CITCO in 48 h. However, higher concentrations of CITCO (X10 mM) inhibited or abolished CAR expression in both T98G-and U87MG-derived BTSCs. Further analyses showed that the BTSCs isolated from DB29 and DB33 (C) gliospheres express lower levels of CAR compared with glioma cells, and treatment with 2.5 mM CITCO resulted in significant increase in the transcription of CAR. Western blot analyses also showed that the expression of CAR protein was low in DB29-BTSCs that increased significantly after treatment with 2.5 mM CITCO in 48 h (C). Higher concentrations of CITCO (X5 mM) abolished the expression of CAR in DB29-BTSCs. These results suggest that CITCO modulates the expression of CAR in BTSCs.

The CITCO inhibits the expansion of BTSCs
To study the use of CAR in the regulation of BTSCs, we examined the effect of CITCO on growth and expansion in culture. Microscopic analyses showed that in vitro culture of T98G, U87MG, DB29 and DB33 glioma cells in NBM with B27 and EGF þ bFGF resulted in the expansion of BTSCs as gliospheres in 5 days (Figures 2A and B). Interestingly, in vitro treatment with 1 and 5 mM CITCO for 48 h induced a dose-dependent decrease in gliosphere size in all four cell types tested. Immunofluorescent microscopy showed that the gliospheres are filled with actively dividing BTSCs as evidenced by Ki-67 staining. Treatment with CITCO resulted in a dose-dependent decrease in Ki-67 þ BTSCs in T98G (C), U87MG (D), DB29 (E) and DB33 (F) gliospheres.
To further determine the use of CAR in targeting BTSCs, we analysed the effect of CITCO on CD133 þ cells by flow cytometry. We found that T98G -BTSCs displayed 43% CD133 þ cells with a mean fluorescence intensity (MFI) of 186 that decreased dose dependently, reaching 16% with a MFI of 49 after treatment with 5 mM CITCO for 48 h (Figure 3). Similarly, U87MG -BTSCs showed 56% CD133 þ cells with a MFI of 254 that decreased to 16% with a MFI of 47 at a dose of 5 mM CITCO. Further analyses showed that the DB29 -BTSCs consist of 31% CD133 þ cells with a MFI of 231 that decreased to 16% with a MFI of 110 at 5 mM CITCO. Similarly,   (Figure 3). These results show that CITCO regulates the expansion of CD133 þ BTSCs, suggesting its significance in the treatment of glioma.

The CITCO inhibits the proliferation of BTSCs
To further test the therapeutic use of CAR in glioma, we examined the effect of CITCO on tumour cell proliferation in culture.
As shown in Figure 4, we found that in vitro culture of T98G (A, B) and U87MG (C, D) cells in DMEM as monolayer (glioma) and in NBM (gliospheres), dissociated gliosphere cells in NBM (BTSCs) and purified CD133 þ gliosphere cells in NBM (CD133 þ BTSCs) resulted in a significant increase in viable cell count (A, C) and proliferation (B, D) as measured by WST-1 and 3 H thymidine uptake assays, respectively. Interestingly, addition of CITCO resulted in a dose-dependent inhibition of viable cell count and proliferation in both T98G and U87MG glioma, and BTSCs.   effect on the morphology, cell viability or proliferation was observed in normal HAs following in vitro culture with CITCO (F), suggesting its selectivity to glioma and BTSCs.

The CITCO induces cell cycle arrest in BTSCs
To explore the intrinsic mechanisms by which CAR regulates growth arrest in BTSCs, we analysed the effect of CITCO on cell cycle progression. We found that BTSCs cultured in NBM with B27 and EGF þ bFGF showed distribution of cells in G0/G1, G2/M and S phases of cell cycle ( Figure 5). Addition of CITCO for 24 h resulted in a dose-dependent cell cycle arrest of BTSCs. Interestingly, data analyses revealed that CITCO induces cell cycle arrest through different mechanisms in different cell types. Although CITCO increased G0/G1 cells with a decrease in G2/M and S phases in T98G -BTSCs, it decreased G0/G1 and increased G2/M without affecting the S-phase cells in U87MG -BTSCs. However, BTSCs isolated from DB29 and DB33 showed comparable responses to CITCO with increased G0/G1 and G2/M with a decrease in S-phase cells ( Figure 5). These results suggest that CITCO induces cell cycle arrest differentially in different BTSCs in culture.

The CITCO induces apoptosis in BTSCs
To further determine the mechanisms by which CAR regulates BTSCs, we examined the effect of CITCO on apoptosis. We found that BTSCs cultured in NBM with B27 and EGF þ bFGF showed very low levels of Annexin V-positive apoptotic cells that increased dose dependently following addition of CITCO ( Figure 6). The BTSCs from T98G and U87MG cultured in the absence of CITCO Targeting BTSCs with CAR agonist CITCO S Chakraborty et al showed 6.8 and 11% Annexin V-positive cells that increased to 62 and 68% following addition of 10 mM CITCO, respectively. Moreover, BTSCs from DB29 and DB33 gliomas showed 3 and 0.5% Annexin V-positive cells that increased to 24 and 41% following treatment with 10 mM CITCO, respectively ( Figure 6). These results suggest that CITCO induces apoptosis in BTSCs in culture.

The CITCO inhibits BTSC xenograft in nude mice
To determine the in vivo effect of CITCO on tumour growth, we used a xenograft model in nude mice. As shown in Figure 7

DISCUSSION
Despite advances in modern medicine, the prognosis from current surgery, radiation and chemotherapy remains poor in brain tumour patients. The recent identification of BTSCs with resistance and recurrence properties has revolutionised the basic approaches on drug discovery and development for brain Nude mice were injected (s.c) with U87MG -BTSCs to induce tumour growth. The mice were treated (intraperitoneal) with 0, 25 or 100 mg CITCO in DMSO on days 22, 24, 26, 30 and 36. (A) The tumour growth was measured using digital Vernier Calipers, and the tumour volume was calculated. The percent inhibition by CITCO was calculated using the mean tumour volume in the group treated with 0 mM CITCO as 100%. The figure is representative of three independent experiments. (B) The mice were euthanised on day 50, and the BTSC xenografts were dissected; 6 mm thick sections were stained with H&E or anti-Ki67 Ab and photographed (100 Â/200 Â ) under fluorescent microscope. *Po0.05, **Po0.001, ***Po0.001.
tumour. Brain tumours often present a small fraction of BTSCs, making it difficult to isolate sufficient cells for drug discovery. EGF and FGF are growth factors that promote the tumourigenicity of glioma cells (Martens et al, 2008). We have shown recently that EGF þ bFGF induce the expansion of CD133 þ BTSCs as gliospheres in culture . In this study, we show for the first time that the CAR agonist CITCO induces growth arrest and apoptosis of BTSCs in culture and in animal model.
Nuclear hormone receptors have therapeutic values in many human diseases (Blumberg and Evans, 1998). Earlier studies have demonstrated deregulated expression of nuclear receptors in glioma and their activation with specific agonists inhibited tumour growth (Berge et al, 2001;See et al, 2004). In this study, we set out to identify novel nuclear receptors that could be used to target BTSCs for the treatment of glioma. We found that BTSCs express altered levels of many nuclear receptors compared with glioma, which are currently under investigation in our laboratory. In this manuscript, we focused on determining the use of CAR agonist CITCO in targeting BTSCs. We found that the expression of CAR was low in BTSCs derived from different gliomas. The downregulation of CAR could be a mechanism by which BTSCs evade antitumour responses. Interestingly, the upregulation of CAR expression by CITCO suggests its use in targeting BTSCs in the treatment of glioma. This is consistent with our earlier report on the induction of PPARg expression by its agonists in BTSCs . Although the constitutively active CAR is retained in the cytoplasm, CITCO induces its translocation into the nucleus and mediates gene transcription (Baes et al, 1994;Kobayashi et al, 2003). Although the exact mechanisms are not known, our findings suggest an autoregulation of CAR expression through CITCO/CAR axis in BTSCs.
Earlier studies have shown that CAR is highly expressed in the liver and small intestine, and promotes the detoxification and elimination of potentially toxic compounds by modulating the phase I and phase II drug-metabolising enzymes (Forman et al, 1998;Xu et al, 2005;Echchgadda et al, 2007;Veith et al, 2009). Although CAR activation can disrupt thyroid hormone homoeostasis , it also showed protective roles in stress response (Forman et al, 1998;Stedman et al, 2005;Xu et al, 2005;Echchgadda et al, 2007). The CAR-mediated expression of xenobiotic-metabolising enzymes is generally protective, but can be deleterious, if toxic metabolites are produced (Xu et al, 2005). The CAR agonists induce hepatocyte proliferation that depends on c-Myc-FoxM1 function (Blanco-Bose et al, 2008). The CAR agonists also inhibits Fas-induced hepatocyte apoptosis, liver injury, and fatalities by depleting the proapoptotic proteins Bak (Bcl-2 antagonistic killer) and Bax (Bcl-2-associated X protein) and increasing the expression of the antiapoptotic effector myeloid cell leukaemia factor-1 (Baskin-Bey et al, 2006). Thus, the xenobiotic properties of CAR and its agonists have been extensively studied, but its anticancer property was not known.
In this study, we have shown that CITCO inhibits the proliferation of glioma cells in a dose-dependent manner, without affecting primary astrocytes. Interestingly, CITCO inhibits the growth and expansion of BTSCs by inducing cell cycle arrest and apoptosis in culture. Moreover, the inhibition of CD133 expression by CITCO indicates the downregulation of BTSC expansion in culture. The inhibition of solid tumour growth by CITCO in xenograft model suggests the use of CITCO in the regulation of BTSCs in vivo. We have shown earlier that PPARg agonists inhibit cytokine-induced activation of Jak-Stat pathway in immune cells (Natarajan and Bright, 2002) and LIF-induced activation of Jak-Stat pathway in mouse embryonic stem cells (Rajasingh and Bright, 2006). We have also demonstrated earlier that the PPARg agonists induce growth arrest and apoptosis in BTSCs by blocking EGF/FGF-induced activation of Tyk2-Stat3 pathway in BTSCs . Although the precise molecular basis of CAR-mediated antineoplastic effect is under investigation, our findings suggest the use of CAR agonists as a new therapy to target BTSCs for the treatment of glioma patients.