ANTICANCER RESEARCH 28 : 2705-2714 (2008)

Anticancer Activity of Novel Extracts from Cautleya gracilis (Smith) Dandy: Apoptosis in Human Medullary Thyroid Carcinoma Cells ZENGXIA LI 1,2 , SONJA STURM 3, BERNHARD SVEJDA 1, HARALD HÖGER 4, ELISABETH SCHRAML 1, ELISABETH INGOLIC 5, VERONIKA SIEGL 1, HERMANN STUPPNER 3 and ROSWITHA PFRAGNER 1

1Department of Pathophysiology and Immunology, Center of Molecular Medicine, Medical University of Graz, Graz, Austria; 2Department of Biochemistry and Molecular Biology, Shanghai Medical School, Fudan University, Shanghai, P. R. China; 3Institute of Pharmacy, Center of Molecular Biosciences, Leopold Franzens University of Innsbruck, Innsbruck; 4Core Unit of Biomedical Research, Div ision of Laboratory Animal Science and Genetics, Medical University of Vienna, Himberg ; 5Research Institute for Electron Microscopy and Fine Structure Research, Technical University of Graz, Graz, Austria

Abstract. Background: Medullary thyroid carcinoma (MTC) gland. In 70-80% of cases, MTC occurs sporadically. The is a calcitonin-producing tumor of the thyroid arising from other 20-30% are inherited and may occur in three distinct the parafollicular C-cells. MTC is poorly responsive to clinical settings: as familial MTC without associated chemotherapy and radiotherapy, hence the only effective endocrinopathies, or combined with other endocrinopathies as therapy is surgery. Based on this fact, alternative strategies multiple endocrine neoplasia type 2A or 2B with autosomal have been sought. Materials and Methods: The effects of dominant inheritance (1-3). At the time of diagnosis, more Cautleya gracilis (Smith) Dandy were investigated for the than 25% of patients have distant metastases. Despite active first time in three human MTC cell lines and in MTC- study, neither chemotherapeutic agents nor irradiation have transplanted mice. Proliferation and viability were quantified proved to be markedly effective in MTC therapy. Apoptosis by cell counting, WST-1 tests, and ATP luminescent cell and proliferation are directly correlated with bcl-2 expression viability assays. Apoptosis was studied by DAPI staining, (4). Resistance to chemotherapy was attributed to the flow cytometry and luminescent assays for caspases 3/7, 8 expression of the multidrug resistance mdr1 gene (5). and 9. Results: A dose-dependent reduction of proliferation Well-characterized human cell lines derived from MTCs and an induction of apoptosis were found in all MTC cell have been very rare. For decades, only one continuous cell lines, while normal fibroblasts were not impaired. Similar line, TT, was available (6). The TT cell line was derived tumor inhibition was seen in heterotransplanted mice. from a hereditary MTC with an exon 11, codon 634 mutation Conclusion: Our in vitro and in vivo findings suggest a new (7). We have established eight continuous MTC cell lines potential clinical effect of Cautleya. (8–10), including MTC-SK (11) and SINJ (12) for deep investigation (13). MTC-SK and SINJ were derived from Medullary thyroid carcinoma (MTC) is a neuroendocrine sporadic forms of MTC. Each of our MTC cell lines showed tumor arising from the parafollicular C-cells of the thyroid an increased expression of the anti-apoptotic protein Bcl-2, allowing the tumor cells to survive. Recently, oriental medicinal herbs with anticancer activity have come into the spotlight as complementary or alternative medicines (14). Correspondence to: Prof essor Roswitha Pfragner, Department of Thus far, several plant-derived drugs have received Food and Pathophysiology and Immunology, Medical University of Graz, Drug Administration (FDA) approval for commercial Heinrichstrasse 31, A-8010 Graz, Austria. Tel: +43 3163804297, Fax: +43 3163809640, e-mail: [email protected] production. Examples of antineoplastic botanical drugs are camptothecin, vinblastine and taxol (14). Key Words: Cautleya gracilis , bioactive agents, medullary thyroid In a preceding screening using MTT cell proliferation tests carcinoma, chemoresistance, apoptosis. (Roche Diagnostics, Vienna, Austria) in HeLa cells, Cautleya

0250-7005/2008 $2.00+.40 2705 ANTICANCER RESEARCH 28 : 2705-2714 (2008) gracilis (Smith) Dandy was selected for the present study cells and transferred into 96-well plates; 10 μl WST-1 labeling (personal communication S Sturm, Institute of Pharmacy, mixture per well was then added per well. After incubation at 37˚C Innsbruck). C. gracilis , sometimes called "hardy ginger", is a for 2 hours, the samples were quantified spectrophoto-metrically by measuring the absorbance of the formazan product at 450 nm with an native of cool forests of the eastern Himalayas, belonging to ELISA plate reader. For TT and HF-SAR adherent cells, cells were the family of Zingiberaceae. The ethnobotanical activity of C. seeded directly into 96-well plates at a density of 1×10 5 cells/ml. gracilis is unclear and C. gracilis extracts have never been After adherence to plates, the cells were treated with extracts for 24, investigated in the treatment of MTC. Therefore, as a first 48 and 72 h and measured as above. Each sample was tested in 6 step to provide scientific evidence for anticancer activity, replicates; means and S.D. were calculated. different fractions of C. gracilis were investigated in three human MTC cell lines and in xenotransplanted MTC mice. ATP luminescent cell viability assay . This assay was performed using the Celltiter-Glo luminescent cell viability assay kit (Promega, The aim of the present study was to assess the in vitro and Madison, WI, USA Cat # G7570). One ml of cell was seeded into a in vivo activity of C. gracilis extracts in chemoresistant MTC. 24-well plate at a density of 2×10 5 cells/ml and incubated in completed Ham’s F12 medium with 10, 25, or 50 μg/ml extracts, or Materials and Methods 5 μl DMSO negative control for 4 h at 37˚C. After equilibrating to room temperature for 30 min, MTC-SK cells were pipetted carefully Plant material. The plant material (aerial parts) was provided by into single cells. Twenty-five μl of each sample were added to white- E. Stöger (Oberndorf, Austria). The dried aerial parts of Cautleya walled 96-well plates (Nunc, Roskilde, Denmark), then a 1:1 ratio of gracilis (Smith) Dandy, Zingiberaceae, were ground and successively Caspase-Glo3/7 reagent volume to sample volume was added into the extracted with petrol eum ether (PE) (CG-1), dichloromethane (DCM) same well. After mixing and incubating at room temperature for (CG-2), and ethyl acetate (EtOAc) (CG-3) in a Soxhlet apparatus. 10 min, all samples were measured in a plate-reading luminometer The solvent was evaporated under reduced pressure. The dry extracts (Mediators PhL, Mediators Diagnostika, Vienna, Austria) as directed were then redissolved in dimethyl sulphoxide (DMSO; Sigma, by the manufacturer. Each treatment was measured for 3 parallel Vienna, Austria) at a concentration of 10 mg/ml and stored at –20˚C. samples, and mean value and S.D. were calculated. The positive control camptothecin (CPT) (Sigma) was also dissolved in DMSO and stored at 4˚C in 1 mM stock solution. DAPI (4’,6’-diamidino-2 phenylindole) staining. TT cells were seeded on FlexiPerm slides (Kendro, Vienna, Austria) overnight for Cell lines and cell culture. The human MTC cell lines MTC-SK adherence and then treated with CG-1 at a concentration of (11) and SINJ (12), and the normal human skin fibroblast cell line 25 μg/ml for 4, 12 and 24 h. The slides with TT cells were washed 2+ 2+ HF-SAR, were established in our laboratory. The human MTC cell with Dulbecco’s phosphate-buffered saline lacking Ca and Mg line TT (6) was purchased from the European Collection of Cell (PBSA) and stained with 0.5 μg/ml DAPI (Sigma) in methanol for Cultures (ECACC, Porton Down, UK; Cat. No. 92050721). MTC- 30 min at room temperature. Cells were viewed using an inverted SK and SINJ cells were maintained in Ham’s F-12 medium phase-contrast fluorescence microscope with ultraviolet (UV) (Biowhittaker, Verviers, Belgium) containing 10% fetal bovine excitation at 300-500 nm (Nikon Eclipse TE300, Tokyo, Japan). serum (PAA Laboratories, Exon, PA, USA) at 37˚C, 5% CO 2. TT Cells with nuclei containing clearly condensed chromatin or cells and HF-SAR cells were cultured in 1:1 Ham’s F-12: M-199 with fragmented nuclei were scored as apoptotic. medium, supplemented with 10% fetal bovine serum, at 37˚C, 5% CO 2. All cell lines were Mycoplasma -free. FACS analysis. To confirm apoptosis, active caspase-3 staining was performed using the FITC-conjugated Monoclonal Active Caspase-3 Cell counting. MTC-cells were seeded at a density of 2×10 5 cells/ ml Antibody Apoptosis Kit (BD Bioscience, Vienna, Austria; Cat into 24-well plates (Sarstedt, Wr. Neudorf, Austria) and incubated for #550480) according to the manufacturer’s instructions: Approximately 24, 48 or 72 h in completed Ham’s F12 medium with DMSO 1×10 5 cells were treated with different concentrations of CG-1 for (control) or supplemented with 10, 25, or 50 μg/ ml of 24 h, washed in cold PBSA, fixed and permeabilized using 500 μl C. gracilis extracts. The cell clusters were pipetted into single cells Cytofix/Cytoperm™ solution for 20 min on ice. Cells were washed and counted automatically (Casy-1 Cell Counter and Analyzer, and resuspended in 100 μl Perm/Wash™ buffer containing 20 μl PE- Schärfe System, Reutlingen, Germany). Each sample was measured 3 conjugated monoclonal rabbit anti-active caspase-3 antibody at room times; the cell counter calculated the mean value and standard temperature for 30 min. After incubation, cells were washed in deviation (S.D.). Perm/Wash™ buffer and analysed with FACScan (BD, Heidelberg, Germany). WST-1 cell proliferation assay. Cell proliferation and viability were quantified using reagent WST-1 Cell Proliferation Reagent (4-[3-(4- TUNEL assay. Cells were detected using the In Situ Cell Death iodophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1, 3-benzene disul- Detection Kit, Fluorescein (Roche, Lewes, UK; Cat phonate) (Roche Diagnostics, Vienna, Austria) according to the #11684795910) according to the manufacturer’s instructions. manufacturer's protocol. This method is based on the ability of viable Briefly, cells were treated with CG-1, CG-2 or CG-3 for 24 h, cells to metabolize tetrazolium salt WST-1 to formazan by washed with PBSA and harvested. The cells were fixed with mitochondrial dehydrogenases. MTC-SK and SINJ cell suspension freshly prepared 4% paraformaldehyde for 60 min at room were seeded at a density of 2×10 5 cells/ ml in 24-well plates temperature and placed on ice for 2 min with 0.1% Triton ® X-100 (Sarstedt). After 24, 48 or 72 h treatment with different concentrations solution (Serva, Heidelberg, Germany). Intracellular DNA of solvent extracts or DMSO, cells were pipetted carefully into single fragments were then labeled by exposing the cells to TUNEL

2706 Li et al : Anticancer Activity of Cautleya gracilis reaction mixture for 1 h at 37˚C, in a humidified atmosphere and injected intratumorally with a) 10 μg/100 μl CG-2 dissolved in protected from light. After washing with PBSA twice, cells were DMSO/ PBSA per tumor for 5 mice , b) DMSO/ PBSA (control 1) transferred to a slide and analyzed under a fluorescence microscope for 5 mice, and c) PBSA only (control 2) for 4 mice. The (Nikon Eclipse TE300, Tokyo, Japan). observation period was 21 days.

Caspase 3/7 activity assay. Caspase 3/7 activities were measured Results with a homogenous luminescent assay kit (Promega, Madison, WI, USA; Cat # G8091) according to the manufacturer’s instructions. Antiproliferative activity of C. gracilis extracts in human One ml MTC-SK cells was seeded into 24-well plate s at a density MTC cell line MTC-SK. First, to ascertain anticancer activity, of 2×10 5 and incubated in completed Ham’s F12 medium with 10, all three extracts of C. gracilis , CG-1, CG-2 and CG-3 , were 25, or 50 μg/ml substances; 5 mM CPT positive control, or 5μl evaluated for their antiprolifer ative activities in the human DMSO negative control for 4 h at 37˚C. After equilibrating to room medullary carcinoma cell line MTC-SK by cell counting. At temperature for 30 min, MTC-SK cells were carefully pipetted into a density of 25 μg/ml, all these extracts prominently single cells; 25 μl of the total volume of each sample were added into white-walled 96-well plates (Nunc), then a 1:1 ratio of Caspase- suppressed proliferation in MTC-SK cells. As shown in Glo 3/7 reagent volume to sample volume was added into the same Figure 1A, after 72 h treatment, cells added with DMSO 5 well. After gently mixing the contents of wells on a plate shaker for solvent reached a density of 5.0×10 cells/ml from the 30 s, they were incubated at room temperature for 1 h and finally starting concentration of 2.35×10 5 cells/ml; however, those measured for luminescence of each sample in a plate-reading cells with C. gracilis extracts only reached a density of less luminometer (Mediators PhL) as directed by the manufacturer. Each than 2.5×10 5 cells/ml with the same starting density. treatment was measured in 3 parallel samples, and the mean value To further test the antiproliferative effect, we used the and S.D. were calculated. WST-1 cell proliferation reagent. Within 48 h of exposure, all Time curve of caspase-8, caspase-9, and caspase-3/7 activity. the extracts had the potency to reduce cell viability in a dose- Caspase-8, caspase-9 and caspase-3/7 activities were measured dependent manner. At a concentration of 50 μg/ml, more than using Promega luminescent assay kits (Cat # G8201, Cat # G8211, 90% of cells lost their viability within 48 h treatment. The and Cat # G8091, respectively). Briefly, MTC-SK cells were seeded IC 50 values of CG-1(PE extract), CG-2 (DCM extract) and 5 at a density of 2 ×10 /ml and 25 μg/ml of CG-1, CG-2, CG-3 or CG-3 (EtOAc extract) were approximately 20, 18 and 15 DMSO solvent were added for 0, 1, 4, 8, 12, and 24 h. Each sample μg/ml, respectively (Figure 1B). was pipetted carefully into single cells, then transferred into three Moreover, we used an even more sensitive method, the 96-well white-walled plates (Nunc) and measured for its caspase-8, caspase-9 and caspase-3/7 activity according to the manufacturer’s ATP luminescent cell viability assay based on the amount of instructions. Each treatment was measured in 3 parallel samples, ATP, to quantify the cell viability changes within a short-term and the mean value and S.D. were calculated. treatment. As shown in Figure 1C, even with 4 h exposure, In addition, the effect of the specific caspase-3 inhibitor Ac- high concentrations of C. gracilis extracts could also reduce DEVD-CHO (Promega, Madison, Wi, USA, cat # G5961) on MTC- MTC-SK cell viability in a dose-dependent manner. At a SK cells treated with 10 μg/ml CG-2 was determined. Thereafter, concentration of 50 μg/ml, the percentage of remaining cell MTC-SK cells were seeded into 24-well plates at a density of 2 ×10 5 viability under treatment with CG-1, CG-2, and CG-3 for 4 cells/ml. One group had been co-treated with specific caspase-3 inhibitor Ac-DEVD-CHO fifteen minutes before CG-2 or h was 29.8%, 50.4% and 42.8%, respectively. camptothecin were added, whereas the other group did not receive the specific inhibitor. Caspase-3/7 activity was measured at 4, 7, 9, Effects of C. gracilis on multicellular aggregation and cell 14, 16, 22 and 24 h and the activity of Ac-DEVD-CHO at 4, 9, 14 morphology of MTC-SK cells . As reported previously by our and 22 h using the Promega luminescent assay kit according to the group (20), MTC-SK cells exhibited an ability to survive and manufacturer’s instructions, as described above. proliferate in the aggregate form. It is believed that the ability of cell populations to survive in the aggregate form is Transmission electron microscopy. MTC-SK and HF-SAR cells a potential index for cell tumorigenicity, which might also were fixed in 3% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2, on ice (Plano, Wetzlar, Germany), postfixed in 1% osmium tetroxide be a reason for drug and radiation resistance in cancer (5-7). (SIGMA) in 0.1 M cacodylate buffer and processed by modified Thus, we aimed to study the effect of C. gracilis extracts on routine methods. Photographs were taken with a FEI Technai 12 intact spheroids of MTC-SK medullary carcinoma cells. equipped with a Gatan CCD Camera Bioscan . Multicellular aggregates were formed by the MTC-SK cells and cultured in 10% FBS for 24 h. At this time, the intact Treatment of MTC xenografts in SCID mice with CG-2 . Female spheroids were treated with and without C. gracilis extracts. SCID mice about 6 weeks old (Department of Laboratory Animal After a 24-h treatment, the spheroids were observed under Science and Genetics, Medical University of Vienna, Himberg, light microscopy and photographed. Figure 2A (a) shows Austria) were injected subcutaneously with two specimens (~2 mm 3) of MTC-SK tumor obtained from a previous tumor induction MTC-SK cells grown as multicellular tumor spheroids from MTC-SK cells (3×10 7 cells/0.3 ml per mouse). After 40 days, without the addition of extracts, whereas Figure 2A (b-j) the tumor diameters had reached approximately 11 mm. Mice were shows the effect of different concentrations of C. gracilis

2707 ANTICANCER RESEARCH 28 : 2705-2714 (2008)

extracts on the previously intact spheroids. Cellular adhesion in the treated cells was significantly disrupted.

Extracts of C. gracilis induced apotosis in MTC-SK cells. To investigate whether the extract-induced inhibition of cell proliferation was associated with cell apoptosis, TUNEL assay ( terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling; Roche, Lewes, UK, cat # 11684795910) was performed to detect apoptotic morphologic changes. MTC-SK cells were incubated with 25 μg/ml CG-1, CG-2, CG-3 or DMSO control for 24 h and stained with fluorescein-12-UTP. As shown in Figure 3, CG-1-, CG-2-, and CG-3-treated MTC-SK cells showed green fluorescence, indicating the fragmented DNA in apoptotic cells, whereas the DMSO control was negative for the staining. To further confirm the induction of apoptosis, FACS analysis with active caspase-3 antibody staining was performed in CG-1-treated MTC-SK cells as shown in Figure 4A. With 24 h treatment, CG-1 increased the proportion of activated forms of caspase-3-positive cells from 7% in control cells to 43.67%, 87.94%, and 90.77% in 10, 25, and 50 μg/ml CG-1-treated cells, respectively.

C. gracilis extracts induced activation of caspases. To confirm further the apoptosis and determine whether caspase pathways were involved in this process, the activities of activated forms of caspase-3/7, caspase-8, and caspase-9 were detected using an in vitro luminescence-based assay. Firstly, the activity of caspase-3/7, which are the final effector caspases in apoptosis, was quantified at different concentrations of CG-1, CG-2 and CG-3 treated MTC-SK cells. As shown in Figure 4B, these three extracts all increased active caspase-3 activity in vitro in a dose-dependent manner. After 4 h exposure at a concentration of 50 μg/ml, the caspase-3 activity of CG-1-, CG-2- and CG-3-treated cells was 6.8-, 8.5- and 7.6-fold more than in control cells. In addition, the caspase-3 activity of MTC-SK cells and the cell viability were found to be inversely correlated (compare Figure 4A to 1C), suggesting that C. gracilis inhibited the growth of MTC-SK cells, at least in part, by inducing apoptosis. The specific caspase-3/7 inhibitor Ac- DEVD-CHO confirmed the results by blocking the caspase- 3/7 activity totally (Figure 5). Besides caspase-3/7 effector caspases, the activation of initiator caspases, caspase-8 and caspase-9 was also Antiproliferative activity of C. gracilis extracts in MTC-SK Figure 1. measured. We observed that in 25 μg/ml C. gracilis -treated cells. A, C. gracilis extracts suppressed MTC-SK cells growth. Cells were incubated in a 24-well microplate with 25 μg/ml of each solvent cells, caspase-8, -9, and -3/7 activities significantly increased extract for 24, 48, and 72 h, and each sample was analyzed in the compared to control cells. Caspase-8, -9, and -3/7 peaked CASY-1 Cell Counter & Analyzer. B, C. gracilis extracts reduced MTC- after about 8 h of treatment (Figure 6 A, B, C). Data of 5 SK cells viability in a dose-dependent way. Cells (2.0×10 ) were treated normal cells are not shown. with different concentrations of each solvent extract for 48 h; cell viability was determined using WST-1 assay. C, ATP luminescent cell viability assay. After treatment with different concentrations of each Different effects of C. gracilis on human MTC cells compared extract for 4 h, cells were measured for viability by ATP luminescent to normal fibroblasts. From previous results, it was deduced cell viability assay. All data are expressed as mean±S.D. that the C. gracilis extracts had strong anticancer effects on

2708 Li et al : Anticancer Activity of Cautleya gracilis

Figure 2. MTC-SK cells (2.0×10 5) were used to form multicellular aggregates. They were pipetted into single cells and treated with CG-1, CG-2, CG-3 or DMSO control for 24 h. C. gracilis extracts dissociated the aggregates in a dose-dependent way, forming a single-cell suspension. Nikon Inverted Microscope Eclipse TE 300 (×400).

MTC-SK cells, though it was unclear whether C. gracilis acted against other cancer cell lines as well. To study generalized applicability, we tested two other medullary carcinoma cell lines, SINJ and TT, compared to the normal human fibroblast cell line HF-SAR for their sensitivity to CG after 48 h treatment (Figure 7 A, B, C). Using the concentration of C. gracilis required to kill 50% of the cells (IC 50 ) as a measure of sensitivity, we detected variable sensitivity among cell lines, with MTC-SK being most sensitive (IC 50 =18 to 23 μg/ml), and SINJ and TT having an intermediate sensitivity (IC 50 =22 to 40 μg/ml and 22 to 40 μg/ml respectively). In contrast, the fibroblast cell line HF- SAR was more resistant (IC 50 >50 μg/ml). The IC 50 difference of CG on MTC cells compared to fibroblasts suggested that C. gracilis displayed special anticancer activity in MTC-SK, SINJ and TT cells, therefore suggesting it to be a potential chemotherapeutic agent for the control of MTC. Figure 3. C. gracilis extracts induced apoptosis in MTC-SK cells. MTC- Electron microscopy. Ultrastructural changes in the tumor cells SK cells (2.0×10 6) were treated with 25 μg/ml CG-1, CG-2, CG-3 or were observed after 24 h treatment with C. gracilis extract s. DMSO control for 24 h and stained with a TdT enzyme buffer The MTC cells showed condensation of chromatin at the containing fluorescein-12- dUTP and analyzed under a fluorescence periphery of the nuclear membrane. The plasma membrane microscope (TUNEL assay). formed many blebs and protrusions (Figure 8A). The control cells, normal human skin fibroblasts, showed no ultrastructural modification under treatment with C. gracilis extracts. The structures of the cell membrane, nucleus and organelles were Similar to the in vitro results, a tendency toward tumor normal, without any apoptotic changes (Figure 8B). inhibition was found in all CG-2-treated mice. No effects were measured in the two control groups: the growth of the Treatment of MTC xenografts in SCID mice with CG-2. untreated tumors was not inhibited, as measured by tumor Tumor growth was followed in each experimental group. diameter (data not shown).

2709 ANTICANCER RESEARCH 28 : 2705-2714 (2008)

Figure 5. Time curve of caspase-3/7 actvity. MTC-SK cells were treated with 10 μg CG-2/ ml for up to 24 h. CG-2 as the most effective extract was analyzed for caspase-3/7 activity. Controls were DMSO (the solvent), the caspase-3/7 specific inhibitor Ac-DEVD-CHO, camptothecin ( CPT) and camptothecin with Ac-DEVD-CHO. CG-2 displayed a different maximum compared to camptothecin. The addition of Ac-DEVD-CHO to CG-2 and to camptothecin blocked caspase-3/7 activity.

50 μg/ml, those extracts had almost no suppressive effect. Almost all conventional anticancer drugs are primarily effective against rapidly dividing cells. The strong proliferative effect and the IC 50 difference between MTC cell lines and the slowly dividing fibroblasts suggested that C. gracilis could be a candidate for effective chemotherapeutic agents, especially in MTC. Figure 4. C. gracilis extracts induced caspase-3 activation. A, FACS Chemotherapeutic drugs are considered to kill tumor cells analysis of active caspase-3 by treatment of CG-1. Cells were exposed to CG-1 (10, 25, 50 μg/ ml) for 24 h and harvested. After fixation and by activating a cascade of events resulting in apoptosis. In permeabilization, active-caspase-3 was stained by fluorescence and agreement with this line of thought, we provided evidence analyzed by FACS. A dose-dependent activation of caspase-3 was noted. that C. gracilis extracts were able to induce apoptosis with B, Caspase-3/7 activity in vitro by luminescence-based assay. The dose- the classic feature of apoptosis including initiator and effector dependent effects of CG1, CG-2 and CG-3 were evaluated and caspase activation and internucleosomal DNA fragmentation. compared with camptothecin (CPT) . Bcl-2 overexpression did not prevent apoptosis. In addition, further analyses revealed that the decrease of cell viability with the treatment was associated with the induction of apoptosis in MTC-SK cells (compare Figure 1C with Figure Discussion 4A, r=0.9, 0.97 and 0.98, respectively), suggesting that cell death could be a main reason for the decrease in cell viability. In this study, we present a first report on the anticancer Moreover, we tried to determine the caspase activation activity of C. gracilis extracts on human medullary pathway in C. gracilis -treated MTC-SK cells. At present, the carcinoma cells. All three extracts from C. gracilis had a receptor-binding and mitochondrial-mediated apoptosis strong inhibitory effect on tumor cell growth, disrupting the pathways are two major modes of caspase activation that have tumor spheriods, and induced tumor cell apoptosis. come to light. In this study, we found that both the initiator CG-1, CG-2 and CG-3 inhibited MTC-SK, SINJ and TT caspase-8, which is mainly involved in the external pathway, and medullary thyroid cells growth in a dose-dependent manner. the internal pathway-related initiator caspase-9 were activated. The growth suppressive function of C. gracilis on MTC cells Caspase-3 and caspase-9 activation implied the involvement of was very strong; even with a 4 h treatment, cell viability was the mitochondrial pathway in C. gracilis - induced apoptosis. largely reduced (Figure 1C). The IC 50 in the medullary cell Caspase-8 activation was induced by extrinsic receptor binding lines MTC-SK, SINJ and TT was between 18-40 μg/ml. In and followed by direct activation of caspase-3 or crosstalk with contrast, in normal fibroblasts, at a concentration lower than the mitochondrially mediated pathway by cleavage of Bid. Our

2710 Li et al : Anticancer Activity of Cautleya gracilis

Figure 6. In MTC-SK cells treated with 25 μg/ ml C. gracilis, caspase-8, Figure 7. After treatment with three C. gracilis extracts, two other medullary -9 and -3/7 activities significantly increased compared to control cells thyroid carcinoma cell lines, SINJ (A) and TT (B), showed a dose-dependent (data of normal cells not shown). Caspase-8, -9 and -3/7 peaked after loss of viability, comparable to the effects in the MTC-SK cells. The normal about 8 h of treatment. control cells, HF-SAR, showed normal viability values (C). data suggest the possible involvement of an extrinsic pathway The C. gracilis extracts also disrupted the intact for the induction of apoptosis by C. gracilis . However, more multicellular tumor spheroids in MTC-SK and SINJ cells. work is required to delineate the relative contribution of this The ability of tumor cell populations to survive in the pathway to apoptosis execution. aggregated form is a potential index for tumorigenicity,

2711 ANTICANCER RESEARCH 28 : 2705-2714 (2008)

aggregates. In our experience, only about 1-5% of MTC-cell lines grow adherently in vitro , while the majority form densely packed spheroidal aggregates. This phenomenon might also account for their chemoresistance. Multicellular resistance reflects a relative intrinsic phenotype and a disruption should actually contribute to growth inhibition. Identical observations were described in previous studies on other bioactive agents, including extracts of Stemona tuberosa Lour (18-20). The in vivo experiment with MTC-xenografted mice resulted in growth retardation of the tumors in the CG-2- treated group. In contrast, the control groups with untreated xenografts showed unimpaired tumor progression. As this pilot experiment involved only a small number of mice, statistical analyses are in progress with a larger number of animals. Compared with the antitumor effects of other bioactive agents, derived from Stemona tuberosa Lour and other plants (18-20), the effects of C. gracilis on proliferation, morphology and apoptotic rates of MTC were clearly stronger: even a very low dose of C. gracilis (10 μg/ ml) was effective in MTC cells, while Stemona effects became manifest from concentrations of 25 μg/ ml upwards. Summarizing, the novel bioactive compounds of C. gracilis seem to be good candidates for further evaluation as effective therapeutic agents acting through reduction of proliferation and induction of apoptosis.

Acknowledgements

This investigation was supported by the Austrian Cancer Aid/ Styria (EF 01/2004). We wish to thank Ms. Eugenia Lamont for critically reviewing the manuscript.

References

1 Vitale G, Caraglia M, Ciccarelli A, Lupoli G, Abbruzzese A, Taggliaferri P and Lupoli G: Current approaches and perspectives in the therapy of medullary thyroid carcinoma. Cancer 91 : 1797-1808, 2001. 2 Eng C and Mulligan LM: Mutations of the RET proto-oncogene in the multiple endocrine neoplasia type 2 syndromes, related sporadic tumors, and Hirschsprung disease. Hum Mutat 9: 97- Figure 8. Effects on ultrastructure. A, Apoptotic changes in an MTC-SK 109, 1993. cell after treatment with C. gracilis extract CG-2 (10 μg/ ml) for 24 h. 3 Mulligan LM, Kwok JB, Healey CS, Elsdon MJ, Eng C, Gardner The nuclear chromatin is condensed at the periphery of the nuclear E, Love DR, Mole SE, Moore JK, Papi L, Ponder MA, Telenius ¡ membrane. The plasma membrane is blebbing ( arrows). TEM image. H, Tunnacliffe A and Ponder BAJ: Germline mutations of the B, Normal human skin fibroblast, HF-SAR, passage 12, treated with C. RET proto-oncogene in multiple endocrine neoplasia syndrome gracilis extract CG-2 (10 μg/ ml) for 24 h. The normal cell shows type 2A. Nature (London) 363 : 458-460, 1993. normal organelle ultrastructure, including an intact plasma membrane, 4 Basolo F, Pollina L, Fontani G, Fiore L, Pacini F and Baldanzi a normally shaped oval nucleus, intact mitochondria and Golgi A: Apoptosis and proliferation in thyroid carcinoma: correlation apparatus. TEM image. with bcl-2 and p53 protein expression. Br J Cancer 75 : 537-541, 1997. 5 Yang KP, Liang YF and Samaan NA: Intrinsic drug resistance in which might be a reason for drug and radiation resistance a human medullary thyroid carcinoma cell line: association with in cancer (15-17). Both MTC-SK and SINJ cell lines exhibit overexpression of mdr1 gene and low proliferation fraction. the phenomenon of growth in vitro as multicellular Anticancer Res 11 : 1065-1068, 1991.

2712 Li et al : Anticancer Activity of Cautleya gracilis

6 Leong SS, Horoszewicz JS, Shimaoka K, Friedmann M, 14 Raskin I, Ribnicky DM, Komarnytsky S, Ilic N, Poulev A, Kawinski E, Song MJ, Chu T, Baylin S and Mirand EA: A new Borisjuk N, Brinker A, Moreno DA, Ripoli C, Yakoby N, O’Neal cell line for study of human medullary thyroid carcinoma. In : M, Cornwell T, Pastor I and Fridlender B: Plants and human Advances in Thyroid Neoplasia. Andreoli M, Monaco H and health in the twenty-first century. Trends Biotechnol 20 : 522- Robbins J (eds.). Field Ed ucational Italia, Rome, pp. 95-108, 531, 2002. 1981. 15 Olive PL and Durand RE: Drug and radiation resistance in 7 Carlomagno F, Salvatore D, Santoro M, de Franciscis V, Quadro spheroids: cell contact and kinetics. Cancer Metastasis Rev 13 : L, Panariello L, Colantuoni V and Fusco A: Point mutation of 121-138, 1994. the RET proto-oncogene in the TT human medullary thyroid 16 Steuer AF, Rhim JS, Hentosh PM and Ting RC: Survival of carcinoma cell line. Biochem Biophys Res Commun 207 : 1022- human cells in the aggregate form: potential index of in vitro cell 1028, 1995. transformation J Natl Cancer Inst 58 : 917-21, 1977. 8 Pfragner R, Behmel A, Ingolic E and Wirnsberger GH: Culture 17 St Croix B and Kerbel RS: Cell adhesion and drug resistance in of human neuroendocrine tumor cells. In : Culture of Specialized cancer. Curr Opin Cell Biol 9: 549-556, 1997. Cells: Culture of Human Tumor Cells. Pfragner R and Freshney 18 Rinner B, Siegl V, Pürstner P, Efferth T, Brem B, Greger H and RI (eds.). Wiley-Liss & Sons, Hoboken, N J, pp. 373-403, 2004. Pfragner R: Activity of novel plant extracts against medullary 9 Pfragner R, Skofitsch G, Hoeger H, Jech M, Rinner B, Siegl V, thyroid carcinoma. Anticancer Res 24 : 495-500, 2004. Niederle B, Gnant M, Friedl J and Stift A: Medullary thyroid 19 Rinner B, Li ZX, Sturm S, Stuppner H, Siegl V and Pfragner carcinoma: autologous tumor cell lines for dendritic cell R: Use of cell lines to define new bioassays for the therapy of vaccination. Anticancer Res 25 : 4225-4230, 2005. medullary thyroid carcinoma. Ann Endocrinol 76 : P29, Supp 10 Pfragner R, Wirnsberger G, Ingolic E and Niederle B: Medullary 4, 2006. thyroid carcinomas in cell culture – models for future therapies. 20 Li ZX, Sturm S, Stuppner H, Schraml E, Aguiriano Moser V, Wien Klin Wochenschr 114 : 279-283, 2002. Siegl V and Pfragner R: The dichloromethane fraction of 11 Pfragner R, Hoefler H, Behmel A, Ingolic E and Walser V: Stemona tuberosa Lour inhibits tumor cell growth and induces Establishment and characterization of continuous cell line MTC- apoptosis of human medullary thyroid carcinoma cells. SK derived from a human medullary thyroid carcinoma. Cancer Biologics: Targets Therapy 1(4) : 455-463, 2007. Res 50 : 4160-4166, 1990. 12 Pfragner R, Wirnsberger G, Behmel A, Wolf G, Passath GA, Ingolic E and Adamiker D: New continuous cell line from human medullary thyroid carcinoma: SINJ. Phenotypic analysis and in vivo carcinogenesis. Int J Oncol 2: 831-836, 1993. 13 Stadler G, Wieser M, Steindl F, Grillari J, Katinger H, Pfragner R and Voglauer R: Development of standardized cell culture Received April 4, 2008 conditions for tumor cells with potential clinical application. Revised June 25, 2008 Cytotherapy 9: 488-498, 2007. Accepte d July 2, 2008

2713