In vitro Cytotoxicity and P-Glycoprotein Modulating ...

Michael Adams1, Anne Mahringer2, Rudolf Bauer1, Gert Fricker2, Thomas Efferth3

Abstract

Abbreviations BBB: blood-brain barrier CNS: central nervous system p-gp: P-glycoprotein PBCEC: porcine brain capillary cells

We recently reported the antimycobacterial activity of geranylated furocoumarins from the fruits of Tetradium daniellii (Benn.) T.G. Hartley (Rutaceae) [1]. To study their cytotoxicity and effects on the cellular efflux pump p-glycoprotein (p-gp), the four coumarins most abundant in T. daniellii fruits [1] were tested against CCRF-CEM leukemia cells and their p-gp over-expressing subline CEM/ADR5000. P-gp, located in the blood-brain barrier (BBB), the physical barrier between and the central nervous system (CNS), alongside other efflux pumps, prevents many xenobiotics from accumulating in the CNS. Over-expression of p-gp in a number of tumors contributes to multidrug resistance [2], [3], [4], [5]. Furthermore, an in vitro microplate-scale assay monitoring the p-gp dependent accumulation of a fluorescent dye in porcine brain capillary cells (PBCECs) was used. Affiliation: 1 Institute of Pharmaceutical Sciences, Department of Pharmacognosy, University of Graz, Graz, Austria ´ 2 Institute of Pharmacy and Molecular Biotechnology, University of Heidelberg, Heidelberg, Germany ´ 3 German Cancer Research Centre, Heidelberg, Germany. Correspondence: Prof. Dr. Thomas Efferth ´ German Cancer Research Centre ´ Im Neuenheimer Feld 280 ´ 69120 Heidelberg, Germany ´ Phone: +49-6221423-426 ´ Fax: +49-6221-423-433 ´ E-mail: [email protected] Received April 13, 2007 ´ Revised September 19, 2007 ´ Accepted September 21, 2007 Bibliography: Planta Med 2007; 73: 1475±1478 Georg Thieme Verlag KG Stuttgart ´ New York ´ DOI 10.1055/s-2007-990261 ´ Published online November 8, 2007 ´ ISSN 0032-0943

Letter

Four antimycobacterial geranylated furocoumarins, from the fruits of Tetradium daniellii (Rutaceae), were tested in a bioassay using CCRF-CEM leukemia cells and their P-glycoprotein (p-gp) over-expressing subline CEM/ADR5000, to asses their cytotoxicity and effects on cellular efflux pumps. All showed considerable cell proliferation inhibition with IC50 values ranging from 1.72 to 11.02 mM against CCRF-CEM and 2.09 to 13.56 mM against CEM/ ADR5000, respectively. An assay monitoring the p-gp-dependent accumulation of a calcein fluorescent dye in porcine brain capillary endothelial cells was used to study interactions of the test substances with this efflux pump. Here, they all were shown to be only weak to moderate modulators of p-gp.

Compounds 1 ± 4 (Fig. 1) showed considerable inhibition of cell proliferation with IC50 values ranging from 1.72 to 11.02 mM in CCRF-CEM and 2.09 to 13.56 mM in CEM/ADR5000, respectively (Table 1). Compound 2 displayed the most potent activity with IC50 values of 1.72 (CCRF-CEM) and 2.09 mM (CEM/ADR5000), respectively. Interestingly, compound 1 was less toxic with IC50 values of 3.9 and 4.8 mg/mL, respectively. The difference between the two is the position of the peroxidized geranyl side chain, on C-5 (2) and C-8 (1). None of the furocoumarins 1 ± 4 showed strong cross-resistance to the multi-resistant cell line CEM/ ADR5000. Xanthotoxin and xanthotoxol, both lacking the geranyl side chain, were inactive at the highest tested concentration of 10 mg/mL in both cell lines which shows that the geranyl moiety increases cytotoxicity. Geraniol, the free geranyl alcohol, showed no activity whatsoever. In the PBCEC assay, compound 1 showed little influence on cellular fluorescence, with the strongest increase of 195 % at 25 mM. This C-8 substituted epoxide was a significantly weaker p-gp inhibitor than its C-5 substituted analogue 2. The activity of 2, however, peaked at 5 mM with a sharp decline to a value below the one of the control at 50 mM, whereas 1, a weak modulator of p-gp, remained between 100 and 200 % up to the highest test concentration. Presumably the decrease in activity of 2 is due its higher cytotoxicity also observed in the cell proliferation assay (Fig. 2). With the two furocoumarins with a non-oxidized geranyl sidechain on C-5 (4) and C-8 (3), activity peaked at 10 mM at approximately 350 and 250 % (Fig. 2). They, together with 2, can be classified as moderate modulator of p-gp. Kawaii et al. [6] examined structure-activity relationships of antiproliferative coumarins. We recently reported the effects of the bicyclic coumarins scopoletin and isoscopoletin from Artemisia argyi (Juss.) Benth. (Asteraceae) in the anti-cell proliferation assay [7]. Bergamottin (4) together with dihydroxybergamottin is a main component of grapefruits Citrus x paradisi Macfad. (Rutaceae) and is responsible for interactions with many drugs due to inhibition of some of the enzyme, particularly [8]. These results

Table 1

IC50 (mM) values and 95% fiducial limits (95 % FD) of geranylated furocoumarins 1 ± 4 from Tetradium daniellii towards CCRF-CEM leukemia cells and multidrug-resistant CEM/ ADR5000 cells IC50 Values (mM)

Compounds

CCRF-CEM

DCR CEM/ADR5000

Mean

95 % FD

Mean

95 % FD

1

11.02

6.78 ± 27.94

13.56

8.33 ± 30.25

1.23

2

1.72

1.16 ± 2.43

2.09

1.16 ± 5.25

1.22

3

3.93

2.63 ± 6.23

5.21

4.70 ± 5.77

1.33

4

5.18

3.99 ± 7.16

4.05

3.58 ± 4.59

0.78 833

Standard cytostatic drugs Doxorubicin

12 nM

10 mM

Vincristine

2 nM

1 mM

550

Paclitaxel

4 nM

890 nM

222

Tests were performed at 10, 3, 1, 0.3, 0.1, 0.03 and 0.01 g/mL in triplicate. For comparison, the IC50 values of the standard cytostatic drugs doxorubicin, vincristine, and paclitaxel are given. (DCR = degree of cross-resistance).

1475

Downloaded by: Universität Heidelberg. Copyrighted material.

In vitro Cytotoxicity and P-Glycoprotein Modulating Effects of Geranylated Furocoumarins from Tetradium daniellii

Fig. 1 Structures of substances 1 ± 4 from Tetradium daniellii, xanthotoxin, xanthotoxol and geraniol.

show that interactions of 4 with p-gp are of little significance for the reported drug interactions.

Materials and Methods General experimental: Test substances 8-geranylpsoralen-6¢¢,7¢¢-epoxide (1), bergamottin-6¢¢,7¢¢-epoxide (2), 8-geranylpsoralen (3), bergamottin (4) were isolated as previously described [1]. The purity of all isolated substances was determined by integration of proton resonances and by HPLC-DAD peak integration and was shown to be 93.6 % for 1, 97.8% for 2, 99.8% for 3 and 94.3 % for 4, respectively. Xanthotoxin was obtained from Fluka (Buchs, Switzerland), xanthotoxol from Roth (Karlsruhe, Germany). Geraniol was obtained from Dragoco (Vienna, Austria). The commercial substances showed no visible impurities in their HPLC chromatograms. Principle of the cell growth inhibition assay: Cells were incubated with test substance for seven days, after which there were counted and the cell number was compared to that of a vehicle-treated (DMSO) control. The results represent the net outcome of cell death and cell proliferation.

Letter ¼ Planta Med 2007; 73: 1475 ± 1478

The in vitro response of CCRF-CEM, and CEM/ADR5000 cells was evaluated as previously described by Efferth et al. [9]. CEM/ADR 5000 cells are a p-gp over-expressing multidrug-resistant subline of CCRF/CEM, resistant to drugs such as doxorubicin, vincristine, and paclitaxel. Briefly, aliquots of 5 104 cells/mL were seeded in 24-well plates. Test compounds were added immediately. The number of cells in four 10 mL aliquots was counted visually seven days after treatment, using a cell counting chamber (improved Neubauer chamber, Brand GmbH + Co. KG.; Wertheim, Germany), under a light microscope. The standard cytostatic drugs doxorubicin (Pharmacia and Upjohn; Nerviano, Italy), vincristine (Bristol-Myers; Neu-Isenburg, Germany), and paclitaxel (Sanofi-Aventis; Berlin, Germany) were used as controls. The results are expressed as % vehicle-treated (DMSO) control cell number and represent the net outcome of cell proliferation and cell death. The compounds were tested at least three times with different cell preparations at dilutions 10, 3, 1, 0.3, 0.1, 0.03 and 0.01 mg/mL, against both CCRF/CEM and CEM/ADR5000 cells. Mean values and the standard deviations were determined (Microsoft, Excel). IC50 values with fiducial limits were calculated by probit-log analysis using SPSS 6.0 for MS Windows as described for a quantitative biological assay [10]. The degree of cross-resistance of a substance can be calculated by dividing the IC50 of


Letter 1476

Fig. 2 Influence of 1 ± 4 on cellular fluorescence due to interaction with the p-glycoprotein efflux pump. Tests were performed in triplicate with porcine brain capillary endothelial cells (PBCECs) at 50, 25, 10, 5, 1, and 0.1 mM, and expressed as % of DMSO treated cells. Verapamil was used as a positive control (mean SEM, n = 3, * p £ 0.01, ** p £ 0.001, *** p £ 0.0001 compared to control).

CEM/ADR5000 by the IC50 of CCRF-CEM. Because p-gp limits the accumulation of xenobiotics in the brain, resistance in CEM/ ADR5000 to substances more toxic to CCRF/CEM cells can serve as a model for blood-brain barrier (BBB) activity of substances [9]. Principle of the PBCEC assay: Calcein, a fluorescent dye, is formed by intracellular hydrolysis of the non-fluorescent calcein-acetoxymethylester (calcein-AM), a p-gp substrate. Test substances, which interact with p-gp, either result in an increased uptake of calcein-AM and therefore in an increased intracellular concentration of calcein, by inhibiting p-gp activity, or the substances induce p-gp efflux activity, thereby reducing fluorescence. Isolation of porcine brain capillary endothelial cells (PBCECs): PBCECs were isolated as previously described [11]: Briefly, cortical grey matter from fresh porcine brains was minced and digested enzymatically using 0.5 % dispase (Roche; Mannheim, Germany). Cerebral microvessels were obtained after centrifugation in 15 % dextran (Sigma-Aldrich; Taukkirchen, Germany) and were

subsequently incubated in buffer containing 1 mg/mL collagenase/dispase (Roche). The resulting cell suspension was supplemented with 10 % horse serum (Biochrom; Berlin, Germany) and filtered through 150-mm nylon mesh, and PBCECs were separated on a discontinuous Percoll (Sigma) gradient consisting of Percoll 1.03 g/mL (20 mL) and 1.07 g/mL (15 mL). Isolated PBCECs were filtered through 80-mm nylon mesh before being seeded, at a density of 250,000 cells/cm2, onto collagen (Roche)-coated 96-well cell culture plates. Cells were cultivated at standard conditions in Medium 199 containing 100 mg/mL streptomycin, 100 mg/mL penicillin G, 10 mmol/L HEPES, 10 % heat-inactivated horse serum and 0.8 mmol/L l-glutamine (all from Biochrom). After reaching confluency, the medium was changed after 6 days in culture and cells were grown in 45 % MEM, 45 % F12HAM, 100 mg/mL streptomycin, 100 mg/mL penicillin G, 10 mmol/ L HEPES and 2 mmol/L glutamine (all from Biochrom). On the following day, medium was removed from each well and cells were washed twice with a Krebs±Ringer buffer (37 8C) and were subsequently used for the transport experiments.

Letter ¼ Planta Med 2007; 73: 1475 ± 1478


Letter 1477

Acknowledgements M. Adams was supported on a research grant form the Heinrich Jörg Stiftung (Graz, Austria) to do these studies.

1478

References 1

Adams M, Ettl S, Kunert O, Wube AA, Haslinger E, Bucar F et al. Antimycobacterial activity of geranylated furocoumarins from Tetradium daniellii. Planta Med 2006; 72: 1132 ± 5. 2 Hendrikse NH, Bart J, de Vries EGE, Groen HJM, van der Graaf WT, Vaalburg W. P-glycoprotein at the blood-brain barrier and analysis of drug transport with positron-emission tomography. J Clin Pharmacol 2001; 41: 48 ± 54. 3 Demeule M, Shedid D, Beaulieu E, Del Maestro RF, Moghrabi M, Ghosn PB et al. Expression of multidrug-resistance p-glycoprotein (MDR1) in human brain tumors. Int J Cancer 2001; 93: 62 ± 6. 4 Ambudkar SV, Dey S, Hrycyna CA, Ramachandra M, Pastan I, Gottesman MM. Biochemical, cellular and pharmacological aspects of the multidrug transporter. Ann Rev Pharmacol Toxicol 1999; 39: 361 ± 98. 5 Schinkel AH. P-glycoprotein, a gatekeeper in the blood-brain barrier. Adv Drug Deliv Rev 1999; 36: 179 ± 94. 6 Kawaii S, Tomono Y, Ogawa K, Sugiura M, Yano M, Yoshizawa Y. The antiproliferative effect of coumarins on several cancer cell lines. Anticancer Res 2001; 21: 917 ± 23. 7 Adams M, Efferth T, Bauer R. Scopoletin and isoscopletin: the anticancer principle in Artemisia argyi, express growth inhibitory effect towards CCRF-CEM leukemia cells and multidrug resistant CEM/ ADR5000 cells. Planta Med 2006; 72: 862 ± 4. 8 Bailey DG, Malcolm J, Arnold O, Spence JD. Grapefruit juice-drug interactions. Br J Clin Pharmacol 1998; 46: 101 ± 10. 9 Efferth T, Davey M, Olbrich A, Ruecker G, Gebhart E, Davey R. Activity of drugs from traditional Chinese medicine toward sensitive and MDR1- or MRP1-overexpressing multidrug-resistant human CCRFCEM leukemia cells. Blood Cells Mol Dis 2002; 28: 160 ± 8. 10 Finney JD. Statistical methods in biological assay. London: Charles Griffin; 1978: 58 ± 65. 11 Fellner S, Bauer B, Miller DS, Schaffrik M, Fankhanel M, Spruss T et al. Transport of paclitaxel (Taxol) across the blood-brain barrier in vitro and in vivo. J Clin Invest 2002; 110: 1309 ± 18. Letter ¼ Planta Med 2007; 73: 1475 ± 1478


Letter

The assay with PBCECs was performed as described by Fellner et al [11]. Briefly, PBCECs were isolated and cultured in 96-well plates, with calcein-acetoxymethyl ester (calcein-AM) in the absence and presence of substrates and test substances. Fluorescent calcein which is a substrate to p-gp is formed by intracellular hydrolysis of the non-fluorescent calcein-AM. Substances interacting with p-gp therefore increase the intracellular concentration of the dye, compared to vehicle-treated cells. After washing and lysing the cells, the extent of intracellularly accumulating fluorescence was monitored with a fluorescence plate reader (Fluoroskan Askent; Labsystems; Helsinki, Finnland) in a concentration-dependent manner. The tests were performed at least three times at the concentrations 0.1, 1, 5, 10, 25 and 50 mM. Data are given as mean values with their standard deviations. Verapamil (Knoll AG; Ludwigshafen, Germany) was used as a positive control. Control and treatment groups were compared by either Student's t-test or one-way analysis of variance, followed by a Dunnett's post-hoc test. Differences were considered significant at ** p < 0.01. The p-gp modulating effect of a substance in this test design can be distinguished as strong p-gp modifiers (effect > 500 % at < 10 mM), moderate (effect of 250 ± 500 % at 10 ± 50 mM), and weak modulators (effect of 125 ± 250 % at 10 ± 50 mM). Substances with an effect < 125 % at 10 ± 250 mM were classified as non-modulators [11].