- Email: [email protected]
4′-O-Alkyaloenin derivatives and their sulfates directed toward overcoming multidrug resistance in tumor cells
Cancer Letters 218 (2005) 15–20 www.elsevier.com/locate/canlet
4 0-O-Alkyaloenin derivatives and their sulfates directed toward overcoming multidrug resistance in tumor cells Guang-zhu Jin, Hong-Ji Quan, Jyunichi Koyanagi, Kazuhiro Takeuchi, Yoshihiko Miura, Fusao Komada, Setsuo Saito* Faculty of Pharmaceutical Sciences, Josai University, Keyakidai 1-1, Sakado, Saitama 350-0295, Japan Received 21 May 2004; received in revised form 15 July 2004; accepted 21 July 2004
Abstract The cytotoxic effects on HCT 116, Hep G2 and HCT 116/VCR 100-1-1 cell lines of synthetic 4 0 -O-alkylaloenins (2–17), 4 0 -O-benzylaloenin (18) and 4 0 -O-allylaloenin (19) were examined by MTT assay, and compared with that of aloenin (1) isolated from Aloe arborescens Mill. Var. natalensis Berger which showed no marked effect (IC50 value: O100 mM). The cytotoxic effects of 4 0 -O-alkylaloenin sulfates (21–29) were also examined on the same cell lines. The introduction of a longer alkyl group at the O-4 0 position of 1 resulted in a higher cytotoxic action on HCT 116 and Hep G2 cells. Among 4 0 -Oalkylaloenins 2–17, 4 0 -O-tetradecylaloenin 14 was the most cytotoxic to both on HCT 116 cells (IC50 value: 5.3G2.3 mM) and Hep G2 cells (IC50 value: 4.0G0.6 mM). Also among 4 0 -O-alkylaloenin sulfates 21–29, 4 0 -O-dodecylaloenin sulfate 29 was the most cytotoxic to both on HCT 116 (IC50 value: 4.8G0.2 mM) and Hep G2 cells (IC50 value: 4.0G0.5 mM). 4 0 -O-Alkylaloenins 7–14 and 4 0 -O-alkylaloenin sulfates 24–29 were also cytotoxic to Hep G2 and HCT 116/VCR 100-1-1 cell lines, which overexpress P-glycoprotein, as well as HCT 116 cell lines which scarcely express it. q 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Cytotoxicity; 4 0 -O-Alkylaloenin; 4 0 -O-Alkylaloenin sulfate; HCT 116; Hep G2; P-glycoprotein
1. Introduction The overexpression of the P-glycoprotein (P-gp) which causes multidrug resistance (MDR 1) in tumor cells has been a significant obstacle for the successful chemotherapy of many cancers [1–3]. Recently to overcome these obstacles, many modulators have
* Corresponding author. Tel.: C81-49-271-7667; fax: C81-49271-7984. E-mail address: [email protected] (S. Saito).
been extensively studied [4,5]. These modulators included calcium channel blockers, calmodulin antagonists, steroidal agents, protein kinase C inhibitors, immunosuppressive drugs and antibiotics as well as surfactants [5]. In a structure–activity relationship study, we previously reported that the anti-HIV activity of various alkyl glycyrrhetinate diglycoside sulfates markedly increased (about 100-fold) compared with that of glycyrrhizin [6]. The enhancement of the activity was probably due to the micelle forming ability of the alkyl glycyrrhetinate diglycoside sulfates.
0304-3835/$ - see front matter q 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2004.07.025
16
G.-z. Jin et al. / Cancer Letters 218 (2005) 15–20
In the present study, we report the synthesis of the 4 0 -O-alkylated aloenin derivatives of aloenin (1) and their sulfate derivatives, which have enhanced the interfacial properties, and the cytotoxicity of the derivatives on human colorectal cancer (HCT 116), HCT 116/VCR 100-1-1 (vincristine-induced resistant variant of HCT 116 cells) and human hepatoma cancer (Hep G2) cell lines using MTT assay [7]. Furthermore, the sensitivity of the derivatives to P-gp is discussed.
2. Materials and methods 2.1. Chemicals Alkylated aloenins (Fig. 1) were obtained from aloenin 1 isolated from the leaves of Aloe arborescens Mill. Var. natalensis Berger (Kidachirokai or Kidachiaroe in Japanese) (Liliaceae) [8,9] as follows: 4 0 -O-methyl aloenin (2) was obtained by methylation of 1 with diazomethane [10], and other 4 0 -Oalkylaloenins were obtained by the following general method. A mixture of 1, alkyl bromides or alkyl iodides (2.0–10.0 eq.) and potassium carbonate (2.0–4.0 eq.) was refluxed in dry acetone. The mixture was evaporated, then poured into water and extracted with ethyl acetate, and further evaporated. The resulting residue was purified by column chromatography (a gradient of 0–20% MeOH in CHCl3) to give the desired 4 0 -O-alkylaloenins (3–17 and 20), 4 0 -Obenzylaloenin (18) and 4 0 -O-allylaloenin (19). 4 0 -OAlkylaloenin sulfates (21–29) (Table 2) were obtained by the reaction of 4 0 -O-alkylaloenins 2, 7, 11, 13–17 and 20 with SO3-pyridine complex in pyridine according to the method previously reported [6]. Test compounds were dissolved in dimethyl sulfoxide (DMSO) and diluted with culture medium to the desired concentrations. MTT (3-(4,5-dimethylthiazol2-yl)-2,5-di-phenyltetrazolium bromide) was obtained from Chemicon International, Inc. (CA, USA).
116/VCR 100-1-1 cell lines were a kind gift from Daiichi Pharmaceutical Co. Lit, Japan. Dulbecco’s modified Eagle’s medium (DMEM), McCoy’s 5A medium, fetal bovine serum (FBS) and penicillin– streptomycin mixture (100 U/ml penicillin and 100 mg/ml streptomycin) were purchased from Wako Pure Chemical Industries Ltd (Osaka, Japan), Sigma (MO, USA), Biosource International (CA, USA) and Bio Whittaker (ND, USA), respectively. The HCT 116, HCT 116/VCR 100-1-1 and Hep G2 cells were maintained in McCoy’s 5A medium and in DMEM, respectively. Each medium was supplemented with 10% FBS and a penicillin–streptomycin mixture at 37 8C in a humidified atmosphere containing 5% CO2. 2.3. Cytotoxicity test Aliquots (200 ml) of 5!103 cells/ml of HCT 116, HCT 116/VCR 100-1-1 and Hep G2 cells were seeded in 96 well flat-bottomed plates (Microteste Tissue Culture Plate, 96 Well, Flat Bottom with Low Evaporation Ltd, Falcon, NJ, USA), and were incubated in a medium containing 10% FBS and a penicillin–streptomycin mixture at 37 8C in a humidified atmosphere of 5% CO2 for 24 h. The test drugs were dissolved in dimethyl sulfoxide (DMSO).
2.2. Cells and cell culture HCT 116 (ATCC No. CCL-247) and Hep G2 (No. RCB0459) cell lines were purchased from Dainippon Pharmaceutical Co. Ltd (Osaka, Japan) and RIKEN Cell Bank (Tsukuba, Japan), respectively. HCT
Fig. 1. Structures of aloenin 1 and 4 0 -O-alkylaloenins 2–19.
G.-z. Jin et al. / Cancer Letters 218 (2005) 15–20
The incubation medium was replaced with each test medium giving a final concentration of 1–500 mmol/l of compounds 1–19 and 21–29 and no drug in 2 ml DMSO over 2 days. The cell survival was determined by the method reported by Mosmann using MTT assay [7]. Each experiment was performed in duplicate wells, and all experiments involving control (DMSO only) and drug treatments were performed separately 3–5 times. Data represent meanGSD values. 2.4. Statistical analysis Regression lines were calculated using StatisticaTM release software, version 5.5 (StatSoft Japan Inc., Tokyo, Japan). Statistical significance was defined as P!0.01.
3. Results and discussion The cytotoxic effects of aloenin 1, 4 0 -O-alkylaloenins (2–17), 4 0 -O-benzylaloenin (18) and 4 0 -Oallylaloenin (19) were tested by MTT assay [7] using HCT 116 and Hep G2 cell lines. The IC50 values were calculated based on the percentage inhibition of cell growth and these are listed in Table 1. Aloenin 1 showed no marked cytotoxicity (IC50Z293.1G 32.7 mM; this value was represented as O100 mM in Table 1) to HCT 116 cells. Hayashi et al. [11] suggested that the appropriate hydrophobicity of flavanone compounds was important for cytotoxicity to KB, KB-VIN and related cell lines. This suggestion let us to introduce various alkyl groups into the aloenin molecule. Aloenin 1 has two kinds of alcohols in the molecule: a phenolic alcohol at the C-4 0 position and aliphatic alcohols on a glucopyranose moiety. In this study, alkyl groups were introduced to the phenolic alcohol in order to enhance the hydrophobicity and to confer the appropriate interfacial property to aloenin 1 to obtain various 4 0 -O-alkylated aloenins (2–17). The cytotoxic effects of these alkylated compounds were tested on two cell lines, HCT 116 and Hep G2, and the mean IC50 values for these cells are listed in Table 1. In the case of HCT 116 cells, the IC50 values of 4 0 O-methyl- (2) and 4 0 -O-ethylaloenin (3) were higher than 100 mM as well as aloenin 1 itself, indicating that
17
they showed no marked cytotoxicity. The cytotoxicity gradually increased on increasing the length of the alkyl chains from propyl to hexyl (IC50 values: 79.5G 38.1, 74.1G2.2, 37.9G12.4, and 17.9G5.5 mM for propyl- 4, butyl- 5, pentyl- 6, and hexylaloenin 7, respectively), and the 4 0 -O-alkylaloenins 8–14 with longer alkyl groups from heptyl to tetradecyl showed IC50 values of 5.3G2.3–10.4G2.5 mM, exhibiting potent cytotoxicity to HCT 116 cells. Among compounds 8–14, compound 14 exhibited the most potent cytotoxicity (IC50 value: 5.3G2.3 mM). However, the IC50 value (16.2G4.9 mM) of 4 0 -O-hexadecylaloenin (15), having a longer alkyl chain by two CH2 units than 14, was slightly greater than that of 8–14, showing a slight reduction in activity. Furthermore, 4 0 -O-octadecyl- (16) and 4 0 -O-eicosylaloenin (17), having a longer alkyl chain than 15, had an IC50 value greater than 100 mM, and exhibited no marked cytotoxicity. In these experiments, while all solutions of samples 1–14 and the cells in the medium were clear, the solution of 15 at a concentration of 150 mM was not completely clear and solutions of 16 and 17 at Table 1 Cytotoxic effects of aloenin 1 and 4 0 -O-alkylaloenins 2–17 on HCT 116 and Hep G2 cells Compound
IC50 (mM)a HCT 116
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
b
O100 O100b O100b 79.5G38.1 74.1G2.2 37.9G12.4 17.9G5.5 8.1G6.8 10.4G2.5 5.7G0.5 8.9G2.1 8.7G2.1 6.6G0.7 5.3G2.3 16.2G4.9 O100b O100b
Hep G2 O100b O100b O100b O100b O100b O100b 36.2G7.1* 28.5G9.2* 23.4G2.2* 14.7G1.5* 13.5G2.3 20.1G3.7* 12.1G3.1* 4.0G0.6 16.0G0.9 O100b O100b
*P!0.01, the significant difference between the activity on HCT 116 cells and that on Hep G2 cells. a IC50 values (meanGSD) are the concentrations at which 50% of the cells are inhibited from growing. b IC50 values more than 100 mM were indicated as O100.
18
G.-z. Jin et al. / Cancer Letters 218 (2005) 15–20
Fig. 2. Structures of 4 0 -O-alkylaloenin sulfates 21–29.
the same concentration were fairly turbid. These lower solubilities may be the cause of the reduction in the IC50 values of compounds 15–17. In the case of Hep G2 cells, while alkylaloenins 2–6 having shorter alkyl groups than a pentyl had IC50 values higher than 100 mM, exhibiting no marked cytotoxicity, compounds 7–14 had IC50 values from 36.2G7.1 to 4.0G0.6 mM. Among alkylaloenins 7–14, tetradecylaloenin 14 showed the most potent cytotoxicity having an IC50 value of 4.0G0.6 mM. Also, the IC50 value (16.0G0.9 mM) of compound 15 was higher than that of 14 (4.0G0.6 mM) and those of compounds 16 and 17 were higher than 100 mM. The increases may be due to the same reason mentioned above. Thus, the cytotoxic effects on HCT 116 and Hep G2 cells were enhanced by the introduction of alkyl groups having an appropriate length at the 4 0 -O position of 1. Tetradecylaloenin 14 showed the most potent cytotoxic effect on HCT 116 cells (IC50 value: 5.3G2.3) and Hep G2 cells (IC50 value: 4.0G 0.6 mM). Here the relationships between the cytotoxic effects on HCT 116 cells (shown by the IC50 values) and the hydrophobicity (shown by C log P values [12]) of 4 0 -O-alkylaloenin derivatives 4–14 were determined and are shown in Fig. 3. The relationships showed positive correlations between the cytotoxic effects of 4 0 -O-alkylaloenin derivatives 4–14 in HCT 116 cells and the hydrophobicities caused by the introduction of alkyl groups at the O-4 0 position of aloenin 1. These relationships were also observed for 4 0 -O-benzylaloenin 18 and 4 0 -O-allylaloenin 19; the more hydrophobic 18 (C log PZ1.81) had a more potent cytotoxic effect (IC50 value: 32.9G4.9 mM), on the other hand the less hydrophobic 19 (C log PZ 0.82) had a weaker effect (IC50 value: 50.2G3.1 mM) on the growth of HCT 116 cells (Fig. 3).
With respect to the relationship between cytotoxic activity and hydrophobicity of compounds, Iwase et al. [13] reported that the extending the alkyl chain at the 8 or 13 position of berberine affected the cytotoxic activity, and the highest level of activity was observed in 8- or 13-hexyl-substituted derivatives of berberin. This agreed with the results of the activities of 4 0 -O-alkylaloenin derivaties 4–14; the introducing of an appropriate hydrophobicity as well as an appropriate size of the substituent into aloenin molecule may enhance the cytotoxic activity, although the exact mechanism for the affinity to the enzyme or intracellular accumulation and so on has not been solved. No relationship was found for hexadecyl- 15, octadecyl- 16 and eicosylaloenin 17 because of the lower solubility of these compounds. In order to enhance the solubility of 15, 16 and 17 in the test medium, the hydroxyl groups on the glucopyranose moiety were sulfated [6] as well as those of 2, 7, 11, 13, 14 and 20. All the synthetic 4 0 -O-alkylaloenin sulfates 21–29 are listed in Fig. 2. The cytotoxic effects of these sulfates on HCT 116 cells and Hep G2 cells are listed in Table 2. In this case, while alkylaloenin sulfates 21–23 having shorter alkyl groups than decyl showed no marked activity, Table 2 Cytotoxic effects of 4 0 -O-alkylaloenins sulfates 21–29 on HCT 116, Hep G2 and HCT 116/VCR 100-1-1 cells Compound
21 22 23 24 25 26 27 28 29
IC50 (mM)a HCT 116
Hep G2
HCT 116/ VCR 100-1-1
O100b O100b O100b 27.1G6.6 47.1G19.6 22.2G4.5 23.2G2.2 12.3G4.0 4.8G0.2
O100b O100b 99.4G9.1 21.4G4.6 24.0G2.0c 18.1G1.6 14.8G5.1c 9.1G0.9 4.5G0.5
O100b O100b O100b 31.4G4.8 27.6G3.1d 21.3G1.3 18.9G1.7 14.9G1.9 19.0G2.6d
a IC50 values (meanGSD) are the concentrations at which 50% of the cells are inhibited from growing. b IC50 values more than 100 mM were indicated as O100. c P!0.01, the significant differences between the activity on HCT 116 cells and that on Hep G2 cells, respectively. d P!0.01, the significant differences between the activity on HCT 116 cells and that on HCT 116/VCR 100-1-1 cells, respectively.
G.-z. Jin et al. / Cancer Letters 218 (2005) 15–20
Fig. 3. Relationships between C log P and the cytotoxic effects of 4 0 -O-alkylaloenins on HCT 116 and Hep G2 cells. Closed circle, HCT 116 cells; open circle, Hep G2 cells. Regression lines for HCT 116 and Hep G2 cells were yZ81.4 eK0.469x (R2Z0.733, P!0.001) and yZ147.5 eK0.489x (R2Z0.827, P!0.001), respectively. Each value shows the meanGSD.
alkylaloenin sulfates 24–29 having longer alkyl groups than docosyl exhibited greater cytotoxicity, and 4 0 -O-dococylaloenin sulfate 29 having the longest alkyl group of 21–29 had the most potent cytotoxic effect on HCT 116 and Hep G2 cells. These results suggest that the cytotoxicity of aloenin 1 is enhanced by having an appropriate length alkyl group thereby giving an appropriate interfacial property to 1. The cytotoxic effects of alkylaloenins 4–14 on HCT 116 cells were more potent than on Hep G2 cells (Table 1 and Fig. 3). This is thought to be due to the difference in physiological properties between HCT 116 and Hep G2 cells. It is known that while HCT 116 cell lines scarcely express the MDR 1 (P-glycoprotein; P-gp) [14], Hep G2 cell lines overexpress it [15]. P-gp acts as an efflux pump to remove several antitumor agents, Ca2C antagonists, cyclosporine, digoxin and other compounds from cells [5,16]. In spite of the overexpression of P-gp in Hep G2 cells, compounds 7–14 had potent cytotoxic effects on the cells. It has been reported that HCT 116/VCR 100-1-1 cells, a mutant of HCT 116 cells, overexpress P-gp, and the IC50 value of a cytotoxic compound such as taxane to the cells is higher than 1000 times by the efflux system such as P-gp of the cells compared with that to HCT 116 cells [14]. Here the compounds 7–14 which were potent active to HCT 116 cells were further measured using HCT 116/VCR 100-1-1 cells and compared with the activities on HCT 116 (Fig. 4).
19
From the figure, while the cytotoxic effects of 7–14 on HCT were from 1.8 to 4.3 times higher than those on HCT 116/VCR 100-1-1, the alkylaloenins 7–14 still had a potent effect on HCT 116/VCR 100-1-1 cell lines which overexpressed P-gp, as was the case with Hep G2. The cytotoxic effects of alkylaloenin sulfates 24–29 on HCT 116, Hep G2 and HCT 116/VCR 100-1-1 cells were also investigated. These sulfates were cytotoxic to Hep G2 and HCT 116/VCR 100-1-1 cells which overexpressed P-gp as well as on HCT 116 which scarcely expressed it (Table 2). This indicates that these sulfates may be lower sensitive substrates for P-gp transport in comparison with the alkylaloenins. In conclusion, aloenin 1 has no marked cytotoxic effect on HCT 116 and Hep G2 cancer cell lines. However, this effect was enhanced by the introduction of an appropriate length alkyl group at the O-4 0 position of 1. Among 4 0 -O-alkylaloenin derivatives 2–17 which were synthesized from 1, 7–15 were markedly cytotoxic not only to HCT 116 cells which scarcely express P-gp but also Hep G2 and HCT 116/VCR 100-1-1 cells which overexpress P-gp. Among 7–15, 4 0 -O-tetradecylaloenin 14 had the most potent cytotoxic effect on HCT 116, HCT 116/VCR 100-1-1 and Hep G2 cell lines. Comparison of the activities of 4 0 -O-alkylaloenin sulfates 21–29 showed that 24–29, having longer alkyl chains, were more potent than sulfates 21–23 with shorter alkyl chains as far as the effects on HCT 116, HCT 116/VCR 100-1-1 and Hep G2 cells were concerned. These results suggest that an appropriate balance
Fig. 4. Cytotoxic effects of 4 0 -O-alkylaloenins 7–14 on the growth of HCT 116 and HCT 116/VCR100-1-1 cancer cell lines. Open bar, HCT 116 cells; closed bar, HCT 116/VCR100-1-1. *P!0.01.
20
G.-z. Jin et al. / Cancer Letters 218 (2005) 15–20
between the hydrophobicity of the aglycon moiety and the hydrophilicity of the sugar moiety in aloenin derivatives is important for exhibiting a potent cytotoxic effect on these cancer cells. Each 4 0 -O-alkylaloenin sulfate of 24–29 was more cytotoxic to P-gp overexpressed Hep G2 cells than the corresponding one to HCT 116 cells (Table 2). Furthermore, 24–29 were markedly cytotoxic to P-gp overexpressed HCT 116/VCR 100-1-1 cells. These results also suggest that these sulfates 24–29 are not influenced by efflux systems such as P-gp as well as alkylaloenins 7–14. Thus the finding that these derivatives 7–14 and 24–29 were markedly cytotoxic to P-gp overexpressed cancer cells without any modulator suggests a new approach to overcome the obstacle posed by the multidrug resistance in cancer. We are continuing our investigations of cytotoxicity using other types of P-gp-overexpressing cells.
References [1] M.M. Gottensman, I. Pasten, Biochemistry of multidrug resistance mediated by the multidrug transporter, Annu. Rev. Biochem. 62 (1993) 385–427. [2] O. Fardel, V. Lecureur, A. Guillonzo, The P-glycoprotein multidrug transporter, Gen. Pharmacol. 27 (1996) 1283–1291. [3] M.M. Gottensman, T. Fojo, S.E. Bates, Multidrug resistance in cancer: role of ATP-dependent transporters, Nat. Rev. Cancer 2 (2002) 48–58. [4] R. Krishna, L.D. Mayer, Multidrug resistance (MDR) in cancer. Mechanisms, reversal using modulators of MDR and the role of MDR modulators in influencing the pharmacokinetics of anticancer drugs, Eur. J. Pharm. Sci. 11 (2000) 265–283.
[5] D.R. Ferry, H. Traunecker, D.J. Kerr, Clinical trial of Pglycoprotein reversal in solid tumours, Eur. J. Cancer 32A (1996) 1070–1081. [6] S. Saito, T. Furumoto, M. Ociai, A. Hosono, H. Hoshino, U. Haraguchi, et al., Synthetic studies on the relationship between anti-HIV activities and micelle forming abilities of various alkylated glycyrrhetinate diglycoside sodium sulfates and related compounds, Eur. J. Med. Chem. 31 (1996) 365–381. [7] T. Mosmann, Rapid colorimetric assay for cellular growth and survival; application to proliferation and cytotoxicity assays, J. Immunol. Methods 139 (1983) 55–63. [8] J.E. Hay, L.J. Haynes, The Aloins. Part I. The structure of barbaloin, J. Chem. Soc. 1956; 3141–3147. [9] N. Okamura, N. Hine, S. Harada, T. Fujioka, K. Mihashi, A. Yagi, Three chromone components from Aloe Vera leaves, Phytochemistry 43 (1996) 495–498. [10] Th. J. De Boer, H.J. Backer, Diazomethane, Org. Synth. IV (1963) 250–253. [11] K. Hayashi, Y. Nakanishi, K.F. Bastow, G. Cragg, H. Nozaki, K.-H. Lee, Anti tumor agents 221. Buceracidins A and B, two new flavanones from Bucida buceras, J. Nat. Prod. 66 (2003) 125–127. [12] C Log P values of derivatives in this study were calculated by ChemDraw Ultra 8.0 (CambridgeSoft, Cambridge Massachusetts, USA). [13] K. Iwase, M. Moriyasu, T. Yamori, T. Turuo, D.-U. Lee, W. Wiegrebe, In vitro Cytotoxicity of the Protoberberine-type Alkoloids. J. Nat. Prod. 64 (2001) 896–898. [14] M. Shionoya, T. Jimbo, M. Kitagawa, T. Soga, A. Tohogo, DJ927, a novel oral taxane, overcomes P-glycoprotein-mediated multidrug resistance in vitro and in vivo, Cancer Sci. 94 (2003) 459–466. [15] G. Lee, M. Piqette-Miller, Influence of IL-6 on MDR and MRP-mediated multidrug resistance in human hepatoma cells, Can. J. Physiol. Pharmacol. 79 (2001) 876–884. [16] K. Takara, T. Sakaeda, Y. Tanigawara, K. Nishiguchi, N. Ohmoto, M. Horinouchi, et al., Effects of 12 Ca(2C) antagonists on multidrug resistance, MDR1-mediated transport and MDR1 mRNA expression, Eur. J. Pharm. Sci. 16 (2002) 159–165.
4 0-O-Alkyaloenin derivatives and their sulfates directed toward overcoming multidrug resistance in tumor cells Guang-zhu Jin, Hong-Ji Quan, Jyunichi Koyanagi, Kazuhiro Takeuchi, Yoshihiko Miura, Fusao Komada, Setsuo Saito* Faculty of Pharmaceutical Sciences, Josai University, Keyakidai 1-1, Sakado, Saitama 350-0295, Japan Received 21 May 2004; received in revised form 15 July 2004; accepted 21 July 2004
Abstract The cytotoxic effects on HCT 116, Hep G2 and HCT 116/VCR 100-1-1 cell lines of synthetic 4 0 -O-alkylaloenins (2–17), 4 0 -O-benzylaloenin (18) and 4 0 -O-allylaloenin (19) were examined by MTT assay, and compared with that of aloenin (1) isolated from Aloe arborescens Mill. Var. natalensis Berger which showed no marked effect (IC50 value: O100 mM). The cytotoxic effects of 4 0 -O-alkylaloenin sulfates (21–29) were also examined on the same cell lines. The introduction of a longer alkyl group at the O-4 0 position of 1 resulted in a higher cytotoxic action on HCT 116 and Hep G2 cells. Among 4 0 -Oalkylaloenins 2–17, 4 0 -O-tetradecylaloenin 14 was the most cytotoxic to both on HCT 116 cells (IC50 value: 5.3G2.3 mM) and Hep G2 cells (IC50 value: 4.0G0.6 mM). Also among 4 0 -O-alkylaloenin sulfates 21–29, 4 0 -O-dodecylaloenin sulfate 29 was the most cytotoxic to both on HCT 116 (IC50 value: 4.8G0.2 mM) and Hep G2 cells (IC50 value: 4.0G0.5 mM). 4 0 -O-Alkylaloenins 7–14 and 4 0 -O-alkylaloenin sulfates 24–29 were also cytotoxic to Hep G2 and HCT 116/VCR 100-1-1 cell lines, which overexpress P-glycoprotein, as well as HCT 116 cell lines which scarcely express it. q 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Cytotoxicity; 4 0 -O-Alkylaloenin; 4 0 -O-Alkylaloenin sulfate; HCT 116; Hep G2; P-glycoprotein
1. Introduction The overexpression of the P-glycoprotein (P-gp) which causes multidrug resistance (MDR 1) in tumor cells has been a significant obstacle for the successful chemotherapy of many cancers [1–3]. Recently to overcome these obstacles, many modulators have
* Corresponding author. Tel.: C81-49-271-7667; fax: C81-49271-7984. E-mail address: [email protected] (S. Saito).
been extensively studied [4,5]. These modulators included calcium channel blockers, calmodulin antagonists, steroidal agents, protein kinase C inhibitors, immunosuppressive drugs and antibiotics as well as surfactants [5]. In a structure–activity relationship study, we previously reported that the anti-HIV activity of various alkyl glycyrrhetinate diglycoside sulfates markedly increased (about 100-fold) compared with that of glycyrrhizin [6]. The enhancement of the activity was probably due to the micelle forming ability of the alkyl glycyrrhetinate diglycoside sulfates.
0304-3835/$ - see front matter q 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2004.07.025
16
G.-z. Jin et al. / Cancer Letters 218 (2005) 15–20
In the present study, we report the synthesis of the 4 0 -O-alkylated aloenin derivatives of aloenin (1) and their sulfate derivatives, which have enhanced the interfacial properties, and the cytotoxicity of the derivatives on human colorectal cancer (HCT 116), HCT 116/VCR 100-1-1 (vincristine-induced resistant variant of HCT 116 cells) and human hepatoma cancer (Hep G2) cell lines using MTT assay [7]. Furthermore, the sensitivity of the derivatives to P-gp is discussed.
2. Materials and methods 2.1. Chemicals Alkylated aloenins (Fig. 1) were obtained from aloenin 1 isolated from the leaves of Aloe arborescens Mill. Var. natalensis Berger (Kidachirokai or Kidachiaroe in Japanese) (Liliaceae) [8,9] as follows: 4 0 -O-methyl aloenin (2) was obtained by methylation of 1 with diazomethane [10], and other 4 0 -Oalkylaloenins were obtained by the following general method. A mixture of 1, alkyl bromides or alkyl iodides (2.0–10.0 eq.) and potassium carbonate (2.0–4.0 eq.) was refluxed in dry acetone. The mixture was evaporated, then poured into water and extracted with ethyl acetate, and further evaporated. The resulting residue was purified by column chromatography (a gradient of 0–20% MeOH in CHCl3) to give the desired 4 0 -O-alkylaloenins (3–17 and 20), 4 0 -Obenzylaloenin (18) and 4 0 -O-allylaloenin (19). 4 0 -OAlkylaloenin sulfates (21–29) (Table 2) were obtained by the reaction of 4 0 -O-alkylaloenins 2, 7, 11, 13–17 and 20 with SO3-pyridine complex in pyridine according to the method previously reported [6]. Test compounds were dissolved in dimethyl sulfoxide (DMSO) and diluted with culture medium to the desired concentrations. MTT (3-(4,5-dimethylthiazol2-yl)-2,5-di-phenyltetrazolium bromide) was obtained from Chemicon International, Inc. (CA, USA).
116/VCR 100-1-1 cell lines were a kind gift from Daiichi Pharmaceutical Co. Lit, Japan. Dulbecco’s modified Eagle’s medium (DMEM), McCoy’s 5A medium, fetal bovine serum (FBS) and penicillin– streptomycin mixture (100 U/ml penicillin and 100 mg/ml streptomycin) were purchased from Wako Pure Chemical Industries Ltd (Osaka, Japan), Sigma (MO, USA), Biosource International (CA, USA) and Bio Whittaker (ND, USA), respectively. The HCT 116, HCT 116/VCR 100-1-1 and Hep G2 cells were maintained in McCoy’s 5A medium and in DMEM, respectively. Each medium was supplemented with 10% FBS and a penicillin–streptomycin mixture at 37 8C in a humidified atmosphere containing 5% CO2. 2.3. Cytotoxicity test Aliquots (200 ml) of 5!103 cells/ml of HCT 116, HCT 116/VCR 100-1-1 and Hep G2 cells were seeded in 96 well flat-bottomed plates (Microteste Tissue Culture Plate, 96 Well, Flat Bottom with Low Evaporation Ltd, Falcon, NJ, USA), and were incubated in a medium containing 10% FBS and a penicillin–streptomycin mixture at 37 8C in a humidified atmosphere of 5% CO2 for 24 h. The test drugs were dissolved in dimethyl sulfoxide (DMSO).
2.2. Cells and cell culture HCT 116 (ATCC No. CCL-247) and Hep G2 (No. RCB0459) cell lines were purchased from Dainippon Pharmaceutical Co. Ltd (Osaka, Japan) and RIKEN Cell Bank (Tsukuba, Japan), respectively. HCT
Fig. 1. Structures of aloenin 1 and 4 0 -O-alkylaloenins 2–19.
G.-z. Jin et al. / Cancer Letters 218 (2005) 15–20
The incubation medium was replaced with each test medium giving a final concentration of 1–500 mmol/l of compounds 1–19 and 21–29 and no drug in 2 ml DMSO over 2 days. The cell survival was determined by the method reported by Mosmann using MTT assay [7]. Each experiment was performed in duplicate wells, and all experiments involving control (DMSO only) and drug treatments were performed separately 3–5 times. Data represent meanGSD values. 2.4. Statistical analysis Regression lines were calculated using StatisticaTM release software, version 5.5 (StatSoft Japan Inc., Tokyo, Japan). Statistical significance was defined as P!0.01.
3. Results and discussion The cytotoxic effects of aloenin 1, 4 0 -O-alkylaloenins (2–17), 4 0 -O-benzylaloenin (18) and 4 0 -Oallylaloenin (19) were tested by MTT assay [7] using HCT 116 and Hep G2 cell lines. The IC50 values were calculated based on the percentage inhibition of cell growth and these are listed in Table 1. Aloenin 1 showed no marked cytotoxicity (IC50Z293.1G 32.7 mM; this value was represented as O100 mM in Table 1) to HCT 116 cells. Hayashi et al. [11] suggested that the appropriate hydrophobicity of flavanone compounds was important for cytotoxicity to KB, KB-VIN and related cell lines. This suggestion let us to introduce various alkyl groups into the aloenin molecule. Aloenin 1 has two kinds of alcohols in the molecule: a phenolic alcohol at the C-4 0 position and aliphatic alcohols on a glucopyranose moiety. In this study, alkyl groups were introduced to the phenolic alcohol in order to enhance the hydrophobicity and to confer the appropriate interfacial property to aloenin 1 to obtain various 4 0 -O-alkylated aloenins (2–17). The cytotoxic effects of these alkylated compounds were tested on two cell lines, HCT 116 and Hep G2, and the mean IC50 values for these cells are listed in Table 1. In the case of HCT 116 cells, the IC50 values of 4 0 O-methyl- (2) and 4 0 -O-ethylaloenin (3) were higher than 100 mM as well as aloenin 1 itself, indicating that
17
they showed no marked cytotoxicity. The cytotoxicity gradually increased on increasing the length of the alkyl chains from propyl to hexyl (IC50 values: 79.5G 38.1, 74.1G2.2, 37.9G12.4, and 17.9G5.5 mM for propyl- 4, butyl- 5, pentyl- 6, and hexylaloenin 7, respectively), and the 4 0 -O-alkylaloenins 8–14 with longer alkyl groups from heptyl to tetradecyl showed IC50 values of 5.3G2.3–10.4G2.5 mM, exhibiting potent cytotoxicity to HCT 116 cells. Among compounds 8–14, compound 14 exhibited the most potent cytotoxicity (IC50 value: 5.3G2.3 mM). However, the IC50 value (16.2G4.9 mM) of 4 0 -O-hexadecylaloenin (15), having a longer alkyl chain by two CH2 units than 14, was slightly greater than that of 8–14, showing a slight reduction in activity. Furthermore, 4 0 -O-octadecyl- (16) and 4 0 -O-eicosylaloenin (17), having a longer alkyl chain than 15, had an IC50 value greater than 100 mM, and exhibited no marked cytotoxicity. In these experiments, while all solutions of samples 1–14 and the cells in the medium were clear, the solution of 15 at a concentration of 150 mM was not completely clear and solutions of 16 and 17 at Table 1 Cytotoxic effects of aloenin 1 and 4 0 -O-alkylaloenins 2–17 on HCT 116 and Hep G2 cells Compound
IC50 (mM)a HCT 116
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
b
O100 O100b O100b 79.5G38.1 74.1G2.2 37.9G12.4 17.9G5.5 8.1G6.8 10.4G2.5 5.7G0.5 8.9G2.1 8.7G2.1 6.6G0.7 5.3G2.3 16.2G4.9 O100b O100b
Hep G2 O100b O100b O100b O100b O100b O100b 36.2G7.1* 28.5G9.2* 23.4G2.2* 14.7G1.5* 13.5G2.3 20.1G3.7* 12.1G3.1* 4.0G0.6 16.0G0.9 O100b O100b
*P!0.01, the significant difference between the activity on HCT 116 cells and that on Hep G2 cells. a IC50 values (meanGSD) are the concentrations at which 50% of the cells are inhibited from growing. b IC50 values more than 100 mM were indicated as O100.
18
G.-z. Jin et al. / Cancer Letters 218 (2005) 15–20
Fig. 2. Structures of 4 0 -O-alkylaloenin sulfates 21–29.
the same concentration were fairly turbid. These lower solubilities may be the cause of the reduction in the IC50 values of compounds 15–17. In the case of Hep G2 cells, while alkylaloenins 2–6 having shorter alkyl groups than a pentyl had IC50 values higher than 100 mM, exhibiting no marked cytotoxicity, compounds 7–14 had IC50 values from 36.2G7.1 to 4.0G0.6 mM. Among alkylaloenins 7–14, tetradecylaloenin 14 showed the most potent cytotoxicity having an IC50 value of 4.0G0.6 mM. Also, the IC50 value (16.0G0.9 mM) of compound 15 was higher than that of 14 (4.0G0.6 mM) and those of compounds 16 and 17 were higher than 100 mM. The increases may be due to the same reason mentioned above. Thus, the cytotoxic effects on HCT 116 and Hep G2 cells were enhanced by the introduction of alkyl groups having an appropriate length at the 4 0 -O position of 1. Tetradecylaloenin 14 showed the most potent cytotoxic effect on HCT 116 cells (IC50 value: 5.3G2.3) and Hep G2 cells (IC50 value: 4.0G 0.6 mM). Here the relationships between the cytotoxic effects on HCT 116 cells (shown by the IC50 values) and the hydrophobicity (shown by C log P values [12]) of 4 0 -O-alkylaloenin derivatives 4–14 were determined and are shown in Fig. 3. The relationships showed positive correlations between the cytotoxic effects of 4 0 -O-alkylaloenin derivatives 4–14 in HCT 116 cells and the hydrophobicities caused by the introduction of alkyl groups at the O-4 0 position of aloenin 1. These relationships were also observed for 4 0 -O-benzylaloenin 18 and 4 0 -O-allylaloenin 19; the more hydrophobic 18 (C log PZ1.81) had a more potent cytotoxic effect (IC50 value: 32.9G4.9 mM), on the other hand the less hydrophobic 19 (C log PZ 0.82) had a weaker effect (IC50 value: 50.2G3.1 mM) on the growth of HCT 116 cells (Fig. 3).
With respect to the relationship between cytotoxic activity and hydrophobicity of compounds, Iwase et al. [13] reported that the extending the alkyl chain at the 8 or 13 position of berberine affected the cytotoxic activity, and the highest level of activity was observed in 8- or 13-hexyl-substituted derivatives of berberin. This agreed with the results of the activities of 4 0 -O-alkylaloenin derivaties 4–14; the introducing of an appropriate hydrophobicity as well as an appropriate size of the substituent into aloenin molecule may enhance the cytotoxic activity, although the exact mechanism for the affinity to the enzyme or intracellular accumulation and so on has not been solved. No relationship was found for hexadecyl- 15, octadecyl- 16 and eicosylaloenin 17 because of the lower solubility of these compounds. In order to enhance the solubility of 15, 16 and 17 in the test medium, the hydroxyl groups on the glucopyranose moiety were sulfated [6] as well as those of 2, 7, 11, 13, 14 and 20. All the synthetic 4 0 -O-alkylaloenin sulfates 21–29 are listed in Fig. 2. The cytotoxic effects of these sulfates on HCT 116 cells and Hep G2 cells are listed in Table 2. In this case, while alkylaloenin sulfates 21–23 having shorter alkyl groups than decyl showed no marked activity, Table 2 Cytotoxic effects of 4 0 -O-alkylaloenins sulfates 21–29 on HCT 116, Hep G2 and HCT 116/VCR 100-1-1 cells Compound
21 22 23 24 25 26 27 28 29
IC50 (mM)a HCT 116
Hep G2
HCT 116/ VCR 100-1-1
O100b O100b O100b 27.1G6.6 47.1G19.6 22.2G4.5 23.2G2.2 12.3G4.0 4.8G0.2
O100b O100b 99.4G9.1 21.4G4.6 24.0G2.0c 18.1G1.6 14.8G5.1c 9.1G0.9 4.5G0.5
O100b O100b O100b 31.4G4.8 27.6G3.1d 21.3G1.3 18.9G1.7 14.9G1.9 19.0G2.6d
a IC50 values (meanGSD) are the concentrations at which 50% of the cells are inhibited from growing. b IC50 values more than 100 mM were indicated as O100. c P!0.01, the significant differences between the activity on HCT 116 cells and that on Hep G2 cells, respectively. d P!0.01, the significant differences between the activity on HCT 116 cells and that on HCT 116/VCR 100-1-1 cells, respectively.
G.-z. Jin et al. / Cancer Letters 218 (2005) 15–20
Fig. 3. Relationships between C log P and the cytotoxic effects of 4 0 -O-alkylaloenins on HCT 116 and Hep G2 cells. Closed circle, HCT 116 cells; open circle, Hep G2 cells. Regression lines for HCT 116 and Hep G2 cells were yZ81.4 eK0.469x (R2Z0.733, P!0.001) and yZ147.5 eK0.489x (R2Z0.827, P!0.001), respectively. Each value shows the meanGSD.
alkylaloenin sulfates 24–29 having longer alkyl groups than docosyl exhibited greater cytotoxicity, and 4 0 -O-dococylaloenin sulfate 29 having the longest alkyl group of 21–29 had the most potent cytotoxic effect on HCT 116 and Hep G2 cells. These results suggest that the cytotoxicity of aloenin 1 is enhanced by having an appropriate length alkyl group thereby giving an appropriate interfacial property to 1. The cytotoxic effects of alkylaloenins 4–14 on HCT 116 cells were more potent than on Hep G2 cells (Table 1 and Fig. 3). This is thought to be due to the difference in physiological properties between HCT 116 and Hep G2 cells. It is known that while HCT 116 cell lines scarcely express the MDR 1 (P-glycoprotein; P-gp) [14], Hep G2 cell lines overexpress it [15]. P-gp acts as an efflux pump to remove several antitumor agents, Ca2C antagonists, cyclosporine, digoxin and other compounds from cells [5,16]. In spite of the overexpression of P-gp in Hep G2 cells, compounds 7–14 had potent cytotoxic effects on the cells. It has been reported that HCT 116/VCR 100-1-1 cells, a mutant of HCT 116 cells, overexpress P-gp, and the IC50 value of a cytotoxic compound such as taxane to the cells is higher than 1000 times by the efflux system such as P-gp of the cells compared with that to HCT 116 cells [14]. Here the compounds 7–14 which were potent active to HCT 116 cells were further measured using HCT 116/VCR 100-1-1 cells and compared with the activities on HCT 116 (Fig. 4).
19
From the figure, while the cytotoxic effects of 7–14 on HCT were from 1.8 to 4.3 times higher than those on HCT 116/VCR 100-1-1, the alkylaloenins 7–14 still had a potent effect on HCT 116/VCR 100-1-1 cell lines which overexpressed P-gp, as was the case with Hep G2. The cytotoxic effects of alkylaloenin sulfates 24–29 on HCT 116, Hep G2 and HCT 116/VCR 100-1-1 cells were also investigated. These sulfates were cytotoxic to Hep G2 and HCT 116/VCR 100-1-1 cells which overexpressed P-gp as well as on HCT 116 which scarcely expressed it (Table 2). This indicates that these sulfates may be lower sensitive substrates for P-gp transport in comparison with the alkylaloenins. In conclusion, aloenin 1 has no marked cytotoxic effect on HCT 116 and Hep G2 cancer cell lines. However, this effect was enhanced by the introduction of an appropriate length alkyl group at the O-4 0 position of 1. Among 4 0 -O-alkylaloenin derivatives 2–17 which were synthesized from 1, 7–15 were markedly cytotoxic not only to HCT 116 cells which scarcely express P-gp but also Hep G2 and HCT 116/VCR 100-1-1 cells which overexpress P-gp. Among 7–15, 4 0 -O-tetradecylaloenin 14 had the most potent cytotoxic effect on HCT 116, HCT 116/VCR 100-1-1 and Hep G2 cell lines. Comparison of the activities of 4 0 -O-alkylaloenin sulfates 21–29 showed that 24–29, having longer alkyl chains, were more potent than sulfates 21–23 with shorter alkyl chains as far as the effects on HCT 116, HCT 116/VCR 100-1-1 and Hep G2 cells were concerned. These results suggest that an appropriate balance
Fig. 4. Cytotoxic effects of 4 0 -O-alkylaloenins 7–14 on the growth of HCT 116 and HCT 116/VCR100-1-1 cancer cell lines. Open bar, HCT 116 cells; closed bar, HCT 116/VCR100-1-1. *P!0.01.
20
G.-z. Jin et al. / Cancer Letters 218 (2005) 15–20
between the hydrophobicity of the aglycon moiety and the hydrophilicity of the sugar moiety in aloenin derivatives is important for exhibiting a potent cytotoxic effect on these cancer cells. Each 4 0 -O-alkylaloenin sulfate of 24–29 was more cytotoxic to P-gp overexpressed Hep G2 cells than the corresponding one to HCT 116 cells (Table 2). Furthermore, 24–29 were markedly cytotoxic to P-gp overexpressed HCT 116/VCR 100-1-1 cells. These results also suggest that these sulfates 24–29 are not influenced by efflux systems such as P-gp as well as alkylaloenins 7–14. Thus the finding that these derivatives 7–14 and 24–29 were markedly cytotoxic to P-gp overexpressed cancer cells without any modulator suggests a new approach to overcome the obstacle posed by the multidrug resistance in cancer. We are continuing our investigations of cytotoxicity using other types of P-gp-overexpressing cells.
References [1] M.M. Gottensman, I. Pasten, Biochemistry of multidrug resistance mediated by the multidrug transporter, Annu. Rev. Biochem. 62 (1993) 385–427. [2] O. Fardel, V. Lecureur, A. Guillonzo, The P-glycoprotein multidrug transporter, Gen. Pharmacol. 27 (1996) 1283–1291. [3] M.M. Gottensman, T. Fojo, S.E. Bates, Multidrug resistance in cancer: role of ATP-dependent transporters, Nat. Rev. Cancer 2 (2002) 48–58. [4] R. Krishna, L.D. Mayer, Multidrug resistance (MDR) in cancer. Mechanisms, reversal using modulators of MDR and the role of MDR modulators in influencing the pharmacokinetics of anticancer drugs, Eur. J. Pharm. Sci. 11 (2000) 265–283.
[5] D.R. Ferry, H. Traunecker, D.J. Kerr, Clinical trial of Pglycoprotein reversal in solid tumours, Eur. J. Cancer 32A (1996) 1070–1081. [6] S. Saito, T. Furumoto, M. Ociai, A. Hosono, H. Hoshino, U. Haraguchi, et al., Synthetic studies on the relationship between anti-HIV activities and micelle forming abilities of various alkylated glycyrrhetinate diglycoside sodium sulfates and related compounds, Eur. J. Med. Chem. 31 (1996) 365–381. [7] T. Mosmann, Rapid colorimetric assay for cellular growth and survival; application to proliferation and cytotoxicity assays, J. Immunol. Methods 139 (1983) 55–63. [8] J.E. Hay, L.J. Haynes, The Aloins. Part I. The structure of barbaloin, J. Chem. Soc. 1956; 3141–3147. [9] N. Okamura, N. Hine, S. Harada, T. Fujioka, K. Mihashi, A. Yagi, Three chromone components from Aloe Vera leaves, Phytochemistry 43 (1996) 495–498. [10] Th. J. De Boer, H.J. Backer, Diazomethane, Org. Synth. IV (1963) 250–253. [11] K. Hayashi, Y. Nakanishi, K.F. Bastow, G. Cragg, H. Nozaki, K.-H. Lee, Anti tumor agents 221. Buceracidins A and B, two new flavanones from Bucida buceras, J. Nat. Prod. 66 (2003) 125–127. [12] C Log P values of derivatives in this study were calculated by ChemDraw Ultra 8.0 (CambridgeSoft, Cambridge Massachusetts, USA). [13] K. Iwase, M. Moriyasu, T. Yamori, T. Turuo, D.-U. Lee, W. Wiegrebe, In vitro Cytotoxicity of the Protoberberine-type Alkoloids. J. Nat. Prod. 64 (2001) 896–898. [14] M. Shionoya, T. Jimbo, M. Kitagawa, T. Soga, A. Tohogo, DJ927, a novel oral taxane, overcomes P-glycoprotein-mediated multidrug resistance in vitro and in vivo, Cancer Sci. 94 (2003) 459–466. [15] G. Lee, M. Piqette-Miller, Influence of IL-6 on MDR and MRP-mediated multidrug resistance in human hepatoma cells, Can. J. Physiol. Pharmacol. 79 (2001) 876–884. [16] K. Takara, T. Sakaeda, Y. Tanigawara, K. Nishiguchi, N. Ohmoto, M. Horinouchi, et al., Effects of 12 Ca(2C) antagonists on multidrug resistance, MDR1-mediated transport and MDR1 mRNA expression, Eur. J. Pharm. Sci. 16 (2002) 159–165.