Screening of 14 alkaloids isolated from Haplophyllum A. Juss. for their cytotoxic properties

Screening of 14 alkaloids isolated from Haplophyllum A. Juss. for their cytotoxic properties

Journal of Ethnopharmacology 105 (2006) 241–245 Screening of 14 alkaloids isolated from Haplophyllum A. Juss. for their cytotoxic properties Olivia J...

375KB Sizes 140 Downloads 71 Views

Journal of Ethnopharmacology 105 (2006) 241–245

Screening of 14 alkaloids isolated from Haplophyllum A. Juss. for their cytotoxic properties Olivia Jansen a , V. Akhmedjanova b , L. Angenot a , G. Balansard c , A. Chariot d , E. Ollivier c , M. Tits a , M. Fr´ed´erich a,∗ a

Laboratory of Pharmacognosy, Natural and Synthetic Drug Research Centre, University of Li`ege, Bˆat. B36, Av. de I’hˆopital 1, 4000 Li`ege, Belgium b Uzbek Academy of Sciences, Yunusov Institute of the Chemistry of Plant Substances, Tashkent, Uzbekistan c Laboratory of Pharmacognosy, Faculty of Pharmacy, University of Aix-Marseille II, France d Laboratory of Medical Chemistry and Medical Oncology, Centre for Cellular and Molecular Therapy, University of Li` ege, Belgium Received 14 March 2005; received in revised form 10 August 2005; accepted 1 November 2005 Available online 2 December 2005

Abstract Further to a systematic chemotaxonomic study of Uzbek Haplophyllum A. Juss. plants selected on ethnopharmacological data, 14 alkaloids were screened for their cytotoxic properties. As a first selection for interesting compounds, each alkaloid was tested against two human cancer cell lines (HeLa and HCT-116), using WST-1 reagent. Of the 14 alkaloids, 5 were cytotoxic when tested against the HeLa line with an IC50 < 100 ␮M. These five compounds consisted of three furoquinolines: skimmianine; haplopine and ␥-fagarine and two pyranoquinolones: flindersine and haplamine. Only haplamine was active against the HCT-116 line. The cytotoxic properties of these five alkaloids were further investigated against five additional human cancer cell lines. Their structure–activity relationships will be discussed. Of these five pre-selected alkaloids, only haplamine showed significant cytotoxic activity against all the tested cell lines. This is the first report of the cytotoxic activity of haplamine. Finally, this pyranoquinolone alkaloid was tested here against 14 different cancer cell lines and against normal skin fibroblasts. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Rutaceae; Haplophyllum; Quinoline alkaloid; Cytotoxicity; Haplamine

1. Introduction Cancer is a very widespread disease, which is responsible for millions of deaths each year worldwide. Chemotherapy is an essential strategy for the treatment of disseminated cancers; however, its efficiency is restricted by both intrinsic and acquired cell resistance to drugs. To circumvent chimioresistance to conventional anticancer drugs, anticancer compounds with new cellular targets are needed. This observation stimulates the search for new anticancer agents, and in this regard, the investigation of naturally originating compounds could be very valuable. The part of the project described in this paper is more focused on the Uzbek flora, and, in particular, on the genus Haplophyl-



Corresponding author. Tel.: +32 4 366 43 31; fax: +32 4 366 43 32. E-mail address: [email protected] (M. Fr´ed´erich).

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.11.001

lum A. Juss. and its alkaloidal content. The genus Haplophyllum (Rutaceae) numbers ca. 70 species. The plants are perennial herbs with a pervasive smell, distributed from the Mediterranean to East Siberia. Fourteen species are found in Uzbekistan (Townsend, 1986). Previous phytochemical analysis of various Uzbek Haplophyllum species has revealed the presence of lignans (Nesmelova et al., 1983), coumarins (Batirov et al., 1982; Yuldashev, 2002), flavonoids (Yuldashev, 2002) and several classes of alkaloids: quinolones (Akhmedjanova et al., 1980; Bessonova, 2000); furoquinolines (Akhmedjanova et al., 1986); dihydrofuroquinolines (Yunusov and Sidyakin, 1955; Akhmedjanova et al., 1980); tetrahydrofuroquinolines (Faizutdinova et al., 1968), and pyranoquinolones (Akhmedjanova et al., 1976). The genus Haplophyllum is widespread in Central Asia and so has been commonly used for a long time in folk medicine by the local population. In the earliest scientific sources, such as the Canon Medicinae by Avicenna, it is indicated that “sadab-

242

O. Jansen et al. / Journal of Ethnopharmacology 105 (2006) 241–245

ruta” (the name given to different Haplophyllum species) could be applied for treating different diseases. It could be used alone or sometimes in combination with other plants, generally using some topical forms to treat, among other diseases, warts, herpes, lichens, tumour of the testes, and erysipelas. These folk uses could indicate biocidal properties (especially cytotoxic properties). More recently, Graham et al. (2000) reported that the aerial parts of Haplophyllum dauricum have been used to treat tumours in Russia. Nevertheless, the cytotoxic properties of Haplophyllum alkaloids have never been studied. The main objective of the INTAS project consists of helping Uzbek and Georgian institutes to select some potentially antitumoral natural compounds from promising endemic plants containing alkaloids and saponins. With this aim in view, 14 quinoline alkaloids isolated from different Haplophyllum species occurring in Uzbekistan were screened for their cytotoxic properties against different human cancer cell lines. The alkaloids consisted of three furoquinolines: skimmianine, ␥-fagarine, and haplopine; two dihydrofuroquinolines: dubinidine and dubinine; two tetrahydrofuroquinolines: perforine and haplophyllidine; two 2-phenylquinolin-4-ones: graveoline and N-methyl-2-phenylquinolin-4-one; three quinolin-2-ones: 4-hydroxyquinolin-2-one, bucharaine, and foliosidine; and two pyranoquinolin-2-ones: flindersine and haplamine. Chemical structures are given in Fig. 1.

2. Methodology 2.1. Plant materials The aerial parts of Haplophyllum leptomerum Licz. et Vved., Haplophyllum bucharicum Litv., Haplophyllum foliosum Vved., Haplophyllum dubium Korov., Haplophyllum perforatum Kar. et Kir. (Rutaceae) were collected during the flowering phase in Uzbekistan. They were then identified by Dr. A.M. Nigmatullaev, Laboratory of Medicinal Plants, S.Yu. Yunusov Institute of the Chemistry of Plant Substances, Academy of Sciences of Uzbekistan, Tashkent. Voucher specimens were deposited at the Institute’s Herbarium. 2.2. Alkaloids and alkaloid solutions The 14 alkaloids studied here were isolated from the above Haplophyllum species at the Institute (Tashkent) according to procedures cited: skimmianine, ␥-fagarine and N-methyl-2phenylquinolin-4-one were extracted from Haplophyllum leptomerum Licz. & Vved. (Akhmedjanova et al., 1986); haplopine, 4-hydroxyquinolin-2-one and bucharaine from Haplophyllum bucharicum Litv. (Bessonova, 2000); dubinidine, graveoline and foliosidine from Haplophyllum foliosum Vved. (Akhmedjanova et al., 1980); dubinine from Haplophyllum dubium (Yunusov and Sidyakin, 1955), perforine and haplophyllidine from Hap-

Fig. 1. Fourteen alkaloids from Haplophyllum A. Juss sp. screened for their cytotoxic properties.

O. Jansen et al. / Journal of Ethnopharmacology 105 (2006) 241–245

lophyllum perforatum Kar. & Kir (Faizutdinova et al., 1968); flindersine and haplamine from Haplophyllum perforatum Kar. & Kir (Akhmedjanova et al., 1976). The alkaloids were identified with authentic samples (Shakirov et al., 1996), according to their physico-chemical constants and spectral data (UV, IR, mass, and 1 H NMR spectra) or by direct comparison of the samples (TLC, mp of a mixed probe, IR spectra). The 14 compounds were dissolved separately in DMSO and each of these stock solutions was diluted with culture medium to desired concentrations. Final DMSO concentration never exceeded 5‰. 2.3. Cells lines 2.3.1. Different human cell lines were used in this study He La: epitheloid cervix carcinoma; HCT-116: colon carcinoma; HT-29: colon adenocarcinoma; OVCAR-3: ovary adenocarcinoma; MCF-7, MDA-MB-231 and SKBR-3: breast adenocarcinoma; Jurkat: acute lymphoblastic leukaemia; HPBALL: acute lymphoblastic leukaemia; ReH: acute lymphoblastic leukaemia; THP-1: acute monocytic leukaemia; HL 60: acute myeloid leukaemia; JOHN: normal skin fibroblasts. The HeLa and JOHN cells were maintained in continuous culture in DMEM medium (Cambrex); HCT-116 and SKBR-3 cells in Me Coy’s 5A modified medium (Cambrex); HT-29, MCF7, OVCAR-3, MDA-MB-231, Jurkat, HL-60, ReH, HPB-ALL and THP-1 cells in RPMI 1640 medium (Cambrex) in a humid atmosphere at 37 ◦ C and 5.5% CO2 . Each medium was supplemented with 10% heat-inactivated foetal bovine serum (Cambrex), 1% l-glutamine (200 mM) (Cambrex) and antibiotics: penicillin (100 UI/ml)–streptomycin (100 ␮g/ml) (Pen-strep® , Cambrex). 2.4. Cytotoxicity assay The assay was based on the cleavage of the tetrazolium salt WST-1 (Roche Biomolecular) producing a soluble formazan salt. This conversion only occurs in viable cells. First, 96-well tissue culture microplates (Micro Test-96® , Falcon, Becton-Dickinson) were seeded with 100 ␮l medium containing x cells in suspension (x = 7000 cells/well for HeLa, HCT-116, MCF-7, HT-29, OVCAR-3, MDA-MB-231 and SKBR-3 cells; 20,000 cells/well for Jurkat, HL-60 and THP-1; 50,000 cells/well for HPB-ALL cells; 70,000 cells/well for REH cells; 5000 cells/well for JOHN cells). After 24 h incubation, cells were treated with a dilution of alkaloid in culture medium. After 48 h incubation in the presence of alkaloid, cell viability was determined by adding WST-1 tetrazolium salt as a cytotoxicity indicator and by reading absorbance at 450 nm with a scanning multiwell spectrophotometer after about one hour’s wait. The level of absorbance directly correlates to the viable cell number. Results were expressed as a percentage of negative control, fixed at 100%. Each condition was realized in triplicate and each set of tests was performed twice, so each condition was repeated in six microwells. Colchicine was used as a positive control.

243

2.5. Statistics Differences betweeen control and the tests samples were analyzed by the Student’s t-test. P-values lower than 0.001 or 0.005 were considered significant (Figs. 3 and 4). 3. Results and discussion In order to conduct a first selection, each of the 14 pure compounds was tested against the two cell lines, HeLa and HCT-116. The results (Table 1) are expressed by the IC50 values (concentration inhibiting 50% of cell growth). IC50 values were calculated from graphs. From this first screening, the alkaloids with an IC50 value lower than 100 ␮M on at least one of the two cell lines (HeLa or HCT-116), were selected. We observed that five alkaloids were cytotoxic on the Hela line, with IC50 < 100 ␮M being the highest tested concentration of drugs. These five most cytotoxic compounds were three furoquinolines: skimmianine, haplopine and ␥-fagarine and two pyranoquinolones: flindersine and haplamine. Only haplamine was moderately active on the HCT-116 colon cancer line. The cytotoxic properties of these five pre-selected alkaloids were further investigated on five additional human cancer cell lines. The limit of activity was defined as IC50 < 100 ␮M (as for the first step of the screening). The results for the first and the second step of the screening are summarised in Table 2. From these five pre-selected alkaloids, we could observe that none but haplamine showed cytotoxic activity with IC50 values ≤100 ␮M on the five newly tested cancer cell lines. On the other hand, if we compare the cell viability at 100 ␮M (the highest tested concentration of alkaloids), haplamine seemed significantly more cytotoxic than the four other pre-selected alkaloids (see Fig. 2). From these results, we can also deduce structure–activity relationships. We observed (on the HeLa line) that the cytotoxic properties of the three furoquinoline alkaloids seem to be correlated with their degree of methoxylation (IC50 skimmianine < IC50 haplopine < IC50 gamma-fagarine). This seems also to be Table 1 IC50 values for the 14 alkaloids tested against HeLa and HCT-116 cells

␥-Fagarine Haplopine Skimmianine Dubinine Dubinidine Perforine Haplophyllidine N-Methyl-2Phenylquinolin-4-one Graveoline 4-Hydroxyquinolin-2-one Foliosidine Bucharaine Flindersine Haplamine Colchicine (positive control)

Hela cells IC50 (␮M)

HCT 116 cells IC50 (␮M)

34.9 ± 9.38 29.37 ± 13.84 11.55 ± 0.20 >100 >100 >100 >100 >100

>100 >100 >100 >100 >100 >100 >100 >100

>100 >100 >100 >100 49.37 ± 6.84 36.50 ± 2.03 1.1 ± 0.6

>100 >100 >100 >100 >100 64.50 ± 14.16 1.3 ± 0.9

244

O. Jansen et al. / Journal of Ethnopharmacology 105 (2006) 241–245

Table 2 Cytotoxicity results for the five pre-selected alkaloids against seven cell lines IC50 (␮M)

HCT 116

HT-29

OVCAR3

MDA-MB-231

SKBR-3

MCF-7

HeLa

␥-Fagarine Haplopine Skimmianine Flindersine Haplamine

>100 >100 >100 >100 64.50 ± 14.46

>100 >100 >100 >100 52.8%a

>100 >100 >100 >100 53.9%a

>100 >100 >100 >100 49.4%a

>100 >100 >100 >100 83.49 ± 17.26

>100 >100 >100 >100 28.52 ± 1.00

34.9 29.37 11.55 49.37 36.50

a

± ± ± ± ±

9.38 13.84 0.20 6.84 2.03

Non-linear relationship between cell viability and log (concentration); % viable cells at 100 ␮M given.

Fig. 2. Comparison of cell viability after treatment with flindersine, haplamine, ␥-fagarine, haplopine and skimmianine at 100 ␮M.

true for the two pyranoquinolone alkaloids (IC50 haplamine < IC50 flindersine) (see Table 2). The cytotoxic properties of haplamine were then investigated on six further cell lines: five acute leukaemia lines (HPB-ALL, Jurkat, REH, THP-1 and HL-60) and one normal skin fibroblast line (John). Results are presented in Figs. 3 and 4. We observed that haplamine was particularly active on the five acute leukaemia lines with IC50 values between 25 and 50 ␮M. At the highest tested concentration (100 ␮M), the cell viability did not reach more than 10% for four of the acute leukaemia lines. Furthermore, haplamine was significantly less cytotoxic against the non-cancer normal skin fibroblasts in comparison with HeLa (see Fig. 4) or with leukaemia cell lines, indicating a selectivity in its cytotoxic properties. In conclusion, this screening revealed the cytotoxic properties of haplamine against cancer cells, reported here for the first time. This alkaloid, tested against 14 human cancer cell lines, showed cytotoxic activity with more or less of a marked effect depending

on the cell line. However, IC50 S were always <100 ␮M. The cytotoxicity of haplamine was especially marked on HeLa cells and on acute leukaemia lines. By contrast, its cytotoxicity was weaker on normal skin fibroblasts (IC50 > 100␮M) and so its cytotoxic properties were shown to be quite selective for cancer cells. Even if haplamine’s IC50 values are somewhat higher than in the case of drugs used in clinical chemotherapy, this alkaloid could become an interesting product for two reasons: first, its low in vivo toxicity (LD50 = 1020 mg/kg, i.p., mice) and second, its sedative and analgesic properties (Akhmedkhodzhaeva et al., 1975). This constitutes a promising and original profile that warrants our further study. Further investigations would require a large amount of haplamine. Various Uzbek Haplophyllum species could be used for extraction for this purpose. Moreover, as the chemical synthesis of haplamine is already described in the literature (Shobana et al., 1989), the product could also be obtained easily in this

Fig. 3. Evolution of cell viability after treatment with haplamine for five acute leukaemia cell lines.

O. Jansen et al. / Journal of Ethnopharmacology 105 (2006) 241–245

245

Fig. 4. Evolution of cell viability after treatment wilh haplamine: normal skin fibroblasts comparing to HeLa cervix carcinoma cells.

way. Our study has shown that the methoxy group added to haplamine seems to increase its cytotoxic potential in comparison with flindersine. This finding provides a clue for further chemical modification and, consequently, some pharmacomodulation experiments could also be performed with the aim of increasing the activity or selectivity of haplamine. The traditional antitumoral uses of the genus Haplophyllum A. Juss. cannot probably be explained by haplamine alone. The antitumour activity should be partially explained by the arylnaphtalene-type lignans found in many Haplophyllum species. These compounds were known to possess antitmour activity (MacRae and Towers, 1984). Acknowledgements The authors wish to thank INTAS (International association for the promotion of co-operation with scientists from the New Independent States of the former Soviet Union) for supporting our project. We are also very grateful to Dr. B. Nusgens (Laboratory of Connective Tissues Biology, University of Li`ege) and Dr. F. Lambert (Laboratory of Haematological Molecular Biology, University of Li`ege) for providing us with the normal skin fibroblast cell lines and the five acute leukaemia cell lines used in this work, respectively. M.F. is a research associate from the Belgian National Fund for Scientific Research (FNRS). References Akhmedjanova, V.I., Bessonova, I.A., Yunusov, S.Yu., 1976. Alkaloids of Haplophyllum perforatum. Khimiya Prirodnykh Soedinenii 3, 320–328. Akhmedjanova, V.I., Bessonova, I.A., Yunusov, S.Yu., 1980. Alkaloids of Haplophyllum foliosum. Khimiya Prirodnykh Soedinenii 6, 803–805.

Akhmedjanova, V.I., Bessonova, I.A., Yunusov, S.Yu., 1986. The structure of leptomerine. Khimiya Prirodnykh Soedinenii 1, 84–86. Akhmedkhodzhaeva, Kh.S., Kurmukov, A.G., 1975. Pharmacology of haplamine. Doklady Akademii Nauk UzSSR 8, 36. Batirov, E.Kh., Matkarimov, A.D., Malikov, V.M., Seitmuratov, E., 1982. Haplophyllum obtusifoliumcoumarins. Khimiya Prirodnykh Soedinenii 6, 691–695. Bessonova, I.A., 2000. Components of Haplophyllum bucharicum. Chemistry of Natural Compounds (Translation of Khimiya Prirodnykh Soedinenii) 36, 323–324. Faizutdinova, Z.Sh., Bessonova, I.A., Yunusov, S.Yu., 1968. Structures of perforine and haplophyllidine. Khimiya Prirodnykh Soedinenii 6, 360– 367. Graham, J.G., Quinn, M.L., Fabricant, D.S., Farnsworih, N.R., 2000. Plants used against cancer—an extension of the work of Jonaihan Hartwell. Journal of Ethnopharmacology 73, 347–377. MacRae, W.D., Towers, G.H.N., 1984. Biological activities of lignans. Phytochemistry 23, 1207–1220. Nesmelova, E.F., Razakova, D.M., Akhmedzhanova, V.I., Bessonova, A.I., 1983. Diphyllin from Haplophyllum albertii-regelii H. bucharicum and H. perforatum. Khimiya Prirodnykh Soedinenii 5, 643–644. Shakirov, R., Telezhenetskaya, M.V., Bessonova, I.A., Aripova, S.F., Israilov, I.A., Sultankhodjaev, M.N., Vinogradova, V.I., Akhmedjanova, V.I.; Tulyaganov, T.S., Salimov, B.T., Telnov, V.A., 1996. The alkaloids. Plants, structure, properties. Khimiya Prirodnykh Soedinenii, 118–189, 244–342, 410–512, 615–681, 761–863, 957–1035. Shobana, N., Yersoda, P., Shanmugan, P., 1989. A convenient approach to the synthesis of prenyl-, furo- and pyranoquinoline alkaloids of the Rutaceae. Tetrahedron 45, 757–762. Townsend, C.C., 1986. Hooker’s Icones Plantarium, vol. XL (parts I–III): Taxonomic Revision of the Genus Haplophyllum. Bentham-Moxon Trustees, Whitstable, pp. 336. Yuldashev, M.P., 2002. Flavonoids and coumarins of Haplophyllum leptomerum and H. dubium. Chemistry of natural compounds 38, 192– 193. Yunusov, S.Yu., Sidyakin, G.P., 1955. Alkaloids of Haplophyllum species. II. Zhurnal Obshchei Khimii 25, 2009–2016.