Pergamw Phytochemistry. Vol. 35, No. 6, pp. 1461-1464, 1994 Copyright 0 1994 Elscvier Scicnm Ltd Printed 10 Great Britain. All rights reserved 0031~9422j94 $6.00+0.00 zyxwvutsrqp ACTIVITY OF EXTRACTS AND NAPHTHYLISOQUINOLINE ALKALOIDS FROM zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA TRIPHYOPHYLLUM PELTATUM , ANCISTROCLADUS ABBREVIATUS AND A. BARTERI AGAINST PLASM ODIUM FALCIPARUM IN VITRO* Gu1w FRANCOIS,t GERHARD BRINGMANN,t$ J. DAVID PHILLIPSON,t$ LAURENT AKB Assr,II CARINR DOCHU, MARTIN R~BENACKER,~CHRISTOPHSCHNEIDER,~MARC WBRY,DAVID C. WARHURS~~and GEOFFREYC. KIRB~~ Laboratorium voor Protozciilogie, Instituut voor Tropische Geneeskunde, Nationalestraat 155, B-2000 Antwerpen, Belgium; SOrganisch Chemisches Institut der Universitiit Wthzburg, Am Hubland, D-97074 Wihzburg, Germany, $Department of Pharmacogucsy, The School of Pharmacy, University of London, 29-39 Brunswick Square, London WClN IAX, U.K.; l/Centre National de Floristique (Conservatoire et Jardin Botaniques), Universite Abidjan, 22 B. P. 582, Abidjan, Ivory Coast; mpartment of Medical Parasitology, London School of Hygiene and Tropical Medicine, University of London, Keppel Street, London WClE 7HT, U.K. (Received 16 April 1993) Key Word I&x-Plasmodiumfalciparuq malaria; in vitro antiplasmodial activity; naphthylisoquinoline alkaloids; dioncopeltine A; dioncophyllines A and B; Ancistrocladus abbreviatus; A. barteri; Ancistrocladaceae; Triphyophylhun peltatum; Dioncophyllaceae; structure-activity relationships. Ah&act-Five extracts from the tropical plant species Triphyophyllum peltatum, Ancistrocladus abbreviatus and A. barteri, and six pure naphthylisoquinoline alkaloids derived from these species have been examined for their antiplasmodial activity. These species are well-known in the traditional medicine of West Africa and are used for the treatment of fevers, malaria and other diseases. The extracts and alkaloids were tested against the asexual erythrocytic stages of two strains of Plasmodium falciparum in vitro (Kl/chloroquine-resistant and NF 54164, clone AtAg/ chloroquine-sensitive). Incorporation of 3H-hypoxanthine was measured in the presence of the test substances after 42 hr of incubation at 37”. All extracts and three alkaloids displayed activity. The two most potent compounds were dioncopeltine A and dioncophylline B. Structure-activity considerations indicate two possible criteria for antiplasmodial activity: an R-configuration at C-3 associated with the absence of an oxygen substituent at C-6 and the absence of N-methylation. INTRODUCIION There is an urgent need for new antimalarial drugs because malaria parasites, especially Plasmodium falciparum, have become resistant to chloroquine and to other clinically used drugs [2-41. Much effort has been made to explore the potential of natural products from higher plants which are used in traditional medicine pantropically [S]. Until now, only artemisinin and its derivatives appear to be near to the stage of commercial availability [6]. There are many potential natural product candidates which are waiting to be examined as possible antimalarial drugs [7, 83. One recently discovered group of interesting natural products, the naphthylisoquinoline alkaloids, occurs in species of the Ancistrocladaceae and Dioncophyllaceae. *Part 52 in the series ‘Acetogenic Isoquinoline Alkaloids’. For Part 51, see ref. [l]. TAuthots to whom correspondence should be addressed. PHYTO 356-6 These small plant families occur in tropical Africa, southern and South East Asia. Several species such as Ancistrocladus tectorius are used for the treatment of malaria and dysentery [9], whilst others, such as Triphyophylhun peltatum, are used to treat malaria and elephantiasis W I. The naphthylisoquinoline alkaloids are unique in several respects. Firstly, their basic structure comprises a biaryl system consisting of a naphthalene and a tetrahydroisoquinoline moiety. Many of these alkaloids display atropisomerism due to the bulky ortho-substituents adjacent to the biaryl axis [ 111. Secondly, this highly unusual structure must result from an unprecedented biogenetic origin for which a polyketide pathway has been assumed [ll, 121. Some naphthylisoquinoline alkaloids in their pure forms exhibit interesting biological activities: ancistrocladidine (from A. heyneanus) has pronounced spasmolytic activity [13] and ancistrotectorine (from A. tectorius) has antitumour activity [9]. Dioncophyllines A (3) and B (1) are active as fungicides [ 143 and dioncophyl- 1461 G. FRANCOIS et al. 1462 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Dioncophylline B Dioncopeltine A Dioncophylline A (1) (2) (3) wopx” “:g; LB” 3 H3CO AkMethyldioncophyltine A Ancistrobrevine D Ancistrocladine (4) (5) (6) line A (3) has an antifeedant effect against the larvae of zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJI of .crude extracts from Table 1. IC,, values (pgrnl-‘) T. peltatum, A. abbreviatus and A. barteri tested against two [15, 161. P. falciparum strains in vitro Because some of these plants are used in folk medicine for the treatment of malaria, it appeared to be worthwhile to evaluate their in vitro antiplasmodial activity. Extracts $&4) Extract from T. peltatum, A. abbreviatus and A. barteri and some of their constituent alkaloids were selected for evaluation T. peltatum root (CH,CI,) 0.017 0.053 against P. falciparum in vitro. N.T. 0.076 T. peltatum shoot (CH,Cl,) Spodoptera littoralis root (CH#ZI,-NH,) shoot (CH,CI,-NH,) root (CH,Cl,) A. abbreoiatus RESULTS AND DISCUSSION A. abbreuiatus A. barteri The IC,, values obtained with the extracts and pure alkaloids against P. falciparum in uitro using Kl (chloroquine-resistant) and NF 54/64 clone AlA (chloroquinesensitive) are given in Tables 1 and 2, respectively. The values obtained for chloroquine and the NF 54164 strain were as follows: (mean values, n = 3 and 95% confidence intervals): IC,, = 0.002 pg ml- ’ ( -0.002-0.006 .ug ml- ‘). The values obtained for chloroquine against the Kl strain were as follows: (mean values, n = 3 and 95% confidence intervals): ICso =0.068 pgml-’ (0.037-0.125 pgml- ‘). There was ca lo-fold difference in the degree of sensitivity of the two strains of P. falciparum to chloroquine, but for the extracts and alkaloids tested the activities showed comparable orders of magnitude although none of them was as active as chloroquine (Tables 1 and 2). Dioncophylline B (1) and dioncopeltine A (2) were highly active against both strains with IC,, values <: 0.5 pg ml - I, whilst dioncophylline A (3) was somewhat less active with I& values in the order of 1 pgml-‘. N-Methyldioncophylline A (4), ancistrobrevine D (5) and ancistrocladine (6) were the least active with IC,, values in the range of 5-24 pgml- ‘. Our results suggest two criteria for antiplasmodial activity in vitro: an R-configuration at C-3 associated with the lack of an oxygen substituent at C-6 and the Kl/chloroquineresistant quine-sensitive. N.T. = not tested. 0.153 N.T. 0.553 0.105 0.03 1 0.648 and NF 54/64, clone AlA9/chIoro- Table 2. IC,, values @g ml- ‘) of six naphthylisoquinoline alkaloids tested against two P.falciparum strains in vitro Alkaloid Dioncophylline B (1) Dioncopeltine A (2) Dioncophylline A (3) N-MethyldioncophyIline Ancistrobrevine D (5) Ancistrocladine (6) Kl/chIoroquine-resistant quine-sensitive. A (4) 0.063 0.330 0.860 5.749 10.440 25.350 0.224 0.021 1.443 13.637 12.222 18.353 and NF 54/64, clone AlA9/chloro- absence of N-methylation. The first of these criteria is typical for many of the alkaloids derived from the Dioncophyllaceae [17]. It might be speculated that both criteria must be fulfilled simultaneously since dioncophyl- Antimalarial activity of alkaloids 1463 line B (l), dioncopeltine A (2) and dioncophylline A (3) Plasmodium falciparum cultures. Continuous in vitro have these structural features. The isoquinoline moiety of cultures of asexual erythrocytic stages of two P. falciparum strains (Kl and NF 54/64, clone AlA9) were mainthe three most active alkaloids is identical, but there are differences in the position of the biaryl axis in the tained following, essentially, the method of ref. [23] at 37” naphthalene moiety (either C-l’ or C-6’) and in the nature under an atm. of 5% CO1, 5% Or and 90% N, (NF of the oxygen functions. The two most active alkaloids, 54/64) or 4% COZ, 3% 0, and 93% N, (Kl). The host cells were human red blood cells (A or 0 Rh+). The dioncophylline B zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA (1) and dioncopeltine A (2) have C-5’ phenolic hydroxyl substituents, but they differ in the culture medium for the NF 54/64 strain was RPM1 1640 position of the axis and in the nature of their C-2 (Gibco), containing HEPES (BDH; 4.57 gl-i), glucose substituents (methyl and hydroxymethylene, see for- (Sigma; 1.54 gl-‘), 5% NaHCO, (Merck; 34.78 mll-‘) mulae). Dioncophylline A (3) is less active than dioncopeland gentamycin (Merck; 8.70 ml l- ‘) supplemented with 10% human plasma (A Rh+). The culture medium for tine A (2) and the two alkaloids differ in the nature of their C-2’ (methyl and hydroxymethylene, respectively) and C- the Kl strain was RPM1 1640 containing TES (Sigma; 7.64 g l- ‘), glucose (BDH; 1.67 gl- ‘), NaHCO, (BDH, 5’ (methoxyl and hydroxyl, respectively) substituents, but they have the same coupling type. The two alkaloids 1.94 gl-i) and gentamycin (Nicholas; 30 mg 1-i) supplemented with 10% human serum (A Rh+). Parasites were which can be most directly compared are dioncophylline A (3) and N-methyldioncophylline A (4), which differ in subinoculated every 3-4 days with initial conditions of 1% parasitaemia and 1% haematocrit (NF 54/64 strain) only one respect: the former has a secondary, the latter, a or 2.5% haematocrit (Kl strain). tertiary amine function. The presence of the methyl substituent on the isoquinoline nitrogen results in a sixIn vitro tests. 1. Plasmodium falciparum (NF 54/64 fold loss of activity against the Kl strain and a nine-fold clone AlA9, chloroquine-sensitive). Each extract and alkaloid was dissolved in DMSO at a concn of 20 ~1mg- ‘. loss of activity against the NF 54/64 strain (Table 2). The These solns were further diluted with physiological saline alkaloids which showed least activity, ancistrobrevine D (5) and ancistrocladine (6), have S-configurations at C-3 to obtain a stock soln of 500 pg ml- l. Each test substance was applied in a series of seven 4-fold dilutions (final and oxygen substituents at C-6. The five extracts tested for antiplasmodial in uitro concns ranging from 50 to 0.012 pg ml- ‘). Chloroquine activity contained relatively complex mixtures of (Nivaquine., Rh6ne-Poulenc) was used in the same expts as a ref. and also applied in seven 4-fold dilutions (final naphthylisoquinoline alkaloids and their ICsO values are concns from 0.5 to 0.0001 pgrnl-’ of chloroquine). Each of the same order of magnitude as those of isolated pure compound was tested in 6-fold repeats. Testing of the alkaloids (Tables 1 and 2). This result may be due to the presence of highly active minor alkaloids or due to extracts and alkaloids was performed in uitro, based upon synergism between individual alkaloids. A series of ex- the method of ref. [24]. The parasites (200~1 of a suspension with initial parasitaemia of 0.5% and haemaperiments to investigate these possibilities is planned. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFED tocrit of 1.5%) were incubated for 24 hr in microtitre plates (Falcon, MicroTest III) in hypoxanthine-free medium in the presence of 25 hl of test soln. The plates FXPJ!JtIMENTAL contained a negative control (6 wells with non-parasitized Extracts and alkaloids. Five extracts were prepd for RBCs, no drug) and a positive control (6 wells with parasitized RBCs, no drug). Thereafter, 25 ~1 of 3HP.filciparum in vitro testing: T. peltatutn CH,Cl, root and shoot extracts, A. abbreuiutus CH,Cl,-NH, root and hypoxanthine (Amersham) was added (0.5 &i well- ‘) and the parasites were incubated for a further period of shoot extracts and A. barteri CH&l, root extracts. 18 hr. Each well was harvested with a Cell Harvester Samples of T. peltatum (January 1988), A. abbreuiatus (Nunc). The filter papers were dried for 2 hr at 52” and (July 1990) and A. barteri (January 1988) were collected their radioactivity was measured by a liquid scintillation by L.A.A. from the West Ivory Coast. Voucher specimens counting in Optiscint HiSafe (LKB Pharmacia). are deposited at the Conservatoire et Jardin Botaniques The mean results, obtained as counts per mm @pm), de l’Universite d’Abidjan, Ivory Coast. Plant materials were air-dried, then freeze-dried and subsequently pow- were expressed as percentages of incorporation or growth inhibition. The sigmoid dose-response curve was then dered. Extractions were carried out in a Soxhlet apparatus and the extracts were evapd to dryness under red. linearized by probit analysis [25], with the aid of software pres. The 6 alkaloids were isolated and characterized as provided by IWONL (Gent), adapted by G. Timperman, and used to derive the I& values. described previously; dioncophylline B (1) [18], dionco2. Plasmodium falciparum (Kl strain, chloroquinepeltine A (2) [19] and dioncophylline A (3) [lo] were resistant). Small aliquots of test material were dissolved in obtained from T. peltatum and N-methyldioncophylline (4) [20], ancistrobrevine D (5) [21] and ancistrocladine (6) 20 ~1 EtOH and RPM1 1640 to give solns of 1 or 10 mg ml- l. Each was applied in a series of 12 4-fold [22] were isolated from A. ubbreuiutus. N-Methyldioncophylline A (4) also occurs in A. barteri [Bringmann, G., dilutions (final concns ranging from 50 to 1.19 x 10e5 pg ml- ‘). Chloroquine diphosphate (Sigma) was Schneider, C. and Aki Assi, L., unpublished results] and ancistrocladine (6) is widespread (A. abbreuiutus, A. hamu- used in the same expts as a ref. and also applied in 12 4fold dilutions (final concns 2 x l0-5-1.19x lo-l2 M). tus, A. heyneanus, A. tectorius and other Ancistrocladw Each compound was tested in duplicate. species) [ 111. 1464 G. FRANCOBet nl. 7. Francois, G., Jaziri, M., Dochez, C., Laurent, A. and Testing was performed zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA in vitro based upon the method W&y, M. (1992) Physiol. Plant. 85, A47. of ref. [24] as modified in refs [26, 271. The parasites 8. Francois, G., Kanyinda, B., Dochez, C., W&y, M. and (50 ~1initial parasitaemia of 1%, haematocrit of 5%) were Vanhaelen, M. (1992) Planta M ed. 58 Suppl. 1, 634. incubated for 24 hr in microtitre plates (Nunc) in normal 9. Ruangrungsi, N., Wongpanich, V., Tantivatana, P., non-hypoxanthine depleted medium in the presence of Cowe, H. J., Cox, P. J., Funayama, S. and Cordell, 50~1 of test soln. Controls were set up in each test by G. A. (1985) 1. Nat. Prod. 48, 529. preparing 12-24 wells with non-parasitized RBCs and no 10. Bringmann, G., Riibenacker, M., Jansen, J. R. and test soln and 12-24 wells with parasitized RBCs and no Scheutzow, D. (1990) Tetra~dron Letters 31, 639. test compound. Thereafter, 5 ~1 of 3H-hy~xanthine 11. B~n~ann, G. (1986) in The Alkaloids, Vol. 29 (Ame~ham) was added (0.2 &i well-‘) and the parasites (Brossi, A., ed.), p. 141. Academic Press, New York. incubated for a further period of 18 hr. Each welt was 12. Bringmann, G., Pokorny, F., Stablein, M., harvested with a Celf Harvester (Skatron). Glass fibre Govindachari, T. R., Ahneida, M. R. and Ketkar, S. filters were dried briefly. Radioactivity was measured by M. (1991) Planta M ed. 57 Suppl. 2, 98. liquid scintillation counting in Ecoscint (National Dia13. Sharma, S. C., Shukla, Y. N. and Tandon, J. S. (1975) gnostics). Phytochemistry 14, 578. The results obtained as cpm were expressed as percent14. Bringmann, G., Akb Assi, L., Rilbenacker, M., ages of incorporation or growth inhibition. The doseAmmermann, E. and Lorenz, G. (1992) D. 0. S. DE related part of the sigmoidal curves obtained were line41 17 080.6, Patent Request of 25.051991, European arized by regression analysis with the aid of software Patent 29. 4. 1992. developed by S. Grant (LSHTM) and used to calculate 15. Grimm, C., Proksch, P., Gramatzki, S. and the ICsO values. Bringmann, G. (1992) Planta M ed. 58 Suppl. 1,630. G., Gramat~i, S., Grimm, C. and 16. Brin~ann, ~Ck~wle~ge~enr~The authors wish to thank Mrs C. Proksch, P. (1992) Phytoc~mistry 31, 1. Van Overmeir and Mr L. Hendrix for their skilful tech17. Bringmann, G., Pokorny, F., Reuscher, H., Lisch, D. nical assistance and Mr R. Zagst for a sample of ancistroand Akt Assi, L. (1990) Planta M ed. 56, 496. brevine D. Financial support by the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 251 18. Bringmann, G., Riibenacker, M., Geuder, T. and Akt Assi, L. (1991) Phytochemistry 30, 3845, ‘tikologie, physiologie und Biochemie pflanzlicher und 19. Bringmann, G., Riibenacker, M., Vogt, P., Busse, H., tier&her Leistung unter StreD’), Wellcome Trust, AM Assi, L., Peters, K. and v. Schnering, H. G. (1991) Medical Research Council and British Public Health Phytochemistry 30, 1691. Laboratory Services (D.C.W.) is gratefully acknowledged. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJI 20. Bringmann, G., Lisch, D., Reuscher, H., Ake Assi, L. and Gunther, K. (1991) Phytochemistry 30, 1307. 21. Bringmann, G., Weirich, R., Lisch, D. and AkC Assi, REFERENCES L. (1992) Planta M ed. 58 Suppl. 1, 703. 1. B~n~ann, G., Zagst, R., SchPffer, M., Hatlock, Y. F., 22. B~ngmann, G., Zags& R., Reuscher, H. and Aki: Assi, Cardellina II, J. H. and Boyd, M. R. (1993) Angew. L. (1992) Phytochemis~y 31,4011. Chem. (submitted). 23. Trager, W. and Jensen, J. B. (1976) Science zyxwvutsrqponmlkjih 193,673. 2. Peters, W. (1987) in Chemotherapy and Drug Resist- 24. Desjardins, R. E., Canfield, C. J., Haynes, J. D. and ance in M alaria, Vol. 2, p. 543. Academic Press, Chulay, J. D. (1979) Antimicrob. Agents Chemother. London. 16, 710. 3. World Health Organization (1989) W eekly Epidem25. Diem, K. (1962) in Scientific Tab&s. Documenta iol. Record 1990, 24. Geigy , Sixth Edition, p. 54. J. R. Geigy S. A., Basle, 4. Bjiirkmann, A. (1991) in: M alaria- W aiting for the Switzerland. Vaccine (Targett, G. A. T., ed.), p. 105. John Wiley, 26. O’Neill, M. J., Bray, D. H., Boardman, P., Phillipson, Chichester. J. D. and Warhurst, D. C. (1985) Planta M ed. 5X, 394. 5. Phillipson, J. D. and Wright, C. W. (1991) Planta 27. Ekong, R. M., Kirby, G. C., G’Neill, M. J., Pate& G., M ed. 57, Suppl. 1, S53. Phillipson, J. D. and W~hu~t, D. C. (1990) Biochem. 6. Klayman, D. L. (1989) Nat~ai history IO, 18. Pharmacof. 40,297.