Phytochemistry 62 (2003) 623–629 www.elsevier.com/locate/phytochem Indolomonoterpenic alkaloids from Strychnos icaja roots Geneviève Philippea, Patrick De Molb, Monique Zèches-Hanrotc, Jean-Marc Nuzillardc, Monique-Hanrot Titsa, Luc Angenota, Michel Frédéricha,b,* a University of Liège, Natural and Synthetic Drugs Research Center, Laboratoire de Pharmacognosie, Avenue de l’Hôpital 1, B36, B-4000 Liège, Belgium b University of Liège, Laboratoire de Microbiologie Médicale, Liège, Belgium c Laboratoire de Pharmacognosie, UMR 6013 CNRS, Bâtiment 18, Moulin de la Housse, F-51687 Reims cedex 2331, France Received 5 September 2002; received in revised form 18 October 2002; accepted 22 October 2002 Abstract In the course of our search for new antiplasmodial alkaloids from Strychnos icaja, we have isolated five alkaloids: three monomers, protostrychnine and genostrychnine, previously described in Strychnos nux-vomica, pseudostrychnine, already found in the leaves of the plant, a new bisindolic alkaloid, named strychnogucine C, and the first naturally occurring trimeric indolomonoterpenic alkaloid: strychnohexamine. This latter trimeric alkaloid presented an antiplasmodial activity against the FCA Plasmodium falciparum line near 1 mM. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Loganiaceae; Strychnos icaja; Indolomonoterpenic alkaloid; Strychnogucine C; Protostrychnine; Strychnohexamine; Plasmodium falciparum 1. Introduction Malaria is the major parasitic infection in many tropical and subtropical regions, leading to more than one million deaths (principally young African children) out of 400 million cases each year (Greenwood and Mutabingwa, 2002). More than half of the world’s population live in areas where they remain at risk of malaria infection. During last years, the situation has worsened in many ways, mainly due to malarial parasites becoming increasingly resistant to several antimalarial drugs. Furthermore, the control of malaria is more and more complicated by the parallel spread of resistance of the mosquito vector to currently available insecticides. Discovering new drugs in this field is therefore a health priority. In this context, the search for new antimalarial compounds from tropical plants could be an economically affordable solution. * Corresponding author. Tel.: +32-4-366-43-38; fax: +32-4-36643-32. E-mail address: m.frederich@ulg.ac.be (M. Frédérich). Strychnos icaja Baill. (Loganiaceae) is a tropical shrub common in tropical forest of central Africa (Congo, Rwanda, Cameroon, . . .). This Strychnos species is mainly used by local populations as an arrow or ordeal poison, but has also been used by Pygmies tribes from Cameroon to treat persistent malaria (Neuwinger, 1996). S. icaja is well known for its toxicity, but the alkaloids responsible for this convulsant effect are mainly strychnine and 12-hydroxystrychnine (Sandberg and Kristianson, 1970). In previous works, some dimeric alkaloids derivated from sungucine (Kambu et al., 1980) and possessing potent and selective (Plasmodium towards human cells) antiplasmodial properties have been isolated from S. icaja (Frédérich et al., 2000, 2001). We report herein the isolation and characterization of strychnogucine C (3), a new bisindolomonoterpenic alkaloid, and of protostrychnine (1), genostrychnine and pseudostrychnine (2). In the course of this work we have also isolated strychnohexamine (4), an original trimeric indolomonoterpenoid alkaloid. Antiplasmodial activities of these alkaloids have then been investigated here for the first time. 0031-9422/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0031-9422(02)00612-X 624 G. Philippe et al. / Phytochemistry 62 (2003) 623–629 2. Results and discussion Starting from an ethyl acetate extract of S. icaja root bark, sequential liquid–liquid extraction and chromatographic separations led to the isolation of the five indolomonoterpenic alkaloids. Alkaloids 1 and 2 were identified as protostrychnine (1) and pseudostrychnine (3-hydroxystrychnine, 2), based on comparison of their ESI–MS, UV and 13C NMR spectral data with literature (Baser et al., 1979; Bisset et al., 1973; Verpoorte et al., 1977; Baser and Bisset, 1982) and by a TLC comparison with an authentic sample for 2. An other alkaloid has been identified as genostrychnine (strychnine-N-oxide) after UV, IR, ESI–MS, and chromatographic comparison with a reference sample and literature (Bisset, 1976). These three compounds (genostrychnine, 1 and 2) were already known to be present in different species of Strychnos, such as S. nux-vomica and S. ignatii (Baser et al., 1979; Bisset, 1976; Datta and Bisset, 1990). Pseudostrychnine (2) has also been previously described in the leaves of S. icaja (Bisset et al., 1973). The identity of 1 and 2 has also been confirmed by 1H NMR and COSY studies of these alkaloids. These data, which were not completely available in the literature, were listed in Table 1. The discovery of protostrychnine in S. icaja is very important from a biosynthetic point of view. Effectively, protostrychnine is a key intermediate in the strychnine biosynthetic pathway hypothesis proposed by Heimberger and Scott (1973), being the last intermediate before the formation of strychnine and isostrychnine. Up to then, protostrychnine (and isostrychnine) had only been described in S. nux-vomica and S. ignatii. The discovery of protostrychnine and, recently, of isostrychnine (Frédérich et al., 2000) in S. icaja, the African strychnine- containing Strychnos species, confirms definitely this biosynthetic hypothesis. Alkaloid 4, named strychnohexamine, exhibiting a strong and fleeting purple coloration with cerium (IV) sulphate reagent, was the first naturally occurring trisindolic monoterpenoid alkaloid which has been isolated to date. Its structure has been described recently (Philippe et al., 2002). Although another trimeric indolomonoterpenic alkaloid has been previously obtained by biotechnology methods, strychnohexamine is the first one isolated directly from a plant source. The former trimeric compound was a derivative of tabersonine and was formed in micro traces (less than 1 mg) after incubation of tabersonine with an enzyme mixture obtained from leaves of mature Catharantus roseus plants (Schübel et al., 1989). Compound 3 gave a blue fluorescence on TLC plate after pulverisation with cerium (IV) sulfate reagent. The UV spectrum was very close to that of strychnine and strychnogucine A (Frédérich et al., 2001), showing an absorption maximum at 254 nm. Based on HR–ESI– MS, C42H42N4O3 was determined as its molecular formula, with a [MH+] at m/z 651,3210. This molecular formula was identical to that of strychnogucine A, and as for strychnogucine C, the ESI–MS fragmentation (daughters peaks) displayed two ions at m/z 335 and 317, corresponding to the masses of strychnine and deoxyisostrychnine, respectively. Compared by TLC with other alkaloids possessing this molecular weight, including strychnogucine A, compound 3 did not present any similarities. NMR spectral data of compound 3 were listed in Table 2. The broad band decoupled 13C NMR spectrum showed 42 carbon signals which were sorted by HSQC and HMBC techniques as one methyl, eight methylenes, 23 methines and 10 quaternary carbons, including two carbons from a carbonyl group. Among the 23 methines, 625 G. Philippe et al. / Phytochemistry 62 (2003) 623–629 Table 1 1 H NMR spectral data of compounds 1 and 2 (recorded respectively at 500 and 400 MHz) in CDCl3 Position Protostrychnine (1) 1 H chemical shiftsa 2 3 5a 5b 6a 6b 9 10 11 12 14a 14b 15 16 17 18a 18b 19 21a 21b 23a 23b a 3.82 3.70 2.90 3.18 1.96 2.53 7.38 7.18 7.29 7.97 1.73 2.18 3.09 1.40 3.97 4.13 4.26 5.79 3.19 3.60 2.39 3.25 (d, 11.2) (m) (m) (m) (m) (m) (d, 7.5) (dt, 7.5 and 0.6) (td, 1.0 and 8.0) (d, 8.0) (dd, 2.5 and 11.3) (dt, 3.3 and 13.9) (d, 3.0) (dt, 3.4 and 10.6) (t, 9.1) (ddd, 1.6; 6.2 and 13.0) (dd, 7.2 and 13.0) (t, 6.4) (m) (d, 17.5) (d, 16.5) (dd, 8.5 and 16.5) Pseudostrychnine (2) COSY H/H correlations 1 H chemical shiftsa 16 14a, 14b, 15, 21b 5b, 6a, 6b 5a, 6a, 6b 5a, 5b, 6b 5a, 5b, 6a 10, 11 9, 11, 12 9, 10, 12 10, 11 3, 14b, 15 3, 14a, 15 3, 14a, 14b, 16, 21b 2, 15, 17 16, 23a, 23b 18b, 19, 21a 18a, 19, 21a 18a, 18b, 21a, 21b 18a, 18b, 21b 3, 15, 19, 21a 17, 23b 17, 23a 3.82 (m) 16 2.85 (m) 3.83 (m) 1.80 (m) 2.26 (m) 7.89 (d, 7.8) 7.05 (m) 7.23 (m) 8.11 (m) 1.80 (d, 13.2) 2.26 (m) 3.28 (m) 1.37 (d, 10.2) 4.27 (m) 4.05 (m) 4.13 (m) 5.91 (t, 7.2) 2.94 (d, 14.9) 3.73 (d, 15.2) 2.66 (dd, 3.4 and 17.4) 3.12 (m) 5b, 6a, 6b 5a 5a, 6b 5a, 6a 10 9, 11 10, 12 11 14b, 15 14a, 15 14a, 14b, 16 2, 15, 17 16, 23a, 23b 19 19 18a, 18b 21b 21a 17, 23b 17, 23a COSY H/H correlations Multiplicities and coupling constants in Hz are in parentheses. 12 were in the aromatic region: the COSY spectrum showed eight aromatic protons, as expected from two indole moieties, and four methines, two of which belonging to two ethylidenic side chains. As in strychnogucine A, 1H and 13C chemical shifts of the portion A of the molecule were quasi-identical to those of strychnine. Among characteristic signals of strychnine, we had a highly shielded shift for H-16, due to the configuration H-16a of strychnine, and a deshielded shift for H-12, due to the influence of the carbonyl C-22. The only differences with strychnine were observed at the 23-position: the C-23 chemical shift was notably shielded (51.9 ppm instead of 42.3 ppm) and only one proton corresponded to H-23, indicating the implication of C-23 in a supplementary C–C bond. These data were very close to these of strychnogucine A. Nevertheless, in the portion B of the molecule, some differences were observed in the 1H and 13C NMR spectra between the two compounds. In the portion B of the alkaloid, H-170 was correlated to a multiplet at
2.77 and to a doublet at
5.98, assigned respectively to H-160 and H-230 . The presence of H-16 and the deshielded shift value for H-230 (methine proton) indicated the presence of a 170 -230 double bond instead of a 160 -170 double bond as in strychnogucine A. The linkage C-23/C-50 between the two portions of 3 was confirmed from the HMBC spectrum (H-50 and the two H-60 were correlated to C-23, while H-17 was correlated to C-50 ), from the COSY spectrum (correlation between H-50 and H-23), and from the correlations between H-23 and H-50 , H-17 and H-60 a, H-17 and H-30 in the ROESY spectrum (Table 2 and Fig. 1). The stereochemistry of 3 was then considered. The configurations H-15a, H-3a, H-2b, 7R, H-150 a, H-30 a, H-20 b and 70 R were those commonly accepted from biogenetic considerations (Klyne and Buckingham, 1974). This configuration is confirmed by the presence, in the CD spectra of a positive Cotton effect near 240 nm. This positive effect is indicative of a 2b,7b configuration in alkaloids with an indoline moiety (Klyne et al., 1965, 1968) and is observed, i.e., in the CD spectra of the monomers protostrychnine, pseudostrychnine (see Experimental) and isostrychnine (Frédérich et al., 2000), but also in the CD spectra of the dimers strychnogucine A, B (Frédérich et al., 2001) and sungucine (Frédérich et al., 2000), whose absolute configuration has been determined by X-ray analysis. In addition, this 2b,7b configuration is always biogenetically linked with a 3a,15a configuration (Klyne and Buckingham, 1974). The H-16a (16R) and H-160 b (160 S) configurations were proposed after comparison of chemical shifts of C2, C-6, C-14, C-7, C-3, C-16, C-21 and C-20 , C-60 , C-140 , C-70 , C-30 , C-160 , C-210 with the published values for retuline, isoretuline (Massiot et al., 1988), strychnine (Verpoorte, 1980), and sungucine (Frédérich et al., 2000), and after observation of the coupling constants between H-2 and H-16 (10.5 Hz, antiperiplanar) and between H-20 and H-160 (6.8 Hz, periplanar). The deshielded value of H-16 (1.23 ppm) and the more 626 G. Philippe et al. / Phytochemistry 62 (2003) 623–629 Table 2 1 H and 13C NMR spectral data of compound 3 (recorded at 500/125 MHz) in CDCl3 Position 1 H NMRa 2 3 5a 5b 6a 6b 7 8 9 10 11 12 13 14a 14b 15 16 17 18a 18b 19 20 21a 21b 22 23 20 30 50 60 a 60 b 70 80 90 100 110 120 130 140 a 140 b 150 160 170 180 190 200 210 a 210 b 220 230 3.98 3.90 2.89 3.17 1.85 1.87 (d, 10.5) (m) (m) (m) (m) (m) 1.52 2.40 3.22 1.23 4.24 4.12 4.16 5.95 3, 14b, 15 3, 14a, 15 14ab, 16, 19 2, 15, 17 16,23 18b, 19 18a, 19 15, 18ab, 21b 2.72 (d, 14.5) 3.69 (d, 14.5) 21b 19, 21a 2.88 4.30 3.52 4.04 1.95 2.21 17, 50 160 140 ab 23, 60 ab 50 , 60 b 50 , 60 a 7.18* (m) 7.08 (m) 7.24 (m) 8.20 (d, 8.1) 100 90 , 110 100 , 120 110 1.75 1.78 2.68 2.77 6.87 1.65 5.28 140 b, 150 140 a, 150 140 ab, 160 20 , 150 , 170 , 230 160 , 230 190 , 210 b 150 , 180 , 210 b (m) (m) (m) (m) (dd, 6.5 and 9.8) (d, 6.0) (d, 6.0) 2 3 5 59.7 60 50 6ab 6ab, 14b, 21a 6a, 21ab 6 42.9 2 7 8 9 10 11 12 13 14 51.7 132.6 122.2 124.4 128.5 116.5 142.1 26.7 2, 6ab, 9, 14b 2, 12, 13 11 12 9 10 2, 9, 10, 11, 12 15 16 17 18 31.9 49.5 79.1 65.1 14b, 21a 2, 14b 2, 18ab, 23, 50 19 20 21 128.4 139.3 52.5 18ab, 21ab 14a, 18ab, 21ab 22 23 20 30 50 60 171 51.9 64.9 65.1 61.7 39.5 23 70 80 90 100 110 120 130 140 51.4 134.2 122.2 124.1 128.5 116.2 141.5 22.5 20 , 60 ab, 90 , 140 ab 30 , 60 a, 120 , 130 110 120 90 100 50 , 90 , 100 , 110 , 120 150 160 170 180 190 200 210 31 40.1 144 12.9 119.3 142.6 49.7 20 , 30 , 140 a, 190 , 210 a 170 , 230 220 230 162 122.8 16 14ab 5b, 6ab 5a, 6ab 6b, 5ab 6a, 5ab 10 9, 11 10, 12 11 (m) (d, 6.8) (d, 4.3) (m) (t, 12) (dd, 12 and 5) HMBCb C!H correlations Position 7.17* (m) 7.06 (m) 7.20 (m) 8.02 (d, 8.0) (d, 14.0) (dt, 14.0 and 4.6) (m) (m) (m) (dd, 12.5 and 5.7) (dd, 12.5 and 6.9) (m) C NMR COSY H/H correlations 3.13 (d, 16.8) 3.62 (d, 16.8) 210 b 190 , 210 a 5.99 (d, 9.8) 160 , 170 13 30 , 60 ab, 170 20 , 60 ab, 210 a 60 a, 210 a, 23 20 , 50 , 23 190 180 , 210 a 50 , 190 *These values could be interchanged. a Multiplicities and coupling constants in Hz are in parentheses. b Correlations from C to the indicated hydrogens. shielded value of H-160 (2.72 ppm) were in agreement with these configurations. The configuration of H-50 was suggested as b (H-50 R) after comparison of the H-60 ab multiplicities and the C-50 , C-60 , H-50 , and H-60 chemical shifts with values for sungucine and isosungucine (Frédérich et al., 2000). This configuration was corroborated by the ROESY coupling between H-23 and H-50 (Fig. 1) when no coupling was observed between H-50 (b) and H-30 (a). 627 G. Philippe et al. / Phytochemistry 62 (2003) 623–629 Table 3 In vitro antiplasmodial activity of compounds 3 and 4, quinine, and chloroquine on FCA line of Plasmodium falciparum Compound IC50 (mM) IC90 (mM) na Sungucine Strychnogucine A Strychnogucine B Strychnogucine C (3) Strychnohexamine (4) Bisnordihydrotoxiferine Chloroquine Quinine 7.816 1.137 2.310 0.304 0.617 0.067 16.057 0.764 1.097 0.099 2.796 1.078 0.02 0.002 0.269 0.006 26.256 6.980 3.785 33.535 4.296 16.409 0.119 1.913 3 2 2 2 3 3 9 3 a Fig. 1. Significant ROESY correlations for compound 4. The H-17a (17R) and H-23b (23S) configurations were proposed after comparison of chemical shifts of C15 through C-23 and C-19 through C-21 with analogous data for strychnine (Sandberg et al., 1969) and strychnogucine A (Frédérich et al., 2001). These hypotheses were corroborated by the ROESY couplings observed between H-17 (a) and H-16 (a), H-17 (a) and H-30 (a), H-23 (b) and H-2 (b). Consequently, alkaloid 3 must have the following stereochemistry: H-15a (15R), H-3a (3S), H-2b (2S), 7R, H-16a (16R), H-17a (17R), H-23b (23S), H-50 b (50 R), H-150 a (150 S), H-160 b (160 S), H-30 a (30 S), H-20 b (20 S), and 70 R. The in vitro antiplasmodial activities of isolated compounds have then been determined against the FCA chloroquine-sensitive strain of Plasmodium falciparum in comparison to chloroquine, quinine and other dimeric alkaloids from S. icaja (see Table 3). Strychnogucine C presented an IC50 between 10 and 20 mM, which was notably less active than other sungucine type alkaloids. This is a confirmation that the presence of a cyclisation of the ring G in the upper part of the molecule (portion A), as in strychnogucine A, or the presence of a 170 -230 double bond, as in sungucine, has a negative impact on antiplasmodial activity. The most active compound from this class of alkaloids (quasi-symmetric dimers possessing a 50 -23 link) remains strychnogucine B, possessing a lower strychnine moiety and an upper isostrychnine II moiety (Frédérich et al., 2001). If the n=Number of experiments. antiplasmodial selectivity of these sungucine-type alkaloids towards human cells has been previously demonstrated (Frédérich et al., 2001), it will be nevertheless important to check, in the future, if these alkaloids are really devoid of any convulsant activity. The monomers have also been tested against Plasmodium but were devoid of any antiplasmodial activity, as expected and previously observed for other monomers (Frédérich et al., 1999). On the other hand, strychnohexamine (4) presented an interesting antiplasmodial activity with an IC50 near 1 mM, which was about two times more potent than bisnordihydrotoxiferine. 3. Experimental 3.1. General UV spectra were recorded on a Kontron Uvikon spectrophotometer, and the IR spectra were recorded on a Perkin-Elmer 1750 FTIR spectrometer. NMR spectra were recorded in CDCl3 on a Bruker 500 MHz NMR spectrometer, with TMS as an internal reference. CD curves were determined on a Jobin Yvon CD6 dichrograph. ESIMS were obtained with a VG Autospec-Q (VG Analytical, Manchester, Liquid s/ms, Cs+, 20 keV, resolution > 5000) apparatus. Analytical TLC was performed on precoated Si gel F254 (Merck, 1.05735) plates. After development, the dried plates were examined under short-wave (254 nm) or longwave (366 nm) UV light and sprayed with one of the following reagents: (a) Dragendorff’s reagent, (b) 1% ceric sulfate in 10% sulfuric acid. LiChroprep Si 60 (15–25 mm, Merck 9336) was used for column chromatography. Si gel 60 PF 254 (Art.1.07747, Merck) was used for purification of alkaloids by preparative TLC (1.25 mm thick, 2040 cm Si gel plates). All solvents used were analytical grade (Merck). The authentic sample of pseudostrychnine has been obtained from the late Professor Bisset, Chelsea College, University of London. The original sample of genostrychnine came from the collections of the laboratory, and had been 628 G. Philippe et al. / Phytochemistry 62 (2003) 623–629 obtained from strychnine by hemi-synthesis (H2O2 treatment). 3.2. Plant material The roots of S. icaja were collected near Kasongo-Lunda (Congo-Zaire). A voucher specimen of the plant (Duvigneaud H787) has been deposited in the herbarium of the Pharmaceutical Institute, at Liège and in the herbarium of the Belgian National Botanical Garden, at Meise. to dryness to yield, from the first fraction, 29 mg of protostrychnine (1) and, from the second, 35 mg of genostrychnine. Further purifications were not necessary. Pseudostrychnine (2) has been isolated from pH 3 extract. This extract has been fractionated by MPLC on Merck LiChroPrep Si 60 using the same mobile phase as for pH 8 extract. The first fraction obtained was pseudostrychnine (79 mg) and was followed by known alkaloids as vomicine, icajine, sungucine, strychnine. 3.4. Protostrychnine (1) 3.3. Extraction and isolation The roots of S. icaja (500 g) were macerated with 300 ml of EtOAc–ethanol–NH4OH (96:3:1) and then percolated with EtOAc then with MeOH until complete extraction of alkaloids. The extract was concentrated under reduced pressure below 60  C to yield 43 g of dry extract and then dissolved in EtOAc and extracted with 4% HOAc. The resulting acidic (pH 3) solution was extracted by CH2Cl2, then basified to pH 8 with Na2CO3 and repeatedly extracted with CH2Cl2. The same extraction was made at pH 10 (alkalinization with NH4OH). The CH2Cl2 extracts obtained were dried over Na2SO4 and concentrated to yield crude alkaloid extracts (respectively 5, 28 and 1 g, at pH 3, 8 and 10). The pH 8 extract was fractionated by medium pressure liquid chromatography (MPLC) on 180 g Merck LichroPrep Si 60 (40–63 mm, Merck 9336) with a gradient of CH2Cl2/MeOH mixtures (0 to 10% MeOH), to give fractions I-XXVI, detected by TLC (EtOAC/2PrOH/NH4OH, 80:15:5), as previously described (Frédérich et al., 2001). Strychnohexamine (4) was present in weak amounts in fractions XVIII to XXII. The purification of 4 (10 mg) has been conducted by MPLC on Merck LiChroprep RP-8 (25–40 mm, 8 g) with MeOH– MeCN–H2O (3:2:1) and finally on a Sephadex LH20 (20 g, Pharmacia Biotech) column with MeOH as mobile phase. Strychnogucine C (3) was present in fractions XVI and XVII (6570 to 8039 ml) along with strychnine, strychnogucine A and bisnordihydrotoxiferine. The four compounds were separated by MPLC on Merck LiChroprep RP8 (25–40 mm, 8 g) with MeOH–MeCN– H2O (3:2:1); strychnine, 240 to 290 ml; strychnogucine A (18 mg) (1), 300 to 360 ml, strychnogucine C, 400 to 450 ml and bisnordihydrotoxiferine, 550 to 700 ml. Protostrychnine (1) and genostrychnine were isolated from the pH 10 extract. This extract has been fractionated by HSCCC. CHCl3, MeOH and water were thoroughly equilibrated in suitable proportions (7:13:8) and the two phases separated. The lower organic phase was used as the stationary phase, and the upper aqueous phase as the mobile phase. We worked in the ascending mode. Compound 1 and genostrychnine were found alone in two successive fractions. These were concentrated under reduced pressure and then evaporated Amorphous buff colored solid. On TLC, gave a pale orange-pink coloration after spraying with cerium sulfate reagent and an orange-red coloration with ferric chloride reagent. The UV, IR, MS and 13C NMR data were in agreement with those from the literature (Baser et al., 1979). CDMeOH, enm : e216 +6.0, e228 0.2, e252 +2.6, e274 +1.1, e292+1.5; 1H NMR data are given in Table 2. 3.5. Genostrychnine Genostrychnine was identified by comparison with an authentic sample (TLC, UV, IR, ESI–MS). 3.6. Pseudostrychnine (2) White needle-crystallized powder. On TLC, gave a fleeting blue coloration after spraying with cerium sulfate reagent and an orange coloration with ferric chloride reagent. CDMeOH, enm : e205 17.48, e214 9.42, e225 20.22, e259 3.72; e277 0.97 FT–IR nmax (C2Cl4) cm1: 3593 (OH), 2952, 1681 (C¼O lactam), 1596, 1477, 1389, 1290. The UV, MS and 13C NMR data (Baser et al., 1979) were in agreement with those from the literature (Verpoorte et al., 1977). 1H NMR data are given in Table 2. 3.7. Strychnogucine C (3) White-yellowish amorphous powder. On TLC, gave a blue fluorescence at 366 nm after spraying with cerium sulfate reagent; UV lmax nm (log e) (MeOH): 208 (4,63), 255 (4,14), 283 (3,93); FT–IR umax (KBr) cm1: 3437, 2926, 1665, 1594, 1482, 1461, 1385, 1287, 1095, 1051, 819, 758, 566, 425; CDMeOH, enm: e205 15.6, e240 0.39, e275 1,56, e298 0,39 ; 1H and 13C NMR data are given in Table 1; ESI–MS m/z 651 [MH+] (90), 335 (50), 317 (100), 274 (5), 134 (20) (daughters); HRESIMS m/z [MH+] 651.3213 (calcd for C42H43N4O3, 651.3335). 3.8. Strychnohexamine (4) White-yellowish amorphous powder. On TLC gave a fleeting purple coloration after spraying with cerium G. Philippe et al. / Phytochemistry 62 (2003) 623–629 sulfate reagent; ESI–MS: m/z 869 (MH+) (100), 855, 674, 629, 583, 546, 510, 438, 391, 345, 258, 247, 234, 231, 222, 208, 194, 144, 122; HRESIMS: m/z 868.48057, (calcd for C59H60N6O1, 868.4828); FT–IR umax (KBr) cm1: 3429, 2926, 1664 (C¼O), 1597 (C¼C), 1486, 1418, 1383, 1261, 1095, 801, 754, 617; UV lmax nm (log e) (MeOH): 210 (3.55), 293 (3.25), 320 (2.99); CDMeOH, enm : e240 +2, e274 16.6, e294 +16.6, e318 33.3. NMR data have been previously described (Philippe et al., 2002). 3.9. Antiplasmodial assays Continuous in vitro cultures of asexual erythrocytic stages of the four P. falciparum strains were maintained following the procedure of Trager and Jensen (1976) and as described previously (Frédérich et al., 2000). Chloroquine diphosphate (Sigma C6628), mefloquine HCl (Roche) and quinine base (Aldrich 14590-4) were used as antimalarial references. Each test sample was applied in a series of eight four-fold dilutions (final concentrations ranging from 20 mg/ml to 0.0012 mg/ml) and was tested in duplicate. Parasite growth was estimated by determination of [3H]hypoxanthine incorporation as described by Desjardins et al. (1979) and modified by Mirovsky et al. (1990). The Student t-test was used to test the significance of differences between results obtained for different samples. Statistical significance was set at P40.05. Acknowledgements The authors wish to thank M.J.C. Van Heugen for providing the MS data, Professor P. Colson (Chimie Macromoléculaire, Université de Liège) for the CD spectra and Professor J. Boniver (Anatomie et Cytologie Pathologique, Université de Liège) for liquid scintillation measurements. This research was supported by the Belgian National Fund for Scientific Research (FNRS) [grant No. 3453201 and fellowship for MF]. References Baser, K.H.C., Bisset, N.G., 1982. Alkaloids of Sri-Lankan Strychnos nux-vomica. Phytochemistry 21, 1423–1429. Baser, K.H.C., Bisset, N.G., Hylands, P.J., 1979. Protostrychnine, a new alkaloid from Strychnos nux-vomica. Phytochemistry 18, 512–514. Bisset, N.G., 1976. The Asian species of Strychnos. Part IV. The alkaloids. Lloydia 39, 263–315. 629 Bisset, N.G., Das, B., Parello, J., 1973. Alkaloids from the leaves of Strychnos icaja Baill. Tetrahedron 29, 4137–4148. Datta, B., Bisset, N.G., 1990. Alkaloids of Strychnos ignatii. Planta Med. 56, 133. Desjardins, R.E., Canfield, C.J., Haynes, J.D., Chulay, J.D., 1979. Quantitative assessment of antimalarial activity in vitro by a semiautomated microdilution technique. Antimicrob. Agents Chemother. 16, 710–718. Frédérich, M., Hayette, M.P., Tits, M., De Mol, P., Angenot, L., 1999. In vitro activities of Strychnos alkaloids and extracts against Plasmodium falciparum. Antimicrob. Agents Chemother. 43, 2328–2331. Frédérich, M., Depauw-Gillet, M.C., Llabres, G., Tits, M., Hayette, M.P., Brandt, V., Penelle, J., De Mol, P., Angenot, L., 2000. New antimalarial and cytotoxic sungucine derivatives from Strychnos icaja roots. Planta Med. 66, 262–269. Frédérich, M., De Pauw, M., Prosperi, C., Tits, M., Brandt, V., Penelle, J., Hayette, M., De Mol, P., Angenot, L., 2001. Strychnogucines A and B, two new antiplasmodial bisindole alkaloids from Strychnos icaja. J. Nat. Prod. 64, 12–16. Greenwood, B., Mutabingwa, T., 2002. Malaria in 2002. Nature 415, 670–672. Heimberger, S.I., Scott, A.I., 1973. Biosynthesis of strychnine. J. Chem. Soc. Chem. Commun., 217-218. Kambu, K., Coune, C., Angenot, L., 1980. Nouveaux alcaloı̈des des racines du Strychnos icaja. Planta Med. 37, 161–164. Klyne, W., Buckingham, J., 1974. Atlas of Stereochemistry: Absolute Configurations of Organic Molecules. Chapman & Hall, London. Klyne, W., Swan, R., Bycroft, B., Schmid, H., 1965. Ermittlung der absoluten Konfiguration von Indolinalkaloiden durch Vergleich der optischen Rotationdispersionen ihrer N(8a)-Acylderivative. Helv. Chim. Acta 49, 833–841. Klyne, W., Swan, R., Gorman, A., Schmid, H., 1968. Optische Rotationsdispersion von Indolinalkaloiden mit Ketogruppen. Helv. Chim. Acta 51, 1168–1184. Massiot, G., Massoussa, B., Jacquier, M.J., Thépénier, P., Le Men Olivier, L., Delaude, C., Verpoorte, R., 1988. Alkaloids from roots of Strychnos matopensis. Phytochemistry 27, 3293–3304. Mirovsky, P., Gay, F., Bustos, D., Mazier, D., Gentilini, M., 1990. Cloning of a fresh isolate of Plasmodium falciparum and drug sensitivity of the clones. Trans. Roy. Soc. Trop. Med. Hyg. 84, 511–515. Neuwinger, H.D., 1996. African Ethnobotany: Poisons and Drugs, Chemistry, Pharmacology, Toxicology. Chapman & Hall, London. Philippe, G., Prost, E., Nuzillard, J.M., Zéches-Hanrot, M., Tits, M., Angenot, L., Frédérich, M., 2002. Strychnohexamine from Strychnos icaja, a naturally occurring trimeric indolomonoterpenic alkaloid. Tetrahedron Lett. 43, 3387–3390. Sandberg, F., Kristianson, K., 1970. A comparative study of the convulsant effects of Strychnos alkaloids. Acta Pharm. Suec. 7, 329–336. Sandberg, F., Lunell, E., Ryrberg, K.J., 1969. Pharmacological and phytochemical investigations of African Strychnos species. Acta Pharm. Suec. 6, 79–102. Schübel, H., Fahn, W., Stöckigt, J., 1989. Formation of the first trimeric monoterpenoid indole alkaloids. Helv. Chim. Acta 72, 147– 150. Trager, W., Jensen, J.B., 1976. Human malaria parasites in continuous culture. Science 193, 673–675. Verpoorte, R., 1980. Assignment of 13C NMR spectra of strychnine and brucine. J. Pharm. Sci. 69, 865–867. Verpoorte, R., Hylands, P.J., Bisset, N.G., 1977. Carbon-13 NMR spectroscopy of some Strychnos alkaloids. Org. Magn. Res. 9, 567– 571.