NPC Natural Product Communications Triterpenoid Acids and Lactones from the Leaves of Fadogia tetraquetra var. tetraquetra (Rubiaceae) 2011 Vol. 6 No. 11 1573 - 1576 Dulcie A. Mulhollanda,b, Abdelhafeez M.A. Mohammeda,c*, Philip H. Coombesa, Shafiul Haqued, Leena L. Pohjalad,e, Päivi S.M. Tammelad and Neil R. Croucha,f a School of Chemistry, University of KwaZulu-Natal, Westville Campus, Private Bag X54001, Durban 4000, South Africa b Division of Chemical Sciences, Faculty of Health and Medical Sciences, University of Surrey, Guildford, Surrey GU2 7XH, United Kingdom c Department of Chemistry, Alzaiem Alazhari University, PO Box 1432, Khartoum, Sudan d Centre for Drug Research, and eDivision of Pharmaceutical Biology, Faculty of Pharmacy, PO Box 56, FIN-00014 University of Helsinki, Finland f Ethnobotany Unit, South African National Biodiversity Institute, PO Box 52099, Berea Road 4007, South Africa d.mulholland@surrey.ac.uk Received: May 24th, 2011; Accepted: July 27th, 2011 Four triterpenoids isolated from the leaves of Fadogia tetraquetra var. tetraquetra, 3-hydroxy-11,12-epoxyoleanan-28,13-olide (1), 3-hydroxyurs-11-en-28,13-olide (2), oleanolic acid (3), and ursolic acid (4), were evaluated for their antiviral and antibacterial properties. Compound 4 showed potent activity against the Semliki Forest virus with an IC50 of 14.7 µM, but was also found to be significantly cytotoxic (68% reduction in cell viability after 24 hours exposure at 50 µM) towards baby hamster kidney (BHK21) host cells. A viability assay on the mammalian human hepatocellular carcinoma (Huh-7) cell line showed no significant effects on intracellular ATP content after 48 hours exposure to compounds 1-4 at this concentration. Compound 4 also inhibited Staphylococcus aureus (MIC 12.5 µM), but was inactive against Enterobacter aerogenes, Escherichia coli, and Pseudomonas aeruginosa. Compounds 1-3 were inactive against all tested bacterial strains at 50 µM concentration. Keywords: Fadogia tetraquetra, Rubiaceae, 3-hydroxy-11,12-epoxyoleanan-28,13-olide, 3-hydroxyurs-11-en-28,13-olide, oleanolic acid, ursolic acid, antiviral, antibacterial, cytotoxic, Semliki Forest virus. The genus Fadogia Schweinf. (Rubiaceae) comprises some forty-five species in tropical Africa, with three, F. homblei De Wild., F. thamnus K.Schum, and F. tetraquetra K.Krause, found in southern Africa. The last occurs as two varieties, F. tetraquetra var. grandiflora (Robyns) Verdc. (syn. F. grandiflora Robyns), and F. tetraquetra var. tetraquetra (syn. F. mucronulata Robyns). The typical variety is the subject of this investigation; its distribution stretches northwards from Swaziland and the Limpopo and Mpumalanga Provinces of South Africa to Zimbabwe, Zambia, Mozambique, western Tanzania and the Democratic Republic of Congo [1]. Another regional genus member, F. homblei, (syn. F. monticola Robyns) has long been known [2] as a cause of the economically-important fatal poisoning syndrome in ruminants, known locally as gousiekte (‘quick disease’). Although a number of other Rubiaceae species from southern Africa, including, inter alia, Pavetta harborii S.Moore, P. schumanniana F.Hoffm. ex K.Schum., Pachystigma pygmaeum (Schltr.) Robyns and P. thamnus Robyns have also been implicated in gousiekte poisoning cases [3,4], the structure of pavettamine, the active principle involved, has remained elusive, finally appearing in print [5] in 2010, more than 15 years after it was first reported isolated in pure form [6]. In view of the likelihood that F. tetraquetra would, on phylogenetic grounds, contain interesting bioactive principles, the present phytochemical and antimicrobial studies were undertaken. Investigation of the leaves of F. tetraquetra afforded two pentacyclic triterpenoid lactones, 3-hydroxy-11,12epoxyoleanan-28,13-olide 1 [7] and 3-hydroxyurs-11en-28,13-olide 2 [8], the pentacyclic triterpenoid acids 3-hydroxy-12-oleanen-28-oic acid (oleanolic acid) 3 [9] and 3β-hydroxy-12-ursen-28-oic acid (ursolic acid, ) 4 [9], two common phytosterols, sitosterol and stigmasterol, and D-sorbitol. Oleanolic acid (3) (0.1%) and ursolic acid (4) 1574 Natural Product Communications Vol. 6 (11) 2011 30 30 29 29 20 19 O 1 2 3 4 HO 11 25 9 10 5 H 23 12 26 O 28 18 13 14 H 21 CO 12 22 15 8 7 27 1 26 25 9 18 13 14 H 4 Compound 1 10 5 H H 23 24 CO 1 22 8 7 27 25 1 2 15 6 24 3 4 10 5 HO H 23 Compound 2 R 9 20 19 12 18 26 13 14 H 11 17 16 3 30 R 20 28 21 2 HO 6 O 11 17 16 H 19 21 17 COOH 16 28 15 8 H 22 27 7 6 24 Compound 3 Compound 4 R = CH3 and 1R = H R = H and 1R = CH3 Figure 1: Structures of compounds 1, 2, 3 and 4. Mulholland et al. Table 2: Primary screening results on the effects of compounds (at 50 µM) 1-4 on Semliki Forest virus (SFV) replication and BHK/Huh-7 cell viability. Compound 1 2 3 4 SFV replication 54.2 ± 7.9 48.0 ± 6.6 97.0 ± 13.1 4.6 ± 6.8 BHK cell viability 85.3 ± 3.2 80.0 ± 1.5 81.8 ± 1.4 31.9 ± 4.2 Huh-7 cell viability 78.2 ± 2.3 103.9 ± 2.4 98.3 ± 1.1 86.9 ± 1.4 Values represent the % remaining virus replication/cell viablity/enzyme activity compared with vehicle-treated control (100%), n = 3. BHK, baby hamster kidney cell line; Huh-7, human hepatocyte cell line. Table 1: Primary screening results on the antibacterial effects of compounds 1-4 at 50 µM concentration. Escherichia coli Pseudomonas aeruginosa Staphylococcus aureus -3.3 ± 0.5 2.2 ± 1.8 10.9 ± 5.7 -0.3 ± 0.9 3.6 ± 2.1 4.2 ± 1.3 -3.8 ± 6.2 1.8 ± 0.8 3 11.9 ± 0.7 6.4 ± 0.4 5.1 ± 1.4 14.0 ± 2.7 4 27.1 ± 2.0 34.1 ± 1.3 8.4 ± 1.2 99.6 ± 1.1 Compound Enterobacter aerogenes 1 2 Values represent % inhibition (avg ± S.E.M., n = 3) of bacterial growth after 24 h compared with a control treated with the vehicle (DMSO) only. (0.2%) were the principal constituents of the n-hexane extract of the dry leaf material, while the methanol extract was found to consist mainly of D-sorbitol (8.0%). Compounds 1-4 were also assayed against the Semliki Forest virus (SFV), an enveloped (+)-stranded RNA virus that is the most widely studied member of the genus Alphavirus. Compound 3 was found to be completely inactive against SFV in the primary screen at 50 µM, while compounds 1 and 2 were moderately active (~50% inhibition) (Table 2). Compound 4 displayed significant antiviral activity [95.4% at 50 µM; IC50, by dose-response assay (Figure 2), 14.7 µM], but it also showed considerable cytotoxicity (68.1% reduction in cell viability after exposure for 24 hours) toward the baby hamster kidney (BHK21) cells hosting the SFV infection. However, exposure of human hepatocellular carcinoma (Huh-7) cells to compounds 1-4 showed no significant differences in cell viability even after 48 hours, which is surprising as 4 has previously been found to be significantly cytotoxic to both human keratinocytes (HaCaT) and human pulmonary embryonic fibroblasts (MRC-5) [10], while the activity of 1 against human oral squamous carcinoma (HSC-2) cells is reported to be equivalent to that of etoposide [11]. Compounds 1-4 were tested for their antibacterial properties against both Gram-negative (Enterobacter aerogenes, Escherichia coli, and Pseudomonas aeruginosa) and Gram-positive (Staphylococcus aureus) bacterial strains. Initial screening (Table 1) showed that all four compounds were inactive (<35% inhibition at 50 µM) against all of the Gram-negative bacterial strains. Compound 4, however, was found to be significantly active (99.6% inhibition) at 50 µM [MIC, by doseresponse assay (results not shown), 12.5 µM] against S. aureus, while compounds 1-3 were inactive. Figure 2: Dose-response curve of compound 4 in the antiviral assay. SFV infections were treated with various concentrations of compound 4 and the data were fitted into a sigmoidal dose-response curve. All data points represent results from six replicates. Experimental General: NMR spectra were recorded at room temperature in CDCl3. on either a 400 MHz Varian UNITY-INOVA or a 400 MHz Bruker AVANCE III spectrometer, HREIMS for compounds 1-4, were recorded on a Kratos 9/50 HRMS instrument. Plant material: The leaves of Fadogia tetraquetra were collected in October 2002 from Buffelskloof Private Nature Reserve in Mpumalanga Province, South Africa. A voucher specimen (N. Crouch & J. Burrows, 1019) has been lodged at the KwaZulu-Natal Herbarium (NH) for verification purposes. Extraction and isolation procedure of compounds: The air-dried, milled leaves (0.50 kg) of Fadogia tetraquetra var. tetraquetra were extracted successively, for 24 h with each solvent, in a Soxhlet apparatus with n-hexane, CH2Cl2, EtOAc and MeOH yielding 15.8, 16.6, 11.3 and 87.8 g of extract, respectively. The n-hexane extract was fractionated using VCC over silica gel (Merck 9385) collecting fractions of 30 cm3.. Fractions 25-40 (100% CH2Cl2) yielded impure 3-hydroxyurs-11-en-28,13olide (2), purified by CC (2% EtOAc : 98% CH2Cl2), fractions 41-43 (100% CH2Cl2) impure 3-hydroxy11,12-epoxyoleanan-28,13-olide (1), purified by CC (5% EtOAc : 95% CH2Cl2), fractions 44-48 (100% CH2Cl2) impure oleanolic acid (3), purified by CC (10% EtOAc : 90% CH2Cl2), and fractions 49-62 (2% MeOH : 98% CH2Cl2), impure ursolic acid (4), purified by CC (10% EtOAc : 90% CH2Cl2). Pentacyclic triterpenoids from Fadogia tetraquetra Natural Product Communications Vol. 6 (11) 2011 1575 Antibacterial assays: Compounds 1-4 were dissolved in dimethylsulfoxide (DMSO) to prepare a 10 mM stock solution. This was further diluted into Mueller Hinton II Broth for the assays (final DMSO concentration 0.5%). wells. After 15 min incubation at RT, luciferase substrate solution was added and the luminometric readout was measured by using a Varioskan Flash plate reader (Thermo Fischer Scientific Inc.). MEM containing 0.2% BSA and 20 mM Hepes (pH 7.2) was used as the medium for all dilutions and infections, and 3'-amino-3'-deoxyadenosine [12] was used as a positive control (at 20 µM, remaining SFV replication was 17%). For dose-response experiments on compound 4, concentrations of 50 µM, 25 µM, 10 µM, 5 µM, 1 µM, 0.5 µM, 0.1 µM and 0.05 µM were used. The data from the dose-response experiment were fitted into a sigmoidal curve using GraphPad Prism 5.0 software. Compounds were assayed against Enterobacter aerogenes ATCC 13048, Escherichia coli ATCC 25922, Pseudomonas aeruginosa ATCC 27853, and Staphylococcus aureus ATCC 25923 at 50 µM concentration by the broth microdilution method on 96well plates according to guidelines set by the Clinical and Laboratory Standards Institute (CLSI). In brief, initial inoculums of 5105 CFU/mL in Mueller Hinton II Broth were used, and antibacterial effects were evaluated against DMSO-treated control after 24 h incubation at 37C. Absorbance at 620 nm (Victor2 V multilabel counter, PerkinElmer) was used as a quantitative measure of bacterial growth. After initial tests, the MIC was determined for compound 4 against S. aureus by using twofold serial dilutions starting from 50 µM concentration. Antiviral activity assay: A luciferase-based reporter gene assay with marker virus Semliki Forest virus (SFV)-Rluc [12] was used for screening for anti-SFV activity. SFV was used as a representative member of the Alphavirus genus, species of which are spread in nature by mosquitoes. Pathogenic viruses in this genus include, for example, Chikungunya and Sindbis viruses [13]. In brief, baby hamster kidney BHK21 cells (ATCC CCL-10, used as the host) were grown in Dulbecco’s Modified Eagle’s Medium (MEM) supplemented with 8% fetal calf serum (FCS), 2% tryptose-broth phosphate, 1% L-glutamine, 100 IU/mL penicillin and 100 µg/mL streptomycin. The culture was kept at 370C with 5% CO2 atmosphere and 95% air humidity. Recombinant SFV strain containing Renilla reniformis luciferace insertion (SFV-Rluc; MOI 0.001 PFU/cell) was used for infecting confluent BHK21 cell cultures in 96-well plates. Renilla luciferase activity of samples was determined after 14 h incubation using Promega’s Renilla Luciferase assay system kit. The infection medium was removed, wells were washed with PBS and cell culture lysis reagent was added into the Cell viability assays: The effects on cell viability were evaluated with two cell lines: baby hamster kidney (BHK21, the host cell line for SFV) and human hepatocellular carcinoma (Huh-7) (for details, see [14]). BHK21 cells were grown as described above. Huh-7 cells were cultured in DMEM supplemented with 10% FBS, 1% L-glutamine, 1% non-essential amino acids, 100 IU/mL penicillin and 100 µg/mL streptomycin in conditions described above. Overnight grown cell cultures on 96-well plates (seeding densities 15000 cells/well for both cell lines) were treated with compounds for either 24 (BHK21) or 48 h (Huh-7), and cell viability was assessed by measuring the intracellular ATP content with Promega’s CellTiter-Glo assay kit according to the manufacturer’s instructions. Polymyxin B (7500 IU/mL) was used as a positive control (remaining cell viability 21% and 20% in BHK21 and Huh-7 cell assays, respectively). Acknowledgments - We thank Mr Dilip Jagjivan for the running of NMR spectra and Dr Philip Boshoff at the Cape Technikon, South Africa for HREIMS of compounds 1-4. The Trustees of the Buffelskloof Private Nature Reserve, Lydenburg, generously permitted collection of plant materials from this site, and the assistance of Mr J. Burrows in this respect is gratefully acknowledged. Dr Tero Ahola and Mr Mohammed Syedbasha are acknowledged for their contributions towards the antiviral studies. References [1] [2] [3] [4] [5] [6] [7] [8] Bridson DA. (1998) Rubiaceae. Flora Zambesiaca, 5(2), 265-281. Hurter LR, Naudé TW, Adelaar TF, Smit JD, Codd LE. (1972) Ingestion of the plant Fadogia monticola Robyns as an additional cause of gousiekte in ruminants. Onderstepoort Journal of Veterinary Research, 39, 71-82. Fourie N, Schultz RA, Prozesky L, Kellerman TS, Labuschagne L. (1989) Clinical pathological changes in gousiekte; a plantinduced cardiotoxicosis of ruminants. Onderstepoort Journal of Veterinary Research, 56, 73-80. Prozesky L, Bastianello SS, Fourie N, Schultz RA. (2005) A study of the pathology and pathogensis of the myocardial lesions in gousiekte, a plant-induced cardiotoxicosis of ruminants. Onderstepoort Journal of Veterinary Research, 72, 219-230. Bode ML, Gates, PJ, Gebretnsae, SY, Vleggaar, R. (2010) Structure elucidation and stereoselective total synthesis of pavettamine, the causal agent of gousiekte. Tetrahedron, 66, 2026-2036. Fourie N, Eramus, GL, Schultz RA, Prozesky L. (1995) Isolation of the toxic responsible for gousiekte, a plant-induced cardiomyopathy of ruminants in southern Africa. Onderstepoort. Journal of Veterinary Research, 62, 77-87. Melek FR, Radwan AS, Ahmed AA, Hamman AA, Aboutabl EA. (1989) Triterpenes from Atractylis carduus L. Die Pharmazie, 735. Tan M-L, Wang, Y, Zhuo, L, Jiang, W. (2007) Pentacyclic triterpenes from Eucalyptus globulus Labill. fruits. Tianran Chanwu Yanjiu Yu Kaifa (Natural Product Research and Development, CAN 148: 49843), 19, 232-234. 1576 Natural Product Communications Vol. 6 (11) 2011 [9] [10] [11] [12] [13] [14] Mulholland et al. Hung C-Y, Yen G-C. (2001) Extraction and identification of antioxidative compounds of Hsian-tsao (Mesona procumbens Hemsl.) Lebensemittel-Wissenschaft und -Technologie (Food Science and Technology, CAN 135:215834), 34, 306-311. Fontanay S, Grare M, Mayer J, Finance C, Duval RE. (2008) Ursolic, oleanolic and betulinic acids: Antibacterial spectra and selectivity indexes. Journal of Ethnopharmacology, 120, 272-276. Kuroda M, Aoshima T, Haraguchi, M, Young MCM, Sakagami H, Mimaki Y. (2006) Oleanane and taraxerane glycosides from the roots of Gomphrena macrocephala. Journal of Natural Products, 69, 1606-1610. Pohjala L, Barai V, Azhayev A, Lapinjoki S, Ahola T. (2008) A luciferase-based screening method for inhibitors of alphavirus replication applied to nucleoside analogues. Antiviral Research, 78, 215-222. Griffin DE. (2001) Alphaviruses. In Fields Virology, 4th ed. Knipe DM, Howley PM. (Eds), Lippincott, Williams & Wilkins, Philadelphia, pp. 917-962. Pohjala L, Tammela P, Samanta SK, Yli-Kauhaluoma J, Vuorela P. (2007) Assessing the data quality in predictive toxicology using a panel of cell lines and cytotoxicity assays. Analytical Biochemistry, 362, 221-228.