PHYTOTHERAPY RESEARCH Phytother. Res. 17, 168–173 (2003) Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ptr.1258 Cytotoxic Sesquiterpene Lactones from Centaurothamnus maximus and Vicoa pentanema Ilias Muhammad,1* Satoshi Takamatsu,1 Jaber S. Mossa,4 Farouk S. El-Feraly,4 Larry A. Walker1,2 and Alice M. Clark3† 1 National Center for Natural Products Research Department of Pharmacology Department of Pharmacognosy, Research Institute of Pharmaceutical Sciences, School of Pharmacy, University of Mississippi, University, MS, 38677, USA 4 Medicinal, Aromatic and Poisonous Plants Research Center (MAPPRC) and Department of Pharmacognosy, College of Pharmacy, King Saud University, P.O. Box 2457, Riyadh 11451, Saudi Arabia 2 3 The aerial parts of Centaurothamnus maximus yielded three cytotoxic guaianolides, chlorojanerin (1), cynaropicrin (2) and janerin (3). The structure elucidation of 1–3 was based on 1H and 13C NMR data, mainly 2D-NMR 1H-1H COSY and 1H-13C HETCOR experiments. Compounds 1–3 showed in vitro cytotoxic activity against human cancer cell lines of malignant melanoma (SK-MEL), epidermoid (KB), ductal (BT-549) and ovarian (SK-OV-3) carcinomas with IC50 values of 2–6 mg/mL. In addition, 12 sesquiterpene lactones (4–15), isolated previously from the aerial parts of Vicoa pentanema, were evaluated for cytotoxic and antimicrobial activities. 2a- Acetoxy-3b-hydroxyalantolactone (10) and 8b-hydroxyparthenolide (14) were found to be the main cytotoxic agents (IC50 values of 2–6 mg/mL against SKMEL, BT-549 and SK-OV-3), while lactones 4, 5, 11 and 15 selectively inhibited the growth of human malignant melonoma (IC50 value of 3.6–7.3 mg/mL). Cell aggregation and cell adhesion assays, using HL60 and HeLa cell lines, evaluated the effect of cytotoxic constituents 1–3, 10 and 14 on immune response and inflammation. Copyright # 2003 John Wiley & Sons, Ltd. Keywords: Centaurothamnus maximus; Vicoa pentanema; Compositae; sesquiterpene lactone; cytotoxic; cell aggregation; cell adhesion. INTRODUCTION Centaurothamnus maximus Wagentz & Dittri. (Fam. Compositae) is a leafy shrub with many branches and about 1.5 m tall (Collenette, 1999). The leaves have a silvery undersurface and the magenta flowers are about 4 cm wide with a faint sweet scent. The plant is found in many localities in the southern part of Saudi Arabia, on cliffs and steep hillsides. It has no known medicinal uses, perhaps because it is difficult to access. This plant has not previously been the subject of phytochemical analysis. Plants from Compositae, especially the genus Centaurea, have yielded a wide array of acetylenes (Bohlman et al., 1973) and sesquiterpene lactones (Massiot et al., 1986; Wang et al., 1991), including germacranolides (Barrero et al., 1989), guaianolides (Oksuz et al., 1994; Youssef and Frahm, 1994) and elemanolides (Tortajada et al., 1988) as the main secondary metabolites. Examination of the aerial parts of C. maximus has led to the isolation and characterization of three cytotoxic sesquiterpene lactones, namely the guainolides chlorojanerin (1), cynaropicrin (2) and janerin (3). * Correspondence to: I. Muhammad, National Center for Natural Products Research, Thad Cochran Research Center, University of Mississippi, University MS, 38677, USA. † Present address: Office of Research, University of Mississippi, University MS, 38677, USA. Contract/grant sponsor: Uehara Memorial Foundation, Tokyo. Contract/grant sponsor: United States Department of Agriculture; Contract/ grant number: 58-6408-7-012. Copyright # 2003 John Wiley & Sons, Ltd. Earlier investigation of the aerial parts of Vicoa pentanema Aitch. & Hemsl., another Compositae plant, yielded 12 sesquiterpene lactones (4–15), of which four were reported as new compounds (Mossa et al., 1997). In this paper we wish to report the isolation and characterization of compounds 1–3 from C. maximus, and the cytotoxic and antimicrobial activities of compounds 1– 15, as well as cell aggregation and adhesion studies of the cytotoxic sesquiterpene lactones. MATERIALS AND METHODS General. The NMR spectra were taken on a Varian instrument at 300 MHz (1H) and 75 MHz (13C), using CDCl3 as solvent and tetramethylsilane (TMS) as internal standard. Multiplicity determination (APT and DEPT) and 2D NMR spectra (COSY and HETCOR) were obtained using a standard Varian pulse-program. ESI-MS was run using a Bruker Bioapex instrument. Optical rotations were obtained at ambient temperature in CHCl3, unless otherwise stated, using a Perkin-Elmer 241 MC polarimeter. TLC was performed on silica gel GF254, using petroleum ether (40 °–60 °)–EtOAc (8:2) as solvent, with visualization using vanillin–H2SO4 spray reagent. Centrifugal preparative TLC (CPTLC, using a Chromatotron1 instrument, Harrison Research Inc. Model 7924) was run with 4 mm silica gel PF254 disc, using a flow rate of 2 mL/min. The isolation and structure elucidation of sesquiterpene lactones (4–15) from the aerial parts of V. Received 8 January 2002 Accepted 5 August 2002 CYTOTOXIC SESQUITERPENE LACTONES pentanema was described previously (Mossa et al., 1997). Plant material. The aerial parts of C. maximus were collected in Abha, Saudi Arabia on 16 May 1995. A voucher specimen (no. 13317) was deposited at the herbarium of the MAPPRC, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia. The aerial parts of V. pentanema were collected in Abha on 13 May 1993 (voucher no. 13063). Extraction and isolation of sesquiterpene lactones from C. maximus. The dried ground leaves of C. maximus (800 g) were percolated at room temperature with 95% EtOH and the extract was evaporated in vacuo to leave 39 g of gummy residue. The active EtOH extract (15 g) was flash chromatographed on silica gel (450 g), † Dictionary of Natural Products on CD ROM (1999). Chapman and Hall, CR Cnet Base, Version 8:1, Chapman and Hall #”s MTS50-D, JZL71-K JZL62-I. Copyright # 2003 John Wiley & Sons, Ltd. 169 using initially CH2Cl2 (1 L) followed by 1%–2% MeCN– CH2Cl2 as solvent, to afford chlorojanerin as needles [1; 50 mg; mp 155 °–156 °C (crystallized from CH2Cl2/nhexane), [a]D ‡69 ° (c 5.0, MeOH); Lit.† mp 151 °– 153 °C, [a]D ‡73 °], followed by mixtures of compounds 2 and 3 (2 g). The mixture (1 g) was subsequently separated by CPTLC (4 mm silica gel P254 disc; solvent: 15% EtOAc in n-hexane) which afforded cynaropicrin [2, 300 mg, [a]D ‡105 ° (c 3.0, MeOH); Lit.† [a]D ‡108.6 °), followed by janerin (3, 350 mg, [a]D ‡75 ° (c 5.0, CHCl3); Lit.† [a]D ‡69.5 °) as transparent gums. Cytotoxicity assay. The in vitro cytotoxic activity was determined against four human cancer cell lines, SKMEL, KB, BT-549 and SK-OV-3 (Table 2), obtained from the American Type Culture Collection (ATCC, Rockville, MD). A primary assay for initial extracts/ fractions used a single concentration (100 mg/mL). In a secondary assay active extracts/pure compounds were tested at three concentrations (10, 3.3 and 1.1 mg/mL), Phytother. Res. 17, 168–173 (2003) 170 I. MUHAMMAD ET AL. using a culture-treated 96-well microplate (Dou et al., 1996). The level of general toxicity of each sample was also determined by measuring their effect on a fibroblast cell line from African green monkey kidney (VERO; non-transformed). The assay is based on the accumulation of neutral red dye in the lysosomes of viable cells (Borenfreund et al., 1990). A subsequent addition of 2propanol will lyse the cells, releasing the dye into solution and the absorbance is measured at 490 nm and 630 nm. Corresponding growth inhibition was calculated and graphed. For secondary assays, IC50 values were determined from logarithmic graphs of growth inhibition values. The cytotoxic agents doxorubicin and 5-fluorouracil were used as positive controls, while the DMSO vehicle was used as a negative control. Antimicrobial assay. The preliminary antimicrobial activities of the crude extracts/fractions and the IC50 / MIC values of compounds 1–5 (Table 3) were determined using a modified microplate assay protocol with a 96-well format, as recommended by the National Committee for Clinical Laboratory Standards (1997). The test organisms used were ATCC strains of Candida albicans B311 (90028), Cryptococcus neoformans (90113), Staphylococcus aureus (6535), methicillinresistant S. aureus (33591). Amphotericin B and rifampicin were used as positive controls, with DMSO as a negative control. Cell aggregation assay. Cell aggregation was measured as previously described (Katagiri et al., 1999). HL-60, the myelomonocytic cell line, was suspended at a density of 1  106 cells/mL. Then 150 mL of the cell suspension was added to each well of a 96-well plate. After incubation with sample for 10 min, phorbol myristate acetate (PMA, Sigma) (10 ng/mL, final) was added. Plates were placed in a CO2 incubator and aggregation of the cells was observed microscopically 16 h after the PMA addition. Cytochalasin B, anti-LFA-1 and anti-ICAM-1 monoclonal antibodies were used as positive control cell aggregation inhibitors. XTT assay for cytotoxicity. Following the cell aggregation assay, the XTT (3'-1[(phenylamono)-carbonyl]-3,4tetrazolium-bis(4-methoxy-6-nitro)benzene sulfonic acid hydrate) assay was performed using the methods described by Scudiero et al. (1998). Briefly, 25 mL of XTT-phenazine methosulfate (PMS) solution (1 mg/mL XTT solution supplemented by 25 mM of PMS) was added to the cells in each well on the microplates. After incubating for 4 h at 37 °C, absorbance at 450 nm was measured by a microplate reader (reference absorbance at 630 nm). Cell adhesion assay. (Musza et al., 1994) HL-60 cells which express LFA-1, were stained with a CFSE (carboxyfluorescein diacetate succinimyl ester, Molecular Probes) (Bronner-Fraser, 1985). CFSE labelled HL-60 cells and potential inhibitors were added to the wells of 96 well microtitre plates which contained confluent monolayers of HeLa cells, a carcinoma cell line which expresses ICAM-1. Then, 50 ng/mL PMA was added to stimulate the HL-60 cells to convert LFA-1 to its high avidity binding state (Martin and Springer, 1987). The cultures were incubated for 45 min at 37 °C. Nonadherent HL-60 cells were washed away, the remaining cells were Copyright # 2003 John Wiley & Sons, Ltd. Table 1. and 3 Carbon 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 13 C NMR chemical shift valuesa for compounds 2 2b 45.3 39.0 73.7 152.2 51.4 78.5 47.5 74.3 37.0 137.3 139.3 169.1 122.7 118.2 113.5 165.3 141.7 62.1 126.7 sd t d s d d d d t s s s t t t s s t t 3c 45.6 36.4 76.0 68.2 53.0 76.6 47.9 74.2 37.6 137.0 139.2 169.0 122.7 118.6 48.4 165.3 141.4 62.1 126.7 sd t d s d d d d t s s s t t t s s t t a Spectra recorded at 75 MHz in CDCl3. Assignments for carbon 8, 10, 13, 14, 15 and 18 of 2 were incorrectly reported by Wang et al. (1991). c Data assigned for comparison; assignment for carbon 10 of 3 was incorrectly reported by Massiot et al (1986). d Multiplicities of the carbon signals are determined by APT and / DEPT experiments. Assignments were aided 2D NMR COSY and HETCOR experiments. b solubilized with 1% Triton X-100 (Sigma) and the fluorescence quantitated using a CytoFluor 2350, Fluorescence Measurement System (Millipore) with an excitation wavelength of 496 nm and emission at 519 nm. Anti-ICAM-1 monoclonal antibody and cytochalasin B were used as positive controls. RESULTS AND DISCUSSION A preliminary cytotoxicity screening showed strong activity of the C. maximus crude extract. A similar cytotoxicity assay was performed on the EtOH extract of V. pentanema, which was also found to be active against human cancer cell lines. The active EtOH extract of C. maximus was flash chromatographed over silica gel to give the sesquiterpene lactones 1–3 in 0.008%, 0.02% and 0.025% yields with respect to the dried plant material, respectively. These compounds were identified as chlorojanerin (1) (Gonzalez et al., 1977; Youssef and Frahm, 1994), cynaropicrin (2) (Corbella et al., 1972) and janerin (3) (Gonzalez et al., 1977) by comparison of their physical and spectroscopic data with those reported previously. The 13C-NMR data for 1 and 3 were generally in agreement with those reported by Youssef and Frahm, (1994) and Massiot et al. (1986), respectively. In addition, a revised 13C-NMR data for cynaropicrin (2) (Wang et al., 1991) is assigned in Table 1, using 2D NMR COSY and HETCOR experiments, as well as by comparison with those of 1 and 3. In addition, a total of 12 sesquiterpene lactones were isolated, including five guaianolides (4–8), four eudesmanolides (9–12), two Phytother. Res. 17, 168–173 (2003) CYTOTOXIC SESQUITERPENE LACTONES 171 Table 2. Cytotoxic activities of sesquiterpene lactones Compound Chlorojanerin (1) Cynaropicrin (2) Janerin (3) 4a,5a-Epoxy-10a,14H-1-epi-inuviscolide (4) 8-Epiconfertin (5) Inuviscolide (6) 10a-Hydroxy-14H-inuviscolide (7) 4a,5a-Epoxy-10a,11b,12H, 14H-1-epi-inuviscolide3b-glucoside (8) 2a-Acetoxy-3b-hydroxyalantolactone (10) 2a,3b-Diacetoxyalantolactone (11) 2a,3a-Dihydroxyalloalantolactone (12) 8b-Hydroxyparthenolide (14) Carabrone (15) Doxorubicin 5-¯uorouracil IC50 (mg/mL) BT-549 SK-OV-3 VEROa SK-MEL KB 2.3 2.0 2.7 4.1 3.6 5.3 >10 ± 6.0 5.5 6.0 ± ± 9.0 ± ± 5.9 5.7 5.8 >10b >10b 5.0 5.0 >10b 6.3 6.3 6.2 >10b >10b 7.3 8.1 >10b 5.9 6.0 6.7 3.7 5.0 5.4 >10 >10b 2.5 7.3 5.5 1.8 5.1 <1.1 6.3 ± ± ± >10 ± <1.1 >10b 5.9 ± >10b 5.0 >10b <1.1 ± 6.3 ± >10b 6.0 >10b <1.1 >10b 4.9 >10b 7.4 5.2 5.0 6.9 >10b SK-MEL, human malignant melanoma; KB, human epidermoid carcinoma; BT-549, human ductal carcinoma; SK-OV-3, human ovary carcinoma. a VERO (kidney, African green monkey) used as normal cell line; , inactive at 20 mg/mL b weak activity. germacranolides (13, 14) and one elemenolide (15), from V. pentanema (Mossa et al., 1997). Most of the sesquiterpene lactones tested for in vitro cytotoxic activity against human cancer cell lines of SKMEL, KB, BT-549 and SK-OV-3 were found to be active (Table 2). Among the guaianolides isolated from C. maximus and V. pentanema, compounds 1–3 and inuviscolide (6) demonstrated the most potent activities (IC50 values between 2 and 9 mg/mL) against these cell lines. The eudesmanolide 10 and germacranolide 14 showed consistent cytotoxicities against SK-MEL, BT549 and SK-OV-3 with IC50 values between 1.8 and 6.3 mg/mL, while the compounds 4, 5, 11 and 15 showed selective activities against human malignant melanoma (SK-MEL), with IC50 values of 4.1, 3.6, 7.3 and 5.1 mg/ mL, respectively. When tested against opportunistic infectious pathogens (Candida albicans and C. neoformans) guaianolides 4 and 5 were found to be weakly active only against C. albicans, with IC50 values at 15 and 20 mg/mL, respectively (Table 3). Interestingly, the main cytotoxic constituents 1–3 from C. maximus, as well as 10 and 14 from V. pentanema were found to be devoid of antimicrobial activity at 50 mg/mL, while most of the sesquiterpene lactones, except 7, 8 and 11 were found to be toxic to the VERO cell line (IC50 of 3.7–7.4 mg/mL). The guaianolide glucoside 8, which is devoid of the cytotoxic a-methylene-g-lactone chromophore (Kupchan et al., 1971), was found to be inactive against these cell lines. Cynaropicrin had previously been reported for its cytotoxic activity against solid and ascites tumours (S180 sarcoma and Ehrlich carcinoma) (Zong et al., 1994) and was also reported to be one of the neurotoxic agents of yellow star thistle that cause nigro-pallidal encephalomalacia in horses upon oral ingestion (Wang et al., 1991). The cytotoxic activity of alantolactone (9) and its ability to induce apoptosis in human T-cell leukaemia cells have recently been reported (Dirsch et al., 2000.), while lipiferolide (13) previously demonstrated activities against KB cell lines (Doskotch et al., 1975). The key cytotoxic sesquiterpene lactones from both Copyright # 2003 John Wiley & Sons, Ltd. Table 3. Antimicrobial activity of compounds 4–6, 10, 11 and 14 IC50/MIC (mg/mL)a Compound C. albicans C. neoformans MR S. aureus 4 5 6 10 11 15 AMB RIF 15/‡b 20/‡ 50/‡ ± 35/‡ 25/‡ 0.015/0.039 NT ± ± ± ± ± 20/‡ 0.045/0.156 NT ± ± ± 50/‡ ± ± ± 0.0025/0.3125 a IC50/MIC values after 48 h of incubation at 37 °C; ‡, MIC not determined; , inactive at 50 mg/mL; NT, not tested; AMB, amphotericin B; RIF, rifampicin. b plants were subjected to lymphocyte-associated antigen1 (LFA-1: CD 11a/CD 18)/intercellular adhesion molecule-1 (ICAM-1: CD 54) mediated aggregation and adhesion assays (Table 4), using HL-60 and HeLa cell lines, in order to determine their effects on immune response and inflammation (Carlos and Harlan, 1994; Hynes, 1992; Springer, 1990). For example, leukocyte adherence to the endothelial cell is an essential event in the process of inflammation and immune recognition (Picker and Butcher, 1992; Osborn, 1990), and the extracellular interactions between specific CAMs expressed on the endothelium and leukocytes will mediate leukocyte entry into tissues, T-cell proliferation and antigen presentation. Therefore, an agent that could inhibit leukocyte adhesion and transmigration would represent a novel mechanism of action as an immunosuppressive and antiinflammatory drug. 8b-Hydroxyparthenolide (14) potently inhibited cell aggregation with a MIC value of <0.15 mg/mL (vs 0.39 mg/mL for cytochalasin B), but was found to be less active in the Phytother. Res. 17, 168–173 (2003) 172 I. MUHAMMAD ET AL. Table 4. Cell proliferation, cell aggregation and cell adhesion activitya of compounds 1–3, 10 and 14 Compound 1 2 3 10 14 Cytochnb Cell agg MIC (mg/mL) (A) 0.5 1.4 1.4 1.4 <0.15 0.39 Cell prolif (HL-60) XTT IC50 (mg/mL) (B) Speci®c index (B)/(A) 0.8 1.1 1.1 4.0 1.2 >30 1.6 0.8 0.8 2.8 >8.0 76.9 Cell adh IC50 (mg/mL) (C) ± ± ± 7.0 5.0 0.25 Speci®c index (B)/(C) ± ± ± 0.6 0.2 >120 a Assay system for inhibitors of LFA-1/ICAM-1-mediated aggregation combined with XTT assay as a primary assay. Following LFA-1/ICAM-1-mediated adhesion assay was performed with HL-60 cells and HeLa cells as a secondary assay. , inactive at 12.5 mg/mL. b Cytochalasin B. secondary cell adhesion assay (IC50 5 mg/mL vs 0.25 mg/ mL for cytochalasin B). On the other hand, guaianolides 1–3 from C. maximus demonstrated potent cytotoxic activity against human leukaemia cell lines in the primary cell proliferation assay (MIC 0.8–1.1 mg/mL), while 8bhydroxyparthenolide (14) was >25 fold more cytotoxic than cytochalasin B as determined by XTT assay. The inhibitory effects on cell aggregation of compounds 1–3 and 10 were probably caused by the toxic effects of these sesquiterpene lactones, while 14 showed a more selective effect on cell aggregation that appears to be irrespective of its cytotoxicity. Thus, 8b-hydroxyparthenolide (14) decreased the expression of ICAM-1 induced by PMA on HL-60 cells. In the cell adhesion assay, between HL-60 cells and HeLa cells, 14 had negligible specificity compared with that of cytochalasin B. Therefore, 14 appeared to inhibit ICAM-1 but had no direct effect on protein-protein interaction between LFA-1 and ICAM-1. Acknowledgements The authors acknowledge the technical assistance of staff of the National Center for Natural Products Research, School of Pharmacy, University of Mississippi, USA: Dr Melissa R Jacob, Mrs Sharon Sanders, Mr Charles Dawson and Mr John Trott for biological work and Mr Frank Wiggers for recording NMR spectra. One of us (S.T.) thanks the Uehara Memorial Foundation, Tokyo, Japan, for the partial award of a research fellowship. This work was supported in part by the United States Department of Agriculture, ARS Specific Cooperative Agreement No. 58-6408-7-012. The College of Pharmacy, King Saud University authors acknowledge Dr Sultanul Abidin for the identification of plant material. REFERENCES Barrero FA, Sanchez JF, Rodriguez I. 1989. Guaianolides from Centaurea melitensis. Phytochemistry 28: 1975±1976. Bohlman F, Burkhardt T, Zdero C. 1973. Naturally Occurring Acetylenes. Academic Press: London, 452±460. Borenfreund E, Babich H, Martin-Alguacil N. 1990. Rapid chemosensitivity assay with human normal and tumor cells in vitro. In Vitro Cell Dev Biol 26: 1030±1034. Bronner-Fraser M. 1985. Alterations in neural crest migration by a monoclonal antibody that affects cell adhesion. J Cell Biol 101: 610±617. Carlos TM, Harlan JM. 1994. Leukocyte-endothelial adhesion molecules. Blood 84: 2068±2101. Collenette S. 1999. Wild¯owers of Saudi Arabia. The National Commission for Wildlife Conservation and Development (NCWCD): Riyadh. Saudi Arabia. Corbella A, Gariboldi P, Jommi G, Samek Z, Holub M, Drozdz B. 1972. Absolute streochemistry of cynaropicrin and related guaianolides. Chem Commun 386. Dirsch VM, Stuppner H, Vollmar AM. 2000. Sesquiterpene lactones induce apoptosis in human T-cell leukemia cells. International Congress and 48th Annual Meeting of the Society of Medicinal Plants Research, Zurich, SL 12. George Thieme Verlag: Stuttgart. Doskotch RW, Kelly SL Jr, Hufford CD, El-Feraly FS. 1975. New sesquiterpene lactones from Liriodendron tulipifera. Phytochemistry 14: 769±773. Dou J, McChesney JD, Sindelar RD, Goins DK, Walker LA. 1996. A new quassinoid from Castela texana. J Nat Prod 59: 73±76. Gonzalez AG, Bermejo J, Cabrera L, Galindo A, Massaret GM. 1977. Chemistry of plant compounds. XXIX. Active principles of Centaurea janeri Graells. An Quim 73: 86. Hynes RO. 1992. Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69: 11±25. Copyright # 2003 John Wiley & Sons, Ltd. Katagiri K, Yokosawa H, Kinashi T et al. 1999. Ubiquitinproteasome system is involved in induction of LFA-1/ ICAM-1- dependent adhesion of HL-60 cells. J Leukoc Biol 65: 778±785. Kupchan SM, Eakin MA, Thomas AM. 1971. Tumor inhibitors. 69. Structure- cytotoxicity relationships among the sesquiterpene lactones. J Med Chem 14: 1147±1152. Martin SD, Springer TA. 1987. Puri®ed intercellular adhesion molecule-1 (ICAM- 1) is a ligand for lymphocyte functionassociated antigen 1 (LFA-1). Cell 51: 813±819. Massiot G, Morfaux A-M, Men-Oliver L et al. 1986. Guaianolides from the leaves of Centaurea incana. Phytochemistry 25: 258±261. Mossa JS, El-Feraly FS, Muhammad I et al. 1997. Sesquiterpene lactones and thymol esters from V. pentanema. J Nat Prod 60: 550±555. Musza LL, Killar LM, Speight R et al. 1994. Potent new cell adhesion inhibitory compounds from the root of Trichilia rubra. Tetrahedron 50: 11369±11378. National Committee for Clinical Laboratory Standards. 1997. Methods for Dilution. Antimicrobial Susceptibility Test for Bacteria that Grow Aerobically. Approved Standard M7-A. National Committee for Clinical Laboratory Standards: Wayne, PA, 4th edn. Oksuz S, Serin S, Topcu G. 1994. Sesquiterpene lactones from Centaurea hermannii. Phytochemistry 35: 435±438. Osborn L. 1990. Leukocyte adhesion to endothelium in in¯ammation. Cell 62: 3±6. Picker LJ, Butcher EC. 1992. Physiological and molecular mechanisms of lymphocyte homing. Annu Rev Immunol 10: 561±591. Scudiero DA, Shoemaker RH, Paull KD et al. 1998. Evaluation of a soluble tetrazolium/formazan assay for cell growth Phytother. Res. 17, 168–173 (2003) CYTOTOXIC SESQUITERPENE LACTONES and drug sensitivity in culture using human and other tumor cell lines. Cancer Res 48: 4827±4833. Springer TA. 1990. Adhesion receptors of the immune system. Nature 346: 425±434. Tortajada A, Picher M-T, Reventos M-M, Amigo J-M. 1988. Structure and stereochemistry of melitensin, an elemanolide from Centaurea aspera Var. stenophylla. Phytochemistry 27: 3549±3550. Wang Y, Hamberger M, Cheng C-H K et al. 1991. 12. Neurotoxic sesquiterpenoids from the yellow star thistle Copyright # 2003 John Wiley & Sons, Ltd. 173 C. solstitialis L. (Asteraceae). Helv Chem Acta 74: 117± 123. Youssef D, Frahm AW. 1994. Constituents of the Egyptian Centaurea scoparia; chlorinated guaianolides of the aerial parts. Planta Med 60: 267±271. Zong Y, Yu M, Huang L, Chang Y, Wang Y, Che C-T. 1994. Studies on Tibetan medicinal plants. II. Antitumor activity of Saussurea eopygmaea. Int J Pharmacogn 32: 284±293. Phytother. Res. 17, 168–173 (2003)