Cytotoxicity of six South African medicinal plant extracts used in the treatment of cancer

Aqueous extracts prepared from six South African medicinal plants, with cancer-related ethnobotanical uses, were tested for their cytotoxic ability in vitro against three human cancer cell lines: DU-145 prostate cancer cells, MDA-MB-231 and MCF-7 breast cancer cells and a nonmalignant breast cell line, MCF-12A. The plants studied were: Bidens pilosa, Centella asiatica, Cnicus benedictus, Dicoma capensis, Hypoxis hemerocallidea and Sutherlandia frutescens. Of these plants, only D. capensis exhibited pronounced cytotoxic effects in two of the cell lines tested: MCF-7 and MCF-12A. © 2006 SAAB. Published by Elsevier B.V. All rights reserved.


Introduction
Reports on plants used for the treatment of cancer are rare in South Africa. Mongelli et al. (2000) found this to be true for Argentina plants, and ascribed it to the fact that cancer involves a complex set of signs and symptoms. It has been recommended that ethnopharmacological usages such as immune and skin disorders, inflammatory, infectious, parasitic and viral diseases be taken into account when selecting plants used to treat cancer, since these reflect disease states bearing relevance to cancer or a cancer symptom (Cordell et al., 1991;Popoca et al., 1998).
Since the majority of cancer chemotherapeutants severely affect the hosts' normal cells (Mascarenhas, 1994), the use of natural products has now been contemplated of exceptional value in the control of cancer (Suffness and Pezzuto, 1990). Furthermore, the search for new sources of biologically active compounds is important for the discovery of new drugs for the treatment of cancer.
This study determined the cytotoxic activities in aqueous extracts of six plants used by traditional healers in South Africa to treat cancer. The plants investigated were: (i) leaves and stems of Bidens pilosa L. (Asteraceae, Mai-Mai market, Johannesburg) used in the treatment of prostate gland tumours and inflammation (Watt and Breyer-Brandwijk, 1962;Hutchings et al., 1996); (ii) Centella asiatica (L.) Urb. (Araliaceae, South African National Biodiversity Institute (SANBI), Tshwane) leaves which are prescribed for skin complaints, rheumatoid arthritis, cancer and fevers (Watt and Breyer-Brandwijk, 1962;Hutchings et al., 1996); (iii) decoctions of Cnicus benedictus L. (Asteraceae, SANBI, Tshwane) taken for the treatment of internal cancer (Watt and Breyer-Brandwijk, 1962;Van Wyk et al., 1997); (iv) leaves and twigs of Dicoma capensis Less. (Asteraceae, gift from Prof. B-E van Wyk, Department of Botany, Rand Afrikaans University) which are prescribed in cases of cancer, high blood pressure and fever (Van Wyk et al., 1997), (v) Hypoxis hemerocallidea Fisch. and C.A. Mey. (Hypoxidaceae, Mai-Mai market, Johannesburg) corms used to treat bladder disorders and testicular tumours (Watt and Breyer-Brandwijk, 1962;Van Wyk et al., 1997) and (vi) leaf decoctions of Sutherlandia frutescens (L.) R.Br. (Fabaceae, provided by Dr. C. Albrecht, Medical Research Council) which are taken for stomach problems, internal cancer, inflammation and viral diseases (Watt and Breyer-Brandwijk, 1962;Rood, 1994;Van Wyk et al., 1997). Cytotoxicity was determined against a prostate carcinoma cell line (DU-145), breast cancer cell lines (MCF-7 and MDA-MB-231) and a nonmalignant breast cell line (MCF-12A). All the remedies investigated in this study are prescribed as infusions prepared by traditional healers, therefore, only aqueous extracts were tested in vitro. Dried plant material (1 g) was suspended in 10 ml deionised water and brewed as a tea by boiling for 15 min. Extracts were allowed to cool, centrifuged and the supernatants passed through 0.45 μm and 0.22 μm filters, consecutively. The yields were 0.75%, 2.85%, 2.49%, 0.55%, 3.69% and 2.38% for B. pilosa, C. asiatica, C. benedictus, D. capensis, H. hemerocallidea and S. frutescens, respectively.
Cells were seeded into 96-well flat-bottomed plates at a concentration of 3.0 × 10 5 cells per ml. After 24 h, cells were treated with plant extract, which was diluted with culture medium to a final concentration of 50 μg/ml (Zee-Cheng, 1997; Itharat et al., 2004). XTT labelling reagent (50 :l) was added and the absorbance (560 nm) read after 72 h (Gerlier and Thomasset, 1986). Experiments were carried out three times in triplicate. Active extracts were considered as those with less than 50% survival after an exposure time of 72 h. Cisplatin, a known anti-tumour agent, was used as positive control.
The effects of the plant extracts on proliferation in DU-145, MDA-MB-231, MCF-7 and MCF-12A cells are presented in Fig. 1A-D, respectively. Results showed B. pilosa to inhibit proliferation by 10% in the carcinoma cell lines. Inhibition of proliferation of leukemia cells in vitro has been reported with hot water extracts of this plant (Chang et al., 2001). In four of the five leukemia cell lines tested by these authors, the IC 50 ranged between 145 and 200 μg/ml, concentrations 3-4 fold above that tested in the present study. There are reports that describe B. pilosa as having antiviral (Chiang et al., 2003), antileukemic (Chang et al., 2001), anti-bacterial (Rabe and van Staden, 1997), anti-malarial (Brandao et al., 1997), antioxidative (Chiang et al., 2004), immunosuppressive and antiinflammatory activity (Jäger et al., 1996;Pereira et al., 1999). The extracts of H. hemerocallidea stimulated DU-145 and MCF-12A cell growth and inhibited the growth of the MCF-7 cells. This plant has been reported to display anti-inflammatory activity (Ojewole, 2002), an activity related to cancer. The aqueous extracts of S. frutescens inhibited growth of the oestrogen dependent cancer cell lines and stimulated the growth of the MCF-12A and MDA-MB-231 cells. Chinkwo (2005) found the crude aqueous whole plant extract of S. frutescens to induce cytotoxicity in cervical carcinoma and Chinese Hamster Ovary cells. Ethanolic extracts of S. frutescens commercial preparations (tablets and powder) have been reported to inhibit proliferation of both MCF-7 and MDA-MB-468 human breast cancer cells, human leukemia Jurkat cells, human promyelocyte HL60 cells and murine RAW 264.7 macrophage/monocyte cells (Tai et al., 2004). S. frutescens has been shown to possess antioxidant potential (Fernandes et al., 2004), which is thought to account for some of the anti-inflammatory properties already known (Ojewole, 2004). In this study extracts of C. asiatica stimulated the growth of three of the cell lines tested. Crude and partially purified fractions of a methanolic extract of C. asiatica have been reported to inhibit proliferation of transformed cell lines (Ehrlich ascites tumour cells: IC 50 62 μg/ml, Dalton's lymphoma ascites tumour cells: IC 50 75 μg/ml), and to be nontoxic to normal human lymphocytes (Babu et al., 1995). Antifungal, antibacterial, anti-inflammatory and anti-allergic activity has been described for C. asiatica (Ponglux et al., 1987). C. benedictus and D. capensis inhibited the proliferation of DU-145, MCF-7 and MCF-12A cells. We found no literature where the cytotoxic activity of C. benedictus or D. capensis has previously been investigated.
The American National Cancer Institute guidelines set the limit of activity for crude extracts at a 50% inhibition (IC 50 ) of proliferation of less than 30 μg/ml after an exposure time of 72 h (Suffness and Pezzuto, 1990). With the exception of D. capensis, the other plant species had IC 50 values higher than 50 μg/ml. The IC 50 values of the latter plant were 30 μg/ml and 31 μg/ml in MCF-7 and MCF-12A cells, respectively. The positive control, cisplatin, had IC 50 values of 0.27 μg/ml and 0.14 μg/ml in MCF-7 and MCF-12A cells, respectively.
Since cytotoxicity of the plants investigated has been reported by other authors, it is evident that different cell lines exhibit different sensitivities towards the plant extracts. Also, some plants are reported to have a cytotoxic effect on cancer cells (Kusuge et al., 1985;Alley et al., 1988) whereas other plant extracts activate several parameters of the immune system as a strategy to destroy cancer (Abuharfeil et al., 2000). Differences in results obtained in this study and those reported in the literature could also be ascribed to differences in extraction procedures and the natural variability in plants.
Since D. capensis showed the most pronounced cytotoxic activity, this plant will be evaluated further for the possible isolation of active compounds.