Evaluation of the cytotoxic activity of extracts from medicinal plants used for the treatment of malaria in Kagera and Lindi regions , Tanzania

Ramadhani S. O. Nondo, Mainen J. Moshi, Paul Erasto, Denis Zofou, Abdel J. Njouendou, Samuel Wanji, Moses N. Ngemenya, Abdul W Kidukuli, Pax J. Masimba, Vincent P.K. Titanji Department of Biological and Pre–Clinical Studies, Institute of Traditional Medicine, Muhimbili University of Health and Allied Sciences, Dar es Salaam, Tanzania. National Institute for Medical Research, Dar es Salaam, Tanzania. Biotechnology Unit, University of Buea, Buea, South West Region, Cameroon. Research Foundation in Tropical Diseases and Environment, Buea, South West Region-Cameroon.


INTRODUCTION
The use of plants as source of medicines for treatment of infectious and non-infectious diseases is an old human tradition (Petrovska, 2012), and the practice is now increasing due to increased global health challenges (WHO, 2002).Malaria is one of diseases treated by herbal medicines originating from different plant parts such as roots, stem bark, leaves, flowers and fruits.It is an old life threatening parasitic disease caused by parasites of the genus Plasmodium.The parasites infect and destroy red blood cells, leading to fever, severe anaemia, cerebral malaria and death may occur if the patient is not treated properly and on time (Fidock et al., 2004;NIAID, 2007).
Exploration of the accumulated indigenous knowledge on the treatment of malaria using medicinal plants enabled the isolation of two important and currently used antimalarial drugs; artemisinin from Artemisia annua and quinine from the bark of Cinchona spp (Wells, 2011).Despite plants being a rich source of useful chemical compounds of various structures and with different pharmacological properties on biological systems (Butler, 2004;Moshi et al., 2009), some of them may be toxic to humans.For example, some of the toxicities associated with the use of medicinal plants include allergic reactions, irritation of the gastrointestinal tract, destruction of red blood cells, and damage of body organs such as the heart and kidney and carcinogenicity (Westendorf, 1999;IARC, 2012).Several medicinal plants have previously been reported to be toxic.Some of the examples include Symphytum officinale L. used for wound healing which contains hepatotoxic pyrrolizidine alkaloids and Valerian officinalis used as a sedative for treatment of insomnia and anxiety which causes hepatitis (Abdualmjid and Segi, 2013).Aristolochia spp contain aristolochic acid I and II that cause renal failure ( Debelle et al., 2008); Drimia sanguinea and Bowiea volubilis which are traditionally used for headache, oedema, infertility and bladder problems contain cardiotoxic bufadienolides (Van der Bijl Jr. and Van der Bijl Sen., 2012).Although the use of herbal medicines is controlled in many countries, information about their efficacy and safety is based on traditional knowledge transmitted through generations over years and not on pre-clinical and clinical evaluation (Chalut et al., 1999).Tanzania shares the same experience of having a number of traditional healers who use traditional medicines for treatment of different diseases and is endowed with over 12,000 plant species, of which at least 10% have medicinal values (Mahunnah et al., 2012).Furthermore, Tanzania is among the six African countries with many reported cases of malaria (WHO, 2012) and because of the long history of the disease, the practice of using medicinal plants to treat malaria is very common (Mahunnah, 1987;Gessler et al., 1995Kinung'hi et al., 2010).Although several antimalarial medicinal plants have been documented in Tanzania, their safety has not been well studied.Therefore in this study the toxicity of crude extracts of medicinal plants used for the treatment of malaria in Kagera and Lindi regions, Tanzania, were assessed using the LLC-MK2 monkey kidney epithelial cell line and the brine shrimp larvae (Artemia salina L.).

Extraction of crude extracts
The powdered plant materials were macerated in 80% ethanol at room temperature for 24h and then filtered through cotton wool.The solid plant materials were macerated again in the same solvent for another 24h and the extracts obtained from the first and the second extractions were pooled and concentrated under vacuo using a Heldolph ® rotary evaporator (Heldolph instruments GmbH, Schwabach, Germany) to obtain viscous extracts which were further dried using a freeze drier (Edwards High Vacuum International, Crawley Sussex, England).The dry extracts were stored at -20°C until use.

Preparation of stock solutions
Stock solutions were prepared by dissolving 4 mg of crude extracts in 100 µL dimethyl sulfoxide and then diluted with RPMI-1640 cell culture medium to make 400 µg/mL.All solutions were sterilized by passing through 0.22 µm syringe-adapted filters and stored at -20°C until use.

Determination of cytotoxic activity on LLC-MK2 cells
Cytotoxicity of the crude extracts was evaluated on LLC-MK2 monkey kidney epithelial cells.Cells were grown in RPMI-1640 culture medium with L-glutamine and 25 mM HEPES (Steinheim, Germany).The medium was supplemented with 2 mg/mL NaHCO 3 (sigma), 10 µg/mL hypoxanthine (Sigma), 11.1 mM glucose (sigma), 10% FBS (BioWhittaker ® , Verviers, Belgium) and 5µg/mL gentamicin.The cells were incubated at 5% O 2 , 5% CO 2 , and 90% N 2 in humidified incubator (SHEL LAB™, Sheldon Mfg Inc, OR, USA) at 37°C until confluent before used for cytotoxicity assay.Trypsinated cells were distributed in 96 well plates at 10,000 cells in 100 µL per well and incubated for 48 h to allow them to attach before adding the extract.After 48 h the medium was removed completely from each well, and 100 µL of fresh culture medium was then added.Thereafter 100 µL of crude extracts (400 µg/mL) were added in row H and then serially diluted to row B to give concentrations ranging from 200 -3.125 µg/mL.Cells in row A served as controls without drug (100% growth).The cells with or without extracts were incubated at 37°C for 72 h before determining their viability.Each concentration level was tested in triplicate.

MTT Assay
Cell viability was determined using MTT assay (Niles et al., 2008;2009).After 72 h of incubation, the culture medium in each well with or without extract was removed completely from the assay plates and replaced by 100 µL of fresh culture medium.Then 10µL of 5 mg/mL Thiazolyl Blue Tetrazolium Bromide, MTT (Sigma) were added into each well to achieve a final concentration of 0.45mg/mL before incubated for 3 h at 37°C.After 3 h, the culture medium with MTT was carefully removed followed by addition of 100µL dimethylsulfoxide to dissolve formazan crystals and then incubated for 1 h before recording the optical density (Emax-Molecular Devices Corporation, California, USA) at 595 nm.

Data analysis
The percentage viability and percentage mortality were calculated from the OD values using Microsoft Excel 2010.The mean results of the percentage mortality were plotted against the logarithms of concentrations using the Fig P computer program Ver 4.189/07 (Biosoft Inc, USA).Regression equations obtained from the graphs were used to calculate the fifty percent cytotoxic concentration (CC 50 ), which is the concentration killing fifty percent of the cells.An extract with CC 50 >30 µg/mL is considered non-toxic (Fadeyi et al., 2013).

Brine shrimp toxicity assay
The brine shrimp lethality assay is a non-specific toxicity assay that is used in natural products research to detect the presence of pharmacologically active chemical constituents.It uses Artemia salina L. (Artemiidae) larvae (Meyer et al., 1982).Solutions of plant extracts were made in dimethylsulfoxide.The brine shrimp toxicity assay was conducted and data analyzed as previously reported (Nondo et al., 2011).An LC 50 (concentration killing fifty percent of the brine shrimp larvae) value greater than 100 μg/mL is considered to represent a non-toxic compound or extract (Moshi et al., 2010).Each extract was tested in duplicate and the concentrations of dimethylsulfoxide were restricted to a maximum of 0.6% in the final volume.
The brine shrimp toxicity assay showed that thirty extracts (65.2%) out of the 46 extracts tested had LC 50 values greater than 100 µg/mL; the cut-off point.Among these, 8 extracts had LC 50 values greater than 1000 µg/mL, while the remaining had LC 50 values between 100 and 800 µg/mL.Only Sixteen extracts (34.8%) showed LC 50 <100 µg/mL, and therefore classified as toxic.Maesa lanceolata leaf extract was the most toxic with LC 50 = 1.55 µg/mL, followed by the extracts from Dalbergia malangensis leaves (16.47 µg/mL), Aspilia natalensis aerial parts (34.93 µg/mL), Desmodium salicifolium stem (36.87 µg/mL), and Dalbergia malangensis stem extract (47.59 µg/mL) (Table 2).The high toxicity of M. lanceolata leaf extract on brine shrimp larvae may be due to the effect of saponins.Previous studies revealed that the leaves of M. lanceolata are rich in triterpenoidal saponins and these compounds were reported to have high molluscicidal and hemolytic activities (Sindambiwe et al., 1998;Apers et al., 2001).According to the American National Cancer Institute (NCI), a crude plant extract is considered to be cytotoxic if its CC 50 value on mammalian cells is <30 µg/mL (Fadeyi et al., 2013).On the other hand the cut-off point to consider a crude plant extract non-toxic in the brine shrimp toxicity assay is LC 50 >100 µg/mL (Moshi et al., 2010).Based on the results obtained in the two bioassays, three extracts were found to be toxic against LLC-MK2 cells and 16 extracts were found to be toxic on brine shrimp larvae.Of these only one extract from A. natalensis aerial parts was found to be toxic in both assays, which may be an indicator of consensus for cytotoxicity.
The extract from A. toxicaria stem bark was ranked the most toxic on the mammalian cells (LLC-MK2 cells) but it was ranked as exceptionally non-toxic using the brine shrimp toxicity assay (with LC 50 >1,000 µg/mL).On the other hand M. lanceolata leaf extract was ranked as the most toxic on brine shrimp assay with LC 50 = 1.55 µg/mL but ranked as non-toxic on LLC-MK2 cells test (Table 1 and 2).These observations suggest that the two models used in this study complement each other for the detection of toxic compounds that may be attributed to different mechanisms of toxicity; although the brine shrimp bioassay was found to be more sensitive in detecting toxic extracts than LLC-MK2 cells.The difference may be explained partly by the non-specificity of the brine shrimp assay in detecting toxic compounds (Meyer et al., 1982) and the differences in the criteria set to define a toxic substance, although in some studies brine shrimp assay has been reported to demonstrate some correlation with cell line results for detecting cytotoxic compounds/extracts (Meyer et al., 1982;Carballo et al., 2002).
The cytotoxicity of A. natalensis aerial parts extract was predicted by both assays; but it exhibited higher toxicity to mammalian cells than to brine shrimp larvae.The cytotoxic activity of A. natalensis (CC 50 = 18.57± 1.04 µg/mL) on LLC-MK2 cells was comparable to that of the standard cytotoxic drug used in this study (Imatinib, Gleevec) which had CC 50 of 18.61 µg/mL.A previous study revealed that an infusion and paste prepared from leaves of A. natalensis are used topically in South Africa to treat skin diseases (Mabona et al., 2013), but information regarding its toxicity was limited.Information from the traditional healers who reported these plants indicated that the decoctions of A. natalensis leaves and A. toxicaria leaves and stem bark are used orally for malaria associated with high fever (''Malaria kali'').They, however, emphasized that the decoction of A. natalensis should be consumed in small quantity because if taken in large quantities it causes stomach pain.These results may support the safety concern raised by traditional healers regarding oral administration of extracts from this plant.
Antiaris toxicaria is a known poisonous plant used in arrow poisoning associated with the presence of a number of cardiac glycosides which are inhibitors of Na + /K + -ATPase pump (Kopp et al., 1992;Shi et al., 2010).In addition, the cardiac glycosides and coumarins isolated from A. toxicaria were reported to have cytotoxic activity on various cancer cell lines (Dai et al., 2009;Liu et al., 2013;Shi et al., 2014).In this study we found that ethanolic extracts of the leaves and stem bark of A. toxicaria were very toxic to non-cancer cells (LLC-MK2).However, these results do not support the questionnaire-based toxicity information collected from traditional healers.
During our ethnobotanical survey, traditional healers reported that decoctions of leaves and stem bark were non-toxic when taken orally for treatment of malaria.This information from the reporting traditional healers is supported by animal studies.Kang et al., (2008) reported that aqueous and ethanolic leaf extracts of A. toxicaria were not toxic to mice even at high doses when given orally.But toxicity was observed when these extracts were administered by intra-peritoneal route.This may suggest that the bioavailability of the cardiac glycosides present in leaves and stem bark is low when given orally compared to when given through other routes.Apart from the toxicity evaluation reported in this study, the plant extracts reported in this study were previously evaluated for in vitro antimalarial activity against P. falciparum Dd2 strains.At a single concentration of 100 µg/mL, ethanolic extracts from A. toxicaria stem bark, M. lanceolata leaves, A. natalensis and D. salicifolium aerial parts inhibited the growth of malaria parasites in vitro (Nondo et al., 2015).Since the LLC-MK2 cells are normal mammalian cells, toxicity against these cells most likely predicts lack of selectivity and thus it will be toxic to mammalian cells, and therefore the traditional healers and patients should be informed on the risk of toxicity that might arise following use of extracts from these plants.18.61 ± 1.30 a= aqueous extract, b= methanol extract.All other extracts are extracted by 80% ethanol.CC50= cytotoxic concentration fifty percent (mean ± SD, n =3).R = root, S = stem, SB = stem bark, L = leaves, AP = aerial parts (stem + leaves), F = fruits, FL = flowers, WP = whole plant CONCLUSION Most of the antimalarial medicinal plants tested were non-toxic, and hence support the traditional healers' claims who believe that the herbal medicines they use are safe.However, further studies using different toxicity models are suggested to confirm their claims.Only the extracts of A. natalensis and A. toxicaria were categorized as toxic to mammalian cells.The evidence of A. natalensis toxicity obtained in this study supports the cautionary note that was given by the collaborating traditional healers.