In vitro antiplasmodial activity and toxicity assessment of some plants from Nigerian ethnomedicine

Abstract

Context: The emergence and spread of Plasmodium falciparum-resistant parasites to nearly all available antimalarial drugs pose a threat to malaria control and necessitates the need to continue the search for new effective and affordable drugs. Ethnomedicine has been shown to be a potential source of antimalarial compounds or source of template for the synthesis of novel antimalarial molecules.

Objective: The antiplasmodial activity and toxicity assessment of 30 plant extracts from eight medicinal plants identified in Nigerian ethnomedicine for the treatment of febrile illnesses were evaluated.

Materials and methods: In vitro antimalarial activity was evaluated using Plasmodium falciparum NF54 (sensitive to all antimalarial drugs) and K1 (chloroquine/pyrimethamine resistant) strains in the [³H]-hypoxanthine incorporation assay. Toxicity was determined against mammalian L6 cells using Alamar blue assay.

Results: The ethyl acetate extract of leaves of Ocimum gratissimum Linn. (Labiatae) and hexane extract of stem bark of Trema orientalis (L.) Blume (Ulmaceae) showed the highest antiplasmodial activity (IC₅₀ 1.8-1.93 µg/mL) against P. falciparum K1 strain but elicited low cytotoxicity (selective index >10). However, hexane, ethyl acetate or methanol extracts of leaves of Terminalia catappa Linn. (Combretaceae), Jatropha curcas Linn. (Euphorbiaceae), Vitex doniana Sweet. (Verbenaceae) and stem bark of Vitex doniana displayed antiplasmodial activity (IC₅₀ 2.3-16.9 µg/mL) with good selectivity (21–120) for malaria parasites.

Discussion and conclusion: The antiplasmodial activity of Terminalia catappa and Vitex doniana against P. falciparum K1 is being reported for the first time in Nigerian ethnomedicine and these plants could be potential source of antimalarial agents.

Keywords::

• Plasmodium falciparum
• L6 cell
• cytotoxicity
• Terminalia catappa
• Vitex doniana

Introduction

Malaria remains a major public health problem in sub-Saharan Africa where the burden is enormous. An estimated 350–500 million clinical malaria episodes occur annually. Most of these are caused by infections with Plasmodium falciparum and Plasmodium vivax (WHO and UNICEF, 2005). Falciparum malaria causes more than 1 million deaths each year. More than 80% of malaria deaths occur in Africa south of the Sahara (WHO and UNICEF, 2005). In the absence of an effective vaccine, prevention and prompt treatment are the mainstays of control of malaria. Despite advances made towards the control of the disease, increasing resistance of vectors to insecticides and the spread of strains of P. falciparum resistant to almost all available antimalarial drugs are major impediments to the control of the disease. This situation underlines the urgent need for the development of new, effective, cheap and safe, antimalarial drugs. Ethnomedicine has given rise to two important antimalarial drugs still in use today: quinine an alkaloid isolated from the bark of the Peruvian plants Cinchona calisaya Wedd. and Cinchona succirubra Pav. ex Klotzsch (Rubiaceae) (Bruce-Chwatt, 1988), and artemisinin from Artemisia annua L. (Asteraceae) (Klayman, 1985). This has inspired many researchers to further intensify the search for antimalarial drugs from the plant compendium. Nigeria, with a wealth of unexplored biodiversity is an ideal place to search for new antimalarial molecules. In continuation of our work on the evaluation of plants used in Nigerian ethnomedicine for the treatment of febrile illnesses (Ajaiyeoba et al., 2002), the antimalarial activity and toxicity of eight medicinal plants were investigated.

Materials and methods

Plant collection and authentication

Eight plant species from six different families (Table 1) were collected between January and March 2006, in Ibadan, Oyo state, Nigeria. These plant species were identified and authenticated by Mr Usang Felix and Mr Osiyemi Seun plant taxonomists at the Forestry Research Institute of Nigeria (FRIN), Ibadan where voucher specimens were deposited (Table 1).

Table 1.  Characteristics of medicinal plants investigated.

Characteristics of medicinal plants investigated				% yield after extraction
Family	Plant species	Part used	Herbarium number	Hexane extract	ETOAc extract	MeOH extract
Verbenaceae	Vitex doniana	Stem bark	FHI 108354	22.4	23.72	25.6
Verbenaceae	Vitex doniana	Leaf	FHI 108354	23.37	23.68	27.19
Ulmaceae	Trema orientalis	Leaf	FHI 107813	24.38	23.31	26.06
Ulmaceae	Trema orientalis	Stem bark	FHI 107813	25.68	24.34	31.59
Leguminosae	Piliostigma thonningii	leaf	FHI 107815	14.63	12.07	20.24
Combretaceae	Terminalia catappa	Leaf	FHI 107812	25.01	23.13	25.86
Labiatae	Ocimum gratissimum	Leaf	FHI 108353	24.02	24.34	26.07
Euphorbiaceae	Jatropha curcas	Leaf	FHI 108355	24.41	23.95	28.18
Euphorbiaceae	Euphorbia hirta	Whole plant	FHI 108351	23.48	23.19	27.16
Euphorbiaceae	Phyllanthus amarus	Leaf	FHI 108356	25.95	23.03	26.3

Extraction of plant material

Plant parts were air-dried, powdered and extracted 3 times at 5 min interval sequentially with hexane, ethyl acetate and methanol using an accelerated solvent extractor (ASE 200 Dionex Corporation, Sunyale, USA), at a temperature of 70°C and pressure 120.0 bar. After removal of solvent, the yield of extracts was determined and extracts stored at −20°C till needed for analysis. In total 30 extracts from 8 plant species were obtained.

In vitro testing for antiplasmodial activity

Plasmodium falciparum strains NF54 (sensitive to all known antimalarial drugs) and K1 (chloroquine/pyrimethamine resistant) were cultivated and maintained in continuous culture according to a previously described method (Trager & Jensen, 1976). The culture medium consisted of RPMI (Roswell Park Memorial Institute medium) 1640 (Gibco, cat. no. 51800043) supplemented with 0.5% AlbuMAX® II (Gibco, cat. no. 11021-045), 25 mM HEPES (4-(2-Hydroxyethyl)piperazine-1-ethanesulonic acid; Sigma, cat. no. H4034), 25 mM NaHCO₃ (Sigma, cat. no. 71628) (pH 7.3), 0.36 mM hypoxanthine (Sigma, cat. no. 56700) and 100 μg/mL neomycin (Sigma, cat. no. N1142). Human erythrocytes served as host cells and parasite cultures kept at 37°C in an atmosphere of 3% O₂, 4% CO₂ and 93% N₂ in humidified modular chambers. In vitro antiplasmodial activity was determined using a modification of the semi-automated micro-dilution technique (Desjardins et al., 1979). Initial plant extract concentration was 50 μg/mL. Two-fold serial dilutions were done with hypoxanthine-free culture medium in 96-well plates to make seven concentrations, the lowest being 0.78 μg/ mL. An asynchronous parasite culture (0.3% parasitemia and 1.25% final hematocrit) was added to the plant extract solutions in 96-well plates for 48 h, at 37°C, prior to the addition of 0.5 µCi of [³H] hypoxanthine (Anawa, Zürich, Switzerland) per well, for 24 h. Parasites were then harvested onto glass-fiber filters and radioactivity was counted using a Betaplate liquid scintillation counter (Wallac, Zurich). The results were recorded as counts/min per well at each drug concentration and expressed as a percentage of the untreated controls.

Toxicity assessment

Rat skeletal myoblast L6 cells maintained in RPMI 1640 media containing 10% fetal calf serum and 1.7 μM l-glutamate was used. The toxicity of plant extracts to rat skeletal myoblast L6 cells was determined according to a previously described alamar blue assay method (Kaminsky et al., 1996). Briefly, L6 cells were seeded in 96-well microtiter plates at a density of 4 × 10³ L6 cells/well in culture medium. A three-fold serial dilution ranging from 450–0.78 μg/mL of plant extract was added to the test plates. The plates were incubated for 70 h at 37°C in 5% CO₂, after which resazurin (0.125 mg/mL) was added. Resazurin, a non-fluorescent redox dye (blue) which is converted by viable, metabolically active cells to fluorescent resorufin (pink) was used to measure the endpoint of the assay. The assay plates were read with a fluorescence microplate fluorometer (Spectramax Gemini XS, Bucher Biotech, Basel, Switzerland) using an excitation wavelength of 536 nm and an emission wavelength of 588 nm. Fluorescence development was expressed as a percentage of the control and the IC₅₀ values were determined. Podophyllotoxin (Polysciences Inc., Warrington, USA) was used as a positive reference.

Data analysis

For the antiplasmodial assay, the drug/plant extract concentration at which growth was inhibited by 50% (IC₅₀) was calculated by linear interpolation (Huber & Koella, 1993). For cytotoxicity determination, the IC₅₀ values were evaluated using a preprogrammed calculus sheet on SOFTmax PRO program. The calculus sheet consisted of (1) a formula that calculated the mean of the duplicates per sample, (2) subtraction of the respective fluorescence background of each dilution of the extract, (3) conversion of the mean fluorescence unit value to a percentage of the response by taking the mean of the negative control as 100% and (4) conversion of the percentage to the IC₅₀ by non-linear regression. Antiplasmodial activity of plant extracts was classified as good (IC₅₀ < 5 µg/mL), moderate (IC₅₀ 5−30 µg/mL) and inactive (IC₅₀ > 30 µg/mL). For the purpose of this study, an extract lacked toxicity to L6 cells by displaying an IC₅₀ value > 90 µg/mL.

Results and discussion

The eight plant species, the parts of plants extracted and percentage yields after extraction are shown in Table 1. The 50% inhibitory concentration (IC₅₀) of plant extracts against the two strains of P. falciparum, mammalian L6 cells and selective indices are presented in Table 2. Of the 30 plant extracts tested against the drug-sensitive NF54 strain of P. falciparum, 5 showed good activity, while 20 showed moderate activity. Against the drug-resistant K1 strain, 13 showed good activity, while 15 showed moderate antiplasmodial activity. The remaining plant extracts were inactive (IC₅₀ > 30 µg/mL). Plant extracts displayed better activity against the drug-resistant strain than the drug sensitive strain of P. falciparum. Overall, the ethyl acetate extract of leaves of Ocimum gratissimum was the most active extract, IC₅₀ = 1.8 ± 0.14 and 3.61 ± 0.22 µg/ mL against KI and NF54, respectively. This was followed by the ethyl acetate extract of leaves of Trema orientalis with IC₅₀ of 1.99 ± 0.3 and 2.29 ± 0.59 µg/mL against P. falciparum KI and NF54, respectively. Furthermore, the IC₅₀ values of plant extracts against mammalian L6 cell were very high for 20 out of the 29 plants extracts evaluated (IC₅₀ > 90 µg/ mL), signifying low toxicity. The least toxic extracts were hexane extract of leaves of Vitex doniana and Jatropha curcas with IC₅₀ values of 431.35 ± 3.21 and 414.76 ± 10.55 µg/mL, respectively. In contrast, the remaining nine extracts displayed IC₅₀ < 90 µg/mL against mammalian L6 cells with hexane extract of leaves of Euphorbia hirta as the most toxic extract with IC₅₀ of 14.19 ± 0.35 µg/mL. On comparison of Euphorbia hirta and reference compound (podophyllotoxin) activities against mammalian L6 cells (reference compound, IC₅₀ = 0.0052 ± 0.0019 µg/mL for L6 cells), it was found that the hexane extract of leaves Euphorbia hirta is relatively non-toxic.

Table 2.  Inhibitory concentration of 50% (IC₅₀) of plant extracts against P. falciparum K1, NF54 strains, mammalian L6 cells and selective indices.

Plant parts	P. falciparum K1	P. falciparum NF54	Mammalian L6 cells	Selectivity index	Selectivity index
	Mean IC50 ± SEM(µg/mL)	Mean IC50 ± SEM(µg/mL)	Mean IC50 ± SEM(µg/mL)	L6/K1	L6/NF54
Trema orientalis (L)
Hexane extract	4.159 ± 0.5	18.16 ± 5.51	226.46 ± 5.91	54.45	12.47
Ethyl acetate extract	1.99 ± 0.3	2.29 ± 0.59	32.53 ± 3.3	16.35	14.21
Methanol extract	6.79 ± 1.92	25.65 ± 0.79	409.02 ± 0.59	60.24	15.95
Trema orientalis (B)
Hexane extract	1.93 ± 0.15	4.9 ± 0.22	52.62 ± 2.54	27.26	10.74
Ethyl acetate extract	2.16 ± 0.36	3.98 ± 0.42	46.16 ± 2.64	21.37	11.6
Methanol extract	12.27 ± 0.09	20.27 ± 0.9	129.38 ± 4.04	10.54	6.38
Vitex doniana (L)
Hexane extract	3.60 ± 0.46	11.53 ± 4.88	431.35 ± 3.21	119.82	37.41
Ethyl acetate extract	3.87 ± 0.55	13.20 ± 4.95	122.6 ± 0.7	31.68	9.29
Methanol extract	34.17 ± 5.2	>50.0	>450	13.17	9.0
Vitex doniana (B)
Hexane extract	6.78 ± 0.48	9.52 ± 1.47	ND		NA
Ethyl acetate extract	13.47 ± 3.08	16.91 ± 0.22	335.47 ± 10.47	24.90	19.84
Methanol extract	29.25 ± 1.32	>50.0	>450	15.38	9.0
Jatropha curcas (L)
Hexane extract	3.08 ± 4.12	11.13 ± 3.36	414.76 ± 10.55	134.66	37.27
Ethyl acetate extract	2.39 ± 0.54	5.06 ± 0.36	126.45 ± 4.45	52.91	24.99
Methanol extract	11.53 ± 2.41	31.09 ± 4.36	>450	39.03	14.47
Terminalia catappa (L)
Hexane extract	10.1 ± 0.61	21.93 ± 2.99	276.75 ± 3.51	27.4	12.62
Ethyl acetate extract	3.05 ± 0.24	6.68 ± 1.37	159.92 ± 2.99	52.43	23.94
Methanol extract	7.42 ± 0.57	9.40 ± 1.08	377.59 ± 6.84	50.89	40.17
Phyllanthus amarus (L)
Hexane extract	8.71 ± 2.01	19.04 ± 4.62	161.45 ± 5.97	18.54	8.48
Ethyl acetate extract	5.62 ± 1.06	8.66 ± 1.05	77.69 ± 3.0	13.82	8.97
Methanol extract	22.32 ± 1.84	23.7 ± 4.12	427.54 ± 3.5	19.16	18.04
Euphorbia hirta (L)
Hexane extract	4.33 ± 0.48	11.55 ± 0.69	14.19 ± 0.35	3.28	1.23
Ethyl acetate extract	7.20 ± 0. 62	10.65 ± 0.68	65.79 ± 0.27	9.14	6.18
Methanol extract	25.04 ± 0.65	30.37 ± 3.07	>450	17.97	14.82
Ocimum gratissimum (L)
Hexane extract	2.45 ± 0.27	2.72 ± 0.43	34.66 ± 4.05	14.15	12.74
Ethyl acetate extract	1.8 ± 0.14	3.6 ± 0.22	60.14 ± 3.75	33.41	16.71
Methanol extract	22.52 ± 2.27	29.67 ± 5.7	424.2 ± 5.24	18.84	14.3
Piliostigma thonningii (L)
Hexane extract	5.85 ± 0.52	8.9 ± 0.16	362.75 ± 12.26	62.01	40.76
Ethyl acetate extract	3.56 ± 0.83	7.14 ± 0.12	56.14 ± 0.51	15.77	7.86
Methanol extract	38.86 ± 6.22	>50.0	>450	11.58	9.0

SEM, Standard error; L, leaves; B, stem bark; ND, not done. NA, not available, Chloroquine IC₅₀ = 0.12 ± 0.036 and 0.0031 ± 0.016 µg/mL for K1 and NF54, respectively. Artemisinin IC₅₀ = 0.001 ± 0.0007 and 0.0009 ± 0.0004 µg/mL for K1 and NF54, respectively. Podophyllotoxin IC₅₀ = 0.0052 ± 0.0019 µg/mL for L6 cells. The IC₅₀ values given are means ± standard error of two (cytotoxicity assay) or three (antiplasmodial assay) independent assays, each assay was run in duplicate.

Selectivity index (SI) defined, as the ratio of IC₅₀ L6 cells to IC₅₀ P. falciparum was also determined. The higher the SI the more promising the extract due to its selectivity for malaria parasites. Fifteen out of the 29 plant extracts evaluated displayed strong selectivity for P. falciparum K1 strain (SI > 20), while only 6 of the plant extracts displayed strong selectivity for P. falciparum NF54 strain (SI > 20). The highest SI values were obtained from hexane extract of leaves of Jatropha curcas (SI = 136). Hexane, ethyl acetate or methanol extracts of leaves of Terminalia catappa, Jatropha curcas, Vitex doniana and stem bark of V. doniana are of particular interest associating antiplasmodial activity (IC₅₀ 2.39–16.9 µg/ml) with high selectivity indices (21–120) against both P. falciparum NF54 and K1 strains.

Some of the plants included in this study have previously been described as having antiplasmodial activity and the activity was confirmed in this study. This was the case for Phyllanthus amarus Schum. & Thonn. (Euphorbiaceae) (Adjobimey et al., 2004), Euphorbia hirta Linn (Euphorbiaceae) (Tona et al., 1999; Liu et al., 2007) and Ocimum gratissimum L. Labiatae (Tchoumbougnang et al., 2005). In contrast, some of the plants evaluated in this study have previously been reported to lack antiplasmodial activity, but activity was detected in this study. This was the case for Trema orientalis L. Blume Ulmaceae (Verotta et al., 2001), Piliostigma thonningii Schum. Leguminosae/Caesalpinioideae (Clarkson et al., 2004) and Jatropha curcas L. Euphorbiaceae (Kaou et al., 2008). Differences in the antiplasmodial activity of some of the plant extracts observed in this study and previous studies may be as a result of many factors such as the season, age, intra-species variation, part collected, soil, climate, methods used for extraction and bioassay (Muregi et al., 2003).

To lend support to the antiplasmodial activity of the tested plants, the reported phytochemical constituents were reviewed. Investigation of dichloromethane and ethyl acetate extracts from trunk and root barks of Trema orientalis afforded compounds such as xanthones, secoiridoid, ursane derivatives, triterpenes and phytosteroids (Tchamo et al., 2001). Xanthones from Calophyllum caledonicum L. Clusiaceae and Garcinia vieillardii Pierre Clusiaceae (Hay et al., 2004), triterpene e.g. lupeol from Cassia siamea Lam. Caesalpiniaceae (Ajaiyeoba et al., 2008) or lupeol-based libraries from triterpenoid lupeol (Srinivasan et al., 2002) have been shown to display antimalarial activity. In addition, piliostigmin, a 2-phenoxychromone, quercetin, quercitrin, 6-C-methylquercetin 3-methyl ether, 6-C-methylquercetin 3,7,3'-trimethyl ether, 6,8-di-C-methylkaempferol 3-methyl ether and 6,8-di-C-methylkaempferol 3,7-dimethyl ether were isolated from the leaves of P. thonningii (Ibewuike et al., 1996). Similar C-methylated flavonols and the oxychromonol isolated from Piliostigma reticulatum showed antimicrobial activity and toxicity in brine shrimp lethality assay (Babajide et al., 2008). Flavonol glycosides afzelin, quercitrin and myricitrin isolated from another plant, e.g. methanol extract of E. hirta aerial parts, showed antiplasmodial activity and little cytotoxic property against human epidermoid carcinoma KB cells (Liu et al., 2007). In addition, lathyrane and podocarpane diterpenoids, stigmasterol, β-sitosterol, triterpenes, and flavonoid glycosides (Khafagy et al., 1977; Ravindranath et al., 2004) were isolated from J. curcas.

Finally, the last two plants evaluated in this study were T. catappa and Vitex doniana of which antiplasmodial activity against P. falciparum K1 and toxicity against mammalian L6 cells is being reported for the first time in Nigerian ethnomedicine. Recently, the activity of these two plants against P. falciparum NF54 was reported using three different in vitro techniques (Abiodun et al., 2010). Previous reports of in vitro antiplasmodial activity of other species of the genus Terminalia had been reported (Okpekon et al., 2004; Shuaibu et al., 2008). A previous study reported that extracts of leaves of Terminalia catappa were more cytotoxic to human hepatoma cells than to normal liver cells (Ko et al., 2003). Flavonoid glycosides, squalene, ursolic acid, 2α, 3β, 23-trihydroxyurs-12-en-28-oic and hydrolysable tannins were isolated from the leaves of T. catappa (Fan et al., 2004; Chen & Li, 2006; Lin et al., 2001). Similarly, hydrolysable tannins such as ellagic acid, flavogallonic acid, punicalagin and terchebulin isolated from stem bark of Terminalia avicennoides showed in vitro antiplasmodial activity with no cytotoxic effect on a mammalian cell line (Shuaibu et al., 2008). Several flavonoids had been shown to exert antiplasmodial activity (Ahmed et al., 2001). Ursolic acid from Mitragyna inermis possessed moderate in vitro antimalarial activity and strong antiproliferative activity on human monocytes (Traore-Keita et al., 2000). In addition, acetone extract of other species of the genus Vitex together with a labdane diterpene isolated from Vitex rehmannii had been shown to display antimalarial activity and cytotoxic activity against Graham cells in a previous study (Nyiligira et al., 2008). In this study, hexane extract displayed the best antiplasmodial activity, lacked cytotoxicity activity against the mammalian L6 cells and also showed great selectivity for P. falciparum K1 and NF54 over L6 cells, whereas the ethyl acetate extract of the leaves of V. doniana displayed antiplasmodial activity with reduced selectivity for P. falciparum NF54. Flavonoids such as luteolin, 5,4′-dihydroxy-3,6,7, 3′-tetramethoxyflavone, artemetin and isorhamnetin, were isolated from the root bark of Vitex aynus castus. With the exception of isorhamnetin, all the other flavonoids isolated showed cytotoxic activity against P388 lymphocytic leukemia cells (Hirobe et al., 1997).

Conclusion

Investigations to identify the active antimalarial compounds from Terminalia catappa and Vitex doniana by bioassay-guided fractionation are now in progress. It is hoped that a lead compound may be identified that could be developed as an antimalarial agent.

Acknowledgments

We are grateful to Matthias Hamburger, Institute of Pharmaceutical Biology, University of Basel, for assistance in plant extraction.

Declarations of interest

This work received support from UNICEF/UNDP/World Bank/WHO special program for Research and Training in Tropical Diseases (ID 980046). Oyindamola Abiodun was supported at the Swiss Tropical Institute by a training fellowship from Medicines for Malaria Venture. The authors have no conflicts of interest to report.