Powdered plant material (400 g) was macerated in 4 l of ethanol/water in the ratio (70:30 v/v) or ethanol (95% v/v) for 48 hours at room temperature. The resulting extract was filtered through a whatman N˚1 filter paper and then dried using a rotary evaporator at 65˚C. The residues which constitute the crude extract were kept at 4˚C until further use.

Fractionation was done by flash chromatography using several solvents or mixture of solvents in the following order: methylene chloride, methylene chloride/methanol (95:5 v/v), methylene chloride/methanol (90:10 v/v), methylene chloride/methanol (50:50 v/v) and methanol (Figure 1). The solvent change was made when the filtrate appeared clear. At the end of the procedure, all filtrates obtained with the same solvent were mixed and then concentrated in a rotary

evaporator. The final fraction obtained was kept in a clean bottle and stored at 4˚C. Each fractionation yield (FY) was calculated according to the formula below.

Fractionation yield = (masse of fraction/masse of extract) × 100

Biphasic medium of Boeck and Drbohlav [25] that involves solid phase (ringer’s solution + egg) and liquid phase (lock’s solution containing nutrients) was used for E. histolytica poloyxenic cultivation. Before inoculation, complete media were pre incubated at 37˚C for 30 min to 1 h and 10 µL of polyxenic culture maintained in the Laboratory of Pharmacology and Toxicology of the University of Yaounde1 containing the clinical isolates of viable E. histolytica trophozoites were introduced in each tube. The tubes were incubated at 37˚C and the E. histolytica growth verified after every 48 or 72 h. Then, the tubes were removed from the incubator and shacked to detach parasites from the solid phase and left for 5 min then the supernatant was decanted to obtain the subculture. The pellet containing the parasites was introduced in a tube containing pre incubated new medium as previously described [18] [26] .

S. rhombifolia aqueous and ethanolic extracts were prepared using sterile DMSO (Sigma-Aldrich, and liquid phase of culture medium leading to concentrations of 200, 20, 2, 0.2 mg/ml respectively). Each mixture was filtered with sterile syringe filters (Ø 22 µm) and aliquots were prepared from these stock solutions. Parasites grown were harvested at midlog phase at the concentration of 1.67 × 10⁷ cells/ml of culture by counting using the haemocytometer (Neubauer, Hausser Scientific) and inoculated in tubes containing new 5 ml media in which 25 µl of plant materials were added. MTZ was used as a standard drug and was tested at 1, 10, 50 and 100 µg/mL. S. rhombifolia extracts were tested at the concentration of 1; 10; 50; 100 and 500 µg/mL µg/ml. One control tube was used in which parasites were incubated on culture medium containing 0.5% DMSO without any drug. Each testing concentration was made in triplicate and the experiment was repeated three times for each compound. All the tested tubes were incubated at 37˚C as previously described [27] and the viability was evaluated by trypan blue method after 24, 48 and 72 h. Amoebicidal activity was evaluated using the method described by [28] . In 1.5 ml micro centrifuge tube, 25 µl of parasite suspension and 225 µl of 0.4% trypan blue solution prepared in 0.9% NaCl was introduced. The mixture was homogenized and 10 µl of this mixture was used for cells counting. The chamber was covered with cover slip and the viable (bright) cells as well as the dead (blue) cells were counted at 40× on a light microscope. The concentration of the cell has been calculated using the following formula:

N = (n × d)/v

where, N = concentration of viable cells/ml; n = number of the viable cells counted in the chamber, d = dilution factor and v = the volume of the chamber (0.1 µl) The percentages of inhibition were calculated also using the formula below and IC50 were determined using the software Graphpad Prism 3.0,

Percentage inhibition (%) = (NC − NT)/NC × 100

NC = Number of viable amoebae in the control tube and NT = Number of viable amoebae in the testing tube.

This assay was carried out following a method previously described [29] . Briefly in a series of test tubes containing 3.1 mL of the methanolic solution of DPPH (40 µg/mL), 50 µL of plant extracts at different concentration was added. In the negative control tubes, the extract was replaced by 50 µL of solvent and the positive control by 50 µL of ascorbic acid. The mixtures were homogenized and incubated in the dark for 30 minutes at room temperature, and the absorbance was measured at 517 nm with a spectrophotometer. The percentages of inhibition were calculated by using the following formula:

% of DPPH scavenging activity = [(OD_control − OD_assay)/OD_control] × 100

where: OD_control: absorbance of the negative control tube;

OD_assay: absorbance of the test tube.

The IC50 value expressed in µg of extract per mol of DPPH for each extract was determined by using a non-linear regression curve of the DPPH scavenging activity against the concentration of extracts tested.

1) Preparation of the liver homogenate

One Wistar rat was sacrificed by cervical dislocation and the liver was excised. The rest of the manipulation was done in ice. The organ was washed in a saline solution of 0.9% NaCl and then spin-dried and weighed. Once the mass of the liver was known, it was cut into small pieces in a solution of 1.15% KCl and crushed using the Teflon plunger of the Potter apparatus and Bleau fountain sand. A 10% homogenate in a 1.15% KCl solution was then prepared taking into account the weight of the liver, divided into several tubes and centrifuged (720 g, 10 min, 4˚C). Each supernatant was collected and the volume noted. Depending on the number of assays to be performed, aliquots were prepared and stored in the freezer until use.

2) Lipid peroxidation inhibition assay

Thiobarbituric acid reactive substances were determined by using previously described method [30] . In each test tube, 50 µL of plant extracts, 1 mL of 10% rat liver homogenate, 50 µL of 0.5 mM FeCl₂ and 50 µL of 0.5 mM H₂O₂ were successively introduced. In the blank tube, FeCl₂ and H₂O₂ were replaced by 100 µL 1.15% KCl while in the negative control tube, the extraction solvent was used instead of the extract. The mixtures were incubated (1 hour, 37˚C). After incubation, 1mL trichloroacetic acid (TCA 15%) and 1 mL 0.67% TBA were added to all tubes and boiled in a water bath for 15 minutes. After cooling and centrifugation (1620 g, 5 min, 4˚C), the supernatants were collected and the absorbance of the pink stain read at 532 nm against the blank. The percentages of inhibition were calculated using the formula below.

% of lipid peroxidation inhibition = [(OD_control − OD_assay)/OD_control] × 100

where: OD_control: absorbance of the negative control tube;

OD_assay: absorbance of the test tube.

The evolution of the percentage of inhibition according to the concentration of extract used allowed us to determine the IC₅₀ in µg/mL.

The ferric reduction power assay was performed as previously described [31] . In each tube 50 µL of plant extracts, 1100 µL of phosphate buffer (0.6 M pH 6.6), 1000 µL of 0.25% potassium ferricyanide were introduced. In the blank, 1100 µL of distilled water was added instead of potassium ferricyanide. After incubation for 20 min at 50˚C, 1 mL of 10% trichloroacetic acid (TCA) was added to all tubes. The whole set was centrifuged (1620 g, 10 min 4˚C) then, to 1mL of supernatant were added 1mL of distilled water and 200 µL of ferric chloride. The whole was well homogenized and then left to stand for 10 min. The absorbance was measured at 700 nm against the blank in a spectrophotometer and the percentages of iron reduction were calculated by the following formula:

% reduction = [(OD_control − OD_assay)/OD_control] × 100

where: OD_control: absorbance of the negative control tube;

OD_assay: absorbance of the test tube.

The evolution of the percentage of reduction as a function of the extract concentration allowed us to determine the EC₅₀.

The TAC was measure following the method previously described [32] . In each test tube 50 µL of plant extracts, 1 mL of 0.6 M sulphuric acid, 1050 µL of 28 mM sodium phosphate and 1050 µL of 4 mM ammonium molybdate were successively introduced. The tubes were capped with the beads and heated for 90 min then cooled on a stream of cold water. The absorbance of the blue staining mixture was measured in a spectrophotometer at 695 nm. The antioxidant capacity of the extracts expressed in g ascorbic acid/mg extract was determined from the calibration curve obtained by using different concentrations of ascorbic acid instead of plant extracts.

The total phenolic and total flavonoid contents of different extracts and fractions were determined using the modified Folin-Ciocalteu method [20] [33] . Gallic acid and Quercetin were use as standard respectivly.

The antiradical activity was evaluated throughout the scavenging of the DPPH radical. C. variegatum leaf extracts and fractions effectively trap the DPPH radical in a concentration-dependent manner between 1 and 500 µg/mL (Figure 4(a)). The IC₅₀ revealed that the extracts and fractions exhibited moderate antiradical potential but this remained lower compared to ascorbic acid. The fractionation process decreased the DPPH antiradical activity of hydroethanolic extract and increased that of ethanolic extract.

The ability of the different extracts and fractions of Codiaeum variegatum to inhibit membrane lipid peroxidation was effective (Figure 4(b)). Moreover, the ethanolic extract shows the highest inhibitory activity. However, ascorbic acid

* = value significantly non different from the standard drug MTZ.

activity was more pronounced compared to that extract (respectively 21.52 ± 9.87 µg/mL and 10.26 ± 4.02 µg/mL). Three fractions (EEF1, EEF3, and EEF5) exhibited moderate inhibitory potentials.

The ferric reducing power assay revealed that ethanolic extract exhibited the highest activity extract with an evolution going in a concentration-dependent manner (Figure 4(c)) However the activity of ethanolic extract was lower than that of ascorbic acid (543.60 ± 42.52 µg/mL and 21.34 ± 2.91 µg/mL respectively) (Table 3). All the fractions exhibited very low ferric reducing powers.

The determination of the total antioxidant capacity was done by the phosphomolybdenum method, the results expressed in milligram ascorbic acid equivalent per gram of plant extract (mgEAA/g extract) showed us that hydroethanolic extract has the greatest ability to reduce the phosphomolybdic complex (54.77 ± 0.08 mgEq AA/g) followed by EEF5 (44.00 ± 0.01 mgEqAA/g) (Table 4). EH and EEF2 exhibited the lowest phosphomolybdic reducing power (13.72 ± 0.0 mgEq AA/g and 12.91 ± 0.13 mgEq AA/g respectively.

Total phenolic and flavonoid contents were determined for the extracts and fractions of C. variegatum. It was noticed that the phenolic ranged from 2 mgGAE/g to 35 mgGAE/g and flavonoid contents ranged from 0.04 mgQE/g to 2 mgQE/g (Table 4). EEF5 exhibited the highest phenolic compound content (35.13 ± 0.01 mgGAE/g) followed by the ethanolic (14.00 ± 0.02 mgGAE/g) and hydroethanolic (4.18 ± 0.04 mgGAE/g). Ethanolic fractions and extract exhibited higher flavonoid contents compared to hydroethanolic extract and fractions. No significant difference was found between the flavonoid content of hydroethanolic extract and those of its fractions.