The two traditional recipes were supplied by a traditional practitioner from Burkina Faso. These recipes are used to treat cervical cancer. The first recipe consists of five plants (Cochlospermum tinctorium, Crateva adansonii, Lannea acida, Balanites aegyptiaca, Vitellaria paradoxa and ingredients) and the second of a single plant (Urginea glaucescens and ingredients). These two recipes are used according to the availability of the constituent species, and given to patients in capsule form. The recipes were used for extraction by infusion, as indicated by the traditional practitioner. 50 g of plant powder from each recipe was placed in a glass bottle and then 500 mL of boiled distilled water was poured into it. The mixture was homogenised and left to infuse for an hour. The infused extract was then filtered and dispersed in 250 mL of distilled water. The filtrate obtained was filtered a second time and centrifuged. The supernatant obtained from the centrifugation was then freeze-dried and the dry extracts obtained were weighed in order to determine the extraction yield according to the formula:

$\text{Yield}\left(\%\right)=\left(M^{\prime }/M\right)\ast 100$

M = initial mass of the recipe;

$M^{\prime }$ = mass of dry extract after extraction and freeze-drying.

The residual moisture content (RMC) was also assessed using watch glasses. The empty mass of the glasses was weighed and one gram of powder was added to the glasses. The watch glasses were placed in an oven for two hours and 30 minutes. After cooling, they were weighed a second time. The RMC was calculated using the following formula:

$\text{RMC}\left(\%\right)=\left(\left(P-P^{\prime }\right)/P\right)\ast 100$

P: Initial weight (g) of plant material;

$P^{\prime }$: Final weight (g) of plant material after steaming.

Total phenolics content was determined using the Folin-Ciocalteu method [11] . Briefly, in 96-well plates, 50 μl of each extract at an initial concentration of 1mg/ml was mixed with 50 µl of Folin-Ciocalteu reagent diluted 1:10. Then 100 μl of Na₂CO₃ (20%) was added. The absorbance was read at 760 nm after 40 minutes of incubation against a blank containing methanol instead of extracts. The same procedure was carried out for gallic acid, the reference compound, but across a range of different concentrations by cascade dilutions in order to establish the calibration curve. The concentration of total phenolics in each extract was expressed as mg gallic acid equivalent per mg extract (mg GAE/mg), and calculated from the gallic acid calibration curve.

The flavonoids content in the various extracts was determined in a 96-well plate; 100 µl of AlCl₃ solution (2%) was added to 100 µl of the extract solution at a concentration of 1mg/ml prepared in methanol. The same procedure was used for quercetin, with concentration ranges diluted by half (from 1 mg/ml) to establish the calibration curve. After 40 minutes incubation, absorbance was read at 430 nm. The concentration of flavonoids in the extracts was deduced from a calibration line established with quercetin and was expressed in mg quercetin equivalent per mg of extract (mg QE/mg).

Spectrophotometric analysis of the activity of extracts to trap ABTS+ cations was determined using the method of [12] . The ABTS+ solution was prepared by dissolving 10 mg of ABTS in 2.6 mL of distilled water. Next, 1.7212 mg of potassium persulphate was added and the mixture was kept in the dark at room temperature for a minimum of 12 hours. The mixture was then diluted with ethanol to give an absorbance of 0.70 ± 0.02 at 734 nm. The dry extracts were weighed and dissolved in ethanol at a concentration of 1 mg/mL. In a 96-well plate, 50 µL of ethanolic extract solution was added to 100 µL of freshly prepared ABTS+ solution with a blank consisted of 50 µL of ethanol and 100 µL of ABTS solution. The mixture in the 96-well plates was then protected from light at room temperature for 15 min and the concentration was read at 734 nm in a spectrophotometer, with trolox as the reference compound. Each test was performed in triplicate and the experience was repeated three times. Inhibition percentages were calculated using the following formula:

$Inhibition\left(\%\right)=\left[\left(BlankAbsorbance-SampleAbsorbance\right)/\left(BlankAbsorbance\right)\right]\ast 100$

Inhibition percentages were then used to calculate the concentration that allowed inhibition of 50% of ABTS (IC₅₀).

The ability of recipes to trap DPPH radicals was determined using the method described by [13] . The extracts and quercetin were dissolved in methanol at a concentration of 1 mg/mL. A cascade dilution of the extracts was performed in a 96-well plate to obtain a volume of 100 µL of extract in each well. 200 μL of a DPPH solution (4 mg/100 mL in methanol) was then added. The blank consisted of 100 μL of methanol and 200 μL of DPPH solution. Absorbances were read at 490 nm using a spectrophotometer after 30 min incubation at room temperature. Each test was performed in triplicate and the experiment was repeated three times. The 50% DPPH radical inhibitory concentration (IC₅₀) was determined using the percentage inhibition corresponding to each concentration. The percentage inhibition was calculated using the following formula:

$Inhibition\left(\%\right)=\left[\left(BlankAbsorbance-SampleAbsorbance\right)/\left(BlankAbsorbance\right)\right]\ast 100$

The ferric reduction potential of recipes was determined in 96-well plates. The FRAP reagent was prepared by mixing 300 mM sodium acetate buffer adjusted to pH 3.6 with 10 mM 2,4,6-tripyridyl-j-triazine (TPTZ) solution and 20mM ferric chloride FeCl₃ in the ratio 10:01:01 (10 parts buffer to one part TPTZ and one part FeCl₃). The extracts were weighed and dissolved in distilled water at a concentration of 1 mg/mL. 10 µL of extract was added to each well and some wells were dedicated to the blank and contained 10 µL of buffer. Next, 300 µL of FRAP reagent was added to each well and the plate was read at 593 nm after 30 minutes incubation in the dark. Absorbances were read against a standard ascorbic acid curve. Rutin and quercetin were used as reference compounds. The potential of the extracts to reduce iron (III) to iron (II) was expressed in millimoles of ascorbic acid equivalent per gram of dry extract (mmol AAE/g) and calculated using the following formula:

$C=\left(c\ast D\right)/\left(M\ast Ci\right)$

where c = concentration read from the sample (calculated from the ascorbic acid curve); D = dilution factor; M = molar mass of ascorbic acid (176.1 g/mol); Ci = concentration of the stock solution.

The aim here was to test the ability of recipes to inhibit 15-lipoxygenase (15-LOX) activity. For this purpose, the spectrophotometric method described by [14] was used with a few modifications. The assay was performed in 96-well microplates. Extracts and the reference compound (indomethacin) were prepared at a final concentration of 100 µg/mL. The experimental protocol is summarized in Table 1 . The experiment was performed in triplicate and the corresponding inhibition percentages were calculated using the following formula:

$Inhibition\left(\%\right)=\left[\left(\text{Enzyme}Absorbance-SampleAbsorbance\right)/\left(\text{Enzyme}Absorbance\right)\right]\ast 100$

where Enzyme Absorbance = Absorbance of enzyme activity − Absorbance of Enzyme Blank; Sample Absorbance = Absorbance of sample activity − Absorbance sample blank.

COX activity inhibition tests were carried out according to the manufacturer’s procedure (Cayman). Tests were performed on a 96-well plate. Extracts were prepared at a final concentration of 100 µg/mL in the wells. A reaction mixture

consisting of 10 µL extract, 10 µL enzyme, 10 µL hemin and 150 µL diluted buffer was dispensed into one well of a 96-well plate. The same mixture was made without the extract but with 10 µL of the extract dilution solvent and dispensed into another well. The blank consisted of 160 µL diluted buffer, 10 µL hemin and 10 µL extract dilution solvent. The plate was homogenized and incubated for 5 minutes. Then 20 µL of substrate (arachidonic acid) and 20 µL of colorimetric substrate were added to the wells. The plate was read at 590 nm after homogenization and incubation for two minutes. Reaction mixtures were run in triplicate. Percentage COX inhibition was calculated using the following formula:

$Inhibition\left(\%\right)=\left[\left(\text{Enzyme}Absorbance-SampleAbsorbance\right)/\left(\text{Enzyme}Absorbance\right)\right]\ast 100$

where Enzyme Absorbance = Absorbance of enzyme activity − Absorbance of Enzyme Blank; Sample Absorbance = Absorbance of sample activity − Absorbance sample blank.

Cervical cancer HeLa and liver cancer HepG2 cell lines were thawed, cultured and maintained under sterile conditions in a laminar flow hood. These cells were grown at 37˚C in an incubator with 5% CO₂ in cell culture flasks, in DMEM medium supplemented with 10% fetal calf serum, one per cent penicillin-streptomycin and one per cent L-Glutamine. Cells at confluence were used to test cell viability.

Overall cell growth was assessed using the re-adapted Mosmann MTT colorimetric assay [15] . The cell proliferation assay is based on the ability of living cells to reduce MTT (yellow) to its metabolite, formazan blue (purple). In 96-well plates, 10,000 cells were seeded in 100 µL per well against a cell-free blank and an extract-free control. After 24 hours incubation, the medium was removed and replaced by the extracts at different concentrations (cascade dilutions) starting at 1 mg/mL. After 72 hours of incubation, 10 µL of tetrazolium salt (5 mg/mL in PBS 1X) was introduced into each well. The plates were then incubated for four hours and the supernatant was removed. Next, 100 µL of isopropanol was introduced into the wells to dissolve the formazan crystals. Plates were shaken for 20 min, then read by spectrophotometer at 570 nm. Growth inhibition was calculated using the following formula:

$Inhibition\left(\%\right)=\left[\left(\text{Control}Absorbance-\text{Blank}Absorbance\right)/\left(\text{Extract Absorbance}-\text{Blank Absorbanc}e\right)\right]\ast 100$

All data were presented as mean ± standard deviation. Data were analyzed by the one-factor ANOVA test followed by Tukey’s multiple comparisons test using R software with P < 0.05 as the criterion for statistical significance. Graphs were constructed in Excel.