Fresh A. cruentus and A. hybridus moisture rates and dried plant material extraction yields from extractions using aqueous acetone (i.e., hydroacetonic) , methanolic or water as a solvent are presented in Table 1. A. cruentus and A. hybridus contain remarkably high amounts of moisture, at 80.86 ± 1.18% and 83.45 ± 0.99%, respectively. Odhav et al. [6] and Aletor and Adeogun [7] reported 83% moisture for A. hybridus and 81.3% for A. cruentus. The two species possess a C4 photosynthetic system [16], which is known to be adapted for tropical regions. The amount of extractable components ranged from 18.55% in A. cruentus (with AE) to 7.43% in A. hybridus (with ME). A. cruentus exhibited the best yields for the three types of extractions. In this study, the water extract investigation was important, as it represents the most used traditional preparation for these species.

Phytochemical analyses revealed the presence of polyphenols, tannins, flavonoids, steroids (including cardenolids), terpenoids (i.e., iridoïds, triterpenes and carotenoids), saponins and betalains (Table 2).

The thin layer chromatography analyses of flavonoids in the HAE showed the presence of rutin (Rf: 0.72) in both genotypes and other unidentified compounds (Figure 1). Anthracenosides, coumarins and alkaloids were not detected, likely because of the extraction methods used. For example, alkaloid extraction requires acidic or alkaline solvents. The two species have presented a similar phytochemical profile. Flavonoids, hydroxybenzoic acids, hydroxycinnamic acids and hydroxycinnamyl amides were detected in the young aerial parts of Amaranthus genotypes [17]. Perdesen et al. [18] reported phenolic amides in other varieties of the Amaranthus genus (A. hypochondriacus and A. mantegazzianus). Hydroacetonic extracts have shown the greatest diversity of secondary metabolites. Acetone dissolves both hydrophilic or lipophilic components and is volatile, so 80:20 acetone-water can serve as an efficient extraction solvent.

The UV–VIS spectra of the hydroacetonic, methanolic and aqueous extracts of both Amaranthus species showed peaks suggesting the presence of flavonoids and betalains (Table 3). Flavonoids induced 250 to 285 nm and 320 to 385 nm peaks [19]. The peaks between 255 and 258 nm should be linked to the presence of phenolic acids [20]. A. cruentus showed a λ_max at 536.6 and 536.9 nm in HAE and ME, respectively. A. hybridus showed a λ_max at 536.9 and 537.2 nm in HAE and ME, respectively. These absorption peaks can be linked to the presence of betacyanins. HAE, ME and AE of the two species did not show characteristic λ_max values (450 to 470 nm) [1,21] related to betaxanthins. In vitro, the solvent property and neutral pH may alter the stability of the betaxanthins [22].

The results of DPPH• scavenging and iron III reducing activities by the two species extracts are summarized in Table 5 and Figure 3 below.

The radical scavenging activity by hydrogen or electron donation is a marker of antioxidant activity [33].The HAE extracts showed the highest free-radical scavenging activity for A. cruentus and A. hybridus with IC₅₀ values of 75.6 ± 0.5 and 56 ± 2 μg/mL, respectively. These activities could be attributed to hydrophilic antioxidants (phenolics) and lipophilic (carotenoids) antiradical substances [15,34]. In addition, RSC (radical scavenging capacity) activity decreased in the AE extracts. Amin et al. [25] obtained similar results. Samarth et al. [35] reported that A. cruentus leaves’ aqueous extract (at 40 °C) had an IC₅₀ value of 548 µg/mL using a DPPH assay. In our work, A. cruentus WE showed higher DPPH-radical scavenging activity with an IC₅₀ value of 330 µg/mL. Antioxidant activities of extracts using the DPPH assay correlated well with total flavonoids. The flavonoid antioxidant properties are linked to their ability to bind metallic ions, which can be involved in free-radical genesis [19]. Radical chain reactions are broken by flavonoid hydrogen donors and also by their capacity to regenerate α-tocopherol [19]. Flavonoids possess a remarkable spectrum of biochemical and pharmacological activities that have been attributed, in part, to their antioxidant and antiradical activities [24,36].

Iron reducing (from ferric to ferrous form) activity were 2.26 ± 0.01 mmol Ascorbic Acid Equivalent (AAE)/g in A. cruentus ME and 2.57 ± 0.07 mmol AAE/g for A. hybidus HAE. A. hybridus HAE (2.57 mmol AAE/g) followed by A. cruentus HAE and AE (2.51 mmol AAE/g) showed the highest activity in the FRAP assay. Antioxidant activity measured in A. cruentus and A. hybridus HAE and ME extracts obtained using DPPH, FRAP assays showed a comparable ranking of activity as reported by Thaipong et al. [15]. We observed that the AE of the two species showed high iron (III) reducing power comparable to HAE activity. Antioxidant activities from the FRAP assay of the extracts were well correlated with the total phenolics.

Extracts (HAE, ME and AE) of each species were tested for XO inhibitory activity at 100 µg/mL in the assay mixture [37]. XO inhibition percentages were 3.18 ± 1.10% and 38.22 ± 2.92% for A. cruentus (ME) and A. hybridus (HAE), respectively (Figure 4 below). Methanolic extracts showed the weakest XO inhibitory activity, specifically those from A. cruentus (3.18%).

A. hybridus showed the best activity and it may present health benefits in the prevention of uric acid accumulation. Polyphenols, flavonoids, and saponins are potent XO inhibitors [38,39]. Kaempferol and quercetin at 50 µg/mL concentrations have shown 85 and 90% XO inhibition, respectively [28]. HAE, containing the highest flavonol content (Figure 4), has shown the best XO inhibitory activity. Flavonols constitute bioactive compounds that may be responsible for XO inhibitory activities. As for the FRAP assay, XO inhibition presented a strong correlation with total phenolic contents.

Two commercial samples (1,500 g each species) of A. cruentus and A. hybridus were purchased for the study in September 2009 (average temperature of 29 °C) from a garden in Ouagadougou (Burkina Faso). The genotypes were cultivated on organic fertilized soils under watering conditions. The soil has a lateritic background. From young plants (one month old and 20 - 30 cm height vegetative growth stage), aerial plant parts (leaves and stems) of each species were harvested. The fresh materials were immediately flushed with water to remove sands and dusts. The plant materials were identified Pr. J. Millogo (botanist) and voucher specimens deposited at the herbarium of University of Ouagadougou. The air-dried (in laboratory conditions during 48 h) aerial parts of the plants were pulverized, and the powders packed in plastic bags were stored in the dark at laboratory’s temperature (25 °C).

Fifty grams of powder material were extracted with 20% aqueous acetone (500 mL) under mechanical shaking for 48 h at room temperature. Hydroacetonic extracts (HAE) were filtered, concentrated under vacuum and freeze-dried (Telstar Cryodos 50 freeze-dryer). Methanolic extracts (ME) were obtained by extracting the plant material in 1/10 (w/v ratio) for 24 h at room temperature. Methanolic extracts were filtered and concentrated with a rotary-evaporation apparatus (Büchi 461) at approximately 40 °C. The remainder of the solvent was removed in air to dryness. Aqueous extracts (AE) of herbal materials were obtained by boiling with distilled water (1/10, w/v) for 30 min and then freeze-dried. The extract residues were weighed prior to packing in waterproof plastic flasks and stored at 4 °C until use. The different crude extract weights and yields were calculated and expressed as grams of extract residues/100 g of dried plant materials.

Moisture was determined using the drying oven method as described by Odhav et al. [6]. Five grams of fresh aerial parts of each plant were dried in an oven at 105 °C for 3 h. To investigate the presence of secondary metabolites in the plant extracts, several methods were used. Flavonoids, tannins, saponins, coumarins, alkaloids, triterpenes and steroids (including cardenolids) were screened according the method of Ciulei [40]. Iridoids have been characterised by the method in Galvez et al. [27]. Betalain identification were performed on fresh plant materials (5 g) that were homogenized in a mortar and extracted with distilled water (25 mL) containing 250 mM ascorbate (water ascorbate extract: WAE) for 15 min. After centrifugation at 6000 rpm for 15 min, the supernatant was removed and stored at 4 °C. Aliquots of the supernatants were used to characterise betalains according to the colour changing method [41].

The thin-layer chromatography (TLC) of flavonoids [42] in the aqueous acetone extract was carried out on silica gel 60 F254 plates (Mercherey-Nagel) using ethyl acetate: formic acid: glacial acetic acid: water (10:1.1:1.1:2.6) as the eluent. Chromatograms were sprayed with diphenylboryloxyethylamin (1% in methanol) and polyethylene glycol 4000 (5% in ethanol) and visualised with a 365 nm lamp. Various flavonoids (rutin, quercetin, quercitrin, kaempferol, apigenin, genistein, chrysin, acacetin, luteolin and myricetin) were used as references.

UV/vis absorption spectra were collected between 200 and 700 nm with a CECIL CE 2041 UV/vis spectrophotometer.

Total polyphenols were determined by the Folin-Ciocalteu method as described by Singleton et al. [43]. Aliquots (125 µL) of extracts in a methanol solution (10 mg/mL) were mixed with Folin-Ciocalteu reagent (625 µL, 0.2 N). After 5 min, Na₂CO₃ aqueous solution (500 µL, 75 g/L) were added and the mixture was vortexed. After 2 h incubating in the dark at room temperature, the optical densities were measured at 760 nm against a blank (0.5 mL Folin-Ciocalteu reagent + 1 mL Na₂CO₃) with the spectrophotometer. The experiments were carried out in triplicate. A standard calibration curve was plotted using gallic acid (0 to 200 mg/L). The results were expressed as mg of gallic acid equivalents (GAE)/100 mg of extract.

The total flavonoids were estimated according to the Dowd method as adapted by Arvouet-Grant et al. [44]. A volume of AlCl₃ methanolic solution (0.5 mL, 2%, w/v) was mixed with methanolic extract solution (0.5 mL, 0.1 mg/mL). After 10 min, the optical densities were recorded at 415 nm against a blank (mixture of 0.5 mL methanolic extract solution and 0.5 mL methanol) and compared to the quercetin calibration curve (0 to 200 mg/L). The data obtained were the means of three determinations. The amounts of flavonoids in the plant extracts were expressed as mg of quercetin equivalents (QE)/100 mg of extract.

The contents of the flavonols were determined by the Almaraz-Abarca et al. [45] method. Aliquots were prepared by mixing ethanolic extract solutions (750 µL, 0.1 mg/mL) and AlCl₃ aqueous solution (750 µL, 20%, w/v). After 10 min of incubation, the optical densities were read at 425 nm against a blank (mixture of 750 µL ethanolic extract solutions and 750 µL ethanol). All determinations were carried out in triplicate. A standard calibration curve was plotted using quercetin (0 to 50 µg/mL). The results were expressed as mg of QE/100 mg of extract.

Total tannin contents were determined by the European Community reference method [46] using tannic acid as the standard. In a test tube, aqueous extract (200 µL), distilled water (1 mL), ammonium ferric citrate (200 µL, 3.5 g/L) and ammoniac (200 µL, 20%) were mixed. After 10 min, the optical densities of the samples were measured at 525 nm against a blank (200 µL aqueous extract + 1.2 mL distilled water). The data are the mean of three determinations. The results were expressed as mg of tannic acid equivalents (TAE) per 100 mg of extract (mg TAE/100 mg extract).

For the determination of betalain contents, the fresh materials were prepared similar to that for betalain detection. Betalains were quantified photometrically at 536 nm (betacyanins) [47] and 475 nm (betaxanthins) [48] using the literature-described molar extinction coefficients (5.66 10⁴ l.cm^-1.mol^-1 for amaranthin, 4.8 10⁴ l.cm^-1 mol^-1 for indicaxanthin). All experiments were performed in triplicate. Betalain contents were calculated using the following formula:

      C = A×V× MW × DF × 100 / [ε × L × W × (1 − M)]

where C is the betalain content (g /100 g dr. wt.); V is the extract volume (l); MW is the molecular weight; DF is the dilution factor; ε is the molar extinction coefficient (726.6 g.mol^-1 for amaranthin; 308 g.mol^-1 for indicaxanthin; L is the path length (1 cm); W is the weight of the fresh material (g); and M is the moisture (%).

The XO activity, with xanthine as a substrate, was measured spectrophotometrically according the method of Ferraz Filha et al. [37] with slight modifications. Extracts were screened for XO inhibitory activity at a final concentration of 100 µg/mL. The assay mixture contained phosphate buffer (150 µL, 1/15 M, pH 7.5), extract solution (50 µL) and enzyme solution (50 µL, 0.28 U/mL in phosphate buffer). The reaction was initiated by adding substrate solution (250 mL, 0.15 mM in water). The absorbance stability at 295 nm was verified for 1 min. Enzymatic kinetics were recorded at 295 nm for 2 min. A negative control (0% XO inhibition activity) was prepared with 1% methanol solution instead of the extract solution. All experiments were performed in triplicate. XO inhibitory activity was expressed as the percentage inhibition of XO, calculated as:

      % inhibition = (1− V/V₀) × 100

where V₀ is the linear change (blank inclination) in absorbance per minute of negative control, and V is the linear change (test inclination) in absorbance per minute.