Table 1 presents the proximate composition and calcium content of the different parts (leaf, stem and stalk) and harvest seasons (summer and winter) of Moringa. Crude protein content of leaves was 24.42–25.29% suggesting that leaves are a good source of protein. Among the summer samples examined, leaves had the highest content of crude protein, crude fat, and ash. The proximate compositions were similar for summer and winter samples, except crude fat and ash for stalk. Similar results of proximate composition for leaves were reported by Gupta et al. [11] and Lowell [12].

The crude protein content (based on wet basis) of Moringa leaves (5.4%) is higher than protein content of alfalfa sprout (3.7%), sweet potato leaves (3.3%), and mung bean sprout (3.1%) being high protein content vegetables usually consumed in Taiwan [13]. According to Hassan and Umar [14], plant food that provides more than 12% (dry basis) of its caloric value from protein is considered a good source of protein. Therefore, Moringa leaves (24.42–25.29% protein) not only meet but even double this requirement.

Moringa leaf also is a good plant source of fat. The crude fat content (1.19–2.77% wet basis) of Moringa leaf is higher compared to reported values (<1.0% wet basis) in most vegetables consumed in Taiwan [13]. Islam et al. [15] investigated fifteen different kinds of leafy vegetables available in Bangladesh and found the highest fat content was observed in Moringa leaves.

Ash content was 8.53–11.0% in Moringa leaves which was lower than the investigation of Moringa leaves (15.09% DW) by Lockett et al. [16].

Calcium builds healthy bones and teeth and assists in blood-clotting. The seasonal effects on calcium content were different. For winter sample, stalk has highest calcium content while the lowest was stem (Table 1). On the other hand, highest calcium content of Morniga for summer sample was leaf while the lowest was stalk. In Taiwan, the leaves and stems of Moringa are used as vegetables and the stalks are used for soup or stew.

The result showed that the macronutritients may not be affected by season; however, the micronutritients may be affected by season. The mineral content may change during different seasons, which induce significantly different calcium and ash contents between summer and winter. The fact that season’s influence the chemical composition of food is well known. Salvador et al. [17] studied the influence of extraction system, production year and area on the chemical compositions of Cornicabra virgin olive oil of five crop seasons and found the crop season was a critical variable.

Methanol was chosen for extraction in this study because it has wide solubility properties for low molecular and moderately polar substances, including the antioxidant-active phenolic compounds. The extraction ratio, scavenging effect on DPPH radicals and total phenolic compounds of methanolic extracts of Moringa samples from different parts and seasons are shown in Table 2. The trend of scavenging effect on DPPH radicals as a function of the part was: leaf > stem > stalk for both of the seasons’ samples investigated. Meanwhile, the highest content of total phenolic compounds was found in leaves for samples from both seasons investigated. The result also showed that the scavenging effect on DPPH radicals was higher in winter than in summer.

One important mechanism of antioxidation involves the scavenging of hydrogen radicals. DPPH has a hydrogen free radical and shows a characteristic absorption at 517 nm. After encountering the proton-radical scavengers, the purple color of the DPPH solution fades rapidly. The extracts of Moringa are able to reduce the unstable radical DPPH to the yellow-colored diphenylpicrylhydrazine. A dose-response relationship was found in the DPPH radical scavenging activity in this study; the activity increased as the concentration increased for each individual sample. The involvement of free radicals, especially their increased production, appears to be a feature of many diseases including cardiovascular disease and cancer [18]. Phenolic compounds of the extracts are probably involved in their antiradical activity.

The content of total phenolic compounds for the stem and stalk were higher from winter than from summer samples, however, similar contents for the leaves from summer and winter (Table 2). The result was similar to the result of Iqbal and Bhanger [19] who studied the effect of season and production location on antioxidant activity of Moringa leaves grown in Pakistan. This may be due to the fact that Moringa leaves grow in June and mature from December to March and phenolic content is the lowest in newly opened leaves, increasing gradually with the maturity of leaves [19]. In this study, the total phenolic content of the selected samples was determined, however, the phenolic composition of the extracts was not analyzed as it was not within the scope of the present investigation.

Phenolic compounds have an important role in stabilizing lipid oxidation and are associated with antioxidant activity because of their scavenging ability due to their hydroxyl groups [20]. It was determined that there were 181.3–200.0 mg, 71.9–134.4 mg and 68.8–93.8 mg catechin equivalent of phenolic compounds in the 100 g of the leaf, stem and stalk of Moringa, respectively (Table 2). These results indicated that there was no correlation between antioxidant activity and total phenolic content (p > 0.05). However, different results were reported on this aspect; some authors found correlation between phenolic content and antioxidant activity [21], whereas the others found no such relationship, since other compounds are responsible for the antioxidant activity [22,23].

Although the phenolic compounds are believed to be the major phytochemicals responsible for antioxidant activity of plant materials [24], Moringa is a rich source of ascorbic acid which is also has the antioxidant activity [25]. To estimate the antioxidant activity contributed from ascorbic acid of Moringa samples, the solvent fraction procedure was performed.

A regular pattern of increase in reducing power as a function of extract concentration of winter’s samples of Moringa was observed (Figure 1). However, there were no significant differences among the three parts of Moringa. Siddhuraju et al. [26] showed that the reducing power of bioactive compounds is directly related to ascorbic acid.

Earlier authors [27,28] have observed a direct correlation between antioxidant activity and reducing power of certain plant extracts. The reducing properties are generally associated with the presence of reductones, which have been shown to exert antioxidant action by breaking the free radical chain by donating a hydrogen atom [25]. Reductones are also reported to react with certain precursors of peroxide, thus preventing peroxide formation. From our data on the reducing power of the tested extracts, we suggest that it is likely that reducing power contributed significantly to the observed antioxidant effect.

Hydrogen peroxide scavenging activities of Moringa extracts are shown in Figure 2. All the Moringa samples were capable of scavenging H₂O₂ in a concentration dependent manner. As shown in the results, remarkable scavenging effects of the Moringa extracts were observed in hydrogen peroxide scavenging assay of all samples, but the stalk showed the least effect. As shown in Table 3 the Moringa extracts showed an EC₅₀ for hydrogen peroxide scavenging effect ranging between 280 and 340 μg/mL, whereas the EC₅₀ for ascorbic acid was found to be ca. 160 μg/mL. This indicates that all the Moringa extracts tested have substantial H₂O₂-scavenging activity, although less than ascorbic acid. Therefore, the reduced amount of hydrogen after this reaction may account for the H₂O₂-scavenging effect of the Moringa extracts observed in the present study.

Hydrogen peroxide can be formed in vivo by many oxidized enzymes such as superoxide dismutase. Being a non-radical oxygen-containing reactive agent, it can form hydroxyl radical, the most highly reactive oxygen radical known, in the presence of transition metal ions and participate in free-radical reaction [29]. It can cross membranes and may slowly oxidize a number of compounds. Hydrogen peroxide itself is not very reactive, but it can sometimes be toxic to cells because it may give rise to hydroxyl radical in the cells. Thus, removing hydrogen peroxide as well as superoxide anion is very important for protection of food systems.

The SOD activity and ascorbic acid content of Moringa extracts are also shown in Table 3. The trend of hydrogen peroxide scavenging activity, SOD activity and ascorbic acid content were similar among the Moringa samples. However, the correlation coefficient between hydrogen peroxide scavenging activity and SOD activity was 0.9410 (p < 0.05), and between hydrogen peroxide scavenging activity and ascorbic acid content was −0.7988 (p > 0.05). Therefore, SOD may affect the hydrogen peroxide scavenging activity of the Moringa samples.

SOD catalyzes the dismutation of superoxide radical in a broad range of organisms, including plants. The dismutation of superoxide into hydrogen peroxide and oxygen constitute the first line of cellular defense to prevent undesirable biological oxidation by oxygen radical generated during cellular metabolism [30]. In plant cells one of the most important detoxification systems is the water-water cycle (WWC) which operates together with SOD as a mechanism of hydrogen peroxide scavenging in intact chloroplasts [31]. The most important function of this cycle is a rapid, immediate scavenging of O₂ and H₂O₂ at the site of its generation prior to their interaction with the target molecules. Ascorbate peroxidase (APX) uses two molecules of ascorbate to reduce H₂O₂ to water, with the concomitant generation of two molecules of monodehydroascorbate (MDHA). MDHA is a radical with a short lifetime, which is reduced directly to ascorbate within the chloroplast at the thylakoid membrane [32].

Moringa oleifera was purchased from Spring Autumn Co. in Taichung, Taiwan, and were separately harvested in July 2004, and January 2005. After harvesting, the samples were divided into leaf, stem, and stalk by hand and followed by freeze drying. The freeze dried materials were then ground into powder, and screened through a 20-mesh sieve (aperture, 0.94 mm) and stored at −80 °C.

Crude protein, crude fat and ash of freeze dried Moringa samples were determined according to AOAC Methods 955.04, 920.39 and 930.05 [33], respectively. Calcium was determined as outlined in AOAC Methods 921.01 [33]. Analysis for calcium was carried out using atomic absorption spectroscopy set at 422.7 nm. The burner height was manually adjusted on the instrument to ensure maximum absorption. The composition of all samples had two determinations per replicate.

Each sample powder (20 g) was extracted with 200 mL of methanol stirred on a stirring plate at room temperature for 24 h. Contents were filtered through #1 filter paper (Whatman Inc., Hillsboro, OR). The filtrate was concentrated to dryness in vacuo to obtain methanolic extract. The dried filtrates were weighed to determine the extraction ratio of soluble constituents.

The methanolic extract was then stored at −20 °C to determine the scavenging effect on α,α-diphenyl-β-picrylhydrazyl (DPPH) radicals. The methanolic extract was also used to determine the total phenolic compounds, ascorbic acid content and reducing power.

The experimental method in the current study was according to Kuo et al. [24]. A 400 μM solution of DPPH was prepared in 100% methanol. 50 μL of samples (methanolic extract or various solvent fractions, at final concentration 0–12,000 μg/mL) and 150 μL of DPPH solution were added to each well in a 96-well flat-bottom EIA microtitration plate. After thorough mixing, the solutions were kept in the dark for 90 min. The absorbency of the samples was measured using an Optimax automated microplate reader (Molecular Devices, Toronto, Canada) at 517 nm against methanol without DPPH as the blank reference. Each sample was tested four times and the values were averaged. For the determination of EC₅₀ (which is the efficient concentration of antioxidant defined as the 50% of the initial DPPH concentration), each sample was measured at least at five different concentrations in the DPPH test. EC₅₀ was obtained by interpolation from linear regression analysis.

Total phenolic compounds were measured with Folin-Ciocalteu reagent using catechin as a standard [34]. A 5 mL of Folin-Ciocalteu reagent (diluted tenfold in distilled water), 2 mL of 200 g/L sodium bicarbonate, and 2 mL of distilled water were added to 1 mL of the raw methanolic extract of Moringa samples. After 1 h at 20 °C, the absorbance at 755 nm was read.

Reducing power was determined following the method reported by Oyaizu [35]. Each 5 mL of methanolic extracts (0.2, 0.4, 0.6 and 0.8 mg/mL) was mixed with phosphate buffer (5.0 mL, 2.0 M, pH 6.6) and 1% potassium ferricyanide (5 mL), and the mixtures were incubated at 50 °C for 20 min. A 5mL of 10% trichloroacetic acid was added and the mixture was centrifuged at 650× g for 10 min. The upper layer of the solution (5 mL) was mixed with distilled water (5 mL) and 0.1% ferric chloride (1 mL), and absorbance was measured at 700 nm. The experiment was conducted in triplicate and results were averaged.

Hydrogen peroxides were measured using the horseradish peroxidase assay [36]. A 1 mL of Moringa extract sample was first mixed with 400 μL of 4 mM H₂O₂ solution and allowed to incubate for 20 min at room temperature. Then 600 μL of phenol red solution (7.5 mM phenol red and 500 μg/mL horseradish peroxidase in 100 mM phosphate buffer) was added to the reaction mixture. After 10 min, the absorbance was measured at 610 nm. Sample absorbances were converted to mM H₂O₂ by interpolating from a standard curve.

Determination of SOD activity was performed by using a kit (Ransod, Randox Labs. cat. no. SD 125, Antrim, UK) based on the method developed by McCord and Fridovich [37]. A 5 μL of Moringa extract sample was added concomitantly with the main reagent (170 μL) to the cuvette. Absorbance was monitored at 500 nm for 150 s after addition of xanthine oxidase (25 μL plus 10 μL of H₂O) as start reagent. Read initial absorbance (A1) after 30 s and start timer simultaneously. Final absorbance (A2) was read after 3 min.

All standard rates and sample diluents rates must be converted into percentages of the sample diluents rate, and subtracted from 100% to give a percentage inhibition.

Each sample was tested four times and the values were averaged. For the determination of EC₅₀, each sample was measured at least at five different concentrations in the test. EC₅₀ was obtained by interpolation from linear regression analysis.

Ascorbic acid was measured by a Merck, RQ flex plus (Darmstadt, Germany) spectrophotometer [38]. A 1 mL of Moringa extract sample was diluted to properly concentration before measurement. An ascorbic acid test strip (25–450 mg/L) was immersed in the liquid for 5 s before leaving a further 55 s and reading the change in color intensity using a RQ Flex plus spectrophotometer.

Data were analyzed by analysis of variance (ANOVA) using general linear model. Duncan’s multiple range test was used to determine the differences among samples. Significant levels were defined as probabilities of 0.05 or less. All processing treatments were done in triplicate.