Influence of extraction technique on yield and antioxidant activity of extracts from Moringa oleifera leaf

Abstract The article presents research on the exploring of extraction process of biologically active substances from the leaves of the Moringa oleifera tree using ethanol. Ethanolic extracts were obtained using three different techniques: maceration with shaking, ultrasound-assisted extraction and extraction in Soxhlet apparatus, in different time variants: 1, 2 and 4-hours. After solvent evaporating and drying, the yields of dry extracts obtained in particular processes were calculated. The antioxidant activity of extracts was analyzed spectrophotometrically using DPPH radical scavenging method, and total phenolic content (TPC) was determined by Folin-Ciocalteu method. By means of gas chromatography with mass selective detector (GC-MS), 11 biologically active compounds present in ethanolic extracts were identified, among which α-tocopherol had the greatest share. Based on the results, the influence of the extraction technique and time on the yield and antioxidant activity of M. oleifera leaf extracts were discussed.


INTRODUCTION
Moringa oleifera (M. oleifera) is the most widely known and cultivated tree from the family of Moringaceae. Its popular names are drumstick tree or horseradish tree, but also miracle tree, tree of life or a wonderful plant -due to the numerous benefi ts resulting from its use in both nutrition and natural therapies 1-3 . It originates from sub-Himalayan areas of India, Pakistan, Bangladesh and Afghanistan but currently is cultivated in many tropical countries 4 . Due to its valuable properties, it has been used by the indigenous people of Asia and Africa as a foodstuff and natural medicine. The leaves, seeds, pods and fl owers of M. oleifera are traditionally used for its nutritional properties. The plant is rich in phytosterols, polyphenols, fatty acids, valuable amino acids, contains vitamin C, B, D, E, minerals including iron, calcium, and zinc 5, 6 . Moringa extracts are used in traditional folk medicines to cure various diseases 5 . Moreover, seeds of the plant are also used as eff ective eco-coagulant for water purifi cation 7, 8 . Recently, the popularity of M. oleifera in Europe is increasing. It is classifi ed into a group of adaptogenic plants -plants which can eff ectively increase resistance to stress and help to keep the body in homeostasis 9 .
Leaves of M. oleifera in various forms, including raw material, juice and extracts, were subject to diff erent studies in vitro and in vivo. They are known for increasing breast milk production during lactation due to the content of phytosterols such as stigmasterol and sitosterol 5, 10 . They also possess antidiabetic and antioxidant properties 5, 11, 12 . Studies on rats have shown that powdered M. oleifera leaves may eff ectively prevent renal impairment by increasing the total protein content in plasma and reducing urea and creatinine levels 13 . Extracts from M. oleifera leaves have also cytotoxic activity and signifi cant anticancer potential 1, 3, 14-16 . It was confi rmed that they decrease cell motility and colony formation in colorectal and breast cancer cell lines 14 and have apoptotic eff ect against prostate cancer cells 16 . Moreover, hydroethanolic leaf extracts signifi cantly improved memory and reduced neurodegeneration in rats. Neuroprotective and memory enhancing eff ects may be the result of oxidative stress lowering and reduction of acetylcholine esterase activi-ty, improving cholinergic function 17 . Gastroprotective activity of M. oleifera leaf extract against aspirin-induced ulcers, with evidence of mucus membrane enhancing activity, was also confi rmed 18 . The antimicrobial activity of M. oleifera leaves and other tissues is also promising, and suggests the potential use of this plant in the control of diff erent pathogenes, including Gram-negative and Gram-positive bacteria (e.g. Staphylococus aureus, Enterococcus faecalis, Aeromonas caviae) and popular viruses like Epstein-Barr virus (EBV) and herpes simplex virus (HSV) 19 . M. oleifera is also traditionally used in the treatment of HIV symptoms, possibly by improving the immune system 19 . The antiretroviral eff ect of powdered moringa leaves in combination with the root or bark was noticed during observation of HIV infected people in Zimbabwe 20 .
Due to the content of numerous biologically active compounds, M. oleifera is increasingly used as a diet supplement and also tested for using a potential drug in diseases such as stomach ulcers, Alzheimer's disease, microbial diseases or even cancer 3, 5, 17, 18 . Extracts from moringa leaves, possessing high antioxidant activity, can be used not only as a dietary supplement but also as active additives in cosmetics, which may help to protect skin against damage caused by free radicals and reduce the signs of aging. Extracts can also have a protective function in cosmetic formulations, acting as eff ective antioxidants.
Most available in the literature reports on the extraction of active substances from moringa leaves are diffi cult to compare because of using plant materials from diff erent countries and applying of various research methodologies. Therefore, the objective of the present study is to investigate the infl uence of extraction conditions on the yield, antioxidant activity and total phenolic contents of dry ethanolic extracts from M. oleifera leaf. Active compounds in extracts were analyzed using gas chromatography with mass selective detector (GC-MS) method.

Extraction techniques
The aim of the research was comparison of different extraction techniques to determine the favorable conditions for obtaining of dry ethanolic extracts from M. oleifera leaf and to study their antioxidant activity. Studied plant material was powdered dry leaves of M. oleifera originated from India (producer: Minvita). Extracts were obtained using 96% ethanol by various techniques: maceration with constant shaking (MS), ultrasound-assisted extraction (UAE) and Soxhlet extraction (SE), in different time variants: 1, 2 and 4 hours. In each condition, three parallel experiments were carried out using 5 g of plant material and 150 cm 3 of 96% ethanol. After extraction and fi ltration through a Filtrak No. 390 paper, the solvent was evaporated using a rotary evaporator. The obtained concentrated ethanolic extracts were transferred to the watch glasses and left in a dark place at room temperature for 48 hours to evaporate the residual solvent. Next, it was dried using a laboratory dryer with air circulation at 35 o C for 6 hours. After drying, the extracts were stabilized to constant weight in a desiccator with silica gel and weighed. For each extraction conditions the yield of the dry extract [mg/g] was calculated as an average of three parallel determinations.

Antioxidant activity
The antioxidant activity of the obtained extracts was measured using DPPH radical scavenging method. For this purpose, 0.01 g/cm 3 solutions of each dry extract in methanol were prepared and then diluted with methanol to obtain working solutions with concentrations in a range of 100-800 g/cm 3 . Directly before the analysis 0.002 mmol/ cm 3 stock solution of DPPH (2,2-diphenyl-1-picrylhydrazyl) in methanol was prepared and diluted ten times with methanol to obtain a DPPH working solution. For determination of antioxidant activity, to 1.5 cm 3 of the extract solution, 3 cm 3 of DPPH working solution was added, mixed and left for 30 minutes incubation in darkness. A blank sample containing 1.5 cm 3 of solvent was prepared analogously. The analyses were carried out using a 1600PC UV-VIS spectrophotometer (VWR) in 1-cm cuvettes, by measurement of the absorbance at 517 nm, with methanol as a reference. The radical scavenging activity (RSA) of particular extracts was calculated from the obtained values of extract sample absorbance after 30 minutes (A 30 ) and blank sample absorbance (A 0 ), using the following formula 21 : Next, for the particular extracts obtained using diff erent techniques, the plots of RSA (%) versus working solutions concentrations C (g/cm 3 ) were prepared and the mathematical equations of these dependencies were determined. From the obtained equations, for the RSA = 50%, the concentrations of extracts causing a 50% inhibition of free radical activity (IC 50 ) were calculated. The IC 50 values are inversely correlated with the radical scavenging activity and antioxidant properties of the sample.

Total phenolic contents
Total phenolic contents in dry M. oleifera extracts were determined by the method with Folin-Ciocalteu (F-C) reagent 22 . For this purpose, 0.5 cm 3 of the extract solution in methanol (C = 5 mg/cm 3 ), 0.5 cm 3 of F-C reagent (Chempur) and 1.5 cm 3 of sodium carbonate solution (C = 200 mg/cm 3 ) were placed in volumetric fl ask and made up to 25 cm 3 with redistilled water. The content of the fl ask was thoroughly mixed and kept at room temperature for 30 min for blue color development, shaking occasionally. After this time the absorbance was measured by using UV-VIS 1600PC spectrophotometer (VWR) in 1-cm cuvettes at 760 nm wavelength. Gallic acid (Sigma-Aldrich) was used as a reference standard (0.05 -0.50 mg/cm 3 ) for calibration curve preparation. The total phenolic contents (TPC) were calculated using a linear regression equation obtained from the calibration curve of gallic acid and expressed as mg of gallic acid equivalent per 1 g of dry extract (mg GAE/g).

GC-MS analysis
The analysis of the obtained extracts was carried out by GC-MS method using a 6890N gas chromatograph with a 5973 Network Mass Selective Detector (Agilent Technologies). Separation was performed using HP--5MSI capillary column (Agilent 19091S-433I: 5%-Phenyl 95%-Methylpolysiloxane Inert column, 30 m x 0.25 mm x 0.25 μm) with the following temperature program: from 80 o C to 320 o C, at a rate of 5 o C/min. The carrier gas was helium (1.2 cm 3 /min). Tested samples (2.0 μl of dry extract solutions, C = 10 mg/cm 3 ) were dosed to the column in a split mode (10:1) using a 7683 Series Injector Autosampler. Electron impact ionization (70 eV) mass spectra were obtained and recorded in the range of 20-600 m/z. The detector temperatures were respectively: quadrupole 150 o C, ion source 230 o C.
Identifi cation of the individual compounds present in the obtained extracts was carried out by comparison of their mass spectra with mass spectra of standards from the NIST 02 library. The identifi cation was confi rmed by comparison of the calculated linear retention indices (LRI) with the values found in the literature and also by comparison of retention times with standards when the standards were available. In order to determine linear retention indices, the standard mixture of the C 7 -C 40 n-alkanes was analyzed under the same chromatographic conditions 23, 24 . The quantitative analysis was performed by the internal normalization method. The relative contents of particular compounds identifi ed in extracts using GC-MS method were evaluated as the percentages of a peak area in a total ion chromatogram (TIC) using the MestReNova 10.0.2 software.

Statistical analysis
The statistical analysis was performed using TIBCO Statistica 13.3 (TIBCO Software Inc.) and Microsoft Excel 2016 (Microsoft). The assays were performed in triplicate and the results were expressed as mean values ± s (standard deviation). To analyze the antioxidant parameters of the extracts, experimental data were subjected to analysis of variance (ANOVA) and the differences among mean values were evaluated by Tukey's HSD post-hoc test at a 5% signifi cance level.

Results and discussion
For all studied extraction techniques, but especially in Soxhlet extraction (SE), the effi ciency of the processes increased with prolongation of the time. The ultrasound--assisted extraction (UAE) was the most effective process in all tested time variations. Conducting ultrasound--assisted extraction for 4 hours allowed to obtain the highest yield among all applied processes (198.0 mg/g). In the case of SE, the 1-hour process was not effective, because the obtained yield was the lowest among all processes (109.9 mg/g). Increasing the time of SE signifi cantly improved the yield and allowed to obtain in a 4-hour process almost the same results (197.4 mg/g) as for a 4-hour UAE. Dependence of the obtained yields of dry extracts [mg/g] on the type and duration of the isolation process is shown in Figure 1.
lowest IC 50 concentration (388.1 g/cm 3 ) and the highest TPC value (76.3 mg GAE/g). In the case of ultrasound-assisted and Soxhlet extraction, some decrease of the antioxidant activity and total phenolic content was observed with the prolongation of the extraction time. It may be caused by possible decomposition of some biologically active compounds under the infl uence of ultrasounds and elevated temperature. Oppositely, time extension in maceration with shaking technique led to increasing of antioxidant activity and contents of phenolics. However, for MS technique the yield, antioxidant activity and TPC values obtained even in the longest time variant (4h) do not reach the results obtained for 1h UAE experiment.
Analyzing the available literature data and comparing them with data obtained in the presented study it can be seen that they show a quite big variation. It may result from the use of plant materials of various origin and diff erent extraction techniques and conditions. Some results obtained by other authors indicate signifi cantly lower antioxidant activity of the extracts. For example, Vats and Gupta 25 received extracts from M. oleifera leaf originated in India using 95% ethanol by maceration with shaking for 24 hours, for which the IC 50 value was 610 μg/cm 3 and total phenolic content TPC = 9.58 mg GAE/g. Extracts obtained at present work using the same technique and solvent and raw material from the same country show signifi cantly higher antioxidant activity and total phenolic content, despite the use of much shorter extraction time (from 1 to 4 hours). In turn, Wright et al. 26 studied M. oleifera leaf extract from plant material derived from Jamaica by means of 24-hour extraction in a Soxhlet apparatus using 80% ethanol, obtaining much lower antioxidant activity with a value of IC 50 = 832.8 μg/cm 3 . Higher values were reported by Vongsak et al. 27 , who studied the extraction process of M. oleifera leaf from Thailand using several diff erent techniques, including simple maceration, Soxhlet extraction, and percolation using 50 and 70% ethanol. The highest DPPH scavenging activity was found for extracts obtained with 70% ethanol during a 20-hour extraction in a Soxhlet apparatus (IC 50 = 55.07 μg/ cm 3 ) and during 72-hour simple maceration (62.94 μg/cm 3 ).
Applying of GC-MS method enabled the identifi cation of 11 compounds present in M. oleifera leaf extracts. On the GC-MS chromatograms of extracts obtained using different techniques relative contents of particular compounds were similar. In Figure 2 chromatogram of the extract obtained using 96% ethanol during 1-hour ultrasound-assisted extraction is presented, while in Table 2 the retention times, retention indices and relative contents of identifi ed compounds are summarized.  The results of antioxidant activity for all obtained extracts, determined by DPPH radical scavenging method and expressed as IC 50 values, are collected in Table 1, together with the results of total phenolic contents (TPC) analyses. The statistical analysis has shown that in most cases observed differences in antioxidant activity and total phenolic content for the particular techniques of extraction were signifi cant, which was indicated by using different letters.
The antioxidant activity of the extracts expressed by the IC 50 parameter varied in a range of 388-434 g/cm 3 . The results are in good correlation with the total phenolic contents (TPC), which increase with the lowering of IC 50 values, as expected. The extract obtained in 1-hour ultrasound-assisted extraction showed the highest activity, represented by the The main component, with share accounted for over 50%, was α-tocopherol -an important biologically active substance with strong antioxidant properties, necessary for the proper functioning of the human body. The other vitamin E analogue, β-tocopherol, was also detected but in small amounts (<1%). The analyzed extracts contained signifi cant amounts of sterols (β-sitosterol, fucosterol), pentacyclic triterpenoids (β-amyrine, lupeol, 24-methylenecycloartanol) and phytol. Small amounts of aliphatic hydrocarbons: pentacosane, heptacosane and nonacosane were also found. Both tocopherols, sterols and pentacyclic triterpenoids are widely distributed in plants and are known to possess a number of bioactive properties and pharmacological effects. β-Amyrin and lupeol are bioactive compounds often found in medicinal plants, known for their anti-infl ammatory, analgesic, anti-ulcerogenic, and hypoglycemic properties 28, 29 . Lupeol is also known as an effective antioxidant and anticancer agent, anticancer properties were also reported for 24-methylenecycloartanol 29, 30 . The presence of 24-methylenecycloartanol in M. oleifera is reported for the fi rst time.

CONCLUSIONS
Comparing the effi ciency of 3 various isolation techniques conducted in diff erent time variants led to the conclusion, that extraction conditions signifi cantly aff ect the yields of the ethanolic extracts from M. oleifera leaf. The highest yields of dry extracts were obtained in 4-hour ultrasound-assisted extraction and Soxhlet extraction processes (~198 mg/g), while the least eff ective was 1-hour Soxhlet extraction (~110 mg/g). The antioxidant activity of the extracts, expressed by the IC 50 parameter, varied in the range of 388.1-433.5 g/cm 3 , while total phenolic content from 66.9 to 76.3 mg GAE/g. The highest antioxidant activity and the highest content of phenolics were found for the extract obtained by means of 1-hour ultrasound-assisted extraction. Among the compounds identifi ed in extracts by using the GC-MS method, the main component was α-tocopherol with relative content over 50%. Other compounds found in smaller amounts, with antioxidant and other biological properties were: β-sitosterol, fucosterol, β-amyrin, lupeol, 24-methylenecycloartanol, phytol and β-tocopherol.
The research carried out using a uniform methodology, in relation to the same raw material, enabled to indicate the most preferred extraction conditions. Among all studied conditions, as the most favorable process 1-hour ultrasound--assisted extraction can be pointed out. This process was characterized by the highest extraction effi ciency and allowed to obtain dry extract with the best antioxidant properties and the maximum total phenolic content in the shortest time. The obtained dry extracts of M. oleifera leaf proved to be a source of valuable biologically active compounds which could be potentially used in pharmacy. These extracts can also fi nd application in cosmetics formulations for protection against reactive oxygen species and as an active anti-infl ammatory agent. Figure 2. GC-MS chromatogram of M. oleifera leaf extract obtained in 1-hour ultrasound-assisted extraction (UAE) using 96% ethanol. Identifi ed compounds: 1) phytol, 2) pentacosane, 3) heptacosane, 4) nonacosane, 5) β-tocopherol, 6) α-tocopherol, 7) β-sitosterol, 8) fucosterol, 9) β-amyrin, 10) lupeol, 11) 24-methylenecycloartanol Table 2. Retention parameters and relative contents of compounds identifi ed in M. oleifera leaf extract obtained in 1-hour ultrasound-assisted extraction using 96% ethanol RT -retention time; LRI exp -linear retention index determined on the HP-5MSI capillary column; LRI lit -retention index from the literature 23, 24 ; Relative content -percentage of the peak area in the total ion chromatogram; s -standard deviation (n = 3)