Antioxidants properties of Murraya koenigii: a comparative study of three different extraction methods

Parithy, M.T., Mohd Zin, Z., Hasmadi, M., Rusli, N.D., Smedley, K.L. and Zainol, M.K. Faculty of Fisheries and Food Sciences, Universiti Malaysia Terengganu, 21030, Kuala Nerus, Terengganu, Malaysia Faculty of Food Science and Nutrition, Universiti Malaysia Sabah, 88400 Kota Kinabalu, Sabah, Malaysia Faculty of Agro Based Industry, Universiti Malaysia Kelantan, Jeli, Kelantan, Malaysia Institute of Food Security and Sustainable Agriculture, Universiti Malaysia Kelantan, Jeli, Kelantan, Malaysia Stratford School, Stratford-upon-Avon, Warwickshire, CV37 9DH, United Kingdom


Introduction
In recent times, medicinal plants are on a rave due to their variant physicochemical properties. These natural products have been shown to have antioxidant properties. They are capable of scavenging free superoxide radicals, thus providing anti-ageing benefits as well as reducing the risk of cancer (Ghasemzadeh et al., 2014). A significant role of dietary phytochemicals human health is mainly to minimize oxidative damage to living cells by deactivating reactive oxygen species (ROS), the byproducts produced during normal cell aerobic respiration (Dharmaraja, 2017). Many studies have been carried out on many local herbs in Malaysia; however, there are still some other herbs such as M. koenigii leaves that remain unexplored in depth and more explorations are needed (Azizah et al., 2014).
The leaves of Murraya koenigii have been widely used in Indian cookery for centuries and have a versatile role to play in traditional medicine (Jain et al., 2012). The M. koenigii leaves are notable for their antitumor, antioxidant, anti-inflammatory, anti-hyperglycemic, and hypoglycemic properties (Dineshkumar et al., 2010). The leaves have a slightly pungent, bitter and weakly acidic taste. They also retain their flavour and other qualities even after drying (Sinha et al., 2012). M. koenigii leaves are generously credited with tonic and stomachic properties. The bark and roots are used as a stimulant and topically to cure eruptions and bites of poisonous animals (Singh et al., 2014). The fresh leaves, dried leaf powder, and essential oils are widely used for flavouring soups, curries, fish and meat dishes, eggbased dishes, traditional curry powder blends, seasoning and other ready to use food preparations (Jain et al., 2012). Traditionally, M. koenigii leaves are boiled with coconut oil until they are reduced to a blanked residue which is then used as an excellent hair tonic for retaining natural hair tone and stimulating hair growth.
In the nutraceutical industry, the extraction process is an important step for the isolation of phytochemicals from herbs and spices (Bak et al., 2012). However, there are some disadvantages with the use of certain types of extraction method such as environmental pollution, lower yields, loss of reactivity, and others. Conventional solvent extraction attracts a higher cost, requires a longer time and is inefficient (Luque-Garcia and Luque de Castro, 2003), therefore alternative extraction needs to be explored to reduce extraction time, reduce solvent consumption, increase extraction yield and improve extract quality. This problem could be hindered by the use of an efficient extraction technique to preserve the beneficial properties of the extract. A comparison with other extraction techniques to obtain beneficial compounds proved to be crucial and important (Medina-Torres et al., 2017). Ultrasound-assisted extraction (UAE) and microwave-assisted extraction (MAE) are recommended for the extraction of analytes from different matrices. The advantages of UAE and MAE are low equipment cost, low energy requirements, reduced solvent quantity and/or time consumption (Jacotet-Navarro et al., 2016). Therefore, the purpose of this study was to determine the antioxidant properties and flavonoid content in M. koenigii leaves extracted using different extraction procedures namely, solvent assisted extraction SAE), microwave assisted extraction (MAE) and ultrasonic assisted extraction (UAE).

Materials and methods
All samples were dried using the oven drying method (AOAC, 2007) prior to all extraction techniques. The samples were then ground and sieved (10-20 mm size) (Waring Commercial, Torrington. CT, U.S.A) and kept in a dark bottle, stored at 5 o C until further analysis.

Solvent assisted extraction
Solvent assisted extraction was conducted using 60% ethanol using a modified method suggested by Zainol et al. (2003). Approximately 10 g of the finely ground curry leaf powder was added with 100 mL of 60% ethanol and left to stand for 1 hr. The solvent was then removed leaving the residue for the second extraction. The solvent was filtered using Whatman No. 4 filter paper using a vacuum pump. Post filtration the filtrate was then concentrated using a rotary evaporator (Buchi, Switzerland).

Microwave assisted extraction
Microwave assisted extraction was conducted using a modified method suggested by Dahmoune et al. (2015). The solution was prepared according to a 28:1 solvent to solid ratio, where 10.7 g of dried curry leaf powder was mixed with 300 mL of 60% ethanol. The solution was then irradiated for 3 min using a microwave set to 300W (Samsung, Korea).

Ultrasonic assisted extraction
Ultrasonic assisted extraction was conducted using a modified method suggested by Zainol et al. (2018). A total of 38 g powdered M. koenigii leaves were mixed with 60% ethanol. The mixture was submerged in an ultrasonic cleaner bath (Buchi, Switzerland) and extracted for 30 min. The extracted samples were then centrifuged at 700 rpm at 4℃ for 10 mins. Ethanol was then removed from the extract using a rotary evaporator and the resulting extract was placed in a bottle in the chiller for the next analysis.

Determination of antioxidant properties
The capacity of trapping of free radical DPPH was evaluated according to the method described by  with slight modification. A 6.1 × 10 -5 M solution of DPPH was prepared in ethanol. Then, 75 μL of the diluted extract was added to 3 mL of the DPPH solution. The absorbance was taken at 515 nm using methanol with DPPH as the negative control whilst quercetin was the positive control after letting it stabilise for 1 hr. All operations or conducted in dark or dim light. The inhibition percentage (IP) of the DPPH by the extract was calculated according to the formula: A0 min is the absorbance of the blank at t = 0 min, and 60 mins is the absorbance of samples at 60 mins. The result was expressed as μmol Trolox equivalent (TE) per gram of sample on a dry basis, through a doseresponse curve for Trolox (0-350 μM).

Ferric thiocyanate (FTC) method
The samples were analysed using methods suggested by Zainol et al. (2003). The adjusted 1mg/mL of sample was dissolved in 4 mL of absolute ethanol (99.5%), added with 4.1 mL of 2.52 % linolenic acid in absolute ethanol. Eight millilitres mL of 0.05M phosphate buffer (pH7) were mixed with 3.9 mL of distilled water and kept in a screw cap bottle and placed in a water bath shaker at 40 o C. Approximately 0.1 mL of samples were added to 9.7 mL of 75% ethanol and 0.1 mL of 30% ammonium thiocyanate finally by 0.1 mL of 0.02M ferrous chloride added in 3.5% Hydrochloric acid into the reaction mixture. The absorbance of the resulting red -blood colour was measured after 3 mins at 500 nm every 24 hrs until the day the absorbance of the control reached the maximum value.

Thiobarbituric acid (TBA) method
The samples were analysed using methods suggested by Malik et al. (2017). An aliquot (1 mL) of sample solution obtained by the FTC method was added to 2 mL of 20% thrichloroacetric acid and 2 mL of 0.67 % 2thiobarbituric acid. The mixture was placed in boiling water at 100°C for 10 mins. Next, the mixture was cooled and then centrifuged at 300 rpm for 20 mins. The absorbance of the supernatant was measured at 552 nm.

Determination of total phenolic content (TPC)
The total phenolic compounds were determined using Folin-Ciocalteu reagent according to the colourimetric method described by Ng et al. (2020). An aliquot (1 mL) of every sample was diluted into 50 mL of stock solution. Approximately 1 mL from the stock solution was added to 17.9 mL of distilled water in 0.5 mL of Folin-Ciocalteu reagent and left to stand for 1 min. Then, 1.5 mL of 20% sodium carbonate was added to the mixture. The sample prepared was then left at room temperature for 2 hrs in the dark. The absorbance value was taken at 765 nm, resulting in mg GAE per gram of sample extract (mg GAE/g) expressed as gallic acid equivalent.

Determination of total flavonoid content (TFC)
Approximately 50 mg of sample was mixed with 1.5 mL methanol, 0.1 mL 10% aluminium chloride, 0.1 mL 1M potassium acetate and 2.8 mL distilled water. The mixture was then incubated at room temperature for 30 mins. The absorbance of the reaction mixture was taken at 415 nm. The result was expressed as mg of quercetin equivalent (QE) per gram of sample extract (MG QE/G) as mg quercetin equivalent (QE)/g of dried plant material (Chong et al., 2018).

Determination of individual flavonoid content using HPLC
All samples were refluxed in 6 M HCl at 90°C for 2 hrs with 60% (v/v) aqueous methanol extracted prior to the HPLC analysis (Mohd Zainol et al., 2009). HPLC analysis was performed using an Analytical High -Performance Liquid Chromatography (HPLC) (Shimadzu, Japan) with a 4 solvent delivery system quaternary pump (LPG 3400 SD) including a diode array detector (DAD 3000) with 5 cm flow cell, a manual sample injection valve equipped with a 20 µL loop and Chromeleon 6.8 system manager as the data processor. A total of 20 mL of sample were injected into the HPLC system for every analysis. One percent (V/V) aqueous acetic acid solution and acetonitrile were used as the gradient mobile phase. The flow rate was fixed at 0.7 mL/min, while the column oven was set at 28℃. The composition of the mobile phase was back to the initial condition (solvent B: solvent A: 10: 90) in 31 mins and allowed to run for another 5 mins before the injection of the next sample. HPLC chromatograms were detected using a photodiode array UV detector at 270 nm. Each compound was identified by its retention time and by spiking with standards under the same conditions (Seal, 2016).

Statistical analysis
The data obtained were subjected to one-way analysis of variance (ANOVA). The mean comparisons from triplicate analysis were carried out using Fisher's Least Significant Difference (LSD) test . Statistical analysis was performed using SPSS software 2004. Table 1 shows the extraction yield (%) of three different extraction methods. The extraction yield of solvent assisted extraction (SAE) showed no significant difference compared to the extraction yield of ultrasonic assisted extraction (UAE) and microwave assisted extraction (MAE). © 2020 The Authors. Published by Rynnye Lyan Resources FULL PAPER The increasing trend of the extraction yield can be expressed in such a manner where the extraction yield of UAE (1.70±0.10%) ˃ MAE (1.60±0.01%) ˃ SAE (1.47±0.04%). Tiwari (2015) stated that UAE is based on the concept of acoustic cavitation capable of damaging the cell wall of the plant matrix and thus favouring the release of bioactive compounds. This method can be used to extract a wide variety of phytochemicals such as phenolic compounds, indicating that it is the best technique for achieving higher yields. The exceptional capability of the acoustic cavitation to damage the cell wall in ultrasonic assisted extraction (UAE) encourages the release of more phytochemicals, thus increasing the extraction yield in this particular extraction method (Ledesma-Escobar et al., 2015).

Ferric thiocyanate method (FTC)
The ferric thiocyanate (FTC) value of the UAE extract did not show a significant difference compared to butylated hydroxytoluene (BHT), Vitamin C and α-Tocopherol (Table 1). This could indicate that the UAE method extracts reacted strongly to the FTC assay in addition to the DPPH assay. As the lowest absorbance reading on the fourth day of incubation dictates the highest antioxidant quality. UAE methods have been shown to preserve more of the antioxidant quality of M. koenigii leaves followed by MAE and SAE methods. The antioxidant quality demonstrated by the samples mimics a number of studies previously conducted. According to Jun et al. (2011), the crude extracts of green tea from the UAE method showed a high radical scavenging capability followed by MAE and SAE. The FTC assay also agrees with previous antioxidant quality patterns demonstrated by the UAE DPPH assay antioxidant quality is the highest compared to MAE and SAE.

Thiobarbituric Acid (TBA) method
The MAE extracts showed the lowest absorbance, indicating that they had the highest antioxidant quality compared to the SAE and UAE methods in the TBA assay (Table 1). Dahmoune et al. (2015) reported that citrus lemon extracts produced using MAE had higher lipid peroxidation compared to SAE and UAE extracts. This explains why MAE peroxidation is lower compared to SAE and UAE methods. Ince et al. (2012) also explained that MAE performed better due to better inhibition of lipid peroxidation by better antioxidant activity against hydroperoxide and free radical formation.

Total phenolic content (TPC) method.
M. koenigii leaves extracted using the MAE method showed the highest total phenolic content (TPC) (120.60±14.81 mg GAE/g sample), followed by UAE (88.79±4.48 mg GAE/g sample) and SAE (58.48±5.46 mg GAE/g sample). Compared to previous assays, it can be observed that the antioxidant activity of the UAE method in TPC was not as favourable as in previous analyses such as DPPH and FTC assays. This could be due to the high frequency in the UAE that could easily degrade the TPC. Degradation of flavonoids and polyphenol compounds is possible at high frequencies in the UAE (Jahromi, 2019). Routray and Orsat (2012) also reported that for total phenolic extraction, microwave applications were observed to have a much higher overall phenolic output than sonication-based extraction. Table 1 shows that there is a significant difference between the total flavonoid content (TFC) of M. koenigii leaves of three different MAE, UAE and SAE extracts. A similar trend was observed in the TPC analysis. However, the total flavonoid content of M. koenigii leaves showed a better significant difference compared to the TPC of M. koenigii leaves. The data suggest that M. koenigii leaves respond better to oxidation induced by aluminium flavonoid complexes. The total flavonoid content of M. koenigii leaves extracted by MAE was also found to be the highest in comparison to UAE and SAE. Analogous results have also been established by Nouha et al. (2017) on Maltese orange peel, where the method with the highest total phenol and flavonoid content is microwave-assisted extraction followed by ultrasoundassisted extraction, conventional solvent extraction. Figure 1 shows the HPLC chromatogram which verifies the presence of the designated standard flavonoid used in this study, identified by its retention time, while Figure 2 shows the HPLC chromatogram confirming the presence of individual flavonoids in the sample undergoing different extraction treatments, in line with the flavonoid standards shown in Figure 1. The results show that rutin, myricetin, kaempferol, gallic acid, catechin, epigallocatechin gallate, quercetin and pcoumaric acid were present after all types of extractions. These figures indicate the presence of the compounds tested confirming that the HPLC technique used was suitable for the detection, identification and measurement of the availability of flavonoids (Moniruzzaman et al., 2014). Table 2 shows the flavonoid content in all three samples of M. koenigii leaves from three different extraction methods (SAE, MAE and UAE) determined using an HPLC analysis. All samples analysed showed the presence of catechin, epigallocatechin gallate, quercetin, gallic acid, myricetin, kaempferol, rutin and pcoumaric acid (Figure 2). The previous study on M. koenigii leaves by Sepahpour et al. (2018) indicated that the 80 %t ethanol extraction yielded 0.9, 5.4, 2.4 and 1.4 mg/g freeze-crude extract of rutin, quercetin-3-glucoside myricetin and quercetin. It is interesting to note that the different extraction techniques used in the study affected the individual flavonoids differently. Results of the study also found that M. koenigii leaves extracted by MAE had the highest concentration of total individual flavonoids, while SAE showed the lowest concentration of total individual flavonoids between samples. Catechin was found to be the concentration compound found in all of the extracts analysed. In contrast, Hertog et al. (1992) reported that quercetin is the major flavonol found in vegetables such as broccoli, kale, French beans, celery, onions and cranberries. In addition, rutin was found to be the least abundant flavonoid in all the samples tested.

Conclusion
M. koenigii leaves extracted using UAE exhibited better antioxidant activity than that of MAE and SAE. The analyses of TPC and TFC showed that the MAE extract M. koenigii leaves showed the best results compared to the UAE and the SAE. M. koenigii leaves extracted by MAE exhibited the highest total number of individual flavonoids compared to M. koenigii leaves extracted using UAE and SAE had the highest concentration of p-coumaric acid, myricetin and quercetin concentration (mg/L). Catechin was the highest flavonoid detected in all the different extraction methods used in the study. The main strength of this study is that it narrows the current discrepancy between the impacts of different extraction methods on the antioxidant properties of M. koenigii.