Comparison of phenolic content and antioxidant activity of two common fruits of Bangladesh in solvents of varying polarities

Phenolic content and antioxidant activity of two common fruits of Bangladesh, namely Phyllunthus emblica and Elaeocarpus floribundus, were measured in water, methanol, ethanol, acetone and hexane extracts. Several in vitro models including phosphomolybdenum assay, DPPH free radical scavenging assay, FRAP assay and reducing power assay were used to assess the antioxidant activity of these extracts in comparison with reference antioxidants. Between the two fruits, P. emblica showed higher phenolic content and antioxidant activity in all the solvents used. In the DPPH scavenging assay, the activity of P. emblica extracts was close to reference antioxidants, ascorbic acid and BHT. Besides, considering the solvents used, extracts of both fruits had the highest phenolic and antioxidant activity in polar solvents. The correlation coefficient between total phenolics and antioxidant activities was found statistically significant. These findings suggest that P. emblica could be an excellent antioxidant resource for industries like food, pharmaceutical, and cosmetics.


Introduction
In a biological system, reactive oxygen species and reactive nitrogen species such as superoxide, hydroxyl, and nitric oxide radicals can damage DNA and lead to the oxidation of lipid and protein in cells, causing some chronic and degenerative diseases (Nunes et al., 2012;Aktumsek et al., 2013;Xu et al., 2017). Recent studies indicate that because of the presence of antioxidants especially, polyphenols and carotenoids, frequent consumption of fruits and vegetables is associated with a lower risk of inflammation, stroke, cancer, diabetes and neurodegenerative diseases (Garsia-Salas et al., 2010;Khoddami et al., 2013;Zhang et al., 2015). Antioxidant molecules block both the initiation and propagation of oxidizing chain reactions, thereby impeding or slowing the oxidation process (Lobo et al., 2010;Nunes et al., 2012).
Fruits of P. emblica and E. floribundus, locally known as amloki and jolpai, respectively, are two wellconsumed fruits of Bangladesh. The plant P. emblica, belonging to the family Euphorbiaceae, is indigenous to the tropical region of Southeast Asia. It is generally found throughout the forests of Chittagong, Chittagong Hill Tracts, Cox's Bazar, Dinajpur, Tangail, Sylhet, and villages of Bangladesh (Uddin et al., 2016). It is highly nutritious and rich in vitamin C, minerals and amino acids. All parts of this plant have medicinal properties, especially the fruits that are used in Ayurveda as a potent Rosayan (rejuvenator). The plant is reported to have various pharmaceutical potentials including antiinflammatory, analgesic and antipyretic, antioxidant, antimicrobial, hepatoprotective, antitumor, immunomodulatory, anti-atherogenic, antiulcerogenic, and adaptogenic activities (Khopde et al., 2001;Krishnaveni and Mirunalini, 2010;Gaire and Subedi, 2014;Uddin et al., 2016;Chaphalkar et al., 2017).
On the other hand, plant E. floribundus belongs to the family Elaeocarpaceae and is found in the South and East Asian region including Bangladesh (Zaman, 2016;Swargiary et al., 2017). The fruits of this plant are traditionally used to prepare pickles (Zaman, 2016) and the seeds are used to prepare oil in Myanmar (Shin et al., 2018). The plant has various biological activities like antioxidant, antibacterial, cytotoxic, anti-ageing and anticancer (Utami et al., 2013;Sircar and Mandal, 2017;Deivasigamani and Devi, 2018).
The phenolic composition of plants depends on many factors such as genetic, seasonal, agronomic, maturation stage and growing conditions like temperature and rainfall (Hilton and Palmer-Jones, 1973; There are many reports on the phenolic contents and antioxidant activities of P. emblica fruits extracted with different solvents and grown in South and East Asian regions especially in India, China, Indonesia, Thailand (Khopde et al., 2001;Charoenteeraboon et al., 2010;Li et al., 2015;Fitriansyah et al., 2018;Li et al., 2019). However, there is a scarcity of studies, if any, showing the effect of solvent types, covering apolar to the polar range, on the phenolic content and antioxidant activities of this fruit. Although the phenolic contents and antioxidant activities of P. emblica fruit grown in Bangladesh has been reported Mondal et al., 2017) but the effect of the solvents on the extraction of phenolic contents and antioxidant activities are not studied. Moreover, detail antioxidant activities of this fruit by different in vitro methods and their comparison with reference antioxidants were not studied yet.
On the other hand, to the best of our knowledge, limited research has been done on extraction, quantification and comparison of phenolic compounds of E. floribundus fruit with different solvents. Sircar and Mandal (2017) reported the phenolic content and DPPH free radical scavenging activity of this fruit in ethanol and aqueous extract. However, the report on the phenolic compositions and antioxidant activities of this fruit grown in Bangladesh is seldom, if any.
In this study, the phenolic contents and antioxidative capacities of P. emblica and E. floribundus fruits grown in Bangladesh have been investigated and compared. Moreover, the effect of solvents on the extraction of phenolic compounds with the antioxidant potential of these two fruits has also been studied in order to find the appropriate solvent for extraction.

Plant materials
Fruits of Phyllanthus emblica L. and Elaeocarpus floribundus Blume (Figure 1) were collected from Chittagong, Bangladesh in October to November, 2018 and were properly authenticated by Dr. Sarder Nasir Uddin, Principal Scientific Officer, Bangladesh National Herbarium, Dhaka, Bangladesh where voucher specimens with accession numbers DACB 51334 and DACB 51335, respectively have been deposited. All the selected fruits were identical in size, shape, colour, ripening stages and were also of eating quality.

Sample preparation
The freshly collected matured fruits were initially washed thoroughly with tap water until the attached dust particles, unicellular algae etc. were removed. Finally, they were washed with distilled water. The fruits were chopped and seeds were removed. The resulting fruit parts were dried in Economy Incubator (Size 2) at 50°C for 10 hrs. The dried fruits were grounded into powder with a grinder. The powders were stored separately in air -tight containers and kept in a refrigerator at -80°C.
For extraction of phenolic compounds, 10 g powder of each fruit was placed separately in a conical flask and soaked in 100 mL of a different solvent system. Each container was sealed and shaken intermittently for 4 days. On the 5 th day, fruit extracts were filtered through 0.45 µm filter paper and the filtrate was stored in a fresh conical flask at 4°C. To the residue, 100 mL solvent was further added and left for 3 days with regular shaking. On the 7 th day, the filtration process was repeated as was done previously and the resulting extract was mixed with those previously stored. Then the preparations were evaporated in a rotary evaporator at 50°C to get the solid crude extract. The extraction yield was measured and expressed as a percentage. The obtained crude extract Five different pure solvent systems including water, methanol, ethanol, acetone and n-hexane (Gorinstein et al., 2010;Lee et al., 2015) with the relative polarity of 1, 0.762, 0.654, 0.355 and 0.009, respectively were used for extraction. To avoid microbial contamination while extracting with water, aseptic conditions were maintained (Aiyegoro and Okoh, 2009). Dimethyl sulfoxide (DMSO) was used to dissolve all extracts for subsequent analysis.

Determination of total polyphenol content
Total polyphenol content was estimated following the method of Ough and Amerine (1988). A 20 µL of sample extract was diluted with 2.58 mL distilled water. Then 100 µL of Folin-Ciocalteu reagent was added. After 1 min, 300 µL of 20% sodium carbonate solution was added and mixed. The mixture was then incubated for 2 hours at room temperature. The absorbance of the resulting blue colour supernatant was measured at 765 nm (UV-1601 Shimadzu, Kyoto, Japan). All samples were analyzed in triplicates. A standard calibration curve of gallic acid (0.002 -0.01 mg/mL, r 2 = 0.997) was plotted. The total phenolic content was expressed as mg gallic acid equivalent (GAE)/g dry fruits weight (DW).

Determination of total flavonoid content
Total flavonoid contents of the fruit extracts were determined following the method of Zhishen et al. (1999). An aliquot (0.2 mL) of fruit extract was mixed with 4.8 mL of distilled water and 0.3 mL of (5% w/v) NaNO 2 was added. After 5 min, 0.3 mL of (10% w/v) AlCl 3 was added and after another minute, 2 mL of NaOH (1 M) was added. To make the final volume 10 mL, 2.4 mL of distilled water was added immediately. The absorbance of the vigorously mixed resulting solution was read at 510 nm (UV-1601 Shimadzu, Kyoto, Japan). A calibration curve was prepared using a standard solution of quercetin (0.002-0.01 mg/mL, r 2 = 0.992). The results were presented as mg quercetin equivalents (QE)/ g DW.

Determination of antioxidant activities 2.6.1 DPPH free radical-scavenging activity (FRSA) assay
The reaction mixture (total volume, 3 mL), consisting of 0.5 mL of 0.5 M acetic acid buffer solution (pH 5.5), 1 mL of 0.2 mM DPPH in ethanol, and 1.5 mL of 50% (v/v) ethanol aqueous solution with different concentrations of extracts, was shaken vigorously (Hossain et al., 2008). After incubation at room temperature (25°C) for 30 mins, the amount of DPPH remaining was determined by measuring absorbance at 517 nm. Ascorbic acid and BHT were used as a reference. The percentage of inhibition of DPPH radicals was calculated using the following formula: Where Acontrol is the absorbance of DPPH solution without extract.

Evaluation of total antioxidant capacity
The total antioxidant capacities of the extracts were evaluated by phosphomolybdenum method according to Prieto et al. (1999). An aliquot of 400 µL of sample solution was combined in a screw cap tube with 4 mL of reagent solution (0.6 M sulfuric acid, 28 mM sodium phosphate, and 4 mM ammonium molybdate). The tubes were capped and incubated in a thermal block at 95 0 C for 90 mins and cooled to room temperature. The absorbance was measured at 695 nm against a blank. The typical blank was prepared as a sample except solvent DMSO was added instead of sample and was incubated under the same conditions as that of a sample. The antioxidant capacity of the sample was expressed as equivalents of ascorbic acid (AAE), utilizing a calibration curve of ascorbic acid in the concentration range from 0.02 mg/mL to 0.32 mg/mL (r 2 = 0.994).

Determination of ferric reducing antioxidant power (FRAP)
The FRAP assay was determined by the method described by Benzie and Strain (1996) with slight modifications. The FRAP reagent was prepared freshly by mixing 300 mM acetate buffer (pH 3.6), 10 mM TPTZ and 20 mM ferric chloride in a ratio of 10:1:1 (v/ v/v). FRAP reagent was pre-warmed at 37°C for 30 mins and different volumes (10,20,40,80,160,240 and 320 µL) of properly diluted fruit extract was mixed with 3 mL of FRAP reagent. The reaction mixture was then incubated in dark for 30 mins. and the absorbance was determined at 593 nm. The antioxidant potential of the fruit extract was determined based on a calibration curve plotted using FeSO 4 .7H 2 O (4.29 -85.8 µM, r 2 = 0.999). Ascorbic acid and BHT were used as a reference.

Reducing power assay
The ferric reducing power was assayed with slight modifications of the method described by Oyaizu (1986). Various concentrations of the extracts (40 µL) were mixed with 2.46 mL of 0.2 M phosphate buffer (pH 6.6) and 2.5 mL of 1% potassium ferricyanide. The mixture was incubated at 50°C for 20 mins. After incubation, followed by cooling for a few minutes, 2.5 mL of 10% trichloroacetic acid was added. The resulting mixture was centrifuged at 650 g for 10 mins. A 2.5 mL of the eISSN: 2550-2166 © 2021 The Authors. Published by Rynnye Lyan Resources FULL PAPER supernatant was mixed with 2.5 mL distilled water and 0.5 mL of 0.1% ferric chloride. The absorbance of the mixture was measured at 700 nm. Ascorbic acid was used as a reference.

Statistical analysis
The SPSS package, version 19.0 (SPSS Inc., Chicago, IL, USA) was used to perform the statistical analysis. Analysis of variance (ANOVA) and Duncan's multiple range method was used to compare the mean values. Data were expressed as mean ± standard deviation of triplicate measurements. Differences were considered significant at P < 0.05. The IC 50 values were calculated by linear regression analysis. Pearson's correlation coefficient was used to do determine correlations.

Results and discussion
In this study, the phenolic contents and antioxidant activities of two commonly consumed fruits of Bangladesh named amloki and jolpai were first compared. To the author's knowledge, there is a single previous study investigating the efficiency of solvents of different polarities (n-hexane, ethyl acetate and hexane) on the extraction of phenolic contents of P. emblica grown in Indonesia (Fitriansyah et al., 2018). Furthermore, study on phytochemical contents and antioxidant activity of E. floribindus fruit is rare. In this study, we used five solvents of differing polarities in the following order: water, methanol, ethanol, acetone, hexane.

Extraction yield and antioxidant components
The influence of solvents on the extraction yield of both fruits is shown in Table 1. Among the different solvent extracts, the water extracts had the highest yield for both fruits, followed by methanol, ethanol, acetone, then n-hexane extracts. The aqueous extracts of plants are commonly found to show higher yields than other solvent extracts (Kong et al., 2012). Of the two fruits, the extraction yield of P. emblica was higher than that of the E. floribundus for all solvents.
The total phenolic and flavonoid contents in the P. emblica and E. floribundus extracts are summarized in Table 2. It was found that the phenolics and flavonoids of P. emblica and E. floribundus were higher in a polar solvent such as water, methanol, and ethanol, implying that most polyphenols in these fruits are polar. A similar effect of the polarity of solvent on phenolic and flavonoid contents of plant materials were reported by Kong et al. (2012), Lee et al. (2015), and Belyagoubi et al. (2016), as well. Similar to the extraction yield, P. emblica had higher phenolic (0.28-103.95 mg GAE/g DW) and flavonoid contents (0.56-30.88 mg QE/g DW) than the phenolic (0.04-3.62 mg GAE/g DW) and flavonoid (0.16-1.42 mg QE/g DW) contents of E. floribundus fruit in all solvents. It is difficult to compare the phenolic and flavonoid content of a particular fruit determined by other studies as the method of extraction, assay and unit of expression varies. However, similar phenolic content (81.5 to 120.9 mg (GAE)/g DW) and flavonoid content (20.3 to 38.7 mg (QE)/g DW) was reported by Liu et al. (2008) in the methanolic extract of the fruit emblica from six regions of China. Microwave extraction of P. emblica fruits, collected from Fujian province of China, with 66% ethanol produced 133.58±15.61 mg GAE/g DW phenolics (Li et al., 2019). The Folin-Ciocalteu reagent method may overestimate TPC, because reducing agents, such as ascorbic acid, may interfere with the results. However, different phenolic compounds respond in the Folin-Ciocalteu method (Ikram et al. 2009

Antioxidant activities
Due to the presence of a variety of antioxidant compounds with different polarities in fruits the antioxidant capacity of fruits differs with the solvent extraction methods. Moreover, antioxidants may respond to different radical or oxidant sources in a different manner. Thus, no single assay can accurately reflect all of the radical sources and antioxidants present in a mixed or complex system as multiple reaction characteristics, mechanisms, and phase localizations are usually involved (Park et al., 2014). Many different methods are used to assay the antioxidant activity of plant extracts (Alam et al., 2013;Pisoschi et al., 2016;Zengin et al., 2018). In this study, we applied four different antioxidant assay methods that would provide a better insight into the true antioxidant potential of the extracts.

2,2-Diphenyl-1-picryl hydrazyl (DPPH) radical-scavenging activity
The assay of scavenging of stable organic radical DPPH has been widely used in antioxidant capacity studies of plant extracts or antioxidant compounds. The antioxidant activity of plant extracts is thought to be due to its ability to transfer hydrogen or electron to DPPH, thus neutralizing its free radical character. In this assay, the purple radical (picrylhydrazyl) is reduced by antioxidant compounds to the corresponding pale yellow hydrazine (picrylhydrazine), in a concentrationdependent manner. The discolouration indicates the FRSA of the tested sample (Rahman et al., 2015).
The abilities of the studied fruit extracts to scavenge DPPH were assessed based on their IC 50 . The IC 50 of an antioxidant is defined as the amount of the antioxidant needed to decrease the radical concentration by 50% and its values are inversely related to the antioxidant capacities. Table 3 shows the FRSA of the extracts of P. emblica and E. floribundus fruits. For both fruits, the antioxidant strength varied with the polarity of the extraction solvent. Similar to its highest total phenolic content in ethanolic extract, the same extract of P. emblica showed the highest FRSA, closer to the reference antioxidant BHT (0.10±0.001 mg/mL) but weaker than ascorbic acid (0.07±0.002 mg/mL), implying its potencies. The methanol and acetone extracts of P. emblica also showed strong FRSA which were close to BHT. Water extract showed relatively poor scavenging activity compared to BHT and n-hexane extract showed the least activity. A similar effect of solvents on the DPPH radical scavenging activity was observed by Do et al. (2014) while studying the solvent effect on the antioxidant activity of L. aromatica. In most instances, the hexane extracts were the least reactive in scavenging the DPPH radicals (Kong et al., 2012). A high total phenolic content is not always associated with a high FRSA. As reported by Bhebhe et al. (2016), ethanol (50%) extracts from S. jambolonum had the highest FRSA, yet they recorded lower TPC than 50% acetone and hot water. In this study, although water extract had higher phenolic content than methanol and acetone extract but the former showed lower FRSA.
The scavenging activity of E. floribundus varied clearly with the polarity of the solvent, i.e. strongest scavenging activity in water but the weakest in acetone, in accordance with its phenolic and flavonoid contents in different solvents. Earlier reports indicated that the FRSA of phenolic extracts varies from plant to plant such that only a suitable solvent for a particular plant may extract phenolics with the highest activity. A solvent may be efficient on one plant and less efficient on another. The phenomena may be explained by the fact that the recovery of antioxidant compounds such as polyphenols from plant materials is influenced by the  FULL PAPER solubility of the phenolic compounds in the solvent used for the extraction. Moreover, the polarity of the solvent will play an important role in increasing the solubility of phenolic compounds (Bhebhe et al., 2016). All extracts of E. floribundus showed lower DPPH radical scavenging activities than P. emblica and standard antioxidants. This can be attributed mainly due to the lower phenolic content of E. floribundus than P. emblica. However, the hexane extract of E. floribundus did not reach the 50% inhibition of the DPPH radicals at the concentration tested.

Total antioxidant capacity
The total antioxidant capacity was determined by phosphomolybdenum method based on reduction of Mo (VI) to Mo(V) by antioxidant substance and subsequent formation of a green phosphate Mo(V) compounds at acidic pH with an absorbance at 695 nm (Aktumsek et al., 2013). The water extracts of P. emblica and E. floribundus fruits had the highest antioxidant capacity (158.81±0.99 and 53.3±0.28 mg AAE/g dw, respectively), whereas hexane extracts had the lowest (1.48±0.02 and 1.12±0.04 mg AAE/g DW, respectively). Besides, P. emblica showed a higher total antioxidant capacity than E. floribundus for all solvent extracts (Table 3).

Ferric reducing antioxidant activity
The assay is based on the ability of antioxidant compounds to a reducing complex, Fe(III)-TPTZ to Fe (II)-TPTZ. This reduction is associated with the colour change from light blue to dark blue with an absorbance maximum at 593 nm (Ezzati Nazhad Dolatabadi et al., 2014). The ferric reducing antioxidant activities of the different extracts of P. emblica and E. floribundus fruits are presented in Table 3. The ferric reducing power of P. emblica fruit was higher than E. floribundus for all solvents. Again, water extract and hexane extract of both P. emblica and E. floribundus fruits showed the highest (1362.15 ±33.49 and 29.50±0.58 mmol Fe 2+ /g DW, respectively) and lowest reducing activity (8.66±0.29 mmol Fe 2+ /g DW and undetectable, respectively), respectively. The ferric reducing antioxidant power of ascorbic acid and BHT were 413.96±12.35 mmol/g and 157.15±6.77 mmol/g, respectively. The highest and lowest FRAP activity was also reported in the water and hexane extract, respectively of black mustard seeds (Lee et al., 2015). Moreover, the FRAP and TAC values of both fruits showed a similar trend. The FRAP activity of hexane extract of E. floribundus could not be measured due to its low activity.

Reducing power activity (Iron(III) to iron(II) reduction)
Literature reports suggest that the reducing properties are generally associated with the presence of reductones. The reductones have been shown to exert antioxidant activity by breaking the free radical chain by donating a hydrogen atom. In the ferric reducing assay, the presence of antioxidants causes the conversion of the Fe 3+ /ferricyanide complex to the ferrous form which can be monitored by measuring the formation of Perl's Prussian blue at 700 nm (Alam et al., 2016). In this study, the ferric reducing activity of different extracts of two fruits was evaluated. Figure 2 shows the dosedependency of ferric reducing power of different solvent extract of both fruits and ascorbic acid. The reducing power of all extracts increased with increased concentration. Dose dependency was also shown by ascorbic acid. Ethanol and water extracts showed the highest reducing activity for P. emblica and E. floribundus fruits, respectively, in accordance with their highest phenolic content whereas, for both fruits, hexane showed the lowest activity similar to other antioxidant assays. All the solvent extracts of P. emblica showed higher ferric reducing activity than ascorbic acid except hexane. The ferric reducing activity of hexane extract of E. floribundus was not found at the concentration tested due to its poor activity.

Correlation analyses
Pearson correlation analyses were done to estimate  (Table 4). The total antioxidant capacity and ferric reducing activity power of both fruits showed a very strong correlation (r > 0.8) with both phenolic and flavonoid content. The lower values in IC 50 of DPPH had higher antioxidant activity. The scavenging activity (IC 50 ) showed a negative correlation with the antioxidant compounds. The scavenging activity of P. emblica showed a strong correlation (r = 0.6-0.79) with both phenolic and flavonoid content. This implies the ability of polyphenols to act as antioxidants and neutralize free radicals. Choe et al. (2014) also reported a negative correlation between phenolic content and DPPH radical scavenging activity of ethanol extract of persimmon peel.

Conclusion
This study indicated that the P. emblica fruit had very high phenolic content and antioxidant activity compared to the E. floribundus fruit. The antioxidative activity of P. emblica grown in Bangladesh was very close to the reference antioxidant BHT and ascorbic acid. The antioxidant activity of both fruits showed a clear correlation with the phenolic content proving that the activity is due to the presence of bioactive compounds. Polar solvents especially water and ethanol showed higher efficiency in extracting the phenolic compounds of P. emblica, resulting in higher antioxidative activity. However, the effect of single or mixed organic solvent or organic solvent-water mixtures on the extraction of phenolic content and antioxidant activity of P. emblica remains to be studied.