A Generalized Method for Determining Free Soluble Phenolic Acid Composition and Antioxidant Capacity of Cereals and Legumes

Summary

Phenolic acids are important phytochemicals that are present in whole grains. They possess bioactive properties such as antioxidant protective functions. This work aimed at reporting on a generalized method for the HPLC identification, total phenolic content estimation, and determination of the antioxidant capacity of phenolic acids in cereals and legumes.

Abstract

Phenolic acids are a class of organic compounds that bear both a phenolic group, and a carboxylic group. They are found in grains and concentrate in the bran of cereals or seed coat of legumes. They possess antioxidant properties that have generated much research interest in recent years, about their potential antioxidant protective health functions. This work presents a generalized method for the extraction of free soluble phenolic acids from whole grains and analysis of their antioxidant capacity. Five whole grain samples comprising two cereals (wheat and yellow corn) and three legumes (cowpea bean, kidney bean, and soybean), were used. The grains were milled into flour and their free soluble phenolic acids extracted using aqueous methanol. The compounds were then identified using a high-pressure liquid chromatograph (HPLC). The Folin-Ciocalteu method was used to determine their total phenolic content while their antioxidant capacities were determined using the DPPH radical scavenging capacity, Trolox equivalent antioxidant capacity (TEAC) and oxygen radical absorbance capacity (ORAC) assays. The phenolic acids identified included vanillic, caffeic, p-coumaric and ferulic acids. Vanillic acid was identified only in cowpea while caffeic acid was identified only in kidney bean. p-Coumaric acid was identified in yellow corn, cowpea, and soybean, while ferulic acid was identified in all the samples. Ferulic acid was the predominant phenolic acid identified. The total concentration of phenolic acids in the samples decreased in the following order: soybean > cowpea bean > yellow corn = kidney bean > wheat. The total antioxidant capacity (sum of values of DPPH, TEAC and ORAC assays) decreased as follows: soybean > kidney bean > yellow corn = cowpea bean > wheat. This study concluded that HPLC analysis as well as DPPH, TEAC, and ORAC assays provide useful information about the phenolic acid composition and antioxidant properties of whole grains.

Introduction

Phenolic acids are among the most important phytochemicals studied in plants due to the vital role they play in plant defense against herbivory and fungal infection, as well as maintaining structural support and integrity in plant tissues^1,2. They are abundant in the bran of cereals and seed coat of legumes³. Structurally, they are divided into two groups: the hydroxybenzoic acids (Figure 1) and hydroxycinnamic acids (Figure 2). The common hydroxybenzoic acids in cereals and legumes include gallic, p-hydroxybenzoic, 2,4-dihydroxybenzoic, protocatechuic, vanillic, and syringic acids, while the common hydroxycinnamic acids include caffeic, p-coumaric, ferulic, and sinapic acids³. Phenolic acids also possess antioxidant properties since they are able to scavenge free radicals, which cause oxidative rancidity in fats, and initiate and propagate radical-induced oxidative stress in physiological systems^4,5. Due to this vital physiological role as antioxidants, they are the subject of recent research. This is because when consumed as components of plant foods, they can exert antioxidant protection.

Cereals and cereal products are major carbohydrate food sources for humans and animals worldwide⁶. Cereals include wheat, rice, corn (maize), barley, triticale, millets, and sorghum. Among them, corn is the most utilized, with an estimated global utilization of 1,135.7 million tonnes in 2019/2020, followed by wheat with an estimated global utilization of 757.5 million tonnes during the same period⁷. Cereal foods are great sources of energy to consumers since they are rich sources of carbohydrates. They also provide some protein, fat, fiber, vitamins and minerals⁶. In addition to their nutritional value, cereals are good sources of phytochemical antioxidants, particularly phenolic acids, which have the potential to protect the physiological system from radical-induced oxidative damage³. Legumes are also good sources of nutrients and are generally higher in protein than cereals. They also contain vitamins and minerals and are used in the preparation of various foods⁸. Additionally, legumes are good sources of a variety of phytochemical antioxidants, including phenolic acids, flavonoids, anthocyanins, and proanthocyanidins^9,10. Different varieties of cereals and legumes may have a different phenolic acid composition. There is therefore the need to study the phenolic acid composition of cereals and legumes and their varieties, in order to know their potential health benefits with respect to phenolic antioxidants.

A number of assays have been reported for measuring the quantity of phenolic acids in cereal and legume grains, and determining their antioxidant activities. The most common methods of analysis for whole grain phenolic acids are spectrophotometry and liquid chromatography¹¹. The aim of this work was to demonstrate a generalized high pressure liquid chromatographic method for determining free soluble phenolic acid composition, and spectrophotometric methods for determining total phenolic content and antioxidant capacity of some whole grain cereals and legumes.

Protocol

1. Type of samples

1. Use five whole grain samples, comprising two cereals (e.g., durum wheat and yellow corn) and three legumes (e.g., Blackeye cowpea bean, soybean, and red kidney bean) for this study.
2. Mill 50 g of each grain in triplicates into flour, using a coffee grinder, and pass them through a 500 µm sieve.
3. Store them at -20 °C.

2. Sample preparation

1. Determination of dry matter content and expression of dry weight basis
   NOTE: Determine the dry matter content of each powdered sample according to the method of AOAC (2000)¹².
   1. Turn on a forced-convection oven and set the temperature to 130 °C.
   2. Dry desiccant (silica gel) in the oven for 30 min to 1 h and transfer the dried silica gel to a desiccator.
   3. Accurately weigh 2 g of each sample into a clean, pre-dried, and weighed aluminum can.
   4. Dry the weighed samples at 130 °C for 1 h in a forced convection oven.
   5. Transfer the dried sample to the desiccator and allow to cool to ambient temperature.
   6. Weigh the dried, cooled sample and record its weight.
   7. Calculate the dry matter content of each sample as follows:
      
   8. Express each parameter measured in dry weight basis using the following formula:
      
2. Phenolic acid extraction
   NOTE: Extract the soluble free phenolic compounds in the grain samples using a modification of the method of Y. Qiu et al.⁵ that enables phenolic extraction from milligram quantities of whole grains.
   1. Accurately weigh 100 mg of the whole grain flour sample directly into an amber-colored 2 mL capacity microcentrifuge tube. The dark color of the tube helps to prevent exposure of the mixture to light.
   2. Add 1 mL of 80% aqueous HPLC-grade methanol to each of the tubes containing the samples.
   3. Vortex briefly to mix the methanol solution and the samples.
   4. Sonicate the samples for 60 min to extract the free soluble phenolic compounds. Put a cover over the samples for the duration of the sonication for added protection from light.
   5. After sonication, centrifuge the mixture at 20,000 × g for 5 min to sediment the solid residues leaving the supernatant on top. Free phenolic compounds will be present in the supernatant after centrifugation.
   6. Transfer the supernatant into a clean microcentrifuge tube.
      NOTE: The supernatant needs to be filtered prior to injecting into the HPLC instrument. To filter the supernatant, remove the plunger of a 3 mL syringe and attach a syringe filter. The filter should have a pore size of no larger than 0.22 µm.
   7. Pipette approximately 0.4 mL of the supernatant into the top of the syringe. Re-insert the plunger and push the liquid through the filter into an HPLC vial containing a vial insert.
   8. Once the instrument has been set up to run the method outlined in the manuscript for HPLC analysis, load the vials into the carousel to correspond with the sample list.
   9. Obtain HPLC chromatograms at 320 nm and 280 nm showing distinct peaks representing different phenolic compounds.
   10. Using appropriate standard curves, quantify hydroxycinnamic acids at 320 nm since they have maximum absorbance at this wavelength. By the same principle, quantify hydroxybenzoic acids at 280 nm.
   11. Store the remaining extracts at -20 °C for other analyses.

3. Phenolic composition

1. Use a high-pressure liquid chromatograph (Table of Materials) to identify and quantify the extracted phenolic compounds in the samples, based on the method of J. Xiang, F. B. Apea-Bah, V. U. Ndolo, M. C. Katundu, and T. Beta ⁴.
2. Prepare phenolic acid standards (vanillic acid, caffeic acid, p-coumaric acid, ferulic acid, and sinapic acid) to identify and quantify constituent phenolic compounds in the extracts.
3. To do this, weigh 1 mg of each standard and dissolve in 1 mL of 50% aqueous methanol to produce 1,000 µg/mL of each standard.
4. Mix equal volumes of all five standards in a 2 mL amber centrifuge tube to produce a cocktail of standards, each having a concentration of 200 µg/mL.
5. Prepare serial dilutions of the standard cocktail by taking a volume into a new tube and diluting with equal volume of the aqueous methanol solvent.
6. Repeat the serial dilutions down to a concentration of 3.125 µg/mL.
7. Also, separately dilute each standard 40 times with the solvent to obtain 25 µg/mL concentration of each standard.
8. Set the column temperature to 35 °C and the sample oven temperature to 15 °C.
9. To prepare mobile phase A (0.1% aqueous formic acid), transfer 1 mL of formic acid into a 1 L capacity volumetric flask and add HPLC-grade water to the 1 L mark. Shake well to mix.
10. To prepare mobile phase B (0.1% formic acid in methanol), transfer 1 mL of formic acid into a 1 L capacity volumetric flask and add HPLC-grade methanol to the 1 L mark. Shake well to mix.
11. Adjust the volumes to the 1 L mark if necessary.
12. Filter both mobile phases through a 0.45 µm hydrophilic filter paper.
13. For the analysis, inject 10 µL of each extract or standard (25 µg/mL standards and the standard cocktails) on to the reverse phase column.
14. Elute with the mobile phases according to the linear gradient program for a 25 min run as follows: 0-3.81 min, 9%-14% B; 3.81-4.85 min, 14%-15% B; 4.85-5.89 min, 15%-15% B, 5.89-8.32 min, 15%-17% B; 8.32-9.71 min, 17%-19% B; 9.71-10.40 min, 19%-19% B; 10.40-12.48 min, 19%-26% B; 12.48-13.17 min, 16%-28% B; 13.17-14.21 min, 28%-35% B; 14.21-15.95 min, 35%-40% B; 15.95-16.64 min, 40%-48% B; 16.64-18.37 min, 48%-53% B; 18.37-22.53 min, 53%-70% B; 22.53-22.88 min, 70%-90% B; 22.88-25.00 min, 90% B.
15. Compare the retention times of chromatographic peaks obtained for the authentic standards of concentrations 25 µg/mL, at 280 nm and 320 nm, with those of the extracts, in order to identify the constituent phenolic compounds in the samples.
16. Plot calibration curves for the phenolic acid standard cocktails, with concentration of the standards on the horizontal axis, and the peak area on the vertical axis
17. Use the calibration curves to estimate the concentrations of the identified phenolic compounds, by comparing the peak areas on the plot with that of the identified compounds at 280 nm and 320 nm as mentioned in the earlier section.

4. Total phenolic content

NOTE: Determine the total phenolic content of the extracts using the Folin-Ciocalteu method described by F. B. Apea-Bah et al.¹³.

1. Prepare ferulic acid and gallic acid standards within the concentration range of 0.025 to 0.150 mg/mL to plot calibration curves in comparison, for the estimation of total phenolic content.
2. To do this, accurately weigh 1 mg each of ferulic acid and gallic acid standards into 2 mL capacity centrifuge tubes.
3. Add 1 mL of 50% aqueous methanol to each standard and vortex to dissolve, producing 1 mg/mL stock of each standard.
4. Prepare a series of dilutions from each stock solution in a total of 500 µL.
5. Pipette 18.2 µL of each extract or standard into a separate well on a 96-well microplate.
6. Add 36.4 µL of 10% (v/v) aqueous Folin-Ciocalteu reagent to each extract or standard.
7. Then, add 145.4 µL of 700 mM sodium carbonate to each reaction mixture.
8. Incubate the reaction mixtures in the dark at room temperature (20-25 °C) for 2 h.
9. Read the absorbance on a microplate reader at 750 nm.
10. Plot calibration curves of change in absorbance against concentration of the phenolic acid standards and use them to estimate the total phenolic contents.
11. Express the results as milligram ferulic acid equivalents per gram of milled sample (mg FAE/g) and milligram gallic acid equivalents per gram of milled sample (mg GAE/g) on dry weight basis.

5. Antioxidant assays

NOTE: Determine the antioxidant capacity of the grain extracts using the following three assays: 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging capacity; 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid (ABTS) radical scavenging capacity, which is also called the Trolox equivalent antioxidant capacity (TEAC); and oxygen radical absorbance capacity (ORAC).

1. Preparation of Trolox standards for standard curves
   1. Use Trolox, a water-soluble analogue of vitamin E, as a standard to estimate the in vitro antioxidant capacity of the whole grain extracts.
   2. Accurately weigh 1 mg of Trolox into a 15 mL tube. Dissolve in 4 mL of 50% aqueous methanol. Vortex to dissolve to prepare a stock solution of 1 mM (1000 µM).
   3. Prepare six concentrations of Trolox, that is, 50, 100, 200, 400, 600, and 800 µM to plot standard curves for the estimation of DPPH radical scavenging capacity and Trolox equivalent antioxidant capacity (TEAC). Similarly, prepare 6.25, 12.5, 25, and 50 µM concentrations of Trolox for estimating oxygen radical absorbance capacity (ORAC). Make up the total volume of each concentration to 500 µL as shown in Table 1.
2. Dilution of sample extracts
   1. Dilute the sample extracts with methanol prior to analysis. Here, yellow corn and cowpea extracts were diluted two times, the wheat and kidney bean were diluted five times while the soybean extract was diluted 10 times with methanol.
3. Trolox equivalent antioxidant capacity (TEAC) assay
   1. Measure the Trolox equivalent antioxidant capacity (TEAC) of the samples using the method described previously by F. B. Apea-Bah et al.¹⁴.
   2. Weigh 8.23 mg of ABTS into a clean 2 mL capacity amber centrifuge tube.
   3. Next, weigh 1.62 mg of potassium persulfate into another clean 2 mL capacity amber centrifuge tube.
   4. Add 1 mL distilled water to each of them and vortex to dissolve them.
      NOTE: This results in 16 mM ABTS stock solution with 6 mM aqueous potassium persulfate solution.
   5. Prepare the ABTS stock solution by mixing the ABTS and potassium persulfate solutions in equal volumes. The solution will immediately change to a dark color.
   6. Incubate the reagent mixture in the dark for 12-16 h.
   7. Dilute the ABTS stock solution 30 times with 200 mM phosphate buffered saline (PBS), to form the ABTS working solution. To do this, add 58 mL of 200 mM PBS to 2 mL of the ABTS stock solution. The working solution will contain 0.27 mM ABTS and 0.1 mM potassium persulfate.
   8. For the analysis, place 10 µL of each diluted extract or Trolox into a 96-well microplate.
   9. Add 190 µL of ABTS working solution to each well and incubate the reaction mixtures for 60 min.
   10. Measure the absorbance of the reaction mixtures at 750 nm in a microplate reader.
   11. Use the Trolox standards in a concentration ranging from 100 to 800 µmol/L to plot a calibration curve.
   12. Estimate the ABTS radical scavenging capacity from the calibration curve.
   13. Express the results as micromole Trolox equivalents per gram (µmol TE/g) sample on dry weight basis.
4. DPPH assay
   NOTE: Determine the DPPH radical scavenging capacity of the samples using the method described previously by F. B. Apea-Bah et al.¹³. The DPPH antioxidant assay requires a radical-generating compound, DPPH (2,2-diphenyl-1-picrylhydrazyl).
   1. Accurately weigh 1.2 mg of DPPH into an empty 50 mL capacity centrifuge tube. Dissolve the DPPH in 30 mL of methanol to prepare a 60 µmol/L methanolic solution.
      NOTE: The DPPH assay tests the ability of the sample extracts to scavenge free radicals produced by DPPH.
   2. For the analysis, add 5 µL of the sample extracts or Trolox solution into the microplate wells.
   3. Next, add 195 µL of the 60 µmol/L DPPH methanolic solution and incubate for 60 min.
   4. Measure the absorbance of the reaction mixture at 515 nm.
   5. Use the Trolox standards (50-800 µmol/L) to plot a calibration curve, with the change in absorbance on the vertical axis and the Trolox concentrations on the horizontal axis.
   6. Estimate the DPPH scavenging capacity from the calibration curve.
   7. Express the results as micromole Trolox equivalents per gram (µmol TE/g) sample on dry weight basis.
5. Oxygen radical absorbance capacity
   1. Determine the oxygen radical absorbance capacity (ORAC) of the samples based on the method of F. B. Apea-Bah et al.¹³.
   2. To begin, prepare Trolox standards of concentrations 6.25, 12.5, 25, and 50 µM from a 1,000 µM Trolox standard stock solution in 75 mM potassium phosphate (K₂HPO₄/KH₂PO₄) buffer.
   3. To do this, weigh 1 mg Trolox powder and dissolve in 4 mL of the buffer solution under sonication to prepare a 1,000 µM stock solution.
   4. Then, take 50 µL of the stock solution into a 2 mL capacity centrifuge tube and dilute with 950 µL of buffer solution to obtain 1,000 µL of 50 µM Trolox solution.
   5. Pipette 500 µL of the 50 µM solution into a new tube and dilute with equal volume of buffer to obtain a 25 µM Trolox solution.
   6. Repeat the serial dilution of the 25 µM to obtain 12.5 µM, and then similarly repeat the dilution of 12.5 µM to obtain 6.25 µM.
   7. Prepare appropriate dilutions of the sample extracts.
   8. Dilute the yellow corn and cowpea extracts 20 times, the wheat and kidney bean extracts 50 times, and the soybean extract 100 times with the buffer solution.
   9. Transfer 200 µL of each sample or Trolox standard into the wells of a black 96-well microplate for automatic pipetting.
   10. Then, fill the three reagent tanks provided with the ORAC equipment, with the following: (1) the buffer solution; (2) 0.816 nM fluorescein in the buffer; and (3) 153 mM of 2,2′-azobis(2-amidinopropane) dihydrochloride (AAPH) in buffer.
   11. Place them in their receptacles for automatic pipetting.
   12. Set up the ORAC machine and automatic pipetting system for analysis, based on the standard operating procedure.
   13. For the analysis, set up the automated pipetting system to transfer 25 µL of each diluted extract or standard to the wells of a clear bottom black 96-well microplate.
   14. Then, automatically add 150 µL of 0.816 nM fluorescein in buffer.
   15. Incubate the reaction mixture at 37 °C for 15 min in the ORAC machine.
   16. Afterwards, automatically add 25 µL of the AAPH to each reaction mixture.
   17. Incubate them at 37 °C for 50 min in the ORAC machine.
   18. Set up the ORAC machine to measure the fluorescence decay, over the incubation period, at excitation and emission wavelengths of 485 nm and 520 nm, respectively.
   19. After the measurements, plot a Trolox standard curve, with Trolox (6.25-50 µM) on the horizontal axis and fluorescence decay on the vertical axis.
   20. Estimate the oxygen radical absorbance capacity of the extracts from the Trolox standard curve.
   21. Express the results as µmol TE/g sample on dry weight basis.
6. Statistical analysis
   1. Present all the results as means ± standard deviation of at least triplicates.
   2. Perform analysis of variance (ANOVA) to determine the effect of grain type on the response variables.
   3. Where significant differences exist at p < 0.05, use least significant difference (LSD) to compare the means.
   4. Perform Pearson correlation analysis to estimate the relationship between the phenolic content and antioxidant capacities.

Representative Results

Table 2 shows the phenolic acids that were identified in the cereal and legume grains. Based on available authentic standards, four phenolic acids were identified in the samples and they are: vanillic, caffeic, p-coumaric, and ferulic acids. Vanillic acid is a hydroxybenzoic acid while the other three are hydroxycinnamic acids. Vanillic acid was identified only in Blackeye cowpea bean while caffeic acid was identified only in kidney bean. p-Coumaric acid was identified in yellow corn, cowpea bean, and soybean. Its concentration ranged between 7.57 and 12.48 µg/g flour on dry weight basis, with yellow corn having the lowest value while soybean had the highest value. Ferulic acid was the only phenolic acid that was identified in all the samples. Its concentration ranged between 5.69 and 41.76 µg/g flour on dry weight basis. Among the samples, ferulic acid concentration was highest in soybean followed by cowpea bean while wheat, yellow corn, and kidney bean had comparable values (Table 2). When all the phenolic acid concentrations were summed up for each sample, their values decreased in the following order: soybean > cowpea bean > yellow corn = kidney bean > wheat.

Table 3 showed the total phenolic content (TPC) and antioxidant capacity of the cereal and legume grains. The antioxidant capacity comprised DPPH radical scavenging capacity, TEAC, and ORAC. The summation of these three values therefore gave the total antioxidant capacity of the samples. TPC was measured both in ferulic acid equivalent and gallic acid equivalent for the purpose of comparison. TPC ranged between 1.16 and 2.78 mg FAE/g flour and 0.63 to 1.48 mg GAE/g flour, on dry weight basis. Irrespective of the equivalent unit, TPC of the samples decreased in the following order: soybean > wheat = yellow corn = kidney bean > cowpea bean.

DPPH radical scavenging capacity of the samples ranged between 4.48 and 14.87 µmol TE/g flour on dry weight basis. Soybean had the highest DPPH scavenging capacity, followed by kidney bean, while yellow corn and cowpea bean had comparable values. Wheat had the lowest ability to scavenge the DPPH radical. The Trolox equivalent antioxidant capacity (TEAC) of the samples ranged between 12.82 and 57.24 µmol TE/g flour on dry weight basis. Again, Soybean had the highest TEAC value, followed by kidney bean, and then wheat, while yellow corn and cowpea bean had comparably low TEAC values. Oxygen radical absorbance capacity of the samples were between 0.35 and 1.67 µmol TE/g flour on dry weight basis, with the cereal grains having the lowest value while soybean had the highest value. When all the antioxidant capacities were summed up, their values in the samples decreased in the following order: soybean > kidney bean > yellow corn = cowpea bean > wheat.

There was a strong positive correlation between the two TPC values, and the values for TEAC, ORAC and total antioxidant capacity (TAC) (Table 4). However, there was weak correlation between the TPC values and DPPH.

Figure 1: Hydroxybenzoic acids identified in cereals and legumes. Please click here to view a larger version of this figure.

Figure 2: Hydroxycinnamic acids identified in cereals and legumes. Please click here to view a larger version of this figure.

Table 1: Preparation of Trolox concentrations for DPPH and ABTS standard curves. Please click here to download this Table.

Table 2: Phenolic acid content of some whole grain cereals and legumes. Please click here to download this Table.

Table 3: Total phenolic content (TPC) and antioxidant capacity of some whole grain cereals and legumes. Please click here to download this Table.

Table 4: Pearson correlation matrix for total phenolic content and antioxidant capacities of some cereal and legume grains. Please click here to download this Table.

Discussion

The whole grains were selected as representative cereal grains and legumes that find wide food applications worldwide. While variations may exist among cultivars of each grain, the focus of this study was to demonstrate a generalized method for free phenolic acid extraction and analysis for whole grains. The extraction method was modified by substantially reducing the amounts of samples and solvents, in order to reduce the amount of chemicals that would be released into the environment when such experiments are conducted. The modification also enables phenolic extraction from milligram quantities of whole grains.

HPLC analysis produces a chromatogram of the constituent phenolic acids in the sample. Each chromatographic peak representing a phenolic acid, can be identified by comparing its retention time, at a predefined wavelength, with that of authentic phenolic acid standards. Ultraviolet (UV) detectors are normally used to identify phenolic acids within the UV wavelength range. Hydroxybenzoic acids are usually identified between 254-280 nm while hydroxycinnamic acids are usually identified between 300-330 nm¹⁵. Photodiode array detectors can be used to scan the entire UV-visible wavelength range, and are particularly useful in determining the UV-visible spectra of compounds present in samples that have never been studied. R. J. Robbins and S. R. Bean¹⁵ reported the following wavelengths for maximum absorption, for vanillic acid, caffeic acid, p-coumaric acid, and ferulic acid, respectively: 260 nm; 325 nm; 310 nm; 325 nm. The HPLC method of identification is normally acceptable for samples of known phenolic acid composition, whose retention times and UV spectral data have been previously established. This is important because some compounds can co-elute when good separation is not achieved on the column. For samples of unknown phenolic acid composition, a mass spectrometer is used in addition to the authentic standards for identification. Published literature provides useful information about phenolic compounds that have been previously identified in the sample under study or in similar samples^3,9,14. It is worth noting that during HPLC analysis, phenolic acid peaks in the chromatogram are identifiable only when authentic standards are available for comparison. There may, therefore, be other peaks that will not be identifiable without their corresponding standards. This is a limitation with HPLC analysis which can be overcome when the HPLC is coupled to a mass spectrometer. Therefore, the total concentration of quantified phenolic acids will depend on the number of phenolic acids identified.

The Folin-Ciocalteu method, which was first published by V. L. Singleton and J. A. Rossi (1965)¹⁶ is an electron transfer-based assay that is used to estimate TPC of an analyte¹⁷. It is based on the ability of the analyte to reduce phosphomolybdate-phosphotungstate in an alkaline medium, thereby converting it from yellow to a dark blue solution¹⁸. The absorbance of the resulting solution can then be measured at 765 nm¹⁹. The reagent is not specific for phenolic compounds alone but can react with several other compounds with reducing properties such as thiols, reducing sugars and some amino acids (e.g., tyrosine). Therefore, it is suggested that TPC could be used to estimate the reducing property of a sample or analyte¹⁷. During the phenolic extraction, most interfering compounds are excluded and since phenolic compounds are the most ubiquitous phytochemicals, Folin-Ciocalteu assay remains a useful method to estimate the TPC of a sample. In most samples whose phenolic composition is not known, gallic acid is used as a standard to draw a calibration curve for estimating TPC. However, not all samples will contain gallic acid. In the current study, we used ferulic acid to compare its results to that of gallic acid. A two-sample T-test showed that TPC determined as ferulic acid equivalent had significantly higher values than TPC expressed as a gallic acid equivalent. Since we confirmed ferulic acid to be present in all the samples based on our HPLC analysis and other reported results, we would lean more toward expressing the TPC results as a ferulic acid equivalent. It is useful to express TPC relative to a predominant constituent phenolic compound in the sample.

Using three different antioxidant assays, it was observed that the different grains responded differently in their abilities to scavenge free radicals. DPPH radical scavenging capacity and TEAC assay are based on electron transfer (ET) mechanism while the ORAC assay is based on hydrogen atom transfer (HAT) mechanism²⁰. Therefore, combining the three assays gives a better indication of the antioxidant capacity of a sample. It is worth noting that, the ORAC assay measures the ability of a sample to scavenge the physiologically-relevant peroxyl radical, which, in a physiological system, can cause lipid peroxidation that is related to foam cell formation and atherogenesis^21,22. On the other hand, DPPH and TEAC assays use radicals that are not of physiological relevance. However, their ease of use makes them useful as rapid methods to estimate the radical scavenging capacity of a sample. ORAC assay also compare well with other in vitro or ex vivo antioxidant assays that use cells and DNA as biomarkers²³.

All the methods described above for the extraction and identification of free phenolic acids, estimation of total phenolic content, and determination of antioxidant properties, are not standardized. Different manufacturers of liquid chromatographic columns recommend different gradient elution systems of solvents to separate phenolic acids for their identification. Although the Folin-Ciocalteu method is the most common for estimating total phenolic content of samples, different laboratories perform this assay differently. The same can be said about the antioxidant methods. There is, therefore, the need for harmonization of these methods to promote inter-laboratory comparison of results. There is also the need for development of robust predictive models that can relate phenolic acid composition to the antioxidant properties measured by these important methods of analysis.

It is concluded from this study that the solvent extraction method used is effective in extracting free soluble phenolic acids from the flour of whole grain cereals and legumes, thereby facilitating their identification and quantification. The HPLC method enables identification and quantification of free soluble phenolic acids in whole grain extracts, provided there are authentic standards for comparison. Using the three different antioxidant methods: DPPH, TEAC, and ORAC, the antioxidant capacities of free soluble phenolic extracts from whole grains can be effectively determined based on the hydrogen atom transfer and electron transfer mechanisms. This study further confirms that whole grains differ in their phenolic acid composition and antioxidant capacities. The correlation analysis demonstrates that the antioxidant capacities measured are related to the constituent phenolic acids in the whole grain free soluble extracts. Although both TEAC and DPPH scavenge free radicals by electron transfer mechanism, they correlated differently with TPC. The difference in behaviors of different antioxidant assays necessitate the need to determine antioxidant capacity of samples using a variety of assays. Among the five whole grains studied, soybean has the highest phenolic acid concentration and correspondingly highest antioxidant capacity. The study also demonstrates that whole grain free soluble phenolic acids can be extracted and their antioxidant capacities studied in as little as 1 mL of aqueous methanolic solution, thereby reducing the amount of laboratory chemicals released into the environment.

Materials

Name	Company	Catalog Number	Comments
15 mL Falcon conical centrifuge tubes	Fisher Scientific	05-527-90
2 mL Amber glass ID Surestop vial	Thermo Scientific	C5000-2W
2 mL Amber microcentrifuge tubes	VWR	20170-084
2,2′-Azobis(2-amidinopropane) dihydrochloride (AAPH)	Sigma-Aldrich	440914-100G
2,2'-Azino-bis(3-ethylbenzothiazoline-6-sulfonic acid (ABTS) (C18H18N4O6S4) ≥98%,	Sigma Aldrich	A1888-2G
2,2-Diphenyl-1pikrylhydrazyl (DPPH) (C18H12N5O6)	Sigma Aldrich	D913-2
6-Hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox) (C14H18O4), ≥98%	Fluka Chemika	56510
9 mm Autosampler Vial Screw Thread Caps	Thermo Scientific	60180-670
96 well flat bottom plates	Fisher Scientific	12565501
Agilent BioTek ELx800 microplate reader	Fisher Scientific	BT-ELX800NB
Agilent BioTek Precision 2000 96/384 Automated Microplate Pipetting System	Fisher Scientific	N/A
Agilent BioTek FLx800 Microplate Fluorescence Reader	Fisher Scientific	N/A
Analytical balance SI-114	Denver Instrument	SI-114.1
Autosampler, Waters 717 Plus	Waters	WAT078900
BD 3 mL syringe Luer-Lok Tip	BD	309657
Bransonic ultrasonic cleaner, Branson 5510	Millipore Sigma	Z245143
Corning LSE Vortex Mixer	Corning	6775
Durapore Filter (0.45 µm PVDF Membrane)	Merck Millipore Ltd	HVLP04700
Durapore Membrane Filters (0.45 µm HV)	Merck Millipore Ltd	HVHP04700
Eppendorf Research plus, 0.5-10 µL	Eppendorf	3123000020
Eppendorf Research plus, 0.5-5 mL	Eppendorf	3123000071
Eppendorf Research plus, 100-1000 µL	Eppendorf	3123000063
Eppendorf Research plus, 10-100 µL	Eppendorf	3123000047
Ethyl acetate, HPLC grade	Fisher Chemical	E195-4
Ferulic acid standard	Sigma Aldrich	128708-5G
Fluorescein	Fisher Scientific	AC119245000
Folin & Ciocalteu phenol reagent	Sigma Aldrich	F9252
Formic acid, 99%	Acros Organics, Janssen Pharmaceuticalaan 3a	27048-0010
Gallic acid standard	Sigma	G7384
High performance liquid chromatograph (HPLC), Waters 2695	Waters	960402
Methanol, HPLC grade	Fisher Chemical	A452-4
Micro pipet tips, 0.5-10 µL	Fisherbrand	21-197-2F
Microcentrifuge Sorvall Legend Micro 21 centrifuge	Thermo Scientific	75002435
Multichannel micropipette, Proline Plus, 30-300 µL	Sartorius	728240
Photodiode array detector, Waters 2996	Waters	720000350EN
Pipet tips, 1000 µL	VWR	83007-382
Pipet tips, 1-5 mL	VWR	82018-840
Potassium persulfate (K2S2O8), ≥99.0%	Sigma Aldrich	216224-100G
Potassium phosphate dibasic anhydrous (K2HPO4)	Fisher Scientific	P288-500
Potassium phosphate monobasic (KH2PO4)	Fisher Scientific	P285-500
PYREX 250 mL Short Neck Boiling Flask, Round Bottom	Corning	4321-250
Reversed phase C18 Analytical Column (100 x 3 mm) Accucore aQ	Thermo Scientific	17326-103030
Roto evaporator, IKA RV 10	IKA	0010005185
Sodium carbonate (NaCO3) anhydrous	Fisher Chemical	S263-1
Sodium chloride (NaCl)	Mallinckrodt AR®	7581
Sodium phosphate dibasic anhydrous (Na2HPO4)	Fisher Scientific	BP332-500
Sodium phosphate monobasic anhydrous (NaH2PO4)	Fisher bioreagents	BP329-500
Standardization pipet tips 0-200µL	Fisherbrand	02-681-134
Syringe Driven Filter unit (0.22 µm)	Millex®-GV	SLGVR04NL
Target micro-serts vial insert (400 µL)	Thermo Scientific	C4011-631
Ultrapure water (Direct Q-3 UV system with pump)	Millipore	ZRQSVP030