Research Article Received: 14 January 2014
Revised: 16 April 2014
Accepted article published: 22 April 2014
Published online in Wiley Online Library: 21 May 2014
(wileyonlinelibrary.com) DOI 10.1002/jsfa.6706
Intra-laboratory validation of microplate methods for total phenolic content and antioxidant activity on polyphenolic extracts, and comparison with conventional spectrophotometric methods Gloria Bobo-García, Gabriel Davidov-Pardo,* Cristina Arroqui, Paloma Vírseda, María R Marín-Arroyo and Montserrat Navarro Abstract BACKGROUND: Total phenolic content (TPC) and antioxidant activity (AA) assays in microplates save resources and time, therefore they can be useful to overcome the fact that the conventional methods are time-consuming, labour intensive and use large amounts of reagents. An intra-laboratory validation of the Folin–Ciocalteu microplate method to measure TPC and the 2,2-diphenyl-1-picrylhydrazyl (DPPH) microplate method to measure AA was performed and compared with conventional spectrophotometric methods. RESULTS: To compare the TPC methods, the confidence intervals of a linear regression were used. In the range of 10–70 mg L−1 of gallic acid equivalents (GAE), both methods were equivalent. To compare the AA methodologies, the F-test and t-test were used in a range from 220 to 320 𝛍mol L−1 of Trolox equivalents. Both methods had homogeneous variances, and the means were not significantively different. The limits of detection and quantification for the TPC microplate method were 0.74 and 2.24 mg L−1 GAE and for the DPPH 12.07 and 36.58 𝛍mol L−1 of Trolox equivalents. The relative standard deviation of the repeatability and reproducibility for both microplate methods were ≤6.1%. The accuracy ranged from 88% to 100%. CONCLUSION: The microplate and the conventional methods are equals in a 95% confidence level. © 2014 Society of Chemical Industry Keywords: Folin–Ciocalteu; DPPH; grape seed extract; apple extract; green tea extract
INTRODUCTION
204
Interest in the research of polyphenols from different natural sources has grown because polyphenols can be utilised as antioxidants in the food industry, and they benefit human health in various ways. The beneficial effects of polyphenols on human health could be due to their free radical scavenger properties, blocking the deleterious action of these molecules on cells.1 In addition to their antioxidant capacity polyphenols can benefit human health in other metabolic ways such as prevention of cardiovascular diseases, anti-inflamatory effects and cancer prevention.2 – 4 Due to all the above mentioned, antioxidants may be the most promising functional ingredient to add to a product.5 Considering that the amount of polyphenols and their antioxidant activity is closely related to their action as food additives or functional ingredients, a fast, cheap and accurate method to measure the total phenolic content and antioxidant activity of plant extracts is needed. Nowadays, the most common method to measure the total phenolic content of all types of sample is the Folin–Ciocalteu method, which is based in the reduction of the phospho-molybdate heteropoly acids Mo(VI) centre in the J Sci Food Agric 2015; 95: 204–209
heteropoly complex to Mo(V), resulting in a blue colouration which is measured at around 750 nm.6 To measure the antioxidants scavenging activity DPPH• is considered a valid; and easy method, because the radical compound (2,2-diphenyl1-picrylhydrazyl, DPPH) is stable and does not have to be generated hours before the analysis, as in other radical scavenging assays.7 The quantification of the antioxidant activity is based on decolourisation of the reagent at around 515 nm.8 Generally, the conventional methods to assess the total phenolic content and antioxidant activity of a sample are time-consuming, labour intensive and use large amounts of reagents.9,10 To overcome these drawbacks, the assays in microplates have been done with positive results on a wide variety of samples: seaweeds,10 sorghum,9
∗
Correspondence to: Gabriel Davidov-Pardo, Public University of Navarra, Food Technology Department, Campus Arrosadia s/n, Pamplona 31006, Spain. E-mail: [email protected] Public University of Navarra, Food Technology Department, Campus Arrosadia s/n, Pamplona 31006, Spain
www.soci.org
© 2014 Society of Chemical Industry
Comparison of microplate and conventional methods for Folin- Ciocalteu and DPPH berries11 – 13 and Cyphostemma digitatum.14 In some cases after analysing a determined number of samples, they showed the values obtained with the new and the reference method. They showed the recovery percentage, precision and reproducibility of the new method.9,15 But statistical comparison between the results obtained with the microplate and the conventional methods have not yet been reported. In the light of this background, the aim of this study was to validate Folin–Ciocalteu and DPPH microplate methods and compare them with the conventional ones, using statistical methodologies.
MATERIAL AND METHODS Samples Grape seed extract (GSE) was provided by Puleva Biosearch (Granada, Spain). Apple extract (AE) was purchased from Principium (Viganello, Switzerland). GSE and AE are commercial extracts, obtained through a hydro-alcoholic extraction. Green tea (Camellia sinensis) extract (GTE) was made by infusing 10 g of green tea (SoriaNatural, Gargay, Spain) in 60 mL of deionised water at 55 ∘ C for 15 min. Then, the plant material was separated using a cheesecloth and the infusion was vacuum filtered.16 The samples were stored at 4 ∘ C until used. Chemicals Methanol, ethanol–water (96:4, v/v) and sodium carbonate pharmaceutical grade and gallic acid 1-hydrate analytical grade were purchased from Panreac (Barcelona, Spain). 2,2-Diphenyl1-picrylhydrazyl (DPPH) and Folin–Ciocalteu reagent analytical grade were purchased from Sigma Chemical Co (St. Louis, MO, USA). 6-Hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox; 970 g kg−1 ) analytical grade was purchased from Aldrich Chemical (Steinheim, Germany). Total phenolic content Conventional method The conventional total phenolic content (TPC) method is based on the Folin–Ciocalteu European Commission Regulation method.6 In a100- mL volumetric flask, 1 mL of the diluted extract, 50 mL of deionised water type I and 5 mL of the Folin–Ciocalteu reagent were added and left to react for 300 s. To complete the reaction, 20 mL of a sodium carbonate solution (200 g L−1 ) were added, and the volumetric flask was filled to its volume with deionised water. After 30 min at room temperature, the absorbance of the samples at 750 nm was measured in polystyrene cuvettes in a Thermo Scientific Multiskan GO spectrophotometer (ThermoFisher Scientific, Vartaa, Finland). The measured absorbance of the same reaction with water instead of the extract or standard was subtracted from the absorbance of the reaction with the sample. The phenolic content was expressed in gallic acid equivalents per litre after the preparation of a standard curve of gallic acid from 10 to 200 mg L−1 .
J Sci Food Agric 2015; 95: 204–209
then 75 μL of sodium carbonate solution (100 g L−1 ) were added and the mixture was shaken at medium-continuous speed for 1 min. After 2 h at room temperature, the absorbance was measured at 750 nm using the microplate reader of a Thermo Scientific Multiskan GO spectrophotometer (ThermoFisher Scientific). The absorbance of the same reaction with water instead of the extract or standard was subtracted from the absorbance of the reaction with the sample. Gallic acid dilutions (10–200 mg L−1 ) were used as standards for calibration. Antioxidant activity DPPH conventional method The conventional antioxidant activity (AA) was evaluated based on the technique by Rivero-Pérez et al.8 In a polystyrene cuvette, 2940 μL of DPPH dissolved in methanol (60 μmol L−1 ) were mixed with 60 μL of the diluted extract. The absorbance at 515 nm was measured after 60 min in the dark using a Thermo Scientific Multiskan GO spectrophotometer (ThermoFisher Scientific). The % DPPH quenched was calculated using Eqn 1 and reported as μmol L−1 of Trolox equivalents after the construction of a standard curve of Trolox (50–500 μmol L−1 ): [ ( )] Asample − Ablank % DPPH quenched = 1 − × 100 (1) Acontrol − Ablank where Asample is the absorbance at 515 nm of 60 μL of extract or standard with 2940 μL of DPPH solution after 60 min, Ablank is the absorbance at 515 nm of 3000 μL of methanol, Acontrol is the absorbance at 515 nm of 60 μL of water with 2940 μL of DPPH solution after 60 min. DPPH microplate method The microplate AA methodology was based on the 96-well plate assay described by Herald et al.9 with some modifications. A total of 20 μL of the diluted sample was added to 180 μL of DPPH solution (150 μmol L−1 ) in methanol–water (80:20, v/v) and shaken
(a)
(b)
Figure 1. Standard curves for microplate assays. (a) Total phenolic content (Folin–Ciocalteu assay); (b) antioxidant activity (DPPH assay).
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
205
Microplate method The microplate TPC method was based on the 96-well microplate Folin–Ciocalteu method given by Al-Duais et al.14 and Müller et al.17 with some modifications. A total of 20 μL of the diluted extract were mixed with 100 μL of 1:4 diluted Folin–Ciocalteu reagent and shaken for 60 s in a flat-bottom 96-well microplate (NUNC, Roskilde, Denmark). The mixture was left for 240 s and
www.soci.org
www.soci.org for 60 s in a 96-well microplate (NUNC). After 40 min in the dark at room temperature, the absorbance was measured at 515 nm in the microplate reader of a Thermo Scientific Multiskan GO spectrophotometer (ThermoFisher Scientific). Trolox was used as a standard at 50–500 μmol L−1 to generate a calibration curve. The % DPPH quenched was calculated using Eqn 1, where Asample is the absorbance at 515 nm of 20 μL of extract or standard with 180 μL DPPH solution after 40 min; Ablank is the absorbance at 515 nm of 20 μL of water with 180 μL methanol–water (80:20, v/v) after 40 min, and Acontrol is the absorbance at 515 nm of 20 μL of water with 180 μL DPPH solution after 40 min. Limit of detection and quantification The limit of detection (LOD) of an individual analytical procedure is the lowest amount of analyte in a sample which can be detected but not necessarily quantified as an exact value.18 The LOD was calculated using Eqn (2): LOD =
3.3𝜎 S
(2)
where 𝜎 is the standard deviation of the blank and S is the slope of the calibration curve.
(a)
G Bobo et al.
The limit of quantification (LOQ) of an individual analytical procedure is the lowest amount of analyte in a sample which can be quantitatively determined with suitable precision and accuracy.18 The LOQ was calculated using Eqn (3): LOQ =
10𝜎 S
(3)
where 𝜎 is the standard deviation of the blank and S is the slope of the calibration curve. Statistical comparison The confidence intervals of the slope and the constant of a linear regression equation were used to compare the TPC conventional and microplate methods.19 A linear regression of five different concentrations of the GSE, AE and GTE was performed, where the results of the conventional TPC method represent the independent variable while the results of the microplate TPC represent the dependent variable. To compare the conventional and microplate AA a statistical comparison based on the F-test and t-test of three concentrations of the GSE, AE and GTE was performed. To consider both methods statistically equal with a confidence of 95%, the P-value for the F-test and t-test has to be higher than 0.05.19 For the TPC method comparison, five concentrations of GSE (12, 24, 36, 48 and 60 mg L−1 ) and five concentrations of AE (24, 48, 54, 60 and 72 mg L−1 ) were prepared in ethanol–water solution (20:80, v/v). Five dilutions (1:10, 1:5, 1:4, 1:3.3 and 1:2.5) from a 1:20 diluted GTE were also used to compare the TPC methods. To compare the AA methods, three concentrations of GSE (35, 40 and 45 mg L−1 ) and AE (50, 60 and 70 mg L−1 ) were prepared as in TPC assays and three dilutions (1:2.5, 1:2 and 1:1.7) from a 1:50 diluted GTE were used. Precision and accuracy Precision is based on the repeatability and reproducibility, which can be expressed as the relative standard deviation (RSD). Accuracy expresses the closeness of a result to a true value.19,20 In this work, the repeatability was the RSD calculated from five repetitions of the analysis of the same sample on the same day and conditions. The reproducibility was the RSD calculated from five repetitions of the analysis in different days and daily prepared reagents. The accuracy was calculated as the percentage of recovery.9,15
(b)
Statistical analyses The statistical analyses were performed using Statgraphics Centurion XVI (Statpoint Technologies, Virginia, USA) and Minitab 16 (Minitab Inc., State College, PA, USA) software.
(c)
Table 1. Confidence intervals at 95% for the estimated coefficients Confidence levels Extract
Parameter
Estimated coefficient
Grape seed
Slope Constant Slope Constant Slope Constant
0.9982 −1.1455 0.9656 −1.3315 0.9838 −0.2523
Apple Green tea
206
Figure 2. Linear regression for total phenolic content methods comparison. (a) Grape seed extract; (b) apple extract; (c) green tea extract.
wileyonlinelibrary.com/jsfa
© 2014 Society of Chemical Industry
Lower
Upper
0.9551 −2.7246 0.9248 −2.8692 0.9674 −1.0406
1.0412 0.4336 1.0064 0.2061 1.0002 0.5361
J Sci Food Agric 2015; 95: 204–209
Comparison of microplate and conventional methods for Folin- Ciocalteu and DPPH
RESULTS AND DISCUSSION Calibration curves and limits of detection and quantitation The standard calibration curves for the microplate methods are presented in Fig. 1. The curves are linear when the concentration of gallic acid is in the range of 10–200 mg L−1 (R2 = 0.9998) and the concentration of Trolox is in the range of 50–500 μmol L−1 (R2 = 0.9999). When comparing these curves with those obtained by Herald et al.,9 who used similar methodologies and the same concentration ranges, it can be seen that the slight changes in the methodology had more influence upon the slope of the AA assay than on the slope of the TPC assay. The slope of the calibration curve for the AA in this study was 0.1647, while the slope in their study was 0.1512; for the calibration curves of the TPC the slopes were 0.0076 and 0.0073, respectively. However, comparing the slopes of this work with those obtained by Horszwald and Andlauer,13 TPC = 3.1761 and DPPH = 0.0591, it can be seen that greater differences in the TPC and DPPH microplate methods had a significant impact on the sensitivity of the methods. That work had also calculated the limits of detection and/or quantification for their methods. In this work, the LOD and LOQ for the Folin–Ciocalteu microplate method were 0.74 and 2.24 mg L−1 GAE, respectively. In the case of the DPPH microplate method the LOD and LOQ were 12.07 and 36.58 μmol L−1 of Trolox equivalents, respectively. As expected, because of the greater sensitivity of the Horszwald and Andlauer method,13 the LOQ (0.015 mg L−1 GAE) of their TPC microplate method is lower than the LOQ of this work. On the contrary, the greater sensitivity of DPPH microplate method of this work, resulted in a lower LOD than that obtained by Horszwald and Andlauer,13 which was 50 μmol L−1 of Trolox. Statistical comparison Preliminary studies The dilutions to perform the TPC and the AA comparisons were established to obtain 10 to 70 mg L−1 of gallic acid equivalents and μmol L−1 concentrations from 220 to 320 Trolox equivalents,
www.soci.org
respectively. These were the broadest concentration ranges in which the three extracts in the microplate method gave determination coefficients higher than 0.999. The determination coefficients were obtained by a linear regression of the dilutions of each extract and the resulted gallic acid or Trolox equivalents concentrations. The regression residues were tested for a normal distribution and homoscedasticity. The Ryan–Joiner coefficient of correlation (P > 0.05) indicated a normal distribution; the Levene test (P > 0.05) confirmed homoscedasticity; finally the Durbin–Watson test (P > 0.05) showed independence of the residues. Total phenolic content To statistically compare the TPC conventional and microplate methods, the confidence intervals of a linear regression were used. To generate the graphic and the linear regression equation, the abscissa represents the values of the reference method (conventional TPC method) and the ordinate shows the values of the method to evaluate (microplate TPC method). Figure 2 shows the linear regression of the TPC comparison using the GSE, AE and GTE. Table 1 shows the confidence values for the slope and the constant. If the slope confidence intervals do not include a value of 1, it means that there is an error proportional to the concentration of the analyte. If the constant confidence intervals do not include a value of zero, it means that there is a systematic excess or defect error.19 The methods in the studied range of concentrations can be considered equivalent because, as shown in Table 1, the confidence intervals of the slope and the constant meet the above mentioned requirements. The changes made in the TPC method used by Al-Duais et al.14 and Müller et al.17 were made to increase the concentration of the Folin–Ciocalteu reagent compared to the concentration of the sample. This increment intended to bring the proportions of the microplate method closer to those used in the conventional method. Due to the increase in the Folin–Ciocalteu reagent the desired statistical values were obtained to consider both method equivalents. It is important to remark that the proportions of
Table 2. Validation of DPPH conventional and microplate methods by using F of the Fisher test and t of the Student test Fisher test Extract Grape seed
Concentration or dilution
DPPH methoda
Mean ± SD (μmol L−1 Trolox)
35 mg L−1
C M C M C M C M C M C M C M C M C M
222.7 ± 6.9 222.3 ± 2.5 250.5 ± 2.7 251.3 ± 1.8 274.6 ± 8.4 277.4 ± 6.3 229.1 ± 3.0 225.8 ± 3.0 264.9 ± 3.3 265.4 ± 2.3 302.3 ± 3.6 305.6 ± 4.6 253.9 ± 2.9 254.2 ± 7.8 308.1 ± 5.7 304 ± 15 324.9 ± 8.5 317 ± 18
40 mg L−1 45 mg L−1 Apple
50 mg L−1 60 mg L−1 70 mg L−1
Green tea
1:2.5 1:2.0 1:1.7
F
Student test
P-value
t
P-value
7.7484
0.07
0.1159
0.91
2.2287
0.46
−0.5255
0.61
1.7686
0.59
−0.5899
0.57
1.0325
0.98
1.7240
0.12
1.9642
0.53
−0.2590
0.80
0.6138
0.65
−1.2998
0.23
0.1378
0.08
−0.0815
0.94
0.1420
0.08
0.6061
0.56
0.2225
0.17
0.8542
0.42
J Sci Food Agric 2015; 95: 204–209
207
a C, conventional; M, microplate. DPPH, 2,2-Diphenyl-1-picrylhydrazyl.
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
www.soci.org reagents and sample of both methods could not be equal because there was bubbling and precipitation during the whole assay and the measurements were unstable. This phenomenon was also reported by Herald et al.9 when they used the Folin–Ciocalteu reagent undiluted. Antioxidant activity To compare the AA conventional and microplate methodologies three different concentration levels of grape seed, apple and green tea extracts were used. Table 2 shows the resulting P-values of the F parameter of the Fisher test and the t parameter of the Student test. In all cases the P-values for the F parameter of the Fisher test were above 0.05, which indicates that the variances among the measurements for each concentration (n = 5) were not significantly different at a 95% confidence level. The lack of significant differences in the variance indicates that both methods are equally precise and allows the performance of the t parameter of the Student test.19 All the P-values for the t parameter of the Student test were above 0.05. It indicates that there are not significant differences between the mean values for both methods at a 95% confidence level.19 Based on the P-values resulting from both statistical tests, it can be said that the conventional and the microplate methods were equivalent in the concentration ranges studied. The equivalency of both methods to measure the AA of the samples indicates that the increase in the concentration of the sample for the microplate analysis was compensated with the increase in the molar concentration of the DPPH. The higher ratio of the sample and the reagent in the microplate methodology could contribute to reduce the time to reach a stable absorbance compared to the time required with conventional method, which is one of the requirements for a correct AA analysis.11 Finally, the addition of 20% of water to the DPPH reagent in the microplate methodology did not affect the results when compared to the conventional methodology and together with the reduction of time, prevented the evaporation of the solvent leading to more stable measurements.11 Precision and accuracy The Ryan–Joiner and Levene test (P > 0.05) confirmed a normal distribution and homoscedasticity for each day and concentration. Table 3 shows the repeatability, reproducibility and percentage of recovery for five gallic acid and Trolox concentrations. The concentrations were chosen within the ranges used to compare the conventional and microplate methods. It can be seen that the RSDs for the repeatability were ≤3.6% and the RSDs for the reproducibility were ≤6.1% for both methods. RSD values below 10% indicate good precision.9 The repeatability and reproducibility results and the fact that in general, the RSDs of the TPC are higher than those of the AA, are in agreement with the findings by Cheng et al.15 and Herald et al.9 The percentages of recovery (87.8–100.3%) suggest an excellent accuracy for both methods within the studied concentrations.19
CONCLUSIONS
208
In the range of 10–70 mg L−1 GAE and 220–320 μmol L−1 of Trolox equivalents for the Folin–Ciocalteu and the DPPH techniques, respectively, the microplate and the conventional methods are equal at a 95% confidence level. The repeatability, reproducibility and percentage of recovery for the TPC and AA microplate
wileyonlinelibrary.com/jsfa
G Bobo et al.
Table 3. Repeatability, intra-laboratory reproducibility and percentage of recovery for total phenol content and antioxidant activity with the conventional and microplate methods Method
Standard Repeatability Reproducibility
Folin–Ciocalteu microplate
DPPH microplate
Recovery (%)
12
3.6
5.2
87.8
24 36 48 60 224 246 268 293 315
2.2 2.4 2.7 2.1 3.4 2.4 1.0 2.2 2.1
2.9 6.1 5.3 3.9 3.2 2.7 1.9 2.4 3.2
97.9 98.3 99.3 100.3 98.3 99.2 97.8 96.4 95.6
For the Folin–Ciocalteu method, the standard was gallic acid (mg L−1 ). For the DPPH method, the standard was Trolox (μmol L−1 ). Repeatability and reproducibility are given as % relative standard deviation for both methods.
methods showed a precision below 6% and accuracy between 88% and 100%. The use of the microplate methods saves resources and time, which makes it more suitable than the conventional ones.
ACKNOWLEDGEMENTS We thank Exxentia (Spain) for kindly providing the samples. This work was partially financed by the Navarra government (Spain) and by the Public University of Navarre (Spain).
REFERENCES 1 Murcia MA, Vera AM, Martínez-Tomé M and Frega N, Sustancias antioxidantes presentes en los alimentos. Acción, dosis y su eficacia en la promoción de la salud. Nueva Imprenta S.A., Madrid (2003). 2 Hollman PCH, Cassidy A, Comte B, Heinonen M, Richelle M, Richling E, et al, The biological relevance of direct antioxidant effects of polyphenols for cardiovascular health in humans is not established. J Nutr 141:989S–1009S (2011). 3 Sies H, Polyphenols and health: Update and perspectives. Arch Biochem Biophys 501:2–5 (2010). 4 Valenzuela A, Tea consumption and health: Beneficial characteristics and properties of this ancient beverage. Revista Chilena de Nutrición Versión Online. 31:72–82 (2004). 5 Jesionkowska K, Sijtsema SJ, Konopacka D and Symoneaux R, Dried fruit and its functional properties from a consumer’s point of view. J Hort Sci Biotechnol 84:85–88 (2009). 6 EEC, Commission Regulation (EEC) N∘ 2676/90 determining community methods for the analysis of wine. pp. 178–179 (1990). [Online]. Available: http://eur-lex.europa.eu/legal-content/EN/ ALL/?uri=CELEX:31990R2676 [11 March 2013]. 7 Sánchez-Moreno C, Review: Methods used to evaluate the free radical scavenging activity in foods and biological systems. Food Sci Technol Int 8:121–137 (2002). 8 Rivero-Pérez MD, Muñiz P and González-Sanjosé ML, Antioxidant profile of red wines evaluated by total antioxidant capacity, scavenger activity, and biomarkers of oxidative stress methodologies. J Agric Food Chem 55:5476–5483 (2007). 9 Herald TJ, Gadgil P and Tilley M, High-throughput micro-plate assays for screening flavonoid content and DPPH-scavenging activity in sorghum bran and flour. J Sci Food Agric 92:2326–2331 (2012). 10 Zhang Q, Zhang J, Shen J, Silva A, Dennis DA and Barrow CJ, A simple 96-well microplate method for estimation of total polyphenol content in seaweeds. J Appl Phycol 18:445–450 (2006).
© 2014 Society of Chemical Industry
J Sci Food Agric 2015; 95: 204–209
Comparison of microplate and conventional methods for Folin- Ciocalteu and DPPH 11 Fukumoto LR and Mazza G, Assessing antioxidant and prooxidant activities of phenolic compounds. J Agric Food Chem 48:3597–3604 (2000). 12 Schmidt BM, Erdman Jr JW and Lila MA, Effects of food processing on blueberry antiproliferation and antioxidant activity. J Food Sci 70:S389–S394 (2005). 13 Horszwald A and Andlauer W, Characterisation of bioactive compounds in berry juices by traditional photometric and modern microplate methods J Berry Res 1:189–199 (2011). 14 Al-Duais M, Müller L, Böhm V and Jetschke G, Antioxidant capacity and total phenolics of Cyphostemma digitatum before and after processing: Use of different assays. Eur Food Res Technol 228:813–821 (2009). 15 Cheng Z, Moore J and Yu L, High-throughput relative DPPH radical scavenging capacity assay. J Agric Food Chem 54:7429–7436 (2006).
www.soci.org
16 Essa HA, Nadir AES and Hamad KI, Effect of anti-browning and anti-microbial constituents of some spices on the quality and safety of apple slices. Modelling, Measurement and Control C 64:15–33 (2004). 17 Müller L, Gnoyke S, Popken AM and Böhm V, Antioxidant capacity and related parameters of different fruit formulations. LWT – Food. Sci Technol 43:992–999 (2010). 18 ICH, Validation of Analytical Procedures: Text and Methodology, Q2(R1). European Medicines Agency, London, pp. 11–12 (2005). 19 Aguirre OL, García GFJ, García JT, Illera FM, Juncadella RM, Lizondo CM, et al, Validación de métodos analíiticos, in Monografías de AEFI, ed. by Pérez CJA and Pujol FM. Asociación Española de Farmacéuticos de la Industria, Barcelona, pp. 79–83 (2001). 20 EURACHEM, The Fitness for Purpose of Analytical Methods: A Laboratory Guide to Method Validation and Related Topics. Eurachem, Middlesex (1998).
209
J Sci Food Agric 2015; 95: 204–209
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
Revised: 16 April 2014
Accepted article published: 22 April 2014
Published online in Wiley Online Library: 21 May 2014
(wileyonlinelibrary.com) DOI 10.1002/jsfa.6706
Intra-laboratory validation of microplate methods for total phenolic content and antioxidant activity on polyphenolic extracts, and comparison with conventional spectrophotometric methods Gloria Bobo-García, Gabriel Davidov-Pardo,* Cristina Arroqui, Paloma Vírseda, María R Marín-Arroyo and Montserrat Navarro Abstract BACKGROUND: Total phenolic content (TPC) and antioxidant activity (AA) assays in microplates save resources and time, therefore they can be useful to overcome the fact that the conventional methods are time-consuming, labour intensive and use large amounts of reagents. An intra-laboratory validation of the Folin–Ciocalteu microplate method to measure TPC and the 2,2-diphenyl-1-picrylhydrazyl (DPPH) microplate method to measure AA was performed and compared with conventional spectrophotometric methods. RESULTS: To compare the TPC methods, the confidence intervals of a linear regression were used. In the range of 10–70 mg L−1 of gallic acid equivalents (GAE), both methods were equivalent. To compare the AA methodologies, the F-test and t-test were used in a range from 220 to 320 𝛍mol L−1 of Trolox equivalents. Both methods had homogeneous variances, and the means were not significantively different. The limits of detection and quantification for the TPC microplate method were 0.74 and 2.24 mg L−1 GAE and for the DPPH 12.07 and 36.58 𝛍mol L−1 of Trolox equivalents. The relative standard deviation of the repeatability and reproducibility for both microplate methods were ≤6.1%. The accuracy ranged from 88% to 100%. CONCLUSION: The microplate and the conventional methods are equals in a 95% confidence level. © 2014 Society of Chemical Industry Keywords: Folin–Ciocalteu; DPPH; grape seed extract; apple extract; green tea extract
INTRODUCTION
204
Interest in the research of polyphenols from different natural sources has grown because polyphenols can be utilised as antioxidants in the food industry, and they benefit human health in various ways. The beneficial effects of polyphenols on human health could be due to their free radical scavenger properties, blocking the deleterious action of these molecules on cells.1 In addition to their antioxidant capacity polyphenols can benefit human health in other metabolic ways such as prevention of cardiovascular diseases, anti-inflamatory effects and cancer prevention.2 – 4 Due to all the above mentioned, antioxidants may be the most promising functional ingredient to add to a product.5 Considering that the amount of polyphenols and their antioxidant activity is closely related to their action as food additives or functional ingredients, a fast, cheap and accurate method to measure the total phenolic content and antioxidant activity of plant extracts is needed. Nowadays, the most common method to measure the total phenolic content of all types of sample is the Folin–Ciocalteu method, which is based in the reduction of the phospho-molybdate heteropoly acids Mo(VI) centre in the J Sci Food Agric 2015; 95: 204–209
heteropoly complex to Mo(V), resulting in a blue colouration which is measured at around 750 nm.6 To measure the antioxidants scavenging activity DPPH• is considered a valid; and easy method, because the radical compound (2,2-diphenyl1-picrylhydrazyl, DPPH) is stable and does not have to be generated hours before the analysis, as in other radical scavenging assays.7 The quantification of the antioxidant activity is based on decolourisation of the reagent at around 515 nm.8 Generally, the conventional methods to assess the total phenolic content and antioxidant activity of a sample are time-consuming, labour intensive and use large amounts of reagents.9,10 To overcome these drawbacks, the assays in microplates have been done with positive results on a wide variety of samples: seaweeds,10 sorghum,9
∗
Correspondence to: Gabriel Davidov-Pardo, Public University of Navarra, Food Technology Department, Campus Arrosadia s/n, Pamplona 31006, Spain. E-mail: [email protected] Public University of Navarra, Food Technology Department, Campus Arrosadia s/n, Pamplona 31006, Spain
www.soci.org
© 2014 Society of Chemical Industry
Comparison of microplate and conventional methods for Folin- Ciocalteu and DPPH berries11 – 13 and Cyphostemma digitatum.14 In some cases after analysing a determined number of samples, they showed the values obtained with the new and the reference method. They showed the recovery percentage, precision and reproducibility of the new method.9,15 But statistical comparison between the results obtained with the microplate and the conventional methods have not yet been reported. In the light of this background, the aim of this study was to validate Folin–Ciocalteu and DPPH microplate methods and compare them with the conventional ones, using statistical methodologies.
MATERIAL AND METHODS Samples Grape seed extract (GSE) was provided by Puleva Biosearch (Granada, Spain). Apple extract (AE) was purchased from Principium (Viganello, Switzerland). GSE and AE are commercial extracts, obtained through a hydro-alcoholic extraction. Green tea (Camellia sinensis) extract (GTE) was made by infusing 10 g of green tea (SoriaNatural, Gargay, Spain) in 60 mL of deionised water at 55 ∘ C for 15 min. Then, the plant material was separated using a cheesecloth and the infusion was vacuum filtered.16 The samples were stored at 4 ∘ C until used. Chemicals Methanol, ethanol–water (96:4, v/v) and sodium carbonate pharmaceutical grade and gallic acid 1-hydrate analytical grade were purchased from Panreac (Barcelona, Spain). 2,2-Diphenyl1-picrylhydrazyl (DPPH) and Folin–Ciocalteu reagent analytical grade were purchased from Sigma Chemical Co (St. Louis, MO, USA). 6-Hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox; 970 g kg−1 ) analytical grade was purchased from Aldrich Chemical (Steinheim, Germany). Total phenolic content Conventional method The conventional total phenolic content (TPC) method is based on the Folin–Ciocalteu European Commission Regulation method.6 In a100- mL volumetric flask, 1 mL of the diluted extract, 50 mL of deionised water type I and 5 mL of the Folin–Ciocalteu reagent were added and left to react for 300 s. To complete the reaction, 20 mL of a sodium carbonate solution (200 g L−1 ) were added, and the volumetric flask was filled to its volume with deionised water. After 30 min at room temperature, the absorbance of the samples at 750 nm was measured in polystyrene cuvettes in a Thermo Scientific Multiskan GO spectrophotometer (ThermoFisher Scientific, Vartaa, Finland). The measured absorbance of the same reaction with water instead of the extract or standard was subtracted from the absorbance of the reaction with the sample. The phenolic content was expressed in gallic acid equivalents per litre after the preparation of a standard curve of gallic acid from 10 to 200 mg L−1 .
J Sci Food Agric 2015; 95: 204–209
then 75 μL of sodium carbonate solution (100 g L−1 ) were added and the mixture was shaken at medium-continuous speed for 1 min. After 2 h at room temperature, the absorbance was measured at 750 nm using the microplate reader of a Thermo Scientific Multiskan GO spectrophotometer (ThermoFisher Scientific). The absorbance of the same reaction with water instead of the extract or standard was subtracted from the absorbance of the reaction with the sample. Gallic acid dilutions (10–200 mg L−1 ) were used as standards for calibration. Antioxidant activity DPPH conventional method The conventional antioxidant activity (AA) was evaluated based on the technique by Rivero-Pérez et al.8 In a polystyrene cuvette, 2940 μL of DPPH dissolved in methanol (60 μmol L−1 ) were mixed with 60 μL of the diluted extract. The absorbance at 515 nm was measured after 60 min in the dark using a Thermo Scientific Multiskan GO spectrophotometer (ThermoFisher Scientific). The % DPPH quenched was calculated using Eqn 1 and reported as μmol L−1 of Trolox equivalents after the construction of a standard curve of Trolox (50–500 μmol L−1 ): [ ( )] Asample − Ablank % DPPH quenched = 1 − × 100 (1) Acontrol − Ablank where Asample is the absorbance at 515 nm of 60 μL of extract or standard with 2940 μL of DPPH solution after 60 min, Ablank is the absorbance at 515 nm of 3000 μL of methanol, Acontrol is the absorbance at 515 nm of 60 μL of water with 2940 μL of DPPH solution after 60 min. DPPH microplate method The microplate AA methodology was based on the 96-well plate assay described by Herald et al.9 with some modifications. A total of 20 μL of the diluted sample was added to 180 μL of DPPH solution (150 μmol L−1 ) in methanol–water (80:20, v/v) and shaken
(a)
(b)
Figure 1. Standard curves for microplate assays. (a) Total phenolic content (Folin–Ciocalteu assay); (b) antioxidant activity (DPPH assay).
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
205
Microplate method The microplate TPC method was based on the 96-well microplate Folin–Ciocalteu method given by Al-Duais et al.14 and Müller et al.17 with some modifications. A total of 20 μL of the diluted extract were mixed with 100 μL of 1:4 diluted Folin–Ciocalteu reagent and shaken for 60 s in a flat-bottom 96-well microplate (NUNC, Roskilde, Denmark). The mixture was left for 240 s and
www.soci.org
www.soci.org for 60 s in a 96-well microplate (NUNC). After 40 min in the dark at room temperature, the absorbance was measured at 515 nm in the microplate reader of a Thermo Scientific Multiskan GO spectrophotometer (ThermoFisher Scientific). Trolox was used as a standard at 50–500 μmol L−1 to generate a calibration curve. The % DPPH quenched was calculated using Eqn 1, where Asample is the absorbance at 515 nm of 20 μL of extract or standard with 180 μL DPPH solution after 40 min; Ablank is the absorbance at 515 nm of 20 μL of water with 180 μL methanol–water (80:20, v/v) after 40 min, and Acontrol is the absorbance at 515 nm of 20 μL of water with 180 μL DPPH solution after 40 min. Limit of detection and quantification The limit of detection (LOD) of an individual analytical procedure is the lowest amount of analyte in a sample which can be detected but not necessarily quantified as an exact value.18 The LOD was calculated using Eqn (2): LOD =
3.3𝜎 S
(2)
where 𝜎 is the standard deviation of the blank and S is the slope of the calibration curve.
(a)
G Bobo et al.
The limit of quantification (LOQ) of an individual analytical procedure is the lowest amount of analyte in a sample which can be quantitatively determined with suitable precision and accuracy.18 The LOQ was calculated using Eqn (3): LOQ =
10𝜎 S
(3)
where 𝜎 is the standard deviation of the blank and S is the slope of the calibration curve. Statistical comparison The confidence intervals of the slope and the constant of a linear regression equation were used to compare the TPC conventional and microplate methods.19 A linear regression of five different concentrations of the GSE, AE and GTE was performed, where the results of the conventional TPC method represent the independent variable while the results of the microplate TPC represent the dependent variable. To compare the conventional and microplate AA a statistical comparison based on the F-test and t-test of three concentrations of the GSE, AE and GTE was performed. To consider both methods statistically equal with a confidence of 95%, the P-value for the F-test and t-test has to be higher than 0.05.19 For the TPC method comparison, five concentrations of GSE (12, 24, 36, 48 and 60 mg L−1 ) and five concentrations of AE (24, 48, 54, 60 and 72 mg L−1 ) were prepared in ethanol–water solution (20:80, v/v). Five dilutions (1:10, 1:5, 1:4, 1:3.3 and 1:2.5) from a 1:20 diluted GTE were also used to compare the TPC methods. To compare the AA methods, three concentrations of GSE (35, 40 and 45 mg L−1 ) and AE (50, 60 and 70 mg L−1 ) were prepared as in TPC assays and three dilutions (1:2.5, 1:2 and 1:1.7) from a 1:50 diluted GTE were used. Precision and accuracy Precision is based on the repeatability and reproducibility, which can be expressed as the relative standard deviation (RSD). Accuracy expresses the closeness of a result to a true value.19,20 In this work, the repeatability was the RSD calculated from five repetitions of the analysis of the same sample on the same day and conditions. The reproducibility was the RSD calculated from five repetitions of the analysis in different days and daily prepared reagents. The accuracy was calculated as the percentage of recovery.9,15
(b)
Statistical analyses The statistical analyses were performed using Statgraphics Centurion XVI (Statpoint Technologies, Virginia, USA) and Minitab 16 (Minitab Inc., State College, PA, USA) software.
(c)
Table 1. Confidence intervals at 95% for the estimated coefficients Confidence levels Extract
Parameter
Estimated coefficient
Grape seed
Slope Constant Slope Constant Slope Constant
0.9982 −1.1455 0.9656 −1.3315 0.9838 −0.2523
Apple Green tea
206
Figure 2. Linear regression for total phenolic content methods comparison. (a) Grape seed extract; (b) apple extract; (c) green tea extract.
wileyonlinelibrary.com/jsfa
© 2014 Society of Chemical Industry
Lower
Upper
0.9551 −2.7246 0.9248 −2.8692 0.9674 −1.0406
1.0412 0.4336 1.0064 0.2061 1.0002 0.5361
J Sci Food Agric 2015; 95: 204–209
Comparison of microplate and conventional methods for Folin- Ciocalteu and DPPH
RESULTS AND DISCUSSION Calibration curves and limits of detection and quantitation The standard calibration curves for the microplate methods are presented in Fig. 1. The curves are linear when the concentration of gallic acid is in the range of 10–200 mg L−1 (R2 = 0.9998) and the concentration of Trolox is in the range of 50–500 μmol L−1 (R2 = 0.9999). When comparing these curves with those obtained by Herald et al.,9 who used similar methodologies and the same concentration ranges, it can be seen that the slight changes in the methodology had more influence upon the slope of the AA assay than on the slope of the TPC assay. The slope of the calibration curve for the AA in this study was 0.1647, while the slope in their study was 0.1512; for the calibration curves of the TPC the slopes were 0.0076 and 0.0073, respectively. However, comparing the slopes of this work with those obtained by Horszwald and Andlauer,13 TPC = 3.1761 and DPPH = 0.0591, it can be seen that greater differences in the TPC and DPPH microplate methods had a significant impact on the sensitivity of the methods. That work had also calculated the limits of detection and/or quantification for their methods. In this work, the LOD and LOQ for the Folin–Ciocalteu microplate method were 0.74 and 2.24 mg L−1 GAE, respectively. In the case of the DPPH microplate method the LOD and LOQ were 12.07 and 36.58 μmol L−1 of Trolox equivalents, respectively. As expected, because of the greater sensitivity of the Horszwald and Andlauer method,13 the LOQ (0.015 mg L−1 GAE) of their TPC microplate method is lower than the LOQ of this work. On the contrary, the greater sensitivity of DPPH microplate method of this work, resulted in a lower LOD than that obtained by Horszwald and Andlauer,13 which was 50 μmol L−1 of Trolox. Statistical comparison Preliminary studies The dilutions to perform the TPC and the AA comparisons were established to obtain 10 to 70 mg L−1 of gallic acid equivalents and μmol L−1 concentrations from 220 to 320 Trolox equivalents,
www.soci.org
respectively. These were the broadest concentration ranges in which the three extracts in the microplate method gave determination coefficients higher than 0.999. The determination coefficients were obtained by a linear regression of the dilutions of each extract and the resulted gallic acid or Trolox equivalents concentrations. The regression residues were tested for a normal distribution and homoscedasticity. The Ryan–Joiner coefficient of correlation (P > 0.05) indicated a normal distribution; the Levene test (P > 0.05) confirmed homoscedasticity; finally the Durbin–Watson test (P > 0.05) showed independence of the residues. Total phenolic content To statistically compare the TPC conventional and microplate methods, the confidence intervals of a linear regression were used. To generate the graphic and the linear regression equation, the abscissa represents the values of the reference method (conventional TPC method) and the ordinate shows the values of the method to evaluate (microplate TPC method). Figure 2 shows the linear regression of the TPC comparison using the GSE, AE and GTE. Table 1 shows the confidence values for the slope and the constant. If the slope confidence intervals do not include a value of 1, it means that there is an error proportional to the concentration of the analyte. If the constant confidence intervals do not include a value of zero, it means that there is a systematic excess or defect error.19 The methods in the studied range of concentrations can be considered equivalent because, as shown in Table 1, the confidence intervals of the slope and the constant meet the above mentioned requirements. The changes made in the TPC method used by Al-Duais et al.14 and Müller et al.17 were made to increase the concentration of the Folin–Ciocalteu reagent compared to the concentration of the sample. This increment intended to bring the proportions of the microplate method closer to those used in the conventional method. Due to the increase in the Folin–Ciocalteu reagent the desired statistical values were obtained to consider both method equivalents. It is important to remark that the proportions of
Table 2. Validation of DPPH conventional and microplate methods by using F of the Fisher test and t of the Student test Fisher test Extract Grape seed
Concentration or dilution
DPPH methoda
Mean ± SD (μmol L−1 Trolox)
35 mg L−1
C M C M C M C M C M C M C M C M C M
222.7 ± 6.9 222.3 ± 2.5 250.5 ± 2.7 251.3 ± 1.8 274.6 ± 8.4 277.4 ± 6.3 229.1 ± 3.0 225.8 ± 3.0 264.9 ± 3.3 265.4 ± 2.3 302.3 ± 3.6 305.6 ± 4.6 253.9 ± 2.9 254.2 ± 7.8 308.1 ± 5.7 304 ± 15 324.9 ± 8.5 317 ± 18
40 mg L−1 45 mg L−1 Apple
50 mg L−1 60 mg L−1 70 mg L−1
Green tea
1:2.5 1:2.0 1:1.7
F
Student test
P-value
t
P-value
7.7484
0.07
0.1159
0.91
2.2287
0.46
−0.5255
0.61
1.7686
0.59
−0.5899
0.57
1.0325
0.98
1.7240
0.12
1.9642
0.53
−0.2590
0.80
0.6138
0.65
−1.2998
0.23
0.1378
0.08
−0.0815
0.94
0.1420
0.08
0.6061
0.56
0.2225
0.17
0.8542
0.42
J Sci Food Agric 2015; 95: 204–209
207
a C, conventional; M, microplate. DPPH, 2,2-Diphenyl-1-picrylhydrazyl.
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
www.soci.org reagents and sample of both methods could not be equal because there was bubbling and precipitation during the whole assay and the measurements were unstable. This phenomenon was also reported by Herald et al.9 when they used the Folin–Ciocalteu reagent undiluted. Antioxidant activity To compare the AA conventional and microplate methodologies three different concentration levels of grape seed, apple and green tea extracts were used. Table 2 shows the resulting P-values of the F parameter of the Fisher test and the t parameter of the Student test. In all cases the P-values for the F parameter of the Fisher test were above 0.05, which indicates that the variances among the measurements for each concentration (n = 5) were not significantly different at a 95% confidence level. The lack of significant differences in the variance indicates that both methods are equally precise and allows the performance of the t parameter of the Student test.19 All the P-values for the t parameter of the Student test were above 0.05. It indicates that there are not significant differences between the mean values for both methods at a 95% confidence level.19 Based on the P-values resulting from both statistical tests, it can be said that the conventional and the microplate methods were equivalent in the concentration ranges studied. The equivalency of both methods to measure the AA of the samples indicates that the increase in the concentration of the sample for the microplate analysis was compensated with the increase in the molar concentration of the DPPH. The higher ratio of the sample and the reagent in the microplate methodology could contribute to reduce the time to reach a stable absorbance compared to the time required with conventional method, which is one of the requirements for a correct AA analysis.11 Finally, the addition of 20% of water to the DPPH reagent in the microplate methodology did not affect the results when compared to the conventional methodology and together with the reduction of time, prevented the evaporation of the solvent leading to more stable measurements.11 Precision and accuracy The Ryan–Joiner and Levene test (P > 0.05) confirmed a normal distribution and homoscedasticity for each day and concentration. Table 3 shows the repeatability, reproducibility and percentage of recovery for five gallic acid and Trolox concentrations. The concentrations were chosen within the ranges used to compare the conventional and microplate methods. It can be seen that the RSDs for the repeatability were ≤3.6% and the RSDs for the reproducibility were ≤6.1% for both methods. RSD values below 10% indicate good precision.9 The repeatability and reproducibility results and the fact that in general, the RSDs of the TPC are higher than those of the AA, are in agreement with the findings by Cheng et al.15 and Herald et al.9 The percentages of recovery (87.8–100.3%) suggest an excellent accuracy for both methods within the studied concentrations.19
CONCLUSIONS
208
In the range of 10–70 mg L−1 GAE and 220–320 μmol L−1 of Trolox equivalents for the Folin–Ciocalteu and the DPPH techniques, respectively, the microplate and the conventional methods are equal at a 95% confidence level. The repeatability, reproducibility and percentage of recovery for the TPC and AA microplate
wileyonlinelibrary.com/jsfa
G Bobo et al.
Table 3. Repeatability, intra-laboratory reproducibility and percentage of recovery for total phenol content and antioxidant activity with the conventional and microplate methods Method
Standard Repeatability Reproducibility
Folin–Ciocalteu microplate
DPPH microplate
Recovery (%)
12
3.6
5.2
87.8
24 36 48 60 224 246 268 293 315
2.2 2.4 2.7 2.1 3.4 2.4 1.0 2.2 2.1
2.9 6.1 5.3 3.9 3.2 2.7 1.9 2.4 3.2
97.9 98.3 99.3 100.3 98.3 99.2 97.8 96.4 95.6
For the Folin–Ciocalteu method, the standard was gallic acid (mg L−1 ). For the DPPH method, the standard was Trolox (μmol L−1 ). Repeatability and reproducibility are given as % relative standard deviation for both methods.
methods showed a precision below 6% and accuracy between 88% and 100%. The use of the microplate methods saves resources and time, which makes it more suitable than the conventional ones.
ACKNOWLEDGEMENTS We thank Exxentia (Spain) for kindly providing the samples. This work was partially financed by the Navarra government (Spain) and by the Public University of Navarre (Spain).
REFERENCES 1 Murcia MA, Vera AM, Martínez-Tomé M and Frega N, Sustancias antioxidantes presentes en los alimentos. Acción, dosis y su eficacia en la promoción de la salud. Nueva Imprenta S.A., Madrid (2003). 2 Hollman PCH, Cassidy A, Comte B, Heinonen M, Richelle M, Richling E, et al, The biological relevance of direct antioxidant effects of polyphenols for cardiovascular health in humans is not established. J Nutr 141:989S–1009S (2011). 3 Sies H, Polyphenols and health: Update and perspectives. Arch Biochem Biophys 501:2–5 (2010). 4 Valenzuela A, Tea consumption and health: Beneficial characteristics and properties of this ancient beverage. Revista Chilena de Nutrición Versión Online. 31:72–82 (2004). 5 Jesionkowska K, Sijtsema SJ, Konopacka D and Symoneaux R, Dried fruit and its functional properties from a consumer’s point of view. J Hort Sci Biotechnol 84:85–88 (2009). 6 EEC, Commission Regulation (EEC) N∘ 2676/90 determining community methods for the analysis of wine. pp. 178–179 (1990). [Online]. Available: http://eur-lex.europa.eu/legal-content/EN/ ALL/?uri=CELEX:31990R2676 [11 March 2013]. 7 Sánchez-Moreno C, Review: Methods used to evaluate the free radical scavenging activity in foods and biological systems. Food Sci Technol Int 8:121–137 (2002). 8 Rivero-Pérez MD, Muñiz P and González-Sanjosé ML, Antioxidant profile of red wines evaluated by total antioxidant capacity, scavenger activity, and biomarkers of oxidative stress methodologies. J Agric Food Chem 55:5476–5483 (2007). 9 Herald TJ, Gadgil P and Tilley M, High-throughput micro-plate assays for screening flavonoid content and DPPH-scavenging activity in sorghum bran and flour. J Sci Food Agric 92:2326–2331 (2012). 10 Zhang Q, Zhang J, Shen J, Silva A, Dennis DA and Barrow CJ, A simple 96-well microplate method for estimation of total polyphenol content in seaweeds. J Appl Phycol 18:445–450 (2006).
© 2014 Society of Chemical Industry
J Sci Food Agric 2015; 95: 204–209
Comparison of microplate and conventional methods for Folin- Ciocalteu and DPPH 11 Fukumoto LR and Mazza G, Assessing antioxidant and prooxidant activities of phenolic compounds. J Agric Food Chem 48:3597–3604 (2000). 12 Schmidt BM, Erdman Jr JW and Lila MA, Effects of food processing on blueberry antiproliferation and antioxidant activity. J Food Sci 70:S389–S394 (2005). 13 Horszwald A and Andlauer W, Characterisation of bioactive compounds in berry juices by traditional photometric and modern microplate methods J Berry Res 1:189–199 (2011). 14 Al-Duais M, Müller L, Böhm V and Jetschke G, Antioxidant capacity and total phenolics of Cyphostemma digitatum before and after processing: Use of different assays. Eur Food Res Technol 228:813–821 (2009). 15 Cheng Z, Moore J and Yu L, High-throughput relative DPPH radical scavenging capacity assay. J Agric Food Chem 54:7429–7436 (2006).
www.soci.org
16 Essa HA, Nadir AES and Hamad KI, Effect of anti-browning and anti-microbial constituents of some spices on the quality and safety of apple slices. Modelling, Measurement and Control C 64:15–33 (2004). 17 Müller L, Gnoyke S, Popken AM and Böhm V, Antioxidant capacity and related parameters of different fruit formulations. LWT – Food. Sci Technol 43:992–999 (2010). 18 ICH, Validation of Analytical Procedures: Text and Methodology, Q2(R1). European Medicines Agency, London, pp. 11–12 (2005). 19 Aguirre OL, García GFJ, García JT, Illera FM, Juncadella RM, Lizondo CM, et al, Validación de métodos analíiticos, in Monografías de AEFI, ed. by Pérez CJA and Pujol FM. Asociación Española de Farmacéuticos de la Industria, Barcelona, pp. 79–83 (2001). 20 EURACHEM, The Fitness for Purpose of Analytical Methods: A Laboratory Guide to Method Validation and Related Topics. Eurachem, Middlesex (1998).
209
J Sci Food Agric 2015; 95: 204–209
© 2014 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
