Influence of high-dose gamma radiation and particle size on antioxidant properties of Maize ( Zea mays L . ) flour

Influence of high-dose gamma radiation and particle size on antioxidant properties of maize (Zea mays L.) flour was studied using response surface methodology. A central composite design based on three levels of each of particle size, in terms of mesh number (40, 60 and 80 meshes), and gamma radiation dose (25, 50 and 75 kGy) was constructed. A statistically significant dose-dependent decrease (p<0.05) in antioxidant properties of gamma irradiated flour was observed. However, an increase in the mesh number (decrease in particle size of flour) resulted in an increase in antioxidant properties. The optimum level of radiation dose to achieve maximum value of responses was found to be 50 kGy for Trolox equivalent total antioxidant activity (TETAOA), 25 kGy for iron chelating ability (ICA), 25 kGy for reducing power (RP) and 75 kGy for linoleic acid reduction capacity (LARC). However, the optimum level of mesh number to achieve desired levels of TETAOA, ICA, RP and LARC was found to be 80 meshes.


INTRODUCTION
Antioxidants are a unique class of substances that are able to protect the biomolecules from oxidative damage caused by endogenous free radicals such as superoxide and hydroxyl radicals produced during various metabolic processes (Nimse, Pal, 2015;Lobo et al., 2010;Sisein, 2014).These free radicals may lead to protein denaturation, lipid per-oxidation, DNA lesions and finally diseased conditions if not captured effectively (Kabel, 2014;Valko et al., 2007).The antioxidants perform their action by trapping the free radicals, reducing the metal ions and preventing the initiation of free radical chain, hydrogen abstraction and peroxide decomposition (Nimse, Pal, 2015).A large number of natural compounds such as ascorbic acid, tocopherols, phenolic acids, flavonoids, anthocyanins, proanthocyanidins and other phytochemical compounds present in food materials have been reported to possess antioxidant properties due to the presence of hydrogen donating groups in their chemical structures (Goufo, Trindade, 2014;Shahidi, 2000).Foods rich in antioxidant compounds have been proved to be effective in decreasing the risk of cardiovascular mortality, destruction of cancer cells and preventing the oxidative lung damage (Miranda-Vilela et al., 2011;Khan et al., 2014).
The processing and preservation techniques significantly influence the nutritional and antioxidant properties of food materials.Gamma-irradiation is an extensively used method in industries for the sterilization of food materials and pharmaceutical products.It is also being used in food science technology to improve the nutritional and functional quality of various food materials (Abdelwhab, Nour, Fageer, 2009;Hassan et al., 2009;Gani et al., 2013).Studies have also shown that gamma irradiations significantly affect the antioxidant composition of plant and other food materials (Kortei et al., 2014;Hussein et al., 2011;Harrison, Were, 2007).
Maize, botanically known as Zea mays L., is the third most important cereal crop after wheat and rice in Pakistan.It is widely used as a source of starch, protein, vitamins and minerals in various food supplementations all over the world.It possesses great medicinal importance due to the presence of considerable amounts of naturally occurring antioxidant compounds including ß-carotene, tocopherols, phenolic acids, flavonoids and anthocyanins (Mboya et al., 2011;Nawaz, Shad, Batool, 2013).Previously, a few data have been reported regarding the gamma radiation induced variations in starch structure and nutritional quality of maize grains (Roushdi et al., 1981;Roushdi et al., 1983).However, no investigations have been reported on the effect of particle size and gamma irradiation on the antioxidant profile of maize flour.Therefore, the present study was designed to optimize the effect of particle size and gamma radiations on phytochemical composition and antioxidant activity of maize flour of different particle sizes using response surface methodology.The data would provide useful information regarding the processing and utilization of maize flour in pharmaceutical and food formulations.

MATERIAL AND METHODS
The maize grains were collected from Maize and Millet Research Institute (MMRI), Yousafwala, Sahiwal, Pakistan.The mature grains were separated manually, dried in air under shade to remove moisture, and ground to fine flour.The grinding was performed discontinuously using a low speed (1000 rpm) electric grinder in order to maintain the temperature at 35±5 °C.The flour was packed in airtight glass bottles and stored in dark at standard laboratory conditions until further processing.

Experimental Design
The cumulative effect of particle size and gamma radiations on antioxidant activity of maize flour of different particle sizes was investigated using response surface methodology (RSM).A bi-factorial facecentered central composite design (CCD) was employed to optimize the effect of two independent variables on antioxidant properties of maize flour.The selected levels of input variables are as follows: X 1 : Particle size of flour in terms of sieve mesh number.Three levels of particle size in terms of sieve mesh number were selected as 40, 60 and 80 meshes.X 2 : Radiation dose Three levels of radiation dose were selected as 25, 50 and 75 kGy.
The codes for the selected levels of the two input variables were calculated using following generalized equations: where X i is the coded value of an input variable, ξ i is the specific location of independent variable, ξ i ¯ is the center point and is the scale factor i.e. the difference between ξ i and ξ i ¯.The specific codes for the selected input variables were calculated by putting the values of ξ i , ξ i ¯ and S i in the generalized equation as: The calculated codes and combination of coded and actual levels of input variables as per chosen by CCD are shown in Table I.
The optimum point of response variables was searched by performing sequential experimentation.A response surface polynomial quadratic model was developed to find the levels of input variables in a region of optimal response.The developed model determines the relationship of antioxidant properties of maize flour against the particle size and the dose of gamma radiations.The study was done in phases based on CCD which consists of 13 points with nf = 4, factorial points, na = 4 axial point and nc = 5 center points.

Sieving
The flour was sieved successively through micro screens of mesh No. 40, 60 and 80 meshes to obtain required levels of particle size.The distribution of particle size was done on the basis of sieve mesh number.Particle size is inversely proportional to sieve mesh number.A gradual increase in mesh number is associated with a respective decrease in particle size.The range of particle size obtained from selected sieves was as follows (Sigma-Aldrich, 2011):

Gamma irradiation of flour
The maize flour was subjected to gamma irradiation in transparent glass bottles at selected levels of radiation dose at a dose rate of 0.26 kGy/h, sample to source distance 1.5 m and average temperature 30 °C using 60 Co (32000 Curies) gamma radiation source at Pakistan Radiation Services (PARAS), Lahore, Pakistan.Cericcereus dosimeters were used to measure the absorbed dose.The gamma irradiated samples and non-irradiated flour at each level of particle size (taken as control) were stored at 25±5 °C in a sterile laboratory environment to minimize the chances of microbial contamination and growth throughout the study period.The exposure of the samples to direct sunlight was prevented throughout the study period in order to minimize the chances of photooxidation of antioxidant compounds.

Preparation of Extracts
The seed flour was soaked in 75% methanol at 1:10 solid to the solvent ratio for 24 h at 25±5 °C with occasional shaking.The contents were filtered through Whatman filter paper (40), the volume of the filtrate was made up to 100 mL with 75% methanol and used for analysis.

Total antioxidant activity (TAOA) by phosphomolybdenum assay
Trolox equivalent total antioxidant activity of extracts was evaluated by Phospho-molybdenum assay described earlier (Prieto, Pineda, Aguilar, 1999).The methanolic extract (0.1 mL) was mixed with reagent solution (600 mM sulphuric acid, 4mM ammonium molybdate, 0.05M sodium phosphate buffer).The contents were incubated at 95 °C for 90 min, cooled to room temperature and absorbance was recorded at 695 nm against blank (without extract).Total antioxidant capacity was calculated as Trolox equivalent g/100g dry weight using regression equation obtained from the standard curve (R 2 = 0.9897).

Reducing power (RP)
The ferric reducing antioxidant power of the samples was estimated according to the method reported earlier (Oyaizu, 1986).The methanolic extract (2.5 mL) was mixed with phosphate buffer solution of pH 6.6 (2.5 mL) and 1% potassium ferricyanide solution (2.5 mL).The mixture was incubated at 50 °C for 20 min followed by the addition of 10% trichloroacetic acid solution (2.5 mL).The mixture was centrifuged at 2000 rpm for 10 min.The supernatant (5 mL) was mixed with distilled water (5 mL) and 0.1% ferric chloride solution (1 mL) and absorbance was recorded at 700 nm.The reducing power was expressed in terms of decrease in absorbance of the reaction mixture.

Linoleic acid reduction capacity (LARC)
The linoleic acid reduction capacity of methanolic extracts was determined by using the method described earlier (Osawa, Namiki, 1981).The methanolic extract (2 mL) was mixed with 2.5% ethanolic solution of linoleic acid (2 mL), 0.05 M sodium phosphate buffer of pH 7.0 (2 mL) and distilled water (2 mL).The reaction mixture was incubated at 40 °C for 24 h.An aliquot (0.1 mL) from above mixture was mixed with 75% methanol (9.7 mL), 20mM FeCl 2 solution (0.1 mL) and 30% ammonium thiocyanate solution (0.1 mL) and allowed to stand for 3 min.The absorbance was recorded at 500 nm and LARC was calculated in terms of percent inhibition of lipid peroxidation in linoleic acid emulsion by following equation: where A c is the absorbance of control (solution without extracts or standards) and A s is the absorbance in the presence of sample or standard.

Iron chelating Ability (ICA)
Iron chelating ability of extracts was estimated by reported method (Puntel, Nogueira, Rocha, 2005) with some modifications.Methanolic extract (1 mL) was mixed with 0. 1M hydrochloric acid solution (1.6 mL), saturated sodium chloride solution (2.18 mL) and 2 mM ferrous sulfate solution (1.5 mL) successively.The mixture was allowed to stand at 25±2 °C for 5 min followed by the addition of 0.25% 1,10-phenanthroline solution (1.3 mL).The absorbance of the reaction mixture was recorded at 510 nm after 5 min.The iron chelating ability was calculated in terms of their EC 50 (effective concentration required for 50% chelation) values for iron using regression equation obtained from the linear curve (R 2 = 0.9734) of series of different dilutions of methanolic extracts.

Statistical Analysis
The results were expressed as means of three parallel replicates.The prediction of optimum levels of response variable as a function of input variables was achieved by creating polynomial quadratic response surface models.The generalized polynomial model for predicting the variation in response variables is given below: where Y i is the predicted response, β o is a constant, β 1 and β 2 are the regression coefficients for the main variable effects, β 11 and β 22 are quadratic effects and β 12 is interaction effect of independent variables.
The significance of estimated regression coefficient for each response was assessed by lack of fit test (F-value) at a probability (p) of 0.05.The coefficient of determination (R 2 ) and the adjusted coefficient of determination (R 2 adj ) were also determined to check the adequacy of the response surface models and to measure the fairness of fit of regression equation respectively.The precision and reliability of experiments were checked by determining the coefficient of variation (CV).A low value of CV suggests a better precision and reliability of the experiments.A ratio greater than 4 indicates an adequate signal.The statistical software, Design Expert 9.0 (Stat-Ease, Inc.) was used for the development of experimental design, data analysis and optimization procedure.
For graphical optimization of particle size and dose of gamma radiation, the three-dimensional plots were constructed between response and independent variables.The adequacy of the response surface models was verified by plotting the experimental values versus those predicted by the final reduced models.The optimum levels of input variables at which the desired goals of responses may be achieved were found by numerical optimization of data at maximum desirability.

RESULTS AND DISCUSSION
Maize flour is a source of carbohydrates and protein.
It also contains some nonnutritional components such as phenolic compounds, flavonoids, tannins, anthocyanins and cardiac glycosides generally known as phytochemicals (Nawaz, Shad, Batool, 2013).These phytochemicals have been reported to possess antioxidant properties.Antioxidants are the natural or synthetic compounds which are generally known to prevent the oxidation of endogenous and exogenous biomolecules due to their hydrogen donating ability.The antioxidants show their action by trapping endogenous free radicals produced as a result of different metabolic processes and protect the lipids, proteins and nucleic acids from the oxidative damage (Kabel, 2014).
The experimental values of antioxidant activities of non-irradiated and gamma irradiated flour at different levels of particle size as per chosen by the experimental design are presented in Table I.Trolox equivalent TAOA (g/100g dw), RP (absorbance at 700 nm) and LARC (%) ranged from 1.50 to 3.48, 0.215 to 0.354 and 25.75 to 68.88 respectively.Iron chelating ability was determined in terms of IC 50 which ranged from 0.304 to 0.350 mg/ mL of extract.Relatively low value of IC 50 shows good chelating ability.Statistically significant difference (p<0.05) in antioxidant activities was observed among non-irradiated maize flours of different particle sizes.

Response surface analysis and optimization of results
The cumulative effect of particle size and gamma radiation on antioxidant activities of maize flour was optimized by response surface methodology.The prediction of an optimum level of each of the independent variables was carried out by using central composite design (CCD).
The following polynomial regression equations were yielded by RSM in order to show an empirical relationship between the process variables and antioxidant activities of maize flour: These equations include the coefficient for intercept, main (linear) effects, interaction terms and quadratic effects.The influence of each factor on the response is shown by the sign and magnitude of the main effect.The main, quadratic and interaction effects of particle size and radiation dose on TETAOA, ICA, RP and LARC of maize flour as obtained by analysis of variance (ANOVA) are given in Table IIa, IIb, IIc and IId respectively.The significance and adequacy of the response surface model were measured in terms of F-value (lack of fit) and p-value (probability) at 5% significance level (p≤ 0.05).The F-value is a measure of the failure of a model to fit the data in experimental domain particularly for reduced points in a randomized experiment.The corresponding variables with relatively larger F-values (F>3.69) and smaller p-values (p<0.05) were considered more significant.The measurement of F-value and p-values indicated the significant linear negative effect of both variables on each of the studied parameters except TETAOA.Interaction effects were found to be nonsignificant on each of the studied parameters.The quadratic effect of particle size was found to be significant on each parameter while that of radiation dose was found to be significant on TETAOA and ICA.It is clear from RSM results that antioxidant activity of maize flour is significantly decreased in response to a decrease in mesh number (increase in particle size) and increase in gamma radiation doses.
The correlation coefficient (R 2 ) measures the variability of the model in the observed response values.A value of R 2 closer to unity gives a better prediction of the response and high significance of the model.The values of R 2 (0.8819-0.9653) indicated that 88-96% of the variability in antioxidant properties of maize flour could be explained by the suggested model.The values of adjusted R 2 (0.7975-0.9406) for these responses also advocate the significance of the model.Relatively low values (4.45-11.34) of the coefficient of variation (CV) and high value of adequate precision (9.937-20.74)suggest a better precision and reliability of the experiments.
Three-dimensional (3D) response surface plots were drawn to show the main and interaction effects of particle size and gamma radiation dose on antioxidant activity of maize flour (Figure 1).To test the applicability of the model, the predicted values of antioxidant activities were calculated from the polynomial regression equations and plotted against the experimental values (Figure 2A-D).A good agreement between the experimental and predicted values of responses was observed with high values of coefficients of determination (R 2 =0.8819-0.9653).The higher values of R 2 prove the applicability of proposed model with greater accuracy to study the effect of particle size and radiation dose on the antioxidant properties of maize flour.
The selection criteria and results for numerical optimization of particle size and radiation dose to achieve the desirable values of antioxidant ability are presented in Table III.The optimum level of particle size in terms of sieve mesh No. was found to be 70.622 for TAOA, 80 for RP and LARC and 57 for ICA.The optimum level of radiation dose at which antioxidant properties were affected to a minimum value was found to be 25 kGy for each parameter.

CONCLUSION
The present study shows an inverse correlation between particle size and gamma radiation dose and antioxidant properties.It is concluded that smaller the  particle size of flour and lower the radiation dose, the lower would be the loss of antioxidant activity.The low values of antioxidant properties of flour with large particle size may be attributed to poor extraction of bound antioxidant compounds due to small surface area.The decrease in antioxidant properties of flour treated at high radiation dose may be due to gamma rays induced oxidative degradation of antioxidant compounds.The present results suggest that sieving of flour through a sieve with higher mesh number and treatment with high-dose gamma radiation decrease    the extraction efficiency and bioavailability of antioxidant compounds from maize flour.

FIGURE 1 -
FIGURE 1 -3D response surface plots of antioxidant properties of maize flour at various levels of particle size and radiation dose.

FIGURE
FIGURE 2A-D -Agreement between experimental values and predicted values of antioxidant properties of maize flour.A: Total antioxidant activity, B: Iron chelation activity, C: Reducing power, D: Linoleic acid reduction capacity.

TABLE I -
The experimental values of antioxidant activity of maize flour at random levels of experimental conditions as per chosen by central composite design * Center points, ** dw: dry weight

TABLE IIA -
Analysis of variance (ANOVA) from the final reduced models for Trolox equivalent total antioxidant activity of gamma irradiated maize flour ** p ≤ 0.05 indicates significant variation at 95% confidence level.

TABLE IIB -
Analysis of variance (ANOVA) from the final reduced models for for iron chelating activity of gamma irradiated maize flour

TABLE IIC -
Analysis of variance (ANOVA) from the final reduced models for reducing power of gamma irradiated maize flour ** p ≤ 0.05 indicates significant variation at 95% confidence level.

TABLE IID -
Analysis of variance (ANOVA) from the final reduced models for linoleic acid reduction capacity of irradiated maize flour * p ≤ 0.05 indicates significant variation at 95% confidence level. *

TABLE III -
Optimum levels of input variables to achieve the desired goals of response variables with maximum desirability