biology

The biology essay below has been submitted to us by a student in order to help you with your studies.

Determination of antioxidant stability

ABSTRACT

Valantina R, Neelamegham P, Accessing antioxidant stability in heated mixture of oils using a neural network approach, Online J Bioinformatics (11) 134-148, 2009. Artificial Neural Networks using a back propagation algorithm was used to compute the percentage of inhibition concentration and antioxidant activity of palm and rice bran oils heated 5 times to 270º C. Rice bran oil and palm oil were blended and anti-oxidative properties were determined by In Vitro ABTS and DPPH free radical scavenging peroxide ion radical. The radical scavenging activity IC 50 value varied with the concentration of heated mixture of oils. Computation of Inhibition concentration at different concentration of the sample using neural network analysis was performed and correlated with an experimental value for the mixture of vegetable oils. The percentage of computed and measured (with ABTS in-vitro) were correlated for RP1 (r = -0.935; p<0.01), RP2 (r = +0.333; p<0.01, RP3 (r = -0.169; p< 0.001) and for DPPH in-vitro RP1 (r = -0.941; p<0.01), RP2 (r = +0.091; p<0.001, RP3 (r = +0.032; p< 0.01). The oil mixture exhibited antioxidant stability during deep-frying, which could reduce the incidence of malignancy, colon cancer and coronary heart diseases.

Key words: Antioxidant, ABTS, DPPH, Neural network.

INTRODUCTION

Vegetable oils undergo extensive oxidative deterioration during deep fat-frying (see for example, Rubalya and Neelameagam, 2008). Lipid oxidation is one of the most common causes of quality deterioration (see for example, Choe et al., 2005). It lowers the sensory perception, nutritional quality and safety of lipids (see for example, Hyun Jung Kim et al., 2007). The oxidation not only makes the food less acceptable to consumers but also causes great economic losses to the food industry (see for example, Min and Boff, 2002). Hydro- peroxide, which is the major oxidation product, decomposes to produce volatile compounds and oxidised dimers, trimers or polymers (see for example, Narwar, 1996). The volatile compounds are esters, aldehydes, alcohols, ketones, lactones, furans and hydrocarbons, which are responsible for undesired rancid flavours of oils (see for example, David Min and Choe, 2003). A paradox in metabolism is that while the vast majority of complex life requires oxygen for its existence, oxygen is a highly reactive molecule that damages living organisms by producing species (see for example, Rubalya Valantina et al., 2009). Consequently, organisms contain a complex network of antioxidant metabolites and enzymes that work together to prevent oxidative damage to cellular components such as DNA, proteins and lipids (see for example, Davies, 1995). In general; antioxidant systems either prevent these reactive species from being formed, or remove them before they can damage vital components of the cell (see for example, Sies, 1997).

In the snack foods industry, palm oil is one of the oils, which is often regarded as heavy frying oil, where re-using frying oil is normal (see for example, Nallusamy, 2006). Palm oil with its inherent frying properties is used due to its techno-economic advantages over other oils and fats. Past studies have demonstrated the frying performance of palm olein during continuous frying of snack foods (see for example, Ahmad Tarmizi and Razali Ismail, 2008). Palm oil contains saturated fatty acids like palmitic acid 44.3%, oleic acid - 38.7% and linoleic acid - 10.5%, vitamin E especially tocotrienols, Vitamin K and magnesium (see for example, Hui, 1999).The antioxidant activity of the palm oil is due to the presence of carotenoids and Vitamin E. Beta carotene the reason for the yellow color of the palm oil may also be an important factor for the free radical scavenging activity gave a commentary on the antioxidant effect of beta carotene and its role in cardio protection (see for example, Tang, 2002).

Rice is a semiaquatic species that grows in tropical and semitropical climates. Crude rice bran oil contains ~96% of saponifiable fractions and ~ 4% unsaponifiable fractions, which include phytosterols, sterolesters, triterpene alcohols, hydrocarbons, and tocopherols (see for example, Rogers, 1993). The unsaponifiable fractions presence of plant sterols; γ-oryzanol and tocotrienols make the oil hypocholesterolaemic (see for example, Orthoefer, 1996) and bioactive so that it provides positive nutritional and health benefits. Oryzanol content is about 2% in crude rice bran oil. γ -Oryzanol a major component of rice bran oil is used to decrease plasma cholesterol, platelet aggregation, cholesterol absorption from cholesterol-enriched diets and aortic fatty streaks (see for example, Sarmento et al., 2006). Using mid range FTIR spectra the structural changes in the composition of heated rice bran oil is found to be not as much of unheated oil (see for example, Rubalya Valantina and Neelamegam, 2008). The antioxidant stability in the heated and unheated rice bran oil using ABTS and DPPH in-vitro decolorisation assay is found to be in great deal.

There is an increasing interest in antioxidants, particularly in those intended to prevent the presumed deleterious effects of free radicals in the human body and to prevent the deterioration of fats and other constituents of foodstuffs (see for example, Schlesier, 2002). Radical scavenging activities are very important due to the deleterious role of free radicals in foods and biological systems Excessive formation of free radicals accelerates the oxidation of lipids in foods and decreases food quality (see for example, Xu, 2001). The antioxidant in the oils breaks oxidation by adding hydrogen atom to free radicals. In biochemistry, 2, 2'-azino-bis (3-ethylbenzthiazoline-6-sulphonic acid) or ABTS is chemical compound used to observe the reaction kinetics of peroxidase. It can be used to quantify the amount of hydrogen peroxide in a sample. To find the stability of free radical diphenylhydrazyl or DPPH method is used to estimate the activity of antioxidants (see for example, Dunford, 2001). In the present application the parameter IC50 (Inhibition concentration to produce 50% reduction of ABTS and DPPH) is estimated for the mixture of rice bran oil and palm oil.

Artificial Neural Networks (ANN) is biologically inspired network based on the organization of neurons and decision making process in the human brain. In other words, it is the mathematical analogue of human nervous system (see for example, Nascimento, 2000). It can be used for prediction, pattern recognition and pattern classification purpose. It has been proved by several authors that ANN can be of great use when the associated system is complex (see for example, Khalid Omatu, 1992). In this study, the common three layer-feed forward type of artificial neural network is considered to find the IC50 for different percentage of rice bran oil, palm oil and mixture of oil at different concentrations.

MATERIALS AND METHODS

Rice bran oil and palm oil have been collected from a local grocery shop located in Thanjavur district of Tamil Nadu, India to assess the possibility of usage of repeatedly heated oil by common people. These oils are mixed at different ratios like 3:1 (RP1), 1:1 (RP2) and 1:3 (RP3)

Sample preparation:

Hundred milliliter of the sample oil has been placed in a copper beaker and heated on an electric device, stirring manually with glass rod. A microcontroller based temperature controller has been designed and has been used to monitor the sample temperature. To mimic the oil oxidation process during frying, the sample has been heated up to 270°C for five times. Initially, the sample was heated to 270°C for half an hour. Then, it was allowed to cool until room temperature is achieved. Similarly, the sample was subjected to heating up to 270°C for 1 hour, 1 ½ hour, 2 hour and 2 ½ hour respectively ensuring that every time the sample is allowed to cool up to room temperature before heating it next time. In order to ensure that the sample has been heated to the temperature greater than its smoke point, it has been exposed to successive heating.

ABTS+˙ radical decolorisation assay:

ABTS+˙ is generated by mixing 2.5 ml of 7 mM ABTS with 14.7 mM ammonium per sulphate and stored in the dark at room temperature for 16 hours. The solution is diluted with water to achieve an absorbance of 0.7 O.D at 734 nm (see for example, Re, 1999). The peroxide level was determined by the reading the absorbance using UV- Spectrophotometer.

ABTS + H2O2 ABTS+˙ + H20 ------ (1)

The radical-scavenging activity is assessed by mixing 2 ml of this ABTS+˙ solution with different concentrations of sample dissolved in chloroform (25, 50, 75, 100 µl). 1.0 ml of chloroform along with 2.0 ml of ABTS+˙ is used as control. The ABTS*scavenging test is used here to determine the antioxidant activity of both hydrophilic and hydrophobic compounds. The reaction between ABTS*and ammonium per sulfate directly generates the blue green ABTS*chromophore, which can be reduced by an antioxidant, thereby resulting in a loss of absorbance at 734 nm. The final absorbance is measured at 734 nm. The antioxidant capacity is expressed as percentage inhibition, calculated using the following formula,

Inhibition (%) = 100 x (A (cont) -A (Test))/A (Cont) ----------- (2)

Where A (cont) is the absorbance of the control, and A (Test) is the absorbance of the sample at 734 nm. IC50 is the antioxidant concentration that inhibits the ABTS+˙ reaction by 50% under the experimental conditions. This is calculated by Graph pad software version 5.0.

DPPH*radical scavenging assay:

Chloroform solutions of oil at different concentrations (25, 50, 75,100 µg/ml of chloroform) are added to 2 ml of a methanol solution of DPPH• free radical or methanol alone (control) (see for example, Brand Williams, 1995 and Schesier, 2002). The DPPH assay is based on the reduction of alcoholic DPPH solution in the presence of a hydrogen donating antioxidant due to the formation of the non-radical form DPPH-H by the reaction. The antioxidants are able to reduce the stable DPPH to yellow coloured diphenyl- picrylhydrazine. This transformation results in a change in color from purple to yellow, which was measured spectrophotometrically by the disappearance of the purple color at 517 nm.

DPPH+ + AH DPPH + A+ --------- (3)

The reaction mixture is shaken by cyclomixer and then kept in the dark for 30 min under ambient conditions. The absorbance is measured at 517 nm, and the capability of scavenge the DPPH+ radical is expressed as percentage inhibition, calculated using the following formula,

Inhibition (%) = 100 X (A (cont) -A (Test))/A (Cont)) ------------ (4)

Where, A (cont) is the absorbance of the control and A (Test) the absorbance of the sample at 517 nm. IC50 is the antioxidant concentration that inhibits the DPPH* reaction by 50% under the experimental conditions. This is calculated by plotting percentage inhibition against different concentrations of oil. Low IC50 values indicate high radical scavenging activity of cation. All analyses were run in triplicate and averaged.

Neural network approach in computation of Inhibition concentration:

Artificial Neural Networks (ANN) can be used for prediction, pattern recognition and pattern classification purpose. It has been proved by several authors that ANN can be of great use when the associated system is complex. In this study, the common three layer-feed forward type of artificial neural network is considered to find the IC50 for different concentrations mixture of oil. Neural networks are computing tools that consist of large number of simple, highly interconnected processors called neurons. A neuron processes an input vector by applying a transfer function to give an output, which can serve as input to other neurons.

In back propagation algorithm one input layer, five hidden layers and one output layer is used to compute IC50 for different concentrations of the mixture of oil. Neural Network is used to determine the IC50 value using back propagation learning. A Neural Network is trained to concentration as input vector and IC50 as output vector by using the back propagation algorithm method. Under Supervised learning, both inputs and output data are given as data for the training. In this process, the weights are modified and the system is trained so as to get the desired output for a given input. The sigmoid function is implemented for both input and output to train the neural network. Having trained, the neural network is used to process the IC50 value to estimate the formation of peroxides in the oils on heating and scavenging assay at different concentrations .mixture of oil.

Statistical analysis:

All data on total antioxidant activity are the average of triplicate. To examine the effect of type of compound and concentration on antioxidant activity, graph pad software version 5.0 was used (r2 =0. 0.9949, p<0.005, n>9). The data were recorded and analysed by SPSS (version 12).One-way analysis of variance was performed by ANOVA procedures. Significant differences between means were determined by Tukey multiple range tests, p-Values <0.05 were regarded as significant and p-value<0.001 were very significant.

RESULTS AND DISCUSSION

ABTS* radical scavenging activity:

Generation of the ABTS radical cation forms the basis of one of the spectrophotometric methods that have been applied to the measurement of the total antioxidant activity of the solutions of pure substances and aqueous mixtures (see for example, Vertuani el al., 2004). A more appropriate format for the assay is a decolorisation technique in that the radical is generated directly in a stable form prior to reaction with putative antioxidants. The improved technique for the generation of ABTS* described here involves the direct production of the blue/green ABTS* chromophore through the reaction between ABTS and ammonium persulfate. Table 1 shows the significance and variance between the percentages of inhibition at different concentration in the three mixtures of oils using ABTS in-vitro analysis.

The values are mean ±SD; Statistical analysis was done by one-way ANOVA and post-hoc by Tukey multiple comparison tests. The * mark indicates comparison with group I & group II; the # mark indicates comparison with group II & group III; the ^ mark indicates comparison with group III & group I. * P < 0.05; ** P <0.01; *** P <0.001; # P < 0.05; ## P <0.01; ### P <0.001; ^ P < 0.05; ^^ P <0.01; ^^^ P <0.001

DPPH* radical scavenging activity:

In the DPPH assay, the antiradical power of antioxidants by measuring of decrease in the absorbance of DPPH by the colour change purple to yellow. The absorbance decreased when the DPPH*was scavenged by an antioxidant through donation of H atom to form a stable DPPH*(diamagnetic) molecule. Table 2 shows the significance and variance between the percentages of inhibition at different concentration in the three mixtures of oils using DPPH in-vitro analysis.

The percentage of inhibition of the mixture RP2 (r2 =0.976, p<0.005, n>9) as in Fig. 5 increases sharply up to100 µgm/ml and its IC 50 value is 60.65%.

Neural Network:

The percentage of inhibition at different concentration of the three mixtures of oils using ABTS* and DPPH* method is given as input data using the back propagation algorithm method. The input and output vectors which are obtained from the experiments is used for learning. The objective of training is to adjust the weights so that application of a set of inputs produces the desired set of outputs. Before the training process, the weights are initialized to small random numbers. Under Supervised learning, both inputs and output data are given as data for the training. In this process, the weights are modified and the system is trained so as to get the desired output for a given input. The training pattern for the input vectors is concentration of the sample and the output vector is percentage of inhibition. The sigmoid function is implemented for both input and output to train the neural network. Having trained, the neural network is used to process the percentage of inhibition of the mixture of rice bran and palm oil at different ratios at various concentrations.

The input quantity is first normalized to a range of 0.15 to 0.8 and then fed into input layer neurons, which in turn, pass them on to the hidden layer neurons after multiplying by a weight. Hidden layer neurons adds up the weighted input received from the input neuron, associate it with a bias and then passes the result on through a non-linear transfer function. The output neurons do the same operation as that of a hidden neuron. Before the application of the problem, the network is first trained, whereby the difference between the target output and the calculated model output at each output neuron is minimized by adjusting the weights and biases through the training algorithm. The learning of ANNs is accomplished by a back-propagation algorithm where information is processed in the forward direction from the input layer to the hidden layer and then the output layer. The program is trained for the measured percentage of inhibition value to the concentration 25 to100 µgm/ml is given as an input and the network is trained.

The percentage of inhibition at different concentration (10 to 100 µgm /ml) is given as an input and the output percentage of inhibition is studied. Figure 7 shows the comparison of percentage of inhibition computed using BPN and experimental for the sample RP1using ABTS. The IC 50 value of computation and experimental is 57.96 and 62.42%.

Similarly Fig.9 depicts the comparison of percentage of inhibition computed using BPN and experimental for the sample RP3 using ABTS. The IC 50 value of computation and experimental is 34.74 and 36.29%. Table 3 shows the significance and variance between the percentages of inhibition at different concentration (10 to 100 µgm /ml) in the three mixtures of oils using ABTS in-vitro analysis.

The values are mean ±SD; Statistical analysis was done by one-way ANOVA and post-hoc by Tukey multiple comparison tests. The * mark indicates comparison with group I & group II; the # mark indicates comparison with group II & group III; the ^ mark indicates comparison with group III & group I. * P < 0.05; ** P <0.01; *** P <0.001; # P < 0.05; ## P <0.01; ### P <0.001; ^ P < 0.05; ^^ P <0.01; ^^^ P <0.001

The IC 50 value of computation and experimental is 62.48 and 69.89%. Figure 11 illustrates the comparison of percentage of inhibition computed using BPN and experimental for the sample RP2 using DPPH*. The IC 50 value of computation and experimental is 50.79 and 60.49%.

Similarly Figure 12 depicts the comparison of percentage of inhibition computed using BPN and experimental for the sample RP3 using DPPH*. The IC 50 value of computation and experimental is 31.56 and 33.19%

CONCLUSIONS

The objective of the back propagation network was to minimize the error function and to generate an output vector. The results have shown that ABTS and DPPH systems provide information on the reactivity of a test compound with a stable free radical. Bleaching of the reagents colour by the test sample represents the capacity for hydrogen or electron donation by the test compound. The IC 50 value of the mixture of rice bran and palm oil at different ratio increases the antioxidant activity in the highly saturated palm oil. The study also reveals shows that the percentage of conversion of unsaturated fatty acid into saturated fatty acid in the repeatedly heated palm oil can be controlled by the addition of high antioxidant rice bran oil and could be used for frying with less adverse effect.

ACKNOWLEGMENT: The authors are thankful to our Vice chancellor, SASTRA University, Tamilnadu, India for providing the facilities to carry out the work successfully

REFERENCES

Ahmad Tarmizi, A. H. and Razali Ismail., (2008), Comparison of the frying stability of standard Palm Olein and special Quality Palm Olein, J. Am. Oil Chem. Soc, 85, 254-251.

Brand William, W., Curvelier M.E. and Berset C. (1995) Use of a free radical method to evaluate antioxidant activity. Lebensm. Wiss Technol. 28: 25-30.

C.A.O. Nascimento, R. Giudici, and R. Guardani, (2000) Neural Network Based Approach for optimization of Industrial Chemical Processes, Comp. Chem. Eng., 24, 2303

Choe, E., Lee, J., David Min, B. (2005), Chemistry for oxidative stability for edible oils, In: Akohh CC, Lai Om (eds) Healthful lipids. AOCS pess, Champaign, pp 558-590.

David Min, B. and Choe, E. (2003), Effects of singlet oxygen oxidation on the flavor of foods and stability of vitamins. Journal of Food Science Biotech, 582-586.

Davies, K. (1995) Oxidative stress: the paradox of aerobic life. Biochem. Soc. Symp.; 61: 1-31.

Dunford, N. T. (2001) Health Benefits and Processing of Lipid-Based Nutritionals. Food Technol. 55 (11): 38-44.

Hui, Y. H. (1999) Bailey's Industrial oil & Fat Products - Edible oil & Fat Products, (Vol 2), A Wiley- Interscience Publication, New York, pp 603-75.

Hyun Jung Kim, Tae Shik Hahm, and David Min B., (2007), Hydroperoxide as a prooxiidant in the oxidative stability of soybean oil, J. Amer oil Chem Soc, 84, 349 - 355.

Lichenstein, A.H., Ausman, L.M. Carrasco W., Gualtieri L.J., Jenner J.L., Ordovas J.M., Nicolosi R.J., Goldin B.R, and Schaefer E.J, Rice Bran Oil Consumption and Plasma Lipid Levels in Moderately Hypercholesterolemic Humans. Arteriosclerosis and Thrombosis 14(4): 549-556 (1994).

Khalid Omatu, M, S. The Neural Network Controller for a Temperature Control System, IEEE control systems, vol. 12, 1992, pp.58-64.

Min, DB, Boff, J. M. (2002) Chemistry and reaction of singlet oxygen in foods. Comp Rev Food Science Food Safty 1, 58-72.

Nallusamy, S. (2006), The role of palm oil in the snack food industry, Presented at International Palm oil trade fair and seminar, Kuala Lumpur.

Narwar, W. W. (1996) Lipid.In: Fennema OR (ed) Food chemistry, 3rd edn. Marcel Dekker, New York, pp 225-320

Orthoefer, F.T. (1996) Rice Bran Oil: Healthy Lipid Source. Food Technology. 50(12): 62-4.

Re, R., Pellegrini, N., Protegente, A., Pannala, A., Yang, M., Rice Evans C. (1999), Antioxidant activity applying improved ABTS radical cation decolorsation assay. Free Radical. Biol. Medicine. 26, 1231-7.

Rogers, E.J., Rice, S.M., Nicolosi, R.J., Carpenter, D.R., McClelland, C.A, Jr. Romanczyk, L.J. (1993) Identification and Quantitation of γ-Oryzanol Components and Simultaneous Assessment of Tocols in Rice Bran Oil. JAOCS. 70(3): 301-7.

Rubalya Valantina, S., Neelameagam, P., (2008) A Study of rheological behaviour and oxidative stability of rice bran and corn oil using FTIR spectra -International journal of pure and applied physics. Vol (4), 77 - 86.

Rubalya Valantina, S., Neelamegham P. and Gayathri, K. (2009) Estimating antioxidant stability in sunflower and rice bran oil on heating using in-vitro analysis. Biomedicine, 29(1); 31-35.

Sarmento, C. M. P., Ferreira, S. R. S. and Hense, H., (2006) Supercritical fluid extraction of rice bran oil to obtain fractions enriched with tocopherols and tocotrienols, Brazilian J. of Chem. Eng., 23, 243- 249.

Schlesier, K. Harwat, M, Bohm, V, Bitsch, R. (2002) Assessment of antioxidant activity by using different In-vitro methods. Free Radic. Res. 36: 177-87.

Sies, H. (1997) Oxidative stress: oxidants and antioxidants. Experimental Physiology. 82(2): 291-95.

Tang, TS, (2002).Fatty acid composition of edible oils in the Malaysian market, with special reference to trans-fatty acid. Journal of Oil Palm research, (14), 1-8.

Vertuani, S., Angusti, A. and Manfredini, S. (2004) The antioxidants and pro-antioxidants network: an overview. Current Parma Des 14: 1677-94.

Xu, Z., Hua, N. and Godber, J. S. Antioxidant Activity of Tocopherols, Tocotrienols, and γ-Oryzanol Components from Rice Bran against Cholesterol Oxidation Accelerated by 2,2'-Azobis (2-methylpropionamidine) Dihydrocholoride, J. Agric. Food Chem. 49, 2077 (2001).


Request Removal

If you are the original writer of this essay and no longer wish to have the essay published on the UK Essays website then please click on the link below to request removal:

Request the removal of this essay


More from UK Essays