Stabilization of sunflower oil by pennyroyal (Mentha piperita) extracts during accelerated storage

ABSTRACT

The aim of this study was to examine the antioxidant activity of pennyroyal (Mentha piperita) extracts of sunflower oil as a replacement for synthetic antioxidant. This study was performed in two stages. In the current study, it was proven that the examined pennyroyal was in all ways similar to Mentha piperita sequence and its blast performance at the National Center for Biotechnology Information (NCBI). The anti-oxidation activities of Mentha piperita extracts were identified using 2,2-diphenyl-1-picrylhydrazyl and total phenolic content. The protective effect of extracts on the stabilization of sunflower oil was tested and compared with tertiary butyl hydroquinone by measuring peroxide values (y₁), p-anisidine (y₂), and rancimat value (y₃) during storage for 4 weeks at 60°C (Schaal test).The results showed that the antioxidant activity of Mentha piperita extracts (500 ppm) was equal with tertiary butyl hydroquinone effect (100 ppm) at 60°C. In the next step, the sunflower oil containing Mentha piperita extracts (500 ppm) and tertiary butyl hydroquinone (100 ppm) as two blocks were tested to determinate the optimum condition of both high temperature (x₁:180–220°C) and time (x₂: 5–15 min) variables on oxidative stability. The optimum conditions were optimized using response surface methodology. The analysis of variance result showed that temperature and time had significant effects on peroxide, p-anisidine, and rancimat values (p < 0.01). The optimal condition for the examined parameters was related to a sunflower oil treatment containing Mentha piperita extracts (500 ppm) at 180°C and 5 min (with the desirability of 92.1%).

KEYWORDS:

• Mentha piperita
• Sunflower oil
• Antioxidant activity
• Oxidation stability
• Extract

Introduction

In recent years, a great deal of research has been devoted to the replacement of chemical materials with natural ones to eliminate or reduce chemical and synthetic constituents in food. Hence, many efforts have been made to find natural antioxidants from herbal sources. Lipid oxidation during food storage and processing not only causes the loss of nutritional and digestive quality of food, it also releases oxidized products such as free radicals. Free radicals can cause many diseases, especially cancer, in biological systems as well as the reduction of sensory indexes of food.^[1] Antioxidants are constituents that effectively prevent lipids oxidation.^[2,3] Currently, synthetic antioxidants (tertiary butyl hydroquinone [TBHQ]) are used for deferring lipid oxidation in industry, but due to the bad nutritional and carcinogenic effects of these constituents, as well as consumers’ tendency toward using natural compounds, use of natural antioxidants has received far more attention.^[4] One of the best sources of natural antioxidants is the phenolic constituents found in herbal samples.^[5] The Labiatae family includes about 220 different kinds and 3,300 species which are widely used for various purposes.^[6] Plants belonging to Labiatae are rich in polyphenolic constituents, and many of them are known for their antioxidant qualities.^[7] Mentha is a member of this family and a flora plant of Iran which contains six species.^[8] The species of Mentha are generally known as “na’na” and “pooneh” in Iran, and are mainly used as herbal tea, flavoring, and herbs.^[9] Several methods have been introduced for examining lipid oxidation so far, but no test is useful for all the oxidation stages and none of them is useful for fat and food by themselves. At best, a test can control and determine only one or few of physical changes in oxidation. Therefore, it is best to apply a combination of several methods to evaluate oxidation. The aim of this study was to evaluate the antioxidation effect of pennyroyal (Mentha Peperita) extract on sunflower oil (SFO) oxidation at low and high temperatures.

Materials and Methods

Plant Material

Pennyroyal, the plant being used in this experiment, was collected in spring during blossoming. It came from Khoy, West Azerbaijan and the genomic DNA was extracted. The primers were designed by Oligo 7 software to amplify ribulose-1, 5-bisphosphate carboxylase/oxygenase large subunit (rbcL gene). The primer sequence was: MRbsL: 5’-GTG TTA AAG AGT ACA AAT TGAC-3’MRbsLR: 5’-AG CAT TCA AGT AA T GCC CTT TG -3’. PCR reactions were performed in an automatic thermal cycler (Bio-Rad, USA) under the following conditions: Initial denaturation at 94°C for 5 min, 35 cycles of 94°C for 60 s, 53°C for 60 s, and 72°C for 90 s, and final extension at 72°C for 10 min and held at 4°C. Polymerase chain reaction (PCR) production was sequenced at Macrogene, Korea and blasted in the National Center for Biotechnology Information (NCBI) and then the species of Mentha piperita was identified.^[10]

SFO and Chemicals

The decolorized and refined SFO, which was completely free from antioxidant was obtained from the oil extraction factory of Piranshar city. According to the information from the factory, this oil contains 62.3% linoleic acid (18: 2 n-6) and 24.74% oleic acid (18: n-90). All reagents used in this study were analytical grade and were mostly obtained from Merck (Darmstadt, Germany).

Preparing of Methanol Extract

The leaves of Mentha were separated and dried in shadow under natural conditions. The extracts were obtained by maceration of dry plant material (100 g) in 400 mL of methanol for 24 h in a dark place at room temperature. The macerates were filtered under a vacuum through a Buckner funnel with filter paper (Whatman #4), while the methanol extract was dried in a rotary evaporator (IKA RV10DS99). The extract was smoothed by dimethyl sulfoxide (DMSO; 10 mL; 100%) and kept in a glass bottle at 4°C for later examination.^[11]

Measuring Total Phenolic Content of the Extract

The total amount of phenolic constituents in pennyroyal extract was examined through colorimetric method using a Folin–Ciocalteu reagent. Measurements were carried out in triplicate and calculated based on a calibration curve of gallic acid. The contents of total phenols were expressed as µg gallic acid equivalent (GAE).g^–1dry weight (DW).^[12]

Free Radical-Scavenging Activity (DPPH)

The examined extract’s potential to combine with hydrogen or electron donation was tested by its potential for decolorizing 2,2-diphenyl-1-picrylhydrazyl solution (DPPH), a stable free radical that is neutralized in the presence of antioxidants. This solution changes to dark purple in ethanol, but if this radical is neutralized, the color intensity reduces and changes to light yellow. As a result, the reduction of light absorbance will be relative to the antioxidant potential of the sample. For performing the examination, 600 µL of DPPH were dispensed into five test tubes; then 20, 40, 60, 80, and 100 mL of the extract were added to the respective test-tubes, and the volume of all the tubes was increased to six using methanol. The sixth test tube was left as a blank. All samples were read after being kept at room temperature for 15 min at 517 nm.^[11] The scavenging stages were computed by below formula:

Antioxidative Activities of the Extract in SFO Emulsion

The methanol extract of Mentha piperita (100, 500, and 1000 ppm) and TBHQ (100 ppm) were added to the refined SFO as shown in Table 1. Oil samples were placed in oven at 60°C for a given time (28 days; Schaal test). The stability of the oil samples was evaluated by a rancimat test, peroxide value and p-anisidine values at the end of every 7 days. In the second stage, based on the results of Schaal test, the best natural antioxidant was selected from among the other ones and a synthetic antioxidant (TBHQ) with acceptable concentration (100 ppm) was added to the SFO. The effect of the temperature (X₁) and the time (X₂) were also examined. The tests of peroxide, p-anisidine, and rancimat values were measured for high temperature.

Table 1. The mixture used for examining anti-oxidative treatments.

Control(SFO)	SFO100*	SFO500*	SFO1000*	TBHQ (T100)
Sunflower oil (SFO) without antioxidant	SFO + 100 ppm	SFO + 500 ppm	SFO + 1000 ppm	100 ppm

*Amount of Mentha extract.

SFO: Sunflower oil; T: TBHQ.

Oil Stability Analysis

The peroxide value was measured in line with Shantha et al. and the p-anisidine value was determined according to Smouse’s study.^[13,14] The stability of oils and fats against oxidation was determined by the rancimat index. The Metrohm rancimat instrument was used to perform the experiment.^[15]

Statistical Analysis

In this study, in the first stage of research (low temperature), the statistical analysis of data obtained from the Schaal test was performed as a factorial experiment on the basis of a completely randomized design by SPSS version 20. Duncan’s multiple range (p ≤ 0.05) was then used for the mean comparison and all the graphs were drawn by EXCEL2007. In the second stage of research, response surface methodology (RSM) with a central composite design (CCD) was employed to optimize the best combination of antioxidant in two blocks to produce SFO in high temperature. The independent variables were temperature (X₁) and time (X₂).The range and central point value of the variables are shown in Table 2. After the experiment was completed, the model was fitted to the data.

Table 2. Independent variables and their coded and actual values used for optimization.

		Coded levels
Independent variables	Mathematical symbol	–α(axial)	–1	0	+1	+α(axial)
Heating temperature °C	X1	171.72	180	200	220	228.28
Heating time (min.)	X2	2.93	5	10	15	17.07

Design-Expert software (version 7) was used to analyze data. Three-dimensional surface plots of the independent variables were drawn to illustrate the best combination based on the regression analysis.^[16,17]

Results and Discussions

Pennyroyal Sequence and Blast by PCR

The examined Pennyroyal sequence and blast was 100% similar to the Mentha piperita recorded at NCBI in terms of molecular taxonomy with a large subset gene of Rubisco enzyme. The result of PCR is shown in Fig. 1. PCR was performed and after that the quality of extracted DNA was confirmed by primer design for amplifying gene span sequence ITS1-5.8S-ITS2. Finally, after purification, the product of PCR was sent to Macrogene in Korea, to be sequenced. Analysis of the sequence obtained was performed by its comparison with recorded sequences of the given genome span of Mentha at NCBI. Primer was designed based on the gene sequence of a large subset of Rubisco enzyme by Oligo7.^[18]

Figure 1. Sample of 1% agarose electrophoresis of large-subunit genes of enzyme Rubisco.

Phenolic Extraction Constituents and Antioxidant Capacity

Antioxidant activities in plant extracts are related to their phenolic contents, which change depending on genetic and environmental factors, as well as on post-harvest processing conditions, so it is necessary to determine the antioxidant capacity of phenolics. In this study, phenolic contents in the Mentha piperita extracts ranged from 87.127 to 100.01 μg GAE/g and the antioxidant capacity shown by IC50 was 0.303 μg /mL. This result was consistent with the results obtained by Kamkar et al.^[19] Furthermore, the study by Luximun-Ramma et al.^[20] showed a linear correlation between antioxidant capacity and the phenolic contents of extracts of plant, fruits, and beverages. Sugihara et al.^[21] and Spencer^[22] also suggested that flavonoids have the potential to scavenge hydroxyl radicals, superoxide anions, and lipid peroxyl radicals.

Antioxidant Activity in Oil Under Storage Conditions

Qualitative parameters of the oil treated with natural and synthetic antioxidants during 28 days of warming up at 60°C

SFO with more than 86% unsaturated fatty acids was exposed so that it was combined with oxygen. The results of the measured parameters for the stability of the oil at 60°C showed that there was a significant difference between treatments within 4 weeks of incubation (p < 0.01).

Peroxide value

As shown in Fig. 2, during 4 weeks of keeping the oil at 60°C, the highest level of change of peroxide value was related to the control sample, where the rate of increase was slower in the first week but sharpened in the following weeks. In the first week of experiment, among treated samples, the lowest level of peroxide was related to the sample of SFO1000. In the following weeks, the increase in peroxide value (SFO₁₀₀₀) can be related to poeroxidation characteristics being produced in a high concentration of anti-oxidant.^[23,24] Nevertheless, in the following weeks, treatments of T₁₀₀ and SFO500 showed a greater effect on deferring the oxidation rather than other treatments, both of whose influence was almost the same.

Figure 2. Change in peroxide value after 4 weeks of storage at 60°C in treated samples.

In similar studies, the antioxidant effect of hydrous, ethanol, and methanol extracts of Iranian Mentha pulegium was examined on soybean oil in different concentrations (0, 200, 400, and 800 ppm). Results of a Schaal test showed that the peroxide value of the extracts was lower in comparison to the control sample and an extract of Mentha pulegium was comparable to the synthetic antioxidant of BHT extract (0.02 ppm) after 35 days at 65°C.^[25] Moreover, in another study, the protective effects of Eriobotrya japonica (Lindl) fruit peel extracts were tested in stabilization of soybean oil and compared to TBHQ by measuring the peroxide values during storage (65 days at 25°C). Results revealed that the fruit peel extract (400 ppm) showed stronger antioxidant activity, but the highest effect was observed in TBHQ.^[26] It can, therefore, be suggested that the effects of SFO₅₀₀ and TBHQ₁₀₀ in our study were similar to the above studies in the stabilization of antioxidant.

Para-anisidine value

As shown in Fig. 3, the longer the storage time (28 days at 65°C) of the samples in oxidation conditions, the higher the values of p-anisidine in all examined treatments. The highest anisidine value was related to the control sample (oil without antioxidant), and the lowest p-anisidine was related to TBHQ in the weeks that the experiment was conducted.

Figure 3. Changes in para-anisidine value after 4 weeks of storage at 60°C in treated samples.

Rancimat Analysis

As shown in Fig. 4, the rancimat analysis was performed at 110°C and the induction period (h) was examined for SFOs both with and without Mentha extracts (100, 500, and 1000 ppm) and TBHQ (100 ppm). In all examined treatments, the longer the storage time of samples, the lower the rancimat value. A treatment containing synthetic antioxidant had the highest resistance in comparison with other treatments by the end of the second week. The control sample had also lost all its resistance by the end of the second week. The extract of Mentha Peperita (500 ppm) had 3 h of oxidative stability at the end of 4 weeks storage at 60°C. However, the other samples had no resistance at the end of 4 weeks.

Figure 4. Changes in rancimat over 4 weeks of storage in the treatments under study.

As mentioned previously, and in accordance with to various stability indicators, it can be concluded that a the extract of Mentha Peperita (500 ppm) acted equal to synthetic TBHQ (100 ppm) in stabilizing SFO during the storage period, and was even more effective than synthetic antioxidant and other concentrations at the end of the final week. The reason for this can be related to both the high and suitable quantities of existing phenolic constituents in the extract (500 ppm) and thymol and carvacrol as a monoterpenoid phenol.^[27]

Response Surface Optimization of SFO Data Containing SFO500 and TBHQ100

In the present study, the relationship between the responses and the variables were identified by using a CCD in two blocks. Further, the heating conditions of the antioxidant activities were optimized. The data for each block, containing 13 runs, based on the experiment design, are shown in Table 3. The lowest peroxide and anisidine value and the highest rancimal value of each block is highlighted in Table 3. Based on these responses, these data were analyzed and they were optimized to obtain the fitted model.

Table 3. Central composite design with experimental and predicted values for peroxide, anisidine, and racimat value.

			Coded variables
Standard	Run	Block	X1Temperature ºc		Response (Y1)	Response (Y2)	Response (Y3) rancimat
11	1	TBHQ100	200.00	10.00	3.59	34.69	5.39
4	2	TBHQ100	220.00	15.00	4.5	39.54	4.1
5	3	TBHQ100	171.72	10.00	2.36	27.43	5.5
7	4	TBHQ100	200.00	2.93	2.1	29	5.1
2	5	TBHQ100	220.00	5.00	3.9	34.65	4.8
3	6	TBHQ100	180.00	15.00	2.82	25.44	4.82
6	7	TBHQ100	228.28	10.00	4.68	39.59	5
12	8	TBHQ100	200.00	10.00	3.34	31.65	5.2
10	9	TBHQ100	200.00	10.00	3	33.55	5
13	10	TBHQ100	200.00	10.00	3.2	33.69	5.4
8	11	TBHQ100	200.00	17.07	4.64	39.87	4.95
9	12	TBHQ100	200.00	10.00	3.59	34.76	5.34
1	13	TBHQ100	180.00	5.00	2.44	26.66	6.11
5	1	SFO500	171.72	10.00	2.36	25.76	5.35
8	2	SFO500	200.00	17.07	4.2	36.42	5.4
1	3	SFO500	180.00	5.00	3.2	23.98	7.6
13	4	SFO500	200.00	10.00	3.4	28	6.45
11	5	SFO500	200.00	10.00	3.3	28	6.43
9	6	SFO500	200.00	10.00	3.51	28	6.72
2	7	SFO500	220.00	5.00	3.8	32.75	5.2
3	8	SFO500	180.00	15.00	2.77	22.76	5.54
4	9	SFO500	220.00	15.00	4.44	33.56	5.32
6	10	SFO500	228.28	10.00	4.2	33.92	5
10	11	SFO500	200.00	10.00	3.92	33.66	4.34
7	12	SFO500	200.00	2.93	2.2	26.57	7
12	13	SFO500	200.00	10.00	3.36	28	6.52

Models

In Table 4, a linear model was fitted for any responses. The results of analysis of variance (ANOVA) showed that the model was significant for all variables except for ranciment (SFO₅₀₀). The significant contribution of each coefficient was determined by p-value of F-test (p < 0.05). The fitness of the model was generally assessed based on lack-of-fit test (p > 0.05). The quality of the fit model can be determined based on the coefficient of determination (R²).

Table 4. Results of ANOVA and sunflower oil modeling containing antioxidant in peroxide, anisidine, and rancimat value.

R^2	Model	Test	Antioxidant
85% – 80%	Y = +1.93559 + 3.70096E-003X1 – 0.45904X2 Non significantY = –2.78056 + 0.14369X1	Peroxide Rancimat p-Anisidine	SFO 500
87%75%85%	Y = –4.67232 + 0.034631X1 + 4.30256E-003X2 + 5.50000E-004 X1X2 Y = +12.05345 – 0.031857X1 – 0.35005X2 Y = +9.79775 + 0.092793X1 – 2.57894X2	Peroxide Rancimat p-Anisidine	TBHQ100

Results of the optimization of SFO containing two natural and synthetic antioxidants being computed by Expert Design software are shown in Tables 5 and 6. Therefore, because of the lowest values of peroxide and anisidine and the highest values of rancimat, the optimized sample was related to a Mentha piperita extract sample (500 ppm) at 180°C for a duration of 5 min and desirability of 92.1%. The 3D response surface as a function of time and temperature is shown in plot (Figs. 5a, 5b, and 5c).

Table 5. Optimization results of TBHQ₁₀₀.

Number	Temperature °C	Time (min)	Rancimat	Pv	Anisidine	Desirability
1	180	5	5.7424	2.7108	25.82	91.1%

Table 6. Optimization Results of SFO-500.

Number	Temperature °C	Time (min)	Rancimat	Pv	Anisidine	Desirability
1	180	5	7.3727	2.01	23.758	92.1%

Figure 5a. Response surface plotting of the effect of temperature and time on change of para-anasidine during frying after adding antioxidant (SFO 500 ppm).

Figure 5b. Response surface plotting of the effect of temperature and time on change of peroxide value during frying after adding antioxidant (SFO 500 ppm).

Figure 5c. Response surface plotting of the effect of temperature and time on change of Rancimat during frying after adding antioxidant (SFO 500 ppm).

Conclusion

The results of peroxide, p-anisidine and rancimat value in SFO by adding the natural antioxidant of Mentha piperita extract and synthetic antioxidant of TBHQ showed that natural antioxidant of Mentha piperita has a significant effect on the stability of SFO. Therefore, Mentha piperita is a natural antioxidant that has a great potential for replacing synthetic antioxidants. The extract could be introduced as a natural antioxidant in food following the successful conclusion of further tests.

Funding

The current article is extracted from a research project funded and supported by the Islamic Azad University of Tabriz and the support of that organization is greatly appreciated.

Additional information

Funding

The current article is extracted from a research project funded and supported by the Islamic Azad University of Tabriz and the support of that organization is greatly appreciated.