Food Chemistry 150 (2014) 280–286
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Analytical Methods
Activity of flavonoids and b-carotene during the auto-oxidative deterioration of model food oil-in water emulsions Sotirios Kiokias, Theodoros Varzakas ⇑ Technological Educational Institute of Peloponnese, Dept. Food Technology, School of Agricultural Technology, Food Technology and Nutrition, Antikalamos 24100, Kalamata, Greece
a r t i c l e
i n f o
Article history: Received 20 March 2013 Received in revised form 4 October 2013 Accepted 24 October 2013 Available online 1 November 2013 Keywords: Antioxidants Flavonoids b-Carotene Auto-oxidative deterioration Oil-in water emulsions
a b s t r a c t The antioxidant effects of flavonoids and b-carotene during the thermal auto-oxidation of food relevant oil-in-water emulsions were spectrophotometrically assessed by measuring the formation of primary oxidation products (conjugated dienes and lipid hydroperoxides). An oxidatively ‘‘sensitive’’ model emulsion was selected as substrate of this study in terms of processing and compositional factors. At a concentration of 1.5 mmol kgr1, only quercetin among the tested compounds significantly reduced the oxidative deterioration of cottonseed oil-in-water emulsions. Structural characteristics (positioning of hydroxyl group) or partitioning behaviour between the emulsion phases may modulate the flavonoid activity. The high oxygen pressure conditions of the experimental system may explain the lack of any antioxidant activity for b-carotene. The antioxidant potential of quercetin increased with its concentration until a specific level. On the contrary, the antioxidant concentration within the same tested range (0.75–3 mmol kgr1) did not impact the activity of catechin and b-carotene. Mixtures of b-carotene with flavonoids did not exert a tendency for increasing the activity of each individual compound. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Lipid oxidation is a major concern for the food industry because it produces rancid odours and flavours, decreases the shelf life, alters texture and colour and decreases the nutritional values of lipid based consumer products (Alamed, Chaiyasit, McClements, & Decker, 2009). Many parameters (such as temperature, oxygen pressure, metal catalysts) affect the lipid oxidation which can be delayed or inhibited in presence of antioxidant compounds (Beker, Bakir, Sonmezoglu, Imer, & Apak, 2011). A wide literature is available about bulk oil oxidation, whilst very little is known about oil oxidative stability when it is present as droplets dispersed in a complex aqueous media. Overall, food emulsions offer good examples of food products that can rapidly degrade by lipid oxidation reactions. An increasing body of research evidence nowadays focuses on oil-in-water emulsions as they form the basis of many innovative food products and their properties define the quality of the final product to a great extent. (Nikovska, 2010). Therefore, a better understanding of the endogenous and exogenous factors which regulate the oxidative deterioration of o/w food ⇑ Corresponding author. Address: Dept. of Food Technology, Higher Institute of Technology, Kalamata, Antikalamos, 24100 Kalamata, Hellas, Greece. Tel.: +30 2721045279; fax: +30 2721045234. E-mail addresses: [emailprotected] (S. Kiokias), [emailprotected] (T. Varzakas). 0308-8146/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodchem.2013.10.112
emulsions would elucidate their lipid oxidation mechanisms during the formulation, production and storage of relevant products such as dressings, mayonnaise etc. (Kiokias & Oreopoulou, 2006). The replacement of synthetic antioxidants by ‘‘safer natural mixtures’’ is being increasingly advocated nowadays by food industry. This trend has been imposed by the worldwide preference of consumers for the use of natural antioxidants, some of which may exist inherently in foods or be added intentionally during their processing (Kiokias, Dimakou, & Oreopoulou, 2009). The antioxidant potential of certain carotenoids and flavonoids has been summarised by a few review papers (Kiokias & Gordon, 2004; Kiokias, Varzakas, & Oreopoulou, 2008). So far, a limited amount of research evidence has been reported in the literature concerning the antioxidant activity of both categories of compounds in multicomponent systems. This paper focuses on several compounds of natural origin, such flavonoids (catechin, and quercetin and b-carotene) which were tested in cottonseed oil-in-water emulsions in order to monitor their thermal auto-oxidative destabilisation. Although these compounds have been commonly investigated in bulk oils or in vivo, their mode of activity is not completely clear yet in dispersed systems (Heinonen, Haila, Lampi, & Piironen, 1997; Mattia, Sacchetti, Mastrocola, & Pittia, 2009). In addition, the effect of endogenous compositional and formulation parameters on the oxidation of emulsion droplets was also investigated in order to optimise an
S. Kiokias, T. Varzakas / Food Chemistry 150 (2014) 280–286
experimental model which was subsequently used to evaluate antioxidant capacities of the tested compounds. 2. Experimental 2.1. Materials The flavonoids and all-trans synthetic b-carotene (purified 95%), standards were supplied by Fluca. Refined cottonseed oil and other vegetable oils were commercially available. Protein preparations were: whey protein (Nutrilac QU7560, Arla Foods, powder containing 75% protein) and Na-caseinate (91.5% protein, Campina DMV). Other chemicals such Tween 20 and various reagents used were of analytical grade. 2.2. Methods 2.2.1. Preparation of the emulsions by use of ultrasonic agitation Each oil-in-water emulsion (10–40% o/w) was prepared by mixing for 1 min in a blender (Waring Commercial, USA): the appropriate mass of tested edible oil (mainly cottonseed- and for some experiments corn-, olive kernel-oil) with distiled water containing the emulsifier Tween 20 (or in certain experiments, whey protein, or sodium caseinaate preparations). The tested compounds were added in the oil phase, after having being dissolved at the desired concentration in hexane (which was then removed under nitrogen). The pre-emulsion was placed on an ice bath and after its sonication (Sonics, Vitracell) for 30 min. 15 ml of each emulsion sample was transferred in 20 ml plastic vials. The vials were wrapped with alimunium foil (to avoid photoxidation) but their lids were only loosely screwed so that air could pass in and out of the headspace above the samples. The aerial autoxidation of the emulsion samples was carried out in a shaking bath at 60 °C. 2.2.2. Measurement of conjugated diene hydroperoxides (CD) A modification of the method described by IUPAC 2.505 has been used for the determination of conjugated diene hydroperoxides. More specifically, the emulsion sample (20 ll) was added to a mixture of 10 mL isooctane/2-propanol (2:1 v/v) and votrexed (1 min). The absorbance was measured at 232 nm using a UV– VIS scanning spectrophotometer (Unicam Helios, Spectronc Unicam EMEA, Cambridge, United Kingdom). The amount of CD in the oxidising emulsions was calculated by monitoring absorbance at 232 nm and using the relative molecular mass (280 g mol1) and the molar absorptivity of linoleic acid (e = 26,000). CD are formed during the propagation stage of oxidation, through reactions of the free radicals that have been generated at the initiation stage with di-unsaturated fatty acids. Common vegetable oils (e.g. sunflower oil, cottonseed oil) are rich in di-unsaturated fatty acids, e.g. linoleic acid while free radicals are regenerated through the reaction. A linear increase of CD values has been observed by researchers in accelerated oxidation of emulsions (Lethuaut, Metro, & Genot, 2002). Therefore, measurement of CD has been increasingly advocated by recent research as a reliable indicator of the oxidative destabilisation in lipid based systems and thereby of the qualitative deterioration of the final product. 2.2.3. Determination of lipid hydroperoxides (ferric thiocyanate method) The ferric thiocyanate method, following its development for specific application in emulsion systems (Kiokias, 2002), was used as an alternative technique for the determination of primary oxidation products. Lipid hydroperoxides (LH) were measured by mixing
281
0.3 mL of emulsion with 1.5 ml of isooctane/2-propanol (3:1 v/v), by vortexing (10 s, three times), and isolation of the organic solvent phase by centrifugation at 1000g for 2 min. The organic solvent phase (200 ll) was added to 2.8 ml of methanol/1-butanol (2:1 v/ v), followed by the addition of 15 ll of 3.97 M ammonium thiocyanate and 15 ll of ferrous iron solution (prepared by adding equal amounts of 0.132 M BaCl2 and 0.144 M FeSO4). After 20 min, the absorbance was measured at 510 nm using a UV–Vis scanning spectrophotometer (Unicam Helios, Spectronc Unicam EMEA, Cambridge, United Kingdom). LH concentrations were calculated, using a standard curve made from cumene hydroperoxides (IUPAC, 1997). The ferric thiocyanate method evaluates the measurement of lipid hydroperoxides which are the main primary products of the oxidative process and linked to the quality deterioration of the lipid based products. This was preferred over the iodometric peroxide value determination, for its direct application in emulsion systems, (Kiokias & Gordon, 2003), whereas it is also much simpler and faster and requires a smaller sample size. 2.2.4. Microstructural evaluation Restricted diffusion-based droplet size measurements were obtained by means of pfg-NMR using a Minispec MQ20 (Bruker) (Kiokias, Reszka, & Bot, 2004). Obtained values of the volume weighted geometric mean diameter d3,3 and the width r of the droplet size distribution are converted to the surface weighted mean diameter d3,2 using the relation d3,2 = d3,3exp(r2/2). Emulsion images were obtained by confocal scanning laser microscopy (CSLM). 2.3. Statistical analysis Oxidation experiments were carried out in triplicates and during analysis each measurement was repeated three times. Results were averaged (n = 9) and statistically analysed with one-way ANOVA test (p < 0.001) by use of a Statgraphics Plus 6.0 statistical programme. Differences between oxidative indicators for the various treatments (e.g. emulsions with varying type and amount of added compound) were calculated by LSD values (least significant differences). 3. Results and discussion 3.1. Selection of emulsion model based on the effect of processing/ formulation factors At the initial stage, consideration was given on the selection of processing/compositional parameters in order to select an optimum experimental model to subsequently evaluate the activity of exogenous antioxidant compounds. A body of research evidence in sunflower oil-in-water emulsions has concluded an increase of oxidative deterioration (a) with increasing oxidation temperature in the range 10–60 °C (Dimakou, Kiokias, Tsaprouni, & Oreopoulou, 2007) (b) with decreasing concentration in the lipid phase (in the range 10–40% w/w) after 20 days of accelerated thermal oxidation at 60 °C (Kiokias, Dimakou, Tsaprouni, & Oreopoulou, 2006). In the current work, the effect of vegetable oil (used to produce the lipid phase of the emulsion) in the oxidation of o/w emulsions under 60 °C has been investigated. Within four tested samples of edible oils, the following order of increasing oxidation rates of emulsions based on these oils, has been established:
Cottonseed oil ¼ Sunflower oil > corn oil > olive oil: This hierarchy seems reasonable thinking of the fatty acid profile of the tested samples and that oils particularly rich in–the more prone to oxidative degradation–linoleic aid (C18:2) present faster oxidation rates.
282
S. Kiokias, T. Varzakas / Food Chemistry 150 (2014) 280–286
Therefore, it was decided that further work will focus on cottonseed oil-based emulsions. Ten percent (10%) cottonseed oil-in-water emulsions were prepared by Tween or two protein preparations (sodium caseinate, whey protein) as emulsifiers at 1% concentration: Protein-stabilized emulsions have been the subject of recent research on their microstructural stability taking into account that they offer the basis for novel food products such as fresh cheese types, coffe creamers etc. (Kiokias & Bot, 2005). The produced emulsions were quite stable as also proved by measurement of diameter d3,2 which was found at approximately 2.3 lm for each type of emulsifier. A CSLM image confirming the stability of these model emulsions is given in Fig. 4. Between the emulsions prepared by 1% emulsifier (Table 1), the oxidation was promoted much faster in the Tween emulsion (175% increase of initial CD232 absorbance) than in the casein-based emulsion (55%), whereas the whey-protein emulsion presented the lowest oxidative degradation (30%). Based on the above mentioned results and literature information, a decision was made in this experimental work to use the following substrate as the most appropriate to promote faster oxidation rates: Emulsions prepared by 1% Tween in the aqueous phase (pH = 7) and 10% cottonseed-oil in the lipid phase, which would be oxidised in the dark under steady elevated temperature (60 °C). Such a ‘‘vulnerable’’ – by oxidative point of view – model emulsion would make it clearer to distinguish between the activities of added compounds. This model system may differ to the realistic conditions in terms of the selected temperature (higher than the one normally expected during use or storage of relevant food product). However, it ‘‘mimics’’ food relevant products in terms of compositional parameters (e.g. content and composition of the lipid phase) to a greater extent than other more diluted emulsion systems commonly examined so far in the literature. 3.2. Effect of flavonoids (quercetin, catechin) and b-carotene against oxidation The activities of these added compounds (at a concentration of 1.5 mmol kgr1) were tested in 10% cottonseed oil in water emulsions prepared with 1% Tween as emulsifier. The selected concentrations of both flavonoids (1.5 mmol kgr1 is equal to 40 mg/ 100 g) are comparable to their average concentration in food rich sources (e.g. onion containing 35 mg/100 g, USDA, 2011). For bcarotene, a higher concentration 2–5 times to their average concentration of natural carotenoid sources has been used (1.5 mmol kgr1 is 50 mg/100gr). Given that earlier experiments at lower concentrations failed to demonstrate antioxidant character during aerial autoxidation, it was decided to use higher concentration under the current experimental conditions. The emulsions were left to autoxidise at 60 °C and measurements of conjugated dienes at 233 nm were taken periodically as the main method of monitoring lipid oxidation. The appearance of conjugated dienes in oxidised lipids is due to the double bond shift following free radical attack on hydrogen atoms of methylene groups separating double bonds. However, the presence of compounds adsorbing in the region of conjugated dienes may interfere with the
measurements. To minimise such interference, for each tested sample, the initial absorbance at 233 nm (reflecting also the level of baseline oxidation level of ‘‘background’’ oxidation that endogenously existed in the oil phase of the emulsions) was subtracted from the subsequent values to assess the formation of conjugated diene hydroperoxides more accurately. Therefore, starting from a zero value in the onset of the experiment (0 h), the measurements taken periodically, reflect more precisely the level of produced conjugated diene hydroperoxides during the temperature-induced autoxidation under 60 °C. According to the results for conjugated dienes, after an induction period of a few hours, the emulsion started to get significantly oxidised. As clearly shown in Fig. 1, the control emulsion (without added compound) presented a significantly faster increase of absorbance at 233 nm when compared to the quercetin-containing emulsion (p < 0.05). However, the control was oxidised similarly to the ones containing catechnin and bcarotene. Therefore, quercetin presented a clear antioxidant character on the contrary to catechin, which not only showed no effect or b-carotene, but that even presented a tendency for a prooxidant character at certain time intervals of the oxidative process. In these experiments, an alternative spectrophotometric technique that provides an estimate of produced lipid hydroperoxides (ferric thiocyanate method, at 510 nm) was also used. As shown by the results (Table 2), a similar magnitude of activities against the oxidation of the emulsions has been observed for the tested compounds. Overall, a good correlation between the two spectrophotometric techniques (Conjugated Dienes at 233 nm and Ferric Thiocyanate at 510 nm) techniques has been concluded in this experimental work. Regarding their mode of antioxidant activity, flavonoids are known to act as free radicals scavengers (Reaction 1) and terminate the radical chain reactions that occur during the oxidation of triglycerides (Rice Evans, Miller, & Paganga, 1996). The ortho-hydroxyl groups on B ring are unambiguously the most active in donating their hydrogen atoms to free radicals. According to Skerget et al. (2005), quercetin can exert a prooxidant character in pure lipids and an antioxidant activity in emulsions due to its hydroxyl group in position-3. The clear antioxidant effect of quercetin in the cottonseed oil-in-water emulsions of the present study is in agreement with the conclusions of earlier research (Oreopoulou & Tsimogiannis, 2004) at the same concentration level (1.5 mmol kgr1) in bulk cottonseed oils. Interestingly, in the earlier study catechin was found to act as antioxidant when added in bulk (though with a significantly lower effect than quercetin) while presented no activity here against the oxidative destabilisation of the emulsified cottonseed oil. That difference may relate to the partitioning behaviour of catechin between the phases of a dispersed lipid system, which could thereby affect its antioxidant potential. In a study carried out by Mattia et al. (2009) in 20% olive oil emulsions emulsified by Tween, catechin presented the highest surface activity among the tested flavonoids, exerting also the highest prooxidant character. In emulsified systems, oxidation is a phenomenon which takes place in the interface between the two phases (water/oil). Thus, a prooxidant activity could be partly due to the surface activity of phenolic molecules that can enhance the emulsifying ability of Tween 20.
Table 1 Effect of emulsifier type (Tween, whey protein, Na-CAS) at 1% concentrations on the oxidative deterioration of of 10% oil-in-water cottonseed emulsions (at 60 °C) as evaluated by the production of conjugated dienes at 232 nm. Emulsifier
Tween-1% Whey protein-1% Na-CAS 1%
d3,2-lm mean ± SD
2.35 ± 0.10 2.30 ± 0.15 2.32 ± 0.12
Measurement of conjugated dienes (g/kg of oil) Baseline mean ± SD
10 days of oxidation mean ± SD
% Change of CD (CD10d – CDbas)⁄100/CDbas
6.30 ± 0.30 5.76 ± 0.20 5.58 ± 0.19
17.38 ± 1.46 7.43 ± 0.16 8.63 ± 0.43
176 ± 7 29 ± 3 55 ± 8
283
S. Kiokias, T. Varzakas / Food Chemistry 150 (2014) 280–286 control
quercetin
catechin
β-carotene
45
conjugated dienes (g/Kg oil)
40 35 30 25 20 15 10 5 0 0
20
40
60
80
100
120
140
160
oxidation time (hours)
Fig. 1. Effect of flavonoids and b-carotene against the thermal auto-oxidation of 10% cottonseed oil-in-water emulsions at 60 °C (results expressed in terms of conjugated dienes measurements at 233 nm).
Table 2 Effect of quercetin, catechin and b-carotene at 1.5 mmol kgr1 against the thermal aerial autoxidation (60 °C) of 10% cottonseed oil-in-water emulsions stabilised by 1% Tween ((results expressed in terms of lipid hydoperoxides measurements with the ferric thiocyanate method at 510 nm) (oxidative instability increases in the order of 1 > 2 > 3). Time Lipid hydroperoxides (mmol/kg oil, mean ± SD) (h) Control Emulsion with Emulsion with emulsion quercetin catechin
Emulsion with bcarotene
0 15 65 90 110 135
0.00 ± 0.00 0.00 ± 0.00 109.601 ± 24.29 375.001 ± 15.00 509.001 ± 33.00 578.001 ± 143.00
0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 1 123.07 ± 1.12 67.072 ± 3.07 1 330.00 ± 73.00 166.702 ± 0.30 410.002 ± 112.00 187.003 ± 158,00 604.001 ± 90.00 327.001 ± 54.00
OH OH
R.
0.00 ± 0.00 0.00 ± 0.00 111.621 ± 69.14 308.001 ± 119.00 404.002 ± 148.00 550.001 ± 216.00
RH
O OH
Fl-O.
Fl-OH O OH
R.
RH
O O
Reaction 1. Free radical scavenging activity of flavonoids.
On the contrary, catechin acted a better antioxidant than quercetin in the copper-induced oxidation of a much more diluted linoleic acid emulsion (0.02 M) also stabilised by Tween (Beker et al., 2011). As noted there, morin, another flavonoid, acted as the strongest antioxidant while catechin and quercetin presented both antioxidant and prooxidant character depending on their concentrations. Further to partitioning behaviours, the difference in the flavonoid activity may relate to the structure of the C-ring, in which quercetin (flavonol) additionally possess a 4-carbolyn group and a double bond at 2,3 position as compared to catechin. In the literature, the superior antioxidant activity of quercetin has been attributed to the formation of a stable aryloxyl radical upon 1-e oxidation due to C2@C3-double bond and the resulting planar geometry which delocalises the radical throughout the entire
molecule, whereas the flavonoid A and B rings are perpendicular to each other on catechnin (Fukahara et al., 2002). Carotenoids are known to act in a different mode compared to flavonoids, mainly as radical scavengers due to the extensive system of conjugated double bonds in their molecule that makes them very susceptible to radical addition. According to this particular mechanism, b-carotene (as an example) is capable of scavenging peroxyl radicals (Reaction 2). The resulting carbon centred radical (ROO-b-CAR) reacts rapidly and reversibly with oxygen to form a new, chain-carrying peroxyl radical (ROO-b-CAR-OO, Reaction 3i). The carbon centred radical is resonance stabilized depending on the oxygen pressure of the system as will be explained below. The lack of antioxidant character of b-carotene in emulsions, as observed in this study, is in general agreement to the findings of earlier autoxidation studies of emulsions, such as those of (a) Keiko, Higashio, and Terao (1999) who did not observe any antioxidant effect of b-carotene in linoleic acid emulsions; and (b) Heinonen et al. (1997) that even observed a prooxidant character of b-carotene when added in 10% oil-in-water emulsions of rapeseed oil triglycerols. This is interesting considering that the experimental model of the current work is rather different from what was used in earlier studies in terms of both composition (e.g. type and quite higher concentration of oil phase and emulsifier) and processing parameters (higher temperature of oxidation, pH etc.). On the other side, b-carotene was found to exert a clear radical scavenging activity during an earlier investigation (Kiokias & Oreopoulou, 2006) following its addition in 10% sunflower oil in water emulsion. The two emulsions studies were quite close in terms of formulation such as identical amount and close profile of the oil phases and emulsifier/comparable carotenoid concentrations. In addition, the processing parameters were also similar, as in both cases, the emulsions were prepared at pH = 7 and was left to autoxidize in the dark under 60 °C in a shaking water bath. However, a basic difference that may explain how b-carotene acted in each case, concerns the mode of oxidation. In the former study, emulsions were oxidised in tightly capped vials by use of AAPH, (2,20 azo-bis[2-amidinopropane]dihydrochloride) as a free radical generator in the aqueous phase that induced very rapid oxidation rates. As the AAPH radical combines very fast with the limited oxygen, pre-existing in the headspace of the sample vial, it is likely to significantly reduce its content in the tested sample. Therefore, under the relatively low oxygen pressure conditions of the former study the equilibrium of Reaction 3i most probably shifts sufficiently to the left, to effectively lower the concentration of peroxyl radicals and hence reduce the amount of autoxidation in the system. Subsequently, the b-carotene radical adduct can also undergo termination by reaction with another peroxyl radical (Reaction 3ii). On the contrary, the current study relies on aerial autoxidation of the emulsion samples, given that a flow of air was allowed to enter the headspace of their vials, which were only loosely closed. In
+ ROOH
O-O-R
Reaction 2. b-Carotene scavenging activity of peroxyl radicals (Reaction 2).
284
S. Kiokias, T. Varzakas / Food Chemistry 150 (2014) 280–286
(i) ROO-β-CAR• + O2
↔ ROO-β-CAR- OO•
(ii) ROO-β-CAR• + ROO• →
100
inactive products
80
Reaction 3. Reactions of carotenoid centred radical leading to antioxidant or prooxidant character.
60
this case, the oxygen pressure is assumed to be high enough, thereby shifting the equilibrium of Reaction 3i to the right, and favouring a prooxidant rather than an antioxidant character of bcarotene. This mechanism may explain the fact that b-carotene did not prohibit the formation of hydroperoxides (as quercetin did) or exerted no action (as catechin did). On the contrary it presented a tendency of promoting the production of hydroperoxides in the emulsion, compared to the control after 4 days of oxidation (Table 2).
3.3. Effect of antioxidant concentration and mixtures against the oxidative destabilisation of the emulsions
4 days oxidation
20
0 0
0.5
1
1.5
2
2.5
3
Fig. 2. Effect of concentration (in the range 0.75–3 mmol kg1) on the antioxidant activity of quercetin during the thermal auto-oxidation of 10% cottonseed oil-inwater emulsions. Antioxidant activity is expressed in terms of the antioxidant factors (AF%), where AF (%) = [A233 (control-em) A233 (quercet-em)]⁄100/A233 (control-em).
25
C 20 D (
g / 15 k g
control
quercetin (0.75 mmol/kg)
o 10 i l
β-carotene (3 mmol/kg)
)
Given the clear ability of quercetin to protect the emulsions from their oxidative destabilisation, another set of experiments was designed in order to examine any effect of concentration on the antioxidant activity. A number of the same model emulsions was prepared, containing 0 (control), 0.75, 1.5, 2.25 and 3 mmol kg1 quercetin, and were left to autoxidise under the previously described experimental conditions. In order to have a clearer picture of the relevant antioxidant activity (with the control emulsion as reference), the results were calculated at each time/concentration level by the following antioxidant factor (AF) based on the absorbances (A233) at 233 nm:AF (%) = [A233 (control-em) A233 (quercet-em)]⁄100/A233 (control-em). According to this equation, positive values indicate antioxidant effect, given that oxidation is going faster in the control emulsion (A233 (control-em) > A233 (quercet-em)]), whereas negative values show a prooxidant character (A233 (control-em) < A233 (quercet-em). Obviously for 0 concentration of quercetin, AF is considered to be 0. An overall plot presenting the change of antioxidant activity (in terms of %AF) against concentration is given in Fig. 2. As shown by the oxidative measurements, the activity increased significantly with concentration up to 2.25 mmol kg1 and then stabilized. The same trends were observed after both 1 and 4 days of autoxidation. The lack of an increase in antioxidant activity with higher than 2.25 mmol kg1 concentrations of quercetin may simply be due to the fact that the oxidation has already been strongly inhibited by lower concentrations of this compound (0.75– 1.5 mmol kg1). Therefore, a significant difference in the absorbance values with further increase in concentration could not be detected as shown by the AF values. An alternative explanation is that the quercetin molecules are only effective at inactivating the free radicals if they are close to the oil–water interface and can no longer pack into the surface if the concentration is >2.25 mmol kg1. Overall, in the experiments of the present study quercetin was found to be the most efficient antioxidant with its protective effect against the oxidative deterioration of the emulsions reaching a plateau of activity at 3 mmol kg1. Therefore, an antioxidant action dependent on concentration can be concluded for quercetin. Another series of experiments proved no effect of concentration on the activity of b-carotene (between 1.5 and 3 mmol kg1), confirming thereby that this compound can exert no antioxidant effect in the emulsion in the tested range under the current experimental conditions. On the contrary, during the azo-initiated oxidation of emulsions in earlier research conducted by (Kiokias & Oreopoulou, 2006), b-carotene (when added between 0.5 and 5 g l1) had
1 day oxidation
40
5 quercetin (0.75 mmol/kg) + β-carotene (3 mmol/kg)
0 0
20
40
60
80
100
120
140
hrs
Fig. 3. Effect of mixture of quercetin (0.75 mmol kg1) with b-carotene (3 mmol kg1 against the thermal auto-oxidation of 10% cottonseed oil-in-water emulsions at 60 °C (results expressed in terms of conjugated dienes measurements at 233 nm).
demonstrated a clear increase in antioxidant activity with concentration up to 1 g l1 (1.9 103 M) whereas a smaller but still clear change occurred in the whole tested range above this value. 3.4. Interaction of quercetin and b-carotene against the lipid autoxidation of the emulsions The effect of combinations of various antioxidants upon lipid oxidation has been widely reported in the literature to determine whether there is a co-operative interaction between them (Bast, Haanen, & den Berg, 2007). Most effective antioxidant systems for foods commonly contain various antioxidants with different mechanisms of action and physical properties (Kiokias et al., 2009). For instance, mixtures of tocopherols and ascorbic acid have been found to exhibit strong synergistic effects in oil model systems because ascorbic acid reduces tocopherol radicals (Filip, Hradkova, & Smidrkal, 2009). In the literature, there is hardly any evidence for antioxidant interactions between carotenoids and flavonoids in food systems. However, this is an interesting topic of discussion although these compounds present a different mode of action against free radicals in the oxidative reactions. In the
S. Kiokias, T. Varzakas / Food Chemistry 150 (2014) 280–286
285
4. Conclusions During the thermal autoxidation of emulsified cotton seed oil, the flavonoids quercetin and catechin presented a different activity which could be linked to their structural characteristics and varying partitioning behaviours within the emulsion phases. Quercetin presented a clear antioxidant effect that was dependent on concentration (0.75–3 mmol kg1) while catechin did not inhibit the oxidative deterioration within the tested range. Furthermore, the tested b-carotene presented a tendency even for a prooxidant rather than an antioxidant character which could be attributed to the high oxygen pressure of the experimental model. Mixtures of flavonoids with b-carotene did not enhance the activity of the individual compounds. Further investigation is required in this area to optimise the conditions (e.g. concentration, oxygen pressure, antioxidant interactions) under which flavonoids and carotenoids may inhibit the oxidative deterioration of food model emulsions, improving thereby the quality of the final product.
References Fig. 4. CSLM image of 10% cottonseed oil-in-water emulsion (photo taken at 10 nm magnification).
(i) CAR• + TOH → CAR (ii) CAR
+ TO•
+ TO• → CAR• + TOH
Reaction 4. Reduction of carotenoid radicals by tocopherols via an electron transfer mechanism.
current study, a number of experiments have been performed to investigate the combining activity of quercetin and b-carotene in various concentrations (Becker et al., 2007). Figs. 3 and 4 presents the results (in terms of changes of conjugated dienes at 233) for an -added in the emulsion-combination of 0.75 mmol kg1 of quercetin (minimum tested concentration of the strongest antioxidant in this study) and 3 mmol kg1 of b-carotene (maximum tested concentration of the weakest compound in this study). As shown there, the antioxidant mixtures did not exert a tendency for a stronger antioxidant effect against the lipid oxidation when compared to the control emulsion or even to the emulsions containing each individual compound (0.75 mmol kg1 of quercetin or 3 mmol kg1 of b-carotene). In the literature, a few carotenoids have been reported to play a role in recycling phenolic antioxidants e.g. tocopherols after oneelectron oxidation. Kiokias and Gordon (2003) reported that mixtures of certain polar carotenoids (bixin, norbixin) with natural a- or d-tocopherols were stronger antioxidants than the individual compounds during the thermal autoxidation of 10% olive oil in water emulsions. A review of various studies on interaction of carotenoids with tocopherols or ascorbic acid has been provided by Kiokias and Gordon (2004). An explanation for the enhanced synergistic effect could be that carotenoid radicals (CAR) are reduced by a- or b-tocopherols via an electron transfer mechanism (Reaction 4i and ii). However, the experiments in the current project did not prove additive or synergistic effect between quercetin and b-carotene. Therefore, no such protective effect of quercetin in re-converting b-carotene radicals and reducing their potential to further promote lipid oxidation, can be concluded under these experimental conditions.
Alamed, J., Chaiyasit, W., McClements, J. D., & Decker, E. A. (2009). Relationships between free radical scavenging and antioxidant activity in foods. Journal of Agricultural and Food Chemistry, 57, 2969–2976. Bast, A., Haanen, G. R., & den Berg, Van. (2007). Antioxidant effects of carotenoids. International Journal for Vitamin and Nutrition Research, 1998(68), 399–403. Becker, E. M., Ntouma, G., & Skibsted, L. H. (2007). Synergism and sntagonism between quercetin and other chain-breaking antioxidants in lipid systems of increasing structural organization. Food Chemistry, 103, 1288–1296. Beker, B. Y., Bakir, T., Sonmezoglu, I., Imer, F., & Apak, R. (2011). Antioxidant protective effect of flavonoids on linoleic acid peroxidation induced by copper (II)/ascorbic acid. Chemistry and Physics of Lipids, 164, 732–739. Dimakou, C., Kiokias, S., Tsaprouni, I., & Oreopoulou, V. (2007). Effect of processing and storage parameters on the oxidative deterioration of oil-in-water emulsions. Food Biophysics, 2, 38–45. Filip, V., Hradkova, I., & Smidrkal, J. (2009). Antioxidants in margarine emulsions. Czech Journal of Food Science, 27, 9–11. Fukahara, K., Nakanishi, I., Kansui, H., Sugiyama, E., Kimura, M., Shimada, T., et al. (2002). Enhanced radical-scavenging activity of a planar catechin analogue. Journal of American Chemical Society, 124, 5952–5953. Heinonen, M., Haila, K., Lampi, M., & Piironen, V. (1997). Inhibition of oxidation in 10% oil in water emulsions by-carotene with and -tocopherols. Journal of American Oil Chemical Society, 74, 1047–1051. IUPAC, (1987). standard Methods of analysis of oils fats and derivatives (7th ed.). In C. Paquot., and H. Hautfenne (Eds.). Oxford, UK: Blackwell Scientific Publications. Keiko, A., Higashio, H., & Terao, S. (1999). Evaluation of antioxidative activity of vegetable extract of linoleic acid emulsions and phospholipid bilayers. Journal of the Science of Food and Agriculture, 79, 2010–2016. Kiokias, S., & Bot, A. (2005). Effect of protein denaturation on temperature cycling stability of heat-treated acidified protein-stabilised o/w emulsions. Food Hydrocolloids, 19, 493–501. Kiokias, S., Dimakou, C., & Oreopoulou, V. (2009). Effect of natural carotenoid preparations against the autoxidative deterioration of sunflower oil-in-water emulsions. Food Chemistry, 114, 1278–1284. Kiokias, S., Dimakou, C., Tsaprouni, I., & Oreopoulou, V. (2006). Effect of compositional factors against the thermal oxidation of novel food emulsions. Food Biophysics, 1, 115–123. Kiokias, S., & Gordon, M. (2003). Antioxidant properties of annatto carotenoids. Food Chemistry, 83, 523–529. Kiokias, S., & Gordon, M. (2004). Properties of carotenoids in vitro and in vivo. Food Reviews International, 20, 99–121. Kiokias, S., & Oreopoulou, V. (2006). Antioxidant properties of natural carotenoid preparations against the AAPH-oxidation of food emulsions. Innovative Food Science and Emerging Technologies, 7, 132–139. Kiokias, S., Reszka, A. A., & Bot, A. (2004). The use of static light scattering and pulsed-field gradient NMR to measure droplet size changes in heat- treated acidified protein stabilised oil-in-water emulsion gels. International Dairy Journal, 14, 287–295. Kiokias, S., Varzakas, T., & Oreopoulou, V. (2008). In vitro activity of vitamins, flavonoids and natural phenolic antioxidants against the oxidative deterioration of oil-based systems. Critical reviews in food science and nutrition, 48, 78–93. Kiokias, S. (2002). In vitro and in vivo antioxidant properties of natural carotenoid mixtures. Ph.D. thesis. faculty of life sciences, school of food biosciences, the university of reading (pp. 77). Lethuaut, L., Metro, F., & Genot, C. (2002). Effect of droplet size on lipid oxidation rates of oil-in-water emulsions stabilized by protein. Journal of American Oil Chemical Society, 79, 425–430.
286
S. Kiokias, T. Varzakas / Food Chemistry 150 (2014) 280–286
Mattia, C. D., Sacchetti, G., Mastrocola, D., & Pittia, P. (2009). Effect of phenolic antioxidants on the dispersion state and chemical stability of olive oil O/W emulsions. Food Research International, 42, 1163–1170. Nikovska, K. (2010). Oxidative stability and rheological properties of oil-in-water emulsions with walnut oil. Advances in Journal Food Science and Technology, 2, 172–177. Oreopoulou, V., & Tsimogiannis, D. I. (2004). Free radical scavenging and antioxidant activity of 5,7,30 ,40 -hydroxy-substituted flavonoids. Innovative Food Science and Emerging Technologies, 5, 523–528.
Rice Evans, C. A., Miller, N. J., & Paganga, G. (1996). Structure antioxidant activity relationships of flavonoids and phenolic acids. Free Radicals Biology and Medicine, 20, 9–15. Skerget, M., Kotnik, P., Hadolin, M., Hras, A., Simonic, M., & Knez, Z. (2005). Phenols, pronahtocyaninds, flavones, and flavonols in some plant materials and their antioxidant activities. Food Chemistry, 89(2), 191–198. USDA Database for the Flavonoid Content of Selected Foods, Release 3. (2011). US Department of Agriculture. Accessible on http://www.ars.usda.gov/ SP2UserFiles/Place/12354500/Data/Flav/Flav_R03.pdf.
