Inhibition of lard oxidation by fractions of different essential oils By

Se examinó la capacidad de los aceites esenciales de Origanum vulgare L. spp. hirtum, Thymus vulgaris L., Thymus serpyllum L., Satureja montana L. y Satureja cuneifolia Ten. para inhibir la oxidación de la manteca de cerdo pura. Excepto Satureja cuneifolia Ten., todos los aceites esenciales mostraron un acusado perfil fenólico caracterizado por la presencia de fenoles monoterpénicostimol y carvacrol. El método Rancimat ha sido aplicado a manteca de cerdo sembrada con los aceites esenciales y sus fracciones. La capacidad de los aceites y sus fracciones para actuar como inhibidores de la oxidación de lípidos fue menor en comparación con la de antioxidante sintéticos (BHA y BHT), ácido ascórbico y α-tocoferol. El efecto antioxidante de las sustancias ensayadas dependió de la dosis. El periodo de inducción de la manteca de cerdo pura no se afectó por la cantidad de muestra presente en el sistema de reacción.


INTRODUCTION
The two most important types of chemical reactions responsible for quality loss in processed foods have been identified as browning and oxidation (Löliger, 1991).The process of lipid oxidation in foods is responsible for the formation of off-flavors and undesirable chemical compounds, which may be detrimental to health (Brand-Williams et al., 1995).The progression of oxidation in a food system occurs from the formation of radicals through primary oxidation products (lipid hydroperoxides) and secondary oxidation products (aldehydes and ketones) to protein damage.
Antioxidants may be active at different stages of the progression of oxidation in food systems.However, the conditions for which the different antioxidative mechanisms contribute to the protection of an actual food product are not well understood.The use of real food systems for detailed studies of antioxidants is complicated by a large number of factors, which are often unknown or cannot be controlled due to the complex nature of foods.Using simplified model systems, which mimic the main features of a given food system, or antioxidant assays for quantifying the antioxidant action, can be very helpful in clarifying the action of potential antioxidants (Andersen et al., 2002).
Because of market requirements, the use of synthetic antioxidants is being replaced more and more by natural antioxidants from plant sources.Many sources of antioxidants of plant origin have been studied in recent years and numerous types of antioxidants with varied activities were identified (Lagouri et al., 1993;Tsimidou and Boskou, 1994, Milos et al., 2000, Radonic and Milos, 2003).It has been clearly demonstrated in numerous model systems that plant phenolic compounds have antioxidative properties (Schwarz et al., 2001).Previous investigations proved the strong phenolic character of essential oils from oregano (Origanum vulgare L.) (Lagouri et al., 1993;Milos et al., 2000;Vichi et al., 2001), thyme (Thymus vulgaris L.) (Schwarz and Ernst, 1996), wild thyme (Thymus serpyllum L.) (Mailhebiau, 1994) and savory (Satureja montana L.) (Radonic and Milos, 2003).On the other hand, the Satureja cuneifolia essential oil (of Croatian origin) was found to contain a low percentage of phenolic compounds (Milos et al., 2001).
In this study we report the investigation of the ability of the previously mentioned herbs to inhibit the process of lard oxidation with essential oils in relation to their chemical composition.The Rancimat method was applied for these investigations because it mimics the process of lipid oxidation in a simple, reproducible and fast manner.The antioxidant effect of different essential oils and their fractions was evaluated in comparison with, the frequently used antioxidants (BHT, BHA, ascorbic acid and α-tocopherol) and pure constituents thymol and carvacrol.

Materials
The plant materials of oregano (Origanum vulgare L. spp.hirtum), thyme (Thymus vulgaris L.), wild thyme (Thymus serpyllum L.) and two savory species (Satureja montana L. and Satureja cuneifolia Ten.) were collected in Central Dalmatia (Croatia).Plant material (flower tops and stalks), after air-drying, was used for the isolation of the essential oils.A hundred grams of each dried plant material was subjected to a 3 h hydrodystillation using a Clevenger-type apparatus.The obtained essential oils were dried over anhydrous sodium sulphate and stored under nitrogen in a sealed vial at -20 o C until needed .The voucher specimens of plant material as well as their essential oils are stored in the Department of Biochemistry and Food Chemistry, Faculty of Chemical Technology, Split, Croatia.
The essential oils of oregano, thyme, wild thyme and two savory species (0.5 g) were fractionated using a silica gel (30-60 µm, Mallinckrodt Baker B.V., Deventer, The Netherlands) column (length 20 cm; i.d. 2 cm).Pentane (50 mL) was used to obtain a fraction, which contained only non-polar hydrocarbons (CH fraction), and diethyl ether (50 mL) was used to obtain a fraction of polar (oxygen containing, CHO fraction) compounds.These fractions were concentrated to 0.5 mL and subjected to thin layer chromatography (TLC) on silica gel plates in order to check results of the column chromatography separation.Different solvents were used as mobile phase: n-hexane for CH fraction and n-hexane:ethyl acetate 85:15 (v/v) for CHO fraction.Two percent vanillin-sulphuric acid was used as a detection reagent.The fractions obtained by column chromatography were also subjected to GC/MS analysis and good separation results were confirmed.
In order to obtain a fraction of phenolic compounds, 0.5 g of the essential oils from Origanum vulgare L. spp.hirtum, Thymus vulgaris L. and Satureja montana L. was dissolved in 5 mL pentane and extracted with sodium hydroxide solution (20%) in water.In this manner, phenolic compounds were removed from the pentane layer.The aqueous phase, containing dissolved phenolic compound sodium salts, was neutralized with hydrochloric acid solution (10%) in water.Finally, isolation of the phenolic compounds was performed by extraction with pentane (5 x 5mL).The effectiveness of this separation method was tested by TLC on silica gel plates (mobile phase: n-hexane:ethyl-acetate 85:15 v/v).Purity of the phenolic compounds fraction was confirmed by GC/MS analysis.
The lard applied in the Rancimat method was home made (free of added antioxidants or preservatives).

GC-MS analysis
The analyses of the volatile compounds were run on a Hewlett-Packard GC-MS system (GC 5890 series II; MSD 5971A, Hewlett Packard).The fused-silica HP-20 M polyethylene glycol column (50 m x 0.2 mm, 0.2 µm thickness, Hewlett-Packard) was directly coupled to the mass spectrometer.The carrier gas was helium (1 mL/min).The linear retention indexes for all the compounds were determined by co-injection of the sample with a solution containing the homologous series of C8-C22 n-alkanes (Van den Dool and Kratz, 1963).The individual constituents were identified by their retention indexes referring to the compounds known from literature data (Adams, 1995); and also by comparing their mass spectra with spectra of either the known compounds or with those stored in the Wiley mass spectral database (Hewlett-Packard).

Induction period of lard oxidation (Rancimat method)
The induction period of lard with and without the addition of antioxidants was determined with the Rancimat model 743 (Metrohm, Switzerland) at 100 o C and the airflow of 20 L/h.The ethanolic solutions of different concentrations of antioxidant (100 µL) were added to the lard (2.5 g) giving a final concentration of 0.16%, 0.12%, 0.08%, 0.04% and 0.016% (w/w) of the antioxidant in the reacting system.
The antioxidant activity index (AI) is calculated from the measured induction times, according to the following formula by Forster et al. (2001).AI = Induction time of lard with antioxidant / Induction time of pure lard Although this technique has been questioned (Frankel, 1993), it is a procedure commonly used in the food industry and governmental analytical laboratories (Parejo et al., 2003).

Chemical composition of essential oils
The chemical composition of total and fractionated essential oils from oregano, thyme and wild thyme is shown in Table I, while Table II show the chemical composition of the total and fractionated essential oil from two savory species, Satureja montana L. and Satureja cuneifolia Ten.Except Satureja cuneifolia Ten., all the essential oils showed qualitative similarities and a strong phenolic character.

Inhibition of lard oxidation (Rancimat method)
The autoxidation of lipids can be inhibited or retarded by adding different antioxidants.They function either by scavenging chain-carrying peroxyl radicals or by diminishing the formation of initiating lipid radicals (Yamamoto and Niki, 1990).The mechanism of the antioxidant action of phenolic antioxidants in lipids has not been completely explained so far (Yanishlieva et al., 1999).
Table III shows the induction times of lard in the presence of the total essential oils and their fractions from oregano, thyme, wild thyme and two savory species in comparison with reference antioxidants (BHT, BHA, α-tocopherol and ascorbic acid) and pure compounds thymol and carvacrol.The concentration of tested ethanolic solutions of antioxidants, which were added to the lard, was 0.16% (w/w).The inhibitory effect of the samples is expressed as the antioxidant activity index (AI).
The higher induction period of the lard with antioxidant added, compared to the control (pure lard) means improved antioxidant activity of that compound (von Gadow et al., 1997).
Reference compounds, BHA and α-tocopherol, were the most potent inhibitors of the autoxidation of lipids with antioxidant activity indexes of 7.2 and 6.3.Synthetic antioxidant BHA has already been approved to control lipid oxidation in foods (Imaida et  al., 1983; Okada et al., 1990; Ayar et al., 2001), while α-tocopherol is known as the main protection factor in a system like LDL, low-density lipoproteins (Jia et  al., 1998; Zhu et al., 1999; Andersen et al., 2002).Ascorbic acid also showed a potent inhibitory effect against the autoxidation of lipid (AI = 4.3).It is known that it is used in patented protection systems for lard and marine oils (Löliger, 1991).Among the reference compounds, BHT showed the lowest antioxidant activity index (3.6),which can be explained by its volatility at the high temperature of the test (100 o C) so BHT could rapidly sweep from the lipid (von Gadow et al., 1997).
In comparison with reference antioxidants, the total essential oils tested and their fractions showed a poor inhibitory effect against the lard autoxidation process under the conditions of this method, but very similar to those obtained for pure thymol and carvacrol.Among the essential oils tested the longest induction time for lard was achieved by the addition of oregano essential oil (AI = 2.0).Results for the other tested essential oils decreased in the order Thymus vulgaris L.>Thymus serpyllum L.>

288
Grasas y Aceites Satureja montana L.>Satureja cuneifolia Ten.Thymol and carvacrol, which were the main components of essential oils of Origanum vulgare L. spp.hirtum, Thymus vulgaris L., Thymus serpyllum L. and Satureja montana L., can be effective in the inhibition of lard autoxidation at 35 o C at a concentration of 0.1% (Lagouri et al., 1993).Lagouri and Boskou (1996) concluded that the inhibition of oxidation by the essential oils from plants of the oregano species was highly dependent on the content of carvacrol and thymol.However, results presented in this study proved the opinion established by Yanishlieva et al. (1999) concerning the volatilities of thymol and carvacrol at 100 o C, which is the main reason for their relatively low inhibitory effect observed by this method.The essential oil from Satureja cuneifolia Ten.showed the poorest inhibitory effect probably because of the non-phenolic character of this oil.
The inhibitory effect of oxygen-containing compound fractions (CHO) was very similar to those for essential oils.Oxygen-containing compound fraction from Thymus vulgaris L. showed the longest induction time (AI = 2.0).Almost all CH fractions showed a prooxidative effect (AI < 1), except Origanum vulgare L. spp.hirtum and Satureja montana L. CH fractions with AI = 1.The strongest prooxidative effect showed CH fraction from Satureja cuneifolia Ten. (AI = 0.1), which could be the reason for the lower AI value of the total oil in comparison to the AI value of the CHO fraction of this oil.Phenolic fractions from Origanum vulgare L. spp.hirtum, Thymus vulgaris L. and Satureja montana L., consisted of thymol and carvacrol (Table I and II), and exhibited lower inhibitory effect in comparison to the effect of oxygen-containing compound fractions, which continues the speculations about the sinergy among minor oxygen containing compounds (Janssen et al., 1988, Kulisic et al., 2004).
The results presented in Table IV prove that the antioxidant effect of tested essential oils was dose-dependent.The induction times of lard were not affected by the different quantity of lard (5.2 h was induction time for 2.0 g, 2.5 g and 3.0 g of lard).
Since the conditions of this test were extreme (temperature -100 o C and air flow -20 L/h), the volatile nature of tested antioxidants reduces their inhibitory effect in the lard autoxidation process.However, due to their strong phenolic character, the essential oils of the herbs tested could find their use in food protection, mostly in food processing at lower temperatures or in the process of food storage.
The program used was 4 min isothermal at 70 o C, then 4 o C/min to 180 o C and 10 min isothermal.The injection port temperature was 250 o C and the detector temperature was 280 o C. Ionization of the sample components was performed in the EI mode (70 eV).

Table II The composition (area %) of Satureja montana L. and Satureja cuneifolia Ten. essential oils
1,2,3 as in TableI.4Tentativelyidentification on basis of the mass spectra (MS) only