A practical study on the feasibility of alpha and gamma-tocopherol quantification for distinguishing Iberian pig feeding systems ; Estudio práctico de la fiabilidad de la cuantificación de alfa y gamma-tocoferol para diferenciar los sistemas de alimentación del cerdo ibérico

Este estudio se llevó a cabo para comprobar la fiabilidad del uso de la cuantificación de alfa y gamma­tocoferol en grasa y músculo para establecer una clasificación correcta de los cerdos ibéricos de acuerdo a su alimentación. Las muestras se obtuvieron durante tres campañas diferentes procedentes de cuatro categorías de cerdos según se descri­ be en la norma de calidad (FREE­OUT: «bellota», FREE­ OUT­FEED: «recebo», FEED­OUT: «campo» y FEED­IN: «cebo»). Se calcularon funciones discriminantes con los da­ tos obtenidos de las campañas 1 y 2 y se validaron. La clasi­ ficación de las muestras de grasa y músculo obtenidas de la tercera campaña en las cuatro categorías de acuerdo a las funciones discriminantes calculadas alcanzó un rango medio de éxito del 76% para diferenciar el origen de las muestras. La cuantificación en el músculo pareció tener una mayor ca­ pacidad de predicción. Las ecuaciones de regresión para cuantificar el peso ganado en montanera en función de la concentración de gamma­tocoferol tuvieron un mayor R pa­ ra el músculo que para la grasa (R= 0.81 vs. 0.62). La unión de alguna de las categorías de alimentación establecidas en la norma de calidad podría incrementar la fiabilidad del méto­ do de cuantificación de los tocoferoles alfa y gamma hasta entre un 89 al 98%.


INTRODUCTION
The Iberian pig, an autochthonous breed mainly reared in the southwest of Spain, is known for the high quality and commercial value of its products.Products from Iberian pigs reared in a freerange production system and fed on natural resources (mainly acorns and grass) reach the highest quality and prices (LópezBote, 1998;García et al., 1996) and are characterized by having high levels of oleic acid (Rey et al., 2006a) and antioxidants such as tocopherols (Rey et al., 2006b, Tejerina et al., 2012) and other compounds (Cantos et al., 2003) provided by acorn and grass consumption (Rey et al., 1998;Tejerina et al., 2011).If the amount of acorns and grass is not enough to cover feeding requirements because of climatic conditions, pigs can be supplemented with feeds that result in a loss in quality (LópezBote, 1998).
To prevent fraud and to properly identify the feeding background of the animals, quality control practices such as the fatty acid analysis of adipose tissue by gas chromatography have been widely used (BOE, 2001).However, this method has been recently devalued (BOE, 2007) because of the incorrect classification of pigs fed in intensive conditions with an oleic acidenriched mixed diet as those that had been fed in free range conditions.Consequently, during the last few years much effort has been invested to finding new methods to differentiate Iberian pigs according to feeding type.One of the possibilities studied has been the quantification of alpha and gamma tocopherols.Hence, in a previous study, Rey et al. (2006a) reported that pigs raised freeranged showed higher gammatocopherol concentration Pigs from the first two seasons were raised in controlled farms or experimental institutions (Valdesequera, Badajoz, Spain), which had implemented a traceability system and a periodic inspection program.Pigs from the last season were raised in practical quality control conditions and they came from a wide variety of farms.
Animals were stunned and slaughtered at a live weight of approximately 152.5 (± 3.7 kg) and samples of subcutaneous fat were obtained from the tail insertion area in the coxal region and kept frozen at -20 °C until analysis.Additionally, a small piece of muscle (Longissimus dorsi) was taken and kept in the same conditions.
Classification was performed according to the tocopherol levels on blind samples so that the feeding group was unknown at the time of analysis.

Alpha and Gamma-Tocopherol Quantification
Concentrations of alpha and gamma tocopherols in muscle (Longissimus dorsi), were quantified by the direct method described by Rey et al. (1996) without previous saponification.Muscle samples (0.8 g) were homogenized in a 0.054 M dibasic sodium phosphate buffer adjusted to pH 7.0 with HCl.After mixing with absolute ethanol and hexane, the upper layer containing tocopherol was evaporated to dryness and dissolved in 200 µL ethanol prior to analysis by reverse phase HPLC (Agilent 1100, Agilent Technologies, Waldbronn, Germany).Separation was carried out on an Agilent Technologies Lichrospher RPC18 column (250 mm × 4 mm i.d., 5 µm particle size), the mobile phase was methanol/water (97:3 v/v) at a flow rate of 2 ml/min and peaks were recorded at 292 nm.The identification and quantification of both tocopherols was carried out by means of a standard curve (R 2 =0.9999) developed with the pure compounds (Sigma, Alcobendas, Madrid).Results were expressed as µg of alphatocopherol or gammatocopherol per g of fresh matter.
Tocopherols in subcutaneous fat were quantified as described by Rey et al. (2006b) in which samples were saponified in the presence of KCl (1.15%) and KOH (50%).Tocopherols were extracted with hexane and the upper layer was evaporated and dissolved in ethanol prior to analysis by reverse phase HPLC (HP 1100, provided with a diode array and fluorescence detector) (Agilent Technologies, Waldbronn, Germany).Analysis by HPLC was carried out following the same procedure described before with the exception that peaks were detected by a fluorescence detector (Agilent technologies series 1200) set at lexcitation: 295 nm and lemission: 330 nm. in muscle and subcutaneous fat than pigs fed a formulated diet in confinement.In an additional experiment, these authors (Rey et al., 2006b) found a direct relationship between gammatocopherol accumulation and time of freerange feeding and weight gained outdoors, which suggests that this could be used as a discriminator for different pig meat qualities.However, this study was carried out under standard conditions.Taking into account the heterogeneity of Iberian pig production, which is similar to other outdoorreared based systems (Pugliese and Sirtori, 2012), more research is needed to evaluate the feasibility of tocopherol quantifications in practical situations.
The objectives of the present study were first to quantify isomers (alpha and gammatocopherol) in subcutaneous fat and Longissimus dorsi muscle from Iberian pigs fed in different commercial feeding situations; and second, to study the feasibility of tocopherol (alpha and gamma) quantification for differentiating and establishing a correct classification according to feeding background in practical situations.

Samples
Samples (subcutaneous fat and Longissimus dorsi muscle) were obtained from Iberian pigs raised in controlled feeding conditions in three different seasons : 20082009, 20092010 and 2010 2011.During the 20082009 season, fat (n=180) and muscle samples (n=188) were collected from the following feeding groups: (1) FREEOUT: pigs raised in freerange conditions and exclusively fed acorns and grass (they put on at least 46 kg or 4 arrobas) in the evergreen oak forest in different geographical areas (Salamanca and Badajoz, Spain) (fat n=61; muscle n=57) (known as "bellota" according to BOE, 2007); (2) FREEOUTFEED: pigs raised in freerange conditions and fed acorns and grass (they put on at least 29 kg or 2,5 arrobas in freerange conditions) and also with a mixed diet during the fattening phase (fat n=12; muscle n=12) (known as "recebo"); (3) FEEDOUT: pigs fed either a conventional diet (fat n=62; muscle n=62) or a fatenriched diet: (fat n=13; muscle n=25) and with access to grass (known as "campo"); (4) FEEDIN: pigs fattened with a standard commercial pig diet (fat n=32; muscle n=32) (known as "cebo").
During the 20102011 season, fat (n=336) and muscle (n=164) samples were collected from very different farms according to the following A.I. Rey, D. AmAzán AnD J. GARcíA-cAsco pigs exclusively fed with acorns and grass in the forest (FREEOUT) had the highest concentration of gammatocopherol (P<0.0001).The FREEOUT FEED group showed intermediate gammatocopherol concentrations and the lowest values were found for the FEEDOUT and FEEDIN groups.The concentration of alphatocopherol was also higher in the fat from pigs fed in freerange conditions (FREE OUT) than the fat from pigs in the other groups (P<0.0001).However, these differences between FREEOUT and FREEOUTFEED were not found in the muscle.In both tissues (fat and muscle), the group fattened with a commercial diet (FEEDIN) had the lowest alphatocopherol concentrations when compared with the other groups (P<0.0001).Intermediate alphatocopherol values were found in the fat and muscle from pigs in the FEEDOUT group.

Discriminant Analysis
The discriminant functions obtained from the tocopherol concentrations in the fat and muscle samples of these two first years are presented in Table 2, their corresponding validation in Table 3 and how the feeding groups were differentiated between them, in Figures 1 and 2. A decimal logarithm was taken for the gammatocopherol variable to normalize values.The R2 was higher for the logarithm of the gammatocopherol than for the alphatocopherol variable (0.68 vs. 0.44) in the linear discriminant functions for fat.In muscle samples, the proportion of gamma:alphatocopherol was used to improve the prediction ability and the gammatocopherol variable had higher R 2 (0.90) than alpha (R 2 =0.43) and the ratio between both tocopherols, gammaalpha (R 2 =0.70).Gamma tocopherol was the main compound that contributed to function 1 in the canonical discriminant functions for fat and muscle samples, while alphatocopherol principally contributed to function 2. Samples were assayed in duplicated for both analyses (muscle and fat).Recovery of gamma tocopherol was not lower than 80% and the internal standard (alphatocopheryl acetate) was added.

Statistical Analysis
The data were analyzed as a completely randomized design using the general linear model (GLM) procedure contained in SAS version 9 (SAS Inst.Inc., Cary, NC).The comparative analysis between means was conducted using the Duncan test.Data are presented as the mean of each group and the SD (standard deviation) together with the significance levels (P value).Differences between means were considered statistically significant when P < 0.05.Linear discriminant functions were developed using the PROC DISCRIM procedure contained in SAS version 9 to assess the suitability of tocopherol concentration for discriminating groups of different feeding backgrounds.Equations to quantify the weight gain depending on gamma tocopherol in fat and muscle samples were developed using the PROC REG statement in SAS.Exponential equations were calculated using the SAS PROC NLIN procedure.The exponential response was calculated using the following equation: y= a+b*exp (-cx) .

Alpha and Gamma-Tocopherol Quantification
The average concentration of alpha and gamma tocopherols and the ratio gamma:alpha in fat and muscle samples of the first two years (seasons 1 and 2) according to the feeding background is presented in Table 1.Fat and muscle samples from the fat samples classified as FREEOUTFEED and 2% as FEEDOUT.However, the FEEDIN group had a high proportion of false positives (20% that were assigned to FEEDOUT and 3.5% assigned to FREEOUTFEED).The FREEOUTFEED group had the highest proportion of false positives that were classified as FREEOUT feeding (31.5%),FEEDOUT (11%) or FEEDIN (2%).
In the muscle samples, the validation results by cross validation of the calculated discriminant functions showed that 76% of the total were correctly classified and 24% were assigned to the The validation results of the discriminant functions (Table 3) show that 24% of the fat samples were not correctly classified giving a 76% success rate.Moreover, 31% of the fat samples from pigs fed freerange (FREEOUT) were assigned to the FREEOUTFREE group.For the group FREEOUTFEED, 72% of the samples were assigned to the correct feeding system however, 19% were considered as FREEOUT, 5% as FEEDOUT and 3.5% as FEEDIN.The group with the lowest proportion of false negatives was FEED IN with 96% correct classification, showing 2% of   showing the relationship and distribution among feeding systems (FREEOUT: pigs raised in freerange conditions with acorns and grass ; FREEOUTFEED: pigs raised in freerange conditions and fed with acorns and grass and also with feed during the fattening phase; FEEDOUT: pigs fed with feed and with access to grass; FEEDIN: pigs fattened with a standard commercial pig feed).
The classification results for fat and muscle samples from season 3 according to the discriminant functions obtained from seasons 1 and 2 are presented in table 4. A high percentage of fat and muscle samples from the FREEOUT feeding group were correctly classified (88% for fat and 98% for muscle samples).The FEED OUT group also had a high percentage of correct classifications (87% for fat and 89% for muscle).The FEEDIN group had intermediate proportions wrong feeding background.The Quantification of tocopherols in muscle for feeding discrimination seemed to be more feasible.The FREEOUT group was correctly classified in 85% of the muscle samples, FREEOUTFEED in 84%, FEEDOUT in 78% and FEEDIN in 95%.The percentage of the muscle samples from FREEOUT that were wrongly assigned to FREEOUTFEED was 15%.Moreover, 11% of the samples from FREEOUT FEED were wrongly considered as FREEOUT and 5% as FEEDOUT.On the other hand, 22% of the muscle samples from the FEEDOUT feeding A prActicAl study on the feAsibility of AlphA And gAmmA-tocopherol quAntificAtion for distinguishing… other regression equations.So, coincidences for the group FREEOUTFEED were only of 25%, while 33% of these samples were considered as FEED OUT feeding and 42% as FEEDIN.

DISCUSSION
The feeding background affected the alpha and gammatocopherol concentrations.Samples from the FREEOUT group had higher (P<0.01)concentrations of gammatocopherol in fat and muscle than those from the FREEOUTFEED, FEEDOUT and FEEDIN groups.These results can be attributed to the high content of gamma tocopherol in acorns, which has been previously reported (Rey et al., 1998;Rey et al., 2006a;Isabel et al., 2009;LópezCarrasco et al., 2011;Tejerina et al., 2011).Since FREEOUTFEED pigs consumed lower amounts of acorns, they had a lower proportion of gammatocopherol in tissues than pigs exclusively fed freerange, but higher than pigs fed with a mixed diet indoors (FEEDIN) or those fed with a mixed diet and consuming grass outdoors (FEEDOUT).So the presence of gamma tocopherol in tissues is directly related to acorn consumption.On the other hand, grass has been reported to provide tocopherol mainly in the form of alpha, while the proportion of gammatocopherol is much lower than in acorns (Rey et al., 1998;Rey et al., 2006b;LópezCarrasco, et al., 2011;Tejerina et al., 2011).Hence, the FEEDIN group exclusively fed a mixed diet had the lowest concentration of alphatocopherol in muscle and fat than the other groups, while FEEDOUT and FREEOUTFEED groups showed intermediate values.A similar trend in the concentrations of both tocopherols has been reported before in different tissues (Rey et al., 1997;Cava et al., 2000;Rey et al., 2006a,b;Rey et al., 2010) for pigs fed in freerange conditions, with comercial feed diet, or fed during a different period in the forest.However, this is the first study examining tocopherol concentration in fat and (64% and 65% for fat and muscle, respectively) and the FREEOUTFEED group showed the lowest percentage of correct assignments (40% for fat and 48% for muscle).Half of the samples from the FREEOUTFEED group were classified as FREE OUTFEED (48% in fat and 44% in muscle) and smaller proportions were assigned to FEEDOUT (12% in fat and 4% in muscle samples).

Regression Equations
Additionally, regression equations were developed to quantify the weight gained by the animals depending on the gammatocopherol concentration in the fat and muscle samples (Table 5).The regression equations (linear, quadratic and exponential) obtained from muscle samples had higher adjustments than those from fat (R 2 = 0.69 vs. 0.60 respectively in the linear equation; R 2 = 0.81 vs. 0.62 respectively in the quadratic equation; R 2 = 0.80 vs. 0.61 respectively in the exponential equation).
The comparison between classifications based on regression equations obtained from the tocopherol concentration of muscle to calculate weight gain and the results obtained by the discriminant functions are presented in Table 6.A high percentage of classification coincidences were found when using linear regression equations or linear discriminant functions.Hence, 100% of the muscle samples from the group FREEOUT FEED were classified in the same group.Moreover, there were 89% of coincidences in samples from the FREEOUT feeding group.However, when using discriminant functions or weight gained, coincidences were lower in the FEEDOUT group.Results were similar when using the quadratic regression equations along with the linear discriminant functions with the exception of 17% of the muscle samples from FREEOUTFEED that were classified as FEEDIN.The application of the exponential regression equation instead of the discriminant functions was not as accurate as the  et al., 2006b), which is again related with the high content of gamma tocopherol found in acorns.Moreover, the canonical discriminant functions (Figures 1 and 2) that represented how feeding groups were differentiated between them showed that function 1, which mainly considered gammatocopherol concentration, separated samples from FEEDIN or FEEDOUT and FREEOUT or FREEOUTFEED.Meanwhile, alphatocopherol (mainly considered in Function 2) separated the group FEEDIN from the others.In the present study it is also of interest to observe that even though the percentage of total classification errors in muscle and fat samples were similar (24%), muscle seemed to have higher predictive ability as demonstrated by the higher percentage of samples classified into the correct feeding group and the lower proportion of false positives and negatives in the cross validation analysis for this tissue (Table 3).Moreover, the gammatocopherol variable obtained from muscle samples of the first two years to develop linear discriminant functions showed higher R2 values than in fat (R 2 = 0.90 vs. 0.68) (Table 2).There is no other information in the literature concerning the feasibility of alpha and gamma tocopherol discriminant analysis for classifying pig samples according to their feeding background.So, this information is of importance for the Iberian pig production and quality sector.muscles from pigs fed with a mixed diet and with access to grass outdoors (FEEDOUT).This group which is known as "campo" was included in the latest quality policy for the Iberian pig (BOE, 2007) but no information was available concerning the alpha and gammatocopherol concentrations of this group or a comparison with the other groups in order to establish its correct classification.
Concerning the quantified concentration of gamma and alpha tocopherols of the present study, values were similar to those reported before (Rey et al., 2006 a,b;Daza et al., 2005), in this type of muscle obtained by direct extraction with differences probably due to variations in acorn and grass consumption, composition, or the delay until analyses were carried out (Rey et al., 2010).However in fat, the gammatocopherol concentrations of the present study were lower than those reported in previous studies (Rey et al., 2006a, 2006b, Rey et al., 2010), due not only to the same variations as in muscle but also in part to small modifications of the analytical procedure which allowed for a more accurate quantification of gammatocopherol.
When discriminant functions were developed from the tocopherol concentration of fat and muscle samples to differentiate the feeding background (Table 2), a better fit for gammatocopherol was observed than for alphatocopherol.A previous study reported closer relationships among the days Iberian pigs after a supplementation of 10 mg alphatocopheryl acetate kg -1 or 125 mg kg -1 in feed (Rey et al., 1997).On the other hand, Daza et al. (2005) found that the muscles from Iberian pigs fed with diets supplemented with alphatocopheryl acetate up to 260 mg kg -1 contained average values of 4.4 µg g -1 and these alphatocopherol levels were similar to those from pigs fed on acorns and grass.However, Ventanas et al. ( 2006) found lower alphatocopherol concentrations in the muscle from pigs fed on natural resources than those supplemented with 250 mg kg -1 in feed.Differences in alphatocopherol accumulation in tissues and the interaction between FEEDOUT and FEEDIN groups or to a smaller extent with other groups could also be affected by variations in the tocopherol composition depending on the type of grass as a function of the predominant species (Mutetika and Mahan, 1993;Lynch et al., 2001) and the date or year (LópezCarrasco et al., 2011;Tejerina et al., 2011).Mutetikka and Mahan (1993) have reported alphatocopherol values in grass of between 29 to 100 mg of alphatocopherol kg -1 DM; other authors found levels of 158.3 mg kg -1 DM (Rey et al., 2006b) and 276 mg kg -1 DM (Isabel et al., 2009).Finally, taking into account that the feeding classification established by the Spanish legislation is in part dependent on the minimum weight gained during the fattening phase (4 arrobas in freerange group; 2,5 arrobas in pigs fed in free range and with a mixed diet feed; and 0 arrobas in pigs fed exclusively with a mixed diet), and the relationship previously observed (Rey et al., 2006b) between the weight gained and the gamma tocopherol concentration, regression equations were calculated (Table 5).Again, muscle samples showed better fit than the fat (R 2 =0.81 vs. 0.62 for quadratic equation; P<0.0001).Moreover, a higher quadratic adjustment than linear (R 2 =0.81 vs. 0.69) was found in muscle while in fat these were similar (R 2 =0.62 vs. 0.60).Previous calculated regression equations (Rey et al., 2006b) showed a saturation effect in muscle but not in subcutaneous fat.When classifications of muscle samples from the last season according to the discriminant functions and weight gained equations were compared (Table 6), a high percentage of similarities were found.It is interesting to point out that because the FEEDOUT classification was not obtained by the regression equations, this group of samples was classified as FEEDIN.So, the three groups (FREEOUT, FREEOUTFEED, and FEEDIN) had a percentage of coincidences higher than 83% by discriminant functions and regression equation (linear and quadratic) classification.There is no other information in the literature concerning the feasibility of using alpha and gammatocopherol concentrations for classification according to the feeding background.To our knowledge, there is a single previous study in which the relationship between gammatocopherol and weight gained was reported and regression equations were calculated In order to corroborate the feasibility of the discriminant equations in practical situations, they were applied to a group of blind samples obtained during the third year (Table 4).Prediction ability was again superior for muscle discriminant functions than for those obtained from fat samples.Muscle samples from pigs fed freerange were almost totally correctly classified (98%), while some of the fat samples from this feeding group (FREE OUT) were assigned as FREEOUTFEED (12%).In both tissues, the group with highest number of classification errors was FREEOUTFEED in which 48% of fat samples and 44% of muscle samples were classified in the FREEOUT group.The incorrect classification of FREEOUTFEED pigs as FREEOUT and vice versa was also observed in the cross validation of the discriminant functions with data from seasons 1 and 2. This would indicate that based on the alpha and gammatocopherol content of these samples, both feeding groups could be considered of similar quality.So, even though pigs from freerange feeding have to put on at least 4 arrobas in the forest, while pigs from FREEOUTFEED group need to reach at least 2,5 arrobas (BOE, 2007), when the pigs' weight gain got close to the lower or upper established limits, the correct feeding classification may be difficult to achieve (Figures 1 and 2).Moreover, interactions between these two feeding groups (FREEOUT and FREEOUTFEED) could be explained by individual differences in the intake of either acorns or grass or by their variable composition.Hence, Tejerina et al. (2011) andLópezCarrasco et al. (2011) reported differences in gamma and alpha tocopherol concentrations of acorns and grass with maturation time and so with the month of freerange feeding.Moreover, LópezCarrasco et al. (2011) have described that the holm oak tree and its location in the forest can also have an influence on the gammatocopherol concentration of acorns.Moreover, the metabolic accumulation of tocopherols may be affected by differences in the consumption of one or the other because the uptake of tocopherol isoforms seems to be different (Ikeda et al., 2001).
It is also of interest to note that 32% of fat and 22% of muscle samples were wrongly classified as FEEDOUT instead of FEEDIN during the third season (table 4).The interaction between these two groups was also observed in the cross validation of the discriminant functions for seasons 1 and 2 (Table 3).The FEEDOUT and FEED IN groups mainly differed in the concentration of alphatocopherol because the concentration of gammatocopherol was similar.Taking into account that alphatocopherol content depends on the alphatocopherol levels in the feed, classification errors between these two feeding groups could be mainly explained by differences in either grass intake (affected by environmental factors), or the alphatocopheryl acetate supplementation of feeds.Hence, a range of values from 2.2 to 3.8 µg g -1 of alphatocopherol were reported in muscle from ( Rey et al., 2006b) but this study was carried out in experimental situations where the groups of pigs were of the same origin and genetics and (n=10 11) fed in a single geographical location (Oropesa, Toledo).So the results obtained in the present study, which takes into account the heterogeneity of the Iberian pig sector, are more applicable to discriminating the feeding background in practical situations.

CONCLUSIONS
In conclusion, the quantification of gamma and alphatocopherol in the fat or muscle of Iberian pigs is a useful method to discriminate the feeding background of the animals and permitted a successful classification in 76% of cases (either by linear discriminant functions or regression equations) into the four qualities defined by the Spanish Quality Policy for the Iberian Pig.Quantification in muscle seemed to be more accurate than in fat and resulted in a lower number of false negatives and positives.Merging of the categories to three (the groups of pigs fed with feed "cebo" and access to grass "campo" as a unique category) may allow for improving the accuracy of the classification using gamma and alphatocopherol quantification to 89% of cases.When the feeding categories are reduced to two: 1) pigs fed with acorns and grass independently of the weight gained (4 or 2,5 arrobas) and 2) pigs fed with feed (without or with access to grass) a 98% success rate could be achieved in classification using gamma and alphatocopherol concentration determination.This information contributes to a great extent to the correct classification of the Iberian pig meat and products in practical situations.
Figure 1Canonical discriminant functions for fat samples (seasons 1 and 2) showing the relationship and distribution among feeding systems (FREEOUT: pigs raised in freerange conditions with acorns and grass; FREEOUTFEED: pigs raised in freerange conditions and fed with acorns and grass and also with feed during the fattening phase; FEEDOUT: pigs fed with feed and with access to grass; FEEDIN: pigs fattened with a standard commercial pig feed).
Figure 2Canonical discriminant functions for muscle samples (seasons 1 and 2) showing the relationship and distribution among feeding systems (FREEOUT: pigs raised in freerange conditions with acorns and grass ; FREEOUTFEED: pigs raised in freerange conditions and fed with acorns and grass and also with feed during the fattening phase; FEEDOUT: pigs fed with feed and with access to grass; FEEDIN: pigs fattened with a standard commercial pig feed).

Table 1 Gamma and alpha-tocopherol concentration in fat and muscle samples according to the feeding background obtained from seasons 1 and 2 µg/g sample FAT MUSCLE
A prActicAl study on the feAsibility of AlphA And gAmmA-tocopherol quAntificAtion for distinguishing…

Table 2 Linear discriminant functions in fat and muscle samples obtained from data of seasons 1 and 2
FREEOUT: pigs raised in freerange conditions and exclusively fed with acorns and grass (known as bellota) 2 FREEOUTFEED: pigs raised in freerange conditions and fed with acorns and grass (they put on at least 29 kg in free conditions) and also with feed during the fattening phase (known as recebo) 1 3 FEEDOUT: pigs fed with either a conventional feed or a fatenriched diet and with access to grass (known as campo) 4 FEEDIN: pigs fattened with a standard commercial pig feed (known as cebo). 5SD: Standard deviation. 6loggtoc: decimal logarithm of gammatocopherol concentration

Table 5 Calculated equations to quantify the weight gained (arrobas) depending on gamma-tocopherol concentration (mg g -1 ) in fat and muscle (from data of seasons 1 and 2)
a SD: Standard deviation b Gtoc: gammatocopherol concentration A.I. Rey, D. AmAzán AnD J. GARcíA-cAsco of freerange or weight gained and the concentration of gammatocopherol (Rey

Table 6 Comparison of classification according to discriminant functions and weight gained equations in muscle samples from campaign 3 (% assigned)
.I. Rey, D. AmAzán AnD J. GARcíA-cAsco Daza A, Rey AI, Ruiz J, LopezBote CJ.2005.Effects of feeding in freerange conditions or in confinement with different dietary MUFA/PUFA ratios and αtocopheryl acetate, on antioxidants accumulation and oxidative stability in Iberian pigs.Meat Sci.69, 151163.Meat Sci.90, 511518.Rey A, LópezBote CJ, Soares J, Isabel B. 1996.Determination of αtocopherol in pork with high intramuscular fat content.Grasas Aceites 47, 331334.Rey AI, LopezBote CJ, Sanz Arias R. 1997.Effect of extensive feeding on α tocopherol concentration and oxidative stability of muscle microsomes from Iberian pigs.Anim.Sci.65, 515520.Rey AI, Isabel B, Cava R, LópezBote CJ. 1998.Dietary acorns provide a source of gammatocopherol to pigs raised extensively.Can.J. Anim.Sci.78, 441443.Rey AI, Daza A, LopezCarrasco C, LopezBote CJ. 2006a.Feeding Iberian pigs with acorns and grass in either freerange or confinement affects the carcass characteristics and fatty acids and tocopherols accumulation in longissimus dorsi muscle and backfat.Meat Sci.73, 6674.Rey A, Daza A, LópezCarrasco C, LópezBote CJ. 2006b.Quantitative study of the alpha and gamma tocopherols accumulation in muscle and backfat from Iberian pigs kept freerange as affected by time of freerange feeding or weight gain.Anim.Sci.82: 901 908.Rey AI, LópezBote, CJ, Daza A, Lauridsen C. 2010.Accumulation and evolution of tocopherols in drycured hams from Iberian pigs as affected by their feeding and rearing system.Food Chem.123, 11701175.SAS.1999.Statistics.In: SAS Use'r guide.Statistical Analysis System Inst.Inc., Cary, NC. A