Effect of the diet on the levels of carotenoids, retinol and alpha-tocopherol in biological fluids and tissues from animals of different species

“Effect of the diet on the levels of carotenoids, retinol and alpha-tocopherol in biological fluids and tissues from animals of different species” Rocío Álvarez Alonso AGRADECIMIENTOS Con esfuerzo y dedicación todo llega y este trabajo es buena prueba de ello, pero no hubiese llegado hasta aquí sin la ayuda y el apoyo de las personas que han creído en mí. Por eso quiero daros las gracias. A mis tres directores por darme la oportunidad de cumplir un sueño y por guiarme en el mundo de la investigación. A la Dra. Mª Jesús Alcalde Aldea por introducirme en el mundo de la producción animal, por tus enseñanzas, por tu confianza en mí (incluso cuando yo la perdía) y por abrirme un mundo de posibilidades. A la Dra. Isabel M. Vicario Romero y al Dr. Antonio J. Meléndez Martínez por introducirme en el increíble mundo de los carotenoides y los retinoides, por vuestras enseñanzas y vuestro apoyo. Gracias a los tres por ser una guía profesional y personal. Gracias por estar ahí. A José Miguel Mejías, gracias de corazón, por tu ayuda constante, tu inagotable paciencia, tus enseñanzas y tu confianza en mí. A todas las personas que me han ayudado desinteresadamente en la toma de muestras a lo largo de estos cuatro años: veterinarios y operarios de los mataderos MercaSevilla, OVISO y CordeSur. A los ganaderos de Acriflor y la Asociación de Payoya. A Ecolactis ® . A TodoMuta por su maravillosa aportación a mi trabajo. A mis niñas y mi niño del “despachito” de la ETSIA: Auro, Ana, Oli, Ester, Rami, Celia, Paloma, Mª José, Mª Luz y Silvia. Por enseñarme que “Sí se puede”, por apoyarme siempre y por hacer fáciles las infinitas horas en la escuela. Al Dr. Francisco José Heredia Mira por introducirme en el mundo de la investigación. A todo el grupo Color y Calidad de Alimentos por ser excelentes compañeros, por su apoyo y por hacerme sentir siempre querida. Muchísimas gracias a Carla y a José Miguel por su paciencia y por estar siempre dispuestos a ayudarme en el laboratorio y a responder a mis preguntas. A Fran y a Julio por arrancarme siempre una sonrisa, incluso cuando no es fácil. A mis compañeras de fatigas con el HPLC, con las extracciones, con los picos infinitos…mi querido “Rincón del Carotenoide” (Ana y Paula), gracias por estar siempre ahí. A Ana por enseñarme a ser más fuerte, a no rendirme nunca y a enfrentarme al mundo con una sonrisa. A Gloria porque “volere è potere”. Thank you to Dr. Alessandro Priolo and Dr. Frank Monahan from Università di Catania and University College of Dublin, for welcoming me into their groups during my two stays abroad. Grazie a “mia sorella” Leti, a Claudia ed a Wanda per la vostra amicizia quando ero lontano da casa, grazie a voi non sono stata sola mai. Thank you Aggeliki for make me believe that “it could be worst” and keep going. A mi tío Santi por guiarme cuando estaba perdida, por ser la referencia a seguir. A Emi porque siempre tiene un buen consejo para darme. A toda mi familia por hacer de mí la persona que soy hoy, por su incondicional apoyo y, sobre todo, por su cariño. A mis padres porque sin vosotros no lo hubiese conseguido. A Edu por ser mucho más que mi hermano. A los Beiztegui, los Mellado y los Villalonga porque sois mi segunda familia. A “mis papis”, Juan y Cris, porque siempre estáis ahí. A “mi bichito” por traer felicidad. A mis amigos, TODOS (sabéis quiénes sois), por vuestros ánimos constantes, por vuestra alegría al celebrar los logros, porque SIEMPRE estáis conmigo. A mi abuelo, a Marisol, a la Lita, al abuelo Pepe y a la abuela porque parte de lo que soy y por lo que he conseguido llegar hasta aquí es por vosotros. “Existimos mientras alguien nos recuerda” (C. R. Zafón) Gracias a todos por formar parte de esta maravillosa experiencia en la que he crecido profesional y personalmente. GRACIAS Para la realización de este trabajo se ha contado con el apoyo institucional y económico de los proyectos P08-AGR03784, AGR-7783-2011 (Consejería de Innovación, Ciencia y Empresa, Junta de Andalucía) y RTA2012-00023-C03-02 (Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria, Ministerio de Economía y Competitividad) y una Beca Predoctoral del IV Plan Propio de la Universidad de Sevilla. A mi abuelo A Marisol Caminante no hay camino, se hace camino al andar (Antonio Machado) Index INDEX List of Tables I List of Figures II ABSTRACT IV RESUMEN VIII INTEREST AND JUSTIFICATION 1 INTRODUCTION 7 Publication: “Carotenoids and vitamin A content in biological fluids and tissues of animals as an effect of the diet. A review” 7 OBJETIVES 58 MATERIAL AND METHODS 60 Sampling methodologies 61 Feedstuffs 61 Animal tissues and biological fluids 61 Plasma 61 Adipose tissue and liver 61 Milk 62 Carotenoids and tocopherols extraction in feedstuffs 62 Carotenoids, retinoids and tocopherols extractions in animal biological fluids and tissues 63 Plasma 63 Adipose tissue 64 Liver 64 Milk 65 High Performance Liquid Chromatography (HPLC) analysis 66 Carotenoids identification and quantification 68 Liquid chromatography-electrospray ionization ion trap/time of flight mass spectrometry (HPLC-ESI/TOF-MS) analysis 70 Index Colour measurement in animal adipose tissue 72 RESULTS AND DISCUSSION 75 Chapter 1. Fatty Acid profile and vitamin A in goat milk: influence of the animal’s diet 76 Antecedentes 76 Objetivos 77 Diseño experimental 77 Resultados 78 Conclusiones 79 Publication: “Fatty Acid profile and vitamin A in goat milk: influence of the animal’s diet” 81 Chapter 2. Effect of pasture and concentrate diets on concentrations of carotenoids, vitamin A and vitamin E in plasma and adipose tissue of lambs 103 Antecedentes 103 Objetivos 104 Diseño experimental 105 Resultados 106 Conclusiones 107 Publication: “Effect of pasture and concentrate diets on concentrations of carotenoids, vitamin A and vitamin E in plasma and adipose tissue of lambs” 108 Chapter 3. Effect of different carotenoid-containing diets on the vitamin A levels and colour parameters in Iberian pigs’ tissues: utility as biomarkers of traceability 115 Antecedentes 115 Objetivos 116 Diseño experimental 116 Resultados 117 Conclusiones 118 Publication: “Effect of different carotenoid-containing diets on the vitamin A levels and colour parameters in Iberian pigs’ tissues: utility as biomarkers of traceability” 119 VI Abstract differentiate stall-fed from pasture-fed lambs. A good discrimination (100% of the animals accurately classified into their group) of lambs according to their diet (pasture vs. concentrate) was obtained when a combination of AVI (integral of the reflectance translated spectrum), retinol and α-tocopherol levels in perirenal fat was used. Thus, these parameters are proposed as a useful tool for feeding traceability studies in sheep livestock. Iberian pigs fed on two rearing systems: Montanera (extensive system where animals are fed on pasture and acorns) and Cebo (intensive system where animals are fed on concentrate) were compared. Both systems showed a different carotenoid profile (P<0.01). It was observed that plasma retinol (1.07 ± 0.69 µg/mL in pigs reared in extensive and 0.68 ± 0.38 µg/mL in pigs reared in intensive) and perirenal fat retinol content (4.14 ± 1.47 µg/g in pigs reared in extensive and 3.69 ± 0.77 µg/g in pigs reared in intensive) in these animals is not significantly affected by the different diets tested. However, retinoids content in liver, as well as L* and h ab values in perirenal fat appeared as useful variables to differentiate these animals according to their feeding system (92.9% and 78.6% of the animals correctly classified, respectively). Finally, the level of carotenoids, retinoids and α-tocopherol were assessed in horse and cattle tissues (plasma, milk, perirenal fat and liver). The animals were fed on different systems (outdoors system based on pasture and indoors based on concentrate). Cattle was selected as a reference model to which compare horse with. In this way, species-specific differences were observed in the accumulation of carotenoids, retinoids and α-tocopherol between equine and cattle livestock. In addition, horse and cattle seemed to have different metabolism for retinoids in liver, such that a higher (P<0.001, P<0.01) accumulation was found in the foal under our experimental conditions. RESUMEN IX Resumen RESUMEN En la presente tesis doctoral hemos estudiado el efecto de diferentes dieras en los niveles de carotenoides, retinoides y α-tocoferol en tejidos biológicos que diferentes especies ganaderas (caprino, ovino, porcino, vacuno y equino). Tanto los carotenoides como las vitaminas liposolubles, entre ellas el retinol y el tocoferol, pueden conferir propiedades saludables a los productos animales y, así, mejorar el punto de vista de los consumidores respecto a ellos. Estos compuestos se encuentran presentes en las dietas de los animales críados en sistemas extensivos, los cuales están más valorados por los consumidores que los intensivos. Además, los mamíferos pueden absorber y metabolizar algunos carotenoides, siendo la dieta uno de los factores más importantes que afecta al contenido de carotenoides, retinoides y tocoferol en tejidos animales. Aunque no todas las especies de animales absorben y metabolizan estos compuestos de igual manera ya que existen diferencias específicas entre especies ganaderas al respecto. Se analizó el efecto de tres dietas distintas respecto a su contenido en carotenoides (pasto, concentrado y una dieta enriquecida con pulpa de naranja) en los niveles de retinol y carotenoides en plasma y leche de ganado caprino. Se observaron niveles más altos (P<0,001) de reinol en la leche de cabras con una dieta rica en carotenoides (dieta basada en pastos o enriquecida con pulpa de naranja) que en la de aquellas con una dieta en estublación a base de concentrado (1,18 ± 0,57, 1,02 ± 0,43 y 0,23 ± 0,18 µg/mL, respectivamente). En el caso del ganado ovino los tejidos y fluidos biológicos analizados fueron plasma, grasa y leche. Se estimó el efecto de la dieta sobre los niveles de carotenoides, retinol y α-tocoferol en relación al sistema de producción. Los niveles de luteína (4,95 ± 0,71 µg/100 mL) y de β-caroteno (2,97 ± 0,68 µg/100 mL) en plasma se mostraron como una herramienta útil para diferenciar a los corderos alimentados exclusivamente con pasto de X Resumen aquellos alimentados con otras dietas, como alimentos concentrados o leche (corderos lechales). También se estudió la posibilidad de combinar los niveles de carotenoides, retinoides y α-tocoferol en tejido adiposo con las medidas de los parámetros de color de dicho tejido para diferenciar los corderos estabulados con una dieta a base de pienso de los corderos alimentados exclusivamente con pasto. Utilizando una combinación de la medida de AVI (integral del espectro de reflectancia trasladado) junto con los niveles de retinol y α-tocoferol en grasa perirrenal, se discriminaron el 100% de los corderos según sus dieta (pastp vs. concentrado). Se compararon dos grupos de cerdo ibérico con dos sistemas de producción disitntos: Montanera (sistema extensivo en el que los cerdos son alimentados con pasto y bellotas) y Cebo (sistema intensivo en el que los animales son alimentados con concentrados). Ambos sistemas de cría presentaron distintos niveles de carotenoides (P<0,01) en la alimentación de los animales. Sin embargo, los resultados obtenidos demuestran que los niveles de retinol en plasma (1,07 ± 0,69 µg/mL y 0,68 ± 0,38 µg/mL, montanera y cebo respectivamente) y grasa perirrenal (4,14 ± 1,47 µg/g y 3,69 ± 0,77 µg/g, montanera y cebo respectivamente) en cerdos ibéricos no se ven afectados por las dos dietas estudiadas. Sin embargo, tanto el contenido en retinoides en le hígado vomo los valores de L* y h ab en grasa perirrenal resultaron útiles para diferenciar estso animales según sus sistema de cría (92,9% y 78,6% de los animales correctamente clasificados, respectivamente). Por último, se estudiaron los niveles de carotenoides, retinoides y α-tocoferol en ganado bovino y equino (plasma, leche, grasa perirrenal e hígado). Estos animales fueron críados en dos sistemas diferentes: sistema extensivo con una dieta en base al pastoreo y sistema intensivo con una dieta en base a concentrado. El ganado bovino se utilizó como un modelo de referencia con el que comparar al ganado equino. De esta forma, se XI Resumen evidenciaron diferencias específicas por especie en la acumulación de carotenoides, retinoides y α-tocoferol entre el ganado bovino y el equino. Además, el caballo y el vacuno parecen tener un metabolismo disinto para los retinoides en el hígado, ya que se observó un almacenamiento mayor (P<0,001, P<0,01) en potros que en terneros, bajo las mismas condicones. INTEREST AND JUSTIFICATION 2 Interest and Justification INTEREST AND JUSTIFICATION Nowadays, there is an increasing interest of consumers about foods with additional health promoting functions (Olmedilla-Alonso et al., 2013). In this sense, consumers are focusing on the healthy properties of products from animals fed on pasture, which are considered as more suitable with a “green-image” of the animal production (Dian et al., 2007a; Nozière et al., 2006b; Prache et al., 2003b; Sheath et al., 2001). For these reasons, research on diet authentication in herbivore products is being conducted and efforts have recently been made to develop analytical tools to quantify specific compounds in animal´s tissues that can act as tracers of the conditions of production (Dian et al., 2007a; Engel et al., 2007). These compounds are transferred from the feed to the final product or transformed by the animal metabolism Thus, they could be used to some extent to authenticate the diet of the animals and, in this way, reassure consumers about the origin of the products and how they are produced (Dian et al., 2007b; Sheath et al., 2001). Carotenoids and fat-soluble vitamins, such as vitamins A (retinol) and E (α-tocopherol) are examples of these compounds. Although species-specific differences exist among livestock species, the diet is one of the main factors that affect the content of these compounds in animal tissues (Prache & Theriez, 1999; Schweigert, 1998). Considering all these facts, it is clear that a deeper knowledge of the factors governing the presence of these health-promoting compounds in biological fluids and tissues of different livestock species with different rearing systems is interesting. European sheep and goat production represents, important economic, environmental and sociological issues, mainly in Mediterranean areas (De Rancourt et al., 2006). However, these productions have low profitability in these countries (Niżnikowski et al., 3 Interest and Justification 2006), so the sector is in a critical economic situation and the production has decreased consistently over the last ten years in Europe (Figure 1). Figure 1. Millions of goat and sheep heads in Europe between 2003 and 2013 Graphics made from FAO (2014) data One of the main reasons behind this problem is the decrease in the consumption due to limited availability of its products together with a relative high price, linked to high production costs (Alcalde et al., 2013; De Rancourt et al., 2006). Thus, to ensure a sustainable future for the small ruminant sector it is desirable to improve the consumption of its products. One strategy would be to add value to the milk, meat, etc. by increasing the content of health-promoting compounds (Nozière et al., 2006b; Prache & Theriez, 1999). Altogether, carotenoids, vitamins A and E, are involved in the nutritional and sensory properties of goat products, mainly in those from animals fed on pasture (Sauvant et al., 2012). Therefore, the study of biomarkers of the feeding system, such as carotenoids and fat-soluble vitamins in small ruminants products could be useful to find ways of conferring them added value related to health benefits and emphasizing their origin. 4 Interest and Justification Figure 2. Florida goat (left) and Segureña sheep (right) reared in an extensive system in Southwest of Spain On the other hand, the Iberian pig breed is one of the most important Mediterranean swine type, both in population size and economic importance (Juárez et al., 2009). According to the new Spanish legislation related to the quality standard for Iberian pig products (BOE, 2014) three different categories are considered based on feeding and rearing system. The highest quality products are related to the typical outdoor rearing system with a nutritional strategy based on acorns and grass (Montanera) (Gaspar et al., 2007). However, the controversy about the different feeding systems in this pig breed and the corresponding quality in derived products is still an unresolved problem. The traditional analysis of fatty acids has been proved to fail in this purpose. In this sense, other minor compounds related to the diet, like provitamin A carotenoids could be of interest to this aim. 5 Interest and Justification Figure 3. Iberian breed pig fed on Montanera rearing system in Southwestern Spain Finally, the situation of equine production in Europe, as well as the economic crisis in most of the South-European countries are deriving in an increment in the slaughters due to the inability of producers to assume the high production costs (Figure 4). Furthermore, mare’s milk production is gaining importance in European countries because of its composition, since it can be a profitable alternative to cow’s milk especially among people with cow’s allergy. In addition, it is being used for cosmetic purposes (Markiewicz-Kęszycka et al., 2014). Figure 4. Slaughtered equines (number of heads) in Spain in the last ten years Graphics made from FAO (2014) data 20000 25000 30000 35000 40000 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 12 Introduction 13 Introduction 14 Introduction 15 Introduction 16 Introduction 17 Introduction 18 Introduction 19 Introduction 20 Introduction 21 Introduction 28 Introduction 29 Introduction 30 Introduction 31 Introduction 32 Introduction 33 Introduction 34 Introduction 35 Introduction 36 Introduction 37 Introduction 44 Introduction 45 Introduction 46 Introduction 47 Introduction 48 Introduction 49 Introduction 50 Introduction 51 Introduction 52 Introduction 53 Introduction 61 Material and Methods MATERIAL AND METHODS Sampling methodologies Feedstuffs Representative samples of the diets of the different livestock species studied were collected in triplicate. In the case of pasture groups, the sampling areas were randomly selected at the prairie where the animals were reared and samples were collected in spring. The pasture was sampled from the ground using definite quadrates arbitrarily established. Finally, the pasture samples were overall characterized at the genus level considering the major herbaceous species present. They were transported to the laboratory, freeze-dried and stored at -80 ºC until analysis. Animal tissues and biological fluids Plasma Blood from living animals (lactating goats, sheep, cows and mares) was taken from the jugular vein (goats, sheep and mares) and the caudal vein in the case of cows. Blood from sacrificed animals (lambs, pigs, foals and calves) was also taken at the moment of slaughter. Li-Heparin was used as anticoagulant in both cases. The samples were transported to the laboratory at 4 ºC, and centrifuged (1500 g, 10 min, 4 ºC). The plasma was collected and stored at -80 ºC until analysis. Adipose tissue and liver Samples of liver and perirenal fat were also taken from the animals that were slaughtered (Chapters 2, 3 and 4). Sampling was made at the moment of the slaughter, during the dressing of the carcasses. Perirenal area was selected for adipose tissue sampling as, according to Priolo et al. (2002), there may be a greater accumulation of 62 Material and Methods carotenoid pigments in perirenal compared to caudal fat in lambs. The samples were transported to the laboratory and stored at -80ºC until analysis. Milk Two aliquots of both morning and afternoon milking were mixed to obtain a representative sample of each animal. In total, 30 mL of milk from each animal were aseptically collected into sterile vials and immediately stored at -80ºC until analysis. The milk samples were collected the same week of lactation for the animals of the same experiment (goats in chapter 1, sheep in chapter 2 and cows and mares in chapter 4). C arotenoids and tocopherols extraction in feedstuffs A good extraction procedure should recover all the analites of interest in solution, without any change in them (Mínguez-Mosquera, 1997). As carotenoids and tocopherols are very unstable some precautions were taken during analysis in order to avoid their oxidation and loss (Meléndez-Martínez et al., 2004; Mínguez-Mosquera, 1997): - Handling time of the extracts was minimized. - Light exposure and heating were avoid in order to reduce the risk of isomerization and oxidation. - Solvents free of impurities, mainly free of peroxides and acids were used. - Oxygen was removed by displacement with nitrogen or working under vacuum atmosphere. - Antioxidants were added to the solvents. Carotenoids and tocopherols analysis in feedstuffs were carried out according to the methodology described previously (Kean et al., 2012; Pickworth et al., 2012) with some modifications. One gram (g) of sample (pasture, concentrate or other specific diet sample) 63 Material and Methods was freeze-dried. 3 milliliters (mL) of hexane/ethanol mixture (1:1, v/v) were added to the freeze-dried material and centrifuged at 3500 rpm and 4 ºC over 10 minutes (Allegra X-22R Centrifuge, Beckman Coulter, USA). The extraction was repeated until colour depletion of the samples. Organic phases were pooled and saponified with 15% (w/v) ethanolic potassium hydroxide solution overnight at room temperature in the dark and under nitrogen atmosphere. Finally, the organic phase was washed several times with water up to all the potassium hydroxide was removed. After that, the organic phase was collected and dried using a concentrator (Concentrator plus, Eppendorf Research ® , Madrid, Spain). The dry residue was stored at -20 ºC until the moment of the HPLC analysis. Carotenoids, retinoids and tocopherols extraction in animal biological fluids and tissues Plasma A methodology described elsewhere was used with some modifications (Lyan et al., 2001). 2 mL of plasma were deproteinized by adding 2 mL of ethanol. Carotenoids, retinol and tocopherol were extracted twice with 2 mL of hexane with centrifugation (3500 rpm, 10 min, 4ºC; Allegra X-22R Centrifuge, Beckman Coulter, USA). The extracts were pooled and evaporated to dryness with a concentrator (Concentrator plus, Eppendorf Research ® , Madrid, Spain). The dry residue was stored at -20 ºC until the moment of the HPLC analysis. 64 Material and Methods Figure 6. Plasma samples Adipose tissue The extraction was carried out following the methodology described by Dunne et al. (2006) with some modifications. 500 milligrams (mg) of perirenal fat were mixed with 1 mL of 3,5-di-ter-4-butylhydroxytoluene (BHT) ethanolic solution (12% w/v) in order to suppress the oxidation of the analites. 5 mL of ethanolic sodium hydroxide solution (30% w/v) and 5 mL of ethanol were also added. The saponification reaction was carried out overnight at room temperature and in the dark. The next morning water was added to stop saponification and the analytes were extracted with 10 mL of a mixture ether/hexane (2:1 v/v). Samples were centrifuged (3500 rpm, 10 min, 4 ºC; Allegra X-22R Centrifuge, Beckman Coulter, USA) and the upper organic phase collected. The extraction was repeated twice. The organic phases were pooled, washed several times with water, collected and dried (Concentrator plus, Eppendorf Research ® , Madrid, Spain). The dry residue was stored at -20 ºC until the moment of the HPLC analysis. Liver In the case of liver samples, the methodology described by Woodall et al. (1996) was followed with some modifications. 0.1 g of liver was freeze-dried and the dried samples were homogenized in 500 µl of saline solution 85% (w/v). Then, compounds were extracted with 1 mL of dichloromethane by vortexing and centrifugation (3500 rpm, 10 65 Material and Methods min, 4 ºC; Allegra X-22R Centrifuge, Beckman Coulter, USA). Lower organic phase was collected and extraction was repeated until colour depletion of the samples. Saponification was not performed in order to identify and quantify retinyl esters with long-chain fatty acids in liver samples. The organic phases were pooled and dried (Concentrator plus, Eppendorf Research ® , Madrid, Spain). The dry residue was stored at -20 ºC until the moment of the HPLC analysis. Milk The extraction of carotenoids, retinol and tocopherol from milk was carried out considering previous methodologies (Hulshof et al., 2006; Jewell et al., 2004; Nozière et al., 2006b). 1 mL of milk was mixed with 0.5 mL of water, 0.25 mL of ammonia aqueous solution 25% (v/v), 1 mL of ethanol and 1 mL of hexane. The mix was then centrifuged (3500 rpm, 10 min, 4ºC; Allegra X-22R Centrifuge, Beckman Coulter, USA) and the upper organic phase collected. The extraction was repeated until colour exhaustion. The organic phases were pooled and 2 mL of ethanolic sodium hydroxide 15% (w/v) was added. The air of the containers was replaced with nitrogen and the saponification reaction was maintained overnight at room temperature in the dark. Finally, the organic phase was washed several times with water, collected and dried (Concentrator plus, Eppendorf Research ® , Madrid, Spain). The dry residue was stored at -20 ºC until the moment of the HPLC analysis. 66 Material and Methods Figure 7. Scheme of the methodology followed for carotenoid, retinol and tocopherol extraction in milk High Performance Liquid Chromatography (HPLC) analysis The analyses were performed on an Agilent 1100 system (Agilent, Palo Alto, CA, United States) fitted with a photodiode array detector, a quaternary pump, a column temperature control module set at 20 ºC and an auto sampler set to draw 20 µl aliquots from the concentrate extracts. YMC C 18 and YMC C 30 columns were used (5µm, 250 x 4.6 mm) (YMC, Wilmington, NC, USA) were used for the analyses of feedstuffs and animals biological fluids and tissues. The mobile phase consisted on: o Phase A: Methanol (MeOH) (Merck, Ref 1.06018). o Phase B: Methyl-ter-butyl ether (MTBE) (Merck, Ref 1.01845) 67 Material and Methods o Phase C: Water MilliQ (filtration equipment NANOpure Dlamond, Barnsted Inc. Dubuque, IO). The linear gradient used is summarized in table 1. Table 1. Gradient elution used in the HPLC determination of carotenoids, retinoids and tocopherol Time (min) % A % B % C 0 90 5 5 12 95 5 0 25 89 11 0 40 75 25 0 50 40 60 0 56 15 85 0 62 90 5 5 The mobile phase was pumped at 1mL/min and the chromatograms were monitored at 450 nm for carotenoids, at 325 nm for retinoids and at 280 nm for tocopherol. Every day when the analyses were finished the column was washed with MTBE:MeOH (50:50) for 20 minutes. 68 Material and Methods Carotenoids identification and quantification Carotenoids were identified by comparison of their spectroscopic and chromatographic characteristics with those of standards obtained from natural sources by recommended procedures as described elsewhere (Meléndez-Martínez et al., 2009). Table 2 shows the spectroscopic characteristics of the main carotenoids identified in this doctoral thesis. Table 2. Spectroscopic characteristics of main carotenoids identified in the samples analyzed in this study Carotenoid Absorption maxima (nm) β-carotene 452 472 β-cryptoxanthin 446 471 Lutein 424, 444, 472 Zeaxanthin 424 450, 476 Anteraxanthin 421 446 473 Violaxanthin 416, 440, 470 Carotenoids in solution follow the Beer-Lambert law, i.e. their absorbance is directly proportional to the concentration, thus their quantification can be performed by relating the spectrophotometry absorbance with a certain wavelength considering an absorption coefficient. The specific absorption coefficient (A) is defined as the theoretical absorbance of a solution with a concentration 1% (w/v) in a cuvette of 1 cm path length. Besides, the molar absorption coefficient (ε) is defined as the theoretical absorbance of a 69 Material and Methods solution 1 molar concentrated. Both coefficients are related according to the following formula: Where ε: molar absorption coefficient A: specific absorption coefficient MM: molecular mass Table 3, summarizes the molar absorption coefficients used for carotenoids quantification. Table 3. Molar absorption coefficients of the main carotenoids. EtOH: ethanol; Hx: hexane Carotenoids ε λ (nm) Solvent β-carotene 140400 450 EtOH β-cryptoxanthin 135700 450 Hx Lutein 144800 445 EtOH Zeaxanthin 140900 450 EtOH Anteraxanthin 137200 446 EtOH Violaxanthin 153000 440 EtOH Source: Britton et al. (1995) Carotenoids quantification was performed by using calibration curves of standards solutions. The concentrations were calculated according to the following formula. C (µg/µl) = [b + (m * A)] * d Where C: concentration b: intercept of the calibration curve ε = A x MM 10 70 Material and Methods m: slope of the calibration curve of the standard X d: factor considering both analyzed and injection volumes Liquid chromatography-electrospray ionization ion trap/time of flight mass spectrometry (HPLC-ESI/TOF-MS) analysis Identification of retinyl esters with long-chain fatty acids was done by HPLCESI/TOF-MS. The chromatographic separations were carried out using a Dionex Ultimate 3000RS U-HPLC system (Thermo Fisher Scientific, Waltham, MA, USA). The HPLC conditions were the same as described above. An YMC C 30 column was used (5µm, 250 x 4.6 mm) (YMC, Wilmington, NC, USA). A split post-column of 0.4 mL/min was introduced directly on the mass spectrometer electrospray ion source. Mass spectrometry was carried out on a micrOTOF-QII High Resolution Time-of-Flight mass spectrometer (UHR-TOF) with Q-TOF geometry (Bruker Daltonics, Bremen, Germany) equipped with an electrospray ionization (ESI) interface. The instrument was operated in positive ion mode using a scan range from m/z 50 to 1200. Mass spectra were acquired in MS fullscan. The instrument control was performed using Bruker Daltonics HyStar 3.2. Preliminary identification of the retinyl esters was based on accurate mass, isotopic pattern and fragmentation profile of the molecular ion. Figure 8 and Table 4 contain the high accuracy measurements obtained from the HPLC and ESI mass spectra and the postprocessing routine applied with the software. 77 Chapter 1 OBJETIVOS El principal objetivo de este estudio fue evaluar el efecto de tres dietas diferentes; pasto, dieta enriquecida con pulpa de naranja y concentrado; en: - Los niveles de carotenoides y retinol en la leche de cabra. - El perfil de ácidos grasos de la leche de cabra. - Los niveles de carotenoides y retinol en plasma caprino. DISEÑO EXPERIMENTAL Se dividieron cuarenta y cinco cabras de la raza Florida en tres grupos según su dieta: el Grupo 1 (G1) eran cabras estabuladas con una dieta basada en alimento concentrado; los animales del Grupo 2 (G2) recibieron una dieta compuesta por alimento concentrado mezclado (TMR, en inglés Total Mixed Ration) y un suplemento de 20% de pulpa de naranja fresca; las cabras del Grupo 3 (G3) fueron alimentadas con pasto. Se recogieron muestras de alimento, plasma y leche de cada grupo. La extracción de carotenoides y retinol en los alimentos se realizó de acuerdo a la metodología descrita en la bibliografía (Kean et al., 2007; Panfili et al., 2004). Para las extracciones en plasma se siguió la metodología descrita por Lyan et al. (2001). Por último, Por último, las extracciones en leche se hicieron considerando diferentes métodos descritos por otros autores para leche de diferentes especies (Hulshof et al., 2006; Jewell et al., 2004; Nozière et al., 2006b). El contenido en carotenoides y retinol de todas las muestras fue analizado por HPLC. Además, la separación y cuantificación de los ésteres de metilo de los ácidos grasos (AG) se hizo con cromatografía de gases de acuerdo a la metodología descrita por Sukhija & Palmquist (1998). 78 Chapter 1 RESULTADOS Se encontraron diferencias cuali y cuantitativas en el perfil de carotenoides de las tres dietas estudiadas. También se encontraron diferencias en el perfil de ácidos grasos de las tres dietas. Respecto a las muestras de los animales, en leche sólo se encontró retinol mientras que en plasma se detectó retinol y luteína. Los niveles más altos (P<0.001) de retinol en leche se detectaron para los animales alimentados con pasto y aquéllos que recibieron un suplemento de pulpa de naranja con la dieta. Por lo tanto, se podría afirmar que la leche de cabra con una dieta rica en carotenoides, es decir, aquella basada en pasto o con suplemento de pulpa de naranja, podría ser útil desde un punto de vista nutricional, así como de cara al consumidor, ya que mostró niveles de retinol más altos que la leche de cabra con una dieta basada en alimentos concentrados. Además, el único carotenoide detectado en plasma fue la luteína, la cual fue detectada en los animales alimentados con pasto (G3) y concentrado (G1), mientras que no fue detectada en los animales que recibieron la suplementación con pulpa de naranja. Por último, se observaron diferencias significativas (P <0.05) en los niveles de retinol en plasma entre los tres grupos de animales, apareciendo los niveles más altos de retinol circulante en el grupo de animales alimentados con concentrado. Con respecto al perfil de AG en la leche, los niveles de la fracción n-3 fueron mayores (P <0,001) en los animales con una dieta basada en pastos que en los otros dos grupos. Sin embargo, la fracción de CLA no mostró diferencias significativas entre los grupos. Además, los niveles de 9-cis CLA y 11-trans (RA) resultaron más altos (P <0,001) en la leche de las cabras alimentadas en pastoreo que en los otros dos grupos. Por otra parte, no se encontraron diferencias significativas en el contenido de AGPI entre G1 (concentrado) y G3 (pasto), pero la cantidad de estos compuestos fue mayor (P <0,001) 79 Chapter 1 en estos dos grupos en comparación con el contenido en el G2 (suplementación con pulpa de naranja). Finalmente, la fracción de Ácidos Grasos Saturados (AGS) fue significativamente mayor (P <0,001) en la leche de los animales que recibieron suplementación de pulpa de naranja (G2) que en los otros dos grupos (G1 y G3). Mediante un análisis discriminante se evaluó el efecto de la dieta sobre el nivel de retinol en la leche. El 100% de las cabras con una dieta a base de concentrado (G1) fueron correctamente clasificadas en su grupo según su alimentación, mientras que en los grupos de animales alimentados con pasto (G3) y TMR suplementado con pulpa de naranja (G2) sólo la mitad, aproximadamente, de los animales (53,8% y 46,7%, respectivamente) fueron clasificados correctamente. De acuerdo con estos resultados, se podría señalar el potencial de este suplemento en la dieta delas cabras para lograr en la leche niveles de retinol similares a los de la leche de animales alimentados con pasto, que es considerada tradicionalmente como de "calidad superior". Se realizó un segundo análisis discriminante considerando las fracciones de SFA y n-6 / n-3 en la leche, en el cual el 100% de las cabras fueron clasificados correctamente de acuerdo con su dieta. Confirmando el efecto de la dieta del animal en el perfil de AG de la leche. CONCLUSIONES Las principales conclusiones que se extraen de los resultados obtenidos en este estudio son las siguientes: 1. Las cabras alimentadas con una dieta rica en carotenoides (pasto o TMR más un suplemento de pulpa de naranja) producen leche con mayor contenido en retinol que las cabras con una dieta basada en alimentos concentrados. 80 Chapter 1 2. El suplemento de pulpa de naranja en la dieta de las cabras alimentadas con TMR tuvo un efecto beneficioso en el contenido de vitamina A en la leche pero no en el perfil de ácidos grasos. 3. La leche de las cabras alimentadas con pasto presentó un perfil de ácidos grasos “saludable”, lo cual es importante desde un punto de vista nutricional. 4. Son necesarios más estudios para establecer la capacidad de las cabras para acumular carotenoides de la dieta. 81 Chapter 1 Fatty Acid profile and vitamin A in goat milk: influence of the animal’s diet R. Álvarez 1 , C.M. Stinco 2 , A.J. Meléndez-Martínez 2 , I.M. Vicario 2 , M.J. Alcalde 1* 1 Dept. Agricultural and Forestry Science, Universidad de Sevilla of Seville, Ctra. Utrera km.1, 41013, Seville, Spain 2 Food Colour & Quality Laboratory, Dept. of Food Science and Nutrition. Universidad de Sevilla, C/ Profesor García González 2, 41012-Sevilla, Spain *Corresponding author: Animal Production Area. Dept. Agricultural and Forestry Science, Universidad de Sevilla, School of Agricultural Engineering, 41013-Sevilla, Spain e-mail: [email protected] 82 Chapter 1 ABSTRACT The aim of this report was to assess the effect of different diets (concentrate, pasture and diet enriched with orange pulp) on carotenoids and retinol levels in goat milk, as well as on the Fatty Acid (FA) profile. Carotenoids and retinol levels in goat plasma were also studied. Higher retinol levels (P<0.001) were found in milk from goats fed an enriched orange pulp diet and grazing goats (1.02 ± 0.43 and 1.18 ± 0.51 µg/mL, respectively) than in milk from goats fed on concentrate (0.23 ± 0.18 µg/mL). Furthermore, n-3 FA fraction was higher (P<0.001) in grazing goats than in the other two feeding system . PUFA and MUFA fractions were higher (P<0.001) in milk from animals fed on concentrate or pasture than in those fed with orange pulp supplements. Both pasture and orange pulp supplemented diets are interesting from the nutritional point of view, as they would improve vitamin A content in milks conferring an added value to dairy goat products. Additionally, milk from goats fed on pasture showed a better and healthier FA profile but the supplementation of orange pulp did not show any improvement on the FA profile in milk. INTRODUCTION Billions of people around the world consume milk and dairy products every day so dairy products suppose a vital source of nutrition (Muchlhoff et al., 2013). Goat milk has an exceptional nutritional quality and it is well tolerated by individuals sensitive and allergic to cow milk and has a benefit effect on health and a high digestibility (Kondyli et al., 2007). Besides, it has been reported (Larsen et al., 2012) that the type and amount of herbage consumed by the animals have an impact on milk composition, so that high levels of grass compared with concentrates based on cereals result in higher milk concentrations 83 Chapter 1 of polyunsaturated fatty acids (PUFA) and conjugated linoleic acid (CLA), as well as carotenoids and fat-soluble vitamins such as retinol or tocopherols. Additionally, it is known (Delgado-Pertiñez et al., 2013) that milk from goats fed on concentrates is rich in Saturated Fatty Acids (SFA) and poor in PUFA. Finally, concentrate feedstuffs, which are usually composed by cereals, tends to present very low levels of carotenoids and fatsoluble vitamins since their oxidation may occur among the manufacture of these concentrate rations (Nozière et al., 2006; Dunne et al., 2009). Moreover, it is known (Agócs et al., 2007) that orange pulp and peel are important sources of carotenoids, including the pro-vitamin A carotenoid β-cryptoxanthin. However, during orange juice production, only around the half of the fresh orange weight is transformed into juice (Rezzadori et al., 2012), producing great amount of carotenoid-rich waste; including pulp, peel or seeds. For this reason, an alternative use for these by-products of the orange juice industry, such as to produce ingredients for animal feed has been proposed (Rezzadori et al., 2012). So that, it could help to add nutritional quality to the animal products and, in this way, present advantages for both orange juice industry and animal production sector. Finally, several information is available on the carotenoids levels in plasma of sheep, lambs and cattle, but few studies have been conducted to explore the profile of these fatsoluble pigments in goats (Yang et al., 1992). Besides, Yang et al. (1992) suggested that the three ruminant species have different biochemical and physiological pathways for dealing with carotenoid absorption, deposition and metabolism. Thus it would be interesting to compare the deposition of these compounds in goat tissues. In consequence, the aim of this report was to study the effect of three different carotenoid-containing diets, i.e. pasture, a diet enriched with orange juice by-products and concentrate, on plasma and milk carotenoid and retinol levels in goats. In addition, 84 Chapter 1 the effect of the three types of diets studied on the fatty acid profile of milk was also assessed. MATERIAL AND METHODS Animals and diets Forty-five female goats, of Florida breed from three herds in commercial farms, with different feeding system from Southwest of Spain were selected (3 × 15). Animals in group 1 (G1) were fed in confinement ad libitum with a diet based on a dry concentrate which was composed by soy (42%), barley (14%), maize (16%), beet pulp (10%), wheat (9%) and rapeseed meal (9%). Those in group 2 (G2) were fed, also indoors, with a total mixed ration (TMR) containing Ray Grass, i.e. Lolium spp. (20%), beet pulp (18%), barley (16%), soybean meal (8%), cottonseed (8%), maize (3%) and mineral corrector (7%) and an additional supplement of 20% of orange pulp, in this case the feed was administrate also ad libitum. Group 3 (G3) consisted of grazing goats fed on natural pasture. The paddock where the goats were reared was mainly composed of plants from Poaceae family, like Agrostis spp., Vulpia spp., Poa spp., Bromus spp. and Lollium spp. Besides, species from Fabaceae family were also present, like Trifolium spp., Ornithopus spp., Medicago spp. and Lotus spp. These goats were grazing an average of 8 hours per day. Besides, these animals received supplementation along the first month of lactation because of production necessities but not during the rest of the milking period. For this study, animals were selected in the middle of their third or fourth lactation with a diet adapted to their production needs. Milk samples were collected in the same conditions for the three herds, it means in the twelfth week of lactation. The mean milk production levels of the animals, at that moment, were 2.5 ± 0.4 L/day for goats of G1, 2.2 ± 0.2 L/day for G2 and 1.9 ± 0.3 L/day for animals of G3. Goats had an average weight of 74 ± 2 kg. 85 Chapter 1 Sampling Representative samples of the different diets were taken by triplicate from the farms. In the case of G3, the sampling areas were randomly selected at the prairie where the animals were reared. The pasture was sampled from the ground; to do this three fixed sampling quadrates (0.5 m by 0.5 m) were randomly established. Finally, pasture samples were characterised overall according to the majority herbaceous species present at the gender level. After collecting, all food samples were transported to the laboratory and, then, they were lyophilized and stored at -80 ºC until analysis. Six millilitres (mL) of blood were taken from the jugular vein of each animal using Li- Heparin as anticoagulant. The blood samples were centrifuged (1500 g, 10 min, 4ºC) and the plasma collected and stored at -80ºC until analysis. What is more, for each group, two aliquots of both morning and afternoon milking were mixed to obtain a representative sample (Delgado-Pertiñez et al., 2009). In total, 30 mL of milk were collected, aseptically into sterile vials, from each goat and immediately stored at -80 ºC until laboratory analysis. Both plasma and milks samples were taken in the twelfth week of lactation. Carotenoids and retinol extractions Feedstuffs The food samples were analysed by triplicate according to the methodology described elsewhere (Panfili et al., 2004; Kean et al., 2007) with some modifications. 3 grams (g) of sample were lyophilized previously to the analysis. The extractions were carried out on the lyophilized extracts and 15 mL of a mixture hexane/ethanol (1:1, v/v) twice. The organic phases obtained were saponified with 15% ethanolic potassium hydroxide (w/v) 86 Chapter 1 overnight at room temperature in the dark and under a nitrogen atmosphere in order to avoid the degradation of carotenoids. Finally the organic phase was washed several times with the same volume of water than organic phase, using a pH indicator to know when all the potassium hydroxide was removed. After that, the organic phase was collected and dried using a concentrator (Concentrator plus, Eppendorf Research ® , Madrid, Spain). The residue was dissolved in 100 µl of ethyl acetate for HPLC analysis. Milk The extraction of carotenoids and retinol from milk was carried out considering the methodology described by other authors in milk samples from different species (Jewell et al., 2004; Hulshof et al., 2006; Nozière et al., 2006b). 15 mL of milk were mixed with 7 mL of water, 3.5 mL of ammonia solution 25% (w/v), 15 mL of ethanol and 15 mL of hexane. Samples were then centrifuged (1500 g, 10 min, 4 ºC) and the upper organic phase collected and extracted twice. The organic phases were pooled and 15 mL of sodium hydroxide in ethanol 15% (w/v) were added. The air of the containers was replaced with nitrogen and the saponification reaction was maintained overnight at room temperature in the dark. Finally, the organic phase was washed several times with water, collected and dried using a concentrator (Concentrator plus, Eppendorf Research ® , Madrid, Spain). The extract was stored at -20 ºC under a nitrogen atmosphere until the moment of the analysis. The residue was dissolved in 35 µl of ethyl acetate for HPLC analysis. Plasma The methodology described by Lyan et al. (2001) was used with some modifications. 3 mL of plasma diluted with the same volume of distilled water were deproteinized by 93 Chapter 1 In fact, β-carotene was not either found in plasma, which could explain its absence in the milk, in agreement with previous studies (Lucas et al., 2008b). The highest retinol levels in milk were detected in grazing animals and in those fed with TMR plus orange pulp, being its concentration fivefold (P<0.001) higher than in milk from animals fed on concentrate. Previous studies (Fedele et al., 2004: Morand-Fehr et al., 2007) reported that retinol content in milk from grazing goat was higher (often significantly) than in milk from animals on an intensive regime. However, to our knowledge, supplementing the diet with orange pulp had not been evaluated previously. It has also been reported elsewhere (Lucas et al., 2008a; Yang et al., 2010) that the amount of retinol in milk is directly dependent on the level of provitamin A (in particular β-carotene) in the diet, as it can be expected beforehand. This fact is also observed in this study as the highest amount of retinol in milk were detected in the goats from the group receiving pasture and those receiving TMR plus orange pulp. Therefore, it could be claimed that milk from goats with a diet rich in carotenoids, i.e. pasture or supplemented with orange pulp, could be useful from a nutritional and consumer’s point of view, since it showed higher retinol levels than milk from goats with a diet based on concentrate. 94 Chapter 1 Table 7. Lutein and retinol in goat milk and plasma (µg/mL ± SD) in the three groups of animals. G1: goats fed on concentrate, G2: goats fed on TMR (Total Mixed Ration)+ orange pulp and G3: goats fed on pasture. Variable G1 G2 G3 Sig. Plasma Lutein 0.01 ± 0.00 n.d. 0.04 ± 0.03 * Retinol 1.18 b ± 0.84 0.64 ab ± 0.37 0.22 a ± 0.09 * Milk Lutein n.d. n.d. n.d. - Retinol 0.23 a ± 0.18 1.02 b ± 0.43 1.18 b ± 0.57 *** Sig.: significant differences. *P<0.05, ***P<0.001. Significant differences (P<0.05) within a row are indicated by superscripts (a, b). n.d.: not detected. On the other hand, the only carotenoid detected in plasma was lutein (Table 3). This observation is in accordance with previous studies in sheep and goats (Yang et al., 1992; Nozière et al., 2006a). Nevertheless, lutein was not either detected in plasma from goats fed on TMR + orange pulp (G2), in spite of lutein was present in their diet (Table 5). In principle, it could be thought that the absence of lutein in these samples could be due to the absence of this compound in the orange pulp but it has to be considered that the percentage of pulp in the diet of these animals was only 20%. In this regard, more studies are needed to determine the cause of the lack of lutein in the plasma of these animals. Finally, significant differences (P<0.05) in plasma retinol levels among the three groups of animals were observed, showing the highest circulating retinol levels those goats fed on concentrate (Table 7). According to the results discussed above, it can be stated that grazing animals showed the lowest amount of retinol in plasma but the highest in milk. It has been hypothesized by previous authors (Schweigert et al., 1990; Calderón et al., 2007) that, in cattle the conversion of β-carotene into retinol by the 15,15’-dioxygenase is also produced in the 95 Chapter 1 mammary gland, causing a rise in the retinol concentration in milk and, as a consequence, a decrease in plasma. These authors observed that the parenteral administration of βcarotene was followed by an increase of retinol concentration in milk but not in plasma. Besides, it can be noticed that, according to our results, retinol levels in milk from animals fed on pasture and those fed on TMR plus orange pulp are not different (Table 7), indicating that supplementing goat diet with orange pulp makes no difference in retinol or other carotenoid levels. Therefore, it could be pointed out the potential of supplementing the diet with orange pulp in order to achieve retinol levels similar to those in the milk from animals feed on grazing, which is traditionally considered as “higher quality”. Fatty acid profile in milk 15-20% of the total fat intake of humans, 25% of which is saturated fat, is derived from dairy products (Elgersma et al., 2006) and the FA composition depends to a large extent on the diet composition fed to the animals (Silanikove et al., 2010). For this reason, current research is focusing on n-3 FA and Conjugated Linoleic Acid (CLA) which benefits human health (Mancilla-Leytón et al., 2013). In this sense n-3 fraction level in milk was higher (P<0.001) in animals with a diet based on pasture (G3) than in the other two groups (Table 4) which is in accordance with previous studies (D’Urso et al., 2008; Mancilla-Leytón et al., 2013). However, the CLA fraction did not show significant differences (Table 8) between groups, i.e. the feeding system did not influence that fraction in milk, as reported previously by Mancilla-Leytón et al. (2013). Besides, CLA cis-9, trans-11 (RA) has been established as the most predominant CLA compound in milk (Lock et al., 2004), which is in accordance with the data obtained in this study (Table 8), being its amount higher (P<0.001) in milk from goats fed on pasture than in milk from 96 Chapter 1 the other two feeding groups. CLA may be beneficial to human health especially for its anticarcinogenic activity and food products coming from ruminants are considered as the main source of CLA in human diets (Lock et al., 2004). In addition, Mancilla-Leytón et al. (2013) also reported in their trial a similar content of PUFA in milk between grazing and indoor-fed with concentrate goats, which also agree with our data since no significant differences were found (Table 8) between G1 and G3 in this sense Besides, under our trial conditions, the n-6/n-3 FA ratio was in accordance with current human health recommendations, 2.0-2.5 (MacRae et al., 2005), (Table 4), for all feeding systems studied. Considering all these data we agree with Silanikove et al. (2010) who affirmed that milk from goats fed on pasture may present an overlooked “treasure trove” with respect to its health promoting lipid profile and its beneficial consequences to human health. Furthermore, it has been reported (Silanikove et al., 2010) that goat fat milk contains a high proportion of medium-chain FA, i.e. caproic (C6: 0), caprylic (C8: 0) and capric (C10: 0), which are partly responsible for the characteristic “goaty” odour of goat milk. These compounds were detected in all milk studied in this report (Table 4) without significant differences in their amount among the three groups, except for capric acid which was present in higher (P<0.001) amount in milk from goats fed on TMR + orange pulp supplementation (G2) than in the other two groups. Finally, in Table 8 it can be observed that SFA fraction, as well as SFA/PUFA and SFA/UFA ration in milk from G2 were significantly higher (P<0.001) than in the other two groups. In addition, PUFA, n-3 and n-6 fractions were significantly lower (P<0.001) in goat milk from G2 than in the other two groups. Therefore, we can conclude that a diet based on TMR with a supplementation of orange pulp for goats did not improve the FA profile in milk. 97 Chapter 1 Table 8. Fatty acid profile (% of total fatty acids) of the milk in Florida goats of three groups of animals: G1 (goats fed on concentrate), G2 (goats fed on Total Mixed Ration + Orange pulp), G3 (goats fed on pasture). Fatty acid a G1 G2 G3 SEM b Sig. C4:0 1.96 1.95 1.78 0.04 n.s. C6:0 2.07 2.03 1.90 0.04 n.s. C8:0 2.23 2.13 2.01 0.04 n.s. C10:0 8.85a 13.36b 8.68a 0.50 *** C11:0 0.06a 0.11b 0.10b 0.01 *** C12:0 3.39a 5.86b 3.75a 0.26 *** C13:0 0.05a 0.08b 0.09b 0.01 *** C14:0 7.45a 9.95b 7.39a 0.27 *** C14:1 0.16a 0.20b 0.53c 0.04 *** C15:0 0.46a 0.75b 0.98c 0.05 *** C15:1 0.02a 0.07b 0.03a 0.01 *** C16:0 25.71b 32.54c 24.48a 0.81 *** C16:1 n-9 1.27c 0.67a 1.00b 0.06 *** C17:0 0.58a 0.57a 0.68b 0.01 *** C17:1 0.31b 0.25a 0.25a 0.01 *** C18:0 13.84b 11.09a 16.28c 0.49 *** C18:1 n-9 trans 1.89b 0.41a 0.49a 0.16 *** C18:1 trans-11 (VA) 1.75b 0.88a 0.84a 0.10 *** C18:1 n-9 cis 22.72b 13.74a 23.96c 1.02 *** C18:2 n-6 trans 0.26a 0.29a 0.59b 0.03 *** C18:2 n-6 cis 2.91c 1.33a 1.92b 0.15 *** C18:3 n-6 0.06 0.07 0.05 0.01 n.s. C20:0 0.42c 0.23a 0.36b 0.02 *** C18:3 n-3 0.21a 0.22a 0.40b 0.02 *** CLA cis-9. trans- 11 (RA) 0.29b 0.22a 0.49c 0.03 *** C20:1 n-9 0.23c 0.08a 0.10b 0.02 *** CLA trans-10. cis-12 0.11b 0.08a 0.12b 0.01 *** C21:0 0.05a 0.06a 0.10b 0.01 *** C20:2 0.06a 0.05a 0.08b 0.01 *** 98 Chapter 1 Table 8 continuation. Fatty acid profile (% of total fatty acids) of the milk in Florida goats of three groups of animals: G1 (goats fed on concentrate), G2 (goats fed on Total Mixed Ration + Orange pulp), G3 (goats fed on pasture). Fatty acid a G1 G2 G3 SEM b Sig. C22:0 0.06b 0.10c 0.04a 0.01 *** C20:3 n-6 0.03b 0.03b 0.02a 0.01 ** C22:1 n-9 0.08b 0.03a 0.02a 0.01 *** C20:4 n-6 (ARA) 0.12b 0.15b 0.07a 0.01 *** C20:3 n-3 0.04b 0.02a 0.02a 0.01 *** C23:0 0.01ab 0.02b 0.01a 0.01 * C20:5 n-3 (EPA) 0.10 0.10 0.11 0.01 n.s. C22:2 0.08b 0.07b 0.04a 0.01 *** C24:0 0.02a 0.04b 0.03ab 0.01 ** C24:1 0.01a 0.03b 0.02a 0.01 ** C22:5 n-3 (DPA) 0.08a 0.11b 0.10b 0.01 *** C22:6 n-3 (DHA) 0.03a 0.06b 0.07b 0.01 *** SFA 67.21a 80.86c 68.67b 1.37 *** MUFA 28.43c 16.35a 27.24b 1.22 *** PUFA 4.36b 2.79a 4.09b 0.16 *** SFA/PUFA 15.47a 29.08b 16.84a 1.39 *** SFA/UFA 2.05a 4.23c 2.19b 0.22 *** CLA 0.52 0.54 0.57 0.01 n.s. n-3 0.45a 0.51b 0.70c 0.02 *** n-6 3.39c 1.86a 2.66b 0.15 *** n-6/n-3 7.54b 3.67a 3.82a 0.41 *** Mean with different alphabets (a, b and c) among columns differ significantly. a VA, vaccenic acid; RA, rumenic acid; ARA, arachidonic acid; EPA, eicosapentaenoic acid; DPA, docosapentaenoic acid; DHA, docosahexaenoic acid; SFA, saturated fatty acids (FA); MUFA, monounsaturated FA; PUFA, polyunsaturated FA; UFA, unsaturated FA; CLA, total conjugated linoleic acid; n-3, all FA with last double bond at 3 rd carbon from the methyl end; n- 6, all FA with last double bond at 6th carbon from the methyl end. b Standard Error of Mean. 99 Chapter 1 Moreover, Table 9 summarizes the results obtained for the correlations coefficients of FA fractions and retinol in milk and plasma, and lutein in plasma. It could be observed that the content of retinol in plasma did not show significant correlations with any of the FA fractions analyzed in milk. Whereas, the n-6/n-3 fraction in milk showed (Table 9) significant and negative correlations with both content of lutein in plasma (r = -0.767, P<0.05) and retinol levels in milk (r = -0.692, P<0.01). In this sense, previous authors (Rafalowski et al., 2014) reported a high positive correlation between the sum of polyenoic and monoeonic FA and the content of vitamin A, among other antioxidants. This affirmation is in disagreement with our data (Table 9). This is probably due to the fact that milk from animals fed on concentrate (G1) showed (Table 8) higher (P<0.001) n-6/n-3 fraction and lower amount of lutein in plasma (P<0.05) and retinol in milk (P<0.001) than the milk from animals with a diet based on TMR + orange pulp supplementation (G2) and goats fed on pasture (G3). Therefore, it could be concluded, that under our trial conditions, there is an effect of the diet on the FA profile in goat milk according to previous studies (Delgado-Pertiñez et al., 2013; Rafalowski et al., 2014), as well as, orange pulp supplementation did not improve the FA profile in goat milk as it was explained above. 100 Chapter 1 Table 9. Correlation coefficients between retinol and FA in milk and lutein and retinol in plasma. Lutein (plasma) Retinol (plasma) Retinol (milk) Retinol (plasma) -0.561 Retinol (milk) 0.535* -0.403* SFA 0.599 -0.263 0.323 MUFA -0.505 0.262 -0.314 PUFA -0.567 0.257 -0.381 n-6/n-3 -0.767* 0.194 -0.692** *Correlation coefficients significant at a level of P<0.05 **Correlation coefficients significant at a level of P<0.01 Effect of the diet on the retinol levels and FA profile in goat milk Finally, in order to assess the effect of the diet on the retinol levels and FA profile in milk two discriminant analyses were carried out. The first one considered retinol levels in milk as variable (Table 10). As a result up to 65.9% of the animals were correctly classified into their group according to their diet. More specifically, 100 % of goats with a diet based on concentrate (G1) were accurately classified, whereas in the groups of pasture (G3) and TMR supplemented with orange pulp (G2) just half of the animals (53.8% and 46.7 %, respectively) were correctly classified. According to these results, it could be stated that supplementing goat diet with orange pulp makes no difference in retinol milk levels. Thus, it could be pointed out the potential of this supplementation in the diet to achieve retinol milk levels similar to those in milk from grazing animals, which are traditionally considered as “higher quality” (Prache et al., 2009). Moreover, in those arid areas where pasture are scarce, supplementing TMR with orange pulp could suppose 101 Chapter 1 a nutritional advantage, since vitamin A levels in goat milk were increased .A second discriminant analysis was performed considering the FA profile in milk. SFA and n-6/n- 3 fractions were considered in the analysis and the 100% of the goats were accurately classified according to their diet Figure 11). Thus, it could be affirmed, once again, that the FA profile of milk is influenced by the diet of the animals. In addition and according to data in Table 8 the SFA fraction was highly influenced by the diet, as the highest (P<0.001) SFA fraction appeared in goats from G2. Therefore, supplementing with orange pulp did not seem useful to enhance the healthy FA profile in goat milk. Table 10. Classification matrix after discriminant analysis based on retinol content in milk: assigned (columns) against real data (rows). Predicted group (%) Group 1 2 3 1 100.0 0.0 0.0 2 20.0 46.7 33.3 3 23.1 23.1 53.8 Group 1: goats with a diet based on concentrate. Group 2: goats fed on TMR (Total Mixed Ration) and orange pulp supplementation. Group 3: pasture goats. 102 Chapter 1 Figure 11. Classification of goats according to their diet (concentrate, TMR + orange pulp supplementation and pasture) considering SFA and n-6/n-3 fraction in milk CONCLUSIONS Retinol levels in milk from goats fed on a diet rich in carotenoids, pasture or supplemented with orange pulp, are higher than in milk from animals fed on a diet based on concentrate. Additionally it could be claimed that supplementing goat diet with orange pulp in arid areas where pasture are scarce, may increase vitamin A levels in milk. In addition, milk from goats fed on pasture showed a healthier FA profile, with PUFA ratios over 4.09. However, orange pulp supplementation to TMR did not show any improvement on the FA profile in milk. 94 Chapter 3 95 Chapter 3 96 Chapter 3 97 Chapter 3 98 Chapter 3 115 Chapter 3 Chapter 3 Effect of different carotenoid-containing diets on the vitamin A levels and colour parameters in Iberian pigs’ tissues: utility as biomarkers of traceability ANTECEDENTES Los consumidores son cada vez más exigentes y reclaman alimentos seguros y de alta calidad, de ahí que el término "trazabilidad" haya ido adquiriendo importancia en los últimos años (Storøy et al, 2013; Van Dijk et al., 2008). La trazabilidad en productos de origen animal puede estar basada en el análisis de micronutrientes que están presentes en la dieta y pueden aparecer como tales o metabolizado en los tejidos de los animales, como los carotenoides o el retinol (Ballin, 2010). Por otra parte, estudios previos (Priolo et al, 2002; Ripoll et al, 2008) informaron que las coordenadas tricromáticas del espacio CIELab fueron útiles para obtener el 100% de discriminación entre corderos alimentados con forraje o con alimentos concentrados. Sin embargo, pocos estudios se han realizado en ganado porcino (Carrapiso y García, 2005). El cerdo ibérico es uno de los más importantes en área Mediterránea, tanto en el tamaño de la población como en su importancia económica (Juárez et al., 2009). De acuerdo con la nueva legislación española (BOE, 2014) existen tres categorías diferentes para los productos del cerdo ibérico según su sistema de alimentación. Los productos de alta calidad están relacionados con el sistema de Montanera, con una dieta a base de bellotas y pastos. Por lo tanto, sería interesante poder diferenciar los productos procedentes de animales criados en Montanera de las otras estrategias de crianza de cerdos ibéricos. 116 Chapter 3 OBJETIVOS El objetivo principal de este estudio fue buscar nuevas herramientas para diferenciar los productos del cerdo ibérico de acuerdo con el sistema de cría de los cerdos. En este sentido, también se propusieron los siguientes objetivos específicos: - Evaluar el efecto de tres dietas con distinto contenido en carotenoides en los niveles de vitamina A en los tejidos de cerdo ibérico y el color de la grasa perirrenal. - Evaluar la utilidad de los niveles de retinoides en diferentes tejidos de cerdo ibérico como una herramienta de trazabilidad para diferenciar los sistemas de cría de estos animales. DISEÑO EXPERIMENTAL Para este estudio se dividieron treinta cerdos ibéricos en dos grupos según su sistema de producción: Montanera (basada en bellotas y pasto) y Cebo (dieta basada en alimentos concentrados). Se recogieron muestras representativas de los alimentos de los animales de cada explotación. Además, se recogieron, en el momento del sacrificio, muestras de sangre, grasa perirenal e hígado de cada animal. Se midió el color (C* ab , L*, a*, b* y h ab ) de la grasa perirenal siguiendo la metodología CIELab. También se calcularon, de acuerdo a la metodología descrita por autores previos (Prache & Theriez, 1999; Priolo et al., 2002; Ripoll et al., 2008; Zawadzki et al., 2013), los valores de los espectros de reflectancia trasladados (TR, de las siglas en inglés) así como el Valor Absoluto de la Integral (AVI, de sus siglas en inglés). La extracción de carotenoides de los alimentos se realizó de acuerdo a métodos publicados por otros autores (Kean et al., 2007; Panfili et al., 2004). Las extracciones en las muestras de plasma y grasa perirenal se hicieron siguiendo la metodología descrita 117 Chapter 3 por Lyan et al. (2001) Dunne et al. (2006), respectivamente. Finalmente, para las extracciones de retinoides y carotenoides en los hígados se tuvo en cuenta el método analítico publicado por Woodall et al. (1996). El contenido en carotenoides y retinoides de todas las muestras fue analizado por HPLC y la identificación de los ésteres de retinol en hígado se realizó mediante análisis HPLC-MS. RESULTADOS La dieta de los animales criados en un sistema de Montanera fue significativamente diferente (P<0.01), en cuanto al perfil en carotenoides, de la dieta del grupo criado en un sistema de Cebo, por lo tanto cabría esperar que el contenido en carotenoides y retinoides en los tejidos de los dos grupos de animales fueses también diferentes. Sin embargo, no se detectaron carotenoides en ninguno de los tejidos biológicos de cerdo ibérico estudiados, aunque sí se detectaron retinoides. Cuatro ésteres de retinol, además de (todo-trans)-retinol, fueron identificados por HPLC-MS en el hígado de cerdos ibéricos. La forma predominante de la vitamina A en hígado fue el palmitato de retinol: 45.8% de la vitamina A en el hígado de los cerdos criados en Montanera y 44.1% de aquellos criados en un sistema de Cebo, siendo significativamente mayor en el primer grupo (P<0.05). El contenido tanto de (todo-t)- retinol como de linoleato de retinol en hígado fue, también, mayor (P<0.05) para los animales alimentados en un sistema de Montanera. Por último, no se encontraron diferencias significativas entre los dos grupos de cerdos ibéricos ni para el oleato de retinol ni para el estearato de retinol. Tampoco se encontraron diferencias significativas entre los dos grupos respecto a la concentración de retinol en plasma y grasa perirenal. Sin embargo, sí hubo diferencias (P<0.01) en el parámetro de color L* en la grasa perirenal según la alimentación de los 118 Chapter 3 animales, aunque el resto de los parámetros de color (a*, b*, C* ab , h ab ) no mostraron diferencias en este sentido. Además, ni TR (450-510 nm) ni AVI fueron diferentes entre los dos grupos de cerdos, probablemente debido a la ausencia de carotenoides en la grasa de los animales. on el fin de evaluar la utilidad de los parámetros estudiados para discriminar los sistemas de alimentación del cerdo ibérico, se realizó un análisis discriminante teniendo en cuenta, por una parte los niveles de retinoides en hígado y por otra los parámetros de color en la grasa perirrenal. En el modelo fueron incluidos los niveles de ésteres de retinol y (todo-t)-retinol en el hígado, lo que permitió una correcta clasificación del 92.9% de los animales. Por otro lado, cuando se consideraron los parámetros de color (L* y h ab ) en el modelo la clasificación de los animales fue peor, ya que este modelo permitió clasificar correctamente el 78,6% de los animales de acuerdo a su sistema de cría. CONCLUSIONES De acuerdo con los resultados obtenidos en este estudio, las principales conclusiones son: 1. El contenido en retinol en plasma y grasa perirenal de cerdo ibérico no se vio afectado por la dieta de los animales. 2. El contenido en retinoides en el hígado de los cerdos ibéricos se mostró como una herramienta útil para diferenciar estos animales según su dieta y, por tanto, para ser usada en estudios de trazabilidad. 3. Algunos parámetros de color (L * y h ab ) en la grasa perirrenal parecieron, también, útiles para diferenciar los cerdos ibéricos de acuerdo a su dieta. 125 Chapter 4 Chapter 4 Carotenoids and fat-soluble vitamins in horse tissues. A comparison with cattle . ANTECEDENTES Según Schweigert (1998) y en base a la acumulación de carotenoides en el tejido adiposo, los mamíferos pueden dividirse en dos grupos: "white-fat", que no absorben carotenoides como el cerdo, la cabra, la oveja y los roedores; y los animales "yellow-fat", que sí absorben carotenoides el vacuno, los caballos y las aves están comprendidos en este grupo. Sin embargo, las razones de las diferencias entre especies en el metabolismo de los carotenoides no se conocen bien todavía. Muchos estudios (Calderón et al., 2007; Majchrzak et al, 2006;.. Martin et al, 2004) han evaluado la acumulación de carotenoides y vitaminas liposolubles (retinol y αtocoferol) en los fluidos y tejidos de vacuno (plasma, leche, grasa, hígado, etc), así como su implicación en el punto de vista de los consumidores respecto a los productos de vacuno. Sin embargo, existen pocos estudios con este objetivo en caballos (Schweigert y Gottwald, 1999), a pesar de la ventaja que podría suponer la mejora de la aceptabilidad de los productos equinos por parte de los consumidores basándose en la presencia de compuestos que promueven la salud (como los carotenoides y las vitaminas liposolubles), cubriendo la necesidad actual de aumentar el consumo de productos del caballo (Lorenzo et al., 2013). Además, la percepción de los colores juega un papel importante en la evaluación de los consumidores en la calidad de los alimentos (O'Sullivan et al., 2003), pero existe poca información disponible acerca de los productos equinos en este sentido. 126 Chapter 4 OBJETIVOS Los objetivos de este estudio fueron: 1. Evaluar la acumulación de carotenoides, retinol y α-tocoferol en fluidos biológicos y tejidos en caballo en comparación con el vacuno. 2. Estudiar las diferencias en el color de la grasa entre las dos especies en relación a su composición en carotenoides. DISEÑO EXPERIMENTAL Cuatro grupos de animales, de 15 animales cada uno, se utilizaron en este estudio. El Grupo 1 (G1) incluía 15 yeguas lactantes criadas en un sistema extensivo basado en pastos. El Grupo 2 (G2) estaba compuesto por 15 vacas lactantes criadas, también, en un sistema extensivo. Los animales del Grupo 3 (G3) eran potros machos con una dieta basada en alimentos concentrados. Por último, el Grupo 4 (G4) estaba formado por terneros machos con una dieta basada, también, en alimentos concentrados. Se recogieron muestras representativas de todas las dietas. También se cogió sangre de todos los animales, leche de las yeguas (G1) y las vacas (G2), así como muestras de hígado y grasa perirenal de los potros (G3) y los terneros (G4) durante el sacrificio. Se midió el color de la grasa perirrenal (L *, a *, b *, C * ab y hab) de acuerdo al sistema CIELab (CIE, 1976). Además, también se adquirieron los espectros de reflectancia en la región visible y se transformaron (TR) para calcular el Valor Absoluto de la Integral (AVI, por sus siglas en inglés) de acuerdo a trabajos anteriores (Prache y Theriez, 1999; Priolo et al., 2002; Ripoll et. al, 2008). Se hicieron extracciones de carotenoides, retinol y α-tocoferoles en todas las muestras. Para los alimentos se siguió la metodología descrita por Pickworth et al. (2012). Las muestras de plasma se extrajeron siguiendo la metodología descrita por Lyan et al. (2001). 127 Chapter 4 Para las extracciones de leche se consideraron los métodos descritos por Hulshof et al. (2006) y Nozière et al. (2006b). En el caso de los hígados, se siguió la metodología descrita por Woodall et al. (1996). Finalmente, las extracciones de grasa perirrenal se llevaron a cabo de acuerdo con Dunne et al. (2006). El contenido en carotenoides, retinoides y α-tocoferol se analizó por HPLC para todas las muestras. Los ésteres de retinol en el hígado se identificaron mediante análisis por HPLC-MS. RESULTADOS Las dietas de los animales con el mismo sistema de producción, es decir extensivo (G1 y G2) o intensivo (G3 y G4), mostraron perfiles de carotenoides y α-tocoferol similares, mientras que no se detectó retinol en ninguna de las dietas estudiadas. Aparecieron diferencias significativas en los niveles de carotenoides en plasma entre yeguas y vacas, siendo la concentración de (13Z) -β-caroteno y (todo-E)-β-caroteno más alta (P <0,001) para las vacas que para las yeguas. Además, se detectó luteína en el plasma de las vacas, pero no en el de las yeguas. Por lo tanto, a pesar de que no había diferencias en las dietas, había diferencias en los niveles circulantes de carotenoides entre bovinos y equinos. Por otro lado, tanto el retinol como el α-tocoferol aparecieron en el plasma de ambas especies. Aunque no se observaron diferencias significativas en los niveles de retinol entre ambas especies, hecho que apoya la hipótesis de la regulación metabólica en la absorción de la vitamina A en ambas especies Con respecto a la leche, el único carotenoide detectado en ambas especies fue βcaroteno, siendo su concentración significativamente mayor (P <0,05) en las vacas que en las yeguas. Estos niveles bajos, a pesar de la elevada presencia de β-caroteno en los alimentos y en el plasma de vacas y yeguas, nos sugieren una transformación similar y limitada de β-caroteno en retinol en ambas especies. Finalmente, el α-tocoferol estaba 128 Chapter 4 presente en la leche de vaca pero no en la de yegua, aunque son necesarios más estudios para establecer el metabolismo de α-tocoferol en ambas especies con el fin de entender las diferencias entre las dos. Tanto en el plasma de los potros como en el de los terneros se detectaron β-caroteno y retinol pero no α-tocoferol, probablemente debido a la ausencia de esta vitamina en el alimento concentrado de ambas especies. Con respecto a los niveles de β-caroteno, el plasma de los terneros (G4) mostró los niveles más altos (P <0,05) que el de los potros (G3), probablemente debido a la diferencia mínima en el contenido en isómeros de βcaroteno en los piensos de ambas especies . Sin embargo, no se observaron diferencias significativas en los niveles plasmáticos de retinol entre potros (G3) y terneros (G4). Por lo tanto, se podría afirmar que la concentración de carotenoides en plasma de ambas especies depende de la dieta, mientras que los niveles de retinol son regulados de manera eficiente por ambas especies, independientemente de la dieta. No se encontraron carotenoides en la grasa ni de los potros (G3) ni de los terneros (G4), pero sí se detectó retinol. Este hecho puede ser debido a la mínima cantidad de carotenoides en las dietas de estos animales que podría resultar en cantidades insignificantes de estos compuestos en su grasa. Además, no se detectó α-tocoferol en la grasa de estos animales probablemente debido a su ausencia en la dieta. Por otro lado, todos los parámetros de color medidos en la grasa perirrenal mostraron diferencias significativas (P <0.01, P <0,001) entre ambas especies. El porcentaje de reflectancia de los espectros en la región entre 450 y 510 nm y los valores de AVI fueron mayores (P <0.01) para los potros que para los terneros, lo que nos lleva a pensar que en los potros acumulan más carotenoides en la grasa perirrenal que los terneros, aunque se necesitan más estudios para confirmarlo. 129 Chapter 4 Por otra parte, el contenido en vitamina A en el hígado de ternero fue menor (P <0,001) que en el de potro. La vitamina A estaba presente como (todo-t)-retinol y los ésteres de retinol linolenato de retinol, linoleato de retinol, oleato de retinol, palmitato de retinol y estearato de retinol. El palmitato de retinol fue la principal forma de vitamina A para ambas especies. Por otra parte, el perfil de ésteres de retinol fue diferente para ambas especies. CONCLUSIONES De acuerdo con los resultados obtenidos en este estudio, las principales conclusiones son: 1. Existen diferencias específicas por especie en la acumulación de carotenoides, retinoides y α-tocoferol entre el ganado equino y bovino 2. Los potros acumularon mejor retinoides en el hígado que los terneros. 3. El hígado de potro puede ser considerada como una buena fuente de vitamina A, lo cual debe ser tenido en cuenta desde el punto de vista nutricional para los seres humanos. 130 Chapter 4 131 Chapter 4 132 Chapter 4 133 Chapter 4 134 Chapter 4