Differential Characteristics of Milk Produced in Grazing Systems and Their Impact on Dairy Products

© 2012 Descalzo et al., licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Differential Characteristics of Milk Produced in Grazing Systems and Their Impact on Dairy Products


Introduction
Functional foods may exhibit health benefits beyond their nutritional value.Many traditional recommendations on food selection have included this view.In more recent years the interest in food with specific health benefits has greatly increased and stimulated the development of respective products for the food market.At the same time large efforts are made to substantiate health claims by validated experimental methods.
Dairy products have come to play an important role in this context.Regarding the nutritional and extra nutritional characteristics of milk, it can be considered a multifunctional food product.Milk is a source of proteins, lipids, vitamins and minerals, but also exerts other beneficial properties due to the presence of numerous bio-active molecules either through the direct consumption of milk or its derivatives (Guimont and others, 1997).
It is widely recognized that pasture, when efficiently grazed, is the cheapest feedstuff (Dillon, 2006) and that the cost of milk production is reduced as the proportion of grazed pasture in the diet of the cow increases (Dillon and others, 2008).Nevertheless, using pasture as the only source of food is inadequate for high milk producing cows.The lower intake of dry matter (DM) and energy would be the main cause of suboptimal production even under conditions of adequate quantity and quality of pasture (Kolver y Muller, 1998;Reis y Combs, 2000;Bargo and others, 2002).
Cows adapt to grazing through increased forage intake, but in order to achieve this, it is necessary to ensure around 60 kg DM/cow/day (Delagarde and others, 2004).Under these conditions, the maximum yield obtainable on high quality pastures seems to be at around 22-25 kg/cow/day in spring and 18 kg/cow/day in autumn.However, at high levels of energy supplementation, rumen pH, the diurnal pattern of volatile fatty acid concentration and fiber digestion may be altered because of changes in rumen fermentation (Tamminga, 1993).These changes may provoke lower consumption of forage and a drop in the fat content of milk.In this regard, studies carried out in EEA INTA Rafaela suggest that the optimal level of supplementation that could maximize milk production without affecting the rumen of dairy cows grazing high quality lucerne, would be 7 kg of concentrate containing approximately 70% corn (Salado and others, 2010 a and b).
On the other hand, it is also known that the amount of conjugated linoleic acid (CLA) in milk fat increases according to the level of pasture included in the diet, with the highest CLA concentrations found in cattle fed 100 % pasture (Dhiman and others, 1999).
Alike other country, Argentina is moving forward to more complex livestock production systems.In this regard, component feeding systems based on the use of grazed pasture are gradually being replaced by confining ones, in which conserved forages (hay, silage) and concentrates (cereal grains, agro-industrial byproducts) are fed to dairy cattle as a total mixed ration (TMR).This phenomenon could be explained by the fact that argentinean dairy farmers are looking for ways to release land for agriculture (mainly for the cultivation of soybean [Glycine max]), which is considered a more profitable and simpler activity than dairy farming.
Nowadays, it is supported that the extreme intensification of dairy production systems would lead to lose not only the advantage that grazed pastures have on the reduction of milk production costs (Garcia and Fulkerson, 2005;Dillon and others, 2008;Macdonald andothers, 2008, McEvoy andothers, 2008) but also the marketing niches that look for dairy products from pasture-fed cattle due to their high CLA content (White and others, 2002).In this respect, pasture-fed dairy cows produced a significantly higher concentration of CLA (83%) than TMR fed-confined dairy cows (0.66 vs. 0.36% of the total fatty acids) (White and others, 2001).
For that purpose, grass-based dairy production should consider some productive aspects such as high-quality pasture and livestock adapted to a high-forage diet.
Pasture is a highly organized structure which changes with season and management.These changes can affect photosynthesis and hence dry matter accumulation, the degree of utilization and the feeding value of herbage.At the same time, these changes in pasture can affect the quality of milk and thereby the quality of dairy products.

Natural antioxidants: Lipid-soluble vitamins
Plant secondary metabolites are incorporated into the tissues through dietary delivery.Among these, alpha-tocopherol is the main active form of vitamin E in animal tissues.It is selectively incorporated among eight isomers (alpha, beta, gamma and delta tocopherol and tocotrienol respectively) that are naturally found in plants.Vitamin E is the main fat-soluble Differential Characteristics of Milk Produced in Grazing Systems and Their Impact on Dairy Products 341 antioxidant incorporated into animal cell membranes and is also a regulator of cellular functions that are mediated by protein kinase C.
Carotenoids are lipophilic secondary plant and algae products consisting of eight isoprene units (C40).Among about 800 known carotenoids, up to ten of them have been determined in forages, namely oxygen-containing xanthophylls (lutein, epilutein, antheraxanthin, neoxanthin, violaxanthin and zeaxanthin), and carotenes (preferably -carotene) of a hydrocarbon nature.Due to the high number o f d o u b l e b o n d s , c a r o t e n o i d s u n d e r g o oxidation and isomerisation (cis-isomers formation from all-trans-ones), in cut forage by the action of light.Carotenoids in cow's milk and consequently in milk products are important for human health and nutrition as natural antioxidants and some of them as precursors of vitamin A, among which all-trans--carotene is the main provitamin.Moreover, milk fat carotenoids cause the yellow color of butter and many cheeses, which is positively perceived by many consumers as ''green image" because of its association with grazing animals.This vitamin is implicated in reproduction, immunity and normal function of retinal cells.Lutein is another carotenoid which is abundant in plants and can be used as a tracer of pasture feeding, as animals are not able to synthesize this molecule and it is stored in the animal's fat after absorption and thus found in milk and meat (Prache 2005).
These molecules are highly variable in forages, silage and hay (Müller 2007;Hidiroglou 1996), and depend on the fermentation form as well as on the conservation method.Praché and others (2009) reported values between 40 and 123 microg/g DM for beta carotene (60 microg/g DM for lucerne) and from 167 to 437 microg/g DM for lutein (142 for lucerne) in pasture collected between Spring (May) and Summer (July).Similar variation in carotenoid levels in mountain grazing was found by Calderón and others (2006) with values that ranged between 18 and 60.4 microg/g DM for beta carotene and between 96 and 262 microg/g DM for lutein.
Data from four separated experiments, which were carried out in INTA, also show a high variability.As shown in table 1, fresh pasture showed the highest input of alpha-tocopherol and beta-carotene among feed components.The variation of its composition is high, depending on seasonal variations, phenological stage and conservation technology, among other factors.Other components of feed are also sources of dietary antioxidants, but they showed less variation and their contribution in vitamins was less significant compared with pasture.

Natural antioxidants: Polyphenols
Polyphenols are antioxidants that may prevent various pathologies (cancer, cardiovascular diseases, immune deficiencies, etc.).Particularly, isoflavones may have a preventive effect against breast and prostate cancers and, to a certain extent, against some disorders related to menopause or andropause.The active molecules are plant-based isoflavones (directly absorbed) as well as a transformation component in the large intestine, equol.The main source of isoflavones in human diet are soy and soy-derived food.Among forage plants, other legumes (certain types of clover and lucerne) contain high levels of isoflavones.They may constitute up to 2% of dry matter in purple clover and are well preserved after silaging.Of the active isoflavones in forage, daidzein and formononetin were transformed in the rumen into equol, which was secreted in the milk of goat and cow species in sufficient quantities to suggest they may exert a biological effect.Moreover, Besle and others (2010) found different phenolic compounds that can be used as tracers for the diet of dairy cows.Data shown in table 1, indicate that coumestrol was the main phytoestrogen found in lucerne (pasture and hay), corn silage, sunflower expeller and wheat bran.Concentrate mixture showed higher amounts of daizein, than other dietary compounds, probably derived from soy.In this experiment, the aglycone and glucoside forms were not differentiated.The extraction was performed in acid medium and therefore, the species were converted into the aglycone form.a For the determination of antioxidant vitamins, samples from individual components of the diet were extracted with hexane (after a saponification step) and quantified using high performance liquid chromatography.Carotenoids were monitored at 445 nm, tocopherols were detected by fluorescence at 296-330 nm ( exc- em respectively) and retinol at 335 nm (Rossetti et al, 2010).b Mean variation coefficient of all vitamins calculated for each dietary compound.

Volatile compounds
Terpenes are direct biomarkers of animal diet as they are not modified by animal metabolism.In addition, the transfer of these compounds into milk fat is very fast and apparently exhibits no memory effect (Viallon and others, 2000).Therefore, their presence in milk and dairy products could be used to discriminate contrasting feeding conditions (Prache and others, 2005).
Lucerne pasture showed the highest volatile production (table 2).The most abundant terpene was d-limonene found in all dietary components.Only three minor compounds, nerol, menthol and linalyl acetate were found exclusively in lucerne.Terpinen-4-ol, was found in lucerne pasture and maize silage in a 2 to 1 proportion.In addition, ocimene was 4.5 times higher in lucerne pasture compared with the other components and linalool was 4 times higher than in maize silage and wheat bran.These results indicate that a group of terpenes, and their abundance and not a single compound could be an indicator of the different components of the diet.(2008)(2009).For this assay, a fiber (CAR/PDMS/DVB 2cm-50/30mm) was used to extract/concentrate volatile compounds in the headspace (HS) of chopped samples.GC-FID was performed on a Shimadzu 14-B GC with a flame ionization detector (Negri et al., 2008).
b Mean variation coefficient of all volatiles calculated for each dietary compound.Lucerne samples also presented a higher proportion of volatile aldehydes (table 3), at the expense of trans-2-hexenal, which was the most abundant volatile after d-limonene.Tava and Peccetti (1997) reported that trans-2-hexenal was the most frequent volatile in 13 genotypes of lucerne.This compound, originated in flowers, should have a physiological role for plant pollination by hymenopterans (Peccetti and others 2002).
Hexanal and trans-2 hexenal are end-products of the oxidation of C18:2 and C18:3 fatty acids respectively.In fact, fresh pasture shows the highest hexanal production, resulting from linoleic acid, whereas trans-2-hexenal indicates a high percentage of C18:3.These fatty acids are oxidized when fresh pasture is cut, producing the corresponding aldehydes.Both aldehydes and terpenes are responsible for the typical odour of different plant species and could be found in milk and dairy products.

Fatty acids
Information available about the fatty acid composition of pasture lipids is scarce.Unsaturated fatty acids, particularly n-3 polyunsaturated fatty acids (PUFA) and conjugated linoleic acid (CLA) may be beneficial to human health.Therefore, maximizing C18:3 and CLA in milk and dairy products would benefit human health and nutrition.Milk and meat are the only significant source of CLA in the human diet and this appears to be related to the consumption by ruminant of fresh pastures (Elgersma and others, 2003).
Sources of variation in pasture lipid composition are plant species, growth stage, temperature and light intensity.Fresh grass contains a high proportion (50-75%) of its total fatty acid content in the form of n-3 linolenic acid.Levels of linolenic acid vary with plant factors such as stage of maturity and with light disponibility.Fatty acid profiles are distinctive to particular species, which confirm that fatty acid composition of grasses is under considerable genetic control (Dewhurst and others 2001).This offers the potential to select for grasses with higher concentrations or altered types of fatty acids.
Garcia et a. ( 2007) evaluated the range of fatty acid concentrations within various pastures and forage crops, at different cutting dates and season in Argentina for beef and milk production.Season affected ether extract (EE%) and fatty acid composition of lucerne cultivars (table 4).No interactions were detected between dormancy group and season (p<0.344).During the fall increased C18:2 and decreased C16:0, C16:1, C18:0.The differences between spring and summer were for C16:1, C18:0, C18:2 and C18:3.The EE% and contribution of C18:2 and C18:3 were higher in fall compared with the other seasons.
The fatty acid composition, percentage of ether extract and contribution (g/100g DM) of C18:2 and C18:3 of several cultivars are shown in tables 5 and 6.Cutting date has affected significantly these parameters.Whole milk powder from pasture produced milk showed the same fatty acids profile as raw milk (table 5), and could constitute a natural source for n-3 fatty acids.In a similar way, antioxidants profile reproduced the composition of raw milk with estimated losses around 20 and 40% being higher for retinol and vitamin D3.

SEASON
Higher natural antioxidants in whole milk powder lowered TBARS values in pasture (A70) samples compared to silage samples (A0) after 18-months shelf life assay (20ºC in sealed bi-laminated plastic pouches under nitrogen atmosphere).

Properties of milk produced on pasture
The nature of cow's forage diet, i.e. botanical composition, maturity stage and preservation mode, strongly influences milk composition in fatty acids, vitamins and carotenoids (Chillard 2001, Hartman 1965).The contents of retinol, α-tocopherol and β-carotene, lutein, xantophylls, saturated and polyunsaturated fatty acids profile in plasma, milk and milk fat are influenced by the diet.In addition, some of these parameters also differed according to sire and stage of lactation (Lucas 2006, Soren 1999).Calderón and others (2007) reported that the incorporation of grass silage and lucerne protein (75:25), in an experimental diet designed for Montbéliarde cows in midlactation induced a rapid increase in plasma concentrations of beta carotene and vitamin E. The incorporation of these compounds varied in a linear form with the proportion of grass within the diet.Similar responses were observed in milk for vitamin E, whereas there was an apparent saturation in milk concentrations of beta carotene at high levels of carotenoid intake, i.e., when plasma beta carotene exceeded 5 microg/ml.
In an experiment conducted in INTA-Rafaela (Santa Fe, Argentina), two isoenergetic diets (1.55 Mcal EN/kgDM) of contrasting nature: 70% lucerne pasture (ALF) and grain sorghum silage (SS) were offered to Holstein cows on their second third of lactation.Milk and protein yields were higher (P < 0.01) in ALF than SS diet (31.78, 26.14 l/v/d and 1.019, 0.841 kg/v/d, respectively).Fat concentrations were lower on ALF than SS diet (P < 0.01).All remaining variables were not significantly different, with the only exception of total solid contents (P < 0.05).No significant differences were found on live weight variation (ALF= 0.084, SS= 0.271 kg/v/d).
As shown in figure 1, fat-soluble vitamins increased over the first 20 days after a dietary shift from silage to grazed pasture (70% lucerne), and the concentrations were stabilized at day 40 with persistence until day 60 (Rossetti and others, 2010).Similarly, Calderón and others (2007) observed a rapid increase during the first 14 days in antioxidant vitamins in plasma and milk of Montbéliarde dairy cows, after a dietary shift from a low-carotenoid diet based on hay and concentrates to a high-carotenoid diet based on grass silage that was prepared from perennial ryegrass (Lolium perenne).The concentration of α-tocopherol, 13-cis-β-carotene, all-trans-βcarotene and lutein reached a plateau between 21 and 28 days and persisted until day 42.The incorporation of zeaxanthine into plasma was reported but it was either not detected in milk (Calderón et al., 2007), or its concentration was around 1% of total carotenoids (Butler 2008).

Effect of increasing pasture content into a total mixed ration diet in relation to health promoting compounds
An increase of pasture in the diet, at the expense of a total mixed ration (TMR), has been shown to improve the incorporation of antioxidant vitamins into milk.Figure 2 shows the relationship between the percentage of pasture (grazed oat pasture) vs. α-tocopherol, retinol, all-trans-β-carotene and lutein in milk after 30-day treatments.Vitamin A in milk is incorporated under the form of retinol and carotenoids.Among these, all-trans-β-carotene was the mayor form recovered followed by lutein, which was consistent with what is generally reported.Although lutein was the major carotenoid found in pasture, its incorporation rate into milk was 10-fold lower than the incorporation of all-trans carotene.The same observation has been reported by Calderón (2007) and Butler (2008).Recovery of these species of vitamin A in milk was directly related to their concentration in feed.Pearson correlation coefficients for all-trans--carotene and lutein in feed and milk were R=0.8323 and R=0.7521 respectively.Concerning the lipoperoxyl radical-scavenging activity of all-trans retinol it behaves as a more effective antioxidant at a low partial oxygen pressure, low retinol concentration and high radical flux (Tesoriere 1997) In a revision article, Nozière and others (2006), have discussed that the extent of carotenoid degradation by microorganisms in the rumen remains uncertain because of the wide range of results, mostly on -carotene, from in vitro and in vivo studies.Whereas some authors reported no degradation (Dawson and Hemington, 1974;Cohen Fernandez and others, 1976a), others found moderate (10-25%; Davison and Seo, 1963;Potkanski and others,1974;Cohen Fernandez and others, 1976b;Mora and others, 1999) or higher -carotene disappearance (40-55%; King and others, 1962).The carotenoid supplement form could explain discrepancies among experiments, because degradation rates were usually higher when carotenoids were supplied as purified products than when provided in forages.This hypothesis was recently confirmed in vitro at the INRA in a study where no apparent degradation was detectable for lutein provided in forage, whereas 50% of the initial amount disappeared when the same quantity of lutein was added as a pure commercially available source.
Vitamin E concentration ranged from 16.5 to 46 microg/g milk-fat for milk from silage or pasture diets respectively.Pasture values were even higher than concentrations reported by Al Mabruk (2004) for grass silage supplemented with vitamin E (22.26 microg/g milk fat).Calderón and others (2007) showed an increase from 8 to 14 microg/g milk fat depending on the proportions of experimental feed (75% grass silage + 25% lucerne protein concentrate) in the basal diet (from 0 to 66%).
Four different assays conducted in INTA (Argentina) indicate that after a shift from silage to pasture diet α-tocopherol augmented from 16.7 to 40.37 (70% grazed lucerne); from 11.59 to 26.05 and 28.53 (35 and 70% grazed lucerne pasture); 22.78 to 36.91 (80% grazed oat pasture) and 24.36 to 30.42 (75% oat pasture).In fact, Schingoethe and others (1978) reported a high variability in the concentrations of vitamin E in milk, depending on the grazing conditions.The concentration augmented from 20 to 50 microg/g milk fat at the time cows started grazing pasture.The same observation was reported by La Terra and others (2010) with an increase of 37% and 68% in -tocopherol for cows fed 30:70 and 70:30 ratio for pasture:TMR, respectively in comparison to TMR diet.Therefore, results from INTA experiments showed the same tendency.However, one experiment conducted in autumn 2007, showed no differences between maize silage or pasture (lucerne) fed cows with an average of 27.5 microg/g milk fat, thus indicating that the quality of the diet and the pasture affects the incorporation of vitamin E into milk.
Vitamin D may be consumed in the diet as either ergocalciferol (D2) from plant sources or cholecalciferol (D3) from animal sources.With the exception of fatty fish, relatively few foods are naturally rich in vitamin D. In mammalians with ample sunlight exposure, the greatest source is endogenous vitamin D produced when 7-dehydrocholesterol in the epidermis and dermis of the skin is converted into vitamin D3 after exposure to ultraviolet B radiation.Vitamins D2 and D3 from dietary sources are transported to the liver, and from there to the tissues.Vitamin D3 from skin is transported through plasma.Therefore more evidence is needed to understand the factors that induce its variation in milk content.Consequently, for vitamin D, the dairy industry proposed technologically vitamin D-enriched dairy products to consumers (Graulet 2010).The vitamin D3 endocrine system has mostly been studied for its role in calcium and phosphorus metabolism and its possible role as an antioxidant has been neglected.Sardar and others (1996) demonstrated that D3 treatment brought about similar reduction in the extent of lipid peroxidation and induction in superoxide dismutase (SOD) activity, as with vitamin E supplementation in rats.
The concentration of vitamin D3 showed different tendencies among experiments.In one experiment, it ranged from 5.851.21;8.012.13 and 10.292.34microg/g milk-fat for diets containing 0, 40 and 80% oat pasture respectively, showing a linear increase with the proportion of pasture in the diet.This observation agreed with the described in figure 1, although values were lower (average of 30 microg/g milk-fat, in figure 1 with 70% of lucerne pasture diet vs. 10 microg/g milk-fat, with an 80% oat pasture diet).In contrast, in a second experiment the concentrations were 0.80; 0.91; 1,98 and 1.34, without significant differences (P >0.05) for diets containing 0, 25, 50 and 75% of oat pasture.Therefore, a tendency for higher values of vitamin D3 with the increase of grazed pasture was observed but further assays are necessary to describe the productive parameters that may be used to improve the concentration of natural vitamin D3 in cow milk.

Effect of increasing concentrate content on grazed pasture.
The addition of concentrate on a pasture based feed was studied in two separated experiments: total feed assignment around 30 and 31.9 kg DM/head/day in experiments 1 and 2 respectively.Pasture consumption diminished with the increase of concentrate within the diet: 18.0; 16.1 and 14.2 kg DM/ head/day for 3.5; 7.0 and 10.5 kg of concentrate respectively in experiment 1 but not in experiment 2. In both cases, the addition of concentrate had no effect on the concentration of antioxidants in milk.The average concentrations were 5.34; 37.25; 1.09 and 6.51 microg/g milk fat for retinol, -tocopherol, tocopherol and -carotene respectively.These values resulted similar to those obtained with pasture proportions above 40% in the diet (figures 1 and 2).Calderón et. al, 2007 demonstrated that the transfer of -carotene from plasma to milk reached a plateau when plasma levels exceeded 5 microg/ml and resulted a limiting factor in terms of the secretion of this antioxidant, that can be due to a limited uptake by the mammary gland or limited transport by binding to transport proteins like -lactoglobulin (Dufuor and Hartle, 1991) and/ or to saturation of milk fat globules.
It can be concluded that good quality pasture favored a linear increase of antioxidant vitamins in milk, and that this increment was not affected by the addition of concentrate in the feed.

Functional fatty acids in milk
There is a growing interest in CLA, considered to be beneficial in prevention of carcinogenesis (Ip and others, 1999), the n-6 to n-3 polyunsaturated fatty acids (PUFA) relationship, and the low percentage of saturated fatty acids, as these conditions are beneficial for the prevention of cardiovascular diseases.Their percentages in milk products can be increased through a suitable dietary regimen.Changes in milk fatty acids profile are possible due to the plasticity of milk fat (reviewed in Chillard and others 2000).
A second experiment was conducted in INTA with increasing percentages of lucerne pasture.Twenty four Holstein cows from the experimental herd (6 heifers and 18 multiparous) in mid lactation, were randomly assigned to three different treatments.During a first pre-experimental period of 7 days, considered as co-variable, all cows were fed the same diet, in a second period each group was assigned to their treatment during 21 days and in the third period of 28 days, each group received the corresponding diet, with 0 (A0 or maize silage control diet), 35 (A35) and 70 % (A70) lucerne pasture.Diets were isoenergetic and isoproteic.Lucern pasture was cut daily and offered to the cows.Rations were fed as TMR (20.34 kg DM/cow/day, in average).
Milk production, protein and urea content showed lower values for lucerne groups compared with milk from silage-fed cows (P<0.05).For other parameters, milk from the three treatments showed a similar overall quality.
As mentioned above, the proportion of pasture in the diet influenced the composition of saturated fatty acids (SFA) in milk (table 8).C4, C15, C16 and C17 FAs were higher in A70 milk compared with A35 and A0 treatments (P < 0.05).C12 FA was higher in A0 than in A35 and A70 milk (P < 0.05).Total C18:2 FA was higher in A0 milk, but differences were observed for individual isomers.Trans-9, trans-12 C18:2 and cis-9, trans-12 C18:2 showed similar percentages in all milk samples.However, trans-9, cis-12 C18:2 increased with the higher content of lucerne in the diet (A70 compared with A35 and A0), and conversely, cis-9, cis-12 C18:2 decreased with an increase of lucerne (P < 0.05).These results differed from Schroeder and others (2003) that reported higher levels of trans-vaccenic and c9 t 11 CLA in milk from pasture (Avena sativa L.) fed cows compared to milk from TMR fed cows.The mean percentage of CLA in argentine milk from the central Region was reported to be 1.2 (Castillo and others, 2006;Páez and others, 2007).The nature of the differences with previous reports could rely on the quality of pasture and its highly variable composition, as discussed in the section above.The mean C18:3 n-3 FA content in lucerne milk was approximately 1.7 and 2.9 times higher (A35 and A70 respectively) than in silage milk (P< 0.01).In addition, C 20:5 n-3 showed an lucerne-dependant increase (A70> A35 > A0).Therefore, the incorporation of lucerne pasture in the diet was associated with a higher proportion of milk n-3 FA compared with maize silage diet.Indeed, the plot of C 18:2 vs. C 18:3 for milk samples showed a differential distribution of milk according to the diet and could therefore be used to differentiate the incorporation of lucerne into the diet of dairy cows (figure 3).
This result is consistent with other results shown for grazed systems (Chillard and others 2001;Castillo and others 2006;La Terra and others 2010).The source of the variability remains to be further studied and strategies different that using pasture as the unique factor to improve trans-vaccenic acid and CLA in milk continue under experimentation.
Low percentages of CLA (around 0.34 %) was found to be associated to cows fed fermented roughage and concentrates (most intensive production farm) whereas high percentages (around 0.80 %) was found in the ecologically produced milk fat.The concentration of CLA correlated positively with and trans-vaccenic acid (Jahreis 1997).The high C18:3 content of young grass and its low fibre content interact to increase the production of CLA or its trans C18:1 precursors (Chillard and others 2000).The presence of lipid precursors is one of the factors described by Griinari and Bauman (1999), to improve CLA biosynthesis.The second factor are the changes associated with the microbial activity associated with ruminal biohydrogenation, which is incomplete and leads to an accumulation of trans vaccenic acid, the precursor of CLA biosynthesis.In addition, factors that regulate these patways as well as the activity of the Δ-9 desaturase in the mammary gland and its regulation are a matter of investigation.

Oxidative status in milk and milk powder
Pasture feeding improved the antioxidant status in raw milk but also the peroxidative index at expenses of an increase of the n-3 fatty acids.However, feeding experiments carried out under the same production conditions showed quite dissimilar results depending on the seasonal variations.
Dry matter consumption differed between pasture and silage feeding regimes and also between Autumn and Spring.This feature is probably attributable to the differences between pasture qualities between assays.Total protein (TP) and butter fat (BF) were lower in pasture than in silage milk (3.38 vs. 3.71 % BF and 3.13 vs. 3.28 % TP respectively).-tocopherol and -carotene were significantly higher in pasture samples (P<0.05), with a significant interaction between treatment and season for retinol and tocopherol, probably due to differences in pasture properties.-carotene was highly correlated with the proportion of lucerne in the diet, as shown in figure 2 for oat pasture.As a consequence of the increase of cartonoids, milk, milk powder and cheese had higher b* values, which was traduced in a yellower fat colour that could differentiate dairy products from milk produced on pasture.Keenan and others., (1970) demonstrated that carotene was the major pigmented material encountered in bovine plasma membrane, and the plasma membrane origin of MFGM (milk globule fat membrane) would explain the origin -carotene in milk, which reflects the dietary origin of this provitamin.
Gamma-tocopherol was the minor isomer of vitamin E in milk and dairy products.Its incorporation into milk was not influenced either by seasonal variation or by dietary treatment.The significance of this minor form of vitamin E incorporated in milk should be further stated.Its concentration remains at least one order below -tocopherol levels in vivo.However, its activity should be taken into account to prevent tissue damage.Particularly, tocopherol has the ability to protect against nitrogen-based free radicals, whilst -tocopherol cannot (Jiang and others, 2001).
Vitamin D3 was influenced by seasonal variation.The concentration in raw milk resulted 10-fold higher in spring than in autumn, probably due to degree of exposure to the sun which induces its biosynthesis in cows.Nevertheless, milk for human consumption is supplemented with 400 UI per quart (1 IU of vitamin D3 corresponds to 0.025 micrograms) to achieve the recommended mean intake of 5 micrograms per day (WHO).Natural vitamin D3 was not found in reggianito cheese.
Pasture feeding also modulated retinol content.Values were higher in pasture milk (p<0.05) with an interaction effect for seasonal variation.These differences were also found in milk powder and cheese, being enhanced in spring. .Therefore the susceptibility to peroxidation due to the number of double bonds in PUFAs in milk, increased with the proportion of pasture in the diet (table 9).
Oxidation in raw milk and milk powder was determined using the thiobarbituric acid reactive substances assay or TBARS, which is one of the oldest and most commonly used methods for assessing lipid oxidation in foods (Ladikos and others 1990).
This method is based on the spectrophotometric determination of extracted malonaldehyde, a minor product of oxidation, and can be performed either directly on the food product or on a steam distillate of the food.Due to the simple procedure and high correlation with sensory scores (Igene and others 1979), many studies have relied on TBARS for determination of oxidative status.The major disadvantage of the TBARS reaction is that it is not specific for malonaldehyde.No significant differences were found for TBARS (P>0.05), but this parameter showed consistently lower values in pasture milk, but was highly influenced by stationary variability.
Other method for the detection of intermediate peroxide compounds is the Peroxide Value.This parameter, as TBARS, was similar for all samples (data not shown).As indicated for feed components, numerous aldehydes are produced during oxidation, including octanal, nonanal, pentanal, and hexanal.Hexanal is the dominant aldehyde produced during oxidation (Dupuy and others 1987;Ajuyah and others 1993).It arises from both the 9 and 13 hydroperoxides of linoleate, and from other unsaturated aldehydes formed during the oxidation of linoleate (Shahidi and Pegg, 1994).It is a useful tool to assess lipid secondary oxidation products in milk (Erickson, 1999;Smet, 2009).This compound was higher in silage milk and could reflect a lower oxidative stability in milk and dairy products derived from silage-fed cows.Whole milk powder from pasture showed the same fatty acids profile as raw milk (table 9), and could constitute a natural source for n-3 fatty acids.In a similar way, antioxidants profile reproduced the composition of raw milk with estimated losses around 20 and 40% being higher for retinol and vitamin D3.
Higher natural antioxidants in whole milk powder lowered TBARS values in pasture (A70) samples compared to silage samples (A0) after 18-months shelf life assay (20ºC in sealed bilaminated plastic pouches under nitrogen atmosphere).
The total antioxidant activity was measured using the Ferric reduction assay (FRAP).Antioxidant compounds such as α-tocopherol, trolox, vitamin C, uric acid and billirubin, among others, are able to reduce ferric-to ferrous-tripyridyltriazine which develops a blue colour (Benzie and Strain 1996) with an adsorption maximum at 593 nm.Smet and others. (2009), demonstrated that FRAP and DPPH assays provide useful information about the oxidation process, particularly about the very early changes in the oxidative stability of milk.FRAP values for A0, A35 and A70 milk, were 332, 428 and 448 micromolar equivalents of Fe+2 respectively (P<0.05) and served as an indicator of the antioxidant capacity of milk.
In milk powder, no differences were detected, thus indicated that some thermo labile antioxidants could be destroyed due to the processing of milk.
It can be concluded that there is not a unique method to describe the antioxidant capacity of biological samples.The complexity and of mechanisms that contribute to the onset of oxidation and the mechanisms that counteract oxidative reactions involve multiple pathways.Also -tocopherol, -carotene and retinol contents were higher in pasture samples throughout the shelf life experiment.

Instrumental sensory odour
The electronic nose (E-nose) has been successfully applied to distinguish seasonal variations in whole milk powder (Biolatto and others., 2005).Also its application permitted the differentiation of virgin olive oil that showed different volatile production and vitamin E degradation patterns during the frying process (Messina and others., 2009).
The application of e-nose measurement to milk samples permitted to demonstrate that a dietary shift from silage to Lucern pasture, induced changes in the odour profile of the raw milk.
As shown in figure 4, at the beginning of the experiment (time 0), all samples were grouped differently than after turn into pasture.Also changes in the profile were detected at time 60 days, thus indicated an interaction between diet and time in the experiment, probably due to changes in the nature of the volatile compounds produced by Lucern plants as the experiment proceeded.
Data were analyzed using a Discriminant Function Analysis (DFA) and were discriminated with 86% of recognition of each sample within each group for the original cases.
A positive correlation was found for e-nose measurements and the content of different antioxidant vitamins in milk.Rossetti and others., 2010, showed than individual e-nose sensors correlated with the concentration of antioxidant vitamins in milk.The sensor called LY2/gCTI correlated positively with α-tocopherol, β-carotene and retinol, and negatively with γ-tocopherol (p<0.05).The sensor P30/1 correlated positively with retinol and the sensor P10/1 showed a positive correlation with γ-tocopherol (p<0.05).
Odour components have shown also to be transferred from raw milk to whole milk powder from cows fed 0, 35 and 70% Lucern pasture.When the value of each sensor was compared in both products, a linear relationship was found with a correlation coefficient of 0.9851 (P<0.05).In reggianito cheese, and after six months maturation, this relation was different (figure 5b), probably due to the development of odour compounds during the fermentation and maturation processes.Odor profile was determined in the head space of milk samples (3 samples per cow per time) with 3 mL of sample in a vial, incubated at 50ºC for 10 min with an agitation speed of 500 rpm (Autosampler HS100, Alpha MOS).For electronic nose analysis (Alpha Fox 4000, Alpha MOS), 1 mL of headspace sample was injected and the acquisition was obtained either with 18-semi-conductor oxide metallic sensors (MOS) or 4 mL of headspace was injected in a quadrupole mass spectrometer (Alpha Kronos, Alpha MOS) with an electronic impact (70 eV) ion source (Rossetti et al., 2010).
Further research will allow explaining the biochemical changes that occur during the transformation of milk from different productive origin; in this concern the use of an e-nose approach represents an alternative powerful tool to traditional methods of odour measurements.

Natural pigments in milk and cheese
Consumer's perception for natural and functional dairy products is an actual concern.In a survey over 179 consulted consumers, 87 % answered that would associate functional and natural concepts.In addition, 47% associated cheese with functional dairy products (other choices were yogurt, milk powder, ice-cream and dietary supplements).In this concern, the concentration of carotenoids, especially -carotene and lutein, could serve as natural pigment in milk fat and cheese.
Changes in beta carotene content were accompanied with changes in the yellow color in milk.As indicated in Calderon and others, (2007), the color index was a valuable tool to discriminate pasture from silage milk.The reflectance spectrum of raw milk samples was measured in the spectral region associated to light absorption by carotenoids, involving wavelengths from 450nm to 530nm (with a resolution of 10nm).In this range spectral data was translated to have reflectance values equal to zero at 530nm.On the translated spectrum, the integral value (IV) was calculated (Prache and others, 1999)  Where Z is the standardized value of the variables -carotene and IV ( ).
This result indicates that the transformation of spectral data in the range of carotenoids, may be used as a representative index of the feeding system.Experimental data obtained with reggianito cheese, which was elaborated within the experimental pilot plant of INTA are shown in table 5. Colour parameters indicate differences in cheese attributable to feeding management.Pasture cheese from A35 and A70 treatments presented higher b* values, which indicates a more yellow color, than A0.Differences were also distinguished by visual appreciation (figure 7).No significant differences were observed for L* parameter.

Conclusions
Productive strategies can be used to develop naturally added value dairy products.These strategies may be considered together with the consumer's perception about the role of functional and natural dairy products on human health.Preliminary studies on "pasture milk", clearly demonstrated the important changes in total lipids and fatty acid composition, especially in the contribution of C18:2 and C18:3 due to cultivar, cutting date and season.The importance of these fatty acids as substrate for CLA and n-3 PUFA concentration in milk lipids need to be considered.
The disadvantage of milk enriched with n-3 and CLA fatty acids is the possibility to suffer oxidation due to its high content of double-bonded molecules, which are prone to oxidation onset.This will not probably affect raw milk, but can bring inconvenient for the processing and commercialization of dairy product.

Figure 2 .
Figure2.Oat pasture vs. TMR feeding.Concentration of -tocopherol (○), all-trans-β-carotene (), retinol (▲) and lutein (x) in milk-fat of dairy cows fed diets differing in the percentage of oat pasture.Linear tendency for -tocopherol and retinol (dashed lines) and all-trans-β-carotene and lutein (solid lines) are shown.All groups were fed corn silage during a pre-experimental period of 6 wk.Each point represents eight cows per treatment.Data were compiled from two separated experiments with comparable diets.

Figure 4 .
Figure 4. Odor profile of ALF (70% Lucerne pasture diet) milk at 0 (time 0), 20 (time 20), 40 (time 40)and 60 (time 60) days after the shift from silage to pasture diet.Odor profile was determined in the head space of milk samples (3 samples per cow per time) with 3 mL of sample in a vial, incubated at 50ºC for 10 min with an agitation speed of 500 rpm (Autosampler HS100, Alpha MOS).For electronic nose analysis (Alpha Fox 4000, Alpha MOS), 1 mL of headspace sample was injected and the acquisition was obtained either with 18-semi-conductor oxide metallic sensors (MOS) or 4 mL of headspace was injected in a quadrupole mass spectrometer (Alpha Kronos, Alpha MOS) with an electronic impact (70 eV) ion source(Rossetti et al., 2010).

Figure 5 .
Figure 5. Relationship between e-nose sensors values in raw milk and its derived whole milk powder or reggianito cheese sample.Each point represents the value for each individual sensor applied to raw milk, and the whole milk powder or reggianito cheese that derived from the same alaboration batch.0, 35 and 70 indicates Lucerne pasture percentage in the diet.Aut: autumn; Spr: spring.
Milk Produced in Grazing Systems and Their Impact on Dairy Products 361

Figure 6 .
Figure 6.Correlation between IV and -carotene content.ALF: 70% grazed Lucerne (•) values in the region of SS: Sorghum Silage diet (○) samples corresponded to milk at the beginning of the assay (time 0)

Table 1 .
Levels of fat-soluble antioxidants in dietary components used in INTA experiments with dairy cows.

Table 2 .
Volatile terpenoid compounds in dietary components used in INTA experiments with dairy cows.

Aldedhydes (Relative Area Units) a Lucerne pasture Oat pasture Concentrate mixture Sunflower expeller Corn silage Wheat Bran 3-methyl-butanal
a Volatile aldehydes were determined as indicated in table 2.b Mean variation coefficient of all volatiles calculated for each dietary compound.

Table 3 .
Volatile aldehydes in dietary components used in INTA experiments with dairy cows.
Differential Characteristics of Milk Produced in Grazing Systems and Their Impact on Dairy Products 345

Table 6 .
Contribution (g/100 g DM) of C18:2 n-6 and C18:3 n-3 of several pastures and forage crops As expected, fatty acids profile was different among dietary components.As shown in table 7, the most abundant fatty acid in lucerne pasture was C18:3 and it can be the molecule that originates branched aldehydes, especially trans-2-hexenal, in this assay.
b mean variation coefficient of all fatty acids calculated for each dietary compound, nd: not detected.

Table 7 .
Fatty acids in dietary components used in INTA experiments with dairy cows.

Table 9 .
Mean values of antioxidant vitamins, color parameters and oxidative stability indicators in two different seasons.
the translated value corresponding to the reflectance intensity at each wavelength.Results in figure6indicate a linear relationship between IV and -carotene.The -carotene content could be predicted in function of IV by the following expression: