The effect of cattle breed , season and type of diet on nitrogen fractions and amino acid profile of raw milk

The aim of the study was to describe the differences in nitrogen fractions and deepen the knowledge in the amino acid profile of raw milk affected by the breed of cattle, season and type of feeding. The study was conducted from June 2005 to February 2007 on 64 bulk milk samples collected from eight herds consisting of Czech Fleckvieh (four herds) and Holstein (four herds) breed. One half of the herds of each breed was grazed while the other half was not. Samples were collected twice in winter and twice in summer. The effect of the breed resulted in differences in milk yield that was lower in Czech Fleckvieh (5 385.50 kg) than in Holstein (7 015.15 kg, P<0.05). The content of nitrogen fractions was higher in Czech Fleckvieh than in Holstein (P<0.05). No effect of the breed on the amino acid profile of milk was observed except on the concentration of Glu (P<0.05). The effect of the season was demonstrated in the decrease of the concentrations of nitrogen fractions and Met during summer in comparison to winter (P<0.05). The effect of the type of feeding resulted in lower milk yield (5 197.50 and 7 203.75 kg) and lower concentrations of nitrogen fractions in grazed herds compared to non-grazed herds (P<0.05), respectively. Furthermore, the amino acid profile of milk differed significantly between grazed and non-grazed herds (P<0.05).


Introduction
In recent years, the demand for food with high nutrient value and high sensory quality has increased.High quality milk is important for the production of high quality pasteurised market milk and dairy products (Elmoslemany et al. 2010).Thus the knowledge of the composition of raw cows' milk is of the greatest importance not only for the dairy industry but also for consumers.Not only the economic impact of variation in fat and protein content has led to much research on manipulation and alteration of these constituents, but also the different amino acid, fatty acid, micronutrient profiles and their consequent impact on health have become current topics of debate.Increased knowledge of the impact of feeding on the quality and quantity of milk production has led to more sophisticated diet formulations for cattle (Schönfeldt et al. 2012, Jenkins & McGuire 2006).Thus, the determination of the AA profile in milk is essential for qualitative evaluation of peptides and proteins that can affect the chemical and nutritional properties of milk (Marino et al. 2010).
The nitrogen fractions of milk can be broadly divided into three categories; casein (CAS), whey proteins (WP) and non protein nitrogen (NPN).Casein comprises the majority of nitrogen in milk (about 78 %) and is a determinant of the technological usability of milk while WP (17 %) are an indicator of the positive health value of milk for humans (Litwińczuk et al. 2011).The smallest fraction, NPN, represents 5 % of the nitrogen in milk (Jenkins & McGuire 2006).The variation in nitrogen fractions of milk affects consistency, yield and quality of dairy products (Mehra et al. 1999).Although genetics, in particular milk protein polymorphisms, play a major role in regulating milk protein composition (Bobe et al. 1999, Graml & Pirchner 2003) less conclusive effects of other factors such as breed (e. g.Chrenek et al. 1987), season (e. g.Lindmark-Månsson et al. 2003, Kriščiunaite et al. 2012), nutrition (e. g.Hadrová et al. 2007, Cant et al. 2003, Bargo et al. 2002) or animal health (e. g.Csapó et al. 1995) have been documented in literature.Therefore, altering milk protein composition by selective breeding or nutrition could be useful for the dairy industry (Bobe et al. 2009).
The aim of the study was to determine the effect of cow breed, season and type of diet on the content of nitrogen fractions and amino acid profiles of raw milk.

Material and methods
The study was conducted from June 2005 to February 2007 on bulk milk samples collected from eight commercial dairy herds consisting of Czech Fleckvieh (four herds) and Holstein (four herds) dairy cows.The average herd size was 185 ± 149 (from 66 to 439) dairy cows and the average milk yield was 6 200.6 ± 1 455.1 (from 3 836 to 8 124) kg (see Table 1 also for details about selected farms environmental characteristics).Cows were fed diets typical in the Czech Republic which consist of maize silages, clover-grass haylages, hay, locally available feedstuffs, concentrate and mineral mixtures according to relevant milk yield and standard requirements.One half of the herds of each breed was grazed during summer season while the other half was not as given in Table 2. Cows were milked twice a day and samples were collected regularly two times in the winter and two times in the summer period resulting in a total of 64 bulk milk samples examined.

Analyses
The crude and true protein and casein (CP, TP and CAS) were determined by reference Kjeldahl's method using the instrument line Tecator with Kjeltec Auto Distillation unit 2 200 (Foss-Tecator AB, Höganäs, Sweden) according to ČSN 57 0530.The instrument was included in international proficiency testing (APLAC and ICAR-CECALAIT) with regularly mostly success ful results.Milk amino acid profile was determined as described in Hadrová et al. (2007).Briefly, samples of lyophilised milk for acid hydrolysis were hydrolysed with 6 mol/L HCl for 24 h at 110 °C.Sulphur amino acids were determined as cysteic acid and methionine-sulphone.All hydrolysates were separated in the automatic amino analyser AAA 400 (Ingos, Prague, Czech Republic) using Na citrate buffer system and quantified by reaction with ninhydrin.For determination of the amino acid content the programme ChromuLan v. 0.7 (http://www.chromulan.org)was used.There were investigated 17 amino acids in the samples.The results were expressed absolutely in g/kg and relatively in %.

Statistical analysis
The GLM procedure of the SAS v. 9 programme package (SAS Institute Inc., Cary, NC, USA) was used for the calculation.Multi-factor analysis of variance with fixed effects as breed, season and feeding was used for statistical evaluation of data set according to following model: where y ijk is the independent variable (investigated milk indicator), μ is the general mean, b i is the effect of breed (i=2), s j is the effect of season (j=2), f k is the effect of feeding (k=2), and e ijk is the random effect.

Results and discussion
The composition of raw milk, especially the content of nitrogen fractions determines, to a large extent, the yield, the nutritional value and technological properties of a number of dairy prod ucts (Heck et al. 2009, Mehra et al. 1999).Thus, variations in milk composition are of a great importance to the manufacturers of dairy products.Recently, several studies have shown a large variation in the relative concentration of the major milk proteins (Bobe et al. 1998, Bordin et al. 2001, Heck et al. 2008) but it has been found that a considerable part of this variation is genetic (Bobe et al. 1999, Graml & Pirchner 2003).The role of nutrition on milk protein com posi tion is also documented in the literature (recently e. g.Bobe et al. 2009, Hadrová et al. 2007, Doepel et al. 2004), however results of these studies are inconsistent and the role of nutrition on manipulating protein composition of milk is less conclusive (Bobe et al. 2009).
Milk contains a mixture of proteins, each having unique attributes for nutritional, bi olog ical and human food ingredient applications (Smithers et al. 1996).The role of milk and dairy products in human nutrition and their impact on consumers' health is one that has been hotly debated by both the food and health industries.However, this discussion often fails to mention the fact that dairy products are among the most versatile protein sources available with a unique amino acid composition (Dunshea et al. 2007).Although milk protein polymorphisms play a major role in regulating the milk protein composition as documented by e. g.Graml & Pirchner (2003), in a recent study Lindmark-Månsson et al. (2003) reported that the amino acid profile of milk not only differs among countries, but also within the same country and between seasons, probably due to feeding changes.From this point of view knowledge concerning not only content and proportion of nitrogen fractions but also amino acid composition of milk is valuable for the assessment of milk quality and its nutritional and manufacturing value (Guy & Fenaille 2006).
To our knowledge, only a very limited number of studies have analysed the amino acid profile of raw milk and the concentration of nitrogen fractions in relation to breed, season and type of feeding.

Effect of the breed
Milk yield and concentrations of nitrogen fractions in dependence on breed, season and type of feeding are presented in Table 3. Milk yield in Czech Fleckvieh breed was 5 385.50 kg and was lower than the yield determined in Holstein breed (7 015.15 kg, P<0.05).Results obtained in our study are in agreement with data describing milk yield of these two breeds in the Czech Republic environment (e. g.Wolfová et al. 2007, Hanuš et al. 2010).There were several significant differences in the content of nitrogen fractions between the two breeds.Czech Fleckviehs produced significantly (P<0.05)higher concentrations of CP, CAS, TP and NPN than Holsteins.The content of WP did not differ significantly between breeds.Similar findings were reported by e. g.Hanuš et al. (2010) and Wolfová et al. (2007).Furthermore, values determined in Holstein are comparable with the data reported in the literature such as Lindmark-Månsson et al. (2003), Kriščuinate et al. (2012), Jõudu et al. (2008) or Wedholm et al. (2006) for this breed.The concentration (g/kg) of amino acids in milk in dependence on cattle breed, season and type of diet are given in Table 4.The total content of amino acids in milk in Czech Fleckvieh was 30.868 g/kg and in Holstein 30.078 g/kg and it was not affected by the cattle breed (P>0.05).The concentration of essential amino acids and non-essential amino acids in milk was similar in both breeds and did not differ significantly (P>0.05).The tendency was found for the total content of semi-essential amino acids (Cys+Tyr), that it was higher in Czech Fleckvieh, being 1.670 g/kg, than in Holstein (1.596 g/kg, P=0.0971).From all individual amino acids the highest concentration of Glu was found in both breeds followed by Pro, Leu and Lys.For the individual essential amino acids no effect of the breed was observed except for the concentration of Glu that was higher in Czech Fleckvieh in comparison to Holstein (P<0.05) and for Tyr that tended to be higher in Czech Fleckvieh than in Holstein (P=0.083).
Our results are comparable with the values published in the literature (Schönfeldt et al. 2012, Lindmark-Månsson et al. 2003, Swaisgood 1995, Davis et al. 1994) confirming the genetical background of amino acid composition of milk proteins with very small breed differences.However, minor differences in the amino acid profile of milk between two cattle breeds and their crosses were mentioned by Chrenek et al. (1987) or Mapekula et al. (2011).

Effect of the season
As well documented in previous studies milk composition varies considerably throughout the seasons (e. g.Jahreis et al. 1996, Auldist et al. 1998, Lindmark-Månsson et al. 2003, Lock & Garnsworthy 2003, Litwińczuk et al. 2011).In our study, during the summer concentration of CP, CAS, TP and WP was lower than in the winter (P<0.05).On the other hand samples taken in the summer contained higher proportion of NPN in comparison to the winter (P<0.05).Similar findings were also reported by Kriščiunaite et al. (2012), Litwińczuk et al. 2011, Heck et al. (2009) who noted the lowest content of milk protein in summer and the highest in autumn/winter or by Lindmark-Månsson et al. (2003); however their milk differed in profile of nitrogen fractions.The mean content of TP in Swedish milk was 3.37 g/100 g and ranged from 3.32 g/100 g found in May to 3.45 g/100 g measured in September while TP content determined in our study was lower, being 3.06 g/100 g in summer and 3.17 g/100 g in winter.
On the other hand the concentration of CAS closely resembled the values of raw Swedish dairy milk as determined in the above mentioned study.The concentration of WP in our study ranged from 0.48 g/100 g found in summer to 0.51 g/100 g found in winter and was lower than that determined by Lindmark-Månsson et al. (2003).Similarly, the concentration of NPN being 0.20 g/100 g in summer and 0.18 g/100 g in winter was lower than given in Lindmark-Månsson et al. (2003).This discrepancy can be explained by a long-term decrease in the CAS content and a corresponding increase in the content of WP in Swedish milk between 1970 and 1996 that indicate a significant change in the proportion of proteins in milk (Lindmark-Månsson et al. 2003).
The concentration of essential, semi-essential, and non-essential amino acids was not affected by the season (P>0.05).The content of individual amino acids found in our study was slightly lower than that determined by Lindmark-Månsson et al. (2003) in Swedish dairy milk with a CP content of 3.67 g/100 g.The concentration of individual amino acids did not differ significantly except for Met whose content was lower during the summer (0.887 g/kg) in comparison to the winter (1.038 g/kg, P<0.05).Further, the concentration of Phe tended to be lower (P=0.0827)and the concentration of Ala tended to be higher (P=0.0652) in summer than in winter.This is in disagreement with Lindmark-Månsson et al. (2003) who determined seasonal variations in all amino acids except for Pro and Tyr.This discrepancy can be caused by differences in the profile of nitrogen fractions in milk as mentioned above.Furthermore, the concentration of free amino acids can also contribute to differences found in the concentrations of total amino acids (Lindmark-Månsson et al. 2003, Csapó et al. 1995).The total of amino acids in milk comprises amino acids incorporated into proteins as well as free amino acids that can be derived from different sources such as from normal metabolic processes, from substances specific to certain kinds of feedstuffs that pass in small quantities from the digestive tract to the milk or from processes connected with health status of cows because the content of free amino acids is highly correlated with udder inflammation (Csapó et al. 1995).

Effect of the feeding type
Milk yield in grazed herds was lower than in non-grazed herds being 5 197.50 and 7 203.75 kg, respectively (P<0.05).This is in agreement with other pasture-based studies (e. g. White et al. 2002, White et al. 2001, Kolver et al. 2000, Kolver & Muller 1998).The differences in milk production between grazed and non-grazed herds are likely related to differences in energy requirements for maintenance related to walking and grazing activity (Agnew et al. 2000), to differences in energy intake that is a primary limiting nutrient for high-yielding dairy cows on pasture (Bargo et al. 2002) and to differences in amounts, sources and ratio of rumen degrad able protein and rumen undegraded protein for grazing cows (Eastridge 2006, Kolver & Muller 1998, DePeters & Cant 1992).As mentioned by Kolver & Muller (1998) the lower milk production from high producing dairy cows consuming only high quality pasture compared with cows consuming a nutritionally balanced total mixed ration (TMR) was due to a lower dry matter intake and energy intake.
The content of all studied protein fractions, which consists of CP, CAS, TP, WP and NPN in milk of grazed herds was lower than in non-grazed herds (P<0.05).Similar results were noted by Bargo et al. (2002) when comparing pasture, TMR and partial TMR based feeding systems while Jõudu et al. (2008) found a significant effect of pasture only on total CAS and some CAS fractions.On the other hand White et al. (2001) or Kriščiunaite et al. (2012) did not find any differences in the protein percentage between pasture and confinement cows or indoor and outdoor housing period, respectively.These contradictory results may have arisen from differences in the experimental design (Jõudu et al. 2008).
A total concentration of amino acids in milk of grazed herds being 26.690 g/kg was lower than in non-grazed herds -31.252 g/kg (P<0.05).Similarly to the overall sum of amino acids, the con cen tra tion of essential, non-essential and semi-essential amino acids was lower in grazed than in non-grazed herds (P<0.05).Most of the amino acids showed significant differ ences, re flecting the above mentioned variation in the major milk proteins.The individual amino acid concentrations of the following amino acids were found to be lower in grazed than in non-grazed herds: Thr, Val, Ile, Leu, Phe, His, Lys, Tyr, Asp, Ser, Glu, Ala, Arg (P<0.05),further con cen tra tion of Cys also tended to be lower in grazed in comparison to non-grazed herds (P=0.0818).

Table 1
Basic information about environment of dairy cow herds raw milk (according to ČSN EN ISO/IEC 17025) in the Research Institute for Cattle Breeding, Ltd. in Rapotin (Czech Republic) for analyses.Data concerning daily milk performance were obtained from animal records kept from milk recording kept by the Czech Moravia Breeders Corporation.

Table 3
Milk yield and protein fractions as affected by the breed, season and type of feeding