Livestock Research for Rural Development 27 (8) 2015 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Prediction of Total Milk Yield in early lactation of dairy cows using Milk Composition Measures

C Alphonsus

Department of Animal Science, Ahmadu Bello University, Zaria-Nigeria
mcdyems@gmail.com

Abstract

This study was conducted to determine the potential of estimating total milk yield (TMY) of dairy cows in early lactation using milk composition measures. Thirteen primiparous and fourty seven multiparous (F1) Friesian x Bunaji cows were used for the study. Cows were milked twice daily (morning and evening) and milk yield was recorded on daily basis. The milk sampled for the determination of milk fat, protein and lactose contents was taken once per week at a weekly interval from 4 to 100 days post-partum. The milk composition analysis was carried out at the Food Science and Technology Laboratory of Institute of Agricultural Research, Ahmadu Bello University, Zaria-Nigeria. Ordinary least square (OLS) regression procedure was used to determine the relationship of milk composition variables with TMY. The results showed that the single most informative milk composition variable that could be used for estimation of TMY in early lactation was milk fat content (MFC). It explained about 46.27% of the variation in TMY with very low prediction error, and has a negative association with TMY. However, when milk protein content (MPC) was included in the model, the prediction ability of the model improved (R² = 53.00%), and the combination of the three milk composition variables (MFC, MPC, MLC) explained about 58.99% of the variation in TMY of the cows. Base on the foregoing it is possible to estimate the total milk yield of dairy cows in early lactation using milk composition measures.

Keywords: Friesian x Bunaji cows, milk fat content, milk protein content, milk lactose content


Introduction

Milk composition (fat, protein and lactose content) is an important trait in dairy cattle and considerable selection pressure is placed on this trait. It determines the quality of milk produce by dairy cow and has economic value since dairy producers are paid a premium for milk of higher than average mentioned composition (Ezikwe and Machebe 2005). It determine the chemical and technological properties of the milk (Sitkowska et al 2013; Mir et al 2014).

The knowledge of the relationship between the milk composition variables and milk yield is very important in predicting the direct and correlated responses due to selection (Alphonsus and Essein 2012). Generally, the correlations between traits of economic importance is essential to; predict the change in one trait in response to selection for another; determine the feasibility of selecting for multiple traits at once; anticipate the overall results of a selection programme (Wattiaux 2002).

The relationship between milk yield and milk composition variables have been investigated by many authors (Schutz et al 1990; Stanton et al 1992; Kay et al 2005; Auldist et al 2007; Alphonsu and Essein 2012). Most of these authors reported negative correlations between milk yield and milk composition variables. It was on the basis of this relationship that this study postulate the use of milk composition variables to predict TMY of dairy cow in early lactation. It is therefore hypothesize that milk composition variables could be used to predict total milk yield of dairy cows in early lactation. One way of validating this hypothesis is to assess the relationship between the milk composition traits and TMY. A clear understanding of this relationship would assist in developing appropriate prediction models that could be used for estimation of total milk yield in early lactation without waiting for the complete lactation of the cow.

If this option has adequate accuracy, it will be an attractive one because it could provide a quick and easy way of evaluating the milk yield potential of dairy cow, and appropriate management and nutritional decision could be taken early without waiting for the complete lactation.

The objective of this study therefore, was to identify the combinations of milk composition measures that could best be used as predictors of total milk yield of dairy cows in early lactation.


Materials and Methods

Experimental site and Animals management

The study was conducted on the dairy herd of National Animal Production Research Institute (NAPRI) Shika, Nigeria, located between latitude 11° and 12°N at an altitude of 640 m above sea level, and lies within the Northern Guinea Savannah Zone (Oni et al 2001).

Thirteen primiparous and fourty seven multiparous (F1) Friesian x Bunaji cows were used for the study. The cows were raised during the rainy season on both natural and paddock–sown pasture, while hay and /or silage supplemented with concentrate mixture of undelinted cotton seed cake and grinded maize, were offered during the dry season. They had access to water and salt lick ad-libitum. Unrestricted grazing was allowed under the supervision of herdsmen for about 7 – 9 hours per day. Routine spraying against ticks and other ecto-parasites was observed.

Milk yield and composition

Cows were milked twice daily (morning and evening) and milk yield was recorded on daily basis. The time of milking was between 0630 and 0830 hours and between 1630 and 1830 hours. The daily milk yield record was used to calculate the total milk yield per lactation of the cows.

The milk sampled for the determination of fat, protein and lactose content was taken once per week at a weekly interval from 4 to 100 days post-partum. The samples collected were preserved with potassium dichromate and frozen immediately after milking and stored at -20°C until analyzed. The milk composition analysis was carried out at the Food Science and Technology Laboratory of Institute of Agricultural Research, Ahmadu Bello University, Zaria-Nigeria. The fat content was determined by Gerber method (FAO 1977) while the crude protein content was determined by Kjeldahl method (AOAC 2000). Lactose content was determined by titration method (Laboratory Manual I, 2005).

Data Analysis

The heritability (h²) of the milk composition traits were estimated by Variance Component Procedure (PROC VARCOMP) of SAS (2000). The fitted random model for paternal half-sib heritability estimation was as follows (Khan and Singh 2002)

Yij= µ +αi + eij

Where:
Yij = records of milk and fertility characteristics of cows of each sire
µ= overall mean
αi = random effect of the ith sire
eij = the uncorrelated environmental and genetic deviations attributed to individual cows within each sire group

Ordinary Least Squares (OLS) regression procedure of SAS, (2000) with multiple explanatory variables was used for the prediction of total milk yield from milk composition variables. The form of the model used was

Y = α + β1X1 + β2X2 + βiXi + ei

Where:
Y = dependent variable, α = intercept, β1, β2, and βi = regression coefficients of explanatory or independent variables X1, X2,Xi, which indicates the average change in Y that is associated with a unit change in X, and ei was the random error.


Results and Discussion

The average milk fat content (MFC), milk protein content (MPC)) and milk lactose content (MLC) in this study (Table 1) were higher than the 3.88MFC and 3.12MPC reported for Ayshire, 3.98MFC and 3.23MPC, reported for Brown Swiss and, 3.67MFC and 2.98MPC reported for Holstein (Vargas 2004). This suggested that the indigenous cows and their crosses although, may produce relatively less quantity of milk than the temperate breeds, but with higher fat and protein content than the temperate breed. It has been reported that percentage milk fat is higher in low yielding than high yielding cows (Auldist et al 2007). This is probably due to the fact that most of the temperate breeds have undergone selection for improved milk yield thereby compromising the genetic merit of the cows for milk fat and protein content.

Table 1. descriptive statistics of milk composition variables
Milk composition Mean ± SE CV (%) Minimum Maximum
Milk fat content (%) 4.30 ± 0.10 7.67 3.74 4.71
Milk protein content (%) 4.16 ± 0.07 5.05 3.89 4.45
Milk lactose content (%) 4.29 ± 0.06 4.39 3.88 4.66
CV= coefficient of variation; SE= standard error

The unfavourable relationship of the milk fat and protein content with milk yield (Table 2) has been reported by many authors (Stanton et al 1992; Kay et al 2005; Auldist et al 2007; Alphonsus and Essien 2012), and it suggested that selection for improved milk yield would decrease the genetic merit of the cows for milk fat and protein content. The observed strong negative relationship between the milk yield and fat content underscored the importance of body fat reserves as biological buffers for milk synthesis, especially during the early lactation when feed intake may not meet up the metabolic demand for the high milk yield during the early lactation (Lucy et al 2001; Buckley et al 2000).

The estimated heritability (h²) values of the milk yield and composition variables were moderate (Table 2) and were within the estimates of some previous studies (Othmane et al 2002; Kadarmideen and Wegmann 2003; Miglior et al 2007; Loker et al 2010). The fact that milk yield and milk composition traits have moderately heritability implies that these traits are heritable enough to yield significant progress from selection, and can therefore, be included in a selection index as direct or correlated traits.

Table 2. Pearson correlations and heritability (diagonal values) of milk yield and composition
Milk variables TMY MFC MPC MLC
Total milk yield (TMY) 0.316
Milk fat content (MFC) -0.680** 0.326
Milk protein content (MPC) -0.214 0.065 0.379
Milk lactose content (MLC) -0.071 0.416* -0.015 0.340
*= P< 0.05; ** = P<0.01

The evaluation criteria of the regression equations for the prediction of total milk yield (TMY) using milk composition variables in early lactation (Table 3) showed that the two bivariate equations that combine milk fat content (MFC) and milk protein content (MPC), and milk lactose content (MLC) and MFC showed better prediction ability (R²) than the model that combine MPC and MLC. Also, the combination of the three milk composition variables (MFC, MPC, and MLC) in a single multivariate equation improved the prediction ability (R²) of the model. However, the single most informative milk composition variable that showed high potential for estimation of TMY in early lactation was MFC. It explained about 46.27% of the variation in TMY and has a negative association with TMY. This suggested that high MFC in early lactation is associated with decrease in total milk yield. However, this does not mean that high milk fat percentage causes low milk yield, it is only an indication that there is simply an association between the two, whereby cows with high milk fat content during early lactation are likely to produce low total milk yield.

Table 3. Regression equations for the prediction of total milk yield using milk composition variables
Predictor variables Prediction equations Evaluation criteria
RMSEP R²/LOS
Milk fat content (MFC) Y= 87.39 – 7.456MFC 2.59 46.27*
Milk protein content (MPC) Y= 41.29 + 3.387MPC 3.45 4.50NS
Milk lactose content (MLC) Y= 60.89 – 1.267MLC 3.52 1.51NS
MFC + MPC Y= 71.09 – 7.643MFC + 3.412MPC 2.62 53.00*
MFC + MLC Y= 73.01 – 8.621MFC + 4.518MLC 2.59 51.67*
MPC + MLC Y= 46.56 + 3.371MPC – 1.211 MLC 2.68 5.05NS
MFC + MPC + MLC Y= 55.20 – 8.879MFC +4.294MPC +4.763MLC 2.61 58.99*
Y= dependent variable (total milk yield), RMSEP = root mean square error of prediction, R² = coefficient of determination (%); LOS=level of significant, * = P<0.05

A biological or physiological basis of this relationship is that in early lactation most of dairy cows experiences low or negative energy balance (NEB) due to the inability of the feed consumed by the animal to meet up the energy required for both maintenance and milk production, thus the animal is force to catabolized its body lipids through the process of lipolysis to meet the deficit. Consequently, the uptake of fatty acids mobilized from the body fat will increase resulting in an increase in fat synthesis in the udder, hence the observed high percentage milk fat. (Gutler and Schweigert, 2005; Buttchereit et al 2010; Alphonsus et al 2014). At the same time, the part of the fatty acids released are metabolized to acetil coA into the krebs cycle for energy production. The large amount of acetil coA accumulated after lipomobilization cannot be completely utilized in this metabolic path ways, because of lack of oxaloacetate resulted from carbohydrate catabolism and glucose depletion. This may results in excess production of ketone bodies and decrease in ruminal activity, consequently, post-calving clinical symptoms such as ketoacidosis and decrease in milk production may result (Rossi et al 2008). Thus, high percentage milk fat could suggest high lipolysis due to negative energy balance which could results in a decrease in milk production if appropriate measures are not taking. Therefore, the dynamics of milk fat content in early lactation could be used to monitor the energy stability of dairy cows.


Conclusion

Milk fat content is a single most informative milk composition trait that could be used to estimate total milk yield of dairy cows in early lactation, it has moderate heritability and is strongly but negatively correlated with total milk yield. It could explain about 46. 27% of the variation in total milk yield of the cow. However, addition of milk protein and lactose content to the model already containing milk fat content improved the prediction ability of the model (R² = 58.99%). Base on the foregoing, it is possible to estimate the total milk yield of dairy cow in early lactation using milk composition measures.


References

Alphonsus C and Essien I C 2012 The relationship estimates amongst milk yield and milk composition characteristics of Bunaji and Friesian × Bunaji cows. African Journal of Biotechnology. Academic Journal Publications. 11(36):8790-8793 Retrieved July, 2013, from www.academicjournals.org/AJB

Alphonsus C, Akpa G N, Nwagu B I, Orunmuyi M and Barje P P 2014 Changes in milk protein content as indicator of energy balance and fertility in dairy cows. Livestock Research for Rural Development, 26 (12) 2014. Retrieved December, 2014, from http://www.lrrd.org

AOAC 2000 Official Method 991.23 Protein Nitrogen content of milk read with 991.20 Nitrogen in milk- Kjeldahl Method and 991.21 Non Protein Nitrogen in whole milk. 17th edition.

Auldist M J, O’Brien G C D, Macmillan K L, Grainger C 2007 Effects of varying lactation length on milk production capacity of cows in pasture-based dairying systems. Journal of Dairy Science. 90: 3234-3241.

Buckley F, Dillon P, Crosse S, Fynn F and M Rath 2000 The performance of Holstein-Friesian dairy cows of high and medium genetic merit for milk production on grass based feeding systems. Livestock Production Science 64:107-119.

Buttchereit N, Stamer E, Junge W and Thaller G 2010 Evaluation of five lactation curve models fitted for fat-protein ratio of milk and daily energy balance. Journal of Dairy Science. 93: 1702-1712.

Cameron N D 1997 Selection indices and prediction of genetic merit in Animal Breeding. CAB international Walking ford U.K. pp.12-31

Ezekwe AG and Machebe N 2005 Milk composition of Muturu cattle under the semi-intensive system of management. Nigerian Journal of Animal Production. 32(2): 287-292

FAO 1977 Manual of food quality control, 14/8 / 1.5 1224- (PEI) – Determination of milk fat by Gerber method. P.8

Gürtler H and Schweigert F J 2005 Physiologie der Laktation. In: Physiologie der Haustiere. Vol 2. W. von Engelhardt and G. Breves, ed. Enke Verlag, Stuttgart, Germany. P. 552-573

Kadarmideen H N and Wegmann S 2003 Genetic parameters for body condition score and its relations with type at production traits in SWISS Holstein. Journal of Dairy Science 86: 3685 – 3693

Kay J K, Weber W J, Moore C E, Bauman D E, Hansen L B, Chester-Jones H, Crooker B A, Baumgard L H 2005 Effects of weak of lactation and genetic selection for milk yield on milk fatty acid composition in Holstein cows. Journal of Dairy Science, 88:3886-3893.

Khan F H and Singh A 2002 Principle of Genetic and Animal Breeding 1st Edition Jaypee Brothers Medical Publishers Ltd EMCA House, New Delhi, 110002 India P.262

Laboratory manual I 2005 Manual of methods of analysis of foods (milk and milk products) Directorate. Directorate general of health services Ministry of health and family welfare Government of India New Delhi. Pp.92

Loker S, Bastin C, Miglior F, Sewalem A, Fatehi J, Schaeffer L R and Jamrozik J J 2010 Genetic Parameters of body condition score and milk production traits in Canadian Holstien

Lucy M C 2001 Reproductive loss in high-producing dairy cattle: where will it end? Journal of Dairy Science 84, 1277-1293

Miglior F, Kemp R A and Burnside E B 1994 Genetic parameter Estimates of conformation and performance traits in Station- tested Limousin bulls. Canadian Journal of Animal Science. 74(2):379 – 383

Mir S N, Ullah O and Sheikh R 2014 Genetic polymorphism of milk protein variants and their association studies with milk yield in Sahiwal cattle. African Journal of Biotechnology 13(4):555-565

Othmane M H, De Lafuente L F, Carriedo J A and San Primitivo 2002 Heritability and Genetic correlation of test day milk yield and composition, Individual Laboratory Cheese yield and somatic cell count for dairy ewes. Journal of Dairy Science, 85:2692-2698

Oni O O, Adeyinka I A, Afolayan R A, Nwagu B I, Malau-Aduli A E O, Alawa C B I and Lamidi O S 2001 Relationships between milk yield, post-partum body weight and reproductive performance in Friesian x Bunaji Cattle. Asian–Australian Journal, 14 (11):1516 – 1519.

Rossi F, Righi F, Romanelli S, Quarantelli A 2008 Reproductive efficiency of dairy cows under Negative energy balance conditions. Annals of Faculty of Veterinary Medicine di Parma, 28: 173 - 180

SAS 2000 SAS User’s Guide Version 8.1. Statistical Analysis system institute Inc, Cary, Nc, USA.

Schutz M M, Hansen L B, Steuernagel G R, Reneau J K, Kuck A L 1990 Genetic parameters for somatic cells,protein, and fat in milk of Holsteins. Journal of Dairy Science, 73:494-502

Sitkowska B, Neja W, Milczewska A, Mroczkowsk S and Markowska A 2013 Milk protein polymorphisms and effect of herds on cows`milk composition. Journal of Central European Agriculture, 2013, 14(1):78-90

Stanton T L, Jones L R, Everett R W, Kachman S D 1992 Estimating milk, fat and protein lactation curves with a test day model. Journal of Dairy Science, 75: 1691–1700.

Vargas G 2004 What Can We Learn From Fat and Protein Levels and Ratios? Dairy and Animal Science Bulletin.

Wattiaux M 2002 Technical dairy guide: Reproduction and Genetic Selection. The Babcock Institute for International Dairy Research and Development. University of Wisconsin, Pp. 111-132.


Received 28 April 2015; Accepted 10 June 2015; Published 1 August 2015

Go to top