The overall goal of this study was to investigate milk flow traits in Italian Holstein-Friesian cows and, in particular, the bimodality of milk flow, defined as delayed milk ejection at the start of milking. Using a milkometer, 2,886 records were collected from 133 herds in northern Italy from 2001 to 2007. All records included 5 time-period measurements for milk flow, somatic cell score (SCS), milk yield, 8 udder type traits, and the presence or absence of bimodality in milk flow. Genetic parameters were estimated using linear animal models for continuous traits such as milk flow, udder type, SCS, and milk production, whereas bimodality was analyzed as a categorical trait. With the exception of decreasing time (which had a very small heritability value of 0.06), heritability values for milk flow traits were moderate, ranging from 0.10 (ascending time) to 0.41 (maximum milk flow). In addition, moderate to high genetic correlations were estimated between total milking time and other time measures (from 0.78 to 0.87), and among time flow traits (from 0.62 to 0.91). The decreasing time was the trait most genetically correlated with udder type traits, with correlation values of 0.92 with rear udder height, 0.85 with rear udder width, and 0.73 with teat placement. Large udders with strong attachments were also associated with greater milk production. Heritability estimated for bimodality was 0.43, and its genetic correlation with milk flow traits and SCS indicated a sizable genetic component underlying this trait. Bimodality was negatively associated with milk production; shorter milking times and greater peak milk levels were genetically correlated with more frequent bimodal flows, indicating that faster milk release would result in an increase in bimodal patterns. The negative genetic correlation of bimodality with SCS (-0.30) and the genetic correlation between milk flow traits and SCS suggest that the relationship between milkability and SCS is probably nonlinear and that intermediate flow rates are optimal with respect to mastitis susceptibility. Quicker milk flow over a shorter period would increase the frequency of bimodal curves in milking, whereas the correlation between bimodality and both ascending and descending time was less clear.

BackgroundMastitis is a major disease of dairy cattle occurring in response to environmental exposure to infective agents with a great economic impact on dairy industry. Somatic cell count (SCC) and its log transformation in somatic cell score (SCS) are traits that have been used as indirect measures of resistance to mastitis for decades in selective breeding. A selective DNA pooling (SDP) approach was applied to identify Quantitative Trait Loci (QTL) for SCS in Valdostana Red Pied cattle using the Illumina Bovine HD BeadChip.ResultsA total of 171 SNPs reached the genome-wide significance for association with SCS. Fifty-two SNPs were annotated within genes, some of those involved in the immune response to mastitis. On BTAs 1, 2, 3, 4, 9, 13, 15, 17, 21 and 22 the largest number of markers in association to the trait was found. These regions identified novel genomic regions related to mastitis (1-Mb SNP windows) and confirmed those already mapped. The largest number of significant SNPs exceeding the threshold for genome-wide significant signal was found on BTA 15, located at 50.43-51.63 Mb.ConclusionsThe genomic regions identified in this study contribute to a better understanding of the genetic control of the mastitis immune response in cattle and may allow the inclusion of more detailed QTL information in selection programs.Electronic supplementary materialThe online version of this article (doi:10.1186/s12863-014-0106-7) contains supplementary material, which is available to authorized users.

A total of 137,753 test day records of 20,745 Italian Brown Swiss dairy cows from 26 provinces of Italy were used to estimate heritability for casein and urea content in milk and their genetic correlations with other production traits and milk somatic cell score. Milk component values were obtained by Fourier Transformed Infrared (IR) Spectroscopy from milk samples collected during national routine recording and were analysed using test day repeatability animal models. Fixed effects included 1,001 levels of herd-test date, 15 classes of days in milk, and 13 classes of age at calving within parity. The variation among cows was large for most of the traits. The heritability value for casein content was 0.31, as for protein content, and genetic and phenotypic correlations between these two traits were large (0.99 and 0.97 respectively). Milk urea content had a heritability of 0.17 and a positive genetic relationship with fat (0.12), null with protein (0.03) and casein (0.002) content and a negative genetic correlation with milk yield (-0.17) suggesting that the genetic improvement for milk urea content would be possible, but genetic gain would be affected by other traits included as selection criteria in the economic index and by their relative economic emphasis

The objectives of this study were to estimate genetic parameters for body weight (BW) and BW change (BWC) and genetic correlations of BW and BWC with diseases and genomic predicted transmitting abilities (PTA) of productive and conformation traits of Holsteins during the first 120 DIM. Daily BW data were from the Afiweigh cow body weighing system (SAE Afikim, Kibbutz Afikim, Israel), which records BW as a cow exits the milking parlor. Disease categories included metabolic diseases, ketosis, infectious diseases, mastitis, reproductive diseases, and other diseases. Edited data included 68,914 and 11,615 daily BW observations from 441 Pennsylvania State University and 72 Virginia Tech Holstein cows, respectively. Two-trait random regression models were used to estimate relationships between BW, BWC, and diseases at 25, 38, and 58 mo of age at calving. Fixed effects for BW were age at calving nested within lactation group, week of lactation, and herd date; random effects for BW included animal, permanent environment, and error. Fixed effects for disease were herd-year-season of calving and age at calving nested within lactation group; random effects for disease were animal, permanent environment (for mastitis only), and error. Correlations of PTA for BW and BWC with genomic PTA for productive and type traits were also estimated with data from 117 cows. Heritability estimates for daily BW ranged from 0.34 to 0.63. Greater BW and less BWC were favorably correlated with ketosis, metabolic diseases, infectious diseases, and other diseases. The genetic correlation estimate between BW and ketosis was strongest at 60 DIM (-0.51), and genetic correlation estimates at 60 DIM with metabolic diseases (-0.52), infectious diseases (-0.81), and other diseases (-0.48) followed the same trend as ketosis. The genetic correlation estimate between BWC and ketosis was strongest for the change from 5 to 20 DIM (0.70) and was similar for metabolic diseases (0.37), infectious diseases (0.74), and other diseases (0.49). Correlations of BW and BWC with reproductive diseases tended to be in the reverse direction of those reported for ketosis. A larger PTA for BW was significantly correlated with smaller genomic PTA for milk yield, dairy form, rear udder height, and udder cleft. Predicted transmitting ability for BWC was negatively correlated with genomic PTA for protein percentage, strength, and hip width (ranging from -0.26 to -0.13 across lactation) and was positively correlated with dairy form, rear udder height, and udder cleft (ranging from 0.20 to 0.37 across lactation). Selection for reduced BW loss can be implemented with automated body weighing systems and may be successful in decreasing disease incidence in the early stages of lactation.

A selective DNA pooling approach was applied to identify QTL for conjugated linoleic acid (CLA), vaccenic acid (VA) and Δ(9) -desaturase (D9D) milk content in Italian Brown Swiss dairy cattle. Milk samples from 60 animals with higher values (after correction for environmental factors) and 60 animals with lower values for each of these traits from each of five half-sib families were pooled separately. The pools were genotyped using the Illumina BovineSNP50 BeadChip. Sire allele frequencies were compared between high and low tails at the sire and marker level for SNPs for which the sires were heterozygous. An r procedure was implemented to perform data analysis in a selective DNA pooling design. A correction for multiple tests was applied using the proportion of false positives among all test results. BTA 19 showed the largest number of markers in association with CLA. Associations between SNPs and the VA and Δ(9) -desaturase traits were found on several chromosomes. A bioinformatics survey identified genes with an important role in pathways for milk fat and fatty acids metabolism within 1 Mb of SNP markers associated with fatty acids contents.