Search

Among Italian dairy cattle, the Holstein is the most reared breed for the production of Parmigiano Reggiano protected designation of origin cheese, which represents one of the most renowned products in the entire Italian dairy industry. In this work, we used a medium-density genome-wide data set consisting of 79,464 imputed SNPs to study the genetic structure of Italian Holstein breed, including the population reared in the area of Parmigiano Reggiano cheese production, and assessing its distinctiveness from the North American population. Multidimensional scaling and ADMIXTURE approaches were used to explore the genetic structure among populations. We also investigated putative genomic regions under selection among these 3 populations by combining 4 different statistical methods based either on allele frequencies (single marker and window-based) or extended haplotype homozygosity (EHH; standardized log-ratio of integrated EHH and cross-population EHH). The genetic structure results allowed us to clearly distinguish the 3 Holstein populations; however, the most remarkable difference was observed between Italian and North American stock. Selection signature analyses identified several significant SNPs falling within or closer to genes with known roles in several traits such as milk quality, resistance to disease, and fertility. In particular, a total of 22 genes related to milk production have been identified using the 2 allele frequency approaches. Among these, a convergent signal has been found in the VPS8 gene which resulted to be involved in milk traits, whereas other genes (CYP7B1, KSR2, C4A, LIPE, DCDC1, GPR20, and ST3GAL1) resulted to be associated with quantitative trait loci related to milk yield and composition in terms of fat and protein percentage. In contrast, a total of 7 genomic regions were identified combining the results of standardized log-ratio of integrated EHH and cross-population EHH. In these regions candidate genes for milk traits were also identified. Moreover, this was also confirmed by the enrichment analyses in which we found that the majority of the significantly enriched quantitative trait loci were linked to milk traits, whereas the gene ontology and pathway enrichment analysis pointed to molecular functions and biological processes involved in AA transmembrane transport and methane metabolism pathway. This study provides information on the genetic structure of the examined populations, showing that they are distinguishable from each other. Furthermore, the selection signature analyses can be considered as a starting point for future studies in the identification of causal mutations and consequent implementation of more practical application.

The aggregate fertility index was added to the national Holstein breeding objective in 2009, and is a linear combination of five selection criteria: angularity, milk yield (MY), interval calving-first service (DTFS), 56 days non-return rate (NR56), and calving interval. The objective of the present study was to include new traits, measured in both productive cows and heifers, to such aggregate index, by exploiting information from linear scoring and fertility-related events. Six subsets of ~15,000 animals calving since 1994 were extracted from the national database. Traits identified were: age at first insemination in heifers (AFI), interval first-last insemination (IFL), NR56, and conception rate (CR) in heifers and cows, and DTFS, MY, and body condition score (BCS) in cows. A multiple-trait animal model was employed to estimate (co)variances of these phenotypes, separately for cows and heifers. Fixed effects per trait were: herd-year of birth (AFI), month of birth (AFI), herd-year-season of birth (IFL, NR56, CR in heifers), month-year of insemination (IFL, NR56, CR in heif-ers), herd-year-season of calving (MY, DTFS, IFL, NR56, CR in cows), month of calving (DTFS, IFL, CR in cows) or insemination (NR56 in cows), age-year of calving (DTFS, IFL, NR56, CR in cows), herd-year-season of classification (BCS), age-stage of lactation at linear scoring (BCS), and year of calving (BCS). Animal additive genetic and residual were treated as random terms. The pedigree was traced back 5 generations. Estimates of G and R matrixes of all the six subsets were averaged to apply selection index methodology for deriving appropriate index weights, by fixing CR in cows and heifers as the only breeding goals. Heritability of cow fertility traits ranged from 0.013 (NR56) to 0.076 (DTFS), whereby estimates of heifer traits were between 0.012 (CR) and 0.071 (AFI). Relative emphasis placed in the selection criteria of the aggregate index for cows were: 0.55 (IFL), 0.16 (DTFS), 0.13 (NR56), 0.09 (MY), and 0.07 (BCS). Index weights for heifer traits were 0.47 (IFL), 0.43 (NR56), and 0.10 (AFI). Both indexes were scaled to have mean 100 and standard deviation 5. The aggregate fertility index published in the official genetic evaluation release is a combined index which includes both indexes by placing 90% emphasis on cow aggregate index and 10% emphasis on heifer aggregate index.

The aggregate index for female fertility was introduced into the national Holstein cattle breeding objectives in 2005, and included five selection criteria, namely days from calving to first service (DTFS), non-return rate at 56 d from first insemination (NR56), calving interval (CI), angularity (ANG), and equivalent-mature milk yield (MY). Conception rate at first service (CR) was the only breeding goal. The objective of the present study was to revise this aggregate index by including new information from both linear scoring (body condition score, BCS), services and pregnancy testing (interval from first to last insemination, IFL). After edits, information was available on 5,466,546 primiparae calving since 1994; to reduce computational time, six subsets of approximately 15,000 cows were extracted. (Co)variance components of DTFS, NR56, IFL, MY, BCS, and CR were estimated simultaneously with a multi-trait animal model, which included the fixed effects of herd-year-season of calving (DTFS, NR56, IFL, MY, and CR), age-year of calving (DTFS, NR56, IFL, CR), month of calving (DTFS, IFL, CR), month of insemination (NR56), herd-year-season of linear scoring (BCS), age-stage of lactation at linear scoring (BCS), and year of calving (BCS). Random terms were animal additive genetic and the residual. At a national level, means of DTFS, NR56, IFL, BCS, MY, and CR were 88 d, 59%, 71 d, 2.98, 10,483 kg, and 35%, respectively. (Co)variances estimated in each subset were averaged to derive G and R matrixes. Heritability estimates of fertility traits ranged from 0.013 (NR56) to 0.076 (DTFS); heritability estimates of BCS and MY were 0.18 and 0.29, respectively. Selection index methodology was employed to derive appropriate index weights from (co)variances between breeding goal (CR) and (among) selection criteria. Results from the present study indicated that the genetic response of the breeding goal using information on BCS and IFL, substituting ANG and CI, respectively, increases by 5.6% after one selection round.

Jersey dairy cattle breed is characterized by relatively low milk yield (MY), but high milk solids content. The low productivity and its udder conformation make Jersey a suitable breed for once-a day-milking, resulting in less stress load for animals and higher milk quality. For these reasons Jersey is appreciated by the dairy industry. The present study aimed to assess the phenotypic variation of MY and milk quality traits in Italian Jersey (IJ), and to evaluate the heterogeneity of variance across IJ herds. Sources of variation were investigated through a mixed linear model, including the fixed effects of days in milk, parity, calving season, milking frequency, recording type, the interaction between days in milk and parity, and the random effects of herd-test-day (HTD) and cow. Heterogeneity of variance was assessed across herds classified as low (L), medium (M) or high (H) according to their average MY and milk quality traits. Results highlighted the high contents of milk fat (5.18%), protein (4.08%) and casein (3.16%) of IJ cows. Somatic cell score (SCS) averaged 3.35 units, which is quite high and suggest that SCS should be monitored carefully in IJ herds, especially through an improvement of management practices. Milk yield showed moderate negative correlations with milk composition and SCS. Milk with greater fat content had also greater protein and casein content, whereas SCS resulted moderately and negatively correlated with lactose content. Phenotypic variances of MY and milk quality traits were mainly related to animal effect; on the other hand phenotypic variance of milk urea nitrogen content was mainly affected by HTD effect. Finally, heterogeneity of variance was observed across L, M and H herds, in a way that highly ranked herds showed greater standard deviation. Future research should investigate the impact of accounting or not for heterogeneity of variance in genetic evaluation of IJ breed.

A new breeding value estimation for mastitis resistance has been implemented in the Italian Holstein dairy genetic evaluation. This breeding value uses indicators derived from test-day somatic cell counts (SCC) in order to predict the breeding value for resistance to mastitis occurrence. The pattern of test-day SCC contains additional information on mastitis resistance beyond its mean value usually used. These breeding values are combined in an “udder health” index, using different weights based on (co)-variance structure among indicators and direct mastitis observations. This new index will be additional to the already used somatic cell score index. The correlation between the two indexes is estimated at 80% meaning that the two are in the same direction, but they differ in terms of ranking individuals. In order to set up the new index, several indicators, derived from test-day somatic cell counts, have been derived. Alternative SCC traits were: mean and standard deviation of somatic cell score within lactation (SCSt and SD_SCSt, respectively), between 5 and 150 DIM (SCS150 and SD_SCS150, respectively), and between 151 and 305 DIM (SCS305 and SD_SCS305, respectively); infection, a dichotomous trait indicating that at least one SCC test-day record was above 100,000 cells mL-1 within lactation; severity, the ratio between the number of test-days with SCC greater than 100,000 cells mL-1 and the total number of test-days; severity, the ratio between the number of test-days with SCC greater than 400,000 cells mL-1 and the total number of test-days; and cell’s peak defined as the number of peaks during lactation (number of times when SCC shows a change from 400,000 mL-1 on three consecutive test-days). Once indicators traits have been defined, these have been validated on a sample data-set with direct mastitis information. Multiple trait animal model has been applied and genetic parameter for these traits including direct mastitis have been estimated. The highest genetic correlations between mastitis itself and indicators were with SCS150, SD_SCS, severity and peak equal to 0.39, 0.44, 0.41 and 0.51 respectively. Selection index methodology was applied to estimate appropriate weights to combined these four traits in the aggregate udder health index. This index has been published for the first time during December 2017 evaluation with mean 100 and standard deviation 5. Initially this index will be published only for national and international bulls. Following steps are to set up the genomic evaluation and, also, to participate to the GMace evaluation.

Feed efficiency in lactating cows is one of the major aspects in modern dairy farm management. Purpose of the present study was to identify sources of variation of body live weight (LW) and feed efficiency traits in a population of 8,516 primiparous Jersey cows. The variables tested included LW, fat corrected milk (FCM), predicted dry matter intake (pDMI), energy corrected milk (ECM), and predicted feed efficiency (pFE_ECM), defined as ECM/pDMI. Linear scoring and age at evaluation were used to estimate LW. Factors associated with such traits were investigated using linear mixed models which included the fixed effects of origin of the paternal grandsire (USA, Denmark, Canada), stage of lactation, year of birth, and age at first calving. Random terms were cow, contemporary group (herd-year-season at linear scoring), and the residual. Mean of LW and pFE_ECM were 435 kg and 1.36, respectively. Feed efficiency was weakly negatively correlated (-0.15) with LW, and pDMI was positive correlated with ECM (0.88). Values of LW ranged from 444.01 to 449.88 kg in cows born between 2006 and 2014. Furthermore, cows with paternal grandsire born in Canada were the heaviest, but they had the lowest values for FCM, pDMI, ECM and pFE_ECM. Feed efficiency, FCM, ECM, and pDMI increased in the first three months of lactation, and then decreased. Kilos of pDMI and FCM increased concurrently with age at first calving and systematically decreased as year of birth proceeds. Energy corrected milk of cows born between 2011 and 2014 was lower than ECM of animals born between 2008 and 2010, consequently resulting in lower pFE_ECM. Late calving Jerseys were less feed efficient compared to early calving contemporaries. Systematic environmental factors associated with feed efficiency will be used to adjust phenotypes for routine genetic evaluation.

Mastitis is one of the most costly diseases in the dairy herd. The direct collection of data on clinical mastitis is not widely implemented worldwide and thus most countries refer to somatic cell score (SCS), i.e. the log transformation of milk somatic cell count (SCC, cells mL-1), as indirect measure to select against mastitis. Recently, alternative SCC traits have been proposed to improve udder health. The aim of this study was to estimate heritabilities of alternative SCC traits in Italian Holstein dairy cattle breed. The dataset of SCC was retrieved from the Italian Holstein Association (ANAFI) and included test-day records of first lactation cows (20 to 40 mo of age) calving between 1999 and 2014 in herds of Padova province. The dataset was edited to include records collected between 5 and 305 days in milk (DIM). Moreover, cows with less than 3 test-day records of SCC, with the first test-day beyond 60 DIM and with interval between two consecutive test-days greater than 70 DIM were discarded. Contemporary groups were defined as cows calving in the same herd-year-season (HYS) and HYS with less than 5 animals were removed. After editing procedure, 46,304 cows from 303 herds were available for subsequent analysis. Five SCC traits were defined: 1) SCSTOT, the lactation-mean SCS, 2) SCS150, the average SCS from 5 to 150 DIM, 3) SCS305, the average SCS from 151 to 305 DIM, 4) infection, a dichotomous trait (0/1), where 1 indicates that the cow had at least one SCC higher than 100,000 cells ml-1 during the lactation, and 5) severity (%), the ratio between the number of lactation test-days on which SCC was greater than 100,000 cells mL-1 and the total number of lactation test-days. Heritability of the aforementioned traits were estimated using an animal model that considered HYS, classes of age of the cow at calving, and number of lactation testdays as fixed effects, and additive genetic animal and residual as random factors. The pedigree included 144,255 animals. Means for SCSTOT, SCS150, SCS305, infection, and severity were 3.43, 3.10, 3.38, 0.82 and 39.0%, respectively, and all traits exhibited large variability. Estimates of heritability for SCSTOT, SCS150, SCS305, infection and severity were 10.6%, 6.4%, 10.3%, 2.9% and 13.2%, respectively. Despite low, these estimates suggest that selection for alternative SCC traits is feasible and they may potentially contribute to improve mastitis resistance in dairy cows.

International audience