Impact of somatic cell count combined with differential somatic cell count on milk protein fractions in Holstein cattle

Udder health in dairy herds is a very important is-sue given its implications for animal welfare and the production of high-quality milk. Somatic cell count (SCC) is the most widely used means of assessing ud-der health status. However, differential somatic cell count (DSCC) has recently been proposed as a new and more effective means of evaluating intramammary infection dynamics. Differential SCC represents the combined percentage of polymorphonuclear neutrophils and lymphocytes (PMN-LYM) in the total SCC, with macrophages (MAC) accounting for the remaining proportion. The aim of this study was to evaluate the association between SCC and DSCC and the detailed milk protein profile in a population of 1,482 Holstein cows. A validated reversed-phase HPLC method was used to quantify 4 caseins (CN), namely alpha S1-CN, alpha S2- CN, & UKappa;-CN, and 0-CN, and 3 whey protein fractions, namely 0-lactoglobulin, alpha-lactalbumin, and lactoferrin, which were expressed both quantitatively (g/L) and qualitatively (as a percentage of the total milk nitro-gen content, %N). A linear mixed model was fitted to explore the associations between somatic cell score (SCS) combined with DSCC and the protein fractions expressed quantitatively and qualitatively. We ran an additional model that included DSCC expressed as PMN-LYM and MAC counts, obtained by multiplying the percentages of PMN-LYM and MAC by SCC for each cow in the data set. When the protein fractions were expressed as grams per liter, SCS was significantly negatively associated with almost all the casein frac-tions and positively associated with the whey protein alpha-lactalbumin, while DSCC was significantly associated with alpha S1-CN, 0-CN, and alpha-lactalbumin, but in the op-posite direction to SCS. We observed the same pattern with the qualitative data (i.e., %N), confirming oppo-site effects of SCS and DSCC on milk protein fractions. The PMN-LYM count was only slightly associated with the traits of concern, although the pattern observed was the same as when both SCS and DSCC were in-cluded in the model. The MAC count, however, gener-ally had a greater impact on many casein fractions, in particular decreasing both 0-CN content (g/L) and proportion (%N), and exhibited the opposite pattern to the PMN-LYM count. Our results show that informa-tion obtained from both SCS and DSCC may be useful in assessing milk quality and protein fractions. They also demonstrate the potential of MAC count as a novel udder health trait