Inferring associations among behavior, hormonal, neurotransmitter and IgA responses of dairy cows housed in automatic milking systems.

This study investigated associations between the hormone cortisol, the neurotransmitters dopamine and serotonin, immunoglobulin A (IgA), heart rate (HR), rectal temperature (RT) and cow behavior pattern in relation to production stress, stress around calving, human handling and adaptation to the milking robot system. Blood samples of 315 Holstein-Friesian cows from five farms with AMS were taken within 14 d after calving (blood sampling date 1 (= BS1) to assess physiological responses during the early postpartum period, followed by human behavioral assessments at blood sampling date 2 (= BS2). Indicators for cow reactions to human handling at BS2 included measurements for the avoidance distance (AD) and scores for the tolerance to tactile interaction (TTI). Dopamine, serotonin and IgA were measured by ELISA tests, and cortisol concentrations were determined by radioimmunoassay. HR and RT as physiological stress indicators were recorded before blood sampling. Associations between blood and physiological biomarkers under different stressors were inferred via generalized linear mixed model applications. In model group 1, analyzing the effects of blood and physiological biomarkers at BS1 on the same biomarkers at BS1, serotonin significantly (P < 0.001) influenced dopamine, and vice versa. Model group 2, examining the effects of behavioral traits on blood and physiological biomarkers at BS2, revealed a significant effect of TTI (P < 0.001) on HR. In model group 3, assessing the effects of biomarkers at BS1 on the same biomarkers at BS2, dopamine, serotonin, IgA and HR at BS1 significantly (P < 0.001) predicted the corresponding biomarker at BS2. In model group 4, effects of daily milk yield on blood and physiological biomarkers at BS2, were non-significant (P > 0.05). In conclusion and from a practical farming perspective, AMS behavioral indicators showed diverse patterns in relation to stress biomarkers, suggesting an index trait for behavioral assessment in a technical environment.