Whole blood optimization and genetic association of ex vivo TNF-α responsiveness to killed E. coli in Danish Holstein cows

Effect of Holstein genotype on ex-vivo cytokine response to lipopolysaccharide (LPS) and lipoteichoic acid (LTA) during the periparturient period

Effects of Holstein genotype on innate immune response were assessed with ex-vivo lipopolysaccharide (LPS) and lipoteichoic acid (LTA) stimulation of whole blood from unselected (UH, n = 10) and contemporary (CH, n = 11) Holsteins that differ in production by more than 4,500 kg/lactation. Blood was collected at -14, 7, 28, and 49 days in milk (DIM), mixed with a pathogen-associated molecular pattern (PAMP) molecule (0.01 or 1.0 µg LPS or 10 or 100 µg LTA per mL blood) and incubated (4 h, 37 °C). Plasma cytokines were quantified by ELISA, log₁₀-transformed and analyzed by repeated measures with DIM as the repeated effect. Cytokine responses increased with PAMP dose and decreased as DIM increased. There was a genotype by LPS dose interaction for IL-1β as response to the low dose was greater in UH but did not differ between genotypes for the high dose. The IL-1β response was greater while the IL-6 response to LTA tended to be greater in UH than in CH cows. The more negative energy balance of CH cows did not impact genotype difference in cytokine responses. Results indicate selection since the mid-1960s has decreased ex-vivo, whole blood cytokine response of CH cows to LPS and to LTA.

Effect of polymyxin B on ex vivo tumor necrosis factor-alpha responsiveness of blood leukocytes in Danish Holstein Friesian cows

Whole blood stimulation assay (WBA) with killed gram-positive and gram-negative udder pathogens were used to investigate the interference of the endotoxin-binding antibiotic polymyxin B (PMB) on the ex vivo TNF-α response. Blood samples were collected from first to third lactating dairy cows in their early lactation (<50 days in milk, n = 32) period. The WBA was stimulated with both inactivated bacteria (e.g., dead Escherichia coli, Staphylococcus aureus, Streptococcus dysgalactiae, Streptococcus uberis), at a concentration of 2.5 × 10⁶/mL; and pathogen-associated molecular pattern molecules, namely E. coli LPS (10 μg/mL), and S. aureus peptidoglycan (PG, 10 μg/mL). The PMB was added at a concentration of 0, 12.5, 25, 50, 100, and 200 μg/mL to each stimulant, respectively. All bacteria stimulants resulted in an increased TNF-α response compared to the negative control. The PMB affected the TNF-α responses of gram-positive (except S. dysgalactaie), gram-negative bacteria; and bacterial cell wall components at a PMB concentration of 25-50 μg/mL. The LPS and E. coli had similar TNF-α response but PG had a lower TNF-α response than gram-positive bacteria. The doses of PMB (≥ 25 μg/mL) should be used with caution when using different types of pathogens or should be avoided in ex vivo TNF-α studies.

Ex vivo tumor necrosis factor-alpha response of blood leukocytes in Danish Holstein-Friesian cows stimulated by Gram-positive and Gram-negative bacteria isolated from mastitic milk

A whole blood stimulation assay was used to investigate the effects of parity, number of weeks after calving and Gram-positive and Gram-negative bacteria on the ex vivo TNF-α responsiveness of Danish Holstein-Friesian cows of first to third lactation (n = 28). Blood samples were collected in weeks 2, 3, 5 and 8 after parturition and stimulated with Escherichia coli LPS (10 μg/mL), Staphylococcus aureus peptidoglycan (PGN, 10 μg/mL) and dead Escherichia coli, Streptococcus uberis, Staphylococcus aureus, and Streptococcus dysgalactiae at a concentration of 2.5 × 10⁶/mL. The antibiotic polymyxin-B (100 μg/mL) was added to the Gram-positive bacteria to avoid the influence of environmental endotoxin by ELISA test. Overall, parity had no effect, whereas number of weeks after calving altered the TNF-α responsiveness of the majority of the stimulants. Ex vivo, Gram-positive bacteria always resulted in a higher TNF-α response than Gram-negative bacteria with large differences within the individual cows. High correlations were found within the Gram-negative stimulants panel (r = 0.83) and within the Gram-positive (r = 0.81 to 0.86) stimulants panel except PGN. The higher TNF-α responsiveness by Gram-positive bacteria is in agreement with in vitro studies in human but in contrast to the in vivo TNF-α responsiveness in bovine udder.