The bovine acute phase response to endotoxin and Gram-negative bacteria

Research output: Book/ReportPh.D. thesis

The overall aims of the work presented in this thesis were to characterize bovine cytokine and acute phase protein (APP) responses to lipopolysaccharide (LPS) and to investigate how LPS-induced clinical and immunoinflammatory responses differed between individual cows. Two kinds of experimental exposure to LPS were employed: 1) intravenous bolus injection of Escherichia coli (E. coli) LPS inducing systemic inflammation (Papers I-III and Reports I and II), and 2) intramammary inoculation with E. coli inducing infection and local inflammation in the mammary gland (Paper IV).

Systemic inflammation
Eight non-lactating, non-pregnant dairy cows each received three intravenous injections of LPS with three-week intervals (doses: 10, 100 and 1000 ng LPS/kg, consecutively). All three LPS doses induced clinical, haematological and APP responses as well as changes in serum levels of microminerals. The responses lasted from a few hours to several days. The magnitude and duration of the responses increased with increasing LPS dose. The experimental design allowed individual differences in the responses to LPS to be evaluated, and it was demonstrated that all dose-dependent clinical, haematological, APP and blood-biochemical responses (except serum calcium concentrations) displayed statistically significant individual variation (Papers I and II). The APP responses of individual cows were of comparable magnitude in consecutive LPS challenges, suggesting that some cows are APP high responders and possess a higher capacity for mounting an APP response than other cows (Paper II). Changes in the cytokine production of peripheral blood mononuclear cells (PBMCs) were investigated by flow cytometry. The percentages of circulating PBMCs with capacity for production of pro-inflammatory cytokines (interleukin-1β, interleukin-8 and tumor necrosis factor-α) decreased within two hours of each LPS injection, and returned to baseline levels within 24 hours. With a few exceptions, the percentages of PBMCs producing anti-inflammatory cytokines (interleukin-4 and -10) either remained unaltered or tended to increase in response to LPS. Pre-challenge levels of PBMCs producing pro-inflammatory cytokines increased significantly from challenge to challenge, whereas pre-injection levels of PBMCs producing IL-4 decreased significantly (Paper III). Serum concentrations of antibodies with anti-LPS activity increased slightly in response to the repeated LPS challenges, whereas the level of antibodies against lipid A did not change (Report I). Levels of PBMCs carrying the LPS receptor (CD14) and CD14-density in the cell membrane differed between cows, but were unaffected by repeated LPS challenges (Report II).

Infection and inflammation of the mammary gland
Six cows were included in this study on the basis of their ability to clear intramammary infection induced by intramammary inoculation of E. coli (mild, intermediate and severe infection). Milk and plasma serum amyloid A (SAA) concentrations were measured sequentially (at 0, 6, 12, 24, 36, 48, 60, and 72 hours after inoculation). Both milk and plasma SAA levels increased in all six cows in response to the infection: milk levels between 6 and 12 hours post-inoculation and plasma levels between 12 and 24 hours post-inoculation. The fact that SAA appeared in milk before it appeared in the systemic circulation suggests that extrahepatic synthesis of SAA takes place in the inflamed udder. SAA levels in plasma and milk were very low prior to inoculation, increased rapidly and exponentially after inoculation, and returned quickly to the baseline after the infection was cleared from the udder. APP measurements in milk may therefore have great potential as mastitis diagnostics. Milk and plasma SAA concentrations seemed to be higher in cows with severe mastitis than in cows with moderate or mild mastitis, which suggests that SAA levels reflect severity of udder infection (Paper IV).

The results presented in this thesis show that even minute amounts of E. coli LPS or bacteria induced long-lasting clinical, haematological, blood-biochemical and immuno-inflammatory responses in the cows. These findings confirm that cattle are very sensitive to LPS. The responses to LPS are all aimed at containing infections, eliminating bacteria and restoring homeostasis. However, during infections or disease complexes in which LPS and/or Gram-negative bacteria persist, disease and pathological changes may result from the prolonged inflammatory response. For example, protracted LPS-induced anorexia and ruminal hypomotility may be detrimental for lactating dairy cows, since these cows are highly dependent on continuous feed intake to avoid severe metabolic disturbances. Furthermore, depression of the cytokine-producing capacity of circulating PBMCs may result in aberrant innate immune responses and hence increased susceptibility to secondary infections. If the cow is not to succumb to the adverse effects of LPS, it is therefore of paramount importance that she is able to orchestrate the LPS-induced inflammation in an appropriate way. Substantial individual variation in this ability was demonstrated in the cows studied in this thesis, and this may explain why cows are not equally suited to cope clinically with LPS-associated disease. However, before the implications of individual differences in the LPS-induced responses can be fully understood, more information is needed about the pathogenesis of bovine acute phase response and the biologically relevant functions of central reactants such as cytokines and APPs. Cytokines are known to induce pathophysiological changes and SAA and haptoglobin have also been suggested to be important modulators of the inflammatory response. Success in mounting appropriate cytokine and APP responses may therefore play a role in the ability of cows to withstand inflammatory insults. Concentrations of APPs are often used as indicators of the severity of disease and to reflect the extent of underlying pathology. The results of this investigation do not undermine this concept as such, but given the significant individual variation in APP response, caution is certainly warranted when absolute concentrations of SAA and haptoglobin are used to interpret disease status in individual cows. In the experimental mastitis model, SAA concentrations in plasma and milk seemed to reflect the severity of disease, and this further emphasizes the potential of APPs as mastitis diagnostics. Repeated or continuous LPS exposure has been shown to result in transient hyporesponsiveness to the effects of LPS. However, three findings suggest either that sensitivity was restored between injections, or that LPS tolerance – in the sense in which this term is generally used – did not develop in the model of repeated LPS injections employed in the main part of this thesis: 1) the intensity of the majority of clinical, haematological and blood-biochemical responses increased with each increase in LPS dose, 2) cytokine responses were not gradually attenuated, and 3) levels of anti-LPS antibodies increased minimally. Moreover, CD14 was not down-regulated after the three LPS injections, and hyporesponsiveness therefore cannot have developed as a consequence of lack of recognition of LPS through this receptor. However, one carry-over effect of LPS was evident, namely that pre-injection levels of PBMCs with capacity for synthesis of pro-inflammatory cytokines increased and levels of PBMCs synthesizing anti-inflammatory cytokines decreased from challenge to challenge. This finding may suggest that cows attain a cytokine profile after LPS exposure that for some period of time allows them to react more promptly when re-exposed to LPS or Gram-negative bacteria. Whether this constitutes tolerance is not known. In conclusion, the work presented in this thesis advances our understanding of the pathogenesis of LPS-induced inflammation. It also emphasizes the potential of APPs as markers of ongoing systemic and local inflammation, and it highlights some of the pitfalls that should be borne in mind when these proteins are being used to diagnose disease in individual cows.
Original languageEnglish
Place of PublicationFrederiksberg
PublisherSamfundslitteratur
ISBN (Print)ISBN 87-989085-1-2
Publication statusPublished - 18 Aug 2003

ID: 119255513