1. The effects of repeated injections of bacterial lipopolysaccharide (LPS) at 3 day intervals on abdominal temperature and systemic release of tumour necrosis factor alpha (TNF)‐like and interleukin‐6 (IL‐6)‐like activity were measured in guinea‐pigs. 2. After the third injection of LPS the fever response was significantly attenuated. 3. TNF‐like activity (peak 1 h after LPS injection) and IL‐6‐like activity (peak 3 h after LPS injection) in plasma changed correspondingly, both being significantly reduced after the third and subsequent injections of LPS. 4. The increase of IL‐6‐like activity in plasma after LPS injection correlated to the febrile change in body temperature. This correlation remained manifest throughout the whole time course of the development of endotoxin tolerance. 5. The reduced production of TNF‐like activity after repeated injections of LPS correlated to the attenuation of the fever index, the integration of the thermal response after LPS application. 6. The results support the hypothesis that one component of the development of endotoxin tolerance is reduced production and release of cytokines in response to repeated injections of the same amount of LPS.

The time course of activity of interleukin-6 (IL-6) and tumor necrosis factor (TNF) was measured in blood plasma and hypothalamic push-pull perfusates during the febrile response to intramuscular injection of bacterial endotoxin (Escherichia coli, 20 micrograms/kg) in 24 guinea pigs. Injection of endotoxin caused a dramatic increase of IL-6 activity in plasma. The logarithmic values of plasma IL-6 activities showed a linear correlation to the febrile change in body temperature (r = 0.898) during the whole time course of fever. IL-6 activity in hypothalamic perfusates increased 12-fold in the first hour after pyrogen application and declined slowly despite the further increase in body temperature. Hypothalamic IL-6 activity did not correlate with the febrile increase in body temperature (r = -0.048). TNF activity in plasma, not detectable before pyrogen application, had its peak in the first hour after endotoxin injection and rapidly declined to 15-20% of the peak activity within the next 2 h and to an undetectable value 5 h after injection. In the hypothalamus TNF was not detectable before endotoxin injection, but it could be monitored in most animals after pyrogen application without a clear correlation to the fever response. These results taken together indicate that endotoxin fever represents a physiological situation in which production and release of cytokines in the peripheral immune system and in the hypothalamus are regulated and stimulated in independent patterns.

No abstract

The role of tumor necrosis factor (TNF) in the febrile and metabolic responses of rats to intraperitoneal injection of a high dose of lipopolysaccharide Injection of a high dose of lipopolysaccharide (LPS) induces a septic-shock-like state, which can be accompanied by phases of hypothermia and phases of fever. In the present study we monitored body core temperature and locomotor activity, both by remote radiotelemetry, as well as changes in food intake, body mass and water intake for 3 days after an intraperitoneal (i.p.) injection of a high dose of LPS (5 mg/kg) along with sterile 0.9% saline or a neutralizing form of the soluble tumor necrosis factor (TNF) type 1 receptor (referred to as TNF-binding protein, TNF bp). Intraperitoneal injection of LPS rapidly induced high concentrations of TNF in the plasma and peritoneal lavage fluid. TNF was undetectable in the plasma and peritoneal lavage fluid of animals co-injected with LPS and TNF bp, implying neutralization of peripheral bioactive TNF. Administration of LPS induced hypothermia by about 1.5 degrees C, which lasted for 5 h after injection. During the light-time periods of days 2 and 3 after injection, the rats developed a robust fever. Treatment with TNF bp resulted in a faster recovery from the LPS-induced hypothermia so that the rats developed a pronounced fever on the day of injection. Locomotor activity during night-time periods was suppressed in LPS-treated animals. The LPS-induced depression of night-time activity was not antagonized by co-injection of TNF bp. On day 1 after the injection of LPS, food intake reduced to virtually zero, water intake fell to about 30% of the control value and body mass dropped by 25 g (about 10% of total body mass). With the exception of body mass, these variables recovered slowly during days 2 and 3 after LPS injection, but did not reach the control values. The LPS-induced decreases in food intake, body mass and water intake were significantly attenuated by the treatment with TNF bp. These results confirm that TNF contributes significantly to the rats' responses to intraperitoneal injection of a high dose of LPS. The fact that treatment with TNF bp accelerated and improved the rats' ability to develop a febrile response supports the view that the fever is beneficial, since all other metabolic responses measured in this study were normalized more effectively in those rats that developed a faster and more pronounced increase in body temperature.

Changes in concentrations of cytokines in plasma and in hypothalamic push-pull perfusates of guinea pigs were measured within the 1 st hour after intramuscular injections of bacterial Hpopolysaccharide (LPS; Escherichia coli, 20 µg/kg) or solvent (0.9% saline). In control animals injected with solvent, interleukin (IL)-1 and tumor necrosis factor alpha (TNF-α) were not detectable in plasma. Only IL-6 was present in picogram quantities. Within 45 min after injection of LPS, the concentrations of IL-1, TNF-α, and IL-6 increased in the plasma: by several orders of magnitude for TNF-α and about tenfold for IL-G. Picogram amounts of biologically active IL-1 were detected in plasma after injection of LPS. No steady state levels of systemic cytokines were reached during the experimental period. In hypothalamic perfusates of animals injected with the solvent, no IL-1 was detectable. TNF-α could be detected at higher concentrations than IL-6. IL-6 was detectable at tenfold lower concentrations than in the plasma. In animals injected with LPS, the hypothalamic concentration of IL-6 started to increase during the period 15-30 min and the concentrations of TNF-α during the period 30-45 min after LPS injection. The concentrations of IL-6 increased by 300-400% and did not exceed picogram values. No progressive increase of hypothalamic levels of these cytokines was observed during the time course of the experiment. The method used did not detect any changes in the amount of biologically active IL-1 in hypothalamic perfusates of LPS-treated animals. No obvious correlation between concentrations of cytokines in plasma and hypothalamic perfusate was observed, indicating the brain origin of the cytokines. Since the increase in IL-6 goes in parallel with resetting of the body thermostat to the higher level, the data support the hypothesis that the increase in the concentration of IL-6 in the brain, occurring during the early phase of the fever, induces the febrile response.