A reappraisal on the ability of leptin to induce fever

Leptin is often regarded as a mediator of fever, even though an in-depth analysis of the dose-dependent effects of leptin on body temperature (T(b)), pro-inflammatory cytokines, and circulating leptin has never been performed. In the present study, such an analysis was performed in rats that were food deprived (lower baseline levels of leptin) or free feeding (higher baseline levels of leptin). In a relatively cool environment (22 degrees C), rats deprived of food for 24 h exhibited mild (approximately 0.5 degrees C) hypothermia. Leptin infusion (250 microg/kg iv) elevated the T(b) of the food-deprived rats to a normothermic level, an effect that peaked (120 min post-infusion) when plasma leptin was at a level (approximately 8 ng/mL) often found in leptin-responsive subjects. Increasing the leptin dose to 1000 microg/kg did not produce any further (febrile) elevation in the T(b) of food-deprived rats. The anti-hypothermic effect of leptin in food-deprived rats was not associated with any rise in the plasma levels of the pro-inflammatory cytokines tumor necrosis factor (TNF)-alpha and interleukin (IL)-6. In free-feeding rats kept in a cooler (22 degrees C) or warmer (28 degrees C) environment, leptin infusion failed to alter T(b) or to produce any surge in plasma TNF-alpha or IL-6, even when the dose infused (3500 microg/kg iv) resulted in excessive, non-physiological rises in plasma leptin (approximately 542 ng/mL at 30 min; approximately 75 ng/mL at 120 min post-infusion). In contrast, free-feeding rats in the same experimental set-up were able to respond to a low dose (2 microg/kg iv) of IL-1beta with a typical biphasic fever, which was associated with surges in plasma TNF-alpha and IL-6. Collectively, our data show that an acute rise in plasma leptin to a level within or fairly above the physiological range does not induce fever. These results challenge the idea that leptin may be a mediator of fever.