Scatterplot of the log of plasma corticosterone levels versus the log of preoptic PGE2 levels from rats treated with LPS combined with ghrelin is defined by its Pearson correlation coefficient (r), which represents direction and strength of the correlation between these two physiological variables [38]. Statistical differences (Statistica™, version 8.0, StatSoft 2008; Tulsa, OK, USA) among groups were assessed by Linear Mixed Model [10] followed by Fisher LSD post hoc

test (Tb time courses) or one-way ANOVA followed by the Tukey post hoc test (thermal index, plasma corticosterone and preoptic region PGE2 levels). Values of P < 0.05 were considered statistically significant. The putative role of ghrelin in modulating LPS-induced fever was studied by evaluating the effect of ghrelin on Tb GSK J4 ic50 of euthermic (saline-treated)

and febrile (LPS-injected) animals. Pyrogen-free saline (1 ml/kg, i.p.) or ghrelin (0.1 mg/kg, 1 ml/kg, i.p.) caused no significant change in Tb of euthermic animals (P > 0.05, Fig. 1A). Conversely, injection of LPS (50 μg/kg, i.p.) elicited the well characterized LPS-induced febrile response (for review see [22]). Interestingly, when LPS administration was associated with ghrelin the febrile Erastin chemical structure response was attenuated at the third phase of fever (P < 0.05, Fig. 1A). Thermal indexes (TIs) (area under curve; indicated by the horizontal bar in Fig. 1A) were calculated to emphasize the effect of ghrelin on LPS-induced fever ( Fig. 1B). As shown in Fig. 1B, injection of LPS induced the highest TI (252.9 ± 12.6 °C min). TI of ghrelin + LPS (169.5 ± 34.3 °C min) was lower than that of LPS alone (P < 0.05), and it was significantly higher than both pyrogen-free saline (86.9 ± 21.8 °C min, P < 0.05) and ghrelin (68.1 ± 24.5 °C min, P < 0.05). Plasma corticosterone levels were assessed to evaluate the putative influence of ghrelin administration on the LPS-induced hypothalamic-pituitary-adrenal

axis activation. As shown in Fig. 2, injection of LPS alone induced a significant increase in plasma corticosterone levels (from 11.0 ± 2.0 to 21.5 ± 2.8 ng/ml; P < 0.05). Administration of ghrelin combined with LPS potentiated the increased secretion of corticosterone induced by LPS (from 11.0 ± 2.0 to 32.6 ± 3.4 ng/ml; P < 0.05), whereas ghrelin alone did not alter MTMR9 the basal levels of plasma corticosterone (from 11.0 ± 2.0 to 13.5 ± 1.6 ng/ml). To address whether the attenuated LPS-induced fever in ghrelin-treated rats resulted from inhibited central COX activity, we measured the levels of PGE2 in the preoptic region of the hypothalamus. Even though PGE2 production tended to be higher in ghrelin-treated rats (51.2 ± 8.4 ng/mg of protein) compared to saline-treated controls (39.1 ± 5.5 ng/mg of protein) no significant difference between these groups was found (P > 0.05). Administration of LPS, on the contrary, evoked a marked increase in preoptic PGE2 levels (from 39.1 ± 5.5 to 163.7 ± 19.