[HTML][HTML] Expression of myeloid differentiation factor 88 in neurons is not requisite for the induction of sickness behavior by interleukin-1β

TP Braun, AJ Grossberg, BO Veleva-Rotse… - Journal of …, 2012 - Springer
TP Braun, AJ Grossberg, BO Veleva-Rotse, JE Maxson, M Szumowski, AP Barnes, DL Marks
Journal of neuroinflammation, 2012Springer
Background Animals respond to inflammation by suppressing normal high-energy activities,
including feeding and locomotion, in favor of diverting resources to the immune response.
The cytokine interleukin-1 beta (IL-1β) inhibits normal feeding and locomotor activity (LMA)
via its actions in the central nervous system (CNS). Behavioral changes in response to IL-1β
are mediated by myeloid differentiation factor 88 (MyD88) in non-hematopoietic cells. It is
unknown whether IL-1β acts directly on neurons or requires transduction by non-neuronal …
Background
Animals respond to inflammation by suppressing normal high-energy activities, including feeding and locomotion, in favor of diverting resources to the immune response. The cytokine interleukin-1 beta (IL-1β) inhibits normal feeding and locomotor activity (LMA) via its actions in the central nervous system (CNS). Behavioral changes in response to IL-1β are mediated by myeloid differentiation factor 88 (MyD88) in non-hematopoietic cells. It is unknown whether IL-1β acts directly on neurons or requires transduction by non-neuronal cells.
Methods
The Nestin-cre mouse was crossed with MyD88lox mice to delete MyD88 from neurons and glia in the CNS (MyD88ΔCNS). These mice were compared to total body MyD88KO and wild type (WT) mice. Mice had cannulae stereotactically placed in the lateral ventricle and telemetry transponders implanted into the peritoneum. Mice were treated with either intracerebroventricular (i.c.v.) IL-1β (10 ng) or vehicle. Food intake, body weight and LMA were continuously monitored for 24 h after treatment. I.c.v. tumor necrosis factor (TNF), a MyD88-independent cytokine, was used to control for normal immune development. Peripheral inflammation was modeled using intraperitoneal lipopolysaccharide (LPS). Groups were compared using two-way ANOVA with Bonferroni post-test. Efficacy of recombination was evaluated using tdTomato reporter mice crossed with the Nestin-cre mouse. MyD88 deletion was confirmed by Western blot.
Results
I.c.v. IL-1β treatment caused a significant reduction in feeding, body weight and LMA in WT mice. MyD88KO mice were protected from these changes in response to i.c.v. IL-1β despite having intact behavioral responses to TNF. Cre-mediated recombination was observed in neurons and astrocytes, but not microglia or endothelial cells. In contrast to MyD88KO mice, the behavioral responses of MyD88ΔCNS mice to i.c.v. IL-1β or intraperitoneal (i.p.) LPS were indistinguishable from those of WT mice.
Conclusion
Sickness behavior is mediated by MyD88 and is dependent on the activity of cytokines within the brain. Our results demonstrate that MyD88 is not required in neurons or astrocytes to induce this behavioral response to IL-1β or LPS. This suggests that a non-Nestin expressing cell population responds to IL-1β in the CNS and transduces the signal to neurons controlling feeding and activity.
Springer