Brain region-specific alterations in the gene expression of cytokines, immune cell markers and cholinergic system components during peripheral endotoxin-induced inflammation

Abstract

Inflammatory conditions characterized by excessive peripheral immune responses are associated with diverse alterations in brain function, and brain-derived neural pathways regulate peripheral inflammation. Important aspects of this bidirectional peripheral immune-brain communication, including the impact of peripheral inflammation on brain region-specific cytokine responses, and brain cholinergic signaling (which plays a role in controlling peripheral cytokine levels), remain unclear. To provide insight, we studied gene expression of cytokines, immune cell markers and brain cholinergic system components in the cortex, cerebellum, brainstem, hippocampus, hypothalamus, striatum and thalamus in mice after an intraperitoneal lipopolysaccharide injection. Endotoxemia was accompanied by elevated serum levels of interleukin (IL)-1β, IL-6 and other cytokines and brain region-specific increases in Il1b (the highest increase, relative to basal level, was in cortex; the lowest increase was in cerebellum) and Il6 (highest increase in cerebellum; lowest increase in striatum) mRNA expression. Gene expression of brain Gfap (astrocyte marker) was also differentially increased. However, Iba1 (microglia marker) mRNA expression was decreased in the cortex, hippocampus and other brain regions in parallel with morphological changes, indicating microglia activation. Brain choline acetyltransferase (Chat ) mRNA expression was decreased in the striatum, acetylcholinesterase (Ache) mRNA expression was decreased in the cortex and increased in the hippocampus, and M1 muscarinic acetylcholine receptor (Chrm1) mRNA expression was decreased in the cortex and the brainstem. These results reveal a previously unrecognized regional specificity in brain immunoregulatory and cholinergic system gene expression in the context of peripheral inflammation and are of interest for designing future antiinflammatory approaches.

Figure 1

Peripheral LPS administration increases serum cytokine and chemokine levels. Mice were injected with LPS (8 mg/kg, IP) or saline (S) and euthanized 4 h later. Serum IL-6, IL-1β, CXCL1, IFN-γ, IL-12p70, IL-10 and TNF were analyzed by using a multiplex platform, as described in Materials and Methods. Data are shown as mean ± SEM (n = 15 mice per group); ****P < 0.0001.

Figure 2

Relative Il1b mRNA expression is differentially increased in various brain regions after peripheral LPS administration. Mice were injected with LPS (8 mg/kg, IP) or saline and euthanized 4 h later. Brain regions were isolated and processed for Il1b mRNA expression analysis by qPCR, as described in Materials and Methods. Data are shown as mean ± SEM and fold-change (n = 15 mice per group); ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05. Because of limited RNA levels in the hippocampus, hypothalamus, striatum and thalamus, three tissue samples (from the same region and treatment group) were combined before RNA extraction; therefore, n = 5 was used for statistical analyses in these regions.

Figure 3

Relative Il6 mRNA expression is differentially increased in various brain regions after peripheral LPS administration. Mice were injected with LPS (8 mg/kg, IP) or saline and euthanized 4 h later. Brain regions were isolated and processed for Il6 mRNA expression analysis by qPCR, as described in Materials and Methods. Data are shown as mean ± SEM and fold-change (n = 15 mice per group); ****P < 0.0001, ***P < 0.0005, **P < 0.001, *P < 0.005. Because of limited RNA levels in the hippocampus, hypothalamus, striatum and thalamus, three tissue samples (from the same region and treatment group) were combined before RNA extraction; therefore, n = 5 was used for statistical analysis in these regions.

Figure 4

Lethal inflammation is not associated with early brain neuronal cell death and morphology alterations. Mice were injected with LPS (8 mg/kg, IP) or saline and euthanized 4 h later. Brain neuronal cell morphology was evaluated (using H&E-stained sections) in the cortex, cerebellum and hippocampus by assessing three visual fields per brain region per mouse. Representative photomicrographs (40×) are shown of H&E-stained sections of cortex-saline (A), cortex-LPS (B), cerebellum-saline (C), cerebellum-LPS (D), hippocampus-saline (E) and hippocampus-LPS (F). Scale bar, 50 μm.

Figure 5

Gfap mRNA expression is increased in specific brain regions during lethal inflammation. Mice were injected with LPS (8 mg/kg, IP) or saline and euthanized 4 h later. Brain regions were isolated and processed for Gfap mRNA expression analysis by qPCR, as described in Materials and Methods. Data are shown as mean ± SEM (n = 15 mice per group); ***P < 0.0001, **P < 0.0002, *P < 0.05. Because of limited RNA levels in the hippocampus, hypothalamus, striatum and thalamus, three tissue samples (from the same region and treatment group) were combined before RNA extraction; therefore, n = 5 was used for statistical analysis in these regions.

Figure 6

Iba1 mRNA expression is decreased in specific brain regions during lethal inflammation. Mice were injected with LPS (8 mg/kg, IP) or saline and euthanized 4 h later. Brain regions were isolated and processed for Iba1 mRNA expression analysis by qPCR, as described in Materials and Methods. Data are shown as mean ± SEM (n = 15 mice per group); **P < 0.0001, *P < 0.05. Because of limited RNA levels in the hippocampus, hypothalamus, striatum and thalamus, three tissue samples (from the same region and treatment group) were combined before RNA extraction; therefore, n = 5 was used for statistical analysis in these regions.

Figure 7

Lethal inflammation is associated with morphological alterations in microglia indicative of activation. Mice were injected with LPS (8 mg/kg, IP) or saline and euthanized 4 h later. Brain tissues were stained for Iba1 immunohistochemistry, as described in Materials and Methods. Iba1-stained microglia were evaluated in the cortex and hippocampus by assessing three visual fields per brain region, per mouse. Microglia were faintly stained with anti-Iba1 in the cortex and the hippocampus of saline-injected mice, while morphological changes of activated microglia, including cell body enlargement, shortening and thickening of processes, were observed in these brain areas of LPS-injected mice. Representative photomicrographs (40×) of Iba1-stained sections are shown: cortex-saline (A), cortex-LPS (B), hippocampus-saline (C) and hippocampus-LPS (D). Enlarged single cell images are shown on the upper right corners. Scale bar, 50 μm.

Figure 8

Lethal inflammation results in decreased Chat mRNA expression in the striatum. Mice were injected with LPS (8 mg/kg, IP) or saline and euthanized 4 h later. Brain regions were isolated and processed for Chat mRNA expression analysis by qPCR, as described in Materials and Methods. Data are shown as mean ± SEM (n = 15 mice per group); *P < 0.01. Because of limited RNA levels in the hippocampus, hypothalamus, striatum and thalamus, three tissue samples (from the same region and treatment group) were combined before RNA extraction; therefore, n = 5 was used for statistical analysis in these regions. No Chat mRNA was detected in the cerebellum and hippocampus.

Figure 9

Lethal inflammation results in region-specific alterations in brain Ache mRNA expression. Mice were injected with LPS (8 mg/kg, IP) or saline and euthanized 4 h later. Brain regions were isolated and processed for Ache mRNA expression analysis by qPCR, as described in Materials and Methods. Data are shown as mean ± SEM (n = 15 mice per group); ***P < 0.004, **P < 0.01. Because of limited RNA levels in the hippocampus, hypothalamus, striatum and thalamus, three tissue samples (from the same region and treatment group) were combined before RNA extraction; therefore, n = 5 was used for statistical analysis in these regions.

Figure 10

Lethal inflammation results in brain region–specific decreases in Chrm1 mRNA expression. Mice were injected with LPS (8 mg/kg, IP) or saline and euthanized 4 h later. Brain regions were isolated and processed for Chrm1 mRNA expression analysis by qPCR, as described in Materials and Methods. Data are shown as mean ± SEM (n = 15 mice per group); ***P < 0.002. Because of limited RNA levels in the hippocampus, hypothalamus, striatum and thalamus, three tissue samples (from the same region and treatment group) were combined before RNA extraction; therefore, n = 5 was used for statistical analysis in these regions. No Chrm1 mRNA was detected in the cerebellum of LPS-treated mice.