Neurogenic exacerbation of microglial and astrocyte responses to Neisseria meningitidis and Borrelia burgdorferi

Abstract

Although glial cells are recognized for their roles in maintaining neuronal function, there is growing appreciation of the ability of resident CNS cells to initiate and/or augment inflammation following trauma or infection. The tachykinin, substance P (SP), is well known to augment inflammatory responses at peripheral sites and its presence throughout the CNS raises the possibility that this neuropeptide might serve a similar function within the brain. In support of this hypothesis, we have recently demonstrated the expression of high affinity receptors for SP (Neurokinin-1 (NK-1) receptors) on microglia and shown that this tachykinin can significantly elevate bacterially induced inflammatory prostanoid production by isolated cultures of these cells. In the present study, we demonstrate that endogenous SP/NK-1R interactions are an essential component in the initiation and/or progression of CNS inflammation in vivo following exposure to two clinically relevant bacterial CNS pathogens, Neisseria meningitidis and Borrelia burgdorferi. We show that in vivo elevations in inflammatory cytokine production and decreases in the production of an immunosuppressive cytokine are markedly attenuated in mice genetically deficient in the expression of the NK-1R or in mice treated with a specific NK-1R antagonist. Furthermore, we have used isolated cultures of microglia and astrocytes to demonstrate that SP can augment inflammatory cytokine production by these resident CNS cell types following exposure to either of these bacterial pathogens. Taken together, these studies indicate a potentially important role for neurogenic exacerbation of resident glial immune responses in CNS inflammatory diseases, such as bacterial meningitis.

FIGURE 1

Bacterially induced increases in CNS cellularity, astrogliosis, and demyelination are markedly reduced in the absence of endogenous SP/NK-1R interactions. N. meningitidis (1 × 10⁶) was administered via i.c. injection into C57BL/6 wild-type (NK1R+/+) and NK-1R-deficient (NK1R-/-) mice. In addition, a group of wild-type mice received s.c. injections (5 mg/kg) of the NK-1R antagonist l-703,606 at days -1, 0, and +1 relative to bacterial challenge. At 72 h postinfection, brain tissue was perfused in situ, isolated, and prepared for histochemical analysis by lightfield microscopy or immunofluorescent microscopy. Upper panels, Representative H&E-stained axial cortical fields (×20 objective) from one of the four animals in each group. Middle panels, Representative GFAP (green) and 4′,6-diamidino-2-phenylindole, dihydrochloride (blue) immunofluorescence (×40 objective) in coronal cortical fields from one of the four animals in each group. Bottom panels, Representative Luxol blue staining of the corpus callosum in coronal fields (×10 objective) from one of the four animals in each group.

FIGURE 2

In vivo astrogliosis and inflammatory cytokine production are markedly reduced in the absence of endogenous SP/NK-1R interactions. A, Wild-type C57BL/6 mice were given vehicle or infected with N. meningitidis (Nm) or B. burgdorferi (Bb) (1 × 10⁶) via i.c. injection and were untreated or received s.c. administration of the NK-1R antagonist l-703,606 (5 mg/kg) at days -1, 0, and +1 relative to bacterial challenge. At 72 h postinfection, glial cells were isolated, permeabilized, and GFAP expression determined by flow cytometry. n = 6; Asterisk indicates significant difference from uninfected (0) animals, and dagger indicates significant difference from infected animals that did not receive the NK-1R antagonist. Fluorescence signal obtained using an isotype and species matched Ab directed against an irrelevant Ag is indicated (Irr). B-D, N. meningitidis (1 × 10⁶) (Nm) or B. burgdorferi (1 × 10⁶) (Bb) was administered via i.c. injection into C57BL/6 wild-type (wt) or NK-1R-deficient (NK1R-/-) mice. In addition, groups of wild-type mice received s.c. injections (5 mg/kg) of the NK-1R antagonist l-703,606 at days -1, 0, and +1 to bacterial challenge (wt inf + ANT). At 72 h postinfection, mRNA or tissue homogenates were isolated for measurement of TNF-α (B), IL-6 (C), and IL-10 (D) mRNA and protein expression by real-time PCR and specific capture ELISA, respectively. n = 8; Asterisk indicates significant difference from uninfected (wt uninf) animals, and dagger indicates significant difference from infected wild-type animals (wt inf).

FIGURE 3

SP augments inflammatory cytokine production by primary microglia following exposure to N. meningitidis or B. burgdorferi in an NK-1R-dependent manner. Cells (2 × 10⁶) from wild-type (NK1R+/+) and NK-1R knockout (NK1R-/-) animals were untreated or exposed to either viable N. meningitidis (Nm: M.O.I. of 10 or 30 bacteria to each glial cell), N. meningitidis Ags (Nm Ag: 1 and 5 μg/ml), viable B. burgdorferi (Bb: M.O.I. of 10 or 30 bacteria to each glial cell), or B. burgdorferi Ags (Bb Ag: 0.5, 1, and 2.5 μg/ml) in the absence (-SP) or presence (+SP) of SP (5 nM). At 12 h following bacterial challenge, culture supernatants were isolated and assayed for the presence of TNF-α (A and C) or IL-6 (B and C) by specific capture ELISA. Data are presented as the mean of triplicate determinations of samples from three separate experiments ± SEM. Asterisks indicate statistically significant differences between cytokine levels in the absence or presence of SP.

FIGURE 4

SP augments bacterially induced inflammatory cytokine production by primary astrocytes in a NK-1R-dependent manner. Cells (2 × 10⁶) from wild-type (NK1R+/+) and NK-1R knockout (NK1R-/-) animals were untreated (0) or exposed to N. meningitidis (Nm: 1, 5, or 10 μg/ml), viable B. burgdorferi (Bb: M.O.I. of 3 or 30 bacteria to each astrocyte) or B. burgdorferi Ags (Bb Ag: 0.5, 1, or 2.5 μg/ml) in the absence (-SP) or presence (+SP) of SP (5 nM). At 12 (A and B) or 24 h (C and D) following bacterial challenge, culture supernatants were assayed for the presence of TNF-α (A and C) and IL-6 (B and D) by specific capture ELISA. Data are presented as the mean of triplicate determinations of samples from three separate experiments ± SEM. Asterisks indicate statistically significant differences between cytokine levels in the absence or presence of SP.

FIGURE 5

Proposed model in which SP exacerbates the early inflammatory responses of resident CNS cells following bacterial infection. Bacterial motifs are recognized by TLR and NLR pattern recognition receptors expressed by perivascular macrophages, microglia, and astrocytes leading to NF-κB activation and resulting in inflammatory cytokine production via NF-κB activation. SP released by neurons and perhaps activated glia acts on these NK-1R bearing CNS cells to alter NF-κB activation or function. In this manner, SP augments TNF-α and IL-6 production by astrocytes and microglia, and limits IL-10 production by perivascular macrophages and/or infiltrating leukocytes, thereby exacerbating CNS inflammation and damage.