ProBDNF promotes sepsis-associated encephalopathy in mice by dampening the immune activity of meningeal CD4+ T cells

Abstract

Background: Sepsis-associated encephalopathy (SAE) increases the mortality of septic patients, but its mechanism remains unclear. The present study aimed to investigate the roles of T lymphocytes, proBDNF, and their interaction in the pathogenesis of SAE.

Methods: Fear conditioning tests were conducted for cognitive assessment in the lipopolysaccharide (LPS, 5 mg kg^-1)-induced septic mice. Meninges and peripheral blood were harvested for flow cytometry or qPCR. FTY720 and monoclonal anti-proBDNF antibody (McAb-proB) were used to investigate the effect of lymphocyte depletion and blocking proBDNF on the impaired cognitive functions in the septic mice.

Results: In the septic mice, cognitive function was impaired, the percentage of CD4⁺ T cells were decreased in the meninges (P = 0.0021) and circulation (P = 0.0222), and pro-inflammatory cytokines were upregulated, but the anti-inflammatory cytokines interleukin (IL)-4 (P < 0.0001) and IL-13 (P = 0.0350) were downregulated in the meninges. Lymphocyte depletion by intragastrically treated FTY720 (1 mg kg^-1) for 1 week ameliorated LPS-induced learning deficit. In addition, proBDNF was increased in the meningeal (P = 0.0042) and peripheral (P = 0.0090) CD4⁺ T cells. Intraperitoneal injection of McAb-proB (100 μg) before LPS treatment significantly alleviated cognitive dysfunction, inhibited the downregulation of meningeal (P = 0.0264) and peripheral (P = 0.0080) CD4⁺ T cells, and normalized the gene expression of cytokines in the meninges. However, intra-cerebroventricular McAb-proB injection (1 μg) did not have such effect. Finally, exogenous proBDNF downregulated the percentage of CD4⁺ T cells in cultured splenocytes from septic mice (P = 0.0021).

Conclusion: Upregulated proBDNF in immune system promoted the pathogenesis of SAE through downregulating the circulating CD4⁺ T cells, limiting its infiltration into the meninges and perturbing the meningeal pro-/anti-inflammatory homeostasis.

Keywords: Brain-derived neurotrophic factor precursor; CD4+ T cells; Encephalopathy; Meningeal immunity; Monoclonal antibody; Sepsis.

Fig. 1

Impaired cognitive dysfunction after lipopolysaccharide (LPS) injection. Mice were i.p. injected with LPS (5 mg kg⁻¹) and behavior tests were conducted at 1 day after LPS injection. a There was a reduction in body weight between saline-treated and LPS-injected mice. n = 8 each group. The data was analyzed by repeated measures ANOVA followed by Bonferroni post hoc test, *P < 0.05, **P < 0.01. b LPS injection did not affect acquisition of fear conditioning test in mice. n = 8 each group. c,d Decreased freezing time in (c) contextual and (d) cued fear conditioning test in LPS-treated mice indicated that 5 mg kg⁻¹ LPS peritoneal injection induced cognitive dysfunction in mice. n = 8 each group. Data c was analyzed by unpaired T test and data b and d were analyzed by repeated measures ANOVA followed by Bonferroni post hoc test, *P < 0.05, **P < 0.01, ***P < 0.001. Data are presented as mean ± SEM. Con = saline injected control. LPS = i.p. LPS injection. i.p. = intraperitoneal

Fig. 2

Alterations in meningeal and peripheral immune cells in the septic mice. Mice were i.p. injected with LPS (5 mg kg⁻¹) and were sacrificed for the measurement of meningeal and peripheral immune cells by flow cytometry analysis. a Representative gating scheme of flow cytometry for meningeal immune cells. b Meningeal CD3⁺ T cells increased at 1 day and 5 days in LPS-treated mice relative to saline control. c, d LPS injection in mice decreased the percentage of (c) CD4⁺ T cells but increased the percentage of (d) CD8⁺ T cells in CD3⁺ T cells in the meninges. e Percentage of CD19⁺ B cells in CD45⁺ cells in the meninges was downregulated but (F) CD11b⁺ cells in the meninges was increased in mice given LPS injection as compared to saline injected control. n = 8 in the Con group and n = 6 in the LPS group. g–k LPS induced significantly increased percentage of g CD3⁺ T cells, i CD8⁺ T cells, and k CD11b⁺ cells among CD45⁺ cells, but decreased percentage of h CD4⁺ T cells and j CD19⁺ B cells in peripheral blood. n = 5 in each group. Data were analyzed by one-way ANOVA and followed by Tukey post hoc test, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Data are presented as mean ± SEM. Con = saline injected control. LPS 1 d = 1 day after i.p. LPS injection. LPS 5 d = 5 days after i.p. LPS injection. i.p. = intraperitoneal

Fig. 3

Distinct gene expression of cytokines in meninges after lipopolysaccharide (LPS) injection. Mice were i.p. injected with LPS (5 mg kg⁻¹), and meninges were collected for qPCR. The meninges of control mice were harvested at 24 h after they were injected with saline. a, b Upregulation of IL-1β and IL-6mRNA levels in the meninges after LPS injection. c–e Meningeal c IL-4, d IFN-γ, and e IL-13 mRNA levels were greatly downregulated after LPS injection. f The gene level of CD4 in the meninges was significantly lower at 1 day and 5 days of LPS injected mice than in saline control. n = 6 in each group. Experiments were performed in triplicate. Data were analyzed by one-way ANOVA and followed by Tukey post hoc test, *P < 0.05, **P < 0.01, ****P < 0.0001. Data are presented as mean ± SEM. Con = saline injected control. LPS 1 d = 1 day after i.p. LPS injection. LPS 5 d = 5 days after i.p. LPS injection. i.p. = intraperitoneal

Fig. 4

FTY720 pretreatment alleviated lipopolysaccharide (LPS)-induced fear conditioning memory deficit. Mice were intragastrically treated with FTY720 daily for a week followed by i.p. LPS (5 mg kg⁻¹) injection. Behavior tests were conducted and peripheral blood and meninges were collected at 1 day after LPS injection. a, b FTY720 significantly cleared a lymphocytes and b white blood cells in peripheral blood. n = 5 in each group. c The absolute number of CD3⁺ T cells in the meninges in FTY720-treated mice were dramatically lower than in saline-treated controls. d–f Neither the percentage of d CD3⁺ T cells in CD45⁺ cells nor the percentage of e CD4⁺ T cell or f CD8⁺ T cells in CD3⁺ T cells changed after LPS injection in FTY720 pre-treated mice. n = 5 in each group. Data a–f were analyzed by unpaired T test, **P < 0.01, ***P < 0.001. g FTY720 alone did not influence weight changes induced by LPS injection in mice. h FTY720 treatment did not influenced the fear conditioning acquiring, but it greatly reversed the reduced freezing time in the i contextual and j cued fear conditioning test induced by LPS injection. n = 6 in each group. Data g, h, and j were analyzed by two-way ANOVA and followed by Tukey post hoc test and data i was analyzed by one-way ANOVA and followed by Tukey post hoc test, *P < 0.05. Data are presented as mean ± SEM. Con = saline control. i.p. = intraperitoneal

Fig. 5

Increased proBDNF expression in meningeal and peripheral blood immune cells in septic mice. Mice were i.p. injected with LPS (5 mg kg⁻¹) and meninges and peripheral blood were harvested for immunofluorescence staining or flow cytometry. a Representative whole mount meningeal immunofluorescence images showed markedly increased proBDNF-positive staining cells in the meninges in mice at 1 day after LPS injection compared to saline injected mice. The high magnification images around the arrows are displayed in a white square as insets. Bar = 100 μm. b–f Representative meningeal single cell flow cytometry images (upper panel) and its statistical analysis (lower panel) indicated that proBDNF MFI was increased in meningeal b CD3⁺ T cells, c CD4⁺ T cells, d CD8⁺ T cells, and f CD11b⁺monocytes/macrophages at 1 day after LPS injection. proBDNF in meningeal e CD19⁺ B cells upregulated until 5 days after LPS injection. n = 10 in the Con group, n = 6 in LPS groups. g–k Upregulation of proBDNF in g CD3⁺ T cells, h CD4⁺ T cells, i CD8⁺ T cells, j CD19⁺ B cells, and k CD11b⁺ monocytes/macrophages in peripheral blood in LPS-injected mice were detected. n = 9 in Con group, n = 8 in the LPS1d group, n = 4 in the LPS 5d group. Data b–k were analyzed by one-way ANOVA and followed by Tukey post hoc test, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Data are presented as mean ± SEM. proB = proBDNF, Con = saline injected control. LPS 1d = 1 day after i.p. LPS injection. LPS 5d = 5 days after i.p. LPS injection. i.p. = intraperitoneal. MFI = mean fluorescence intensity

Fig. 6

Systemic blockade of proBDNF ameliorated cognitive dysfunction and restored meningeal and peripheral CD4⁺ T cell ratio in septic mice. Mice were i.p. injected with proBDNF 30 min before LPS (5 mg kg⁻¹) injection. Fear conditioning testing was performed 1 day after LPS injection. Meninges and peripheral blood were harvested 5 days after LPS injection for flow cytometry. a McAb-proB did not influence the weight of mice or b fear conditioning acquiring. c,d McAb-proB greatly alleviated memory deficit induced by LPS injection in mice as indicated by the increased freezing time in (c) contextual and (d) cued fear conditioning tests in the McAb-proB group relative to the IgG control. n = 8 in each group. Data a, b, and d were analyzed by repeated measures ANOVA and followed by Bonferroni post hoc tests. Data c was analyzed by one-way ANOVA and followed by Tukey post hoc test, *P < 0.05. e–f Representative meningeal e single cell flow cytometry and f analysis of its results indicated that McAb-proB greatly increased the percentage of CD4⁺ T cells among CD3⁺ T cells than IgG control in LPS injected mice. g–j McAb-proB did not change the infiltration of immune cells such as g CD3⁺ T cells, h CD8⁺ T cells, i CD19⁺ B cells, and j CD11b⁺monocytes/macrophages in the meninges after LPS injection relative to IgG control. n = 4 in each group. k–o McAb-proB greatly reversed the decreased percentage of l CD4⁺ T cells and the increased percentage of m CD8⁺ T cells in CD3⁺ T cells, but had no effect on percentage of k CD3⁺ T cells, n CD19⁺ B cells, or o CD11b⁺ monocytes/macrophages in CD45⁺ cells in the peripheral blood of septic mice. n = 6 in each group. Data f–o were analyzed by unpaired T test, *P < 0.05, **P < 0.01, ****P < 0.0001. Data are presented as mean ± SEM. Con = saline injected control, McAb-proB = monoclonal anti-proBDNF antibody injection, IgG = isotype Igg injected control

Fig. 7

Systemic blockade of proBDNF restored meningeal pro-inflammatory microenvironment in septic mice. Mice were i.p. injected with proBDNF 30 min before LPS (5 mg kg⁻¹) injection. Meninges were harvested 5 days after LPS injection for qPCR. a The level of CD4 gene expression was higher in the meninges of the McAb-proB group than in IgG controls in septic mice. b–f Gene levels were significantly lower in b IL-1β and c IL-6 but higher in d IL-4, e IFN-γ, and f IL-13 in the meninges after LPS injection in the McAb-proB group as compared to IgG control. n = 5 in each group. All experiments were performed at least in triplicate. Data were analyzed by unpaired T test, *P < 0.05, **P < 0.01, ****P < 0.0001. Data are presented as mean ± SEM. McAb-proB = monoclonal anti-proBDNF antibody injection, IgG = isotype Igg injected control

Fig. 8

Effect of i.c.v. injection of anti-proBDNF antibody on fear memory and meningeal immune activity in the septic mice. Mice were bilateral i.c.v. injected with 1 μg McAb-proB 3 days before LPS injection. Behavior tests were performed 1 day after LPS injection. a Representative images showed the broad and thorough diffusion of drugs in cerebroventricular lumen following i.c.v. injection with methylene blue. b McAb-proB i.c.v. injection did not influence weight of mice. c–e There was no difference of c fear conditioning acquiring performance in each group, nor McAb-proB increased the freezing time of d contextual or e cued fear conditioning test as compared to IgG control after LPS injection. n = 6 in each group. Data b, c, and e were analyzed by repeated measures ANOVA and followed by Bonferroni post hoc test and data d was analyzed by one-way ANOVA and followed by Tukey post hoc test, *P < 0.05, **P < 0.01. f–p Meninges were harvested 5 days after LPS injection for flow cytometry analysis and qPCR. f–j Flow cytometry indicated that McAb-proB i.c.v. injection did not alter immune cell subpopulation, except for the greater percentage of j CD11b⁺ monocytes/macrophages in CD45⁺ cells in the meninges, as compared to IgG control. n = 4 in each group. k–p No difference in CD4 gene level or gene expression of IL-1β, IL-6, IL-4, IFN-γ, and IL-13 in the meninges between McAb-proB-injected mice and isotype-injected control after LPS injection. n = 5 in each group. Experiments were performed at least in triplicate. Data f–p were analyzed by unpaired T test, *P < 0.05, **P < 0.01. Data are presented as mean ± SEM. Con = saline injected control, McAb-proB = monoclonal anti-proBDNF antibody injection, IgG = isotype Igg injected control, i.c.v. = intracerebroventricular

Fig. 9

Exogenous proBDNF protein reduced CD4⁺ T cells but increased CD8⁺ T cells of septic mice in vitro. Mice injected with saline or LPS (5 mg kg⁻¹) for 5 days and the splenocytes were isolated and cultured for 3 days in vitro. Exogenous proBDNF did not alter the percentage of a CD3⁺ T cells in CD45⁺ cells or the percentage of b CD4⁺ T cells or c CD8⁺ T cells in CD3⁺ T cells in splenocytes from mice treated with saline. d Exogenous proBDNF did not alter the percentage of CD3⁺ T cells in CD45⁺ cells in splenocytes in septic mice. e–f ProBDNF treatment significantly decreased the percentage of e CD4⁺ T cells but increased the percentage of f CD8⁺ T cells in CD3⁺ T cells in splenocytes in LPS-treated mice. n = 4 in each group. Data were analyzed by one-way ANOVA and followed by Tukey post hoc test, *P < 0.05, **P < 0.01. Data are presented as mean ± SEM. NT = no treated control. proB = exogenous proBDNF protein treatment

Fig. 10

Schematic diagram showing how proBDNF dampens CD4⁺ T cell activity and contributes to the pathogenesis of SAE. In sepsis, proBDNF expression is increased in peripheral blood and meningeal immune cells, which then decreases the infiltration of CD4⁺ T cells in the meninges. As a result, meningeal pro-inflammatory cytokines such as IL-6 and IL-1β are upregulated, but anti-inflammatory cytokines including IL-4 and IL-13 are downregulated, finally leading to SAE. SAE, sepsis-associated encephalopathy