Rapamycin Modulates the Proinflammatory Memory-Like Response of Microglia Induced by BAFF

Abstract

Background: Recently trained immunity of microglia provided an opportunity to study the chronic effect of microglial activation and its metabolic rewiring in neuroimmunological diseases. Since elevated levels of B cell-activating factor (BAFF) have been proved to be associated with some chronic neuroimmunological disorders. Here, we used the trained innate immunity model to analyze the effect of BAFF, a vital regulator of the adaptive immune system, on long-term microglial activation and metabolic reprogramming in vitro and in vivo.

Methods and results: In vitro, BV2 cells and mouse primary microglial cells were incubated with BAFF for 24 h (BAFF priming). After 5 days of resting, microglia were restimulated with LPS (LPS restimulation) or BAFF (BAFF restimulation). BAFF priming induced a pro-inflammatory trained immunity-phenotype of both BV2 cells and primary microglial cells, which was indicated by morphological change, secretion of pro-inflammatory cytokine and chemokine upon LPS restimulation or BAFF restimulation. The production of lactate and NAD+/NADH ratio were elevated 5 days after BAFF priming. The activation of the Akt/mTOR/HIF-1α pathway was induced by BAFF priming and lasted for 5 days. Pretreating the BV2 cells or mouse primary microglial cells with rapamycin blocked mTOR/HIF-1α activation and cellular metabolic reprogramming induced by BAFF training. Consistently, rapamycin efficiently suppressed the trained immunity-like responses of microglia triggered by BAFF. In vivo, adult male mice were treated with BAFF by intracerebroventricular injection for priming and 7 days later with BAFF for restimulation. BAFF training activated microglia in the cortex and hippocampus. The production of proinflammatory cytokines and chemokines was elevated after BAFF training.

Conclusion: Our current data, for the first time, demonstrate that BAFF priming induces a proinflammatory memory-like response of microglia not only to LPS but also to BAFF itself. Rapamycin inhibits microglial priming triggered by BAFF through targeting the mTOR/HIF-1α signaling pathway. Our data reveal a novel role of BAFF in trained immunity and that rapamycin may be a potential therapeutic target of neuroimmunological diseases.

Keywords: B cell-activating factor; aerobic glycolysis; mammalian target of rapamycin; microglia; trained immunity.

Figure 1

BAFF priming induced a proinflammatory memory response of BV2 cells to LPS and BAFF itself. (A) Graphical outline of the in vitro methods. BV2 cells or primary mouse microglial cells were incubated for 24 h with vehicle or BAFF (defined as ‘BAFF priming’), and after a 5-day rest period, microglial cells were restimulated with 40 ng/ml BAFF (BAFF restimulation) or 100 ng/ml LPS (LPS restimulation) for 24 h. In (B–D), BV2 cells were primed with different doses of BAFF (2.5, 5, 10 ng/ml) or vehicle for 24 h, rested for 5 days, on day 6 treated with LPS restimulation for 24 h. TNF-α (B), IL-6 (C), and IL-1β (D) were measured in the supernatants by ELISA on day 7(n = 12). On day 6, BV2 cells were restimulated with BAFF instead of LPS. On day 7, TNF-α (E), IL-6 (F) and IL-1β (G) were measured in the supernatants (n = 12). Data are means ± SED; Kruskal-Wallis test with Turkey post hoc test; *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 2

BAFF priming was associated with the induction of glycolysis. BV2 cells were primed with BAFF or vehicle for 24 h and then rested for 5 days. The production of lactate in the supernatants (n = 12) and NAD+/NADH ratio (n = 12) in the whole-cell lysates were assayed 24 h after BAFF priming (A, B) and on day 6 before restimulation (C, D) respectively. Data are means ± SED; Kruskal-Wallis test with Turkey post hoc test; *p < 0.05 and **p < 0.01 compared with CTL.

Figure 3

Inhibition of glycolysis by 2-DG blocked proinflammatory cytokines secretion of BAFF-training BV2 cells. BV2 cells were pretreated with 1mM 2-DG for 1 h, then primed with vehicle or BAFF for 24 h. After a rested of 5 days, cells were restimulated with LPS or BAFF for 24 h. The amounts of TNF-α (A) and IL-6 (B) were measured in the culture supernatants and the cytotoxicity of 2-DG was measured by MTT (C) on day 7 upon LPS restimulation (n = 12). The secretion of TNF-α (D) and IL-6 (E) and the cytotoxicity of 2-DG (F) were measured on day 7 upon BAFF restimulation (n = 12). The Data are mean ± SEM; Kruskal-Wallis test with Turkey post hoc test; ***p < 0.001. ns: no significance between groups.

Figure 4

BAFF priming induces an epigenetic trained immunity phenotype in BV2 cells. (A, B) BV2 cells were treated with 10ng/ml BAFF or vehicle for 24 h and kept for 5 days. ChIP assay was performed using an antibody against histone 3 lysine 4 trimethylation (H3K4me3) or control IgG. The enrichment of H3K4me3 at the promoters of TNF-α (A) and IL-6 (B) was quantified by quantitative real-time PCR. The results are expressed as % input and data are means ± SED (n = 6); t-test; *p < 0.05. (C, D) BV2 cells were treated as indicated with 10 ng/ml BAFF, 20 mM MTA (histone methyltransferase inhibitor) or vehicle for 24 h, rested for 5 days, on day 6 treated with LPS restimulation for 24 h. TNF-α (C) and IL-6 (D) were measured in the supernatants by ELISA on day 7(n = 12). On day 6, BV2 cells were restimulated with BAFF instead of LPS. On day 7, TNF-α (E) and IL-6 (F) were measured in the supernatants (n = 12). Data are means ± SED (n = 12); Kruskal-Wallis test with Turkey post hoc test; *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 5

TACI and Akt/mTOR/HIF-1α signaling were involved in BAFF priming of BV2 cells. BV2 cells were primed with 10ng/ml BAFF or vehicle for 24 h. (A) RNA was collected and the relative mRNA expression of TACI, BAFF-R, and BCMA, relative to GAPDH, are assessed in BV2 cells by quantitative real-time PCR. (B, C) The expression of BAFF receptors on the surface of microglia was detected by flow cytometry. BV2 cells are stained with primary antibodies for TACI, BAFF-R, and BCMA (solid lines without filling) or an isotype control Ab (solid lines filled with gray). (B) The graph represents the relative fluorescence intensity of the primary antibody subtracting its isotype control antibody. (D) 6 days after BV2 cells were primed by BAFF or vehicle, whole-cell extracts were collected and levels of Akt, p-Akt, mTOR, p-mTOR, and HIF-1α were detected by western blot. Densitometric quantification of p-Akt/Akt (E), p-mTOR/mTOR (F), and HIF-1α/tubulin (G) in (D). Data are mean ± SEM (n = 9); Kruskal-Wallis test with Turkey post hoc test; *p < 0.05, **p < 0.01 compared with CTL.

Figure 6

BAFF-induced metabolic reprogramming was blocked by rapamycin in BV2 cells. (A) On day 0, BV2 cells were pretreated with 100 nM rapamycin for 1 h and then treated with vehicle or BAFF (10 ng/ml) for 24 h, rested for 5 days. BV2 cells were lysed on day 6. Whole-cell extracts were subjected to western blot for the expression level of mTOR, p-mTOR, and HIF-1α. Densitometric quantification of p-mTOR/mTOR (B) and HIF-1α/tubulin (C) in (A) (n = 9). (D) Lactate was measured in the culture supernatants on day 6 (n = 12). (E) The ratio of NAD+/NADH was detected in whole-cell lysates on day 6 (n = 9). Data are mean ± SEM; Kruskal-Wallis test with Turkey post hoc test; **p < 0.01, and ***p < 0.001.

Figure 7

Rapamycin inhibited BAFF-induced trained immunity phenotype of BV2 cells. BV2 cells were pretreated with 100 nM rapamycin for 1 h, then primed with vehicle or BAFF for 24 h. After a rested of 5 days, cells were restimulated with BAFF or LPS for 24 h. (A) Representative micrographs of BV2 cells on day 7(n = 9). The amounts of TNF-α (B) and IL-6 (C) were measured in the culture supernatants on day 7 upon LPS restimulation (n = 12). The amounts of TNF-α (D) and IL-6 (E) were measured in the culture supernatants on day 7 upon BAFF restimulation (n = 12). Data are mean ± SEM; Kruskal-Wallis test with Turkey post hoc test; *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 8

Inhibition of Akt or HIF-1α suppressed BAFF training of BV2 cells in a dose-dependent manner. (A–C) BV2 cells were pretreated with different doses of MK2206, a specific Akt inhibitor (50, 100, 200 nM) for 3 h, and then primed with BAFF for 24 h. After a rested of 5 days, cells were restimulated with BAFF for 24 h. The amounts of TNF-α (A) and IL-6 (B) were measured in the culture supernatants and the cytotoxicity of MK2206 was measured by MTT (C) on day 7(n = 12). (D-F) BV2 cells were pretreated with different doses of 2-Methoxyestradiol (2-ME), a specific HIF-1α inhibitor (0.5, 1, 2 μM) for 3 h, and then primed with BAFF for 24 h. The secretion of TNF-α (D) and IL-6 (E) and the cytotoxicity of 2-ME (F) were measured on day 7 upon BAFF restimulation (n = 12). The data are mean ± SEM; Kruskal-Wallis test with Turkey post hoc test; *p < 0.05, **p < 0.01.

Figure 9

Rapamycin suppressed HIF-1α and metabolic reprogramming in primary microglial cells induced by BAFF priming. Primary microglia were pretreated with 100 nM rapamycin for 1 h, then treated with BAFF (10 ng/ml) or vehicle for 24 h, and rested for 5 days. (A) Whole-cell extracts were collected on day 6 and levels of Akt, p-Akt, mTOR, p-mTOR, and HIF-1α were detected by Western blot. Densitometric quantification of p-Akt/Akt (B), p-mTOR/mTOR (C), and HIF-1α/tubulin (D) in (A). (E) The production of lactate in the supernatants was measured on day 6; (F) The ratio of NAD+/NADH in the whole-cell lysates was measured on day 6. Data are mean ± SEM (n = 9); Kruskal-Wallis test with Turkey post hoc test; *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 10

Rapamycin suppressed proinflammatory memory-like state of primary microglial cells induced by BAFF training. Primary microglia were pretreated with 100 nM rapamycin for 1 h and then treated with BAFF (10 ng/ml) or vehicle for 24 h. On day 6, microglial cells were restimulated with BAFF (40 ng/ml) for 24 h. (A) The representative micrographs of primary microglial cells on day 7. The amounts of TNF-α (B) and IL-6 (C) were measured in the culture supernatants on day 7. (D) Cell viability of primary microglial cells on day 7 was tested by MTT assay. Data are mean ± SEM (n = 9); Kruskal-Wallis test with Turkey post hoc test; *p < 0.05, **p< 0.01, and ***p< 0.001, ns: no significance between groups.

Figure 11

BAFF induced microglial training in vivo. Immunofluorescence staining of Iba1 in the cortex (A) and hippocampus (C). Comparisons of the numbers of Iba1-positive microglia in the cortex (B) and hippocampus (D) (n = 6). A representative Iba-1+ cell from the cortex is shown magnified in the red box in the image’s right upper corner. The expression of TNF-α (E), IL-6 (F), CCL2 (G) were measured in the supernatants by ELISA on day 7(n = 6). Data are mean ± SEM; Kruskal-Wallis test with Turkey post hoc test; **p < 0.01, and ***p < 0.001.