25-Hydroxycholesterol secreted by macrophages in response to Toll-like receptor activation suppresses immunoglobulin A production

Abstract

25-Hydroxycholesterol is produced in mammalian tissues. The function of this oxysterol is unknown. Here we describe a central role for 25-hydroxycholesterol in regulating the immune system. In initial experiments, we found that stimulation of macrophage Toll-like receptors (TLR) induced expression of cholesterol 25-hydroxylase and the synthesis of 25-hydroxycholesterol. Treatment of naïve B cells with nanomolar concentrations of 25-hydroxycholesterol suppressed IL-2-mediated stimulation of B cell proliferation, repressed activation-induced cytidine deaminase (AID) expression, and blocked class switch recombination, leading to markedly decreased IgA production. Consistent with these findings, deletion of the mouse cholesterol 25-hydroxylase gene caused an increase in serum IgA. Conversely, inactivation of the CYP7B1 oxysterol 7alpha-hydroxylase, which degrades 25-hydroxycholesterol, decreased serum IgA. The suppression of IgA class switching in B cells by a macrophage-derived sterol in response to TLR activation provides a mechanism for local and systemic negative regulation of the adaptive immune response by the innate immune system.

Fig. 1.

TLR4 activation in intraperitoneal and bone marrow–derived macrophages leads to induction of CH25H. (A) Time course of 25-hydroxycholesterol synthesis in intraperitoneal macrophages challenged with 100 ng/mL KDO vs. PBS. Error bars in panels A and D show the mean ± SEM for data obtained from 3 independent experiments. Inset: structure of 25-hydroxycholesterol. (B) Time course of CH25H mRNA induction in intraperitoneal macrophages treated with KDO vs. PBS. C_T values for mRNA levels determined by qPCR at time 0 and 8 h are indicated. Data in panels B and E are representative of >10 experiments. (C) Induction of CH25H protein in response to KDO in intraperitoneal macrophages isolated from wild-type mice (WT) and Ch25h^−/− mice. The positions of the CH25H enzyme and a cross-reacting protein of unknown identity (X) are indicated on the blot. A calnexin control for protein loading is shown in the bottom film. Data in panels C and F are from 1 of 3 experiments. (D) Time course of 25-hydroxycholesterol synthesis in bone marrow–derived macrophages challenged with KDO vs. PBS. (E) Time course of CH25H mRNA induction in bone marrow–derived macrophages treated with KDO vs. PBS. C_T values for mRNA levels determined by qPCR at time 0 and 4 h are indicated. (F) Induction of CH25H protein in response to KDO in bone marrow–derived macrophages isolated from wild-type mice (WT) and Ch25h^−/− mice. A calnexin control for protein loading is shown in the bottom film.

Fig. 2.

Induction and inhibition of CH25H gene expression in intraperitoneal macrophages treated with different TLR agonists and NF-κB/MAPK antagonists. (A) Time course of CH25H mRNA induction in response to KDO and E. coli and S. enterica LPS. Data in panels A–D are representative of 3 experiments. (B) Induction of CH25H mRNA by activation of TLR4 (KDO), TLR2/6 (LTA, lipoteichoic acid), TLR2 (PG, peptidoglycan from S. aureus), or TLR3 (poly I:C). (C) Inhibition of JNK, a MAPK family member, with SP600125 blocks TLR4-mediated induction of CH25H mRNA. (D) Inhibition of NF-κB with curcumin blocks TLR4-mediated induction of CH25H mRNA.

Fig. 3.

Induction of CH25H mRNA, protein, and product in mice treated with KDO. (A) C57BL6/J wild-type mice (n = 3) were treated with PBS or KDO for 8 or 16 h, the indicated tissues were harvested, and CH25H mRNA levels determined in pools of total RNA by qPCR. Data in panels A and B are from 1 of 2 experiments. (B) Microsomal membrane proteins isolated from the lung and liver of wild-type or Ch25h^−/− mice treated for 24 h with PBS or KDO were blotted for CH25H. A loading control consisting of the calnexin protein is shown in the lower blot. (C) 25-Hydroxycholesterol levels were quantified in the lungs and sera of wild-type or Ch25h^−/− mice (n = 3 per genotype, 1 experiment) treated for 16 or 24 h with PBS or KDO. Error bars show the mean ± SEM.

Fig. 4.

Ig levels in wild-type, Ch25h^−/−, and Cyp7b1 mice. (A) IgM_H, IgA_H, and IgJ mRNAs were measured by qPCR in the indicated tissues of wild-type (WT) or Ch25h^−/− mice (n = 3 per genotype, 1 experiment). C_T values determined for IgA_H mRNA in each tissue are indicated above the black histogram bars. LN, axiliary and inguinal lymph nodes. (B) Levels of IgA in serum, BAL (lung), and small intestinal mucosal fluid isolated from individual wild-type, Ch25h^−/−, and Cyp7b1^−/− mice as determined by a qualitative assay. Horizontal bars indicate means. *P values for differences between IgA levels in WT vs. Ch25h^−/− mice were 1 × 10⁻¹⁰ for serum, 2 × 10⁻⁶ for lung, and 1 × 10⁻⁸ for mucosa. P values for differences between IgA levels in wild-type vs. Cyp7b1^−/− mice were 1 × 10⁻⁹ for serum, 1 × 10⁻⁷ for lung, and 2 × 10⁻³ for mucosa. Data are derived from 2–6 experiments involving indicated the numbers of mice. (C) Levels of IgA in serum, lung, and small intestinal mucosa from wild-type and Ch25h^−/− mice as determined by a quantitative assay. *P values for differences between IgA levels in wild-type vs. Ch25h^−/− mice were 5 × 10⁻⁴ in serum, 2 × 10⁻³ in lung, and 5 × 10⁻³ for mucosa. Data are averages from 3 experiments (n = 8 per genotype per experiment for serum and lung, n = 4 per experiment per genotype for mucosa).

Fig. 5.

25-Hydroxycholesterol suppresses IgA CSR in vitro. (A) Induction of CSR in splenic B220⁺ cells by cytokines. Cells were treated with LPS for 16 h, followed by different combinations of cytokines. On the indicated day, IgA levels were determined in the cellular medium by ELISA. Data in panels A–C are averages from 3 experiments. (B) Time course of 25-hydroxycholesterol suppression of CSR. B220⁺ cells were treated with LPS and cytokines to induce CSR. On the indicated day, 250 nM 25-hydroxycholesterol was added to the medium and IgA levels determined on day 6 of the experiment. (C) 25-Hydroxycholesterol dose–response data. B220⁺ cells were treated with LPS and cytokines in the presence of different amounts of oxysterol added on day 0 of the experiment. IgA levels were determined in the medium on day 6. (D) Oxysterol specificity for CSR suppression. B220⁺ cells were induced to undergo CSR by LPS and cytokine treatments in the presence of the indicated sterol added on day 0 of the experiment. IgA levels were determined on day 6. Sterols and concentrations used were as follows: 25-hydroxycholesterol (25-HC), 250 nM; cholesterol (C), 250 nM; 27-hydroxycholesterol (27-HC), 250 nM; 22(R)-hydroxycholesterol (22-HC), 250 nM; 24(R/S)-hydroxycholesterol (24-HC), 500 nM; and 24-dihydrolanosterol (DHL), 250 nM. Data are averages from 2 experiments.

Fig. 6.

Effect of 25-hydroxycholesterol on B cell proliferation and AID expression. (A) Splenic B220⁺ cells were treated with the indicated agents in the presence of ³H-thymidine. Incorporation of the radiolabeled nucleotide into acid-precipitable DNA was determined 56 h later. The decreases measured in the presence of 25-hydroxycholesterol did not reach statistical significance for any of the conditions tested. Data in panels A and B are averages from 3 experiments. (B) B220⁺ cells were treated for 16 h with LPS, followed by cytokine combinations and different amounts of 25-hydroxycholesterol as indicated. On day 6 of the experiment, cellular DNA was quantified using a fluorescence-based assay. (C) Splenic B220⁺ cells were treated with the indicated agents for 3 days. Levels of AID, IgA_H, RelA, and IgM_H mRNAs were measured by qPCR. Data are from 1 of 2 experiments.