Cholesterol Accumulation in CD11c+ Immune Cells Is a Causal and Targetable Factor in Autoimmune Disease

Abstract

Liver X receptors (LXRs) are regulators of cholesterol metabolism that also modulate immune responses. Inactivation of LXR α and β in mice leads to autoimmunity; however, how the regulation of cholesterol metabolism contributes to autoimmunity is unclear. Here we found that cholesterol loading of CD11c⁺ cells triggered the development of autoimmunity, whereas preventing excess lipid accumulation by promoting cholesterol efflux was therapeutic. LXRβ-deficient mice crossed to the hyperlipidemic ApoE-deficient background or challenged with a high-cholesterol diet developed autoantibodies. Cholesterol accumulation in lymphoid organs promoted T cell priming and stimulated the production of the B cell growth factors Baff and April. Conversely, B cell expansion and the development of autoantibodies in ApoE/LXR-β-deficient mice was reversed by ApoA-I expression. These findings implicate cholesterol imbalance as a contributor to immune dysfunction and suggest that stimulating HDL-dependent reverse cholesterol transport could be beneficial in the setting of autoimmune disease.

Keywords: LXR; autoantibodies; autoimmune disease; reverse cholesterol transport.

Figure 1. Deletion of Lxrβ on an ApoE-null background is sufficient to provoke autoimmune disease

(A) Kidney sections from indicated ages of Apoe^−/− and Apoe^−/−Lxrβ^−/− mice immunostained for IgG, B220, CD3 and CD68. (B) Plasma samples from 8-week-old and 24-week-old Apoe^−/− and Apoe^−/−Lxrβ^−/− mice were analyzed for anti-nuclear antibody (ANA) titers by ELISA. (C) Flow cytometric analysis of CD19⁺B220⁺ B cells from lymph node of 8-week-old Apoe^−/− and Apoe^−/− Lxrβ^−/− mice. (D) Percentages of indicated cell populations in lymph node of Apoe^−/− and Apoe^−/−Lxrβ^−/− mice analyzed by flow cytometry. (E) Cell counts of the indicated cell populations in lymph node of ApoE^−/− and Apoe^−/−Lxrβ^−/− mice analyzed by flow cytometry. N=4–6 per group. Statistical analysis was performed with Student’s t test. *p < 0.05, **p < 0.01, NS, not significant. Error bars represent means +/− SEM. See also Figure S1.

Figure 2. Feeding a high cholesterol diet promotes autoimmunity

(A) Flow cytometric analysis of CD19⁺B220⁺ B cells from lymph node of wild-type and Lxrβ^−/− mice fed Western diet for 16 weeks. (B) Percentages and cell counts of the indicated cell populations in lymph node of wild-type and Lxrβ^−/− mice fed chow or Western diet for 8 weeks or 16 weeks analyzed by flow cytometry. (C) Flow cytometric analysis of CD19⁺B220+ B cells from lymph node of wild-type and Lxrαβ^−/− mice fed Western diet for 12 weeks. (D) Percentages and cell counts of the indicated cell populations in lymph node of wild-type and Lxrαβ^−/− mice fed chow or Western diet for 12 weeks analyzed by flow cytometry. (E) Plasma samples from wild-type and Lxrαβ^−/− mice fed chow and Western diet for 12 weeks analyzed for the titer of ANA by ELISA. (F) Plasma samples from wild-type and Lxrαβ^−/− mice fed chow or Western diet for 12 weeks analyzed for the titers of total IgM, IgM against Cu-OxLDL, MDA-LDL and EO6 by ELISA. N=4–6 per group. Statistical analysis was performed with two-way ANOVA. *p < 0.05, **p < 0.01, NS, not significant. Error bars represent means +/− SEM. See also Figure S2.

Figure 3. Cholesterol accumulation in lymphoid organs promotes the production of Baff and April

(A) Lipid was extracted from lymph node, spleen and liver of 8-week-old Apoe^−/− and Apoe^−/−Lxrβ^−/− mice. Total masses of cholesterol and triglyceride were determined by colorimetric methods. N=3. (B) Gene expression in lymph node of 8-week-old Apoe^−/− and Apoe^−/−Lxrβ^−/− mice analyzed by real-time PCR. (C) Plasma Baff concentration in 8-week-old ApoE^−/− and Apoe^−/−Lxrβ^−/− mice determined by ELISA. (D) Gene expression in lymph node (upper) and spleen (bottom) of wild-type and Lxrβ^−/− mice fed chow or Western diet for 16 weeks analyzed by real-time PCR. (E) Gene expression in lymph node (upper), spleen (middle) and CD11c⁺ antigen-presenting cells (APC) (bottom) of wild-type and Lxrαβ^−/− mice fed chow or Western diet for 12 weeks analyzed by real-time PCR. (F) Plasma Baff concentration in wild-type and Lxrαβ^−/− mice fed chow or Western diet for 12 weeks determined by ELISA. (G) Pan B cells, pan T cells and APC were isolated from spleen of wild-type mice. Gene expression was analyzed by real-time PCR. N=4–6 per group. Statistical analysis was performed with Student’s t test (A–C) and two-way ANOVA (D–F). *p < 0.05, **p < 0.01, NS, not significant. Error bars represent means +/− SEM.

Figure 4. Loss of LXRβ expression in hematopoietic cells promotes the development of autoimmune disease

Bone marrow cells from Apoe^−/− and Apoe^−/−Lxrβ^−/− mice were transplanted into Apoe^−/− mice. Mice were analyzed 12 weeks after the transplantation. (A) Gene expression in from lymph node and spleen of Apoe^−/− mice transplanted with bone marrow cells from Apoe^−/− or Apoe^−/− Lxrβ^−/− mice analyzed by real-time PCR. (B) Percentages of indicated cell populations in lymph node and spleen of Apoe^−/− mice transplanted with bone marrow cells from Apoe^−/− or Apoe^−/−Lxrβ^−/− mice analyzed by flow cytometry. (C) Plasma Baff concentration in Apoe−/− mice transplanted with bone marrow cells from Apoe^−/− or Apoe^−/−Lxrβ^−/− mice determined by ELISA. N=6 per group. Statistical analysis was performed with Student’s t test. *p < 0.05, **p < 0.01, NS, not significant. Error bars represent means +/− SEM.

Figure 5. Loss of LXRβ in lymphocyte does not affect the development of autoimmune disease

(A) Pan-B cells and non-B cells were isolated from spleen of Lxrβ^F/F (F/F) and Lxrβ^F/F; CD19-Cre (B-KO) mice. Cells were treated with GW3965 (1μM) overnight. Gene expression was analyzed by real-time PCR. (B) Pan-T cells and non-T cells were isolated from spleen of Lxrβ^F/F (F/F) and Lxrβ^F/F; LCK-Cre (T-KO) mice. Cells were treated with GW 3965 (1μM) overnight. Gene expression was analyzed by real-time PCR. (C) Percentages and cell counts of the indicated cell populations in lymph node of Apoe^−/−Lxrβ^F/F and ApoE^−/−Lxrβ^F/F; CD19-Cre mice analyzed by flow cytometry. (D) Gene expression in from lymph node of Apoe^−/−Lxrβ^F/Fand Apoe^−/−Lxrβ^F/F; CD19-Cre mice analyzed by real-time PCR. (E) Percentages and cell counts of the indicated cell populations in lymph node of Apoe^−/−Lxrβ^F/Fand Apoe^−/−Lxrβ^F/F; Lck-Cre mice analyzed by flow cytometry. (F) Gene expression in lymph node from Apoe^−/−Lxrβ^F/Fand Apoe^−/−Lxrβ^F/F; Lck-Cre mice analyzed by real-time PCR. N=4–5 per group. Statistical analysis was performed with Student’s t test. *p < 0.05, NS, not significant. Error bars represent means +/− SEM. See also Figures S3 and S4.

Figure 6. Altered antigen-presenting cell function in LXR-deficient mice

(A) Lipid content in CD11c⁺ APCs was analyzed by staining cells from lymph node and spleen of wild-type and Lxrαβ^−/− mice fed Western diet for 12 weeks with BODIPY. MFI in CD11c⁺ APC population was determined by flow cytometry. (B) CFSE dilution of adoptively-transferred OT-1 T cells from lymph node and spleen of wild-type and Lxrαβ^−/− mice fed Western diet for 16 weeks. On day 3 after challenge with ovalbumin (200 μg/mouse), CFSE-labeled OT-1 T cells were assessed by flow cytometry gating on CD8⁺TCR Va2⁺. (C) Bone marrow-derived dendritic cells were incubated with hydroxypropyl-β-cyclodextrin (HβCD, 10 mM) for 1 hr. Gene expression was analyzed by real-time PCR. (D) Bone marrow–derived dendritic cells were pretreated with GW3965 (1 μM) overnight, followed by stimulation with Pam3CSK4 (100 ng/ml) or CpG (1 μM) for 24 hr. Gene expression was analyzed by real-time PCR. (E) CD11c⁺ and CD11c⁻ cells were isolated from spleen of Apoe^−/−Lxrβ^F/F and Apoe^−/−Lxrβ^F/F; Cd11c-Cre mice. Cells were treated with GW3965 (1μM) overnight. Gene expression was analyzed by real-time PCR. (F) Cell counts of the indicated cell populations in lymph node of ApoE^−/−Lxrβ^F/F and ^−/−Lxrβ^F/F; Cd11c-Cre mice at 8 weeks of age analyzed by flow cytometry. (G) Gene expression in lymph node and spleen of Apoe^−/−Lxrβ^F/F and Apoe^−/−Lxrβ^F/F; CD11c-Cre mice analyzed by real-time PCR. N=4–6 per group. Statistical analysis was performed with Student’s t test (A, C, F, G) and two-way ANOVA (D). *p < 0.05, **p < 0.01. Error bars represent means +/− SEM. See also Figures S5 and S6.

Figure 7. ApoA-I expression ameliorates autoimmune disease in mice deficient in ApoE and LXRβ

6 week-old Apoe^−/− and Apoe^−/−Lxrβ^−/− mice were injected with Hd-Ad-control or Hd-Ad-ApoA-I. The mice were analyzed on 12 weeks after the injection. (A) Gene expression in liver of Apoe^−/− and Apoe^−/−Lxrβ^−/− mice injected Hd-Ad-control or Hd-Ad-ApoA-I. (B) Percentages of indicated cell populations in lymph node and spleen of Apoe^−/− and Apoe^−/− Lxrβ^−/− mice injected Hd-Ad-control or Ad-ApoA-I were analyzed by flow cytometry. (C) Gene expression in lymph node and spleen of Apoe^−/− and Apoe^−/−Lxrβ^−/− mice injected with Hd-Ad-control or Ad-ApoA-I was analyzed by real-time PCR. (D) Plasma samples from Apoe^−/− and Apoe^−/−Lxrβ^−/− mice injected with Hd-Ad-control or Ad-ApoA-I were analyzed for the presence of Baff (upper) and ANA (bottom) by ELISA. N=4–6 per group. Statistical analysis was performed with two-way ANOVA. *p < 0.05, **p < 0.01. Error bars represent means +/− SEM. See also Figure S7.