Class A scavenger receptors regulate tolerance against apoptotic cells, and autoantibodies against these receptors are predictive of systemic lupus

Abstract

Apoptotic cells are considered to be a major source for autoantigens in autoimmune diseases such as systemic lupus erythematosus (SLE). In agreement with this, defective clearance of apoptotic cells has been shown to increase disease susceptibility. Still, little is known about how apoptotic cell-derived self-antigens activate autoreactive B cells and where this takes place. In this study, we find that apoptotic cells are taken up by specific scavenger receptors expressed on macrophages in the splenic marginal zone and that mice deficient in these receptors have a lower threshold for autoantibody responses. Furthermore, antibodies against scavenger receptors are found before the onset of clinical symptoms in SLE-prone mice, and they are also found in diagnosed SLE patients. Our findings describe a novel mechanism where autoantibodies toward scavenger receptors can alter the response to apoptotic cells, affect tolerance, and thus promote disease progression. Because the autoantibodies can be detected before onset of disease in mice, they could have predictive value as early indicators of SLE.

Figure 1.

Apoptotic cells bind MARCO and SR-A, and apoptotic cells are trapped in the marginal zone of the spleen. (A) PKH26-labeled (red) apoptotic cells were injected i.v. in BALB/c mice, and spleens collected after 30 min were stained with anti-CD11c (DCs; green) and anti-B220 (B cells; pseudo-colored blue). Bar, 100 μm. (B; top) A serial cryostat section stained with anti-MARCO (MZMs; green). Bar, 100 μm. (Bottom) Higher magnification of MZM-binding apoptotic cells (red) stained with both anti-MARCO and anti–SR-A (green). Arrows indicate apoptotic cells taken up by MZMs. Bar, 10 μm. (C) In vitro binding assay using CHO cells transfected with mouse MARCO or SR-A and incubated with labeled apoptotic cells (red). (Left) Cells stained with anti-MARCO or anti–SR-A (green) and DAPI nuclear staining (blue). Bar, 10 μm. (Right) Quantification of apoptotic cell binding showing the amount of apoptotic cells binding transfected or nontransfected cells on the same slide. 10–15% of the total amount of CHO cells was successfully transfected. In A and B the Leica confocal system was used, and in C the Leica DMRB microscope was used (refer to Materials and methods).

Figure 2.

Class A scavenger receptors regulate tolerance against i.v. injected apoptotic cells. (A) 10⁷ syngeneic apoptotic cells were injected i.v. four times weekly in WT and MARCO^−/−/SR-A^−/− DKO mice (C57BL/6 background). IgM and IgG anti-DNA responses in serum were measured pre-immune (PI) at days 12 and 19. Data are shown as mean ± SEM (n = 8 per genotype). (B) The anti-DNA response in WT, MARCO^−/−, SR-A^−/−, and DKO mice at days 12 (IgM) and 19 (IgG). Individual data and mean are presented (n = 8 per genotype). As a control in these experiments, a quantitative analysis of serial dilution of pooled MRL^lpr sera that develop spontaneous disease revealed that the levels of IgM and IgG anti-DNA antibodies are ∼5–40 times that of WT mice injected with apoptotic cells. The PI values for MARCO^−/− and SR-A^−/− were not statistically different from WT controls (not depicted). (C) Subclass analysis of the anti-DNA response at day 26, after the fourth injection of apoptotic cells. Data are shown as mean ± SEM of the OD 405-nm ratio between IgG_2a/IgG_2b and IgG₁ (n = 8 per genotype). (D) Representative ANA pattern from DKO and WT mice after the fourth injection (d26). Bar, 50 μm. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (nonparametric Mann-Whitney U-test).

Figure 3.

Anti-MARCO antibodies are found in SLE-prone mice. (A and C) IgG anti-DNA levels in 2–8-mo-old (NZB x NZW)F1 mice and in 3–14-mo-old FcγRIIB^−/− mice. Control mice are 3-mo-old C57BL/6 mice. (B and D) IgG anti-MARCO reactivity in 2–8-mo-old (NZB x NZW)F1 mice, 3–14-mo-old FcγRIIB^−/− mice, and 3-mo-old C57BL/6 mice. (E) Binding to MARCO in the anti-MARCO ELISA blocked with a rat monoclonal antibody (ED31) against MARCO, but not with an isotype control (n = 2). (F) CHO cells transfected with mouse MARCO or SR-A stained with sera from (NZB x NZW)F1 mice and anti–mouse IgG-FITC. Arrows indicate transfected cells stained by the mouse sera. Bar, 20 μm.

Figure 4.

Anti-MARCO antibodies affect TI-2 B cell responses in C57BL/6 mice. WT mice were injected with 5 μg of the TI-2 antigen TNP-Ficoll alone (n = 7) or with 100 μg rat IgG anti-MARCO (n = 7) or a rat isotype control (n = 4). The data are shown for IgG (A), IgM (B), and IgG3 (C) anti-TNP response pre-immune (PI) at days 5 and 14. In A, data are presented as mean ± SEM, and in B and C, individual data and mean are shown. *, P < 0.05 (nonparametric Mann-Whitney U-test. TI-2 plus anti-MARCO compared with TI-2 or TI-2 plus control group).

Figure 5.

SLE patients have circulating anti-MARCO antibodies. (A) IgG anti-MARCO reactivity is found in sera from SLE patients. sMARCO or blocking buffer–coated ELISA plates were incubated with sera from SLE patients (n = 20) and healthy individuals (n = 19). Data are shown as anti-MARCO data minus anti-block buffer data to reduce the level of binding to the block buffer. (B) IgG anti-DNA activity in SLE patients and healthy individuals was measured by ELISA. The patient indicated by an arrow had an OD⁴⁰⁵ value of 1.44. (C) Correlation between anti-MARCO and anti-DNA levels in the patient group. The two measures do not correlate as deduced by linear regression analysis (R² = 0.038). (D) A model describing the proposed mechanism where anti-scavenger receptor antibodies influence SLE development by altering the marginal zone response to apoptotic cells. (1.) Anti-scavenger receptor autoantibodies alter the uptake of apoptotic cells by MZMs in the marginal zone leading to an increased local self-antigen load. (2.) Subsequently, autoreactive B cells have increased access to self-antigens in this area. (3.) In addition, engagement of the MARCO receptor by specific antibodies has the ability to send an activating signal to MZBs, suggesting a dual effect by the anti-scavenger receptor autoantibodies resulting in disease promotion.