Evidence for a functional role of IgE anticitrullinated protein antibodies in rheumatoid arthritis

Abstract

Rheumatoid arthritis (RA) is a systemic autoimmune disease involving inflammation of the joints. Among the autoantibodies described in RA, anticitrullinated protein antibodies (ACPAs) are highly specific and predictive for RA. In addition, ACPAs have been implicated in the pathogenesis of RA. However, a direct functional response of immune cells from ACPA(+) RA patients toward citrullinated proteins has not been demonstrated. In this study, we show that exposure to citrullinated antigens leads to activation of basophils from ACPA(+) RA patients within 20 minutes. This was not observed after exposure of basophils to noncitrullinated control antigens or after stimulation of basophils from ACPA(-) RA patients and healthy controls. Basophil activation was correlated with the binding of citrullinated proteins to basophils. Furthermore, serum from ACPA(+) RA patients in contrast to that from ACPA(-) RA patients could specifically sensitize human FcepsilonRI expressing rat basophil cells (RBL), enabling activation by citrullinated proteins. Mast cell degranulation products such as histamine levels were enhanced in synovial fluid of ACPA(+) RA patients as compared with ACPA(-) RA and osteoarthritis patients. In addition, histamine levels in synovial fluid from ACPA(+) RA patients correlated with IgE levels, suggesting degranulation of mast cells by cross-linking IgE. Immunohistochemistry on synovial biopsies demonstrated an increased number of degranulated CD117(+) mast cells in ACPA(+) RA patients; IgE and FcepsilonRI expression in synovial mast cells from ACPA(+) RA patients was increased. In conclusion, our results show an immunological response of immune cells from ACPA(+) RA patients in a citrulline-specific manner. Moreover, these data indicate a role for IgE-ACPAs and FcepsilonRI-positive cells in the pathogenesis of RA.

Fig. 1.

IgE-ACPAs. (A) Anti-IgE antibodies against noncitrullinated fibrinogen (NC-FB) (○) and citrullinated fibrinogen (C-FB) (●) in IgG-depleted serum samples from control (CTR; n = 5), ACPA⁻ RA (n = 4), and ACPA⁺ RA (n = 8) patients. (B) Anti-IgE-ACPAs in IgG-depleted serum samples from CTR (n = 5), ACPA⁻ RA (n = 4), and ACPA⁺ RA (n = 6) patients. P = 0.003 in ACPA⁺ RA vs. CTR. P = 0.007 in ACPA⁻ RA vs. ACPA⁺ RA.

Fig. 2.

Basophil activation. Whole heparinized blood was stimulated with HBSS buffer, 1 μg/mL anti-immunoglobulin E (anti-IgE), 0.2 U/mL PAD, 10 μg/mL noncitrullinated fibrinogen (NC-FB), and 10 μg/mL citrullinated fibrinogen (C-FB) in control (CTR; n = 26), ACPA⁻ RA (n = 15), and ACPA⁺ RA (n = 29) patients. Bars represent median values. P < 0.0001 for HBSS vs. aIgE in CTR, ACPA⁻ RA, and ACPA⁺ RA. P < 0.0001 for C-FB vs. NC-FB in ACPA⁺ RA. P < 0.0001 for C-FB in ACPA⁺ RA vs. ACPA⁻ RA. P < 0.0001 for C-FB in ACPA⁺ RA vs. CTR.

Fig. 3.

Basophil activation with citrullinated (C)-MBP, heat-aggregated IgG (HAIgG), and IgG. Whole heparinized blood was stimulated with HBSS buffer, 1 μg/mL anti-IgE, 10 μg/mL noncitrullinated (NC)-MBP, 10 C-MBP, 10 μg/mL NC-fibrinogen (FB), 10 μg/mL C-FB, 10 μg/mL monomeric IgG, and 10 or 100 μg/mL HAIgG in healthy control (CTR; n = 3), ACPA⁻ RA (n = 2), and ACPA⁺ RA (n = 6) patients. P = 0.02 in ACPA⁺ RA vs. CTR and ACPA⁻ RA.

Fig. 4.

Elution of IgG. Basophil activation before and after elution of Ig with acid glycine buffer in ACPA⁺ RA patients (n = 5). Basophils were stimulated with HBSS buffer, 1 μg/mL anti-IgE, 10 μg/mL noncitrullinated fibrinogen (NC-FB), 10 μg/mL citrullinated fibrinogen (C-FB), and 1 μg/mL fMLP before and after elution of Ig with acid glycine buffer in ACPA⁺ RA patients (n = 5). P = 0.04 for C-FB PRE elution vs. C-FB POST elution.

Fig. 5.

Passive sensitization of FcεRI-transfected RBLs. After sensitization with sera from control (CTR; n = 10), ACPA⁻ RA (n = 39), and ACPA⁺ RA (n = 50) patients for 18 hours, RBLs were stimulated with Thyrod’s buffer, 1 μg/mL anti-IgE, 10 μg/mL noncitrullinated fibrinogen (NC-FB), or 10 μg/mL citrullinated fibrinogen (C-FB). Results are expressed as the percentage of total β-hexoaminidase release (treated with Triton-X) after correction for spontaneous release in Thyrod’s buffer. P < 0.0001 for C-FB vs. NC-FB in ACPA⁺ RA. P < 0.0001 for C-FB in ACPA⁺ RA vs ACPA⁻ RA. P = 0.001 for C-FB in ACPA⁺ RA vs CTR.

Fig. 6.

Surface IgE expression on CD117-allophycocyanin (APC⁺) synovial mast cells. Synovial mast cells of ACPA⁺ RA (n = 10), ACPA⁻ RA (n = 5), and OA (n = 18) patients were stained with Alexafluor-conjugated anti-IgE. Results are expressed as difference of mean fluorescence intensity (MFI) between isotype control MFI and anti-IgE MFI. P = 0.008 in ACPA⁺ RA vs. ACPA⁻ RA. P < 0.0001 in ACPA⁺ RA vs. OA.

Fig. 7.

Histamine levels in RA synovial fluid. Histamine levels (ng/mL) in synovial fluid of OA (n = 14), ACPA⁻ RA (n = 36), and ACPA⁺ RA (n = 39) patients. Bars represent median values. P = 0.004 in ACPA⁺ RA vs. OA. P = 0.008 in ACPA RA⁺ vs. ACPA RA⁻.

Fig. 8.

Microscopic features of the synovial tissue in ACPA⁻ RA and ACPA⁺ RA. (A) CD117 staining of mast cells. (Magnification: ×200.) (B) Staining of surface CD117 on mast cells. (Magnification: ×400.) (C) CAE staining of mast cell. (Magnification: ×400.) (D) CAE staining of granules in mast cells. (Magnification: ×1,000.) (E) Double CAE/ CD117 staining (whole granules not visible after antigen retrieval). (Magnification: ×400.) (F) Nondegranulated mast cell positive for CAE and CD117 (Left) and degranulated mast cell positive for CD117 and negative for CAE (Right). (Magnification: ×1,000.)