Essential role of mannose-binding lectin-associated serine protease-1 in activation of the complement factor D

Abstract

The complement system is an essential component of innate immunity, participating in the pathogenesis of inflammatory diseases and in host defense. In the lectin complement pathway, mannose-binding lectin (MBL) and ficolins act as recognition molecules, and MBL-associated serine protease (MASP) is a key enzyme; MASP-2 is responsible for the lectin pathway activation. The function of other serine proteases (MASP-1 and MASP-3) is still obscure. In this study, we generated a MASP-1- and MASP-3-deficient mouse model (Masp1/3-/-) and found that no activation of the alternative pathway was observed in Masp1/3-/- serum. Mass spectrometric analysis revealed that circulating complement factor D (Df) in Masp1/3-/- mice is a zymogen (pro-Df) with the activation peptide QPRGR at its N terminus. These results suggested that Masp1/3-/- mice failed to convert pro-Df to its active form, whereas it was generally accepted that the activation peptide of pro-Df is removed during its secretion and factor D constitutively exists in an active form in the circulation. Furthermore, recombinant MASP-1 converted pro-Df to the active form in vitro, although the activation mechanism of pro-Df by MASP-1 is still unclear. Thus, it is clear that MASP-1 is an essential protease of both the lectin and alternative complement pathways.

Figure 1.

Masp1/3^−/− mice show no ability to activate the alternative pathway. (A) Rabbit erythrocytes (2.5 × 10⁶) were incubated for 1 h with mouse serum in GVB containing Mg²⁺-EGTA. Hemolysis was measured in a microplate reader at 405 nm. Two individual C4-deficient (C4^−/−), MASP-1/-3, and C4 triple-knockout (M1/3^−/−C4^−/−) mice and MASP-2/sMAP and C4 triple-knockout (M2^−/−C4^−/−) mice were analyzed. (B) Sera from wild-type C57BL6 (WT), Masp1/3^−/− (M1/3^−/−), and Masp2/sMap^−/− (M2^−/−) mice diluted with GVB containing Mg²⁺-EGTA were incubated in zymosan-coated microwells for 1 h. Bound C3 was detected by fluorescein-conjugated anti–mouse C3 antibody. Values are presented as relative fluorescence units (RFUs). Data are presented as means ± SD of three individual mice for each strain. Data from one out of two (A) or three (B) independent experiments with similar results are shown.

Figure 2.

Df restores the deficiency of alternative pathway activation in Masp1/3^−/− mice. (A) Immunoblot analysis of Bf incubated with the sera of wild-type C57BL/6 (WT), MBL-A, and -C double-deficient (Mbl null) and Masp1/3^−/− mice at 37°C for 30 min. Each serum type was incubated with (+) or without (−) 1.25 µg C3(H₂O). Serum from Masp1/3^−/− mice was also incubated with (+) human Df (0.2 µg). Note: an additional band (∼26 kD) was also detected with Ba, but it was not identified. (B) A hemolytic assay using rabbit erythrocytes was performed. Percent hemolysis induced by sera (25 µl) from wild-type (WT), Masp1/3^−/− (M1/3^−/−), and Masp1/3^−/− mice supplemented with 0.2 µg human Df (huDf) is indicated. (C) A C3 deposition assay on immobilized zymosan was performed. 5 µl of sera from wild-type (WT), Masp2/sMap^−/− (M2^−/−), and Masp1/3^−/− (M1/3^−/−) mice supplemented with 0.2 µg human Df (huDf) were analyzed. Data are presented as the means ± SD of three independent experiments.

Figure 3.

Circulating Df in Masp1/3^−/− mice is a zymogen. (A) Endogeneous Df was immunoprecipitated and immunoblotted with a polyclonal anti-Df antibody. Precipitated Df was incubated with N-glycosidase F (Endo-F) to remove N-glycosylations. Sera from wild-type (WT) and Masp1/3^−/− mice were analyzed. (B) Endogenous Df was immunoprecipitated from the sera of wild-type (C57BL6; top) and Masp1/3^−/− mice (bottom). V8 protease-digested fragments were analyzed by ESI-TOF-MS. The range from 590 to 630 (m/z) is presented. (C) Western blotting of Df in culture supernatants from undifferentiated 3T3-L1 cells (lane 1) and differentiated 3T3-L1 cells (lane 2) were analyzed. As a control, precipitated Df from the sera of wild-type (lane 3) and Masp1/3^−/− mice (lane 4) was used in immunoblotting experiments. Data from one out of three (A and C) or two (B) independent experiments with similar results are shown.

Figure 4.

MASP-1 activates pro-Df. (A) Recombinant pro-Df (80 ng) was incubated with 120 ng (lane 2) and 240 ng (lane 4) of recombinant MASP-1 (rMASP-1K) at 37°C for 1 h. As a control, pro-Df was incubated without rMASP-1K (lanes 1 and 3). After removing N-glycosylations by N-glycosidase F treatment, samples were separated by SDS-PAGE and immunoblotted with anti-Df antibody. Data are representative of two independent experiments. (B) A C3 deposition assay on immobilized zymosan was performed. 5 µl of the sera of wild-type (WT) and Masp1/3^−/− (M1/3^−/−) mice supplemented with 0.2 µg of pro-Df alone (+pro-Df), 0.03 µg of recombinant MASP-1K alone (+rM1K) or pro-Df (0.2 µg) that was preincubated with rMASP-1K (0.03 µg; +pro-Df+rM1K) are analyzed. Data are presented as the means ± SD of three independent experiments. (C) A reconstitution experiment using purified components, i.e., CVF, Bf, and recombinant pro-Df was performed. Samples were subjected to electrophoresis under nonreducing conditions and stained with Coomassie. Before reconstitution with CVF and Bf, pro-Df was preincubated with rMASP-1K (lane 4 and 5). Data are representative of two independent experiments.

Figure 5.

Specific detection of pro-Df and kinetic analysis for cleavage of its activation peptide. (A) Pro-Df was detected by a sandwich ELISA. Endogenous Df in mouse sera was captured in anti-Df antibody-coating wells, followed by detection of pro-Df with antibody specific for the activation peptide of mouse pro-Df. (B) Sera (0.5 µl) from WT (lane 1) and Masp1/3^−/− (lane 2) mice were analyzed by immunoblotting using anti-Df (top) and anti–pro-Df (bottom) under reducing conditions. (C) Masp1/3^−/− serum was incubated with same volume of WT serum at 37°C for 1 h (lane 2), 2 h (lane 3), and 3 h (lane 4). As controls, Masp1/3^−/− serum (lane 1) or WT serum (lane 5) was incubated for 3 h. Incubated samples corresponding to 0.5 µl of serum were immunoprecipitated with anti–pro-Df, followed by treatment of N-glycosidase-F. Captured pro-Df was detected by immunoblot with anti-Df. (D) Kinetic analysis to cleave a fluorogenic synthetic substrate (H-QPRGR-MCA) by thrombin and rMASP-1K. Each enzyme (1 nM) was used to cleave 20 µM of substrate. (E) After MASP-1 was pulled down from the purified rMASP-1K using anti–MASP-1 antibody (open circles) or the irrelevant antibody as control (filled circles), and the samples were used for cleavage of the synthetic substrate. All data represent one of similar results obtained from at least two experiments.