Figure 1.
Modular structure of the proteases of the C1r/C1s/MASP family and their role in complement activation.
(A) The generic modular structure of the proteases of this family is shown. MASP-1 and -3, the two proteases coded by the MASP1 gene, exhibit a common core (yellow) and unique serine protease (SP) domains. The target of the activation cleavage and the functional domain subdivision are illustrated. The N-terminal interaction domain mediates the binding to the cognate recognition protein and calcium-dependent protease dimerization. In C1r, C1s and MASP-2, the catalytic activity of the C-terminal SP domain is modulated by the preceding CCP1 and CCP2 modules involved in substrate recognition or dimerization [22], [31], [63]. ap: activation peptide. CCP: complement control protein module; CUB: complement C1r/C1s, Uegf, Bmp1 module (B) The activation cascades triggering the lectin and classical complement (C) pathways, and their inhibition by C1-inhibitor, are illustrated. The proteases are associated in large complexes with collagen defence recognition proteins. A possible role for MASP-3 in the activation of the complement alternative pathway needs to be confirmed and its other possible implications outside complement are to be deciphered.
Figure 2.
SDS-PAGE analysis of the recombinant MASP-3 SP domain.
The purified SP domain of MASP-3 was analyzed by SDS-PAGE under non-reducing (NR) and reducing (R) conditions followed by Coomassie blue staining. The molecular masses of reduced and non-reduced standard proteins (expressed in kDa) are shown on the right and left sides of the gel, respectively.
Table 1.
Amidolytic activity of MASP-3 SP, thrombin and trypsin.
Table 2.
Kinetic parameters for selected substrates of MASP-3 SP and thrombin.
Figure 3.
Comparative SPR analysis of the interaction of MASP-3 and other trypsin-like proteases with ecotin.
Ecotin (1,000 RU) was immobilized on a CM5 sensor chip as described under Materials and Methods. Sixty microliters of varying concentrations of the MASP-3 SP domain (A), the MASP-2 CCP1/2-ap-SP fragment (B), trypsin (C), and factor XIa (D) were injected over immobilized ecotin in 145 mM NaCl, 50 mM triethanolamine-HCl, pH 7.4 containing 0.005% surfactant P20 at a flow rate of 20 µl/min. The specific binding signals shown were obtained by subtracting the background signal over a reference surface with no protein immobilized. Fits are shown as dotted lines and were obtained by global fitting of the data using a 1∶1 Langmuir binding model.
Table 3.
Kinetic and dissociation constants for the interaction of selected proteases with immobilized ecotin.
Figure 4.
Gel filtration analysis of ecotin-MASP-3 SP complexes.
(A) Equimolar amounts of the MASP-3 SP domain and ecotin (225 pmol) were preincubated for 20 min at room temperature before injection on a TSK G-3000 SW column equilibrated in 145 mM NaCl, 50 mM triethanolamine-HCl, pH 7.4 and run at 1 ml/min. (B) SDS-PAGE analysis under reducing conditions and Coomassie blue staining of the protein content of peak 1. The molecular masses of reduced standard proteins (expressed in kDa) are shown on the left side of the gel.
Figure 5.
Structure of the ecotin/MASP-3 SP tetramer.
(A) Overall and (B) schematic view of the tetramer highlighting its primary (1) and secondary (2) interfaces. The same color codes are used for MASP-3 and ecotin in the remainder of the figure. The two ecotin and MASP-3 monomers are labeled C, D and E, F, respectively. (C) Zoom showing the MASP-3 active site, clamped in-between two ecotin molecules. The positions of the mutations related to the 3MC syndrome are underlined. The active site triad is in black. The similar tetrameric structure of the Y69F/D70P ecotin mutant complexed to the D102N rat anionic trypsin (PDB code 1EZU) is superimposed, with trypsin colored in orange and the two ecotin chains colored in pink and light blue. Note the large insertion loop B (480–493) in MASP-3 (green) compared to trypsin. (D) Zoom on the ecotin/MASP-3 primary interface. For clarity purposes, only the catalytic triad and the residues or main-chain stretches contributing to the closest contacts are shown. (E) Zoom on the ecotin/MASP-3 secondary interface. Dotted lines in panels D and E indicate conserved (red) or specific (dark) H-bonds. Here the term ‘conserved’ means that these bonds are observed in the other ecotin-protease complexes.
Table 4.
Crystallographic data and refinement statistics.
Figure 6.
Structural comparison of MASP-3 with other trypsin-like SP domains.
(A) and (B) overall view of the superimposition of the MASP-3 SP domain with different complement and coagulation proteases. The proteases are identified by labels with the corresponding color. The substrate-like ecotin loop 80–86 is displayed in yellow in (B). The two similar inhibitor loops are displayed in red (ecotin), orange (SGMI-1) and yellow (SGMI-2) in (A). The loop definition is according to [56]: 1: 633–638 (184–188); 2: 669–679 (217–225); 3: 612–624 (169–176); A: 448–463 (33–42); B: 480–498 (59–64); C: 529–533 (97–101); D: 576–594 (144–151); E: 505–514 (72–82); the corresponding PDB codes are 3TVJ (MASP-2/SGMI-2), 4DJZ (MASP-1/SGMI-1), 1MD8 (C1r), 1ELV (C1s), 3KCG (fIXa), 1POS (fXa). (C) Sequence alignment of human MASPs, C1r, C1s, coagulation fX and fIX including loops 1 and 2. The red color highlights a sequence stretch highly specific of MASP-3. The sequences are listed in a decreasing order of sequence identity of their SP domain compared to MASP-3 (91.1 to 29.7%). The sequences corresponding to MASP-3 in other species are the following: chicken (Q6Q1Q8), mouse (Q8CIR9), xenopus (Q8AXQ8), and amphioxus (Q868H6). (D) Zoom on the substrate-binding site, superimposed onto trypsin (dark blue). (E) Zoom on the SP loops 1 to 3 of the three MASPs structures in a slightly different orientation. (F) Zoom on the fairly similar peptide inhibitor conformations observed for ecotin (red) and SGMI-2 (yellow).
Figure 7.
Ecotin-induced displacement and proposed conformational equilibrium of MASP-3.
(A) Displacements of the homologous Phe549 and Phe529 in MASP-1 and MASP-2 (in cyan and blue for the free enzyme, respectively) upon SGMI inhibitor binding (left in orange). (B) The position of the homologous Tyr531 in MASP-3 (green) clusters with the ‘displaced’ positions and not with the ‘free’ positions of MASP-1 and MASP-2. The homologous side-chain in free C1r (dark blue) perfectly fits between those of free MASP-1 and MASP-2. It is also the case for C1s (not shown). (C) Scheme illustrating the proposed conformational equilibrium between an inactive E* conformation and the active E conformation observed upon interaction with ecotin. The segment 215–217, which collapses in the substrate-binding site in the E* or Z* states is shown in grey. (D) Position of the segment 667–671 (in grey) homologous to 215–217, in close proximity to Tyr531 and the substrate binding site (see ecotin in yellow). This segment includes the specific MASP-3 insertion in loop-2 illustrated in Fig. 6.