Chimeric Proteins Containing MAP-1 and Functional Domains of C4b-Binding Protein Reveal Strong Complement Inhibitory Capacities

Abstract

The complement system is a tightly regulated network of proteins involved in defense against pathogens, inflammatory processes, and coordination of the innate and adaptive immune responses. Dysregulation of the complement cascade is associated with many inflammatory disorders. Thus, inhibition of the complement system has emerged as an option for treatment of a range of different inflammatory diseases. MAP-1 is a pattern recognition molecule (PRM)-associated inhibitor of the lectin pathway of the complement system, whereas C4b-binding protein (C4BP) regulates both the classical and lectin pathways. In this study we generated chimeric proteins consisting of MAP-1 and the first five domains of human C4BP (C4BP^1-5) in order to develop a targeted inhibitor acting at different levels of the complement cascade. Two different constructs were designed and expressed in CHO cells where MAP-1 was fused with C4BP^1-5 in either the C- or N-terminus. The functionality of the chimeric proteins was assessed using different in vitro complement activation assays. Both chimeric proteins displayed the characteristic Ca²⁺-dependent dimerization and binding to PRMs of native MAP-1, as well as the co-factor activity of native C4BP. In ELISA-based complement activation assays they could effectively inhibit the lectin and classical pathways. Notably, MAP-1:C4BP^1-5 was five times more effective than rMAP-1 and rC4BP^1-5 applied at the same time, emphasizing the advantage of a single inhibitor containing both functional domains. The MAP-1/C4BP chimeras exert unique complement inhibitory properties and represent a novel therapeutic approach targeting both upstream and central complement activation.

Keywords: C4BP; MAP-1; chimeric protein; classical pathway; complement activation; complement inhibition; lectin pathway.

Figure 1

Recombinant constructs and purified proteins. (A) Diagram of the domain distribution of MAP-1, C4BP α-chain and the recombinant proteins. (B) Instant blue stain of purified MAP-1:C4BP¹⁻⁵, C4BP¹⁻⁵:MAP-1, and C4BP¹⁻⁵via affinity chromatography. (C) The identity of the protein bands was confirmed by Western blotting using anti-MAP-1 mAb and rMAP-1 as positive control.

Figure 2

Gel filtration chromatography of chimeric proteins. (A) Gel filtration profile of MAP-1:C4BP¹⁻⁵ (M:C) and C4BP¹⁻⁵:MAP-1 (C:M) under physiological calcium conditions or 10 mM EDTA. The identity of the gel filtration profiles was confirmed by analyzing the elution fractions of M:C (B) or C:M (C) by sandwich ELISA using anti-C4BP as capture antibody and anti-MASP-1/-3/MAP-1 as detection. Relative abundance of the chimeric proteins in the elution fractions following gel filtration under calcium conditions or EDTA is expressed as OD. mAU, milli absorption units.

Figure 3

MAP-1:C4BP¹⁻⁵ (M:C), C4BP¹⁻⁵:MAP-1 (C:M) and MAP-1 binding to MBL (A) and CL-11 (B). rMBL and rCL-11 were immobilized onto mannan-coated plates. Recombinant proteins were applied in a two-fold dilution in the presence of calcium or EDTA. Binding was determined using an anti-MAP-1 mAb. Connecting lines display nonlinear fitting using the equation specific binding with hill slope. Results are representative of three independent experiments and error bars represent minimum and maximum values of triplicate measurements.

Figure 4

Chimeric and rMAP-1 complement inhibition after pre-incubation on rMBL/mannan. Serial dilutions of MAP-1:C4BP¹⁻⁵ (M:C), C4BP¹⁻⁵:MAP-1 (C:M), and rMAP-1 were allowed to form complexes with mannan-bound rMBL prior to addition of MBL-defect serum as a source of complement. Detection of C4 (A), C3 (B), and TCC (C) deposition was quantified using anti-C4, anti-C3, and anti-TCC mouse mAbs. Connecting lines are four parameters nonlinear fitting using the equation inhibitor concentration vs. slope. Data are reported as [(OD_inhibitor-OD_background)/(OD_noinhibitor-OD_background)]100. Error bars represent the SEM of three independent experiments, and the dashed line the 50% inhibition level. *P < 0.05; ***P < 0.001; ****P < 0.0001.

Figure 5

Chimeric and rMAP-1 complement inhibition after co-incubation with NHS. Serial dilutions of MAP-1:C4BP¹⁻⁵ (M:C), C4BP¹⁻⁵:MAP-1 (C:M), and rMAP-1 were co-incubated with 2% NHS in non-adsorbent titration plates for 30 min prior to addition to mannan-bound rMBL. Detection of deposition of C4 (A), C3 (B), and TCC (C) was determined as described before. Connecting lines are four parameters nonlinear fitting using the equation inhibitor concentration vs. slope. Dashed line Data are reported as [(OD_inhibitor – OD_background)/(OD_noinhibitor – OD_background)]100. Error bars represent the SEM of three independent experiments, and the dashed line the 50% inhibition level. *P < 0.05; ***P < 0.001.

Figure 6

Cofactor activity in factor I-mediated C4b and C3b cleavage. Purified C3b or C4b was incubated with fI and different cofactors for 2 h at 37°C. The reactions were stopped with LDS buffer and subjected to western blotting under reducing conditions using C4d and C3d antibodies. (A) Incubation of C4b with fI in the presence of C4BP, fH or C4BP-containing chimeric proteins resulted in the generation of a band corresponding to the inactive degradation products iC4b and C4d. (B) Incubation of C3b with fI in the presence of C4BP, fH or C4BP-containing chimeric proteins results in the generation of the 68 kDa degradation fragment. M:C, MAP-1:C4BP¹⁻⁵; C:M, C4BP¹⁻⁵:MAP-1. The blots are representative of three independent experiments.

Figure 7

Inhibition of complement deposition on tubular proximal kidney epithelial cells. (A) HK-2 cells were incubated with increasing concentrations of rMBL at 4°C and bound MBL was determined using an anti-MBL mAb. (B) Binding of MBL results in the deposition of C4. HK-2 cells were incubated with rMBL followed by 10% MBL defect serum for 1 h at 4°C. Bound C4 was determined using an anti-C4c pAb. (C) rMBL-mediated C4 deposition after MAP-1:C4BP¹⁻⁵ (M:C), C4BP¹⁻⁵:MAP-1 (C:M), rMAP-1 or full length C4BP co-incubation with 10% MBL defect serum. Inhibition is reported as the ratio of the MFI of the inhibitor to the MFI of the bank (MFI_inh:MFI_blank). Significance was tested for each concentration of the inhibitors compared to the blank. Data are reported as the mean ± SEM of three independent experiments.

Figure 8

Wielisa total complement screen. Serial dilutions of MAP-1:C4BP¹⁻⁵ (M:C), C4BP¹⁻⁵:MAP-1 (C:M), rMAP-1, rC4BP¹⁻⁵, and rMAP-1 and rC4BP¹⁻⁵ together (M+C) were incubated with NHS and subsequently the protein/serum mix was applied to pre-coated Wielisa plates. Complement activation was quantified using an anti-TCC mAb. (A) Lectin pathway activation on mannan-coated plates. (B) Classical pathway activation on IgM-coated plates. (C) Alternative activation on LPS-coated plates. Connecting lines are nonlinear fitting using the equation inhibitor concentration vs. slope. Values represent OD readings normalized to the OD of the wells without inhibitor times 100. Significance was tested for best-fit IC50 values of data sets with an adjusted R-squared > 0.8. Error bars represent the SEM of three independent experiments, and the dashed line the 50% inhibition level. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.