Polyphosphate is a novel cofactor for regulation of complement by a serpin, C1 inhibitor

Abstract

The complement system plays a key role in innate immunity, inflammation, and coagulation. The system is delicately balanced by negative regulatory mechanisms that modulate the host response to pathogen invasion and injury. The serpin, C1-esterase inhibitor (C1-INH), is the only known plasma inhibitor of C1s, the initiating serine protease of the classical pathway of complement. Like other serpin-protease partners, C1-INH interaction with C1s is accelerated by polyanions such as heparin. Polyphosphate (polyP) is a naturally occurring polyanion with effects on coagulation and complement. We recently found that polyP binds to C1-INH, prompting us to consider whether polyP acts as a cofactor for C1-INH interactions with its target proteases. We show that polyP dampens C1s-mediated activation of the classical pathway in a polymer length- and concentration-dependent manner by accelerating C1-INH neutralization of C1s cleavage of C4 and C2. PolyP significantly increases the rate of interaction between C1s and C1-INH, to an extent comparable to heparin, with an exosite on the serine protease domain of the enzyme playing a major role in this interaction. In a serum-based cell culture system, polyP significantly suppressed C4d deposition on endothelial cells, generated via the classical and lectin pathways. Moreover, polyP and C1-INH colocalize in activated platelets, suggesting that their interactions are physiologically relevant. In summary, like heparin, polyP is a naturally occurring cofactor for the C1s:C1-INH interaction and thus an important regulator of complement activation. The findings may provide novel insights into mechanisms underlying inflammatory diseases and the development of new therapies.

Figure 1

PolyP enhances the capacity of C1-INH to dampen C1s-mediated cleavage of C4 and C2 in a concentration- and size-dependent manner. SDS-PAGE analysis under reducing conditions was used to evaluate the effect of polyP concentration and size, on C4 and C2 cleavage. Concentrations of polyP reflect the concentration of the monomer. (A) C4 alone is shown in lane 1. C1s alone is not detectable at the concentration used. C1-INH and C1s were reacted with C4 in the presence of varying concentrations of polyP₁₃₀ or 500 μM of monoP (lane 5) as shown and described in supplemental Materials and methods. C4 cleavage products were detected by Coomassie staining after SDS-PAGE. (B) C1-INH and C1s were reacted with C2 in the presence of varying concentrations of polyP₁₃₀ or 500 μM of monoP (lane 5). (C) C1-INH and C1s were reacted with C4 in the presence of 500 μM monoP (P1), P3, polyP₁₄ (P14), polyP₆₀ (P60), or polyP₁₃₀ (P130) for 30 minutes, after which C4 cleavage products were visualized by SDS-PAGE. Gels are representative of experiments performed a minimum of 5 times.

Figure 2

Heparin and polyP increase the rate of complex formation between C1-INH and C1s. For panels A-D, proteins were separated by 10% SDS-PAGE and gels were stained with Coomassie blue R-250. Concentrations of polyP reflect the concentration of the monomer. (A) 1 μM C1-INH was reacted at 4°C with 1 μM recombinant C1s (C1s SP domain), in the presence of increasing concentrations of polyP₁₃₀ (P130) for the indicated times. Purified C1-INH is shown in the lane between the 1 and 5 minute panels. (B) 1 μM C1-INH was reacted at 4°C for 0 to 30 minutes with 1 μM recombinant C1s (C1s SP domain) in the absence (C, control) or presence of either 50 μg/mL heparin or 160 μM polyP₁₃₀ (left). Densitometry performed on 3 gels (right). The percent C1-INH complex is based on C1-INH alone being assigned as 100%. (C) 1 μM C1-INH was reacted at 4°C for 0 to 120 minutes with 1 μM plasma C1r in the absence (C) or presence of either 50 μg/mL heparin or 166 μM polyP₁₃₀ (left). Densitometry performed on 3 gels (right). The percent C1-INH complex is based on C1-INH at t = 0 being assigned as 100%. (D) 10 nM C1-INH was reacted at 4°C for varying periods of time as shown, with 10 nM C1s (CCP12SP), 1 μM C4, and either buffer alone (C), 50 μg/mL heparin, or 166 μM polyP₁₃₀ (left). Densitometry performed on 3 gels (right). The percent C4 α chain is based on C4α at 5 hours being assigned as 100%. CCP, complement control protein-like domain; H, heparin; P, polyP₁₃₀.

Figure 3

Effect of polyP₁₃₀ and heparin on the observed rates of association between C1s and C1-INH. The effect of polyP₁₃₀ (A) or heparin (C) on the observed rate of association between recombinant WT CCP12SP C1s and C1-INH was determined by adding C1s (2 nM) to C1-INH (20 nM), Z-Lys-SBzl (0.2 mM), and DTDP (0.6 mM) in the presence of increasing concentrations of polyP₁₃₀ (0-2000 μM) (A) or heparin (0-100 μg mL⁻¹) (C). The pseudo-first-order rate constant (k_obs) is plotted as a function of polyP₁₃₀ or heparin concentration. The right-hand axis shows the ratio of the k_obs at a particular polyP₁₃₀ or heparin concentration to that in the absence of a cofactor. The effect of polyP₁₃₀ (B) or heparin (D) on the observed rate of association between recombinant exosite (A1) mutant CCP12SP C1s and C1-INH was determined by adding 1.5 nM C1s A1 mutant to 30 nM C1-INH, 0.2 nM Z-Lys-SBzl, and 0.9 mM DTDP in the presence of increasing concentrations of polyP₁₃₀ (0-2000 μM) (B) or heparin (0-500 μg mL⁻¹) (D). The pseudo-first-order rate constant (k_obs) is plotted as a function of polyP₁₃₀ or heparin concentration. The right-hand axis shows the ratio of the k_obs at a particular polyP₁₃₀ or heparin concentration to that in the absence of a cofactor. Results are representative of 3 independent experiments.

Figure 4

Interactions of C1-INH and C1s with heparin and polyP. (A) Analytical affinity chromatography indicates binding of heparin to C1s and C1-INH. C1-INH (gray line), plasma-derived C1s (black line), a 1:1 mixture of both (pre-incubated for 60 minutes at 37°C) (dotted black line), or the C1s A1 mutant (orange solid line), were separately applied to a HiTrap-Heparin column and eluted with a NaCl gradient as described in “Materials and methods.” Conductivity measurements are shown on the right-hand axis. Eluted fractions were analyzed by SDS-PAGE to confirm the identity of the proteins. (B) SPR was used to quantify binding of C1s (CCP12SP) to biotinylated polyP₂₅₀ attached to streptavidin immobilized to the chip. A representative experiment (n = 3) shows curves for 0 to 625 nM C1s flowed over the immobilized polyP₂₅₀. Inset: The response units obtained at equilibrium for each concentration of C1s were plotted and fitted using a one-site binding model on GraphPad Prism (regression coefficient = 0.99). Similar results were obtained in 2 additional independent experiments.

Figure 5

PolyP inhibits C4d deposition on endothelial cells in a concentration-dependent manner. HMEC-1 were exposed to (A) NHS, (B) MBL-deficient serum, or (C) C1q-deficient serum in the presence of varying concentrations of polyP₁₃₀ (P130) or monoP (P1). For each serum, a corresponding HIS control was tested (flow cytometry graphics on right panels). After 1 hour, the reactions were stopped, and C4d deposition was quantified by flow cytometry. Data shown are representative of 3 independent experiments for each serum condition. Histograms on the left show normalized MFI levels of C4d deposited on endothelial cells. Percent MFI values are mean ± SEM of duplicates. On the right, flow cytometry graphic reveals inhibition of C4d deposition, reflected by reduced fluorescence. HIS, heat-inactivated serum; MFI, mean fluorescence intensity; SEM, standard error of the mean.

Figure 6

PolyP and C1-INH colocalize in activated human platelets. Human platelets were plated onto glass coverslips in the resting state (top panels) or after 2 minutes of activation with 100 nM of the phorbol ester, phorbol 12-myristate 13-acetate, and 1 μM of calcium ionophore A23187. After fixation and permeabilization, the platelets were stained as described in “Materials and methods” to detect C1-INH (left panels, green) and polyP (middle panels, red) by confocal microscopy. The right panels reveal the merged images. In resting platelets, C1-INH is detected in α granules and on the cell surface, and polyP is found in dense granules. After activation, C1-INH and polyP coalesce in the center of the platelets where they colocalize (yellow, right bottom panel). Size bars, 10 μm. Results are representative of 3 independent experiments.

Figure 7

Mechanisms by which polyP regulates complement activation. Upon activation, polyP and C1-INH can colocalize, after which they are secreted. PolyP triggers a conformational change in FXII, resulting in generation of FXIIa, which can activate prekallikrein and/or FXI to FXIa. Kallikrein or plasmin (not shown) can further cleave FXIIa to generate βFXIIa, which may activate C1r and thus promote complement activation. C1-INH dampens that pathway by inhibiting FXIIa, βFXIIa, and kallikrein. C1s cleaves C4 and C2 to generate the C4b2a C3 convertase, which ultimately leads to formation of the C5b,6 complex, and assembly of the C5b-9 MAC. PolyP or heparin potentiate the inhibitory function of C1-INH via direct interactions with C1-INH and the target protease, C1s. PolyP also destabilizes C5b,6, thereby dampening the formation of the MAC. The over-riding effect of polyP in a serum-based endothelial cell culture system is to suppress complement activation. K, kallikrein; MAC, membrane attack complex; PK, prekallikrein.