HMGB1 cleavage by complement C1s and its potent anti-inflammatory product

Abstract

Complement C1s association with the pathogenesis of several diseases cannot be simply explained only by considering its main role in activating the classical complement pathway. This suggests that non-canonical functions are to be deciphered for this protease. Here the focus is on C1s cleavage of HMGB1 as an auxiliary target. HMGB1 is a chromatin non-histone nuclear protein, which exerts in fact multiple functions depending on its location and its post-translational modifications. In the extracellular compartment, HMGB1 can amplify immune and inflammatory responses to danger associated molecular patterns, in health and disease. Among possible regulatory mechanisms, proteolytic processing could be highly relevant for HMGB1 functional modulation. The unique properties of HMGB1 cleavage by C1s are analyzed in details. For example, C1s cannot cleave the HMGB1 A-box fragment, which has been described in the literature as an inhibitor/antagonist of HMGB1. By mass spectrometry, C1s cleavage was experimentally identified to occur after lysine on position 65, 128 and 172 in HMGB1. Compared to previously identified C1s cleavage sites, the ones identified here are uncommon, and their analysis suggests that local conformational changes are required before cleavage at certain positions. This is in line with the observation that HMGB1 cleavage by C1s is far slower when compared to human neutrophil elastase. Recombinant expression of cleavage fragments and site-directed mutagenesis were used to confirm these results and to explore how the output of C1s cleavage on HMGB1 is finely modulated by the molecular environment. Furthermore, knowing the antagonist effect of the isolated recombinant A-box subdomain in several pathophysiological contexts, we wondered if C1s cleavage could generate natural antagonist fragments. As a functional readout, IL-6 secretion following moderate LPS activation of RAW264.7 macrophage was investigated, using LPS alone or in complex with HMGB1 or some recombinant fragments. This study revealed that a N-terminal fragment released by C1s cleavage bears stronger antagonist properties as compared to the A-box, which was not expected. We discuss how this fragment could provide a potent brake for the inflammatory process, opening the way to dampen inflammation.

Revisiting the interaction between complement lectin pathway protease MASP-2 and SARS-CoV-2 nucleoprotein

“…Notably, the flexible linker sequence contains several arginine residues corresponding to potential cleavage sites for trypsin-like proteases. MASP-2 has trypsin-like specificity and is known to cleave itself (autoactivation), complement C2 and C4 proteins, C1-Inhibitor, prothrombin ( 22 , 37 , 38 ) and more recently the HMGB1 alarmin ( 20 ). It is homologous to the C1s protease of the classical complement pathway and has been shown to have higher catalytic efficiency than C1s ( 39 ).…”

“…Each plasmid was used for stable transfection of Freestyle 293-F cells (ThermoFisher Scientific) and the recombinant proteins were purified from cell culture supernatants by anti-FLAG and C1q affinity chromatography, as described by Lorvellec et al. ( 20 ). MASP-2 molar concentration was estimated using a M r value of 75,100 and an absorbance (0.1%, 1 cm) at 280 nm (A 0.1%, 1 cm ) of 1.57.…”

“…Recombinant human MASP-2 with a C-terminal Flag tag was produced by stable transfection of 293-F cells using a pcDNA3.1_MASP-2-Flag expression vector as described by Lorvellec et al (20). The expression plasmid coding for recombinant MASP-2 stabilized in a proenzyme form by mutation of the active site Ser 618 into Ala (mature protein numbering) was generated with the QuickChange II XL site-directed mutagenesis kit (Agilent Technologies), using the mutagenic primers described by Zundel et al (21) and the pcDNA3.1_MASP-2-Flag plasmid as template.…”

“…The expression plasmid coding for recombinant MASP-2 stabilized in a proenzyme form by mutation of the active site Ser 618 into Ala (mature protein numbering) was generated with the QuickChange II XL site-directed mutagenesis kit (Agilent Technologies), using the mutagenic primers described by Zundel et al (21) and the pcDNA3.1_MASP-2-Flag plasmid as template. Each plasmid was used for stable transfection of Freestyle 293-F cells (ThermoFisher Scientific) and the recombinant proteins were purified from cell culture supernatants by anti-FLAG and C1q affinity chromatography, as described by Lorvellec et al (20). MASP-2 molar concentration was estimated using a M r value of 75,100 and an absorbance (0.1%, 1 cm) at 280 nm (A 0.1%, 1 cm ) of 1.57.…”

“…This plasmid served as a template to generate the expression plasmid coding for the S618A mutant of MASP-2 catalytic domain. Both catalytic domains were purified from cell culture supernatant on an anti-FLAG M2 agarose column (Sigma-Aldrich) as described by Lorvellec et al (20). Most of the recombinant proteins were recovered in the flow through, dialyzed against 20 mM Na 2 HPO 4 , 5 mM EDTA, pH 8.6 and loaded onto a Q-Sepharose Fast Flow column (Cytiva) equilibrated in the same buffer.…”