The complement alternative pathway (AP) is a key component of the innate immune system. Not only does it maintain a low level of spontaneous activity in the fluid phase but it also amplifies complement activity initiated by the classical and lectin pathways. C3b deposited on cell surfaces by any complement pathway acts as a locus for AP activation. Factor B binds C3b and is subsequently cleaved by Factor D to form the C3 convertase, C3bBb. C3bBb cleaves more C3 to C3b that can bind covalently to cells and recruit Factor B to form additional AP C3 convertases. C3b bound, or in close proximity, to C3bBb [C3bBb(C3b)_n_] switches the specificity of the enzymatic complex from C3 to C5, leading to terminal complement activation that generates the effector molecules C5a and the membrane attack complex (MAC). Due to its spontaneity and its ability to amplify all complement activity, the human host has developed many regulatory mechanisms to prevent excessive AP activation.
Factor H (FH) is the key regulator of the AP in the fluid phase and also functions to limit AP activity selectively on self-cell surfaces. It accelerates the decay of C3/C5 convertases, and acts as a cofactor for Factor I-mediated cleavage of C3b to inactive C3b (iC3b) to prevent further convertase formation. Twenty homologous complement control protein domains comprise FH. The four most N-terminal domains (domains 1–4) contain all of the protein’s regulatory function, whereas the remaining domains help anchor the protein to cell surfaces for efficient regulation by domains 1–4. At least two distinct C3b and polyanion binding sites have been described between domains 7–20. While these binding sites have cell- and tissue-specific roles in the human host, the C-terminal domains, 19–20, contain both C3b and polyanion binding sites and are key for FH interactions with erythrocytes, platelets, endothelial cells and neutrophils, among other ligands.
Mutations in domains 19 and 20 that impair their ability to bind to C3b and polyanions are associated with atypical hemolytic uremic syndrome (aHUS), a disease characterized by renal failure, hemolytic anemia, and thrombocytopenia. Thrombocytopenia is a result of excessive platelet activation leading to increased thrombi formation. While platelet–platelet and platelet–endothelium interactions are key for thrombi formation, activated platelets can also bind granulocytes to form platelet/granulocyte aggregates (PGAs). Increased PGA levels enhance thromboinflammation in animal models of vascular disease and are found in patients suffering from a variety of inflammatory vascular conditions, including acute coronary syndromes, inflammatory bowel and lung disease, and diabetes, however, it is not known whether increased PGA formation contributes to aHUS pathophysiology.
Complement activity enhances PGA formation in human whole blood stimulated with thrombin receptor-activating peptide (TRAP). Our laboratory has previously shown that the AP is key to the effects of complement on PGA formation and that a competitive inhibitor of FH C-terminal domains 19 and 20 (rH19–20), significantly increases AP activity and enhances PGA formation in TRAP-stimulated human whole blood. By limiting critical FH cell-surface interactions and maintaining fluid-phase AP regulation, rH19–20 simulates the pathophysiological mechanisms involved in aHUS, suggesting that increased PGA formation could enhance vascular pathology in patients suffering from the disease. Here, we sought to better characterize the interaction of FH at the platelet/granulocyte interface and to determine the effects that mutations in domains 19 and 20 have on control of PGA formation in human whole blood.
Modified HEPES/Tyrode’s (HT) buffer (137 mM NaCl, 2.8 mM KCl, 1 mM MgCl 2 6H 2 O, 12 mM NaHCO 3, 0.4 mM Na 2 HPO 4, 10 mM HEPES, 0.35% BSA, 5.5 mM glucose; pH 7.4); PBS (10 mM NaH 2 PO 4, 140 mM NaCl, pH 7.4); Tyrode’s buffer [136.9 mM NaCl, 2.7 mM KCl, 983.8 µM MgCl 2 6H 2 O, 3.2 mM Na 2 HPO 4, 3.5 mM HEPES, 0.35% BSA, 5.5 mM dextrose, 2 mM CaCl 2 (pH 7.4)]; Tyrode’s + EDTA (Tyrode’s buffer + 10 mM EDTA); Tyrode’s + heparin + PGE 1 (Tyrode’s buffer + 1 µM PGE 1 and 2 IU/ml heparin); and MgEGTA [0.1 M MgCl 2 and 0.1 M EGTA (pH7.3)].
Murine monoclonal antibodies (all IgG1κ): antihuman CD42b-APC (BioLegend), antihuman CD45-PE (BioLegend), antihuman/mouse C3/C3b/iC3b-FITC and unlabeled (Cedarlane), antihuman CD11b-PerCP/Cy5.5 and -PE (BioLegend), antihuman CD62P-PE/Cy5 (BioLegend), and goat antimouse-AF488. Isotype controls: APC- and PerCP/Cy5.5-labeled (BioLegend), PE-labeled (BioLegend), and FITC-labeled (Cedarlane).
Properdin-depleted serum and C8-depleted serum were purchased from Complement Technologies.
RH19–20 was produced as previously described. Additional coding sequence for residues rH1–3, rH2–4, rH3–5, rH6–8, rH7–9, rH8–10, rH10–12, rH12–14, rH12–15, rH13–15, rH14–16, rH15–17, rH16–18, and rH17–19 were PCR amplified from a full-length FH cDNA template, accession number NM_000186. The SapI-digested PCR products were cloned into pPICZ-α vector (Invitrogen) containing a His-6 tag and modified to include SapI sites in its multiple cloning site, as previously described. Following transformation into Pichia pastoris strain KM71H, protein expression was induced with methanol following manufacturer’s instructions, deglycosylated by incubation with endoglycosidase H (New England Biolabs, Ipswich, MA, USA), and the proteins were purified from the media by affinity chromatography using Ni-NTA Agarose (Qiagen).
Table 1. Primers used to amplify coding sequences of several fragments of Factor H.
RH19–20 mutants were produced as previously described. Two additional mutants, W1183L and R1215Q, were produced as described in the cited reference. Concentrations for all proteins were determined (280 nm), in triplicate, using theoretical E 1 cm 1% values (e.g., 1.49 for W1183L and W1183R and 1.85 for the rest of the mutants) using the Expert Protein Analysis System (EXPASY; https://www.frankenthalerfoundation.org
Cp20 (Ac-I[CV-1MeW-QDW-Sar-AHRC]mI-NH 2), a potent compstatin analog, was used to inhibit convertase-mediated C3 activation, and was produced by solid phase synthesis. PMX53 was produced as described. SALO was a generous donation from Dr. Jesus Valenzuela (National Institutes of Health). Eculizumab was purchased from Creative Biolabs, and OmCI was a generous donation from Dr. Susan Lea (University of Oxford). Human IgG4 isotype control (BioLegend) was used as a control for Eculizumab.
Properdin was isolated from human plasma, as previously described. For some experiments, properdin dimers, trimers, and tetramers (P 2, P 3, and P 4) were separated from non-physiological aggregates (P n), stored at 4°C, and used within 2 weeks of separation, as previously described.
Human whole blood was obtained via venipuncture from volunteer donors. The Institutional Review Board from the University of Toledo College of Medicine and Life Sciences approved the protocols, and written informed consent was obtained from all donors, in accordance with the Declaration of Helsinki. Blood was drawn into vacutainer tubes (Becton Dickinson) containing 50 µg/ml final concentration of the thrombin inhibitor lepirudin (Celgene). 20 µl of whole blood was gently mixed with modified HT buffer + 10 µM TRAP (Bachem) and complement modulators, for a final volume of 80 µl. All groups were set up in duplicate. Samples were incubated at 37°C for 15 min and the reaction was stopped by addition of 800 µl RBC lysation/fixation solution (BioLegend). Samples were fixed for 10 min at room temperature (RT) before being washed, stained with detection antibodies for 15 min at RT, then diluted with 800 µl RBC fixation/lysation solution. Finally, samples were spun at 200 g for 15 min at 4°C, then 500 µl supernatant was removed and the samples run on a FACSCalibur (Becton Dickinson) flow cytometer. 10,000 events were acquired from a gate encompassing granulocyte and monocyte populations. Using FlowJo software version 7.6 (Tree Star), granulocytes were gated based on CD45 and side scatter, and the percent of granulocytes positive for CD42b fluorescence, as well as the C3- and CD11b-associated geometric mean fluorescent intensities (GMFIs) on gated granulocytes, were determined.
Whole blood assays were set up as described, but in triplicate. Following the 37°C incubation, one replicate was immediately placed on ice then spun at 300 g for 10 min at 4°C. The other two replicates were processed via flow cytometry, as described above. Supernatants were spun at 13,000 g for 5 min at 4°C, then immediately frozen at −80°C until use. Supernatants were diluted 1/10 and C5a levels determined via standard ELISA kit (Abcam) following manufacturer’s instructions.
Platelets were isolated from human whole blood as previously described. Briefly, blood from volunteer donors was drawn into acid citrate dextrose vacutainers via venipuncture then spun at 200 g for 15 min at RT without braking. Platelets were washed twice with acid citrate wash buffer at 440 g for 10 min at RT, then resuspended in Tyrode’s buffer. Platelets (1 × 10 8/ml) were activated with thrombin (1 U/ml) for 30 min at 37°C and the reaction stopped by the addition of Tyrode’s + heparin + PGE 1, as previously described.
Neutrophils were isolated from EDTA-anticoagulated human whole blood via Polymorphprep gradient centrifugation, using the manufacturer’s protocol. Briefly, whole blood was layered over Polymorphprep in a 1:1 ratio, then spun at 500 g for 35 min (without brake) at RT. The neutrophil layer was obtained from the gradient and washed with an equal volume of half-strength HBSS, without Ca+2 or Mg+2 (Life Technologies), at 400 g for 10 min at RT. Contaminating erythrocytes were lysed by a 7-min incubation in RBC lysis buffer (BioLegend) at
