Complement activation on heme-treated endothelial cells results in cell-bound C3 fragments and C5b9. HUVECs or GEnCs were stimulated with increasing concentrations of heme in M199 medium for 20 minutes at 37°C. If not stated otherwise, cells washed with PBS with Ca²⁺/Mg²⁺ (Gibco) were incubated for 30 minutes at 37°C in NHS diluted 1:4 in M199 medium. (A) Washed cells were detached with PBS, 1% bovine serum albumin, 10 mM EDTA, 0.1% sodium azide, without (HUVECs) or with 5 mg/mL lidocaine (GEnC) to be analyzed by flow cytometry after labeling with anti-C3c monoclonal antibody (mAb) (Quidel) and phycoerythrin-labeled secondary antibody (mean ± SD of mean fluorescent intensities relative to the mAb isotype control (RFI, n = 3). (B) Washed cells were fixed with 4% paraformaldehyde and 2% saccharose for 20 minutes at room temperature for analysis by cell-ELISA after labeling with biotin–anti-C3c (Quidel) in Tris-buffered saline–bovine serum albumin and peroxidase-extravidin (Sigma) revealed with 3,3′,5,5′-tetramethylbenzidine substrate (Pierce) (mean ± SD of triplicate wells from one of 3 similar experiments). (C) HUVECs were fixed with paraformaldehyde as in (B) to be analyzed by confocal microscopy after labeling with a mouse anti-CD31 mAb and a rabbit anti-C3 Ab revealed with Alexa555 (red)- and Alexa488 (green)-labeled secondary antibodies (Molecular Probes); scale bar, 50 µm. (D) Heme-exposure, under the experimental conditions used, did not induce cell death. HUVECs were exposed to 100 µM heme, quickly detached with trypsin, stained with Annexin V and propidium iodide, and analysed by flow cytometry. Apopto-necrotic HUVECs spontaneously detached overnight from a confluent monolayer served as a positive control for Annexin V and propidium iodide staining. Representative dot plots out of 3 independent experiments are shown. (E) Heme-induced C3 deposition on HUVECs were measured by flow cytometry with an anti-C3c mAb, under conditions favoring a fluid phase complement activation (heme added to the cells together with serum, “heme in serum”), direct effect on the cell surface (heme exposure as in panels [A], [B] and [C], “heme, wash, serum”); or both (cells pre-exposed to heme, then treated with serum with no washing step, “heme, serum”) (mean ± SD of 3 independent experiments). (F) The presence of uncleaved C3 on the surface of heme-exposed HUVECs were tested by staining with anti-C3a mAb the cells exposed with 100 µM heme and NHS as in (E). Blue, resting cells; purple, “heme, wash, serum”; green, “heme, serum”; black, unstained cells; red, isotype control IgG1. (G) Acidic wash of the uncovalently bound C3 forms from the heme-exposed cells. HUVECs were exposed to 100 µM heme, washed and incubated with NHS. Subsequently, 3 washes with PBS (purple histogram) or with a pH 2.7 PBS-25 mM Glycine buffer (green line) were performed. The cells were stained with anti-C3c, anti-C3a or anti-iC3b neoepitope antibody (Quidel) and analyzed by flow cytometry. Signal from unstained cells is presented in black and the isotype control IgG1 is in red; representative histograms of 3 independent experiments. (H) Heme-induced C3 deposition is dependent on the alternative complement pathway. HUVECs treated as in (A), with or without 100 μM heme, were incubated with 1:4 diluted NHS, with or without 10 mM EDTA or 10 mM EGTA and 5 mM MgCl₂, C2-depleted or FB-depleted serum, with and without 150 mg/ml purified FB. C3 deposits were measured, as in (A), by flow cytometry (mean ± SD RFI, n = 4). (I) HUVECs treated as in (A) were labeled with a mouse anti-C5b-9 mAb kindly provided by Prof Paul Morgan (Cardiff, United Kingdom) (n=3).