The monosialoganglioside GM1a protects against complement attack

The complement system is a part of the innate immune system in the fluid phase and efficiently eliminates pathogens. However, its activation requires tight regulation on the host cell surface in order not to compromise cellular viability. Previously, we showed that loss of placental cell surface sialylation in mice in vivo leads to a maternal complement attack at the fetal-maternal interface, ultimately resulting in loss of pregnancy. To gain insight into the regulatory function of sialylation in complement activation, we here generated trophoblast stem cells (TSC) devoid of sialylation, which also revealed complement sensitivity and cell death in vitro. Glycolipid-analysis by multiplexed capillary gel electrophoresis coupled to laser-induced fluorescence detection (xCGE-LIF) allowed us to identify the monosialoganglioside GM1a as a key element of cell surface complement regulation. Exogenously administered GM1a integrated into the plasma membrane of trophoblasts, substantially increased binding of complement factor H (FH) and was sufficient to protect the cells from complement attack and cell death. GM1a treatment also rescued human endothelial cells and erythrocytes from complement attack in a concentration dependent manner. Furthermore, GM1a significantly reduced complement mediated hemolysis of erythrocytes from a patient with Paroxysmal nocturnal hemoglobinuria (PNH). This study demonstrates the complement regulatory potential of exogenously administered gangliosides and paves the way for sialoglycotherapeutics as a novel substance class for membrane-targeted complement regulators.

changed every other day.For differentiation experiments, standard differentiation medium (TSC medium without FGF4, Heparin and MEF conditioned medium) was used.In differentiation experiments with murine complement active serum (Innovative research), standard differentiation medium was used without FCS, but with 20 % murine complement active serum.Bovine Ig depleted FCS was generated by incubation of FCS with Protein G beads (Amersham, UK) over night at 4°C with gentle agitation.

Hemolysis assay.
Human whole blood was obtained by venipuncture and erythrocytes were isolated by centrifugation for 5 min at 1.000 x g.RBCs were washed two times in PBS and stored in Alsever's solution at 4°C with light agitation.Desialylation was performed under continuous agitation with 0.13 mg/ml purified Arthrobacter ureafaciens neuraminidase (AU 54) (1) for 30 min at 37°C in PBS.
Simultaneously, sensitising of erythrocytes was achieved by incubation with anti-CD59 antibody (2 µg/ml).For the incorporation of GM1a or GM3, erythrocytes were incubated for 1 h at 37°C in PBS in agitation with the respective concentrations.Erythrocytes were washed and resuspended in DGHB-Mg-EGTA (HEPES 4.2 mM, NaCl 59 mM, MgCl2 7 mM, EGTA 10 mM, Glucose 2.08 % (w/v), Gelatine 0.08 % (w/v)) (2) buffer to a final concentration of 5*10 7 erythrocytes/ml.In a 96-well microtiter plate 50 µl erythrocytes were mixed with human serum diluted in DGHB Mg-EGTA to a final volume of 100 µl and incubated for 30 min at 37°C with continuous shaking.Subsequently, 150 µl NaCl 0.9 % were added and the plate was centrifuged at 1000 xg for 5 min.Supernatant was transferred to a flat-bottomed 96well plate and absorption at 414 nm was detected in a plate reader.All samples were measured in duplicates.Negative controls contain no serum.100 % erythrocyte lysis was carried out by adding 200µl RBC lysis buffer (NH4Cl 155 mM, NaHCO3 12 mM, 0.1 mM EDTA) to 50 µl RBC.For hemolysis assays of PNH-RBC incubation with anti-CD59 was omitted and 100µM GM1a and complement active human serum serotype AB were used.Experiments with human samples were approved by the local ethics committee (9371_BO_S_2020).Hemolytic assay with sheep red blood cells (SRBC) (Innovative research, Novi, MI, USA) were performed as described for human RBC.However, inucubation with anti-CD59 was omitted.
Calcein release assay.In brief, EA.hy926 cells (10 4 per well) were treated for 30 minutes at 37°C with Neu, incubated for 30 minutes with Calcein AM (25 µM) and GM1a or GM3 (Avanti Polar Lipids, United States) of varying concentrations was added for 1h at 37°C.To induce a complement attack the cells were then incubated with 15 % human serum together with anti-CD59 (10 µg/ml).After 30 minutes at 37°C fluorescence of the supernatant at 494 nm was measured.

Supplemental Tables
Figure S1.Genetic deletion of Cmas does not alter the stem cell status of TSC.(A) Scheme of the murine Cmas gene and sequencing data of one representative Cmas -/-clone with indels at the target site in exon 4. (B) The table depicts the size of indels in exon 4 of Cmas -/-clones determined by sequencing.(C) Brightfield images of control and one representative Cmas -/-TSC cultivated under standard conditions.Scale bars = 250 µm.(D) Detection of EOMES and CDX2 by indirect immunofluorescence analysis of control and one representative Cmas -/-TSC clone.Nuclei were stained with DAPI and are shown in blue.Scale bars = 50 µm.

Figure S2 .
Figure S2.Cmas -/-TSC lack sialylated glycosphingolipids.(A) Quantification of neutral glycans derived from glycosphingolipids of control and one representative Cmas -/-TSC clone using xCGE-LIF.Glycan notations for GM2, GD1a and GM1a refer to the respective glycosphingolipid derived glycan.For inter-sample comparisons signal intensities were normalized to Man6 (nRFU).MTU: migration time unit.Symbol nomenclature according to 3 .n=7 Mean values are depicted with standard deviation.(B-C) xCGE-LIF analysis of glycans derived from glycosphingolipids.Glycan notations for GM2, GD1a and GM1a refer to the respective glycosphingolipid derived glycan.Overlay of xCGE-LIF electropherograms of APTS-labeled GSL derived glycans of control (B) and Cmas -/-TSC (C) with or without prior treatment with Neu.For inter-sample comparisons signal intensities were normalized to Man6 (nRFU).MTU: migration time unit.Symbols nomenclature according to (3).

Figure S4 .
Figure S4.FCS induced complement activation on Cmas -/-TSC is driven by bovine immunoglobulins.(A) Brightfield images of control and Cmas -/-TSC cultured in differentiation medium supplemented either with complement active FCS or complement active murine serum.Scale bars = 250 µm.(B) Detection of bovine Ig (shown in green) by indirect immunofluorescence analysis of differentiated Cmas -/-TSC.Nuclei were stained with DAPI and are shown in blue.Scale bars = 50 µm.(C)Brightfield images of control and Cmas -/-TSC cultured for two days in differentiation medium supplemented either with untreated (active complement), heat-inactivated (inactive complement) or protein G treated (immunoglobulin depleted, active complement) FCS.Scale bars = 250 µm.

Figure S5 .
Figure S5.Gating strategy for flow cytometry analysis.Red blood cells were selected from a forward scatter-area vs sideward scatter-area dot plot.Single cells were subsequently selected in a sideward scatter-area vs sideward scatter width dot plot.This population was analyzed for FH/CtxB reactivity.