Beyond complement: Patient and in vitro evidence for alternative mechanisms of Monocyte–Platelet aggregation in PNH

Abstract Background: Paroxysmal nocturnal haemoglobinuria (PNH) is a clonal haematopoietic stem cell disorder characterised by the absence of glycosylphosphatidylinositol (GPI)-anchored surface proteins and a markedly increased risk of thrombosis. The mechanisms underlying monocyte–platelet interactions in PNH remain poorly defined. We investigated the formation of monocyte–platelet aggregates (MPAs) in PNH patients and established an in vitro model using a GPI-deficient monocytic cell line to dissect the contributing cellular mechanisms. Additionally, we compared different monocytic cell lines to primary human monocytes to assess their suitability for modelling platelet aggregation behaviour. Methods: Peripheral blood samples from 13 PNH patients and 13 age- and sex-matched healthy controls were analysed by flow cytometry. Monocytes were identified using CD64, PNH clones distinguished by loss of CD14, and MPAs defined by co-expression of CD42a. Furthermore, a CRISPR/Cas9-engineered GPI-deficient monocytic cell line (PIGA knockout) was generated with 97% knockout efficiency (validated by FLAER staining). Aggregation assays were performed following co-incubation with platelet-rich plasma (PRP), with or without pre-treatment using 1 µg/ml PMA, Zymosan A, PNH patient serum, or control serum. A cell–cell interaction model was established by co-culturing wild-type (WT) and GPI-deficient cell lines prior to stimulation. Results: In patient samples, flow cytometric analysis revealed that WT- monocytes (CD14⁺) from PNH patients exhibited a significantly higher median rate of platelet aggregation (16%) compared to clonal PNH monocytes (CD14⁻, 6%) and monocytes from healthy controls (4.3%). These findings suggest that thromboinflammatory activation in PNH is not limited to GPI-deficient cells but may extend to non-clonal, bystander monocytes. In vitro, stimulation of primary human monocytes and a monocytic cell line with either PNH patient serum or healthy control serum did not result in increased platelet aggregation. These findings exclude direct serum- or complement-mediated activation under the tested conditions. A co-culture assay combining WT and the generated PIGA knockout cell line suggests thrombogenic activation by the surrounding WT cells via direct or paracrine mechanisms. These findings are currently under investigation. Conclusion: Our data suggest that the prothrombotic phenotype in PNH may not be solely intrinsic to GPI-deficient cells but could involve intercellular mechanisms that enhance platelet aggregation by bystander wild-type monocytes. This insight may provide a new perspective on systemic thromboinflammation in PNH and inform therapeutic strategies beyond complement inhibition.