Vascular endothelial-cadherin is involved in endothelial cell detachment during thrombotic thrombocytopenic purpura
Raphael Cauchois 1)2)3), Marie Lagarde 1)2), Romain Muller 1)2)3, Julien Faccini 1)2), Aurélie Leroyer 1), Laurent Arnaud 4), Pascale Poullin 2)5), Françoise Dignat-George 1)4), Gilles Kaplanski 1)2)3), Edwige Tellier 2)3)
1 – Institut national de la santé et de la recherche médicale (INSERM), Institut national de recherche pour l’agriculture, l’alimentation et l’environnement (INRAE), Centre de Recherche en CardioVasculaire et Nutrition (C2VN), Aix Marseille University, Marseille, France
2 – French Reference Center for Thrombotic Microangiopathies, Paris, France
3 – Assistance Publique – Hôpitaux de Marseille (APHM), Centre Hospitalier Universitaire (CHU) Conception, Service de médecine interne et d’immunologie clinique, Marseille, France
4 – Assistance Publique – Hôpitaux de Marseille (APHM), Centre Hospitalier Universitaire (CHU) Conception, Département d’hématologie et de biologie vasculaire, Marseille, France
5 – Assistance Publique – Hôpitaux de Marseille (APHM), Centre Hospitalier Universitaire (CHU) Conception, Service d’Hémaphérèse, Marseille, France
Abstract
Background
Immune thrombotic thrombocytopenic purpura (i-TTP) is a life-threatening thrombotic microangiopathy linked to ADAMTS-13 deficiency. It has long been assumed that the activation of endothelial cells is the triggering factor for the thrombotic thrombocytopenic purpura crisis. Circulating endothelial cells (CECs) have been shown to be a biomarker of vascular damage and are associated with the clinical severity of i-TTP. However, the mechanisms leading to endothelial cell detachment remain unclear.
Objectives
We investigated junctional destabilization the mechanisms underlying cell detachment in thrombotic thrombocytopenic purpura.
Methods
We quantified CECs in i-TTP patients and investigated the effect of plasmas in vitro by measuring phosphorylation and internalization of vascular endothelial (VE)-Cadherin and in vivo in a vascular permeability model.
Results
In plasma from i-TTP patients, we show that CEC count is associated with severity and correlated to intracellular calcium influx (P < .01). In vitro, serum from i-TTP patients induced stronger detachment of human umbilical vein endothelial cells than serum from control patients (P < .001). Plasma from i-TTP patients induced a higher calcium-dependent phosphorylation (P < .05) and internalization (P < .05) of VE-cadherin compared with plasma from control patients. This effect could be reproduced by immunoglobulin (Ig)G fraction isolated from patient plasma and, in particular, by the F(ab)’2 fragments of the corresponding IgG. In addition, subcutaneous injection of i-TTP plasma into mice resulted in higher vascular permeability than plasma from control patients. An inhibitor of endothelial calcium influx, ITF1697, normalized this increase in permeability.
Conclusion
Our results suggest that plasma-induced endothelial activation also leads to an increase in vascular permeability. They contribute to the understanding of the mechanisms behind the presence of elevated CECs in patients’ blood by linking endothelial activation to endothelial injury.
Introduction
Thrombotic thrombocytopenic purpura (TTP) is a life-threatening thrombotic microangiopathy (TMA) associated with severe a disintegrin and metalloproteinase with thrombospondin motif 13 (ADAMTS-13) deficiency. This deficiency can be caused by mutations in the ADAMTS-13 gene but is most commonly caused by anti–ADAMTS-13 antibodies in immune-mediated TTP (i-TTP) [1]. The current pathophysiological paradigm suggests that ADAMTS-13 deficiency leads to an accumulation of ultralarge von Willebrand factor (UL-VWF) multimers, which thus form prothrombotic strings in microvessels. They induce massive platelet adhesion with the rapid formation of disseminated microthrombi, leading to the characteristic triad of thrombotic microangiopathies: organ ischemia, thrombocytopenia, and mechanical hemolytic anemia [2].
In addition to the central role of ADAMTS-13 deficiency in TTP pathogenesis, endothelial cells, which are the major storage sites of UL-VWF, have long been considered major players in TMA, especially in TTP [3]. Indeed, endothelial damage is observed in TTP, with the presence of histopathological features (swelling and necrosis of endothelial cells and subendothelial hyaline deposits), biological markers associated with endothelial activation, and commonly reported environmental triggers in patients and animal models [4,5]. Among endothelial markers, circulating endothelial cells (CECs), consisting of mature endothelial cells detached from the vessel wall [6], were identified in a TTP patient in 1993 [7]. This was confirmed in a prospective multicenter study of 22 patients: CECs were elevated in the acute phase and normalized during remission, and a high CEC count at diagnosis was associated with clinical severity [8]. High CEC counts have also been documented as a prognostic marker in other TMAs [9].
We have recently shown that plasma from patients with TTP induced in vitro a rapid calcium (Ca2+)-dependent endothelial activation (ie, type 1 endothelial cell activation [10]) with the release of UL-VWF [11]. Moreover, the immunoglobulin (Ig)G fraction present in the patients’ plasmas seems to be mainly responsible for this endothelial degranulation. Intracellular Ca2+ flux acts as a danger signal that confers a prothrombotic and proinflammatory phenotype to endothelial cells and also induces increased vascular permeability, involving the degradation of vascular endothelial (VE)-cadherin, the major transmembrane protein of adherens junctions, also known as CD144 [12].
Although CECs have long been observed in TTP, the mechanisms underlying cell detachment remain unclear. Intercellular adherens junctions are important components for the integrity of the endothelial monolayer. We, therefore, hypothesized that endothelial cell (EC) detachment is a consequence of strong endothelial activation leading to destabilization of intercellular adherens junctions.
Section snippets
Plasma and serum obtention
Plasma was obtained from patients enrolled in the prospective study Endo13 investigating endothelial biomarkers in i-TTP. Informed consent was obtained according to the Declaration of Helsinki. Fifty-three patients were diagnosed with i-TTP based on the following criteria: ischemic clinical events associated with mechanical hemolytic anemia (hemoglobin < 100 g/L), presence of schistocytes > 2% on blood smears, peripheral thrombocytopenia (<150 × 109/L), low plasmatic ADAMTS-13 activity (<5%, N:
The number of CECs is associated with disease severity and correlates with intracellular Ca2+ influx
We first quantified the CEC count in the plasma of i-TTP patients according to the severity of i-TTP. Severe patients had a higher CEC count than nonsevere patients (Figure 1A; 78 [24-255] vs 22 [5-40] cells/mL; P < .01). To gain an initial insight into the relationship between endothelial activation (essentially Ca2+-dependent) and cell detachment, we quantified intracellular Ca2+ fluxes on HUVECs stimulated by plasma from the same patients. We observed a significant positive statistical
Discussion
We demonstrated that i-TTP plasmas induce Ca2+-dependent phosphorylation of VE-cadherin, leading to its internalization. This destabilization of adherens junctions is involved in the detachment of endothelial cells observed in vitro and may explain the high level of CECs occurring in TTP crisis. Furthermore, we have shown that IgG from patients mimics these findings and is likely involved in plasma-induced endothelial activation. We have confirmed that plasmas from i-TTP patients can induce
Acknowledgments
We thank Italfarmaco for providing ITF1697 and Karen Vanhoorelbeke for providing the IgG fraction.
Author contributions
R.C., J.F., R.M., M.L., A.L., L.A., and E.T. conducted the experiments and interpreted the data; P.P., F.D.G., and G.K. contributed to sample collection and analysis of clinical data; R.C. and G.K. took care of patients; E.T. and G.K. supervised the project; R.C. and E.T. drafted the manuscript.
Declaration of competing interests
P.P. is a member of the scientific advisory boards of Ablynx-Sanofi. The other authors declare no