¿ROLE OF ENDOTHELIAL DYSFUNCTION IN THE PATHOGENESIS OF UNPROVOKED VENOUS THROMBOEMBOLISM¿

Background and Aim Venous thromboembolism (VTE) is a multifactorial disease that includes deep vein thrombosis (DVT) and pulmonary embolism (PE). VTE patients can be classified on the basis of the risk factor(s) that trigger the thromboembolic event. Indeed, up to 50% of VTE events occur in absence of predisposing conditions and are therefore classified as unprovoked VTE (uVTE). Being uVTE pathogenesis still unknown, uVTE patients are characterised by a high risk of recurrence, and the therapeutic management of these patients to prevent VTE recurrence is still controversial. A better comprehension of uVTE pathogenesis and the identification of the molecular mechanisms underlying VTE triggering in these patients are needed to guide patient treatment through the definition of new clinicopathological entities and the definition of molecular targets for the development of novel therapeutic strategies. Endothelial dysfunction (ED) plays a crucial role in promoting thrombus formation, yet its presence in uVTE patients has been poorly investigated, so far. Therefore, in this study we investigated ED in uVTE, using patient-derived endothelial colony-forming cells (ECFCs) that represent an optimal non-invasive model to assess the functionality of the endothelial compartment. In particular, in this project we aimed to: • set-up and optimize ECFC-based functional assays aimed at assessing thrombogenic endothelial properties; • investigate the presence of constitutive ED in uVTE patients and the involved underlying molecular mechanisms; • investigate whether ED in uVTE patients is further promoted by pro-inflammatory stimuli; • evaluate whether ED contribute to VTE recurrence risk in uVTE patients; • assess whether ECFC features of ED are restricted to uVTE or shared with secondary VTE. Methods Optimization of functional assays. ECFC-based TGA and thrombogenesis assay were preliminarily set up on commercially-available human umbilical venous endothelial cells (HUVECs), then tested on ECFCs isolated and expanded from 8 healthy donors (HDs). ECFCs were used at passages 4-6, in basal conditions or after TNF-stimulated activation. TNFinduced EC activation was assessed as upregulation of the adhesion molecule VCAM-1 and the pro-coagulant molecule Tissue Factor (TF) by flow cytometry. EC function was assessed by analyzing: a) thrombin generation assessed by TGA on ECs - a modification of Hemker’s method; b) platelet deposition and fibrin formation under flow conditions, assessed in thrombogenesis assay. Assessment of ED role in uVTE pathogenesis. 78 VTE patients and 27 HDs were enrolled in the study. VTE patients were classified as: i) provoked VTE (pVTE) (n=28) when the VTE event was associated with major risk factors, ii) weakly provoked VTE (wpVTE) (n=28) when the VTE event was associated with minor risk factors; iii) unprovoked VTE (uVTE) (n=22) when the VTE event occurred in the absence of any provoking risk factors. A peripheral blood sample was collected at a single timepoint from all the enrolled subjects for ECFC isolation and plasma collection. In VTE patients, blood withdrawal was executed 3 months after the index VTE event, along with simultaneous assessment of validated predictors of VTE recurrence (namely, post thrombotic syndrome, D-dimer levels, and presence of residual venous obstruction). To investigate the presence of constitutive ED, ECFCs obtained from the different subgroups of VTE patients and HDs were characterized by assessing: i) their immunophenotype; ii) their ability to promote thrombosis in ECFC-based TGA and thrombogenesis assay; iii) their secretion of soluble ED markers (soluble adhesion molecules including the soluble forms of ICAM-1, VCAM-1, PECAM-1 and E-Selectin) in culture supernatants; to assess the in vivo relevance of these markers, they were also measured in the plasma of the same patients; iv) their transcriptomic profile assessed by RNA sequencing to investigate possible molecular pathways underlying ED in uVTE. To investigate whether ED in uVTE patients is further promoted by pro-inflammatory stimuli, the phenotype and functionality of TNF-treated ECFCs were assessed by using the same functional assays used to study constitutive ED. To investigate the role of ED in VTE recurrence, ECFC features of ED were analyzed in patients stratified according to their recurrence risk. Results Optimization of functional assays. TGA and the thrombogenesis assay were optimized to the use of ECs as a substrate. Both assays demonstrated the ability to discriminate between unstimulated and activated ECs, either HUVECs or ECFCs, whose activation was confirmed by the upregulation of VCAM-1 and TF expression. In particular, TNF-activated HUVECs and ECFCs were similarly characterized by significantly higher thrombin generation, platelet deposition and fibrin formation compared with their unstimulated counterparts. Based on these results, both ECFC-based functional assays were used for subsequent experiments addressing ED in uVTE, using patient-specific ECFCs. Assessment of ED role in uVTE pathogenesis. ECFCs were isolated from the three subgroups of VTE patients with a similar efficiency as HDs, but they were characterized by increased early senescence. The presence of constitutive ED was observed only in uVTE patients, as assessed as: a) a faster and higher peak of thrombin generation observed in TGA; b) increased platelet deposition observed in the thrombogenesis assay, in uVTE patients compared with HDs. Relevant to the increased platelet deposition, higher levels of soluble ICAM-1 and PECAM-1 were detected in the culture supernatants of ECFCs obtained from uVTE patients, the in vivo relevance of this finding being suggested by the observation of increased levels of soluble ICAM-1 also in the plasma of the same patients. Based on the preliminary analysis of the transcriptomic ECFC profiling, the upregulation of these adhesion molecules in uVTE patients may be supported by a downregulated gene expression of GSTM1 and PTGIR that were downregulated in uVTE ECFCs compared with ECFCs obtained from both HDs and patients with secondary VTE. A higher propensity of ECFCs obtained from uVTE patients to promote thrombus formation was also confirmed in inflammatory conditions, as only TNF-stimulated ECFCs from these patients showed a shorter lag time in TGA and higher platelet deposition in the thrombogenesis assay, compared with HDs. Notably, ECFC treatment with TNF increased the amount of thrombin generated in TGA only in HDs but not in uVTE, wpVTE or pVTE patients, likely related to direct oral anticoagulants (DOACs) taken by the vast majority of VTE patients at the time of blood collection. Finally, by stratifying patients according to their risk to develop VTE recurrences based on the presence of residual venous obstruction, we showed that several ECFC features of ED were broadly higher in uVTE and wpVTE patients with a high risk of recurrence compared with low-risk patients, thus suggesting that ED may also play a role in VTE recurrence in these patients. Conclusions In this study, we validated the use of ECFCs for assessing endothelial functions relevant to the study of human thromboembolic diseases. By applying properly optimized assays, we performed a functional characterization of patient-derived ECFCs, demonstrating that uVTE patients have endothelial prothrombotic alterations that may play a relevant role in the pathogenesis of the disease and in the risk of VTE recurrence. By extending the ECFC characterization on a larger cohort of patients, deepening the molecular characterization of ED, and by addressing the impact of anticoagulation on ED and VTE recurrence, this experimental approach has the potential to further dissect the molecular pathways underlying ED in uVTE patients, with the final aim to identify molecular targets for the development of novel therapeutic strategies aimed at preventing VTE recurrences in these patients.