Platelet Endoplasmic Reticulum Stress Promotes Thrombosis in an in Vitro Laser-Induced Endothelial Injury Model

Aim: Platelet endoplasmic reticulum (ER) stress has recently been associated with platelet activation (Jain et al. Circ Res 2022). ER stress is triggered by calcium depletion from the ER which rapidly triggers downstream activation pathways. On the other hand, platelet ER stress activates the ER stress sensor, inositol-requiring enzyme 1 α (IRE1), which increases platelet secretion and reactivity (Lay et al. Blood Adv 2023). However, the role of IRE1 activation in platelet calcium mobilization is underexplored in the context of vascular thrombosis. Here, we utilized an endothelialized microfluidic platform that we have developed to study the relation of ER stress and platelet calcium mobilization in real time, under conditions closely mimicking the development of a vascular thrombus. This was achieved by perfusing whole blood across an endothelialized microfluidic device and inducing a site-specific endothelial injury using a high-powered laser and comparing with a standardized in vivo laser-induced venous thrombosis model. We further investigated the role of IRE1 using a chemical inhibitor of IRE1.

Method: Initially, we standardized the kinetics of platelet, fibrin and neutrophil accumulation in a laser-induced mesenteric vein thrombosis in 3-month-old C57/BL6 mice. This was used as a comparator for the laser-induced thrombus formation on our endothelialized device at shear rates of 100 s^-1 and 800 s^-1. The laser injury was performed using a straight channel microfluidic device fabricated from polydimethylsiloxane and endothelialised with human umbilical vein endothelial cells (HUVECs). Citrated (0.32% w/v) whole blood was collected from healthy volunteers and stained with anti-CD41-FITC (0.5 μg/mL), anti-fibrin-Alexa Fluor 594 (0.5 μg/mL) antibodies, and Annexin V-Alexa Fluor 647 (1:200) to visualize platelets, fibrin, and phosphatidylserine exposure respectively. For calcium mobilization, whole blood was stained with an anti-CD42a-PE (0.5 μg/mL) antibody and Cal520 (2 μM). To examine the effects of anticoagulants in our endothelialized device, whole blood was treated with heparin (0.5 U/mL) or thrombin inhibitor argatroban (0.5 μg/mL). We then compared with the effects of IRE1 activator IXA4 (10- 300 μM). IXA4 was incubated with whole blood for 2 hrs prior to perfusion following whole blood recalcification with CaCl2 (10 mM). A localized injury was then induced at the junction of endothelial cells by pulsing a 300 mW, 355nm laser at 10% power, 5 times over 2 sec, in a 1 um square area parfocal to the endothelial monolayer. Subsequent platelet adhesion, fibrin formation, and phosphatidylserine exposure was imaged over 15 min and quantified based on the surface area of fluorescence. Statistical analysis was by paired student t-test after confirming normality.

Results: Laser injury of the endothelial monolayer resulted in the Annexin V binding to endothelial cells directly adjacent to the injury site with no loss of endothelial cell junctions. Platelet adhesion and fibrin formation was restricted to Annexin V positive endothelial cells, forming a tear drop-shaped thrombus within 6 mins of injury. This closely followed the kinetics formation in vivo. Compared to vehicle, both argatroban and heparin inhibited platelet adhesion (argatroban: 2856 vs 226.0 um²; heparin: 2856 vs 56.70 um², P<0.05) and fibrin formation (argatroban: 8421 vs 1176 um^2; heparin: 8421 vs 9.651 um², P<0.05), without affecting Annexin V binding. IXA4 enhanced the accumulation of calcium in platelets after laser injury (0.07080 vs 0.1470 AU, P<0.05). Furthermore, it enhanced platelet adhesion (1709 vs 1957 um²; P<0.05) and fibrin formation (3045 vs 7716 um²; P< 0.05).

Conclusion: The kinetics of our in vitro, laser-induced vascular thrombosis model closely mirrors laser-induced thrombus formation in vivo, as measured by intravital microscopy. Thrombus formation in the endothelialized device was Factor Xa and/or thrombin dependent. We propose our model offers a robust assay for the study of calcium kinetics in platelets under conditions of vascular thrombosis. This model offers a feasible alternative to mouse models of thrombosis using laser injury for the study of calcium and novel therapeutic agents. Activation of IRE1 increased calcium mobilization in the platelets and thrombus formation. The above introduce the novel concept of ER stress-dependent calcium mobilization in vascular thrombosis.