Several studies have demonstrated that platelets can interact with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and consequently undergo programmed cell death, extracellular vesicle release, and increased platelet reactivity in coronavirus disease 2019 (COVID-19) patients in response to low doses of α-thrombin.[1 ]
[2 ]
[3 ] Such hyper-sensitivity could be partially due to increased mitogen-activated protein kinase (MAPK) signaling pathway activation, protein kinase C delta (PKCδ) phosphorylation, and thromboxane synthesis.[1 ]
[2 ] These posttranslational modifications are often triggered following activation of one or more receptors on the platelet surface.
GPIbα (CD42b) in the GPIb-IX-V receptor complex is the major binding site for α-thrombin associated with platelets, and through this function may support procoagulant activities and contribute to platelet activation and aggregation.[4 ] Importantly, blocking α-thrombin binding to the N-terminal region of GPIbα (His1-Glu282) can be achieved by a blocking antibody (SZ2) that selectively inhibits the α-thrombin binding site on GPIbα.[4 ] According to recent findings, GPIbα is the receptor through which SARS-CoV-2 spike protein binds to platelets as well as activates their increased expression of ligands.[5 ] However, the role of GPIbα in the α-thrombin-induced platelet hyper-responsiveness observed during SARS-CoV-2 infection is not well understood.
At the low enzyme concentration, we identified α-thrombin/GPIbα interaction as a novel mechanism triggering the hyper-reactivity observed in COVID-19 patients. Our findings showed that during SARS-CoV-2 infection, pretreatment of platelets with human anti-GPIbα antibody (SZ2) prevents platelet hyper-aggregation and degranulation. Interestingly, we found that platelets derived from patients with Bernard–Soulier syndrome (BSS) and who are infected with SARS-CoV-2 are not hyper-reactive. We thus have identified low α-thrombin–GPIbα interaction as a novel prothrombotic pathway which triggers the formation of procoagulant platelets in COVID-19 patients.
COVID-19 patients (n = 10) and COVID-19 patients with BSS (n = 3) who were admitted to the Cheikh Zaid Hospital (Rabat, Morocco) were included in this prospective, observational study conducted from December 7, 2020 to January 12, 2022. Patients with BSS have a homozygous mutation on GP9 which prevents GPIbα expression on platelets. All patients with COVID-19 were studied (on average) at 5.8 ± 1.2 hours after receiving a nasopharyngeal swab that showed positivity for SARS-CoV-2. Sex- and age-matched (age: 47.5, interquartile range: 38–52.4, 50% female) healthy blood donors (n = 10) were used as controls. None of the patients or healthy volunteers were treated with antithrombotic drugs (either premorbidly or for the treatment of their COVID-19 infection), and that could affect platelet functions or coagulation. All human blood studies were approved by the Local Ethics Committee of Cheikh Zaid Hospital, Rabat, Morocco; Project: CEFCZ/PR/2020-PR04. Informed consent was obtained from each subject and all experiments were conducted according to the principles set out in the Declaration of Helsinki.
Washed platelets were prepared as previously described.[1 ] Markers of platelet activation, α-granule release (CD62P or P-selectin expression) ([Fig. 1Ai, Aii ]) and dense granule secretion, as assessed by ATP release ([Fig. 1Aiii ]) and loss of mepacrine fluorescence ([Fig. 1Aiv ]), were significantly enhanced in the presence of sub-threshold concentrations of α-thrombin in platelets from COVID-19 patients. We sought to investigate the contribution of GPIbα to the regulation of platelet hyper-activation. As shown in [Fig. 1A(i–iv) ], this process was prevented following pretreatment of platelets with a selective human anti-GPIbα antibody (SZ2), whereas a control immunoglobulin G (IgG) antibody had no effect. This suggests that SARS-CoV-2 triggered platelet degranulation in an α-thrombin- and GPIbα-dependent manner. This is consistent with a report that GPIbα is the major α-thrombin binding site on the platelet surface with evidence for a high affinity site within its extracellular domain adjacent to the von Willebrand factor (VWF) binding site.[6 ] Also, at a higher α-thrombin concentration, the protease-activated receptors (PAR1 and PAR4) likely become significantly engaged and further enhance degranulation and extent of platelet hyper-activation and aggregation.[7 ]
Fig. 1 (A ) GPIbα positively regulates human platelet degranulation induced by low doses of α-thrombin in COVID-19 patients. (Ai) Platelet P-selectin expression was measured (percent of CD62P positive platelets) in washed human platelets by flow cytometry at baseline in 10 healthy donors and 10 COVID-19 patients. Platelets were pretreated with human anti-GPIbα antibody (SZ2, 20 µg/mL) or its isotype IgG control for 5 minutes at 37°C. Degranulation was then initiated by α-thrombin at 0, 0.05 (priming dose), and 0.2 U/mL (positive control). Histogram represents the mean of data ± standard deviation (SD) of plots for P-selectin (CD62P) expression (n = 10); **p < 0.01; ****p < 0.0001. Statistical significance was analyzed using one-way analysis of variance (ANOVA) with subsequent Dunnett's t -test for comparison against a single group. (Aii) Effect of anti-GPIbα on P-selectin (CD62P) expression in COVID-19 patients, as assessed by flow cytometry. Left plots represent resting platelets. Right plots represent platelets in the presence of a priming concentration of α-thrombin (0.05 U/mL). (Aiii) ATP release was measured by a Lumi-Aggregometer (Luciferase assay; Chrono-Lume, Chrono-log). Results are expressed as a measure of increase in luminescence and (Aiv) dense granule secretion was evaluated by measuring the loss of mepacrine fluorescence following activation by α-thrombin (0.05 U/mL). Histogram represents the mean of data ± SD of (Aiii) ATP release (fold increase) and (Aiv) remaining mepacrine fluorescence (n = 10); ****p < 0.0001. The reference range (resting platelets) is represented by the shaded gray region. (B ) Pretreatment with a human anti-GPIbα antibody prevents COVID-19-induced potentiation of platelet aggregation while Bernard–Soulier Syndrome (BSS) platelets are not hyper-reactive. (Bi) Platelets were pretreated with the human anti-GPIbα antibody (SZ2, 20 μg/mL) for 5 minutes at 37°C. Aggregation was then initiated by low concentrations of α-thrombin (0.05 U/mL). Histogram represents the mean of data ± SD of aggregation traces (n = 10); ****p < 0.0001. (Bii) Representative traces of platelet aggregation induced by a priming dose of α-thrombin (0.05 U/mL). (Biii) Specific blockade of the thrombin-binding site on GPIbα, by either SZ2 or VM16d blocking monoclonal antibodies, prevented the potentiation of thrombin-induced platelet aggregation in COVID-19 patients. (Biv) Platelets isolated from COVID-19 patients with (or without) BSS (deficiency in GPIb-V-IX complex) were stimulated by low doses of α-thrombin (0.05 U/mL). (Bv) Representative traces of platelet aggregation (shown in Biv) induced by a priming dose of α-thrombin (0.05 U/mL). (C ) Scheme of the potential role the α-thrombin/GPIbα axis plays in platelet hyper-reactivity in COVID-19 patients. During SARS-CoV-2 infection, α-thrombin stimulates its high-affinity receptor GPIbα leading to α- and dense granule release and platelet hyper-aggregation. Receptor blockade or deficiency prevents the hyper-coagulable state occurring in COVID-19 patients. IgG, immunoglobulin G.
Similarly, we assessed whether the α-thrombin–GPIbα axis plays a role in the hyper-aggregation of platelets from COVID-19 patients. As anticipated, priming, but not high concentration of α-thrombin-induced aggregation, was significantly inhibited in platelets that were pretreated with a human anti-GPIb antibody (SZ2) (Beckman Coulter, sodium azide free), as compared with COVID-19 and to COVID-19 control IgG groups ([Fig. 1Bi, Bii ]). Thus, specific inhibition of the high-affinity binding site of α-thrombin on GPIbα was found to inhibit potentiated washed human platelet secretion and aggregation response, indicating that the α-thrombin/GPIbα axis can potentiate platelet function in the presence of suboptimal α-thrombin concentrations.
To further assert our finding based on the mouse-anti-human antibody, clone SZ2, that targets the anionic/sulfated tyrosine sequence 269 to 282 of GPIbα, we performed a similar experiment using a mouse anti-human antibody, clone VM16d (Abcam, dialyzed against phosphate-buffered saline in our laboratory to remove sodium azide), that maps the C-terminal flanking sequence 226 to 268 of GPIbα.[8 ] Our results show that specific blockade of the thrombin-binding site on GPIbα, by either SZ2 or VM16d ([Fig. 1Biii ]) blocking monoclonal antibodies, prevented the potentiation of thrombin-induced platelet aggregation in COVID-19 patients.
To confirm our pharmacological-based approach, the key role for GPIbα was assessed using platelets isolated from COVID-19 patients with BSS (deficiency in GPIb-V-IX complex causing an absence of GPIb expression). Platelets from these patients were found to have markedly diminished aggregation in response to low concentrations of α-thrombin compared with COVID-19 patients without BSS ([Fig. 1Biii, Biv ]). Indeed, SARS-CoV-2 infection failed to trigger platelet hyper-aggregation in patients with BSS suggesting that GPIbα positively regulates platelet aggregation downstream of α-thrombin in COVID-19 patients ([Fig. 1C ]).
BSS is characterized by prolonged bleeding time, thrombocytopenia,[9 ]
[10 ] and giant platelets lacking the surface membrane glycoprotein GPIb of the GPIb-IX-V complex.[10 ]
[11 ]
[12 ] The GPIb-IX-V complex is a platelet-specific adhesion-signaling complex, and it consists of GPIbα linked to GPIbβ via a disulfide bond and to GPIX and GPV noncovalently. Here, GPIbα is the major ligand-binding subunit and binds thrombin and the adhesive ligand VWF.[8 ] Given that platelets of BSS patients hardly express GPIb, BSS platelets lack GP1b-specific response to α-thrombin. Similarly, SZ2, an anti-GPIbα monoclonal antibody, is known to inhibit VWF binding to platelet and platelet aggregation.[13 ] Thus, both SZ2 antibody-treated platelets and platelets of BSS lack GP1bα function and GP1bα-driven response to α-thrombin. Other factors may contribute to the diminished response of BSS platelets to thrombin and SARS-CoV-2 infection, including the additional loss of GPV and GPIX. The loss of surface-membrane glycoprotein in BSS platelets has been closely linked to in vitro and in vivo functional impairment.[9 ]
[10 ]
Our finding was supported by a recent study[14 ] that worked on S100A8/A9, also known as “calprotectin” or “MRP8/14,” an alarmin primarily secreted by activated myeloid cells with antimicrobial, proinflammatory, and prothrombotic properties. This research group identified the S100A8/A9-GPIbα axis as a novel targetable prothrombotic pathway inducing procoagulant platelets and fibrin formation, in particular in diseases associated with high levels of S100A8/A9, such as COVID-19.
Taken together, GPIbα interaction with low concentration α-thrombin is a potential contributor to the formation of procoagulant platelets in COVID-19 patients. Such interaction is probably only one aspect of many that likely regulate platelet activity in COVID-19 patients. These observations provide a valuable foundation for understanding the disease pathophysiology and to identify a new target for treatment options.
