Antibody-induced procoagulant platelets in severe COVID-19 infection

Abstract

The pathophysiology of COVID-19-associated thrombosis seems to be multifactorial. We hypothesized that COVID-19 is accompanied by procoagulant platelets with subsequent alteration of the coagulation system. We investigated depolarization of mitochondrial inner transmembrane potential (ΔΨm), cytosolic calcium (Ca2+) concentration, and phosphatidylserine (PS) externalization. Platelets from COVID-19 patients in the intensive care unit (ICU; n = 21) showed higher ΔΨm depolarization, cytosolic Ca2+, and PS externalization compared with healthy controls (n = 18) and non-ICU COVID-19 patients (n = 4). Moreover, significant higher cytosolic Ca2+ and PS were observed compared with a septic ICU control group (ICU control; n = 5). In the ICU control group, cytosolic Ca2+ and PS externalization were comparable with healthy controls, with an increase in ΔΨm depolarization. Sera from COVID-19 patients in the ICU induced a significant increase in apoptosis markers (ΔΨm depolarization, cytosolic Ca2+, and PS externalization) compared with healthy volunteers and septic ICU controls. Interestingly, immunoglobulin G fractions from COVID-19 patients induced an Fcγ receptor IIA-dependent platelet apoptosis (ΔΨm depolarization, cytosolic Ca2+, and PS externalization). Enhanced PS externalization in platelets from COVID-19 patients in the ICU was associated with increased sequential organ failure assessment score (r = 0.5635) and D-dimer (r = 0.4473). Most importantly, patients with thrombosis had significantly higher PS externalization compared with those without. The strong correlations between markers for apoptosic and procoagulant platelets and D-dimer levels, as well as the incidence of thrombosis, may indicate that antibody-mediated procoagulant platelets potentially contributes to sustained increased thromboembolic risk in ICU COVID-19 patients.

Graphical abstract

Figure 1.

Platelet apoptosis in COVID-19 patients. (A-C) Changes in apoptosis pathways were analyzed by assessing the depolarization of the ΔΨm (A), cytosolic calcium concentration (B), and PS externalization (C) in platelets from COVID-19 patients in the ICU or COVID-19 patients not in the ICU, as well as ICU non–COVID-19 patients (control group) and healthy donors, respectively. (D) Quantification of cleaved-caspase 9 level in platelets from COVID-19 patients in the ICU normalized to healthy donors. (E) Representative western blot showing GAPDH and cleaved-caspase 9 proteins in platelets from COVID-19 patients in the ICU. Protein bands were detected with the infrared imaging system (Odyssey, LI.COR, Lincoln, NE). (F) Diagram indicating the number of COVID-19 patients in the ICU positive for each apoptotic parameter: ΔΨm depolarization, cytosolic calcium concentration, and PS externalization. Data are presented as mean ± standard error mean (SEM) of the measured fold increase compared with control. Not significant, *P < .05, **P < .01, ***P < .001, ****P < .0001. The number of patients and healthy donors tested is reported in each graphic. Dashed lines represent the cutoffs determined from healthy donors as mean of fold increase (FI) + 2× SEM.

Figure 2.

Association among platelet apoptosis and clinical biomarkers, thromboembolic complications, and mortality. The correlations between platelet apoptosis parameters and PLT count, as well as D-dimer and SOFA score measured at the same day of platelet testing, were assessed. (A-B) An association was observed between PLT count and PS externalization (A) and cytosolic calcium concentration (B). (C) Moreover, a significant correlation was detected for D-dimer and PS externalization. (D) The clinical relevance of PS externalization was assessed using the SOFA score and revealed a significant positive correlation. Pearson’s correlation coefficients were calculated and are shown in the panels. (E-F) The PS externalization (E) and the cytosolic calcium concentration (F) were determined and compared between COVID-19 patients in the ICU depending on the incidence of thromboembolic complications and mortality, respectively. Data are presented as mean of the measured FI compared with control. Not significant, *P < .05, **P < .01, ***P < .001, ****P < .0001. The number of patients and healthy donors tested is reported in each graphic. Dashed lines represent the cutoffs determined from healthy donors as mean of FI + 2× SEM.

Figure 3.

Impact of sera from COVID-19 patients on platelet apoptosis. (A-C) Changes in apoptosis pathways induced by sera from COVID-19 patients in the ICU, non–COVID-19 patients in the ICU (control group), and healthy donors were investigated by assessing the depolarization of the ΔΨm (A), cytosolic calcium concentration (B) (n = 19 because of the lack of biomaterial for 2 patients), and PS externalization (C). (D) Quantification of cleaved caspase 9 level in platelets from healthy donors after incubation with patient sera, normalized to sera from healthy donors. (E) Representative western blot of GAPDH and cleaved caspase 9 proteins. Protein bands were detected with the infrared imaging system (Odyssey, LI.COR). (F) Diagram indicating the number of COVID-19 patients in the ICU positive for each apoptotic parameter: ΔΨm depolarization, cytosolic calcium concentration, and PS externalization. The 2 sera tested only for ΔΨm depolarization and PS externalization are in parentheses. Data are presented as mean ± SEM of the measured FI compared with control. Not significant, *P < .05, **P < .01, ***P < .001, ****P < .0001. The number of sera tested is reported in each graphic. Dashed lines represent the cutoffs determined testing sera from healthy donors as mean of FI + 2× SEM.

Figure 4.

IgG fractions from COVID-19 patients in the ICU induce platelet apoptosis via crosslinking FcγRIIA. (A-C) Changes in apoptosis pathways induced by IgG fraction from COVID-19 patients in the ICU or COVID-19 patients not in the ICU, as well as non–COVID-19 patients in the ICU (control group) and healthy donors, were analyzed by assessing the depolarization of the ΔΨm (A), cytosolic calcium concentration (B), and PS externalization (C) in platelets from 3 different healthy donors. (D-F) The same assays were performed in the presence of the IV.3 monoclonal antibody (mAb) to block FcγRIIA signaling. Each dot represents 2 experiments with platelets from 2 different donors. Data are presented as mean ± SEM of the measured FI compared with control. Not significant, *P < .05, **P < .01, ***P < .001, ****P < .0001. The number of sera tested is reported in each graphic.

Figure 5.

The impact of CsA and Q-VD-OPh on antibody-mediated procoagulant platelets. (A-B) CsA and Q-VD-OPh inhibition of COVID-19–induced ΔΨ depolarization (A) and PS externalization (B). (C-D) CsA significantly inhibits the depolarization of ΔΨm and PS expression in COVID-19–positive patients. Q-VD-OPh was not able to inhibit ΔΨm depolarization (C) but significantly reduced PS externalization (D). Data are presented as mean ± SEM of the measured FI compared with control. Not significant, *P < .05, **P < .01, ***P < .001, ****P < .0001. The number of sera tested is reported in each graphic.