Introduction

The coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), results in significant morbidity and mortality [1, 2]. Platelet hyperreactivity is a hallmark of COVID-19 and contributes to an increased risk of thrombotic disorders observed in many COVID-19 patients [3, 4]. Platelets are small and anucleate blood cells that play crucial roles in hemostasis, thrombosis, and immune response [5]. Excessive platelet aggregation and activation contribute to occlusive thrombus formation and the development and progression of atherothrombosis [6]. As mounting evidence supports a link between COVID-19 and platelet hyperreactivity, the underlying mechanism for this association has multiple theories.

SARS-CoV-2 infects host cells via binding of spike protein to angiotensin-converting enzyme-2 (ACE2). However, it remains controversial whether platelets express ACE2 and if SARS-CoV-2 interacts with platelets through an ACE2-mediated mechanism. It is demonstrated that the SARS-CoV-2 spike protein has an RGD (arginine-glycine-aspartate) sequence in the receptor-binding domain, a motif that could bind to cell-surface integrins [7]. Integrin αIIbβ3, a highly abundant heterodimeric platelet receptor, plays a central role in platelet functions, hemostasis, and arterial thrombosis. Integrin αIIbβ3 can bind to several RGD-containing ligands, including fibrinogen and von Willebrand factor (vWF), which triggers various outside-in signaling events that further facilitate platelet spreading, stabilizes platelet aggregation, and so on [8]. Since Xiaoying Ma and their colleagues have found that SARS-CoV-2 spike protein RBD binds platelet β3 integrin [9], whether spike protein upregulates outside-in signaling to potentiate platelet function is still unknown. In this study, we demonstrated that SARS-CoV-2 spike protein directly potentiated platelet aggregation and spreading by interacting with platelet integrin αIIbβ3 and upregulating the outside-in signaling pathway, suggesting a potential therapeutic target for the prevention and treatment of platelet hyperreactivity in subjects with COVID-19.

Materials and methods

Chemicals

Detailed descriptions of the reagents and antibodies are provided in Supplement material: Expanded Materials and Methods.

Platelet preparation, platelet functional assays and Western Blotting

This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the School of public health (Shenzhen), Sun Yat-Sen University [[2020] No.013]. Healthy subjects aged between 18 to 30, recruited from Sun Yat-Sen University, participated in the study. Subjects who had self-reported cardiovascular disease or taken any antiplatelet drugs or nutrients within the previous 2 weeks were excluded. Whole human blood was drawn and gel-filtered platelets were prepared according to our previously described methods [10].

Detailed descriptions of the methods of platelet aggregation, activation, spearing on immobilized spike protein and western blotting are descripted in Supplement material: Expanded Materials and Methods.

Statistical analysis

Data were expressed as mean ± standard deviation of the means (SD) of at least three independent experiments. One-way analysis of variance followed by Tucky post hoc analysis was used for multiple comparisons and paired Student’s t test was used for two paired sample comparisons. Statistical analysis was performed using SPSS 25.0. P<0.05 was considered statistically significant.

Results

Spike protein potentiates platelet aggregation and upregulates platelet integrin αIIbβ3 outside-in signaling in vitro

According to a previous study that has demonstrated that spike protein (0.001-1 μg/ml) induced platelet aggregation in a dose-dependent manner [11], we selected 1 μg/ml spike protein as the standard concentration for subsequent experiments and observed that 1 μg/ml spike protein significantly promoted platelets aggregation in response to thrombin, collagen, and ADP (Fig. 1A). However, we found no significant effects of spike protein on platelet granule secretion, such as the surface expression of P-selectin (Fig. 1B). The inside-out signaling to integrin αIIbβ3 leads to its activation and ligand binding [12]. Using PAC-1 binding assay, we observed that spike protein did not affect integrin αIIbβ3 inside-out signaling (Fig. 1C).

Fig. 1
figure 1

Spike protein potentiates platelet aggregation and upregulates platelet integrin αIIbβ3 outside-in signaling in vitro. (A) Platelets were preincubated with spike protein at 1 μg/mL for 10 min, and then gel-filtered platelets were stimulated with 0.1 U/mL thrombin or 1 μg/mL collagen, and PRP stimulated with 5 μM ADP. Aggregation was assessed under stirring at 1200 rpm (n=4). (B-C) Gel-filtered platelets were incubated with 1 μg/mL spike protein for 10 min and then stimulated with 0.1 U/ml thrombin, and then the expression of P-selectin (B) (n=3) or PAC-1 binding (C) (n=3) was analyzed using a flow cytometer. (D) Gel-filtered platelet was allowed to spread on the spike protein (1, 10 μg/ml) or positive control (100 μg/mL fibrinogen) coated surfaces (blocked with 1% BSA) for 60 min at 37 °C, and the representative images and summary data of the mean platelet surface area were shown (n=3). (E) After aggregation in (A), the platelets were collected and then lysed with lysis buffer. The cell lysates were prepared and Western Blot was performed to assess the phosphorylation levels of β3 (n=4) and c-Src (n=3). * P<0.05, ** P<0.01, *** P<0.001, indicate significant difference compared with control group. # P<0.05, ## P<0.01, ### P<0.001, indicate significant difference between the two group. NS: no significant difference between the two groups. Statistical analyses were performed using paired two-tailed Student’s t test to compare the difference between control and spike protein groups in (A-C) and (E). One-way ANOVA with Tukey’s post hoc test was performed in (D)

Full size image

Platelet spreading is primarily regulated by integrin αIIbβ3 outside-in signaling [8]. Spike protein has an RGD-containing motif that enables it to bind with platelet integrin β3 [9], indicating platelets could adhere to immobilized spike protein similarly to immobilized fibrinogen and further inducing platelet spreading [13]. We found that platelets spreading on immobilized BSA (the control group) remained discoid, while the surface area of platelets spreading on immobilized spike protein was significantly greater (Fig. 1D). We further analyzed whether spike protein can directly affect platelet integrin αIIbβ3 outside-in signaling by measuring phosphorylation of β-integrin tail and c-Src in platelets. We further found that only spike protein could potentiate the phosphorylation of β-integrin tail and c-Src in resting platelets (Supplemental Figure 1), although spike protein cannot induce platelet activation without agonists [14]. Moreover, spike protein significantly potentiated the phosphorylation of β-integrin tail and c-Src (Fig. 1E) in thrombin-activated human platelets. These findings suggest that spike protein directly upregulates integrin αIIbβ3 outside-in signaling.

Spike protein potentiates platelet aggregation via interacting with platelet integrin αIIbβ3

Firstly, whether ACE2 participants in the process of interaction between SARS-CoV-2 and platelet exits contradiction. Western blot assay showed that human platelets exhibited trace expression of ACE2 and TMPRSS2. These levels were far below those in PBMC (Supplemental Figure 2), indicating spike protein may use other receptors instead of ACE2 to interact with platelets.

As spike protein contains RGD sequence to bind cell-surface integrins and could bind to platelet β3 integrin [9], we further explore whether spike protein potentiates platelet aggregation via interacting with integrin αIIbβ3. To test whether spike protein blocks fibrinogen binding integrin αIIbβ3, we performed the fibrinogen-binding assay, in which platelets were incubated with labeled fibrinogen in the presence or absence of spike protein, to examine the formation of the platelet-fibrinogen complex. As shown in Fig. 2A, we found that 10 μg/ml spike protein significantly inhibited platelet fibrinogen binding and had no significant difference to 100 μg/ml fibrinogen from human plasma which directly interacts with integrin αIIbβ3. We next use tirofiban, the integrin αIIbβ3 specific antagonist, to prevent spike protein binding to αIIbβ3. We found that spike protein did not further potentiate platelet aggregation (Fig. 2B) and the phosphorylation of β3 (Fig. 2C) in the presence of tirofiban. These results suggest that spike protein potentiates platelet aggregation via interacting with platelet integrin αIIbβ3.

Fig. 2
figure 2

Spike protein potentiates platelet aggregation via interacting with platelet integrin αIIbβ3 and upregulating outside-in signaling. (A) Platelets were incubated with spike protein (10 μg/mL) or fibrinogen (100 μg/mL) for 10 min and stimulated with thrombin (0.1 U/mL) for 10 min. Platelet fibrinogen binding was analyzed using a flow cytometer (n=4). (B) Human PRP was incubated with spike protein at 1 μg/mL for 10 min in the presence of tirofiban (10 nM) for 1 min. Platelet aggregation assays were performed after stimulated with 5 μM ADP (n=3). (C) After aggregation in (B), the cell lysates were prepared and the phosphorylation levels of β3 were determined by Western Blot (n=3). (D-E) Platelets were preincubated with 10 μM PP2 for 10 min and incubated with spike protein at 1 μg/mL for 10 min. Platelet aggregation assays were performed after stimulated with 0.1 U/ml thrombin (n=4) (D). After platelet aggregation, the level of p-c-Src (E) was determined by Western blotting (n=3). * P<0.05, ** P<0.01, *** P<0.001, indicate significant difference between two groups. NS: no significant difference between the two groups. Statistical analyses were performed using One-way ANOVA with Tukey’s post hoc test.

Full size image

Spike protein potentiates platelet aggregation via upregulating outside-in signaling

As the c-Src is proved to be the dominant effector of αIIbβ3 outside-in signaling [15, 16], we used 4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolo[3,4-d]pyrimidine (PP2), the Src inhibitor, to block c-Src activation and thus inhibiting outside-in signaling. We observed that spike protein did not further potentiate platelet aggregation (Fig. 2D) and the phosphorylation of c-Src (Fig. 2E) in the presence of PP2 (10 μM). These observations indicate that the potentiating effect of spike protein on platelet aggregation is predominantly mediated by upregulating integrin αIIbβ3 outside-in signaling.

Discussion

COVID-19 causes platelet hyperreactivity and severe thrombotic disorders, leading cause of mortality and morbidity worldwide. Multiple studies have found that platelets isolated from COVID-19 patients show increased platelet aggregation, integrin αIIbβ3 activation and P-selectin expression [4, 17]. Moreover, the pathogenicity of SARS-CoV-2 may be attributed to spike protein engaging with host cell receptors. In this study, we observed that spike protein potentiates platelet aggregation which plays a critical role in hemostatic plug and thrombus formation [18].

The integrins, a superfamily of cell adhesion receptors, play critical roles in cell adhesion and cell-extracellular matrix (ECM) interaction [19]. Integrin αIIbβ3 is the most abundant surface-expressed integrin in platelets and can bind to several RGD-containing ligands. Upon stimulation with agonists, the binding of activated αIIbβ3 and soluble ligands can be detected within seconds [20]. Since spike protein has an RGD sequence and could bind to platelet β3 integrin [9], we further perform Fg-binding assay to verify this viewpoint. While a low concentration of spike protein (2 μg/ml) did not affect platelet fibrinogen binding when stimulated with thrombin [14], we found that a high concentration of spike protein (10 μg/ml) attenuated platelet fibrinogen binding, similar to 100 μg/ml fibrinogen that directly binds to αIIbβ3. Additionally, we observed that the spike protein-potentiated platelet aggregation and the phosphorylation of β-integrin tail were abolished after being pre-incubated with tirofiban, a selective and reversible platelet integrin αIIbβ3 receptor antagonist. Consistent with this, another study also observed that RGD peptide, an αIIbβ3 inhibitor, blocked spike protein-induced platelet aggregation [11]. Moreover, subthreshold doses of GPIIb/IIIa blockers like eptifibatide and tirofiban could prevent thrombus formation in COVID-19 [21]. These results suggest that spike protein interacts directly with integrin αIIbβ3 to affect platelet aggregation, and indicate a potential target for SARS-CoV-2 interacting with platelet.

Platelet integrin αIIbβ3 outside-in signaling induces multiple events, such as stabilized platelet aggregation, spreading, and granule secretion, playing a critical role in the process of thrombosis [8]. We observed that spike protein upregulated platelet integrin αIIbβ3 outside-in signaling, like increasing exposure to immobilized spike protein spread and the phosphorylation of β-integrin tail, c-Src. Since spike protein could bind to platelet integrin αIIbβ3, we used immobilized spike protein to assess platelet spread function, which better reflects the effect of spike protein on platelet spread. Following ligand binding to integrin αIIbβ3, integrin αIIbβ3 clustering promotes Src activation, and the activated Src phosphorylates and supports the activation of numerous signaling proteins [8]. Therefore, using PP2, to inhibit c-Src activity, we observed that the spike protein did not further enhance platelet aggregation. The PP2 concentration (10 μM) chosen in blocked experiment was referred to previous study [22]. Thus, we conclude that the potentiated effect of spike protein on platelet aggregation is mainly mediated by the upregulation of integrin αIIbβ3 outside-in signaling.

In the present study, our data shows that spike protein upregulates integrin αIIbβ3 outside-in signaling pathway to potentiate platelet aggregation. Other studies found that SARS-CoV-2 spike protein could potentiate platelet aggregation via ACE2 [17] or CD42b [11] -mediated signaling pathway, suggesting that spike protein may interact with platelets through a variety of targets. Subsequent experiments need to further investigate which target plays a more important role in this process. In conclusion, this integrin αIIbβ3-mediated outside-in signaling mechanism may be a potential therapeutic target to preventing and treating platelet hyperreactivity in COVID-19.