NLRP3 inflammasome-mediated platelet hyperreactivity in sickle cell mice is targetable by BTK inhibition

TL;DR

This study shows that inhibiting BTK can reduce platelet hyperreactivity in sickle cell disease mice by targeting the NLRP3 inflammasome.

Contribution

The study demonstrates that BTK inhibition can target NLRP3 inflammasome-mediated platelet hyperreactivity in sickle cell disease.

Findings

Platelet function assays in SCD mice show elevated responses reduced by NLRP3 or BTK inhibitors.

BTK inhibition with ibrutinib reduces platelet hyperreactivity in SCD mice.

Nigericin partially reverses the effects of ibrutinib on platelet function.

Abstract

The platelet nucleotide-binding domain leucine-rich repeat-containing protein 3 (NLRP3) inflammasome is upregulated in sickle cell disease (SCD) and promotes platelet aggregation. We previously identified Bruton tyrosine kinase (BTK) as a critical regulator of the platelet NLRP3 inflammasome. However, whether NLRP3 contributes to platelet function beyond aggregation in SCD and whether these effects can be modulated through BTK inhibition, has been incompletely understood. Here, we show that platelet secretion, platelet spreading, platelet aggregation, and in vitro thrombus formation in response to collagen are elevated in SCD mice and are reduced following treatment of mice with the NLRP3 inhibitor MCC950 or the BTK inhibitor ibrutinib. The NLRP3 activator nigericin partially reversed the inhibitory effects of ibrutinib across all platelet function assays. Together, we identify the NLRP3 inflammasome as a critical mediator of platelet hyperreactivity in SCD mice, which can be targeted via BTK inhibition.

Figures

D](#F1)) and dense granules (p = 0.0199; [Fig. 1E](#F1)), while significantly reducing platelet P-selectin expression (p < 0.0001; [Fig. 1F](#F1)), platelet ATP secretion (p < 0.0001; [Fig. 1G](#F1)), platelet spreading (p = 0.0004; [Fig. 1H](#F1) and [I](#F1)), platelet aggregation (p < 0.0001; [Fig. 1J](#F1)), and in vitro thrombus formation (p < 0.0001; [Fig. 1K](#F1)). In AA mice, MCC950 treatment had a significant inhibitory effect on platelet aggregation (p = 0.0153; [Fig. 1J](#F1)), but not on any of the other platelet readouts ([Fig. 1D](#F1), [E](#F1), [F](#F1), [G](#F1), [H](#F1), [I

C](#F2)), platelet ATP secretion (p < 0.0001; [Fig. 2D](#F2)), platelet spreading (p = 0.0039; [Fig. 2E](#F2) and [F](#F2)), platelet aggregation (p < 0.0001; [Fig. 2G](#F2)), and in vitro thrombus formation (p = 0.0004; [Fig. 2H](#F2)). Nigericin had no significant effect in vehicle-treated SS mice ([Fig. 2C](#F2)–[H](#F2)). In ibrutinib-treated AA mice, nigericin significantly increased platelet P-selectin expression (p = 0.0148; [Fig. 2C](#F2)), platelet ATP secretion (p = 0.0206; [Fig. 2D](#F2)), and platelet aggregation (p = 0.0023; [Fig. 2G](#F2)), but had no significant effect on platel

Full text

Introduction

Patients with sickle cell disease (SCD) exhibit elevated platelet activation [1,2], which contributes to an increased risk of thrombosis and inflammation [3]. The nucleotide-binding domain leucine-rich repeat-containing protein 3 (NLRP3) inflammasome is a key inflammatory mediator that, upon activation, assembles with the adaptor apoptosis-associated speck-like protein containing a CARD (ASC) to form a multiprotein complex promoting caspase-1 activation [4]. In platelets, the NLRP3 inflammasome was first reported in the setting of dengue fever [5], and has since been implicated in various other diseases associated with abnormal coagulation and inflammation [6]. We have recently shown that the platelet NLRP3 inflammasome is abnormally activated in both patients with SCD and the Townes SCD mouse model [7–9], with the latter closely recapitulating key features of human SCD [10]. Platelet NLRP3 inflammasome activation in SCD mice leads to microvascular obstruction of lung arterioles through the formation of platelet-neutrophil aggregates [11] and promotes platelet aggregation [7–9]. However, the role of the NLRP3 inflammasome in driving overall platelet hyperreactivity in SCD and its potential as a therapeutic target remain incompletely understood.

Growing interest in targeting the NLRP3 inflammasome has spurred the development of multiple NLRP3 inhibitors, but their clinical progress has been limited by hepatotoxicity concerns and other adverse effects [12]. We and others have identified Bruton tyrosine kinase (BTK) as a critical regulator of NLRP3 inflammasome activation [13–16]. In SCD mice, we showed that the BTK inhibitor ibrutinib interferes with upregulated platelet aggregation [7,8], although it was unclear in these studies whether this effect was mediated through the NLRP3 inflammasome.

Here, we studied the role of the NLRP3 inflammasome in platelet hyperreactivity in Townes SCD mice and tested whether BTK inhibition with ibrutinib interferes with these responses in an NLRP3-dependent manner.

Materials and methods

Study approval

2.1.

All studies were approved by the NIH Clinical Center Animal Care and Use Committee (DPM23–01, DPM23–03) and complied with the NIH Guide for the Care and Use of Laboratory Animals.

Animals and inhibitor injections

2.2.

We used Townes SCD mice, the B6; 129-Hba^tm1(HBA)Tow^ Hbb^tm2(HBG1,HBB^^)Tow*^/Hbb^tm3(HBG1,HBB)Tow^/J strain [10]. SS mice express human HbS (HbSS) and exhibit a severe SCD phenotype mirroring that of SCD patients, whereas AA mice express human HbA (HbAA) and served as controls. All experiments included balanced numbers of age-matched male and female mice. Mouse breeding and genotyping were performed as reported [17]. Mice received IV injections of the NLRP3 inhibitor MCC950 (Cayman Chemical, 50 mg/kg) or the BTK inhibitor ibrutinib (Selleckchem, 10 mg/kg) 4 h prior to blood collection by cardiac puncture into heparinized syringes. Procedures were conducted under isoflurane anesthesia. Treatments were well tolerated and did not significantly alter complete blood counts (Element HT5, Heska) (data not shown).

Platelet isolation

2.3.

Murine platelets were isolated as previously described [18]. Platelet-rich plasma (PRP) was obtained by centrifugation at 260×g for 5 min, followed by another centrifugation step at 640×g for 5 min to pellet the platelets. Collagen (ChronoLog) at a concentration of 1 μg/mL was used as platelet agonist for all assays unless otherwise indicated.

Immunofluorescence staining of platelets and confocal microscopy

2.4.

Isolated platelets were stained with anti-NLRP3 monoclonal antibody (2 μg/mL, mouse IgG2b, AdipoGen) and anti-ASC polyclonal antibody (2 μg/mL, rabbit IgG, Biorbyt) as previously described [9]. Alexa Fluor 488-conjugated goat anti-mouse IgG (1:100, Invitrogen) and Alexa Fluor 568-conjugated donkey anti-rabbit IgG (1:100, Invitrogen) served as secondary antibodies. Corresponding IgG antibodies served as controls. Confocal stacks were acquired on a Zeiss 880 using a 63x oil objective (NA 1.4) and Alexa 488/568 were excited at 488/561 nm with emission at 490–562/568–700 nm. Images were collected at 0.8–1.0 AU with voxel dimensions of 0.08 × 0.08 × 0.37 μm. Image deconvolution and colocalization/3D volumetric analyses were performed using Huygens Essential v23.10 (Scientific Volume Imaging) and Imaris v10.1 (Oxford Instruments), respectively. Pearson correlation coefficients were calculated following background masking and Costes automatic thresholding [19].

Transmission electron microscopy

2.5.

Platelets were fixed in 2.5 % glutaraldehyde and 1 % paraformaldehyde prepared in 0.1 M cacodylate buffer and processed for transmission electron microscopy (TEM). Samples were post-fixed with 1 % OsO_4_, stained with 1 % uranyl acetate, dehydrated, and resin-embedded (EMBed 812, Electron Microscopy Sciences). Ultrathin sections (80 nm) were stained with UranyLess and lead citrate and imaged by TEM (JEM-1200EXII, JEOL) at 80 kV using an XR611 M CCD camera (Advanced Microscopy Techniques). Intraplatelet α-granules and dense granules were identified by a hematopathologist in representative TEM images and quantified independently by two investigators. Investigators were blinded to the experimental groups. Granule classification followed established platelet ultrastructural criteria [20]. Mean numbers of α-granules and dense granules per platelet were calculated from 4 random images per mouse.

Flow cytometric evaluation of P-selectin expression on platelets

2.6.

To assess platelet α-granule degranulation, platelet surface expression of CD62P (P-selectin) was measured by flow cytometry using an APC-conjugated anti-CD62P monoclonal antibody (mouse IgG1κ; 17–0626-82, eBioscience). A PE-conjugated anti-CD41 antibody (rat IgG1κ; 133906, Biolegend) was added to identify platelets. Flow cytometry was performed with a FACS Calibur (BD Biosciences). Where indicated, platelets were incubated for 30 min with the NLRP3 inflammasome activator nigericin (10 μM, Cayman Chemical).

ATP release from platelets

2.7.

ATP release from platelets was determined to study the secretion of platelet dense granules as reported [21]. After calibration of one whole blood sample with ATP standard (ChronoLog), ATP concentrations were determined using the ChronoLume luciferin assay (ChronoLog) on a Lumi-Aggregometer (Model 700, ChronoLog) according to the manufacturer’s protocol. Where indicated, blood was incubated for 30 min with the NLRP3 inflammasome activator nigericin (10 μM).

Platelet spreading and scanning electron microscopy

2.8.

Isolated murine platelets were allowed to adhere and spread on collagen-coated coverslips (100 μg/mL) for 30 min. Where indicated, platelets were incubated for 30 min with the NLRP3 inflammasome activator nigericin (10 μM). Adherent platelets were fixed in 2.5 % glutaraldehyde, post-fixed with 1 % OsO_4_, dehydrated, critical-point dried (Samdri-795, Tousimus), gold-coated (5 nm), and imaged by SEM (Zeiss Evo MA10, Carl Zeiss) at 8 kV, 25 pA, and 8,000x. Images are representative of at least 3 random platelets for each sample. Platelet area (μm^2^) was quantified using machine learning software (Aivia v9.0, SVision LLC) [8,22] and mean platelet area was determined for each sample.

Platelet aggregometry

2.9.

Platelet aggregation was evaluated using whole-blood impedance aggregometry (Model 700, ChronoLog) as described [21]. Where indicated, blood was incubated for 30 min with the NLRP3 inflammasome activator nigericin (10 μM). Aggregation was evaluated after 6 min at 37 °C with a stir speed of 1200 rpm. Analysis was performed using Aggrolink-8 software (ChronoLog). Data are reported as area under the curve (AUC).

Flow chamber

2.10.

Heparinized murine blood was diluted 2:1 with Tyrodes-HEPES buffer and perfused through a transparent flow chamber (slit depth 50 μm) over a collagen-coated surface (100 μg/mL) with shear rates at 1700 s^−1^ as reported [21]. Where indicated, blood was incubated for 30 min with the NLRP3 inflammasome activator nigericin (10 μM). Pictures were taken from at least 2 microscopic areas (using 20x optical objectives) for each sample. Thrombus area was analyzed with ImageJ software (NIH), and mean percentage values of thrombus area were determined for each run.

Statistical analysis

2.11.

Data were analyzed using two-tailed Student’s t-test or analysis of variance (ANOVA), as appropriate. For each dependent variable, an ANOVA model was fitted with in vivo treatment, genotype, and in vitro treatment as independent variables, including interaction terms. Model diagnostics were examined to confirm that data met ANOVA assumptions. Experiments involving repeated measurements were analyzed using repeated-measures ANOVA or mixed-effects models (restricted maximum likelihood), as appropriate. Control of the false discovery rate for multiple hypothesis testing was performed using the two-stage step-up method of Benjamini, Krieger, and Yekutieli, and adjusted p-values are reported. Statistical analyses were performed using GraphPad Prism (v10). Data are presented as mean ± standard deviation (SD).

Results

The NLRP3 inflammasome promotes platelet secretion, platelet spreading, platelet aggregation, and in vitro thrombus formation in sickle cell mice

3.1.

Consistent with previous findings [7–9], the platelet NLRP3 inflammasome was upregulated in SCD mice, as demonstrated by significantly increased colocalization of the adaptor ASC (red) with NLRP3 (green) in collagen-activated platelets from SS mice compared with AA controls (Fig. 1A; Pearson’s coefficient, p < 0.0001; Fig. 1B). We next conducted a comprehensive platelet assessment in SCD mice, which revealed substantial morphological (Fig. 1C–E) and functional (Fig. 1F–K) platelet abnormalities. TEM imaging showed pronounced shape distortions in platelets from SS mice (Fig. 1C), along with a non-significant reduction in α-granules (p = 0.1392; Fig. 1D) and a significant reduction in dense granules (p < 0.0001; Fig. 1E) in response to collagen, compared with platelets from AA mice (both vehicle-treated). No significant differences in α-granules or dense granules were observed in resting, non-activated platelets (data not shown). Functionally, collagen-induced platelet P-selectin expression (α-granule-derived; p < 0.0001; Fig. 1F), platelet ATP secretion (dense granule-derived; p < 0.0001; Fig. 1G), platelet spreading (p = 0.0003; Fig. 1H and I), platelet aggregation (p < 0.0001; Fig. 1J), and in vitro thrombus formation (p < 0.0001; Fig. 1K) were significantly increased in isolated platelets or blood from vehicle-treated SS mice compared with that from vehicle-treated AA mice.

Compared to vehicle control, treatment of SS mice with the NLRP3 inhibitor MCC950 reversed these platelet abnormalities, significantly increasing intraplatelet α-granules (p = 0.0223; Fig. 1D) and dense granules (p = 0.0199; Fig. 1E), while significantly reducing platelet P-selectin expression (p < 0.0001; Fig. 1F), platelet ATP secretion (p < 0.0001; Fig. 1G), platelet spreading (p = 0.0004; Fig. 1H and I), platelet aggregation (p < 0.0001; Fig. 1J), and in vitro thrombus formation (p < 0.0001; Fig. 1K). In AA mice, MCC950 treatment had a significant inhibitory effect on platelet aggregation (p = 0.0153; Fig. 1J), but not on any of the other platelet readouts (Fig. 1D, E, F, G, H, I, K).

BTK inhibition interferes with platelet hyperreactivity in sickle cell mice in an NLRP3-dependent manner

3.2.

We next assessed the effect of the BTK inhibitor ibrutinib on platelet hyperreactivity in SCD mice and the role of the NLRP3 inflammasome in mediating this effect. Compared to vehicle, ibrutinib treatment significantly (p < 0.0001) reduced NLRP3/ASC colocalization in collagen-activated SS but not AA platelets (Fig. 2A and B), suggesting interference with NLRP3 inflammasome activation in SS platelets. In SS mice, compared to vehicle, ibrutinib significantly reduced upregulated platelet P-selectin expression (p = 0.0026; Fig. 2C), platelet ATP secretion (p < 0.0001; Fig. 2D), platelet spreading (p < 0.0001; Fig. 2E and F), platelet aggregation (p < 0.0001; Fig. 2G), and in vitro thrombus formation (p = 0.0001; Fig. 2H) in response to collagen. In AA mice, ibrutinib significantly decreased platelet ATP secretion (p = 0.0223; Fig. 2D) and aggregation (p = 0.0007; Fig. 2G) but had no significant effect in the other platelet function assays (Fig. 2C, E, F, H).

To assess whether the NLRP3 inflammasome contributes to ibrutinib’s inhibitory effect on platelet hyperreactivity in SCD mice, we incubated platelet or blood samples from ibrutinib-treated mice with the NLRP3 activator nigericin prior to collagen exposure. In samples from SS mice, nigericin partially reversed the inhibitory effects of ibrutinib, significantly increasing platelet P-selectin expression (p = 0.0034; Fig. 2C), platelet ATP secretion (p < 0.0001; Fig. 2D), platelet spreading (p = 0.0039; Fig. 2E and F), platelet aggregation (p < 0.0001; Fig. 2G), and in vitro thrombus formation (p = 0.0004; Fig. 2H). Nigericin had no significant effect in vehicle-treated SS mice (Fig. 2C–H). In ibrutinib-treated AA mice, nigericin significantly increased platelet P-selectin expression (p = 0.0148; Fig. 2C), platelet ATP secretion (p = 0.0206; Fig. 2D), and platelet aggregation (p = 0.0023; Fig. 2G), but had no significant effect on platelet spreading (Fig. 2E and F) and in vitro thrombus formation (Fig. 2H). In vehicle-treated AA mice, nigericin increased platelet P-selectin expression (p = 0.0003; Fig. 2C), platelet ATP secretion (p = 0.0052; Fig. 2D), platelet aggregation (p = 0.0189; Fig. 2G), and in vitro thrombus formation (p = 0.0269; Fig. 2H), but had no significant effect on platelet spreading (Fig. 2E and F).

Discussion

In this study, we show that the NLRP3 inflammasome is a critical driver of platelet hyperreactivity in SCD mice. Platelet secretion, spreading, aggregation, and in vitro thrombus formation in response to collagen were markedly increased in SCD mice and were suppressed by the NLRP3 inhibitor MCC950. Inhibition of BTK with ibrutinib decreased platelet hyperreactivity in an NLRP3-dependent manner, as the NLRP3 activator nigericin partially reversed the inhibitory effects of ibrutinib.

Circulating platelets are activated in patients with SCD [1,2], with similar findings reported in SCD mouse models [23,24]. Targeting platelet activation has therefore been explored as a therapeutic strategy, but clinical outcomes have been mixed. While crizanlizumab, a humanized monoclonal antibody against P-selectin, initially showed benefit in reducing pain crises in patients with SCD [25], this effect was not confirmed in a subsequent phase 3 trial [26]. Likewise, the P2Y12 receptor antagonists prasugrel [27] and ticagrelor [28] failed to significantly reduce vaso-occlusive events in pediatric patients. As platelet activation in SCD is driven by complex inflammatory mechanisms [3, 29], targeting P-selectin or purinergic signaling alone might therefore not be sufficient to fully suppress platelet hyperreactivity in this disease. This highlights the need to identify critical inflammatory drivers of platelet activation in SCD. Our findings position the NLRP3 inflammasome as one such driver of platelet hyperreactivity and support NLRP3 inflammasome inhibition as a novel antiplatelet approach that may prove beneficial in SCD.

Despite growing interest in NLRP3 inflammasome suppression, the clinical development of direct NLRP3 inhibitors has been hampered by hepatotoxicity and other adverse effects observed in early clinical trials [12]. Furthermore, NLRP3 activation is regulated by diverse upstream signals, including potassium efflux, lysosomal disruption, endoplasmic reticulum stress, and reactive oxygen species, and it is difficult to selectively inhibit NLRP3 in platelets without compromising its essential role in host defense. We and others have demonstrated that BTK is a critical regulator of NLRP3 inflammasome activation through NLRP3 tyrosine phosphorylation, oligomerization, and ASC polymerization, suggesting that BTK inhibition may represent a feasible strategy to modulate NLRP3 activity [13–16]. We previously reported that the BTK inhibitor ibrutinib interferes with platelet aggregation in SCD mice, although it remained unclear whether this effect was mediated through the NLRP3 inflammasome [7,8]. Here, we demonstrate that ibrutinib suppresses platelet secretion, spreading, aggregation, and in vitro thrombus formation in SCD mice, and that the inhibitory effect of ibrutinib observed across all platelet function readouts was partially reversed by the NLRP3 activator nigericin, supporting an NLRP3-dependent mechanism. The therapeutic potential of BTK inhibition in SCD is currently being evaluated in a phase 3 randomized trial using the next-generation BTK inhibitor rilzabrutinib (NCT06975865) [30]. Given that BTK inhibitors are associated with bleeding events and other systemic adverse effects, careful definition of the therapeutic window and mechanistic targets of BTK inhibition will be essential as evaluation of their therapeutic benefit in the context of SCD continues.

Our findings provide the foundation for further mechanistic and translational studies evaluating NLRP3 inflammasome inhibition via BTK inhibitors as a novel antiplatelet strategy in SCD.