YAP1 regulates thrombopoiesis by binding to MYH9 in immune thrombocytopenia Free

Immune thrombocytopenia (ITP) is an acquired hemorrhagic autoimmune disease characterized by severe thrombocytopenia.¹ Despite advances in knowledge, its intricate mechanisms remain inadequately elucidated. The pathogenesis revolving around immune dysregulation, particularly among T cells, B cells, dendritic cells, and myeloid-derived suppressor cells,^1,2 manifested as accelerated platelet destruction and decreased production.³ Autoantibodies trigger abnormal apoptosis of platelets and inhibit megakaryocyte (MK) maturation, hindering platelet release.⁴ Although cytotoxic T lymphocytes (CTLs) mediate direct platelet destruction⁵ and indirectly suppress platelet production by inhibiting MK apoptosis.⁶ A synergistic interplay between humoral and cellular immunity fuels thrombocytopenia, leading to subsequent bleeding manifestations in various tissues such as skin, mucous membranes, and internal organs.⁷ 

Megakaryopoiesis involves the intricate differentiation of MK progenitors to functional platelets from hematopoietic stem cells (HSCs).⁸ A critical step in proplatelet formation (also known as thrombopoiesis) is the migration of MKs from the bone marrow (BM) periosteal niche toward the BM sinusoidal blood vessels.⁹ Once localized in the proximity of marrow sinusoids, MKs extend to form proplatelets through the endothelial barrier.¹⁰ Studies indicate significant reductions in MK maturation and distribution within the BM periosteal niche in ITP, suggesting impaired megakaryopoiesis is critical in the pathogenesis of ITP.¹¹ However, the intricate mechanisms underlying this process during ITP remain poorly understood.

Yes-associated protein 1 (YAP1), identified as a consequence of its interaction with the Src-homology 3 domain of the tyrosine kinase yes,¹² serves as a core regulator in the Hippo signaling pathway.¹³ Once entering the nucleus, YAP1 forms complexes with TEA domain transcription factors to activate the transcription of genes involved in cell proliferation, invasion, and metastasis.¹⁴ Functionally, YAP1 is essential for regulating cytoskeletal tension¹⁵ and GATA-binding protein 1 (GATA1)–mediated tumor suppressor Ras association domain family member 1a, which acts upstream of YAP1, during the transition from pluripotency to differentiation.¹⁶ Although YAP1 has been extensively studied in tumors and hematologic malignancies,^17-19 its role in megakaryopoiesis is inconclusive. Royet al showed YAP1 depletion hindered platelet generation due to mitochondrial defect in MKs,²⁰ whereas Lorthongpanich et al revealed YAP1 upregulation promotes expansion of CD41⁺ immature MKs.²¹ 

Our study disclosed a reduced YAP1 expression in ITP MKs, linked to a defective cytoskeleton. In ITP mice, decreased YAP1 levels correlated with aberrant MK distribution and insufficient thrombopoiesis. Mechanistically, YAP1 was regulated by GATA1, promoting MK maturation. Furthermore, phosphorylated YAP1 (P-YAP1) partnering with myosin heavy chain 9 (MYH9) facilitated cytoskeletal rearrangement for proplatelet formation. Using a YAP1^+/− ITP mouse model and MKs from patients with ITP, we showed that the administration of XMU-MP-1, a YAP1 agonist, restored thrombopoietic function in YAP1^+/− mice, implicating YAP1 as a promising therapeutic target for ITP.

More details on the materials and methods are provided in the supplemental Material, available on the Blood website.

Healthy donors and patients with ITP were recruited at the First Affiliated Hospital of Soochow University. This study was approved by the Medical Ethical Committees of the First Affiliated Hospital of Soochow University. Written informed consent was obtained from each patient and healthy donor in accordance with the Declaration of Helsinki.

All animal protocols were approved by the Soochow University Animal Care and Use Committee. YAP1^+/− mice were generated via clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated 9 (CRISPR/Cas9) technology and IV injected with an anti-platelet monoclonal CD41 antibody at an initial dose of 0.2 mg/kg body weight for 7 days.

CD34⁺ HSCs were isolated from peripheral blood using an immunomagnetic bead cell-sorting system and cultured with the human cytokines thrombopoietin, stem cell factor, and interleukin-3 over a period of 12 days.

The cells were crosslinked with 1% formaldehyde and quenched with glycine for 5 minutes. Pellets were lysed and sonicated after centrifugation. Sheared chromatin was then incubated with the primary antibody overnight, followed by elution and reverse crosslinking at 65°C overnight. Tris-EDTA buffer was added, followed by RNase treatment at 37°C for 30 minutes and proteinase K treatment at 51°C for 1 hour, after which the DNA was subsequently isolated and purified.

HEK293T cells were cultured and transfected with pcDNA3.1-GATA1 and pGL3-YAP1-wild type/mutant type using Lipofectamine 2000. Five hours after transfection, the medium was replenished with a medium containing curcumin. Forty-eight hours after transfection, luciferase activity was measured using the Dual-Luciferase Reporter Assay System.

All data are shown as the mean ± standard error of the mean and were analyzed using PRISM software (GraphPad software version 8.0). Statistical significance was established using the Student t test, as described in the legends, or the 1-way analysis of variance, followed by the pairwise multiple comparison procedures. A P value <.05 indicated statistically significant results.

YAP1, a vital component of the Hippo signaling pathway, governs cell proliferation and differentiation to maintain homeostasis,²² and patients with ITP exhibit decreased platelet production due to MK maturation disorders. To investigate the potential role of YAP1 in ITP, we performed a flow cytometry analysis of MKs from the BM of patients with ITP. Patients with ITP displayed a lower percentage of YAP1⁺ MKs, both overall MKs (ITP 45.29% vs healthy 95.25%; P < .05; Figure 1A) and among mature MKs (ITP 38.77% vs healthy 82.34%; P < .01; Figure 1B). Messenger RNA (mRNA) analysis of ITP (n = 9) and healthy (n = 7) donor MKs showed a 54.87% reduction in YAP1 mRNA expression (P < .0001), consistent with other MK markers (CD41, CD42a, CD42b, and CD61; Figure 1C). Intriguingly, compared with the even distribution in the nucleus and cytoplasm in healthy MKs, YAP1 exhibited cytoplasmic dominance in MKs from patients with ITP (Figure 1D).

To improve our understanding of the reduced expression and abnormal distribution of YAP1 in patients with ITP, we conducted an RNA sequencing analysis of total BM RNA from patients with ITP and healthy controls (supplemental Table 1). Gene ontology enrichment analysis (supplemental Figure 1A) revealed that the differentially expressed genes between the 2 groups were mainly associated with immune response and membrane vesicle formation. Notably, among 7 randomly selected samples, at least 42.86% of patients with ITP showed a reduction in YAP1 expression compared with healthy donors (supplemental Figure 1B). Taken together, these results indicate an indispensable role of YAP1 in MKs from patients with ITP.

To investigate the role of YAP1 in ITP development, we first generated an ITP mouse model (supplemental Figure 2A), as previously described.²³ Consistent with expectations, the expression of YAP1, along with CD41 and CD42a, was reduced in WT MKs after ITP induction (Figure 1E; supplemental Figure 2B-C). Considering that YAP1 knockout is embryonically lethal in mice, next, we used CRISPR/Cas9 technology to generate a heterozygous YAP1 mouse model. After ITP induction, submaximal platelet recovery was observed in YAP1^+/− mice compared with that in WT mice with ITP (Figure 1F). Furthermore, compared with WT mice with ITP, YAP1^+/− mice exhibited a decreased number of total MKs (Figure 1G, bottom left; P < .0001) with reduced localization around the sinusoidal endothelium (Figure 1G, bottom right; P < .0001), suggesting a role for YAP1 in MK migration toward the BM sinusoids.

After single-cell RNA sequencing of BM cells from WT and YAP1^+/− mice after ITP induction (supplemental Figure 2D), 12 cell subpopulations (HSCs, MK-erythroid progenitors, MKs, basophils, plasma dendritic cells, natural killer cells, monomacrophages, macrophages, T cells, monocytes, B cells, and neutrophils) were identified using a t-distributed stochastic neighbor embedding plot based on characteristic markers (Figure 1H; fraction of cells and characteristic marker genes of each subset are presented in supplemental Figures 2E and 3). Notably, the total proportion of MKs was reduced in YAP1^+/− mice after ITP induction, whereas the proportion of MK-erythroid progenitors remained unchanged compared with that in WT mice (supplemental Figure 2G-H). Based on the expression of immature MK markers (GATA2, Stat3, and FOXP1) and mature MK markers (PF4, Mef2c, Cited2, and CD61), we further subdivided the MKs into early MKs and late MKs (supplemental Figure 2F). The abundance of total MKs (CD41⁺), especially late MKs, was reduced in YAP1^+/− mice (in total MKs, YAP1 128 vs WT 239; in early MKs, YAP1 56 vs WT 87; in late MKs, YAP1 72 vs WT 152; supplemental Figure 2I-J). To investigate the distinct functions of these 2 subsets, we analyzed the differential gene expression profiles (supplemental Figure 2K-L). Gene set enrichment analysis revealed that early MKs were involved in immune regulation and antigen presentation, whereas late MKs focused on genes linked to the cytoplasmic membrane, pivotal for thrombopoiesis (Figure 1I; supplemental Table 6). Furthermore, gene ontology analysis of the differential genes between the late MKs from WT and YAP1^+/− mice highlighted disruptions in proplatelet formation in YAP1^+/− MKs (Figure 1J). Taken together, these results indicate that a reduction in YAP1 expression alters the proportion of late MKs and the regulation of thrombopoiesis during ITP development.

Next, we sought to examine whether YAP1 expression facilitates MK maturation using a CD34⁺ HSC–based in vitro induction system (supplemental Figure 4A) and showed that YAP1 expression gradually increased with MK maturation (Figure 2A). Notably, the diameter of MKs (supplemental Figure 4B), CD41/CD42 expression (supplemental Figure 4C), and the proportion of CD42a⁺ MKs and CD41a⁺CD42a⁺ mature MKs (Figure 2C) were remarkably increased by XMU-MP-1 (YAP1 activator) treatment. Conversely, inhibiting YAP1 with peptide 17 (YAP1 inhibitor) showed a decrease in these maturation indicators. To consolidate these findings, we manipulated YAP1 levels through overexpression (by PCDH-YAP1) or knockdown (using YAP1-short hairpin RNA) in CD34⁺ HSCs for MK induction. As expected, YAP1 overexpression in the PCDH-YAP1 group increased CD41a⁺ and CD42a⁺ mature MKs compared with the vehicle control (PCDH group; Figure 2D). In contrast, YAP1 knockdown (YAP1-short hairpin RNA) led to reduced maturation of MKs compared with vehicle controls (Figure 2D). These results collectively indicate that YAP1 contributes to MK maturation.

GATA1, a crucial transcription factor in hematopoietic development,²⁴ plays a pivotal role in MK maturation. GATA1 mutations within the N-terminal zinc finger domain are associated with myeloproliferative disorders²⁵ and acute megakaryoblastic leukemia.²⁶ Embryonic stem cells with decreased GATA1 expression exhibit insufficient expression of megakaryocytic markers,²⁷ underscoring its essential role in the maturation of MKs. Upon investigation, we observed lower GATA1 mRNA levels in the BM of 14 patients with ITP than in those of 15 healthy individuals (Figure 2E; P < .05). To explore the interplay between GATA1 and YAP1, GATA1-overexpression and GATA1-knockdown HEK293T cells (supplemental Figure 4D-E) were generated. Overexpression of GATA1 resulted in the upregulation of YAP1, whereas GATA1 knockdown suppressed YAP1 expression (Figure 2F-I). A luciferase assay demonstrated that transfection with the GATA1 plasmid enhanced YAP1 transcriptional activity (Figure 2J), suggesting that GATA1 directly regulates YAP1 by binding to its promoter. Subsequently, binding sites for GATA1 in the YAP1 promoter were screened using the JASPAR and PROMO databases, and 3 potential binding sites (named sites 1, 2, and 3) were proposed (Figure 2K). Luciferase experiments revealed that site 3 (CTCCAATCTAT) was a specific site for GATA1 binding (Figure 2L). Moreover, the chromatin immunoprecipitation–quantitative polymerase chain reaction assay validated the interaction of binding between GATA1 and YAP1 promoter (supplemental Figure 4F). Collectively, these results suggest GATA1 as a regulator of YAP1 expression via its binding to the YAP1 promoter.

Cytoskeletal regulation in MKs, especially the activation of microtubules and F-actin assembly, is crucial for proplatelet formation. Thus, we examined the potential role of YAP1 in the MK cytoskeletal activation. Pretreatment with XMU-MP-1 promoted actin polymerization and abundant stress fiber formation (Figure 3A, left and middle). However, F-actin presented as an entwined pattern at the periphery of the MKs, with a disrupted assembly of F-actin (Figure 3A, left, yellow) after YAP1 inhibition. As a consequence of defective actin cytoskeletal dynamics, the spreading area was diminished in YAP1-inhibited MKs compared with normal controls (252.9 μm² of peptide 17 vs 470.2 μm² of normal control; P < .01; Figure 3A, right). Moreover, there were no differences in tubulin levels or distribution between the YAP1 upregulation and knockdown groups (Figure 3A, left, green), suggesting that YAP1 is a regulator of actin rather than tubulin dynamics in MKs. Consistently, MKs from healthy donors exhibited well-developed actin stress fibers with larger spreading areas than those from patients with ITP (healthy 644.7 μm² vs ITP 133.2 μm²; P < .0001; Figure 3B). In line with these observations, YAP1-deficient MKs derived from YAP1^+/− mice displayed defective cytoskeletal activation with reduced spreading area (159.3 μm² for YAP1 vs 528.7 μm² for WT; P < .0001; Figure 3C). Taken together, these findings emphasize the critical role of YAP1 in MK maturation, where it functions as a key regulator of actin dynamics during platelet biogenesis.

To explore the target cytoskeletal proteins downstream of YAP1 in MKs, proteins that potentially interacted with YAP1 were identified by mass spectrometry (supplemental Table 3). Notably, MYH9 was validated as a binding protein, and the interactions were confirmed by coimmunoprecipitation (Figure 3D). Confocal microscopy images further demonstrated the colocalization of YAP1 with MYH9 (Figure 3E). Upon YAP1 overexpression, P-YAP1 and MYH9 levels were upregulated (supplemental Figure 5A). Coimmunoprecipitation experiments substantiated the interaction between P-YAP1 and MYH9, verifying that P-YAP1 interacts with MYH9 to activate the downstream cytoskeleton (Figure 3F; supplemental Figure 5B). However, no interaction binding was detected between unphosphorylated YAP1 and MYH9 after P-YAP1 coimmunoprecipitation in MKs (Figure 3G), indicating that the phosphorylated conformation of YAP1 plays a fundamental role in YAP1-MYH9 binding in MKs.

Using ZDOCK combined with Discovery Studio, several potential binding sites were identified between MYH9 and the WW domains of YAP1 (which are universally acknowledged binding sites for interacting proteins²⁸). Subsequently, mutant constructs were generated to determine specific binding sites (supplemental Figure 5C). The absence of WW1 led to a decrease in P-YAP1 levels, whereas WW2 deletion and double WW1-WW2 deletion did not decrease or even increase P-YAP1 levels (Figure 3H). Coimmunoprecipitation pinpointed the WW2 domain as the functional binding site (Figure 3H-I), as WW2 deletion led to an increased level of dysfunctional P-YAP1 with crucial defects in MYH9 binding (Figure 3H-I). Immunofluorescence staining of full-length YAP1 and its mutants for colocalization of MYH9 and YAP1 further confirmed the defective interaction after WW2 deletion (supplemental Figure 5F). In summary, these data suggested that P-YAP1 interacts with MYH9 through its WW2 domain, supporting the potential role of YAP1 in orchestrating actin filament dynamics during thrombopoiesis.

Because F-actin in ITP MKs displayed peripheral localization without filamentous networks, further proplatelet formation was decreased in the ITP group compared with that in healthy individuals (ITP 1.83 vs healthy 14.83; P < .0001; Figure 4A). Meanwhile, YAP1^+/− mice showed reduced proplatelet formation compared with WT (6.8 for YAP1^+/− vs 15.3 for WT; P < .0001; Figure 4B). Moreover, XMU-MP-1 treatment enhanced proplatelet production (Figure 4C, red arrows), with a 52.70% increase in thrombopoiesis (Figure 4D, bottom; P < .01) but a 40.42% decrease upon peptide 17 inhibition (Figure 4D, bottom; P < .05). Collectively, these observations suggested that YAP1 contributes to thrombopoiesis.

We first uncovered direct binding between YAP1 and MYH9, implicated in cytoskeletal activation. Next, we explored the potential role of MYH9 in YAP1-regulated thrombopoiesis by examining MK differentiation (D7, D10, and D12), mobility, and proplatelet formation. Notably, treatment with the specific MYH9 inhibitor, blebbistatin (B), significantly compromised YAP1-stimulated (XMU-MP-1 group) MK differentiation (Figure 4E). Considering the association of MYH9 with F-actin for force generation in cell motility,²⁹ the colocalization of YAP1 and actin was validated through coimmunoprecipitation and confocal imaging (supplemental Figure 5D-E). Thus, we used blebbistatin, the actin polymerization inhibitor cytochalasin D (Cy), and the actin polymerization activator narciclasine (N) in MKs pretreated with the YAP1 activator (XMU-MP-1) or inhibitor (peptide 17). Suppression of blebbistatin and cytochalasin D on YAP1-mediated adhesion and migration was observed, whereas narciclasine rescued MK motility defects induced by YAP1 inhibition (Figure 4F; supplemental Figure 5G-H). Mechanistically, blebbistatin and cytochalasin D impeded actin activation downstream of YAP1 in MKs (Figure 4H-J, yellow), diminishing augmented proplatelet formation. Narciclasine reversed the actin disruption caused by YAP1 inhibition, counteracting defects in proplatelet formation (Figure 4K-L). In summary, YAP1-MYH9 interactions orchestrate actin regulation, driving MK thrombopoiesis.

To investigate the role of YAP1 in platelet recovery, we first conducted BM transplantation (BMT) between WT and YAP1^+/− mice, and ITP was induced after recovery (Figure 5A). Remarkably, WT mice that received YAP1^+/− BM cells exhibited submaximal platelet recovery compared with those that received WT BM cells (Figure 5B). Further analysis uncovered a considerable decline in total MKs and mature MKs around the sinusoids in both groups receiving YAP1^+/− BM, regardless of their genetic background (Figure 5C). Additionally, we isolated MKs from the BM of both WT and YAP1^+/− recipients and showed that reduced YAP1 expression led to impaired cytoskeletal activation, resulting in hindered proplatelet formation (Figure 5D), suggesting a crucial role of YAP1 in regulating the MK cytoskeleton and further thrombopoiesis.

Subsequently, to further elucidate the role of megakaryocytic YAP1 in thrombopoiesis, we established optimized BMT models using mouse HSCs (ckit⁺ HSCs), manipulated early (MKs cultured for 3 days [D3-MKs]) and late (MKs cultured for 8 days [D8-MKs]) stages of MKs (Figure 5E). Distinctly, platelet recovery in WT donors outpaced that of YAP1^+/− donors, with the late-stage MKs (D8-MKs) from WT mice exhibiting the fastest recovery, followed by the early-stage MK (D3-MKs) group and HSC group from WT mice. In contrast, the slowest recovery was observed in the HSC group from YAP1^+/− mice (Figure 5F). This pattern aligns with the typical BM reconstruction timeline, which typically commences on the seventh day after transplantation. At this time point, the transplanted D8-MKs differentiated into mature MKs and actively produced proplatelets. MKs in the early MK group (D3-MKs), as well as in the HSC group, have not yet fully matured, limiting thrombopoiesis. Late MK group (D8-MKs) from WT mice exhibited the highest MK population in the BM, followed by the early MK group (D3-MKs) from WT mice via immunofluorescence. Conversely, the HSC group of YAP1^+/− mice displayed the lowest density of MKs (Figure 5G). Taken together, these findings demonstrate that megakaryocytic YAP1, especially in late MKs, is crucial for the efficient restoration of platelets in ITP mice.

As an important transcription factor upstream of YAP1, GATA1 binds to the promoter of YAP1, regulating its transcription and promoting MK maturation. Moreover, decreased YAP1 and GATA1 expression correlates closely with differentiation defects in ITP MKs, suggesting a critical role of the GATA1-YAP1 axis in megakaryopoiesis. To address this, BMT mouse models using GATA1 knockdown BMs were established, followed by restoring YAP1 expression using the YAP1 activator XMU-MP-1 in GATA1-knockdown mice (Figure 6A; supplemental Figure 6A). These results showed that thrombopoietic defects triggered by GATA1 deficiency were rescued by YAP1 administration. In details, both platelet counts (Figure 6B) and MK distribution (Figure 6C) in the BM were restored after YAP1 treatment, thereby highlighting the pivotal role of YAP1 in mediating the function of GATA1 in megakaryopoiesis and hematopoietic homeostasis.

Given the connection between reduced YAP1 expression and disrupted MK cytoskeleton with impaired thrombopoiesis, we aimed to determine whether the functional defects observed in YAP1^+/− MKs and patients with ITP could be corrected by pharmacological administration of YAP1. Thus, we administered XMU-MP-1 intraperitoneally to YAP1^+/− mice for 3 consecutive days after ITP induction (supplemental Figure 6B). As expected, YAP1 administration restored platelet counts in YAP1^+/− mice after ITP induction to levels comparable with those in WT mice (Figure 6D) and augmented the total MK count, especially around marrow sinusoids (Figure 6E). Moreover, YAP1 injection corrected cytoskeletal defects and enhanced MK spreading (Figure 6F; supplemental Figure 6C-D) in YAP1^+/− mice. We further studied the impact of XMU-MP-1 on thrombopoiesis of MKs from patients with refractory ITP and chronic ITP and revealed that YAP1 administration improved cytoskeletal organization and stress fiber formation after XMU-MP-1 treatment (supplemental Figure 6E). Furthermore, more proplatelets were formed in MKs from refractory ITP and chronic ITP after XMU-MP-1 treatment (Figure 6G), suggesting a fundamental regulatory role of YAP1 in thrombopoiesis. In conclusion, these results indicate that the administration of YAP1 effectively corrects MK cytoskeletal dysregulation resulting from reduced YAP1 expression during ITP development.

Dysregulation of megakaryopoiesis, which can be derived from cell-extrinsic immune stresses and cell-intrinsic regulation in response to immune attack, has been recognized as the pathogenesis of ITP.⁵ Here, we identified a novel mechanistic connection between the hematopoietic regulatory functions of YAP1 and ITP pathogenesis. First, YAP1 expression was reduced in ITP MKs, which delayed MK maturation. Moreover, reduced YAP1 expression weakened the interactions between P-YAP1 and the cytoskeletal protein MYH9, impairing downstream cytoskeletal actin organization and resulting in thrombocytopenia (Figure 6H). Contrary to previous reports implicating a suppressive role of YAP1 in the differentiation and proplatelet formation of the megakaryocytic cell line MEG-01,²¹ our investigation used HSCs to emulate primary megakaryopoiesis and demonstrated that YAP1 promotes thrombopoiesis, thus sidestepping artifacts of cell-specific phenotypes that deviate from authentic MK maturation dynamics and proplatelet production. Divergent findings across various research platforms, methodologies, and focuses expose disparities. By adopting a novel perspective, our findings differ from the conventional focus on proliferation and embrace a deeper exploration of intracellular mechanisms during MK maturation, particularly exploring the interactions between YAP1 and cytoskeletal-related proteins and their roles in orchestrating cytoskeletal rearrangements to facilitate thrombopoiesis.

MYH9 regulates proplatelet formation in MKs by tethering F-actin and affecting its organization.^30,31 Our study unveiled a novel interaction between YAP1 and MYH9, crucial for cytoskeletal regulation. Specifically, we identified the WW2 domain as a precise binding site for the YAP1-MYH9 interaction. However, considering the nonspecificity of Blebbi as an MYH9 inhibitor, MK-specific MYH9 knockout mice can be generated and used to further confirm the YAP1-MYH9 interaction in future studies.

We observed that decreased YAP1 was closely related to defects in differentiation and thrombopoiesis in ITP MKs. Impaired MK maturation and platelet production in response to autoimmune attack have been considered as one of the important pathogenesis mechanisms of ITP.^32,33 However, the precise mechanisms by which immunity diminishes YAP1 expression remain unidentified. To address this, we used anti-CD41 antibody (supplemental Figure 7A-C), BM supernatant from ITP (supplemental Figure 7D-G), and CTLs (supplemental Figure 7H-J) to coculture with MKs and demonstrated that CTLs, instead of humoral immune responses, resulted in YAP1 reduction. Thus, dysregulation of cellular immune responses, leading to downregulation of YAP1, ultimately contributes to persistent MK maturation defects and thrombocytopenia. Our observations across both patients with ITP and animal models consistently link a reduced YAP1 expression to a reduction in peripheral platelet counts. However, further investigation is warranted to unravel the specific mechanisms.

Single-cell RNA sequencing revealed that YAP1 is essential for thrombopoiesis. YAP1 knockdown impeded cytoskeletal activation, resulting in thrombocytopenia, which was further verified in primary MKs and mouse models. We identified 2 distinct MK subsets and demonstrated that a reduction in YAP1 leads to reductions in both early and late MK subpopulations, especially in the late MKs associated with thrombopoiesis, exhibiting a more severe impairment. Given the functional enrichment analysis of these subsets, YAP1 emerges as pivotal not only in thrombopoiesis but also in immunoregulation during ITP pathogenesis. However, several limitations should be noted in our study. First, our observations were obtained on day 11 after ITP establishment, which remains the greatest differences in platelet levels. Collection at additional time points (such as days 7 and 14) may provide more comprehensive insights into the pathogenic process of ITP. Second, bioinformatics analysis of differentially expressed genes was performed using only 2 MK subsets. Further investigations into the various functional MK subsets implicated in YAP1-associated ITP are necessary. Last but not least, the passive ITP model has contributed significantly to elucidating the pathophysiology of ITP but incompletely recapitulates chronic ITP.⁶ Thrombocytopenia correlates with CTL expansion and exacerbates immune damage. Therefore, although the passive ITP model does not directly address CTL-mediated autoimmune responses, thrombocytopenia-induced cellular immune responses indirectly contribute to its pathogenesis.

In conclusion, we demonstrated that YAP1 upregulation via the binding of GATA1 to its promoter promotes MK maturation. P-YAP1 promotes cytoskeletal activation by binding its WW2 domain to MYH9, facilitating thrombopoiesis. Conversely, during ITP development, YAP1 expression is decreased in MKs in response to autoimmune responses, impairing thrombopoiesis and leading to the deterioration of ITP. Our work provides novel insights into the YAP1-mediated thrombopoiesis mechanism involving the cytoskeleton, suggesting potential for the development of diagnostic markers and therapeutic strategies for ITP.

The authors thank the Tissue Specimen Library of the Hematology Department at The First Affiliated Hospital of Soochow University for providing the clinical samples for this study.

This work was supported by grants from the National Key Research and Development Program (2022YFC2502700) (Y.X.), the National Natural Science Foundation of China (82000140 [S.H.], 81730003 and 82020108003 [D.W.], 81870120 and 82070187 [Y.X.], 82170466 [L.Z.]), the Natural Science Foundation of Jiangsu Province (BK20200197) (S.H.), the Social Development Project of Jiangsu Province (BE2019655) (Y.X.), the Jiangsu Province Key R&D Program (BE2019798) (D.W.), the Suzhou Science Project (SKY2021104) (Y.X.), and the Priority Academic Program Development of Jiangsu Higher Education Institutions (D.W.).

Contribution: S.H., Y. Liu, X.Z., X.W., M.C., and Y. Li performed the experiments and analyzed the data; S.H., Y. Liu, Y.X., and L.Z. designed the research, interpreted the results, and wrote the manuscript; J.Y. and Y. Li collected the samples; C.R., L.Z., Y.X., and D.W. provided overall guidance; and all authors reviewed the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Depei Wu, The First Affiliated Hospital of Soochow University, Soochow University, Suzhou 215123; email: wudepei@suda.edu.cn; and Yang Xu, The First Affiliated Hospital of Soochow University, Soochow University, Suzhou 215123; email: yangxu@suda.edu.cn.