Hyperhomocysteinemia potentiates megakaryocyte differentiation and thrombopoiesis via GH-PI3K-Akt axis

Abstract

Hyperhomocysteinemia (HHcy) is closely associated with thrombotic diseases such as myocardial infarction and stroke. Enhanced platelet activation was observed in animals and humans with HHcy. However, the influence of HHcy on thrombopoiesis remains largely unknown. Here, we reported increased platelet count (PLT) in mice and zebrafish with HHcy. In hypertensive patients (n = 11,189), higher serum level of total Hcy was observed in participants with PLT ≥ 291 × 10⁹/L (full adjusted β, 0.59; 95% CI 0.14, 1.04). We used single-cell RNA sequencing (scRNA-seq) to characterize the impact of Hcy on transcriptome, cellular heterogeneity, and developmental trajectories of megakaryopoiesis from human umbilical cord blood (hUCB) CD34⁺ cells. Together with in vitro and in vivo analysis, we demonstrated that Hcy promoted megakaryocytes (MKs) differentiation via growth hormone (GH)-PI3K-Akt axis. Moreover, the effect of Hcy on thrombopoiesis is independent of thrombopoietin (TPO) because administration of Hcy also led to a significant increase of PLT in homozygous TPO receptor (Mpl) mutant mice and zebrafish. Administration of melatonin effectively reversed Hcy-induced thrombopoiesis in mice. ScRNA-seq showed that melatonin abolished Hcy-facilitated MK differentiation and maturation, inhibited the activation of GH-PI3K-Akt signaling. Our work reveals a previously unrecognized role of HHcy in thrombopoiesis and provides new insight into the mechanisms by which HHcy confers an increased thrombotic risk.

Trial Registration clinicaltrials.gov Identifier: NCT00794885.

To the Editor,

Elevated total serum level of homocysteine (tHcy), known as hyperhomocysteinemia (HHcy) [1], is a risk factor of cardiovascular disease, ischemic stroke, and venous thromboembolism [2, 3]. If untreated, approximately 50% of patients with severe HHcy due to genetic defects suffer from thrombotic events [4, 5], while even moderate HHcy increase the risk of thrombosis [6]. However, the underlying mechanisms remains unclear. Platelets, anucleated cytoplasmic fragments derived from megakaryocyte (MK) [7, 8], are key protagonists in thrombotic disease [9,10,11]. The aim of the present study is to investigate the impact of HHcy on thrombopoiesis.

Analysis data from 11,189 participants of the China Stroke Primary Prevention Trial [12] revealed that, compared with low PLT (Q1), higher PLT (Q4) was positive correlated with higher tHcy (β = 0.59; 95% CI 0.14,1.04, p = 0.010) (Additional file 1: Tables S1–S3 and Fig. S1). Analogous to human, a significant increase in PLT was detected in both male and female mice with HHcy (Fig. 1A; Additional file 1: Fig. S2, S3). Elevated thrombocytes were also observed in a zebrafish transgenic reporter line Tg(mpl:eGFP)smu4 treated with Hcy (Fig. 1B, C). In a thrombopenia model, the days needed for 50% recovery of PLT from the nadir was shorter in HHcy mice (Fig. 1D). The possible contribution of the spleen to Hcy-increased PLT was excluded since the elevation of PLT remained unaffected in HHcy mice with splenectomy and the half-life of platelet was not affected (Fig. 1E; Additional file 1: Fig. S4). Furthermore, elevation in PLT was observed in Mpl mutant mice and zebrafish with HHcy (Fig. 1F–H), suggesting that HHcy facilitates thrombopoiesis independent of TPO.

Fig. 1

Hcy facilitates MKs differentiation and platelet production. A Peripheral PLT in C57BL/6J mice. Significance according to two-tailed unpaired t test (n = 8). B Representative images for staining and C quantification of mpl:eGFP⁺ cells (green) in zebrafish Tg(mpl:eGFP)smu4 larvae caudal hematopoietic tissue (CHT) region. Scale bars, 50 μm. Significance according to one-way ANOVA with Tukey multiple comparisons test (n = 10). D PLT recovery after platelet depletion by monoclonal rat anti-mouse CD42b antibody (Anti-CD42b MoAb). Significance according to two-tailed unpaired t test (n ≥ 5). E PLT in splenectomized mice. Significance according to two-tailed unpaired t test (n = 8). F PLT in C57BL/6J-Mpl^hlb219/J mice. Significance according to two-tailed unpaired t test (n ≥ 5). G Representative images for staining and H quantification of mpl:eGFP⁺ cells (green) at mpl-mutational zebrafish Tg(mpl:eGFP)smu4;mpl^smu3 larvae CHT region. Scale bars, 50 μm. Significance according to Welch ANOVA test with Dunnett T3 multiple comparisons test (n = 10). I 13 cell clusters were displayed by UMAP. Colors indicate cell types. CMP, common myeloid progenitor; GMP, granulocyte-monocyte progenitor; MEMP, megakaryocyte-erythroid-mast cell progenitor; MK, megakaryocyte. J Bar diagram showing the representative GO biological process terms of MKs subpopulations. K Bar diagram showing the percentage of each MKs subpopulation number to all cells. L Venn diagram visualizing the up-regulated biological processes of each MKs subpopulation in Hcy group compared with control group. M Representative images of PPF detected by phase contrast imaging (left) and confocal microscopy (right). β1-tubulin (green) and DAPI (blue) were stained; Scale bars, 20 μm. N Histogram showing the number of PPF-MKs. Significance according to two-tailed unpaired t test (n = 6). O Representative images and P quantification of CD41⁺ MKs (green) in mice femurs bone marrow by immunofluorescence staining. Scale bars, 50 μm. Significance according to two-tailed unpaired t test (n = 8). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns not significant

To explore the influence of Hcy on the developmental trajectories of MKs, scRNA-seq was performed utilizing cells collected from a hUCB-derived CD34⁺ cell differentiation system. Total 13 clusters including four MK clusters were identified (Fig. 1I; Additional file 1: Fig. S5). Gene ontology (GO) and high-activity regulons (HARs) analysis showed that MK0 represents a less mature population. MK2 is a mature thrombocyte-forming cluster. MK1 serves as a transition state between MK0 and MK2. MK3 highly expresses genes associated with apoptotic signaling (Fig. 1J; Additional file 1: Fig. S6A). Pseudotrajectory analysis revealed a continuous development from MK0 to MK3, which is characterized by a “wave-like” fluctuating gene expression pattern (Additional file 1: Fig. S6B, C). Hcy increased the percentage of MKs, especially MK2 (Fig. 1K). Gene set variation analysis (GSVA) revealed that, compared with control group, biological processes such as megakaryocyte differentiation and megakaryocyte development were significantly up-regulated in MK0-MK2 of Hcy group, indicating Hcy promotes MKs differentiation (Fig. 1L). We further validated the scRNA-seq data. The proplatelet (PPF: β1-tubulin⁺) was enlarged and the amount was increased with Hcy. The number of MKs (CD41⁺) was increased in HHcy mice (Fig. 1M–P). Flow cytometry showed the proportion of MKs (CD41⁺CD42b⁺) was elevated with Hcy (Additional file 1: Fig. S7).

By integrating the scRNA-seq and bulk RNA-seq data from mouse bone marrow MKs (Fig. 2A, B), PI3K-related pathway and response to GH stimulus were the common up-regulated pathways in MKs of Hcy group. Western blotting confirmed Hcy increased the levels of phosphorylated PI3K (p-PI3K) and phosphorylated Akt (p-AKT) (Fig. 2C, D). LY294002, a PI3K inhibitor, blocked Hcy-facilitated MKs differentiation and platelet production (Fig. 2E, F; Additional file 1: Fig. S8). Moreover, when the receptor of GH (GHR) was knocked down in Meg-01 cells by siRNA, Hcy-induced p-PI3K and p-Akt was dramatically attenuated (Additional file 1: Fig. S9). Consistently, no obvious increase in PLT nor the activation of PI3K-Akt signaling was observed in Ghr^−/− mice with HHcy (Fig. 2G–J). These data indicated that Hcy promotes MK differentiation via boosting the GH-PI3K-Akt axis.

Fig. 2

Hcy promotes MKs differentiation via GH-PI3K-Akt axis. A Venn diagram visualizing the elevated Pathway Interaction Database pathways of MKs subpopulations in Hcy group. B Heatmap showing the relative GSVA scores for each gene set based on bulk RNA-seq of BM MKs (n = 4). C and D Western blot analysis of p-PI3K and p-AKT (Ser473) in Meg-01 cells after exposure to Hcy (100 μM) for indicated time period. Total PI3K, AKT and GAPDH were used as loading control. Significance according to one-way ANOVA with LSD multiple comparisons test (n = 3). E Culture-derived MKs and F platelets were analyzed by flow cytometry. Significance according to one-way ANOVA with Tukey multiple comparisons test (n = 3). G PLT in Ghr^−/− mice. Significance according to two-tailed unpaired t test (n = 6). H Quantification of CD41⁺ MKs in femurs bone marrow of Ghr^−/− mice. Significance according to Mann–Whitney test (n = 6). I and J Western blot analysis of p-PI3K and p-AKT (Ser473) in the bone marrow MKs after exposure to Hcy (100 μM) for 30 min. Significance according to one-way ANOVA with Tukey multiple comparisons test (n = 6). K Bar diagram showing the percentage of MK2 to all cells. L Peripheral PLT. Significance according to one-way ANOVA with Tukey multiple comparisons test (n = 8). M Representative images and N quantification of CD41⁺ MKs (green) in femurs bone marrow of mice. Scale bars, 50 μm. Significance according to one-way ANOVA with Tukey multiple comparisons test (n = 8). O Bar chart showing the significantly down-regulated GO-BP terms and P PID pathways in MKs. T values are from the linear model in the limma package. Q–R Western blot analysis the level of p-PI3K and p-AKT (Ser473) in Meg-01 cells after exposure to Hcy (100 μM) with or without MT (1 μM) for 30 min. Significance according to one-way ANOVA with Tukey multiple comparisons test (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns not significant

Finally, scRNA-seq showed Hcy-increased MK2 proportion was reversed by melatonin (MT) (Fig. 2K). MT blockaded Hcy-facilitated thrombopoiesis in mice without decreasing the level of tHcy (Fig. 2L–N; Additional file 1: Fig. S10A). In addition, Hcy-increased thrombocytes in Tg(mpl:eGFP)smu4;mpl^smu3 zebrafish was also reduced by MT (Additional file 1: Fig. S10B, C). As expected, Hcy-elevated MK-associated functions and PI3K-Akt signaling were attenuated by MT (Fig. 2O–R).

Overall, our work demonstrated a role of HHcy in MKs differentiation and characterized the underlying mechanism. Further studies are needed to evaluate the impact of HHcy-promoted thrombocytosis on thrombotic disease.