Myelofibrosis (MF) is characterized by clonal hematopoiesis, bone marrow (BM) fibrosis, and dysplastic megakaryocyte (MK) hyperplasia. Recently, murine studies have challenged the traditional view of MKs as a homogeneous population, instead identifying MK subtypes with distinct functional roles (PMID:34115843). Comprehensive single-cell(sc) transcriptional characterization of human MKs has been hindered by technical and biological challenges with profiling being limited to fewer than 1000 MKs. In MF dysplastic MKs accumulate abnormally in the BM, yet their cellular and molecular heterogeneity remains poorly characterized largely due to the inability to isolate significant numbers of MKs from fibrotic BMs. To gain access to large numbers of human MKs, we have utilized a two-phase culture system (PMID:40643151) that generatesrobust numbers of MKs from MF patients. This approach allowed us to perform scRNA sequencing and provided the first documentation of MF MK cellular and transcriptional heterogeneity. Single-cell capture was performed using the BD Rhapsody platform and a cDNA library was prepared using their Enhanced Cartridge Reagent Kit V3, WTA Amplification Kit; and sequenced using the Illumina Novaseq X plus platform. Scanpy, Scvi-tools (v1.3.0) were used for quality control filtering, integration, and downstream analysis. Fast Gene Set Enrichment Analysis (fgsea) was performed using the MSigDB. Flow cytometric analysis demonstrated robust ex vivo generation of MF MKs (76.18 ± 18.59% CD41+) by day 13 of culture. scRNA sequencing of 62,847 total cells from these cultures identified 50,219 MK transcriptomes (MF1: 21,299; MF2: 18,103; MF3: 10,817). Unsupervised clustering of the transcriptomes revealed 22 distinct clusters, which segregated into six functionally discrete MK subpopulations and one non-MK population. MKs were identified using a panel of 45 human MK transcripts (ITGA2B, FLI1, MYH9, MPIG6B, TGFβ, MEIS1 etc). While these transcripts were variably expressed across clusters, their consistent enrichment enabled the classification of MK subsets and distinction from non-MK cells. Furthermore, fgsea of the clusters revealed groups of pathways that were specifically up or downregulated reflecting the function of the proposed MF MK subtype. The following 6 subtypes of MF MKs were identified: 1: Endomitotic MKs(23.8%) Characterized by high expression of cell cycle regulators (STMN1, AURKA/B, MCM2-7, MKI67, TOP2A, CCNB1, UBE2C) and condensin complex components. They represent MKs undergoing endoreduplication. 2: Platelet-Generating MKs(22.6%) Highly express transcripts associated with mature MKs primed for platelet production and homeostasis, including MPIG6B, ITGA2B, GP1BA, TUBB1, PF4, PPBP, MYH9. 3: HSC Niche-Supporting MKs(16.6%) Overexpress transcripts that serve to regulate the MF hematopoietic stem cell(HSC) niche (ANGPT1, MEF2C, MMRN1) and promote fibrosis (PFN1, LTBP1). 4: Inflammatory MKs(3.4%) Express transcripts for MHC class II molecules -(HLA-DRA, HLA-DRB1), antigen presentation components (CD74), inflammatory mediators (S100A8/A9/A10/A11, CHI3L1, LGALS1), and immune signaling genes (TYROBP, LSP1, CD53). These MKs likely play role in T-cell activation and cytokine production. Fgsea revealed exclusive upregulation of inflammatory pathways in this subset. 5: Translating MKs(9.8%) Express transcripts for ribosomal protein (RPL/RPS families) and translation initiation factors (EIF genes). They represent a metabolically active, transitional state that contributes to MK cytoplasmic maturation. 6: MK Progenitor Cells(MK-PCs)(3.8%) Express stem/progenitor transcripts(CD34, MYB, RBM8A, RUNX1, ETV6, MECOM) and MK transcripts indicating that they include MK biased HSC/MK-PCs. Non MKs(20.0%): Lack expression of MK-defining transcripts but express other lineage markers such as CD3E, PAX5, and MS4A1 indicating that they include lymphocytes and mast cells. Our data provides for the first robust characterization of human MKs in MF patients. Single cell analysis led to clustering and identification of six distinct MK types and a non-MK population. The unique distribution of MK subtypes among the cultures with cells from 3 different patients suggests that transcriptional diversity of MK subtypes likely contributes to the diverse clinical phenotypes of MF patients.

This content is only available as a PDF.
2025
Sign in via your Institution