CD133 Marks a Stem Cell Population That Drives Human Primary Myelofibrosis

Primary Myelofibrosis (PMF) is a chronic myeloproliferative neoplasm of alleged stem cell origin. To define the characteristics of malignant PMF stem cells previous studies have focused on the isolation and xenotransplantation of circulating and/or splenic, PMF patient - derived CD34+ stem/progenitor cells. Despite the reported engraftment of CD34+ cell pool, former analyses failed to reproduce major PMF parameters attributed to abnormal human myeloid cell differentiation. The focus of our work was to identify the stem cell population responsible for initiation and development of PMF. To assess the presence of malignant stem cells we analyzed peripheral blood of 30 PMF patients for expression of LT-HSC antigen CD133. To exclude committed myeloid and lymphoid circulating progenitors we performed lineage depletion of PBMCs and isolated CD133+ and/or CD34+ stem cells. Variable CD133+/CD34 ± and CD133-/CD34+ stem cell fractions from 15 PMF patients were assessed for their clonogenic potential in semisolid media and for reproduction of PMF morbidity in a xenotransplantation mouse model. JAK2V617F mutation was used as a genetic marker to track clonal evolution both in vitro and in vivo. In patients' PBMC we detected the consistent presence of a CD133+ population ranging from 0.3% to >30%, which varies in the expression of CD34. CD133 marks overlapping but also distinct cell populations as compared to CD34. To determine the differentiation potential of disparate stem cell populations, CD133+CD34+, CD133-CD34+ and CD133+CD34- cells were subfractionated from PB of 7 patients and assessed for clonogenic capacity. Strikingly, CD133+CD34+ cells exhibited multipotent, bipotent, and unipotent myeloid (including erythroid) and endothelial-like output, whereas CD133-CD34+ cells gave rise predominantly to lineage-restricted granulocyte/monocyte (GM) progenitors or endothelial-like progenitors. Thus, in contrast to circulating CD133-/CD34+ cells in PMF patients, CD133+ cells have a broader and more robust differentiation capacity to all myeloid cell types, including megakaryocyte /erythrocyte lineages. Four JAK2V617F+ patient samples were used to assess mutation burden at the single-cell level from representative colony types. Obtained results demonstrate an early acquisition of JAK2V617F mutation in the primitive CD133+ stem cell compartments, but also revealed an unexpected variability in the genotypes of emerging progenitors. Homozygous JAK2617F/617F progenitors were detectable in all analyzed patient samples, even if a relative low JAK2V617F burden (30%) was determined from the initial pool of CD133+ cells. A disproportionately high incidence of a homozygous JAK2V617F genotype was observed in erythroid progenitors, indicating a skewing for this lineage. Homozygosity was additionally detected in megakaryocytic and multipotent progenitors. In vivo xenotransplantation experiments of various subfractions confirm the origin of multipotent JAK2V617F+ progenitors from CD133+/CD34± stem cells. Transplantation of PMF patient-derived CD133+/CD34± stem cells in immuno-compromised mice induces abnormal human JAK2V617F+ erythroid, megakaryocytic, and monocytic differentiation, splenomegaly, bone marrow/splenic fibrosis and anemia, reproducing many aspects of PMF development. Our data provide the first evidence for the existence of a CD133+ LT-HSC population responsible for development of PMF. It is for the first time demonstrated that JAK2V617F mutation in PMF occurs at the level of a multipotent stem cell, from which all abnormal myeloid cells emanate during evolution of the disease. Identification of the stem cell compartment involved in the triggering and progression of PMF provides the basis to elucidate the nature of the complex niche interactions in myeloproliferative neoplasms.