An animal model for myelofibrosis

Idiopathic myelofibrosis (IM) is an acquired, insidious disorder characterized by collagen deposition in the bone marrow, splenomegaly, and extramedullary hematopoiesis. Marrow fibrosis, the hallmark histologic feature, is believed to represent a reaction toward underlying clonal proliferation of hematopoietic cells. Indeed, progressive accumulation of some lineages, especially atypical megakaryocytes, has long been recognized in IM, and other syndromes of megakaryocyte hyperproliferation in man and mice are frequently associated with secondary fibrosis. Although it is often assumed that megakaryocyte products induce local collagen deposits, the molecular basis of IM is unknown and is likely to be multifactorial. Moreover, there is considerable heterogeneity in the extent and speed with which myelofibrosis develops in response to diverse forms of megakaryocyte hyperplasia.

One mouse model of megakaryocytosis results from deficiency of the transcription factor GATA1. Vannucchi and colleagues (page 1123) now extend observations first hinted at by Takahashi et al (Blood. 1998;92:434-442) that, with age, GATA1-deficient mice develop a clinical disorder that resembles IM. The phenotype is variably severe and affected mice show no evidence for the most fearsome complication of human IM, transformation to acute leukemia. This accurately reflects the fundamental difference in underlying pathophysiology between the human disease and the mouse model: the root cause of IM is clonal cell proliferation, whereas GATA1-deficient mice principally model the late sequelae, presumably, of nonclonal megakaryocyte excess. These findings highlight the complex effects of GATA1 in regulating megakaryocyte maturation, including the extracellular milieu associated with incomplete differentiation.

Identifying the factors that drive collagen deposition in IM represents an important challenge, and a genetically defined animal model could prove invaluable in focused molecular studies. But the results reported by Vannucchi et al suggest that many candidates may need to be considered before the key molecular defects are known. Regarding the question of pathogenic molecular lesions, it will also be interesting to know whether and how acquired, clonal loss of GATA1 function may contribute to a whole range of megakaryocyte hyperplastic states.