The myelodysplastic syndromes (MDS) are a heterogeneous group of clonal hematopoietic stem cell disorders characterized by ineffective hematopoiesis and cellular dysplasia. Peripheral blood cytopenias and progression to AML are important clinical sequelae of MDS. 10–20% of MDS cases are a consequence of prior treatment with alkylators. The molecular basis of therapy-related MDS (t-MDS) is poorly understood. Point mutations of RAS family members and inactivation of the p53 and p15 tumor suppressor genes by mutation or hypermethylation represent the most frequently reported molecular abnormalities in MDS. Clonal cytogenetic changes, usually involving loss of material from chromosomes 5 and/or 7 are present in >90% of t-MDS cases. These recurring deletions suggest that myeloid tumor suppressor genes may be present in these regions, although their identify has not yet been established. Progress in understanding the genetic basis of human MDS has been hampered by a lack of suitable animal models. To develop a mouse model of t-MDS, we screened 32 inbred strains for susceptibility to t-MDS after treatment with the prototypical alkylating agent, N-nitroso-N-ethylurea (ENU). Mice (n=12 per strain) received two doses of ENU (100mg/kg, IP) or no treatment (n=12 per strain) at 9 and 10 weeks of age. Among the strains tested, SWR/J mice were found to be highly susceptible to myeloid malignancies (MDS/AML). We confirmed this in a second cohort. 10 of 33 (30%) ENU-treated SWR/J mice developed key features of MDS, including anemia (mean Hb=10.9 ±1.1 g/dL, compared to mean Hb=14.0 ±0.3 g/dL in 32 untreated age and sex-matched SWR/J mice, p=0.0006) and erythroid dysplasia (megaloblastic maturation, nuclear budding and blebbing in normoblasts) with a latency of approximately 30 weeks after ENU exposure. There was also evidence of dysplasia in the megakaryocytic lineage, manifested by numerous micromegakaryocytes with unilobar nuclei. Mild dysplastic features were detected rarely in untreated controls from this strain. The t-MDS mice developed significant splenomegaly (mean=0.49 ±0.19 g compared to control mean=0.15 ±0.01 g, p=0.004) with histologic evidence of increased extramedullary hematopoiesis. No significant immunophenotypic differences were detected in bone marrow cells from the t-MDS cases compared to controls. Iron stores were normal with no evidence of ringed sideroblasts. In 2 of the 10 affected mice, MDS evolved to AML, manifested by rapid breathing, circulating myeloid blasts, and leukocytosis (21-621,000 cells/μL). An additional 3 ENU-treated SWR/J mice developed AML without evidence of a preceding MDS phase. All AML cases had a Kit+Gr1+CD34- phenotype with no expression of lymphoid markers. The blasts were myeloperoxidase negative. This mouse model recapitulates many key features of human alkylator-associated t-MDS/AML and should be useful for discovery of mutations involved in the pathogenesis of this syndrome. We are employing array-based comparative genomic hybridization and candidate gene resequencing as tools for mutation discovery in this model. Because of their unique sensitivity to t-MDS, the SWR/J strain will also be particularly useful for identification of germline polymorphisms that affect susceptibility to alkylator-associated t-MDS/AML.

Author notes

Corresponding author