Expression of HMGA2 in Blood Cells from Patients with Paroxysmal Nocturnal Hemoglobinuria.

Paroxysmal nocturnal hemoglobinuria (PNH) is caused by a somatic mutation of PIG-A gene in one or few hematopoietic stem cells and subsequent clonal expansion of mutant stem cells that leads to development of symptoms. It is known that PIG-A mutation is insufficient to account for the clonal expansion required for clinical manifestation of PNH. We are proposing a 3-step model of PNH pathogenesis. Step 1 involves the generation of a GPI-deficient hematopoietic stem cell by somatic mutation of the PIG-A gene. Step 2 involves the immunological selection of GPI-deficient hematopoietic stem cells. Based on the close association of PNH with aplastic anemia, it has been suggested that the selection pressure is immune mediated. However, in spite that over 60% of patients with aplastic anemia have subclinical population of GPI-deficient hematopoietic cells at diagnosis, only 10% develop clinical PNH, suggesting that step-1 and 2 are insufficient to cause PNH. Under immune mediated selection pressure, GPI-deficient cells not only survive, but also proliferate much more frequently than usual to compensate for anemia. This elevated proliferation rate increases the chance that additional genetic mutations are acquired, in turn leads to Step 3. Step 3 involves a second somatic mutation that bestows on PIG-A mutant stem cell a proliferative phenotype. According to this hypothesis, we searched for the candidate gene for Step 3. We reported 2 patients with PNH whose PIG-A mutant cells had an acquired rearrangement of chromosome12, making the break within the 3’ untranslated region in HMGA2. This gene encodes an architectural transcription factor which is deregulated in many benign mesenchymal tumors (Blood. 2006 vol.108 no.13, p4232). Recently, many reports show that truncation of 3’ untranslated region of HMGA2 disrupts binding of miRNA, let-7, which regulates both transcription and translation of HMGA2. In fact, these two PNH patients with chromosomal abnormalities had ectopic expression of HMGA2 in the bone marrow. Based on these, we consider HMGA2 as a candidate gene, ectopic expression of which causes proliferation. We have established the method for stable isolation of mRNA and miRNA from blood and bone marrow cells from PNH patients and analyzed the expression of HMGA2 and let-7 by quantitative RT-PCR. We have analyzed the peripheral blood from 8 healthy volunteers and 12 PNH patients. The samples from patients had significantly higher expression of HMGA2 than those from normal volunteers (relative mRNA expression, 4.8±2.4 vs 1.3±0.3, p<0.05). We analyzed the genomic sequence of three patients including one who has highest HMGA2 expression and found no mutation in 3’ untranslated region. We also analyzed the expression of miRNA and found significantly lower expression of let7b and c in patients. Surprisingly, truncated form without 3’ untranslated region is predominantly expressed in patients. There maybe deregulation of alternative splicing of HMGA2 gene in patients, which needs further investigations. We are now analyzing more PNH samples including bone marrow, where proliferation of stem cells takes place, to investigate whether high expression of HMGA2 contributes to the pathogenesis of PNH. In addition we are going to analyze whether high expression of HMGA2 causes the clonal expansion of PNH cells using PNH mouse model.