Myelodysplasia is a bone marrow failure syndrome with a tendency to progress to leukemia. A characteristic finding in blood cell precursors of some individuals with myelodysplasia is the ringed sideroblast, a cell that accumulates large amounts of toxic iron in mitochondria. Recently, the presence of these abnormal mitochondria has been correlated with mutations of the splice factor SF3B1. We plan to investigate the mitochondria of these cells with perturbed SF3B1, aiming to gain insight into mitochondrial causes of myelodysplasia. This may point to new therapies.
Myelodysplasia is a bone marrow failure syndrome with a tendancy to progress to leukemia. Recently mutations in the splice factor SF3B1 were linked to myelodysplasia and sideroblastic anemia. We used a drug meayamycin B (made to Dr K. Koide) to inhibit SF3B1 activity and to mimic the disease in bone marrow precursor cells. We have found that there are changes in iron metabolism that are unique to these precursor cells and that may be important for causing the disease.
Blood cells develop in the bone marrow and are released into the blood. Myelodysplasia refers to a group of disorders characterized by ineffective blood cell formation and a tendency to progress to leukemia, or cancer of the blood. Treatments of myelodysplasia are indirect and unsatisfactory, typically directed at mitigating the secondary effects of the disease and in reducing complications. With the advent of low cost DNA sequencing, scientists have been able to correlate changes in DNA of bone marrow cells with particular types of myelodysplasia. In particular, a high incidence of mutations in a gene, SF3B1, has been correlated with subtypes of myelodysplasia with ringed sideroblasts. Ringed sideroblasts are red blood cell precursors in the bone marrow that fail to develop normally because of iron trapped in the mitochondria. The goal of our project was to mutate or inhibit SF3B1 in immortal cell lines and in cells derived from healthy bone marrow to see if we could generate ringed sideroblasts. If inhibition of SF3B1 led to the formation of ringed sideroblasts, then we would biochemically characterize these cells and provide information that might lead to new therapeutic approaches.
Our initial experiments used an immortal (cancer) cell line (K562) that was derived from a cell with the potential to form red blood cells. These cells are difficult to manipulate genetically, so we employed a chemical inhibitor of SF3B1 called meayamycin. This was a chemical developed by our collaborator Kazunori Koide (University of Pittsburgh). Meayamycin did alter cellular iron metabolism, but not in the way that we hypothesized. The drug treated cells behaved in a way that would limit iron accumulation. Indeed when we measured iron uptake in cells that had been treated with meaymycin, we observed that cells accumulated less iron, not more. This was surprising because this drug had been reported (Visconte et al. 2012, Blood 120: 3173-3186) to trigger the formation of ringed sideroblasts in healthy bone marrow (blood precursor) cells. We attempted to repeat this published observation, but were unsuccessful. Instead the drug proved to be toxic with prolonged exposure and killed most of the bone marrow cells. We presented these results at the American Society for Hematology meeting in December 2014, at which another investigator informed us that he had seen the same results as us and did not see the formation of ringed sideroblasts. We decided to try a molecular genetic approach to inhibit SF3B1 called siRNA.
For the siRNA studies we had to use a different cell line (HeLa) that would more readily take up DNA to permit genetic manipulation. We used gene biomarkers to assess the effects of the siRNA. The siRNA technique was successful in that the level of SF3B1 in the cell was decreased by five-fold. We also observed changes in a gene called ABCB7, depletion of which had previously been shown to cause iron accumulation in mitochondria. Other changes, particularly those concerning genes involved in iron metabolism were the opposite to what we had observed with meaymycin and the K562 cell line. We tested the effect of meayamycin on the HeLa cells and observed the same results that we had obtained with the siRNA approach, indicating that the downstream effects of inhibiting SF3B1 are dependent on cell type.
In summary using two different methods of SF3B1 inhibition and different cell types we have shown that less SF3B1 does lead to altered cellular iron metabolism. However, the changes were complex and dependent on the cell type. Our research suggests that simple loss of function of SF3B1 in bone marrow cells does not cause formation of ringed sideroblasts. However the type of mutations in SF3B1 that have been correlated with the formation of ringed sideroblasts may be "gain of function" mutations. Recently a technique called CRISPR has been developed and we are currently using this to see if we can test this possibility.