Lenalidomide Induces Lipid Raft Assembly to Enhance Erythropoietin Receptor Signaling in Myelodysplastic Syndrome Progenitors

Erythropoietin receptor (EpoR) signaling is impaired in patients with Myelodysplastic Syndromes (MDS) despite appropriate growth factor production and cellular receptor display. We previously reported that EpoR signaling is dependent upon receptor localization within membrane lipid raft microdomains, and that disruption of raft integrity abolishes signaling capacity (McGraw KL, et al. PLoS One 2012). Here, we show that MDS erythroid progenitors display markedly diminished raft assembly (p=0.005) and smaller raft aggregates (p=0.023) compared to normal controls. Because lenalidomide triggers raft coalescence in T-lymphocytes to promote immune synapse formation, we assessed the effects of lenalidomide on raft assembly in MDS erythroid precursors and UT7 cells. Lipid rafts were isolated from UT7 cells using ultracentrifugation and identified by GM-1 dot blot and Lyn kinase western blot. Lenalidomide rapidly induced lipid raft formation in UT7 cells which was confirmed by confocal microscopy visualization of GM-1 fluorescence. Lenalidomide also significantly induced lipid raft formation in pooled MDS erythroid progenitors (CD71⁺, cKit⁺) from 11 patients [mean raft size, control (n=569) vs. lenalidomide treatment (n=659), p<0.001], with no significant change observed in pooled erythroids from 3 normal donors (n= 327 for control and n=365 for lenalidomide treated, p=0.37). Interestingly, lipid rafts were significantly larger in erythroid progenitors from patients who responded (n=5) to lenalidomide treatment compared to non-responders (n=3) (75.52 ±13.68 vs. 35.85 ±8.56, p=0.02). Although lenalidomide increased raft size in erythroid progenitors from both responders (p=0.0007) and non-responders (p=0.013), mean raft size was greater in erythroid precursors from responding patients after treatment (p=0.11). Increased raft aggregation after lenalidomide treatment was accompanied by EpoR recruitment into raft fractions together with STAT5, JAK2, and Lyn kinase, whereas the JAK2 phosphatase, CD45, a negative regulator of EpoR signaling, was displaced from raft fractions. Incubation with lenalidomide prior to Epo stimulation enhanced both JAK2 and STAT5 phosphorylation in UT7 cells and primary MDS erythroid precursors. Bone marrow specimens from 12 non-del(5q) IPSS lower risk, lenalidomide naive MDS patients were analyzed by flow cytometry to compare changes in STAT5 phosphorylation in response to Epo stimulation in the presence or absence of lenalidomide. We found a 79.1% mean increase in p-STAT5 mean fluorescence intensity (MFI 95th percentile) in CD45^dim, CD71^high, Gly^low erythroid precursors in 7 of the 12 patient specimens following lenalidomide exposure. Furthermore, increased STAT5 phosphorylation was accompanied by increased DNA binding of the transcription factor in UT7 cells, and improved erythroid colony forming capacity in both UT7 and primary MDS bone marrow cells. Raft induction was associated with F-actin polymerization that was blocked by Rho kinase inhibition and confirmed by lipid raft isolation followed by dot blot with western blot and confocal microscopy. These data provide new insight into abnormalities in the EpoR signaling platform that underlie impaired Epo responsiveness in MDS erythroid precursors. Our findings that deficient raft integrity impairs EpoR signaling provides a novel strategy to enhance EpoR signal fidelity in non-del(5q) MDS. These data also warrant investigation in a larger data set to determine whether lipid raft size may be a predictive biomarker for lenalidomide response.