Knockdown of eIF6 expression improves ribosome association in SDS patient cells but does not rescue the marrow failure phenotype. (A) Lysates from healthy controls versus SDS patients were sedimented through sucrose gradients under native conditions. Exogenous recombinant luciferase protein was added in equal amounts to each fraction before protein extraction to control for potential variations in protein recovery between fractions. Proteins from each fraction were immunoblotted for eIF6, RPL3, and luciferase and quantitated by densitometry. Levels of eIF6 in the 60S fractions were normalized to the 60S subunit protein RPL3 in the 60S fractions as an internal control to allow comparison between different gradient blots and the results of 3 experiments are graphed. The fraction of free eIF6 versus 60S-associated eIF6 was also calculated as a percentage of total eIF6. No statistically significant difference was found between patients versus controls (P = .4). (B) SDS patient-derived marrow stromal cells were transfected with a scrambled RNAi sequence (Scr) or RNAi targeting eIF6. Knockdown of eIF6 in SDS patient cells was confirmed by Western blot with tubulin used as a loading control. Cells were lysed in the presence of 0.25mM MgCl2 to dissociate the ribosomal subunits and assayed in “Ribosomal subunit association assay.” Representative results for cells transfected with the indicated RNAi target sequences are shown. 80S formation relative to the 40S peak was quantitated for 3 independent experiments with cells from SDS patients (FHCRC18, FHCRC30, UW1) of different SBDS genotypes (*P < .05). (C) CD34+ cells were infected with the indicated lentiviral vectors (+) encoding an RNAi sequence targeting SBDS, eIF6, or scrambled control (Scr). Hematopoietic progenitor colony formation in methylcellulose was quantitated in a blinded fashion in triplicate for each condition.
