Attenuated Protein Synthesis Drives the Hematopoietic Defects in Shwachman-Diamond Syndrome

Shwachman-Diamond syndrome (SDS) is an autosomal recessive disorder characterized by bone marrow failure and a striking propensity to develop poor prognosis myelodysplastic syndrome and acute myeloid leukemia. In 90 % of cases the disease is caused by biallelic mutations in the gene encoding SBDS. We have shown previously that SBDS is a cytoplasmic ribosome assembly factor that catalyzes the release of the eukaryotic initiation factor 6 (eIF6) from the subunit joining interface of 60S ribosomal subunit (Menne et al, 2007; Finch et al, 2011). Deficiency of SBDS therefore results in aberrant retention of eIF6 on the 60S subunits that in turn perturbs ribosomal subunit joining and the formation of translation-competent 80S ribosomes. However, the mechanism linking defective ribosome assembly to marrow failure and leukemia in SDS remain poorly understood.

Lack of viable mouse models presents a barrier to progress in understanding SDS disease pathophysiology and to evaluate novel therapies. We hypothesized that induced overexpression of eIF6 would mimic the consequences of SBDS deficiency by reducing the cytoplasmic pool of free 60S subunits and impairing translation. To test this hypothesis we have generated a novel transgenic eIF6 mouse model for SDS using KH2 embryonic stem cells that constitutively express the M2-reverse tetracycline transactivator at the Rosa26 locus with the EIF6 gene targeted downstream of the Col1a1 locus. This strategy permits systemic doxycycline-inducible and graded overexpression of eIF6 through control of the transgene copy number.

We have validated that eIF6 overexpression promotes an increase in eIF6-bound cytoplasmic 60S subunits with a concomitant reduction in 80S ribosomes and polysomes in c-kit+ hematopoietic progenitor cells isolated from the transgenic eIF6 mice, thereby recapitulating the ribosomal subunit joining defect observed in patients with SDS. In vitro, the hematopoietic progenitor cells exhibit a strict eIF6 dose-dependent expansion defect. In vivo, mice with graded eIF6 overexpression are viable but develop macrocytic anemia with reticulocytopenia, thrombocytosis and mild leukopenia. Bone marrow transfer experiments demonstrate that the phenotype is autonomous to the hematopoietic system. Longitudinal phenotypic analyses in primary and transplanted animals are ongoing. Flow cytometric analysis of the bone marrow from transgenic eIF6 mice reveals a significant increase in the frequencies of preCFU-E and CFU-E erythroid progenitor cells and erythroblasts, but a significant reduction in the frequency of reticulocytes. Furthermore, we observe a striking accumulation of abnormal orthochromatic erythroblast-like cells that appear to have failed to enucleate, comprising approximately 1.5 % of the total bone marrow cells. Amnis ImageStream analysis, which combines flow cytometry with fluorescent microscopy, reveals a significant decrease in the frequency of erythroblasts that are able to complete the enucleation process. To address the underlying mechanism, we hypothesized that by impairing the formation of translation-competent 80S ribosomes, eIF6 overexpression would reduce the global rate of protein synthesis. Indeed, O-propargyl-puromycin incorporation assays established that the erythroblasts from the transgenic eIF6 mice have an approximately 3-fold reduction in global protein synthesis rate. Furthermore, our preliminary data suggest that the erythroid phenotype is p53-independent. Finally, erythroblasts from the transgenic eIF6 mice show a significant increase in levels of reactive oxygen species, but the functional significance of this finding remains unclear.

We conclude that reduced rates of global translation drive defective hematopoiesis in the transgenic eIF6 mice. Importantly, eIF6 overexpression in vivo phenocopies SBDS depletion in human CD34+ cells (Sen et al, 2011). Together with the recent discovery of DNAJC21 (the human homologue of the 60S ribosomal assembly factor JJJ1 in yeast) as an SDS disease gene, our data support the hypothesis that deregulated cytoplasmic 60S subunit maturation and reduced translation are the primary drivers of the hematopoietic defect in SDS. Our viable transgenic eIF6 mouse model provides a unique tool to further dissect the mechanisms that underlie bone marrow failure and malignant transformation in SDS and for the development of novel therapeutics.