Abstract
Abstract 3296
Genetic modification of hematopoietic stem cells (HSCs) has the potential to benefit acquired and congenital hematological disorders. Despite the use of so-called “tissue-specific” promoters to drive expression of the desired transgene, off-target (and consequent deleterious) effects have been observed. MicroRNAs (miRNAs) are important regulators of gene expression. They associate with Argonaute proteins and most typically target 3'UTRs, where complementary base-pairing results in repressed gene expression via RNA decay and translation inhibition. Most miRNAs are ubiquitously expressed, and although some are claimed to be “tissue specific,” such claims have generally not been rigorously validated. The long-term goal of this work is identifying “cell preferential” miRNA expression that could be exploited in expression vectors to minimize off-target transgene expression in HSCs.
Initially, total RNA was extracted with Trizol from the megakaryocyte and T-lymphocyte cell lines, Meg-01 and Jurkat, and miRNAs were profiled by Nanostring technology (Nanostring Technologies, Denver, CO). MiR-495 was determined to be highly expressed in Meg-01 and very low in Jurkat cells. A luciferase reporter construct was generated with four canonical binding sites for miR-495 in the 3'UTR and transfected into both cell lines. Compared to control vector without miR-495 binding sites, luciferase expression showed a 50% reduction in Meg-01 cells, but no knock down in Jurkat cells. These experiments indicated that different levels of endogenous miRNA levels can regulate transgene expression through a novel design in the 3'UTR. We next turned our attention to human hematopoietic cells. We reasoned that the long-term goal of minimal off-target transgene expression in HSCs would require knowledge of miRNAs that had little or no detectable expression (“selectively reduced [SR]”) in one cell type and were highly expressed in other cell types. In this manner, the transgene expression would be dampened only in the non-target cells. As a surrogate for bone marrow progenitors and as proof of principle, we used primary cells in normal human peripheral blood. T-cells, B-cells, platelets and granulocytes were purified by density centrifugation followed by immunoselection from five healthy human donors. Flow cytometry using membrane specific markers demonstrate >97% purity of each specific cell preparation. Total RNA was extracted and miRNAs were profiled as above. First, we identified 277 miRNAs that were differentially expressed between any pair of cell types (p-value<0.05 by ANOVA). Second, we performed ranked pair-wise comparisons across all cell types to determine SR miRNAs. This analysis revealed 5 platelet SR-miRNAs, 6 B-cell SR-miRNAs, 2 T-cell SR-miRNAs and 4 granulocyte SR-miRNAs. Lastly, we considered which of these 17 SR-miRNAs would be the best single SR-miRNA within and across cell types. SR-miRNAs were normalized to let-7b, a miRNA we determined to be equivalently expressed across all cell types, and hence, an ideal normalizer. Lineage-specific SR-miRNAs were selected based on extremely low expression in only one cell type and highest fold change of expression compared to the other cell types. The best SR-miRNAs were miR-29b (SR in platelets), miR-125a-5p (SR in B-cells) and miR-146a (SR in granulocytes). The SR expression levels of these 3 miRNAs were validated by qRT-PCR. Our analysis identified no good SR-miRNAs in T-cells. On-going experiments are testing the selective effects of the SR miRNAs in lentiviral vector infection of cord blood CD34+ cells differentiated along specific lineages. In summary, we have demonstrated in hematopoietic cell lines that SR endogenous miRNAs can regulate the expression of transgenes via tandem arrangement of their target sites in the 3'UTR. Additionally, we have identified miRNAs that are specifically expressed at a very low level in one blood cell type and at high levels in other cell types. These miRNAs could potentially be utilized as new biological tools in gene therapy for hematological disorders to restrict transgene expression and avoid the negative consequences of off-target expression.
Disclosures:
No relevant conflicts of interest to declare.
Sources of off-target expression from recombinase-dependent AAV vectors and mitigation with cross-over insensitive ATG-out vectors
