467. Lentiviral Vector (LV) Non-Coding Elements That Impart High Titers and Expression to β-Globin (hβ) Cassettes

Abstract

Top of pageAbstract Thalassemias are the most common monogenic defects that result from absent/reduced h|[beta]| in RBC. However, the complexities of the h|[beta]| gene and its regulatory elements have been significant obstacles for optimization of gene therapy approaches using |[gamma]|-oncoretroviral vectors. Recently, LV have been shown to stably transfer and express h|[beta]| at high levels. The goal of our study was to identify non-coding elements in LV required for this effect. A series of LV carrying different elements from the gag and env fragments were made. All vectors encoded h|[beta]| gene driven by |[beta]|-regulatory elements (|[beta]|-promoter, HS234 and |[beta]| 3' enhancer), contained the same 5' and 3' SIN-LTR: sBG-0 was a gutted LV vector containing only the packaging (|[Psi]| region including 40bp of the 5' gag).In sBG-1 the central polypurine tract (cPPT) element was added to sBG-0.In sBG-2, the rev response element (RRE) was added to sBG-1. In sBG-3 and sBG-4, the wt or the mutated splice acceptor (SA) in the env sequence was added to sBG-2, respectively. In sBG-5, |[sim]|600bp of gag and RRE was added to sBG-1. In sBG-6, the |[sim]|400bp of gag was added to sBG-3. All vectors were packaged using the standard transient co-transfection system in 293T cells, concentrated 1400-fold and assayed in differentiated mouse erythroleukemia (MEL) cells. The level of h|[beta]| expression in MEL cells was analyzed by RNase protection assay (RPA) and FACS analysis with 3 independent viral lots. sBG-6 had the highest titers (6|[times]|10E8 TU/mL) while the gutted LV vector sBG-0 and sBG-1 had very poor titers (Fig. 1A). Addition of RRE (sBG-2) increased the titers very significantly. Addition of the env fragment (sBG-3) further increased titers nearly 2-3 fold. However, when the SA in the env gene was mutated (sBG-4), the titers of sBG-4 vectors dropped to levels lower than sBG-2, despite sBG-3 and sBG-4 containing RRE, suggesting that some contribution of titers by the env fragment in sBG-3 reside in the SA, and its mutation has a negative effect on RRE. Replacement of a |[sim]|600bp of gag in place of the env (+SA) in sBG-5 was sufficient to improve titers to 84% of sBG-6. These results were confirmed at the RNA level (RPA in Fig. 1B). When probed with an LCR fragment, Northern blot analysis showed stable transmission of full-length LV mRNA in 293T cells, with varying ratio of spliced and unspliced LTR transcripts in different vectors. The mean fluorescence intensity of h|[beta]| expression in differentiated MEL cells was similar with all vectors, suggesting that upon integration, viral cis elements probably play a minimal role in improving expression of the h|[beta]| transcripts, although this is under investigation; as is RNA export and stable transmission from these vectors. In conclusion, while the rev/RRE in HIV is likely the most important element imparting high titers to h|[beta]| cassettes, the env SA and regions of gag appear to contribute to the titers from SIN-LV.