The development of HIV-1 derived gene transfer technology: optimisation of vector safety, processing and production.

Abstract

Vectors derived from Human Immunodeficiency Virus type 1 (HIV-1) are being widely developed for gene therapy applications, principally because they are able to transduce both dividing and non-dividing cells and result in stable, long term gene expression. However, these vectors are difficult to produce in high titres and sufficient volumes for large scale experiments and clinical application. Therefore, an investigation into methods to improve the production of HIV-1 derived gene transfer vectors was undertaken. One factor that limits the production of recombinant virus is the amount of viral genomic RNA available for packaging into virions. Therefore, a transfer vector was modified with the aim of increasing the amount of genomic RNA produced. Substitution of the polyadenylation (pA) signal, mutation splice donor sites and removal of unnecessary sequences were all examined. pA signal readthrough was quantified to determine the effect of these modifications on the rate of pA signal readthrough. Insertional mutagenesis and vector mobilisation are recognised risk factors with all integrating vectors. Self inactivating (SIN) vectors, which contain a deletion of U3 sequences in the 3’ LTR, demonstrate a reduced rate of mobilisation. Transduction with these vectors results in a provirus containing no viral promoter elements, with transcription of the transgene being controlled from an internal promoter. However, LTR repair of SIN vectors occurs at an appreciable frequency. Therefore, the extent of this deletion was maximised and the effect on the frequency of the repair examined. The production of lentiviral gene therapy vectors by large-scale transient transfection is both time consuming and technically difficult. Therefore, methods to increase the scale of production without compromising virus titre were developed. This resulted in fewer transfections and less handling of the cells when making virus on a large scale (3-4 L). In order to process the virus on this scale in a single day (i.e. 8 hours), new concentration and purification methods were established. The protocol consisted of low speed centrifugation, 0.45 μm filtration, 750 kDa ultrafiltration, 0.8 μm filtration and ultracentrifugation. However, the use of ultracentrifugation means that this protocol is not amenable to further scale up. Therefore, the replacement of the ultracentrifugation step with anion exchange was investigated. A number of different resins and anion exchange devices were investigated, two of which show promise for large scale purification of HIV-1 derived gene transfer vectors. In an ideal world, HIV-1 derived gene transfer vectors would be produced using stable packaging cell lines engineered to produce the desired virus. However, previous attempts to produce such a cell line with the desired properties have had limited success and have generally used outdated helper systems. Therefore, in an attempt to combine the efficiency advantages of having a single helper plasmid with the safety advantages of expressing each protein separately, a single packaging construct that contained separate transcription units for each of the required proteins was produced. Transcription of cyotoxic proteins was controlled by inducible promoters. Initial results suggest that such a system is technically feasible but that further work is required to optimise the expression of helper functions.