Lineage-specific and high-level expression of transgenes introduced into hematopoietic stem cells have been the holy grail of gene therapy for over a decade. Initial excitement and competitiveness to be first quickly faded into frustration and disappointment as 2 major hurdles were identified using standard retroviral vectors. First, early human clinical trials demonstrated the deficiencies of murine and in vitro models for predicting gene transfer efficiency into true repopulating stem cells, and levels of gene transfer were far below those required for therapeutic efficacy. Second, the inclusion of elements required for high-level and tissue-specific expression in retroviral vectors, studied most extensively in relation to hemoglobin transgenes, resulted in unstable vector genomes and resulting low vector production.
Progress toward clearing the first hurdle began several years ago, when relatively simple modifications in culture conditions used during transduction of primitive nonhuman primate and human hematopoietic cells increased vector uptake and integration. These included the use of flt3 ligand in combination with other early-acting cytokines to stimulate the required passage through the cell cycle while retaining in vivo engraftment ability, and use of a supportive surface composed of a carboxy-terminal fragment of fibronectin, important both for colocalizing vector particles and primitive cells, and maintaining stem cell properties of cultured cells. These advances resulted in encouraging results in nonhuman primates and, finally, an unequivocally successful human clinical trial, with correction of the immune defect in children with X-linked severe combined immunodeficiency by Fischer and coworkers in France.
But overall levels of stem- and progenitor-cell retroviral transduction are still insufficient for many applications, and recent interest has focused on the development of safety-modified lentiviral vectors, based on their ability to transduce cells without requirement for nuclear membrane breakdown and passage through the M phase of the cell cycle. Results in murine-human xenograft models are encouraging, and extensive modifications of the HIV sequences in the helper and vector genomes at this point reassure investigators that recombination and productive HIV infection using these vector systems is extraordinarily unlikely. But results in nonhuman primates to date have not indicated a significant improvement over optimized standard retroviral vectors, and thus the advantages of lentiviral vectors simply for increasing stem-cell transduction efficiency are not yet clear.
The paper by Cui and colleagues (page 399), however, highlights progress toward overcoming the second hurdle, utilizing a potentially more uniquely useful property of lentiviral vectors for gene therapy applications. They have focused on achieving lineage-specific expression of a transgene in dendritic cells derived from transduced hematopoietic stem and progenitor cells. Using modern safety-modified self-inactivating (SIN) lentiviral vectors, the authors have been able to accommodate the HLA-DR class II MHC promoter to drive lineage-enhanced expression of a green fluorescent protein specifically in dendritic cells. None of the problems encountered with the inclusion of strong nonviral LTR promoter/enhancer elements in standard retroviral vectors were encountered, such as a drop in titer or recombination of the vector in the producer cells. The ratio of expression of GFP in dendritic cells to other lineages generated from transduced human CD34+ cells in vitro, or after engraftment in NOD/SCID mice, was much higher using the MHC promoter than with a constitutive housekeeping EF-1 alpha promoter. This is the second demonstration that high-level, lineage-specific expression can be achieved using lentiviral vectors, the first being the recent report from Sadelain's group that lentiviral vectors can overcome the myriad problems encountered using retroviral vectors to drive erythroid-specific high-level hemoglobin gene expression.
Expression of transgenes specifically in dendritic cells may be important for tumor vaccine strategies, but the results are more important globally as a demonstration that lentiviral vectors may be able to accommodate large, strong, and complex genetic regulatory elements conferring high-level expression and lineage-specificity, critical to the success of many experimental and therapeutic gene-transfer strategies.