High-efficiency identification of genes by functional analysis from a retroviral cDNA expression library

Retroviral gene transfer efficiently delivers genes of interest stably into target cells, and expression cDNA cloning has been shown to be highly successful. Considering these two advantages, we now report a method by which one can identify genes stimulating cell growth through functional analysis. The first step requires the construction of a retroviral cDNA expression library and the optimization of transfection of vector DNA into virus packaging cells. The second step involves the cocultivation of target cells with libraries of retrovirus-producing cells, resulting in the amplification of target cells transduced with a gene(s) stimulating cell growth. Under standardized conditions of transfection, we detected an average of 4,000 independent clones per dish, among which expression of a retroviral beta-galactosidase gene at an abundance of 0.2% could be detected. Next, we demonstrated the augmentation of the sensitivity of the assay by retroviral infection and functional analysis. We did this by cocultivating factor-dependent (FD) cells with dishes of GP/E cells transfected with plasmids containing various molar ratios of pN2-IL3 DNA and retroviral library cDNA and by determining the highest dilution of pN2-IL3 which still resulted in the conversion of FD cells to factor independence. The retroviral interleukin-3 gene at an abundance as low as 0.001% could be detected. Indeed, we were able to detect from FD cells the development of factor-independent colonies with different phenotypes after retroviral transfer of cDNAs from an immortalized hemopoietic stem cell line. Thus, the combination of a standardized high-efficiency DNA transfection and retrovirus-mediated gene transfer should facilitate the identification of genes capable of conferring to target FD cells a detectable new function or phenotype. By scaling up the size of the experiment realistically during screening, the assay can detect cDNA at an abundance of lower than 0.0001%.