Promoter trapping reveals significant differences in integration site selection between MLV and HIV vectors in primary hematopoietic cells

Recent reports have indicated that human immunodeficiency virus (HIV) and murine leukemia virus (MLV) vectors preferentially integrate into active genes. Here, we used a novel approach based on genetic trapping to rapidly score several thousand integration sites and found that MLV vectors trapped cellular promoters more efficiently than HIV vectors. Remarkably, 1 in 5 MLV integrations trapped an active promoter in different cell lines and primary hematopoietic cells. Such frequency was even higher in growth-stimulated lymphocytes. We show that the different behavior of MLV and HIV vectors was dependent on a different integration pattern within transcribed genes. Whereas MLV-based traps showed a strong bias for promoter-proximal integration leading to efficient reporter expression, HIV-based traps integrated throughout transcriptional units and were limited for expression by the distance from the promoter and the reading frame of the targeted gene. Our results indicate a strong propensity of MLV to establish transcriptional interactions with cellular promoters, a behavior that may have evolved to enhance proviral expression and may increase the insertional mutagenesis risk. Promoter trapping efficiency provides a convenient readout to assess transcriptional interactions between the vector and its flanking genes at the integration site and to compare integration site selection among different cell types and in different growth conditions.