A new experimental approach was designed to test different predictions of current models of the nuclear architecture with respect to the topography of transcription. We constructed a plasmid, termed pIndi, which carries a reporter gene coding for a red cytoplasmic fluorescent reporter protein. Transcription of the reporter gene is regulated by the inducible promoter of the human immunodeficiency virus (HIV) and is strongly dependent on the HIV-1 Tat protein. Expressing the red fluorescent reporter protein allowed us to distinguish between cells with active and silent reporter genes. Importantly, transient transfection resulted in the clustering of plasmids, forming one or several extra-chromosomal pIndi bodies. Repetitive lac operator sequences in pIndi allowed us to visualize these bodies in living cells by the binding of LacI proteins tagged with a fluorescent protein. Using this model, we analyzed the three-dimensional nuclear topography of pIndi bodies with active or silent reporter genes. Our results are compatible with predictions of the chromosome territory-interchromatin compartment (CT-IC) model. We demonstrate that pIndi bodies localize in the IC, both in the silent and active state. Activation of transgene transcription resulted in the recruitment of RNA polymerase II and NFkappaB and a closer positioning to splicing speckles.