Abstract It is now clear that nuclear context is playing an essential role in gene expression. For this reason we have developed methods to study gene expression in situ. Transcription takes place in discrete foci in the nu- cleoplasm of mammalian cells. These sites concentrate several RNA polymerases II and factors involved in the production of mRNA. Moreover, these sites also contain the active machinery to carry out protein synthesis. Once the mRNA leaves the transcription site, it interacts with the nuclear pore complex and the mRNA is exported using the filaments of the nuclear pore complex in to the middle part of the nuclear pore complex and finally is released in the cytoplasm. Keywords RNA polymerase · Transcription · Nuclear pore · Nuclear translation Introduction Three different types of polymerising complex make RNA in mammalian nuclei: RNA polymerase I is dedi- cated to the synthesis of 45S ribosomal RNA within nu- cleoli, while polymerases II and III in the nucleoplasm transcribe all protein-coding genes and many small RNAs, respectively (Sentenac 1985; Young 1991; Maniatis and Reed 2002). I will describe the path that the transcripts, made by the major activity RNA poly- merase II, take from the synthetic site in the nucleus to the cytoplasm. Our ultimate goal is to map the route and the flux of molecules along it. The starting point: transcription sites in the nucleus The number of active polymerases As background, we estimated that there were 320,000 molecules of the largest (catalytic) subunit of polymer- ase II in a HeLa cell by quantitative immunoblotting, us- ing the bacterially expressed subunit as a reference (Kimura et al. 1999). How many of these are active? Two independent approaches gave similar results. In one, we used the strong detergent, sarkosyl, to disassem- ble nuclei and strip from the template all proteins except for engaged polymerases (Hawley and Roeder 1987) and their transcripts; about one quarter remained engaged (Kimura et al. 1999). In the second, we determined the number of nascent transcripts in a HeLa cell. Cells were encapsulated in agarose microbeads to protect them dur- ing subsequent manipulation, permeabilised and washed to remove endogenous nucleotide triphosphates. Then, transcripts were trimmed with RNase A to leave a few nucleotides attached to still-engaged polymerases. Next, those polymerases were allowed to extend their trun- cated transcripts by a few nucleotides in a limiting con- centration of [ 32 P]UTP, and the number of 32 P molecules incorporated determined. Finally, the resulting [ 32 P]RNA was sized on a gel. As incorporation of a known number of labelled nucleotides into all transcripts is associated with a known increase in length of each one, the total number of growing transcripts can be calculated. There were 90,000 nascent transcripts (and so engaged polym- erases) per cell, and the use of inhibitors showed that 15,000, 65,000 and 10,000 were made by polymerases I, II and III, respectively (Jackson et al. 1998; Pombo et al. 1999). The number of transcription sites Estimating the number of transcription sites per cell proves to be a difficult problem. If, for example, the 65,000 active molecules of polymerase II were distrib- Robert Feulgen Prize Lecture 2002 presented at the 44th Sympo- sium of the Society for Histochemistry in Vlissingen, The Nether- lands, on 28 September 2002 F.J. Iborra ( ✉ ) Sir William Dunn School of Pathology, University of Oxford, South Parks Road, Oxford, OX1 3RE, UK e-mail: fiborra@molbiol.ox.ac.uk Tel.: +44-1865-275527, Fax: +44-1865-275515 Histochem Cell Biol (2002) 118:95–103 DOI 10.1007/s00418-002-0441-z ROBERT FEULGEN PRIZE LECTURE Francisco J. Iborra The path that RNA takes from the nucleus to the cytoplasm: a trip with some surprises Accepted: 11 June 2002 / Published online: 4 July 2002 © Springer-Verlag 2002