The pore–transcription connection

The transcriptional state of a gene is connected to its association with the nuclear pore, according to Jason Casolari, Pamela Silver, and colleagues (Harvard Medical School, Boston, MA). Although the silencing of certain loci was known to rely on their localization to the nuclear periphery, where the pores lie, the new results suggest that yeast pores prefer transcriptionally active genes. “You get more bang for your buck if highly active genes are at the pore,” says Casolari, because it may expedite export of the transcripts.

The nuclear pore is, however, important for all sorts of genes—active and inactive, and the boundaries between them. The group shows that pore proteins are fond of both active and inactive genes with binding sites for the Rap1 transcription factor. Rap1 has boundary activity—it shields intervening sequences from the transcriptional state of the outlying DNA. Some nuclear transport proteins also have boundary activity. Association with the nuclear pore might thus prevent the unwanted spreading of either activation or silencing into nearby regions.

Only one nuclear transport protein examined, the Prp20 RanGTP exchange factor, strongly favored inactive genes. As Prp20 helps to release cargo from their import carriers, the authors speculate that it might lie near inactive genes so that it is ready to release any imported transcription factors for fast gene activation.

Prp20 was found at silent GAL genes but was replaced by other pore proteins when the genes were activated. These proteins included both structural and shuttling components of the nuclear pore, which were most often found at strongly transcribed genes. Some of the favorite targets of the pore proteins were genes involved in protein biosynthesis, whose export should be even more efficient as the genes are coexpressed and found in clusters.

The active GAL genes preferred the nuclear periphery, but it is not clear whether transcription precedes the change in localization. How the DNA reaches the periphery is also unknown, but may be a consequence of protein–protein interactions between the transcriptional machinery and pore proteins, perhaps via hnRNPs. ▪

Reference: