The Geneva group had earlier found factors linking transcription activation to the nuclear pore. They now use a nuclease fused to a nuclear pore protein, which clips any DNA that comes to the nuclear pore for a visit.
The regions preferentially cleaved by the nuclease were just upstream of ∼40% of genes on chromosome VI. Constitutively expressed genes use different promoter elements and may not need the activation conferred by the visit. But back-of-the-envelope calculations suggest that there are enough pores for all activated genes to get in a quick visit, even if they do so for every round of transcription.
Laemmli suggests that “the pore is an active participant in forming a complex between enhancer and promoter.” Constrained enhancers could no longer influence more distant promoters.
In higher eukaryotes, Laemmli thinks that so-called transcription factories may act as promoter attachment and assembly sites, and thus as the intranuclear equivalent of nuclear pores. In yeast, the Geneva group is mutating a newly isolated gene that should confirm whether the visit to the pore is an essential event.