|
Abstract
|
The splicing of two small introns of 58 and 67 nucleotides (nts) of the Drosophila gamma carboxylase (DγC) in cultured human cells was examined. The two introns are below the minimum size, which was proposed as a requirement for an intron to be spliced in human cells. None of the well-known software examined was able to provide a proper prediction for the DγC intron 1 in a mammalian host. Our in vivo experiments recognized only donor splice sites (not the acceptor site) of the DγC intron 1 by the mammalian cell, which, together with a new splicing acceptor site, generated an intron of 159 nts that was not predictable in silico. The experimental results showed that, for an intron to be spliced in mammalian splicing machinery, a minimum intron-length and acceptable branch point (BP) are more important than a pyrimidine-rich stretch in the 3'-region and wild-type splice sites. The in vivo approach, such as the one reported in this work, seems reliable for inspecting the interchangeability of introns among species.
|