Reciprocal Effects of Splicing and Polyadenylation on Human Immunodeficiency Virus Type 1 Pre-mRNA Processing

Abstract

Insertion of a functional splicing cassette into a construct containing the HIV-1 poly(A) site followed by the adenovirus L3 poly(A) site results in both specific stimulation of 3′ end processing at the HIV-1 site and an increase in the steady-state levels of RNA processed at both sites. To further evaluate this influence of splicing on processing of the HIV-1 poly(A) site, defined mutations which alter or abolish splicing of the intron were made and analyzed for their effects on polyadenylation and steady-state levels of RNA. The data show that a point mutation at the 3′ splice site caused activation of a cryptic splice acceptor that is as efficient as the wild-type acceptor. Substitution of this mutant intron for the wild-type intron resulted in stimulation of the HIV-1 poly(A) site to levels equivalent to those caused by the wild-type intron. This mutant did not, however, have as great an effect on steady-state RNA levels as the wild-type intron. A second construct containing a mutated branch point and polypyrimidine tract resulted in abolishment of splicing and a decrease in both poly(A) site use and steady-state levels of RNA. These data demonstrate that the enhanced use of the HIV-1 poly(A) site is a direct result of the splicing reaction, and not merely due to the sequences that were inserted. The effect that poly(A) site strength has on splicing was also addressed. Using activation of the cryptic splice acceptor to indicate changes in splicing efficiency resulting from alterations in poly(A) site strength, it was determined that poly(A) site strength does have an effect on the efficiency of the splicing reaction.