The inactivity or occlusion of the HIV-1 poly(A)
signal when in the 5′ long terminal repeat (LTR)
has been mechanistically investigated. First we show that
neither the homologous HIV-1 promoter nor the close proximity
of this RNA processing signal to the transcript initiation
site is required for the occlusion effect. Instead we demonstrate
that the major splice donor (MSD) site positioned about
200 bp downstream maintains the poly(A) site in an inactive
state. Although mutation of MSD results in activation of
the 5′ LTR poly(A) signal, this effect can be suppressed
by targeting U1 snRNAs near to the mutated MSD by base
pairing. We show that hybrid U7-U1 snRNAs can also suppress
the poly(A) signal and that this suppression is dependent
on the U1 stem-loop 1. In particular the binding site for
the U1 snRNP protein 70K that binds to the loop structure
of stem-loop 1 is associated with poly(A) site occlusion.
These experiments were carried out with an HIV-1 proviral
construct and as such emphasize the physiological importance
of this splice donor–poly(A) site interaction.