Strong natural pausing by RNA polymerase II within 10 bases of transcription start may result in repeated slippage and reextension of the nascent RNA.

We find that immediately following transcript initiation, RNA polymerase II pauses at several locations even in the presence of relatively high (200 microM) levels of nucleoside triphosphates. Strong pauses with half-lives of >30 s were observed at +7, +18/19, and about +25 on the template used in these experiments. We show that the strong pause at +7, after the synthesis of 5'-ACUCUCU, leads to repeated cycles of upstream slippage of the RNA-DNA hybrid followed by re-pairing with the DNA and continued RNA synthesis. The resulting transcripts are 2, 4, and 6 bases longer than predicted by the template sequence. Slippage is efficient when transcription is primed with the +1/+2 (ApC) dinucleotide, and it occurs at even higher levels with the +2/+3 primer (CpU). Slippage can occur at high levels with ATP initiation, but priming with CpA (-1/+1) supports very little slippage. This latter result is not simply an effect of transcript length at the point of pausing. Slippage can also occur with a second template on which the polymerase can be paused after synthesizing ACUCU. Slippage is not reduced by an ATP analog that blocks promoter escape, but it is inhibited by substitution of 5Br-U for U in the RNA. Our results reveal an unexpected flexibility of RNA polymerase II ternary complexes during the very early stage of transcription, and they suggest that initiation at different locations within the same promoter gives rise to transcription complexes with different properties.