Transcription and RNA processing

Eukaryotes have three nuclear RNA polymerases: RNA polymerase I (28S, 18S and 5.8S rRNA, in the nucleolus), RNA polymerase II (pre-mRNA and most snRNAs and microRNAs) and RNA polymerase III (tRNA, 5S rRNA and other small RNAs). Pol II cannot find promoters alone: general transcription factors assemble at the core promoter — TFIID binds the TATA box (≈25–30 bp upstream of the start site) through its TATA-binding protein — and TFIIH unwinds the DNA and phosphorylates the C-terminal domain of Pol II to release it into elongation. Gene-specific activators and repressors bound to enhancers and silencers, often tens of kilobases away, communicate with the promoter through the Mediator complex and DNA looping.

The polymerase reads the template (antisense) strand 3′ → 5′ and builds RNA 5′ → 3′, so the transcript has the same sequence as the coding (sense) strand, with U in place of T. RNA polymerases need no primer and do not proofread.

Pre-mRNA is processed co-transcriptionally: (1) a 5′ 7-methylguanosine cap protects the RNA and is needed for export and translation initiation; (2) splicing by the spliceosome (snRNPs U1, U2, U4, U5, U6) removes introns (which begin with GU and end with AG) as lariats and joins exons; alternative splicing, which affects ≈95 % of multi-exon human genes, lets one gene encode several proteins; (3) cleavage downstream of an AAUAAA signal and addition of a poly(A) tail (≈200 adenines) stabilise the mRNA. The mature mRNA is exported through nuclear pores.