The transcriptional landscape

Control of non-productive RNA polymerase II transcription via its early termination in metazoans

Transcription establishes the universal first step of gene expression where RNA is produced by a DNA-dependent RNA polymerase. The most versatile of eukaryotic RNA polymerases, RNA polymerase II (Pol II), transcribes a broad range of DNA including protein-coding and a variety of non-coding transcription units. Although Pol II can be configured as a durable enzyme capable of transcribing hundreds of kilobases, there is reliable evidence of widespread abortive Pol II transcription termination shortly after initiation, which is often followed by rapid degradation of the associated RNA. The molecular details underlying this phenomenon are still vague but likely reflect the action of quality control mechanisms on the early Pol II complex. Here, we summarize current knowledge of how and when such promoter-proximal quality control is asserted on metazoan Pol II.

Discovery of widespread GTP-binding motifs in genomic DNA and RNA

Biological RNAs that bind small molecules have been implicated in a variety of regulatory and catalytic processes. Inspired by these examples, we used in vitro selection to search a pool of genome-encoded RNA fragments for naturally occurring GTP aptamers. Several aptamer classes were identified, including one (the "G motif") with a G-quadruplex structure. Further analysis revealed that most RNA and DNA G-quadruplexes bind GTP. The G motif is abundant in eukaryotes, and the human genome contains ~75,000 examples with dissociation constants comparable to the GTP concentration of a eukaryotic cell (~300 μM). G-quadruplexes play roles in diverse cellular processes, and our findings raise the possibility that GTP may play a role in the function of these elements. Consistent with this possibility, the sequence requirements of several classes of regulatory G-quadruplexes parallel those of GTP binding.

Analysis of mutations that influence pre-mRNA splicing

A rapidly increasing number of human diseases are now recognized as being caused by the selection of wrong splice sites. In most cases, these changes in alternative splice site selection are due to single nucleotide exchanges in splicing regulatory elements. This chapter describes the use of bioinformatics tools to predict the influence of a mutation on alternative pre-mRNA splicing and the experimental testing of these predictions. The bioinformatic analysis determines the influence of a mutation on splicing enhancers and silencers, splice sites and RNA secondary structures. This approach generates hypotheses that are tested using splicing reporter constructs, which are then analyzed in transfection assays. We describe a recombination-based system that allows for the generation of splicing reporter constructs in the first week and their subsequent analysis in the second week.