A new genomic tool defines the human transcriptional landscape with single-nucleotide precision

How genomes are transcribed by RNA polymerases in living cells is poorly defined. Here, we developed a genomic approach that visualizes RNA polymerase II transcription at single-nucleotide resolution in human cells. This approach reveals new general features of gene transcription and defines the transcriptional landscape of human cells.

HFSP Long-Term Fellow Andreas Mayer and colleagues
authored on Tue, 02 June 2015

Recent technological advances have revealed that eukaryotic genomes are pervasively transcribed. A key enzyme in genome transcription is RNA polymerase II (Pol II). Pol II synthesizes all protein-coding RNAs and many non-coding RNAs. However, the mechanisms that underlie genome transcription in vivo are poorly understood, mainly due to a lack of quantitative genomic approaches that monitor transcription at nucleotide resolution in living cells.

Here we developed a simple and robust genome-wide approach -called native elongating transcript sequencing (NET-seq)- that maps the position of RNA polymerase across the human genome at single-nucleotide resolution (Figure 1A). This strategy is based on deep sequencing of the 3’ ends of nascent transcripts providing a DNA strand-specific quantitative measure of RNA polymerase density genome-wide (Figure 1A,B). As this technique captures transcripts as they are being produced, unstable transcripts that are usually rapidly degraded via the RNA exosome are also detected. NET-seq is a new genomic tool to study the full spectrum of transcriptional activities in unperturbed human cells.

Figure 1: Human NET-seq reveals Pol II genome transcription at single-nucleotide resolution. (A) Schematic view of the human NET-seq approach; after the quantitative purification of elongating RNA polymerases the associated nascent RNA is isolated; 3’ end high-throughput sequencing of the nascent RNA identifies the last incorporated nucleotide (red) revealing the exact genomic position of RNA polymerase; (B) DNA strand-specific Pol II density with single-nucleotide resolution at a representative gene in HeLa S3 cells as determined by human NET-seq; strong peaks of Pol II occupancy indicate pausing.

Human NET-seq reveals pervasive Pol II pausing at the majority of expressed genes. Prominent Pol II pauses are observed in the promoter proximal region of most genes confirming that early transcription elongation represents a key rate-limiting step in RNA production (Figure 1B). Human NET-seq also uncovers strong Pol II pauses throughout the gene body region of active genes demonstrating that productive transcription elongation is a very discontinuous process in vivo, frequently interrupted by pausing events (Figure 1B). Very strong Pol II pauses occur at boundaries of exons that are retained in the mature RNA but not at skipped exons (Figure 2A). This indicates that Pol II slows down at retained exons which is consistent with a model where Pol II transcription elongation is kinetically coupled with RNA splicing.

Figure 2: Human NET-seq uncovers (A) Pol II pausing at the boundaries of retained exons but not at exons that are skipped and (B) convergent transcription where RNA polymerases face each other in the promoter proximal region of a large set of genes.

Human NET-seq discovers a new class of anti-sense transcription: convergent anti-sense transcription. We show that this new transcriptional activity arises locally from an open chromatin region downstream of the +1 nucleosome and converges on canonical sense transcription (Figure 2B). Genes with convergent transcription in their promoter proximal regions are lower expressed supporting the view that convergent anti-sense transcription interferes with the release of Pol II from promoter proximal pausing into productive elongation. Future studies aim to reveal the biological function of convergent anti-sense transcription and to identify the key regulatory factors that underlie Pol II pausing in vivo.

Reference

Native elongating transcript sequencing reveals human transcriptional activity at nucleotide resolution. Mayer A, di Iulio J, Maleri S, Eser U, Vierstra J, Reynolds A, Sandstrom R, Stamatoyannopoulos JA, Churchman LS. Cell 161 (3), 541-54.

Pubmed link

Cell link