by ecancer reporter Clare Sansom

On the cellular level, all cancers arise from the excessive growth and division of cells, and, on the molecular level, this process is orchestrated by changes in the genes that regulate these processes.

The protein products of many of these genes are targets for anti-cancer drugs.

However, the mechanisms through which their expression is regulated are still poorly understood.

Cell differentiation is a particularly complex process at the genetic level, involving the re-programming of gene expression pathways to control mRNA splicing and small RNA stability and to synthesise a different and wider range of proteins.

The DEAD box or DDX proteins are a conserved family of helicases – proteins that bind to and separate nucleic acid strands – that are involved in different aspects of RNA metabolism.

Two members of this protein family, DDX5 and DX17, are known to be involved in several important processes of cell differentiation, including the regulation of muscle differentiation and the differentiation of epithelial cells into fibroblasts.

They are also involved in the biogenesis of one type of small, non-coding RNA, microRNA (miRNA) and in regulating alternative splicing.

Two different mechanisms have been proposed for the role of these proteins in the control of alternative splicing, but neither has been proved.

A group of researchers led by Cyril Bourgeois and Didier Auboeuf at the Centre de Recherche en Cancérologie de Lyon, Lyon, France have now explored the role of these proteins in controlling transcription, miRNA stability and splicing in cancer cells.

Firstly, they conducted a genome-wide study of the activity of these proteins in regulating mRNA splicing in epithelial MCF7 breast cancer cells.

A total of 233 single cassette exons were found to be more skipped and 139 exons were more included following depletion of DDX5 and DDX17 using small interfering RNA (siRNA).

Exons that were included more often when DDX5 and DDX17 were depleted (class I exons) had a low base inclusion rate and were flanked by large introns.

In contrast, exons that were skipped more often (class S exons) were highly included in control cells and were flanked by short introns.

Both these types of exon had weak 5’ splice sites (5’ss).

Bourgeois, Auboeuf and their colleagues suggested that the high inclusion rate of class S exons in control cells despite their weak 5’ splice sites may depend on DDX5 and DDX17.

They tested this by searching for motifs known to be associated with intronic splicing downstream of class S exons, and found these regions of DNA to be enriched in sequences that bind to heterogeneous ribonucleoprotein (hnRNP) H/F splicing factors.

Previous research has shown that many of the class S exons are also skipped if this splicing factor is depleted.

The researchers showed that depletion of both hnRNP H/F and DDX5/DDX17 further enhanced the skipping of those class S exons with downstream hnRNP binding sites, indicating that the helicases act in cooperation with the splicing factors.

These splicing factors are known to bind to guanine-rich sequences that can form stable, four-stranded structures known as G-quadruplexes; these sequences are often found downstream of class S exons and of those immediately preceding class I exons.

Bourgeois and Auboeuf proposed that the RNA helicase activity of the DEAD box proteins might enhance the binding of hnRNP H/F to these quadruplex structures, and presented several pieces of experimental evidence to back up this hypothesis.

The researchers then confirmed the cooperation between the helicases and the ribo-nucleoprotein in a genome-wide analysis of another breast cancer cell line, MCF10A following RNAi depletion of DDX5/DDX17.

They showed that cooperation between DDX5/DDX17 and hnRNP H/F was involved in specific splicing programs that are associated with the differentiation of epithelial cell differentiation and with myogenesis.

Interestingly, the down-regulation of the DEAD box proteins was found to play an important part in each of these differentiation processes as part of a feedback loop that involves small interfering RNA (siRNA) and is regulated by proteins in the SMAD signalling pathway.

Nevertheless, high levels of these proteins are required in earlier stages of cell differentiation.

Taken together, these results suggest that the RNA helicases DDX5 and DDX17 act in concert with hnRNP H/F to coordinate a specific program of RNA splicing that is necessary for but tightly controlled in the differentiation of several important cell types.

Reference

Dardenne, E., Espinoza, M.P., Fattet, L. and 14 others (2014). RNA Helicases DDX5 and DDX17 Dynamically Orchestrate Transcription, miRNA, and Splicing Programs in Cell Differentiation. Cell Reports 7, 1900–1913.