RNA Biology

Synopsis

The role of mRNA processing in gene expression variability is poorly characterized. This study investigates the extent of cell‐to‐cell variability of alternative RNA splicing in mammalian cells using single‐molecule imaging of CAPRIN1 and MKNK2 splice isoforms.

Biological gene expression is a complex process which includes transcription, mRNA processing, and translation. As gene expression is a fundamental aspect of biological behavior, a central question within the fields of molecular and cellular biology is how effectively cells control the abundance of their gene expression products, mRNA and protein.

Previous experimental and theoretical studies have shown that there can be substantial cell‐to‐cell variation in gene expression, even between genetically identical cells grown in uniform conditions. This variation was shown to be important in a variety of biological contexts such as development, virology, immune system function, and cancer treatment. One major source of variability was shown to be transcriptional bursting, or the process in which genes are expressed sporadically separated by long durations of inexpression. Additionally, since the biochemical reactions that govern gene expression are often mediated by molecular species that are present in low numbers, variability can arise from stochastic effects owing to the random chance that an individual biochemical reaction will occur.

The role of mRNA processing in gene expression variability has not been examined thoroughly, particularly with respect to alternative splicing. Alternative RNA splicing is a form of mRNA processing which leads to the synthesis of multiple different mRNAs from a single gene. In this process, the nascent mRNA (pre‐mRNA) of a gene contains sequences known as introns that can be excised in different combinations to generate multiple gene products, known as isoforms. As alternative splicing occurs in the vast majority of human genes, it presents a potentially major source of cell‐to‐cell variability in gene expression.

In this study, we sought to characterize the extent of cell‐to‐cell variability that arises from alternative RNA splicing. To do so, we utilized a single‐molecule imaging approach based on fluorescent in situ hybridization to study the cell‐to‐cell variability in isoform ratios of two genes, CAPRIN1 and MKNK2, which each contain two splice isoforms (Figure 2 from the manuscript). Using a clonally derived, diploid, non‐transformed cell line (Rpe1 cells—retinal pigment epithelial cells), we found that variability is remarkably close to the minimum possible given the probabilistic chance of individual splicing events. In contrast, we found that isoform ratio variability was substantially larger in clonally derived HeLa cells, a cancerous cell line with an unstable karyotype. To explain the differences between the two cell lines, we further examined the potential origins of isoform ratio variability. We first studied several known sources of mRNA variability, such as transcriptional bursting, but found that they did not contribute significantly to the difference between cell lines. However, when we examined the role of splicing factors in controlling cell‐to‐cell variability, we found that lesser control over the regulation of alternative splicing is likely to be the primary source of this difference.

Cell‐to‐cell variability in gene expression owing to alternative splicing is an inevitable feature of biology. Since spliced isoforms can have different and even opposing cellular functions, it would be interesting to see if such variability can have phenotypic consequences in various biological settings. We anticipate that future work will shed light on the extent of cell‐to‐cell variability of alternative splicing for additional genes, and may identify splicing events where heterogeneity has an important functional role.

Synopsis

The function of non‐coding antisense RNAs in yeast remains to be fully understood. Steinmetz and colleagues provide evidence for a general regulatory effect of antisense expression on sense genes and for a role in spreading regulatory signals between neighboring genes.

Antisense expression, the RNA expression on the opposite strand of coding genes, is widespread but its general role has remained elusive. By expression profiling yeast in different environments and genetic backgrounds, the authors observed that genes with antisense are more frequently switched‐off and show higher expression variability. This effect is the outcome of repression that specifically acts on low levels of sense expression—a model that is experimentally validated for the SUR7 locus. Furthermore, antisense expression is shown to connect the regulation of neighbouring loci in a setting where the bidirectional promoter of a gene initiates expression antisense to an upstream gene. Together, these findings underline the regulatory potential of the downstream region of genes as promoters of antisense transcripts and indicate antisense expression as a regulatory mechanism to enhance switch‐like expression for stress–response and condition‐specific genes.