Get up to speed with RNA velocity

In a cell, at any particular time, only a subset of genes will be expressed. This subset can vary over time as the cell adapts to both external and internal conditions. Various techniques can be used to determine what these subsets of genes are, information which can be used to understand the function and state of various cell types. However, not only are different subsets of genes expressed in cells at different times, but the rate of expression can vary per gene over time. How can we deduce which genes are becoming more highly expressed and which are decreasing in expression? A challenging question, but a method, recently published in Nature by La Manno et al.1, that uses data gained through RNA sequencing, may provide the solution to measure this RNA velocity.

rna velocity4.png
Figure 1: Which RNAs have a greater velocity?

So what would be the ideal solution?

RNA sequencing is a technique that can quantitate the global expression level of different genes in a cell. However, in order to extract RNA from cells to sequence it, you kind of have to kill the cells in the process… This means we can only get a snapshot of the genes expressed in a cell at a particular time. Instead, we want a movie. If there was a way to measure the RNA transcribed continuously over time, that would revolutionise our study of gene expression**. But we’re not quite there…yet. Anyway, you may be thinking – ‘but why not sacrifice only a few cells and repeat this over time for a population of cells?’ – Well, you’d get results, but the issue is, the population of cells that you have is heterogeneous and so they can be at different stages of the cell cycle or at different points of differentiation and could be different due to all the ‘noise’ in gene expression. Even synchronising cells so that they begin at the same stage won’t completely solve this issue. However, this can be circumvented through single cell RNA sequencing. But how can RNA velocity be extracted from this data set?

 

**We can technically already do this, but only for a limited number of genes at a time, not for a global analysis!

RNA velocity – taken from the balance of spliced and unspliced transcripts

Gene expression encompasses all the steps going from gene to gene product (e.g a protein or non-coding RNA). After transcription, a key processing step for many genes is splicing, a step that removes fragments of RNA that are not part of the mature transcript. The fragments removed are referred to as introns and the remaining sections, exons (Figure 2). Nascent transcripts will therefore still have introns, whilst in mature transcripts, the introns would have been removed. After a while, the mature transcripts are eventually degraded (Figure 3).

mrna structure.png
Figure 2: Removing introns from mRNA in a process known as splicing

So how would you distinguish a gene increasing, from a gene decreasing in expression levels? Well, if a gene was beginning to increase in expression, you would expect to see an increase in the abundance of nascent transcripts, before the abundance of mature transcripts increased. Conversely, if a gene was decreasing in expression, you would notice a drop in unspliced transcripts before the spliced transcripts decrease in number. The ratio of spliced and unspliced transcripts present in a cell therefore provides a glimpse into the future state of cell’s transcriptional state. To be more precise then, RNA velocity is the first time derivative of the spliced mRNA abundance.

steps.png
Figure 3: (1) Transcription resulting in the production of nascent transcripts which are then processed via splicing to form (2) mature transcripts. Over time mature transcripts are degraded (3).

 

By exploiting splicing to distinguish nascent from mature transcripts, chemical labelling of RNA can be avoided, allowing total RNA sequencing data from single cells to be used…handy.

But be careful..

…Some genes can be alternatively spliced. Alternative splicing results in the formation of different mRNA isoforms of a gene, whereby a different subset of introns are included in the mature transcript. This could hugely implicate the lifespan and outcome of a transcript. For example, if an intron had a sequence that recruited decay factors then the decay rates for a gene would now vary depending on whether that intron was kept. But for most genes, alternative splicing need not be a concern to this analysis.

So experimentally, to calculate RNA velocity, the process is the same as RNA sequencing, but the interpretation and analysis of the data collected sets La Manno’s study apart, allowing the future fate of a cell to be predicted.

But, how can RNA velocity be applied?

By studying single cells, individual cell trajectories can be studied providing a greater understanding into the main drivers (genes) of cells fate. Primarily then, RNA velocity measurements will aid quantitative cellular computational simulations and can be used to infer the future state of a cell. The latter could be particularly useful for tracking and understanding changes in cells during development. Other interesting processes to follow would be disease progression. It will be interesting to see how else RNA velocity is applied. However, to gain the most from this technique, it should be complemented with other techniques such as live imaging that can track the spatial location of cells during development and disease progression so that additional factors such as cell lineage can be included into studies of cell fate.

Further Reading

  1. Manno, G. La et al. RNA velocity in single cells. bioRxiv 206052 (2017). doi:10.1101/206052

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s