Featured Post

91 A kinetic model of pre-mRNA splicing / RNA剪接的模擬


/*—————divider—————*/
Hello everyone! Today I am going to introduce an interesting work about a kinetic model of pre-mRNA splicing. Their method is very straightforward yet provides lots of insights about this fundamental molecular process, and it really inspired me a lot. I hope you guys enjoy!

/*—————divider—————*/

In eukaryotes, the primary transcripts of protein-encoding genes, or the pre-mRNAs, have to go through a splicing process that removes the unwanted portions, or the introns, before they become fully functional mRNAs. Variations in the splicing products are known to affect development. For example, the DNA methylation of a gene called “alk” would affect the final mRNA, which in turn determines whether a honeybee will become a queen or a worker [1]. Mutations in splicing sites are also known to cause diseases such as the limb girdle muscular dystrophy [2]. Previous studies relied heavily on statistical methods to infer the factors, such as DNA methylation or RNA binding proteins, that may affect the proportion of different splicing products. Despite its importance, a quantitative and mechanistic model that could simulate and explain the splicing process was lacking.

The model of co-transcriptional splicing, which showed the detailed kinetics of the transitions among different splicing intermediates. (Figure source: Figure 1B of our suggested reading)

In 2018, a group of researchers in San Diego Center for Systems Biology (SDCSB) successfully developed a kinetic model that deepens our understanding about the splicing process. In their model, various splicing intermediates are denoted as different nodes, and the transitions among these nodes are calculated by Markov Chain methods with a transition matrix. Using their model, they found that when the splicing starts could make a huge difference in the final splicing products: If splicing and transcription happen together (or the “co-transcriptional splicing”), the final mRNAs are more likely to retain all of their exons. On the other hand, if splicing only happens after the transcription is finished (or the “post-transcriptional splicing”), the final mRNAs could lose their exons easily. This is in fact quite intuitive, as the spliceosomes can only remove the introns that are fully transcribed. For an internal exon that is surrounded by its upstream and downstream introns, if the 3’ end of the downstream intron is not transcribed yet, the spliceosomes can only remove the upstream intron, rather than remove both introns and the exon altogether. 

Their model implies that we can increase the number of retained exons by several ways. For example, if we tune down the transcription rate of a gene with co-transcriptional splicing, we can increase the number of retained exons in our final products. Another example would be an internal exon followed by a short downstream intron. Normally, this kind of internal exon would easily be excised with its surrounding introns altogether. However, if we make the spliceosomes bind to the 3’ splicing site more rapidly, we can rescue the internal exon.

Previous studies about RNA splicing usually assume a post-transcriptional splicing mechanism implicitly. However, this study showed that co-transcriptional splicing can be an important mechanism to ensure high fidelity in final mRNAs. The study also provided a feasible kinetic framework that can be readily adjusted by including more intermediate states or specifying more parameters and variables in the transition matrix, which will enable us to study how the interactions among RNA polymerases, spliceosomes, enhancers, and silencers affect pre-mRNA splicing quantitatively. This work is definitely a major step towards a more quantitative understanding of this important mechanism.

/*—————divider—————*/
Reference:
[1] Foret, S., Kucharski, R., Pellegrini, M., Feng, S., Jacobsen, S., Robinson, G. and Maleszka, R. (2012). DNA methylation dynamics, metabolic fluxes, gene splicing, and alternative phenotypes in honey bees. Proceedings of the National Academy of Sciences, 109(13), pp.4968-4973.
[2] Muchir, A. (2000). Identification of mutations in the gene encoding lamins A/C in autosomal dominant limb girdle muscular dystrophy with atrioventricular conduction disturbances (LGMD1B). Human Molecular Genetics, 9(9), pp.1453-1459.

Suggested reading:
Davis-Turak, J., Johnson, T. and Hoffmann, A. (2018). Mathematical modeling identifies potential gene structure determinants of co-transcriptional control of alternative pre-mRNA splicing. Nucleic Acids Research, 47(3), pp.1602-1603.

Comments