Positive Charges Clog the Ribosomal Tunnel
Back when I was a lad we used to do lots of experiments, generate some data, analyse them, and try to come up with something interesting to say. Nowadays of course, experiments are for schmucks; the sheer quantity and availability of data means that you can cut out those first tedious and messy steps and just do the analysis on someone else’s data instead (assuming those data are openly available, of course).
That’s what Catherine Charneski and Laurence Hurst have done in an intriguing paper just published in this month’s PLOS Biology. You can also read about it in my accompanying Synopsis. The openly accessible data came from a 2009 Science paper from Jonathan Weissman’s lab, and this isn’t the first time that people have pored over this valuable dataset.
The problem that the authors tackled relates to the ribosome, an incredible nanomachine that trundles along gene transcripts (mRNAs), translating them three letters at a time into the proteins that run most of the workings of the cell. The translation process requires little L-shaped adapters called tRNAs which line up the appropriate amino acids; the ribosome then glues the amino acids together and shoves the growing protein chain out through an exit tunnel in the back of the ribosome.
But does the ribosome move at a more-or-less constant rate along mRNA molecules, or is its speed is influenced by factors beyond its control? There are several reasons we might care about this. Evolutionary biologists are interested in the natural optimization of protein translation and the forces that shape it, while biotechnologists would like to optimize translation speed artificially to maximize protein productivity. The speed of translation is also known to affect how proteins fold and where they end up.
We already know that translation speed is indeed highly variable, and it’s been thought that this is largely down to the abundance of the tRNA molecule needed to interpret each codon. Cells contain varying amounts of the different tRNAs, so a codon needing a rare tRNA might slow translation down because the ribosome has to hang around longer to encounter that tRNA, and vice versa.
And what about mRNA secondary structure? If the ribosome needs to plough through a tangled segment of transcript, then we’d expect it to slow down as energy is expended to unravel the obstacle.
This is where large datasets come in handy; rather than relying on anecdotal data from individual transcripts, the recently developed technique of ribosome profiling gives us a snapshot of where the ribosomes are across all the transcripts in the cell. These data can be used to infer a high-resolution map of ribosome speeds that has the potential to tell us exactly what might be slowing ribosomes down.
So what is the answer? In their paper, Charneski and Hurst show that codon/tRNA abundance and mRNA secondary structure don’t seem to matter very much. Instead, with amazing simplicity, the prime determinant of ribosome speed isn’t actually a direct feature of the mRNA itself, but a feature of the newly synthesised protein emerging from the exit tunnel. Here’s the key: if a ribosome has just translated one or more positively charged amino acids then it slows down.
The proposed explanation for this surprisingly simple outcome is that the positive charges interact electrostatically with the negatively charged lining of the ribosomal exit tunnel, gumming it up. The longer the run of positive charges, the greater the effect. As a throwaway, the authors also speculate about the polyA tails that decorate the ends of mRNAs; these would encode long strings of positively charged amino acids if translated, so might they have evolved as a sand-trap for runaway ribosomes?
An elegantly simple explanation for the observed variation in ribosome speed, a complex emergent consequence of an electrostatically charged exit tunnel, and a fine exemplar of data re-use. Nice paper.
Charneski, C., & Hurst, L. (2013). Positively Charged Residues Are the Major Determinants of Ribosomal Velocity PLoS Biology, 11 (3) DOI: 10.1371/journal.pbio.1001508