User:ElNando888/Blog/New switches

(preformatting for possible publication on the main site)

A new kind of switch will soon be introduced in EteRNA. They will be experimentally tested in <a href="http://eterna.cmu.edu/web/player/126931/" target="_blank">Johan Andreasson</a>'s chip-based pipeline, just like the first <a style="font-weight: bold;" href="http://eterna.cmu.edu/web/lab/4736274/" target="_blank">MS2 Riboswitches On Chip</a> lab was.

The principles for these new switches are essentially the same as before:

there are 2 states, OFF (unbound) and ON (bound)
an external molecule provides a free energy bonus to make the shape-shifting happen

There are a few differences though:

the molecule isn't a ligand like FMN or Theophylline, it's another (usually shorter) RNA strand: a microRNA
there's no predefined binding site, it will be up to the designers to decide where to place it
the free energy bonus will logically depend on the pairings chosen by the designer

There is also a more fundamental difference: microRNAs are being very intensively investigated by about everyone, pharmaceutic companies and of course, academic researchers. A quick look at the <a href="http://en.wikipedia.org/wiki/MicroRNA" target="_blank">Wikipedia entry for microRNAs</a> tells already a lot about the numerous fields of application and how important this topic is in current medical research. And we at EteRNA will soon get to investigate them ourselves!

The interface that will allow us players to manipulate these RNA dimer switches (as player puzzles and as lab designs) is nearly ready, as can be seen in the screenshot, but still in development. One thing is for sure though: they are coming very soon!

I realize that some of you may not be familiar with the word « dimer ». Pictures are better than a thousand words, they say. Let's see

Since novelties tend to be disconcerting at first, I thought that maybe some of you might be interested by the preparation work I did about these switches. First, a few words about the goals towards which we're going to strive. Let's take an example.

<a href="/wiki/images/MiRNA_switch_puzzle_OFF_state.png" target="_blank"><img style="float: left; margin-right: 8px; width: 150px;" src="/wiki/images/MiRNA_switch_puzzle_OFF_state.png" alt="" /></a>
This would simply mean that we don't care how the sequence will fold in the absence of a certain miRNA. The only thing we should care about here is that the MS2 hairpin should not be present in the MFE structure.

<a href="/wiki/images/MiRNA_switch_puzzle_ON_state.png" target="_blank"><img style="float: left; margin-left: 24px; margin-right: 24px; width: 110px;" src="/wiki/images/MiRNA_switch_puzzle_ON_state.png" alt="" /></a>And then, we "just" need to make the MS2 hairpin appear when the miRNA is added to the solution.

In other words, the structure with the MS2 hairpin should be sub-optimal compared to the other state, but should emerge as better than optimal thanks to the added bonus provided by the pairings with the miRNA strand.

Right... And how do we do that? I know, it sounds a little complicated, but I'm gonna let you on a little secret: there are known 'techniques' to solve this problem, and before we try to invent our own techniques, I'm going to to show you one of them, using... pen & paper (I'm not even kidding).

(See? I really wasn't kidding!) On the left, we have the MS2 hairpin, no introductions needed I assume, and on the right, the miRNA we're going to work with. For those interested, this sequence is found in many mammals, it's got a technical name, hsa-mir-208a, and you can read <a href="http://www.mirbase.org/cgi-bin/mirna_entry.pl?acc=MI0000251" target="_blank">more about it on mirBASE</a>.

<a href="/wiki/images/MiRNA_switch_puzzle_-_diag_2.png" target="_blank"><img style="float: left; margin-left: 10px; margin-right: 18px; width: 280px;" src="/wiki/images/MiRNA_switch_puzzle_-_diag_2.png" alt="" /></a> So, we're going to need a suboptimal and a "super-optimal".

A simple way to make a suboptimal is presented on the left. It simply consists in replicating a few bases from the sequence, at the start or at the end.

The MS2 hairpin is about -5~6 kcal solid. The construct at the bottom drops to about -3 kcal stability.

Now, how are we going to create a "super-optimal"?

<a href="/wiki/images/MiRNA_switch_puzzle_-_diag_3.png" target="_blank"><img style="float: right; margin-left: 8px; width: 350px;" src="/wiki/images/MiRNA_switch_puzzle_-_diag_3.png" alt="" /></a> Well, that's fairly easy. We will simply add a few pairs. Let's imagine we add six. On average, it should give us a 6 x -1.5 = -9 kcal boost in stability, which should be plenty enough to beat the MS2 hairpin.

Great! Now we have both a suboptimal and a super-optimal. Now the question is: how to use the miRNA strand to make the super-optimal break apart and let the MS2 hairpin take the advantage?

How about this? If we place the miRNA like presented above, we can derive all the nucleobases that we're going to need, by making them complementary. For what purpose?

The plan is to make the 6 basepairs long stack "slide" from one side to the other. The scientists call it a "strand displacement". As you can see this is pretty simple. Thermodynamically speaking, it should be easy for the molecules to evolve from the top to the bottom and vice-versa (the transition is completely reversible).

But, the strands that will reach the state figured at the bottom of the previous diagram, rather than returning to the original state, have the possibility to evolve in a different direction.

Do you recognize the highlighted section? Yes! that's our engineered suboptimal. Which means that locally, the RNA has a better option: the MS2 hairpin.

And voila! <sound-effect type="TADAAA" volume="MAX" /> 8P

Cool, right? :)
Wanna try it? Ok, ok, just because it's you, go try the <a href="http://nando.eternadev.org/web/puzzle/3378672/" target="_blank">dimer tutorial puzzle</a> (very easy), or if you feel confident, you may want to check out <a href="http://nando.eternadev.org/web/puzzle/3378629/" target="_blank">the lab preview</a>, but shhhh, it's all a secret :P

And as a last word, we have a little "surprise":

(maybe better watch it fullscreen, you'll see better)

-- Nando

Among the various examples of this technique that can be found in the scientific literature, the following one is my favorite:
Bhadra, S., & Ellington, A. D. (2014). <a href="http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3985647/" target="_blank">Design and application of cotranscriptional non-enzymatic RNA circuits and signal transducers.</a> Nucleic Acids Research, 42(7), e58. doi:10.1093/nar/gku074
(in particular, observe figures 5 & 7)