Prokaryotic Saturday: molecular switches and Travis' experiment

You won’t believe how much I love Saturdays. It’s very difficult to be depressed in a Saturday, seriously. Who can be depressed on Saturday? I mean, you can be depressed on Monday, that’s fine, or Wednesday, I understand. But Saturdays are just perfect. See, no matter at what time you sleep or wake up, you won’t be late to anywhere and you can also work whenever you want and it won’t be wrong.  It’s perfect!

And another good thing I have on Saturdays is the Prokaryotic Saturday, yeeey!!! 

For today’s writing I had three papers from where to choose. The options were:

1) Sorek, R., and Cossart, P. 2010. Prokaryotic transcriptomics: a new view on regulation, physiology and pathogenicity. Nat. Rev. Genet. 11 (1): 9-16. (this one is behind a paywall right here, but I have it, so if you would like to read it, email me –hope this is not considered an illegal statement-)

2) Joint, I., Doney, S.C. and Karl, D.M. 2011. Will ocean acidification affect marine microbes? ISME J. 5(1): 1-7(Available online here).

3) Smolke, C. 2005. Molecular switches for cellular sensors. Engineering and Science. 68 (4): 28-37. (Available online here).

So I chose the last one. Anyway, if you would like me to write about the other two, just let me know. I will probably write about Sorek and Cossart paper next week, the other of ocean acidification I’m not really sure, maybe later. 

Now, let me tell you why I chose Smolke’s paper. Reason #1, because I liked it so much, reason #2, because I had a test about this paper and although I read it my test wasn't very good AT ALL... :( aaaaaaah!! This was my face when I saw my result:

 So I want to use this Prokaryotic Saturday to redeem myself (gosh, I'm such a nerd).  

The paper is a little old but it is very enjoyable (something you can’t say about many scientific readings) because Smolke writes this particular paper in a very “friendly manner” (maybe that’s why I didn't take it too seriously-damn!). 

So Smolke’s team designs biosensors using RNA molecules. So basically she is trying to engineer sensors based on the RNA property to associate with proteins. Remember that RNA (and DNA) can form tertiary structures and create binding sites. In the case of RNAs, they can actually act as regulatory elements forming structures called switches, or riboswitches. These riboswitches are in bacteria, they are RNA structures that can fold and bind specifically to certain molecules (or ligands), these can be a protein, a mineral, etc. And as you can see in the picture, the binding between the aptamer (the part of the RNA that binds to the target) and the target molecule modify the structure of the entire RNA molecule, “hiding “ either transcription start sites or ribosome binding sites, in both cases controlling (turning on and off) gene expression. 

Other regulatory RNA elements are the antisense RNAs. These are small RNA pieces that are complementary to mRNA, so when they both, mRNA and antisense RNA bind, translation is blocked. 

Smolke in her paper explains how she takes advantage of these RNA tricks in the design of sensors. She is not giving any methodology, just an idea of how the process works. The first step is to combine the target molecule (the molecule you want your sensor to detect) with RNA, and RNA can be synthesized in the lab from DNA, and you can buy nucleic acids from several companies. Which DNA you use to make your RNA and exactly how do you decide which sequences you use it is not explained here. But once you have your RNAs and your target molecule, you incubate both. Then you will see that some RNAs will bind with the target, some will not. Later you take those RNAs that worked, you amplify and you continue again, with the incubation of the target molecule and the now smaller pool of RNAs. It is, as Smolke defines it, a talent search. 

But the sensors also need to be measurable. So you can add a sequence that codes for a fluorescent protein in a way that if the target molecule binds, the RNA will or will not express the protein. So imagine, you place these sequences in a cell, you add your target molecule and you wait and see. If the cell glows (or “produce” the fluorescence), then your sensor either is on or off, depending on how you designed. And according to Smolke, it is also possible to design biosensors that are able to detect concentrations of the target molecule, this is, it is not only an on and off response, but either how much of the target is in the medium. .. wooow!! 

At some point she explains the experiment of one of her students, Travis, and at the end you can’t conclude something else than “Travis is a really smart guy”. And now you will see why. What Travis did is to create a biosensor that detects a chemical called theophylline, which is found in tea and is similar in its structure to caffeine. He used yeast to insert the sensor. And this is the figure of the results of Travis’ experiment:

In the y-axis we have the expression of GFP (green fluorescent protein), so basically what he did was measure how much the yeast cells were glowing depending on the amount of target molecule (theophylline), x-axis. 

So the blue line is the switch response. And in the beginning the binding wasn't changing, but a dramatic change was observed when the target molecule reached something around 1 mM, meaning that the biosensor needed to have a certain concentration of target so this can actually be “detected”. So Travis’ biosensor worked, after the binding with theophyline, GFP production stopped, in other words, the biosensor was off when the target molecule was detected. And the picture is also showing how specific the biosensor is. The orange line is caffeine, an analogue of teophyline, as I told you before, while the green line is the biosensor only with no target and the red line is the biosensor plus an antisense RNA complementary to the binding site of the theophyline, so GFP was never produced here. As you can see, both, green and orange line shows that GFP was constantly produced, therefore, the biosensor wasn't detecting the target, because there was no target! Isn't this super interesting?  

Later Smolke explains a couple of other biosensors her students synthesized, and how they are applied. In the future, she concludes, biosensors can be engineered to detect unhealthy levels of molecules in blood samples, or to detect growth factors that drive cells to go carcinogenic. A very interesting promise. 

For now I’m going to stop here because this is already to long. But if you read the paper and would like to discuss more, not only about this experiment with the theophyline, but also the one Travis run with the caffeine, you can email me or write something in the comments. Also, if you find more recent applications of biosensors, or something cool related with this field of research, please, share it! :)

See you next week! 

The picture of the riboswitches was taken from Kim, J. N.; Breaker, R. R., Purine sensing by riboswitches. Biology of the Cell 2008, 100, (1), 1-11. Although I copied it from here. And obviously the picture of Travis' results was taken from Smolke paper. I mean, just saying.