Field of Science

Programming bacteria for search and destroy

ResearchBlogging.orgAs iGEM season is now properly underway, I thought I'd have a look at a synthetic biology paper and found this fairly awesome one about programming bacteria to hunt out and destroy atrazine, a chemical herbicide pollutant. One of the most exciting things about this work was that it didn't just involve bacteria with the ability to remove atrazine from the environment but to actively migrate towards the chemical and then destroy it.
The chemical structure of atrazine

The bacteria are controlled using riboswitches - little RNA pieces that can bind directly to a ligand (or signal molecule) and cause a change in gene expression, changes which can include switching on or off the genes involved in cell movement. Atrazine is a good molecule to start with because as well as being a relevant pollutant it also contains plenty of N-H bonds which are good for forming hydrogen-bond interactions with RNA. As well as that it has a well-characterized breakdown pathway, all components of which have been expressed successfully in E. coli.

The first stage in creating these seek and destroy bacteria was finding RNA sequences that would bind to atrazine. This was done by attaching the atrazine to a solid support and running bits of RNA past it, to see which ones would bind. They then took these successful binders and tested for riboswitch activity, i.e whether the binding to atrazine caused a conformational change in the RNA that lead to the turning on of a gene. They did this by putting a sequence complimentary to the isolated RNA upstream of the DNA for the CheZ gene, which controls motility in E. coli and then carrying out the selection process shown below:

In the absence of atrazine the CheZ is not synthesized and the bacteria stay where they are. When atrazine is added to the plate, the bacteria start to move...

Dose-dependent assays were then done on the successful RNA sequences, to characterise the reaction and check that it actually was the atrazine levels that lead to movement rather than some other confounding factor. Sequencing and examination of the binding site also helped to characterize the riboswitch and determine how it was working. The genes for atrazine-consuming ability were then added to the bacteria that moved towards the atrazine, leading to a little search-and destroy module capable of seeking out a dangerous pollutant and removing it from the environment.

Sinha J, Reyes SJ, & Gallivan JP (2010). Reprogramming bacteria to seek and destroy an herbicide. Nature chemical biology, 6 (6), 464-70 PMID: 20453864

Follow me on Twitter!


Captain Skellett said...

Way cool. It took me a while to remember where I'd heard of atrazine, but it was recently linked to sex change in frogs. Sounds like good research, but it would be even better if atrazine was made illegal. If wishes were fishes...

Lab Rat said...

From a report in 2003: "yield loss plus increased herbicide cost may result in an average estimated loss of $28 per acre" if atrazine were unavailable to corn farmers.

That would be why it's not illegal... I hadn't heard about the emasculated frogs but it made for interesting (if sad) reading.

Lucas Brouwers said...

Wow, great post and some awesome and elegant 'design' work (I've always wondered if artificial selection can be seen as design)!
There's one thing I don't understand though: why would the atrazine now work as a chemoattractant? The motility induced by CheZ is like random movement right? Sorry if I misunderstood the experiment!

Lab Rat said...

CheZ works by moving the bacteria when it sees chemical attractant (albeit in random directions), and stopping movement when the attractant goes away. This leads to whats called a 'biased random walk', random movement that overall heads towards a chemoattractant.

What the experiment did was to introduce an atrazine sensor and make that turn on CheZ. So the bacterial 'biased random walk' will head in the direction of the atrazine.

CheZ is a wonderfully elegant simple walking system, I should probably do a post on it some time.