Functional mining of transporters using synthetic selections

After our successful Meet the Fellows series, we are starting a new project: The metaRNA Fellows will introduce papers new and old that they find interesting or inspiring. Today, Josi Buerger starts  with a recent publication on riboswitches.

On this blog, we have talked at length about aptamers and biosensors. There has been a post on aptamers in general, the specific SPINACH aptamer, and an application on how biosensors can be targeted to specific cells.

Today I want to introduce a biosensor for the vitamin thiamine. This biosensor detects a phosphorylated version of the thiamine molecule and this detection is sensitive enough to sense the very low thiamine concentrations inside the cell. A recent publication describes how this biosensor can explore bacterial biodiversity and how this has ramifications for the discovery of new drug targets.


The thiamine biosensor consists of a riboswitch which is placed in front of a gene that makes a bacterial cell resistant to antibiotics. When thiamine is present, the riboswitch is turned ON and the bacteria cell can survive the antibiotic. However, if there is no thiamine present in the cell, the resistance gene can not be utilised and the cell is killed by the antibiotic. This has the distinct advantage that those cells that do not have  thiamine are killed by the riboswitch-modulated system.

Riboswitch: An RNA molecules that switches conformation upon binding of a particular target molecules. Riboswitches are often used to control gene expression.

This riboswitch was used to look at how bacterial cells transport thiamine from their environments through the cell wall into the cell. The image on the right is a nice representation of this – transporters are a hole in the cell membrane that can allotransporterw particular molecules to enter or exit the cell. There are particular strains of E. coli that do not have a naturally occurring importer for thiamine and instead synthesise whatever they need. The trick for the selection system here is that the strain is engineered to lose its biosynthetic capability, so that it is dependent on external thiamine – but it can’t take up the external thiamine!

In order to survive, the cell is exposed to a “metagenomic library” which is a pool of many fragments of DNA from many different microorganisms. Then, the cell randomly takes up bits of this external DNA. If it manages to take up a gene that encodes a transporter, voila! It can survive.

So where does the riboswitch come in? There are countless types of transporters in the metagenomic gene pool and some of them are non-specific. But as this study wanted to find highly specific and efficient thiamine transporters, it needed to increase the amount of thiamine required for survival. Therefore, coupling the presence of thiamine to the ability to survive high levels of antibiotics can do just that. The image below shows how the cells survive only if they manage to find a transporter from the metagenomic library. Furthermore, this transporter has to be functional.


From Genee et al., 2016

So what are the results of the riboswitch selection? The researchers found a type of thiamine transporters that was previously undiscovered. In particular, this type of thiamine transporter, called PnuT, is prevalent in bacteria of the human gut. Moreover, some known human pathogens have only the new transporter type for their thiamine uptake and therefore, their survival depends on it. This means that the Pnu transporters constitutes an interesting new drug target for human health.

The publication is a nice example of how synthetic biology, cell metabolism, and metagenomic DNA can all come together and point to a new direction for drug targets.


Genee, Hans J., et al. “Functional mining of transporters using synthetic selections.” Nature Chemical Biology 12.12 (2016): 1015-1022.