Biological "AND" gate

Promoter paraBAD • cyclic Adenosine MonoPhosphate • Gene transcription • A new characterization of the paraBAD promoter • Conclusion

Promoter paraBAD

In order to develop our device we needed a biological "AND" gate. We found a promoter which can be activated by 2 molecules : CRP-cAMP complex and the AraC protein. The paraBAD promoter has two states; in absence of L-arabinose the paraBAD promoter is repressed by AraC, whereas the paraC promoter is activated (unless an excess of AraC is present) [Schemes].

In presence of L-arabinose and the CRP-cAMP complex, the promoter is activated thus enabling the transcription of the downstream elements.

This AND gate provides a filter to biological noise. Check out our main results page to for the AND gate characterization.

Cyclic Adenosine MonoPhosphate

In order to make our AND gate work we need to produce cyclic Adenosine Monophosphate (cAMP). cAMP is produce by adenyl cyclase (encoded by the cyaA gene). It is an enzyme which catalyses the conversion of ATP to 3’-5’-cAMP. In E. coli, cAMP is involved in carbon catabolite repression [1] and binds to the cAMP receptor protein (CRP). The corresponding complex (CRP-cAMP) is a transcriptional factor controlling the expression of more than 220 operons [2]. It has been known for a long time that E. coli actively exports cAMP into the growth medium [3][4].

Gene transcription

The transcription of the cyaA gene and the crp gene is negatively regulated by CRP-cAMP [5] [6].
The translation of adenylate cyclase mRNA is ineffective as to prevent excessive synthesis of adenylate cyclase. This can be attributed to the fact that overproduction of cAMP is lethal to E. coli possibly due to an accumulation of methylglyoxal [7] [8]. As we want to use paraBAD like an AND Gate and because high a concentration of cAMP is lethal for E. Coli we make our device in a BW25113 ΔcyaA.

A new characterization of the paraBAD promoter

In order to characterize the paraBAD promoter we used a transcriptional fusion of the promoter and gfpmut2 from Alon on pUA66 in BW25113 ΔcyaA.
We followed the fluorescent expression of the GFP versus different concentrations of arabinose AND cAMP over time. We made this experiment in different growth media:

M9 complement with 0.03% of glucose and 0.03% of acetate (1)
M9 complement with 0.1% of glycerol (2)
M9 complement with 0.1% of acetate (3)

See the protocol.

Those different media did not affect bacterial growth.
NB : You can click on each figure to have a larger view !

medium (1)

medium (2)

medium (3)

In a first place, we tested the robustness of the AND gate.

medium (1)

medium (2)

medium (3)

As you can see on this figure even after few hours there is no fluorescent expression if one activator is absent.
Now for our system we studied the half-expression of GFP versus the cAMP concentration at 1.6% of arabinose.

medium (1)

medium (2)

medium (3)

As you can see in the two first media we have the half expression of GFP at 0.4mM of cAMP whereas in the third medium its concentration is: 0.8mM. We want a sensitive detector, therefore the third medium conditions are not optimum.

NB : if there is no arabinose we do not have any GFP expression.

medium (1)

medium (2)

medium (3)

One of ours imperative is : a fast response. Consequently, we need to know the time it takes to reach the half expression.

medium (1)

medium (3)
In this figure we can see that GFP expression is faster in the medium with glycerol.

Conclusion

With those different experiments we demonstrated that paraBAD works as expected. It will be activated only if there is arabinose and cAMP in the medium. It appears that both cAMP concentration and time response required for a half expression of GFP depends on the medium. As we want a quick response as well as a satisfying sensivity, these analysis shows that the best medium for our device to come would be the M9 medium complemented with 0.1% of glycerol (second medium).