Team:Newcastle/Stochasticity

=Stochastic Switch=

Introduction
One of the most exciting aspects of our project is our synthetic stochastic switch. The switch regulates the decision to become a non-germinating metal container spore, or a spore that can go on to germinate as part of the normal life cycle. Whilst stochastic oscillators have been implemented before using transcriptional regulators, our switch makes use of an invertable DNA segment to ensure that the decision is heritable.

By differentially controlling the expression of the Hin invertase, we designed our switch to be tunable to achieve a biased heads or tails response, allowing a range of probabilities of orientation of the invertable segment to be achieved.



Novelty in this sub-project
We designed a synthetic stochastic switch by using an invertible segment of DNA flanked by a pair of promoters. Depending on the orientation of the invertible sequence, coding sequences will be expressed which reflect the decision to be a metal container or not. We also tuned the natural stochasticity of the sporulation system towards greater sporulation rates by altering the rate of Spo0A phosphorylation.



BioBrick constructs
There are a number of bricks involved within the stochastic switch construct.

The stochastic brick construct uses the Hin invertase system in order to flip a region between Hix sites. The directionality of the promoter determines whether the switch is 'on' or 'off'. When the promoter is facing right it allows transcription of genes that control:


 * Prevention of germination
 * Upregulation of sporulation rate
 * Expression of the metal sponge (SmtA)
 * Decreased cadmium efflux
 * Upregulation of cadmium import

Importantly, Hin is differentially expressed depending on the levels of the two inducible promoters that flank the invertable segment on which it lies. This means the segment can be biased in a predictable and controllable fashion to favour one orientation or the other.

The following diagram shows our stochastic construct:





Prevention of germination
The prevention of germination is governed by another invertase switch. When the sequence faces right, a FimE protein is expressed which inverts a further promoter region. This promoter controls expression of the cwlD and sleB genes. If their promoter is in the correct orientation then the cell will be able to germinate and continue as a vegetative cell. However if their promoter has been flipped, the cell can not germinate following sporulation, and will be trapped as a metal containing spore.

Upregulation of sporulation rate
The upregulation of sporulation involves increasing KinA expression. kinA codes a kinase protein that phosphorylates the Spo0A protein to its active form. When the promoter region within our stochastic brick faces right, there will be increased KinA expression, and thus a greater sporulation rate.

Metal sponge and cadmium influx/efflux
Our stochastic switch determines whether the spores can germinate, or whether they are commited to be metal containers that cannot germinate again. We need this switch as we cannot totally interrupt the natural life cycle of the bacteria, since a proportion of cells have to go on to seed the next generation. Expression of the metallothionein fusion protein (cotC-gfp-smtA), cadmium import channel (mntH) and the cadmium efflux channel (cadA) is also governed by the direction of the stochastic promoter. When the direction of promoter faces right, the metallothionein fusion protein's expression will be triggered, ant will soak up the cadmium. While the import channel is upregulated, the efflux system's activity will be slowed down to increase the amountof cadmium inside the cell.

Stochastic Brick
We decided to get our stochastic construct synthesised, as trying to build the construct manually would be too time consuming. The following sequencher diagram shows the components of the construct we had synthesised.

Testing construct
In order to test our construct we had to redesign using inducible promoters governing Hin invertase expression. We used the promoters pSpac and pxylA (Induced by IPTG and Xylose) to test our system. We include cut sites around these promoters in order to replace them with SigmaA promoters once the construct has been characterised.(See sequencher diagram above)

Degradation controller
In order to have another level of control over the orientation of the promoter within the flipping region we added a degradation tag to the Hin invertase protein. The following paper describes how proteins including modified ssrA tags can be located to the ClpXP protease by an Sspb protein. This means that inducible Sspb expression can requlate degradation levels of the tagged protein.

Inducible protein degradation in Bacillus subtilis using heterologous peptide tags and adaptor proteins to target substrates to the protease ClpXP

We decided to put the Sspb protein under the control of an arabinose inducible promoter as the following diagram illustrates. We also included a region of the sac gene in our construct, so that the region will integrate into the Bacillus genome at a region other than amyE.



We added a modified version of ssrA degradation tag onto the C-terminus of the Hin protein. Expressed proteins are therefore degraded by ClpXP. However mutations on the ssrA tag weaken the recognition by ClpX, and the modified tags require the SspB adaptor protein to be recognized. When the SspB protein is expressed the proteins tagged with modified version of ssrA tag are targeted for degradation. Otherwise they remain stable.

In B. subtilis there is no sspB orthologue and SspB from E. coli works in B. subtilis. By regulating the levels of SspB by arabinose, we implemented an inducable protein degradation device.



The wild type E. coli ssrA tag is AANDENY-ALAA (SspB recognition site – ClpX recognition site). As suggested in the paper, we took one of the modified ssrA tags to use in our system.

AANDENY-SENY-ALGG (SspB recognition site – SENY +4 Linker - ClpX recognition site)

This tag works well in B. subtilis. However, degradation tags can affect the activity of proteins. Different degradation tags may effect the activity of different proteins. It has been shown that this tag effected the activity of ComA(1).


 * 1) Griffith, K. L., and A. D. Grossman. 2008. Inducible protein degradation in Bacillus subtilis using heterologous peptide tags and adaptor proteins to target substrates to the protease ClpXP. Mol. Microbiol. 70:1012-1025.

Stochastic Modelling Tools
Matlab can be used for stochastic modelling. The Glasgow team used Matlab to implement the Gillespie algorithm for incorporating noise among cells. They also used deterministic modelling using ODEs and compared their results. When the number of cells increase two approaches become similar since the noise is cancelled out.

Stocks 2 is another stochastic simulation tool which also uses Gillespie’s direct method and supports SBML.

We used computational modelling in Matlab to try to determine how to make our system tuneable.

Please see our modelling page for Matlab files on our stochastic switch model.

FimE switch
The FimE switch is a similar switch to the Hix system. However, it acts as a latch, meaning that once flipped the segmant will not flip back. A Tightly Regulated Inducible Expression System Utilising the fim Inversion Recombination Switch.(E. Coli) Timothy S. Ham, Sung Kuk Lee, Jay D. Keasling,Adam P. Arkin,Received 21 December 2005; accepted 2 March 2006 Published online 13 March 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/bit.20916
 * 1) fimE switch for DNA re-arrangement

We decided to use FimE to switch off or on the production of a protein of our choice, such as the genes involved in germination.


 * 1) Control of the Arabinose Regulon in Bacillus subtilis by AraR In Vivo: Crucial Roles of Operators, Cooperativity, and DNA Looping
 * 2) Binding of the Bacillus subtilis spoIVCA product to the recombination sites of the element interrupting the sigma K-encoding gene =>...DNA rearrangement that depends on the spoIVCA gene product...

Bistability in Bacillus subtilis
Read this page to find more options for natural stochastic switches in Bacillus subtilis. Natural stochastic switches:Bistability in Bacillus subtilis

And to find out how we are tuning sporulation using our stochastic switch choice see the sporulation tuning page.

Lab strategies
To carry out our labwork we needed cloning strategies for all of our bricks and devices. Please see our cloning strategies page for details on how we cloned our devices.