Endy:Chassis engineering/Orthogonal protein synthesis

Dedicated systems

An independent protein production channel, dedicated to the synthesis of engineered system proteins would contribute to decoupling the operation of systems and their cellular chassis

.

Engineered biological systems use host cells as a power supply and chassis. In an ideal chassis-system relationship, the operation of the chassis should be decoupled from the operation of the system and vice versa. This is not the case for today' engineered biological systems. A prime example of where this ideal relationship is not being achieved is the protein production process for engineered systems. Engineered systems use the same transcription and translational machinery as the cellular chassis. The engineered system and the chassis also share a pool of nucleotides, amino acids, and tRNAs. Variations in machinery and material levels affect the operation of the engineered system and high demands for machinery and materials by the system perturb the function of the chassis.

A solution to this problem of the coupled function of the engineered system and chassis is to construct a second protein production system inside the chassis, which is independent from, or orthogonal to, the endogenous protein production system. Ideally, this dedicated system channel would use transcription, translation machinery that is orthogonal from that of the cellular chassis and also draw from a different material pool than the chassis. The levels and regulation of the components of the dedicated system channel could be modified independent from the levels and regulation of the protein production system of the chassis itself.

If this dedicated system channel could be standardized and designed to be able function in multiple cell types, then the dedicated systems could form the basis of a biological virtual machine

Currently, we are working to enable dedicated transcription and dedicated translation in an E. coli strain MG1655 chassis. Details on this work can be found on the following pages -

• Dedicated Transcription
• Dedicated Translation

Reporter devices for the transcription/translation systems

	General Translation	VM Translation
General Transcription	I7101	I7102
VM Transcription	E7104	E7103

VM Transcription

Introduction

Implementing Dedicated Transcription

The T7 expression system developed by Studier and coworkers is essentially orthogonal from the E. coli transcription system. T7 RNAP does not recognize E. coli promoters and E. coli RNAP does not recognize T7 promoters.

Effects of producing dedicated transcription machinery in a cellular chassis

Growth curves for three colonies of BL21(DE3) in the presence or absence of maximally inducing levels of IPTG.

Three colonies of BL21(DE3) were grown in the presence or absence of 0.4mM IPTG. The IPTG induces the production of T7 RNAP. Results suggest that the presence of dedicated transcription machinery in BL21 does not noticeably affect the growth rate of the cellular chassis.

While this implies that the presence of a dedicated transcription system has negligible effect on a cells ability to supply a demand, it is possible that the added demand of these systems pushs the cellular chassis much closer to its maximum demand level before physiological changes become evident.

GFP production from T7 promoters of varying strength

Time courses of GFP accumulation per cell using promoters of different strengths

• I built GFP reporter systems that were driven by T7 promoters of various strengths, based on the in vitro data of Imburgio et al..

• As expected, the initial accumulation rate of GFP is qualitatively in keeping with the in vitro promoter strengths. It is possible that there are saturation effects evident as the accumulation rate in the two strongest promoters is very similar. What was less expected was that the accumulation rate after 3 hours of induction dropped off significantly for the strong promoters and much more slowly for the weaker promoters.

• The suggestion has been made that the transition to stationary phase affects the strength of the promoters. This seems unlikely, since the T7 RNAP, which is orthogonal to E. coli RNAP, should have no interaction with stationary phase E. coli sigma factors, nor is there any reason for there to be inhibition of the T7 RNAP during stationary phase. Another point that is important is that the four promoters all differ from the consensus sequence (as used E7104) by just one base. This makes it unlikely that one of them has a cryptic E. coli promoter or stationary phase sigma factor binding site.

• I need to repeat this data again, and fit a trendline to the initial accumulation rates to see if they are quantitatively in keeping with the in vitro data. It might also be good to run the samples in a chemostat and see which has the highest level of GFP accumulation.

Dedicated Translation

Dedicated systems/Dedicated translation

RBS design

The first RBS I designed was BBa_B0036. This part uses the RBS described by Brink et al.. The spacing between the RBS and the start codon was set to be the same as that of the consensus E. coli RBS sequence.

The RiPS from BBa_B0036 are not as high as I would like. Hence I am designing a new RBS. For this new RBS, I will use the spacing between RBS and start codon used by Brink and coworkers. I have obtained high levels of translation using this spacing as shown in the protein gel below using the reporter construct pSD-rbsK2 (obtained from Chris Hayes)

To increase the spacing, I am using the same 5bp sequence as used by Brink et al. This sequence doesn't appear to form a restriction site (it is the first 5bp of an XbaI site) or an E. coli RBS.

Current Work

1. Demonstrate specificity of combined dedicated systems - Combinations of dedicated systems should allow highly specific production of proteins with no non-specific protein production by E. coli systems. This can be demonstrated by stably maintaining a plasmid-borne copy of a highly toxic gene in a chassis. It should be possible to turn on high level expression of this toxic gene by inducing the dedicated systems.
   • The current candidates for a toxic gene are CCDB, T7 gene 5.3 or T7 gene 7.7.
   • If you have a gene you would like to stably maintain with zero expression, let me know and I'll have a go!
   • A quick check to see if this will work can be done by looking at the expression level of GFP when the dedicated systems are not induced. The initial attempt to show this didn't look good, expression level of GFP from E7103 was the same as E7104 and both were a little bit higher than BL21(VM1.0) with no GFP coding plasmid. This may be due to the fact that the expression of the dedicated ribosomes required by E7103 is leaky due to the P_BAD promoter. To get around this, I'm going to redo this experiment using glucose as a carbon source to further repress the promoter.
2. Demonstrate decoupled function of the system and the cellular chassis - While the above point shows that the dedicated systems are highly specific, the real objective of dedicated systems is to decouple the function of system and cellular chassis, such that perturbations to one should not be transmitted to the other. For example a sudden increase in the protein production rate of the system should not cause the chassis to stop growing. Alternatively a reduction in the growth rate of the chassis due to a scarcity of nutrients or another environmental effect should not affect the rate of protein accumulation in the system.
   • GFP accumulation as chassis make the transition from log to stationary phase should indicate whether system performance is less affected by the change in the chassis state.
   • Suddenly turning on the high level protein expression of a system should reduce growth rate of the chassis when the system uses dedicated systems than when it does not. Initial experiments have suggested that this might only be true in a minimal media such as M9.
3. Better understanding of dedicated system performance - I'm still trying to understand some of the details of how the systems are operating.
   • The initial rate of GFP accumulation seems to be same when using dedicated translation and chassis translation. Since there was no attempt to make sure that there were the same numbers of ribosomes available for each or to make the RBS strenghts the same it is unlikely that the translation rates for reporters using both should be the same. The most likely explanation appears to be that some factor downstream of translation machinery is limiting reporter production. Here are some possibilities -
     • Supply of charged tRNAs. This would assume that there are large numbers of transcripts and large numbers of ribosomes translating the messages such that the production is being limited by the supply of charged tRNAs.
     • Alternatively, translation may not be limiting at all but some step related to folding of the GFP peptide is limiting. Not sure what chaperones are involved in GFP folding and maturation.