Team:ETH Zurich/Wetlab/Genome Reduction

Genome reduction

Goal

Goal of our project is to randomly delete chromosomal fragments of E. coli in order to reduce the physical amount of genomic DNA.

Method

To reach our goal of random deletion of chromosomal fragments, we want to express a frequently cutting restriction enzyme along with the simultaneous, shortly delayed, or even continuous expression of a ligase. The restriction enzyme will cut genomic DNA in a random fashion in vivo, while the ligase performs its job of religation. Assuming that the genomic DNA is cut at several sites within one cell, religation will lead to the exclusion of chromosomal fragments in some cells. Multiple rounds of restriction and religation will therefore lead to a markedly reduced genome.

Proof of concept

In 1997 Ren et al. showed that in vivo religation of linearized vector and insert is possible by overexpression of the T4 ligase (1). Our idea, however, relies on the assumption that in vivo restriction and religation is possible and leads to the exclusion of chromosomal DNA. We are trying to verify these assumptions in several experiments which will from now on be refered to as our “proof of concept”.

Our proof of concept relies the following construct (not shown are ribosomal binding sites behind both SceI restriction sites and a terminator following the RFP):

construct for proof of concept

The above construct was ordered at GeneArt and is supposed to be integrated into the genome of a wild-type E. coli strain (MG1655). We are planning to use the lambda red recombination system established by Wanner and Datsenko (2) to integrate our construct into the tryptophanase A gene resulting in the knockout of the latter.

Additionally, we ordered DNA encoding the T4 ligase and the SceI restriction enzyme. Transformation of these plasmid-encoded enzymes into cells carrying the proof of concept construct is supposed to yield bacteria that can be induced to express T4 and SceI.

Before induction of T4 and SceI, cells carrying the proof of concept construct do not express RFP. After induction of the restriction enzyme, SceI will cut the bacterial chromosome at the sites indicated above. The ligase will then religate the construct, leading to the exclusion of the sensitivity gene and the terminator in some of the cells. Hence, these cells would express the RFP reporter protein and could therefore easily be identified:

proof of concept

Additionally, adding sucrose to the medium would kill all cells that have not eliminated the sacB sensitivity gene. Hence, only RFP-expressing bacteria would survive, which would presumably facilitate detection by several orders of magnitude.

Modification of proof of concept

Unfortunately, up to now we have not received our construct for the proof of concept. Therefore, we are working on a modified construct:

modified construct for proof of concept

For this purpose, a commonly used RFP reporter plasmid is used. SceI restriction sites are introduced by cutting the plasmid in front of and after the RFP gene and ligating oligonucleotide duplexes containing the desired restriction sites. After transformation of this plasmid, the SceI restriction enzyme, and the T4 ligase into bacteria, induction of the enzymes will lead to in vivo cutting and religation of the RFP-encoding plasmid, leading to the exclusion of RFP. Therefore, if cells lose the RFP signal after induction of the restriction enzyme and ligase, we have proven that in vivo restriction and ligation is possible and can lead to the exclusion of DNA fragments:

modified proof of concept

SceI restriction enzyme

SceI is a site-specific homing endonuclease. It is extremely rare-cutting as it recognizes an 18 bp sequence. These properties make SceI a perfect restriction enzyme for our proof of concept for which the restriction enzyme should only cut the sites of our constructs indicated above. However, for our goal of minimizing E. coli’s genome we will need to use a frequently cutting restriction enzyme which will potentially lead to severe damage of the cell’s genome. In order to limit this damage, we want to be able to pulse the expression of the restriction enzyme. Therefore, we are cloning an arabinose-inducible promoter (I0500) also in front of SceI.

T4 ligase

T4 is the commonly used ligase for in vitro cloning. For in vivo cloning, high levels of T4 are advantageous not only for improving the efficiency of religation leading to the exclusion of chromosomal fragments, but also for limiting DNA damage. However, constitutive overexpression of T4 might lead to immediate religation without the exclusion of chromosomal fragments. Therefore, we are cloning T4 behind an IPTG-inducible promoter (R0010), and, as alternative approach, behind several constitutive promoter of differing strengths. Another idea would be to clone both SceI and T4 behind the same promoter, so that induction would lead to simultaneous expression of both enzymes.

References

(1) Ren Z. J., Baumann R. G., Black L. W. (1997): Cloning of linear DNAs in vivo by overexpressed T4 DNA ligase: construction of a T4 phage hoc gene display vector. Gene 22 195(2):303-11.

(2) Datsenko K. A., Wanner B. L. (2000): One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci 97 (12):6640-5.