Recombineering refers to the strategic use of recombination in vivo in order to reach a defined goal. In the case of BioBytes, a method is required to target the final construct to insertion at a specific place on the E. coli chromosome.

To do this successfully, three components must be taken into account:

• There must be a system for targeting the construct to a specific site for insertion

• Activation of the recombination system must be under experimenter control

• It must be possible to select for and verify colonies in which the insertion was successful

The BioBytes team has chosen to use a recombination system from bacteriophage lambda. Lambda Red recombinase specifically recombines on the ends of a linear fragment of DNA. If the ends of this fragment are homologous to two separate sites on the E. coli chomosome, the genetic material between these two homologous regions will be exchanged. This will be the basis for our targeting system.

The homologous regions must be a minimum of 50 base pairs in length for recombination to occur at a significant frequency. This can be achieved in different ways:

First 5' extensions corresponding to the homologous sequence can be added to any gene using PCR amplification. This would allow a PCR product to be targeted to a specific site for insertion. Because our constructs will be recircularized and grown (see DNA assembly), this would require us to PCR each plasmid construct seperately in order to add the homologous regions to the ends.

An alternative is to use the genes on either end of our construct as the homologous regions. As an example, we could first locate a region of genes which were deemed inessential through literature review and our Matlab modelling. This region would necessarily be flanked by an essential gene at either end. We would then assemble a plasmid containing these two essential genes. If the insertion is successful, we would be left with a chromosome without this region of inessential genes.

'''Attempt I:'''

Our first attempt at recombineering entailed the insertion of an ampicillin resistance cassette into a gene-less region of the host chromosome. Amp cassette primers were engineered with 50 bp extenstion that were homologous to flanking portions of the region of the chromosome to be excised. The linear construct produced by these primers was then electroporated into pre-Arabinose pKD46 containing competent cells. Cells were left to incubate at 30C for four hours, then plated on ampicillin containing agarose plates.

Colonies with proper recombination were selected for by PCR verification. PCRing across the excised region was expected to show markedly different fragment sizes between wildtype and ampicillin resistant cells, as the ampicillin cassette was a substantially different size than the region excised.

Upon PCR verification, we saw no difference from wildtype in ampicillin resistant cells. This has lead us to believe that recombination had been delocalized.

'''Attempt II:'''

In order to increase the degree of non-specific recombination, we attempted to insert a linear construct with whole-gene homology to the chromosome on either side of a chloramphenicol resistance cassette. The region selected for excision contained no known essential genes.

Our 5 byte construct was built using our Biobyte system and looked like:

• orn - 75% promoter - amp - 75% promoter - yjeE

The gel purified construct was once again electroporated into pre-Arabinose pKD46 containing competent cells. Cells were left to incubate at 30C for four hours, then plated on chloramphenicol containing agarose plates. This time no colonies grew.

This shows that either recombination did not occur, or the cassette was non-functional. We did not have time for further trouble-shooting.