A new model for chromosomal evolution:

ELIMINATING ANTIBIOTIC RESISTANCE

We constructed plasmids pSB6A1-T7ccdB (for use in CIChE)and pSB1C3-T7ccdB (to submit in the BioBrick Registry).Through these plasmids, pressure can be put on the CIChEstrains simply by adding IPTG. All plasmids we obtained andsequenced had mutations in the last part of the ccdB gene.We performed back mutations to restore the ccdB genes.

We also disposed of three other plasmids, all having differentcopy numbers. In this way, we were able to test CIChE withdifferent degrees of toxin pressure.

All plasmids we obtained were transformed into the straincontaining the CIChE construct using electroporation.

Plasmids containing T7ccdBKnock-in CIChE construct

A construct containing the homologous regions, theantitoxin ccdA and gfp was inserted in the genome of an E.coli strain using the method of Datsenko and Wanner withLambda Red recombinase.

The E. coli strain must contain recA which is necessary toperform gene duplication by homologous recombination.

Our goal is to improve an existing technique,called chemically inducible chromosomalevolution (CIChE).CIChE raises the yield of biochemicalproducts by incorporating and duplicating agene of interest in the bacterial genome. Thistechnique, however, entails one majordisadvantage: the creation of bacterialstrains with a large amount of antibioticresistance genes. We have tried to eliminateCIChE’s need for antibiotic resistance genesby using a toxin-antitoxin system.

Our Goal

When the desired gene copy number is reached recA is deleted, thereby fixing the copy number. The main disadvantageof this system is that after chromosomal evolution a strain with multiple antibiotic resistance genes is obtained. Thecreation of such a strain raises several safety concerns as these resistance genes can be passed on to other, possiblypathogenic, bacteria in the environment through horizontal gene transfer.

‘Chemically inducible chromosomal evolution’ or CIChE isa technique for the stable, high copy expression of a geneof interest in E. coli. It was developed in 2009 by Tyo et al.In CIChE, a construct containing the gene of interest andan antibiotic marker flanked by homologous regions isintegrated into the E. coli genome. The construct can beamplified in the chromosome through recA homologousrecombination. The strain containing the CIChE constructis cultured in increasing antibiotic concentrations,providing a growth advantage for cells with increasedrepeats of the construct and thereby selecting for bacteriawith a higher gene copy number. This process is calledchromosomal evolution.

Meet the team

The students: Eva, Eline, Renate , Tine , Jules, Tim & Marlies

Our instructors:Prof. dr. ir. Marjan De Mey,

ir. Pieter Coussement, ir. Frederik De Bruyn, ir. Gert Peters & Charlotte Verniers

GOIM

antibiotics

Transduction

GO

IM

= Gene of interest

= Marker gene

= Homologous region

nGOIMGOIMnGOIMGOIM

@ iGEM UGent - BBEu PP

@ IGEM team UGent- Bio Base Europe Pilot Plant

2013.igem.org/Team:UGent (scan our QR code!)

Our new model for CIChE

We performed CIChE with plasmids with copy numbers ranging from five to twenty(pSB6A1-T7ccdB, p5SpFRT-T7ccdB, p10SpFRT-T7ccdB and p20SpFRT-T7ccdB). To evaluateour strains at different points during the CIChE process, we measured OD600 (opticaldensity measured at a wavelength of 600nm) and fluorescence intensity (F). Strains withmore copies of the gfp gene should result in higher fluorescence intensity, indicating ahigher construct copy number. Because fluorescence also depends on OD600, we dividedfluorescence by OD600 (F/OD600) to analyze our results. Finally we plotted F/OD600 infunction of the IPTG concentration. In every graph the bar on the left represents theaverage F/OD600 of a reference strain having only one gfp copy. If our model for CIChEworks, strains obtained after CIChE (containing more gfp copies) should have an averageF/OD600 higher than the one of the reference strain, raising with increasing IPTGconcentrations.

Looking at the graphs we see that most of our strains result in an average F/OD600 ofabout the same as the average result of the reference strain. According to these data, wemust conclude that no duplication of the construct occurred, hence no chromosomalevolution was achieved. We presume that it is the malfunction of CcdB due to mutationsthat lead to these unexpected results (see CIChE and its stumbling blocks).

Results

CIChE

We used a CcdB/CcdA toxin-antitoxin system in our new modelfor CIChE. CcdB interferes with DNA gyrase and is used to putpressure on the cells. CcdA can inhibit CcdB from performing itstoxic function by forming a CcdA-CcdB complex.

CcdB is added to the cells by placing it on a plasmid behind aninducible T7 promoter. In this way CcdB concentrations inside thecell can be regulated by adding different concentrations of theinducer IPTG in a range from 0 mM to 240 mM. The ccdA gene isplaced inside the CIChE construct next to the gene of interest,instead of the antibiotic resistance gene. Increasing CcdB pressurestimulates the cell to duplicate the CIChE construct containingccdA. Through construct amplification the desired expression levelof the gene of interest can be obtained. Using this model,chromosomal evolution can be achieved without using antibioticsor antibiotic resistance genes.

Stumbling blocksIt is known from previous research that the partessential for CcdB’s toxic activity is situated at thelast three amino acids (Trp-Gly-Ile) which interactwith the DNA gyrase A subunit (GyrA). Sequencing ofT7ccdB plasmids at several steps during our researchall showed a deletion either in this toxic part, or in

the stopcodon of the ccdB gene. Both deletions cause a frameshift, resulting in a prolongedprotein of 154 amino acids instead of the native structure with 102 amino acids. In the firstcase, the amino acids responsible for CcdB’s toxic activity are mutated. We designed modelsfor the protein structure that, put together with literature, strongly suggest that interactionwith DNA gyrase is no longer possible with these mutated CcdB proteins.

We believe the high mutation frequency of the ccdB gene during our project has tworeasons. The first is leaky expression of the T7 promoter during cloning. Cloning wasperformed in Top10 cells which do not contain T7 RNA polymerase. Taking into account thetroubles we had with the construction of the plasmids with the T7ccdB insert, it seemshighly likely other polymerases which are not specific for the T7 promoter were able totranscribe some ccdB. This probably only occurs at very low frequencies, but as in thesestrains there is no CcdA present at all, even the lowest amount of working CcdB can belethal for these cells. This means that random mutations in the part of the ccdB geneassociated with its toxicity will provide cells with a great growth advantage over cells withintact ccdB genes.

Another factor contributing to this apparently high mutation rate is that CcdB causesactivation of the SOS pathway when not enough CcdA is present. One of the aspects of thisSOS pathway is enhancing the capacity of mutagenesis. We postulate that these mutationscan also affect the ccdB gene itself.

Red: native CcdB structureYellow: one of our mutated Ccdb

structures

gfpccdA

T7 RNAP

T7 RNAP

ccdBT7

gfpccdA gfpccdA gfpccdAn

CcdA-CcdB complex

IPTG

gfpccdA

ccdB

T7T7

RNAP

Interaction of intact CcdB (above) with DNA gyrase (below)

F/OD600 for three strains (copy number 5, 10 and 15-20) at different points during chromosomalevolution. The reference gene contains one gfp copy. Other bars: F/OD600 of strains during CIChE atseveral IPTG concentrations.