Knocking DNAs in and out at random with transposons, and moving genome loci between strains by phage transduction or conjugation
Bacterial and Bacteriophage Genetics — ISBN 978-3540942702
Conjugation, Transposition, and Transduction are old-school ways of performing genome manipulation that pre-date recombinant DNA methods. Despite their age, they are still commonly used techniques. Conjugation involves the transfer of DNAs from one cell to another through mating. Transduction involves the transfer of DNAs from one cell to another via a phage particle intermediate. Transposition moves a DNA to a random position in the genome.
Section divider: Conjugation, Transposition, and Transduction. Knocking DNAs in and out at random with transposons, and moving genome loci between strains by phage transduction or conjugation.
Conjugation
For a strain that will not take DNA at all
Some organisms cannot be transformed or electroporated, and no amount of technique fixes it
Put an oriT on your plasmid — that is the only change to the DNA
Transform it into a donor E. coli carrying the RP4 machinery — usually WM3064
Mix donor and recipient on a plate, and the plasmid is donated cell to cell
Then select for the plasmid and against the donor
This is the one to reach for when a strain simply will not take DNA. Transformation and electroporation both work by damaging the envelope enough to let DNA through, and there are plenty of organisms where that either does not work at all or kills the cell first. Conjugation does not go through the outside of the cell in that way: the donor makes physical contact and the DNA is handed across. And the part that makes it usable is how little you have to do to your own construct.
Conjugation, in five points: some strains cannot be transformed at all; put an oriT on the plasmid; transform it into a donor carrying the RP4 machinery, usually WM3064; mix donor and recipient on a plate so the plasmid is donated cell to cell; then select for the plasmid and against the donor.
Transposons
For putting DNA in the genome when you do not care where
Random insertion, and that is the point — one library of strains, each with the payload somewhere different
Put terminal repeats on the ends of the DNA you want in
Mix with transposase, transform, and it inserts wherever it lands
Or encode it on a plasmid and conjugate it in — for a strain that will not transform
In human cells the same job is usually viral — lentivirus integrates at random
Transposons are for the case where random is what you want. If you are asking which gene does something, or hunting for an insertion that gives you a phenotype, you do not want one defined edit. You want ten thousand strains with the DNA in ten thousand different places, and then a screen. Everything else in this lecture has been about control over where the DNA goes; this is the one method that deliberately gives that up. The picture is worth walking: the transposase grabs both terminal repeats and holds them against each other, which forces everything between them into a loop, and that complex of enzyme and DNA is what is called a transposome. It is a real thing you can buy in a tube, preloaded, and electroporate straight into cells. Once it is in, it cuts at both repeats and pastes the whole loop somewhere else, and nothing about the choice of somewhere else is yours.
Transposons, in five points: random insertion is the point; put terminal repeats on the ends of the DNA; mix with transposase and transform, and it inserts wherever it lands; or encode it on a plasmid and conjugate it in; and in human cells the same job is usually done with a lentivirus. Beside them, a three-panel figure: the construct with a terminal repeat at each end, then transposase gripping both repeats so the DNA between them loops out, which is the transposome, and then the transposon pasted into the middle of a gene in the genome.
P1 generalized transduction
Moves ~90 kb chunks at a time · some strains, DH10B among them, are P1 resistant
P1 generalized transduction involves the transfer of random DNA fragments from one strain of E. coli to another. First, the donor cell is infected with P1vir phage. When a lytic phage infects a cell, usually it will package its own DNA. However, the process incorporates a piece of genomic DNA instead around 2% of the time, and the sequence incorporated into the phage head is a random one. The cell lysate is then sterilized with chloroform to kill any un-lysed bacteria and then added to a sample of recipient cells. For the rare phages that contain a piece of genome, this DNA will be injected into the recipient cell where it will undergo homologous recombination with the recipient genome. If that fragment of DNA contains a selectable marker, these transduced cells can be selected. P1 transduction moves 90kb random chunks of the genome. It is thus useful for transferring single genes or even large gene clusters from one genome to another. Often people will use homologous recombination or phage att integration methods to first modify one cell's genome, and then combine several such mutant genomes into one using P1 transduction. Note that some of the popular E. coli strains are P1 resistant.
Two cells side by side, a donor on the left with a marker in its genome and a recipient on the right.
glgS troubles
Susan did a markerless glgS knockout of MG1655, a second knockout with CmR of the rfaE gene in another MG1655, and then tried to move the rfaE::Cm marker into the glgS knockout strain by P1 transduction. Her products only have rfaE knocked out. What happened?
In PubMed, protein-search each gene, then navigate from the protein to the CDS to the GenBank sequence
Look up where the two genes actually are:
glgS in MG1655 NC_000913 REGION: 3191739..3191939
rfaE in MG1655 NC_000913 REGION: 3195320..3196753
You cannot move one without the other. The fragment that brings rfaE::Cm in brings the donor’s wild-type glgS with it, so selecting chloramphenicol selects for exactly the event that undoes her first knockout.
Give them a moment with this one before clicking. Everything needed to answer it is on the slide already, plus one number from the transduction slide a few clicks back.
A question for the room about a P1 transduction that lost a knockout, with the hint that you should look up where the two genes sit on the genome. The answer arrives in two steps.
Questions?
That is the lecture, and it adds up to the seven names on the map at the start. Four ways to put DNA into a genome: a phage att site, homologous recombination, CRISPR forcing that recombination, and a transposon when you do not care where it lands. One way to take the marker back out afterwards, which is Flp. And two ways to move a change between strains once you have made it: conjugation, and P1.
A slide reading Questions.
