He's right to have _some_ concern about resistance, given that wild
relatives of B.subtilis are commonplace (same as E.coli, or most other
free-living mesophiles really) and pretty compatible with subtilis genes.
Of course this certainly fits into the usual "inappropriate use of
antibiotics by farmers/doctors does far more harm" argument, which is
absolutely true. But, all the same I think he's right to take
responsibility for his work.
As to self-excision.. if you can afford the DNA, it's easy. You just
place two LoxP sites in the correct orientation around the DNA you want
to delete, and include the Cre enzyme in that cassette with an inducible
promoter. When induced, the Cre enzyme will trigger excision of the
LoxP-flanked DNA. It will also trigger re-integration of this DNA, but
over time the DNA is more likely to be lost than retained as long as you
maintain the induction system for some generations.
So, your DNA would look like this:
--- Desired Gene ---LoxP--AmpR--Cre--LoxP---
Induction gives this:
--- Desired Gene ---LoxP---
And a circular cutout:
(=AmpR==Cre=LoxP=)
Eventually leading to just:
--- Desired Gene ---LoxP---
Orientation of LoxP sites is important: look up the system and learn how
it works. If you put the sites in the wrong orientation, you'll end up
with a rapidly flipping cassette, rather than one that cuts itself out
of the genome. That's cool and all, but totally useless!
LoxP is, if I recall, somewhat long: about 30bp. In a bacterial genome,
this means it's unlikely to occur by chance, so odds of massive genetic
upheaval due to Cre use are low. You could use any number of
bacteriophage integrases to do the same thing, using the non-identical
integration sites. Efficiency will be higher as the reaction is one-way,
but using Cre does have the advantage of leaving a LoxP site behind that
could be used later for integrating more interesting DNA.. :)
On 16/07/12 22:25, Nathan McCorkle wrote:
> On Mon, Jul 16, 2012 at 5:01 PM, Andreas Sturm <
masters...@gmail.com> wrote:
>> Yeah, but those don't have self-excising marker genes.
>
> I don't think there is any system that is self-excising in use today
> in the synbio world, it could probably be made using a stripped down
> transposon type system though.
>
> That said, if you integrate a marker, it doesn't have to be antibiotic
> resistance. Even if it was, B. subtilis 168 is a lab strain, so it's
> not likely to survive outside the lab. And it's just as easy to remove
> the resistance yourself once you get your gene of interest in the
> genome, you just knock it out using the same integration vector
> technique.
>
--
www.indiebiotech.com
twitter.com/onetruecathal
joindiaspora.com/u/cathalgarvey
PGP Public Key:
http://bit.ly/CathalGKey