"CNS specific knock-outs for cre-lox"

z un691cs at genius.embnet.dkfz-heidelberg.de
Fri Sep 27 06:23:48 EST 1996

In artikeltje  <960926191314_318231441 at emout12.mail.aol.com>,  26 Sep 1996 16:14:10 
-0700,  Neurobiol at aol.com  zei:...

>First, I believe there was a report at the mouse meeting of a cre-lox
>knockout made by crossing a cre transgene driven by the nestin promoter into
>a strain that was knocked out for something (I forget what ) that is lethal
>in embryogenesis when homozygous. 

Wasn't that the genentech mouse made by Heidi Phillips ? I believe it was
NGF they knocked out. these mice died of severe phenotypes in the periphery.
by introducing NGF behind a peripheral promoter they could surpress
that phenotype, and study the brain. Last time i heard this story
is a while back, not sure whether thiswas the actual strategy...

>I've been too lazy to look up the
>expression pattern of nestin, but I gather from your comments that the PNS
>should have been affected, too.  Does anyone have a better knowledge of this
>work?  I wasn't at the meeting and heard it secondhand...

nestin and vimentin are expressed in several peripheral organs, alas.

>Second, Martin, I agree with you that it is very nice but not sufficient to
>restrict the knockout to a particular cell type by breeding to a tissue
>specific promoter-cre mouse line.  

you argue the wrong way round: naturally one tries to make a general
knock-out, but if the animal dies during embryogenisis, or if the
phenotype (say: heart failure after birth) completely covers up the 
changes in the tissues that your are studying, the 2nd step would be
a cre-lox mouse.

>I have a couple of reservations.  The
>first is practical.  Not a lot of promoters are really tissue specific, and
>it doesn't take much cre expression to irrevokably alter a cell that might be
>the first in a lineage of any organ.  I suspect there will be a few
>undiagnosed chimeras caused by this method if people don't look at the
>animals thoroughly. 

True. But again: why does one make a cre-lox ? simply because the major
phenotype of a normal knock-out is so overwhelming, that one needs a 
more restricted approach. Than the weak expression of cre in unwanted
tissues is a calculated risk.

>The second reason I think that the cre needs to be
>delivered temporally is of course the issue you bring up-  that delivery of a
>knockout to a specific cell type- by retrovirus? naked DNA? liposomes?- is
>the next step, AFTER the simple knockout is replaced with the
>tissue-restricted knockout.

this i can't understand. as far as i know one makes two mice: a cre mouse
under nestin promoter, and a mouse with the lox sites flankin' the
gene to be knocked out. when you then cross these two (phenotipically normal)
animals, your gene is knocked out in all those cell lines of which
a progenitor expressed nestin during any stage.
or am i mistaken. I never heard of cre delivery they way you suggest

>Interesting stuff.  I'd like to continue this discussion.  Anyone else?
>Jeanne Loring


More information about the Methods mailing list