Northern Normalization

Przemko Tylzanowski przemko at sgi.celgen.kuleuven.ac.be
Wed Aug 21 02:24:14 EST 1996


Heinz-Juergen Schaefers wrote:
> 
> Susan Patrick wrote:
> >
> > I was hoping to use an internal standard, a so-called constituitively
> > expressed gene, to normalize my northerns.  I have read some about how
> > much these genes may vary in some cells, during differentiation, etc.
> > Any suggestions on how I should go about insuring loading consistent
> > amounts of RNA?  (I am working with differentiating F9 (murine) cells.)
> 
> Hi Susan,
> 
> the best internal standard for RNA loading is 18s or 28s rRNA.
> 
> bye
>         Heinz

Well, that assumes of course work with total RNA, not PolyA+. Now, the
problem of internal standard is, IMHO, an unsurmountable one. The
problem lies in the fact that, no matter what you use as as standard,
you have to be sure that "it" is not regulated under your experimental
conditions. How do you do that, well you compare it to something
else...See the loop now? It becomes almost Turing's halting problem.

If you work with total RNA, ribosomal probes are indeed the best. Since
ribosomal RNA takse up to 80% of total, that kind of normalization makes
sense. Of course, you have to be sure that RNAs you put on the gel were
isolated by the same method. CsCl method gives you more ribosomal RNA
than Chomczynski, so you should not load on one gel RNA isolated by
different procedures.

Anywa, my 2c worth

Przemko



More information about the Methods mailing list