Human Quant. PCR cDNA Control

Eivind Hovig ehovig at
Tue Feb 14 17:04:30 EST 1995

In article <3hq92n$q09 at>, bltihi at (Timo Hiltunen) wrote:

> Pat Charmley (patcharm at wrote:
> : Dear Everyone,
> : I'm looking for a good PCR primer set for a human "structural gene" that 
> : can be used as an internal control in my cDNA PCRs from various tissues.  
> : I've read that actin is inducible, and thus I'm not interested.  I've 
> : heard that GAPDH is a good and stabley expressed gene.  Does anyone have 
> : a good reference, or personal experience?
> GAPDH is not stably expressed in the samples that we have analysed.
> Up to 5-fold induction seems to be possible.
> Anyhow, GAPDH has been rather widely used, and in some cases may be stably 
> expressed.

And my opinion, although it does not weigh all that much, is that it will
not be possible to find such a stable cDNA. And it will anyway not be
possible to ***prove*** that anything is stable, as long as there is no
stochiometric relationship known between a given gene copy number, its
regulation, and cell status. What one can do is either pretend that such a
relationship exists ("a number of people are are using gene X, therefore
it must be usable, or at least referees will likely not mind"), or test a
number of high, middle, and low expressed transcripts to get a general
impression of cellullar status, or something like that. The same goes for
northerns. And consider a 5-fold variation: how will such a variation give
sensitive results when covaried with often very unreliable
semiquantitative rtPCR assays like limited cycle number. Not exactly
precise science...

Eivind Hovig PhD  email= ehovig at  Dept. of Tumor Biology
S=EHOVIG; OU=radium;  O=uio; ADMD= ; PRMD=no   Inst. for Cancer Res.
Phone:-47-22935416                         The Norwegian Radium Hosp.
Fax:  -47-22522421                             0310 Oslo, Norway

More information about the Methods mailing list