house keeping standard

Tracy Aquilla aquilla at
Tue Jun 21 18:08:31 EST 1994

In Article <940621094812.2040151b at Frodo.MGH.Harvard.EDU>,
>    I am trying to examine relative changes in the expression of cytokine 
>genes in the antigen stimulated human cells.  This is being done by 
>"direct" PCR using labeled nucleotide (not by competitive PCR).
>    Does anyone have any suggestions as which house keeping gene should I 
>use as a standard?  From my initial experiments I know that the expression
>of actin is altered upon antigen stimulation.  So is G3PDH. 
>    Does anyone have experience with beta-tubulin or pyruvate kinase 
>expression under the same or similar conditions? Any other suggestions?
>    Thanks in advance,
>    Andrzej Konieczny
    I have been doing quantitative Northerns for some time now, and I have
spent a great deal of my time pondering this question. Under various
experimental conditions, the steady-state levels of many (if not most) RNAs
change, and it can be argued that there is no "gold standard" to use for an
RNA internal control. I have also found that when I normalize with different
probes as internal controls, the results of the experiment are interpreted
quite differently. I have tried using GAPDH, actins, and rRNAs, and all of
these seem to be influenced to some extent by my treatments, however 28S
rRNA seems to be the best I have found.
    I finally came to the conclusion that it is quite important to select
the best control for the chosen experimental conditions, based on empirical
evidence. I suggest that you try one of the rRNAs, and test whether or not
its absolute level is affected by your experimental treatment. You can also
test whether rRNA levels change relative to mRNAs using an oligo dT probe. I
generally prefer this method, since these markers seem to have the least
variability in my experiments. I hope this helps some.
Tracy Aquilla, Post-doctoral Research Associate
Department of Molecular Physiology and Biophysics
University of Vermont, College of Medicine
aquilla at

More information about the Methods mailing list