PCR & Housekeeping genes: An undigested summary

From: Mario Roederer (Roederer@drmr.com)
Date: Wed Jul 05 2000 - 22:12:06 EST


As usual, the community was extremely responsive in my query to using
housekeeping genes for RT-PCR.  Since so many people wrote to me
asking for a summary of the responses, I present here slightly edited
versions of most of the EMails.  Some of these were already sent to
the Cytometry list, but I thought it would be useful to list them
all.  I chose to remove personal contact information in case some of
the people providing advice preferred anonymity by default; if anyone
would like to get in touch with a specific contributor (most of whom
offered to provide more detailed information upon request), I can
EMail your request on to the appropriate person.

Again, thanks to everyone for their advice, especially on this
slightly off-topic query.

mr

>In the last experimental design evaluating TAP1/2 mRNA (TAC-MAN) and nef- vs.
>nef+ gene constructs it appeared that transferrin or GAPDH mRNA was
>superior to
>beta-actin.  This was measured on PBMC's stimuated with PHA/IL2 over
>14 days and
>infected over 6 days.  I know the GAPDH does take advantage of the splice
>juction but I am not sure about the others.


>I have used G3PDH (glyceraldehyde 3-phosphate dehydrogenase) for semi
>quantitiave PCR using RNA.  The gene contains introns so you can design
>primers that overlap the junction in order to amplify RNA products and not
>genomic DNA.  This worked very well for the system that I was using at the
>time (cytokine receptor transcripts in stimulated PBMCs).  I have used it as
>a control in several PBMC populations (T cells, B cells, macrophages and
>neutrophils) and have not seen any amplification even using strong mitogens
>such as PMA/ionomycin.


>we use GAPDH and L32 for our RNAse protection assay.


>We have found glycerol-3-phosphate dehydrogenase to be a more suitable
>housekeeping gene that beta-actin for quantitative RT-PCR.  It is
>interesting that the primers we use amplify cDNA from a number of
>different species.  I am pretty sure our primers do take advantage of
>spanning an intron or a splice junction as you worded it.  We have found
>similar levels of expression on an cellular basis regardless of whether
>the cells are unstimulated in culture or stimulated with mitogens or
>other stimuli.  Of course, this was also true of beta-actin, but not
>quite as consistent.


>We are using the GAPDH
>housekeeping gene and DO take advantage of an intron to be able to
>distinguish genomic DNA and  cDNA  PCR amplification products. It is my
>understanding that the sequences and primers were originally designed by
>Francois Villinger. They're based on the human GAPDH gene (but are
>conserved in rhesus). They are also using a different GAPDH primer
>set for Taqman amplification and we're currently developing more Taqman
>probes for a number of genes to follow during SIV infection (including
>viral RNA/DNA) (no other housekeeping genes as far as I know). Hope this
>helps.


>Try looking at 18S or Gus,  ABI sell Taqman Endogenous Control using the
>7700 Taqman gene expression  technology.  Look at the ABI site, PEbio.com
>under PCR   -PDAR's


>RT-PCR  is  performed with the Superscript One-Step RT-PCR kit according to
>the  instructions of the manufacturer (GibcoBRL). In a first reaction tube,
>the  total RNA is "titrated" by measuring the amount of 28S rRNA (ribosomal
>RNA)  with  primers  specific  for  the  28S rRNA after addition of a known
>amount  of an internal standard RNA which is co-amplified with the cellular
>28S  rRNA  but  yields  a  different  (shorter) amplification product. This
>internal  standard  is  a custom construct obtained through a local company
>called  SwissClone.  The  RT-PCR  products are then separated on an agarose
>gel, stained with ethidium bromide and a digital picture is obtained with a
>Kodak  DC-40  camera. The intensity of the two bands (28S rRNA and internal
>standard)  is  measured  on  the  digital  picture file using Kodak Digital
>Science  1D  software and the R1 ratio calculated as follow: R1 = intensity
>of the 28S RNA signal / intensity of the internal standard signal.
>In  a  second  tube,  the  IL-1ß  mRNA (for example) is amplified by RT-PCR
>together  with  a known amount of internal standard (pAW109 RNA from Perkin
>Elmer  in  this case) which is added in the same tube as internal standard.
>Again,  two  distinct  amplification products are obtained, separated on an
>agarose gel, quantified using the Kodak system and the R2 ratio calculated:
>R2 = intensity of the IL-1ß signal / intensity of the pAW109 signal.
>The  relative  IL-1ß  mRNA expression is then deduced by computing R2 / R1.
>The results are expressed in % of the negative control expression.


>  The PDH beta subunit (Accession No
>D90086) is a good one.  The mRNA of this gene is constitutively express in
>lymphocytes and suffers minor changes after activativation.  I have used
>primers: GTTTGGCAGCGGATAGAGGACACG and CATTCTCTAGCACCACCACTGGAT that I
>designed myself with the aid of the DNA Star program.  The amplicon size is
>637 bp (spanning at least one intron)


>At the moment we are amplifying the abl message and found the information
>present in "Debate Round-table: Multiplex PCR for quality control of
>template RNA/cDNA in RT-PCR assays.  F Watzinger and T Lion, Leukemia year
>1998, page 1984-1986 (can't find the volume reference)" very useful: these
>people report on the use of bcr, abl, b2-MG and PBGD (porphobilinogen
>deaminase) in a multiplex assay!


>We are using the ABI Prism 7700 Taqman system for our QPCR.  This assures
>more reliable quantification and reduced workload compared to classic
>competitive PCR methods (which we still use for some targets).  The demands
>that we make for housekeeping genes are the following:
>- the expression level should be similar between the control gene and the
>target gene
>- expression of the control gene is not effected by therapy, infections,
>etc...
>Besides this we try to avoid genes for which many pseudogenes are known
>(such as beta-actin) and we always place our primers over an exon junction.


>I use GAPDH, but have primer designs for pyruvate dehydrogenase and histone
>3A. One potential problem in GAPDH is the presence of pseudogenes although I
>have not personally experienced this problem.
>We do not have a 7700 so one work-around we have used with partial success
>has been the Ambion competimer strategy. About equivalent to our success
>with competitive PCR (not good enough). However that took considerable
>fiddling to get it right (which Ambion blamed on the rarity of our target
>message)
>Beta 2 microglobulin was recently suggested to me on the basis that ALL
>cells express it (even my meso/epithelial cells where it too is a rare
>message).


>GAPDH is my choice for a good housekeeping gene.  Yes, I always try to use
>primers that span an intron (or a few) if I can.  Also, you could run a PCR
>reaction with RNA only to see what level of genomic DNA contamination you
>might have.


>in terms of housekeeping gene controls, for mouse gapdh is
>pretty standard and in human b2 micro globulin (i'm pretty sure, it's
>been a while). someone's email response mentioned beta actin, we've
>found that it is more highly regulated than gapdh or b2m. all exon
>probes. for the determination of gene regulation as well as specific
>cellular gene expression you have to use different cells, types and
>lines, under different but morphologically, phenotypically, or
>genetically defined conditions.


>I have just finished "troubleshooting" housekeeping gene primers for
>quantitative real time RT-PCR.  After some fairly extensive work, I found
>HPRT to be the best.  Both beta actin and GAPDH (a.k.a. G3PDH) amplify
>genomic DNA.



This archive was generated by hypermail 2b29 : Sat Mar 10 2001 - 19:31:25 EST