> Weird results in PCR

Mr. T. Weissensteiner tweissen at hgmp.mrc.ac.uk
Wed Nov 19 13:20:55 EST 1997

Actin used to be a fairly popular internal control in RT-PCR for the
amount and integrity of endogenous RNA. I don't know whether you are 
planning to do anything (semi)quantitative, but if so consider that 
a) actin can be upregulated in dividing cells, 
b) there are several pseudogenes, some of which may give PCR products 
   identical to those of the cDNA template (I am not sure whether 
   all actin pseudogenes have been sequenced).

The "weird effect" you describe can IMHO easily explained by a
competition between the larger and the smaller amplimer for common
primers, dNTPs, and Taq.

If one of the templates is preferentially amplified in each cycle,
then doing more cycles (or rather, more cycles in the exponential
phase of the amplification) will increase the final difference in
product yield. In your case, there seems to be a bias towards
amplification of the smaller (RT-PCR) fragment, so starting 
with more material and doing less PCR cycles will increase the 
relative product yield from the bigger (genomic) template. 

Note that this final relative yield is largely determined by 
the PCR conditions and tells you very little about the amount 
of genomic DNA contamination!

For a more exhaustive treatment of these and other "weird effects"
check out:

Weissensteiner, T. / Taberlet,P. (1997) Guidelines for microsatellite
allele typing. Nucleic Acids Research 25(3), Scientific Correspondence

T Weissensteiner & JS Lanchbury : Strategy for controlling
preferential amplification and avoiding false negatives in PCR typing.
Biotechniques (1996) 21 : 1102- 1108

> From: Volker Sievert <sievert at remove_this.genetik.biologie.fu-berlin.de
> [1] Weird results in RT-PCR
> Date: Tue Nov 18 20:44:13 GMT 1997
> Hello!
> Presently I do RT-PCR on crude total RNA or directly on little
> amounts of tissue with Actin-primers which span one little
> intron (these are tests during establishment of a sophisticated
> single cell RT-PCR protocol). During these experiments I observed
> weird effects.
> - crude RNA or whole cells are "contaminated" with gDNA, right?
>    Therefore it may happen that a fragment is amplified from DNA
>    rather from RNA. In my experiment, I can distinguish these frag-
>    ments from genuine RT-PCR fragments by their size (the DNA-
>    fragments are larger because they contain an intron).
> - I set up three types of reactions: one containing every reagent
>    including RTase, but no RNA. The second one contains template,
>   and every reagent but the RTase. The third one contains template,
>   and every reagent including RTase.
> What do I excpect?
> Ideally, only the third reaction gives a signal of the expected
> size for cDNA, the first two should show no signal at all. If I
> take into account the contaminating DNA, then I expect to see nothing
> in the first reaction, a larger than expected fragment in the
> second reaction and two fragments in the third reaction (the expec-
> ted one and the larger genomic one).
> What I see
> is a strange dose dependant effect. If I use very little
> material, the results came close to the ideally expected. When I
> use "intermediate" amount of RNA/cells then I see a strong signal
> of the expected size in reaction three, nothing else, but _two_
> (weaker) signals in reaction two, one of cDNA, the other of genomic
> size (and nothing at all in the first reaction). When I use lots of
> RNA both the second and the third reaction showed two fragments of
> either cDNA or genomic size.
> OK, the practical lession from that is: "Don't use too much tem-
> plate for RT-PCR". But I am still curious how the results with
> intermediate or much RNA can happen. The RTase-activity of Taq-
> polymerase alone may be sufficient to explain the reactions with
> "lots" of RNA, but not the the one with "intermediate" amount where
> presence of RTase seems to "suppress" amplification of the "wrong"
> DNA-template.
> Any ideas?
> cu, Volker

Thomas Weissensteiner
Autoimmunity Group
Jenner Institute for Vaccine Research
Compton / Newbury
Berkshire RG20 7NN

T    :  0044 1635 577920
FAX  :  0044 1625 577901

email : tweissen at RETURNMEhgmp.mrc.ac.uk
(please remove capitalized letters when using this address)

        ---{#$:>  <:$#}-~~

More information about the Methods mailing list