RT-PCR protocol

Krzysztof W. Wasowicz wasowicz at uwm.edu.pl
Wed Feb 6 02:50:21 EST 2002

I have noticed also that my reverse transcription reactions done with
MMLV RT are somehow inconsistent. They sometimes work, sometimes not
(as judged by results of subsequent PCR). I was doing RT with exactly
the same preparations of total RNA with variable results. For example,
I was doing RT with porcine total spinal cord RNA twice (!) without
success, then I fell in doubts whether there is any RNA left in the
preparation. I checked the quality of RNA on gel and it was OK. I have
done RT with the same RNA again and this time it was working
beautifully. I didn't notice any difference in the procedure. Seems
that MMLV RT is quite sensitive to some not identified by me factors.
I have never had this kind of problems with AMV RT which I was using
for years, but had to switch to MMLV RT because it is much cheaper.
Not much help, just sharing experience. If I had money I would prefer
to switch back to AMV RT.

>I'm trying an RT-PCR of a bacterial gene with random hexamere primers
>(Roche) and MMulv, 2-tube from 1ug total RNA.  The reactions seem to work,
>but (there's always a `but'):
>	- the reactions seem not to be reproducible
>	- ...and not very efficient.
>I'm struggling to quantify the cDNA using real time PCR.  The PCR reactions
>seem to work nicely (standard curve is OK), and I get very similar results
>using different primer-pairs (different reactions) for the same cDNA and
>the same gene.  "No template controls" (NTCs) seem to be clean (no
>contamination).  However, the values are very, very low, and I got strange
>artifacts.  For example, even though the measured cDNA quantity for the two
>different primer-pairs is the same, it is lower then the measured quantity
>in the NPC (no polimerase control), which, basically, is impossible or very
>unlikely (in 18 different probes???).  
>I arrived at the conclusion that the reason for my troubles is an
>inefficiently designed RT reaction, because my real time PCR measurements
>seem to be highly reproducible with small or no errors.  Therefore, my plea
>-- I'm looking for a good protocol for a cDNA synthesis of bacterial genes
>using random hexamere primers and, preferably, MMulv.  
>Thanks in advance,
>His ideas of first-aid stopped short of squirting soda water.

Krzysztof W. Wasowicz
Department of Animal Anatomy
University of Warmia & Mazury
ul. Oczapowskiego 13 (Bldg. 105J)
10-957 Olsztyn-Kortowo
Phone: +48-89-5233688
Fax: +48-89-5234986
e-mail: wasowicz at moskit.uwm.edu.pl
ICQ: 58434323

More information about the Methods mailing list