Problems with RT-PCR

Gary Paterson g.paterson at
Thu Jun 25 06:15:48 EST 1998

I am interested in doing semi quantitative RT-PCR on targets that have
very low expression levels. To standardise all my reactions I have been
amplifying tubulin (T26 variant) in the same reaction tube as the
target. In the past, I have varied the cycle number from 20-40 cycles
with the T26 plateauing out at about 32 cycles, whereas, the target
starts to appear around 30 cycles.  This worked the first couple of
times about a month ago but since then I have failed to get any target
bands.  I am even now having difficulty in getting bands when no T26 is
amplified in the same tube.  I am DNAse1 treating my samples
,conditions, 7ug RNA, 3ul 10x PCR buffer, 3ul 25mM MgCl2, 40u Rnasin,
7.5u DNAse1 in a final vol of 30ul, incubate 37degs for 15min then heat
inactivate at 65degs for 10min.  I use 5ul of this to make first strand
cDNA using the Superscript kit ( Life technologies).  I do not use
Rnasin during this step and I also reduce the amount of buffer/Mg added
to the sample since the RNA sample already contains salt ( but still
give the correct final concentration)  I then use 2ul and 0.4ul for
PCR.  I have been worrying that I have got RNase contamination but the
tubulin always works well.  I have since replaced everything and it's
still not working.  The only thing i can think of, is that I'm getting
very low level of RNA degradation somewhere along the line which is
sufficient to degrade the target RNA but still leave plenty of T26 RNA
which is much more abundant.  Is it better to use primer sets that span
exon/intron boundaries and so remove the need for DNAse digestion?  I am
at a loss, so any help will be very appreciated.

Thanks Gary

More information about the Methods mailing list