RT PCR woes...

parakeet s535290 at aix1.uottawa.ca
Fri Dec 13 11:25:57 EST 1996

hi all,

I need some wisdom...

I have cloned the genomic copy of my Drosophila gene. Sequenced it, and to
my chagrin, it has an intron. I want to express the protein so I have been
trying to use RT-PCR to get the spliced copy from RNA. I have tried using
both a poly-T and a specific gene primer as cDNA synthesis primers. I'm
using Superscript II and using their specifications. I have done northern
blots on the total RNA that I'm using as a template for cDNA synthesis and
I can detect the transcript. This should work, right ?

wrong...it's not working at all. The expected product is of about 1.6 Kb
which encompasses the region from the start to stop codons. The oligos
were designed using primer design software (not fail safe, I know, but
they checked alright). I run controls with both genomic DNA and a lambda
genomic clone that contains the gene, and I am able to get the 2.3 Kb band
expected from the intron bearing genomic copy. The primers are obviously
working. I also do a control using the 3' primer (at the stop codon) in
conjunction with a 5' primer somewhere in the middle of the gene, and 
which I designed during the sequencing of the genomic clone. I do
manage to amplify a 0.8 Kb subfragment that I am expecting. It just appears 
that the full length copy amplifying using the primer from the start codon
region is not working out and I guess the only thing that could be going
wrong is that during cDNA synthesis, not enough of the full lenght cDNA is
being synthesized. I mean, I know the primers themselves are not at fault
because they check alright amplifying the genomic copy. Is there something
else I could do ?  I was thinking of subcloning two halves of the gene 
separately (amplified from random primed cDNA) and making a chimaeric copy
for expression, since it seems that the lenght is the problem. Though I
wouldn't think that synthesis of a 1.6 Kb cDNA molecule would be that hard
to make, would it ? I'm at a loss...it really seemed like this would be a
simple thing to do...

molecular bio, the grinch that stole christmas...

thanks for your help...


Ed Taboada
Molecular Genetics
University of Ottawa
Ottawa, Canada

More information about the Methods mailing list