An RT-PCR problem to sink your teeth into

Andrew J. Doherty doherty at
Tue Sep 3 02:34:00 EST 1996

K J C 944 wrote:
> OK, Bionetters, here's a problem you can sink your teeth into:
> I'm doing RT-PCR on mussel larvae total RNA to determine at what
> developmental stages a particular gene is transcribed.  I tried out my
> primers on a cDNA of the gene which I have cloned and sequenced.  The
> primer set works great giving a single strong band of the correct size.  I
> then constructed an internal standard by cutting out a piece of the cDNA,
> religating and transcribing into RNA.  (Yes I transcribed the right way,
> etc.)  Using the religated cDNA, I tested my primers again and got a
> single band of the correct size.   I also transcribed my cDNA  into RNA
> (without deletion) to use as a standard to test the system. Everything was
> checked on gels, quantified on the spec, etc. So far, so good.
>  I diluted my standards in total RNA from tissue which does not express
> the protein (my negative control RNA) and did RT on the following:
> 1. My deletion standard, diluted in negative control RNA.
> 2. My standard without deletion, diluted in negative control RNA.
> 3. Negative control RNA.
> 4. Positive control RNA (total RNA extracted from the tissue where the
> gene is expressed).
> 5. Positive control RNA without the RT enzyme added.
> After the RT reaction (using random haxamers), I did PCR (hot start,
> touchdown) using my gene specific primers.  I included a positive PCR
> control (the cDNA mentioned above) and a no template negative control.  I
> got a single strong band for my positive PCR control, no PCR products for
> my negative PCR control, my negative control RNA sample, and my positive
> control sample without the RT added, as expected.  And I got a single
> strong band for my deletion standard (correct size).  Here's where things
> get strange:  Both my positive control RNA and my standard RNA without the
> deletion gave a very weak band (barely visible) of the correct size and
> two even weaker smeary bands that were smaller by 100-200 bp.  I should
> mention at this point that I have done RT-PCR using random haxamers and
> the same RNA samples, but different PCR primers, several times before with
> success.
>  This morning I pulled out one of my previous (and successful) RT samples
> (from the same tissue RNA extraction) to compare to my new RT prep, and
> ran two sets of PCRs on each RT prep:  One with a set of primers for the
> same gene but further upstream. (I've used these primers before.)  And the
> second with the primers described above.  Using the upstream primers, I
> got good amplification for both templates.  Using the downstream primers,
> I got the same results on both templates; same as before:  three very weak
> bands.
> At this point, I know my primers are no good, but why?  Can anyone give me
> a good explanation as to why the RNA template works ok with these primers
> if a small section has been cut out, but not if left intact?  Any Ideas on
> how to solve this problem, without going through the trouble of testing
> new primers, making new standards, etc.  (The upstream primer set is not
> suitable due to polymorphism in the upsteam gene sequence.)
> I apologize if this is too long, just trying to provide as many details as
> possible.
> thanks in advance,
> Kathryn J. Coyne
> College of Marine Studies
> University of DE
> Lewes, DE  19958
Hi Kathryn, is this a secondary structure problem? The fact that the
primers work well on the deletion standard but not the full length one
would suggest that cut something out which prevents the amplification
and/or RT process. This may be a loop in the RNA, especially as you are
also getting a couple of smaller bands amplifying (maybe RT "jumping"
across the junction?). Just don't ask why this should not be a problem
with your standard RNA!!

Hope it helps and I'll keep chewing the cud.

   Dr Andrew Doherty                          
   Department of Anatomy                   
   School of Medical Sciences             
   University Walk                               
   BS8 1TD

   e-mail  Doherty at


More information about the Methods mailing list