How to eliminate gDNA from RT-PCR?

John Watson watson_j at
Thu Jul 24 16:19:02 EST 1997

John Ladasky wrote:
> Hi there,
>         I designed an RT-PCR which gave me a product of the expected size
> (1 kB).  I prepared my mRNA from a cell line using the Invitrogen Fast
> Track mRNA Kit.  When I cloned and sequenced the RT-PCR products, I was
> surprised to discover that all nine of my first-round clones were not
> cDNA sequences, but instead were nearly identical to a recently-published
> processed pseudogene sequence.  I had the same in-frame, premature stop
> codons that were reported, so there's no doubt that I've amplified from
> contaminating genomic DNA.  If I replace my 5' primer with one substan-
> tially downstream, or if I try a different gene that does not have a par-
> alogous processed pseudogene, I get cDNA sequences with no problem.
>         I'm willing to bet that many of us are using mRNA preps that con-
> tain contaminating genomic DNA.  However, because the genomic sequences are
> usually longer owing to the presence of introns, it is easy to eliminate
> them from RT-PCR.  Processed pseudogenes are nasty because they have no
> introns, and thus they are the same length as the cDNA sequence.
>         I am entertaining ideas about how I can eliminate the gDNA contamination.

<interesting stuff snipped to allow Netscape posting>

We often face this problem since many of our desired targets are
intronless G protein-coupled receptors. We use TRIzol reagent from
Gibco-BRL followed by a couple of rounds of acid-phenol extraction. The
total RNA is almost always subjected to DNaseI digestion using the
enzyme and conditions supplied in the Gibco-BRL Preamplification/First
Strand synthesis kit.  Often there is a small amount of contaminating
DNA (as adjudged by PCR of the -RT template), but ususally we find that
this eliminates the problem.

AJW (not affilated with Gibco-BRL -- it just looks that way)

John Watson
Bristol-Myers Squibb Co.
watson_j at
"If you're not part of the solution, you're part of the precipitate."

More information about the Methods mailing list