RT-PCR blues; help please!

Andre Hamel hamel at ccu.umanitoba.ca
Fri Apr 2 11:34:54 EST 1993

Have you considered trying different primers corresponding to a different
part of your mRNA, for example, if you current primers amplify a 300 bp
region close to the 3' end of the mRNA, try primers that amplify a product
towards to 5' end of the mRNA. Perhaps there's some sequences in the part
of your mRNA that are too closely similar to other RNAs (such as other

Also, have you tried your current primers out on cDNA insert (such as
found in plasmid clone)? Or on in vitro transcript of same such cDNA?
Serial dilution of such in vitro transcripts? Also, compare other RT
enzymes, such as Superscript (RNaseH deficient MMLV RT, from BRL), it
works up to 50oC, and seems to produce less "artifacts" based upon
sequence of cloned cDNAs, also is very processive (up to 10 kb routinely
made, eg./ 32P labelled BRL RNA size marker (0.24 to 9 kb) yields distinct
9 kb cDNA on autorads after alk.den. 0.8% agarose gel). ALso New England
Biolabs MMLV RT works very nice too. I've had mixed results with the AMV
RTs and personally keep away from them (costs more too!). 


Andre Hamel
Manitoba Vet.Virol.
Winnipeg, Manitoba

email: hamel at ccu.umanitoba.ca

In article <1993Apr1.145936.8442 at mbcf.stjude.org> neale at mbcf.stjude.org writes:
>G'day all,
>I am curious to know what people are using for RT-PCR (reverse transcriptase
>We have been using AMV RT using a specific oligo for cDNA synthesis, followed
>by PCR using more of the specific 3'oligo  coupled with a 5' specific oligo.
>Problems we have been having are rather weird to say the least. When the
>abundance of the mRNA is low we find that more product is synthesized in the
>ABSENCE of RT. When the abundance is high we find the expected result in that
>-RT reactions are usually clean.
>Now before you think to the obvious solution of DNA contamination in the
>sample, let me point out that all RNA samples are DNase I digested, and in some
>cases digested additionally with restriction endonuclease between primer sites
>to avoid such pitfalls. RNA samples tested with primers that would amplify DNA
>if present do not give products after these treatments. (ie the 5' primer is
>targeted  upstream of the transcription initiation site)
>There are two issues in my mind:
>(1) There is inherent RT activity in the Taq polymerase (purchased from
>Promega, and said to be the same as that from Perkin ELmer).
>  I know of a
>couple papers where this activity has been reported for Taq Pol, and I'm not
>referring to the Tth pol which is being sold for this type of assay.
>I spoke to a rep from Perkin Elmer (Promega's rep was not helpful in this area)
>who said that others had experienced a similar observation that could not be
>attributed to DNA contamination (different sized products for spliced mRNA vs
>genomic DNA), but this usually occurred with high abundant RNA (>10,000 copies
>per reaction). 
>(2) The RT appears to inhibit PCR amplification of the product since there
>is more product in the -RT assay than in the +RT assay. I currently heat
>inactivate the RT at 95 C for 15 min prior to PCR, yet there still seems to be
>some effect at low amounts of target RNA. In control experiments where we
>dilute an abundant RNA species from one cell with RNA from a non-expressing
>cell we find that the switch over occurs around 10-4 dilution (Idon't know what
>that is in copy number). At dilutions to 10-3 there is more product in the +RT
>assays but when you reach 10-4 dilution the opposite occurs where the -RT shows 
>better than the +RT.
>In these experiments all water controls are clean and the primers work fine
>with DNA templates.
>So my questions are:
>(i) What gives? If anyone has experienced similar weirdness (there is another
>lab in our dept which also has this problem) do you know why?
>(ii) Does anyone have recommendations as to the best (ie most infallible)
>method of RT-PCR. (Someone else at the institute swears by the invitrogen cDNA
>copy kit for first strand synthesis). Should we try priming with oligo dT
>instead of a specifiic 3' oligo (as I assume this kit does)?
>Thanks in advance,
>Geoffrey Neale Ph.D
>neale at mbcf.stjude.org
>Dept. of Virology and Molecular Biology
>St. Jude Children's Research Hospital
>Memphis, TN
>Voice: (901) 522-0400
>Fax: (901) 523-2622
>"Such is life" - Ned Kelly, infamous Australian bushranger before his hanging

More information about the Methods mailing list