RNase Protection Assay; non-specific?

David L. Haviland, Ph.D. dhavilan at IMM2.IMM.UTH.TMC.EDU
Mon Feb 23 01:13:14 EST 1998

At 21:53 2/20/98 -0800, Rolands G. Aravindan wrote:


OK... I'll take a stab at this...

>I have never actually done RPA, but a Professor in the department showed
>me a RPA blot that had just returned from a journal after initial
>review.The reviewer's criticism of the blot was rather strange to me. I
>wonderhow one could explain/rebutt it.

I've done RPA, hands on.  I've not diddled with any sort of kit, I
conducted the assay per the Harvard manual as a means to confirm 5' CAP
site of mRNA.  

>Let me explain the blot, in short. It is rather clean with clearly
>treatment-dose-dependent increase in the expression of POMC mRNA and the
>standard was constant. Both signals are remarkably strong and the
>background is remarkably white and clean; except for another minor
>(about1/100 the specific band) band, also appearing dose-dependent, about
>100 bases lesser in size. Considering the strength of the specific signal,
>the blot could have been sent in with slightly decreased exposure to film
>without compromising authenicity (?; wrong? coments appreciated) in the
>first place.

Could you have used a lesser exposure?  Possibly.  Depends on info that I
need from you.

1)  You make no mention of the "ever-needed" tRNA control lane.  This is
almost always shown and should be absolutely blank.  There should be no
hybridzing signal and theoretically no signal from the probe either.

2)  What is the purpose of the RPA?  5'-end detection?  Overall message
detection?  In any case, the signal should be visibly less (base pair wise)
than that of the probe alone.  I usually design my probes to be 200-250 bp
long with the expected fragment to be about ~100-150 bp.  The expected
(guesstimate +/- 10 bp) size should be known ahead of time.  If the
expected size is the same size as the probe then it is paramount that the
tRNA lane is shown and is blank - suggesting adequate RNAse digestion.
Personally, I've found complete (as in absolutely blank) digestion of the
probe is sometime hard to achieve.  I've dropped the signal to where it
barely exposes the film, but rarely no signal at all.

3)  Can the additional signal possibly be due to the probe folding on
itself?  GCG -RNA fold might lend some information. 

4)  Is the sample RNA poly A+ message or total?   I found using primer
extension that numerous stray signals in total RNA dissappeared when tried
again with PA+ mRNA.

>Strangely, the reviewer has asked "whether the minor POMC band was also
>scanned"! Being quite familiar with Northerns and the like, I would have
>thought that there really is no confusion as to the fact that the minor
>band is not related to POMC and that it really is nonspecific. Or am I
>wrong? Anybody out there have any suggestions as to how one could
>explain/rebutt the reviewer's comment?
>I would be very grateful for all replies, either directly or via the NG.

Even when using formamide, RPA's are prone to folding in the RNA rendering
false signals.  

I understand that it is a dose response increase of your message that
you're looking for.   However, would primer extension be of use?  Normally,
these two methods are played "off each other" as a dual evidence for
suggesting that transcription starts at a particular nucleotide, given how
RNA folding can render a false signal.  If your RPA is based on the very
5'end of your message, this might be something to consider.

Hope this helps,

 David L. Haviland, Ph.D.
 Asst. Prof. Immunology 
 University of Texas - Houston, H.S.C.
 Institute of Molecular Medicine  
 2121 W. Holcombe Blvd.  
 Houston, TX  77030 
 Internet:"dhavilan at imm2.imm.uth.tmc.edu" 
 Voice: 713.500.2413  FAX: 713.500.2424
Never take life seriously. Nobody gets out alive anyway.

More information about the Methods mailing list