Problems with S1 digestion

[Dr. Peter Gegenheimer]

On Fri, 13 Feb 1998 12:38:53, January Weiner 
<jweiner1 at aixterm2.urz.uni-heidelberg.de> wrote:

> Hi,
> I recently tried to map 3' end of a bacterial RNA species using a
> 32-P 3' labelled PCR product as a probe. I have repeated the experiment NN
> times under various conditions, but it still doesn't work - apart of
> getting me a little frustrated. 
> 	I have used the standard protocol by Maniatis, and I am quite sure
> that I ruled out the usual suspects (like old buffer, cooling below 37 C
> after hybridisation and so on). The S1 is from Boehringer and seems to work
> in other cases. Everything I get is an awful smear, even in the case of a
> heavy overdigestion. 
> 	Thank you in advance for your help,
> 						January


?!? You said the oligonucleotide probe was [32P]-labelled at its 
_3'_ end. Is this right? I am not aware of any method for 3' 
end-labelling a probe other than with [5'-32P]pCp and RNA ligase, or
tailing with [alpha-32P](d)NTPs and TdT or another polymerase. 
Unless the added 3' base (for example, C) of the probe is 
complementary to the RNA target, it will be removed by the nuclease.


For example, the sequence of a PCR product may be written 
5'-hoNpN.....pNoh-3', which after ligation to *pCp (*p denotes the 
5'-32P) will be 5'-hoNpN.....pN*pCp. Suppose the RNA:DNA hybrid is 
written as

            (RNA) 3'-RpRpRp---------------pRpRpR-5'
                         : : : : : : : : : :
                         Dp---------------pD
           (DNA) 5'-DpDp/                   \*pCp-3'


Endonucleases of the P1 family (P1, S1, and mung-bean nucleases) all
cleave their substrate 5' to the phosphate. The reaction products 
will thus be 

            (RNA)   3'-  Rp---------------pRp -5'
                         : : : : : : : : : :
            (DNA)   5'- pDp---------------pD  -3'
                                                  +  *pC and Pi     

I think the simplest way to do the mapping you want is to use an 
internally-labelled probe. The PCR reaction can be done in the 
present of the 4 normal dNTPs and however much [32P]dNTP you need to
achieve the desired specific activity (the spec. act. of the product
is determined solely by the specific activity of the labelled dNTP).
You need to make sure that the probe is long enough to be really 
specific for your transcript. 

In the nuclease reaction itself, you should use P1 nuclease rather 
than S1. P1 digestions can be done at neutral pH and in the absence 
of Zn(2+); there are no special buffer requirements. The omission of
divalent ion reduces RNA fragmentation, and the neutral pH makes the
RNA:DNA hybrid substantially more stable than it would be at pH 4.5 
(commonly used for S1 digestions). Although P1 is a Zn-containing 
metallonuclease, the catalytic Zn is tightly bound and no Zn need be
supplied in the reaction buffer. For example, 0.2 M NH4OAc (ammonium
acetate, pH unadjusted) works well; HEPES or Tris (pH 7-7.5) with 
0.5 M KCl should work as well. (Our P1 analyses use 20 mM NH4OAc to 
prevent base-pairing, so I don't have first-hand data on the salt 
effect. Most of these enzymes are very salt-tolerant.) Remember that
the shorter the labelled probe, the higher the salt concentration 
will be needed to keep the RNA:DNA hybrid perfectly paired.  



o-------------------------------------------------------------------
--o
| Dr. Peter Gegenheimer       |  Vox: 785-864-3939  FAX: 
785-864-5321 |
| Departments of Biochemistry |    PGegen at UKans.edu                 
 |
|  and of Botany              |    http://RNAworld.Bio.UKans.edu/   
 |
|                             |                                     
 |
| University of Kansas        |                                     
 |
| 2045 Haworth Hall           |  "The sleep of reason produces      
 |
| Lawrence  KS  66045-2106    |   monsters."                  Goya  
 |
o_____________________________|_____________________________________
__o