Differential Display :

Stephen R. Lasky Stephen_Lasky at brown.edu
Wed Apr 20 08:01:43 EST 1994

In article <2p1e7f$b9n at mother.usf.edu>, schnetze at chuma.ec.usf.edu. (Joann
Schnetzer (BIO)) wrote:

> I'm having problems getting the results of my differential display 
> experiment. I followed the established protocol for this procedure 
> (Liang,et.al. I believe - Don't hold me to it though) and my gel shows 
> streaking/fuzzes; yet no clear banding patterns like sequencing (which is 
> how it is supposed to be). 

How many times have you done the procedure?  It gets better as you do it
more.  Are you using Chomczynski's RNA prep technique?  If so, how long do
you let the RNA sit before you do your first strand synthesis? We found
that using it right away (like the day that you precipitate it and
resuspend the RNA in an aqueus solution) gives the best results.  We used
to think that a high background was due to the presence of RNA in
sequencing reactions.  Sometimes you see a high background in DD also.

>Does anyone know whether the problem is in the 
> RNA (I used total, although this isn't supposed to cause problems) the 
> DNAse treatment of the RNA (placental ribonuclease inhibitor added with 
> DNAse I), the RT, the PCR, or ???. 

Did you use the DNase kit that Liang's group sells?  I have used that and
RQ-DNase with good results.  One of the things that we found when cloning
RT-PCR products is that the quality of your first strand synthesis is a
limiting variable.  If you have really good first strand synthesis, it is
easy to clone out of PCR (using the TA-Cloning techniques) but if your
first strand isnt great, its very difficult to get what you want.  I
carried this idea into DD and got much better and more consistent results
than by just using the RT part of Liang's kits.  I used the Cycle DNA kit
to do two rounds of 1st strand synthesis and got clearer bands that came up
much more quickly.  The trick with the CycleDNA kit (unfortunately) was the
methyl mercury denaturation of the RNA.  They have now removed the MeHg
from the kit since it is toxic and OSHA doesnt like it (or so I was told).
So the cycleDNA kit is the same as everyone else's cDNA kit.  Liang said he
didn't need to do this, but I couldn't get my DD gels to look like his
using his kit.  By optimizing the first strand synthesis they look more
like the advertized gels (still not as many bands though which is probably

Like I said, DD gets better the more times you do it.  The thing to
remember is that once you get some bands you really need to confirm them by
northern analysis and by repeating the DD and seeing the same band come up
differentially.  There are a very large percent of false positives.

I'll send this to your e-Mail address as well as the list because I want to
see what others have to say about their use of this tech. 

Good luck.

> schnetze at chuma.cas.usf.edu
> with any helpful suggestions, (or lighthearted jokes, if you're witty)

Stephen R. Lasky, Ph.D.       Brown University/Roger Williams Medical
e-mail: Stephen_Lasky at brown.edu         LandLine: 401-456-6572
A nuclear war could ruin your whole day.

More information about the Methods mailing list