Phenol/CHCl3 rendering dsDNA single-stranded

Benoit Hebert benoith at
Sun Dec 13 12:20:45 EST 1998

"R.N. Leach" wrote:

> Recently I've been seeing a strange phenomenon in my plasmid DNA preps.
> I cut my plasmid, gel purify it by cutting the appropriate band out of an
> agarose gel and eluting the DNA by centrifugation. After EtOH/NaoAc
> precipitation and resuspension I dephosphorylate and then phenol/chloroform
> extract the cut, dephosphorylated plasmid DNA ready to ligate in my insert.
> However, when I check a proportion of the phenol/chloroform extracted plasmid
> on a gel I see two bands. It is not uncut plasmid as the same thing happens
> to my PCR products and a vector of a different size, the expected product and
> a product half the size.
> I've narrowed it down to the phenol/chloroform step. I don't heat inactivate
> prior to the extraction (only if I'm feeling really paranoid) but I do
> vigorously vortex my phenol/chloroform+DNA suspensions. I don't believe for a
> moment that mechanical shearage is what I'm seeing so what is happening?
> Logically, I'm going to try a colleague's stock of phenol/chloroform but
> ideally I'd like to do away with the step altogether. Would gel purification
> be sufficient to remove contaminating phosphatases and dephosphatases that
> could creep into ligation reactions?
> Any ideas/information/tips gratefully received.
> Rob Leach, Ph.D.,
> School of Biochemistry & Molecular Biology,
> Uinversity of Leeds,
> W. Yorks., England
> U.K.

Phenol/Chloroform will re-anneal +/- strands, not denature them. Could someone
have contaminated your phosphatase with a restriction enzyme?

Anyway, you don't need to get rid of the phosphatase before ligation, just heat
inactivation is OK. To avoid problems and get the best results, use Shrimp
intestinal phosphatase (Roche/B.Mannheim).

have fun...



Benoit Hebert
Post-Doctoral Research Associate

Purdue University
Dept. of Biological Sciences
Structural Studies
1392 Lilly Hall of Life Sciences
West Lafayette, IN  47907-1392
email       : bhebert at  or benoit at
web site    :

More information about the Methods mailing list