query about methanol vs paraformaldehyde fixation

=?big5?B?xKyo2Kri?= via methods%40net.bio.net (by bmpfsu from ibms.sinica.edu.tw)
Fri Sep 7 02:56:26 EST 2007

Dear Dr. Schul,=20

I found the following mail from google when I tried to find out what's =
the differences between methanol and paraformaldehye fixation. Is these =
messenger true? Which articles can I use for reference, if I need to =
discuss with my boss?

Thank you very much.

Pei-Fen Su
Postdoctoral fellow=20
Institute of Biomedical Sciences
Academia Sinica

Dear Deana,

You have a classical case of the fixation paradox. Fixation does not
lock everything in place instantaneously. Methanol/Acetone will
precipitate proteins and permeabilize the cells at the same time. A
higly soluble protein may actually be extracted by this method,
especially from the cytoplasm, and not detected. The well known PCNA
staining to detect cells in S-phase depends on this trick because PCNA
is present in both the cytoplasm and the nucleus in unbound form and
therefore gives a very low signal in cells after methanol fixation
(extraction) except in S-phase when PCNA is attached to DNA so these
cells give a strong nuclear PCNA labelling. Formaldehyde fixation
crosslinks proteins and DNA with very little extraction (that's why
you have to permeabilize with Triton afterwards to get your antibodies
in). In the case of PCNA it results in a strong cytplasmic and nuclear
labelling in all phases of the cell-cycle.
In your case, is it possible that in formaldehyde fixed cells there is
a very weak nuclear labelling that is hard to detect against the
strong cytoplasmic labelling while the cytoplasmic protein is lost
during methanol/acetone fixation so the nuclear labelling is
detectable? Confocal laser scanning microscopy may show you the
nuclear labelling in the formaldehyde fixed cells.
Formaldehyde crosslinking may also cause problems with permbeability
and mutilation of your epitope. Maybe your antibodies can't reach the
nucleus after formaldehyde fixation; shorter fixation and longer
Triton treatment may answer that.
Good and reliable immunolocalization is more than just a standard
protocol, you have to try and optimize. Also overexpression of a GFP
fusion protein can give you the wrong picture.
Glutaraldehyde is similar to formaldehyde but don't use it with
immunofluorescent labelling because it gives strong autofluorescent

Good luck,

Wouter Schul

More information about the Methods mailing list