[Histonet] RE: detection of HRP in FFPE sections- ?antibody

[John Kiernan]

You are quite right. The peroxidase activity of HRP
does not survive dehydration and paraffin embedding,
but it does survive brief formaldehyde fixation.

If you obtain positive "peroxidase" staining in
paraffin sections it is probably due to the catalytic
action of certain metalloproteins including
haemoglobin in erythrocytes and "myeloperoxidase"
in leukocytes. Some neurons in the CNS have
"endogenous peroxidase" activity that shows up
with DAB-H2O2.

In neurobiology HRP and HRP-labelled proteins have
been used as in vivo tracers for more than 30 years.
Endogenous peroxidase-like activity has always been
a curse because there is no way to block it without
also inhibiting the HRP. 

The optimal methods for preparing sections were
worked out in the 1970s, especially by MM Mesulam
and colleagues. Their paper, "Sensitivity in 
horseradish peroxidase neurohistochemistry: a
comparative and quantitative study of nine methods"
is a must-read for anyone who needs to detect
HRP that has been introduced into a living animal. The
paper is in J. Histochem. Cytochem. 27:763-773 (1979).

Briefly, you thoroughly perfuse the animal with a 
neutral buffered formaldehyde solution, put it
in a poly bag in the fridge for a few hours,
take out the brain and put it in a strong sucrose 
solution for cryoprotection (24-48h, 4C), then 
freeze and cut frozen sections. Don't use this
summary as a set of instructions. Read Mesulam's 
paper and some of the earlier ones cited in it.

For many neuro purposes thick (50-200 um) sections 
are needed to see infrequently occurring labelled 
cells or axons. If there is a high density of HRP 
labelling, you may need to cut cryostat sections 
(5-20 um), or cut unfrozen wet sections (100-200 um) 
with a vibrating micotome, do the DAB-H2O2 peroxidase
histochemistry, osmicate, then cut out small areas of
interest, and then process these into a resin for
ultrathin sectioning and electron microscopy.

There is a huge body of literature in this field.
It's mostly in journals, and you'll need the
full text, not just the PubMed abstract. If you
are a researcher you will have access to a library.
Use it.

If you have already made paraffin sections of your
HRP-labelled tissue your only hope may lie with 
immunohistochemistry directed specifically at the
horseradish plant's peroxidase protein. There's
probably something to be found if you look in
enough catalogues of antibodies and antisera, but
it may need many trials to get valid results on
your material. It may well be less expensive to
repeat the tracing experiments and prepare the 
tissues optimally for HRP activity. This would
also increase the chances of eventually getting 
the results published in a good journal.
--
  John A. Kiernan
  Department of Anatomy & Cell Biology 
  The University of Western Ontario
  London,  Canada.
  http://publish.uwo.ca/~jkiernan/
___________________________________________
Shaumik Adhya <shaumik.adhya <@t> ucl.ac.uk> wrote:
> Date  Thu, 21 Oct 2004 16:42:57 +0100
> To  Histonet <@t> lists.utsouthwestern.edu
> Subject  [Histonet] detection of HRP in FFPE sections- ?antibody
> Hi histonetters,
> 
> I was wondering if any of you out there have experience of detection of
> HRP in formalin fixed, paraffin embedded tissue.  I assume that the
> process of fixation and tissue processing will destroy the enzymatic
> activity of HRP, rendering the usual histochemical reagents useless.
> Are there antibodies out there that anyone's used?  Thanks a lot for
> your help,
> 
> Shaumik Adhya
> 
> _______________________________________________
> Histonet mailing list
> Histonet <@t> lists.utsouthwestern.edu
> http://lists.utsouthwestern.edu/mailman/listinfo/histonet
-----------------------------------------------------------------