In principle it is possible, but you have to count on the autofluorescence of formaldehyde- fixed tissue.

I have no experience with brain tissues (except from mice) but we do almost all our double IF on FFPE tissues (formalin-fixed paraffin-embedded). In mouse embryos there is no problem with autofluorescence problem in the brain.

Generally, the resolution and quality of the sections is pretty good compared to frozen sections especially since it is a little bit easier to cut very thin sections. I also have the feeling (no hard data) that more antibodies work in FFPE tissues than frozen sections, but that is just a general observation I made over the years.

There are actually ways to reduce autofluorescence. There has been a discussion about this here on Researchgate. Unfortunately, I can't find it that quickly anymore. One suggested PBS/glycin 3%, one ammonium hydroxide, 50mM NH4Cl in PBS for 30 min, another some other "exotic" chemical. There seem to be options out there to reduce autofluorescence. There is also MaxBlock Autofluorescence Reducing Kit, or check out https://www.researchgate.net/post/Does_anyone_have_a_good_protocol_to_reduce_remove_autofluorescence_in_tissues?_sg=o%2FBAogwRLBnNw380lXj9bXXnsfaktNTXVDhHhGp55VEwVMOss8JAubxhKyMND%2Fdn_io1YHd2v5A1scDJcoMGdPIHNBGF0Vq6VQD3nwRmRjesw%2FK8dx10%2BuEdED37tfsAZ_6Fieq7OtahGgWeSGiyEGVgUqnTQfupMa2%2F6QIvRRgol%2BafBUD5RNwY3z3pwD2JyY&_tpcectx=search.

Thanks for your answers!! I fixed my tissue in 4 % paraformaldehyde in PBS, so probably I don´t have the autofluorescence problem. What do you think about it?

Gudrun is probably the expert. PFA is basically the same as formalin without methanol. I don' think using PFA versus formalin makes a substantial differences when it comes to autofluorescence. Actually autofluorescence can occur in frozen material as well. 

I don't use glutaraldehyde very often-basically never- but the one time I used it, it gave me this nasty "background". Must have messed up the entire section to produce a weird background autofluorescence? But glutaraldehyde is a much more powerful fixative than PFA/formalin.

If you are concerned about autofluorescence, just leave one section untouched (just apply PBS to it during staining). This will show you the autofluorescence background level.

Thomas is right, that formalin and paraformaldehyde is both based on formaldehyde. Formalin is the aequous solution of formaldehyde produced via gas bubbled through water. And paraformaldehyde is the dry substance of polymerizised formaldehyde-molecules, that is brought again into solution. The difference is, that commercial formalin stock solution usually contents also methanol to stabilize the product. Formaldehyde binds to aminoacid-residues and builds methylen-bridges beween them. It may happen, that is also produces intra-molecular-bridges, that form carbon-rings, that act similar to fluorophors.  The background is, that through this reaction free pi-electrons are produced, that are able to absorbe the excitation light.

Glutardialdehyde has double-bonds with pi-electrons in the formula and as polymerizate it has many binding sites for proteins to produce a very close-meshed net.

In practice we saw a small decrease in autofluorescence after heat antigen retrieval (HIER). But this was done for in-situ-hybridization. The longer the fixation - the stronger the autofluorescence. But with a strong specific signal, this may be ingored.

Hya, i do agree with the answers above and i would add that yes it is possible. Your fixation should be fine re the autofluorescence. I haven't tried brain tissue but soft tissue and skin tumours in the past and they seemed to work ok. In general it depends on the abundance of your target and the efficiency of your antibody as well. make  sure you use antibodies specific for use with parafin embeded samples as experience have taught me in the past! good luck

Finally, I did fluorescence immunohistochemistry in paraffin sections and it worked, the only problem is the background of the staining. How can reduce it?

Can you specify what you mean with background? Not autofluorescence? background from the primary antibody, from the secondary system? Background relative to what? Unstained sections, sections with just secondary, sections with a control serum (mouse IgG, rabbit Ig?). There are many potential "layers" of background depending on what kind of quality reagents you have used.

Background from the primary antibody. I can not detect well the labelled protein

A good starter to read if one is not sure what the primary antibody is doing is this article:Griffiths G1, Lucocq JM: Antibodies for immunolabeling by light and electron microscopy: not for the faint hearted. Histochem Cell Biol. 2014 Oct;142(4):347-60. doi: 10.1007/s00418-014-1263-5. Epub 2014 Aug 24. I found this through Researchgate!!! This is an honest summary of the quality of antibody stainings, what to do to make the situation better, and what to be aware of. And it has interesting references within.

Basically it gives you an idea how bad things are with antibodies. I have seen many dubious publications where stainings were basically "background" or better false positive staining. And you can read many comments on Researchgate about bad antibodies and bad antibody sellers.

I assume you know where your antigens are expressed (nucleus, membrane etc). Then you can expect that the antibody stainings are restricted to these subcelluar structures. If your primary antibody produces "background" then there is very little you can do except reabsorb it versus the background epitope. I have seem many polyclonal rabbit sera that like to bind to keratins and this will "overshadow" the good antibodies in the serum.

You can try to use ordinary blocking that in a perfect world should be a useless reagent since you often block with something like BSA . That will not fundamentally make a bad antibody good unless miraculously the "background epitopes" are in the blocking solution. Very unlikely. Blocking for IF is probably one of the most overrated procedures.

You can try to vary your antigen retrieval in the hope that the background will not be activated by certain methods. I never tried this but it may be worth the effort if there are no alternative antibodies available. Since you intend double-stainings this may not be an option.

I hope I understood your info correctly and my comments will be helpful.

Somebody knows some product to prevent the unglue of brain sections embedded in paraffin? It should be apply at the beginning of the inmunohistochemistry to avoid that the solutions detach the edges of the brain sections.

You can use slides electrostatically charged or  solutions (Amino-Propil-Trietoxisilane -APES/TESPA/SILANE. I use this for IHC, i do not know if will interfere with IF.

I would go for the slides (thermo scientific-superfrost)