Crypt cell markers in the olfactory organ of Poecilia reticulata: analysis and comparison with the fish model Danio rerio.

Olfactory crypt neurons have been observed in several bony fishes and chondrichtyans. Although their morphology is uniform in all fish, very few is known about their antigenic properties, usually studied in zebrafish, but quite overlooked in other species. We tested in Poecilia reticulata (guppy) the two antibodies recognized to mark zebrafish crypt cells: while anti-S100 showed an immunohistochemical pattern comparable to what reported in zebrafish, anti-TrkA gave no signal. Western blot analysis revealed that S100-antiserum bound an antigen of expected weight, probably belonging to the S100 family. On the contrary, anti-TrkA detected more bands, but the protein/s might be too much diffused and/or diluted in the tissue to be detected with immunohistochemistry. Because of the high level of conservation in the Trk family proteins of the kinase domain, on which anti-TrkA was produced, we also tested anti-TrkB to exclude cross reactivity. Immunohistochemistry and Western blot confirmed that anti-TrkB displayed high specificity to its target and a different staining pattern compared to anti-TrkA, but, as anti-TrkA, it did not label crypt neurons. Finally, we documented that calretinin, a known marker of zebrafish ciliated and microvillous olfactory cells, in the guppy is expressed also by a subpopulation of S100-positive crypt neurons. These results reveal differences in antigen expression between zebrafish and guppy crypt cells. Together with the already known species-specific projections to the olfactory bulb and a heterogeneous panel of odorants, our findings support the possibility that crypt cells are functionally less uniform as supposed.