How the fixation-embedding protocol affects the specificity and efficiency of immunocytochemical stains for gonadotropin subunits.

This report describes a study designed to test factors that may affect the efficiency and specificity of stains for gonadotropins. These include chemical or freeze-fixation and dehydration, heat polymerization of the plastic embedding compound, dehydration in organic solvents, and etching. Specifically, postembedding stains for LH or FSH subunits were applied to 1-μm sections of (1) Araldite-embedded pituitaries that were either chemically fixed and dehydrated or freeze-fixed and freeze-dried; (2) Aldehyde-fixed pituitaries that were dehydrated in water-soluble glycol methacrylate (GMA) and embedded in GMA at 4°C; and (3) p-formaldehyde-fixed pituitaries that were embedded in paraffin. A fourth group of pituitaries was dispersed and grown in monolayers for 1–3 days. These were stained following glutaraldehyde fixation. The optimal dilution of the primary antisera varied with the protocol; however, the percentage of cells staining for beta subunits did not change. In contrast, postembedding stains showed that alpha subunit reactivity is masked or destroyed in pituitaries that are fixed in glutaraldehyde and embedded in Araldite. Alpha chain reactivity was detected (in 14% of cells) either after freeze-fixation and freeze-drying followed by Araldite embedding, or after 4% paraformaldehyde fixation and GMA embedding (in 17% of cells). Staining in paraffin-embedded pituitaries was seen in only 10% of the cells. Preembedding stains for alpha chains were strikingly sensitive, however, and immunoreactivity was seen in 18–26% of the population of monolayer cells. Thus, whereas the percentages of cells staining for beta subunits do not change following the use of most of the fixation and embedding protocols, alpha chain reactivity is destroyed by all but the mildest. These findings show that one can control or improve the specificity of the stains for LH and FSH by the fixation-embedding protocol.