Optimization of non-radioactive in situ hybridization: image analysis of varying pretreatment, hybridization and probe labelling conditions

Using detection of proopiomelanocortin (POMC) mRNA in rat pituitary as a model, varying conditions of tissue pretreatment, hybridization and probe labelling have been tested. Results were evaluated both by visual assessment and by image analysis of coded specimens. Good correlations between visual gradation, optical densities and cell area percentages were obtained. However, determinations of optical densities (or pixel values) provided most detailed information. The data obtained emphasize the interdependence of fixation and permeabilization conditions and clearly show that the stronger the primary fixation, the more efficient the permeabilization by proteinase K must be. The hybridization temperature is also of importance and temperatures between 40–45° C produced the best signal to noise ratio. The POMC-directed 24-mer probe had a theoretical melting point (Tm) of 49.4° C (in the absence of formamide) and four individual experimental determinations of Tm produced a mean value of 48.9° C. Detection of the biotinylated probe was best accomplished with monoclonal antibiotin antibodies and the alkaline phosphatase-anti-alkaline phosphatase (APAAP) system. Short washes at high-stringency (0.1×SSC, 45° C) produced an optimal signal to noise ratio. Inclusion of 50% formamide in the hybridization buffer produced an enhanced signal to noise ratio, in spite of a higher background staining. The probe employed for most studies was a synthetic 24-mer oligodeoxynucleotide, complementary to the MSH[4–11]-coding region of POMC mRNA. It was labelled with biotinylated dUTP and unlabelled dCTP using terminal transferase. Chromatographical analyses revealed the labelled probe to be heterogeneous in tail length. Image analyses of stainings obtained with individual probe fractions, varying in tail length, proved that probes extended to 27-mer size produced the best results. Finally, similar experiments using a 24-mer probe complementary to the dynorphin B[10–18]-coding region of prodynorphin mRNA revealed best hybridization results with a tail length of 4 nucleotides. Purification and testing of labelled probes can result in major improvements in hybridization detection efficiency.