The combination of ISH with immunohistochemistry allows the simultaneous detection of nucleic acids and proteins. This approach has been successfully used in a variety of settings. ISH for the detection of DNA requires heat denaturation of tissue and probe DNA which may destroy antigens. Similarly, the prehy-bridisation treatment and the formamide used in many ISH protocols (DNA or RNA) may interfere with subsequent antigen detection. Also, dextran sulphate, which is a constituent of hybridisation mixtures in most published protocols, can bind to proteins and impair their antigenic properties. Therefore, it is usually recommended to perform immunohistology before ISH (Herbst et al., 1992; Niedobitek et al., 1997a; Roberts et al., 1989; Van der Loos et al., 1989). However, methods performing ISH prior to immunohistology have been described for some stable antigens (Wolber and Lloyd, 1988). When immunohistochemistry is conducted prior to ISH, RNase inhibitors must be added to antibodies and other solutions to prevent RNA degradation. Heparin, placental RNase inhibitor and yeast tRNA are useful to block RNase (Hofler et al., 1987). Also, low concentrations of diethylpyrocarbonate (DEPC) may be used, but high DEPC concentrations can also degrade immunoglobulins and antigens. However, in spite of all precautions, RNA degradation will occur to some extent during immunohistology. Therefore, control slides hybridised to the probe without previous immunohistology should be included to assess RNA loss and avoid erroneous evaluation.
Immunohistochemistry can be readily combined with ISH employing either radioactive or non-radioactive probes and various immunohistochemical detection systems. Use of radioactive probes is advantageous in some settings. The detection of radiolabelled probe requires only dipping of slides into a nuclear track emulsion. By eliminating the immunohistochemical detection reagents necessary for immunoenzymatic probe detection, there is no risk of cross-reactivity of any reagents. However, in some situations it may be more appropriate to use non-radioactive instead of radioactive ISH. The tissue resolution usually is better when enzymatic procedures are used for the demonstration of bound probe. Also, overexposure of radiolabelled probes may result in an intense signal leading to saturation of the emulsion which may hide the underlying immunohistochemical staining product.
Immunohistochemistry on frozen or paraffin sections as well as on cytological preparations is performed as required for the respective antibody. After completion of immunohistology, including enzymatic development of the reaction, a proteolytic treatment may be required to unmask the target nucleic acid. Treatment of sections with a 3M KCl solution in addition to the proteolytic digestion has proved helpful in our hands (Niedobitek and Herbst, 1991). Some attention has to be given to the choice of enzymes and chromogens. If both immunohis-tology and ISH are developed by enzymatic procedures, then chromogens have to be chosen to give good contrast. Diaminobenzidine (DAB) for peroxidase and new fuchsin or Fast Red for alkaline phosphatase result in stable precipitates still clearly visible even after photographic development of radioactive ISH. Nitro blue tetrazolium (NBT) results in a dark blue to brown precipitate which contrasts well with the red colour of new fuchsin or Fast Red. It is important to bear in mind that some of these chromogens are not alcohol resistant. This may make modifications of the ISH protocol necessary.
When setting up double staining experiments using two immunoenzymatic detection systems, it is important to chose reagents so as to avoid cross reactivities. Using DAB as the chromogen for the first step is advantageous since the reaction product will mask all underlying antibody reagents thus reducing the risk of cross reactivity. Double staining using two alkaline phosphatase-based detection systems is possible but must be carefully controlled to exclude that the alkaline phosphatase used to detect the first reaction is still reactive in the second step. In our experience this is never entirely possible. Of course, detection systems employing two different fluorochromes may also be employed (Zaidi et al.,
2000). Double-labelling techniques can also be applied to the simultaneous detection of two different nucleic acids. In this setting, usually one radioactively labelled probe is combined with one non-radioactive probe. Both probes are hybridised simultaneously. The non-radioactive probe is detected by immu-noenzymatic techniques including colour development followed by dipping into nuclear track emulsion. Finally, it is possible to perform double-labelling ISH using probes labelled with 3H and 35S or 33P (Salin et al., 2000). However, discrimination of the signals generated by these probes requires sophisticated technical equipment which is not widely available (Salin et al., 2000).
Was this article helpful?