IMMUNOHISTOCHEMICAL ANALYSIS OF P110(RB) EXPRESSION IN HUMAN-CELLS AND TISSUES - A REAPPRAISAL AND CRITICAL-REVIEW OF THE LITERATURE

We tested a variety of techniques to optimize the immunolocalization o f p110(RB) in deparaffinized formalin- and methacarn-fixed tissue, inc luding the use of heat-induced epitope retrieval (HIER) using a microw ave oven and DNase as pretreatments. By testing a number of commercial ly available antibodies and employing a series of different buffers at different pHs, as well as varying microwave exposure times, we obtain ed optimal nuclear immunostaining comparable to that observed in snap- frozen tissue sections and cell blocks. Employing very low pH buffers (pH 1.6) resulted in the highest signal-to-noise ratio, but also resul ted in nonspecific nuclear immunostaining that could be duplicated wit h monoclonal antibodies to cell-surface antigens (e.g., CD20) using si milar conditions. Even in optimally preserved tissues using optimal im munostaining techniques, a range of p110(RB) immunostaining patterns w as noted in normal tissues, including a subset of cells in which p110( RB) was not reliably detectable. We conclude that whereas HIER techniq ues can permit the immunolocalization of p110(RB) in deparaffinized se ctions of human tissues and tumors, given the large variations in appa rent p110(RB) immunostaining intensities, immunohistochemistry must be used with great caution as a method of assessing alterations of retin oblastoma (RE) protein function and should be used only in the context of adequate internal controls and corroborative methods of assessment of p110(RB) status.