Abstract
It has proved difficult to obtain good immunohistochemical localization of cell surface antigens in nerve for a number of reasons, prominent among which are problems of fixing this class of molecule without destroying their antigenicity. In the course of developing a fixation procedure suitable for one such antigen. Thy-1, we have quantitatively assessed the effect of different fixation parameters upon the retention of Thy-1 antigenicity and upon the extent of cross-linking of the antigen in the tissue. The former was measured using radioimmunoassays adapted for membrane antigens in fixed tissue, the latter by measuring the proportion of antigen rendered insoluble to the detergent, sodium deoxycholate, and by examining the size of the antigen on sodium dodecyl sulfate--polyacrylamide gels. These approaches demonstrated that minor modifications of the standard vascular perfusion fixation of brain, using both glutaraldehyde and paraformaldehyde, were sufficient to fix the Thy-1 molecule, and at the same time substantially spare its antigenicity. In this study we measured Thy-1 using both a conventional rabbit antiserum and a mouse monoclonal antibody to the Thy-1.1 antigenic determinant. The multiple antigenic determinants recognized by the rabbit antibodies were cumulatively more resistant to fixation than the single antigenic determinant recognized by the monoclonal antibody.