TUNEL (terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling) is a method of choice for rapid identification and quantification of the apoptotic cell fraction in cultured cell preparations. However, TUNEL application has been restricted to a narrow spectrum of sample conditions, and only detergents have been proposed as labeling enhancers. This study was aimed at extending TUNEL to variously fixed cells and improving TUNEL sensitivity by optimized pretreatments, the specificity being assessed by reference to the apoptotic morphology. Comparative TUNEL was performed with three protocols on CEM-C7 cells, a model of glucocorticoid-induced apoptosis. Samples were submitted to six modalities of fixation and TUNEL was performed after each of the following conditions: no pretreatment; detergent permeabilization; proteolytic digestion; microwave irradiation; and a recently published combination of the latter two. The proportion of TUNEL-stained elements within the cell fraction, with and without apoptotic morphology, was quantified. Our results showed that: (a) with an adequate pretreatment, reliable TUNEL can be obtained after each fixative tested; (b) detergent was inefficient in improving sensitivity; (c) whatever the fixation, microwave pretreatment provided the best TUNEL sensitivity without notable loss of specificity; (d) under adaptive technical conditions, TUNEL can be associated with detection of various proteins by double labeling; and (e) the existence of a limited population of intensely TUNEL-positive cells that lacked apoptotic morphology contributes to the current debate about a preapoptotic state.