DNA fragmentation is not the primary event in glucocorticoid-induced thymocyte death in vivo

Thymocyte death has been recognized as one of the best models for studying apoptosis. Our recent study, however, indicated that most thymocytes die without DNA fragmentation and become terminal deoxynucleotidyl transferase (TdT)-mediated dUTP-biotin nick end-labeling-positive (TUNEL⁺) only after being phagocytosed by macrophages. In this study, we used histological techniques using the TUNEL method, histochemistry, immunohistochemistry, and transmission electron microscopy as well as flow cytometry to examine in vivo the effect of glucocorticoid (GC), a well-known agent for inducing thymocyte apoptosis in vitro, on thymocyte death to determine whether or not DNA fragmentation was the first event of GC-induced thymocyte death. At 2 h and 4 h after GC injection, a large number of cortical thymocytes were TUNEL⁺. Most TUNEL⁺ cells were aggregated to form clusters. Double staining of the section showed that the TUNEL⁺ thymocytes were phagocytosed by acid phosphatase⁺ and Mac-2⁺ macrophages. An ultrastructural study indicated that a far greater number of small pyknotic thymocytes were present in the cortex of the GC-treated thymus than were observed in the control thymus, that all those pyknotic thymocytes were TUNEL^-, and moreover, that at the electron microscopic level, TUNEL⁺ cells were all phagocytosed by macrophages. Flow cytometric analysis did not detect a single TUNEL⁺ thymocyte even 4 h after the GC treatment, suggesting that virtually no free dead thymocytes were present after DNA fragmentation. These results indicate that, consistent with our previous findings with normal thymocyte death and B cell death in the germinal centers, DNA fragmentation is not involved in the cell death process of the GC-induced rapid thymocyte death in vivo.