142 Biosynthetic Phenotype of Thymic

The thymus is the site of considerable cell proliferation and turnover (1-3). Virtually all lymphoid cells in the organ can be labeled after 3-4 d of exposure to tritiated thymidine. It is striking, therefore, that the small cells forming the dominant population are themselves inactive in DNA synthesis. Proliferation takes place only in a minority class of large and medium-sized lymphoblasts, which traverse a full cell cycle in as little as 8 h and give rise to the small, postmitotic cells by asymmetric division (4-6). It has been estimated that the period of rapid division lasts only 1-1.5 d, after which the accumulated small cells remain in the thymus for 3-4 d before death or export (5, 6). The lymphoblasts thus represent the immediate precursors of most other thymocytes and, very likely, the ultimate precursors of T cells in the periphery. Although a great deal has been learned in recent years about the signals that trigger proliferation in mature, functional T cells (7-10), those that drive the intense mitotic activity of thymic lymphoblasts are completely obscure. It is not known, for example, whether any part of the thymoeyte antigen receptor must be engaged. For mature lymphocytes, the linkage between antigen recognition and mitogenesis is a central feature of immune specificity, and, to the extent that they have been stimulated successfully in vitro, small thymocytes also seem to require contact with antigen or a corresponding lectin (11-13). If, in contrast, proliferation signals for thymic blast cells bypass their antigen receptors, then the processes converting these blasts into small lymphocytes take on profound functional significance. On the other hand, if interaction with stromal histocompatibility antigens turns out to be required, it may be possible to formulate mechanisms in precise terms to explain the apparent influence of the thymic stroma on the antigens that T cells later recognize as "self" (14, 15). Before the triggering requirements of thymic blast cells can be compared with those of small thymocytes and peripheral T cells, the blasts must be identifiable by some characteristics other than size and mitotic activity. The studies reported here were designed to search for patterns of gene expression that distinguish the proliferating cells in the thymus from the majority populations in both cortex and medulla. To this end, populations highly enriched in cycling cells have been isolated by centrifugal elutriation and characterized by biosynthetic labeling. This has permitted a direct * Supported by grant AI 16769 from the National Institutes of Health to E. Rothenberg, and by core grant CA 14195 from the National Cancer Institute to the Armand Hammer Cancer Center.