T CELL DEVELOPMENT 
Philippa Marrack, Ph.D., Investigator 
T cell development occurs most efficiently in the 
thymus. In this organ, precursor cells progress 
through a number of transformations, which are 
crucial to the eventual appearance of mature T cells 
that will be useful to the animal. These transforma- 
tions include rearrangement and expression of the 
genes that code for the aP or "yS receptors, which 
the cell will eventually bear, and induction of ex- 
pression of a number of other genes that are re- 
quired for the function of mature T cells, in particu- 
lar CD4 and CDS. These events appear to be 
common to almost all precursors. Subsequently the 
developing a(3-bearing cell must go through two 
other tests, based on the specificity of its receptor: 
1) positive selection for self-restriction [i.e., cells 
are picked out to mature if their ap receptors will 
be able to recognize foreign antigens bound to self 
major histocompatibility complex (MHC) mole- 
cules] and 2) clonal deletion of potentially au- 
toreactive cells. Only after a thymocyte has been 
successfully selected for self-restriction and has 
proved not to be self-reactive does it mature and 
become a functional peripheral T cell. 
A few years ago this laboratory was the first to 
show that tolerance to self is achieved, in part, by 
clonal elimination of potentially self-reactive cells in 
the thymus. This was done in studies on the fate of 
T cells bearing particular Vas as part of their ap-re- 
ceptors. In some cases the presence of a particular 
vp on the surface of a T cell is predictive of its reac- 
tion with a particular superantigen. T cells bearing 
VP3, for example, react with the mouse-expressed 
superantigens Mls-2^ or Mls-3^. Expression in the 
animal of a particular superantigen caused the dis- 
appearance in the thymus of cells bearing the vp 
that this superantigen could stimulate. At the time, 
however, reagents were not available to allow a de- 
termination of exactly when in thymocyte develop- 
ment such deletion occurred. 
Recently, in collaboration with Dr. Ralph Kubo 
(National Jewish Center for Immunology and Respi- 
ratory Medicine) a monoclonal antibody was devel- 
oped against all mouse aP-receptors that could be 
used in such an experiment. With this reagent it 
was shown that self antigens caused the disappear- 
ance of all mature, and about half the immature, 
potentially reactive thymocytes. This suggested that 
clonal deletion could occur when the target thy- 
mocyte was an immature, CD4"'^, CD8"^, cortical thy- 
mocyte. 
This discovery posed a new problem, because se- 
lection for self-restriction also occurs at this stage of 
thymocyte development. Because both positive se- 
lection and clonal deletion depend on reaction 
with the a P- receptor, then both must be signaling 
the thymocyte through its receptor; thus the prob- 
lem arises of how the thymocyte distinguishes be- 
tween a selecting or a deleting signal. It has been 
suggested that the distinction depends on the cell 
with which the thymocyte is interacting, because 
different MHC-presenting cells might provide differ- 
ent secondary signals to the thymocyte. Interac- 
tions between receptor and MHC molecules on thy- 
mus cortical epithelial cells, for example, might 
lead to positive selection, whereas interactions be- 
tween receptors and MHC on bone marrow-de- 
rived macrophages or dendritic cells might lead to 
clonal deletion. There is a good deal of evidence to 
support this notion. Other mechanisms have not, 
however, been excluded. 
This problem has recently been tackled by mem- 
bers of this laboratory, in collaboration with Dr. 
Kubo and Dr. John Cambier (National Jewish 
Center for Immunology and Respiratory Medi- 
cine). Thymocyte development was studied in 
vitro in cultured intact thymus lobes from fetal 
mice. In such cultures thymocyte development 
proceeds normally, with the exception that 
the lobes gradually accumulate mature thymocytes 
and become depleted in precursor cells, because 
prothymocytes in the organ are not self-regener- 
ating but are replenished in the intact animal by 
fresh precursors from bone marrow. Therefore, in 
such cultures, aP-receptor~, CD4", and CDS" pre- 
cursors gradually disappear, giving rise to the ap- 
receptor"*^, CD4^, CDS"*^ intermediate cells and 
eventually to CD4"^ or CDS^, single, positive, ma- 
ture thymocytes. Addition of anti-aP-antibody 
to such cultures caused the disappearance of all 
mature thymocytes and half of the immature ap^ 
cells; addition of an antibody to CD3, the complex 
of proteins with which ap is associated on cell 
surfaces, led to the elimination of all ap^ cells, both 
mature and immature. Study of the ability of the 
different antibodies to induce increases in intracel- 
lular Ca^"*^ showed that anti-CD3 caused Ca^^ in- 
creases in all receptor-bearing cells, whereas 
anti-aP caused increases in all mature thymocytes 
but only half of the immature aP^ cells, the half 
that were eliminated by anti-ap-antibody. It is likely 
Continued 
423 
