Molecular Approaches to Lymphocyte Recognition and Differentiation 
transgenic for T cell receptors of defined specific- 
ity and have followed the expression of these mol- 
ecules during T cell differentiation. We have ob- 
tained evidence of selection both for and against 
particular T cell receptors, depending on the 
MHC genotype of the host. 
Our evidence concerning positive selection in- 
dicates that this is an inefficient process that can 
be significantly influenced by the density or loca- 
tion of the MHC molecules required. Experi- 
ments also indicate that MHC alleles that differ in 
the binding site of MHC profoundly alter or abol- 
ish this type of selection, suggesting that specific 
peptides (or other molecules occupying the site) 
may play a role. Thus it seems probable that in 
addition to the complex of foreign antigen plus 
MHC that a given T cell receptor may encounter 
in the periphery, there is first another MHC + 
"X" complex that must be successfully encoun- 
tered in the thymus. 
We have also developed a transgenic model for 
self-tolerance of T cells by introducing a fusion 
protein that contains the target peptide seen by 
our T cell receptor transgenic mice. This, to- 
gether with direct peptide injection/slow deliv- 
ery systems, has enabled us to control the timing 
and dose of an antigen seen as "self." We find 
evidence for both paralysis and deletion of T 
cells, with deletion, which occurs in both the 
thymus and the periphery, the favored outcome 
as the dose of self antigen is increased. 
Genes Controlling Lymphocyte 
Differentiation 
The last area of interest involves the isolation 
and characterization of genes that might control 
differentiation in lymphocytes. We have worked 
extensively on the XLR gene, which encodes a 
small (25-kDa) previously unidentified nuclear 
protein that is specifically turned on in late-stage 
B cells and medium- to later-stage T cells. One 
novel feature of the protein is that it is stabilized 
in its nuclear location by a relatively low concen- 
tration of zinc ions. A number of other proteins 
have the same characteristics, and there may be 
some novel zinc-dependent structures in the nu- 
cleus. We are currently trying to deduce what the 
product of this gene does. 
A major effort is also under way to isolate other 
potential regulatory genes, using new subtractive 
hybridization and cloning schemes and nuclear 
localization as a screen. We are particularly inter- 
ested in genes that are turned on late in B cell 
differentiation and in those that are specific for 
early thymocytes. Nuclear localizing proteins in 
either category may provide important clues 
about the regulation of differentiation in these 
cells, clues that may be applicable to cellular dif- 
ferentiation in general. 
106 
