pocampus and to have specific learning impair- 
ments. These disabilities indicate that a-CaMKII has 
a prominent role in spatial learning but is not essen- 
tial for some types of nonspatial learning. The data 
thus considerably strengthen the contention that the 
synaptic changes exhibited in LTP are the basis for 
spatial memory. 
Dr. Tonegawa's laboratory is carrying out further 
characterization of the a-CaMKII mutant mice with 
respect to a possible defect in the modulation of the 
amygdala-based acoustic startle response (a collabo- 
ration with Dr. Michael Davis at Yale University) and 
to the kindling behavior (a collaboration with Dr. 
James McNamara at Duke University). Dr. Tone- 
gawa's group is also producing several other mice, 
each with a mutation in a gene of neurobiological 
significance. 
DNA Rearrangement 
To test the hypothesis that somatic DNA rearrange- 
ment akin to immunoglobulin (Ig) or T cell recep- 
tor (TCR) V(D)J (V, variable; D, diversity; J, joining) 
recombination plays a pivotal role in the develop- 
ment or functions of the central nervous system 
(CNS), Dr. Tonegawa's laboratory produced trans- 
genic (Tg) mice with the bacterial /3-galactosidase 
gene 0-lacZ ) whose expression would be depen- 
dent on a V(D)J or V(D)J-like inversional recombina- 
tion occurring in its flanking regions. Despite the 
previous report of others to the contrary, Dr. Tone- 
gawa's group obtained no evidence of somatic re- 
combination in the CNS of the Tg mice. They con- 
cluded that the previously presented evidence is 
disputable and that somatic rearrangement in the 
CNS has yet to be demonstrated. 
a/? T Cells 
Using the ES cell gene-targeting technique. Dr. 
Tonegawa's laboratory produced mice with muta- 
tions in the TCR a, /3, or b gene as well as in the 
RAG-1 gene, whose product is required for somatic 
rearrangement of Ig and TCR genes. Each mutation 
blocks T cell differentiation at a distinct stage. In 
RAG-1 mutants, the thymocyte differentiation is 
blocked at a CD4-CD8 double-negative (DN) state. 
Rearrangement and/or expression of the TCR /3 gene 
is required and sufficient for the transition of the DN 
to the double-positive (DP) state and for the expan- 
sion of the DP cells to the wild-type level. TCR a 
gene rearrangement and/or expression is irrelevant 
for these events but required for the transition of the 
DP to CD4 or CDS single-positive state. Neither TCR 
a nor TCR /3 is required for the generation of 76 T 
cells. Reciprocally, the TCR b mutation does not 
block a{i T cell differentiation. 
While it is clear that TCR |8 gene rearrangement 
occurs prior to TCR a gene rearrangement in thymic 
ontogeny, neither is required for the other to occur. 
In the thymus and the lymph nodes of TCR a mutant 
mice, a small fraction of T cells express /3 only (i.e., 
a-less) TCRs associated with CD3 molecules. In the 
lymph nodes, these T cells are exclusively CD4- 
positive, suggesting that they interact with MHC 
class II molecules during differentiation. The func- 
tions of these novel T cells are unknown. 
Dr. Tonegawa's laboratory also produced mice 
with mutations in the peptide-transport gene TAP-1. 
In these mice, cell surface expression of MHC class I 
molecules and the generation of CDS-positive T 
cells are severely impaired. Expression of an MHC 
class I molecule can be restored specifically in vivo 
or in vitro upon injection of or treatment with a 
peptide known to bind with that molecule. Thus 
these mutant mice provide an excellent system for 
studying both the role and the nature of self pep- 
tides involved during intrathymic positive selection 
of T cells. (This project has been supported in part 
by grants from the National Institutes of Health.) 
76 T Cells 
T cells with ■yb receptors are composed of subsets 
that utilize one or a few distinct V gene segments 
each to encode its TCR, which appears at a particu- 
lar time in ontogeny. By making use of the fact that 
TCR 7 and b gene rearrangement can occur in the 
TCR b mutant mice without cell surface expression 
of the receptor (and therefore without TCR- 
mediated cellular selection). Dr. Tonegawa's lab- 
oratory demonstrated that developmentally pro- 
grammed TCR gene rearrangement plays a pivotal 
role in the temporarily differential generation of the 
76 T cell subsets. 
Using the same mice, the laboratory also demon- 
strated that the remarkable sequence homogeneity 
in the V(D)J junctions of the TCR genes expressed in 
the skin- and uterus-associated yb T cell subsets is 
accomplished primarily by evolutionarily acquired 
sequence features at the junction of these genes and 
only auxiliarily by cellular selection. 
On the other hand, the laboratory demonstrated 
the importance of cellular selection in the matura- 
tion of 75 T cells in a Tg mouse system. In the thy- 
mus or the spleen of mice produced by crossing 76 
TCR-Tg mice with others genetically deficient in 
182-microglobulin (iSjiTti) . no mature Tg 76 T cells are 
generated. Hence interaction between the Tg yb 
TCR and a /32m-associated molecule is required for 
the generation of mature Tg 76 T cells. 
Dr. Tonegawa and his colleagues used the TCR a 
and TCR /? mutant mice to identify the function of yb 
364 
