mice transgenic for a single, rearranged T cell re- 
ceptor can produce both types of effector cell, a 
result now confirmed in Dr. Bottomly's laboratory. 
Second, different antigen-presenting cells activate 
different subsets of CD4 T cell. This is being 
examined using MHC class 11 transgenic mice that 
express the class II transgene on different antigen- 
presenting cell subsets. The third possible explana- 
tion for the role of MHC class II molecules in deter- 
mining the outcome of priming is that the density of 
ligand presented to the T cell receptor plays a key 
role. 
Data obtained in Dr. Bottomly's laboratory (under 
a grant from the National Institute of Allergy and 
Infectious Diseases, National Institutes of Health) 
strongly support the third hypothesis: 1 ) MHC class 
II, I-A molecules that present the peptide to give 
dominant Thl responses bind the peptide more 
strongly by several orders of magnitude than do I-A 
molecules that present the peptide for dominant 
Th2 responses; 2) adjusting peptide dose alters the 
functional outcome of immunization; and 3) mu- 
tant peptides with different I-A binding properties 
give rise to different functional outcomes. All of 
these data suggest that the density of ligand pre- 
sented to naive CD4 T cells may dictate the func- 
tional outcome of immunization. The mechanism of 
this effect, and its generalization to other peptide 
systems, are now being explored. 
Activation of CD4 T Cells Into Effector 
and Memory T Cells 
>XTien CD4 T cells first encounter antigen, they 
proliferate and subsequently differentiate into mem- 
ory or effector T cells. The first question being ad- 
dressed (under a grant from the National Cancer In- 
stitute, National Institutes of Health) is the nature of 
the memory T cell itself and its discrimination from 
effector T cells. Immunological memory is a critical 
feature of the adaptive immune response, allowing 
lasting protection from infectious disease by vacci- 
nation. However, it is not clear whether there exists 
a distinctive set of long-lived memory cells. Memory 
may simply be more cells of a given specificity, re- 
sulting from proliferation without differentiation 
following priming with antigen, or it may reflect 
continued activation of cells by antigen. 
Naive and memory CD4 T cells have been studied 
using in vitro analysis of primary and secondary re- 
sponses. Measurement of the CD4 T cell responses 
to non-self MHC molecules and to antigen, using T 
cell receptor transgenic mice, indicates that the pro- 
liferative potential of both naive and memory CD4 T 
cells is identical. The differences in the magnitude 
and lag time of the primary and memory responses 
described previously for B cells can be accounted 
for entirely by changes in the frequency of respond- 
ing cells. However, these two populations are not 
the same, because memory cells produce different 
cytokines from naive CD4 T cells. Interestingly, 
naive CD4 T cells make interleukin-2 (IL-2), a cyto- 
kine involved in T cell proliferation, while the 
memory population makes predominantly IL-4 and 
interferon-7 (IFN-7), cytokines involved in the ma- 
jor effector functions of CD4 T cells. 
These observations indicate that memory cells 
have differentiated so that they rapidly achieve ef- 
fector function upon antigen activation, while naive 
cells must proliferate before effector function can 
be acquired. Thus immunological memory in CD4 T 
cells involves changes in expression of cell surface 
molecules and their organization (see below), in- 
creases in frequency of specific cells, and differen- 
tiation such that effector status is rapidly achieved 
upon antigen activation. 
The second question being addressed is whether 
production of IL-2 and/or IL-4 is an obligatory pre- 
cursor to the generation of CD4 effector or memory 
T cells. To answer this, mice have been generated, 
in collaboration with Dr. Richard Flavell (HHMI, 
Yale University) , that carry the herpes simplex thy- 
midine kinase gene under control of the IL-2 or lL-4 
promoter sequences. When T cells are activated to 
make IL-2 or IL-4, they produce herpes simplex thy- 
midine kinase, and such cells are eliminated by 
treating cultures with the drug ganciclovir. 
Initial studies indicate that elimination of cells 
producing IL-2 within a primed memory cell popu- 
lation prevents the activation of memory CD4^ T 
cells to secrete IL-4 and IFN-7 and interferes with 
the ability of memory CD4 T cells to activate B cells. 
These results suggest that even within the memory 
population, cells first produce IL-2 in response to 
antigen and subsequently differentiate into cells 
producing IL-4 and IFN-7 as well. 
Role of CD45 in T Cell Signaling 
and T Cell Differentiation 
Subsets of CD4 T cells can be distinguished by 
their expression of unique isoforms of the trans- 
membrane tyrosine phosphatase CD45. There are at 
least six distinct CD45 isoforms that differ in the 
splicing of three exons that contribute to the ecto- 
domain of the CD45 molecule. Naive CD4 T cells 
express mainly the unspliced isoforms of CD45, 
whereas memory CD4 T cells express the CD45RO 
314 
