T Lymphocyte Biology and Genetics 
We initially reported that MHC class II molecules 
are the specific cellular receptors for a prototype 
of this family of molecules, staphylococcal en- 
terotoxin A (SEA) . 
Interest in the structural basis of superantigen 
activities led us to assess their binding affinities 
to MHC class II molecules and the effects of bio- 
chemical modifications on binding and function. 
We observed a direct correlation between bind- 
ing affinity as a linear function and the logarithm 
of the effective dose for T cell stimulation. Thus 
relatively small changes in binding affinity had 
marked effects on enterotoxin potency. A com- 
mon feature of members of the enterotoxin fam- 
ily is a small intrachain disulfide loop in the cen- 
tral region of the molecule. We assessed the role 
of this loop and adjacent amino acids on func- 
tional activity and demonstrated that the disul- 
fide bond was essential to T cell mitogenic activ- 
ity but had little effect on MHC class II binding or 
monocyte stimulation. We concluded that the in- 
ability of these modified toxins to induce T cell 
proliferation reflected their failure to interact ef- 
fectively with the T cell receptor for antigen 
rather than with class II molecules. 
For several years we have also studied "non- 
classical" MHC class I genes and antigens of the 
mouse. These genes, which constitute the major- 
ity of class I genes, were termed nonclassical be- 
cause they had not been thought to be involved in 
conventional antigen presentation and self- 
nonself discrimination. We have been specifi- 
cally interested in the maternally transmitted an- 
tigen (Mta). Discovered by Kirsten Fischer Lin- 
dahl (HHMI, University of Texas Southwestern 
Medical Center at Dallas), Mta is recognized by 
cytolytic T cells and is unique in its mode of in- 
heritance. Offspring of mice with different forms 
of the antigen always express the maternal form. 
Both nuclear and mitochondrial genes are in- 
volved in Mta expression, but the maternally in- 
herited polymorphism is determined by a mito- 
chondrial gene product. This is a hydrophobic 
peptide encoded at the amino terminus of the 
NDl mitochondrial gene. We have investigated 
structural properties of this peptide that enable it 
to bind and be presented by a specific nonclassi- 
cal class I gene product, Hmt. The essential fea- 
ture of this peptide is a biochemical substitution 
(an amino-terminal formyl residue) that is char- 
acteristic of the first amino acid of mitochondrial 
proteins but is not found on proteins synthesized 
from nuclear genes. Proteins synthesized within 
mitochondria share this property with proteins 
synthesized by bacteria. We therefore investi- 
gated the capacity of certain bacterial peptides to 
bind similarly to the Hmt class I molecule. Three 
of the first six synthetic bacterial peptides char- 
acterized displayed this property, but if and only 
if they also had the terminal formyl residue. Un- 
modified peptides or peptides modified with 
other biochemical radicals were unable to bind. 
These data suggest that the Hmt class I mole- 
cule may represent a specialized host defense 
mechanism for recognition of proteins of bacte- 
rial origin. As such they may provide insight into 
the biological functions of the nonclassical MHC 
class I antigens. Additionally, they offer the first 
evidence of true biochemical specialization in 
the binding of antigenic peptides to a specific 
MHC molecule. 
370 
