MOLECULAR GENETICS OF MEMBRANE AND SECRETORY PROTEINS 
Mary-Jane Gething, Ph.D., Investigator 
Investigations in this laboratory focus on the mo- 
lecular genetics of membrane and secretory pro- 
teins. Experiments involve three proteins: 1) the 
hemagglutinin (HA) of influenza virus, which is 
being utilized as a marker molecule for specific cell 
populations in transgenic mouse experiments; 2) 
human tissue- type plasminogen activator (t-PA), a 
serine protease that is produced in endothelial cells 
and is involved in fibrinolysis; and 3) BiP/GRP78, a 
luminal protein of the endoplasmic reticulum (ER) 
that appears to be involved in the initial mobiliza- 
tion of proteins that traverse the secretory pathway. 
This work is undertaken in close collaboration with 
the laboratory of Dr. Joseph F. Sambrook (Univer- 
sity of Texas Southwestern Medical Center at 
Dallas). 
L Studies on Transgenic Mice Expressing Influenza 
Hemagglutinin. 
Dr. Gething and her colleagues are using trans- 
genic mice to study the development of immuno- 
logical responses to a well-characterized cell sur- 
face antigen. RIPHA mice, which express HA from 
the rat insulin II promoter/enhancer only in the P- 
cells of the pancreas, promise to provide a valuable 
model for the study of immune tolerance and auto- 
immune diabetes. Additional lines of transgenic 
mice were recently developed that express wild- 
type (cell surface) and secreted forms of HA under 
the control of the mouse metallothionein pro- 
moter. These MTHA mice, which express high lev- 
els of HA in liver and kidney, will enable investiga- 
tion of how the immunological response to HA 
differs when the protein is expressed in major cell 
populations. 
From birth, transgenic RIPHA-33 mice have slight- 
ly raised blood glucose levels (190 ±60 mg/dl) 
compared with those measured in control animals 
(136 ± 24 mg/dl). Although histological analysis of 
the pancreata of young RIPHA-33 mice reveals some 
disorganization of the normally ordered architec- 
ture of the islets, these animals display no physio- 
logical problems until —4-5 months of age. At this 
time, increases in the blood glucose levels of indi- 
vidual mice to >300 mg/dl begin to be observed, 
and shortly afterward these mice develop hypergly- 
cemia (blood glucose >400-650 mg/dl), which is 
responsive to administration of insulin. Such 
changes in blood sugar levels are not observed in 
control animals. By 9 months a significant fraction 
of the RIPHA-33 mice (60% of the males, 10% of the 
females) have developed frank diabetes mellitus. 
Histological analysis of the pancreata of such ani- 
mals reveals many islets with disrupted morphol- 
ogy, as well as destruction of (3-cells and evidence of 
insulitis. The majority of the invading immune cells 
stain with anti-CD4 antibodies, suggesting that they 
are helper/DTH cells; others stain with anti-CD8 
antibodies and may be cytotoxic T cells. Analyses of 
the sera of diabetic animals often reveal the pres- 
ence of anti-HA antibodies, as well as antibodies 
against islet cell antigens. 
II. Structure-Function Studies on Tissue-Type Plas- 
minogen Activator. 
The biological function of t-PA is to convert the 
inactive zymogen, plasminogen, into the active pro- 
tease plasmin. In the circulation the level of t-PA ac- 
tivity is controlled by the interaction of the mole- 
cule with three other proteins. First, the affinity of 
the enzyme for its substrate plasminogen is in- 
creased several hundredfold by binding to fibrin. 
Second, t-PA is rapidly inactivated by the serpin 
plasminogen activator inhibitor- 1 (PAI-1). PAI-1 acts 
as a suicide substrate and forms a covalent bond 
with Ser-478 in the active site of t-PA. Finally, the 
enzyme is efficiently cleared from the circulation by 
specific t-PA receptor (s) on hepatic cells. 
Synthesized and secreted as a single polypeptide 
chain, t-PA is subsequently cleaved into two sub- 
units held together by a single disulfide bond. The 
carboxyl-terminal light chain constitutes the 
catalytic domain of the molecule and shares homol- 
ogy with other members of the serine protease fam- 
ily. The heavy chain is composed of a number of in- 
dependent structural domains that are encoded by 
individual exons in the t-PA gene. These include 1) 
a "finger" domain having homology to the fibrin- 
binding finger domains of fibronectin, 2) an epider- 
mal growth factor (EGF)-like domain, and 3) two 
"kringle" structures having homology to similar do- 
mains found in numerous other serum proteins. A 
set of mutant enzymes lacking individual structural 
domains of the heavy chain was generated to deter- 
mine which domains of t-PA interact with the vari- 
ous effector molecules. Although the finger and 
EGF-like domains are involved in the initial, high-af- 
finity binding of t-PA to fibrin, stimulation of t-PA ac- 
Continued 
61 
