sus proteolysis question will be possible. Pro- 
teolytically inactive thrombins that bind the throm- 
bin receptor would also be invaluable as affinity re- 
agents for receptor purification. 
C. Thrombin receptor. Previous attempts to identify 
the thrombin receptor have not succeeded. Binding 
and crosslinking studies have been confounded by 
a number of high-affinity thrombin-binding pro- 
teins involved in terminating thrombin's action 
rather than mediating it. In addition, the nature of 
thrombin's interaction with its receptor (occupancy 
versus proteolysis) is not known. Several strategies 
have been adopted to identify and obtain the cDNA 
for the thrombin receptor. The first strategy, expres- 
sion cloning in Xenopus oocytes, requires no as- 
sumptions regarding the nature of thrombin's in- 
teraction with its receptor. The only criterion is that 
the molecule cloned by this technique must specif- 
ically confer thrombin responsiveness on the oo- 
cytes; i.e., it is a thrombin receptor. Dr. Coughlin's 
laboratory first screened a number of mammalian 
cell lines for thrombin responsiveness using cal- 
cium mobilization, an early biochemical response 
to thrombin stimulation, as an endpoint. HEL cells 
(a human megakaryocyte-like cell line) demon- 
strated a particularly robust response. Dr. 
Coughlin's laboratory then showed that microinjec- 
tion of Xenopus oocytes with mRNA prepared from 
HEL cells conferred thrombin responsiveness on 
the oocytes (in this case, thrombin-induced in- 
creases in calcium mobilization were assessed by 
thrombin-induced increases in ''^Ca efflux from pre- 
labeled oocytes). Size fractionation of the HEL 
mRNA suggested that the thrombin receptor is en- 
coded by a single mRNA species of ~4 kb. These 
experiments establish that oocytes will translate an 
mRNA species encoding the thrombin receptor and 
properly process the product. Moreover, the signal 
transduction machinery necessary to couple this re- 
ceptor to a response is present and functional in 
the oocytes. 
The oocyte expression system thus constitutes an 
assay for thrombin receptor mRNA. This can be ex- 
ploited to obtain a cDNA clone encoding the recep- 
tor. This general strategy has been successful for 
the serotonin receptor and substance K receptors. 
A cDNA library is made in a vector that allows in 
vitro transcription of the cDNA insert. The library is 
divided into pools of a specific complexity (e.g., 
20,000 clones/pool). DNA from each pool is pre- 
pared and transcribed in vitro. The resulting syn- 
thetic mRNA is assayed for receptor activity in the 
oocyte system. Once a positive pool is found, it is 
plated at a lower complexity (e.g., 1 pool of 20,000 
would be divided into multiple pools of 2,000). 
DNA from each of these pools is then transcribed, 
and the resulting mRNA is again assayed in the oo- 
cyte system. By repeated iterations, a single clone 
encoding the receptor is obtained. This tedious ap- 
proach has the advantage that once a positive pool 
is found, the path to the receptor clone is straight- 
forward. By virtue of the selection method, the 
molecule that is obtained will be the functional re- 
ceptor. 
Dr. Coughlin's laboratory has characterized the 
screening system with a model phage containing 
cDNA encoding the serotonin receptor (5HTlc); 
these studies established that the screening 
method was sensitive enough to detect a single se- 
rotonin receptor clone among 20,000 other phage. 
Two HEL cDNA libraries and one endothelial cDNA 
library are currently being screened. 
An alternative approach to expression cloning, 
using novel recombinant thrombins as affinity re- 
agents to purify the receptor, is also planned. 
Once a cDNA clone is obtained, the following 
questions will be addressed: 1) Is the receptor 
cleaved by thrombin, and what is the relationship 
of cleavage to activation? 2) Do other thrombin re- 
ceptor-Uke molecules exist? Does the thrombin 
receptor define a family of protease receptors? 
'Where are they expressed? 3) What sequences 
within thrombin and its receptor mediate throm- 
bin-receptor interaction? Can a soluble receptor ex- 
tracellular domain or portions thereof be used to 
block the platelet activation by thrombin? 4) What 
signal transduction molecules associate with the 
thrombin receptor and mediate its actions? 
II. Cells and Molecules Mediating Graft 
Atherosclerosis. 
Dr. Coughlin's laboratory has established a con- 
sortium to study the pathogenesis of accelerated 
transplant atherosclerosis as one model of patho- 
logical vascular smooth muscle cell proliferation. 
The laboratories of Drs. Bruce Hall, David Gordon, 
and Josiah Wilcox are participating. 
Accelerated transplant atherosclerosis in the cor- 
onary arteries of transplanted hearts is the major 
limitation to long-term survival of transplant recipi- 
ents. The disease is distinct from nontransplant ath- 
erosclerosis in its tempo (at least 50% have signifi- 
cant disease within 5 years) and its pathology (a 
diffuse concentric proliferative lesion that extends 
Continued 
48 
