K. LAKI 
475 
arises from a clottable protein dissolved in the 
plasma. The fibrinogen-like clottable protein in 
lobster's blood is clotted by a tissue enzyme, 
now recognized to be a transamidase. 
The elucidation of the clotting of lobster's 
blood progressed in three stages. In 1954, Du- 
chateau and Florkin^ reported the clotting of 
lobster's "fibrinogen" by a muscle enzyme (coag- 
uline). Later Lorand and co-workers^ showed 
that liver extracts can clot lobster's blood and 
suggested that transamidase in the liver extract 
was responsible for the clot formation. That a 
transamidase can clot fibrinogen was demon- 
strated by Farrell and Laki^^ who clotted bovine 
fibrinogen by pure transglutaminase isolated 
from guinea pig liver. 
In order to put the clotting of lobster's blood 
in perspective, I am going back to our discovery 
of the fibrin stabilizing factor. In 1945 we 
reported^i that thrombin, under certain condi- 
tions, does not clot fibrinogen — only carries out 
some small alteration on the fibrinogen 
molecules.^2 It developed that the alteration 
was the splitting of some small peptides. The 
observation that the altered fibrinogen mole- 
cules (fibrin molecules) could be kept in solu- 
tion suggested to us that it should be possible to 
bring a clot of purified fibrinogen into solution. 
Although, according to the literature, normal 
blood clot is insoluble in concentrated urea, we 
tried and found that urea readily dissolved a 
clot of purified fibrinogen. Not only did the clot 
dissolve in urea, but after dialyzing out the 
urea, the clot returned. Obviously, urea did not 
denature the fibrin molecules, only dispersed 
them. It was a simple matter for Lorand and 
me^^ to demonstrate that the clot will not dis- 
solve in urea if we added a small amount of 
serum or plasma to our purified fibrinogen be- 
fore clotting it. 
Eventually it turned out from the clinical 
studies of Duckert and co-workers^^ that the 
absence of the plasma component which renders 
the clot insoluble in urea gives rise to a clotting 
defect. The clot stabilizing factor of plasma, or 
the Laki-Lorand factor, became known as Fac- 
tor XIII. 
We know now that Factor XIII is a precursor 
of transglutaminase, the role of which is to con- 
nect fibrin molecules in the clot with primary 
bonds. 
The first indication that activated Factor 
XIII was a transamidase came when it was ob- 
served that ammonia was liberated in the stabili- 
zation reaction!"^ and that the enzyme catalyzed 
the incorporation of amines into glutamine 
residues of fibrinogen, i*' Then came the 
demonstration that 4-6 y-glutamyl-lysine bonds 
were formed in fibrin during clot stabiliza- 
tion.^'^ From the work of Doolittle and oth- 
ers, it became quite clear that the first two 
cross-bonds form near the C-terminal portion of 
the y-chains of human and bovine fibrinogens. 
The muscle enzyme of lobster which clots lobs- 
ter fibrinogen is also a transglutaminase and 
connects fibrinogen molecules with primary 
bonds similar to the transglutaminase of the 
plasma of vertebrates with the exception that 
the plasma enzyme connects fibrin molecules. 
However, as we have demonstrated," transglu- 
taminase can also clot fibrinogen molecules, not 
only fibrin molecules. In fact, it appears that in 
atherosclerosis, the disease of the arteries, 
transglutaminase does clot fibrinogen directly. 
We found^'' in experimentally-induced athero- 
sclerosis in rabbits that the sclerotic plaques con- 
tain increased amounts of transglutaminase. It 
was also found by Shainoff et al.,^^ that scler- 
otic plaques also contained increased amounts 
of fibrinogen. However, ShainoflF et al. did not 
demonstrate the presence of free fibrinogen, 
only increased amounts of the protein that re- 
leased fibrinopeptides upon addition of throm- 
bin. Microscopic examination reveals network 
structure in sclerotic plaques which people 
thought was a fibrin clot. It appears more 
likely, based on Shainoff's and our work, that 
the network structure in the sclerotic plaques is 
actually fibrinogen network. This suggests that 
in this diseased condition, the arterial tissue 
reverts to an ancient clotting habit operative in 
lobster's blood but which, under normal condi- 
tions, became outmoded in vertebrates. 
CASCADE SYSTEM 
The normal blood coagulation in man and in 
other vertebrates depends on half a dozen clot- 
ting factors of the plasma operating in a se- 
