REGULATION OF CACHECTIN/TUMOR NECROSIS FACTOR BIOSYNTHESIS 
Bruce A. Beutler, M.D., Assistant Investigator 
Dr. Beutler and his colleagues are studying the 
mechanisms by which biosynthesis of cachectin 
[tumor necrosis factor (TNF)] is controlled. This 
important hormone, which is a product of periph- 
eral blood monocytes and tissue macrophages, is 
produced in response to invasive stimuli, notably 
bacterial endotoxin. When released in large quanti- 
ties, it acts to mediate many of the untoward effects 
of endotoxin, including shock, tissue injury, and 
coagulopathy. When released over a long period of 
time, it eventuates a state in which anorexia and 
unabating protein and lipid catabolism lead to 
weight loss and death. Cachectin/TNF has also been 
implicated as a mediator of several inflammatory 
processes in humans and animals, including the 
acute phase of graft-versus-host disease and the 
vascular changes that are observed in cerebral ma- 
laria. Thus a clear understanding of cachectin/TNF 
regulation would be medically significant. 
Transcription of the cachectin/TNF gene is accel- 
erated threefold after activation of the macrophage 
by endotoxin. However, cachectin/TNF mRNA levels 
rise by 100-fold or more, and at least a 10,000-fold 
increase in synthesis may be observed at the pro- 
tein level. Thus much of the regulation of 
cachectin/TNF biosynthesis is post-transcriptional. 
This is supported by the observation that mRNA en- 
coding cachectin/TNF and other cytokines often ex- 
ists in an untranslated form. 
Recent work in Dr. Beutler's laboratory has indi- 
cated that regulation of cachectin/TNF biosynthesis 
is largely exercised at a translational level. Reporter 
constructs, in which a chloramphenicol acetyl- 
transferase (CAT) -coding sequence is driven by a 
constitutively active promoter unresponsive to the 
effects of endotoxin, but followed by varying por- 
tions of the cachectin/TNF 3 -untranslated region 
(UTR), have revealed that endotoxin-responsive se- 
quences reside in the 3' -UTR of cachectin/TNF 
mRNA. The most important element within the 3 - 
UTR is the UpA-rich sequence commonly observed 
in many short-lived cytokine and proto-oncogene 
mRNAs, which has elsewhere been shown to con- 
fer translational suppression. Endotoxin appears 
to cause "derepression" of translation, leading to a 
> 200-fold increase in the rate of reporter protein 
synthesis. Dexamethasone, which inhibits cachec- 
tin/TNF biosynthesis, also strongly inhibits the 
derepressive effect of endotoxin. 
When cachectin/TNF is produced, it is synthe- 
sized as a prohormone containing (in the case of 
the murine protein) 79 propeptide amino acids ap- 
pended at the amino terminus. The propeptide 
portion of the molecule is more highly conserved 
among species than the mature cachectin/TNF. It 
has been suggested that the protein is initially an- 
chored in the plasma membrane and that mem- 
brane-associated cachectin/TNF may exist as an ac- 
tive species. Dr. Beutler and his colleagues have 
examined the processing of murine cachectin/TNF 
and have determined that a fraction of the protein 
exists as a glycosylated product. Moreover, they 
have found that a major alternative cleavage site 
exists, wherein the prohormone is cleaved before 
Leu yielding a soluble but inactive product. 
This suggests that the addition of several residues 
to the amino terminus is not tolerated, at least in 
the case of the murine protein. The extended form 
of the hormone retains a relatively high affinity for 
the cachectin/TNF receptor. This indicates that it 
may be possible to design competitive antagonists 
of cachectin/TNF binding. Given the inactivity of 
the precursor forms of cachectin/TNF, it is also 
possible that the protein may be regulated by 
proteolytic cleavage. This possibility is under 
study. 
Dr. Beutler has also been interested in determin- 
ing the nature of the signal evoked by endotoxin. 
This signal, which leads to the biosynthesis of 
cachectin/TNF, is apparently inoperative in C3H/HeJ 
mice. These mice are genetically unresponsive to 
endotoxin and therefore tolerate the administra- 
tion of endotoxin in doses that are fatal to sensitive 
animals. Using macrophages of a line derived from 
the C3H/HeJ mouse as an assay substrate and add- 
ing back proteins derived from endotoxin-activated 
normal macrophages by electroporation. Dr. 
Beutler and his co-workers have demonstrated that 
a cytoplasmic protein induced by endotoxin acts to 
trigger the biosynthesis of cachectin/TNF. Currently 
research is directed toward the isolation and char- 
acterization of this protein, which may be central to 
the activation process. 
The chief cellular in vivo sources of cachec- 
tin/TNF have remained mysterious, since the hor- 
mone is rapidly cleared and degraded after release. 
A nonsecreted marker protein, produced under cir- 
cumstances identical to those evoking the release 
of cachectin/TNF, would help to clarify this issue. 
Using a marker construct in which the cachec- 
Continued 
21 
