Kunimoto, M., Otto, E., and Bennett, V. 1991. A 
new 440-kD isoform is the major ankyrin in neona- 
tal rat hvzm. J Cell Biol 115:1319-1331. 
Michaely, P., and Bennett, V. 1992. ANK repeats: a 
ubiquitous motif involved in macromolecular rec- 
ognition. Trends Cell Biol 2:127-129. 
Nehls, v., Drenckhahn, D., Joshi, R., and Bennett, 
V. 1991. Adducin in erythrocyte precursor cells 
of rats and humans: expression and compartmen- 
talization. 78:1692-1696. 
Peters, L.L., Birkenmeier, C.S., Bronson, R.T., 
White, R.A., Lux, S.E., Otto, E., Bennett, V., Hig- 
gins, A., and Barker, J.E. 1991. Purkinje cell 
degeneration associated with erythroid ankyrin 
deficiency in nb/nb mice. / Cell Biol 114: 
1233-1241. 
CONTROL OF TUMOR NECROSIS FACTOR GENE EXPRESSION 
Bruce A. Beutler, M.D., Associate Investigator 
Dr. Beutler's laboratory has continued to focus on 
the regulation and function of tumor necrosis factor 
(TNF), a protein originally isolated as a mediator of 
shock and wasting and also shown to be a selective 
cytolytic agent capable of destroying a variety of 
transplantable tumors in animal models. TNF is now 
known to participate in several inflammatory and 
catabolic disease processes. Dr. Beutler and his co- 
workers are seeking to understand not only the 
pathologic effects of TNF but also the physiologic 
effects that have justified its phylogenetic conserva- 
tion. Their method of approach has consisted of 1) 
an inquiry into the mechanisms by which TNF gene 
expression is regulated, 2) a determination of the 
anatomic distribution of biosynthesis in vivo under 
normal conditions, 3) a study of the signaling path- 
way that leads to TNF biosynthesis in response to 
endotoxin and other stimuli, 4) an analysis of TNF 
biosynthesis in endotoxemic animals and in animals 
exposed to ultraviolet light or to bacterial exotox- 
ins, and 5) an analysis of the effect of a recombinant 
inhibitor of TNF bioactivity when administered to 
normal animals or when expressed as the product of 
a transgene. 
TNF biosynthesis is regulated through restriction 
of TNF gene accessibility, through modulation of 
TNF gene transcriptional activity, and through mod- 
ulation of the efficiency with which TNF mRNA is 
translated. In past studies, Dr. Beutler and his co- 
workers demonstrated that the TNF promoter/en- 
hancer region responds to inducing stimuli by driv- 
ing transcription at a more rapid rate, whereas the 
TNF 3'-untranslated region (UTR), in its mRNA 
form, responds to inducing stimuli by permitting 
enhanced translation of the TNF mRNA. The inde- 
pendent action of these two regions of the TNF gene 
leads to a very high net inducibility of TNF biosyn- 
thesis at the protein level. In connection with these 
studies, it was noted that the TNF promoter is ubiq- 
uitously active when transfected into mammalian so- 
matic cells; apparently the promoter is well utilized 
by difi'erent tissues. 
More recently, the accessibility of the TNF gene 
was analyzed through somatic cell hybridization ex- 
periments. It was determined that the TNF gene is 
differentially methylated in different cellular envi- 
ronments. In cells that do not express TNF, a highly 
methylated form of the gene is apparent, and inactiv- 
ity of the gene seems to be imposed as a dominant 
trait. Therefore, when the TNF gene in macrophages 
is moved to a hybrid cell environment through fu- 
sion of macrophage and fibroblast cell lines, the 
normally active TNF locus is silenced. Through the 
use of reporter constructs, the DNA sequence that is 
recognized in gene inactivation has been circum- 
scribed to a region 3 2 kb in length, which includes 
the TNF promoter/enhancer region, the TNF 5'-UTR, 
and the TNF 3'-UTR. 
Gross mutations of the TNF 3'-UTR are known to 
cause disregulation of TNF biosynthesis. This has 
been shown in transgenic animals that have received 
TNF genes in which the 3'-UTR of TNF is supplanted 
by the 3'-UTR of an unrelated gene. Dr. Beutler and 
his co-workers have now shown that the 3'-UTR of 
the TNF gene in mice is relatively polymorphic; 
some of the mutations detected may have functional 
consequences related to accessibility or transla- 
tional control of TNF synthesis. 
When the TNF promoter is used to drive the ex- 
pression of a chloramphenicol acetyltransferase 
(CAT) coding sequence, which in turn is followed 
by the TNF 3'-UTR, an excellent mimic of TNF bio- 
synthesis is achieved. Cells transfected with such a 
reporter construct produce CAT activity under cir- 
cumstances in which the TNF gene itself is ex- 
pressed, provided that the latter is accessible. Since 
CAT remains confined to the cytoplasm of cells in 
which it is produced, the same reporter construct, 
22 
