Control of Cell Growth and Phenotype by 
Transforming Growth Factors 
Joan Massague, Ph.D. — Investigator 
Dr. Massague is also Member of the Cell Biology and Genetics Program at Memorial Sloan- Kettering 
Cancer Center and Professor of Cell Biology at Cornell University Graduate School of Medical Sciences, 
New York. He received his Ph.D. degree in biochemistry from the University of Barcelona, Spain. His post- 
doctoral training was with Michael Czech on the identification of insulin-like growth factor receptors. He 
was Assistant and Associate Professor of Biochemistry at the University of Massachusetts Medical School 
(Worcester) before moving to his present position. 
THE purpose of our research is to elucidate the 
mechanism of action of transforming growth 
factor-^S (TGF-/3) and other factors. Our apprecia- 
tion of the complexity of the TGF-/? system has 
escalated dramatically in the last five years. TGF- 
(8 is now known to represent a vast family of para- 
crine polypeptides thought to be involved in 
many processes of tissue development, morpho- 
genesis, recycling, and repair. This family in- 
cludes homodimeric and heterodimeric products 
of more than 20 genes. The distribution of TGF-/3- 
related factors is widespread in organisms from 
fruit flies to humans, and their evolutionary con- 
servation is unusually strict. They are actively ex- 
pressed throughout embryo development and 
into adulthood, in discrete regions of many tis- 
sues with characteristic temporal patterns. 
TGF-(8 exists in multiple isoforms, including 
TGF-j8l , -(82, and -|83 in mammals. These isoforms 
show many similarities in their biological proper- 
ties, but also some important differences. TGF-j8 
is exceptional in its multifunctional nature, act- 
ing, for example, as a paracrine inhibitor of cell 
proliferation and a regulator of cell differentia- 
tion, cell adhesion, and extracellular matrix de- 
position. Defects in TGF-/? activity as a growth 
suppressor may lead to oncogenesis, and its ex- 
cessive activity to fibrotic disorders. 
Mechanisms of TGF-/? Action 
One of our goals is to identify, clone, and char- 
acterize the membrane receptors that mediate 
TGF-/3 action. To this end, we have identified 
membrane proteins that bind TGF-(S with high af- 
finity. These include TGF-/? receptor components 
I and II, which are distinct cell-surface glycopro- 
teins of 53 and 70 kDa, respectively. They are 
expressed ubiquitously, but in low numbers, in 
most normal and transformed cells, with the nota- 
ble exception of human retinoblastoma cells. 
These receptors markedly discriminate between 
the various TGF-/3 isoforms. A third TGF-(8-bind- 
ing component, betaglycan, has an unusual struc- 
ture; it is a membrane-anchored proteoglycan. 
Which of these binding components mediates 
TGF-j8 action? The resolution of this question 
made use of one of the most remarkable effects of 
TGF-/3, inhibition of cell growth. A panel of mu- 
tant cell clones resistant to the growth inhibitory 
action of TGF-/3 was isolated by selection of chem- 
ically mutagenized cells that would grow in the 
continuous presence of TGF-/3. Characterization 
of these mutants showed that the mutations af- 
fected the expression of TGF-/? receptors I and II 
but not betaglycan. Moreover, somatic cell hy- 
brids between cell mutant clones demonstrated 
that all TGF-j8 responses were regained when ex- 
pression of receptors I and II was restored by 
complementation between different cell mu- 
tants. This evidence linked receptors I and II to 
mediation of the pleiotropic action of TGF-/3 and 
indicated that both receptor components cooper- 
ate to mediate the multiple effects of TGF-/?. We 
are developing methodology to use the TGF-^S re- 
ceptor-defective cells to clone receptor cDNAs. 
Progress has also been made in the study of be- 
taglycan. This integral membrane proteoglycan 
contains heparan/chondroitin sulfate glycosami- 
noglycan (GAG) chains attached to a 110- to 
1 30-kDa core glycoprotein. The GAGs are not re- 
quired for TGF-^ binding or expression of beta- 
glycan on the cell surface; the TGF-;8-binding site 
resides in the core protein. Soluble forms of be- 
taglycan lacking membrane anchor are released 
into the medium and can be found in extracellu- 
lar matrices. 
Betaglycan is intriguing as a novel example of 
regulatory molecules involved in the biology of 
growth factors. It is widely distributed in tissues, 
but absent from certain cell types that, neverthe- 
less, respond to TGF-/?. No evidence is presently 
available for a direct role of betaglycan in signal 
transduction. It may be involved, however, in li- 
gand presentation to TGF-/3 receptors I and II and 
could function as a pericellular reservoir of 
bioactive TGF-jS. Like other membrane proteogly- 
cans, betaglycan may also participate in cell ad- 
hesion and recognition. We have purified it to 
homogeneity in order to investigate it further. 
Recent progress linked growth arrest by TGF-/? 
in late Gl to its ability to control the phosphory- 
lation state of an intracellular growth suppressor. 
301 
