Control of Cell Growth and Phenotype by Transforming Growth Factors 
RB, the nuclear phosphoprotein product of the 
retinoblastoma susceptibility gene. TGF-/? arrests 
cell growth at precisely the cell cycle point 
where RB phosphorylation is due to take place. 
This observation led to the finding that TGF-/3 in- 
hibits RB phosphorylation, retaining this protein 
in its presumed growth suppressor state. 
Expression of the SV40 transforming protein, 
large T antigen, which binds unphosphorylated 
RB and disrupts its growth-suppressive function, 
prevented growth inhibition by TGF-/3 without 
affecting its inhibition of RB phosphorylation or 
other effects. Thus TGF-|S's inhibition of RB phos- 
phorylation is not an irrelevant event but, possi- 
bly, one that mediates cell cycle arrest by this 
factor. These results represent a case in which a 
prototypic intracellular growth suppressor (RB) 
participates in the mechanism of action of a 
growth-inhibitory paracrine agent (TGF-|S) . 
The mechanisms involved in growth suppres- 
sion by TGF-/3 might also relate to those that lead 
to its control of cell differentiation. This effect 
was best studied with skeletal muscle myoblasts. 
TGF-18 inhibited myoblast differentiation in part 
by interfering with the expression of myogenin, a 
differentiation-determining gene. Other studies 
showed that TGF-jS is a major regulator of the 
cell adhesion apparatus — cell adhesion recep- 
tors and extracellular matrix protein expression 
— in many cell types. These effects could in turn 
explain the ability of TGF-)8 to affect the morphol- 
ogy and phenotype of cells and, most impor- 
tantly, explain its presumed roles in tissue mor- 
phogenesis and fibrotic disorders. Effects of 
TGF-|S on gene expression might be mediated by 
its ability to control the levels or activity of cer- 
tain transcription factors, a possibility that is the 
focus of current studies. 
Cell-Cell Communication by 
Membrane-anchored Growth Factors 
Our research interests extend also to the area of 
growth factors that are synthesized as part of 
transmembrane protein precursors. The model 
system in our studies is transforming growth fac- 
tor-a (TGF-a). This is a 50-amino acid polypep- 
tide initially identified in culture fluids of trans- 
formed cells, hence its name. It is not related to 
TGF-/5 but shares structural homology with 
members of the epidermal growth factor (EGF) 
family. TGF-a is most prevalent in tumor-derived 
cells and causes a high incidence of mammary 
and liver neoplasias in transgenic mice that 
chronically overexpress it. The factor stimulates 
cell growth through the EGF membrane receptor, 
which is a ligand-activated tyrosine kinase. 
Paracrine growth factors and polypeptide hor- 
mones are generally synthesized as larger soluble 
precursors that are cleaved by specific peptidases 
to release the bioactive domain. In contrast to 
this, a remarkable feature of TGF-a, shared with 
the other members of its family, is its generation 
by cleavage of a membrane-anchored precursor, 
pro-TGF-a. This is a 1 60-amino acid polypeptide 
with the TGF-a sequence in the extracellular do- 
main. The proteolytic process that releases ma- 
ture TGF-a from its precursor is inefficient in 
most cell types. Consequently, substantial levels 
of pro-TGF-a can accumulate on the cell surface. 
Membrane pro-TGF-a can bind EGF receptors 
on the surface of adjacent cells. This interaction 
leads to signal transduction and DNA replication. 
Pro-TGF-a also functions as a mediator of cell- 
cell adhesion. Thus a membrane-anchored 
growth factor and its receptor can function simul- 
taneously as mediators of cell-cell adhesion and 
as initiators of mitogenic stimulation by cell-cell 
contact. The membrane-anchored forms could 
be important in tissue development processes 
whose guidance depends on discrete cell-cell in- 
teractions incompatible with the diffusible na- 
ture of soluble factors. Given their structure, the 
membrane forms might function as signaling re- 
ceptors. This possibility is under study. 
302 
