Role of the Time Factor in Signaling Specificity:
Application to Mitogenic and Metabolic Signaling by the Insulin
and Insulin-Like Growth Factor-I Receptor Tyrosine Kinases
Pierre De Meyts, Claus T. Christoffersen, Birgitte Urso, Brenda Wallach, Karen Gronskov, Fumiatsu Yakushiji,
and Ronald M. Shymko
The signal transduction pathways activated by hormones, growth factors, and cytokines show an extraordinary degree of
cross-talk and redundancy. This review addresses the question of how the specificity conferred at the binding step is
maintained through the signaling network despite the convergence of multiple signals on common efferent pathways such as
mitogen-activated protein (MAP) kinase. The mechanism of receptor activation by ligand-induced dimerization provides a
signaling device with both a switch and a timer. The role of the time factor, ie, of signaling kinetics, as a determinant of
selectivity is discussed with emphasis on the receptor tyrosine kinases and cytokine receptors, and especially mitogenic versus
metabolic signaling by insulin and insulin-like growth factor-I (IGF-I).
Copyright © 1995 by W.B. Saunders Company
H
ORMONES and growth factors exert their intracellu-
lar effects through the activation of a complex array
of signaling pathways that are closely interconnected (see
review by Seedorf in this issue). Although the starting point
in the signaling cascades, the cell membrane-bound recep-
tor, is highly specific for a given ligand (and sometimes a
small number of closely related molecules), this apparent
molecular selectivity is rapidly lost in the next steps of the
signaling network given the convergence of multiple recep-
tor-initiated signals on common pathways such as the
ras/raf/MAPK/ERKkinase (MEK)/mitogen-activated pro-
tein (MAP) kinase cascade or the phosphatidylinositol-3
kinase (for review, see De Meyts et al, 1 and Seedorf, this
issue). Moreover, cross-talk between the pathways trig-
gered by cytokine receptors, receptor tyrosine kinases, and
seven-transmernbrane-domain receptors further compli-
cates the understanding of how a specific starting signal
travels through the network to generate a specific end-point
pattern of cellular responses. Although a number of ele-
ments are involved in the selectivity of signals (see below),
this brief review will focus on the importance of the kinetic
aspects (eg, transient v sustained) of the activation of
signaling molecules, including the receptors themselves, in
deciding which of multiple possible bifurcating pathways
will actually be followed.
CELLULAR PROTEIN KINASES
The activity of a large number of cellular proteins is
regulated through their reversible phosphorylation by en-
zymes called protein kinases, 2 which transfer the third (g)
phosphate of the cellular energy carrier adenosine triphos-
phate to an amino acid side-chain containing a hydroxyl
(OH) group. Most protein kinases, of which there may be
more than 1,000, 3 phosphorylate proteins on serine or
threonine. In contrast, the enzymes encoded by oncogenes
From the Departments of Molecular Signaling and Scientific
Computing, Hagedorn Research Institute, Gentofle, Denmark.
Address reprint requests to Pierre De Meyts, MD, PhD, Hagedom
Research Institute, Niels Steensens Vej 6, DK 2820 Gentofle, Den-
mark.
Copyright © 1995 by W..B. Saunders Company
0026-0495 / 95/4410-4002503. O0/0
such as the prototypical src, 4 as well as a number of growth
factor receptors, instead phosphorylate tyrosines. Although
tyrosine phosphorylation may represent less than 0.1% of
total cellular phosphorylation events, it plays a crucial role
in the control of normal cell growth (both early and late
mitogenic events) and cell transformation, as well as in a
number of metabolic signaling pathways.
The reverse reaction, protein dephosphorylation, is
equally important in cell regulation, and involves a vast
array of phosphoprotein phosphatases with either serine/
threonine or tyrosine specificity that will not be discussed
here (for review, see Hunter, 5 Cohen, 6 and Tonks et a17).
Members of the protein tyrosine kinase family show
considerable structural diversity: the conserved catalytic
domain is combined with a wide variety of extracatalytic,
regulatory domains (some of which are receptor-binding
domains for extracellular ligands), s There are indeed two
groups of protein tyrosine kinases: the nonreceptor, cyto-
solic tyrosine kinases, and the transmembrane, receptor
tyrosine kinases. 9
Several families of cytosolic kinases have been identified
and many members molecularly cloned that are not recep-
tors, such as the src, fes, abl, and JAK families.8 Some
modules of src molecular anatomy, the so-called src homol-
ogy (SH) domains SH2 and SH3, are found in other
unrelated signaling molecules, where they play an impor-
tant role in their binding to motifs containing phosphory-
lated tyrosines (SH2) or proline-rich motifs (SH3) on
activated protein tyrosine kinases or intermediary docking
molecules. Many of the cytosolic tyrosine kinases are
activated by growth factors and other cellular activators. 1°
Upon activation, some of these kinases are recruited to the
plasma membrane and associate with a non-tyrosine kinase
receptor to become part of its signaling complex, eg, JAK2
with the growth hormone receptor. 11
OVERVIEW OF THE SUPERFAMILY OF RECEPTOR PROTEIN
TYROSINE KINASES
The members of this family receive environmental infor-
mation from regulatory domains expressed on the cell
surface. 12These receptors are allosteric molecules in which
high-affinity ligand binding outside the cell induces confor-
mational changes that activate the intracellular kinase
2 Metabolism, Vo144, No 10, Suppl4 (October), 1995: pp 2-11