Introduction
It has been 28 years since the first papers describing signaling bias
were published but we are just beginning to fully capitalize on this
unique property of GPCRs; at present this represents a prominent example
of a scientific mechanism failing to live up to it’s promise. GPCR
signaling bias is the result of natural protein allostery and is an
established property of receptors and agonists. Simple in vitroassays can identify biased agonists which hopefully then can be used to
produce unique (and in many cases beneficial) in vivo phenotypic
agonism. A practical barrier to the effective exploitation of this
property, however, is the confounding effect of the cellular host. It is
not clear if this is the reason for biased molecule failures in
vivo but it could be a consideration. Thus while signaling bias remains
a viable valuable property of new synthetic agonists, unless ways to
predict accurate translation can be solved, it may remain an untapped
resource in drug therapy.
Agonists produce cell response because they have affinity for the
receptor and also the fact that, when bound, they change the
conformation of the receptor through the property of efficacy. Molecular
dynamics predicts that binding is not a passive property but rather that
the binding of a ligand to a protein necessarily will alter the
conformation of that protein (Kenakin and Onaran, 2002). Therefore,
ligand binding is expected to change receptor conformation which, in
turn, will be discerned by the cell. Historically, whole cell or tissue
response has revealed graded strengths of efficacy but since whole cell
response or single signaling pathways were chosen for assays, no texture
in the quality of efficacy could be discerned. When pharmacologists
acquired the ability to selectively measure different signaling pathways
emanating from the same receptor with agonist activation, a rich
allosteric world of texture in agonist response was revealed.
In the time honored Pharmacologic tradition of ‘Occam’s Razor’ (keep it
simple), receptors were thought of as switches that were activated by
agonists to form a state which then interacted with cellular signaling
components to produce response. Though receptors are known to be
pleiotropic with respect to the number of signaling pathways they
influence, it was assumed that the full cadre of available pathways
coupled to the receptor were activated by agonists in a generally
uniform manner as a function of strength of signal. The main reason why
this could not be challenged at the time was the paucity of readouts of
agonist efficacy, i.e. these were complex outcomes of agonist activation
such as whole cell response or monotonic signals chosen by the assay
such as cyclic AMP or calcium.
The first indications that this simple model was not tenable were
reports that, when more than one signaling pathway coupled to a single
receptor could be measured, there were deviations from homogeneous
activation. Although the mechanism for this was not specified, reports
began to indicate the possibility that different agonists might select
different signaling pathways (i.e Roth and Chuang, 1987). A subsequent
wave of publications indicated that signaling was more heterogenous than
previously thought; various groups around the world published these
ideas and each had their own name for the phenomenon ( ‘Stimulus
Trafficking’; Kenakin, 1995; ‘Biased Agonism’ ; Jarpe et al, 1998 ;
‘Functional Selectivity’, Lawler et al, 1999; ‘Functional Dissociation’
Whistler et al, 1999; ‘Biased Inhibition’, Kudlacek et al, 2002;
‘Differential Engagement’,Manning, 2002; ‘Collateral Efficacy’, Kenakin,
2005). The first proposed mechanism for this was the selective
stabilization of different active receptor conformations by different
agonists (Kenakin, 1995; Kenakin and Morgan, 1989).