Mapping the Genotype–Phenotype Relationship in Cellular Signaling Networks: Building Bridges Over the Unknown - Systems Biology: Networks, Models, and Applications - page 141

like lipids (e.g., PIP3) and metabolites (e.g., ATP). Rather, genomic
technologies identify which proteins participate in a given cellular
signaling network.
COMPONENT INTERACTION: BRIDGING GENOTYPES
TO INTERMEDIATE PHENOTYPES
Relatively quick (on the order of seconds to minutes), measurable
responses to a signaling event are here called “intermediate phenotypes”
(see figure 5.1). For example, the rapid increase in intracellular calcium
concentrations is an intermediate phenotype because it is an observable
response to a triggering signaling event. These intermediate pheno-
types may be transient responses or may lead to irreversible changes in
gene expression (an endpoint phenotype). Most technologies devel-
oped to date to characterize cellular signaling networks, which are
elaborated in table 5.2, measure these intermediate phenotypes.
Perhaps the most widely used technique to characterize signaling
functions is the yeast two-hybrid approach and its associated varia-
tions [55]. Simply, the genes for target and bait proteins are fused to
DNA binding and transcriptional activity protein domains so that if the
target and bait proteins interact, the expression of a corresponding
gene occurs. While this approach has been criticized for its lack of con-
textual specificity (e.g., two proteins might never be expressed at the
same time, and yet they may interact in a two-hybrid screen), tremen-
dous amounts of data have generated novel hypotheses regarding
signaling network function. For example, yeast two-hybrid screens
were recently used to construct a proteome-wide protein-protein inter-
action, or “interactome,” map of Caenorhabditis elegans [56], including
a network of 71 interactions among 59 proteins for the C. elegans
DAF-7/TGF-b signal transduction pathway [57].
FRET-based technologies underpin another set of methods for
measuring intermediate phenotypes [36]. This technology involves the
fusion of two fluorophores to two proteins. Each fluorophore emits
light at a distinct wavelength; however, when two proteins associate,
the wavelength of the emitted light is different. Thus, associating pro-
teins and their dynamic characteristics can be identified. For instance,
a genetically encoded fluorescent indicator called a phocus, a mutant
of the traditional green fluorescent protein (GFP), was recently devel-
oped specifically for the purpose of visualizing signal transduction
based upon protein phosphorylation in vivo [58]. A recent study
also used FRET to observe real-time changes in Ras signaling, which
functions as a molecular switch in many signaling cascades [59].
Fluorescent technologies have also been used for high-throughput
identification of cellular localization as well.