Mapping the Genotype–Phenotype Relationship in Cellular Signaling Networks: Building Bridges Over the Unknown - Systems Biology: Networks, Models, and Applications - page 157

TEMPORAL REGULATION OF CELLULAR SIGNALING NETWORKS
In addition to the three tiers of information in a cellular signaling
network, across all these scales there is a temporal component. For
example, the binding of a ligand to its receptor may result in the acti-
vation of a transcription factor. This transcription factor may induce the
expression of a gene whose corresponding protein inhibits the activity
of the original ligand-receptor complex. The inhibition of the signaling
function may in turn arrest some migratory activity. An understanding
of the function of the ligand-receptor complex and the inhibitory protein
does not reflect the temporal elements of how the components are related
to each other, that is, the “sequence of events.” Temporal components
of cellular signaling systems have been perhaps most characterized in
autocrine signaling loops for which temporal characteristics are critical
elements (for an example, see [76]). While temporal regulation is critical
to cellular signaling network function, its analysis at a system level is
only beginning as methods for quantitative mapping of components,
interactions, and resultant phenotypes are emerging.
QUANTITATIVE MAPPING OF GENOTYPE-PHENOTYPE
RELATIONSHIPS
There are efforts to provide quantitative descriptions of components
and their relationships in each tier of cellular signaling networks. With
respect to the genotype component of cellular signaling, extensive
bioinformatics analysis has identified the complete set of protein
kinases in the human genome [77]. These bioinformatics analyses pro-
vide the foundation for more comprehensive quantitative analysis of
relationships between components.
Three poignant examples illustrate the quantitative mapping
to cell signaling phenotypes from the cataloging of genotype. First,
an extensive model incorporating G-protein coupled receptors,
G-proteins, and GTPase-activating proteins (GAPs) was analyzed to
characterize 16 distinct “states” or intermediate phenotypes [78]. These
“states” allowed for the clarification of previously paradoxical informa-
tion regarding the lack of GAP regulation of current amplitude
of G-protein-activated ion channels. Second, a mathematical representa-
tion of NF-kB and IkB dynamics clarified the role of various IkB isoforms
[79]. These results were experimentally verified with murine cell knock-
outs of the associated IkB isoforms. Third, an exhaustive stoichiometric
reconstruction of JAK-STAT signaling in the human B-cell resulted in a
novel, quantitative analysis of crosstalk and pathway redundancy [72].
Quantitative models of endpoint phenotypes have also generated
novel characterizations of genotype-phenotype relationships in cellular
signaling networks. Cell migration is the endpoint phenotype that has
been perhaps most thoroughly analyzed with quantitative models.