Biomedical Engineering Reference
In-Depth Information
respect to a large portion, if not all, of the "gene dimensions" of the state space,
this would be strongly suggestive of an entry into an attractor state. Initial stud-
ies for in vitro neutrophil differentiation suggest that this is in fact the case (Fig-
ure 4).
Such network perturbation experiments would, when performed systemati-
cally and in many cell types, allow us to obtain a first glance at the structure of
the "attractor landscape" of the genome without knowledge of all the details of
the "wiring diagram" of the genomic regulatory network. Unfortunately, the
manipulation of the activation state of individual genes in living cells is still
cumbersome compared to the situation in computer-simulated networks, such
that systematic network perturbations that may reveal more detailed information
about the structure of the attractor are still limited. Nevertheless, such experi-
ments are a first small step toward the molecular characterization of Wadding-
ton's "epigenetic landscape" (Figure 5) and an essential intermediate step toward
our understanding of how the genome maps into the phenome.
4.2.5. Hierarchical Considerations: Signaling Networks Beyond the Cell
Similar to the structural networks discussed above, information networks
will extend beyond the limits of intracellular regulation. Cells in various states
(attractors) signal to each other via physical cell-cell contacts, soluble cytokines,
and insoluble matrix scaffolds, thus forming an extracellular communication
network. The dynamics of such "cellular networks" can also be viewed in a
framework of state space concepts, with stable behavioral modes that involve
many cell types and their secreted products representing a coherent, robust
physiological program of the tissue, such as inflammation, immune response,
regeneration, development, and toxicity. These distinct "tissue fates" also exhibit
properties of state space trajectories and attractors. For instance, immune system
decisions between mutually exclusive, robust responses are common place, as in
the Th1/Th2 dichotomy in the T-cell immune response (54; see also Part III,
chapter 4.1, by Segel, this volume). Moreover, it now appears that cancer is not
just a cell-autonomous disease in which mutant cells evolve to proliferate in an
unconstrained fashion (see this volume, Part III, chapters 6.1, by Pienta, and 6.2,
by Solé, Gonzales Garcia, and Costa), but also involves a dysregulation at the
tissue level in which non-mutant cells of the tumor bed (stroma) and its blood
vessels play a central role—thus, the cancerous tumor itself may be an unfortu-
nately stable "tissue fate" (see also this volume, Part III, chapter 6.3, by Mansury
and Deisboeck).
Biomedical research is only at the beginning of appreciating these higher-
level interactions as formal networks, because most of leading edge "systems
biology" research is still carried out on single-cell model organisms focusing on
individual molecular pathways. But experiments in the near future that system-
