Enforcing Order on Signaling

See allHide authors and affiliations

Science  12 Mar 2010:
Vol. 327, Issue 5971, pp. 1335-1336
DOI: 10.1126/science.1187865

Mechanical forces provide important regulatory information that directs development. Even throughout adulthood, tissue homeostasis remains tightly linked to tensional homeostasis, the perturbation of which often leads to chronic conditions such as cardiovascular disease, arthritis, and cancer (1, 2). Environmental cues affect cell behavior by triggering signal transduction networks, but how cells actually integrate mechanical cues with these biochemical networks remains largely unresolved. On page 1380 in this issue, Salaita et al. describe how mechanical force, spatial organization of large clusters of cell surface receptors, and receptor-mediated signal transduction are coupled (3). Disruption of this mechanical-coupling mechanism in tumor cells may explain the invasive characteristics of aggressive, metastatic cancers.

Most models of cellular “mechanotransduction” incorporate the idea that proteins subjected to force undergo conformational changes that alter their function. One example is illustrated by the force-dependent assembly of integrin adhesion structures called focal adhesions. An integrin is a cell surface transmembrane receptor that interacts with the extracellular matrix outside the cell and with cytoskeletal and signaling molecules through its intracellular domain. In a “protein-centric” model of mechanotransduction, cytoskeletal tension on integrins and associated focal adhesion proteins drives the unfolding of key adaptor proteins that are linked to the “tensed” integrin. These conformational changes reveal binding sites that permit association with a network of proteins essential for signal transduction (4).

In contrast to this classic protein-based mechanotransduction model, Salaita et al. postulate that mechanical cues can impose spatial patterning on cell surface receptors that alter their signaling function. Many transmembrane receptors at the cell surface assemble into dimers, trimers, or higher-order oligomers to initiate intracellular signaling. In a “membrane-centric” mechanotransduction mechanism (5), Salaita et al. propose that either applied force or cellular tension can affect lateral receptor movement within the membrane to change the size (degree of oligomerization) of receptor clusters, or alter their spatial organization throughout the membrane. Consequently, different cellular responses emerge depending on the spatial configuration of receptors.

Opposing forces.

EphA2 receptors on one cell membrane bind to ephrin ligands on an apposing membrane. In normal cells, receptors oligomerize and activate signaling pathways that control cell growth, survival, and movement. Contractile forces of the cytoskeleton drive oligomer assembly, whereas mechanical impediments (whose source is unclear) on receptor or ligand movement restrict assembly and dictate the organization of oligomers in the membrane. In tumor cells, receptor movement is more efficient, resulting in larger clusters that trigger distinct responses including cytoskeletal changes and more robust activation of signaling molecules. Normal mechanical constraints may be lost in tumor cells.


To test their predictions, Salaita et al. evaluated signaling between the transmembrane receptor tyrosine kinase EphA2 and its ligand, ephrin-A1, which is normally presented on an apposed cell membrane. After binding to their ligands, Eph receptors assemble into dimers or small oligomers, become active, and trigger signaling cascades (6). Salaita et al. show that over time, ligand-bound EphA2 receptors expressed in a human breast cancer cell line condensed into very large macroscopic clusters exceeding several micrometers in diameter and presumably containing thousands of receptors (see the figure). Two opposing forces controlled this process: Myosin motors, which act on the cytoskeleton, transported associated receptors into the clusters, whereas mechanical restriction of ligand (or receptor) movement in the membrane impaired clustering. These results suggest that EphA2 receptor organization is coupled to the local mechanical environment.

Salaita et al. further developed a model membrane system to mechanically control the spatial organization of receptor clusters. A fluid lipid bilayer containing ephrin ligand was supported on a glass substrate and then presented to live cells expressing EphA2. With advanced lithography techniques, grid-like barriers were introduced into the model membrane to physically restrict ligand within corrals of defined size. These “spatial mutations” of ligand perturbed the mobility of ligand-bound EphA2 receptors and consequently modified receptor cluster size and distribution throughout the cell membrane. Although EphA2 was locally activated regardless of spatial constraint, cellular response such as actin organization depended on receptor organization.

Cell signaling networks are compromised in cancer, and deregulation of mechanical coupling to these networks may represent a major yet under-appreciated underlying disease mechanism (1, 7, 8). Salaita et al. found that the extent of active EphA2 receptor movement into clusters correlated positively with cell invasiveness and metastatic potential. They also identified the association of Src and CD44, molecules involved in tumorigenesis and metastasis, with increased radial movement of EphA2 into clusters. Thus, tumor cells may deregulate the normal mechanical constraints on EphA2 movement in the membrane to acquire an aggressive phenotype, possibly by collaborating with molecules already implicated in cancer.

Eph receptors guide cell migration and tissue patterning during specific stages of embryogenesis and tissue development (6). Given the exquisite mechanical movements that underlie these fundamental processes (9), every push, twist, and turn made by cells could relay contextual information essential for proper coordinated development. Subtle changes in spatial organization of these and other transmembrane receptors could lead to developmental abnormalities and functional aberrations. Receptors that regulate programmed cell death (apoptosis) also form lower-order clusters (trimers) upon ligand binding and assemble into higher-order structures after association with the actin cytoskeleton (10, 11). These larger clusters are commonly formed in tumor cells after exposure to their cognate ligand (TRAIL) (11). The notion that normal mechanical constraints on these receptors may be lost in tumors could explain the enhanced sensitivity of transformed cells to TRAIL.

Further studies should clarify the resistance of Eph receptors to clustering in normal cells, and whether ephrin ligands are naturally constrained in cell membranes. For instance, it is not clear whether local lipid organization in the membrane could influence these dynamics. Nor is it clear why large EphA2 clusters efficiently elicit distinct cellular signals, whereas smaller but more numerous complexes do not. Physical perturbations unique to tumors could be targeted with the goal of overcoming resistance to treatments that so frequently plague current antitumor therapies.


View Abstract

Stay Connected to Science

Navigate This Article