Professional Documents
Culture Documents
of Pages 12
Review
Towards a Dynamic
Understanding of Cadherin-
Based Mechanobiology
Brenton D. Hoffman1,* and Alpha S. Yap2
Cadherin-based cell–cell adhesions are a primary determinant of tissue struc- Trends
ture. For several decades, it had been thought that the primary function of these Cadherin complexes bear mechanical
ubiquitous structures was to resist external mechanical loads. Here we review forces that arise from cellular contrac-
tility in adherent cells. Coupling to acto-
recent evidence that cadherins also couple together the force-generating acto- myosin via association with myosin-
myosin cytoskeletons of neighbouring cells, serve as potent regulators of the bound F-actin leads to contractile
actomyosin cytoskeleton, and activate diverse signalling pathways in response forces being applied to cadherins and
their associated proteins.
to applied load. In considering the force sensitivity of the molecular-scale
processes that mediate these events, we propose a dynamic picture of the Bond properties within the cadherin–
force-sensitive processes in cell–cell contacts. This quantitative and physical catenin complex influence mechano-
sensitivity of cadherin junctions. Catch
understanding of the mechanobiology of cadherin cell–cell junctions will aid bonds have been identified to mediate
endeavours to study the fundamental processes mediating the development adhesive binding between cadherin
ectodomains and the association of
and maintenance of tissue structure.
the cadherin–catenin complex with
F-actin.
framework has guided research in the field for many years. Recently, though, it has come to be
appreciated that the relationship between force and adhesion is more complex. A key insight is
that many of the forces that cadherins experience are generated by their host and neighbouring
cells [4–8], which can contribute to the tension that tissues display under physiological circum-
stances [9]. Cadherin adhesion thus serves to couple the contractile actomyosin cortices of cells
together [10]. Furthermore, cell signalling at cell–cell junctions can promote actin assembly and
myosin activation, leading to the active generation of contractility [11]. It is also increasingly
1
apparent that cells sense, interpret, and respond to those applied forces through a poorly Department of Biomedical
Engineering, Duke University, Durham,
understood process termed mechanotransduction [12,13]. NC 27708-0281, USA
2
Division of Cell Biology and
The application of force to junctions affects many aspects of their biological and mechanical Molecular Medicine, Institute for
Molecular Bioscience, The University
function. For instance, on the molecular scale, force is a key determinant of the strength of of Queensland, Brisbane QLD 4072
cadherin–cadherin interactions as well as the composition of cell–cell adhesions [14–16]. On Australia
larger length scales, this leads to the emergence of tissue level tensions that are at least partially
driven by actomyosin contractility within the constituent cells [8,9]. Such tension contributes to
*Correspondence:
morphogenetic events such as cellular rearrangements and cell extrusion. Thus, the active brenton.hoffman@duke.edu
generation, transmission, and sensing of mechanical forces play integral roles in cadherin (B.D. Hoffman).
biology. Arguably, this is one of the major recent developments in our understanding of the cell
biology of these ubiquitous adhesion systems.
The realisation that cadherin junctions are mechanically sensitive structures implies that physical
principles guide and constrain the behaviour of these biological systems. In short, cadherin
junctions and the tissues that they support must obey quantitative physical laws and a
consideration of these laws can guide our understanding of their cell biology. In this article
we review recent developments important in our understanding of the mechanobiology of
cadherin adhesion systems. We first focus on molecular developments that reveal the important
roles that bond properties play in the biochemistry of cadherin complexes and discuss a range of
mechanosensitive biological processes that involve cadherin interactions. But this poses an
important challenge: molecular mechanisms and biological outcomes operate on vastly different
time scales that do not readily map to one another. Instead, there must be regimes that operate
on intermediate time scales to mediate between these two levels. How those intermediate time
scales arise is an important challenge for future research and we discuss some ways in which
this problem might be addressed through the application of quantitative and physical principles.
(D) Mechanoresponse
(Short mes)
Force
Force
(E) Mechanoresponse
(Long mes)
Force
Force
Mechanical loading can alter molecular associations, ion channels, and even affect cytoskeletal
networks [20]. The detection of this signal (mechanosensing, Figure 2C), typically through forma-
tion of new protein–protein interactions, then elicits a response in a signalling pathway, gene
transcription, or cell behaviour (mechanoresponse, Figure 2D,E). Notably, mechanoresponses can
occur over a variety of time scales. These events are often treated as switch-like as there is no
accounting for the dynamics, or potential reversibility, within, or between, each step. Thus, in this
picture, if a strong enough force is applied, protein conformation changes are induced and
mechanoresponses are initiated. However, it should be noted that force-sensitive molecular
interactions are often allosteric [21], and it is their probability of residing in a particular conformation
that is biased by the application of force.
The crystal structures of a number of classical cadherins indicate that isolated cadherin
ectodomains can adopt either of two distinct trans-binding interactions [28,29] (Figure 1B).
In one conformation, opposing cadherins insert a conserved tryptophan residue into a hydro-
phobic pocket in their binding partner generating a strand-swapped dimer. In the second
conformation, extensive interactions between the N termini of the ectodomains yield an X-dimer
[30]. The X-dimer is thought to be an intermediate stage towards generating the strand-
swapped dimer. Interestingly, E-cadherin mutants that favour the X-dimer displayed adhesive
bond lifetimes that increased when force was applied [27,31]. Thus, X-dimers have features of
catch bonds, whose ability to strengthen in response to force would be predicted to allow initial
bonds to resist tensile force and thereby stabilise cell–cell interactions. However, the strand-
swapped dimers, which are thought to be more prevalent in mature adhesions, mediate longer
interactions at no or low forces compared with the X-dimer configuration [27]. Together, these
data suggest that the X-dimer plays a role in stabilising nascent linkages, while strand-swap
dimers may mediate force-enhanced dynamics to enable cell rearrangements at large loads. At
high loads strand-swapped dimers of E-cadherin also displayed slip bond-like properties [27],
implying that other mechanisms, such as cadherin clustering, might be necessary to stabilise
junctions under high load [32–35].
Force sensitivity also influences the bonds between cadherins and the actin cytoskeleton.
Extensive biological evidence indicates that cadherins mechanically interact with actin filaments
[10,22,36,37]. This association was often thought to entail /-catenin, which can bind directly to
F-actin. However, a long-standing conundrum arose when minimal cadherin–catenin com-
plexes reconstituted from purified components failed to bind F-actin [38,39]. This issue has now
been resolved by the demonstration that the association between a minimal cadherin/catenin
complex and F-actin is force-sensitive (Figure 1C). Recently, single molecule optical tweezers
were used to show that applying force substantially increases the lifetime of interactions between
minimal cadherin complexes and actin filaments, an interaction which required /-catenin.
Kinetic modelling was consistent with a system where the cadherin/catenin–F-actin interaction
represented a two-state catch bond [21]. Catch bonds may then represent a common mecha-
nism that reinforces the physical and mechanical coupling of cadherin adhesion to the cyto-
skeleton upon application of force.
There are likely to be many other force-sensitive proteins within adherens junctions. Proteins
such as /-actinin [45] and formins [46] that are found at cadherin junctions have been reported
to show force sensitivity [47,48], although in other contexts. It is also possible that both force-
sensitive and force-insensitive interactions exist at junctions. For example, although the associ-
ation between vinculin and /-catenin appears to require the application of force [40,41,43],
vinculin can also bind to b-catenin [15]. Whether this latter association is modulated by force
remains to be determined, but it raises the interesting possibility that the incorporation of some
proteins into cadherin complexes may be tuned by force, rather than be an all-or-nothing
phenomenon.
Cadherin-Dependent Mechanoresponses
The cellular processes that respond to cadherin mechanosensing are only just beginning to be
characterised, but they appear to encompass diverse phenomena that operate on fairly
disparate time scales (Figure 2D,E). When analysed on short time scales (seconds–minutes)
cadherin mechanotransduction most evidently affects dynamic processes of the junctional actin
cytoskeleton. These processes often result in adhesion strengthening, or the mechanical
reinforcement of the load-bearing structure (i.e., the adherens junction). For example, the
recruitment of vinculin to cadherin junctions via its association with /-catenin is force-sensitive
[40] (Figure 2D). Activated vinculin can bind F-actin directly [49], which likely provides additional
linkages to reinforce cadherin–actin binding in response to force [50]. On slightly longer time
scales (tens of seconds to tens of minutes), load-bearing structures can also be reinforced
through enhanced assembly. Vinculin can recruit proteins of the Ena/VASP family, which
promote barbed-end growth of actin filaments to support junctional actin assembly [16], to
the adherens junction in response to applied loads (Figure 2E). There are likely to be many other
mechanisms that contribute to adhesion strengthening in different ways, as several other
proteins have been implicated in the maintenance of cell–cell contacts [51,52].
On very long time scales (hours), cadherin mechanosensing can potentially influence processes
such as single [53] or collective [54] cell migration, and mitotic spindle orientation [55,56]. These
are known to be influenced by tissue mechanics and require cadherins. However, whether they
are linked by cadherin mechanotransduction remains to be directly tested. If so, they are likely to
involve signalling pathways that act over longer time scales than those that influence the dynamic
junctional cytoskeleton. This is exemplified by the recent demonstration that MDCK monolayers
were induced to enter the cell cycle and proliferate when subjected to mechanical strain over
time courses of hours. This was attributable to the activation of Yap1 and canonical b-catenin
signalling through mechanisms that were perturbed by expression of a dominant-negative
E-cadherin mutant [57].
The importance of considering a role for dynamics is well illustrated by two sets of observations.
First, protein linkages in cell–cell adhesions are dynamic. In single molecule measurements, both
cadherin–cadherin linkages and the linkages of cadherin complex with actin last at most several
seconds [21,25]. While in cells these linkages appear somewhat longer lived, potentially due to
the stabilising effects of interactions not present in the single molecule experiments, the
structures are still dynamic, as indicated by both fluorescence recovery after photobleaching
(FRAP) [39,58] and optical highlighting techniques [59]. Thus, even mature junctions exhibit
dynamics and display a steady-state turnover of constituent proteins. A second example comes
from the recent observation that while force-dependent conformational changes in /-catenin
were detected almost instantaneously after the application of load (<5 s), its binding partner,
vinculin, was recruited at a rate that was approximately six times slower [42]. Neither the
presence of continuous molecular dynamics within nor the separation of time scales associated
with mechanotransduction, for example, the conformation change in /-catenin, and mecha-
nosensing, for example, vinculin binding, fit with a switch-like model for force-sensitive pro-
cesses. Instead these data imply that on the molecular level, the mechanobiology of cadherin
Physical principles and quantitative measurements can be used to reveal key aspects to
understanding how dynamics contribute to cadherin mechanobiology. One such insight is that
duration of mechanical signalling will be influenced by the effect of force on the mechanical
Force Force
Mechanotransmission
Mechanotransducon
Force Force
Mechanosensing
Force Force
Mechanoresponse
(Short mes)
Force Force
Mechanoresponse
(Long mes)
Force Force
Stable reinforcement
Figure 3. A Dynamic Picture of the Mechanobiology of Cadherin–Catenin. Transmission of forces through the
cadherin–catenin complex will be limited by the ability of the molecular constituents to resist applied loads. Thus, while force
can catalyse force-activated protein conformation changes and cell signalling events (shown in the right branch), it will also
enhance the rates of force-induced bond disassociation in the case of slip bonds and in certain force regimes in catch–slip
bonds (shown in left branch). Thus, there is an implicit competition between the termination of mechanotransmission and in
the initiation of mechanosensitive signalling in loaded cadherin–catenin complexes. Notably, force-induced dissociation
could happen through several distinct processes, which are shown within in the dashed box.
linkages within the cadherin complex (Figure 3). On a molecular level, mechanotransmission is
dictated by the ability of protein–protein interactions to resist mechanical loading. In the extreme
case, breaking any link in a mechanotransmission pathway will terminate the mechanical signals
and cause any mechanosensitive proteins to revert to their unloaded state [43]. A key issue is the
location where breakages may occur in the cadherin–catenin complex. Measurements of the
affinity between cadherin–catenin complex constituents in their unloaded configuration suggest
that the weakest link is between /-catenin and actin [39,60,61]. However, bond strength under
load can be very different [2_TD$IF]from that suggested by the equilibrium-binding affinities, especially in
the case of catch bonds. For instance, biotin–streptavidin interactions, thought to be among the
strongest noncovalent interactions in nature are slip bonds [62] and rupture more readily than
/5b1 integrin and fibronectin interactions [63], which are catch bonds [64]. Based on the multiple
break points found in focal adhesions [65,66], we predict the existence of several positions for
force-induced dissociation of the cadherin–catenin complex (Figure 3). Additionally, while we
discussed these phenomena mostly in the context of an individual protein complex, cell–cell
contacts are composed of a large number of mechanosensitive linkages. Thus, even small shifts
in the typical lifetime of a key protein–protein interaction, or within a subpopulation of molecules
engaged in this interaction, could have significant consequences at the cell or tissue level.
Regardless of the exact molecular details, there is likely to be an innate competition between
mechanotransmission and the subsequent steps in force-sensitive processes. This can be
understood through an evaluation of time scales associated with these processes. If load can be
maintained on time scales longer than those associated with mechanotransduction and sensing,
which are likely dictated by protein diffusion and binding kinetics, then force-sensitive responses
will be initiated. This scenario is depicted in the right branch of Figure 3. Shown in the left branch
are the consequences of the rupture of a mechanical linkage, which leads to the unloading of the
complex and the termination of force-sensitive signalling. Notably, the sequence of events
depends on the exact rates of unbinding and the effect of force on these rates. As many bonds
are actually catch–slip bonds, strengthening at low forces and then weakening at larger forces
[25], applied load could be expected to either enhance assembly or disassembly of cadherin
adhesions in various contexts. These time scale-dependent effects could be particularly impor-
tant in mediating differential response to time-dependent mechanical stimuli, such as the forces
exerted on the vessel wall during blood flow or generated by a beating heart [20].
Similarly, cell–cell contacts can have diverse effects on the actomyosin cytoskeleton. For
instance, applied loads at cell–cell junctions often result in the enhanced stabilisation [71] or
assembly of actin [16] or of non-muscle myosin II [14] junctions. Conversely, applied loads can
Force-induced
disassembly
Cadherin
delivery
Actomyosin
assembly
Actomyosin
disassembly
Cadherin
removal
Force-induced
assembly
also induce the disassembly of actomyosin networks. Such stress-induced actin turnover is
acutely influenced by the state and organisation of the actin filaments [72]. Compressed or
buckled actin filaments are more likely to undergo spontaneous fragmentation than unloaded or
tensed filaments [73], while myosin activity preferentially disassembles branched actin networks
[74], compared with bundled networks. Indeed, stress-induced actin turnover can limit the
forces that can be generated in cell–cell contacts [75,76]. In the latter case, branched actin
networks found at lateral cell–cell junctions were preferentially affected by myosin-induced Outstanding Questions
contractility, whereas the bundled networks found at the apical zonulae adherentes were What other mechanical linkages within
relatively protected. Therefore, the cytoskeletal responses to junctional loads are likely to be cadherin junctions are force-sensitive?
Cadherin and /-catenin adhesive
complex and influenced by the organisational state of the actin networks that are bearing load, bonds are unlikely to be the sole
as well as the magnitude of that load itself. force-sensitive linkages in the cadherin
adhesome.
The effects of these rearrangement processes on force-sensitive signalling can be subtle, but
What signalling pathways are sensitive
may influence how long key pathways are active within cell–cell contacts. Epithelial monolayers to cadherin mechanotransduction?
can rapidly dissipate extrinsic stresses [55,56] and mechanisms that reduce load will tend to The induction of signalling resulting
terminate mechanical signalling, acting over time scales that would reflect how quickly load is from conformation changes in /-cat-
enin is unlikely to be the only force-
altered. This may occur by the dispersion of loads across reinforced scaffolds that bear the load
sensitive event capable of initiating
or the biochemically regulated decoupling of mechanical linkers. Specifically, the force-sensitive downstream signalling.
accumulation [67] or oligomerization [35] of adhesion receptors might be predicted to distribute
the stresses at those junctions, potentially unloading force-sensitive proteins that are found How do molecular mechanisms for
cadherin mechanotransduction inter-
there. Consistent with this idea, the molecular tension across E-cadherin was found to be
act across various time and length
constant between junctions that displayed varying cadherin intensity and mechanical load [69], scales to enable emergent, tissue level
implying that variations in the forces across cell–cell contacts are controlled by local accumula- morphogenetic events?
tion of E-cadherin, instead of differential loading of these proteins. Cadherin accumulation and
Does mechanical stimulation ultimately
adhesive strengthening may then constitute a mechanism that tends to limit the duration of affect cell signalling through local
mechanical signalling. mechanical feedback that is mediated
through cadherin mechanotransduc-
One important unanswered question is the identity of key factors that determine if an applied tion? Cadherin mechanotransduction
responds to contractile events gener-
load results in reinforcement or disassembly of cell–cell contacts. One possibility is that the ated within neighbouring cells. Can
resting tension within a cell–cell contact may dictate these responses. For example, the actin these local events form an intercon-
assembly necessary for junctional actomyosin and contractility [77] is itself tension-sensitive, nected network that couples adhesion,
signalling, and contractility resulting in
being decreased when myosin was blocked and increased when contractility was stimulated
emergent properties that are evident at
[16]. Taken together, these observations suggest the hypothesis that when cell-generated the tissue level?
forces are low, applied forces can be supported for time scales long enough to initiate signals
that enhance assembly of the junction. By contrast, if cell-generated forces are high, applied
loads cannot be supported for sufficient times, force-sensitive stabilisation pathways are
therefore not activated, and ultimately disassembly of cell–cell contacts occurs. Competition
between these processes may ultimately serve to return the system towards its set point and
promote mechanical homeostasis.
Concluding Remarks
The ultimate goal of most studies of cadherin mechanobiology is to identify the principles that
determine tissue structure (see Outstanding Questions). Major progress has come from the
realisation that many morphogenetic events are driven by actomyosin-based contractility that
is coupled to cadherin-based cell–cell junctions [78]. These include common, conserved pro-
cesses such as tissue invagination [79] and cell intercalation during convergent-extension move-
ments [80], which entail forces that are generated by pulsatile apical actomyosin networks [7,81].
Here we suggest that a dynamic understanding of cadherin mechanobiology may provide a clearer
insight into these processes. Overall, it is possible to regard the response to force of adhesion and/
or its associated cytoskeleton as a form of mechanical feedback [82], which will ultimately influence
mechanical signalling as has recently been observed in studies of cell intercalation during Dro-
sophila melanogaster germband extension [83]. It is likely that further work focused on integrating
the advances discussed in this review with optical techniques capable of detecting force-activated
protein dynamics, conformation, and biochemical activity (e.g., FRAP, FRET, and optical highlight-
ing) will be important to building a complete picture of how biological forces promote tissue
structure. As cancer and atherosclerosis are associated with alterations in tissue structure, it is
possible that defects in the dynamic force-sensitive feedback mechanism are associated with the
onset of these and other such force-sensitive diseases.
Acknowledgments
B.D.H. is supported by a Searle Scholars Award, a Basil O’Connor Starter Scholar Award from the March of Dimes, and a
CAREER Award from the National Science Foundation. A.S.Y. is supported by the National Health and Medical Research
Council of Australia (1044041, 1037320).
References
1. Levine, E. et al. (1994) Selective disruption of E-cadherin function 25. Rakshit, S. and Sivasankar, S. (2014) Biomechanics of cell adhe-
in early Xenopus embryos by a dominant negative mutant. Devel- sion: how force regulates the lifetime of adhesive bonds at the
opment 120, 901–909 single molecule level. Phys. Chem. Chem. Phys. 16, 2211–2223
2. Takeichi, M. (1988) The cadherins: cell–cell adhesion molecules 26. Roca-Cusachs, P. et al. (2012) Finding the weakest link: exploring
controlling animal morphogenesis. Development 102, 639–655 integrin-mediated mechanical molecular pathways. J. Cell Sci.
3. Delva, E. et al. (2009) The desmosome. Cold Spring Harb. Per- 125, 3025–3038
spect. Biol. 1, a002543 27. Rakshit, S. et al. (2012) Ideal, catch, and slip bonds in cadherin
4. Conway, D.E. et al. (2013) Fluid shear stress on endothelial cells adhesion. Proc. Natl. Acad. Sci. U.S.A. 109, 18815–18820
modulates mechanical tension across VE-cadherin and PECAM- 28. Harrison, O.J. et al. (2011) The extracellular architecture of adhe-
1. Curr. Biol. 23, 1024–1030 rens junctions revealed by crystal structures of type I cadherins.
5. Borghi, N. et al. (2012) E-cadherin is under constitutive actomyo- Structure 19, 244–256
sin-generated tension that is increased at cell–cell contacts upon 29. Brasch, J. et al. (2012) Thinking outside the cell: how cadherins
externally applied stretch. Proc. Natl. Acad. Sci. U.S.A. 109, drive adhesion. Trends Cell Biol. 22, 299–310
12568–12573 30. Harrison, O.J. et al. (2010) Two-step adhesive binding by classical
6. Rauzi, M. et al. (2010) Planar polarized actomyosin contractile flows cadherins. Nat. Struct. Mol. Biol. 17, 348–357
control epithelial junction remodelling. Nature 468, 1110–1114 31. Manibog, K. et al. (2014) Resolving the molecular mechanism of
7. Martin, A.C. et al. (2009) Pulsed contractions of an actin–myosin cadherin catch bond formation. Nat. Commun. 5, 3941
network drive apical constriction. Nature 457, 495–499 32. Yap, A.S. et al. (1997) Lateral clustering of the adhesive ectodo-
8. Martin, A.C. et al. (2010) Integration of contractile forces during main: a fundamental determinant of cadherin function. Curr. Biol.
tissue invagination. J. Cell Biol. 188, 735–749 7, 308–315
9. Fernandez-Gonzalez, R. et al. (2009) Myosin II dynamics are 33. Wu, Y. et al. (2015) Actin-delimited adhesion-independent clus-
regulated by tension in intercalating cells. Dev. Cell 17, 736–743 tering of E-cadherin forms the nanoscale building blocks of adhe-
10. Maitre, J.L. et al. (2012) Adhesion functions in cell sorting by rens junctions. Dev. Cell 32, 139–154
mechanically coupling the cortices of adhering cells. Science 34. Truong Quang, B.A. et al. (2013) Principles of E-cadherin supra-
338, 253–256 molecular organization in vivo. Curr. Biol. 23, 2197–2207
11. Ratheesh, A. and Yap, A.S. (2012) A bigger picture: classical 35. Strale, P.O. et al. (2015) The formation of ordered nanoclusters
cadherins and the dynamic actin cytoskeleton. Nat. Rev. Mol. Cell controls cadherin anchoring to actin and cell–cell contact fluidity. J.
Biol. 13, 673–679 Cell Biol. 210, 333–346
12. Leckband, D.E. and de Rooij, J. (2014) Cadherin adhesion and 36. Lambert, M. et al. (2002) Dynamics of ligand-induced, Rac1-
mechanotransduction. Annu. Rev. Cell Dev. Biol. 30, 291–315 dependent anchoring of cadherins to the actin cytoskeleton. J.
13. Ladoux, B. et al. (2010) Strength dependence of cadherin-medi- Cell Biol. 157, 469–479
ated adhesions. Biophys. J. 98, 534–542 37. Kwiatkowski, A.V. et al. (2010) In vitro and in vivo reconstitution of
14. Smutny, M. et al. (2010) Myosin II isoforms identify distinct func- the cadherin–catenin–actin complex from Caenorhabditis elegans.
tional modules that support integrity of the epithelial zonula adhe- Proc. Natl. Acad. Sci. U.S.A. 107, 14591–14596
rens. Nat. Cell Biol. 12, 696–702 38. Drees, F. et al. (2005) Alpha-catenin is a molecular switch that
15. Peng, X. et al. (2010) Vinculin regulates cell-surface E-cadherin binds E-cadherin–beta-catenin and regulates actin-filament
expression by binding to beta-catenin. J. Cell Sci. 123, 567–577 assembly. Cell 123, 903–915
16. Leerberg, J.M. et al. (2014) Tension-sensitive actin assembly 39. Yamada, S. et al. (2005) Deconstructing the cadherin–catenin–
supports contractility at the epithelial zonula adherens. Curr. Biol. actin complex. Cell 123, 889–901
24, 1689–1699 40. Yonemura, S. et al. (2010) alpha-Catenin as a tension transducer that
17. Roux, K.J. et al. (2012) A promiscuous biotin ligase fusion protein induces adherens junction development. Nat. Cell Biol. 12, 533–542
identifies proximal and interacting proteins in mammalian cells. J. 41. Choi, H.J. et al. (2012) /E-Catenin is an autoinhibited molecule that
Cell Biol. 196, 801–810 coactivates vinculin. Proc. Natl. Acad. Sci. U.S.A. 109, 8576–8581
18. Guo, Z. et al. (2014) E-cadherin interactome complexity and 42. Kim, T.J. et al. (2015) Dynamic visualization of alpha-catenin
robustness resolved by quantitative proteomics. Sci. Signal. 7, rs7 reveals rapid, reversible conformation switching between tension
19. Van Itallie, C.M. et al. (2014) Biotin ligase tagging identifies proteins states. Curr. Biol. 25, 218–224
proximal to E-cadherin, including lipoma preferred partner, a reg- 43. Yao, M. et al. (2014) Force-dependent conformational switch of
ulator of epithelial cell–cell and cell–substrate adhesion. J. Cell Sci. alpha-catenin controls vinculin binding. Nat. Commun. 5, 4525
127, 885–895 44. Carisey, A. et al. (2013) Vinculin regulates the recruitment and
20. Hoffman, B.D. et al. (2011) Dynamic molecular processes mediate release of core focal adhesion proteins in a force-dependent
cellular mechanotransduction. Nature 475, 316–323 manner. Curr. Biol. 23, 271–281
21. Buckley, C.D. et al. (2014) The minimal cadherin–catenin complex 45. Tang, V.W. and Brieher, W.M. (2012) /-Actinin–4/FSGS1 is
binds to actin filaments under force. Science 346, 1254211– required for Arp2/3-dependent actin assembly at the adherens
1254218 junction. J. Cell Biol. 196, 115–130
22. Ratheesh, A. et al. (2012) Centralspindlin and alpha-catenin regu- 46. Carramusa, L. et al. (2007) Mammalian diaphanous-related formin
late Rho signalling at the epithelial zonula adherens. Nat. Cell Biol. Dia1 controls the organization of E-cadherin-mediated cell–cell
14, 818–828 junctions. J. Cell Sci. 120, 3870–3882
23. Cai, D. et al. (2014) Mechanical feedback through E-cadherin 47. Kozlov, M.M. and Bershadsky, A.D. (2004) Processive capping by
promotes direction sensing during collective cell migration. Cell formin suggests a force-driven mechanism of actin polymerization.
157, 1146–1159 J. Cell Biol. 167, 1011–1017
24. Grashoff, C. et al. (2010) Measuring mechanical tension across 48. Ferrer, J.M. et al. (2008) Measuring molecular rupture forces
vinculin reveals regulation of focal adhesion dynamics. Nature 466, between single actin filaments and actin-binding proteins. Proc.
263–266 Natl. Acad. Sci. U.S.A. 105, 9221–9226
49. Carisey, A. and Ballestrem, C. (2011) Vinculin, an adapter protein 66. Maruthamuthu, V. et al. (2010) Conserved F-actin dynamics
in control of cell adhesion signalling. Eur. J. Cell Biol. 90, 157–163 and force transmission at cell adhesions. Curr. Opin. Cell Biol.
50. le Duc, Q. et al. (2010) Vinculin potentiates E-cadherin mechano- 22, 583–588
sensing and is recruited to actin-anchored sites within adherens 67. Liu, Z. et al. (2010) Mechanical tugging force regulates the size of
junctions in a myosin II-dependent manner. J. Cell Biol. 189, cell–cell junctions. Proc. Natl. Acad. Sci. U.S.A. 107, 9944–9949
1107–1115 68. Maruthamuthu, V. et al. (2011) Cell–ECM traction force modulates
51. Bazellieres, E. et al. (2015) Control of cell–cell forces and collective endogenous tension at cell–cell contacts. Proc. Natl. Acad. Sci. U.
cell dynamics by the intercellular adhesome. Nat. Cell Biol. 17, S.A. 108, 4708–4713
409–420 69. Sim, J.Y. et al. (2015) Spatial distribution of cell–cell and cell–ECM
52. Takeichi, M. (2014) Dynamic contacts: rearranging adherens junc- adhesions regulates force balance while maintaining E-cadherin
tions to drive epithelial remodelling. Nat. Rev. Mol. Cell Biol. 15, molecular tension in cell pairs. Mol. Biol. Cell 26, 2456–2465
397–410 70. Collinet, C. and Lecuit, T. (2013) Stability and dynamics of cell–cell
53. Weber, G.F. et al. (2012) A mechanoresponsive cadherin–keratin junctions. Prog. Mol. Biol. Transl. Sci. 116, 25–47
complex directs polarized protrusive behavior and collective cell 71. Kovacs, E.M. et al. (2011) N-WASP regulates the epithelial junc-
migration. Dev. Cell 22, 104–115 tional actin cytoskeleton through a non-canonical post-nucleation
54. Vedula, S.R. et al. (2012) Emerging modes of collective cell migra- pathway. Nat. Cell Biol. 13, 934–943
tion induced by geometrical constraints. Proc. Natl. Acad. Sci. U. 72. De La Cruz, E.M. and Gardel, M.L. (2015) Actin mechanics and
S.A. 109, 12974–12979 fragmentation. J. Biol. Chem. 290, 17137–17144
55. Campinho, P. et al. (2013) Tension-oriented cell divisions limit 73. Murrell, M.P. and Gardel, M.L. (2012) F-actin buckling coordinates
anisotropic tissue tension in epithelial spreading during zebrafish contractility and severing in a biomimetic actomyosin cortex. Proc.
epiboly. Nat. Cell Biol. 15, 1405–1414 Natl. Acad. Sci. U.S.A. 109, 20820–20825
56. Wyatt, T.P. et al. (2015) Emergence of homeostatic epithelial 74. Reymann, A.C. et al. (2012) Actin network architecture can deter-
packing and stress dissipation through divisions oriented along mine myosin motor activity. Science 336, 1310–1314
the long cell axis. Proc. Natl. Acad. Sci. U.S.A. 112, 5726–5731
75. Moore, T. et al. (2014) Self-organizing actomyosin patterns on
57. Benham-Pyle, B.W. et al. (2015) Cell adhesion. Mechanical strain the cell cortex at epithelial cell–cell junctions. Biophys. J. 107,
induces E-cadherin-dependent Yap1 and beta-catenin activation 2652–2661
to drive cell cycle entry. Science 348, 1024–1027
76. Wu, S.K. et al. (2014) Cortical F-actin stabilization generates
58. de Beco, S. et al. (2015) Mechanosensitive adaptation of E-cad- apical–lateral patterns of junctional contractility that integrate cells
herin turnover across adherens junctions. PLoS ONE 10, into epithelia. Nat. Cell Biol. 16, 167–178
e0128281
77. Verma, S. et al. (2012) A WAVE2–Arp2/3 actin nucleator appara-
59. Daneshjou, N. et al. (2015) Rac1 functions as a reversible tension tus supports junctional tension at the epithelial zonula adherens.
modulator to stabilize VE-cadherin trans-interaction. J. Cell Biol. Mol. Biol. Cell 23, 4601–4610
208, 23–32
78. Heisenberg, C.P. and Bellaiche, Y. (2013) Forces in tissue mor-
60. Hansen, S.D. et al. (2013) /E-Catenin actin-binding domain alters phogenesis and patterning. Cell 153, 948–962
actin filament conformation and regulates binding of nucleation
79. Martin, A.C. and Goldstein, B. (2014) Apical constriction: themes
and disassembly factors. Mol. Biol. Cell 24, 3710–3720
and variations on a cellular mechanism driving morphogenesis.
61. Pokutta, S. et al. (2014) Structural and thermodynamic character- Development 141, 1987–1998
ization of cadherin.beta-catenin.alpha-catenin complex formation.
80. Walck-Shannon, E. and Hardin, J. (2014) Cell intercalation from
J. Biol. Chem. 289, 13589–13601
top to bottom. Nat. Rev. Mol. Cell Biol. 15, 34–48
62. Merkel, R. et al. (1999) Energy landscapes of receptor–ligand
81. Bertet, C. et al. (2004) Myosin-dependent junction remodelling
bonds explored with dynamic force spectroscopy. Nature 397,
controls planar cell intercalation and axis elongation. Nature 429,
50–53
667–671
63. Jurchenko, C. et al. (2014) Integrin-generated forces lead to
82. Priya, R. et al. (2015) Feedback regulation through Myosin II
streptavidin–biotin unbinding in cellular adhesions. Biophys. J.
confers robustness on RhoA signaling at E-cadherin junctions.
106, 1436–1446
Nat. Cell Biol. Published online September 14, 2015. http://dx.doi.
64. Kong, F. et al. (2009) Demonstration of catch bonds between an org/10.1038/ncb3239
integrin and its ligand. J. Cell Biol. 185, 1275–1284
83. Munjal, A. et al. (2015) A self-organized biomechanical network
65. Chan, C.E. and Odde, D.J. (2008) Traction dynamics of filopodia drives shape changes during tissue morphogenesis. Nature 524,
on compliant substrates. Science 322, 1687–1691 351–355