E-Cadherin: A Conductor of Cellular Signaling

Abstract

Upon engagement of E-cadherin or when mechanical force is applied, E-cadherin recruits cytoskeletal proteins and triggers various signal transduction cascades including PI3K, Src, Rho family GTPases, kinases, YAP/TAZ, AMPK, and other metabolic enzymes. These cascades modulate E-cadherin’s stability, viscosity, and its connection to the actin cytoskeleton, thereby reinforcing cell-cell adhesion.

Introduction

Adherens junctions act like cellular bridges, connecting neighboring epithelial cells and linking them to the cytoskeleton. This connection allows cells to sense and transmit forces, sparking biomechanical signaling and communication. E-cadherin was originally identified as one of the glycoproteins responsible for the homophilic cell-to-cell adhesion of neighboring cells. Since those initial observations were made, it has become evident that the role of E-cadherin is not limited to cell-cell adhesion. Among other functions, cadherins also sense and respond to external and internal forces and “tune” their adhesion to withstand mechanical stresses. Also, since their discovery over four decades ago, many signaling pathways that emanate from cadherin engagement, clustering, or mechanical strain have been identified, and the pace of discovery has not slowed down. In this brief review, the focus is restricted to a few selected topics, particularly concentrating on the signaling pathways downstream of epithelial or E-cadherin that impact the organization of the cytoskeleton and the cellular response to external cues.

Acting alone vs in concert

There is a long history of examining signaling downstream of E-cadherin in cells actively forming junctions. In many of these studies, E-cadherin engagement is monitored in cell-cell junctions which have been disassembled by removing calcium from the culture media and reassembled by restoring calcium to the cultures. Alternatively, signals have been studied in cells treated with cadherin ligands or with force applied. However, it is becoming increasing evident that cadherins do not act alone and their engagement using these types of approaches may also trigger other proteins in the cell membrane that could impact downstream signaling. For example, E-cadherin and EGFR co-localize and physically bind in cell-cell junctions, and there is some evidence of reciprocal regulation between the two with force on E-cadherin stimulating EGFR release from the complex^{1; 2}. Similarly, E-cadherin binds mechanosensitive ion channels known as Piezo1 and Piezo2, thereby providing the Piezo channels with a link to the actin cytoskeleton³. This tether allows the cytoskeleton to transmit mechanical forces to the Piezo channels, leading to their opening and activation³. Additionally, E-cadherin interacts with the Na⁺/HCO₃^- co-transporter, NBCn1, which is a key contributor to epithelial pH homeostasis⁴. These are only a sampling of the many proteins that have been reported to be affected by cadherins, and they demonstrate that while we generally attribute one signal to one downstream affect, that in the case of cadherins, other proteins may be involved.

Signaling through the catenins

Signals emanating from E-cadherin are mediated in part by the catenins, a family of proteins indirectly linking cadherins to the actin cytoskeleton. Among these, β-catenin directly binds to the E-cadherin cytoplasmic tail and also binds to α-catenin and vinculin, both of which connect to the actin cytoskeleton^5–8(Figure 1). β-catenin recruits α-catenin to the cadherin adhesion complex, where α-catenin unfurls under force, enabling its interaction with vinculin^{9; 10}. β-catenin or increased myosin II contractily which subsequently unfurls ⍺-catenin enhances the recruitment of vinculin^{8; 11–13}. Most adherens junction components including afadin and vasodilator-stimulated phosphoprotein (VASP) can bind complexes lacking ⍺-catenin¹⁴. In the absence of ⍺-catenin, β-catenin binds and maintains vinculin in its open conformation¹⁵.

Figure 1. Components of adherens junctions.

E-cadherin links neighboring cells to the actin cytoskeleton via adaptor proteins. The extracellular portion of E-cadherin contains five extracellular cadherin (EC) domains, the first of which interacts homophilically with the first EC domain on the adjacent cell. The intercellular portion of E-cadherin interacts with the adaptor proteins β-catenin and p120-catenin. β-catenin binds ⍺-catenin and vinculin which both connect to actin serving as a link to the cytoskeleton. Afadin interacts with ⍺-catenin and promotes its ability to interact with actin.

While ⍺-catenin is dispensable for recruitment of adherens junctional components, it is needed to promote adhesion strength through binding F-actin via a catch bond which strengthens under force^16–18. α-Catenin preferential interacts with regions of spiraled F-actin which form from myosin force generation in regions distal from motor binding sites¹⁹. The interaction of α-catenin with F-actin is critical for the clustering of the cadherin-catenin complex and is enhanced when tension unfolds α-catenin^{20; 21}.

Vinculin is phosphorylated at Y822 by Abelson tyrosine kinase which enables its recruitment to the cadherin adhesion complex in response to mechanical force²². This phosphorylation is essential for maintaining the balance between cell-cell and cell-matrix adhesions²³. Vinculin stabilizes the open conformation of ⍺-catenin and promotes its binding to F-actin^24–26. Vinculin, together with afadin, relieves ⍺-catenin autoinhibition enhancing the ability of α-catenin to bind to F-actin, resulting in stronger cellular adhesion and signaling²⁷.

Cadherin-mediated signaling to the Rho family of GTPases

E-cadherin is a key player in activating various Rho family GTPases, which are crucial for regulating the actin cytoskeleton²⁸. Among these, RhoA is activated at adherens junctions by several upstream guanine-nucleotide-exchange factors, such as p114 Rho guanine nucleotide exchange factor, RH-Rho guanine nucleotide exchange factor subfamily (p115RhoGEF/PDZ-RhoGEF/LARG), and ECT2^29–31, which catalyze the exchange of GDP for GTP (Figure 2). Furthermore, scaffolding proteins can further modulate the activities of guanine nucleotide exchange factors. For example, centralspindlin localization of Ect2 and plakophilin 4 localization of ARHGEF2 to adherens junctions is essential for RhoA and its effectors, myosin light chain kinase and myosin light chain, to facilitate cortical actin ring formation and actomyosin contraction^{29; 32}. Once activated, RhoA promotes actin assembly initiated by the formin, mDia1,^{33; 34} and boosts contractility by a signal transduction cascade which increases myosin II phosphorylation³⁵. This dynamic interplay stabilizes E-cadherin at adherens junctions, particularly during epithelial migration, where cells are constantly rearranging their cell-cell contacts³⁶.

Figure 2. Adherens junctions mediate parallel actin filament assembly via RhoA activation.

Linear actin bundles support junctional strength by generating intercellular tension. Various RhoA guanine nucleotide exchange factors (GEFs) are recruited or scaffolded by centralspindlin and plakophilin 4 (PKP4). These GEFs stimulate the activation of RhoA into the activite GTP bound form which activates Rho kinase (ROCK) and the formin, mDIA1. ROCK phosphorylates non-muscle myosin II, and mDIA1 assembles parallel actin filaments, collectively supporting adherens junction strength.

Regulating RhoA at cell-cell junctions is a finely tuned process that hinges on the activity of GTPase activating proteins which inactivate the small GTPase. These proteins, including the RhoA-specific GTPases, myosin-IXA and p190RhoGAP, are strategically localized to cell-cell junctions where they play a crucial role in negatively regulating RhoA^{37; 38}. Interestingly, the knockdown of p190RhoGAP-B, but not its close relative p190RhoGAP-A, is essential for the proper regulation of RhoA activity in response to mechanical tension³⁹. Adding another layer of complexity, Rho family GTPase 3 safeguards junctional RhoA by preventing the recruitment of p190RhoGAP-B⁴⁰. This intricate interplay underscores the meticulous regulation of RhoA at cell-cell junctions.

At nascent cell-cell adhesions, E-cadherin plays a crucial role by activating Rac1 (Figure 3)^{28; 41}. This activation prevents the endocytosis of E-cadherin⁴² and promotes the assembly of branched actin through the WAVE2–Arp2/3 pathway⁴³. The WAVE2 and Arp2/3 complexes are essential for forming dynamic, lamellipodia-like actin protrusions that maintain the integrity of adherens junctions. These processes ultimately stabilize cell-cell contacts, which are vital for regulating the polarity and barrier function of the epithelium, ensuring the preservation of epithelial tissue integrity and homeostasis.

Figure 3. Adherens junctions mediate branched actin assembly via Rac1 signaling.

Branched actin is necessary for nascent adherens junctions such as during migration. E-cadherin interacts with phosphoinositide 3-kinase (PI3K) and epidermal growth factor receptor (EGFR) which promote the activation of Rac1. Rac1 activation is promoted by various GEFs. Active GTP bound Rac1 promotes branched actin assembly via WAVE2–Arp2/3. Vinculin inhibits WAVE2–Arp2/3 decreasing branched actin polymerization regulating migration.

There are numerous mechanisms for Rac1 activation at cell-cell contacts. Signaling from E-cadherin to Rac1 involves EGFR and/or PI3K^{44; 45}. Rac1 is activated by guanine nucleotide exchange factors, such as Asef, ECT2, Tiam1, Tiam2, Trio, and Vav2⁴⁶. Additionally, ELMO2, a protein that forms a complex with the unconventional GEF Dock180, is required for Rac1 activation in cell-cell junctions⁴⁷. The sheer number of Rac activators motivated an examination of whether multiple pathways operate within a single cell or if different cell types utilize distinct activators. This analysis revealed that EGFR and DOCK180 both regulate Rac1 activation at nascent cell-cell adhesions in the same cell but control separate cellular events that work together to stabilize cell-cell junctions⁴⁴. More recently, the guanine nucleotide exchange factor, Tiam1 can modulate cell-cell junctions by binding, in the nucleus, to TRIM28, a protein involved in transcriptional regulation⁴⁸. Thus, multiple Rac1 activators are necessary in various locations within a cell to modulate cell-cell junctions effectively.

Like Rac1, Cdc42 activation is required for the formation of cell-cell contacts and plays a role in their stabilization, but its actions appear to be limited to a subset of epithelial cells⁴⁹. Activation of Rap1, another small GTPase, promotes the activation of Cdc42 and formation of adherens junctions^{50; 51}. Rap1 is stimulated by the guanine nucleotide exchange factors, C3G and RapGEF2, which interact with E-cadherin and β-catenin, respectively^{50; 52}. Additionally, Rap1 activates afadin and enhances afadin’s recruitment to multicellular junctions⁵³.

Thus, the Rho GTPases are like master conductors in the cellular orchestra, modulating the activity of various actin-binding proteins and other downstream effectors within the cell-cell contacts. This dynamic interplay allows cells to respond swiftly to extracellular cues. The intricate crosstalk among Rho GTPases and their context-dependent regulation of actin organization highlights their remarkable adaptability, complexity, and crucial role in shaping the epithelium.

Signaling to YAP/TAZ

When epithelial cells come into contact or experience mechanical force, two key players, YAP (Yes-associated protein) and TAZ (transcriptional coactivator with PDZ-binding motif), are affected. These transcriptional coactivators are crucial for controlling cell fate, proliferation, and survival. E-cadherin binding activates the Hippo signaling pathway, which inhibits YAP/TAZ activity and reduces cell proliferation⁵⁴. However, responses to other stimuli can be more complex and may vary depending on the cell type. For example, when epithelial cells are densely packed and inactive, applying mechanical force to E-cadherin stimulates the nuclear localization of YAP first, followed by β-catenin. This upregulation of YAP is necessary for cells to re-enter the cell cycle⁵⁵. Additionally, when cells are at high density, the contraction of the circumferential actin belt releases a protein called Merlin from E-cadherin. Merlin then moves into the nucleus and helps export YAP/TAZ back into the cytoplasm⁵⁶. Thus, the presence and activity of YAP/TAZ are finely regulated by cell contact, mechanical forces, and the Hippo signaling pathway, highlighting their critical roles in cellular behavior.

Regulation of viscosity of cell-cell junctions

Cells undergo dramatic shape changes to accommodate local mechanical constraints while maintaining cell–cell contact to ensure monolayer integrity. Central to these distortions is an ability to modulate the deformability of the cytoplasm (intracellular viscosity) and the junctions between cells (intercellular viscosity). Decreased intracellular viscosity promotes cell migration while increased intracellular viscosity inhibits cell migration⁵⁷. However, the mechanisms governing viscosity remain largely unexplored. It has been suggested that viscosity can be modulated by controlling the number of actin crosslinks⁵⁸. Recent studies indicate that engagement of E-cadherin at nascent cell-cell adhesions triggers a ligand independent phosphorylation and activation of EGFR which subsequently activates Rac1. Active Rac1 promotes the activation of two actin modifying proteins (WAVE2 and Arp2/3), thereby fine turning the junctional actin network by balancing the number of branched and linear junctional actin structures⁵⁹. Additionally, other research suggests that the viscosity of the Drosophila wing epithelium is reduced by the loss of p120-catenin. Further investigation is needed to understand how p120-catenin controls viscosity, but its ability to regulate E-cadherin appears to be crucial in this process⁶⁰.

Signaling to metabolic pathways

E-cadherin senses and responds to mechanical force by stimulating pathways that help cells reinforce their actin cytoskeletons and strengthen their connections with neighboring cells in a process known as cell stiffening. This process is energetically demanding, requiring increased actin polymerization, rearrangements, and elevated gene expression. To fuel these activities, E-cadherin triggers key metabolic pathways to supply the necessary energy^{61; 62}(Figure 4).

Figure 4. Adherens junctions coordinate metabolic response to provide energy for cellular stiffening.

The cellular response to force creates huge energy demands which cells meet by activating multiple pathways. E-cadherin is linked via Ankyrin G to GLUT1 enabling increased glucose uptake and glycolysis in response to force. Liver kinase B1 (LKB1) and calcium/calmodulin-dependent protein kinase kinase 2 (CAMKK2) phosphorylate and activate the master regulator of cell metabolism, AMP-activated protein kinase (AMPK). AMPK activates multiple downstream pathways including (1) p21-activated protein kinase 2 (PAK2) which inhibits the proapoptotic protein, BCL2 associated agonist of cell death (BAD), increasing cell survival. (2) AMPK directly phosphorylates non-muscle myosin II as well as indirectly promotes its phosphorylation by activating RhoA/ROCK. ROCK also promotes GLUT1 localization to the plasma membrane resulting in increased glucose uptake. (3) AMPK activates Abelson kinase (Abl) downstream of PAK2 which phosphorylates vinculin enabling its junctional presence supporting barrier function.

E-cadherin activates AMP-activated protein kinase (AMPK), a master regulator of cell metabolism. During the assembly of cell-cell junctions, AMPK can be activated by calcium/calmodulin-dependent protein kinase kinase 2 (CAMKK2) in regions of myosin II-containing actin bundles, and CAMKK2 inactivation leads to a loss of peripheral actomyosin bundles⁶³. Similarly, application of shear or tensile force onto E-cadherin triggers recruitment of liver kinase B1 (LKB1) to the cadherin adhesion complex and activation of AMPK⁶⁴. Inactivation of either LKB1 or CAMKK2 leads to a loss of actin bundles at sites of cell-cell contacts, suggesting a critical role for AMPK in maintaining the actin cytoskeleton in cells under force.

Force stimulated AMPK then stimulates a series of events that call for the increased uptake of glucose and its metabolism to ATP. AMPK recruits and phosphorylates p21-activated protein kinase 2 (PAK2) which suppresses the pro-apoptotic protein BAD, promoting cell survival under mechanical stress. PAK2 also activates Abelson tyrosine kinase (Abl), which phosphorylates tyrosine 822 in vinculin, resulting in increased RhoA activity and heightened contractility. Additionally, ankyrin G is recruited to E-cadherin, and ankyrin G tethers, through the spectrin actin cytoskeleton, E-cadherin to the glucose transporter 1 (GLUT1), facilitating increased glucose uptake at cell-cell contacts⁶⁵. Interestingly, in non-epithelial cells, RhoA/ROCK signaling also promotes GLUT1 localization to the plasma membrane, boosting glucose uptake⁶⁶.

Finally, force stimulated increases in AMPK signaling can also phosphorylate and activate PGC-1α, a master regulator of mitochondrial protein expression, which stimulates the hydrolysis of lipid droplets into fatty acids which provide fuel for cytoskeletal rearrangements⁶⁷. It is worth noting that cells have AMPK-independent mechanisms for providing energy for the cytoskeleton. Cyclic stretch increases phosphofructokinase-2/fructose-2,6-bisphosphatase 3 production of fructose-2,6-bisphosphate, an allosteric activator of phosphofructokinase-1, the enzyme that catalyzes the rate-limiting step in glycolysis⁶⁸. Thus, E-cadherin responds to force by stimulating multiple metabolic pathways that yield energy to support and reinforce its connection to the underlying cytoskeleton.

Signaling by kinases

For the sake of completeness, it is worth briefly mentioning two kinases activated by E-cadherin, Src and PI3K, have been the focus of much research over the years. Src kinase, activated by E-cadherin, has shown conflicting roles in cell-cell adhesion. Early studies revealed that constitutive activation of Src disrupts cell-cell adhesions and promotes epithelial to mesenchymal transition^{69; 70}. However, other studies suggest that Src might also positively influence cell-cell adhesion^{71; 72}. This discrepancy can be explained by the observation that the effect of Src on cell-cell adhesion depends on its level of activation. At lower levels, Src supports adhesion, while at higher levels, it inhibits it⁷².

Similarly, E-cadherin binds PI3K, activating PI3K/Akt signal transduction that modulates cell survival, proliferation, and differentiation^{73; 74}. PI3K is essential for forming productive adhesive contacts following initial homophilic ligation⁷⁴. However, in cancer contexts, E-cadherin presence is thought to suppress PI3K signaling, and its loss may contribute to dysregulated PI3K/Akt signaling, promoting uncontrolled growth and survival⁷⁵. Thus, Src and PI3K likely support the initial formation of cell-cell contacts. Beyond that, their roles are complex and depend on the cellular context and activation magnitude.

Endocytosis and recycling of E-cadherin

E-cadherin undergoes endocytosis, sorting, and recycling back to the plasma membrane, with a half-residence time of just two minutes⁷⁶. Its stability at the membrane is ensured by p120-catenin, which binds to the cadherin juxtamembrane domain, preventing endocytic adaptor proteins from binding. However, under low glucose conditions or oxidative stress, p120-catenin undergoes glutathionylation, dissociates from E-cadherin, and is degraded⁷⁷. Mechanical stress also increases E-cadherin turnover by releasing p120-catenin into the cytosol⁶⁰.

Members of the Rho family of proteins play a crucial role in controlling E-cadherin stability. Rac1 signaling mediates the dynamin-dependent endocytosis of E-cadherin into recycling endosomes^{60; 78}. Interferon regulatory factor 6, which regulates the expression of ArhGAP29 (a RhoA GTPase activating protein), controls the trafficking of E-cadherin back to the plasma membrane⁷⁹. Additionally, SGEF, a guanine nucleotide exchange factor for RhoG, forms a complex with Scribble and Dlg1 to stabilize the E-cadherin-catenin complex and prevent its internalization via endocytosis⁸⁰. Thus, the presence of E-cadherin on the cell surface is a highly regulated process, modulated by binding partners and small GTPases of the Rho family.

Conclusions

Over four decades of research have unveiled a vast array of cellular pathways activated by the engagement and force on E-cadherin. Significant progress has deepened our understanding of well-known signal mediators like catenins and the Rho family of GTPases. However, surprising connections have also emerged, linking the adhesive machinery to enzymes that regulate metabolism, transcription, viscosity, and E-cadherin trafficking. These new discoveries have prompted a re-examination of older studies, revealing an additional layer of complexity previously unknown.

More work is needed to identify signal transduction pathways that lie at the nexus of two pathways. Much research focuses on understanding how a single pathway produces an outcome. Emerging links between cell metabolism and mechanics, and their regulation of epithelial barrier function, have introduced a new level of complexity by integrating two previously separate signaling modules. Understanding how multiple signal transduction pathways connect to produce a biological outcome is crucial.

Additionally, most research to date has focused on the application of stimuli at a single time, but less is known about how different durations of signaling impact biological outcomes. There is evidence suggesting that different biological signaling and outcomes depend on the amplitude of force applied. For example, epithelial cells in the lung can experience vastly different amplitudes of force depending on the context, such as during sleep, vigorous exercise, or coughing. Muscle cell studies indicate that different amplitudes of force stimulate different AMPK isoforms^81–83. Thus, more work is needed to resolve these important unknowns, promising to shed light on these complex processes.