Introduction

Epidermal renewal is accompanied by continuous reorganization of the keratinocyte's adhesive interactions. This is particularly evident in the basal layer of the epidermis, where the keratinocytes proliferate and, after exiting the cell cycle and commitment to terminal differentiation, disassemble their cell–matrix attachments to move upwards and rebuild the overlying epidermis (for review see Blanpain and Fuchs, 2006; Watt et al., 2006). These dynamic adhesive interactions could endow epidermal keratinocytes with a unique and potent system to survey the changing environment and direct the cell fate when coupled to cell signaling. Hence, this article tackles the emerging knowledge on outside-in signals emanating from adhesive interactions in epidermal keratinocytes of the "transitional phase", that is from the time point the epidermal keratinocyte exits the stem cell compartment and becomes a transit amplifying cell until it is growth arrested and has committed to terminal differentiation (Figure 1).

Figure 1.

Diagram depicting the time frame in which the "transitional phase" takes place in cultured keratinocytes. Top panel: Left; human epidermis was stained with Ki67 and Hoechst 33258 to show one proliferating cell per approximately 10 basal keratinocytes; middle; cultured mouse keratinocytes at confluency were stained with Hoechst 33258 to show asynchronous division; about one cell in 10 is dividing, similar to what is observed in basal epidermal keratinocytes stained with Ki67. Right: dividing mouse keratinocytes of the transitional phase are loosely attached to the culture dish. Bottom panel: Model discussed in this review. After seeding, keratinocytes proliferate under influence of RTK (specifically discussed for EGF-R) and integrins. Three to four days post-seeding, the keratinocytes reach confluency. Integrin- and cadherin-mediated signals in cooperation then oppose each other to lead the keratinocyte into cell-cycle exit, growth arrest, and onset of terminal differentiation. Figure adapted from Kolly et al. (2005).

Full figure and legend (141K)Download Power Point slide (454 KB)

The integrins are the main family of adhesion molecules that connect keratinocytes to the underlying basement membrane. In human and mouse epidermis, the primary integrins expressed are 64 (adhering to laminin-332 (formerly laminin 5)), 21 (adhering to collagen I), 31 (adhering to laminin-332), and the less abundant v5 (adhering to vitronectin) (Watt, 2002; Wilhelmsen et al., 2006). Of these, 64 assembles with plectin, BP 230, and the collagen-type molecule BP180 to form hemidesmosomes that tether to keratin intermediate filaments, whereas the other -isomers assemble into focal adhesions that link to actin. Like their suprabasal neighbors, basal keratinocytes also adhere tightly to each other via anchoring junctions, that are adherens junctions and desmosomes (for review see Green and Gaudry, 2000; Vasioukhin and Fuchs, 2001). The transmembrane proteins mediating the intercellular adhesive interactions belong to the cadherin superfamily. In basal keratinocytes, E- and P-cadherin are the major adhesion-mediating molecules in adherens junctions. They assemble with plaque proteins p120^cnt, -catenin, plakoglobin (-catenin), and -catenin into functional junctions. Alternatively, the strongest intercellular adhesion between these cells is provided by desmosomal cadherins (for example, in basal keratinocytes desmoglein (Dsg) 3 and desmocollin 3 and low levels of Dsg2) forming desmosomes together with plaque proteins such as plakophilins and plakoglobin, whereas desmoplakin anchors desmosomes to the intermediate filament network.

The adhesive interactions of keratinocytes are of crucial importance for the mechanical integrity of the epidermis, as exemplified by bullous autoimmune diseases that disable adhesive functions (for review see Suter et al., 1998; Schmidt and Zillikens, 2000). However, evidence is emerging that the functional units of cell adhesion, termed here adhesion receptors, can in addition serve as transducers of outside-in signals to control proliferation and terminal differentiation in keratinocytes of the transitional phase (for example, integrins, Mainiero et al., 1997; Nikolopoulos et al., 2005; or cadherins, Pece and Gutkind, 2000; Calautti et al., 2005; Williamson et al., 2006; Perrais et al., 2007; Xie and Bikle, 2007).

To date, no enzymatic activity has been attributed to the cytoplasmic tails of adhesion receptors. Their signaling capability emanates from the recruitment of intracellular signaling components such as kinases and phosphatases that physically link to the adhesion receptors' cytoplasmic tail, to adaptor proteins, or plaque proteins (for example, Fuchs et al., 1996; Mainiero et al., 1996; Shaw et al., 1997; Calautti et al., 2005). Although outside-in signals transmitted by ligand-activated integrin receptors have been extensively described and are known to control cell migration, survival, and proliferation in many cell types (for review see Miranti and Brugge, 2002; Guo and Giancotti, 2004; Janes and Watt, 2006; Wilhelmsen et al., 2006), evidence for outside-in signals triggered by the engagement of cadherin receptors through transadhesion is only starting to emerge.

Recent studies in various epithelial and endothelial cell types, including keratinocytes, now underscore that cadherins can play a key role in controlling, for example, cell growth, differentiation, and survival (Takahashi and Suzuki, 1996; Carmeliet et al., 1999; Pece et al., 1999; Cabodi et al., 2000; Pece and Gutkind, 2000; Calautti et al., 2002, 2005; Laprise et al., 2002, 2004; Dejana, 2004; Qian et al., 2004; Pang et al., 2005; Liebner et al., 2006; Williamson et al., 2006; Hofmann et al., 2007; Kang et al., 2007; Perrais et al., 2007; Xie and Bikle, 2007). Moreover, there is increasing evidence that cadherins like integrins associate with and control the action of receptor tyrosine kinases (RTKs), in particular of the epidermal growth factor receptor (EGF-R) family (integrins, Miranti and Brugge, 2002; Guo and Giancotti, 2004; Janes and Watt, 2006; Wilhelmsen et al., 2006; cadherins, Hoschuetzky et al., 1994; Takahashi and Suzuki, 1996; Pece and Gutkind, 2000; Qian et al., 2004; Calautti et al., 2005; Kang et al., 2007; Perrais et al., 2007). While signals generated by ligand-activated integrins and RTK are in principle pro-proliferative, signals emanating from cadherins and RTKs can be pro-proliferative or antiproliferative depending on the cadherin and the associated RTK; for instance, N-cadherin was suggested to interact with fibroblast growth factor receptor to promote invasion and tumor-cell behavior, whereas E-cadherin has been described to associate/cooperate with EGF-R to modulate its activity in confluency-dependent events preceding growth arrest (Takahashi and Suzuki, 1996; Pece and Gutkind, 2000; Qian et al., 2004; Calautti et al., 2005; Perrais et al., 2007; Theisen et al., 2007).

Conventional submerged cultures of mouse, human, and canine keratinocytes share many features with epidermal keratinocytes in vivo, and thus were instrumental in gaining invaluable insight into the control of epidermal homeostatic processes (for example, Yuspa et al., 1989; Pillai et al., 1990; Poumay and Pittelkow, 1995; Kolly et al., 2005). In low-density cultures and in media containing EGF or related growth factors, keratinocytes proliferate and migrate, and are most similar to keratinocytes in re-epithelializing skin. At confluency, a continuous monolayer is formed and ample cell–cell contacts promote the stabilization of junctional components at the plasma membrane, similar to the situation in basal or transit-amplifying keratinocytes in vivo. At that stage the keratinocytes are said to become "contact inhibited", and like transit-amplifying cells in the epidermis, undergo two to four mitotic divisions while committing to terminal differentiation (Poumay and Pittelkow, 1995; Kolly et al., 2005; Watt et al., 2006). During their mitotic divisions, the proliferation rate of confluent keratinocyte cultures remains constant in spite of the steady increase of antiproliferative factors such as cyclin-dependent kinase inhibitors p21^WAF and p27^KIP (Missero et al., 1996; Kolly et al., 2005). This suggests that during the transitional phase pro-proliferative and antiproliferative events occur simultaneously within the same cell, but the antiproliferative effectors accumulate over time and only override pro-proliferative signals when they have reached a certain threshold level (Figure 1; Kolly et al., 2005).

The concept explored in this review is that integrins and cadherins sense the cells immediate environment and cooperate in directing the transitional phase of epidermal renewal by outside-in signaling. As suggested for other epithelial cell types, ligand-activated integrins and cadherins may thereby serve as signaling receptors, which exert their function before assembling into fully organized junctions firmly anchoring to the cytoskeletal network (for example, Mariotti et al., 2001; Kovacs et al., 2002; Laprise et al., 2002; Calautti et al., 2005). This conceivably facilitates lateral movement, supporting their association and cooperation with EGF-R family members in addition to their control.

To begin, this review focuses on the pathways through which integrins can stimulate proliferation, and how groundbreaking work on 4-integrin-mediated signaling using a novel approach (Nikolopoulos et al., 2005; Guo et al., 2006) integrates with the broader body of related literature. The review then continues with an analysis of the emerging knowledge on the involvement of cadherins in growth arrest and commitment to terminal differentiation, and how recent studies on E-cadherin-mediated signaling in keratinocytes (Calautti et al., 2005; Perrais et al., 2007; Xie and Bikle, 2007) combine with established results on E-cadherin and new data on Dsg3-mediated signaling (Williamson et al., 2006) to form a comprehensive picture, which now allows us to simultaneously discuss a possible cross talk between integrins and cadherins with EGF-R family members in epidermal keratinocytes of the transitional phase (Figure 1).

Integrin-mediated pro-proliferative signals

Keratinocytes in the basal proliferative layer of the epidermis can be distinguished from their suprabasal postmitotic neighbors by their cell–matrix attachment (Watt, 2002). If this connection is disrupted by placing human or mouse keratinocytes in suspension, they start to terminally differentiate (Green, 1977; Adams and Watt, 1989; Rodeck et al., 1997). Blocking phosphatidylinositol trisphosphate kinase (PI3K) or alternatively, EGF-R of suspension cells, induces apoptosis. Inversely, if integrins are cross-linked, proliferation is maintained (Adams and Watt, 1989; Watt et al., 1993; Rodeck et al., 1997; Nikolopoulos et al., 2005). This indicates that normal homeostatic keratinocytes require integrin-mediated cell-substrate adhesion to proliferate. Moreover, when cell–substrate adhesion is lost, differentiation-promoting and cell survival cascades are activated, which depend on PI3K and EGF-R. Collectively, these findings suggest that in epidermal keratinocytes, functional integrin receptors either support proliferation or alternatively suppress terminal differentiation or both.

Reports on the lack of a proliferative phenotype in 1-integrin-knockout keratinocytes in vitro and more recently in conditional Cre-mediated 1-integrin-knockout epidermis of adult mice appeared to exclude 1-integrin as a regulator of the proliferative process during epidermal turn-over (Raghavan et al., 2003; Lopez-Rovira et al., 2005). Moreover, deletion confined to small stretches (to prevent loss of cell–matrix adhesion) of 4-integrin, the second major -subunit in the epidermis, also failed to impede normal proliferation and/or terminal differentiation, raising further doubts about a pro-proliferative activity of integrins during epidermal renewal (Raymond et al., 2005). These findings were surprising in view of the many signaling events reported downstream of integrin receptors (for review see Miranti and Brugge, 2002; Guo and Giancotti, 2004; Janes and Watt, 2006; Wilhelmsen et al., 2006).

Studying the role of adhesion receptors is complicated by the fact that their genetic deletion not only affects signaling but also adhesion. In addition, deletion of adhesion components in skin can trigger secondary events such as inflammation, which induce hyperproliferation as a bystander effect. As an example, varying grades of inflammation were discussed as a reason for conflicting results obtained from different 1-integrin-knockout mice (Lopez-Rovira et al., 2005). Consequently, for many studies on adhesion receptors, it is difficult to discriminate between signals related to homeostatic processes, and those linked to lack of tissue integrity or to secondary events like inflammation. In the case of 4-integrin, this problem was recently elegantly circumvented by expressing a C-terminal deletion mutant (from aa 1,355 onwards) in mouse epidermis, which is not critical for adhesion but abrogates signaling (Nikolopoulos et al., 2005).

Transgenic mice (both newborn and adult) expressing the signaling-defective 4-integrin mutant showed a twofold decrease in the epidermal proliferative index, confirming that 4-integrin actually contributes to proliferation (Nikolopoulos et al., 2005). Furthermore, keratinocytes from 4-integrin mutant mice failed to proliferate and migrate in vitro in response to EGF, suggesting an essential cooperation between 4-integrin and an RTK in proliferative epidermal keratinocytes. The follow-up of this study with comparative expression of full-length and signaling-defective 4-integrin in a transgenic mouse mammary carcinoma model with constitutive ErbB2 activation (an EGF-R family member; Ishizawar and Parsons, 2004), confirmed the pro-proliferative activity of 4-integrin; that is, loss of 4-integrin signaling-suppressed tumor onset, and invasive growth. More importantly, this study demonstrated that 4-integrin has to physically associate with an EGF-R family member to amplify pro-proliferative signals, and provided novel insights into the mechanism of the 4-integrin-mediated cooperation (Guo et al., 2006). Specifically, after association, 4-integrin mediates the phosphorylation of ErbB2 on Tyr⁸⁷⁷ by a Src-family kinase. This phosphorylation site resides in the catalytic domain of ErbB2 and corresponds with Tyr⁸⁴⁵ in EGF-R, which is known to be essential for mitogenesis in terms of DNA synthesis (Ishizawar and Parsons, 2004).

The combination of the more detailed study in mammary epithelium with the findings in mouse epidermis using the unique approach of signaling-defective (but adhesion competent) 4-integrin now enables a clear definition of signals relevant for the 4-integrin-mediated pro-proliferative and pro-survival activity (Nikolopoulos et al., 2005; Guo et al., 2006). Like in mammary epithelium, 4-integrin was found to cooperate with EGF-R in epidermal keratinocytes to enhance proliferation in a Src- but not Ras-dependent manner. Signaling-wise this involves mainly the mitogen-activated protein kinase (MAPK) kinase module JNK/c-Jun and contributing activation of signal transducer and activator of transcription 3 (STAT3), whereas the PI3K/Akt kinase survival axis (protein kinase B) was suggested to be activated by 4-integrin and/or RTK alone (see Figure 2, figure legend for full names of effectors). The relevance of these signals for the processes governing epidermal proliferation and cell survival is, discussed in more detail in the following three sections.

Figure 2.

Opposing signaling events discussed in this review to emanate from 4-integrin and E-cadherin receptors within one keratinocyte. Depicted are effectors contributing to proliferation (4-integrin) and cell-cycle exit (E-cadherin), in addition to cell survival in keratinocytes. Note that the respective cascades were investigated independently of each other but under similar experimental settings. JNK/c-Jun was found to be crucial for proliferation. STAT3 was described to contribute to the disruption of tight junctions and adherens junctions in mammalian carcinoma cells (Guo et al., 2006). Akt generates survival signals, in addition to possibly contributing to proliferation and terminal differentiation depending on the receptor and Akt isoform. Yellow star on JNK1: 4-integrin/EGF-R-mediated nuclear targeting of phosphorylated JNK1; purple stars on EGF-R: Src kinase-mediated phosphorylation of EGF-R. Abbreviations: -cat, -catenin; EGF-R, epidermal growth factor receptor; ERK1/2, extracellular signal-related kinase; GSK3, glycogen synthase kinase; c-Jun, Janus kinase; JNK, c-Jun N-terminal protein kinase; PDK-1, phosphoinositide-dependent kinase-1; PG, plakoglobin; PI3K, phosphatidylinositol triphosphate kinase; Srcasm, Src-activating and signaling molecule; STAT; signal transducers and activator of transcription.

Full figure and legend (244K)Download Power Point slide (504 KB)

MAPK activation downstream of 4-integrin/EGF-R: pro-proliferative events

In epithelial cells, three MAPK modules have been identified, which lead up to extracellular signal-regulated kinase (ERK)1/2, c-Jun N-terminal kinase (JNK), or p38 activation, respectively (Cowan and Storey, 2003). Of these, JNK-activated c-Jun was found to centrally control the proliferative activity downstream of 4-integrin and EGF-R in mouse and human keratinocytes plated on laminin-332, as well as downstream of 4-integrin/ErbB2 in the mammary carcinoma model (Mainiero et al., 1997; Nikolopoulos et al., 2005; Guo et al., 2006; Figure 2). Importantly, both 4-integrin/EGF-R and 4-integrin/ErbB2 did not activate/phosphorylate JNK, but instead triggered the nuclear translocation of activated JNK. Although the mechanism for nuclear targeting is currently not known, it implies 4-integrin-independent JNK activation through EGF-R or another RTK (Zenz and Wagner, 2006), followed by 4-integrin/EGF-R-dependent nuclear targeting (Figure 2).

An important role of JNK in epidermal proliferation is compatible with the contribution of JNK1 (but not of JNK2; Sabapathy and Wagner, 2004; Weston et al., 2004) and c-Jun to normal epidermal proliferation as well as pathological conditions like psoriasis and tumor formation (Zenz and Wagner, 2006). Consistently, JNK drives proliferation in embryonic mouse epidermis with NF-B/RelA deficiency (Zhang et al., 2004), whereas experimentally blocked NF-B was found to act in a laminin-332 and 4-integrin-dependent manner to promote human epidermal tumorigenesis in cooperation with oncogenic Ras (Dajee et al., 2003). In contrast to the involvement of NF-B blockade in tumorigenesis, activation but not inhibition of NF-B/RelA was implicated in 4-integrin/EGF-R-dependent proliferation in addition to JNK/c-Jun in mouse keratinocyte cultures, raising the possibility that NF-B can function in a bidirectional manner in epidermal keratinocytes (Nikolopoulos et al., 2005).

In parallel to JNK, two other MAPKs, ERK1 and ERK2, were also found to be targeted to the nucleus in a 4-integrin-dependent manner in keratinocytes (Nikolopoulos et al., 2005). Nonetheless, proliferation was not dependent on ERK in either keratinocytes or the mammary carcinoma model downstream of 4-integrin (Nikolopoulos et al., 2005; Guo et al., 2006). This is in line with persistence of hyperproliferation in NF-B-blocked human keratinocytes treated with pharmaco-logical ERK inhibitors (Zhang et al., 2004), as well as lack of evidence for a contribution of ERK to mitogenesis downstream of Tyr⁸⁴⁵EGF-R (Ishizawar and Parsons, 2004).

Janus kinase–STAT activation downstream of 4-integrin/EGF-R: proliferation-dependent events

In confluent mammary carcinoma cells in vitro, 4-integrin/ErbB2 activation prevented contact-induced growth inhibition (Guo et al., 2006). Simultaneously, adherens junctions and tight junctions were disrupted in spite of the fact that expression levels of junctional proteins remained unaffected. Junctional disruption was forestalled by signaling-defective 4-integrin or by inhibitors to ErbB2, and in part by a dominant-negative STAT3 mutant. This indicates that in addition to supporting proliferation via JNK, 4-integrin/ErbB2 cross talks to intercellular junctions via STAT3 in the mammary carcinoma cells (Figure 2).

4-integrin STAT3-induced junctional disruption has as yet not been addressed in other cell types. It is however of particular interest as it differs from endocytosis and downregulation of junctional components known to accompany Src/PI3K-induced snail-related tumorigenic transformations (Thiery, 2002). It instead resembles normal mitosis-induced disruption and endocytosis of adherens junctions, which occurs without degradation of junctional proteins in dividing Madin–Darby canine kidney cells at confluency (Bauer et al., 1998). Consistent with the activation of 4-integrin signaling during mitotic processes (Nikolopoulos et al., 2005; Guo et al., 2006), cell–substrate adhesion was reduced in dividing Madin–Darby canine kidney cells. Reduced cell–substrate adhesion is a long known phenomenon of dividing cells and is also observed in confluent mouse keratinocyte cultures (Figure 1, top). In several epithelial cell types, including HaCaT keratinocytes, reduced cell-substrate adhesion was ascribed to Src-family kinase-induced disassembly of hemidesmosomes preceding 4-integrin signaling (Mariotti et al., 2001), a phenomenon that might also account for Madin–Darby canine kidney cells. Disassembly of hemidesmosomes was discussed to uncouple 64-integrin from the adhesion complex and from the underlying cytoskeletal, to favor its colocalization with EGF-R in signaling platforms known as lipid rafts (Mariotti et al., 2001; Gagnoux-Palacios et al., 2003).

In vivo, 4-integrin amplifies pro-proliferative signals in basal epidermal keratinocytes (Nikolopoulos et al., 2005), where STAT3 is expressed and can be activated by EGF/EGF-R (Nishio et al., 2001; Chan et al., 2004; Li et al., 2007). This raises the possibility of a 4-integrin/EGF-R to cadherin cross talk via STAT3 also in vivo. Although this needs to be confirmed, transient dissolution of intercellular junctions without degradation of junctional proteins would perfectly meet the requirements of dividing keratinocytes in epidermal tissue. These cells need to rapidly reorganize their cell–matrix and cell–cell attachment to preserve the integrity of the epidermal sheet. Such reorganizational processes would however only concern one dividing cell per epidermal proliferative unit (10–20 basal keratinocytes; Figure 1; Gambardella and Barrandon, 2003), and possibly only in selected areas of the plasma membrane. Consistently, inspection of basal keratinocytes by electron microscopy did not reveal apparent junctional disruption (Baker and Garrod, 1993). Altogether, this suggests that the 4-integrin/EGF-R/STAT3-induced junctional disruption without degradation could be a general mechanism, which reversibly disassembles intercellular junctions during mitosis (Figure 2). This would further permit 4-integrin to simultaneously control pro-proliferative signals and STAT3-related antiproliferative signals that emanate from transadhering E-cadherin (as will be discussed below).

PI3K/Akt activation downstream of 4-integrin/EGF-R: survival and potential pro-proliferative events

Akt is a serine/threonine kinase with a well-known role in cell survival (for review see Vivanco and Sawyers, 2002). Accordingly, Akt is activated and ensures cell survival in keratinocytes devoid of cell–matrix attachment (Frisch and Screaton, 2001). In addition to its pro-survival function, Akt has also been involved in cellular processes ranging from migration, proliferation, to terminal differentiation (for review see Vivanco and Sawyers, 2002; Toker and Yoeli-Lerner, 2006). These functions might depend on the Akt isoform and the activating receptor. Class IA PI3K, the main activators of Akt, can be recruited to receptors such as activated RTK (for review see Vivanco and Sawyers, 2002; Engelman et al., 2006), integrins (Mainiero et al., 1997; Shaw et al., 1997; Miranti and Brugge, 2002), but also transadhering cadherins (as further discussed below).

As outlined above, 4-integrin signaling involves reduced cell–substrate adhesion secondary to disruption of hemidesmosomes (Mariotti et al., 2001). Accordingly, Akt is activated/phosphorylated to ensure survival in mouse keratinocytes and mammary carcinoma cells with signaling active 4-integrin, and this activation in part depends on RTK (Nikolopoulos et al., 2005; Guo et al., 2006). In case of signaling-defective 4-integrin, cellular stress such as starvation or enhanced proliferation resulted in apoptosis in both cell types. This implies that under circumstances requiring high Akt activity, such as stress, the Akt pool becomes limiting in absence of 4-integrin signaling. Hence, this provides circumstantial evidence that 4-integrin itself can activate Akt in these experimental settings. A possibility, which so far lacks experimental proof, is that, in addition to supporting cell survival, 4-integrin-mediated Akt activation might be involved in nuclear targeting of JNK and thereby contribute to the proliferative process. This is compatible with the observation that Akt can amplify growth factor-induced signals (Toker and Yoeli-Lerner, 2006).

In normal homeostatic mouse epidermis, Akt activation is mainly confined to suprabasal cells, with the exception of scattered keratinocytes in the basal layer with a loosely attached appearance (Calautti et al., 2005). Overall, this is consistent with the fact that Akt is activated in keratinocytes without cell–substrate attachment to promote cell survival (Frisch and Screaton, 2001). Hence, Akt-positive cells in the basal layer can comprise keratinocytes committed to terminal differentiation that have disassembled their hemidesmosomes to start migrating upwards. Alternatively, as outlined above, Akt can be activated in actively dividing cells that have disassembled their hemidesmosomes in presence of 4-integrin signaling. In the basal layer, the Akt-positive cells could therefore also result from and indicate 4-integrin signaling (one out of 10–20 cells in the epidermis; Figure 1), a possibility, which warrants further investigations.

Conclusion on 64-integrin signaling in epidermal keratinocytes

In summary, in epidermal keratinocytes of the transitional phase (Figure 1), ligand-activated 4-integrin receptors cooperate with EGF-R to amplify pro-proliferative and survival signals. This occurs in an Src- but not Ras-dependent manner and mainly involves JNK and Akt (Figure 2). The cooperation between 4-integrin and EGF-R may be brought about by uncoupling 64-integrin from hemidesmosomes, which facilitates the association with EGF-R in lipid rafts (Mariotti et al., 2001). In addition to pro-proliferative signals, 4-integrin/EGF-R may control antiproliferative events emanating from transadhering E-cadherin receptors via STAT3 (Figure 2). The collective findings on 4-integrin signaling in keratinocytes are supported by complementary observations in a mammary carcinoma model, while the evoked signaling pathways align with current knowledge on pro-proliferative and survival signals driving epidermal renewal.

With respect to the mechanisms, a number of important questions remain to be answered, including how JNK is targeted to the nucleus, whether STAT3 is activated in a 4-integrin/EGF-R-dependant manner in the epidermis and whether it contributes to junctional disassembly, as well as to which extent Akt is involved in the pro-proliferative process.

Based on the studies using signaling-defective 4-integrin, the central query "does -4-integrin-mediated signaling amplify epidermal proliferation in vivo" seems to have been answered in the affirmative. Outstanding is whether and to which extent other -integrin subunits and eventually also 6-integrin (Owens et al., 2003) contribute to this event, and whether they act synergistically or redundantly with 4-integrin. Redundancy is actually suggested by the discrepant observation that the signaling-defective 4-integrin but not its gene deletion displays a proliferative phenotype in the epidermis (Nikolopoulos et al., 2005; Raymond et al., 2005). Signaling-defective 4-integrin might compete with wild-type integrins for the binding of essential proliferation-promoting factors, without being able to propagate their activity. This conceivably fails to happen in the knockout situation. Redundancy is also supported by the fact that the amplification of pro-proliferative signals in the manner described for 4-integrin equally applies for 1-integrin. 1-integrin phosphorylation by Src results in EGF-R activation (Miranti and Brugge, 2002), and 1-integrin expression is required for the initiation of mouse mammary tumors expressing the polyoma virus middle T antigen (White et al., 2004). Furthermore, both, 4- and 1-integrin amplify signals of other RTKs than of ErbB2 and EGF-R, such as RON, the hepatocyte growth factor receptor Met, and tumor necrosis factor receptor, and all have been involved in carcinogenesis also in the skin (for review see Miranti and Brugge, 2002; Guo and Giancotti, 2004; Janes and Watt, 2006). Collectively, these results strongly argue in favor of an obligatory role of the integrin family members in controlling and propagating pro-proliferative and survival signals in basal keratinocytes of the transitional phase.

E-cadherin-mediated antiproliferative events

In contrast to integrins, evidence for cadherin-induced outside-in signaling came into focus only slowly. Published over the last 10 years, a number of studies in both epithelial and endothelial cells now appear to agree that signaling cascades emanating from transadhering or ligated E-cadherin or vascular (V) E-cadherin play an important role in confluency-dependent growth arrest and differentiation (Takahashi and Suzuki, 1996; Carmeliet et al., 1999; Pece et al., 1999; Cabodi et al., 2000; Pece and Gutkind, 2000; Calautti et al., 2002, 2005; Laprise et al., 2002, 2004; Qian et al., 2004; Liebner et al., 2006; Hofmann et al., 2007; Kang et al., 2007; Perrais et al., 2007; Xie and Bikle, 2007). Proof of concept that E-cadherin actually acts as an adhesion receptor in keratinocytes as well as in other epithelial cells, was provided by homophilic ligation of E-cadherin with recombinant protein, which results in direct activation of signaling cascades and growth arrest (Kovacs et al., 2002; Pang et al., 2005; McLachlan et al., 2007; Perrais et al., 2007), and inversely, by function disrupting E-cadherin antibodies, which abrogate for instance E-cadherin signaling, growth arrest, or terminal differentiation (Takahashi and Suzuki, 1996; Pece et al., 1999; Noren et al., 2001; Laprise et al., 2002; Qian et al., 2004; Calautti et al., 2005; McLachlan et al., 2007).

As far as analyzed, the mode by which E-cadherin transmits growth-inhibiting outside-in signals seems to follow a strikingly similar scheme than that of integrins involved in proliferation. Discussed in more detail below, signaling is initiated through homophilic transadhesion between E-cadherin molecules at points of cell–cell contact (Figure 2). The E-cadherin complex then associates and cooperates with an EGF-R family member to activate the PI3K/Akt pathway and possibly MAPK in a Src-family kinase-dependent manner, as indicated in a number of complementary studies on keratinocytes and other cell types (Hoschuetzky et al., 1994; Takahashi and Suzuki, 1996; Pece and Gutkind, 2000; Qian et al., 2004; Calautti et al., 2005; Perrais et al., 2007). Ultimately, this results in the loss of the cells' responsiveness to EGF through inhibition of the EGF-R pro-mitotic activity. Recently reported in A431 cells, this can occur through E-cadherin-mediated inhibition of transphosphorylation on Tyr⁸⁴⁵EGF-R (Perrais et al., 2007), the site, which catalyzes EGF-induced mitogenesis (Ishizawar and Parsons, 2004).

As outlined above also, studies in human and mouse keratinocytes highlight the involvement of E-cadherin in outside-in signaling. In addition, they demonstrate cooperation between E-cadherin and EGF-R in activating PI3K/Akt and inducing growth arrest (Figure 2; Pece and Gutkind, 2000; Calautti et al., 2005; Perrais et al., 2007; Xie and Bikle, 2007). Although not yet confirmed more broadly, this perfectly fits the observations that inhibition of PI3K or EGF-R is sufficient to prevent terminal differentiation in keratinocytes in suspension (Rodeck et al., 1997; Nikolopoulos et al., 2005), and that function disrupting antibodies to E-cadherin abrogate Akt phosphorylation inducing growth arrest and terminal differentiation in conventional submerged mouse keratinocyte cultures at confluency (Calautti et al., 2005). The underlying mechanism of the apparent joint signaling through E-cadherin/EGF-R is currently not known. In mouse keratinocytes, it was however shown to result in Src kinase-mediated phosphorylation of EGF-R, presumably on Tyr⁹²⁰ (Calautti et al., 2005). This can generate a docking site for p85, a regulatory subunit of PI3K (Stover et al., 1995; Calautti et al., 2005), with potential activation of diverse signaling cascades (for review see Engelman et al., 2006), including the MAPK described in human HaCaT cells (Pece and Gutkind, 2000).

In support of a possible involvement of E-cadherin in growth control in keratinocytes in vivo, E-cadherin-knockout epidermis displays some hyperproliferation, which however appears to be limited due to compensatory upregulation of P-cadherin (Young et al., 2003; Tinkle et al., 2004; Tunggal et al., 2005) or (as will be further discussed below) compensatory desmosomal cadherins. Furthermore, compatible with a dual contribution of EGF-R, PI3K, and Akt to pro-proliferative and antiproliferative events in epidermal keratinocytes (Figure 2), EGF-R- and Akt1/2-knockout mice as well as mouse keratinocytes overexpressing PI3K class IA catalytic subunits display a broad range of epidermal phenotypes, which extend from impaired migration and proliferation to impaired terminal differentiation (Miettinen et al., 1995; Peng et al., 2003; Pankow et al., 2006).

PI3K/Akt activation downstream of cadherins/EGF-R: antiapoptotic and antiproliferative events

In keratinocytes, exogenous expression of membrane tethered, constitutively active Akt was found to be sufficient in driving cell-cycle exit and onset of terminal differentiation, underscoring the importance of Akt in this process (Calautti et al., 2005). Accordingly, Akt activation downstream of E-cadherin engagement has been analyzed in quite some detail in mouse and human keratinocytes (Figure 2). The sequence of events outlined in the following was mainly established in several follow-up studies in mouse keratinocytes by Calautti and co-workers, and was sequentially confirmed by other investigators for mouse or human keratinocytes.

What is known so far is that transadhering E-cadherins are stabilized through association with p120^cnt (for review see Anastasiadis, 2007), which results, in the context of confluent mouse keratinocytes, in recruitment and activation of the small GTPase RhoA (Calautti et al., 2002; Anastasiadis, 2007) necessary for the association of a Src-family tyrosine kinase with the nascent cadherin tails at the plasma membrane (Cabodi et al., 2000; Calautti et al., 2002). Src-family kinases then phosphorylate the plaque proteins -catenin, plakoglobin, p120^cnt, and -catenin, which is required for their association with the E-cadherin tail (Calautti et al., 1998, 2005). In mouse keratinocytes, Src kinase-mediated phosphorylation was found to generate a docking site for PI3K (p85) on plakoglobin in an EGF-R-dependent manner, to a minor extent on p120^cnt, but not on -catenin, which apparently lacks a PI3K (p85)-docking motive (Calautti et al., 2005). Consistently, in the former study the catalytic subunit of PI3K (p110) was mainly associated with plakoglobin upstream of PI3K/Akt activation. This contrasts with data from human keratinocytes in which -catenin and p120^cnt but not plakoglobin were found to be involved in PI3K recruitment to transadhering E-cadherin upstream of Akt (Xie and Bikle, 2007). On the other hand, PI3K recruitment was related to -catenin but not p120^cnt in SW480 human colon carcinoma cells (Perrais et al., 2007).

The discrepant results on the involvement of -catenin or plakoglobin in tethering PI3K to the E-cadherin tail in keratinocytes might be explained by a fundamental difference in the experimental set ups (Calautti et al., 2005; Xie and Bikle, 2007). In the context of human keratinocytes, PI3K association was investigated in the first minutes after E-cadherin engagement, whereas it was addressed 1 hour later in mouse keratinocytes (Calautti et al., 2005; Xie and Bikle, 2007). Hence, it is conceivable that the PI3K recruitment occurs in a biphasic time course, at first for E-cadherin-mediated signaling involving p120^cnt and -catenin or -catenin-dependent molecules, and then second for Dsg3-mediated signaling involving plakoglobin (SID Meeting abstract; Schulze et al., 2007), while potentially underlying the cross talk between E-cadherin and desmosomal cadherins that plakoglobin was described to mediate many years ago (Lewis et al., 1997). Although this hypothesis needs to be confirmed, it is noteworthy that quite a few studies on various mouse cell types have concluded on the association of PI3K with -catenin based on co-immunoprecipitation experiments using E-cadherin antibodies (for example, Espada et al., 1999; Woodfield et al., 2001). Given that plakoglobin and -catenin coexist in adherens junctions (Butz et al., 1992), they co-precipitate and hence cannot be distinguished (for example, Calautti et al., 2005). Although the docking partners for PI3K on E-cadherin in keratinocytes require further detailed investigations, the studies on human and mouse keratinocytes agree on Akt activation following PI3K recruitment to transadhering E-cadherin (Calautti et al., 2005; Xie and Bikle, 2007).

Two possible targets inhibited through Akt-induced phosphorylation are glycogen synthase kinase (GSK3) and (for review see Vivanco and Sawyers, 2002). This inhibition is known to promote, among many other activities, the stabilization and nuclear translocation of -catenin and plakoglobin in various cell types (Aberle et al., 1997; Kodama et al., 1999). Accordingly, confluent mouse and human keratinocytes subjected to Akt activation exhibit phosphorylated GSK3 and GSK3 (Calautti et al., 2005; Thrash et al., 2006), which might trigger the confluency-dependent translocation of -catenin and plakoglobin into the nucleus that was described in an unrelated study on mouse keratinocytes (Figure 2; Williamson et al., 2006). One of the reported consequences of this translocation is the plakoglobin-mediated suppression of the pro-proliferative proto-oncogene c-Myc, which is critical for growth inhibition in these cells (Arnold and Watt, 2001; Waikel et al., 2001; Kolly et al., 2005; Figure 2). Consistently, maximal accumulation of nuclear plakoglobin as well as nuclear -catenin chronologically correlates with reduced cyclin D1 levels and keratinocyte growth arrest (Williamson et al., 2006). It is noteworthy that nuclear accumulation of plakoglobin and/or -catenin has not yet been functionally linked to transadhering E-cadherin, plakoglobin was however related to Dsg3-induced signaling as will be further discussed below (Williamson et al., 2006).

Besides -catenin and plakoglobin, another potential effector activated downstream of E-cadherin/Akt and GSK is NF-B/RelA, which also can control cell-cycle exit in human and mouse keratinocytes (Dajee et al., 2003; Zhang et al., 2004). Upstream of Akt, E-cadherin/PI3K activates phospholipase-C-1, which is a crucial element in the mobilization of intracellular calcium and terminal differentiation (Xie and Bikle, 2007). The relative contribution of the Akt pathway and PI3K/phospholipase-C-1 to growth arrest versus terminal differentiation in keratinocytes downstream of transadhering E-cadherin, remains to be investigated.

MAPK activation downstream of cadherins/EGF-R: potential antiproliferative events

Only a small number of reports have investigated the activation of MAPK modules downstream of transadhering E-cadherin and cooperating EGF-R in keratinocytes and other epithelial cells. However, some evidence exists that ERK1/2 and p38 are transiently activated (Pece and Gutkind, 2000; Laprise et al., 2004). In human HaCaT keratinocytes for instance, ERK1/2 phosphorylation increases in dependence of EGF-R recruitment to transadhering E-cadherin, and E-cadherin-mediated activation of MAPK was also reported in intestinal epithelial cells (Pece and Gutkind, 2000; Laprise et al., 2004). Of particular interest is that JNK2 was found to be necessary for cell-cycle arrest in keratinocytes, as exemplified by hyperproliferation of JNK2-knockout epidermis (Sabapathy and Wagner, 2004; Weston et al., 2004). In light of the central pro-proliferative role attributed to JNK/c-Jun downstream of 4-integrin/EGF-R in keratinocytes (Nikolopoulos et al., 2005), it is noteworthy that JNK2 was suggested to destabilize JNK1-induced c-Jun (Figure 2). As E-cadherin ligation or transadhesion is sufficient for growth arrest and onset of terminal differentiation in cultured keratinocytes (Calautti et al., 2005; Perrais et al., 2007; Xie and Bikle, 2007), JNK2 might potentially be activated via Src kinase-mediated phosphorylation of EGF-R on Tyr⁹²⁰ (Calautti et al., 2005), to counteract the pro-proliferative c-Jun activity, a hypothesis requiring further detailed investigations.

Inhibition of ERK, Janus kinase/STAT3, and JNK1: antiproliferative events

E-cadherin signaling seems to be able to oppose keratinocyte proliferation by transient activation of the PI3K/Akt/GSK3 axis (Calautti et al., 2005; Figure 2). It can also contribute to destabilizing JNK1-induced c-Jun by JNK2 (Sabapathy and Wagner, 2004), and to silencing the mitotic activity of EGF-R by inhibiting Tyr⁸⁴⁵ phosphorylation (Perrais et al., 2007). Furthermore, in human keratinocytes, E-cadherin/PI3K also activates the terminal differentiation-inducing phospholipase-C-1 (Xie and Bikle, 2007), whereas the Src tyrosine kinase Fyn can activate protein kinase-C, which was suggested to be the major protein kinase-C isoform involved in terminal differentiation downstream of EGF-R (Denning et al., 1995). The latter downregulates, for example, 64-integrin expression, an event necessary to support the upward migration of keratinocytes (Alt et al., 2004). Furthermore, the increasing activity of the Src-family kinase Fyn measured during the transitional phase (Calautti et al., 1998, 2002) was found to exert a negative feedback on EGF-R-triggered pro-proliferative effectors via the Src-activating and signaling molecule (Srcasm) in mouse keratinocytes (Li et al., 2007). Downstream of EGF-R activation, Srcasm was shown to block ERK1/2 and STAT3 in addition to PDK-1, the latter being a crucial activator of the PI3K/Akt axis (Vivanco and Sawyers, 2002). In other terms, Srcasm appears to be able to inhibit all pro-proliferative signals described downstream of integrin/EGF-R, except c-Jun, which was not addressed and might be destabilized by JNK2 (Figure 2). Finally, Srcasm itself exerts a negative feedback, as it inhibits its activator, the Src kinase Fyn (Li et al., 2007). Although not investigated so far, this could potentially terminate both 4-integrin and E-cadherin-mediated signaling, which in each case is transient and depends on Src-family kinases (Figure 2). Following inhibition of Src-family kinases, phosphatases likely dephosphorylate adherens junction components to allow their assembly into stable junctions at the plasma membrane (Gumbiner, 2000), similarly as was suggested for the termination of 4-integrin-mediated signaling in HaCaT keratinocytes (Mariotti et al., 2001).

It has long been known that Src-mediated tyrosine phosphorylation shifts cadherin-based cell adhesion from a strong to a weak adhesive state (Takeda et al., 1995; for review see Gumbiner, 2000). As Src activation is a prerequisite for E-cadherin-induced signaling (Calautti et al., 2005), this would imply that signaling active E-cadherin receptors are of an intermediate state between no and strong intercellular adhesion. The transient nature of signaling cascades triggered by E-cadherin actually entails the formation of an intermediate state and would best explain the bidirectional activity of Src-family kinases that are involved in junction assembly as well as disassembly (Takeda et al., 1995; Calautti et al., 1998; McLachlan et al., 2007).

Indeed, based on observations in Chinese hamster ovary and intestinal cells by immunofluorescence analyses, signaling activity was assigned to E-cadherin molecules without cytoskeletal anchorage because E-cadherin-mediated signaling preceded the assembly of adherens junctions (Kovacs et al., 2002; Laprise et al., 2002). A similar possibility was discussed for mouse keratinocytes (Calautti et al., 2005) and would correspond with the properties evoked for the signaling-active 4-integrin pool (Mariotti et al., 2001). Although this topic needs to be addressed in detail, the possibility that the signaling pool of E-cadherin resides outside of junctions and has no or weak cytoskeletal connection is compatible with the notion that transadhesion between E-cadherins (and desmosomal cadherins) does not necessarily involve cytoskeletal anchorage (Roh and Stanley, 1995; Marcozzi et al., 1998). In addition, the capacity of sensing the environment by "ultrafast probing" was attributed to C-cadherin, which has no cytoskeletal interaction site (Pierres et al., 2007), while the lateral movement thought to facilitate the association of E-cadherin with EGF-R in lipid rafts (Seveau et al., 2004), was ascribed to E-cadherin receptors without cytoskeletal anchorage, which reside outside of cell–cell contacts in a mouse keratinocyte cell line (Kusumi et al., 1993). Support for the possibility that non-junctional E-cadherin without strong cytoskeletal anchorage exists in vivo was provided by the observation that E-cadherin antibodies bind to the plasma membrane between adherens junctions in mouse epidermis (Boller et al., 1985).

Conclusion on E-cadherin/EGF-R signaling in epidermal keratinocytes

Compelling but so far limited studies implicate transadhering E-cadherin receptors and cooperating EGF-R in outside-in signaling during growth arrest and terminal differentiation in keratinocytes, principally via PI3K/Akt and potentially MAPK (Figures 1 and 2). More ample results from other epithelial cell types and also from V(E)-cadherin in endothelial cells corroborate these findings, suggesting a potent cadherin-mediated mechanism to sense cell–cell contact and regulate cell fate. The capability to sense the environment might be an attribute of non-junctional E-cadherin, which in the absence of cytoskeletal anchorage has some leeway that allows the association with EGF-R in signaling platforms known as lipid rafts.

As all investigations to date were conducted with cultured cells, it cannot currently be judged whether E-cadherin signaling is of relevance for growth arrest in vivo. Indirect support is provided, as already mentioned, by E-cadherin gene deletion, which results in some hyperproliferation in the epidermis (Young et al., 2003; Tinkle et al., 2004; Tunggal et al., 2005). Furthermore, gene ablation of -catenin in the epidermis resulted in a severe phenotype resembling pre-neoplastic transformations (Vasioukhin et al., 2001), whereas, unexpectedly, p120^cnt knockout did not impact tissue homeostasis if inflammation was prevented (Perez-Moreno et al., 2006). The -catenin gene deletion correlated with ultrastructurally disturbed adherens junctions as well as reduced desmosomes and tight junctions. In contrast, only adherens junctions were affected in the case of the p120^cnt knockout. Collectively, these findings may suggest that if E-cadherin-mediated growth inhibition and onset of terminal differentiation cannot occur in the epidermis in vivo, it can be compensated by other cadherins, including desmosomal cadherins and, as discussed below, possibly by Dsg3.

Desmosomal cadherin-mediated antiproliferative events

Dsg3-mediated antiproliferative events

Anti-Dsg3 adhesion-disrupting antibodies are produced in the majority of Pemphigus vulgaris patients (Amagai et al., 1991; Payne et al., 2004) or experimentally in the mouse (monoclonal anti-Dsg3 (AK23)) (Tsunoda et al., 2003). When added to cultured mouse keratinocytes at the beginning of the transitional phase, they induce, besides other events (for review see Amagai et al., 2006), sustained proliferation, in correlation with a dramatic upregulation of c-Myc (Williamson et al., 2006). Moreover, these antibodies disrupt the desmosomal organization, but initially do not disrupt the oranization of adherens junctions (de Bruin et al., 2007).

In agreement with the in vitro data, high c-Myc expression, sustained proliferation (in terms of increased Ki67-positive cells), and consistently, transepidermal keratin-14 expression was observed as a general feature of epidermis and oral mucosa in all human and canine Pemphigus vulgaris patients analyzed so far ((Williamson et al., 2006, 2007a, 2007b; de Bruin et al., 2007; Muller et al., 2007) and Lietti et al., unpublished observations). Discovered only recently, the fact that sustained proliferation does not result in hyperplasia in the patient's tissue might explain why this phenomenon has not been recognized previously. Hyperproliferation without hyperplasia can for instance result from more rapid outward migration of basal epidermal keratinocytes, as was observed in keratin-10-knockout mice in presence of increased c-Myc expression (Reichelt and Magin, 2002).

In analogy to results on E-cadherin (for example, Calautti et al., 2005; Perrais et al., 2007), the phenotype of sustained proliferation induced by anti-Dsg3 antibodies suggested a link between transadhering Dsg3 and cell-cycle control. Recent data on Dsg3-mediated signaling in keratinocytes are in agreement with such a possibility (Figure 3, left panel).

Figure 3.

Dsg3-mediated signaling pathways discussed to be activated during the transitional phase, under control conditions or in keratinocytes exposed to Pemphigus vulgaris antibodies. Note that increased proliferation and c-Myc accumulation is observed in epidermis and oral mucosa of all human and canine patients analyzed so far (Williamson et al., 2006, 2007a, 2007b). Yellow stars, constitutively activated proteins. Note that the association of Dsg3/EGF-R in control cells as well as the disrupted association between PG and DP of fully assembled desmosomes is based on circumstantial evidence, as is constitutive activation of GSK3 in Pemphigus vulgaris-exposed keratinocytes. Results on Dsg3-mediated activation of the PI3K/Akt axis upstream of PG are unpublished (Lietti et al., Schulze et al., unpublished observations). For abbreviations see Figure 2. DP, desmoplakin.

Full figure and legend (169K)Download Power Point slide (683 KB)

At the beginning of the transitional phase, desmocollin 3 and Dsg3 are the main desmosomal cadherins displayed at the cell surface of keratinocytes (Green and Gaudry, 2000; Dusek et al., 2007). Similar to E-cadherin, they are stabilized following RhoA activation (Waschke et al., 2006), Dsg3-associated plakoglobin becomes tyrosine phosphorylated in a Src-family kinase-dependent manner (Calautti et al., 1998) and PI3K (p85) associates with the Dsg3/plakoglobin complex (SID Meeting abstract; Schulze et al., 2007). The formation of a Dsg3/PI3K complex is followed by an increase in Akt activity and growth arrest (Lietti et al., unpublished observation) (Figure 3, left panel). If cultured keratinocytes are challenged at that stage with antibodies from Pemphigus vulgaris patients, they in contrast exhibit (with regards to Dsg3-mediated signaling) (a) reduced RhoA activity (demonstrated in Pemphigus vulgaris antibody-treated organotypic cultures; Waschke et al., 2006), (b) enhanced turnover of membrane-soluble, non-junctional Dsg3 (Calkins et al., 2006; Williamson et al., 2006), and plasma membrane-anchored non-junctional Dsg3-associated plakoglobin (Williamson et al., 2006), (c) decreased non-junctional Dsg3 protein levels (Aoyama and Kitajima, 1999; Calkins et al., 2006; Williamson et al., 2006; Yamamoto et al., 2007), (d) reduced nuclear import of plakoglobin (Williamson et al., 2006), (e) abrogation of plakoglobin-mediated transcriptional repression of c-Myc, and (e) sustained proliferation concurrent with reduced expression of differentiation markers (Figure 3, right panel) (Williamson et al., 2006, 2007a, 2007b). Furthermore, inhibition of both GSK3 and c-Myc prevents loss of intercellular adhesion in neonatal mice injected with Pemphigus vulgaris antibodies or anti-Dsg3 mouse monoclonal antibodies AK23. This firstly provides circumstantial evidence that GSK3 remains activated in presence of pathogenic anti-Dsg3 antibodies and secondly underscores the pathogenic nature of this event (Williamson et al., 2006; Figure 3).

Taken collectively, the signaling pathways downstream of transadhering Dsg3 and changes upon binding of Pemphigus vulgaris antibody or AK23 are reminiscent of events triggered by transadhering E-cadherins. They suggest Dsg3-mediated activation of the PI3K/Akt pathway and growth arrest contrasting with disruption of the PI3K/Akt pathway and continuing proliferation at the expense of terminal differentiation in presence of anti-Dsg3 antibodies (compare Figures 2 and 3). At that stage, Pemphigus vulgaris antibodies neither visibly affect the localization or turn-over in individual components of adherens junctions, nor do they prevent the nuclear import of -catenin (Calkins et al., 2006; Williamson et al., 2006; de Bruin et al., 2007; Muller et al., 2007). This suggests that Dsg3 engagement is synergistic with and possibly amplifies E-cadherin-mediated signaling to induce growth arrest via PI3K/Akt/GSK3 and potentially EGF-R.

In support of a role of Dsg3 in growth control in vivo, a twofold increase in the proliferative index was noted in terms of Ki67 staining in the basal layer of Dsg3-knockout mice (our unpublished observation, Koch et al., 1997). Consistently, expression of a truncated Dsg3 transgene in basal epidermal keratinocytes, which lacks the transadhesion domain, induced a hyperproliferative phenotype (Allen et al., 1996). Moreover, suprabasal mis-expression of full-length mouse Dsg3 under the control of an involucrin promoter disrupted the barrier function but failed to induce hyperproliferation, which is expected if Dsg3 has antiproliferative activity (Elias et al., 2001). In contrast, human Dsg3 transgene expression driven by the keratin-1 promoter displayed a hyperplastic phenotype (Merritt et al., 2002). The exact reason for this discrepancy is currently not known.

Does Dsg3 control EGF-R signaling?

Indirect evidence for a control of Dsg3 over EGF-R has been provided by activation of EGF-R sustained in Pemphigus vulgaris antibody-treated keratinocytes and A431 epithelial cells. Upon exposure to Pemphigus vulgaris antibodies, EGF-R becomes rapidly hyperphosphorylated, and in addition to Src, all three MAPK modules, ERK1/2 and c-Jun, followed by p38, are activated, (Sanchez-Carpintero et al., 2004; Berkowitz et al., 2005; Frusic-Zlotkin et al., 2006; Chernyavsky et al., 2007; Figure 3). Simultaneously, EGF-R was found to be internalized (Frusic-Zlotkin et al., 2006), reminiscent of vascular endothelial growth factor receptor 2-mediated signaling from intracellular stores (Lampugnani et al., 2006). These collective signaling events involving EGF-R are characteristic of proliferating cells, which correlates with increased proliferation observed in Pemphigus vulgaris antibody-challenged mouse keratinocytes as well as in the patient (Williamson et al., 2006, 2007a, 2007b). Alternatively, Pemphigus vulgaris antibodies can also activate apoptotic pathways in cultured keratinocytes, as is supported by the presence of apoptotic keratinocytes in the blister roof of Pemphigus vulgaris patients (Frusic-Zlotkin et al., 2005, 2006; Chernyavsky et al., 2007). Induction of apoptotic effectors subsequent to inhibition of PI3K/Akt is expected, but only if EGF-R is blocked, that is in growth-arrested keratinocytes (Adams and Watt, 1989; Watt et al., 1993; Rodeck et al., 1997; Calautti et al., 2005; Nikolopoulos et al., 2005). Such a correlation can for instance be observed in cultured human keratinocytes challenged with Pemphigus vulgaris antibodies (Chernyavsky et al., 2007).

In agreement with sustained EGF-R activation, basal keratinocytes in the Pemphigus vulgaris patient's epidermis as well as in culture ultimately show acantholysis or loss of intercellular adhesion (Payne et al., 2004). This is a known phenomenon in epithelial cells treated with EGF and was termed "cell rounding" (Peppelenbosch et al., 1993). Mechanistically, EGF-induced modulation of intercellular adhesion was suggested to rely on EGF-dependent tyrosine phosphorylation of plakoglobin that compromises the link between desmosomal cadherins and the intermediate filament cytoskeleton (Gaudry et al., 2001). This appears to correspond with the "keratin retraction" described in human Pemphigus vulgaris antibody-treated mouse keratinocytes (Jones et al., 1984; Caldelari et al., 2001), as well as with the plakoglobin-mediated disruption of the desmosomal plaque (Caldelari et al., 2001; de Bruin et al., 2007) (Figure 3; right panel).

Cell rounding in EGF-treated HaCaT keratinocytes can be prevented simply by application of GSK3 inhibitors (Koivisto et al., 2006), exactly as was shown in neonatal mice exposed to Pemphigus vulgaris antibodies or AK23 (Williamson et al., 2006). This underscores that cell rounding relies on GSK3 activation related to EGF-R signaling. Hence, these events are consecutive and interdependent, with GSK3 being upstream of EGF-R (Figure 3). This would agree with the finding that c-Myc gene transcription (downstream of GSK3) is required to propagate Ras/MAPK induced uncontrolled proliferation (Oskarsson et al., 2006), whereby Ras activates the MAPK pathways downstream of EGF-R. Further supporting the GSK3–EGF-R hierarchy, plakoglobin-knockout keratinocytes do not undergo keratin retraction in response to Pemphigus vulgaris antibodies, despite elevated c-Myc levels (Caldelari et al., 2001; Williamson et al., 2006), because plakoglobin was found to integrate EGF-R-mediated signaling in mouse keratinocytes in terms of loss of keratin anchorage (Gaudry et al., 2001). Furthermore, c-Myc inhibition alone is sufficient to prevent acantholysis (without inhibiting EGF-R) (Williamson et al., 2006). Finally, the fact that only basal keratinocytes are responsive to Pemphigus vulgaris antibodies in terms of "cell rounding" could relate to the finding that Ras activation downstream of EGF-R is dependent on cell–substrate adhesion (Mainiero et al., 1997) and hence is switched off in suprabasal keratinocytes (Dajee et al., 2002).

Is Dsg3 signaling triggered by non-junctional Dsg3?

Pathogenic Dsg3 antibodies target the N-terminal domain (patients or mouse monoclonal (AK23); Tsunoda et al., 2003) and interfere with growth arrest. E-cadherin antibodies that bind to the N-terminal domain, disrupt E-cadherin engagement (Hoschuetzky et al., 1994) and prevent PI3K/Akt activation (Pece et al., 1999; Calautti et al., 2005). Due to the similarity of the evoked signaling pathways, pathogenic Dsg3 antibodies may also exert their activity by disrupting transadhesion between Dsg3 molecules. This mechanism was addressed as "steric hindrance" (Futei et al., 2000; Tsunoda et al., 2003). Several lines of evidence suggest that Dsg3 antibodies further activate cell signaling by binding to transadhering non-junctional Dsg3, like discussed here for integrin and E-cadherin receptors; (a) non-keratin-anchored Dsg3 is the primary target of Pemphigus vulgaris antibodies observed in time-lapse studies in vitro (Sato et al., 2000), (b) Pemphigus vulgaris antibodies bind to the interdesmosomal space as observed in Pemphigus vulgaris patients biopsies (Suter et al., 1990; Bedane et al., 1996), (c) split formation between epidermal keratinocytes in mice injected with Pemphigus vulgaris antibodies first occurs between desmosomes (Takahashi et al., 1985), and (d) in contrast to non-junctional Dsg3, the desmosomal fraction is not affected at the time changes in diverse effectors including nuclear targeting of plakoglobin are measured (Williamson et al., 2006). Dsg3 in the desmosomal fraction is reduced at later time points, which likely represents an important event preceding acantholysis (Koch et al., 1997; Aoyama and Kitajima, 1999; Williamson et al., 2006; Yamamoto et al., 2007).

Conclusion on Dsg3/EGF-R signaling in epidermal keratinocytes

Reminiscent of studies on E-cadherin, investigations on Dsg3 highlight its capability to activate signaling cascades involving PI3K/Akt and potentially EGF-R, to exert growth control in vitro and in vivo. Although still limited, the current view on Dsg3-mediated signaling is supported by the consequences of anti-Dsg3 antibodies on cell growth seen in Pemphigus vulgaris patients.

The emerging concept is that anti-Dsg3 antibodies oppose Dsg3-mediated signaling by disrupting transadhesion between non-junctional Dsg3 through steric hindrance. This provides a signal of "no cell" contact, GSK3 and EGF-R fail to be inhibited, and the keratinocytes in the basal layer of the epidermis continue to proliferate. As a consequence, the desmosomal anchorage is reduced through a plakoglobin/EGF-R-mediated process (Figure 3), Dsg3 is depleted from desmosomes, and adhesiveness is weakened resulting in the formation of blisters. Noticeable, hypotheses other than the one discussed here have been raised for Pemphigus vulgaris pathogenesis. They do however not address Dsg3-mediated growth control, and have been discussed in detail elsewhere (for review see Amagai et al., 2006).

With respect to the detailed mechanism of Dsg3-mediated signaling, some aspects described here are confirmed, some are based on circumstantial evidence, while many others such as Akt inhibition in Pemphigus vulgaris challenged keratinocytes, disruption by steric hindrance, and the relationship GSK3–EGF-R need to be filled in. Nonetheless, these collective results assembled with respect to Dsg3 raise the possibility that non-junctional Dsg3 represents an extraordinary sensory system of basal keratinocytes in the way suggested for E-cadherin.

The relative importance of E-cadherin, Dsg3, and other desmosomal cadherins in mediating antiproliferative signals in basal epidermal keratinocytes requires further detailed studies. Importantly, however, E-cadherin distribution and -catenin nuclear translocation are not initially affected by Pemphigus vulgaris antibodies (Williamson et al., 2006; de Bruin et al., 2007; Muller et al., 2007), suggesting that the pathways downstream of E-cadherin and Dsg3 are separable, as could also be suggested by the results on PI3K recruitment to the E-cadherin tail (Calautti et al., 2005; Xie and Bikle, 2007).

Conclusion and Outlook

Epidermal homeostasis requires a balance of pro- and antiproliferative signals in addition to survival signals (Figures 2 and 3). The current knowledge on adhesion receptor signaling summarized herein highlights the critical control the integrin and cadherin receptor network in cooperation with EGF-R (or other RTK) can impose on most of these crucial signaling events. It is further proposed that the adhesion receptors exert their control by communicating with the cell's immediate environment in absence of their cytoskeletal anchorage. This would facilitate the association with EGF-R in lipid rafts. Although we are far from understanding the complex interplay between the entire adhesive network, the fact that experimental disruption of adhesion or mimicry of adhesion is sufficient to change cell fate, strongly suggests that adhesion receptors have a fundamental role in regulating important homeostatic processes also in vivo. Consistent with such a scenario, constitutive activation or disruption of these receptor functions in vivo, including disturbed expression of adhesion components, leads to uncontrolled proliferation, resulting in severe pathologies such as cancer, psoriasis, and as suggested more recently, also Pemphigus vulgaris. Further investigations of the interdependent signaling pathways between integrins, cadherins, and RTK during epidermal renewal will contribute to a better understanding of their complex interplay and provide novel insights into the mechanisms underlying these severe pathologies.

Conflict of Interest

The authors state no conflict of interest.

Notes

Editor's Note

In this issue of the JID, the Perspectives Series on keratinocyte adhesion and cell junctions concludes with a thought-provoking review of the potential role of integrins and cadherins as molecules that not only hold the keratinocytes together, but also serve as signaling proteins for the epidermis. In this review, Eliane Müller and colleagues discuss the hypothesis that adhesion proteins may function as signaling molecules that help communicate important extracellular signals to the keratinocytes, and by so doing play an important role in keratinocyte growth and differentiation. This article will provide readers with a different perspective on the function of keratinocyte cell-surface adhesion proteins and stimulate new thought and investigation into how these proteins may affect epithelial biology.

"To raise new questions, new possibilities, to regard old problems from a new angle, requires creative imagination and marks real advance in science", Albert Einstein.

Russell P. Hall, III
Deputy Editor

References

1. Aberle H, Bauer A, Stappert J, Kispert A, Kemler R (1997) Beta-catenin is a target for the ubiquitin–proteasome pathway. EMBO J 16:3797–3804 | Article | PubMed | ISI | ChemPort |
2. Adams JC, Watt FM (1989) Fibronectin inhibits the terminal differentiation of human keratinocytes. Nature 340:307–309 | Article | PubMed | ISI | ChemPort |
3. Allen E, Yu QC, Fuchs E (1996) Mice expressing a mutant desmosomal cadherin exhibit abnormalities in desmosomes, proliferation, and epidermal differentiation. J Cell Biol 133:1367–1382 | Article | PubMed | ISI | ChemPort |
4. Alt A, Gartsbein M, Ohba M, Kuroki T, Tennenbaum T (2004) Differential regulation of alpha6beta4 integrin by PKC isoforms in murine skin keratinocytes. Biochem Biophys Res Commun 314:17–23 | Article | PubMed | ChemPort |
5. Amagai M, Ahmed AR, Kitajima Y, Bystryn JC, Milner Y, Gniadecki R et al. (2006) Are desmoglein autoantibodies essential for the immunopathogenesis of pemphigus vulgaris, or just "witnesses of disease"? Exp Dermatol 15:815–831 | Article | PubMed | ISI | ChemPort |
6. Amagai M, Klaus-Kovtun V, Stanley JR (1991) Autoantibodies against a novel epithelial cadherin in pemphigus vulgaris, a disease of cell adhesion. Cell 67:869–877 | Article | PubMed | ISI | ChemPort |
7. Anastasiadis PZ (2007) p120-ctn: a nexus for contextual signaling via Rho GTPases. Biochim Biophys Acta 1773:34–46 | Article | PubMed | ISI | ChemPort |
8. Aoyama Y, Kitajima Y (1999) Pemphigus vulgaris-IgG causes a rapid depletion of desmoglein 3 (Dsg3) from the Triton X-100 soluble pools, leading to the formation of Dsg3-depleted desmosomes in a human squamous carcinoma cell line, DJM-1 cells. J Invest Dermatol 112:67–71 | Article | PubMed | ISI | ChemPort |
9. Arnold I, Watt FM (2001) c-Myc activation in transgenic mouse epidermis results in mobilization of stem cells and differentiation of their progeny. Curr Biol 11:558–568 | Article | PubMed | ISI | ChemPort |
10. Baker J, Garrod D (1993) Epithelial cells retain junctions during mitosis. J Cell Sci 104(Part 2):415–425 | PubMed |
11. Bauer A, Lickert H, Kemler R, Stappert J (1998) Modification of the E-cadherin–catenin complex in mitotic Madin–Darby canine kidney epithelial cells. J Biol Chem 273:28314–28321 | Article | PubMed | ISI | ChemPort |
12. Bedane C, Prost C, Thomine E, Intrator L, Joly P, Caux F et al. (1996) Binding of autoantibodies is not restricted to desmosomes in pemphigus vulgaris: comparison of 14 cases of pemphigus vulgaris and 10 cases of pemphigus foliaceus studied by western immunoblot and immunoelectron microscopy. Arch Dermatol Res 288:343–352 | Article | PubMed | ISI | ChemPort |
13. Berkowitz P, Hu P, Liu Z, Diaz LA, Enghild JJ, Chua MP et al. (2005) Desmosome signaling. Inhibition of p38MAPK prevents pemphigus vulgaris IgG-induced cytoskeleton reorganization. J Biol Chem 280:23778–23784 | Article | PubMed | ISI | ChemPort |
14. Blanpain C, Fuchs E (2006) Epidermal stem cells of the skin. Annu Rev Cell Dev Biol 22:339–373 | Article | PubMed | ISI | ChemPort |
15. Boller K, Vestweber D, Kemler R (1985) Cell-adhesion molecule uvomorulin is localized in the intermediate junctions of adult intestinal epithelial cells. J Cell Biol 100:327–332 | Article | PubMed | ISI | ChemPort |
16. Butz S, Stappert J, Weissig H, Kemler R (1992) Plakoglobin and beta-catenin: distinct but closely related. Science 257:1142–1144 | Article | PubMed | ISI | ChemPort |
17. Cabodi S, Calautti E, Talora C, Kuroki T, Stein PL, Dotto GP (2000) A PKC-eta/Fyn-dependent pathway leading to keratinocyte growth arrest and differentiation. Mol Cell 6:1121–1129 | Article | PubMed | ISI | ChemPort |
18. Calautti E, Cabodi S, Stein PL, Hatzfeld M, Kedersha N, Paolo Dotto G (1998) Tyrosine phosphorylation and src family kinases control keratinocyte cell-cell adhesion. J Cell Biol 141:1449–1465 | Article | PubMed | ISI | ChemPort |
19. Calautti E, Grossi M, Mammucari C, Aoyama Y, Pirro M, Ono Y et al. (2002) Fyn tyrosine kinase is a downstream mediator of Rho/PRK2 function in keratinocyte cell-cell adhesion. J Cell Biol 156:137–148 | Article | PubMed | ISI | ChemPort |
20. Calautti E, Li J, Saoncella S, Brissette JL, Goetinck PF (2005) Phosphoinositide 3-kinase signaling to Akt promotes keratinocyte differentiation versus death. J Biol Chem 280:32856–32865 | Article | PubMed | ISI | ChemPort |
21. Caldelari R, de Bruin A, Baumann D, Suter MM, Bierkamp C, Balmer V et al. (2001) A central role for the armadillo protein plakoglobin in the autoimmune disease pemphigus vulgaris. J Cell Biol 153:823–834 | Article | PubMed | ISI | ChemPort |
22. Calkins CC, Setzer SV, Jennings JM, Summers S, Tsunoda K, Amagai M et al. (2006) Desmoglein endocytosis and desmosome disassembly are coordinated responses to pemphigus autoantibodies. J Biol Chem 281:7623–7634 | Article | PubMed | ISI | ChemPort |
23. Carmeliet P, Lampugnani MG, Moons L, Breviario F, Compernolle V, Bono F et al. (1999) Targeted deficiency or cytosolic truncation of the VE-cadherin gene in mice impairs VEGF-mediated endothelial survival and angiogenesis. Cell 98:147–157 | Article | PubMed | ISI | ChemPort |
24. Chan KS, Carbajal S, Kiguchi K, Clifford J, Sano S, DiGiovanni J (2004) Epidermal growth factor receptor-mediated activation of Stat3 during multistage skin carcinogenesis. Cancer Res 64:2382–2389 | Article | PubMed | ISI | ChemPort |
25. Chernyavsky AI, Arredondo J, Kitajima Y, Sato-Nagai M, Grando SA (2007) Desmoglein vs non-desmoglein signaling in pemphigus acantholysis: characterization of novel signaling pathways downstream of pemphigus vulgaris antigens. J Biol Chem 282:13804–13812 | Article | PubMed | ISI | ChemPort |
26. Cowan KJ, Storey KB (2003) Mitogen-activated protein kinases: new signaling pathways functioning in cellular responses to environmental stress. J Exp Biol 206:1107–1115 | Article | PubMed | ChemPort |
27. Dajee M, Lazarov M, Zhang JY, Cai T, Green CL, Russell AJ et al. (2003) NF-kappaB blockade and oncogenic Ras trigger invasive human epidermal neoplasia. Nature 421:639–643 | Article | PubMed | ISI | ChemPort |
28. Dajee M, Tarutani M, Deng H, Cai T, Khavari PA (2002) Epidermal Ras blockade demonstrates spatially localized Ras promotion of proliferation and inhibition of differentiation. Oncogene 21:1527–1538 | Article | PubMed | ISI | ChemPort |
29. de Bruin A, Caldelari R, Williamson L, Suter MM, Hunziker T, Wyder M et al. (2007) Plakoglobin-dependent disruption of the desmosomal plaque in pemphigus vulgaris. Exp Dermatol 16:468–475 | Article | PubMed | ChemPort |
30. Dejana E (2004) Endothelial cell–cell junctions: happy together. Nat Rev Mol Cell Biol 5:261–270 | Article | PubMed | ISI | ChemPort |
31. Denning MF, Dlugosz AA, Williams EK, Szallasi Z, Blumberg PM, Yuspa SH (1995) Specific protein kinase C isozymes mediate the induction of keratinocyte differentiation markers by calcium. Cell Growth Differ 6:149–157 | PubMed | ISI | ChemPort |
32. Dusek RL, Godsel LM, Green KJ (2007) Discriminating roles of desmosomal cadherins: beyond desmosomal adhesion. J Dermatol Sci 45:7–21 | Article | PubMed | ISI | ChemPort |
33. Elias PM, Matsuyoshi N, Wu H, Lin C, Wang ZH, Brown BE et al. (2001) Desmoglein isoform distribution affects stratum corneum structure and function. J Cell Biol 153:243–249 | Article | PubMed | ISI | ChemPort |
34. Engelman JA, Luo J, Cantley LC (2006) The evolution of phosphatidylinositol 3-kinases as regulators of growth and metabolism. Nat Rev Genet 7:606–619 | Article | PubMed | ISI | ChemPort |
35. Espada J, Perez-Moreno M, Braga VMM, Rodriguez-Viciana P, Cano A (1999) H-Ras activation promotes cytoplasmic accumulation and phosphoinositide 3-OH kinase association of {beta}-catenin in epidermal keratinocytes. J Cell Biol 146:967–980 | Article | PubMed | ISI | ChemPort |
36. Frisch SM, Screaton RA (2001) Anoikis mechanisms. Curr Opin Cell Biol 13:555–562 | Article | PubMed | ISI | ChemPort |
37. Frusic-Zlotkin M, Pergamentz R, Michel B, David M, Mimouni D, Brégégère F et al. (2005) The interaction of pemphigus autoimmunoglobulins with epidermal cells: activation of the fas apoptotic pathway and the use of caspase activity for pathogenicity tests of pemphigus patients. Ann N Y Acad Sci 1050:371–379 | Article | PubMed | ChemPort |
38. Frusic-Zlotkin M, Raichenberg D, Wang X, David M, Michel B, Milner Y (2006) Apoptotic mechanism in pemphigus autoimmunoglobulins—induced acantholysis-possible involvement of the EGF receptor. Autoimmunity 39:563–575 | Article | PubMed | ChemPort |
39. Fuchs M, Muller T, Lerch MM, Ullrich A (1996) Association of human protein-tyrosine phosphatase kappa with members of the armadillo family. J Biol Chem 271:16712–16719 | Article | PubMed | ISI | ChemPort |
40. Futei Y, Amagai M, Sekiguchi M, Nishifuji K, Fujii Y, Nishikawa T (2000) Use of domain-swapped molecules for conformational epitope mapping of desmoglein 3 in pemphigus vulgaris. J Invest Dermatol 115:829–834 | Article | PubMed | ISI | ChemPort |
41. Gagnoux-Palacios L, Dans M, van't Hof W, Mariotti A, Pepe A, Meneguzzi G et al. (2003) Compartmentalization of integrin alpha6beta4 signaling in lipid rafts. J Cell Biol 162:1189–1196 | Article | PubMed | ISI | ChemPort |
42. Gambardella L, Barrandon Y (2003) The multifaceted adult epidermal stem cell. Curr Opin Cell Biol 15:771–777 | Article | PubMed | ISI | ChemPort |
43. Gaudry CA, Palka HL, Dusek RL, Huen AC, Khandekar MJ, Hudson LG et al. (2001) Tyrosine-phosphorylated plakoglobin is associated with desmogleins but not desmoplakin after epidermal growth factor receptor activation. J Biol Chem 276:24871–24880 | Article | PubMed | ISI | ChemPort |
44. Green H (1977) Terminal differentiation of cultured human epidermal cells. Cell 11:405–416 | Article | PubMed | ISI | ChemPort |
45. Green KJ, Gaudry CA (2000) Are desmosomes more than tethers for intermediate filaments? Nat Rev Mol Cell Biol 1:208–216 | Article | PubMed | ISI | ChemPort |
46. Gumbiner BM (2000) Regulation of cadherin adhesive activity. J Cell Biol 148:399–404 | Article | PubMed | ISI | ChemPort |
47. Guo W, Giancotti FG (2004) Integrin signalling during tumour progression. Nat Rev Mol Cell Biol 5:816–826 | Article | PubMed | ISI | ChemPort |
48. Guo W, Pylayeva Y, Pepe A, Yoshioka T, Muller WJ, Inghirami G et al. (2006) Beta 4 integrin amplifies ErbB2 signaling to promote mammary tumorigenesis. Cell 126:489–502 | Article | PubMed | ISI | ChemPort |
49. Hofmann C, Obermeier F, Artinger M, Hausmann M, Falk W, Schoelmerich J et al. (2007) Cell–cell contacts prevent anoikis in primary human colonic epithelial cells. Gastroenterology 132:587–600 | Article | PubMed | ChemPort |
50. Hoschuetzky H, Aberle H, Kemler R (1994) Beta-catenin mediates the interaction of the cadherin–catenin complex with epidermal growth factor receptor. J Cell Biol 127:1375–1380 | Article | PubMed | ISI | ChemPort |
51. Ishizawar R, Parsons SJ (2004) c-Src and cooperating partners in human cancer. Cancer Cell 6:209–214 | Article | PubMed | ISI | ChemPort |
52. Janes SM, Watt FM (2006) New roles for integrins in squamous-cell carcinoma. Nat Rev Cancer 6:175–183 | Article | PubMed | ChemPort |
53. Jones JCR, Arnn J, Staehelin LA, Goldman RD (1984) Human autoantibodies against desmosomes: possible causative factors in pemphigus. Proc Natl Acad Sci USA 81:2781–2785 | Article | PubMed | ChemPort |
54. Kang HG, Jenabi JM, Zhang J, Keshelava N, Shimada H, May WA et al. (2007) E-cadherin cell–cell adhesion in Ewing tumor cells mediates suppression of anoikis through activation of the ErbB4 tyrosine kinase. Cancer Res 67:3094–3105 | Article | PubMed | ChemPort |
55. Koch PJ, Mahoney MG, Ishikawa H, Pulkkinen L, Uitto J, Shultz L et al. (1997) Targeted disruption of the pemphigus vulgaris antigen (desmoglein 3) gene in mice causes loss of keratinocyte cell adhesion with a phenotype similar to pemphigus vulgaris. J Cell Biol 137:1091–1102 | Article | PubMed | ISI | ChemPort |
56. Kodama S, Ikeda S, Asahara T, Kishida M, Kikuchi A (1999) Axin directly interacts with plakoglobin and regulates its stability. J Biol Chem 274:27682–27688 | Article | PubMed | ISI | ChemPort |
57. Koivisto L, Jiang G, Hakkinen L, Chan B, Larjava H (2006) HaCaT keratinocyte migration is dependent on epidermal growth factor receptor signaling and glycogen synthase kinase-3alpha. Exp Cell Res 312:2791–2805 | Article | PubMed | ChemPort |
58. Kolly C, Suter MM, Muller EJ (2005) Proliferation, cell cycle exit, and onset of terminal differentiation in cultured keratinocytes: pre-programmed pathways in control of C-Myc and Notch1 prevail over extracellular calcium signals. J Invest Dermatol 124:1014–1025 | Article | PubMed | ISI | ChemPort |
59. Kovacs EM, Ali RG, McCormack AJ, Yap AS (2002) E-cadherin homophilic ligation directly signals through Rac and phosphatidylinositol 3-kinase to regulate adhesive contacts. J Biol Chem 277:6708–6718 | Article | PubMed | ISI | ChemPort |
60. Kusumi A, Sako Y, Yamamoto M (1993) Confined lateral diffusion of membrane receptors as studied by single particle tracking (nanovid microscopy). Effects of calcium-induced differentiation in cultured epithelial cells. Biophys J 65:2021–2040 | PubMed | ISI | ChemPort |
61. Lampugnani MG, Orsenigo F, Gagliani MC, Tacchetti C, Dejana E (2006) Vascular endothelial cadherin controls VEGFR-2 internalization and signaling from intracellular compartments. J Cell Biol 174:593–604 | Article | PubMed | ChemPort |
62. Laprise P, Chailler P, Houde M, Beaulieu JF, Boucher MJ, Rivard N (2002) Phosphatidylinositol 3-kinase controls human intestinal epithelial cell differentiation by promoting adherens junction assembly and p38 MAPK activation. J Biol Chem 277:8226–8234 | Article | PubMed | ISI | ChemPort |
63. Laprise P, Langlois MJ, Boucher MJ, Jobin C, Rivard N (2004) Down-regulation of MEK/ERK signaling by E-cadherin-dependent PI3K/Akt pathway in differentiating intestinal epithelial cells. J Cell Physiol 199:32–39 | Article | PubMed | ChemPort |
64. Lewis JE, Wahl JK III, Sass KM, Jensen PJ, Johnson KR, Wheelock MJ (1997) Cross-talk between adherens junctions and desmosomes depends on plakoglobin. J Cell Biol 136:919–934 | Article | PubMed | ISI | ChemPort |
65. Li W, Marshall C, Mei L, Gelfand J, Seykora JT (2007) Srcasm corrects Fyn-induced epidermal hyperplasia by kinase downregulation. J Biol Chem 282:1161–1169 | Article | PubMed | ChemPort |
66. Liebner S, Cavallaro U, Dejana E (2006) The multiple languages of endothelial cell-to-cell communication. Arterioscler Thromb Vasc Biol 26:1431–1438 | Article | PubMed | ChemPort |
67. Lopez-Rovira T, Silva-Vargas V, Watt FM (2005) Different consequences of beta1 integrin deletion in neonatal and adult mouse epidermis reveal a context-dependent role of integrins in regulating proliferation, differentiation, and intercellular communication. J Invest Dermatol 125:1215–1227 | Article | PubMed | ChemPort |
68. Mainiero F, Murgia C, Wary KK, Curatola AM, Pepe A, Blumemberg M et al. (1997) The coupling of alpha6beta4 integrin to Ras–MAP kinase pathways mediated by Shc controls keratinocyte proliferation. EMBO J 16:2365–2375 | Article | PubMed | ISI | ChemPort |
69. Mainiero F, Pepe A, Yeon M, Ren Y, Giancotti FG (1996) The intracellular functions of alpha6beta4 integrin are regulated by EGF. J Cell Biol 134:241–253 | Article | PubMed | ISI | ChemPort |
70. Marcozzi C, Burdett ID, Buxton RS, Magee AI (1998) Coexpression of both types of desmosomal cadherin and plakoglobin confers strong intercellular adhesion. J Cell Sci 111:495–509 | PubMed | ISI | ChemPort |
71. Mariotti A, Kedeshian PA, Dans M, Curatola AM, Gagnoux-Palacios L, Giancotti FG (2001) EGF-R signaling through Fyn kinase disrupts the function of integrin alpha6beta4 at hemidesmosomes: role in epithelial cell migration and carcinoma invasion. J Cell Biol 155:447–458 | Article | PubMed | ISI | ChemPort |
72. McLachlan RW, Kraemer A, Helwani FM, Kovacs EM, Yap AS (2007) E-cadherin adhesion activates c-Src signaling at cell–cell contacts. Mol Biol Cell 18:3214–3223 | Article | PubMed | ChemPort |
73. Merritt AJ, Berika MY, Zhai W, Kirk SE, Ji B, Hardman MJ et al. (2002) Suprabasal desmoglein 3 expression in the epidermis of transgenic mice results in hyperproliferation and abnormal differentiation. Mol Cell Biol 22:5846–5858 | Article | PubMed | ISI | ChemPort |
74. Miettinen PJ, Berger JE, Meneses J, Phung Y, Pedersen RA, Werb Z et al. (1995) Epithelial immaturity and multiorgan failure in mice lacking epidermal growth factor receptor. Nature 376:337–341 | Article | PubMed | ISI | ChemPort |
75. Miranti CK, Brugge JS (2002) Sensing the environment: a historical perspective on integrin signal transduction. Nat Cell Biol 4:E83–E90 | Article | PubMed | ISI | ChemPort |
76. Missero C, Di Cunto F, Kiyokawa H, Koff A, Dotto GP (1996) The absence of p21Cip1/WAF1 alters keratinocyte growth and differentiation and promotes ras-tumor progression. Genes Dev 10:3065–3075 | Article | PubMed | ISI | ChemPort |
77. Muller EJ, Hunziker T, Suter MM (2007) Keratin intermediate filament retraction is linked to plakoglobin-dependent signaling in pemphigus vulgaris. J Am Acad Dermatol 56:890–1
78. Nikolopoulos SN, Blaikie P, Yoshioka T, Guo W, Puri C, Tacchetti C et al. (2005) Targeted deletion of the integrin beta4 signaling domain suppresses laminin-5-dependent nuclear entry of mitogen-activated protein kinases and NF-kappaB, causing defects in epidermal growth and migration. Mol Cell Biol 25:6090–6102 | Article | PubMed | ChemPort |
79. Nishio H, Matsui K, Tsuji H, Tamura A, Suzuki K (2001) Immunolocalisation of the Janus kinases (JAK)–signal transducers and activators of transcription (STAT) pathway in human epidermis. J Anat 198:581–589 | Article | PubMed | ISI | ChemPort |
80. Noren NK, Niessen CM, Gumbiner BM, Burridge K (2001) Cadherin engagement regulates rho family GTPases. J Biol Chem 276:33305–33308 | Article | PubMed | ISI | ChemPort |
81. Oskarsson T, Essers MA, Dubois N, Offner S, Dubey C, Roger C et al. (2006) Skin epidermis lacking the c-Myc gene is resistant to Ras-driven tumorigenesis but can reacquire sensitivity upon additional loss of the p21Cip1 gene. Genes Dev 20:2024–2029 | Article | PubMed | ISI | ChemPort |
82. Owens DM, Romero MR, Gardner C, Watt FM (2003) Suprabasal {alpha}6{beta}4 integrin expression in epidermis results in enhanced tumorigenesis and disruption of TGF{beta} signalling. J Cell Sci 116:3783–3791 | Article | PubMed | ISI | ChemPort |
83. Pang J-H, Kraemer A, Stehbens SJ, Frame MC, Yap AS (2005) Recruitment of phosphoinositide 3-kinase defines a positive contribution of tyrosine kinase signaling to E-cadherin function. J Biol Chem 280:3043–3050 | Article | PubMed | ChemPort |
84. Pankow S, Bamberger C, Klippel A, Werner S (2006) Regulation of epidermal homeostasis and repair by phosphoinositide 3-kinase. J Cell Sci 119:4033–4046 | Article | PubMed | ISI | ChemPort |
85. Payne AS, Hanakawa Y, Amagai M, Stanley JR (2004) Desmosomes and disease: pemphigus and bullous impetigo. Curr Opin Cell Biol 16:536–543 | Article | PubMed | ISI | ChemPort |
86. Pece S, Chiariello M, Murga C, Gutkind JS (1999) Activation of the protein kinase Akt/PKB by the formation of E-cadherin-mediated cell–cell junctions. Evidence for the association of phosphatidylinositol 3-kinase with the E-cadherin adhesion complex. J Biol Chem 274:19347–19351 | Article | PubMed | ISI | ChemPort |
87. Pece S, Gutkind JS (2000) Signaling from E-cadherins to the MAPK pathway by the recruitment and activation of epidermal growth factor receptors upon cell–cell contact formation. J Biol Chem 275:41227–41233 | Article | PubMed | ISI | ChemPort |
88. Peng XD, Xu PZ, Chen ML, Hahn-Windgassen A, Skeen J, Jacobs J et al. (2003) Dwarfism, impaired skin development, skeletal muscle atrophy, delayed bone development, and impeded adipogenesis in mice lacking Akt1 and Akt2. Genes Dev 17:1352–1365 | Article | PubMed | ISI | ChemPort |
89. Peppelenbosch MP, Tertoolen LG, Hage WJ, de Laat SW (1993) Epidermal growth factor-induced actin remodeling is regulated by 5-lipoxygenase and cyclooxygenase products. Cell 74:565–575 | Article | PubMed | ISI | ChemPort |
90. Perez-Moreno M, Davis MA, Wong E, Pasolli HA, Reynolds AB, Fuchs E (2006) p120-catenin mediates inflammatory responses in the skin. Cell 124:631–644 | Article | PubMed | ISI | ChemPort |
91. Perrais M, Chen X, Perez-Moreno M, Gumbiner BM (2007) E-cadherin homophilic ligation inhibits cell growth and epidermal growth factor receptor signaling independent of other cell interactions. Mol Biol Cell 18:2013–2025 | Article | PubMed | ChemPort |
92. Pierres A, Prakasam A, Touchard D, Benoliel AM, Bongrand P, Leckband D (2007) Dissecting subsecond cadherin bound states reveals an efficient way for cells to achieve ultrafast probing of their environment. FEBS Lett 581:1841–1846 | PubMed | ChemPort |
93. Pillai S, Bikle DD, Mancianti ML, Cline P, Hincenbergs M (1990) Calcium regulation of growth and differentiation of normal human keratinocytes: modulation of differentiation competence by stages of growth and extracellular calcium. J Cell Physiol 143:294–302 | Article | PubMed | ISI | ChemPort |
94. Poumay Y, Pittelkow MR (1995) Cell density and culture factors regulate keratinocyte commitment to differentiation and expression of suprabasal K1/K10 keratins. J Invest Dermatol 104:271–276 | Article | PubMed | ISI | ChemPort |
95. Qian X, Karpova T, Sheppard AM, McNally J, Lowy DR (2004) E-cadherin-mediated adhesion inhibits ligand-dependent activation of diverse receptor tyrosine kinases. EMBO J 23:1739–1748 | Article | PubMed | ISI | ChemPort |
96. Raghavan S, Vaezi A, Fuchs E (2003) A role for alphabeta1 integrins in focal adhesion function and polarized cytoskeletal dynamics. Dev Cell 5:415–427 | Article | PubMed | ISI | ChemPort |
97. Raymond K, Kreft M, Janssen H, Calafat J, Sonnenberg A (2005) Keratinocytes display normal proliferation, survival and differentiation in conditional beta4-integrin knockout mice. J Cell Sci 118:1045–1060 | Article | PubMed | ISI | ChemPort |
98. Reichelt J, Magin TM (2002) Hyperproliferation, induction of c-Myc and 14-3-3sigma, but no cell fragility in keratin-10-null mice. J Cell Sci 115:2639–2650 | PubMed | ChemPort |
99. Rodeck U, Jost M, Kari C, Shih DT, Lavker RM, Ewert DL et al. (1997) EGF-R dependent regulation of keratinocyte survival. J Cell Sci 110(Part 2):113–121 | PubMed | ISI | ChemPort |
100. Roh JY, Stanley JR (1995) Intracellular domain of desmoglein 3 (pemphigus vulgaris antigen) confers adhesive function on the extracellular domain of E-cadherin without binding catenins. J Cell Biol 128:939–947 | Article | PubMed | ChemPort |
101. Sabapathy K, Wagner EF (2004) JNK2: a negative regulator of cellular proliferation. Cell Cycle 3:1520–1523 | PubMed | ChemPort |
102. Sanchez-Carpintero I, Espana A, Pelacho B, López Moratalla N, Rubenstein DS, Diaz LA et al. (2004) In vivo blockade of pemphigus vulgaris acantholysis by inhibition of intracellular signal transduction cascades. Br J Dermatol 151:565–570 | Article | PubMed | ChemPort |
103. Sato M, Aoyama Y, Kitajima Y (2000) Assembly pathway of desmoglein 3 to desmosomes and its perturbation by pemphigus vulgaris-IgG in cultured keratinocytes, as revealed by time-lapsed labeling immunoelectron microscopy. Lab Invest 80:1583–1592 | Article | PubMed | ISI | ChemPort |
104. Schmidt E, Zillikens D (2000) Autoimmune and inherited subepidermal blistering diseases: advances in the clinic and the laboratory. Adv Dermatol 16:113–157 | PubMed | ChemPort |
105. Schulze K, Lietti C, Strauss C, Williamson L, Suter MM, Muller EJ (2007) Pemphigus vulgaris identifies non-junctional desmoglein 3 as regulator of tissue homeostasis via plakglobin signaling. J Invest Dermatol 127:485
106. Seveau S, Bierne H, Giroux S, Prevost MC, Cossart P (2004) Role of lipid rafts in E-cadherin—and HGF-R/Met—mediated entry of Listeria monocytogenes into host cells. J Cell Biol 166:743–753 | Article | PubMed | ISI | ChemPort |
107. Shaw LM, Rabinovitz I, Wang HH, Toker A, Mercurio AM (1997) Activation of phosphoinositide 3-OH kinase by the alpha6beta4 integrin promotes carcinoma invasion. Cell 91:949–960 | Article | PubMed | ISI | ChemPort |
108. Stover DR, Becker M, Liebetanz J, Lydon NB (1995) Src phosphorylation of the epidermal growth factor receptor at novel sites mediates receptor interaction with Src and P85 alpha. J Biol Chem 270:15591–15597 | Article | PubMed | ISI | ChemPort |
109. Suter MM, de Bruin A, Wyder M, Warm S, Credille K, Crameri FM et al. (1998) Autoimmune diseases of domestic animals: an update. Adv Vet Dermatol 3:321–337
110. Suter MM, Wilkinson JE, Dougherty EP, Lewis RM (1990) Ultrastructural localization of pemphigus vulgaris antigen on canine keratinocytes in vivo and in vitro. Am J Vet Res 51:507–511 | PubMed | ChemPort |
111. Takahashi K, Suzuki K (1996) Density-dependent inhibition of growth involves prevention of EGF receptor activation by E-cadherin-mediated cell–cell adhesion. Exp Cell Res 226:214–222 | Article | PubMed | ISI | ChemPort |
112. Takahashi Y, Patel HP, Labib RS, Diaz LA, Anhalt GJ (1985) Experimentally induced pemphigus vulgaris in neonatal BALB/c mice: a time-course study of clinical, immunologic, ultrastructural, and cytochemical changes. J Invest Dermatol 84:41–46 | Article | PubMed | ISI | ChemPort |
113. Takeda H, Nagafuchi A, Yonemura S, Tsukita S, Behrens J, Birchmeier W et al. (1995) V-src kinase shifts the cadherin-based cell adhesion from the strong to the weak state and beta catenin is not required for the shift. J Cell Biol 131:1839–1847 | Article | PubMed | ISI | ChemPort |
114. Theisen CS, Wahl JK III, Johnson KR, Wheelock MJ (2007) NHERF links the N-cadherin/catenin complex to the PDGF receptor to modulate the actin cytoskeleton and regulate cell motility. Mol Biol Cell 18:1220–1232 | Article | PubMed | ChemPort |
115. Thiery JP (2002) Epithelial–mesenchymal transitions in tumour progression. Nat Rev Cancer 2:442–454 | Article | PubMed | ISI | ChemPort |
116. Thrash BR, Menges CW, Pierce RH, McCance DJ (2006) AKT1 provides an essential survival signal required for differentiation and stratification of primary human keratinocytes. J Biol Chem 281:12155–12162 | Article | PubMed | ISI | ChemPort |
117. Tinkle CL, Lechler T, Pasolli HA, Fuchs E (2004) Conditional targeting of E-cadherin in skin: insights into hyperproliferative and degenerative responses. Proc Natl Acad Sci USA 101:552–557 | Article | PubMed | ChemPort |
118. Toker A, Yoeli-Lerner M (2006) Akt signaling and cancer: surviving but not moving on. Cancer Res 66:3963–3966 | Article | PubMed | ISI | ChemPort |
119. Tsunoda K, Ota T, Aoki M, Yamada T, Nagai T, Nakagawa T et al. (2003) Induction of pemphigus phenotype by a mouse monoclonal antibody against the amino-terminal adhesive interface of desmoglein 3. J Immunol 170:2170–2178 | PubMed | ISI | ChemPort |
120. Tunggal JA, Helfrich I, Schmitz A, Schwarz H, Günzel D, Fromm M et al. (2005) E-cadherin is essential for in vivo epidermal barrier function by regulating tight junctions. EMBO J 24:1146–1156 | Article | PubMed | ISI | ChemPort |
121. Vasioukhin V, Bauer C, Degenstein L, Wise B, Fuchs E (2001) Hyperproliferation and defects in epithelial polarity upon conditional ablation of alpha-catenin in skin. Cell 104:605–617 | Article | PubMed | ISI | ChemPort |
122. Vasioukhin V, Fuchs E (2001) Actin dynamics and cell–cell adhesion in epithelia. Curr Opin Cell Biol 13:76–84 | Article | PubMed | ISI | ChemPort |
123. Vivanco I, Sawyers CL (2002) The phosphatidylinositol 3-kinase AKT pathway in human cancer. Nat Rev Cancer 2:489–501 | Article | PubMed | ISI | ChemPort |
124. Waikel RL, Kawachi Y, Waikel PA, Wang XJ, Roop DR (2001) Deregulated expression of c-Myc depletes epidermal stem cells. Nat Genet 28:165–168 | Article | PubMed | ISI | ChemPort |
125. Waschke J, Spindler V, Bruggeman P, Zillikens D, Schmidt G, Drenckhahn D (2006) Inhibition of Rho A activity causes pemphigus skin blistering. J Cell Biol 175:721–727 | Article | PubMed | ISI | ChemPort |
126. Watt FM (2002) Role of integrins in regulating epidermal adhesion, growth and differentiation. EMBO J 21:3919–3926 | Article | PubMed | ISI | ChemPort |
127. Watt FM, Kubler MD, Hotchin NA, Nicholson LJ, Adams JC (1993) Regulation of keratinocyte terminal differentiation by integrin–extracellular matrix interactions. J Cell Sci 106(Part 1):175–182 | PubMed | ISI | ChemPort |
128. Watt FM, Lo Celso C, Silva-Vargas V (2006) Epidermal stem cells: an update. Curr Opin Genet Dev 16:518–524 | Article | PubMed | ISI | ChemPort |
129. Weston CR, Wong A, Hall JP, Goad ME, Flavell RA, Davis RJ (2004) The c-Jun NH2-terminal kinase is essential for epidermal growth factor expression during epidermal morphogenesis. Proc Natl Acad Sci USA 101:14114–14119 | Article | PubMed | ChemPort |
130. White DE, Kurpios NA, Zuo D, Hassell JA, Blaess S, Mueller U et al. (2004) Targeted disruption of beta1-integrin in a transgenic mouse model of human breast cancer reveals an essential role in mammary tumor induction. Cancer Cell 6:159–170 | Article | PubMed | ISI | ChemPort |
131. Wilhelmsen K, Litjens SH, Sonnenberg A (2006) Multiple functions of the integrin alpha6beta4 in epidermal homeostasis and tumorigenesis. Mol Cell Biol 26:2877–2886 | Article | PubMed | ISI | ChemPort |
132. Williamson L, Hunziker T, Suter MM, Muller EJ (2007a) Nuclear c-Myc: a molecular marker for early stage pemphigus vulgaris. J Invest Dermatol 127:1549–1555 | Article | ChemPort |
133. Williamson L, Raess NA, Caldelari R, Zakher A, de Bruin A, Posthaus H et al. (2006) Pemphigus vulgaris identifies plakoglobin as key suppressor of c-Myc in the skin. EMBO J 25:3298–3309 | Article | PubMed | ISI | ChemPort |
134. Williamson L, Suter MM, Olivry T, Wyder M, Muller EJ (2007b) Upregulation of c-Myc may contribute to the pathogenesis of canine pemphigus vulgaris. Vet Dermatol 18:12–17 | Article |
135. Woodfield RJ, Hodgkin MN, Akhtar N, Morse MA, Fuller KJ, Saqib K et al. (2001) The p85 subunit of phosphoinositide 3-kinase is associated with beta-catenin in the cadherin-based adhesion complex. Biochem J 360:335–344 | Article | PubMed | ISI | ChemPort |
136. Xie Z, Bikle DD (2007) The recruitment of phosphatidylinositol 3-kinase to the E-cadherin/catenin complex at the plasma membrane is required for calcium-induced phospholipase C-gamma 1 activation and human keratinocyte differentiation. J Biol Chem 282:8695–8703 | Article | PubMed | ChemPort |
137. Yamamoto Y, Aoyama Y, Shu E, Tsunoda K, Amagai M, Kitajima Y (2007) Anti-desmoglein 3 (Dsg3) monoclonal antibodies deplete desmosomes of Dsg3 and differ in their Dsg3-depleting activities related to pathogenicity. J Biol Chem 282:17866–17876 | Article | PubMed | ChemPort |
138. Young P, Boussadia O, Halfter H, Grose R, Berger P, Leone DP et al. (2003) E-cadherin controls adherens junctions in the epidermis and the renewal of hair follicles. EMBO J 22:5723–5733 | Article | PubMed | ISI | ChemPort |
139. Yuspa SH, Kilkenny AE, Steinert PM, Roop DR (1989) Expression of murine epidermal differentiation markers is tightly regulated by restricted extracellular calcium concentrations in vitro. J Cell Biol 109:1207–1217 | Article | PubMed | ISI | ChemPort |
140. Zenz R, Wagner EF (2006) Jun signalling in the epidermis: from developmental defects to psoriasis and skin tumors. Int J Biochem Cell Biol 38:1043–1049 | Article | PubMed | ISI | ChemPort |
141. Zhang JY, Green CL, Tao S, Khavari PA (2004) NF-kappaB RelA opposes epidermal proliferation driven by TNFR1 and JNK. Genes Dev 18:17–22 | Article | PubMed | ISI | ChemPort |

Acknowledgments

We are most grateful to Professor K Green and Dr S Getsios, Departments of Pathology and Dermatology, Northwestern University, Chicago, Professor T Hunziker, Department of Dermatology, University of Bern, Switzerland and Dr E. Calautti, MBC, University of Turin, Italy, for helpful comments on the paper, as well as Peter Girling, CELLnTEC advanced cell systems, Bern, Switzerland, for editorial support. The work from our laboratory was supported by the Martha Stiftung (Zürich) and the Swiss National Science Foundation (Grant 3100A0-107243).