Ezrin Orchestrates Signal Transduction in Airway Cells

Ezrin modulates signal pathways mainly by connecting membrane receptor complexes to signaling pathways ( Youn et al. 2009 ; Pore et al. 2015 ; Clucas and Valderrama 2014 ). The RhoA/ROCK pathway is a critical signaling pathway that regulates both ezrin and airway smooth muscle contractions in asthma ( Kosako et al. 2000 ). RhoA interacts with various effectors to regulate the downstream signaling pathways and the actin cytoskeleton ( Schmieder et al. 2004 ; Anastasiadis et al. 2000 ). RhoA activates ezrin, which promotes the assembly of stress fibers and the polymerization of cortical actin in multiple cells, including fibroblasts, A431 and NIH3T3 cells ( Mackay et al. 1997 ; Yonemura et al. 2002 ; Song et al. 2000 ). The activation of RhoA increases the intracellular levels of PIP 2 , which induces the phosphorylation of ezrin threonine-567 and therefore releases the ezrin C-terminal domain, allowing it to interact with F-actin ( Castellani et al. 2012 ). The EGF-induced ezrin activation is mediated by RhoA because the inhibition of RhoA by fasudil (a potent Rho-kinase chemical inhibitor and vasodilator) prevents the activation of ezrin ( Ma et al. 2013 ). Pretreatment with fasudil before the EGF stimulation decreased the phosphorylation of RhoA and the expression of ezrin in MDA-MB-231 cells. These results show that RhoA acts as an upstream factor that regulates ezrin ( Ma et al. 2009 ). The inhibition of RhoA by C3 transferase (an exoenzyme that inhibits the addition of ADP-ribose moieties to Rho-like proteins) in cortical neurons abrogates the potential of netrin-1 to induce ERM phosphorylation via the DCC receptor ( Lawrence et al. 2016 ). Ezrin also interacts with Rho-GDI and enhances Rho activation ( Takahashi et al. 1997 ; Nethe and Hordijk 2010 ). The binding of ezrin to Rho-GDI contributes to the activation of RhoA in MDCK cells after stimulation with podocalyxin ( Schmieder et al. 2004 ). Once activated, RhoA maintains the ezrin activation and allows the connection of podocalyxin to actin at the apical cell membrane ( Schmieder et al. 2004 ). Clinical evidence supports the relationship between RhoA and ezrin because more than 60% of ezrin-positive osteosarcomas show RhoA overexpression ( Chiappetta et al. 2014 ). RhoA colocalizes with ezrin at the membrane ruffles of human endothelial cells that contribute to cellular migration during development ( Menager et al. 1999 ). The phosphorylation of podocalyxin prevents its binding to ezrin and its subsequent dissociation from Rho-GDI. Thus, RhoA prevents the phosphorylation of ezrin by blocking its C-terminal actin-binding domain and inducing the dissociation of the podocalyxin/ezrin complexes from the actin cytoskeleton ( Fukasawa et al. 2011 ). Treatment with podoplanin activates RhoA, which induces an epithelial mesenchymal transition in MDCK cells, and the activation of RhoA during the epithelial mesenchymal transition requires the binding of ezrin to the cytoplasmic tail of podoplanin (RKMSGRYSP) ( Martin-Villar et al. 2006 ). The blockade of ezrin via RNA interference reduced the migration of ectopic endometrial cells and decreased the expression levels of RhoA and ROCK. These results suggest that the Ezrin/RhoA/ROCK pathway is a potential therapeutic target for the treatment of endometriosis ( Jiang et al. 2012 ). ROCK regulates the cytoskeleton by phosphorylating the myosin regulatory light chain at serine-19 during cytokinesis ( Totsukawa et al. 2000 ). Recent studies have indicated that ROCK regulates the phosphorylation of ezrin, but some of these results are controversial. The ROCK inhibitor Y27632 prevents the phosphorylation of ezrin threonine-567 and its binding to the cytoskeleton after the transfection of RhoGEFs Net and Dbl in NIH3T3 fibroblasts ( Tran Quang et al. 2000 ). However, other studies have indicated that the ROCK inhibitors Y27632 and HA1077 enhanced the total phosphorylation of ezrin in human glioblastoma U251 cells during cytokinesis ( Kosako et al. 2000 ). These results are consistent with previous study showing that cisplatin induces the phosphorylation of ezrin by ROCK ( Rebillard et al. 2010 ). The cAMP/PKA pathway regulates ezrin phosphorylation, airway smooth muscle cell contraction, and cytokine production in asthma ( Horvat et al. 2012 ). Ezrin is highly expressed in human airway smooth muscle cells and modulates β 2 -adrenergic receptor signaling and muscle contraction in asthma ( Horvat et al. 2012 ). Ezrin promotes the interaction between PKA and other proteins, such as the chloride channel cystic fibrosis trans-membrane regulator (CFTR), the transporter Na + /H + exchanger type 3 (NHE3), and the scaffold protein EBP50 ( Vallée 2000 ; Soares et al. 2016 ). Ezrin is associated with cystic fibrosis, and several investigators have reported its binding to F508del CFTR, which is the most common mutation associated with this disease. In primary cystic fibrosis airway cells, the phosphorylation of ezrin threonine-567 enhances its binding to F508del CFTR and the actin cytoskeleton, activates the cAMP/PKA pathway, and rescues the F508del CFTR-dependent chloride secretion ( Abbattiscianni et al. 2016 ). In human airway epithelial cells, the formation of the CFTR-ezrin complex increases the cAMP-mediated activation of CFTR ( Ribas et al. 2007 ). In human airway smooth muscle cells, a disruption of the ezrin-PKA complex does not affect the β-agonist-induced accumulation of cAMP, but it increases the duration of plasma membrane-delineated cAMP ( Horvat et al. 2012 ). Similarly, ezrin enhances PKA signaling during the Na + /H + exchanger activation. In the mammalian kidney, EBP50 enhances the phosphorylation of NHE3 by tethering NHE3 with the PKA-ezrin complex ( Weinman et al. 2000 ). EBP50 also forms a complex with ezrin and the type-2 Na-Pi co-transporter in opossum kidney cells. Ezrin anchors PKA and triggers the phosphorylation of EBP50, which releases Npt2a and thereby inhibits the transport of phosphate ( Wang et al. 2012 ). PKA promotes the phosphorylation of ezrin threonine-567 and its localization at the actin membrane ruffles, which activates the cAMP-induced exchange protein in HEK293T-EPAC1 cells ( Parnell et al. 2015 ). Ezrin, CD99, soluble adenylyl cyclase, and PKA form a signaling complex at the endothelial junctions that regulates the movement of the recycling compartments to the site of the transendothelial migration ( Watson et al. 2015 ). The PKA-ezrin-Cx43 complex regulates the cAMP-induced gap junction connections ( Pidoux and Tasken 2015 ). Ezrin binds to the adenosine A2b receptor and stabilizes the receptor complex after an adenosine stimulation in intestinal epithelial cells ( Sitaraman et al. 2002 ). In contrast, ezrin, PKA, CFTR, and E3KARP (NHE3 kinase A regulatory protein) form a regulatory complex at the apical membranes of human airway epithelial cells to enhance the cAMP-induced activation of CFTR ( Sun et al. 2000 ). Because the cAMP/PKA pathway contributes to asthma, these results suggest that the modulation of this pathway by ezrin can contribute to asthma.

Ezrin is considered a key regulator of airway cells by modulating membrane-cortex interactions. The clinical implications of ezrin in asthma are growing and recent results suggest that ezrin is a potential therapeutic target in asthma. First, ezrin may participate in the repair of the bronchial epithelium during the early stages of asthma ( Fig. 3 ). EGF is a key factor in bronchial epithelial repair and CD44 enhances the repair efficiency ( Holgate 2000 ). The epithelial 3v isoform of CD44 is overexpressed in damaged tracheal epitheliums. Thus, EGF can induce the phosphorylation of ezrin, which links CD44 to the cortical actin cytoskeleton ( Holgate 2000 ). Ezrin has also been reported to induce the secretion of mucin 5 AC (a typical feature of airway remodeling in asthma) after a neutrophil elastase attack in human airway epithelial cells. Second, ezrin appears to be required for the activation, morphological change, and apoptosis of T lymphocytes in asthma ( Burkhardt et al. 2008 ). Ezrin is recruited to the immunological synapse between the T lymphocytes and APC ( Roumier et al. 2001 ). TCR activates Rac1 to dephosphorylate ezrin, and the phosphorylation of ezrin can be restored by PIP 2 binding ( Burkhardt et al. 2008 ). This regulation of ezrin collapses the microvilli, detaches the membrane from the actin cytoskeleton, and modulates the morphology of T lymphocytes ( Fig. 4 , upper panel) ( Brown et al. 2003 ). Ezrin also plays an important role in the distal pole complex (DPC) at the opposite site of the immunological synapse ( Allenspach et al. 2001 ; Shaffer et al. 2009 ). Ezrin binding is crucial for the phosphorylation of CD43 during T lymphocyte activation ( Cannon et al. 2011 ). Similarly, the activation of Rac1 induces ezrin dephosphorylation and regulates its binding to membrane receptors at the DPC of T lymphocytes ( Fig. 4 , lower panel) ( Burkhardt et al. 2008 ; Nijhara et al. 2004 ). Chemokine stimulation induces the dephosphorylation of ezrin, which triggers a rapid microvillar collapse in T lymphocytes, leading to the arrest of circulating T lymphocytes ( Brown et al. 2003 ). Meanwhile, ezrin antisense oligonucleotides protect T lymphocytes from CD95-mediated apoptosis ( Parlato et al. 2000 ). These results suggest that ezrin plays a role in T lymphocyte apoptosis and has clinical implications in asthma. Third, ezrin can control the contraction of airway smooth muscle cells. Phosphorylated ezrin colocalizes with internalized β 2 -adrenergic receptors and increases the re-sensitization of these receptors to cAMP stimulation, thus relaxing the airway smooth muscle cells ( Horvat et al. 2012 ; Cant and Pitcher 2005 ). Furthermore, phosphorylated ezrin binds to Rho-GDI, enhances RhoA activation, and induces airway smooth muscle cells relaxation ( Fig. 5 ) ( Takahashi et al. 1997 ; Fukata et al. 2001 ). Altogether, these results suggest that ezrin can contribute to asthma by affecting bronchial epithelium repair, T lymphocyte regulation, and airway smooth muscle cell relaxation.

Ezrin was discovered in multiple cellular processes and was thought to be different proteins because of its different electrophoretic mobility on the sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) ( Gould et al. 1989 ; Chambers and Bretscher 2005 ). Ezrin was first identified in 1981 as a polypeptide with an apparent molecular weight of 81-kDa on SDS-PAGE that was quickly phosphorylated after the stimulation of human A431 carcinoma cells with EGF ( Hunter and Cooper 1981 ). In 1983, ezrin was purified as a polypeptide with an apparent molecular weight of 80-kDa on SDS-PAGE from the microvillus cytoskeleton in chicken intestinal epithelial cells ( Bretscher 1983 ). In 1986, both polypeptides were compared and identified as ezrin (true molecular weight of 69-kDa), which was confirmed by immunoblotting, immunoprecipitation, two-dimensional gel electrophoresis, and protein sequencing ( Gould et al. 1986 ). Ezrin was also purified as an 82-kDa tumor antigen from the cytosol of a methylcholanthrene-induced sarcoma ( Ullrich et al. 1986 ; Fazioli et al. 1993 ). In 1988, ezrin was isolated from the microvillar membranes of human choriocarcinoma cells as a 75-kDa protein named cytovillin ( Pakkanen et al. 1987 ; Turunen et al. 1989 ). Ezrin was also purified as a 78-kDa cyclic AMP-dependent kinase anchoring protein (originally named AKAP78) enriched in murine gastric parietal cells ( Dransfield et al. 1997 ). These studies showed that ezrin is involved in multiple cellular processes ranging from the EGF stimulation of human carcinoma cells to tumor antigens. These results also indicated that the original confusion was due to the different electrophoretic mobility of ezrin on SDS-PAGE ( Gould et al. 1989 ). These differences were mostly due to its phosphorylation, indicating that ezrin is regulated by phosphorylation, which induces structural changes that affect its electrophoretic mobility. Ezrin contains an N-terminal FERM (four-point one, ezrin, radixin, moesin) domain (~300 residues), a central linker region (~200 residues), and a C-terminal ERM-associated domain (C-ERMAD, ~80 residues; Fig. 1 ) ( Jayasundar et al. 2012 ). The N-terminal FERM domain consists of the following three subdomains: F1–F3. These subdomains have structural (but not sequence) homology to known folded proteins. F1 is similar to ubiquitin, F2 is similar to acyl-CoA-binding protein, and F3 is similar to the PTB (phosphotyrosine binding)-domain. All ERMs have a central helical linker region composed of the predicted α-helical domain. Ezrin and radixin, but not moesin, have a proline-rich linker domain (~470–497 residues) preceding the C-ERMAD. Ezrin is normally a dormant inactive protein due to the intramolecular interaction between the N- and C-terminal domains. Ezrin is activated by phosphorylation, which dissociates the intramolecular interaction between the N- and C-terminal domains and allows the N-terminal domain to interact with membrane receptor complexes and the C-terminal domain to interact with F-actin ( Bretscher et al. 1997 ). Thus, the interaction between ezrin and other proteins is regulated by its phosphorylation in multiple domains by various kinases ( McRobert et al. 2003 ; Fehon et al. 2010 ). A detailed list of the ezrin phosphorylation sites, kinases, and biological activity is provided in Table 1 . The most common ezrin phosphorylation site is threonine-567 in the C-terminal domain ( Zhu et al. 2007 ). Ezrin threonine-567 is phosphorylated by calcineurin homologous protein-1 (CHP1), and the inhibition of CHP1 abrogates the interaction between ezrin and the Na + /H + exchanger 3 ( Di Sole et al. 2009 ). The phosphorylation of threonine-567 via phosphatidylinositol 4,5-bisphosphate (PIP 2 ) is necessary for ezrin activation during kidney epithelial LLC-PK1 cell morphogenesis ( Fievet et al. 2004 ). Ezrin threonine-567 is also phosphorylated by Akt, which induces microvilli in human BeWo trophoblastic cells ( Miura et al. 2015 ). The phosphorylation of threonine-567 enhances the binding of ezrin to the actin cytoskeleton and is crucial for establishing epithelial polarity in epithelial LLC-PK1 cells ( Gautreau et al. 2000 ). The phosphorylation of ezrin threonine-567 contributes to the migration and invasion of fibroblast-like synoviocytes in rheumatoid arthritis ( Xiao et al. 2014 ). The ezrin C-terminal domain is also regulated by the phosphorylation of threonine-477. The inhibition of Src by the PP2 chemical inhibitor prevented the phosphorylation of ezrin threonine-477 that is induced by pervanadate in human embryonic kidney 293 cells ( Heiska and Carpen 2005 ). The phosphorylation of threonine-477 by Src enhances the binding of ezrin to the Kelch-repeat and BTB/POZ domain containing 2 (KBTBD2) complex, which suggests that ezrin regulates cellular morphology and adhesion ( Heiska and Carpen 2005 ). The phosphorylation of the N-terminal domain regulates the binding of ezrin to a variety of proteins and membrane complexes. PKA phosphorylates ezrin serine-66 during the remodeling of the apical membrane during the acid secretion of gastric parietal cells ( Zhou et al. 2003 ). CDK5 (cyclin-dependent kinase 5) induces the phosphorylation of ezrin threonine-235, which regulates the cellular shape and cytoskeletal activity in human osteosarcoma SAOS-2 cells ( Yang and Hinds 2003 ). The phosphorylation of threonine-235 by CDK5 causes ezrin to dissociate from the Rho GDP-inhibitor (Rho-GDI) and prevents senescence-associated flat cell formation in SAOS-2 cells and human diploid fibroblasts ( Yang and Hinds 2006 ). Ezrin is also regulated by tyrosine phosphorylation in both the N-terminal domain (Y145) and the helical linker (Y353/4) ( Fehon et al. 2010 ). No tyrosine phosphorylation has been reported in the C-terminal domain, suggesting that tyrosine phosphorylation does not regulate the binding of the ezrin C-terminal domain to F-actin ( Fehon et al. 2010 ). As mentioned above, ezrin was originally identified as a protein that was quickly phosphorylated at tyrosine-353 after the stimulation of human epidermoid carcinoma A431 cells with EGF ( Hunter and Cooper 1981 ; Krieg and Hunter 1992 ). Ezrin is phosphorylated at both tyrosine-145 and 353 during carcinoma differentiation and invasion ( Saygideger-Kont et al. 2016 ; Bretscher 1989 ). The phosphorylation of ezrin tyrosine-353 is related to tumor differentiation associated with positive lymph node metastasis and shorter survival times in human invasive pancreatic carcinomas ( Wang et al. 2014 ; Fehon et al. 2010 ; Cui et al. 2010 ). The phosphorylation of ezrin tyrosine-353 by Akt (but not ERK1/2 or ROCK1) has been associated with the metastasis of tongue squamous cell carcinomas ( Wang et al. 2014 ). This phosphorylation induced invasive ductal carcinoma in human pancreatic intraepithelial neoplasia ( Oda et al. 2013 ). These results indicate that ezrin is regulated by multiple kinases (PKA, Akt, and Src ) during critical cellular processes including carcinoma tumor differentiation, survival, and metastasis.

Ezrin is a critical protein that orchestrates membrane receptor complexes and intracellular transduction pathways. Ezrin is also a complex protein in terms of its structural features and functional regulation. Ezrin has multiple phosphorylation sites, but the biological and clinical implications of some of these sites remain unknown. Ezrin interacts with key receptors and adaptors, which modulates their signaling pathways, including the RhoA/ROCK and the cAMP/PKA pathways, in critical cellular processes associated with asthma. The clinical implications of ezrin in asthma are growing, and recent results support that ezrin is a potential therapeutic target for the treatment of asthma. These results warrant future studies to elucidate the interactions between ezrin and membrane receptor complexes and the regulation of airway cells in physiological and clinical conditions, such as asthma.

Asthma is a clinical challenge in modern medicine that affects over 300 million people and causes over 250,000 deaths annually worldwide ( D’Amato et al. 2016 ; Lambrecht and Hammad 2015 ). Structural airway cells, such as smooth muscle and epithelial cells, are critical factors that contribute to asthma. Multiple studies suggest that alterations in the actin cytoskeleton cause a pathological contraction of the structural airway cells, which contributes to asthma ( Fletcher and Mullins 2010 ; Noble et al. 2014 ). In clinical studies, biopsied tissue from asthmatic patients showed reduced β-actin mRNA levels ( Glare et al. 2002 ). Altogether, these results encouraged investigators to study the actin-binding proteins in the airway cells and their clinical implications in asthma ( Tang 2015 ). This review analyzes recent studies on ezrin and its potential implications in asthma. Ezrin is a principal member of the ERM (ezrin–radixin–moesin) protein family, which includes actin-binding proteins of the band 4.1 superfamily because their N-termini are similar to those of the erythrocyte cytoskeletal protein band 4.1 ( Sagara et al. 1995 ; Vaheri et al. 1997 ; Gould et al. 1989 ; Ng et al. 2001 ). The ERM proteins are known as structural organizers that link membrane proteins to the underlying actin cytoskeleton. In addition to their structural role, the ERMs also regulate the interaction between receptor complexes and intracellular proteins, thereby modulating signal transduction pathways, such as the RhoA, Rho-associated protein kinase (ROCK), and cyclic AMP/protein kinase A (cAMP/PKA) pathways ( Iontcheva et al. 2004 ; Ponta et al. 2003 ; Celik et al. 2015 ). The ERMs are expressed in a developmental and tissue-specific pattern. Ezrin is mainly expressed in lymphocytes and epithelial cells, moesin is expressed in endothelial cells, and radixin is expressed in hepatocytes. Ezrin has recently attracted the attention of multiple investigators due to its role in key biological processes such as the immunological synapsis in T lymphocytes and the epidermal growth factor (EGF)-induced stimulation of human carcinoma tumor differentiation and metastasis ( Bretscher et al. 1997 ; Yoshida et al. 2016 ). The interaction between ezrin and various receptor complexes and intracellular targets is mainly regulated by phosphorylation ( Neisch and Fehon 2011 ). In this study, we review recent results of ezrin regulation and its physiological and clinical implications.