Encyclopedia of Signaling Molecules

2018 Edition
| Editors: Sangdun Choi

Eps8 (Epidermal Growth Factor Receptor Pathway Substrate 8)

  • Francesca Milanesi
  • Niels Volkmann
  • Giorgio Scita
  • Dorit Hanein
Reference work entry
DOI: https://doi.org/10.1007/978-3-319-67199-4_165

Synonyms

Historical Background

Eps8 (EGFR pathway substrate #8) is an actin-binding and signaling molecule with a molecular weight of 97KDa encoded by a gene comprising 21 exons located on the human chromosome 12p12. It was originally identified through an expression cloning approach designed to isolate intracellular substrates for the tyrosine kinase of the Epidermal Growth Factor Receptor (EGFR) (Fazioli et al. 1993). Eps8 is efficiently phosphorylated on tyrosine residues by a variety of both receptor and non-receptor tyrosine kinases (Fazioli et al. 1993). Following stimulation with neurotrophic factor BDNF, which critically controls growth and differentiation processes in the brain during development, through the activation of Trk tyrosine kinase receptors, Eps8 can also be phosphorylated, in a MAPK-dependent manner, on serine and threonine residues. Notably, while the physiological and functional implication of Eps8 tyrosine phosphorylation remains to be identified, BDNF induced, MAPK-dependent phosphorylation of S624 and T628 of mouse Eps8 are critical for regulation of its localization and functions (Menna et al. 2009) (see below).

Eps8 is the founding member of a family of eps8-related proteins, comprising three additional gene products in vertebrates, named eps8L1, eps8L2, and eps8L3 (Fig. 1a, b). All Eps8Ls display collinear topology and 27–42% identity to Eps8, sharing a similar modular organization into distinct structural and functional domains that comprise an N-terminal Phosphotyrosine Binding domain (PTB), a central Src homology-3 domain (SH3), two Proline rich regions, and a C-terminal “effector actin-binding region” (ABR) that mediates Eps8 interactions with actin filaments (Offenhauser et al. 2004). EPS8-like orthologues are also present in nonmammalian species (Fig. 1a, b). In Drosophila melanogaster there are two like genes called arouser and LP01469p. A single eps8 gene is, instead, present in Caernorabditis elegans that gives rise to two splicing isoform eps-8A and eps-8b, differing at their carboxyl termini. While the function of eps8 in the fly is unknown, in the nematode, eps-8 is essential for embryonic development (Croce et al. 2004). Furthermore, EPS-8A, but not EPS-8B protein, is specifically required for proper morphogenesis of intestinal cells. This latter phenotype could be precisely correlated with the evolutionarily conserved actin-binding activity, which is present in the C terminus of the EPS-8A isoform, providing genetic evidence that the actin-related properties of EPS-8 are critical for its function in the whole organism.
Eps8 (Epidermal Growth Factor Receptor Pathway Substrate 8), Fig. 1

The C-terminal region of EPS8 is highly conserved among metazoan homologues. (a) Multiple sequence alignment of EPS8 homologues. Protein sequences were identified by BLAST searches against all known protein sequences at NCBI or predicted from the genomic sequences of the specific organism using the GeneWise software (http://www.ebi.ac.uk/Wise2/). Sequences were aligned using the ClustalW program and the picture was produced using Jalview. Domains are boxed and indicated. PTB, phosphotyrosine binding domain; SH3, Src homology domain 3; ABR, Actin-binding region, EPS8 possesses a barbed end capping activity encoded by its evolutionarily conserved, C-terminal region (residues 648–821 for murine EPS8) (a). This region of EPS8 displays no sequence similarity with other known capping proteins or actin-binding motifs, but displays a similarity (similarity, E = 60.8; confidence E = 2.4) with the profile of the SAM Pointed (SAM_PNT) domain (a). SAM_PNT domains, which belong to the larger family of sterile alpha motif (SAM) (Slupsky et al. 1998), are present in a subset of the ETS-like transcription factors, including mammalian Ets1, Ets2, Erg, Fli1, GABPa, and Tel, and Drosophila Pnt and Yan (Slupsky et al. 1998). The C-terminal portion of EPS8 and its family members can be aligned with known SAM_PNT domain sequences, and its predicted secondary structure is superimposable to that determined for Ets1 (not shown). (b) Phylogenetic tree was produced with ClustalW and illustrated with FigTree (software by Andrew Rambaut, http://evolve.zoo.ox.ac.uk/software.html/id=figtree). GenBank accession numbers for each protein or coordinates of the genomic sequences from where they were reconstructed are provided in Table 1

Domain Organization and Interacting Partners

The modular organization of Eps8 and its family members supports the notion that Eps8L family proteins generally act as adaptors for the assembly of macromolecular complexes required for the transduction of signals from receptor tyrosine kinases (RTKs) leading to regulation of actin remodeling and cell migration (Fig. 1a). This section will describe what is currently known about the PTB and SH3 domain of Eps8, while more detailed information on the effector ABR will be provided in the section entitled “Eps8 and Actin Dynamics.”

Among the distinctive protein:protein interaction domains, the least functionally characterized is the N-terminal PTB domain. Originally, PTB domains were identified as protein–protein interaction modules, binding to a variety of NPXY containing motifs with high affinity when the tyrosine residue is phosphorylated, such as in response to RTK stimulation. This latter feature, however, is not shared by the majority of PTB domains that bind NPXY motifs irrespective of ligand phosphorylation or preferentially recognize unphosphorylated ligands, such as the PTB of the endocytic adaptor NUMB (Forman-Kay and Pawson 1999). The high structural homology between Eps8 and the latter PTB domain suggests that also Eps8 PTB may preferentially recognize unphosporylated NPXY motifs (Smith et al. 2006). In vitro evidence in this direction has, indeed, been provided; however, the lack of physiological validation of any of the putative binding partners of Eps8 PTB domain has prevented not only to validate the mode of binding of Eps8 PTB domain, but also to elucidate the functional consequence of this interaction module.

A canonical SH3 fold consists of two antiparallel beta sheets packed against each other at an approximate right angle. The first crystal structure of the SH3 domain of Eps8 revealed an intertwined dimer, characterized by “strand exchange,” in which the two antiparallel beta sheets are contributed by different polypeptide chains (Kishan et al. 1997). Surprisingly, this results in half-dimers whose folds are superimposable to that of a canonical SH3 module. An important consequence of the dimeric configuration is that the proline helix-binding groove is partially occluded in the context of the hybrid-dimer, impeding binding to ligands (Kishan et al. 1997). More recently, a monomeric structure of the SH3 domain of Eps8 has been reported, as obtained from crystals grown at low pH. In this case, the SH3 domain of Eps8 displays a canonical SH3 fold (Kishan et al. 2001). While this latter configuration is likely the one adopted under physiological conditions, dimerization of Eps8 SH3 domain may occur under some circumstances and might function as an “OFF” signal, which can be switched “ON” as the molecule becomes monomeric thereby allowing protein–protein interactions at the sites occluded in the dimer to occur. It should be noted, however, that the formation of a strand-exchanged dimeric Eps8–SH3 domain results in an extensive dimerization interface, much greater than usually observed for reversible regulated protein–protein associations in signal transduction. Thus, the physiological relevance of a dimer–monomer equilibrium in vivo remains to be established.

The identification of the optimal binding peptides for the SH3 domain of Eps8 provided another unexpected result. In spite of the overall conservation of the primary structure and the similarity to the canonical SH3 fold, the SH3 domain of Eps8 binds preferentially to peptides containing a PXXDY, instead of the XPXXP, consensus sequence (Mongiovi et al. 1999). This binding specificity is conserved among the three related genes (Mongiovi et al. 1999). Solution by NMR of the Eps8L1 SH3 domain in complex with the PPVPNPDYEPIR peptide from the CD3epsilon cytoplasmic tail further highlighted the specific molecular requirements at the basis of this unusual mode of binding (Aitio et al. 2008). Indeed, the polyproline peptide binds Eps8L1 SH3 in a class II orientation, but neither adopts a polyproline II helical conformation nor engages the first proline-binding pocket of the SH3 ligand-binding interface. Critical conserved residues in the SH3 domain of Eps8L family members, that are unique with respect to the other SH3 domains, account for this specificity. Most notably, an Eps8L-conserved isoleucine (I531 in EpsL1) is present instead of Y or F in the hydrophobic pocket rendering it nonoptimal for binding to conventional PxxP peptides. Furthermore, specific electrostatic and hydrogen bonding between the D and Y residue of the ligand peptide with positive charge of R and carboxyl group of E in the SH3 domain of Eps8L1, respectively, account for the critical role of the PxxDY motif tyrosine residue in binding to Eps8 family SH3 (Aitio et al. 2008). Thus, the SH3 of Eps8 represents the prototype of a unique family of SH3 modules, which do not bind to canonical XPXXP-containing peptides, and contract specific and distinct interactions in the signaling network. This contention is further supported by the isolation of two physiological interactors of the SH3 domain of Eps8, Abi1 (Abelson Binding Interactor 1) and RN-tre (a GAP for the endocytic Rab5) which display the PXXDY motif and, as it will be discussed later on, establish a novel network involved in the regulation of RTK-dependent signaling.

Recently, the human lanthionine synthetase C-like protein 1 (LanCL1), a eukaryotic homologue to prokaryotic lanthionine cyclases, has been identified as a novel interactor of the SH3 domain of Eps8 (Zhang et al. 2009). Lanthionine cyclase in prokaryotic cells is a peptide-modifier enzyme frequently required to generate antibiotic peptides. However, the function of LanCL1 in mammalian cells remains ill defined. Remarkably, the interaction with Eps8 appears to be required for LanCL1-mediated neuritogenesis (Zhang et al. 2009). Furthermore, the biochemical basis of the association is unusual since it is not mediated by a linear PXXDY motif of LanCL1, but presumably requires an extensive molecular surface interaction from both of these molecules, and is regulated by redox signaling, thus establishing a novel paradigm in the binding and regulation of SH3-mediated protein:protein interactions (Zhang et al. 2009).

Multiple Functions of Eps8 in the Small GTPase Pathways Control Actin Dynamics–Based Processes

Eps8 has emerged as a key regulator in controlling actin cytoskeleton remodeling. The structural organization of Eps8 in multiple protein–protein interaction domains suggests that it acts both by participating in signaling cascades downstream RTKs receptors, through its interaction with Abi1, Sos-1, and RN-tre, and by directly controlling the actin cytoskeleton dynamics and structural organization through its capping and a bundling activity. It is noteworthy that virtually all these functions are mediated by Eps8 C-terminal domain (aa 648–821), where both the Sos-1 and the actin-binding surfaces are located.

Eps8 in the Rho GTPase signaling cascade. Eps8 as a molecular adaptor participates in the RTKs signaling cascade leading to Rac-dependent actin cytoskeleton remodeling. At the molecular level, Eps8 was shown to enter into a multimolecular complex with Abi1, p85, and Sos-1 (for a review of this pathway see Di Fiore and Scita (2002)). Abi1 is the non-catalytic subunit of PI3K, and Sos-1 is the dual GEF for Ras and Rac (Fig. 2). Within this complex, Abi1 acts as a scaffold facilitating the interaction between Eps8, p85, and Sos1 (Di Fiore and Scita 2002). The assembly of this signaling unit is critical to unmask the otherwise silent Rac-specific GEF activity of Sos-1, through mechanisms that remain to be fully elucidated, mediating actin cytoskeleton remodeling in response to growth factors stimulation (Di Fiore and Scita 2002). Notably, the C-terminal region of Eps8 mediates in vitro a direct, low-affinity interaction with Sos-1, sufficient to activate the Rac-specific GEF of the latter protein in cells, suggesting that this region acts as a bona fide effector region within Eps8, mediating all the functions so far ascribed to this protein both as signal adaptors and actin regulatory binding interactors (Di Fiore and Scita 2002).
Eps8 (Epidermal Growth Factor Receptor Pathway Substrate 8), Fig. 2

Eps8-based complexes in the midst of GTPases signaling controlling actin dynamics–based processes. The presence of multiple protein–protein interaction domains enables Eps8 to form diverse multimolecular complexes that act on various small GTPases-dependent pathways. (1) Through its unique SH3 domain, Eps8 can bind to Abi1, which forms the scaffold for a larger complex together with Eps8, Sos-1, and p85 the regulatory subunit of phosphatidyl-inositol-3- kinase (PI3K) mediating the propagation of signaling leading to Rac activation and Rac-dependent actin remodeling from activated PDGFR either directly or through the Grb2-Sos-1-Ras pathways (Di Fiore and Scita 2002). (2) The SH3 domain of Eps8 mediates also the interaction with RN-tre, a GAP for the small G protein Rab5 (Lanzetti et al. 2000), which, in turn, controls RTK internalization and trafficking, thus regulating the duration and cellular localization of RTK signaling, resulting in the formation of specialized actin-rich membrane protrusion (not shown) (Lanzetti et al. 2004). (3) Through its C-terminal actin-binding region, Eps8, which is localized at the very leading edge of actin-rich lamellipodial protrusions as well as in rocketing tails propelling endosomal vesicles and intracellular pathogens (Disanza et al. 2004), binds and caps actin filaments thereby controlling their dynamic and architectural organization. Remarkably, the capping activity of full-length Eps8 is inhibited and can be induced, at least in vitro, through binding to Abi1, which is also localized at the very leading edges of lamellipodia.(4) Eps8 can also cross-link actin filament when it is bound to IRSp53 (Hertzog et al. 2010; Disanza et al. 2004). IRSp53 dimerizes, thus enabling the formation of Eps8 clusters that may act to enhance the cross-linking activity of the complex, which can be recruited by activated Cdc42 to plasma membrane sites for filopodia initiation resulting from the combined membrane deformation ability of IRSp53 and actin-binding property of Eps8. It is of note that the actin-related activities of Eps8 are conserved throughout evolution and have been shown to mediate the morphogenesis of intestinal microvilli in nematodes (Hertzog et al. 2010) and mice (Tocchetti et al. 2010) as well as the development of stereocilia (not shown). Both microvilli and stereocilia are formed by parallel array of bundled filaments whose elongation is tightly controlled by a variety of actin regulatory proteins

The above scenario became more complex with the identification of another Eps8 interactor, IRSp53, which primarily acts as downstream effector of Cdc42 (Fig. 2). IRSp53 was originally discovered as a substrate of the insulin receptor. Subsequently, IRSp53 was shown to participate in the formation of various Cdc42- and Rac-dependent, actin-based protrusions, including filopodia and lamellipodia, neurites and dendritic spines, through a mechanism that remains to be fully elucidated. The role of IRSp53 in filopodia is more clearly established. Filopodia are actin-rich, finger-like structures that protrude from the cell membrane of a variety of cell types and play important roles in cell migration, neurite outgrowth, and wound healing (Scita et al. 2008). They are characterized by a small number of long and parallel actin filaments that deform the cell membrane, giving rise to protrusions. Within this context, IRSp53 is thought to connect actin regulatory complexes with the extending membranes. In keeping with this notion, IRSp53 possesses an N-terminal helical domain, which belongs to the BAR (Bin-amphipysin-Rvs) family of domains and folds into zeppelin-shaped dimers. BAR domains themselves are banana-shaped structures that induce curvature in membranes via their concave face promoting membrane invagination and the generation of tubules and vesicles important in endocytic and intracellular trafficking processes. In contrast to these standard BAR domains, the I(inverted)-BAR domain of IRSp53 induces negative curvature in membranes, thus inducing protrusions rather than invagination and contributing to the formation of filopodia (Scita et al. 2008). In this latter process, IRSp53 is thought to act as an effector of Cdc42, which in its activated GTP-bound form can bind to a Cdc42-specific CRIB-like sequence of IRSp53. This interaction leads to relocalization of IRSp53 to sites where filopodia are initiated. At this location, IRSp53 can physically link the protruding membrane to the underlying actin filaments by binding, through its SH3 domain, a number of regulators of actin dynamics and architectural organization, including Dia1, Mena, WAVE2, and Eps8 (Scita et al. 2008). It is noteworthy that, among these interactors, Eps8 is the only protein that can form a stable complex with IRSp53 both in vitro and in vivo (Hertzog et al. 2010). The assembly of a Eps8::IRSp53 complex endows this unit with actin bundling activity in vitro, that may initiate, in vivo, the formation of actin bundles, which would subsequently become tightly cross-linked by other more efficient bundlers, such as fascin. Thus, Eps8 together with IRSp53 may act at the interface of membranes and the actin cytoskeleton, initiating in a Cdc42-dependent fashion the bundling of filaments, which may subsequently elongate and become tightly bundled supporting filopodia extension.

Eps8 and Rab5 endocytic networks. Eps8 involvement in endocytic processes derives from its ability to bind yet another SH3 interactor, RN-tre. RN-tre is a GAP specific for Rab family proteins, including Rab5 (Lanzetti et al. 2000), indicating its involvement in trafficking processes. Most notably, RN-tre was originally demonstrated to function both in vitro and in vivo on Rab5 (Lanzetti et al. 2000), which is in turn essential for regulating multiple steps of internalization of various membrane receptors, including early endocytic events and early endosomal formation, maturation, and trafficking. In particularly, Rab5 is essential for EGF receptor endocytosis upon ligand binding. Interaction between Eps8 and RN-tre has been shown to be required for the activation of the Rab5-GAP specific activity of the latter protein. This leads to the inhibition of the small GTPase accompanied by increased EGFR at the cell surface due to impaired receptor internalization and increased signaling (Lanzetti et al. 2000) (Fig. 2).

Thus, Eps8 may regulate Rac activity either directly through assembly of a plasma-membrane-localized signaling complex that promotes optimal transmission of motogenic RTK signaling, and indirectly through endocytsosis that confers a spatial dimension to signaling that is ultimately required for the regulation of polarized function, first and foremost actin dynamics–based directional migration.

Eps8 and Actin Dynamics

Eps8 is able to directly control actin cytoskeleton remodeling through inherent capping and bundling activities. Eps8 can accelerate actin treadmilling by blocking polymerization at filaments ends, acting as a capping protein. Eps8 can also regulate the architectural organization of the actin cytoskeleton by cross-linking actin filaments, in this case acting as a bundling protein. These properties in conjunction with in vivo localization data suggest that Eps8 critically participates both in the regulation of cell motility and in the architectural organization of specific actin-based structures, including cell protrusions at the leading edge of migrating cells, microvilli of intestinal cells, and stereocilia. Both capping and bundling activities are mediated and retained by the C-terminal region of the protein (amino acids 648–821) but the full-length protein is tightly regulated by protein–protein interactions. The bundling activity can be isolated and attributed to the C-terminal four helices of the region (Hertzog et al. 2010). The capping activity can also be isolated and can be attributed to the first helix of the region that is tethered to the C-terminal four helices by a flexible linker (Hertzog et al. 2010). Thus, the C-terminal effector region of Eps8 contains two structurally distinct actin-binding modules, one responsible for capping and one responsible for bundling (Fig. 3a).
Eps8 (Epidermal Growth Factor Receptor Pathway Substrate 8), Fig. 3

Eps8 C-terminal actin-binding region. (a) Structure of the conserved C-terminal region of Eps8 shown in cartoon representation. The region contains two distinct actin-binding sites responsible for bundling (blue) and capping (green) respectively. These regions are connected by a flexible linker (red). (b) Low resolution representation of an actin filament barbed end (white surface) with the Eps8 C-terminal domain bound and the capping domain occupying the hydrophobic groove of the barbed end actin subunit (arrowhead). This interaction prevents the addition of new filament subunits. (c) Low resolution representation of an actin filament (white surface) with the Eps8 C-terminal domain bound to the side (arrowhead), contacting three successive actin subunits. In conjunction with dimerization, this interaction is responsible for the bundling activity of Eps8. The capping domain does not occupy the hydrophobic groove of actin in the configuration

Eps8 capping activity. In vitro, the conserved C-terminal region of Eps8, in substoichiometric amounts with respect to actin, inhibits barbed end growth by up to 90% and enables propulsion of N-WASP-coated beads in the absence of gelsolin or other cappers (Disanza et al. 2004). In vivo, Eps8 localizes in rocketing tails propelling phosphatidylinositol-4,5-phosphate-(PI45P) enriched vesicles or intracellular pathogens such as Listeria monocytogenes or Shighella flexneri (Disanza et al. 2004). Structurally, the Eps8 capping activity is mediated by binding of the amphipathic first helix of the Eps8 C-terminal region within the hydrophobic pocket at the barbed ends of actin (Hertzog et al. 2010). Occupation of this site on actin filaments by the helix blocks further addition of actin monomers (Fig. 3b). The capping activity of full-length Eps8 is regulated by interactions with Abi1 (Disanza et al. 2004). While the exact mechanism of this regulation remains obscure, it is likely that the binding of Abi1 induces a conformational change in full-length Eps8 that exposes the amphipathic first helix within the C-terminal region and makes it available for interactions with the filament barbed ends.

Eps8 bundling activity. The C-terminal four-helix region of Eps8 binds to the sides of actin filaments as a compact helix bundle (Hertzog et al. 2010) and possesses bundling activity in vitro when dimerization domains such as GST are attached to the construct (Disanza et al. 2006). The four-helix bundle interacts with three adjacent subunits along the short-pitch helix of the actin filament (Fig. 3c). Full-length Eps8 can form a constitutive, stable complex with IRSp53 that bundles actin filaments (Disanza et al. 2006). While Eps8 expression alone has no effect on filopodia formation in vivo, the concomitant expression of the two proteins leads to a dramatic increase in filopodia number and length (Disanza et al. 2006). The Eps8:IRSp53 interaction mediates the bundling activity of the side-bound full-length Eps8 most likely through dimerization of IRSp53, with a possible contribution from the inherent, weak actin-binding activity of the N-terminal IMD domain of IRSp53, which is auto-inhibited in full-length IRSp53 but may be released by Eps8 binding.

Physiological Roles of Eps8: From Cells to Multicellular Organisms

The dissection of the biochemical and signaling properties of Eps8 together with its ability to participate in various, distinct macromolecular complexes has provided a rational framework to account for a number of apparently unrelated and diverse cell biological phenotypes associated with this protein. Thus, for example, the presence of Eps8 in actin dynamics–based motile structures (Hertzog et al. 2010; Disanza et al. 2004) reflects the ability of this protein to cap the barbed ends of actin filament. Conversely, Eps8 localization to microspikes and filopodia is accounted for, at least in part, by the ability to form a complex with IRSp53 at the plasma membrane where Eps8 may both “protect” plus ends from stronger cappers, such as CP, while promoting the convergent association and bundling of filament ends, a presumably limiting step in filopodia initiation (Hertzog et al. 2010). This context allows to begin rationalizing how the actin-related activities of Eps8, controlled by distinct molecular partners in different cellular contexts, may account for seemingly paradoxical and opposite effects on filopodia whereby genetic removal of Eps8 reduced filopodia in HeLa cells (Disanza et al. 2004), but increased them in hippocampal neurons (Menna et al. 2009). It is likely that the dynamic and context-dependent interactions established by Eps8, IRSp53, and Abi1 with actin filaments generate a flexible signaling network that governs the multifunctional activities of its components in the formation of diverse actin-based structures: a notion that is being supported by computational modeling of this network (GS personal communication). Whatever the case, Eps8 capping and bundling activities are emerging as critical in diverse phenotypes both at the cellular and organism levels, revealing the involvement of actin dynamics in a variety of diverse and frequently unexpected physiological processes. Specific impairment of bundling activity, for example, of the only EPS-8 family member in nematodes leads to a severely altered intestinal morphogenesis that correlates with a disruption of the stereotypical orderly organization of actin-bundle-rich microvilli (Croce et al. 2004). This specific function appears evolutionary conserved since also 8 null mice display shorter intestinal microvilli (Tocchetti et al. 2010). Notably, this alteration is accompanied by a defective intestinal fat absorption that results in calorie-restricted metabolism and increased life span, providing genetic evidence of actin dynamics as a novel variable in the determination of longevity (Tocchetti et al. 2010). It is also noteworthy that the three other Eps8L family members do not compensate for the lack of Eps8 in the intestine suggesting that differential tissues expression and cellular localization (in this case to microvilli) contribute to functional specificity within the Eps8L family. A similar situation may also account for the recent discovery that eps8 null mice are deaf. Hearing impairment is a consequence of disrupted stereocilia morphology, where Eps8 localizes together with the Myosin XVa and the adaptor protein whirlin (Manor et al. 2011). While the specific contribution of Eps8 actin-related activities to proper stereocilia elongation has not yet been established, it is reasonable to assume that, in analogy to intestinal microvilli, which share a similar architectural organization in parallel actin bundles with stereocilia, the actin cross-linking property of Eps8 is responsible for the phenotype. However, a role of Eps8 capping in fine-tuning filament elongation at the basis of the precise staircase pattern of stereocilia cannot be ruled out at the present stage.

Finally, a specific functional role of Eps8 with unexpected physiopathological consequences has emerged when the function of this protein was analyzed in the central nervous system. Eps8 is uniquely expressed in the hippocampus as well as the cerebellum (Offenhauser et al. 2006). In cerebellar neurons, Eps8 is significantly more abundant. Consistently, genetic removal of 8 alters actin dynamics, particularly at postsynaptic sites, where Eps8 is part of the N-methyl-D-aspartate (NMDA) receptor complex. Moreover, in Eps8 null mice, NMDA receptor currents and their sensitivity to inhibition by ethanol are abnormal. In addition, 8 null cerebellar neurons are resistant to the actin-remodeling activities of NMDA and ethanol (Offenhauser et al. 2006). These alterations presumably explain why 8 null mice are resistant to acute intoxicating effects of ethanol and show increased ethanol consumption (Offenhauser et al. 2006), further supporting the notion that proper regulation of the actin cytoskeleton by Eps8 is a key determinant of cellular and behavioral responses to ethanol.

In cortical neurons, the capping activity of Eps8 is, instead, required downstream of Brain Derived Neurotrophic Factor (BDNF) for the regulation of axonal filopodia: a process with crucial impacts on neuronal development and synapse formation (Menna et al. 2009). In this cell system, Eps8 capping activity is negatively regulated by BDNF-mediated, MAPK-dependent Eps8 phosphorylation, which reduces the affinity of the Eps8:Abi1 complex for barbed ends, thus promoting the elongation of uncapped actin filament and the generation of axonal filopodia (Menna et al. 2009). Notably, filopodia and short axonal branches are frequently extended and retracted to form boutons onto postsynaptic structures, or to originate en passant boutons, which are added and eliminated along the axonal shaft: all processes that characterize synaptic plasticity. Consistent with this possibility, eps8 null brains display a significantly higher number of presynaptic boutons (Scita 2008) that might originate from the increased axonal filopodia during neuronal development, thus suggesting a role of the protein in controlling synapse formation and presynaptic plasticity phenomena in the developing brain and, possibly, in the adult brain.

Eps8 and Cancer

While the involvement of Eps8 in cytoskeletal reorganization and its molecular modes of action are gradually being elucidated, experimental evidence has also accumulated supporting the implication of Eps8 in tumorigenesis. Since its discovery, Eps8 was shown to enhance mitogenesis and, in some circumstances, also cell transformation in response to growth factor treatment (reviewed in Di Fiore and Scita (2002)). Furthermore, constitutive tyrosine phosphorylation of Eps8, frequently accompanied by the elevated phosphotyrosine levels of a number of RTK signaling molecules, was detected in a large variety of tumor cells (reviewed in Di Fiore and Scita (2002)). These initial observations supported the notion that Eps8, presumably through its signal-transducing adaptor functions, may be relevant in promoting mitogenic signaling particularly in those tumors where RTKs signaling is elevated. This latter contention was subsequently extended by the finding that constitutive tyrosine phosphorylation of Eps8 was also frequently coupled to elevated protein expression, such as in cells transformed by non-receptor tyrosine kinases, v-SRC (reviewed in Di Fiore and Scita (2002)), and in a variety of human tumors, including colorectal (Maa et al. 2001), cervical (Chen et al. 2008; Wang et al. 2009), pituitary (Xu et al. 2008), pancreatic (Welsch et al. 2007), and oral cancers (Yap et al. 2009). Consistent with this, the overexpression of Eps8 confers the ability of fibroblasts to form foci in culture dishes and to grow tumors in mice (Maa et al. 2001). These findings confirm the initial observations and further corroborate the oncogenic potential of Eps8. In agreement with this notion, Eps8 attenuation retarded cellular growth in v-SRC transformed cells and human cancer cells (Maa et al. 2007). Frequently, Eps8 overexpression in tumors correlates with increased MAPK and/or AKT signaling. While the mechanisms and pathways through which Eps8 mediates these effects remain ill defined, it is reasonable to propose that it is the ability of Eps8 to form diverse multimolecular signaling complexes acting in the Ras-PI3K-Rac (Di Fiore and Scita 2002) or RTK-p66 shc (Bashir et al. 2010) transduction cascades that accounts for its role in promoting mitogenesis or increased resistance to apoptosis. It should be noted, however, that an additional, much less explored possibility, is that deregulated levels of Eps8 may disturb the endocytosis or intracellular trafficking of RTKs, in turn, affecting the duration and spatial localization of RTK-dependent signaling, ultimately resulting in enhanced tumorigenesis. In this latter respect, it is noteworthy that in C. elegans elevated and basolaterally localized EPS-8 in vulval precursor cells (VPCs) blocks EGFR internalization, leading to enhanced signaling. Conversely, low levels of EPS-8 in the neighboring secondary VPCs result in the rapid degradation of the EGFR, allowing these cells to adopt a secondary cell fate (Stetak et al. 2006). Thus, Eps8 regulation of EGFR trafficking may be an evolutionary conserved mechanism to control signaling in pattern formation during normal organogenesis or to promote enhanced mitogenic signaling, when its levels are deregulated, such as in tumor development.

Finally, it is important to point out that at least in specific tumor types, EPS8 expression levels correlate with an increase in their capacity to migrate and invade surrounding tissues. This is the case for head and neck squamous cell carcinomas (HNSCCs) (Yap et al. 2009) and for a subset of colon carcinoma (Maa et al. 2007). In HNSCC, Eps8 is upregulated in 32% of primary tumors compared with normal oral mucosa, and its expression significantly correlated with lymph node metastasis, suggesting a disease-promoting effect. Consistent with this notion, the ablation of Eps8 from high-expressing HNSCC derived from tumor patients impaired cell migration and invasion both in in vitro organotypic and in mouse model systems (Wang et al. 2009; Yap et al. 2009). Some of these effects of Eps8 were attributed to elevated Rac activity (Yap et al. 2009; Bashir et al. 2010) or to indirect regulation of transcription factors, such as FOXM1, in turn controlling signaling proteins important for cell cycle progression and cell migration (Wang et al. 2010). However, it is tempting to speculate that the actin regulatory activities of Eps8 may directly contribute to the acquisition of highly migratory and invasive properties of advanced stage of squamous cell carcinoma. The possibility to dissect at the molecular levels the diverse actin regulatory functions of Eps8 will be instrumental in the years to come to test this hypothesis.

Summary

Eps8, originally identified as a substrate of the EGFR, is the founding member of a family of proteins that includes three additional members in mammals. They all display a modular organization into various protein–protein interaction domains that suggest the involvement of this protein family in the transduction of receptor tyrosine kinases (RTKs) signals. Consistently, Eps8 can participate in various, distinct macromolecular complexes that either mediate the activation of small GTPases Rac and Rac-dependent actin remodeling or regulate RTK signal duration by controlling receptor internalization and trafficking. Remarkably, Eps8 is able to bind to actin filaments and to control actin dynamics, through its ability to associate with and block the plus growing ends of filaments. Eps8 can also regulate the architectural organization of actin meshworks and bundles by cross-linking adjacent filaments. These two properties can be dissected in vitro and are differentially regulated by Eps8 interactors: Abi1 activates actin capping, and IRSp53 promotes actin bundling. Thus, Eps8 by participating in multiple and distinct molecular complexes plays roles in a variety of actin-based processes, including motile cellular protrusions such as lamellipodia, rocketing tail, and filopodia, as well as sensory structures such as microvilli and stereocilia. The cellular roles of Eps8 are mirrored at the organism level by the requirement of Eps8 in diverse and specific physiological actin-based processes. Thus, Eps8 null mice display a range of actin-dependent phenotypes ranging from defective extension of neuronal filopodia and synaptogenesis that can result in deregulation of behavioral responses to ethanol or altered morphogenesis of intestinal microvilli and stereocilia that can cause fat malabsorption and deafness, respectively. Finally, emerging evidence implicates Eps8 in tumourigenesis whereby elevated levels of this protein either favors cell proliferation or enhances cell migration and invasion of various cancer cells. Thus, Eps8 represent a unique actin regulatory protein endowed with signal-transducing properties accounting for its pleiotropic functional roles.
Eps8 (Epidermal Growth Factor Receptor Pathway Substrate 8), Table 1

Accession numbers for the protein sequences used in Fig. 1

Name in alignment

GenBank accession

Description

Organism

Hs_EPS8

Q12929

Epidermal growth factor receptor kinase substrate Eps8

Homo sapiens

Hs_EPS8L1

AAL76117.1

Epidermal growth factor receptor pathway substrate 8 related protein 1

Homo sapiens

Hs_EPS8L2

AAL76118.1

Epidermal growth factor receptor pathway substrate 8 related protein 2

Homo sapiens

Hs_EPS8L3

AAL76119.1

Epidermal growth factor receptor pathway substrate 8 related protein 3

Homo sapiens

Mm_EPS8

Q08509

Epidermal growth factor receptor kinase substrate eps8

Mus musculus

Mm_EPS8L1

AAL76120.1

Epidermal growth factor receptor pathway substrate 8 related protein 1

Mus musculus

Mm_EPS8L2

AAH05492.1

Similar to hypothetical protein FLJ21935

Mus musculus

Mm_EPS8L3

AAL76121.1

Epidermal growth factor receptor pathway substrate 8 related protein 3

Mus musculus

Rn_EPS8

Predicted

ENSEMBL gene prediction ENSRNOT00000009328.3

Rattus norvegicus

Rn_EPS8L1

XP_341783.1

Similar to epidermal growth factor receptor pathway substrate 8-like protein 1

Rattus norvegicus

Rn_EPS8L2

Predicted

ENSEMBL gene prediction ENSRNOT00000024725.3

Rattus norvegicus

Rn_EPS8L3

XP_215677.4

Similar to epidermal growth factor receptor pathway substrate 8_Elike protein 3

Rattus norvegicus

Gg_EPS8

XP_416405.2

Similar to Epidermal growth factor receptor pathway substrate 8

Gallus gallus

Gg_EPS8L2

XP_001232098.1

Hypothetical protein

Gallus gallus

Xl_EPS8

AAH68768.1

MGC81285 protein

Xenopus laevis

Xt_EPS8

NP_001072508.1

Epidermal growth factor receptor pathway substrate 8

Xenopus tropicalis

Xt_EPS8L2

Predicted

Genewise gene prediction from xenopus genome scaff0296old:324,626–357,189

Xenopus tropicalis

Dr_EPS8

NP_956536.1

Similar to epidermal growth factor receptor pathway substrate 8

Danio rerio

Dr_EPS8L1

Predicted

N-SCAN gene predictions (chr2.1002.1)

Danio rerio

Dr_EPS8L3

XP_691442.2

Hypothetical protein

Danio rerio

Tn_EPS8L1

Predicted

Geneid gene prediction (chr2_362.1)

Tetraodon nigroviridis

Tn_EPS8L2

Predicted

Geneid gene prediction (chr13_663.1)

Tetraodon nigroviridis

Tn_EPS8L3

CAG06824.1

Unnamed protein product

Tetraodon nigroviridis

Ci_EPS8

Predicted

Genewise gene prediction from ciona genome chr05q:2,271,763–2,296,713

Ciona intestinalis

Dm_EPS8

AAL13963.1

Lp01469p

Drosophila melanogaster

Dp_EPS8

XP_001359077.1

GA21402-PA

Drosophila pseudoobscura

Cb_EPS8

CAE68196.1

Hypothetical protein CBG13858

Caenorhabditis briggsae

Ce_EPS8

NP_001041047.1

EPS (human endocytosis) related family member,eps-8

Caenorhabditis elegans

Notes

Acknowledgments

This study was supported by the NIH Cell Migration Consortium grant U54 GM064346 from the National Institute of General Medical Sciences (NIGMS) to DH and NV; from the Associazione Italiana per la Ricerca sul Cancro (AIRC), the Italian Ministries of Education-University-Research (MIUR-PRIN) and of Health, AICR (International Association For Cancer Research – AICR), and from IFOM Foundation to GS; FM is supported by a fellowship from Fondazione Italiana Ricerca sul Cancro (FIRC).

References

  1. Aitio O, Hellman M, Kesti T, Kleino I, Samuilova O, Paakkonen K, et al. Structural basis of PxxDY motif recognition in SH3 binding. J Mol Biol. 2008;382(1):167–78.PubMedCrossRefGoogle Scholar
  2. Bashir M, Kirmani D, Bhat HF, Baba RA, Hamza R, Naqash S, et al. P66shc and its downstream Eps8 and Rac1 proteins are upregulated in esophageal cancers. Cell Commun Signal. 2010;8:13.PubMedPubMedCentralCrossRefGoogle Scholar
  3. Chen YJ, Shen MR, Maa MC, Leu TH. Eps8 decreases chemosensitivity and affects survival of cervical cancer patients. Mol Cancer Ther. 2008;7(6):1376–85.PubMedCrossRefGoogle Scholar
  4. Croce A, Cassata G, Disanza A, Gagliani MC, Tacchetti C, Malabarba MG, et al. A novel actin barbed-end-capping activity in EPS-8 regulates apical morphogenesis in intestinal cells of Caenorhabditis elegans. Nat Cell Biol. 2004;6(12):1173–9.PubMedCrossRefGoogle Scholar
  5. Di Fiore PP, Scita G. Eps8 in the midst of GTPases. Int J Biochem Cell Biol. 2002;34(10):1178–83.PubMedCrossRefGoogle Scholar
  6. Disanza A, Carlier MF, Stradal TE, Didry D, Frittoli E, Confalonieri S, et al. Eps8 controls actin-based motility by capping the barbed ends of actin filaments. Nat Cell Biol. 2004;6(12):1180–8.PubMedCrossRefGoogle Scholar
  7. Disanza A, Mantoani S, Hertzog M, Gerboth S, Frittoli E, Steffen A, et al. Regulation of cell shape by Cdc42 is mediated by the synergic actin-bundling activity of the Eps8-IRSp53 complex. Nat Cell Biol. 2006;8(12):1337–47.PubMedCrossRefGoogle Scholar
  8. Fazioli F, Minichiello L, Matoska V, Castagnino P, Miki T, Wong WT, et al. Eps8, a substrate for the epidermal growth factor receptor kinase, enhances EGF-dependent mitogenic signals. EMBO J. 1993;12(10):3799–808.PubMedPubMedCentralGoogle Scholar
  9. Forman-Kay JD, Pawson T. Diversity in protein recognition by PTB domains. Curr Opin Struct Biol. 1999;9(6):690–5.PubMedCrossRefGoogle Scholar
  10. Hertzog M, Milanesi F, Hazelwood L, Disanza A, Liu H, Perlade E, et al. Molecular basis for the dual function of Eps8 on actin dynamics: bundling and capping. PLoS Biol. 2010;8(6):e1000387.PubMedPubMedCentralCrossRefGoogle Scholar
  11. Kishan KV, Scita G, Wong WT, Di Fiore PP, Newcomer ME. The SH3 domain of Eps8 exists as a novel intertwined dimer. Nat Struct Biol. 1997;4(9):739–43.PubMedCrossRefGoogle Scholar
  12. Kishan KV, Newcomer ME, Rhodes TH, Guilliot SD. Effect of pH and salt bridges on structural assembly: molecular structures of the monomer and intertwined dimer of the Eps8 SH3 domain. Protein Sci. 2001;10(5):1046–55.PubMedPubMedCentralCrossRefGoogle Scholar
  13. Lanzetti L, Rybin V, Malabarba MG, Christoforidis S, Scita G, Zerial M, et al. The Eps8 protein coordinates EGF receptor signalling through Rac and trafficking through Rab5. Nature. 2000;408(6810):374–7.PubMedCrossRefGoogle Scholar
  14. Lanzetti L, Palamidessi A, Areces L, Scita G, Di Fiore PP. Rab5 is a signalling GTPase involved in actin remodelling by receptor tyrosine kinases. Nature. 2004;429(6989):309–14.PubMedCrossRefGoogle Scholar
  15. Maa MC, Hsieh CY, Leu TH. Overexpression of p97Eps8 leads to cellular transformation: implication of pleckstrin homology domain in p97Eps8-mediated ERK activation. Oncogene. 2001;20(1):106–12.PubMedCrossRefGoogle Scholar
  16. Maa MC, Lee JC, Chen YJ, Chen YJ, Lee YC, Wang ST, et al. Eps8 facilitates cellular growth and motility of colon cancer cells by increasing the expression and activity of focal adhesion kinase. J Biol Chem. 2007;282(27):19399–409.PubMedCrossRefGoogle Scholar
  17. Manor U, Disanza A, Grati M, DeAndrade L, Lin H, Di Fiore P, et al. Regulation of stereocilia length by myosin XVa and whirlin depends on the actin regulatory protein Eps8. Curr Biol. 2011;21(2):167–72.PubMedPubMedCentralCrossRefGoogle Scholar
  18. Menna E, Disanza A, Cagnoli C, Schenk U, Gelsomino G, Frittoli E, et al. Eps8 regulates axonal filopodia in hippocampal neurons in response to brain-derived neurotrophic factor (BDNF). PLoS Biol. 2009;7(6):e1000138.PubMedPubMedCentralCrossRefGoogle Scholar
  19. Mongiovi AM, Romano PR, Panni S, Mendoza M, Wong WT, Musacchio A, et al. A novel peptide-SH3 interaction. EMBO J. 1999;18(19):5300–9.PubMedPubMedCentralCrossRefGoogle Scholar
  20. Offenhauser N, Borgonovo A, Disanza A, Romano P, Ponzanelli I, Iannolo G, et al. The eps8 family of proteins links growth factor stimulation to actin reorganization generating functional redundancy in the Ras/Rac pathway. Mol Biol Cell. 2004;15(1):91–8.PubMedPubMedCentralCrossRefGoogle Scholar
  21. Offenhauser N, Castelletti D, Mapelli L, Soppo BE, Regondi MC, Rossi P, et al. Increased ethanol resistance and consumption in Eps8 knockout mice correlates with altered actin dynamics. Cell. 2006;127(1):213–26.PubMedCrossRefGoogle Scholar
  22. Scita G, Confalonieri S, Lappalainen P, Suetsugu S. IRSp53: crossing the road of membrane and actin dynamics in the formation of membrane protrusions. Trends Cell Biol. 2008;18(2):52–60.PubMedCrossRefGoogle Scholar
  23. Slupsky CM, Gentile LN, Donaldson LW, Mackereth CD, Seidel JJ, Graves BJ, et al. Structure of the Ets-1 pointed domain and mitogen-activated protein kinase phosphorylation site. Proc Natl Acad Sci USA. 1998;95(21):12129–34.PubMedPubMedCentralCrossRefGoogle Scholar
  24. Smith MJ, Hardy WR, Murphy JM, Jones N, Pawson T. Screening for PTB domain binding partners and ligand specificity using proteome-derived NPXY peptide arrays. Mol Cell Biol. 2006;26(22):8461–74.PubMedPubMedCentralCrossRefGoogle Scholar
  25. Stetak A, Hoier EF, Croce A, Cassata G, Di Fiore PP, Hajnal A. Cell fate-specific regulation of EGF receptor trafficking during Caenorhabditis elegans vulval development. EMBO J. 2006;25(11):2347–57.PubMedPubMedCentralCrossRefGoogle Scholar
  26. Tocchetti A, Soppo CB, Zani F, Bianchi F, Gagliani MC, Pozzi B, et al. Loss of the actin remodeler Eps8 causes intestinal defects and improved metabolic status in mice. PLoS One. 2010;5(3):e9468.PubMedPubMedCentralCrossRefGoogle Scholar
  27. Wang H, Patel V, Miyazaki H, Gutkind JS, Yeudall WA. Role for EPS8 in squamous carcinogenesis. Carcinogenesis. 2009;30(1):165–74.PubMedCrossRefGoogle Scholar
  28. Wang H, Teh MT, Ji Y, Patel V, Firouzabadian S, Patel AA, et al. EPS8 upregulates FOXM1 expression, enhancing cell growth and motility. Carcinogenesis. 2010;31(6):1132–41.PubMedPubMedCentralCrossRefGoogle Scholar
  29. Welsch T, Endlich K, Giese T, Buchler MW, Schmidt J. Eps8 is increased in pancreatic cancer and required for dynamic actin-based cell protrusions and intercellular cytoskeletal organization. Cancer Lett. 2007;255(2):205–18.PubMedCrossRefGoogle Scholar
  30. Xu M, Shorts-Cary L, Knox AJ, Kleinsmidt-Demasters B, Lillehei K, Wierman ME. Epidermal growth factor receptor pathway substrate 8 (Eps8) is overexpressed in human pituitary tumors: role in proliferation and survival. Endocrinology. 2008. doi:10.1210/en.2008-1265.Google Scholar
  31. Yap LF, Jenei V, Robinson CM, Moutasim K, Benn TM, Threadgold SP, et al. Upregulation of Eps8 in oral squamous cell carcinoma promotes cell migration and invasion through integrin-dependent Rac1 activation. Oncogene. 2009;28(27):2524–34.PubMedCrossRefGoogle Scholar
  32. Zhang W, Wang L, Liu Y, Xu J, Zhu G, Cang H, et al. Structure of human lanthionine synthetase C-like protein 1 and its interaction with Eps8 and glutathione. Genes Dev. 2009;23(12):1387–92.PubMedPubMedCentralCrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  • Francesca Milanesi
    • 1
  • Niels Volkmann
    • 2
  • Giorgio Scita
    • 1
    • 3
  • Dorit Hanein
    • 2
  1. 1.IFOMMilanItaly
  2. 2.Bioinformatics and Systems Biology ProgramSanford-Burnham Medical Research InstituteLa JollaUSA
  3. 3.Dipartimento di Medicina, Chirurgia ed OdontoiatriaUniversita’ degli Studi di MilanoMilanItaly