FGD1 as a central regulator of extracellular matrix remodelling - lessons from faciogenital dysplasia

Hypothesis

3265

FGD1 as a central regulator of extracellular matrix remodelling – lessons from faciogenital dysplasia Elisabeth Genot1,2,3,4, Thomas Daubon1,2,3, Vincenzo Sorrentino5 and Roberto Buccione6,* 1

Universite´ de Bordeaux, Physiopathologie du Cancer du Foie, U1053, F-33000 Bordeaux, France INSERM, Physiopathologie du Cancer du Foie, U1053, F-33000 Bordeaux, France 3 European Institute of Chemistry and Biology, 2 rue Robert Escarpit, 33607 Pessac, France 4 CHU de Bordeaux, F-33076 Bordeaux, France 5 Department of Neurosciences, Molecular Medicine Section, University of Siena, Siena 53100, Italy 6 Tumour Cell Invasion Laboratory, Consorzio Mario Negri Sud, S. Maria Imbaro, Chieti 66030, Italy 2

*Author for correspondence ([email protected])

Journal of Cell Science

Accepted 27 March 2012 Journal of Cell Science 125, 3265–3270 ß 2012. Published by The Company of Biologists Ltd doi: 10.1242/jcs.093419

Summary Disabling mutations in the FGD1 gene cause faciogenital dysplasia (also known as Aarskog-Scott syndrome), a human X-linked developmental disorder that results in disproportionately short stature, facial, skeletal and urogenital anomalies, and in a number of cases, mild mental retardation. FGD1 encodes the guanine nucleotide exchange factor FGD1, which is specific for the Rho GTPase cell division cycle 42 (CDC42). CDC42 controls cytoskeleton-dependent membrane rearrangements, transcriptional activation, secretory membrane trafficking, G1 transition during the cell cycle and tumorigenic transformation. The cellular mechanisms by which FGD1 mutations lead to the hallmark skeletal deformations of faciogenital dysplasia remain unclear, but the pathology of the disease, as well as some recent discoveries, clearly show that the protein is involved in the regulation of bone development. Two recent studies unveiled new potential functions of FGD1, in particular, its involvement in the regulation of the formation and function of invadopodia and podosomes, which are cellular structures devoted to degradation of the extracellular matrix in tumour and endothelial cells. Here, we discuss the hypothesis that FGD1 might be an important regulator of events controlling extracellular matrix remodelling and possibly cell invasion in physiological and pathological settings. Additionally, we focus on how studying the cell biology of FGD1 might help us to connect the dots that link CDC42 signalling with remodelling of the extracellular matrix (ECM) in physiology and complex diseases, while, at the same time, furthering our understanding of the pathogenesis of faciogenital dysplasia. Key words: Aarskog-Scott syndrome, ECM remodelling, Faciogenital dysplasia, FGD1, Invadopodia, Podosomes

Introduction Mutations in the FGD1 gene, which is localized to the proximal short arm of the X chromosome (Xp11.21, Fig. 1), cause the human X-linked developmental disorder faciogenital dysplasia [FGDY, also known as Aarskog-Scott syndrome (OMIM 305400)] (Pasteris et al., 1994; Scriver et al., 2001). Patients with FGDY present a distinguishing set of craniofacial features, which result in a dysmorphic facial appearance, a disproportionately short stature that is accompanied by skeletal anomalies, urogenital malformations and, in a number of cases, mild mental retardation (Lebel et al., 2002; Orrico et al., 2010; Pasteris et al., 1994; Scriver et al., 2001). The FGD1 gene encodes the FGD1 protein, which is a member of the DBL family of guanine nucleotide exchange factors (GEFs) that regulate the activation of the Rho GTPases (Schmidt and Hall, 2002). A study of the spatiotemporal expression pattern of FGD1 in mouse embryos has provided important clues for the understanding of FGDY pathogenesis (Gorski et al., 2000). In fact, within the developing mouse skeleton, FGD1 is almost exclusively expressed in pre-cartilaginous mesenchymal condensations, the perichondrium and periostium, proliferating chondrocytes and osteoblasts. More precisely, FGD1 is not expressed in the earlier phases of skeletogenesis, but is detected

in the ossifying skeletal components of craniofacial bones, vertebrae, ribs, long bones and phalanges, the skeletal components derived from the neural crest, and the periaxial and lateral mesoderm (Gorski et al., 2000). Therefore, the expression pattern of FGD1 is consistent with most of the skeletal malformations observed in FGDY patients, which suggests that FGD1, at the very least, is an important regulator of bone development. Rho GTPases interact with their effector proteins to drive a large variety of biological processes, including reorganisation of the actin cytoskeleton (thereby affecting changes to cell shape, cell polarity, cell movement and cytokinesis), gene transcription, cell cycle progression, cell adhesion and oncogenic transformation. GEFs catalyze the exchange of GDP for GTP on the GTPases, thereby activating a molecular switch that leads to GTPase activation. By specifically activating the Rho GTPase cell division cycle 42 (CDC42) (Fig. 2), FGD1 controls cytoskeletondependent membrane rearrangements, transcriptional activation, secretory membrane-trafficking, transition through G1 during the cell cycle and tumorigenic transformation (Olson et al., 1996). However, two recent studies have introduced a new twist with regards to FGD1 function by unveiling potential roles for this GEF in the control of extracellular matrix (ECM) remodelling

3266

Journal of Cell Science 125 (14)

Chromosome X

Xp 11.21

N

C

PRD

Journal of Cell Science

1

DH and PH 391

FYVE 713

821

PH 961

Fig. 1. FGD1 domain structure. FGD1 is encoded by a gene that is located on the short arm of the X-chromosome at position Xp11.21 (Pasteris et al., 1994; Scriver et al., 2001). FGD1, like other GEF proteins, contains an Nterminal PRD (amino acids 7-355) that negatively regulates GEF activity and contains two putative SH3-binding domains (amino acids 171–179 and 179– 187). The PRD is followed by a DH and PH domain; the DH domain (amino acids 391–561) lies adjacent to a PH domain (amino acids 590–707), which catalyzes the exchange of bound GDP for GTP on CDC42. The DH and PH domains are followed by a FYVE domain (amino acids 713–813), which constitutes a PtdIns(3)P-binding region and a cysteine-rich zinc-finger, and a second, C-terminal PH domain (amino acids 821–922) (Gorski et al., 2000; Pasteris et al., 1994). Three types of genetic FGD1 alterations have been associated with FGDY patients: (1) deletion of the entire coding frame encoding FGD1, (2) nonsense and frameshift mutations that result in truncated FGD1 proteins, or (3) missense point mutations located in some of the signalling motifs that are contained within the functional domains. Whereas most point mutations are found in the catalytic region (E380A, R402Q, R408Q, R442H, R443L or R443H, M466V, R522H and S558W) and the adjacent PH domain (R610Q), others are found in the PRD (S205I and P312L) or in the FYVE (K748E) domains.

in pathology and normal physiology, namely during cancer progression and vascular function (Ayala et al., 2009; Daubon et al., 2011). Here, we will tell the story of how the cell biology of an extremely rare Mendelian disorder has unexpectedly provided important clues towards our understanding of the physiological (e.g. development, vascular and bone remodelling) and pathological (e.g. cancer and vascular disease) events regulated by FGD1. We will discuss how cell biological studies of FGD1 and its interactors not only further our understanding of the complexity of CDC42 function and FGDY pathogenesis, but are also helping us to reconstruct the pathways that are involved in physiological processes and common multifactorial diseases, by highlighting shared trait components. Finally, we will present the hypothesis that FGD1 is a signalling hub in the regulation of ECM remodelling events and that investigation of its function will further our understanding of the connections between CDC42 signalling and ECM remodelling in physiology and complex diseases. From FGDY to the cell biology of FGD1 Rho, Rac and CDC42 are members of the Rho family of low molecular weight GTPases. They are molecular switches that control many cell functions by interacting with, and stimulating, various effector targets, including protein kinases, actin nucleators and phospholipases (Hall, 2005; Ellenbroek and Collard, 2007). These GTPases, thereby, affect fundamental physiological and pathological events, such as cell shape, morphology, polarization, motility and metastatic dissemination (Etienne-Manneville and Hall, 2002; Hall, 2005; Ellenbroek and Collard, 2007). Rho GTPases cycle between an inactive (GDPbound) and an active (GTP-bound) form. The exchange of GDP

for GTP induces a conformational change that allows interaction with effectors, which is terminated by hydrolysis of bound GTP to GDP. However, the small GTPases have limited hydrolytic and guanine nucleotide exchange activity, and thus, require accessory proteins to ensure the proper regulation of this cycle. These accessory proteins are part of three main classes; the GTPase activating proteins (GAPs), the guanine nucleotide dissociation inhibitors (GDIs) and the GEFs. The GAP proteins accelerate the intrinsically low GTPase activity, thus, increasing the deactivation rate of GTPases. GDI proteins interact with the GDP-bound forms of the small GTPases and act as negative regulators that control cycling between membranes and the cytosol. Finally, GEFs stimulate the exchange of GDP for GTP to generate the activated form of the GTPases. Aberrant signalling through Rho GTPases has often been implicated in human disease (Ellenbroek and Collard, 2007; Vega, 2008). However, in most cases disease-causing mutations do not occur at the level of the GTPases themselves, but in their upstream regulators – the GEFs – some of which were, for example, initially identified as oncogenes. The number of GEFs that have been identified greatly exceeds the number of known GTPases (Schmidt and Hall, 2002), and they probably hold the key in regulating the specificity of downstream signalling from Rho GTPases in diverse cellular functions and cell systems (Zhou et al., 1998). For instance, GEFs might be important components involved in invasion and metastasis (Vigil et al., 2010), as deregulated expression of GEF can lead to aberrant growth, invasiveness and/or increased metastatic potential. Although it is still poorly understood how GEFs are regulated, they clearly represent powerful candidates as spatial and temporal Rho-GTPase regulators. One such GEF is FGD1, which possesses a GEF activity that is specific for CDC42. Like most Rho GEFs, FGD1 contains a DBL homology (DH) domain that is positioned adjacent to a pleckstrin homology (PH) domain. Together, these two domains catalyze the exchange of CDC42-bound GDP for GTP (Fig. 1). FGD1 also features an N-terminal proline-rich domain (PRD), a cysteine-rich zinc-finger Fab-1, YOTB, Vps27 and EEA1 (FYVE) domain, and a second C-terminal PH domain (Gorski et al., 2000; Pasteris et al., 1994). The PRD has potential binding sites for Src homology 3 (SH3)-, WW- and Ena/VASP-homology 1 (EVH1, also known as SPRE1)-domain-containing proteins and profilin. The only two known FGD1 interactors, however, are cortactin and mammalian actin-binding protein (mAbp1, also known as DBNL) (Hou et al., 2003). Obviously, the binding of FGD1 to other cytoskeletal or signal transduction protein ligands cannot be excluded at this time (Estrada et al., 2001). The FGD1 N-terminal region is thought to be recruited to the subcortical actin cytoskeleton and the Golgi complex through protein–protein interactions involving polyproline tracts. In addition, the N-terminus negatively regulates the guanine nucleotide exchange activity (Estrada et al., 2001). The N-terminal PH domain could help the targeting of FGD1 to the appropriate intracellular location (Schmidt and Hall, 2002), whereas the C-terminal PH domain does not seem to serve this function (Estrada et al., 2001). FYVE domains usually bind with high affinity and specificity to phosphatidyl-inositol-3-phosphate [PtdIns(3)P], and this interaction is crucial for targeting FYVEdomain-containing proteins to PtdIns(3)P-enriched membranes, such as those of the endosomal system (Hayakawa et al., 2007). The FYVE domain of FGD1, however, appears to be atypical in this respect, as it does not direct the localization of FGD1 to such membranes (Estrada et al., 2001). This might be because the FGD1

FGD1 in ECM remodelling

3267

FGD1 signalling and functions TGF-β receptor

EGF

TGF-β

EGF receptor

PM

ECM Cortactin

Src

mAbp1

P Y

Invadopodia

Degradation by proteasome

FGD1 S P GSK3β

Golgi complex

CDC42

CDC42 GTP

GDP

JNK

Trafficking MLK3 ERK1/2

Journal of Cell Science

ECM remodelling

p38

P FAK

Transcription Metastasis

Podosomes

Cytoskeleton remodelling

Key

Phosphorylation Described pathway Interaction Cell programme

Fig. 2. FGD1 signalling and functions. FGD1 is a soluble protein that partially localizes to the subcortical actin cytoskeleton and Golgi membranes (Estrada et al., 2001), and translocates to the plasma membrane in response to extracellular stimuli, such as epidermal growth factor (EGF) in epithelial cells (Oshima et al., 2010) or TGF-b in endothelial cells (Daubon et al., 2011). During TGF-b signalling, binding of the cytokine to its receptor regulates the Src family kinase(s) (Varon et al., 2006), which subsequently phosphorylate FGD1 at tyrosine residues (Daubon et al., 2011). This presents a possible general mechanism for FGD1 activation. Activation of FGD1 leads to GDP/GTP exchange on CDC42, which, in turn, mediates a number of cellular events. Some of these events include cytoskeleton remodelling through the activation of JNK, which leads to cell migration (Olson et al., 1996; Oshima et al., 2011), the activation of ERK1/2 and subsequent focal adhesion kinase (FAK) activation, which leads to metastasis, or the activation of ERK1/2 and/or MLK3 and p38, which leads to osteoblast differentiation (Zou et al., 2011). FGD1 also regulates ECM remodelling through the formation of invadopodia in tumour cells (Ayala et al., 2009) or podosomes in endothelial cells (Daubon et al., 2011). Finally, FGD1 controls the export of cargo proteins from the Golgi complex through CDC42 activation (Egorov et al., 2009). The FGD1 SH3-binding domain binds directly to the SH3 domain of mAbp1 or cortactin (Hou et al., 2003), and FGD1–cortactin binding promotes CDC42-independent actin assembly by the Arp2/3 complex (Kim et al., 2004). Phosphorylation at serine 205 by glycogen synthase kinase 3-b (GSK3-b) mediates the degradation of FGD1 at the proteasome (Hayakawa et al., 2005).

domain of FYVE is lacking one of the three conserved signature motifs (the WxxD motif) that are involved in PtdIns(3)P binding and exhibits altered lipid binding specificities. In fact, the WxxD motif is important for the recognition of the 4-, 5- and 6-hydroxyl groups of the inositol ring (Sankaran et al., 2001; Kutateladze, 2006). As mentioned above, impaired FGD1 function, owing to mutations in its gene, causes the X-linked disease FGDY. In this context, it is of interest that three types of FGD1 gene alterations have been found in FGDY patients: (1) deletion of the entire FDG1 coding frame; (2) nonsense and frameshift mutations yielding truncated FGD1 proteins; or (3) missense point mutations located within the functional domains (Fig. 1). Whereas most point mutations are located in the catalytic region (E380A, R402Q, R408Q, R442H, R443L or R443H, M466V, R522H and S558W) and the adjacent PH domain (R610Q), some have also been described in the PRD (S205I,

P312L) and in the FYVE (K748E) domains. The extent to which these mutations impair FGD1 function, however, has not yet been precisely determined. It could be inferred that the known point mutations impair FGD1 function directly or indirectly by altering catalytic activity, localisation and/or stability of the protein. For instance, whereas FGD1 stimulates directed cell migration, the FGD1 (S205I) mutant fails to do so (Oshima et al., 2011). Furthermore, the same mutation appears to confer increased stability to the protein (Hayakawa et al., 2005). Thus, the naturally occurring mutations identified in FGDY patients aid the study of FGD1 by pointing to specific residues that can be tested for their involvement in the regulation of FGD1 function. FGD1 potently stimulates the reorganization of the actin cytoskeleton (Olson et al., 1996). In particular, when microinjected into Swiss 3T3 cells, it stimulates the formation of peripheral actin micro-spikes, and this effect can be reversed by the expression of dominant-negative CDC42. Interestingly, the direct

Journal of Cell Science

3268

Journal of Cell Science 125 (14)

binding of FGD1 to the multi-domain protein cortactin stimulates actin nucleation independently of the FGD1 guanine nucleotide exchange activity (Kim et al., 2004). Furthermore, two independent downstream CDC42 effectors, c-Jun N-terminal Kinase (JNK) and p70 S6 Kinase (S6K, also known as RPS6KB1), which are both involved in transcriptional regulation, become activated in FGD1-expressing cells (Olson et al., 1996). FGD1 also regulates persistent directional cell migration, and this function appears to involve the PRD domain (Oshima et al., 2011). In addition, expression of the DH and PH domains or the DH domain alone induces G1 cell cycle progression and entry into S phase as efficiently as constitutively active CDC42, suggesting that FGD1 facilitates G1 progression both through CDC42-dependent and CDC42-independent mechanisms (Nagata et al., 1998). Finally, recent data suggest that through the modulation of CDC42 activation, FGD1 is involved in the regulation of secretory protein export from the Golgi complex, the central processing unit of the secretory pathway (Egorov et al., 2009). Together, the data converge in positioning FGD1 upstream of CDC42 in these events (Fig. 2). However, the possibility remains that FGD1 mediates some of its biological activities through non-CDC42 targets (Whitehead et al., 1998). Two members of the CDC42 subfamily, TC10 (also known as RhoQ) and TCL (also known as RhoJ) (Billottet et al., 2008), might be possible mediators of these activities. From the cell biology of FGD1 to physiology and disease Important clues that aid our understanding of how FGDY ensues during development have recently emerged. FGD1 was found to be expressed in human mesenchymal stem cells (hMSCs) isolated from adult bone marrow (Gao et al., 2011). hMSCs can differentiate into many cell types, including fibroblasts, osteoblasts, adipocytes and chondrocytes, and are thought to have a function in musculoskeletal tissue maintenance. Indeed, during osteogenic differentiation of hMSCs in vitro, FGD1 expression and CDC42 activity were shown to be upregulated. Furthermore, the overexpression of wild-type FGD1 and a dominant-negative, inactive form of FGD1 promote and suppress hMSC osteogenesis, respectively. In another recent study, mixed lineage kinase (MLK3, also known as MAP3K11) was found to function downstream of FGD1 in the regulation of ERK1 and ERK2 (also known as MAPK1 and MAPK3, hereafter referred to as ERK1/2) and p38 MAPK, which in turn modulate the master regulator of osteoblast differentiation, runt-related transcription factor 2 (RUNX2), by phosphorylating it (Fig. 2) (Zou et al., 2011). These results are consistent with the clinical manifestations of FGDY, but also suggest that the role of FGD1 in bone remodelling extends beyond development into adult life. As mentioned above, two recent studies have opened up new perspectives by unveiling other potential functions for FGD1 in different contexts (Fig. 2). In the first study, it was shown that FGD1 is associated with invadopodia and, through direct modulation of CDC42, is essential for their formation and function in ECM degradation (Ayala et al., 2009). Invadopodia are actin-driven, proteolytically active membrane protrusions formed by tumour or transformed cells that are capable of ECM degradation (reviewed in Caldieri et al., 2009). They are thought to recapitulate, in vitro, the initial steps of tumour cell invasion and dissemination, namely the focal pericellular degradation of the ECM. These findings position FGD1 as the upstream

regulator of CDC42 in the formation and function of invadopodia to direct focal ECM degradation. In the same study, FGD1 was also found to be expressed in human prostate and breast cancer cells (but not in normal prostate and breast tissue). In addition, higher expression levels were found to correlate with an increase in tumour aggressiveness. Another recently published study (Daubon et al., 2011) has demonstrated that FGD1 is expressed in aortic endothelial cells (Box 1) and that its function is regulated by transforming growth factor beta (TGF-b) (Box 2). The study has also shown that FGD1 regulation occurs through phosphorylation at tyrosine residues and requires the participation of cortactin. In the model proposed on the basis of these observations, TGF-b-mediated FGD1 activation stimulates CDC42-dependent podosome assembly. Podosomes are dynamic, actin-rich plasma membrane adhesion microdomains that are endowed – in a similar manner to invadopodia – with ECM-degrading activity (Linder et al., 2010). Although they are mostly found in monocyte-derived cells, some non-myelomonocytic cells, such as endothelial cells (Box 1), can also be induced to assemble podosomes in response to cytokine stimulation (Osiak et al., 2005; Varon et al., 2006). The physiological relevance of these in vitro observations was revealed when it was shown that, in the endothelium of explanted murine aortic segments exposed to TGF-b, podosomes are present in situ and produce local alterations of the underlying basement membrane (Rottiers et al., 2009). Clearly, podosome formation is associated with

Box 1. Endothelial cell physiology The term ‘endothelium’ refers to the thin monolayer of endothelial cells lining the inner face of all blood vessels of the circulatory system. Located at the interface between the blood and the remainder of the vessel wall, the endothelium occupies a strategic position, where it monitors the transport of molecules between the plasma and tissues through bidirectional receptor-mediated and receptor-independent transcytosis and endocytosis. In larger blood vessels, the vascular endothelium also actively regulates the vascular tone, the breakdown of lipids, thrombogenesis, inflammation and vessel growth. Because they are exposed to mechanical forces exerted by the blood flow as well as noxious substances, endothelial cells are under extreme stress, and these cells react to stressors by modulating their cellular functions. As a consequence, endothelial dysfunction (defined by the loss or impairment of function or the acquisition of new functions) is a key step in most cardiovascular diseases. Microvessels are endowed with distinct functions, as they ensure the supply of oxygen and nutrients to tissues. In response to angiogenic factors, endothelial cells from the microvasculature drive the growth of new blood vessels. In the adult, angiogenesis is restricted to, and required for, reproductive functions and wound healing. However, tumour angiogenesis accompanies the growth of solid tumours and, thereby, constitutes one of the most deleterious aspects of cancer diseases. At the ultrastructural level, the characteristic features of endothelial cells are specialised endothelial-cell junctions involving vascular endothelial (VE)-cadherin, a particularly high number of caveolae equipped with numerous receptors and the presence of transendothelial channels. The description of podosomes in endothelial cells represents a recently discovered aspect of vascular biology with far-reaching implications in the understanding and, possibly, the treatment of vascular diseases.

FGD1 in ECM remodelling

Journal of Cell Science

Box 2. TGF-b signalling TGF-b is a pleiotropic cytokine that controls embryonic development and tissue homeostasis in adult organisms. It has also been implicated in pathologies, such as cancer, fibrosis or vascular diseases. TGF-b signalling is classified into two distinct pathways, SMADdependent and SMAD-independent pathways. In the SMADdependent pathways (also referred to as canonical pathways), the initial step is the binding of active TGF-b to a type II receptor (a serine/threonine kinase receptor), which, in turn, recruits and phosphorylates a type I receptor (another serine/threonine kinase receptor). The type I receptor then phosphorylates receptorregulated SMADs (R-SMADs), which form a complex with SMAD4 (co-SMAD). R-SMAD–co-SMAD transcriptional complexes subsequently accumulate in the nucleus to regulate gene expression (Pardali et al., 2010). In the SMAD-independent pathways (also referred to as noncanonical pathways) a number of other cellular signals can be directly activated following the binding of TGF-b to its receptors. For example, the PI3K–Akt–mTOR pathway regulates various cellular activities including proliferation, differentiation and apoptosis. Furthermore, Ras and TGF-b-activated kinase 1 (TAK1) control the MAPKs, which are also involved in transcriptional activation. Finally, TGF-b regulates RhoGTPases by activating GEFs, such as FGD1, thereby leading the remodelling of the cytoskeleton (Zhang, 2009).

vessel remodelling, a process which fits well with the recognized function of TGF-b in the maintenance of vascular homeostasis during adult life. These results, thus, implicate FGD1 in normal endothelial cell-mediated ECM remodelling and, therefore, potentially, in vascular pathology. Understanding FGD1-mediated ECM remodelling How can we bring together these apparently unrelated findings into a coherent picture? Can we generate a unifying hypothesis that allows us to place the pathogenesis of FGDY into the framework of the recent findings on invadopodia formation, tumour progression and endothelial physiology? The key to answering these questions might lie in regulated ECM remodelling and tissue invasion events. These processes are crucial for physiological processes such as inflammatory responses, wound healing as well as tissue remodelling, embryonic morphogenesis and differentiation. However, the pathological subversion of these events can drive tumour cell invasion, vascular diseases originating from endothelial cells, bone-remodelling diseases and possibly many other pathological disorders. ECM remodelling and tissue invasion require the integration of several processes. These include the local modulation of cytoskeleton structure and contractile forces, and the generation of specialized, transient domains that mediate the protease-dependent focal degradation of the extracellular matrix (ECM), which is required to make way for migrating cells and/or to unmask and activate cryptic sites and stimulatory factors embedded within the ECM (Page-McCaw et al., 2007). On the basis of the evidence obtained from the observation of both physiological and pathological processes, we suggest that FGD1 has an important function in the regulation of ECM remodelling in different tissues, different contexts and beyond development into adult life. The clinical outcome of FGD1 loss during development (in patients with FGDY) is certainly

3269

consistent with such a role, at the very least in bone remodelling during development. The fact that osteogenic differentiation is also FGD1-dependent (Gao et al., 2011; Zou et al., 2011) potentially extends the validity of this notion to adults. Indeed, our recent findings that describe the function of FGD1 in TGF-b-stimulated, endothelial-cell-mediated ECM remodelling fuels research in two directions. First, it positions TGF-b as a potential upstream regulator of FGD1 in osteoblast differentiation. Second, it suggests a regulatory function for FGD1 in angiogenesis, vasculogenesis and vascular remodelling. Indeed, because TGF-b is an important regulator of bone biogenesis, we could even speculate that the defects observed in faciogenital dysplasia might result, at least in part, from altered TGF-b signalling. An extensive search of the available literature has revealed some very interesting support for this hypothesis. For instance, dysplastic carotid arteries, artery malformations and aneurysms were incidentally discovered by cerebrovascular imaging (MRI) in an FGDY patient (DiLuna et al., 2007). It must not be forgotten, however, that this reflects only a single case and could, thus, represent a coincidental observation. Furthermore, a microarray-based study found that expression of FGD1 is downregulated in abdominal aortic aneurysm and arterial occlusive disease, compared with human healthy aorta (Armstrong et al., 2002), which might also be connected to altered ECM remodelling in arterial walls (Jones et al., 2009). Again, interestingly, TGF-b might have an important function in this process (Wang et al., 2010). Finally, the fact that FGD1 is required for ECM degradation in vitro by a number of cancer-derived cell lines and that it is expressed in human prostate and breast cancer cells, but not in normal prostate and breast tissue (Ayala et al., 2009), suggests that in adults, aberrant expression of FGD1 is linked to the pathological events associated with tumorigenesis. Namely, the GEF could be involved in the processes involved in ECM remodelling that are associated with tumour cell infiltration, invasion and/or dissemination. Conclusions Here, we have provided an overview of the current knowledge on the rare Mendelian disorder FGDY and the CDC42-specific guanine nucleotide exchange factor FGD1, which, when impaired by spontaneously occurring mutations, leads to the onset of this disease. On the basis of the evidence available so far, which stems from our own work and that of many others, we hypothesise that FGD1 might be an important regulator of events that control ECM remodelling and possibly cell invasion in both physiological and pathological settings. FGDY clearly testifies an important role of FGD1 in bone development, but the functions for this GEF might also continue throughout adult life in normal bone turnover, and perhaps even in certain bone disorders that are unrelated to FGDY. Classical cell biological approaches have allowed us to unveil further possible implications of FGD1 function in different contexts. For instance, normal FGD1 function might be important for vascular repair and/or formation, whereas aberrant expression might underlie pathogenic ECM remodelling, such as, for instance, in tumour cell invasion and/or infiltration and, perhaps, in certain types of vascular diseases. Clearly, the clinical outcome of the loss of FGD1 function during development (as in FGDY) is consistent with such a role. Needless to say, much work still lies ahead and the field is wide open for the verification of each of these scenarios.

3270

Journal of Cell Science 125 (14)

However, an important lesson can be taken away from the knowledge that monogenic diseases provide us with spontaneous ‘human allelic knockouts’ that afford us a unique view into pathophysiology. Mendelian disorders represent virtual gold mines to further our knowledge of gene and protein function, as well as with the potential to substantially improve our understanding of the molecular and functional basis of common, complex diseases by helping us to reconstruct normal and pathological pathways, as discussed previously (Antonarakis and Beckmann, 2006). Through the power of cell biological approaches, new hypotheses can be created and novel pathway components can be monitored and extended to physiology, and to common, complex diseases as a result of findings obtained through the study of inherited disorders. Acknowledgements

We thank Elena Fontana for the artwork.

Journal of Cell Science

Funding

Work in the authors’ laboratories was supported by the Italian Association for Cancer Research (AIRC, Milano, Italy to R.B.), the the European Union Seventh Framework Programme [grant number 237946 to E.G. and R.B.], the French National Research Agency [grant number ANR 2010 BLAN 1237 01 to E.G.], the France Cancer Research Association [grant number ARC n ˚ 5040 to E.G.) and the CNRS GDR [grant number 2823 to E.G.]. T.D. is the recipient of a France Cancer Research Association fellowship. References Antonarakis, S. E. and Beckmann, J. S. (2006). Mendelian disorders deserve more attention. Nat. Rev. Genet. 7, 277-282. Armstrong, P. J., Johanning, J. M., Calton, W. C., Jr, Delatore, J. R., Franklin, D. P., Han, D. C., Carey, D. J. and Elmore, J. R. (2002). Differential gene expression in human abdominal aorta: aneurysmal versus occlusive disease. J. Vasc. Surg. 35, 346-355. Ayala, I., Giacchetti, G., Caldieri, G., Attanasio, F., Mariggio`, S., Tete`, S., Polishchuk, R., Castronovo, V. and Buccione, R. (2009). Faciogenital dysplasia protein Fgd1 regulates invadopodia biogenesis and extracellular matrix degradation and is up-regulated in prostate and breast cancer. Cancer Res. 69, 747-752. Billottet, C., Rottiers, P., Tatin, F., Varon, C., Reuzeau, E., Maıˆtre, J. L., Saltel, F., Moreau, V. and Ge´not, E. (2008). Regulatory signals for endothelial podosome formation. Eur. J. Cell Biol. 87, 543-554. Caldieri, G., Ayala, I., Attanasio, F. and Buccione, R. (2009). Cell and molecular biology of invadopodia. Int. Rev. Cell Mol. Biol. 275, 1-34. Daubon, T., Buccione, R. and Ge´not, E. (2011). The Aarskog-Scott syndrome protein Fgd1 regulates podosome formation and extracellular matrix remodeling in transforming growth factor b-stimulated aortic endothelial cells. Mol. Cell. Biol. 31, 4430-4441. DiLuna, M. L., Amankulor, N. M., Johnson, M. H. and Gunel, M. (2007). Cerebrovascular disease associated with Aarskog-Scott syndrome. Neuroradiology 49, 457-461. Egorov, M. V., Capestrano, M., Vorontsova, O. A., Di Pentima, A., Egorova, A. V., Mariggio`, S., Ayala, M. I., Tete`, S., Gorski, J. L., Luini, A. et al. (2009). Faciogenital dysplasia protein (FGD1) regulates export of cargo proteins from the golgi complex via Cdc42 activation. Mol. Biol. Cell 20, 2413-2427. Ellenbroek, S. I. and Collard, J. G. (2007). Rho GTPases: functions and association with cancer. Clin. Exp. Metastasis 24, 657-672. Estrada, L., Caron, E. and Gorski, J. L. (2001). Fgd1, the Cdc42 guanine nucleotide exchange factor responsible for faciogenital dysplasia, is localized to the subcortical actin cytoskeleton and Golgi membrane. Hum. Mol. Genet. 10, 485-495. Etienne-Manneville, S. and Hall, A. (2002). Rho GTPases in cell biology. Nature 420, 629-635. Gao, L., Gorski, J. L. and Chen, C. S. (2011). The Cdc42 guanine nucleotide exchange factor FGD1 regulates osteogenesis in human mesenchymal stem cells. Am. J. Pathol. 178, 969-974. Gorski, J. L., Estrada, L., Hu, C. and Liu, Z. (2000). Skeletal-specific expression of Fgd1 during bone formation and skeletal defects in faciogenital dysplasia (FGDY; Aarskog syndrome). Dev. Dyn. 218, 573-586. Hall, A. (2005). Rho GTPases and the control of cell behaviour. Biochem. Soc. Trans. 33, 891-895. Hayakawa, A., Hayes, S., Leonard, D., Lambright, D. and Corvera, S. (2007). Evolutionarily conserved structural and functional roles of the FYVE domain. Biochem. Soc. Symp. 74, 95-105.

Hayakawa, M., Kitagawa, H., Miyazawa, K., Kitagawa, M. and Kikugawa, K. (2005). The FWD1/beta-TrCP-mediated degradation pathway establishes a ‘turning off switch’ of a Cdc42 guanine nucleotide exchange factor, FGD1. Genes Cells 10, 241-251. Hou, P., Estrada, L., Kinley, A. W., Parsons, J. T., Vojtek, A. B. and Gorski, J. L. (2003). Fgd1, the Cdc42 GEF responsible for Faciogenital Dysplasia, directly interacts with cortactin and mAbp1 to modulate cell shape. Hum. Mol. Genet. 12, 1981-1993. Jones, J. A., Spinale, F. G. and Ikonomidis, J. S. (2009). Transforming growth factorbeta signaling in thoracic aortic aneurysm development: a paradox in pathogenesis. J. Vasc. Res. 46, 119-137. Kim, K., Hou, P., Gorski, J. L. and Cooper, J. A. (2004). Effect of Fgd1 on cortactin in Arp2/3 complex-mediated actin assembly. Biochemistry 43, 2422-2427. Kutateladze, T. G. (2006). Phosphatidylinositol 3-phosphate recognition and membrane docking by the FYVE domain. Biochim. Biophys. Acta 1761, 868-877. Lebel, R. R., May, M., Pouls, S., Lubs, H. A., Stevenson, R. E. and Schwartz, C. E. (2002). Non-syndromic X-linked mental retardation associated with a missense mutation (P312L) in the FGD1 gene. Clin. Genet. 61, 139-145. Linder, S., Wiesner, C. and Himmel, M. (2011). Degrading devices: invadosomes in proteolytic cell invasion. Annu. Rev. Cell Dev. Biol. 10, 185-211. Nagata, K., Driessens, M., Lamarche, N., Gorski, J. L. and Hall, A. (1998). Activation of G1 progression, JNK mitogen-activated protein kinase, and actin filament assembly by the exchange factor FGD1. J. Biol. Chem. 273, 15453-15457. Olson, M. F., Pasteris, N. G., Gorski, J. L. and Hall, A. (1996). Faciogenital dysplasia protein (FGD1) and Vav, two related proteins required for normal embryonic development, are upstream regulators of Rho GTPases. Curr. Biol. 6, 1628-1633. Orrico, A., Galli, L., Faivre, L., Clayton-Smith, J., Azzarello-Burri, S. M., Hertz, J. M., Jacquemont, S., Taurisano, R., Arroyo Carrera, I., Tarantino, E. et al. (2010). Aarskog-Scott syndrome: clinical update and report of nine novel mutations of the FGD1 gene. Am. J. Med. Genet. 152A, 313-318. Oshima, T., Fujino, T., Ando, K. and Hayakawa, M. (2010). Proline-rich domain plays a crucial role in extracellular stimuli-responsive translocation of a Cdc42 guanine nucleotide exchange factor, FGD1. Biol. Pharm. Bull. 33, 35-39. Oshima, T., Fujino, T., Ando, K. and Hayakawa, M. (2011). Role of FGD1, a Cdc42 guanine nucleotide exchange factor, in epidermal growth factor-stimulated c-Jun NH2-terminal kinase activation and cell migration. Biol. Pharm. Bull. 34, 54-60. Osiak, A. E., Zenner, G. and Linder, S. (2005). Subconfluent endothelial cells form podosomes downstream of cytokine and RhoGTPase signaling. Exp. Cell Res. 307, 342-353. Page-McCaw, A., Ewald, A. J. and Werb, Z. (2007). Matrix metalloproteinases and the regulation of tissue remodelling. Nat. Rev. Mol. Cell Biol. 8, 221-233. Pardali, E., Goumans, M. J. and ten Dijke, P. (2010). Signaling by members of the TGF-beta family in vascular morphogenesis and disease. Trends Cell Biol. 20, 556567. Pasteris, N. G., Cadle, A., Logie, L. J., Porteous, M. E., Schwartz, C. E., Stevenson, R. E., Glover, T. W., Wilroy, R. S. and Gorski, J. L. (1994). Isolation and characterization of the faciogenital dysplasia (Aarskog-Scott syndrome) gene: a putative Rho/Rac guanine nucleotide exchange factor. Cell 79, 669-678. Rottiers, P., Saltel, F., Daubon, T., Chaigne-Delalande, B., Tridon, V., Billottet, C., Reuzeau, E. and Ge´not, E. (2009). TGFbeta-induced endothelial podosomes mediate basement membrane collagen degradation in arterial vessels. J. Cell Sci. 122, 4311-4318. Sankaran, V. G., Klein, D. E., Sachdeva, M. M. and Lemmon, M. A. (2001). Highaffinity binding of a FYVE domain to phosphatidylinositol 3-phosphate requires intact phospholipid but not FYVE domain oligomerization. Biochemistry 40, 85818587. Schmidt, A. and Hall, A. (2002). Guanine nucleotide exchange factors for Rho GTPases: turning on the switch. Genes Dev. 16, 1587-1609. Scriver, C. R., Beaudet, A. L., Sly, W. S., Valle, D., Childs, B., Kinzler, K. W. and Vogelstein, B. (2001). The Metabolic and Molecular Bases of Inherited Diseases. New York, NY: McGraw-Hill. Varon, C., Tatin, F., Moreau, V., Van Obberghen-Schilling, E., Fernandez-Sauze, S., Reuzeau, E., Kramer, I. and Ge´not, E. (2006). Transforming growth factor beta induces rosettes of podosomes in primary aortic endothelial cells. Mol. Cell. Biol. 26, 3582-3594. Vega, F. M. and Ridley, A. J. (2008). Rho GTPases in cancer cell biology. FEBS Lett. 582, 2093-2101. Vigil, D., Cherfils, J., Rossman, K. L. and Der, C. J. (2010). Ras superfamily GEFs and GAPs: validated and tractable targets for cancer therapy? Nat. Rev. Cancer 10, 842-857. Wang, Y., Ait-Oufella, H., Herbin, O., Bonnin, P., Ramkhelawon, B., Taleb, S., Huang, J., Offenstadt, G., Combadie`re, C., Re´nia, L. et al. (2010). TGF-beta activity protects against inflammatory aortic aneurysm progression and complications in angiotensin II-infused mice. J. Clin. Invest. 120, 422-432. Whitehead, I. P., Abe, K., Gorski, J. L. and Der, C. J. (1998). CDC42 and FGD1 cause distinct signaling and transforming activities. Mol. Cell. Biol. 18, 4689-4697. Zhang, Y. E. (2009). Non-Smad pathways in TGF-beta signaling. Cell Res. 19, 128-139. Zhou, K., Wang, Y., Gorski, J. L., Nomura, N., Collard, J. and Bokoch, G. M. (1998). Guanine nucleotide exchange factors regulate specificity of downstream signaling from Rac and Cdc42. J. Biol. Chem. 273, 16782-16786. Zou, W., Greenblatt, M. B., Shim, J. H., Kant, S., Zhai, B., Lotinun, S., Brady, N., Hu, D. Z., Gygi, S. P., Baron, R. et al. (2011). MLK3 regulates bone development downstream of the faciogenital dysplasia protein FGD1 in mice. J. Clin. Invest. 121, 43834392.