The Ubiquitin-Proteasome System Meets Angiogenesis

Abstract

A strict physiological balance between endogenous pro-angiogenic and anti-angiogenic factors control endothelial cell functions, such that endothelial cell growth is normally restrained. However, in pathologic angiogenesis a shift occurs in the balance of regulators favoring endothelial growth. Much of control of angiogenic events is instigated through hypoxia-induced VEGF expression. Ubiquitin-proteasome system plays a central role in fine-tuning function of core pro-angiogenic proteins including VEGF, VEGFR-2, angiogenic signaling proteins including PLCγ1 and PI3 kinase/AKT pathway and other non-VEGF angiogenic pathways. The emerging mechanisms by which ubiquitin modification of angiogenic proteins control angiogenesis involve both proteolytic and non-proteolytic functions. Here, I review the recent advances linking ubiquitin-proteasome system to regulation of angiogenesis and highlights the potential therapeutic of ubiquitin-proteasome system in angiogenesis-associated diseases.

Introduction

Angiogenesis, the growth of new blood vessels from pre-existing vessels, is an important physiological process in the body required for normal wound healing and female reproduction. Pathologic angiogenesis, either excessive or insufficient, is now recognized as a “common denominator” underlying a number of deadly and debilitating human diseases like cancer, age- related macular degeneration (AMD), diabetic retinopathy, and cardiovascular diseases (1,2). Although, the etiology and the mechanisms of development of these diseases in many ways are distinct, they all share abnormal angiogenesis. For example, while cancer has nothing to do with AMD, it does share one of its characters, angiogenesis, with AMD (Figure 1). Acquisition of angiogenesis by tumor cells is considered the most critical step in tumor growth and metastasis. To grow beyond two millimeters in diameter a tumor needs to acquire angiogenesis which is often established by hypoxia-induced expression of vascular endothelial growth factor-A (VEGF-A) and other angiogenesis inducing molecules (3). To support the growth of expanding tumor an “angiogenic switch” is turned on ensuing in normally quiescent endothelial cells to proliferate and sprout (76). It is now clear that induction of VEGF-A and other related angiogenesis inducers and reduction in the expression of angiogenesis inhibitors such as thrombospondin (TSP-1) governs tumor-induced angiogenic switch (77,78). Although it was initially thought angiogenesis to be significant when tumor mass reaches to a macroscopic size, it is now increasingly apparent that angiogenesis is instigated in early stage of tumor development (77,79), further underscoring angiogenesis as a vital components of tumor growth and metastasis.

Figure 1.

Schematic presentation of tumor-induced angiogenesis and choroidal neovascularization as manifested in wet AMD. Expression of VEGF is responsible for induction of angiogenesis which is associated with tumor progression and worsening of wet AMD.

VEGF-A also plays a central role in the development of choroidal neovascularization (CNV) and indeed is responsible for both neovascularization and vascular leakage in wet AMD (4). The hallmarks of wet AMD are drusen formation (i.e., focal deposition of debris between the retinal pigment epithelium (RPE) and Bruch’s membrane), CNV, RPE (retinal pigmented epithelial) cells detachment, fibrovascular scarring and vitreous hemorrhages. Aberrant blood vessel growth as well as blood vessel leakage subsequently results in the loss of central vision (80). Many cell types in the eye including retinal pigmented epithelial cells (RPE), pericytes, endothelial cells, glial cells, muller cells and ganglion cells synthesis and secrete VEGF. In addition to vital importance of VEGF in pathology of AMD, elevated VEGF levels also strongly correlate with retinal ischemia-associated neovascularization in diabetic retinopathy and retinopathy of prematurity (5,4).

Once angiogenic switch is activated, different sequential steps takes place, including the activation of various proteases from activated endothelial cells resulting in the degradation of the basement membrane surrounding the existing vessel, migration of the endothelial cells into the interstitial space, endothelial cell proliferation, sprouting, lumen formation, generation of new basement membrane with the recruitment of pericytes, and fusion of the newly formed vessels (81). In general and in most of pathological conditions angiogenesis starts when cells within a tissue respond to hypoxia (i.e., low oxygen) or in certain circumstances by oncogenic gene products such as Ras and Myc by inducing expression of VEGF along with other hypoxia- inducible genes (6). The VEGF family growth factors include placenta growth factor (PlGF), VEGF-A, VEGF-B, VEGF-C, VEGF-D and VEGF-E (7). VEGF-A isoforms are resultant of alternative splicing of messenger RNA (mRNA) and VEGF-165 is considered the most common VEGF-A isoform (8). Three different VEGF gene products including, PIGF, VEGF-A, VEGF-B identified as ligands for VEGFR-1. VEGF-A also along with VEGF-D and VEGF-C bind to and activate VEGFR-2 (7). VEGF-C and VEGF-D are also known to recognize VEGFR-3 (also called FLT4), an RTK that is expressed predominantly by lymphatic endothelial and hematopoietic progenitor cells (9). VEGF-E is a virally encoded VEGF-like protein and selectively binds to VEGFR-2 (10). VEGF family proteins also interact with non-RTK cell surface receptors including Neuropilin-1 and Neuropilin-2 which are characterized as co-receptors for VEGF family ligands (7,11). Among all the VEGF receptors and co-receptors, activation of VEGFR-2 by VEGF family ligands is considered the most critical event in angiogenesis (12,13). Beyond VEGF-mediated VEGFR-2 activation, recent studies demonstrate that VEGFR-2 also under certain circumstances could be activated by non-VEGF family ligands including heparan sulfate proteoglycans (HSPGs) (14) and Galectin-3, a glycan binding protein (15). HSPGs is proposed to potentiate VEGFR-2 activation through cis- and trans-binding with VEGFR-2 and VEGF complex (14) where galectin-3 is thought to potentiate VEGFR-2 activation by prolonging its presence in the plasma membrane (15). It is needless to say that although VEGF family proteins are prominent regulators of angiogenesis, however several other growth factors and cytokines including, angiopoietin-1, Del-1, FGF, HGF, Interleukin-8 (IL-8) and Leptin all are known to stimulate angiogenesis (82) adding further complexities to regulation of angiogenesis.

The role of VEGFR-2 in angiogenesis is well established and more comprehensive reviews on the role of VEGFR-2 in angiogenesis have been recently published (7,11,13). Figure 2 summarizes VEGF superfamily ligands and their interactions with VEGF receptors (Figure 2). It is increasingly evident that angiogenic signaling is established through utilization of an elegant and complex system of VEGF and non-VEGF ligands and VEGF receptors and co-receptors resulting in homo- and heterodimeric activation of VEGF receptors leading to complex processes of angiogenesis.

Figure 2.

VEGF superfamily ligands and receptors: The schematic of VEGF ligands, their interaction with VEGF receptors are shown.

Ubiquitin-Proteosome System

Ubiquitin is an evolutionarily conserved 76 amino acids polypeptide and was named for its ubiquitous expression in eukaryotes. Ubiquitin is activated by a ubiquitin-activating enzyme, E1 in an ATP-dependent manner and it is transferred to a ubiquitin conjugating enzyme (E2) and eventually a ubiquitin-protein ligase (E3) specifically attaches ubiquitin molecule to a target protein through the ε-amino group of a lysine residue (Figure 3). E3 ubiquitin ligases are a large family of proteins (almost 700 in human genome) that are known to involve in the regulation of the turnover and activity of many target proteins (16). E3 ligases are divided into two large groups; the homology to the E6-associated protein carboxyl terminus (HECT) domain- containing E3 ligases, and the really interesting new gene (RING) domain-containing E3 ligases. In the RING-type E3 ligases, there are single subunit E3 ligases (such as Cbl family E3 ligases) and multi-subunit E3 ligases (such as Cullin-RING ubiquitin ligases). In the recent years additional E3 ligases also have been identified that uses different domain to recognize E2 conjugating enzymes such as plant homeodomain (PHD) domain-containing E3 ligases and the U-box E3 ligases (17,16). Although, conjugation of ubiquitin to target proteins was initially recognized as a signal for protein degradation by 26S proteasome, however, it is now recognized that ubiquitination regulates a broad range of cellular functions including, protein processing, membrane trafficking and transcriptional regulation (16,18). The recent studies show that ubiquitination also can impact cell signaling by targeting activation of proteins in a proteolysis-independent manner (19,56,63).

Figure 3.

The schematic of ubiquitin-proteasome system is shown. Ubiquitin (Ub) is activated by the ubiquitin-activating enzyme (E1) followed by its transfer to a ubiquitin-conjugating enzyme (E2). E2 transfers the activated ubiquitin moieties to the protein substrate that is bound specifically to a particular ubiquitin ligase (E3). The transfer of ubiquitin takes place either directly in the case of RING finger ligases or via an additional thiol-ester intermediate on the ligase in case of HECT domain ligases. Repeated conjugation of ubiquitin moieties to one another generates a polyubiquitin chain that serves as the binding and degradation signal for the 26S proteasome. The protein substrate is degraded generating short peptides, and free ubiquitin which could be further re-used.

Multiple ubiquitin molecules can be attached to a target protein either by means of mono- ubiquitination (i.e., attachment of a single ubiquitin to one or multiple lysine residues). Mono- ubiquitination is regarded as a signal for non-proteolytic events such as endocytosis, histone regulation, DNA repair, virus budding and nuclear export (18). Alternatively, ubiquitin can be attached to a target protein in the form of poly-ubiquitination, where multiple ubiquitin molecules are attached to a single lysine residue. There are seven different lysine residues in ubiquitin that can potentially be employed for ubiquitin-chain assembly. Lys48- and Lys-29-linked poly- ubiquitination generally is associated with degradation of target proteins by the 26S proteasome where Lys63-linked poly-ubiquitination is involved in DNA repair, signal transduction and endocytosis, but not degradation (16,18). Clearly, ubiquitin machinery system is evolved to play a versatile role in protein functions ranging from protein turnover, subcellular localization to kinase activation and consequently it is highly relevant to pathobiology of many human diseases.

The ubiquitin proteasome system (UPS) is consists of two major components including substrate-recruiting and substrate-degrading enzymes. The former is composed of three enzymes, the first of which (E1) activates the polypeptide ubiquitin in an ATP-dependent manner, enabling its transfer onto a ubiquitin carrier enzyme (E2). Activated ubiquitin is then transferred by a ubiquitin protein ligase (E3) to a substrate protein (83). The substrate-recruiting components of UPS then catalyze the formation of an isopeptide bond between the C-terminal glycine residue of ubiquitin and the ε-amino group of a substrate protein lysine residue. Continual addition of ubiquitin moieties onto substrate (i.e., polyubiquitination) facilitates recognition of substrate by the proteolytic machinery of the UPS, the 26S proteasome (83, 84). The 26S proteasome complex is essentially composed of one 20S and two 19S units (Figure 3). The 19S complex has two multi-subunit components and often described as the “base” and the “lid”. The base contains six ATPases which they belong to the TRIPLE-A family of ATPases plus two non-ATPase subunits which binds to the 20S catalytic core. The lid contains up to ten non-ATPase subunits (85,86). Collectively, the base and lid together function in the recognition of ubiquitinated substrates and their subsequent binding. The 20S complex is composed of 28 related subunits (14 different subunits) which are arranged as four heptameric staggered rings. The two outer rings contain the α subunits (α1-α7). The two inner rings contain two copies of the β subunits (β1-β7). Within the 20S proteasome, subunits β1, β2, and β5 exhibit postglutamyl peptide hydrolyzing (PGPH), trypsin-like and chymotrypsin-like cleavage activity, respectively. More comprehensive reviews on the function and composition of 26S-proteosome have been recently published (83, 84,86,87).

Regulation of VEGF expression by von Hippel-Lindau (VHL) E3 ubiquitin ligase

Molecular mechanism by which hypoxia sets off expression of VEGF has been extensively studied (6,20). In normoxic conditions (i.e., normal oxygen level) VEGF expression is generally inhibited due to interaction of von Hippel-Lindau (pVHL) E3 ubiquitin ligase with hypoxia-induced transcription factor, HIF-1α leading to its ubiquitination which targets HIF-α for degradation by 26S-proteasome. Among three HIF-α isoforms, HIF-1α and HIF-2α are closely related and are each able to interact with hypoxia response elements (HER) to induce VEGF expression (21,22), where HIF-3α appears to be involved in the negative regulation of hypoxia- induced gene expression (23) .

Oxygen-mediated posttranslational modification of HIF-α through non-heme and iron-dependent oxygenases that uniquely hydroxylate specific HIF-α at proline residues regulates its transcriptional activity. Hydroxylation of human HIF-1α at two proline residues (Pro402 and Pro564) by prolyl hydroxylase domain (PHD) proteins creates binding sites for VHL E3 ubiquitin ligase complex that targets HIF-1α for proteasomal degradation (24,25). These proline hydroxylation sites contain a conserved LxxLAP (X, any amino acids) motif which is recognized by PHD proteins leading to HIF-1α hydroxylation by PHD (24,26). Interestingly, hydroxylation of an asparagine residue (Asn803) in the C-terminal activation domain of HIF-1α by HIF asparagine hydroxylase, termed FIH, factor inhibiting HIF, inhibits HIF-1α activity by blocking interaction of the HIF-1α C-terminal activation domain with the transcriptional co-activator, p300 (27). In hypoxic conditions, however VEGF is generally over-produced which leads to pathological angiogenesis. In response to low oxygen, pVHL E3 ubiquitin ligase is S- nitrosylated, the covalent attachment of a nitrogen monoxide group to the thiol side chain of cysteine, which blocks HIF-α interaction with pVHL E3 ubiquitin ligase. HIF-1α escapes from ubiquitin-mediated degradation as a consequence of S-nitrosylation of pVHL (Figure 4). NO- mediated S-nitrosylation of HIF-1α at cysteine residue (C800) permits interaction of HIF-α with p300 prompting its transcription activity and VEGF expression (28).

Figure 4.

Role of pVHL in expression of VEGF and angiogenesis: In the presence of oxygen proline residues in the oxygen-dependent degradation (ODD) domain of HIF are hydroxylated. This allows HIF-α to interact with pVHL. The interaction between HIF and pVHL causes degradation of HIF through ubiquitination. In response to low oxygen (i.e., hypoxia), pVHL is S- nitrosylated, preventing HIF-α to interact with and pVHL and the degradation of HIF-α is disallowed. HIF-α stimulates gene expression such as VEGF which stimulates angiogenesis as manifested in cancer progression. S-nitrosylation is the covalent attachment of a nitrogen monoxide group to the thiol side chain of cysteine.

Another important aspect and added complexity of HIF-α regulation is the function of heat shock protein, Hsp90. Hsp90 often is upregulated under cellular stress conditions such as hypoxia (88) which prevents HIF-1α degradation apparently in a pVHL independent manner (88,89,102). Consistent with the protective role of Hsp90 in HIF-1α agents that inhibit Hsp90 activity also have been shown to promote ubiquitin-mediated degradation of HIF-1α (89,90). More recent study indicates that inhibition of Hsp90 by hemin, a derivative of protoporphyrin compound, increases HIF-1α ubiquitination and with it angiogenesis (91).

β-Trcp (β-transducin repeat-containing protein) ubiquitin E3 ligase controls ubiquitination and degradation of VEGFR-2

Activation of VEGFR-2 by VEGF family ligands mediates most of the known VEGF cellular responses (7,13,29). Central to proper regulation of angiogenic activity of VEGFR-2 is the process by which VEGFR-2 triggers its own internalization and degradation, consequently terminating its angiogenic signaling. Upon stimulation with VEGF family proteins, VEGFR-2 is removed from cell membrane and undergoes through clathrin-dependent endocytosis (30) which initiates its degradation (31) and recycling (32). Interestingly, VEGFR-2 internalization is stabilized by cadherin-5 (34). Cadherin-5 dependent stabilization of VEGFR-2 is established by reducing tyrosine phosphorylation of VEGFR-2 perhaps by recruiting tyrosine phosphatases to VEGFR-2 (30,33). Ligand-mediated degradation of VEGFR-2 requires its tyrosine kinase activity and activation of protein kinase C (PKC) pathway accelerates its degradation (31). On the other hand, activation of p38 MAPK has been shown to stabilize VEGFR-2 (34), suggesting that VEGFR-2 stability and degradation in endothelial cells is highly fine-tuned by activity of PKC and p38 MAPK pathways. Initial studies demonstrated that carboxyl terminal of VEGFR-2 plays a pivotal role in VEGFR-2 stability and degradation. Progressive deletion of carboxyl terminal of VEGFR-2 has been shown to inhibit ligand-dependent degradation of VEGFR-2 (29,31). A recent study has identified presence of a PEST domain in the carboxyl domain of VEGFR-2 which may account for critical role of carboxyl domain in VEGFR-2 degradation (34). PEST motif (rich in proline (P), glutamic acid (E), serine (S), and threonine (T) is considered a signature of short-lived proteins degraded by the ubiquitin pathway (35). It is thought that PEST sequences are unstructured regions in certain protein sequences possibly serving as a phospho-degron for the recruitment of F-box containing ubiquitin E3 ligases leading to ubiquitination and degradation (36,37). Phosphorylation of Ser1188 and Ser1191 of PEST domain of VEGFR-2 recruits SCF-βTrcp1 E3 ubiquitin ligase to VEGFR-2 leading to SCF- βTrcp1-dependent ubiquitination and degradation of VEGFR-2. Degradation of VEGFR-2 is mainly attained through Lys48-linked polyubiquitination (34).

β-TrCPs (also called FWD1) belong to a larger family of Fbw (F-box/WD40 repeat containing) proteins whose general features are the presence of a 42-48 amino-acid F-box motif at the N- terminus and seven WD40 repeats at the C-terminus (38). β-TrCPs are highly conserved across the species, in particular within the F-box motif and WD40 repeats motif. Human β- TrCP1 and β-TrCP2 exist in multiple isoforms due to alternatively spliced mRNA, but all are conserved in F-box and all seven WD40 repeats (39,40). The most notable differences in sequences between β-TrCP1 and β-TrCP2 are in their N-terminal regions, which are proximal to the F-box motif. The N-terminal sequences of β-TrCP1 and β-TrCP2 is thought to allow these proteins to undergo homo- and hetero-dimerization (41). Interestingly, it appears that both β- TrCP1 and β-TrCP2 can promote ubiquitination of VEGFR-2 suggesting that perhaps to some extent they may act in a redundant fashion in VEGFR-2 ubiquitination (34). A redundant role of mammalian β-TrCP1 and b-TrCP2 in ubiquitination and degradation of other proteins including IκB and β-catenin was also suggested (42,43).

Role of ubiquitination in PLCγ1 activation and angiogenesis

Activation of phospholipase Cγ1 (PLCγ1) in endothelial cells is considered among the chief mediators of the angiogenic signaling of VEGFR-2. It catalyzes the formation of inositol 1,4,5- trisphosphate (IP3) and diacylglycerol (DAG) from phosphatidylinositol 4,5-bisphosphate (PIP2). Phosphorylation of Tyr1173 on the mouse VEGFR-2 (corresponding to Tyr1175 on human VEGFR-2) is identified as a primary site responsible for the recruitment of PLCγ1 to VEGFR-2 (19, 44,45). Substantial information has been obtained from animal models that link PLCγ1 to angiogenesis. The initial evidence linking PLCγ1 to endothelial cell function and angiogenesis was provided by targeted deletion of PLCγ1, which resulted in early embryonic lethality between E (embryonic day) 9.5 and E10.5 due to significantly impaired vasculogenesis and erythrogenesis (46). Inactivation of PLCγ1 in zebrafish also is shown required for the function of VEGF and arterial development (47). Additional evidence for the importance of PLCγ1 in angiogenic signaling of VEGFR-2 comes from the use of pharmacological inhibition of PLCγ1, where U73122, a potent PLCγ1 inhibitor have been shown to inhibit endothelial cell tube formation in vitro (45) and angiogenesis in vivo in CAM assay (48). Silencing expression of PLCγ1 in primary endothelial cells by siRNA strategy also inhibits VEGF-mediated endothelial cell tube formation and proliferation (49) further underscoring the importance of PLCγ1 pathway for angiogenic signaling of VEGF.

PLCγ1 is a multi-domain protein it contains two SH2 domains and one SH3 domain between the catalytic domains. The SH2 domains recognize phosphotyrosine 1173 on VEGFR-2 (45), while the SH3 domain recognizes proline-rich sequences, PXXP motif. In addition to its SH domains, PLCγ1 also contains a C2 domain, EF hand and two putative PH domains. The presence of both N- and C-terminus SH2 domains is required for optimal binding of PLCγ1 with VEGFR-2 (45). PLCγ1 also interacts with c-Cbl through its proline rich motif in a non-inducible manner (31). As a result of activation by VEGFR-2, c-Cbl is recruited to VEGFR-2 and distinctly inhibits phosphorylation of tyrosine 783 on PLCγ1 in an ubiquitination-dependent manner (19,49). c-Cbl negatively regulates PLCγ1 activation in a proteolysis-independent manner. Instead of targeting it for degradation, c-Cbl distinctively mediates ubiquitination of PLCγ1 and suppresses its phosphorylation on Y783 (19). The conundrum of how PLCγ1 escapes from ubiquitin-mediated degradation but renders to a less enzymatic active state warrants further investigation.

Endothelial cells derived from c-Cbl knockout mice also showed that loss of c-Cbl results in an increased in phosphorylation of PLCγ1 with no apparent effect on its half-life (49). Over- expression of c-Cbl in endothelial cells also has been shown to inhibit tube formation and sprouting of endothelial cells. Conversely, over-expression of c-Cbl (70Z/3-Cbl), an E3 ligase deficient variant form of c-Cbl or silencing its expression by siRNA elevated sprouting of endothelial cells (19). A recent study has demonstrated that in c-Cbl nullizygous mouse the VEGF- and tumor induced angiogenesis is highly elevated (48,49). It appears that role of c-Cbl in angiogenesis is widespread since laser-induced angiogenesis in c-Cbl knockout mice also results in enhanced retinal neovascularization (48).

Role of ubiquitination in PI3 kinase/AKT pathway

Phosphoinositide3-kinase (PI3K) signal transduction pathway is one of the main signaling routes that VEGFR-2 employs to stimulate endothelial cell survival and proliferation (50,51,52). VEGFR-2 activates PI3K through recruitment of p85 of PI 3 kinase involving tyrosine 799 and tyrosine 1173 (48,49). PI3K consists of an 85-kDa regulatory subunit and a 110-kDa catalytic subunit. It is a lipid kinase that converts the plasma membrane lipid phosphatidylinositol 4,5- bisphosphate (PIP2) to phosphatidylinositol-3,4,5-triphosphate (PIP3). Proteins with pleckstrin- homology (PH) domains, such as protein kinase B (PKB/AKT), phosphoinositide-dependent kinase-1 (PDK-1) and PDK-2, bind to PIP3. AKT is activated by PIP3, PDK1 and PDK2 leading to phosphorylation of a host of other proteins that affect cell proliferation, cell cycle progression and cell survival (54). Cbl-b ubiquitin E3 ligase is known to interact with the p85-SH2 domain and catalyzing p85 polyubiquitination (55,56). Interestingly, the Cbl-mediated ubiquitination does not lead to degradation of p85 (56).

AKT is a serine/threonine protein kinase and is one of the key PI3K substrates that plays a central role in mediating VEGFR-2 mediated cellular events in endothelial (57,58). It has recently been shown that carboxyl terminus of Hsc-70-interacting protein (CHIP) interacts with AKT and induces its ubiquitination (59). In addition, tetratricopeptide repeat domain 3 (TTC3) containing E3 ligase recently has been linked to AKT ubiquitination and degradation (60,61). Interestingly, TTC3 interacts only with active AKT, but not inactive AKT, in the nucleus (60,61), indicating that perhaps TTC3-mediated AKT ubiquitination is important for controlling AKT signaling in nuclear. TTC3 itself is target of AKT and is phosphorylated at S378 by AKT and this phosphorylation appears to be necessary for TTC3 E3 ligase activity (60,61). BRCA1 is another E3 ligase that also interacts with activated AKT and targets it for ubiquitination and degradation (62). Recent studies also found that AKT is ubiquinated by TRAF6 E3 ligase. TRAF6 directly interacts with and induces AKT ubiquitination (63). TRAF6-mediated AKT ubiquitination takes place through the K63-linked modification and does not trigger AKT degradation. Interestingly, K63 chain polyubiquitination of AKT contributes to its membrane localization where it gets phosphorylated (63). Figure 5 summarizes various ubiquitin E3 ligases involved in the fine- tuning abundance and activation of key angiogenic proteins (Figure 5). Some ubiquitin E3 ligases such as Cbl family proteins target multiple angiogenic proteins where in some other cases more than one ubiquitin E3 ligases involved in the ubiquitination of an angiogenic protein as illustrated for AKT (Figure 5).

Figure 5.

Regulation of core angiogenic proteins by ubiquitin E3 ligases. Expression of VEGF is regulated by activity of pVHL. βTrcp1 ubiquinates VEGFR-2 and subjects it for proteasomal degradation. c-Cbl catalyzes ubiquitination of numerous angiogenic signaling proteins including VEGFR-1, Eph, Tie2 and PLCγ1. AKT abundance and activity is regulated TRAF6, CHIP and TTC3. Itch regulates degradation of Notch1 receptor.

Role of Ubiquitination in Wnt Signaling

Wnt pathway is another key play of angiogenesis. Binding of Wnt to its seven-span transmembrane receptor, Frizzled (Fz), and its co-receptor, Lrp5/6, at the cell surface initiates a signaling cascade that mediates angiogenesis and other key developmental processes including stem cell maintenance, growth and cell fate specification, and cell migration (64). Deregulation of activation of Wnt/β-catenin signaling is linked to a range of human diseases including cancer (64,65). During the resting state of canonical Wnt signaling, several key Wnt- associated signaling proteins including β-catenin via ubiquitination are targeted for degradation. Initially, the Adenomatous Polyposis Coli (APC) protein forms a complex with glycogen synthase kinase 3β (GSK-3β) and axin. This complex is then binds to β-catenin in the cytoplasm which leads to phosphorylation of β-catenin by CK1 (casein kinase 1), and GSK-3β. Phosphorylation leads to creation of a phospho-degron motif on β-catenin which allows ubiquitin E3 ligases such as β-Trcp ubiquitin E3 ligases, and Jade-1 to recognize β-catenin which targets β-catenin for ubiquitination leading to its 26S-proteasome mediated degradation (66,67). In addition, other ubiquitin E3 ligases such as Siah1 and Ozz also target β-catenin for degradation in cell type or context specific manner (92,93).

Removal of cytosolic β-catenin through ubiquitin-proteasome system prevents β-catenin from translocating into the nucleus, where it acts as a transcription factor for genes associated with angiogenic events such as proliferation of endothelial cells. In contrast, activation of the canonical Wnt signaling pathway results in inhibition of β-catenin degradation leading to increased cytosolic β-catenin, which then translocates to the nucleus. In the nucleus, β-catenin associates with at least one of a family of Tcf/Lef transcription factors and induces expression of numerous genes such as cyclinD1 and c-myc (68), which are implicated in cellular proliferation. Besides Wnt pathway mediated ubiquitination of β-catenin, the stability of β-catenin protein also is regulated by ubiquitination of cadherins. For example c-Cbl-related ubiquitin E3 ligase, Hakai associates with E-cadherin promoting its degradation resulting in destruction of cadherin-β- catenin complex formation and its degradation (69).

Interestingly, in addition to regulation of β-catenin protein levels by ubiquitination, the levels and sub-cellular functions of Dvl are tightly regulated via multiple ubiquitin-dependent pathways. Binding of the Kelch-like 12 (KLHL12) E3 ligase to Dvl is regulated by Wnt stimulation. Subsequent ubiquitylation of Dvl leads to its proteasomal degradation (70), suggesting that KLHL12 acts as a Wnt-mediated negative regulator of the Wnt pathway by inducing the degradation of Dvl. Surprisingly, in neuronal cells, Dvl uniquely is ubiquitinated by HECT-type E3 ligase, NEDL1 not by KLHL12 (71), suggesting a cell type specific regulation of Dvl by ubiquitin-proteasome system.

Ubiquitination system as a potential target for anti-angiogenesis and anti-cancer therapy

Given that expression and degradation of core pro-angiogenic proteins are regulated by ubiquitin-proteosome system selective targeting the different components of this pathway may represent a potentially effective means for anti-angiogenesis treatments. Our expanding understanding of the ubiquitination system and its role in angiogenesis has generated significant interest in the development of novel strategies to block pathological angiogenesis. The role of ubiquitination pathway in human diseases in general and angiogenesis, in particular, is still in its infancy, because the molecular mechanisms and gene products involved in ubiquitin- proteosome system is not fully understood. Moreover, there has been no comprehensive analysis of functional importance in the ubiquitin-proteosome system in angiogenesis- associated human diseases. While there are potentially many different components of the ubiquitin-proteosome system that might be targeted for inhibition or stimulation in milieu of angiogenesis, the therapeutic value of existing proteasome inhibitor, Bortezomib /Velcade™ (Millennium Inc.), a first drug targeting the ubiquitin-proteosome system approved by FDA for treatment of relapsed or refractory multiple myeloma (72) could be explored in angiogenesis- associated diseases.

Recent studies have already linked potential therapeutic value of inhibition of proteosome pathway to angiogenesis. Indeed, treatment of endothelial cells in cell culture system with proteasome inhibitors inhibit capillary tube formation of endothelial cells and blood vessel formation in embryonic chick chorioallantoic membrane (CAM) assay (73,74). Moreover, Bortezomib have been shown to inhibit tumor angiogenesis in murine xenograft model (75). More recently it has been shown that small molecules such as nutlins™ (Roche, Inc.) and RITA can block p53 ubiquitination by inhibiting the activity of MDM2 ubiquitin ligase which is known to mediate p53 ubiquitylation (94,95, 101). Loss of p53 activity is linked to both tumor growth and angiogenesis suggesting that possible application of nutlins™ in cancer treatment in principle could target both tumor cells and angiogenesis. PR-171™ (Proteolix, Inc.) a synthetic analog of epoxomycin is another proteosome inhibitor which was reported irreversibly to inhibit the chymotryptic site of the 26S proteasome and initial studies suggest that it has more potent anti- cancer activity than Velcade™ ((Millennium Inc.) (96). Moreover, the recent patent applications indicate that ubiquitin activating enzyme (E1) also could be targeted for possible therapeutic use (97,98). More comprehensive reviews on the inhibition of 26S-proteosome and drug discoveries have been recently published (99,100). Numerous ubiquitin E3 ligases are involved in the regulation of angiogenesis, however it remains to be determined if a particular ubiquitin E3 ligase can indeed be exploited as a target for molecular therapeutic approaches in angiogenesis-associated diseases. Regardless of whether a particular ubiquitin E3 ligase could be targeted for therapeutic approaches in angiogenesis-associated diseases, broad studies are clearly required to obtain better understanding for their role in angiogenesis. Based on the current drug discovery activities targeting ubiquitin-proteosome system, it’s evident that by harnessing ubiquitin-proteasome system one can design strategies to different components of ubiquitin-proteasome system either to selectively target proteins for ubiquitylation/degradation or inhibit protein degradation. Hence, it is not unreasonable to expect more drug discovery efforts based on ubiquitin are made toward targeting proteins with pro-angiogenesis and anti- angiogenesis activities in the near future.

Conclusions and Perspectives

The emerging role of ubiquitin-proteasome system in regulation of angiogenesis underlines the importance of investigating this pathway in milieu of angiogenesis. The reports outlined in this review enumerate examples of regulation of angiogenesis by various components of ubiquitin- proteasome system although these represent only the beginning of our understanding of this important pathway in regulation of angiogenesis. Characterizing the nature of this system in angiogenesis and added complexity of ubiquitin-proteasome system will undoubtedly have many therapeutic applications. Understanding the molecular basis of ubiquitin-proteasome system and the target protein substrates in endothelial cells may also provide a foundation for learning to stimulate or inhibit angiogenesis. Finally, it is reasonable to envision ubiquitin- proteasome system as a key component of angiogenic switch in cancer and other pathological angiogenesis.