The Ubiquitin-Proteasome System
The Ubiquitin-Proteasome System
Regulated protein degradation via the ubiquitin-proteasome system ([Fig. 1]) is an essential aspect of cell signaling pathways, functioning from cell-cycle
control and transcription to development [1], [2], [3]. In the ubiquitin-proteasome system, the client proteins are ubiquitinated by the
ubiquitin system and degraded by the 26S proteasome in an ATP-dependent manner. Ubiquitin
is composed of 76 amino acids and attaches to a client protein (ubiquitination) prior
to degradation. In the ubiquitin system, ubiquitination requires the sequential actions
of three enzymes, ubiquitin-activating enzyme (E1), ubiquitin-conjugating enzyme (E2),
and ubiquitin-protein ligase (E3), which results in the formation of the polyubiquitin
chain. Then, the polyubiquitin chain, tagged to the client protein, is recognized
by the 26S proteasome, an intracellular high-molecular weight protease subunit complex
[1], [2], [3], [4], [5], [6]. The 26S proteasome consists of two subcomplexes, the 20S core particle (also known
as the 20S proteasome) and the 19S regulatory particle ([Fig. 1]), and the protein portion of a client protein is degraded by the proteolytic active
sites in a cavity of the 20S proteasome. The 20S proteasome is classified as a threonine
protease that contains two pairs of three different catalytic sites. The β1, β2, and
β5 subunits contain catalytic sites with caspase-like, trypsin-like, and chymotrypsin-like
activities, respectively. These activities are capable of hydrolyzing proteins into
oligopeptides. The level of the 26S proteasome is increased especially in tumor cells.
Therefore, it is reasonable to develop specific compounds targeting proteasome-mediated
proteolytic degradation for cancer treatment.
Fig. 1 Schematic diagram of the ubiquitin-proteasome system. This system consists of the
ubiquitin system, the 26S proteasome, and the delivery system. Ub, ubiquitin; E1,
Ub-activating enzyme; E2, Ub-conjugating enzyme; E3, Ub-protein ligase.
The recent approval of bortezomib, a synthetic proteasome inhibitor, for the treatment
of relapsed multiple myeloma has opened the way to the discovery of drugs targeting
the ubiquitin-proteasome system consisting of the proteasome, ubiquitinating enzymes,
the delivery system, and deubiquitinating enzymes (see below). To date, various natural
and synthetic products have been reported to inhibit the components of the ubiquitin-proteasome
system. In this review, we focus on recent progress concerning natural products targeting
the ubiquitin-proteasome system as well as synthetic compounds with potent inhibitory
effects.
Targeting the 20S Proteasome
Targeting the 20S Proteasome
Inhibitory mechanisms of proteasome inhibitors
Proteasome inhibitors show antitumor activity against various tumor cells that are
resistant to conventional chemotherapeutic agents. Structurally diverse proteasome
inhibitors have been developed by chemical synthesis and also by searching natural
sources and chemical libraries as drugs for the clinical treatment of cancer and also
as molecular tools for the investigation of cellular events. To date, various synthetic
peptides including MG132 (Z‐Leu-Leu-Leu-al, 1) ([Fig. 2]) [7] and bortezomib (PS-341, Velcade®, 2) ([Fig. 3]) [8], [9], and natural products including salinosporamide A (NPI-0052, 3) ([Fig. 4]) [10] have been reported to inhibit proteasomal activity. Bortezomib (2), a synthetic proteasome inhibitor developed by Millennium Pharmaceuticals, Inc.,
was approved for myeloma therapy in the United States in 2003, which strongly indicates
that the proteasome would be a novel target for cancer treatment [11]. Salinosporamide A (3), isolated from the culture of a marine actinomycete, was found to be more effective
than 2 and is now undergoing clinical trials. Several proteasome inhibitors have been developed
and are classified into five groups, peptide aldehydes, peptide boronates, β-lactones,
epoxyketones, and macrocyclic vinyl ketones, on the basis of inhibitory mechanisms.
Fig. 2 Structures of proteasome inhibitors containing peptide aldehydes and their inhibitory
mechanism.
Fig. 3 Structures of proteasome inhibitors containing peptide boronates and their inhibitory
mechanism. Note that bortezomib (2) and CEP-18770 (8) are under evaluation in clinical trials for cancers.
Fig. 4 Structures of proteasome inhibitors containing β-lactones and their analogs, and
their inhibitory mechanism. Note that 19 and 20 lack β-lactones but contain boronates and that salinosporamide A (3) is in clinical trials for cancers.
Proteasome inhibitors: peptide aldehydes
Peptide aldehydes ([Fig. 2]), the first proteasome inhibitors [7], [12], are the most widely used as molecular tools for the investigation of various cellular
events. Generally, peptide aldehydes act against serine and cysteine proteases. The
aldehyde functional group of the inhibitor is readily attacked by a hydroxy or thiol
group at the active site of the protease. In the case of the proteasome, the N-terminal threonine residue at its active site carries out a nucleophilic attack on
the aldehyde moiety of the inhibitor and then a covalent hemiacetal adduct is formed
between the inhibitor and the threonine residue of the proteasome ([Fig. 2]). The formation of the covalent adduct is mediated by a reversible reaction under
physiological conditions, and aldehyde inhibitors are rapidly oxidized into inactive
acids in cells and transported out of the cell by the multidrug resistance (MDR) carrier
system. Therefore, in experiments with cultured cells, effects of aldehyde inhibitors
can be rapidly reversed by removal of the inhibitors. Although MG132 (1) is a potent and selective inhibitor of the chymotrypsin-like activity of the proteasome,
it is not suitable for use as a therapeutic agent.
Tyropeptin A (4) was isolated from the culture broth of Kitasatospora sp. [13], [14] and contains an aldehyde moiety at the C-terminal. This compound inhibits the chymotrypsin-like and trypsin-like activities
of the proteasome with IC50 values of 0.1 and 1.5 µg/mL, respectively, but scarcely inhibits the caspase-like
activity. In rat pheochromocytoma (PC12) cells, 4 induced neurite outgrowth, and this compound is thought to be capable of permeating
cells and inhibiting intracellular proteasomal activity [15]. Momose et al. designed and synthesized several derivatives of 4 [16]. Among them, TP-104 (5) is 20-fold more potent than 4 in terms of its inhibitory effect on the chymotrypsin-like activity of the mammalian
20S proteasome, while TP-110 (6) specifically inhibits the chymotrypsin-like but not the trypsin-like or caspase-like
activity of the proteasome.
Fellutamide B (7) was originally isolated from the marine-derived Penicillium fellutanum as a cytotoxic compound [17] and found to induce the release of nerve growth factor (NGF) from fibroblasts and
glial-derived cells [18]. Because of its structural similarity with 1, 7 was tested for inhibitory activity against the proteasome and found to potently inhibit
the chymotrypsin-like activity with an IC50 value of 9.4 nM along with the trypsin-like and caspase-like activities albeit less
potently with IC50 values of 2.0 and 1.2 µM, respectively [19]. In addition, 7 increased NGF gene transcription [19]. Therefore, it can be inferred that 7 exerts neurotrophic activity through stabilization of a short-lived unidentified
transcription factor that upregulates NGF gene expression.
Proteasome inhibitors: peptide boronates
The development of a proteasome inhibitor suitable for therapeutic use without the
defects of MG132 (1) led to the introduction of boronic acid as a functional group ([Fig. 3]). Although boronic acid-containing inhibitors bind with the hydroxy group of the
N-terminal threonine residue in the proteasome via a non-covalent bond, their inhibitory
potency and selectivity towards the proteasome are excellent in comparison with other
proteasome inhibitors. Since the boron atom is able to receive the oxygen lone pair
of the N-terminal threonine residue of the proteasome, inhibitors containing boronate can
form a stable tetrahedral intermediate ([Fig. 3]). Although peptide aldehydes are readily oxidized into inactive acids, boronates
are not inactivated by oxidation and are not rapidly removed from the cell by the
MDR system. In 2003, bortezomib (2), a synthetic boronate-containing inhibitor, was first approved by the FDA for treating
relapsed multiple myelomas [20], [21]. This inhibitor selectively inhibits the proteasome in a reversible manner and induces
apoptosis of malignant cells through the inhibition of NF-κB signaling and stabilization
of proapoptotic proteins. Clinical trials show promising results for the combination
of 2 and DNA cross-linking agents in the treatment of myeloma and ovarian cancer [22].
CEP-18770 (8), a synthetic boronate derivative, is an orally active proteasome inhibitor [23], which inhibits the chymotrypsin-like activity of the proteasome at a low nanomolar
concentration. This inhibitor exhibits the concentration-dependent induction of apoptotic
cell death in human multiple myeloma and tumor-derived cell lines, and exhibits a
more favorable cytotoxicity profile toward normal human endothelial cells, bone marrow
progenitors, and bone marrow-derived stromal cells than 2 [24]. Recently, it has been reported that 8 enhances the antimyeloma activity of 2 and melphalan in the xenograft models [25]. The in vitro and in vivo antitumor and anticlastogenic pharmacological profiles of 8 and its reduced cytotoxicity against a variety of normal human cell lineages compared
with tumor cells provide the rationale for further studies evaluating its preclinical
and clinical efficacies in multiple myeloma and other hematological malignancies.
At present, 8 is undergoing phase I and II trials.
Proteasome inhibitors: β-lactones
Lactacystin (9) ([Fig. 4]), the first natural proteasome inhibitor, was originally isolated from the culture
broth of Streptomyces sp. as an inducer of neurite outgrowth in a murine neuroblastoma cell line, Neuro-2a
[26], [27]. Subsequently, this inhibitor was found to inhibit proteasomal activity by binding
with the N-terminal threonine residue in the proteasome via a stable covalent bond ([Fig. 4]) [28]. The active component of 9 is clasto-lactacystin β-lactone (omuralide, 10), which is derived from 9 by elimination of the N-acetyl cysteine moiety to form a lactone ring [29], [30]. The cocrystallization of 9 with the proteasome revealed that the side chain residues of 10 are closely related to the selectivity of the inhibitor. PS-519 (11), which has an n-propyl group instead of the methyl group found in omuralide, is the most clinically
advanced lactacystin analogue [31] and has been used in a clinical trial to treat acute stroke.
Salinosporamide A (3) was isolated from the culture of a marine bacterium of the new genus Salinispora tropica [10], [32] and showed strong cytotoxic activity against HCT-116 human colon carcinoma. Because
of its structural similarity to 10, 3 was tested for inhibitory activity against the proteasome and found to inhibit the
chymotrypsin-like activity 35 times more potently than 10. The dechloro derivative, salinosporamide B (12), is ten times less potent, suggesting that the chloro substituent in 3 is important for its inhibitory ability [33]. Several potent salinosporamide analogues have been chemically synthesized, bioengineered,
and isolated from microorganisms. Antiprotealide (13), which was originally synthesized as a molecular hybrid of 3 and 10 [34], was recently isolated from a large-scale fermentation of the bacterium [35]. Fluorosalinosporamide (14), a fluoro analogue, was biosynthesized and showed reversible binding ability toward
the proteasome with an IC50 value of 1.5 nM (the value of salinosporamide A is 0.7 nM) [36]. Cinnabaramide A (15), an analogue with a structural difference in the alkyl side chain, was isolated
from a terrestrial streptomycete and showed inhibition with an IC50 value of 1 nM [37]. Although bortezomib (2) inhibits most strongly the chymotrypsin-like activity and moderately the caspase-like
activity of the proteasome [38], [39], [40], 3 inhibits most strongly the chymotrypsin-like activity and moderately the trypsin-like
activity of the proteasome [40]. A phase I study with 3 is currently proceeding. Although 2 is effective for the treatment of relapsed or refractory multiple myeloma, its prolonged
use can be associated with toxicity and the development of drug resistance. Importantly,
3 overcomes resistance to conventional treatment and 2 [41], [42]. In vivo studies using human multiple myeloma xenografts revealed that 3 shows well tolerated/prolonged survival and reduces tumor recurrence [42]. In addition, 3 was a more effective proapoptotic agent than 2 in isolated chronic lymphocytic leukemia cells [43], and the combination of 3 and a histone deacetylase inhibitor, MS-275 or valproic acid, induced greater cell
death than did the combination of 2 and these inhibitors [44]. It should be noted that 3 is orally bioavailable and cytotoxic to multiple myeloma cells with reduced toxicity
against normal cells compared to 2. Furthermore, in vivo studies with a human plasmacytoma xenograft mouse model showed that a low dose combination
of 3 and 2 is well tolerated and triggers synergistic inhibition of tumor growth and chymotrypsin-like,
caspase-like, and trypsin-like proteasomal activities in tumor cells [45]. Combination of 3 and lenalidomide (Revlimid®) induces synergistic anti-multiple myeloma activity in
multiple myeloma cell lines or patient multiple myeloma cells [46], implying the preclinical rationale for clinical protocols evaluating lenalidomide
together with 3 to improve patient outcome in multiple myeloma. The preclinical and clinical results
with 3 strongly suggest that the discovery of 3 indicates the importance of metabolites from marine microorganisms for drug discovery
and development [47].
Belactosins A (16) and C (17), which were isolated from a Streptomyces sp., contain the same β-lactone ring as 10 and inhibit the chymotrypsin-like activity of the proteasome in the same fashion
as 10 [48], [49]. Derivatives of 16 with more potent inhibitory effects than 16 have been synthesized [50]. Homobelactosin C (18), a modified derivative, has an IC50 value in the low nanomolar range [51]. The structural data for the complex of the proteasome and 18 provide an explanation for the involvement of immunoproteasome subunits in the generation
of antigen, and open the way for the rational design of compounds that exclusively
inhibit constitutive proteasomes or immunoproteasomes [51]. Recently, analogues of 17 were synthesized, and their inhibitory activity was investigated [52]. Among them, its two boronate-containing peptide analogues (19 and 20) showed significant inhibition of the chymotrypsin-like activity of the 20S proteasome
with IC50 values of 0.28 and 0.51 µM, respectively. Furthermore, the development of potent
proteasome inhibitors based on a stereochemical diversity-oriented strategy with 16 and its stereo- and regioisomers is underway [53].
Proteasome inhibitors: epoxyketones
The proteasome inhibitor epoxomicin (21) ([Fig. 5]) was originally isolated from an unidentified actinomycete strain and exhibited
in vivo antitumor activity against B16 melanoma [54]. This inhibitor contains an α,β-epoxyketone moiety that is involved in the formation
of a morpholino adduct with the N-terminal threonine residue in the proteasome, which results in inactivation of the
proteasome ([Fig. 5]) [55]. Although clinical studies with bortezomib (2) have validated the proteasome as a therapeutic target for the treatment of multiple
myeloma and some forms of non-Hodgkin's lymphoma [56], significant toxicity against normal cells has restricted the dosage. Furthermore,
many patients have tumors that do not respond to 2 and others develop resistance. This has led to the need for other proteasome inhibitors
with enhanced activity. Carfilzomib (PR-171, 22), which was derived from 21, is an irreversible proteasome inhibitor [57]. In multiple myeloma cells, 22 specifically inhibits the chymotrypsin-like activities of the proteasome and immunoproteasome
[58]. In comparison to 21, 22 exhibits greater selectivity toward the chymotrypsin-like activity of the proteasome
and is active against 2-resistant multiple myeloma cell lines. Since 22 also overcomes resistance to conventional agents and acts synergistically with dexamethasone
to enhance cell death, this compound is currently under evaluation in phase I clinical
trials in patients with multiple myeloma and non-Hodgkin's lymphoma. The first phase
I study of 22 shows that the drug is well tolerated, and produces signals of activity in patients
with multiple myeloma and that 22 did not produce any neuropathy [59]. These data support the further development of 22 in patients with hematologic malignancies.
Fig. 5 Structures of proteasome inhibitors containing epoxyketones and their inhibitory
mechanism. Note that carfilzomib (22) is in clinical trials for cancers.
Proteasome inhibitors: macrocyclic vinyl ketones
Syringolin A (23) ([Fig. 6]) was originally isolated as a virulence factor of the plant pathogen Pseudomonas syringae pv. syringae [60] and found to induce a change in the gene expression profile similar to that in cells
treated with proteasome inhibitors [61]. The crystal structure of the yeast proteasome in a complex with 23 revealed that the hydroxy group of the catalytic threonine residue exhibits a Michael
type 1,4-addition to the vinyl ketone moiety in the 14-membered ring of 23 ([Fig. 6]) [61]. This is a new mode of inhibition, and 23 irreversibly inhibits all three types of proteasomal activity. Recently, synthetic
and structural analyses of 23 revealed critical determinants of the selectivity and potency of its inhibitory effect
[62]. Furthermore, it was found that a rhodamine-tagged 23 selectively binds to and labels the active sites of the proteasome at therapeutic
concentrations [63]. In the same fashion, glidobactin A (24), another microbial metabolite [64], inhibits the chymotrypsin- and trypsin-like activities of the proteasome and reacts
with the threonine residues of the respective active sites [61]. Both 23 and 24 inhibit the proliferation and induce the apoptosis of malignant cells [61].
Fig. 6 Structures of proteasome inhibitors containing macrocyclic vinyl ketones and their
inhibitory mechanism.
Proteasome inhibitors: cyclic peptides
TMC-95A (25) ([Fig. 7]) was isolated from a fermentation broth of Apiospora montagnei [65], [66] by direct screening on the basis of inhibitory activity against the proteasome.
This compound inhibits the chymotrypsin-like, trypsin-like, and caspase-like activities
of the proteasome with IC50 values of 5.4, 200, and 60 nM, respectively. The cocrystallization of 25 with the yeast proteasome revealed that 25 is bound to the core particle of the proteasome through specific hydrogen bonds and
specifically blocks the proteasomal active sites non-covalently [67], [68]. In addition, 25 was found to induce neurite outgrowth in rat PC12 cells [69].
Fig. 7 Structures of cyclic peptides as proteasome inhibitors.
Argyrin A (26) was originally isolated from the myxobacterium Archangium gephyra as an immunosuppressive cyclic peptide [70] and identified as a small molecule capable of promoting the accumulation of p27kip1, a cyclin-dependent kinase inhibitor, in cancer cells by a high-throughput whole-cell
assay [71]. It should be noted that the level of p27kip1, a tumor suppressor protein and a proteasome substrate, is often reduced in human
cancer cells and that the expression of a degradation-resistant p27kip1 mutant reduced the number of intestinal adenomatous polyps that developed into invasive
carcinomas. It was found that 26 inhibits most strongly the chymotrypsin-like activity, moderately the caspase-like
activity, and weakly the trypsin-like activity of the proteasome [71].
Other proteasome inhibitors
Various proteasome inhibitors isolated from natural sources are shown in [Fig. 8]. Gliotoxin (27) is a fungal epipolythiodioxopiperazine toxin and contains a heterobicyclic structure
with a disulfide bridge. This compound was originally identified as a potent inhibitor
of NF-κB activation in T and B cells [72]. Additional studies showed that 27 inhibits the chymotrypsin-like activity of the proteasome and that the disulfide
bridge is responsible for the inhibition [73].
Fig. 8 Structures of proteasome inhibitors other than six groups shown in [Figs. 2]–[7].
We performed a screening of marine organisms and marine-derived fungi based on the
inhibition of the chymotrypsin-like activity of the proteasome and isolated agosterol
derivatives [e.g., agosterol C (28)], polyhydroxysterols, from a marine sponge, Acanthodendrilla sp., as proteasome inhibitors [74]. Agosterols were originally isolated from the marine sponge Spongia sp. and found to reverse multidrug resistance in tumor cells [75], [76]. Among the agosterol derivatives, 28 most strongly inhibited the chymotrypsin-like activity of the proteasome with an
IC50 value of 10 µg/mL. Interestingly, they differ in their inhibitory potency despite
their structural similarity. We also isolated another proteasome inhibitor, secomycalolide
A (29), together with known compounds, mycalolide A (30) and 30-hydroxymycalolide A (31), from a marine sponge of the genus Mycale [77]. Among the mycalolides, 29 has the most potent effect on the chymotrypsin-like activity of the proteasome with
an IC50 value of 11 µg/mL.
Extensive searches for proteasome inhibitors in plants and foods have been performed.
Activity-guided fractionation of a chloroform-soluble extract of the leaves of Ormosia sumatrana led to the isolation of a new cerebroside, sumatranoside (32), as a proteasome inhibitor with an IC50 value of 30 µM [78]. Curcumin (33), the major active ingredient of turmeric (Curcuma longa) used in South Asian cuisine for centuries, markedly inhibited the chymotrypsin-like
activity of a purified rabbit 20S proteasome with an IC50 value of 1.85 µM and also inhibited the cellular 26S proteasome [79]. Four dietary flavonoids, apigenin (34), kaempferol (35), quercetin (36), and myricetin (37), inhibited the chymotrypsin-like activity of the 26S proteasome in intact Jurkat
T cells with IC50 values of 1, 11, 2, and 12 µM, respectively [80]. Genistein (38), a soy isoflavone [81], (−)-epi-gallocatechin gallate (EGCG) (39), a green tea polyphenol [82], withaferin A (40), a steroidal lactone from the medicinal plant “Indian Winter Cherry” (Withania somnifera) [83], and celastrol (41), a triterpene from the Chinese “Thunder of God Vine” (Tripterygium wilfordii) [84], also inhibited the chymotrypsin-like activity of a purified 20S proteasome with
IC50 values of 26, 0.30, 4.5, and 2.5 µM, respectively.
Targeting the Ubiquitin System
Targeting the Ubiquitin System
The ubiquitin-proteasome pathway consists of the ubiquitin system and the protein
degradation system (the 26S proteasome) (see [Fig. 1]). The former contains the ubiquitin-activating enzyme (E1), ubiquitin-conjugating
enzyme (E2) and ubiquitin ligase (E3), and catalyzes the ubiquitination of client
proteins. In addition to inhibitors targeting the proteasome, various inhibitors of
the ubiquitin system consisting of E1, E2 and E3 enzymes have been developed.
Ubiquitin ligase (E3) inhibitors
Among enzymes in the ubiquitin system, E3s are a large family that recognize huge
numbers of client proteins and target them for degradation [85], [86]. They are classified into three major groups, the RING, HECT, and U‐box families,
on the basis of their domain structures and substrate recognition mechanisms. As E3
definitively determines which client proteins are ubiquitinated, a specific inhibitor
against an E3 recognizing a key client protein could be a good lead for the treatment
of diseases associated with degradation of the key client protein. Among many E3s,
MDM2 (mouse double minute 2) or HDM2 (human double minute 2), a RING-type E3 for p53
protein, is frequently used as a target for inhibitor development [87]. Although HDM2 is normally expressed at a low level, it is overexpressed in a variety
of human cancers. On the other hand, p53, a tumor suppressor, induces growth arrest
and apoptosis upon activation by various stimuli such as DNA damage [88]. The crystal structure of the 109-residue amino-terminal domain of MDM2, which binds
to a 15-residue transactivation domain peptide of p53, revealed that MDM2 has a deep
hydrophobic cleft, to which the p53 peptide binds [89]. Therefore, targeting MDM2/HDM2 is a promising way to reactivate p53, inducing apoptosis
in human cancer cells. For example, nutlin-3 (42) ([Fig. 9]) was discovered by screening a chemical library as an MDM2 antagonist and found
to suppress tumor progression in nude mice bearing subcutaneous human cancer xenografts
[89]. The development of 42 strongly indicates that MDM2 antagonists would be promising candidates for leads
in the treatment of cancer. Recently, it has been reported that the combination of
42 with bortezomib (2) mediates additive cytotoxicity against 2-sensitive multuple myeloma cell lines and synergistic activity against epithelial
carcinoma cell lines and that non-genotoxic activation of the p53 pathway using 42 can sensitize epithelial carcinoma cells to 2 in a manner that is not suppressed by microenvironmental interactions, i.e., even
in the presence of stromal cells [90].
Fig. 9 Structures of ubiquitin ligase (E3) inhibitors.
Chlorofusin (43), the first MDM2 antagonist from natural sources, was isolated from the culture of
a Fusarium sp. with an IC50 value of 4.6 µM [91]. The absolute configuration of chlorofusin was determined by total synthesis [92], [93]. Furthermore, inhibition of MDM2-p53 binding was evaluated with seven chromophore
diastereomers of 43 [94]. The biosynthesis of 43 was also investigated [95].
By bioassay-guided isolation, (−)-hexylitaconic acid (44) was isolated from a culture of marine-derived fungus as an inhibitor of p53-HDM2
interaction [96]. Since 44 is unable to inhibit the interaction of p53 with COP1, another E3 of the HECT-type
for p53, it can be inferred that 44 binds to HDM2 protein. Recently, the stereogenic center of 44 was determined to be R by vibrational circular dichroism (VCD) spectroscopy [97]. Its synthetic S-(+)-enatiomer also had an inhibitory effect on the p53-HDM2 interaction, which was
comparable to that of the natural R-(−)-enatiomer (44) [97].
The interaction of chalcones [e.g., chalcone C (45)] with the p53-MDM2 system was analyzed by NMR spectroscopy (1H-15 N HSQC spectrum) [98], since chalcone derivatives were reported to inhibit tumorigenesis [99]. Chalcones were found to bind to a subsite of the p53-binding cleft of MDM2. On
the other hand, RITA (46) was discovered by screening a chemical library as an inhibitor of p53-MDM2 interaction
through its binding to p53 protein [100].
Ubiquitin-activating enzyme (E1) inhibitors
Ubiquitin is first activated by the ubiquitin-activating enzyme (E1) in the ubiquitin-proteasome
system ([Fig. 1]). In the E1-mediated ubiquitin-activation process, ubiquitin and ATP bind to different
sites in the E1 enzyme, and E1 catalyzes the formation of a ubiquitin-adenylate intermediate
from ubiquitin and ATP, and subsequently the binding of ubiquitin to a cysteine residue
at the E1 active site via a thiol ester linkage. Then, the high-energy intermediate
of ubiquitin is transferred to the thiol group of the active cysteine residue in the
ubiquitin-conjugating enzyme (E2) and then to the client proteins, mediated by ubiquitin
ligase (E3) [1], [5]. Since E1 activity is thought to be essential for the ubiquitin-proteasome system,
developing inhibitors against E1 is another possible route of drug development for
the treatment of cancer.
Two natural E1 inhibitors, panepophenanthrin (47) ([Fig. 10]) [101] and himeic acid A (48) [102], have been isolated from microorganisms. As the first natural E1 inhibitor, 47 was isolated from a mushroom strain, Panus rudis. This compound inhibits the formation of an E1-ubiquitin thioester intermediate with
an IC50 value of 17 µg/mL. On the other hand, during the screening of extracts of marine
organisms and marine-derived microorganisms, a culture of the fungus Aspergillus sp., isolated from the mussel Mytilus edulis galloprovincialis, showed strong E1 inhibitory activity, and bioassy-guided fractionation of the culture
afforded 48 as an E1 inhibitor [102]. This compound inhibited the formation of the E1-ubiquitin intermediate by 65 %
at a concentration of 50 µM and was found to inhibit the binding of ubiquitin to the
ubiquitin-binding site in the E1 enzyme. But two congeners, himeic acids B (49) and C (50), were inactive even at 100 µM. As 48 cannot inhibit E1-like enzymes for other ubiquitin-like modifiers, at least, SUMO-1
and ISG15, this compound could be a specific inhibitor of the ubiquitin E1 enzyme.
Fig. 10 Structures of ubiquitin-activating enzyme (E1) inhibitors.
Recently, PYR-41 (51), a synthetic pyrazone derivative, was identified using a commercial screening library
as a cell-permeable E1 inhibitor [103]. This compound blocks protein degradation and cytokine-induced activation of NF-κB,
activates p53 in cells, and preferentially kills cells transformed with wild-type
p53.
Ubiquitin-conjugating enzyme (E2) inhibitors
Ubiquitination performs proteolytic and non-proteolytic functions [104], [105]. The lysine48 (K48)-linked polyubiquitin chain is related to proteasome-dependent
protein degradation, while the K63-linked chain plays non-proteolytic roles in various
cellular events including signal transduction and DNA repair. The formation of the
latter chain is catalyzed by a hetero-dimer formed by the ubiquitin E2 enzyme Ubc13
and an inactive ubiquitin-conjugating enzyme variant (Uev1A or Mms2) [106], and a functional difference between the two Ubc13 complexes (Ubc13-Uev1A and Ubc13-Mms2
complexes) was suggested [107]. The report that the knockdown of Ubc13 led to an increase in p53 activity [108] led us to speculate that an inhibitor of Ubc13, that is, one preventing the formation
of the Ubc13-Uev1A complex, would be a lead for an anticancer agent. Subsequently,
we carried out a search for inhibitors of Ubc13-Uev1A interaction in natural resources
and found leucettamol A (52) ([Fig. 11]) in the marine sponge Leucetta aff. microrhaphis [109]. Although 52 was originally reported as an antimicrobial compound with a racemic nature [110], it was recently found that 52 is chiral with the configuration 2R,3S,28S,29R, as revealed by deconvoluted exciton coupled circular dichroism (ECCD) spectroscopy
[111]. Furthermore, the recent report that the association of p53 with Ubc13 on polysomes
requires ongoing translation and results in p53 ubiquitination which interferes with
its tetramerization [112] also supports the use of inhibitors of the formation of the Ubc13-Uev1A complex
in anticancer therapy.
Fig. 11 Structure of a ubiquitin-conjugating enzyme (E2) inhibitor.
In summary, targeting of the E1, E2, and E3 enzymes in the ubiquitin system offers
promising prospects for drug discovery.
Targeting the Delivery System Connecting the Ubiquitin System to the Proteasome
Targeting the Delivery System Connecting the Ubiquitin System to the Proteasome
As described above, the ubiquitin-proteasome system consists of two subsystems, the
ubiquitin system and the protein degradation system (the 26S proteasome). Recently,
a third system, the so-called delivery system, was proposed to function in the delivery
of ubiquitinated proteins to the 26S proteasome (see [Fig. 1]). The proteasome subunit Rpn10 functions as an intrinsic ubiquitin receptor of the
proteasome [113] and Rpn13 was recently found to be an additional intrinsic ubiquitin receptor [114], [115]. The delivery system consisting of intrinsic ubiquitin receptors Rpn10/Rpn13 and
extrinsic ubiquitin receptors such as Rad23 and Dsk2 functions as the third system
in the ubiquitin-proteasome pathway, although the mechanism for the discrimination
of ubiquitinated client proteins by the respective ubiquitin receptors remains unclear
[116], [117]. Therefore, specific rather than general targeting of the respective ubiquitin receptors
could be an effective approach in the treatment of specific client protein-associated
diseases.
Girolline (53) ([Fig. 12]) was originally isolated as an antitumor compound from a marine sponge, but a phase
I clinical study with this compound showed severe side effects in patients and no
apparent antitumor activity [118]. We found that 53 induces G2/M cell cycle arrest in several tumor cell lines. In addition, ubiquitinated
p53 but no other protein accumulated in cells treated with 53, but this compound does not inhibit proteasomal activity [119]. Although the exact target of 53 has not been identified, it was proposed that this compound inhibits the delivery
of ubiquitinated p53 protein to the proteasome.
Fig. 12 Structures of inhibitors targeting the delivery system connecting the ubiquitin system
to the 26S proteasome.
Ubistatin (54) was discovered by searching a chemical library for an inhibitor of destruction box-dependent
protein degradation [120]. It was found that 54 binds to the ubiquitin chain of ubiquitinated proteins and inhibits ubiquitin-dependent
proteolysis.
Thus, compounds inhibiting the delivery system for ubiquitinated proteins could serve
as novel inhibitors targeting the ubiquitin-proteasome system.
Targeting Deubiquitinating Enzymes
Targeting Deubiquitinating Enzymes
In the polyubiquitin chain, the C-terminal carboxyl group of one ubiquitin monomer is covalently linked with the ε-amino
group of the lysine residue in another ubiquitin monomer or the client protein via
an isopeptide bond. Upon degradation of the polyubiquitinated client protein by the
proteasome, the polyubiquitin chain is recognized by intrinsic ubiquitin receptors
of the 26S proteasome as well as extrinsic ubiquitin receptors in the third system
and is converted to ubiquitin monomers by the actions of intrinsic and extrinsic deubiquitinating
enzymes (DUBs) [121]. Thus, it is now known that deubiquitination mediated by the actions of various
DUBs plays important regulatory roles in various cellular events [122].
Prostaglandins (PGs) function as intracellular signal mediators in the regulation
of a variety of physiological processes, including inflammation and immune responses.
J series PGs [e.g., Δ12-PGJ2 (55)] ([Fig. 13]) with a unique exocyclic α,β-unsaturated ketone inhibit ubiquitin isopeptidase activity
in the proteasome-mediated proteolytic pathway, while PGA1, PGB1, PGE2, and 15-keto-PGE2 are inactive [123]. The former J series PGs cause apoptosis, independently of p53-mediated gene transactivation.
Punaglandins [e.g., punaglandin 4 (56)] isolated from the soft coral Telesto riisei are highly functional cyclopentadienone and cyclopentenone prostaglandins chlorinated
at the endocyclic α-carbon position, and inhibit ubiquitin isopeptidase activity and
exhibit antiproliferative effects more potently than J series PGs [124]. Recently, a small-molecule inhibitor (HBX 41,108, a cyano-indenopyrazine derivative,
57) of USP7/HAUSP, a member of the ubiquitin-specific protease (USP) family of DUBs,
was isolated by high-throughput screening [125]. It was found that the treatment of cancer cells with 57 results in the accumulation of p53 and that 57 inhibits cancer cell growth and induces apoptosis. This suggests that compounds inhibiting
specific DUBs could act as novel anticancer drugs. Although drug discovery targeting
DUBs is in its early stages, a better understanding of the DUBs with regard to their
mechanisms of action and substrate recognition may enable the development of small-molecule
inhibitors for effective anticancer treatment.
Fig. 13 Structures of deubiquitinating enzyme (DUB) inhibitors.
Future Perspectives
Future Perspectives
The ubiquitin-proteasome system controls a wide range of cellular events including
cell cycle progression, and defects associated with this system result in various
diseases including cancer and neurodegenerative disorders. Thus, the ubiquitin-proteasome
system is emerging as a significant target in anticancer therapies. Bortezomib (2), a synthetic proteasome inhibitor, is already on the market for the treatment of
patients with multiple myeloma and is also undergoing clinical trials for other cancers.
In preclinical studies, 2 showed antitumor activity against a variety of solid tumors, including breast, gastric,
colon, pancreas, and non-small lung cancers [126]. In addition, several natural and synthetic inhibitors targeting the proteasome,
salinosporamide A (3), CEP-18770 (8), and carfilzomib (22), are also in clinical trials for cancers. Inhibitors targeting the ubiquitin system,
the delivery system, and deubiquitinating enzymes are also candidates for anticancer
drugs and several compounds are now undergoing preclinical and clinical trials for
cancers. In 2004, Ciechanover, Hershko, and Rose received the Nobel Prize in Chemistry
for the discovery of ubiquitin-mediated protein degradation. Although the mechanisms
and functions of the ubiquitin-proteasome system have been investigated extensively,
a comprehensive understanding of the complex ubiquitin-proteasome system as well as
the development of inhibitors of this system by searching natural sources and chemical
libraries and also by chemical synthesis is needed to develop efficient anticancer
drugs in the future.
