Proteasome Inhibitors

Christopher J. Kirk

Brian B. Tuch

Shirin Arastu-Kapur

Lawrence H. Boise

BIOCHEMISTRY OF THE UBIQUITIN-PROTEASOME PATHWAY

The ubiquitin proteasome system is involved in the degradation of more than 80% of cellular proteins, including those that control cell-cycle progression, apoptosis, DNA repair, and the stress response.¹ A key step in this process is the tagging of proteins targeted for degradation with multiple copies of ubiquitin, a 76-amino acid protein whose primary sequence and structure is highly conserved in organisms ranging from yeasts to mammals.²^,³ Once polyubiquitinated, proteins targeted for degradation bind to the 26S proteasome, a holoenzyme composed of two 19S regulatory complexes capping a central 20S proteolytic core. The 20S core is a hollow “barrel” consisting of four stacked heptameric rings. The subunits of the rings are classified as either β subunits (outer two rings) or β subunits (inner two rings). The 19S regulatory complex consists of a lid that recognizes ubiquitinated protein substrates with high fidelity, and a base that contains six adenosine triphosphatases, unfolds protein substrates, removes the polyubiquitin tag, and threads them into the catalytic chamber of the 20S particle in an adenosine triphosphate-dependent manner.⁴^,⁵ Unlike typical proteases, the 20S proteasome in eukaryotic cells contains multiple proteolytic activities resulting in the cleavage of protein targets after many different amino acids. In most cells, the 20S core particle contains the catalytic subunits β5 (PSMB5), β1 (PSMB1), and β2 (PSMB2), accounting for chymotrypsinlike (CT-L), caspaselike (C-L), and trypsinlike (T-L) activities, respectively, each differing in their substrate preference.⁶ However, in cells of hematopoietic origin, such as lymphocytes and monocytes, the proteasome catalytic subunits are encoded by homologous gene products: LMP7 (PSMB8), LMP2 (PSMB9), and MECL-1 (PSMB10).⁷ These immunoproteasome subunits are also induced in nonhematopoietic cells following exposure to inflammatory cytokines such as interferon-γ (IFN-γ) and tumor necrosis factor alpha (TNF-α).⁸ In the immunoproteasome, the 19S regulatory complex can be replaced with proteasome activators such as PA28, whose expression is also induced in cells following exposure to IFN-γ. Hybrid proteasomes, both for the catalytic subunits and regulatory particles, have been described.⁹

Given its key role in maintaining cellular homeostasis, the ubiquitin proteasome system appeared to be an unlikely target for pharmaceutical intervention. However, a variety of groundbreaking studies in the 1990s suggested that inhibitors of proteasome function might prove to be viable therapeutic agents.¹⁰ Initial studies used substrate-related peptide aldehydes to investigate the proteolytic functions and specificity of the proteasome.¹¹ In vitro and in vivo studies with these inhibitors demonstrated their ability to induce apoptosis as well as inhibit tumor growth.¹²^,¹³^,¹⁴^,¹⁵ It was subsequently discovered that several natural products with antitumor activity exert their action via proteasome inhibition, providing additional rationale for the development of selective proteasome inhibitors (PIs).¹⁶^,¹⁷

PROTEASOME INHIBITORS

Chemical Classes of Proteasome Inhibitors in Clinical Development

As of the writing of this overview, six different proteasome inhibitors comprising three distinct chemical classes have been tested in clinical trials (Table 24.1) and include: (1) dipeptide boronic acids, (2) peptide epoxy ketones, and (3) β-lactones.¹⁸^,¹⁹ Bortezomib (PS-341, Velcade), a dipeptide boronic acid, was developed by Millennium Pharmaceuticals (Cambridge, MA) and was the first PI approved for clinical use.²⁰ Two additional dipeptide boronic acids have entered clinical development, ixazomib/MLN 9708 (Millennium), currently in phase III studies, and delanzomib/CEP-18770 (Teva Pharmaceuticals; Frazer, PA), the clinical development of which has been halted. Carfilzomib (Onyx Pharmaceuticals; San Francisco, CA), a tetrapeptide epoxy ketone, received U.S. Food and Drug Administration (FDA) approval in 2012.²¹ A second peptide epoxy ketone proteasome inhibitor, oprozomib (Onyx), entered clinical study in 2010. The third class of proteasome inhibitors, β-lactones, is represented by NPI-0052 (salinosporamide A [Marizomib]) and is currently being developed by Nereus Pharmaceuticals, Inc. (San Diego, CA). The initial approvals for both bortezomib and carfilzomib were in multiple myeloma (MM), a plasma cell neoplasm and the second most common hematologic cancer. However, the activity of PIs in other B-cell neoplasms has resulted in an expansion of the clinical utilization of this drug class.

Preclinical Activity of Proteasome Inhibitors

Each of the three classes of inhibitors has a distinct chemical mechanism of proteasome inhibition.²² Peptide boronates form stable but reversible tetrahedral intermediates with the γ-hydroxyl (γ-OH) group of the catalytic N-terminal threonine of the proteasome active sites.²³^,²⁴ β-lactones also interact with this γ-OH, but form a completely irreversible interaction.²⁵ Similarly, peptide epoxy ketones form irreversible covalent adducts with the active site threonine but do so via a dual covalent adduction of γ-OH group and the free amine.²⁶ This interaction is highly specific for N-terminal threonine-containing hydrolases and renders peptide epoxy ketones the most selective proteasome inhibitors yet described.²⁷^,²⁸

The primary targets of these PIs within the constitutive and immunoproteasomes are the CT-L subunits, β5 and LMP7, respectively. Despite accounting for less than 50% of total protein turnover by the proteasome, these subunits are essential for cell survival.²⁹ In MM cell lines, inhibiting both subunits (β5 and LMP7) is necessary and sufficient for tumor cell death.³⁰ Cytotoxicity of other tumor cell types requires the inhibition of multiple active sites beyond the CT-L activity. The combination of inhibitors specific for either the T-L or C-L activities, which have no cytotoxic activity on their own, augments the cytotoxic potential of the CT-L-specific inhibitors.³¹^,³²

TABLE 24.1 Proteasome Inhibitors in Clinical Development

Agent	Other Names	Drug Class	Stage of Development	Tumor Types	Route of Administration	Dose Levels	Schedule of Administration
Bortezomib	Velcade PS-341	Peptide boronate	FDA/EMEA approved	Multiple myeloma, mantle cell lymphoma	Intravenous, subcutaneous	1.3 mg/m²	Days 1, 4, 8, & 11 (21-day cycle)
Ixazomib	MLN 9708 MLN 2238	Peptide boronate	Phase III	Multiple myeloma, AL, amyloidosis	Oral	4 mg	Once weekly (21-day cycle)
Delanzomib	CEP-18770	Peptide boronate	Phase I (discontinued)	Multiple myeloma	Intravenous	0.1-1.8 mg/m²	Days 1, 4, 8, & 11 (21-day cycle)
Carfilzomib	Kyprolis PR-171	Peptide epoxy ketone	FDA approved, phase III	Multiple myeloma	Intravenous	20/27 mg/m²	Days 1, 2, 8, 9, 15, & 16 (28-day cycle)
Oprozomib	ONX 0912 PR-047	Peptide epoxy ketone	Phase I/II	Multiple myeloma	Oral	150-240 mg (dose escalation ongoing)	Days 1, 2, 8, & 9 (14-day cycle) Days 1-5 (14-day cycle)
Marizomib	NPI-0052 Salinosporamide A	β-lactone	Phase II	Multiple myeloma	Intravenous	0.075-0.6 mg/m²	Days 1, 4, 8, & 11 (21-day cycle)
EMEA, European Medicines Agency; AL, amyloid light chain.

Given its status as the first proteasome inhibitor approved for marketed use, the antitumor potential and preclinical activity of other proteasome inhibitors have generally been compared to bortezomib.¹⁹ Carfilzomib showed equivalent antitumor activity to bortezomib in vitro against a panel of tumor cell lines under standard culture conditions but was >10-fold more potent at inducing tumor cell death when cells were exposed to drug for a 1-hour pulse, which mimics the pharmacokinetics of both compounds.³³ MLN2238 (the active agent of ixazomib) was active in the same mouse models of human tumors as bortezomib, but demonstrated greater levels of proteasome inhibition in the tumors.³⁴ In biochemical assays of proteasome activity, delanzomib had an identical potency and subunit activity profile to bortezomib, but in tumor cytotoxicity assays, potency relative to bortezomib was 2- to 10-fold less.³⁵ In addition, delanzomib appeared to be less cytotoxic than bortezomib to normal cells and had a differential effect on cytokine release in bone marrow stromal cells, suggesting a different pharmacologic activity. Oprozomib is 10-fold less potent than carfilzomib in proteasome activity assays, but showed similar antitumor activity in mouse tumor models.³⁶^,³⁷ Marizomib displayed greater potency against the non-CT-L active sites of the proteasome than bortezomib.³⁸ Interestingly, this agent synergized with bortezomib in killing tumor cells in vitro.³⁹ All of the secondgeneration inhibitors have shown activity in tumor cells made resistant to bortezomib and/or MM cells isolated from patients relapsed from bortezomib-based therapies³⁵^,³⁶^,⁴⁰^,⁴¹^,⁴²

The inhibition of tumor cells with proteasome inhibitors induces cell death via the induction of apoptosis through death effector caspase activation.¹⁰ Although the mechanism underlying the induction of cell death remains to be fully elucidated, extensive research suggests a complex interplay of multiple pathways. PIs have been shown to affect the half-life of the BH3-only members of the Bcl-2 family, specifically BH3-interacting-domain death agonist (Bid) and Bcl-2 interacting killer (Bik).⁴³ Moreover the BH3-only protein NOXA is upregulated at the transcription level by PIs.⁴⁴^,⁴⁵^,⁴⁶^,⁴⁷^,⁴⁸ Proteasome inhibition also upregulates the expression of several key cell-cycle checkpoint proteins that include p53 (an inducer of G0/G1 cell-cycle arrest through accumulation of the cyclindependent kinase [CDK] inhibitor p27); the CDK inhibitor p21; mammalian cyclins A, B, D, and E; and transcription factors E2F and Rb.⁴⁹^,⁵⁰ The transcription factor nuclear factor kappa B (NF-κB), an important regulator of cell survival and cytokine/growth factor production,⁵¹ is also affected by proteasome inhibition in multiple ways. The net effect on NF-κB signaling is not consistent across various assays and cell lines, and its relative importance in the antitumor effects of PIs remains unclear. Although it is interesting to note that patients whose myeloma harbor NF-κB-activating mutations (˜20%) respond better to bortezomib than those without NF-κB-activating mutations.⁵²^,⁵³^,⁵⁴ In MM cell lines, there is growing evidence that the major determinant of sensitivity to proteasome inhibition is the relative load of protein flux to the proteasome.⁵⁵^,⁵⁶^,⁵⁷ These data suggest that induction of the terminal unfolded protein response may drive cell death. Whether proteotoxic stress induced cell death reflects sensitivity to proteasome inhibitors in other tumor types remains to be determined.

Pharmacokinetics and Pharmacodynamics of Proteasome Inhibitors in Animals

Following intravenous (IV) administration to animals and humans, proteasome activity is inhibited in a dose-dependent fashion within minutes; however, PIs such as bortezomib and carfilzomib are also rapidly cleared from circulation.⁵⁵^,⁵⁶^,⁵⁸^,⁵⁹^,⁶⁰^,⁶¹ Recovery of proteasome activity in animals occurs in tissues with a half-life of approximately 24 hours, mirroring the recovery time of cells exposed to sublethal concentrations of PIs in vitro and likely reflecting new protein synthesis.³³^,⁶²

Only gold members can continue reading. Log In or Register to continue