Structure–Activity Relationships of Noncovalent Immunoproteasome β5i-Selective Dipeptides

Abstract

The immunoproteasome (i-20S) has emerged as a therapeutic target for autoimmune and inflammatory disorders and hematological malignancies. Inhibition of the chymotryptic β5i subunit of i-20S inhibits T cell activation, B cell proliferation, and dendritic cell differentiation in vitro and suppresses immune responses in animal models of autoimmune disorders and allograft rejection. However, cytotoxicity to immune cells has accompanied the use of covalently reactive β5i inhibitors, whose activity against the constitutive proteasome (c-20S) is cumulative with the time of exposure. Herein, we report a structure–activity relationship study of a class of noncovalent proteasome inhibitors with picomolar potencies and 1000-fold selectivity for i-20S over c-20S. Furthermore, these inhibitors are specific for β5i over the other five active subunits of i-20S and c-20S, providing useful tools to study the functions of β5i in immune responses. The potency of these compounds in inhibiting human T cell activation suggests that they may have therapeutic potential.

Immunoproteasome and Non-Covalent Inhibition: Exploration by Advanced Molecular Dynamics and Docking Methods

“…Promising immunoproteasome inhibitors, both covalent and non-covalent, have been recently identified. In covalent inhibitors, the presence of a reactive warhead may cause significant off-target activities against other proteins, which may result in side effects (e.g., liver toxicity and idiosyncratic adverse reactions) and reduced selectivity over time [ 8 , 9 ]. For these reasons, the attention was focused on non-covalent immunoproteasome inhibitors.…”

“…Non-covalent inhibition is therefore strongly desirable, because it is free of these disadvantages. Lacking a reactive warhead, non-covalent inhibitors may have an improved selectivity and less reactivity and instability and, therefore, may not exhibit the side effects that occur in covalent inhibitor therapies (e.g., liver toxicity and idiosyncratic adverse reactions) [ 8 , 9 ]. Furthermore, the enzyme–inhibitor complexes have reduced lifetimes, and this promotes an extensive tissue distribution of the drug [ 10 ].…”

A Nut for Every Bolt: Subunit-Selective Inhibitors of the Immunoproteasome and Their Therapeutic Potential

“…Although DPLG3 should not target subunits other than β5i at the applied concentrations, in vivo inhibition values have not been reported and so it remains to be further investigated whether single-subunit inhibition is really sufficient for the observed therapeutic effects [47]. Due to its adverse physicochemical properties, i.e., high hydrophobicity and poor water solubility, as well as poor cell penetration ability, efforts to improve DPLG3 were undertaken and afforded several derivatives, among them PKS21272 with 500-fold β5i-selectivity and IC 50 values of 0.0012 µM for β5i and 0.61 µM for β5c, as well as < 50% inhibition of other subunits at 33.3 µM [104]. This compound is cell-penetrating and inhibits proteasomes in vivo but at lower selectivity (81-fold selectivity for iCP) and potency (0.053 µM).…”

“…This compound is cell-penetrating and inhibits proteasomes in vivo but at lower selectivity (81-fold selectivity for iCP) and potency (0.053 µM). In addition, PKS21272 can inhibit activation and proliferation of T cells [104]. Notably, derivatives of DPLG3 are claimed to be 'active site-directed non-competitive inhibitors' [64,104], meaning that their inhibition potency is not affected by accumulating substrate concentrations.…”

“…In addition, PKS21272 can inhibit activation and proliferation of T cells [104]. Notably, derivatives of DPLG3 are claimed to be 'active site-directed non-competitive inhibitors' [64,104], meaning that their inhibition potency is not affected by accumulating substrate concentrations. Considering their peptidic or peptide-like structure and their binding to the natural β5-substrate binding pockets [64], a non-competitive mode of action is questionable.…”