Tripeptide epoxyketones with a linear hydrocarbon chain at the P3 position have higher specificity toward immunoproteasome catalytic subunits than normal tetrapeptide epoxyketones such as epoxomicin.

(A) LMP7 and X as well as LMP2 protein bands are competed away by dihydroeponemycin (2) or its analog (9) on western blot. (B) Proteasome subunit (LMP7/X) bands are efficiently competed awayby excess epoxomicin or dihydroeponemycin on western blot. EL4cells were preincubated with proteasome inhibitors for 30 min beforebeing treated with assay probes (biotin-epoxomicin or biotin-dihy-droeponemycin). After 1 hr of incubation, cells were lysed and analyzed by western blot by using streptavidin-HRP and ECL.

Reagents and conditions for 12. (a) 2-Methoxyethoxymethyl Chloride, i-Pr₂EtN, CH₂Cl₂, 0°C → rt; (b) Benzonitrile, H₂O₂, i-Pr₂EtN, MeOH, 0°C, 3 hr; (c) 5, TFA, CH₂Cl₂, 30 min; (d) HBTU, HoBt,i-Pr₂EtN, CH₂Cl₂, rt, 12 hr; (e) TBAF, THF, 1 hr. *(e) precedes (d): 7a was deprotected with TBAF before it was coupled to 7.

(A–C) EL4 cells were preincubated with compounds before the treatment of assay probes for visualization of proteasome subunits that are not targeted by compounds 12 and 15. (A) Compounds 10, 11, and 13 nonselectively target proteasome subunits, whereas compound 14 does not bind proteasome subunits. (B) Compounds 12 and 15 selectively target the immunoproteasome subunit LMP2. (C) Compounds 12 and 15 do not target other catalytic subunits (LMP7, X, Z, and MECL1).
(D) Cells were incubated for 1 hr before SDS-PAGE and western blot analysis with anti-LMP7, X, and Y antibodies. Compound 15 covalently modifies the immunoproteasome subunit LMP2, but not other subunits in EL4 cells.
Biotin-EPN, Biotinylated dihydroeponemycin; Biotin-EPX, Biotinylated epoxomicin.

(A) Inhibition of chymotrypsin-like activity of the immunoproteasomes by compound 15. Proteasome kinetic studies were performed with purifiedimmunoproteasome. Purified immunoproteasome was preincubated with lactacystin, compound 15, or cotreatment for 30 min before CT-L fluoro-genic substrate was added (Suc-Leu-Leu-Val-Tyr-AMC). Data were collected over a 1.5 hr period at room temperature. For compound 15, K_obs/[I] (M⁻¹S⁻¹) = 83 ± 27. The range of inhibitor concentrations used was 5-20 μM.
(B) Compounds 15 and 12 do not inhibit the constitutive proteasome. Human endothelial cell spheroids were seeded in collagen I gels in a 96-wellplate and were stimulated with vascular endothelial growth factor (VEGF; 20 ng/ml) to induce angiogenic sprouting. In replicate wells, VEGF-treatedspheroids were coincubated with epoxomicin (Epox), dihydroeponemycin (Epn), compound 15, or compound 12. Representative photographicimages of spheroids taken after 24 hr show invasive growth of vessel structures with VEGF alone and compound 15 cotreatment or compound 12 cotreatment, but a potent inhibitory effect with dihydroeponemycin (0.2 μM and 0.5 μM) or epoxomicin (0.2 μM) cotreatment.

(A) LMP2 expression levels of prostate cancer cells.
(B and C) PC3 cells were preincubated with compounds before the treatment of biotinylated dihydroeponemycinto visualize biotinylated proteasome subunits that are not targeted by compounds 12 and 15. (B) Compounds 12 and 15 selectively target the immunoproteasome subunit LMP2 in PC3 prostate cancer cells. (C) Compounds 12 and 15 do not compete with epoxomicin or dihydroeponemycin for binding Z, MECL1, LMP7, and X sub-units.
(D) Compound 15 covalently modifies the immunoproteasome subunit LMP2 in PC3 prostate cancer cells. Cells were incubated with compound 15, biotin-epoxomicin, or biotin-dihydroeponemycin for 1.5 hr. The mobility shift of the LMP2 subunit was analyzed by western blotting with anti-LMP2 antibody.
DPN, dihydroeponemycin; EPX, epoxomicin.