Proteolysis-targeting chimera (PROTAC) is a novel technology that can hijack cellular degradation machinery and degrade target proteins. It is a heterobifunctional small molecule consisting of three components: a molecule that binds to a protein of interest (POI), an E3 ligase-recruiting ligand, and a linker for connecting the two moieties.¹⁾ PROTACs have several advantages over traditional inhibitors like unique mechanism of action (MOA) to degrade POI rather than inhibition, targeting the undruggable target protein, overcoming the drug resistance, and improved selectivity profiles.^2-4) Having such a great potential, only a limited number of ubiquitin E3 ligase binders (VHL, CRBN, MDM2, cIAP1, DCAF15, RNF4, RNF114) (Fig. 1) have been used in the targeted protein degradation (TPD).^5-8) It has also been confirmed that not every POI-E3 ligase combination can form a ternary complex, resulting in the failure of the ubiquitin-proteasome system (UPS).^{9, 10)} This clearly indicates the need for a novel E3 ligase ligand for TPD.

Fig. 1. Structures of E3 ligase binder for PROTACs technology.
(A) VHL (B) CRBN (C) MDM2 (D) cIAP1 (E) DCAF15 (F) RNF4 (G) RNF114

Kelch-like ECH-associated protein 1 (KEAP1) is a master regulator of oxidative and electrophilic stress. It acts as an adaptor protein for the Cullin 3 (CUL3) E3 ligase complex, which has three modules: the CUL3 scaffold protein, RING-box protein 1 (RBX1), and a bric-a-brac tramtrack board (BTB). KEAP1 ubiquitinates the nuclear factor erythroid 2–related factor 2 (Nrf2), a key transcription factor responsible for the expression of antioxidant response elements (ARE) to protect cells from oxidative damage. Under oxidative stress, KEAP1 cysteine residues are chemically modified, resulting in the dissociation of CUL3 and other conformational changes, leading to the release of Nrf2, which translocates to the nucleus and stimulates the expression of the ARE-directed gene.^11-13) Nrf2 activation has been reported to exhibit protective activity against diseases such as cancer, cardiovascular diseases, inflammation, neurodegenerative diseases, respiratory disorders, liver diseases, and diabetes.^14-20)

Targeting the KEAP1-Nrf2 pathway has become a hot topic in the development of a plethora of inhibitors.²¹⁾ Lu et al. designed and synthesized a Nrf2 ETGE motif peptide-based KEAP1 recruiting degrader which degrades tau protein via the UPS pathway.²²⁾ Due to the peptide-based nature of the degrader, there are some limitations to its application. Tong et al. reported bardoxolone as a covalent ligand of KEAP1, where a linker connected to JQ1 (a POI ligand) enabled degradation of bromodomain-containing protein 4 (BRD4). Nevertheless, bardoxolone is a non-selective KEAP1 ligand that induces the self-degradation of KEAP1 via an unknown mechanism.²³⁾ Wei et al. and Du et al. also synthesized KEAP1 E3 ligase-based PROTACs, depicting the degradation of the target protein BRD4.^{24, 25)} This ensures that expanding the toolbox of diverse E3 ligase recruiters will significantly contribute to the TPD platform.

To demonstrate our proof-of-concept, we adapted a selective, potent inhibitor of KEAP1 as an E3 ligase recruiter for degrading the BRD4 protein by synthesizing KEAP1 BRD4 targeting PROTAC compounds, considering JQ1 as a POI ligand with a flexible aliphatic linker to connect the two moieties. SD-2406 degraded BRD4 in an orderly manner, reinforcing our concept and thereby exploiting the limited E3 ligase toolbox for TPD.

Cell culture and reagents

LNCaP cells were purchased from the Korean Cell Line Bank (KCLB, Seoul, Republic of Korea). Cells were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (HyClone, South Logan, Utah, USA) supplemented with 10 % fetal bovine serum at 37 °C in a humidified incubator containing 5% CO₂. Anti-BRD4 antibody was purchased from Abcam (Cambridge, UK). The anti-MYC antibody was purchased from ABclonal Science (Woburn, MA, USA). Anti-β-actin antibody was purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Anti-rabbit and anti-mouse secondary antibodies were purchased from Cell Signaling Technology (Danvers, MA, USA).

Cell viability assay

LNCaP cells were cultured in 6-well plates at a density of 1 × 10⁵ cells/well. The cells were then treated with PROTAC at 37 °C for 72 h. Cell viability was determined using the EZ-Cytox assay kit (DoGenBio, Seoul, Korea). Absorbance was measured at 450 nm using a BioTeK microplate reader (Winooski, VT, USA). Images of live cells were captured using an IncuCyte (Panasonic, Tokyo, Japan).

Western blot analysis

LNCaP cells were lysed on ice in radioimmunoprecipitation assay (RIPA) buffer containing a protease inhibitor cocktail. Proteins were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS–PAGE) and transferred to Immun-Blot PVDF membranes (Bio-Rad, Hercules, CA, USA). After transfer, membranes were blocked with 5% skim milk in Tris-buffered saline with 0.1% Tween 20 (TBS-T) and probed with the specified primary antibodies overnight at 4 °C. The membranes were washed and incubated with secondary antibodies in TBS-T for 2 h, washed three times with TBS-T for 10 min, and developed using Immobilon Western Chemiluminescent HRP Substrate (Sigma-Aldrich, St. Louis, MO, USA). Protein expression levels were measured using a ChemiDoc XRS+ imaging system (Bio-Rad). The respective protein levels were quantified using Image J 1.8.0 software.

Quantitative real-time polymerase chain reaction (RT-qPCR)

Total RNA was isolated using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA). RNA was reverse transcribed using an iScript cDNA synthesis kit (Bio-Rad), and the resulting cDNA was amplified in real time using a LightCycler 480 II (Roche, Indianapolis, IN, USA). The primer used for the RT-qPCR reactions are as follows (forward and reverse respectively): PSA, 5′-TCACAGCTACCCACTGCATCA-3′ and 5′-AGGTCGTGGCTGGAGTCATC-3′; TMPRSS2, 5′-CCTGCAAGGACATGGGCTATA-3′ and 5′-CCGGCACTTGTGTTCAGTTTC-3′; KLK2, 5′-GCTGCCCATTGCCTAAAGAAG-3′ and 5′-TGGGAAGCTGTGGCTGACA-3′; MYC, 5′-CTCTCAACGACAGCAGCTCG-3′ and 5′-CAACATCGATTTCTTCCTCATCTTC-3′; 18S, 5′-GAGGATGAGGTGGAACGTGT-3′ and 5′-TCTTCAGTCGCTCCAGGTCT-3′. Data are presented as mean ± SD (n=3). The mRNA expression levels were normalized to those of 18S.

Statistics

Statistical analyses were performed using the GraphPad Prism 8 software (GraphPad Inc., San Diego, CA, USA). Significant differences between groups were analyzed using one-way analysis of variance (ANOVA), followed by Dunnett’s test. P < 0.05 was considered statistically significant.

Chemistry

Unless otherwise specified, all the reactions were performed in oven-dried glassware using a magnetic stirrer under a nitrogen balloon. Commercially available reagents were purchased from various companies (TCI, Sigma-Aldrich, Alfa Aesar, and Junsei) and were used throughout the synthesis. Anhydrous solvents [dichloromethane (DCM), dimethyl formamide (DMF), dimethyl sulfoxide (DMSO), and 1,4-dioxane] were purchased from commercial sources. Anhydrous tetrahydrofuran (THF) was prepared from Na metal, benzophenone, and anhydrous triethylamine (TEA) using KOH pellets. The progress of the reaction was checked using a thin layer chromatography (TLC) plate Silica Gel 60 F₂₅₄ by visual monitoring under UV light (254 nm and 365 nm). The products were concentrated using a Buchi rotary evaporator and purified by column chromatography using Zeochem silica gel or Biotage Medium Pressure Liquid Chromatography (MPLC). The NMR samples were prepared using CDCl₃ or (CD₃)₂SO purchased from Cambridge Isotope Laboratories. ¹H NMR spectra were obtained using a Bruker Avance III 600 MHz instrument. Chemical shifts (δ) are expressed as parts per million (ppm) versus tetramethyl silane (TMS). HRMS (ESI) data were obtained using a DART-MS machine.

6-(Benzyloxy)-4-bromo-N1-methylbenzene-1,2-diamine (3)

2-Amino-3-nitro phenol 2 (5.0 g, 32.4 mmol) was dissolved in 75 mL of EtOH. Potassium carbonate (6.7 g, 48.7 mmol) was then added to the flasks. Then, BnBr (4.80 mL, 40.6 mmol) was slowly added to the flask and the mixture was stirred heating at 70 °C for 4 h. Completion of the reaction was confirmed by TLC, and the solvent was removed using a rotary evaporator. The reaction mixture was diluted with 50 mL of water and extracted with 50 mL of EtOAc (×2). EtOAc layer was collected and dried over anhydrous MgSO₄ powder then removed under reduced pressure to obtain the crude product. Purification was performed by column chromatography EtOAc/n-Hex 1:5 to furnish benzylated yellow solid product 2a (6.40 g, 81%). ¹H NMR (600 MHz, Chloroform-d) δ 7.75 (dd, J = 8.9, 1.3 Hz, 1H), 7.43 – 7.41 (m, 4H), 7.40 – 7.36 (m, 1H), 6.96 (dd, J = 7.8, 1.3 Hz, 1H), 6.59 (dd, J = 8.9, 7.7 Hz, 1H), 6.45 (s, 2H), 5.12 (s, 2H). HRMS (ESI) calculated for C₁₃H₁₂N₂O₃ (M + Na)⁺ was 267.0745 and found to be 267.0757.

The obtained product 2a (2.50 g, 10.2 mmol) was dissolved in 40 mL acetic acid and sodium acetate (1.34 g, 16.4 mmol) was also added to the flask. Bromine (0.60 mL, 11.3 mmol) was added to the flask using a syringe with caution. The final reaction mixture was stirred at room temperature for 30 minutes, and the reaction was monitored by TLC. Acetic acid was removed under reduced pressure and the reaction mixture was diluted with 50 mL of water, followed by extraction with 50 mL of DCM (×2). DCM layer was collected and dried over anhydrous MgSO₄ powder then removed under reduced pressure to obtain the crude product. Purification was done using column chromatography EtOAc/n-Hex 1:10–1:5 to furnish brominated red oily product 2b (3.17 g, 96%). ¹H NMR (600 MHz, DMSO-d₆) δ 7.73 (d, J = 2.1 Hz, 1H), 7.53–7.51 (m, 2H), 7.42 (dd, J = 8.4, 6.8 Hz, 2H), 7.38–7.34 (m, 1H), 7.33 (d, J = 2.1 Hz, 1H), 7.18 (s, 2H), 5.28 (s, 2H). HRMS (ESI) calculated for C₁₄H₁₃BrN₂O₃ (M+H)⁺ was 323.0055 and found to be 323.0031.

The above desired brominated product 2b (3.16 g, 9.77 mmol) was dissolved in 30 mL DMF at 0 °C and sodium hydride (282 mg, 11.7 mmol) was also added to the solution. Then, the mixture was stirred for 30 minutes at 0 °C and then methyl iodide (0.67 mL, 10.7 mmol) was transferred to the flask containing starting material. The reaction was stirred at r.t for 1 h and the progress of the reaction was checked using TLC. DMF was removed under reduced pressure and the reaction mixture was diluted with 40 mL of water, followed by extraction with 40 mL of DCM (×2). DCM layer was collected and dried over anhydrous MgSO₄ powder then removed under reduced pressure to obtain the crude product. Purification was performed by column chromatography EtOAc/n-Hex 1:5 to furnish N-methylated orange solid product 2c (2.62 g, 79%). ¹H NMR (600 MHz, DMSO-d₆) δ 7.63 (d, J = 2.3 Hz, 1H), 7.50 (dd, J = 7.0, 2.0 Hz, 2H), 7.42 (tt, J = 8.1, 2.0 Hz, 3H), 7.32 (d, J = 2.3 Hz, 1H), 7.29 (q, J = 4.7 Hz, 1H), 5.21 (s, 2H), 2.89 (d, J = 5.3 Hz, 3H). HRMS (ESI) calculated for C₁₄H₁₃BrN₂O₃ (M + H)⁺ was 337.0213 and found to be 337.0187.

N-Methylated product 2c (2.61 g, 7.74 mmol) was dissolved in 40 mL of acetic acid. Zinc dust (3 g, 46.4 mmol) was cautiously added to the flask at 0 °C. The reaction mixture was then stirred at 45 °C for 2 h. TLC was performed to confirm the completion of the reaction. Acetic acid was removed under reduced pressure and the reaction mixture was diluted with 35 mL of water, followed by extraction with 35 mL of EtOAc (×2). EtOAc layer was collected and dried over anhydrous MgSO₄ powder then removed under reduced pressure to obtain the crude product. Purification was performed by column chromatography EtOAc/n-Hex 1:5 ~ 1:1 to furnish reduced brown oily product 3 (1.39 g, 59%). ¹H NMR (600 MHz, DMSO-d₆) δ 7.45 (d, J = 7.2 Hz, 2H), 7.39 (dd, J = 8.5, 6.8 Hz, 2H), 7.34 – 7.31 (m, 1H), 6.47 (d, J = 2.1 Hz, 1H), 6.45 (d, J = 2.2 Hz, 1H), 5.05 (s, 2H), 4.96 (s, 2H), 3.57 (s, 1H), 2.55 (s, 3H). HRMS (ESI) calculated for C₁₄H₁₅BrN₂O (M+H)⁺ was 307.0480 and found to be 307.0484.

7-(Benzyloxy)-5-bromo-1-methyl-1H-benzo[d][1,2,3] triazole (4)

3 (1.38 g, 4.49 mmol) was dispersed in 25 mL of 10% sulfuric acid solution (aq.) and NaNO₂ (436 mg, 6.28 mmol) was slowly added to the flask. After stirring for 30 minutes, water (50 mL) was added to the flask and stirred overnight. TLC confirmed the completion of the reaction. Extraction was performed using 50 mL of DCM (×2). DCM layer was collected and dried over anhydrous MgSO₄ powder. DCM was removed under reduced pressure to obtain the crude product. Purification was performed by column chromatography with EtOAc/n-Hex 1:5~1:1 to furnish yellow solid product 4 (1.25 g, 87%). ¹H NMR (600 MHz, DMSO-d₆) δ 7.84 (d, J = 1.4 Hz, 1H), 7.57 – 7.55 (m, 2H), 7.45 (t, J = 7.5 Hz, 2H), 7.40 – 7.37 (m, 1H), 7.24 (d, J = 1.4 Hz, 1H), 5.36 (s, 2H), 4.39 (s, 3H). HRMS (ESI) calculated for C₁₄H₁₂BrN₃O (M+H)⁺ was 318.0255 and found to be 318.0242.

tert-Butyl (E)-3-(7-(benzyloxy)-1-methyl-1H-benzo[d] [1,2,3] triazol-5-yl) acrylate (5)

4 (1.24 g, 3.89 mmol) was dissolved in 25 mL DMF at r.t. tert-Butyl acrylate (2.81 mL, 19.6 mmol), palladium tetrakis triphenyl phosphine (223 mg, 0.190 mmol), tri(o-tolyl) phosphine (237 mg, 0.77 mmol) and DIPEA (1.69 mL, 9.72 mmol) were also added to the flask. Solvent was degassed using nitrogen balloon for 10 minutes and was stirred at 95 °C for 5h. The progress of the reaction was confirmed by TLC, and DMF was removed under reduced pressure. The reaction mixture was diluted with 40 mL of water and extracted with 40 mL of EtOAc (×2). EtOAc layer was collected and dried over anhydrous MgSO₄ followed by concentration under reduced pressure to get crude product. Purification was performed by column chromatography with EtOAc/n-Hex 1:5 to furnish yellow solid product 5 (1.12 g, 79%). ¹H NMR (600 MHz, Chloroform-d) δ 7.72 (s, 1H), 7.67 (d, J = 15.8 Hz, 1H), 7.49 – 7.43 (m, 5H), 7.02 (d, J = 1.2 Hz, 1H), 6.36 (d, J = 15.9 Hz, 1H), 5.25 (s, 2H), 4.44 (s, 3H), 1.56 (s, 9H). HRMS (ESI) calculated for C₂₁H₂₃N₃O₃ (M + H)⁺ was 366.1818 and found to be 366.1834.

tert-Butyl-3-(7-(benzyloxy)-1-methyl-1H-benzo[d][1,2,3] triazol-5-yl)-3-(3(hydroxymethyl) -4-methylphenyl) propanoate (6)

5 (1.11 g, 3.29 mmol) was dissolved in a solvent mixture of 45 mL of dioxane/water (2:1). Rhodium catalyst (75 mg, 0.16 mmol) and Et₃N (0.64 mL, 4.93 mmol) were added to the flask and stirred for a few minutes. Then, 13 (1.14 g, 4.93 mmol) separately dissolved in 15 mL solvent mixture was transferred to the flask. The final reaction mixture was degassed using nitrogen balloon for 10 minutes and was stirred heating at 95 ^oC for 5h. The progress of the reaction was monitored by TLC. The solvent was removed using a rotary evaporator, and the reaction mixture was diluted with 35 mL of water, followed by extraction with 35 mL of EtOAc (×2). EtOAc layer was collected and dried over anhydrous MgSO₄ powder then was removed under reduced pressure to obtain the crude product. Purification was performed by column chromatography EtOAc/n-Hex 1:5 to furnish yellow oily racemate product 6 (788 mg, 53%). ¹H NMR (600 MHz, Chloroform-d) δ 7.50 (s, 1H), 7.40 (dt, J = 4.5, 0.7 Hz, 5H), 7.23 – 7.22 (m, 1H), 7.08 (s, 1H), 7.07 (d, J = 1.9 Hz, 1H), 6.69 (d, J = 1.1 Hz, 1H), 5.12 (s, 2H), 4.64 (s, 2H), 4.57 (t, J = 8.1 Hz, 1H), 4.39 (s, 3H), 3.01 (d, J = 8.5 Hz, 1H), 2.99 – 2.94 (m, 2H), 2.30 (s, 3H), 1.31 (s, 9H). HRMS (ESI) calculated for C₂₉H₃₃N₃O₄ (M+H)⁺ was 488.2577 and found to be 488.2549.

tert-Butyl-3-(7-(benzyloxy)-1-methyl-1H-benzo[d][1,2,3] triazol-5-yl)-3-(4-methyl-3-(((R)-4-methyl-1,1-dioxido-3,4-dihydro-2H-benzo[b] [1,4,5] oxathiazepin-2-yl) methyl) phenyl) propanoate (7)

6 (785 mg, 1.70 mmol) and 16 (525 mg, 2.55 mmol) were dissolved in 35 mL anhydrous THF. Triphenyl phosphine (847 mg, 3.40 mmol) was also dissolved in 20 mL anhydrous THF (separate flask), and DIAD (0.66 mL, 3.40 mmol) was added to the flask, which was then stirred at r. t. until the formation of a yellow-white precipitate. 10 mL of the precipitate was transferred to a flask containing the starting materials via a syringe and stirred for 3h. The progress of the reaction was confirmed by TLC, and solvent was removed under reduced pressure. The reaction mixture was diluted with 25 mL of water, followed by extraction with 25 mL of EtOAc (×2). EtOAc layer was collected and dried over anhydrous MgSO₄ powder then was removed under reduced pressure to obtain the crude product. Purification was performed by column chromatography EtOAc/n-Hex 1:5 ~ 1:1 to furnish yellow solid racemate product 7 (817 mg, 75%). ¹H NMR (600 MHz, DMSO-d₆) δ 7.78 (dt, J = 7.8, 1.8 Hz, 1H), 7.65 (ddt, J = 8.1, 7.4, 1.8 Hz, 1H), 7.51 (tt, J = 8.0, 1.4 Hz, 2H), 7.47 (t, J = 1.6 Hz, 1H), 7.41–7.32 (m, 5H), 7.31 (td, J = 2.9, 1.1 Hz, 1H), 7.28–7.25 (m, 1H), 7.12 (dd, J = 7.9, 2.3 Hz, 1H), 7.05 (dd, J = 21.2, 1.2 Hz, 1H), 5.27 (s, 2H), 4.48 – 4.36 (m, 3H), 4.34 (d, J = 5.2 Hz, 3H), 3.79 (dd, J = 14.0, 4.1 Hz, 1H), 3.62–3.55 (m, 1H), 3.08 (d, J = 8.1 Hz, 1H), 3.04 (dd, J = 15.5, 8.6 Hz, 1H), 2.73 (ddd, J = 56.2, 15.3, 1.4 Hz, 1H), 2.24 (d, J = 3.7 Hz, 3H), 1.22 (d, J = 5.3 Hz, 9H), 1.15 (d, J = 6.4 Hz, 2H), 1.03 (d, J = 6.3 Hz, 1H). HRMS (ESI) calculated for C₃₈H₄₂N₄O₆S (M+H)⁺ was 683.2880 and found to be 683.2870.

tert-Butyl-3-(7-hydroxy-1-methyl-1H-benzo[d][1,2,3] triazol-5-yl)-3-(4-methyl-3-(((R)-4-methyl-1,1-dioxido-3,4-dihydro-2H-benzo[b][1,4,5]oxathiazepin-2-yl) methyl) phenyl) propanoate (8)

7 (810 mg, 1.23 mmol) was dissolved in 30 mL solvent mixture of MeOH/THF (1:5) and catalytic amount of palladium (10% on Carbon) was also added to the flask. Vacuum was then applied to the flask, which was sealed with a septum, and hydrogen balloon was inserted via the septum. The reaction mixture was stirred for a few hours and the progress of the reaction was monitored using TLC. Solvent was filtered through celite powder washing with 20 mL EtOAc (×2), and collected clear organic layer was removed under reduced pressure to furnish yellow solid racemate product 8 (685 mg, 98%). ¹H NMR (600 MHz, DMSO-d₆) δ 7.77 (ddd, J = 7.8, 4.8, 1.7 Hz, 1H), 7.66–7.63 (m, 1H), 7.35 (tdd, J = 7.6, 3.9, 1.2 Hz, 1H), 7.29 (dt, J = 8.1, 1.4 Hz, 1H), 7.27 (s, 1H), 7.23 (d, J = 1.9 Hz, 1H), 7.16 (ddd, J = 8.0, 4.0, 1.9 Hz, 1H), 7.11 (dd, J = 7.9, 2.5 Hz, 1H), 6.55 (d, J = 9.6 Hz, 1H), 4.42–4.35 (m, 2H), 4.33 (d, J = 3.7 Hz, 3H), 3.80 (dd, J = 14.0, 3.0 Hz, 1H), 3.63–3.55 (m, 1H), 3.02–2.87 (m, 3H), 2.81–2.71 (m, 1H), 2.24 (s, 3H), 1.22 (d, J = 1.8 Hz, 9H), 1.17 (d, J = 6.4 Hz, 2H), 1.07 (d, J = 6.4 Hz, 1H). HRMS (ESI) calculated for C₃₁H₃₆N₄O₆S (M+H)⁺ was 593.2435 and found to be 593.2422.

2-((5-3-(tert-Butoxy)-1-(4-methyl-3-(((R)-4-methyl-1,1-dioxido-3,4-dihydro-2H-benzo[b][1,4,5]oxathiazepin-2-yl) methyl)phenyl)-3-oxopropyl)-1-methyl-1H-benzo[d][1,2,3] triazol-7-yl) oxy) acetic acid (9)

8 (250 mg, 0.42 mmol) was dissolved in THF (25 mL) and NaH (25 mg, 0.63 mmol) was added at 0 °C. The resulting mixture was stirred at room temperature for 30 min and then bromoacetic acid (88 mg, 0.63 mmol) was added to the flask. The final reaction mixture was refluxed for 2h. TLC confirmed the completion of the reaction, and THF was removed using a rotary evaporator. The reaction mixture was diluted with 40 mL of water, followed by extraction with 35 mL of EtOAc (×2). Collected EtOAc layer was dried over anhydrous MgSO₄ powder and concentrated in rotary evaporator to get racemate product 9 (259 mg, 95%). ¹H NMR (600 MHz, DMSO-d₆) δ 8.88 (s, 1H), 7.78 (ddd, J = 7.7, 3.7, 1.7 Hz, 1H), 7.68 (dd, J = 7.8, 1.7 Hz, 1H), 7.67–7.64 (m, 1H), 7.60 (td, J = 7.7, 1.7 Hz, 1H), 7.48 (s, 1H), 7.38–7.34 (m, 1H), 7.33 (dd, J = 7.7, 2.0 Hz, 1H), 7.31–7.30 (m, 1H), 7.11 (dd, J = 8.0, 2.5 Hz, 1H), 6.97 (dd, J = 18.2, 1.1 Hz, 1H), 4.88 (d, J = 9.5 Hz, 2H), 4.77 (p, J = 6.3 Hz, 2H), 4.42 (dd, J = 11.7, 3.8 Hz, 2H), 4.39 (d, J = 4.1 Hz, 3H), 3.80–3.77 (m, 2H), 3.06 (d, J = 8.2 Hz, 1H), 2.24 (d, J = 3.6 Hz, 3H), 1.31 (d, J = 6.4 Hz, 3H), 1.21 (d, J = 5.7 Hz, 9H). HRMS (ESI) calculated for C₃₃H₃₈N₄O₈S (M + Na)⁺ was 673.24, and found to be 673.2426.

General procedure for the HATU mediated amide coupling and tert-butyl group deprotection

9 (50 mg, 0.076 mmol), HATU (35 mg, 0.091 mmol) and DIPEA (0.02 mL, 0.114 mmol) were dissolved in 15 mL DMF solution of four different RB flasks. After stirring for a few minutes, 19a-d (0.076 mmol) were added to the respective flask, and the reaction was allowed to proceed for overnight. DMF was removed under reduced pressure and the reaction mixture was diluted with 20 mL of water, followed by extraction with 20 mL of EtOAc (×2). EtOAc layer was collected, dried over anhydrous MgSO₄ powder and concentrated in rotary evaporator to get crude product, which on purification furnished amide coupled products. Amide products were subjected to tert-butyl group deprotection using 9 mL of 4N HCl solution in dioxane and upon completion of reaction, solvent was removed in rotary evaporator washing with 10 mL MeOH (×2) to give the final racemate products 10a-d.

3-(7-(2-((4-(2-((S)-4-(4-Chlorophenyl)-2,3,9-trimethyl-6H-thieno[3,2-f][1,2,4]triazolo[4,3-a][1,4]diazepin-6-yl) acetamido) butyl) amino)-2-oxoethoxy)-1-methyl-1H-benzo[d] [1,2,3]triazol-5-yl)-3-(4-methyl-3-(((R)-4-methyl-1,1-dioxido-3,4-dihydro-2H-benzo[b][1,4,5]oxathiazepin-2-yl)methyl) phenyl) propanoic acid (10a, SD-2163)

Yield = 38 mg (86%), ¹H NMR (600 MHz, DMSO-d₆) δ 12.11 (s, 1H), 8.21 (t, J = 5.7 Hz, 1H), 8.11 (q, J = 5.4 Hz, 1H), 7.79–7.76 (m, 1H), 7.65 (td, J = 7.7, 1.7 Hz, 1H), 7.49–7.47 (m, 3H), 7.41 (d, J = 8.3 Hz, 2H), 7.35 (td, J = 6.1, 3.3 Hz, 1H), 7.33–7.29 (m, 2H), 7.26–7.22 (m, 1H), 7.11 (d, J = 7.9 Hz, 1H), 6.86 (d, J = 13.6 Hz, 1H), 4.66 (d, J = 7.0 Hz, 2H), 4.50 (dd, J = 8.1, 6.2 Hz, 1H), 4.46 (dd, J = 10.7, 5.3 Hz, 2H), 4.42 (d, J = 4.3 Hz, 1H), 4.37 (d, J = 4.3 Hz, 3H), 3.78 (d, J = 13.8 Hz, 1H), 3.60 (dt, J = 15.5, 9.8 Hz, 2H), 3.22 (dd, J = 13.6, 7.1 Hz, 2H), 3.13–3.06 (m, 6H), 2.58 (s, 3H), 2.40 (s, 3H), 2.24 (d, J = 5.2 Hz, 3H), 1.61 (s, 3H), 1.43 (dt, J = 6.9, 3.4 Hz, 4H), 1.23 (s, 1H), 1.20 (s, 1H), 1.09 (d, J = 6.3 Hz, 1H). HRMS (ESI) calculated for C₅₂H₅₅ClN₁₀O₈S₂ (M + Na)⁺ was 1069.33, and found to be 1069.3358.

3-(7-(2-((6-(2-((S)-4-(4-Chlorophenyl)-2,3,9-trimethyl-6H-thieno[3,2-f][1,2,4]triazolo[4,3-a][1,4]diazepin-6-yl) acetamido) hexyl) amino)-2-oxoethoxy)-1-methyl-1H-benzo[d] [1,2,3]triazol-5-yl)-3-(4-methyl-3-(((R)-4-methyl-1,1-dioxido-3,4-dihydro-2H-benzo[b][1,4,5]oxathiazepin-2-yl)methyl) phenyl) propanoic acid (10b, SD-2164)

Yield = 43 mg (94%), ¹H NMR (500 MHz, DMSO-d6) δ 12.10 (s, 1H), 8.18 (t, J = 5.7 Hz, 1H), 8.08 (q, J = 5.6 Hz, 1H), 7.77 (dt, J = 7.7, 1.8 Hz, 1H), 7.67 – 7.63 (m, 1H), 7.49–7.46 (m, 3H), 7.41 (d, J = 8.5 Hz, 2H), 7.35 (ddd, J = 8.9, 4.9, 2.3 Hz, 1H), 7.32–7.29 (m, 2H), 7.23 (ddd, J = 10.0, 7.7, 1.9 Hz, 1H), 7.10 (d, J = 7.9 Hz, 1H), 6.84 (d, J = 13.8 Hz, 1H), 4.66 (d, J = 7.2 Hz, 2H), 4.50 (dd, J = 8.0, 6.2 Hz, 1H), 4.45 (td, J = 7.2, 2.5 Hz, 2H), 4.41 (d, J = 3.6 Hz, 1H), 4.38 (d, J = 4.4 Hz, 3H), 3.77 (d, J = 13.8 Hz, 1H), 3.60 (dt, J = 15.3, 10.2 Hz, 2H), 3.23 (dd, J = 15.1, 7.0 Hz, 2H), 3.07 (qd, J = 7.6, 2.1 Hz, 6H), 2.60 (s, 1H), 2.58 (s, 3H), 2.41 (s, 1H), 2.40 (s, 3H), 2.24 (d, J = 5.1 Hz, 3H), 1.61 (s, 3H), 1.43–1.35 (m, 4H), 1.28–1.23 (m, 4H), 1.19 (d, J = 2.6 Hz, 1H), 1.08 (d, J = 6.4 Hz, 1H). HRMS (ESI) calculated for C₅₄H₅₉ClN₁₀O₈S₂ (M + Na)⁺ was 1097.36 and found to be 1097.3631.

3-(7-(2-((8-(2-((S)-4-(4-Chlorophenyl)-2,3,9-trimethyl-6H-thieno[3,2-f][1,2,4]triazolo[4,3-a][1,4]diazepin-6-yl) acetamido)octyl)amino)-2-oxoethoxy)-1-methyl-1H-benzo[d] [1,2,3]triazol-5-yl)-3-(4-methyl-3-(((R)-4-methyl-1,1-dioxido-3,4-dihydro-2H-benzo[b][1,4,5]oxathiazepin-2-yl)methyl) phenyl) propanoic acid (10c, SD-2406)

Yield = 41 mg (95%), ¹H NMR (600 MHz, DMSO-d₆) δ 12.04 (s, 1H), 8.18 (t, J = 5.7 Hz, 1H), 8.08 (q, J = 5.8 Hz, 1H), 7.75 (dt, J = 7.8, 1.7 Hz, 1H), 7.64–7.62 (m, 1H), 7.47 (d, J = 8.7 Hz, 2H), 7.44 (d, J = 8.5 Hz, 1H), 7.41 (d, J = 8.5 Hz, 2H), 7.33 (tdd, J = 7.5, 3.4, 1.2 Hz, 1H), 7.27 (ddd, J = 6.4, 4.9, 2.8 Hz, 1H), 7.23 (s, 1H), 7.17 (d, J = 7.7 Hz, 1H), 7.05 (dd, J = 8.0, 3.7 Hz, 1H), 6.75 (d, J = 12.9 Hz, 1H), 4.62 (d, J = 4.1 Hz, 1H), 4.61 (s, 1H), 4.49 (dd, J = 8.3, 5.9 Hz, 2H), 4.39 (d, J = 5.0 Hz, 1H), 4.37 (d, J = 4.7 Hz, 3H), 3.78 (dd, J = 13.8, 2.7 Hz, 1H), 3.60–3.56 (m, 1H), 3.22 (d, J = 8.2 Hz, 1H), 3.19 (d, J = 6.0 Hz, 1H), 3.12–3.09 (m, 1H), 3.07 (d, J = 6.5 Hz, 1H), 3.04–3.00 (m, 2H), 2.82 (d, J = 15.3 Hz, 2H), 2.73 (d, J = 15.0 Hz, 1H), 2.58 (s, 3H), 2.39 (s, 3H), 2.21 (d, J = 7.2 Hz, 3H), 1.61 (s, 3H), 1.49 (p, J = 5.5 Hz, 1H), 1.41 (t, J = 7.1 Hz, 2H), 1.34 (dq, J = 13.8, 6.5 Hz, 3H), 1.28–1.18 (m, 8H), 1.15 (d, J = 6.1 Hz, 1H), 1.04 (d, J = 6.3 Hz, 1H). HRMS (ESI) calculated for C₅₆H₆₃ClN₁₀O₈S₂ (M+H)⁺ was 1103.40, and found to be 1103.4020.

3-(7-(2-((10-(2-((S)-4-(4-Chlorophenyl)-2,3,9-trimethyl-6H-thieno[3,2-f][1,2,4]triazolo[4,3-a][1,4]diazepin-6-yl) acetamido)decyl)amino)-2-oxoethoxy)-1-methyl-1H-benzo[d] [1,2,3]triazol-5-yl)-3-(4-methyl-3-(((R)-4-methyl-1,1-dioxido-3,4-dihydro-2H-benzo[b][1,4,5]oxathiazepin-2-yl)methyl) phenyl) propanoic acid (10d, SD-2407)

Yield = 43 mg (90%), ¹H NMR (600 MHz, DMSO-d₆) δ 8.17 (t, J = 5.6 Hz, 1H), 8.08 (d, J = 5.7 Hz, 1H), 7.75 (dt, J = 7.8, 1.7 Hz, 1H), 7.65–7.62 (m, 1H), 7.48–7.45 (m, 2H), 7.44 (d, J = 8.8 Hz, 1H), 7.42 (d, J = 8.5 Hz, 2H), 7.35–7.32 (m, 1H), 7.28 (ddd, J = 8.1, 2.6, 1.2 Hz, 1H), 7.23 (s, 1H), 7.17 (d, J = 7.8 Hz, 1H), 7.05 (dd, J = 8.0, 3.7 Hz, 1H), 6.75 (d, J = 13.0 Hz, 1H), 4.62 (d, J = 5.1 Hz, 1H), 4.61 (s, 1H), 4.49 (dd, J = 8.4, 5.7 Hz, 2H), 4.40 (s, 1H), 4.37 (d, J = 4.6 Hz, 3H), 3.79 (dd, J = 14.0, 3.0 Hz, 1H), 3.59 (d, J = 5.2 Hz, 1H), 3.23 (d, J = 8.4 Hz, 1H), 3.18–3.09 (m, 3H), 3.04 (dt, J = 21.6, 6.6 Hz, 3H), 2.83 (d, J = 15.1 Hz, 2H), 2.74 (d, J = 14.8 Hz, 1H), 2.58 (s, 3H), 2.43 (t, J = 6.8 Hz, 1H), 2.40 (s, 3H), 2.21 (d, J = 7.2 Hz, 3H), 1.61 (s, 3H), 1.51–1.48 (m, 1H), 1.42 (t, J = 7.1 Hz, 2H), 1.38–1.31 (m, 3H), 1.24 (d, J = 12.0 Hz, 4H), 1.19 (d, J = 4.2 Hz, 6H), 1.16 (d, J = 6.2 Hz, 1H), 1.04 (d, J = 6.3 Hz, 1H). HRMS (ESI) calculated for C₅₈H₆₇ClN₁₀O₈S₂ (M+Na)⁺ was 1153.4300 and found to be 1153.4309.

Design and synthesis of KEAP1 BRD4 PROTACs

In order to design our KEAP1 BRD4 PROTAC (Fig. 2), we first selected an E3 ligase binder i.e. compound 1 on the basis of its’ binding affinity IC₅₀ = 12.50 nM by fluorescence polarization (FP) and K_d = 1.30 nM by isothermal titration calorimetry (ITC). Furthermore, the crystal structures of compound 1 and the KEAP1 protein (PDB: 5FNU) revealed the solvent-exposed region of compound 1 which could provide crucial details about the position for attaching the linker.²⁶⁾

Fig. 2. Design of KEAP1 BRD4 PROTAC.
Compound 1 is reported as an inhibitor of KEAP1 (IC₅₀ < 15 nM, Kd = 1.3 nM), and was used as an E3 ligase binder connected to a flexible linking vector augmented with JQ1 (POI binder).

Scheme 1 depicts the synthesis of KEAP1 BRD4 PROTACs. Starting with 2-amino-3-nitrophenol 2, the sequential O-benzylation of the phenolic hydroxyl groups with benzyl bromide, para-bromination of aniline with bromine, N-methylation of aniline with methyl iodide, and reduction of the nitro group to amine using zinc dust gave aniline 3. N-methyl-1,2-aminobenzene 3 was cyclized to benzotriazole 4 by treating it with NaNO₂ and 10% H₂SO₄, and the Heck reaction with tert-butyl acrylate using Pd(PPh₃)₄ and P(O-Tol)₃ ligand produced ester 5. The olefin was coupled with the phenylboronic acid pinacol ester 13 using a rhodium catalyst [RhCl(Cod)]₂] to afford alcohol 6, which was transformed into sulfonamide 7 via the Mitsunobu reaction (DIAD, PPh₃). Debenzylation of 7 using H₂ and 10% Pd/C provided the key intermediate, phenol compound 8 that was subjected to O-alkylation reaction using bromoacetic acid to yield compound 9. The final compounds SD-2163, SD-2164, SD-2406 and SD-2407 were obtained by the HATU-mediated amide coupling reaction of compound 9 and linker-augmented JQ1 compounds 19a-d followed by tert-butyl group deprotection in 4N HCl/Dioxane condition.

Scheme 1. Synthesis of KEAP1 BRD4 PROTAC.
Reagent and conditions: (a) BnBr, K₂CO₃, EtOH, 70 ^oC, 81%; (b) Br₂, AcONa, AcOH, r.t, 96%; (c) MeI, NaH, DMF, 0 ^oC to r.t, 79%; (d) Zn dust, AcOH, 0 ^oC to 45 ^oC, 59%; (e) NaNO₂, 10% H₂SO₄ (aq.), 87%; (f) tert-Butyl acrylate, Pd(PPh₃)₄, P(O-Tol)₃, DMF, 95 ^oC, 79%; (g) 13, [RhCl(Cod)]₂, Et₃N, Dioxane/H₂O (2:1), 95 ^oC, 53%; (h) 16, DIAD, PPh₃, anhydrous THF, r.t, 75%; (i) H₂, Pd/C, MeOH, r.t, 98%; (j) NaH, Bromo acetic acid, THF, reflux, 95%; (k) (i) 19a-d, HATU, DIPEA, DMF, r.t, 45-60%; (ii) 4N HCl/Dioxane, r.t, 86-95%

Degradation of BRD4 by PROTAC compounds in LNCaP cells

BRD4 protein levels were assessed by western blot analysis in LNCaP cells treated with SD-2163, SD-2164, SD-2406, or SD-2407 for 24 h (Fig. 3). AT-1 was used as the positive control. All four compounds degraded BRD4 protein levels in a concentration-dependent manner, with SD-2406 exhibiting the lowest DC₅₀ of 0.045 μM. AT-1 showed DC₅₀ of 0.371 μM. These results suggested that the PROTAC compounds SD-2163, SD-2164, SD-2406, and SD-2407 downregulated BRD4 protein levels in LNCaP cells and that SD-2406 was the most potent BRD4 degrader.

Fig. 3. Degradation of BRD4 by PROTAC compounds in LNCaP cells.
(A) LNCaP cells were treated with 0.01–10 μM of BRD4 PROTAC compounds for 24 h. Western blotting was used to measure BRD4 and β-actin protein levels. AT-1 was used as a positive control compound. (B) Changes in relative BRD4 protein levels by PROTAC compound treatment in LNCaP cells. The respective protein levels were quantified using Image J software. (C) BRD4 proteolytic activity of AT-1, SD-2163, SD-2164, SD-2406, and SD-2407 was presented as DC₅₀. (D) BRD4 Degrader AT-1

Inhibition of LNCaP cell proliferation

Inhibition of BRD4 by small molecules or PROTAC compounds has been reported to inhibit prostate cancer cell proliferation.^27,28) Therefore, the proliferation of LNCaP cells treated with PROTAC was observed using a live cell imaging system (Fig. 4). All the tested compounds inhibited LNCaP proliferation in a concentration-dependent manner. The most significant results were observed with SD-2406 treatment, which inhibited cell proliferation more strongly than positive control AT-1. SD-2163, SD-2164, and SD-2407 also inhibited cell proliferation in a concentration-dependent manner, but had weaker inhibitory effects than AT-1.

Fig. 4. Inhibition of LNCaP cell proliferation.
(A) LNCaP cells were treated with 1 and 10 μM of AT-1, SD-2163, SD-2164, SD-2406, and SD-2407 for 72 h. Cell images were captured using a live cell imaging system. (B) Cell viability of BRD4 PROTAC-treated LNCaP cells. Results are presented as the mean ± S.D. *P<0.05, **P<0.01 vs CTR.

Down-regulation of BRD4 target gene expression

BRD4 acts as a co-activator of the androgen receptor (AR) and is attributed to the upregulation of AR target genes such as MYC, an important gene responsible for cell cycle regulation and anti-apoptotic function in prostate cancer cells.²⁹⁾ Therefore, we determined their mRNA and protein expression levels to evaluate the effects of KEAP1 BRD4 PROTAC compounds on BRD4 target gene expression. SD-2164, SD-2406, and SD-2407 significantly down-regulated the mRNA expression of PSA, MYC, TMPRSS2, and KLK2 in LNCaP cells in a concentration-dependent manner (Fig. 5A). SD-2163 significantly downregulated the mRNA expression of PSA and KLK2 but had no effect on the mRNA expression of MYC and TMPRSS2. Furthermore, MYC protein levels in LNCaP cells treated with the PROTAC compounds were examined by western blot analysis. SD-2406 and SD-2407 significantly attenuated MYC protein level of MYC (Fig. 5B). These results suggested that the expression of BRD4-regulated genes was effectively downregulated by the degradation of BRD4 by PROTAC compounds.

Fig. 5. Down-regulation of BRD4 target gene expression by BRD4 PROTAC.
(A) RT-qPCR analysis of relative PSA, MYC, TMPRSS2, and KLK2 mRNA levels in LNCaP cells treated with 1 and 10 μM of BRD PROTAC compounds for 24 h. Data are presented as mean ± SD (n=3). (B) LNCaP cells were treated with SD-2406 and SD-2407 (0.01–10 μM) for 48 h. MYC and β-actin protein levels were detected using western blotting. Results are presented as the mean ± S.D. *P<0.05, **P<0.01, ***P<0.001 vs CTR.

Novel E3 ligase that can be applied to PROTAC technology may offer new possibilities in terms of cleavage activity of the POI, selectivity, and toxicity in PROTAC technology. A crucial element in formulating PROTACs compounds is the incorporation of an appropriate E3 ligase binder with a linker-bound POI ligand. The challenge lies in the limitations of the available E3 ligase binders, making it difficult to select a suitable binder, although several E3 ligase binders have been successfully applied in this technology. In this context, we chose KEAP1 E3 ligase as a component of PROTAC. Several known compounds inhibit KEAP1-Nrf2 interactions. The first step in designing PROTAC with a KEAP1 E3 ligase inhibitor is to determine where to introduce the linker from the selected KEAP1 E3 ligase inhibitor. Fortunately, the co-crystal structures of KEAP1 E3 ligase and its inhibitor are already known. By analyzing the binding structures, we determined the solvent-facing part of the binding structure and devised a new synthetic scheme based on an inhibitor to introduce the linker outward. We selected BRD4 as the target protein to be degraded by PROTAC and used JQ1 compound, which is a well-established BRD4 binder that has been applied to other PROTACs. The type and nature of the linker were found to significantly affect the degradation activity. The length of the linker between the POI binder and E3 ligase binder is usually 6–12 atoms to evaluate the activity of PROTAC. Therefore, in this study we designed four different linker lengths (5, 7, 9, and 11 from C=O of the KEAP1 binder to NH of JQ1) and introduced a flexible alkyl chain to prove this possibility. The degradation activity of the synthesized PROTACs were evaluated to reveal that the length of the linker was crucial for degrading the BRD4 target protein. Our compound, SD-2406, with a DC₅₀ value of 0.043 µM, showed the highest degradation activity and outperformed the positive control AT-1, which has a DC₅₀ of 0.371 µM.

In summary, our results suggest that a small-molecule inhibitor targeting the KEAP1 E3 ligase-Nrf2 interaction holds promise as an E3 ligase-recruiting ligand for designing PROTAC compounds using the TPD approach. Furthermore, this study provides insights into the repertoire of novel E3 ligases on the PROTAC platform.

This research was funded by grants from the National Research Foundation (NRF) of Korea (grant numbers NRF- 2020R1A6A1A03043708 and NRF-2021R1A2C1012280).

All authors declare that they have no conflict of interest.

Raju Gurung : Graduate student

Jae Rim Lee : Graduate student

Min Ju Cho : Graduate student

Jin Ah Jeong : Graduate student

Sung Jean Park : Professor

Kwang Won Jeong : Professor

Dongyun Shin : Professor