Disordered region of cereblon is required for efficient degradation by proteolysis-targeting chimera

Proteolysis targeting chimeras (PROTACs) are an emerging strategy for promoting targeted protein degradation by inducing the proximity between targeted proteins and E3 ubiquitin ligases. Although successful degradation of numerous proteins by PROTACs has been demonstrated, the elements that determine the degradability of PROTAC-targeted proteins have not yet been explored. In this study, we developed von Hippel-Lindau-Cereblon (VHL-CRBN) heterodimerizing PROTACs that induce the degradation of CRBN, but not VHL. A quantitative proteomic analysis further revealed that VHL-CRBN heterodimerizing PROTACs induced the degradation of CRBN, but not the well-known immunomodulatory drug (IMiD) neo-substrates, IKAROS family zinc finger 1 (IKZF1) and −3 (IZKF3). Moreover, truncation of disordered regions of CRBN and the androgen receptor (AR) attenuated their PROTAC-induced degradation, and attachment of the disordered region to stable CRBN or AR facilitated PROTAC-induced degradation. Thus, these results suggest that the intrinsically disordered region of targeted proteins is essential for efficient proteolysis, providing a novel criterion for choosing degradable protein targets.


General tert-butylester deprotection for 4
To a solution of 3 (100 mg) in DCM (1 mL) was added TFA (1 mL) and stirred at rt for 1 h. The reaction mixture was concentrated in vacuum to afford 4 as a crude product. The crude was used in the next step without purification after LC/MS analysis.

General synthetic method for 6
To a solution of 4 (1 eq.), VHL ligand 5 (1 eq.), HATU (1.5 eq.) in DMF was DIPEA (2 eq.) and stirred at rt overnight. The reaction mixture was diluted with water and extracted with EtOAc. The organic layer was washed with brine, dried over MgSO4, and concentrated. The crude residue was purified on PREP-TLC using MeOH/DCM (5% as an eluent) to afford the desired product 6.

TMT Labeling and Peptide Fractionation
The proteins from three replicates were labeled with Amine-Reactive Tandem Mass Tag Reagents (TMT6 Label Reagents, #90064; Thermo Scientific) according to the manufacturer's protocol. The peptide was solubilized in 100 mM TEAB. The TMT labels were reconstituted in 41 μL of acetonitrile prior to labeling, and 51 μL was added to the sample for labeling over 1 h at 25 °C.
Eight microliters of 5% hydroxylamine were added to quench each reaction. After 15 min, the samples were combined. Finally, the labeled peptides were dried using vacuum centrifugation. To

Identification of Differentially Expressed Proteins
The reporter ion intensities of the identified peptides were normalized using quantilenormalization methods 3 . We then selected the sets of proteins that have more than three nonredundant unique peptides. For the selected proteins, the relative protein abundances were computed from unique peptide intensities as previously described 4

Pharmacokinetic studies in mice
In vivo PK of compounds was examined in male ICR mice (6-7 weeks, OrientBio, Kyunggido, South Korea). For TD-165, three doses (2, 10, and 50 mg/kg) were administered via intraperitoneal route. Dosing vehicles were composed of 47% PEG300, 3% Cremophor EL and 50% Saline. After the retro-orbital blood sample collection at each time point, protein precipitation was conducted for the plasma sample preparation by adding acetonitrile at a 9:1 ratio to the plasma.

Equilibrium solubility
Excess amount of solid compound were added to the deionized water and samples were incubated at 25'C for 24 h on an orbital shaker. After filtration using the syringe filter (0.2 µm pore size, Whatman plc, UK), the compound concentration in the filtrate was determined using the LC-MSMS system.

Parallel artificial membrane permeability assay
Permeability of TD-165 was evaluated using the PAMPA Explorer Test SystemTM (Pion Inc.).
PAMPA sandwich was formed from a 96-well filter plates coated with GIT-0 lipid solution (Pion Inc.).
The initial donor sample concentration was 10 uM and the acceptor compartment was filled with acceptor sink buffer (Pion Inc.). Then the PAMPA sandwich was incubated for 4 h in the Gut-BoxTM.
After the incubation, the compound concentration in acceptor and donor was determined by UV spectrophotometry with an Epoch plate reader instrument (Biotek Inc., Winnoski, VTS) and permeability coefficient was determined using the PAMPA Explorer software.

Plasma protein binding
The unbound fractions to the human and mouse plasma protein were estimated using a rapid equilibrium dialysis (RED) system following the manufacturer's protocol (Thermo Scientific, Waltham, MA). Briefly, plasma sample with TD-165 (100 µM) and PBS were placed into the assembled RED system and then, incubated at 37'C on an orbital shaker for 4 h. After the incubation, samples were obtained from the device, equal volume of blank plasma or PBS was added to avoid matrix effect. Protein precipitation was conducted for the ample preparation by adding acetonitrile and after centrifugation the drug concentration of supernatant were analyzed using the LC-MSMS system. The unbound fraction of the drug was calculated.