Development of a potent small-molecule degrader against oncogenic BRAF^V600E protein that evades paradoxical MAPK activation

2.1 Design and synthesis of PROTAC compounds

We designed PROTAC compounds in which a BRAF inhibitor is connected to a ligand for an E3 ligase via a linker. The chemical synthesis, schemes, and physicochemical data for compounds are provided in the Doc. S1 and Scheme S1-S23.

2.2 Reagents

Tissue culture plastics were purchased from Greiner Bio-One. RPMI 1640 medium, DMEM, Eagle's minimum essential medium, and kanamycin were from Merck. McCoy's 5a medium and Leibovitz's L-15 medium were from Thermo Fisher Scientific. Benzyloxycarbonyl-L-leucyl-L-leucyl-L-leucinal (MG132) was from Peptide Institute. MLN7243 and MLN4924 were from Active Biochem.

2.3 Cell culture

Human melanoma A375 cells, N-RAS mutant-A375 isogenic cells, and K-RAS mutant-A375 isogenic cells were maintained in DMEM containing 10% FBS and 100 μg ml⁻¹ kanamycin. Human colon carcinoma COLO205, RKO, and NCI-H508 cells; human lung carcinoma NCI-H23, NCI-H1395, NCI-H2087, NCI-H1755, and NCI-H1666 cells; human melanoma SK-MEL-30 cells; human ovarian carcinoma OVCAR8 cells; and human B cell leukemia JVM-3 cells were maintained in RPMI 1640 medium containing 10% FBS (5% FBS for NCI-H1666) and 100 μg ml⁻¹ kanamycin. Human melanoma SK-MEL-28 cells were maintained in Eagle's minimum essential medium containing 10% FBS and 100 μg ml⁻¹ kanamycin. Human colon carcinoma HCT116 cells were maintained in McCoy's 5a medium containing 10% FBS and 100 μg ml⁻¹ kanamycin. Human breast carcinoma MDA-MB-231 cells were maintained in Leibovitz's L-15 medium containing 10% FBS and 100 μg ml⁻¹ kanamycin. SK-MEL-30 and JVM-3 cell lines were purchased from DSMZ. OVCAR8 cell line was described elsewhere.³⁴ The other cell lines were purchased from ATCC. These cells were treated with various concentrations of compounds for the indicated periods of time.

2.4 Western blotting

Cells were lysed with SDS lysis buffer (0.1 M Tris-HCl at pH 8.0, 10% glycerol, 1% SDS) and immediately boiled for 10 minutes to obtain clear lysates. Protein concentrations were measured using the BCA method (Pierce). Lysates containing equal amounts of proteins were separated by SDS-PAGE and transferred to PVDF membranes (Merck) for Western blot analysis using the appropriate antibodies. Immunoreactive proteins were visualized using the Immobilon Western chemiluminescent HRP substrate (Merck) or Clarity Western ECL substrate (Bio-Rad); light emission intensity was quantified using an LAS-3000 lumino-image analyzer equipped with Image Gauge v2.3 software. The antibodies used in this study were: anti-BRAF antibody (14814; Cell Signaling Technology), anti-CRAF antibody (53745; Cell Signaling Technology), anti-MEK1/2 antibody (8727; Cell Signaling Technology), anti-phospho-MEK1/2 (Ser217/221) antibody (9121; Cell Signaling Technology), anti-p44/42 MAPK (ERK1/2) antibody (4695; Cell Signaling Technology), anti-phospho-p44/42 MAPK (ERK1/2) (Thr202/Tyr204) antibody (4377; Cell Signaling Technology), anti-RAS antibody (8955; Cell Signaling Technology), anti-CRBN antibody (71810; Cell Signaling Technology), and anti-β-actin antibody (A5316; Merck).

2.5 siRNA transfection

A375 cells were transiently transfected with an equal mixture of three CRBN-specific stealth short interfering RNAs (stealth siRNAs) (HSS121807, HSS121808, HSS121809; Life Technologies) or a negative control stealth siRNA (12935300; Life Technologies) using Lipofectamine RNAi MAX reagent (Life Technologies) according to the protocols provided by the manufacturer.

2.6 Cell viability assay

Cell viability was determined using water-soluble tetrazolium WST-8 (4-[3-(2-methoxy-4-nitrophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3-benzenedisulfonate) reagent in a spectrophotometric assay according to the manufacturer's instructions (Dojindo). Cells treated with compounds were incubated with WST-8 reagent for 0.5 hour at 37°C in a humidified atmosphere of 5% CO₂. The absorbance of the medium at 450 nm or 490 nm was measured using an EnVision Multilabel Plate Reader (PerkinElmer).

3.1 CRBR(BRAF)-24 induces efficient and potent degradation of BRAF^V600E

To target BRAF mutant proteins for degradation, we designed and synthesized several chimeric molecules that combine various BRAF inhibitors with four E3 ligase ligands (LCL161 for IAP,²⁸ VH032 for VHL,²⁷ pomalidomide for CRBN,³⁵ and RG7388 for MDM2³⁰). After evaluating the ability of the chimeric compounds to degrade BRAF^V600E mutant proteins in A375 cells (a cell line with homozygous BRAF^V600E), we identified an active compound CRBN(BRAF)-1 composed of a clinical BRAF inhibitor PLX8394^{23, 36, 37} and pomalidomide (Figure 1). As shown in Figure 2A, CRBN(BRAF)-1 effectively reduced BRAF protein levels after 24 hours of treatment. CRBN(BRAF)-2, in which the same ligands are conjugated with a different linker, showed attenuated activity to reduce BRAF^V600E (Figure 1 and Figure 2A). Chimeric compounds recruiting other E3 ligases (IAP[BRAF]-7, VHL[BRAF]-2, and MDM2[BRAF]-2) that replaced the pomalidomide of CRBN(BRAF)-1 with LCL161, VH032, and RG7388, respectively, did not reduce BRAF levels (Figure S1).

We further developed a series of chimeric compounds (CRBN[BRAF]-23, CRBN[BRAF]-24, CRBN[BRAF]-12, and CRBN[BRAF]-8) with improved degradation activity by modifying CRBN(BRAF)-1 with linkers of different lengths and structures (Figure 1). CRBN(BRAF)-24, one of the most potent compounds, was selected as a representative on the basis of its ability to reduce BRAF protein levels in A375 cells (Figure 2B,C). CRBN(BRAF)-24 showed a DC₅₀ (the concentration at which 50% of maximum degradation was observed) value of 6.8 nM and D_MAX (the percentage of maximum degradation observed) of approximately 80% in A375 cells at 24 hours after treatment (Figure 2B,C). The reduction in BRAF protein was maintained for at least 72 hours (Figure 2D). In SK-MEL-28 cells (another cell line with homozygous BRAF^V600E), CRBN(BRAF)-24 also showed effective BRAF-reducing activity and inhibited the downstream pathway, MEK, as shown by suppressed ERK phosphorylation (Figure 2E). We further synthesized an inactive degrader, CRBN(BRAF)-29, by methylating the nitrogen of the glutarimide ring in the pomalidomide moiety of CRBN(BRAF)-24, which abolished the recruitment of CRBN³⁵ (Figure 1). CRBN(BRAF)-29 did not decrease BRAF protein levels, suggesting that the binding to CRBN with a pomalidomide moiety is necessary for the BRAF^V600E reduction by CRBN(BRAF)-24 (Figure 2E). In addition, the activity of CRBN(BRAF)-29 to inhibit the downstream kinase signaling was approximately 10-fold weaker than that of CRBN(BRAF)-24, suggesting that the BRAF degradation activity of CRBN(BRAF)-24 contributes to better inhibition of downstream kinase signaling. A moderate hook effect was occasionally observed in the CRBN(BRAF)-24–induced BRAF reduction at higher concentrations (Figure 2B and D), probably because excess amounts of bivalent compounds inhibited the formation of the ternary complex composed of target (BRAF), degrader (CRBN[BRAF]-24), and E3 ligase (CRBN).^{24, 27}

3.2 CRBN(BRAF)-24 degrades BRAF^V600E via UPS

To determine whether CRBN(BRAF)-24 reduces the protein level of BRAF^V600E by the expected molecular mechanism, we first investigated the effect of UPS inhibitors. When A375 cells were treated with a proteasome inhibitor, MG132, or a ubiquitin-activating enzyme inhibitor, MLN7243, together with CRBN(BRAF)-24, the CRBN(BRAF)-24–induced reduction in BRAF levels was completely suppressed (Figure 3A). Similarly, the effect of CRBN(BRAF)-24 was also inhibited by MLN4924, a NEDD8-activating enzyme inhibitor that inhibits the activity of the Cullin-RING E3 ligases (CRLs), including CRL4^CRBN. These results suggest that CRBN(BRAF)-24 degrades BRAF^V600E via UPS using the CRL-type E3 ligases. It was also confirmed that treatment with a mixture of the BRAF ligand (PLX8394-L) and the CRBN ligand (pomalidomide) did not induce BRAF degradation (Figure 1; Figure 3B). In addition, cotreatment with an excess amount of PLX8394-L (target ligand competition) or pomalidomide (E3 ligand competition) inhibited the BRAF degradation by CRBN(BRAF)-24 (Figure 3B). Silencing the expression of CRBN by RNA interference (RNAi) inhibited the degradation of BRAF by CRBN(BRAF)-24 (Figure 3C). These results collectively indicate that CRL4^CRBN is required for BRAF^V600E degradation by CRBN(BRAF)-24.

3.3 Neither BRAF degradation nor paradoxical activation of MAPK signaling is induced by CRBR(BRAF)-24 in BRAF^WT cells

PLX8394, the BRAF ligand used for CRBN(BRAF)-24, has been shown to bind to BRAF^V600E, BRAF^WT, and CRAF, although the binding is approximately three- and sixfold more selective for BRAF^V600E than for BRAF^WT and CRAF, respectively.²³ Therefore, we next examined if CRBN(BRAF)-24 can degrade BRAF^WT and CRAF proteins. In cell lines that do not bear the BRAF mutant alleles (HCT116, NCI-H23, SK-MEL-30, and OVCAR8), CRBN(BRAF)-24 had little effect on BRAF^WT and CRAF protein levels (Figure 4B), strongly suggesting that CRBN(BRAF)-24 did not induce the degradation of BRAF^WT and CRAF proteins. In cell lines bearing BRAF^V600E alleles, CRBN(BRAF)-24 showed marginal (A375, COLO-205, and RKO) or moderate (SK-MEL-28) activity to reduce CRAF protein (Figure 4A). However, the CRAF reduction mechanism in these cells seems to be different from the BRAF^V600E degradation because depletion of CRBN did not affect the CRAF reduction (Figure S2A), and CRAF protein levels were also reduced by PLX8394 (Figure S2B). Importantly, CRBN(BRAF)-24 did not activate MAPK signaling in any BRAF^WT cells regardless of the presence of oncogenic RAS (Figure 4B), indicating that CRBN(BRAF)-24 does not induce paradoxical activation even in cells in which BRAF is constantly activated by the oncogenic RAS mutants.

3.4 CRBN(BRAF)-24 selectively induces growth inhibition of cell lines expressing BRAF^V600E

Next, we investigated the effect of CRBN(BRAF)-24 on cancer cell proliferation and compared its efficacy with that of CRBN(BRAF)-29 and PLX8394. Consistent with the effective inhibition of the downstream MAPK pathway (Figure 4A), CRBN(BRAF)-24 strongly inhibited proliferation in three of the four BRAF^V600E cell lines (A375, SK-MEL-28, and COLO205) at 10 nM and higher (Figure 5A). In these cell lines, CRBN(BRAF)-24 demonstrated better antiproliferative activity than CRBN(BRAF)-29 and PLX8394. In RKO cells, CRBN(BRAF)-24 showed a weak activity to inhibit proliferation. In contrast, none of the drugs showed antiproliferative activity against BRAF^WT cell lines (Figure 5B). These results indicate that CRBN(BRAF)-24 selectively inhibits proliferation of cancer cells harboring BRAF^V600E mutation, as does PLX8394.

3.5 CRBN(BRAF)-24 exerts a potent and sustained inhibitory effect by degrading BRAF^V600E

Because CRBN(BRAF)-24 uses a BRAF inhibitor as a warhead, it inhibits and degrades BRAF^V600E protein. Hence, we investigated the advantage of CRBN(BRAF)-24 as a degrader/inhibitor over a mere inhibitor. First, to determine whether CRBN(BRAF)-24 potently inhibits downstream signaling by degrading BRAF^V600E, we compared ERK phosphorylation in BRAF^V600E-positive cells treated with CRBN(BRAF)-24 and CRBN(BRAF)-29 in the presence or absence of a proteasome inhibitor, MG132. The active degrader CRBN(BRAF)-24 inhibited ERK phosphorylation more potently than the inactive degrader CRBN(BRAF)-29, but the better inhibitory effect of CRBN(BRAF)-24 was diminished by MG132 (Figure 6A). These results strongly suggest that CRBN(BRAF)-24 potently suppresses downstream ERK signaling by both inhibiting and degrading BRAF^V600E.

As for the advantages of PROTACs, catalytic degradation of target proteins has been reported to result in sustained drug action.^{27, 38, 39} To examine the sustained suppression of ERK phosphorylation, cells were treated with CRBN(BRAF)-24, PLX8394, and CRBN(BRAF)-29 for 24 hours, and then further incubated in drug-free medium (Figure 6B,C). After 24 hours of drug treatment (washout 0 hour), BRAF protein was reduced only in the cells treated with CRBN(BRAF)-24. At this timepoint, CRBN(BRAF)-24 and PLX8394 at 100 nM almost completely inhibited the level of phosphorylated ERK in SK-MEL-28 cells, while CRBN(BRAF)-29 required a concentration that was 10 times higher to achieve equivalent inhibition (Figure 6B). The suppressed ERK phosphorylation was maintained for 24 hours in CRBN(BRAF)-24–treated cells even after drug removal but was gradually restored in cells treated with PLX8394 and CRBN(BRAF)-29, especially at low concentrations. Surprisingly, the level of BRAF^V600E protein was further reduced in the CRBN(BRAF)-24–treated cells during 24 hours incubation in drug-free medium. In line with this, treatment of SK-MEL-28 cells with CRBN(BRAF)-24 for 24 hours effectively inhibited cell proliferation in drug-free medium for the subsequent 72 hours (Figure 6C). In a parallel experiment, PLX8394 and CRBN(BRAF)-29 had little effect. These results suggest that CRBN(BRAF)-24 shows potent and sustained suppression of downstream ERK signaling by degrading BRAF^V600E, thereby inhibiting the proliferation of BRAF^V600E-driven cancer cells more potently than kinase inhibitor.