Selective CRAF Inhibition Elicits Transactivation

Abstract

Discovering molecules that regulate closely related protein isoforms is challenging, and in many cases the consequences of isoform-specific pharmacological regulation remains unknown. RAF isoforms are commonly mutated oncogenes that serve as effector kinases in MAP kinase signaling. BRAF/CRAF heterodimers are believed to be the primary RAF signaling species, and many RAF inhibitors lead to a "paradoxical activation" of RAF kinase activity through transactivation of the CRAF protomer; this leads to resistance mechanisms and secondary tumors. It has been hypothesized that CRAF-selective inhibition might bypass paradoxical activation, but no CRAF-selective inhibitor has been reported and the consequences of pharmacologically inhibiting CRAF have remained unknown. Here, we use bio-orthogonal ligand tethering (BOLT) to selectively target inhibitors to CRAF. Our results suggest that selective CRAF inhibition promotes paradoxical activation and exemplify how BOLT may be used to triage potential targets for drug discovery before any target-selective small molecules are known.

Figure 1

Designing BOLT ligands and sites of CRAF tethering. (A) Structures of noncanonical amino acids (ncAA) BocK and BCNK. ncAA are site-specifically incorporated into amber (UAG) variants of the CRAF gene. (B) Schematic of tethering between BOLT ligand, containing tetrazine (blue), linker (green), and pharmacophore (red), and BCNK containing proteins following an inverse electron demand Diels–Alder reaction between the BCNK and tetrazine. (C) Chemical structures of parent pharmacophores and the corresponding BOLT ligands. RAF pharmacophores, AZ628 (type II), and PLX4720 (type I) are shown in red. BOLT ligands AZ13-tet and AZ181-tet are shown; the tetrazine moiety is in blue, the linker in green, and the pharmacophore in red. (D) Structural superposition of MEK (gray) and CRAF (blue) kinase structures. A small-molecule inhibitor (yellow) occupies the ATP binding pocket. Spheres highlight positions tested for amber suppression expression and tethering, MEK (red) and CRAF (orange). Figure created using Pymol. PDB: 3ZLS MEK; 3OMV CRAF. (E) Immunoblot of the indicated CRAF variants showing full length and ncAA-dependent expression. Variants (small arrow) include C terminal epitope tags, FLAG, and HA (3xFLAG-HA) to ensure immunoprecipitation and detection of full length CRAF. Plasmids containing 4x[tRNAPyl] and CRAF(S357TAG) or CRAF(Q436TAG) variants were transfected into HCT116* cells. Cells were grown with indicated ncAA (2 mM BocK, 200 μM BCNK). Lysates were collected after 48 h of expression. Extended screening and testing of CRAF(YXXXTAG) alleles are available in Figure S3; XXX indicates the position at which the codon for a canonical amino acid (Y) is replaced with the amber codon (TAG).

Figure 2

BOLT of CRAF variants elicits kinase activation. (A) Immunoblot of kinase assays using CRAF (S357TAG) expressed with BCNK or BocK in HCT116*. Cells were washed to remove excess ncAA and then treated with indicated BOLT ligand (AZ13-tet or AZ181-tet, 2 μM). CRAF(S357BCNK) or CRAF(S357BocK) were immunoprecipitated via their C-terminal FLAG tag. The immunoprecipitate was assayed for RAF kinase activity in vitro, using catalytically dead MEK1 as a substrate. All CRAF amber alleles are HA tagged. (B) Immunoblot of kinase assays using CRAF (Q436TAG) expressed with BCNK or BocK in HCT116*. Cells were washed to remove excess ncAA and then treated with indicated BOLT ligand (AZ13-tet or AZ181-tet, 2 μM). CRAF(Q436BCNK) or CRAF(Q436BocK) were immunoprecipitated via their C-terminal FLAG tag. The immunoprecipitate was assayed for RAF kinase activity in vitro, using catalytically dead MEK1 as a substrate. All CRAF amber alleles are HA tagged. (C) Immunoblot of kinase assays using CRAF (S357TAG) expressed with BCNK or BocK in HCT116* treated with varying concentrations of BOLT ligands. The experiment was performed as described in panel A. Ligand concentrations were 2000, 200, and 20 nM. (D) Immunoblot of kinase assays using CRAF (Q436TAG) expressed with BCNK or BocK in HCT116* treated with varying concentrations of BOLT ligands. The experiment was performed as described in panel B. Ligand concentrations were 2000, 200, and 20 nM.

Figure 3

Dimerization of cellular RAF using BOLT. (A) Cellular RAF dimerization assay based on the BRET donor nanoluciferase(nLuc) and BRET acceptor chloroalkane conjugate and Halo-tagged (HT) species of BRAF kinases. Addition of nLuc substrate, furimazine, results in a dimerization-dependent energy transfer, as detected by emission of the fluorescent-dye chloroalkane conjugate. (B) Heterodimizeration of CRAFnLuc variants (wild-type and CRAF(S357TAG) incorporating BCNK or BocK) in response to an increasing concentration of AZ628 and AZ13-tet. Error bars correspond to standard deviation across four biological replicates. The continuous line corresponds to nonlinear regression (four variable) completed using Prism software. Data are color-coded as shown in the associated table. (C) Heterodimizeration of CRAFnLuc variants (wild-type and CRAF(Q436TAG) incorporating BCNK or BocK) in response to increasing concentrations of AZ628 and AZ13-tet. Error bars correspond to standard deviation across four biological replicates. Continuous line corresponds to nonlinear regression (four variable) completed using Prism software. Data are color-coded as in panel B. (D) Summary of calculated apparent EC₅₀ values across the different CRAF variants and ligands, mean values shown with error bars representing standard deviation between at least two independent dose response experiments. Error bars correspond to standard deviation between calculated EC₅₀ values based on nonlinear regression (four variable) modeling. Calculated EC₅₀ values with confidence intervals shown in Figure S11.