Specific purified N-terminal domains are involved in HRAS interactions. (A) Diagram of BRAF-NT constructs. Top panel is full-length BRAF, followed by proteins NT1-4 expressed in E. coli and purified. Not shown: 6xHis/MBP tag on the N-terminal of NT proteins. BSR: BRAF specific region; RBD: RAS binding domain; CRD: cysteine rich domain. (B) Coomassie stained gels of purified NT1-4 and GST-HRAS. (C) Western blot of HRAS-GMPPNP or HRAS-GDP pulled down on glutathione resin to probe for NT1 binding. (D) Western blot of BRAF NT1-4 pulled down on amylose resin to probe for HRAS-GMPPNP binding.

HDX-MS revealed conformational changes of BRAF N-terminal domains in response to active HRAS binding. (A, C) Representative NT2 (A) and NT3 (C) peptides identified from HDX-MS in the absence (blue) and presence (pink) of HRAS. Peptides with no difference in H-D exchange rate are consistent in both NT2 and NT3. Presented peptide plots displaying differences in H-D exchange rate are representative of a trend of multiple overlapping peptides in the BSR and/or RBD. Gray dotted lines represent the theoretical exchange behavior for specified peptide that is fully unstructured (top) or for specified peptide with a uniform protection factor (fraction of time the residue is involved in protecting the H-bond) of 100 (lower). (B, D) Deuteron uptake differences of NT2 (B) and NT3 (D) mapped on the predicted BRAF AlphaFold structure, where deuterium exchange is decreased (red) or increased (cyan).

BRAF specific region (BSR) in conjunction with the cysteine rich domain (CRD) reduces binding affinity for HRAS. (A-D) OpenSPR binding curves of HRAS flowed over NT1-4 immobilized on NTA sensors and the best fit curves (black) produced from a 1:1 fitting model kinetic evaluation. Representative of independent experiments with similar results each (NT1: n=7, NT2: n=2, NT3: n=2, NT4: n=3). (E) Diagram of the average dissociation constant (KD) ± standard deviation from independent OpenSPR experiments of HRAS flowed over immobilized NT1-4. (F-G) Western blot of purified His/MBP-NT1 (F) or -NT3 (G) binding to active GST-HRAS on glutathione resin in a pulldown assay and subsequent competition with NT3 (F) or NT1 (G). Representative of 2 independent experiments each with similar results.

BSR differentiates the BRAF-KRAS interaction from the BRAF-HRAS interaction. (A) Western blot of purified His/MBP-NT1-4 binding to GST-KRAS on glutathione resin in a pulldown assay. Representative of 2 independent experiments with similar results. (B) Diagram of the average dissociation constant (KD) ± standard deviation from independent OpenSPR experiments of KRAS flowed over immobilized NT1-4. (C-F) OpenSPR binding curves of KRAS flowed over NT1-4 immobilized on NTA sensors and the best fit curves (black) produced from a 1:1 fitting model kinetic evaluation. Representative of at least 2 independent experiments with similar results each. (G) OpenSPR binding curves of KRAS and HRAS flowed over FL-BRAF immobilized on NTA sensors and the best fit curves (black) produced from a 1:1 fitting model kinetic evaluation. K- and HRAS flowed over at increasing concentrations of 12.3, 37, 111, 333, 1000, and 3000 nM at 30 µL/min. Representative of 2 independent experiments with similar results each. (H) Diagram of the average dissociation constant (KD) ± standard deviation from independent OpenSPR experiments of H/KRAS flowed over immobilized FL-BRAF. (I) Representative NT2 peptides identified from HDX-MS in the absence (blue) and presence (pink) of KRAS. Peptides with no difference in H-D exchange rate are exemplified under “no difference”. Presented peptide plots displaying differences in H-D exchange rate are representative of a trend of 4+ overlapping peptides in the RBD. Gray dotted lines represent the theoretical exchange behavior for specified peptide that is fully unstructured (top) or for specified peptide with a uniform protection factor (fraction of time the residue is involved in protecting the H-bond) of 100 (lower).

BSR and CRD promote BRAF autoinhibitory interactions. (A) Western blot of purified His/MBP-NT1-4 binding to biotinylated His-KD on streptavidin beads in a pulldown assay. Representative of 3 independent experiments with similar results. (B) Diagram of the average dissociation constant (KD) ± standard deviation from independent OpenSPR experiments of NTs flowed over immobilized KD. (C-D) OpenSPR binding curves of NT1 and NT3 at 5, 15, 44, 133, 400, and 1200 nM (NT3 only) flowed over KD and the best fit curves (black) produced from a 1:1 fitting model kinetic evaluation. Representative of independent experiments with similar results (NT3: n=2, NT1: n=3). (E-F) No binding of NT2 or NT4 to immobilized KD was observed by OpenSPR even at high concentrations (NT2: 1.125, 2.25, 4.5 µM; NT4: 1.5, 3, 6 µM). Representative of 2 independent experiments each with similar results. (G) Diagram of BRAF-NT5 (top) and Coomassie stained gel of recombinant NT5 (lower). (H) NT5 at 1.5, 3, and 6 µM flowed over immobilized KD on a carboxyl sensor. Representative of 3 independent experiments.

HRAS and KDD594G disrupt BRAF autoinhibition. (A) Western blot of pulldown assay of pre-incubated His/MBP-NT1 and His-KD added to purified active GST-HRAS on glutathione resin. Representative data of 2 independent replicates with similar results. (B) OpenSPR experiments in which NT1 at 5, 15, 44, 133, 400 nM (black) and NT1 + HRAS-GMPPNP (1:1) at 5, 15, 44, 133, and 400 nM (red) flowed over KD immobilized on carboxyl sensors. Representative data of 3 independent replicates with similar results. (C) OpenSPR experiments of NT1 at 5, 15, 44, 133, and 400 nM flowed over immobilized KDWT (black) or KDMUT (D594G; red) on carboxyl sensors. Representative data of 3 independent replicates with similar results. (D) Western blot of purified His/MBP-NT1 binding to either biotinylated His-KDWT or His-KDMUT (D594G) in pulldown assays on streptavidin beads. Representative data of 3 independent replicates with similar results.

Model of BRAF activation. (A) BRAF is initially an autoinhibited monomer in the cytosol, in which signaling through the RAS-RAF-MEK-ERK cascade is not promoted. The BRAF N-terminal region (NT1; aa 1-288), including the BRAF Specific Region (BSR), Cysteine Rich Domain (CRD), and RAS Binding Domain (RBD), interacts with active GTP-bound RAS at the membrane in an isoform specific manner. BRAF has higher affinity for KRAS due to isoform differences, as shown through the different dissociation constants (KD) determined through OpenSPR. (B) Once bound to active H- or K-RAS, BRAF is unable to remain in the autoinhibited conformation and is subsequently activated upon dimerization, which stimulates signaling for events such as cell growth, proliferation, and differentiation. (C) Tight binding is observed with the BRAF Kinase Domain (KD; aa 442-723) and BRAF NT1, revealing the concerted action of the BSR, RBD, and CRD domains to reinforce the autoinhibited conformation and restrict signaling without upstream activation. (D) Oncogenic BRAFD594G stimulates activation of MAPK pathway through a decreased ability to remain in the autoinhibited conformation and an increased potential to dimerize with CRAF. Evading autoinhibitory regulation leads to overactivation of the signaling cascade and tumorigenesis.

SEC of active GST-HRAS and GST-KRAS.

(A) GST-HRAS monomer is ∼45 kDa and elutes as a dimer at ∼90 kDa. HRAS elution profile (green) on a Superdex 200 (Cytiva) overlayed with protein size standard elution profile (gray). (B) GST-KRAS elution profile details the same as GST-HRAS (A). Fractions from ∼12-14 mL were collected and concentrated. (C) Coomassie stained gel of GST-KRAS final purification product.

Stripe Plots for peptides identified in BRAF-NT2 (a) and NT3 (b). Peptide coverage shown begins at the start of BRAF-NT2 or NT3 (not shown-coverage for MBP-tag N-terminal to the BRAF sequence). (C) Protein sequences used in MS analysis.

Peptides from NT3 in the CRD region. Plots on left represent peptides that could have slowed deuterium uptake. Plots on right represent peptides in the same region that show essentially no change. Blue= NT3-apo; Magenta= NT3+HRAS. BRAF residues 232-284 (CRD)= peptide residues 491-543. Gray dotted lines represent the theoretical exchange behavior for specified peptide that is fully unstructured (top) or for specified peptide with a uniform protection factor (fraction of time the residue is involved in protecting the H-bond) of 100 (lower).

Peptides from NT2 in the BSR region (aa. 82-99). Three representative peptide plots that have increased deuterium uptake. Blue= NT2-apo; Magenta= NT2+HRAS. BRAF residues 82-99 correspond with peptide residues 490-507. Gray dotted lines represent the theoretical exchange behavior for specified peptide that is fully unstructured (top) or for specified peptide with a uniform protection factor (fraction of time the residue is involved in protecting the H-bond) of 100 (lower).

HRAS-NT1 OpenSPR shows slow association.

(A) OpenSPR injections of HRAS (111, 333, 660, and 999 nM) at 5 µl/min over His/MBP-NT1 immobilized on an NTA sensor. (B) Western blot of GST-HRAS on glutathione resin and NT1 binding through pulldown assay. HRAS was first added to resin for 1 hour. After washing to remove unbound HRAS, NT1 was added in a 1:1 molar ratio and incubated at 4 °C for 5, 30, and 60 minutes. HRAS was probed with GST antibody and NT1 with His antibody.

(A) Binding curves of 1 μM GST-KRAS (active/GMPPNP= blue; inactive/GDP= orange) flowed over NTA-immobilized BRAF-NT2 at 30 μl/min. (B) Binding curves of 1 μM GST-KRAS (active/GMPPNP-loaded= red) and purified GST (green) flowed over NTA-immobilized BRAF-NT2 at 30 μl/min. (C) Coomassie stained gel of GST protein, purified following the same protocol as GST-HRAS.

Characterizing untagged KRAS

(A) Untagged KRAS is displayed as a monomer with a molecular weight of ∼22 kDa including N-terminal linker residues. KRAS-GMPPNP (purple) and KRAS-GDP (green) elution profile on a Superdex 75 (Cytiva) overlayed with protein size standard elution profile (gray). (B) Coomassie stained gel of untagged KRAS final purification product. (C) Western blot of purified KRAS binding to His/MBP-NT1-4 on amylose resin in a pulldown assay. Representative of 3 independent experiments with similar results. (D) OpenSPR binding curves of KRAS flowed over NT2 and the best fit curves produced from a 1:1 fitting model kinetic evaluation. Representative of 2 independent experiments with similar results each. KRAS was flowed over at 30 μL/min (20 mM HEPES pH 7.4, 150 mM NaCl, 0.05% Tween-20, 1% w:v BSA) for 10 mins at increasing concentrations (62.5, 125, 250, 500, 1000, 2000 nM). (E) Diagram of the average binding constant (KD) ± standard deviation from independent OpenSPR experiments of KRAS flowed over immobilized NT2. (F) Binding curves of 2 μM KRAS (active/GMPPNP= blue; inactive/GDP= orange) flowed over NTA-immobilized BRAF-NT2 at 30 μl/min.

(A) Coomassie stained gel of purified recombinant full-length (FL) BRAF with copurified chaperone proteins. (B) Binding curves of 1 μM GST-KRAS (active/GMPPNP = blue; inactive/GDP= orange) flowed over NTA-immobilized FL-BRAF at 30 μl/min.

(A) OpenSPR binding curves of BRAF-NT1 without MBP flowed over immobilized BRAF-KD at 30 μl/min in increasing concentrations. Best fit curves (black) are produced from a 1:1 fitting model kinetic evaluation.