Mechanism of BRAF Activation through Biochemical Characterization of the Recombinant Full-Length Protein

Abstract

BRAF kinase plays an important role in mitogen-activated protein kinase (MAPK) signaling and harbors activating mutations in about half of melanomas and in a smaller percentage in many other cancers. Despite its importance, few in vitro studies have been performed to characterize the biochemical properties of full-length BRAF. Herein, a strategy to generate an active, intact form of BRAF protein suitable for in vitro enzyme kinetics is described. It is shown that purified, intact BRAF protein autophosphorylates the kinase activation loop and this can be enhanced by binding the MEK protein substrate through an allosteric mechanism. These studies provide in vitro evidence that BRAF selectively binds to active RAS and that the BRAF/CRAF heterodimer is the most active form, relative to their respective homodimers. Full-length BRAF analysis with small-molecule BRAF inhibitors shows that two drugs, dabrafenib and vemurafenib, can modestly enhance kinase activity of BRAF at low concentration. Taken together, this characterization of intact BRAF contributes to a framework for understanding its role in cell signaling.

Keywords: BRAF; cancer; cell signaling; enzymes; proteins; reaction mechanisms.

Figure 1.

(A) Size-exclusion chromatography of purified full-length BRAF. The elution volumes of protein standards (690, 158, 44, and 17 kDa) are indicated by arrows. (B) Coomassie blue–stained SDS–PAGE analysis of the purified full-length BRAF. FL-BRAF and HSP70 bands are indicated by arrows. Molecular weight markers (M) are indicated. (C) One microgram of purified full-length BRAF was incubated with various concentrations of BS³ (0, 7.8, 15.6, 31.25, 62.5, 125, and 250 μM) for 30 min at room temperature. The samples were separated with SDS-PAGE and stained with coomassie blue staining. BRAF dimers and BRAF monomers are indicated by arrows. (D) Quantification of the activity of 5 nM BRAF with or without the indicated amount (1, 5, and 25 nM) of HSP70 protein. Reaction mixtures were subjected to western blot analysis with anti-pMEK (top) and anti-HSP70 (bottom) antibodies. Representative images were shown from one of three independent experiments. (E) Specific enzyme activity of full-length BRAF (FL-BRAF) and kinase domain of BRAF (CD-BRAF). The kinase activity of BRAF was quantified via western blotting for MEK phosphorylation. The levels of phospho-MEK were analyzed using ImageJ software and three independent replicates were included.

Figure 2.

(A) Comparison of the enzymatic activity of dephosphorylated versus native BRAF. The dephosphorylated BRAF was obtained after treatment of CIP protein phosphatase. BRAF protein purified from HEK293F cells is defined as native protein. The kinase activity of BRAF was quantified via western blotting for MEK phosphorylation. (B) Time-course of auto-phosphorylation of the activation loop. Purified BRAF was incubated with ATP for the indicated time 2.5, 5.0, 7.5, 10, 15, 20, and 30 min. (C) Auto-phosphorylation of the BRAF activation loop was increased by the presence of kinase-dead MEK substrate. BRAF was incubated with or without ATP and kinase-dead MEK for 5 min at 30 °C and probed with phospho-Thr599 BRAF antibody. (D) Autoradiograph showing incorporation of radioactive phosphate (³²P) into BRAF and MEK simultaneously. BRAF and ³²P-labeled ATP were incubated with or without kinase-dead MEK for 30 min at 30 °C. The top band represents auto-phosphorylated BRAF (p-BRAF). The bottom band represents phosphorylated MEK (p-MEK).

Figure 3:

Enzymatic characterization of BRAF. (A&B) The kinase activity of BRAF is linear versus enzyme concentration (25, 50, 100, and 200 nM) (A) as well as reaction time (5.0, 7.5, 10, and 12.5 min) (B). (C) Steady-state kinetic analysis of BRAF. The values of K_m for ATP and k_cat were obtained with varying concentrations of ATP (3.13, 6.25, 12.5, 25.0, 50.0 μM) and fixed kinase-dead MEK concentration (200 nM) and BRAF enzyme concentration (50 nM).

Figure 4.

Dose-responsive inhibition curves of ATP-competitive inhibitors. (A&B) Effects of different concentrations of dabrafenib and vemurafenib on the activity of full-length BRAF. (C) Dose-response curves of various ATP-competitive inhibitors against full-length BRAF. Summary of percent of activity (± s.d.) calculated from three independent experiments is shown.

Figure 5:

(A) Coomassie blue–stained SDS-PAGE analysis of the purified HRAS, as indicated by the arrow. (B) Top: Size-exclusion chromatography of the purified HRAS. The elution volumes of protein standards (690, 158, 44, and 17 kDa) are indicated by arrows; Bottom: Individual fractions were analyzed by SDS-PAGE and stained with coomassie blue. (C) In vitro affinity pull-down experiments. After Flag-tagged BRAF proteins were bound to Flag-M2 antibody conjugated agarose beads, the samples were incubated with either GDP-HRAS or GppNHp-HRAS. After washing, resin-bound proteins were probed with anti-GST antibody (upper panel). The corresponding two lower panels demonstrate comparable protein loading of the lanes. (D) Comparison of the activity of 5 nM BRAF with or without the indicated amounts (0, 2.5, 5.0, 10, 20, 50, 75, 100, and 200 nM) of GppNHp-HRAS.

Figure 6:

(A) Comparison of the specific activities of 20 nM BRAF with and without 100 nM CRAF. The activity was measured by radiometric kinase assay and normalized to the activity of 20 nM BRAF. The activity of 100 nM CRAF by itself is below the detection limit. (B) A pull-down assay shows the formation of a BRAF/CRAF complex. (C) The kinase activity of the BRAF/CRAF mixture is non-linear versus reaction time. (D) Autoradiograph showing incorporation of radioactive phosphate (³²P) into BRAF and MEK, simultaneously. The top band represents phosphorylated BRAF (p-BRAF). The bottom band represents phosphorylated MEK (p-MEK). No phosphorylated CRAF was detected. (E) The mixture of BRAF/CRAF is more resistant to dabrafenib. Autoradiograph showing incorporation of radioactive phosphate (³²P) into BRAF and MEK simultaneously. The results are representative of three independent tests. Error bars indicate standard deviation.