CRAF mutations in lung cancer can be oncogenic and predict sensitivity to combined type II RAF and MEK inhibition

Abstract

Two out of 41 non-small cell lung cancer patients enrolled in a clinical study were found with a somatic CRAF mutation in their tumor, namely CRAF^P261A and CRAF^P207S. To our knowledge, both mutations are novel in lung cancer and CRAF^P261A has not been previously reported in cancer. Expression of CRAF^P261A in HEK293T cells and BEAS-2B lung epithelial cells led to increased ERK pathway activation in a dimer-dependent manner, accompanied with loss of CRAF phosphorylation at the negative regulatory S259 residue. Moreover, stable expression of CRAF^P261A in mouse embryonic fibroblasts and BEAS-2B cells led to anchorage-independent growth. Consistent with a previous report, we could not observe a gain-of-function with CRAF^P207S. Type II but not type I RAF inhibitors suppressed the CRAF^P261A-induced ERK pathway activity in BEAS-2B cells, and combinatorial treatment with type II RAF inhibitors and a MEK inhibitor led to a stronger ERK pathway inhibition and growth arrest. Our findings suggest that the acquisition of a CRAF^P261A mutation can provide oncogenic properties to cells, and that such cells are sensitive to combined MEK and type II RAF inhibitors. CRAF mutations should be diagnostically and therapeutically explored in lung and perhaps other cancers.

Fig. 1

CRAF^P261A activates the ERK pathway in HEK293T and BEAS-2B cells, and signals as a dimer. HEK293T (a) or BEAS-2B cells (b) cells were transiently transfected with a CRAF^P261A, CRAF^P207S, or CRAF^WT expression vectors alone. The CRAF^P261A mutation hyperactivates the ERK pathway. The prevention of CRAF dimerization (by CRAF^R401H) abolishes CRAF^P261A hyperactivation of the ERK pathway in HEK293T cells (c). Co-expression of CRAF^WT, CRAF^P261A, or CRAF^P207S with BRAF^WT (as previously described [22]) in HEK293T cells do not result in higher (or lower) ERK pathway activation compared with single transfections (d). Forty-eight hours post-transfection cells were lysed and subjected to western blotting analysis to detect the indicated proteins (a, b, c, d). EV stands for empty vector. Lipofectamine was added in the mock condition

Fig. 2

CRAF^P261A induces anchorage-independent growth. BEAS-2B (a), NIH3T3 (b), and MEF (c) cells were stably transduced with empty vector, CRAF^WT, CRAF^P207S (only in BEAS-2B and NIH3T3), or CRAF^P261A as indicated (a–c). Total number of colonies per well were counted (representative experiment) after 16 (c) or 18 (a, b) days in culture. The experiments were performed with three biological repeats and were represented as a bar chart (means ± SEM). d Representative whole well image of the nitro blue tetrazolium chloride stained MEF colonies after 16 days in culture. e Representative high-magnification images of the MEF colonies prior to staining. f Graphical representation of the relative size of the MEF colonies as determined with OpenCFU software [55]. Dots represent individual colonies (the three biological repeats are pooled), lines represent the means

Fig. 3

CRAF^P261A-induced ERK pathway activation is suppressed by type II RAF inhibitors. a BEAS-2B cells were transiently transfected with different CRAF expression vectors (wild type or mutant). Forty-eight hours post transfection, cells were treated for 2 h with DMSO, Dabrafenib (1 µM), or AZ628 (1 µM), or LY3009120 (1 µM), then lysed and subjected to western blotting analysis to detect the indicated proteins. b, c BEAS-2B cells were transiently transfected with CRAF^P261A. Forty-eight hours post transfection, cells were treated for 2 h with DMSO, Vemurafenib, Dabrafenib, AZ628, LY3009120, or Sorafenib (drug concentrations are indicated). b Cells were lysed and subjected to western blotting analysis to detect the indicated proteins. c Graphical representation of the relative p-ERK signals (normalized to actin and DMSO group) based on at least two independent experiments as shown in b. Dots represent individual data points, lines represent the mean value ± SEM. d MEF cells stably expressing CRAF^P261A were treated with DMSO, Vemurafenib, Dabrafenib, AZ628, LY3009120, or Sorafenib (all at 1 µM but Sorafenib at 5 µM) for a duration of 72 h. Cell viability was determined using CellTiter-Glo. The dots represent the means of the independent experiments, the horizontal lines with error bars represent the mean ± SEM of three independent experiments each performed at least in triplicate. Statistical significance was indicated by *** and represents a p-value < 0.001

Fig. 4

CRAF^P261A predicts sensitivity to combination of LY3009120 and Trametinib. a BEAS-2B cells were transiently transfected with different CRAF expression vectors (wild type or mutant). Forty-eight hours post transfection, cells were treated for 2 h with DMSO, LY3009120 (1 µM), and/or Trametinib (25 nM), then lysed and subjected to western blotting analysis for the indicated proteins. b MEF cells stably expressing CRAF^P261A were treated with DMSO, Sorafenib, LY3009120, Trametinib, and the combinations of Sorafenib (5 µM) or LY3009120 (1 µM) with Trametinib (10 nM), for a duration of 72 h. Cell viability was determined using CellTiter-Glo. The dots represent the means of the independent experiments, the horizontal lines with error bars represent the mean ± SEM of three independent experiments each performed at least in triplicate. Statistical significance was indicated by ** and ***, which represent p-values < 0.01 and 0.001, respectively