Editors' ChoiceCancer

The Double Life of RAF

See allHide authors and affiliations

Science Translational Medicine  03 Mar 2010:
Vol. 2, Issue 21, pp. 21ec35
DOI: 10.1126/scitranslmed.3001000

Last week, BRAF was the darling of cyberspace. Among the New York Times most–e-mailed articles was Amy Harmon’s three-part series on the personalized treatment of melanoma with inhibitors of a mutated form of this protooncogene product. At the same time, the journal Nature posted an Advance Online Publication by Neal Rosen and colleagues that added yet another layer of complexity to the mechanism(s) of action of RAF inhibitors.

A serine-threonine protein kinase, BRAF is an important component of the RAS/RAF/MEK/ERK cell-signaling pathway, which mediates cell-growth signals and is aberrantly activated in a variety of solid tumors through mutations in either the RAS or RAF genes (but not both). The RAF family proteins are downstream of RAS in the signaling pathway, and RAS activates RAF by promoting its dimerization. Early RAF inhibitors such as sorafenib interfere in a competitive manner with wild-type RAF’s ability to bind adenosine triphosphate (ATP), which is required for RAF’s protein kinase activity. However, when a recurrent BRAF-activating mutation (V600E) was identified in melanoma and other cancers, scientists developed a new class of drugs that specifically inhibits the mutated version of BRAF. This attempt to minimize the effect of BRAF inhibitors on normal cells has yielded the promising yet complex results obtained with the drug PLX4032 and chronicled in Harmon’s articles.

In the new work, researchers show that all RAF inhibitors, whether mutant-selective or not, activate rather than inhibit the RAS/RAF/MEK/ERK pathway in BRAF wild-type cells at low doses. This pathway activation required the presence of CRAF, which normally forms homodimers or heterodimers with BRAF upon RAS activation. The inhibitor appears to bind directly to CRAF and yet stimulates its kinase activity. To explain this remarkable paradox, the investigators propose the following model: Within a CRAF/CRAF (or CRAF/BRAF) dimer, binding of the inhibitor to one partner leads to activation of the other. Of course, the active site of one partner in the dimer must be free of inhibitor, which explains why the activating effect is seen only at low drug concentrations. RAS activation is also required to promote dimerization of RAF.

To test their model, the researchers turned to another RAF inhibitor called JAB, which binds RAF only when a “designer mutation” is introduced that allows JAB access to the active site. Like the other RAF inhibitors, JAB activated the signaling pathway downstream of RAF. The authors then added another mutation that incapacitated the active site, and JAB had no effect. However, when wild-type CRAF (which does not bind JAB) was mixed in with the JAB-binding, kinase-dead CRAF, JAB again became a pathway activator. This observation provides elegant proof that one RAF partner binds the inhibitor, whereas the other activates the pathway. In tumors carrying the BRAF (V600E) mutations, RAS is not activated because RAS and RAF mutations tend to be mutually exclusive. Thus, in cells with RAF mutations the inhibitors do not cause the paradoxical activation because RAF does not undergo dimerization. This explains the remarkable clinical activity of PLX4032 in BRAF-mutant cancers. It remains to be seen whether circumstances exist in certain cancer patients that permit this paradoxical activation of the RAS/RAF/MEK/ERK pathway by RAF inhibitors to occur.

P. I. Poulikakos et al., RAF inhibitors transactivate RAF dimers and ERK signalling in cells with wild-type BRAF.Nature 13 February 2010 (doi: 10.1038/nature08902). [Abstract]

Navigate This Article