Portal de Periódicos da CAPES

The Brothers RAF

2010; Cell Press; Volume: 140; Issue: 2 Linguagem: Inglês

10.1016/j.cell.2010.01.013

1097-4172

Lawrence N. Kwong, Lynda Chin,

Cancer-related Molecular Pathways

Targeted molecular therapies for cancer treatment have shown promise, but also have limitations. In this issue, Heidorn et al., 2010Heidorn S.J. Milagre C. Whittaker S. Nourry A. Niculescu-Duvas I. Dhomen N. Hussain J. Reis-Filho J.S. Springer C.J. Pritchard C. Marais R. Cell. 2010; (this issue)PubMed Google Scholar find that a class of targeted molecular therapies with clinical effectiveness against one melanoma subtype may have adverse clinical effects in another. Targeted molecular therapies for cancer treatment have shown promise, but also have limitations. In this issue, Heidorn et al., 2010Heidorn S.J. Milagre C. Whittaker S. Nourry A. Niculescu-Duvas I. Dhomen N. Hussain J. Reis-Filho J.S. Springer C.J. Pritchard C. Marais R. Cell. 2010; (this issue)PubMed Google Scholar find that a class of targeted molecular therapies with clinical effectiveness against one melanoma subtype may have adverse clinical effects in another. Comprehensive multidimensional analysis of cancer genomes has revealed a staggering level of complexity and variability even within tumors of the same histopathological subtype. These data highlight the breadth and depth of the constellation of genomic alterations in cancer. Although canonical signaling pathways seem to be universally deregulated in cancer (Cancer Genome Atlas Research Network, 2008Cancer Genome Atlas Research NetworkNature. 2008; 455: 1061-1068Crossref PubMed Scopus (5223) Google Scholar), different components of a pathway can be altered in different tumors. This variability affects tumor responses to targeted therapies and could explain the limited activity observed clinically when targeted therapies are deployed across a group of cancer patients with the same histopathological subtype of tumor. Hence, great effort is being devoted to designing and developing therapeutic regimens tailored to patients whose tumors carry particular molecular features. A marquee example is the drug trastuzumab, used for treating breast cancer. Patients with breast tumors harboring amplification or overexpression of the HER2/NEU gene, the target of trastuzumab, are more likely to show a clinical response than patients with tumors that do not (Smith et al., 2007Smith I. Procter M. Gelber R.D. Guillaume S. Feyereislova A. Dowsett M. Goldhirsch A. Untch M. Mariani G. Baselga J. et al.HERA study team.Lancet. 2007; 369: 29-36Abstract Full Text Full Text PDF PubMed Scopus (1271) Google Scholar). But not all patients with such signature mutations show clinical benefit. One study reported that only six out of 100 breast cancer patients whose tumors carried HER2 amplification or overexpression derived survival benefits from trastuzumab treatment (Smith et al., 2007Smith I. Procter M. Gelber R.D. Guillaume S. Feyereislova A. Dowsett M. Goldhirsch A. Untch M. Mariani G. Baselga J. et al.HERA study team.Lancet. 2007; 369: 29-36Abstract Full Text Full Text PDF PubMed Scopus (1271) Google Scholar). Therefore, understanding what dictates responsiveness to therapy in the clinic is one of the greatest challenges the oncology community faces, with significant medical and economic implications. Now Heidorn et al., 2010Heidorn S.J. Milagre C. Whittaker S. Nourry A. Niculescu-Duvas I. Dhomen N. Hussain J. Reis-Filho J.S. Springer C.J. Pritchard C. Marais R. Cell. 2010; (this issue)PubMed Google Scholar, reporting in this issue, identify a new mechanism that underlies differential biochemical responsiveness to a targeted cancer therapy. They show, in melanoma, that drugs that specifically inhibit the oncogene BRAF are effective in the subpopulation of melanomas (∼45%) harboring BRAF mutations. However, of concern, they discovered that these drugs also unleash cancer-promoting effects in melanomas that harbor mutations in the RAS oncogene. These findings may come as a surprise, as the signaling proteins RAS and RAF are depicted as components of the same pathway, and RAF is in fact considered to be an immediate downstream signaling surrogate for RAS in the linear ERK activation cascade. That said, groundwork hinting at the complex interplay between RAS and RAF has been laid by studies on the role of CRAF in melanoma. RAS proteins (NRAS, KRAS, and HRAS) directly activate RAF proteins (ARAF, BRAF, and CRAF) as part of the oncogenic RAS-RAF-MEK-ERK signal transduction cascade. The key output of this pathway is the phosphorylation of the oncogenic kinase ERK to pERK, its active form. Thus, activation of any upstream element will result in ERK phosphorylation. Indeed, overexpression of RAS, RAF, or MEK isoforms (with the possible exception of ARAF) potently activates ERK. In melanoma, the majority of tumor cells with BRAF mutations harbor the V600E mutation, and activation of the signaling cascade is presumed to be mediated by deregulated BRAF (Figure 1A). In contrast, in melanoma cells with NRAS mutations, overactive NRAS signals through CRAF to activate downstream MEK (Figure 1B) (Dumaz et al., 2006Dumaz N. Hayward R. Martin J. Ogilvie L. Hedley D. Curtin J.A. Bastian B.C. Springer C. Marais R. Cancer Res. 2006; 66: 9483-9491Crossref PubMed Scopus (224) Google Scholar). This latter result underlies the drug responses observed by Heidorn et al., 2010Heidorn S.J. Milagre C. Whittaker S. Nourry A. Niculescu-Duvas I. Dhomen N. Hussain J. Reis-Filho J.S. Springer C.J. Pritchard C. Marais R. Cell. 2010; (this issue)PubMed Google Scholar when they targeted the redundant yet intricately interacting RAF proteins in melanoma cells. Armed with two specific BRAF inhibitors, PLX4720 and 885-A, the authors found that ERK phosphorylation was suppressed as expected in BRAF mutant melanoma cell lines (Figure 1C), but, surprisingly, was induced in RAS mutant melanoma cell lines (Figure 1D). To delineate the underlying mechanism of this paradoxical response, the authors then used either the MEK inhibitor PD184352 or the pan-RAF-inhibitor sorafenib to show that these compounds can block ERK phosphorylation in both BRAF mutant and NRAS mutant cell lines. The efficacy of PD184352 in both BRAF and NRAS mutant tumor cells suggests that the mechanism lies upstream of MEK, whereas the efficacy of sorafenib points to CRAF as a critical node. Notably, the authors show that the specific BRAF inhibitors induce physical binding of BRAF to CRAF, upon which BRAF can serve as an activating scaffold to enhance CRAF signaling to ERK. The authors use an array of different BRAF and CRAF mutants to dissect the underlying mechanisms: (1) non-RAS binding CRAFR89L and BRAFR188L mutant proteins demonstrate that physical interactions of RAF proteins with RAS are required to mediate the signal, (2) "gatekeeper threonine" BRAFT529N and CRAFT421N mutants that are drug resistant show that binding of a specific BRAF inhibitor to BRAF (and hence inhibition of BRAF activity) is required for CRAF activation and that concomitant inhibition of CRAF by pan-RAF inhibitors explains their suppression of ERK phosphorylation, and (3) kinase-dead BRAFD594A or BRAFD594V mutants mimic the effect and biochemistry of drug-mediated inhibition of BRAF with the V600E mutation. These findings are further supported by the demonstration that in a genetically engineered mouse model, targeted expression of mutant KRAS in melanocytes with wild-type BRAF produces only hyperpigmentation but when combined with expression of a kinase-dead BRAF (which may mimic the effects of specific BRAF inhibitors) results in malignant melanoma. This study and other work illuminates an intriguing and clinically relevant aspect of the dynamic interplay between BRAF and CRAF and emphasizes the importance of genetic context in dictating how these interactions impact biological phenotypes. BRAFV600E-mediated ERK activation is enhanced by the genetic inhibition of CRAF or by blocking the binding of active CRAF to BRAF; conversely, this ERK activation is suppressed by CRAF overexpression (Karreth et al., 2009Karreth F.A. DeNicola G.M. Winter S.P. Tuveson D.A. Mol. Cell. 2009; 36: 477-486Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar). Other work shows that when oncogenic BRAF is inhibited, an increase in CRAF levels can sometimes enable melanoma cells to acquire drug resistance (Montagut et al., 2008Montagut C. Sharma S.V. Shioda T. McDermott U. Ulman M. Ulkus L.E. Dias-Santagata D. Stubbs H. Lee D.Y. Singh A. et al.Elevated CRAF as a potential mechanism of acquired resistance to BRAF inhibition in melanoma.Cancer Res. 2008; 68: 4853-4861Crossref PubMed Scopus (413) Google Scholar). Additionally, use of short hairpin RNAs to inhibit both BRAF and CRAF in NRAS mutant melanoma xenografts potently suppresses tumor growth in vivo, whereas inhibition of RAF alone does not. This recapitulates the differential effects of specific BRAF versus pan-RAF drug inhibitors (Jaiswal et al., 2009Jaiswal B.S. Janakiraman V. Kljavin N.M. Eastham-Anderson J. Cupp J.E. Liang Y. Davis D.P. Hoeflich K.P. Seshagiri S. PLoS ONE. 2009; 4: e5717Crossref PubMed Scopus (96) Google Scholar). Notably, however, when CRAF is knocked down in these xenografts, ERK phosphorylation is virtually extinguished despite continued tumor growth, suggesting that BRAF may exhibit ERK-independent oncogenic activity. Finally, in melanoma cells harboring the low-activity BRAF mutations, D594G or G469E, inhibition of CRAF with either sorafenib or small interfering RNAs decreases ERK phosphorylation and induces apoptosis (Smalley et al., 2009Smalley K.S. Xiao M. Villanueva J. Nguyen T.K. Flaherty K.T. Letrero R. Van Belle P. Elder D.E. Wang Y. Nathanson K.L. Herlyn M. Oncogene. 2009; 28: 85-94Crossref PubMed Scopus (176) Google Scholar). Collectively, these studies highlight how the two RAFs interact in vivo. First, there are clear dissimilarities between BRAF and CRAF. When CRAF is inhibited in BRAFV600E cells, CRAF's inhibition of BRAF is lifted; when BRAF is inhibited in NRAS mutant cells, it binds to and positively activates CRAF (Figure 1D). Whether this depends on inherent differences between BRAF and CRAF or on the V600E mutation in BRAF remains to be determined. Second, dual inhibition of RAF may be a valid therapeutic approach either alone or in combination with MEK inhibition, if CRAF is responsible for acquired drug resistance. However, existing pan-RAF inhibitors such as sorafenib suffer from either nonspecific kinase targeting or weak in vivo activity. Thus, new second-generation pan-RAF inhibitors need to be developed. Finally, BRAF and CRAF exemplify the complexity of functionally redundant proteins: their effects on one another, including but not limited to direct binding, belie any straightforward interpretation. This RAF story will inform the study of other families of proteins whose redundant nature may be deeper than it seems. The Heidorn et al. study also provides a cautionary tale for those enamored with the linear pathways presented in textbooks, particularly when it comes to designing or developing targeted therapy clinical trials in cancer. The nonlinear complex interplay between RAS and RAF proteins described here is likely to be the norm rather than the exception. As drug design continues to focus on molecular targeting (e.g., herceptin, sorafenib, and imatinib), the nuances of a target's biological activity in specific cellular and genetic contexts will be crucial for predicting patient responses to the drug and potential toxicity. The full molecular effects of a single or even a multiple-target regimen must be thoroughly explored in a preclinical setting prior to, as well as alongside, clinical trials. Genome-wide unbiased analyses such as genomic, transcriptomic, and proteomic profiling will be necessary to identify unanticipated points of weakness as well as paths to drug resistance (Stommel et al., 2007Stommel J.M. Kimmelman A.C. Ying H. Nabioullin R. Ponugoti A.H. Wiedemeyer R. Stegh A.H. Bradner J.E. Ligon K.L. Brennan C. et al.Science. 2007; 318: 287-290Crossref PubMed Scopus (728) Google Scholar) (Linardou et al., 2008Linardou H. Dahabreh I.J. Kanaloupiti D. Siannis F. Bafaloukos D. Kosmidis P. Papadimitriou C.A. Murray S. Lancet Oncol. 2008; 9: 962-972Abstract Full Text Full Text PDF PubMed Scopus (636) Google Scholar). Heidorn and colleagues demonstrate that molecular analyses of established tumor cell lines reveal the potential risk of a promising drug type, the specific BRAF inhibitors, if used in unselected melanoma patient populations. For personalized medicine to be realized, we need not only a complete atlas of genomic events in cancers and how such events functionally interact, but also an understanding of how such dynamic networks respond to genetic and chemical interference in both preclinical and clinical settings. Such mechanistic insights into the biology of therapeutic targets and the drugs developed against them requires a seamless integration of basic, translational, and clinical efforts in cancer drug development. We thank R. Depinho and G.F. Draetta for helpful comments. L.N.K. is supported by the American Cancer Society; L.C. is the recipient of the Abby S. and Howard P. Milstein Innovation Award in Melanoma Research and is supported by the National Institutes of Health and the Belfer Institute for Applied Cancer Science. Kinase-Dead BRAF and Oncogenic RAS Cooperate to Drive Tumor Progression through CRAFHeidorn et al.CellJanuary 22, 2010In BriefWe describe a mechanism of tumorigenesis mediated by kinase-dead BRAF in the presence of oncogenic RAS. We show that drugs that selectively inhibit BRAF drive RAS-dependent BRAF binding to CRAF, CRAF activation, and MEK–ERK signaling. This does not occur when oncogenic BRAF is inhibited, demonstrating that BRAF inhibition per se does not drive pathway activation; it only occurs when BRAF is inhibited in the presence of oncogenic RAS. Kinase-dead BRAF mimics the effects of the BRAF-selective drugs and kinase-dead Braf and oncogenic Ras cooperate to induce melanoma in mice. Full-Text PDF Open Access