Portal de Periódicos da CAPES

Targeting the (Un)differentiated State of Cancer

2018; Cell Press; Volume: 33; Issue: 5 Linguagem: Inglês

10.1016/j.ccell.2018.04.007

1878-3686

Lajos V. Kemény, David E. Fisher,

Cancer Immunotherapy and Biomarkers

Dedifferentation in cancer is associated with intrinsic and acquired resistance to therapies. In this issue of Cancer Cell, Tsoi et al. identify four differentiation states in melanoma and provide evidence that melanoma cells develop drug resistance through a stepwise dedifferentiation process, making them vulnerable to ferroptotic cell death-inducing compounds. Dedifferentation in cancer is associated with intrinsic and acquired resistance to therapies. In this issue of Cancer Cell, Tsoi et al. identify four differentiation states in melanoma and provide evidence that melanoma cells develop drug resistance through a stepwise dedifferentiation process, making them vulnerable to ferroptotic cell death-inducing compounds. Modern cancer therapeutics have produced a medical revolution by attacking highly rational molecular targets. These range from mutated oncoproteins stuck in an "on" position to immune-based attacks that suppress tolerance in order to destroy neoantigen-laden malignancies. Clinical successes of these approaches have significantly altered patient outcomes, thereby stoking well-founded optimism. However, they have been met at every step by therapeutic resistance—either from the outset of treatment (acute) or following an initial response (acquired). Although durable major responses can now be a reality using these approaches, such successes remain frustratingly unpredictable and, for many cancers, rare. The search for mechanisms of resistance has been energetic and highly enlightening. Much resistance can be traced to acquired or selected mutations that "work around" drug-targeted vulnerabilities. These range from genomically selected alterations in drug targets or their pathways to mutations that disrupt antigen presentation capabilities. However, other resistance phenotypes lack clearly traceable mutational workarounds, instead apparently exploiting variations in epigenetic states to confer altered dependency (or immunologic evasion) that permits therapeutic escape. In this issue of Cancer Cell, Tsoi and colleagues (Tsoi et al., 2018Tsoi J. Robert L. Paraiso K. Galvan C. Sheu K.M. Lay J. Wong D.J.L. Atefi M. Shirazi R. Wang X. et al.Multi-stage differentiation defines melanoma subtypes with differential vulnerability to drug-induced iron-dependent oxidative stress.Cancer Cell. 2018; 33 (this issue): 890-904Abstract Full Text Full Text PDF PubMed Scopus (354) Google Scholar) present analyses of therapeutic resistance in melanoma traceable to undifferentiated lineage states. The authors highlight four differentiation states in melanoma identified by computational analyses of transcriptional profiles, verifying the presence of these distinct profiles in both cell lines and tumor tissue (Figure 1). Their findings highlight the plasticity of melanoma and offer a better understanding of varied states of differentiation, identifying intermediate transcriptional programs between fully differentiated and dedifferentiated states. They provide evidence that cells undergo this dedifferentiation while developing resistance to targeted therapies or to immune cytokines. Recent studies have suggested that some of these non-genomic resistance programs share a signature that has been identified in epithelial-to-mesenchymal transition (EMT), innate PD-1 resistance (IPRES), high AXL and/or ZEB1 expression, low MITF, or MAPK inhibitor resistance programs. A common feature of these programs in the context of melanoma is the expression of a dedifferentiated cellular state, even though melanoma is not an epithelial cell formally capable of undergoing EMT. The processes leading to this dedifferentiated state are not well understood, despite the high demand to identify better predictive markers and potential new therapeutic options. Interestingly, single-cell expression analysis of melanoma has identified fluidity across this spectrum within individual tumors and melanoma cell lines, which exhibit clear evidence of simultaneous high versus low MITF populations, whereas selection in vivo and in vitro for the undifferentiated cellular state occurs after BRAF/MEK inhibitor exposure (Müller et al., 2014Müller J. Krijgsman O. Tsoi J. Robert L. Hugo W. Song C. Kong X. Possik P.A. Cornelissen-Steijger P.D. Geukes Foppen M.H. et al.Low MITF/AXL ratio predicts early resistance to multiple targeted drugs in melanoma.Nat. Commun. 2014; 5: 5712Crossref PubMed Scopus (389) Google Scholar, Tirosh et al., 2016Tirosh I. Izar B. Prakadan S.M. Wadsworth 2nd, M.H. Treacy D. Trombetta J.J. Rotem A. Rodman C. Lian C. Murphy G. et al.Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq.Science. 2016; 352: 189-196Crossref PubMed Scopus (2026) Google Scholar). Differentiation of the melanocyte lineage (from which melanoma derives) is quite well understood. The master transcriptional regulator MITF plays a central role in conferring melanocytic differentiation from neural crest precursors (Kawakami and Fisher, 2017Kawakami A. Fisher D.E. The master role of microphthalmia-associated transcription factor in melanocyte and melanoma biology.Lab. Invest. 2017; 97: 649-656Crossref PubMed Scopus (132) Google Scholar, Lin and Fisher, 2007Lin J.Y. Fisher D.E. Melanocyte biology and skin pigmentation.Nature. 2007; 445: 843-850Crossref PubMed Scopus (851) Google Scholar). MITF responds to extracellular signals during development, as well as to environmental cues such as UV (tanning), by inducing expression of the differentiation-associated pigmentation machinery, including pigment enzymes, melanosomal factors, and proteins participating in melanosome export. Although MITF expression is a commonly employed diagnostic marker for melanoma (as a neoplasm of melanocytic origin), heterogeneity of MITF expression in melanoma has long been observed, with high versus low MITF expression explicitly suggested to confer different phenotypic behaviors (Hoek et al., 2008Hoek K.S. Eichhoff O.M. Schlegel N.C. Döbbeling U. Kobert N. Schaerer L. Hemmi S. Dummer R. In vivo switching of human melanoma cells between proliferative and invasive states.Cancer Res. 2008; 68: 650-656Crossref PubMed Scopus (490) Google Scholar). Importantly, aside from its pigmentation/differentiation activity, MITF shares DNA binding features with the Myc oncoprotein family, and MITF has been identified as an amplified or mutated oncogene in sporadic and familial melanoma (Lo and Fisher, 2014Lo J.A. Fisher D.E. The melanoma revolution: from UV carcinogenesis to a new era in therapeutics.Science. 2014; 346: 945-949Crossref PubMed Scopus (274) Google Scholar). The "MITF-low" state recognized by Tsoi and colleagues (Tsoi et al., 2018Tsoi J. Robert L. Paraiso K. Galvan C. Sheu K.M. Lay J. Wong D.J.L. Atefi M. Shirazi R. Wang X. et al.Multi-stage differentiation defines melanoma subtypes with differential vulnerability to drug-induced iron-dependent oxidative stress.Cancer Cell. 2018; 33 (this issue): 890-904Abstract Full Text Full Text PDF PubMed Scopus (354) Google Scholar) has been similarly described by other investigators in settings of treatment resistance (Tirosh et al., 2016Tirosh I. Izar B. Prakadan S.M. Wadsworth 2nd, M.H. Treacy D. Trombetta J.J. Rotem A. Rodman C. Lian C. Murphy G. et al.Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq.Science. 2016; 352: 189-196Crossref PubMed Scopus (2026) Google Scholar). Tsoi et al. further interrogated pharmacogenomics databases to identify drug vulnerabilities that correlate inversely with differentiation. They identified ferroptosis inducers as agents producing strikingly more potent lethality in less differentiated states. These findings buttress recent reports by Schreiber and colleagues (Hangauer et al., 2017Hangauer M.J. Viswanathan V.S. Ryan M.J. Bole D. Eaton J.K. Matov A. Galeas J. Dhruv H.D. Berens M.E. Schreiber S.L. et al.Drug-tolerant persister cancer cells are vulnerable to GPX4 inhibition.Nature. 2017; 551: 247-250Crossref PubMed Scopus (635) Google Scholar, Viswanathan et al., 2017Viswanathan V.S. Ryan M.J. Dhruv H.D. Gill S. Eichhoff O.M. Seashore-Ludlow B. Kaffenberger S.D. Eaton J.K. Shimada K. Aguirre A.J. et al.Dependency of a therapy-resistant state of cancer cells on a lipid peroxidase pathway.Nature. 2017; 547: 453-457Crossref PubMed Scopus (753) Google Scholar) that revealed distinctive vulnerabilities to ferroptosis-inducing agents in cancers with a variety of undifferentiated and other drug-resistant/persistence phenotypes. Ferroptosis refers to an iron-dependent, non-apoptotic cell death mediated by accumulation of lipid reactive oxygen species (ROS). Lipid peroxidation induced by cellular metabolic activities may be ameliorated by the action of phospholipid hydroperoxide glutathione peroxidase (GPX4), a selenium-containing enzyme that utilizes reduced glutathione to detoxify these lipid oxidation products. Antagonism of GPX4 results in accumulation of lipid peroxides that can participate in iron-dependent cellular lethality that is distinct mechanistically from apoptotic death. Agents that antagonize GPX4 to trigger ferroptosis may potentially act at varied levels upstream of GPX4 synthesis (statins) or through interference with glutathione reduction, but direct GPX4 targeting may be most likely to target ferroptosis while minimizing collateral toxicity. Nonetheless, it remains uncertain whether there is strong cancer specificity or therapeutic index available for in vivo induction of ferroptosis, because certain stem cell or progenitor populations within adults might be predicted to exhibit similar vulnerabilities. Indeed, Gpx4 knockout has been seen to produce neuronal loss and lethality, even when induced post-natally in adult mice (Yoo et al., 2012Yoo S.E. Chen L. Na R. Liu Y. Rios C. Van Remmen H. Richardson A. Ran Q. Gpx4 ablation in adult mice results in a lethal phenotype accompanied by neuronal loss in brain.Free Radic. Biol. Med. 2012; 52: 1820-1827Crossref PubMed Scopus (141) Google Scholar). For the melanocyte lineage, hair follicle stem cells are notably low in MITF (Lin and Fisher, 2007Lin J.Y. Fisher D.E. Melanocyte biology and skin pigmentation.Nature. 2007; 445: 843-850Crossref PubMed Scopus (851) Google Scholar) and might be anticipated to exhibit ferroptosis vulnerability, though gray hair would be more a pharmacodynamic marker of successful targeting than a medical concern. How might the dedifferentiated, mesenchymal-like state connect mechanistically to produce a distinctive dependency on cellular antagonism of ferroptosis? Perhaps the generation of lipid peroxidation products is distinctly elevated in this cell population. Alternatively, other endogenous ROS-antagonizing mechanisms might be less efficient at preventing formation of lipid oxidation products, although Tsoi and colleagues (Tsoi et al., 2018Tsoi J. Robert L. Paraiso K. Galvan C. Sheu K.M. Lay J. Wong D.J.L. Atefi M. Shirazi R. Wang X. et al.Multi-stage differentiation defines melanoma subtypes with differential vulnerability to drug-induced iron-dependent oxidative stress.Cancer Cell. 2018; 33 (this issue): 890-904Abstract Full Text Full Text PDF PubMed Scopus (354) Google Scholar) did not observe similar sensitivity features for non-ferroptosis ROS-inducing agents. MITF has been shown to regulate oxidative metabolism via direct targeting of the mitochondrial master regulator PGC1α, resulting in increased ROS detoxification capacities; however, the role of this pathway in regulating lipid ROS is incompletely understood. An additional intriguing feature of the MITF-low state is highlighted by the red hair/light-skin phenotype, which is associated with low MITF, elevated ROS (including oxidized lipid moieties in skin), and accelerated melanomagenesis (Lo and Fisher, 2014Lo J.A. Fisher D.E. The melanoma revolution: from UV carcinogenesis to a new era in therapeutics.Science. 2014; 346: 945-949Crossref PubMed Scopus (274) Google Scholar). It remains to be determined whether common mechanisms or distinct lineage-specific ones are ultimately responsible for the heightened ferroptotic vulnerability phenotype. Regardless of the underlying mechanism(s) producing ferroptosis vulnerability, the susceptibility appears to be defined by distinctive state(s) of cellular differentiation. The differentiated state, stated differently, represents altered epigenetic control of gene expression. Certain critical genes, such as MITF, regulate both lineage-specific differentiation targets and oncogenic/malignancy-associated ones. A burgeoning focus of cancer therapeutics is the identification of druggable epigenetic factors—some of which are approved (HDAC inhibitors) and many others of which are undergoing active preclinical and clinical investigation. Perhaps the dedifferentiated state defined by Graeber (Tsoi et al., 2018Tsoi J. Robert L. Paraiso K. Galvan C. Sheu K.M. Lay J. Wong D.J.L. Atefi M. Shirazi R. Wang X. et al.Multi-stage differentiation defines melanoma subtypes with differential vulnerability to drug-induced iron-dependent oxidative stress.Cancer Cell. 2018; 33 (this issue): 890-904Abstract Full Text Full Text PDF PubMed Scopus (354) Google Scholar), Schreiber (Viswanathan et al., 2017Viswanathan V.S. Ryan M.J. Dhruv H.D. Gill S. Eichhoff O.M. Seashore-Ludlow B. Kaffenberger S.D. Eaton J.K. Shimada K. Aguirre A.J. et al.Dependency of a therapy-resistant state of cancer cells on a lipid peroxidase pathway.Nature. 2017; 547: 453-457Crossref PubMed Scopus (753) Google Scholar), and others will exhibit additional vulnerabilities that were not examined in the pharmacogenomic datasets that identified ferroptosis inducers. Epigenetic factors required to retain the undifferentiated state or to maintain survival in that context may produce parallel druggable opportunities. And perhaps such approaches could, in the future, be combined with ferroptosis inducers to ameliorate high-dose single-agent toxicities while "triangulating" upon the common dedifferentiated state of aggressive, treatment-refractory cancer cells. D.E.F. gratefully acknowledges funding support from the NIH (2R01 AR043369-22, 1R01CA222871, 5P01 CA163222-05) and the Dr. Miriam and Sheldon G. Adelson Medical Research Foundation. D.E.F. has a financial interest in Soltego, Inc., a company developing SIK inhibitors for topical skin darkening treatments. Multi-stage Differentiation Defines Melanoma Subtypes with Differential Vulnerability to Drug-Induced Iron-Dependent Oxidative StressTsoi et al.Cancer CellApril 12, 2018In BriefTsoi et al. show that melanoma can be categorized into four subtypes following a differentiation trajectory with subtype-specific sensitivity to ferroptosis induction, which presents a therapeutic approach to target the differentiation plasticity to increase the efficacy of targeted and immune therapies. Full-Text PDF Open Archive