Systems biology analysis of mitogen activated protein kinase inhibitor resistance in malignant melanoma
- 1.9k Downloads
Kinase inhibition in the mitogen activated protein kinase (MAPK) pathway is a standard therapy for cancer patients with activating BRAF mutations. However, the anti-tumorigenic effect and clinical benefit are only transient, and tumors are prone to treatment resistance and relapse. To elucidate mechanistic insights into drug resistance, we have established an in vitro cellular model of MAPK inhibitor resistance in malignant melanoma.
The cellular model evolved in response to clinical dosage of the BRAF inhibitor, vemurafenib, PLX4032. We conducted transcriptomic expression profiling using RNA-Seq and RT-qPCR arrays. Pathways of melanogenesis, MAPK signaling, cell cycle, and metabolism were significantly enriched among the set of differentially expressed genes of vemurafenib-resistant cells vs control. The underlying mechanism of treatment resistance and pathway rewiring was uncovered to be based on non-genomic adaptation and validated in two distinct melanoma models, SK-MEL-28 and A375. Both cell lines have activating BRAF mutations and display metastatic potential.
Downregulation of dual specific phosphatases, tumor suppressors, and negative MAPK regulators reengages mitogenic signaling. Upregulation of growth factors, cytokines, and cognate receptors triggers signaling pathways circumventing BRAF blockage. Further, changes in amino acid and one-carbon metabolism support cellular proliferation despite MAPK inhibitor treatment. In addition, treatment-resistant cells upregulate pigmentation and melanogenesis, pathways which partially overlap with MAPK signaling. Upstream regulator analysis discovered significant perturbation in oncogenic forkhead box and hypoxia inducible factor family transcription factors.
The established cellular models offer mechanistic insight into cellular changes and therapeutic targets under inhibitor resistance in malignant melanoma. At a systems biology level, the MAPK pathway undergoes major rewiring while acquiring inhibitor resistance. The outcome of this transcriptional plasticity is selection for a set of transcriptional master regulators, which circumvent upstream targeted kinases and provide alternative routes of mitogenic activation. A fine-woven network of redundant signals maintains similar effector genes allowing for tumor cell survival and malignant progression in therapy-resistant cancer.
KeywordsCancer systems biology Precision medicine Omics RNA-Seq Transcriptomics Upstream regulator analysis Transcription factor Master regulator Regulome Non-genomic Rewiring Adaptation Genetic selection Drug resistance Therapy resistance Melanoma Melanogenesis
B-Raf proto-oncogene, serine/threonine kinase
dual specific phosphatase
half maximal inhibitory concentration
mitogen activated protein kinase
next generation mRNA transcriptome sequencing
RNA-Seq single-end reads in reads per kilobase million
real-time quantitative polymerase chain reaction
signal transducer and activator of transcription
Therapy resistance in cancer
Cancer drug resistance is a major obstacle in achieving durable clinical responses with targeted therapies. This highlights a need to elucidate the underlying mechanisms responsible for resistance and identify strategies to overcome this challenge. In malignant melanoma, activating point-mutations in the mitogen activated protein kinase (MAPK) pathway in BRAF kinase (B-Raf proto-oncogene, serine/threonine kinase, Gene ID: 673) [1, 2, 3] made it possible to develop potent kinase inhibitors matched to genotyped kinase mutations in precision medicine approaches [4, 5, 6]. In tumors expressing the oncoprotein BRAF(V600E), the inhibitor molecules vemurafenib, dabrafenib, and encorafenib are designed to lock the ATP binding site into an inactive conformation of the kinase , the preferred state of wild-type RAF proteins. Trametinib and cobimetinib target MAP2K7 (MEK, mitogen-activated protein kinase kinase 7, Gene ID: 5609), the BRAF target and downstream effector molecule. In MAPK signaling, combinations of specific inhibitors have proven to be superior to single-agent regimens: BRAF inhibitors (BRAFi) in combination with MEK inhibitors (MEKi) improved survival compared to single MAPK inhibitors (MAPKi) [7, 8, 9, 10]. However, many patients responding to small molecule inhibition of the MAPK pathway will develop resistance. Ultimately, disease progression will take place and patients relapse with lethal drug-resistant disease.
Mechanism of resistance beyond mutations
Acquired resistance has been shown to involve a diverse spectrum of oncogenic mutations in the MAPK pathway [11, 12, 13, 14, 15]. In addition, non-genomic activation of parallel signaling pathways has been noted . Cell-to-cell variability in BRAF(V600E) melanomas generates drug-tolerant subpopulations. Selection of genetically distinct, fully drug-resistant clones arise within a set of heterogeneous tumor cells surviving the initial phases of therapy due to drug adaptation . Non-genomic drug adaptation can be accomplished reproducibly in cultured cells, and combination therapies that block adaptive mechanisms in vitro have shown promise in improving rates and durability of response . Thus, better understanding of mechanisms involved in drug adaptation is likely to improve the effectiveness of melanoma therapy by delaying or controlling acquired resistance.
Cellular models of malignant melanoma
SK-MEL-28 and A375 are human skin malignant melanoma cell lines with BRAF(V600E) activation that are tumorigenic in xenografts [19, 20, 21, 22] (HTB-72 and CRL-1619, American Type Culture Collection, Manassas, VA). The cell lines are maintained in DMEM medium supplemented with 10% fetal bovine serum and antibiotics (10–017-CV, 35–010-CV, 30–002-CI Corning, Corning, NY). All experimental protocols were approved by the Institutional Review Boards at the University of California Merced and Irvine. The study was carried out as part of IRB UCM13–0025 of the University of California Merced and as part of dbGap ID 5094 on somatic mutations in cancer and conducted in accordance with the Helsinki Declaration of 1975.
BRAFi-resistant (BRAFi-R) models were obtained by challenging cancer cell lines with incrementally increasing vemurafenib (PLX4032, PubChem CID: 42611257, Selleckchem, Houston, TX) concentrations in the culture media. Starting at 0.25 μM, which matched the naïve half maximal inhibitory concentration (IC50) of the parental cell lines, the vemurafenib concentrations were increased every 7 days in an exponential series up to 100-fold the naïve IC50 concentrations. Following this 6-week selection protocol, vemurafenib-adapted, cancer therapy resistance models were maintained in media supplemented with 5.0 μM vemurafenib.
Transcriptomic profiling and differential gene expression analysis
Total RNA from malignant melanoma cells was extracted using a mammalian RNA mini preparation kit (RTN10-1KT, GenElute, Sigma EMD Millipore, Darmstadt, Germany) and then digested with deoxyribonuclease I (AMPD1-1KT, Sigma EMD Millipore, Darmstadt, Germany). Complementary DNA (cDNA) was synthesized using random hexamers (cDNA SuperMix, 95,048–500, Quanta Biosciences, Beverly, MA). The purified DNA library was sequenced using a HighSeq2500 (Illumina, San Diego, CA) at the University of California Irvine Genomics High-Throughput Facility. Purity and integrity of the nucleic acid samples were quantified using a Bioanalyzer (2100 Bioanalyzer, Agilent, Santa Clara, CA). Libraries for next generation mRNA transcriptome sequencing (RNA-Seq) analysis were generated using the TruSeq kit (Truseq RNA Library Prep Kit v2, RS-122-2001, Illumina, San Diego, CA). In brief, the workflow involves purifying the poly-A containing mRNA molecules using oligo-dT attached magnetic beads. Following purification, the mRNA is chemically fragmented into small pieces using divalent cations under elevated temperature. The cleaved RNA fragments are copied into first strand cDNA using reverse transcriptase and random primers. Second strand cDNA synthesis follows, using DNA polymerase I and RNase H. The cDNA fragments are end repaired by adenylation of the 3′ ends and ligated to barcoded adapters. The products are then purified and enriched by nine cycles of PCR to create the final cDNA library subjected to sequencing. The resulting libraries were validated by qPCR and size-quantified by a DNA high sensitivity chip (Bioanalyzer, 5067–4626, Agilent, Santa Clara, CA). Sequencing was performed using 50 base pair read length, single-end reads, and more than 107 reads per sample. Raw sequence reads in the file format for sequences with quality scores (FASTQ) were mapped to human reference Genome Reference Consortium GRCh38 using Bowtie alignment with an extended Burrows-Wheeler indexing for an ultrafast memory efficient alignment within the Tuxedo suite followed by Tophat to account for splice-isoforms [23, 24]. Read counts were scaled via the median of the geometric means of fragment counts across all libraries. Transcript abundance was quantified using normalized single-end RNA-Seq reads in read counts as well as reads per kilobase million (RPKM). Since single-end reads were acquired in the sequencing protocol, quantification of reads or fragments yielded similar results. Statistical testing for differential expression was based on read counts and performed using EdgeR in the Bioconductor toolbox . Differentially expressed genes were further analyzed using Ingenuity Pathways Analysis (IPA, Qiagen, Rewood City, CA), classification of transcription factors (TFClass), and gene set enrichment analysis (GSEA, Broad Institute, Cambridge, MA) [26, 27]. For real-time quantitative polymerase chain reaction (RT-qPCR) validation of RNA-Seq signals of differentially expressed target genes in BRAFi-R melanoma cells, gene expression profiles were analyzed using the ΔΔ threshold cycle (CT) method. Oligonucleotides spanning exon-exon-junctions of transcripts were used for RT-qPCR validation (Additional file 1: Table 1). Triple replicate samples were subjected to SYBR green (SYBR green master mix, PerfeCTa® SYBR® Green FastMix®, 95072-05k, Quanta Biosciences, Beverly, MA) RT-qPCR analysis in an Eco system (Illumina, San Diego). CT values were normalized using multiple housekeeping genes like actin beta (ACTB, Gene ID: 60), cyclophilin A (PPIA, peptidylprolyl isomerase A, Gene ID: 5478) and RNA polymerase II subunit A (POLR2A, GeneID: 5430).
Inhibitor cytotoxicity studies
Chemical BRAFi against BRAF(V600E), vemurafenib, was dissolved in dimethyl sulfoxide (DMSO, Sigma) as a 10.0 mM stock solution and used in treatments in final concentrations between 0.01 μM and 50.0 μM. Melanoma control experiments were carried out in the presence of equivalent amounts of DMSO solvent without drug. Cell viability was determined using a 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide (XTT, X4626, Sigma EMD Millipore, Darmstadt, Germany) absorbance assay by subtracting background readout at 650 nm from response readout at 570 nm wavelength. IC50 concentrations were determined after 72 h of drug treatment between 0.01–100 μM in two-fold dilution series. Analysis was performed using CalcuSyn (v2.0, Biosoft, Cambridge, UK).
Melanin pigment production of cultured cells was determined by colorimetric measurements normalized for total protein levels in arbitrary units [28, 29]. Melanoma cells were harvested by centrifugation at 3000 rpm (3830 g, Z326K, Labnet International, Edison, NJ) and dissolved in either 1.0 N NaOH for melanin assay or lysis 250 for protein assay. The cell lysates were sonicated, incubated at room temperature for 24 h, and cleared by centrifugation at 13,000 rpm for 10 min (17,000 g, Z326K, Labnet International, Edison, NJ). The absorption of the supernatant was measured at 475 nm in a spectrophotometer (Smartspec3000, Bio-Rad, Carlsbad, CA). Cells were lysed in mild denaturing conditions in lysis 250 buffer (25 mM Tris, [pH 7.5], 5 mM EDTA, 0.1% NP-40, 250 mM NaCl) containing proteinase inhibitors (10 μg/ml aprotinin, 10 μg/ml leupeptin, 10 μg/ml pepstatin, 5 μg/ml antipain, 1 mM phenylmethylsulfonyl fluoride). The total protein amount in the lysates was quantified using a colorimetric Bradford assay (5000001, Bio-Rad, Richmond, CA) at 595 nm and an incubation time of 30 min .
Generation of BRAFi-resistant melanoma cell lines
Some BRAFi-R cell lines showed structures typically observed in differentiated melanocytes (Fig. 1b-c). In the presence of 5 μM vemurafenib, however, the parental cells were not able to grow but the resistant cells proliferated comparable to naïve cell lines (Fig. 1d-e). For the SK-MEL-28 cell line, two resistant sublines were established. The resistant sublines displayed IC50 values of 11.5 ± 0.9 μM and 13.3 ± 1.2 μM for SK-MEL-28-BRAFi-R1 and SK-MEL-28-BRAFi-R2 respectively, which is approximately 10–20 fold of the IC50 in a low micro-molar range for the parental cells with 0.74 ± 0.05 μM. For the A375 cell line, the IC50 of the A375-BRAFi-R cell line was observed at 17.7 ± 1.5 μM, 22.7 fold of IC50 for the parental A375 cells with 0.78 ± 0.22 μM (Fig. 1f).
Transcriptomic profiling identifies non-genomic rewiring of treatment-resistant cancer cells
Upstream regulator analysis suggests control by transcription factor families
Validation of pathway rewiring in drug resistance in multiple cell lines by transcriptomics arrays
Transcriptome analysis of reversible drug resistance identified distinct pathways that allowed for circumvention of BRAF blockage (Fig. 4a). Cell-to-cell variability in combination with drug exposure selects for distinct sub-populations of MAPKi-resistant (MAPKi-R) cell lines. In a hierarchical fashion, transcriptional master regulators promote a distinct set of target genes resulting in circumvention of MAPK inhibition. Receptor activation by fibroblast growth factor 1 (FGF1, Gene ID: 2246) or PDGFC can lead to activated receptor tyrosine signaling parallel to canonical MAPK signaling  (Fig. 4b). In addition, downregulation of tumor suppressors reengages mitogenic signaling. The dual specific phosphatases, DUSP1 and DUSP2, have the ability to switch MAPK signaling off and rank among the top downregulated hits. Thus, downregulation of dual specific phosphatases facilitates and reinforces alternative MAPK effector activation under BRAF blockage (Fig. 4b). One of the mitogen-activated protein kinase 1 (MAPK1, ERK2, Gene ID: 5594) effector targets, transcription factor EPAS1, showed upregulation and the ability to maintain its transcriptional program. The pro-apoptotic program of TGFB3 was downregulated including SMAD family member 9 (SMAD9, Gene ID: 4093) and DUSP1/2 (Fig. 4c). Adenylate cyclase, G-protein, and phospholipase signaling are alternative cascades observed in cutaneous and uveal melanoma (Fig. 4d). Upregulation of ADCY1, endothelin receptor type B (EDNRB, Gene ID: 1910), phospholipase C beta 4 (PLCB4, Gene ID: 5332), and cAMP responsive element binding protein 3 (CREB3, Gene ID: 10488) promote MITF activity, the master transcription factor for pigmentation genes. Downstream metabolic enzymes, TYR and DCT, are both MITF target genes and contribute to enhanced eumelanin production observed in some therapy-resistant cell lines. The observed pigmentation showed a wide range of from 1.3-fold to up to 16.8-fold upregulation (Fig. 4d). While both cell lines showed dysregulation of melanogenesis, the regulators and effectors involved were different. SK-MEL-28-BRAFi-R2 has ASIP prominently expressed (TYR (2.1), DCT (2.8), tyrosinase related protein 1 (TYRP1, OCA3, Gene ID: 7306) (0.5), MITF (0.7), agouti signaling protein (ASIP, Gene ID: 434) (18.9)), while A375-BRAFi-R showed strongest regulation of TYRP1 and MITF (TYR (0.34), DCT (0.24), TYRP1 (41.8), MITF (2.94), ASIP (0.41)).
Activation of the MAPK pathway is the central and most common oncogenic event in the pathogenesis of malignant melanoma [3, 33]. About 50% of all melanoma patients have activating somatic mutations in the activator loop involving L597, T599, V600, and K601 switching proto-oncogene BRAF into a constitutively active protein kinase and cancer driver. Such activation is supported by somatic copy number amplifications of chromosome 7 , often coinciding with somatic V600E/G/K/M/R mutations. Another 20–30% of the patients show non-genomic activation of BRAF by transcriptional upregulation or post-translational modification induced by somatic mutations of upstream signaling molecules like KIT proto-oncogene receptor tyrosine kinase (KIT, Gene ID: 3815), proto-oncogene neuroblastoma RAS viral oncogene homolog (NRAS, Gene ID: 4893), or loss-of-function neurofibromin 1 (NF1, Gene ID: 4763). Constitutively activated BRAF phosphorylates MAPK1 and downstream kinases resulting in mitogenic signaling, proliferation, and cell growth. Integrated into this cellular program is negative feedback resulting in reduction of NRAS expression [35, 36].
Genomic and non-genomic mechanisms of therapy resistance
Genomic sequencing has facilitated the understanding of acquired resistance mechanisms to MAPKis [14, 15, 16, 37, 38, 39, 40]. Detected genetic aberrations included mutations in NRAS, MAPK1/2, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA, Gene ID: 5290), and phosphatase and tensin homolog (PTEN, Gene ID: 5728). Somatic melanoma mutations provide examples of how single, well-defined genomic events can confer resistance against vemurafenib treatment. In contrast, transcriptomic as well as epigenomic regulation can provide insight into resistance states that may involve larger networks. Eventually, resistance-conferring genomic, epigenomic, and transcriptomic alterations result in sustained mitogenic effector signaling and persist to promote proliferation.
Network rewiring of therapy-resistant melanoma
The transcriptomic profiles revealed a network of genes involved in adenylate cyclase signaling conferring resistance and contributing to melanogenesis. ADCY1 and CREB3 are prominent members of the melanogenesis pathway exhibiting mitogenic control and MITF activation. Similarly, a gain-of-function screen confirmed a cyclic-AMP-dependent melanocytic signaling network including G-protein-coupled receptors, adenylate cyclase, protein kinase cAMP-activated catalytic subunit alpha (PRKACA, Gene ID: 5566), and cAMP responsive element binding protein 1 (CREB1, Gene ID: 1385) . The MAPK pathway negatively regulates MITF protein level as well as activity , which in turn regulates a series of cell cycle regulating genes. In particular, P16INK4A and P21CIP1, gene products of cyclin dependent kinase inhibitor 2A (CDKN2A, Gene ID: 1029) and cyclin dependent kinase inhibitor 1A (CDKN1A, Gene ID: 1026), respectively, differentiation genes TYR, DCT, TYRP1 as well as survival genes B-cell lymphoma 2 apoptosis regulator (BCL2, Gene ID: 596) and BCL2 family apoptosis regulator (MCL1, Gene ID: 4170) are effector genes under the control of MITF. Indeed, inhibition of MITF increases sensitivity to chemotherapy drugs . In contrast, upregulation of MITF in therapy-resistance may present itself as a survival mechanism, which coincides with upregulation of melanin, hence it may serve as prognostic biomarker for drug adaptation.
Dual specific phosphatases (DUSPs) act downstream of BRAF on phosphorylated MAPK members to provide attenuation of signal. Loss of DUSP activity results in constitutive activation of the pathway. Prominent members of this family DUSP1 and DUSP2 are consistently downregulated at the transcriptional level. In prior clinical studies, somatic mutation of DUSP4 in MAPKi-R has been reported . Although in that case a genomic mechanism of resistance was utilized, the outcome of reduced DUSP activity by genomic or transcriptomic changes is equivalent and leads to persistent triggering of MAPK effectors.
Metabolic support of therapy resistance
Metabolic genes support the rewiring of acquired resistance and have been shown to play an intricate role in the malignancy of skin cutaneous tissues. Glutamine and glucose metabolism showed sensitivity to combinations of MAPKi and metabolic inhibitors in preclinical studies . The transciptomic profiles identified key enzymes in related, branching glycolytic pathways of serine, folate and pyrimidine metabolism. A cancer systems biology analysis of skin cutaneous melanoma brought forward a new master regulator and diagnostic target in cancer metabolism. Somatic mutations of DPYD have the ability to reconfigure and activate pyrimidine metabolism promoting rapid cellular proliferation and metastatic progression .
Concertation of transcriptional regulators
The forkhead box family of transcription factors is an important downstream target of the MAPK pathway and is currently being considered as a new therapeutic target in cancer, including melanoma therapy . In epithelial cells, these transcriptional factors are directly involved in the expression of cyclin dependent kinase inhibitors and CDKN2A gene under the control of TGFβ [46, 47]. Both downregulation of anti-apoptotic targets as well as activation of proliferative metabolism have been observed as mechanisms contributing to MAPKi-R. Downregulation of FOXF2 has been shown to promote cancer progression, EMT, and metastatic invasion . In contrast, a different member of the FOX family, the stem cell transcription factor forkhead box D3 (FOXD3) has been identified as an adaptive mediator of the response to MAPK pathway inhibition selectively in mutant BRAF melanomas [49, 50].
We have discovered non-genomic rewiring of pathways in chemotherapy resistance by RNA-Seq data and validated gene targets in two cell lines by transcriptomics arrays. Perturbation of these resistance pathways by drug molecules, RNA interference, or genomic editing will corroborate the translational impact of identified gene targets. The established cell culture models of treatment resistance provide a broadly applicable platform to utilize high-throughput screening tools in the search for effective combinations of targeted therapies in cancer.
The MAPK pathway undergoes major rewiring at the transcriptional level while acquiring inhibitor resistance. The outcome of such transcriptional plasticity is dysregulation at the level of different upstream master regulators, while maintaining similar effector genes. Combination therapies including targeted approaches and immune checkpoint inhibition are promising and rapidly improving. For these therapies to show durable, progression-free success in the clinical setting, adaptation mechanisms of treatment resistance need to be understood. Cellular model systems in combination with transcriptome-wide analyses provide insight into how non-genomic drug adaptation is accomplished. Ongoing efforts are focused on utilizing the established preclinical models to overcome drug adaptation as well as precision medicine profiling of cancer patients. Over time, a better understanding of mechanisms involved in drug adaptation is likely to improve the effectiveness of melanoma therapy by delaying or controlling acquired resistance.
We would like to thank Angela Garcia, Charles Fagundes, Garja Suner, Sandeep Sanghera, Taran Kaur, Kirandeep Kaur, Keedrian Olmstead, Stephen Wilson, and Rohit Gupta for help with maintaining the cellular models of drug-resistant cancer cells.
Availability of preprint publication
FVF is grateful for the support of grant CA154887 from the National Institutes of Health, National Cancer Institute. The research of the University of California Merced Systems Biology and Cancer Metabolism Laboratory is generously supported by University of California, Cancer Research Coordinating Committee CRN-17-427258, National Science Foundation, University of California Senate Graduate Research Council, and Health Science Research Institute program grants. FLS is supported by grant CA160756 from the National Institutes of Health, National Cancer Institute. FLM and FLS are in part supported by the Waltmar and Oxnard Foundations and Aldrich Chair Endowment.
Conception and design: FLS, FVF Establishing of cellular models, data acquisition, and analysis of data: HZ, DT, ZW, AF, PP, JL, SS, AS, AB, SYT, FLM, FLS, FVF. Preparation of figures, data analysis, interpretation, writing, review, and revision of the manuscript: FVF. Study supervision: FLS, FVF. All authors read and approved the final manuscript.
Ethics approval and consent to participate
All experimental protocols were approved by the Institutional Review Boards at the University of California Merced and Irvine. The study was carried out as part of IRB UCM13–0025 of the University of California Merced and as part of dbGap ID 5094 for study accession phs000178.v9.p8 on somatic mutations in cancer and conducted in accordance with the Helsinki Declaration of 1975.
There is no competing financial interest. FLM is co-Founder and Medical director of Cancer Prevention Pharmaceuticals with no direct implications on the conducted study on melanoma resistance.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 15.Moriceau G, Hugo W, Hong A, Shi H, Kong X, Yu CC, et al. Tunable-combinatorial mechanisms of acquired resistance limit the efficacy of BRAF/MEK cotargeting but result in melanoma drug addiction. Cancer Cell. 2015;27(2):240–56. https://doi.org/10.1016/j.ccell.2014.11.018.CrossRefPubMedPubMedCentralGoogle Scholar
- 27.Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102(43):15545–50. https://doi.org/10.1073/pnas.0506580102.CrossRefPubMedPubMedCentralGoogle Scholar
- 29.Liu F, Singh A, Yang Z, Garcia A, Kong Y, Meyskens FL, Jr. MiTF links Erk1/2 kinase and p21 CIP1/WAF1 activation after UVC radiation in normal human melanocytes and melanoma cells. Mol Cancer 2010; 9:214. doi: https://doi.org/10.1186/1476-4598-9-214.
- 32.Filipp FV. Epioncogenes in cancer—identification of epigenomic and transcriptomic cooperation-networks by multi-omics integration of ChIP-Seq and RNA-Seq data. Syst Biol. Meth Mol Biol. 2019;1800:101–21.Google Scholar
- 34.Tiffen J, Wilson S, Gallagher SJ, Hersey P, Filipp FV. Somatic copy number amplification and Hyperactivating somatic mutations of EZH2 correlate with DNA methylation and drive epigenetic silencing of genes involved in tumor suppression and immune responses in melanoma. Neoplasia. 2016;18(2):121–32. https://doi.org/10.1016/j.neo.2016.01.003.CrossRefPubMedPubMedCentralGoogle Scholar
- 35.Lito P, Pratilas CA, Joseph EW, Tadi M, Halilovic E, Zubrowski M, et al. Relief of profound feedback inhibition of mitogenic signaling by RAF inhibitors attenuates their activity in BRAFV600E melanomas. Cancer Cell. 2012;22(5):668–82. https://doi.org/10.1016/j.ccr.2012.10.009.CrossRefPubMedPubMedCentralGoogle Scholar
- 37.Wagle N, Van Allen EM, Treacy DJ, Frederick DT, Cooper ZA, Taylor-Weiner A, et al. MAP kinase pathway alterations in BRAF-mutant melanoma patients with acquired resistance to combined RAF/MEK inhibition. Cancer Discov. 2014;4(1):61–8. https://doi.org/10.1158/2159-8290.CD-13-0631.CrossRefPubMedGoogle Scholar
- 39.Johnson DB, Menzies AM, Zimmer L, Eroglu Z, Ye F, Zhao S, et al. Acquired BRAF inhibitor resistance: a multicenter meta-analysis of the spectrum and frequencies, clinical behaviour, and phenotypic associations of resistance mechanisms. Eur J Cancer. 2015;51(18):2792–9. https://doi.org/10.1016/j.ejca.2015.08.022.CrossRefPubMedPubMedCentralGoogle Scholar
- 40.Filipp FV. Precision medicine driven by cancer systems biology. Cancer Metastasis Rev. 2017;36(1):91–108.. https://doi.org/10.1007/s10555-017-9662-4.
- 42.Garraway LA, Widlund HR, Rubin MA, Getz G, Berger AJ, Ramaswamy S, et al. Integrative genomic analyses identify MITF as a lineage survival oncogene amplified in malignant melanoma. Nature. 2005;436(7047):117–22. https://doi.org/10.1038/nature03664.
- 51.Zecena H, Tveit D, Wang Z, Farhat A, Panchal P, Liu J, et al. Systems biology analysis of mitogen activated protein kinase inhibitor resistance in malignant melanoma bioRxiv; 2017. https://doi.org/10.1101/231142.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.