Melanoma is a highly aggressive form of skin cancer that originates from melanocytes. Approximately, 60% of melanoma tumors harbor a BRAF mutation, most often BRAF^V600E, which leads to hyperactivation of the mitogen-activated protein kinase (MAPK) pathway (Davies et al., 2002). Drugs targeting the BRAF and MAPK pathways are clinically important, but almost invariably, resistance arises within a short time period (Kugel and Aplin, 2014). Melanoma inherits its aggressive nature from its multipotent neural crest precursors that gives rise to various cells including melanocytes, glia, and adrenal cells (Le Douarin and Kalcheim, 1999; Le Douarin and Dupin, 2018). The developmental programme of neural crest cells is believed to be reinitiated during melanoma progression and dysregulation of neural crest genes is predictive of metastatic potential and negative prognosis in melanoma (Mascarenhas et al., 2010; Bailey et al., 2012; Kulesa et al., 2006). Various different studies, including gene expression studies of tumors, immunohistochemical analysis of melanoma samples and single-cell sequencing studies of patient-derived xenografts suggest the existence of different cell types in melanoma tumors. This cellular heterogeneity is believed to reflect the associated ability of tumor cells to switch their phenotype from proliferative, non-invasive cells to quiescent, invasive cells and back, thus allowing metastasis and the escape from therapeutic intervention (reviewed in Rambow et al., 2019). This has been summarized in the phenotype switching model which suggests that melanoma cells can switch between invasive and proliferative states allowing them to either grow and form tumor or metastasize to a new site (Rambow et al., 2019; Hoek and Goding, 2010). Understanding the molecular mechanisms underlying the phenotypic plasticity of melanoma cells is key to addressing the metastatic potential of melanoma cells.

The microphthalmia-associated transcription factor (MITF) is essential for melanocyte differentiation, proliferation, and survival. MITF is also important during melanomagenesis (reviewed in Goding and Arnheiter, 2019). This is best evidenced by the observations that the rare germline mutation E318K of MITF increases the susceptibility to melanoma and MITF has been shown to be amplified in 15% of melanoma tumors (Bertolotto et al., 2011; Garraway et al., 2005; Yokoyama et al., 2011). Importantly, MITF activity has been used as a proxy for the phenotype switching model with MITF^high cells characterized as proliferative, whereas MITF^low cells have been assigned a quiescent invasive phenotype (Carreira et al., 2006; Hoek et al., 2006). In fact, MITF has been proposed to act as a rheostat where the levels of MITF activity determine the phenotypic state of melanoma cells (reviewed in Rambow et al., 2019). Since MITF expression and activity are regulated by the various signaling pathways, the tumor microenvironment has been proposed to instruct phenotypic changes in melanoma cells and thus foster disease progression (Feige et al., 2011; Miskolczi et al., 2018; Widmer et al., 2013; Riesenberg et al., 2015). However, antibody staining suggests that cells lacking MITF are abundant in melanomas (Goodall et al., 2008) and single-cell sequencing of human xenotransplants and of zebrafish melanoma models suggest the existence of cells with very low MITF expression (Rambow et al., 2018; Travnickova et al., 2019). These cells belong to a population of cells believed to represent minimal residual disease, cells that remain viable upon drug exposure.

The extracellular matrix (ECM) is an important component of the tumor microenvironment as it provides cells with biochemical and structural support. In melanoma, expression of ECM proteins such as tenascin and fibronectin increases during disease progression (Frey et al., 2011). Focal adhesions not only offer physical attachment of cells to the ECM through the integrin receptor, but also initiate signaling cascades that regulate cell proliferation, migration, and survival (Mitra et al., 2005; Geiger et al., 2001; Playford and Schaller, 2004). A key focal adhesion signaling protein is Focal Adhesion Kinase (FAK), which activates the ERK pathway via Grb-FAK interactions (Schlaepfer et al., 1999). An important scaffolding protein at the focal adhesion complex is Paxillin (PXN) which recruits other proteins to the focal adhesion sites when phosphorylated by FAK and SRC (Deakin and Turner, 2008). Importantly, phosphorylation of PXN is critical for activation of RAF, MEK, and ERK and has been shown to confer drug resistance by activating Bcl-2 through ERK signaling (Wu et al., 2016; Sen et al., 2010; Ishibe et al., 2003; Sen et al., 2012; Hirata et al., 2015). This highlights the importance of identifying a molecular mechanism that confers cells with the ability to circumvent drug inhibition through phenotypic changes.

In this study, we show that MITF represses the expression of focal adhesion and ECM genes in melanoma cells and tissues. Our findings reveal a new role for MITF in regulating the expression of genes that are essential for creating the melanoma microenvironment, establishing a link to melanoma progression and drug resistance.

In this study, we have shown that MITF directly binds to and represses the expression of ECM, EMT and focal adhesion genes in human melanoma cells. We first observed this using our MITF-KO cells but verified our observations in other cell models by overexpression and knockdown of MITF using siRNA and inducible microRNA against MITF (miR-MITF) in melanoma cells. Importantly, we showed that MITF^low tumors in humans as well as in zebrafish have increased expression of ECM and focal adhesion genes. Together, our findings indicate that MITF acts as a transcriptional repressor of genes involved in ECM and focal adhesion.

A role for MITF as a repressor has been described in both melanoma cells and immune cells (Riesenberg et al., 2015; Hu et al., 2007). In myeloid precursor cells, MITF was shown to interact with EOS to recruit co-repressors to target genes (Hu et al., 2007), whereas in melanoma cells MITF bound directly to an E-box located in an enhancer of the c-JUN gene, leading to reduced expression of the gene (Riesenberg et al., 2015). Our results show that many of the genes whose expression is repressed by MITF are bound by MITF and contain E-boxes in their regulatory regions (Figure 2h). This suggests that direct binding of MITF is involved in their repression. Since we observed differences in secondary motifs between the repressed and activated genes, different co-factors may be involved in mediating the repression in each case.

The MITF-dependent rheostat model predicts that high MITF activity promotes proliferation, whereas low activity promotes invasion (Carreira et al., 2006). Consistent with the rheostat model, proliferation was severely reduced upon MITF knockout (Figure 7a,b). Unexpectedly, however, the migrative and invasive properties were reduced in both MITF-KO and MITF-KD (miR-MITF) cells (Figure 7c–h). Immunohistochemistry and single-cell sequencing studies of melanoma tumors have shown the existence of cells with low or no MITF expression (Goodall et al., 2008; Rambow et al., 2018; Travnickova et al., 2019). The involvement of MITF in migration has mostly been characterized using knockdown studies in melanoma cell lines using either siRNA or shRNA and by using Matrigel-coated Boyden chambers (Carreira et al., 2006; Bianchi-Smiraglia et al., 2017; Cheli et al., 2011); in these studies, knocking down MITF resulted in increased migration properties. Cheli et al., 2012 also injected melanoma cells into the tail vein of mice and showed increased formation of metastasis when MITF was knocked down. Two different pathways involved in migration were shown to be regulated by MITF; DIAPH1, a gene implicated in actin polymerization (Carreira et al., 2006), and the guanosine monophosphate reductase (GMPR) gene encoding an enzyme involved in regulating intracellular GTP levels (Bianchi-Smiraglia et al., 2017). Surprisingly, more recent studies by Falletta et al., 2017 and Vlčková et al., 2018 failed to observe any effects on migratory/invasive properties upon MITF knockdown using the same cell lines as were used in the previous studies. Falletta et al., 2017 suggested that the translation factor eIF2α was needed along with the nutrient sensor ATF4 to mediate invasion of melanoma cells (Falletta et al., 2017). Clearly, the experimental system used and additional triggers such as nutrient limitations might play a role in mediating the invasive phenotype. Interestingly, knocking down SMAD7 in melanoma cells resulted in a dual invasive-proliferative phenotype without affecting MITF expression (Tuncer et al., 2019).

Thus, we might speculate that the migratory phenotype is a transient event and that in order to achieve migration the cells need to be tested at a narrow time interval where MITF activity is decreasing. Carmit Levy's group has recently shown MITF oscillations upon UV exposure in order to synchronize stress response and pigmentation (Malcov-Brog et al., 2018). It is possible that such a mechanism exists, for example during melanoblast development where oscillating MITF expression might ensure that proliferation and migration can both take place but not at the same time. Previous work has suggested oscillations in the dependence of melanoblasts on the receptor tyrosine kinase KIT (Yoshida et al., 1996; Hou et al., 2000). It is plausible that activation of invasion genes requires discontinuous presence of MITF along with other co-factors at specific time windows. According to the phenotype switching model of melanoma the reversible switch between proliferative and invasive state is needed for melanoma progression (Arozarena and Wellbrock, 2019). In our case, the complete loss of MITF might trap the cells in a state where the MITF switch is dead and migration therefore not possible. This would then explain why we do not see the invasive gene signature and no changes in the expression of DIAPH1, p27, or GMPR. Our analysis of published transcriptomic signatures of melanoma cells and tumors validated that upon MITF loss the expression of differentiation and proliferation genes was diminished and the expression of drug resistant and neural crest-like program was enriched (Figure 2c). Importantly, comparing our gene signature with that of 86 melanoma cells (Hoek et al., 2006) showed that the ∆MITF-X6 cells displayed a more prominent loss of proliferative genes than a gain of invasive genes (Figure 2c). Similarly, Rambow et al., 2015 showed that melanoma cell lines (G1, T1, 501Mel, MNT-1, SKMel3) which express high levels of MITF displayed a high proliferation potential but less migratory, invasive, and subcutaneous tumor growth capabilities than cell lines with low MITF expression (WM1366, WM793, WM852, LU1205, A375M) (Rambow et al., 2015). The expression of TRIM63 and CAPN3 was over-represented in the cell lines with high MITF expression and knocking down MITF led to a reduction of their expression. Interestingly, knocking down either TRIM63 or CAPN3 enhanced the invasive potential of 501Mel cells. In the ∆MITF-X6 cells, we observed reduction only of TRIM63 expression (Rambow et al., 2015). Thus, the transcriptome of our ∆MITF-X6 cells does not fully recapitulate the gene signature of invasive cells.

We identified MITF as an important transcriptional regulator of ECM and focal adhesion genes. Interestingly, we observed increased expression of TGFß1, encoding an important regulator of ECM-related genes, in the MITF-KO cells and MITF^low melanoma tumors (Figure 4b,c). It has been shown that TGFß1 supresses the expression of MITF in melanoblasts, thereby inhibiting differentiation into melanocytes (Nishimura et al., 2010). This autocrine signaling of TGFβ is retained in melanoma cells (Javelaud et al., 2008). According to Hoek et al., 2006, the MITF^low transcriptional state is dictated by TGFß1 signaling, which can suppress MITF expression resulting in an invasive and drug-resistant phenotype (Hoek et al., 2006; Miskolczi et al., 2018). This suggests that the genes induced upon MITF loss are partly due to induction of TGFß signaling. However, our results suggest that MITF is directly involved in mediating the observed effects on the expression of ECM and focal adhesion genes. In addition, the relationship between MITF and the expression of TGFß1 is not clear. Our observations suggest that knocking down MITF leads to a major increase in TGFß1 mRNA expression in the melanoma cells, suggesting that the effects are cell-autonomous and driven by MITF. However, there are no MITF-peaks in or near the TGFß1 gene in melanoma cells, leading us to hypothesize that the effects must be mediated through a hitherto unknown intermediary.

Enhanced expression and phosphorylation of paxillin has been linked to therapy resistance in other cancer cell types, such as lung cancer (Wu et al., 2016). In melanoma, an inverse relation between BRAF inhibition and the expression of ECM genes has been described as a marker of de-differentiated drug-resistant cells (Fallahi-Sichani et al., 2017). Our data showed that the number of paxillin-positive dots was induced in both MITF-KO and miR-MITF cells as compared to controls (Figure 6—figure supplement 1a,b,f,g h) and paxillin expression was inversely correlated with MITF expression in melanoma tissues and cell lines (Figure 6—figure supplement 1c–e). Interestingly, we found that treating cells devoid of MITF with a BRAF inhibitor resulted in an increase in formation of focal adhesions (Figure 6a–f). It is worth mentioning that an increase in the number of focal adhesions was restricted to SkMel28 melanoma cells in which MITF protein level was reduced upon vemurafenib treatment (Figure 6d,f, Figure 6—figure supplement 2a, b). However, we did not observe a significant increase in the number of focal adhesions in the 501Mel cells that gained MITF upon vemurafenib treatment (Figure 6c,e, Figure 6—figure supplement 2a, b). This highlights the role of MITF as a mediator of focal adhesion formation. However, how the synergistic effects of MITF and vemurafenib on focal adhesion formation are mediated is unclear. One way to explain an increase in the formation of focal adhesions is that it is due to integrin clustering that is essential for the activation of focal adhesion pathways (Harburger and Calderwood, 2009; Humphries et al., 2006). We observed an increase in the expression of several integrins including ITGA1, ITGA2, ITGA6, ITGA10, and ITGB3 in the MITF-KO cells, as well as in the siMITF 501Mel and SkMel28 cell lines (Supplementary file 5). In addition to this, the FLT1 receptor tyrosine kinase (VEGFR1) and its ligand VEGFA, which activate a pathway that phosphorylates FAK, a key mediator of focal adhesions, were increased in expression. Interestingly, both FLT1 and VEGFA have MITF-binding sites in their promoters and MITF has previously been shown to regulate VEGFA expression (Louphrasitthiphol et al., 2019). Exposure of melanoma cells to BRAF and MEK inhibitors has been shown to slow growth and result in increased expression of NGFR and ECM and focal adhesion genes (Fallahi-Sichani et al., 2017). Consistent with these findings, we observed an up to 200-fold induction of the NGFR transcript in the MITF-KO cells compared to EV-SkMel28 cells, and we identified an MITF peak in the 3’UTR of NGFR in both the MITF CUT-and-RUN data from SkMel28 cells and in the MITF ChIP-seq data from COLO829 cells (Webster et al., 2014); expression of the melanocyte differentiation marker and MITF target MLANA was 50- to 80-fold reduced in the MITF-KO cells (Figure 2—figure supplement 2a–d). Thus, it is possible that MITF affects focal adhesions by both directly regulating expression of genes involved in the process and indirectly by activating the expression of signaling processes involved. Importantly, we found that cell survival of MITF- KO cells might be dependent on FAK pathway (Figure 6g,h); therefore, this might be the therapeutic vulnerability of MITF-low melanoma cells and can potentially enhance the current treatment options of melanoma.

Upon MITF loss, an EMT-like process has been described to be involved in driving drug resistance in melanoma (Denecker et al., 2014; Caramel et al., 2013). In addition, the degree of plasticity between EMT and mesenchymal to epithelial transition (MET) has been suggested to lead to high metastatic potential as well as therapeutic resistance (Stylianou et al., 2019; Pastushenko et al., 2018; Thompson and Nagaraj, 2018). Indeed, we observed changes in important EMT markers and regulators such as ZEB1, CDH1 (E-Cadherin), CDH2 (N-Cadherin), SNAI2, and TGFß1 in the MITF-KO cells (Figure 4b–e) as well as in TCGA melanoma samples. Also, the MITF-KO cells showed increased expression of SOX2, which is important for neuronal stem cell maintenance and has been suggested to be important for self-renewal of melanoma tumor cells (Taranova et al., 2006; Santini et al., 2014; Figure 4b). Importantly, the effects of MITF on the expression of E-Cadherin, N-Cadherin, and ECM genes (ITGA2 and SERPINA3) is reversible (Figure 5e–j). This suggests that MITF enables epithelial to mesenchymal plasticity (EMP) that allows the formation of a hybrid state between EMT and MET to enforce the aggressiveness of melanoma. The binary effects of MITF on the expression of EMT genes may be the molecular mechanism that explains its rheostat activity.

The minimal residual disease (MRD) is a major reason for relapse in cancer. We found that ∆MITF-X6 cells are positively correlated with the gene signature of a population of MRD cells in melanoma tumors as determined by single-cell RNA-seq of human PDX samples and zebrafish melanoma models (Figure 6—figure supplement 2c–e; Figure 2c; Rambow et al., 2018). This gene signature was different from the ‘invasive gene signature’ that the authors observed in a different set of melanoma cells (Rambow et al., 2018). Interestingly, the MRD melanoma cells in zebrafish express little to no MITF protein and have increased expression of ECM genes (Figure 6—figure supplement 2e). This suggests that the induced expression of ECM genes and low expression of MITF is one of the markers of MRD in melanoma. Thus, permanently losing MITF reprograms gene expression toward the drug-resistant state, suggesting that MITF-KO cells can be a tool to study drug resistance in melanoma. In the absence of MITF, melanoma cells may become MRD cells by reshaping their ECM, enhancing their attachment to the surface, thus forming quiescent cells which wait for an opportunity to change their phenotype and re-emerge as proliferative melanoma cells. Since melanoma cells can mediate these effects on their own, in the absence of the tumor microenvironment, this suggests that this process is cell-autonomous and under the direction of MITF which instructs the cells to create their own microenvironment.

MITF CUT&RUN sequencing data have been deposited in GEO under accession codes GSE153020 and the RNA-Seq data discussed in this publication are available under the accession number GSE163646.

1. 1. Dilshat R

   (2020) NCBI Gene Expression Omnibus

   ID GSE163646. MITF reprograms the extracellular matrix and focal adhesion in melanoma.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE163646
2. 1. Kenny C
   2. Cornell RA

   (2021) NCBI Gene Expression Omnibus

   ID GSE153020. Cut and Run data of MITF in SKmel28 cells.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE153020

1. 1. Webster DE
   2. Barajas B
   3. Bussat RT
   4. Yan KJ

   (2014) NCBI Gene Expression Omnibus

   ID GSE50681. MITF ChIP-seq in primary melanocyte and melanoma as a function of oncogenic BRAF.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE50681
2. 1. Laurette P
   2. Strub T
   3. Koludrovic D
   4. Keime C

   (2015) NCBI Gene Expression Omnibus

   ID GSE61965. BRG1 recruitment by transcription factors MITF and SOX10 defines a specific configuration of regulatory elements in the melanocyte lineage (ChIP-seq).

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE61965

1. Software

   1. Andrews S

   (2010) FastQC: a quality control tool for high throughput sequence data, version GPL v3 

   Bioinformatics Babraham.

   http://www.bioinformatics.babraham.ac.uk/projects/fastqc
2. 1. Arozarena I
   2. Wellbrock C

   (2019) Phenotype plasticity as enabler of melanoma progression and therapy resistance

   Nature Reviews Cancer 19:377–391.

   https://doi.org/10.1038/s41568-019-0154-4

   • PubMed
   • Google Scholar
3. 1. Bailey CM
   2. Morrison JA
   3. Kulesa PM

   (2012) Melanoma revives an embryonic migration program to promote plasticity and invasion

   Pigment Cell & Melanoma Research 25:573–583.

   https://doi.org/10.1111/j.1755-148X.2012.01025.x

   • PubMed
   • Google Scholar
4. 1. Bertolotto C
   2. Lesueur F
   3. Giuliano S
   4. Strub T
   5. de Lichy M
   6. Bille K
   7. Dessen P
   8. d'Hayer B
   9. Mohamdi H
   10. Remenieras A
   11. Maubec E
   12. de la Fouchardière A
   13. Molinié V
   14. Vabres P
   15. Dalle S
   16. Poulalhon N
   17. Martin-Denavit T
   18. Thomas L
   19. Andry-Benzaquen P
   20. Dupin N
   21. Boitier F
   22. Rossi A
   23. Perrot JL
   24. Labeille B
   25. Robert C
   26. Escudier B
   27. Caron O
   28. Brugières L
   29. Saule S
   30. Gardie B
   31. Gad S
   32. Richard S
   33. Couturier J
   34. Teh BT
   35. Ghiorzo P
   36. Pastorino L
   37. Puig S
   38. Badenas C
   39. Olsson H
   40. Ingvar C
   41. Rouleau E
   42. Lidereau R
   43. Bahadoran P
   44. Vielh P
   45. Corda E
   46. Blanché H
   47. Zelenika D
   48. Galan P
   49. Aubin F
   50. Bachollet B
   51. Becuwe C
   52. Berthet P
   53. Bignon YJ
   54. Bonadona V
   55. Bonafe JL
   56. Bonnet-Dupeyron MN
   57. Cambazard F
   58. Chevrant-Breton J
   59. Coupier I
   60. Dalac S
   61. Demange L
   62. d'Incan M
   63. Dugast C
   64. Faivre L
   65. Vincent-Fétita L
   66. Gauthier-Villars M
   67. Gilbert B
   68. Grange F
   69. Grob JJ
   70. Humbert P
   71. Janin N
   72. Joly P
   73. Kerob D
   74. Lasset C
   75. Leroux D
   76. Levang J
   77. Limacher JM
   78. Livideanu C
   79. Longy M
   80. Lortholary A
   81. Stoppa-Lyonnet D
   82. Mansard S
   83. Mansuy L
   84. Marrou K
   85. Matéus C
   86. Maugard C
   87. Meyer N
   88. Nogues C
   89. Souteyrand P
   90. Venat-Bouvet L
   91. Zattara H
   92. Chaudru V
   93. Lenoir GM
   94. Lathrop M
   95. Davidson I
   96. Avril MF
   97. Demenais F
   98. Ballotti R
   99. Bressac-de Paillerets B
   100. French Familial Melanoma Study Group

   (2011) A SUMOylation-defective MITF germline mutation predisposes to melanoma and renal carcinoma

   Nature 480:94–98.

   https://doi.org/10.1038/nature10539

   • PubMed
   • Google Scholar
5. 1. Bianchi-Smiraglia A
   2. Bagati A
   3. Fink EE
   4. Moparthy S
   5. Wawrzyniak JA
   6. Marvin EK
   7. Battaglia S
   8. Jowdy P
   9. Kolesnikova M
   10. Foley CE
   11. Berman AE
   12. Kozlova NI
   13. Lipchick BC
   14. Paul-Rosner LM
   15. Bshara W
   16. Ackroyd JJ
   17. Shewach DS
   18. Nikiforov MA

   (2017) Microphthalmia-associated transcription factor suppresses invasion by reducing intracellular GTP pools

   Oncogene 36:84–96.

   https://doi.org/10.1038/onc.2016.178

   • PubMed
   • Google Scholar
6. 1. Bray NL
   2. Pimentel H
   3. Melsted P
   4. Pachter L

   (2016) Near-optimal probabilistic RNA-seq quantification

   Nature Biotechnology 34:525–527.

   https://doi.org/10.1038/nbt.3519

   • PubMed
   • Google Scholar
7. 1. Cano A
   2. Pérez-Moreno MA
   3. Rodrigo I
   4. Locascio A
   5. Blanco MJ
   6. del Barrio MG
   7. Portillo F
   8. Nieto MA

   (2000) The transcription factor snail controls epithelial-mesenchymal transitions by repressing E-cadherin expression

   Nature Cell Biology 2:76–83.

   https://doi.org/10.1038/35000025

   • PubMed
   • Google Scholar
8. 1. Caramel J
   2. Papadogeorgakis E
   3. Hill L
   4. Browne GJ
   5. Richard G
   6. Wierinckx A
   7. Saldanha G
   8. Osborne J
   9. Hutchinson P
   10. Tse G
   11. Lachuer J
   12. Puisieux A
   13. Pringle JH
   14. Ansieau S
   15. Tulchinsky E

   (2013) A switch in the expression of embryonic EMT-inducers drives the development of malignant melanoma

   Cancer Cell 24:466–480.

   https://doi.org/10.1016/j.ccr.2013.08.018

   • PubMed
   • Google Scholar
9. 1. Carreira S
   2. Goodall J
   3. Denat L
   4. Rodriguez M
   5. Nuciforo P
   6. Hoek KS
   7. Testori A
   8. Larue L
   9. Goding CR

   (2006) Mitf regulation of Dia1 controls melanoma proliferation and invasiveness

   Genes & Development 20:3426–3439.

   https://doi.org/10.1101/gad.406406

   • PubMed
   • Google Scholar
10. 1. Cheli Y
    2. Giuliano S
    3. Guiliano S
    4. Botton T
    5. Rocchi S
    6. Hofman V
    7. Hofman P
    8. Bahadoran P
    9. Bertolotto C
    10. Ballotti R

    (2011) Mitf is the key molecular switch between mouse or human melanoma initiating cells and their differentiated progeny

    Oncogene 30:2307–2318.

    https://doi.org/10.1038/onc.2010.598

    • PubMed
    • Google Scholar
11. 1. Cheli Y
    2. Giuliano S
    3. Fenouille N
    4. Allegra M
    5. Hofman V
    6. Hofman P
    7. Bahadoran P
    8. Lacour JP
    9. Tartare-Deckert S
    10. Bertolotto C
    11. Ballotti R

    (2012) Hypoxia and MITF control metastatic behaviour in mouse and human melanoma cells

    Oncogene 31:2461–2470.

    https://doi.org/10.1038/onc.2011.425

    • PubMed
    • Google Scholar
12. 1. Colaprico A
    2. Silva TC
    3. Olsen C
    4. Garofano L
    5. Cava C
    6. Garolini D
    7. Sabedot TS
    8. Malta TM
    9. Pagnotta SM
    10. Castiglioni I
    11. Ceccarelli M
    12. Bontempi G
    13. Noushmehr H

    (2016) TCGAbiolinks: an R/Bioconductor package for integrative analysis of TCGA data

    Nucleic Acids Research 44:e71.

    https://doi.org/10.1093/nar/gkv1507

    • PubMed
    • Google Scholar
13. 1. Davies H
    2. Bignell GR
    3. Cox C
    4. Stephens P
    5. Edkins S
    6. Clegg S
    7. Teague J
    8. Woffendin H
    9. Garnett MJ
    10. Bottomley W
    11. Davis N
    12. Dicks E
    13. Ewing R
    14. Floyd Y
    15. Gray K
    16. Hall S
    17. Hawes R
    18. Hughes J
    19. Kosmidou V
    20. Menzies A
    21. Mould C
    22. Parker A
    23. Stevens C
    24. Watt S
    25. Hooper S
    26. Wilson R
    27. Jayatilake H
    28. Gusterson BA
    29. Cooper C
    30. Shipley J
    31. Hargrave D
    32. Pritchard-Jones K
    33. Maitland N
    34. Chenevix-Trench G
    35. Riggins GJ
    36. Bigner DD
    37. Palmieri G
    38. Cossu A
    39. Flanagan A
    40. Nicholson A
    41. Ho JW
    42. Leung SY
    43. Yuen ST
    44. Weber BL
    45. Seigler HF
    46. Darrow TL
    47. Paterson H
    48. Marais R
    49. Marshall CJ
    50. Wooster R
    51. Stratton MR
    52. Futreal PA

    (2002) Mutations of the BRAF gene in human Cancer

    Nature 417:949–954.

    https://doi.org/10.1038/nature00766

    • PubMed
    • Google Scholar
14. 1. Deakin NO
    2. Turner CE

    (2008) Paxillin comes of age

    Journal of Cell Science 121:2435–2444.

    https://doi.org/10.1242/jcs.018044

    • PubMed
    • Google Scholar
15. 1. Denecker G
    2. Vandamme N
    3. Akay O
    4. Koludrovic D
    5. Taminau J
    6. Lemeire K
    7. Gheldof A
    8. De Craene B
    9. Van Gele M
    10. Brochez L
    11. Udupi GM
    12. Rafferty M
    13. Balint B
    14. Gallagher WM
    15. Ghanem G
    16. Huylebroeck D
    17. Haigh J
    18. van den Oord J
    19. Larue L
    20. Davidson I
    21. Marine JC
    22. Berx G

    (2014) Identification of a ZEB2-MITF-ZEB1 transcriptional network that controls melanogenesis and melanoma progression

    Cell Death & Differentiation 21:1250–1261.

    https://doi.org/10.1038/cdd.2014.44

    • PubMed
    • Google Scholar
16. 1. Fallahi-Sichani M
    2. Becker V
    3. Izar B
    4. Baker GJ
    5. Lin JR
    6. Boswell SA
    7. Shah P
    8. Rotem A
    9. Garraway LA
    10. Sorger PK

    (2017) Adaptive resistance of melanoma cells to RAF inhibition via reversible induction of a slowly dividing de-differentiated state

    Molecular Systems Biology 13:905.

    https://doi.org/10.15252/msb.20166796

    • PubMed
    • Google Scholar
17. 1. Falletta P
    2. Sanchez-Del-Campo L
    3. Chauhan J
    4. Effern M
    5. Kenyon A
    6. Kershaw CJ
    7. Siddaway R
    8. Lisle R
    9. Freter R
    10. Daniels MJ
    11. Lu X
    12. Tüting T
    13. Middleton M
    14. Buffa FM
    15. Willis AE
    16. Pavitt G
    17. Ronai ZA
    18. Sauka-Spengler T
    19. Hölzel M
    20. Goding CR

    (2017) Translation reprogramming is an evolutionarily conserved driver of phenotypic plasticity and therapeutic resistance in melanoma

    Genes & Development 31:18–33.

    https://doi.org/10.1101/gad.290940.116

    • PubMed
    • Google Scholar
18. 1. Feige E
    2. Yokoyama S
    3. Levy C
    4. Khaled M
    5. Igras V
    6. Lin RJ
    7. Lee S
    8. Widlund HR
    9. Granter SR
    10. Kung AL
    11. Fisher DE

    (2011) Hypoxia-induced transcriptional repression of the melanoma-associated oncogene MITF

    PNAS 108:E924–E933.

    https://doi.org/10.1073/pnas.1106351108

    • PubMed
    • Google Scholar
19. 1. Fock V
    2. Gudmundsson SR
    3. Gunnlaugsson HO
    4. Stefansson JA
    5. Ionasz V
    6. Schepsky A
    7. Viarigi J
    8. Reynisson IE
    9. Pogenberg V
    10. Wilmanns M
    11. Ogmundsdottir MH
    12. Steingrimsson E

    (2019) Subcellular localization and stability of MITF are modulated by the bHLH-Zip domain

    Pigment Cell & Melanoma Research 32:12721.

    https://doi.org/10.1111/pcmr.12721

    • Google Scholar
20. 1. Frey K
    2. Fiechter M
    3. Schwager K
    4. Belloni B
    5. Barysch MJ
    6. Neri D
    7. Dummer R

    (2011) Different patterns of fibronectin and tenascin-C splice variants expression in primary and metastatic melanoma lesions

    Experimental Dermatology 20:685–688.

    https://doi.org/10.1111/j.1600-0625.2011.01314.x

    • PubMed
    • Google Scholar
21. 1. Garraway LA
    2. Widlund HR
    3. Rubin MA
    4. Getz G
    5. Berger AJ
    6. Ramaswamy S
    7. Beroukhim R
    8. Milner DA
    9. Granter SR
    10. Du J
    11. Lee C
    12. Wagner SN
    13. Li C
    14. Golub TR
    15. Rimm DL
    16. Meyerson ML
    17. Fisher DE
    18. Sellers WR

    (2005) Integrative genomic analyses identify MITF as a lineage survival oncogene amplified in malignant melanoma

    Nature 436:117–122.

    https://doi.org/10.1038/nature03664

    • PubMed
    • Google Scholar
22. 1. Geiger B
    2. Bershadsky A
    3. Pankov R
    4. Yamada KM

    (2001) Transmembrane crosstalk between the extracellular matrix--cytoskeleton crosstalk

    Nature Reviews Molecular Cell Biology 2:793–805.

    https://doi.org/10.1038/35099066

    • PubMed
    • Google Scholar
23. 1. Giuliano S
    2. Cheli Y
    3. Ohanna M
    4. Bonet C
    5. Beuret L
    6. Bille K
    7. Loubat A
    8. Hofman V
    9. Hofman P
    10. Ponzio G
    11. Bahadoran P
    12. Ballotti R
    13. Bertolotto C

    (2010) Microphthalmia-associated transcription factor controls the DNA damage response and a lineage-specific senescence program in melanomas

    Cancer Research 70:3813–3822.

    https://doi.org/10.1158/0008-5472.CAN-09-2913

    • PubMed
    • Google Scholar
24. 1. Goding CR
    2. Arnheiter H

    (2019) MITF-the first 25 years

    Genes & Development 33:983–1007.

    https://doi.org/10.1101/gad.324657.119

    • PubMed
    • Google Scholar
25. 1. Goodall J
    2. Carreira S
    3. Denat L
    4. Kobi D
    5. Davidson I
    6. Nuciforo P
    7. Sturm RA
    8. Larue L
    9. Goding CR

    (2008) Brn-2 represses microphthalmia-associated transcription factor expression and marks a distinct subpopulation of microphthalmia-associated transcription factor-negative melanoma cells

    Cancer Research 68:7788–7794.

    https://doi.org/10.1158/0008-5472.CAN-08-1053

    • PubMed
    • Google Scholar
26. 1. Harburger DS
    2. Calderwood DA

    (2009) Integrin signalling at a glance

    Journal of Cell Science 122:159–163.

    https://doi.org/10.1242/jcs.018093

    • PubMed
    • Google Scholar
27. 1. Hirata E
    2. Girotti MR
    3. Viros A
    4. Hooper S
    5. Spencer-Dene B
    6. Matsuda M
    7. Larkin J
    8. Marais R
    9. Sahai E

    (2015) Intravital imaging reveals how BRAF inhibition generates drug-tolerant microenvironments with high integrin β1/FAK signaling

    Cancer Cell 27:574–588.

    https://doi.org/10.1016/j.ccell.2015.03.008

    • PubMed
    • Google Scholar
28. 1. Hoek KS
    2. Schlegel NC
    3. Brafford P
    4. Sucker A
    5. Ugurel S
    6. Kumar R
    7. Weber BL
    8. Nathanson KL
    9. Phillips DJ
    10. Herlyn M
    11. Schadendorf D
    12. Dummer R

    (2006) Metastatic potential of melanomas defined by specific gene expression profiles with no BRAF signature

    Pigment Cell Research 19:290–302.

    https://doi.org/10.1111/j.1600-0749.2006.00322.x

    • PubMed
    • Google Scholar
29. 1. Hoek KS
    2. Eichhoff OM
    3. Schlegel NC
    4. Döbbeling U
    5. Kobert N
    6. Schaerer L
    7. Hemmi S
    8. Dummer R

    (2008) In vivo switching of human melanoma cells between proliferative and invasive states

    Cancer Research 68:650–656.

    https://doi.org/10.1158/0008-5472.CAN-07-2491

    • PubMed
    • Google Scholar
30. 1. Hoek KS
    2. Goding CR

    (2010) Cancer stem cells versus phenotype-switching in melanoma

    Pigment Cell & Melanoma Research 23:746–759.

    https://doi.org/10.1111/j.1755-148X.2010.00757.x

    • PubMed
    • Google Scholar
31. 1. Hou L
    2. Panthier JJ
    3. Arnheiter H

    (2000)

    Signaling and transcriptional regulation in the neural crest-derived melanocyte lineage: interactions between KIT and MITF

    Development 127:5379–5389.

    • PubMed
    • Google Scholar
32. 1. Hu R
    2. Sharma SM
    3. Bronisz A
    4. Srinivasan R
    5. Sankar U
    6. Ostrowski MC

    (2007) Eos, MITF, and PU.1 recruit corepressors to osteoclast-specific genes in committed myeloid progenitors

    Molecular and Cellular Biology 27:4018–4027.

    https://doi.org/10.1128/MCB.01839-06

    • PubMed
    • Google Scholar
33. 1. Humphries JD
    2. Byron A
    3. Humphries MJ

    (2006) Integrin ligands at a glance

    Journal of Cell Science 119:3901–3903.

    https://doi.org/10.1242/jcs.03098

    • PubMed
    • Google Scholar
34. 1. Ishibe S
    2. Joly D
    3. Zhu X
    4. Cantley LG

    (2003) Phosphorylation-dependent paxillin-ERK association mediates hepatocyte growth factor-stimulated epithelial morphogenesis

    Molecular Cell 12:1275–1285.

    https://doi.org/10.1016/S1097-2765(03)00406-4

    • PubMed
    • Google Scholar
35. 1. Javelaud D
    2. Alexaki VI
    3. Mauviel A

    (2008) Transforming growth factor-beta in cutaneous melanoma

    Pigment Cell & Melanoma Research 21:123–132.

    https://doi.org/10.1111/j.1755-148X.2008.00450.x

    • PubMed
    • Google Scholar
36. 1. Javelaud D
    2. Alexaki VI
    3. Pierrat MJ
    4. Hoek KS
    5. Dennler S
    6. Van Kempen L
    7. Bertolotto C
    8. Ballotti R
    9. Saule S
    10. Delmas V
    11. Mauviel A

    (2011) GLI2 and M-MITF transcription factors control exclusive gene expression programs and inversely regulate invasion in human melanoma cells

    Pigment Cell & Melanoma Research 24:932–943.

    https://doi.org/10.1111/j.1755-148X.2011.00893.x

    • PubMed
    • Google Scholar
37. 1. Jönsson G
    2. Busch C
    3. Knappskog S
    4. Geisler J
    5. Miletic H
    6. Ringnér M
    7. Lillehaug JR
    8. Borg A
    9. Lønning PE

    (2010) Gene expression profiling-based identification of molecular subtypes in stage IV melanomas with different clinical outcome

    Clinical Cancer Research 16:3356–3367.

    https://doi.org/10.1158/1078-0432.CCR-09-2509

    • PubMed
    • Google Scholar
38. 1. Jónsson H
    2. Sulem P
    3. Kehr B
    4. Kristmundsdottir S
    5. Zink F
    6. Hjartarson E
    7. Hardarson MT
    8. Hjorleifsson KE
    9. Eggertsson HP
    10. Gudjonsson SA
    11. Ward LD
    12. Arnadottir GA
    13. Helgason EA
    14. Helgason H
    15. Gylfason A
    16. Jonasdottir A
    17. Jonasdottir A
    18. Rafnar T
    19. Besenbacher S
    20. Frigge ML
    21. Stacey SN
    22. Magnusson OT
    23. Thorsteinsdottir U
    24. Masson G
    25. Kong A
    26. Halldorsson BV
    27. Helgason A
    28. Gudbjartsson DF
    29. Stefansson K

    (2017) Whole genome characterization of sequence diversity of 15,220 icelanders

    Scientific Data 4:170115.

    https://doi.org/10.1038/sdata.2017.115

    • PubMed
    • Google Scholar
39. 1. Kugel CH
    2. Aplin AE

    (2014) Adaptive resistance to RAF inhibitors in melanoma

    Pigment Cell & Melanoma Research 27:1032–1038.

    https://doi.org/10.1111/pcmr.12264

    • PubMed
    • Google Scholar
40. 1. Kulesa PM
    2. Kasemeier-Kulesa JC
    3. Teddy JM
    4. Margaryan NV
    5. Seftor EA
    6. Seftor RE
    7. Hendrix MJ

    (2006) Reprogramming metastatic melanoma cells to assume a neural crest cell-like phenotype in an embryonic microenvironment

    PNAS 103:3752–3757.

    https://doi.org/10.1073/pnas.0506977103

    • PubMed
    • Google Scholar
41. 1. Langmead B
    2. Salzberg SL

    (2012) Fast gapped-read alignment with bowtie 2

    Nature Methods 9:357–359.

    https://doi.org/10.1038/nmeth.1923

    • PubMed
    • Google Scholar
42. 1. Laurette P
    2. Strub T
    3. Koludrovic D
    4. Keime C
    5. Le Gras S
    6. Seberg H
    7. Van Otterloo E
    8. Imrichova H
    9. Siddaway R
    10. Aerts S
    11. Cornell RA
    12. Mengus G
    13. Davidson I

    (2015) Transcription factor MITF and remodeller BRG1 define chromatin organisation at regulatory elements in melanoma cells

    eLife 4:e06857.

    https://doi.org/10.7554/eLife.06857

    • Google Scholar
43. 1. Le Douarin NM
    2. Dupin E

    (2018) The "beginnings" of the neural crest

    Developmental Biology 444:S3–S13.

    https://doi.org/10.1016/j.ydbio.2018.07.019

    • PubMed
    • Google Scholar
44. 1. Le Douarin N
    2. Kalcheim C

    (1999) The migration of neural crest cells

    Developmental and Cell Biology Series 131:23–59.

    https://doi.org/10.1017/CBO9780511897948.005

    • Google Scholar
45. 1. Leroy B
    2. Girard L
    3. Hollestelle A
    4. Minna JD
    5. Gazdar AF
    6. Soussi T

    (2014) Analysis of TP53 mutation status in human Cancer cell lines: a reassessment

    Human Mutation 35:756–765.

    https://doi.org/10.1002/humu.22556

    • PubMed
    • Google Scholar
46. 1. Lindsay H
    2. Burger A
    3. Biyong B
    4. Felker A
    5. Hess C
    6. Zaugg J
    7. Chiavacci E
    8. Anders C
    9. Jinek M
    10. Mosimann C
    11. Robinson MD

    (2016) CrispRVariants charts the mutation spectrum of genome engineering experiments

    Nature Biotechnology 34:701–702.

    https://doi.org/10.1038/nbt.3628

    • PubMed
    • Google Scholar
47. 1. Lister JA
    2. Capper A
    3. Zeng Z
    4. Mathers ME
    5. Richardson J
    6. Paranthaman K
    7. Jackson IJ
    8. Elizabeth Patton E

    (2014) A conditional zebrafish MITF mutation reveals MITF levels are critical for melanoma promotion vs. regression in vivo

    Journal of Investigative Dermatology 134:133–140.

    https://doi.org/10.1038/jid.2013.293

    • PubMed
    • Google Scholar
48. 1. Livak KJ
    2. Schmittgen TD

    (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta delta C(T)) Method

    Methods 25:402–408.

    https://doi.org/10.1006/meth.2001.1262

    • PubMed
    • Google Scholar
49. 1. Louphrasitthiphol P
    2. Ledaki I
    3. Chauhan J
    4. Falletta P
    5. Siddaway R
    6. Buffa FM
    7. Mole DR
    8. Soga T
    9. Goding CR

    (2019) MITF controls the TCA cycle to modulate the melanoma hypoxia response

    Pigment Cell & Melanoma Research 32:792–808.

    https://doi.org/10.1111/pcmr.12802

    • Google Scholar
50. 1. Ma W
    2. Noble WS
    3. Bailey TL

    (2014) Motif-based analysis of large nucleotide data sets using MEME-ChIP

    Nature Protocols 9:1428–1450.

    https://doi.org/10.1038/nprot.2014.083

    • PubMed
    • Google Scholar
51. 1. Magnúsdóttir E
    2. Dietmann S
    3. Murakami K
    4. Günesdogan U
    5. Tang F
    6. Bao S
    7. Diamanti E
    8. Lao K
    9. Gottgens B
    10. Azim Surani M

    (2013) A tripartite transcription factor network regulates primordial germ cell specification in mice

    Nature Cell Biology 15:905–915.

    https://doi.org/10.1038/ncb2798

    • PubMed
    • Google Scholar
52. 1. Malcov-Brog H
    2. Alpert A
    3. Golan T
    4. Parikh S
    5. Nordlinger A
    6. Netti F
    7. Sheinboim D
    8. Dror I
    9. Thomas L
    10. Cosson C
    11. Gonen P
    12. Stanevsky Y
    13. Brenner R
    14. Perluk T
    15. Frand J
    16. Elgavish S
    17. Nevo Y
    18. Rahat D
    19. Tabach Y
    20. Khaled M
    21. Shen-Orr SS
    22. Levy C

    (2018) UV-Protection timer controls linkage between stress and pigmentation skin protection systems

    Molecular Cell 72:444–456.

    https://doi.org/10.1016/j.molcel.2018.09.022

    • PubMed
    • Google Scholar
53. 1. Mascarenhas JB
    2. Littlejohn EL
    3. Wolsky RJ
    4. Young KP
    5. Nelson M
    6. Salgia R
    7. Lang D

    (2010) PAX3 and SOX10 activate MET receptor expression in melanoma

    Pigment Cell & Melanoma Research 23:225–237.

    https://doi.org/10.1111/j.1755-148X.2010.00667.x

    • PubMed
    • Google Scholar
54. 1. Meers MP
    2. Tenenbaum D
    3. Henikoff S

    (2019) Peak calling by sparse enrichment analysis for CUT&RUN chromatin profiling

    Epigenet Chromatin 12:42.

    https://doi.org/10.1186/s13072-019-0287-4

    • Google Scholar
55. 1. Miskolczi Z
    2. Smith MP
    3. Rowling EJ
    4. Ferguson J
    5. Barriuso J
    6. Wellbrock C

    (2018) Collagen abundance controls melanoma phenotypes through lineage-specific microenvironment sensing

    Oncogene 37:3166–3182.

    https://doi.org/10.1038/s41388-018-0209-0

    • PubMed
    • Google Scholar
56. 1. Mitra SK
    2. Hanson DA
    3. Schlaepfer DD

    (2005) Focal adhesion kinase: in command and control of cell motility

    Nature Reviews Molecular Cell Biology 6:56–68.

    https://doi.org/10.1038/nrm1549

    • Google Scholar
57. 1. Moreno-Bueno G
    2. Portillo F
    3. Cano A

    (2008) Transcriptional regulation of cell polarity in EMT and Cancer

    Oncogene 27:6958–6969.

    https://doi.org/10.1038/onc.2008.346

    • PubMed
    • Google Scholar
58. 1. Nishimura EK
    2. Suzuki M
    3. Igras V
    4. Du J
    5. Lonning S
    6. Miyachi Y
    7. Roes J
    8. Beermann F
    9. Fisher DE

    (2010) Key roles for transforming growth factor beta in melanocyte stem cell maintenance

    Cell Stem Cell 6:130–140.

    https://doi.org/10.1016/j.stem.2009.12.010

    • PubMed
    • Google Scholar
59. 1. Pastushenko I
    2. Brisebarre A
    3. Sifrim A
    4. Fioramonti M
    5. Revenco T
    6. Boumahdi S
    7. Van Keymeulen A
    8. Brown D
    9. Moers V
    10. Lemaire S
    11. De Clercq S
    12. Minguijón E
    13. Balsat C
    14. Sokolow Y
    15. Dubois C
    16. De Cock F
    17. Scozzaro S
    18. Sopena F
    19. Lanas A
    20. D'Haene N
    21. Salmon I
    22. Marine JC
    23. Voet T
    24. Sotiropoulou PA
    25. Blanpain C

    (2018) Identification of the tumour transition states occurring during EMT

    Nature 556:463–468.

    https://doi.org/10.1038/s41586-018-0040-3

    • PubMed
    • Google Scholar
60. 1. Pimentel H
    2. Bray NL
    3. Puente S
    4. Melsted P
    5. Pachter L

    (2017) Differential analysis of RNA-seq incorporating quantification uncertainty

    Nature Methods 14:687–690.

    https://doi.org/10.1038/nmeth.4324

    • PubMed
    • Google Scholar
61. 1. Playford MP
    2. Schaller MD

    (2004) The interplay between src and integrins in normal and tumor biology

    Oncogene 23:7928–7946.

    https://doi.org/10.1038/sj.onc.1208080

    • PubMed
    • Google Scholar
62. 1. Rambow F
    2. Job B
    3. Petit V
    4. Gesbert F
    5. Delmas V
    6. Seberg H
    7. Meurice G
    8. Van Otterloo E
    9. Dessen P
    10. Robert C
    11. Gautheret D
    12. Cornell RA
    13. Sarasin A
    14. Larue L

    (2015) New functional signatures for understanding melanoma biology from tumor cell Lineage-Specific analysis

    Cell Reports 13:840–853.

    https://doi.org/10.1016/j.celrep.2015.09.037

    • PubMed
    • Google Scholar
63. 1. Rambow F
    2. Rogiers A
    3. Marin-Bejar O
    4. Aibar S
    5. Femel J
    6. Dewaele M
    7. Karras P
    8. Brown D
    9. Chang YH
    10. Debiec-Rychter M
    11. Adriaens C
    12. Radaelli E
    13. Wolter P
    14. Bechter O
    15. Dummer R
    16. Levesque M
    17. Piris A
    18. Frederick DT
    19. Boland G
    20. Flaherty KT
    21. van den Oord J
    22. Voet T
    23. Aerts S
    24. Lund AW
    25. Marine JC

    (2018) Toward minimal residual Disease-Directed therapy in melanoma

    Cell 174:843–855.

    https://doi.org/10.1016/j.cell.2018.06.025

    • PubMed
    • Google Scholar
64. 1. Rambow F
    2. Marine JC
    3. Goding CR

    (2019) Melanoma plasticity and phenotypic diversity: therapeutic barriers and opportunities

    Genes & Development 33:1295–1318.

    https://doi.org/10.1101/gad.329771.119

    • PubMed
    • Google Scholar
65. 1. Riesenberg S
    2. Groetchen A
    3. Siddaway R
    4. Bald T
    5. Reinhardt J
    6. Smorra D
    7. Kohlmeyer J
    8. Renn M
    9. Phung B
    10. Aymans P
    11. Schmidt T
    12. Hornung V
    13. Davidson I
    14. Goding CR
    15. Jönsson G
    16. Landsberg J
    17. Tüting T
    18. Hölzel M

    (2015) MITF and c-Jun antagonism interconnects melanoma dedifferentiation with pro-inflammatory cytokine responsiveness and myeloid cell recruitment

    Nature Communications 6:ncomms9755.

    https://doi.org/10.1038/ncomms9755

    • Google Scholar
66. 1. Santini R
    2. Pietrobono S
    3. Pandolfi S
    4. Montagnani V
    5. D'Amico M
    6. Penachioni JY
    7. Vinci MC
    8. Borgognoni L
    9. Stecca B

    (2014) SOX2 regulates self-renewal and tumorigenicity of human melanoma-initiating cells

    Oncogene 33:4697–4708.

    https://doi.org/10.1038/onc.2014.71

    • PubMed
    • Google Scholar
67. 1. Schlaepfer DD
    2. Hauck CR
    3. Sieg DJ

    (1999) Signaling through focal adhesion kinase

    Progress in Biophysics and Molecular Biology 71:435–478.

    https://doi.org/10.1016/S0079-6107(98)00052-2

    • PubMed
    • Google Scholar
68. 1. Sen A
    2. O'Malley K
    3. Wang Z
    4. Raj GV
    5. DeFranco DB
    6. Hammes SR

    (2010) Paxillin regulates androgen- and epidermal growth Factor-induced MAPK signaling and cell proliferation in prostate Cancer cells*

    Journal of Biological Chemistry 285:28787–28795.

    https://doi.org/10.1074/jbc.M110.134064

    • Google Scholar
69. 1. Sen A
    2. De Castro I
    3. DeFranco DB
    4. Deng F-M
    5. Melamed J
    6. Kapur P
    7. Raj GV
    8. Rossi R
    9. Hammes SR

    (2012) Paxillin mediates extranuclear and intranuclear signaling in prostate Cancer proliferation

    Journal of Clinical Investigation 122:2469–2481.

    https://doi.org/10.1172/JCI62044

    • Google Scholar
70. 1. Shirley SH
    2. Greene VR
    3. Duncan LM
    4. Torres Cabala CA
    5. Grimm EA
    6. Kusewitt DF

    (2012) Slug expression during melanoma progression

    The American Journal of Pathology 180:2479–2489.

    https://doi.org/10.1016/j.ajpath.2012.02.014

    • PubMed
    • Google Scholar
71. 1. Skene PJ
    2. Henikoff S

    (2017) An efficient targeted nuclease strategy for high-resolution mapping of DNA binding sites

    eLife 6:e21856.

    https://doi.org/10.7554/eLife.21856

    • PubMed
    • Google Scholar
72. 1. Stylianou N
    2. Lehman ML
    3. Wang C
    4. Fard AT
    5. Rockstroh A
    6. Fazli L
    7. Jovanovic L
    8. Ward M
    9. Sadowski MC
    10. Kashyap AS
    11. Buttyan R
    12. Gleave ME
    13. Westbrook TF
    14. Williams ED
    15. Gunter JH
    16. Nelson CC
    17. Hollier BG

    (2019) Correction: a molecular portrait of epithelial-mesenchymal plasticity in prostate Cancer associated with clinical outcome

    Oncogene 38:2436.

    https://doi.org/10.1038/s41388-018-0587-3

    • PubMed
    • Google Scholar
73. 1. Subramanian A
    2. Tamayo P
    3. Mootha VK
    4. Mukherjee S
    5. Ebert BL
    6. Gillette MA
    7. Paulovich A
    8. Pomeroy SL
    9. Golub TR
    10. Lander ES
    11. Mesirov JP

    (2005) Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles

    PNAS 102:15545–15550.

    https://doi.org/10.1073/pnas.0506580102

    • PubMed
    • Google Scholar
74. 1. Taranova OV
    2. Magness ST
    3. Fagan BM
    4. Wu Y
    5. Surzenko N
    6. Hutton SR
    7. Pevny LH

    (2006) SOX2 is a dose-dependent regulator of retinal neural progenitor competence

    Genes & Development 20:1187–1202.

    https://doi.org/10.1101/gad.1407906

    • PubMed
    • Google Scholar
75. 1. Thompson EW
    2. Nagaraj SH

    (2018) Transition states that allow Cancer to spread

    Nature 556:442–444.

    https://doi.org/10.1038/d41586-018-04403-x

    • PubMed
    • Google Scholar
76. 1. Tirosh I
    2. Izar B
    3. Prakadan SM
    4. Wadsworth MH
    5. Treacy D
    6. Trombetta JJ
    7. Rotem A
    8. Rodman C
    9. Lian C
    10. Murphy G
    11. Fallahi-Sichani M
    12. Dutton-Regester K
    13. Lin JR
    14. Cohen O
    15. Shah P
    16. Lu D
    17. Genshaft AS
    18. Hughes TK
    19. Ziegler CG
    20. Kazer SW
    21. Gaillard A
    22. Kolb KE
    23. Villani AC
    24. Johannessen CM
    25. Andreev AY
    26. Van Allen EM
    27. Bertagnolli M
    28. Sorger PK
    29. Sullivan RJ
    30. Flaherty KT
    31. Frederick DT
    32. Jané-Valbuena J
    33. Yoon CH
    34. Rozenblatt-Rosen O
    35. Shalek AK
    36. Regev A
    37. Garraway LA

    (2016) Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq

    Science 352:189–196.

    https://doi.org/10.1126/science.aad0501

    • PubMed
    • Google Scholar
77. 1. Travnickova J
    2. Wojciechowska S
    3. Khamseh A
    4. Gautier P
    5. Brown DV
    6. Lefevre T
    7. Brombin A
    8. Ewing A
    9. Capper A
    10. Spitzer M
    11. Dilshat R
    12. Semple CA
    13. Mathers ME
    14. Lister JA
    15. Steingrimsson E
    16. Voet T
    17. Ponting CP
    18. Patton EE

    (2019) Zebrafish MITF-Low melanoma subtype models reveal transcriptional subclusters and MITF-Independent residual disease

    Cancer Research 79:5769–5784.

    https://doi.org/10.1158/0008-5472.CAN-19-0037

    • PubMed
    • Google Scholar
78. 1. Tsoi J
    2. Robert L
    3. Paraiso K
    4. Galvan C
    5. Sheu KM
    6. Lay J
    7. Wong DJL
    8. Atefi M
    9. Shirazi R
    10. Wang X
    11. Braas D
    12. Grasso CS
    13. Palaskas N
    14. Ribas A
    15. Graeber TG

    (2018) Multi-stage differentiation defines melanoma subtypes with differential vulnerability to Drug-Induced Iron-Dependent oxidative stress

    Cancer Cell 33:890–904.

    https://doi.org/10.1016/j.ccell.2018.03.017

    • PubMed
    • Google Scholar
79. 1. Tuncer E
    2. Calçada RR
    3. Zingg D
    4. Varum S
    5. Cheng P
    6. Freiberger SN
    7. Deng C-X
    8. Kleiter I
    9. Levesque MP
    10. Dummer R
    11. Sommer L

    (2019) SMAD signaling promotes melanoma metastasis independently of phenotype switching

    Journal of Clinical Investigation 129:2702–2716.

    https://doi.org/10.1172/JCI94295

    • Google Scholar
80. 1. Verfaillie A
    2. Imrichova H
    3. Atak ZK
    4. Dewaele M
    5. Rambow F
    6. Hulselmans G
    7. Christiaens V
    8. Svetlichnyy D
    9. Luciani F
    10. Van den Mooter L
    11. Claerhout S
    12. Fiers M
    13. Journe F
    14. Ghanem G-E
    15. Herrmann C
    16. Halder G
    17. Marine J-C
    18. Aerts S

    (2015) Decoding the regulatory landscape of melanoma reveals TEADS as regulators of the invasive cell state

    Nature Communications 6:ncomms7683.

    https://doi.org/10.1038/ncomms7683

    • Google Scholar
81. 1. Vlčková K
    2. Vachtenheim J
    3. Réda J
    4. Horák P
    5. Ondrušová L

    (2018) Inducibly decreased MITF levels do not affect proliferation and phenotype switching but reduce differentiation of melanoma cells

    Journal of Cellular and Molecular Medicine 22:2240–2251.

    https://doi.org/10.1111/jcmm.13506

    • PubMed
    • Google Scholar
82. 1. Webster DE
    2. Barajas B
    3. Bussat RT
    4. Yan KJ
    5. Neela PH
    6. Flockhart RJ
    7. Kovalski J
    8. Zehnder A
    9. Khavari PA

    (2014) Enhancer-targeted genome editing selectively blocks innate resistance to oncokinase inhibition

    Genome Research 24:751–760.

    https://doi.org/10.1101/gr.166231.113

    • PubMed
    • Google Scholar
83. 1. Widmer DS
    2. Hoek KS
    3. Cheng PF
    4. Eichhoff OM
    5. Biedermann T
    6. Raaijmakers MIG
    7. Hemmi S
    8. Dummer R
    9. Levesque MP

    (2013) Hypoxia contributes to melanoma heterogeneity by triggering HIF1α-Dependent phenotype switching

    Journal of Investigative Dermatology 133:2436–2443.

    https://doi.org/10.1038/jid.2013.115

    • Google Scholar
84. 1. Wu DW
    2. Chen CY
    3. Chu CL
    4. Lee H

    (2016) Paxillin confers resistance to tyrosine kinase inhibitors in EGFR-mutant lung cancers via modulating BIM and Mcl-1 protein stability

    Oncogene 35:621–630.

    https://doi.org/10.1038/onc.2015.120

    • PubMed
    • Google Scholar
85. 1. Yokoyama S
    2. Woods SL
    3. Boyle GM
    4. Aoude LG
    5. MacGregor S
    6. Zismann V
    7. Gartside M
    8. Cust AE
    9. Haq R
    10. Harland M
    11. Taylor JC
    12. Duffy DL
    13. Holohan K
    14. Dutton-Regester K
    15. Palmer JM
    16. Bonazzi V
    17. Stark MS
    18. Symmons J
    19. Law MH
    20. Schmidt C
    21. Lanagan C
    22. O'Connor L
    23. Holland EA
    24. Schmid H
    25. Maskiell JA
    26. Jetann J
    27. Ferguson M
    28. Jenkins MA
    29. Kefford RF
    30. Giles GG
    31. Armstrong BK
    32. Aitken JF
    33. Hopper JL
    34. Whiteman DC
    35. Pharoah PD
    36. Easton DF
    37. Dunning AM
    38. Newton-Bishop JA
    39. Montgomery GW
    40. Martin NG
    41. Mann GJ
    42. Bishop DT
    43. Tsao H
    44. Trent JM
    45. Fisher DE
    46. Hayward NK
    47. Brown KM

    (2011) A novel recurrent mutation in MITF predisposes to familial and sporadic melanoma

    Nature 480:99–103.

    https://doi.org/10.1038/nature10630

    • PubMed
    • Google Scholar
86. 1. Yoshida H
    2. Kunisada T
    3. Kusakabe M
    4. Nishikawa S
    5. Nishikawa SI

    (1996)

    Distinct stages of melanocyte differentiation revealed by analysis of nonuniform pigmentation patterns

    Development 122:1207–1214.

    • PubMed
    • Google Scholar
87. 1. Yu G
    2. Wang LG
    3. Han Y
    4. He QY

    (2012) clusterProfiler: an R package for comparing biological themes among gene clusters

    OMICS: A Journal of Integrative Biology 16:284–287.

    https://doi.org/10.1089/omi.2011.0118

    • PubMed
    • Google Scholar
88. 1. Yu G
    2. Wang LG
    3. He QY

    (2015) ChIPseeker: an R/Bioconductor package for ChIP peak annotation, comparison and visualization

    Bioinformatics 31:2382–2383.

    https://doi.org/10.1093/bioinformatics/btv145

    • PubMed
    • Google Scholar

Acceptance summary:

In this elegant report, Dilshat and collaborators perform a number of carefully designed experiments in cell lines that show that the transcription factor MITF directly represses the expression of genes associated with the extracellular matrix, focal adhesion and epithelial-mesenchymal transition pathways in human melanoma cells. They then confirm their observations in human tumours from The Cancer Genome Atlas and zebrafish melanomas. These results illustrate the complexity of the MITF regulatory network and are of major importance for understanding resistance of melanoma cells to targeted and immune therapy.

Decision letter after peer review:

Thank you for submitting your article "MITF reprograms the extracellular matrix and focal adhesion in melanoma" for consideration by eLife. Your article has been reviewed by three peer reviewers, including C Daniela Robles-Espinoza as the Reviewing Editor and Reviewer #1, and the evaluation has been overseen by Richard White as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Heinz Arnheiter (Reviewer #2); Stacie Loftus (Reviewer #3).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

We would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). Specifically, we are asking editors to accept without delay manuscripts, like yours, that they judge can stand as eLife papers without additional data, even if they feel that they would make the manuscript stronger. Thus the revisions requested below only address clarity and presentation.

In this manuscript, Dilshat and collaborators perform a number of carefully designed experiments in cell lines that show that the transcription factor MITF directly represses the expression of genes associated with the extracellular matrix (ECM), focal adhesion and EMT pathways in human melanoma cells, and confirm their observations in human tumours from TCGA and zebrafish melanomas. The reviewers all agree that the experiments are elegant and the results are sound, and raise only a number of relatively minor issues that would need to be addressed before this manuscript can be published.

Essential revisions:

1) The proliferative/non-invasive and non-proliferative/invasive states of melanoma cells are based on previous gene expression profiling and in vitro experiments. Perhaps due to different experimental conditions, it has also been shown, however, that downregulation of MITF may not lead to large alterations in proliferation and invasion, that nutrient starvation reduces MITF without changing invasive properties, or that normal levels of MITF are compatible with a dual proliferative/invasive phenotype. While it is already difficult to find a common thread to explain these discrepant findings, Dilshat et al. now report that in the absence of MITF or after its downregulation, both proliferation AND invasion are reduced, which is compatible with their finding that a reduction of MITF leads to an increase in focal adhesions. The Discussion would be greatly enhanced if it could be amended to include further insights into why their results are different from those previously published. Can the authors specify if the state they describe represents an additional one to the two mentioned at the beginning of this paragraph? Also, an attempt to reconcile the various findings on MITF high and low states would be welcome in the Discussion section.

2) Regarding the MITF-Χ2 cells, the reviewers have pointed out that additional information regarding the WGS and transcriptomic profile of the CRISPR clones may be necessary to aid the interpretation of experiments involving this clone. One reviewer wrote, "Although the molecular nature of the Χ2 protein would suggest that it is capable of dimerization and DNA binding, Χ2 cells show virtually the same cellular phenotype as MITF-negative X6 cells or occasionally a somewhat milder phenotype than the X6 cells [such as in effects on the expression of certain genes (Figures 3, 4 and 5)]. However, a transcriptomic profile of the Χ2 cells, or a direct analysis of the transcriptional performance of the truncated Χ2 protein, is missing." Can a description of the additional mutations carried by this clone, if any, be included, along with a description of its transcriptional profile?

3) The analysis of MITF expression profiles within the 470 TCGA cutaneous melanoma tumors dataset as described in Figure 3I suggests that analysis was performed on both primary and metastatic tumors. Reviewers suggest that the authors should display or indicate which type of tumors are being analyzed in this analysis. While authors have removed the 130 tumors with "fibroblast marker" expression to assess the MITF high and MITF low tumor profiles, the work would benefit from authors clearly indicating whether the high vs. low MITF profiles are solely reflective of metastatic tumor data, primary tumor data or if this profile holds true regardless of tumor's location.

Essential revisions:

1) The proliferative/non-invasive and non-proliferative/invasive states of melanoma cells are based on previous gene expression profiling and in vitro experiments. Perhaps due to different experimental conditions, it has also been shown, however, that downregulation of MITF may not lead to large alterations in proliferation and invasion, that nutrient starvation reduces MITF without changing invasive properties, or that normal levels of MITF are compatible with a dual proliferative/invasive phenotype. While it is already difficult to find a common thread to explain these discrepant findings, Dilshat et al. now report that in the absence of MITF or after its downregulation, both proliferation AND invasion are reduced, which is compatible with their finding that a reduction of MITF leads to an increase in focal adhesions. The Discussion would be greatly enhanced if it could be amended to include further insights into why their results are different from those previously published. Can the authors specify if the state they describe represents an additional one to the two mentioned at the beginning of this paragraph? Also, an attempt to reconcile the various findings on MITF high and low states would be welcome in the Discussion section.

The rheostat model for the role of MITF in melanoma predicts that high MITF activity promotes proliferation whereas low activity promotes invasion. The reviewers correctly point out that our MITF knockout and knockdown cells do not behave as expected by the model with respect to migratory properties although they do with respect to proliferation. As explained in the Discussion of our manuscript, previous work used siRNA/shRNA to achieve partial and transient depletion of MITF resulting in increased migratory properties of melanoma cells upon MITF reduction (Carreira et al., 2006; Cheli et al., 2012; Javelaud et al., 2011; Bianchi-Smiraglia et al., 2016). Most of the studies used the 501mel and Skmel28 cell lines. Two different pathways involved in migration were shown to be regulated by MITF; DIAPH1, a gene implicated in actin polymerization and the guanosine monophosphate reductase (GMPR) gene encoding an enzyme involved in regulating intracellular GTP levels (Carreira et al., 2006; Bianchi-Smiraglia et al., 2017). We did not observe changes in DIAPH1 or GMPR expression in our cells or in our analysis of melanoma tumors of TCGA. Importantly, not all studies have resulted in effects on migration, sometimes using the same cells and methods as were used in the previous studies (Falletta et al., 2017 and Vlckova et al., 2018). And interestingly, a recent study showed that reducing MITF alone was not sufficient to mediate invasion; the translation factor eIF2α was needed along with the nutrient sensor ATF4 to mediate invasion of melanoma cells (Falleta et al., 2017). We did not observe changes in eIF2α transcript expression in our cells but did not assess the activity of eIF2α by determining phosphorylation status. Clearly, the experimental system used and the conditions they are performed in play a role in mediating the invasive phenotype.

In our ∆MITF-X6 cell model we have a complete loss of functional MITF. The resulting transcriptional signature was similar to that of the “Rambow neural crest like” cells as determined by single cell RNA-sequencing of melanoma tumors treated with BRAFi (Rambow et al., 2018). These cells are believed to represent minimal residual disease. This gene signature was different from the “invasive gene signature” that the author observed in a different set of melanoma cells (Rambow et al., 2018). Furthermore, comparing our gene signature with that derived from 86 melanoma cells (Hoek et al., 2006), showed that the ∆MITF-X6 cells displayed a more prominent loss of proliferative genes than a gain of invasive genes. Similarly, Rambow et al., 2015, showed that melanoma cell lines (G1, T1, 501Mel, MNT-1, SKMel3) which express high levels of MITF displayed a high proliferation potential but less migratory, invasive and subcutaneous tumour growth capabilities than cell lines with low MITF expression (WM1366, WM793, WM852, LU1205, A375M). The expression of TRIM63 and CAPN3 was over-represented in the cell lines with high MITF expression and knocking down MITF led to a reduction of their expression. Interestingly, knocking down either TRIM63 or CAPN3 enhanced the invasive potential of 501Mel cells. In the ∆MITF-X6 cells we observed reduction only of TRIM63 expression. Thus, the transcriptome of our ∆MITF-X6 cells does not fully recapitulate the gene signature of invasive cells. To emphasize, we did not observe an increased migratory phenotype either using siRNA/miRNA in melanoma cells so this feature is not restricted to our knockout lines.

We could speculate that the migratory phenotype is a transient event and that in order to achieve migration the cells need to be tested at a specific time interval where MITF activity is gradually decreasing. Carmit Levy‘s group has recently shown MITF oscillations upon UV exposure that synchronizes stress response and pigmentation (Malcov-Brog et al., 2018). It is possible that such a mechanism exists, e.g. during melanoblast development where oscillating MITF-expression might ensure that proliferation and migration can both take place. Previous work has suggested oscillations in the dependence of melanoblasts on the receptor tyrosine kinase KIT ( Yoshida et al., 1996; Hou et al., 2000). It is plausible that activation of invasion genes requires discontinuous presence of MITF along with other co-factors at specific time windows. According to the phenotype switching model of melanoma the reversible switch between the proliferative and invasive states is needed for melanoma progression (Arozarena et al., 2019). In our case, the complete loss of MITF might trap the cells in a state where the MITF switch is dead and migration therefore not possible. This would then explain why we do not see the invasive gene signature and no changes in DIAPH1, p27 or GMPR expression. However, this would not explain why we do not observe this either using knockdown strategies.

In order to improve the Discussion with respect to this issue we edited the text considerably: The most significant change is the following text:

“Thus, we might speculate that the migratory phenotype is a transient event and that in order to achieve migration the cells need to be tested at a narrow time interval where MITF activity is decreasing. Carmit Levy‘s group has recently shown MITF oscillations upon UV exposure in order to synchronize stress response and pigmentation (Malcov-Brog et al., 2018). [...] Thus, the transcriptome of our ∆MITF-X6 cells does not fully recapitulate the gene signature of invasive cells.”

2) Regarding the MITF-Χ2 cells, the reviewers have pointed out that additional information regarding the WGS and transcriptomic profile of the CRISPR clones may be necessary to aid the interpretation of experiments involving this clone. One reviewer wrote, "Although the molecular nature of the Χ2 protein would suggest that it is capable of dimerization and DNA binding, Χ2 cells show virtually the same cellular phenotype as MITF-negative X6 cells or occasionally a somewhat milder phenotype than the X6 cells [such as in effects on the expression of certain genes (Figures 3, 4 and 5)]. However, a transcriptomic profile of the Χ2 cells, or a direct analysis of the transcriptional performance of the truncated Χ2 protein, is missing." Can a description of the additional mutations carried by this clone, if any, be included, along with a description of its transcriptional profile?

The mutation induced in the ∆MITF-Χ2 line was characterized by both WGS and by PCR-amplification of both genomic DNA and cDNA and subsequent cloning of the resulting fragments into the puc19 vector for Sanger sequencing. The WGS sequencing of the ∆MITF-Χ2 line was included in the original manuscript and we have now added Figure 1—figure supplement 1C showing the Sanger sequencing-results of PCR amplified cDNA fragments. The PCR-seq results are consistent with the WGS results in that we observed an early stop codon at position Y22 (M-MITF) in 5 of 9 clones and a stop codon at position K43 in 4 of 9 clones (Figure 1—figure supplement 1C). We did not observe additional mutations in MITF neither in our WGS sequencing nor in PCR amplified cDNA fragments. The sequence analysis did not reveal any alternative splice products in our mutant clones and searching for them using RT-PCR failed to unravel any such products. Western blotting reveals only the shorter MITF variants in the ∆MITF-Χ2 line that are also observed in wild type but now at a much higher concentration. The size of the smaller MITF variants is estimated at approximately 45 kD (top-most band on Figure 1D) and 40 kD (strongest band on Figure 1D). The difference from the size of the full-length MITF 55 kD protein is 10 and 15 kD, respectively, which corresponds to about 90 and 135 amino acids (excluding possible post-translational modifications that might alter mobility). The fact that these products are detectable using the C5 antibody suggests that they contain the epitope which is located between amino acids 120 and 170 of M-MITF encoded by exons 4 and 5 (Fock et al., 2019). There are several ATGs in the sequence between the initiator MET and the region encoding exon 4, including Met62, Met75, Met77, Met105, Met114, MET116, MET138 and MET144. Some of these have elements of a Kozak-sequence and might represent the alternative products. Mass spec or protein sequencing would be the best way to determine what these products are and at present we do not have the capacity to perform such analysis.

As we were not able to characterize the precise nature of the shorter protein isoforms of MITF that predominate in the Χ2 cells, we hesitated to perform RNA-seq on these cells. However, we performed RT-PCR on RNA isolated from these cells and always observed an intermediate effect on gene expression to that observed in the X6 line, suggesting that this is indeed a hypomorphic mutation.

To provide further detail on the nature of the Χ2 mutation, we have added the following text to the Results section and hope this adds clarity:

“Sanger sequencing of DNA clones containing PCR amplified cDNA fragments from the MITF-gene of ∆MITF-Χ2 cells verified that the ∆MITF-Χ2 cells have the same mutations at similar frequency as were observed in the WGS data (Figure1—figure supplement 1C).” And later in the same paragraph:

“Similarly, neither the WGS nor RT-PCR studies showed evidence for transcripts lacking exon2 from the ∆MITF-Χ2 cDNA indicating that the truncated MITF proteins are most likely products of alternative translation start sites.”

3) The analysis of MITF expression profiles within the 470 TCGA cutaneous melanoma tumors dataset as described in Figure 3I suggests that analysis was performed on both primary and metastatic tumors. Reviewers suggest that the authors should display or indicate which type of tumors are being analyzed in this analysis. While authors have removed the 130 tumors with "fibroblast marker" expression to assess the MITF high and MITF low tumor profiles, the work would benefit from authors clearly indicating whether the high vs. low MITF profiles are solely reflective of metastatic tumor data, primary tumor data or if this profile holds true regardless of tumor's location.

The reviewers are correct that we analysed gene expression in both metastatic and primary tumors before and after withdrawing the fibroblast markers. The melanomas in TCGA are mostly metastatic tumors (371) and primary tumors are much fewer (109). We have now separated the primary tumors from the metastatic ones and then divided them into MITF-high (highest 10%) and low (lowest 10%) tumors. We then performed gene set enrichment analysis and showed that the MITF low tumor samples from both primary and metastatic melanomas showed enrichment for the ECM genes that are affected by MITF loss.

We have edited the paragraph to make these results more clear:

“Next, we analysed whether the collagens that were differentially expressed in the MITF-KD or KO melanoma cell lines were also affected by MITF in melanoma tumors in TCGA. […] The enrichment for ECM genes was observed in both primary and metastatic tumors (Figure 3I).”

Author details

1. Ramile Dilshat

   Department of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

   Contribution

   Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - original draft, Project administration, Writing - review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2126-2902

2. Valerie Fock

   Department of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

   Contribution

   Conceptualization, Resources, Investigation, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

3. Colin Kenny

   Department of Anatomy and Cell biology, Carver College of Medicine, University of Iowa, Iowa City, United States

   Contribution

   Data curation, Software, Formal analysis, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

4. Ilse Gerritsen

   Department of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

   Contribution

   Data curation, Software, Formal analysis, Methodology

   Competing interests

   No competing interests declared

5. Romain Maurice Jacques Lasseur

   Department of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

   Contribution

   Methodology

   Competing interests

   No competing interests declared

6. Jana Travnickova

   MRC Institute of Genetics and Molecular Medicine, MRC Human Genetics Unit, University of Edinburgh, Edinburgh, United Kingdom

   Contribution

   Data curation, Methodology

   Competing interests

   No competing interests declared

7. Ossia M Eichhoff

   Department of Dermatology, University Hospital Zurich, Zurich, Switzerland

   Contribution

   Data curation, Formal analysis, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-3319-1312

8. Philipp Cerny

   Department of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

   Contribution

   Data curation, Formal analysis, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

9. Katrin Möller

   Department of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

   Present address

   Institute of Molecular Life Sciences, University of Zurich, Zurich, Switzerland

   Contribution

   Methodology

   Competing interests

   No competing interests declared

10. Sara Sigurbjörnsdóttir

    Department of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

    Contribution

    Methodology

    Competing interests

    No competing interests declared

11. Kritika Kirty

    Department of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

    Contribution

    Data curation, Methodology

    Competing interests

    No competing interests declared

12. Berglind Ósk Einarsdottir

    Department of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

    Contribution

    Data curation

    Competing interests

    No competing interests declared

13. Phil F Cheng

    Department of Dermatology, University Hospital Zurich, Zurich, Switzerland

    Contribution

    Data curation

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0003-2940-006X

14. Mitchell Levesque

    Department of Dermatology, University Hospital Zurich, Zurich, Switzerland

    Contribution

    Resources, Data curation

    Competing interests

    No competing interests declared

15. Robert A Cornell

    Department of Anatomy and Cell biology, Carver College of Medicine, University of Iowa, Iowa City, United States

    Contribution

    Resources, Writing - review and editing

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0003-4207-9100

16. E Elizabeth Patton

    MRC Institute of Genetics and Molecular Medicine, MRC Human Genetics Unit, University of Edinburgh, Edinburgh, United Kingdom

    Contribution

    Resources, Writing - review and editing

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0002-2570-0834

17. Lionel Larue

    Institut Curie, CNRS UMR3347, INSERM U1021, Centre Universitaire, Orsay, France

    Contribution

    Resources, Writing - review and editing

    Competing interests

    No competing interests declared

18. Marie de Tayrac

    1. Service de Génétique Moléculaire et Génomique, CHU, Rennes, France
    2. Univ Rennes1, CNRS, IGDR (Institut de Génétique et Développement de Rennes), Rennes, France

    Contribution

    Resources, Data curation, Formal analysis, Methodology, Writing - review and editing

    Competing interests

    No competing interests declared

19. Erna Magnúsdóttir

    Department of Anatomy, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

    Contribution

    Resources, Data curation, Formal analysis, Methodology, Writing - review and editing

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0002-3369-4390

20. Margrét Helga Ögmundsdóttir

    Department of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

    Contribution

    Resources, Methodology, Writing - review and editing

    Competing interests

    No competing interests declared

21. Eirikur Steingrimsson

    Department of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Reykjavik, Iceland

    Contribution

    Conceptualization, Resources, Supervision, Investigation, Methodology, Writing - original draft, Project administration, Writing - review and editing

    For correspondence

    eirikurs@hi.is

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0001-5826-7486

• 2,825

  Page views

• 385

  Downloads

• 26

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, Scopus, PubMed Central.