Inverse Expression States of the BRN2 and MITF Transcription Factors in Melanoma Spheres and Tumour Xenografts Regulate the NOTCH Pathway
Total Page：16
File Type：pdf, Size：1020Kb
Oncogene (2011) 30, 3036–3048 & 2011 Macmillan Publishers Limited All rights reserved 0950-9232/11 www.nature.com/onc ORIGINAL ARTICLE Inverse expression states of the BRN2 and MITF transcription factors in melanoma spheres and tumour xenografts regulate the NOTCH pathway AE Thurber1, G Douglas1, EC Sturm1, SE Zabierowski2, DJ Smit1, SN Ramakrishnan1, E Hacker3, JH Leonard3, M Herlyn2 and RA Sturm1,2 1Institute for Molecular Bioscience, Melanogenix Group, The University of Queensland, Brisbane, Queensland, Australia; 2The Wistar Institute, Philadelphia, PA, USA and 3Queensland Institute of Medical Research, Brisbane, Queensland, Australia The use of adherent monolayer cultures have produced Introduction many insights into melanoma cell growth and differentia- tion, but often novel therapeutics demonstrated to act on Despite several decades of research on the causes and these cells are not active in vivo. It is imperative that new potential treatments for melanoma, little improvement methods of growing melanoma cells that reflect growth has been made in the prognosis of this cancer, which in vivo are investigated. To this end, a range of human remains at less than 15% survival after 5 yrs for patients melanoma cell lines passaged as adherent cultures or diagnosed with metastatic disease (Miller and Mihm, induced to form melanoma spheres (melanospheres) in 2006). One issue slowing progress is the large disparity stem cell media have been studied to compare cellular that often exists between experimental results and characteristics and protein expression. Melanoma spheres clinical outcomes. In order to screen new therapeutic and tumours grown from cell lines as mouse xenografts drugs more quickly and cost effectively, new culture had increased heterogeneity when compared with adherent models representative of the clinical setting are needed. cells and 3D-spheroids in agar (aggregates). Furthermore, This has lead to a gain in popularity of the three- cells within the melanoma spheres and mouse xenografts dimensional (3D) culture systems as an alternative to each displayed a high level of reciprocal BRN2 or MITF adherent monolayer culture. Fang et al. (2005) demon- expression, which matched more closely the pattern seen strated that growth of melanoma cell lines in stem cell in human melanoma tumours in situ, rather than the media to form non-adherent spheres increased their propensity for co-expression of these important melano- tumorigenic grafting ability, making this an interesting cytic transcription factors seen in adherent cells and 3D- new model to study. Growth of melanoma spheres was spheroids. Notably, when the levels of the BRN2 and initially developed because it was thought that the stem MITF proteins were each independently repressed using cell conditions would enrich an otherwise small ‘cancer siRNA treatment of adherent melanoma cells, members of stem cell’ population (Schatton et al., 2008). However, the NOTCH pathway responded by decreasing or recent evidence has dispelled the notion of only a increasing expression, respectively. This links BRN2 as selected subpopulation of cancer cells having the ability an activator, and conversely, MITF as a repressor of the to form a melanoma tumour (Quintana et al., 2008), NOTCH pathway in melanoma cells. Loss of the BRN2- making it likely that other unknown factors are MITF axis in antisense-ablated cell lines decreased the responsible for the increased tumorigenesis seen with melanoma sphere-forming capability, cell adhesion during melanoma sphere formation and growth. As yet, the 3D-spheroid formation and invasion through a collagen relationship between the sphere growth characteristics, matrix. Combined, this evidence suggests that the expression profiles and behaviour in relation to mela- melanoma sphere-culture system induces subpopulations noma in situ has not been fully investigated. of cells that may more accurately portray the in vivo An important consideration in melanoma treatment is disease, than the growth as adherent melanoma cells. the high degree of heterogeneity found in tumours Oncogene (2011) 30, 3036–3048; doi:10.1038/onc.2011.33; (Hoek et al., 2008), as clearly illustrated by the inverse published online 28 February 2011 expression pattern of two transcription factors, BRN2 and MITF, important in proliferation and metastasis Keywords: cell adhesion; invasion; melanoma; (Carreira et al., 2006; Goodall et al., 2008). A metastasis; NOTCH; transcription factor phenotype-switching model in melanoma progression based on tumour microenvironmental cues has recently been highlighted (Hoek and Goding, 2010). These two proteins have been implicated in the regulation of Correspondence: Dr RA Sturm, Institute for Molecular Bioscience, downstream pathways linked to invasive abilities, as Melanogenix Group, The University of Queensland, 306 Carmody has the NOTCH pathway (Pinnix and Herlyn, 2007), Road, Brisbane, Qld. 4072, Australia. E-mail: [email protected] but as yet a relationship between these pathways has not Received 21 August 2010; revised and accepted 12 January 2011; been described. In human melanoma tumour samples, published online 28 February 2011 BRN2 and MITF mark two distinct subpopulations BRN2-MITF regulate NOTCH in melanoma AE Thurber et al 3037 predicted to have different proliferative and invasive This method utilized stem cell growth conditions with abilities, whereas in adherent cultures, they are seen to mouse embryo fibroblast-conditioned media (MEF) to be co-expressed. Therefore, the pattern of these two promote the growth of non-adherent spheres. In addition proteins may provide a measure of tumour-like hetero- to standard melanoma (MEL) and stem cell media geneity, seen using in vitro culture systems and have (human embryonic stem cell–knockout serum replace- been investigated here using adherent and non-attached ment, hESCM–KSR), we also assayed for sphere growth melanoma-cell growth states. in stem cell media supplemented with 10% fetal bovine serum (FBS; hESCM–FBS) instead of 20% KSR. Cells were allowed to grow undisturbed in each media for 14 days or until melanoma spheres formed. Results Melanoblasts were unable to sustain growth in any of these media. Of the melanoma cell lines tested, only 16% Media conditions required for melanoma sphere formation had the capacity for detached cell growth using MEL A range of diverse melanoma cell lines including those media with only a minor increase to 19%, seen using of known BRAF, NRAS, PTEN gene mutation status hESCM–FBS; however, the majority of cell lines, 87% (Pavey et al., 2004; Johansson et al., 2007), several with (27 out of 31), were capable of forming non-adherent b-catenin mutations, non-pigmented and pigmented melanoma spheres when cultured in hESCM–KSR. strains, four ocular melanomas and a cell line showing Morphological differences between cell lines were spontaneous sphere growth were available for this observed during initial sphere formation. Some lines study. In all, 31 parental melanoma cell lines (Supple- produced small spheres within the first few days of mentary Table 1) and several primary melanoblast culture, whereas others grew large sheets of cells strains (Cook et al., 2003) were investigated for their detaching after 14 days of culture. Two examples ability to form spheres in three different types of media, showing some of the variation in sphere formation are using the method first developed by Fang et al. (2005). illustrated using the pigmented A11 (Figure 1a) and the Figure 1 Melanoma sphere induction in human melanoma cell lines. (a, b) A11 and D10 cell lines were plated as adherent monolayers and grown to confluence in T75 flasks before changing to the media indicated above the panels as MEL, hESCM–FBS or hESCM– KSR. Digital pictures were taken of cells under bright-field microscopy on the days as indicated to the right of the panels (D4, D6, D9, D13). Floating-cell aggregates were harvested from the flasks at D14 or earlier and passaged as melanoma spheres, as described. Scale bars in black represent 200 m.(c) Loss of pigmentation in A11 melanoma cells was apparent upon changing from adherent to detached growth states. The colour of the cell pellets from adherent cells (passage P0) and those from serially-passaged melanoma spheres in hESCM–KSR medium (passages P1, P2, P3) demonstrates the cells become depigmented. Oncogene BRN2-MITF regulate NOTCH in melanoma AE Thurber et al 3038 spontaneous sphere forming D10 (Figure 1b) cell lines. Assaying of RNA transcript levels in the adherent and Melanoma spheres were harvested and passaged by melanoma spheres was performed using quantitative dissociation into single cell suspensions and replated. RT–PCR as shown by the representative data presented After passaging, the melanoma spheres tended to be for the MM96L cell line (Figure 2b). This revealed smaller and to have a more regular globular shape. In significant reductions in mRNA levels for BRN2, agreement with previous data (Fang et al., 2005), some MITF, NOTCH1 and DLL1 in line with the reduction cell lines lost pigmentation during the first few passages in their protein levels. There was inconsistency between in hESCM–KSR media. The A11 cells were initially the protein and mRNA responses seen for JAG2, with a darkly pigmented when cultured as an adherent mono- decrease in protein but not transcript in MM96L layer in MEL media and became depigmented after spheres. Notably spheroid growth did not alter the three serial passages as detached cells in hESCM–KSR transcript level of the ABCB5 gene, which had media (Figure 1c). Depigmented sphere cells did not previously been described as a cancer-initiating cell regain pigmentation after culturing as an adherent marker (Schatton et al., 2008). monolayer for 14 days in MEL media. In contrast, this Given the cell heterogeneity found in melanoma, we change in pigmentation pattern was not seen in A11 were interested to examine the protein expression and melanoma spheres derived using hESCM–FBS media. localization at a cellular level of several of these proteins Because of the greater success of sphere formation in using immunofluorescence, as well as measuring total hESCM–KSR media, only melanoma spheres generated protein levels.