We present here a cell culture method for inducing mesenchymal-epithelial transitions (MET) in sarcoma cells based on combined ectopic expression of microRNA-200 family members and grainyhead-like 2 (GRHL2). This method is suitable for better understanding the biological impact of phenotypic plasticity on cancer aggressiveness and treatments.
Phenotypic plasticity refers to a phenomenon in which cells transiently gain traits of another lineage. During carcinoma progression, phenotypic plasticity drives invasion, dissemination and metastasis. Indeed, while most of the studies of phenotypic plasticity have been in the context of epithelial-derived carcinomas, it turns out sarcomas, which are mesenchymal in origin, also exhibit phenotypic plasticity, with a subset of sarcomas undergoing a phenomenon that resembles a mesenchymal-epithelial transition (MET). Here, we developed a method comprising the miR-200 family and grainyhead-like 2 (GRHL2) to mimic this MET-like phenomenon observed in sarcoma patient samples.We sequentially express GRHL2 and the miR-200 family using cell transduction and transfection, respectively, to better understand the molecular underpinnings of these phenotypic transitions in sarcoma cells. Sarcoma cells expressing miR-200s and GRHL2 demonstrated enhanced epithelial characteristics in cell morphology and alteration of epithelial and mesenchymal biomarkers. Future studies using these methods can be used to better understand the phenotypic consequences of MET-like processes on sarcoma cells, such as migration, invasion, metastatic propensity, and therapy resistance.
Phenotypic plasticity refers to a reversible transition between cellular phenotypes, and is commonly divided into two types, epithelial-to-mesenchymal (EMT) transitions and mesenchymal-to-epithelial transitions (MET). This phenotypic plasticity plays an important role in normal processes of multicellular organisms, such as development and wound healing1; however, these same pathways and gene expression programs can also lead to disease, such as fibrosis (reviewed in2,3,4) and carcinoma metastasis (reviewed in references5,6,7,8). During metastasis, for example, EMT disrupts cell polarity, cell-cell interactions, and promotes invasion9,10. Together, EMT contributes to a phenotypic state that facilitates cancer cell dissemination. In addition, EMT also leads to a host of other phenotypic alterations that drive an aggressive phenotype, including deregulation of cancer cell metabolism6, development of drug resistance11,12, increased tumor-initiation ability13,14 and host immune evasion15.
Phenotypic plasticity has been well studied in carcinoma progression; however, sarcomas also exhibit phenotypic plasticity. Interestingly, it appears as if some of the same drivers of phenotypic plasticity in carcinomas also contribute to sarcoma plasticity and aggressiveness. For instance, circulating tumor cells (CTCs) from sarcoma patients have been shown to express EpCAM, a cell surface protein that is typically found on epithelial cells16. Additionally, 250 soft tissue sarcoma samples were categorized as epithelial-like or mesenchymal-like based on gene expression. Patients in the epithelial-like biomarker signature had a better prognosis than patients with the mesenchymal-like biomarker signature17. This is consistent with many carcinomas, in which patients with more epithelial-like carcinomas have better outcomes compared with patients with more mesenchymal-like tumors18.
While some sarcomas display biomarkers and gene expression pathways consistent with MET, the molecular underpinnings of this phenotypic plasticity remain poorly understood. To study the mechanisms and drivers of MET in sarcoma we developed a model of MET induction using two epithelial-specific factors, the microRNA (miR)-200 family and grainyhead-like 2 (GRHL2). The miR-200s are a family of small non-coding RNAs that regulate gene expression by binding to the 3' UTRs of messenger RNA and preventing translation into protein. The miR-200 family consists of two subgroups – one containing miR-141 and miR-200a, and the other including miR-200b, miR-200c, and miR-429. Members of the miR-200 family are enriched in epithelial tissues, and the loss of miR-200s is associated with metastasis in carcinomas19. The miR-200 family is also downregulated in soft tissue sarcomas compared to normal tissue20. Similar to the miR-200s, GRHL2 is a key regulator that is important for epithelial development21. The GRHL2 transcription factor acts in two ways to upregulate epithelial genes, such as E-cadherin: 1) In epithelial cells, GRHL2 directly represses the EMT master regulator, ZEB122; and 2) GRHL2 directly activates transcription of epithelial genes23. Our previous investigations have shown that combined expression of miR-200s and GRHL2 in sarcoma cells induces an MET-like phenotype24. Here, we present a detailed protocol to create an in vitro model of MET induction in sarcoma cells using ectopic expression of miR-200s and GRHL2.
1. Preparation of Reagents
2. Lentiviral Transduction of GRHL2
Day 1
Day 2
Day 3
Day 4
Day 5
Day 7
3. Reverse Transfection of miR-200s
4. RNA extraction, Reverse transcription, and qPCR
5. Immunofluorescence Staining
6. Western Blotting
7. Anchorage-independent Growth Assays
NOTE: For a detailed soft agar assay protocol, see 28.
Schema for MET induction in sarcoma cells
A general timeline for the induction of MET-like changes in sarcoma cells is shown in Figure 1. The protocol begins by transducing GRHL2 (Figure 1A), followed by transfection of the miR-200 family (Figure 1B). GRHL2 or miR-200 family members were not able to impact the appearance of RD cells when expressed alone, but ectopic expression of GRHL2 and miR-200s together results in epithelial-like morphological changes in RD cells. Cells transition from a spindle-shaped form to a more rounded appearance with increased cell-cell contact (Figure 1C).
Induction of MET-like changes in sarcoma cells
The morphological change in RD cells upon GRHL2 and miR-200 over-expression was accompanied by upregulation of the epithelial marker E-cadherin (Figure 2A). Addition of miR-200s upregulated E-cadherin alone, but combined miR-200s and GRHL2 overexpression synergistically enhanced E-cadherin expression (Figure 2A; not log scale). In addition, there was an increase at cell-cell junctions of epithelial adhesion molecules, EpCAM and TJP1, white arrows (also known as zona occludens 1, ZO-1) (Figure 2B).
MET induction decreases the colony formation ability of sarcoma cells
Upregulation of epithelial proteins was accompanied by downregulation of mesenchymal genes Zeb1 and Notch1 (Figure 3A), which are known targets for miR-200. Induction of MET reduced the anchorage-independent growth of RD cells as measured by colony number (Figure 3B). This growth inhibition was driven by miR-200s alone; as overexpression of GRHL2 led to an increase in anchorage independent growth (Figure 3). This is consistent with previous reports showing GRHL2 expression induces anchorage-independent growth22.
Figure 1: Timeline of MET Induction with GRHL2 Over-expression and miR-200 Transfection. (A) Timeline of ectopic over-expression of GRHL2 in target cells. (B) Timeline of MET induction via reverse transfection of miR-200 family members in target cells. (C) GRHL2 and miR-200 over-expression led to changes in morphology of target cells consistent with MET. Scale bar = 75 µm Please click here to view a larger version of this figure.
Figure 2: Concurrent Over-expression of GRHL2 and miR-200s Led to MET in Target Cells. (A) Expression of miR-200s led to elevated E-cadherin expression while combined expression of GRHL2 and miR-200s produced a synergistic effect on E-cadherin expression at both the mRNA (mean values ± standard deviation) and protein levels. (B) Combined expression of GRHL2 and miR-200s led to increased expression of EpCAM and TJP1 at cell-cell contacts (arrows). Scale bar = 20 µm. * indicates p< 0.05 analyzed using ANOVA with Tukey's post-hoc correction.This figure has been modified from reference24. Copyright © American Society for Microbiology, Molecular Cell Biology, volume 36, Issue 19, 2503-2513, 2016. Please click here to view a larger version of this figure.
Figure 3: Expression of miR-200s Suppresses Mesenchymal Markers and Decreases Anchorage-independent Growth in Sarcoma Cells. (A) Over–expression of miR-200s but not GRHL2 inhibited Zeb1 and Notch1 mRNA expression (mean values ± standard deviation). (B) Over-expression of miR-200s but not GRHL2 inhibited anoikis resistance in sarcoma cells. Representative images of stained colonies (Scale bar = 200 µm) and quantification of colony number (mean values ± standard deviation) are shown. * indicates p <0.05 analyzed using ANOVA with Tukey's post-hoc correction.This figure has been modified from reference24. Copyright © American Society for Microbiology, Molecular Cell Biology, volume 36, Issue 19, 2503-2513, 2016. Please click here to view a larger version of this figure.
Supplemental Video : RD cells expressing empty vector and negative control miRNAs. Videos were compiled from images of RD cells transfected with empty vector and negative control miRNAs acquired every two hours using an automated live-cell imager. Please click here to view this video. (Right-click to download.)
Supplemental Video 2: RD Cells Expressing GRHL2-EGFP and Negative Control miRNAs. Videos were compiled from images of RD cells transfected with GRHL2-EGFP and negative control miRNAs acquired every two hours using an automated live-cell imager. Please click here to view this video. (Right-click to download.)
Supplemental Video 3: RD Cells Expressing Empty Vector and miR200 miRNAs. Videos were compiled from images of RD cells transfected with empty vector and miR200 miRNAs acquired every two hours using an automated live-cell imager. Please click here to view this video. (Right-click to download.)
Supplemental Video 4: RD Cells Expressing GRHL2-EGFP and miR200 miRNAs. Videos were compiled from images of RD cells transfected with GRHL2-EGFP and miR200 miRNAs acquired every two hours using an automated live-cell imager. Please click here to view this video. (Right-click to download.)
Sarcomas are rare, but highly aggressive cancers of a mesenchymal lineage. Despite their mesenchymal lineage, a subset of sarcomas appears to undergo a phenotypic transition to a more epithelial-like state. This MET-like switch has prognostic relevance, as patients with more epithelial-like tumors are less aggressive24. Despite their clinical relevance, there are few studies addressing the molecular mechanisms driving these phenotypic transitions in sarcomas.
To examine MET-like transitions in sarcoma cells, we have developed an MET-induction model by combining expression of epithelial factors GRHL2 and the miR-200 family. This method rapidly induces sarcoma cells to become more epithelial-like as measured by alterations in morphology and gene expression. Using this protocol to induce MET in sarcoma cells facilitates study of the impact of these transitions on the phenotypes that drive sarcoma aggression, such as migration, invasion, proliferation and death resistance and how each of these changes in biology can affect drug resistance.
In the context of epithelial-derived carcinomas, expression of one mesenchymal factor is often sufficient to induce EMT14. However, in this sarcoma-derived model of MET the expression of two epithelial factors, GRHL2 and miR-200s, are required. Interestingly, the miR-200s alone had a stronger effect on most biomarkers of MET than GRHL2, while GRHL2 was able to robustly activate epithelial genes only in the presence of miR-200-based repression of epithelial gene repressors, such as ZEB124. It is possible that for some mesenchymal cell types epithelial genes need to be both de-repressed (e.g. via miR-200s) and activated (e.g. via GRHL2) to drive MET.
We have experienced variation in the levels of GRHL2 expression across different cell lines. To overcome this, we used a GRHL2 expression plasmid that also expresses EGFP25 to sort by flow cytometry EGFP positive cells prior to experiments. It is also critical to validate the functionality of miR-200s and GRHL2 in different cell types. EMT is viewed as a spectrum based on expression of epithelial and mesenchymal-associated genes, which can have context-dependent variation based on cell line, cancer type, treatment, etc. Therefore, we analyzed five genes regulated by miR-200 or GRHL2 to account for cell context-dependent changes. Another caveat of this assay is the reliance on a transient transfection of miR-200s for MET induction. Thus, long-term experiments can become costly and complicated when multiple repeat transfections become necessary. This can be overcome by using miR-200 expression plasmids.
Other studies have also reported evidence of MET in sarcomas. For instance, MET was observed in a subset of leiomyosarcomas and was associated with better survival for patients. Mechanistically, Yang et al. found that in a leiomyosarcoma cell line inhibition of Slug with siRNA was sufficient to induce MET-like changes29. Likewise, depletion of yet another mesenchymal factor, Snail, in mesenchymal stem cells reduced sarcoma formation in mice30. Thus, it is clear from the literature that there are multiple pathways to an MET-like phenotype, which may vary by cell type. While this is not the only way to induce MET, we have used our method to induce MET in two sarcoma subtypes, including rhabdomyosarcoma (RD cells) and osteosarcoma (143B cells). In the future, it would be interesting to compare these different methods in a broader range of sarcoma cells. For example, do certain sarcoma subtypes have underlying genetic or epigenetic alterations that make them more or less susceptible to MET induction?
Identifying the impact of MET on a variety of biological outputs in sarcomas cells could provide a better understanding of why patients with more epithelial-like sarcomas have an improved prognosis. In addition, understanding the impact of treatment on driving the phenotypic transitions between states would deepen our biological understanding and inform upon response to current therapies in sarcoma patients.
The authors have nothing to disclose.
JAS acknowledges support from the Duke Cancer Institute, The Duke University Genitourinary Oncology Laboratory, and the Duke University Department of Orthopaedics. HL was supported by the National Science Foundation (NSF) Center for Theoretical Biological Physics (NSF PHY-1427654) and NSF DMS-1361411, and as a CPRIT (Cancer Prevention and Research Institute of Texas) Scholar in Cancer Research of the State of Texas at Rice University. KEW was supported by the NIH F32 CA192630 MKJ and HL benefited from useful discussions with Mary C. Farach-Carson, J. N. Onuchic, Samir M. Hanash, Kenneth J. Pienta, and Donald S. Coffey.
Countess automated counter | Life technologies | AMQAX1000 | |
Countess cell counting chamber slides | Invitrogen | C10283 | |
SimpliAmp Thermal Cycler | Thermo Fisher | A24811 | |
Odyssey Fc | LI-COR Inc | ||
ViiA7 Real Time PCR System | Thermo Fisher | 4453536 | |
PCR microplate | Corning | 321-29-051 | |
KAPA SYBR Fast Universal qPCR Kit | KAPA Biosystems | KK4602 | |
Starting Block (PBS) Blocking Buffer | Thermo Fisher | 37538 | BSA-based blocking buffer |
Agarose General Purpose LE | Genesee Scientific | 20-102 | |
10X Tris/Glycine/SDS Buffer | Bio-Rad Laboratories Inc | 161-0732 | Running buffer |
10X Tris/Glycine Buffer | Bio-Rad Laboratories Inc | 161-0734 | Transfer buffer |
RIPA Buffer | Sigma Life Sciences | SLBG8489 | |
Amersham Protran 0.45 μm nitrocellulose | GE Healthcare Lifesciences | 10600012 | |
Quick-RNA MiniPrep Kit | Genesee Scientific | 11-358 | |
Laemmli Sample Buffer (4X) | Bio-Rad Laboratories Inc | 1610747 | |
Mini Trans-Blot Cell | Bio-Rad Laboratories Inc | 1703930 | |
Mini-Protean Tetra Cell | Bio-Rad Laboratories Inc | 1658005EDU | |
DPBS | Life technologies | 14190-144 | |
0.05% Trypsin-EDTA | Life technologies | 11995-065 | |
DMEM | Life technologies | 11995-065 | |
Lipofectamine RNAi Max | Thermo Fisher | 13778150 | |
Lipofectamine 2000 Ragents | Thermo Fisher | 11668019 | |
Penicillin Streptomycin | Life technologies | 15140-122 | |
miRVana miRNA mimic negative control #1 | Thermo Fisher | 4464058 | neg miRNA |
hsa-miR-200 mirVana miRNA mimic | Thermo Fisher | 4464066 | miR200A |
has-miR-200 mirVana miRNA mimic | Thermo Fisher | 4404066 | miR200B |
has-miR-200 mirVana miRNA mimic | Thermo Fisher | 4404066 | miR200C |
Opti-MEM | Life technologies | 11088-021 | serum-free media |
anti-Ecadherin antibody | BD Bioscience | 610182 | |
anti-beta actin | Santa Cruz Biotechnology | sc-69879 | |
anti-EpCam | Ab Serotec | MCA18706 | |
anti-ZO1 | Invitrogen | 402200 | |
IRDye 800W | LI-COR Inc | 925-32210 | |
IRDye 680 | LI-COR Inc | 926-32223 | |
anti-mouse AlexaFluor 647 | Thermo Fisher | A211241 | |
anti-rabbit AlexaFluor 647 | Thermo Fisher | ab150075 | |
Halt Protease and Phosphatesse Inhibitor | Thermo Fisher | 1861281 | |
Precision Plus Protein Dual Color | Bio-Rad Laboratories Inc | 161-0374 | |
Partec CellTrics | Sysmex | 04-004-2326 | 30 μm filter for flow |
GAPDH-F | IDT | AGCCACATCGCTCAGACAC | |
GAPDH-R | IDT | GCCCAATACGACCAAATCC | |
Ecadherin-F | IDT | TGGAGGAATTCTTGCTTTGC | |
Ecadherin-R | IDT | CGCTCTCCTCCGAAGAAAC | |
ZEB1-F | IDT | GCATACAGAACCCAACTTGAACGTC | |
ZEB1-R | IDT | CGATTACACCCAGACTGC | |
NOTCH-F | IDT | GGCAATCCGAGGACTATGAG | |
NOTCH-R | IDT | CTCAGAACGCACTCGTTGAT | |
nitro blue tetrazolium | Sigma | N5514 | |
hexadimethrine bromide | Sigma | H9268 | polybrene |
3 mL syringe | BD Bioscience | 309657 | |
Sterile syringe filter | VWR | 28145-505 | |
5mL polypropylene round-bottom tube | 352063 | flow cytometry tubes | |
High-Capacity cDNA Reverse Transcription Kit | Thermo Fisher | 4368814 | reverse transcription kit |
4% paraformaldyhyde | Santa Cruz Biotechnology | sc-281612 | |
Triton-X100 | Sigma | 93443 | |
bovine serum albumin | Sigma | A7906 |