Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 May 13;11(1):2380.
doi: 10.1038/s41467-020-16167-y.

YAP1/TAZ drives ependymoma-like tumour formation in mice

Affiliations

YAP1/TAZ drives ependymoma-like tumour formation in mice

Noreen Eder et al. Nat Commun. .

Erratum in

Abstract

YAP1 gene fusions have been observed in a subset of paediatric ependymomas. Here we show that, ectopic expression of active nuclear YAP1 (nlsYAP5SA) in ventricular zone neural progenitor cells using conditionally-induced NEX/NeuroD6-Cre is sufficient to drive brain tumour formation in mice. Neuronal differentiation is inhibited in the hippocampus. Deletion of YAP1's negative regulators LATS1 and LATS2 kinases in NEX-Cre lineage in double conditional knockout mice also generates similar tumours, which are rescued by deletion of YAP1 and its paralog TAZ. YAP1/TAZ-induced mouse tumours display molecular and ultrastructural characteristics of human ependymoma. RNA sequencing and quantitative proteomics of mouse tumours demonstrate similarities to YAP1-fusion induced supratentorial ependymoma. Finally, we find that transcriptional cofactor HOPX is upregulated in mouse models and in human YAP1-fusion induced ependymoma, supporting their similarity. Our results show that uncontrolled YAP1/TAZ activity in neuronal precursor cells leads to ependymoma-like tumours in mice.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. YAP1 is present in ependymal layer and its regulation is critical for brain development.
a Coronal view of control P0 mouse brain sections immunofluorescence stained for YAP1. Arrowhead pointing at strong positive staining in the ependymal layer, scale bars = 1 mm. bc Higher magnifications of ependymal layer/ventricular zone in control mice at P0. b Immunostainings with YAP1 and Nestin and c. Immunostainings with YAP1 and Ki67 show colocalization (arrows point to cells with colocalizations, overlap shown in white colour). Scale bars = 10 µm. d Breeding scheme to generate mice expressing nlsYAP5SA under the control of NEX-Cre (referred to as nlsYAP5SA). Mice with only one allele of NEX-Cre without the YAP1 transgene were used as controls. e Sagittal vibratome sections of the whole brain of Control and nlsYAP5SA mice are shown at P14. In nlsYAP5SA mice, ependymal layer is destructed, multiple tumours are formed across the brain as shown by DAPI stained nuclei and hippocampus is not developed. Arrowheads point at hippocampus. Scale bar = 1 mm.
Fig. 2
Fig. 2. Uncontrolled YAP1 in NEX-Cre-expressing cells is sufficient for tumour formation at the expense of hippocampal differentiation.
a Immunofluorescence staining of YAP1 in coronal vibratome sections of P0 animals. Neurons in nlsYAP5SA mouse cortex show strong nuclear staining of YAP1. Hippocampus formation is altered in nlsYAP5SA (arrowheads). Scale bars = 1 mm. b Arrowheads pointing at tumours visible in coronal vibratome sections of a P0 nlsYAP5SA animal. Scale bars = 1 mm. c Immunofluorescence staining in coronal sections of tumours in a nlsYAP5SA mice. Tumours are outlined with yellow-dashed line. Nestin, Ki67 and MUC1 demonstrate highly intense staining in the tumours while NeuN is absent. Scale bars = 100 µm.
Fig. 3
Fig. 3. Conditional knockout of LATS1/2 under control of NEX-Cre leads to clonal YAP1-positive tumours.
a Breeding scheme to generate LATS1/2 conditional knockout (LATS1/2 cKO) mice and control littermates. b Tumours associated with the ependymal layer are found in LATS1/2 cKO mice at P20 (arrowheads). Multiple clonal tumours can be found in each brain (middle and right images are from the same brain at different coronal planes). Scale bars = 1 mm. c A coronal section of LATS1/2 cKO brain at P6 shows strong YAP1 immunofluorescence staining in a tumour arising from the ependymal layer (arrowhead). Scale bars =  mm. de Confocal images of immunofluorescence stainings in LATS1/2 cKO brains. Scale bars = 10 µm. d YAP1 staining is increased in the tumour and YAP1 is nuclear in some cells (arrowheads). YAP1 expression does not overlay with GFAP+ cells. e YAP1 and Nestin are co-expressed in the tumour cells (arrows).
Fig. 4
Fig. 4. YAP1 and TAZ are necessary for tumour development in LATS1/2 cKO mice.
a Breeding scheme to generate a rescue model. Available animals (Lats1f/f Lats2f/f, YAPf/f TAZf/+ and NEXCre/Cre) were crossed to generate mice with shown genotypes. b Measurements of size of tumours found in LATS1/2 cKO (n = 16 tumours), Mutant YAPf/+ (n = 15 tumours) and Mutant YAPf/f (n = 11 tumours) from n = 3 mice for each genotype. Colour shade indicates tumours from the same animal. LATS1/2 cKO tumour volumes compared with those measured in Mutant YAPf/f animals show a significant difference with padj = 0.0006 (two-sided Dunn’s multiple comparison test). Error bars = SEM. c Coronal vibratome sections of P20 brains stained for DAPI. No tumour formation identified in Rescue mice. Tumours (arrowheads) can be found in LATS1/2 cKO, Mutant YAPf/f and Mutant YAPf/+ animals. Scale bars = 1 mm.
Fig. 5
Fig. 5. LATS1/2 cKO mouse tumours display histological features of ependymoma.
ad The tumour (arrowhead) appear to arise from the ependymal layer of the anterior temporal horn of the lateral ventricle at the level of hippocampus (* CA3). a H&E, ×4, scale bar = 500 µm; the lesion has compressive margin. b H&E, ×10, scale bar = 200 µm and it is composed of fascicles of spindle cells with focal nuclear clustering. c. H&E, ×40, scale bar = 50 µm. A few tumour cells contain intracytoplasmic, paranuclear vacuoles (arrowheads). d. H&E, ×60, scale bar = 20 µm. Tumour cells demonstrate. e nuclear and cytoplasmic YAP1 expression (immunoperoxidase, ×20), f weak cytoplasmic MUC1 expression (immunoperoxidase, ×20) and g they are focally positive for GFAP (immunoperoxidase, ×20) and h cytokeratin 18/CK18 (immunoperoxidase, ×10). Reactive astrocytes surrounding the lesion are also GFAP-positive (g—immunoperoxidase, ×20). Scale bars = 50 µm. i TEM image (×9900) within the tumour showing multiple ependymal features (Yellow arrowheads = microvilli, red arrows = tight junctions). Scale bar = 1 µm. j, k. Enlargement of the yellow and red insets in i, respectively. Scale bars = 200 nm. j. High magnification of microvilli and k. a tight junction found in the tumour are shown.
Fig. 6
Fig. 6. Transcriptome and proteome analysis of nlsYAP5SA mice exhibit further parallels to human ependymoma.
ae Gene expression analysis for nlsYAP5SA versus control. Five animals per genotype (control = NEX+/+ YAP1nlsYAP5SA/+; nlsYAP5SA = NEXCre/+ YAP1nlsYAP5SA/+) were analysed utilising the Illumina HiSeq 4000. a Heat map of all significantly differently expressed genes (2035 genes) between control and nlsYAP5SA animals. Rows represent genes and columns show individual animals. Legend indicates per gene median centred fold change. b Gene enrichment analysis (GSEA) of complete gene list (30,280 detected genes) from RNA sequencing (RNASeq) with gene sets derived from the KEGG pathway database. The top 10 annotated pathways with the highest normalised enrichment score (NES) are shown with number of genes in the data set linked to the indicated pathway (Size). c GSEA of KEGG annotation “Pathways in Cancer” with highest NES in analysis. d Gene list of YAP1 and RELA association (see methods) was used to assign specificity to detected genes in our expression data from nlsYAP5SA screen (350 genes of the 354 in reference list are present in our data set). Log2 fold change of control vs. nlsYAP5SA are ranked from highest to lowest. Teal and red colours are used to denote YAP-specific and RELA-specific genes, respectively. e The 10 highest ranked genes with assigned specificity all of these are associated with YAP1. f Venn diagrams illustrating numbers of identified genes in Transcriptome (RNAseq, blue) and proteome quantified by liquid chromatography mass spectrometry (LC-MS) (red). Left: all genes, right: significantly upregulated genes in nlsYAP5SA. g Density scatter plot of 6439 genes present in both proteome and transcriptome plotted by their expression difference between control and nlsYAP5SA (Welch difference is used for proteomics and stat value is used for transcriptomics). Data sets show a correlation coefficient of 0.5. Each point represents one gene. h Volcano plot of difference in protein levels between control and nlsYAP5SA cohort. Each point represents one protein. The x axis shows log2 transformed fold change and the y axis shows significance by –log10 transformed p value obtained by two-sided Welch t test. A gene is called significantly and differentially expressed if its FDR is <0.05 and s0 > 0.1, which is indicated by the black line. Proteins associated with astrocytes (magenta) and microglia (green) are displayed. Cell type association was generated by integrating previously published data (see methods). i Area shown in h, enlarged to display and name significantly upregulated proteins in nlsYAP5SA brains.
Fig. 7
Fig. 7. Validation of markers identified in our proteomics and transcriptomics screen in mouse and human models of ependymoma.
a Western blots and quantification of proteins levels comparing control and nlsYAP5SA animals (control = NEX+/+ YAP1nlsYAP5SA/+; nlsYAP5SA = NEXCre/+ YAP1nlsYAP5SA/+) (n = 3 mice, each). Significant increase in YAP1 (p = 0.0042), Vimentin (p = 0.0017), GFAP (p < 0.0001) and HOPX (p = 0.0002) are observed in nlsYAP5SA (two-tailed unpaired t tests). No significant difference in protein level was observed for TAZ nor CTGF. Error bars = SEM. b Immunofluorescence stainings for selected YAP1 downstream genes in LATS1/2 cKO tumours at P20. Strong staining of ANKRD1, AMOTL2 and AXL are observed in the tumour. ANKRD1 was uniformly expressed in throughout all tumour cells, AMOTL2 and AXL exhibited higher levels in tumour near its border. Scale bar = 100 μm. c Immunofluorescence of HOPX and YAP1 in a LATS1/2 cKO tumour at P20, showing higher staining at tumour edge, HOPX co-localises with YAP1 in this area (white shows colocalisation, Scale bar = 100 μm). d A subset of HOPX expressing cells have clear, nuclear YAP1 staining, example cell shown from a different LATS1/2 cKO tumour (Scale bar = 10 μm). e Two YAP1-MAMLD1-fused ependymoma characterised by uniform cells forming pseudo-rosettes in a fibrillary background (H&E, scale bar = 100 μm (Patient 1), 200 μm (Patient 2)); tumour cells show YAP1 nuclear expression (immunoperoxidase, scale bar = 50 μm); there is cytoplasmic and nuclear expression of HOPX (immunoperoxidase, scale bar = 100 μm), whereas p65 is negative in tumour cells (immunoperoxidase, scale bar = 100 μm).

References

    1. Pan D. The hippo signaling pathway in development and cancer. Dev. Cell. 2010;19:491–505. doi: 10.1016/j.devcel.2010.09.011. - DOI - PMC - PubMed
    1. Totaro A, Panciera T, Piccolo S. YAP/TAZ upstream signals and downstream responses. Nat. Cell Biol. 2018;20:888–899. doi: 10.1038/s41556-018-0142-z. - DOI - PMC - PubMed
    1. Harvey KF, Zhang X, Thomas DM. The Hippo pathway and human cancer. Nat. Rev. Cancer. 2013;13:246–257. doi: 10.1038/nrc3458. - DOI - PubMed
    1. Zanconato F, Cordenonsi M, Piccolo S. YAP/TAZ at the roots of cancer. Cancer Cell. 2016;29:783–803. doi: 10.1016/j.ccell.2016.05.005. - DOI - PMC - PubMed
    1. Irvine KD. Integration of intercellular signaling through the Hippo pathway. Semin. Dev. Biol. 2012;23:812–817. doi: 10.1016/j.semcdb.2012.04.006. - DOI - PMC - PubMed

Publication types

MeSH terms