Loss of the tumor suppressor Snf5 leads to aberrant activation of the Hedgehog-Gli pathway - PubMed (original) (raw)

. 2010 Dec;16(12):1429-33.

doi: 10.1038/nm.2251. Epub 2010 Nov 14.

E Lorena Mora-Blanco, Courtney G Sansam, Elizabeth S McKenna, Boris Wilson, Dongshu Chen, Justin Klekota, Pablo Tamayo, Phuong T L Nguyen, Michael Tolstorukov, Peter J Park, Yoon-Jae Cho, Kathy Hsiao, Silvia Buonamici, Scott L Pomeroy, Jill P Mesirov, Heinz Ruffner, Tewis Bouwmeester, Sarah J Luchansky, Joshua Murtie, Joseph F Kelleher, Markus Warmuth, William R Sellers, Charles W M Roberts, Marion Dorsch

Affiliations

Loss of the tumor suppressor Snf5 leads to aberrant activation of the Hedgehog-Gli pathway

Zainab Jagani et al. Nat Med. 2010 Dec.

Abstract

Aberrant activation of the Hedgehog (Hh) pathway can drive tumorigenesis. To investigate the mechanism by which glioma-associated oncogene family zinc finger-1 (GLI1), a crucial effector of Hh signaling, regulates Hh pathway activation, we searched for GLI1-interacting proteins. We report that the chromatin remodeling protein SNF5 (encoded by SMARCB1, hereafter called SNF5), which is inactivated in human malignant rhabdoid tumors (MRTs), interacts with GLI1. We show that Snf5 localizes to Gli1-regulated promoters and that loss of Snf5 leads to activation of the Hh-Gli pathway. Conversely, re-expression of SNF5 in MRT cells represses GLI1. Consistent with this, we show the presence of a Hh-Gli-activated gene expression profile in primary MRTs and show that GLI1 drives the growth of SNF5-deficient MRT cells in vitro and in vivo. Therefore, our studies reveal that SNF5 is a key mediator of Hh signaling and that aberrant activation of GLI1 is a previously undescribed targetable mechanism contributing to the growth of MRT cells.

PubMed Disclaimer

Conflict of interest statement

COMPETING FINANCIAL INTERESTS

The authors declare competing financial interests: details accompany the full-text HTML version of the paper at http://www.nature.com/naturemedicine/.

Figures

Figure 1

Figure 1

Snf5 interacts with GLI1 and localizes to Gli1 regulated promoters. (a) All proteins precipitated by TAP-GLI and detected by mass spectrometry are indicated by a single data point. The y axis shows the log10 fold ratios comparing the frequency with which each mouse protein was detected in the TAP-GLI1 protein purifications relative to the 26 TAP-protein purifications in our entire mouse database. The x axis shows the associated expected (E) values calculated using binomial statistics corrected for multiple hypotheses (false discovery rate–corrected P values not shown). (b) Immunoblot of GLI1 in whole-cell lysates from TM3 cells transfected with a vector control (pcdna) or GLI1-V5-tagged vector (shown in the first two lanes), and from GLI1-V5 expressing lysates subjected to immunoprecipitation (IP) of endogenous Snf5 with a Snf5-specific antibody or a control IgG (as indicated in lanes three through six). (c) Schematic of the mouse Gli1 and Ptch1 promoters showing locations of primers relative to the ATG translation initiation site. The arrow shows the location of the transcriptional start site (TSS) and the asterisk (*) denotes proximity of the primer set to sequences resembling the Gli1 recognition sequence. (d,e) Quantitative PCR (qPCR) with primers binding the locations depicted in c, showing percentage input recoveries of Snf5 and GLI1 at the Gli1 (d) and Ptch1 (e) promoters in ChIP performed with TM3 GLI1-V5 cells with a Snf5-specific antibody, a V5-specific antibody or a rabbit IgG. qPCR was performed in triplicate and input recovery (%) is shown as mean ± s.d. Ctl, control.

Figure 2

Figure 2

Loss of Snf5 leads to activation of the Hh-Gli pathway in vitro and in vivo. (a) Immunoblot showing reduction of Snf5 protein in TM3 cells expressing Snf5-targeting shRNA but not TM3 cells expressing nontargeting control (Ctl) shRNA. Gapdh, glyceraldehyde 3-phosphate dehydrogenase. (b) Quantitative RT-PCR showing expression of Snf5, Gli1, Gli2, Ptch1 and Smo mRNA in TM3 cells expressing shRNA targeting Snf5. Values are shown as mean ± s.d. (c) Immunoblot showing loss of Snf5 protein in Cre recombinase–treated _Snf5_fl/− (fl1 and fl2 are duplicate samples) primary MEFs; β-actin is shown as a loading control. (d) Quantitative RT-PCR showing expression of Gli1 mRNA in the above (same as in c) Cre-treated _Snf5_fl/− MEFs (labeled as _Snf5_fl/− 1, _Snf5_fl/− 2). The experiment was performed in duplicate, and expression is shown as mean ± s.d. (e) In situ hybridization showing the expression of Gli1 in the limb buds of littermate control (Snf5_fl/+; Prx1_-Cre) and Snf5-deficient (_Snf5_fl/fl; _Prx1_-Cre) embryos at day 11.5. Boxed regions are enlarged beneath their respective images. WT, wild type.

Figure 3

Figure 3

The Hh-Gli pathway is activated in MRT cell lines and primary tumors. (a) Quantitative RT-PCR of expression of GLI1 relative to β-actin in MRT cells (A204, G401, BT12 and BT16) and in glioblastoma (U87MG and LN229), medulloblastoma (DAOY), melanoma (A2058 and A375), multiple myeloma (LP1 and RPMI-8226), breast (MDA-MB231 and MCF7) and pancreas (MiaPaca, PANC1) cancer cell lines. (b) Immunoblot of SNF5 expression in G401 cells after 5 d of selection following retroviral infection with a SNF5-expressing vector or a control vector (Ctl); β-actin is shown as a loading control. (c) Quantitative RT-PCR showing GLI1 expression in Ctl vector–transduced and SNF5-expressing G401 cells. The experiment shown is representative of three independent experiments, and values represent mean ± s.d. of triplicate samples. (d) Heat map showing the single-sample signature profiles using the Hh signaling pathway, basal cell carcinoma (BCC) and GLI1-induced gene expression gene sets in a multi-tumor gene expression panel. ATRT, primary brain MRTs; MRT, three MRT cell lines; NC, normal cerebellum; MD, medulloblastoma; MD SHH, medulloblastoma with Hh pathway activation.

Figure 4

Figure 4

Inhibition of GLI1 impairs proliferation of MRT cells. (a) Immunoblot showing reduction of GLI1 protein in G401 cells upon induction with doxycycline (dox) of GLI1 shRNAs or of a nontargeting control (Ctl) shRNA. GLI1 is overexpressed upon doxycycline treatment in cells carrying both the inducible GLI1 shRNA and inducible GLI1 vectors; Gapdh is included as a loading control. (b) BrdU assay showing cell proliferation over a time course of doxycycline induction of GLI1 shRNA-1 or Ctl shRNAs in G401 cells. (c) BrdU time-course assay in G401 cells showing cell proliferation following doxycycline-induced expression of a control (Ctl) shRNA, GLI1 shRNA-2 or GLI1shRNA-2 accompanied by expression of a doxycycline-inducible GLI1 construct not containing the shRNA recognition sequence. (d) BrdU assay in A204 cells showing cell proliferation following doxycycline induction of Ctl, GLI1 shRNA-1 and GLI1shRNA-2. All assays were performed in triplicate, and BrdU incorporation in doxycycline-treated cells is represented as percentage of BrdU label compared to uninduced cells. Values are shown as mean ± s.d. (e) Colony formation in A204 cells in the absence and presence of doxycycline induction of Ctl shRNA, GLI1 shRNA-1 and GLI1 shRNA-2. A representative experiment of at least three independent experiments performed in triplicate is shown. (f) Tumor volumes over a time course of vehicle- and doxycycline-treated animals bearing G401 MRT cells containing inducible GLI1 shRNA-1. Tumor volume is reported as mean ± s.e.m.; n = 8 per treatment group.

Comment in

References

    1. Jiang J, Hui CC. Hedgehog signaling in development and cancer. Dev Cell. 2008;15:801–812. -PMC -PubMed
    1. Ruiz i Altaba A, Mas C, Stecca B. The Gli code: an information nexus regulating cell fate, stemness and cancer. Trends Cell Biol. 2007;17:438–447. -PMC -PubMed
    1. Gailani MR, et al. The role of the human homologue of Drosophila patched in sporadic basal cell carcinomas. Nat Genet. 1996;14:78–81. -PubMed
    1. Raffel C, et al. Sporadic medulloblastomas contain PTCH mutations. Cancer Res. 1997;57:842–845. -PubMed
    1. Kinzler KW, et al. Identification of an amplified, highly expressed gene in a human glioma. Science. 1987;236:70–73. -PubMed

MeSH terms

Substances

LinkOut - more resources