Helen Poppleton

Prominin 1 marks intestinal stem cells that are susceptible to neoplastic transformation

Cancer stem cells are remarkably similar to normal stem cells: both self-renew, are multipotent and express common surface markers, for example, prominin 1 (PROM1, also called CD133). What remains unclear is whether cancer stem cells are the direct progeny of mutated stem cells or more mature cells that reacquire stem cell properties during tumour formation. Answering this question will require knowledge of whether normal stem cells are susceptible to cancer-causing mutations; however, this has proved difficult to test because the identity of most adult tissue stem cells is not known. Here, using an inducible Cre, nuclear LacZ reporter allele knocked into the Prom1 locus (Prom1C-L), we show that Prom1 is expressed in a variety of developing and adult tissues. Lineage-tracing studies of adult Prom1+/C-L mice containing the Rosa26-YFP reporter allele showed that Prom1+ cells are located at the base of crypts in the small intestine, co-express Lgr5 (ref. 2), generate the entire intestinal epithelium, and are therefore the small intestinal stem cell. Prom1 was reported recently to mark cancer stem cells of human intestinal tumours that arise frequently as a consequence of aberrant wingless (Wnt) signalling. Activation of endogenous Wnt signalling in Prom1+/C-L mice containing a Cre-dependent mutant allele of β-catenin (Ctnnb1lox(ex3)) resulted in a gross disruption of crypt architecture and a disproportionate expansion of Prom1+ cells at the crypt base. Lineage tracing demonstrated that the progeny of these cells replaced the mucosa of the entire small intestine with neoplastic tissue that was characterized by focal high-grade intraepithelial neoplasia and crypt adenoma formation. Although all neoplastic cells arose from Prom1+ cells in these mice, only 7% of tumour cells retained Prom1 expression. Our data indicate that Prom1 marks stem cells in the adult small intestine that are susceptible to transformation into tumours retaining a fraction of mutant Prom1+ tumour cells.

Subtypes of medulloblastoma have distinct developmental origins.

Medulloblastoma encompasses a collection of clinically and molecularly diverse tumour subtypes that together comprise the most common malignant childhood brain tumour. These tumours are thought to arise within the cerebellum, with approximately 25% originating from granule neuron precursor cells (GNPCs) after aberrant activation of the Sonic Hedgehog pathway (hereafter, SHH subtype). The pathological processes that drive heterogeneity among the other medulloblastoma subtypes are not known, hindering the development of much needed new therapies. Here we provide evidence that a discrete subtype of medulloblastoma that contains activating mutations in the WNT pathway effector CTNNB1 (hereafter, WNT subtype) arises outside the cerebellum from cells of the dorsal brainstem. We found that genes marking human WNT-subtype medulloblastomas are more frequently expressed in the lower rhombic lip (LRL) and embryonic dorsal brainstem than in the upper rhombic lip (URL) and developing cerebellum. Magnetic resonance imaging (MRI) and intra-operative reports showed that human WNT-subtype tumours infiltrate the dorsal brainstem, whereas SHH-subtype tumours are located within the cerebellar hemispheres. Activating mutations in Ctnnb1 had little impact on progenitor cell populations in the cerebellum, but caused the abnormal accumulation of cells on the embryonic dorsal brainstem which included aberrantly proliferating Zic1+ precursor cells. These lesions persisted in all mutant adult mice; moreover, in 15% of cases in which Tp53 was concurrently deleted, they progressed to form medulloblastomas that recapitulated the anatomy and gene expression profiles of human WNT-subtype medulloblastoma. We provide the first evidence, to our knowledge, that subtypes of medulloblastoma have distinct cellular origins. Our data provide an explanation for the marked molecular and clinical differences between SHH- and WNT-subtype medulloblastomas and have profound implications for future research and treatment of this important childhood cancer.

Cross-species genomics matches driver mutations and cell compartments to model ependymoma

Understanding the biology that underlies histologically similar but molecularly distinct subgroups of cancer has proven difficult because their defining genetic alterations are often numerous, and the cellular origins of most cancers remain unknown. We sought to decipher this heterogeneity by integrating matched genetic alterations and candidate cells of origin to generate accurate disease models. First, we identified subgroups of human ependymoma, a form of neural tumour that arises throughout the central nervous system (CNS). Subgroup-specific alterations included amplifications and homozygous deletions of genes not yet implicated in ependymoma. To select cellular compartments most likely to give rise to subgroups of ependymoma, we matched the transcriptomes of human tumours to those of mouse neural stem cells (NSCs), isolated from different regions of the CNS at different developmental stages, with an intact or deleted Ink4a/Arf locus (that encodes Cdkn2a and b). The transcriptome of human supratentorial ependymomas with amplified EPHB2 and deleted INK4A/ARF matched only that of embryonic cerebral Ink4a/Arf−/− NSCs. Notably, activation of Ephb2 signalling in these, but not other, NSCs generated the first mouse model of ependymoma, which is highly penetrant and accurately models the histology and transcriptome of one subgroup of human supratentorial tumour. Further, comparative analysis of matched mouse and human tumours revealed selective deregulation in the expression and copy number of genes that control synaptogenesis, pinpointing disruption of this pathway as a critical event in the production of this ependymoma subgroup. Our data demonstrate the power of cross-species genomics to meticulously match subgroup-specific driver mutations with cellular compartments to model and interrogate cancer subgroups.

Stem cells of ependymoma

Ependymomas are tumours that arise throughout the central nervous system. Little is known regarding the aberrant cellular and molecular processes that generate these tumours. This lack of knowledge has hampered efforts to reduce the significant mortality and morbidity that are associated with ependymoma. Here, we review recent data that suggest that radial glia are cells of origin of ependymoma, and discuss the processes that might transform these neural progenitors into ependymoma cancer stem cells.

A procedure to statistically evaluate agreement of differential expression for cross-species genomics

MOTIVATION Animal models play a pivotal role in translation biomedical research. The scientific value of an animal model depends on how accurately it mimics the human disease. In principle, microarrays collect the necessary data to evaluate the transcriptomic fidelity of an animal model in terms of the similarity of expression with the human disease. However, statistical methods for this purpose are lacking. RESULTS We develop the agreement of differential expression (AGDEX) procedure to measure and determine the statistical significance of the similarity of the results of two experiments that measure differential expression across two groups. AGDEX defines a metric of agreement and determines statistical significance by permutation of each experiments group labels. Additionally, AGDEX performs a comprehensive permutation-based analysis of differential expression for each experiment, including gene-set analyses and meta-analytic integration of results across studies. As an example, we show how AGDEX was recently used to evaluate the similarity of the transcriptome of a novel model of the brain tumor ependymoma in mice to that of a subtype of the human disease. This result, combined with other observations, helped us to infer the cell of origin of this devastating human cancer. AVAILABILITY An R package is currently available from www.stjuderesearch.org/site/depts/biostats/agdex and will shortly be available from www.bioconductor.org.

Abstract LB-63: Cell ground state dictates cancer susceptibility across organs

Cancers are distributed unevenly across the body, suggesting that cell context dictates susceptibility to transformation. To test this hypothesis, we mapped cellular susceptibility to tumor initiation across all organs in adult mice. Ten combinations of potent oncogenic mutations, including activating mutations of Kras, Notch1, Ctnnb1 and deletions of tumor suppressors Tp53,Pten and Cdkn2a, were targeted by conditional recombination into Prom1+ (CD133+) cells in the organs of >800 Prom1-Cre/+ mice. Life time clinical surveillance, necropsy and gene expression profiling of over 250 animals, across all genotypes, identified extensive tumor formation that was biased to certain organs including small intestine, stomach, liver, skin, prostate and uterus. Extensive histology and cross-species genomics confirmed these tumors recapitulated >250 tissue-matched human cancers. Detailed mapping of the number and proliferative activity of Prom1+ cells in organs demonstrated that these variables did not dictate cancer incidence. Rather, parallel, life-long linage tracing of Prom1+ populations showed that the capacity of these cells to repair and regenerate their host organ is directly correlated with their tumorigenic capacity. Only Prom1+ cells capable of long-term tissue renewal generated cancers; quiescent and differentiated Prom1+ cells rarely formed tumors regardless of targeted oncogenic mutation. The repair/regeneration capacity of Prom1+ populations also dictated developmental and damage-response patterns of tumorigenesis. Normally quiescent and tumor-resistant Prom1+ cells in the adult liver became highly regenerative upon liver damage, dramatically increasing their tumorigenic capacity. Conversely, Prom1+ cells are highly proliferative in the neonatal liver - driving the development of this organ - and were highly susceptible to transformation without inducing liver damage. Remarkably, neonatal Prom1+ cells formed hepatoblastoma, the pediatric form of liver cancer. Our large-scale, comprehensive mapping of tumor susceptibility shows that cell ground-state, and specifically the capacity to repair and regenerate organs, is a common determinant of cancer susceptibility across all major organs. Citation Format: Liqin Zhu, Helen Poppleton, Geoffrey Neale, Arzu Onar-Thomas, Stanley Pounds, Armita Bahrami, Sarah Utley, Kasper S. Wang, Dolores H. Lopez-Terrada, David W. Ellison, Richard J. Gilbertson. Cell ground state dictates cancer susceptibility across organs. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr LB-63. doi:10.1158/1538-7445.AM2014-LB-63

Abstract 3448: Subtypes of medulloblastoma have distinct developmental origins

Proceedings: AACR 102nd Annual Meeting 2011‐‐ Apr 2‐6, 2011; Orlando, FL Medulloblastoma encompasses a collection of clinically and molecularly diverse tumor subtypes that together comprise the most common malignant childhood brain tumor. These tumors are thought to arise within the cerebellum, with approximately 25% originating from granule neuron precursor cells (GNPCs) following aberrant activation of the Sonic Hedgehog pathway (hereafter, SHH-subtype). The pathological processes that drive heterogeneity among the other medulloblastoma subtypes are not known, hindering the development of much needed new therapies. Here, we provide evidence that a discrete subtype of medulloblastoma that contains activating mutations in the WNT pathway effector CTNNB1 (hereafter, WNT-subtype), arises outside the cerebellum from cells of the lower rhombic lip (LRL). We found that genes marking human WNT-subtype medulloblastomas are more frequently expressed in the LRL and embryonic dorsal brainstem than in the upper rhombic lip (URL) and developing cerebellum. Magnetic resonance imaging (MRI) and intra-operative reports showed that human WNT-subtype tumors infiltrate the dorsal brainstem, while SHH-subtype tumors are located within the cerebellar hemispheres. Activating mutations in Ctnnb1 had little impact on progenitor cell populations in the cerebellum, but caused an aberrant accumulation of proliferating precursor cells within the LRL. These lesions persisted in the dorsal brainstem of all mutant adult mice and in 15% of cases in which Tp53 was concurrently deleted, progressed to form medulloblastomas that modeled faithfully the anatomy and gene expression profiles of human WNT-subtype medulloblastoma. We provide the first evidence that subtypes of medulloblastoma have distinct cellular origins. Our data provide an explanation for the marked molecular and clinical differences between SHH and WNT-subtype medulloblastomas and have profound implications for future research and treatment of this important childhood cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3448. doi:10.1158/1538-7445.AM2011-3448

Abstract 4759: Integrated in vitro and in vivo screening of tumor and normal neural stem cells identifies potential new treatments of ependymoma

Ependymomas are rare brain tumors that are incurable unless completely excised. The low incidence of the disease and lack of pre-clinical models has limited research efforts to advance understanding of biology and treatment. Recently, we used interspecies genomics to match specific driver mutations with distinct types of mouse neural stem cell (NSC) to accurately model human ependymoma. Here we report the use of these models for high throughput drug screening (HTS). Stem-like mouse ependymoma cells (mEPCs), non-ependymoma mouse brain tumor cells (mBTCs) and control transduced NSCs (mNSCs) were cultured in neurosphere conditions adapted for use in an automated small molecule HTS. We first performed replicate primary screens of 7,579 agents drawn from a bioactive library, FDA approved drug library and kinase library. Primary screens were conducted in a single concentration format (10µM). The primary screen was highly reproducible and ROC analysis of primary screen data was used to assess predictive power of the screen (ROC AUC>0.89 (0.85-0.92 95% CI)). A total of 602 compounds representing diverse drug classes progressed from primary to secondary screening. These included full 10-point dose response assays that identified a total of 181 agents with activity in at least one cell population. In all, 2.3% of compounds displayed anti-mEPC activity and were enriched for anti-cancer drugs (Fishers Exact P=1.9 × 10-7: Bonferroni correction threshold, p=0.0016). Since our HTS strategy included non-ependymoma tumor cells and mNSCs, we were able to refine our classification of activity to define compounds more active against mEPCs than other cells (0.08%); equally active against mEPCs and mBTCs relative to mNSCs (0.04%); equally active against all four cell types (2.2%); more active against mBTCs relative to all other cells (0.2%); more active against mNSCs relative to tumor cells (0.8%); inactive against mEPCs relative to all other cells (0.1%); and inactive against all four cell types (96%). Interestingly, anticancer compounds displayed patterns of cell-selective activity that varied according to their mechanism of action with some drug classes appearing significantly more toxic to mNSCs than either mEPCs or mBTCs. These HTS studies identified treatments for ependymoma including drugs that were relatively non-toxic to normal NSCs. Five agents were selected for assessment of in vivo against the originating ependymoma mouse models. Using the xenogen system to monitor tumor growth and assessing animal survival, we identified FDA-approved agents with activity in ependymoma which have not previously been implicated in the disease, and may be translated directly into the clinic. In summary, this approach identified a number of potential new treatments with potent activity against ependymoma relative to normal NSCs, and could be used to develop effective therapies for other rare cancers. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4759. doi:10.1158/1538-7445.AM2011-4759

Abstract A52: An interspecies genomics based high‐throughput screen for novel treatments of ependymoma

Brain tumors are often chemoresistant and these diseases are heterogeneous complicating efforts to discover effective new therapies. We describe a powerful interspecies genomics approach that meticulously matches subgroup specific driver mutations with cellular compartments to model cancer subgroups for drug screening. First, we performed a comprehensive genomics analysis to identify disease subgroups among >200 ependymomas. Subgroup specific alterations included amplifications and homozygous deletions of genes not yet implicated in ependymoma. We then identified CNS cell compartments most likely to give rise to ependymoma subgroups by matching the transcriptomes of human tumors to those of distinct types of mouse neural stem cell (NSC). Remarkably, activation of oncogenes in appropriate NSCs generated ependymomas that modelled the histology and transcriptome of the human disease. Stem‐like tumor cells isolated from these mouse ependymomas were then cultured under conditions that promote stem cell growth as neurospheres. These conditions were adapted for use in an automated high‐throughput screening system. Control transduced NSCs were also included to identify ependymoma‐selective agents. Tumor and control spheres were seeded in 384 well plates and drug treated using pin tool transfer 24 hours after plating. Cells were exposed to drug for 96 hours before compound cytotoxicity measured using the cell‐titre glo luciferase based assay. We first performed replicate primary screens of a large ‘bioactive library’ including natural products, bioactive alkaloids and marketed drugs (5760 compounds [3460 unique in structure]) in a single concentration format (10uM). Active compound hit rate varied from 0.9% to 4.8% in spheres derived from mouse tumors and 3.1% in control NSCs. The primary screen was highly reproducible (263 and 261 hits in replicate assays, of which 226 of were common in an example tumor line). ROC analysis of primary screen data was used to assess predictive power of the screen. For all cells the ROC AUC was >0.89 (0.85–0.92 95% CI). We next performed secondary screens of all primary screen hits. These included full 10‐point dose response assays that identified a total of 292 active agents with activity in at least one cell population. Following analysis of the ‘bioactive library, we screened a collection of 320 FDA approved active pharmaceutical ingredients representing all drug classes and all approved chemotherapy agents in a 10‐point dose response format. The hit rate of this drug collection in all cell types was between 10.36% to 15.4% and hit compounds represented a variety of drug classes including chemotherapeutics, NSAIDs, antibiotics and dopamine‐like agents. These compounds and the related families and structures will be described in detail as well as the results of ongoing in vivo activity against the originating ependymoma mouse models. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):A52.