Despite numerous advances, direct reprogramming of fibroblasts to induced cardiomyocytes remains an inefficient process with largely unknown underlying mechanisms. To understand the molecular events associated with direct cardiac reprogramming, we characterized genome-wide transcriptional changes in individual and bulk reprogramming cell populations, examined genomic changes in chromatin accessibility and DNA methylation levels during reprogramming, and mapped DNA binding by Gata4, Mef2c, and Tbx5 (GMT) that initiates the transition from fibroblast to induced cardiomyocyte. We discovered that the transition to cardiac-like gene expression occurs within 48 hours of initiating transdifferentiation. This change is accompanied by rapid epigenomic remodeling and includes dynamic chromatin regions enriched for transcription factor motifs in addition to GMT, with different motif families associated with specific patterns of chromatin and DNA methylation changes. We found evidence for binding of the reprogramming factors to dynamically-opening distal regulatory regions linked to cardiac function. In addition, we document the mechanism through which TGFβ and Wnt signaling pathway inhibition improves reprogramming efficiency by limiting aberrant reprogramming outcomes. These results provide insight into the mechanisms that drive direct cardiac reprogramming and provide a hypothesis-building framework for improving this process.

Low-grade glioma (LGG) is a diffuse and infiltrative adult brain tumor. Due to the invasive nature of the tumor, surgical resection is rarely curative. LGG often undergo malignant progression and recur as a high-grade glioblastoma (GBM). The genomic and epigenomic landscapes of these tumors, particularly at recurrence, are understudied yet clinically important. Recurrent tumors may be driven by a distinct set of genetic and epigenetic alterations than their initial tumors, yet therapeutic decisions are often made based on profiling of initial malignancies. Here we comprehensively profiled 33 LGGs and their patient-matched recurrences, including cases with 2-6 intratumoral samples, with exome sequencing to identify somatic mutations, with the Infinium 450K array to investigate DNA methylation changes, and with RNA sequencing to measure gene expression changes. We found a wide range in the degree of evolution from initial to recurrence, in terms of both somatic mutations and DNA methylation changes. Tumors treated with the alkylating chemotherapy temozolomide (TMZ) often recurred with a hypermutation phenotype that was suggestive of therapy-driven malignant progression to GBM. Therapy-associated mutational activation of the AKT-mTOR pathway was a consistent feature of these hypermutated recurrences, which promoted a new clinical trial combining the anti-tumor activity of TMZ with an mTOR inhibitor. Recurrence as GBM was associated with aberrations to cell cycle genes through convergence of both genetic and epigenetic mechanisms. Moreover, we found that the evolutionary history of a tumor is similar whether inferred from genetic or epigenetic data, suggesting co-evolution of the genome and epigenome. Finally, we identified cases in which mutations in IDH1, which are the earliest known alteration in LGG and drive gliomagenesis, were either deleted or amplified at recurrence. Mutant IDH1 reprograms the epigenome and these recurrences showed partial reversion of these epigenomic alterations. Together, these findings highlight the heterogeneity and continual evolution of LGG and emphasize the importance of studying recurrent tumors for a more complete understanding of tumor evolution and to make more informed treatment decisions.

Telomeres are repetitive sequences of DNA that protect the ends of chromosomes and are gradually lost each cycle of cell division. In cells such as stem cells and germ cells, telomeres are indefinitely maintained through the use of the enzymatic complex telomerase. In order to achieve replicative immortality and form a tumor, cancer cells must find a way to replenish telomeres early on during tumorigenesis. The most common way that cancer cells enable immortality is by reactivating expression of the catalytic subunit, Telomerase Reverse Transcriptase (TERT), which is normally silenced in somatic cells. Activating mutations in the promoter region of TERT gene are the most common mechanism through which tumor cells reactivate telomerase, allowing for indefinite telomere maintenance and enabling cellular immortalization. These mutations specifically recruit the multimeric ETS factor GABP, which can form two functionally independent transcription factor species – a dimer or a tetramer.

We have identified GABPβ1L, the tetramer-forming isoform of GABP that is dispensable for normal development, as being specifically recruited to the mutant TERT promoter in glioblastoma cells. We show that genetic disruption of GABPβ1L results in TERT silencing in a TERT promoter mutation-dependent manner. Reducing TERT expression by disrupting GABPβ1L culminates in telomere loss and cell death through exclusively in TERT promoter mutant cells. Orthotopic xenografting of GABPβ1L-reduced, TERT promoter mutant glioblastoma cells rendered lower tumor burden and longer overall survival in mice. These results highlight the critical role of GABPβ1L in enabling immortality in TERT promoter mutant glioblastoma.

TERT promoter mutations are the third most common mutation in human cancer, and the single most common mutation in glioblastoma. Understanding how the promoter mutation leads to tumor cell immortality could uncover potential targets to undermine immortality and reduce tumor growth. TERT promoter mutations selectively recruit the transcription factor GABP to activate TERT expression across multiple types of cancer. Our results suggest that the normally dispensable GABPβ1L isoform of GABP is a key to tumor cell immortality in TERT promoter mutant brain tumors. Therefore, inhibiting GABPβ1L may be an approach to reverse tumor cell immortality while sparing TERT promoter wild-type cells.

Meningiomas arising from the meningothelial lining of the central nervous system are the most common primary intracranial tumors, and a significant cause of neurologic morbidity and mortality. There are no effective medical therapies for meningioma patients, and new treatments have been encumbered by limited understanding of meningioma biology. DNA methylation profiling provides robust classification of brain tumors, but has not informed new treatments for patients.

Here we use DNA methylation profiling on 565 meningiomas, controlled for confounding artifacts from copy number variants and integrated with genetic, transcriptomic, biochemical, and single-cell approaches, to show meningiomas are comprised of 3 groups with distinct clinical outcomes, biological drivers, and therapeutic vulnerabilities. Merlin-intact meningiomas have the best outcomes and are distinguished by NF2/Merlin regulation of glucocorticoid signaling and apoptosis. Immune-enriched meningiomas have intermediate outcomes and are distinguished by immune infiltration, HLA expression, and lymphatic vessels. Hypermitotic meningiomas have the worst outcomes and are distinguished by convergent genetic mechanisms misactivating the cell cycle. Translating these findings, we show cell cycle inhibitors block Immune-enriched and Hypermitotic meningioma growth in cell culture, organoids, xenografts, and patients.

To extend the relevance of our findings, we use single-cell RNA sequencing of 8 human meningiomas, 3 canine meningiomas, and 1 human hemangiopericytoma to identify Notch3+ perivascular cells as a potential cell of origin for meningeal tumors. This perivascular cell population is conserved across meningeal tumors in both humans and canines and across normal meningeal tissue in both humans and mice. These findings set the stage for developing new mouse models of meningioma, which will accelerate and enhance our ability to study meningioma biology, such as interrogating the pathways and phenotypes identified from DNA methylation groups. Moreover, new mouse models will also allow us to study meningiomas in the context of the immune system and test the efficacy of drugs in a more physiologic setting. Our results establish a framework for understanding meningioma biology, provide the basis for new meningioma treatments, and lay the foundation for future meningioma studies.

Although DNA methylation is commonly found in the bodies of genes, its biological significance is unclear. Using a novel method for high-resolution analyses of the DNA methylation status of CpG islands, we identified and validated many new intragenic CpG islands that were methylated in a tissue-specific manner in normal human tissues, one of which was in SHANK3.

Through a comparative epigenomics approach, we discovered that the tissue-specific DNA methylation levels of the intragenic CpG islands of SHANK3 were evolutionarily conserved in humans and mice. In addition, the expression levels of SHANK3 were tissue-specific and evolutionarily conserved. SHANK3 encodes a structural protein in the postsynaptic densities of excitatory neurons and altered copy number, and presumably aberrant expression levels of SHANK3, underlie the abnormal neurological phenotypes in patients afflicted with 22q13 deletion syndrome and autism spectrum disorders.

The known function of SHANK3, its association with neurological diseases, and the persistence of SHANK3 tissue-specific CpG island methylation and gene expression patterns over 75 million years suggests that the intragenic DNA methylation might have a conserved function in gene expression. Because DNA methylation is known to influence the activity of promoter sequences, we searched for genetic and epigenetic evidence of in vivo promoter activity embedded within SHANK3 in mouse and human tissues. Using this integrated and cross-species approach, we identified two intragenic regions with promoter activity in vitro, are differentially methylated in vivo, and facilitate the tissue-specific transcription of novel and potentially protein-coding SHANK3 transcripts. These correlative data suggest that the intragenic methylation negatively influences promoter activity. Through a series of in vitro and in vivo DNA methylation experiments, we directly demonstrated that methylation of the intragenic promoters precluded expression.

Additionally, we found that 68.2% of intragenic sequences exhibiting tissue-specific DNA methylation identified from a genome-wide screen harbored features of promoters more commonly associated with the 5'-end of genes.

Altogether, these results support a role for intragenic DNA methylation in regulating the activity of alternate, intragenic promoters that may be generally applicable. Importantly, this study also provides insight into the transcriptional regulation of SHANK3, the level of which is critical for normal brain development.

Gliomas are a group of central nervous system tumors that are classified based upon histopathological and molecular features. IDH-mutant low-grade gliomas (LGG) are prognostically favorable tumors and are often treated with temozolomide, but can undergo malignant progression via TMZ-induced hypermutation. Cell cultures that faithfully model this genetically distinct and clinically relevant hypermutated (HM) tumor subgroup are lacking, and are necessary to advance our understanding of HM tumors. We established patient-derived cell culture and xenograft models of HM oligodendroglioma and astrocytoma that faithfully recapitulate the molecular and functional features observed in their tumor of origin. We thoroughly characterized these cell lines and established that they are suitable for studying HM glioma, IDH-driven phenotypes, TERT promoter- and alternative lengthening of telomeres-driven cellular immortality, therapy-driven evolution, and intratumoral heterogeneity. These models have been widely shared with the neuro-oncology community and accelerated research at institutions across the United States.

The molecular underpinnings of invasion, a hallmark of cancer, have been defined in terms of individual mediators but crucial interactions between these mediators remain undefined. In xenograft models and patient specimens, we identified a c-Met/β1 integrin complex that formed during significant invasive oncologic processes: breast cancer metastases and glioblastoma invasive resistance to antiangiogenic VEGF neutralizing antibody, bevacizumab.

Inducing c-Met/β1 complex formation through an engineered inducible heterodimerization system promoted features crucial to overcoming stressors during metastases or antiangiogenic therapy: migration in the primary site, survival under hypoxia, and extravasation out of circulation. c-Met/β1 complex formation was up-regulated by hypoxia, while VEGF binding VEGFR2 sequestered c-Met and β1 integrin, preventing their binding.

Complex formation promoted ligand-independent receptor activation, with integrin-linked kinase phosphorylating c-Met and crystallography revealing the c-Met/β1 complex to maintain the high-affinity β1 integrin conformation. Site-directed mutagenesis verified the necessity for c-Met/β1 binding of amino acids predicted by crystallography to mediate their extracellular interaction. Far-Western blotting and sequential immunoprecipitation revealed that c-Met displaced α5 integrin from β1 integrin, creating a complex with much greater affinity for fibronectin (FN) than α5β1. Thus, tumor cells adapt to microenvironmental stressors induced by metastases or bevacizumab by co-opting receptors, which normally promote both cell migration modes: chemotaxis, movement toward concentrations of environmental chemoattractants, and haptotaxis, movement controlled by the relative strengths of peripheral adhesions. Tumor cells then redirect these receptors away from their conventional binding partners, forming a powerful structural c-Met/β1 complex whose ligand-independent cross-activation and robust affinity for FN drive invasive oncologic processes.