Peter M. Glazer, MD, PhD

Tumor hypoxia, DNA repair, and cancer therapy: Our work established that hypoxia is a key driver of genetic instability in solid tumors. We have shown that tumor hypoxia causes down-regulation of specific DNA repair genes, particularly the homology-dependent repair factors, BRCA1 and RAD51. This down-regulation of DNA repair in hypoxic cancer cells renders them vulnerable to therapeutic strategies that exploit the specific repair deficiencies, providing the basis for novel, rationally designed cancer therapies. We also discovered that the anti-angiogenic agent, cediranib, not only damages tumors by interrupting their blood supply and thereby inducing hypoxia but also down-regulates DNA repair (Kaplan, Science Translational Medicine), sensitizing cancer cells to PARP inhibitors and highlighting a therapeutic strategy that is currently being pursued in clinical trials.

Oncometabolites and DNA repair. In collaborative work with the Bindra lab published in Science Translational Medicine, Nature Genetics, and Nature, we found that elevated levels of the metabolites, 2-hydroxyglutarate, fumarate, and succinate, generated in human malignancies by neomorphic IDH mutations or by inherited mutations in the fumarate hydratase or succinate dehydrogenase genes, also suppress DNA repair via dysregulation of chromatin signaling and induce PARP inhibitor sensitivity, providing a new approach to treat these cancers that is being tested in current clinical trials.

DNA repair inhibitors for cancer therapy. We have made the unexpected finding that a cell-penetrating lupus autoantibody, 3E10, has potential as a targeted therapy for DNA-repair deficient malignancies (Hansen, Science Translational Medicine 2012). We found that 3E10 is synthetically lethal to BRCA2- or PTEN-deficient human cancer cells and sensitizes such cells to radiation and doxorubicin. Mechanistically we found that the 3E10 antibody binds to and inhibits the DNA repair factor, RAD51 (Turchick, NAR, 2019), providing a basis for its effects.

Tumor specific targeting of microRNAs and other undruggable targets. We have examined microRNA regulation of DNA repair, identifying several key microRNAs that mediate the stress response to hypoxia and to radiation. This work led to a collaborative effort to target peptide nucleic acids (PNAs) to tumors in mice using pH sensitive peptides as a means to inhibit oncogenic microRNA pathways (Cheng, Nature, 2015). We recently applied this approach to deliver antisense PNAs to inhibit the otherwise undruggable DNA repair factor, Ku80, to sensitize tumors to radiation (Kaplan, Molecular Cancer Research, 2020).

Gene editing via triple helix formation: From an interest in studying cellular DNA repair and recombination pathways, we recognized the utility of DNA triple helix formation as a mechanism for the site-specific induction of gene editing in human cells. We are focusing on triplex forming peptide nucleic acids (PNAs), delivered via polymer nanoparticles, as tools to mediate targeted modification of human disease-related genes. In collaborative work with the Saltzman, Egan, Gallagher, and Kumar labs, we have been optimizing this approach for application in human hematopoietic stem cells and in mouse models of human genetic diseases (Bahal, Nature Communications 2016, Ricciardi, Nature Communications 2018).

Novel pathways that regulate the DNA damage response. Recently, in collaborative work with the Bindra lab, we discovered that the oncometabolite, 2-hydroxyglutarate, generated by neomorphic IDH mutations in gliomas and other maligancies suppresses homologous recombination and confers PARP inhibitor sensitivity, identifying a new approach to treat these malignancies (Sulkowski, Science Translational Medicine, 2017). Similarly, we found that elevated level of the Krebs cycle intermediates, fumarate and succinate, associated with the hereditary cancer syndromes, Hereditary Leiomyomatosis and Renal Cell Cancer (HLRCC) and Succinate Dehydrogenase-related Hereditary Paraganglioma and Pheochromocytoma (SDH PGL/PCC), also suppress the homologous recombination pathway, rendering these tumors vulnerable to synthetic lethal targeting with PARP inhibitors, pointing to a new therapeutic approach for advanced HLRCC and SDH PGL/PCC, both incurable when metastatic (Sulkowski, Nature Genetics, 2018). The use of PARP inhibitors in these malignancies is currently being tested in several clinical trials directly based on our work. Mechanistically, we determined that oncometabolites suppress homology dependent repair (HDR) by inhibiting the histone lysine demethylase, KDM4B. This causes aberrant hypermethylation of H3K9 across the genome and thereby disrupts the normal temporal and spatial pattern of HDR factor recruitment to sites of DNA DSBs (Sulkowski, Nature, 2020). We also recently characterized a novel pathway by which mitochondrial DNA damage mediates signaling to upregulate nuclear DNA repair (Wu, Nature Metabolism, 2019).

Novel approaches to cancer therapy. Effective cancer treatment depends on achieving a therapeutic window in which there is greater toxicity to the malignant cells than to normal, healthy tissue. We have sought to exploit synthetic lethality by seeking agents that are more toxic to cancer cells deficient in homology-dependent DNA repair. This applies to cancers with genetic defects in genes such as BRCA1, BRCA2, PALB2, and PTEN. But it also applies to hypoxic cancer cells in which there is down-regulation of the HDR genes, BRCA1 and RAD51. Our efforts in this area have focused on a novel lupus-derived, cell-penetrating antibody that functions as a DNA repair inhibitor to radiosensitize cells (Hansen, Science Translational Medicine 2012) and a promising class of natural products (Colis, Nature Chemistry 2014). We have also sought to exploit the acidic tumor microenvironment using a novel pH-dependent trans-membrane delivery peptide (pHLIP) to introduce anti-microRNA peptide nucleic acids (PNAs) antisense agents to disrupt oncomiR addiction (Cheng, Nature 2015). We recently applied this approach to deliver antisense PNAs to inhibit the otherwise undruggable DNA repair factor, Ku80, to sensitize tumors to radiation (Kaplan, Molecular Cancer Research, 2020). Recently, we have shown that the anti-angiogenic agent, cediranib, not only damages tumors by interrupting their blood supply and thereby inducing hypoxia but also directly down-regulates DNA repair, sensitizing cancer cells to PARP inhibitors and suggesting a strategy for targeted treatment that is currently being pursued in several clinical trials at Yale and elsewhere (Kaplan, Science Translational Medicine, 2019).

Hypoxia causes genetic instability, down-regulates DNA repair, and promotes gene silencing. In the mid 1990's, we developed the hypothesis that the hypoxic tumor microenvironment could be a cause of genetic instability in cancer. This was contrary to the dogma at the time, because it was thought that under hypoxia there would be less oxidative damage to DNA. We showed that growth of cancer cells in vivo in tumor xenografts produces an elevated mutation rate compared to growth of the same cells in culture (Reynolds, Cancer Res. 1996), and we went on to demonstrate that this effect could be attributed to hypoxia. Since then, we have systematically dissected the mechanisms underlying this effect. These mechanisms include transcriptional downregulation of the homology-dependent repair genes RAD51 and BRCA1 (Bindra, Cancer Res. 2006), of the Fanconi pathway gene FANCD2(Scanlon, Molecular Cancer Res. 2014) and of the mismatch repair gene, MLH1 (Mihaylova, Mol. Cell. Biol. 2003) plus induction of microRNAs 210 and 373 that suppress expression of several DNA repair factors (Crosby, Cancer Res. 2009). In addition, besides acute transcriptional downregulation, our laboratory discovered that hypoxia promotes silencing of the DNA repair genes, BRCA1 and MLH1, in a pathway that is dependent on the histone lysine demethylase, LSD-1 (Lu, Mol. Cell. Biol. 2011 and Lu, Cell Reports 2014).

Triplex DNA provokes DNA repair, and triplex-forming oligonucleotides can stimulate site-specific gene editing in vivo. In the early 1990's, triplex-forming oligonucleotides (TFOs) were being touted as tools to suppress gene expression by binding to promoter sites to block transcription factor access. However, I realized that the site-specific binding properties of TFOs could also be used to mediate sequence-specific gene editing. We discovered that triplex formation, itself, constitutes a helical alteration sufficient to induce DNA repair at the site of the triplex (Wang, Science 1996 and Vasquez, Science 2000), via the nucleotide excision repair (NER) pathway. This activates the target site for recombination with "donor DNAs" via homology-dependent repair. After a systematic evaluation of DNA analogs for improved triple helix formation in cells, we have focused on peptide nucleic acids (PNAs), which have a neutral polyamide backbone and bind DNA with high affinity. Using PNAs, we demonstrated successful editing of the beta-globin gene in human primary hematopoietic stem cells. In collaboration with the Saltzman lab, we developed a strategy to encapsulate the PNAs and donor DNAs in polymer-based, biocompatible nanoparticles to achieve effective in vivo delivery in mice with minimal toxicity. Recent work with the Egan lab has demonstrated the ability of nanoparticles containing PNAs and donor DNAs to mediate editing of the F508del CFTR gene mutation in airway epithelia in vivo in a mouse model of cystic fibrosis (McNeer, Nature Communications 2015) and to mediate substantial correction of anemia and in mice with thalassemia by simple intravenous injection of PNA and DNA containing nanoparticles in adult mice (Bahal, Nature Communications 2016) and in fetal mice via in utero injection (Ricciardi, Nature Communications, 2018).