Direct Regulation of Estrogen Receptor Transcriptional Activity by NF1
Germline mutation in NF1 causes Neurofibromatosis type 1, a common inherited disorder that predisposes individuals to both benign and malignant tumors of the nervous system1, as well as an increased risk for breast cancer. Analysis of TCGA data has shown that NF1 is mutated in a wide range of common cancers (e.g., melanoma, lymphoma, and cancers of the lung, breast, and colon). In this regard, NF1-deficiency underlies the formation and/or progression of a large number of cancers, as well as many neurological disorders, and the development of therapies targeted to NF1-deficient malignancies would have broad impact. NF1 loss induces tamoxifen resistance in ER+ breast cancer. Approximately 2/3 of all breast cancers (150,000 diagnosed annually) are dependent on estrogen receptor-α (ER) to grow. Although great strides have been made in targeting the ER pathway for treating this type of tumor, relapse and death is common and is closely linked to resistance to ER-targeting agents. As a result, the majority of deaths from breast cancer still come from ER+ tumors.
To discover drivers for endocrine resistance, we have sequenced tumor DNAs from a cohort of >600 patients treated by 5-year tamoxifen monotherapy with a median 10.4 years follow up. This is a valuable cohort for detecting genes involved in tamoxifen resistance because it is not confounded by simultaneous chemotherapies. It also has a long follow up of over 10 years (as compared to <4 years in TCGA), which is important because relapse risk for ER+ tumors extends over a very long period. Our preliminary data show that poor outcomes were selectively associated with NF1 mutations that caused protein truncations. We note that NF1 was also identified in an unbiased in vitro siRNA screen for drivers of tamoxifen resistance. In addition, in a previous study comparing mutations in primary versus metastatic breast cancers, NF1 loss was undetectable in primary but more common in metastatic breast cancer. This project centers on the NF1 (Neurofibromin 1) gene whose loss is proposed to induce aggressive tumor behavior (e.g., resistance to tamoxifen, a very common form of endocrine therapy) and relapse. As such, successful completion of the proposed study will address one overarching challenge “Identify what drives breast cancer growth; determine how to stop it.”
Drivers of Endocrine Therapy Resistance
The primary goal of my research is to improve breast cancer diagnosis, treatment and survival by precision data science.
Breast cancer is a diverse and heterogeneous disease, hence one therapy will not be uniformly beneficial to all patients. To dissect the molecular level heterogeneity of these tumors, we employ multi-omics and big data analysis on patient’s next generation sequencing data. We implement bioinformatics to understand the role of mutations, structural variations, gene expressions, copy number variations, and genomic aberrations in promoting breast cancer.
In the past, we have shown role of DNA damage repair defects as a new class of endocrine treatment resistance driver. Exploring the cause and effect relationship further, we are trying to understanding the contribution of immune-checkpoint and other immune-tolerance factors in therapeutic resistant and advancement of estrogen-receptor positive disease. This warrants for integration of patient genome, transcriptome and metabolome with their clinical data.
Development of Microscaled Proteomic Profiling to Identify and Target Aberrant Protein Kinases Within Human Breast Tumors
The breast cancer research community has made substantial progress in analyzing the DNA and RNA of cancer patients. However, changes in DNA and RNA that ultimately affect protein activities are often very difficult to predict by analyzing only DNA and/or RNA. The resulting translated protein encoded by such DNA/RNA experiences post-translational modifications that sometimes deviate from their templates. In addition, kinases, an enzymatic classification of proteins, are regarded as the “molecular switches” that can turn cellular pathways on and off.
A common cause of cancer occurs when these molecular switches become defective and remain “on.” These kinases are also highly-susceptible therapeutic candidates based on their well-defined molecular structures to which drugs can be designed to lodge within their catalytic core.
Our team has developed a tool to enrich protein kinases (kinome) with existing protein kinase inhibitors and profile changes of kinases in a particular patient using mass spectrometry technology. Through rigorous optimization, we can now use this innovative technology to analyze human breast tumor samples approximately the size of a grain of rice. This micro-scaling of our technology positions us within striking distance of surveying actual tumor biopsies for aberrant protein kinases that may be responsible for driving each tumor; thereby, exposing the faulty molecular switch for precision therapeutics. For instance, when comparing kinase networks across breast cancer subtypes, we have identified a key molecule that can drive the low expression of claudin which defines a very aggressive form of triple negative breast cancer that is currently without suitable targeted therapy. Thus, our kinome profiling approach may shed light on how to treat this as well other forms of breast cancer.
Non-Canonical HER2 Activation in Human Cancer
Investigating the impact of HER-targeted drugs on HER2/EGFR non-amplified solid tumors
Targeting HER family members is one of the greatest successes in oncology. It is, therefore, essential to fully identify all cancers that are driven by these oncogenes in order to take full advantage of the wide array of HER targeting agents available. Until recently, HER-targeted therapy is only effective in HER2/EGFR-amplified cancers. However, using genome-sequencing approaches we showed that HER2-mutation status can predict response to HER-targeted therapy. We are currently using proteomic approaches to identify novel subsets of breast cancer patients who could benefit from HER-targeted therapy using patient-derived xenografts (PDXs), publicly available breast cancer datasets such as the Cancer Genome Atlas (TCGA), and our breast cancer patient sequencing studies. Additionally, our research focuses on understanding the biological underpinnings of sensitivity and resistant to HER-targeted therapy in breast and colorectal cancers.
To test the impact of tyrosine kinase (TK) mutations in human cancers
The Cancer Genome Atlas (TCGA) have identified several activating HER2 mutations in breast and colorectal cancers. We provided the functional and clinical impact of HER2 mutations in breast and colorectal cancer patients using cancer cell lines, PDX models, and in a clinical trials setting. Genome sequencing of ER+ breast cancer patients we have identified novel tyrosine kinase mutations that drive poor prognosis but our understanding of the biochemistry of these mutations in ER+ breast cancer is extremely limited. We will, therefore, study how specific poor prognosis tyrosine kinase mutations, discovered in our sequencing analysis, reprogram the kinome as a prelude to developing an effective therapeutic approach.
ESR1 Gene Fusion-Induced Therapeutic Resistance and Metastasis in Estrogen Receptor Positive Breast Cancer
Gene fusions resulting from chromosomal rearrangements have been recognized as key drivers of oncogenesis. Emerging evidence now indicates chromosomal rearrangements involving estrogen receptor alpha gene (ESR1), are able to drive endocrine therapy resistance in estrogen receptor positive (ER+) breast cancer.
Our study characterizes the role of understudied ESR1 translocation events identified from advanced ER+ breast cancer in driving not only endocrine therapy resistance but also metastasis, highlighting how these two lethal processes can be linked together. We propose two unbiased proteomic approaches to elucidate active ESR1 interacting proteins and ESR1 fusion-induced downstream kinases underlying transcriptional activation, growth, and metastatic biology.
DPYSL3 Modulates Mitosis, Migration and Epithelial-to-Mesenchymal Transition in Claudin-Low Breast Cancer
Claudin-low breast cancer is an intrinsic subtype of breast cancer that is reportedly associated with poor survival. A Clinical Proteomic Tumor Analysis Consortium (CPTAC) proteogenomic analysis prioritized dihydropyrimidinase-like-3 (DPYSL3) as a multilevel (RNA/protein/phosphoprotein) expression outlier specific to the claudin-low subset of triple-negative breast cancers.
Our paper (Matsunuma et al., 2018) demonstrated that DPYSL3 knockdown in DPYSL3-positive CLOW cell lines lead to reduced proliferation, yet enhanced motility and increased expression of epithelial-to-mesenchymal transition (EMT) markers, suggesting that DPYSL3 is a multifunctional signaling modulator. Slower proliferation in DPYSL3-negative CLOW cells was associated with accumulation of multinucleated cells, indicating a mitotic defect that was associated with a collapse of the vimentin microfilament network and increased vimentin phosphorylation.
Therefore, we will focus on the multifunctional roles of DPYSL3 in claudin-low breast cancers and explore the underlying mechanisms by which DPYSL3 modulates different aspects of claudin-low breast cancer biology. Specifically, we will study how DPYSL3 interplays with cytoskeletal proteins during mitotic process and identify critical targets for DPYSL3-modulated mitosis and proliferation in claudin-low breast cancers. These results will shed the light on how to treat claudin-low breast cancers with targeted therapies.