Using theoretical and applied biophysical methods, we discovered how a patient mutation in a class of glutamate receptors responsible for excitatory neurotransmission is affected at the atomic level. This multifaceted approach provides the necessary fine detail for improving the therapeutic landscape for patients affected by this mutation. This work was done in the lab of Dr. Edward Twomey and in close collaboration with the lab of Dr. Albert Lau.

During the past two years, I have been working in Dr. Jennifer Pluznick’s lab investigating the mechanism of a specific receptor and sex differences in blood pressure. Our lab has found that when this receptor is absent in the kidney, anticipated sex differences in blood pressure are gone. My project has found that two sex hormones, luteinizing and follicle stimulating hormones, do not activate this receptor like hypothesized. The other part of this project has been seeing if there are differences in proteins between male and female mice that have or lack this receptor. Currently, I am still analyzing this data and trying to understand what proteins are worth further investigating and may explain the phenotype we see. This research is important because men and women have the same diagnostic measure for hypertension even though women incur significant cardiovascular risks at lower blood pressures than men. Better understanding of the sex differences in blood pressure would support the need for better sex-specific diagnostic measures and treatments in the clinic.

My research uncovered the molecular mechanism by which the brain’s major neurotransmitter, glutamate,  activates its principal receptor, the AMPA-subtype ionotropic glutamate receptor (iGluR). I demonstrated how both physiological and hyperthermic temperatures affect iGluRs. Using cryo-EM, I captured the first glutamate-activated structure of an iGluR. This discovery provides valuable insights into the molecular details of excitatory neurotransmission and offers a foundation for designing drugs targeting neurological disorders. This work was primarily conducted in Dr. Edward Twomey’s lab, in collaboration with Professor Vasanthi Jayaraman’s lab at the University of Texas Health Science Center.

In Dr. Hiromi Sesaki’s lab, I study how mitochondrial fission and fusion are essential to animal physiology. We discovered that the synergy between two stress-responsive systems — the ubiquitin E3 ligase Parkin and the metalloprotease OMA1 — safeguards mitochondrial structure and genome by regulating mitochondrial fusion. Whereas the individual loss of Parkin or OMA1 does not affect mitochondrial integrity, their combined loss results in small body size, low locomotor activity, premature death, mitochondrial abnormalities and innate immune responses. These findings account for previous puzzling reports that the individual loss of Parkin–PINK1, whose defects are known to cause Parkinson’s disease, shows minimal mitochondrial phenotypes in mice.

Diabetic retinopathy (DR) is the leading cause of vision loss among working-age people in the developed world. Currently, optimizing glucose management is the cornerstone for preventing diabetic eye disease. However, tight control of serum glucose concentration (TGC) has been associated with an early, paradoxical worsening of DR. Until recently, why TGC promotes an early worsening of DR has not been fully understood. One of the challenges of TGC is preventing hyperglycemic episodes without promoting increased episodes of hypoglycemia. We therefore proposed examining how the neurosensory retina responds to acute episodes of hypoglycemia, and how these responses may contribute to the development and/or progression of diabetic eye disease. In several cell-based, ex vivo and animal models, we observed that transient episodes of hypoglycemia resulted in HIF-1α accumulation and HIF- 1-dependent expression of angiogenic genes. In mice diabetic for as early as three months, prior to overt microvascular injury, hypoglycemia accelerated a synergistic increase in the accumulation of HIF-1α and the expression of HIF-regulated angiogenic factors. This, in turn, was sufficient in diabetic mice — but not control mice — to promote breakdown of the inner blood- retinal barrier (iBRB) and increased vascular permeability. Collectively,  these observations help explain why patients with diabetes-initiating TGC have worsening of their DR. While TGC remains essential for preventing microvascular injury in patients with diabetes, our findings suggest that this should not be achieved at the expense of transient episodes of hypoglycemia. My whole work was conducted in Dr. Sodhi’s lab.

The high mortality rate associated with cancer is, in part, due to diagnosis at late-stage disease. Unfortunately, there are currently no available screening tests for most cancer types. For those with screening modalities, several challenges have emerged, such as false positives and negatives and low adherence. In healthy individuals, DNA is released in the blood, which we refer to as cell-free DNA. When an individual has cancer, tumor cells also release cell-free DNA into the blood, termed circulating tumor DNA, providing a potential blood-based avenue for early cancer detection. To this end, I developed a prototype blood test for screening for multiple cancer types that requires a single tube of blood. The approach uses cost-efficient sequencing of cell-free DNA and machine learning of mutation frequencies across the genome to predict whether an individual has cancer. As a proof-of-concept demonstration of the approach, we detected cancers of the lung, liver, skin and white blood cells. This work was done in the Cancer Genomics Lab and was jointly supervised by Victor Velculescu and Rob Scharpf.

Our study defines a novel germline cancer predisposition syndrome, the long telomere syndrome, with a mechanism distinct from traditional classifications of tumor suppressors and oncogenes. We found that  individuals with heterozygous loss-of-function variants in POT1, which encodes telomere binding protein POT1, had extremely long telomere length and a pan-tissue risk for clonality and malignancy. Typically, telomere shortening with cell division acts as a type of tumor suppressor, limiting the expansion of clones with advantageous mutations. However, individuals with germline telomere lengthening variants have a capacity to support the expansion of these clones, allowing increased numbers of somatic mutations to be acquired with a standard mutational rate, thus leading to increased risk for neoplasia. This offers insight into an extensive literature of population-level association of telomere length with clonality and malignancy. It also provides a cautionary tale against the manipulation of telomere length in pursuit of the extension of human lifespan. This work was done in the lab of Dr. Mary Armanios.

My research established a detailed understanding of how E3 and E2 enzymes, Bre1 and Rad6, cooperatively mediate histone H2B ubiquitination, and provided insights into how cancer-associated mutations in human Bre1, Rad6 and the nucleosome disrupt this process, potentially driving cancer development. Furthermore, these findings facilitate the discovery of potential drug targets and the rational design of therapeutics for disease treatment. I did this project in Cynthia Wolberger’s Lab.

I did this research in the Green lab, where we study ribosomes, the molecular machines that carry out protein synthesis. Cells regulate ribosome levels to support robust protein synthesis in favorable conditions and preserve resources in stress conditions. Mammalian cells degrade intact ribosomes in response to stress, but the mechanisms have been unclear. We found that during amino acid starvation, 40S ribosomal subunits are marked with ubiquitin on specific ribosomal proteins. The atypical kinase RIOK3 then specifically binds these ubiquitylated 40S ribosomes and induces their degradation by triggering  decay of their ribosomal RNA. Together, our work describes a pathway that regulates ribosome levels during stress by connecting ribosome ubiquitylation to downstream degradation of the ribosomal RNA itself.

I discovered that a specialized branch of the neuron has a different shape than has been assumed for the past 120 years. Rather than looking like a small tube, the small neuron branch called the axon looks like a pearl necklace. I found that this shape easily and dynamically adapts to signals, such as those commonly associated with forming new memories. This shape has been overlooked because many parts of neurons are so small that they cannot be studied under traditional microscopes, so in the Watanabe Lab, I used a special technique to zoom in further using electron microscopy. With this new understanding of neuron branch shape, we can better study how our brain  and nervous system function in our everyday lives.

The ribosome is a macromolecular complex that is essential for life. My work provides deeper understanding of how mammalian cells regulate the synthesis of ribosomes through a key protein called LARP1. My work also identified a physical interaction between LARP1 and ribosomes that enables LARP1 to sense ribosome concentration and titrate its own protein expression accordingly. This work was done in the lab of Rachel Green in the Department of Molecular Biology and Genetics.

Mitochondrial fatty acid oxidation is facilitated by the combined activities of Carnitine Palmitoyltransferase 1 (Cpt1) and Cpt2, which generate and utilize acylcarnitines, respectively. We compared the response of mice with liver-specific deficiencies in the liver-enriched Cpt1a or the ubiquitously expressed Cpt2 and discovered that they display unique metabolic, physiological and molecular phenotypes. The loss of Cpt1a or Cpt2 results in the induction of the muscle- enriched isoenzyme Cpt1b in hepatocytes in a Ppara-dependent manner. However, hepatic Cpt1b does not contribute substantively to hepatic fatty acid oxidation when Cpt1a is absent. Both the liver-specific double knockout of Cpt1a and Cpt1b, or Cpt2 eliminates the mitochondrial oxidation of nonesterified fatty acids. However, Cpt1a/Cpt1b double-knockout mice retain fatty acid oxidation by utilizing extracellular long-chain acylcarnitines that are dependent on Cpt2. These data demonstrate the noncell-autonomous intercellular metabolism of fatty acids in hepatocytes. Much of the work done targeting mitochondrial ß-oxidation targets Cpt1, but this work shows that there are other metabolites that can feed into the same long-chain fatty acid metabolism. This work was done in Dr. Michael Wolfgang’s lab.

Immunotherapy (IO) has made great strides in combating metastatic non-small cell lung cancer (NSCLC), but patients with tumors that elicit weak immune responses (“immunogenically cold”) often derive little benefit from IO. We investigated the hypothesis that radiation followed by immunotherapy would stimulate greater systemic immune responses than immunotherapy alone in patients with immunogenically cold metastatic NSCLC. By studying pre- and post- treatment tumor biopsies from non-irradiated sites and peripheral blood samples, we confirmed this hypothesis. The increased systemic immune upregulation observed in patients receiving radio-immunotherapy also correlated with improved clinical response. This suggests that radio- immunotherapy is a promising avenue of treatment for these patients, and that clinical trials should investigate the use of pre-IO radiation in patients with immunogenically cold NSCLC. This research was performed in the lab of Dr. Valsamo Anagnostou.

Endosomal ion homeostasis is critical for the proper function of many cells, including neurons of the central nervous system. While many neurological disorders are linked to defects in endolysosomal processing, many molecular mechanisms are still enigmatic. In Dr. Zhaozhu Qiu’s lab, I studied the proton-activated chloride (PAC) channel and its role in synaptic plasticity, which is thought to be the molecular basis of learning and memory. We discovered that PAC contributes to synaptic long-term depression (LTD), a form of plasticity that weakens synaptic connections and is important for animal physiology and behavior. This research provides evidence for a previously unexplored mechanism for synaptic plasticity along with novel approaches to study LTD.

Understanding how the brain links internal states — like sleepiness or alertness — to behavior is a central question in neuroscience. While scientists have long known that small movements, or microbehaviors, can reflect changes in an animal’s internal state, these subtle actions are hard to measure with existing tools. To address this, I developed FlyVISTA in Dr. Mark Wu’s lab — a machine-learning system that captures fine-scale behavior in freely moving fruit flies using high-resolution video and deep learning to track 35 body parts. Using this platform, I discovered a previously unknown behavior called the haltere switch, a small movement that only occurs during deep sleep in flies. This suggests that, like mammals, flies also have different stages of sleep — one of which might be comparable to REM sleep. By revealing these detailed behavioral signatures of sleep and linking them to physiology (such as muscle atonia), FlyVISTA enables rigorous, quantitative analysis of behavior at a scale not previously possible. This discovery illustrates that fruit flies can serve as  a powerful model to study the fine structure of sleep, helping to uncover its biological functions in a genetically tractable system.

It has been known that some cells can communicate by exchanging materials through tiny bridge-like nanotubes, but whether neurons use such connections in the brain remained unknown. In this study, conducted in the lab of Dr. Hyungbae Kwon in the Department of Neuroscience, I discovered a new type of neuronal connection called dendritic nanotubes (DNTs),  which form direct links between dendrites. Using advanced imaging and machine-learning analysis, we identified DNTs in mouse brains and found  that they transport calcium and small molecules, including amyloid-beta  (Aß), a key protein in Alzheimer’s disease. Remarkably, DNT levels increased before amyloid plaques appeared in a mouse model of Alzheimer’s, with computational simulations supporting their role in early neurodegeneration. Our findings reveal a previously unrecognized neuronal communication network in the brain, providing new insights into how neurons interact beyond synapses and potentially opening avenues for understanding and treating neurodegenerative disorders.

In the lab of Joshua Modell, I study the adaptive immune systems of bacteria known collectively as CRISPR-Cas systems. CRISPR-Cas systems generate immunological memories of infecting bacterial viruses to provide defense when  these threats are encountered again. We discovered that bacterial CRISPR-Cas systems exploit a unique portion of an infecting virus’s life cycle, a dormant state known as lysogeny, to establish and carry out the steps required for adaptive immunity against these viruses. Our discoveries shed light on the enigmatic mechanisms by which bacteria create and use immunological memories to defend themselves against viruses and other threats.

The mechanotransduction (MET) channel in the hair cells can convert sound-induced vibration into electrical signal, and dysfunction of this MET channel will cause hearing loss. This MET channel is a large molecular machinery that contains potential pore subunit proteins, such as TMC1 and TMIE. The mechanisms by which these proteins are regulated by mechanical forces remain unclear. This ion channel complex is expressed throughout the animal kingdom, ranging from invertebrates to vertebrates, and it has fundamental roles in the perception of not only sound but also other mechanical signals, such as those provided by food texture or touch. Unfortunately, all efforts have so far failed to determine the structure of the mammalian ion channel complex at atomic resolution. However, the structure of this ion channel complex from C. elegans has been solved and provides a blueprint to study the mammalian channel complex. Studies in C. elegans suggest that nematode TMC proteins are components of several ion channel complexes (the mechanosensitive ion channel, alkaline pH-activated channel and Na+-leak channel) with distinct functions. The mechanisms by which nematode TMCs contribute to such diverse physiological processes and their functional relationship to mammalian mTMCs is unclear. We show that association with accessory proteins tunes nematode TMC-1 to divergent sensory functions. In addition, different protein domains in nematode TMC-1 are required for different sensory modalities, and these protein domains have been segregated during evolution in mammalian TMC proteins into distinct proteins. Our findings demonstrate that sequence diversification and association with accessory proteins has led to the emergence of TMC protein complexes with diverse properties and physiological functions. This work was done in Dr. Ulrich Mueller’s lab.

Parkinson’s disease is characterized by the progressive loss of dopamine-producing neurons, partly due to the toxic buildup and aggregation of the α-synuclein (SNCA) protein. Overproduction of SNCA is one of the greatest risk factors for Parkinson’s; therefore, reducing SNCA levels represents a promising therapeutic strategy. In the McCallion lab, we discovered a genetic “switch” (or enhancer) that helps control SNCA levels in this vulnerable population of dopamine-producing neurons. When this switch is active, SNCA is produced at high levels, but when the enhancer element is removed, SNCA production is diminished. To test whether turning off this switch could be protective, we created mice in which this enhancer element was removed. These mice exhibited a reduction in SNCA levels and, as a result, were resistant to Parkinson’s-like motor symptoms and had healthier neurons with significantly less toxic protein buildup, inflammation and neurodegeneration. Current therapeutic interventions for Parkinson’s mostly help manage disease symptoms, but don’t stop or slow neuron death. My findings suggest that targeting this genetic switch could slow or even stop disease progression, offering a potential gene therapy approach for Parkinson’s and other disorders linked to SNCA, such as dementia with Lewy bodies.

As a postdoctoral fellow in Dr. Shuying Sun’s  lab in the Department of Physiology, I investigated how RNA repeats regulate RNP granule homeostasis and contribute to neurodegeneration. My research focuses on the (GGGGCC)n repeat expansion in C9ORF72, the leading genetic cause of frontotemporal dementia and amyotrophic lateral sclerosis (C9-FTD/ALS). I have demonstrated that repeat RNA disrupts nuclear speckle dynamics, sequestering SRRM2 into cytoplasmic poly-GR inclusions, which leads to RNA splicing defects and neuronal toxicity. These findings uncover a novel mechanism linking nuclear speckle dysfunction to RNA misprocessing in C9- FTD/ALS, providing insights into potential biomarkers and therapeutic targets.

Stroke is caused by a lack of oxygen delivery to the brain due to blood vessel blockage, leading to neuronal death and long-term disability. Current lifesaving therapies focus on restoring blood to the brain; however, they do not address the consequence of brain injury in stroke. Specifically, brain inflammation following ischemic stroke is known to exacerbate neuronal injury and long- term functional outcomes. Early infiltration of immune cells into the brain after ischemia is correlated with increased risk of subsequent stroke, post- stroke depression and higher three-month mortality rates in stroke patients. Despite decades of awareness of neurogenic inflammation in stroke, clear inciting events in this inflammatory cascade remain unclear, preventing the development of successful therapeutic interventions. This study sought to identify inciting events that drive brain inflammation after stroke. We show that a mast cell receptor is activated early after stroke injury and mediates immune cell infiltration into the brain. Collectively, our study identifies the Mrgprb2 mast cell receptor as a critical meningeal gatekeeper for immune cell migration from bone marrow reservoirs into the brain, and it provides a specific and druggable target to attenuate post-stroke brain inflammation.

We developed a molecular blood test to identify and track acute spinal cord injury (SCI). This will allow for more rapid diagnosis and precise treatment of patients with SCI. This work was conducted with the guidance of Dr. Chetan Bettegowda and Dr. Nicholas Theodore.

T cells can be engineered to target and kill cancer cells via synthetic proteins called chimeric antigen receptors (CARs). In the clinic, CAR T cells are targeted to single antigens overexpressed on cancer, but also kill normal cells because those antigens are not entirely cancer specific. For some blood cancers, the resulting normal toxicity is manageable, but this is not the case for most other cancers. During my Ph.D. in the lab of Dr. Kenneth Kinzler and Dr. Bert Vogelstein at The Ludwig Center at Johns Hopkins, I focused on engineering new synthetic receptors that enable CAR T cells to kill cancer cells based on combinations of antigens with Boolean logic. These new receptors allow engineered T cells to more precisely kill cancer cells and spare normal cells, yielding a safe and more effective cancer therapy.

Ras GTPases play a critical role in cell proliferation and differentiation and are mutated in approximately 30% of cancers. In the laboratory of Peter N. Devreotes, I investigate Ras regulation in cell migration. As Ras is activated at the leading edge of many migrating cells, my collaborator and I explored whether RasGAPs, natural Ras inhibitors, could halt migration and potentially prevent cancer metastasis. While RasGAP activation effectively suppressed Ras signaling, it unexpectedly induced cell polarization. This finding has important implications that directly inhibiting Ras may unintentionally enhance polarization and promote metastasis, highlighting careful considerations for future drug design targeting oncogenic Ras.

I joined Dr. Zhaozhu Qiu’s lab for postdoctoral training in September 2019. I’m interested in studying the role of epigenetic factors in neurological diseases. Intellectual disability (ID) affects approximately 2% of the  population, and ID-associated genes are enriched for epigenetic factors, including those encoding the largest family of histone lysine acetyltransferases (KAT5-KAT8). Among them is KAT6A, whose mutations cause KAT6A syndrome, with ID as a common clinical feature. However, the underlying molecular mechanism remains unknown. Our results demonstrate that KAT6A-RSPO2-Wnt axis plays a critical role in regulating hippocampal CA3 synaptic plasticity and cognitive function, providing potential therapeutic targets for KAT6A syndrome and related neurodevelopmental diseases.

To coordinate billions of electrical impulses that must arrive with millisecond precision, the brain employs myelin — fatty insulation that modulates the speed of signal transmission. Loss of myelin, and associated temporal control, is thus commonly found in neurodegenerative diseases, leading to impaired learning, disrupted motor control and seizures. Myelin is particularly interesting as it is produced throughout life, unlike most other substrates in the brain, meaning that the entire myelin landscape can shift to support learning and promote recovery from injury. In the Bergles laboratory, I sought to map this incredibly diverse and dynamic myelin landscape at the brain-wide level, using whole-brain imaging and deep-learning tools. Altogether, these interrogations revealed unique differences in vulnerability and regenerative potential across the brain, while also highlighting enhanced plasticity in the prefrontal cortex and hippocampus in advanced age. These insights are critical to understanding the dynamics of myelin patterning throughout life, while also providing a robust platform to assess the mechanisms that maintain and disrupt the myelin landscape in health and disease.

During my postdoc in the Holland lab at the Department of Molecular Biology and Genetics, I uncovered how the protein TRIM37 controls key structures involved in cell division by sensing when they form large assemblies  and triggering their breakdown. How a single enzyme could selectively recognize and act on such large assemblies remained unclear. My work elucidated the molecular steps that activate TRIM37 — beginning with substrate recognition, followed by enzymatic activation and culminating in degradation. This mechanism ensures cells divide with high fidelity and helps explain how TRIM37 mutations lead to the Mulibrey nanism disorder, as well as why certain cancers are vulnerable to therapeutic strategies aimed at targeting these structures. More broadly, these findings reveal a general principle for how the TRIM protein family regulates large cellular assemblies, providing insight into their diverse roles in human health and disease.