I study how cells can adapt to their physical surroundings, and I discovered how modifications on specific cellular components called microtubules enable cells to modulate their mechanical properties and migration behavior. I performed my research in the lab of Dr. Takanari Inoue in the Department of Cell Biology.

Chimeric antigen receptor (CAR) T cells have produced dramatic outcomes in patients with blood cancers, but have had limited success in solid cancers like pancreatic cancer or glioblastoma. One major challenge in solid tumors is the lack of cancer-specific targets that minimize toxicity to normal tissues. The genetic alterations essential for the development of cancer can also be used as immuno-therapeutic targets to specifically kill cancer cells and spare normal tissues. During my Ph.D. with Ken Kinzler and Bert Vogelstein in The Ludwig Center at Johns Hopkins, I developed two types of CAR T cells to target these cancer-specific genetic alterations. The first CAR approach utilized pairs of activating and inhibitory CARs to target the loss of heterozygous genes within cancer cells. In the second approach, components of a CAR were fused with the T cell receptor to target “hot spot” mutations that are presented as peptides on human leukocyte antigens. Both CAR approaches were able to cure mice of cancers containing the targeted genetic alteration without affecting cells representing normal tissues.

My project in the Bergles laboratory focused on identifying how we form connections from an early age between the inner ear and brain to detect and process sound, and if these mechanisms are altered in conditions of early hearing loss. To prepare for sound processing, the immature brain undergoes a training period before any external environmental sounds can be perceived. This training comes from the immature inner ear, which generates “spontaneous” neural activity in precise patterns before the ear canal opens. We discovered that these patterns of spontaneous activity are critical for establishing the sensitivity of the auditory system to sound loudness and for defining normal brain representations of sound pitch, both of which are crucial components of communication with each other and the world around us. Next, we generated a model of the most common genetic form of hearing loss, due to disruption of the gene GJB2, and remarkably found that pre-hearing spontaneous activity was preserved and sufficient to induce circuit maturation despite profound deafness following ear canal opening. The preservation of this highly stereotyped activity may explain why patients with GJB2-mediated hearing loss respond well to cochlear prostheses, motivating further development of early therapeutic interventions to restore function.

Organisms must be able to adapt to changes in nutrient availability in order to maintain normal physiology. The mechanisms involved in these homeostatic responses are of interest in many disease contexts, such as diabetes, cardiovascular disease and cancer. In the Espenshade lab, we are interested in studying lipid metabolism in cancer. Specifically, we are interested in targeting lipid metabolic pathways to treat pancreatic cancer, which historically has very poor patient outcomes. For my project, I designed a CRISPR knockout screen to identify genes that are required for pancreatic tumor growth in mice. Through subsequent follow-up experiments in vitro and in vivo, I demonstrated that the isoprenoid geranylgeranyl diphosphate is essential for tumor growth. These findings validate a method for conducting a genetic screen in mice, and provide evidence that implicates this pathway in pancreatic cancer.

Postsurgical pain causes significant suffering. Continuous reliance on opioid analgesics can lead to severe side effects and accidental death. Therefore, there is an urgent need to develop nonopioid therapies for managing postsurgical pain. We demonstrated that local application of a naturally occurring biologic derived from human birth tissues (amniotic membrane, AM) attenuated established postsurgical hypersensitivity without exhibiting known side effects of opioid use in mice. Importantly, preemptive treatment also prevented postsurgical pain from transitioning to a prolonged state. This effect was achieved through diverse modes of actions, including direct inhibition of nociceptive dorsal root ganglion neurons via CD44-dependent pathways and indirect pain relief by attenuating immune cell recruitment. We further purified the major matrix component, the heavy chain-hyaluronic acid/pentraxin 3 (HC-HA/PTX3) from human AM that has greater purity and water solubility. HC-HA/PTX3 replicated neuronal and pain inhibition. Mechanistically, HC-HA/PTX3 induced cytoskeleton rearrangements to inhibit sodium current and high-voltage activated calcium current on nociceptive neurons, suggesting it is a key bioactive component mediating pain relief. Collectively, our findings highlight the potential of a naturally derived biologic from human birth tissues as an effective nonopioid treatment for postsurgical pain, and unravel the underlying mechanisms. This study was done in Yun Guan’s lab in the Department of Anesthesiology and Critical Care Medicine.

Most cells in our body contain two copies of DNA — one copy from each of our parents. However, many cancer cells undergo a process called whole-genome doubling, through which they acquire approximately twice the amount of DNA of a normal cell. This process is associated with drug resistance and metastasis throughout the body, so it is an important step in cancer progression. In the Regot Lab, my work has focused on understanding how this process happens; first discovering various causes, then identifying what the causes have in common, and finally demonstrating how they cause it. These findings changed our understanding of the wiring of the human cell division cycle, and provide a mechanistic basis for the high incidence of whole-genome doubling in cancer.

Hepatocytes, polarized epithelial cells of the liver, play a crucial role in maintaining systemic glucose and lipid homeostasis in the face of an ever-changing nutritional environment. Hepatocytes are both incredibly metabolically flexible and also have an efficient intracellular trafficking system that allows them to transport cargo across several apical and basolateral domains with high fidelity. Due to this necessity for efficient endocytosis and secretion, the family of Rab GTPases, principal regulators of the intracellular trafficking network, play a particularly important role in hepatocytes. However, despite their importance in vesicle trafficking, the physiological roles of many Rabs in animals have yet to be characterized. Using a combination of mouse models and cell culture studies, my thesis work has defined the role for the Golgi-localized family member Rab30 in liver physiology during fasting. I have uncovered that Rab30 expression is specifically induced by fasting and not by other dietary interventions in the mouse liver. Fasting-induced Rab30 expression is regulated by a master transcriptional regulator of lipid metabolism in the liver called Pparα. Additionally, Rab30 expression is further amplified in liver-specific carnitine palmitoyltransferase 2 knockout mice (Cpt2^L-/-) that lack the ability to oxidize fatty acids and have fasting-induced hepatic steatosis, serum dyslipidemia and a hyper-induction of Pparα transcriptional activity, leading us to the hypothesis that Rab30 contributes to lipid homeostasis. Live-cell super-resolution imaging and biochemical in vivo proximity labeling demonstrate that Rab30-marked vesicles are highly dynamic and interact with proteins at the Golgi apparatus and throughout the secretory pathway. While analysis of liver-specific knockouts of Rab30 reveals its dispensability in the fasting response, analysis of Rab30;Cpt2 double knockout mice, which have a potentiated Pparα response and enable us to amplify the effect of loss of Rab30 in the liver, reveals a retention of proteins within the livers of fasted mice, a reversal of serum dyslipidemia observed in Cpt2^L-/- mice, and a corresponding decrease in serum apolipoprotein A4 levels. Together, these data suggest a role for Rab30 in the sorting of secreted proteins and lipoproteins to influence hepatic and circulating triglyceride levels during fasting, particularly during times of excessive lipid burden.

I did my postdoctoral research with Peter Devreotes in the Department of Cell Biology. Traditionally, research in the Devreotes lab has been focused on chemotaxis using the soil amoeba Dictyostelium as the model organism of choice. Chemotaxis is the directed movement of cells along extracellular gradients, which plays an important role in development and immunity. Although chemotaxis studies of the model system have been invaluable, there has been an urgent need to understand mechanisms of migration in immune cells for the research to be useful to human health and disease. Since studies in immune cell lines are technically more difficult than in the model system, it has discouraged all but a few laboratories from making significant progress in these cellular systems.   When I joined Peter’s lab, I saw this roadblock as an opportunity to make fundamental discoveries in the field whilst honing my technical and analytical skills. To this end, I developed new approaches for powerful optogenetic studies in human neutrophils and macrophages. With these optical tools, I elucidated the role of the Ras/PI3K/Akt pathway in immune cell migration. Ras signaling is typically associated with cell growth, but not direct regulation of motility or polarity. However, my studies demonstrated that local activation of Ras-mediated classical growth-control pathways directly modulate actin polymerization, cell shape and migration modes. Thus, my work provided the first mechanistic description of the role of Ras GTPases and Akt in directly controlling human cell migration. I further continued my pursuit of the role of Ras growth pathways in immune cell migration by noting the effects of locally inhibiting Ras GTPases. Based on my previous results, we expected that suppressing Ras activity would stop migration. But I saw the opposite: Reducing Ras activity on the cell membrane polarized cells and improved their migratory ability by increasing actomyosin contractility at the back. The surprising ability of Ras inhibition to promote migration presents a warning that targeting Ras to inhibit proliferation could have the unanticipated effect of promoting migration and metastasis.

In response to environmental stressors such as nutrient deprivation, bacteria activate a conserved stress response pathway called the stringent response (SR). In addition to stress survival in non-pathogenic contexts, SR activation is implicated in the virulence and antibiotic tolerance of pathogens. In the Goley lab, we use the freshwater bacterium Caulobacter crescentus as a tractable model to study bacterial adaptation to stress. During SR activation, adaptation and survival are promoted over growth and anabolism. In Caulobacter, a major regulator of anabolic genes is the widely conserved transcription factor CdnL. If and how CdnL is controlled during the SR and why that might be functionally important is unclear. Through my thesis research, I have elucidated the mechanisms contributing to downregulation of CdnL during the SR. Preventing CdnL degradation during nutrient deprivation causes misregulation of ribosomal and metabolic genes. Functionally, I found that CdnL clearance allows for efficient adaptation to nutrient repletion, and that cells with the inability to clear CdnL during starvation are outcompeted by wild-type cells when subjected to nutrient fluctuations. These findings indicate that clearance of CdnL during the SR is critical for altering the transcriptome to permit cell survival during nutrient stress. Because CdnL homologs are broadly found and are in important pathogens such as Borrelia burgdorferi and Mycobacterium tuberculosis, I hypothesize that CdnL regulation is a conserved mechanism of stress adaptation across bacteria.

795,000 people suffer from stroke annually in the United States. 87% of these strokes are ischemic. This condition is a leading cause of serious long-term disability and the second leading cause of death worldwide. Current management strategies for ischemic stroke, such as alteplase treatment and mechanical thrombectomy, are focused on reperfusion of the ischemic brain regions. However, neurons continue to die even after these reperfusion therapies are instituted, leading to persistent neurological deficit. No clinically available therapeutics can prevent this phenomenon, limiting the capacity to preserve neurologic function in this patient population. As such, there is a great need to develop therapeutic strategies capable of preventing this secondary form of neurologic injury. Ischemic stroke and Parkinson’s disease share many molecular alterations, including the downregulation of parkin, an E3 ubiquitin ligase with critical neuroprotective functions. Cytosolic substrates of parkin, such as parkin interacting substrate (PARIS), have been implicated as drivers of neurodegeneration in Parkinson’s disease. However, no study to date has examined the role of these substrates in ischemic stroke. My work, under the mentorship of Drs. Ted and Valina Dawson, has been focused on characterizing the role of PARIS in neuronal death following cerebral ischemia. We demonstrated that PARIS becomes upregulated in ischemic stroke through enhanced proteasomal degradation of parkin. Using a PARIS gene knockout mouse model, we discovered that PARIS knockout dramatically reduces infarct volume and neurological deficit  following cerebral ischemia. Finally, we identified that PARIS gene knockdown attenuates both cell autonomous and non-cell autonomous modes of neuronal death using an in vitro model of cerebral ischemia. Together, our results suggest that PARIS is a promising therapeutic target in the context of ischemic stroke.

Regulation of directed axon guidance and branching during development is essential for the generation of neuronal networks. My research in the Kolodkin laboratory aims to uncover molecular principles of interstitial axon branching, fundamental phenomena allowing central nervous system neurons to connect to multiple targets that are spatially distinct. We use novel in vivo single cell labeling approaches that enable sparse and robust visualization of individual cortical excitatory neurons, along with the capacity to perform genetic manipulations. These new techniques allow for quantitative assessment of axonal morphologies at the single cell resolution. We have found that balancing cytoskeletal dynamics through action of microtubule-binding proteins and tubulin posttranslational modifications is an important regulatory element of collateral axon elaboration in cortical projection neurons in vivo. Our data describe one of the very first initial intracellular signaling pathways known to cell-autonomously regulate interstitial axon branching in the developing neocortex

As part of my T32 research fellowship, I joined the cardio-immunology lab of Dr. Luigi Adamo. Despite modern therapies, mortality and morbidity of chronic heart failure remains high. Inflammation plays a critical role in the development and progression of heart failure after a heart attack, but this connection is complex precluding the development of specific immunomodulatory therapies to treat heart failure. We have found that B cells contribute to adverse cardiac remodeling through antigen presentation, which may represent a potential novel therapeutic target in chronic heart failure.

I joined Dr. Jennifer Pluznick’s lab for postdoctoral training in October 2018. My research focused on two different aspects of gut microbiome and host-interaction: (a) elucidating the influence of gut microbes on glomerular filtration rate in health and disease, and (b) uncovering a novel role of a sensory receptor (OLFR558) that responds to gut microbial metabolites in blood pressure control. Males have a higher blood pressure than premenopausal women with ~10 mmHg. In the second project, I found that OLFR558 is required for sex differences in blood pressure. Olfr558 KO females exhibit increased blood pressure whereas KO males have decreased diastolic blood pressure. A rare OR51E1 (human ortholog) missense variant has a statistically significant sex interaction effect with diastolic blood pressure, increasing diastolic blood pressure in women but decreasing it in men. Uncovering the origin of sex differences in blood pressure regulation can help to move the field toward a more thoughtful approach to blood pressure management in both men and women.

Gene regulatory DNA elements, such as enhancers, are enriched with pathogenic mutations associated with devastating diseases such as cancer and schizophrenia. Despite the significant medical implication, functional characterization of individual enhancers has been difficult. For example, many human genes are regulated by complex networks of enhancers, and it remains unclear how these network properties affect phenotypes and how their disruptions cause disease. Further, putative pathogenic regulatory variants are often tested through their conserved counterparts in mice, but mapping human enhancers to mice has long been a computational challenge due to both rapid evolution and sequence complexities of enhancers. In Dr. Michael Beer’s lab, we addressed these challenges in two orthogonal approaches. First, we collaborated with multiple labs to functionally characterize a large number of enhancers near genes of high medical significance using CRISPR. For example, we epigenetically perturbed enhancers that drive stem cell differentiation, and discovered that enhancers can regulate the speed of cell-state transitions. Fetal development is a finely controlled dynamic process with high spatiotemporal precision, and our discovery will help us understand how enhancer mutations may cause developmental disorders. Second, we developed a novel genome-alignment algorithm (gkm-align) that can detect more than 20,000 novel distal enhancers conserved between human and mouse. Using our novel method, we published an expanded catalogue of conserved enhancers, which we believe will streamline functional characterization of human enhancers. I aspire to contribute to advancements in the diagnosis and treatment of regulatory diseases through our research efforts.

The lab of Drs. Robert and Janet Siliciano focuses on HIV cure. Antiretroviral therapy (ART) prevents new cycles of HIV replication and disease progression, but it is not curative. If ART is stopped, viral rebound will occur due to the persistence of a population of infected cells that can hide from our immune system, known as the latent reservoir. The development of effective ART regimens was approximately 27 years ago, so there are few studies looking at the population of people with HIV on ART for over 20 years. Previously, our lab showed that the latent reservoir decays slowly during the first seven years of ART, enough to guarantee lifetime persistence of HIV, but it was unknown if that decay continued after long-term ART (greater than 20 years), or if ART should continue indefinitely. In this study, we discovered that the latent reservoir does not continue to decay and actually increases slowly. These results emphasize that ART should not be discontinued even after 20 years of treatment, and that the latent reservoir remains a barrier to cure.

Loss of neurons is the key pathological feature of many retinal degenerative diseases that often result in permanent blindness. While there is currently no effective regenerative therapy to replace neurons, one of the most potentially promising strategies is through direct reprogramming of endogenous Müller glia into retinal neurons. Despite successful applications in vitro, in vivo implementation has been hampered by low efficiency. In our study, we present a highly efficient strategy for reprogramming retinal glial cells into neurons by simultaneously inhibiting key negative regulators. Our discovery of the near-complete conversion of glia to neurons demonstrates that there is no clear intrinsic barrier to glial reprogramming in the retina, further strengthening the feasibility of reprogramming as a viable therapeutic strategy for retinal degeneration. This work was done in Dr. Seth Blackshaw’s lab.

I focused on studying the local clock mechanism in regulating rhythmic behaviors. More than 50 years ago, the suprachiasmatic nucleus (SCN) was determined to house the master circadian pacemaker that coordinates nearly all daily biological rhythms in mammals. Since that time, the dominant paradigm for the organization of the circadian network has postulated that the SCN acts to synchronize local clocks throughout the body. In the brain, these local oscillators are thought to regulate individual rhythmic behaviors, but a discrete extra-SCN brain oscillator has never been previously identified. In addition to cyclical clock gene expression, brain oscillators should also exhibit rhythms of electrical activity. However, no genetic marker exists that labels electrically rhythmic neural circuits. In this project, we leveraged these observations and found that mWAKE labels an extra-SCN brain oscillator in the lateral amygdala (LA). We first show that mWAKE is enriched in a molecularly defined subregion of the LA (anterior-dorsal LA/adLA). Although the core clock protein Per2 cycles throughout the LA, we show that mWAKE-positive, but not mWAKE-negative, adLA neurons exhibit rhythmic intrinsic excitability. In contrast, expressing a clock-dominant negative virus in mWAKE-negative LA neurons does not affect Per2 cycling, even in the cells directly expressing the viral transgene. At a molecular level, the mechanisms mediating rhythmic excitability outside of the SCN are poorly understood. We show that mWAKE levels rise at night in adLAmWAKE neurons and upregulate BK current to inhibit the excitability of these cells at night. Finally, we show that, rather than modulating an individual behavior, adLAmWAKE neurons utilize distinct projections to produce clock- and mWAKE-dependent rhythmic changes in two different behaviors: touch sensitivity and anxiety. Our investigation of the first discrete extra-SCN brain oscillator also reveals new insights into the nature of the circadian timing network.

In the Green Lab and Doloff Lab, what I’ve had the privilege of exploring is unique in that I’m exploring both arms of the immune system when it comes to dysregulation and disease. My research dissertation aims to develop and screen biomaterial-based mRNA nanoparticle formulations to target immune cells, such as antigen presenting cells, for oncology and autoimmune therapies. There is an urgent need for a therapeutic that can engineer the immune system to prevent disease progression at early stages and have a long-lasting and transformative effect on patients. Recent novel nonviral delivery biotechnologies have great potential to impact medicine and are being investigated in clinical trials with a few in the clinic today, such as the current Moderna and Pfizer/BioNTech mRNA SARS-CoV-2 vaccine in addition to Alnylam’s RNAi therapy for rare hereditary diseases — setting precedent for subsequent gene delivery methodologies such as my dissertation.

As a Post-baccalaureate Research Education Program (PREP) Scholar, I am researching under the mentorship of Dr. Netz Arroyo in the Department of Pharmacology and Molecular Sciences. My project focuses on developing an electrochemical platform for investigating intercalation as a mechanism of DNA binding. Intercalators are small molecules that insert a part of themselves between the base pairs of double-stranded DNA, causing the DNA to unwind and inhibit replication and other cell functions. With many uses in chemotherapy drugs and treatments for various infections, intercalators are important in developing new medicines to treat such health issues. Medicinal chemists use indirect and complex techniques such as viscometry and X-ray crystallography to confirm intercalation, slowing down drug development efforts to find new intercalators. To provide a direct and efficient way of detecting DNA intercalation, I am developing a DNA-hairpin electrochemical sensor that contains self-complementary DNA, allowing intercalation to occur. I have used phenanthridine as a model system to optimize sensor binding kinetics and gain an improved understanding of sensor performance under variable environmental conditions. Studying the effect of temperature on sensor performance in the absence and presence of an intercalator further confirmed intercalation was occurring as the melting temperature increased in the presence of an intercalator.

Dr. Sohn’s lab studies various biological stress-sensor proteins, including mammalian cyclic GMP-AMP (cGAMP) synthesis (cGAS). cGAS plays a crucial role in activating innate immune responses against cytosolic dsDNA in mammals. My research on cGAS involved solving multiple crystal structures of cGAS bound to different substrates, dinucleotide intermediates and divalent metals at various catalytic stages. Our structural findings, combined with biochemical measurements, redefined the molecular mechanism by which cGAS generates 2’-5’/3’-5’-linked cGAMP. The novel structures and mechanistic insights from our study have the potential to guide researchers in developing small molecule therapeutics targeting different active states of cGAS.

I discovered a novel mechanism that enables neuronal contact sites called synapses to rapidly strengthen their signaling and maintain it during high demand. Synapses communicate via signaling molecules called neurotransmitter. Neurotransmitter is packed into synaptic vesicles, and upon neuronal activity, synapses fuse these vesicles at release sites and release neurotransmitter onto postsynaptic receptors for signal transduction. After vesicle fusion, these release sites become vacated and require new vesicles for continued neurotransmitter release. My work uncovered the function of two proteins that coordinate the replenishment of these release sites after their use by keeping so-called replacement vesicles nearby. This finding is significant to the field of synaptic cell biology as it provides an answer to a long-standing question — how is it that synaptic vesicles are kept close enough to release sites to rapidly replace the vesicles fused during the previous round of release? I conducted my research in the lab of Dr. Shigeki Watanabe.

My work uncovered the mechanism of action for a class of anti-epileptic drugs that target the most common neurotransmitter receptor in the brain. This discovery will help design future drugs to target epilepsy and other neurological disorders. I led a collaboration between Rick Huganir and Ed Twomey’s labs, which initiated this work, and we also collaborated with Albert Lau’s lab at Johns Hopkins and Vasanthi Jayaraman’s lab at the University of Texas Health Science Center.

Species-specific vocalizations are important for the survival and social interactions of both humans and vocal animals. My research projects focus on understanding the underlying neural mechanisms for species-specific vocalization processing in the brain. We use a unique nonhuman primate, the common marmoset (Callithrix jacchus, New World primate), as our animal model. The marmoset is one of the two key laboratory primate models, and provides several important advantages over other nonhuman primates: a rich vocal repertoire, a high reproductive rate while in captivity, a relatively short lifespan, a similar hearing range as humans, and a smooth brain allowing easy access to all parts of the cerebral cortex. We demonstrated the continuity and divergence of the dual auditory pathways in the primate brains along the evolutionary path and highlight human-specific brain specialization for speech and language processing, suggesting that the putative neural networks supporting human speech and language processing might have emerged early in primate evolution. Moreover, we revealed the existence of voice patches in the auditory cortex of marmosets, and support the notion that similar cortical architectures are adapted for recognizing communication signals for both vocalizations and faces in different primate species. These findings are significant because they will give us a unique opportunity to provide critical new evidence to modify and expand current models of speech processing, and will open exciting avenues of research in understanding the roles of the auditory dorsal and ventral neural networks in vocal perception and production. Ultimately, understanding these mechanisms will provide new diagnostic and therapeutic avenues for people with speech and communication disorders. I conducted my research in Dr. Xiaoqin Wang’s lab in the Department of Biomedical Engineering. Our lab is devoted to understanding the neural basis of auditory perception and vocal communication in a naturalistic environment.

As a postdoctoral fellow in Dr. Shuying Sun’s lab, I study how the dysregulation of RNA metabolism contributes to neurodegeneration, particularly in amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). We provided evidence that showed a prevalent mRNA modification, N6-methyladenosine (m6A), as a previously unknown pathogenic mechanism in the most common familial C9ORF72-associated ALS and FTD. On one hand, we found that the abnormal m6A reduction leads to the transcriptome-wide mRNA dysregulation with significant enrichment in synaptic activity and neuronal functional pathways. On the other hand, we found that m6A reduction disturbed the decay of the causative repeat expansion in the C9ORF72 gene. Our strategies to elevate the m6A levels in the diseased neurons rescued a broad spectrum of post-transcriptional dysregulation and disease-related pathologies, which holds great therapeutic potential.

Lipid nanoparticles, renowned for their role in delivering messenger RNA for COVID-19 vaccines, have emerged as promising vehicles for cancer immunotherapy. Earlier studies concentrated on enhancing lipid nanoparticles to stimulate robust responses from T helper 1 (Th1) cells, which are pivotal in enabling the immune system to recognize and combat cancer cells. Working in Hai-Quan Mao’s lab, I utilized an innovative screening technique to optimize the lipid nanoparticle composition, tailoring it to enhance and maximize immune response activation. Through this endeavor, I found lipid nanoparticles capable of eliciting responses through dual pathways, effectively presenting tumor antigens to both Th1 and Th2 cells, another subset of helper cells. This research stands out for demonstrating the potential of lipid nanoparticles to enhance both Th1 and Th2 responses, orchestrating concerted attacks on cancer by diverse immune cell populations. Such findings offer a versatile strategy for vaccine development applicable across various diseases, thereby broadening the scope of mRNA lipid nanoparticle-based immunotherapies.

Exosomes are small extracellular vesicles of 30 to 200 nm in diameter. They have the same topology as the cell, and carrying selected proteins, lipids and nucleic acids. They are secreted by all cells, having important roles in health and disease. Under the mentorship of Dr. Stephen Gould at Johns Hopkins, my work focused on exosome biogenesis. In the prevailing model, syntenin drives the biogenesis of CD63 exosomes by recruiting Alix and ESCRT machinery to endosomes and leading to multivesicular endosomes mediated pathway of exosome biogenesis. Here I show syntenin drives CD63 exosome biogenesis by simply blocking CD63 endocytosis, thereby allowing CD63 to accumulate at the plasma membrane, and budding from there subsequently. I also showed the role of endocytosis is not limited to syntenin and CD63. I propose a new model based on our observations: Exosomes primarily bud from the plasma membrane, and endocytosis inhibits the protein loading into exosomes. This model revolutionizes our understanding of exosome biogenesis and has a profound impact on exosome-based drug pharmaceutical industry.