The research projects at NYU Langone’s Parekh Center for Interdisciplinary Neurology all capitalize on cross-disciplinary collaboration. Each focuses on commonalities between multiple neurological disorders and strives to build a framework for how the gut microbiome, peripheral immune system, and immune and glial cells in the central nervous system interact at every stage of neurological disorder.

Restoration of Healthy Gut Flora with Postbiotics in Parkinson’s Disease

Principal Investigators: Un Jung Kang, MD; Jonas Schluter, PhD; Aasma Shaukat, MD, MPH; Giulietta M. Riboldi, MD, PhD; and Kara G. Margolis, MD

It has become increasingly evident that body systems beyond the nervous system are impacted in neurological disease. Among these, gastrointestinal (GI) health and the role of the gut microbiome have emerged as critical components of nervous system disorders, including in synucleinopathies such as Parkinson’s disease (PD), dementia with Lewy bodies (DLB), and multiple system atrophy (MSA). As GI abnormalities are common in PD and may precede the manifestation of the movement disorder, the GI system likely contributes to PD pathogenesis. Significant perturbation of the gut microbial community has been consistently reported in PD patients compared with healthy individuals. However, whether this contributes to the disease process in PD or is a consequence of the disease remains unknown.

We plan to employ novel microbiome modulation methodology in PD patients to study its effect on disease pathogenesis and clinical symptoms, which would answer the scientific question of whether the microbiome differences in PD are the cause or a consequence of the disease. We will investigate whether the postbiotics, a novel approach for altering the gut microbial community, can sustainably modify the gut microbiome in PD patients to promote a microbial composition more associated with health, determine whether this can alleviate GI symptoms associated with PD, and ultimately delay PD progression.

The Influence of IFN Signaling on Astrocyte Function and Neuronal Vulnerability Following Ischemic Stroke

Principal Investigators: Shane A. Liddelow, PhD, and Youssef Z. Wadghiri, PhD

Stroke is a leading cause of disability and the fifth leading cause of death in the United States. It is associated with neuron cell death, vascular damage, and infiltration of peripheral immune cells—all of which provoke central nervous system (CNS) inflammation. Neuron death is tightly regulated by neuroinflammation following the initial event, but also in the subacute and chronic phases—with many pathological hallmarks like glial scarring persisting for years. Despite their critical supportive role in normal brain homeostasis, microglia and astrocytes have been increasingly implicated in the progression of degeneration and worsening of many diseases. Understanding the altered functions of distinct substates of reactive astrocytes and their impact on neuronal vulnerability will pave the way for novel strategies to protect neurons in following stroke and a number of other neurodegenerative and neurodevelopmental disorders.

One substate is strongly interferon (IFN)-responsive, which hints at an important immune regulatory role. We have modeled these IFN-Responsive Reactive Astrocytes (IRRAs) in vitro and found that they do not induce apoptosis in neuronal cultures. We will investigate whether IRRAs respond to peripheral signals to preserve brain homeostasis and maintain neuronal connectivity and function following stroke.

Gut Microbiome as a Mechanism of Drug Resistance in Epilepsy

Principal investigators: Claude Steriade, MD, and Deepak Saxena, MS, PhD

There are an estimated 3.4 million people with active epilepsy in the United States. Despite an exponential increase in approved medications to treat epilepsy, one-third of patients remain refractory to medications, and mechanisms of drug resistance are not completely understood.

Drug-resistant epilepsy presents with alterations in the gut microbiome, and ketogenic diet studies have suggested that the gut microbiome influences seizure control. Therefore, manipulation of the gut microbiome may be an effective epilepsy therapy. Seizures are also associated with neuro-inflammation, specifically upregulation of IL1β and IL6, and inflammatory dysregulation itself, such as altered levels of IL1β, can in turn lead to dysbiosis in animal models. Thus, neuroinflammation is a potential mechanism by which seizures may impact gut dysbiosis, and vice versa.

This project is improving our understanding of the independent role of seizures in the gut microbiome and explores neuro-inflammation as a potential mediator of the relationship between seizures and gut dysbiosis. In doing so, it will inform the design of rational and targeted microbiome interventions in the treatment of epilepsy.

Converging Lines of Evidence Between Alzheimer’s Disease and Epilepsy

Principal Investigator: Jayeeta Basu, PhD, and Orrin Devinsky, MD

Converging lines of evidence support that Alzheimer’s disease (AD) and epilepsy share underlying mechanisms: AD patients have an increased risk of seizures, and developing epilepsy while temporal lobe epilepsy (TLE) patients suffer from memory loss and cognitive impairments that mirror AD symptoms. The hippocampus (HC) and entorhinal cortex (EC) are common seizure foci in TLE and are affected early in AD. While these brain regions are extensively studied in rodents, the cellular and circuit organization of human EC-HC network is not well defined. What are the underlying changes at the level of cell types, molecular and synaptic physiological, that disrupt circuit dynamics causing hyperexcitability in TLE and AD?

A critical gap remains in understanding the consequences of seizure activity on HC neurons and circuit dynamics that limits our ability to effectively treat seizures and prevent or reverse cognitive and behavioral comorbidities. In an unprecedented, multidisciplinary collaboration, our study will decipher the synaptic, cellular, and circuit-level mechanisms for neuronal hyperexcitability in human epilepsy patients, and in rodent models of AD prone to epilepsy. By identifying common brain regions and cell-types of vulnerability in TLE and AD, we seek to better understand the mechanisms of epileptogenesis, neurodegeneration, and memory and cognitive dysfunctions, and to identify novel therapeutic targets for early intervention, to prevent or reverse cognitive deficits.

Inhibition of the CAMKK2-AMPK Dyad to Prevent Sleep-Deficit Alzheimer’s Disease Pathogenesis

Principal Investigators: Timothy Cardozo, MD, PhD, Ricardo Osorio, MD, and Thomas Wisniewski, MD

Individuals with sleep problems appear to have nearly double the risk for developing cognitive impairment (CI) or Alzheimer’s Disease (AD) and up to 15% of AD prevalence may be directly attributable to sleep dysfunction. Poor sleep quality and daytime sleepiness have been associated with β-amyloid (Aβ) and tau accumulation in cerebrospinal fluid (CSF). Conversely, Aβ, the pathological hallmark of AD, appears to directly impair sleep and its benefits, which is observable in rodent models of AD.

The CAMKK2-AMPK kinase pathway is over-activated by Aβ42 oligomers (Aβ42o) and triggers activation of several downstream effectors, leading to early loss of excitatory synapses (synaptotoxic effects) in neurons before plaque formation and without compromising neuronal viability. This strongly suggests that synaptotoxicity is an early event in AD progression, triggered by Aβ42o.

Sleep disturbance enhances AMPK phosphorylation and CAMKK2 levels in the brain. Thus, CAMKK2-AMPK, sleep, and AD are linked, suggesting that both Aβ42o and sleep disturbance impose a stress on brain tissues that is responded to deleteriously by the CAMKK2-AMPK pathway. AMPK is ubiquitously expressed by all cells in the body, but CAMKK2 expression is largely restricted to AD relevant regions of the brain, and thus constitutes an ideal drug target. The Cardozo lab developed a small molecule CAMKK2 inhibitor, which has properties suggesting it could be delivered orally and penetrate the brain to have its effects. Thus, this research team is in a unique position to establish a highly innovative, long-term research program into the key signaling pathway controlling the intersection of sleep disturbance and AD.

Deciphering the Neurobiological Mechanisms of Visual Hallucinations

Principal Investigator: Biyu J. He, PhD

Visual hallucination is a mysterious condition that afflicts patients with a variety of clinical conditions, including age-related vision loss, neurodegenerative disorders, and psychiatric illnesses. The occurrence of visual hallucinations constitutes an important safety risk, is distressing to patients and families, and predicts worse clinical outcomes. We hypothesize that several neural factors contribute to visual hallucinations, including hyperactive spontaneous brain activity, overactive top-down feedback, and an overreliance on prior knowledge learned from past experiences.

To elucidate the neurobiological mechanisms of visual hallucination, we will study two distinct patient populations: those with ophthalmological conditions, including Charles Bonnet Syndrome (CBS), and those with neurodegenerative disorders, including Parkinson’s disease (PD) and dementia with Lewy body (DLB). In these conditions, the pathology lies either in the input pathways (CBS) or in the central brain (PD and DLB). The He Lab will integrate findings across patient groups to build a unified framework of the neurobiological underpinnings of visual hallucinations. Our long-term goal is to identify biomarkers and potential treatment targets for visual hallucinations across diseases to meet a major unmet clinical need.

Loss of Energy Homeostasis Drives Neurodegeneration

Principal Investigators: Liam J. Holt, PhD; Michael E. Pacold, MD, PhD; and Nicolas Tritsch, PhD

Protein aggregation and mislocalization are hallmarks of neurodegeneration (e.g., TDP-43), but the factors driving these molecular pathologies remain to be elucidated for both familial and sporadic disease. ATP-dependent metabolic activity is essential for molecular motion and to prevent aggregation of proteins. The cost of neuronal activity is extremely high, and hyperexcitability decreases ATP leading to decreased solubility of TDP-43. Metabolic deficiencies related to Type 2 diabetes are a major risk factor for AD. We hypothesize that ATP depletion disrupts the dynamics of assembly, disassembly, and motion of molecules, leading to initiation of neurodegeneration.

By studying ATP flux and its effect on neurons as well as TDP-43 transport and aggregation we will gain a better understanding of brain energy metabolism, determine if a causal link exists between hyperexcitability, loss of energetic homeostasis, protein aggregation, and neurodegeneration, and provide a new framework to explore metabolic interventions to treat or prevent neurodegeneration.

Deciphering the Role of the Microbiome in Parkinson’s Disease Pathogenesis

Principal Investigators: Khalil Ramadi, PhD, and Thong C. Ma, PhD

The human gastrointestinal (GI) tract is densely populated by over 1013 bacteria of various phyla, and they can have a huge impact on health. Dysbiosis has been linked to various neurological disorders such as Parkinson’s disease (PD). The GI microbiome is heterogeneous and varies significantly across the GI tract (GIT). Microbiome analyses are generally done through fecal sampling, which mostly represents luminal bacteria in the colon and does not reveal the full breadth of the microbiome along the GIT.

Other than endoscopic biopsy or postmortem histology, few techniques exist to interrogate bacteria in the small intestine, and those adherent to the mucosa along the GIT. This is particularly significant in PD, as, for example, microbes in the small intestine may regulate the bioavailability and efficacy of levodopa, the gold standard therapeutic for symptomatic control. We aim to develop ingestible devices for comprehensive microbiome analysis to study the impact of α-synuclein pathology (which leads to PD) on the microbiome and use these findings to provide greater resolution of the bacterial species present and their alterations.

Neurodegenerative and Neuroinflammatory Plasma Biomarkers Associated with Epilepsy in Alzheimer’s Disease

Principal Investigators: Orrin Devinsky, MD, Jaqueline French, MD, Thomas Wisniewski, MD

Alzheimer’s Disease (AD) and epilepsy are reciprocally related: AD increases risk for late-onset seizures that occur in 10 to 22 percent of patients, and epilepsy increases risk for cognitive impairment in up to 80 percent of patients. Epileptiform activity and cognitive deficits are linked with tau pathology, with total-tau and phosphorylated tau (pTau) increased in AD and epilepsy hippocampi. Specific pTau sites are linked to early AD stages, with increased pTau217 and pTau231 in brain and plasma. However, plasma tau has only been studied in AD and some epilepsy groups.

Tau has not been studied in plasma or neuron derived exosomes that may better correspond to findings in the brain from AD cases with and without epilepsy. We seek to better understand how epilepsy contributes to the etiology of Alzheimer’s disease (AD) by identifying biomarkers in AD with and without epilepsy, which will inform on follow up mechanistic, diagnostic, and prognostic studies.

High-Throughput Drug Screening in mESCs for Transposable Element-Associated Neurological Disorders

Principal Investigators: Weimin Zhang, PhD

X-Linked Dystonia Parkinsonism (XDP) is a devastating neurodegenerative disorder that mainly affects Filipino males and their descendants across the world. Genetic studies have pinpointed an SVA retrotransposon insertion in TAF1 gene as the major causal variation for XDP. SVA mobilization in neural cells influence the regulation of nearby genes. To counteract that, hominoid genomes have co-evolved KRAB zinc-finger proteins to repress SVA’s regulatory activities which has minimized the pathologic impact of SVA, and therefore increases the difficulty of finding effective cures for the associated neurological disorders.

High-throughput drug screening is a powerful approach for finding effective treatments—a strong phenotype and clear readout is the prerequisite. Humanizing the disease-linked SVA along with the affected human gene in a model organism with a clean genome without coevolved repressive mechanisms, could amplify the pathologic phenotypes of a neurological disease. Using in vitro high-throughput drug screening, we seek to identify compounds for one SVA-associated neurological disorder, XDP, which can become candidates for treatment of other SVA-associated neurological disorders.

Using Artificial Intelligence to Establish Diagnostic Criteria for Neuropathological Classification of the Hippocampus in SUDC

Principal Investigators: Orrin Devinsky, MD, and Thomas M. Wisniewski, MD

The normal morphology of the hippocampus in the developing child is not well understood, and neuropathologists are inconsistent in identifying “abnormal” findings as diagnostic criteria lack standardization. A better understanding of the normal hippocampus would allow us to distinguish normal variants from pathologies in the heterogeneous group of sudden unexplained deaths in childhood (SUDC; death of a child >12 months that remains unexplained after a thorough autopsy and investigation). Each year, approximately 400 children in the United States die from SUDC, accounting for more than 31,000 life-years lost annually. Most cases are healthy toddlers, aged 1 to 4 years -- the 5th leading category of death in this age group.

Utilizing artificial intelligence (AI), we plan to create and test an image recognition algorithm to standardize diagnostic criteria of hippocampal morphology among young children. This proof-of-concept approach has potential to impact the evaluation of multiple neurological disorders in which hippocampal pathology is a prominent feature across the lifespan, including epilepsy and Alzheimer’s disease.

Defining the Role of Cortical Disinhibition in the Development of Post-stroke Neurologic Complications

Principal Investigators: Sean N. Kelly, MD, PhD

There are a host of unexpected disabling complications that patients develop in the days and weeks after having a stroke, including poststroke seizures (3 to 7% of patients), precipitous cognitive decline (11%), movement disorders (4%), and mood disorders (31%). These secondary complications lead to a massively underappreciated burden of disability in the more than 795,000 patients who experience stroke in the United States each year. The underlying cellular and circuit mechanisms causing these complications are not yet known, and effective targeted treatments are very limited. Moreover, we have no biomarkers or predictive indexes to reliably determine which patients will be most likely to experience them.

We hypothesize that the pathologic disruption of cortical inhibition in peri-ischemic regions surrounding stroke beds is a key mechanism for igniting aberrant circuit activity. Additionally, we hope to identify activity signature and possible biomarkers related to secondary stroke complications.

Neuroimaging in the Time of Artificial Intelligence: How Computer Learning Is Advancing Neurology and Neurosurgery Research at NYU Langone Health

Principal Investigators: Rachel Kenney, PhD; Eric Oermann, MD; and Kimberly O’Neill, MD

We are at the cusp of a new era in neurological disease research, education, diagnosis, and treatment, due to the burgeoning field of artificial intelligence (AI). Computer learning has the ability to transform medicine as we know it and improve outcomes for our patients. This is especially true in the fields of neurology and neurosurgery. Drawing researchers together to encourage idea and expertise exchange has proven to be a productive manner in which to embark on new projects across many disciplines. We will accomplish this by hosting a symposium for researchers with interest in AI and neuroimaging to catalyze collaborations for future research.

This symposium will lead to a multidisciplinary consensus on best practices and a roadmap to create a comprehensive multimodal neuroimaging dataset of the central nervous system (CNS) and clinical data. By bringing together NYU’s experts in the field of AI, the symposium will advance the field of neuroimaging and clinical applications of machine learning techniques. It will lay the groundwork for ongoing collaboration and innovation with dialogue that continues beyond the event, facilitating sustained progress towards our long-term goals. The insights gained with this symposium and subsequent meetings can be applied across neurological diseases and beyond.

Neuroimaging in the Time of Artificial Intelligence: How Computer Learning Is Advancing Neurology and Neurosurgery Research at NYU Langone Health

Principal Investigators: David Schoppik, PhD, and Molly C. Gale Hammell, PhD

The microtubule-associated protein tau has been implicated in the pathogenesis of tauopathies, various neurodegenerative disorders. Patients suffering from tauopathies often present with sleep disruption and locomotor dysfunction; treatments are largely palliative, and prognosis is poor. Pathologically, glial cells called astrocytes exhibit inclusions of abnormal tau in many tauopathies, including progressive supranuclear palsy (PSP), globular glial tauopathy (GGT), and corticobasal degeneration (CBD).

We seek to define and validate transcriptional signatures of astrocytic tauopathy, particularly with respect to datasets from tauopathic patients/models. We hypothesize that these behavioral disruptions result from transcriptional changes in astrocytes that express tau. By bringing together the expertise of the Schoppik and the Molly Gale Hammell labs to conduct a high-risk and high-reward set of experiments to illuminate the cellular correlates of tauopathy, we will investigate if transcriptional changes to tauopathic astrocytes lead to cellular dysfunction that causes locomotor and sleep deficits.