Thesis Seminars

Allison Murphy - PhD Candidate, Advisor: Farran Briggs, PhD

In the visual system, information enters as light hitting the retina of the eye. This signal is transmitted to the dorsal lateral geniculate nucleus (LGN) in the thalamus, then to primary visual cortex (V1). In highly visual animals, this feedforward visual pathway is organized into parallel informational streams, known as the magnocellular, parvocellular, and koniocellular streams in primates and the somewhat homologous X, Y, and W streams in carnivores, including ferrets. While these feedforward stream have been extensively studied, relatively little is known about the corticogeniculate (CG) feedback pathway that projects from V1 to the LGN. Although corticogeniculate feedback synapses in the LGN greatly outnumber feedforward retinal inputs, their modulatory nature has made their functional role difficult to study. The purpose of this thesis is to examine the structure of this pathway and its functional role in vision by answering the following questions: 1) Is a parallel stream structure maintained in the feedback pathway? 2) What is the functional impact of feedback in the LGN? To answer these questions, we performed neurophysiological recordings simultaneously in LGN and V1 of anesthetized animals. For the first question, we found functionally connected pairs of corticogeniculate and LGN neurons using cross-correlation of their spike trains. We analyzed the physiological response properties of these pairs of neurons to determine whether distinct populations of CG neurons maintain stream-specific projections to the LGN. Additional anatomical analyses provide support for distinct morphological subtypes of CG neurons. For the second question, we used a modified rabies virus to selectively infect corticogeniculate neurons with a light sensitive ion channel and performed optogenetic manipulation during presentation of visual stimuli. We then examined how manipulation of corticogeniculate feedback affected visual response properties of LGN neurons, including the variability and information content of their spike trains. Together, these experiments demonstrate that corticogeniculate feedback is a pathway with a complex organizational structure that plays a subtle role in shaping LGN visual responses.

 Apr 11, 2023 @ 11:00 a.m.

 Medical Center | K-307 (3-6408)

Thomas Delgado - PhD Candidate, Advisor: Gail Johnson, PhD

Astrocytes are essential for maintaining neuronal function in resting and disease states. Following injury, astrocytes take on reactive phenotypes that grade from neurotoxic to neuroprotective, which impact subsequent neuronal recovery processes, such as axonal regeneration. Only recently has the heterogeneity of reactive astrocyte populations begun to be described, and little is known about the contextual requirements and molecular inflection points that underlie these graded responses. Accumulating data from my lab suggests that one of these inflection points is transglutaminase 2 (TG2). TG2 is complex in that it regulates signaling of numerous molecular pathways at the cell membrane, in the cytosol, and in the nucleus; thus, it can provide multiple levels of context-dependent input into gene regulation. Importantly, when TG2 is depleted from astrocytes, they better protect neurons in culture from oxygen-glucose deprivation (OGD), improve motor function recovery in a mouse spinal cord injury model, and better facilitate neurite outgrowth in vitro on an injury-relevant, growth-inhibitory matrix. Additionally, TG2 depletion upregulates lipid handling (lipid uptake and formation of lipid droplets) and lipid metabolism pathways in an injury-dependent manner. As stressed neurons accumulate reactive oxygen species and peroxidated lipids, and cannot metabolize these toxic lipids on their own, they export them to astrocytes. Astrocytic lipid uptake and metabolism eliminates these oxidized lipids and recycles them to produce energy substrates for neurons. This mechanism may partly underlie the protective effect of TG2-/- astrocytes after injury.

In aim 1, I will test the hypothesis that depletion of astrocytic TG2 improves axonal regeneration through an inhibitory extracellular matrix (ECM) after CNS injury using an optic nerve crush model in mice. These studies are built on previous spinal cord injury data from my lab, with the unique addition that, in the optic nerve, axonal regeneration can only occur through the inhibitory ECM deposited at the crush site, while in the spinal cord, regrowth may occur via collaterals around the injury site, through permissive matrices.

In aim 2, I will identify the major molecular pathways and mechanisms of gene and protein regulation underlying the unique function of TG2-/- astrocytes. As TG2 is a known interactor of transcription machinery and chromatin regulators in the nucleus, I will use ATAC-seq to analyze differences in chromatin accessibility between TG2-/- and wild type (WT) astrocytes and integrate these data with differentially regulated transcripts identified through RNA sequencing. I will compare differential enrichment in upstream signaling pathways with differential protein expression identified by tandem mass spectrometry to better approximate the functional impact of these molecular changes.

In aim 3, following data that lipid handling and metabolism is differentially regulated in TG2-/- astrocytes, I hypothesize that TG2-/- astrocytes more efficiently uptake and metabolize lipids that are released by stressed neurons (as peroxidated lipids), ultimately providing better control of oxidative stress and increased energy supply to neurons in injury conditions. Therefore, I will measure the ability of WT and TG2-/- astrocytes to take up and metabolize lipids released by neurons, in control and stressed conditions, as indicated by lipid droplet accumulation. Further, I will measure metabolic flux through ketone and cholesterol synthesis pathways, downstream of fatty acid oxidation, in WT and TG2-/- astrocytes by labeling them with C13-palmitate and measuring C13-labeled metabolites by LC-MS. I will then pair neurons with either astrocyte group to track the export of C13-labeled energy substrates to neurons.

Overall, this project will explore the role of TG2 as a key regulator of astrocyte reactivity after CNS injury and, within this scope, better characterize the astrocytic metabolic pathways that are integral for neuronal and functional recovery.

 Apr 07, 2023 @ 2:00 p.m.

 Medical Center | Lower Adolph Auditorium (1-7619)

Garrick Salois - PhD Candidate, Advisor: Margot Mayer-Proschel, PhD

Iron deficiency is the most common micro-nutrient deficiency worldwide and is especially common amongst pregnant women. Gestational iron deficiency (gID) has been correlated with a wide range of timing, dose, and duration-dependent behavioral and neurophysiological consequences that are often not corrected by iron supplementation later in life. Despite its prevalence and consequences, little is known about how gID affects the developing human nervous system.

Data from mouse models suggests that gID may affect the balance of excitation and inhibition in adult neural circuits and that the timing of iron deficiency may alter both the resulting phenotype as well as the potential for treatment via iron supplementation. Using immunofluorescence in our mouse model of gID, in which overt brain iron deficiency begins by embryonic day 12, we observed increased embryonic expression of Nkx2.1, a transcription factor critical for fate determination of inhibitory neural precursors in the ganglionic eminence. Furthermore, despite iron supplementation at birth, the offspring show a disruption in cortical inhibitory interneuron subtypes that persists at least until day 100.

To examine whether these observations in mice extend to human neurodevelopment, we established a human ventral forebrain organoid model of gID that recapitulates relevant iron levels and allows for the identification of changes in the transcription factor Nkx2.1 as well as subsequent impairments in the proliferation, differentiation, and maturation of developing interneurons. We then used a variety of techniques including immunofluorescence, confocal microscopy, ICPMS, qRT-PCR, flow cytometry, and sc-RNAseq to measure the effect of iron deficiency on organoid development. Using this novel organoid system, we observed that three major hallmarks identified in our animal model are preserved: (i) disrupted divalent metal homeostasis, (ii) increased Nkx2.1 expression, (ii) and sustained changes in cell type specification. Our novel human model of gID allows us to now decipher the mechanisms that leads to these iron deficiency-associated disruptions and offers a unique opportunity to gain insight into the impact of this prevalent condition on fetal brain development.

 Mar 29, 2023 @ 2:00 p.m.

 Medical Center | 1-7619 Adolph Lower Aud.

Host: University of Rochester School of Medicine and Dentistry
The Neuroscience Graduate Program

Dennisha P. King - PhD Candidate, Advisor: Julie Fudge, MD

In primates, the amygdala’s basal nucleus is evolutionarily expanded, and matures postnatally. The ventral (“parvicellular”) basal nucleus (Bpc) is the last region to develop in the fetus; postnatally, it is bordered by the paralaminar nucleus (PL) which contains immature neurons. Between infancy and adolescence, the PL gains mature neurons (concomitantly reducing immature ones) implying that synapse formation is also occurring. Microglia form, remodel and maintain neural circuitry by pruning synapses that are non-functional, weak, or redundant. Our preliminary data have shown that the density, area, and perimeter of the microglia differ between the PL and adjacent Bpc, and increase in both regions from infancy to adolescence. Aim 1 will first characterize microglial morphology and their role in synaptic pruning in the PL and Bpc in control infant and adolescent monkeys during normal development. Aim 1A will extend preliminary morphologic analyses of Iba1-immunoreactive (IR) cells (microglial marker), using larger cell numbers, and more sophisticated analyses (Imaris software) in control infant and adolescent PL and Bpc. I expect that the microglia will increase their process length and complexity between infancy and adolescence in the PL and Bpc. Aim 1B will examine connectivity between pre-synaptic (synapsin-1) contacts and excitatory spines (PSD-95) in the PL and Bpc between infancy and adolescence. I hypothesize that synaptic contacts onto excitatory spines are greater in the Bpc which is more cellularly mature than the PL, and that contacts in both regions increase between infancy and adolescence. Aim 1C will focus on microglia’s role in pruning excitatory spines by examining PSD-95 engulfment by microglial lysosomes (PSD95 colocalization with CD68/Iba1IR). I hypothesize that PSD-95 engulfment, representing spine pruning, occurs in both the PL and Bpc more frequently in adolescence correlating with increased phagocytic activity in the microglia (CD68/Iba1).

During development, the PL’s postmitotic neurons continue differentiating and thus may be susceptible to being influenced by early life events. Our preliminary data shows that early life stress (ELS) in the form of maternal deprivation, results in a downregulation of PL transcripts that promote neural growth and migration, and an upregulation in complement pathways, a key mediator of synaptic pruning, during infancy. However, we still do not understand the dynamic relationship between the neurons and microglia in the PL and Bpc, which are both impacted by ELS. Aim 2A will determine whether ELS is associated with changes in infant monkey PL microglia morphology, suggesting increased synaptic pruning. Aim 2B will determine whether there is a net increase or decrease in synapses in the infant PL after the ELS conditions. In Aim 2C, I will interrogate PL transcription changes in ELS versus control infants to test the idea that complement mediated mechanisms are increased in ELS. Using a curated list of microglia transcripts, I hypothesize that many upregulated transcripts will be complement and immune-related molecules. I will also perform a pathway analysis (whole genome) to investigate the most changed pathways following ELS.

In Aim 3, I will validate the most changed microglial transcripts (Aim 2C) and examine whether ELS leads to increased microglial engulfment of glutamatergic spines in the infant PL, through a complement mechanism, e.g. a net increase of PSD95/C3 engulfment by CD68/Iba1. Using the results of Aim 2C, I will first choose the most changed transcripts involved in microglial pruning for which antibodies are available and ensure that these are changing at the protein level. I will then examine microglia (Iba1/CD68/complement receptor) engulfment of dendritic spines (PSD95, complement protein) in the PL of ELS versus control infants. Depending on the results of the microarray and validation experiments, I may focus on different members of the complement pathway.

Ultimately this project will examine the microglia’s role in supporting synaptic formation and maintenance in the primate PL and Bpc, and investigate whether and how ELS alters this role during infancy.

 Mar 15, 2023 @ 11:30 a.m.

 Medical Center | Ryan Case Method (1-9576)

Michael Giannetto - PhD Candidate, Advisor: Maiken Nedergaard, MD, DMSc

The brain is the most metabolically active organ in the body, yet it lacks a traditional lymphatic system for waste clearance. Instead, the brain utilizes the glymphatic system, a network of fluid filled spaces surrounding blood vessels, termed perivascular spaces (PVSs), which facilitates movement of cerebrospinal fluid (CSF) into the brain along arteries and waste clearance out of the brain along veins. Astrocytic endfeet comprise the outer boundary of the PVS and serve as a point of regulation for CSF flow into the brain. CSF flow in pial artery PVSs is well characterized to follow the same direction as blood flow driven by cardiac pulsations, and CSF flow is increased by large arterial dilations associated with neuronal activity. However, the function, and even existence, of the PVS along capillaries remains unclear. Additionally, the physiological function of pericytes, mural cells that cover capillaries, remains controversial. Some groups claim pericytes are important in blood flow regulation while others have demonstrated they are irrelevant to blood flow, but instead could maintain the extracellular matrix and phagocytose waste. Capillaries make up the bulk of vasculature surface area in the brain, with no brain tissue further than 30 micrometers away from a capillary, and capillaries are continuously covered by pericytes. Thus, capillary PVSs and pericytes are well positioned to clear waste but remain understudied. In this proposal, I will determine if there is directional fluid flow in capillary PVSs, test if pericytes contribute to capillary PVS function, and finally test pericyte and capillary PVS function in aging.

Aim1, I will test the hypothesis that fluid flow in the capillary PVS is directional, following the same direction of blood flow, similar to CSF flow in arterial PVSs. I will utilize in vivo 2-photon imaging of secreted fluorescent protein to label capillary PVSs and measure fluorescent recovery after photobleaching.

Aim 2 will test whether pericytes play a role in clearing waste or maintaining the structure of the capillary PVS. I will use approaches developed in Aim 1 combined with inducible genetic manipulations of pericytes to ablate them or impair their phagocytic and cell matrix maintaining functions, then measure capillary PVSs and glymphatic flow.

Aim3, I will test the hypothesis that capillary PVSs and pericyte dysfunction contribute to glymphatic impairment. I will use the same methods to label capillary PVSs to determine if functional fluid flow decreases, the structure of capillary PVSs changes, or pericyte functions decrease in a cohort of aged mice. I will then attempt to rescue pericyte function in aged mice using PDGF-beta supplementation to improve glymphatic function.

Ultimately this project will answer longstanding questions concerning function of pericapillary PVSs and pericytes, and determine their effect on the glymphatic system in normal aging.

 Feb 08, 2023 @ 2:00 p.m.

 Medical Center | Ryan Case Method (1-9576)

Berke Karaahmet - PhD Candidate in Neuroscience, Advisor: Kerry O’Banion, MD, PhD

Alzheimer’s Disease (AD) is a chronic neurodegenerative disorder that clinically manifests as the most common form of dementia. Due to their surveillance functions and immunocompetence as resident macrophages of the Central Nervous System (CNS), microglia are well-equipped to respond to perturbation of tissue homeostasis. Therefore, they are regarded as promising translational targets in modulating the impact of amyloid and Tau pathologies observed in AD. Here, we sought to elucidate the effect of two peripherally administrated pharmacologic approaches that are hypothesized to modulate microglial activation phenotypes in AD-like models.

In our first approach, we investigated the use of the multiple sclerosis (MS) drug, Glatiramer Acetate (GA), in murine models of aggressive amyloid pathology (5xFAD) or amyloid and Tau pathology combined (3xTg). In response to GA treatment, we observed improvements in cognitive function and molecular pathology in female 3xTg mice. These were associated with minimal transcriptomic changes in microglia, in which Dcstamp was the most upregulated gene. Follow-up analyses of Aβ plaque burden in 5xFAD; DCSTAMP knockout mice showed that the females of this genotype had increased plaque numbers, but this effect did not reach significance. In female 5xFAD mice, we found that GA treatment did not impact plaque burden if started early in life, showed a trend towards decreased plaque burden if started during early disease progression and, unexpectedly, increased plaque burden if started during late disease stages. No changes in plaque burden were observed in 5xFAD males.

In our second approach, we sought to investigate whether microglia that were depleted with a CSF1R inhibitor containing diet and allowed to repopulate could attenuate levels of pathological markers in aged 3xTg mice. We observed no changes in amyloid pathology but found differential effects in several markers of phosphorylated Tau. Single-cell transcriptomic analysis of microglia revealed a cluster that was strongly characterized by Cxcl13 expression. In situ analysis of Cxcl13 showed that it was localized to regions of AD-like pathology in 3xTg mice. This suggests that Cxcl13 upregulation in repopulated microglia responding to AD-like pathology might be one of the driving factors behind the changes observed in phosphorylated Tau levels. Further studies are warranted to establish mechanistic links between these observations.

Altogether, our data indicate that immunomodulatory therapeutics may be beneficial in restricting certain aspects of AD pathology. However, caution must be exercised when designing these therapies since outcomes may depend on the pathological stage of AD.

 Feb 01, 2023 @ 12:00 p.m.

 Medical Center | 1-7619 (Lower Adolph Auditorium)

Host: Dr. Kerry O’Banion, Neuroscience, University of Rochester School of Medicine & Dentistry