Traumatic spinal cord injuries (SCI) are devastating and lead to permanent and irreversible functional impairments. While the primary injury causes immediate damage, the progression of secondary injury is characterized by neuroinflammation. Microglia, the resident immune cells of the central nervous system, can further contribute to tissue damage. Although extensive research has been done focusing on microglial activation at the lesion site, their presence in regions distal to the injury at the chronic stage (4-6 weeks after injury) remains understudied. Here, we aimed to investigate the correlation between microglial activation in the lumbar region caudal to the injury site and cervical region rostral to the injury site with different grades of injury severity. We hypothesize that as injury severity increases, microglial activation in the surrounding region will increase as well. Contusion injuries were created at T8/9 in Long Evans rats using three impact forces: 200 kilodynes, 200 kilodynes and 2-second dwell time, and 250 kilodynes and 2-second dwell time. After 8 weeks post injury, our histological analyses revealed a significant increase in microglial activation in the caudal region, with activation levels reaching 136.03%, 205.08%, and 254.81% of sham levels in the 200, 200/2s, and 250/2s groups, respectively (n = 5/group). The most severe injury had the highest microglial inflammation. We anticipate a similar trend in the cervical region rostral to the injury. Combined, our data indicate a significant microglial response in regions distal to the injury, which correlates with impaired functional recovery. We believe that modulating microglial responses in the regions distal to the site of SCI could provide a novel approach to reducing chronic secondary injury and enhancing recovery following traumatic SCI.

Cervical spinal cord injuries (SCIs) often result in impaired locomotion and forelimb mobility. Assessed with a forelimb locomotor scale (FLS), previous research demonstrated improved forelimb locomotion with task-specific physical therapy (PT) training following a unilateral 150 kdyne C6 SCI in a rat model. Improvement in locomotion was calculated using a recovery ratio, which measures the difference in FLS scores normalized to the animal’s baseline score. There was a significant improvement in the recovery ratio (3dpi-14wpi) for animals that underwent task-specific PT training (p ≤ 0.0014). The underlying molecular mechanisms behind this improvement are still unclear. A potential explanation for these functional changes is the task-specific PT-induced increase in brain-derived neurotrophic factor (BDNF) targeted at the injury site. Task-specific PT stimulates neurons, microglia, and astrocytes to release BDNF, promoting neuron survival, growth, and synaptic plasticity. These processes are crucial for mediating the effects of cervical SCI trauma. This study investigates whether BDNF expression in the spinal cord varies in rats that underwent task-specific PT training following cervical SCI. PT-trained rats are hypothesized to have increased BDNF within the spinal cord, especially proximal to the injury site, which would help mitigate SCI trauma and enhance forelimb locomotion and mobility. To this end, we performed immunohistochemistry staining and quantification of BDNF in the spinal cord and compared it across three groups: PT + SCI, SCI-only, and Sham. The expression levels of BDNF in neuronal cells, astrocytes, and microglia were then correlated with their FLS scores to determine if there was a positive correlation between BDNF expression and forelimb locomotion. This would indicate that BDNF supports synaptic plasticity and recovery from SCI-induced trauma.

Initial work from our lab has demonstrated that decision-making in OCD patients varies along the approach-avoidance behavioral axis depending on their response to deep brain stimulation (DBS) treatment. Patients who do not respond to DBS exhibit heightened risk aversion and consistently make avoidant decisions, while responders balance risk and reward in their decision-making process. In contrast, patients who receive excessive stimulation display disinhibited behavior, making choices that maximize reward regardless of potential risk. While this 2D task has provided valuable insights into approach-avoidance behavior in OCD, it does not fully capture the naturalistic behavioral responses observed in daily life. To address this limitation, we are developing a virtual reality (VR) task designed to quantify approach-avoidance behavior in a dynamic 3D environment. Participants complete probabilistic decision-making trials while wearing a VR headset, allowing for precise tracking of eye movements, hand positioning, and body dynamics. This project aims to provide a naturalistic task to analyze movement velocity and behavioral trends to identify neural biomarkers associated with approach-avoidance tendencies. This work will act as a stepping stone, enabling deeper insights into how neuromodulation regulates these behaviors. We are currently finalizing the development of this VR task and will demonstrate the feasibility of using this task to collect unique behavioral data on approach-avoidance behaviors. In the future we hope to use this task to identify biomarkers of approach-avoidance behavior and use that information to further refine neuromodulation treatment for psychiatric disorders.

Dravet Syndrome (DS) is a severe developmental epileptic encephalopathy often associated with SCN1A mutations. DS is predominantly caused by a heterozygous loss-of-function mutation in the SCN1A gene, which codes for the pore-forming alpha subunit of the Na_v1.1 voltage-gated sodium channel. The disease is marked by seizures that are resistant to treatment, ataxia, developmental delays, cognitive impairment, and higher rates of premature mortality, primarily due to sudden unexpected death in epilepsy (SUDEP). At this time however, there is no effective intervention against these devastating outcomes. Anecdotal evidence from family members of children with DS suggests that sensory stimulation during these seizures might reduce their severity and duration. This study investigates whether sensory stimulation can reduce SUDEP in DS using a preclinical mouse model with the SCN1A knocked out. We created a closed-loop responsive system that detects seizure onset and triggers sensory stimulation in real time by utilizing piezoelectric sensors, a Teensy microcontroller, and a 12V computer fan to deliver airflow-based stimulation as a response to spontaneous seizures. Using the modified Racine scale, the system successfully identified scale 4 seizures (generalized tonic-clonic while lying on the belly), as well as scale 6 seizures (generalized tonic-clonic with tonic extension). However, it was unable to detect scale 5 seizures (by sudden, erratic jumping movements). Particularly, for scale 6 seizures, typically fatal in all cases, activating the fan completely prevented SUDEP, resulting in zero mortality. In contrast, for scale 5 seizures that went undetected and did not trigger the fan, mortality remained at 100%. These findings emphasize the potential of airflow-based sensory stimulation as a promising, non-invasive intervention for SUDEP. Future research will focus on improving seizure detection algorithms to enhance sensitivity across a wider range of seizure types.

Alzheimer's Disease (AD) is a neurodegenerative disorder with classical behavioral and pathophysiological presentations such as memory impairment, cerebral vascular alterations, and most notably, the accumulation of amyloid-beta plaques and tau fibers in humans. While these symptoms are hallmark signs of AD, the severity of these symptoms differs for individual patients. As such, clinicians face challenges in misdiagnosing or giving a late diagnosis using a few of these AD markers. My research project aims to characterize AD through 3 concurrent modalities with the goal of early diagnosis and better quality of care for AD patients: [1] motor activity using a motion capture chamber, [2] spatial memory and learning using Barnes Maze Trials, and [3] real-time cerebral hemodynamics with ultrafast contrast-enhanced ultrasound (CEUS) in a rat model of AD. This project utilized control wild-type (WT; n=3) and transgenic (AD; n=3) rats. We hypothesize that the WT and AD rats will express similar movement and cognitive metrics measured until they are 12 months old, whereas the AD rats show a gradual motor decline and cognitive impairment. The 12-month metric is used as a benchmark set by prior research using this rat model. Intravital CEUS imaging will be conducted at 16-19 months old. We expect CEUS imaging will reveal that cerebral blood flow in the parietal and hippocampal regions of the brain is reduced, and microvascular response is impaired in AD rats. The expected outcome of all 3 experiments is that there is a strong correlation between motor and cognitive decline with impaired hemodynamics in AD rats. Results from our study will allow for systematically chronologizing the progression of AD in greater detail than before which allows for greater diagnostic ability.  

Leigh Syndrome (LS) is a neurodegenerative disease due to the dysfunction of mitochondria. This disease usually begins in infancy and affects approximately 1 in 40,000 individuals, with children experiencing a progressive decline in their cognitive and motor functions often accompanied by severe treatment-resistant epileptic seizures. Mutations in Ndufs4, the gene that encodes a subunit of mitochondrial complex I have been linked to LS. Using mouse models, our lab has previously demonstrated that GABAergic interneurons play an important role in the pathophysiology of LS. Specifically, mice with Ndufs4 knockout (KO) in GABAergic neurons located across all brain regions exhibit seizures. However, seizures in epilepsy patients and animal models typically originate from forebrain structures. In this project, we examined whether inactivation of Ndufs4 in GABAergic neurons of the forebrain alone is sufficient to cause seizures in mice. To inactivate the Ndufs4 gene in the interneurons of the forebrain, homozygotes floxed Ndfus4 (Ndufs4flx/flx) mice were crossed with Dlx56Cre+ mice. Ndufs4flx/flx; Dlx56Cre+ mice obtained from this cross were used as experimental mice. We hypothesized that mice carrying the gene KO in this region will exhibit seizures and related mortality. Thermal seizure testing was conducted on 9 experimental mice and 10 control mice. Our results show that mice with Dlx56Cre KO exhibit a high seizure susceptibility to both spontaneous and thermally induced seizures. In addition, these mice exhibit a very reduced life span with nearly all mice dying by age P60. These findings indicate that inactivation of Ndufs4 in GABAergic neurons of the forebrain is sufficient to induce seizures and mortality in mice.

Glioblastomas (GBMs) are highly aggressive brain tumors with poor patient prognosis, necessitating improved preclinical models to evaluate therapeutic strategies. My lab develops cerebral organoids from human pluripotent stem cells, seeded with primary patient tumors to model GBM progression and therapeutic screening. Developing biologically relevant neural organoids provides a platform for integrating patient-derived GBM samples, enabling disease modeling and treatment testing. This study aims to optimize the embedding, cryosectioning and immunofluorescence (IF) staining protocols used to screen key molecular markers and cell populations within the organoids to validate their suitability for GBM tumor engraftment. Fixed organoids, along with embryonic and adult mouse brain tissues, are embedded in OCT to preserve structure and cryosectioned (12–20 μm). IF staining is optimized by adjusting fixation time, permeabilization, blocking reagents, and antibody concentrations to improve specificity and reduce background fluorescence. Markers analyzed so far include SOX2 (neural precursors), PAX6 (radial glia), FOXG1 (forebrain), and TUJ1 (neuronal differentiation). Mouse brain cryosections from newborn (P0) and adult (P56) stages serve as positive controls to validate antibody specificity and distinguish true signals from autofluorescence or non-specific staining. Images are acquired via Olympus scanner and analyzed using OlyViA and NIH Fiji (Enhanced ImageJ). Current efforts focus on optimizing section thickness for clearer images and refining blocking conditions to minimize non-specific binding. We expect the detected fluorescent markers will mirror known cellular and tissue expression patterns, confirming that the organoids exhibit normal human fetal neurodevelopmental characteristics and are biologically relevant for GBM modeling. Future work will expand marker validation to include GFAP (astrocytes), DCX (neurogenesis marker), TBR2 (intermediate progenitors), OLIG2 (oligodendrocyte progenitors), PTPRZ1 (radial glia), IBA1 (microglia) and other cell lineage-specific markers. Establishing reliable staining and imaging conditions is a crucial step toward developing our organoid model to be suitable for exploring GBM tumor biology and potential therapeutic responses.

Spinal cord injury (SCI) is a destructive neurological and pathological state that causes major motor, sensory and autonomic dysfunctions with an estimated global rate between 250,000 and 500,000 individuals every year. Many therapeutic strategies have been proposed to overcome neurodegenerative events and reduce secondary neuronal damage. Available treatments are limited and only provide supportive relief to patients with lifetime disability. The severity of impairment is related to the function of the remaining viable neural resources since the central neurons cannot yet be repaired or replaced, only reorganized.   Use-dependent movement therapies have been proven to increase neuronal plasticity. In addition, electrical stimulation can directly induce neuronal plasticity, enhancing therapeutic efficacy. Using a well-known rat model of Acute Spinal Cord Injury (ASCI) available in our laboratory, we hypothesized that targeted, activity-dependent spinal stimulation (TADSS) with physical retraining enhances motor recovery after SCI by facilitating and directing intrinsic synaptic plasticity in specific spared motor circuits below SCI. Long-Evans rats will undergo training and testing for pellet reaching four-legged assessment test, and CatWalkXL test for 4 weeks followed by a moderate to severe unilateral dorsal spinal contusion at the C4/C5 border ipsilateral to the dominant forelimb, resulting in a marked and persistent inability to extend the elbow, wrist, and digits for injured group. Following injury, a neurochip is implanted which delivers closed-loop electrical stimulation below the lesion point throughout the weekdays of training (for 6-8 hours per day). All groups will resume training for another 40 weeks and data will be collected and analyzed. Based on our initial data, we expect to prove that electrical stimulation combined with physical training improves the functional recovery of limb use after acute unilateral spinal cord injury.  

Understanding how blood flow can be influenced by the use of drag-reducing polymers (DRPs) is crucial for addressing secondary injury mechanisms in spinal cord injuries (SCI). SCI disrupts spinal blood flow due to increased intraspinal pressure, altered vascular topology and increased resistance, exacerbating hypoperfusion resulting in hypoxia additional cell death. Thus, mitigating the impact of these secondary mechanisms is critical for better outcomes. We hypothesize that DRPs may reduce vascular resistance by reducing turbulent flow in injured spinal cords; specifically, by reducing the effect of flow separation in larger vessels. The major experiments of this study are to (1) test multiple DRP concentrations to find optimal restoration of hemodynamics after injury and (2) to design 3D models of in-vivo vasculature structures based on ultrasound scans. We have currently tested 2 different DRP concentrations and determined an ideal injection volume in a non-injured rat to increase blood flow—we hope to further these experiments via an injury model and analyzing effects of DRP. Hemodynamic analyses will be conducted from contrast-enhanced ultrasound (CEUS) scans at baseline, post-DRP injection at 30, 60, 90 minutes where a microbubble bolus injection will be delivered. Specifically, we will examine arrival time delay (ATD) which represents relative vascular resistance and area under the curve (AUC) which represents total blood flow volume. Preliminary results showed improvement of flow attributed to the DRP injection (~ 15-20% decrease and increase in ATD and AUC respectively). I will also design 3D models of intraspinal vessels informed by imaging and bioinks to explore blood flow behavior in controlled in vitro settings. Combined, these studies will serve to understand how DRPs can be effective as mitigating secondary injury mechanisms of SCI and improve recovery