Traumatic spinal cord injuries (SCI) are a catastrophic type of injury that disrupt many functions of daily life such as mobility, temperature regulation, sexual function, and bowel/bladder use. Approximately 80% of individuals who suffer from SCI develop neurogenic bladders, and subsequently require catheterization. Patients with sacral SCI have different neurogenic bladders than those with higher injuries; specifically, severing the peripheral innervation at the sacral level causes a non-contracting and low pressure bladder, without the scar tissue found commonly in higher injuries. This suggests that peripheral denervation decreases the incidence of non-compliant bladder complications. Bladders primarily have two uses - storage and voiding during urination. One common strategy to improve the storage function of neurogenic bladders is to denervate them with direct injections of Onabotulinumtoxin-A (ONA). However, this treatment is only applied after other treatments have failed, leaving time for significant damage to occur. In the current study, we propose to study the effects of direct ONA injections acutely after SCI. Our two aims are 1) to determine the treatment window for chemodenervation that best reduces bladder wall hypertrophy and fibrosis, and 2) to determine whether bladders remain compliant after chemodenervation. Using a rat model, we applied two treatments - SCI + Saline, SCI + ONA, with both saline and ONA directly injected into the bladder wall. A control group received laminectomy without injury. After treatment, we plan to measure bladder function using a urinary incontinence scale, as well as histological measurements of collagen deposition and muscle area to quantify bladder wall hypertrophy and fibrosis. We anticipate that early ONA injection (as defined in Aim 1) will be associated with less fibrous and more compliant bladders in the long term when compared with non-chemodenervated bladders. If our hypotheses are correct, we plan on getting IRB approval for a clinical trial.

Breathing rhythms are initiated in the ventrolateral medulla, in a region known as the preBötzinger complex (preBötC). The preBötC generates normal inspiratory activity and breaths of greater volume called sighs. Sighs prevent the spontaneous collapse of alveoli in the lungs, termed atelectasis, and mice that lack the ability to sigh die of atelectasis after birth. For many years, respiratory physiologists have tried to understand how one network, the preBötC, is able to generate two rhythms with distinct timing mechanisms. In order to test this hypothesis that glia drive the sigh rhythm through purinergic signaling, I will make recordings of neuronal population activity from the preBötC in the horizontal brainstem slice preparation, and apply caged ATP. Brief light pulses will uncage the ATP. We predict that ATP and its conversion to ADP will bind to purinergic receptors to facilitate calcium spread and recruit neurons along the inspiratory column to generate these large breaths. Then, I will apply the purinergic receptor antagonist that we have established is effective in blocking sighs. We observed this result in a preliminary experiment, but more data needs to be collected to confirm this relationship. Next, we will confirm that glial increases in intracellular calcium coincide with sigh generation. We will perform calcium imaging using two photon microscopy in the horizontal slice preparation isolated from transgenic cre mice that express GcAMP6f driven by the astrocyte-specific promoter, Aldh1l1. We predict that calcium activity in astrocytes will correspond with the sigh rhythm, and that we can inhibit sighs by applying purinergic antagonists.

Traumatic spinal cord and brain injuries may lead to a devastating loss of neurological function. After the traumatic primary insult, a secondary injury phase ensues, which greatly increases the extent of the injury. Inflammation is a dominant component of secondary injury, which includes the immune response in which free radicals and proinflammatory cytokines are released that induce the death of surrounding neurons. Dexamethasone, a commonly used corticosteroid, has been shown to produce a neuroprotective effect by inhibiting inflammation and reducing cytokine release. However, the clinical application of large systemic doses of steroids is limited by side effects, such as sepsis and pneumonia. For this reason, a localized microneedle delivery system may be favored to allow for a therapeutic dose at the injury site, with reduced side effects. The goal of this work is to design a local, controlled drug release system, in which the drug could be given for a sustained duration at a therapeutic dose at the site of injury. Using a mixture of polyvinylpyrrolidone (PVP) and poly (ethylene glycol) diacrylate (PEG-DA), a biodegradable microneedle array can be formed that can encapsulate dexamethasone that then slowly release into the tissue over 72 hours. With a 3D printed system, various sizes can be quickly developed and produced to create customizable arrays for various target locations. Using a 2-photon polymerization laser, Nanoscribe, arrays can be formed on a scale of <300 μm as well as larger arrays using traditional printers. The needles are developed using a three-stage molding method, casting customized 3D printed arrays in a silicon elastomer, PDMS, and polymerizing our monomer-drug solution to the desired shape. Current results indicate sustained release over the target window and are currently branching into trans-membrane in-vitro studies.

Recent work published by my colleagues and I showed that diagnostic ultrasound, applied directly over the visual cortex of human participants, increased the likelihood that they would observe visual effects while looking at a visual target, with that likelihood increasing over the course of the experiment. However due to a lack of EEG data, it is impossible to know the biophysical mechanisms and neural pathways that generate the observed effects on brain function. To better understand the full effect and mechanism of our initial findings, I helped develop a surgical protocol for a mouse model allowing for the collection of EEG data while exposing the animal to a combination of light and diagnostic ultrasound stimuli. EEG data was collected from 5 mice using a stimulus paradigm that we believed would increase the rodent's susceptibility to light stimulus. I utilized Matlab for processing, visualization, and statistical analysis of the data to determine the effects and hypothesize potential biophysical mechanisms of the stimuli. My analysis focused on determining whether the ultrasound stimulus successfully increased the susceptibility of the visual cortex, and which brain frequency bands were modulated by the stimulation.

Traumatic cervical spinal cord injury (SCI) results in a wide range of outcomes from partial paralysis to complete tetraplegia depending on the location of injury along the length of the cervical spinal cord. Importantly, there is a high density of motor neurons in the cervical region which are involved in important motor outputs such as breathing and hand function. The present study aims to minimize the secondary damage to the spinal cord after the primary insult, by addressing two substantial contributors to neuron death: first, surgical decompression is conducted to reduce local tissue swelling after injury, and second, administration of the metabolite oxaloacetate (OAA) to minimize excitotoxicity by stimulating glutamate transport away from injured neurons. We hypothesized that animals treated with decompression, OAA, or both, would have increased neuronal survival, general tissue sparing, and improved behavioral outcomes than those without treatment, and that combined treatment would be more effective than each individual treatment. We tested the treatments using a clinically relevant rat model for bilateral, moderately severe cervical spinal cord injury. Following treatments, we determined effectiveness by assessing animals’ forelimb function and quantifying motor neuron and white matter sparing in the injured tissue. Results consistent with the hypothesis would have meaningful impacts for future cervical SCI patients, as even a limited increase in tissue sparing in the cervical region have profound functional outcomes for patients’ independence and opportunities. Future studies will work to visualize parameters for segmental tissue at risk after acute injury in order to specify each patient’s treatment and maximize their opportunities for recovery.

Hydrocephalus is a condition in which the cerebrospinal fluid (CSF) accumulates in the brain, increasing pressure inside the skull. A common method of treatment in pediatric patients is the placement of a shunt inside the fluid-filled ventricle of the brain. Overtime, shunts will become obstructed, fail to work, and require brain surgery and reimplantation of a new shunt. We will apply a intense focus ultrasound to send a pulse through the catheters obstructed by astrocyte tissue cultures in a three-dimensional model for shunt failure. When the method has been refined, we plan to apply ultrasound on catheters clogged from cerebrospinal fluid (CSF) that will be obtained from shunt replacement surgeries performed at Seattle Children’s hospital. Our aim is to improve the flow rate through the catheter, thus providing a less invasive method for shunt clearing in pediatric patients with hydrocephalus.

Human cerebral organoids (HCOs, also known as mini brains) generated from induced pluripotent stem cells (iPSCs) provide exciting opportunities to study neurological disorders in ways not previously possible from traditional animal, brain slice, or 2D cell culture models commonly used in research. Recent studies have demonstrated that HCOs can reproduce complex neural development pathways and mimic several biomarkers of neurodegenerative disorders. However, it remains uncertain if these HCOs can model the characteristic electrophysiological activities of epilepsy. We have developed a technique to record electrocorticography (ECoG) directly from these HCOs to first characterize their electrographic behavior as a step toward evaluating their potential as a clinically relevant model of epilepsy. We aimed to determine the baseline ECoG characteristics of HCOs and the changes in ECoG characteristics after exposure to common proconvulsants with different modes of action in order to assess their ability to model seizure activity. We recorded ECoG while the HCOs were bathed in normal artificial cerebrospinal fluid (ACSF) and while they were in ACSF solutions containing pentylenetetrazol (PTZ, a GABA A receptor antagonist), kainic acid (KA, a glutamate receptor agonist), or an elevated bath potassium concentration, which induces neuronal hyperexcitability. We observed an increase in spikes and seizure-like activity after high potassium and KA exposure. In contrast, there was no change in activity associated with PTZ exposure. These findings are consistent with the fact that our HCOs are primarily composed of excitatory neurons, with only a minimal number of inhibitory interneurons. Developing reliable brain organoid models of epilepsy is an important next step in the field that will revolutionize studies of precision therapy for epilepsy.

 Iaccarino et al exposed one hour of light flickering at 40Hz to awake 5XFAD Alzheimer's Disease (AD) mouse models, consequently generating action potentials at 40 Hz and activating mircoglia. Consequent colocalization of microglia with Aß plaque actuely and clearing of Aß plaque after seven days was observed, but only in the visual cortex. We hypothesized transcranially delivered, near diagnostic ultrasound (tnDU) can replicate the results of Iaccarino et al but throughout its area of application, thus not limited to the visual cortex. We exposed sedated 5XFAD mice to tnDU at 40Hz with 400microsecond-long pulses for one hour, targeting one hemisphere of brain centered on its hippocampus. Chronic studies targeted comparable brain in each hemisphere for one hour/day for five days. Histology and EEG recoding revealed acute application of tnDU activated more microglia that colocalized with Aß plaque, relative to the contralateral hemisphere of treated brain with brain activation at 40Hz. Chronic application reduced their Aß plaque burden by nearly half relative to paired sham animals. Our results compare to those of Iaccarino et al but throughout the area of ultrasound-exposed brain. Our results also compare to those achieved by medications that target Aß for a substantially shorter period of time. The proximity of our ultrasound protocol to those shown as safe for non-human primates and humans may motivate its rapid translation to human studies. 

Leigh syndrome (LS), the most common type of pediatric mitochondrial disease, has been associated with loss-of-function mutations in genes that encode for proteins in complex 1 of the electron transport chain. Mutations in a gene called NADH dehydrogenase (ubiquinone) iron sulfur protein 4 (NDUFS4) are linked to LS in several groups of patients. Mice carrying a whole body or central nervous system-specific deletion of this gene develop several symptoms reminiscent of those found in humans. Recent studies in our lab showed that knockout (KO) of Ndufs4 in GABAergic interneurons are the key driver of seizures in the mouse model. In this study, we investigated which subtypes of interneurons are the leading cause of epilepsy. We focused on two major subtypes of GABAergic interneurons: interneurons that express parvalbumin (PV), a high-affinity calcium binding protein, and interneurons that express somatostatin (SST), a neuropeptide. PV interneurons make up 40% of GABAergic cells and have faster spike firing patterns when compared to SST cells. We hypothesize that PV neurons will contribute more to seizure susceptibility than SST interneurons because mitochondrial dysfunction is known to affect cells with high firing rates due to their high energy demand. To test our hypothesis, we generated mice with Ndufs4 KO restricted to PV or SST interneurons using LoxP Cre technology. We evaluated the general health of mice and tested their susceptibility to thermal seizures through behavioral checks and thermal induction tests. We found that Ndufs4 KO in both PV and SST interneurons led to seizure phenotypes. However, this gene KO did not have any detectable effect on body weight, motor activity, breathing patterns, physical appearance, and limb extension. A better understanding of mechanisms underlying epileptic seizures will lead to the development of effective treatments for LS-related epilepsy.