



Oral Presentations
Dr. Jianhua Xiong
Assistant Professor
Department of Urology, Emory University School of Medicine
Endothelial Fatty Acid β-oxidation Prevents Endothelial-to-Mesenchymal Transition and Kidney Fibrosis
Kidney fibrosis is a central pathological feature of chronic kidney disease (CKD) and a major contributor to progressive renal dysfunction in urologic disorders. Emerging evidence highlights endothelial dysfunction and endothelial-to-mesenchymal transition (EndoMT) as key drivers of fibrotic remodeling. Endothelial fatty acid β-oxidation (FAO) plays an essential role in maintaining cellular homeostasis, yet its contribution to endothelial phenotypic stability and renal disease progression remains poorly defined. We hypothesized that endothelial FAO functions as a metabolic checkpoint that preserves endothelial identity and prevents EndoMT and kidney fibrosis.
Methods
To test this, we generated mice with endothelial-specific deletion of carnitine palmitoyltransferase 2 (Cpt2), a key enzyme required for mitochondrial FAO. Renal structure and function were evaluated using histopathology, collagen deposition assays, and measurements of urinary protein excretion. EndoMT was assessed by examining the expression of endothelial and mesenchymal markers in renal tissues and isolated endothelial cells. Metabolic analyses, including ATP and acetyl-CoA quantification, were performed to define the underlying metabolic alterations.
Results
Endothelial FAO functions as a metabolic checkpoint that preserves endothelial identity and limits EndoMT and renal fibrosis. Disruption of this pathway promotes kidney injury and structural remodeling, highlighting endothelial metabolism as a critical regulator of non-malignant renal disease processes.
Dr. Razvan Voicu
Assistant Professor of Robotics and Mechatronics Engineering
Kennesaw State University
A Home-Like Multimodal Sensing Platform for Measuring Balance, Mobility, and Functional Performance Metrics
Balance, mobility, and functional performance are central indicators of independence, rehabilitation status, and the ability to safely participate in daily life. However, these measures are often captured through brief clinical observation or laboratory-based assessment, which can limit understanding of how movement is performed in environments that more closely resemble the home. This work presents a compact multimodal sensing platform for measuring movement in a residential-style research setting.
The device integrates vision-based sensing and onboard processing to capture human movement locally and convert raw observations into structured summary metrics related to balance, mobility, and functional performance. The platform was developed to operate as a practical measurement tool with a small deployment footprint, repeatable setup, and local data-processing workflow. Emphasis was placed on capturing movement in context while reducing dependence on continuous manual observation or complex participant instrumentation.
Preliminary implementation demonstrated that the platform could be installed, calibrated, and operated in a home-like environment while maintaining stable local processing. The system successfully captured movement data during repeated functional activities and generated organized outputs suitable for reviewing movement consistency and performance across sessions. Importantly, the focus is on practical measurement rather than diagnosis, enabling the platform to serve as a research instrument for comparing movement over time and across conditions.
These findings support the feasibility of using a compact, home-like sensing platform to collect objective movement data outside traditional clinical assessment spaces. The work contributes an early measurement framework that may complement existing rehabilitation assessments by providing additional context on how individuals move in realistic environments. In future studies, this approach can support Veteran-centered rehabilitation research, assistive technology evaluation, and longitudinal monitoring of functional change. By focusing on practical deployment and interpretable movement metrics, the platform helps bridge the gap between clinic-based assessment and real-world function.
Dr. Matthew Ryan Smith
Health Research Scientist; Assistant Professor of Medicine
VA Healthcare System of Atlanta; Division of Pulmonary, Allergy, Critical Care, and Sleep Medicine at Emory University
Metabolome-wide association study of firearms, pyrotechnic and smokescreen-derived particulate matter in mouse lung explants
Background: Active military personnel are exposed to a range of inhalants that may elevate inflammatory biomarkers and cytokines, but the links to metabolic and mitochondrial dysfunction remain unclear. Elemental analysis of these exposures identifies varying components, including redox-active metals, which are known to disrupt redox balance; therefore, we sought to answer whether these exposures were mitochondrial toxins. Hypothesis: Exposure to particulate matter derived from firearm discharge or ordnance causes significant metabolic disruption and is associated with significant inflammatory processes.
Methods: High-resolution metabolomics analysis using dual column and electrospray ionization (RP-C18-/ HILIC+) was performed on the lungs of female CD-1 mice (n=12) that were exposed to a single dose (100 µg) of previously extracted particulate matter from military ordnance or from firearm discharge (20 µg) via oropharyngeal aspiration and euthanized at 4 and 24 hours post-exposure. Metabolome-wide association studies (MWAS) were conducted using LIMMA with the Benjamini-Hochberg false discovery rate correction (0.2), implemented in the R package xmsPANDA. Pathway analysis using mummichog v2 was conducted, and metabolite annotations were performed using xmsAnnotator. Inflammatory biomarkers and cytokines were measured in the BALF fluid for lung toxicity markers.
Results: MWAS of the military exposures revealed metabolic perturbations associated with each exposure. Pathway analysis revealed significant perturbations to fatty acids, redox imbalance, and mitochondrial dysfunction compared with the negative controls. Bubble plot analysis revealed profiles for each exposure in lung tissue, suggesting distinct metabolic dysregulation significantly associated with redox imbalance and inflammatory signaling.
Discussion: Military service personnel are subject to a host of unique exposures during both training and active combat situations, such as exposure to ordnance and firearm discharge. Our preliminary findings show that significant metabolic perturbations remain in the lungs of mice exposed to these agents, suggesting acute inflammatory stress and redox imbalance. Future studies should focus on the underlying molecular mechanisms of mitochondrial toxicity, including whether these perturbations can be remediated.
Dr. Aimee Bozeman
Postdoctoral Fellow
Translational Neurotrauma Research Laboratory
Emory University
Department of Emergency Medicine
A Clinically Grounded Porcine Model of SCI Induced Neuropathic Pain Reveals Human Like Sensory Phenotypes Through Quantitative Sensory Testing
Neuropathic pain (NP) is one of the most debilitating and treatment resistant consequences of spinal cord injury (SCI), affecting an estimated 50–80% of individuals living with SCI. Progress toward effective therapies has been hindered by a heavy reliance on rodent models, which despite their value in uncovering mechanistic underpinnings, primarily assess pain through reflexive withdrawal and fail to capture the supraspinal, perceptual dimensions that define the human NP experience. These limitations, combined with substantial anatomical, physiological, and pharmacokinetic differences between rodents and humans, have contributed to the persistent translational gap in SCI NP research. To overcome these barriers, we developed a reverse translated porcine model of SCI induced neuropathic pain (SCI NP) that enables the use of quantitative sensory testing (QST) and supraspinal pain assessment paradigms directly adapted from human clinical practice. Pigs offer a uniquely human relevant platform due to their comparable spinal cord size, dermatomal organization, metabolic profile, and neurophysiology, making them ideally suited for evaluating sensory perception and higher order pain processing. Using reverse translated QST protocols in male and female adult Yucatan minipigs with midthoracic SCI, we assessed both reflexive and supraspinal responses to mechanical and thermal stimuli. We found that pigs develop robust, human like NP phenotypes, including below level mechanical allodynia, with increased sensitivity to normally non painful stimuli; altered thermal thresholds, including heightened heat sensitivity and reduced cold tolerance; and enhanced wind up ratios, indicating amplified temporal summation of pain. These results show that pigs exhibit sensory abnormalities strikingly similar to those reported by individuals with SCI NP, supporting the model’s potential to more accurately predict clinical efficacy of emerging analgesics. By integrating clinically grounded QST measures with a large animal SCI platform, this porcine model addresses critical shortcomings of rodent research and provides a powerful translational bridge for accelerating the development of effective therapies for neuropathic pain. This work is supported by a VA BLRD Merit Award (I01BX005203) and the NIH HEAL Initiative (RF1NS135504).
Dr. Asim Gazi
Postdoctoral Fellow in the Departments of Computer Science and Statistics at Harvard University
Incoming Assistant Professor in the Department of Biomedical Engineering at Georgia Tech and Emory University
Characterizing the Stress-Reducing Effects of Non-Invasive Vagus Nerve Stimulation During Trauma Recall in Posttraumatic Stress Disorder
Background: Individuals with posttraumatic stress disorder (PTSD) recall traumatic memories in everyday environments, precipitating debilitating episodes of acute stress that occur outside the context of clinical care. Non-invasive vagus nerve stimulation (nVNS), delivered cervically or auricularly, can counteract acute stress by upregulating the parasympathetic ("rest and digest") branch of the autonomic nervous system (ANS) and downregulating the sympathetic (“fight or flight”) branch of the ANS. However, the precise time course of nVNS effects on physiological markers of the ANS remains poorly understood, impeding the design and implementation of wearable nVNS systems that can provide just-in-time support for acute stress.
Objective: To develop personalized predictive models capturing how nVNS and trauma recall elicit underlying changes in the ANS, and to accordingly simulate the potential benefit of just-in-time nVNS delivery during trauma recall.
Methods: We analyzed data from a double-blind, randomized, sham-controlled clinical trial of nVNS in 50 participants with history of prior psychological trauma, 24 with PTSD. Person-specific linear state-space models were trained (~3,500 s of data) and tested (~2,000 s of held out data) per participant to predict nine physiological markers' dynamics. Validated models were used to simulate just-in-time nVNS (and sham stimulation), activated 30 seconds into a trauma recall episode.
Results: Our models significantly outperformed baseline forecasting methods in predicting future physiological dynamics (P<.05; N=50). For the models personalized to participants with PTSD, just-in-time nVNS significantly attenuated—and briefly reversed—the stress response, compared to no attenuation observed for sham stimulation. This difference was statistically significant 15 s after the onset of stimulation (P=.04; N=24).
Conclusions: Our modeling approach captures underlying ANS dynamics due to increases and decreases in acute stress associated with trauma recall and nVNS, respectively. Just-in-time nVNS has the potential to rapidly counteract the physiological stress response to trauma recall for individuals with PTSD.