Within the field of anesthesia, the process of arousal following general anesthesia is still little understood. Of particular concern is the way a state of pain can modulate arousal, with the nucleus accumbens (NAc) as a brain region of interest as it serves many functions including controlling mood, pain states, and reward motivation. This study investigates how NAc principal cells change their firing during arousal and are influenced by a pre-existing pain state. To assess this, we leverage the partial sciatic nerve ligation model of chronic neuropathic pain to examine the influence of pain state on arousal. We then test groups of mice in their behavioral responses, including open field test, von Frey, hot plate test, and the return of righting reflex as a measure of arousal from isoflurane anesthesia-induced unconsciousness. In some mice, we record 1-photon calcium-related population activity in the NAc to analyze neural activity during arousal from anesthesia. Our findings will serve to illuminate the underlying brain circuitry involved in arousal from anesthesia and the influence of pain state, which can help improve anesthesia recovery and may reveal non-opioid or endogenous mechanisms for pain relief.

Despite the widespread use of general anesthesia, our understanding of mechanisms by which anesthetics and analgesics induce unconsciousness remains limited. This study used a transgenic mouse model (FosTRAP2) to investigate neural circuits that are active during isoflurane-induced anesthesia. FosTRAP2 mice were retro-orbitally injected with an AAV-PHP.eB virus expressing Cre-conditional DREADDs (designer receptors engineered to be activated by designer drugs), which was followed by general anesthesia exposure, where 4-hydroxytamoxifen was injected to chemogenetically label isoflurane-activated cells brain-wide with DREADDs. We subsequently implanted the mice with wireless mechano-acoustic (MA) devices to record peripheral physiologic data such as heart rate, respiratory rate, temperature, and physical activity. To determine the functional impact of isoflurane-activated circuits, chemogenetic reactivation was induced via clozapine-N-oxide (CNO) injection, with concurrent video recording for pose estimation with Social LEAP Estimates Animal Poses (SLEAP), and behavior classification with Simple Behavioral Analysis (SimBA). Combining both physiologic data and machine learning provided us with an experiment pipeline that allowed us to better study brainwide isoflurane-activated neural ensembles. We found that reactivation of these circuits led to a significant reduction in heart rate, body temperature, physical activity, accompanied by a reduction in typical active behaviors, such as grooming and rearing. By gaining a deeper understanding of how general anesthetics alter neural circuits, we hope to uncover the complex relationships between brain activity and consciousness, with important implications for improving anesthetic practices and developing novel sedatives or analgesics in the future.

The current understanding of mechanisms of anesthesia and the function of anesthesia-activated brainwide neural circuitry is very limited. Additionally, there is an urgent need to develop new non-opioid analgesic drugs, and targeting anesthesia neural circuitry could provide a novel path to pain relief. To investigate the function of this circuitry, we use a brainwide approach to perform chemogenetic manipulations in a FosTRAP2 transgenic mouse model. Briefly, FosTRAP2 (Fos-2a-Cre) mice receive retroorbital injections of a Cre-dependent virus expressing chemogenetic DREADDs. Mice then undergo isoflurane anesthesia exposure and midway through the exposure, they receive an intraperitoneal injection of 4-hydroxytamoxifen to induce activity-dependent chemogenetic labeling of isoflurane-activated brainwide ensembles. I use behavioral analysis pipelines to analyze how the activation of these ensembles affects thermal nociceptive processing after mice are induced into a lightly anesthetized state and subjected to analgesia testing. Mice underwent the warm water tail withdrawal and hot plate assays. I then use a combination of manual annotation and pose estimation approaches with supervised machine learning using Social LEAP Estimates Animal Poses (SLEAP) followed by Simple Behavioral Analysis (SimBA) to provide insight into behavioral signatures and classifications. I identify a number of occurrences for behaviors such as tail withdrawal, latency to jump, and paw grooming, which is used to infer thermal anti-nociception in open field testing. I also helped develop five distinct behavioral classifiers: rearing, grooming, freezing, circling, and Straub tail response. With the resulting behavioral analysis, I can investigate how targeting brainwide anesthesia-activated neural ensembles produces anti-nociception. Anesthesia, although used for many common procedures, is not widely available to the general public and must be administered by a medical professional. Understanding the mechanisms behind its effect on pain processing is a gateway for revolutionary research that could potentially eliminate the need for opioid medications in the future.

This research project focuses on developing and enhancing an AI auditing system to assess diversity and fairness in large language modeling (LLMs) systems. By replicating an existing Python-based audit framework, originally created by my Principal Investigator (PI), this study extends its functionality to specifically evaluate how race and ethnicity are represented in AI-generated outputs related to professional occupations. The enhanced auditing system cross-references race and ethnicity data with job positions to identify potential biases, providing a deeper understanding of whether AI systems (specifically GPT-4) disproportionately associate certain ethnic groups with specific professions. These findings contribute to the ongoing discourse on fairness in AI, offering insights into how LLM models may perpetuate or mitigate biases in career representation. This research is critical for the development of more equitable AI systems that reflect diversity across various social and professional contexts, highlighting the importance of fairness in the deployment and usage of AI technology.

The sensation of acute pain is fundamental to survival, indicating tissue damage that motivates an animal to engage in adaptive protective behaviors. Chronic pain, however, is persistent pain beyond typical recovery window and serves little adaptive function. The negative emotional component inherent in chronic pain contributes to the development of comorbid psychiatric disorders such as depression, social aggression, and social withdrawal. Our research aims to understand the bidirectional relationship between pain and social behavior, by evaluating mechanical sensitivity and changes in social motivation, reward, and interaction following a neuropathic injury. Using social self-administration (SSA), pair-housed mice were placed in operant chambers and underwent voluntary lever press trials for the reward of social interaction with their cage mate. Mice also underwent mechanical hypersensitivity response assays called von Frey where increasing weights of plastic filament were applied to the hind paw. Following baseline von Frey testing and the acquisition of the SSA task, mice then received a spared nerve injury (SNI) to induce neuropathic pain. After surgery recovery, mice were returned to the lever press and von Frey trials at different post-operative windows. Pain sensitivity was determined by the filament weight in which the animal withdrew their paw during von Frey. Changes in social behavior were measured via changes in lever press frequency and interactions during trials. Behavior changes were quantified using Simple Behavior Analysis (SimBA) machine learning to classify interactions during social trials. Once the trials were completed, brain tissue from regions associated with reward and social neural circuitry was collected and investigated using transcriptomic methods. Our data found sexually divergent social adaptations and gene expression following chronic pain. Future experiments will further delineate these sex-specific adaptations following a traumatic injury. This research can inform social intervention as an adjunct or alternative treatment to pharmacological pain intervention and its comorbidities.