Competition Abstracts

Poster 1  (Post-Candidacy)

Targeting MET–LGR5 Crosstalk using an Antibody-Drug Conjugate Combination with Diverse Payloads to Overcome Adaptive Resistance in Colorectal Cancer  

Shraddha Subramanian^1,2, Zhengdong Liang¹, Peyton C. High^1,2, Cara Guernsey-Biddle^1,2, Adela M. Aldana¹, and Kendra S. Carmon^1,2
1Center for Translational Cancer Research, The University of Texas Health Science Center, Houston, TX, USA; ²The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX, USA

Off-target side effects and drug resistance often hamper the therapeutic benefits of existing anti-cancer therapies against colorectal cancer (CRC), the second deadliest malignancy worldwide. CRCs contain tumor-initiating cancer stem cells (CSCs) that survive toxic drugs, thereby producing mortality-inducing metastases. Antibody-drug conjugates (ADCs) leverage monoclonal antibody (mAb) specificity to deliver cytotoxic payloads to cancer cells. We generated ADCs targeting leucine-rich repeat-containing G protein-coupled receptor 5 (LGR5), a bona fide CSC biomarker frequently overexpressed in CRCs. Our LGR5 ADC exhibited high potency in CRC xenograft models with minimal toxicity; however, it failed to prevent tumor relapse after treatment cessation. Interestingly, our findings demonstrated that recurrent tumors evade LGR5 ADC-mediated elimination through transient conversion into an LGR5-negative (LGR5^–) state, accompanied by concomitant MET upregulation and downstream pathway activation. MET is a receptor tyrosine kinase frequently upregulated in CRCs and promotes metastatic progression. Thus, this study aims to elucidate the mechanisms of MET-mediated resistance to LGR5 ADCs and to evaluate the therapeutic efficacy of combining MET-targeted therapies with next-generation LGR5 ADCs in preventing the recurrence and metastasis of CRC. We identified that the activation of STAT3, a downstream effector protein in the MET pathway, undermines LGR5 ADC efficacy in CRC cells. To eliminate drug-resistant LGR5^– cells, we generated MET-targeted ADCs by conjugating a highly selective MET mAb (ABT700) backbone to the pyrrolobenzodiazepine (PBD) dimer using a site-specific conjugation methodology. Our MET ADC (ABT700-PBD) demonstrated dose-dependent cytotoxicity in CRC cells. Furthermore, ABT700-PBD did not affect CRC cells with genetically induced MET ablation, demonstrating its specificity. Safety studies in immunocompetent mice showed that ABT700-PBD exhibited a favorable safety profile. Moreover, in CRC patient-derived xenograft (PDX) models, ABT700-PBD induced marked tumor regression. However, in tumors that recurred following dose-dependent ABT700-PBD monotherapy, LGR5 protein expression was upregulated. Together, these findings suggest that MET and LGR5 function within an epistatic pathway that governs cellular plasticity and therapeutic resistance in CRC. Cytotoxicity assays evaluating the combination of ABT700-PBD with our next-generation LGR5 ADC showed additive anti-proliferative activity in vitro. The LGR5 ADC was produced by attaching a previously characterized LGR5 mAb (8E11) backbone to a camptothecin (CPT2)-derived payload. Combination treatment with ABT700-PBD and 8E11-CPT2 in CRC PDX models extended survival compared to LGR5 or MET ADC monotherapies. These results provide a preclinical framework supporting combination therapy with MET and LGR5-targeted ADCs to overcome resistance-induced tumor relapse in advanced CRC.

Poster 2 (Post-Candidacy)

Rapid Progressors in Alzheimer’s Disease: Differences in Cognitive Domain Decline

Madison Shyer¹, Wentao Li, PhD², Xiaotian Ma, PhD², Kristofer Harris, MPH, RN¹, Livia Merrill, PhD¹, Yejin Kim, PhD², and Paul E. Schulz, MD^1
1The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX, USA
²UTHealth Houston McWilliams School of Biomedical Informatics, Houston, TX, USA

Background: One of the most pressing challenges in Alzheimer’s disease (AD) research is the unpredictable variability in disease progression across patients. Identifying different progression patterns, including a subset of patients who rapidly decline (rapid progressors, RPs), is essential for prognosis and clinical trial design. Furthermore, it is possible that a different underlying pathology differentiates a RP from normal disease progression and can be assessed with changes in six cognitive and functional domains: social cognition, complex attention, executive function, perceptual-motor function, language, and learning & memory. This study explores whether changes in these cognitive domains and early change trajectories can differentiate RPs from non-RPs to better understand differences in disease progression.

Methods: Using placebo-arm data from three EXPEDITION trials, RPs were defined as the 10% of patients with the greatest decline over 80 weeks on four cognitive tests: the Mini-Mental State Exam (MMSE), Clinical Dementia Rating Scale – Sum of Boxes (CDR-SB), Alzheimer’s Disease Assessment Scale-Cognitive Subscale (ADAS-Cog) and the Alzheimer’s Disease Cooperative Study Activities of Daily Living (ADCS ADL). Demographics, and clinical characteristics were compared via Chi-squared, Fisher’s exact test, or ANOVA. Deep survival models were trained at four time points (4, 12, 28, and 80 weeks) to predict probability of RP status at 80 weeks. Model performance was assessed with Brier score, AUROC, AUPRC, and compared with Cox proportional hazards models. Feature importance was quantified via permutation‑based attribution to identify cognitive domains contributing to risk discrimination.

Results: A total of 1603 patients were included; 150 were classified as RPs by the CDR-SB definition. Survival models successfully stratified patients into RP or non-RP across timepoints (AUROC range: 0.930-0.951, AUPRC range: 0.751-0.865). Early decline in social cognition at 12 weeks and executive function at 28 weeks significantly and most strongly differentiated RPs from non-RPs on a clinically-relevant timescale for intervention. The ADAS-Cog14 and CDR-SB models showed clearer separation of progression risk over time as opposed to MMSE and ADCS-ADL models, reflecting the cognitive versus functional emphasis of each scale.

Conclusions: This study shows early (12-28 week) changes in specific cognitive domains carry significant predictive value for rapid progression. Deep survival modeling applied to RCT data can identify individuals at high risk of rapid progression and highlight domain-specific predictors; predictive models unique to this subset of patients are therefore necessary when evaluating prognostic biomarkers or stratifying randomization in clinical trials. Further analyses correlating cognitive domain decline and its relationship with neuroanatomical changes in RPs will determine how domain-specific decline relates to underlying neurobiology and biochemical changes, further elucidating the combination of factors that result in a RP.

Poster 3 (Masters)

Hepatic Sexual Dimorphism in Cerebral Amyloid Angiopathy

Shivi Garg^1,2, Lillianne Liu², Danye Jiang, PhD², Louise D. McCullough, MD, PhD^1,2
1The University of Texas MD Anderson UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX, USA
²UTHealth Houston McGovern Medical School, Houston, TX, USA

Background: Cerebral amyloid angiopathy (CAA) is a form of vascular dementia characterized by accumulation of amyloid-beta (Aβ) in the walls of cerebral and leptomeningeal blood vessels, leading to vascular dysfunction, neuroinflammation, and cognitive impairments. Impaired clearance of Aβ, especially by peripheral organs, is hypothesized to be a major driver of amyloid-related pathologies. The liver is the primary peripheral organ involved in Aβ clearance and hepatic inflammation or dysfunction is associated with exacerbated CAA pathology and progression. Preliminary data from preclinical models has suggested sexual dimorphism in hepatic inflammation and lipid regulation, but the exact progression of these alterations and their impact on cerebral pathology remains unknown. Thus, the goal of this project is to define the progression of sexually dimorphic changes in hepatic inflammation across several stages of CAA pathology using a transgenic mouse model.

Hypothesis: We hypothesize that females will exhibit earlier development of hepatic inflammation and dysfunction, compared to males. This is expected to lead to a greater cerebral amyloid burden and worse performance on tests of cognition and memory.

Methods: To answer these questions, tissues will be harvested from mice of both sexes at various stages of pathology (ranging from early to advanced). Inflammatory status will be evaluated using flow cytometry and immunofluorescence. Hepatic function will be assessed using serum ELISAs for common markers such as ALT and AST. Lastly, cerebral Aβ burden will be quantified using western blots and immunofluorescence.

Results: We observed persistently higher macrophage abundance in females, in addition to dichotomies in macrophage polarization and lymphoid cell phenotypes. Interestingly, females also exhibit higher levels of lipid accumulation in the liver compared to males, as well as signs of fibrosis at later stages. This is associated with higher cerebral Aβ burden in females at advanced stages of pathology.

Summary: Overall, this study provides a novel role of hepatic dysfunction in predicting CAA disease trajectory and outcomes. Furthermore, it provides evidence for sexual dimorphism in the liver-brain axis and highlights the need for deeper investigation to enhance our understanding of, and eventually improve outcomes for the disproportionate disease burden faced by women.

Poster 4 (Post-Candidacy)

Epiregulin and amphiregulin as dual antibody-drug conjugate targets in colorectal cancer.

Cara Guernsey-Biddle^{1, 2}, Joan Jacob³, Zhengdong Liang², Kendra Carmon^2
1MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences, Houston, TX, 77030
²Center for Translational Cancer Research, The Brown Foundation Institute for Molecular Medicine, The University of Texas Health Science Center, Houston, TX, 77030
³Pediatrics Oncology-AICT, Baylor College of Medicine, Houston, TX 77030

Colorectal cancer (CRC) is deadly at late metastatic stage. Existing therapies are often constrained by toxicity or drug resistance, including constitutively activating mutations (i.e., KRAS) that limit epidermal growth factor receptor (EGFR)-targeted therapy. Novel therapeutic strategies are needed. Antibody-drug conjugates (ADCs) act as guided missiles where a monoclonal antibody (mAb) conjugated to cytotoxic payloads binds its tumor-specific target, internalizes, and traffics to lysosomes for payload release, inducing tumor cell death while sparing normal tissues. We previously developed ADCs targeting the EGFR ligand epiregulin (EREG), which is overexpressed in KRAS wildtype and mutant (MUT) primary and metastatic CRC tumors. These ADCs inhibited CRC patient-derived xenograft tumor growth. Yet, tumors regrew upon treatment termination, and residual tumors showed EREG downregulation as a potential resistance mechanism, suggesting EREG ADC monotherapy may be insufficient. Co-targeting additional cell-surface antigens may enhance intratumoral payload delivery and overcome resistance. Analyses of CRC tumors, including RNA-seq data from The Cancer Genome Atlas Colorectal Adenocarcinoma cohort, show frequent co-overexpression of EREG and the EGFR ligand amphiregulin (AREG). We hypothesize that co-targeting EREG and AREG will enhance ADC efficacy compared to single-target monotherapy. We generated EREG knockout, KRAS-MUT CRC cell lines, which exhibit reduced proliferation, indicating that EREG promotes CRC growth despite constitutive RAS activation. We also generated AREG-overexpression cell lines to screen a panel of novel AREG-targeted mAbs for target-specific binding. Additionally, we subcloned and purified a chimaeric AREG-targeted mAb, which shows specific binding, internalization, and lysosomal trafficking in CRC cells, supporting AREG as a viable ADC target. Future work will generate AREG knockout cell lines and characterize the AREG mAbs to develop an AREG ADC as well as test dual EREG- and AREG-targeting strategies.

Poster 5 (Post-Candidacy)

Therapeutic Potential of Estrogen-Related Receptor Alpha in a Pre-Clinical Model of Duchenne Muscular Dystrophy

Sophia Huang^1,2, Thi Thu Hao Nguyen Nguyen², Svitlana Poliakova², Danesh Sopariwala², Vihang Narkar^1,2
1The University of Texas MD Anderson Cancer Center UTHealth, Graduate School of Biomedical Sciences, Houston, TX, USA
²Brown Foundation Institute of Molecular Medicine, McGovern Medical School, Center for Metabolic and Degenerative Diseases, Houston, TX, USA

The orphan nuclear receptor, estrogen-related receptor alpha (ERRα), has emerged as a central regulator of skeletal muscle aerobic capacity and exercise tolerance. ERRα regulates muscle endurance through transcriptional programs that drive mitochondrial biogenesis in skeletal muscle. Despite its established role in aerobic function, the therapeutic potential of ERRα in muscle regeneration and orphan myopathies remains poorly characterized. The most severe myopathy, Duchenne Muscular Dystrophy (DMD), is a progressive muscle degenerative disease characterized by impaired muscle regeneration, limited treatment options, and poor prognosis. We have discovered that ERRα and its target mitochondrial genes are repressed in skeletal muscles of the mdx mice, a murine model of DMD. Transgenic overexpression of ERRα, selectively in mdx skeletal muscle, reduced muscle damage marked by decreased serum creatine kinase, improved muscle histology, and restored contractile function and exercise tolerance. Notably, ERRα overexpression in mdx mice enhances muscle regeneration after acute barium chloride injury, demonstrating its regenerative potential in promoting repair in dystrophic muscle. Insights into mitochondrial function and transcriptional program, obtained through transmission electron microscopy and RNA sequencing, respectively, reveal restoration of mitochondrial number and health, and robust activation of the mitochondrial gene program through ERRα in dystrophic muscle. Notably, gene therapy via AAV9-mediated delivery of ERRα to dystrophic skeletal muscle in mdx mice largely recapitulated the effects observed in the transgenic mouse model, indicating translational potential. Additionally, we found a deficiency in ERRα expression in dystrophic muscle stem cells (MuSCs), and restoring receptor expression promoted differentiation and the recovery of mitochondrial gene programs. Our data demonstrates a critical deficiency in ERRα signaling in dystrophic muscles, possibly originating in dystrophic MuSCs. Restoring ERRα through transgenesis and gene therapy mitigates muscle damage in mdx mice by activating the mitochondrial gene program and promoting regeneration to restore function. Collectively, our studies identify ERRα as a novel dystrophin-independent therapeutic target for muscle function enhancement and repair in DMD.

Poster 6 (Post-Candidacy)

Evaluation of EGFR and LGR5 Antibody-Drug Conjugate-Based Dual-Targeting Therapeutic Strategies for the Improved Treatment of Colorectal Cancer

Peyton High^1,2, Zhengdong Liang¹, Maya Cappellino², Tiffani Blackburn², Cara Guernsey-Biddle^1,2, Shraddha Subramanian^1,2, Yueh-Ming Shyu^1,2, Adela Aldana¹, Yukimatsu Toh¹, Kendra S. Carmon^1
1Center for Translational Cancer Research, The Brown Foundation Institute of Molecular Medicine, University of Texas Health Science Center at Houston, Houston, TX, 77030, USA; ²The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX, 77030, USA

Colorectal cancer (CRC) is the second-leading cause of cancer-associated death in the United States, indicating a need for improved therapies. Antibody-drug conjugates (ADCs) are a class of therapeutics that employ the specificity of monoclonal antibodies to direct highly potent drug payloads to cancer cells while sparing normal tissues. We have developed ADCs targeting leucine-rich repeat-containing G-protein coupled receptor 5 (LGR5), a marker of normal adult intestinal stem cells and CRC stem-like cells with roles in CRC tumorigenesis, tumor progression, metastasis, and drug resistance. While LGR5-targeting ADCs incorporating microtubule inhibitors promoted tumor regression in select CRC xenograft models and were well-tolerated, tumors eventually relapsed following treatment cessation due to LGR5 downregulation and suboptimal ADC payload selection. Thus, this work aims to investigate multi-targeting therapeutic strategies with improved LGR5-targeting ADCs to improve CRC care and prevent relapse. Our work demonstrates that therapies targeting the epidermal growth factor receptor (EGFR), including cetuximab (CTX) which is approved for KRAS^WT metastatic CRC, increase LGR5 protein levels in CRC cell lines, tumor organoids, and mouse models independent of KRAS mutational status. To evaluate EGFR and LGR5 dual-targeting approaches, we generated a novel LGR5-targeting ADC (8E11-CPT2) incorporating a topoisomerase I inhibitor payload via site-specific conjugation that was well-tolerated in immunocompetent mice. Importantly, combination of CTX with 8E11-CPT2 significantly enhances anti-tumor efficacy and extends survival as compared to ADC and CTX monotherapies in RAS^MUT CRC patient-derived xenografts. Still, tumors eventually relapse following combination treatment, warranting investigation into alternative dosing regimens and dual-targeting modalities. We have therefore generated two EGFR:LGR5 bispecific antibodies (bsAbs) that, mechanistically, promote lysosome-mediated EGFR degradation in an LGR5-dependent fashion. EGFR:LGR5 bsAbs demonstrate minimal cytotoxicity in CRC cells, necessitating drug conjugation to generate an EGFR:LGR5 bispecific ADC (bsADC; E⨯L-1-CPT2). Importantly, E⨯L-1-CPT2 exerts 100- to 1000-fold enhanced potency over 8E11-CPT2 with identical linker-payload in LGR5/EGFR-expressing CRC cell lines of varying genetic backgrounds. Furthermore, E⨯L-1-CPT2 significantly reduced tumor burden and extended survival in RAS-mutant CRC xenografts. These promising findings warrant further investigation into the anti-tumor efficacy of EGFR:LGR5 bsADC monotherapy versus combination therapy of CTX and LGR5 ADCs in CRC xenograft models. Taken together, this work strongly rationalizes dual-targeting of EGFR and LGR5 as an effective therapeutic option for CRC and other EGFR- and LGR5-expressing cancers.

Poster 7 (Ineligible for Prizes)

Methylomic and Transcriptomic Signatures of Suicide Death in Postmortem

Brains of Individuals with Bipolar Disorder 

Valeria Arredondo¹, Steven De La Garza², Salahudeen Mirza³, Iago Junger-Santos⁴, Camila N.C. Lima², Alexandra Del Favero-Campbell², Alexandre Rubinsten², Christopher Busby², Laura Stertz², Rohit Jha², Consuelo Walss-Bass², Jair C. Soares², Gabriel R. Fries^2
1The University of Texas Health Science Center at Houston School of Behavioral Health Sciences, Houston, TX, USA; ²UTHealth Houston, McGovern Medical School, Houston, TX, USA; ³Child Study Center, Yale School of Medicine, New Haven, CT, USA; ⁴Department of Physiology and Biophysics University of São Paulo, São Paulo, Brazil

Background: Individuals with bipolar disorder (BD) have elevated suicide death (SD) rates, and current literature postulates gene expression differences – as a result of epigenetic mechanisms – as increasing suicide risk in vulnerable individuals. The present study aimed to identify brain genome-wide DNA methylation (DNAm) and transcriptomic changes associated with SD in BD using post-mortem dorsolateral prefrontal cortex (BA9/46) tissue.
Methods: Samples were obtained from individuals with BD (n=211) who died by suicide (n=98, BD/SD), or who died by causes other than suicide (n=113, BD/non-SD), and age- and sex-matched non-psychiatric controls (n=135). DNAm was assessed in all tissues with the Illumina EPIC BeadChip v1, transcriptomic changes were analyzed in subsets of BD/SD (n=53), BD/non-SD (n=56), and controls (n=55) by using bulk RNA-sequencing (Illumina NovaSeq, 2x150bp), and weighted gene expression network analysis (WGCNA) was used to analyze different modules of co-expressed genes between groups.
Results: Two differentially methylated positions (DMPs, p<9×10-8) – cg16259714 (SDK1) and cg04828068 (STAM2) – were found downregulated in BD, and thirteen differentially expressed genes (DEGs, p<0.05, logFC≥0.5) were identified between BD and controls. Subset BD analysis found two significant DMPs – cg07521902 (PODNL1) downregulated and cg14325002 (DCST1) upregulated in SD – and 125 DEGs between BD/SD and BD/non-SD. Seven modules of co-expressed genes were differentially expressed between BD/SD and BD/non-SD, in which the highest module – composed of genes related in pathways for insulin resistance, PPAR signaling, and adipocytokine signaling – was downregulated in BD/SD. When examining overlaps between nominally significant DMPs and DEGs, the long intergenic non-protein coding RNA 1354 (LINC01354) was identified, noting increased methylation and decreased expression in BD/SD.
Conclusions: In accordance with previous studies, the results demonstrate differences in epigenetic and transcriptomic changes in individuals with BD who died by suicide and non-suicide. These gene expression differences may serve as potential targets for understanding and developing treatments for suicide-vulnerable individuals with bipolar disorder. Particularly, LINC01354 may serve as a gene of interest in better understanding lithium’s suicide-reducing effects, as this gene is implicated in the stabilization of B-catenin, a protein associated with regulating neuroprotective factors in neurons and glia.

Poster 8 (Post-Candidacy)

Prdm1/Blimp-1 acts as a corepressor of FoxA to suppress esophageal gland cell fate in schistosome parasites

Ryan Sloan^1,2,3,5, Pallavi Yadav³, Pengyang Li⁴, Sabona B. Simbassa^1,3, Bo Wang⁴, Jayhun Lee^1,2,3,6
1Microbiology and Infectious Diseases Program, UT MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences. ²Genetics and Epigenetics Program, UT MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences. ³ Department of Microbiology and Molecular Genetics, McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX 77030. ⁴ Department of Bioengineering, Stanford University, Stanford, CA 94305. ⁵[email protected]. ⁶[email protected].

Background: Schistosomes are parasitic flatworms that cause the major disease schistosomiasis, which affects over 250 million people. Only one drug is currently available for treatment of this disease, which does not prevent reinfection and has had reports of resistance. In schistosomes, the forkhead transcription factor FoxA regulates the development/maintenance of the esophageal gland (EG), a specialized organ essential for immune evasion and survival in vivo.  In addition to its EG expression, prior single-cell studies show foxA expression in a limited number of neural subsets and the intestine. However, we lack a comprehensive understanding of the cell-type-specific expression and function of FoxA. A pioneer factor such as FoxA works with cofactors to promote or repress lineage-specific gene expression by altering chromatin accessibility. Thus, we hypothesize that schistosome FoxA works with tissue-specific cofactors to regulate the production and/or function of non-EG cell types. Methods: To enrich for foxA-expressing cell types and identify potential FoxA cofactors, we performed a head-enriched (HE) single-cell (sc) RNA-seq analysis coupled with in situ hybridization (ISH) and RNAi of candidate genes. Results: HE scRNA-seq shows foxA expression in the EG, stem cells, intestinal progenitors, and a subset of neurons. Highly associated with foxA expression in these non-EG cell types is prdm1/blimp-1, encoding a transcriptional repressor known in other model organisms as a cofactor of FoxA. ISH showed prdm1/blimp-1 enrichment in the neural and intestinal subsets and confirmed co-expression with foxA. RNAi-mediated prdm1/blimp-1 knockdown significantly increased ectopic EG gene expression in the parasite body posterior to the EG, suggesting its role in suppressing the EG fate in non-EG cells. Motif analysis identified both FoxA and Prdm1/Blimp-1 motifs in the foxA regulatory region, suggesting this suppression may occur via negative regulation of foxA. Conclusions: We conclude that Prdm1/Blimp-1 acts as a corepressor of EG cell fate in non-EG foxA expressing cells. Future work aims to further define the roles of prdm1 in gene regulation, cell type production, and survival in the host. Comprehensive characterization of these cell types and genes may identify factors that are vital for tissue homeostasis and function, with potential to be targeted for treatment of this disease.

Poster 9 (Post-Candidacy)

Activation of a Two-Component System Confers Resistance to Novel Bam Complex Inhibitors

Teresa Sullivan^1,2, Anna Konovalova^1,2
1The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030
²Department of Microbiology and Molecular Genetics, The University of Texas Health Science Center at Houston, Houston, TX 77030

Gram-negative bacteria present a significant problem for antimicrobial therapies due to the presence of an outer membrane, which restricts the entry of many compounds, including antibiotics. One strategy to overcome the outer membrane permeability barrier is to develop inhibitors that target essential proteins on the bacterial surface. One such protein is BamA, an essential component of the β-barrel assembly machinery (Bam) complex that folds and inserts all outer membrane proteins. BamA emerged as an attractive target, and its small molecule, peptide, and antibody inhibitors have been reported.

MRL-494 is a small molecule inhibitor of BamA, but mechanisms of action and resistance remain poorly understood. We isolated several novel MRL-494 resistant mutations, including those that target the two-component system, PmrAB. This signaling pathway is well-studied for controlling enzymes that modify the lipid A portion of LPS, reducing its negative charge and conferring high resistance to polymyxin antibiotics. Through genetic analysis, I showed that the activation of PmrAB leads to resistance to MRL-494 in the manner dependent on lipid A modifications. Moreover, I demonstrated that this resistance is not MRL-494 specific. It applies more broadly to Bam complex inhibiting conditions and promotes resistance to another BamA inhibitor, darobactin.

LPS modifications are a major threat for clinical resistance to last-resort antibiotics like polymyxin B and colistin. The observation that the same LPS modifications confer cross-resistance to Bam complex inhibitors presents significant concerns and highlights the urgent need to address resistance mechanisms in order to develop effective antimicrobial therapies.

Poster 10 (Post-Candidacy)

Conditional Hap40 deletion in GABAergic neurons preserves vertebrate development and physiologic homeostasis across the lifespan

Stephen M Farmer^1,2,3, Xin Ye¹, Jing Cai^1,2, Vicky Chuong^2,4, Zhengmei Mao⁵, Zhiyu Cao⁶, Qingchun Tong^1,2,7, Sheng Zhang^1,2,4
1Center for Metabolic and Degenerative Diseases, The Brown Foundation Institute of Molecular Medicine, University of Texas Health Science Center at Houston (UTHealth), Houston, TX, USA
²Program in Neuroscience, The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences (MD Anderson UTHealth GSBS), Houston, TX, USA
³Program in Molecular and Translational Biology, The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences (MD Anderson UTHealth GSBS), Houston, TX, USA
⁴Department of Neurobiology and Anatomy, McGovern Medical School at the University of Texas Health Science Center at Houston (UTHealth), Houston, TX, USA
⁵Microscopy Core, The Brown Foundation Institute of Molecular Medicine, University of Texas Health Science Center at Houston (UTHealth), Houston, TX, USA
⁶Department of Biosciences, Rice University, Houston, TX, USA
⁷Center for Neuroimmunology and Glial Biology, The Brown Foundation Institute of Molecular Medicine, University of Texas Health Science Center at Houston (UTHealth), Houston, TX, USA

Huntington’s disease (HD) is a fatal autosomal-dominant neurodegenerative disorder caused by CAG expansion in Huntingtin (HTT), producing an expanded polyglutamine tract that confers toxic gain-of-function effects while disrupting HTT’s normal physiological roles. These combined mechanisms drive progressive degeneration of striatal and cortical projection neurons within corticostriatal and cortico-basal ganglia-thalamocortical circuits, leading to motor, cognitive, and behavioral decline. Although HTT-lowering strategies are being actively pursued, the normal regulation and physiological functions of HTT in mammalian systems remain incompletely defined. HAP40 is a conserved, obligate HTT-binding partner that stabilizes HTT and regulates its conformation and abundance. Our group previously demonstrated that global Hap40 loss in Drosophila ameliorates mutant HTT toxicity, suggesting that reducing HAP40 may be protective; however, the physiological consequences of Hap40 depletion alone in mammals have not been evaluated. Here, we generated Hap40-floxed mice to enable conditional deletion in broad or tissue-specific contexts. Consistent with their obligate partnership, we found that endogenous HTT and HAP40 show broadly overlapping expression across mouse embryonic development and in the adult brain. Deletion using Nestin-Cre, which targets neuroepithelial and radial glial progenitors that give rise to most neurons and glia in the CNS, resulted in severe brain atrophy, neuroinflammation, and postnatal lethality prior to weaning. We therefore deleted Hap40 using Vgat-Cre, which targets GABAergic neurons including striatal medium spiny neurons that are selectively vulnerable in HD, and performed longitudinal physiological, behavioral, and histological analyses across aging. Hap40 conditional knockout (Hap40-cKO) mice remained broadly healthy into advanced age (>24 months), with preserved survival and fecundity, body composition, locomotor activity, and indirect calorimetry measures, indicating intact metabolic homeostasis. Under increased motor demand, young adult (3-6 month) cKO exhibited reduced exercise endurance without grip-strength deficits, whereas endurance, grip strength, and/or motor coordination were preserved in old-aged (18-24 month) cohorts. Since GABAergic circuits regulate affective and exploratory behaviors in addition to motor output, we assessed behavior in open-field, light-dark transition, and social interaction assays: two-year-old cKO displayed increased locomotor output and reduced latency to enter avoidance zones, consistent with reduced anxiety-like behavior. Notably, aged Hap40-cKO mice also developed prominent thalamic calcifications, yet showed no overt gliosis, reduced survival, or overall health deficits. Together, these findings indicate that Hap40 loss in GABAergic neurons is largely tolerated across aging and support HAP40 targeting as a potential HTT-lowering approach.

Poster 11 (Post-Candidacy)

Pediatric moyamoya disease is driven by heterozygous pathogenic variants disrupting smooth muscle cell differentiation

Kiara E. Bornes^1,2, Amelie Pinard¹, Jose Emiliano Esparza Pinelo¹, Dianna M. Milewicz¹, Callie S. Kwartler^1
1Department of Internal Medicine, Division of Medical Genetics, UTHealth Houston McGovern Medical School, ²Genetics and Epigenetics Program, The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences

Moyamoya disease (MMD) is a rare cerebrovascular disorder that predisposes patients to transient ischemic attacks and stroke. Pediatric cases arise in the absence of chronic vascular stressors, strongly suggesting a genetic trigger. Although multiple genes have been associated with MMD, including a highly prevalent East Asian founder variant in RNF213, these variants explain only a subset of familial cases, and the molecular mechanisms underlying disease initiation remain unclear. Histopathologic studies of affected internal carotid arteries consistently reveal occlusive lesions composed of cells positive for the smooth muscle cell (SMC) marker smooth muscle α-actin (SMA), supporting a hypothesis that SMCs migrate from the arterial media into the lumen and proliferate to form stenotic lesions. Heterozygous ACTA2 p.R179 mutations are associated with bilateral MMD-like carotid occlusive lesions. Prior work demonstrated that ACTA2 p.R179 mutant SMCs exhibit incomplete differentiation, enhanced migration, and increased proliferation. These findings raise a broader question: Is incomplete SMC differentiation a shared pathogenic mechanism across genetically distinct forms of MMD? To address this, I examined SMCs harboring  heterozygous RNF213 p.F4120L mutation that occur de novo and cause severe, early-onset pediatric MMD. SMCs isolated from a knock-in mouse model (Rnf213^F4070L/+, the murine homolog of human F4120L) exhibit impaired differentiation, increased proliferative capacity, and reduced oxidative metabolism—paralleling the abnormalities observed in ACTA2 p.R179 SMCs. Additionally, human iPSC-derived SMCs carrying RNF213 p.F4120L exhibit maturation defects aligning with the explanted mouse SMC phenotype. Together, these data provide preliminary but compelling evidence that incomplete SMC differentiation may represent a unifying molecular mechanism across genes with pathogenic variants that cause MMD. Ongoing work will determine if Rnf213^F4070L/+ mice have MMD-like disease in vivo. Defining how these mutations disrupt normal SMC maturation will be essential for understanding disease initiation and progression and may ultimately guide the development of targeted therapeutic strategies to restore proper SMC differentiation and prevent vascular occlusion.

Poster 12 (Post-Candidacy)

Comparison of potent microtubule inhibitors in docetaxel-resistant metastatic castration-resistant prostate cancer.

 Jack Adams^1,2, Servando Hernandez Vargas^1,2, Majid Momeny¹, and Ali Azhdarinia^1
 1Institute of Molecular Medicine, McGovern Medical School, The University of Texas Health Science Center at Houston (Houston, TX)
² Graduate School of Biomedical Sciences, The University of Texas MD Anderson Cancer Center and UTHealth Houston (Houston, TX)  

Metastatic castration-resistant prostate cancer (mCRPC), a lethal form of prostate cancer, arises from reactivation of the androgen receptor (AR) pathway despite pharmacologically depriving malignant tissues of androgens and is often diagnosed in the metastatic setting. Microtubule inhibitors (MTIs) are the only chemotherapies that improve median overall survival (mOS) in mCRPC due to blocking the cell cycle and AR nuclear trafficking. While efficacious, hematological, gastrointestinal, and neurological toxicities have limited clinical use of non-tumor-specific MTIs in an advanced age population. Radioligand therapy (RLT; ¹⁷⁷Lu-PSMA-617) was the first FDA-approved precision therapy for taxane-experienced mCRPC patients and consists of a small molecule that targets prostate-specific membrane antigen (PSMA) to selectively deliver the DNA-damaging, beta-emitting radionuclide. Despite producing a clinically meaningful 4-month increase in mOS in the phase 3 VISION trial, nearly all patients undergoing RLT will experience disease progression. Thus, strategies that capitalize on the persistent expression of PSMA and leverage the precision targeting properties of RLT are warranted.

PSMA-targeted drug conjugates build on the proven mCRPC targeting of RLT by incorporating a chemotherapeutic payload in place of a DNA-damaging radionuclide. Despite six PSMA drug conjugates entering phase I/II clinical trials, none have received FDA approval. While these drug conjugates primarily use MTIs, such as monomethyl auristatin E (MMAE), as a payload, studies in docetaxel-resistant (DR) cells demonstrate a significant reduction in MMAE potency. A novel MTI, MTI19, has been shown to retain potency in multidrug-resistant cells. To test those effects in mCRPC and identify an optimal payload for drug conjugate development, we compared MTI19 to MMAE in docetaxel-sensitive (DS) and DR mCRPC cells. We hypothesized that MTI19 would induce stronger and more durable antitumor responses in DS and DR cells, making it a preferred chemotherapy agent for a PSMA-targeted drug conjugate.

Using mCRPC cell lines, we compared MTI19 to MMAE via proliferation, cell cycle, and clonogenic assays. We then performed Western blot analysis to assess cell death induction via markers of apoptosis. To recapitulate the clinical scenario of taxane resistance, we continuously exposed mCRPC cells to docetaxel to generate DR cells, and then evaluated the expression of anti-apoptotic proteins and drug efflux transporters. We then conducted the same series of assays as the DS cell lines to determine efficacy in the setting of DR compared to MMAE.

MTI19 demonstrated 140-fold greater cytotoxicity than MMAE and inhibited the clonogenicity of DS cells at picomolar concentrations. We show that MTI19 induces apoptosis at 100 pM compared to 1 nM MMAE, representing a 10-fold increase in potency over a benchmark MTI payload. While the DR cells were resistant to MMAE, they remained sensitive to MTI19 in proliferation and clonogenic assays, despite upregulation of the drug efflux transporter and the anti-apoptotic proteins BCL-xL and survivin. Inhibition of efflux transporters restored MMAE activity without affecting MTI19 potency, suggesting the potential of MTI19 to evade drug efflux pumps and overcome taxane resistance.

Collectively, these results suggest that MTI19 may outperform MMAE as a payload for mCRPC due to its ability to evade resistance mechanisms. Future studies are planned to synthesize MTI19 tumor-targeted drug conjugates and evaluate their pharmacological properties in cell and animal models.

Poster 13 (Post-Candidacy)

Linker Optimization Improves the Performance of SSTR2-Targeted Peptide–Drug Conjugates for Neuroendocrine Tumor Therapy

Tyler M. Bateman¹, Ebaston Thankarajan², Solmaz AghaAmiri¹, Sukhen C. Ghosh¹, Jack T. Adams¹, Servando Hernandez Vargas¹, Majid Momeny¹, Vahid Khalaj¹, Martin J. Schnermann², Ali Azhdarinia^1
1McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX 77054, USA; ²Chemical Biology Laboratory, Center for Cancer Research, National Cancer Institute, Frederick, MD 21702, USA

Peptides have been used as therapeutic agents in oncology since the introduction of leuprolide in the 1980s, serving to modulate receptors or deliver payloads as targeting ligands. In the past decade, peptide-based targeting has produced a major clinical breakthrough with the development and clinical approval of peptide-receptor radionuclide therapy (PRRT) using ¹⁷⁷Lu-DOTA-TATE, a radiolabeled somatostatin analog that delivers cell-killing beta radiation to neuroendocrine tumors (NETs) expressing the somatostatin receptor subtype 2 (SSTR2). Currently, ¹⁷⁷Lu-DOTA-TATE is limited to four FDA-approved cycles of therapy due to concerns over cumulative radiation toxicity, and most patients relapse after treatment discontinuation. These tumors retain SSTR2 expression after treatment with ¹⁷⁷Lu-DOTA-TATE, allowing for the development of chemotherapeutic SSTR2-targeted peptide-drug conjugates (PDCs) to provide further treatment options post-PRRT. Such PDCs can be generated by incorporating potent chemotherapeutics into PRRT-derived scaffolds, but optimal payload conjugation strategies must be identified to maximize effectiveness and translational potential.

To determine optimal payload‑linkage strategies, we developed a custom somatostatin analog (MMC-TOC) and selected monomethyl auristatin E (MMAE) as a prototype payload. A library of MMC(MMAE)-TOC conjugates was synthesized using click chemistry and amide linkage in combination with cleavable and non-cleavable linkers. Agents were evaluated for SSTR2 affinity and specificity using cell binding assays and CellTiter Glo^® assays. Biodistribution was assessed 4 h post injection (p.i.) in nude mice with bilateral SSTR2⁺ and SSTR2^– tumor implants to identify the conjugation strategy that provides the most favorable combination of tumor specificity and clearance. The lead compound was further evaluated in healthy CD‑1 mice at 1, 6, and 24 h p.i., and additional imaging studies in NCI‑H69 (SSTR2⁺ NET) xenografted nude mice are ongoing to evaluate pharmacokinetics.

Results indicate that cleavable amide‑based linkers preserve SSTR2 affinity more effectively than the non-cleavable or click-chemistry-based linkers. The cleavable amide‑conjugated constructs also demonstrated improved cytotoxicity in SSTR2⁺ cells compared to SSTR2^– cells, with no difference between cell lines observed when the non-cleavable linker is used. Cleavable amide-linked conjugates enhanced renal clearance and led to a 5-fold increase in selective SSTR2⁺ tumor accumulation compared to other tested conjugation strategies. These findings highlight linker chemistry, particularly cleavable amide linkers, as a critical determinant of pharmacologic performance in SSTR2‑targeted PDCs. This data also supports the potential of PDCs as next‑line therapies for patients with SSTR2⁺ NETs after PRRT, offering a targeted option that could overcome limitations associated with radiation‑based treatments.

Poster 14 (Post-Candidacy)

Circadian targeting improves glioma cell response to temozolomide

Paula Bender^1,2, Kristin Eckel-Mahan^1,2
1The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX, 77030, USA
²Center for Metabolic and Degenerative Diseases, UTHealth Houston McGovern Medical School Brown Foundation Institute of Molecular Medicine, Houston, TX, 77030

Gliomas are common and lethal brain tumors, with high-grade variants exhibiting aggressive growth and therapeutic resistance. Standard treatment with surgery, radiation, and temozolomide (TMZ) chemotherapy offers limited survival benefits, highlighting a need for improved treatment strategies. Chronotherapy aims to enhance treatment efficacy by aligning drug administration with the body’s natural 24-hour cycles (i.e., circadian rhythms). The circadian clock, regulated by the transcription factors BMAL1 and CLOCK, regulates DNA repair, cell cycle progression, and apoptosis. Nobiletin (NOB), a flavonoid found in citrus peels, enhances circadian rhythms by increasing BMAL1 expression and has been demonstrated by our lab to exhibit anti-cancer effects in hepatocellular carcinoma and acute myeloid leukemia.

To investigate circadian regulation in glioma, we synchronized several murine and patient-derived glioma cell lines using serum shock. Glioma cells displayed disrupted rhythmicity of clock genes, which persisted even after overexpression of BMAL1 and CLOCK. Overexpression, however, did alter the expression of additional circadian and tumor-related genes. In treatment studies, combination treatment with NOB and TMZ reduced glioma cell viability more effectively than either treatment alone, and the reduction in viability showed a circadian pattern. Single and combination treatments also differentially modulated circadian and tumor-related gene expression and rhythmicity.

Together, these findings suggest that circadian disruption may contribute to glioma progression and aggressiveness, and that targeting circadian pathways could enhance the efficacy of chemotherapy. Ongoing studies are expanding this work to additional glioma cell lines, in vivo syngeneic models, and patient-derived xenografts to further examine how NOB and timed TMZ influence circadian gene activity and tumor growth. Ultimately, this research may help establish circadian-based strategies to improve outcomes for glioma patients.

Poster 15 (Post-Candidacy)

Developing GPR56-Targeted Antibody-Drug Conjugates for Triple-Negative Breast Cancer Treatment

Yueh-Ming Shyu^1,2, Joan Jacob³, Carla Godoy², Treena Chatterjee², Zhengdong Liang², Kendra S. Carmon²

¹The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030; ²Center for Translational Cancer Research, The Brown Foundation Institute of Molecular Medicine, The University of Texas Health Science Center at Houston, Houston, TX 77030; ³Department of Pediatrics, Oncology AICT, Baylor College of Medicine, Houston, TX 77030

Triple-negative breast cancer (TNBC) is an aggressive subtype lacking durable targeted therapies, and resistance continues to drive poor clinical outcomes. The clinical success of antibody–drug conjugates (ADCs), which combine antibody specificity with potent cytotoxic payloads, highlights their growing clinical impact. Although the approved anti-TROP2 ADC, sacituzumab govitecan, provides clinical benefit, it causes toxicity in normal tissues, underscoring the need for more selective therapeutic strategies. Adhesion-mediated signaling through the FAK–SRC axis is a major survival pathway under therapeutic stress. Our previous study showed that GPR56, an adhesion G protein-coupled receptor (GPCR), activates the FAK–SRC pathway and that targeting GPR56 with an ADC elicited potent antitumor efficacy in colorectal cancer models. Given these results, we examined GPR56 as a potential therapeutic target in TNBC. While TROP2 is a validated ADC target in TNBC, its expression in normal tissues limits tumor selectivity. GPR56, by contrast, is highly expressed in TNBC and associated with poor prognosis, yet shows limited normal tissue expression, potentially conferring a superior therapeutic index for GPR56-targeted ADCs. Our first-generation anti-GPR56 ADC (10C7-Duo), incorporating the DNA-damaging payload duocarmycin, exhibited target-dependent cytotoxicity in GPR56-positive TNBC cell lines, confirming that GPR56 is a viable therapeutic target. However, the monoclonal antibody (mAb) 10C7 exhibited agonistic activity, which could limit its therapeutic potential. Thus, we developed 9E3, a non-agonist anti-GPR56 mAb that internalizes efficiently and traffics to lysosomes for payload delivery. 9E3 was conjugated to a more potent pyrrolobenzodiazepine (PBD) payload using site-specific chemistry, and analyses confirmed successful conjugation, stability, and preserved antigen binding. Functional studies in breast cancer models showed that GPR56 knockdown suppressed, while overexpression enhanced, FAK–SRC phosphorylation, tumor cell growth, and invasion. Currently, we are evaluating 9E3-PBD for in vitro cytotoxic potency and selectivity, as well as in vivo safety and antitumor efficacy in TNBC cell line xenografts and patient-derived models. These findings help establish GPR56 as a clinically relevant adhesion GPCR and introduce a novel ADC approach to target therapeutic resistance in TNBC. By integrating receptor biology with targeted payload delivery, this work lays the foundation for potentially safer and more selective therapeutic options for TNBC patients.

Poster 16 (Post-Candidacy)

Rheb membrane orientation dynamics and functional consequences elucidated by molecular simulations, single-molecule-FRET and signaling assays

Chase M. Hutchins^1,2, Alemayehu A. Gorfe^1
1McGovern Medical School, University of Texas Health Science Center at Houston, Department of Integrative Biology and Pharmacology, 6431 Fannin St., Houston, Texas 77030, USA
²UTHealth MD Anderson Cancer Center Graduate School of Biomedical Sciences, Houston, 6431 Fannin St., Texas 77030, USA

Background: Rheb is a small GTPase in the Ras family that localizes to the endoplasmic reticulum, lysosomes and other endomembranes promotes cell growth and proliferation by directly activating mTORC1. Overexpression or mutation or Rheb can lead to increased tumorigenesis and poor prognosis in colorectal or hepatocellular carcinoma. Like other Ras family GTPases, Rheb is a peripheral membrane protein consisting of a ~160 residue catalytic domain bound to membranes by a C-terminal farnesyl tail at the end of a flexible intrinsically disordered region called the hypervariable region (HVR). The three human Ras proteins (K-, H-, and NRAS), have been shown to switch between several distinct orientation membrane states, some of which display impaired effector binding due to membrane occlusion. Although previous simulation and experimental studies have shown that Rheb also adopts multiple orientations, the functional role of Rheb’s membrane dynamics in more native-like environments remains unexplored.

Methods: To map the membrane orientation landscape of Rheb, we conducted 20 µs simulations of GDP- and GTP-bound Rheb in an endoplasmic reticulum (ER) model bilayer. We then performed single-molecule FRET assays of Rheb in cell-derived lipid nanodiscs to experimentally map Rheb’s dynamics. We assessed the functional consequences of disrupting those states through simulation guided mutagenesis assays of Rheb’s ability to stimulate mTORC1 activity in Rheb dKO HeLa cells. We explored the kinetic pathways through each orientation state with Hidden Markov Modeling to propose a mechanism for orientation-dependent modulation of Rheb activity.

Results: We identified four distinct membrane orientations of Rheb, which displayed a near perfect alignment with populations observed in smFRET assays, validating our simulations. When aligning simulation structures with a recent experimental structure of the mTORC1-Rheb complex on a membrane, we found that one orientation state (OS3) was capable of forming a favorable binding configuration. We identified several distinct residue patches in each orientation state, and mutating these residues in two of our four orientations resulted in a bi-directional modulation of mTORC1 activity. Hidden Markov Modeling revealed that Rheb sequentially transitions between two stable basins, through two obligate intermediate states.

Conclusions: Our findings revealed that Rheb adopts four distinct orientation states with a highly ordered transition path between them. When combined with our functional assays, our kinetic findings indicate that orientation-dependent regulation of Rheb is a more complex process than observed in  Ras, implying that orientation’s functional role is more complex than a binary “occluded” vs. “unoccluded” paradigm. Furthermore, they act as proof of concept that membrane orientation dynamics are a functionally relevant processes in GTPases beyond Ras, and suggests that the process may be generalizable to the entire  protein class.

Poster 17 (Pre-Candidacy)

The Dark Side of the Medicine Cabinet: Impact of Medications on Candida Resistance & Fitness

Weerakkody Sanduni Ranasinghe^1,2, Madhvi Bhakta¹, Diana M. Proctor^1,2
1The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences, Houston, TX, USA
²Department of Microbiology and Molecular Genetics, UTHealth Houston McGovern Medical School, Houston, TX, USA

Use of antibacterials is a well-known risk factor for Candida infection, including bloodstream infection. The prevailing dogma holds that antibacterials disrupt colonization resistance, thereby creating a niche for Candida. However, this model cannot explain why antibacterials are also associated with infection by antifungal-resistant Candida. Large-scale drug screens have identified medications that exert collateral effects on Candida antifungal resistance, though the mechanism remains unknown. Since many drugs are known to induce mitochondrial dysfunction in human cells, we hypothesize that these medications, including antibacterials, increase antifungal resistance by inducing mitochondrial dysfunction in Candida. To test this hypothesis, we screened a panel of 22 drugs against C. albicans for minimum inhibitory concentration (MIC) using broth microdilution tests, interactions with fluconazole using checkerboard assays, and effects on mitochondrial function using assays such as MitoROS. Similar to all antifungals (4 out of 4), we found that 1/4 of cardiac meds and a statin inhibited Candida growth in a dose-dependent manner. The checkerboard assay for vancomycin (Van) showed increased Candida growth (30-600%) compared to Van-free controls in a medium- and dose-dependent manner, but only in the presence of fluconazole. Although Van modulated medium pH (4-6.5) at high drug concentrations, permutation testing revealed a statistically significant effect of Van on C. albicans growth in the presence of fluconazole that could not be explained by acidification alone. Several drugs (36%) induced mitochondrial superoxide production (4/5 antifungals, 2/2 antidepressants, 1/1 statin, and 1/4 cardiac medications), a hallmark of mitochondrial dysfunction. We are currently probing other axes of mitochondrial function that may explain these altered phenotypes while expanding our dataset to include over 1000 FDA-approved drugs.

Poster 18 (Ineligible for Prizes)

Development of brain endothelial cell-derived extracellular vesicles from bioreactor cultures

 Balaji Govindaswamy¹, Paromita Paul Pinky², Vivek Basudkar², Devika Soundara Manickam^1
1Department of Neurology, The University of Texas Health Science Center at Houston, TX 77030, United States of America
²Graduate School of Pharmaceutical Sciences, Duquesne University, Pittsburgh, PA 15282, United States of America

Background
Extracellular vesicles (EVs) are carriers of cellular cargo such as proteins, lipids, and nucleic acids that mediate intercellular communication and hold promise as a novel therapeutic/diagnostic platform. Efficient and scalable production of EVs remains a critical challenge for translational applications. Our laboratory has previously established a batch isolation process for EVs from cultured brain endothelial cells (BECs) grown in T-175 tissue culture flasks.^1–3
Hypothesis
We hypothesize that bioreactor cultures of brain endothelial cells will support continuous collection of EVs resulting in greater EV yields with preserved biological functionality and will serve as a cost-efficient process.
Methods
Mouse BECs (b.End3) were cultured in a CELLine AD 1000 bioreactor, and EVs were isolated by sequential centrifugation of EV-conditioned medium.^4,5 We collected both large EVs (lEVs > 200 nm) and small EVs (sEVs < 200 nm). The sequential centrifugation involved low-speed centrifugation to remove cell debris (300xg for 11 mins at 4°C), apoptotic bodies (2,000xg for 22 mins at 4°C) and ultracentrifugation to pellet lEVs (20,000xg for 45 mins at 4°C) and sEVs (120,000xg for 70 mins at 4°C). Both EV subsets were characterized for total EV protein content using micro-BCA assay, while dynamic light scattering (DLS) was used to measured particle diameter (nm) and surface charge (mV), and nanoparticle tracking analysis (NTA) was performed to determine the particle concentration (particle/mL). Western blotting was used to identify EV subtype-specific markers. Functional bioactivity of EVs was evaluated using an ATP assay in oxygen-glucose deprived BECs to mimic ischemic stress, and the relative intracellular ATP levels were quantified using a luminescence-based ATP assay.⁶
Results
Across three independent isolations, lEVs exhibited larger particle diameters (200 – 230 nm; PDI 0.25 – 0.38), with total EV protein content ranging from (2.6 to 5.3 mg), while sEVs showed a smaller particle diameter (75-150 nm; 0.37 – 0.55), with total EV protein content ranging from (2.7 to 5.3 mg). Surface charge measurements confirmed both EV subtypes carried a net negative charge (-25 to -32 mV), consistent with stable colloidal suspensions. NTA revealed that particle concentrations of lEVs ranged from 1.2 x 10¹⁰ ± 1.3 x 10¹⁰ to 7.7 x 10¹⁰ ± 2.1 x 10⁸ particles/mL, whereas sEVs exhibited concentrations ranging from 6.3 x 10¹⁰ ± 3.2 x 10¹⁰ to 5.2 x 10¹⁰ ± 8.3 x 10⁸ particles/mL. Western blotting confirmed expression of subtype-specific markers, with CD63 being expressed in sEV fraction and Arf6 present in lEVs. Functional assays revealed that both EV subtypes enhanced relative ATP levels in recipient BECs, an EV dose-dependent restoration of ATP levels was noted in sEV-treated BECs compared to lEVs. These effects are attributed to the presence of heat shock proteins in the EV cargo¹, suggesting that bioreactor BEC cultures produce EVs with properties consistent with that of BECs batch-cultured in T-175 tissue culture flasks.
Conclusion
Bioreactor BEC cultures enable scalable, high-yield production of functionally active EVs. These findings highlight their potential for further development for translational therapeutic applications involving metabolic restoration.

Poster 19 (Pre-Candidacy)

Mechanistic Understanding of MED13 Regulation and Its Promise as a Cancer Therapeutic Target

Hsiang-Ching Tseng^1,2, Leon Palao III³, Tao Li², Shin-Fu Chen², Kenji Murakami³, and Kuang-Lei Tsai^1,2*
1The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030; ²Department of Biochemistry and Molecular Biology, McGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX 77030; ³ Department of Biochemistry and Biophysics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104

Dysregulation of gene expression is a hallmark of numerous diseases, including cancer, and the Mediator complex is believed to play a crucial role in coordinating RNA Polymerase II (Pol II)-dependent transcription. This complex consists of the Core Mediator (cMED) and a dissociable CDK8 kinase module (CKM), which includes MED12, MED13, CDK8, and Cyclin C. While the CKM is known to drive transcriptional reprogramming in tumors, the molecular mechanisms governing its recruitment and dissociation remain poorly defined.

A key component within this CKM is MED13, which contains a ~700-amino-acid-long intrinsically disordered region (IDR). Using cross-species sequence alignment, we identified evolutionarily conserved motifs within the MED13-IDR that are essential for the CKM–cMED association. Our in vitro yeast transcription assay and competition binding assays demonstrate that the IDR sequences are crucial for robust transcriptional repression. Disruption of these motifs weakens CKM–cMED binding, leading to impaired transcriptional control and phenotypic defects, including temperature sensitivity and abnormal cellular aggregation.

Notably, in human cells, MED13-IDR overexpression suppresses cancer-related gene expression, underscoring its regulatory importance. Given that the MED13-IDR is extensively phosphorylated, I hypothesized that post-translational modifications (PTMs) act as a molecular fine-tuner of CKM association. Specifically, luciferase screening assays indicated that phosphorylation at specific residues profoundly affects transcription levels, as evidenced by the expression of phospho-mimetic or phospho-deficient mutants. To further elucidate this mechanism, pull-down assays and Transmission Electron Microscopy (TEM) observation will be employed to characterize how CDK8-mediated phosphorylation of the MED13-IDR promotes CKM dissociation, thereby de-repressing RNA Pol II-dependent transcription. Lastly, by integrating RNA-seq with CUT&Tag profiling, I aim to map the global dynamics of MED13-IDR in the phosphorylated and unphosphorylated states and their broader impact on transcriptional control in cancer.

Our findings provide a molecular basis for CKM-mediated repression and highlight a novel regulatory mechanism that could inform future therapeutic strategies, such as chemical-induced proximity (CIP) to target transcriptional dysregulation in cancer.

Acknowledgements: This work is supported by the US National Institutes of Health grants R01 GM143587 (K.L.T.), the Welch Foundation (AU-2050-20200401), and Cancer Prevention & Research Institute of Texas (CPRIT RP210045).

Poster 20 (Post-Candidacy)

Decoding p53(R249S)-mutated liver cancer: from developmental origins to targeted therapies

Megan E. Fisher^1,2, Yu-Wen Huang², Mo-Fan Huang^1,2, Ruiying Zhao², Dung-Fang Lee^1,2
1The University of Texas MD Anderson Cancer Center UTHealth, Graduate School of Biomedical Sciences, Houston, TX, USA
^,2Department of Integrative Biology and Pharmacology, McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX, USA

Hepatocellular carcinoma (HCC) is the most common form of primary liver cancer and a leading cause of cancer-related deaths worldwide. Unlike other cancers, HCC incidence has sharply increased over the past four decades, while therapeutic advances remain limited. Current treatment options, including surgical resection and locoregional therapies, provide modest survival benefit, underscoring the need for mechanistically informed and mutation-specific therapeutic strategies.

One of the most frequent genetic alterations in HCC is the hotspot mutation p53(R249S), which has been linked to aggressive disease and poor prognosis. Although p53 is canonically recognized as a tumor suppressor, the R249S mutation exhibits gain-of-function properties that promote tumorigenesis. However, most studies investigating this mutation rely on established cancer cell lines harboring numerous additional genomic alterations, making it difficult to define the direct consequences of mutant p53 activity.

To address this limitation, we utilized a human pluripotent stem cell (hPSC) model engineered to harbor only the p53(R249S) mutation. These cells were differentiated into hepatocyte-like cells, enabling investigation of mutant p53 function within a controlled, non-transformed genetic background that recapitulates early hepatocellular development. Transcriptomic profiling and pathway enrichment analyses revealed robust upregulation of FOXM1, a transcription factor that regulates cell cycle progression and genomic stability, as a key downstream effector associated with p53(R249S). Elevated FOXM1 expression correlates with poor clinical prognosis in HCC patients, supporting its potential clinical relevance.

To evaluate therapeutic vulnerability within this axis, we treated HCC cell lines with thiostrepton, a reported FOXM1 inhibitor. Thiostrepton significantly reduced cell viability and impaired competitive growth, suggesting that FOXM1-associated proliferative programs are functionally important for tumor cell maintenance.

Ongoing studies aim to assess thiostrepton efficacy in vivo, both as monotherapy and in combination with standard-of-care treatments. In parallel, integrated computational genomic and chromatin analyses are being performed to define the mechanistic basis of p53(R249S)-mediated FOXM1 upregulation. Collectively, our work identifies a novel p53(R249S)-FOXM1 oncogenic axis and highlights the power of stem cell-based modeling to uncover mutation-specific therapeutic vulnerabilities in HCC.

Poster 21 (Post-Candidacy)

Muscle estrogen-related receptors a/g mitigate muscle wasting in cancer cachexia

Anna DeBruine^1,2, Svitlana Poliakova², Danesh Sopariwala, Ph.D., Vihang Narkar, Ph.D.^1,2
1Therapeutics and Pharmacology Program, The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, Texas, USA;
²Center for Metabolic and Degenerative Diseases, Institute of Molecular Medicine, McGovern Medical School, UTHealth Houston, Texas, USA

Cancer cachexia is a devastating muscle-wasting condition leading to poor quality of life and higher mortality rates in cancer patients. While cachexia is responsible for nearly 30% of cancer deaths, there are currently no effective clinical interventions, highlighting a critical need for therapeutic target discovery for managing cachexia. We previously showed that estrogen-related receptors α and γ (ERRα/γ) are not only indispensable muscle fitness factors but also promising therapeutic targets for peripheral arterial disease and Duchenne Muscular Dystrophy. Given the integral role of ERRs in skeletal muscle homeostasis, we postulate that ERRα/γ could play a mitigative role in cachectic muscle wasting. Our aims to test this hypothesis are twofold: 1) Identify and measure the effect of cancer-derived factors on ERRα/γ signaling in skeletal muscle, and 2) Investigate how ERRα/γ modulation in skeletal muscle affects cancer cachexia development. We have found that ERRα/γ protein and transcripts, as well as their mitochondrial and angiogenic gene targets, are repressed in atrophic muscles from murine models of cancer cachexia (lung and pancreatic). To recapitulate cancer cachexia in vitro, we treated differentiated C2C12 mouse muscle cells with conditioned media from various cancer cells, which resulted in myofiber atrophy. Overexpression of ERRα/γ alone or in combination via adeno-associated virus prevented cancer cell-conditioned media induced atrophy and repressed expression of cachexia-associated wasting genes. Conversely, an ERRα inverse agonist (XCT-790) in conjunction with cancer cell conditioned media worsened atrophy and expression of cachectic wasting markers. For Aim 2, we use the KPC mouse model, a Kras/p53 mutant pancreatic cancer orthotopic model, which exhibits severe muscle wasting.  Intramuscular injection of an adeno-associated virus (AAV9) carrying Esrra (ERRα) gene reduced cachexia-related wasting genes, while increasing metabolic and angiogenic genes in the atrophic skeletal muscle. Strikingly, intramuscular AAV9-Esrra injection not only mitigated cancer-induced muscle wasting but also restored muscle contractile function in the KPC mice. In summary, our data suggest that deficiency of ERRs in skeletal muscles of tumor-bearing mice could contribute to cachexia and gene therapy-mediated restoration of ERR pathways ameliorates wasting. This project presents ERRs as a novel, muscle-specific therapeutic target for managing cancer-induced cachexia.

Poster 22 (Ineligible for Prizes)

Prefrontal cortex astrocytes modulate distinct neuronal populations to control anxiety-like behavior

Eunyoung Kim^1,2, Brandon L. Brown³, and Xinzhu Yu^2
1Neuroscience Program, University of Illinois at Urbana-Champaign, Urbana, IL
²Institute of molecular medicine, Center for Neuroimmunology and Glial Biology, UThealth Houston, Houston, TX
³UT Physicians Interventional Psychiatry Clinic, UThealth Houston, Houston, TX

Anxiety disorders are one of the most common psychiatric diseases with a detrimental impact on the quality of life of patients with anxiety disorders. Unfortunately, current anxiolytic medications are ineffective since around half of patients fail to respond to initial treatment. In order to develop novel therapeutic treatments for anxiety disorders, understanding the neurobiological mechanism is crucial. The medial prefrontal cortex (mPFC) enacts an essential role as a central hub of microcircuits that regulate emotions including anxiety. Numerous clinical and research studies indicate essential neuronal contribution to anxiety in this brain region, yet the mechanisms of non-neuronal cells governing this emotional state remain poorly understood. Astrocytes are one of the most abundant non-neuronal cells in the brain and interact closely with neurons and other glial cells. Dissimilar to neurons, astrocytes utilize intracellular calcium signals rather than electrical signals to interact with other cells. However, the role of astrocytes in mPFC and their effects on anxiety-like behavior are largely unknown. Here, we show that non-neuronal astrocytes in the mPFC encode anxiogenic environmental cues. Silencing mPFC astrocyte Ca²⁺ signaling heightens anxiety-like behavior and disrupts the balance between excitatory and inhibitory neuronal population activity. Moreover, this effect is differentially pronounced across behaviorally tuned neuronal subpopulations on single-cell and network levels. Collectively, our findings suggest a novel homeostatic plasticity mechanism by which prefrontal astrocytes regulate discrete neuronal populations related to anxiety, offering insights into the pathophysiology and potential therapeutic interventions for emotional disorders.

Poster 23 (Post-Candidacy)

To Beat or Not to Beat: Hippo Signaling Controls Development of the Heart’s Natural Pacemaker

Julianna Quinn^1,2, Xiaotong Chen^2,3, Nam Troung², Mingjie Zheng², Xiao Li³, Jun Wang^1,2.
¹The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences, Houston, TX, USA
²The University of Texas Health Science Center Houston, Department of Pediatrics, Houston, TX, USA
³Texas Heart Institute at Baylor College of Medicine, Houston, TX, USA

Cardiac arrhythmias are a leading cause of cardiac-related mortality worldwide and can arise from sinoatrial node (SAN) dysfunction. The SAN is the heart’s natural pacemaker and contains specialized cardiomyocytes (CMs) known as pacemaker cells (PCs). PCs generate spontaneous action potentials to drive the heartbeat. Despite its essential role, the SAN remains the least understood cardiac component, and the molecular mechanisms underlying its development are poorly defined. Technical challenges have limited studies of the developing SAN, and only a few transcription factors (TFs) are known to regulate PC differentiation. Our lab recently demonstrated that canonical Hippo signaling is required for maintaining adult SAN homeostasis. During development, Hippo signaling is well known to regulate heart size by inhibiting CM proliferation. Preliminary data suggest Hippo signaling is selectively upregulated in SAN PCs compared to surrounding CMs. We therefore hypothesize that Hippo activity regulates SAN morphogenesis and PC differentiation during development. To test this, we conditionally knocked out (CKO) the Hippo kinases Lats1/2 in embryonic SAN PCs. Lats1/2 CKO mice survived postnatally but developed severe arrhythmias persisting into adulthood. Embryonic SANs exhibited increased PC proliferation, leading to SAN enlargement. Single-cell RNA sequencing revealed a unique population of PCs in embryonic Lats1/2 CKO SANs. This new population has increased expression of cell cycle genes, including regulators within the IGF signaling pathway. Furthermore, core TFs required for PC differentiation were dysregulated. Together, these findings demonstrate that Hippo signaling is a key regulator of PC proliferation and differentiation during SAN development. This study offers new insights into the molecular mechanisms that govern SAN formation and arrhythmia susceptibility. Future studies will investigate how transcriptional changes induced by Hippo inhibition regulate PC function using patch-sequencing technology.

Poster 24 (Post-Candidacy)

Contrasting roles of PVH^CRH neuron neurotransmission in diet-induced obesity

Yuhan Cao^1,2,5, Zhiying Jiang^1,5, Michelle He¹, Maojie Yang¹, Jing Cai^1,2, Hongli Li¹, Yuanzhong Xu¹, Shibin Li³, Yamaguchi Hiroshi³, Luis de Lecea³, Benjamin R Arenkiel⁴, and Qingchun Tong^{1,2, *
1}The Brown Foundation Institute of Molecular Medicine for the Prevention of Human Diseases, The University of Texas Health Science Center at Houston, Texas, USA, 77030.
²MD Anderson Cancer Center & UTHealth Houston Graduate School for Biomedical Sciences,University of Texas Health Science at Houston, TX 77030.
³Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine,Stanford, CA. Wu Tsai Neurosciences Institute, Stanford University, Stanford, CA 94305, USA.
⁴Department of Molecular and Human Genetics and Department of Neuroscience, Baylor College of Medicine, and Jan and Dan Duncan Neurological Research Institute, Texas Children’s Hospital
⁵These authors contribute equally to this work

*Correspondence: Qingchun Tong, Ph.D.
The Brown Foundation Institute of Molecular Medicine for the Prevention of Human Diseases, The University of Texas Health Science Center at Houston,1825 Pressler Street, Houston, Texas, USA, 77030

Stress-induced obesity has become an increasingly prevalent and realistic issue in today’s society. The hypothalamic-pituitary-adrenal (HPA) axis is known to connect psychological stress to peripheral bodily responses. Chronic stress can disrupt HPA axis function, leading to increased cortisol or corticosterone (Cort) levels, promoted appetite, the preference for calorie-dense foods, and abdominal fat accumulation. Conversely, obesity itself can act as a stressor, triggering psychological distress and further activating the stress response. While this bidirectional relationship between obesity and stress is well recognized, the underlying neural mechanisms remain unclear.

Corticotropin-releasing hormone (CRH)-expressing neurons in the paraventricular hypothalamus (PVH^CRH neurons) play a key role in the HPA axis by releasing CRH, which stimulates downstream stress-related hormones. PVH^CRH neurons are responsive to stress, but their activity can be suppressed by consuming calorie-dense foods, suggesting a role in stress-related obesity. These neurons release both CRH and glutamate. Embryonic deletion of the CRH gene in PVH^CRH neurons leads to adrenal gland deficits, underscoring the essential role of CRH in development, whereas the role of glutamate release from these neurons is less understood.

In this study, we found that disrupting glutamate release from PVH^CRH neurons in adulthood reduced diet-induced obesity (DIO), whereas deleting CRH had no effect. Conversely, increasing CRH release promoted DIO, but enhancing glutamate release did not. Notably, CRH deletion in PVH^CRH neurons did not affect basal corticosterone levels, but it did reduce corticosterone levels under stress. These findings demonstrate the contrasting roles of CRH and glutamate release from PVH^CRH neurons in DIO and suggest that CRH is not required to maintain basal corticosterone levels in the absence of stress.

Poster 25 (Pre-Candidacy)

Targeting GPR56 to Suppress Colorectal Cancer Stemness and Metastasis through Next-Generation Antibody–Drug Conjugates

Wanqing Cheng^1,2, Zhengdong Liang¹, Yueh-Ming Shyu^1,2, Kendra Carmon^1,2
1Center for Translational Cancer Research, The Brown Foundation Institute of Molecular Medicine, University of Texas Health Science Center, Houston, TX, 77030, USA; ²The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX, 77030, USA

Distant metastasis and relapse are the leading causes of mortality in Colorectal cancer (CRC), as approximately 20-25% of patients develop metastatic disease, where the 5-year overall survival is about 15%. Despite the availability of therapies for specific subtypes of mCRC, outcomes remain poor for most patients due to tumor heterogeneity/mutation status, therapy-related toxicity and resistance, and the persistence of cancer stem-like cells (CSCs), highlighting the urgent need for new targeted therapeutics. Antibody-drug conjugates (ADCs) are an emerging class of anticancer agents that harness the specificity of monoclonal antibodies (mAbs) to precisely deliver potent cytotoxic payloads to tumor cells, while minimizing toxicity. Our previous studies have demonstrated that adhesion G protein–coupled receptor 56 (GPR56) is a viable therapeutic target for the treatment of CRC. GPR56 is aberrantly overexpressed in 65–80% of CRC and is associated with the microsatellite stable (MSS) subtype and poor clinical outcomes. We have shown that GPR56 promotes tumor growth, chemoresistance, and alters expression of the CSC marker LGR5. Furthermore, we developed GPR56-targeted ADCs incorporating in-house mAb 10C7 and the DNA-alkylating agent duocarmycin SA (DMSA), which exhibited strong selectivity and potent antitumor activity in GPR56-high CRC cells and patient-derived xenografts (PDXs) without overt toxicity. However, incomplete tumor elimination indicates the need for improved therapeutic efficacy and strategies. We therefore hypothesize that GPR56 helps promote stemness and the metastatic potential of CRC cells, which can be effectively inhibited using a novel optimized GPR56 ADC with enhanced efficacy. Using CRISPR/Cas9, we generated GPR56 knockout metastatic CRC cells expressing mCherry–luciferase to enable orthotopic implantation and in vivo imaging of metastatic progression. We also generated GPR56-overexpressing human and murine CRC cell lines that exhibit enhanced proliferation and are evaluating its impact on CSC phenotypes (e.g., LGR5 and CD133), clonogenicity, migration/invasion and chemoresistance. Further, we are developing optimized GPR56-targeted ADCs incorporating diverse cytotoxic payloads to enhance therapeutic efficacy. Affinity maturation was used to generate a novel anti-GPR56 mAb, M6, which displays >6-fold higher binding affinity compared to the 1st generation 10C7 mAb. Additionally, M6 was engineered for site-specific conjugation to payloads. We are now generating M6 ADCs incorporating DMSA payloads to evaluate its improved potency and antitumor efficacy compared to our 1st generation 10C7 ADC. This project aims to dissect the mechanistic role of GPR56 in CRC metastasis and to develop and optimize next-generation GPR56 ADC therapies for advanced CRC. By validating GPR56 as a therapeutic target in CRC, our work has the potential to expand the treatable patient population, particularly within the predominant MSS subtype, for which effective treatment options remain limited.

Poster 26 (Ineligible for Prizes)

The Role of Astrocytic Glutamate Dysregulation in the Paraventricular Nucleus of Thalamus Drives Adult Anxiety Susceptibility Induced by Adolescent Repeated Alcohol Exposure

Aubrey Bennett¹, Hyunjung Kim¹, and Seungwoo Kang¹

¹Department of Neuroimmunology and Glial Biology, Institute of Molecular Medicine, The University of Texas Health Science Center, Houston, Texas, USA.

Adolescence is a critical period for the maturation of brain development, and alcohol exposure during this developmental window increases vulnerability to long-lasting physical and psychiatric disorders. Repeated ethanol exposure during adolescence is strongly associated with the emergence of anxiety disorders in adulthood, yet the underlying neural mechanisms remain poorly understood. The paraventricular nucleus of the thalamus (PVT) is a key hub for anxiety regulation and is particularly sensitive to early-life experiences. Here, we investigated how adolescent intermittent ethanol exposure (AIE) alters PVT activity and contributes to anxiety-like behaviors in adulthood. Using in-vivo calcium imaging, chemogenetic manipulation, and magnetic resonance spectroscopy, in combination with behavioral and molecular assays, we assessed neuronal and astrocytic adaptations within the PVT following AIE. Adult mice previously exposed to AIE exhibited increased anxiety-like behaviors accompanied by elevated neuronal firing rates and enhanced calcium signaling in PVT neurons compared to control mice. Chemogenetic inhibition of PVT neurons significantly attenuated anxiety-like behaviors in AIE mice, confirming a functional role for PVT hyperactivity in ethanol-induced adult anxiety susceptibility. Mechanistically, increased PVT neuronal activity was mediated, at least in part, by reduced expression of GLT1, an astrocyte-dominant glutamate transporter (also known as EAAT2, Slc1a2). Magnetic resonance spectroscopy revealed elevated glutamate/GABA ratios in the thalamus of mice with GLT1 knockdown, which was associated with heightened anxiety-like behaviors. Importantly, selective deletion of GLT1 in PVT astrocytes of ethanol-naïve mice mimicked anxiety-like behaviors, whereas enhancement of GLT1 expression in PVT astrocytes of AIE-exposed mice alleviated AIE-induced anxiety. Together, these findings demonstrate that repeated adolescent ethanol exposure induces persistent anxiety susceptibility through astrocyte-mediated glutamatergic dysregulation within the PVT. Specifically, astrocytic GLT1 in the PVT plays a critical role in regulating neuronal excitability and anxiety-like behaviors in adulthood, highlighting GLT1 as a potential therapeutic target for alcohol use disorder and comorbid emotional disorders.

Poster 27 (Ineligible for Prizes)

Linking Metabolism and Differentiation in Vascular Smooth Muscle Cells.

Jose Emiliano Esparza Pinelo¹, Alexei Robin¹, Aminat Dosunmu¹, Alexis Richard¹, Jeison Garcia Serrano¹, Angie Danhi Gonzalez¹,Hannah Krenz¹, Jessica Chen¹, Mario Carrera¹, Anita Kaw¹, Dianna M. Milewicz¹, Callie S. Kwartler^1
1Division of Medical Genetics, Department of Internal Medicine, McGovern Medical School, The University of Texas Health Science Center at Houston, TX, USA.

Phenotypic modulation of smooth muscle cells (SMCs) is a major driver of cardiovascular disease. This is shown in patients harboring pathogenic variants in ACTA2 altering arginine 179, causing Smooth Muscle Dysfunction Syndrome, which manifests as thoracic aortic disease, moyamoya-like cerebrovascular disease, and other vascular complications. Like tumor cells, modulated SMCs can have an altered metabolic profile of increased glycolysis and decreased oxidative phosphorylation (OXPHOS). Acta2^R179C/+SMCs fail to fully differentiate and retain stem cell phenotypes, consistent with a burdensome reliability in glycolysis. Our findings suggest an inherent relationship between metabolism and differentiation in SMCs. For instance, siRNA-mediated knockdown of electron transport chain Complex IV component Cox4i1 leads to dedifferentiation of wild-type (WT) SMCs. Treatment with nicotinamide riboside (NR) shifts Acta2^R179C/+ SMC metabolism from glycolysis towards OXPHOS, while simultaneously decreasing migration and increasing expression of contractile SMC markers. However, the mechanisms linking SMC phenotype and metabolism remains poorly understood. Ten-eleven translocation 2 (TET2), an established regulator of SMC differentiation, is a DNA demethylase which oxidizes 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC), leading to DNA demethylation and gene activation. TCA cycle metabolite 2-hydroxygluterate (2HG) competitively inhibits α-keto-gluterate-(αKG)-dependent dioxygenases, including TET2. Glucose flux experiments show a higher accumulation of 2HG in Acta2^R179C/+ SMCs. We hypothesize that accumulation of 2HG in SMCs with high glycolytic activity leads to inhibition of TET2, as well as subsequent epigenetic changes that decrease expression of SMC differentiation genes. NR treatment decreases flux through glycolysis and TCA cycle metabolites including 2HG in both WT and Acta2^R179C/+ SMCs. Chromatin immunoprecipitation (ChIP) assays determined that TET2 binding to SMC differentiation genes is decreased in Acta2^R179C/+ SMCs and increased with NR. Global and site-specific decreases in 5mC and increases in 5hmC following NR treatment was confirmed using Hydroxy methylated DNA immunoprecipitations. To determine if these findings represent a generalizable mechanism linking SMC metabolism with differentiation, WT SMCs were treated with 2HG, which induced dedifferentiation and decreased 5hmC and TET2 levels. Knockdown of enzyme L2HGDH, which metabolizes L-2HG back to αKG, leads to an accumulation of L-2HG in WT SMCs. This accumulation in turn reduces key SMC marker expression due to changes in 5hmC patterns via TET2 inhibition. Finally, the addition of exogenous 2HG prevents NR from increasing contractile SMC markers in Acta2^R179C/+ SMCs. These findings suggest that the 2HG-TET2 axis bridges a metabolic switch with SMC differentiation. This work suggests novel metabolic treatment targets to restore SMC differentiation and prevent pathologic phenotypic modulation in cardiovascular diseases.

Poster 28 (Ineligible for Prizes)

Metabolic Signatures of Suicide Risk: The Role of Diet and Metabolic Function

 Ana Cristina Ruiz¹, Steven De La Garza², Valeria Arredondo¹, Gabriel R. Fries^1,2
1School of Behavioral Health Sciences, The University of Texas Health Science Center at Houston, Houston, TX, USA; ²Faillace Department of Psychiatry and Behavioral Sciences, McGovern Medical School, The University of Texas Health Science Center at Houston, McGovern Medical School, Houston, TX, USA.

Background: An association between mental health and abnormalities in several metabolic parameters has been consistently reported across studies. Specifically, changes in blood glucose levels, lipid profiles, and metabolic syndrome have been linked to a range of psychiatric conditions. Although the relationship between metabolic dysfunction and psychiatric disorders is well established, far fewer studies have examined how specific metabolic parameters may influence suicide risk. Given the substantial number of individuals who die by suicide each year, it is critical to deepen our understanding of how abnormalities in these metabolic factors might contribute to heightened risk for suicidal ideation (SI) and suicidal behaviors.

Aims: To assess the potential effects of multiple metabolic parameters (i.e., dietary inflammatory index (DII), triglyceride glucose index (TyG), dietary sugar intake, and cardiometabolic index (CMI)) on suicide risk.

Methods: Publicly available data from the National Health and Nutrition Examination Survey (NHANES) were analyzed, specifically from the 2005–2023 collection cycles (n = 50,255 with available Patient Health Questionnaire (PHQ-9) data). Suicide risk was assessed using item #9 of the PHQ-9,  which specifically evaluates SI. Weighted logistic regression models were used to examine the association between the predictors (TyG, CMI, dietary sugar intake, and DII) and SI. Model performance was evaluated using receiver operating characteristic curves and corresponding area under the curve values. Covariates were included based on emerging evidence suggesting their potential associations with both the predictor variables and suicide outcomes.

Results: When controlling for age, sex, race/ethnicity, education, poverty index ratio, and marital status, a higher TyG index significantly predicted SI (p = 0.04; OR = 1.28; 95% CI = [1.008, 1.636]). Using the same covariates, DII was also a significant predictor of SI (p = 0.018; OR = 1.06; 95% CI = [1.01, 1.113]), as well as dietary sugar intake (p = 0.01; OR = 1.002; 95% CI = [1.0, 1.003]). In contrast, CMI was not significantly associated with suicide risk (p < 0.08; OR = 1.00; 95% CI = [1.00, 1.001]).

Conclusions: In this preliminary analysis, higher TyG index and dietary sugar intake were associated with a greater predicted probability of suicide risk. DII, which reflects diet quality and inflammation in individuals with a high risk of metabolic conditions, also significantly predicted suicide risk. In contrast, CMI, which has been identified as a central component of metabolic syndrome, did not show a significant association with suicide risk in this sample. Overall, these findings highlight meaningful associations between key metabolic parameters and suicide risk in the general population, providing potential targets for suicide prevention strategies.

Poster 29  (Ineligible for Prizes)

Hepatic SIRT3 Restrains MASH-Associated Hepatocarcinogenesis through the Regulation of Choline-Betaine Metabolism

Qixiu Hou^1#, Shuying Wang^1#, Dali Han¹, Dongqing Yin¹, Constance Atkins¹, David R. Hall², Holger K. Eltzschig¹, Mark T. Miedel³, Cynthia Ju¹, Juli Bai^1
1Department of Anesthesiology, Critical Care and Pain Medicine, McGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX, 77030, USA
²Department of Surgery, McGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX, 77030, USA
³Organ Pathobiology and Therapeutics Institute, Department of Pharmacology and Chemical Biology, University of Pittsburgh School of Medicine, Pittsburgh, PA, 15219, USA

# These authors contribute equally to the work.

Metabolic dysfunction-associated steatohepatitis (MASH) is increasingly recognized as a major etiological driver of hepatocellular carcinoma (HCC). MASH is characterized by hepatocytic lipid accumulation, inflammation, fibrosis and cellular injury. However, notably, not all individuals with MASH progress to HCC. Therefore, elucidating the molecular mechanisms governing this transition represents a significant knowledge gap. Sirtuin 3 (SIRT3) is a mitochondrial NAD⁺-dependent deacetylase that regulates mitochondrial metabolism and cellular stress responses. While previous studies have linked SIRT3 deficiency to the development of hepatic steatosis, MASH, or HCC in both humans and mouse models, its role in diseases progression, particularly in the transition from MASH to HCC, remains largely unexplored. Interestingly, we found that hepatocyte-specific SIRT3 knockout (SIRT3^HepKO) mice exhibited a markedly increased incidence of HCC, along with severe MASH features, including steatosis, injury, and fibrosis, when subjected to 12 weeks of choline-deficient, amino acid-defined, high-fat-diet (CDAHFD), a duration that typically induces only mild MASH in controls. In contrast, under high-fat-diet conditions, SIRT3^HepKO mice displayed comparable hepatosteatosis and metabolic homeostasis to controls, suggesting that SIRT3 plays a critical role in choline metabolism and thereby influences MASH progression to HCC. Consistent with these findings, SIRT3 deficiency in hepatocytes reduced betaine production, suppressed monomethyl-phosphatidylethanolamine (MMPE) and phosphatidylcholine (PC) synthesis, and inhibited VLDL secretion. Mechanistically, we found that SIRT3 interacted with choline dehydrogenase (CHDH) and betaine aldehyde dehydrogenase (BADH), two mitochondrial enzymes responsible for choline-betaine conversion. Collectively, our findings identify hepatic SIRT3 as a critical metabolic gatekeeper that restrains MASH progression and hepatocarcinogenesis by regulating choline-betaine metabolism. This work establishes a mechanistic connection between mitochondrial protein deacetylation and metabolic liver cancer, highlighting SIRT3-dependent pathways as promising therapeutic targets for preventing MASH-associated HCC.

Poster 30 (Post-Candidacy)

Increased firing rate activity and dopamine release in the prelimbic cortex correlate with risky behavioral choice during an approach-avoidance conflict test in rats

Vicky Chuong¹, Guillermo Aquino-Miranda¹, Thays Brenner dos Santos¹, Fabricio H. Do Monte¹
Department of Neurobiology & Anatomy, The University of Texas Health Science Center at Houston, Houston, TX, USA¹

Neurons in the prelimbic (PL) subregion of the medial prefrontal cortex change their firing rates in response to fear- and reward-associated cues. PL activity is essential for the retrieval of fear- and reward-related memories. However, it remains unknown how the activity of PL neurons is modulated to shape behavioral choice during situations of motivational conflict when reward and fear cues co-occur. To explore this question, adult male Long-Evans rats previously implanted with single-unit recording electrodes in PL were initially trained to press a lever for sucrose during the presentation of audiovisual cues. Next, rats were exposed to an approach-avoidance conflict test consisting of three phases: (i) a reward phase, when only food cues were presented, (ii) a cat odor phase, when only fear-inducing cat odor was presented, and (iii) a conflict phase, when food cues were concomitantly presented with cat odor. Compared to the reward phase, rats displayed increased defensive behaviors and reduced food-seeking responses during the cat odor and conflict phases, respectively. During a contextual fear test in the following day in the absence of cat odor, rats displayed individual variability in their behavioral choice to the food cue trials. During food cue presentations, rats either approached the food area and pressed the lever for sucrose reward (risk-taking trials) or avoided the odor area and did not press the lever (risk-avoiding trials). Analysis of PL activity during food cue onset in the contextual fear test revealed a similar proportion of responsive neurons during the risk-taking trials (~25% of the cells) and risk-avoiding trials (~24% of the cells). However, only a fraction of these cells (~7%) responded in both trials, suggesting that distinct subpopulations of food cue responsive neurons are recruited according to the animal’s behavioral choice. Moreover, a trial-by-trial analysis of PL activity revealed that cells excited in response to food cues during risk-taking trials were attenuated in response to food cues during risk-avoiding trials, and vice versa. These results suggest that modifications in PL food cue responses may reflect distinct behavioral choices during reward and conflict phases. Given the role of the dopaminergic system during motivated behaviors, we next quantified the release of dopamine in PL during the motivational conflict using fiber photometry combined with the fluorescence dopamine sensor, GRAB_DA. We observed that dopamine levels increased in response to food cues during the reward phase, but this increase was attenuated during the conflict phase. Interestingly, during the contextual fear test, dopamine levels in response to food cues were significantly greater during risk-taking trials compared to risk-avoiding trials. Together, these findings suggests that increased PL activity and greater dopamine release in PL in response to food-associated cues is correlated with risky behavioral choice during motivational conflict. Ongoing experiments are testing the causal role of cue-induced increases in firing activity and dopamine release in the PL during approach-avoidance conflict.

Poster 31 (Post-Candidacy)

Functional Characterization of Hsp110 in Drosophila Reveals its Essential and Dosage-Sensitive Role in Nervous System Integrity

Beatriz Rios^1,2, , Shiyu Xu¹, Stephen M Farmer^1,2,3, Xin Ye¹, Lili Ye¹, Kevin A. Morano^4*, Sheng Zhang^1,5,6*
1Center for Metabolic and Degenerative Diseases, The Brown Foundation Institute of Molecular Medicine, McGovern Medical School at the University of Texas Health Science Center at Houston (UTHealth Houston), Houston, Texas, USA
²Program in Neuroscience, The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences (MD Anderson UTHealth Houston GSBS), Houston, Texas, USA
³Program in Molecular and Translational Biology, The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences (MD Anderson UTHealth Houston GSBS), Houston, Texas, USA
⁴Department of Microbiology & Molecular Genetics, McGovern Medical at the University of Texas Health Science Center at Houston (UTHealth Houston), Houston, Texas, USA
⁵Program in Genetics and Epigenetics, The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences (MD Anderson UTHealth Houston GSBS), Houston, Texas, USA
⁶Department of Neurobiology and Anatomy, McGovern Medical School at the University of Texas Health Science Center at Houston (UTHealth Houston), Houston, Texas, USA

The Hsp70 molecular chaperone system is the front line of defense in maintaining cellular proteostasis. In eukaryotes, ATP/ADP nucleotide exchange in the Hsp70 chaperone cycle is stimulated by Hsp110, a divergent member of the Hsp70 chaperone superfamily and co-chaperone of Hsp70. Hsp110 is also a known modifier of neurodegenerative and other protein misfolding-related disorders. Biochemical aspects of Hsp110 chaperone functions have been characterized in vitro, and pathway interactions have been extensively characterized genetically in yeast model systems; however, a detailed understanding of its physiological roles in metazoans, particularly in the nervous system has not been carried out. Taking advantage of the single Hsp110-encoding gene in the Drosophila genome, we conducted a comprehensive investigation of its expression and function in this animal model. Notably, Drosophila and human Hsp110 share significant similarity in their sequence, structure, and splicing variants. At the protein level, Hsp110 is ubiquitously expressed, with both cytosolic and nuclear distribution in a tissue-dependent manner. Functionally, while Hsp110 is dispensable for cell proliferation in developing larvae, it is essential for long-term cell survival and normal development of the nervous system, including non-autonomous effects on neuronal differentiation and glial cell migration. Furthermore, despite being identified as a potent suppressor of protein aggregation and neurotoxicity in multiple neurodegenerative diseases, higher levels of Hsp110 are detrimental in flies. Overexpression of Hsp40, another key co-chaperone of Hsp70, can mimic this effect. However, simultaneous overexpression of both Hsp40 and Hsp110 does not further exacerbate their detrimental effect. Together, these results demonstrate a critical role of Hsp110 in neuronal development and cell survival, and further suggest that in vivo, the levels and activities of Hsp110 and Hsp40 co-chaperones need to be properly balanced. Furthermore, this work supports the contention that the Hsp70 chaperone network must be considered as a whole when targeted for potential therapeutic purposes to meet the complex pathophysiological demands in multicellular organisms.

Poster 32 (Pre-candidacy)

Spatial transcriptomic analysis of astrocytes in Alzheimer’s disease mouse model

Yoon-Ze Shin¹, Eun-Young Kim^2,3, Xinzhu Yu^3
1 The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX, USA
² Neuroscience Program, University of Illinois at Urbana-Champaign, Urbana, IL, USA
³ Center for Neuroimmunology and Glial Biology, Brown Foundation Institute of Molecular Medicine, The University of Texas Health Science Center at Houston McGovern Medical School, Houston, TX, USA

Extensive studies have highlighted astrocytes as not only key regulators of brain homeostasis but also contributors to the pathogenesis of neurodegenerative disease such as Alzheimer’s disease (AD). Astrocytes undergo morphological, molecular, and functional changes and become reactive in response to pathological environment and could exacerbate or ameliorate disease progression. However, the heterogeneity of astrocytes and its interaction with adjacent cells in the context of AD are poorly understood. Here, using 5xFAD AD mice, we performed high-resolution spatial transcriptome analysis to elucidate alterations in the transcriptomic and interaction profiles of astrocytes during AD progression.  Consistent with previous research, reactive astrocyte signature genes were significantly upregulated in old AD astrocytes compared to old WT, whereas no sign of reactivity was observed in young WT or young AD. We further identified subpopulations of astrocytes that evolve during the AD progression. Moreover, we found different spatial patterns and gene enrichment of astrocyte subpopulations. Together, these analyses provide a spatiotemporal characterization of diverse groups of cortical astrocytes in the context of AD.

Poster 33  (Ineligible for Prizes)
Developing Phage Cocktails for Treatment of AMR-Staphylococcus aureus Orthopaedic Device-Related Infections

Cora Kosnik¹, Raquel C. Luna¹, Violeta Chavez², Catherine G. Ambrose³, Heidi B. Kaplan^1
1Microbiology and Molecular Genetics, ²Pathology and Laboratory Medicine, and ³Orthopedic Surgery, McGovern Medical School, UTHealth, Houston, TX

Orthopedic device-related infections (ODRIs) affect ~200,000 Americans annually and are a complication in approximately 3% of orthopedic device-related surgeries. The primary causative agent is Staphylococcus aureus, which is a challenging pathogen to treat due to its high incidence of antimicrobial resistance (AMR), ability to form biofilms, and prevalence in hospital settings. This task is further complicated by low blood flow to the affected bone, reducing the effectiveness of intravenous (IV) treatments. One alternative to IV antibiotic treatment is the use of bacteriophages (phage), which are viruses that selectively lyse bacteria while leaving host cells intact. To address these issues, we are designing a phage treatment for AMR S. aureus ODRIs using biodegradable microspheres to localize anti-staphylococcal phage directly to the site of infection.

The AMR S. aureus can also encode or evolve resistance to the phage used in treatment. As a result, many phage therapies employ a mixture of multiple phage, termed a phage cocktail, to reduce bacterial phage resistance and improve efficacy. A phage cocktail will commonly contain three to five different phage, each with a broad host range. We are designing an anti-staphylococcal phage cocktail by analyzing the host ranges of ten lytic phage. Each phage has been tested for its ability to lyse 63 S. aureus musculoskeletal clinical isolates from our biorepository using a plaque assay. Once four phage with a combined maximal host range are identified, they will be tested as single phage and in various combinations on our in vitro biofilm ODRI model.

Currently, I have identified four phage that together successfully lyse 56 (87.5%) of the 63 S. aureus clinical isolates in our biorepository of deep musculoskeletal infection strains and our reference strain S. aureus UAMS-1 (ATCC 49230). All seven of the remaining strains are partially or fully resistant to all ten phage in our library. I have also isolated 5 phage-resistant mutants of UAMS-1, which is an osteomyelitis clinical isolate, by selecting for growth in the presence of individual phage from our library. Specifically, I collected surviving cells after coculture with one phage on agar plates at a low multiplicity of infection (MOI). We are currently analyzing the whole genome sequence of the resistant strains to characterize the resistance mechanisms of at least a subset of these strains based on the locations of single nucleotide polymorphisms (SNPs). Finally, we plan to use a subset of these resistant strains to isolate phage that attach to different cell surface receptors and expect to include some of these new phage in an expanded phage cocktail.

We hypothesize that using these methods, our expanded phage cocktail will increase treatment efficiency, as most S. aureus strains will be susceptible to the cocktail, and resistance should not develop during treatment.

Poster 34 Post-Candidacy)

Autophagy influences neurodegenerative vulnerability in distinct neuron types during aging 

*Mya Rodriguez^{1, 2} Grace Augustine³, & Andrea Stavoe^{1, 2
1}The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX, USA
²Department of Neurobiology & Anatomy, McGovern Medical School, UTHealth Houston, Houston, TX, USA
³Trinity University, Department of Neuroscience, San Antonio, TX, USA

Neurodegenerative diseases like Alzheimer’s and Parkinson’s rob millions of their memories, movement, and independence. The biggest risk factor for developing these diseases is aging, highlighting the importance of studying how the brain changes with age. Multiple homeostatic and proteostatic processes become dysregulated during aging; one of these key processes is autophagy. Neuronal autophagy declines with age at a gross scale, however, autophagy’s neuron-type-specific age-related trends and contributions to age-related degeneration remain poorly understood. Here, we used C. elegans to systematically examine autophagic trends during aging in distinct neuron types. Using spinning disk confocal microscopy, we quantified endogenously labeled LGG-1 (LC3) autophagosomes in serotonergic and dopaminergic neurons across adulthood. Autophagosome number increased with age in serotonergic neurons but decreased in dopaminergic neurons, consistent with reduced autophagosome biogenesis. Notably, dopaminergic neurons exhibited age-dependent degeneration, whereas serotonergic neurons remained intact.

Building on prior work showing that enhancing WIPI2, an early autophagy component, restores autophagosome biogenesis in aged dorsal root ganglion neurons, we tested whether modulating WIPI2 activity could do the same in vivo and delay dopaminergic neurodegeneration. Using CRISPR/Cas9, we generated a phosphodead WIPI2 mutant and quantified neurodegeneration by assessing dopaminergic dendrite integrity during aging. Phosphodead mutants displayed significantly reduced neurodegeneration compared to wild-type animals at late adulthood. Together, these findings identify neuron-type-specific trends in autophagy during aging and are the first to demonstrate that modulating early autophagy components can mitigate age-related dopaminergic neurodegeneration, revealing a potential mechanistic link between the modulation of autophagosome biogenesis and neuronal vulnerability. Based on these results, we hypothesize that phosphodead WIPI2 restores autophagosome biogenesis, which will be further tested in the future using endogenous autophagosome markers. Other future directions include assessing the other side of the WIPI2 phosphorylation spectrum by using a phosphomimetic mutant.

Poster 35 (Ineligible for Prizes)

Determining the Role of Astrocytic Nf1 Mutation in Driving Ras-dependent Neurological Deficits in Neurofibromatosis Type 1 (NF1) 

Alesandra Echeandía Marrero¹, Anand Singh¹, Momo Harris¹, Cheng-En Shen¹, Chhay Hok¹, Kechen Ban¹, Renae E. Bertrand¹, Sama A. Alnassiry¹, Benjamin Deneen², Yuan Pan^1,3
1Department of Symptom Research, the University of Texas MD Anderson Cancer Center, Houston TX
²Center for Cell and Gene Therapy, Department of Neuroscience, and Department of Neurosurgery, Baylor College of Medicine, Houston TX
³Department of Neuro-Oncology, the University of Texas MD Anderson Cancer Center, Houston TX

Neurofibromatosis Type 1 (NF1) is a genetic disorder affecting ~1 in 3,000 individuals and is best known as a tumor-predisposition syndrome. About half of patients inherit an NF1 mutation, while the rest arise from de novo mutations. All patients carry a monoallelic NF1 mutation, and tumor formation or other complications can occur when a second somatic “hit” causes biallelic loss. The NF1 gene encodes neurofibromin, a Ras-GTPase activating protein that accelerates conversion of Ras-GTP to Ras-GDP. Loss of neurofibromin leads to Ras hyperactivation and increased downstream MAPK/ERK and PI3K/AKT signaling. Although Ras signaling is widely studied in cancer, it also plays essential roles in cognition and behavior. NF1 patients exhibit features such as café-au-lait macules, neurofibromas, and optic pathway gliomas; however, up to 75% experience neurological or cognitive impairments, including learning deficits, attention-related issues, and motor dysfunction. These symptoms overlap with other Ras/MAPK-related disorders (RASopathies), including Noonan and Costello syndromes, all of which display neurobehavioral phenotypes. Early studies in global Nf1-mutant mice demonstrated learning and motor impairments driven by Ras hyperactivation, and pharmacologic Ras inhibition rescued these deficits. While neuron-, oligodendrocyte-, and microglia-specific Nf1 models have provided insight into cell-type-specific contributions to behavior, the role of astrocytes remains poorly defined. Astrocytes regulate CNS homeostasis, including neurotransmitter uptake, ion balance, BBB support, and synaptic development and refinement. Recent work shows that Ras-MAPK pathway mutations disrupt astrocyte function and can impair cognition, suggesting astrocytes may play a critical role in Nf1-associated neurological deficits. To address this gap, we generated an astrocyte-specific Nf1-mutant mouse model to determine how Nf1 loss alters astrocyte signaling and contributes to behavioral dysfunction. We will isolate astrocytes to examine Ras pathway hyperactivation and transcriptional changes, and assess cognition and motor behavior in vivo using the Puzzle Box, Open Field, Novel Object Place Recognition, and Beam Walk tests. Preliminary results indicate emerging behavioral deficits in astrocyte-specific Nf1 mutants. This work will clarify astrocyte-specific mechanisms underlying NF1 and may reveal new therapeutic targets for NF1-associated neurobehavioral symptoms.

Poster 36 (Ineligible for Prizes)

Investigating the role of astrocytes in Alzheimer’s disease pathology

Hairuo Du^1,2, Eunyoung Kim^1,2, Yoon Ze Shin^2,3, Xinzhu Yu^2
1Neuroscience Program, University of Illinois at Urbana-Champaign, Urbana, IL, USA
²Center for Neuroimmunology and Glial Biology, Brown Foundation Institute of Molecular Medicine, McGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX, USA
³The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX, USA

Alzheimer’s disease (AD) is the most common cause of dementia worldwide and represents a major and growing public health challenge. While extensive research has focused on neuronal mechanisms underlying AD progression, the contribution of non-neuronal cell types in AD pathology remains insufficiently understood. As one of the most abundant resident cells tiling the entire central nervous system, astrocytes form intimate associations with neurons. Astrocytes express numerous transporters, channels and receptors that are essential to monitor, integrate and maintain neuronal activity. A key mechanism of such interactions is astrocyte Ca²⁺ signaling. Previous research has implied that astrocytes play an essential role in regulating neural homeostasis, synaptic function, ion balance, and blood–brain barrier integrity, processes that are severely disrupted in Alzheimer’s disease (AD), suggesting astrocyte participation in AD pathology. To investigate the role of astrocytes in AD progression, we used the 5xFAD mouse model to assess behavioral performance and astrocyte Ca²⁺ signaling at young (3–4 months) and old (12–13 months) stages. We found that old, but not young, 5xFAD mice exhibited impaired spatial working memory and reduced anxiety-like behavior compared with age-matched controls. Ex vivo calcium imaging revealed age-dependent alterations in astrocyte calcium activity. Young 5xFAD astrocytes showed decreased spontaneous Ca²⁺ activity, whereas old 5xFAD astrocytes exhibited increased spontaneous activity. In addition, dopamine-evoked Ca²⁺ responses were reduced in 5xFAD astrocytes at both young and old stages. Together, these findings demonstrate that astrocyte Ca²⁺ signaling undergoes dynamic, age-dependent alterations during AD progression accompanying behavioral impairments. Our results provide new insight into the potential contribution of astrocyte dysfunction to AD pathology and offer a basis for further investigation of the molecular mechanisms by which astrocytes contribute to behavioral alterations in AD.

Poster 37 (Masters)

Role of Interferon Regulatory Factor 7 (IRF7) in Macrophage in Dermal Fibrosis in Systemic Sclerosis​

Trithi Sunder¹², Minghua Wu², Brian Skaug², Elena Dyukova², Maureen Mayes², Shervin Assassi^2
1MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences​, Houston, TX, USA, 77030 ²Division of Rheumatology, Department of Internal Medicine, McGovern Medical School, The University of Texas Health Science Center, Houston, TX, USA, 77030

Systemic sclerosis (SSc) is a rare autoimmune disease characterized by immune dysregulation, vasculopathy, and progressive fibrosis of the skin and internal organs. Interstitial lung disease is a leading cause of morbidity and mortality. Type I interferon signaling is consistently elevated in SSc and correlates with disease severity. Interferon regulatory factor 7 (IRF7), a transcription factor central to type I interferon responses, has emerged as a potential regulator of fibrotic pathways. While fibroblast-intrinsic IRF7 has been linked to profibrotic signaling, IRF7 expression is also increased in macrophages within SSc skin and in experimental models, suggesting that macrophage-specific IRF7 may drive inflammation and fibroblast activation. However, its precise contribution in macrophages remains poorly defined.

To characterize IRF7 expression in human disease, we analyzed skin sections from 10 patients with SSc and matched healthy controls. Immunophenotypic assessment demonstrated increased macrophage infiltration in SSc skin, with a higher proportion of IRF7-expressing macrophages compared with controls. Bulk RNA sequencing of patient and matched healthy skin samples was performed to evaluate profibrotic gene expression and define the broader transcriptomic landscape associated with SSc.

To directly interrogate the role of macrophage IRF7 in fibrosis, we developed and validated a macrophage-specific IRF7 knockout mouse model. Dermal fibrosis was induced using the bleomycin model to interrogate the role of macrophage-specific IRF7 in fibrotic remodeling in vivo. Ongoing histologic and molecular analyses will assess dermal thickening, collagen deposition, and expression of key fibrotic and inflammatory genes in IRF7-deficient mice compared with controls.

To define the molecular programs regulated by IRF7 in macrophages, transcriptomic profiling of IRF7-deficient macrophages is in progress to characterize changes in inflammatory and profibrotic gene expression. In parallel, human SSc precision-cut skin slice experiments are being conducted to evaluate the effects of type I interferon pathway blockade on macrophage activation and tissue fibrosis, with the goal of establishing translational relevance and serving as an effective platform for pre-clinical studies.

Together, these studies define a functional role for IRF7 in macrophages as a driver of dermal fibrosis in SSc and support therapeutic targeting of the type I interferon pathway as a strategy to attenuate macrophage-mediated inflammation and fibroblast activation.