United States

Background
Breast cancer (BC) is the most common cancer affecting women and is associated with multiple genetic and environmental factors. To better understand individual’s genetic risk for disease, polygenic risk scores (PRSs) were developed using genome-wide association studies (GWASs). BC PRS have been developed with the hopes of improving risk stratification, but their replication is often limited. We worked with a population of genetically similar Dutch and American women to investigate the use of BC PRSs.

Methods
Two published PRSs utilizing 313 and 3280 single nucleotide polymorphisms (SNPs) were used to score 468 BC cases and 4,337 cancer negative controls from the integrated Cancer Repository for Cancer Research and cancer negative controls from the Netherland Twin Register. Blood samples were used to extract DNA and subjects were genotyped using the Illumina Global Screening Array and underwent standardized quality control, and 20 averaged runs of imputation, and polygenic scoring. Percentile scores were then compared via ANOVA testing, variability assessment, and area under the curve analysis.

Results
Results showed that the PRSs of BC cases were significantly greater than the controls at 52.81 and 48.07 (p&lt;0.001), respectively, for the 313 PRS and 55.45 and 49.81 (p&lt;0.001) for the 3820 PRS, respectively. The AUC was 0.596 (95% CI: 0.569-.624) for the 313 PRS and 0.603 (95% CI: 0.576-0.630) for the 3820 PRS, with both slightly underperforming the originally published AUC of 0.639 and 0.646 for the 313 and 3820 PRS, respectively. Variability of PRSs across imputation was 2.17% for the 313 PRS and 2.99% for the 3820 PRS, with a subset with increased score ranges.

Conclusion
Our study provides further evidence that BC PRSs are reproducible in our study populations with slightly less predictive capacity. In addition, we found that scores varied 2-3% between various imputation runs with some increased variability in subjects carrying many rare alleles, indicating that imputation methodology can have an impact an individual’s reported PRS. This may affect the clinical implementation of PRSs and may be worth considering as this new tool enters clinical use.

AMA Research Challenge 2023

Exploration of breast cancer polygenic risk scores in a population of breast cancer patients

Background
Breast cancer (BC) is the most common cancer affecting women and is associated with multiple genetic and environmental factors. To better understand individual’s genetic risk for disease, polygenic risk scores (PRSs) were developed using genome-wide association studies (GWASs). BC PRS have been developed with the hopes of improving risk stratification, but their replication is often limited. We worked with a population of genetically similar Dutch and American women to investigate the use of BC PRSs.

Methods
Two published PRSs utilizing 313 and 3280 single nucleotide polymorphisms (SNPs) were used to score 468 BC cases and 4,337 cancer negative controls from the integrated Cancer Repository for Cancer Research and cancer negative controls from the Netherland Twin Register. Blood samples were used to extract DNA and subjects were genotyped using the Illumina Global Screening Array and underwent standardized quality control, and 20 averaged runs of imputation, and polygenic scoring. Percentile scores were then compared via ANOVA testing, variability assessment, and area under the curve analysis.

Results
Results showed that the PRSs of BC cases were significantly greater than the controls at 52.81 and 48.07 (p<0.001), respectively, for the 313 PRS and 55.45 and 49.81 (p<0.001) for the 3820 PRS, respectively. The AUC was 0.596 (95% CI: 0.569-.624) for the 313 PRS and 0.603 (95% CI: 0.576-0.630) for the 3820 PRS, with both slightly underperforming the originally published AUC of 0.639 and 0.646 for the 313 and 3820 PRS, respectively. Variability of PRSs across imputation was 2.17% for the 313 PRS and 2.99% for the 3820 PRS, with a subset with increased score ranges.

Conclusion
Our study provides further evidence that BC PRSs are reproducible in our study populations with slightly less predictive capacity. In addition, we found that scores varied 2-3% between various imputation runs with some increased variability in subjects carrying many rare alleles, indicating that imputation methodology can have an impact an individual’s reported PRS. This may affect the clinical implementation of PRSs and may be worth considering as this new tool enters clinical use.

poster

Welcome to the **largest national, multi-specialty research** event featuring research from nearly 700 medical students, residents, fellows, and international medical graduates.

Use the buttons below or navigation menu to access the event.
 


[![](https://assets.underline.io/uploads/markdown_image/1/image/3e33a8c754cc49610b30950617d813e7.png)](https://underline.io/events/434/sessions?searchGroup=all)
 
[![](https://assets.underline.io/uploads/markdown_image/1/image/551404d429c4e274f9a54cbdd7081918.png)](https://underline.io/events/434/posters?searchGroup=all)
 
[![](https://assets.underline.io/uploads/markdown_image/1/image/677d92f566b880cebd36905fc738fc10.png)](https://underline.io/events/434/lounge)
 
[![](https://assets.underline.io/uploads/markdown_image/1/image/c7392833c7261c3874c5b43fa2f4df19.png)](https://underline.io/events/434/help-desk)
 

#### Event Details

The 2023 American Medical Association Research Challenge includes a poster symposium hall and a semifinals hall with the top scored research. **Log in at any time throughout the event to:**

* **View posters and presentations** on a variety of topics and specialties
* **Score posters in the semifinals** to help decide the top five finalists (AMA member exclusive opportunity)
* **Connect with participants** and provide feedback to help advance their research
* **Attend education sessions** with physician experts on key research topics

At the conclusion of this event, the AMA will announce the top five finalists who will advance to the AMA Research Challenge finals, present their research to an elite panel of judges, and compete to win a $10,000 grand prize sponsored by Laurel Road. 

### Grand Prize Sponsor

AMA would like to thank Laurel Road for sponsoring a $10,000 grand prize for the winner of the AMA Research Challenge. Laurel Road is not involved in the judging or scoring process to select the winner. The winner will be announced on February 4, 2024 during the AMA Research Challenge finals. Payment of the grand prize will be directly from Laurel Road. See the AMA Research Challenge Official Rules for terms and conditions.

![](https://assets.underline.io/uploads/markdown_image/1/image/611e331d4a0eba0920f240cfc04b4d77.png)

**There is no registration fee for this event! **

Register to explore research in a variety of topics. AMA members get exclusive access to score posters in the semifinals.

**Please use this link [here](https://www.ama-assn.org/about/events/ama-research-challenge) to register for free.**

Registration Is Required

Join the AMA Research Challenge poster symposium and semifinals to explore top research posters, score posters in the semifinals (an AMA member exclusive opportunity) and attend education sessions with experienced physicians to learn tips to advance your research and career. 

Background
Pancreatic ductal adenocarcinoma (PDAC) is a lethal malignancy, largely due to its high rate of early metastasis. PDAC remains almost uniformly refractory to T cell-based immunotherapies that have transformed treatment in many other cancers, and there is a critical need to identify new therapeutic approaches. The abundance and functions of eosinophils in PDAC have not been explored despite increasing evidence for important roles of eosinophils in other tumor types. Here, we quantified tumor eosinophilia in preclinical PDAC models and a large PDAC patient cohort, and we assessed whether modulation of eosinophils impacts primary tumor growth or metastasis. 

Methods
We used flow cytometry and immunohistochemistry to quantify eosinophil abundance in PDAC patient samples and in subcutaneous, orthotopic, and genetically engineered mouse models (GEMMs) of PDAC. We used single cell RNA-sequencing (scRNA-seq) and cytokine arrays to study the mechanism by which eosinophils are recruited into pancreatic tumors. Using an eosinophil-deficient mouse model (dblGATA), genetically engineered cytokine-overexpressing cell lines, and systemic injection of exogenous interleukin-33 (IL-33), we studied the effects of eosinophil depletion, expansion, and activation on primary tumor growth and metastasis in murine PDAC. We used scRNA-seq to study mechanisms by which intratumoral eosinophils may impact PDAC metastasis. 

Results
Eosinophils were abundant in PDAC samples from human patients and murine models. They are likely recruited into tumors via local production of eotaxin-1 and eotaxin-2 by distinct populations of cancer associated fibroblasts and macrophages. Eosinophil expansion or depletion does not impact the growth of primary pancreatic tumors, but mice lacking eosinophils are more likely to develop spontaneous lung metastases. Treating mice with recombinant IL-33 expands eosinophils systemically and reduces rates of metastasis. Wild-type and eosinophil-deficient mice have similar metastasis burdens after intravenous injection of tumor cells, suggesting that eosinophils do not limit metastasis by impacting the pre-metastatic niche. Rather, eosinophils may interact with other intratumoral cell types to limit the entry of cancer cells into circulation. Indeed, scRNA-seq reveals that eosinophil deficiency is associated with a significant shift in the proportions of intratumoral endothelial cell (EC) subpopulations from tip-like to stalk-like ECs. 

Conclusion
Eosinophils are recruited by locally produced cytokines into the PDAC TME, where they shape the endothelial compartment and regulate the metastatic spread of tumor cells. Further studies are warranted to better delineate the mechanisms underlying this phenotype and to determine whether tumor eosinophilia can serve as a useful therapeutic target or prognostic biomarker in PDAC patients.

Eosinophils regulate metastasis in pancreatic cancer

In order to function correctly, cells must regulate gene expression; chemical modification of DNA is an important method by which they achieve this. For example, cytosine can be methylated to form 5-methylcytosine (5mC), a repressive epigenetic marker. 5mC can be oxidized by Ten-Eleven-Translocation (TET) enzymes to produce additional forms of cytosine. One of these derivatives, 5-hydroxymethylcytosine (5hmC), is an activator of gene expression and ͏can be further oxidized to 5-formylcytosine (5fC) and 5-carboxylcytosine (5caC). Subsequently, 5caC is excised and replaced with unmodified cytosine. This process by which 5mC is converted back to cytosine is known as active demethylation (AD).
Improper control of methylation and demethylation can impair development and is a hallmark of many cancers. Ubiquitin-like Modifier with PHD and Ring Finger Domain 2 (UHRF2) is a DNA-interacting protein that plays an important but poorly characterized role in active demethylation. Previous work suggests that UHRF2 is required for the maintenance of 5hmC levels in somatic cells but does not demonstrate a mechanism. Furthermore, a paralog of this protein, UHRF1, binds to 5mC via its Set and Ring Associated (SRA) domain to preserve methylation levels in DNA. Thus, we proposed a related model where UHRF2 binds to 5hmC via its SRA domain and prevents further oxidation by TET enzymes. 
In this study, we overexpressed UHRF2 and TET2 in wild-type U2OS (human osteosarcoma) cells and U2OS cells with endogenous UHRF2 expression knocked out. We also conducted domain-deletion experiments to determine which domains of UHRF2 are required to preserve 5hmC levels.
As expected, we demonstrated that excess TET2 increased 5hmC levels in both cell lines, but concurrent overexpression of UHRF2 appeared to amplify this effect. We also determined that the plant homeodomain (PHD), tandem Tudor domain (TTD), and SRA domain are each necessary for UHRF2 to preserve 5hmC levels. Furthermore, we found that simultaneous overexpression of TET2 and UHRF2 increased levels of 5fC more than overexpressing TET2 alone, suggesting that UHRF2 may have a more complex role in regulating the AD cycle. Eventually, a better understanding of UHRF2 function will elucidate its role in cancer and may guide future research and therapeutics.

The epigenetic modifier UHRF2 has a complex role in 5hmC maintenance

Background
Neurofibromatosis type 2 (NF2) is cancer predisposition syndrome characterized by the development of bilateral vestibular (VS) and spinal schwannomas secondary to loss of heterozygosity of NF2 in Schwann cells or their precursors. Largely benign, tumors may result in deafness, vertigo, facial muscle weakness, chronic neuropathic pain, and death. There is an urgent need to develop pharmaceutical approaches for NF2 patients. Receptor tyrosine kinase inhibitors, brigatinib and dasatinib, were identified as potentially efficacious agents for the treatment of VS and shown to target the Ephrin A2 receptor (EPHA2). EPHA2 is a transmembrane receptor tyrosine kinase involved in cell contact-mediated motility, adhesion, and migration. 

Methods
We analyzed RNA Sequencing data and found EPHA2 expression to be increased in NF2-deficient Schwann cells. Data obtained from the TCGA Pancancer database demonstrated increased EPHA2 expression in NF2-/- cancers. We hypothesized that EPHA2 could be a potential target for the treatment of NF2-associated VS. Immunohistochemistry confirmed increased EphA2 expression in NF2-deficient murine vestibular schwannoma. 

ProteomicsDB was used to identify drugs predicted to target EPHA2 based on kinome analysis. We identified ponatinib, an ABL1 inhibitor, as an agent that targets EPHA2. Dose response curves obtained by cell titer glo luminescence assay and colony forming assays performed in human NF2-deficient schwann cells in vitro demonstrated impaired viability following treatment with ponatinib. Kinome profiling identified EPHA2 as the most significantly inhibited kinase following 24hour treatment with 100nM ponatinib. Western blot analysis was performed and EPHA2 protein expression was significantly decreased following ponatinib treatment.

We performed selective pharmacologic and genetic inhibition of EPHA2 with ALW-II-41-27 and siRNA in human NF2-deficient schwann cells in vitro. Dose response curves were obtained by cell titer glo luminescence assay and colony forming assays performed in human NF2-deficient schwann cells in vitro. 


Results

We demonstrate that EPHA2 expression is increased in NF2-/- Schwann cells and NF2-/- cancers. We identify ponatinib, a RTK inhibitor targeting ABL1 for CML, as an agent that targets EPHA2. We demonstrate that ponatinib treatment impairs the viability of both human and murine NF2-/- Schwann cells in vitro and decreases EPHA2 protein expression. Pharmacologic and siRNA-mediated inhibition of EPHA2 also impaired the growth of human NF2-/- Schwann cells in vitro. Lastly, we demonstrate that both ponatinib and EPHA2 inhibition induce morphological changes in NF2-/- Schwann cells. 

Conclusion
Our findings suggest that ponatinib or the direct targeting of EPHA2 may be efficacious for the treatment of NF2-associated vestibular schwannoma. Future in vivo efficacy studies are warranted. 


EPHA2 is a Potential Target for the Treatment of NF2-/- Vestibular Schwannoma
Kéyana Foster1, Dana K. Mitchell1, Alyssa Flint1, Brooke Rodriguez1, Henry Mang1, Chris Davis1, Steven P. Angus1, D. Wade Clapp1, Charles Yates2
1Department of Pediatrics, Herman B Wells Center for Pediatric Research, Indiana University School of Medicine, Indianapolis, IN, USA
2Department of Otolaryngology, Indiana University School of Medicine, Indianapolis, IN, USA


Single cell RNA sequencing (scRNA-seq) can detect gene expression and its dynamics in groups of cells. Its application to time-series samples can thus identify temporal gene expression programs active in different cell types. However, current scRNA-seq analysis needs improvement with regards to low gene detection, low replicate count, and inflated statistics when using cells as individual sampling data points. To address these issues, we explore whether “metacells” can be used as “replicates” through time. Toward this, we applied the use of metacells to a time-series scRNA-seq dataset from PBMCs after SARS-Cov-2 infection, which were then used in maSigPro for quadratic regression to find significantly differentially expressed genes (DEGs) over time, followed by clustering analysis of the expression velocity trends. Results demonstrated that this strategy could overcome the limitation of 1-2 replicates, as it correctly identified the known ISG15 interferon response program in almost all PBMC cell types. Its application further led to the uncovering of additional gene expression programs that potentially modulate different levels of host response after infection.

Identifying cell type transcriptome changes in COVID-19 patient PBMCs using metacells

Background 
Exosomes are extracellular vesicle (EV) nanocarriers responsible for delivering biologics including lipids, proteins, and nucleic acids. EVs are an attractive therapeutic weapon for delivering drugs and biologics to cells and tissues. Utilizing EVs for clinical applications is based on their low toxicity, ability to target to specific cells and tissues, and can be bioengineered to enhance their activities. Significant challenges remain before exosomes can be widely used in clinical settings. These include developing improved methods for purifying exosomes, enhancing the targeting of exosomes to diseased tissue, and boosting the efficiency of engineering exosomes for disease-specific applications. 
Exosomes need to be engineered in a more efficient way for broad therapeutic application. The lack of efficient intracellular bioengineering technologies to modify produced exosomes is a major obstacle to the widespread use of EVs for therapeutic applications. 
Advances in mammalian synthetic chromosome design and engineering offer a potential improvement in EV manufacturing. Our aim is to adopt the Artificial Chromosome Expression System (ACE System), an autonomous chromosome-based “circuit-board” designed to carry single or multiple copies of genes or DNA elements of interest for EV engineering.

Methods 
We have demonstrated that human miRNAs expressed from the ACE platform residing in a CHO-based cell line failed to process human miRNAs into CHO exosomes. This suggests that the processing of miRNAs from primary gene transcripts may be regulated in a species-specific manner. Thus, the ACE chromosome has been transferred into human HEK293 cells. Engineered miRNAs have been cloned downstream of a modified green fluorescent protein (AcGFP1) expressed from a doxycycline inducible promoter system retrofitted to contain ACE platform recombination system. 

Results 
Our results indicated that the ACE synthetic chromosome platform can be successfully engineered to express miRNAs as part of an initial doxycycline inducible AcGFP1 transcript. Furthermore, isolation of resulting EVs from the ACE engineered HEK293 cells indicated the presence of the expressed miRNAs. 

Conclusion 
From our work, we propose that the ACE synthetic platform has the potential to serve as a genetic circuit-board for modifying EVs of therapeutic value. In addition, the ACE synthetic chromosome is autonomous to the host cellular genome, i.e., non-integrating and non-mutagenic, thus would provide an increasing safety profile for using our developed synthetic chromosome technology in an ex vivo cellular therapy format. Future efforts will focus on enhancing the efficiency of our bioengineering approach for a specific disease indication.

The Engineering of Human Synthetic Chromosomes for Exosome-based Therapeutics

Background
Chromosome-level mutations play an essential role in the mechanisms behind speciation and disease. Analyzing massive genomic data to elucidate orthologous chromosome pairings between primate species is currently limited by computational constraints. Specifically, computational approaches have been limited by their inability to determine balanced chromosomal mutations rapidly, accurately, and cost-effectively. As the Alu family of transposable elements is widely conserved and prevalent (10%) among primate genomes, it is ideal for interspecies comparison and employed in our novel method. Our investigation employs unique, advanced data science techniques to classify orthologous chromosomes across primates, detect chromosomal inversions, and identify chromosomal translocations. We hypothesize that this unique approach will produce accurate results when compared to established literature, perform in less time relative to current methodologies, and be more cost-effective than its counterparts. 

Methods
We obtained and subsequently aligned Alu elements from 11 distinct primate species’ genomes on the University of California Santa Cruz’s Genome Table Browser. After computationally aggregating and statistically analyzing all primate Alu sequences, orthologous chromosome pairs were identified based on the highest frequency of homologous pairings across any two compared primate species. Using the orthologous chromosomes’ loci-specific pairing data, we detected statistically significant balanced inversions and translocations between species. 

Results
We successfully classified orthologous chromosomes between two primates with high z-scores (minimum z-score of 2.5 and maximum above 4), confirmed with previously published lists. Graphical representations of primate species, utilizing both subfamily-specific and all Alu elements, allowed for clear, statistically significant detection of chromosomal inversions, which were then validated with previously published genomic differences (10/10). Further loci-specific analysis identified every reported chromosomal translocation or change in orientation greater than 5 Mbp translocations (maximum p < 0.007 and regularly p < 1e–4). 

Conclusion
Our novel approach can be used to detect chromosomal abnormalities in primate genomes, making it a powerful tool for evolutionary and oncological research. Furthermore, it does so in a manner that is far less expensive, lengthy, and resource-intensive, while retaining the accuracy of chromosomal mutation detection. As primates serve as an important model organism for basic science research in various subspecialties of medicine, this approach could have a significant impact on numerous biomedical fields. Future investigations will explore specific data-size constraints in interspecies comparisons and the method’s applicability in detecting mutational patterns in diseases affecting humans and other primate species.

Primate Chromosomal Abnormalities Detection Through Selective Sampling Across Large-Scale Genomic Data

Background
Kidney transplantation is a lifesaving treatment for patients with end-stage renal failure. However, the success of kidney transplantation is hindered by chronic kidney graft rejection. On average, transplanted kidneys last 9 years due to chronic rejection, requiring patients to undergo dialysis or re-transplantation. While advances have reduced the rates of acute rejection, little is known about preventing chronic rejection. In mice and humans, chronic rejection is characterized by a persistent T cell infiltrate. The cytokine IL-15 is elevated in renal graft rejection. However, it is unknown whether IL-15 signaling supports graft-specific T cells that drive chronic rejection. I hypothesize that IL-15 signaling is required for the long-term survival of graft-specific T cells, and that disrupting IL-15 signaling can prevent chronic rejection.

Methods
We transplanted MHC-mismatched kidneys (Balb/c.OVA x B6) to recipient mice (B6) followed by injection of OVA-specific CD8 T cells (OT-I cells). OT-I cells allow for tracking immune response in kidney grafts. Recipient native kidneys were removed during transplantation. Kidney grafts underwent analysis via flow cytometry, immunofluorescence, and histology.

The Cre-loxP system was used for temporal genetic deletion of CD122 or CD215. Anti-CD122 blocking antibody was administered intravenously to transplant mice. Graft rejection was monitored by signs of uremia (lethargy, decreased mobility, ruffled hair), serum creatinine, or death.

Results
We found evidence that IL-15 signaling was taking place in kidney grafts, with cells expressing CD215, a chaperone protein for IL-15, and CD8 T cells expressing CD122, a receptor subunit for sensing IL-15. Using a combination of conditional knockouts of CD215 and CD122, we found that genetic deletion of either CD215 or CD122 significantly reduced graft-specific CD8 T cells. Given the importance of IL-15 signaling, we treated transplant mice with anti-CD122 antibody, which significantly reduced graft-specific CD8 T cells, decreased graft infiltration and fibrosis, and prevented graft rejection.

Conclusion
IL-15 is a key component in the survival and function of graft-specific T cells and, ultimately, graft failure. Disrupting IL-15 signaling with anti-CD122 blocking antibody could serve as a potential approach to the prevention of chronic rejection.

Disrupting the Survival of Pathogenic T cells to Ensure the Survival of Kidney Transplants

Introduction: Metabolic Syndrome (MetS) is a constellation of risk factors including glucose intolerance, central obesity, hypertension, and dyslipidemia. Low-grade inflammation, possibly originating in inflamed fat, has been implicated in the genesis and pathophysiological consequences of MetS. MetS occasionally co-exists with renal artery stenosis (RAS) and aggravates outcomes partly through the release of pro-inflammatory cytokines. Mesenchymal stem/stromal cells (MSCs) are multipotent cells with robust regenerative and immunomodulatory potential. This study tested the hypothesis that MSC delivery would alternate fat inflammation in MetS+RAS.
Methods: Twenty-four pigs were randomly divided into four groups: Lean, MetS, MetS+RAS, and Mets+RAS+MSC (n=6 each). In MetS+RAS+MSC pigs autologous adipose tissue-derived MSCs (107) were injected into the renal artery 6 weeks after RAS induction. Four weeks later, renal function was studied in-vivo using CT. After euthanasia, adipose tissue samples were stained to assess inflammatory markers, regulatory T-cells, resident stem cells, T-bet+ B-cells, adipocyte size, and fat fibrosis. Pro-inflammatory monocyte chemoattractant protein-1 (MCP-1), interleukin-6, and tumor necrosis factor-alpha immunoreactivity and levels were measured in fat and systemic blood, respectively.
Results: Injection of MSCs into MetS+RAS pigs did not affect renal function but slightly decreased blood pressure and circulating MCP-1 levels in systemic blood. Furthermore, in subcutaneous adipose tissue MSCs restored T-regulatory cells and endogenous stem cells, decreased T-bet+ B-cells, and downregulated inflammatory marker immunoreactivity. MSCs also restored adipocyte size and reduced fat fibrosis. 
Conclusion: Adipose tissue-derived MSCs attenuate fat inflammation in pigs with MetS+RAS. Hence, MSCs may represent a promising therapeutic strategy in metabolic disorders.

Mesenchymal Stem Cells Reverse Inflammatory Changes in the Adipose Tissue of Swine with Metabolic Syndrome and Renovascular Hypertension

Background
Prior information is routinely integrated in neocortex during decision-making. The orbitofrontal and posterior parietal cortex (OFC, PPC) directly represent and use stimulus history to efficiently guide ongoing behavior. To integrate prior stimulus information selectively and direct behavior flexibly, a mechanism for routing information is necessary. Prior evidence suggests critical structures in the mesial temporal lobe (MTL) route information during navigation. The neural dynamics underlying the behavioral effects of prior experiences remain poorly understood. Through studying the evolution of electrophysiologic signatures in human intracranial recordings during learning, we investigate how experiences are routed to modify processing in neocortex and direct behavior.

Methods
We trained eleven treatment-refractory epilepsy participants with a reward-based decision-making task with reversal learning in two contexts. We used a Bayesian framework to infer expectation for reward, a computation that depends on prior choices and reward outcomes. We simultaneously recorded neural ensemble activity from high frequency (HF, 30-120 Hz) neural power in intracranial electrodes implanted across the brain. To measure the representation of history-dependent computations, we performed a trial-wise regression of HF neural power on expectation. We monitored the effect of experience on neocortical processing by measuring how variations in representations changed with variations in expectation using representational similarity and multivariate pattern analyses. We tracked how much representations are predictive of choice behavior throughout learning using a linear decoder to measure the influence of prior experiences on behavior.

Results
Representations of expectation in MTL were conditioned on the history of rewards and were greater when current reward outcomes were consistent with prior compared to inconsistent. This feature was not observed in neocortical regions. These history-dependent computations in MTL reliably explained increases in the separation of HF neural power in the OFC. Representations of expectation across several neocortical regions were predictive of choice behavior significantly above chance. Choice prediction accuracy increased with learning using representations in OFC and not other neocortical regions.

Conclusion
Our findings suggest that prior reward experiences are represented in MTL structures, which then direct the formation of neocortical representations. OFC representations during learning are associated with MTL dynamics and are behaviorally relevant, suggesting that the OFC is a critical locus for mediating effects of prior experience on behavior. These findings add to accumulating evidence that flexible behavior depends on interactions between MTL and neocortex. Our study offers electrophysiological signatures for history-dependent computations to supervise the generation of representations that can redirect behavior.


Prior experiences fine-tune representations in human neocortex to direct behavior

Background
Diffuse midline glioma (DMG) is an aggressive, almost universally lethal form of high-grade glioma that occurs in children. One reason that DMG are so difficult to treat is that epigenetically unstable tumor cells evolve to resist radiation, chemotherapy, and other forms of treatment. Treating DMG cells with HDAC inhibitors (HDACi) reduces cell viability in mouse models, yet HDACi do not improve clinical outcomes when provided as monotherapies. The overall goal of this research is to identify combination therapies to synergistically reduce the growth and survival of DMG cells. A CRISPR-based genetic screen in DMG cell lines revealed that knockout of polo-like kinase 1 (PLK1), a key regulator of cell cycle progression, may sensitize DMG cells to HDACi. The purpose of this study is to develop PLK1i+HDACi combination therapies for DMG, which have few treatment options.

Methods
We conducted dose curve analyses to analyze changes in DMG cell growth and survival in response to increasing doses of HDACi combined with PLK1 inhibitors using the CellTiterGlo assay to measure cell viability. We also conducted immunofluorescence (IF) imaging of DMG cell lines treated with different HDACi and PLK1i to monitor cell cycle (phos-H3-Ser10) and DNA damage (yH2AX) markers.

Results
We found that co-treatment with PLK1i sensitizes DMG cells to HDAC inhibitors, as shown in the dose curve analyses where the addition of PLK1i (GSK461364, BI2536) lowered the IC50 values for two different HDACi (entinostat and panobinostat) compared to HDACi alone. Analysis of the IF images showed higher %(+) signals for both yH2AX and phos-H3-Ser10 in PLK1i and HDACi treated cells compared to either vehicle-treated controls, or DMG cells treated with HDACi and PLK1i alone. 

Conclusion
We conclude that PLK1i + HDACi drug combinations synergistically reduce DMG viability. We also observe that PLK1i + HDACi combinations affect DMG cell phenotypes via increased DNA damage and induced mitotic arrest. The importance of this work is that it potentially opens more therapeutic approaches for patients with DMG who would otherwise have a poor prognosis and limited treatment options. Future steps for this study are to determine the optimal PLK inhibitors to best sensitize DMG to HDAC inhibitors and to detect changes in signal transduction pathways and gene transcription downstream of PLK1i+HDACi combination therapy by phospho-proteomics and RNA sequencing.

Next from AMA Research Challenge 2023

Eosinophils regulate metastasis in pancreatic cancer