| Poster | Name | Lab | Title | Abstract |
|---|---|---|---|---|
| 2 | Lily Miller | Denu Lab | Comparative Enzymatic Analysis of Histone H3 Lysine 9 Methyltransferases | Histone lysine methylation is an epigenetic post-translational modification that regulates chromatin function and is impacted by metabolite availability. Histone methyltransferases rely on the metabolite S-adenosylmethionine (SAM) as the sole methyl donor, resulting in altered in vivo activity during SAM depletion. Surprisingly, histone mono-methylation, but not di- or tri-methylation, is maintained at specific heterochromatic loci, such as histone H3 lysine 9 (H3K9), despite extremely low SAM availability. Here, we investigated whether the intrinsic kinetic properties of histone H3 lysine 9 (H3K9) methyltransferases (HMTs) support a preference for mono-methylation previously observed in vivo. To determine this, we analyzed the steady-state kinetics and substrate binding parameters of the nuclear H3K9 HMTs (i.e., Ehmt1/2 and Suv39H1/2) in vitro using varying methylated forms of H3 substrates (i.e., unmodified, mono-, and di-methylated H3K9 peptides and nucleosomes). With SAM concentration varied over a broad range, HMTs displayed higher catalytic efficiency (kcat/KM, SAM) for mono-methylation compared to di- or tri-methylation. Under limiting H3 substrate conditions, the catalytic efficiency (kcat/KM, H3) for mono-methylation was 1-2 orders of magnitude higher than tri-methylation. Further, under saturating SAM and H3 substrate conditions, Ehmt1/2 and Suv39H1 displayed significantly faster turnover rates (kcat) for the mono-methylation reaction by 1.4- to 6.4-fold over di-methylation, and 7- to 19-fold over tri-methylation. In contrast, Suv39H2 displayed a similar kcat value for all three substrates. Interestingly, binding studies indicated all enzymes exhibited minimal differences in affinity for H3K9 methylation states, suggesting HMT substrate preference is not dependent on binding but occurs during catalysis for the enzymes besides Suv39H2. This detailed kinetic analysis directly compares multiple nuclear H3K9 HMTs across all substrate methylation states under conditions that mimic physiological fluctuations in SAM availability. The data provides a plausible mechanism in which the intrinsic rate of HMT catalysis drives the observed maintenance of H3K9me1 during SAM depletion. |
| 4 | Dekun Zhou | Del Pia Group | An SDP Relaxation for the Sparse Integer Least Square Problem | In this paper, we study the sparse integer least square problem (SILS), which is the NP-hard variant of the least square problem, where we only consider sparse {0, ±1}-vectors. We propose an l1-based SDP relaxation to SILS, and provide sufficient conditions for our SDP relaxation to solve SILS. The class of data input which guarantee that SDP solves SILS is broad enough to cover many cases in real-world applications, such as privacy preserving identification, and multiuser detection. To show this, we specialize our sufficient conditions to two special cases of SILS with relevant applications: the feature extraction problem and the integer sparse recovery problem. We show that our SDP relaxation can solve the feature extraction problem with sub-Gaussian data, under some weak conditions on the second moment of the covariance matrix. Next, we show that our SDP relaxation can solve the integer sparse recovery problem under some conditions that can be satisfied both in high and low coherence settings. We also show that in the high coherence setting, our SDP relaxation performs better than other l1-based methods, such as Lasso and Dantzig Selector. |
| 6 | Junha Shin | Roy Lab | Single-cell transcriptome analysis based on Tree-guided non-negative Matrix Factorization for studying genome editing effects on different culture platforms of human retinal organoids | Human stem-cell-derived retinal organoids offer powerful platforms for studying molecular programs associated with normal and disrupted eye function as well as for rapid assessment of potential targets of genome editing. The efficiency of genome editing in traditional 3D organoid systems is limited because of the low rate of transduction with non-AAV based delivery platforms. We observe that this transduction efficiency could be enhanced by removing the extracellular barriers for viral vector delivery via organoid dissociation and replating the cells into a 2D culture for transduction. However, it is unclear whether 2D platforms faithfully capture the cell type composition and gene expression programs of 3D organoids. To compare the transcriptome states of cells in 2D dissociated cultures and 3D organoids, we performed single cell RNA sequencing (scRNA-seq) of the two different platforms by taking samples from six different time points corresponding to different stages of retinal cell fate specification. To enable integrative analysis of this complex multi-sample dataset, we developed and applied a novel multi-task matrix factorization method named Tree-guided non-negative Matrix Factorization (TMF), which performs sample-specific dimensionality reduction while incorporating the relatedness of the samples. We used the TMF results to define cell clusters that were enriched in different retinal cell types. Both 2D and 3D platforms show the presence of a diverse set of cell types including Mller glia, photoreceptors, and horizontal cells. We identified a total of 22.4% of cells from 2D cultures as photoreceptors compared to 51.2% of cells from the 3D organoid cultures, while 2D cultures had a higher proportion of RPE/progenitor cell types compared to 3D cultures. Photoreceptors identified in either platform show expression of known cone and rod marker genes, which validate the identity of these cell types. The photoreceptor cells maintain functionality, as they maintain visual cycle function after delivery of the genome editor. Overall, this work shows that 2D platforms retain a sizable photoreceptor population, making them an effective platform for examining molecular signatures of genome-editing in a higher throughput manner. |
| 8 | Mitchell Josvai | Crone Lab | In Vitro iPSC-derived Cardiomyocyte Platform for Mechanical Analysis and Disease Modeling | Cardiovascular disease continues to be a leading cause of death worldwide motivating the need for models of cardiac function for basic science and clinically translational research. Cardiomyocytes (CMs) derived from human induced pluripotent stem cells provide a source for developing in vitro cardiac models, however current in vitro culture strategies and analysis techniques provide only a portion of the necessary means to fully characterize functionality. Specifically, there is a need for additional techniques to quantify and understand the contractile behavior of CMs as they interact collectively with each other and their surroundings. In this research, Digital Image Correlation (DIC) was used to quantify the mechanical function of iPSC-CMs on a tailorable, biologically relevant platform. The functionality of this platform as a disease model was demonstrated, first as a model of hypertrophic cardiomyopathy (HCM) using CRISPR/Cas9 modified iPSCs deficient in the cMyBP-C. The model recapitulated the increased contractile function of CMs with a homozygous knockout prior to hypertrophic remodeling. A catecholaminergic polymorphic ventricular tachycardia (CPVT) disease model was them generated using iPSC-CMs from a CPVT patient (RyR2-H2464D mutation) and a healthy familial control. The maximum contractile strain was found to be consistently higher in iPSC-CMs derived from the patient compared to the familial control, and the patient cell line had a statistically significantly slower intrinsic contraction rate than the control. These combined results highlight the utility of the platform as an in vitro cardiovascular model and will allow for further understanding of the interplay of genetics, environment, and genotype-phenotype relationships. |
| 10 | Brady Lundin | Ashton Lab | Neural Rosette Arrays as Neural Tube Defect Risk Prediction Screens | Neural tube defects (NTDs) are the second most common congenital malformation in the US and beyond. Given the multifactorial etiology and complicated genetic landscape underlying NTDs, current in vivo and in vitro models of neural development have limited utility for investigating clinical NTD pathophysiology and discovering preventative interventions. This is further confounded by species differences and lack of scalability with typical animal models. Human pluripotent stem cell (hPSC)-derived neural organoids partially address these shortcomings as they recapitulate incipient neural morphogenesis, i.e., neural tube formation, in vitro. However, their spontaneous emergence causes inconsistencies in tissue composition and cytoarchitecture, thus limiting their application. We hypothesized that this variability is a of product of non-standard and insufficiently controlled biophysical and biochemical cues. As such, we developed a microarray culture platform that enables spatiotemporal control of microenvironmental cues to reproducibly direct early neural organoid morphogenesis. The resulting human Rosette ArrayTM platform standardizes derivation of 3-D, singularly polarized neuroepithelial rosette tissues from various rostral/caudal CNS regions with high efficiency and reproducibility. Here, validation of the Rosette Array platform for environmental and genetic NTD risk assessment is presented. Using neural rosette emergence as a principal metric, various pharmaceuticals and agrochemicals known to cause NTDs through multiple modes of action were screened to evaluate the Rosette ArrayÕs ability to assess NTD risk; results displayed high sensitivity and specificity in line with similar in vitro assays. This includes a demonstrated ability to detect disruption of the developmentally critical folic acid metabolism pathway. Additionally, preliminary experiments using gene edited hPSCs containing clinically-relevant NTD mutations showed that the platform could detect correlated perturbations to neural rosette morphogenesis in a CNS region-specific manner. Thus, the Rosette Array platform is a scalable, Òoff-the-shelfÓ screen for quantitatively assessing clinical NTD risk and prophylactics. This technology has direct regulatory applications, and future work involving NTD patient-derived hPSCs evoke the possibility of patient-specific risk assessment and discovery of personalized prophylactics. |
| 12 | Abbey Manning | Tiny Earth | Searching for bioactive compounds of two morphologically different Pseudomonads | |
| 14 | Chris Endemann | Data Science Hub | The Data Science Hub: Empowering people to tackle data-rich research projects | The Data Science Hub in the Wisconsin Institute for Discovery provides data science training and implements data science into research practices across campus. The Data Science Hub executes a mission for community engagement and learning opportunities for campus researchers through a variety of services, including consultations with Data Science Facilitators who can recommend learning pathways and project strategies, community events and co-sponsored seminars that bring together researchers and other partners around relevant data science topics, and regular trainings (including workshops) around fundamental data science and computational skills |
| 16 | Shaoqin Gong | Gong Lab | Nanomedicines for Various Applications | |
| 18 | Jose Moran | Denu Lab | Kinetics and Regulation of Nucleosomal Deacetylation by the NAD+-dependent histone deacetylase, SIRT6 | The NAD+-dependent histone deacetylase sirtuin 6 (SIRT6) regulates various biological functions including DNA repair, telomere maintenance, and transcriptional repression of metabolic genes. As opposed to its activity in vivo, recombinant SIRT6 deacetylase activity in vitro is roughly 10,000 less efficient at deacetylating peptide substrates than other sirtuins. In the cell, SIRT6 deacetylates the multimeric Histone:DNA packaging complex known as the nucleosome core particle (NCP). Previous studies lack describing SIRT6 kinetics in the presence of NCPs and/or acetylated NCPs towards peptide substrates and what molecular features of the NCP or SIRT6 are necessary for this activity. Using timecourse deacetylation assays and subsequent kinetic analyses, we have showed that recombinant NCPs potently activate SIRT6 deacetylase activity towards peptide substrates and is a more catalytically efficient substrate (10,000 times more efficient than peptide alone and 60x than when the NCP stimulates SIRT6 activity toward peptide substrate). We have demonstrated that deacetylation of the NCP is dependent on SIRT6Õs interaction with the acidic binding patch, a chromatin-binding protein region on the NCP. Finally, utilizing a split-luciferase system, we have shown that SIRT6 dissociation from the NCP is linked to nucleosomal deacetylation, suggesting that SIRT6 may become a more dynamic enzyme in response to fluctuations in NAD+ levels. This work suggests SIRT6 is activated by and prefers to deacetylate NCPs, and its association along chromatin are likely regulated by local available pools of NAD+, suggesting that perturbations along various NAD+ producing pathways may also affect SIRT6 localization and deacetylase activity along chromatin. |
| 20 | Tolulope Perrin-Stowe | Yin Lab | Emergence and evolution of defective viral genomes in natural hosts | Defective viral genomes (DVGs) are incomplete genomes that emerge as a byproduct of virus infection. DVGs only replicate in the presence of full-length genomes by exploiting the resources of the standard virus (STV). DVGs interfere with the replication of STVs, co-transmit, and over multiple generations may enter a co-evolutionary arms race; as DVGs emerge and increase, they inhibit STVs which selects for STV variants that can replicate uninhibited until the process repeats. DVGs can also elicit immune responses that reduce viral load and thus play a role in the severity of infections (e.g. some DVGs can reduce SARS-CoV-2 viral load). Vesicular stomatitis virus (VSV) spreads from insect vectors to mammalian species, including humans, similar to other zoonotic viruses of concern (e.g., dengue & zika). An unknown aspect of viral infections is how the evolutionary interactions of DVGS & STVs and the innate immune pathways triggered by DVGs in initial vectors will impact the severity and spread of zoonotic viruses to humans. We aim to: (i) elucidate how expression of genes associated with innate immune responses impacts emergence and enrichment of DVGs during STV culture, and (ii) characterize co-evolution between DVGs and STVs over multiple generations. Cell lines from arthopod vectors of VSV will be infected with VSV and serially passaged for 10 generations. We expect to observe, via RNA sequencing of each passage, DVG emergence, changing fitness levels of DVGs and oscillations of STVs and DVGs as they co-evolve. We also anticipate upregulation of innate immunity genes as DVGs emerge, enrich, and adapt. Our findings will be a first step toward a deeper understanding of how co-evolution between STVs and DVGs and innate immunity gene regulation in a vector will affect zoonotic viral infections and how that may impact spillover to humans. |
| 22 | Bryce Sprecher | Ponto Lab | TBD | |
| 24 | Chung Yin Lin | Turng Lab | Bridging engineering and life sciences for manufacturing of cell --/tissue based therapeutic products | |
| 26 | Subhojyoti Mukherjee | Nowak Group | ReVar: Strengthening Policy Evaluation via Reduced Variance Sampling | In off-policy RL we evaluate a target policy on a pre-collected dataset before deploying it in real-world. This dataset is collected using a different behavior policy. Most work in policy evaluation for RL assumes a fixed behavior policy that generates data to evaluate a particular target policy. In our work, we study the optimal choice for the behavior policy and introduce a method that can learn the optimal behavior policy during data collection. |
| 28 | Daniel Pimentel-Alarcon | Pimentel-Alarcon Group | Subspaces & The Double Banana | |
| 30 | Keerthana Shankar | Saha Lab | Targeted Transgene Insertion in Natural Killer Cells using CRISPR-Cas9 | Natural killer (NK) cells are innate cytotoxic lymphocytes capable of killing virally infected cells and malignant tumors. Unlike T cells, NK cells are HLA-agnostic and cause little to no Graft vs. Host Disease in allogenic transfusions, making them excellent candidates for off-the-shelf therapies. However, current techniques to insert the CAR gene into NK cells primarily employ viral vectors. Viral methods can give rise to complications such as insertional mutagenesis, which can lead to gene silencing or oncogene activation. Here we have developed a non-viral editing strategy in primary NK cells. With this method, the transgene is encoded in a linear double stranded DNA (dsDNA) template and produced by polymerase chain reaction (PCR). The template includes homology to the intended target for insertion that is defined by nuclease mediated double strand break formation by a Cas9-single guide RNA (gRNA) complex. We show efficient (>60%) CRISPR knock-out of the inhibitory NKG2A receptor encoded by the KLRC1 gene within primary NK cells. Optimization of electroporation timing showed that delivery of 3 ug of donor DNA on day 6 of ex vivo expansion results in gene knock-in rates of up to 6.5% for transgenes as large as 2.5 kb. Lastly, we report that the use of K562-mbIL15-41BBL feeder cells improves expansion of CRISPR edited cells by approximately twenty-fold. This data provides a proof-of-principle for on-target integration of a long transgene within primary NK cells via CRISPR-Cas9 genome editing, which could be further optimized to enable the non-viral manufacture of genetically-programmed NK immunotherapies. |
| 32 | Kirstan Gimse | Saha Lab | Transcriptomic profiling of the mouse hippocampus after intracerebral injection of Cas9 nanocapsule genome editors | Evaluating the editing efficiency and cell-type specificity of genome editors is a critical task for developing somatic cell genome editing strategies, especially for those that target the brain. Standard methods rely on deep sequencing at the on-target site combined with immunohistochemistry within treated animal model systems to enumerate the types of edited cells in select tissues after administering genome editors. Animal reporter systems that express fluorescent proteins after successful on-target genomic editing provide robust platforms to evaluate the number of edited cells but typically incorporate limited opportunities to co-register cell-type markers with the reporter protein. Thus, molecular characterization of an edited cell typically is limited to 2-4 cell-type markers that can be imaged simultaneously with the fluorescent reporter. Here, we have utilized single nuclei RNA sequencing (snRNA seq) to deeply characterize the transcriptome of edited cells within the hippocampus of Ai14-tdTomato reporter mice. We performed snRNA-seq on nuclei isolated from the hippocampus after intracranial injection of nanoparticle genome editors. Our research team has developed a biodegradable nanocage (NC) capable of delivering preassembled SpyCas9 protein-gRNA ribonucleoprotein complexes (RNPs). RNPs targeting the Ai14 loxP-STOP cassette were encapsulated into these NCs and delivered into the hippocampus via intracranial injection. Two weeks post-injection, nuclei were isolated from the hippocampi of treated Ai14 mice and single nuclei RNA sequencing was performed. After performing snRNA-seq, we observe the capture of ~500-15000 unique transcripts per nuclei and robust Ai14-tdTomato reporter expression in nuclei from neuronal, glial, and oligodendrocytic cells. Additionally, differential cell-type specific transcriptomic shifts were identified primarily in immune and cell signaling pathways between treatment with recombinant adeno-associated viral vectors (rAAV) or NC. Overall, transcriptional profiling provides a high-resolution and complementary method to examine the cellular outcomes from genome editing within animal reporter systems, which has high potential to inform the clinical development of genome editing therapeutics. |
| 34 | Andrew Hanus | Illuminating Discovery Hub | Illuminating Discovery | |
| 36 | Austin Hall | Handelsman Lab | Bacteria respond to interactions by adapting new ways to move |
Click any poster to see a larger version. You may need to scroll.
We Thank Our Sponsors
Gold Sponsor
Lightning Talk Sponsors
