Abstracts for 39th International Mammalian Genome Conference

1 , P-22

Genome-wide association study of opioid addiction-related behaviors in Heterogenous Stock rats

Montana Kay Lara1, Lieselot L. G. Carrette1, Thiago Missfeldt Sanches1, Oksana Polesskaya1, Alicia Avelar2, Angela Beeson3, Hassiba Beldjoud1, Brent Boomhower1, Molly Brennan1, Denghui Chen1, Riyan Cheng1, Lindsey China1, Apurva S. Chitre1, Dana Elizabeth Conlisk4, McKenzie Fannon1, Benjamin B. Johnson1, Elaine Keung1, Adam Kimbrough5, Jenni Kononoff4, Angelica Renee Martinez1, Lisa Maturin1, Khai-Minh Nguyen1, Alex Morgan1, Joseph Mosquera1, Dyar Othman1, Sonja L. Plasil1, Jarryd Ramborger1, Paul Schweitzer1, Sharona Sedighim1, Osborne Seshie3, Kokila Shankar1, Benjamin Sichel1, Sierra Simpson1, Lauren Cassandra Smith1, Elizabeth A. Sneddon1, Lani Tieu1, Nathan Velarde1, Selene Zahedi1,6, Leah C. Solberg Woods3, Marsida Kallupi1, Giordano de Guglielmo1, Abraham A. Palmer1,7*, Olivier George1*

1 Department of Psychiatry, University of California San Diego, La Jolla, CA, USA

2 Jamestown Community College, Olean, NY, USA

3 Department of Internal Medicine, Section on Molecular Medicine, Wake Forest University School of Medicine, Winston-Salem, NC, USA

4 The Scripps Research Institute, La Jolla, CA, USA

5 Purdue University, West Lafayette, IN, USA

6 Institut de Neurosciences de la Timone, Aix-Marseille Université, Marseille, 13005, France

7 Institute for Genomic Medicine, University of California San Diego, La Jolla, CA, USA

Substance use disorders are complex polygenic traits and highly heritable. Despite growing numbers of genome-wide association studies (GWAS) in humans, the genetic etiology of opioid addiction-related behaviors remains incomplete. To improve our ability to identify genetic loci associated with risk for distinct behavioral components of addiction, we use Heterogeneous Stock (HS) rats. The HS rat population has been outbred for over 100 generations and provides high genetic and phenotypic heterogeneity for genetic mapping studies. We phenotyped approximately ~1,700 HS rats for opioid self-administration with either oxycodone or heroin. HS rats exhibit wide variability in escalation of drug intake and total drug taking, as well as multiple measures related to drug self-administration. GWAS for these distinct opioid drug studies identified genetic loci associated with vulnerability or resilience to opioid addiction-like behaviors for oxycodone and heroin. We also extend our genetic analysis across both drugs, examining genetic correlations between the two drugs and integrating the relevant drug-taking traits for both oxycodone and heroin to examine shared genetic liability. This study represents the largest GWAS of opioid self-administration behaviors in HS rats that integrates genetic risk across opioid drugs, allowing us to dissect both drug-specific and shared genetic associations.

2, P-34

Reversal of epigenetic memory of an American diet exposure in C57BL/6J male mice through dietary intervention

Alexandra Naron1, 2, 3, Oscar Camacho Martinez5 , Anna Salvador4 , Andrew Feinberg6,7 , David Threadgill1,2,3

1Interdisciplinary Program in Genetics and Genomics, Texas A&M University, College Station, TX, USA

2Department of Cell Biology and Genetics, Texas A&M University, College Station, TX, USA 3Department of Nutrition, Texas A&M University, College Station, TX, USA

4Center for Gastrointestinal Biology and Disease, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA

5Center for Epigenetics, Johns Hopkins University School of Medicine, Baltimore, MD, USA

6Center for Epigenetics, Johns Hopkins University School of Medicine, Baltimore, MD, USA

7Department of Medicine, Biomedical of Engineering and Public Health, Johns Hopkins University, Baltimore, MD, USA

Metabolic disorders such as obesity and diabetes are rising in prevalence worldwide highlighting a need for more effective interventions. The American (Western) diet is high in carbohydrates and fats and supports obesity and metabolic dysregulation. Carbohydrate restriction through ketogenic dietary interventions has emerged as a promising treatment strategy for obesity and metabolic dysfunction. We previously investigated the efficacy of ketogenic dietary interventions in genetically distinct mice. Our group demonstrated that when C57BL/6J (B6) males are exposed to the American diet,their fat percentage is 1.77-fold higher than B6 males consuming a ketogenic diet, suggesting these mice are responsive to carbohydrate restriction in the form of a ketogenic diet. B6 mice exposed to an American diet for as short as two weeks no longer responded to a ketogenic diet intervention. To test whether B6 mice exposed to the American diet gained an epigenetic memory preventing their response to the ketogenic intervention, mice on the American diet for three months were provided a diet that is deficient of methyl donors for two weeks before going onto the ketogenic diet intervention. Body composition measurements were collected at each stage of the dietary treatment and up to 6 months of the ketogenic intervention. After the methyl intervention diet, mice on the methyl donor deficient diet had a significant loss of total fat mass with little impact on lean mass. One and six months post intervention, we observed that exposure to the methyl donor deficient diet resulted in mice gaining only 3.9 ± 2.2 gm and 12.48 ± 4.68 gm of fat mass, respectively, compared to mice who received a diet supplemented in methyl donors. Adipocytes from gonadal fat pads were isolated and sequenced from. Numerous differentially methylated regions (DMRs) induced by the American diet were reversed by the methyl donor deficient treatment. Based on these responses, we speculate that the exposure to a methyl donor deficient diet can reverse memory of the prior American diet exposure.

3, P-6

Background-dependent response to ERBB3 inhibition in colorectal cancer

Kaitlyn E. Carter1,2,3, David W. Threadgill1,2,3

1 Interdisciplinary Program in Genetics and Genomics, Texas A&M University, College Station, TX, USA

2 Department of Cell Biology and Genetics, Texas A&M University, College Station, TX, USA

3 Department of Nutrition, Texas A&M University, College Station, TX, USA

Colorectal cancer (CRC) is the second leading cause of cancer-related deaths in the United States. Advances in precision medicine have contributed significantly to early detection, diagnosis, and treatment, helping to reduce this substantial health burden. Increased understanding of molecular mechanisms which govern CRC progression has further supported this progress, as mortality rates have trended downward over the last five decades. However, death rates for individuals under fifty are increasing, highlighting the need for more effective targeted therapeutic approaches.

This preclinical research seeks to explore how genetic differences in the host influence tumor molecular profiles and treatment response, aiming to address variable outcomes observed in diverse patient populations in clinical trials. The ERBB receptor tyrosine kinase (RTK) family regulates numerous complex biological processes and has been linked to aberrant cell growth and survival, making these receptors key therapeutic targets and promising candidates for preclinical therapies. ERBB3 is a pseudo-kinase that lacks intrinsic kinase activity and depends on transactivation by other ERBB receptors. In homogenous preclinical mouse models, ERBB3 deletion in gastrointestinal epithelia significantly reduces polyp number in the intestine and colon. In contrast, clinical trials testing ERBB3 inhibitors have shown little to no efficacy and have been associated with poor patient outcomes.

Building on this understanding, we investigated whether patient heterogeneity contributes to the failure of translating preclinical findings into clinical success. Using heterogenous preclinical mouse models, we discovered that the effect of ERBB3 inhibition on tumor growth is genetic-context dependent. ERBB3 inhibition on the 129S1/SvlmJ (129) background reduces tumor growth, whereas on the C57BL/6J (B6) background, tumor growth is promoted- offering a potential explanation for translational failures. This study aims to leverage mouse models to identify genetic modifiers linked to loss of ERBB3, with the goal of improving outcomes in failed clinical trials.

We generated an F2 population from a 129 and B6 cross to identify regions associated with polyp count distribution through quantitative trait loci analysis (QTL). The distribution of the F2 population encompasses the range observed in both parental backgrounds. The F2 mice have been genotyped and we have identified significant regions of interest on Chromosomes 2 and 11 using R/qtl2. Candidate genes include Stk39, which has been shown to contribute to cancer progression through the MAPK/ERK pathway; Hmmr, recently identified as regulator of proliferation and CRC tumor progression through mTOR signaling (PI3K/AKT); and Gabrg2, which isinvolved in tumor growth through the MEK/ERK axis. Analysis leveraging human transcriptomic databases indicates that expression patterns of all candidate genes, when combined with low ERBB3 expression, affect patient prognosis. Further work will involve investigating the mechanisms underlying these modifiers to better understand their roles in CRC progression and identify potential therapeutic targets.

4, P-4

Longitudinal standardised behavioural phenotyping pipelines for knock-in mouse models of amyloid accumulation in Alzheimer’s disease

Sevda T. Boyanova1,2, G. Banks1,3,4, M. H. Correa2, T. Lipina1,5, R. S. Bains3, M. Stewart3, S. Wells3, and F. Wiseman1,2

s.boyanova@ucl.ac.uk

1. UK Dementia Research Institute at University College London, London, UK.

2. UCL Queen Square Institute of Neurology, University College London, London, UK.

3. The Mary Lyon Centre at MRC Harwell, Oxfordshire, UK.

4. School of Science and Technology, Nottingham Trent University, Nottingham, UK

5. University of Toronto, Department of Pharmacology & Toxicology, Toronto, Canada

Alzheimer’s disease (AD) is characterised by decline in cognitive functions and is often associated with behavioural and psychological symptoms (Sasaguri et al., 2017). AD is known to have genetic underpinnings, including autosomal dominant cases caused by mutations in the amyloid precursor protein (APP) gene. These mutations affect APP processing and result in an altered aggregation of the Aβ peptide (Dai et al., 2018). How the accumulation of Aβ leads to the cascade of neuropathological changes and the development of AD clinical features is not fully understood.

Robust phenotyping of animal models of AD is critical to improving our understanding of the molecular mechanisms that cause disease symptoms (Sasaguri et al., 2017). New generation of knock-in gene-edited mice express disease-associated protein at endogenous levels, with correct temporal and spatial expression, reducing model artefacts. These models also carry humanised Aβ region of the App gene promoting key aspects of AD biology (Saito et al., 2014, Serneels et al., 2020).

Here we use longitudinal standardised behavioural testing pipelines to characterise knock-in mouse models of Aβ accumulation in AD - Apptm2.1Tcs/Apptm2.1Tcs (AppNL-F/NLF) carrying the Swedish (KM670/671NL) and Iberian mutation (I716F), Apptm3.1Tcs/Apptm3.1Tcs (AppNL-G-F/NL-G-F), carrying the Swedish, Iberian and Arctic (E693G) mutation, and the control humanised line Appem1Bdes/Appem1Bdes (Apphu/hu) which carries only humanisation of the Aβ region, without any autosomal mutations involved in AD.

One behavioural pipeline included classical out of cage tests including Y-maze test of short-term memory, Crawley three-chamber test of social preference, optokinetic drum test of visual acuity, olfactory habituation-dishabituation test, as well as 24-h home-cage analysis of activity. The second pipeline measured cognitive flexibility and paired associate learning tasks using touchscreen-based assay. All protocols used in this study were done in accordance with the Animals (Scientific Procedures) Act 1986 (UK) at the Mary Lyon Centre at MRC Harwell.

I will present intermediate results from the pipelines, including short-term memory, social preference, olfaction, and visual acuity, home-cage activity, paired-associate learning and cognitive flexibility. These data can be used to select the most useful model and time-point to inform fundamental research into the cellular and molecular causes and consequences of Aβ pathology and for proof of principle testing of interventions.

5, P-1

Genetic regulation of diet-induced thermogenesis in mice

Emanuele Baldassarri1,2, Anna C. Salvador1,2, Alexandra Naron2,3, Ahmed Elsaadi2, Aaron Vanwettering2, Thomas Wong2, Ryan McGovern2, and David W. Threadgill1,2,3,4

1 Department of Nutrition, Texas A&M University, College Station, TX, 77843, USA

2 Department of Cell Biology and Genetics, Texas A&M University, College Station, TX, 77843, USA

3 Interdisciplinary Program in Genetics and Genomics, Texas A&M University, College Station, TX, 77843, USA

4 Department of Biochemistry & Biophysics, Texas A&M University, College Station, TX, 77843, USA

The prevalence of obesity and other metabolic diseases has increased, despite advancements in precision medicine in recent decades. There are many reasons why individuals who eat the same foods, consume the same calories, or exercise at the same level may experience different results. One key factor is thermogenesis, the production of heat through metabolic processes. Thermogenesis, and consequently body composition, is influenced by diet, exercise, genetics, or a combination of these factors. This aspect is not unique to humans and can also be observed in mice, which makes them a good model to study how genetically distinct individuals respond to the same diet.

We previously demonstrated that A/J, C57BL/6J (B6), NOD/ShiLtJ, and FVB/NJ mice have different metabolic responses when exposed to a ketogenic diet (high fat, no carbohydrate) or an American diet (high fat, high carbohydrate). While a significant increase in heat expenditure was observed across the 4 strains exposed to the ketogenic diet, the magnitude of this response varied between strains. A/J had the steepest increase in heat expenditure in response to the ketogenic diet relative to American diet, whereas B6 showed a more modest response. The differences in heat expenditure could not be attributed to rates of activity or food consumption. Interestingly, B6 mice had the sharpest decrease in body fat percentage in response to ketogenic diet relative to the American diet. In contrast, A/J mice had a more modest, but still significant, decrease in body fat percentage in response to ketogenic diet relative to the American diet.

To identify the genetic loci regulating thermogenesis during carbohydrate restriction, we generated F2 populations from B6 and A/J crosses and exposed them to a ketogenic diet for 3 months. Mice were phenotyped for body composition and metabolic rate, then genotyped using the Mouse Universal Genotyping Array (MUGA) with 7854 markers. Quantitative trait loci (QTL) mapping was performed to identify genomic regions driving heat expenditure, body fat gain, and activity. For heat expenditure, linkage analysis revealed QTLs on chromosomes 1 and 7 (Heatq1, Heatq2). Interestingly, the model for activity revealed nearly identical QTL on Chr1 (Actq1). Similarly, the most prominent QTL for percentage of body fat gain was observed on chromosomes 7 (Bfgq3), being in a nearly identical location to Heatq2. Two independent Chromosome Substitution Strains (CSS) lines were obtained to validate the effects of the AJ Chr1 (CSS1) and AJ Chr7 (CSS7) on the B6 background.

By better understanding the interpersonal differences in genetic background, which is responsible for variation in rates of thermogenesis, there could be an opportunity to develop precision nutrition approaches to better understand metabolic disorders and address the obesity epidemic.

A NOVEL WILD-DERIVED MHC-LINKED LOCUS REGULATES HOST IMMUNITY TO GAMMAHERPESVIRUS INFECTION

Courtney M. Waytashek1, Emily A. Nelson1, Katherine J. Sessions1, Kaylin Dalhberg1, Meghan Fondakowski1, Edward J. Usherwood2, and Dimitry N. Krementsov1.

1Department of Biomedical and Health Sciences, University of Vermont

2Department of Microbiology and Immunology, Geisel Sch. of Med., Dartmouth Col.

Inefficient host control of infection by the gammaherpesvirus Epstein-Barr virus (EBV) is a risk factor for several autoimmune diseases and cancers. However, naturally occurring host genetic determinants mediating immune control of gammaherpesvirus replication are largely unknown. Previously, we discovered that wild-derived and genetically divergent PWD/PhJ (PWD) mice infected with MHV-68, a gammaherpesvirus homologous to EBV, control viral burden at levels much lower than C57BL/6 (B6) mice. Here, we used several approaches to identify host genetic determinants of gammaherpesvirus control in PWD mice. Analysis of B6PWDF1 and B6PWDF2 mice suggested the existence of multiple host genetic elements contributing to this superior control, with one acting in a dominant manner. To identify these host genetic elements, we performed a genetic screen in C57BL/6J-ChrPWD chromosome substitution (consomic) mice, which identified a consomic strain carrying PWD chromosome 17 (Chr17PWD) as fully capturing the MHV-68 resistance of PWD mice. Genetic mapping using a subconsomic Chr17SPWD strain demonstrated that a PWD-derived interval at 27.6-49.4 Mb on Chr 17, which includes the MHC/H-2 complex, was required for MHV-68 resistance. B6Chr17PWDF1 experiments confirmed a dominant inheritance pattern for the resistance locus on Chr 17. An N2/N3 breeding strategy is ongoing and 11 new subconsomic lines have been generated to narrow the locus of interest on Chr 17, demonstrate its sufficiency for resistance, and allow for in-depth analysis of immunological traits linked to superior control of viral replication. In search of the remaining genetic element(s) identified in the B6PWDF2 mice, and given that Chr17PWD mice, unlike PWD, showed reduced viral loads independent of NK cells, we also tested C57BL/6J-Chr6PWD (Chr6PWD) consomic mice, which carry the highly polymorphic NK receptor complex (NKC). We found that Chr6PWD mice were not resistant to MHV-68 and that Chr6PWDChr17PWDF1 mice showed no evidence of epistasis between the MHC and NKC. Taken together, our results identify a novel MHC-linked locus on Chr 17 in mice that regulates control of gammaherpesvirus viral burden, in agreement with emerging studies in humans linking MHC variants to control of EBV.

7, P-35

Modeling background dependent tumor growth in response to ERBB3 inhibition using mouse derived organoids

Wyatt W. Porter1, Kaitlyn E. Carter1,2, David W. Threadgill1,2

1 Department of Nutrition, Texas A&M University, College Station, TX, USA

2 Interdisciplinary Program in Genetics and Genomics, Texas A&M University, College Station, TX, USA

Colorectal cancer (CRC) is the second leading cause of cancer-related death worldwide, highlighting the urgent need for effective therapeutic strategies. This project investigates one such potential target, ERBB3, by examining differences in cell viability and proliferation in response to ERBB3 inhibition in C57BL/6J (B6) and 129S1s/SvlmJ (129) ApcMin/+ mice. Using mouse-derived organoids to model colorectal cancer (CRC), we aim to explore why preclinical studies suggested ERBB3 inhibition could reduce polyp number, whereas clinical trials have shown limited success, emphasizing the potential role of genetic context in therapeutic outcomes.

Organoids from both strains werearecollected and utilizedgenerated and subsequently treated with EGFR, ERBB3, and dual EGFR-ERBB3 inhibitors. Growth rate inhibition, proliferation, and cell viability assays are performed to measure and validate long-term kinetic assays using live cell imaging systems. We expect to see increased cell proliferation and increased cell viability in C57BL/6JB6 Apcmin/+ mice and decreased cell proliferation and decreased cell viability in 129 Apcmin/+ mice, to support what was found in previous in vivo workanalyses. This research will provide increased evidence that background differences between strains can impact outcome. Furthermore, by conducting this experiment with organoids instead of a mouse model, we plan to demonstrate that organoids can add robustness to a study when combined with the mouse model.

Dissecting the role of cis- and trans- regulatory elements in human-mouse hybrid cells

Gilbert Giri1, Grant Goda2, Jose C. Martinez3, Daniel Dominguez4

1Department of Genetics, UNC Chapel Hill

2Department of Chemistry, UNC Chapel Hill

3Department of Hematology, UNC Chapel Hill

4Department of Pharmacology, UNC Chapel Hill

Background: Protein-coding sequences are highly conserved across mammals, yet species display significant phenotypic divergence largely driven by regulatory differences at transcriptional, RNA processing, and translational levels. These differences arise from interactions between cis-acting sequence elements (e.g. promoters, RNA motifs) and transacting factors (e.g. DNA and RNA binding proteins), which together determine gene expression and transcript diversity. However, the evolution and interaction of cis- and trans-regulatory mechanisms in shaping species-specific gene regulation remains poorly understood. Additionally, most functional and genomic cross-species studies were performed by conducting experiments in each species’ native cellular environment, making it difficult to disentangle cisand trans-specific effects due to uncontrolled cellular contexts.

Design: To address this gap, I am using novel human–mouse hybrid cell models combined with next-generation sequencing to directly compare cis- and trans-acting regulatory factors genome-wide. These cells contain human and mouse chromosomes in a shared nuclear environment, enabling us to determine gene expression and RNA processing of human and mouse genes in same cellular environment. Furthermore, comparing hybrids with their parent cells reveals the impact of trans factors while controlling for cis content at chromosome level.

Results: I developed a novel computational pipeline to align human and mouse sequencing reads from a single sample, removing cross-aligning reads for improved mapping accuracy and accurate gene expression quantification. Comparison of one-to-one human-mouse ortholog expression in the hybrid environment revealed substantial expression divergence, i.e. human and mouse copies differed in their expression levels. Also, comparison of gene expression with respective parent cells revealed varied degree of divergence. These differences were not fully explained by sequence differences in gene and promoter region, thus implicating the role of trans-acting regulatory elements. To further explore this possibility, we will employ ATAC-seq to access whether species-specific differences in chromatin accessibility correlate with underlying sequence divergence and contribute to observed differences in expression. Furthermore, splicing analysis revealed differential alternative splicing of human transcripts in hybrid cells, likely driven by mouse splicing machinery acting on human RNAs. Despite splicing factors being highly conserved between mouse and humans, these altered splicing patterns reveal a novel cross-species interplay between cis- and trans- acting elements.

Conclusion: Our work represents one of the few genome-wide analysis of factors affecting gene regulation and first one to study RNA processing in a hybrid cellular environment. By dissecting cis- and trans-acting regulators of transcription and splicing, we intend to reveal their influence in shaping species-specific gene regulation. These findings offer unprecedented insight into molecular basis of evolutionary divergence and will help design better cross-species pharmacological studies.

9, P-30

Geography Outweighs Host Genetics in Shaping Gut Microbiota of Giant Pandas: Implications for Conservation and Rewilding

Sudhanshu Mishra1,2 *, Ying Li1

1Chemistry Department, School of Physical Sciences, DIT University, Dehradun- 248009, INDIA

2Farm Animal Genetic Resources Exploration and Innovation Key Laboratory of Sichuan Province, Sichuan Agricultural University, Chengdu 611130, Sichuan, China.

The giant panda (Ailuropoda melanoleuca), which survives on a highly specialized bamboo-based diet low in nutrition and energetic value, presents unique conservation challenges. A better understanding of its gut microbiota composition and associated functions is therefore essential for evaluating health, adaptability, and long-term survival. Although host genetics, diet, and environmental conditions are all known to influence gut microbial communities, their relative contributions in wild giant panda populations remain insufficiently resolved. In this study, we investigated the factors driving gut microbiota variation in wild giant pandas, with particular emphasis on the relationships among microbial community structure, host genetic diversity, and geographic distribution. In addition, we assessed microbiota functional profiles to identify potential ecological risks associated with current panda management and conservation practices. A total of 78 fecal samples were collected from wild pandas across multiple geographic regions. Host genetic diversity was evaluated using the mitochondrial Cytochrome-b and D-loop markers while gut microbiota composition was characterized through high-throughput pyrosequencing of the V4–V5 region of 16S rRNA gene. Functional analyses identified antibiotic resistance genes (ARGs), virulence factors (VFs), and heavy metal tolerance genes. UniFrac distances evaluated genetic and geographic influences on microbiota diversity. The gut microbiota of wild giant pandas was dominated by Proteobacteria (57%), Bacteroidetes (22%), and Firmicutes (18%) with pronounced variation observed within and among populations. Genetic analysis of the same pandas revealed moderate to high genetic diversity, however, genetic distances showed no significant association with microbiota composition (P > 0.05). In contrast, geographic location exerted significant influence on gut microbiota composition (P < 0.05). Notably, microbiomes of captive pandas showed higher abundance of ARGs, VFs, and heavy metal tolerance genes, raising concerns about the potential dissemination of these traits to wild populations following reintroduction. Our findings highlight geography as a primary determinant of gut microbiota structure in wild giant pandas, while captivity-induced microbial alterations pose ecological risks. These results suggest that conservation strategies must prioritize habitat protection and restoration to preserve region-specific microbial communities and mitigate risks of rewilding captive pandas. Future research on the functional consequences of microbiota shifts will be critical for refining evidence-based conservation and management practices.

Keywords: Giant panda, gut microbiota, biogeographic Influence, conservation metagenomics, host genetics

10, P-7

Understanding the genotype-phenotype correlation of ZIC1 variants

Michelle Chang1, Isaac Walton2, Lisa Leinhos2, Karsten Boldt3, Tina Beyer3, Stephen Twigg2, Ruth Arkell1

1 John Curtin School of Medical Research, Australian National University, Canberra, Australia

2 Clinical Genetics Group, MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford, UK

3 Institute for Ophthalmic Research, Center for Ophthalmology, University of Tübingen, Tübingen, Germany

The zinc finger of the cerebellum genes (ZIC1-5) encode transcription factors that share a highly conserved zinc finger domain that functions in protein interaction, nuclear localisation and DNA binding. The ZIC family play critical roles in early embryonic development and, when disrupted, cause distinct congenital defects. Loss-of-function alleles (i.e., heterozygous deletions) of the tandemly paired ZIC1 and ZIC4 genes on chromosome 3q, are associated with cerebellar abnormalities including Dandy-Walker malformation. More recently, probands with heterozygous mutations of ZIC1 alone were identified and can be grouped into two different clinical presentations; neurodevelopmental delay (NDD) with or without craniosynostosis.

Variants associated with both NDD and craniosynostosis were clustered at the C-terminal end of the protein and were predominantly nonsense variants predicted to escape nonsense mediated decay. Variants associated with NDD alone included nonsense variants in exon 1 and missense substitutions within exon 2 predicted to disrupt the normal function of the DNA binding zinc fingers. Here, using a luciferase reporter assay, we have demonstrated that variants associated with NDD alone show significantly decreased ZIC1 transcriptional activity compared to wildtype ZIC1, while variants associated with NDD and craniosynostosis demonstrate transcriptional activity similar to that of wildtype.

As these C-terminal variants occur near an evolutionarily conserved NEWYV motif resembling a PDZ binding motif, our current working model is that the loss of the well-characterised ZIC1 transcription factor function produces the NDD phenotype. In contrast, the mutations disrupting the ZIC1 C-terminus drive a gain-of-function effect causing craniosynostosis and NDD via ZIC function in a process other than transcription. Strep/FLAG Tandem Affinity Purification (SF-TAP) of full length and truncated FLAG-tagged ZIC1 constructs have identified candidate proteins that interact with full length, but not truncated ZIC1. These include members of the PAR complex (PAR3, PAR6, and aPKC) that are crucial in establishing and maintaining cell polarity during morphogenetic processes required for the developing coronal sutures. Investigating this protein-protein interaction will provide a novel hypothesis for ZIC function beyond that of transcription.

11, P-41

Comparison of skeletal phenotypes between genetically distinct Down syndrome mouse models using a composite score derived from principal component analysis

Kourtney Sloan1, Pathum Randunu Nawarathna Kandedura Arachchige2 , Charles R. Goodlett3, Gayla R. Olbricht2, Randall J. Roper1

1Department of Biology, Indiana University, Indianapolis, Indiana, USA

2Department of Mathematics and Statistics, Missouri University of Science and Technology, Rolla, Missouri, USA

3Department of Psychology, Indiana University, Indianapolis, Indiana, USA

More than 20 mouse models have been generated to represent Down syndrome (DS), which is caused by the triplication of human chromosome 21 (Hsa21). Due to genomic differences between humans and mice, Hsa21 orthologous genes are found on three different mouse chromosomes. This contributes to the complexity of modeling DS in mice regarding which genes and how many are triplicated, how extra genomic content is transmitted (aneuploidy vs. segmental duplication or translocation), and what constitutes the genetic background (the source of modifier genes). Skeletal structure of long bones, including trabecular and cortical bone, has been quantified in over 10 mouse models of DS, but direct comparisons between mouse models have not been performed. In addition to the variations in mouse models, multiple variables are assessed for each bone compartment, and each variable may contribute to the overall structure yet may be highly correlated, presenting statistical challenges for between-model comparisons. To compare the skeletal structure changes directly between Dp1Tyb, Ts66Yah, and TcMAC21 DS model mice during skeletal growth (6-weeks-old) and at skeletal maturity (16-weeks-old), a composite score was generated by first using z-score normalization to minimize effects of genetic background and differences in imaging methodology. Then, principal component analysis of the standardized data was performed to determine the contribution of each variable to overall trabecular (5 variables) or cortical (7 variables) structure before reduction to a single variable for each compartment through linear combination of the standardized variables and principal component 1 loadings of each variable. The trabecular composite score was significantly decreased in male mice of all three DS models compared to male euploid littermates both during skeletal growth and at skeletal maturity, whereas female mice were unaffected. The cortical composite score was significantly decreased in both sexes during skeletal growth. At skeletal maturity, this cortical composite score deficit persisted in both sexes of Dp1Tyb and male Ts66Yah mice yet were not present in either sex of TcMAC21 nor female Ts66Yah mice. Additionally, there were instances where the composite score was significantly decreased despite a lack of significant deficits in any individual variable, suggesting this method can detect low severity or subclinical phenotypes. Together, the composite scores allowed for straightforward quantitative determination of phenotypic differences, direct comparisons between animal models, and detection of subclinical phenotypes. Future evaluation of this technique with other multivariate phenotypes or subphenotypes and models of genetic conditions is necessary. Modifications may be needed if the variables are uncorrelated or the first principal component does not explain a sufficient proportion of the variance.

12, P-11

Effects of a functional reduction of Dyrk1a on cerebellar structure in otherwise trisomic male and female Down syndrome model mice

Andrew Folz1, Charles R. Goodlett1,2, Randall J. Roper1

1Department of Biology, School of Science, Indiana University Indianapolis, IN, USA

2Department of Psychology, School of Science, Indiana University Indianapolis, IN, USA

Down syndrome (DS) is caused by the triplication of human chromosome 21 (Hsa21), which includes the gene DYRK1A. Spatiotemporal regulation of DYRK1A levels is necessary for proper brain development, thus both over- and under-expression of DYRK1A are associated with abnormal brain development and impaired cognitive phenotypes in those with DS and those with DYRK1A syndrome, respectively. Overexpression of Dyrk1a in DS mouse models results in similar neurological phenotypes as seen in individuals with DS. When Dyrk1a copy number is genetically reduced from three to two at conception in B6EiC3Sn a/A-Ts(1716)65Dn/J (Ts65Dn) mice, some cerebellar phenotypes are rescued, such as granule and molecular layer size as well as granule and Purkinje cell densities. We have shown trisomic Ts65Dn mice have elevated cerebellar DYRK1A protein levels in both male and female mice at postnatal day (P)6. This suggests that P6 may encompass a treatment window to target Dyrk1a overexpression in Ts65Dn mice. Utilizing a Tet-On Cre system in Ts65Dn mice, one copy of Dyrk1a was functionally reduced at birth in pups (Ts65Dn,Dyrk1a+/+/dox-cre), simulating a postnatal DYRK1A normalization treatment that may be used for individuals with DS. After functionally reducing Dyrk1a at P0, the cerebella of P21 mice are removed, sectioned, and stained with Nissl (all cells) or Calbindin (Purkinje cells) staining. Using systematic stereological analysis, total cerebellar, granule cell layer, and molecular cell layer volumes along with cell counts and cell densities of the granule layer, molecular layer, and Purkinje cell layer are quantified. Preliminary results show decreased cerebellar and molecular layer volume and granule cell count in female trisomic mice as compared to female euploid mice, but males show no significant differences in any of the measured cerebellar parameters between trisomic and euploid mice. Male Ts65Dn,Dyrk1a+/+/dox-cre mice show a reduction of total cerebellar and granule cell layer volumes as compared to the trisomic mice, but no differences between female trisomic and Ts65Dn,Dyrk1a+/+/dox-cre mice. Examining sex differences, only male Ts65Dn,Dyrk1a+/+/dox-cre mice have a decreased granule layer volume as compared to female Ts65Dn,Dyrk1a+/+/dox-cre mice. These preliminary results suggest that the functional reduction of Dyrk1a might be detrimental to trisomic male mice at this time point, whereas female mice are showing a non-significant trend towards a rescue effect in the Ts65Dn,Dyrk1a+/+/dox-cre model.

13, P-17

Identification of Significant Sex Specific FV Expression Differences in Mice

Adrianna M. Richards, Kailey H. MacFadyen, Caitlin D. Schneider, Marisa A. Brake, Arina Rodionova and Randal J. Westrick

Oakland University, Rochester Michigan

Coagulation Factor 5 (F5) is a protein in the coagulation cascade found in two circulating compartments with 80% in plasma and 20% in platelet alpha-granules. Human F5 is made by liver hepatocytes, contributing to the F5 in the plasma, and plasma F5 is endocytosed by megakaryocytes and stored in platelet alpha granules. In mice, hepatocytes produce plasma F5, and megakaryocytes directly produce platelet F5. F5 deficiency causes bleeding of the mucosal lining as well as post-trauma. Complete absence of F5 in humans results in increased bleeding. In mice, F5 deficiency is incompatible with life. Currently, little is known about the genetic regulation of this important protein at the center of coagulation. We previously generated Bacterial Artificial Chromosome (BAC) F5 transgenic mice to dissect the individual physiological roles of plasma F5. F5 gene expression was restricted to the liver using a BAC construct with the albumin promoter conferring liver specificity (F5LiverTg). We bred the F5LiverTg onto the F5 deficient (F5+/- ) background and performed F5LiverTg F5+/- x F5+/- intercrosses to generate a series of genotypes for F5 analyses. We analyzed the plasma of F5LiverTg F5+/- x F5+/- offspring for F5 activity and antigen levels. In addition, we analyzed the plasma of C57BL/6J (B6), DBA/2J (DBA), A/J, CAST/EiJ (CAST), and 129S1/SvImJ (129S1) strains for F5 antigen levels to obtain information on these strains. The F5 activity assay resulted in significant sex differences in the percent F5 activity in the following genotypes (F5LiverTg F5-/- (p=0.0175), F5LiverTg F5+/- (p=0.0079), and F5LiverTg F5+/+ (p=0.0148). There was no significant difference in the F5+/- mice. Analysis of the F5 antigen levels via ELISA revealed a significant difference between the males and females in F5 levels in the F5LiverTg F5-/- mice (p=0.0002). There was also a trend towards significance in the F5LiverTg F5+/- (p=0.1316) mice. In all genotype categories analyzed, the males have higher percent activity and plasma F5 levels. Additionally, analysis of the mouse strains revealed significant sex differences in the DBA and CAST strains (p=0.0011 and 0.0236 respectively). Our findings indicate a significant sex difference due to the F5LiverTg. This suggests that the F5 transgene is susceptible to regulation by sex hormones. Additional analysis of mouse strains identified significant sex differences in F5 levels in the CAST and DBA strains. The F5LiverTg DNA construct sequence and the CAST and DBA regulatory sequence may contain hormone response elements. This is an area of active investigation. Future work of identifying these regulatory elements will increase our understanding of sex specific gene regulation. This may reveal novel targets or strategies for modulating FV in humans, which could be used to treat hemostatic disease.

Androgens mediate sex differences in a mouse model of Pilarowski-Bjornsson Syndrome

Kimberley Jade Anderson1,2, Eirný Tholl Thorolfdóttir1, Hans Tómas Björnsson1,2,3,4

1 Department of Genetics and Molecular Medicine, Landspítali University Hospital; Reykjavík, Iceland.

2 Louma G. Laboratory of Epigenetic Research, Faculty of Medicine, University of Iceland; Reykjavík, Iceland.

3 McKusick-Nathans Department of Genetic Medicine, Johns Hopkins University School of Medicine; Baltimore, MD, USA.

4 Department of Pediatrics, Johns Hopkins University; Baltimore, MD, USA.

Email: kimberl@hi.is

Pilarowski-Bjornsson Syndrome (PILBOS, OMIM: 617682) is a rare neurodevelopmental disorder characterized by growth retardation, hypotonia and autism spectrum disorder caused by heterozygous variants in the chromatin remodeler gene CHD1. To investigate the mechanism of disease, we generated a patient-specific knock-in mouse model harboring the heterozygous Chd1R616Q/+ variant using CRISPR-Cas9. An in vitro nucleosome remodeling assay demonstrated impaired chromatin remodeling activity of CHD1 harboring the p.R616Q variant compared to the wildtype protein, supporting loss of function as the disease mechanism. Phenotypic analysis revealed striking female-limited effects, with heterozygous females exhibiting significant reductions in body weight (P<0.001) impaired motor performance (P<0.01), and increased anxiety-like behavior (P<0.05) relative to wildtype littermates, whereas mutant males were largely unaffected. Female mutants also displayed elevated hypothalamic oxytocin levels (P<0.05). We hypothesized that androgens may protect against CHD1-associated phenotypes and tested this through hormonal manipulation. Orchiectomy of male mice at postnatal day 15 uncovered a previously masked growth deficit in Chd1R616Q/+ males compared to orchiectomized WT littermates (P<0.01). Conversely, testosterone supplementation in female mice led to rescue of the growth phenotype. At the cellular level, embryonic neural progenitor cells (NPCs) isolated from Chd1R616Q/+ mice displayed a significant proliferation defect (P<0.05). We observed that testosterone treatment of pregnant dams was able to normalize proliferation rates in the embryonic mouse cortex (P<0.001). Consistently, in vitro dihydrotestosterone treatment broadly rescued transcriptional dysregulation observed in Chd1R616Q/+ NPCs, suggesting that androgens may counteract CHD1 dysfunction by promoting the normal transcriptional program. Finally, analysis of the gnomAD and UK Biobank human population datasets revealed a significant enrichment of rare (< 1% frequency) missense alleles in CHD1 within the male population compared to females, suggesting that males may be protected from the disease phenotype even at the population level. These findings establish a sexually dimorphic mouse model of PILBOS and identify androgen signaling as a modifier of CHD1-associated neurodevelopmental phenotypes, with implications for sex-biased penetrance in Mendelian forms of autism.

Host genetic background shapes susceptibility and neurological outcomes of Tick-Borne Encephalitis in Collaborative Cross mice

Luc Fery-Simonian1, Laurine Conquet1, Léa Santerre1, Bruno Passet2, Nolwenn Jouvenet3, Sarah Bonnet2, Sandra Martin-Latil4, Caroline Manet1

1Mouse Genetics, Immunity and Infections, Institut Pasteur, Université Paris Cité, Paris, France

2Ecology and emergence of arthropod-borne pathogens Unit, Institut Pasteur, Université Paris Cité, CNRS UMR 2000, INRAE USC 1510, Paris, France

3Virus Sensing and Signaling Unit, Department of Virology, Institut Pasteur, Université Paris Cité, CNRS UMR 3569, Paris, France

4ANSES, INRAE, ENVA, UMR Virologie, F-94700 Maisons-Alfort, France

Tick-borne encephalitis virus (TBEV) is a neurotropic orthoflavivirus transmitted primarily by Ixodes ticks and represents the most significant tick-borne viral infection in Europe and Asia. Human infection is characterized by striking inter-individual variability, ranging from asymptomatic cases to severe encephalitis associated with long-term neurological sequelae. This heterogeneity strongly points to a major role of host genetic factors in shaping susceptibility and outcomes. However, identifying these genetic determinants remains difficult in humans due to limited case numbers, complex exposure histories, and genetic heterogeneity. To address this challenge, we use the Collaborative Cross (CC), a panel of genetically diverse recombinant inbred mouse strains. Multiple CC strains were infected subcutaneously with TBEV and were monitored for clinical signs, body weight evolution, and survival. Tissue collection enabled the quantification of viral loads in peripheral organs and the brain to characterize systemic infection and neuroinvasion. Across CC mice, we observed a broad spectrum of susceptibility, from resistant strains to highly susceptible ones exhibiting rapid disease progression and early mortality after symptom onset, with intermediate profiles in between. Notably, CC040 mice developed marked neurological signs, including tremors and loss of balance, yet survived, providing a unique model to investigate non‑fatal neurological disease and its potential sequelae. Importantly, initial analyses revealed that brain viral loads correlate with clinical severity, supporting a link between viral neuroinvasion and neurological disease expression. Besides, Oas1b, an interferon-stimulated restriction factor known to limit infection by West Nile and Powassan viruses, likely contributes to TBEV control. Indeed, several CC strains carrying a functional Oas1b allele showed no mortality or clinical signs. However, CC039 mice, despite also harboring a functional allele, remained highly susceptible, with ~40% mortality, marked body-weight loss, and clinical signs, indicating that Oas1b alone does not confer full resistance to TBEV. This work provides the first characterization of TBEV infection in CC mice, revealing strong genetic control of disease outcome. These novel models offer a unique resource to capture the diversity of human disease and dissect host determinants of susceptibility and neuroinvasion. Ongoing work will extend this approach to a natural tick–mouse transmission model to examine how host genetics also shape viral acquisition and transmission between host and vector.

Generation and Characterization of an Syt1-D365E mouse model for Baker-Gordon Syndrome using CRISPR-Cas9 Targeted Allele Editing

Samantha M. Norris1*, Sai Goutham Reddy Yeddula1, Elaine Su2, Klancey Vandeloecht2, and Daniel J. Davis1,2

1Department of Pathobiology and Integrative Biomedical Sciences (PIBS), University of Missouri, Columbia, Missouri

2Animal Modeling Core, University of Missouri, Columbia, Missouri

Baker-Gordon syndrome is a rare neurodevelopmental disorder diagnosed in children with severe developmental delays and characterized by mutations in the synaptotagmin-1 (SYT1) gene. The SYT1 gene codes for the protein synaptotagmin-1 (SYT1), which is a neuronal synaptic vesicle protein required to facilitate proper neurotransmitter release. Clinically in patients, the mutant SYT1 protein produces a dominant-negative phenotype by desynchronizing neurotransmitter release which interferes with wildtype SYT1 protein function. For this reason, homozygous SYT1 mutations are incompatible with life, and heterozygous mutations can result in severe phenotypes after birth such as developmental delays, neonatal hypotonia, and feeding difficulties. Currently, no treatment for Baker-Gordon syndrome exists, and little is known about the factors driving variable symptoms in patients. To address these gaps, we sought to develop a heterozygous mouse model carrying a patient-identified SYT1 mutation that faithfully recapitulates the disease. CRISPR-Cas9 mediated genome editing was used to introduce a patient-derived, disease-associated D365E missense mutation into the endogenous mouse Syt1 locus. However, the combination of lethal biallelic targeting with a severe heterozygous phenotype, complicates the generation of a mammalian disease model with standard CRISPR-Cas9 gene editing procedures. To circumvent this complication, we forced heterozygosity by first incorporating a synonymous Y364Y silent mutation immediately upstream of the desired pathogenic variant. A guide RNA specific to the Y364Y sequence was then designed to selectively target and edit the modified allele, enabling introduction of the D365E mutation while leaving the wildtype allele intact. This strategy successfully generated heterozygous Syt1-D365E mice, as confirmed by both DNA and RNA sequencing. Mutant animals exhibited a pronounced failure-to-thrive phenotype, closely mirroring the clinical features observed in individuals diagnosed with Baker-Gordon syndrome. Additionally, supportive care provided in the early stages of life resulted in increased survival. Given the similarity in genotype and phenotype, the Syt1-D365E mouse will provide a robust and translational platform for investigating disease mechanisms and evaluating therapeutic strategies for Baker-Gordon syndrome.

Rhbdl2 modulates inflammation, coagulation and vascular responses in experimental pneumococcal meningitis

D. Bakker, R. Koning, Marian A. van Roon, V. Jaspers, M.C. Brouwer, D. van de Beek

Amsterdam UMC

Background

A genome-wide association study performed by our group identified a single nucleotide polymorphism located in the intronic region of the Rhomboid-related protein 2 (RHBDL2) gene that was associated with increased incidence of cerebral infarctions in patients with bacterial meningitis (rs143464422, MAF = 0.05; odds ratio (OR) = 3.15; p = 5.0 × 10–8). Here, we studied the effect of Rhbdl2 gene knock out in a murine model of pneumococcal meningitis.

Methods

Male and female Rhbdl2tm1b(KOMP)Wtsi knockout (n = 24) and C57BL/6N mice (n = 24), aged 8-12 weeks, were inoculated intracisternally with 0.5*10^5 CFU of Streptococcus pneumoniae (D39 WT). Mice were treated with ceftriaxone (100 mg/kg) at 20 and 32 hours post inoculation (hpi) and were euthanized at 20hpi or 44hpi. Disease severity was determined according to a predefined clinical severity score. Bacterial outgrowth was determined in the brain, blood and spleen. Levels of endothelial and coagulation markers were determined in plasma with ELISA. The left hemisphere of the brain was cut and haemorrhagic spots were counted at 0.3 mm intervals.

Results

At 20 hpi, bacterial outgrowth and expression of pro-inflammatory markers were significantly reduced in the brain of Rhbdl2 -/- mice compared to controls (Fig. 1). Levels of soluble ICAM-1, soluble VCAM-1and D-dimer were increased in plasma of Rhbdl2 -/- mice compared to controls, while thrombomodulin and IL-6 levels remained unchanged (Fig. 2). We observed no difference in the number and size of haemorrhagic spots between the two groups at 44hpi (Fig. 3).

Conclusion

Despite being identified as a genetic risk factor for cerebral infarction in patients with pneumococcal meningitis previously, we did not observe an effect of Rhbdl2 on the number of haemorrhages. This might be due to the fact that cerebral infarctions do not occur in this experimental setting. Interestingly, Rhbdl2 deletion resulted in lower bacterial burden and inflammation, while increasing markers of endothelial activation and coagulation. These findings suggest potential pathways through which genetic variation in Rhbdl2 may contribute to infarction risk in patients with pneumococcal meningitis.

Figure 1. Decreased bacterial load and inflammation in Rhbdl2-/- mice compared to wild-type mice (WT) at 20 hours post infection. A. Bacterial load (CFU/organ) in the brain at 20 hours post infection. B. Gene expression of Il-1β, Cxcl2 and Il6 in the brain at 20 hours post infection. N=12 mice per group. Differences between groups were calculated with a Mann Whitney U test. Ns = not significant, * = P < 0.05, ** = P < 0.01.

Figure 2. Effect of Rhbdl2 on inflammatory, endothelial and coagulation markers. Levels of IL-6, soluble ICAM-1 (sICAM-1), soluble VCAM-1 (sVCAM-1), D-dimer and thrombomodulin were measured in the blood of Rhbdl2-/- mice and wild-type mice (WT) at 20 hours post infection. For control animals; N=6 mice per group, for infected animals; N=12 mice per group. Differences between groups were calculated with a Kruskal Wallis test. Ns = not significant, * = P < 0.05, ** = P < 0.01.

Figure 3. The number and size or haemorrhagic spots in the left hemisphere of the brain of Rhbdl2-/- mice and wild-type mice (WT) at 44 hours post infection . For control animals; N=12 mice per group. Differences between groups were calculated with a Mann Whitney U test. Ns = not significant.

Slc1a4 knockout mice model motor deficits associated with SPATCCM

Teresa M. Gunn1-3, Megan L. Ratz-Mitchem1, Allison Mace1,2, Heather Wyman1-3, Derek Silvius1,3, Andrea Grindeland Panter1-3, Michael P. Kavanaugh1,4

1The McLaughlin Research Institute, Great Falls, MT, USA;

2 Weissman Hood Institute at Touro University, Great Falls, MT, USA;

3Touro University College of Osteopathic Medicine – Montana, Great Falls, MT USA;

4The Division of Biological Sciences, The University of Montana, Missoula, MT, USA

SLC1A4 (solute carrier family 1 member 4) encodes a sodium-dependent neutral amino acid transporter also referred to asASCT1(alanine/serine/cysteine/threonine-preferring transporter 1). SLC1A4 is highly expressed on astrocytes in the brain, where it is proposed to regulate neurotransmitter homeostasis and N-methyl-D-aspartate (NMDA) receptor signaling through the uptake of L- and D-serine. Biallelic mutations in SLC1A4 cause the rare autosomal recessive neurodevelopmental disorder spastic tetraplegia, thin corpus callosum, and progressive microcephaly (SPATCCM, OMIM 616657). Although most SPATCCM-associated mutations are missense mutations, patients with putative null alleles typically present with a more severe clinical phenotype. We previously reported that Slc1a4 knock-in mice expressing the most common pathogenic human SLC1A4 variant (p.Glu256Lys) exhibit microcephaly and a thin corpus callosum but lack overt cognitive or motor deficits. Here, we describe Slc1a4 knockout mice (Slc1a4em1Tmg) generated on the C57BL/6J background using CRISPR/Cas9-gene editing. While homozygotes were smaller than wildtype mice, their brain weight was proportionate to body weight and corpus callosum thickness appeared normal. In contrast to the knock-in model, Slc1a4 knock-out mice developed a resting tremor by 6-months-of-age, showed impaired performance on the rotarod, and exhibited gait abnormalities consistent with reduced motor coordination and stability. These findings indicate that complete loss of Slc1a4 preferentially disrupts motor function in mice and suggest that this knockout model may provide a valuable platform for dissecting the mechanisms underlying motor impairments in SPATCCM patients and evaluating potential therapeutic interventions.

The Verne Chapman Lecture

My Journey in Mouse Genetics: a 30 Year Detour

Fernando Pardo Manuel de Villena

University of North Carolina at Chapel Hill

In 1994 in joined the laboratory of Carmen Sapienza in Philadelphia, changing the trajectory of my career by selecting the laboratory mice as the major tool for my research program. It also was my introduction of the IMGS community, and in particular to the many of its members that over the years have served as mentors, collaborators and colleagues. Looking back at the past three decades, I have tried to juggle a selfish passion about some obscure aspects of mammalian genetics, evolution and the search for “exceptions” to basic genetic principles, with a growing interest in the development of shared resources and the realization that service at all levels may be, on the final count, my most impactful professional contribution to the field. I will share my journey and a few lessons learned, from meiotic drive, to the Collaborative Cross, and to institutional programs to improve the rigor and impact of the science that we do. Verne Chapman is widely remembered for his passion for shared resources and his community focus (“I have a mouse that may interest you” comes to mind). In this time when uncertainty and misinformation are rampant, I hope that this lecture will contribute to renew the IMGS commitment to excellence in science, meaningful transparency and real-world impact on the career of our trainees.

Evidence-Based, Accessible Mouse Genetic Quality Control: Educating Researchers Through MMRRC Strain GQC

Matthew W. Blanchard1,2, James Amos-Landgraf3, Timothy A. Bell1, Jennifer Brennan2, Charisse Carlson4, Dominic Ciavatta1,2, Paul Cotney1, Steve A. Murray5, Samit Patel4, Laura Reinholdt5, Franklin Sandridge1, John Sebastian Sigmon2,6, Fernando Pardo Manuel de Villena1,2,7

1Department of Genetics, The University of North Carolina at Chapel Hill,

2Mutant Mouse Resource and Research Centers, University of North Carolina at Chapel Hill

3Mutant Mouse Resource and Research Centers, University of Missouri

4Mutant Mouse Resource and Research Centers, University of California at Davis

5Mutant Mouse Resource and Research Centers, The Jackson Laboratory

6Department of Computer Science, The University of North Carolina at Chapel Hill

7Lineberger Comprehensive Cancer Center, The University of North Carolina at Chapel Hill

The mission of the MMRRC is to archive and distribute laboratory mouse strains carrying alleles of interest to the biomedical research community. A key aspect of this mission, strain-level Genetic Quality Control (GQC) was recently adopted as part of MMRRC standard operations; we are actively generating and distributing Strain GQC Reports for ordered strains or strains that are maintained live on the shelf. This rollout also involves educating the broader mouse research community about the process and purpose of Strain GQC. The MMRRC Strain GQC process will be reported in a policy paper accepted in Science entitled “Improve genetic quality control to increase rigor and reproducibility of mouse research.” The two main shortcomings of the current GQC as identified in this paper are that strain names do not match the genetics for almost half of the strains studied, and that the standard nomenclature cannot capture the genetic complexity of 80% of congenic strains. These results demonstrate the immediate need for an improved GQC process. The paper describes the MMRRC solution to these shortcomings in the form of MMRRC Strain GQC report. The MMRRC at UNC is creating a series of four videos: a high-level justification of the need for an improved GQC process; an overview targeted at the lay audience that succinctly describes how to read and interpret a Strain GQC report; and two technical videos on the strain classifier and the estimation of genome replicability. To increase the outreach of these videos, each provides a brief, high-level presentation, accessible to all members of the mouse community, and uses real-world examples to illustrate a few key concepts. We expect that the Strain GQC process and associated videos will improve the rigor and reproducibility, the quality and impact of mouse-based research, and increase adherence to the 3R principles of vertebrate animal research.

James Amos-Landgraf

3Rs

Age-related Behavioural and Molecular Landmarks in New Mouse Models for Studying Alzheimer’s Disease in Down Syndrome

Monika Rataj Baniowska1,#, Paige Mumford2,#, Francesca Prestia3, Pauline Stephan1, Millie Beament2, Marie-Christine Birling4, Chiara Lanzillotta3, Letizia Ciafardini3, Eugenio Barone3, Gloria Lau2, Claire Chevalier1, Chadia Nahy1, Nadia Messaddeq1, Thais Lestra2, Yixing Wu2, Valérie Nalesso1, Fabio Di Domenico3,* , Frances Wiseman2,* and Yann Herault1,4,5, *

#, * equal contributions

1 Université de Strasbourg, CNRS, Inserm, Institut de Génétique Biologie Moléculaire et Cellulaire, IGBMC, UMR 7104- UMR-S 1258, F-67400 Illkirch, France

2 UK Dementia Research Institute at UCL, London, United Kingdom.

3 Department of Biochemical Sciences A. Rossi Fanelli, Sapienza University of Rome, Piazzale Aldo Moro 5, 00185, Rome, Italy

4 Université de Strasbourg, CNRS, Inserm, PHENOMIN-Institut Clinique de la Souris, PHENOMIN-ICS, CELPHEDIA, F-67400 Illkirch, France

5 CNRS, CELPHEDIA Core, UAR2052, Villejuif, France

Down syndrome (DS) is the most prevalent genetic cause of intellectual disability and Alzheimer’s disease (AD), with over 90% of individuals developing AD-related dementia (DSAD). The triplication of the APP gene on chromosome 21 drives early amyloid-β (Aβ) accumulation, but other Hsa21 genes contribute to neurodevelopmental and neurodegenerative changes. Despite extensive research, current DSAD models are limited by species-specific differences in Aβ sequence, hindering accurate recapitulation of human pathology. Here, we generate and characterize two novel DSAD mouse models that incorporate partial humanization of Aβ. These models exhibit key features of early AD, including cognitive impairment, hyperactivity, changed response to novelty or to risk, tau hyperphosphorylation, and endolysosomal dysfunction. APP processing is shifted toward the β-secretase pathway, resulting in increased CTF-β levels and altered Aβ dynamics. Notably, Aβ humanisation modulates behavioural outcomes, improving performance in specific cognitive tasks while enhancing anxiety-related traits. Myelinosome formation and autophagic flux impairment further align these models with human AD pathology. Our findings underscore the complex interplay between trisomy, APP dosage, and Aβ sequence in shaping DSAD progression. These models provide a valuable resource for investigating early mechanisms of AD in DS and for evaluating targeted therapeutic strategies, pending the development of a fully humanised three-copy DSAD model.

Novel MIDY-mouse models and a surprising lack of phenotype severity in female mice

Vitus Wenig1,2, Maximilian Schmidtke1,2, Aliona Harten1,2, Gerhard K.H. Przemeck1,2, Martin Hrabě de Angelis1,2,3

1Institute of Experimental Genetics, Helmholtz Zentrum München, Neuherberg, Germany

2German Center for Diabetes Research (DZD), Neuherberg, Germany

3Chair of Experimental Genetics, TUM School of Life Sciences, Technische Universität München, Weihenstephan, Germany

Mutant INS-gene Induced Diabetes of Youth (MIDY) is a rare and severe monogenic form of neonatal diabetes, typically manifesting within the first six months of life. As no curative therapy exists, patients rely on life-long insulin treatment, which carries a substantial burden and risk of complications even with well-controlled glycemia. The disease is caused by heterozygous point mutations in the INS gene, leading to misfolding proinsulin, massive endoplasmic reticulum (ER) stress and finally severe disruption of insulin production. Understanding the MIDY pathomechanism, particularly how distinct mutations affect proinsulin folding, is critical for the development of targeted therapeutic approaches. The use of different genetic backgrounds allows for broader insight into how genetic factors modulate MIDY manifestation, providing a framework for understanding the impact of genetic diversity in humans. In this study, we performed a comprehensive phenotypic characterization of two novel mouse models, Ins2C109G and Ins2V26D, and compared them to the gold-standard Ins2Akita model. Both lines were derived from an ENU mutagenesis screen and are on a C3HeB/FeJ genetic background, which is generally more robust than the commonly used C57BL/6J strain of Ins2Akita models. All three mutant lines developed severe, early-onset diabetes, as evidenced by elevated blood glucose levels from 4 weeks onwards and impaired glucose tolerance at 6 weeks of age. Glucose-stimulated insulin secretion of isolated pancreatic islets was strongly impaired in vitro at 4 and 12 weeks of age. Despite the diabetic phenotype, the islets retained the ability to secrete insulin in response to glucose and KCl. However, total insulin content in the islets was severely reduced, leading to a stark decrease in the amount of insulin being secreted. Female mutants showed a milder reduction in islet insulin content, consistent with a better glycemic control and generally a less severe diabetic phenotype. Further, male mice at 12 weeks of age showed reduced plasma insulin levels and a disrupted islet architecture. However, substantial beta-cell mass persisted even after prolonged diabetes at 22 weeks of age. Notably, female mice were largely protected across all measured parameters. Overall, the newly developed models are viable MIDY models that complement the Ins2Akita line. Comparisons between the C3HeB/FeJ and C57BL/6J backgrounds show that genetic background does not substantially affect the development of the diabetic phenotype, which strengthens the translatability of these findings to humans, who exhibit high genetic diversity. Despite carrying the same underlying mutations, female mice maintained better glycemic control, suggesting a sex-specific mechanism of protection. This can largely be explained by the sex-specific levels of estrogen, which, via multiple mechanisms, protects beta cells, enhances peripheral insulin sensitivity and exerts antioxidative and anti-inflammatory effects. The relatively mild phenotype in females is striking because the molecular defect in insulin folding is identical in both sexes, highlighting that systemic factors can significantly modulate disease severity. The combination of multiple mutations and different genetic backgrounds provides a robust framework for understanding how MIDY manifests under diverse physiological contexts, offering insights that would be missed using a single mutation or background alone. In conclusion, this study demonstrates that these models faithfully recapitulate key aspects of MIDY, reveal sex-specific differences in disease severity, and provide translationally relevant systems for investigating disease mechanisms and testing therapeutic interventions.

RARA mutation enhances a novel interaction with AEBP1 to promote breast cancer

Anthony J Schulte1*, Caroline Downes2, Carson D Broeker3, Mylena Ortiz4, John Vusich5, Eran R Andrechek2

1Michigan State University, Pharmacology and Toxicology, East Lansing, MI;

2Michigan State University, Physiology, East Lansing, MI;

3Michigan State University, Biochemistry and Molecular Biology, East Lansing, MI;

4Michigan State University, Genetics and Genome Sciences, East Lansing, MI;

5Michigan State University, Cell and Molecular Biology, East Lansing, MI.

Tumor evolution enhances aggressiveness, metastasis, and drug resistance, resulting in patient mortality. Secondary genetic alterations are frequent and change tumor biology, yet are understudied. As evidence, the transcription factor retinoic acid receptor alpha (RARA) is mutated 30% of the time in breast fibroepithelial tumors. Previous literature fails to clearly elucidate the role of RARA and how to functionalize the targeting of it in breast cancer. We therefore sought to establish mechanisms by which RARA drives this disease. Remarkably, we have discovered a novel interaction of adipocyte enhancer binding protein 1 (AEBP1) with RARA. We hypothesize this RARA-AEBP1 interaction is driving tumorigenic mechanisms in RARA altered breast cancer. In the MMTV-Myc mouse model of breast cancer, whole genome sequencing has revealed a conserved mutation of A255D in RARA in the microacinar histological subtype. Additional sanger sequencing confirmed 32 of 38 tumors with this alteration, compared to the papillary subtype with 0 of 16 tumors with this mutation. This evolutionarily selected genetic alteration mirrors mutations seen in human breast fibroepithelial tumors, providing an insightful model to understand how RARA can drive breast cancer. In a proximity-based protein labeling experiment, BioID, we identify a novel interaction of RARA with AEBP1, a known transcriptional repressor. CUT&RUN confirms genomic binding for these proteins in the same location and moreover informs that AEBP1 binding is dependent upon RARA. Interestingly, in mutant RARA, this interaction persists despite ligand activation whereas in wild-type RARA, activation ablates this interaction. Further investigation reveals mutant RARA tumors exhibit decreased signaling of RARA target genes, such as RARB and CDKN1B, that are crucial to induce differentiation and reduce cell proliferation respectively. These data indicate this mutation enhances RARA interaction with the corepressor AEBP1, reducing RARA receptor activity to alter programs favorable for tumor growth. Overall, the results suggest aberrant RARA-AEBP1 interactions enhances tumorigenesis and thus elucidates AEBP1 as a vulnerability to reduce tumor growth and improve survival in breast fibroepithelial tumor patients.

An In Vivo Study of Novel Genetic Modifiers in Autosomal Recessive Polycystic Kidney Disease (ARPKD)

Paraskevi. Goggolidou¹, S.A. Malik¹

¹University of Wolverhampton, Wolverhampton, United Kingdom

Autosomal recessive polycystic kidney disease (ARPKD) is a severe inherited ciliopathy characterised by bilaterally enlarged cystic kidneys, hepatic fibrosis and pulmonary hypoplasia. The disease is primarily caused by mutations in PKHD1 (Polycystic Kidney and Hepatic Disease 1) and DZIP1L genes, though variable disease severity among patients suggests the involvement of genetic modifiers. To investigate these mechanisms, we generated a novel CRISPR/Cas9-engineered mouse model carrying the T36M missense mutation in Pkhd1, which represents the most common and severe mutation in human ARPKD patients.

Comprehensive phenotypic characterization of Pkhd1T36M/T36M mice revealed a progressive, multiorgan disease phenotype. Histological analysis demonstrated tubular dilation in kidneys from postnatal day 2, with immunofluorescence studies in 9-month-old mice showing diminished Fibrocystin expression and cytoskeletal disorganisation, evidenced by disrupted F-actin staining patterns. Importantly, cilia length and number remained unaffected. Hepatic manifestations included progressive liver fibrosis developing by 9 months of age, with increased expression of fibrosis markers compared to wild-type controls.

Unexpectedly, we identified a novel pulmonary phenotype in 4-month-old Pkhd1T36M/T36M mice, characterised by reduced alveoli and enlarged airspaces. Molecular analysis revealed significant reductions in mRNA expression of Pkd1, Atmin and Vangl2 in affected lungs. This lung pathology occurred independently of severe renal manifestations, suggesting a primary role for pulmonary involvement in ARPKD pathogenesis rather than a secondary consequence of kidney disease.

This Pkhd1T36M/T36M mouse model recapitulates key features of human ARPKD and provides a valuable platform for identifying genetic modifiers and investigating the molecular mechanisms underlying disease variability. The identification of lung-specific gene expression changes offers new insights into ARPKD as a multisystem disorder and may reveal novel therapeutic targets for this currently untreatable condition.

Keywords: ARPKD, PKHD1, mouse model, genetic modifiers, ciliopathy, pulmonary hypoplasia

Early onset diabetic phenotype and impaired beta cell maturation in two novel MIDY mouse models

Maximillian R Schmidtke1,2; A Harten1,2; GKH Przemeck1,2; M Hrabě de Angelis1,2,3

1Functional Genetics, Institute of Experimental Genetics, Helmholtz Zentrum München, 85764, Neuherberg, Germany

2German Center for Diabetes Research (DZD), 85764, Neuherberg, Germany

3Chair of Experimental Genetics, TUM School of Life Sciences, Technische Universität München, 85354, Weihenstephan, Germany

Mutant INS-gene-induced diabetes of youth (MIDY) is a severe form of monogenic diabetes characterized by increased endoplasmic reticulum (ER) stress due to misfolding mutant proinsulin. This results in reduced insulin availability and early-onset diabetes. So far, most research has focused on the postweaning phase, neglecting early disturbances to beta cell differentiation and maturation. In mice, these issues may begin as early as embryonic day 13.5 when insulin is first detectable and thus misfolding mutant proinsulin also being expressed. Given that mutant proinsulin is nearly incapable of reaching its native folding state, this immediately disturbs ER homeostasis. This study aims to address this gap in knowledge by analysing the preweaning phase in two novel MIDY mouse models, Ins2C109G and Ins2V26D, with a particular focus on beta cell maturity. In addition, we also compared our findings against the gold standard MIDY model, the Ins2Akita mouse.

Analysing those mouse models, we observed slightly elevated fasted blood glucose levels already at P21 in all lines, regularly exceeding 250 mg/ml by P28. Interestingly, plasma insulin levels were not altered compared to wild-type controls at all analysed timepoints (P7, P15, P21, P28). As this contrasts with the marked hyperglycaemia, we evaluate the functional state of Ins2Mutant beta cells to dig deeper into the underlying disease mechanism. Confocal microscopy of immunofluorescent stains on FFPE pancreatic tissue revealed a drastic reduction in beta cells positive for MAFA at all timepoints, indicating impaired maturation, while NKK6.1 positivity, responsible for beta cell differentiation and identity, remained unaffected. Remarkably, islet architecture was largely preserved, even amidst severe hyperglycaemia at P28. In line with the reduced MAFA positivity beta cells of mutant mice failed to upregulate expression of GLUT2, the primary insulin-independent glucose transport of beta cells, with a complete lack of expression by P28. Interestingly beta cell dedifferentiation, as assessed by the expression of the beta cell disallowed gene Aldh1a3, only manifested after P28, suggesting dedifferentiation is a secondary event caused by elevated blood glucose levels. This was despite increased BIP expression at P7 already, indicative of ER stress.

Our findings highlight an early onset of the diabetic phenotype in MIDY mouse models during the preweaning phase. The reduction in MAFA and GLUT2 positive cells as early as P7 indicates an arrest in beta cell maturation without progression to functional maturity. Contrary to common assumptions, our data suggest that beta cells lose functionality long before the onset of diabetic symptoms, despite largely healthy islet architecture and beta cell mass. Our findings suggest that beta cells become locked in an immature, dysfunctional state shortly after insulin is being expressed due to overwhelming ER stress. Further investigations are highly advisable to elucidate underlying mechanisms and inform potential new treatment strategies for MIDY.

Loss of RNA cytidine acetyl transferase NAT10 alters enhancer organization via p300 mislocalization and suppresses lung metastasis

Ruhul Amin, Ngoc-Han Ha, Tinghu Qiu, Huaitian Liu, Ronald Holewinski, Andy D. Tran, Maxwell P. Lee, Thorkell Andresson, Jordon L. Meier and Kent W. Hunter

National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892.

The development of secondary tumors, metastasis, is associated with the majority of breast cancer mortality. Despite the importance of this process in patient survival, the etiology of metastasis is poorly understood. Recent data from human sequencing studies and work from our laboratory have demonstrated that unlike tumor etiology, there do not appear to be high frequency somatic events associated with progression to metastasis, suggesting that the metastatic process is more likely associated with transcriptional modulation through epigenetic or inherited polymorphic events. Previously, our laboratory identified a number of genes associated with inherited predisposition to breast cancer metastasis. Intriguingly, protein-protein interaction analysis has revealed many interactions between these independently identified genes, as well as other common interaction partners. One common partner is the RNA acetylase NAT10 and we have demonstrated that loss of NAT10, and NAT10’s interacting protein co-factor THUMPD1, significantly decreases lung metastasis in genetically engineered mouse models of breast cancer metastasis. Further investigation has revealed that NAT10 operates upstream of our previously identified mechanosensitive pathway associated with the nuclear pore protein, NUP210. Moreover, loss of NAT10’s ability to acetylated leucine-tRNA molecules results in substantial reorganization of the genome and alteration of enhancer organization through interference with p300 histone acetylation. This leads to an inability of tumor cells to recruit pro-metastatic neutrophils necessary for the initial steps of the metastatic cascade, potentially through alterations in the biophysical properties of transcriptional condensates.

The National Center of Rabbit Models for Translational Research: Expanding Access to Genetically Engineered Rabbits

Zachary T Freeman1,2, Brooke D Pallas1,3, Jifeng Zhang3,4, Jie Xu3,4 Dongshan Yang3,4 Y Eugene Chen3,4

1Unit for Laboratory Animal Medicine, University of Michigan

2Transgenic Animal Model Core

3Center for Advanced Models for Translational Sciences & Therapeutics

4Division of Cardiovascular Medicine, University of Michigan

Rabbit models have been critical to diverse scientific breakthroughs, including the development of in-vitro fertilization, the HPV vaccine, and numerous antibodies for various scientific applications. Despite these successes, advances in genetic engineering have primarily favored mice, due to the accessibility of embryonic stem cells. Advancements in CRISPR/Cas9 allow for precise genetic modifications in large animal species including rabbits, overcoming past technical barriers and enabling the creation of translationally relevant genetically engineered (GE) rabbit models at scale. While species such as pigs and non-human primates can be genetically modified, rabbits combine biological and practical benefits including shorter generation times, cost effective housing, while maintaining increased translational relevance. Compared to rodents, GE rabbits offer increased translational relevance for many fields, including cardiovascular, ocular, neurology, immunology, and infectious disease. Our team has developed a robust genome editing pipeline, expertise in rabbit embryology and cryopreservation, and an improved rabbit reference genome to facilitate the generation of more than 50 different translationally relevant GE rabbit lines. With support from the NIH, we are launching the National Center of Rabbit Models for Translational Research (NCRM). The NCRM mission is to expand access to highly relevant GE rabbit models with unique relevance to human disease. The Center will generate and maintain a diverse portfolio of humanized GE rabbits, distribute these models to the scientific community, and offer hands-on workshops to promote effective use and further innovation. By leveraging our combined expertise in GE rabbit models and genome engineering, a centralized NIH resource will improve the generation of and access to, translationally relevant GE rabbits.

Dysfunctional mitochondria in KMT2D-deficient chondrocytes drive premature hypertrophy and cellular senescence

Sara Tholl Halldorsdottir1,2, Agnes Ulfig1, Stefán Pétursson1, Hans Tomas Bjornsson1-4

1Louma G. Laboratory of Epigenetic Research, Faculty of Medicine, University of Iceland; Reykjavik, Iceland

2Department of Genetics and Molecular Medicine, Landspitali University Hospital; Reykjavik, Iceland

3McKusick-Nathans Department of Genetic Medicine, Johns Hopkins University School of Medicine; Baltimore, MD, USA

4Department of Pediatrics, Johns Hopkins University; Baltimore, MD, USA

Kabuki syndrome type 1 (KS1) is a genetic disorder caused by heterozygous pathogenic variants in the gene encoding the histone methyltransferase KMT2D. KS1 is a multi-organ disorder commonly associated with intellectual disability, skeletal abnormalities, and growth deficiency. However, the mechanisms underlying these phenotypic manifestations remain poorly understood. Longitudinal bone growth occurs through endochondral ossification, a process tightly regulated by chondrocyte differentiation within the growth plate. Disruptions of this process are known to lead to skeletal growth abnormalities. Our studies have investigated the effects of KMT2D deficiency on chondrocyte maturation and identified premature chondrocyte differentiation, hypertrophy and senescence as key drivers of skeletal abnormalities in KS1. In an established KS1 mouse model, we previously observed reduced femur and tibia length, along with altered growth plate architecture, particularly affecting the size of the hypertrophic zone. Here, we show with sc-RNA sequencing that KMT2D-deficient chondrocytes exhibit decreased cell cycling accompanied by accelerated differentiation and early senescence when exposed to supraphysiological oxygen levels (20% O2). These changes are associated with altered cellular metabolism, including aberrant upregulation of the hypoxia response and glycolysis, driven by defects in the mitochondrial electron transport chain. This mitochondrial dysfunction results in oxidative stress and increased mitochondrial reactive oxygen species (ROS). By pharmacologically neutralizing the excessive ROS or differentiating the cells in hypoxic environments, we were able to mitigate the premature hypertrophy and restore normal chondrocyte differentiation. Together, our findings demonstrate that loss of KMT2D induces oxidative stress-driven chondrocyte hypertrophy, disrupting the balance in cartilage growth. These results provide a novel insight into the role of KMT2D in the Kabuki disease phenotype and identify mitochondrial ROS regulation as a potential therapeutic target for skeletal defects in KS1.

Dyrk1a dosage in Cortical Interneuron Development: insights for DYRK1A and Down syndromes

Victorine ARTOT1, Véronique BRAULT1, Yann HERAULT1,2

1Université de Strasbourg, Centre National de la Recherche Scientifique (CNRS) UMR7104, Institut National de la Santé et de la Recherche Médicale (INSERM) U1258, Institut de Génétique et de Biologie Moléculaire et Cellulaire (IGBMC), IllkirchGraffenstaden, F-67404, France.

2CELPHEDIA, PHENOMIN Institut Clinique de la Souris (ICS), Illkirch-Graffenstaden, F-67000, France.

Proper development of excitatory and inhibitory neural systems is essential for normal brain function. Disruptions in the balance between these systems are a hallmark of various neurodevelopmental disorders, including Down syndrome (DS) and DYRK1A-haploinsufficiency syndrome. In DS, the excitatory/inhibitory imbalance is commonly attributed to gene dosage effects, particularly involving DYRK1A, a gene located on chromosome 21. While Dyrk1a overexpression has been implicated in altered neurogenesis and intellectual disability, its role in GABAergic interneuron development remains underexplored. Conversely, Dyrk1a haploinsufficiency, caused by gene mutations, is associated with epilepsy, intellectual disability, and autism spectrum disorder, suggesting that both increased and decreased dosage of Dyrk1a can impair GABAergic circuit formation. In this study, we investigated the role of Dyrk1a in interneuron development using complementary approaches in mouse models of both DS and DYRK1A-haploinsufficiency. In both models, we used a conditional genetic approach based on the Dlx5a-Cre driver (1), which targets postmitotic GABAergic interneurons. Starting from the Dp(16)1Yey DS model (2) that carries three copies of Dyrk1a, we generated mice with either two or one functional copies of the gene in interneurons by combining the Dlx5a-Cre transgene with a floxed Dyrk1a allele. This strategy allowed us to normalize Dyrk1a dosage specifically in interneurons (two copies) or mimic haploinsufficiency (one copy), enabling a direct comparison of the effects of gene dosage imbalance on interneuron development. We found that Dyrk1a dosage critically controls multiple stages of GABAergic interneuron development, including progenitor proliferation, tangential migration, and cortical integration. Both overexpression and haploinsufficiency of Dyrk1a disrupted these processes, albeit through distinct cellular mechanisms. In the DS model, overdosage altered interneuron proliferation and delayed cortical migration, leading to cognitive deficits, rescued by dosage normalization. In the haploinsufficiency model, reduced Dyrk1a impaired interneuron migration through defective actomyosin cytoskeleton remodeling, resulting in behavioral abnormalities and epileptic activity.

(1) Brault V. et al., 2021. Dyrk1a gene dosage in glutamatergic neurons has key effects in cognitive deficits observed in mouse models of MRD7 and Down syndrome. PLoS Genet. doi: 10.1371/journal.pgen.1009777.

(2) Li Z. et al., 2007. Duplication of the entire 22.9 Mb human chromosome 21 syntenic region on mouse chromosome 16 causes cardiovascular and gastrointestinal abnormalities. Hum Mol Genet. Doi: 10.1093/hmg/ddm086.

The Bill Dove Lecture

Phenotype-driven Genetics and the Joy of Discovery

(Or: why do we do this)

David Beier, M.D., Ph.D.

Seattle Children’s Research Institute/University of Washington

Forward genetics is a method in which one identifies an organism with an abnormal phenotype and then uses molecular analysis to identify the mutated causal gene. In the mouse, this approach has been powerfully enabled by efficient methods of creating mutations (such as treatment with ENU), diligent assessment of phenotypes, and positional cloning strategies using genetic mapping. Bill Dove was one of a small cadre of mouse geneticists to embrace these methods, and this talk is dedicated to his insight and mentorship.

My own experience with phenotype-driven genetic analysis began as a post-doc with the ascertainment of a spontaneous mutation causing polycystic kidney disease (PKD), and the ultimate identification of the causal gene as the previously uncharacterized kinase Nek8, which has been shown to be aberrant in the human cystic disorder NPHP9. In my career as an independent investigator, I employed ENU mutagenesis with collaborators including David MacDonald, Pam Tran, Rolf Stottman and Theresa Gunn to identify genes required for normal mammalian development, which would potentially be causal for human congenital defects. We characterized over 20 of such loci, developing along the way methods of genetic mapping that took advantage of state-of-art technologies for genomic analysis.

This work resulted in numerous interesting discoveries, including identifying a hypomorphic mutation in Zfpm2 (Fog2) as causal for a mouse model of congenital diaphragmatic hernia; variants in this gene are presently the most commonly ascertained mutations identified in the human disorder. We also found a mutation in Ttc21b, a cilial intraflagellar transport gene whose loss results in an unexpected increase in SHH activity, and which led me to an as-yet unfinished journey down the rabbit hole of Hedgehog signaling.

The development of robust methods of genome editing and the application of these to mice in a high through-put manner (e.g., via the KOMP effort) yielded the unexpected result that about a third of mammalian genes are required for normal embryonic development and post-natal survival. This has made the utilization of mutagenesis for this analysis less compelling. We have consequently directed our attention to a different phenotype – which is presumptive haploinsufficiency in humans. In collaboration with Shamil Sunyaev, we examined human sequence data and developed a metric for assessing selection against the heterozygous null state. Remarkably, while the highest decile of genes includes many well-known to affect mammalian development, most have not been associated with any human disorders, and a large number are not at all functionally annotated. Our focus as developmental geneticists has been overly underneath the streetlamp, and there remains an abundance of discovery yet to be done.

Deciphering the consequences of CLCN4 mutations in rat models for a better understanding of the rare X-linked CLCN4-related neurodevelopmental condition

Tania Sorg,Marie-Christine Birling, Géraldine Prevost, Fabrice Riet, Yann Hérault.

Université de Strasbourg, CNRS, INSERM, CELPHEDIA, PHENOMIN, Institut Clinique de la Souris (ICS), Illkirch, France

CLCN4-related neurodevelopmental condition (CLCN4-NDC) is a rare X-linked condition caused by inherited or de novo pathogenic variants of the CLCN4 gene, affecting both sexes and being associated with intellectual disability, epilepsy, gastrointestinal issues, movement and behavioural disorders.

CLCN4 encodes the CIC-4 protein, a 2Cl−/H+ exchanger present in various tissues, more highly expressed in brain and skeletal tissue. The function of CIC-4 is not yet well understood, and how CLCN4 pathogenic variants lead to the observed clinical symptoms remains unclear. The medical needs of CLCN4-NDC patients remain largely unmet with only supportive multidisciplinary therapies available. Thus, there is a need to better understand the pathophysiology of CLCN4-NDC by using an appropriate animal model. Given that the rat Clcn4 gene is on the X chromosome like in humans, in opposite to the mouse, we generated a rat Clcn4 LOF “KO” line and a “KI” line harbouring the A549V GOF variant. We then generated the experimental cohorts of animals to evaluate the suitability of these rat lines as animal models for CLCN4-NDC. We carried out a detailed phenotypic characterisation of CIC-4 LOF and GOF hemizygous males and heterozygous females using a test battery covering learning and memory, coordination, locomotor activity, neuromuscular function, compulsive behavior, anxiety, depression, blood analysis, proteomics and histological analysis of relevant organs. We could show that Clcn4 GOF males display behavioural and neurological phenotypes, while heterozygous females show only a few mild phenotypes. In contrast, Clcn4 LOF rats no not display any phenotype.

Overall, this project will enable us to evaluate how well the rat model fits with the clinical spectrum of CLCN4-NDC in a LOF or GOF context and determine its usability for future research on the CLCN4-NDC pathophysiology and potential therapeutic developments.

Assessing the potential of in vivo prime editing to rescue neuro-developmental phenotypes

Alexandria Reyes, Teresa M. Gunn

Touro College of Osteopathic Medicine - Great Falls; Weissman-Hood Institute at Touro (McLaughlin Research Institute)

SOX10 is essential for oligodendrocyte (OL) differentiation and survival. Mutations in this gene cause central nervous system (CNS) de/dysmyelination in humans and SOX10 dysfunction is associated with inherited leukodystrophies. Mice homozygous for the Sox10gt mutation (an A to G transition that changes a highly conserved glutamate to glycine in the dimerization domain) develop early-onset spongiform encephalopathy and myelination defects. This model provides an optimal platform for evaluating the CNS effects of disrupted SO10 function and testing in vivo prime editing as a therapeutic strategy for CNS myelination disorders. To better understand disease pathogenesis and define the critical treatment window, we examined whether Sox10gt/gt OLs undergo apoptosis and determined the temporal progression of myelin abnormalities. Preliminary results indicated decreased myelination in the cerebellum of Sox10gt/gt mice by postnatal day 12. To test the efficacy of prime editing in ameliorating CNS phenotypes of Sox10gt/gt mice, we are using a split-AAV system with the PHP.eB serotype, which efficiently crosses the blood–brain barrier, to deliver the pegRNA and the PE3 prime editor to neonatal pups via temporal vein injection. Transduction efficiency and distribution are first being assessed using AAV particles that express an mCherry reporter under control of the minimal human myelin associated glycoprotein (MAG) promoter. Litters of pups from Gt/Le-Sox10gt/+ intercrosses will be infused with AAVs expressing prime editing components, and rescue assessed 2-3 weeks later using histology and immunohistochemistry (ICH) to evaluate vacuolation and myelination. This work aims to establish proof-of-concept for prime editing as a treatment for inherited leukodystrophies and to provide foundational data necessary for developing future gene-based therapies for CNS myelination disorders.

One hundred and sixty generations of breeding work – how to effectively use opposingly selected lines of mice?

Marta Gajewska, Norbert Gałka, Paulina Stermach, Julia Maciocha, Aleksandra Garbacz, Wiesław Świderek

Department of Animal Genetics and Conservation, Institute of Animal Sciences, Warsaw University of Life Sciences, Warsaw, Poland

Nearly 60 years ago, a breeding experiment began, which continues to this day. A cross between four inbred mouse strains created an outbred stock, from which lines opposingly selected for weight gain have been derived. For over 160 generations, breeding work has continued with subsequent generations of researchers, and new research methods have enabled more detailed information about the animals being maintained. The selected lines, known as LIGHT (Kghz:L) and HEAVY (Kghz:C), currently differ not only in body weight but also in behavior, growth rate, maturation, aging, the occurrence of spontaneous diseases, lifespan, and even the composition of the gut microbiome. Because they are maintained in a specific facility (a part of an animal breeding faculty), they have been used primarily as models of production traits in farm animals. New research results suggest that they can also be successfully used as models for the study of polygenic diseases in human populations – sites that provide a stable, predictable phenotype while maintaining much greater genetic variability than that found in inbred strains.

Multigenic Regulation of Obesity in the Berlin Fat Mouse: Bbs7 and Acat2 Variants

Gudrun A. Brockmann¹,†, Danny Arends¹˒², Deike Hesse¹

¹ Albrecht Daniel Thaer-Institut für Agrar- und Gartenbauwissenschaften, Humboldt-Universität zu Berlin, Unter den Linden 6, 10099 Berlin, Germany
² Department of Applied Sciences, Northumbria University, Newcastle upon Tyne, UK

† Corresponding author

The Berlin Fat Mouse Inbred (BFMI) line is a genetic model for juvenile obesity. Quantitative trait locus and fine-mapping studies identified Bardet–Biedl syndrome 7 (Bbs7) as a major contributor to this phenotype. Bbs7 expression is reduced by about 60% in brain and adipose tissue of BFMI compared with the lean reference strain C57BL/6N (B6N), explaining approximately 40% of the variance in fat mass.

Whole-genome sequencing revealed multiple BFMI-specific variants in Bbs7. Using CRISPR/Cas9, we introduced a 1,578 bp BFMI-derived intronic deletion containing a CTCF binding site into B6N. Complementation tests showed partial rescue (13–15%) of obesity, indicating that this deletion contributes but is not solely responsible for the phenotype. Hi-C analysis confirmed altered three-dimensional chromatin conformation in the Bbs7 region of BFMI mice.

Dual-luciferase assays comparing BFMI and B6N promoter haplotypes with 16 sequence variants demonstrated reduced reporter activity for the BFMI allele. SNP rs29947545 in the 5′UTR (Chr3:36,613,350) reduced transcriptional efficiency; however, the same allele occurs in other strains with normal Bbs7 expression, implying additional compensatory variants that are absent in BFMI.

We propose that cumulative effects of cis-regulatory and coding mutations in Bbs7 and Acat2 drive the obesity phenotype. A deleterious Acat2 mutation further promotes hepatic steatosis by disrupting acetyl-CoA metabolism. These combined variants define a multigenic framework of obesity in BFMI, paralleling the polygenic architecture of human metabolic disorders.

Additionally, a deleterious mutation in Acat2 on the obese BFMI background promotes ectopic hepatic lipid storage by impairing the acetyl-CoA metabolism in adipose tissue. We propose that cumulative effects of multiple cis-regulatory variants in Bbs7 and coding variants in Acat2 underlie the BFMI obesity phenotype. Together, these data illustrate how interacting genomic variants shape complex metabolic traits in BFMI, paralleling polygenic obesity in humans.

Novel mechanisms of MITF regulation identified in a mouse suppressor screen

Hong Nhung Vua, Matti Már Valdimarssonb, Sara Sigurbjörnsdóttira, Kristín Bergsteinsdóttira, Julien Debbachec, Keren Bismuthc, Deborah A. Swingd, Jón H. Hallssona, Lionel Laruee, Heinz Arnheiterc, Neal G. Copelandd,f, Nancy A. Jenkinsd,f, Petur O. Heidarssonb, and Eiríkur Steingrímssona,1

aDepartment of Biochemistry and Molecular Biology, BioMedical Center, Faculty of Medicine, University of Iceland, Sturlugata 8, 102 Reykjavík, Iceland.

bDepartment of Biochemistry, Science Institute, School of Engineering and Natural Sciences, University of Iceland, Sturlugata 7, 102 Reykjavík, Iceland.

cMammalian Development Section, NINDS, NIH, Bethesda, MD 20892-3706.

dMouse Cancer Genetics Program, NCI, Frederick, MD 21702-1201.

eInstitut Curie, PSL Research University, INSERM U1021, Normal and Pathological Development of Melanocytes, 91405, Orsay, France.

fCurrent address: Genetics Department, MD Anderson Cancer Center, Houston, TX 77030

Mutations which suppress the phenotypic effects of other mutations has led to many important insights into gene functions and genetic pathways. We have generated an intragenic suppressor mutation in the mouse Mitf gene, a transcription factor gene with major roles in regulating melanocyte development and melanoma. The suppressor mutation corrects the pigmentation phenotype of other Mitf mutations, in some cases from nearly entirely white to normal coat color. The novel suppressor mutation, termed Mitf-spotless (Mitfmi-sl), terminates MITF at the K316 SUMOylation site leading to the loss of the C-terminal intrinsically disordered region (IDR). Intriguingly, however, Mitfmi-sl homozygotes display a brownish coat color instead of the normal black. The MITF-SL protein is more nuclear but less stable than wild-type MITF and retains DNA-binding ability. Interestingly, as a dimer, it can translocate wild-type and mutant MITF partners into the nucleus, improving its own stability thus ensuring nuclear MITF supply. smFRET analysis showed the effects of the truncation are largely due to the K316 SUMOylation and S409 phosphorylation sites; mutating these sites simultaneously replicates the K316 deletion. The altered dynamics and localization conferred by the suppressor mutation highlight the critical role of the MITF C-terminus in modulating protein function. Notably, the recurrent melanoma-associated E318K mutation in MITF, which affects K316 SUMOylation, also alters protein regulation in concert with S409. This suggests that residues K316 and S409 of MITF are impacted by SUMOylation and phosphorylation, respectively, mediating effects on nuclear localization and stability through conformational changes. Our work provides a novel mechanism of genetic suppression, and an example of how apparently deleterious mutations can lead to normal phenotypes.

Characterization of a highly active Murine Leukemia Virus transposable element and its effect on the genomic stability of the CC015/Unc colony

Martin T. Ferris1,2, Timothy A Bell1, Pablo Hock1, Paul Cotney1, Anwica Kashfeen1, Leonard McMillan3, Rachel M Lynch1,2, Fernando Pardo-Manuel de Villena1,2,3

1Department of Genetics, University of North Carolina at Chapel Hill
2Systems Genetics Core Facility, University of North Carolina at Chapel Hill
3Department of Computer Science, University of North Carolina at Chapel Hill

We have recently discovered that the Collaborative Cross mouse strain, CC015/Unc, has a highly active Murine Leukemia Virus (MuLV) transposable element (TE). A second element (RLTR4*) is hitchhiking using MuLVs retrotransposition machinery in CC015 as well. We have assessed the other 62 CC strains, as well as several dozen other inbred strains, and found no evidence of such activity for these elements. Since our initial description of this activity, we have used two modalities of sequencing (short and long reads) in two complementary ways: determining the timing of new insertions and determining the activity of novel insertions. Utilization of high coverage short-read sequencing within nuclear pedigrees can allow for determining the timing of activity by determining the fraction of reads within an offspring containing a novel insertion not present in the parents, where 0.5 would indicate gametic activity, with reducing fractions indicated later stages in embryogenesis. Concurrently, long read sequencing can be used to directly associate single nucleotide variants arising during retrotransposition with de novo insertions in pedigree progeny, allowing us to determine whether relatively new insertions are themselves active. Using our data from a nuclear pedigree of two parents and their 3 male offspring, we have shown that insertion occurs exclusively at the 2-cell embryo stage (with 25% of autosomal reads or 50% of X-chromosome reads at de novo insertion loci containing evidence of a TE), and that multiple de novo MuLV insertions are active within the CC015 colony.

Mechanisms of motivation to approach humans in mice selected for active tameness from a wild-derived heterogeneous stock

Tsuyoshi Koide 1,2, Bharathi Venkatachalam 1,2, Sarah Dagher 3, Shimpei Ishiyama 4

1. Mouse Genomics Resource Laboratory, National Institute of Genetics, Mishima, Shizuoka, Japan.

2. Department of Genetics, SOKENDAI, Mishima, Shizuoka, Japan

3. Max Planck Institute for Metabolism Research, Köln, Germany

4. Central Institute of Mental Health, Mannheim, Germany

We aim to elucidate the mechanisms underlying animal domestication using a domesticated mouse model derived from wild mice. We generated a wild-derived heterogeneous stock (WHS) population from eight wild mouse strains and conducted selective breeding for tameness, defined as the propensity of animals to voluntarily approach humans (active tameness). As a result, the selected lines exhibited significantly higher levels of active tameness than control lines.

Behavioural analyses of both selected and non-selected lines further revealed that the selected mice displayed higher levels of social behaviour compared with controls. Additional analyses demonstrated that the selected mice responded to human hand-tickling by emitting ultrasonic vocalisations resembling “laughter” and exhibited play-like behaviours, such as chasing the human hand.

To identify brain regions associated with active tameness, we examined behaviour-induced neuronal activation using immunohistochemical analysis of c-FOS expression as a marker. In non-selected mice, brain regions including the dorsal premamillary nucleus (PMd), the dorsomedial hypothalamus (DMH), and the ventromedial hypothalamus (VMH), which are associated with fear and avoidance responses, were activated following behavioural testing, whereas such activation was not observed in the selected mice. These findings indicate that active tameness is associated with suppression of neural circuits involved in fear and avoidance.

Together, our results suggest that active tameness is promoted by mechanisms related to intraspecific social behaviour, while fear- and threat-related circuits exert inhibitory effects. We are currently developing experimental platforms to enable detailed genetic analyses of active tameness, and we would like to discuss these approaches in this presentation.

Genetic and molecular basis for a large X-effect hybrid incompatibility causing obesity in deer mouse hybrids

M. Wyatt Touré1,2, Riya Rampalli1,2, Adib Jalali Rabbani1,2, Dasha Dworkin-Cantor1,2, Andrés Bendesky1,2

1Zuckerman Mind Brain Behavior Institute, Columbia University, New York, NY, USA.

2Department of Ecology, Evolution and Environmental Biology, Columbia University, New York, NY, USA

Antagonistic genetic interactions causing sterility or inviability of hybrids between diverging lineages, simply known as hybrid incompatibilities, can reinforce reproductive isolation leading to the formation of new species. Mammalian hybrid incompatibilities map disproportionately to the X chromosome, an observation known as the large X-effect. Moreover, mammalian hybrid inviability is often driven by growth abnormalities. Despite consistent evidence that the X chromosome is important in speciation, the specific X-linked loci and molecular mechanisms generating hybrid inviability across mammals remain poorly understood. To discover genetic and molecular mechanisms that contribute to mammalian growth abnormalities, we studied growth disruptions in deer mouse (Genus: Peromyscus) hybrids. Crosses between a P. maniculatus bairdii (BW) mother and P. polionotus subgriseus (PO) father produce viable and fertile offspring that are smaller than both parents. In contrast, PO mothers crossed with a BW father produce extremely overgrown offspring and inviable pregnancies. Through quantitative trait mapping, we established that part of the hybrid overgrowth is due to a 7 Mb interval containing PO ancestry on chromosome X (mexlA, mass-effect X-linked locus A) interacting with autosomal BW ancestry. To dissect the molecular basis of hybrid overgrowth, we isolated mexlA in a congenic line. We found that mexlA overgrowth is primarily driven by an increase in body fat, with mexlA hybrids having over double the body fat percentage of either pure species. Additionally, mexlA mice consume more calories than control siblings, suggesting mexlA increases body fat, at least partly, by regulating food intake. We transcriptionally profiled the hypothalamus—the master brain region for appetite control—and identified Bex1 as the top candidate gene within the mexlA interval driving overgrowth. Our results also indicate that mexlA effects are independent of the genomic imprinting disruptions that have been shown to contribute to hybrid incompatibilities in these species and other mammalian taxa. Together, these results reveal the phenotypic, genetic, and molecular basis of a large X-effect growth incompatibility, with consequences for the biomedical understanding of obesity. Due to frequent omission of chromosome X from genome-wide association studies, only nine X-linked variants associated with obesity have been identified in humans—none of which overlap mexlA. Our study of deer mice thus uncovers novel X-linked genetic loci and molecular mechanisms underlying obesity and driving hybrid incompatibility through imprinting-independent effects in a naturally diverged mammalian taxa.

Inbreeding Drives Systematic Shifts in Mouse Genome Variation and Structure

Beth L. Dumont1,2,3, Sam Bigalke1, Michael W. Nachman4, and Jaroslav Piálek5

1. The Jackson Laboratory, Bar Harbor, Maine, USA

2. Graduate School of Biomedical Sciences, Tufts University, Boston, Massachusetts, USA

3. Graduate School of Biomedical Science and Engineering, The University of Maine, Orono, Maine, USA

4. Department of Integrative Biology, Museum of Vertebrate Zoology, and Center for Computational Biology, University of California, Berkeley, Berkeley, California, USA

5. Studenec Research Facility, Institute of Vertebrate Biology, Czech Academy of Sciences, Brno, Czech Republic

Inbred laboratory mouse strains have been at the forefront of biomedical discovery for nearly a century. Standardized inbred strain backgrounds enable testing of genetically identical cohorts across different experimental perturbations and offer reproducible platforms for mechanistic investigations into gene and pathway function. Despite these strengths, inbred strains fail to fully model the genetic complexity of human biology. Notably, inbreeding enforces homozygosity at loci across the genome, such that recessive variants associated with lethality, infertility, or severe reductions in fitness are selectively purged from the genome. As a result, the distribution of fitness effects in inbred genomes is likely distorted compared to natural organisms, where heterozygosity can buffer the effects of deleterious variants to allow their persistence. While these impacts on inbred strain genomes are tacitly recognized, the dynamics of genome evolution during inbreeding have not, to our knowledge, been systematically evaluated using modern sequencing.

To address this gap, we generated long-read PacBio HiFi whole-genome sequences from six “wild-inbred mouse triad”. Each triad is comprised of two wild-caught house mice that were brother-sister mated for >15 generations to derive an inbred strain. We uncover equal founder haplotype contributions to the autosomal genome of each inbred strain, ruling out founder dominance and providing a baseline for interpreting allele fates under inbreeding. As expected, we observe a consistent depletion of predicted deleterious alleles in inbred strains relative to their wild founders. The degree of purging evident in each inbred strain is variable across triads, likely reflecting the unique demographic history and deleterious allele burden of the different source populations from which the wild-caught founders of each triad derive. Additionally, preliminary analyses of repeat content uncover a systematic expansion of satellite DNA fractions during inbreeding, with especially dramatic expansions observed in centromeric satellite domains. These observations suggest that inbreeding provides an opportune setting for the structural remodeling of centromere architecture, a phenomenon potentially mediated by selfish centromere drive.

Taken together, our work provides unprecedented insight into the genomic consequences of inbreeding, including its effects on the distribution of variant effects and genome architecture. Our results also underscore the value of wild-caught genomes as more faithful models of human genomic complexity, positioning wild mice as essential complements to inbred strains in genetic and biomedical research.

Allelic Variants of MTOR Associated with Cancer Susceptibility and Radiation Sensitivity

Shuling Zhang1, Wendy DuBois1, Rand Gabriel M. Buenaventura1, Emily Xu1, Alexandra Mora1, Joe T. Nguyen2, Aleksandra Michalowska1, and Beverly A. Mock1

1Laboratory of Cancer Biology and Genetics1 and Surgical Oncology Program2, CCR/NCI, NIH, Bethesda, MD 20892

MTOR allelic variants (R/C) at amino acid 628 are linked to plasmacytoma (PCT) susceptibility in BALB/cAn and NZB/BINJ (628C) mice and PCT resistance in most inbred strains (R628), including C57BL/6 (B6). The 628C allele exhibits lower activity in in vitro kinase assays compared to the R628 common/wild-type (WT) allele. After 8 Gy irradiation, male 628C knock-in (KI) mice (backcrossed to N15 on C57BL/6 background) showed reduced survival rates compared to male WT littermates. Mouse embryonic fibroblast cells from 628C mice exhibited decreased sensitivity to rapamycin and irradiation-induced growth inhibition compared to those from WT (R628 allele) mice. We developed doxycycline inducible (TET-on) HEK293Trex cell lines stably expressing Flag-tagged R628 or 628C MTOR alleles; the 628C cells showed reduced G2 cell cycle arrest after irradiation compared to R628-expressing cells. Immunoprecipitation Western blots revealed stronger RPTOR binding to R628 than to 628C, both under doxycycline induction and irradiation stress. Concomitantly, 628C cells, had lower phosphorylation levels of mTOR substrates EIF4EBP1 and RPS6KB1. RNA-seq analysis revealed global gene expression differences associated with MTOR alleles and treatment conditions. Differential expression analysis identified an irradiation (IR)-responsive gene signature dependent on the Mtor genotype. Pathway analysis showed significant enrichment of DNA damage repair and replication between the two alleles. The G2/M checkpoint pathway was upregulated in R628 cells and downregulated in 628C cells. Ongoing efforts include extensive pathway analyses of DNA damage, response, replication and repair affected by the two alleles to understand effects on both upstream and downstream signaling pathways.

Integrative statistical modeling of genetic regulation and immune mediation of coronavirus disease severity

Ellen Risemberg1,2, Sarah R Leist3, Alexandra Schäfer3, Kalika Kamat1,2, Timothy A Bell1, Pablo Hock1, Colton L Linnertz1, Darla R Miller1, Ginger D Shaw1,4, Fernando Pardo Manuel de Villena1,4, Ralph Baric3,4,5, Martin T Ferris1, William Valdar1,4

1Department of Genetics, UNC Chapel Hill

2Curriculum in Bioinformatics & Computational Biology, UNC Chapel Hill

3Department of Epidemiology, UNC Chapel Hill

4Lineberger Comprehensive Cancer Center, UNC Chapel Hill

5Department of Microbiology & Immunology, UNC Chapel Hill

Zoonotic coronaviruses have caused three severe epidemics in the 21st century, including SARS and COVID-19, while climate change and increasing human-animal interaction raise the likelihood of future outbreaks. The enormous public health burden imposed by coronaviruses motivates improved understanding of viral pathogenesis and mechanisms of susceptibility to severe disease. Several studies have found evidence for genetic regulation of coronavirus disease severity1,2, and it is known that critical SARS and COVID-19 are caused, in part, by a pathological immune response and subsequent damage to the lungs3,4. Given the evidence that host genetics modulates the immune system5, it is likely that this immunopathology is genetically driven. Here, we integrate genetic, immune, and disease severity data to study the genetic basis of immune-driven severe coronavirus disease. Bayesian variable selection reveals immune traits that are predictive of disease severity; infection-stratified and combined genotype-by-treatment QTL mapping reveal extensive genetic regulation of the immune system at homeostasis and in response to infection; and mediation analysis with Bayesian model selection supports causal pathways linking genetic loci, immune responses and disease severity. Overall, we observe an immune response to coronavirus disease that largely recapitulates that observed in humans, underscoring the relevance of mouse models for infectious disease research. We further identify genetic regulation of this immune response, highlighting loci that may contain therapeutic targets.

Citations:

1. Dos Santos ACM, Dos Santos BRC, Dos Santos BB, De Moura EL, Ferreira JM, Dos Santos LKC, et al. Genetic polymorphisms as multi-biomarkers in severe acute respiratory syndrome (SARS) by coronavirus infection: A systematic review of candidate gene association studies. Infection, Genetics and Evolution. 2021 Sept;93:104846.

2. Cappadona C, Rimoldi V, Paraboschi EM, Asselta R. Genetic susceptibility to severe COVID-19. Infection, Genetics and Evolution. 2023 June;110:105426.

3. Cameron MJ, Bermejo-Martin JF, Danesh A, Muller MP, Kelvin DJ. Human immunopathogenesis of severe acute respiratory syndrome (SARS). Virus Research. 2008 Apr;133(1):13–9.

4. Yang OO. The immunopathogenesis of SARS-CoV-2 infection: Overview of lessons learned in the first 5 years. The Journal of Immunology. 2025 June 1;214(6):1095–104.

5. Brodin P, Davis MM. Human immune system variation. Nat Rev Immunol. 2017 Jan;17(1):21–9.

Mechanisms of Syk gain-of-function inflammatory disease

Henry Taylor1, Claire Pearson1, Holm Uhlig2, Fiona Powrie1

1Kennedy Institute of Rheumatology, University of Oxford, U.K.

2Centre for Human Genetics, University of Oxford, U.K.

Spleen tyrosine kinase (Syk) is a signalling intermediate that is highly expressed in innate immune cells such as macrophages and neutrophils, as well as in B lymphocytes, where it has critical roles downstream of multiple immunoreceptors including pattern recognition receptors and the B cell receptor. Human gain-of-function (GOF) variants in Syk have previously been shown to cause severe immune dysregulation, leading to very early onset inflammatory bowel disease, multi-system inflammation including arthritis, and humoral immunodeficiency. A SykS544Y GOF mouse model of the disease recapitulates many of the human disease phenotypes. Studying such monogenic immune disorders may provide valuable insights into mechanisms that drive inflammation both within monogenic and polygenic disease.

We found that Syk GOF causes a complex B cell immunodeficiency that starts within the bone marrow and is more profound in the periphery. These abnormalities are associated with impaired immunity to intestinal Citrobacter rodentium infection. Single-cell analysis of intestinal B cells from a SYK GOF patient revealed conserved features between the human and murine phenotypes.

Syk GOF mice also develop intestinal inflammation following colonisation with Helicobacter hepaticus, a gram-negative pathobiont normally tolerated by the mucosal immune system. Inflammation is likely to be driven by innate cells since Rag-deficient Syk GOF mice also develop disease. Both Rag-sufficient and deficient mice also develop a spontaneous spondyloarthropathy which shares clinical, and histological features with human axial spondyloarthritis, and develops in the absence of microbes. A shared transcriptional signature between colitis and arthritis was characterised by upregulation of genes associated with pattern recognition receptor signalling and pro-inflammatory cytokines.

Our findings deepen our understanding of SYK as a key regulator of innate immune responses, suggest broader relevance of SYK signalling pathways in the pathogenesis of IBD and spondyloarthropathy, and support the potential of SYK as a therapeutic target in inflammatory diseases.

Jordan Genome Project: characterizing baseline genetic variation in 89 healthy Jordanians

Tawfiq Froukh

Dept. of Biotechnology and Genetic Engineering, Philadelphia University, Amman-Jordan

Background: Whole is exome sequencing (WES) is widely used to characterize population-specific genetic variation to identify rare variants that may have an influence on rare and common diseases. Many rare genetic variants were identified in worldwide populations, but they are underrepresented in the Middle Eastern populations with few exceptions such as the Qatari population. This study is considered the first in Jordan to establish baseline genetic variation in healthy cohort which pave the road for future clinical interpretation and population genetics research.

Methods: WES from 89 healthy unrelated Jordanians were extracted from the Jordan Genome Project (JGP) database in Philadelphia University. Variants were called and annotated using the standardized bioinformatics pipeline used in the JGP which include the tools: Bcftools and snpEff. In order to identify novel and population specific genetic variant, the identified variants were annotated and compared with the variants in the following public databases: dbsnp157, ClinVar, gnomAd, 1000 genome project, TopMed, ALFA, Regeneron, Qatari and Turkish Genome Project.

Results:
-Total Variants: A total of 758,118 variants were identified across the 89 exomes, with an average of 8,518 variants per individual.

-Novel variants: 12,973 novel variants were identified; not previously reported in public databases.

-Pathogenic variants: Screening against ClinVar identified 119 pathogenic/likely pathogenic variants in 104 genes. While 113 disease associated variants were identified in a heterozygous state, 6 variants were found in a homozygous state in 19 individuals. Notable homozygous variants are recorded to be in association with the Duffy blood group system and G6PD deficiency. Other homozygous variants were linked to antithrombin Pittsburgh-related hemorrhagic diathesis, distal hereditary motor neuronopathy, and acute febrile neutrophilic dermatosis.

-Population structure: Principal Component analysis (PCA) identified two distinct genetic clusters which indicates a degree of genetic heterogeneity or admixture within the sampled population.

Conclusion:

This study successfully established a foundational baseline of genetic variation for Jordan which fulfill the need for underrepresented Middle Eastern populations in global genetic databases. Identifying thousands of novel and population-specific variants, along with 119 pathogenic variants, providing a critical resource for local clinical interpretation and diagnostics. The established baseline, coupled with the observed genetic structure (admixture), paves the way for future large-scale population genetics research and enables more accurate, regionally appropriate genetic counseling within Jordan.

Genetic variants ARMS2 rs10490924G/T, VEGF rs699947A/C and COL10A1 rs1064583A/G influence susceptibility to Age-related macular degeneration

Akshay Ananda, Priya Battua,b, Manjari Raina, Kaushal Sharmaa, Suresh Sharmac, Ramandeep Singhd

aNeuroscience Research Lab, Department of Neurology, Post Graduate Institute of Medical Education and Research, Chandigarh

bIndian Institute of Science Education and Research, Mohali, Punjab

c Department of Biostatistics, Panjab University, Chandigarh

d Advanced Eye Centre, Post Graduate Institute of Medical Education and Research, Chandigarh

Introduction: Age-related macular degeneration (AMD) is a multifactorial eye disease that typically develops with old age, leading to central vision loss in individuals who are ≥50 years of age. A comprehensive understanding of the genetic, environmental and molecular basis can help translate research findings into clinical practice. The study investigated the association of 6 single nucleotide polymorphisms (SNPs) with AMD risk.

Methods: The study included 178 AMD and 84 controls. TaqMan SNP genotyping assay was used for genotyping SNPs. Bivariate logistic regression was performed to analyse the risk association. Multifactor Dimensionality of Reduction (MDR) analysis was used to estimate epistatic interactions between the SNPs. The statistical modeling was performed for generating an equation to predict AMD risk.

Results: ARMS2 rs10490924G/T, VEGF rs699947A/C and COL10A1 rs1064583A/G were significantly associated with AMD. The ARMS2 rs10490924T (p<0.0001; Odd’s ratio (OR) =2.307) and VEGF rs699947C (p=0.026; OR=1.592) associated with AMD risk, whereas the COL10A1 rs1064583G (p=0.012; OR=0.588) conferred protection for AMD. MDR revealed that two-locus model between ARMS2 rs10490924G/T and COL10A1 rs1064583A/G was the best model for AMD risk prediction (p<0.0001; testing accuracy = 0.639; cross-validation consistency = 10/10). Statistical modeling revealed that ARMS2 and VEGF could be good predictors of AMD with area under ROC curve of 69.2%.

Conclusion: ARMS2 rs10490924G/T and VEGF rs699947A/C had risk enhancing effect while COL10A1 rs1064583A/G had protective effect for AMD.

Beyond Lifespan: Anti-Aging Interventions to Unlock Cognitive Longevity

Shannon J. Moore1,2, Geoffrey G. Murphy1

1Molecular. & Integrative Physiology, 2Neurology, Univ. of Michigan, Ann Arbor, MI

Aging is associated with declines in both physical health and cognitive function, and is the greatest risk factor for developing neurodegenerative diseases, like Alzheimer’s disease (AD). The geroscience hypothesis suggests that successfully slowing the aging process could broadly and concurrently improve multiple age-related deficits; thus, the Interventions Testing Program seeks to identify “anti-aging” interventions, using lifespan extension as a proxy for delayed aging. A number of lifespan-extending interventions have been identified, with subsequent studies showing their efficacy to ameliorate several physiological aspects of aging; however, their impact on cognitive function - a crucial aspect of aging - has remained largely unexamined.

To address this gap in knowledge, we investigated whether a successful lifespan-extending drug, acarbose, was also effective in improving cognitive function in a genetically heterogenous mouse model of “normal aging” (using HET3 mice) and AD (by incorporating the 5xFAD transgene onto the HET3 background; AD-HET3). Using the Morris water maze, learning and memory were assessed at three timepoints (young adult, middle age, and aged) in individual cohorts of mice treated with acarbose (1000 ppm in chow, started at 4 mo of age) or maintained on standard “control” chow. We found that acarbose effectively promoted cognitive function in normal aging but failed to ameliorate cognitive impairments that emerge with age in AD-HET3 mice. These results suggest that there are at least partially overlapping mechanisms that mediate lifespan and cognitive function, but that additional, distinct mechanisms contribute to the unique pathological processes underling AD-specific cognitive deficits.

Novel genetic contributions to congenital craniofacial and brain anomalies

Paul Iyyanar1, Iftekhar Showpnil1, Mike Lape2, Logan Willeke1, Katarina Dunn1, Russell Chuah1, Jesus Leal1, Matthew Weirauch2, Rolf Stottmann1

1Nationwide Childrens Hospital

2Cincinnati Children's Hospital Medical Center

We are far from understanding the complete set of genes guiding craniofacial and brain development and contributing to congenital anomalies. We will summarize our recent efforts to identify new genetic variants contributing to congenital craniofacial and brain anomalies using complementary approaches in human and mouse. First, we will review a multi-institutional effort to perform trio sequencing and traditional variant analysis on over 200 families with congenital anomalies to identify novel genetic contributions. We will also highlight a new method we have developed to identify novel non-coding regulatory variants that also contribute to disease. One of the most promising candidates from this new approach has been validated in a mouse model. Secondly, we will report on a complementary approach using the power of unbiased forward genetics in the mouse. Building on our previous success with ENU mutagenesis, we have started a new screen for gross morphological phenotypes at E18.5 in the mouse. We have also included the Patched1-lacZ allele to genetically sensitize the screen. In this new screen, we have analyzed 85 lines in the past year and recovered 27 craniofacial alleles. This is a much more robust yield than any screen we have previously performed or have seen in the literature. All of these alleles from the screen have been inherited into future generations showing high penetrance and suggesting they are almost certainly monogenic traits. While genome sequencing and data analysis are ongoing, we have already identified strong candidates for causal variants in 16 of these 27 craniofacial phenotype lines to date. Several of these genes are completely unknown to the field of craniofacial genetics. Together, these approaches show that unbiased gene discovery efforts are still quite valuable and contribute to our understanding of craniofacial development and disease.

CRISPR knockout screens reveal genes and pathways essential for neuronal differentiation and implicate PEDS1 in neurodevelopment

Stephan C Collins1,2, # , Alana Amelan3, # , Nadirah S Damseh4, # , Nanako Hamada5 , Ahd Salim3 , Elad Dvir3, Galya Monderer-Rothkoff 3, Tamar Harel6, 7, Koh-Ichi Nagata5, 8, Sagiv Shifman3, # and Binnaz Yalcin1,2, #

1INSERM Unit 1231, Université de Bourgogne-Europe, Dijon, France

2Institut NeuroMyoGène, Unité Physiopathologie et Génétique du Neurone et du Muscle, CNRS UMR 5261, Inserm U1315, Université Claude Bernard Lyon 1, Lyon, France

3Department of Genetics, The Alexander Silberman Institute of Life Sciences, The Hebrew University of Jerusalem, Jerusalem, Israel

4Department of Pediatrics & Genetics, Makassed Hospital & Al-Quds Medical School, East Jerusalem, Palestine 5Department of Molecular Neurobiology, Institute for Developmental Research, Aichi Developmental Disability Center, Kasugai, Japan

6Department of Genetics, Hadassah Medical Center, Jerusalem, Israel

7Faculty of Medicine, The Hebrew University of Jerusalem, Jerusalem, Israel

8Department of Neurochemistry, Nagoya University Graduate School of Medicine, Nagoya, Japan

#Contributed equally

Neurodevelopmental disorders (NDDs) arise from disruptions in brain development, yet the underlying pathways remain incompletely understood. Here we demonstrate that genome-wide CRISPR knockout screens in mouse embryonic stem cells differentiating into neural lineages identify hundreds of essential genes, only a minority of which are currently implicated in NDDs. Dominant NDD genes were enriched for transcriptional regulators, whereas recessive NDD genes were predominantly involved in metabolic processes. Mouse models for eight genes (Eml1, Dusp26, Dynlrb2, Mta3, Peds1, Sgms1, Slitrk4 and Vamp3) revealed marked neuroanatomical abnormalities, including microcephaly in half of the cases. Focusing on PEDS1, a key enzyme in plasmalogen biosynthesis, we identified a bi-allelic variant in individuals with microcephaly, global developmental delay and congenital cataracts. In mice, Peds1 deficiency led to accelerated cell-cycle exit and impaired neuronal differentiation and migration. These pathways required for neural differentiation provide a genetic framework for discovering additional NDD genes.

The Mary Lyon Lecture

Host genetic diversity in action: determinants of outcome and spread in vector‑borne viral diseases

Caroline Manet

Institut Pasteur, Université Paris Cité, Mouse Genetics, Immunity and Infections Laboratory, 75015 Paris, France

Every infection tells a different story. Even when exposed to the same virus, individuals can experience radically different trajectories: some develop severe illness, while others experience only mild symptoms or silent infection. In my team, we see this diversity not as experimental noise but as a source of biological insight. By leveraging mammalian genetic diversity, we aim to uncover how host variation shapes the entire spectrum of vector‑borne viral diseases, from individual pathology to transmission potential. The public‑health questions at stake are clear: which genetic factors determine severe outcomes, or drive viral transmission, even from moderate or asymptomatic infections? Addressing these questions requires accounting for genetic diversity, a key dimension that remains rarely accessible through experimental, mechanistic models. Yet understanding these complex, multiscale phenotypes demands precisely such controlled systems. Using the Collaborative Cross mouse panel, we have established genetically diverse and tractable infection models that mirror the variability of outcomes seen in humans. Through quantitative trait mapping and functional investigations, we have identified loci influencing Zika virus (ZIKV) infection outcome and are validating candidate genes using reverse genetics and congenic strains. These tools now enable a mechanistic exploration of host-virus-vector interactions across genetically diverse backgrounds. In our most recent projects, we apply this framework to flaviviruses of emerging relevance, ZIKV and tick‑borne encephalitis virus (TBEV), to study disease severity and transmission. One line of work examines how host genetic background modulates neuroinvasion, inflammation, and neurological sequelae. Another will investigate how genetically driven immune, metabolic, and physiological differences modulate ZIKV transmission to mosquitoes and TBEV milk-borne transmission from mother to offspring. We are also extending this work to a tripartite design that integrates host, vector, and viral genetic diversity to decipher how their combined variation drives arboviral outcomes. Beyond identifying the host genetic determinants that drive infection severity and transmission, our ultimate aim is to unravel actionable mechanisms they engage, and integrate these insights into a broader framework linking hosts, vectors, and viruses to inform control strategies.

An improved version of an artificial intron cassette for rapid and efficient generation of conditional alleles

Marie-Christine Birling1, Laurence Schaeffer1, Philippe André1, Pascale Mercier3, Frédéric Fiore4, Yann Hérault1,2

1CELPHEDIA, PHEN-ICS, Institut Clinique de la Souris, PHENOMIN, CNRS, INSERM U964, Université de Strasbourg, 1 rue Laurent Fries BP 10142 Parc d’Innovation 67404 Illkirch, France

2Institut de Génétique Biologie Moléculaire et Cellulaire (IGBMC), CNRS, INSERM, Université de Strasbourg, UMR7104, UMR964, Illkirch, France

3Institut de Pharmacologie et de Biologie Structurale (IPBS), CNRS UMR5089, Toulouse, France

4Centre d’Immunophénomique (CIPHE), Aix Marseille Université, INSERM, CNRS, Marseille, France

Rapid generation of conditional loss-of-function mouse models (cKOs) can be achieved by inserting a short artificial intron into a target exon using CRISPR/Cas9 electroporation. This single-guide strategy greatly simplifies the process compared to traditional approaches that require two guide RNAs and dual insertion sites. The method, known as DECAI (Degradation based on Cre-regulated Artificial Intron) or eSPLIT (exon split), supports normal gene expression in the absence of Cre recombinase because the inserted cassette is recognized as an intron and efficiently removed by the splicing machinery. Upon Cre-mediated recombination, excision of key intronic elements disrupts normal splicing and leaves residual sequences carrying stop codons in all reading frames, resulting in premature translation termination and gene inactivation.

We systematically evaluated an optimized version of this artificial intron across genes with diverse structural features to define its versatility and limitations. The tested models included Ufsp1 and Gjb2 (single-exon coding genes), Nr4a3 (a large exon), Eng, Mbnl1, and Jam3 (small exons), Il33 (multiple small exons in the same reading frame), and Vma21 (an X-linked pseudogene with conserved introns and small exons). These analyses confirmed the efficiency of the improved intron design and enabled the derivation of practical guidelines to enhance editing efficiency while minimizing the risk of unintended knockout alleles caused by aberrant splicing. Although not all exons are suitable for artificial intron insertion, our results demonstrate that exons of sufficient size with unique sequences—lacking homology to gene family members or pseudogenes—can be reliably targeted through appropriate design to generate conditional knockout mouse models in a rapid and cost-effective manner. Importantly, the same design principles are directly transferable to rat models and to genome engineering in cultured cell lines, supporting the broad translational applicability of this strategy across mammalian experimental systems.

Characterization of Oog1-Cre BAC Transgenic Mice: Efficient Maternal Cre Recombination and Transgene Integration Mapping

Monica Meng-Chun Shih1, Ting-Shuan Kung1, Hsien-Pin Chiu1, Chien-Hsing, Lin2, Po-Hao Cheng1, Si-Tse Jiang1

1National Center for Biomodels, National Institutes of Applied Research, Taiwan

2Phalanx Biotech Group, Hsinchu, Taiwan

The generation of whole-body knockout mice from floxed alleles typically requires an additional generation of breeding to segregate Cre transgenes, thereby increasing time, cost, and animal usage. Oogenesin 1 (Oog1) is a maternally derived gene that influences early embryonic development through proteins expressed in oocytes, with expression spanning from oogenesis through early embryogenesis.

Taking advantage of this maternal expression pattern, we established an Oog1-Cre bacterial artificial chromosome (BAC) transgenic mouse line that mediates highly efficient gene excision via maternally deposited Cre recombinase during early embryogenesis. Here, we present a comprehensive genomic and functional characterization to validate this model as a reliable and efficient genetic tool. PacBio long-read sequencing was employed to precisely map transgene integration sites within the host genome, ensuring transgene integrity and copy number accuracy, and thereby establishing a robust framework for quality control and reproducibility.

Recombination efficiency was evaluated using two independent floxed mouse lines, Slitrk1 and Hk2. In the Slitrk1 model, offspring derived from Oog1-Cre females showed complete conversion of the floxed allele to a null allele. Importantly, this recombination persisted in progeny lacking the Cre transgene, demonstrating a strong and reliable maternal Cre effect. Comparable high-efficiency excision was also observed in the Hk2 floxed line.

Systemic recombination was confirmed by PCR analysis across 15 major tissues, revealing uniform whole-body excision of floxed alleles. Collectively, these findings establish C57BL/6-Tg(Oog1-Cre; CAGGS-flEmGFP-TyrmiR)19Narl/Narl as a ubiquitous Cre strain that enables the rapid generation of global knockout mice in a single generation. Compared with conventional breeding strategies, this approach accelerates breeding timelines by approximately three months and reduces animal usage by ~30%, directly supporting the Reduction principle of the 3Rs and enhancing overall animal welfare.

Using Bxb1 and Pa01 integrases to re-engineer and improve the hUBC-Creert2 transgenic line

Lydia Teboul , Rosie Bunton-Stasyshyn, Adam Caulder, Skevoulla Christou-Smith, James Cleak, Gemma F. Codner, Sarah Dowding, Alex Fower, Wendy Gardiner, Christy J. Greenwood, Jorik Loeffler, Matthew Mackenzie, Edward Ward, Michelle E. Stewart, Sara Wells

The Mary Lyon Centre, MRC Harwell, Harwell Campus, Didcot, Oxon, OX11 0RD, UK.

l.teboul@har.mrc.ac.uk

The B6.Cg-Ndor1Tg(UBC-cre/ERT2)1Ejb transgene (IMSR_JAX:008085) was obtained by random integration and is an efficient allele to support widespread tamoxifen-induced cre deletion, including in challenging tissues such as the brain. This infers that the transgene landed in a locus that permits high levels of widespread expression. Two drawbacks of the model are that it was generated in embryos of mixed genetic 129SvEv;C57BL/6 genetic background and that it interrupts the Ndor1 gene that is homozygous lethal when its function is affected. We have sequenced the original insertion and reproduced the transgene. We have introduced in the same locus and next to the Ndor1 coding sequence a landing pad for the Bxb1 integrase in C57BL/6J and C57BL/6N mice. We have used Bxb1 to integrate the UBC-Creert2 transgene in these inbred genetic backgrounds. We showed that the new alleles within and next to Ndor1 have similar activity to the original transgene and that an integration outside of Ndor1 is homozygous viable. We also integrated a landing pad for Pa01 next to the Col1a1 gene and again obtained integrations of the transgene. This is the first illustration of the use of the Pa01 integrase for the integration of transgenes in embryos. We will present the new lines we have generated and how they were validated at the genomic and activity levels. We will discuss the advantages and challenges associated with using integrases for transgene integration.

Expanding GWAS/QTL Resources at the Rat Genome Database (RGD)

Stanley J F Laulederkind1, Logan Lamers1, Shur-Jen Wang1, Carissa A Park1, Mary L Kaldunski1, Melissa Eilbes1, Marek Tutaj1, Jennifer R Smith1, Jeff De Pons1, Melinda R Dwinell1, Anne E Kwitek1 and the RGD team1

1 Rat Genome Database, Department of Physiology, Medical College of Wisconsin, Milwaukee, WI, USA

For over 25 years, the Rat Genome Database (RGD) has curated and housed comprehensive data including gene associations with diseases, pathways, molecular functions, biological processes, cellular components, chemical interactions, phenotypes, quantitative trait loci (QTL), and strains for the laboratory rat. Because the rat is predominantly studied as a human disease model, RGD also curates, ingests, and integrates complementary information for humans, mice, and several other mammalian disease models. To further integrate genetic risk data into RGD, we recently incorporated data from the NHGRI-EBI Catalog of human genome wide association studies (GWAS; https://www.ebi.ac.uk/gwas/). The human GWAS catalog leverages the Experimental Factor Ontology (EFO) to describe and standardize the associated traits. To integrate these data into RGD, we leveraged automated and manual curation practices to map the EFO terms to multiple ontologies used at RGD. EFO terms were mapped to RDO (RGD Disease Ontology) terms, HP (Human Phenotype) and MP (Mammalian Phenotype) ontology terms, CMO (Clinical Measurement Ontology) terms, and VT (Vertebrate Trait ontology) terms. Standardized mapping files are posted at https://download.rgd.mcw.edu/ontology/mappings/. We also created a user search for human GWAS_QTL from the traits or SNP/rs IDs to link out to comprehensive report pages that include the various trait annotations, references, associated variants and other genome features, and external links. Given this standardized infrastructure, we are integrating complementary GWAS data for the rat, which provides a platform for additional comparative genetic and genomic studies across the species in RGD.

Sex-specific phenotypes and their response to neonatal Dyrk1a reduction in the Ts65Dn Down syndrome mouse model

Alyssa Duerst1, Laura Hawley1, Linnea Johnson1, Drew Folz1, Zaina Obeid1, Laura Snellenberger1, Charles R. Goodlett2, and Randall J. Roper1

1Department of Biology, Indiana University Indianapolis, IN, USA

2Department of Psychology, Indiana University Indianapolis, IN, USA

Children with Down syndrome (DS) experience delays in cognitive, physical, and motor development. Overexpression of Dual-specificity tyrosine phosphorylation-regulated kinase 1A (DYRK1A), a gene located on human chromosome 21 (Hsa21) and triplicated in individuals with Trisomy 21 (Ts21), is thought to be a major contributor to some neurodevelopmental delays associated with DS and is a candidate for targeted intervention therapies to improve neurodevelopmental phenotypes. Male and female Ts65Dn DS model pups are trisomic for ~100 Hsa21 orthologs including Dyrk1a and have significant DYRK1A protein overexpression on postnatal day 6 (P6) in the hippocampus, cerebral cortex, and cerebellum. We hypothesized that early postnatal normalization of Dyrk1a copy number, prior to DYRK1A overexpression, would diminish molecular, physical, and behavioral outcomes in the Ts65Dn mice thus providing a standard for comparison to measure the success of interventions targeting Dyrk1a. Dyrk1a was reduced from three to two copies in otherwise trisomic Ts65Dn pups at P0. At P3-P21, Ts65Dn pups showed physical, motor, and behavioral deficits as compared to euploid pups, often differing between the sexes. Dyrk1a normalization in Ts65Dn pups, improved only few sex-specific outcomes, perhaps because of limited correlations between Dyrk1a copy number and the developmental regulation of RNA transcripts and DYRK1A protein levels in the brains of these animals. Improvements in DS related Ts65Dn behavioral phenotypes that may emerge later in life cannot be ruled out. Additionally, combinations of different overexpressed protein targets in DS or an earlier Dyrk1a reduction may be more effective at normalizing perinatal neurobehavioral deficits.

A Novel Mouse NPC Differentiation Protocol Enables Parallel CRISPR Screening

Katrín Möller1, Afrooz Razi2, Juan Ouyang1, Stefán Pétursson1, Kasper D Hansen2,3,4, Hans T Björnsson1,4,5

1Louma G. Laboratory of Epigenetic Research, Faculty of Medicine, University of Iceland; Reykjavík, Iceland.

2Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD USA.

3Department of Bioengineering, Johns Hopkins University School of Medicine, Baltimore, MD, USA.

4Department of Genetic Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA.

5Department of Genetics and Molecular Medicine, Landspitali University Hospital, Reykjavik, Iceland.

Primary mouse neuronal progenitor cells (NPCs) are a powerful tool for studying neurodevelopment, lineage specification, and the roles of disease-associated genes. While protocols for deriving both neurons and glial cells from iPS cells are abundant, reliable methods for differentiating primary mouse NPCs remain underdeveloped. This lack of standardized and reproducible protocols has hindered the use of mouse NPCs in comparative neurodevelopmental studies.

To address this, we developed a novel differentiation protocol that has defined transitions between growth, induction, differentiation, and maturation phases, involves standardized media and handling procedures and yields reproducible cultures of mixed neuronal and astrocyte compositions. Besides its utility as a baseline differentiation method, the protocol is particularly useful when combined with CRISPR-based gene knockout strategies.

We have used this model to perform a parallel CRISPR screen on 60 different epigenetic machinery (EM) genes and 10 scrambled controls, to study the effects on gene expression, lineage composition and differentiation trajectories using single-cell RNA sequencing. We found that many EM knockouts (KOs) show convergent gene expression changes, and that the majority have a significant enrichment of differentially expressed Autism and Schizophrenia risk genes (OR>1, FDR<0.05). Several EM KOs further show changes in cell-type compositions, for example Crebbp-KOs are biased towards immature neurons after 12 days of differentiation, and Setd1b-KOs have more astrocytes, compared to controls.

By combining reproducible primary mouse NPC differentiation with parallel CRISPR perturbation, we provide a scalable framework for mapping how epigenetic machinery shapes neurodevelopment. Our findings show that multiple EM knockouts have shared transcriptional abnormalities and a disruption of key risk genes for other more common neurological disorders, including autism and schizophrenia. This suggests the possibility of shared therapeutic development for EM-related and other neurodevelopmental disorders.

THE CORRECT ALLOCATION AND DIFFERENTIATION OF ENDODERM POPULATIONS DURING GASTRULATION IS A CRITICAL PRECURSOR OF HEART FORMATION

Ruth M. Arkell1, Kristen S. Barratt1, Koula E. M. Diamand1, Di Pan3, Kelsey Walsh1, Alaa Alzahrani1, Kathryn Dickson1, Josie Carberry2, Maja Adamska3

1 The John Curtin School of Medical Research, The Australian National University, Canberra, Australia

2 Department of Mechanical and Aerospace Engineering, Monash University, Clayton, VIC, Australia

3 Research School of Biological Sciences, The Australian National University, Canberra, Australia

The importance of WNT target gene repression during embryonic lineage allocation and axis formation is reflected in the phenotypes associated with the loss-of-function of negative canonical WNT regulators. Mouse embryos in which WNT target gene expression is inappropriately de-repressed have forebrain truncations, and abnormalities of somites and the midline mesoderm. These phenotypes are found in embryos that lack the Tcf7l1 transcription factor (the mediator of WNT signals at this stage of development) indicating its main function at gastrulation is to repress transcription. Here we show that loss-of-function of another transcription factor, Zic3, phenocopies Tcf7l1 loss, suggesting that Zic3 also represses WNT-dependent transcription during gastrulation. Mutation of ZIC3 has long been associated with heterotaxy in man and mouse; a congenital defect that occurs when left-right (L-R) axis establishment early in embryogenesis is perturbed. We have explored the relationship between Zic3, WNT signalling, lineage allocation and L-R axis formation. We find that Zic3 mutation causes elevated TCF-dependent transcription at gastrulation driving increased activity of the canonical WNT signalling pathway. Using single cell RNA sequencing of normal and Zic3 null embryos we show that Zic3 controls the allocation of cells to the mesendoderm and endoderm lineages and that cell identity and polarity is disrupted in specific anterior mesoderm and endoderm populations. This leads to aberrant movement of the endoderm which is normally required for the morphogenesis of the node, a transient structure responsible for conferring the L-R axis. This impedes the induction of left-sided NODAL signalling leading to abnormal heart development.

Trans-regulation of heterochromatin underlies genetic variation in 3D genome contacts

Haley J Fortin1,2, Anna Z Struba1, Christopher L Baker1,2*

1The Jackson Laboratory, Bar Harbor, ME 04609, USA

2Graduate School of Biomedical Sciences, Tufts University, Boston, MA 02111, USA

Genetic variation drives phenotypic diversity and disease susceptibility. Trans-acting genetic variation coordinates genome-wide chromatin changes, yet the molecular mechanisms underlying this regulation remain unclear. Here, we use the power of mouse genetics to investigate how genetic variation at trans-acting loci regulates 3D chromatin interactions. Using HiChIP to map H3K27ac-associated regulatory elements in C57BL/6J and DBA/2J embryonic stem cells (ESCs), we identified 4,962 strain-differential interactions, 71% of which overlapped chromatin accessibility quantitative trait loci (QTL), establishing chromatin interaction variation is predominantly heritable. These differential interactions showed coordinated changes in chromatin state and gene expression, with stronger interactions associated with increased accessibility and transcription. Notably, loci regulated in trans exhibited a unique chromatin signature where weaker interactions were enriched for H3K9me3-marked heterochromatin. Analysis of F1 hybrids revealed dominant repressive effects, consistent with heterochromatin-mediated trans-regulation. To causally test this mechanism, we generated reciprocal congenic mouse strains carrying a Chr13 trans-QTL region. Integrated multiomic profiling of congenic ESC lines demonstrated that this single locus coordinates changes in H3K9me3, H3K27ac, chromatin accessibility, and 3D contact frequency at hundreds of distal targets. Consistent with our expectations, 73-83% of differential interactions at Chr13 trans-QTL targets changed in the predicted direction, demonstrating that heterochromatin-mediated trans-regulation coordinately regulates hundreds of regulatory loci. This work establishes heterochromatin formation as a mechanism by which genetic variation at trans-acting loci coordinates changes across chromatin accessibility, histone modifications, and 3D genome organization, providing a framework for understanding how early developmental chromatin states could generate phenotypic variation while preserving essential developmental programs.

Reading Between the Bases: A Proteomic Screen Reveals a Novel Noncanonical DNA Methylation Reader in the Brain

Mandy Eckhardt1, Tiffany Chin1, Eric Sun1, Hume Stroud1

1Department of Neuroscience, Peter O’Donnell Jr. Brain Institute, UT Southwestern Medical Center, Dallas, TX, USA.

DNA methylation is an evolutionarily ancient key chemical modification of DNA bases essential for development and maintaining homeostasis. DNA methylation typically occurs at cytosine-guanine dinucleotide sequences (mCG) in mammals. More recently, other forms of DNA methylation, most notably at cytosine-adenine sites (mCA), have been observed in certain cell types and tissues. Interestingly, mCA is by far most abundant in neurons where mCA levels exceed that of mCG. Despite its peculiar abundance in the brain, mCA was thought to be an artifact until recently, and thus little is known about its function and the pathway by which it operates. The DNA methyltransferase 3A (DNMT3A) protein is responsible for all mCA deposition in the brain, but knowledge of mCA recognition proteins is less informed. Currently, only one mCA-binding protein is known: methyl-CpG binding protein 2 (MECP2). Curiously, Dnmt3a knockout (KO) neurons and Mecp2 KO neurons vary significantly in gene expression, implying other mCA reader proteins exist but have not been identified. The importance of methylation in the brain is underscored by human neurological disorders caused by mutations in DNMT3A or MECP2, including autism, Rett syndrome, and overgrowth syndromes. Characterizing the mCA pathway may reveal new insights into understanding and treating various neurodevelopmental disorders. One approach to elucidate the mCA pathway is to identify and characterize proteins that preferentially interact with mCA. To this end, we executed a proteomics screen using mouse brain protein lysates. Consistent with current knowledge, this screen revealed MECP2 preferentially binds mCA, validating the screen’s reliability. Excitingly, we also identified a novel putative mCA-binding protein that is poorly studied, as its function, structure, and role in the brain are not well understood. Like mCA, this protein is most highly prevalent in neurons and is conserved across vertebrates, suggesting both are important for proper vertebrate brain function. Current efforts seek to determine the in vivo DNA-binding signatures of this candidate mCA-interacting protein through chromatin immunoprecipitation sequencing, characterize molecular and behavioral phenotypes in a knockout mouse line, and reveal the structure of this mCA-protein complex.

Large Scale Analysis of Early Lethal Phenotypes

Jesse Mager*

Department of Veterinary and Animal Sciences. University of Massachusetts. Amherst, MA. 01003, USA.

*Correspondence should be addressed to: J.M. (jmager@umass.edu).

Early mammalian development requires that embryos accomplish specific milestones including implantation, axis specification and gastrulation. Since these events require precise timing of epigenetic regulation and complex morphogenetic movements for success, they are extremely vulnerable to genetic mutation and embryo lethality. Human data supports this idea with estimates as high as 80% of pregnancy losses occurring during the first trimester. Large scale projects designed to assess gene function in model organisms such as the Knock Out Mouse Project (KOMP) have also highlighted the stringent requirement of these early stages where ~12% of all null alleles result in lethality prior to E9.5. We have analyzed more than 250 novel early lethal single-gene knockouts produced on the same genetic background, which allows us to define specific phenotypic bottlenecks as well as predict essential gene networks. Rather than a distribution, we find that the majority of homozygous mutant embryos have one of two specific phenotypes – a failure to implant or a failure to initiate gastrulation. Our data suggest that distinct mechanisms limit progression through each of these two stages and indicate that implantation failures occur due to molecular and/or transcriptional programing defects while gastrulation failures result from deficiencies in proliferation and/or cell number. Here we will share lessons learned from analysis of such a large cohort of phenotypes as well as highlight some recent gene-specific findings which include lineage defects and failure to establish axes in the early mammalian embryo.

Abstracts for 39th International Mammalian Genome Conference

Evidence-Based, Accessible Mouse Genetic Quality Control: Educating Researchers Through MMRRC Strain GQC

Age-related Behavioural and Molecular Landmarks in New Mouse Models for Studying Alzheimer’s Disease in Down Syndrome

A Novel Mouse NPC Differentiation Protocol Enables Parallel CRISPR Screening

©2024 INTERNATIONAL MAMMALIAN GENOME SOCIETY﻿

©2024 INTERNATIONAL MAMMALIAN
GENOME SOCIETY