Symposium

Advisor: Mohamed Salem
Co-authors: Guglielmo Raymo, Rafet Al-Tobasei, Ali Ali, Tim Leeds

Efficient feed conversion into muscle growth and enhanced fillet quality are economically important traits in aquaculture breeding programs. In this study, we investigated the integration of gut metagenomics (microbiota) profiles with host genomic and transcriptomic data to improve the accuracy of genetic predictions for growth and muscle-related traits in rainbow trout (Oncorhynchus mykiss). We analyzed genome, transcriptome, and gut microbiome data from rainbow trout families exhibiting divergent phenotypes for growth, muscle yield and fillet quality. Genome-wide association studies (GWAS), microbiome-wide association studies (mGWAS), and differential gene expression and microbial abundance analyses were conducted to identify host and microbial factors associated with trait variation. Additionally, we assessed the effect of incorporating microbiome-derived features into genomic prediction models. Our analyses identified multiple differentially expressed host genes and microbial biomarkers linked to growth performance and fillet quality. Notably, V-type proton ATPase subunit C 1-A and gamma-aminobutyric acid receptor subunit rho-1 emerged as candidate genes associated with whole-body weight and muscle development. Incorporating microbiome information into genomic models resulted in modest but consistent improvements in prediction accuracy across five-fold cross-validation analyses. In conclusion, we identified host genomic and microbial markers associated with key production traits in rainbow trout. While the gut microbiome explains a limited portion of phenotypic variance, its inclusion in genomic prediction models enhances the accuracy of selection for complex traits, offering a promising avenue to refine breeding strategies in aquaculture.

Advisor: Younggeon Jin

Mucosal healing has emerged as a pivotal therapeutic goal in inflammatory bowel disease (IBD), linked to sustained remission. While mesenchymal stem cells (MSCs) are currently explored for their immunomodulatory function during mucosal inflammation, their pro-regenerative potential to improve epithelial regeneration remains underexplored. To address this, we develop novel priming methods to enhance immunomodulatory and pro-regenerative properties in MSCs. We hypothesized that human bone marrow-derived MSC (HBMMSC) could directly communicate with injured human colonoids and be primed to enhance the secretion of immunomodulatory and pro-regenerative factors. We used coculture models to investigate the molecular response of HBMMSC to normal and injured colonoid, with injury induced using IFN-γ and TNF-α cocktail in anoxic conditions to mimic intestinal inflammatory injury. To better understand HBMMSC whole transcriptomic response to injured colonoids, we performed RNA sequencing for HBMMSC cocultured with normal or injured colonoids and HBMMSC-only groups. Under normal coculture settings, we observed HBMMSC support colonoid spheroidal growth. Pro-WNT ligand genes (WNT2B, WNT4, WNT5A, and RSPO3) were enhanced in HBMMSC co-cultured with normal colonoids compared to the HBMMSC-only group. In contrast, HBMMSCs cocultured with injured colonoids substantially expressed the pro-WNT signaling genes relative to those with normal colonoids, suggesting their pro-regenerative support for injury repair processes. The transcriptomic data identified significant differentially expressed genes (q-value ≤ 0.05) related to chemokines (CXCL11, 10, 9 and CCL5, 8, etc.), anti-inflammatory cytokines (IDO1 and TGF-b, etc.), and WNT signaling amplifiers (a 3.64fold change of RSPO3 and a 1.83fold change of WNT5A) in HBMMSC primed by injured colonoids. Gene ontology (GO) and KEGG pathway analyses suggest potential crosstalk between MAPK and WNT signaling mediators. The results show that HBMMSCs primed by damaged colonoids enhance cellular signaling supporting immunomodulation and epithelial regeneration. The findings from our new priming model could advance MSC-based therapies by targeting their dual therapeutic capacities to resolve mucosal healing in IBD.

Advisors: Philip Johnson, Li Ma

Immense diversity is conferred on T-cell receptors due to V(D)J gene segments recombination and the modifications at the junctions of the gene segments. As a result, the theoretical population of T-cell receptors (TCRs) vastly exceeds the actual T-cell population in an individual. This makes the presence of identical TCRs across different individuals very unlikely if receptor generation were entirely random. However, some TCRs termed “public TCRs,” due to their higher prevalence in the population, are observed at a frequency higher than is theoretically expected, suggesting a biased generation process. Public TCRs have been implicated in various immune responses in humans. This study aims to leverage machine learning from human data to examine the functional profile of public T cells in cattle. A convolutional neural network (CNN) classifier was trained on a dataset of about 100 million unique TCRβ CDR3 nucleotide sequences from 786 humans to distinguish between public and private TCR sequences. Despite an extreme class imbalance, the classifier achieved a precision of 71%, a recall of 89%, and F1 score of 81%. This was validated by a significant separation in the generation probabilities of the public and private TCR sequences. Integrated gradients showed an enrichment of the 5′ region of the sequences in the network’s ability to classify the sequences as public or private. This indicates the significance of the V gene segment in TCR publicity. The classifier was directly applied to predict the publicity of TCRβ sequences from CD4+ and CD8+ T cells isolated from bovine peripheral blood mononuclear single-cell RNA seq data. Interestingly, more public TCRs were predicted in cattle compared to humans. This shift could be a species effect or the extensive microbiome-influenced immune milieu in cattle. Based on the prediction, public and private T cells were assorted, and their transcriptomes were evaluated. Crucially, the transcriptome profile of pseudo-bulk bovine T cells classified as “public” showed upregulation of immune system processes compared to those classified as “private.” These transcriptome changes confirm that the classifier identified a meaningful biological difference and potentially indicates a unique role of public T cells in disease resistance and tolerance in cattle. 

Advisor: Andrew Broadbent
Co-authors: Sofia Egana-Labrin, Vishi Reddy, Andrew Broadbent

Infectious bursal disease virus (IBDV) is of major economic importance to the poultry industry. The global population of IBDV isolates was recently classified into nine genogroups by phylogenetics analysis based on the amino acid sequence of the hypervariable region (HVR) of the capsid protein (VP2). This classification system has become the gold-standard for the selection of vaccines. However, how the genogroups relate to each other antigenically is poorly understood. To address this, we inoculated groups of chickens with a panel of seven recombinant IBDVs that contained HVRs from six different genogroups from Europe, South America, China, Australia, and the USA. We obtained the serum, conducted a cross-neutralization assay, and applied antigenic cartography analysis to the results. Interestingly, sequence differences (as measured by amino acid substitution count) correlated weakly with antigenic distance (Pearson’s Coefficient = 0.467, R² =0.218), suggesting that the antigenic relationships of IBDV strains could not be accurately predicted based on the HVR sequence alone. We hypothesized that the classifying IBDV based on the structural similarity of the VP2 protein might lead to a better correlation with antigenic distance. To test this hypothesis, we used Alphafold3 to predict the structure of the VP2 timers of the seven recombinant strains and we developed a computational algorithm to compare the structures. Briefly, atoms of the VP2 timer were scored according to properties including electronegativity, polarizability, ionization energy, and physical position (Euclidian distance), and an overall score was determined for each strain. The overall score produced by our algorithm showed a substantially improved correlation with the observed antigenic distance (Pearson’s coefficient = 0.784, R² =0.614). By leveraging the accurate protein structure prediction enabled by AlphaFold3 in conjunction with our algorithm, it might be possible in the future to classify IBDV strains into a “structural phylogeny”, to better inform vaccine antigen selection. 

Advisor: Dr. Fabiana Cardoso
Co-authors: Niraj Suresh, Mai Zarrinkar, Jeff Semler, Tatiane Fernandes, Victor Malacco

Whole-plant corn silage (WPCS) is an indispensable forage in a dairy cow`s diet due to its high energy content. Therefore, the energy availability, fermentation profile, and silage stability are all dependent on WPCS quality. Short-stature corn (STC) is a relatively new hybrid of corn that is suggested to be efficient in growth size, harvesting, and ensiling compared to conventional corn (CC). However, further research is needed to validate its effectiveness and forage quality. This study aimed to assess the physical properties at time of harvest and evaluate the effect of organic acids on nutritional quality, fermentation profile and silage aerobic stability of the two corn varieties. The addition of a commercial organic acid blend (Silage SAVOR^® Kemin Industries, Iowa) was used as the treatment (ORG) versus control (CON) (water; no organic acids). Experimental silos were used to analyze samples at specific ensiling days (d 0, 7, 15, 30, 60, and 120), as well as all days sent to a commercial laboratory (Dairy One, Ithaca, NY) for analysis. Data were analyzed using the MIXED procedure in SAS. The particle size distribution was 14 and 15% on the upper sieves, 56 and 58% in the middle, 15 and 17% on the lower sieves, and 13% in the pan for STC and CC, respectively. The DM % for CC and STC was 32% and 34%, respectively. There was an interaction between variety x day for pH (P <0.01). STC had a greater pH (5.13) compared to CC (4.88) on day 0 compared to day 120, where STC had a lower pH (3.82) compared to CC (3.84). Total acids % were greater for CC (8.0%) (P = 0.05) compared to STC (7.7%), and greater (P < 0.01) for ORG (8.4%) compared to CON (7.3%). Yeast was greater (P < 0.01) for STC (2.19x10³ CFU/g) compared to CC (4.47x10² CFU/g) and greater (P < 0.01) for CON (2.75x10⁴ CFU/g) compared to ORG (3.47x10¹ CFU/g). The results indicate a potential for STC to be a nutritionally viable forage source, and the addition of organic acids improving fermentation profile by increasing total acids to protect against mold, with interaction affects from days ensiled, variety and treatment. 

Advisor: Iqbal Hamza
Co-Authors: Sohini Dutt, Xiaojing Yuan, James Wohlschlegel

Heme, an iron-containing organic ring, acts as an essential co-factor in numerous proteins. Since heme is a hydrophobic and cytotoxic molecule, it must be transported in a well-controlled manner through membranes via specific intra- and inter-cellular pathways. However, the molecular components of heme trafficking remain poorly defined. Our previous studies in Caenorhabditis elegans uncovered HRG-7, an aspartic protease homolog, that mediates inter-organ heme signaling between the intestine and extra-intestinal tissues. Intestinal HRG-7 functions as a secreted signaling factor during heme starvation and is regulated through DBL-1, a BMP5 homolog secreted from neurons. Intestinal expression of hrg-7 is regulated via the transcription factor SMA-9. Here, we identify hrg-11.1, and its paralogs hrg-11.2, hrg-11.3 and hrg-11.4 as additional components of the HRG-7 mediated inter-organ heme communication network. Loss of hrg-11 leads to a heme deficiency signal in the heme sensor strain, phenocopying loss of hrg-7. hrg-11 mutants have significantly lower HRG-7 protein level compared to wildtype. Loss of hrg-11 results in mislocalization of HRG-7 and lower heme levels in muscle and neurons. hrg-11.1 and hrg-11.2 mutants show a heme-dependent growth, phenocopying hrg-7 mutants. Depletion of either hrg-7 or hrg-11 paralogs in hrg-1(heme importer) mutants results in lethality, which are rescued with heme. hrg-1 mutants show elevated HRG-7 even at excess heme that was suppressed in hrg-11 mutants. Intestine-specific loss of hrg-11 significantly depletes HRG-7 protein abundance, phenocopying systemic hrg-11 loss. Knockdown of either hrg-7 or hrg-11 causes AMsh glia to over-migrate and reduces the number of GABAergic synapses, phenotypes that are rescued by heme. Collectively, these results strongly support a role for hrg-11 in regulating HRG-7 secretion, stability, and inter-organ heme signaling function.

Advisor: Zhengguo Xiao

Gastrointestinal (GI) diseases in cattle pose significant challenges to animal health and cause substantial economic losses worldwide. The intestinal epithelium is constantly exposed to dietary antigens, commensal microbes, and pathogens. Intraepithelial T lymphocytes (T-IELs) within this layer are crucial for maintaining immune tolerance against benign harmless stimuli while mounting effective defenses against pathogens. In recent work, we characterized bovine T-IEL subsets for the first time. These subsets showed slight differences in the composition but had cytokine profiles similar to those observed in humans and mice. However, it remains unclear how each T-IEL subset regulates their activity within the intestine, especially in cattle. Given the central role of T cell receptor (TCR) in antigen recognition by T cells, we hypothesized that bovine T-IEL subsets modulate their activation and function through TCR signaling in response to luminal antigens. To test this hypothesis, we isolated four bovine T-IEL subsets: TCRαβ⁺CD8αβ⁺, TCRαβ⁺CD4⁺, TCRγδ⁺CD8⁻, and TCRγδ⁺CD8⁺. Each subset was stimulated in vitro on plates coated with anti-CD3 antibody to directly induce TCR signaling without requiring specific antigen ligands. Our findings revealed differential responses among these bovine T-IELs. TCR stimulation in the TCRαβ⁺CD8αβ⁺ and TCRαβ⁺CD4⁺ subsets upregulated activation marker CD25 while simultaneously suppressing production of pro-inflammatory cytokines such as IFNγ and TNFα. This suggests a mechanism for the regulatory role of T-IELs in tolerating luminal antigens at the intestinal mucosal interface. In contrast, the TCRγδ⁺ subsets (both CD8⁻ and CD8⁺) exhibited minimal to no response under similar conditions, indicating a lesser involvement in TCR-mediated activation pathways. Furthermore, considering the proximity of intestinal epithelial cells (IECs) to T-IELs and their role in antigen presentation, our future research will investigate the bidirectional interactions between bovine T-IEL subsets and IECs to clarify how IECs may influence T-IEL functions. Our studies provide novel insights into mechanisms underlying bovine mucosal immune regulation and highlights the potential for leveraging these findings to develop targeted vaccines and therapeutic interventions for GI diseases in cattle.

Advisor: Tom E. Porter
Co-authors: Madison Berger, Christina R. Ferreira, Brenda Flack, Charlene Hanlon, Kristen Diehl

Early embryonic mortality (EEM) occurs within the first week of incubation, and significantly reduces hatchability and flock productivity. This study explores the yolk lipidome of Cobb broiler breeder embryos, comparing lipid profiles from EEM and healthy, no embryonic mortality (NEM) eggs to identify potential lipid biomarkers associated with early embryonic mortality. A previous discovery-phase lipidome trial in our laboratory utilizing Multiple Reaction Monitoring (MRM) detected a total of 3,706 lipid species across 10 lipid classes in broiler breeder embryonic day 7 yolk samples. For the current trial, yolk samples were collected on days one, two, or three of incubation from fertilized broiler breeder eggs subjected to (1) either 2 or 10 days of storage (n=4/day/storage duration) and (2) with normal development (NEM) or succumbing to EEM (n=3/day; only collected on days 2 and 3 of incubation). Yolk samples were screened for the previously detected lipid species using MRM profiling. Total abundance of each lipid class and relative ion intensities for each lipid species were compared by t-test between experimental groups. Significant effects (p<0.05) were detected for the phosphatidylserine (PS) lipid class with respect to storage at day one and for the PE lipid class with respect to EEM at day three. Metaboanalyst 6.0 identified a total of 134 lipid species that were significantly affected by storage and 46 by mortality. Lipid ontology enrichment analysis (LION) highlighted changes in energy storage pathways, particularly lipid droplet formation and triacylglycerol metabolism, suggesting a crucial role for lipid reserves in early embryonic development. This study provides the first MRM-based lipidomic profiling of broiler breeder egg yolks, revealing storage-related lipidomic shifts and identifying lipid profiles differentiating NEM from EEM samples. These findings offer insights into early embryonic development and mortality mechanisms, supporting future lipid biomarker discovery. Further research is needed to determine how storage impacts yolk composition and to determine if lipidomic shifts are subsequent to embryonic mortality, or if yolk composition is contributing to EEM.

Advisor: Debabrata Biswas

Staphylococcus epidermidis (SE) is a prevalent common microorganism found on the skin of most mammals including humans. However, SE is also known as an opportunistic pathogen that can cause diseases if the skin microbiome becomes skewed to favor its overgrowth. This situation can be worsened if the causative strains are antibiotic resistant. This study investigates Lactobacillus rhmanosus (LR) with or without prebiotic and their antimicrobial activity as an alternative method of limiting SE growth and survivability as well as virulence properties. Multidrug-resistant SE strain was co-cultured with LR in the presence of prebiotics, and antimicrobial activity of LP was determined using colony count of SE at various timepoints. The activity of LR was further studied by collecting and treating SE with cell-free cultural supernatant (CFCS) to determine the mechanism of inhibitory effects. The concentration of SE in a co-culture with LR in the presence of prebiotics was shown to decrease over time compared to SE grown alone, with 2.5% inulin showing the most significant decrease at 12hr (p<0.0001) and a complete reduction at 48hr. CFCS collected from LR grown for 72hr in the presence of prebiotics showed a significant decrease in SE concentration after 6hr of treatment (p=0.0003) and a complete reduction at 72hr, with CFCS in the presence of 2.5% inulin causing a complete reduction of SE at 48hr. Additionally, treatment with CFCS has shown a significant decrease of SE concentration in biofilm after 24hr (p=0.0211), as well as a significant decrease in SE biofilm formation after 48hr (p<0.05). Gene expression analysis showed that in the presence of CFCS, there was an overall downregulation of the SE genes arcC, mur1, and ftsZ, and an upregulation of mutS and gtr, although not statistically significant. Fluorescent microscopy showed that treatment with CFCS resulted in decreased cell wall integrity, shown by a visible decrease in fluorescence intensity of HADA and increase in propidium iodide (PI). LR showed antimicrobial activity against multidrug-resistant SE and in combination with prebiotic (inulin), this antimicrobial activity can be amplified, which could be an alternative option to control SE colonization.

Advisor: Dr. Andrew Schiffmacher
Co-author: Hannah Burchard

The mammary gland is unique in that its function and efficiency is not realized until lactation, an advanced physiologic stage. Thus, understanding how the mammary environment develops in utero is crucial to proactively address problems in lactation and dairy production. While gross development of the bovine gland has been described for nearly a decade, the specific gene regulatory networks governing tissue differentiation remain largely unknown. Uncovering this information would facilitate the establishment of bovine mammary stem cell technologies, thereby offering unique approaches to improving dairy performance and welfare. By studying bovine mammary development, we aim to understand the mechanisms regulating fetal mammary cell stemness, identify stem cell populations throughout bovine development, and develop cellular tools for translational research purposes. Through use of IHC, qRT-PCR, and multiplexed HCR RNA-FISH, fetal bovine tissue was used to define spatiotemporal gene expression of putative master regulators of the fetal multipotent rudiment before basal/luminal segregation. Co-immunolabelling of TRP63 and cytokeratins 5, 8, and 14 at the mammary hillock to primary sprout stage indicate that early mammary cells remain uncommitted to a basal or luminal fate. By the secondary sprout stage, TRP63 localization is restricted to the future basal cell layer. Tissue sections were also used in RNA-FISH to validate probes for a 12-gene panel that explores the transcriptome relevant to mammary stemness and cell fate decisions. Gene expression data will serve as in vivo references to determine the developmental potential and identity of six cultured fetal mammary-derived cell lines. Gene expression of these 2D cell lines will be evaluated by qRT-PCR and RNA FISH. Cell lines demonstrating multipotency will be used to generate 3D models for the genetic study of early mammary lineage differentiation. Upon completion, we will have established a crucial model for use in mechanistic applications that truly reflects the multipotent in vivo mammary environment.

Advisor: Nishanth E. Sunny
Co-authors: Courtney Carlson, Ipsa Mishra, Parama Bhattacharjee

Embryonic chicken liver relies on high rates of lipid oxidation, while the hatchling dramatically upregulates hepatic lipogenesis. A clear crosstalk between branched-chain amino acids (BCAAs; Valine, Leucine, Isoleucine), lipid metabolism, and mitochondrial function is also evident. We tested whether in ovo BCAAs will promote whole-body lipid oxidation and mitochondrial function in the muscle via. modulation of BCAA degradation. In study 1, White Leghorn embryos (day 9) were subjected to daily in ovo (100µL) saline or BCAAs (114 mM Valine, 138 mM Leucine, 83 mM Isoleucine) for 5 days. In study 2, leg muscles from broiler (Strain: Ross 708) chicken at embryonic (E) days 9, 11, 13, 15, and 18, 2- and 4-days post-hatch (PH) were collected. Plasma and tissues were utilized for targeted metabolomics using gas-chromatography mass-spectrometry. Plasma BCAAs (µM±SEM) were higher with in ovo BCAAs (Valine; 369±20 vs. 647±31; P≤0.05). The keto acids of BCAAs were also higher with in ovo BCAAs (µM±SEM, α-ketoisovaleric acid (KIV) of valine; 33.1±1.9 vs. 44.3±2.5; P≤0.05). Plasma β-hydroxybutyrate (µM±SEM; 2881±127 vs. 3307±207; P=0.09) and hepatic Carnitine palmitoyltransferase IA (CPT1A; P≤0.05) trended higher with in ovo BCAAs. These data suggest that in ovo BCAAs resulted in higher BCAA degradation and alterations in lipid metabolism. β-hydroxybutyrate (µg/g ± SEM) was also higher in embryonic muscles vs. PH (E18; 847±34.98 vs. PH2; 87.2±5.59, P ≤ 0.05). On the contrary, the muscle tricarboxylic acid (TCA) cycle intermediates (µg/g ± SEM) were higher during PH (eg., Succinate; E18; 95±16.1 vs. PH2; 262.1±40.5; P ≤ 0.001). Keto acids of BCAAs were higher in the embryonic muscle compared to PH (e.g., KIV; E18; 4.52±0.55 vs. PH2; 1.29±0.17; P ≤ 0.01). These keto acids correlated negatively with muscle TCA cycle intermediates (e.g., Succinate vs. KIV; r(55) = -0.56, P ≤ 0.01) suggesting an interaction between BCAA degradation and TCA cycle activity. In summary, we characterized the patterns of BCAA degradation during embryonic-to-PH and evaluated the impact of BCAAs on whole-body and muscle lipid metabolism/ mitochondrial function. Identifying mechanisms by which BCAAs impact mitochondrial metabolism will allow us to improve the overall metabolic health of the embryos and hatchlings.