POSTER ABSTRACTS - In Alphabetical Order
Poster 1
Disease biomarker discovery using patterns of protein glycosylation
Robert A. Amos (r.amos@uga.edu) and Parastoo Azadi
Complex Carbohydrate Research Center, UGA
Diseases including cancers and autoimmune disorders can disrupt patterns of protein glycosylation that result from abnormal regulation of cell cycle control and changes to metabolic pathways. Comparison of protein glycosylation patterns in healthy versus diseased patients has the potential to be used in the development of new diagnostic tests for diseases that do not currently have well-established biomarkers. Liquid chromatography-mass spectrometry (LC-MS) has risen in popularity within clinical laboratory settings, primarily for the detection and quantitation of small molecules such as hormones and vitamins. Assays for these simple molecules are amenable to the creation of standardized reagents, control analytes, and established linear ranges of detection that exhibit high quantitative reproducibility. Adaption of LC-MS to analyze glycosylation of proteins offers the potential for expanding the library of diagnostic tools for diseases that have complicated patterns of pathogenesis.
We have examined glycosylation patterns of proteins present in serum, plasma, human tissue samples, and induced pluripotent stem cells. The complex matrix of proteins present in these samples imposes challenges in the consistent detection of glycopeptides, particularly for proteins that are present in low abundance relative to the entire biological sample. The chemical heterogeneity of N- and O-glycosylation on proteins also creates challenges because the total signal for a particular target peptide becomes diluted across many different glycoforms present at a single site.
We are developing strategies for overcoming these challenges including glycopeptide enrichment, sample fractionation, immunoprecipitation, and depletion methods to simplify complex samples. One of the major constraints to developing LC-MS approaches that can be widely adapted to clinical lab settings is the time required and potential for user error during manual data analysis when measuring glycopeptide peak areas for cross-sample quantitative comparisons. We are testing software methods to determine the reliability of automated glycopeptide quantitation methods when compared against traditional methods for manual analysis of LC-MS data. The long-term goals of this project are to determine if glycosylation-based biomarkers produce consistent patterns when comparing healthy controls against diseased patient samples and to develop high-throughput methods suitable for widespread adoption within clinical laboratory settings.
Poster 2
Analysis of two diseases associated with human protein O-Fucosyltransferase 1 variants
Siri Anne (siri.anne@uga.edu), Shannon E. Ferry, Kelvin B. Luther, Robert S. Haltiwanger
Complex Carbohydrate Research Center, UGA
Protein O-fucosyltransferase 1 (POFUT1) adds O-fucose to a consensus sequence in Epidermal Growth Factor-like Repeats (EGFs) of NOTCH1. The NOTCH1 signaling receptor has 36 tandem EGFs in its extracellular domain, and the O-fucose modification is crucial for proper folding of the protein, ligand binding, and signaling through the receptor. A pediatric patient presents with symptoms indicative of a NOTCH1 receptor signaling disorder, and whole exome sequencing shows that they are heterozygous for POFUT1 variant alleles. Haploinsufficiency of POFUT1 is expected to show symptoms of Dowling-Degos Disease (DDD). Neither of the patient’s parents have phenotypes indicative of any POFUT1 mutations, suggesting that the variant alleles they have passed down are partially active instead of completely inactive. The goal of this study is to analyze the effects of these mutations on the ability to rescue O-fucosylation and NOTCH1 receptor function. The central question: what are the in vitro and in cellulo effects of POFUT1 point mutations, specifically P233L and S357A, on NOTCH1 fucosylation and function? The in vitro effect of the mutations has been analyzed via in vitro enzyme assays showing decreased fucosylation for the S357A mutant compared to WT. The P233L mutant has not produced sufficient protein to be visualized by SDS-PAGE and cannot be assayed in vitro. The in celluloeffect of these mutants on receptor fucosylation will be investigated by co-transfecting the mutant POFUT1 plasmids into POFUT1 knockout cells with a mouse Notch1 EGF1 to EGF 5 (N1(1-5)) reporter construct. The amount of fucosylation on the EGFs of the N1(1-5) protein will then be analyzed via mass spectrometry. We hypothesize that these variants will retain partial enzymatic activity and can partially rescue O-fucosylation of Notch1 EGFs in POFUT1 knockout cells. Lastly, N1(1-5) protein from patient fibroblasts will be analyzed to determine the total level of in cellulo fucosylation produced by these mutants working in concert. This study will clarify how specific POFUT1 mutations affect NOTCH1 signaling and O-fucosylation in order to understand the molecular basis of disorders such as DDD.
Poster 3
HCD-Triggered 213nm Ultraviolet Photodissociation (UVPD) Fragmentation for More Complete Glycopeptide Mapping
Stephanie Archer-Hartmann (sarcher@uga.edu)(1), Parastoo Azadi (1)
(1) Analytical Services and Training, Complex Carbohydrate Research Center, UGA
Although collisionally-induced fragmentation methods (HCD, CID) have been traditionally used for tandem mass spectrometry, the emergence of newer fragmentation methods has emerged as exciting options to better preserve labile structural features. Ultraviolet photodissociation, or UVPD, has recently become a research focus as a powerful and versatile ion activation method. Unlike CID or HCD, UVPD dissociation involves the absorption of high-energy photons to energize ions within the mass spectrometer to introduce new dissociation pathways that may be inaccessible with conventional collisional techniques. While UVPD using 193nm photons has been shown to efficiently generate a wide variety of fragments across a broad m/z range, UVPD recently has been released as a fragmentation option available on commercial Tribrid Orbitrap mass spectrometers through the Thermo Scientific. Preliminary work has found that, unlike the literature reports of 193nm UVPD, the commercial 213nm UVPD fragmentation results in less cross-ring fragmentation but also many glyco-fragments that preserve labile groups such as sialic acid or acetylation modifications.
In this study, we characterize the fragmentation using HCD-triggered 213nm UVPD fragmentation to map out a variety of different glycopeptides focusing on glycan mapping coupled with the compositional information generated from traditional HCD fragmentation.
Poster 4
Poster 4
GlyTableMaker: The glycomics and glycoproteomics data repository
Sena Arpinar (sena@uga.edu), Mindy Porterfield, Mike Tiemeyer, and Rene Ranzinger
Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia, USA
Advances in glycomics and glycoproteomics have enabled detailed characterization of glycan structures, protein glycosylation sites, and glycosylation patterns across organisms, tissues, and cell types. While these technologies generate valuable datasets, their growing scale presents major challenges for data organization, standardization, and reuse. In particular, a substantial part of the generated data is published without deposition in public data repositories, limiting accessibility for the research community and hindering integration into bioinformatics workflows.
To address this need, we developed GlyTableMaker, a web-based, public data repository designed to streamline the submission, standardization and dissemination of glycomics and glycoproteomics datasets. GlyTableMaker supports data deposition by both experimental data generators as well as curators. The platform supports entry of glycans, glycoproteins, and site-specific glycosylation information, together with essential biological and experimental metadata, including species, tissue, cell line, disease context, and analytical method. GlyTableMaker accommodates both published and unpublished datasets and allows data providers to select data licenses appropriate for their data. After public release of user datasets, these will be accessible thorough an intuitive web portal and via machine-readable formats that use standardized dictionaries, controlled vocabularies, and ontologies to ensure data consistency and interoperability. With the free web portal and machine-readable APIs, GlyTableMaker supports both exploratory and programmatic workflows for the data access.
As a freely available repository for data, GlyTableMaker plays a key role in improving the accessibility of glycan and glycoprotein datasets. Following FAIR (Findable, Accessible, Interoperable, and Reusable) data principles, the platform strengthens data sharing and data integration within the glycoscience community.
Poster 5
Discovery of Cell Surface Glycoproteomic Biomarkers for Pancreatic Cancer via Selective Exo-Enzymatic Labeling of Tumor-Associated Carbohydrate Antigens
Sydney Bedillion(1)(sydney.bedillion@uga.edu), Peng Zhao(1), Digantkumar Chapla(1), Jeremy L. Praissman(1), Kelley W. Moremen(1), Michael Tiemeyer(1), and Lance Wells(1)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, GA
Pancreatic adenocarcinoma (PAAD) is one of the most aggressive and deadly cancers with a less than 15% 5-year survival rate. Due in part to late-stage diagnoses, high recurrence rate, and vague symptomology, PAAD is the third leading cause of cancer-related mortalities. Sensitive and specific detection through biomarkers is crucial for increasing survival rates and improving prognosis. Here, we utilize a live cell surface labeling technique to probe for alterations in the glycoproteomic profile of a pancreatic cancer cell model. While the connection between cancerous phenotypes and abnormal glycan presentation is well documented, the proteins to which these glycans are attached are poorly catalogued. Our Selective Exo-Enzymatic Labeling (SEEL) technique is a one-step reaction that removes all terminal sialic acids while simultaneously installing a synthetic sialic acid containing a biotin linker onto precise glycan-types of interest. We take advantage of recombinant ST6GAL1, ST3GAL1, and ST6GALNAC1 to selectively install this chemical reporter onto complex N-linked glycans, complex O-linked glycans, and truncated O-linked glycans, respectively. SEEL labeling of cancerous and non-cancerous immortalized human pancreatic nestin-expressing (HPNE) cells generated an extensive glycoproteomic profile of the cell surface of pancreatic cancer cells. Once labeled by the enzyme of choice, glycoproteins were enriched by streptavidin pull-down followed by S-Trap (PROTIFI) capture for protein digestion and clean up. Shotgun tandem mass spectrometry proteomics was performed subsequently using ThermoFisher Orbitrap Ascend and Bruker timsTOF fleX mass spectrometers. Significant differences were detected in the proteome, secretome, and cell surface glycoproteome of these cells, including proteins involved in insulin signaling, growth factor signaling, transporter activity, and scaffolding. This work expands our SEEL methodology by applying other sialyltransferases and by exploiting the cancerous phenotype of truncated O-linked glycans. Precise strategies that can capture the specific proteins which are modified by tumor-associated carbohydrate antigens of interest will yield new functional insights into cancer pathophysiology and provide substrates for generating high resolution biomarkers and novel therapeutic targets.
Poster 6
Multiplexed Monitoring of Recombinant Glycoprotein Glycosylation Patterns
George N. Bendzunas(1), Sylvain Lehoux(1), Sheng-Cheng Wu(1), Robert J. Woods(2), Loretta Yang(3)
(1)Lectenz Bio, Athens, GA, USA; (2)Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA; (3)Lectenz Bio, San Diego, CA, USA
Accurate monitoring of glycosylation remains a major challenge in the development and manufacture of therapeutic glycoproteins. GlycoSense™, a microsphere-based multiplex lectin array platform, enables rapid near real-time profiling of glycosylation patterns without the need for time-intensive mass spectrometry workflows. Here, we demonstrate expanded applications of GlycoSense™ for monitoring recombinant glycoprotein production and culture-dependent glycosylation changes in mammalian expression systems. Using recombinant erythropoietin (EPO) and an in-house expressed human IgG antibody (AB1) as model glycoproteins, we compared glycosylation profiles across CHO and HEK293F expression hosts, and evaluated the effects of culture conditions including temperature and media supplementation. GlycoPrint analyses revealed distinct host-dependent glycosylation signatures, including differential α2,3- and α2,6-sialylation patterns consistent with known differences between CHO and HEK-derived systems. To further expand platform utility, we incorporated a novel N-GlyFind reagent into the GlycoSense™ workflow to provide a broad measure of total N-linked glycosylation alongside terminal glycan feature analysis. Together, these results demonstrate the utility of GlycoSense™ as a flexible platform for rapid monitoring of glycoprotein quality attributes and process-dependent glycosylation changes during biologics production.
Supported by NIH grant R44GM153145.
Poster 7
Arabidopsis FRIABLE1 (FRB1) produced in a prokaryotic expression system reveals rhamnogalacturonan-I rhamnosyltransferase activity and critical catalytic residues
Madison N. Blea(1), Kayleigh J. Robichaux(2), Robert A. Amos(1), Chin Huang(1), Debra Mohnen(1), Ian S. Wallace(1,2)
(1)Department of Biochemistry and Molecular Biology, Complex Carbohydrate Research Center, University of Georgia, Athens, GA USA; (2)Department of Biochemistry and Molecular Biology, University of Nevada, Reno, NV USA
The plant cell wall is a heterogeneous mixture of structurally and functionally complex polysaccharides, which are critical in plant growth, cellular adhesion, and cell shape and architecture. Pectin is one critical component of the plant cell wall that influences these processes, but the microheterogeneity and complexity of this class of cell wall polysaccharides make it difficult to study. Rhamnogalacturonan-I (RG-I) is one structural domain of pectin composed of a repeating disaccharide backbone of [4-⍺-D-GalA-1,2-⍺-L-Rha-1,] decorated with varying branched glycans. Understanding complex polysaccharides like RG-I necessitates the identification and biochemical characterization of the enzymes that synthesize them. Plant cell wall glycosyltransferases have traditionally relied on expression in insect, fungal, or human cell culture for successful expression studies, while prokaryotic expression systems have been generally unsuccessful. Here, we show that Arabidopsis FRIABLE1 (FRB1)/rhamnogalacturonan-I rhamnosyltransferase 8 (RRT8) can be produced in Escherichia coli RosettaGami2 as N-terminal maltose-binding protein fusion proteins containing C-terminal 6X-His-tags and report the catalytic constants of FRB1/RRT8 for UDP-Rhamnose and rhamnogalacturonan-I (RG-I). We also examine the activities of mutated FRB1 proteins based on an AlphaFold 3-generated FRB1 structural model with a virtually docked UDP-Rha donor. Enzymatic characterization of the mutated and wildtype FRB1/RRT8 protein confirmed that mutation of predicted catalytic site amino acid residues resulted in a 20-fold reduction in RRT activity. Finally, we demonstrate that when combined with RG-I Galacturonosyltransferase 1, FRB1 polymerizes RG-I in vitro. These results show how a robust E. coli expression system combined with artificial intelligence tools can be used to increase understanding of plant cell wall glycosyltransferase structure and function.
Poster 8
Potential for the Terminal Skp1 Glycosyltransferase to Exert Non-Enzymatic Control of Skp1 in Toxoplasma gondii
Donovan A. Cantrell(1,4)(dac24283@uga.edu), Elizabet Gas-Pascual(1,2,4), Msano N. Mandalasi (1,2), Hyun W. Kim(1,3), *Christopher M. West(1,2,3,4)
(1)Department of Biochemistry and Molecular Biology, University of Georgia, Athens, GA, USA; (2)Tropical and Emerging Diseases, University of Georgia, Athens, GA, USA; (3)Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA; (4)Center for Center for Molecular Medicine, University of Georgia, Athens, GA, USA
The Skp1/Cul1/F-Box (SCF) complex is an E3 ubiquitin ligase responsible for targeting a range of proteins for degradation by the 26S proteosome. Within this complex, a variety of F-box protein (FBP) substrate receptors link to the SCF complex via the Skp1 adaptor protein allowing for differential substrate recognition. Within the intracellular parasite Toxoplasma gondii, Skp1 is subject to oxygen dependent regulation. Under normoxic conditions, the oxygen-dependent prolyl hydroxylase PhyA hydroxylates Skp1 priming it for modification by five Skp1 specific glycosyltransferase activities. Glycosylation plays an important role in controlling Skp1 availability by weakening the tight Skp1 homodimer and altering Skp1’s FBP binding repertoire. This is believed to partially occur through a fuzzy protein interface in Skp1’s C-terminal region. However, the presence of the terminal Skp1 glycosyltransferase, Gat1, in the Skp1 interactome regardless of its glycosylation status is atypical for an enzyme. This behavior is reminiscent of its functional counterpart AgtA in Dictyostelium which exerts non-catalytic functions on Skp1. Utilizing sedimentation velocity analytical ultracentrifugation, we demonstrate that the Gat1 dimer forms a high affinity complex with Skp1 monomers with the stoichiometry dictated by its glycostate. Computational modeling supported by mutagenesis studies indicates that Gat1 competes with the same core hydrophobic interface utilized by FBPs and the Skp1 homodimer. This interface is modulated by variable, transient interactions made by Skp1’s intrinsically disordered C-terminal region (CTR) that are in turn constrained by the glycan. Remarkably, substoichiometric levels of Gat1 mediate monomerization of Skp1 in a CTR-dependent manner, indicating that Gat1 has the kinetic potential to promote Skp1 monomer availability for FBP binding in cells. This weakening of the Skp1 dimer is replicated in a scrambled CTR variant indicating potential leveraging of Skp1’s fuzzy interface in this process. In co-immunoprecipitation studies of Skp1 from cells analyzed by mass spectrometry, the FBP binding repertoire of Skp1 from gat1-KO cells is distinct from that of phyA-KO and wild-type parasites. This is consistent with a novel role for Gat1 in regulating FBP incorporation into SCF complexes.
Poster 9
Phosphorylation of Core O-Mannose: a Gatekeeper for the Dystroglycanopathy pathway
Terrell Carter (Terrell.Carter@uga.edu), Sree Seenivasan, David Steen, Jeffery Fairley, Chin Huang, Erin Suh, Emily Kimbrell, Sandeep Yerraguntla, Robert Bridger, Linda Zhao, Jeremy Praissman, Osman Sheikh, Digant Chapla, Kelley Moremen, Lance Wells
Complex Carbohydrate Research Center, Department of Biochemistry and Molecular Biology, University of Georgia
Dystroglycanopathy results from the failure to generate a specific glycan structure on alpha-dystroglycan (α-DG) that serves as a binding partner for extracellular matrix components with LG-domains, such as laminin. Dystroglycanopathy often results from pathogenic mutations in the genes encoding the glycosyltransferases that build this specific glycan structure referred to commonly as the functional M3 (fM3) glycan that terminates in a repeating disaccharide referred to as matriglycan that is directly involved in the protein-glycan interactions. Biosynthesis of the fM3 glycan requires the activity of 11 enzymes and appears to only be present on one protein, α-DG, at 2(3) sites. We hypothesize that localization and specificity of the M3 pathway specific enzymes ensure that only α-DG is modified with the fM3 glycan. Utilizing recombinant enzymes and sugar nucleotides, we synthesized the entire fM3 glycan in vitro and investigated the requirements of the M3 enzymes’ activities against a variety of canonical and non-canonical acceptor substrates. We demonstrate that specificity appears to be regulated by both enzyme localization, especially for early step enzymes in the endoplasmic reticulum, and highly specific substrate specificity, especially for many of the Golgi enzymes. While protein specificity seems to be primarily regulated by POMGNT2, POMK serves as the writer for Glycan specificity with FKTN/FKRP serving as the readers. Molecular modeling coupled with in vitro activity assays reveal how these M3 gatekeeping enzymes utilize key residues to recognize and coordinate acceptor substrates. This work clearly illustrates the substrate specificities of the enzymes involved in fM3 synthesis. The work also highlights that core mannose phosphorylation, by POMK in the endoplasmic reticulum, ensures that the reader enzymes, FKTN/FKRP in the Golgi, assemble fM3 only on this phosphotrisaccharide structure found exclusively on α-DG.
Poster 10
An Automation Platform for Chemo-Enzymatic Syntheses of Complex Sulfated and Branched Glycans
Saptashwa Chakraborty(1)(sc82800@uga.edu), Kyle Minder(1,2), Anthony Robert Prudden(1), Geert-Jan Boons(1,2,3)*
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia 30602, United States
(2)Department of Chemistry, University of Georgia, Athens, Georgia 30602, United States
(3)Chemical Biology and Drug Discovery, Utrecht Institute for Pharmaceutical Sciences, and Bijvoet Center for Biomolecular Research, Utrecht University, 3584 CG Utrecht, The Netherlands
Diverse collections of well-defined glycans are needed to investigate the molecular mechanisms by which these biomolecules mediate biological and disease processes. Several automation approaches have been introduced to accelerate the enzymatic synthesis of complex glycans. These methodologies, however, have provided only relatively simple oligosaccharides due to limitations in glycosyltransferase selectivity. Here, we describe an automation platform that makes it possible, for the first time, to prepare sulfated polylactosamines and asymmetric multiantennary complex N-glycans via sequential enzymatic and chemical reaction cycles in an automated fashion. This platform is based on a re-engineered ISYNTH AI SWING robotic system, which is equipped with an in-house-built evaporation unit. It integrates glycosyltransferase-catalyzed glycosylations, the use of the unnatural sugar nucleotide donor 5′-diphosphate-2-deoxy-2-trifluoro-N-acetamido-glucose (UDP-GlcNHTFA), and chemical manipulations including base-mediated trifluoroacetamido (TFA) removal, azido transfer and azido reduction, tert-butyloxycarbonyl (Boc) protection, acid-mediated deprotection, and amine acylation. The latter transformations are important for stop-and-go chemoenzymatic synthetic strategies in which unnatural monosaccharides are introduced to temporarily disable specific sites from enzymatic modification. Nascent saccharides are modified by a hydrophobic 2-naphthyl methyl (Nap) tag at the reducing end linker. After completion of a reaction, the reaction mixture is passed through a nickel-NTA resin-packed column to capture His6-tagged recombinant glycosyl transferases. Further, the flow-through is passed through a C18 reverse phase cartridge to ‘catch’ Nap-tagged oligosaccharides, while other reagents can be washed with water. Next, the oligosaccharides can be ‘released’ using 40% acetonitrile in water and transferred to the evaporation unit where acetonitrile is removed, leaving an aqueous solution of product that is ready for the next reaction cycle by simply adding buffer and other reagents or enzymes. For reactions that are conducted in organic solvents, a somewhat modified catch-and-release approach is applied, involving solvent evaporation first, followed by solid-phase extraction on a C18 resin. We have found that standard protocols can catch and release oligosaccharides ranging from disaccharides to multi-antennary N-glycans and highly charged oligosaccharides containing multiple sulfate groups and sialic acids. Removal of Nap tag can be performed by a simple hydrogenation reaction, which makes the oligosaccharides ready for microarray printing or bioconjugation.
Poster 11
A simplified and robust genotyping method
Tessa Ciavarro(1)(tec75557@uga.edu), Liang Zhang(1,2), Tasleem Javaid(1), Breeanna Urbanowicz(1,2)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens 30605, GA, United States.
(2)Department of Biochemistry and Molecular Biology, University of Georgia, Athens 30605, GA, United States.
Genotyping is a fundamental tool used to connect genes to observable traits, with applications in medicine, forensics, and agriculture. DNA extraction is a crucial step in genotyping but often involves unique challenges having to do with contamination, inhibitors, and complications specific to sample type. In plant tissues, DNA extraction can be difficult due to rigid cell walls and PCR-inhibiting secondary metabolites. This study tests a simple, low-cost, and reproducible DNA extraction method using forceps to physically release DNA from leaf tissue and directly introduce it into polymerase chain reaction (PCR) mixtures. The method was tested on duckweed and compared to the previously successful results obtained with Arabidopsis, two model plant species with distinct structural and biochemical properties.
PCR reactions were prepared using a standard master mix, gene-specific primers, and molecular-grade water, with three technical replicates to ensure reliable and consistent results. We developed a positive control using previously successful primers to confirm the PCR was functioning, while a negative control checked for contamination through the absence of a DNA template. Fresh leaf tissue was crushed using forceps and a small amount was immediately transferred into PCR mixtures. Amplification was performed under optimized thermal cycling conditions, and products were analyzed using 1% agarose gel electrophoresis.
The forceps-based extraction method successfully produced high-quality DNA suitable for PCR amplification in both duckweed and Arabidopsis. Clear, distinct bands corresponding to expected fragment sizes were consistently observed across replicates, indicating successful amplification and minimal PCR inhibition. Positive controls yielded strong bands, confirming primer efficiency, while negative controls showed no bands, thus exhibiting effective contamination control.
These results demonstrate that mechanical DNA extraction using forceps is a reproducible and versatile method across plant species with differing cellular complexity. This approach eliminates the need for additional chemicals or specialized materials, providing a simplistic and accessible alternative to traditional extraction methods for plant genotyping applications.
Poster 12
Mass Spectrometric Determination of Site-Specific O-Acetylation in Rhamnogalacturonan-I Oligomers
Liyanage Devthilini Fernando(1)(ldfl8884@uga.edu)*, Xu Yang(1), Stephanie Archer-Hartmann(1), Lubana Shahin(1), Liang Zhang(1), Breeanna R. Urbanowicz(1), Christian Heiss(1), and Parastoo Azadi(1)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, GA 30602, USA
O-Acetylation of rhamnogalacturonan I (RG-I) plays vital roles in plant growth, stress adaptation, and pathogen defense, yet pinpointing the exact positions of acetyl groups remains challenging. Nuclear magnetic resonance (NMR) demands large amounts of pure material, while tandem mass spectroscopy (MS) often suffers from acetyl migration or O-acetyl group during analysis. We present a stabilization strategy that “locks” native acetyls by introducing trideuteroacetyl and propionyl groups onto free hydroxyls of isolated RG-I oligosaccharides. Combined matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF-MS) and electrospray ionization (ESI) MS with MS/MS or tandem mass spectrometry (MSn) enabled precise mapping of monosaccharides and acetyl positions in partially O-acetylated RG-I oligomers. This approach overcomes long-standing analytical barriers, offering a powerful tool to probe the regiospecificity of pectin O-acetyltransferases and to define acylation patterns in complex oligosaccharides.
Poster 13
EXTL3 Negatively Regulates EXT1/EXT2 Stability to Control Heparan Sulfate Chain Elongation
Xiaolin Dong(1,2)(xd80450@uga.edu), Peng Zhao(1), Kaleigh E. Gosnell(2), Jack Carignan Moore(1,2), Brady Kuelpman(2), Digantkumar Chapla(1), Eric Cester(2), Camilo Perez(2), Ana Ramirez Vaca(1,2), Kelley W Moremen(1,2), Lance Wells(1,2), Ryan J. Weiss(1,2)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia 30602, United States
(2)Department of Biochemistry and Molecular Biology, University of Georgia, Athens, Georgia 30602, United States
Heparan sulfate (HS) is a critical cell surface and extracellular matrix polysaccharide that regulates diverse physiological processes, including cell signaling, adhesion, and tissue morphogenesis. HS biosynthesis is driven by a network of Golgi-localized glycosyltransferases, including exostosin-1 (EXT1) and exostosin-2 (EXT2), which form a heterodimeric co-polymerase complex responsible for HS chain elongation. Loss-of-function mutations in EXT1 or EXT2 cause the congenital skeletal disorder, Multiple Hereditary Exostosis (MHE), which is characterized by the formation of cartilage-capped bony outgrowths near the epiphyses of long bones. Despite the clinical significance of EXT1 and EXT2, the mechanisms governing their stability and function remain poorly characterized. To identify regulatory interactions, we performed co-immunoprecipitation (co-IP) coupled with mass spectrometry of endogenous EXT1 and EXT2 in epitope-tagged TC28a2 chondrocytes. Unexpectedly, exostosin-like-3 (EXTL3), the initiating enzyme of HS synthesis, emerged as the top interactor of the EXT1/EXT2 complex. This interaction was orthogonally validated across multiple cell types by reciprocal co-IP pulldowns and confirmed in situ in the Golgi via Proximity Ligation Assays (PLA). Domain-swap experiments further demonstrated that the extended coiled-coil stem domain of EXTL3 is required for complex formation. Strikingly, while ablation of EXT1 or EXT2 had no significant effect on EXTL3 localization, expression, or stability, knockout of EXTL3 caused marked upregulation and prolonged half-life of both EXT1 and EXT2. Pharmacologic inhibition of lysosomal or ubiquitin pathway components recapitulated this accumulation, implicating ubiquitin-dependent lysosomal turnover as the degradative mechanism. Together, these findings establish EXTL3 as a negative regulator of EXT1/EXT2 stability, revealing a previously unrecognized regulatory axis that controls HS chain elongation with potential implications for MHE pathogenesis and HS-associated disorders.
Poster 14
Structural and mechanistic characterization of human Phosphatidylinositol-glycan biosynthesis class W protein (PIGW)
Mahboobeh Faraminlashkarian(1,2)(mf90407@uga.edu), Ana S. Ramírez(1,2)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA
(2)Biochemistry and Molecular Biology, University of Georgia, Athens, GA, USA
Glycosylphosphatidylinositol-anchored proteins (GPI-APs) are widely expressed across eukaryotes. In humans, this GPI anchors mediate the cell-surface localization of approximately 150 proteins involved in essential physiological processes, including embryogenesis, neurogenesis, immune regulation, and development. Phosphatidylinositol-glycan biosynthesis class W protein (PIGW) is an endoplasmic reticulum membrane-resident acyltransferase and the fourth enzyme in the GPI-anchor biosynthesis pathway, catalyzing the inositol acylation of glucosamine-phosphatidylinositol (GlcN-PI). Autosomal recessive mutations in PIGW gene lead to severe neurodevelopmental disorders; however, the molecular basis by which PIGW recognizes its substrate and catalyzes the reaction remains poorly understood. Although other membrane-bound O-acyltransferases (MBOATs) share conserved structural features and acyl-CoA binding sites, it is unclear how PIGW accommodates its complex lipid acceptor substrate and coordinates donor and acceptor molecules within its catalytic site.
We hypothesize that structural and mechanistic characterization of human PIGW will provide critical insight into its enzymatic function and enable rational interpretation of disease-associated mutations. To address this, we optimized the recombinant expression and detergent-based purification of human PIGW. To overcome challenges associated with its relatively small size for cryo-electron microscopy (cryo-EM) studies, we employed a fusion partner strategy in combination with high-affinity Fab fragments to increase particle size and rigidity. Preliminary results indicate that BRIL-fused PIGW constructs form a stable complex with a synthetic nanobody recognizing BRIL and introduce additional structural features that improve particle alignment during initial cryo-EM data analysis. Future studies will focus on defining PIGW function through complementary biochemical and cellular assays. In vitro comparisons of wild-type and mutant enzymes will identify key residues involved in substrate recognition and catalysis, while cell-based rescue experiments will evaluate the restoration of GPI-anchored protein surface expression. Together, these studies will establish a comprehensive framework linking PIGW structure to function and disease.
Poster 15
Biochemical Formation of the Apoplastic PAM Interface in Sorghum Root Cortex
Jonah Friedmann(1,2)(jhf03150@uga.edu), Ajaya Biswal(2), Melani Atmodjo(1,2), Debra Mohnen(1,2), Jonathan Arnold(2)
(1,2)Department of Biochemistry and Molecular Biology: University of Georgia
(2)Complex Carbohydrate Research Center
Arbuscular mycorrhizal fungi (AMFs) are a heterogenous group of symbiotic fungi that associate with over 90% of plant roots. Sorghum bicolor is the 5th most important cereal crop that has been cultivated for thousands of years. Sorghum is a C4 plant that is resilient to drought, saline conditions, and heat stress, making it well-suited for recent green initiatives. Amidst symbiosis, the AMF facilitates the bidirectional exchange of essential micronutrients for up to 20% of host photosynthate derivatives. This exchange occurs across a shared cell wall interface called the peri-arbuscular membrane (PAM), and importantly, little is knownabout the cell wall structure along the PAM interface. To utilize AMF as mutualistic biofertilizers there needs to be a biochemical understanding of how AMF uses intraradical hyphae to penetrate the root cortex and form arbuscules without compromising plant host cell wall integrity. We hypothesize that apoplastic PAM material in sorghum roots originates from a host driven remodeling response to AMF colonization, in which root cortex cells utilize a specific combination of plant cell wall (PCW) degrading and biosynthetic enzymes to both loosen pre-existing polysaccharides and build new polysaccharides present in the PAM. We further propose that this results in controlled remodeling of the wall in a manner that permits AMF hyphal penetration while evading a plant host cell damage response. Current efforts focus on developing a robust system to study this process in sorghum. Towards this end, we have generated cross sections of sorghum fine roots inoculated with AMF who are known to associate with sorghum, verified via India ink staining of fungal structures. These sections will be used for immunofluorescence staining with CCRC glycan reactive antibodies to view changes in cell wall carbohydrates within the PAM interface. We will create a working biomarker & glycan immunolabeling system within a monoxenic hairy root organ culture that can synchronize the PCW remodeling to each of the four phases of AMF colonization. This system will enable researchers to understand how a plant host organizes its cell wall to establish the PAM interface between plant and AMF, providing insight into the biochemistry of glycan-based PAM formation.
Poster 16
Four Glycobiological Examples from Protist Parasites
Elisabet Gas-Pascual (elisabet@uga.edu), Center for Molecular Medicine, Dept. of Biochemistry & Molecular Biology, University of Georgia
Robert Amos, Complex Carbohydrate Research Center, University of Georgia
Carolina Koeller, Federal University of Rio de Janeiro
Julia Alvarez, Dept. of Molecular and Cell Biology, University of California, Merced
Megna Tiwari, Dept. of Biochemistry & Molecular Biology, University of Georgia
Manish Goyal, Dept. of Molecular and Cell Biology, Boston University
Parastoo Azadi, Complex Carbohydrate Research Center, University of Georgia
Norton Heise, Federal University of Rio de Janeiro
Kirk Jensen, Dept. of Molecular and Cell Biology, University of California, Merced
John Samuelson, Dept. of Molecular and Cell Biology, Boston University
Chris West, Center for Molecular Medicine, Dept. of Biochemistry & Molecular Biology, University of Georgia
Protein glycosylation of protists is understudied but no less important than for their animal and plant counterparts. Four examples from recent collaborations are described.
α3-Gal modifications of O-linked glycans in Trypanosoma cruzi, agent for Chagas disease. Metacyclic forms exhibit stage-specific terminal α3-Gal detectable serologically. It has been posited to be a pro-inflammatory signal acting to the parasite’s advantage in infection. In collaboration with Carolina Koeller and Norton Heise, we predicted the responsible glycosyltransferase and generated KOs. With Robert Amos, nLC/MS of alkali-released O-glycans in conjunction with 13C-standards we verified the presence of a trisaccharide sensitive to treatment with α-galactosidase, and absence of the trisaccharide in the mutants. Mutants exhibit deficits in ability of the parasite to infect cells and to properly emerge from host cells.
GPI anchors and GIPLs from Toxoplasma gondii, agent for chronic toxoplasmosis. These glycosylphosphatidylinositols are characterized by a distinctive disaccharide sidechain, which has been explored as a biomarker and vaccine target. In collaboration with Kirk Jensen, we predicted the responsible glycosyltransferases and tested their role by analysis of GPIs and GIPLs from KO strains. nLC/MS analyses confirmed predictions and allowed for functional analysis of the disaccharide-free forms. KO strains exhibited generally increased virulence in mouse infection models, which was associated with changes in host cell protein interactions.
O-Fuc modifications of nucleocytoplasmic proteins in T. gondii. The modification of numerous nucleocytoplasmic proteins with the monosaccharide α-L-fucose was described by John Samuelson and Giulia Bandini, and our group identified the responsible glycosyltransferase as OFT, a paralog of the well-known OGT that mediates O-GlcNAcylation. By epitope tagging O-Fuc proteins in the presence or absence of OFT, we ascertained that O-Fuc modestly supports the abundance of two proteins while not affecting the localization of one in cytoplasm.
O-Fuc in Acanthamoeba castellanii, agent for debilitating corneal infections. Probing permeabilized trophozoites suggested presence of fucose in their nuclei, which was confirmed to be O-Fuc using specialized anti-O-Fuc pAbs. In collaboration with John Samuelson, a proteomics survey of affinity-purified material suggests O-fucosylation of a range of nucleocytoplasmic proteins, including nuclear pore proteins, that parallels findings in the social amoeba Dictyostelium discoideum.
Poster 17
Macrophage Priming Against Glyco-Antigens in the Host Microbiome
Leandre M. Glendenning(1)(leandre.glendenning@uga.edu), Ashley M. Rogers(1,2), Robert T. Patry(1,2,3) and Christine M. Szymanski(1,2)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, GA 30602
(2)Department of Microbiology, University of Georgia, Athens, GA 30602
(3)Current Address: Department of Medical Microbiology and Immunology, University of Alberta, Edmonton, AB T6G 2E1
INTRODUCTION/RATIONALE:
The mammalian immune system recognizes and responds to a broad variety of glycan antigens, most notoriously including lipopolysaccharides (LPS). LPS lacking O-antigens, known as lipooligosaccharides (LOS), can also cause strong immune responses, including Guillain-Barré syndrome (GBS), the leading cause of autoimmune flaccid paralysis in the USA. This condition can develop following resolution of food-borne illness and is commonly associated with infection by Campylobacter jejuni, which produces LOS structures that mimic neuronal gangliosides.
METHODS:
Previously, we showed that macrophage-like THP-1-derived monocytes can be tolerized against the GM1 ganglioside antigen. Pre-incubation with GM1-mimicking C. jejuni (or purified LOS) results in dampened pro-inflammatory cytokine responses following challenge with GM1-mimicking C. jejuni. Here, we continue utilizing this model, complemented with transcriptomics, to probe macrophage responses to C. jejuni challenge following pre-incubation with opportunistic Enterococcus species originating from guts of multiple animals.
RESULTS:
We show that cholera toxin B subunit (GM1 ganglioside probe) cross-reacts to glyco-antigens produced by opportunistic Enterococcus species found within the gut microbiome. Further, a subset of these species can prime THP-1 cells towards a pro-inflammatory cytokine response upon subsequent exposure to C. jejuni.
CONCLUSIONS:
Together, these data further our understanding of how the host immune system recognizes and responds to microbial glycans within the gut microbiome and how this can impact subsequent autoimmune development. These data suggest that colonization with certain Enterococcus species may serve as a potential risk factor for GBS development.
Poster 18
Identifying Potential Protein Interactors of CesA1 N-terminal pS167 in Arabidopsis
Alyssa E. Guarino(1,2)(alyssa.guarino@uga.edu), Morgan Murff(1,2), C. Frankie Griffith(1,2), Payton Gjesvleik(2), Robert Bridger(1), Ian S. Wallace(1,2)
(1)Complex Carbohydrate Research Center, (2)Department of Biochemistry and Molecular Biology, University of Georgia, Athens, GA 30602
The plant cell wall is a dynamic carbohydrate matrix that is essential for growth, structural integrity, and environmental adaptation. Cellulose, the primary load-bearing component of the cell wall, is composed of β-1,4-linked glucose chains synthesized at the plasma membrane by the Cellulose Synthase Complex (CSC). The CSC is organized as a hexamer of trimeric subunits called cellulose synthase subunits (CesA), which each contain an intrinsically disordered cytosolic N-terminal domain (NTD), a glycosyltransferase catalytic domain, seven transmembrane domains, and a C-terminal domain. The NTD of CesA subunits are extensively phosphorylated, with the CesA1 subunit containing nine phosphorylation sites. Despite the abundance of phosphorylation events on the CesA1 NTD, it is still unknown how these modifications may contribute to the recruitment of regulatory protein interactors involved in cellulose biosynthesis regulation. To investigate how phosphorylation influences CesA1 protein interactions, we utilized a phosphoserine orthogonal translation system to produce recombinant CesA1 NTD proteins that contain site-specific phosphoserine. Using recombinantly produced CesA1 NTD with phosphoserine encoded at site S167 (CesA1 pS167), we performed a quantitative proteomics approach to identify potential protein interactors. This experiment identified six candidate protein interactors for site pS167. Preliminary phenotypic characterization and subcellular localization analyses were used to assess if the candidate proteins exhibited CSC-related localization or phenotypes. Among the six candidates, PMI2 and ABIL4 were chosen as top candidates to perform further analysis due to their cytoskeletal co-localization patterns and distinct root phenotypes. To determine whether these proteins directly bind CesA1 pS167, we will utilize in vitro direct binding assays and co-immunoprecipitation experiments with recombinant CesA1 pS167. Additionally, cellulose content analyses in Arabidopsis knockout lines will test if the loss of PMI2 or ABIL4 alters cellulose deposition. These techniques will help us to determine the physiological and biochemical functions of PMI2 and ABIL4 and provide insight into how phosphorylation at CesA1 S167 contributes to CSC regulation. Overall, this works aims to inform a more comprehensive model of post-translational regulation of cellulose biosynthesis.
Poster 19
Mining Domain-Specific O-glycosylation in Human Plasma from Public Glycoproteomic Databases
Huilin Hao(1)(hh84581@uga.edu and Robert S. Haltiwanger(1)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, GA 30605
Human plasma is a clinically relevant resource that has been extensively profiled in large-scale proteomics repositories. While these datasets enable broad coverage of the plasma proteome, domain-specific O-glycosylation remains poorly annotated. This modification involves the addition of O-glucose, O-fucose or O-GlcNAc to Epidermal Growth Factor-like (EGF) repeats, Thrombospondin Type 1 repeats (TSRs) and Elastin Microfibril Interface (EMI) domains. Many O-glycosylated proteins circulate in plasma and represent potential disease biomarkers. Here, we reanalyzed public human plasma glycoproteome datasets using an integrated search pipeline combining Byonic and FragPipe, followed by Python-based data filtering and motif annotation. Three recently published human plasma glycoproteome databases were selected, covering both DDA and narrow-window DIA data from commercial plasma and clinical samples acquired on Thermo Tribrid and Sciex ZenoTOF instruments. Parallel searches were performed with parameters optimized for domain-specific O-glycosylation detection, and search results were processed using Python scripts incorporating consensus motifs for Protein O-fucosyltransferases (POFUTs), Protein O-glucosyltransferases (POGLUTs) and EGF-domain-specific O-GlcNAc transferase (EOGT) to extract motif-matched glycopeptides.
This approach identified over 11,000 glycoPSMs across more than 30 glycosites from 18 confidently assigned glycoproteins. Most glycosites reside within consensus motifs of EGF, TSR or EMI domains, indicating canonical substrates of POFUTs, POGLUTs or EOGT. A subset of glycosites occurs within flexible motifs, such as O-fucosylation in C9 EGF1 containing a mispositioned cysteine, representing noncanonical sites. Notably, 10 of the 18 proteins are previously uncharacterized substrates. Five proteins (PROS1, FA10, C9, HGFA and HABP2) were selected for validation. Proteins were expressed in HEK293T wild-type and enzyme knockout cells and analyzed by LC-MS/MS. POGLUT2/3-mediated O-glucosylation of PROS1 EGF2 and FA10 EGF2 was confirmed with near-complete stoichiometry and was absent in POGLUT2/3 double knockout cells. Mutating O-glucose sites reduced PROS1 and FA10 secretion by ~30%, with similar effects observed in POGLUT2/3 knockout cells. POFUT1-mediated O-fucosylation and POGLUT1-mediated O-glucosylation of HGFA EGF1/2 were also confirmed, and POGLUT1 knockout reduced HGFA secretion by ~30%. Together, these results highlight the power of our integrated pipeline to uncover deeply buried domain-specific O-glycosylation in public proteomic resources and link these glycans to protein function.
Poster 20
N-Glycan Remodeling as a Signature of Drug-Induced Metabolic Shifts Revealed by Single-Cell MALDI-MSI
Lauren E. Hill(1), Lyndsay E.A. Young(1,2), Anand Mehta(1,2), Richard R. Drake(1,2)
(1)Department of Pharmacology and Immunology, College of Medicine, Medical University of South Caroline, 68 President Street, Charleston, SC, 29425, USA, (2)Hollings Cancer Center, Medical University of South Carolina, 86 Jonathon Lucas Street, Charleston, SC, 29455, USA
Reprogramming of cellular metabolism and evasion of the immune system are defining hallmarks of disease and a predictive determinant of therapeutic response. With N-glycosylation serving as both a reflection and result of these processes, glycans play predominant roles in cell signaling, cellular-matrix interactions, metastasis, and immune modulation. However, the direct impact of therapeutic agents on the cellular glycome remain poorly understood, especially in the context of drug-induced metabolic shifts on heterogeneous cell populations like those that comprise the immune system. Here, we leverage our established single-cell matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) platform to profile the unique changes in N-glycosylation following drug exposure across bulk and single-cell populations. Each drug was selected based on their conjugated structural ring systems and membrane-localizing properties, as well as their distinct metabolic targets. Immortalized lymphoma cell lines and clinical PBMC samples were treated with chosen drugs at a range of concentrations for 24 and 48 hours. Following treatments, cells were washed and seeded at 20,000-30,000 cells in 1ul spots for bulk analysis. Single cells were captured using PDMS stamps coated with specific immune cell marker antibodies (e.g., CD4) for unbiased selection while preserving intrinsic cellular heterogeneity. SoloCell software was used to identify and localize approximately 15,000 single cells per slide. N-glycans were released with PNGaseF enzyme, and CHCA matrix was applied post incubation for data acquisition using a timsTOFfleX instrument (Bruker) and SCILs Lab software. Drugs were readily ionizable by MALDI, with their specific m/z values detected. Current findings have demonstrated perturbation of central carbon metabolism (brequinar), altered signal transduction (imatinib), and inhibition of N-glycan and glycosphingolipid biosynthesis (NG1-I and Eliglustat) result in distinct, reproducible remodeling of the N-glycome. These findings support a direct coupling between metabolism and glycan maturation. Notably, N-glycan alterations were detected within each immune cell class (CD4, CD8, CD19, CD14, and CD16), and single cell populations exhibited heterogenous drug uptake. Ongoing efforts aim to evaluate pathway-selective inhibitors to analyze site-specific structural modulation of glycans. By linking targeted therapeutic mechanisms to distinct patterns of glycomic remodeling, this work positions glycans as functional readouts and molecular tools to modulate cellular therapies.
Poster 21
O-GlcNAc Transferase Intellectual Disability Variants Alter Interaction with and Function of the SET1/COMPASS Complex
Naomi L. Hitefield(1)(nlhitefield@uga.edu), Peng Zhao(1), Hannah M. Stephen(1), & Lance Wells(1)
(1)From the Department of Biochemistry and Molecular Biology, Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia, 30605, United States of America
Mutations in O-GlcNAc transferase (OGT) have been found to be causal for X-Linked Intellectual Disability (XLID). Patients with OGT-XLID present with an IQ<70 and limitations in adaptive functioning alongside a high penetrance of dysmorphic facial features, clinodactyly, and eye and ear abnormalities. Mutations in OGT linked to OGT-XLID span the entire gene including both the encoded N-terminal tetratricopeptide (TPR) domain and the C-terminal catalytic domain. OGT is the sole glycosyltransferase that adds a single, non-extended N-acetylglucosamine (GlcNAc) moiety onto thousands of serine/threonine residues of nuclear and cytoplasmic proteins. The O-GlcNAc modification has been shown to regulate basic processes like transcription and translation as well as larger biological roles like neural development and degeneration. The first five XLID-linked variants of OGT reported are localized to the TPR domain of OGT, responsible for protein-protein interactions and substrate selection. Initial work in our lab confirmed no gross biochemical deficiencies for these five OGT-XLID variants, thus we proposed that the protein-protein interactome of OGT variants might drive the phenotype. We utilized male neuroblastoma cells transfected with full-length wildtype or OGT-XLID variant OGTs tagged with the promiscuous biotin ligase TurboID and a FLAG tag to obtain the WT and OGT-XLID OGT interactomes. Via LC-MS/MS, we identified 22 proteins significantly reduced in at least one OGT-XLID variant interactome. Altered proteins have functions in chromatin remodeling and the SET1/COMPASS complex including HCFC1 and SETD1A. We Cas9-engineered BE(2)-M17 cells to house an endogenous OGT-XLID variant, in which we confirmed the reduced interactions of OGT-XLID OGT with HCFC1 and SETD1A. As the SET1/COMPASS complex is responsible for histone 3 lysine 4 tri-methylation (H3K4me3), we investigated H3K4me3 levels and found they were reduced in the variant cell line. Comparing the transcriptome, via RNA-Seq, of the OGT-XLID variant line to wildtype cells identified multiple altered transcripts with roles in processes related to neuronal development and function. In conclusion, we have discovered a common change in the interactome of the TPR variants studied that have downstream impacts on gene expression that may play a role in the phenotype observed in OGT-XLID patients.
Poster 22
Modeling glycans with AlphaFold 3: capabilities, caveats, and limitations
Chin Huang*(1,2)(chin@uga.edu), Natarajan Kannan(1,3), Kelley W. Moremen*(1,2)
(1)Department of Biochemistry and Molecular Biology, University of Georgia, Athens, GA, USA
(2)Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA
(3)Institute of Bioinformatics, University of Georgia, Athens, GA, USA
Glycans are complex carbohydrates that exhibit extraordinary structural complexity and stereochemical diversity, playing essential roles in immune regulation, pathogen recognition, cell communication, and many other biological processes. In humans, more than half of all proteins are glycosylated, particularly those in secretory and membrane-associated pathways, highlighting the importance of glycans in health and disease. The recent release of the AlphaFold 3 (AF3) source code enables customizable modeling of proteins and glycan-containing biomolecular complexes. We evaluated AF3 glycan modeling across several input formats and identified a hybrid syntax using Chemical Component Dictionary (CCD) building blocks linked by ‘bondedAtomPairs’ (BAP) as most effective in producing stereochemically valid models. This workflow was used to create a library of AF3 input templates and corresponding structural models for diverse glycan classes. We further examined modeling capabilities, limitations, and remediation strategies for challenging structures. Glycan interactions with glycosyltransferases, glycosylhydrolases, lyases and lectins were also modeled, benchmarked, and validated against known crystal structures. This protocol-driven strategy is valuable for generating stereochemically plausible, static models of glycan-protein interactions to inform hypothesis development and experimental validation. To support broader use, we developed a graphical web-based interface that automates generation of valid AF3 JSON input files. The tool allows users to draw glycans with an integrated SugarDrawer module or input GlycoCT strings, which are automatically converted into CCD+BAP syntax. This functionality streamlines JSON file assembly, minimizes errors, and facilitates modeling of glycans and glycan-macromolecule complexes. All other custom AF3 functions remain fully supported. By providing benchmarked CCD+BAP examples and an intuitive tool, we aim wider application of AF3 for studying glycan-related mechanisms in biosynthesis, signaling, infection, and disease.
Poster 23
Exploring the biosynthesis of pectic rhamnogalacturonan I (RG-I) and its biological function in Arabidopsis embryo development
Pradeepa Jayawardhane(1,2)(pradeepa.jayawardhane@uga.edu), Haruka Sunazaki(3), Melani Atmodjo(1), Robert Amos(1), Chin Huang(1,2), Jeffrey Ecker(1), Jonah Friedmann(4), Clifford Okoye(1,2), Takeshi Ishimizu(3), Kelley Moremen(1,2), Debra Mohnen(1,2)
(1)Complex Carbohydrate Research Centre, University of Georgia, Athens, USA
(2)Department of Biochemistry and Molecular Biology, University of Georgia, Athens, USA
(3)Ritsumeikan University, Kusatsu, Japan
(4)Lawrence University, Appleton, Wisconsin
Keywords: rhamnogalacturonan I, biosynthesis, embryo defective
Rhamnogalacturonan I (RG-I) is a pectic polymer essential for plant growth and has significant industrial applications. The synthesis of RG-I's backbone, composed of galacturonic acid (GalA) and rhamnose (Rha) [-4-α-d-GalpA-1,2-α-l-Rhap-1-], requires the activities of RG-I:GalATransferase (RGGAT) and RG-I:RhaTransferase. At1g28240 (RGGAT1), the first identified RG-I:GalAT, is the founding member of the Arabidopsis GT116 family. We hypothesize that the seven other RGGAT1 homologs in GT116 may also exhibit RG-I:GalAT activity with tissue- or acceptor-specific functions. To test this, we are heterologously expressing RGGAT1 homologs in HEK293 cells and assessing their RG-I:GalAT activity. So far, we have identified four additional GT116 family members as RG-I:GalATs, each with distinct enzymatic rates with RG-I acceptors. The identified RGGATs exhibit optimal activity within a pH range of 5.5-6.5. All RGGATs preferentially utilize long RG-I acceptor oligomers (DP > 14) for activity. To investigate the biological significance of these enzymes, we focused on the embryo mutant emb2756, a T-DNA insertion mutant in At1g34550 (RGGAT4). This mutant displays a severe embryo-defective phenotype, with embryo arrest at the globular and heart-shaped stages. Segregation analysis revealed a significant germination defect, with only four homozygous plants identified from a segregating population of 175. Only 40% of the homozygous seeds germinated, compared to 100% in wild type (WT). Homozygous mutants exhibited reduced stem height, rosette diameter, and both the size and number of siliques compared to WT. qPCR analysis revealed a 90% reduction in RGGAT4 expression relative to WT. Cell wall analysis of stem, root and leaf tissues of mutant revealed an increased GalA content and decreased Rha content, suggesting disrupted RG-I biosynthesis. Molecular complementation of the emb2756mutant with RGGAT4 is ongoing to restore the defective phenotype. Understanding RG-I biosynthesis and function will enhance its industrial applications and enable targeted gene manipulation.
Poster 24
Genome-wide CRISPR screening reveals genetic regulators of anticoagulant heparin biosynthesis in mast cells
Mingchang Jin(1,2)(mj41066@uga.edu), Amrita Basu(1,2), Jack Moore(1,2), Xiaolin Dong(1,2), Neil Patel(1,2), Zhangjie
Wang(3), Jian Liu(4), Ryan Weiss(1,2)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, GA, 30605, USA
(2)Department of Biochemistry and Molecular Biology, University of Georgia, Athens, GA30605, USA
(3)Glycan Therapeutics, 617 Hutton Street, Raleigh, North Carolina 27606, United States
(4)Division of Chemical Biology and Medicinal Chemistry, Eshelman School of Pharmacy, University of North Carolina, NC
Heparin is the most widely prescribed anticoagulant drug in the world and is used routinely for the treatment and prevention of deep vein thrombosis and pulmonary embolism. Currently, pharmaceutical heparin is a fractionated form of heparan sulfate derived from animal sources, predominantly from connective tissue mast cells in pig mucosa sourced from China. While essential and widely used, outbreaks of animal disease and contamination issues pose a serious threat to the heparin supply. Therefore, identifying alternative methods for heparin production is critical. Mast cells are the only known cell type capable of synthesizing anticoagulant heparin, which is stored in intracellular granules and released during the inflammatory response. Structurally, the heparin polysaccharide backbone is highly sulfated and contains unique sulfated pentasaccharide sequences that provide high affinity binding sites for antithrombin (AT), a serine protease inhibitor that neutralizes thrombin and Factor Xa in the coagulation cascade. Intriguingly, while the enzymes involved in heparin biosynthesis are identical to those for heparan sulfate, the regulatory mechanisms that give rise to mast cell-specific biosynthesis of heparin and its anticoagulant activity remain poorly understood, representing a critical barrier to bioengineered heparin production. To identify genetic regulators of heparin biosynthesis, we performed a genome-wide CRISPR knockout screen in a murine mastocytoma (MST) cell line using exogenous AT as a probe for intracellular heparin. Candidate hits were prioritized using publicly available RNA sequencing data comparing bone marrow-derived precursors and mature mast cells. Validation of top candidates revealed a subset of targets that significantly increased 3-O sulfation and total heparin content in MST cells, measured using established LC-MS/MS methods, suggesting the identity of novel negative regulators of anticoagulant heparin biosynthesis. These findings identify novel molecular targets that could inform cell-based bioengineering a safer, animal-free recombinant heparin.
Poster 25
Rapid and Accurate 3D Structure Modeling of Glycoproteins with GLYCAM-Web
Rajan Kandel, Oliver C Grant, Tobias Grelsson, Eliza Gazaway, Robert J Woods*
Complex Carbohydrate Research Center and Department of Biochemistry and Molecular Biology, University of Georgia, 315 Riverbend Road, Athens, 30602, Georgia
Given the presumed flexibility and heterogeneity of glycans in glycoproteins, obtaining well-resolved glycan structures through experimental methods remains challenging. Researchers often resort to molecular dynamics (MD) simulations and enhanced sampling approaches to understand their structural properties. However, these approaches can be computationally expensive, and may not be able to accurately depict the ensemble of glycan shapes relative to the protein surface in the time accessible to current MD simulations. To address this challenge, we developed an online computational tool “gpBuilder” at GLYCAM‑Web (www.glycam.org) that rapidly generates ensembles of glycan conformations on protein surfaces within the bounds of experimentally observed constraints for all glycosidic linkages and for the asparagine side chain torsion angles in N-linked glycans. Even in the challenging case of N-linked glycans in the Fc domain of an antibody, where many computational methods have failed to generate reasonable models, we demonstrate that gpBuilder produces ensembles that recapitulate the experimentally observed shapes of glycan conformations and asparagine side‑chain torsion angles, and has similar performance across additional glycoproteins. The gpBuilder tool assembles glycans sequentially in the glycosite, rapidly eliminating the steric conflicts in real time as it samples the appropriate rotamers. This strategy enables the tool to generate an ensemble of the glycan conformations for any glycan composition in the glycome that functionalizes a glycoprotein, thereby facilitating the comprehensive characterization of its glycan repertoire.
Poster 26
N-Glycan Recognition and Substrate Specificity of GnT-VI from G. gallus
Authors: Keshav Kishor(1,2)(keshav.kishor@uga.edu), Thomas Buckley(1,2), Balasaheb Ghotekar(1), Seema Bhagwat(1), Evana Sharma(3), Geert-Jan Boons(1,2,4)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens Georgia, 30602, United States
(2)Department of Chemistry, University of Georgia, Athens, Georgia, 30602, United States
(3)Department of Biochemistry & Molecular Biology, University of Georgia, Athens, Georgia, 30602, United States
(4)Chemical Biology and Drug Discovery, Utrecht Institute for Pharmaceutical Sciences, and Bijvoet Center for Biomolecular Research, Utrecht University, 3584 CG Utrecht, The Netherlands
N-acetylglucosaminyltransferase-VI (GnT-VI, MGAT6, UniprotID: Q9DGD1) is a glycosyltransferase derived from chicken (Gallus gallus) that transfers GlcNAc from UDP-GlcNAc to form a ß1,4-GlcNAc on the α1,6-mannoside of complex N-glycans. Humans contain an ortholog of this gene (MGAT4C, UniprotID: Q9UBM8) in which activity has not been elucidated. To examine the acceptor specificity of MGAT6, the first 31 amino acids were truncated and fused with an N-terminal His6 tag and recombinantly expressed in CHO cells. A range of N-glycans having different patterns of antennae was prepared and examined for the ability to be modified by MGAT6 using a UDP-Glo assay. It confirmed that MGAT6 is most active on a substrate containing a ß1,6-GlcNAc antenna with a similar level of activity for the bisecting variant. Kinetic analysis also showed a >50-fold decrease in activity for substrates with just the MGAT1 arm. In-depth NMR analysis confirmed that in the absence of the MGAT5 arm, GnT-VI transfers a ß1,4 GlcNAc at the α1,3 mannoside. Utilizing molecular docking and all-atom molecular dynamics simulations, we generated a structure of GnT-VI in complex with glycans with and without the MGAT5 arm. Our findings suggest a novel substrate specificity of MGAT6 in higher branched multi-antennary N-glycans.
Poster 27
O-Linked Glycosylation Mapping of Wild-Type Human NOTCH3
Jordan Kula (Jak69404@uga.edu), Robert S. Haltiwanger, Youxi Yuan
CCRC
Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy (CADASIL) is a genetically inherited neurovascular disorder affecting the small vessels in the brain. It is brought about by variants in the NOTCH3 protein, with most of them occurring in the extracellular part of the protein, specifically in the epidermal growth factor (EGF) repeats. We postulate that the variants will interfere with O-linked glycosylation on the EGF repeats. Prior research established a comprehensive mouse NOTCH3 (mN3) glycosylation map which allows us to gain a deeper understanding of NOTCH3 glycosylation and facilitate the study of CADASIL-associated variants. We recently obtained a construct which expresses the entire extracellular part of human NOTCH3 (hN3-ECD). However, the expressed hN3-ECD protein was incompatible with its secretion into media. To address this, we truncated the hN3-ECD gene and sub-cloned it into a pSecTag vector, which has a signal peptide and a C-terminal MYC-His6 tag. The present study is focused on determining the O-fucosylation and O-glucosylation sites of hN3 EGF1–34 using the pSecTag vector. This pSecTag construct allowed for effective secretion of hN3-ECD protein, resulting in a significant improvement in protein yield. The secreted hN3 was purified using Ni-NTA affinity chromatography, yielding a purified protein for further glycoproteomic analysis. This mapping of O-glucose and O-fucose modifications establishes a baseline for future comparative studies examining how CADASIL-associated mutations alter glycosylation patterns. The difference in O-linked glycosylation between wildtype NOTCH3 and CADASIL variants in NOTCH3 will allow us to determine if disease-associated mutations alter O-glycan occupancy or structure.
Poster 28
High-Specificity Affinity Tools for Detection and Enrichment of N-Glycoconjugates
Sylvain Lehoux(1), Sheng-Cheng Wu(1), Christian Gerner-Smidt(1), Lu Meng(1), George N. Bendzunas(1), Robert J. Woods(2), Loretta Yang(3)
(1)Lectenz Bio, Athens, GA, USA; (2)Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA; (3)Lectenz Bio, San Diego, CA, USA
Introduction: Asparagine-linked glycans (N-glycans) are central to numerous biological processes, and they are increasingly recognized as potential disease biomarkers and therapeutic targets. However, the complexity of N-glycans—characterized by extensive branching and variable linkages—continues to challenge efforts in detection, purification, and structural characterization. Although advances in analytical methods and instrumentation, particularly in mass spectrometry, have improved N-glycome analysis, there remains a critical need for high-affinity, high-specificity reagents to enable simple and effective interrogation and enrichment of biological samples.
Methods: Lectenz Bio has developed N-GlyFindTM, an N-glycan binding reagent engineered from the mouse F box only protein 2 (FBXO2 or Fbs1) using a combination of molecular dynamics (MD) simulations and screening of combinatorial yeast display libraries.
Results: The resulting reagent and its immobilized resin (N-GlyRichTM) derivative specifically bind the core-structure (Man-GlcNAc-GlcNAc-Asn) present in all N-glycoproteins and N-glycopeptides. The N-GlyFindTM reagents have been thoroughly validated by multiple assays including glycan microarray, Western blot, Bio-Layer Interferometry, ELISA and flow cytometry. The N-GlyRichTM resin, designed for affinity chromatography, showed remarkable enrichment of N-glycopeptides and N-glycoproteins from various sample types.
Supported by NIH grant R44OD035390.
Poster 29
From days to hours: A fast One-Flow method for unified N- and O-Glycan release in high-throughput glycomics
Bhoj Kumar (bhoj.kumar@uga.edu), Parastoo Azadi
Complex Carbohydrate Research Center, University of Georgia
Glycoproteins, decorated with N- and O-linked glycans, are central to virtually all biological systems, influencing protein folding, stability, trafficking, cell signaling, and immune recognition. Because of their critical role, N- and O-linked glycosylation patterns serve as highly informative biomarkers for health, disease, and therapeutic efficacy. However, existing workflows for comprehensive glycan profiling remain slow, labor-intensive, and prone to inefficiencies. Conventional methods typically require sequential, multi-day steps: overnight enzymatic release of N-glycans (PNGase F), followed by a separate overnight chemical β-elimination for O-glycans, along with repeated molecular weight cut-off (MWCO) filtration, desalting, and cleanup steps. These cumbersome procedures not only extend turnaround times but also risk sample loss and introduce contaminants that compromise downstream derivatization chemistries such as permethylation.
Here, we present a novel Fast One-Flow method that achieves the simultaneous release of both N- and O-glycans under a unified buffer and processing system, enabling completion of the entire workflow within a single workday. This streamlined approach is fully compatible with downstream analytical steps, eliminating the need for additional cleanup or enrichment. We demonstrate its robustness through successful application of permethylation derivatization—one of the most impurity-sensitive techniques—yielding enhanced ionization efficiency, stabilization of labile sialic acids, and compatibility with LC-MS platforms.
By reducing preparation time from several days to less than 24 hours, the Fast One-Flow method represents a transformative advance in glycomics, offering a cost-effective, high-throughput solution for the structural analysis of glycans from complex biological samples. This approach not only accelerates biomarker discovery but also provides a scalable platform for large-cohort studies and clinical translation.
Poster 30
Not Presented
Poster 31
A Novel Induced-Fit Strategy Using Condensed Amino Acid Representations for Protein–Ligand Modeling
Alexander H. Lee(1,2)(Alexander.Lee@uga.edu), Robert J. Woods(1,2)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA
(2)Department of Biochemistry and Molecular Biology, University of Georgia, Athens, GA, USA
Modeling induced fit remains a key limitation in computational ligand screening, as most approaches rely on rigid receptor structures, limited side-chain flexibility, or ensembles derived from molecular dynamics simulations of the apo receptor that may not sample the required binding-competent conformation. These constraints can prevent accurate representation of side-chain rearrangements required for ligand accommodation, leading to incorrect binding poses and reduced predictive accuracy. Here, we present a novel induced-fit strategy based on condensed amino acid representations that enables backbone-responsive flexibility during ligand binding. In this framework, side-chain conformations are compacted to alleviate steric barriers, allowing ligands to access otherwise inaccessible binding configurations. Following ligand placement, energy minimization restores full residue geometry while enabling the binding site to adapt in response to both the ligand and surrounding protein environment. This approach is designed to expand accessible conformational space during docking while maintaining physically realistic protein structures. Validation focuses on comparison to experimentally determined protein–ligand complexes and benchmarking against conventional docking strategies using structural and interaction-based metrics. This work establishes a foundation for improved modeling of induced fit and aims to enhance the accuracy of binding pose prediction in structure-based drug design.
Poster 32
Kinetic defects in intellectual disability causal variants of O-GlcNAc Transferase provide opportunities for therapeutic intervention
Jonathan Mayfield (Jjman95@uga.edu)
Department of Biochemistry and Molecular Biology, Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia, 30605, United States of America
O-GlcNAc Transferase (OGT) is the sole enzyme responsible for the O-GlcNAc modification onto thousands of nucleocytoplasmic proteins, and mutations in this X-chromosome gene are causal for an intellectual disability syndrome termed OGT-Congenital Disorder of Glycosylation (OGT-CDG). Currently, we are characterizing novel catalytic domain variants of OGT, T570A, Y835C, and A952V, discovered in new OGT-CDG families. Our studies have shown that all three variants are active glycosyltransferases, but they all fail to recapitulate the activity of wildtype OGT. This finding is in alignment with the molecular modeling of the catalytic domain variants, which suggested the variants are likely Km variants for the donor sugar nucleotide, UDP-GlcNAc. It has been previously suggested that glucosamine supplementation may be a therapeutic option for patients as it increases available UDP-GlcNAc, and if the novel catalytic domain variants are Km mutants, supplementation should alleviate the defects seen in glycosyltransferase activity. We have shown that all three catalytic domain variants are responsive to glucosamine supplementation, and the variants T570A and A952V reach wildtype levels of O-GlcNAcylation upon glucosamine supplementation. Y835C does not reach wildtype levels with supplementation likely due to a stability defect as evidenced by low levels of expression. Furthermore, we have used chemical inhibitors of OGA currently in clinical trials for other diseases to show that these therapeutics may also be used to alleviate global glycosylation defects seen in these catalytic domain variants. Since these three novel variants all likely result in hypoglycosylation of key substrates as the result of an altered Km, therapeutic intervention with substrate supplementation may prove beneficial. We have purified full-length recombinant enzymes and are currently working to fully characterize the enzymatic activity towards protein and peptide substrates as well as the stability defects of Y835C. In characterizing the variants, we have shown that none of them display issues with dimerization by size-exclusion chromatography, and we have shown the variants have lowered glycosyltransferase activity in vitro. Defining which OGT-CDG variants are defective in donor Km may help to define those patients that are more likely to be responsive to substate enhancement therapy.
Poster 33
CRISPR activation screens reveal core protein-mediated regulation of heparan sulfate sulfation and ligand specificity
Jack C. Moore(1,2)(Jack.Moore@uga.edu), Haruki Takeuchi(2), Caitlien Nguyen(2), Chin Huang(1,2), Digantkumar Chapla(2), Amrita Basu(2), Zhangjie Wang(3), Jian Liu(4), Kelley Moremen(1,2), Ryan J. Weiss(1,2)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, GA
(2)Department of Biochemistry and Molecular Biology, University of Georgia, Athens, GA
(3)Glycan Therapeutics Corp, Raleigh, NC
(4)Division of Chemical Biology and Medicinal Chemistry, Eshelman School of Pharmacy, University of North Carolina, Chapel Hill, NC
Heparan sulfate proteoglycans (HSPGs) are essential cell surface and extracellular matrix biomolecules that mediate diverse biological processes through interactions between their heparan sulfate (HS) chains and extracellular ligands. While accumulating evidence suggests that HSPGs and their heterogeneously sulfated HS chains have distinct functionality across cell types and disease states, the molecular mechanisms by which cells regulate HS assembly to modulate HS-ligand interactions remain incompletely understood. To systematically identify genetic determinants of HS-protein interactions, we performed genome-wide CRISPR activation (CRISPRa) screens in HEK293T cells using binding of antithrombin (AT), which selectively recognizes rare 3-O-sulfated HS motifs, or the N-sulfation–specific antibody 10E4 as functional readouts. Unexpectedly, the screens converged on proteoglycan core proteins as key determinants of HS function. Specifically, syndecan-1 (SDC1) emerged as a dominant enhancer of AT binding compared to other syndecan family members. Targeted CRISPRa-mediated upregulation of endogenous syndecan 1-4 expression revealed that only SDC1 significantly increased AT binding despite all syndecans elevating total HS levels. Biochemical analyses revealed that HS chains associated with recombinant SDC1 ectodomain displayed increased 6-O- and 3-O-sulfation relative to SDC2. In vitro 3-O sulfotransferase assays further demonstrated that HS chains isolated from SDC1 were more readily modified, indicating a higher prevalence of precursor sequences permissive for 3-O sulfation. Time-course studies following proteolytic removal of cell surface HSPGs revealed distinct trafficking kinetics between SDC1 and SDC2, suggesting that SDC1 undergoes slower recycling that prolongs biosynthetic processing. Collectively, this study establishes a functional genomics framework for dissecting the regulation of HS-protein interactions and identifies proteoglycan core proteins as active regulators of HS sulfation patterning, thus impacting ligand interactions and cell biology.
Poster 34
Identification and Study of the First Glycosyltransferases that Extend Pectic Rhamnogalacturonan I Backbone with Homogalacturonan
Clifford Okoye(1,2)(cliffordokoye@uga.edu), Robert Amos(1,2), Chin Huang(1,2), Pradeepa Jayawardane(1,2), Digantkumar Chapla(1,2), Jiri Vlach(2), Parastoo Azadi(2), Nika Recto(3), Jeff Ecker(2), Magdalena Bezanilla(3), M., Kelley Moremen(1,2), Debra Mohnen(1,2).
(1)Department of Biochemistry and Molecular Biology, University of Georgia, B122 Life Sciences Bldg., Athens, GA, 30602, USA.
(2)Complex Carbohydrate Research Center, University of Georgia, 315 Riverbend Rd, Athens, GA, 30602, GA, USA.
(3)Life Sciences Center, Dartmouth, Life Sciences Center, Room 231
HB 6044, NH, USA.
Keywords: Pectin, homogalacturonan, rhamnogalacturonan, GAUT13, GAUT14
Pectin, a major plant cell wall polysaccharide, is comprised of the glycan domains homogalacturonan (HG), rhamnogalacturonan I (RG-I), and rhamnogalacturonan II (RG-II) that are synthesized by distinct glycosyltransferase families. In plants, HG exists in the wall as a homopolymer, is the backbone of RG-II, and is covalently linked to the RG-I backbone. The enzymes that link these different pectic domains are unknown. Here, we show that two Arabidopsis α-1,4-galacturonosyltransferases, GAUT13 and GAUT14, in addition to previously reported HG acceptor-dependent and HG de novo synthesis activities, also initiate and elongate α(1→4)-linked HG onto the non-reducing end of RG-I. We find that other GAUTs do not initiate HG synthesis on RG-I, although they elongate such HG once initiated on RG-I by the pollen-tube growth-associated GAUT13/14. We further show that the GAUT13/14 homolog in an early-diverging non-vascular seedless plant, Physcomitrium, retains this trifunctional activity, and that a CRISPR-Cas9 knockout of this gene results in stunted protonemal growth (early filamentous stage) and prevents the formation of leafy reproductive structures (gametophores). Through a combination of computational modeling, site-directed mutagenesis, and in vitroenzyme assays, we identify a unique RTFASKLQSRS motif in the hypervariable region 1 (HV1) of Arabidopsis GAUT13 and GAUT14 as critical for RG-I:HG extension, and show that GAUT10, which normally lacks this activity, gains RG-I:HG extension when this motif is introduced. Together, these findings identify GAUT13 and GAUT14 as key players in pectin heteropolymer synthesis and provide catalysts for probing specific pectic polymer function in plant growth and cell-wall architecture across evolutionary lineages.
Poster 35
A non-canonical EZH2/TRIM28 epigenetic axis drives heparan sulfate remodeling and melanoma metastasis
Neil G. Patel(1,2)(ngp94529@uga.edu), Alexandra Drakaki(3), Farhan Valummel(2), Jack C. Moore(1,2), Amrita Basu(1), Bin Hu(4), Xiaolin Dong(1,2), Peng Zhao(1), Mees Botman(3), Emily Spector(1), Mark C. Mochel(5), Lance Wells(1,2), Jennifer E. Koblinski(4,5), Marten A. Hoeksema(3), Ryan J. Weiss(1,2)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia 30602, United States
(2)Department of Biochemistry and Molecular Biology, University of Georgia, Athens, Georgia 30602, United States
(3)Department of Medical Biochemistry, Amsterdam University Medical Center, University of Amsterdam, Amsterdam, The Netherlands
(4)Massey Cancer Center, Virginia Commonwealth University; Richmond, VA, United States
(5)Department of Pathology, Virginia Commonwealth University; Richmond, VA, United States
Melanoma progression and metastasis are driven not only by oncogenic alterations but also by epigenetic programs that dynamically remodel the tumor microenvironment. Heparan sulfate (HS) proteoglycans are key extracellular matrix components that integrate growth factor signaling, cell-matrix interactions, and migratory behavior by controlling ligand availability and receptor engagement, yet how chromatin-associated factors regulate HS remodeling in cancer remains poorly defined. Here, we identify the histone methyltransferase EZH2 as a key regulator of HS biosynthesis in melanoma. Integrated bioinformatic and genomic analyses revealed enrichment of EZH2 and additional Polycomb Repressive Complex (PRC) factors at regulatory regions of HS biosynthetic genes. CRISPR-mediated loss of EZH2 altered expression of multiple HS-modifying enzymes, most notably the secreted endosulfatases SULF1 and SULF2, resulting in enhanced HS 6-O sulfation and altered ligand binding at the cell surface. Unexpectedly, EZH2 promoted SULF1 expression through a methyltransferase-independent mechanism via a non-canonical interaction with TRIM28, whereas SULF2 was regulated through canonical PRC2-mediated repression. Functionally, SULF1 depletion impaired melanoma cell migration and invasion in vitro and reduced spontaneous metastasis in an orthotopic xenograft model. Together, these findings define an epigenetic axis linking chromatin regulation to extracellular glycan remodeling and identify HS-modifying enzymes as candidate targets to limit melanoma metastasis.
Poster 36
GlyGen: A Resource for Exploring Glycans and Glycoproteins
Rene Ranzinger(1)(rene@ccrc.uga.edu)*, Mike Tiemeyer(1), Raja Mazumder(2) and the GlyGen Consortium
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia, USA
(2)Department of Biochemistry & Molecular Medicine, The George Washington School of Medicine and Health Sciences, Washington, DC, USA
Glycosylation is a fundamental and dynamic biological process that plays critical roles in many biological processes including development, cellular communication, and disease progression. Despite its importance, progress in glycoscience is often limited by the complexity, heterogeneity, and fragmentation of data spanning glycan structures, glycoproteins, glycosylation enzymes, and associated genomic and proteomic information. GlyGen was developed to address these challenges by providing an integrated platform that democratizes access to glycoscience data and enables cross-domain biological discovery.
GlyGen integrates diverse glycoscience datasets through international collaborations with major data providers and repositories, including EBI, NCBI, PDB, GlyTouCan, and GlyCosmos, as well as through direct contributions from data generators. Data from these sources is standardized, harmonized, and cross-linked to enable seamless navigation across multiple biological domains. The integrated data is freely accessible through a user-friendly web portal (https://glygen.org), complemented by programmatic access via application programming interfaces (APIs), a dataset portal, and a SPARQL endpoint, supporting both exploratory and computational research workflows.
In GlyGen each glycan and protein is represented by a detailed, interlinked summary page that consolidates information from multiple resources into a concise presentation. These web pages enable users to explore relationships among glycosylated proteins, their associated glycans, biosynthetic enzymes, and relevant annotations. Flexible search interfaces allow users to query data based on structural features, functional annotations, or biological properties. In addition, GlyGen offers visualization and mapping tools to study glycosylation and the cooccurrence with other site annotation; compare incomplete and related glycan structures; or map identifiers across database namespaces.
By bridging data silos and promoting interoperability, GlyGen enhances the accessibility, usability, and impact of glycoscience data.
Poster 37
Antibodies, the immune system’s double-edge sword: a comparative analysis of GBS and gastroenteritis in humans
Ashley M. Rogers(1,2)(amr39315@uga.edu, Jessica McAlpine(3), Xu Yang(2), Stephanie Archer-Hartmann(2), Israt Jahan(4), Zhahirul Islam(4), Parastoo Azadi(2), Nadja Zeltner(3), and Christine M. Szymanski(1,2)
(1)Department of Microbiology, University of Georgia, Athens, GA, USA; (2)Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA; (3)Center for Molecular Medicine, University of Georgia, Athens, GA, USA; (4)ICDDR,B, GPO Box 128, Dhaka 1000, Bangladesh.
Campylobacter jejuni is a gram-negative organism that is the worldwide leading cause of bacterial gastroenteritis. C. jejuni is coated in an outer layer of lipooligosacchrides (LOS) that are composed of glycans that mimic the structure of glycans found in human gangliosides. Typically, the immune system treats this antigen as “self” which allows C. jejuni to be tolerated by the immune system. However, in approximately 1 in every 1,000 cases of C. jejuni infection, the immune system will generate antibodies against the ganglioside-mimicking LOS which subsequently cross-react to the gangliosides, found abundantly on neurons. Opsonized neurons are then destroyed via recruitment of complement and phagocytes, leading to a paralysis known as Guillain-Barré Syndrome (GBS).
Our team has collected serum, peripheral blood mononuclear cells and fecal samples from former GBS patients, C. jejuni-only enteritis patients, and healthy donors from the same population. With these samples, we analyzed gut microbial composition via 16S rRNA sequencing, anti-ganglioside responses via ELISA, and memory B cell repertoire via fluorescence-activated cell sorting and sequencing. To further explore the immune response, we used human pluripotent stem cell-derived sensory neurons to develop a GBS model which assessed neuropathy by measuring oxidative stress and chromatolysis. Additionally, immune responses were analyzed by western blotting and opsonophagocytosis assays using C. jejuni expressing ganglioside-mimicking LOS.
Through understanding the differences in immune responses in former GBS patients, enteritis-only patients, and healthy donors, therapeutics preventing GBS development and improved GBS treatment can be developed.
Poster 38
Stepwise Enzymatic Synthesis of N-Glycoprotein Standards for
Site-Specific Glycopeptide Characterization by Ion Mobility and
High-Resolution Tandem Mass Spectrometry
Sree Hari Seenivasan(SreeHari.Seenivasan@uga.edu), Peng Zhao, Digantkumar Chapla, Kelley W. Moremen & Lance Wells
Complex Carbohydrate Research Center,
Department of Biochemistry and Molecular Biology
University of Georgia, Athens, Georgia, USA
N-glycosylation is a ubiquitous and critical post-translational modification involved in essential biological processes, including protein folding, trafficking, and cellular communication. Specific glycan structures on therapeutic proteins are monitored as a critical quality attribute in biologics development. Despite this importance, the site-specific characterization and quantification of N-glycans on glycoproteins remains a major analytical challenge. This is primarily due to the vast complexity of glycan structures, which exhibit both site-specific macroheterogeneity (variable occupancy) and microheterogeneity (multiple structures at the same site). Numerous isomeric species are also frequently present in low abundance in native proteome samples. These analytical bottlenecks represent a significant knowledge gap that hinders in-depth glycoproteomic analysis. To address this challenge, we report the enzymatic synthesis of a comprehensive panel of N-glycoprotein standards. We used reporter N-linked glycoproteins expressed in MGAT1-null HEK293 cells, which provide a uniform, oligomannose substrate. Through sequential in vitro enzymatic reactions with purified glycosyltransferases, we generated a diverse set of homogeneous, structurally defined, site-specific N-linked glycoforms, ranging from oligomannose and hybrid types to complex, including core-fucosylated glycoforms. Following proteolytic digestion, these hundreds of glycopeptide standards were then extensively characterized using advanced high-resolution mass spectrometry workflows, including stepped HCD and trapped ion mobility ExD tandem mass spectrometry, to capture their characteristic ion fragmentation profiles. This systematic characterization yielded a spectral library rich with diagnostic ion fragmentation patterns for each glycoform, serving as a reference for complex sample analyses. The generated N-glycoprotein standards and corresponding spectral library will be useful for the unambiguous identification and characterization of N-linked glycoproteins. Furthermore, they provide a crucial foundation for developing and validating robust quantitative workflows that enable more precise and efficient glycoproteomic analyses for both fundamental research and biopharmaceutical development.
Poster 39
Unraveling the Molecular Basis of Substrate Selectivity in Pectin O-acetyltransferases
Lubana Shahin(1,2), Jiri Vlach(2,3), Emily Mathus(2,4), Digant K. Chapala(1,2), Kelley Moremen(1,2), Parastoo Azadi(2,3), Debra Mohnen(1,2), and Breeanna Urbanowicz(1,2)
(1)Department of Biochemistry & Molecular Biology, University of Georgia, Athens, GA
(2)Complex Carbohydrate Research Center, University of Georgia, Athens, GA
(3)Analytical Services, DOE Center, Glycomics & Glycoproteomics, University of Georgia, Athens, GA
(4)Department of Biological Sciences, University of Georgia, Athens, GA
Pectins comprise up to 35% of the primary plant cell wall with roles in cell-cell adhesion/interaction and defense mechanisms. Pectin O-acetylation influences its physiochemical properties and roles in environmental adaptation. Pectin O-acetyltransferases (POATs) have been identified as members of the TRICHOME BIREFRINGENCE-LIKE (TBL) gene family in Arabidopsis thaliana, however, POATs are poorly understood in regards to the specific sites of their acetylation within the pectic glycans. For example, the substrate selectivity of many POATs, rhamnogalacturonan-I (RG-I) vs. homogalacturonan (HG), remains enigmatic and the specific hydroxyl group(s) in the Rha or GalA that are acetylated remain largely unknown. Recently, we biochemically characterized TBR/POAT1 and showed that it acetylates both RG-I and HG with a preference for RG-I. We further showed that TBR is a 3-O rhamnose specific RG-I acetyltransferase (RROAT). Here we describe the biochemical characterization of additional POATs including their biochemical properties and substrate selectivity. To achieve this, we developed methods to precisely determine their regiospecificities through generation and use of defined acceptor substrates combined with detailed structural analyses of the reaction products by mass spectrometry (MS) and nuclear magnetic resonance (NMR) spectroscopy. We are using biochemical data to investigate structure-function relationships via protein modeling and prediction. The results provide a foundation for understanding the design rules for future enzyme engineering efforts and to elucidate the effects of specific of non-glycosyl pectin modifications on plant phenotype, cell wall chemotype, and functions in plant growth and development.
Poster 40
α-1,3-Glucan-Driven Remodeling of the Conidial Cell Wall in an Aspergillus fumigatus Vaccine Strain Alters Innate Immune Recognition
Kalpana Singh(1)#, Ankur Ankur(1)#, Jayasubba Reddy Yarava(1), Caroline Mota Fernandes(2), Gianluca Vascelli(3), Alessia Sulla(3), Teresa Zelante(3), Maurizio Del Poeta(2,4,5)*, and Tuo Wang(1),*
(1)Department of Chemistry, Michigan State University, East Lansing, MI 48824, USA
(2)Department of Microbiology and Immunology, Stony Brook University, Stony Brook, NY 11794, USA
(3)Department of Medicine and Surgery, University of Perugia, Perugia, 06123, Italy
(4)Division of Infectious Diseases, Stony Brook University, Stony Brook, NY 11794, USA
(5)Veterans Affairs Medical Center, Northport, NY 11768, USA
Aspergillus fumigatus is a leading cause of invasive aspergillosis in immunocompromised individuals, and current antifungal treatments are constrained by toxicity, emerging resistance, and limited long-term efficacy, with no approved vaccines available. A mutant lacking the sterylglucosidase gene (sglA) has shown potential as a vaccine candidate due to its ability to elicit protective immune responses; however, the underlying structural basis remains poorly understood. In this work, we employ cellular solid-state NMR spectroscopy to investigate the conidial cell wall architecture of the ΔsglA mutant in comparison to the wild-type strain. Our results reveal significant cell wall remodeling in ΔsglA, characterized by elevated α-1,3-glucan content, increased structural heterogeneity, enhanced interactions with β-glucans, reduced hydration, and restricted molecular dynamics. These changes collectively produce a more rigid cell wall framework with decreased β-glucan accessibility. Importantly, these structural alterations correlate with modified neutrophil responses and shifts in innate immune signaling. This study establishes a connection between cell wall organization and immune recognition, providing insights relevant to the development of antifungal immunotherapeutic strategies.
Poster 41
Determining Novel Tissue Specific Regulators of O-mannosylation
David E. Steen, Jr.(1), Yushu Wang(2), Zheng Zhang(2), Sydney Bedillion(1), Terrell Carter(1), Xiaolin Dong(1), Erin Suh(1), Ryan Weiss(1), Pengpeng Bi(2), Lance Wells(1)
(1)Department of Biochemistry and Molecular Biology, Complex Carbohydrate Research Center, University of Georgia, Athens, Georgia, 30605, United States of America; (2)Department of Genetics, Center for Molecular Medicine, University of Georgia, Athens, Georgia, 30605, United States of America
A group of approximately 12 enzymes work together to add an extended O-Mannose structure (referred to as the functional M3 glycan) on the protein alpha-dystroglycan (a-DG) that terminates in a repeating disaccharide motif referred to as matriglycan. Defects in any of the M3 enzymes result in hypoglycosylation of a-DG and are causal for a family of congenital muscular dystrophies referred to as dystroglycanopathies. These degenerative skeletal muscle diseases can also lead to devastating brain and eye abnormalities along with very short lifespans in the most severe forms. Despite multiple gene products assigned as causal for dystroglycanopathies, a purported 30% of cases are still of unknown genetic etiology. Here we present a CRISPR-Cas9 based genetic screen aimed at identifying novel gene products involved in the synthesis and regulation of the functional M3 glycan in an effort to elucidate genes that are causal for these unexplained dystroglycanopathies. We performed the genetic screen in a novel myoblast cell line with the MYMK gene knocked out that can differentiate to myocytes but is incapable of fusion. We compared the transcriptome and proteome of Wild type and MYMK Knockouts finding minimal differences. This cell line provides us, for the first time, the opportunity to Cas9 screen myocytes, that have functional a-DG unlike myoblasts, via FACS sorting of single cells. Utilizing the clinical antibody IIH6, which specifically recognizes matriglycan, we can sort cells that have lost a gene that prevents them from producing the functional M3 glycan on cell surface a-DG. We have sequenced these genes and are functionally characterizing the novel gene products elucidated from the screen. Further screens in other disease relevant cell lines such as neural and hepatocarcinoma cell lines are in progress. Finally, the characterized gene list will be evaluated by our clinical collaborators to potentially explain undefined cases and to provide new gene product targets for possible novel therapeutic intervention.
Poster 42
High-resolution ultraviolet dissociation of permethylated glycans using 213nm laser
Sonali Sunsunwal (sonali.sunsunwal@uga.edu), Parastoo Azadi
Ultraviolet photodissociation (UVPD), which absorbs high-energy photons, has emerged as a powerful tool in identification and molecular characterization of proteins, lipids, nucleic acids and carbohydrates. While collision-induced dissociation (CID) continues to be the established benchmark for glycan analysis, recent studies have shown UVPD as a promising alternative by enabling more comprehensive structural characterization of glycans with the help of diagnostic cross-ring fragments. UVPD has been explored with deprotonated and permethylated glycans, but in low-resolution ion trap instruments. In this study, we employed UVPD for structural characterization of permethylated glycans using 213nm laser and high resolution Orbitrap mass spectrometry for improved mass accuracy of the diagnostic cross-ring fragments. This approach gave clear diagnostic cross-ring fragments for permethylated α-2,3 and α-2,6 sialyllactoses. In addition, it also provided diagnostic fragment ions for bisecting N-acetylglucosamine (GlcNAc) in human IgG (immunoglobulin G) N-glycans at MS2 level. These findings highlight the ability of 213nm UVPD with high-resolution Orbitrap mass spectrometry for the identification of diagnostic fragments for linkage information and bisecting N-glycans already at the MS2 level. Ongoing work will extend this method for comprehensive structural glycomics in identifying the diagnostic fragments in complex biological samples from diseased or injured tissues to identify glycan structural changes related to pathophysiological states.
Poster 43
Complementary Distribution of Glycoconjugates Revealed by Multi-Enzyme MALDI Mass Spectrometry Imaging
Mehrnoush Taherzadeh Ghahfarrokhi*(1)(Mehrnoush.Taherzadeh@uga.edu), Emily Sekera(2), John Bowling(2), Parastoo Azadi(1)
(1)Complex Carbohydrate Research Center, UGA
(2)St Jude's Research Hospital, Memphis TN
Introduction
Glycoconjugates play crucial roles in cancer biology, particularly in tumor progression and invasion. However, their spatial distribution and potential functions in tumor microenvironments remain poorly understood due to technical limitations in analyzing complex glycan structures in situ. Matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) has emerged as a powerful tool for spatial analysis of biomolecules, but its application to glycoconjugate imaging has been limited by the complexity of glycan structures. Here, we present a combination of specific enzymatic treatments with MALDI-MSI that offers a novel approach to overcome these limitations, enabling detailed spatial mapping of different glycoconjugate classes. This technique can unravel biochemical implications of tumor growth and help in biomarker discovery and margin detection studies.
Method
We developed a multi-enzyme MALDI-MSI workflow incorporating sequential applications of chondroitinase and PNGase F for comprehensive glycoconjugate analysis. Mouse brain tissues containing xenograft tumors are cryosectioned and subjected to systematic enzyme treatments. MALDI-MSI analysis is performed using 9-amino acridine and α-cyano-4-hydroxycinnamic acid as MALDI matrices for negative and positive modes, respectively. Data acquisition is conducted on a MALDI-timsTOF mass spectrometer. Image reconstruction and statistical analysis are performed using SCiLS Lab software. H&E staining is used for biological validation.
Preliminary Results
Glycoconjugates usually have large structures with heterogeneous domains that are not as easily ionized in MALDI as more traditionally MALDI-MSI targets such as lipids and metabolites, making them a more difficult target for imaging experiments. Our multi-enzyme MALDI-MSI approach addresses these issues and shows distinct spatial distributions of various glycoconjugate classes in mouse brain with tumor tissues. Aberrant chondroitin sulfate (CS) accumulation in tumor tissue has been reported, making CS a valuable biomarker for pathological experiments using mass spectrometry imaging. CS is a sulfated glycosaminoglycan made of long chains of alternating N-acetyl galactose and glucuronic acids. The combination of long chains and sometimes multiple sulfations that are labile to ionization in MALDI source, make mass spectrometry imaging experiments challenging for these compounds. Chondroitinase (A, B, or C) treatment breaks different types of CS to smaller disaccharide units that are more easily ionized in the MALDI source and require lower laser energy that reduced the in-source decay and loss of sulfation. This treatment enables us to expose a statistically significant enrichment of chondroitin sulfate disaccharides within tumor regions compared to surrounding healthy tissue. Employing complementary enzymatic digestions in addition to the use of higher resolution MSI can provide sharp visualizations of different tissue types. The results show satisfactory compliance with histopathology results. Next step, N-glycan imaging following PNGase F with a more established workflow demonstrates different distribution pattern for different N-glycans in the mouse brain tissue. The spatial patterns observed through MALDI-MSI shows satisfactory correlation with H&E staining, confirming the precise localization of different glycan species is achievable and notably comparable to histopathological experiments that are more standardized. Additionally, enzyme-facilitated MALDI-MSI of glycoconjugates provides valuable biochemical insights. Imaging data of glycoconjugate distributions of tumor and non-tumor regions, demonstrates distinct boundary patterns that could potentially serve as molecular markers for tumor margins.
Novel Aspect
Demonstration of complementary glycoconjugate distributions in tumor vs healthy tissue using multi-enzyme MALDI-MSI, revealing potential new therapeutic targeting strategies.
Poster 44
Elucidating the glyco-code using functional genomics
C. Kimberly Tsui1(1,2)(ckimberlytsui@gmail.com)
(1)Department of Microbiology and Immunology, LSU Health Shreveport
(2)Incoming Assistant Professor at the Center for Molecular Medicine, UGA
Cell-to-cell interactions and signaling events are modulated throughout the body by a diverse suite of cell surface glycans. While recent advances in chemical tools and techniques have enabled us to modify and characterize this glycome, our understanding of the cellular regulation and recognition of glycans remains limited: How do cells coordinate the synthesis of the tens of thousands of unique glycan structures? How do the mere ~200 human lectins recognize and interpret the massive glycan repertoire? And ultimately, how do these glycan-lectin interactions result in functional cell-cell interactions?
Our group combines functional genomics and glycomics techniques to answer these questions. We aim to (1) elucidate the expanded network of genes that regulate the cell surface glycome and (2) map the endogenous glycan targets of human lectins. Together, this will provide critical insight into how glycans are written and read, allowing us to connect the two sides and reveal how altering cell surface glycans can impact cell-cell signaling. We will initially be focusing on the glycan-lectin interactions at the blood-brain barrier, uncovering the mechanisms by which these interactions can impact neuroinflammation.
In August 2026, the Tsui lab will be moving to the Center for Molecular Medicine at UGA, joining the vibrant glycoscience community in Georgia. We are actively recruiting postdocs, students, and technicians to join our group. Please come chat with me or contact me at ckimberlytsui@gmail.com for more information.
Poster 45
Protein glycosylation in moon snail egg collars
Jose Villalobos(1), Karla Piedl(2), Stephanie Archer-Hartman(1), Emily Mevers(2), and Parastoo Azadi(1)
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA
(2)Department of Chemistry, Virginia Tech, Blacksburg, VA, USA
Moon Snails (Naticidae) cover their eggs with a mucus secreted by the parent before solidifying them with sand, forming a characteristic collar shape. As these egg collars age, they undergo a process of degradation which is mediated by a microbiota dominated by Flavobacteriia. Recent studies revealed Flavobacteriia produce a diverse set of carbohydrate-active enzymes. To understand the potential substrates for these enzymes, we characterized the N- and O-glycans within the egg collars using the analytical tools available at the CCRC.
Utilizing a mixture of GC-MS and LC-MS techniques, we identified a novel core extension using glucuronic acid as well as methylated galactose. Analysis of permethylated O-glycans from moon snail eggs showed a diverse array of glycans containing fucose, xylose and glucuronic acid, including core-fucosylated structures. Compositional analysis of the O-glycans indicated the presence of methyl galactose, which was confirmed by follow-on analysis of the O-glycan fraction in its native form.
Poster 46
Characterizing the Regulation of Lipopolysaccharide Biosynthesis in Neisseria gonorrhoeae
Victoria Walden (Victoria.walden@uga.edu)
Department of Microbiology
The Gram-negative bacterial cell envelope is a selectively permeable multilayered structure that is rich in glycans. In particular, its inner membrane (IM) is surrounded by a cell wall composed of peptidoglycan and an asymmetric outer membrane (OM) with a lipopolysaccharide (LPS) outer leaflet. The LPS-rich OM serves as a potent barrier that prevents the influx of environmental toxins, including antibiotics, into the cell. The OM serves as a major protective feature of Gram-negative organisms that drives intrinsic antimicrobial resistance and thus understanding how its biosynthesis is regulated has the potential to inform therapeutic design. In Escherichia coli (Ec) LpxC, the committed enzyme of LPS synthesis, is negatively regulated by LapB-mediated allosteric inhibition and FtsH proteolysis. Although LpxC regulation has been well-characterized in Ec, LpxC regulatory systems are not broadly conserved, leading to diverse mechanisms in differing bacterial species that are largely uncharacterized. In an effort to expand the known repertoire of systems that control production of LPS, we have set out to investigate the regulation of LpxC in the priority pathogen Neisseria gonorrhoeae (Ng). Our work has identified a LapB homolog in Ng, known as Ght, as a potential regulator of LpxC. Specifically, we have found that NgGht directly binds to NgLpxC in vitro, that NgGht depletion resulted in decreased LPS production, and that NgGht directly increases LpxC catalytic activity in vitro. These findings suggest that NgGht may serve as an activator of NgLpxC to promote LPS production in response to certain stimuli, which is intriguingly opposite of LapB’s impact on EcLpxC activity. Further exploration of additional post-translational factors that influence NgLpxC activity, as well as continued elucidation of the putatively contrasting LPS biosynthesis regulation systems in Ec and Ng, will provide valuable insights into the evolution of LPS production regulators in phylogenetically distinct species.
Poster 47
N-glycan mass spectrometry imaging characterization of intraductal papillary mucinous neoplasms and mucinous pancreatic cancer tissues
Abigail Weatherford, Caroline Kittrell(1), David Bayadyan(1), Kristin Clift(2), Yan Bi(2), and Richard R. Drake(1)
(1)Department of Pharmacology and Immunology, Medical University of South Carolina, Charleston SC
(2)Mayo Clinic, Department of Internal Medicine, Jacksonville FL
Intraductal papillary mucinous neoplasms (IPMN) are common precancerous mucin producing cystic lesions that arise in the ductal epithelium of the pancreas. As some high-grade IPMNs can develop into invasive cancers, there is great clinical need for early detection of these subtypes. A cohort of human IPMN tissue samples representative of the different
histopathologies were analyzed by N-glycan MALDI mass spectrometry imaging (MSI)
workflows to characterize IPMN-associated N-glycans. A subset of tissues also had mucinous pancreatic cancer present, allowing comparison of N-glycan compositions with IPMN. Each tissue had claudin 18.2 immunohistochemistry stainingto definitively identify each IPMN lesion. Spatial N-glycosylation profiles were determined for each IPMN tissue, with over 200 N-glycans compositions detected in some tissues. Notably, lower mass N-glycans between 1500-1800 m/z were detected at high levels in IPMN lesions, as well as larger mass glycans >3000 m/z.
Broadly, IPMN N-glycans contained predominantly bisecting N-acetylglucosamine and multi-fucosylated structures. One notable structural feature of the IPMN and mucinous tumor N-glycans were the abundance of Lac-diNAc (GalNAc-GlcNAc) groups, with one tissue having more than 50 di-LacNAc N-glycans. Tissue distribution and relative abundance of these individual di-LacNAc species were compared with their more common lactosamine (Gal-GlcNAc) counterparts. Correlation of di-LAcNAc abundance relative to clinical properties is ongoing. Also, if present, pancreatic islets also had specific N-glycans that were sulfated bi-antennary di-LacNAc structures. A subset of tissues were chemically amidatedto stabilize sialic acid linkages and determine anomeric configurations. Sialylation of IPMN and mucinous tumor N-glycans was modest compared to fucosylation, and tended to be predominantly singly sialylated in alpha 2,6 linkages. Ongoing studies are focusing on identifying the specific IPMN glycoproteins present using mass spectrometry and spatial proteomics, linked with spatial transcriptomics.
Poster 48
The Study of Binding Interactions of Microfibrillar-Associated Protein 2 to Fibrillin-1 with or without O-linked Glucose
Aura Wilson, Kruti Dalal, Robert S. Haltiwanger
Complex Carbohydrate Research Center, Department of Biochemistry and Molecular Biology, University of Georgia, Athens, GA, USA Email: Alw55146@uga.edu, Kruti.Dalal@uga.edu, rhalti@uga.edu
O-glucosylation of Epidermal Growth Factor-like (EGF) repeats plays an important role in protein folding, trafficking, and function. Two novel protein O-glucosyltransferases (POGLUT) 2 and 3 modify EGF repeats on extracellular matrix (ECM) proteins including Fibrillins (FBNs) which are essential for formation of microfibrils and elastic fibers in the ECM. Glycoproteomics analysis revealed that approximately half of the EGF repeats in FBN1, are modified by POGLUT2/3. Without O-glucosylation, FBN1 is unable to properly fold and reduces its secretion in conditioned medium and within ECM in mouse Poglut2/3 double-knockout (DKO) dermal fibroblasts at E18.5. Another glycoprotein that is decreasing in ECM of DKO dermal fibroblasts is Microfibrillar-associated protein 2 (MFAP2) suggesting that O-glucosylation on hFBN1 may be required to bind to MFAP2. MFAP2 is known to bind with FBN1, but the explicit details of O-glucosylation on the interaction remains unexplored. Current work aims to compare the binding interactions between MFAP2 and FBN1 with or without O-glucose modifications using BioLayer Interferometry (BLI). It is hypothesized that β O-linked glucosylation on hFBN1 is required for the binding interaction with MFAP2. Transfections were performed using constructs encoding His-tagged N-terminal FBN1 (EGF1-26) and MFAP2 in wild-type HEK293T and FBN1 construct was transfected into POGLUT2/3 DKO HEK293T. Protein in culture media were purified using Ni-NTA beads followed by SDS-PAGE and mass-spectrometry analysis. Ongoing BLI experiments measure the association and dissociation kinetics to compare MFAP2-FBN1 binding interactions with or without O-glucosylation. Differences in binding affinity will reveal how O-glucose modulates ECM assembly and microfibril function.
Poster 49
Dry lab oriented wet lab design to enable alginate mannuronic/guluronic acid ratio quantification by mass spectrometry
Xu Yang(1)(xu.yang1@uga.edu), Hongrui Wu(2), Xinyi Ni(2), Ambrish Kumar(1), Yuefan Song(3), Jonathan S. Dordick(3), Christian Heiss(1), Pengyu Hong(2), Parastoo Azadi(1)
(1)Analytical Services & Training, University of Georgia Complex Carbohydrate Research Center, Athens, GA, United States.
(2)Department of Computer Science, Brandeis University, Waltham, MA, United States.
(3)Center for Biotechnology and Interdisciplinary Studies, Rensselaer Polytechnic Institute, Troy, NY, United States.
The linear backbone of alginate is made of a pair of C-5 epimers, α-L-guluronic acid (G) and β-D-mannuronic acid (M) in apparently irregular lengths of blocks. The M/G ratio is a crucial property of alginate materials relevant to their gel-forming ability, pH response, and biofunctions. Conventional 1H-NMR and FTIR spectroscopy methods analyze bulk solution/water-free specimens, but only indirectly portray structural information of alginate. We have developed a complementary higher collisional energy dissociation (HCD) based mass spectrometry (MS) method enabled by machine learning (ML) interpretation to better describe the M/G ratio of alginate oligosaccharides (AOS) at the molecular level. Intensive purification and benchmarking synergy were used to establish the quantitative relation between M/G ratio of individual degree of polymerization (DP) 3-7 AOS and their corresponding HCD-MS2 spectra. We were able to imitate MG-block fragmentation behavior by formulating MM-block and GG-block AOS. The mean absolute error (MAE) of M/G ratio prediction model could reach 1-2% on adequately hydrolyzed samples. For hydrolysis-resistant GG-block-rich alginate, prediction based on DP3-4 has a deviation of 12% from the overall 1H-NMR result. Our MS-ML M/G ratio quantification approach bypasses dependencies on pure standards, quantitative derivatization/hydrolysis, and unambiguous spectral annotation required in conventional carbohydrate analysis. The wet lab design is completely forward thinking, actively fueling the dry lab demands.
Poster 50
Polysaccharide Double Helix Formation: The Amylose Story
Zhe Yang(1)(zy67785@uga.edu), Xiaocong Wang(1), Robert J. Woods(1)*
(1)Complex Carbohydrate Research Center, University of Georgia, Athens, GA 30602, USA
Understanding how certain polysaccharides form stable double helices while others do not is a fundamental question in carbohydrate science. Amylose is a linear polysaccharide made of α(1→4)-linked glucose units that provides an ideal system to study this question. This study aims to explain how polysaccharide chain length affects double helix formation and stability.
Molecular dynamics (MD) simulations indicate that the stability of amylose double helices depends strongly on their length, and, in agreement with experiment, suggest that a minimum length of 12 monosaccharides is necessary for stability. Molecular Mechanics-Generalized Born Surface Area (MM-GBSA) energy calculations were combined with normal-mode entropy calculations to obtain the free energy (ΔG) of double helix formation. This analysis indicated that double helix formation is enthalpically favored, but is entropically unfavorable. For shorter chain lengths, entropy overcomes enthalpy and the chains dissociate spontaneously, while over approximately 12-residues, enthalpy wins over enthalpy and the chains remain together.
At the cusp of stability, the MD simulations showed spontaneous sliding movements of each polysaccharide chain along the axial spiral structure. These sliding motions offer important clues into how polysaccharides dissociate and raise key questions about what molecular factors drive or resist these movements. This ongoing work seeks to reveal how chain length and molecular interactions together govern polysaccharide double helix stability, contributing to the broader understanding of polysaccharide behavior in biological and material systems, and potentially guiding the design of novel glycomaterials.
Poster 51
Targeting Sialic Acid Uptake to Disrupt Bacterial Persistence in COPD
Helia Yazdiani (hy34459@uga.edu), Robert Woods
Department of Biochemistry and Molecular Biology, CCRC, University of Georgia
Chronic obstructive pulmonary disease (COPD) is a leading cause of mortality worldwide and is frequently exacerbated by persistent bacterial infections in the airway. One of the most prevalent pathogens in COPD patients, nontypeable Haemophilus influenzae (NTHi), survives by scavenging host-derived sialic acid (Neu5Ac) and incorporating it into its surface to evade immune recognition. This process enhances biofilm formation, complement resistance, and long-term persistence. Despite its importance, no current therapies target bacterial sialic acid acquisition, highlighting a critical gap in current anti-virulence treatment strategies.
Sialic acid uptake in NTHi is mediated by the TRAP transporter system, in which the periplasmic binding protein SiaP captures Neu5Ac with high affinity and undergoes ligand-induced conformational changes that initiate transport. However, the structural mechanisms governing this activation process remain poorly understood.
In this study, we investigate SiaP as a ligand-activated molecular switch and explore its potential as a druggable target. Using molecular dynamics simulations based on the closed SiaP–Neu5Ac structure (PDB: 3B50), we separated the protein into its two domains to evaluate their individual contributions to ligand binding. Preliminary results indicate that Domain II plays a dominant role in stabilizing Neu5Ac interactions, supported by enhanced binding stability and favorable interaction energies. These findings define key residues and structural features that can be selectively targeted.
Building on these insights, we implemented a structure-guided virtual screening strategy to design sialic acid analogs by grafting functional moieties onto the carbohydrate scaffold to exploit identified binding pockets. Recombinant SiaP has been successfully expressed and purified, and ligand binding is being evaluated using NMR-based assays to quantify affinity and competitive inhibition.
Overall, this work establishes a mechanistic framework for SiaP-mediated transport and introduces a novel anti-virulence strategy to disrupt bacterial persistence by targeting nutrient acquisition rather than viability. This approach may reduce infection severity in COPD while minimizing selective pressure for antibiotic resistance.
Poster 52
Multi-omic Single cell MALDI-MSI reveals N-glycome and lipidome remodeling in PBMCs and TCR-engineered Cells
Lyndsay E.A. Young(1,2), Lauren E. Hill1, Satyajit Das(2,3), James W. Dressman(1), M. Furkan Bayram(1), Shikhar Mehrotra(2,3), Anand Mehta(1,2), Richard R. Drake(1,2)
(1)Department of Pharmacology and Immunology, College of Medicine, Medical University of South Caroline, 68 President Street, Charleston, SC, 29425, USA, (2)Hollings Cancer Center, Medical University of South Carolina, 86 Jonathan Lucas Street, Charleston, SC, 29455, USA (3)Department of Surgery, Medical University of South Carolina, 86 Jonathan Lucas Street, Charleston, SC, 29425
Metabolic fitness is a fundamental determinant of immune cell function, influencing activation, persistence, and effector responses. N-glycan expression and the cellular lipidome further regulate membrane structure, signaling pathways, and the molecular identity of immune cells. Single-cell omics advances have revealed how cellular heterogeneity influences immune function and therapeutic outcomes, prompting us to develop a single-cell array platform for profiling peripheral blood mononuclear cells (PBMCs). PBMCs comprise immune subsets including lymphocytes, monocytes, and dendritic cells that coordinate innate and adaptive immunity. Human clinical samples were processed using a high-throughput single-cell array-based slide assay. Cells were patterned into a grid via PDMS stamp at 60 μm spacing, and the custom software SoloCell was developed to identify and record the location of individually captured cells. Biomolecular profiling of lipids and N-glycans was performed by Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry Imaging (MALDI-MSI) on a Bruker timsTOF fleX, with data analyzed in SCiLS software. PBMC subpopulations were captured using CD4, CD8, CD14, CD19, and CD56 antibodies, revealing clear separation of cell populations based on distinct lipid and glycan signatures indicating strong cell-type specificity. A total of 4,979 single cells were captured via antibodies and ConA lectin on the same slide, ensuring unbiased selection that preserves the natural heterogeneity of PBMC isolates. We extended our analyses to transduced and untransduced TCR cells cultured with either IL-2 or a T1/T17 hybrid programming cocktail. Samples were captured, using T cell-specific markers (CD4 and CD8) and ConA lectin. This approach yielded 11,797 CD4 or CD8-captured single cells for comprehensive molecular profiling. Comparative analyses revealed pronounced shifts in the surface N-glycome between CD4+ and CD8+ TCR cells following transduction, suggesting glycan-associated modulation of T-cell activation. Lipidomic profiling identified condition-specific remodeling patterns, including increased phosphoethanolamine species in hybrid-programmed TCR cells and elevated lysophosphatidylethanolamine species in IL-2–cultured TCR cells. Together, these findings demonstrate that coordinated N-glycan and lipid remodeling shapes immune cell identity and function, with this flexible platform offering a powerful tool for immunotherapy development and cell-based therapy optimization.