Glycan node profiling of soluble and membrane glycoproteins in whole cell lysates.

Glycan node analysis (GNA) is a molecularly bottom-up glycomics technique based on the relative quantification of glycan linkage-specific monosaccharide units ("glycan nodes"). It was originally applied to blood plasma/serum, where it detected and predicted progression, reoccurrence, and survival in different types of cancer. Here, we have adapted this technology to previously inaccessible membrane glycoproteins from cultured cells. The approach is facilitated by methanol/chloroform precipitation of cell lysates and a "liquid phase permethylation" (LPP) procedure. LPP gave better signal-to-noise, yield and precision for most of the glycan nodes from membrane glycoproteins/glycolipids than the conventional solid phase permethylation approach. This GNA approach in cell lysates revealed that specific glycan features such as antennary fucosylation, N-glycan branching, and α2,6-sialylation were elevated in hepatocellular carcinoma (HepG2) cells relative to leukemia cells (THP-1 and K562) and normal donor PBMCs. Additional nodes commonly associated with glycolipids were elevated in the leukemia cells relative to HepG2 cells and PBMCs. Exposure of HepG2 cells to a fucosyltransferase inhibitor resulted in a significant reduction in the relative abundance of 3,4-substituted GlcNAc, which represents antennary fucosylation-providing further proof-of-concept that downregulation of glycosyltransferase activity is detected by shifts in glycan node expression-now detectable in membrane glycoproteins.

Glycan Node Analysis Detects Varying Glycosaminoglycan Levels in Melanoma-Derived Extracellular Vesicles.

Extracellular vesicles (EVs) play important roles in (patho)physiological processes by mediating cell communication. Although EVs contain glycans and glycosaminoglycans (GAGs), these biomolecules have been overlooked due to technical challenges in comprehensive glycome analysis coupled with EV isolation. Conventional mass spectrometry (MS)-based methods are restricted to the assessment of N-linked glycans. Therefore, methods to comprehensively analyze all glyco-polymer classes on EVs are urgently needed. In this study, tangential flow filtration-based EV isolation was coupled with glycan node analysis (GNA) as an innovative and robust approach to characterize most major glyco-polymer features of EVs. GNA is a molecularly bottom-up gas chromatography-MS technique that provides unique information that is unobtainable with conventional methods. The results indicate that GNA can identify EV-associated glyco-polymers that would remain undetected with conventional MS methods. Specifically, predictions based on GNA identified a GAG (hyaluronan) with varying abundance on EVs from two different melanoma cell lines. Enzyme-linked immunosorbent assays and enzymatic stripping protocols confirmed the differential abundance of EV-associated hyaluronan. These results lay the framework to explore GNA as a tool to assess major glycan classes on EVs, unveiling the EV glycocode and its biological functions.

Extracellular vesicle glucose transporter-1 and glycan features in monocyte-endothelial inflammatory interactions.

Monocyte-induced endothelial cell inflammation is associated with multiple pathological conditions, and extracellular vesicles (EVs) are essential nanosized components of intercellular communication. EVs derived from endotoxin-stimulated monocytes were previously shown to carry pro-inflammatory proteins and RNAs. The role of glucose transporter-1 (GLUT-1) and glycan features in monocyte-derived EV-induced endothelial cell inflammation remains largely unexplored. This study demonstrates that EVs derived from endotoxin-stimulated monocytes activate inflammatory pathways in endothelial cells, which are partially attributed to GLUT-1. Alterations in glycan features and increased levels of GLUT-1 were observed in EVs derived from endotoxin-stimulated monocytes. Notably, inhibition of EV-associated GLUT-1, through the use of fasentin, suppressed EV-induced inflammatory cytokines in recipient endothelial cells.

Surface plasmon resonance microscopy identifies glycan heterogeneity in pancreatic cancer cells that influences mucin-4 binding interactions.

Membrane proteins are the main targets of therapeutic drugs and most of them are glycosylated. Glycans play pivotal roles in several biological processes, and glycosylation changes are a well-established hallmark of several types of cancer, including pancreatic cancer, that contribute to tumor growth. Mucin-4 (MUC-4) is a membrane glycoprotein which is associated with pancreatic cancer and metastasis, and it has been targeted as a promising vaccine candidate. In this study, Surface Plasmon Resonance Microscopy (SPRM) was implemented to study complex influences of the native N-glycan cellular environment on binding interactions to the MUC-4 receptor as this is currently the only commercially available label-free technique with high enough sensitivity and resolution to measure binding kinetics and heterogeneity on single cells. Such unique capability enables for a more accurate understanding of the "true" binding interactions on human cancer cells without disrupting the native environment of the target MUC-4 receptor. Removal of N-linked glycans in pancreatic cancer cells using PNGase F exposed heterogeneity in Concanavalin (Con A) binding by revealing three new binding populations with higher affinities than the glycosylated control cells. Anti-MUC-4 binding interactions of enzymatically N-linked deglycosylated pancreatic cancer cells produced a 25x faster association and 37x higher affinity relative to the glycosylated control cells. Lastly, four interaction modes were observed for Helix Pomatia Agglutinin (HPA) binding to the glycosylated control cells, but shifted and increased in activity upon removal of N-linked glycans. These results identified predominant interaction modes of glycan and MUC-4 in pancreatic cancer cells, the kinetics of their binding interactions were quantified, and the influence of N-linked glycans in MUC-4 binding interactions was revealed.

The antibody-binding Fc gamma receptor IIIa / CD16a is N-glycosylated with high occupancy at all five sites.

The antibody-binding Fc γ receptors (FcγRs) trigger life-saving immune responses and many therapeutic monoclonal antibodies require FcγR engagement for full effect. One proven strategy to improve the efficacy of antibody therapies is to increase receptor binding affinity, in particular binding to FcγRIIIa/CD16a. Currently, affinities are measured using recombinantly-expressed soluble extracellular FcγR domains and CD16a-mediated antibody-dependent immune responses are characterized using cultured cells. It is notable that CD16a is highly processed with multiple N-glycosylation sites, and preventing individual N-glycan modifications affects affinity. Furthermore, multiple groups have demonstrated that CD16a N-glycan composition is variable and composition impacts antibody binding affinity. The level of N-glycosylation at each site is not known though computational prediction indicates low to moderate potential at each site based on primary sequence (40-70%). Here we quantify occupancy of the extracellular domains using complementary mass spectrometry-based methods. All five sites of the tighter-binding CD16a V158 allotype showed 65-100% N-glycan occupancy in proteomics-based experiments. These observations were confirmed using intact protein mass spectrometry that demonstrated the predominant species corresponded to CD16a V158 with five N-glycans, with a smaller contribution from CD16a with four N-glycans. Occupancy was likewise high for the membrane-bound receptor at all detected N-glycosylation sites using CD16a purified from cultured human natural killer cells. Occupancy of the N162 site, critical for antibody binding, appeared independent of N169 occupancy based on analysis of the T171A mutant protein. The weaker-binding CD16a F158 allotype showed higher occupancy of >93% at each site.

Glycosylation Profiling of Glycoproteins Secreted from Cultured Cells Using Glycan Node Analysis and GC-MS.

Glycan "node" analysis is the process by which pooled glycans within complex biological samples are chemically deconstructed in a way that facilitates the analytical quantification of uniquely linked monosaccharide units (glycan "nodes"). It is based on glycan methylation analysis (a.k.a. linkage analysis) that has historically been applied to pre-isolated glycans. Thus, when using glycan node analysis, unique glycan features within whole biospecimens such as "core fucosylation," "α2-6 sialylation," "ß1-6 branching," "ß1-4 branching," and "bisecting GlcNAc," are captured as single analytical signals by GC-MS. Here we describe the use of this methodology in cell culture supernatant and in the analysis of IgG (alpha-1 antitrypsin) glycans. The effect of IL-6 and IL-1ß cytokines on secreted hepatocyte protein glycan features is demonstrated; likewise, the impact of neuraminidase treatment of IgG is illustrated. For the majority of glycan nodes, the assay is consistent and reproducible on a day-to-day basis; because of this, relatively subtle shifts in the relative abundance of glycan features can be captured using this approach.

Oxidized-Desialylated Low-Density Lipoprotein Inhibits the Antitumor Functions of Lymphokine Activated Killer Cells.

Elevated concentrations of circulating low density lipoprotein (LDL) that is abnormally oxidized and desialylated is both a precursor to and a hallmark of atherosclerosis. Peripheral blood mononuclear cells (PBMCs) treated in vitro with interleukin-2 (IL-2) become lymphokine activated killer (LAK) cells, the primary effectors of which are NK cells and NKT cells. LAK cells display antitumor functions such as increased cytotoxicity and IFN-γ production, and they have been evaluated as a potential cancer therapeutic. Atherosclerotic processes may influence innate immunity against cancer. Because prior studies have shown that low density lipoprotein (LDL) reduces T-cell and NK cell antitumor functions, we asked whether oxidized-desialylated LDL affects the functionality of LAK cells in vitro. We show here that LAK cells take up oxidized-desialylated LDL to a significantly greater extent than native LDL over a period of 72 hours. This resulted in a significant downregulation of LAK cell cytotoxicity against K562 cells. In particular, the expression of IFN-γ, CD56, and NKG2D were reduced upon oxidized-desialylated LDL treatment of LAK cells and, conversely, their expression was enhanced with native LDL. It was also observed that as the number of CD56 and NKG2D positive cells decreased upon treatment with oxidized-desialylated LDL, the number of CD3 positive cells increased in proportion. Additionally, only a slight inhibition of LAK cell cytotoxicity was observed with desialylation alone of LDL, and no significant inhibition was observed with oxidation alone of LDL. Thus, this study describes a new role of oxidized-desialylated LDL as an inhibitor of the antitumor functions of LAK cells. These observations have implications for how atherosclerosis processes, namely oxidation and desialylation of LDL, may influence LAK cell antitumor activity.

Glycan Node Analysis of Plasma-Derived Extracellular Vesicles.

Blood plasma is a readily accessible source of extracellular vesicles (EVs), i.e., cell-secreted nanosized carriers that contain various biomolecules, including glycans. Previous studies have demonstrated that glycans play a major role in physiological and pathological processes, and certain plasma glycans have been associated with disease conditions. However, glycome studies have been limited by a lack of analytical techniques with the throughput capacity necessary to study hundreds of clinical samples. This study is the first to characterize the EV plasma glycome based on all major glycan classes. The results based on glycan node analysis revealed, as expected, that plasma-derived EVs have distinct glycan features from donor-matched whole plasma. Specifically, glycan nodes corresponding to those observed in chondroitin sulfate, dermatan sulfate, type I keratan sulfate, and type II keratan sulfate were enriched on EVs. The identification of specific differences in glycan features in plasma vs. plasma-derived EVs is relevant for understanding the physiological role of EVs and as a reference for future diagnostic studies. Additionally, the results indicate that EV glycan nodes do not substantially differ among a small set of healthy donors. These results lay the framework for the further evaluation of all EV glycan classes as diagnostic markers, therapeutic targets, and biologically active components in health and disease.