Abstract
N-glycolylneuraminic acid (Neu5Gc) is an immunogenic sugar of dietary origin that metabolically incorporates into diverse native glycoconjugates in humans. Anti-Neu5Gc antibodies are detected in all human sera, though with variable levels and epitope-recognition profiles. These antibodies likely play a role in several inflammation-mediated pathologies including cardiovascular diseases and cancer. In cancer, they have dualistic and opposing roles, either stimulating or repressing disease, as a function of their dose, and some of these antibodies serve as carcinoma biomarkers. Thus, anti-Neu5Gc antibodies may signify risk of inflammation-mediated diseases, and changes in their levels could potentially be used to monitor disease progression and/or response to therapy. Currently, it is difficult to determine levels of anti-Neu5Gc antibodies in individual human samples because these antibodies recognize multiple Neu5Gc-epitopes. Here we describe a simple and specific method for detection and overall estimation of human anti-Neu5Gc antibodies. We exploit the difference between two mouse models that differ only by Neu5Gc-presence (wild-type) or Neu5Gc-absence (Cmah(-/-) knockout). We characterize mouse serum from both strains by HPLC, lectin and mass-spectrometry analysis and show the target Neu5Gc-epitopes. We then use Cmah(-/-) knockout sera to inhibit all non-Neu5Gc-reactivity followed by binding to wild-type sera to detect overall anti-Neu5Gc response in a single assay. We applied this methodology to characterize and quantify anti-Neu5Gc IgG and IgA in sera of patients with Kawasaki disease (KD) at various stages compared to controls. KD is an acute childhood febrile disease characterized by inflammation of coronary arteries that untreated may lead to coronary artery aneurysms with risk of thrombosis and myocardial infarction. This estimated response is comparable to the average of detailed anti-Neu5Gc IgG profile analyzed by a sialoglycan microarray. Both assays revealed an elevated response in acute KD patients with normal coronaries compared to patients with aneurysm or dilated coronaries. Implications of these findings are discussed.
