Patient-derived organoids to study glycosylation dynamics during gastric disease

ABSTRACT BACKGROUND AND AIMS Aberrant cellular glycosylation remains a key event that accompanies and actively sustains gastric neoplastic transformation. Patient-derived organoids (PDOs) have recently emerged as a promising ex vivo model to study human gastric disorders. Since the PDOs glycosylation landscape remains unknown, this study aims to evaluate PDOs as potential avatars of in vivo tissue glycosylation profiles in the gastric context. METHODS Fresh gastric mucosa samples derived from non-tumoral obese patients (n=11), adjacent tumor mucosa samples (n=29), and tumor tissue samples derived from gastric cancer (GC) patients (n=30) were used to establish a biobank of gastric PDOs (n=56). The N - and O- glycophenotypes of normal, adjacent, and tumor PDOs and respective in vivo tissues were thoroughly characterized by immunostaining. Additionally, a comparative glycan analysis was performed over time, upon PDO biobanking and xenografting in mice. The binding of two Helicobacter pylori ( H. pylori ) isogenic strains with distinct glycan-binding affinities was assessed in parental gastric mucosa tissues and compared with the respective PDOs before and after modulation of their glycan landscape. RESULTS Our results show that PDOs mimic different phenotypes of the carcinogenic cascade and recapitulate parental gastric tissues’ glycosylation profile. Tumor PDOs recapitulate the inter- and intra-heterogeneity features observed in GC, which is maintained over time, upon biobanking and xenografting. We demonstrated that the expression of type I and type II Lewis antigens is dynamically controlled by PDOs differentiation status, which results in differential binding to H. pylori strains displaying distinct glycan-binding adhesins, mirroring the gastric epithelium tissue interactions. CONCLUSIONS This study established PDOs as invaluable ex vivo tools to study the complex glycan dynamics in both gastric physiological and pathological settings.