Activity profiling of natural compounds identifies potent chemotypes active in parental and drug-resistant lymphoma models

Background. Despite the big improvements achieved during the years, too many patients affected by lymphomas still succumb to their disease. Plants represent an extremely rich source of structurally diverse secondary metabolites, which can provide novel active compounds. Thus, we screened a curated library of 37 natural (mostly plant-derived) and nature-derived compounds in cell lines derived from marginal zone lymphoma (MZL), including those with acquired resistance to targeted agents. The library consisted of molecules selected through a cheminformatics approach for their diversity, and spanned 16 chemical classes, including anthranoids, flavonoids, chalcones, triterpenes, alkaloids, acetophenones, benzophenones, coumarins, and a dibenzofuran derivative. Methods. The library was screened for anti-proliferative activity in 6 MZL lymphoma-derived models (VL51, and 3 BTK/PI3K/BCL2 inhibitors (i) resistant; Karpas1718, and BTKi/PI3Ki resistant). Cells were treated with each compound at 0.4 µM, 4 µM, and 40 µM for 72 hours. Cell viability was assessed using an MTT assay, normalized to DMSO controls. Data were analyzed by class and by individual compound across doses and cell-line subgroups, using a ≥30% reduction in proliferation as the threshold for activity. Results. At 0.4 µM, most compounds displayed minimal inhibition, with only vismione B (anthranoid) showing partial activity. At 4 µM, modest reductions were observed in selected chalcones and in usnic acid (dibenzofuran). At 40 µM, clear class- and compound-dependent effects emerged. The most active compounds, osajin (flavonoid), ursolic acid (triterpene), usnic acid (dibenzofuran), cordoin, and 4,4′-O-methoxy-chalcone, reduced viability to <25% across all models. These compounds were also validated in a dose-dependent study. By class, chalcones, flavonoids and anthranoids were among the most active chemical classes. A comparison between parental and resistant sublines revealed that most active compounds retained their activity in both settings, indicating resistance-independent mechanisms of action. Some of the tested flavonoids showed preferential activity in VL51 rather than Karpas1718. Finally, selected compounds were tested against the CB-33 lymphoblastoid cell line, with a particular focus on usnic acid (benzofuran), which demonstrated more cytotoxicity in lymphoma cells than in the non-lymphoma cell line. Conclusions. This screen of structurally diverse natural compounds identified chalcones, flavonoids, and anthranoids as the most active classes in lymphoma models. The reproducible, dose-dependent activity of osajin, ursolic acid, usnic acid, cordoin, and 4,4′-O-methoxy-chalcone across all models highlights them as promising lead scaffolds for further mechanistic studies and medicinal chemistry optimization toward novel anti-lymphoma agents.