Integrating a combinatorial structure database and feature-based molecular networking to identify mycosporine-like amino acids in the aerophytic green alga Apatococcus ammoniophilus

Apatococcus ammoniophilus is a recently identified green alga forming green biofilms on the bark and twigs of coniferous trees. It contains high amounts of an unknown mycosporine-like amino acid (MAA), a compound class known for its photoprotective, antioxidant, and wound-healing properties. To identify this MAA, a workflow combining UHPLC-VWD-HRMS2 analysis, feature-based molecular networking, and metabolite annotation using a rationally designed combinatorial database was developed. By leveraging the limited set of known MAA core structures and amino acid-related substituents, a theoretical compound library was created in silico and implemented into the bioinformatics platform SIRIUS. Applying the workflow to extracts of A. ammoniophilus and Stereum gausapatum, a fungus anticipated to contain the same or a similar MAA, enabled the putative annotation of several new MAAs and led to the isolation, structure elucidation, and biological activity evaluation of three compounds: mycosporine-serinol, gadusol, and the novel compound algasporine-glycine. Altogether, this study provides valuable insights into the MAA profile of an aerophytic microalga and establishes a broadly applicable workflow for the discovery and annotation of MAAs in natural extracts.