Events

Discovery of Nucleobase Amino Acid Metabolism Using AI and Ultrasensitive Metabolomics

Life's chemistry is built on two sequence-encoded molecular alphabets: amino acids and nucleic acids. Molecules that bridge these alphabets-nucleobase amino acids-could enable new forms of molecular recognition, cellular engineering, information storage, and catalysis, yet their availability has been limited by lengthy and costly chemical synthesis. During my postdoctoral work in the laboratory of George M. Church at Harvard and in collaboration with Regina Barzilay and Gregory Stephanopoulos at MIT, I uncovered enzymes for the de novo biosynthesis of nucleobase amino acids incorporating all five canonical nucleobases. By developing and combining a machine-learning pipeline for enzyme and cofactor discovery with ultrasensitive metabolomics for experimental validation, I identified and characterized nucleobase amino acid synthases that catalyze a previously unrecognized mode of C-N bond formation between O-acetylserine and free canonical nucleobases. Phylogenetic and metabolomic analyses suggest that nucleobase amino acid biosynthetic activity may be widespread across the three domains of life, revealing a previously unrecognized link between amino acid and nucleic acid metabolism. By uncovering nature's capacity to merge amino acids and nucleobases, my work opens new avenues toward nucleobase-programmable peptides, proteins, and hybrid biological codes.