Yizhi Sheng

Life on Earth depends on N 2 ‐fixing microbes to make ammonia from atmospheric N 2 gas by the nitrogenase enzyme. Most nitrogenases use Mo as a cofactor; however, V and Fe are also possible. N 2 fixation was once believed to have evolved during the Archean‐Proterozoic times using Fe as a cofactor. However, δ 15 N values of paleo‐ocean sediments suggest Mo and V cofactors despite their low concentrations in the paleo‐oceans. This apparent paradox is based on an untested assumption that only soluble metals are bioavailable. In this study, laboratory experiments were performed to test the bioavailability of mineral‐associated trace metals to a model N 2 ‐fixing bacterium Azotobacter vinelandii . N 2 fixation was observed when Mo in molybdenite, V in cavansite, and Fe in ferrihydrite were used as the sole sources of cofactors, but the rate of N 2 fixation was greatly reduced. A physical separation between minerals and cells further reduced the rate of N 2 fixation. Biochemical assays detected five siderophores, including aminochelin, azotochelin, azotobactin, protochelin, and vibrioferrin, as possible chelators to extract metals from minerals. The results of this study demonstrate that mineral‐associated trace metals are bioavailable as cofactors of nitrogenases to support N 2 fixation in those environments that lack soluble trace metals and may offer a partial answer to the paradox.