Biotic and abiotic pathways of phosphorus cycling in minerals and sediments: insights from oxygen isotope ratios in phosphate.

A key question to address in the development of oxygen isotope ratios in phosphate (δ(18)O(p)) as a tracer of biogeochemical cycling of phosphorus in ancient and modern environments is the nature of isotopic signatures associated with uptake and cycling of mineral-bound phosphate by microorganisms. Here, we present experimental results aimed at understanding the biotic and abiotic pathways of P cycling during biological uptake of phosphate sorbed to ferrihydrite and the selective uptake of sedimentary phosphate phases by Escherichia coli and Marinobacter aquaeolei. Results indicate that a significant fraction of ferrihydrite-bound phosphate is biologically available. The fraction of phosphate taken up by E. coli attained an equilibrium isotopic composition in a short time (<50 h) due to efficient O-isotope exchange (between O in PO(4) and O in water; that is, actual breaking and reforming of P-O bonds) (biotic pathway). The difference in isotopic composition between newly equilibrated aqueous and residual sorbed phosphate groups promoted the ion exchange (analogous to isotopic mixing) of intact phosphate ions (abiotic pathway) so that this difference gradually became negligible. In sediment containing different P phases, E. coli extracted loosely sorbed phosphate first, whereas M. aquaeolei preferred Fe-oxide-bound phosphate. The presence of bacteria always imprinted a biotic isotopic signature on the P phase that was taken up and cycled. For example, the δ(18)O(p) value of loosely sorbed phosphate shifted gradually toward equilibrium isotopic composition. The δ(18)O(p) value of Fe-oxide-bound phosphate, however, showed only slight changes initially but, when new Fe-oxides were formed, coprecipitated/occluded phosphate retained δ(18)O values of the aqueous phosphate at the time of formation of new Fe oxides. Concentrations and isotopic compositions of authigenic and detrital phosphates did not change, suggesting that these phosphate phases were not utilized by bacteria. These findings support burgeoning applications of δ(18)O(p) as a tracer of phosphorus cycling in sediments, soils, and aquatic environments and as an indicator of paleo- environmental conditions.