The Evolution of the Sulfur Cycle

There are 2 principal avenues of inquiry relevant to reconstructing the history of the sulfur cycle. One avenue relies on the comparison of molecular sequences derived from biologically essential proteins and genetic material. Most sequence information is available from the small subunit of the rRNA molecule and from these sequences a Tree of Life emerges providing a reconstruction of the evolutionary relationships among organisms. Near the root of the tree are numerous bacteria1metabolizing sulfur species including organisms living from dissimilatory elemental sulfur reduction, dissimilatory sulfate reduction, and anoxygenic photosynthesis. These metabolisms are likely very ancient. Many of the deep-branching bacteria of the sulfur cycle are active at very high temperatures (hyperthermophiles) and are commonly found in modern sulfide-rich hydrothermal systems. One can imagine a primitive early Earth terrestrial ecosystem housed around active hydrothermal areas with anoxygenic photosynthesis producing organic matter and oxidized sulfur species. These oxidized sulfur species could have been used as electron acceptors in the mineralization of organic matter, completing the carbon cycle. The evolution of oxygenic photosynthesis provided for dramatically increased rates of carbon production, and a much wider range of ecosystems for both carbon production, and carbon oxidation. Either associated with, or following, the evolution of oxygenic photosynthesis is the emergence of lineages housing most of the bacteria of which we are familiar, including most of the bacteria of the sulfur cycle. The geologic record can provide direct evidence for the state of chemical oxidation of the Earth-surface, with possible indications of when specific bacterial metabolisms first occurred. We offer the following scenario for the evolution of the Earth-surface environment based on the available geological evidence. By 3.5 Ga anoxygenic photosynthesis was established and provided a weak source of sulfate to the global ocean with sulfate concentrations likely 91 mM. In some instances locally high concentrations of sulfate could accumulate and precipitated as evaporitic sulfate minerals. There is no compelling evidence for sulfate reduction at this time. The first evidence for sulfate reduction is found between 2.7 and 2.5 Ga, and the first evidence for oxygen production by oxygenic photosynthesis is found at around 2.8 Ga. Even so, levels of seawater sulfate remained low, below 1 mM, and did not increase to G1 mM until around 2.3 Ga. This increase in sulfate levels may have been promoted by a rise in atmospheric oxygen concentration at this time. Throughout the Archean and early Proterozoic the deep oceans contained appreciable concentrations of dissolved ferrous iron, and banded iron formations (BIFs) were a common form of chemical sediment. Sulfate levels increased slowly, and by 1.8 Ga sulfate concentrations were sufficient to increase rates of sulfate reduction to greater than the delivery flux of iron to the oceans. Sulfide accumulated and precipitated ferrous iron from solution. It is proposed that the oceans remained sulfide-rich until the Neoproterozoic, where renewed deposition of banded iron formations occurred at around 0.75 Ga. It is possible that during the Neoproterozoic, decreased carbon production resulted from an ice covered ‘‘Snowball Earth’’ reducing rates of sulfate reduction below rates of iron delivery to the oceans, promoting BIF formation. At around this time high carbon burial rates increased levels of atmospheric oxygen toG10 percent present-day levels, promoting the widespread oxidation of marine surface sediments and an evolutionary radiation of sulfide oxidizing bacteria. * Institute of Biology, Odense University, Campusvej 55, 5230 Odense M, Denmark. ** Department of Earth Sciences, Leeds University, Leeds LS2 9JT, United Kingdom. 1 Throughout this text the term bacteria refers to prokaryotic organisms from both of the Domains Bacteria and Archaea. When written as Bacteria, organisms of the Domain Bacteria are referred to. [AMERICAN JOURNAL OF SCIENCE, VOL. 299, SEPT./OCT./NOV., 1999, P. 697–723]