9714214 Onstott Recent investigations have identified microbial communities in various crustal environments down to 2800 meters below the surface (mbls). Only a hand full of deep microbial samples from continental crust (>500 mbls.) exists, however, because coring is expensive. The gold mines of the 2.9 Ga Witwatersrand Supergroup in South Africa, however, provide a unique opportunity to study microbial communities at depths ranging from 2000 to 3500 mbls. And eventually up to 5000 mbls. Reconnaissance samples of a uranium-rich, gold-bearing, carbonaceous rock and of water from a gallery borehole were collected for microbial analyses from mined depth of 3200 mbls., where the rock and water temperatures were 50 to 55(C. Measures were taken to avoid contamination during mining and sampling. Samples were shipped to the U.S.A. in sterile, anaerobic canisters on ice, processed under sterile anaerobic conditions and distributed to other microbiology labs. Microscopic observations indicated the presence of intact cells in both types of samples. Electron microscopy reveals filamentous microorganisms in the rock samples. Phospholipid fatty acid and DNA analyses indicate that the rock samples contain Cyanobacteria and sulfate-reducing bacteria (SRB). Growth was detected in aerobic and anaerobic enrichments of the rock samples. The water sample yielded a strain of Thermus that is the first reported Thermus to reduce Fe (III) and the first facultative, thermophilic Fe (III) reducing bacteria (IRB). Funds are requested to return to South Africa, set up a sample-processing laboratory on site, and carry out a three-month field study to address the following issues raised by this sample reconnaissance: ( To what extent are rock and water samples contaminated by mining activity and what sampling strategies must be developed to reduce and quantify this contamination? Specifically, do the Cyanobacteria represent mining contamination? ( Does the radiation emanating from the uranium-rich, car bonaceous layer provide a significant source of energy for indigenous microbial communities? In particular, does the radiolytic reaction with water generate sufficient oxygen for the growth of facultative microorganisms like IRB-SA? ( If these microbial communities have been present in the rock since the time of the last thermal episode at 2.0 Ga, then does the in situ activity of the SRB and IRB explain the occurrence of framboidal pyrite and filamentous aggregates of gold? To address the first question, samples of mining water, borehole water, and filtered air samples will be collected and analyzed. For the second question, rock samples of the carbonaceous layer and host rock will be collected with aseptic methods in a three dimensional grid (3x3x12 meter) during excavation of an access tunnel. To answer the second and third question, the relationships between the mineralogy and the bacteria in the rocks will be examined in situ and the capability of the cultured microorganisms to precipitate minerals at in situ conditions will be examined. To constrain the habitation time of the microbial community, the present day and paleo hydrology will be modeled using thermal constraints provided by borehole temperatures and thermochronometry. Incubations of all samples will be initiated on site. Off site analyses during the subsequent 21 months will include 1) phospholipid, glycolipid, ether lipid analyses (Univ. of Tenn.); 2) DNA extraction, polymerase chain reaction (PCR) amplification, cloning, and sequencing (Pacific Northwest National Laboratories-PNNL); 3) acridine orange direct counts (AODC), in situ probe, and field emission gun scanning electron microscopy (FEG-SEM) (Princeton-Rutgers); 4) chemical and isotopic analyses (Princeton-Indiana University); 5) fission track apatite analyses (Princeton-Univ. of Penn.). 6) Phosphorimaging of SRB and uranium reducing activity (PNNL-Princeton).
