Microbial Activity at Gigapascal Pressures Anurag Sharma, James H. Scott, George D. Cody, Marilyn L. Fogel, Robert M. Hazen, Russell J. Hemley, and Wesley T. Huntress

Supplementary Material

1. Diamond anvil cells [W. A. Bassett, A. H. Shen, M. Bucknum, I. M. Chou, Rev. Sci. Instrum. 64, 2340 (1993); H. K. Mao, R. J. Hemley, in Rev. Mineral., R. J. Hemley, Ed. (Mineral. Soc. Am., Washington, DC) 37, 1-32 (1998)] enclose a sample chamber between two diamonds, which when squeezed can produce very high pressures on the sample. For biological work, these were modified by reducing the optical working distance for high magnification microscopic observations by decreasing the thickness of carbide supports for the diamonds and enlargement of the opening angle of the diamond anvil cell.

2. Shewanella oneidensis MR1 (formerly Shewanella putrefaciens MR1), a metal-reducing, facultative anaerobe [C. R. Meyers, K. H. Nealson, Science 240, 1319 (1988)] and Escherichia coli strain MG1655, were utilized in this study because of their ability to oxidize formate. Although both MR1 and MG1655 are non-piezophilic, the genus Shewanella appears to have some species predisposed to pressure tolerance [C. Kato, K. Nakasone, M. H. Qureshi, K. Horikoshi, in Env. Stressors and Gene Responses, K. B. Storey, J. Storey, Eds, 277-291 (2000)]. The inocula was grown overnight (16 to 24 hrs) to stationary phase at 30°C in Luria Bertani broth (LB) [J. Sambrook, E. Fritsch, T. Maniatis, Molecular Cloning: A Lab Manual (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1989)]. The cells were washed in 50 mM potassium phosphate buffer (pH 7.4) 3 times and resuspended in either a saturated solution of 2 M formate and 0.2 M sodium fumarate, 0.1 grams per liter of yeast, 0.2 M sodium formate, 0.2 sodium fumarate and 10 mM potassium phosphate (pH 7.4) or LB medium. The bacteria were immediately aseptically loaded in a Au-lined sample chamber and sealed between diamond anvils.

3. Raman spectra were obtained using 514-nm Ar+ laser excitation with about 30 mW laser power. Spectra were acquired with a TRIAX 550 spectrometer equipped with an ANDOR Peltier cooled (-90°C) charge-coupled device detector. Raman spectroscopy of the formic bearing solutions gave us a control on the extent of metabolic activity. Using the ratio of C-H stretching vibration peak for formic (2950 cm^-1) to the O-H stretching vibration peak for water (~3500 cm^-1), we were able to monitor the metabolic formic oxidation rate at various pressures. Pressures in the diamond anvil cells were determined using Raman peak shifts in quartz [K. J. Dean, W. F. Sherman, G. R. Wilkinson, Spectrochimica Acta 38A, 1105 (1982); R. J. Hemley, in High Pressure Research in Mineral Physics, M. H. Manghnani, Y. Syono, Eds, 347-359 (1987)] added as an internal calibrant, and are accurate to within 20 MPa. The pressures were typically increased very quickly (within 5-30 seconds) and allowed to equilibrate (for 5-60 minutes) before making measurements.

4. Cell numbers were obtained by direct counts at high pressure by analyzing 100 to 150 frames selected randomly through the sample. At t = 0 hr, the bacteria were observed through the entire depth of the sample chamber (E. coli: 200-micron thickness, 300-micron diameter; Shewanella MR1: 200-micron thickness, 500-micron diameter), while after 30 hrs only one distinct layer of cells was observed due to sedimentation. The total cell numbers were then calculated by integrating the average of each count over the entire volume of the sample chamber [total cell number = average cell number * (sample chamber volume/area of each frame)]. For E. coli: at 0 hour, average cells per frame (n = 11) = 44 (std = 9.26; variance = 21%); at 30 hrs, average cells per frame (n = 11) = 60 (std = 9.3; variance = 15%). For Shewanella MR1: at 0 hour, average cells per frame (n = 12) = 56 (std = 10.3; variance = 18%); at 30 hours, average cells per frame (n = 12) = 43 (std = 8.1; variance = 19%).

---

To view these movies, download a QuickTime viewer.

• Movie 1
  Digital movie segment of E. coli within fluid-inclusions in ice-VI at 1250 MPa.

• Movie 2
  Digital movie segment of E. coli at 1000 MPa, after depressurizing from 1600 MPa for >30 hours.

• Movie 3
  Digital movie segment of Shewanella MR1 at 1400 MPa in organic-rich fluid vein within ice-VI.