Sharma et al. used formate oxidation "as a probe of metabolic viability." Because the process of formate oxidation can take place with purified enzymes and cell extracts (6), however, their observations of formate oxidation up to a pressure of 324 ± 20 MPa simply show the occurrence of enzyme activity. Formate dehydrogenase activity to this pressure is in harmony with recent findings that formate dehydrogenase is active to pressures above 300 MPa (7). Although Sharma et al. referred to the cells as being viable and living, not a single observation they made supports that contention.

Sharma et al. did not test cells in any of their experiments to determine the fraction of the cells capable of forming colonies at atmospheric pressure following exposure to high pressures. The authors stated, for example, that "microscopic observations to pressures of 1060 MPa confirmed continued cell viability (intact and motile) over a duration of more than 30 days." In most cases, simple visual observations of bacteria provide insufficient and nonoperationally defined criteria of cell viability (8). If cells are intact, they may be dead (8). If cells appear motile, that may be due to a nonbiological cause. For most bacteria inhabiting atmospheric-pressure environments, exposure to pressures greater than around 50 MPa is inhibitory to growth. Lethal effects increase with increasing pressure and depend on the cell type, on the metabolic state of cells prior to compression, and on medium parameters at pressure. In 1914, experiments on the sterilization of foods (9) showed that several types of microorganisms survive or die during exposure to pressures as high as 690 MPa. There is now a large literature on prolonging shelf life by exposing foods to high pressure to attenuate the level of contaminant microorganisms (8, 10). In none of those reports is there evidence showing the growth of atmospheric-pressure inhabitants at pressures even approaching 100 MPa, let alone 1 gigapascal (GPa).

The experiments of Sharma et al. on ice formation at high pressure (1) were deficient. The authors provided no discussion of pressure gradients within the diamond anvil cell containing ice. Analysis was also lacking for the possibility that convection in the melting portions of the ice-liquid mixture led to particle motion. The use of apparent motility as a criterion of cell viability is unconvincing without supporting evidence, such as colony-forming ability by cells at high pressure and at 0.1 MPa following pressure treatment. Freezing and thawing at pressures above 300 MPa kills bacterial cells (11).

Finally, the speculation by Sharma et al. (1) that microorganisms are viable in Earth's subduction zones is incorrect. Because of the geothermal gradient, the temperature in a subduction zone (12) at depths where the pressure is 1 GPa far exceeds the permissible limit for life.

A. Aristides Yayanos
Scripps Institution of Oceanography
University of California, San Diego
La Jolla, CA 92093-0202, USA
E-mail: ayayanos{at}ucsd.edu

REFERENCES

Response: Our study (1) showed cell activity at gigapascal pressures using Raman spectroscopy and optical microscopy. With diamond anvil cells, we were able to characterize the microbe-fluid environment as a closed system with no pressure gradients or convective gradients, as evidenced by a simultaneous phase transformation throughout the whole sample. Raman spectroscopy was used to measure formate rates at various pressures. By including several controls (dead cells, cell-free experiments, and cyanide-inhibited cells), we demonstrated that the observed formate oxidation was primarily due to biological activity. Furthermore, optical microscopic observations showed cell motility at these high pressures and after depressurization. These observations were unique to the samples with live cells and did not occur in the controls with inorganic particles or those containing dead cells.

Although we appreciate the concerns expressed by Yayanos, they appear in most instances to reflect differences in semantics--primarily because Yayanos accepts cell replication as the only proof of viability. In our report, we included several in situ observations, such as biological formate oxidation rates at high pressures, as clues to continued microbial activity at these extreme pressures. We recognize that the study used novel methods and at present we have not focused on the recovery of cells. However, that does not make our in situ observations less relevant than those done using traditional approaches. Furthermore, several previous studies done at lower pressures (~600 to 800 MPa) have extracted and cultured Escherichia coli cells to show their viability at high pressures (2-4).

The comment by Yayanos suggesting that the observed formate oxidation could merely stem from enzymatic activity suggests that he has not considered all of the controls used in our study to distinguish microbial activity from enzymatic activity alone. We also note that the reason we used formate oxidation as the primary metabolic proxy is because the formate dehydrogenases employed by S. oneidensis and E. coli during anoxia are primary dehydrogenases--that is, they couple formate oxidation directly to the reduction of a terminal electron acceptor, using an Fe-S cluster containing proteins and b-type cytochromes as mediators for electron transport (5, 6), unlike the formate dehydrogenase mentioned in the study cited by Yayanos. We agree with Yayanos that experiments done in 1914 showed an increase in the shelf life of food upon pressure treatment. More recent studies, however, have established that E. coli can become resistant to such pressures (~600 MPa) and can continue to be viable even after such extreme pressure treatment (2-4).

Ultimately, the effects of pressure, volume, and temperature are intertwined. Even though temperature has not been examined as an implicit variable so far, it is not far-fetched to think that survival at higher temperatures may be favorable at higher pressures. Therefore, it is not idle speculation that life can be viable within cold subducting plate environments (7), where the geotherm conditions bring together potentially suitable conditions of pressure, temperature, and nutrients.

Anurag Sharma
James H. Scott
George D. Cody
Marilyn L. Fogel
Robert M. Hazen
Russell J. Hemley
Wesley T. Huntress
Geophysical Laboratory
Carnegie Institution of Washington
5251 Broad Branch Road, N.W.,
Washington, DC 20015, USA
E-mail: sharma{at}gl.ciw.edu

REFERENCES