- Email: [email protected]
Survivability of Bacteria in Hypervelocity Impact
Icarus 154, 545–547 (2001) doi:10.1006/icar.2001.6738, available online at http://www.idealibrary.com on
NOTE Survivability of Bacteria in Hypervelocity Impact Mark J. Burchell and Jo Mann Centre for Astrophysics and Planetary Science, Physics Laboratory, University of Kent, Canterbury, Kent CT2 7NR, United Kingdom E-mail: [email protected]
and Alan W. Bunch and Pedro F. B. Brand˜ao Research School of BioSciences, University of Kent, Canterbury, Kent CT2 7NJ, United Kingdom Received March 5, 2001; revised August 8, 2001
Bacteria belonging to the genus Rhodococcus have been tested for their survivability in hypervelocity impacts at 5.1 ± 0.1 km s−1 . This is similar to the martian escape velocity for example but is slower than the mean velocities typical of impacts from space on planets like Mars (typically 14 km s−1 ) and Earth (typically 20– 25 km s−1 ). The bacteria fired were loaded on a projectile using a two-stage light-gas gun. The targets were plates of nutrient media. Analysis techniques including pyrolysis mass spectrometry and selective growth in acetonitrile confirmed that the bacterium grown on a target plate after a shot was the original strain. The indication is that, if fired on a projectile, bacteria can survive a hypervelocity impact and subsequently grow. This holds implications for the study of possible natural migration of life around the Solar System on minor bodies which end up impacting target planets, thus transferring life if the bacteria can survive the resulting hypervelocity impact. c 2001 Elsevier Science (USA) Key Words: exobiology; impact processes; meteorites.
1. Introduction. A topic of perennial interest is the possibility of life naturally migrating around the Solar System. This has been present in scientific literature for over 100 years and regularly reappears (e.g., Hoyle 1998). More sophisticated arguments exist that it may not so much be life that migrates through space as the necessary ingredients for a biosphere (Delsemme 2000). If however the discussion concerns life itself, then a common assumption is that life (a bacterium) is traveling on a small parent body which may have originated in space (e.g., a comet) or may have been thrown off a planetary surface at some stage in the past. Discussions of this type (e.g., Mileikowsky et al. 2000) emphasize the issue of survivability of the bacteria in space (e.g., hard vacuum, the radiation environment, the length of time spent in space, etc.) as well as the acceleration that occurs upon launch into space and arrival at a “new home.” It should be stressed however, that the acceleration/deceleration is important, as escape velocities are high and orbital mechanics indicates that the impact at the destination is also at very high speed. For example, material arriving at Earth from interplanetary space typically does so at 20 to 25 km s−1 (Hughes and Williams 2000), and if only near-Earth asteroids are considered the mean impact speeds on Earth and Mars are 17.2 and 13.6 km s−1 respectively (Bottke et al. 1994). It should be noted, however, that these values are mean impact speeds. The distribution of
impact speeds is quite wide and planetary impacts at slower speeds are thus not excluded, although in this context the Earth and martian escape velocities (11.1 and 5.03 km s−1 respectively) should be noted to set a minimum scale. Nevertheless, impacts at speeds of more than a few kilometers per second are termed hypervelocity impacts and are commonly held to involve at least some, if not near complete, vaporization of the projectile. However, a planet such as the Earth possesses an atmosphere which can help decelerate small bodies. The results are the well-known influx of dust into the Earth’s atmosphere (Love and Brownlee 1993), the collection of small meteorites on the Earth’s surface, and the explosion of some larger bodies in the atmosphere due to the shock they undergo during descent (e.g., Tunguska) (Bronshten 1999). Although destructive of the parent body, the latter process can shower the region below with small components of the initial body. The evidence of elevated temperatures in meteorites due to their heating during capture in the atmosphere (e.g., formation of a fusion crust etc.) can be confined to just the few millimeters below their surface (Sears 1975, Genge and Grady 1999). Indeed, recent work (Weiss et al. 2000) on the martian meteorite ALH84001 indicates that throughout the whole cycle of ejection from the martian surface and arrival on the Earth, parts of the meteorite’s interior were not elevated in temperature by more than 40◦ C. Thus the earth and similar planets have accumulated relatively unprocessed material from space via these mechanisms. If life can survive the transfer through space from another source of life then the next question is, Can it survive the impact upon arrival? Traditionally this question is not well highlighted. Searches have concentrated on finding evidence for amino acids in meteorites (Cronin 1989), modeling whether amino acids in comets and asteroids could survive a hypervelocity impact (Pierazzo and Chyba 1999), or jumping straight to looking for fossil evidence of life (e.g., on martian meteorites) (McKay et al. 1996). Laboratory tests of survivability of bacteria have focused on use of ballistic velocities, i.e., at a few hundreds of meters per second (Mastrapa et al. 2000, Mileikowsky et al. 2000), use of centrifuges to simulate high accelerations but for relatively long duration (e.g., 4.27 × 106 m s−2 with a rise time of 2 min by Mastrapa et al. 2000), or flyer plate experiments to generate peak shocks at some tens of GPa (Horneck et al. 2001). While revealing (all provide positive results for survivability), none of these actually involves a hypervelocity impact of bacteria or spores onto a target. It is thus interesting to directly test in the laboratory the plausibility of bacteria surviving hypervelocity impacts. Earlier papers (Burchell et al. 2000, 2001) have reported on impacts at 4–5 km s−1 , where bacteria were loaded onto projectiles and fired at a variety of solid targets (metal, glass, rock). In no case was bacteria recovered from the impact site. However, these papers do intriguingly report that bacteria launched on the 545 0019-1035/01 $35.00 c 2001 Elsevier Science (USA) All rights reserved.
546
BURCHELL ET AL.
projectile were recovered mixed with the ejecta from the crater (an ejecta capture system was included in that work). This suggests that somehow the bacteria transferred from the projectile directly to the ejecta. If correct, this at the very least suggests that bacteria were successfully launched alive in a light-gas gun. The question that then arises is whether it is possible to capture live bacteria at the impact site itself, this being an unambiguous test of whether bacteria can survive an impact event. There is, however, no need to use solid targets. An ideal target would be one where the bacteria could grow in situ after the impact, e.g., nutrient gel in agar plates. Accordingly, this work presents a program of shots (at typically 5 km s−1 ) of bacteria-laden projectiles into nutrient gel. 2. Method. The bacterium used in this work was Rhodococcus erythropolis, which is able to grow in various nitrile compounds. This is a hardy bacterium that is very resistant to many breakage methods. Rhodococci are known to survive on Earth in relatively harsh environments, including deep-sea sediments at depths of up to 6475 m (Colquhoun et al. 1998). They are 1 to 2 µm in length and colonies of these bacteria are reddish in color. The strain used here was DSM 13002 and was obtained from the German culture collection for microorganisms. The samples used were from a late logarithmic stage of a bacteria culture. DSM 13002 is a bacteria that is resistant to many chemicals that are considered toxic and has a strong wall structure. For an extensive discussion of this bacterium see Goodfellow and Alderson (1998) and successive papers in the same journal. The projectiles used in this work were pieces of ceramic, which were washed in alcohol and baked at 100◦ C before use. The bacteria were loaded onto the porous ceramic by soaking the ceramic in heavily concentrated suspensions of the bacteria. After soaking, several pieces of ceramic, each approximately 100– 200 µm across were placed in the disposable sabot of a two-stage light-gas gun (Burchell et al. 1999). When fired, the gun projected the sabot at speeds of typically 4 or 5 km s−1 . The sabot was discarded in flight and only the projectiles mounted within it proceeded to the target. The speed of flight was measured by passage of the projectiles through two laser light curtains positioned with known separation along the line of flight. In the worst case the velocity is accurate to ±4%. During a shot the range of the gun, including target chamber, was maintained at a vacuum of typically 0.2–1.0 mbar, to avoid slowing of the projectiles in flight. The targets used in this work were plates of glucose yeast extract medium (GYE). The plates were 10 cm across and filled to a depth of typically 5 mm. Tests showed that the plates of GYE resisted the exposure to low pressure without significant loss of mass. During the impact work staff wore protective clothing including face masks and sterile gloves. Surfaces were washed with ethanol to reduce risk of contamination. Separate staff handled the projectiles and target plates. Agar plates were kept in clean air cupboards and sealed with laboratory sealing film when not in use in the gun. Standard biological practice was followed during the culturing and subsequent analysis of plates after the shots. After each shot the surface of the agar around the impact sites was scored with a sterile blade to permit oxygen to reach the trapped projectiles. The plates were then incubated at a temperature of 30◦ C for 5 days to permit colony growth to occur. When colonies were found they were restreaked onto new GYE agar plates to check their purity. Standard gram staining tests were performed on pure cultures. These cultures were subject to pyrolysis mass spectrometry (PyMS) analysis (Goodacre and Kell 1996, Colquhoun et al. 2000, Bull et al. 2000), to generate a phenotypic database and sort the putative rhodococci into clusters alongside reference strains of Rhodococcus species. Pure cultures were finally inoculated in liquid mineral medium supplemented with 20 mM acetonitrile and incubated at 30◦ C at 200 rpm. This was done to selectively grow the nitriledegrading Rhodococcus erythropolis strain. In addition, tests were carried out without firing the gun to investigate the possibility of accidental, nonimpact related transport of Rhdococcus to the target. First, doped projectiles were loaded into the gun and then unloaded manually and cultured. Prolific Rhodococcus growth was obtained, indicating that the bacteria were present on the projectiles when loaded in the gun. Next, the whole prefiring procedure, involving mounting a target in the target chamber, loading doped projectiles in the gun, evacuating and repressurizing the target chamber, etc. was carried out. The sabot was then unloaded manually from the barrel end of the gun and the projectiles were inspected. Again, prolific bacteria growth was obtained
TABLE I Results of the Shot Program
Type
Impact velocity (km s−1 )
Original growth visibility
Gram stain
Acetonitrile growth
PyMS (% similarity to R. erythropolis type strain)
Control Control Control Control Control Control Doped Doped Doped Doped Doped Doped Doped
5.24 5.20 5.00 5.01 5.00 5.00 5.24 5.20 4.98 5.00 5.30 5.00 5.20
No No No No No No No No No No Yes Yes Yes
— — — — — — — — — — Negative Negative Positive
— — — — — — — — — — No No Yes
— — — — — — — — — — 42 42 100
from the unshot doped projectiles. In addition, the target in this “nonshot” was allowed to incubate, but it did not show any evidence of Rhodococcus growth. 3. Results. Two types of shot were carried out: controls (i.e., with clean projectiles) and doped shots (i.e., with projectiles loaded with Rhodococcus). An alternating sequence of doped and control shots was followed. The results of 13 shots, 6 controls and 7 doped, are reported here (Table I). Following each shot the first test checked if any Rhodococcus-like colony was visible on the target plates after incubation (“Original growth visibility” in Table I). No control shot produced such a result; however three of the doped shots did. It should be noted that these three plates took between 5 and 7 days to show significant growth, whereas Rhodococcus bacteria subcultured directly to a plate grew well within 3–4 days. Next gram staining was carried out on the colonies of the three plates, with the results given in Table I. This staining technique is a general procedure used to differentiate bacteria into two groups based on the structure of their cell walls (e.g., see Gerhardt 1981 for more details). It is thus a differential test permitting exclusion of bacteria of the wrong type. The Rhodococcus bacteria used in the shots were gram positive. In general most strains of bacteria are gram negative. Only one of the colonies tested gave a positive result. All three of the colonies identified by visible growth characteristics then underwent the selective growth test in 20 mM acetonitrile. Growth was observed in only one of the three bacterial strains recovered from the target plates. This was the one which had the (correct) gram positive result. Finally, all three colonies identified by visual inspection on the original plates were subject to PyMS. The two gram negative samples (which failed to grow on acetonitrile) grouped together in a cluster with a 42% similarity to the Rhodococcus erythropolis reference strain. In such similarity analyses this is considered as a clear failure to match the test strain. A successful match requires a similarity of at least 95% to be of interest, and preferably 100% for a positive match. The gram positive colony which successfully grew in acetonitrile had 100% similarity to the original bacterium R. erythropolis. This sequence of tests confirms the identity of the recovered colony with that used in the shot program and eliminates the false results. Estimating the rate of bacteria survivability involves determining several factors. By assuming that 10% of a doped projectile’s volume was bacteria, that a bacterium was rod shaped (1 × 1 × 2 microns in size), and that 10 projectile pieces (each 100 µm across) per shot hit the target (a typical value from visual examination of the plates after impact), then with seven doped shots fired, the number of bacteria used in the program was approximately 7,000,000 (i.e., of order 107 ). With survival observed in just one shot and assuming the successful colony grew from just one bacterium, the survival rate was thus of order 1 in 107 . This should be compared to the results of Horneck et al. (2001) who found survival rates of up to 1 in 104 for spores subject to shocks in the GPa range.
NOTE 4. Discussion. The result that bacteria growth was observed after impact is remarkable. Previous studies of survivability (Mastrapa et al. 2000, Horneck et al. 2001) have concentrated on spores, rather than bacteria themselves. This is partly because the spore is widely held to be hardier than a bacterium, and so more likely to survive an impact event, and also because during the transfer between planets spores are considered more likely to survive than bacteria. However, this study shows that bacteria themselves can survive hypervelocity impacts. Recent work by Mastrapa et al. (2000) argues that launch into space from the martian surface (carried on ejecta following an impact from space) can involve an acceleration of 3 × 106 m s−2 , with a duration of 0.5 × 10−3 s and a change in acceleration (or “jerk”) of 6 × 109 m s−3 . In the work here, assuming the projectile accelerates uniformly over the length of the launch tube in the gun, the acceleration in the firing of the gun is 1.8 × 107 m s−2 , with a duration of 0.28 × 10−3 s, corresponding to a jerk of 6 × 1010 m s−3 . These conditions exceed the requirements for launch to escape velocity from the martian surface. Furthermore, the deceleration in the impact event is even greater than the original acceleration and the time scale is even shorter. If a crater depth of 2 mm is taken for impacts in the nutrient gel (this was typical of the depth of the pit in which captured projectiles sat) the figures are deceleration of 6.25 × 109 m s−2 and duration of 0.8 × 10−6 s, corresponding to a jerk of 7.8 × 1015 m s−3 . 5. Conclusion. The results suggest that bacteria can successfully be accelerated to speeds of the order of 5 km s−1 and survive the subsequent hypervelocity impact with a target. The implications of such hypervelocity impact survivability are potentially significant. Results from a wide range of sources are systematically addressing the issue of whether bacteria can survive severe radiation environments and vacuum and live (Vreeland et al. 2000) for the long periods typically associated with objects drifting in the inner Solar System (Gladman 1997). Taken together, these results, plus those reported here for survivability in a hypervelocity impact, are starting to form a picture in which it is no longer totally implausible that bacteria can naturally migrate through space. Acknowledgment. We thank M. Cole for operation of the light-gas gun, whose use is financed by the Particle Physics and Astronomy Research Council (United Kingdom), Professor Alan T. Bull of the Biosciences Department at the University of Kent for permitting the PyMS work, and Professor Christoph Syldatk, Institute of Biochemical Engineering, University of Stuttgart, for providing access to the Rhodococcus strains used in this study.
REFERENCES Bottke, W. F., M. C. Nolan, R. Greenberg, and R. A. Kolvoord 1994. Collisional lifetimes and impact statistics of near-Earth asteroids. In Hazards due to Comets and Asteroids (T. Gehrels, Ed.), pp. 337–357. Univ. of Arizona Press, Tucson. Bronshten, V. A. 1999. The nature of the Tunguska meteorite. Meteorit. Planet. Sci. 34, 723–728. Bull, A. T., A. C. Ward, and M. Goodfellow 2000. Search and discovery strategies for biotechnology: The paradigm shift. Microbiol. Mol. Biol. Revi. 64, 573– 606. Burchell, M. J., M. J. Cole, J. A. M. McDonnell, and J. C. Zarnecki 1999. Hypervelocity impact studies using the 2 MV Van de Graaff accelerator and two-stage light gas gun of the University of Kent. Measurement Sci. Technol. 10, 41–50. Burchell, M. J., N. R. G. Shrine, A. Bunch, and J. C. Zarnecki 2000. Exobiology: Laboratory tests of the impact related aspects of Panspermia. In Impacts and the Early Earth (I. Gilmour and C. Koeberl, Eds.), pp. 1–26. Springer-Verlag, Berlin.
547
Burchell, M. J., N. R. G. Shrine, J. Mann, A. W. Bunch, P. Brand˜ao, J. C. Zarnecki, and J. A. Galloway 2001. Laboratory investigations of the survivability of bacteria in hypervelocity impacts. Adv. Space Res., in press. Colquhoun, J. A., S. C. Heald, J. Tamaoka, C. Kato, K. Horikoshi, and A. T. Bull 1998. Taxonomy and biotransformation activities of some deep-sea actinomycetes. Extremophiles 2, 269–277. Colquhoun, J. A., J. Zulu, M. Goodfellow, K. Horikoshi, A. C. Ward, and A. T. Bull 2000. Rapid characterisation of deep-sea actinomycetes for biotechnology screening. Antonie van Leeuwenhoek Int. J. Gen. Mol. Microbiol. 77, 359–367. Cronin, J. R. 1989. Origin of organic compounds in carbonaceous chondrites. Adv. Space Res. 9, 54–64. Delsemme, A. H. 2000. Cometary origin of the biosphere. Icarus 146, 313– 325. Genge, M. J., and M. M. Grady 1999. The fusion crusts of stony meteorites: Implications for the atmospheric reprocessing of extraterrestrial materials. Meteorit. Planet. Sci. 34, 341–356. Gerhardt, P. 1981. Chapter 3. In Manual Methods for General Biochemistry, pp. 21–34. American Society for Microbiology, Washington, DC. Gladman, B. 1997. Destination Earth, martian meteorite delivery. Icarus 130, 228–246. Goodacre, R., and D. B. Kell 1996. Pyrolysis mass spectrometry and its applications in biotechnology. Curr. Opinions Biotechnol. 7, 20. Goodfellow, M., and G. Alderson 1998. Editorial. Antonie van Leeuwenhoek Int. J. Gen. Mol. Microbiol. 74, 1. Hoyle, F. 1998. Comets: A matter of life or death. In The Universe Unfolding (H. Bondi and M. Weston-Smith, Eds.), pp. 3–22. Oxford Univ. Press, London. Horneck, G., D. St¨offler, U. Eschweiler, and U. Hornemann 2001. Bacterial spores survive simulated meteorite impact. Icarus 149, 285–290. Hughes, D. W., and I. P. Williams 2000. The velocity distributions of periodic comets and stream meteoroids. Mon. Not. R. Astron. Soc. 315, 629–634. Love, S. G., and D. E. Brownlee 1993. A direct measurement of the terrestrial mass accretion rate of cosmic dust. Science 262, 550–553. Mastrapa, R. M. E., H. Glanzberg, J. N. Head, H. J. Melosh, and W. L. Nicholson 2000. Survival of Bacillus subtilis spores and Deinococcus radiodurans cells exposed to extreme acceleration and shock predicted during planetary ejection. Lunar Planet. Sci. 31, 2045 (abstract). McKay, D. S., E. K. Gibson, Jr., K. L. Thomas-Keprta, H. Vali, C. S. Romanek, S. J. Clemett, X. D. F. Chillier, C. R. Maechling, and R. N. Zare 1996. Search for past life on Mars: Possible relic biogenic activity in martian meteorite ALH84001. Science 273, 924–930. Mileikowsky, C., F. A. Cucinotta, J. W. Wilson, B. Gladman, G. Horneck, L. Lindegren, J. Melosh, H. Rickman, M. Valtonen, and J. Q. Zheng 2000. Natural transfer of viable microbes in space. Icarus 145, 391– 427. Pierazzo, E., and F. Chyba 1999. Amino acid survivability in large cometary impacts. Meteorit. Planet. Sci. 34, 909–918. Sears, D. W. 1975. Temperature gradients in meteorites produced by heating during atmospheric passage. Mod. Geol. 5, 155–164. Vreeland, R. H., W. D. Rosenzweig, and D. W. Powers 2000. Isolation of a 250 million year old halotolerant bacterium from a primary salt crystal. Nature 407, 897–900. Weiss, B. P., J. L. Kirschvink, F. J. Baudenbacher, H. Vali, N. T. Peters, F. A. Macdonald, and J. P. Wikswo 2000. A low temperature transfer of ALH84001 from Mars to Earth. Science 290, 791–795.
NOTE Survivability of Bacteria in Hypervelocity Impact Mark J. Burchell and Jo Mann Centre for Astrophysics and Planetary Science, Physics Laboratory, University of Kent, Canterbury, Kent CT2 7NR, United Kingdom E-mail: [email protected]
and Alan W. Bunch and Pedro F. B. Brand˜ao Research School of BioSciences, University of Kent, Canterbury, Kent CT2 7NJ, United Kingdom Received March 5, 2001; revised August 8, 2001
Bacteria belonging to the genus Rhodococcus have been tested for their survivability in hypervelocity impacts at 5.1 ± 0.1 km s−1 . This is similar to the martian escape velocity for example but is slower than the mean velocities typical of impacts from space on planets like Mars (typically 14 km s−1 ) and Earth (typically 20– 25 km s−1 ). The bacteria fired were loaded on a projectile using a two-stage light-gas gun. The targets were plates of nutrient media. Analysis techniques including pyrolysis mass spectrometry and selective growth in acetonitrile confirmed that the bacterium grown on a target plate after a shot was the original strain. The indication is that, if fired on a projectile, bacteria can survive a hypervelocity impact and subsequently grow. This holds implications for the study of possible natural migration of life around the Solar System on minor bodies which end up impacting target planets, thus transferring life if the bacteria can survive the resulting hypervelocity impact. c 2001 Elsevier Science (USA) Key Words: exobiology; impact processes; meteorites.
1. Introduction. A topic of perennial interest is the possibility of life naturally migrating around the Solar System. This has been present in scientific literature for over 100 years and regularly reappears (e.g., Hoyle 1998). More sophisticated arguments exist that it may not so much be life that migrates through space as the necessary ingredients for a biosphere (Delsemme 2000). If however the discussion concerns life itself, then a common assumption is that life (a bacterium) is traveling on a small parent body which may have originated in space (e.g., a comet) or may have been thrown off a planetary surface at some stage in the past. Discussions of this type (e.g., Mileikowsky et al. 2000) emphasize the issue of survivability of the bacteria in space (e.g., hard vacuum, the radiation environment, the length of time spent in space, etc.) as well as the acceleration that occurs upon launch into space and arrival at a “new home.” It should be stressed however, that the acceleration/deceleration is important, as escape velocities are high and orbital mechanics indicates that the impact at the destination is also at very high speed. For example, material arriving at Earth from interplanetary space typically does so at 20 to 25 km s−1 (Hughes and Williams 2000), and if only near-Earth asteroids are considered the mean impact speeds on Earth and Mars are 17.2 and 13.6 km s−1 respectively (Bottke et al. 1994). It should be noted, however, that these values are mean impact speeds. The distribution of
impact speeds is quite wide and planetary impacts at slower speeds are thus not excluded, although in this context the Earth and martian escape velocities (11.1 and 5.03 km s−1 respectively) should be noted to set a minimum scale. Nevertheless, impacts at speeds of more than a few kilometers per second are termed hypervelocity impacts and are commonly held to involve at least some, if not near complete, vaporization of the projectile. However, a planet such as the Earth possesses an atmosphere which can help decelerate small bodies. The results are the well-known influx of dust into the Earth’s atmosphere (Love and Brownlee 1993), the collection of small meteorites on the Earth’s surface, and the explosion of some larger bodies in the atmosphere due to the shock they undergo during descent (e.g., Tunguska) (Bronshten 1999). Although destructive of the parent body, the latter process can shower the region below with small components of the initial body. The evidence of elevated temperatures in meteorites due to their heating during capture in the atmosphere (e.g., formation of a fusion crust etc.) can be confined to just the few millimeters below their surface (Sears 1975, Genge and Grady 1999). Indeed, recent work (Weiss et al. 2000) on the martian meteorite ALH84001 indicates that throughout the whole cycle of ejection from the martian surface and arrival on the Earth, parts of the meteorite’s interior were not elevated in temperature by more than 40◦ C. Thus the earth and similar planets have accumulated relatively unprocessed material from space via these mechanisms. If life can survive the transfer through space from another source of life then the next question is, Can it survive the impact upon arrival? Traditionally this question is not well highlighted. Searches have concentrated on finding evidence for amino acids in meteorites (Cronin 1989), modeling whether amino acids in comets and asteroids could survive a hypervelocity impact (Pierazzo and Chyba 1999), or jumping straight to looking for fossil evidence of life (e.g., on martian meteorites) (McKay et al. 1996). Laboratory tests of survivability of bacteria have focused on use of ballistic velocities, i.e., at a few hundreds of meters per second (Mastrapa et al. 2000, Mileikowsky et al. 2000), use of centrifuges to simulate high accelerations but for relatively long duration (e.g., 4.27 × 106 m s−2 with a rise time of 2 min by Mastrapa et al. 2000), or flyer plate experiments to generate peak shocks at some tens of GPa (Horneck et al. 2001). While revealing (all provide positive results for survivability), none of these actually involves a hypervelocity impact of bacteria or spores onto a target. It is thus interesting to directly test in the laboratory the plausibility of bacteria surviving hypervelocity impacts. Earlier papers (Burchell et al. 2000, 2001) have reported on impacts at 4–5 km s−1 , where bacteria were loaded onto projectiles and fired at a variety of solid targets (metal, glass, rock). In no case was bacteria recovered from the impact site. However, these papers do intriguingly report that bacteria launched on the 545 0019-1035/01 $35.00 c 2001 Elsevier Science (USA) All rights reserved.
546
BURCHELL ET AL.
projectile were recovered mixed with the ejecta from the crater (an ejecta capture system was included in that work). This suggests that somehow the bacteria transferred from the projectile directly to the ejecta. If correct, this at the very least suggests that bacteria were successfully launched alive in a light-gas gun. The question that then arises is whether it is possible to capture live bacteria at the impact site itself, this being an unambiguous test of whether bacteria can survive an impact event. There is, however, no need to use solid targets. An ideal target would be one where the bacteria could grow in situ after the impact, e.g., nutrient gel in agar plates. Accordingly, this work presents a program of shots (at typically 5 km s−1 ) of bacteria-laden projectiles into nutrient gel. 2. Method. The bacterium used in this work was Rhodococcus erythropolis, which is able to grow in various nitrile compounds. This is a hardy bacterium that is very resistant to many breakage methods. Rhodococci are known to survive on Earth in relatively harsh environments, including deep-sea sediments at depths of up to 6475 m (Colquhoun et al. 1998). They are 1 to 2 µm in length and colonies of these bacteria are reddish in color. The strain used here was DSM 13002 and was obtained from the German culture collection for microorganisms. The samples used were from a late logarithmic stage of a bacteria culture. DSM 13002 is a bacteria that is resistant to many chemicals that are considered toxic and has a strong wall structure. For an extensive discussion of this bacterium see Goodfellow and Alderson (1998) and successive papers in the same journal. The projectiles used in this work were pieces of ceramic, which were washed in alcohol and baked at 100◦ C before use. The bacteria were loaded onto the porous ceramic by soaking the ceramic in heavily concentrated suspensions of the bacteria. After soaking, several pieces of ceramic, each approximately 100– 200 µm across were placed in the disposable sabot of a two-stage light-gas gun (Burchell et al. 1999). When fired, the gun projected the sabot at speeds of typically 4 or 5 km s−1 . The sabot was discarded in flight and only the projectiles mounted within it proceeded to the target. The speed of flight was measured by passage of the projectiles through two laser light curtains positioned with known separation along the line of flight. In the worst case the velocity is accurate to ±4%. During a shot the range of the gun, including target chamber, was maintained at a vacuum of typically 0.2–1.0 mbar, to avoid slowing of the projectiles in flight. The targets used in this work were plates of glucose yeast extract medium (GYE). The plates were 10 cm across and filled to a depth of typically 5 mm. Tests showed that the plates of GYE resisted the exposure to low pressure without significant loss of mass. During the impact work staff wore protective clothing including face masks and sterile gloves. Surfaces were washed with ethanol to reduce risk of contamination. Separate staff handled the projectiles and target plates. Agar plates were kept in clean air cupboards and sealed with laboratory sealing film when not in use in the gun. Standard biological practice was followed during the culturing and subsequent analysis of plates after the shots. After each shot the surface of the agar around the impact sites was scored with a sterile blade to permit oxygen to reach the trapped projectiles. The plates were then incubated at a temperature of 30◦ C for 5 days to permit colony growth to occur. When colonies were found they were restreaked onto new GYE agar plates to check their purity. Standard gram staining tests were performed on pure cultures. These cultures were subject to pyrolysis mass spectrometry (PyMS) analysis (Goodacre and Kell 1996, Colquhoun et al. 2000, Bull et al. 2000), to generate a phenotypic database and sort the putative rhodococci into clusters alongside reference strains of Rhodococcus species. Pure cultures were finally inoculated in liquid mineral medium supplemented with 20 mM acetonitrile and incubated at 30◦ C at 200 rpm. This was done to selectively grow the nitriledegrading Rhodococcus erythropolis strain. In addition, tests were carried out without firing the gun to investigate the possibility of accidental, nonimpact related transport of Rhdococcus to the target. First, doped projectiles were loaded into the gun and then unloaded manually and cultured. Prolific Rhodococcus growth was obtained, indicating that the bacteria were present on the projectiles when loaded in the gun. Next, the whole prefiring procedure, involving mounting a target in the target chamber, loading doped projectiles in the gun, evacuating and repressurizing the target chamber, etc. was carried out. The sabot was then unloaded manually from the barrel end of the gun and the projectiles were inspected. Again, prolific bacteria growth was obtained
TABLE I Results of the Shot Program
Type
Impact velocity (km s−1 )
Original growth visibility
Gram stain
Acetonitrile growth
PyMS (% similarity to R. erythropolis type strain)
Control Control Control Control Control Control Doped Doped Doped Doped Doped Doped Doped
5.24 5.20 5.00 5.01 5.00 5.00 5.24 5.20 4.98 5.00 5.30 5.00 5.20
No No No No No No No No No No Yes Yes Yes
— — — — — — — — — — Negative Negative Positive
— — — — — — — — — — No No Yes
— — — — — — — — — — 42 42 100
from the unshot doped projectiles. In addition, the target in this “nonshot” was allowed to incubate, but it did not show any evidence of Rhodococcus growth. 3. Results. Two types of shot were carried out: controls (i.e., with clean projectiles) and doped shots (i.e., with projectiles loaded with Rhodococcus). An alternating sequence of doped and control shots was followed. The results of 13 shots, 6 controls and 7 doped, are reported here (Table I). Following each shot the first test checked if any Rhodococcus-like colony was visible on the target plates after incubation (“Original growth visibility” in Table I). No control shot produced such a result; however three of the doped shots did. It should be noted that these three plates took between 5 and 7 days to show significant growth, whereas Rhodococcus bacteria subcultured directly to a plate grew well within 3–4 days. Next gram staining was carried out on the colonies of the three plates, with the results given in Table I. This staining technique is a general procedure used to differentiate bacteria into two groups based on the structure of their cell walls (e.g., see Gerhardt 1981 for more details). It is thus a differential test permitting exclusion of bacteria of the wrong type. The Rhodococcus bacteria used in the shots were gram positive. In general most strains of bacteria are gram negative. Only one of the colonies tested gave a positive result. All three of the colonies identified by visible growth characteristics then underwent the selective growth test in 20 mM acetonitrile. Growth was observed in only one of the three bacterial strains recovered from the target plates. This was the one which had the (correct) gram positive result. Finally, all three colonies identified by visual inspection on the original plates were subject to PyMS. The two gram negative samples (which failed to grow on acetonitrile) grouped together in a cluster with a 42% similarity to the Rhodococcus erythropolis reference strain. In such similarity analyses this is considered as a clear failure to match the test strain. A successful match requires a similarity of at least 95% to be of interest, and preferably 100% for a positive match. The gram positive colony which successfully grew in acetonitrile had 100% similarity to the original bacterium R. erythropolis. This sequence of tests confirms the identity of the recovered colony with that used in the shot program and eliminates the false results. Estimating the rate of bacteria survivability involves determining several factors. By assuming that 10% of a doped projectile’s volume was bacteria, that a bacterium was rod shaped (1 × 1 × 2 microns in size), and that 10 projectile pieces (each 100 µm across) per shot hit the target (a typical value from visual examination of the plates after impact), then with seven doped shots fired, the number of bacteria used in the program was approximately 7,000,000 (i.e., of order 107 ). With survival observed in just one shot and assuming the successful colony grew from just one bacterium, the survival rate was thus of order 1 in 107 . This should be compared to the results of Horneck et al. (2001) who found survival rates of up to 1 in 104 for spores subject to shocks in the GPa range.
NOTE 4. Discussion. The result that bacteria growth was observed after impact is remarkable. Previous studies of survivability (Mastrapa et al. 2000, Horneck et al. 2001) have concentrated on spores, rather than bacteria themselves. This is partly because the spore is widely held to be hardier than a bacterium, and so more likely to survive an impact event, and also because during the transfer between planets spores are considered more likely to survive than bacteria. However, this study shows that bacteria themselves can survive hypervelocity impacts. Recent work by Mastrapa et al. (2000) argues that launch into space from the martian surface (carried on ejecta following an impact from space) can involve an acceleration of 3 × 106 m s−2 , with a duration of 0.5 × 10−3 s and a change in acceleration (or “jerk”) of 6 × 109 m s−3 . In the work here, assuming the projectile accelerates uniformly over the length of the launch tube in the gun, the acceleration in the firing of the gun is 1.8 × 107 m s−2 , with a duration of 0.28 × 10−3 s, corresponding to a jerk of 6 × 1010 m s−3 . These conditions exceed the requirements for launch to escape velocity from the martian surface. Furthermore, the deceleration in the impact event is even greater than the original acceleration and the time scale is even shorter. If a crater depth of 2 mm is taken for impacts in the nutrient gel (this was typical of the depth of the pit in which captured projectiles sat) the figures are deceleration of 6.25 × 109 m s−2 and duration of 0.8 × 10−6 s, corresponding to a jerk of 7.8 × 1015 m s−3 . 5. Conclusion. The results suggest that bacteria can successfully be accelerated to speeds of the order of 5 km s−1 and survive the subsequent hypervelocity impact with a target. The implications of such hypervelocity impact survivability are potentially significant. Results from a wide range of sources are systematically addressing the issue of whether bacteria can survive severe radiation environments and vacuum and live (Vreeland et al. 2000) for the long periods typically associated with objects drifting in the inner Solar System (Gladman 1997). Taken together, these results, plus those reported here for survivability in a hypervelocity impact, are starting to form a picture in which it is no longer totally implausible that bacteria can naturally migrate through space. Acknowledgment. We thank M. Cole for operation of the light-gas gun, whose use is financed by the Particle Physics and Astronomy Research Council (United Kingdom), Professor Alan T. Bull of the Biosciences Department at the University of Kent for permitting the PyMS work, and Professor Christoph Syldatk, Institute of Biochemical Engineering, University of Stuttgart, for providing access to the Rhodococcus strains used in this study.
REFERENCES Bottke, W. F., M. C. Nolan, R. Greenberg, and R. A. Kolvoord 1994. Collisional lifetimes and impact statistics of near-Earth asteroids. In Hazards due to Comets and Asteroids (T. Gehrels, Ed.), pp. 337–357. Univ. of Arizona Press, Tucson. Bronshten, V. A. 1999. The nature of the Tunguska meteorite. Meteorit. Planet. Sci. 34, 723–728. Bull, A. T., A. C. Ward, and M. Goodfellow 2000. Search and discovery strategies for biotechnology: The paradigm shift. Microbiol. Mol. Biol. Revi. 64, 573– 606. Burchell, M. J., M. J. Cole, J. A. M. McDonnell, and J. C. Zarnecki 1999. Hypervelocity impact studies using the 2 MV Van de Graaff accelerator and two-stage light gas gun of the University of Kent. Measurement Sci. Technol. 10, 41–50. Burchell, M. J., N. R. G. Shrine, A. Bunch, and J. C. Zarnecki 2000. Exobiology: Laboratory tests of the impact related aspects of Panspermia. In Impacts and the Early Earth (I. Gilmour and C. Koeberl, Eds.), pp. 1–26. Springer-Verlag, Berlin.
547
Burchell, M. J., N. R. G. Shrine, J. Mann, A. W. Bunch, P. Brand˜ao, J. C. Zarnecki, and J. A. Galloway 2001. Laboratory investigations of the survivability of bacteria in hypervelocity impacts. Adv. Space Res., in press. Colquhoun, J. A., S. C. Heald, J. Tamaoka, C. Kato, K. Horikoshi, and A. T. Bull 1998. Taxonomy and biotransformation activities of some deep-sea actinomycetes. Extremophiles 2, 269–277. Colquhoun, J. A., J. Zulu, M. Goodfellow, K. Horikoshi, A. C. Ward, and A. T. Bull 2000. Rapid characterisation of deep-sea actinomycetes for biotechnology screening. Antonie van Leeuwenhoek Int. J. Gen. Mol. Microbiol. 77, 359–367. Cronin, J. R. 1989. Origin of organic compounds in carbonaceous chondrites. Adv. Space Res. 9, 54–64. Delsemme, A. H. 2000. Cometary origin of the biosphere. Icarus 146, 313– 325. Genge, M. J., and M. M. Grady 1999. The fusion crusts of stony meteorites: Implications for the atmospheric reprocessing of extraterrestrial materials. Meteorit. Planet. Sci. 34, 341–356. Gerhardt, P. 1981. Chapter 3. In Manual Methods for General Biochemistry, pp. 21–34. American Society for Microbiology, Washington, DC. Gladman, B. 1997. Destination Earth, martian meteorite delivery. Icarus 130, 228–246. Goodacre, R., and D. B. Kell 1996. Pyrolysis mass spectrometry and its applications in biotechnology. Curr. Opinions Biotechnol. 7, 20. Goodfellow, M., and G. Alderson 1998. Editorial. Antonie van Leeuwenhoek Int. J. Gen. Mol. Microbiol. 74, 1. Hoyle, F. 1998. Comets: A matter of life or death. In The Universe Unfolding (H. Bondi and M. Weston-Smith, Eds.), pp. 3–22. Oxford Univ. Press, London. Horneck, G., D. St¨offler, U. Eschweiler, and U. Hornemann 2001. Bacterial spores survive simulated meteorite impact. Icarus 149, 285–290. Hughes, D. W., and I. P. Williams 2000. The velocity distributions of periodic comets and stream meteoroids. Mon. Not. R. Astron. Soc. 315, 629–634. Love, S. G., and D. E. Brownlee 1993. A direct measurement of the terrestrial mass accretion rate of cosmic dust. Science 262, 550–553. Mastrapa, R. M. E., H. Glanzberg, J. N. Head, H. J. Melosh, and W. L. Nicholson 2000. Survival of Bacillus subtilis spores and Deinococcus radiodurans cells exposed to extreme acceleration and shock predicted during planetary ejection. Lunar Planet. Sci. 31, 2045 (abstract). McKay, D. S., E. K. Gibson, Jr., K. L. Thomas-Keprta, H. Vali, C. S. Romanek, S. J. Clemett, X. D. F. Chillier, C. R. Maechling, and R. N. Zare 1996. Search for past life on Mars: Possible relic biogenic activity in martian meteorite ALH84001. Science 273, 924–930. Mileikowsky, C., F. A. Cucinotta, J. W. Wilson, B. Gladman, G. Horneck, L. Lindegren, J. Melosh, H. Rickman, M. Valtonen, and J. Q. Zheng 2000. Natural transfer of viable microbes in space. Icarus 145, 391– 427. Pierazzo, E., and F. Chyba 1999. Amino acid survivability in large cometary impacts. Meteorit. Planet. Sci. 34, 909–918. Sears, D. W. 1975. Temperature gradients in meteorites produced by heating during atmospheric passage. Mod. Geol. 5, 155–164. Vreeland, R. H., W. D. Rosenzweig, and D. W. Powers 2000. Isolation of a 250 million year old halotolerant bacterium from a primary salt crystal. Nature 407, 897–900. Weiss, B. P., J. L. Kirschvink, F. J. Baudenbacher, H. Vali, N. T. Peters, F. A. Macdonald, and J. P. Wikswo 2000. A low temperature transfer of ALH84001 from Mars to Earth. Science 290, 791–795.