Exploration of Methodologies to Investigate Bacterial Survival in Planetary Impacts

Introduction

The survivability of bacteria in planetary impacts has previously been investigated using a two-stage Light Gas Gun (as seen in Figure 1). During these experiments, projectiles were doped with bacteria (Rhodococcus erythropolis) and fired at hypervelocity into rock [1,2], metal [2], water-ice [3,4] and agar [4,5] targets. These experiments were designed to investigate the feasibility of the panspermia theory, which details how indigenous life forms may be spread beyond their host body via the ejecta created by hypervelocity impacts and go on to seed neighbouring bodies. These studies showed that bacteria do survive hypervelocity impact; however, the methods do not accurately quantify the survival rate on the target compared to the initial cell load on the projectile, nor do they quantify or characterise any sub-lethal effects.

Figure 1: The Light Gas Gun impact facility at the University of Kent.

 

Creating A Successful Target

Attempts have been made to develop new methods that give greater quantitative insights into both survival and sub-lethal effects of hypervelocity impacts. As well as investigating the survival rate of the bacterial population, we are looking at whether the transient extremes of pressure which occur as a result of hypervelocity impact modulate phenotypic change amongst the surviving bacteria, and thus could be a factor in the evolution of life.

The bacteria (in this case Escherichia coli) in our experiments have been placed inside the target instead of the projectile. This decision was made in order to remove the potential issues of the loss or death of cells as a result of the acceleration of the projectile. Several different target designs were trialled, largely involving the use of agar as a medium for housing the bacteria. These attempts led to a set of criteria being defined for a successful target, including efficient propagation of the shock wave through the sample to ensure that the bacteria are experiencing the intended conditions, and clean recovery of the majority of the sample with little or no contaminants.

 

The Liquid Target Setup

Following much experimentation, a liquid target setup was created, as seen in Figure 2. A 50 ml liquid sample containing the bacteria is housed inside a thin polythene bag and placed inside a steel tube, which upon impact collects the liquid and allows for ease of recovery and analysis. The impacted plastic bag produces a large amount of debris within the collected sample, which interferes with some of the optical data gathered during post-impact analysis of the bacteria. Also, there is concern that the entirety of the sample is not experiencing the full force of the impact. To address these issues, we are designing a secondary tube with a much smaller diameter which can be inserted and secured inside the primary steel tube; this should mean that more of the sample volume will experience higher shock pressures in the range of several GPa, which can be verified by simulating the impacts using Autodyn modelling. To replace the polythene bag, and thus attempt to minimise the quantity of debris entering the system, the liquid sample will be sealed directly inside the secondary tube with an extremely thin foil.

Impacts have been completed across the velocity range of 1-5 km/s using this setup. The following analysis methods have been applied to the recovered samples post-impact:

• OD₆₀₀ (optical density) recordings to understand changes in whole cell numbers by measuring the number of particles within a given sample
• Protein assays to understand the amount of physical damage or lysis to the cells
• Growth on agar plates to understand how the viability of the population has changed via the counting of colony forming units (CFUs)
• Oxygen electrode analysis to see if the metabolic pathways of the cells have been affected by measuring cellular respiration in the presence of glucose

So far, no significant changes to the survival rate or the phenotype of the population have been observed following these impacts using the analysis methods described.

 

Figure 2: The liquid target setup within the target chamber of the Light Gas Gun.

 

Influence of Exposure Time to Impact Conditions

The results from the liquid target impacts have raised the question of whether the extremely short duration of the shock pressure is insufficient to lead to any meaningful change in the bacterial population. To investigate this, E. coli samples prepared in the same manner as for the impacts are instead placed inside a sonicator, where the sound waves generate pressures within the sample of around 200 MPa, compared to the impact shock pressures of several GPa created in the Light Gas Gun. The sonication exposure times were varied, using 4 bursts of 0, 5, 10, 30 and 60 seconds, compared to the milliseconds or less that the peak shock pressures are applied to the sample in the Light Gas Gun.

A steady decline of surviving bacteria and a significant increase in cell lysis was observed as the exposure time was increased, supporting the idea that time is the key factor in generating populational changes. At burst exposure times of 30 seconds or more, an unusual phenotype emerges following growth of the sonicated samples in the form of a new colony type displaying a concentric ring pattern of growth, as shown in Figure 3. This is currently being investigated further with repeat experiments, antibiotic testing and 16S rRNA sequencing to confirm that this is indeed a change in phenotype and not a result of a separate factor such as contamination.

Figure 3: An agar plate spread with a sample of E. coli cells following 4 30-second bursts of sonication.

 

References

[1] Burchell et al. (2000) In Gilmour I., Koeberl C. (eds) Impacts and the Early Earth. Lecture Notes in Earth Sciences, vol 91.

[2] Burchell et al. (2001) Adv. Space. Res. 28(4), 707-712.

[3] Burchell et al. (2003) Origins of Life and Evolution of the Biosphere 33, 53-74 (2003).

[4] Burchell et al. (2004) Mon. Not. R. Astron. Soc. 352, 1273-1278 (2004).

[5] Burchell et al. (2001) Icarus 154, 545-547 (2001).