B American Society for Mass Spectrometry, 2014

J. Am. Soc. Mass Spectrom. (2014) DOI: 10.1007/s13361-014-0827-x

RESEARCH ARTICLE

Bacterial Spores Survive Electrospray Charging and Desolvation Sara N. Pratt, Daniel E. Austin Department of Chemistry and Biochemistry, Brigham Young University, Provo, UT 84602, USA

Abstract. The survivability of Bacillus subtilis spores and vegetative Escherichia coli cells after electrospray from aqueous suspension was tested using mobility experiments at atmospheric pressure. E. coli did not survive electrospray charging and desolvation, but B. subtilis did. Experimental conditions ensured that any surviving bacteria were de-agglomerated, desolvated, and electrically bacteria charged. Based on mobility measurements, B. subtilis spores survived even with + + charged + + + 2,000–20,000 positive charges. B. subtilis was also found to survive introduction into vacuum after either positive or negative electrospray. Attempts to measure Desolvated, + de-agglomerated, the charge distribution of viable B. subtilis spores using electrostatic deflection in charged bacteria vacuum were inconclusive; however, viable spores with low charge states (less than 42 positive or less than 26 negative charges) were observed. Key words: Electrospray, Bacillus subtilis, Bacteria viability, Charging, Desolvation, Extremophile + + + Solvated, no charge + + on bacteria + + + + + Agglomerate: + only exterior + +

Received: 5 June 2013/Revised: 3 January 2014/Accepted: 3 January 2014

Introduction

E

lectrospray ionization (ESI) has revolutionized access to biological molecules and systems for mass spectrometric and ion mobility analysis. The process of electrospray consists of spraying a solution of the desired analyte in a strong electric field [1]. The solution flows through a capillary and forms a Taylor cone at the end from which charged droplets are emitted. As solvent evaporates, the droplets break apart when the force of Columbic repulsion overcomes the surface tension, a point known as the Rayleigh instability limit. As solvent continues to evaporate, that process is repeated, eventually leaving individually charged molecules or particles once all the solvent has evaporated. Although mainly used for proteins and other molecules, ESI has also been used to charge intact biological systems such as viruses [2–5] and bacteria [6, 7]. Some researchers have even demonstrated that viruses are not only intact, but also viable after the electrospray process. Siuzdak et al. were the first to test viruses for viability after electrospraying [2]. They wanted to see whether noncovalent interactions were m a i n t a i n e d u n d e r el e c tro spra y co nd itions . T he electrosprayed tobacco mosaic virus particles that had landed on a glycerol-coated brass plate were examined with transmission electron microscopy. The viruses retained their quaternary structure. To determine further whether the native

Correspondence to: Daniel E. Austin; e-mail: [email protected]

structure was retained, tobacco plants were inoculated with the tobacco mosaic virus. The viruses were viable and infected the plants. Hogan et al. electrosprayed bacteriophages MS2 and T2 and T4 [8]. They found that the smaller MS2 bacteriophage survived whereas the larger T2 and T4 bacteriophages did not. The tails of T2 and T4 were separated from the heads during the electrospray process. Kim et al. used electrospray to generate an aerosol of viable Staphylococcus epidermidis and Escherichia coli [6]. However, bacteria in these experiments were not desolvated, as evidenced by the measured aerodynamic size of the resulting aerosol droplet being significantly larger than the known size of the bacteria. Further, no effort was made to exclude agglomerates, or to determine whether the bacteria that survived were electrically charged. Viability under these conditions must be addressed before ESI-generated bacteria can be used in mass spectrometry (MS) or ion mobility spectrometry (IMS) applications. In the present work, experiments using E. coli, which is the same species used by Kim et al. [6], allow for comparison to their results. The other species chosen for these studies was Bacillus subtilis, because it can survive many extreme conditions as an endospore. Nicholson et al. provide a good review of the B. subtilis sporulation process and extreme conditions that it can survive [9]. Sporulation is triggered by a nutritional deficit of carbon, nitrogen or phosphorus. In the spore state, they are resistant to desiccation, vacuum, gamma radiation, UV radiation, oxidizing agents, dry heat, and wet heat [9]. Other methods

S. N. Pratt and D. E. Austin: Bacterial Spores Survive Electrospray

besides electrospray have been used to study charged gas phase bacteria. Mainelis et al. used both electrostatic precipitation to collect different microorganisms from the air and a mobility analyzer that separated the microorganisms by charge [10–13]. The species used included vegetative Pseudomonas fluorescens and B. subtilis var. niger (BG) spores. In this process, the bacteria were nebulized, charged with an ionizer, and then deflected. In some experiments, the bacteria were collected on agar plates. In others [10], they were sorted with a mobility analyzer and counted with an optical particle counter [11, 13]. They found that both species survived these experiments, but that the sensitive P. fluorescens had a greater loss of viability [10]. In these experiments, there was no way to prevent agglomeration of bacteria, which may have contributed to the survival rates. There are a number of scenarios under which bacteria could survive ionization conditions if not desolvated or deagglomerated, as illustrated in Figure 1. It is possible that some species could survive electrospray only when solvated or agglomerated because the bacteria cell itself is not experiencing a strong electric field. This is because all of the charge is concentrated on the surface of the water droplet or the exterior of an agglomerate, with minimal electric field in the interior. If bacteria only survive electrospray while agglomerated, it would not be useful for coupling with mass spectrometry or ion mobility spectrometry because bacteria could not be separated based on individual mass. This is also the case if bacteria only survive while solvated. Therefore, for the results to be meaningful for MS and IMS, the bacteria that survive must not be solvated or agglomerated. Herein, experimental results are presented that conclusively show that individual B. subtilis spores remain viable after becoming electrically charged and desolvated using ESI. This could lead to enhanced capabilities in ion mobility

spectrometry or mass spectrometry characterization of bacterial species by coupling with other biological techniques, controlled selection and deposition of different bacteria species into arrays, and other applications such as providing the means to electrically accelerate bacteria for studies of impact survivability.

Experimental Electrospray survivability of both vegetative E. coli and B. subtilis spores was tested first with atmospheric mobility experiments and then under vacuum conditions. The aim of the atmospheric mobility experiments was to see whether bacteria could survive electrospray while charged, desolvated, and de-agglomerated. The experiments were designed to collect only bacteria cells that met all these criteria simultaneously. Experiments under vacuum were intended to determine the charge distribution of bacteria that remained viable. Because of the need to detect bacteria based on viability, mobility exclusion under atmospheric conditions was used. An electric field with opposing flow of gas excluded bacteria and particles that did not contain charge above a certain threshold. This is in contrast to conventional ion mobility spectrometry (IMS), in which a group of ions is released together by a shutter [14], and after drifting through an electric field, are characterized by the time needed to reach a Faraday-style detector. In the present experiments, detection required both collection and culturing of the bacteria, making time-resolved mobility measurements impractical. Therefore, in each experiment, a continuous stream of charged bacteria from ESI entered the drift tube instead of a pulsed sample. This resulted in integrated data because under a given set of applied voltage and gas flow rate, all bacteria with charge above the threshold would be detected together. Bacteria with a lower charge state, no charge, or a charge of opposite sign would not be detected. There could be a small possibility of obtaining negatively charged bacteria from positive electrospray based on the work by Maze et al., but they would not be detected with this set-up [15].

Preparation of Bacterial Samples

Figure 1. Three different cases in which bacteria could survive electrospray. In this figure, solid bacteria are alive and striped bacteria are dead. (a) In this case, the bacterium survives in the interior of a droplet, without experiencing an electric field. (b) In this case, the interior bacteria are uncharged and survive. The outside bacteria are charged and die. (c) In this case, the bacterium itself is charged and survives

In order to minimize false positive results and simplify sterilization procedures, antibiotic-resistant strains of both bacteria were used for all experiments. Experiments with E. coli used strains L99A and M1060, which are resistant to ampicillin. Petri dishes containing 50 μg/mL ampicillin were used for culturing in the E. coli experiments. B. subtilis experiments used strain 1A308, which is resistant to rifampicin. All petri dishes used to culture B. subtilis in this experiment consisted of nutrient agar with 2 μg/mL rifampicin to ensure that only the strain of B. subtilis we used in the experiment would grow on them. Thus, it was only necessary to ensure absence of these strains on the

S. N. Pratt and D. E. Austin: Bacterial Spores Survive Electrospray

instrument before each experiment through sterilization and negative controls. The strains chosen were resistant to two separate antibiotics to prevent cross-contamination of one species in the experiments of the other species. E. coli cells were suspended in sterile nutrient broth for the experiments. B. subtilis 1A308 was sporulated as a lawn plate on Leighton-Doi media. The plates were incubated at 32 °C for 2 weeks [9]. The shape of the cells was examined under a microscope to ensure sporulation. The spores were then harvested, washed, and suspended in sterile HPLC water. The concentrations of each were determined through serial dilutions. A small known volume of each serial dilution was cultured on a petri dish. The cultured cells were counted and multiplied by the dilution factor to determine the concentration of the original solution.

Atmospheric-Pressure Mobility Instrument and Experiments The aim of the atmospheric-pressure mobility experiments was to determine whether the species tested survives electrospray. The instrument was designed so that uncharged bacteria would be excluded. It was also designed with the ability to vary conditions (i.e., applied voltage, gas flow rate) to explore the range of charges of surviving bacteria. A similar drift tube has very recently been reported by Oberreit et al. [16] for time-resolved ion mobility spectrometry experiments on very small aerosols (2–11 nm range). A diagram of the instrumental setup used for the mobility experiments is shown in Figure 2. Some features illustrated in the diagram were different for E. coli and B. subtilis experiments. For the E. coli experiments, the ESI source consisted of a 32-gauge stainless steel needle (Small Parts, Inc., Logansport, IN, USA) connected using 1/16 in. PEEK tubing and PEEK fittings to a 500 μL syringe (Hamilton Gastight #1750; Reno, NV, USA) fixed in a syringe pump (KD Scientific model # 780100 V; Holliston, MA, USA). A high voltage power supply (Stanford Research Systems PS350; Sunnyvale, CA, USA) was connected to the electrospray needle. The drift tube consisted of a lead silicate glass tube (Burle Fieldmaster; Sturbridge, MA,

Figure 2.

USA) that was doped to provide 20 mega ohms resistance. The drift tube was used to provide a uniform electric field and to direct the flow of gas. It was 20-cm long and had a 57-mm inner diameter. Tungsten wire mesh covered the inlet of the drift tube so that the applied voltage was uniform across the inlet. A stainless steel collection plate was secured to the back of the drift tube. Dry nitrogen gas was connected to an inlet in the back of the drift tube and was able to flow uniformly around the circumference of the collection plate and through the tube. The counter-gas flow rate was monitored with a flowmeter. A picoammeter (Keithley 6485; Cleveland, OH, USA) was connected to the collection plate to monitor the current from the spray. The electrospray was monitored with a microscope or LED light to ensure that it was in stable cone-jet mode. No measurements of current were made. For the experiments with B. subtilis, additional instrument modifications were made to improve the setup. To ensure that the B. subtilis spores were desolvated, dry nitrogen nebulizing gas flowing around the electrospray needle was added. The N2 nebulizing gas was set up with an Upchurch P727 Tee. The small, 32-gauge electrospray needle went straight through the Tee connection with the gas flowing around it. Both the counter-gas and the nebulizing gas flow rates were monitored with a flowmeter. A 2 μm frit (Upchurch Scientific Frit-In-A-Ferrule) was placed immediately before the electrospray needle to exclude agglomerated bacteria. To generate a higher potential difference across the drift tube, the front of the drift tube was connected to ground and a Stanford Research Systems PS350 High Voltage Power Supply was connected to the back of the drift tube and the collection plate. In the E. coli atmospheric-pressure mobility experiments, suspensions of E. coli in nutrient broth were introduced via electrospray into the inlet of the drift tube. The nutrient broth consisted of 3.0 g beef extract and 5.0 g peptone suspended in a liter of HPLC water. The final pH was 6.8±0.2. A large positive voltage, typically 4,000–5,000 V, was applied to the electrospray needle. From the needle, bacteria were electrosprayed into the inlet of the drift tube. A positive voltage lower than the voltage applied to the electrospray

An illustration of the instrumentation used for the atmospheric mobility experiments

S. N. Pratt and D. E. Austin: Bacterial Spores Survive Electrospray

needle, typically around 2,500–3,000 V, was applied to the front of the drift tube whereas the back of the drift tube was grounded. This created an electric field across the drift tube that drew electrically-charged species through the tube. A countercurrent of dry nitrogen flowing in the opposite direction of the electric field prevented neutral and negatively-charged bacteria from reaching the end of the drift tube. Positively-charged species traversed the drift region and landed on a metal collection plate. The collection plate was sterilized with methanol before each experiment. Following sterilization, a swab from the collection plate was taken and cultured on a nutrient agar petri dish containing ampicillin to ensure that there was no E. coli on the plate before running the experiment. After each experiment, the contents of the collection plate were washed onto a petri dish. The plate was washed by pipetting 300 μL of sterile water onto it, swirling the water around, and then transferring the water with a pipette onto the petri dish. The sample was spread on the petri dish with a sterile spreader. The petri dish was cultured overnight at 39 °C. The electrospray voltage, the voltage applied to the front of the tube, and the flow rate of the bacteria solution could be adjusted in all of the experiments. To ensure that the E. coli could survive drying on the collection plate, some of the electrospray solution was pipetted onto the collection plate. It was left to dry and then the collection plate was swabbed and cultured. To test if the E. coli was not sticking to the drift tube, the drift tube was sterilized before an experiment and then swabbed and cultured afterward to look for growth from bacteria that had fallen off the collection plate and landed in the tube. In the first experiments with B. subtilis, the voltage was applied in the same way that it was in the E. coli experiments (the front of the tube was at a lower positive voltage than the electrospray needle and the back of the tube was grounded). Later, to get a larger potential across the tube, the front of the drift tube was grounded and a constant negative voltage (could be set between –1,000 and – 5,000 V) was applied to the back end of the drift tube. In the experiments, we could adjust the electrospray voltage, the voltage applied to the tube, the flow rate of the nebulizing gas, the flow rate of the counter gas (up to 0.3 m/s), and the flow rate of the B. subtilis suspension (typically 100 μL/h). Before each trial with B. subtilis, the collection plate was sterilized with bleach. Negative controls were made as above. Suspensions of B. subtilis in sterile water were introduced via electrospray into the inlet of the drift tube. After each trial, the contents of the collection plate were washed with 300 μlL of water onto a petri dish with rifampicin-containing agar. The method of washing was the same as in the E. coli experiments. The concentration of the spore solution was previously determined by counting the colonies from cultures of serial dilution. Typically, a solution with a concentration of 66,000 colony-forming units per milliliter was used. This concentration is low

enough that the droplets formed in the spray will rarely, if ever, contain more than a single spore, eliminating the possibility of agglomeration in the spray plume. Percent survivability was determined by comparing the live counts from the culture of the collection plate to the number of spores contained in the solution that was electrosprayed into the drift tube. Negative controls were run with the gas in the absence of bacteria to ensure that no growth was observed from the experiment when the electrosprayed solution contained no bacteria.

Vacuum Deflection Theory Because the above method produced integrated data on charge states of bacteria that survived electrospray, it was desirable to conduct additional experiments to determine the charge state distribution. These measurements were made using electrostatic deflection in vacuum. A charged particle of mass m and charge q moving through an electric field E experiences an acceleration [17] aE ¼

qE : m

This acceleration is inversely proportional to the mass-tocharge ratio. Therefore, if a stream of particles with uniform velocity passes through a transverse electric field, particles with higher charge or lower mass will be deflected more strongly. This principle is illustrated in Figure 3. In this case, physical dispersion is necessary because the bacteria must be in separate locations to be cultured, which is the only way to determine viability. Using the equation for acceleration of a charged particle due to an electric field, the equation to solve for particle charge in terms of total distance deflected is found to be as follows: q¼

ytot 1 vL2 vLD þ 2 2 mdvx mdv2x

Figure 3. A beam of charged particles (in this case, bacteria) with uniform velocity that pass between two plates with an electric potential between them will be spatially dispersed based on m/z

S. N. Pratt and D. E. Austin: Bacterial Spores Survive Electrospray

In this equation, q is the charge, ytot is the total deflected distance in the y direction, v is the voltage, m is the mass of the particle, L is the distance from the deflector to the collector in the x direction, d is the distance between deflector plates, D is the length of the deflector, and vx is the velocity in the x direction. Because DGL, the second term in the denominator may be negligible. A diagram illustrating these variables is shown in Figure 4.

Vacuum Deflection Instrument and Experiments The electrospray source and power supply were identical to those used during mobility experiments above. Again, a 2μm frit filter was placed immediately before the electrospray needle to exclude agglomerated bacteria. Both positive and negative voltages on the electrospray needle were used for the experiments with B. subtilis. The nitrogen nebulizing gas was heated to 26–29ºC using heat tape wrapped around the gas line. The gas flow was monitored with a flow meter. The vacuum system consists of three differentiallypumped stages. The electrospray plume was directed at a 150-μm orifice, beyond which a 38-cm long beam tube with a 4-mm i.d. was used to collimate the particle beam and ensure that the spores achieved a uniform velocity. Skimmers separated this section from the two subsequent stages, both of which were evacuated using turbopumps. The final stage has a pressure in the 10–5 Torr range. The “Bug Trap” was placed after the second skimmer cone. The Bug Trap consists of a deflector and a collector. The deflector is two parallel stainless steel plates with a potential applied between them. The acrylic collector has a series of slots or channels in which bacteria accumulate after deflection by the applied voltage. The Bug Trap was mounted onto a flange so that it could be easily removed after each experiment for culturing and sterilization. One of the deflector plates was grounded and the other was connected to either a high- or low-voltage power supply (Stanford Research Systems PS350 or Agilent Triple Output DC Power Supply). An acrylic collector on a Delrin stand with seven slots was placed after the deflector plates. To better trap the spores, the back of the slots is deeper than the entrance, and the slots are angled to reduce

Figure 4. Diagram of the terms used in the electrostatic deflection calculations

the possibility that the spores bounce out of the slot upon striking the wall. A diagram of the complete instrument is shown in Figure 5. The deflection/collection setup (Bug Trap) is illustrated in detail in Figure 6. The collector was washed out with and submerged overnight in a 10% bleach solution to sterilize it. Before each experiment, the collector was then washed out three times with sterile water to remove the bleach. Each slot was then swabbed and plated on a rifampicin-containing petri dish to ensure that no B. subtilis was present before the experiment. For each experiment, a 0.5-mL suspension of B. subtilis spores in HPLC-grade water was electrosprayed at 100 μL/h into the inlet of the vacuum system. Pure water was used to ensure that the B. subtilis remained in the spore state. Freshly vortexed solution was used for each experiment to fix a previously identified problem of spores settling out of solution. The typical concentration of the spore solution was 660,000 colony-forming units per milliliter, which was higher than the atmospheric mobility experiments because losses due to entry into vacuum were expected. However, even at this concentration, the likelihood of droplets containing multiple spores was very low. The nebulizing gas was set to 1 L/min. The electrospray voltage was set to give a stable electrospray (i.e., observed Taylor cone using microscope objective and illumination), which occurred at 3,700 V when the needle was 1 cm from the inlet. The electrosprayed spores passed through the beam tube and skimmers and between the two metal plates. One of the plates was always grounded. The other plate was set at a voltage between 0 and 5,000 V for each experiment. The spores landed in different slots on the collection vessel depending on the charge on the spore, the spore velocity, and the voltage applied to the plates. The average (36.2 m/s) and (standard deviation of 4.4 m/s) range of spore velocities was measured previously using this instrument configuration with an image-charge detector (unpublished work).

Figure 5. Diagram of the vacuum deflection instrumentation. See Figure 6 for a detailed view of the Bug Trap

S. N. Pratt and D. E. Austin: Bacterial Spores Survive Electrospray

with both positive and negative electrospray were performed.

Results and Discussion Atmospheric-Pressure Mobility Experiments

Figure 6. Design drawing of the Bug Trap on the inner part of the flange it was mounted on. Also, a top view detail of the collection channels

After each experiment, the flange containing the Bug Trap was removed from the vacuum system. The collector was detached and each slot was individually washed out with sterile water and plated on a rifampicin-containing nutrient agar petri dish. The washing procedure consisted of filling up a slot with sterile water using a sterile syringe. Then, the syringe was used to take the water back out of the slot and transfer it to the petri dish where the water was spread with a sterile spreader. The petri dishes were incubated overnight at 39 °C. The next day, the colonies that had grown on the plates were counted. Experiments

E. coli inconsistently survived the atmospheric mobility experiments where it was electrosprayed through the drift tube. In the absence of a counter-gas flow, the E. coli sometimes survived, but what looked like dried water droplets were visible on the plate. When dried water droplets were not visible, the E. coli did not survive. This indicates that E. coli only survived electrospray when it was still solvated at the time the droplets contacted the collection plate and were neutralized, because even E. coli with a very low charge state would have made it to the end of the drift tube in the absence of a counter-gas flow. With a countergas flow, no viable E. coli was collected. To ensure that the E. coli was not just falling down off of the collection plate onto the drift tube, the drift tube was sterilized before an experiment and then it was swabbed afterwards to see if the E. coli survived, but did not stick to the collection plate. There was no growth, which indicates that the E. coli did not survive. Because E. coli survived the drying control, we can infer that it was the charging of the cell that caused death. Impact with the collection plate may contribute to the lack of survival, but the E. coli would have had a range of charges and thus velocities. The cells with a low charge state would have had an especially low velocity because of the the drift gas, so they would have had a softer impact. Kim et al. collected E. coli with an Anderson Impactor and saw viability under their conditions [6]. It is unlikely that the impact would have killed all of the E. coli. From this, we conclude the E. coli does not survive all of the conditions of electrospray under normal atmospheric conditions. Unlike the vegetative E. coli, B. subtilis spores did survive desolvation and charging at atmospheric pressure. In these experiments, we never observed dried water droplets on the collection plate or any other sign that the solvent was reaching the plate. This is likely due to the addition of the nebulizing gas. The difference in survival is likely due to the difference between the two types of cells. E. coli is gram negative and does not form spores, whereas B. subtilis is gram positive and does form spores. Because the B. subtilis was sporulated, the most important factor in the survival of the B. subtilis spores was likely their spore coat. Other research into B. subtilis extremophile behavior has shown the spore coat to be instrumental in B. subtilis survival of harsh conditions [9]. Further experiments done with unsporulated B. subtilis, other Bacillus spores, and other gram negative and gram positive bacteria are necessary to further explore the factors behind survivability. With the counter-gas flowing in the drift tube, the minimum number of charges that allowed an individual

S. N. Pratt and D. E. Austin: Bacterial Spores Survive Electrospray

bacterium to reach the collection plate in these experiments can be determined using their mobility K: K¼

q f

In this equation, q is the charge on a particle and f is the friction factor. In these experimental conditions, the radius of the bacteria is much greater than the mean free path of the gas molecules, meaning that the friction factor can be determined using the equation of friction factor at continuum limit [12]. f ¼ 6⋅π⋅μ⋅rp

In this equation, rp is the radius of spherical particle and μ is the gas dynamic viscosity. The mobility limit of the bacteria can be calculated from a force balance between the electric field and the counter gas. In the case where the electric field is balanced by the friction force: Force ¼ q⋅E ¼ v⋅ f

Here, E is the electric field and v is the velocity of the counter gas. The electric field is defined as the applied voltage, V, divided by the length of the drift tube, d. E¼

V d

From the definition of mobility, we can see that: K¼

q v ¼ : f E

By these given relations with V = 3,000 V, v=0.03 m/s, d=20 cm, and approximating rp = 0.5 μm, then ¼ 2x10−6 ˙ ð1=T Þ. From this, we can calculate that each B. subtilis spore that reached the end of the drift tube while the counter-gas was flowing had a minimum of 2×103 elementary charges. We observed spores surviving with voltages as low as 315 V, which meant that at least some of the spores had a charge of at least 2×104 elementary charges. The design for the B. subtilis atmospheric mobility experiments ensured that any bacteria detected at the end of the tube must be charged, desolvated, and de-agglomerated. Specifically, the nitrogen counter gas and voltage applied to the drift tube ensured that the bacteria that were collected at the end were charged. The dry nitrogen nebulizing gas and counter gas are believed to be adequate to desolvate the bacteria, although this was not verified other

than the absence of any observable water building up on the collection plate after many hours of operation. In their experiments, Kim et al. used CO2 to isolate the electrospray from air, but they did not specifically use a nebulizing gas or counter gas [6]. The filter in the electrospray line in our experiment ensured that the bacteria were not agglomerated. Therefore, any bacteria that were cultured from the collection plate were charged, desolvated, and de-agglomerated. The E. coli and B. subtilis had different outcomes in the atmospheric mobility experiments. Eninger et al. looked at the influence of gas composition on viability while electrospraying viruses [18]. They found that adding CO2 to the sheath gas increased virus viability, possibly because it lowered the production of reactive oxygen species compared with normal air. Owing to the sheath gas in those experiments, the viruses were likely desolvated at the electrospray source and thus would have been exposed to the reactive oxygen species in the air. Kim et al. placed their electrospray in CO2, but it was not a flowing sheath gas, therefore the CO2 did not likely influence survival in their experiments because the bacteria were not desolvated at the electrospray source. During our E. coli experiments, the atmospheric mobility instrument did not yet have a nebulizing sheath gas at the electrospray source, so the E. coli likely did not become desolvated until contacting the nitrogen drift gas. This is evidenced by the appearance in some of the experiments without a countergas of apparent dried water droplets on the collection plate. Thus, it is unlikely that the gas composition at the source influenced E. coli survival in our experiments either. In the experiments with B. subtilis, the sheath gas was nitrogen, which would have also reduced reactive oxygen species. In light of the conclusions of Eninger et al., a study of different nebulizing sheath gas compositions and the impact on the electrospray survivability of bacteria could be useful. The measured recovery rate of B. subtilis spores was less than 20% at atmospheric pressure, but the actual survival rate could be higher because all of the charged, viable spores may not have been collected. There are a few reasons why not all of the spores in the solution would have made it to the end of the drift tube. Not all of the bacteria in the solution would be charged. Some were probably lost in the spray before they reached the drift tube. It could also be possible that some cells had a charge lower than was needed to make it to the end of the drift tube. In any case, at least 15–20% of the B. subtilis spores satisfied simultaneously all the criteria of being charged, de-agglomerated, desolvated, and still viable. The percent recovery of the bacteria changed with different voltages or nitrogen counter gas flow rate, but the variation was dominated by the settling of the spores in the aqueous suspension, making it difficult to correlate recovery with mobility (and hence, charge). In control experiments where voltage and flow rate were held constant, the variation in recovery rate was as large as that observed by varying voltage and flow rate (Table 1). There was a strong

S. N. Pratt and D. E. Austin: Bacterial Spores Survive Electrospray

correlation between the time since vortexing and the recovery rate, other factors held constant, implying that this variable obscured the others. As a result, we were able to constrain, but not determine, the charge distribution on B. subtilis from the mobility experiments. For instance, the spores observed viable in Table 1 all had at least 3600 elementary charges. It was important to ensure that the bacteria were deagglomerated and desolvated. If the electrosprayed bacteria were still agglomerated, the bacteria on the inside of the agglomerate would not have become charged. Thus, the bacteria that were cultured would not necessarily have been charged. Desolvation is important for a similar reason. If the spore was surrounded by water, it would not experience a net electric field. Thus, the bacteria that survived may not have been charged. Because of the steps to ensure desolvation and de-agglomeration, the B. subtilis spores that survived in the present experiments were definitely charged. In these experiments, the vegetative E. coli did not survive, whereas the B. subtilis spores did. This result is similar to what Mainelis et al. found in their studies [10]. In those studies, they found that BG spores had a higher survival rate than vegetative P. fluorescens, but that P. fluorescens did survive their ionization method. They had no way of preventing the micro-organisms from being agglomerated, so that may have contributed to the survival of the P. fluorescens. Also, perhaps that species is hardier than E. coli. Further electrospray experiments using P. fluorescens could be valuable. This work addresses limitations of other previous work [6]. Kim et al. studied the use of electrospray for viable bioaerosols with S. epidermidis and E. coli. They found that the bacteria survived, but they report their electrosprayed bacteria droplet size greater than the size of the bacteria. Therefore, we hypothesize that the bacteria were not desolvated. Also, there was no method applied to ensure the bacteria were not agglomerated besides relying on the repulsive Columbic force and vortexing the solution. In their work, they reported that the bacteria survived electrospray, but the E. coli was not desolvated, so they did not truly survive the complete process. They were only looking at electrospray to form a bacterial aerosol, which was successful. However, there was not enough information to generalize their findings of viability and apply it to future mass

Table 1. Results from Atmospheric Mobility Experiments with B. subtilis Spores Under the Same Conditions with 1,800 V Across the Tube Run Number 1 2 3 4 5 6 7

% Recovery 12.9 14.3 17.0 8.1 0.1 5.6 2.2

spectrometry studies. This shortcoming is the area that this work addresses.

Vacuum Deflection Experiments The B. subtilis spores also survived electrospray with subsequent introduction into vacuum, although the vacuum recovery rate was lower (G0.1%). The entrance into the vacuum system was much smaller than the entrance of the drift tube, so lower recovery rates were expected because fewer bacteria would make it into the system in the first place. Also, bacteria could be lost before reaching the end of the vacuum system. When no voltage was applied to the deflection plates, we only observed bacteria deposited into the central channel, as expected for an undeflected beam. This demonstrated that the channel geometry was appropriate for spore collection and recovery. We observed minimal deflection of the spores with positive electrospray and applied deflection voltages ranging from 0 to –5,000 V. Even when the maximum voltage of –5,000 V was applied to the right deflection plate, the furthest the spores were deflected was by one slot from the beam axis. The spores would have had less than or equal to approximately 42 charges (in units of e) and greater than 26 charges to be deflected this far. No spores were collected from the remaining channels. This observation could be the result of several possible mechanisms: (1) the spores are not becoming as highly charged as expected from our previous atmospheric mobility experiments, (2) spores with a high charge state lose viability in vacuum, or (3) some form of charge reduction occurs as the spores pass through the vacuum interface. The last of these three—charge reduction upon introduction to vacuum—seems to be the most likely, although our experiments cannot rule out the second possibility. Perhaps the combined stress of both a high charge state and low pressure is enough to kill the spores. It is also possible that highly-charged spores died on impact because of a charge interaction with the plastic surface of the collector. In order to determine whether all the spores are in a low charge state or whether the spores have a variety of charge states and only the ones with little charge are surviving or being collected, more experiments would be needed. The bacterial spores also survived negative electrospray into vacuum conditions. They were electrosprayed again in sterile HPLC-grade water. Even with a maximum deflection voltage of +5,000 V, the spores were not deflected. This again implies that under these experiment conditions, one of three processes is occurring: (1) only the spores in a low charge state survive the full experiment including vacuum and collection, (2) none of the spores acquiring high charges are successfully collected, or (3) the charge of the spores is reduced during the experiment, perhaps as a result of introduction into vacuum. The spores would have had fewer than 26 negative charges to not be deflected at all. Because the negatively electrosprayed spores were not deflected as

S. N. Pratt and D. E. Austin: Bacterial Spores Survive Electrospray

much as the positively electrosprayed spores, it may be that the negatively electrosprayed spores had a lower number of charges than the positively electrosprayed spores, or they only survived with a lower number of charges than the positively electrosprayed spores. The mechanism of charge reduction is expected to differ between positive and negative charges, so if this is occurring, the difference is not surprising.

Potential Applications for Results The fact that bacteria can survive electrospray charging opens up possibilities for different types of bacterial analyses to be coupled together. For example, electrospray can be used to introduce samples and separate bacteria by nondestructive mass spectrometry or ion mobility spectrometry followed by analysis using standard microbiological techniques. The bacteria could also be manipulated and deposited into arrays based on characteristics measured by IMS or MS. There is also currently a need for better methods to study bacteria survival in high-velocity impacts. When looking for signs of extraterrestrial life, it is important that spacecraft do not contaminate a sample. If spacecraft crash, they could potentially contaminate other planets with bacteria and give false positive results when looking for signs of life. We are currently conducting experiments to determine whether bacteria can survive high-velocity impacts using electrospray to introduce them into an acceleration instrument. A few experiments testing microorganism survivability of extreme shock or acceleration have been done, but they were aimed at demonstrating survivability under conditions simulating the interior of a meteorite [19–21]. These studies show widely variable survivability rates for different bacteria [20]. Still, the upper limits of survivability have not been tested, and they were shock rather than impact experiments. These experiments were very costly and time-consuming. Electrospray ionization (ESI) is a much less expensive alternative, costing almost nothing per experiment once the equipment is constructed and the method working. In order to use ESI for this purpose, the microorganisms must survive the charging process.

Conclusion We have presented experimental evidence that B. subtilis spores survive electrospray charging and, further, that the spores remaining viable were also electrically charged, deagglomerated, and desolvated. The discovery that bacterial spores survive electrospray charging could enable sorting of bacteria and mass spectrometry followed by culturing. It could also enable the use of electrospray as a source of charged bacteria that could be electrically accelerated for high velocity impact research. The fact that bacteria can survive electrospray charging opens up possibilities for different types of bacterial analyses to be coupled together.

The above results are also interesting in the context of bacteria extremophile behavior. These experiments demonstrate another harsh condition that B. subtilis spores survive—being electrically charged in the gas phase. Although this characteristic probably does not represent an evolutionary adaptation to environmental pressure, it nonetheless can be added to the long list of extreme conditions that the spores can withstand. Further, electrical charging of bacterial spores may occur in space, an environment that B. subtilis spores are known to survive [22].

Acknowledgments This project was funded in part by NASA’s Planetary Protection Program, with additional funding from the Office of Graduate Studies and the College of Physical and Mathematical Sciences, Brigham Young University.

References 1. Kebarle, P., Tang, L.: From ions in solution to ions in the gasphase—the mechanism of electrospray mass-spectrometry. Anal. Chem. 65, A972–A986 (1993) 2. Siuzdak, G., Bothner, B., Yeager, M., Brugidou, C., Fauquet, C.M., Hoey, K., Chang, C.M.: Mass spectrometry and viral analysis. Chem. Biol. 3, 45–48 (1996) 3. Fuerstenau, S.D., Benner, W.H., Thomas, J.J., Brugidou, C., Bothner, B., Siuzdak, G.: Mass spectrometry of an intact virus. Angew. Chem. Int. Ed. 40, 542–544 (2001) 4. Thomas, J.J., Bothner, B., Traina, J., Benner, W.H., Siuzdak, G.: Electrospray ion mobility spectrometry of intact viruses. Spectrosc. Int. J. 18, 31–36 (2004) 5. Bothner, B., Siuzdak, G.: Electrospray ionization of a whole virus: analyzing mass, structure, and viability. Chem. Biochem. 5, 258–260 (2004) 6. Kim, K., Kim, W., Yun, S.H., Lee, J.H., Kim, S., Lee, B.U.: Use of an electrospray for the generation of bacterial bioaerosols. J. Aerosol Sci. 39, 365–372 (2008) 7. Kim, K., Lee, B.U., Hwang, G.B., Lee, J.H., Kim, S.: Drop-on-demand patterning of bacterial cells using pulsed jet electrospraying. Anal. Chem. 82, 2109–2112 (2010) 8. Hogan, C.J., Kettleson, E.M., Ramaswami, B., Chen, D.R., Biswas, P.: Charge reduced electrospray size spectrometry of mega- and gigadalton complexes: whole viruses and virus fragments. Anal. Chem. 78, 844– 852 (2006) 9. Nicholson, W.L., Munakata, N., Horneck, G., Melosh, H.J., Setlow, P.: Resistance of bacillus endospores to extreme terrestrial and extraterrestrial environments. Microbiol. Mol. Biol. Rev. 64, 548–572 (2000) 10. Mainelis, G., Adhikari, A., Willeke, K., Lee, S.A., Reponen, T., Grinshpun, S.A.: Collection of airborne microorganisms by a new electrostatic precipitator. J. Aerosol Sci. 33, 1417–1432 (2002) 11. Mainelis, G., Gorny, R.L., Reponen, T., Trunov, M., Grinshpun, S.A., Baron, P., Yadav, J., Willeke, K.: Effect of electrical charges and fields on injury and viability of airborne bacteria. Biotechnol. Bioeng. 79, 229–241 (2002) 12. Mainelis, G., Willeke, K., Baron, P., Grinshpun, S.A., Reponen, T.: Induction charging and electrostatic classification of micrometer-size particles for investigating the electrobiological properties of airborne microorganisms. Aerosol Sci. Technol. 36, 479–491 (2002) 13. Mainelis, G., Willeke, K., Baron, P., Reponen, T., Grinshpun, S.A., Gorny, R.L., Trakumas, S.: Electrical charges on airborne microorganisms. J. Aerosol Sci. 32, 1087–1110 (2001) 14. Eiceman, G.A., Karpas, Z.: Ion mobility spectrometry, 2nd edn. CRC Press, Boca Raton (2005) 15. Maze, J.T., Jones, T.C., Jarrold, M.F.: Negative droplets from positive electrospray. J. Phys. Chem. A 110, 12607–12612 (2006)

S. N. Pratt and D. E. Austin: Bacterial Spores Survive Electrospray

16. Oberreit, D.R., McMurry, P.H., Hogan, C.J.: Mobility analysis of 2 nm to 11 nm aerosol particles with an aspirating drift tube ion mobility spectrometer. Aerosol Sci. Tech. 48, 108–118 (2014) 17. Raymond A Serway; John W. Jewett, J.: Physics for Scientists and Engineers, 7th ed. Thomson Brooks/Cole: USA Vol. II (2008). 18. Eninger, R.M., Hogan, C.J., Biswas, P., Adhikari, A., Reponen, T., Grinshpun, S.A.: Electrospray versus nebulization for aerosolization and filter testing with bacteriophage particles. Aerosol Sci. Technol. 43, 298–304 (2009) 19. Roten, C.-A.H., Gallusser, A., Borruat, G.D., Udry, S.D., Karamata, D.: Impact resistance of bacteria entrapped in small meteorites. Bulletin de la Societe Vaudoise des Sciences Naturelles 86, 1–17 (1998)

20. Mastrapa, R.M.E., Glanzberg, H., Head, J.N., Melosh, H.J., Nicholson, W.L.: Survival of bacteria exposed to extreme acceleration: Implications for panspermia. Earth Planetary Science Lett. 189, 1–8 (2001) 21. Horneck, G., Stoffler, D., Eschweiler, U., Hornemann, U.: Bacterial spores survive simulated meteorite impact. Icarus 149, 285–290 (2001) 22. Wassmann, M., Moeller, R., Rabbow, E., Panitz, C., Horneck, G., Reitz, G., Douki, T., Cadet, J., Stan-Lotter, H., Cockell, C.S., Rettberg, P.: Survival of spores of the UV-resistant Bacillus subtilis strain mw01 after exposure to low-earth orbit and simulated Martian conditions: data from the space experiment ADAPT on EXPOSE-E. Astrobiology 12, 498–507 (2012)