Reactions of Microorganisms with Atomic Oxygen Radical Anions: Damage of Cells and Irreversible Inactivation

Reactive oxygen species play important effects on organisms not only in vivo but also in vitro. The atomic oxygen radical anion (O) has shown extremely high oxidation and reactivity towards small molecules of hydrocarbons. However, the O effects on cells of microorganisms are scarcely investigated. This work showed the evidence that O could react quickly with microorganisms (Escherichia coli, Bacillus subtilis, Staphylococcus aureus, Aspergillus niger, Saccharomyces cerevisiae, and Actinomycetes (5046)) and damaged the cell walls seriously as well as their intrinsic structures, arising a fast and irreversible inactivation. SEM and TEM micrographs were used to reveal the structure changes of cells before and after reacting with O radicals. The inactivation efficiencies of the microorganisms depended on the O intensity, the initial population of microorganisms, the exposed area, the environment, and the microorganisms’ types. Over 99% reduction of an initial 1:0 × 107 colony-forming unit (cfu), E. coli population only required less than 2 minutes while exposed to a 0.23 μA/cm O flux under dry argon atmosphere (30°C, 1 atm). The observation of anionic intermediates (CO, CO, H2O , and anionic hydrocarbons) by time-of-flight (TOF) mass spectrometry and the neutral volatile products (CO, CO2, and H2O) by quadrupole mass spectrometry (Q-MS) provided an evidence of the reactions of O with hydrocarbon bonds of the microorganisms. The inactivation mechanism of microorganisms induced by O was discussed.


Introduction
Reactive oxygen species (ROS), such as the superoxide anion radical (O 2 -), the hydroxyl radical (OH), and the singlet oxygen ( 1 O 2 ), are key chemical species in the biochemical processes and exert a very important effect on organisms [1][2][3][4][5][6]. Organisms can suffer lipid peroxidation, protein denaturation, DNA injuries, and enzyme inactivation by ROS [7][8][9][10][11][12][13]. Due to high reactivity of ROS, all the cellar components, lipids, proteins, nucleic acids, and carbohydrates may be damaged by their reactions with ROS, giving rise to metabolic and cellular disturbances or cell death [14][15][16][17]. There are increasing evidences showing that ROS are responsible for many diseases such as inflammation, lung injury, ischaemia-reperfusion injury, cancer, and aging [18][19][20]. On the other hand, ROS are not always harmful to organ-isms. For instance, ROS may be used as the intracellular signaling species for the introduction of defense gene expression and destroying malignant cells or tissue [21,22].
Atomic oxygen radical anion (O -) is a monovalent anion (or monovalent negative ion) through the attachment of an electron to atomic oxygen (O). At the same time, Ois also considered as a radical because it has an unpaired electron in its outermost orbit. It was found that atomic Ohad extremely high oxidation and reactivity towards the hydrocarbons' small molecules [23][24][25][26]. Moreover, Omay be one of the most reactive oxygen species and therefore has various potential applications, such as a one-step synthesis of phenol from benzene [27], the reduction of NO [28], and the dissociation and oxidation of bio-oil [29,30]. Previously, our work showed that the Ocould quickly react with the B. subtilis cells and seriously damage the cell walls as well as their other contents, leading to a fast and irreversible inactivation [31].
Herein, we investigated the Oeffects on the microorganisms' inactivation (including the B. subtilis cells), the changes of cell structures caused by O -, and the products formed by the reaction of the microorganisms with O -. Our results demonstrate that Ocan quickly react with the microorganisms, including Gram-positive bacteria (B. subtilis, S. aureus), Gram-negative bacteria (E. coli, B. A. niger), and fungi (S. cerevisiae, Actinomycetes (5046)), and seriously damage the cell walls and their intrinsic structures, leading to a fast and irreversible inactivation. The observed intermediates and volatile products provided an important evidence of the reactions of Owith hydrocarbon bonds of the microorganisms.  [32,33]. The emitted species from the C12A7-Osurface were about 90% Oand weak electron (<10%). A reflective field was used to reflect the electrons. We controlled the beam intensity of O -(0.01-2.0 μA/cm 2 ) by changing the temperature of the C12A7-Osample and the ion extraction field.

Microorganisms' Preparation.
The microorganisms investigated in our experiments were Gram-positive bacteria (B. subtilis, S. aureus), Gram-negative bacteria (E. coli, B. A. niger), and fungi (S. cerevisiae, Actinomycetes (5046)) which were purchased from China General Microbiological Culture Collection Center and maintained on nutrient agar slants at 5-6°C. Before the inactivation experiments, microorganisms were harvested by centrifuging (4000 rpm, 4°C) for 10 min and the microorganisms' pellets were washed three times with sterile water. After pouring upper water, E. coli, B. subtilis, S. aureus, S. cerevisiae, and Actinomycetes (5046) were inoculated to L-broth (Bacto peptone 10 g/L, Bacto yeast extract 10 g/L, NaCl 5 g/L, pH = 7:2) and then incubated at 37°C on a shaking tray for 12 hours. A. niger was cultured in potato dextrose broth (200 g peeled potato with a size of 4:0 × 2:0 × 2:0 cm 3 were cooked in 1000 mL boiling distilled water for 8 min. After passing through four lay cheesecloths, the broth was collected and 20 g of dextrose was added.) and incubated at 27°C on a shaking tray for 12 hours. The density of microorganisms was controlled at 10 7~1 0 8 cfu/mL. Then, we transferred 0.1 mL resulted microorganisms to glass slide (20 mm × 20 mm) for exposure to Oflux under the dry argon environment (1 atm). All petri dishes and glass microscopy slides used for holding the microorganisms were first autoclaved at 125°C for 1 h. The numbers of colonies on the slides before and after Oexposure were determined by the spread plate method with nutrient agar grown at 37°C for 12 h.

Inactivation Experiments.
The experimental arrangement for measuring the inactivation efficiency and the reactions of microorganisms with Owas made up of three major parts: an Oradical source, a reaction chamber, and a product analysis system. The inactivation experiments were carried out in a cylindrical quartz reactor with a length of 60 cm and an inner diameter of 15 cm. The microorganism samples prepared were maintained by Petri dishes or slides, which were supported by a copper dish. The inactivation experiments can be performed under different environmental conditions by flowing different gases (such as Ar, He, N 2 , H 2 O, and O 2 ). In order to study the atomic oxygen radical anion effects, all inactivation measurements were carried out in the dry argon environment. The reaction temperature was varied by cooling or heating copper dish with a watercycled system. The Oflux emitted from the C12A7-Osample was irradiated onto the microorganisms. The absolute Oion flux was measured by a picoammeter and corrected by a TOF mass spectrometer. The numbers of colonies on the slides before and after Oexposure were determined by the spread plate method with nutrient agar grown at 37°C for 12 h.

Reaction Product Analysis.
For online analysis of reaction products, a time-of-flight (TOF) mass spectrometry and a quadrupole mass (Q-MS) spectrometry were connected to the reactor. A time-of-flight (TOF) mass spectrometer was used for measuring the anionic intermediates formed by the reactions of microorganisms with O -. The neutral products were detected by a quadrupole mass (Q-MS) spectrometry.
2.5. SEM and TEM Measurements. SEM (scanning electron microscopy) and TEM (transmission electron microscopy) experiments were performed to study the structural changes of cells before and after exposure to O -. The preparation of test samples was similar to the inactivation experiments mentioned above. After exposure to O -, the suspensions of E. coli were fixed with an equal volume of 3% glutaraldehyde buffered at pH = 7:2 with phosphate for about 1 h. Then, the samples were centrifuged, and the resulting bacterial pellets were exposed overnight to additional phosphate-buffered 3% glutaraldehyde solution. The glutaraldehyde-fixed bacteria were embedded in 1% agar and washed with phosphate buffer. After that, the samples were fixed in buffered 1% osmium tetroxide in cacodylate buffer for 1 h at room temperature and then dehydrated by successive soakings in 50, 70, 90, and 100% ethanol. The dried samples were rinsed 2 Journal of Nanomaterials twice with propylene oxide and infiltrated with propylene oxide-epoxy resin mixtures until the samples were in pure epoxy resin. Finally, the samples were placed in polyethylene capsules and resin polymerized at 60°C overnight [35]. Thin sections were cut using an ultracrotome and were mounted on 200 mesh copper grids and stained with uranyl acetate and lead citrate. The sections were examined with X-650 SEM (HITACHI) operating at 30 kV and H-800 TEM (HITACHI) operating at 100 kV.  Figures 1(a) and 1(b) present the SEM photographs for E. coli cells before (Figure 1(a)) and after the Oexposure (Figure 1(b)). The irradiated cells appeared dramatically swollen and collapsed with a large amount of fragments, which were associated with the damages to the cell walls and subsequent lysis due to the Oirradiation. Figures 1(c) and 1(d) show the representative TEM micrographs of the initial (Figure 1(c)) and the O --irradiated (Figure 1(d)) E. coli cells, respectively. The lighter part of the cell was the nuclear region containing some DNA fibrils and electron-dense ribosomes. The nucleoids after the exposure had contracted and a coarse precipitation of DNA appeared (Figure 1(d)), which indicated that Omight be able to penetrate the cells, and reacted with proteins or numerous enzymes involved in the control of DNA conformation in the nucleoids, resulting in the precipitation of DNA. Some of the cells appeared to be disrupted, and fragments of lysed cells were observed. The TEM results reveal that the secondary structure of the DNA-binding proteins of the cells was destroyed by the exposure to O -. These results are in good agreement with our earlier research [31].

O -Effects on the Microorganisms' Inactivation.
The correlation between the inactivation efficiency of the microorganisms and the Ointensity was also demonstrated by measuring the survival curves under different Ointensity. Figure 2(a) displays a series of survival diagrams under different Ointensity for E. coli intact cells, which exhibits biphasic curves (fast and slow processes). The decay of the survival numbers of E. coli cells, both in the fast and slow phases, depends on the Ointensity. On the other hand, the survival curves of E. coli spheroplasts in the fast processes are much different from those of E. coli intact cells (Figure 2(b)). The survival numbers of E. coli spheroplasts quickly decreased to the slow phases within 1 min while E. coli intact cells require about 5 min at the same condition. This result indicates that Odestroy cell walls firstly, which agree with the result by the SEM measurements. For the biphasic inactivation processes, we deduce that during the first phase; the Oradicals react with the cell walls and form the volatile small molecules (such as CO, CO 2 , and H 2 O), resulting in quickly irreversible damage and lysis of the cells in the top layer (Figure 1(b)). As the microorganisms are killed in the first phase, the inactivated cells may stack and form an isolated layer. The occurrence of slow processes may be attributed to protection resulting from the contents of dead cells, which shielded the remaining survivors or aggregation during the treatment. The second phase, therefore, would mainly reflect the time required for sufficient erosion by reacting with the debris on the top layer.
Oeffects on the other microorganisms' inactivation were also observed. As shown in Figure 3, the O --induced inactivation efficiencies for B. subtilis, S. aureus, A. niger, S. cerevisiae, and Actinomycetes (5046) were also investigated. We found Actinomycetes (5046) were most resistant to the Ointeraction, because Actinomycetes (5046) are terrestrial prokaryotes that grow mainly in mycelia and reproduce by spore. On the other hand, fungi (A. niger, S. cerevisiae) were easiest to inactivate in the same Ointensity. It appears that Gram-negative bacteria (E. coli) are more sensitive to Othan Gram-positive bacteria (B. subtilis, S. aureus). Thus, the inactivation efficiencies depend on the microorganisms' types, because the microorganisms with different components may have different resistant ability to O -.
Moreover, the inactivation efficiencies also depend on the reaction temperature and the exposure area. Figure 4 presents the survival curves of E. coli at different reaction temperature (29°C, 34°C, and 42°C). With elevatory reaction temperature, the inactivation efficiency of E. coli increases, which indicate that the inactivation process of E. coli by O - 3 Journal of Nanomaterials is a thermally enhanced process. In addition, with enlarged exposure area, the inactivation efficiency also increases.

Products Formed by the Reaction of E. coli with O -.
The reaction products had been detected by the online analysis of the trail gases when Ois exposed onto the microorganisms. The anionic intermediates and the neutral products were detected by a time-of-flight spectrometry (TOF) and a quadrupole mass (Q-MS) spectrometry, respectively. Figure 5(a) displays a typical TOF spectrum for the O -/E. coli reaction sys-tem. The anionic species of H -, OH -, CO -, CO 2 -, H 2 O -, and anionic hydrocarbons were clearly identified ( Figure 5(a)). In addition, the neutral products of CO, CO 2 , and H 2 O were also observed by the Q-MS spectra (Figure 6(a)). It was also found that the peak intensities both for the anionic species and the neutral products increased with the increasing of the Oflux intensity ( Figure 5(b) and Figure 6(b)). The microorganisms, briefly, are macromolecules mainly consisted of carbon, hydrogen, oxygen, and nitrogen atoms. Among these elements, carbon atom is the only one not self-associating to make volatile molecules such as CO and Survival curves of E. coli spheroplasts (N 0 = 1:0 × 10 7 cfu/mL) exposed to a series of Oflux intensity (exposed size: 1.8 cm 2 ), reaction temperature: 30°C.  Journal of Nanomaterials CO 2 , and it must therefore form volatile compounds with other atoms in order to be removed from the cell surface. The released CO and CO 2 would originate from the reactions of Owith hydrocarbon bonds of the microorganisms' cells. Based on the observations mentioned above, the active Oradical anions can react with the microorganisms, damage their cell structures, and finally result in a fast and irreversible inactivation. The inactivation mechanism of microorganisms induced by Owas schematically described in Figure 7.

Conclusions
The active Oradical anions possess high reactivity towards the microorganisms and damage the cell walls as well as their intrinsic structures, resulting in a fast and irreversible inactivation. The inactivation efficiencies depend on the Oflux intensity, the initial population of microorganisms, the exposed area, environment, as well as the microorganisms' types. The observation of anionic intermediates (CO -, CO 2 -, H 2 O -, and anionic hydrocarbons) by time-of-flight (TOF)

Data Availability
The data used to support the findings of this study are available from the corresponding authors upon request.

Conflicts of Interest
The authors declare no conflict of interest.  Journal of Nanomaterials