Cancer-specific cytotoxicity of Ringer’s acetate solution irradiated by cold atmospheric pressure plasma

Abstract Cold atmospheric pressure plasmas are promising medical tools that can assist in cancer treatment. While the medical pathology mechanism is substantially understood, knowledge of the contribution of reactive species formed in plasma and the mode of activation of biochemical pathways is insufficient. Herein, we present a concept involving antitumoral plasma-activated organics, which is envisaged to increase cytotoxicity levels against cancer cells. Ringer′s acetate solution was irradiated by low-temperature plasma at atmospheric pressure and possible reaction pathways of the compound generation are presented. The chemical compounds formed by plasma treatment and their effects on non-tumorigenic breast epithelial cells (MCF-10A) and breast cancer cells (MCF-7) were investigated. The cell viability results have shown that plasma-derived compounds have both, stimulatory and inhibitory effects on cell viability, depending on the concentration of the generated compounds in the irradiated liquids. Previous studies have shown that oxidative stresses involving reactive oxygen and nitrogen species (RONS) can be used to kill cancer cells. Hence, while RONS offers promising first-step killing effects, cell viability results have shown that plasma-derived compounds, such as acetic anhydride and ethyl acetate, have the potential to play important roles in plasma-based cancer therapy.


Introduction
With the promising perspective of becoming a new type of oncological therapy, cold atmospheric plasmas (CAPs) and plasma-activated liquids have recently entered the limelight [1][2][3][4][5]. A CAP is a mixture of ionized gases produced at near-room temperature conditions that contain a high number of reactive species, ions, electrons, and metastable particles within an electromagnetic field created by weak ultraviolet (UV) and vacuum ultraviolet (VUV) radiation [6]. While thermodynamically stable under normal conditions, such plasma-originated chemical activities can be transferred to liquids through kinetically controlled secondary reactions [7]. These non-equilibrium interactions hereafter referred to as non-equilibrium atmospheric pressure plasma (NEAPP) [8], have been experimentally used to irradiate more than 30 compounds suspended in Dulbecco's modified Eagle minimum essential medium (DMEM) to produce plasma-activated medium (PAM), which has been experimentally shown to be capable of selectively killing cancer cells while leaving normal cells unharmed [9,10]. Additionally, in vivo xenografted tumors in a mouse model were decreased by the subcutaneous injection of PAM, as reported by Utsumi et al. [11]. Hence, it can be seen that the antitumor effects of plasma irradiated liquids have the potential to provide novel cancer therapeutic applications and are therefore being actively developed to realize the fourth type of therapy alongside the three major conventional therapies of surgery, chemo, and radiation [12].
NEAPP sources are designed to operate in ambient air conditions, generating a significant amount of gaseous reactive oxygen and nitrogen species (RONS) that are relevant due to their biological effects [13,14]. The actual RONS concentrations in PAM were investigated and found that H 2 O 2 and NO 2 À combined synergistic-with the case when H 2 O 2 was used alone [15,16]. Presently, the synergistic effects of the NO 2 À /H 2 O 2 ratio have been attracting attention [17] and efforts aimed at controlling the production of RONS in aqueous solutions have been widely reported [18][19][20]. Nonetheless, when the NO 2 À /H 2 O 2 ratio in the PAM is taken into consideration, it is clear that there are mysterious hindering mechanisms behind the antitumor effects that have not yet been clarified. Plasma-activated Ringer's lactated solution (PAL) has also shown an antitumor effect [21][22][23]. The selective killing effect on cancer cells of PAL was evidenced in tumor cells caused by cell apoptosis and cell proliferation suppression [24,25]. To date, no specific chemical compounds have been assigned completely yet for the main role in inducing the antitumor effect. For this, knowledge of plasma-derived liquid chemistry is becoming important to identify any species responsible for the selective killing of cancer cells. Apart from RONS, different compounds may be generated in a plasma-activated medium, depending on the chemical composition of the liquid precursor. A careful investigation of the reaction pathways occurring in various types of plasma-irradiated liquids is of paramount importance in understanding the effect of plasma-generated compounds on cancer cells.
Ringer's solutions are used as solvents for anticancer drugs administered intraperitoneally [26,27]. Ringer 0 s acetate was found to be more useful in metabolic acidosis treatment than Ringer 0 s lactate, as the lactated liquids increased the levels of plasma lactate and pyruvate compared to the acetated solutions after infusion [28]. Due to the rapid metabolization and clinical usefulness, Ringer 0 s acetate solution is also considered a useful extracellular fluid replacement. Here, we report the antitumor effect of cold atmospheric plasma-activated Ringer 0 s acetate solution (PAA) on breast cancer cell lines (MCF-7) and its selectivity in killing these cells, compared to non-tumorigenic epithelial cell lines (MCF-10A). The plasma-derived liquid chemistry was investigated to identify the chemical species formed in the discharge with a potential antitumor effect. The plasma irradiated liquids were thoroughly analyzed by liquid chromatography-tandem mass spectrometer (LC-MS/MS). Moreover, we investigated H 2 O 2 and NO 2 À generation in liquids by electrochemical chromatography detection (HPLC-ECD) and Oxiselect In Vitro Nitric Oxide colorimetric assay, respectively. Proton nuclear magnetic resonance ( 1 H-NMR) was used to determine the structure of PAA samples. Lastly, colorimetric MTS assays were used to evaluate the viability levels of cells incubated in the prepared plasma-activated or non-activated solutions.

Materials and methods
Sample preparation using low-temperature plasma irradiation Plasma was ignited between copper electrodes using a high-voltage alternating current (AC) 60 Hz power supply set at a peak-to-peak voltage of 9 kVp-p. Ar gas with a flow rate of 2.0 standard liters min À1 was used to generate a glow-like discharge plasma ( Figure S1). The plasma source has a multi-hole structure for uniformly distributing flowing gas. The plasma plume extended down approximately 10 mm in length from the plasma head exit. The details of this plasma source were described previously [9]. The electron density was 2 Â 10 16 cm À3 , and the density of oxygen atoms contributing to the generation of reactive species was 4 Â 10 14 cm À3 . A quartz dish was used to irradiate 8 mL of the as-received, Ringer 0 s acetate (Solyugen V R F) solution for 5 min.

Liquid chromatography-mass spectrometry (LC-MS)
Liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) analysis was performed on a SCIEX X500 QTOF mass spectrometer (CB21311709, AB Sciex, USA) equipped with a Turbo V TM Ion Source (Electrospray Ionization; ESI), two ExionLC AC pumps, ExionLC AC Autosampler (Cat. No. 5036650), and PDA detector (5036657) manufactured and distributed by AB Sciex, USA). The chromatographic separation was performed on a 150 mm Â 2.1 mm Sypron AX-2, 5 mm column (GL Sciences). The column temperature was set at 40 C.
The mobile phases consisted of 50/50 H 2 O/CH 3 CN (reagent A) and NH 4 HCO 2 100 mM 50% in H 2 O (reagent B), delivered at a flow rate of 0.20 mL/min. The injection volume was 1 mL, and the total run time was 10 min. The instrument was operated in negative-ion mode. The experimental parameters were set as follows: Curtain Gas at 25 psi, Ion Spray Voltage at À4500 V, and Heater gas temperature at 300 C. Nitrogen was used for the collision gas. The data were acquired and processed with SCIEX OS (AB SCIEX) software (USA). Information Dependent Acquisition (IDA) is an artificial intelligence-based product ion scan mode providing automatic "on-the-fly" MS to MS/MS switching. IDA performed informationdependent scanning at two different fragmentation energies (-10 eV and À35 eV), and two collision-induced dissociation product ion spectra for each of the detected compounds are generated. Therefore, information-rich MS/MS spectra were obtained from precursor ions not known beforehand. The "soft ionization" of the compounds predominantly gives the molecular ion and isotopic peaks which are helpful in the determination of the accurate mass and putative formula of the analyte. When coupled with the fragmentation spectra, it is possible to elucidate the structure of the analyte.
The concentration of acetate and some of the compounds identified in plasma irradiated Ringer 0 s acetate solution was calculated. Non-irradiated samples were prepared using sodium acetate, pyridoxamine dihydrochloride, pyridoxine, acetic anhydride, nitromethane, glycolic acid, sodium pyruvate, and glyoxylic acid monohydrate analytical standards. All samples were filtered over a 0.2 mm PTFE filter (mdi Membrane Technologies). The concentration of the compounds found in plasma irradiated samples was calculated by using the linear regression of the calibration curves performed for the non-irradiated compounds mentioned above. The calibration curves were made by plotting the solutions with known concentrations (0.01 mM À 120 mM) and their integrated areas after MS/MS measurement. The area of the analyte peaks in the calibration curve was plotted against the known concentrations. Subsequently, a line was fitted to the points. The parameters of the linear regression equations (slope, intercept, and correlation coefficient (R > 9.91) were calculated by Sciex OS Analytics software.

Electrochemical chromatography detection
After the plasma irradiation, concentrations of H 2 O 2 were measured by the high-performance liquid chromatography electrochemical detection (HPLC-ECD) system with a column (Inertsil, GL Science, Tokyo), a column oven (CTO-20AC, Shimadzu, Kyoto), a column-ECD oven controller (ATC-700, EiCOM, Kyoto) and an ECD (ECD-700, EiCOM, Kyoto). The electrolyte was 50 mM of sodium sulfate (Na 2 SO 4 ) mixed with 100 lM of EDTAÁ2Na solution (C 10 H 14 N 2 Na 2 O 8 Á2H 2 O) as the mobile phase. The flow rate of the electrolyte was set to 0.5 ml min À1 at a pressure of 12.0 MPa at maximum. The column oven temperature was set at 40 C. A series of diluted H 2 O 2 solutions with known concentrations were prepared as standards with a calibration curve and used to calculate the H 2 O 2 concentration.

Colorimetric assay
Nitrite (NO 2 À ) determination was performed using Oxiselect In Vitro Nitric Oxide (Nitrite/Nitrate) Assay Kit (Cell Biolabs Inc., San Diego, CA, USA). Phosphate buffer solution (PBS) was added to the sample to adjust the pH. The sulfanilamide and N-(1-naphthyl) ethylenediamine dihydrochloride solutions were also added to the sample and maintained for 10 min at room temperature. The calibration curve was made by plotting the solutions with 3 varied concentrations of nitrite standard solution of 140 lM, 35 lM, and 8.75 lM. The NO 2 À concentrations in the plasma irradiated acetated samples were measured in the same way using the Ringer 0 s acetate samples before plasma irradiation as background (only Ringer's acetate solution), subtracted from the absorbance of each measured sample, and calculated using the calibration curve prepared above. Since it absorbs light at a wavelength of 540 nm, the quantitative measurement of nitrite ions in the sample is possible by measuring the absorbance at this wavelength by a plate reader.

Cell viability test
MTS viability tests were performed on MCF-10A (human mammary epithelial cell line) and MCF-7 (breast cancer cells) obtained from the American Type Culture Collection (ATCC; Manassas, VA). The cell lines were maintained in a humidified incubator containing 5.0% CO 2 at 37 C. Cells were suspended in DMEM (Dulbecco's Modified Eagle Medium) with fetal bovine serum and antibiotics. Cell lines were subcultured regularly in flasks and healthy cells were used. The technique was presented in detail in previous work [23,25]. MCF-10A and MCF-7 were suspended in fresh medium (with FBS and P/S) and 200 mL was placed in each well of a 96-well plate. Following incubation for 24 h (37 C, 5% CO 2 ), the medium was replaced with PAA samples or with plasmatreated pyridoxamine dihydrochloride solutions (without or with 0.5 mg/mL catalase), glycolic acid, formic acid, and sodium pyruvate in DMEM (without FBS and P/S) of known concentration in the range of 0.01 mM to 120 mM for 2 h. Then, the culture medium (with FBS and P/S) was replaced with them. MTS assay was used to measure the cell viability after 24 h of incubation with plasma irradiated liquid. After 24 h of incubation, the cell viability was measured by the MTS assay. Mass spectrometry was carried out in informationdependent acquisition (IDA) mode scanning at two different fragmentation energies (À10 eV and À35 eV), and two collision-induced dissociation product ion spectra for each of the detected compounds were generated. Information-rich mass spectra were obtained from ions not known beforehand and identified in the PAA. Among these species, pyridoxamine (4-(aminomethyl) 5(hydroxymethyl)-2-methylpyridine-3-ol;   Figure 2(E)). They were attributed to 2-methyl-1,3-oxazole-4,5-dicarboxylate (C 6 H 3 NO 5 À2 ) and pyridoxal oxime (C 8 H 10 N 2 O 3 Figure S2D). Other intermediates were also detected at m/z 124.8905, assigned to 5-ethoxy-4methyl oxyoxazole (C 6 H 9 NO 2 ) ([M-2H] 2-) eluted at RT 7.33 min ( Figure S2C). The information obtained in the mass spectra and extracted ion chromatograms of PAA related to the compounds mentioned above was also confirmed by analyzing a solution of pyridoxamine dihydrochloride (Á2HClÁxH 2 O) irradiated by plasma for 5 min ( Figures  S3A and B, Figures S4A, B, C, and Supplementary text). Several reaction pathways were similar for PAA and PAP solutions. After 5 min of plasma irradiation, the concentration of pyridoxamine decreased to 0.5 mM, suggesting its decomposition and formation of new chemical structures during the discharge ( Figure S3A).

Cell viability test
Cytotoxic effects of the plasma-derived products:

MTS assay
The cell viability test was furthermore performed by incubating for 24 h the MCF-7 and MCF-10A cell lines  (Figure 3). It was observed that for the serial dilution rates from 4 to 16, both types of cells were killed. The same behavior was observed for the MTS assay performed when PAA was used to incubate the cells (Figure 1). The plasma-activated pyridoxamine solution (PAP) showed enhanced selectivity in destroying the MCF-7 cells for a dilution rate of 32 (14.1% ±13.9 viability), but also for the 64-fold diluted (45.2% ± 13.2) (Figure 3), compared to the PAA, for which the viability of cancer cells was only 30% ± 5 for the 32-fold dilution, while the 64-fold diluted PAA had no antitumor effect (Figure 1). These results show that even if the pyridoxamine was generated in a very low concentration in PAA, it may play a role in the selective killing effect of MCF-7 cells, possibly through intermediate compounds and reaction pathways suggested later in this study.
Additional organic substances detected in the plasma-irradiated acetated solution. Plasma oxidation products Mass-spectrometric analysis (IDA TOF-MS ion scan mode) of acetated solutions of control and irradiated samples gave precursor molecular ions representing acetate. The mass spectra were extracted from the TIC (Figure 2(A)). In the ionization source, acetic acid is deprotonated to acetate (m/z 59.01326, Figure 4(A)). The extracted ion chromatogram shows that acetate was eluted at a retention time of 5.306 min for both samples (Figure 4(B)). The concentration of acetate decreased from 28.27 mM (untreated sample) to 12.22 mM for 5 min PAA. This shows that acetic acid was decomposed under plasma irradiation, resulting in the formation of other chemical species in the plasma. As an example, peracetic acid was detected at m/z 75.0087 ( Figure 5) and generated as a consequence of acetic acid oxidation by H 2 O 2 through the following reaction pathway: H 2 O 2 concentration in the plasma-treated acetated solution was measured by electrochemical detection of high-performance liquid chromatography (HPLC-ECD) method and it was 650.87 mM for 5 min PAA. The hydrogen peroxide generation was modeled in detail in our previous work [29]. The OH radicals and H 2 O 2 react with acetic acid, leading to the generation of peracetic acid (Eq. 1). This compound was also observed in the untreated sample (at a slightly lower intensity), because of electrospray ionization (ESI). Samples introduced through the TurboIon Spray probe of ESI are ionized within the tubing by the application of high voltage (Ion Spray voltage), producing a corona discharge of a nebulized jet of hot dry nitrogen gas. Since the peak intensity is slightly higher for the irradiated sample, it may be assumed that it is also formed in the PAA sample. Figure 5 shows the peaks formed at m/z 87.008 for both control and PAA samples, which could correspond to pyruvate. An enlarged image of the mass spectra  acquired for this compound may be seen in Figure S5A. However, the extracted ion chromatograms (XIC) have not shown any eluted compound for the peak identified at m/z 87.008 Da in the mass spectra but could be observed for the peaks at m/z 86.9756 and 87.0433 at RT 1.56 min and 1.38 min, respectively (Figures S5B, C, and D). To confirm whether the pyruvate can be detected or not, a non-irradiated pyruvate solution (1 mM) was measured by LC-MS/MS. The extracted ion chromatogram showed the pyruvate eluted peak at RT 5.53 min ( Figure S6). Therefore, pyruvate may not be formed in PAA or the amount generated was too low to be detected in the extracted ion chromatograms. It was however detected in the proton nuclear magnetic resonance ( 1 H-NMR) spectra of the PAA irradiated for 5 min ( Figure S7, Supplementary text for Figure S7). The concentration of sodium acetate treated by plasma was much higher (500 mM diluted in 8 mL D 2 O) and the pyruvate could be clearly identified at 2.287 ppm. Along with the pyruvic acid, another new peak was also detected in the spectra at 3.01 ppm, which might be attributed to N,N-dimethylacetamide (C 4 H 9 NO, m/z 86.0241 Da, RT 1.35 min, Figure S8).
Ethyl acetate and 2-oxidoiminoacetate were assigned to the peaks identified at m/z 87.0443 and m/z 86.9756, respectively ( Figure 5). The extracted ion chromatograms of the plasma irradiated samples showed additional peaks at RT 1.3 min and 5.29 min for the compound extracted at m/z 75.008 Da (peracetic acid, Figure S9A). It may be concluded that in plasma there are generated compounds that have a different chemical structure, even if they are identified in the mass spectra at the same m/z value as in the control sample. The type of species corresponding to these eluted peaks is currently under investigation.
Among the chemical structures identified in the mass spectra for the control and PAA was also glycolate (m/z 74.976, Figure S9B) and glyoxylate (m/z 72.94249), as seen in Figure 5. 1H-NMR spectra have also shown a peak at 4.259 ppm corresponding to glycolic acid ( Figure S7). In a previous study of Ringer 0 s lactate irradiated by the same plasma configuration, pyruvic acid,   glycolic acid, and glyoxylic acid (as well as formic acid) were also identified for the plasma irradiated Ringer 0 s lactate samples by NMR measurements [30]. However, their effect on cancer cells was not well understood. In this paper, the antitumoral activities of pyruvic acid, formic acid, glycolic acid, and glyoxylic acid, non-irradiated by plasma, were investigated on MCF-7 cells ( Figure  6(A)). Pyridoxamine was identified in the mass spectra only for the PAA solutions, and its sole effect on the cancer cells, without being exposed to plasma discharge, is also presented in Figure 6 (Figure 6(B)). Higher solution concentrations have led to the destruction of both types of cells, caused by an increased amount of hydrochloride and subsequent bleaching effect. It may be therefore once again concluded that plasma irradiation is of paramount importance in the selective killing effect of MCF-7 cells by potentiating the effect of the chemical species formed in plasma.

Inorganic reactive oxygen and nitrogen species (RONS) and derived organic compounds
A major role of these plasma-derived RONS is assumed in the selective killing of cancer cells [31]. Hydrogen peroxide, hydroxyl radicals, or nitric oxide could be identified in the plasma-treated liquids and assumed to have a significant contribution to the modification of biomolecules. The effect of the plasma-activated acetated solution on the MCF-7 cancer cells is again shown in Figure 7   ( Figure 7(A)). These results show once again that only hydrogen peroxide and nitrite ions containing solutions cannot kill the cancer cells as effectively as the plasmaactivated liquids, which provide the most efficient chemically active species mixture capable of stimulating an enhanced antitumor activity in cancer cells.
The cytotoxic effect of hydrogen peroxide in plasmatreated samples was further investigated by treating the PAA and PAP with catalase (0.5 mg/ml in PAA) for 2 h, before the MCF-7 cells' incubation with these liquids. The hydrogen peroxide concentration was less than 50 nM after catalase treatment. The cytotoxic effect was assessed by the MTS assay (Figure 7(B)). We found that PAP with catalase completely killed the cancer cells, while the PAA treated with catalase showed a cell viability of only 13% ± 6, suggesting that components other than RONS in PAA are responsible for the cytotoxicity of PAA.
Acetic anhydride (C 4 H 6 O 3 , m/z 101.0046), nitromethane (CH 3 NO 2 m/z 60.9927), and ethyl acetate (C 4 H 8 O 2 , m/z 87.0443) were identified in the mass spectra as a consequence of acetate reaction with hydrogen peroxide, hydroxides, and nitric oxide species (discussed later in this work). The cytotoxic effect of these compounds on MCF-10A and MCF-7 cells was investigated by incubating them with the above-mentioned solutions of 0.05 mM, 0.1 mM, 0.5 mM, 1 mM, 5 mM, 10 mM, and 20 mM prepared in Ringer 0 s acetate solutions (Figure 8(A, B)). Acetic anhydride has shown a selective killing effect on cancer cells from 0.05 mM to 0.5 mM. Higher concentrations were lethal for both types of cells, with 5 mM solution killing them completely (Figure 8(A)). The nitromethane solutions of all concentrations had no effect on MCF-10A cells but enhanced the growth of the MCF-7 cells up to $ 170% in the case of 0.05 mM and 0.5 mM solutions. Ethyl acetate (Figure 8(C)) showed a slight selective killing effect on cancer cells for all solution concentrations (0.05 mM À 20 mM). The cytotoxic effect of the 1 mM sample on MCF-7 cells was most pronounced, the viability was 65% ± 5). This suggests that in-plasma-generated compounds may have both inhibitory and stimulatory effects on metabolism or the structural integrity of the cells, depending on their chemical composition.

Discussion
Plasma derivation mechanismsoxygenation, ring-formation, and amination The mechanism of formation of these compounds is suggested as follows. Sodium acetate is generated by reacting acetic acid with sodium hydroxide (water as solvent). Ringer 0 s acetate solutions contain 130 mEq/L of Na þ . This hydroxide forms ethanoic compounds through electrolytic reactions. The electrolytic reactions are well known in Kolb es electrolytic method. First, carboxylates are dimerized via electrochemical decarboxylation. These processes are for the preparation of substituted hydrocarbons from the substituted carboxylic acids by using the electric discharge method where carbon dioxide gas is released [32]: Sodium hydroxide can attack chlorinated ethane in the presence of chlorine and some heat and light.
Reaction 3 in plasma leads to the generation of ethanol. The alcohol is oxidized by oxygen, leading to acetaldehyde.
The reaction of acetaldehyde with ammonia generates alanine (C 3 H 6 NO 2, RT 1.35 min) in a reaction similar to the Strecker amino acid synthesis: Ethyl acetate (m/z 87.0443, RT 1.35 min) is synthesized through the esterification reaction of ethanol with acetic acid: In the presence of a strong base, ethyl acetoacetate (m/z 129.0548) is formed through the condensation reaction of ethyl acetate: Next, ring formation is discussed. Diels-Alder reaction is a synthetic route involving alanine and ethyl acetoacetate through condensation of 4,5-substituted-oxazoles and diverse ene-substrates to synthesize pyridinebased structures [33,34]. Methyl-oxazoles represent the key substrates of this synthetic path. The ring is commonly closed through a reaction with phosphorus pentoxide and oxides of alkaline earth metals [34]. Considering this synthetic strategic pathway, similar modifications may occur in plasma. Alkaline species present in the Ringer 0 s acetate composition can contribute to the ring-closing, with the formation of 2-methyl-1,3-oxazole-4,5-dicarboxylate (C 6 H 3 NO 5 À2 ) and 5ethoxy-4-methyl oxyoxazole (C 6 H 9 NO 2 ). The oxazole compound reaction with 2-methyl-4,7-dihydro-1,3-dioxepine (C 6 H 10 O 2 , m/z 113.0603). The formation pathway mechanism in the plasma of this compound includes acetaldehyde diethyl acetal (m/z 117.1665) and acetaldehyde compounds ( Figure S10, Supplementary text for Figure S10, reaction [1] and Scheme S1). Lastly, pyridoxamine is synthesized in plasma by amination of pyridoxine through alcohol group oxidation, as in Scheme 1. According to this mechanism, pyridoxamine is considered formed through the oxidation of pyridoxine followed by the reaction with amine substrates, with the formation of intermediate compounds, such as an aldehyde (C 8 H 10 NO 3 ) and pyridoxal oxime (C 8 H 10 N 2 O 3 ) (Scheme 1). It should be mentioned that nitrogen gaseous species from the atmosphere environment are necessary for the generation of these compounds. The formation of reactive nitrogen species (such as NO 2 À ) during plasma discharge was discussed in our previous work [25]. A positive reaction outcome was reported to be also ensured by the excess addition of sodium acetate [35,36]. In our case, acetate and sodium were the initial components of the solution used in the experiment. However, the yield was pretty low, the pyridoxamine concentration calculated for the 5 min irradiated acetate solutions was found in the range of 0.003 À 0.01 mM. These results imply that acetate is used for the synthesis of intermediate compounds needed for the synthesis of pyridoxine and pyridoxamine, but also for the generation of other compounds in plasma discharges, as suggested further in this study.

Plasma induction oxygenation of solutes
The acetated solution containing acetate, sodium, potassium, chlorine, and calcium is ionized in the plasma leading to the formation of acetic anhydride (C 4 H 6 O 3 , m/z 101.0046, RT 1.35 min, Fig. S11A) and glycolic acid (C 2 H 4 O 3 , m/z 74.976, RT 1.35 min) ( Figure 5) from peracetic acid, as in the reactions below: The origin of methyl radicals can be explained by the homolytic cleavage of excited state acetic acid with the formation of formyloxyl COOH and methyl CH 3 radicals. The chlorination of acetic acid with acetic anhydride (m/z 101.004) as a catalyst resulted in chloroacetic acid. This compound reacts with sodium hydroxide, leading to sodium chloroacetate formation. The reacidification of sodium chloroacetate is followed by the formation of glycolic acid. Acetate ions are oxidized as presented by Tanaka et al. [22]. Nitrate is oxidizing the acetate through a series of intermediates, such as nitromethane (m/z 60.99271, RT 1.35 min), producing nitroacetate at m/z 103.92062 ( Figure S11B and C) and ammonia [38]:

Plasma induction nitration of solutes
Cytotoxic effect of plasma-activated liquids and derived compounds Plasma irradiated Ringer 0 s acetate solution (PAA) showed a cytotoxic effect on cancer cells when 32-fold diluted in the untreated solution. A major role of these plasma-derived RONS was assumed in the selective killing of cancer cells [31]. Hydrogen peroxide could be identified in the plasma-treated liquids and their effect on the MCF-7 cancer cells was shown in Figure 7(A Plasma-irradiated Ringer 0 s acetate and pyridoxamine solutions (non-diluted in untreated Ringer 0 s acetate) were furthermore treated with catalase to evaluate the cytotoxic effect of hydrogen peroxide on cancer cells. The PAA treated with catalase showed a cell viability of only 13% ± 6, while the PAP with catalase completely killed the cancer cells (Figure 7(B)), suggesting once again that components other than RONS are also responsible for the cytotoxicity of plasma-activated liquids.
Mass spectra of PAA showed that the acetate concentration decreased from 28.27 mM (untreated sample) to 12.22 mM for 5 min PAA, with the formation of numerous other chemical compounds of a more complex structure compared to that of the precursor. This implies the fact that some compounds formed in the plasma, such as acetic anhydride, have the potential of providing good selectivity, as also ethyl acetate, which showed slight selectivity for all solution concentration ranges (0.05 mM À 20 mM) (Figure 8(A,C)). It is well known that acetic anhydride may interact with ethanol (generated in plasma) resulting in the acetic acid formation and esters, such as ethyl acetate, in the presence of pyridine as solvent [39]. In PAA was formed pyridoxamine, which has the functional groups (hydroxy, methyl, hydroxymethyl, and aminomethyl groups) substituted on pyridine. However, the role of pyridoxamine in PAA is still unclear and being investigated. The plasma-activated pyridoxamine solution (PAP) of 0.01 mM was indeed more effective in destroying the MCF-7 cells for a dilution rate of 32 (14.1% ±13.9 viability), and also for the 64-fold diluted (45.2% ± 13.2), possibly through decomposition products (such as pyridoxal oxime) or by favoring the formation in plasma of some other compounds with cytotoxic effect. These results suggest that even if the compounds generated in PAA were found in low concentrations, they may play a role in the selective killing effect of MCF-7 cells through intermediate compounds or by reaction pathways with other structures formed in plasma, intricate mechanisms which are currently under investigation.

Conclusions
It may be concluded that a variety of chemically active species are generated in Ringer 0 s acetate solutions irradiated by plasma, starting from the same precursors, but following different reaction pathways under the influence of electromagnetic fields, electron density, and temperature, as well as excited species formed during treatment. Hydroxyl radicals, sodium hydroxide, hydrogen peroxide, nitrite, and nitrate ions are of paramount importance in the reaction with acetate and the generation of chemical species, such as acetic anhydride, ethyl acetate, pyridoxamine, nitromethane, and other intermediates. These compounds have stimulatory or inhibitory effects on cell viability, depending on the concentration and chemical structure of the generated compounds in the irradiated liquid. Plasma is producing a mixture of chemicals in the exact amount able of inducing a selective cytotoxic effect on cancer cells, without being harmful to normal cells. The adjustment of the plasma parameters may give the possibility of engineering liquids that induce the desired effect on cells. The chemical species found in the mass spectra of PAA using the configuration presented in this paper and the possible reaction pathways occurring in plasma are summarized in Figure 9.