Electron density and gas temperature of different gas mixture plasmas. Optical emission spectra were acquired while irradiating Ringer lactate solutions by discharges in pure Ar, Ar/N2 0.9/0.1, Ar/N2/O2 0.8/0.1/0.1, and Ar/O2 0.9/0.1 gas mixtures at 1 slm of the total gas flow. The discharge voltage (V = 9 kV) and the repetition frequency (f = 9 kHz) were the same for each type of gas mixture. As shown in Fig. 1, the typical emission spectrum at a typical discharge condition (i.e. containing Ar and N2 gas mixture) mainly consists of atomic hydrogen (Balmer series, Hα and Hβ) and atomic oxygen (3p5P → 3s5S0 at 777.5 nm and 3p3P → 3s3S0 at 844.6 nm) emission lines, as well as several molecular rotational bands, such as the OH A2Σ → X2Π (0, 0) band (at about 310 nm), NO A2Σ+ → X2Π γ-system bands (in the 220–260 nm range), as well as the N2 (C2Σ+ → B2Σ+) second positive system bands (visible in the 300–450 nm range). The intensity of the O emission peak at 777 nm increases when O2 is added to the mixture. The traces of N2 (B2 Σ-A2 Σ+) first positive system bands (above 600 nm) and the H2 Fulcher system (at about 600 nm) are also visible under some conditions.
The gas temperature calculated for Ar, Ar/N2, Ar/N2/O2, and Ar/O2 gas mixtures- PAL solutions, is shown in Fig. 2. It should be mentioned that in the water vapor containing discharge the OH rotational distribution may reveal a double slope, which is related to higher rotational-translational (R-T) relaxation rates at low rotational numbers. At low rotational numbers, R-T relaxation is fast, providing the proximity between the “low” rotational temperature and the gas temperature. For high rotational numbers such equilibrium does not happen (the higher the rotational number the higher the deviation from R-T equilibrium) and the rotational temperature given by the Boltzmann plot is always higher than the gas temperature, representing a certain balance between the excitation of the rotational levels and R-T relaxation. In our case the small (but visible) double-sloped nature of the rotational distributions takes place in the OH case, making about a 700 K difference between the low and high rotational temperatures. Keeping in mind the error of temperature determination, it is barely possible to make a conclusion about any temperature trend depending on the studied solution. We can only say that the overall gas temperature is about 1500 K (Fig. 2), whereas the rotational temperature of NO and OH, obtained as a result of a single Boltzmann fit, is about 2200 K. Rotational temperature is somewhat elevated in the presence of oxygen in the gas mixture, which might be related to the excitation kinetics of the OH A2Σ molecular state rather than to changes in kinetic gas temperature.
The trends found for electron density (Ne) are shown in Fig. 3. We found that the double Lorentzian line profile is an optimum fit for the experimentally obtained hydrogen spectral lines, see Fig. 3a. As a result of double fitting, the high- and low- electron populations found are often different by about two orders of magnitude. As clear from Fig. 3a, a single Lorentzian fit is not suitable for our plasma conditions, clearly pointing out two-electron density groups, which likely correspond to the core and its periphery. In Fig. 3b the electron density in the streamer core lies in the range 0.5–1x1016 cm− 3 showing no decent trends. Unfortunately, in spite of a rather precise fitting, the narrow Lorentzians always give the electron density below 1014 cm− 3, which is below the level of applicability of the semi-empirical fitting expressions built-in25. Due to this fact, we can only state that the low-density electron population is below 1014 cm− 3 for all the examined discharge conditions.
Compounds identified in plasma irradiated Ringer`s lactate solution. Total ion chromatograms (TIC) were performed for untreated (control) and plasma irradiated Ringer`s lactate (PAL) solutions for 10 min. The gas mixtures used for these measurements were Ar 1 slm, Ar/N2 0.9/0.1, Ar/N2/O2 0.8/0.1/0.1, and Ar/O2 0.9/0.1. The peaks at retention time (RT) of 1.35 min (A), 5.52 min (C), and 6.61 min (D) for the control and plasma-treated samples correspond to lactate precursor and fragments formed because of electrospray ionization (Fig. 4a). An additional peak was separated at 5.15 min (B) only for the plasma-treated samples, corresponding to compounds formed after the plasma irradiation, lately discussed in this work. A shoulder at RT 7.37 min (E) was observed only for the Ar/N2, Ar/N2/O2, and Ar/O2 gas mixture plasma-treated samples, attributed to fragments and chemical species newly formed in plasma. Although fragmentation occurs in the plasma discharge, some fragment products are also formed during the electrospray ionization (ESI) analysis. ESI is a so-called 'soft ionization' technique since there is little fragmentation. The advantage is that it may produce multiple-charged ions, effectively extending the mass range of the analyzer to accommodate higher orders of the magnitude observed for compounds of higher molecular masses compared to that of lactate ions, possibly formed during plasma irradiation. Therefore, for this study, further analysis of the mass spectra was focused on the identification of the parent ion (lactate) and product compounds corresponding to peaks observed mostly for the plasma-treated samples.
The IDA TOF-MS ion scan mode for the control and PAL solutions gave precursor molecular ions representing lactate (Fig. 4b). The mass spectra were extracted from the TIC shown in Fig. 4a. The lactic acid is deprotonated to lactate, m/z 89.02469 Da (theoretical mass 89.02387 u). Polylactic acid at m/z 187.04289 Da (PLA linear n = 2)30 was also detected for all samples. Extracted ion chromatograms showed that lactate ion eluted at RT 5.5 min for all samples (Fig. 5c). The concentration of lactate decreased from 28.28 ± 4.03 mM to 20.78 ± 3.4 mM, 18.75 ± 2.92 mM, 19.91 ± 3.14 mM, and 20.08 ± 3.06 mM for the Ar, Ar/N2, Ar/N2/O2, and Ar/O2 gas mixture plasma irradiated samples, respectively (Table 1). The decrease in the concentration indicates the decomposition of lactate and the formation of different chemical compounds due to the interaction with the reactive species. 2-hydroperoxyacetate ion (C2H2O4−) at m/z 89.98911 (theoretical mass 89.9953 u) was identified in the mass spectra of Ar, Ar/N2/O2, and Ar/O2 PAL (Fig. 5a).
Sample
|
Lactate
|
Pyruvate
|
Glycerate
|
Citrate
|
Tricarballylate
|
Table 1
Concentration [mM] of lactate, pyruvate, glycerate, citrate, and tricarballylate calculated for the control and plasma irradiated samples
Control
|
28.28 ± 4.03
|
0.38 ± 0.002
|
0
|
0
|
0
|
Ar
|
20.78 ± 3.4
|
0.32 ± 0.002
|
0.18 ± 0.005
|
0.01732 ± 0.002
|
0.75 ± 0.04
|
Ar/N2
|
18.75 ± 2.92
|
1.35 ± 0.04
|
0.11 ± 0.004
|
0.16195 ± 0.03
|
0.66 ± 0.04
|
Ar/N2/O2
|
19.91 ± 3.14
|
1.86 ± 0.05
|
0.09 ± 0.003
|
0.16393 ± 0.02
|
0.72 ± 0.02
|
Ar/O2
|
20.08 ± 3.06
|
1.39 ± 0.06
|
0.04 ± 0.002
|
0.17354 ± 0.01
|
0.52 ± 0.03
|
The cleavage of a methyl radical may be an important step for further structural modifications:
Pyruvic acid (C3H4O3) was found at m/z 87.009 for all samples (Fig. 5b). Pyruvate formation through catalytic oxidation of lactic acid by ·OH radicals formed by atmospheric pressure plasma irradiation was discussed in detail in our previous work18. Pyruvate was identified in the mass spectra for the control sample 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 (IonSpray voltage), producing a corona discharge of a nebulized jet using hot dry nitrogen gas. However, except for the Ar treated sample, the concentration of pyruvate increased after plasma exposure for the Ar/N2, Ar/N2/O2, and Ar/O2 samples, the highest value being measured for the Ar/N2/O2 sample (~ 1.86 mM) (Table 1). The Ar sample showed a very slight decrease to ~ 0.32 mM, compared to ~ 0.38 mM measured for the control sample. The extracted ion chromatograms were also different for the Ar/N2, Ar/N2/O2, and Ar/O2 samples, a shoulder was additionally formed at RT ~ 7.38 min compared to the control and Ar plasma sample, which showed only the common peak at RT 5.45 min (Fig. 5d). This suggests the formation of an additional compound (m/z 87.04632) of a different chemical structure compared to that of pyruvate. Figure 5b also shows an additional peak in the mass spectra of Ar/N2, Ar/N2/O2, and Ar/O2 and it was attributed to ethyl acetate C4H8O2, an ester formed between acetic acid and ethanol (Supplementary Fig. 1) [2].
It may be concluded that argon gas mixed with nitrogen and oxygen produces different structures compared to the samples irradiated only with argon gas. The exact identification and the generation possibility of such additional compounds are currently under investigation.
Furan (C4H4O) at m/z 67.0198 Da and 1,3,4-oxadiazole-2-ol (C2H2N2O2) at m/z 85.0299 were identified in the mass spectra and formed in plasma (Supplementary Fig. 2a and b, Fig. 3, supplementary information text and references1,2, and reaction S1) as a consequence of the substitution of two methylene groups (= CH) from furan with two pyridine type nitrogen’s (-N=)31.
Further examination of the mass spectra revealed a series of other new peaks formed in case of plasma irradiated samples. Glyceric acid C3H5O4 (m/z 105.01977), fumaric acid C4H4O4 (m/z 115.04), and oxaloacetic acid C4H4O5 (m/z 131.0386) were identified in the mass spectra and extracted ion chromatograms after plasma treatment of all samples (Fig. 6a, b, c, d, e, and f). Glycerol C3H8O3 (m/z 91.004) could be generated from the reaction of acrolein C3H4O (m/z 55.0193) with propylene oxide32 C3H6O (m/z 57.035), which through further oxidation in plasma resulted in the formation of glyceric acid (GA):
Glycerate concentration was 0.18 mM for the Ar plasma-treated sample and decreased when nitrogen and oxygen gas mixture plasma was used to irradiate the samples (Table 1). The lowest concentration of glycerate was measured for the Ar/O2 plasma-treated sample (0.04 mM). Glycerate and glycerol interact with oxygen reactive species in plasma [4] forming alcohols and radicals. These species may further interact with HCl and KCN33, resulting in the formation of citric acidC6H8O7 (m/z 191.0206, Fig. 7b and d). Another route of citric acid formation may be through pyruvate-oxalacetate solutions34. Oxaloacetic acid is the result of the condensation of pyruvic acid with carbonic acid (H2CO3) in biochemistry, driven by the hydrolysis of ATP. In plasma, the energy required for the reaction to occur is supplied by the discharge. Oxaloacetate ion (C4H3O5−) could not be identified for Ar irradiated samples (Fig. 6c and f). But pyruvate was found in the argon PAL sample (Figs. 5c and d). It may be assumed that the biosynthesis route of oxaloacetate from pyruvate does not occur in Ar plasma. The formation pathway may instead include oxalate (C2HO4−, m/z 89.024) and acetate (C2H3O2−, m/z 59.015) (oxaloacetate + H2O ⇌ oxalate + acetate), in which the enzymatic process (catalyzed by oxaloacetase)35 acting on C-C bonds may be substituted by the plasma which provides electrons and reactive species able to break the chemical bonds.
Tricarballylic acid (TA) was formed for all plasma irradiated samples, through dehydration of the citric acid to aconitic acid, after which this unsaturated compound has to be hydrogenated to obtain TA (tricarballylate ions: C6H7O6 − 1, m/z 175.02514 and C6H5O6 − 3, m/z 173.026 eluted at RT 5.7 min and 7.07 min (Fig. 7a and c), as shown in reaction [5]. This first dehydration step is an equilibrium reaction and is enhanced at temperatures above 200°C36. The gas temperature in plasma is 2200 K for all gas mixtures used, therefore the conditions for TA generation are favorable in this experimental configuration. It may be observed that the TA is formed also for the Ar sample, and even in the highest concentration (0.75 ± 0.04 mM, Table 1). Therefore, in Ar plasma, only some chemical reactions occur, compared to the other gas mixture plasmas.
The peak area of tricarballylate and glycerate was highest for the Ar sample and lowest for the Ar/O2 samples (Fig. 8). However, some other compounds, such as oxaloacetate, citrate, and fumarate could not be identified in the mass spectra of Ar irradiated samples. The addition of nitrogen and oxygen in a controlled amount is essential for the generation of a certain type of chemical species.
Lactones were formed by intramolecular esterification of the corresponding hydroxycarboxylic acids, which takes place spontaneously when the ring formed is five- or six-membered, in acidic media. Plasma is generating H+ ions, therefore, the D-arabinono-1,4-lactone (C5H8O5, m/z 147.03113, Supplementary Fig. 4) may be easily synthesized from their corresponding hydroxycarboxylic acid (D-arabinonic acid, C5H10O6, m/z 165.03) [6]:
A summary of the key compounds formed in PAL samples is presented in Fig. 9. Pyruvate and acrolein generation in plasma leads to the further synthesis of a series of compounds that may play an important role in the metabolism of the cells incubated with these plasma irradiated liquids.
Chemical compounds formed in plasma have inhibitory or stimulatory effects on normal and cancer cells. The cell viability test was furthermore performed by incubating for 24 hours the MCF-7 and MCF-10A cell lines with D-glyceric acid sodium salt and tricarballylic acid solutions (prepared in Ringer`s lactate) which were not irradiated by plasma (Fig. 10). Concentrations of 0.05 mM, 0.5 mM, 1 mM, 5 mM, 10 mM, and 50 mM for each compound were used in this experiment. The control represents the MCF-7 and MCF-10A cells only incubated in non-treated plasma Ringer lactate solution. It was observed that glyceric acid has no antitumor effect on MCF-7 cells, but on the contrary, it is increasing their viability up to 185% ± 40 (1 mM) (Fig. 10b). The MCF-10A viability increased for all glyceric acid solutions, the highest value being measured for the 0.5 mM sample (360% ± 45).
Tricarballylic acid (TA) showed a more intense cytotoxic effect on both types of cells, killing them completely when incubated with solutions in the range of 1–50 mM concentrations (Fig. 10c). The viability of MCF-7 cells increased when incubated with 0.05 mM (205 ± 43%) and 0.5 mM (160 ± 41%) TA acid solutions, while that of MCF-10A decreased to 71 ± 11% and 16 ± 10%, respectively. Table 1 showed that the concentration of TA was highest (0.75 ± 0.04 mM) for Ar plasma and lowest for Ar/O2 plasma (0.52 ± 0.03 mM). To confirm whether this compound is toxic for the cancer cells in this concentration range, an MTS assay for only MCF-7 cells was performed for solution concentrations of 0.6 mM, 0.7 mM, 0.8 mM, and 1 mM (Fig. 10d).
The cell viability decreased to 10.12 ± 0.5 for the 0.6 mM solution, while for the higher concentrations the cells were entirely destroyed. It may be therefore concluded that D-glyceric acid has a stimulatory effect on the growth of the cells, especially on the MCF-10A cells, while the tricarballylic acid has an inhibitory effect, possibly producing impairments in the mitochondrial tricarboxylic acid cycle and cell death.