*2.2. Nitrite and Acidification Have Synergistic Effects on Ca2+ Influx and Cytotoxicity*

Next, we assessed the effects of nitrate and nitrite on the cytoplasmic Ca2+ levels using the fluorescent calcium indicator fura-2 AM. Cells of the melanoma cell line Mel Im (derived from metastasis) were treated with 1 mM nitrate or 2 mM nitrite in a physiological HEPES-buffered solution. However, no alterations of cytoplasmic calcium concentrations were observed. Since previous studies revealed CAP-induced acidification to be essential for cellular effects of plasma-treated solutions, we combined acidic buffer solutions (pH 3.9) with nitrate and nitrite to resemble these aspects of CAP treatment. Interestingly, a synergistic effect of acidification and nitrite was observed, leading to a strong and immediate increase in cytoplasmic Ca2+ fluorescence (Figure 2A). Acidification alone, however, did not show significant effects. We also investigated the combination of nitrate and acidification but could not detect any increase in cytoplasmic calcium levels (Figure S3). Functional consequences of treatment with acidic nitrite solution were additionally analyzed using the melanoma cell line Mel Juso (derived from the primary tumor). Cell viability was assessed using the tetrazolium-based XTT assay, revealing significant cytotoxicity of a 5 min treatment in Mel Juso (Figure 2B) and Mel Im (Figure S4A). To exclude any contribution of HEPES to this effect, we repeated the experiment with a phosphate-buffered solution without HEPES. Similar cytotoxicity was observed, indicating that HEPES does not actively contribute to the effects of acidic nitrite solutions (Figure 2C and Figure S4B). Since phosphate-buffered solutions are not particularly suited for such low pH levels, all further experiments of this study used HEPES-buffered solutions. On a molecular level, cytotoxicity was accompanied by a strong induction of cell cycle inhibitor p21, which was found both on mRNA and protein levels (Figure 2D–F and Figure S4C–E). Immunofluorescent stainings of promyelocytic leukemia protein (PML) were used to evaluate DNA damage in Mel Juso. The number of PML nuclear bodies was significantly increased in response to treatment with acidic nitrite solution (Figure 2G,H). Finally, we assessed whether acidic nitrite solutions exhibited the same tumor selectivity that was previously reported for CAP. As expected, normal human fibroblasts showed no significant response to the treatment as compared to melanoma cell lines Mel Juso and Mel Im (Figure S4F).

#### *2.3. Molecular Effects of Acidic Nitrite Solution Compared to CAP Treatment*

Multiple studies proposed the generation of peroxynitrite (ONOO-) to be a major consequence of CAP treatment [30,31]. We aimed to estimate intracellular ONOO- generation indirectly via quantification of 3-nitrotyrosin, a marker that was previously reported to be a major consequence of peroxynitrite-dependent protein nitration [32,33]. Therefore, Western blot analysis of 3-nitrotyrosine was performed directly after a 5 min treatment with acidic nitrite solution (Figure 3A). Interestingly, we could not detect relevant amounts of 3-nitrotyrosine even after prolonged treatment periods of 1 h. A physiological buffer solution treated with 2 min CAP was used as a positive control and resulted in protein nitration after it was applied on the cells for 5 min and strong induction of 3-nitrotyrosine after 1 h incubation. Another established feature of CAP was the introduction of membrane damage, which was assessed next. We used a 5 min propidium iodide (PI) staining without fixation or permeabilization to detect ruptures in the cellular membrane. While there was a minor increase in PI signal after treatment with acidic nitrite solution, we could not find significant induction of membrane damage (Figure 3B). Again, a buffer solution treated with 2 min CAP was used as a positive control and led to a significant increase in PI signal. Next, we analyzed MAPK activity by Western blot analysis of ERK1/2 phosphorylation and found a significant elevation of pERK1/2 after 5 min treatment with acidic nitrite solution or indirect CAP treatment (Figure 3C).

**Figure 2.** Nitrite and acidification have synergistic effects on melanoma cells. (**A**) Time course of cytoplasmic Ca2+ levels due to a 4-min treatment with acidic ECS (HCl), nitrite solution (NaNO2), or a combination of both. Ca2+ levels were quantified by calculating the area under the curve (AUC) of the first 120 s after the start of the application (F(2,828) = 1891, *p* <0.0001, *n* = 260–293). (**B**) Cell viability analysis 24 h after a 5 min treatment with untreated ECS (Ctrl) or solutions described in (**A**) (F(3,8) = 10.93, *p* = 0.0033). (**C**) Cell viability analysis 24 h after a 5 min treatment with phosphate-buffered ECS without HEPES (Ctrl) or a combination of acidic phosphate-buffered ECS without HEPES and nitrite (Student's *t*-test). (**D**) Expression analysis of p21 during the time span of 24–72 h after treatment (F(11,24) = 13.82, *p* < 0.0001). (**E**,**F**) Western blot analysis and quantification of p21 protein levels with similar incubation time as (**D**) (F(11,24) = 10.44, *p* < 0.0001). (**G**,**H**) Immunofluorescent stainings of PML and DAPI to assess DNA damage. The amount of PML nuclear bodies was quantified in the bar chart (F(3,8) = 55.13, *p* < 0.0001). Scale bars: 50 µm. Traces are mean with 95% confidence interval, bars are mean ± SEM (ANOVA followed by Tukey's HSD post-hoc test vs. Ctrl, *n* = 3, \*: *p* < 0.05). **Figure 2.** Nitrite and acidification have synergistic effects on melanoma cells. (**A**) Time course of cytoplasmic Ca2+ levels due to a 4-min treatment with acidic ECS (HCl), nitrite solution (NaNO<sup>2</sup> ), or a combination of both. Ca2+ levels were quantified by calculating the area under the curve (AUC) of the first 120 s after the start of the application (F(2,828) = 1891, *p* <0.0001, *n* = 260–293). (**B**) Cell viability analysis 24 h after a 5 min treatment with untreated ECS (Ctrl) or solutions described in (**A**) (F(3,8) = 10.93, *p* = 0.0033). (**C**) Cell viability analysis 24 h after a 5 min treatment with phosphate-buffered ECS without HEPES (Ctrl) or a combination of acidic phosphate-buffered ECS without HEPES and nitrite (Student's *t*-test). (**D**) Expression analysis of p21 during the time span of 24–72 h after treatment (F(11,24) = 13.82, *p* < 0.0001). (**E**,**F**) Western blot analysis and quantification of p21 protein levels with similar incubation time as (**D**) (F(11,24) = 10.44, *p* < 0.0001). (**G**,**H**) Immunofluorescent stainings of PML and DAPI to assess DNA damage. The amount of PML nuclear bodies was quantified in the bar chart (F(3,8) = 55.13, *p* < 0.0001). Scale bars: 50 µm. Traces are mean with 95% confidence interval, bars are mean ± SEM (ANOVA followed by Tukey's HSD post-hoc test vs. Ctrl, *n* = 3, \*: *p* < 0.05).

*2.3. Molecular Effects of Acidic Nitrite Solution Compared to CAP Treatment* 

ment with acidic nitrite solution or indirect CAP treatment (Figure 3C).

Multiple studies proposed the generation of peroxynitrite (ONOO-) to be a major consequence of CAP treatment [30,31]. We aimed to estimate intracellular ONOO- generation indirectly via quantification of 3-nitrotyrosin, a marker that was previously reported to be a major consequence of peroxynitrite-dependent protein nitration [32,33]. Therefore, Western blot analysis of 3-nitrotyrosine was performed directly after a 5 min treatment with acidic nitrite solution (Figure 3A). Interestingly, we could not detect relevant amounts of 3-nitrotyrosine even after prolonged treatment periods of 1 h. A physiological buffer solution treated with 2 min CAP was used as a positive control and resulted in protein nitration after it was applied on the cells for 5 min and strong induction of 3-nitrotyrosine after 1 h incubation. Another established feature of CAP was the introduction of membrane damage, which was assessed next. We used a 5 min propidium iodide (PI) staining without fixation or permeabilization to detect ruptures in the cellular membrane. While there was a minor increase in PI signal after treatment with acidic nitrite solution, we could not find significant induction of membrane damage (Figure 3B). Again, a buffer solution treated with 2 min CAP was used as a positive control and led to a significant

**Figure 3.** Molecular effects of acidic nitrite solution and CAP. (**A**) Western blot analysis of protein nitration directly after 5 min treatment with acidified nitrite solution or indirect CAP (F(6,14) = 9.437, *p* = 0.0003). Reference samples of 1 h treatment serve as a positive control. (**B**) Propidium iodide staining in combination with flow cytometry to assess membrane damage after treatment (F(2,6) = 11.62, *p* = 0.0086). (**C**) Western blot analysis of pERK1/2 and ERK1/2 after treatment with acidic nitrite solution or indirect CAP (F(4,10) = 7.404, *p* = 0.0049). Control treatment (Ctrl) refers to ECS without **Figure 3.** Molecular effects of acidic nitrite solution and CAP. (**A**) Western blot analysis of protein nitration directly after 5 min treatment with acidified nitrite solution or indirect CAP (F(6,14) = 9.437, *p* = 0.0003). Reference samples of 1 h treatment serve as a positive control. (**B**) Propidium iodide staining in combination with flow cytometry to assess membrane damage after treatment (F(2,6) = 11.62, *p* = 0.0086). (**C**) Western blot analysis of pERK1/2 and ERK1/2 after treatment with acidic nitrite solution or indirect CAP (F(4,10) = 7.404, *p* = 0.0049). Control treatment (Ctrl) refers to ECS without nitrite at pH 7.4. Bars are shown as mean ± SEM (ANOVA followed by Tukey's HSD post-hoc test vs. Ctrl, *n* = 3, \*: *p* < 0.05).

#### **3. Discussion**

Generation of nitrite and nitrate after CAP treatment was addressed in some publications before, but their presence and quantity highly depend on the used plasma device and experimental conditions. Since we wanted to combine these molecules with the previously reported acidification after CAP treatment, detection and quantification of both molecules were an essential part of developing a valid RNS-based treatment comparable to CAP. We used Raman spectroscopy to identify long-lived reactive species and found increased levels of nitrate and nitrite, which were then quantified. The use of Raman spectroscopy has shown to be especially practicable due to the all-optical assessment allowing contact-free and label-free quantification of samples. Our results of up to mM ranges are supported by studies on plasma-treated aqueous solutions using other plasma devices [34,35] and display an important baseline for comparison of CAP effects.

When monitoring cytoplasmic calcium levels in melanoma cells, we detected a significant difference between treatments using acidic nitrite and nitrate solutions. It, therefore, seems important to differentiate between these two species when assessing RNS-related effects of CAP, which is not reliably done to date. While it is possible that nitrate contributes to the CAP effect on melanoma cells, most probably due to interaction with ROS and other components of the plasma [36], its role is most definitely minor in comparison to acidic nitrite solutions.

The combination of inorganic nitrite and acidification was previously referred to as *acidified nitrite* [37–39]. On a molecular level, such treatment mainly results in the generation of nitrous acid (HNO2), an unstable compound that degrades to NO and NO2. However, the strong acidification might result in further protonation and production of additional reactive species such as H2NO<sup>2</sup> <sup>+</sup> or N2O<sup>3</sup> [40,41]. Identification and evaluation of these reactive species were not addressed in this study, but since the CAP and nitrite effects both depend on strong extracellular acidification, it is possible that their cytotoxicity is not solely based on HNO2. Acidified nitrite solutions have been well studied due to their antimicrobial activity and positive effects on wound healing, including successful clinical trials using acidified nitrite creams [42,43]. The similarity of these effects with CAP indicates a potential role during plasma treatment. However, the antitumor effects of acidified nitrite are mostly unknown to date. To our knowledge, there is only one study by Morcos et al. [44] showing that 50 µM sodium nitrite inhibits human bladder tumor cells at pH 5.5 to 6 by interfering with DNA replication. However, due to the narrow methodological spectrum of this study and the focus on a physiological rather than therapeutic setting, their work was hardly proof of antitumor effects of acidified nitrite. Nevertheless, it supports our findings and indicates that such treatment might be able to inhibit a wide variety of cancer cells. In the present study, acidified nitrite caused a strong reduction of cell viability in human melanoma cells. Surviving cells were characterized by significant DNA damage and activation of cell cycle inhibitor p21, thereby indicating that this treatment causes lasting damage to melanoma cells. Due to these severe cytotoxic effects, we propose that acidification and nitrite are important components of CAP. Cytotoxicity in normal human fibroblasts was not significant, unlike both melanoma cell lines used in this study, indicating tumor selectivity similar to CAP. However, further comparative studies will be necessary to validate this observation. It was previously reported that HEPES might undergo chemical changes in response to reactive species [45,46]. Since the products of such reaction were found to be cytotoxic, it was necessary to rule out any contribution of modified HEPES to the effects observed in this study. The cytotoxicity of acidified nitrite, however, was still present after interchanging HEPES with a phosphate buffer. We, therefore, conclude that molecular changes of HEPES play a negligible role during our experiments. Thus far, molecular studies of CAP show a strong tendency towards ROS, mainly due to their superior reactivity and cytotoxicity in comparison to RNS. It is, therefore, not surprising that previous studies on CAP-induced nitrite only used it in combination with ROS (namely H2O2) to assess antitumor effects [47,48].

On a molecular level, CAP effects on tumor cells were previously linked to oxidative stress, such as the generation of peroxynitrite (ONOO−) and the resulting formation of nitrated proteins. For example, we recently reported increased levels of 3-nitrotyrosine after short CAP treatment of melanoma cells [28]. Furthermore, several studies proposed a causative role of ONOO− during antimicrobial and even cytotoxic effects of CAP [49,50]. In the present work, we used an experimental setup solely based on RNS and could not detect significant amounts of protein nitration, most probably due to the absence of ROS. Nevertheless, acidified nitrite was found to be strongly cytotoxic, indicating that molecular mechanisms independent of peroxynitrite exist and might also be involved in CAP effects. Another common feature of CAP is the induction of membrane damage in prokaryotic and eukaryotic cells [51,52]. Such alterations of the plasma membrane were previously used to improve drug delivery to cells and through tissues [53] but also have a potential role in antimicrobial and antitumor effects. Interestingly, we could not detect increased membrane damage in response to treatment with acidified nitrite. A possible explanation can be found in the studies of He et al. [52,54], which proposed ROS-dependent lipid peroxidation as the main cause of CAP-induced membrane damage. Our findings of strong cytotoxicity without membrane damage suggest that this process might not be a major contributor to CAP-induced killing of cancer cells. At this stage, however, we are not able to draw a final conclusion on the importance of protein nitration and membrane damage during the antitumor effects of CAP. It is likely that CAP utilizes a broader range of molecular

mechanisms to induce cell death, some of which may depend on protein nitration or membrane damage.

When assessing signaling pathways, we found phosphorylation of ERK1/2 to be increased in response to CAP treatment, which was not described before in tumor cells. However, a few articles reported such an activation of the MAPK pathway in normal cells, leading to diverse cellular effects [55–57]. Since a similar increase in pERK1/2 was found after treatment with acidified nitrite, the MAPK pathway might be involved in the observed antitumor effects. This hypothesis is supported by an already established linkage between MAPK activation and apoptosis [58,59].

In summary, this study highlights the importance of acidified nitrite during CAP treatment and calls for further research on CAP-induced RNS. We could show that acidified nitrite is a potent inhibitor of melanoma cells, although it represents only a fraction of all reactive species involved in CAP. Additionally, the comparison of acidified nitrite and CAP treatment is a useful approach for the identification of molecular mechanisms and their evaluation in the context of antitumor effects. Our observations, therefore, contribute to a better understanding of CAP action on tumor cells and facilitate development of CAP-based anti-cancer therapies.

#### **4. Materials and Methods**

#### *4.1. Chemicals and Solutions*

Extracellular solution (ECS) and phosphate-buffered ECS (pbECS) were prepared as previously described [28]. For ECS, the following chemicals were diluted in bi-distilled water: 145 mM NaCl, 5 mM KCl, 10 mM glucose, 1.25 mM CaCl2, 1 mM MgCl2, and 10 mM HEPES. For preparation of pbECS, 133 mM NaCl, 3.5 mM KCl, 10 mM glucose, 1.25 mM CaCl2, 1 mM MgCl2, 1.5 mM KH2PO4, and 8.1 mM Na2HPO<sup>4</sup> were diluted in bi-distilled water. Both solutions were adjusted to pH 7.4. Sources for further chemicals: Fura-2 AM and pluronic F-127 (Biotium, Fremont, CA, USA), ionomycin (Enzo Life Sciences, Farmingdale, NY, USA), KNO<sup>3</sup> (Carl Roth, Karlsruhe, Germany), NaNO<sup>3</sup> (Acros Organics, Fair Lawn, NJ, USA), NaNO<sup>2</sup> (Sigma Aldrich, Steinheim, Germany), salicylic acid (Carl Roth), Sulfuric acid (Carl Roth, Karlsruhe, Germany), sulfanilamide (Sigma Aldrich, Steinheim, Germany), naphthylethylenediamine dihydrochloride (Sigma Aldrich, Steinheim, Germany).

#### *4.2. CAP Treatment*

CAP treatment of aqueous solutions was described previously [28]. Eight droplets of ECS (20 µL each) were distributed evenly inside a 35 mm petri dish. A MiniFlatPlaster device was placed directly above the dish, resulting in a distance of approximately 10 mm between electrode and sample. Air circulation was minimized by contact of the device to the plastic dish with careful application of pressure. After treatment was finished, all droplets were collected and transferred to reaction tubes for further processing. ECS treated with 2 min CAP was used for the indirect treatment of melanoma cells.

#### *4.3. Raman Spectroscopy*

A laser diode with a wavelength of 785 nm (Laser-785-LAB-ADJ-S, Ocean Optics, Dunedin, FL, USA) was used in combination with a low noise spectrometer (QE65000 Pro-Raman, Ocean Optics, Dunedin, FL, USA) and a Raman probe (General Purpose Raman, RIP-RPB-785-SMA-SMA, Ocean Optics, Dunedin, FL, USA). Before each measurement, a reference 'dark spectrum' was recorded and subtracted. An enclosed aluminum chamber was used to measure liquid samples (Figure S5). In this custom design, a 100 µL sample carrier was placed at the working distance of the Raman probe. The molecular concentration of each solution was quantified before the experiment, and a negative control sample of ultra-pure water (Merck-Millipore Chemicals GmbH, Darmstadt, Germany) served as negative control (0 mmol). From each solution, 100 subsequent spectra were recorded with 2 s integration time each. Data processing of the Raman spectra was performed as

previously described [60]. The raw spectra were cropped to the spectral range 500–1500 cm−<sup>1</sup> . Denoising was achieved by a median filter and discrete wavelet denoising (DWT) (k = 2, Jmax = 2) [61]. The autofluorescence background was modeled by asymmetric least square [62] with λ = 73 and *p* = 0.001 and then subtracted. Finally, each spectrum was normalized to its entire area under the curve. The peak value was determined within the spectral resolution of 12 cm−<sup>1</sup> .

#### *4.4. Photometric Nitrate Assay*

Quantification of nitrate was based on the transnitration of salicylic acid. A volume of 2 µL of the sample solution was combined with 8 µL of 5% salicylic acid in concentrated sulfuric acid and allowed to incubate for 20 min. Then, 200 µL of 8% NaOH was added to achieve basic pH. The solution was mixed thoroughly, measured at 410 nm using a Clariostar Plus Multiplate reader (BMG Labtech, Ortenberg, Germany), and quantified using a standard curve of NaNO3.

#### *4.5. Photometric Nitrite Assay*

A modified Griess diazotization reaction was used to quantify nitrite levels. Briefly, 1.5 µL of the sample solution was transferred to a 96-well plate, followed by 100 µL 1% sulfanilamide in 1 M HCl and 100 µL 0.2% naphthylethylenediamine dihydrochloride (NED) in bi-distilled water. After 15 min incubation, the solution was mixed thoroughly, measured at 540 nm using a Clariostar Plus Multiplate reader, and quantified using a standard curve of NaNO2.

#### *4.6. Cell Culture*

Melanoma cell line Mel Im and normal human fibroblasts were cultivated in DMEM low glucose, while the Mel Juso cell line required RPMI 1640 medium with 2% sodium bicarbonate. All media were supplemented with 10% FCS and 1% penicillin/streptomycin. Cells were incubated at 37 ◦C and 8% CO<sup>2</sup> until approximately 80% confluence. Following a washing step with PBS, a solution of 0.05% trypsin and 0.02% EDTA in PBS was applied to detach the cells. After centrifugation and removal of the trypsin solution, cells were counted using a Neubauer counting chamber. Mycoplasma contamination was regularly excluded for all cell lines. All cell culture chemicals and media were obtained from Sigma Aldrich.

#### *4.7. Calcium Imaging*

The experimental setup and procedures were described elsewhere [27]. Briefly, 200,000 cells were seeded in 35 mm cell culture dishes. On the next day, cells were stained with fura-2 AM (3 µM) in ECS with 0.02% pluronic for 30 min at 37 ◦C and 8% CO2. After a 5 min washing step with ECS, the solution was removed before the dish was mounted on an inverted microscope, and the perfusion outlet was placed within 1 mm distance to the cells. The imaging procedure started with 1 min background measurement, followed by 4 min treatment with the sample solution. During this time, cells were alternatingly excited at 358 nm and 391 nm while recording fura-2 fluorescence. Intracellular calcium levels were evaluated by calculation of the F358 nm/F391 nm ratio. The area under the curve (AUC) refers to the first 120 s after treatment began, relative to the fluorescence 10 s before treatment. To ensure responsiveness of the cells and validate the staining, 2 µM ionomycin was applied after the treatment. Consequently, non-responsive or erratic cells were excluded from analysis. Areas of interest were placed on individual cells, and their fluorescence ratio time courses were calculated after background subtraction. Further information on data evaluation and imaging equipment can be found in a previous publication [63].

#### *4.8. Nitrite Treatment*

A stock solution of 50 mM NaNO<sup>2</sup> was prepared freshly using ECS or pbECS with the according pH. Cells were washed with PBS to remove all cell culture media and cell debris, followed by the addition of a 2 mM NaNO<sup>2</sup> solution. Unless otherwise specified,

treatment duration was 5 min at 37 ◦C. The solution was removed afterward, and cells were cultivated for 24–72 h in their regular cell culture medium.

#### *4.9. Cell Viability Assay*

One day prior to treatment, 6000 cells/well were seeded in a 96-well plate. Following treatment and subsequent incubation for 24 h, cell viability was assessed using the Cell Proliferation Kit II (Roche, Basel, Switzerland) according to the manufacturer's instructions. Photometric detection was realized with a Clariostar Plus Multiplate reader. Absorbance values were normalized to control. The resulting ratios were visualized as % of control.

#### *4.10. Analysis of mRNA Expression by Real-Time PCR*

Approximately 150,000 cells/well were seeded in 6-well plates 1 day before treatment. RNA isolation was performed 24 h, 48 h, and 72 h after treatment using E.Z.N.A.® Total RNA Kit (Omega Bio-Tek, Norcross, GA, USA) according to manufacturer's instructions, followed by cDNA generation using reverse transcriptase reaction as previously described [64]. Real-time PCR was carried out in LightCycler® 480 II devices (Roche, Basel, Switzerland) with forward and reverse primers from Sigma-Aldrich: p21\_for: 5 0 - CGAGGCACCGAGGCACTCAGAGG-30 ; p21\_rev: 50 -CCTGCCTCCTCCCAACTCATCCC-30 ; 18s\_for: 50 -TCTGTGATGCCCTTAGATGTCC-30 ; 18s\_rev: 5 0 -CCATCCAATCGGTAGTAGCG-30 .

#### *4.11. Western Blot Protein Analysis*

Approximately 150,000 cells/well were seeded in 6-well plates 1 day before treatment. Total protein isolation was performed 24 h, 48 h, and 72 h after treatment by addition of radio-immunoprecipitation assay buffer (Roche, Basel, Switzerland) as described elsewhere [65]. Detection of 3-nitrotyrosin required immediate protein isolation after treatment. 20 µg protein were loaded on a 10.00% or 12.75% SDS polyacrylamide gel for electrophoresis and subsequently blotted onto a PVDF membrane (Bio-Rad, Hercules, CA, USA). After a short incubation in methanol, Ponceau S staining was performed to quantify total protein load. Membranes were then washed with double distilled water and incubated in 5% non-fat dried milk/TBS-T for 1 h to block unspecific binding sites. Primary antibodies against p21 (1:5000 in 5% NFDM, Abcam, ab109199), 3-nitrotyrosine (1:1000 in TBST, Merck Millipore, 06-284), pERK and ERK (1:1000 in 5% BSA, Cell Signaling, 4370 and 9102) were incubated overnight shaking at 4 ◦C. Secondary antibodies conjugated to horseradish peroxidase (HRP, Cell Signaling, 7074) were applied for 1 h at room temperature. Visualization of HRP-conjugated antibodies was achieved by the addition of Clarity™ Western ECL Substrate (Bio-Rad) in combination with a Chemostar chemiluminescence imager (Intas, Goettingen, Germany). Signal intensity was then quantified using LabImage software version 4.2.3 (Kapelan Bio-Imaging GmbH, Leipzig, Germany).

#### *4.12. Immunofluorescent Staining*

Approximately 35,000 cells were seeded on 18 mm round coverslips the day before treatment. 24 h after treatment, cells were fixed and stained as previously described [66]. The following antibodies were used: anti-PML (1:200, Santa Cruz, Dallas, TX, USA), Cy3 anti-mouse (1:400, Thermo Fisher, Waltham, MA, USA). Cells were eventually stained with DAPI (1:10,000, Sigma Aldrich, Steinheim, Germany). Final stainings were stored at 4 ◦C and analyzed using an Olympus IX83 inverted microscope in combination with Olympus CellSens Dimension software (Olympus, Tokio, Japan).

#### *4.13. Detection of Membrane Damage*

Approximately 200,000 cells/well were seeded in 6-well plates. After cultivation for 24 h, cells were washed with PBS, treated with each sample solution, and washed again. Staining was achieved by the addition of 1 mL propidium iodide solution (10 µg/mL, PromoCell, Heidelberg, Germany) and 5 min incubation at room temperature. The staining solution was then removed, cells were washed with PBS and detached from the plate

using trypsin. Following another washing step, cells were eventually resuspended in 1% BSA/PBS and analyzed by flow cytometry (LSRFortessaTM, BD Biosciences, San Jose, CA, USA). Data analysis was done using FACSDiva 9.0 software (BD Biosciences).

#### *4.14. Statistical Analysis*

Experimental results were analyzed and visualized using GraphPad Prism 7 software (GraphPad Software Inc., San Diego, CA, USA). If not otherwise specified, at least 3 biological replicates were measured, and statistical analysis was performed by one-way ANOVA. A significant F-test was followed by Tukey's HSD post-hoc tests. A critical value of *p* < 0.05 was set for statistical significance. All results were given as mean ± standard error of the mean (SEM).

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/ijms22073757/s1, Figure S1: Calibration of Raman spectroscopy to potassium nitrate solutions of known concentration, Figure S2: Calibration of Raman spectroscopy to sodium nitrite solutions of known concentration, Figure S3: Acidified nitrate solution does not induce a cytoplasmic Ca2+ release in melanoma cells, Figure S4: Effects of acidic nitrite solution in melanoma cell line Mel Im and normal human fibroblasts, Figure S5: Custom designed experimental chamber for Raman spectroscopy.

**Author Contributions:** Conceptualization, T.Z., A.-K.B.; methodology, T.Z., L.A.G., L.K., O.F., M.J.M.F., A.-K.B.; formal Analysis, T.Z., L.A.G., L.K.; investigation, T.Z., L.A.G., L.K., C.S.; resources, O.F., M.J.M.F., A.-K.B.; writing—original draft, T.Z., L.K., A.-K.B.; writing—review and editing, T.Z., L.A.G., L.K., C.S., S.A., S.K., O.F., M.J.M.F., A.-K.B.; visualization: T.Z., L.K.; supervision, O.F., M.J.M.F., A.-K.B.; funding acquisition, O.F., M.J.M.F., A.-K.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was funded by the German Research Foundation (DFG) (BO1573/23 and TRR241-C01) and the Interdisciplinary Center for Clinical Research (IZKF) Erlangen (D31).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** We thank Michaela Pommer, Ingmar Henz, Sebastian Staebler, Chafia Chiheb and Ines Boehme for technical and methodological assistance and discussions.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Ewa Tyczkowska-Siero ´n <sup>1</sup> , Tadeusz Kału˙zewski <sup>2</sup> , Magdalena Grabiec <sup>2</sup> , Bogdan Kału˙zewski <sup>2</sup> and Jacek Tyczkowski 3,\***


Received: 15 September 2020; Accepted: 27 October 2020; Published: 30 October 2020

**Abstract:** We treated *Candida albicans* cells with a sublethal dose of nonequilibrium (cold) atmospheric-pressure He plasma and studied alterations in the genome of this fungus as well as changes in the phenotypic traits, such as assimilation of carbon from carbohydrates, hydrolytic enzyme activity, and drug susceptibility. There is a general problem if we use cold plasma to kill microorganism cells and some of them survive the process—whether the genotypic and phenotypic features of the cells are significantly altered in this case, and, if so, whether these changes are environmentally hazardous. Our molecular genetic studies have identified six single nucleotide variants, six insertions, and five deletions, which are most likely significant changes after plasma treatment. It was also found that out of 19 tested hydrolytic enzymes, 10 revealed activity, of which nine temporarily decreased their activity and one (naphthol-AS-BI- phosphohydrolase) permanently increased activity as a result of the plasma treatment. In turn, carbon assimilation and drug susceptibility were not affected by plasma. Based on the performed studies, it can be concluded that the observed changes in *C. albicans* cells that survived the plasma action are not of significant importance to the environment, especially for the drug resistance and pathogenicity of this fungus.

**Keywords:** *Candida albicans*; cold plasma treatment; genome; hydrolytic enzyme activity; carbon assimilation; drug susceptibility

#### **1. Introduction**

The application of nonequilibrium (cold) atmospheric-pressure plasma for the inactivation of microorganisms in the areas of medicine, biotechnology, and food processing has attracted a rapidly growing interest in recent years. In particular, much attention has been paid to plasma medicine and healthcare [1–4]. Numerous attempts to use cold plasma in oncology [5], dermatology [6–8], and wound healing [9], as well as in disinfection and sterilization [10,11], are already quite advanced. The studies have long since gone beyond in vitro experiments and are conducted in vivo in animals as well as in humans as clinical trials [12–16]. A significant problem that has emerged as a result of these studies is the finding that plasma treatment does not necessarily kill all microorganisms in the area of plasma action. Some of them survive, and it is suspected that certain changes in their genome and phenotypic traits occur [17,18]. Particularly dangerous changes would be the increase in virulence and drug

resistance, which would place the application of plasma technology in a disadvantageous position in medicine. Although researchers have just turned their attention to the plasma-induced sublethal effects in living organisms [19,20], also studying them on isolated DNA [21–23] or enzymes [24], the precise mechanism of this phenomenon, especially the types of DNA damage, is poorly characterized [25]. This problem, therefore, requires further and more advanced studies. *Int. J. Mol. Sci.* **2020**, *21*, x FOR PEER REVIEW 2 of 15 sublethal effects in living organisms [19,20], also studying them on isolated DNA [21–23] or enzymes

Following this research trend, in this paper, we present the results of studies on *Candida albicans* cells in the sublethal state, which survived plasma treatment, looking for genotypic and phenotypic changes that would be a consequence of such a treatment. *C. albicans* was chosen for two reasons. Firstly, it is a well-characterized microorganism [26], which is recognized as a model in the study of fungal pathogens [27], but most of all—secondly—it is the most important cause of fungal infections in humans [28,29], commonly known as candidiasis, which poses a serious therapeutic problem, among others, due to its constantly increasing resistance to antifungal agents [30–32]. It is not surprising then that new and effective strategies of fighting against candidiasis are being searched for. Considering that superficial candidiasis, mainly skin infections, are the most numerous and widespread group of all fungal infections [33,34], cold plasma, which is especially suitable for surface treatment, seems to be a very promising therapeutic method in this case. As this method is becoming increasingly popular and is beginning to be applied in practice, the determination of changes that may occur in *C. albicans* cells that survived plasma treatment is particularly justified. [24], the precise mechanism of this phenomenon, especially the types of DNA damage, is poorly characterized [25]. This problem, therefore, requires further and more advanced studies. Following this research trend, in this paper, we present the results of studies on *Candida albicans* cells in the sublethal state, which survived plasma treatment, looking for genotypic and phenotypic changes that would be a consequence of such a treatment. *C. albicans* was chosen for two reasons. Firstly, it is a well-characterized microorganism [26], which is recognized as a model in the study of fungal pathogens [27], but most of all—secondly—it is the most important cause of fungal infections in humans [28,29], commonly known as candidiasis, which poses a serious therapeutic problem, among others, due to its constantly increasing resistance to antifungal agents [30–32]. It is not surprising then that new and effective strategies of fighting against candidiasis are being searched for. Considering that superficial candidiasis, mainly skin infections, are the most numerous and widespread group of all fungal infections [33,34], cold plasma, which is especially suitable for surface treatment, seems to be a very promising therapeutic method in this case. As this method is becoming

Several tests with the use of cold plasma have already been carried out on *C. albicans*. Apart from determining the influence of plasma parameters and treatment time on the survival rate of the cells of this fungus using various types of plasma sources [35–39], a few preliminary studies have also been carried out on the changes that occur in sublethal cells. For example, Rahimi-Verki et al. [40] found a reduction in the activity of phospholipase and proteinase enzymes for plasma-treated *C. albicans* samples compared to untreated controls. A decrease in the amount of ergosterol after plasma treatment was also noticed. On the other hand, Borges et al. [41] did not observe the effect of plasma on exoenzyme (phospholipase and proteinase) production but found a promising impact of plasma treatment on morphogenesis, with an almost 40-fold reduction of the filamentation rate compared to the nonexposed group. As can be seen, just taking enzymes as an example, the response of *C. albicans* to plasma treatment is inconclusive and requires further detailed investigation. increasingly popular and is beginning to be applied in practice, the determination of changes that may occur in *C. albicans* cells that survived plasma treatment is particularly justified. Several tests with the use of cold plasma have already been carried out on *C. albicans*. Apart from determining the influence of plasma parameters and treatment time on the survival rate of the cells of this fungus using various types of plasma sources [35–39], a few preliminary studies have also been carried out on the changes that occur in sublethal cells. For example, Rahimi-Verki et al. [40] found a reduction in the activity of phospholipase and proteinase enzymes for plasma-treated *C. albicans* samples compared to untreated controls. A decrease in the amount of ergosterol after plasma treatment was also noticed. On the other hand, Borges et al. [41] did not observe the effect of plasma on exoenzyme (phospholipase and proteinase) production but found a promising impact of plasma treatment on morphogenesis, with an almost 40-fold reduction of the filamentation rate compared to the nonexposed group. As can be seen, just taking enzymes as an example, the response of *C. albicans*

#### **2. Results and Discussion** to plasma treatment is inconclusive and requires further detailed investigation.

treatment.

colonies were collected for further investigations.

Surface plasma treatment of *C. albicans* culture induced the growth inhibition zone with an elliptical-like shape, as shown in Figure 1. The zone edge has been defined as a place where cell survival reaches an average of 10 % compared to the plasma-untreated region [39]. Due to the appropriately selected plasma parameters (see Section 3.1) and the relatively short time of plasma action (see Section 3.2), not all *C. albicans* cells in the elliptical zone were killed. Some of them survived, which were revealed, after incubation, as spot colonies inside this zone (Figure 1). These colonies were collected for further investigations. **2. Results and Discussion**  Surface plasma treatment of *C. albicans* culture induced the growth inhibition zone with an elliptical-like shape, as shown in Figure 1. The zone edge has been defined as a place where cell survival reaches an average of 10 % compared to the plasma-untreated region [39]. Due to the appropriately selected plasma parameters (see Section 3.1) and the relatively short time of plasma action (see Section 3.2), not all *C. albicans* cells in the elliptical zone were killed. Some of them survived, which were revealed, after incubation, as spot colonies inside this zone (Figure 1). These

**Figure 1.** The growth inhibition zone induced by plasma that was generated by the linear microdischarge jet. The red arrow indicates one of the spot colonies, which grew after plasma **Figure 1.** The growth inhibition zone induced by plasma that was generated by the linear microdischarge jet. The red arrow indicates one of the spot colonies, which grew after plasma treatment.

*2.1. Genomic Alterations* 

mitochondrial DNA is not discussed further.

#### *2.1. Genomic Alterations* evaluation. Particular attention was paid to genes known to alter the *C. albicans* physiology— ALS1−ALS12, HWP1, EAP1, ECM33, MP65, PHR1, SAP1−SAP10, SAP30, SAP98, SAP99, SAP155,

The results of nuclear DNA sequencing are presented on a circular plot (Figure 2). The variant calling allows the identification of numerous genomic alterations in each evaluated sample. However, since no genomic changes affecting the mitochondrial genome have been recorded, mitochondrial DNA is not discussed further. SAP190, RSR1, BIG1, LIP1−LIP10, PMT1, PMT4, BCR1, TEC1, ACE2, EFG1, ZAP1, MDR1, CDR1, CDR2, EFG1, CPH1, CPH2, ECE1, TUP1, NRG1, RFG [42], and ERG3, ERG11 [43]). There were no significant changes in the sequence of those genes. A detailed summary of the molecular findings is presented in Table 1.

control sample, but they were significantly less frequent, according to the VarScan software

*Int. J. Mol. Sci.* **2020**, *21*, x FOR PEER REVIEW 3 of 15

The results of nuclear DNA sequencing are presented on a circular plot (Figure 2). The variant calling allows the identification of numerous genomic alterations in each evaluated sample. However, since no genomic changes affecting the mitochondrial genome have been recorded,

In this study, we focused on the single nucleotide variants (SNVs), insertions, and deletions, which were present in two investigated samples after 12 cycles of plasma treatment (12\_1 and 12\_2) and were significantly more frequent than in the control sample. This approach led to the identification of six single nucleotide variants, six insertions, and five deletions. The following genes were affected by high-confidence SNVs: TCP1 (p.Gly>Arg and p.Gly>Glu), UBP1 (p.Gln>STOP), and PUF3 (p.Gly>Gly). Short insertions were present in the sequence of genes: C5\_03980W\_A and C7\_02250W\_A. The sequence of C7\_01080C\_A was interfered with by deletion. All of the

**Figure 2.** Genome diagram: (**1**) *C. albicans* SC5314 A22 ideogram; (**2**) control sample coverage; (**3**) 12\_1 sample coverage; (**4**) 12\_2 sample coverage; (**5**) chromosomal coordinates; (**6**) genomic alternations (grey—low confidence single nucleotide variants (SNVs), red—high confidence SNVs, blue deletions, green—insertions); (**7**) genomic alternations statistically significant (red—high confidence SNVs, blue—deletions, green—insertions); (**8**) genes affected by high-confidence SNVs (red), deletion (blue), insertions (green). **Figure 2.** Genome diagram: (**1**) *C. albicans* SC5314 A22 ideogram; (**2**) control sample coverage; (**3**) 12\_1 sample coverage; (**4**) 12\_2 sample coverage; (**5**) chromosomal coordinates; (**6**) genomic alternations (grey—low confidence single nucleotide variants (SNVs), red—high confidence SNVs, blue—deletions, green—insertions); (**7**) genomic alternations statistically significant (red—high confidence SNVs, blue—deletions, green—insertions); (**8**) genes affected by high-confidence SNVs (red), deletion (blue), insertions (green).

In this study, we focused on the single nucleotide variants (SNVs), insertions, and deletions, which were present in two investigated samples after 12 cycles of plasma treatment (12\_1 and 12\_2) and were significantly more frequent than in the control sample. This approach led to the identification of six single nucleotide variants, six insertions, and five deletions. The following genes were affected by high-confidence SNVs: TCP1 (p.Gly>Arg and p.Gly>Glu), UBP1 (p.Gln>STOP), and PUF3 (p.Gly>Gly). Short insertions were present in the sequence of genes: C5\_03980W\_A and C7\_02250W\_A. The sequence of C7\_01080C\_A was interfered with by deletion. All of the abovementioned mutations were not exclusive to the tested samples. They were also present in the control sample, but they were significantly less frequent, according to the VarScan software evaluation. Particular attention was paid to genes known to alter the *C. albicans* physiology—ALS1−ALS12, HWP1, EAP1, ECM33, MP65, PHR1, SAP1−SAP10, SAP30, SAP98, SAP99, SAP155, SAP190, RSR1, BIG1, LIP1−LIP10, PMT1, PMT4, BCR1, TEC1, ACE2, EFG1, ZAP1, MDR1, CDR1, CDR2, EFG1, CPH1, CPH2, ECE1, TUP1, NRG1, RFG [42], and ERG3, ERG11 [43]). There were no significant changes in the sequence of those genes. A detailed summary of the molecular findings is presented in Table 1.


*Int. J. Mol. Sci.* **2020**, *21*, 8100

– C7\_02250W\_A Chr7B:490089 +TTCCAA 0.0022 0.0208

CDC317: CPAR2\_301140, *C. tenuis* NRRL Y-1498: CANTEDRAFT\_ 135055, and *Debaryomyces hansenii* CBS767: DEHA2E07678g

The genome of *C. albicans* has a length of 28,605,418 bp (including the nuclear and mitochondrial genome) and consists of eight pairs of chromosomes [44]. A characteristic feature of this organism is the occurrence of frequent genetic alternations (translocations, insertions, and deletions on both the chromosomal and nucleotide levels), which are part of its strategy to adapt to environmental conditions [45]. The abovementioned circumstances, along with the alternative codon usage (CUG translated into serine rather than leucine), make it a challenging organism to study.

In the performed analysis, we identified several changes in the nuclear genome of the examined strain. During the analysis, we proved the presence of thousands of SNVs, insertions, and deletions in each of the samples, including the control. The genomic alternations presented in the results were selected after a detailed statistical analysis, taking into account the methodology of the experiment. However, there is no evidence that any of them appeared as a result of plasma treatment, and it is not simply a matter of coincidence. To gain a deeper understanding of possible plasma-induced genetic changes in *C. albicans*, phenotypic characteristics such as carbon assimilation, enzyme activity, and susceptibility to antifungal drugs should be investigated, and, if changes occur due to the plasma treatment, to try to link them to the genetic changes. The following part of this paper presents the results of such studies.

#### *2.2. Phenotypic Changes*

#### 2.2.1. Carbon Assimilation and Hydrolytic Enzyme Activity

Tests of the ability to assimilate carbohydrates as a carbon source did not show any changes in the investigated *C. albicans* strain after repeated sublethal treatments with the use of cold atmospheric plasma. The analyzes were performed for the untreated culture as well as for cultures after 1, 7, and 12 plasma treatments. Both before and after the plasma treatment, the studied fungal cells were characterized by the ability to assimilate carbon from the same 13 compounds. As an example, Table 2 shows the API 20C AUX test results for the untreated culture and after 12 cycles of plasma treatment. A plus sign indicates that *C. albicans* grows on this carbohydrate medium.


**Table 2.** The ability of the tested strain to absorb carbon from 19 carbohydrates, assessed using the API 20C AUX system.

The situation is different, however, in the case of the activity of hydrolytic enzymes, which is associated with the virulence of *C. albicans* [46–48] and therefore requires more careful analysis. Among 19 tested enzymes (Table 3), 10 of them showed activity in the investigated strain of *C. albicans*. In nine cases, the activity decreased in the course of successive cycles of plasma treatment, even falling to zero, while in one case, it increased. Relative average activities for successive plasma treatment cycles of these enzymes are shown in Figure 3.


**Table 3.** List of hydrolytic enzymes tested, using the API ZYM system and the characteristic wavelengths, at which the absorbance of the products of the enzymatic reactions was measured.

After 12 cycles, further screening tests were performed without plasma to determine the persistence of the observed changes. As shown in Figure 4, the activity of the enzymes decreased due to the plasma action returning to the initial state, while the activity of the enzyme No. 12 (naphthol-AS-BI-phosphohydrolase), which increased, remained unchanged. The next 12 screenings did not change the activity of enzyme No. 12, indicating a permanent change.

Based on these results, we have two problems to solve. First, what is the reason for the unstable decrease in the activity of nine of the tested enzymes, which returned to their original state after the first screening without plasma treatment? Secondly, why did enzyme No. 12 permanently change its activity?

The reversible changes in enzyme activity, observed for nine enzymes, may be due to cell-to-cell transmission of information from dying cells about danger during the plasma action [49], for example, via cytoplasmic flow [50]. It can also be caused by stress induced by changes in the environment [51], for example, the byproducts of killed cells, which are constantly present during the growth of colonies after plasma treatment. The fact that virtually all cells growing in a given colony revealed reduced enzyme activity directly after plasma treatment and while being reinoculated onto a new medium, they forgot this change and returned to the state of original activity, indicates a stress effect that is not a reflection of genetic changes. To find out more precisely what could be causing the stress, the strain was inoculated on the medium that had been treated with plasma for 12 min. No changes in enzyme activity were observed in this case. This indicates that it is not the changes in the substrate due to the action of the plasma but the products derived from killed cells that are a source of stress.

cycles.

*Int. J. Mol. Sci.* **2020**, *21*, x FOR PEER REVIEW 7 of 15

**Figure 3.** Activities of hydrolytic enzymes for *C. albicans* as a function of the number of plasma **Figure 3.** Activities of hydrolytic enzymes for C. albicans as a function of the number of plasma cycles.

*Int. J. Mol. Sci.* **2020**, *21*, x FOR PEER REVIEW 8 of 15

**Figure 4.** Summary of normalized activities of hydrolytic enzymes for *C. albicans*, before plasma treatment, after 12 plasma cycles, and after subsequent plasma-free screenings. All activity values for active enzymes before plasma treatment were normalized to 1. **Figure 4.** Summary of normalized activities of hydrolytic enzymes for *C. albicans*, before plasma treatment, after 12 plasma cycles, and after subsequent plasma-free screenings. All activity values for active enzymes before plasma treatment were normalized to 1.

2.2.2. Susceptibility to Antifungal Drugs Susceptibility studies of *C. albicans* to antifungal drugs as a function of the number of cycles of plasma treatment did not reveal any differences in comparison to the untreated culture. Table 4 shows the MIC (minimum inhibitory concentration) values determined for the tested drugs for the untreated culture and the culture after 12 plasma treatment cycles, i.e., after the highest plasma exposure we have used for cells that survived such treatment. These results were confirmed by repeating the measurement series three times. Although the dose of plasma in each cycle was large On the other hand, the sustained increase in naphthol-AS-BI-phosphohydrolase activity after plasma treatment seems to have a genetic basis. Changes in the genome could be generated by both the cell-to-cell communication process during plasma operation and the stress induced by plasma products. Unfortunately, the lack of a connection in the literature between this enzyme and specific genetic sequences makes it difficult to determine the exact mechanism that caused these changes. However, it should be added that the naphthol-AS-BI-phosphohydrolase enzyme is not, so far, considered to be an important hydrolytic enzyme that affects the virulence of *C. albicans* [48,52].

#### enough to kill the vast majority of cells in its area of action (Figure 1), it did not alter drug 2.2.2. Susceptibility to Antifungal Drugs

susceptibility in this small number of surviving cells. No increase in drug resistance for cells that have survived the plasma action bodes well for the future progress of using plasma techniques in the fight against superficial candidiasis. **Table 4.** Minimum inhibitory concentration **(**MIC) values (μg/mL) of antifungal agents for the investigated *C. albicans* strain. **Antifungal Agent Before Plasma Treatment [MIC Value] After 12 x Plasma Treatment [MIC Value]**  Voriconazole 0.094 0.094 Susceptibility studies of *C. albicans* to antifungal drugs as a function of the number of cycles of plasma treatment did not reveal any differences in comparison to the untreated culture. Table 4 shows the MIC (minimum inhibitory concentration) values determined for the tested drugs for the untreated culture and the culture after 12 plasma treatment cycles, i.e., after the highest plasma exposure we have used for cells that survived such treatment. These results were confirmed by repeating the measurement series three times. Although the dose of plasma in each cycle was large enough to kill the vast majority of cells in its area of action (Figure 1), it did not alter drug susceptibility in this small number of surviving cells. No increase in drug resistance for cells that have survived the plasma action bodes well for the future progress of using plasma techniques in the fight against superficial candidiasis.


Caspofungin 0.125 0.125 Amphotericin B 0.125 0.125 **Table 4.** Minimum inhibitory concentration (MIC) values (µg/mL) of antifungal agents for the investigated *C. albicans* strain.

Fluconazole 2.0 2.0

details on the design of the plasma jet, its principle of operation, and process characteristics can be found in [39]. In this study, plasma was generated at 13.56 MHz in helium as a working gas. The helium flow rate was 1.9 L/min, while the discharge power was set at 17 W. The plasma beam had a cylindrical shape, with a length of 40 mm and a diameter of about 1.5 mm, and was aligned parallel

#### **3. Materials and Methods**

#### *3.1. Plasma Source*

As a source of nonequilibrium (cold) atmospheric plasma for the treatment of *Candida* cells, we used a linear microdischarge jet (produced in-house), sometimes called a plasma razor jet [53]. More details on the design of the plasma jet, its principle of operation, and process characteristics can be found in [39]. In this study, plasma was generated at 13.56 MHz in helium as a working gas. The helium flow rate was 1.9 L/min, while the discharge power was set at 17 W. The plasma beam had a cylindrical shape, with a length of 40 mm and a diameter of about 1.5 mm, and was aligned parallel to the treated surface at a distance of 5 mm. The experimental system is shown in Figure *Int. J. Mol. Sci.* **2020**, *21*, x FOR PEER REVIEW 5. 9 of 15

**Figure 5.** The experimental system: (**a**) schematic diagram of the setup; (**b**) photo of the plasma razor jet. **Figure 5.** The experimental system: (**a**) schematic diagram of the setup; (**b**) photo of the plasma razor jet.

#### *3.2. Microbiological and Plasma Treatment Procedures 3.2. Microbiological and Plasma Treatment Procedures*

The reference strain of the American Type Culture Collection (ATCC) *Candida albicans* ATCC® 10231 (National Collection of Pathogenic Fungi (NCPF), Salisbury, UK) was used in this study. A 24 h culture of the fungus for further studies was prepared by uniformly spreading 100 μL of phosphate buffer solution (PBS) containing 5 × 107 CFU/mL (5 units of the McFarland scale, determined with a DEN-1B type densitometer, Biosan, Riga, Latvia) onto a Petri dish with Sabouraud dextrose agar (bioMérieux, Marcy-l'Etoile, France), after which it was exposed to the plasma treatment for 1 min and then incubated at 37 °C for 24 h. *C. albicans* cells that were not killed in the plasma-induced growth inhibition zone were the protoplasts of growing spot colonies, which were the starting material for further studies. They also served as the basis for the preparation of the next culture and its plasma treatment, exactly as it was done the first time. This procedure was repeated 12 times in each measurement series. Then, subsequent screenings without plasma treatment were performed to determine the durability of occurring changes that had taken place under the influence of the plasma. As control samples, a 24-h culture of the fungus was used. The reference strain of the American Type Culture Collection (ATCC) *Candida albicans* ATCC® 10231 (National Collection of Pathogenic Fungi (NCPF), Salisbury, UK) was used in this study. A 24-h culture of the fungus for further studies was prepared by uniformly spreading 100 µL of phosphate buffer solution (PBS) containing 5 <sup>×</sup> <sup>10</sup><sup>7</sup> CFU/mL (5 units of the McFarland scale, determined with a DEN-1B type densitometer, Biosan, Riga, Latvia) onto a Petri dish with Sabouraud dextrose agar (bioMérieux, Marcy-l'Etoile, France), after which it was exposed to the plasma treatment for 1 min and then incubated at 37 ◦C for 24 h. *C. albicans* cells that were not killed in the plasma-induced growth inhibition zone were the protoplasts of growing spot colonies, which were the starting material for further studies. They also served as the basis for the preparation of the next culture and its plasma treatment, exactly as it was done the first time. This procedure was repeated 12 times in each measurement series. Then, subsequent screenings without plasma treatment were performed to determine the durability of occurring changes that had taken place under the influence of the plasma. As control samples, a 24-h culture of the fungus was used.

#### *3.3. DNA Isolation and Genome Bioinformatics Analysis 3.3. DNA Isolation and Genome Bioinformatics Analysis*

treatment.

The molecular genetic research was based on a comparison between the control sample ("control") and two test samples after 12 exposure periods ("12\_1" and "12\_2") of the selected *Candida albicans* strain. The analysis aims to identify possible genetic changes caused by cold plasma The molecular genetic research was based on a comparison between the control sample ("control") and two test samples after 12 exposure periods ("12\_1" and "12\_2") of the selected *Candida albicans* strain. The analysis aims to identify possible genetic changes caused by cold plasma treatment.

which the CTAB extraction buffer (OPS Diagnostics LLC, Lebanon, NJ, USA), proteinase K (Promega, Madison, WI, USA), and RNase-A (Promega, Madison, WI, USA) were added. The prepared material was transferred to sterile Eppendorf tubes and incubated in a ThermoMixer C (Eppendorf, Hamburg, Germany) at 70 °C for 30 min. The tubes were centrifuged in an Eppendorf Centrifuge. The collected 200 μL of supernatant fluid was transferred to Maxwell® 16LEV Plant DNA Kit cartridges (Promega, Madison, WI, USA), and the isolation continued in a Maxwell® 16 machine (Promega, Madison, WI, USA). The amount and purity of DNA were assessed using a Nanodrop ND 2000C (ThermoFisher Scientific, Waltham, MA, USA) and a Qubit fluorimeter (ThermoFisher Scientific, Waltham, MA, USA). Sequencing libraries were prepared using the TruSeq DNA PCR-free reagent set (350 bp insert;

The DNA isolation was performed with 50 mg of 24-h cultures of the *C. albicans*, before and after plasma treatment. The biological material was triturated in liquid nitrogen, in a sterile mortar, to which the CTAB extraction buffer (OPS Diagnostics LLC, Lebanon, NJ, USA), proteinase K (Promega, Madison, WI, USA), and RNase-A (Promega, Madison, WI, USA) were added. The prepared material was transferred to sterile Eppendorf tubes and incubated in a ThermoMixer C (Eppendorf, Hamburg, Germany) at 70 ◦C for 30 min. The tubes were centrifuged in an Eppendorf Centrifuge. The collected 200 µL of supernatant fluid was transferred to Maxwell® 16LEV Plant DNA Kit cartridges (Promega, Madison, WI, USA), and the isolation continued in a Maxwell® 16 machine (Promega, Madison, WI, USA). The amount and purity of DNA were assessed using a Nanodrop ND 2000C (ThermoFisher Scientific, Waltham, MA, USA) and a Qubit fluorimeter (ThermoFisher Scientific, Waltham, MA, USA). Sequencing libraries were prepared using the TruSeq DNA PCR-free reagent set (350 bp insert; Illumina, San Diego, CA, USA), and the type of prepared libraries was Illumina-Shotgun. Next-generation sequencing was carried out on the HiSeq X Ten platform (Illumina, San Diego, CA, USA) to generate 2 × 150 bp paired-end reads, assuming a path coverage in the flow cell for each sample of 20 %.

The quality of the obtained raw reads was checked using FastQC v0.11.4 software [54]. Then, the adapter sequences and low-quality regions of raw reads were trimmed using the Trimmomatic v0.36 tool [55], with the following operating parameters: initial and final regions quality >20, average read quality >30, minimum reads length = 90. The quality of processed reads has been confirmed by rechecking the samples with the FastQC program. The obtained reads were mapped using STAR mapper v2.7.3a [56] to the current, up-to-date version of the *C. albicans* reference genome (*C. albicans* SC5314 A22) taken from the Candida Genome Database [44], implementing the default operating parameters. To identify single nucleotide variations, as well as insertions and deletions in the nuclear sequences, the data was converted to mpileup format using SAMtools v1.10 software [57]. The variant calling and comparison between the samples was performed by VarScan v2.4.4 [58], with default parameters. The effect of selected SNVs on protein products was evaluated using IGV browser v2.7.2 [59]. The obtained data were visualized in the Perl environment with the usage of Circos v.0.69.9 [60]. The coverage data for visualization was obtained by deepTools v2.0 [61].

#### *3.4. Carbohydrate Assimilation Test*

To evaluate the changes in the biochemical properties of the strain after plasma treatment, investigations were carried out using the API 20C AUX system (bioMérieux, Marcy-l'Etoile, France). It is a biochemical identification series based on the assessment of the ability of fungi to absorb carbon (auxanogram) from 19 compounds, i.e., D-glucose (GLU), glycerol (GLY), 2-keto-D-gluconate (2KG), L-arabinose (ARA), D-xylose (XYL), adonitol (ADO), xylitol (XLT), D-galactose (GAL), inositol (INO), D-sorbitol (SOR), methyl-αD-glucopyranoside (MDG), N-acetyl-D-glucosamine (NAG), D-cellobiose (CEL), D-lactose (LAC), D-maltose (MAL), D-sucrose (SAC), D-trehalose (TRE), D-melesitose (MLZ), and D-raffinose (RAF). The API 20C AUX strip consists of 20 cupules containing the 19 dehydrated substrates and a place for the control sample, where assimilation tests are performed. A semiliquid starvation medium is introduced into the cupules. Yeast-like fungi grow if they can use a given substrate as the only carbon source. The API strips were prepared according to the manufacturer's instructions and incubated at 30 ◦C for 72 h, after which the carbohydrate assimilation patterns were read.

#### *3.5. Hydrolytic Enzyme Activity*

The hydrolytic enzyme activities for the investigated fungal strain were determined using an API ZYM system (bioMérieux, Marcy-l'Etoile, France). The system enables rapid determination of the activity of 19 enzymes (5 peptidases, 3 lipases, 3 phosphatases, and 8 carbohydrases) using very small amounts of unpurified samples. The list of the enzymes is presented in Table 3. Although it is a qualitative test based on visual comparison of the colors produced by enzymatic reactions, with a printed color standard to more quantitatively determine the enzyme activities, we used spectrophotometric analysis.

Based on the manufacturer's instructions, enzymatic activity measurements were carried out by suspending a given fungal sample in 2 mL of sterile distilled water with turbidity between 5 and 6 on the McFarland scale and then transferring 65 µL of this suspension to each cupule of the test strip. The strips were incubated for 4 h at 37 ◦C. After incubation, one drop of ZYM A reagent (trihydroxymethylaminomethane, 37% hydrochloric acid, lauryl sulfate, distilled water) and one drop of ZYM B reagent (fast blue 2 BB, methoxyethanol) were added to each cupule to stop the reaction. After 5 min of color development, the test strip was exposed to intense visible light for 10 s to eliminate a yellow tint due to an excess of unreacted fast blue 2 BB. Then, a 5-µL sample was taken from each cupule and placed in a microvolume Nano Stick-S for UV–vis spectrophotometry (PIKE Technologies, Fitchburg, WI, USA). The absorbance of each sample was measured using a spectrophotometer UV–VIS Jasco V-630 (ABL&E-JASCO Polska, Cracow, Poland) for the wavelength characteristic of the given enzyme reaction (Table 3) [62]. Two identical tests were performed at the same time for each sample, while each series of measurements was repeated three times. The maximum relative uncertainty in determining the activity of a given enzyme for a given sample is estimated to be ±5%.

## *3.6. Susceptibility Testing*

The in-vitro activity of typical antifungal agents (voriconazole, fluconazole, amphotericin B, caspofungin, micafungin, and anidulafungin) was determined by the Etest strips (bioMérieux, Marcy- l'Etoile, France). This method consists of placing a narrow plastic strip soaked with a given agent of increasing concentration along this strip on a fungal culture. This allows us to estimate the minimum inhibitory concentration (MIC), which is a measure of drug activity. Inoculum suspensions in sterile saline (0.85 % NaCl) were prepared from primary and plasma-treated *C. albicans* cultures, with an optical density of 0.5 McFarland standard (approximately 5 <sup>×</sup> <sup>10</sup><sup>6</sup> CFU/mL). The suspensions were inoculated directly onto plates, with RPMI-1640 agar (bioMérieux, Marcy-l'Etoile, France) as the base medium on which the Etest strips were placed, according to the manufacturer's instructions. MIC values were read 24 h after incubation at 35 ◦C, accurate to scale on the Etest strips. Each series of measurements was repeated three times.

#### **4. Conclusions**

One of the problems associated with the development of plasma medicine is the risk that microorganisms surviving the plasma action may unfavorably change their characteristics; for example, their virulence or resistance to drugs may increase. The studies conducted on the *C. albicans* fungus have not confirmed these concerns. Although both genotypic and phenotypic changes were observed as a result of repeated plasma treatment, they did not have a significant effect on the virulence and drug susceptibility of the tested strain.

Among the observed changes, the alteration in enzyme activity is of particular interest. The results showed that 9 out of 19 tested enzymes reduced, some significantly, their activity after plasma treatment, which could be interpreted as a decrease in the virulence of this fungus. However, when such cells were inoculated onto a new medium devoid of plasma products, for example, killed cells, the activities of these enzymes returned to their original state. This most likely indicates the occurrence of a stress effect, which should be further investigated in more detail.

Another interesting result is the permanent increase in the activity of one of the enzymes (naphthol-AS-BI-phosphohydrolase) under plasma treatment. Multiple screening of cells on the medium, which had not been in contact with the plasma, did not change the elevated enzyme activity. This permanent shift in the activity must result from alterations in the genotype. Additionally, although we have identified some of the genetic alterations that were possibly plasma-induced in *C. albicans*, their link to the change in enzyme behavior, as well as explaining the genesis of this phenomenon, requires further research.

In summary, it should be noted that the performed studies, on the one hand, have shown changes in both the genotype and phenotype of *C. albicans* under the influence of plasma, which, however, are not dangerous in terms of virulence and drug resistance, and, on the other hand, have revealed some interesting effects related to the activity of hydrolytic enzymes.

**Author Contributions:** E.T.-S. contributed to manuscript writing, plasma treatment, measurements of carbon assimilation from carbohydrates, hydrolytic enzyme activity, and drug sensitivity; T.K. contributed to manuscript writing and genetic analysis; M.G. contributed to genetic analysis; B.K. was the consultant in the field of molecular genetic research; J.T. conceptualized, supervised, wrote, and edited the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was financially supported by the Ministry of Science and Higher Education (MNiSW) as part of statutory research (No. 7189 and No. 7297 (2018)).

**Acknowledgments:** The authors would like to thank Jan Paweł Jastrz˛ebski (Department of Plant Physiology, Genetics and Biotechnology University of Warmia and Mazury in Olsztyn, Poland) for valuable discussions and Ryszard Kapica and Justyna Markiewicz (Department of Molecular Engineering, Lodz University of Technology, Poland) for their technical assistance.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **Abbreviations**


#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
