The relation between doses or post-plasma time points and apoptosis of leukemia cells induced by dielectric barrier discharge plasma

Recently, the application of low-temperature dielectric barrier discharge (DBD) plasma in biomedicine has been drawn much attention from biologists, physicists and clinicians. Plasma contains large amounts of reactive oxygen species (ROS) and reactive nitrogen species (RNS), as well as other short- and long-lived active substances, such as O, O₂, O₂⁻, OH, NO and NO₂.^1,2 Atmospheric pressure plasma has the advantages of high density and rich chemical agents without elevation of the substrate temperature.

Many studies suggest that atmospheric pressure plasma can effectively sterilize medical equipment, biological materials, food packaging materials, and other items;^3–6 that it can rapid coagulate on biological tissue;^7,8 and that it can be used to disinfect teeth or treat dental caries.^9–11 More recent studies have revealed that plasma can also induce tumor cell death; hence, it is being explored as a new cancer treatment method. So far, low-temperature plasma has been found to exhibit a strong inhibitory effect on brain tumors, oral cancer, ovarian cancer, skin melanoma and other malignant cell types.^12–17 When applied to human bronchial epithelial cells (HBE) and lung cancer cells (SW900 line), Keidar, et al. observed low-temperature plasma is capable of killing cancer cells without damaging human bronchial epithelial cells.¹⁸ 

Although the latest studies show low-temperature plasma can induce tumor cell death, the apoptotic mechanisms underlying this process remain unclear. Thus, in our study, the impact of dielectric barrier discharge (DBD) low-temperature plasma on leukemia cells is investigated at different doses, using the rat acute myeloid leukemia LT-12 cell line as a model. We observed the occurrence of apoptosis at different doses, as well as changes of apoptotic gene and protein expression after plasma treatment.

The plasma instrument used in experiments, shown in Figure 1, was purchased from Nanjing Suman Plasma Technology Co., Ltd. (China), including voltage regulator, high voltage (HV) power supply and DBD experimental facility. The voltage regulator regulates input voltage from 0–250V and output voltage was amplified more than 1000 times by high voltage power supply to provide the DBD reactor to ignite the plasma jet. As the equipment photograph in Figure 1(b) illustrates, the DBD plasma reactor consisted of upper and lower stainless steel electrodes (50-mm diameter) with two pieces of quartz glass (90-mm diameter, 1-mm thickness) placed 6 mm apart to act as a medium. One of the electrodes was connected to a high voltage (HV) power supply while the other was grounded for safety. Use of quartz glass not only avoids arcing edge the electrodes but also cools them down.

In our experiments, operational parameters of the DBD plasma source were carefully chosen to obtain a stable plasma jet. Our sine wave high voltage power supply (CTP-2000K) was used to drive the powered electrode at a frequency of 10 kHz. From a digital oscilloscope (Rigol DS6102, 1 GHz, 2 Channel), the electrical characteristics were read as 235 kV peak to peak and 812 mV peak to peak using output voltage and current testing interfaces, respectively, along with a sampling resistance at output current testing interface of 50 Ω. During experiments, input voltage from the regulator was 18V, with a 1.6A current. The plasma source produced lots of ozone, which was detected for a long time even after power was turned off.

LT-12 cells were cultured at 37°C in a 5% CO₂ incubator with RPMI 1640 (Hyclone, Thermo Fisher Scientific, Waltham, MA) containing 10% inactivated fetal bovine serum, 100 μg/mL penicillin and 100 μg/mL streptomycin (Gibco, Thermo Fisher, Waltham, MA). Cells were passaged once every 2–3 d and logarithmic growth phase cells were used for experiments. For plasma treatments, 35-mm petri dishes with 2mL cell suspensions of 1 × 10⁵–1 × 10⁶ cells/mL were placed between the two plasma reactor electrodes for different doses respectively.

To detect the lethal effect of low-temperature plasma on LT-12 cells, the cell killing rate was analyzed at 2, 4, 6, 8, 12, 16, 20 and 24 h post-plasma treatment, 5 mg/mL MTT reagent (Sigma-Aldrich, St. Louis, MO) was added to cultures. After 4 h incubation, 10% SDS/50% dimethylformamide (volume fraction, Sigma) were added to dissolve MTT overnight. The absorbance (A) at 570 nm was measured using a microplate reader (Thermo) and cell killing rate was calculated as (1 − A_{experimental group}/A_{control group}) × 100%.

As shown in Figure 2(a), at a dose of 240 s, the killing rate of cells increased from 21.6% at 2 h to 60.4% at 24 h post-plasma treatment. When comparing minimal (30 s) and maximal doses (240 s) at 8 h post-plasma, the killing rate of cells increased from 1.5% to 36.7%. These data indicated that the plasma lethal effects on LT-12 cells was related to significantly doses and post-plasma time points. Additionally, the cell killing rate significantly increased after 12 h, indicating plasma-mediated death of tumor cells is neither a fast nor immediate process. Further, normal spleen leukocytes treated maintained 20% cell killing rate at the same doses and at the same time points, as shown in Figure 2(b); thus, plasma does not exhibit a significant lethal effect on normal leukocytes.

During plasma treatment, an infrared thermometer was employed to monitor changes in culture media temperature. As shown in Figure 2(c)–2(h), the temperature of culture media rose from 23.9°C to 31.7°C at the different doses from 0 s to 240s, demonstrating that under our experimental doses conditions, the influence of temperature effects on cell death can be ignored.¹⁹ 

To better elucidate apoptotic mechanisms underlying these changes, cells exposed to plasma for 60, 90 or 120 s were assessed at 8 h and 12 h post-plasma treatment using a variety of the following analytical methods. First, to assess cell apoptosis, flow cytometry was performed using a binding buffer containing Annexin V-Alexa Fluor 488 and PI (Beyotime). After incubating for 15 min at room temperature in the dark, cell apoptosis rate was detected using the C6 instrument (BD Bioscience, San Diego, CA, USA) at an excitation wavelength of 488 nm. The cells were distinguished as living (Annexin V-/PI-), early apoptotic (Annexin V + /PI-), late apoptotic (Annexin V + /Percentage PI-) or necrotic (Annexin V-/PI  +); results are shown in Figure 3(a). Without plasma treatment, the apoptosis rate is 1.9%, 8 h later after plasma doses from 60 s to 120 s, the apoptosis rate increased from 23.8% to 28%. At 12 h post-plasma points, this apoptosis rate increased from 48% to 55.3% at the same doses, respectively. Therefore, the apoptosis rate increases significantly post-plasma treatment, further supporting the cell apoptosis is related to the plasma doses and the post-plasma time points.

In addition, at the same time points with the same doses, the images of cell apoptosis were simultaneously collected using a high-content cellular imaging system (IN Cell Analyzer 2000, GE Healthcare LifeSciences, Piscataway, NJ) to analyze post-plasma cells. Medium was removed from cultured cells before they were suspended in a solution of Annexin V-Alexa 488, PI and Hoechst 33342 (Beyotime) in PBS (Sigma) for 15 min at room temperature. After incubation, cells were washed with PBS and centrifuged at 1500 × g for 30 s for analysis. Excitation wavelengths of 490/20 nm, 543/22 nm and 400/50 nm were used for Annexin V-Alexa 488, PI and Hoechst 33342, respectively. Corresponding emission wavelengths for these biomarkers were 525/30 nm, 605/64 nm and 455/50 nm, respectively; results are shown in Figure 3(b). 8-h later after plasma treatment, few cells undergoing necrosis (PI indicated by red fluorescence) and apoptosis (Annexin V-FITC indicated by green fluorescence) are visible. However, 12-h later, many cells in the late stages of apoptosis or necrosis are visible, indicating that plasma treatment indeed induces apoptosis within leukemia cells.

In order to explore the apoptotic mechanisms underlying this process, we detected ROS concentrations after plasma treatment, We use the 90 s doses as a representative by using ROS detection kit. Briefly, 5 μmol/L of 2’ and 7’-dichloro fluorescein diacetate (DCFH-DA, Beyotime Biotechnology, Shanghai, China) was added to cells at 0, 2, 4, 6, 8, 12, 14, 16, 18, 20, 22 and 24 h post-plasma respectively, after incubating for 20 min at incubator, then an Accuri™ C6 Flow Cytometer (BD Biosciences, San Diego, CA) was employed to detect fluorescence intensity of dichlorofluorescein (DCF), excitation wavelength was 488nm, emission wavelength was 525nm. Intracellular DCF fluorescence intensity was continuously monitored for 24 h after plasma treatment. Finally, using treated group divided untreated group, the rate was relative DCF fluorescence intensity, which can indirectly reflect the intracellular ROS levels. As shown in Figure 4, ROS rises immediately post-plasma by 1.24-fold and continues to increase slowly. At 20 h, ROS increases sharply to 5.39 times pre-treatment levels, finally reaching 13.88-fold original levels at 24 h (p < 0.05).

We detect expression of apoptosis-related genes at the same time points with the same doses, four genes (caspase-3, caspase-8, Bcl-2, and Bid) were evaluated by real-time quantitative PCR (qRT-PCR). The primers (Invitrogen, Carlsbad, CA) were shown in Table I. qRT-PCR reactions were conducted with SYBR GREEN qPCR mix (ComWin Biotech, Beijing, China) using a denaturation temperature of 95°C, annealing temperature of 60°C, and extension temperature of 72°C for 35–40 cycles. Results, as shown in Figure 5, were calculated using the ΔΔCT method.²⁰ At 8 h post-plasma treatment, expression of caspase-3 increased by 2.57-fold (60 s), 2.85-fold (90 s) and 3.70-fold (120 s) compared with untreated groups. Further, expression of caspase-8 and Bid genes also increased in a dose-dependent manner. As an anti-apoptotic gene, Bcl-2’s expression level showed the strongest inhibition at 90 s doses (p < 0.05). At 12 h post-plasma treatment, gene expression patterns were similar; however, Bid expression was no longer active.

After confirming expression of apoptosis-related genes, a western blot assay was performed to detect expression of apoptosis-related proteins in post-treatment cells. Briefly, cells were collected and lysed in RIPA buffer before being centrifuged at 12000 × g for 5 min. The apoptosis antibody detection kit used for these experiments was purchased from Cell Signaling Technology (Danvers, MA). Caspase activation is a key event during apoptosis. Activation of caspase-9 and caspase-3, key regulators for initiating the entire caspase cascade, represents an essential step for the occurrence of cell apoptosis. Caspase-3 and caspase-7 transduce the signal, and PARP performs downstream activities common to all apoptotic-signaling pathways. As shown in Figure 6, at 8 h post-plasma, expression of cleaved caspase-3, -7 and -9, and PARP gradually increased in a dose-dependent manner compared with controls. This was sustained at 12 h post-plasma, and expression of these proteins proved even stronger than at 8h time points.

All statistical analyses were carried out using SPSS 22.0 statistical software and data are expressed as $X ̄ ±S$ (mean ± standard deviation). Analysis of variance was used as the analytical method, with ^∗P < 0.05 considered statistically significant.

It is possible to generate atmospheric low-temperature plasma by applying a high voltage between two electrodes to allow for gas discharge. The electron energy inside plasma is quite high, making it capable of exciting, dissociating and ionizing reactant molecules into activated reactive species. The resulting active ingredients, including ultraviolet (UV) and charged particles (electrons, positive and negative ions), chemically reactive particles, reactive oxygen species (ROS), reactive nitrogen species (RNS), and others, can react within cells to induce apoptosis.

In the process of apoptosis induced by low temperature plasma, doses and post-plasma points play important roles. Barekzi et al. observed that the lethal effects of plasma are not immediate, but occurring with a certain delay, and the rate of cell death increases in a dose-dependent manner. In their investigation, when plasma doses were more than 3 min, it resulted higher death rates than less dose at 12 h post-plasma treatment later. The authors suggested that with increasing plasma doses, the concentration of generated active substances increases which activates intracellular apoptotic signaling pathway and induces programmed cell death.²¹ Related studies have also shown that the lethal effect of plasma on tumor cells is dose-dependent.²² When observed at post-plasma 8 h later with plasma doses over 60 s, the lethal effect on leukemia cells gradually became significant in our MTT experiment, supporting the delayed lethal effect of plasma. Meanwhile, our flow cytometry analysis indicates that at 8 h post-plasma treatment, increasing plasma doses (60, 90 and 120 s) result in a increased proportion of dead cells (including apoptosis and necrosis cells), 26%, 34.7% and 42.8%, respectively. At the same doses, they at 12 h post-plasma group was consistently higher than at 8 h post-plasma group, indicating that the significant lethal effects of plasma on tumor cells were related to doses and time has passed after plasma.

To further explore mechanisms underlying the lethality of plasma on LT-12 cells, we examined ROS levels and expression levels of several common apoptosis-related genes and proteins after plasma exposure. Additionally, simultaneous monitoring of intracellular ROS indicated gradual increases over time. At 24 h post-plasma treatment, ROS levels rose sharply to 13.88 times pre-treatment levels. This may explained the delay observed in the lethal effect of plasma. Upon facing the stresses induced by plasma, cells first show self-healing; however, failure in self-healing leads to an outbreak of ROS in vivo that results in cell death.²³ Caspases and Bcl-2 family members play an important role in plasma-induced apoptosis. Kim et al. demonstrated that plasma is capable of activating the apoptosis executioner caspase-3 to promote release of cytochrome C from mitochondria and change membrane potential.²⁴ Our results demonstrate that with the increase of the plasma doses, expression of apoptosis-related genes, including caspase-3, -8 and Bid, significantly increase, expression of activated caspase-3, -7 and -9, and PARP proteins also increase, and the 120-s group showed the strongest protein activation. Obviously, expression in the 12-h group was significantly stronger than that of the 8-h group. The results of our study show that after plasma treatment in LT-12 cells, caspases family proteins are activated and may result in activation of complex intracellular apoptotic net and, ultimately, promote apoptosis. However, the exact signaling pathway remains unknown. We need more information to evaluate which signaling pathway - mitochondria mediated signaling pathway or death receptor signal transduction pathway or others plays a domain role in the apoptosis induced by plasma.

In summary, our results provide insights into possible mechanisms of DBD plasma-induced apoptosis. Low-temperature plasma exhibits legitimate anticancer effects and almost no injury to normal cells, so its adoption as a new method for treating tumors is possibly expected. Previous plasma inactivation tumor studies have mostly focused on superficial tumors because plasma demonstrates a poor capacity to penetrate tissue. Our study used leukemia cells as a treatment object because the cyclic photodynamic method developed in our group enables us to make direct contact between the plasma and blood cells in treatment;^25,26 this will provide the opportunity to transfer anti-tumor plasma treatments from in vitro to in vivo experiments and represents the future direction for our work.

This work was supported by the Tianjin Science and Technology Committee (14CDZSY00037).