A human non-small-cell lung cancer (NSCLC) cell line consisting of lung adenocarcinoma epithelial A549 cells (ATCC, United States) was cultured in a humidified 5% CO₂ incubator at 37°C. Cells were seeded in complete RPMI-1640 medium (10% (v/v) fetal bovine serum, 2 mM l-glutamine, and 1% (v/v) 1002 U/ml penicillin/streptomycin). All cell experiments were performed under “hypoxic conditions” (1% O₂, 5% CO₂, 85% N₂, and 37°C) and compared with the control “normoxic conditions” (20% O₂, 5% CO₂, 75% N₂, and 37°C) (Lee et al., 2018). Unless otherwise noted, chemicals and supplements were purchased from Sigma/Aldrich, VA, United States. Carboplatin (Novartis, Basel, Switzerland) and Perftoran^® (Scientific Industrial Company Perftoran, Pushchino, Russia) were used in the study.

To assess the cytotoxicity of carboplatin with/without Perftoran^®, the method of 3,4,5-dimethylthiazol-2,5-diphenyl tetrazolium bromide (MTT) was used (Hansen et al., 1989). Cells were incubated with different concentrations of carboplatin for 24 h in the presence or absence of 5% Perftoran^® emulsion. For oxygenation, Perftoran^® emulsion was oxygen-bubbled for 5 min immediately before cell treatment. Cell viability was assayed by liquefying insoluble formazan, using dimethyl sulfoxide, at the end of the MTT assay, and the reading was recorded at 570 nm using a FlOUstar Optima multi-detection system (BMG, Germany). Viability results were expressed as percentage of untreated control cells (mean ± standard deviation). Half-maximal inhibitory concentration (IC₅₀) was calculated by linear approximation regression of the percentage survival versus the drug concentration.

A549 cells (1 × 10⁴ cells per chamber) were seeded onto 8-chamber glass slides for cell culture (SPL Life Sciences, Korea). The cells were treated with carboplatin IC₅₀ for 24 h in the presence and absence of a 5% oxygenated Perftoran^® emulsion. Then the cells were stained with dual DNA staining (acridine orange/ethidium bromide; AO/EB in PBS) at a concentration of 100 μg/ml (V/V) (Baskić et al., 2006). The cells were analyzed and counted using a fluorescence microscope (Axio Imager Z2, Carl Zeiss, Germany) and were subclassified based on fluorescence color via an image analyzer: vital cells (green), early apoptotic cells (yellowish green), and necrotic and late apoptotic cells (orange to red) (n = 8). A percentage of each cell population was calculated (mean ± SE) and tabulated.

DNA damage in A549 cells seeded with IC₅₀ of carboplatin with/without oxygenated Perftoran^® (5%) was examined by the comet assay (Collins, 2004) for 24 h at 37°C under normoxic and hypoxic conditions. The corresponding control cells were incubated using the same conditions. After trypsinization, 0.5% normal agarose/PBS (100 ml) was dropped onto a frosted glass microslide, covered with a coverslip, and cooled at 4°C for 10 min before the coverslip was removed. The cells (1 × 10³) were mixed with low melting agarose (1%) at 37°C, applied to the gel-coated slide, covered with a coverslip, and cooled for 10 min at 4°C. The coverslip was removed, and a similar third coat of low melting agarose (0.5%) was applied. The slides were kept in the dark for 2 h at 4°C in ice-cold lysis solution (2.5 M NaCl, 10 mM Tris, 100 mM EDTA, 90 mM sodium sarcosinate, NaOH, pH 10, 10% DMSO, and 1% Triton X-100). DNA was allowed to unwind in electrophoresis buffer (300 mM NaOH, 1.2 mM EDTA) for 30 min. The slides underwent electrophoresis for 20 min at 300 mA and 25 V and were washed in a neutralization buffer (400 mM Tris–HCl, pH 7.5) and transferred to EB solution (20 g/ml) in the dark. After 20 min, the slides were washed in distilled water and air-dried, and the comets were scored under a fluorescent microscope (400 × magnification). The cells were analyzed according to the mean of their tail-length range—0 (no tail) to 4 (most of the DNA is in the tail, with the smallest head) (Olive, 1999; Collins, 2004)—and DNA damage index (ID) was calculated according to Collins (2004) using the percentage of DNA in the tail obtained using the image analyzer.

To quantify the formation of carboplatin/DNA adduct (Yang et al., 2000), the cells were seeded at 37°C for 1, 6, 12, and 24 h with a fixed concentration of carboplatin (10 µM) with/without oxygenated Perftoran^® (5%). The medium was then discarded, and the cells were collected using 2 min of trypsinization and washed with ice-cold phosphate-buffered saline (PBS). The cells were mixed at 37°C for 5 h with lysis buffer (10 mM Tris–HCl, pH 8.0; 0.5% (w/v) SDS, 0.1 M NaCl, 20 μg/ml RNase, and 0.1 mM EDTA; the mixture was then incubated overnight with 100 μg/ml proteinase H at 50°C. DNA was isolated using the phenol/chloroform protocol and dissolved in Tris/EDTA. DNA concentration was evaluated at 260 nm, and the platinum content was measured via atomic absorption spectroscopy. DNA platination was presented as platinum (pg)/DNA (µg) (n = 8). The analysis of cellular samples and standards was performed using a Zeeman 5000 Atomic Absorption Spectrometer (Perkin-Elmer, United States) fitted with a platinum hollow cathode lamp, HGA500 graphite furnace, and autosampler (Buck Scientific, East Norwalk, Conn., United States). Gradual dilutions of commercial atomic absorption platinum standard (0–100 ng/ml in 5% HCl; Sigma) were used. Readings were recorded at 265.9 nm (slit width 0.7 nm; lamp current 20 mA).

The degree of hypoxia was monitored in A549 cells after treatment with 20% of the corresponding IC₅₀ for carboplatin in the presence and absence of 5% oxygenated Perftoran^® emulsion under the same hypoxic conditions (1% O₂) for 1, 6, 12, and 24 h. The cellular hypoxia degree was traced via a microplate fluorometer using pimonidazole, an anoxic indicator (Challapalli et al., 2017). In another experiment, the cells were digested using Cell Lysis Solution (#LSKCLS500; Merck, United States) supplemented with Protease Inhibitor Cocktail (#P8340; Merck, United States). Lysates were investigated for alterations in HIF-1α and HIF-2α concentrations using the Human HIF-1α ELISA Fluorescent Kit (#ab229433; Abcam, Germany) and Human HIF-2-alpha ELISA Kit (ab227898; Abcam, Germany), respectively.

Total RNA, including miRNA, was extracted from A549 cells using the miRNeasy RNA extraction kit (217004, Qiagen, Germany). Reverse transcription was accomplished using the miScript II RT kit (218161, Qiagen, Germany), and qRT-PCR amplification was assayed using the miScript Sybr green PCR kit (218073, Qiagen, Germany). A battery of miRNAs was inspected using the following Qiagen miRCURY LNA miRNA detection probe kits: hsa-miR-21 (MS00009079), hsa-miR-181a (MIMAT0000270), hsa-miR-210 (MS00003801), and RNU6 (MS00033740). Relative miRNA expression was calculated using the ΔΔCt method (Livak and Schmittgen, 2001), and the values were normalized to the U6 expression in non-treated controls.

A549 cells (5 × 10³ cells per chamber) were seeded onto 8-chamber cell culture glass slides (SPL Life Sciences, Korea). The cells were treated with 20% of IC₅₀ of carboplatin or Perftoran^®/carboplatin for 24 h and then fixed for 20 min with absolute methanol. The cells were incubated in blocking buffer (3% bovine serum albumin/PBS) for 1 h. Slides were immersed overnight in MRP-2 primary antibody (ab240169 Abcam, Germany) at 4°C and then incubated in Goat Anti-Rabbit IgG Alexa-Flour-488 (A11034, Invitrogen). Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI, 1 μg/ml, Sigma). The cells were analyzed under a fluorescence microscope at 200 × magnification (Axio Imager Z2, Carl Zeiss, Germany).

Data were expressed as mean ± SE, and group results were statistically analyzed using a one-way ANOVA. Significance was considered at p < 0.05.

As shown in Figure 1A, treatment of A549 cells with carboplatin alone induced a concentration-dependent decrease in cell viability, as expressed by mitochondrial dehydrogenase activity in the MTT assay. From the concentration/viability curve, the calculated IC₅₀ for carboplatin was 25.21 µM (under normoxic conditions) and 32.83 µM (under hypoxic conditions), while the combination of carboplatin with the oxygenated Perftoran^® emulsion (5%) decreased the resistance of A549 cells to carboplatin by potentiating its cytotoxicity, as concluded from the lower IC₅₀ of 14.78 µM (p < 0.01) with carboplatin/Perftoran^®. For the hypoxia-related cellular and genomic analyses, 20% of each IC₅₀ was used.

Cell death can occur by different modes when the cells were cultured with cytotoxic agents, but the two most common types of cell death are necrosis and apoptosis. Apoptosis was characterized by nuclear condensation, DNA fragmentation, cell contraction, and the membrane blebbing formation of apoptotic bodies (Ercin et al., 2019; Zhang et al., 2018). To elucidate the effect on the cell death type of carboplatin combined with Perftoran^®, an AO/EB dual DNA staining was performed. As shown in Table 1, treatment with carboplatin IC₅₀ for 24 h yielded a high percentage of the apoptotic cell population (35.4%) for both early and late phases (p < 0.0001) and in necrotic cells (p < 0.0001) compared to the A549 cells under hypoxic conditions. On the other hand, IC₅₀ of Perftoran^®/carboplatin led to a high increase in the total percentage of apoptotic cells (54.4%, p < 0.01) compared to carboplatin-treated cells, particularly in the early apoptotic population (p < 0.01). Additionally, the presence of Perftoran^® inhibited necrosis as a cause of death (p < 0.0001) compared to carboplatin-treated cells. These findings suggest that the combination of carboplatin with oxygenated Perftoran^® diminishes carboplatin resistance and shifts its cytotoxicity from necrosis to high apoptosis.

The degree of platinum-enhanced DNA damage was investigated using an alkaline comet assay (Figure 1B; Table 2). Following the exposure of A549 cells to the corresponding IC₅₀, length grades of the comet tails were analyzed; analysis results are listed in Table 2. Comparison of cells under normoxic and hypoxic conditions suggests that hypoxia may prevent the formation of reactive oxygen species (ROS) and protect DNA from damage, as indicated by the lower degree of DNA damage (ID) under hypoxic conditions (p < 0.01). For hypoxic cells, the presence of oxygenated Perftoran^® shifted the ID value toward the normoxic cell range. A comet assay was used to assess the degree of DNA damage in A549 cells treated with carboplatin alone and with both Perftoran^® and carboplatin, as presented in Table 2. The treatment of cells with carboplatin alone showed significant induced damage in DNA (ID 39) compared to hypoxic cells (p < 0.0001), as demonstrated by surprising induced percentages of length grades of the comet tails (p < 0.01–0.0001), whereas 61% of the cells exhibited intact DNA. The combination of carboplatin with oxygenated Perftoran^® yielded a noticeable induction in the DNA damage index (ID 57) compared to the cells treated only with carboplatin (p < 0.01), as demonstrated by the elevated percentages of length grades 1–4 (p < 0.05–0.001), whereas only 43% of the cells maintained intact DNA. These findings suggest that the presence of oxygenated Perftoran^® may enhance the damaging ability of carboplatin by inducing ROS supply in A549 cells.

To measure platinum/DNA adduct formation, cells were incubated for 1, 6, 12, and 24 h with a fixed concentration of carboplatin (10 µM) in the absence and presence of a fixed concentration of oxygenated Perftoran^® (5%). Figure 1C presents the results of DNA platination analysis. Results showed that presence of oxygenated Perftoran^® strongly induced the formation of platinum/DNA adducts, even after only 1 h of incubation, while the highest concentration of platinum/DNA adducts was detected after 6 h of incubation. Findings also revealed that after 24 h, carboplatin-treated cells lost most of the binding affinity between DNA and platinum, while in Perftoran^®/carboplatin-treated cells, the adduct level after 24 h was as high as the adduct level in carboplatin-treated cells after 1 h. These findings suggest that the inhibition of hypoxia induced and maintained the level of DNA platination.

To evaluate the degree of hypoxia in A549 cells and the effect of Perftoran^® on this, a microplate fluorometric determination for the relative fluorescent intensity (FU) of pimonidazole adducts was performed to monitor hypoxia adduct formation after different incubation times (1, 6, 12, and 24 h). As shown in Table 3, the formed fluorescent adducts were dependent on time, gradually increasing in cells under hypoxic conditions up to 24 h (p < 0.001) compared to normoxic cells. Under hypoxic conditions, the presence of oxygenated Perftoran^® decreased adduct formation to a range lower than that of normoxic cells. The treatment of cells with carboplatin alone (20% of IC₅₀) significantly induced hypoxia over time, from 6 to 24 h of incubation, compared with the hypoxic cells (p < 0.05), as shown in Table 3. On the other hand, treatment with Perftoran^®/carboplatin (20% of IC₅₀) suppressed the formation of hypoxia adducts (p < 0.01) compared with carboplatin-treated cells at all incubation intervals tested. These findings confirm the ability of oxygenated Perftoran^® to deplete carboplatin-associated hypoxia in A549 cells, as depicted in Table 3.

Under hypoxic conditions, HIF-1α and HIF-2α both play key roles in lung tumorigenesis, and both participate in suppressing apoptosis and inducing cell growth (Ercin et al., 2019). To ascertain the involved hypoxia mediators in the anti-hypoxic role of Perftoran^®, we investigated the protein concentrations of HIF-1α and HIF-2α, which may help interpret the decrease in hypoxia adducts. The treatment of A549 cells with carboplatin under normoxic condition resulted in an elevation in the concentrations of HIF-1α (p < 0.05) and HIF-2α (p < 0.05). The concentrations of HIF-1α (p < 0.01) and HIF-2α (p < 0.0001) were elevated in cells under hypoxic conditions. This elevation was suppressed by oxygenated Perftoran^® alone (p < 0.05 and p < 0.01 for HIF-1α and HIF-2α, respectively), as presented in Figures 2A,B. The incubation of A549 cells for 24 h with carboplatin alone (20% of IC₅₀) strongly increased the concentration of HIF-1α (p < 0.0001; Figure 1A) and HIF-2α (p < 0.05; Figure 1B) compared with the hypoxic cells, whereas the combined treatment of Perftoran^®/carboplatin (20% of IC₅₀) noticeably diminished both HIF-1α (p < 0.01; Figure 1A) and HIF-2α (p < 0.01; Figure 1B) concentrations compared to their carboplatin-treated counterparts. Thus, it can be suggested that the presence of Perftoran^® successfully diminishes both the degree of hypoxia and levels of hypoxia mediators in lung A549 cells, which may explain the lower resistance to carboplatin.

In the present study, the propagation of cells under hypoxic conditions elevated the expression of all hypoxamiRs—miR-21 (p < 0.05), miR-181a (p < 0.05), and miR-210 (p < 0.01)—while the addition of oxygenated Perftoran^® to hypoxic cells led to significant inhibition in the expression of those hypoxamiRs—miR-21 (p < 0.05), miR-181a (p < 0.05), and miR-210 (p < 0.01)—compared to hypoxic cells (Figure 3). Treating cells under hypoxic conditions with carboplatin dramatically induced the tested hypoxamiRs—miR-21 (p < 0.0001), miR-181a (p < 0.01), and miR-210 (p < 0.0001)—compared to the hypoxic cells (Figure 3). However, the presence of Perftoran^® with carboplatin effectively repressed the expression of miR-21 (p < 0.001), miR-181a (p < 0.05), and miR-210 (p < 0.0001) compared to carboplatin alone, as shown in Figure 3. These results thus suggest that the strong inhibitory effect of Perftoran^® on hypoxia is enabled by the repression of multiple hypoxia-regulating miRNAs.

The effect of Perftoran^® on the protein expression of the drug resistance transporter MRP-2 was explored in carboplatin-treated A549 cells. Cells were immunocytochemically stained with Alexa-flour-488-conjugated-MRP-2 antibody (green), while nuclei were counterstained with DAPI (blue). Microscopic examination indicated that the cells treated with Perftoran^®/carboplatin showed noticeably lower MRP-2 concentrations than those observed in the carboplatin-treated cells, as shown in Figure 4. This finding suggests that drug resistance against carboplatin was repressed as a consequence of inhibited hypoxia due to the presence of Perftoran^®.