Differential effects of sevoflurane on the growth and apoptosis of human cancer cell lines

The human breast cancer cell line MDA-MB-231; colon cancer cell lines HCT116, HT-29, and RKO; lung cancer cell line NCI-H1299; and a spontaneously immortalized human breast epithelial cell line, MCF10A, were obtained from the American Tissue Culture Collection (ATCC; Manassas, VA, USA). The human breast cancer cell line MCF7, colon cancer cell line DLD-1, and lung cancer cell line A549 were obtained from the National Institutes of Biomedical Innovation (NIBIO; Osaka, Japan). MCF10A cells were cultured in Dulbecco's modified Eagle’s medium/nutrient mixture F-12 (D-MEM/F-12; NacalaiTesque, Kyoto, Japan) supplemented with 5% horse serum (Biowest, Nuaille, France), 20 ng/mL epidermal growth factor, 10 μg/mL insulin, 0.5 μg/mL hydrocortisone, and 0.1 μg/mL cholera toxin (Sigma-Aldrich, St. Louis, MO, USA). Cell lines other than MCF10A were cultured in D-MEM (NacalaiTesque) supplemented with 5% fetal bovine serum (Atlas Biologicals, Ft. Collins, CO, USA) and 1% penicillin–streptomycin (NacalaiTesque). Cell lines were maintained at 37 °C with 5% CO₂ in a humidified incubator (Thermo Fisher Scientific, Waltham, MA, USA), except when the cells were exposed to sevoflurane. Cells were reseeded every 3–4 days prior to reaching confluence.

Cell lines were plated in tissue culture plates (TPP, Trasadingen, Switzerland) with their respective media and placed in a CO₂ incubator (Thermo Fisher Scientific). At 24 h post seeding, adherence of the cells to the wells and the confluency between 40 and 80% were confirmed by phase-contrast microscopy. Cells were then subjected to 2–8 h exposure to 1–3% (v/v) sevoflurane (Maruishi Pharmaceutical, Osaka, Japan) with 5% CO₂ in atmospheric air at a flow rate of 0.5 L/min, at 37 °C in a humidified CO₂ incubator (ASTEC, Fukuoka, Japan). The specific time courses of sevoflurane exposure in respective experiments are schematically shown in the corresponding figures. The concentrations of sevoflurane and CO₂ in the incubator were monitored using a Vamos Plus Anaesthetic Gas analyzer (Dräger, Lubeck, Germany). After exposure to sevoflurane, cells were transferred to a CO₂ incubator (Thermo Fisher Scientific) pre-set to a standard culture condition, and further incubated until they were examined in assays. Untreated controls were prepared by seeding the cells to tissue culture plates together with the other samples, left unexposed to sevoflurane, incubated, and examined in parallel with the other samples.

Cell lines were seeded in 6-well tissue culture plates at a density of 4 × 10⁴ cells/cm² in 1.0 mL media, exposed to 1–3% (v/v) sevoflurane for predetermined periods of time, and then incubated for 48 h. In the experiments in which 1% sevoflurane was applied, cell lines were dissociated with trypsin (Nacalai Tesque) and enumerated using a Coulter counter (Beckman Coulter, Brea, CA, USA). When cell lines were treated with 1.5–3.0% sevoflurane, confluency of the cells was assessed using the Incucyte Zoom live-cell microscopy incubator system (Essen BioScience, Ann Arbor, MI, USA) as a surrogate for cell number. During assays, the exponential growth of untreated control cells, as well as the proliferation of cells without reaching confluence in any condition, was confirmed by phase-contrast microscopy.

Cell invasion to Matrigel was assessed using Transwell cell culture chambers with 8.0 μm pores (Corning, Oneonta, NY, USA). The Transwell inserts were coated with 200 μL/well of 300 μg/mL Matrigel (Corning) and air dried for 2 h at 37 °C in a regular CO₂ incubator. Filters in the Transwell inserts were reconstituted with serum-free D-MEM immediately before use. Cells were incubated with or without 1% sevoflurane in a CO₂ incubator for 4 h, dissociated with trypsin, and seeded in triplicate into the Transwell inserts (5 × 10⁴ cells/well). The lower compartment was filled with D-MEM supplemented with 5% FBS. After 24 h of incubation at 37 °C with 5% CO₂, the filters were fixed using 0.5% paraformaldehyde and stained with 0.1% crystal violet in PBS. The migration through Matrigel was scored under microscopy at a magnification of 40× (AXIO, Carl Zeiss Microscopy, Oberkochen, Germany).

Cells seeded in 6-well tissue culture plates at a density of 8 × 10⁴ cells/well were exposed to 1–3% sevoflurane and dissociated and collected after 24 h of incubation. Collected cells were washed with PBS once, stained with fluorescein isothiocyanate (FITC)-conjugated Annexin V (100-fold dilution; Medical & Biological Laboratories, Aichi, Japan) for 20 min, again washed once with PBS, and then stained with 7-aminoactinomycin D (7-AAD; 500-fold dilution; AAT Bioquest, Sunnyvale, CA, USA) for 15 min. The ratios of FITC-positive and 7-AAD-positive cells were determined using flow cytometry in a Gallios Flow Cytometer (Beckman Coulter). Data were analyzed using Kaluza software (Beckman Coulter).

Cells seeded in 6-well tissue culture plates at 6 × 10⁴ cells/well were exposed to 1% sevoflurane in air for the predetermined duration, and incubated for 24 h with or without the pan-caspase inhibitor Z-VAD-FMK (20 µM; AdooQBioScience, Irvine, CA, USA). Whole-cell extracts were lysed by sonication in RIPA buffer (NacalaiTesque), and protein concentration in each extract was determined using a BCA protein assay (Thermo Scientific, Rockford, IL, USA). Equal amounts of protein (50 µg) were resolved via SDS-PAGE using Real Gel plate (Bio Craft, Tokyo, Japan), electro-transferred to an Immobilon-P polyvinylidene fluoride membrane (Merck KGaA, Darmstadt, Germany), and probed with primary and horseradish-peroxidase-conjugated secondary antibodies. The primary antibodies used in this study are anti-caspase-3 rabbit polyclonal antibody (1:1000 dilution; Cell Signaling Technology, Danvers, MA, USA) and anti-β-actin mouse monoclonal antibody (1:2000 dilution; Medical & Biological Laboratories). Chemiluminescent signals were captured and imaged using Amersham Imager 600 (GE Healthcare Life Sciences, Pittsburgh, PA, USA).

Data are reported as mean ± SD values; n represents the number of times an independent assay was performed. One-way-ANOVA and Scheffe’s post-hoc test were performed for cell proliferation, Annexin-V, and 7-AAD assays, whereas two-sided Student’s t-test was applied for cell invasion assays. All statistical analyses were performed using IBM SPSS statistics 24 (SPSS Inc., Chicago, IL, USA). A P value less than 0.05 indicated statistical significance.

To comprehensively understand the impact of sevoflurane exposure on cancer cell growth, we performed cell proliferation assays using eight cancer cell lines. The cell lines examined included two non-small cell lung cancer cell lines (NCI-H1299 and A549), two breast cancer cell lines (MDA-MB-231 and MCF-7), and four colon cancer cell lines (HCT116, DLD-1, HT29, and RKO). A spontaneously immortalized human breast epithelial cell line MCF10A was also included in this assay as a non-cancerous cell control. These cell lines were exposed to 1% sevoflurane in a CO₂ incubator for 2–8 h. After exposure, the cells were allowed to grow for 48 h, and the numbers of cells was determined (Fig. 1a). The results revealed that 4–8 h exposure to sevoflurane augmented proliferation in NCI-H1299 by 1.3-fold (P = 0.01), MDA-MB-231 by 1.4-fold (P = 0.01), DLD-1 by 1.3-fold (P = 0.03), HCT116 by 1.6-fold (P = 0.001), RKO by 1.3-fold (P = 0.0001), and HT29 by 1.2-fold (P = 0.0001), compared with their respective untreated controls. In NCI-H1299 and MDA-MB-231, 8 h exposure to sevoflurane resulted in a suppression of cell growth compared with 4 h exposure, suggesting that a long-term exposure to sevoflurane may cause modest cellular toxicity and cancel the growth promotion conferred by sevoflurane itself in these cell lines. In the cell lines A549, MCF-7, and MCF10A, sevoflurane reduced cell proliferation by 24% (P = 0.0001), 13% (P = 0.002), and 21% (P = 0.04), respectively, largely in an exposure time-dependent fashion (Fig. 1b).

Collectively, these results demonstrate that sevoflurane differentially perturbs cell proliferation among human cancer cell lines regardless of their organ origins, and it attenuates the proliferation of non-cancerous cells as described previously [16].

Invasion to extracellular matrix is a notable property of cancer cells, which in fact was found to be suppressed after sevoflurane exposure in a few types of cancer cells [13, 14]. To address how sevoflurane exposure in our experimental condition affects cancer cell invasion, we performed Boyden chamber-based Matrigel invasion assays using four cancer cell lines originating from lung (NCI-H1299 and A549) and breast (MDA-MB-231 and MCF-7). These cell lines were employed because sevoflurane triggered contradictory changes in proliferation capacity in these pairs of cell lines derived from the same organs.

As shown in Fig. 2, sevoflurane significantly suppressed Matrigel invasion by A549 (23% reduction, P = 0.02) and MCF7 (14% reduction, P = 0.007), two cell lines exhibiting attenuated growth capacities after sevoflurane exposure. In contrast, invasion into Matrigel was upregulated (MDA-MB-231; 2.0-fold, P = 0.007) or not altered significantly (NCI-H1299) upon sevoflurane exposure in the cell lines whose proliferation was enhanced after sevoflurane exposure. These data suggest that alteration in Matrigel invasion occurs in parallel with that in proliferation capacity in cancer cell lines. Of note, changes in the number of Matrigel-invading cells after sevoflurane exposure may result from altered cellular capacity to invade Matrigel, altered proliferation rate, or both.

We subsequently addressed apoptosis, one of the biological processes that may be associated with attenuated cell proliferation, in cells exposed to sevoflurane. We used the same four cell lines derived from lung (NCI-H1299 and A549) and breast (MDA-MB-231 and MCF-7) cancers in this assay. Incidence of early apoptosis was evaluated by staining cells with fluorescence-labeled Annexin V, a cellular protein that specifically binds to phosphatidylserine exposed to the outer surface of the plasma membrane in an early phase of apoptosis.

The cells were exposed to sevoflurane for 2–8 h and, after a 24-h incubation period, incubated with FITC-conjugated Annexin V (Fig. 3a). Fluorescence flow cytometric analyses demonstrated 4.1-fold (P = 0.0001) and 1.5-fold (P = 0.01) increases in the ratio of apoptotic cells relative to the untreated controls in A549 and MCF-7, respectively. Conversely, the apoptotic ratio of NCI-H1299 was reduced by 40% (P = 0.0001), and no significant change in apoptosis was observed in MDA-MB-231, after sevoflurane exposure (Fig. 3b, c).

Induction of early apoptosis in the four cell lines was also assessed by a distinct assay quantifying cleaved (i.e., activated) caspase-3, a protein playing a critical role in the signaling pathway initiating apoptotic processes. Cells were exposed to 1% sevoflurane for 4–8 h and processed for immunoblotting against caspase-3 after a 24 h incubation period (Fig. 4a). This assay demonstrated that cleaved caspase-3 accumulated in A549 and MCF-7 cells but decreased in NCI-H1299 and MDA-MB-231 cells after sevoflurane exposure (Fig. 4b). These data collectively suggest that sevoflurane elicits early apoptosis only in cell lines wherein proliferation is hampered by sevoflurane exposure.

We subsequently investigated the effects of sevoflurane on cell death using 7-AAD. 7-AAD is a fluorescent dye that exclusively marks nonviable cells by moving into the nuclei and intercalating into genomic DNA only when the cell membrane is disintegrated during the late phase of apoptosis. The four cell lines (NCI-H1299, MDA-MB-231, A549, and MCF-7) were first exposed to 1% sevoflurane for 4–8 h, maintained for 24 h after cessation of exposure, stained with 7-AAD, and then analyzed using flow cytometry (Fig. 5a). As shown in Fig. 5b, c, the proportion of 7-AAD-stained cells among A549 and MCF-7 cells significantly increased by 1.7-fold (P = 0.0001) and 1.2-fold (P = 0.006), respectively, in an exposure-time-dependent manner. In contrast, NCI-H1299 and MDA-MB-231 showed no statistically significant difference in the proportion of 7-AAD-stained cells relative to the non-exposed controls. Altogether, these results suggest that the suppression, but not the acceleration, of cell proliferation after sevoflurane exposure is accompanied by an increase in early apoptosis and cell death.

Lastly, we exposed the same set of cell lines (NCI-H1299, MDA-MB-231, A549, and MCF-7) to higher concentrations of sevoflurane and evaluated its effects on cell proliferation, early apoptosis, and cell death. In the assessment of cell proliferation, the cell lines were treated with 1.5%, 2.0%, and 3.0% sevoflurane for 4 h, incubated for 48 h, and analyzed by whole well imaging using the Incucyte Zoom system to determine cell confluency (Fig. 6a). We found that exposure to sevoflurane at any tested concentration augmented cell growth in NCI-H1299 and MDA-MB-231 cells, while attenuating cell growth in A549 and MCF-7 cells (Fig. 6b), when compared to unexposed controls in the same cell lines. Increasing the sevoflurane concentration from 1% (v/v) to a higher concentration resulted in slight reduction of the cell numbers/confluency in NCI-H1299, MDA-MB-231, and A549 cells, suggesting potential cellular toxicity caused by high-dose sevoflurane (Figs. 1b and 6b).

To assess early apoptosis and cell death after sevoflurane exposure, cell lines were exposed to 2% and 3% sevoflurane for 4 h, maintained for 24 h, fluoro-stained for Annexin V and 7-AAD, and analyzed by flow cytometry (Fig. 7a). The results demonstrated that both early apoptosis and cell death were reduced in NCI-H1299 and MDA-MB-231 cells, but increased in A549 and MCF-7 cells when cells were exposed to sevoflurane (Fig. 7b, c). Collectively, these data indicate that high-dose sevoflurane differentially elicited cell proliferation, early apoptosis, and cell death in individual cell lines, similar to the results obtained with 1% sevoflurane.