The FaDu and SNU-1041 human HSCC cell lines were purchased from the Korean Cell Line Bank (Seoul, Republic of Korea) and cultured in either Minimum Essential Medium (FaDu) or RPMI-1640 medium (SNU-1041) from Welgene, Inc. containing 10% FBS (Welgene, Inc.) and 1% penicillin/streptomycin (P/S). The IHOK cell line was kindly provided by Yonsei University (Seoul, Republic of Korea) and cultured in KBM™ Gold Keratinocyte Growth Basal Medium (Lonza Group, Ltd.) supplemented with KGMTM-2 Keratinocyte Growth Medium-2 SingleQuots™ Supplements and Growth Factors (Lonza Group, Ltd.). The cells were incubated at 37°C in a humidified environment with 5% CO₂. All experiments were conducted when the cells had grown to ~50% confluence. All chemicals were dissolved in DMSO and stored at −20°C. CAPE was purchased from Santa Cruz Biotechnology, Inc. Each cell line was divided into two groups for further analysis of cellular responses to CAPE treatment: i) The vehicle control group and ii) the CAPE (70 µM for FaDu and 100 µM for SNU-1041 and IHOK) treatment group. Vehicle- and CAPE-treated cells were incubated at 37°C for 24 h. For pre-treatment of inhibitors for 1 h (Z-VAD-FMK) or 2 h (CHX and MG132) at 37°C, each cell line was divided into following two groups: i) Vehicle control group and ii) Z-VAD-FMK (10 µM), CHX (100 ng/ml for FaDu and 50 ng/ml for SNU-1041), and MG132 (800 nM) treatment group.

FaDu, SNU-1041 and IHOK cells at 50% confluency were treated with vehicle or CAPE. Trypan blue solution in a concentration of 0.4% (Corning, Inc.) was used to stain the cells at a 1:1 ratio at room temperature (RT). Immediately, after staining, cell viability was evaluated automatically using a CytoSMART cell counter (Corning, Inc.).

Basal Medium Eagle (BME; Sigma-Aldrich,) was dissolved in distilled water (DW) supplemented with sodium bicarbonate, resulting in a 2X BME solution, then filtered using a 0.2-µm filter (Sartorius AG). A mixture containing 1.25% agar, 2X BME, FBS, PBS, l-glutamine and gentamicin was then prepared. A total of 3 ml of the 1.25% agar mixture containing the indicated amounts of CAPE (70 µM for FaDu and 100 µM for SNU-1041) was added to each well of the 6-well plates. The agar mixture was then allowed to solidify for 2 h at RT. Subsequently, a 10% BME solution was prepared by diluting 2X BME with DW and supplementing with l-glutamine, gentamicin and FBS. FaDu and SNU-1041 cells were suspended in the 10% BME solution and mixed with 1.25% agar. After mixing, 1 ml of the mixture containing 0.8×10⁴ cells was added directly onto the 6-well solid-bottom agar plates. The plates were incubated at RT for 2 h for agar solidification and then placed in a humidified incubator at 37°C with 5% CO₂ for a period of ~4 weeks. The 6-well plates were supplemented once a week with 150 µl of complete medium containing the indicated concentrations of CAPE, to prevent agar desiccation. Images of cell colonies were captured using a CKX53 microscope (Olympus Corporation) and colony counts were analyzed automatically using the ImageJ software, version 1.53t (National Institutes of Health).

Cells were suspended in a serum-free medium supplemented with 0.01X N-2 and B-27 supplements, 25 ng/ml of human basic fibroblast growth factor (bFGF; Invitrogen; Thermo Fisher Scientific, Inc.) and human epidermal growth factor (EGF; Thermo Fisher Scientific, Inc.) and 1% P/S. The cells (1×10⁴ cells/well) were cultured in ultralow attachment 6-well plates (Corning, Inc.) for 5 days (FaDu) and 11 days (SNU-1041). Spheres were randomly photographed using an inverted light microscope (Nikon Corporation) and counted automatically using ImageJ software, version 1.53t.

FaDu and SNU-1041 cells, which were treated with either a vehicle or CAPE when they reached ~50% confluence were fixed with 70% ethanol at −20°C overnight and stained with a DAPI solution (2 µg/ml; MilliporeSigma) on a glass slide for 10 min at RT. Images of the nuclear morphological changes occurring in apoptotic cells were captured by fluorescence microscopy (Leica DMi8; Leica Microsystems GmbH).

After treating the cells with vehicle or CAPE when they reached ~50% confluence, FaDu and SNU-1041 cells were fixed with 70% ethanol overnight at −20°C and stained with a propidium iodide (PI) solution (20 µg/ml; MilliporeSigma) containing RNase A (20 µg/ml; Thermo Fisher Scientific Inc.) at 37°C for 15 min. Cell-cycle distribution was then assessed by an LSRFortessa X-20 flow cytometer (BD Biosciences) and analyzed using the FlowJo software, version 10.8.1 (FlowJo LLC).

FaDu and SNU-1041 cells, which were treated with either a vehicle or CAPE when they reached approximately 50% confluence were stained with FITC Annexin V Apoptosis Detection Kit (cat. no. 556547; BD Pharmingen; BD Biosciences), according to manufacturer's instructions. In brief, the cells were stained with Annexin V for 15 min at RT and then stained with PI on ice just before evaluation using a flow cytometer. The apoptotic population was evaluated using an LSRFortessa X-20 flow cytometer (BD Biosciences) and analyzed using the FlowJo software, version 10.8.1.

Total protein lysates were extracted using 1X RIPA buffer (MilliporeSigma) including cOmplete™, Mini protease inhibitor cocktails (Roche Diagnostics) and Pierce™ phosphatase inhibitor tablets (Thermo Scientific; Thermo Fisher Scientific, Inc.). A quantitative analysis of protein concentrations was carried out using a DC Protein Assay Kit (Bio-Rad Laboratories, Inc.). Equal amounts of protein were loaded in each lane of the gel, but this amount was adjusted according to the protein examined. Protein lysates were separated using SDS-PAGE. The separated proteins were then electrotransferred onto a PVDF membrane. The membrane was blocked with 5% skim milk in TBST including 0.05% Tween 20 (Duchefa Biochemie) for 1 h 30 min at RT, to prevent non-specific binding of antibodies to the membrane. Subsequently, the membrane was thoroughly rinsed with TBST, incubated with the designated primary antibodies overnight at 4°C, then exposed to HRP-conjugated secondary antibodies for 4 h at 4°C. The protein bands of interest were visualized using the WestGlow™ PICO PLUS chemiluminescent Substrate (BIOMAX) on an X-ray film or using the Image Quant LAS 500 system (Cytiva). β-actin was used as a loading control. The intensity of protein bands was quantified using ImageJ software, version 1.53t. Information concerning the antibodies used as well as the amount of loaded protein and the percentages of the gels for the experiments conducted are presented in Table I.

The effects of CAPE on the ΔΨm were determined using a MitoScreen JC-1 kit (cat. no. 551302; BD Pharmingen; BD Biosciences), according to the manufacturer's protocols. Briefly, FaDu and SNU-1041 cells were washed twice with PBS and incubated with a 1X JC-1 working solution for 30 min at 37°C. The stained cells were washed and resuspended in 1X assay buffer, then analyzed using an LSRFortessa X-20 flow cytometer (BD Biosciences) and interpreted using the FlowJo software, version 10.8.1.

Vehicle- or CAPE-treated FaDu and SNU-1041 cells were separated into cytosolic and mitochondrial extracts using the Mitochondria/Cytosol Fractionation Kit (cat. no. ab65320; Abcam), according to the manufacturer's protocols. In brief, the cells were resuspended in 1X cytosol extraction buffer mix including protease inhibitors and DTT, vortexed for 2 min and placed on ice for 10 min, to allow complete disruption of cell membranes. Following centrifugation for 15 min at −10,000 × g at 4°C, the supernatants (including the cytosolic fraction of the cells) were carefully transferred to new microtubes. The remaining cell pellets were washed with PBS, resuspended in 1X mitochondrial extraction buffer mix containing protease inhibitors and DTT, vortexed for 10 sec, and centrifuged using the aforementioned conditions. The resulting supernatants (including the mitochondrial fraction of the cells) were collected in new microtubes.

TRIzol^® reagent (Thermo Fisher Scientific, Inc.) was utilized to extract total RNA from FaDu and SNU-1041 cells and 1 µg of the extracted RNA was reverse transcribed into cDNA using the AMPIGENE cDNA Synthesis Kit (Enzo Life Sciences, Inc.): 42°C for 30 min and 85°C for 10 min. The resulting cDNA was used as the input for PCR using the AMPIGENE qPCR Green Mix Hi-Rox reagent (Enzo Life Sciences, Inc.). Real-time PCR was carried out on an Applied Biosystems StepOne Plus Real-Time PCR System (Applied Biosystems) using the following conditions for all genes: 95°C for 2 min, followed by 40 cycles at 95°C for 10 sec and 60°C for 30 sec. The amount of the GAPDH gene was used to normalize the relative amount of each gene, which was measured using the 2^−ΔΔCq method (19). The qPCR primers used in this experiment were as follows: Survivin forward, 5′-ACTTGGCCCAGTGTTTCTT-3′ and reverse, 5′-GACAGAAAGGAAAGCGCAAC-3′; XIAP forward, 5′-TGGTATCCAGGGTGCAAATATC-3′ and reverse, 5′-GTTCTTACCAGACACTCCTCAAG-3′; and GAPDH forward, 5′-GTGGTCTCCTCTGACTTCAAC-3′ and reverse, 5′-CCTGTTGCTGTAGCCAAATTC-3′. The primers were designed based on the NCBI-BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi) reference sequences: Survivin-CR541740.1, XIAP-BC032729.1 and GAPDH-BC083511.1. The detailed primer binding sites of respective proteins are included in the Fig. S1, Fig. S2, Fig. S3.

All graphs were generated using GraphPad Prism, version 8.0 (GraphPad Software; Dotmatics) and all statistical analyses were performed using SPSS, version 26.0 (IBM, Corp). The results of three independent biological experiments are presented as the average (mean) value together with the measure of variation [standard deviation (SD)] for all data. Statistical significance was determined by conducting either an unpaired two-tailed Student's t-test or a one-way ANOVA with Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

To examine the impacts of CAPE on HSCC cell growth, a trypan blue assay was performed on two HSCC cell lines, FaDu and SNU-1041, which had been treated with CAPE for 24 h. As revealed in Fig. 1A, the growth of the two cell lines was significantly reduced after treatment with CAPE, but not with the vehicle control. Notably, it was observed that CAPE worked less effectively in the IHOK cell line, immortalized human oral keratinocytes, compared with HSCC cell lines, suggesting that CAPE could selectively attenuate the growth of cancer cells rather than affect normal cells (Fig. S4). A soft agar assay was then conducted to further evaluate the influence of CAPE on the anchorage-independent growth of the HSCC cell lines. As displayed in Fig. 1B, the clonogenic activity of the two cell lines was attenuated by CAPE treatment, as indicated by the reduced size and number of the cell colonies. Furthermore, to examine the tumor suppressive capability of CAPE in a 3D culture system, a sphere-formation assay was carried out. As demonstrated in Fig. 1C, the sphere-forming ability of both cell lines was strongly inhibited in CAPE-treated cells compared with control cells. These findings suggested that CAPE thwarts cell growth and carcinogenesis in these HSCC cell lines.

To confirm whether the observed cytotoxic effect of CAPE on HSCC cell lines was caused by the activation of apoptotic signaling, three apoptosis detection assays were employed. First, the morphological changes in the nuclei of CAPE-treated cells compared with vehicle-treated cells were evaluated; fragmented or condensed nuclei were observed, which are regarded as typical morphological signs of apoptosis (Fig. 2A). Subsequently, whether cell populations within the sub-G₁ phases expanded after CAPE treatment was investigated. As illustrated in Fig. 2B, a significant increase in the sub-G₁ population in both cell lines was observed after the application of CAPE. Finally, apoptotic cell populations were measured using annexin/PI double staining, to investigate further the effect of CAPE on the induction of apoptosis. It was also revealed that CAPE treatment enhanced the populations within the early and late apoptotic compartments (Fig. 2C). Based on these results, whether the apoptosis stimulated in the cells exposed to CAPE was dependent on the activation of caspase signaling, was examined. As shown in Fig. 3A, the CAPE treatment notably stimulated the formation of cleaved forms of caspase-3 (c-caspase-3) and induced the expression of cleaved poly(ADP-ribose) polymerase (c-PARP). To obtain evidence to corroborate these results, the cells were treated with Z-VAD-FMK, which is a caspase-inhibitor, for 1 h before exposing them to CAPE treatment for 24 h, to investigate whether CAPE can induce caspase-dependent apoptosis in the two cell lines. As expected, the induction of the expression of c-caspase-3 and c-PARP and annexin V-positive compartments by the CAPE treatment was reversed by pre-treatment of the cells with Z-VAD-FMK (Fig. 3B and C). Taken together, these results indicated that CAPE considerably attenuates the growth of HSCC cells, possibly by triggering caspase-dependent apoptotic processes.

Apoptosis can be triggered via two primary pathways, the extrinsic or intrinsic pathway, with formation of cleaved caspase-8 and caspase-9 being considered a typical marker of each pathway, respectively. Therefore, the pathway which was responsible for the apoptotic cell-killing effect in cells that were subjected to CAPE treatment was investigated. After exposure of both cell lines to CAPE for 24 h, an increase in the expression of cleaved caspase-9 (c-caspase-9), but not of cleaved caspase-8 (c-caspase-8), Bid, and t-Bid were observed, indicating that CAPE may promote intrinsic apoptosis signaling (Fig. 4A). In the intrinsic pathway, mitochondria lose their ΔΨm and exhibit increased membrane permeabilization, which causes the release of proteins from the mitochondrial intermembrane spaces, such as cytochrome c (20). To demonstrate further the impact of CAPE on the activation of the intrinsic apoptotic pathway, changes in ΔΨm were measured by employing the JC-1 fluorescent dye. As expected, the number of JC-1 aggregates (red fluorescence) was decreased in cells treated with CAPE, indicating that the treatment caused a reduction in the ΔΨm (Fig. 4B). Next, it was investigated whether the changes in the ΔΨm caused by CAPE led to the outer mitochondrial membrane permeabilization, causing the release of cytochrome c. Cytochrome c oxidase subunit 4I1 (Cox IV), a cytochrome c oxidase enzyme located within the inner membrane of the mitochondria (21), and α-tubulin, a cytoplasmic marker protein (22), were used as mitochondrial and cytosol loading controls, respectively. As demonstrated in Fig. 4C, it was found that cytochrome c was transferred from the mitochondria to the cytosol, as indicated by the decrease in cytochrome c detected in the mitochondria and the increase in cytochrome c observed in the cytosol in response to CAPE. Accordingly, the apoptosis observed in the two CAPE-treated cell lines was attributed to an intrinsic mitochondrial signaling process.

As aforementioned, survivin and XIAP are the major members of the IAP family, and both hinder apoptosis, leading to the increased survival of cancer cells (23). Therefore, the expression of survivin and XIAP were investigated to determine the exact mechanism underlying CAPE-mediated apoptosis. As expected, CAPE administration resulted in the downregulation of both proteins (Fig. 5A). Next, to investigate whether the changes observed for the two proteins were caused by transcriptional regulation, the corresponding mRNA expression levels were measured. As revealed in Fig. 5B, CAPE downregulated the survivin mRNA, but not the XIAP mRNA, suggesting that survivin alone was regulated at the transcriptional level. Therefore, it was hypothesized that XIAP is regulated at the post-translational level, and it was investigated whether CAPE compromises the stability of this protein. In advance of implementing cycloheximide (CHX) and MG132 experiments, a time-course experiment was conducted to verify the exact time point of XIAP degradation by CAPE and to minimize its potential effect on cell viability. In Fig. S5, XIAP was downregulated after 6 h of CAPE treatment in both cell lines. The time was determined when a 1-h pretreatment of inhibitors was followed by subsequent CAPE treatment for 6 h. As depicted in Fig. 5C, the level of the XIAP protein was decreased in the two cell lines after co-treatment with CHX (a protein synthesis inhibitor) and CAPE, which may indicate that XIAP was controlled by post-translational regulation, particularly via protein degradation. Thus, the cells were subsequently treated with MG132 (a proteasome inhibitor) for 2 h prior to CAPE treatment, to explore whether XIAP protein degradation was accelerated by CAPE in a proteasome-dependent manner. In Fig. 5D, MG132 treatment restored the downregulated levels of expression of the XIAP protein in the CAPE-treated groups in both cell lines. Consequently, these data indicated that CAPE exerts its apoptosis-inducing effects on HSCC cells by downregulating two anti-apoptotic proteins, survivin and XIAP, via either transcriptional or post-translational regulation.