The chronic myelogenous leukemia cell line, K562 (cat. no. RCB-0027), was obtained from the RIKEN BioResource Center and maintained in RPMI-1640 medium (FUJIFILM Wako Pure Chemical, Osaka, Japan) supplemented with 10% fatal bovine serum (FBS; Sigma-Aldrich; Merck KGaA). The cervical cancer cell line, HeLa (cat. no. RCB0007), was also obtained from the RIKEN BioResource Center. The liver cancer cell line, HepG2, (cat. no. HB-8065) and the colon cancer cell line, SW480, (cat. no. CCL-228) were obtained from the American Type Culture Collection (ATCC). These cells were maintained in DMEM supplemented with 10% FBS and 2 mM-glutamine (Gibco; Thermo Fisher Scientific, Inc.). These cells were maintained at 37˚C in humidified air with 5% CO₂.

Transcriptional regulation regions, promoters on the 5'-flanking region of the human PD-1 or PD-L1 gene were obtained from human genomic DNA (Promega Corporation) by PCR using PrimeStar GXL DNA polymerase (Takara Bio, Inc.) and inserted into a luciferase reporter plasmid pGL3 (Promega Corporation) at the KpnI sites. The plasmids, hPD-1-luc and hPD-L1-1uc, were generated. DNA sequences were analyzed using a Genetic analyzer 3500 (Thermo Fisher Scientific, Inc.).

Coffee extracts were generated by the dissolution of instant coffee powder Nescafe Excella (Nestle) with sterilized water. Caffeine, chlorogenic acid and caffeic acid were purchased from Nacalai Tesque, Inc. Quinic acid was purchased from Alfa Aesar; Thermo Fisher Scientific, Inc. These reagents were dissolved in dimethyl sulfoxide (DMSO) (FUJIFILM Wako Pure Chemical). Caffeine was used at a final concentration 1, 2 or 4 mM and the other reagents were used at final a concentration 10 µM as reoffered to in previous studies (34,35).

The plasmids were transfected into K562 cells by electroporation with Nucleofector2b (Program T-016) (Lonza Group, Ltd.). Cells and plasmids were applied into 100 µl buffer of the AMAXA Cell Line Nucleofector Kit V (Lonza Group, Ltd.) in a cuvette equipped with aluminium electrodes of the kit and immediately pulsed at room temperature. Following electroporation, the cells were recovered with RPMI-1640 with 10% FBS. Subsequently, 24 h following incubation at 37˚C, the cells were treated with the reagents mentioned above for 24 h, followed by analyses. In the HeLa, HepG2 and SW480 cells, the plasmids were transfected by lipofection with Lipofectamine 2000^® (Thermo Fisher Scientific, Inc.). Plasmid hPD-1-luc or hPD-L1-luc (1.25 µg) which were generated by the insertion of the promoter region into pGL3 (Promega Corporation) and phRL-TK (Promega Corporation) (0.0625 µg) in 25 µl of Opti-MEM I (Life Technologies; Thermo Fisher Scientific, Inc.) and 0.5 µl Lipofectamine 2000^® in 25 µl Opti-MEM I were incubated for 5 min at room temperature. These solutions were then mixed, incubated for 20 min at room temperature and added to the cells supplemented in 250 µl RPMI-1640 with 10% FBS. At 24 h following transfection, the cells were treated with the reagents mentioned above for 24 h and collected for use in the experiments.

The cells were transfected by electroporation [Nucleofector2b (Lonza Group, Ltd.)] for the K562 cells and by lipofection [Lipofectamine 2000^® (Thermo Fisher Scientific, Inc.)] for the HeLa, HepG2 and SW480 cells. The cells transfected with the plasmids, hPD-1-luc or hPD-L1-luc, which were generated by the insertion of the promoter region derived from the human PD-1 or human PD-L1 gene into pGL3 (Promega Corporation) and phRL-TK (Promega Corporation) were incubated at 37˚C for 24 h. The cells were then treated with the reagents mentioned above for 24 h, harvested and lysed with a 1X concentration of Passive lysis buffer (Promega Corporation) and cell lysate were used for luciferase assay with Dual Luciferase Reporter Assay System (Promega Corporation) and a luminometer (Lumat LB9507, Berthold Detection Systems). The Renilla reniformis luciferase plasmid phRL-TK was also transfected with the reporter plasmids to standardize the transfection rate.

Total RNA was prepared from the cultured cells using Isogen II (Nippon Gene Co., Ltd.). Complementary DNA (cDNA) was generated from total RNA using oligo (dT) (Thermo Fisher Scientific, Inc.), dNTP mix (Thermo Fisher Scientific, Inc.) and superscript IV (Thermo Fisher Scientific, Inc.). RT-qPCR was performed with specific primers for human PD-1 and human PD-L1 synthesized by Eurofins Genomics, THUNDERBIRD™ SYBR qPCR mix (Toyobo Life Science), generated cDNA and the StepOnePlus real-time PCR system (Thermo Fisher Scientific, Inc.). The thermocycling conditions were as follows: 95˚C for 10 sec, followed by 40 cycles of 95˚C for 15 sec and 60˚C for 1 min, then 95˚C for 15 sec, 60˚C for 1 min and 95˚C for 15 sec. The quantification method used was the ΔΔCq method (36). The primer sequences were the following: Human PD-1 forward, 5'-GACAGCGGCACCTACCTCTGTG and reverse, 5'-GACCCAGACTAGCAGCACCAGG; human PD-L1 forward, 5'-GGAGATTAGATCCTGAGGAAAACCA and reverse, 5'-AACGGAAGATGAATGTCAGTGCTA; and human actin forward, 5'-GCTGTGCTACGTCGCCCTG and reverse, 5'-GGAGGAGCTGGAAGCAGCC.

The number of viable cells was evaluated using a Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Inc.). The cells were incubated with the CCK-8 reagent at 37˚C, for 4 h. The absorbance (450 nm) was determined using SpectraMax iD3 (Molecular Devices, LLC). The percentage growth was determined relative to the control (sterilized water for the coffee extracts and DMSO for caffeine).

The cultured cells were harvested and suspended in propidium iodide (PI) (Nacalai Tesque, Inc.) solution (PBS containing 10 µg/ml PI, 0.1% Triton X-100) for 5 min at room temperature. Cell cycle and cell death were analyzed by flow cytometry using a FACS Canto II flow cytometer (BD Biosciences) and BD FACS Diva software v6.1.3 (BD Biosciences). The sub-G1 population was estimated as dead cells.

For the analysis of cell apoptosis, the cells were counterstained using the Annexin V-FITC Apoptosis Detection Kit (Nacalai Tesque, Inc.) for 15 min at room temperature, and were analyzed using a BD Accuri C6 Flow Cytometer (BD Biosciences) and BD Accuri C6 Plus Software (BD Biosciences).

The cells were collected following treatment with caffeine at 4 mM and incubated with anti-PD-1 antibody labeled FITC (cat. no. 11-9969-42, Invitrogen; Thermo Fisher Scientific, Inc.) or anti-PD-L1 antibody labeled PE (cat. no. 12-5983-42, Invitrogen; Thermo Fisher Scientific, Inc.) for 1 h on ice. The analysis was performed using flow cytometry with a FACS Canto II flow cytometer (BD Biosciences) and BD FACS Diva software v6.1.3 (BD Biosciences).

K562 cells were attached to a glass slide using a cytospin (cytospin 3, Shandon; Thermo Fisher Scientific). The glass slide was first stained with May-Grunwald's eosine-methylene blue solution (Merck KGaA) for 3 min at room temperature and then stained with Giemsa's azur eosine methylene blue solution (Merck KGaA) for 30 min at room temperature. The slide glass was washed with water. Cell morphology was observed under a microscope (Eclipse E800 optical microscope, Nikon Corporation).

All cells were treated with 4 mM caffeine for 24 h. The HepG2 and SW480 cells were inoculated in a 12-well plate with cover glasses at the bottom of the plate prior to caffeine treatment. The K562 cells were attached to glass slides using a cytospin after caffeine treatment. The cells were fixed with 100% methanol for 2 min at room temperature, washed with PBS and blocked with PBS with 2% of FBS for 30 min at room temperature. Anit-PD-1 rabbit monoclonal antibody (cat. no. 86163, Cell Signaling Technology, Inc.) or anit-PD-L1 rabbit monoclonal antibody (cat. no. 13684, Cell Signaling Technology, Inc.) was reacted with cells at a 1/200 dilution for 1 h at room temperature. The cells were then reacted with AlexaFluor 488 goat anti-rabbit IgG (cat. no. A11008, Life Technologies; Thermo Fisher Scientific, Inc.) at a 1/500 dilution for 1 h at room temperature. The nuclei were then stained using Prolong Gold antifade reagent with DAPI (P36931, Invitrogen; Thermo Fisher Scientific, Inc.). Immunostaining was observed with a BZ-X710 microscope (Keyence Corporation) at a magnification x400 (scale bar, 20 µm). The mean of fluorescence intensity per cell was measured using ImageJ software (https://imagej.net/ij/).

Data are presented as the mean of three results ± standard deviation (SD). To confirm the reproducibility, we performed the experiments more than three times. Data were statistically analyzed using a two-tailed Student's t-test with Microsoft Excel (Microsoft Corporation) for comparisons among two groups, or one-way ANOVA followed by Tukey's post hoc test with GraphPad Prism Version 8.4.3 (Dotmatics) for comparisons among multiple groups. A value of P<0.05 was considered to indicate a statistically significant difference.

To examine whether coffee affects the expression of the immune checkpoint factors, PD-1 and PD-L1, first, the cloning of the PD-1 and PD-L1 promoter regions was performed. A part of human genomic DNA upstream transcription-start site of human PD-1 or human PD-L1 was amplified by PCR and ligated to the luciferase assay vector pGL3, and hPD-1-luc and hPD-L1-luc were generated (Fig. 1A). To confirm that the cloned regions functioned as a promoter, hPD-1-luc and hPD-L1-luc were transfected into HeLa cells and luciferase assay was performed. The values of luciferase assay for hPD-1-luc and hPD-L1-luc were compared to those of the empty vector, pGL3 (Fig. 1B). The luciferase activities of hPD-1-luc and hPD-L1-luc were markedly increased. The results indicated that the cloned regions functions as promoters of the PD-1 and PD-L1 genes. Subsequently, the transcriptional regulations of PD-1 and PD-L1 by coffee extracts were examined. The coffee extracts were generated by instant coffee powder dissolved in sterilized water. The concentrations of the coffee extracts used in the present study were as referred to in a previous study (6). The HeLa cells transfected with hPD-1-luc or hPD-L1-luc were cultured in medium containing coffee extracts and luciferase assay was performed with the cell lysate. No changes in transcriptional activity were not observed in the cells transfected with hPD-1-luc; however, the coffee extracts increased the promoter activity of hPD-L1-luc-transfected cells compared to the control (Fig. 2A). The same experiment was then performed using the K562 cells. The transcriptional activities of PD-1 and PD-L1 were markedly increased by the coffee extracts in a concentration-dependent manner (Fig. 2B).

Subsequently, the components contained in coffee were examined. Caffeine, chlorogenic acid, caffeic acid and quinic acid were used. The concentrations used in the present study were as referred to in previous studies (34,35). Luciferase assay was performed with these components in HeLa cells; however, no change was observed (Fig. 3A). When the luciferase assay was performed in K562 cells, caffeine but not the other components increased the PD-1 and PD-L1 transcriptional activities (Fig. 3B). The enhancements of the PD-1 and PD-L1 transcriptional activities by caffeine occurred in a concentration-dependent manner (Fig. 4A). It was then examined whether endogenous mRNA levels were affected by caffeine using RT-qPCR. Treatment with caffeine increased the mRNA expression levels of PD-1 and PD-L1 (Fig. 4B). Moreover, the cell surface expression of PD-1 and PD-L1 was examined at the protein level, following treatment with caffeine. As shown in Fig. 4C, caffeine increased PD-1 and PD-L1 protein expression on the cell surface.

The present study then used flow cytometry to reveal whether coffee extracts and caffeine affect the cell cycle and death of K562 cells. When the cells were treated with PI solution, Triton X-100 permeabilized the nucleus and nucleic DNA was stained by PI. Thus, DNA contents were measured. When the cells died, DNA became fragments and the sub-G1 population, which had a lower DNA content than the G1 population, then appeared. Sub-G1 cells were estimated as dead cells. The sub-G1 population of K562 cells treated with the coffee extracts increased in a concentration-dependent manner (Fig. 5A). Treatment with caffeine also increased the sub-G1 cell population, although the effect was less prominent than that of the coffee extracts (Fig. 5B). In addition, the G2/M cell population was slightly increased following treatment with caffeine. These data indicate that coffee extracts and caffeine induce the death of malignant tumor cells.

Subsequently, cell proliferation was examined using CCK-8 assay. The coffee extracts and caffeine inhibited cell proliferation in a concentration-dependent manner (Fig. 5C). The induction of apoptosis was analyzed using Annexin V/PI staining (Figs. 5D and S1). Coffee extracts and caffeine increased the number of Annexin V-positive cells, indicating that apoptosis was induced by treatment with coffee extracts and caffeine. The cell morphologies of the cells treated with the coffee extracts or caffeine were observed (Fig. S2). Neither coffee extracts nor caffeine induced cell differentiation, such as megakaryocytes, a change which was previously reported in 12-O-tetradecanoyl phorbol 13-acetate-treated K562 cells (37). In the cells treated with the coffee extracts, a number of cells exhibited shrunken shapes and nuclear fragmentation, indicating the induction of apoptosis. The present study examined whether the overexpression of PD-1 and PD-L1 induced cell death; however, a high level of PD-1 and PD-L1 expression did not induce cell death (Fig. S3).

The present study demonstrated that PD-1 and PD-L1 expression were increased at a transcriptional level in chronic myeloid leukemia K562 cells. Subsequently, the present study examined whether this activation also occurs in cell lines derived from other tissues. Liver cancer HepG2 cells and colon cancer SW480 cells were used. As shown in Fig. 6A, coffee extracts and caffeine activated both the PD-1 and PD-L1 promoters in a concentration-dependent manner in HepG2 cells. Moreover, treatment with caffeine increased the PD-1 and PD-L1 mRNA levels in HepG2 cells (Fig. 6B). In addition, the promoter activities of PD-1 and PD-L1 increased by the coffee extracts and caffeine in colon cancer SW480 cells (Fig. 7A). PD-1 and PD-L1 mRNA expression also increased following treatment with caffeine in SW480 cells (Fig. 7B). In addition, immunostaining revealed the increase in both PD-1 and PD-L1 protein levels in HepG2, SW480 and K562 (Figs. S4 and S5). These results indicate that coffee extracts and caffeine regulate PD-1 and PD-L1 transcription not only K562 cells, but also in HepG2 and SW480 cells.