Abstract
Thyroid hormone, l-thyroxine (T4), has been shown to promote ovarian cancer cell proliferation via a receptor on plasma membrane integrin αvβ3 and to induce the activation of ERK1/2 and expression of programmed death-ligand 1 (PD-L1) in cancer cells. In contrast, resveratrol binds to integrin αvβ3 at a discrete site and induces p53-dependent antiproliferation in malignant neoplastic cells. The mechanism of resveratrol action requires nuclear accumulation of inducible cyclooxygenase (COX)-2 and its complexation with phosphorylated ERK1/2. In this study, we examined the mechanism by which T4 impairs resveratrol-induced antiproliferation in human ovarian cancer cells and found that T4 inhibited resveratrol-induced nuclear accumulation of COX-2. Furthermore, T4 increased expression and cytoplasmic accumulation of PD-L1, which in turn acted to retain inducible COX-2 in the cytoplasm. Knockdown of PD-L1 by small hairpin RNA (shRNA) relieved the inhibitory effect of T4 on resveratrol-induced nuclear accumulation of COX-2- and COX-2/p53-dependent gene expression. Thus, T4 inhibits COX-2-dependent apoptosis in ovarian cancer cells by retaining inducible COX-2 with PD-L1 in the cytoplasm. These findings provide new insights into the antagonizing effect of T4 on resveratrol’s anticancer properties.
Introduction
Integrin αvβ3 is a structural component of the plasma membrane that is abundantly expressed in rapidly dividing endothelial cells and tumor cells (Davis et al. 2013). This integrin contains receptors for a panel of hormones and hormone-like small molecules, including resveratrol, dihydrotestosterone (DHT), thyroid hormones (THs) and their derivatives, all of which induce different behaviors in cancer cells (Davis et al. 2013). Resveratrol, a stilbenoid enriched in grape skin, has been shown to have antiproliferative properties in different cancer cell lines (Lin et al. 2013, Chin et al. 2015, Ryu et al. 2015, Wang et al. 2015, Selvaraj et al. 2016, Nana et al. 2018) and to inhibit cancer growth in vivo (Aggarwal et al. 2004, Tseng et al. 2004, Tan et al. 2016, Nana et al. 2018). By binding to its receptor on integrin αvβ3, resveratrol decreases the proliferation of human breast cancer cells, prostate cancer, thyroid cancer, glioma and ovarian cancer cells (Lin et al. 2011b ). Resveratrol exerts its anticancer activity by stimulating the activation of mitogen-activated protein kinase (MAPK; extracellular signal-regulated kinase-1 and -2 (ERK1/2)), leading to the phosphorylation and nuclear translocation of ERK1/2 and nuclear accumulation of the inducible form of cyclooxygenase (COX)-2 (Lin et al. 2013, Chin et al. 2015). Nuclear COX-2 and nuclear phosphorylated ERK1/2 (pERK1/2) in conjunction, trigger the phosphorylation of p53 at Ser-15 and consequent antiproliferation. The nuclear accumulation of COX-2 is essential for resveratrol-induced apoptosis via the integrin αvβ3 axis (Lin et al. 2011b ).
In contrast, THs such as l-thyroxine (T4) and 3, 5, 3′-triiodo-l-thyronine (T3) have been shown to enhance cancer cell proliferation (Davis et al. 2016, Lin et al. 2016b , Lee et al. 2018). Indeed, T4 and T3 bind to a receptor on integrin αvβ3 – that is proximal to, but distinct from the receptor for resveratrol – to non-genomically initiate downstream target genes expression that leads to cell proliferation and antiapoptosis (Lin et al. 2016b ). Recent studies also indicate that TH promotes human ovarian cancer cell proliferation via cross-talk between estrogen receptor α (ERα) and cell surface receptors for TH on integrin αvβ3 (Hsieh et al. 2017).
Programmed death-ligand 1 (PD-L1), also known as cluster of differentiation 274 (CD274) or B7 homolog 1 (B7-H1), is a type I transmembrane protein that is implicated in physiological regulation of the immune response at the cellular level. By binding to its receptor, programmed death 1 (PD-1), expressed on the surface of activated T lymphocytes, PD-L1 exerts an inhibitory effect on immune cells, preventing an unwanted immune reaction involving healthy cells (Keir et al. 2008). However, the ability of cancer cells to express PD-L1 and avoid activated T cell-mediated immune destruction is well described (Dong et al. 2002, Iwai et al. 2002, Zitvogel & Kroemer 2012). As a result, PD-L1 is considered a biomarker of cancer aggressiveness, and a number of anti-PD-L1 and anti-PD-1 agents are currently in use or are under development in anticancer immunotherapy (Fankhauser et al. 2015, Mazel et al. 2015, Patel & Kurzrock 2015). Expression of PD-L1 by tumor cells depends upon a variety of intracellular signaling pathways, including those involving nuclear factor (NF)-κB, MAPK, phosphoinositide 3-kinase (PI3K), mammalian target of rapamycin (mTOR) and Janus kinase/signal transducers and activators of transcription (JAK/STAT) (Ritprajak & Azuma 2015). PD-L1’s induction is regulated by receptor-mediated signaling pathways that affect the cell cycle, cell proliferation, apoptosis and cell survival, and also by intrinsic cell alterations linked to carcinogenesis (Ritprajak & Azuma 2015). In addition, the protein stability and immunosuppression function of PD-L1 are modulated by post-translational modifications, including phosphorylation, glycosylation and ubiquitination under the regulation of EGFR signaling (Li et al. 2016).
We previously showed that T4 can upregulate PD-L1 gene expression and consequent PD-L1 protein abundance in colon and breast cancer cells via activation of ERK1/2 (Lin et al. 2016a ). THs, specifically T4, blocks the antiproliferative effect of resveratrol in thyroid cancer cells and glioma cells (Lin et al. 2016b ). This inhibitory effect may relate to disruption of the pERK1/2-inducible nuclear COX-2 complex, the formation of which is induced by resveratrol, but the molecular mechanisms involved in the blockade by T4 of the anticancer activity of resveratrol are yet to be clarified. In the current report, we studied the inhibitory action of T4 on the anti-proliferative effect of resveratrol in three different ovarian cancer cell lines, OVCAR-3, A2780 and ES-2 cells. In these cells, the blockade of proliferation by resveratrol was dependent upon induction of COX-2, but the antagonism of the stilbene’s effect by T4 was shown to depend upon the expression and cytoplasmic increase of PD-L1. This effect antagonized resveratrol-induced nuclear accumulation of COX-2 and antiproliferation.
Materials and methods
Reagents
T4 was obtained from Sigma-Aldrich. T4 was prepared as a stock of 10−4 M solution in 0.04 N KOH and 4% propylene glycol. Resveratrol was purchased also from Sigma-Aldrich and was dissolved in ethanol as a 100 mM stock solution. The final concentrations of solvents in which reagents were dissolved were tested for activity and did not affect the experimental outcomes. Polyclonal rabbit anti-COX-2 (used for Western blot analysis, Catalog No.: ab15191) was acquired from Abcam, polyclonal goat anti-COX-2 (used for confocal microscopy, Catalog No.: SC-1745) was acquired from Santa Cruz Biotechnology; monoclonal mouse anti-PD-L1 (Catalog No.: #13684) was from Cell Signaling and polyclonal rabbit anti-Lamin B1 (Catalog No.: GTX103292) was from Genetex (San Antonio, TX, USA). Goat anti-rabbit immunoglobulin G (IgG) (Catalog No.: #7074) and goat anti-mouse IgG (Catalog No.: #7076) were acquired from Cell Signaling, and the Amersham ECL Western Blotting Detection Reagents were from GE Healthcare.
Cell culture
Human ovarian cancer OVCAR-3 cells were purchased from the Bioresource Collection and Research Center (BCRC, Hsinchu, Taiwan). Two other ovarian cancer cell lines, A2780 and ES-2 cells, were kindly gifted from Professor Tsui-Chin Huang (PhD Program for Cancer Molecular Biology and Drug Discovery, Taipei Medical University) who purchased them from the same center. The BCRC itself, acquired cell lines from the American Type Culture Collection (ATCC) and proceeded to their authentication. Cells were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS) and 0.01 mg/mL bovine insulin, and then applied for experiments until passage 15. All cell cultures were maintained in a 5% CO2/95% air incubator at 37°C. Prior to treatment, cells were placed in 0.25% hormone-stripped FBS-containing medium for 2 days.
Nuclear and cytosolic protein extraction and Western blot
Nuclear and cytosolic protein extracts were prepared with the NE-PER Nuclear and Cytoplasmic Extraction Kit (Thermo Scientific). The nuclear extracts were quantitated, separated on discontinuous sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred by electroblotting to polyvinylidene difluoride membranes (Millipore), as described previously (9–13). Membranes were blocked with 5% skim milk in Tris-buffered saline containing 0.1% Tween 20 and then incubated with selected antibodies overnight. Secondary antibodies were either horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG or goat anti-mouse IgG (Cell Signaling), depending on the origin of the primary antibody. Membranes were developed by chemiluminescence and measured with Amersham Imager 600 (GE Healthcare), and the integrated optical density of bands were quantitated by ImageJ software (NIH Image).
Cell viability assay
Cells (2 × 103 cells per well) were seeded in 96-well plates and either treated or left untreated for 96 h. Cell proliferation was determined by incubating cells with 200 mL of fresh medium containing 1 mg/mL 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Sigma-Aldrich) for 4 h at 37°C. After removal of the MTT solution, the resulting formazan crystals were completely dissolved in an ethanol/dimethyl sulfoxide mixture (1:1), and the plates were read using a microplate reader (VersaMax Tunable Microplate Reader, Molecular Devices, Sunnyvale, CA) by measuring the absorbance at 490 nm. Triplicate wells were assayed for each experiment, and three independent experiments were performed.
Confocal microscopy
Exponentially growing ovarian cancer cells were seeded on sterilized cover glasses (Paul Marienfeld, Lauda-Königshofen, Germany). After exposure to 0.25% stripped FBS-containing medium for 2 days, ovarian cancer cells were treated with 10 μM resveratrol, 10−7 M T4 or their combination for 24 h. Cells were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) for 30 min and then permeabilized in 0.06% Triton X-100 for 30 min. Cells were incubated with a polyclonal goat-anti-COX-2 antibody or monoclonal rabbit anti- PD-L1 antibody, followed by an Alexa-488-labeled goat anti-rabbit antibody, Alexa-488-labeled donkey anti-goat antibody or Alexa-647-labeled goat anti-rabbit antibody (Abcam) and mounted in EverBrite Hardset mounting medium with DAPI (Biotium, Fremont, CA). The fluorescent signals were recorded and analyzed with the TCS SP5 Confocal Spectral Microscope Imaging System (Leica Microsystems). The figures shown are representative of at least four fields for each experimental condition.
Quantitative real-time PCR (qPCR)
Total RNA was extracted using the NucleoSpin RNA Kit (macherey-Nagel, Germany) and complementary (c) DNA was synthesized with the RevertAid H Minus First Strand cDNA Synthesis Kit (Thermo Scientific). A quantitative real-time PCR was conducted using a QuantiNova SYBR Green PCR kit (Qiagen). Sequences of the primers amplified were Homo sapiens COX-2, forward 5′-GCCAAGCACTTTTGGTGGAG-3′ and reverse 5′-GGGACAGCCCTTCACGTTAT-3′ (accession no.: AY462100.1); PD-L1, forward 5′-GTTGAAGGACCAGCTCTCCC-3′ and reverse 5′-ACCCCTGCATCCTGCAATTT-3′ (accession no.: NM_014143.3); MYC (c-myc), forward 5′-TTCGGGTAGTGGAAAACCAG-3′ and reverse 5′-CAGCAGCTCGAATTTCTTCC-3′ (accession no.: NM_002467); B-cell lymphoma 2 (BCL-2), forward 5′-TGTTGGGCCTGTGCCTTAG-3′ and reverse 5′-CCTGAGTACCTCGTGGGAAC-3′ (accession no.: NM_000657); proliferating cell nuclear antigen (PCNA), forward 5′-TCTGAGGGCTTCGACACCTA-3′ and reverse 5′-TCATTGCCGGCGCATTTTAG-3′ (accession no.: Bbib62439.1); Bcl-2-associated death promoter (BAD), forward 5′-CTTTAAGAAGGGACTTCCTCGCC-3′ and reverse 5′-AAGTTCCGATCCCACCAGGA-3′ (accession no.: NM_004322); FAS receptor (FAS), forward 5′-TGAAGGACATGGCTTAGAAGTG-3′ and reverse 5′-GGTGCAAGGGTCACAGTGTT-3′ (accession no.: NM_152873); cyclin D1 (CCND1), forward 5′-CAAGGCCTGAACCTGAGGAG-3′ and reverse 5′-GATGACTCTGGAGAGGAAGCG-3′ (accession no.: NC_000011.10); tumor protein p53 (p53), forward 5′-AAGTCTAGAGCCACCGTCCA-3′ and reverse 5′-CAGTCTGGCTGCCAATCCA-3′ (accession no.: NM_000546.5); and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), forward 5′- TGCCAAATATGATGACATCAAGAA-3′ and reverse 5′- GGAGTGGGTGTCGCTGTTG-3′ (accession no.: NM_002046). The real-time PCRs were performed on a CFX Connect Real-Time PCR Detection System (Bio-Rad Laboratories) using the following conditions: 2 min at 50°C, 10 min at 95°C, 40 cycles of 15 s at 95°C and 1 min at 60°C. Data calculations of relative gene expressions (normalized to GAPDH reference gene) were performed according to the ΔCT method. Fidelity of the PCR was determined by a melting temperature analysis.
Transfection of shRNA
Ovarian cancer cells were seeded in 6-well tissue culture plates, grown at 60~80% confluence (105 cells/well), and maintained in the absence of antibiotic for 24 h before transfection. Just prior to transfection, the culture medium was removed, and cells were washed once with PBS, then transfected with shRNA or negative control (0.2 µg/well, RNAi core, Academia Sinica, Taipei, Taiwan) using Lipofectamine 2000 (2 µg/well) in Opti-MEM I medium according to the instructions of the manufacturer. After transfection, cultures were incubated at 37°C for 4 h and then placed in fresh culture medium for an additional 24 h. These cells were utilized in the experiments.
Statistical analysis
Immunoblots, nucleotide densities and quantitative data of confocal microscopy were analyzed with IBM SPSS Statistics software (SPSS). One-way analysis of variance was conducted for multiple groups’ comparison and two-tailed Student’s t-test was also conducted. The P value <0.05 was considered statistically significant.
Results
Resveratrol-induced antiproliferation is inhibited by thyroid hormone in ovarian cancer cells
To assess the antiproliferative effect of resveratrol on human ovarian cancer cells, OVCAR-3 cells were treated with 10 µM of resveratrol alone, T4 10−7 M alone or the combination of both agents for 96 h. Cell viability was then examined by MTT assays. As expected, resveratrol significantly inhibited cancer cell proliferation (Fig. 1A). Conversely, ovarian cancer cell proliferation was significantly stimulated when cells were treated with T4 alone. Additionally, T4 blocked the inhibitory effect of resveratrol in the co-treatment group, suggesting that resveratrol may lose its antiproliferative abilities in the presence of the thyroid hormone (Fig. 1A). Additionally, resveratrol-induced COX-2 gene expression but inhibited the cancer proliferative gene CCND1 (Fig. 1B and C). In contrast, T4-induced CCND1 expression but downregulated COX-2. Furthermore, T4 compromised resveratrol-induced COX-2 gene expression in the co-treatment group (Fig. 1B and C).
Thyroid hormone inhibits resveratrol-induced antiproliferation and nuclear accumulation of COX-2 in OVCAR-3 cells. (A) MTT assay showed cell proliferation rates of OVCAR-3 cells treated with 10 μM resveratrol (RV), 10−7 M thyroid hormone (T4) or their combination for 96 h. (B) Cells were treated with resveratrol, T4 or their combination for 24 h and collected for analysis of expression levels of COX-2 (B) and CCND1 (C) by a qPCR. (D) Cells were seeded on cover slips and received the same treatments for 24 h. Cells were fixed for confocal microscopic analysis of COX-2 nuclear accumulation. The quantitative data was also shown. Each experiment was repeated 6 times (for MTT assay) and 3 times for other assays. (Data are expressed mean ± s.d.; compared to the untreated control, *P < 0.05, **P < 0.01, ***P < 0.001; compared to resveratrol alone, ## P < 0.01, ### P < 0.001.)
Citation: Endocrine-Related Cancer 25, 5; 10.1530/ERC-17-0376
Thyroid hormone (T4) blocks resveratrol-induced COX-2 nuclear accumulation in human ovarian cancer cells
Thyroid hormone and resveratrol have receptors on integrin αvβ3, which also contains receptors for hormones such as DHT. However, while DHT and T4 promote cancer growth (Chin et al. 2015, Lin et al. 2016b ), resveratrol inhibits cancer cell proliferation via a mechanism that involves ERK1/2 phosphorylation, nuclear COX-2 accumulation and p53-Ser15 phosphorylation (Lin et al. 2011a , b ). In this process, nuclear COX-2 accumulation is essential for resveratrol-induced antiproliferation in cancer cells (Lin et al. 2011b ). We previously demonstrated that DHT and T4 prevent resveratrol-induced apoptosis in breast cancer cells (Chin et al. 2015) and glioma cells (Lin et al. 2008), respectively, by hindering resveratrol-induced COX-2 nuclear translocation. To determine whether T4 prevents resveratrol-induced COX-2 nuclear accumulation in ovarian cancer, OVCAR-3 cells were seeded on cover slips. Cells were starved for 48 h and refed with RPMI 1640 medium containing 1% hormone-stripped serum, and then treated with either resveratrol 10 µM alone, T4 10−7 M alone, or the combination of agents for another 24 h. Studies of confocal microscopy indicated that 35.8% of resveratrol-treated cells exhibited nuclear COX-2 accumulation compared to the control (7.4%). However, T4 co-treatment with resveratrol significantly reduced COX-2 nuclear translocation to 16.2% compared to resveratrol treatment alone (Fig. 1D).
This inhibitory effect of T4 on resveratrol-induced nuclear COX-2 accumulation in human OVCAR-3 cells was further confirmed in Western blot of nuclear extracts showing a parallel decrease in COX-2 protein when cells were co-incubated with T4 (Fig. 2A). In fact, treatment with T4 led to the retention of COX-2 protein in the cytosol even in the presence of resveratrol, as demonstrated in confocal microscopy experiments (Fig. 1D).
Thyroid hormone reduces resveratrol-induced nuclear accumulation of COX-2 in ovarian cancer cell lines. OVCAR-3 cells (A), A2780 cells (B) and ES-2 cells (C) were treated with 10 μM resveratrol, 10−7 M T4, or their combination for 24 h and their nuclear extracts were examined by Western blot. Resveratrol induced the nuclear accumulation of COX-2 in these three cell lines, but T4 reduced this nuclear accumulation. (Each experiment was repeated 3 times; data are expressed mean ± s.d.; compared to the untreated control, *P < 0.05, **P < 0.01, ***P < 0.001; compared to resveratrol alone, # P < 0.05, ### P < 0.001.)
Citation: Endocrine-Related Cancer 25, 5; 10.1530/ERC-17-0376
Two more ovarian cancer cell lines were used to investigate this phenomenon. Resveratrol-induced nuclear accumulation of COX-2 in ovarian cancer A2780 cells (Fig. 2B) and ES-2 cells (Fig. 2C), which was reduced by the co-incubation of T4. Taken together, these results suggested that T4 antagonized both resveratrol-induced COX2 gene expression (Fig. 1B) and nuclear translocation of COX-2 in human ovarian cancer cells.
Thyroid hormone (T4) and resveratrol oppositely modulate PD-L1 expression in ovarian cancer
Next, we sought to examine the effects of T4 and resveratrol on PD-L1 expression in ovarian cancer. OVCAR-3 cells, A2780 cells and ES-2 cells were treated with 10 µM of resveratrol, T4 at 10−7 M or their combination for 24 h. PD-L1 gene expression was significantly induced by T4, but compromised under resveratrol’s treatment (Fig. 3A). Later, we examined the accumulation of PD-L1 and nuclear accumulation of COX-2 by confocal microscopy imaging system. The accumulation of PD-L1 protein (green fluorescence) was enhanced in presence of T4, regardless of resveratrol’s treatment (Fig. 3B). Further exploration on A2780 cells and ES-2 cells also showed comparable results (Fig. 4A and B). Elsewhere, consistent with our previous findings, resveratrol induced the nuclear accumulation of COX-2 in A2780 cells (Fig. 4A) and in ES-2 cells (Fig. 4B) at 72.6% and 83.3%, respectively (Fig. 4B), compared to untreated cells. However, in the presence of T4, the nuclear accumulation of COX-2 dropped to 24.7% (A2780 cells) and 30.0% (ES-2 cells).
Thyroid hormone (T4) enhanced PD-L1 expression in ovarian cancer cell lines. (A) Ovarian cancer cells, OVCAR-3 cells (left), A2780 cells (middle) and ES-2 cells (right) were treated with T4, resveratrol or combination for 24 h and collected for analysis of expression levels of CD274 (PD-L1) by a qPCR. (B) OVCAR-3 cells with the same treatments were fixed for confocal microscopic analysis of PD-L1 expression. Perinuclear region of the merge panel was magnified; PD-L1 punctuates (arrows) only showed in combination of T4 and resveratrol. Squares in merged photos were further enlarged. PD-L1 was shown in green fluorescence and nuclear was shown in blue. (Each experiment was repeated 3 times; qPCR data are expressed mean ± s.d.; compared to the untreated control, ***P < 0.001; compared to resveratrol alone, ## P < 0.01, ### P < 0.001.)
Citation: Endocrine-Related Cancer 25, 5; 10.1530/ERC-17-0376
Thyroid hormone enhanced PD-L1 expression and reduced resveratrol-induced nuclear accumulation of COX-2 in ovarian cancer cells. (A) A2780 cells and (B) ES-2 cells were treated with 10 μM resveratrol, 10−7 M T4, or their combination for 24 h, and then examined by confocal microscopy. T4 enhanced PD-L1 (red) expression, but that was reduced by combined treatment of resveratrol in these two cells. Resveratrol-induced nuclear accumulation of COX-2 in A2780 cells and ES-2 cells, as indicated by the appearance of a purple color due to the superimposition of COX-2 (green) and PD-L1 (red) images. The quantitative data of nuclear-accumulated COX-2 were also shown. Each experiment was repeated 3 times. (Data are expressed mean ± s.d.; compared to the untreated control, **P < 0.01, ***P < 0.001; compared to resveratrol alone, ## P < 0.01, ### P < 0.001.)
Citation: Endocrine-Related Cancer 25, 5; 10.1530/ERC-17-0376
Inhibition of PD-L1 reduces the inhibitory effect of T4 on resveratrol-induced antiproliferation
We have shown that the inhibitory effect of T4 on resveratrol-induced COX-2 nuclear accumulation was demonstrable in OVCAR-3, A2780 and ES-2 ovarian cancer cells. We also found that while resveratrol impaired PD-L1 expression, T4, in contrast, promoted the immune-suppressive gene, alone or when co-administered with resveratrol. To ascertain the contribution of PD-L1 on this inhibitory effect of T4 on COX-2 nuclear accumulation, we knocked down the PD-L1 gene by shRNA techniques and then treated cells with resveratrol, T4, or their combination. We then examined whether the inhibitory effect of T4 was depleted by PD-L1 knockdown. Resveratrol-induced COX-2 gene expression was enhanced by PD-L1 knockdown in OVCAR-3 and ES-2 ovarian cancer cells (Fig. 5A and C) but not in A2780 cells (Fig. 5B). On the other hand, the inhibitory effect of thyroid hormone on resveratrol-induced COX-2 was reversed in all three ovarian cancer cell lines (Fig. 5) supporting the concept of participation of PD-L1 in the T4 inhibitory effect.
Inhibitory effects of thyroid hormone on the gene expression of COX-2 are blocked by PD-L1 knockdown in ovarian cancer cells. (A) OVCAR-3 cells, (B) A2780 cells and (C) ES-2 cells were seeded on 6-well plates and transfected with PD-L1 shRNA for 48 h. After another treatment with 10 μM resveratrol, 10−7 M T4 or their combination for 24 h, cells were collected for analysis of expression levels of CD274 (PD-L1) and COX-2 by a qPCR. The knockdown of PD-L1 enhanced resveratrol-induced COX-2 expression in OVCAR-3 cells and ES-2 cells, but not in A2780 cells. (Each experiment was repeated 3 times; data are expressed mean ± s.d.; compared to the untreated control; **P < 0.01, ***P < 0.001; compared to resveratrol alone, # P < 0.05, ### P < 0.001; compared to the mock, $ P < 0.05, $$ P < 0.01, $$$ P < 0.001.)
Citation: Endocrine-Related Cancer 25, 5; 10.1530/ERC-17-0376
Western blot analyses showed that although resveratrol-induced COX-2 dropped in PD-L1-knockdown OVCAR-3 cells, the inhibitory effect of T4 on resveratrol-induced nuclear COX-2 was reduced in PD-L1-knockdown cells (Fig. 6A). This observation suggested that PD-L1 abundance increased in the cytosol under T4 treatment. The increase in the cytosolic fraction of PD-L1 then prevented the nuclear translocation of COX-2 that is normally triggered by resveratrol treatment and is necessary for resveratrol-induced antiproliferation via integrin αvβ3 (Lin et al. 2008). We also examined PD-L1 knockdown in studies of confocal microscopy with OVCAR-3 cells. In PD-L1 knockdown samples, the total COX-2 intensity dropped while its nuclear portion increased (Fig. 6B). In these knockdown experiment images, COX-2 is displayed as condensed punctuate accumulations in both the cytosol and nucleus.
Inhibitory effects of thyroid hormone on nuclear accumulation of COX-2 are blocked by PD-L1 knockdown. OVCAR-3 cells were seeded on 10-cm petri-dishes or cover slips and transfected with PD-L1 shRNA for 48 h. After another treatment with 10 μM resveratrol, 10−7 M T4 or their combination for 24 h, cells were collected or fixed for analysis of nuclear-accumulated COX-2 by a Western blot or confocal microscopy. (Each experiment was repeated 3 times; data are expressed mean ± s.d.; compared to the untreated control; ***P < 0.001; compared to resveratrol alone, # P < 0.05; compared to the shRNA of PD-L1 only, † P < 0.05.)
Citation: Endocrine-Related Cancer 25, 5; 10.1530/ERC-17-0376
To further explore the involvement of PD-L1, mRNA levels of proliferation or apoptosis-related genes were determined in OVCAR-3 cells, A2780 cells and ES-2 cells (Fig. 7). The abundance of BAD and FAS was low in mock-transfected cells but elevated in PD-L1-knockdown cells. Meanwhile, T4-induced c-Myc expression was reduced by resveratrol and further decreased in PD-L1-compromised cells.
Inhibition of T4-induced PD-L1 reduces the inhibitory effect of T4 on antiproliferation induced by resveratrol. (A) OVCAR-3 cells, (B) A2780 cells and (C) ES-2 cells were transfected with PD-L1 shRNA for 48 h and treated with 10 μM resveratrol, 10−7 M T4 or their combination for another 24 h. Cells were collected for analysis of expression levels of FAS, BAD and c-Myc by a qPCR. (Each experiment was repeated 3 times; data are expressed mean ± s.d.; compared to the untreated control; *P < 0.05, **P < 0.01, ***P < 0.001; compared to resveratrol alone, # P < 0.05, ## P < 0.01, ### P < 0.001; compared to the mock, $ P < 0.05, $$ P < 0.01, $$$ P < 0.001.)
Citation: Endocrine-Related Cancer 25, 5; 10.1530/ERC-17-0376
Discussion
In this study, we demonstrated that T4 blocks resveratrol-induced COX-2 nuclear accumulation and antiproliferation in ovarian cancer cells via upregulation of PD-L1. Resveratrol triggered COX-2 nuclear accumulation, and this was significantly decreased when cells were co-treated with T4. T4 increased PD-L1 expression and the inhibitory effect of T4 on the action of resveratrol was inhibited by knockdown of the PD-L1 gene, suggesting that PD-L1 plays a key role in this phenomenon. Confocal microscope studies in a PD-L1-knockdown model further demonstrated that under T4 treatment, PD-L1 induced the retention of COX-2 in the cytosol and prevented its nuclear translocation; the latter is required for resveratrol’s antiproliferative effect.
The interaction between chemotherapeutic agents and the microenvironment of cancer cells has recently been emphasized. Resveratrol is a stilbene, which has antitumor properties in a variety of human cancer cells and in animal models (Aggarwal et al. 2004, Nana et al. 2018). We and others have demonstrated that resveratrol is able to induce p53-dependent apoptosis via the signal transduction processes of ERK1/2 that mediates nuclear COX-2 accumulation and the formation of phosphorylated p53-COX-2 complexes (Lin et al. 2011b ). We recently showed, however, that this clinically desirable action of resveratrol may be blocked by T4 (Lin et al. 2007) or DHT (Chin et al. 2015) in cancer cells, via a mechanism that impairs the nuclear accumulation of resveratrol-induced COX-2. Of course, additional factors may be involved in this inhibitory effect of T4. In the current study, we found that, in response to T4, the immune modulator, PD-L1, induced the sequestration of COX-2 in the cytosol and impeded its shuttling to the nucleus that is required for the antiproliferative effect of resveratrol. Post-translational modifications of PD-L1 are responsible for its immunosuppression function (Li et al. 2016), yet, at this point, whether such modifications affect its capability of cytosolic COX-2 retention remains unrevealed.
Although resveratrol has been extensively demonstrated in a variety of models to be beneficial against a range of diseases, including cancer, its translation to clinical application still suffers from important drawbacks (Singh et al. 2015). The main reason for this has been seen to be the poor bioavailability of the stilbene. After oral administration, as little as 1% of a resveratrol dose reaches the bloodstream (Walle 2011), and the agent is extensively metabolized by the small intestine and liver (Almeida et al. 2009, Cottart et al. 2010). PD-L1 is expressed by cancer cells as a defense that enables escape from anticancer immune attack by activated T cells (Shien et al. 2016). The present study indicates that, in addition to resveratrol’s low bioavailability, cancer cell PD-L1 represents a supplementary obstacle that may hinder the polyphenol’s anticancer features. Furthermore, these anti-resveratrol properties of PD-L1 are supported by physiological concentrations of T4.
In the current study, PD-L1 was barely detectable in a Western blot analysis but was visualized with confocal microscopy and its abundance was enhanced by T4 treatment. Grenga et al. (2014) and Mahoney et al. (2015) have also demonstrated low PD-L1 content in OVCAR-3 cells. Notwithstanding this low quantity, PD-L1 still mediated T4’s inhibitory effect on resveratrol. We previously found that resveratrol was able to induce COX-2 nuclear accumulation in MDA-MB-231 cells (Chin et al. 2015) which express generous amounts of PD-L1 (Grenga et al. 2014). Therefore, it is possible that the antagonistic effect of PD-L1 that prevents COX-2 nuclear shuttling preferentially occurs in settings where T4 is in ample amounts. On the other hand, our results also showed that resveratrol downregulated PD-L1 gene expression. Thus, there is a complex interplay between resveratrol and PD-L1 that presents mutually antagonistic effects. High expression of the immunosuppressive PD-L1 is a cancer cell defense that allows tumor cells to grow in stealth mode, undetected by the immune system. This effect has generated much interest in the development of anti-PD-L1 agents, and these have been shown to have clinically important effects (Ott et al. 2013, Powles et al. 2014, Robert et al. 2014).
PD-L1 is variously expressed in cancers. Indeed, various signals and microenvironment contingencies may affect tumors' ability to generate this immune-escape protein. PD-L1 is expressed in response to various stimuli including (NF)-κB, MAPK, (PI3K), (mTOR) and (JAK/STAT) (Ritprajak & Azuma 2015) and T4 as demonstrated in the current work. Therefore, within the same type of tumor, different patients may exhibit different levels of PD-L1 on the tumor sites that dictate different outcomes as summarized by Wang and colleagues (Wang et al. 2016). That different ovarian cancer cells express different levels of PD-L1 has also been described. Nonetheless, high levels of PD-L1 is correlated with peritoneal dissemination, the most frequent metastatic and recurrent site for ovarian cancer and one of the most important prognostic factors (Abiko et al. 2013). More investigations in clinical settings are needed to determine the association between thyroid hormone levels and PD-L1 expression in their subsequent prognostic value in ovarian cancer.
In summary, we have demonstrated that PD-L1 is involved in the inhibitory effect of the thyroid hormone (T4) on resveratrol-induced antiproliferation in ovarian cancer cells. By downregulating PD-L1 gene expression in cancer cells, resveratrol represents an attractive agent that could possibly be employed in combination with other anti-PD-L1 drugs. In response to T4, PD-L1 impaired the nuclear translocation of COX-2, which is the essential mechanism of resveratrol’s anticancer property via the integrin αvβ3 axis.
Declarations of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this review.
Funding
This work was supported in part by a TMU Affiliated Hospital Collaborating Program (104TMU-TMUH-13) and by grants from the Ministry of Science and Technology, Taiwan (MOST104-2320-B-038-009, MOST105-2320-B-038-006 and MOST104-2314-B-038-046-MY3).
Author contribution statement
Y T C, Y H, and H Y L designed experiments; M T H, C A C, A W N, Y R C, Y T C, and Y J S performed experiments; Y T C, A W N, Y C Y, Y H, and A H analyzed the data; A W N and H Y L wrote the manuscript; H Y L, P L W, and P J D supervised the work.
References
Abiko K, Mandai M, Hamanishi J, Yoshioka Y, Matsumura N, Baba T, Yamaguchi K, Murakami R, Yamamoto A, Kharma B, et al.2013 PD-L1 on tumor cells is induced in ascites and promotes peritoneal dissemination of ovarian cancer through CTL dysfunction. Clinical Cancer Research 19 1363–1374. (https://doi.org/10.1158/1078-0432.CCR-12-2199)
Aggarwal BB, Bhardwaj A, Aggarwal RS, Seeram NP, Shishodia S & Takada Y 2004 Role of resveratrol in prevention and therapy of cancer: preclinical and clinical studies. Anticancer Research 24 2783–2840.
Almeida L, Vaz-da-Silva M, Falcao A, Soares E, Costa R, Loureiro AI, Fernandes-Lopes C, Rocha JF, Nunes T, Wright L, et al.2009 Pharmacokinetic and safety profile of trans-resveratrol in a rising multiple-dose study in healthy volunteers. Molecular Nutrition and Food Research 53 (Supplement 1) S7–S15. (https://doi.org/10.1002/mnfr.200800177)
Chin YT, Yang SH, Chang TC, Changou CA, Lai HY, Fu E, HuangFu WC, Davis PJ, Lin HY & Liu LF 2015 Mechanisms of dihydrotestosterone action on resveratrol-induced anti-proliferation in breast cancer cells with different ERalpha status. Oncotarget 6 35866–35879. (https://doi.org/10.18632/oncotarget.5482)
Cottart CH, Nivet-Antoine V, Laguillier-Morizot C & Beaudeux JL 2010 Resveratrol bioavailability and toxicity in humans. Molecular Nutrition and Food Research 54 7–16. (https://doi.org/10.1002/mnfr.200900437)
Davis PJ, Mousa SA, Cody V, Tang HY & Lin HY 2013 Small molecule hormone or hormone-like ligands of integrin alphaVbeta3: implications for cancer cell behavior. Hormones and Cancer 4 335–342. (https://doi.org/10.1007/s12672-013-0156-8)
Davis PJ, Glinsky GV, Lin HY & Mousa SA 2016 Actions of thyroid hormone analogues on chemokines. Journal of Immunology Research 2016 3147671. (https://doi.org/10.1155/2016/3147671)
Dong H, Strome SE, Salomao DR, Tamura H, Hirano F, Flies DB, Roche PC, Lu J, Zhu G, Tamada K, et al. 2002 Tumor-associated B7-H1 promotes T-cell apoptosis: a potential mechanism of immune evasion. Nature Medicine 8 793–800. (https://doi.org/10.1038/nm730)
Fankhauser CD, Curioni-Fontecedro A, Allmann V, Beyer J, Tischler V, Sulser T, Moch H & Bode PK 2015 Frequent PD-L1 expression in testicular germ cell tumors. British Journal of Cancer 113 411–413. (https://doi.org/10.1038/bjc.2015.244)
Grenga I, Donahue RN, Lepone L, Bame J, Schlom J & Farsaci B 2014 PD-L1 and MHC-I expression in 19 human tumor cell lines and modulation by interferon-gamma treatment. Journal for Immunotherapy of Cancer 2 P102. (https://doi.org/10.1186/2051-1426-2-S3-P102)
Hsieh MT, Wang LM, Changou CA, Chin YT, Yang YSH, Lai HY, Lee SY, Yang YN, Whang-Peng J, Liu LF, et al. 2017 Crosstalk between integrin alphavbeta3 and ERalpha contributes to thyroid hormone-induced proliferation of ovarian cancer cells. Oncotarget 8 24237–24249. (https://doi.org/10.18632/oncotarget.10757)
Iwai Y, Ishida M, Tanaka Y, Okazaki T, Honjo T & Minato N 2002 Involvement of PD-L1 on tumor cells in the escape from host immune system and tumor immunotherapy by PD-L1 blockade. PNAS 99 12293–12297. (https://doi.org/10.1073/pnas.192461099)
Keir ME, Butte MJ, Freeman GJ & Sharpe AH 2008 PD-1 and its ligands in tolerance and immunity. Annual Review of Immunology 26 677–704. (https://doi.org/10.1146/annurev.immunol.26.021607.090331)
Lee YS, Chin YT, Shih YJ, Nana AW, Chen YR, Wu HC, Yang YSH, Lin HY & Davis PJ 2018 Thyroid hormone promotes beta-catenin activation and cell proliferation in colorectal cancer. Hormones and Cancer [epub]. (https://doi.org/10.1007/s12672-018-0324-y)
Li CW, Lim SO, Xia W, Lee HH, Chan LC, Kuo CW, Khoo KH, Chang SS, Cha JH, Kim T, et al. 2016 Glycosylation and stabilization of programmed death ligand-1 suppresses T-cell activity. Nature Communications 7 12632. (https://doi.org/10.1038/ncomms12632)
Lin HY, Tang HY, Shih A, Keating T, Cao G, Davis PJ & Davis FB 2007 Thyroid hormone is a MAPK-dependent growth factor for thyroid cancer cells and is anti-apoptotic. Steroids 72 180–187. (https://doi.org/10.1016/j.steroids.2006.11.014)
Lin HY, Tang HY, Keating T, Wu YH, Shih A, Hammond D, Sun M, Hercbergs A, Davis FB & Davis PJ 2008 Resveratrol is pro-apoptotic and thyroid hormone is anti-apoptotic in glioma cells: both actions are integrin and ERK mediated. Carcinogenesis 29 62–69. (https://doi.org/10.1093/carcin/bgm239)
Lin C, Crawford DR, Lin S, Hwang J, Sebuyira A, Meng R, Westfall JE, Tang HY, Lin S, Yu PY, et al. 2011a Inducible COX-2-dependent apoptosis in human ovarian cancer cells. Carcinogenesis 32 19–26. (https://doi.org/10.1093/carcin/bgq212)
Lin HY, Tang HY, Davis FB & Davis PJ 2011b Resveratrol and apoptosis. Annals of the New York Academy of Sciences 1215 79–88. (https://doi.org/10.1111/j.1749-6632.2010.05846.x)
Lin HY, Delmas D, Vang O, Hsieh TC, Lin S, Cheng GY, Chiang HL, Chen CE, Tang HY, Crawford DR, et al.2013 Mechanisms of ceramide-induced COX-2-dependent apoptosis in human ovarian cancer OVCAR-3 cells partially overlapped with resveratrol. Journal of Cellular Biochemistry 114 1940–1954. (https://doi.org/10.1002/jcb.24539)
Lin HY, Chin YT, Nana AW, Shih YJ, Lai HY, Tang HY, Leinung M, Mousa SA & Davis PJ 2016a Actions of l-thyroxine and nano-diamino-tetrac (nanotetrac) on PD-L1 in cancer cells. Steroids 114 59–67. (https://doi.org/10.1016/j.steroids.2016.05.006)
Lin HY, Chin YT, Yang YC, Lai HY, Wang-Peng J, Liu LF, Tang HY & Davis PJ 2016b Thyroid hormone, cancer, and apoptosis. Comprehensive Physiology 6 1221–1237. (https://doi.org/10.1002/cphy.c150035)
Mahoney KM, Sun H, Liao X, Hua P, Callea M, Greenfield EA, Hodi FS, Sharpe AH, Signoretti S, Rodig SJ, et al. 2015 PD-L1 antibodies to Its cytoplasmic domain most clearly delineate cell membranes in immunohistochemical staining of tumor cells. Cancer Immunology Research 3 1308–1315. (https://doi.org/10.1158/2326-6066.CIR-15-0116)
Mazel M, Jacot W, Pantel K, Bartkowiak K, Topart D, Cayrefourcq L, Rossille D, Maudelonde T, Fest T & Alix-Panabieres C 2015 Frequent expression of PD-L1 on circulating breast cancer cells. Molecular Oncology 9 1773–1782. (https://doi.org/10.1016/j.molonc.2015.05.009)
Nana AW, Chin YT, Lin CY, Ho Y, Bennett JA, Shih YJ, Chen YR, Changou CA, Pedersen JZ, Incerpi S, et al. 2018 Tetrac downregulates beta-catenin and HMGA2 to promote the effect of resveratrol in colon cancer. Endocrine-Related Cancer 25 279–293. (https://doi.org/10.1530/ERC-17-0450)
Ott PA, Hodi FS & Robert C 2013 CTLA-4 and PD-1/PD-L1 blockade: new immunotherapeutic modalities with durable clinical benefit in melanoma patients. Clinical Cancer Research 19 5300–5309. (https://doi.org/10.1158/1078-0432.CCR-13-0143)
Patel SP & Kurzrock R 2015 PD-L1 expression as a predictive biomarker in cancer immunotherapy. Molecular Cancer Therapeutics 14 847–856. (https://doi.org/10.1158/1535-7163.MCT-14-0983)
Powles T, Eder JP, Fine GD, Braiteh FS, Loriot Y, Cruz C, Bellmunt J, Burris HA, Petrylak DP, Teng SL, et al. 2014 MPDL3280A (anti-PD-L1) treatment leads to clinical activity in metastatic bladder cancer. Nature 515 558–562. (https://doi.org/10.1038/nature13904)
Ritprajak P & Azuma M 2015 Intrinsic and extrinsic control of expression of the immunoregulatory molecule PD-L1 in epithelial cells and squamous cell carcinoma. Oral Oncology 51 221–228. (https://doi.org/10.1016/j.oraloncology.2014.11.014)
Robert C, Ribas A, Wolchok JD, Hodi FS, Hamid O, Kefford R, Weber JS, Joshua AM, Hwu WJ, Gangadhar TC, et al. 2014 Anti-programmed-death-receptor-1 treatment with pembrolizumab in ipilimumab-refractory advanced melanoma: a randomised dose-comparison cohort of a phase 1 trial. Lancet 384 1109–1117. (https://doi.org/10.1016/S0140-6736(14)60958-2)
Ryu J, Yoon NA, Seong H, Jeong JY, Kang S, Park N, Choi J, Lee DH, Roh GS, Kim HJ, et al. 2015 Resveratrol induces glioma cell apoptosis through activation of tristetraprolin. Molecules and Cells 38 991–997. (https://doi.org/10.14348/molcells.2015.0197)
Selvaraj S, Sun Y, Sukumaran P & Singh BB 2016 Resveratrol activates autophagic cell death in prostate cancer cells via downregulation of STIM1 and the mTOR pathway. Molecular Carcinogenesis 55 818–831. (https://doi.org/10.1002/mc.22324)
Shien K, Papadimitrakopoulou VA & Wistuba II 2016 Predictive biomarkers of response to PD-1/PD-L1 immune checkpoint inhibitors in non-small cell lung cancer. Lung Cancer 99 79–87. (https://doi.org/10.1016/j.lungcan.2016.06.016)
Singh CK, Ndiaye MA & Ahmad N 2015 Resveratrol and cancer: challenges for clinical translation. Biochimica et Biophysica Acta 1852 1178–1185. (https://doi.org/10.1016/j.bbadis.2014.11.004)
Tan L, Wang W, He G, Kuick RD, Gossner G, Kueck AS, Wahl H, Opipari AW & Liu JR 2016 Resveratrol inhibits ovarian tumor growth in an in vivo mouse model. Cancer 122 722–729. (https://doi.org/10.1002/cncr.29793)
Tseng SH, Lin SM, Chen JC, Su YH, Huang HY, Chen CK, Lin PY & Chen Y 2004 Resveratrol suppresses the angiogenesis and tumor growth of gliomas in rats. Clinical Cancer Research 10 2190–2202. (https://doi.org/10.1158/1078-0432.CCR-03-0105)
Walle T 2011 Bioavailability of resveratrol. Annals of the New York Academy of Sciences 1215 9–15. (https://doi.org/10.1111/j.1749-6632.2010.05842.x)
Wang G, Dai F, Yu K, Jia Z, Zhang A, Huang Q, Kang C, Jiang H & Pu P 2015 Resveratrol inhibits glioma cell growth via targeting oncogenic microRNAs and multiple signaling pathways. International Journal of Oncology 46 1739–1747. (https://doi.org/10.3892/ijo.2015.2863)
Wang X, Teng F, Kong L & Yu J 2016 PD-L1 expression in human cancers and its association with clinical outcomes. OncoTargets and Therapy 9 5023–5039. (https://doi.org/10.2147/OTT.S105862)
Zitvogel L & Kroemer G 2012 Targeting PD-1/PD-L1 interactions for cancer immunotherapy. Oncoimmunology 1 1223–1225. (https://doi.org/10.4161/onci.21335)
