2020 Volume 45 Issue 4 Pages 227-236
A growing body of experimental evidence strongly suggests that cannabidiolic acid (CBDA), a major component of the fiber-type cannabis plant, exerts a variety of biological activities. We have reported that CBDA can abrogate cyclooxygenase-2 (COX-2) expression and its enzymatic activity. It is established that aberrant expression of COX-2 correlates with the degree of malignancy in breast cancer. Although the reduction of COX-2 expression by CBDA offers an attractive medicinal application, the molecular mechanisms underlying these effects have not fully been established. It has been reported that COX-2 expression is positively controlled by peroxisome proliferator-activated receptor β/δ (PPARβ/δ) in some cancerous cells, although there is “no” modulatory element for PPARβ/δ on the COX-2 promoter. No previous studies have examined whether an interaction between PPARβ/δ-mediated signaling and COX-2 expression exists in MDA-MB-231 cells. We confirmed, for the first time, that COX-2 expression is positively modulated by PPARβ/δ-mediated signaling in MDA-MB-231 cells. CBDA inhibits PPARβ/δ-mediated transcriptional activation stimulated by the PPARβ/δ-specific agonist, GW501516. Furthermore, the disappearance of cellular actin stress fibers, a hallmark of PPARβ/δ and COX-2 pathway activation, as evoked by the GW501516, was effectively reversed by CBDA. Activator protein-1 (AP-1)-driven transcriptional activity directly involved in the regulation of COX-2 was abrogated by the PPARβ/δ-specific inverse agonists (GSK0660/ST-247). Thus, it is implicated that there is positive interaction between PPARβ/δ and AP-1 in regulation of COX-2. These data support the concept that CBDA is a functional down-regulator of COX-2 through the abrogation of PPARβ/δ-related signaling, at least in part, in MDA-MB-231 cells.
Contrary to some earlier findings, accumulating experimental evidence now strongly suggests that cannabidiolic acid (CBDA) (Fig. 1A), a constituent of the fiber-type cannabis plant, is a biologically active component. For example, CBDA is reported to exert anti-bacterial effects (Appendino et al., 2008), cyclooxygenase-2 (COX-2) enzyme inhibition (Takeda et al., 2008; Takeda, 2013), down-regulation of COX-2 expression (Takeda et al., 2014a, 2017; Suzuki et al., 2017), and anti-nausea/emetic effects (Bolognini et al., 2013; Rock and Parker, 2013).
PPARβ/δ-mediated signaling is involved in the modulation of COX-2 expression in MDA-MB-231 cells cultured in an incomplete system. (A) The structure of CBDA. (B) Real-time RT-PCR analysis of COX-2 after 48 hr of transfection with pcDNA3.1 plasmid (60 ng) (mock; indicated as –) or the PPARβ/δ expression plasmids (60 ng) (indicated as +). The MDA-MB-231 cells were cultured in a complete system (medium with phenol red + non-DCC-FBS) or cultured in an incomplete system (medium without phenol red + DCC-FBS). The data are expressed as the mean ± S.E. (n = 6) of the fold induction from the mock control in the incomplete system. Significant differences (two-way ANOVA followed by Tukey-Kramer’s post-hoc test) from the vehicle-treated control are marked with an asterisk (*P < 0.05). (C, D) The effects of GSK0660 on PPARβ/δ-mediated transcriptional activity on MDA-MB-231 cells are shown. MDA-MB-231 cells were transiently transfected with a PPRE-luciferase reporter plasmid in combination with the PPARβ/δ expression plasmids (C) or with a PPRE-luciferase reporter plasmid alone (D). After transfection, the cells were treated with vehicle (indicated as Ctl.) or GSK0660 (0.2, 1, and 5 μM). After 24 hr, the cells were harvested and assayed for luciferase activity, and all the transfections were normalized for efficiency using the internal Renilla control plasmid. The data are expressed as the mean ± S.E. (n = 3) of the fold induction as compared to the vehicle-treated control. Significant differences (one-way ANOVA, followed by Dunnett’s post-hoc test) from the vehicle-treated control are marked with an asterisk (*P < 0.05). (E) Real-time RT-PCR analyses of COX-2 in the MDA-MB-231 cells after 48 hr of treatment with vehicle (indicated as Ctl.) and GSK0660 (0.2, 1, and 5 μM). The data are expressed as the mean ± S.E. (n = 6) of the fold induction as compared to the vehicle-treated control. Significant differences (one-way ANOVA, followed by Dunnett’s post-hoc test) from the vehicle-treated control are marked with an asterisk (*P < 0.05). All data shown are representative of three experiments, and n (= 3 or 6) indicates the number of technical replicates in the experiments.
The expression of COX-2 in certain cancer cells, including breast cancer cells, is highly up-regulated and is associated with abnormal growth character (Majumder et al., 2016). Based on these aspects, there is an interest to develop therapeutic modalities to abrogate the levels of COX-2 expression. Previously, we demonstrated that CBDA can interfere with COX-2 expression in MDA-MB-231 breast cancer cells, a highly aggressive triple-negative breast cancer cell line, and that CBDA exhibits selective inhibition of COX-2, as shown with purified COX-2 used as an enzyme source (Takeda et al., 2008, 2014a, 2017; Takeda, 2013; Suzuki et al., 2017). The MDA-MB-231 cell line is also an accepted model of estrogen receptor α (ERα)-negative breast cancer as the cells do not require estrogens for growth, and MDA-MB-231 cells express all the subtypes of the peroxisome proliferator-activated receptors, PPARα, PPARβ/δ, and PPARγ (Suchanek et al., 2002a, 2002b; Takeda et al., 2013, 2014b; Hirao-Suzuki et al., 2019b). Many studies have examined the modulatory effects of COX-2 expression at various levels; for example, COX-2 is likely a critical contributor to the downstream effects of activator protein-1 (AP-1), nuclear factor-κB (NF-κB), and PPARβ/δ (Dong et al., 1997; Schmedtje et al., 1997; Glinghammar et al., 2003; Wang et al., 2014). We reported that CBDA abrogates AP-1-mediated transcription though the down-regulation of c-fos, a component of AP-1 heterodimers, that in turn is coupled to the reduction of COX-2 expression (Suzuki et al., 2017; Takeda et al., 2017). However, the inhibitory degree by CBDA between AP-1-mediated transcription (~30% inhibition) and COX-2 expression (~50% inhibition) was different; thus, additional mechanism(s) seem to be involved. It has been reported that COX-2 expression is positively controlled by PPARβ/δ in some cancerous cells, whereas there is no modulatory element for PPARβ/δ on the COX-2 promoter. It remains unclear whether CBDA affects PPARβ/δ-related signaling, and thereby potentially leading to the down-regulation of COX-2.
We selected the human breast cancer MDA-MB-231 cell line as a model to investigate this unaddressed question since these cells express functional COX-2, a feature associated with their highly invasive nature (Miller et al., 2005; Yoon et al., 2015; Majumder et al., 2016). In this study, we investigated the effects of CBDA on transcription mediated by PPARβ/δ in MDA-MB-231 cells. We report that CBDA abrogates PPARβ/δ transcription leading to the reduced expression of COX-2 and reverses the reduction of actin stress fiber formation, a hallmark feature of activated COX-2 signaling pathways (Bos et al., 2004; Takai et al., 2013). AP-1-driven transcriptional activity was abrogated by two kinds of inverse agonists specific for PPARβ/δ. Thus, it is implicated that there is positive interaction between PPARβ/δ and AP-1 in regulation of COX-2. These data suggest the concept that CBDA is a functional down-regulator of COX-2 through the abrogation of PPARβ/δ-related signaling.
CBDA (purity: ≥ 96.5%), GSK0660 (purity: ≥ 98%), GW501516 (purity: ≥ 98%), and ST-247 (purity: ≥ 98%) were purchased from Sigma-Aldrich (St. Louis, MO, USA).
Cell culturesCell culture conditions/methods were based on procedures described previously (Takeda et al., 2014a, 2017; Suzuki et al., 2017). Briefly, MDA-MB-231 cells (obtained from the American Type Culture Collection, Rockville, MD, USA) were routinely grown in phenol red-containing minimum essential medium α (MEMα) (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan), supplemented with 10 mM HEPES [4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid], 5% fetal bovine serum (FBS), penicillin (100 U/mL), and streptomycin (100 μg/mL) at 37ºC in a 5% CO2–95% air-humidified incubator. Prior to the 24 hr chemical exposure, the culture medium was changed to phenol red-free MEMα (FUJIFILM Wako Pure Chemical Corporation) supplemented with 10 mM HEPES, 5% dextran-coated charcoal-treated FBS (DCC-FBS), penicillin (100 U/mL), and streptomycin (100 μg/mL). CBDA was prepared in ethanol. GSK0660 and GW501516 were prepared with dimethyl sulfoxide (DMSO). As a control, only these solvents were added for incubation.
Real-time reverse transcription-polymerase chain reaction (RT-PCR)Real-time RT-PCR was performed as described previously (Hirao-Suzuki et al., 2019a, 2019b). The following primers were used: COX-2 (sense), 5’-CCGGGTACAATCGCACTTAT-3’; COX-2 (antisense), 5’-GGCGCTCAGCCATACAG-3’; FA2H (sense), 5’-AACGAGCCTGTAGCCCTTGA-3’; FA2H (antisense), 5’-ACTGCCACCGTGTACTCTGTT-3’; adiponectin (sense), 5’-GGTCTCGAACTCCTGGCCTA-3’; adiponectin (antisense), 5’-TGAGATATCGACTGGGCATGGT-3’; β-actin (sense), 5’-GGCCACGGCTGCTTC-3’; β-actin (antisense), 5’-GTTGGCGTACAGGTCTTTGC-3’. The primers for β-actin, COX-2, FA2H, and adiponectin were previously described (Takeda et al., 2013, 2014b; Roelofs et al., 2014; Okazaki et al., 2015). COX-2, FA2H, and adiponectin mRNA levels were normalized to β-actin.
Transfection and luciferase reporter assay (dual-luciferase assay)The experiments were performed as described previously with slight modification (Okazaki et al., 2015; Takeda et al., 2014b; Hirao-Suzuki et al., 2019b). Twenty-four hours before transfection, MDA-MB-231 cells (5 × 104 cells) were seeded onto 24-well plates containing MEMα. Each expression plasmid was transfected using Lipofectamine® LTX with PLUSTM reagent (Thermo Fisher Scientific, Waltham, MA, USA). The DNA mixtures containing 300 ng of the PPAR response element (PPRE)-Luc plasmid encoding the rat acyl-CoA oxidase PPRE or pGL4.44[luc2P/APl RE/Hygro] vector for AP-l (Promega, Madison, WI, USA) with/without 100 ng of human PPARβ/δ expression plasmid and 100 ng of human retinoid X receptor α (RXRα) expression plasmid were co-transfected with 2 ng of the Renilla luciferase reporter plasmid (pRL-CMV). Human PPARβ/δ expression plasmid, human RXRα expression plasmid, and PPRE reporter plasmid were kindly gifted by Dr. Curtis J. Omiecinski (Pennsylvania State University, PA, USA). The cells were washed with phosphate-buffered saline 24 hr post-transfection and changed to phenol red-free MEMα supplemented with 5% DCC-FBS followed by the respective chemical treatment. The cell extracts were then prepared using 100 μL of passive lysis buffer (Promega), and 20 μL was used to perform the firefly luciferase and Renilla luciferase assays (Dual-Luciferase Reporter Assay System, Promega). The ratio of firefly luciferase activity to Renilla luciferase activity in each sample served as a measure of the normalized luciferase activity.
Staining of actin stress fibersThe MDA-MB-231 cells were seeded onto NuncTM Lab-TekTM II Chamber SlideTM systems (Thermo Fisher Scientific) at 1 × 104 cells/chamber, followed by 4 hr incubation. The cells were then exposed to vehicle, 10 nM GW501516, or 10 nM GW501516/5 μM CBDA for 48 hr. After chemical treatment, the cells were fixed in 4% paraformaldehyde, permeabilized with 0.5% Triton X-100 in PBS, and blocked with Blocking One Histo (Nacalai Tesque, Kyoto, Japan). The F-actin structure of the cells was visualized by Alexa Fluor® 488 Phalloidin (Thermo Fisher Scientific). Staining of F-actin was performed for 1 hr at room temperature. The coverslips were mounted with FLUOROSHIELDTM Mounting Medium (ImmunoBioScience Corp., Mukilteo, WA, USA). The cells were visualized by using a laser scanning confocal microscope, FV-300/IX71 (Olympus, Tokyo, Japan).
Data analysisThe IC20 values were calculated by fitting the dose-response curves using SigmaPlot11 software (Systat Software, Inc., San Jose, CA, USA). Differences were considered significant when the P-value was < 0.05. Differences between the two groups were calculated by Student’s t-test. ANOVA with Dunnett’s or Tukey-Kramer’s post-hoc test was used for the comparison of more than two groups. These analyses were performed using Statview 5.0J software (SAS Institute Inc., Cary, NC, USA).
An analysis of the expression profiles of the individual PPAR subtypes revealed that all the PPARs were detected in the MDA-MB-231 cells (data not shown) (Suchanek et al., 2002a, 2002b; Yao et al., 2014; Hirao-Suzuki et al., 2019b). Although a positive interplay between PPARβ/δ and COX-2 was implicated in human hepatocellular carcinoma cells (HepG2 cells) and lung carcinoma cells (H358 and H441 cells) (Glinghammar et al., 2003; Genini et al., 2012), no previous studies have reported whether a positive interaction between PPARβ/δ-mediated signaling and COX-2 expression exists in the MDA-MB-231 cell line. Initially, we determined the optimal experimental conditions to detect the interaction by focusing on the difference of culture conditions, that is, MDA-MB-231 cells were cultured in a complete system (i.e., medium with phenol red + non-DCC-FBS) or cultured in an incomplete system (i.e., medium without phenol red + DCC-FBS). As clearly shown in Fig. 1B, when compared to the incomplete system, the COX-2 expression was up-modulated (~1.8-fold) in the complete system. Since hormones and phenol red (which are involved in normal serum and medium, respectively) can affect signaling pathways mediated by some nuclear receptors such as ERs, to simply detect the effects of compounds the incomplete system without them is preferred. In addition, we previously reported that CBDA down-regulates COX-2 expression in MDA-MB-231 cells cultured in the incomplete system (Suzuki et al., 2017; Takeda et al., 2017). Since the MDA-MB-231 cell line can grow in an estrogens-independent manner, together with the above-mentioned phenomena, we selected the incomplete system in the subsequent study. We next sought to establish the potential functional involvements of PPARβ/δ in the modulation of COX-2 in MDA-MB-231 cells by means of several experiments: i) introduction of human PPARβ/δ cDNA (60 ng; exhibiting the maximum induction) into the cells further stimulated the expression of COX-2 (~2.3-fold), although this stimulation was not significantly detected in the complete system (Fig. 1B), ii) the transcriptional activity mediated by ectopically expressed PPARβ/δ and basally expressed PPAR (non-transfection) was suppressed by GSK0660, a selective PPARβ/δ antagonist (Shearer et al., 2008), in a concentration-dependent manner (Fig. 1C and D), and iii) the antagonist of PPARβ/δ inhibited COX-2 expression (Fig. 1E). These results confirm that COX-2 expression is positively modulated by PPARβ/δ-mediated signaling in the MDA-MB-231 cells in the incomplete system. Although the notion that PPARβ/δ increases COX-2 expression is sometimes contradicted by other reports, even in the case with the findings in HepG2 cells by Glinghammar et al., 2003, it might be indicated that there is a positive correlation between PPARβ/δ and COX-2 in some experimental settings.
Prior to the following analyses, cell viability assays were performed with CBDA. CBDA did not affect the viability of MDA-MB-231 cells even after 48 hr exposure to the respective concentration ranging from 1 to 50 μM (data not shown). Similar to our previously reported results (Takeda et al., 2014a, 2017; Suzuki et al., 2017), the expression of COX-2 was suppressed by CBDA at a concentration of 5 μM (Fig. 2B). This CBDA concentration was selective in reducing the COX-2 gene in the MDA-MB-231 cells as compared to the other target genes tested. Fatty acid 2-hydroxylase (FA2H) and adiponectin, which are regulated by PPARα and PPARγ, respectively (Takeda et al., 2013, 2014b; Okazaki et al., 2015; Hirao-Suzuki et al., 2019b), were not modulated (Fig. 2A and C). The mRNA expression level of PPARβ/δ was not affected by 5 μM CBDA itself (data not shown). Taken these lines of results into consideration, it is suggested that CBDA can abrogate the PPARβ/δ-mediated positive modulation of COX-2, at least in part, in the MDA-MB-231 cells.
CBDA selectively down-regulates COX-2. Real-time RT-PCR analyses were used to analyze the expression levels of FA2H (a PPARα-regulated gene) (A), COX-2 (a PPARβ/δ-regulated gene) (B), and adiponectin (a PPARγ-regulated gene) (C) in the MDA-MB-231 cells after 48 hr of treatment with vehicle (indicated as Ctl.) or 5 μM CBDA. The data are expressed as the mean ± S.E. (n = 6) of the fold induction as compared to the vehicle-treated control. Significant differences (by Student’s t-test) to the vehicle-treated control are marked with an asterisk (*P < 0.05). All data shown are representative of three experiments, and n (= 6) indicates the number of technical replicates in the experiments.
It is important to determine more mechanistically how CBDA may inhibit PPARβ/δ activity, in turn leading to the down-regulation of COX-2. We utilized GW501516, a highly selective agonist for PPARβ/δ, which displays 1000-fold selectivity over the other human PPARs (Oliver et al., 2001). In our experimental conditions, GW501516 exhibited the maximum activation potential for PPARβ/δ at 10 nM (~1.5-fold activation) (Fig. 3A). If CBDA affects PPARβ/δ-mediated transcriptional activity through a possible antagonistic action, CBDA might weaken the GW501516-activated stimulation of PPARβ/δ. In support of this reasoning, CBDA was found to inhibit GW501516 activation in a concentration-dependent manner with an IC20 value of 6.12 μM (Fig. 3B). It is important to determine whether the negative interaction between GW501516 and CBDA in the PPARβ/δ modulation was functionally relevant. As shown in Fig. 4A, a significant up-regulation of COX-2 expression (~3-fold) was observed at 10 nM concentration of GW501516, and the GW501516-stimulated increase in COX-2 expression was inhibited by CBDA (Fig. 4B). It has been reported that the transcriptional activities of PPARs can be modulated by positive/negative cross-talk with other signaling pathways mediated by nuclear receptors, such as ERα/β (Keller et al., 1995; Wang and Kilgore, 2002). MDA-MB-231 cells do express functional ERβ (Lazennec et al., 2001; Takeda et al., 2013). If CBDA behaves as a ligand (i.e., an agonist/antagonist) for ERβ, the function of PPARβ/δ might be affected by such interaction with ERβ. However, 17β-estradiol-induced ERβ activity was not modulated in the presence of the highest concentration of CBDA used (25 μM) (data not shown), implying that CBDA tends to attenuate the transcriptional activities of PPARβ/δ expressed in the MDA-MB-231 cells.
CBDA abrogates PPARβ/δ agonist-induced transcriptional activity. (A and B) Effects of GW501516 on PPARβ/δ-mediated transcriptional activity (B) and the effects of CBDA on GW-mediated stimulation of PPARβ/δ (B) in MDA-MB-231 cells. The cells were transiently transfected with a PPRE-luciferase reporter plasmid in combination with the PPARβ/δ expression plasmids. After transfection, the cells were treated with vehicle (indicated as Ctl.) and 10 nM GW501516 (A) or 10 nM GW501516 alone (indicated as Ctl.) and 10 nM GW501516 in combination with 1 μM or 5 μM CBDA (B). After 24 hr, the cells were harvested and assayed for luciferase activity, and all transfections were normalized for efficiency using the internal Renilla control plasmid. The data are expressed as the mean ± S.E. (n = 3) of the fold induction as compared to the vehicle-treated control (A) or 10 nM GW501516-treated alone control (B). Significant differences [Student’s t-test for (A), and one-way ANOVA followed by Dunnett’s post-hoc test for (B)] from each control are marked with an asterisk (*P < 0.05). All data shown are representative of three experiments, and n (= 3) indicates the number of technical replicates in the experiments.
CBDA abrogates PPARβ/δ agonist-induced COX-2 expression and reverses PPARβ/δ agonist-mediated interference with SFs. (A and B) Real-time RT-PCR analyses of COX-2 in MDA-MB-231 cells after 48 hr of treatment with vehicle (indicated as Ctl.) and 10 nM GW501516 (A) or 10 nM GW501516 alone (indicated as Ctl.) and 10 nM GW501516 in combination with 1 μM or 5 μM CBDA (B). The data are expressed as the mean ± S.E. (n = 6) of the fold induction compared to the vehicle-treated control (A) or 10 nM GW501516-treated alone control (B). Significant differences [Student’s t-test for (A), and one-way ANOVA followed by Dunnett’s post-hoc test for (B)] from each control are marked with an asterisk (*P < 0.05). (C) The MDA-MB-231 cells were treated with vehicle (left panel), 10 nM GW501516 (middle panel), or 10 nM GW501516/5 μM CBDA (right panel) for 48 hr. Actin stress fibers were detected with phalloidin under permeabilized conditions. All data shown are representative of three experiments, and n (= 6) indicates the number of technical replicates in the experiments.
When COX-2 signaling is positively up-regulated, the formation of cellular actin stress fibers (SFs) is disrupted allowing invasion associated with migration (Bos et al., 2004; Bravo-Cordero et al., 2012; Takai et al., 2013). SFs are known to play a pivotal role in the regulation of various cell functions, including cell adhesion and motility (Bravo-Cordero et al., 2012). Based on this interplay of COX-2/SFs, we sought to investigate whether GW501516 can abrogate the formation of the SFs (i.e., possibly via activation of PPARβ/δ and COX-2 signaling) in MDA-MB-231 cells. As shown in Fig. 4C, when compared to vehicle-treated control (left panel), SFs were disappeared in the cells treated with 10 nM GW501516 (middle panel). The GW501516-mediated interference with SFs was effectively reversed with CBDA exposures (right panel). These results indicate the presence of negative cross-talk between GW501516- and CBDA-mediated signaling (possibly on PPARβ/δ) in the MDA-MB-231 cells.
As shown in Fig. 1 and Fig. 4A, PPARβ/δ positively modulated COX-2 expression in MDA-MB-231 cells, which is consistent with other groups’ reports demonstrated in HepG2 and H358/H441 cells (Glinghammar et al., 2003; Genini et al., 2012). Although no PPRE has been identified in the COX-2 promoter, GW501516 induces COX-2 expression through increased activity of the COX-2 promoter (−327/+59) containing response elements for AP-1 and NF-κB (Glinghammar et al., 2003), suggesting that PPARβ/δ indirectly regulates COX-2 expression. It is being recognized that PPARβ/δ mediates its transcriptional activity via interaction with other transcriptional factors (Di-Poï et al., 2002). Because the AP-1 complex has been shown to be important for the regulation of COX-2 (Park et al., 2005), it is possible that PPARβ/δ indirectly modulates COX-2 expression via AP-1 in MDA-MB-231 cells. Given that PPARβ/δ interacts with AP-1 (possibly forming complex) on the promoter of COX-2, PPARβ/δ inverse agonists (GSK0660 and ST-247) might abrogate AP-1-mediated transcriptional activity. As shown in Fig. 5B, GSK0660 inhibited the transcriptional activity mediated by AP-1 in a concentration-dependent manner. The other inverse agonist ST-247 suppressed AP-1 activity as well (Fig. 5C), possibly supporting our above-mentioned. Since GSK0660/ST-247 induce recruitment of corepressors, SMRT (silencing mediator of retinoic acid and thyroid hormone receptors) and HDAC3 (histone deacetylases 3) (Naruhn et al., 2011). It should be noted that SMRT and HDACs have been shown to interact with not only PPARβ/δ but also AP-1, which result in suppression of PPARβ/δ- and AP-1-mediated transcriptional activities (Lee et al., 2000). Thus, it is suggested that AP-1-mediated transcriptional activity can be suppressed by the corepressors recruitment triggered by GSK0660 and ST-247 because of the interaction between PPARβ/δ and AP-1. In addition, CBDA has also the potential to recruit corepressors, similarly to GSK0660 and ST-247 (see Fig. 5A).
CBDA potentiates PPARβ/δ antagonist-mediated inhibition of COX-2 expression, and PPARβ/δ antagonists inhibited AP-1-mediated transcriptional activation. (A) Real-time RT-PCR analyses of COX-2 in MDA-MB-231 cells after 48 hr of treatment with 1 μM GSK0660 alone and 1 μM GSK0660 in combination with 5 μM CBDA. The data are expressed as the mean ± S.E. (n = 6) of the fold induction compared to 1 μM GSK0660-treated alone. Significant differences (by Student’s t-test) to1 μM GSK0660-treated alone are marked with an asterisk (*P < 0.05). (B and C) The effects of PPARβ/δ antagonists on AP-1-mediated transcriptional activity on MDA-MB-231 cells are shown. MDA-MB-231 cells were transiently transfected with a AP-1-luciferase reporter plasmid. After transfection, the cells were treated with vehicle (indicated as Ctl.) and GSK0660 (0.2, 1, and 5 μM) (B), or vehicle (indicated as Ctl.) and ST-247 (0.2, 1, and 5 μM) (C). After 24 hr, the cells were harvested and assayed for luciferase activity, and all the transfections were normalized for efficiency using the internal Renilla control plasmid. The data are expressed as the mean ± S.E. (n = 3) of the fold induction as compared to the vehicle-treated control. Significant differences (one-way ANOVA, followed by Dunnett’s post-hoc test) from the vehicle-treated control are marked with an asterisk (*P < 0.05). All data shown are representative of three experiments, and n (= 3 or 6) indicates the number of technical replicates in the experiments.
In this study, we have suggested PPARβ/δ as one of the key elements contributing to the down-regulation of COX-2 by CBDA. Although a positive correlation has been reported between aberrant COX-2 expression and enhanced malignant phenotype of breast cancers, for example, acceleration of metastasis, it is puzzling to consider how cancer cells positively regulate COX-2 expression. Accumulating experimental evidence suggests that the expression of COX-2 is coordinately regulated by at least three representative transcriptional factors, AP-1, NF-κB, and PPARβ/δ (Dong et al., 1997; Schmedtje et al., 1997; Glinghammar et al., 2003; Wang et al., 2014), although the biological effects of PPARβ/δ on COX-2 expression is confusing (Glinghammar et al., 2003; Genini et al., 2012; Peters et al., 2015). CBDA can reduce the AP-1-mediated transcriptional activity in MDA-MB-231 cells, which may result in the reduction of COX-2 (Suzuki et al., 2017; Takeda et al., 2017). Since in general, cancer cells over time often exhibit reduced responsiveness to anti-cancer therapy, there is a need to develop additional modalities for cancer treatment, including the development of “multiple action points” to down-regulate the expression of COX-2. Taken together with our previous findings, it is suggested that CBDA offers a chemotherapeutic potential for inhibiting the transcriptional activities of both AP-1 and PPARβ/δ, which when combined, may lead to the effective down-regulation of COX-2. However, to generalize our findings, further studies are needed.
This study was supported in part by a Grant-in-Aid for Scientific Research (C) [17K08402, (to S.T.)] from the Japan Society for the Promotion of Science (JSPS) KAKENHI.
Conflict of interestThe authors declare that there is no conflict of interest.
