Fatty Acids of CLA-Enriched Egg Yolks Can Induce Transcriptional Activation of Peroxisome Proliferator-Activated Receptors in MCF-7 Breast Cancer Cells

In our previous study, we showed that fatty acids from CLA-enriched egg yolks (EFA-CLA) reduced the proliferation of breast cancer cells; however, the molecular mechanisms of their action remain unknown. In the current study, we used MCF-7 breast cancer cell line to determine the effect of EFA-CLA, as potential ligands for peroxisome proliferator-activated receptors (PPARs), on identified in silico PPAR-responsive genes: BCAR3, TCF20, WT1, ZNF621, and THRB (transcript TRβ2). Our results showed that EFA-CLA act as PPAR ligands with agonistic activity for all PPAR isoforms, with the highest specificity towards PPARγ. In conclusion, we propose that EFA-CLA-mediated regulation of PPAR-responsive genes is most likely facilitated by cis9,trans11CLA isomer incorporated in egg yolk. Notably, EFA-CLA activated PPAR more efficiently than nonenriched FA as well as synthetic CLA isomers. We also propose that this regulation, at least in part, can be responsible for the observed reduction in the proliferation of MCF-7 cells treated with EFA-CLA.

It has been suggested that ligand-dependent activation of PPARs results in the inhibition of proliferation in some model cancer cell lines [5][6][7]. In particular, PPAR isoform was shown to reduce cancer cell proliferation as well as regulate cell differentiation, activate apoptosis, and inhibit angiogenesis [8][9][10]. Specifically, the administration of specific PPAR agonist resulted in cells arrest in G1 phase and inhibited proliferation [5,11]. However, available literature presents also contradicting results. In some studies, PPAR specific antagonist, T0070907, significantly reduced proliferation and migration of breast cancer cells [12,13].
PPARs act as transcription factors and regulate the expression of dependent genes by binding to their PPREs. A significant number of genes regulated by PPARs have been described; however, the list is not exhaustive and is constantly being updated as new results are being published from both experimental data and bioinformatics analyses of promoter regions and PPRE consensus sequences. In the current study, we applied those tools to identify in silico PPRE selected genes involved in cell cycle progression and proliferation. Next, we analyzed the effect of synthetic cis9,trans11CLA and trans10,cis12CLA isomers as well as a mixture of fatty acids extracted from CLA-enriched and nonenriched egg yolk on the expression of those genes. To the best of our knowledge, our study is the first to address the effect of CLA incorporated in fatty acids profile of the egg yolk; we expect that activity of CLA in such a "bioorganic" form may deviate from that of a synthetic form. The presence of other fatty acids in an egg yolk, which themselves can act as potential ligands for PPARs, may modulate the action of CLA; therefore, our data may be particularly important for the evaluation of CLA-enriched food products.

Production of CLA-Enriched Egg Yolks.
Production of CLAenriched egg yolks was performed in the National Research Institute of Animal Production in Krakow (Poland), as per the recommendations of the Local Animal Ethics Committee (approval number: 851/2011) as described previously [21]. Eggs were collected and stored at 4 ∘ C, and yolks were separated from albumen, homogenized, and frozen at −20 ∘ C. Samples were then lyophilized and again stored at −20 ∘ C until further analyses.

Extraction and Analysis of Fatty Acids Composition.
Lipids from control and CLA-enriched yolks were extracted by using modified Folch method [22] as described previously [23]. 10 mg of each lipid extract was subjected to saponification with 0.5 M KOH/methanol followed by methylation with 14% (v/v) BF3/methanol and extraction with hexane. Fatty acid methyl esters (FAME) were analyzed by GC/MS as described previously [23].  2.6. Plasmids. PPAR expression vectors were prepared using Gateway5 Cloning System (Thermo Fisher, USA). Briefly, PPARA (CR456547_1), PPARD (NM_006238.4), and PPARG (NM_015869.4) ORF sequences were synthesized, optimized for the expression in human cells, and cloned into the pDONR221 Entry Vectors (GeneArt, Thermo Fisher, USA). Subsequently, the ORF inserts were transferred into pcDNA6.2/N-EmGFP-DEST Destination Vectors (Thermo Fisher, USA) under the CMV promoter control via Clonase II Recombination Reaction.

Cell Transfection with PPAR Encoding Plasmids.
Cell lines with PPARA, PPARD, and PPARG overexpression were obtained via transient transfections with pcDNA6.2/N-EmGFP-DEST vectors containing respective human PPAR ORF. MCF-7 cells were seeded on 12-well plates, at 1 × 10 5 cells per well. 24 h after seeding, cells were transiently transfected with 1.5 g of PPAR encoding plasmids using Lipofectamine (Thermo Fisher Scientific, MA, USA) in OPTI-MEM medium (Thermo Fisher Scientific, MA, USA). 24 h after transfection, the growth medium was replaced with selective MEM medium with 10% FBS and 5.0 g/mL blasticidin (BioShop, Canada). Transfected cells were cultured until confluency.
Real-time PCR and western blot method were performed to confirm the presence of PPAR plasmids after transfection ( Figure S1 and Table S2, Supplementary Material available online at https://doi.org/10.1155/2017/2865283).
2.9. Dual-Luciferase Assay. 24 hours after transfection with PPRE plasmid, the medium was again replaced with MEM medium containing 10% FBS and appropriate experimental treatment as described above. 24 hours after treatment, cells were harvested for isolation of protein luciferase.

In Silico Selection and Experimental Confirmation of PPAR-Dependent Genes (PPAR-Responsive mRNAs)
. PPARresponsive genes were selected in silico by searching for peroxisome proliferator hormone response elements (PPREs, AGGTCANAGGTCA) within promoters and/or 5cis-regulatory regions of the promoters of genes involved in cell cycle progression and proliferation. This search was performed with NCBI Gene and Blast tools.
Experimentally, 24 hours after transfection with respective PPAR plasmids, the medium was replaced with MEM medium containing 10% FBS and appropriate experimental treatment as described above. 48 hours after treatment, cells were harvested for mRNA isolation and RT-qPCR.

RNA Isolation, cDNA Synthesis, and RT-qPCR Analysis.
Total RNA was isolated from the cells using RNA isolation kit for cell cultures (A&A Biotechnology, Poland). Reverse transcription was performed on 1 g of total RNA using Maxima First-Strand cDNA Synthesis kit for RT-qPCR (Thermo Scientific, MA, USA). Quantitative verification of genes was performed using CFX96 Touch6 Real-Time PCR Detection System instrument (Bio-Rad, CA, USA) and SYBR Green Precision Melt Supermix kit (Bio-Rad, CA, USA). Conditions of individual PCR reactions were optimized for given pair of oligonucleotide primers (Table S1, Supplementary Material). Basic conditions were as follows: 95 ∘ C for 10 min, 45 PCR cycles at 95 ∘ C, 15 s; 59 ∘ C, 15 s; 72 ∘ C, 15 s, followed by melting curve analysis (65-97 ∘ C with 0.11 ∘ C ramp rate and 5 acquisitions per 1 ∘ C). Results were normalized using at least two reference genes (GAPDH, HPRT1, ACTB, or HSP90AB1) and were calculated using the 2 −ΔΔC T method [24].

Protein Isolation and Western Blot Analysis.
Cell lysis was carried out using Cell Lysis Buffer (Cell Signaling Technology, MA, USA) as per the manufacturer's protocol. Total protein quantification was performed using Pierce BCA6 Protein Assay Kit (Thermo Fisher Scientific, MA, USA).
To remove the antibodies from the membrane, we used western blot stripping buffer (Thermo Scientific, MA, USA).

Statistical Analysis.
All experiments were performed at least three independent times and measured in triplicate. Shapiro-Wilk's test was applied to assess normality of distribution. An independent samples -test was applied to compare unpaired means between two groups. < 0.05 was considered statistically significant. All analyses were performed using Statistica ver.12 (StatSoft, Tulsa, OK, USA).  at 72 h. The reductive effect of cis9,trans11CLA isomer was less evident and statistically significant only after 72 h (overall reduction in viability by 15%).

Selective Effect of FA on Transcriptional Activity of
PPARs. The selective effects of the studied FA as potential PPAR ligands are shown in Figures 4(a)-4(d). EFA-CLA was determined to be the most specific for PPAR (3.5-fold increase in activity, < 0.001; Figure 4(a)). EFA extract acted as an antagonist towards both PPAR and PPAR , while it exhibited only negligible agonist activity on PPAR (1.44fold increase in activity, > 0.05, Figure 4(b)). cis9,trans11 isomer showed agonist properties towards all PPAR isoforms, with the strongest effect on PPAR (2.37-fold increase in activity, < 0.005; Figure 4(c)). trans10,cis12 isomer showed no significant effect on transactivation of both PPAR and PPAR ( > 0.05, Figure 4(d)), while it showed an antagonist activity towards PPAR ( < 0.01, Figure 4(d)).

Prediction of Potential PPRE-Dependent Genes In Silico.
The prediction of potential PPRE-responsive genes was performed in silico. NCBI database was searched for the presence of specific PPRE (peroxisome proliferator response element) consensus sequences (AGGTCAAAGGTCA, AGGTCA-GAGGTCA, AGGTCACAGGTCA, or AGGTCATAGGTC-A) in the 5 region of genes linked to oncogenesis and cell cycle ( Figure 5). Seven genes were identified: BCAR3, LZTS, SLC5A1, TCF20, WT1, ZNF621, and THRB (transcript TR 2), potentially regulated by PPARs (      gene was identified by the presence of the PPRE consensus sequence in a region of the alternative promoter for TR 2 isoform (intron between the 4th and 5th exon). Among identified potential PPRE-dependent genes, few were selected for further experimental analyses, including TCF20, WT1 ZNF621, and THRB.

Effects of EFA-CLA on the Expression of PPAR-Regulated
Genes. Expression of selected PPAR-responsive genes (containing PPRE) has been tested in response to various experimental fatty acids as potential ligands for PPARA, PPARD, or PPARG. Our results showed both agonist and antagonist effects of studied experimental FA.
EFA-CLA added to the PPAR -overexpressing cells elevated the expression of TCF-20 over 3.2-fold and ZNF621 over 3.1-fold, while decreasing the expression of WT1 gene 1.2fold. However, the latest may be explained, at least in part, from the fact that WT1 gene is cotranscribed with interfering long, noncoding antisense RNA (WT1-AS) from the same bidirectional promoter. For cells overexpressing PPAR , EFA-CLA treatment resulted in the elevated expression of TCF-20 over 3-fold, while for the PPAR -overexpressing cells ZNF621 gene was upregulated 1.8-fold.
The strongest enhancement of TCF-20 expression (over 13-fold) was observed in PPAR -and PPAR -overexpressing cells after treatment with trans10,cis12CLA. Interestingly, the

Discussion
Chicken egg enriched with conjugated linoleic acid (CLA) via feed modification meets the criteria of the functional food product. Based on Roberfroid's [25] classification, CLAenriched egg can be considered as a conventional food product that is intended to be consumed as a part of a normal diet but is modified to contain biologically active substances, that is, CLA isomers. It has been shown to have a beneficial effect on physiological functions of the human body, in a way that goes beyond its nutritional value, specifically by lowering the risk of developing atherosclerosis [26]. Our previous studies showed additional beneficial properties of CLAenriched eggs in reducing proliferation of breast cancer and melanoma cells [23,27]. The current manuscript supports those findings as our new results showed that fatty acids extract from CLA-enriched egg yolks (EFA-CLA) reduced the viability of MCF-7 breast cancer cell line ( Figure 1). However, the molecular mechanism is not fully understood. Comparison of the effect on cancer cell proliferation between extracts from CLA-enriched and nonenriched egg yolks could lead to the conclusion that it is simply the result of the presence of CLA isomers incorporated in the egg yolk lipids. Available literature would support such a hypothesis as numerous studies showed an inhibitory effect, especially for cis9,trans11CLA isomer, on tumor cells [28][29][30][31][32]. Indeed, our analysis of FA profile of CLA-enriched egg yolk showed that cis9,trans11CLA was incorporated more efficiently (3 : 1 ratio) than trans10,cis12 isomer [21] and therefore could predominate in EFA-CLA. Interestingly, comparison of the effect of synthetic CLA isomers with CLA-EFA from egg yolk showed the advantage of the latter in reducing cancer cell viability (Figure 1). The analysis of fatty acids profiles between enriched and nonenriched egg yolks revealed not only CLA incorporation but also unexpected, significant change in SFA/MUFA ratio, specifically an increase in total SFA concentration at the expense of MUFA. Thus, a question 8 PPAR Research arises of whether it is an individual or combined effect of CLA and modified SFA/MUFA ratio in enriched egg yolks on MCF-7 cell line [23]. We observed that results of CLA-EFA are most likely achieved by the effect of both: incorporated CLA isomers and other fatty acids in eggs modified organically through hens' diet [23]; however, this issue requires further research.
It has been shown that PPAR agonists have different properties for individual PPAR isoforms, with different absorption and distinctive gene expression profiles. To our knowledge, this is the first study focused on the effect of FA from CLAenriched egg yolks on transcriptional activation of PPARs (PPAR , PPAR , and PPAR ). All experiments included as controls synthetic CLA isomers as well as standard agonists and antagonists of different PPARs. Our results showed that EFA-CLA extract exhibits the properties of agonists for all PPAR isoforms (Figures 3(a)-3(c)); however, those properties seem to be most selective towards PPAR (Figure 4). Interestingly, PPAR has been associated with the greatest impact on cancer cell proliferation, survival, and differentiation, and its ligands are associated with anticancer properties [33,34]. In addition, as observed for EFA-CLA, transactivation of PPAR receptors is more effective compared to fatty acids extracted from a nonenriched egg yolk (EFA) (Figures 3(a)-3(c)). Since cis9,trans11CLA isomer showed PPAR agonist activity (Figures 3(a)-3(c)) and since this isomer was 3-fold more efficiently incorporated into egg yolks than trans10,cis12CLA [23], it could be hypothesized that cis9,trans11CLA plays a significant role in EFA-CLA-mediated activation of PPARs.
The effect of synthetic CLA isomers provided us with important information about their specificity. While cis9,trans11 isomer acted as a PPAR agonist (Figures  3(a)-3(c)), the antagonist effect was observed for trans10,cis12 isomer, specifically on PPAR (Figure 3(c)). Available literature is consistent with our results. cis9,trans11 isomer has been reported to inhibit cell growth [15,16] showing antitumor properties [17][18][19][20]. It has been found as well that the presence of trans10,cis12 isomer may abrogate the antiproliferative activity of cis9,trans11 [18] and even inhibit the activity of synthetic PPAR agonists [15]. Thus, it is even more interesting that our results showed more efficient reduction in cancer cells proliferation for EFA-CLA treatment than using a pure synthetic cis9,trans11CLA isomer that may suggest other factors including modified SFA/MUFA ratio in enriched egg yolks [23], supporting antiproliferative action of cis9,trans11CLA isomer.
PPARs act as transcription factors and regulate the expression of dependent genes by binding to their PPREs. Available literature gives a number of genes regulated by PPARs; the ligand-dependent transcription factors [35] and the expression of those genes can be both inhibited or activated depending on the ligand, suggesting selectivity [36]. CLA isomers have been found to act as PPAR ligands and shown to be involved in the inhibition of transcription of genes including TNF [37], NFKB1 [38], and NR1I3 [39] as well as transactivation: TGFB1 [40], BRCA1 [41], PTEN [42], p21/WAF1/CDKN1A [43], CEBPA [44], ABCB4 [45], and AOX [46]. Although a significant number of genes regulated by PPARs have been described, the list is not exhaustive and is constantly updated as new results are being published from both experimental data and bioinformatics analyses of promoter regions containing PPRE consensus sequences (AGGTCANAGGTCA) ( Figure 5).
In the current study, we applied bioinformatic tools to find genes with PPRE and analyze the effect of CLA on the expression of these genes. To our knowledge, we proposed several new genes that could be potentially PPARregulated: BCAR3, LZTS, SLC5A, TCF20, WT1, ZNF621, and THRB (transcript TR 2) ( Table 2). Since preliminary data showed that some of them were strongly regulated by PPARs, we studied the expression of TCF20, WT1, THRB (TR 2), and ZNF621 genes in the context of various PPAR ligands, including EFA-CLA.
First one TCF20 can act as a phosphoserine-specific repressor of estrogen receptors (ER) in estrogen-dependent tumors [47]. MCF-7 human breast carcinoma cell line is estrogen receptor (ER) positive; thus, the expression of TCF20 should inhibit ER and consequently impair the viability of the tumor cells. Our results confirm these assumptions, showing elevated TCF20 mRNA level in cells treated with EFA-CLA. This effect was much stronger than for EFA (Table 2). Interestingly, the most pronounced effect was found for trans10,cis12CLA isomer (Table 2), which may explain its advantages over the cis9,trans11CLA in reducing the viability of MCF-7 (positively correlates with its effect on the reduction in cell viability) (Figure 1). In contrast to Pariza et al. [18], this result also suggests that trans10,cis12CLA isomer could support antiproliferative action of cis9,trans11CLA in EFA-CLA via transcription-enhancing effects on TCF20.
Available literature addresses the relationship between receptors encoded by PPAR and THRB genes [48][49][50]. THRB encodes three isoforms of human thyroid hormone receptor: TR 1 and tissue-specific TR 2 and TR 4, which are thought to be engaged in cell cycle control and metabolism [51]. Recently, THRB has been studied as a tumor suppressor [52]. Although TR 1 isoform has been found to play a role in the competitive inhibition of the PPAR transactivation [53], there is limited information on the relationships between TR 2 and PPAR receptors. TR and PPAR receptors are linked by the same obligatory coreceptor, retinoid X receptor (RXR), that binds to their heterodimeric partners before binding to DNA. Although RXR plays a central role in regulating the activity of a number of nuclear hormone receptors including TR and PPARs by acting as a heterodimeric partner, this receptor is known to be constitutively expressed in cells [53]; therefore, focusing on PPARs, we do not show the expression of RXR in this paper. Nevertheless, it has been reported that TR and PPAR receptors can compete for binding to RXRs in the nucleus [54]. Since we have found PPRE within the sequence of TR 2-specific promoter, located in intron IV of THRB gene, the bidirectional regulation of TR 2 and PPARs is thought to be more complex. Results presented in the current manuscript indicated enhanced transactivation of TR 2 by all PPARs isoforms in response to the treatment with experimental FA ( Table 2) that may be evidence of the functional activity of the TR 2-specific PPRE; however, this needs further studies. The most significant effect was measured for the synthetic CLA isomers, especially trans10,cis12 (Table 2). Taken together, our findings showed that transcription levels of TR 2 are elevated by PPARs and their agonists. Simultaneously, TR 1 isoform has been shown to compete with PPAR for access to the RXR coreceptor or for PPRE binding sites in promoter regions of regulated genes [50] that could suggest TR 1-mediated inhibitory role in expression of TR 2 isoform and possibly other PPARresponsive genes.
WT1 gene, as a transcription factor, directly or indirectly interacts with a number of genes involved in cell cycle and neoplasia, including HIF1A, AREG, SRY, NROB1, SOX9, IGF2, MDM4, BRCA1, TP53, and SP1 (NCBI Gene). Available literature suggests an oncogenic nature of WT1 and has shown its overexpression in various tumors and tumor cell lines, especially in breast cancer cells and melanoma [55,56]. In addition, decreased levels of WT1 gene expression correlated with reduced cell proliferation in both melanoma and breast cancer cells [57,58]. WT1 has also been linked with malignant transformation in breast cancer, and its overexpression associated with reduced susceptibility to drug treatment. Indeed, it has been shown for estrogen-dependent lines that WT1 positively regulates the expression of EGFR and HER2 [55], contributing to the resistance to hormone therapy [59,60]. In melanoma, in vitro WT1 silencing resulted in decreased cell proliferation, followed by apoptosis induction with caspase-3 activation [61], while in vivo it reduced the melanoma metastatic to lungs [56]. On the other hand, some studies indicate that pharmacologic activation of PPAR by its agonists (GW0742 and GW501516) inhibited proliferation of the murine melanoma cells, accompanied by downregulation of WT1 [62]. It was suggested that PPAR can act via the PPRE in the WT1 promoter and directly suppress its activity; however, our results do not support this hypothesis. Although the use of a known PPAR agonist, GW0742, resulted in PPAR activation (Figure 3(b)), no decrease in the expression of WT1 was measured (Table 1). This contradiction may result from the use of different biological materials suggesting cell/tissue-specific regulation and/or association/dissociation of different corepressors or coactivators to transcription machinery. Interestingly, we showed that treatment with EFA-CLA and cis9,trans11 reduced expression of WT1 via the activation of PPAR (Table 2). A similar effect was observed for other experimental FA (Table 2) suggesting that various PPAR ligands may exert different effects in different cells; however, this hypothesis should be studied.

Conclusion
In conclusion, potential tumor suppressor properties of PPAR receptors make their ligands attractive candidates for the development of new chemopreventive, anticancer agents. Here, we show for the first time a functional food product, CLA-enriched egg (EFA-CLA), that is more effective in reducing of MCF-7 cancer cells proliferation than synthetic CLA isomers. This EFA-CLA effect could result from the high content of cis9,trans11 isomer, altered SFA/MUFA ratio in enriched egg yolks, and/or supportive role of trans10,cis11 isomer in regulation of specific genes. Our results indicate that EFA-CLA can act as a ligand of PPARs, showing an agonist activity, specifically towards the PPAR isoform. Control, synthetic cis9,trans11 isomer of CLA exerted an agonist effect on all PPAR receptors, while trans10,cis12 showed no effects or even acted as an antagonist of PPAR . However, this isomer was able to regulate some specific genes containing PPREs such as TCF20 involved in cell cycle arrest. Simultaneously, cis9,trans11 isomer upregulated THRB suppressor and downregulated WT1 oncogene showing a small part of a PPAR action that in case of EFA-CLA leads to the observed reduction in proliferation of the breast cancer cells. It seems therefore that CLA-enriched eggs could be considered as food products with anticancer potential.

Conflicts of Interest
The authors report no financial or other conflicts of interest relevant to the subject of this article.