PTGDR expression is upregulated through retinoic acid receptors (RAR) mechanism in allergy

Functional studies suggest that promoter polymorphisms of the Prostaglandin D Receptor (PTGDR) gene can be involved in asthma. All-trans Retinoic acid (ATRA) has also been linked to allergic diseases. We have previously described the PTGDR promoter activation mediated by ATRA through response elements (RARE) at position -549T> C. In this study we aimed to analyze the effect of retinoic acid (RA) on the expression of PTGDR, the production of cytokines as well as to evaluate the binding of RA receptors to RA-Response Elements (RARE) sequences. A549 cells were transfected with vectors carrying different PTGDR haplotypes and treated with all-Trans Retinoic Acid (ATRA). PTGDR expression was measured by qPCR. Chromatin Immunoprecipitation assays (ChIP) were performed in ATRA stimulated KU812 cells and in PBMCs of patients carrying CTCT, CCCC or CCCT haplotypes. In addition, a broad panel of cytokines was analyzed by cytometric bead assay in A549 cells. The expression of PTGDR increased in A549 cells transfected with PTGDR-variants. The CCCC haplotype showed a significantly higher expression compared with CTCT. However, we found that RA up-regulated PTGDR expression through RARα mainly in the CTCT variant. Experiments on PBMCs from allergic patients carrying the -549T and -549C variant of the PTGDR promoter after ATRA and RAR antagonist administration confirmed the modulation of PTGDR by ATRA. The cytokine analysis showed that IL4 and IL6 levels were significantly increased in A549 cells transfected with PTGDR. In addition, ATRA treatment decreased the levels of IL4, IL6 and TNFα in A549 cells, whereas it increased IL4 and TNFα levels in PTGDR-transfected cells. We observed genetic differences in the regulation of PTGDR by ATRA that could contribute to the phenotypic differences observed in allergic patients. Our findings showed that RAR modulation by PTGDR might have an impact on Th2 responses, suggesting that RAR could be a potential therapeutic target in allergic inflammation.


Introduction
Asthma is a chronic inflammatory disease affecting more than 358 million people [1]. The inflammatory response in asthma involves the participation of the respiratory epithelium, the innate immune system and the adaptive immunity [2,3]. Prostaglandin D2 (PGD2) is the major cyclooxygenase (COX) metabolite of arachidonic acid produced in response to allergens and has been proposed as a mast cell activation marker [4]. There are two types of transmembrane receptors of PGD2, the D prostanoid receptor, known as PTGDR, DP or DP1, and the chemoattractant receptor-homologous expressed on Th2 (CRTH2) or DP2 [4,5]. Several PTGDR polymorphisms have been previously reported [4,[6][7][8]. Promoter variants have shown a different binding of transcription factors that controls the expression of PTGDR, and this could be related to the development of asthma susceptibility [4,6].
Epidemiological studies have related vitamin A levels [through its active metabolite, the alltrans (AT)-retinoid acid (RA)] with the prevalence of allergic diseases [9]. Some studies have associated vitamin A deficiency with an increase frequency of atopy [9,10], although others have reported that vitamin A supplementation associated with increased airway hyperresponsiveness [11,12]. Dawson et al reported that ATRA promotes the synthesis of the human type 2 cytokines IL4, IL5 and IL13 while decreasing IFNγ and, TNFα expression, and IL12 synthesis in activated human T-cells [13]. RA has previously been associated with the prostaglandin pathway. ATRA has been shown to induce an increase of the expression of cyclooxygenase 2 (COX-2), one of the cyclooxygenases involved on PGD2 synthesis, suggesting ATRA as a main regulator of COX-2 expression [14]. In addition, the synthesis of PGD2 is mediated by the lipocalin-type prostaglandin D synthase (LPGDS), which is a retinoid transporter able to bind ATRA [9]. RA activates nuclear retinoic acid receptors (RARα, β, γ), which dimerize with retinoid X receptors (RXRα, β, γ) and function as ligand-dependent transcriptional regulators by binding to RA response elements (RARE) on target genes [15,16]. We have previously shown that RA activates the PTGDR promoter; in addition, we have identified RAREs in the promoter region, and demonstrated that some transcription factor motifs were affected by genetic variants [17]. Considering our previous results, the objective of this study was to deeper explore the regulatory mechanisms involved in the expression of PTGDR by ATRA and its effect on cytokine production. In addition, we aimed to evaluate the role of different factors involved in this regulation, like promoter polymorphisms and the different nuclear retinoic acid receptors.

Subjects
The study included 6 adult patients (3 males and 3 females) sensitized to pollen that had been evaluated and diagnosed by allergists. All of them signed an informed written consent. The study was approved by the Ethics Committee of Clinical Investigation of the hospital (PI 1/07/ 2013).

Isolation of peripheral blood cells
Peripheral Blood Mononuclear Cells (PBMC) were obtained by venous puncture on Lithium Heparin vacuum tubes. Blood was centrifuged on Ficoll-Paque (GE-Healthcare Life Science, Chicago, IL, USA) density gradients. Interphase cells were washed three times with Dulbecco's phosphate-buffered saline (PBS, GIBCO-Thermo Fisher Scientific, Whaltham, MA, USA).

Cell culture
Isolated PBMCs were resuspended at a concentration of 10x10 6 cells/ml in complete medium, consisting of RPMI 1640 supplemented with 2mM L-glutamine, 1% penicillin-streptomycin (P/S) (GIBCO-Thermo Fisher Scientific, Waltham, MA, USA) and 10% autologous human serum from each patient and cultured for 48h in an incubator at 37˚C and 5% CO2.

Expression assays
Total RNA isolation, RT-PCR and qPCR were performed as described in detail in the Method Section of the S1 Appendix. Briefly, mRNA was retro-transcribed, and relative qPCR was performed using SYBR-Green-I-Master in a LightCycler480 (Roche Applied Science, Mannheim, Germany). Calculations were made by the comparative Ct method [19]. All procedures were performed following MIQE guidelines [20].

Chromatin Immunoprecipitation (ChIP) Assay
ChIP was performed using EZ-Magna ChIP A kit (Millipore, Burlington, MA, USA) on KU812 cells and human PBMCs isolated from subjects bearing the PTGDR haplotypes [17]. Immunoprecipitated DNA using RARα and RARβ antibodies (Santa Cruz Biotechnology, Dallas, TX, USA) and input samples were subjected to qPCR using primers spanning the -549C/T PTGDR-promoter. Additional details are provided in the Method Section of the S1 Appendix.

Cytokine analysis
Cell supernatants were harvested, centrifuged (400g, 10 min), and stored at -80˚C. Cytokine levels were determined using the Bio-Plex Pro Human Cytokine standard 27-plex, Group I (Bio-Rad, Hercules, CA, USA). Fluorescence was measured by Luminex 100IS (Bio-Rad, Hercules, CA, USA) with Bio-Plex High-throughput fluidics system, powered by the Luminex X-Map Technology (Luminex, Austin, TX, USA). Data were acquired and processed by the Bioplex Manager Software version 4.1.1 (Bio-Rad, Hercules, CA, USA). A preliminary assay showed that 48h was the optimal time for the determination of most cytokines. Each sample was analyzed in triplicate. In addition, expression analysis of cytokine mRNAs after ATRA treatment was performed in A549 cells as previously described.

Statistical analysis
Data analysis was performed using pairwise comparison by analysis of variance (ANOVA), unpaired one-sample t-test, Kruskal Wallis and Pearson's correlation coefficient using the SPSS Software (version 23) (IBM, Armonk, NY, USA). Data were representative of at least three independent experiments. A P<0.05 was considered significant.

The expression of PTGDR increased significantly with retinoic acid and is dependent on promoter variants
Transfection assays with the above-mentioned construction vectors were performed. Culture cells were treated with 1μΜ ATRA or DMSO and collected after 24 and 48h. All haplotypic variants exhibited higher PTGDR expression after ATRA-treatment at 24h compared with control cells (P<0.001) (Fig 2).

ATRA stimulation promoted binding of RARα and RARβ to the PTGDR promoter
To interrogate whether RARs directly bound to RARE elements in ATRA-stimulated KU812 cells, a ChIP assay was carried out. The expression of RARs (α, β, γ) and RXRs (α, β, γ) on  Accordingly, ChIP assays were performed with RAR-antibodies.
ATRA stimulation promoted the binding of RARα to the proximal region of the PTGDR promoter ( Fig 3A). No binding of RAR was detected when the distal PTGDR intron region ChIP analysis of RAR isoforms on the PTGDR promoter. A. KU812 cells were exposed to 1μM ATRA or DMSO for 24h. ChIP assays were performed with antibodies against RARα and RARβ. Normal Rabbit IgG was used as the negative control for immunoprecipitation. Immunoprecipitated DNA was amplified by qPCR using primers spanning the polymorphic variant -549T>C in the proximal region of the PTGDR promoter or using the distal PTGDR intron region (negative primers). Results are presented as the fold enrichment of chromatin DNA precipitated by the specific antibody compared with normal rabbit IgG. ATRA data were normalized versus DMSO. Values are means from three independent experiments. �� P<0.01 for anti-RARα immunoprecipitated chromatin in the -549T promoter region versus the distal region of PTGDR (negative control), n = 3. B. PBMCs isolated from subject bearing the CTCT, CCCC or TCCC-PTGDR haplotype variants were subject to ChIP assays with anti-RARα and RARβ antibodies as previously described, n = 1.

Genetic variants determined the occupancy of PTGDR RAREs by RARs α and β
To further determine if promoter haplotype variants influence the binding of RARα and RARβ to a RARE motif, a ChIP assay using PBMCs from individuals bearing the CTCT, CCCC and the CCCT haplotype variants was performed (Fig 3B). An increase in immunoprecipitated DNA in the CTCT compared with the CCCC and CCCT haplotype was detected when the -549T was present.

RA up regulated the gene expression of PTGDR through RAR in the KU812 cell line
In the KU812 cell line, the mRNA expression of PTGDR was upregulated by ATRA in a timedependent manner. In the absence of RA, we could not have observed PTGDR expression in this cell line. However, it was strongly induced 48h after ATRA stimulation. We could barely detect expression after 24h (Fig 4A). The obtained data were: DMSO-24h: 1±0; ATRA-24h: 1.04±0.33; DMSO-48h: 1.0±0.0; ATRA-48h: 141.92±27.15 (P<0.05 for ATRA 48h compared with DMSO 48h).

ATRA-treatment modified the release of cytokines in A549 cells
After ATRA-treatment, a decrease in the concentration of IL4, IL6, Eotaxin, FGFβ, IP10, MIP1α, MIP1β, RANTES and TNFα was detected in control cells compared to DMSO. On the other hand, IL8, IL10, IL12, IL13, IFNγ, and VEGF were increased after ATRA-induction in control cells ( Table 2).

The effect of ATRA treatment on cytokine release is mediated by PTGDR
We detected differences in some of the secreted cytokines in PTGDR transfected cells compared with control cells. An increase in most of the above-mentioned cytokines was detected. A significant increment in IL4, Eotaxin, MIP1α, MIP1β, and TNFα was observed in contrast to a decrease in IL8 (Table 2). No significant differences were observed between the CTCTand CCCC-PTGDR-variants.
These results were confirmed by qPCR, mRNA expression in A549 transfected cells after RA-treatment showed that IL4 mRNA levels were elevated too (S2 Fig). This data correlated with cytokine measurements (P<0.01).

PTGDR is stimulated by ATRA in PBMCs carrying the -549T variant
We performed an in vitro ATRA stimulation in PBMC cultures of two allergic patients who carried the -549T variant in the PTGDR promoter, and one allergic patient who carried the -549C. PBMCs were pre-treated with RARα antagonist 1h before the cells were stimulated with 1μM ATRA or DMSO. Cells were collected 48h after and RNA was isolated. The relative expression of PTGDR was increased in the two patients carrying the -549T variant, and this increment reverted by the RARα antagonist ( � P<0.05, for ATRA compared with DMSO, Kruskal Wallis test). Conversely, the -549C patient did not show any increment of PTGDR after ATRA treatment. (Fig 5).

In PBMCs carrying the -549T variant the production of Th2 cytokines was stimulated by ATRA
We analyzed the in vitro ATRA induced cytokine profile in PBMC cultures of two allergic patients who carried the -549T and -549C variants, respectively, as described above. mRNA of IL4, IL6, IL8 and IL13 were measured. The patient -549T showed increased levels of IL6, IL8 and IL13 after treatment with ATRA. This increase was attenuated when pre-treated with the antagonist, thus showing a specific effect of ATRA. On the contrary, IL4 levels were lower than DMSO. However, in the patient carrying the -549C variant ATRA treatment reduced IL4 and

Discussion
Epidemiological studies have related vitamin A with allergy by affecting the Th1-Th2 balance [9,14]. However, the role of RA in asthma and allergy is still controversial, [11,12,14]. We have previously reported that RA activated the PTGDR promoter, identifying RAREs in the promoter region and showing that transcription motifs were affected by genetic variants [17]. In the present study, we report for the first time the molecular mechanism of regulation of PTGDR expression by RA. A significant increase in the PTGDR expression was detected in transfected cells with different haplotypic variants. CCCC showed the highest PTGDR expression, both at 24 and 48h. In previous reports, this haplotypic variant was associated with higher promoter activity and it was also associated to asthma [8,17,18]. Interestingly, the wt-CTCT-variant, which has been associated with lower expression, was more frequent in non-allergic controls than in allergic patients [4,6]. These data are consistent with our previous studies in which the higher expression of PTGDR corresponded to CCCC-and CCCT-haplotypes [18] that were mainly detected in allergic patients [21].
Noteworthy, we observed an increase in the PTGDR expression of all variants after ATRAtreatment, although the greatest increase was shown with the CTCT-variant. In previous studies we have reported that the promoter proximal region showed the greatest increase in luciferase activity in response to ATRA [17]. Sequence analyses revealed a RARE motif in the CTCTvariant, which exhibited the highest level of PTGDR expression. In this sense, we have reported a potential extra RARE motif located in the -549T>C promoter position of PTGDR in the CTCT-variant [17]. This extra RARE could explain the highest sensitivity of CTCT to ATRA.
We have also observed that PTGDR mRNA is controlled by genetic variants located within a binding site for RXR/RAR receptors. Differential allelic occupancy at -549 position of RARE determines modifications in binding, which could explain the differential regulation. Therefore, we used ChIP experiments to examine the enrichment of RARs in the PTGDR promoter region. ATRA stimulation significantly promoted the binding of RARα to the proximal region of PTGDR. Using anti-RARα we detected the biggest enrichment of immunoprecipitated DNA-protein in the sequence carrying the -549T (wt) in the KU812 cell line. In addition, we detected an increase in immunoprecipitated DNA in the PBMCs bearing the CTCT haplotype compared to those bearing the CCCC and TCCC haplotypes. The -549T>C SNP is located within the binding site for RAR transcription factors. RA increases PTGDR transcription and different SNPs can modify its binding through RARα and RARβ isoforms. We hypothesize that these SNPs could mediate the RA response and associate to allergic diseases.
We have also showed an upregulation of PTGDR after ATRA-treatment in KU812 cells. In order to explain our hypothesis of the activation of RARs by ATRA, the cells were treated with RARα and RARβ antagonists, which attenuated the response, thus confirming that the regulation of the expression of PTGDR by ATRA occurred through RARs. These results were confirmed in PBMCs from allergic patients who carried the -549T variant, in which the PTGDR expression levels were elevated after treatment with ATRA and attenuated in the presence of the antagonist. Interestingly, these responses were not observed in the patient carrying the -549C variant.
A better understanding of the influence of PTGDR and ATRA on cytokine production can help to unravel molecular bases of allergic diseases, providing the field for better clinical interventions. In this sense, we have evaluated the effect of PTGDR-variants on the production of cytokines in A549 cells. Thus, significant increase in the Th2 cytokine IL4, and the proinflammatory cytokines IL6, IL8 and TNFα, was detected in pCTCT-PTGDR transfected cells compared to un-transfected cells. However, in pCCCC-PTGDR transfected cells, only significant increases in IL4 and IL6 were detected. Robinson et al., observed an increase in IL4 mRNA in the bronchi of patients with atopic asthma [22]. By transcriptomic massive sequencing, we have previously reported an increased expression in IL4RA mRNA in B-cells of patients with allergic asthma [23]. IL4R up-regulation facilitates IL4 signaling, associated to IgE class-switch recombination [24]. It has also been reported that PGD2 increases the Th2 polarization of naïve Th cells, as evidenced by an increase of IL4 and a decrease of IFN-γ [25]. Our data corroborate the activation of IL4 in presence of PTGDR, which could have implications in allergic diseases.
IL6 has been involved in the synthesis of PGE2 [26], it is increased in asthmatic patients, and has been detected in the bronchoalveolar lavage fluid (BALF) of severe asthmatic patients [27]. IL6 and IL8 are pro-inflammatory mediators that have been detected in culture supernatant of A549 cells after epithelial-mesenchymal transition induction in A549 [28]. Hirano et al reported that IL8 production induced by TNF-alpha was regulated by the prostanoid DP receptor [29]. In addition, IL4, IL8 and TNFα have been related with the pulmonary function in obstructive airway diseases and could be potential markers of asthma [30].
We have previously reported that RA stimulates the promoter activity driven by the 5'flanking region of PTGDR [17], which suggests that ATRA regulation occurs at a transcriptional level. In the present work, all variants exhibited higher PTGDR expression after ATRA treatment comparing with DMSO, corroborating previous results [17]. To interrogate the effect of RA regulation on PTGDR in inflammation, we analyzed cytokine levels in the supernatant of transfected cell after ATRA treatment. In control cells, we observed a decrease in the concentration of IL4 and IL6 and an increase on IL8 after treatment with ATRA. In PTGDR transfected cells ATRA also induced a significant decrease in IL6 and IL8, but there was a significant increase in IL4. These differences in the cytokine production between controls and transfected cells were statistically significant and seemed to be dependent on the expression of PTGDR. Nevertheless, we have not observed significant differences between CTCT and CCCC variants regarding the cytokine profile production after ATRA treatment. RA has been described as a key regulator of TGFβ-dependent immune responses and inhibited IL6 driven induction of Treg cell differentiation [31]. In addition, several studies have reported an antiinflammatory activity of RA [32,33]. Furthermore, Sheffel et al, reported that ATRA significantly inhibited IL6 secretion in human B-cells stimulated with anti-CD40 plus IL4 and that the inhibition of IgE by ATRA depended on the inhibition of IL6 [34]. Babina et al. reported that ATRA increased TNFα and IL8 in mast cells while there was no impact on IL6 [35]. These data support our results in IL6 and IL8 levels. However, it is noteworthy that IL8 levels decreased in PTGDR-transfected cells after ATRA treatment. We also observed an increase in TNFα in transfected cells after ATRA treatment. Thus, the activation of TNFα seemed to be regulated by PTGDR, which is in line with reports from Hirano et al, who showed that the PGD2 receptor up-regulated the cytokine production by TNFα in THP-1 cells. Dawson et al. reported that RARα mediates human T-cell activation and Th2 production [36].
There are some limitations in our study. The use of different cell types, as lung epithelial cells, basophils, and PBMCs, can difficult the interpretation of the results, given the complexity of regulatory interactions and the possible differences among the different cellular models. Nevertheless, all cellular models point to the influence of ATRA in the expression of PTGDR, which is the main finding of this study, although this is not so clear in the case of cytokine production. Therefore, we believe that this particular aspect needs to be prospectively confirmed in larger series of well characterized patients.
In conclusion, our results suggest that ATRA regulates the expression of PTGDR, which could be important in the regulation of Th1 and Th2 responses in allergic diseases. In addition, the -549T> C polymorphism modulates the binding of ATRA to regulatory elements of the PTGDR promoter and this may contribute to the phenotypic differences observed in allergic patients. The potent induction of PTGDR by ATRA and its inhibition by specific antagonists support that these effects are mediated through RAR receptors, pointing to RAR as a potential therapeutic target in allergic diseases.
Supporting information S1 Appendix. Methods. (DOCX) S1 Fig. qPCR analysis of cytokines gene expression. A549 cells were transfected with pCTCT-PTGDR and pCCCC-PTGDR expression vectors and with carrier DNA (control condition). Cells were collected after 24 and 48h followed by qPCR analysis of IL4, IL6 and IL8 cytokines genes. Gene expression was normalized to GAPDH mRNA levels. Data are shown as fold increase relative to mRNA levels for control cells. ( ��� P<0.001 for IL6 in CTCT versus control at 24h and for IL6 in CCCC versus control at 48h; �� P<0.01 for IL6 in CCCC versus control at 24h, and for IL4 in CCCC versus control at 24h; � P<0.05 for IL6 in CTCT versus control at 48h and IL8 in CCCC versus control at 24h).