Investigation of PPARβ/δ within Human Dental Pulp Cells: A Preliminary In Vitro Study

Controlling the inflammatory response to restore tissue homeostasis is a crucial step to maintain tooth vitality after pathogen removal from caries-affected dental tissues. The nuclear peroxisome proliferator-activated receptor beta/delta (PPARβ/δ) is a ligand-activated transcription factor with emerging anti-inflammatory roles in many cells and tissues. However, its expression and functions are poorly understood in human dental pulp cells (hDPCs). Thus, this study evaluated PPARβ/δ expression and assessed the anti-inflammatory effects evoked by activation of PPARβ/δ in lipopolysaccharide- (LPS-) induced hDPCs. Our results showed that hDPCs constitutively expressed PPARβ/δ mRNA/protein, and treatment with LPS increased PPARβ/δ mRNA expression. The selective PPARβ/δ agonist GW0742 significantly decreased inflammation-related mRNA expression in hDPCs (IL6, IL1β, TNFα, MMP1, and MMP2) and RAW264.7 cells (Il6 and Tnfα). Further, PPARβ/δ agonist attenuated MMP2/9 gelatinolytic activity in hDPCs. Previously LPS-conditioned hDPCs increased the migration of RAW264.7 cells through the membrane of a Transwell coculture system. Conversely, pretreatment with GW0742 markedly decreased macrophage recruitment. These findings provide among the first evidence that hDPCs express PPARβ/δ. In addition, they suggest that activation of PPARβ/δ by GW0742 can attenuate some cellular and molecular in vitro aspects related to the inflammatory process, pointing out to investigate its potential target role in dental pulp inflammation.


Introduction
After traumatic injuries and dental caries, a natural defense response takes place within the dentin-pulp complex. When controlled and self-limited, resolving inflammation stimulates regenerative events [1]. These culminate in reactionary dentin production by the primary odontoblasts [2] or, if these cells die, in reparative dentin production by the stem/progenitor cells present in the pulp tissue [3]. The signalling events related to stem cell recruitment and differentiation into a new generation of odontoblast-like cells are complex and not fully understood. However, it is now evident that many molecules, which classically act as inflammatory mediators, including bacterial components, reactive oxygen species (ROS), and cytokines, are also involved in repair response, in a time-and concentrationdependent manner [4][5][6]. Further, activation of canonically related inflammatory pathways, such as NFκB and MAPK, can also signal in favour of the repair process [7,8]. Potentially, while relatively low levels of cytokines and growth factors can stimulate repair, high amounts of these molecules, as a result of more intense/persistent bacterial challenges and inflammation, can actively inhibit tertiary dentinogenesis [9]. These findings, together with the tissue breakdown caused by the course of the immune/inflammatory process within an inextensible environment, highlight the need for regenerative approaches based on therapeutic targets to attenuate the inflammation.
Peroxisome proliferator-activated receptor beta/delta (PPARβ/δ) is a ligand-activated transcription factor that belongs to the nuclear hormone receptor (NR) superfamily. Amongst endogenous ligands are fatty acids, prostaglandins, and leukotrienes, while synthetic agonists include GW0742, a selective high-affinity agonist, widely used in research to explore the role of PPARβ/δ [10]. Besides major functions in the metabolism, PPARβ/δ activation displays anti-inflammatory/immune roles, by negatively interfering with proinflammatory transcription factor signalling pathways [11]. Additionally, PPARβ/δ can act in cell proliferation, differentiation, apoptosis, and angiogenesis, key cellular processes involved in healing and regeneration [12]. Together, such pleotropic functions make PPARβ/δ a potential therapeutic target to be explored in the dentin-pulp context. However, despite being broadly expressed, this NR has not been reported in dental pulp cells, nor have its related roles. Thus, the aim of this study was to examine whether human dental pulp cells (hDPCs) express PPARβ/δ and to gain insight into its anti-inflammatory effects.

Materials and Methods
This study was approved by the Ethics Board of the School of Health Science at the University of Brasilia (number 1400631), and informed consent was obtained from all participants.

Cell Cultures and Treatments.
Primary hDPCs were harvested from extracted nonerupted, caries-free third molars with partially root formation of eleven young donors (aged 18 to 21), without systemic disorder, and with no history of regular medication intake. These criteria were considered because the result reproducibility can be affected by the donor tooth conditions, such as stage of development [13] or retained/erupted tooth [14], the donor age [15], and the presence of systemic diseases [16]. Primary cultures were established by the explant outgrowth method [17]. Briefly, immediately after tooth extraction, dental pulps were removed, and the tissues were minced into small fragments and placed into 35 mm culture dishes with high-glucose Dulbecco's modified Eagle medium (DMEM; Sigma-Aldrich, St Louis, MO) supplemented with 20% fetal bovine serum (FBS; Invitrogen, Carlsbad, CA) and antibiotics (50 U/mL penicillin and 50 mg/mL streptomycin; Sigma-Aldrich). The fragments were stabilized with a glass coverslip. Cells were incubated at 37°C in a humidified incubator at 5% CO 2 and culture medium was replaced every 2-3 days. Confluent cells were subcultured with DMEM supplemented with 10% FBS and antibiotics. All procedures were done in biohazard laminar flow hood and under the sterile conditions, following a rigid laboratory routine with the best practices to avoid misidentification or cross-contamination. Cells at passage 4 were used for all experiments. The RAW264.7 murine macrophage cell line was purchased from the American Type Culture Collection (ATCC® TIB71™, Manassas, VA) and kindly provided by Dr. Paul Webb from the Methodist Research Institute, Houston, TX. RAW264.7 cells were maintained in high-glucose DMEM containing 10% FBS, 50 U/mL penicillin, and 50 μg/mL streptomycin. All experiments were conducted using serum-free medium, with serum starved 24 h before each experiment.

PPARβ/δ Gene Expression Was Upregulated in LPS-
Stimulated hDPCs. Treatment with 2 μg/mL LPS slightly but significantly increased the PPARβ/δ level (p < 0:05) (Figure 1(c)). A similar but more pronounced increase was obtained after the hydrogen peroxide (H 2 O 2 ) stimulus (Supplementary Fig. 3A). This compound was used to mimic a more pronounced inflammatory process as the carious lesion progresses, with increased levels of ROS and cytokines [28].

PPARβ/δ Agonist Attenuated Inflammatory Gene
Expression. The PPARβ/δ agonist GW0742 was first tested for cytotoxicity, and data showed that none of the concentrations used affected cell viability at any time-point considered (Figure 2(a)). We also tested the effect of LPS alone or in association with GW0742 in cell viability, and we did not find any statistical difference between vehicle-treated cells and cells treated with LPS alone or with GW0742 ( Supplementary  Fig. 4). Then, we assessed IL6, IL1β, and TNFα expression. As expected, exposure of hDPCs to 2 μg/mL LPS increased inflammatory gene expression. Conversely, GW0742 pretreatment significantly reduced LPS-induced IL6 and IL1β at 0.1 μM, and LPS-induced IL6, IL1β, and TNFα at 1.0 μM concentration (Figure 2(b)). GW0742 repressing proinflammatory gene expression was also observed with H 2 O 2 stimulus ( Supplementary Fig. 3B). The results with RAW264.7 cells showed that the pretreatment with GW0742 (0.01 μM) significantly reduced Il6 and Tnfα levels in LPS-stimulated cells, compared with the group exposure to LPS alone (Figure 2(c)).

PPARβ/δ Agonist Attenuated MMP Expression and
Gelatinolytic Activities. After exposing cells to 2 μg/mL LPS, data revealed an increase in MMP1 level, with no impact in MMP2 expression. When hDPCs were pretreated with GW0742, a significant downregulation of MMP1 and MMP2 levels at 1.0 μM concentration was observed (Figure 3(a)). Similar results were found with H 2 O 2 stimulus ( Supplementary Fig. 3C). In accordance with gene expression data, LPS stimulation did not affect MMP2 proteolytic activity, but pretreatment with 1.0 μM GW0742 caused a slight reduction on the gelatinolytic activity that was statistically significant (Figure 3(b)). In these conditions, we did not detect the activity of MMP9. However, when hDPCs were treated with a higher dose of LPS (10 μg/mL), a slim proteolytic band of MMP9 was observed. Densitometric analysis showed a slight enhancement in 3 PPAR Research MMP9 activity after exposure cells to LPS, and a dosedependent decrease in the proteolytic band when cells were pretreated with GW0742 (Figure 3(c)). To confirm them to be MMPs with gelatinolytic activity, an EDTA inhibition assay was performed (data are available on request), excluding other MMPs with low specific activity for gelatin [29].

Discussion
In this study, we provide among the first evidence of PPARβ/δ mRNA and protein expression in dental pulp cells. We also revealed that PPARβ/δ activation by the specific ligand GW0742 improved the inflammatory profile by attenuating some aspects related with the inflammatory process, including proinflammatory cytokine gene expression, MMP gene expression, gelatinase activity, and macrophage recruitment.
To explore whether PPARβ/δ is expressed by hDPCs, we firstly screened the mRNA expression by using real-time qPCR and confirmed the protein expression and cell localization by immunofluorescence. The basal PPARβ/δ localization was predominantly on the nucleus of hDPCs, in agreement with other reports [30,31]. Like many NRs, PPARβ/δ is generally localized on the nucleus, binding to the promoter regions of its target genes as a heterodimer with a retinoid X receptor. Canonically, in the absence of agonists, PPARβ/δ mediates gene repression, while gene expression is induced in the presence of its agonists [32]. In addition, activated PPARβ/δ can also repress genes independently of DNA binding, by interacting with transcription factors, for example, [33]. Thus, its localization on the nucleus is compatible with the mode of action and suggests that PPARβ/δ might exert some roles in hDPCs. Indeed, when hDPCs were stimulated with LPS, a component of Gram-negative bacteria that triggers the innate immune response [34,35], PPARβ/δ expression significantly increased. Such an increase was also observed after stimulus with H 2 O 2 , suggesting a pathophysiologic role of this NR.
To gain insights into PPARβ/δ anti-inflammatory function, we pretreated LPS-stimulated cells with three noncytotoxic concentrations of GW0742, and then, the IL6, IL1β, and TNFα mRNA expression was assessed. In

PPAR Research
our study, treatment with GW0742 significantly reduced proinflammatory cytokine gene expression in hDPCs, in agreement with other reports that ascribe to PPARβ/δ a regulatory role on transcription and inflammatory mediator production [19,[36][37][38][39]. The potential of GW0742 in repressing proinflammatory gene expression in hDPCs was also supported by a second inflammatory in vitro model with H 2 O 2 . Because macrophages seem to play a critical role in the progression of pulpal inflammation, we next tested whether PPARβ/δ agonist could modulate cytokine gene expression in LPS-stimulated RAW264.7 cell line, and similar to dental pulp cells, a reduction in Il6 and Tnfα mRNA levels was also observed. The mechanisms by which the PPARβ/δ ligand reduces inflammatory response in hDPCs should be assessed. However, data from macrophages and other cell lines suggested a direct inhibition of NFκB and STAT transactivation by activated PPARβ/δ without direct DNA contact [19,37]. The asso-ciation with the transcriptional repressor protein B cell lymphoma 6 (BCL6) has also been described: unliganded PPARβ/δ can physically associate with BCL6, thus preventing BCL6 to repress proinflammatory genes. Conversely, in the presence of the agonist, PPARβ/δ dissociates from BCL6, releasing it to suppress proinflammatory pathways [38,40]. Canonical direct transcriptional induction of anti-inflammatory genes, such as TGFβ [41] and antioxidative genes [38,39], might be the other way activated PPARβ/δ exerts its actions.
Because pulp tissue destruction involves extracellular matrix breakdown by the action of proteolytic enzymes, such as matrix metalloproteinases, released to facilitate immune cells recruitment [28], the effects of GW0742 in MMP gene expression and gelatinase activity were also investigated. In our study, pretreatment with 1.0 μM GW0742 significantly reduced the expression of the collagenase MMP1 and the gelatinase MMP2. A reduction in MMP2 proteolytic activity

PPAR Research
was also observed. Interestingly, LPS did not increase MMP2 gene expression and activity. We attributed these results to the posttreatment period evaluated (24 h) [42]. The MMP9 levels were relatively low, with undetectable basal transcript levels and protein activity for most cultures investigated. However, when we increased LPS concentration, an MMP9 proteolytic activity was observed. Additionally, pretreatment with GW0742 effectively decreased its activity. Finally, to further explore the biologic relevance of the PPARβ/δ activation, hDPCs previously treated with GW0742/LPS were tested for their chemotactic effect on macrophage recruitment. Among immune cells, macrophages likely predominate in health and inflamed pulp tissue [43]. Further, their number increases with the progression of dental caries, playing a key role in the course of pulp inflammation and necrosis [44]. Thus, a tight regulation of macrophage recruitment might protect the pulp from excessive inflammation and collateral damage. Here, our study revealed that pretreated LPS-stimulated hDPCs with GW0742 markedly altered their chemotactic gradient, resulting in suppression of RAW264.7 migration, with likely phenotypic alteration, as recently suggested [27]. Indeed, LPS-stimulated RAW264.7 cells cocultured with hDPCs expressed less proinflammatory factors compared to LPSstimulated ones. Also, RAW cells cocultured with DPSCs appeared more morphologically elongated than cells cul-tured without hDPCs, which seem clustered and roundshaped [27].
Taken together, our findings indicate that GW0742 can contribute to attenuate the proinflammatory environment in the pulp, protecting it from excessive inflammation and destructive damage. The anti-inflammatory properties together with repair induction might further increase the therapeutic potential of activated PPARβ/δ in the dentinpulp complex. Our preliminary results from a pilot assay showed an increase in nodule formation by GW0742treated hDPCs ( Supplementary Fig. 5). A previous study showed that ligand-activated PPARβ/δ induced osteogenic differentiation of osteoblasts, with an increase in bone nodule formation and alkaline phosphatase expression. They suggested that PPARβ/δ activation can amplify Wntdependent and β-catenin-dependent signalling through transcriptional regulation of the low-density lipoprotein receptor-related protein 5 (LRP5) and direct interaction with β-catenin [45]. Due to the importance of the Wnt/β-catenin signalling pathway in dentin formation and repair, its interaction with PPARβ/δ might be a new field to investigate.
The present study has several limitations. The preventive design of experiments, despite being widely used in agonism studies, does not match with clinical reality. The sample size within some experiments is another limitation, and further studies are suggested to endorse reproducibility.

PPAR Research
Furthermore, although we have shown promising gene expression results, it is also important to investigate whether such effects are reproducible in the protein levels. Thus, future protein detection experiments are necessary to better support the present results. Also, further experiments are needed to elucidate the effects of GW0742 itself in gene expression, to investigate the signalling pathway involved with the anti-inflammatory activity of activated PPARβ/δ, to explore how activated PPARβ/δ alters hDPC chemotaxis, and to confirm whether the effects are receptor-dependent or independent. Nevertheless, our findings highlight a new target to be explored in future research and open potential new therapeutic avenues for the treatment of pulp diseases, which could be used in association with microbial control.

Conclusion
In conclusion, this study is the first to demonstrate PPARβ/δ expression in human dental pulp cells, and it suggested that activated PPARβ/δ has anti-inflammatory effects in hDPCs by preventing proinflammatory and MMP gene expression, suppressing gelatinase activity and macrophage recruitment.

Data Availability
Data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
The authors declare that there is no conflict of interest regarding the publication of this paper.

Supplementary Materials
Supplementary information may be found in the online version of this article: Supplementary methods. Quantitative real-time polymerase chain reaction: all quantitative real-time PCR assays were performed on the Applied Biosystems StepOnePlus™ Real-Time PCR Systems, and the data were generated by StepOne v2.1 Software. The cycle conditions were 50°C for 2 minutes, 95°C for 10 minutes, followed by 40 cycles at 95°C for 15 s and 60°C for 1 minute. Samples were normalized to beta-actin (hDPCs) or to glyceraldehyde 3-phosphate dehydrogenase enzyme (Gapdh; RAW264.7 cells). To assure specificity, a melt curve was obtained for all qPCR products. Further, standard curves were obtained for each primer pair to assess the efficiency of amplification. Mineralization assay: confluent hDPCs were cultured in mineralization medium (alpha minimum essential medium (α-MEM) with 10% FBS, antibiotics, 10 mmol/L β-glycerophosphate, 10 nmol/L dexamethasone, and 50 μg/mL ascorbic acid) containing GW0742 (1.0 μM) or vehicle (DMSO 0.01%). After 28 days, cells were rinsed with PBS, fixed with ethanol for 30 minutes at RT, and stained for 10 minutes with 2% Alizarin red S (Sigma-Aldrich) solution, pH 4.2, at RT. Cells were then rinsed 3 times with distilled water to reduce nonspecific staining. This experiment was performed in triplicate. Supplementary Table 1 Gene and primer sequences. Supplementary Figure  1 MTT cell viability assay to assess DMSO safety. hDPCs were incubated with DMEM/2% FBS containing DMSO 0.1% or left untreated (hDPCs), and metabolic activity was evaluated daily during 6 days by MTT colorimetric assay. Supplementary Figure 2 Diagrams of experimental protocol for GW0742 treatment followed by inflammatory stimulus with LPS in hDPCs (2 μg/mL) and RAW264.7 cells (100 ng/mL), for (a) gene expression and gelatinolytic activity, and for (b) chemotaxis assay using coculture Transwell system. Supplementary Figure 3