PPARβ/δ-Interfering Peptide Enhanced Mesenchymal Stromal Cell Immunoregulatory Properties

Background Mesenchymal stem/stromal cells (MSCs) have been widely used for their therapeutic properties in many clinical applications including osteoarthritis. Despite promising preclinical results showing the ability of MSC to reduce the clinical severity of osteoarthritis (OA) in experimental animal models, the benefits of intra-articular injection of MSC in OA patients are limited to the short term. In this regard, it is anticipated that improving the properties of MSC may collectively enhance their long-term beneficial effects on OA. Methods and Results Recently, we have shown that PPARβ/δ inhibition using a commercially available antagonist in murine MSC increases their immunoregulatory potential in vitro as well as their therapeutic potential in an experimental murine arthritis model. Here, we relied on an innovative strategy to inhibit PPARβ/δ:NF-κB TF65 subunit interaction in human MSC by designing and synthesizing an interfering peptide, referred to PP11. Through RT-qPCR experiments, we evidenced that the newly synthesized PP11 peptide reduced the expression level of PDK4, a PPARβ/δ target gene, but did not modify the expression levels of ACOX1 and CPT1A, PPARα target genes, and FABP4, a PPARγ target gene compared with untreated human MSC. Moreover, we showed that human MSCs pretreated with PP11 exhibit a significantly higher capacity to inhibit the proliferation of activated PBMC and to decrease the frequency of M1-like macrophages. Conclusions We designed and synthesized an interfering peptide that potently and specifically blocks PPARβ/δ activity with concomitant enhancement of MSC immunoregulatory properties.


Introduction
Strategies to improve mesenchymal stem/stromal cell (MSC) therapeutic properties for osteoarthritis (OA) treatment by enhancing in particular, immunoregulatory, and cytoprotective properties are a matter of intense research. These strategies consist, in part, of genetic or pharmacological preconditioning of MSC to further enhance the expression of one candidate gene or to activate a specific pathway.
Recently, we demonstrated that PPARβ/δ, a lipid ligandinducible transcription factor described for its metabolic functions, plays a key role in MSC derived from murine bone marrow (mMSC) through its capacity to regulate their NO production [1]. This suppressive role of PPARβ/δ was mediated through the regulation of NF-κB activity, a major player in MSC immunosuppressive properties [1,2]. Murine MSC from PPARβ/δ knockout mice exhibited increased immunosuppressive potential which was interestingly correlated with an increased NF-κB activity both at a steady state and upon activation with proinflammatory cytokines [1]. Thus, PPARβ/δ inhibition in MSC increased their immunoregulatory potential in vitro, on activated splenocytes or Tcell subsets, and their therapeutic potential in an experimental murine arthritis model [3]. Indeed, primary function of PPARs during the inflammatory response is to promote NF-κB inactivation. Possible mechanisms of inactivation could include inhibition through direct binding and ubiquitination leading to proteolytic degradation of NF-κB TF65 as well as indirect effects on NF-κB [4].
Additionally, whether PPARβ/δ modulation using commercially available PPARβ/δ antagonists in human MSC could affect their immunoregulatory potential has never been investigated. Moreover, the nonspecificity of the current available PPARβ/δ antagonists that can modulate the activity of other members of the PPAR family is well documented [5], as well as the fact that the MSC culture media serum contains PPARβ/δ ligands [6]. Indeed, PPAR ligands are present in the serum of the cell culture media and can be produced by the cells themselves [6]. Thus, according to the cell type used in studies (for instance human versus mouse MSC), different concentrations of PPAR ligands can accumulate within the culture media.
In addition to the influence of ligands on the conformation of PPARβ/δ, the molecular environment of the nuclear receptor (presence or absence of coactivators/corepressors) might also influence its activity [7,8]. Moreover, at this time, despite the role of PPARβ/δ in many cellular mechanisms, the synthesis of PPARβ/δ antagonists or reverse agonists are limited to three compounds (antagonists: GSK3787, SR13904 or reverse agonist: GSK0660), and no synthetic ligand of PPARβ/δ has received marketing authorization.
PPARβ/δ can interact directly with other transcription factors, including NF-κB, by blocking their nuclear translocation and/or their binding to the DNA. Given that the NF-κB signaling pathway is required for the initial priming of immunosuppressive function in human MSC [2], in the present study, we relied on an innovative strategy to inhibit PPARβ/δ:NF-κB TF65 subunit interaction by designing an interfering peptide. Then, we studied the specificity of this peptide mimicking the NF-κB TF65 subunit in adipose tissue-derived MSC (AD-MSC) by evaluating (i) the expression levels of PPAR family member target genes, (ii) their immunoregulatory properties on activated human peripheral blood mononuclear cells (PBMC) and the expression level of MSC immunoregulatory mediators, and (iii) their immunoregulatory properties on macrophages.
The following libraries were synthesized twice: PepScan (15-mer overlapping peptides with a shift of 3 amino acids); length analysis (18-mer peptide shorted from N-and C-terminus); substitutional analysis (each amino acid of the PP11 peptide replaced by the 20 L-amino acids).

2.2.
Peptide Array Binding Studies. The cellulose membranebound peptides were prewashed once with ethanol (1 × 10 min), with tris-buffered saline (TBS), pH 8.0, (3 × 10 min), and then blocked for 3 h with blocking buffer (Sigma-Aldrich) in TBS pH 8.0, containing 5% sucrose (Sigma-Aldrich). The membranes were incubated with human recombinant PPARβ/δ protein (Leu-260-Glu426; PPARD-15H human, Interchim) having an N-terminal His-tag for the detection (5 μg/ml in blocking buffer overnight at 4°C). The next day, the membrane was washed with TBS (3 × 10 min) and incubated with a primary anti-His antibody (H1029, Sigma-Aldrich; 1 : 1,000 in blocking buffer) for 2.5 h at room temperature. After TBS washes (3 × 10 min) to remove the primary antibody, the membrane was incubated with a peroxidase-labeled anti-mouse-HRP (#7076S, Cell Signaling; 1 : 2,000 in blocking buffer) for 1.5 h at room temperature. To remove antibody excess, the membrane was washed with TBS (3 × 10 min) and then incubated with Pierce ECL Plus Western Blotting Substrate (Thermo Fisher). Images were acquired with an Imager 600 (Amersham), and spot intensities were measured using the "protein array analyzer" macro of ImageJ. An arbitrary threshold was set for each membrane at 100,000 to determine the "binders" (see Supplementary materials, Tables S1, S2, and S3).

Human Adipose-Derived Stromal Cell Isolation and
Expansion in Hollow-Fiber Bioreactor. Human adiposederived stromal cells (hAD-MSCs) were isolated from abdominal adipose tissue obtained from 6 healthy donors (N = 6) receiving aesthetic surgery and were expanded in vitro. In brief, lipoaspirate from each donor was washed twice with a buffered-saline solution to remove residual blood. The adipose tissue was digested by incubation with collagenase NB4 (Nordmark). The collagenase was neutralized with cell culture media (Minimum Essential Medium, α-MEM, Gibco) containing 5% human platelet lysate (hPL Stemulate, Sexton Biotechnologies), and the suspension was filtered, centrifuged, and resuspended in complete 2 Stem Cells International media. The number of mononucleated cells (MNCs) in the isolated stromal vascular fraction (SVF) and cell viability was determined using an automated cell counter (NucleoCounter® NC-200™, ChemoMetec). SVF was then resuspended in the cell culture media at the desired concentration and loaded into the intracapillary space of the hollow fibers of the Quantum® Cell Expansion System (Terumo BCT). Expansion of MSC isolated from adipose tissue or bone marrow has been previously described by others ( [11,12], respectively). Briefly, 24 h before cell loading, the bioreactor was coated with fibronectin from human plasma (Corning, NY) to support attachment and expansion of MSC. Over the course of cell expansion (7-8 days), the feed rate was adjusted according to the manufacturer's instructions. After trypsinization and washing steps with a buffered-saline solution, hAD-MSC (passage P0) were cryopreserved in a cryopreservation media (Cryostor® CS10, BioLife Solutions). hAD-MSCs used in this study were obtained after a second expansion step in the Quantum® system (passage P1, 6 days of cultivation) and cryopreserved in Cryostor® CS10 (BioLife Solutions). Of note, the telomereassociated-protein Rap1 (Trf2IP) was not or weakly expressed in hAD-MSC from the 6 donors used in that study. To determine the percentage of inhibition of PBMC proliferation by hAD-MSC, PHA-activated PBMC was considered as 100% of proliferation; hAD-MSC decrease the percentage of PHA-activated PBMC proliferation; the difference between both conditions corresponds to the percentage of inhibition.
2.9. Indoleamine 2,3-Dioxygenase (IDO) Activity. In order to quantify the IDO activity, we analyzed the tryptophan-tokynurenine conversion using a photometric detection method. 60 μl of cell supernatant (conditioned medium) was pipetted to a 96-well culture plate, and 30 μl of 30% trichloroacetic acid solution (Sigma) was added for 30 min at 50°C. After a centrifugation step, 75 μl of the samples were added to 75 μl of freshly prepared Ehrlich's solution (Sigma, MO, USA), and photoabsorbance was read at 450 nM on Multiskan plate reader (Thermo Fisher).

Statistical
Analysis. GraphPad Prism 7.0v software (GraphPad, CA, USA) was used for the presentation of the results and statistical analysis. All over the manuscript, results were expressed as the mean ± SEM. The statistical analysis was performed using the nonparametric Kruskal-Wallis followed by Dunn's post hoc test for multiple comparisons. For comparison between the two groups, the Mann-Whitney U or Wilcoxon (IDO activity) tests were used. Statistical significance was noted as ns for p > 0:05, * for p < 0:05, * * for p < 0:01, * * * for p < 0:001, and * * * * for p < 0:0001.

Screening PPARβ/δ Interactions to Develop an
Interfering Peptide. It is known that PPARβ/δ can bind to proteins including nuclear factor NF-κB TF65 subunit (also called transcription factor p65, TF65) and attenuate NF-κBdependent signaling [13]. However, to our best knowledge, no structural information showing the interface of this interaction is published so far. To screen PPARβ/δ:NF-κB TF65 interaction, we used the SPOT technology allowing the parallel synthesis of cellulose-bound peptide libraries for the screening of protein-protein interaction (Figure 1(a)).
First, a PepScan analysis was performed by dissecting the primary sequence of TF65 in overlapping peptides to determine the NF-κB TF65 epitope involved in the interaction with PPARβ/δ (Figure 1(b)). Because a direct correlation between signal intensity and affinity could not be performed using this technology, we set an arbitrary threshold (values > 100,000) to determine "binders" (Supplementary materials, Table S1). Many potent epitopes were detected after PPARβ/δ protein incubation. To select the more potent one that is able to interrupt PPARβ/δ:TF65 interaction, we used the two published crystal structures of NF-κB TF65 [14,15] to evaluate the position of the different epitopes ( Figure 2). The epitope situated in one of the TF65 loops corresponding to the spots n°24 to 25 (70-GTVRISLVTKDPPHRPHP-87) of the PepScan was selected based on its accessibility for protein-protein interaction (red square in Figure 1(b), red part in the two structures of Figure 2).
Afterward, a length analysis was performed on this 18mer peptide to define the optimal length for PPARβ/δ binding by cutting successively the N-and C-terminal regions of the peptide. After PPARβ/δ incubation, the 12mer sequence of spot n°8 (73-RISLVTKDPPHR-84) was determined as the peptide having one of the highest signal intensities (Figure 1(c) and Supplementary materials, Table S2). Finally, this 12-mer peptide was evaluated by substitutional analysis (SubAna, amino acid mutation at each sequence position) to determine the potent key positions of PPARβ/δ:TF65 interaction (Figure 1(d) and Supplementary materials, Table S3). The first column of the SubAna corresponds to the wild-type (wt) sequence and the following column to the exchanged amino acids. The SubAna revealed 4 key positions of the peptide: isoleucine, leucine, lysine, and arginine (RISLVTKDPPHR) which could not be arbitrarily replaced. In many cases, SubAnas with a clear and defined pattern of key positions will demonstrate a specific proteinpeptide interaction [16][17][18].
The determined 12-mer peptide will act as a mimic of the NF-κB TF65 subunit and prevent PPARβ/δ interaction to the NF-κB TF65 subunit. To evaluate its cellular activity, the found decoy peptide was coupled to the commonly used cell-penetrating peptide Tat [19] to allow cellular internalization. Tat conjugated to the PPARβ/δ-interfering peptide is afterward called PP11 (Figure 2(c)).

PP11 Peptide Downregulates the Expression Level of
PPARβ/δ Target Genes without Affecting those of PPARα or PPARγ. To determine the specificity of PP11, we assessed the expression level of PDK4 and ANGPTL4 and PPARβ/δ target genes, as well as the expression levels of ACOX1 and CPT1A, PPARα target genes, and FABP4, a PPARγ target gene in untreated hAD-MSC compared to human hAD-MSC pretreated with either GSK3787 (a PPARβ/δ antagonist) or PP11. Using RT-qPCR, we observed that the 24 h pretreatment of hAD-MSC with 0.1 μM of PP11 or GSK3787 significantly reduced the expression level of PDK4 (Figure 3(a)) as compared to untreated hAD-MSC (dashed line). In contrast, the ANGPTL4 expression level was affected only after the pretreatment with GSK3787 but not with PP11, as compared to untreated hAD-MSC (Figure 3(b)). Then, we assessed the expression levels of PPARα target genes and found that the 24 h pretreatment of hAD-MSC with 0.1 μM of PP11 did not modify the expression level of the two PPARα target genes tested, whereas 0.1 μM GSK3787 significantly increased ACOX1 (Figure 3(c)) and decreased CPT1A (Figure 3(d)). Similarly, while PP11 pretreatment did not modulate the FABP4 expression level, a PPARγ target genes, GSK3787 significantly increased it (Figure 3(e)). Altogether, these results indicate the specificity of the newly synthesized PP11 peptide which reduces specifically the expression level of PDK4, a PPARβ/δ target gene, but not the expression levels of PPARα or PPARγ target genes.

Human AD-MSC Pretreated with PP11 Peptide Exhibit
Enhanced Immunoregulatory Properties. We investigated the role of PPARβ/δ modulation using either the PP11 peptide or GSK3787 on hAD-MSC immunoregulatory functions. To that end, hAD-MSCs were pretreated with 0.1 μM of either PP11 or GSK3787 for 24 h before being cocultured with phytohemagglutinin-(PHA-) activated PBMC at a ratio of 1 : 10 for 4 days. First, we confirmed the immunosuppressive properties of hAD-MSC on CellTrace Violet-(CTV-) labeled PBMC activated with PHA (Figures 4(a) and 4(b)). Although the inhibition of PPARβ/δ activity using the GSK3787 antagonist tended to increase the capacity of hAD-MSC to inhibit the proliferation of PHAactivated PBMC, no significant change in hAD-MSC immunoregulatory potential was observed compared to untreated hAD-MSC (Figures 4(a) and 4(c)). In contrast, the pretreatment of hAD-MSC with PP11 peptide significantly enhanced their immunoregulatory properties compared to their untreated counterpart (Figures 4(a) and 4(d)). Altogether, these results suggest that the PPARβ/δ-mediated repression of TF65, lifted/removed by the PP11 peptide, is pivotal for hAD-MSC immunosuppressive functions.
Then, we assessed the effect of PPARβ/δ activity modulation on the indoleamine 2,3-dioxygenase (IDO) activity, involved in MSC immunoregulatory properties [20], in hAD-MSC cocultured with PHA-activated PBMC. Compared to untreated hAD-MSC, we showed that hAD-MSC treated with GSK3787 exhibited a significantly lower level of IDO activity while hAD-MSC pretreated with PP11 peptide had a significantly higher IDO activity (Figure 4(e)). Moreover, we evaluated the expression level of the cyclooxygenase-2 (COX2) gene, another MSC immunoregulatory mediator [21] in hAD-MSC pretreated with GSK3787 and PP11. Using RT-qPCR, we observed    Stem Cells International an increased mRNA expression level of COX2 in response to the two pretreatments as compared to untreated hAD-MSC, but a significant increase was obtained only with the PP11 peptide (Figure 4(f)).

Human AD-MSC Pretreated with PP11 Peptide Exhibit an Enhanced Immunoregulatory Property on Macrophages.
Finally, we assessed the effect of the modulation of PPARβ/δ activity on MSC immunoregulatory functions focusing on macrophages that play a pivotal role in OA pathogenesis. MSC regulates the macrophage response by inducing M1-like macrophage polarization toward a M2like phenotype in a COX2-dependent manner [22][23][24]. We thus wondered whether the modulation of PPARβ/δ activity in human hAD-MSC using the PP11 peptide could modify their capacity to control the macrophage response. To address that hypothesis, we performed coculture experiments with M1-like macrophages and hAD-MSC pretreated with PP11 peptide (Figure 5(a)). First, we confirmed that the coculture of hAD-MSC with M1-like macrophages (Mϕ+hAD-MSC) significantly increased the percentage of M2-like macrophages positive for CD163 and CD206 ( Figure 5(b)) and decrease the percentage of M1-like macrophages positive for CD86 and HLA-DR, as compared to M1-like macrophages cultured alone (Mϕ) (Figure 5(c)). When cocultured with hAD-MSC pretreated with PP11 peptide, the percentage of M2-like macrophages positive for CD163 and CD206 was significantly increased compared to macrophages cultured alone but no significant difference was observed compared to macrophages cultured with untreated hAD-MSC ( Figure 5(b)). In contrast, while both untreated hAD-MSC and PP11-treated hAD-MSC significantly decreased the percentage of M1-like macrophages positive for CD86 and HLA-DR, PP11treated hAD-MSC exhibited a more potent immunoregulatory property ( Figure 5(c)). This result suggests that the repression of PPARβ/δ interaction with TF65, a subunit of the NF-κB transcription factor complex, promoting the nuclear translocation of NF-κB in hAD-MSC, enhances the immunoregulatory potential of hAD-MSC on proinflammatory M1-like macrophages.

Discussion
In the present study, we design and synthesize a specific inhibitor of PPARβ/δ:NF-κB TF65 interaction, PP11, and demonstrate in hAD-MSC its potent effect in controlling PPARβ/δ activity without affecting the expression levels of the target genes of other PPAR family members. Human AD-MSC treated with PP11 exhibit enhanced capacity to inhibit the proliferation of activated PBMC and to regulate the macrophage response associated with an increased IDO activity.
Deciphering protein-protein interactions (PPIs) involved in various diseases is an interesting strategy to engineer interfering peptides. Indeed, a complex and dynamic network of PPIs regulates most cellular functions and in particular signaling cascades leading to cell proliferation or death. Specific inhibitors of PPIs are thus emerging as a new class of drugs with many potential applications in basic and clinical research. Unfortunately, PPIs have often been considered nondruggable for several reasons. First, the PPI architecture is often not sufficiently known to allow the rational design of inhibitors. Second, and more importantly, PPIs occur through large, flat   [25,26]. Due to their good solubility, favorable pharmacokinetic profile, low toxicity/mitogenicity, and infinite possibility to improve their binding stability/affinity by modifications, peptides are considered as interesting therapeutic molecules [27]. Therefore, the pharmaceutical industry considered, nowadays, peptides as therapeutics as evidenced by more than 60 peptide drugs approved by the Food Drug Administration and a continuously growing market over the last two decades [28].
To develop an interfering peptide of the complex formed between PPARβ/δ and the nuclear factor NF-κB TF65 subunit, we have synthesized and screened different peptide libraries by SPOT synthesis. This method allows us to elucidate the binding region of the NF-κB TF65 subunit important for PPARβ/δ interaction (70-GTVRISLVTKDPPHRPHP-87, 18-mer), to reduce this epitope to a 12-mer peptide and to determine the key positions (73-RISLVTKDPPHR-84). By coupling the decoy peptide to the cell-penetrating peptide Tat (10-mer), the cellular internalization of this conjugate, called afterward PP11, was ensured (data not shown). Within the cell, PP11 will interact as peptidyl antagonist by mimicking   Pyruvate dehydrogenase kinase 4 (PDK4) is expressed by adipocytes, and increased expression of PDK4 is associated with inhibited glucose uptake [29]. PDK4 acts as an inhibitor of pyruvate oxidation and is involved in the change of cellular metabolism [30]. The activation of the three main isoforms of the PPAR family including PPARα, PPARβ/δ, and PPARγ increases PDK4 expression which involves a coordinated metabolic transfer from glucose to fatty acids as the primary energy fuel [30]. Here, we show that PP11 as well as GSK3787, a potent antagonist of PPARβ/δ, significantly decreased the expression of PDK4 in human AD-MSC without altering the expression levels of PPARα and PPARγ target genes. This result not only demonstrates that PP11 is as effective as GSK3787 in regulating the expression of a PPARβ/δ target gene but also that it is more specific than a commercially available antagonist. This decreased expression level of PDK4 induced by PP11 paralleled the enhanced immunoregulatory properties of human AD-MSC on PHA-activated PBMC. Based on our previous work, showing that the proglycolytic switch of MSC mediated by oligomycin significantly enhanced their immunoregulatory properties in vitro [31]; it is tempting to speculate that AD-MSC pretreatment with PP11 induces a metabolic transfer from fatty acids as the primary energy fuel to glucose in cells. Of course, this hypothesis will require further investigations.
Angiopoietin-like 4 protein (ANGPTL4) is a member of the angiopoietin-like family of proteins structurally related to factors modulating angiogenesis first identified as an adipokine exclusively involved in lipid metabolism [32]. ANGPTL4 has since been shown to be involved in cell differentiation, tumorigenesis, energy and glucose homeostasis, lipid metabolism, wound healing, and inflammation regulation [33], ANGPTL4 transcriptional regulation can be regulated by nutritional and hormonal conditions as well as several transcription factors, including PPARα, PPARβ/δ, PPARγ [33], and HIF1α [34]. Unexpectedly, while the treatment of hAD-MSC with a PPARβ/δ antagonist, GSK3787, significantly reduced the expression of ANGPTL4, the treatment of the cells with PP11 did not. This could be explained by the fact that the transcriptional regulation of ANGPTL4 is mediated by different factors other than PPARβ/δ that could be present in the culture medium of hAD-MSCs or expressed by the cells themselves. Further investigations are needed to clarify that point and confirm the specificity of PP11 for PPARβ/δ and its targets. Although puzzling, this result is very intriguing and could explain the increased immunoregulatory effect obtained with hAD-MSC treated with PP11 as compared to GSK3787 in the in vitro functional tests. Indeed, we showed that PP11-interfering peptide blocks PPARβ/δ activity without affecting ANGPTL4 expression in human AD-MSC. Moreover, the treatment of hAD-MSC with PP11 but not GSK3787 significantly increased their IDO activity compared to the untreated cells and increased the expression level of COX2 that regulates the production of prostaglandin E2 (PGE2), well-known mediators of MSC immunoregulatory properties. Altogether, these effects mediated by PP11 on ANGPTL4 and COX2 expression and IDO activity in human AD-MSC might explain their enhanced immunoregulatory properties. While the role of COX2 and IDO on MSC immunosuppressive properties was described a while ago, the role of ANGPTL4 is more recent. Indeed, MSC overexpressing COX2 more potently inhibited the activation and proliferation of PBMC [21,35]. Regarding IDO, it is one of the most critical factors in human MSC-directed immunoregulation [35,36]. For the ANGPTL4 part, it was more recently shown that in conditions of coculture with macrophages, MSC markedly expressed ANGPTL4 to dampen macrophage polarization toward the proinflammatory phenotype. MSCs lacking ANGPTL4 were not successful in inhibiting the inflammatory phenotype of macrophages [37]. Altogether, these results suggest that the enhanced immunoregulatory properties of human AD-MSC treated with PP11 or GSK3787 compared to naïve MSC or GSK3787 are associated with the higher expression of ANGPTL4 and COX2 and the higher activity of IDO. Further studies should be investigated to confirm the pivotal role of ANGPTL4 induced by PP11 on MSC immunosuppressive properties.
In this study, we have developed PP11, an interfering peptide of the complex formed between PPARβ/δ and TF65 subunit of the nuclear factor NF-κB, that plays a pivotal role in MSC immunosuppressive properties. Although we demonstrated that PP11 enhanced MSC immunosuppressive properties, the effect of PP11 on NF-κB activity in human AD-MSC and on the regulation of NF-κB target genes remains to be studied. Blocking the interaction between PPARβ/δ and NF-κB TF65 subunit may also impact NF-κB TF65 pathway itself resulting in (1) a more rapid proteasomal degradation, (2) a potent higher interaction with other cytoplasmic or nuclear proteins, and (3) an exceeding formation of various NF-κB subunit dimer combinations inducing a heterogeneous regulation of NF-κB target genes. However, the effect of PP11 on NF-κB TF65 signaling pathways is likely to be the subject of another manuscript.

Conclusions
In conclusion, we designed and synthesized an interfering peptide that specifically blocks PPARβ/δ activity and enhances MSC immunoregulatory properties in vitro both on activated PBMC and proinflammatory macrophages. These promising results are an important step to better understand the MSC immunoregulatory properties even if additional biochemical experiments will be necessary to highlight the PP11 effect on the PPARβ/δ:NF-κB TF65 complex. Our study raises the possibility that PP11 will be a more specific PPARβ/δ antagonist than the commercially available drugs.

Data Availability
Data are available on request through the authors (Gautier Tejedor and Farida Djouad). 10 Stem Cells International

Ethical Approval
The study was conducted in accordance with the Declaration of Helsinki and approved by the Regional Investigational Review Board "Comité de Protection des Personnes SUD-EST II, Lyon, France" (Protocol 2019-A01067-50 approved on September 11 th , 2019).

Consent
Informed consent was obtained from all subjects involved in the study.