1. HDAC Activity Is Required for p65/RelA-Dependent Repression of PPARδ-Mediated Transactivation in Human Keratinocytes 
Lene Aarenstrup, Esben Noerregaard Flindt, Kristian Otkjaer, Morten Kirkegaard, Jens Skorstensgaard Andersen, Karsten Kristiansen 
Journal of Investigative Dermatology 
Volume 128, Issue 5, Pages (May 2008)
DOI: /sj.jid
Copyright © 2008 The Society for Investigative Dermatology, Inc Terms and Conditions

2. Figure 1
Activated NF-κB represses PPARδ-mediated transactivation in HPKs. HPKs were transiently transfected with the CMV-β-galactosidase construct for normalization and the PPRE-containing Tk-Luc reporter construct with or without pCMV-IκBαM. The cells were treated with GW and IL-1β or vehicle (0.1% DMSO) for 3hours before harvesting. All transfections were performed in triplicates, measured in duplicates, and the luciferase activities were normalized to the corresponding β-galactosidase activities. Results represent means±SD from three independent experiments.
Journal of Investigative Dermatology  , DOI: ( /sj.jid )
Copyright © 2008 The Society for Investigative Dermatology, Inc Terms and Conditions

3. Figure 2
Only the p65/RelA subunit of NF-κB is a strong repressor of PPAR-mediated transactivation in HPKs. HPKs were cotransfected with the PPRE-Tk-Luc and different combinations of expression vectors encoding hPPARδ, (a) p65/RelA, (b) p50, (c) RelB, (d) c-rel, and (e) p52. The cells were treated with L or vehicle for 24hours before harvesting. The SV40-β-galactosidase construct was used for normalization. All transfections were performed in triplicates, measured in duplicates, and the luciferase activities were normalized to the corresponding β-galactosidase activities. Results represent means±SD from three independent experiments. *P=3.39 × 10−9 when cells transfected with the highest amount of p65/RelA encoding vector were compared with p65/RelA-null-transfected cells.
Journal of Investigative Dermatology  , DOI: ( /sj.jid )
Copyright © 2008 The Society for Investigative Dermatology, Inc Terms and Conditions

4. Figure 3
Intracellular localization of PPARδ, p65/RelA, and p50 in HPKs after stimulation with IL-1β. HPKs were stimulated with IL-1β (10ngml−1) for the indicated time before fixation and immunostaining of PPARδ, p65/RelA, and p50. The different proteins were detected using Alexa488-conjugated secondary antibodies and DNA was visualized by SYTOX. The magnification used was × 630. One representative data set out of three independent experiments is shown.
Journal of Investigative Dermatology  , DOI: ( /sj.jid )
Copyright © 2008 The Society for Investigative Dermatology, Inc Terms and Conditions

5. Figure 4
The PPARδ DNA binding to the PPARδ/RXR recognition site from the human ADRP promoter is not affected by activated NF-κB. Nuclear extracts were isolated from HPKs treated with either vehicle or IL-1β for 30minutes. (a) PPARδ binding to the ADRP, (b) a mutated PPARδ/RXR recognition site, and (c) NF-κB binding to the P53 κB site were assayed by gel-shift assays. To demonstrate the specificity of the binding, unlabelled homologous oligonucleotides were added to the binding reactions (a and c, lanes 3 and 6; b, lanes 3 and 5). Specific antibodies were added to the binding reactions to supershift the DNA/protein complex (a and c, lanes 4 and 7). *Indicates that the PPAR binding deviates 5.1% when comparing lane 2 with lane 5. One representative data set out of three independent experiments is shown.
Journal of Investigative Dermatology  , DOI: ( /sj.jid )
Copyright © 2008 The Society for Investigative Dermatology, Inc Terms and Conditions

6. Figure 5
DNA binding via the PPARδ DNA-binding domain is required for p65/RelA-mediated repression. HPKs were cotransfected with UAS-Tk-Luc (a and c) or PPRE-Tk-Luc (b and d) and p65/RelA, and different expression plasmids encoding Gal4-PPARδ (full length) (a and b), Gal4-PPARδ (ΔAB) (c), and PPARδ (ΔAB) (d), respectively, and the cells were treated with L or vehicle for 24hours before harvesting. The SV40-β-galactosidase construct was used for normalization. All transfections were performed in triplicates, measured in duplicates, and the luciferase activities were normalized to the corresponding β-galactosidase activities. Results represent means±SD from three independent experiments. *P=0.014, **P= , ***P= , and ****P= when compared with p65/RelA-null-transfected cells, respectively.
Journal of Investigative Dermatology  , DOI: ( /sj.jid )
Copyright © 2008 The Society for Investigative Dermatology, Inc Terms and Conditions

7. Figure 6
TSA abolishes p65/RelA-mediated repression of PPARδ-dependent transactivation. (a) HPKs were cotransfected with PPRE-Tk-Luc and different expression plasmids encoding PPARδ (20ng) and p65/RelA. The cells were treated with L or vehicle, TSA, or VPA for 24hours before harvesting. The SV40-β-galactosidase construct was used for normalization. All transfections were performed in triplicates, measured in duplicates, and the luciferase activities were normalized to the corresponding β-galactosidase activities. Results represent means±SD from three independent experiments. *P=0.27, **P=0.018, and ***P=5.24 × 10−6 when compared with p65/RelA-null-transfected cells, respectively. (b) HPKs were treated with TSA or VPA for the indicated time before whole-cell preparations, and proteins were separated in SDS–polyacrylamide gels and analyzed by western blotting. TFIIB was included to confirm equal loading and protein transfer. Bands are quantified by normalization to TFIIB and values are shown below each band.
Journal of Investigative Dermatology  , DOI: ( /sj.jid )
Copyright © 2008 The Society for Investigative Dermatology, Inc Terms and Conditions

8. Figure 7
Ligand-induced ADRP expression is repressed by TNFα in the absence but not in the presence of TSA in HPKs. (a) HPKs were treated with GW or vehicle for the time indicated before harvest. *P= when compared with the vehicle-treated cells (24hours). (b) HPKs were pretreated with TSA for 14hours before being stimulated with L for 2.5hours and TNFα (50μgml−1) for the last 0.5hour as illustrated in lower part of (b). After harvesting (a and b), RNA was purified from cell extracts and the ADRP mRNA content was assessed by real-time PCR. The mRNA level was normalized to TBP and results represent means±SD from three independent experiments. Significant repression in the absence of TSA, ***P=0.0019, but no significant repression in the presence of TSA, **P=0.949, when compared with cells not treated with TNFα.
Journal of Investigative Dermatology  , DOI: ( /sj.jid )
Copyright © 2008 The Society for Investigative Dermatology, Inc Terms and Conditions

9. Figure 8
Interaction between PPARδ, p65/RelA, and HDAC1 in 293T cells in vivo. 293T cells were transiently transfected with different expression plasmids encoding PPARδ myc-tagged (a) or untagged (b), p65/RelA (a–c) or p65/RelA-RHR (c), and myc-HDAC1 (b) or HDAC1 (a). The cells were treated with L for 24hours, and after cell lysis, proteins were immunoprecipitated using myc–sepharose beads. After immunoprecipitation, the beads were collected and proteins were denatured and separated in SDS–polyacrylamide gels and analyzed by western blotting. The membranes were stripped for antibody and reprobed for detection of the different proteins. One representative data set out of four independent experiments is shown.
Journal of Investigative Dermatology  , DOI: ( /sj.jid )
Copyright © 2008 The Society for Investigative Dermatology, Inc Terms and Conditions

10. Figure 9
Interaction between PPARδ and HDAC1 in vitro. (a) Plasmids encoding PPARδ, GST-HDAC1(1–432), or GST were expressed in E. coli, and the bacteria lysates were incubated at 1:1 as indicated overnight, together with glutathione–sepharose beads and L or vehicle. Proteins from the pull downs were separated on SDS–polyacrylamide gels and analyzed by western blotting. One representative data set out of three independent experiments is shown. (b) HPKs were transiently transfected with PPRE-Tk-Luc and expression plasmids encoding hPPARδ (20ng) and HDAC1, and the cells were treated with L or vehicle for 24hours before harvesting. The SV40-β-galactosidase construct was used for normalization. All transfections were performed in triplicates, measured in duplicates, and the luciferase activities were normalized to the corresponding β-galactosidase activities. Results represent means±SD from three independent experiments. *P=0.009 when compared with HDAC1-null-transfected cells.
Journal of Investigative Dermatology  , DOI: ( /sj.jid )
Copyright © 2008 The Society for Investigative Dermatology, Inc Terms and Conditions