XB-ART-50023Nucleic Acids Res January 1, 2007; 35 (3): 890-901.
Pax6 transcription is under the control of two main promoters (P0 and P1), and these are autoregulated by Pax6. Additionally, Pax6 expression is under the control of the TGFbeta superfamily, although the precise mechanisms of such regulation are not understood. The effect of TGFbeta on Pax6 expression was studied in the FHL124 lens epithelial cell line and was found to cause up to a 50% reduction in Pax6 mRNA levels within 24 h. Analysis of luciferase reporters showed that Pax6 autoregulation of the P1 promoter, and its induction of a synthetic promoter encoding six paired domain-binding sites, were significantly repressed by both an activated TGFbeta receptor and TGFbeta ligand stimulation. Subsequently, a novel Pax6 binding site in P1 was shown to be necessary for autoregulation, indicating a direct influence of Pax6 protein on P1. In transfected cells, and endogenously in FHL124 cells, Pax6 co-immunoprecipitated with Smad3 following TGFbeta receptor activation, while in GST pull-down experiments, the MH1 domain of Smad3 was observed binding the RED sub-domain of the Pax6 paired domain. Finally, in DNA adsorption assays, activated Smad3 inhibited Pax6 from binding the consensus paired domain recognition sequence. We hypothesize that the Pax6 autoregulatory loop is targeted for repression by the TGFbeta/Smad pathway, and conclude that this involves diminished paired domain DNA-binding function resulting from a ligand-dependant interaction between Pax6 and Smad3.
PubMed ID: 17251190
PMC ID: PMC1807973
Article link: Nucleic Acids Res
Species referenced: Xenopus
Genes referenced: cat.2 hpgds pax6 smad1 smad3 smad4 smad4.2 tgfb1
Article Images: [+] show captions
|Figure 1. TGFβ represses Pax6 expression and autoregulation. (A) TGFβ inhibits Pax6 expression in human lens epithelial cells. FHL124 cells were serum starved for 24 hours and stimulated with 10 ng/ml TGFβ1 or TGFβ2 for a further 24 h. Total RNA was extracted and reverse transcribed. Endogenous Pax6 expression was quantified by quantitative real-time PCR. Data were normalized with mGAPDH control and represent mean +/−SEM (n = 4). (B) Constitutively activated TGFβ receptor I inhibits expression and autoregulation of the Pax6 P1 promoter in reporter assays. HEK-293 cells were transiently transfected with 2 μg of P1-Luc in combination with 5 μg of caTGFβRI or Pax6. Cells were lysed 48 h post-transfection and luciferase assays were performed. The data represent mean +/−SEM (n = 9). (C) Constitutively activated TGFβ receptor I inhibits Pax6 protein function in reporter assays. HEK-293 cells were transiently transfected with 2 μg of P6CON-Luc in combination with 5 μg of caTGFβRI or Pax6. Cells were lysed 48 h post-transfection and luciferase assays were performed. The data represent mean +/−SEM (n = 4). (D) Time-course-dependent repression of P1-Luc and P6CON-Luc in FHL124 cells by TGFβ ligand stimulation. FHL124 cells were transfected with 1 μg of Pax6-P1-Luc or P6CON-Luc using Lipofectamine 2000, serum-starved overnight, and stimulated with 5 ng/ml TGFβ for the times indicated. Cells were lysed and luciferase assays were performed (n = 3).|
|Figure 2. Pax6 paired domain mediates direct autoregulation of promoter P1. (A) An isolated Pax6 paired domain (PD) inhibits the function of full-length Pax6 in dominant negative manner. HEK-293 cells were transiently transfected with 2 μg of P6CON-Luc in combination with 5 μg of Pax6 or PD. Cells were lysed 48 h post-transfection and luciferase assays were performed. The data represent mean +/−SEM (n = 3). (B) Dominant negative paired domain (PD) inhibits the basal activity of promoter P1 indicating an autoregulatory component. HEK-293 cells were transiently transfected with 2 μg of P1-Luc in combination with 5 μg of PD. Cells were lysed 48 h post-transfection and luciferase assays were performed. The data represent mean +/−SEM (n = 5). (C) Alignment of the human and Xenopus Pax6 P1 promoters reveals an evolutionary conserved putative paired domain-binding site. Sequences of the human and Xenopus Pax6 P1 promoters were aligned in the region corresponding to the reporter construct P1-Luc. This alignment was then aligned to the consensus-paired domain binding sequence, P6CON, which is boxed (Note: this is the reverse-complement P6CON sequence as the putative binding site is encoded 5′ to 3′ on the opposite strand). A broken arrow indicates orientation of the P6CON consensus. Nucleotides matching the majority sequence are shaded in grey, while the TATA and CCAAT boxes are shaded in black. Bold letters indicate the location of PCR primers used to generate DNA probes for gel-shift assays, and solid lines indicate primer orientation. Human transcription start site is indicated as +1. (D) Deletion of a putative paired domain binding site in promoter P1 (P1(▵PBS)-Luc) disrupts Pax6 autoregulation. HEK-293 cells were transiently transfected with 2 μg of P1(▵PBS)-Luc in combination with 5 μg of Pax6. Cells were lysed 48 h post-transfection and luciferase assays were performed. The data represent mean +/−SEM (n = 5). (E) Gel mobility shift assay of GST-Pax6-PD and GST-Pax6-HD binding to the P1 promoter. Equal amounts of GST-Pax6-PD and GST-Pax6-HD were used as described in the materials and methods. Probe 1 was amplified using the PCR primers FL-P1-F and FL-P1-R, and Probe 2 using S-P1-F and FL-P1-R as indicated in the P1 sequence shown in Figure 2C. The actual primer sequences used here are provided in the materials and methods. The migration of the major shifted bands is indicated as (*).|
|Figure 3. Interaction of Smads with Pax6. (A) Pax6 interacts with Smad1, 3, 4 and 5 in vitro. Lysates were prepared from HEK-293 cells that had been transiently transfected with p3xFlag-Pax6. These were then incubated with GST-Smad proteins bound onto glutathione beads. Following extensive washing, Pax6 associated with the Smad-GSTs was identified by elution of beads with SDS-Laemmli buffer, separation by 10% SDS-PAGE, and western blotting using a specific Pax6 antibody (Santa Cruz, cat. No. SC-20). GST alone was used as a control. The presence of the GST proteins was confirmed by staining gels with Coomassie Blue (C/Blue). (B) Pax6 interacts with Smad3 in the presence of constitutively activated TGFβ receptor I in vivo. HEK-293 cells were transfected as indicated. Pax6-Flag was immunoprecipitated with anti-FLAG antibody. Samples were separated by 10% SDS-PAGE and immunoblotted with anti-GFP antibody. (C) Pax6 and Smad3 interact endogenously in response to TGFβ stimulation of human lens epithelial cells.|
|Figure 4. The MH1 domain of Smad3 interacts primarily with the paired domain of Pax6. (A) Lysates were prepared from HEK-293 cells that had been transiently transfected with pcDNA3-Pax6. These were then incubated with GST-Smad truncated proteins as indicated in the left-hand schematic diagram and, following extensive washing, associated proteins were identified by elution of beads with SDS-Laemmli buffer, separation by 10% SDS-PAGE, and western blotting using a specific Pax6 antibody. The presence of the GST proteins was confirmed by staining gels with Coomassie Blue (C/Blue). (B) Pax6 constructs used for in vitro translation are shown in the left-hand panel. GST pull-down assays were performed with full-length Smad3 and different domains of Smad3 fused to GST and immobilized on glutathione-agarose beads and Pax6, Pax6ΔHD and Pax6ΔPD produced by in vitro transcription and translation in the presence of [35S]-methionine. Ten-microliter portions of the in vitro translation reactions were preincubated with GST immobilized on glutathione-agarose beads before incubation with the GST fusion proteins. The GST beads, GST-Pax6ΔHD beads and GST-Pax6ΔPD beads were washed several times before they were boiled and run on a 10% SDS–polyacrylamide gel. Two microlitres of the in vitro translated proteins were run on the same gel to visualize the signal from 20% of the input as shown in the middle panel. (C) GST-Pax6 sub-paired domain constructs are shown in the left-hand schematic panel. The right panel shows the results of GST pull-down assays with the paired domain (PD) of Pax6 and the two sub-domains, PAI and RED fused to GST and immobilized on glutathione–agarose beads and Smad3 or Smad4 produced by in vitro transcription and translation in the presence of [35S]-methionine. Samples were prepared and separated as described above in Section B.|
|Figure 5. Smad3 prevents Pax6 paired domain from binding DNA. (A) Schematic representation of the experimental protocol for the DNA absorption assay together with the sequence of the biotinylated P6CON aptamer. (B) Biotinylated double-stranded P6CON oligonucleotide immobilized on streptavidin beads was incubated with lysates of HEK-293 cells transfected with the indicated expression plasmids. Pax6 bound to P6CON was analysed by immunoblotting using anti-Flag antibodies. The lower panels show the expression levels of Pax6-Flag and Smad3-HA proteins as analysed in immunoblots of the cell lysates.|
|Figure 6. Summary of the mechanism for the repression of Pax6 function by Smads. Our results show that TGFβ represses Pax6 promoter activity by inducing the nuclear translocation of Smad3. (A) Once in the nucleus, Smad3 interacts with the RED sub-domain of the paired domain in Pax6 and releases Pax6 from its DNA binding site. (B) Thus, the Smad/TGFβ signalling pathway turns off Pax6 expression by preventing it from autoregulating its own promoter.|
References [+] :
Aota, Pax6 autoregulation mediated by direct interaction of Pax6 protein with the head surface ectoderm-specific enhancer of the mouse Pax6 gene. 2003, Pubmed