Transcriptional activation of TINF2, a gene encoding the telomere-associated protein TIN2, by Sp1 and NF-κB factors

Abstract

The expression of the telomere-associated protein TIN2 has been shown to be essential for early embryonic development in mice and for development of a variety of human malignancies. Recently, germ-line mutations in TINF2, which encodes for the TIN2 protein, have been identified in a number of patients with bone-marrow failure syndromes. Yet, the molecular mechanisms that regulate TINF2 expression are largely unknown. To elucidate the mechanisms involved in human TINF2 regulation, we cloned a 2.7 kb genomic DNA fragment containing the putative promoter region and, through deletion analysis, identified a 406 bp region that functions as a minimal promoter. This promoter proximal region is predicted to contain several putative Sp1 and NF-κB binding sites based on bioinformatic analysis. Direct binding of the Sp1 and NF-κB transcription factors to the TIN2 promoter sequence was demonstrated by electrophoretic mobility shift assay (EMSA) and/or chromatin immunoprecipitation (ChIP) assays. Transfection of a plasmid carrying the Sp1 transcription factor into Sp-deficient SL2 cells strongly activated TIN2 promoter-driven luciferase reporter expression. Similarly, the NF-κB molecules p50 and p65 were found to strongly activate luciferase expression in NF-κB knockout MEFs. Mutating the predicted transcription factor binding sites effectively reduced TIN2 promoter activity. Various known chemical inhibitors of Sp1 and NF-κB could also strongly inhibit TIN2 transcriptional activity. Collectively, our results demonstrate the important roles that Sp1 and NF-κB play in regulating the expression of the human telomere-binding protein TIN2, which can shed important light on its possible role in causing various forms of human diseases and cancers.

Figure 1. Luciferase assay analysis of TINF2 promoter truncations defines two significant drops in promoter activity.

A: The name of each TIN2 reporter construct was assigned according to the 5′-end nucleotide numbers of the promoter sequences inserted upstream of the ATG initiation codon. Basic refers to the pGL3-Basic vector. For each transfection, the firefly luciferase activity was normalized to the β-galactosidase activity expressed from a co-transfected β-galactosidase expression vector. The means from three independent experiments are shown for each construct; bars, SD. (***p<0.001). B: Finer promoter mapping of the region between −450 and −351, expressed as in panel A. C: Finer promoter mapping of the region between −148 and −24, expressed as in panel A. D: Verification of the functionality of the TINF2 promoter construct in various cell lines. For each transfection, the firefly luciferase activity was normalized to the Renilla activity expressed from the co-transfected pRL-CMV expression vector and expressed as in panel A. E: Schematic of the core TIN2 promoter construct showing putative Sp1 and NF-κB binding sites predicted using bioinformatic tools. “Drop 1” and “Drop 2” refer to the drops in luciferase reporter activity shown in panel A.

Figure 2. Sp1 binds to putative binding sequences in the TIN2 promoter in vitro and in vivo.

A: EMSA showing the ability of Sp1 to bind to its two predicted sites in vitro. Sp1 protein was in vitro translated using the rabbit reticulocyte lysate (RRL) system and incubated with end-labeled DNA oligos encompassing the putative binding sites. These complexes (lanes 3 and 9) could be competed away with unlabeled wild-type oligos (lanes 4 and 10), but not with a mutant form (lanes 5 and 11). Furthermore, the complexes could be super-shifted using an anti-Sp1 antibody (lanes 6 and 12). B: The ability of Sp1 to bind to the endogenous TINF2 promoter in vivo was shown using ChIP. The 407 base pair fragment could be amplified from a reaction including the anti-Sp1 antibody (lane 7), but not from reactions containing anti-c-Myc, normal IgG, or no antibody (lanes 4–6). C: Sp1 is the major Sp transcription factor that can activate the TINF2 promoter. Drosophila melanogaster SL2 cells were co-transfected with the minimal promoter reporter construct pGL3-P406 and various amounts of expression vectors encoding either Sp1 or Sp3. For each transfection, the firefly luciferase activity was normalized to the total protein concentration. The means from three independent experiments are shown; bars, SD.

Figure 3. NF-κB can bind to and activate the TINF2 promoter.

A: The ability of NF-κB to bind to the endogenous TINF2 promoter in vivo was shown using ChIP. The 407 base pair fragment could be amplified from a reaction including the anti-p65 antibody (lane 5), but not from reactions containing no antibody or normal IgG (lanes 3 and 4). B: The ability of NF-κB to activate the minimal TINF2 promoter was verified by co-transfection of vectors encoding p50 and/or p65 protein and pGL3-P406 into NIH 3T3 (p50⁻/p65⁻) cells. For each transfection, the firefly luciferase activity was normalized to the total protein concentration. The means from three independent experiments are shown; bars, SD. (***p<0.001).

Figure 4. Mutating the putative binding sites abolishes Sp1 and NF-κB transcriptional activation of the TINF2 promoter.

A: Mutating core consensus nucleotides in predicted Sp1 or NF-κB binding sites (see Materials and Methods) results in reduced TINF2 promoter-driven luciferase activity in a minimal promoter context. For each transfection, the firefly luciferase activity was normalized to the Renilla activity expressed from the co-transfected pRL-CMV expression vector. The means from three independent experiments are shown; bars, SD. (p<0.001). B: Mutating core consensus nucleotides in the predicted Sp1 binding site at position −(88−74) results in reduced TINF2 promoter-driven luciferase activity as compared to wild-type control. Results are shown as in panel A. (***p<0.001).

Figure 5. Pharmacological inhibitors can interfere with Sp1- and NF-κB-mediated TINF2 promoter activation.

A: Mithramycin A, an Sp1 inhibitor, reduces TINF2 promoter-driven luciferase activity. 293T cells are transfected with either the promoter-less pGL3-Basic plasmid, the NFAT-responsive pNFAT-Luc plasmid, the SV40 promoter-containing pGL3-Control plasmid, or the minimal TIN2 promoter P406 plasmid and incubated with the indicated concentrations of drug for 24 hours. For each transfection, the firefly luciferase activity was normalized to Renilla luciferase activity expressed from a co-transfected Renilla luciferase expression vector. The means from three independent experiments are shown; bars, SD. (**p<0.01, ***p<0.001). B: Bay11-7082, an NF-κB inhibitor, reduces TINF2 promoter-driven luciferase reporter activity. Cells are processed and values expressed as in panel A. (***p<0.001). C: PDTC, an NF-κB inhibitor, reduces TINF2 promoter-driven luciferase reporter activity. Cells are processed and values expressed as in panel A. (***p<0.001). D. Mithramycin A and Bay11-7082 (Bay11) reduce endogenous TINF2 gene expression. The levels of TINF2 mRNA were normalized to those of the housekeeping GAPDH gene. The means from three independent experiments are shown.