Trans-activation activity of HOXB5 on RET promoter was assayed by measuring the luciferase activity from human RET promoter luciferase reporter construct. Luciferase activity was normalized with Renilla luciferase to obtain relative luciferase unit. Data are shown as the fold increase (mean±SD) in relative luciferase unit compared to that of pRC/CMV.

A, EMSA assay with D2 and nuclear extract of HOXB5 transfected SK-N-SH cells or purified GST-HOXB5 protein. Binding of D2 as indicated by retarded migration of D2 biotin probe (arrow) was specifically detected in lanes with HOXB5 transfected SK-N-SH nuclear extract or purified HOXB5 protein. Addition of excess unlabeled D2 probe (D2 cold probe) either reduced or completely abolished the binding of labeled D2 probe. B, ChIP was performed with HOXB5 transfected SK-N-SH cells, and binding of D2 to HOXB5 was determined by real-time quantitative PCR using UD1 forward and UD1 reverse primer. x-axis indicated fold enrichment normalized to the control IgG. C, Binding of D2A oligonucleotide to purified HOXB5 protein as indicated by retarded migration of D2A probe. Addition of excess unlabeled D2A probe reduced the binding markedly. No competition was observed with the addition of non-specific unlabled probe, and no binding was detected with GST protein.

A, Immuno-precipitation (IP) on nuclear extract of HOXB5 transfected SK-N-SH cells with αHOXB5 antibody. Protein complexes precipitated with αHOXB5 or non-specific IgG were resolved by SDS-PAGE, and NKX2-1 in the complex was detected by immuno-blotting using αNKX2-1 antibody. NKX2-1 (38KD) was specifically detected in the lane of protein complex precipitated with αHOXB5. B, Chromatin immuno-precipitation (ChIP) of HOXB5 transfected SK-N-SH cells and αHOXB5 or αNKX2-1 antibody. D2 was amplified from chromatin precipitated with αHOXB5, and αNKX2-1, but not in the preparation with non-specific IgGs (mIG; rIG). Bindings of HOXB5 and NKX2-1 to D2 were detected by real-time quantitative PCR. x axis is fold enrichment normalized to the non-immune mouse IgG control.

A, Schematic diagram of the human RET promoter. “+1” denoted transcription start of RET gene, and the first ATG of RET gene was indicated. The 5′ ends of sequential deletion constructs of RET promoter were indicated with nucleotide positions. Location of the D2 element was indicated. B, Nucleotide sequence of the human RET gene promoter was shown with the locations of various primers used for the generation of deletion constructs. Primers pair used for the generation of D2 PCR fragment for EMSA, and for the PCR amplification in ChIP assays were indicated. The PCR primers (UD1 forward and reverse primer) used for real-time quantitative PCR for the binding of D2 to HOXB5 were indicated with ↔. The D2A oligonucleotide probe (bold and italic) which was specifically bound by HOXB5 as shown by EMSA was underlined with dotted arrow, and the putative HOXB5 binding site in D2 was indicated by filled boxed. NKX2-1 binding site was boxed, and the two SNPs affecting NKX2-1 binding were indicated with downwards arrows. Putative binding sites of other transcription factors were boxed and labeled accordingly. Abbreviations: TFIIB, transcription factor II B; SREBP, sterol regulatory element-binding protein; MTE, human motif ten element. C, Trans-activation of HOXB5 on full-length and 5′ deletion RET promoter was assayed by measuring the luciferase activity from RET promoter luciferase reporter constructs. Luciferase activity was normalized with Renilla luciferase to obtain relative luciferase unit. Percentage trans-activation of deletion constructs was determined by comparing luciferase activity to that of full-length promoter construct, which was arbitrarily regarded as 100%. D, Trans-activation of HOXB5 from D2 was assayed by luciferase activity from a D2-SV40 promoter luciferase reporter. Luciferase activity was normalized with Renilla luciferase to obtain relative luciferase unit. Fold increase (mean±SD) was determined relative to luciferase unit of pRC/CMV which was arbitrarily regarded as 1.

A, Wild-type or mutated (SNPs -5G>A and -1A>C) or (-1A>C) full-length RET promoter was transfected with HOXB5 or NKX2-1 alone or in combination into SK-N-SH cells. Relative luciferase units for each combinations were determined, and fold increase (mean±SD) was calculated as compared to luciferase unit of empty vector (pXP1 Basic), which was arbitrarily regarded as 1. B, Expression levels of NKX2-1 mRNA were determined by real-time RT-PCR quantitation in non-transfected SH-N-SH cell and SH-N-SH cell transfected with NKX2-1 siRNA or negative control siRNA (Ctrl siRNA). Relative expressions of NKX2-1 were calculated with reference to non-transfected cell, which was arbitrarily regarded as 100%. C, full-length RET promoter was transfected with HOXB5 alone or in combination with NKX2-1 siRNA or negative control siRNA (Ctrl siRNA). Relative luciferase units for each combination were determined, and fold increase (mean±SD) was calculated as compared to luciferase unit of empty vector (pRC/CMV), which was arbitrarily regarded as 1.