Ments using nuclear proteins from cells expressing the mouse GH receptor and wild-type Stat5b after GH treatment. FP = unbound probe. The arrow indicates protein-DNA complexes. Right panels: binding curves with Kds listed (mean 6 S.E., n = 3 
experiments). doi:10.1371/journal.pone.0050278.gR53?4 or R13?3.5 (Fig. 1B). To test the hypothesis that `inactive’ Stat5b could either differentially activate or inhibit target gene transcription via individual Stat5b responsive elements, studies were performed in the absence of GH, using expression plasmids encoding either previously-validated wild type (WT), dominant-negative (DN), or constitutively-active (CA) Stat5b [31], and Igf1 promoter 2 – reporter genes containing individual intact enhancers or enhancers in which all Stat5b binding sites weredisrupted by point mutations. For 4 of the native enhancer promoter – reporter plasmids tested, `inactive’ Stat5bWT and Stat5bDN had little differential effect on gene transcription, although in all cases Stat5bCA was stimulatory by 3-8-fold (Fig. 3A, R2?, R13, R34?5, R53?4). The exceptions were R57?9 and R60?1, in which `inactive’ Stat5bWT was able to drive promoter function to 3?-fold higher levels than Stat5bDN, although only to ,25 of the values obtained with Stat5bCAFigure 5. Defining a hierarchy of binding affinities of Stat5b for individual DNA sites within the rat Igf1 locus. A. Gel-mobility shift experiments were performed with the Cy5.5-labeled double-stranded probe R34, 2 mg of nuclear protein from Cos-7 cells transfected with expression plasmids for the mouse GH receptor and rat Stat5b, and incubated with rat GH [40 nM] for 1 h, and various concentrations of competitor DNAs as indicated. Two representative individual competition experiments are shown. The arrow indicates the location of protein-DNA complexes (NS, no Stat5b in nuclear protein extract, FP = unbound probe). B. The graph illustrates results of competition experiments for 4 different unlabeled doublestranded competitor DNAs (mean 6 S.E., n = 3 independent experiments, with 4 data points/experiment). C. Results for all probes have been tabulated (n = 3 independent experiments, with 4 data points/experiment) and are presented as IC50 values (DNA concentration at which binding of labeled probe is reduced to 50 of starting value). The 95 confidence interval (CI) also is indicated and each Stat5b core DNA binding sequence is listed. doi:10.1371/journal.pone.0050278.gDefining GH-Activated Stat5b Enhancersreporter genes with mutated enhancer elements (Fig. 3A). Levels of expression of transfected Stat5bWT, Stat5bDN, and Stat5bCA were nearly identical (Fig. 3B), but examination of their sub-cellular location in the absence of GH treatment showed that Stat5bCA was found in the cytoplasm and nucleus and was tyrosine phosphorylated, that Stat5bDN was in the cytoplasm, and that a small amount of Stat5bWT was nuclear and tyrosine ML 281 phosphorylated (Fig. 3C and D). Taken together, these results demonstrate a selective transcriptional stimulatory effect of Stat5b on 2 of 6 Stat5b-responsive enhancers in the absence of GH-induced activation, implying that individual Igf1 locus Stat5b-regulated responsive elements have different functional properties.DNA Binding Strength and Transcriptional Methionine enkephalin site FunctionQuantitative in vitro DNA-protein binding experiments [31] revealed a ,15-fold difference in affinities of GH-activated wildtype Stat5b for the 3 different Stat5 sites studied with this method: R58, R3.Ments using nuclear proteins from cells expressing the mouse GH receptor and wild-type Stat5b after GH treatment. FP = unbound probe. The arrow indicates protein-DNA complexes. Right panels: binding curves with Kds listed (mean 6 S.E., n = 3 experiments). doi:10.1371/journal.pone.0050278.gR53?4 or R13?3.5 (Fig. 1B). To test the hypothesis that `inactive’ Stat5b could either differentially activate or inhibit target gene transcription via individual Stat5b responsive elements, studies were performed in the absence of GH, using expression plasmids encoding either previously-validated wild type (WT), dominant-negative (DN), or constitutively-active (CA) Stat5b [31], and Igf1 promoter 2 – reporter genes containing individual intact enhancers or enhancers in which all Stat5b binding sites weredisrupted by point mutations. For 4 of the native enhancer promoter – reporter plasmids tested, `inactive’ Stat5bWT and Stat5bDN had little differential effect on gene transcription, although in all cases Stat5bCA was stimulatory by 3-8-fold (Fig. 3A, R2?, R13, R34?5, R53?4). The exceptions were R57?9 and R60?1, in which `inactive’ Stat5bWT was able to drive promoter function to 3?-fold higher levels than Stat5bDN, although only to ,25 of the values obtained with Stat5bCAFigure 5. Defining a hierarchy of binding affinities of Stat5b for individual DNA sites within the rat Igf1 locus. A. Gel-mobility shift experiments were performed with the Cy5.5-labeled double-stranded probe R34, 2 mg of nuclear protein from Cos-7 cells transfected with expression plasmids for the mouse GH receptor and rat Stat5b, and incubated with rat GH [40 nM] for 1 h, and various concentrations of competitor DNAs as indicated. 
Two representative individual competition experiments are shown. The arrow indicates the location of protein-DNA complexes (NS, no Stat5b in nuclear protein extract, FP = unbound probe). B. The graph illustrates results of competition experiments for 4 different unlabeled doublestranded competitor DNAs (mean 6 S.E., n = 3 independent experiments, with 4 data points/experiment). C. Results for all probes have been tabulated (n = 3 independent experiments, with 4 data points/experiment) and are presented as IC50 values (DNA concentration at which binding of labeled probe is reduced to 50 of starting value). The 95 confidence interval (CI) also is indicated and each Stat5b core DNA binding sequence is listed. doi:10.1371/journal.pone.0050278.gDefining GH-Activated Stat5b Enhancersreporter genes with mutated enhancer elements (Fig. 3A). Levels of expression of transfected Stat5bWT, Stat5bDN, and Stat5bCA were nearly identical (Fig. 3B), but examination of their sub-cellular location in the absence of GH treatment showed that Stat5bCA was found in the cytoplasm and nucleus and was tyrosine phosphorylated, that Stat5bDN was in the cytoplasm, and that a small amount of Stat5bWT was nuclear and tyrosine phosphorylated (Fig. 3C and D). Taken together, these results demonstrate a selective transcriptional stimulatory effect of Stat5b on 2 of 6 Stat5b-responsive enhancers in the absence of GH-induced activation, implying that individual Igf1 locus Stat5b-regulated responsive elements have different functional properties.DNA Binding Strength and Transcriptional FunctionQuantitative in vitro DNA-protein binding experiments [31] revealed a ,15-fold difference in affinities of GH-activated wildtype Stat5b for the 3 different Stat5 sites studied with this method: R58, R3.
