SOCS3 gene is an early direct target of Sox6. (A) Mapping of the SOCS3 conserved region that contained the double Sox6 binding site. University of California Santa Cruz (UCSC) map indicates a block of conservation of approximately 100 nt centered on the Sox6 site. The double site was fully conserved in rat, mouse, and human. (B) Sox6 was bound in vitro to the SOCS3 enhancer in electrophoretic mobility shift assay. Nuclear extracts from Sox6-K562 (lanes 1-9) or from EV-K562 (lane 10) were incubated with either the WT (lanes 2-10) or the mutated (lane 1; MUT) SOCS3-labeled probes. The retarded band generated by the Sox6-FLAG protein was supershifted by the anti-FLAG antibody (lane 2) and competed for both by the SOCS3 oligonucleotide itself (lanes 4-5) and by a known Sox6 consensus (lanes 8-9; Col2a1²⁹ ). The SOCS3 probe mutated in the Sox6 consensus failed to give any binding (lane 1) or to compete for the Sox6 band (lanes 6-7). The WT probe alone (lane 11) yielded no bands. Ns indicates no specific binding. (C) ChIP on chromatins from transduced K562 cells. The SOCS3 enhancer element was immunoprecipitated by the anti-FLAG antibody (which recognizes the Sox6 transduced protein) but not by the corresponding preimmune serum, immunoglobulin G (IgG). (Top panel) SOCS3 region; (bottom panel) glyceraldehyde-3-phosphate dehydrogenase (GAPDH) locus, as negative control. Lanes 1 and 2 show input chromatins; lanes 3-4, IgG; lanes 5-6, anti-FLAG antibody; and lane 7, water. EV indicates empty vector; Sox6, Sox6 vector. (D) The SOCS3 enhancer containing the Sox6 double site was activated by Sox6 in cotransfection experiments in K562 cells. A 70-nt fragment containing either the WT (EnhWT) or the mutated Sox6 double site (EnhMut) was cloned upstream of the −393 + 12 region that corresponded to the SOCS3 minimal promoter (SOCS3prom). (Left panel) Mutations in the Sox6 consensus were the same as in panel B and were proven to abolish Sox6 binding in an electrophoretic mobility shift assay. (Right panel) All constructs were cotransfected in K562 cells together with increasing amounts (0.2 μg and 1 μg) of a Sox6-expressing plasmid. The EnhWT construct was activated in a dose-dependent manner by the addition of the cotransfected Sox6 plasmid (solid columns), whereas the corresponding mutated element, EnhMut, was insensitive to Sox6 cotransfection (open columns on right). (E) Real-time quantification of SOCS3 expression on Sox6 transduction in K562 (3 and 72 hours after transduction) and in primary cord blood–derived (CB) erythroblasts (48 and 96 hours after transduction). (F) SOCS3 expression during CD34⁺-derived erythroblast terminal maturation, determined by real-time PCR. (G top panel) Real-time quantification of Sox6 expression at day 9 and day 12. (Bottom panel) ChIP performed on chromatins from primary cells on the same days as above demonstrated that at day 12, Sox6 was bound to the SOCS3 enhancer. IgG antibodies were used as a control. Statistical significance is indicated above the chart.