Identification of human STAT5-dependent gene regulatory elements based on interspecies homology - PubMed (original) (raw)

Identification of human STAT5-dependent gene regulatory elements based on interspecies homology

Erik A Nelson et al. J Biol Chem. 2006.

Abstract

STAT5 is a transcription factor essential for hematopoietic physiology. STAT5 functions to transduce signals from cytokines to the nucleus where it regulates gene expression. Although several important transcriptional targets of STAT5 are known, most remain unidentified. To identify novel STAT5 targets, we searched chromosomes 21 and 22 for clusters of STAT5 binding sites contained within regions of interspecies homology. We identified four such regions, including one with tandem STAT5 binding sites in the first intron of the NCAM2 gene. Unlike known STAT5 binding sites, this site is found within a very large intron and resides approximately 200 kb from the first coding exon of NCAM2. We demonstrate that this region confers STAT5-dependent transcriptional activity. We show that STAT5 binds in vivo to the NCAM2 intron in the NKL natural killer cell line and that this binding is induced by cytokines that activate STAT5. Neither STAT1 nor STAT3 bind to this region, despite sharing a consensus binding sequence with STAT5. Activation of STAT4 and STAT5 causes the accumulation of both of these STATs to the NCAM2 regulatory region. Therefore, using an informatics based approach to identify STAT5 targets, we have identified NCAM2 as both a STAT4- and STAT5-regulated gene, and we show that its expression is regulated by cytokines essential for natural killer cell survival and differentiation. This strategy may be an effective way to identify functional binding regions for transcription factors with known cognate binding sites anywhere in the genome.

PubMed Disclaimer

Figures

FIGURE 1

FIGURE 1. NCAM2 genomic structure

The upper panel shows a schematic of the NCAM2 gene with the approximate locations of exons (boxes) and introns (horizontal lines). The location of the STAT5 consensus binding sites is indicated with an arrow. The distances of the STAT5 consensus sites to the first and second exons are noted below the gene structure. The lower panel shows the exact sequence identified in the homology screen using a screen capture of the UCSC genome browser and shows the sequence homology between five different species. The STAT5 consensus sites are boxed. The base pair numbers and the genomic assembly from the UCSC genome browser for each species is as follows: human, chr21:21,494,503–21,494,577 (May 2004); chimpanzee, chr22: 21,644,947–21,645,021 (November 2003); mouse, chr16:81,746,580–81,746,651 (May 2004); rat, chr11:20,599,259–20,599,330 (June 2003); dog, chr31: 20,113,821–20,113,895 (July 2004).

FIGURE 2

FIGURE 2. STAT5a and STAT5b activation by IL-2 leads to NCAM2 expression in NKL cells

a, starved NKL cells were either treated with IL-2 for 15 min or left untreated. Immunoprecipitation of STAT5a and STAT5b using specific antibodies was performed, followed by Western blot analysis with the indicated antibodies. b, starved NKL cells were treated with IL-2 for the indicated times. NCAM2 expression was analyzed by quantitative real time PCR. GAPDH was used as an invariant control. c, starved NKL cells were treated for 7 h with IL-2 at the indicated concentrations (units/ml). NCAM2 expression was analyzed by quantitative real time PCR. Each RT-PCR experiment was repeated two times.

FIGURE 3

FIGURE 3. The NCAM2 intronic region confers transcriptional activity that depends on the STAT5 binding sites

a, schematic of the luciferase construct containing the NCAM2 intronic region. Numbering is in reference to the first nucleotide of the NCAM2 mRNA sequence (accession number U75330). The wild type (wt) consensus binding sites are indicated as well as the nucleotides mutated in each construct. b, T47D cells were transfected with empty vector (pLuc) or with a luciferase construct containing the NCAM2 intronic region (_NCAM2_-Luc) and then were stimulated with prolactin overnight or were left untreated, and luciferase activity was measured. c, 293 cells were transfected with the NCAM2 intronic region together with either an empty vector (_NCAM2_-Luc/vector) or a STAT5a1*6 expression construct (_NCAM2_-Luc/STAT5a1*6), and luciferase activity was quantitated 24 h after transfection. d, T47D cells were transfected with either the wild type NCAM2 intronic element (_NCAM2_-Luc) or the NCAM2 intronic element with mutations in STAT5 consensus site 1 (Mut1-Luc), STAT5 consensus site 2 (Mut2-Luc), or both STAT5 consensus sites (Mut1 + 2-Luc) and either left untreated (light bars) or treated with prolactin (dark bars). Luciferase activity was determined and was normalized to untreated for each construct. Each luciferase assay was repeated four times.

FIGURE 4

FIGURE 4. STAT5a, STAT5b, and components of the basal transcriptional machinery bind to the NCAM2 intronic element in vivo

a, ChIP assays were performed on NKL cells that had been starved and then left untreated or stimulated with IL-2 for 30 min. Immunoprecipitations were performed using antibodies (Ab) to STAT5a, STAT5b, or a nonspecific antibody (IgG). The upper panel shows PCR analysis using primers specific for the NCAM2 intronic region, whereas the lower panel shows PCR analysis using primers specific for a control region. The right panels show PCR of input DNA. b, ChIP analysis was performed using antibodies to Pol II, p300, total STAT5, or a nonspecific IgG. PCR was performed on the NCAM2 intronic region using both ChIP product (left) and input DNA (right). Each ChIP experiment was repeated two times.

FIGURE 5

FIGURE 5. Neither STAT1 nor STAT3 bind to the NCAM2 STAT binding region

ChIP assays were performed on NKL cells that had been starved and then left untreated or stimulated with IL-2 for 30 min. Immunoprecipitations were performed using antibodies (Ab) specific for STAT1, STAT3, STAT5, or a nonspecific antibody (IgG). DNA was quantitated by real time PCR and was normalized to input and expressed relative to nonspecific IgG. Each ChIP experiment was repeated three times.

FIGURE 6

FIGURE 6. IFN-α induces expression of NCAM2

a, starved NKL cells were either treated with the indicated cytokine for 15 min or left untreated. Western blots were performed with the indicated antibodies. b, starved NKL cells were treated with IL-2 or IFN-α for 15 min or left untreated. Immunoprecipitation was performed using an antibody to STAT4. Western blot analysis was performed against phosphotyrosine (P-Tyr) or STAT4. c, starved NKL cells were treated with the indicated cytokines for 7 h, and NCAM2 expression was analyzed by quantitative real time PCR. GAPDH was used as an invariant control. d, starved NKL cells were treated for the indicated times with IFN-α. NCAM2 expression was analyzed by quantitative real time PCR. GAPDH was used as an invariant control. Each RT-PCR experiment was repeated two times.

FIGURE 7

FIGURE 7. STAT4 and STAT5 bind to the NCAM2 intronic region after IFN-α stimulation

ChIP assays were performed on NKL cells that had been starved and then left untreated or had been stimulated with IFN-α for 30 min. Immunoprecipitations were performed using antibodies specific for STAT4, STAT5, or a nonspecific antibody (IgG). DNA was quantitated by real time PCR and was normalized to input and expressed relative to nonspecific IgG. Each ChIP experiment was repeated three times.

FIGURE 8

FIGURE 8. IL-2 and IFN-α synergize to regulate NCAM2 expression

a, starved NKL cells were treated with the indicated cytokines for 7 h. NCAM2 expression was analyzed by quantitative real time PCR. GAPDH was used as an invariant control. b and c, starved NKL cells were treated with the indicated cytokine for 2 h. The expression of IFN-γ (b) and CIS (c) was analyzed by quantitative real time PCR. GAPDH was used as an invariant control. d, ChIP assays were performed on NKL cells that had been starved and then left untreated or stimulated with the indicated cytokines for 30 min. Chromatin immunoprecipitations were performed using antibodies specific for STAT4, STAT5, or a nonspecific antibody (IgG). Data were normalized to input and expressed relative to nonspecific IgG. Each RT-PCR and ChIP experiment was repeated two times.

Similar articles

Cited by

References

    1. Bowman T, Garcia R, Turkson J, Jove R. Oncogene. 2000;19:2474–2488. - PubMed
    1. Frank DA. Cancer Treat. Res. 2003;115:267–291. - PubMed
    1. Sternberg DW, Gilliland DG. J. Clin. Oncol. 2004;22:361–371. - PubMed
    1. Horita M, Andreu EJ, Benito A, Arbona C, Sanz C, Benet I, Prosper F, Fernandez-Luna JL. J. Exp. Med. 2000;191:977–984. - PMC - PubMed
    1. Grimley PM, Dong F, Rui H. Cytokine Growth Factor Rev. 1999;10:131–157. - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources