Isolation of Unique STAT5 Targets by Chromatin Immunoprecipitation-based Gene Identification

STAT5a and STAT5b are two highly related transcription factors that control essential cellular functions. Several STAT5 targets are known, although it is likely that most remain uncharacterized. To identify a more complete set of STAT5-regulated genes, we used a modification of the chromatin immunoprecipitation procedure, which does not presuppose any information regarding these targets. Employing Ba/f3 cells in which STAT5 is activated by interleukin-3, we have identified novel STAT5 binding sites that may be regulatory regions for nearby genes. These sites are typically found far from transcription start sites, and most do not contain CpG islands, indicating that they are not in traditional promoter regions. Nonetheless, when the expression of genes near these STAT5 binding sites was examined, all were expressed in Ba/f3 cells, and most were modulated by interleukin-3. Furthermore, genes identified by this strategy show unique expression patterns in acute leukemias, tumors characterized by activated STAT5. Whereas both STAT5 isoforms bound to all promoters tested, STAT5a and STAT5b bound with different kinetics, suggesting that at least some of the differences between the functions of these two proteins are mediated by their DNA binding activity. Therefore, this method of transcription factor target identification represents an effective strategy to isolate transcription factor targets in an unbiased fashion, and it has revealed many novel STAT5-dependent regulatory regions outside of traditional promoters. STAT5a and STAT5b are two highly related transcription factors that control essential cellular functions. Several STAT5 targets are known, although it is likely that most remain uncharacterized. To identify a more complete set of STAT5-regulated genes, we used a modification of the chromatin immunoprecipitation procedure, which does not presuppose any information regarding these targets. Employing Ba/f3 cells in which STAT5 is activated by interleukin-3, we have identified novel STAT5 binding sites that may be regulatory regions for nearby genes. These sites are typically found far from transcription start sites, and most do not contain CpG islands, indicating that they are not in traditional promoter regions. Nonetheless, when the expression of genes near these STAT5 binding sites was examined, all were expressed in Ba/f3 cells, and most were modulated by interleukin-3. Furthermore, genes identified by this strategy show unique expression patterns in acute leukemias, tumors characterized by activated STAT5. Whereas both STAT5 isoforms bound to all promoters tested, STAT5a and STAT5b bound with different kinetics, suggesting that at least some of the differences between the functions of these two proteins are mediated by their DNA binding activity. Therefore, this method of transcription factor target identification represents an effective strategy to isolate transcription factor targets in an unbiased fashion, and it has revealed many novel STAT5-dependent regulatory regions outside of traditional promoters. STAT 1The abbreviations used are: STAT, signal transducers and activators of transcription; ChIP, chromatin immunoprecipitation; RT, reverse transcription; IL-3, interleukin-3; IP, immunoprecipitation; EST, expressed sequence tag; ssa, similar to sperm antigen; Arnt, aryl hydrocarbon receptor nuclear translocator; ssa, similar to sperm antigen. proteins are a family of transcription factors that regulate genes important for growth, differentiation, and apoptosis (1Frank D.A. Cancer Treat. Res. 2003; 115: 267-291Crossref PubMed Google Scholar, 2Bowman T. Garcia R. Turkson J. Jove R. Oncogene. 2000; 19: 2474-2488Crossref PubMed Scopus (1593) Google Scholar). Upon cytokine or growth factor stimulation, STATs become phosphorylated and translocate to the nucleus, where they bind to promoters of their target genes (3Aaronson D.S. Horvath C.M. Science. 2002; 296: 1653-1655Crossref PubMed Scopus (1070) Google Scholar). In addition to essential functions in normal cells, STATs have been implicated in the pathogenesis of numerous solid and hematopoietic malignancies (4Battle T.E. Frank D.A. Blood. 2003; 102: 3016-3024Crossref PubMed Scopus (47) Google Scholar, 5Bromberg J. J. Clin. Invest. 2002; 109: 1139-1142Crossref PubMed Scopus (755) Google Scholar). STAT5 refers to two highly related proteins, STAT5a and STAT5b, which are encoded by separate genes. These proteins are more than 90% identical at the amino acid level (6Grimley P.M. Dong F. Rui H. Cytokine Growth Factor Rev. 1999; 10: 131-157Crossref PubMed Scopus (188) Google Scholar). Despite their similarities, some functional differences exist between these two proteins. For example, there may be differences in the DNA binding affinities of STAT5a and STAT5b (7Boucheron C. Dumon S. Santos S.C. Moriggl R. Hennighausen L. Gisselbrecht S. Gouilleux F. J. Biol. Chem. 1998; 273: 33936-33941Abstract Full Text Full Text PDF PubMed Scopus (74) Google Scholar, 8Verdier F. Rabionet R. Gouilleux F. Beisenherz-Huss C. Varlet P. Muller O. Mayeux P. Lacombe C. Gisselbrecht S. Chretien S. Mol. Cell. Biol. 1998; 18: 5852-5860Crossref PubMed Scopus (131) Google Scholar), although this has not been uniformly seen (9Ehret G.B. Reichenbach P. Schindler U. Horvath C.M. Fritz S. Nabholz M. Bucher P. J. Biol. Chem. 2001; 276: 6675-6688Abstract Full Text Full Text PDF PubMed Scopus (312) Google Scholar). Mice lacking STAT5a or STAT5b show phenotypic differences in both mammary development (10Liu X. Robinson G.W. Wagner K-U. Garrett L. Wynshaw-Boris A. Hennighausen L. Genes Dev. 1997; 11: 179-186Crossref PubMed Scopus (927) Google Scholar, 11Udy G.B. Towers R.P. Snell R.G. Wilkins R.J. Park S-H. Ram P.A. Waxman D.J. Davey H.W. Proc. Natl. Acad. Sci. 1997; 94: 7239-7244Crossref PubMed Scopus (838) Google Scholar) and hematopoiesis (12Imada K. Bloom E.T. Nakajima H. Horvath-Arcidiacono J.A. Udy G.B. Davey H.W. Leonard W.J. J. Exp. Med. 1998; 188: 2067-2074Crossref PubMed Scopus (279) Google Scholar, 13Pestina T.I. Jackson C.W. Exp. Hematol. 2003; 31: 1198-1205Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 14Feldman G.M. Rosenthal L.A. Liu X. Hayes M.P. Wynshaw-Boris A. Leonard W.J. Hennighausen L. Finbloom D.S. Blood. 1997; 90: 1768-1776Crossref PubMed Google Scholar). These studies indicate that there are differences between STAT5a and STAT5b that may result from either their expression patterns or differential target selection. Clarifying this point is an essential step in understanding the biology of STAT5. Whereas there are genes known to be directly regulated by STAT5 (15Silva M. Benito A. Sanz C. Prosper F. Ekhterae D. Nunez G. Fernandez-Luna J.L. J. Biol. Chem. 1999; 274: 22165-22169Abstract Full Text Full Text PDF PubMed Scopus (250) Google Scholar, 16Matsumoto A. Masuhara M. Mitsui K. Yokouchi M. Ohtsubo M. Misawa H. Miyajima A. Yoshimura A. Blood. 1997; 89: 3148-3154Crossref PubMed Google Scholar, 17Matsumura I. Kitamura T. Wakao H. Tanaka H. Hashimoto K. Albanese C. Downward J. Pestell R.G. Kanakura Y. EMBO J. 1999; 18: 1367-1377Crossref PubMed Scopus (294) Google Scholar), most STAT5 targets are unknown. The location of STAT5-dependent promoters has been shown to be close to the transcription start site, such as the STAT5 binding sites in the CIS and cyclin D1 genes (18Rascle A. Lees E. Nucleic Acids Res. 2003; 31: 6882-6890Crossref PubMed Scopus (41) Google Scholar, 19Brockman J.L. Schroeder M.D. Schuler L.A. Mol. Endocrinol. 2002; 16: 774-784Crossref PubMed Scopus (137) Google Scholar). However, the STAT5 binding site in the bcl-x promoter is found in the first intron (15Silva M. Benito A. Sanz C. Prosper F. Ekhterae D. Nunez G. Fernandez-Luna J.L. J. Biol. Chem. 1999; 274: 22165-22169Abstract Full Text Full Text PDF PubMed Scopus (250) Google Scholar, 20Grillot D.A. Gonzalez-Garcia M. Ekhterae D. Duan L. Inohara N. Ohta S. Seldin M.F. Nunez G. J. Immunol. 1997; 158: 4750-4757PubMed Google Scholar), and thus, it is likely that the STAT5 binding sites for other genes may be located in introns as well. To identify STAT5 target genes in an unbiased fashion, we utilized the technique of chromatin immunoprecipitation (ChIP), which is a method to identify DNA regions that are bound by a transcription factor in vivo. We combined ligation-mediated PCR with subtraction steps to isolate pure STAT5 targets. We present data showing that this strategy is efficient at identifying STAT5 targets in an unbiased fashion, and we show that the STAT5 binding sites are often in introns. We also show that STAT5a and STAT5b bind to all targets tested, although each isoform binds to any given promoter with different kinetics compared with the other isoform. This demonstrates that STAT5a and STAT5b have distinct differences in their promoter binding behaviors that may underlie their functional differences. Cell Lines—Ba/f3 cells were obtained from the DSMZ-German Collection of Microorganisms and Cell Cultures (Braunschweig, Germany), and were maintained in RPMI supplemented with 10% fetal calf serum and 10% WEHI-3B conditioned media. WEHI-3B cells were obtained from ATCC (Manassas, VA). Antibodies and Cytokines—Antibodies recognizing STAT5a (sc-1081) and total STAT5 (sc-835) were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Anti-STAT5b (71-2500 and 13-5300) was obtained from Zymed Laboratories Inc. (South San Francisco, CA). Phosphospecific anti-STAT5 (935-1L) was obtained from Cell Signaling (Beverly, MA). Recombinant murine IL-3 was obtained from R 102: 3016-3024Crossref PubMed Scopus (47) Google Scholar). each cells were was with for at by a of were washed with were in of containing and and and for were at in a and were in of nuclear and for of was and the chromatin was by to an of was by the was and of were to each of were and the were at of and of (Santa Cruz were and the were for at The were washed with for each at by with for at with were twice of for at were and the were at The DNA was a PCR of each was and was used for PCR was the for for and for for are in the was as that two were the the first IP, the were from the of each of was and the was as was from the method used to isolate expressed genes Proc. Natl. Acad. Sci. U. S. A. PubMed Scopus Google Scholar). were in of the DNA to the MA). The DNA was in with and of were by by to the are as were by of these were to the DNA of of the was by PCR the of the as the The PCR was as either or The DNA was and was used to these from the Therefore, DNA used the DNA used in a total PCR of each and and DNA was a PCR of DNA was to a of The and DNA were to an of and This was at for and to which the was at for of of and of were and to which the was to the and and at of in with were and at for of was to the a were and the DNA was with and of The DNA was resuspended in of were and the DNA was the DNA was a DNA was at either the Cancer or by for these are in the were the of Santa Cruz, site the CpG were identified the CpG the at the to the were and were at least separate for are in the data for the S. P. R. S. M. C. M. E. T. M. Proc. Natl. Acad. Sci. U. S. A. 2001; PubMed Scopus Google Scholar) were obtained from the site the The of the STAT5 targets were identified the and the at and for these genes were obtained from the at was with C. Proc. Natl. Acad. Sci. U. S. A. 2001; PubMed Scopus Google Scholar) as the STAT5 in Ba/f3 the Ba/f3 as a to in the identification of STAT5 targets. IL-3 STAT5 in these cells, which to the expression of the F. Blood. 2000; PubMed Google Scholar, S. Santos S.C. F. L. C. P. Gisselbrecht S. Gouilleux F. Oncogene. 1999; 18: PubMed Scopus (137) Google Scholar). We first that both STAT5a and STAT5b are present in Ba/f3 cells and that both STAT5 isoforms are phosphorylated in to of from Ba/f3 cells were for STAT5a or phosphorylated STAT5a or STAT5b was seen in However, of both STAT5 isoforms was seen in cells Therefore, IL-3 in the of both STAT5a and STAT5b in Ba/f3 cells, we that these cells an to identify the targets of both STAT5 proteins. STAT5a and STAT5b to the bcl-x STAT5 binding site in the bcl-x promoter has been by and it to the sequence for STAT5 S. Santos S.C. F. L. C. P. Gisselbrecht S. Gouilleux F. Oncogene. 1999; 18: PubMed Scopus (137) Google Scholar). In to to identifying STAT5 we to the in binding of both STAT5 isoforms to the bcl-x were to this and were used in a Ba/f3 cells were of IL-3 and or with chromatin was and were to STAT5a or to In normal rabbit IgG was used as a STAT5a and STAT5b were bound to the bcl-x promoter IL-3 some binding be in cells, this is with IL-3 the immunoprecipitation is for STAT5a and STAT5b, there was of the bcl-x promoter when a was used for immunoprecipitation immunoprecipitation information the DNA binding of a transcription factor in 1999; 19: PubMed Scopus Google Scholar). is by of to DNA and these with a this method isolate all of the targets that are bound by the transcription factor at the of we that these targets be identified by a ligation-mediated PCR In this the is to and the sequence for this the DNA be and To also identify any differences in the targets of STAT5a and STAT5b, we a subtraction which is the technique to identify differential expression of in different Proc. Natl. Acad. Sci. U. S. A. PubMed Scopus Google Scholar). are two to this any differences in targets between STAT5a and STAT5b be and any DNA bound the be the in a of targets. of STAT5 identify STAT5 we the strategy from STAT5a and STAT5b chromatin the we targets from the and targets from the We the location of these in the the the STAT5a and identified unique targets. The of the STAT5b targets were unique targets. We a total of separate STAT5 binding of these were of the STAT5a targets were also from the STAT5b and targets from were in of the revealed that most are close to genes and be regulatory regions. the we that sequence and found that a STAT as by CpG are present in of genes at their F. Cell Mol. Sci. 2003; PubMed Scopus Google Scholar), have combined with from CpG DNA to identify targets 2003; PubMed Scopus Google Scholar, T. Mol. Cell. Biol. 2001; PubMed Scopus Google Scholar, J. M. T. Oncogene. 2003; PubMed Scopus Google Scholar). Therefore, we to or not the STAT5 binding sites that we have identified CpG We that CpG islands, Furthermore, with CpG were most often found near the of a as F. Cell Mol. Sci. 2003; PubMed Scopus Google Scholar). Therefore, the STAT5 binding sites that we have identified are most often found near genes, most contain a STAT site, and most do not contain CpG Genes near STAT5 STAT5 functions to directly regulate the expression of target genes, a is to identify where these identified STAT5 binding sites in to known genes. of these binding sites were of the transcription start the of the STAT5 binding sites were introns. in which the of the is not of STAT5 binding sites were located in the first intron Genes located near the STAT5 binding sites are in I. binding sites the and the other binding sites were near genes. These known and genes. of the STAT5 binding sites were near a were near more than For example, was found in the first intron of and this was found in the first intron of the important is that of the genes identified known STAT5 and they genes of signal DNA and of STAT5 binding sites in to known binding site first first of intron of intron of and first first first and from first first intron in a of STAT5 to that the binding sites that we have identified were STAT5 we these binding sites and PCR from obtained STAT5a and STAT5b We targets and found that are bound by STAT5 there was some binding of STAT5 to these targets binding with IL-3 stimulation, as was the with a signal in cells that was not with IL-3 This binding is for there is binding to when a IgG is used in the immunoprecipitation and binding to although there was in the binding of STAT5 to these promoters in to IL-3 not Therefore, they are likely STAT5 binding sites, the in the data does not to the to which IL-3 both promoters. of the targets were by a IgG indicating that the binding is for STAT5 of the binding sites to binding and we that this is not a STAT5 target not Therefore, we show that of the DNA STAT5 binding sites, with regulated by IL-3, and that all are bound by both STAT5a and STAT5a and STAT5b to with STAT5a and STAT5b bound all of the targets tested, the of binding of each isoform was different between promoters Therefore, we to or not binding of STAT5a or STAT5b We the binding of STAT5 proteins to promoters and of IL-3 The binding of STAT5a was similar at each point the promoter STAT5b binding at The binding of STAT5a to the promoter at and STAT5b binding is at and at STAT5a binding to the bcl-x promoter STAT5b is bound at The kinetics of of both STAT isoforms is similar not suggesting that any differences of DNA binding are not to differences in STAT5a and STAT5b promoters when both are activated in the STAT5 by IL-3 in Ba/f3 the of STAT5 with a binding site has functional we any of the genes containing these binding sites were regulated by IL-3 in Ba/f3 We of the identified STAT5 target genes for a we the known STAT5 target bcl-x and found that expression is IL-3 and were by IL-3 and bcl-x at of IL-3 stimulation, the of are at and to The STAT5 binding sites with and show an in STAT5 binding However, this is not the for the STAT5 binding site for STAT5 is bound to the of IL-3 or Therefore, any in transcription be to STAT5 binding may other STAT5 such as the binding of The aryl hydrocarbon receptor nuclear and to sperm were IL-3 the of are of IL-3 stimulation, the of to Therefore, the genes that we have identified as STAT5 targets by are expressed in Ba/f3 cells and show expression in to These also that STAT5 may as an or of STAT5 Genes in the genes identified by this method are targets of we a in which STAT5 is known to be STAT5 has been to be activated in acute (4Battle T.E. Frank D.A. Blood. 2003; 102: 3016-3024Crossref PubMed Scopus (47) Google Scholar, 5Bromberg J. J. Clin. Invest. 2002; 109: 1139-1142Crossref PubMed Scopus (755) Google Scholar). To these genes unique expression in leukemias, we the which expression data from a of Whereas this data set information the expression of genes these it does not contain normal control and thus, this method expression in the tumors are The genes identified in that the were for expression patterns in the These genes a distinct of expression in leukemias, with many of these tumors showing or expression of these STAT5 targets Therefore, of the identified STAT5 targets are expressed in a of acute leukemias, suggesting that this identify STAT5 targets. immunoprecipitation has become the to the binding of transcription factors to DNA in vivo. a of the DNA binding of transcription factors in In addition to the binding of a to a known DNA has been a to identify targets of transcription have combined with to identify targets of transcription factors in These from CpG DNA 2003; PubMed Scopus Google Scholar, Genes Dev. 2002; 16: PubMed Scopus Google Scholar, N. G.W. J. P.A. E. Science. PubMed Scopus Google promoter containing DNA near the of genes H. Y. T. J. Genes Dev. 2002; 16: PubMed Scopus Google Scholar, N. G.W. J. P.A. E. Science. PubMed Scopus Google and containing the of and S. S. P. D. A. J. R. J. M. S. S. H. G. K. Cell. Full Text Full Text PDF PubMed Scopus Google Scholar, G. T.E. P. R. J.L. F. P. M. S. M. Mol. Cell. Biol. PubMed Scopus Google Scholar, R. G. P. S. T.E. J.L. P. M. S. M. Proc. Natl. Acad. Sci. U. S. A. 2003; PubMed Scopus Google Scholar). of these studies have given important the genes by transcription However, each of these has For example, of genes contain CpG in their promoters F. Cell Mol. Sci. 2003; PubMed Scopus Google Scholar), the of to identify all of the binding sites of a transcription not found at the of genes. promoter contain DNA near the transcription start site H. Y. T. J. Genes Dev. 2002; 16: PubMed Scopus Google Scholar, N. G.W. J. P.A. E. Science. PubMed Scopus Google Scholar) and are the of genes. The data in this show that a of STAT5 binding sites are not near the transcription start have also shown that the binding sites for other transcription factors are often located far from the of genes S. S. P. D. A. J. R. J. M. S. S. H. G. K. Cell. Full Text Full Text PDF PubMed Scopus Google Scholar, G. T.E. P. R. J.L. F. P. M. S. M. Mol. Cell. Biol. PubMed Scopus Google Scholar, R. G. P. S. T.E. J.L. P. M. S. M. Proc. Natl. Acad. Sci. U. S. A. 2003; PubMed Scopus Google Scholar). However, although it is that these they of the Therefore, there is at present to identify all of the binding sites of a given transcription to identifying all targets of a transcription factor is a such as that used to identify targets T. Mol. Cell. Biol. 2001; PubMed Scopus Google Scholar). This the of identifying novel transcription factor binding sites in and we this technique to a ligation-mediated PCR step to the of a of is that this PCR step may regions. However, most of the identified targets are STAT5 binding sites, this method does not many In it has the of the of a of DNA from chromatin We used this technique to isolate targets of STAT5. The of the identified STAT5 binding sites are located with most in the first is not known most STAT5 binding sites are located in there is for this in that the STAT5 binding site in the bcl-x is also located in the first intron (15Silva M. Benito A. Sanz C. Prosper F. Ekhterae D. Nunez G. Fernandez-Luna J.L. J. Biol. Chem. 1999; 274: 22165-22169Abstract Full Text Full Text PDF PubMed Scopus (250) Google Scholar, 20Grillot D.A. Gonzalez-Garcia M. Ekhterae D. Duan L. Inohara N. Ohta S. Seldin M.F. Nunez G. J. Immunol. 1997; 158: 4750-4757PubMed Google Scholar). We not identify any identified STAT5 targets in However, this is not studies that there may be of binding sites in the for any transcription factor a given set of S. S. P. D. A. J. R. J. M. S. S. H. G. K. Cell. Full Text Full Text PDF PubMed Scopus Google Scholar, G. T.E. P. R. J.L. F. P. M. S. M. Mol. Cell. Biol. PubMed Scopus Google Scholar, R. G. P. S. T.E. J.L. P. M. S. M. Proc. Natl. Acad. Sci. U. S. A. 2003; PubMed Scopus Google Scholar). STAT5 has been in and has been shown to be a of the STAT5 the expression of genes that are known to regulate growth and such as cyclin D1 and bcl-x I. Kitamura T. Wakao H. Tanaka H. Hashimoto K. Albanese C. Downward J. Pestell R.G. Kanakura Y. EMBO J. 1999; 18: 1367-1377Crossref PubMed Scopus (294) Google Scholar, F. Blood. 2000; PubMed Google Scholar) as as the G. C. S. N. PubMed Scopus Google Scholar). The novel genes that we have identified in this of STAT5 targets to in and in addition to a of genes. This the of STAT5 and it also the of known functions in which STAT5 may be In of the genes identified are known to be with This which is with T. M. K. G. A. N. Y. Y. K. 1998; PubMed Scopus Google the of which is for R. 1997; PubMed Scopus Google Scholar, C. M. Mol. Med. 1998; Full Text Full Text PDF PubMed Scopus Google and which is with L.A. Blood. PubMed Scopus Google Scholar). is that STAT5 may a in the pathogenesis of these In addition to identifying novel STAT5 this to there are functional differences between STAT5a and STAT5b related to their DNA binding site selection. We have shown that both STAT5a and STAT5b bind to all of the sites the binding of STAT5a and STAT5b any given promoter is with STAT5a and STAT5b binding for not of The of this binding is it may differential of other factors for This level of is an of STAT5 and it that STAT5a and STAT5b may have different functions in the of transcription when both are present and activated in a STAT5 functions by the expression of genes, and all of the genes that we are expressed in Ba/f3 cells, most modulated by We also that these genes show unique in the of STAT5 as an or of some of these genes are in acute leukemias, and some are This is an important of STAT5 is known to be a of (1Frank D.A. Cancer Treat. Res. 2003; 115: 267-291Crossref PubMed Google Scholar). We have used this identification technique to identify targets of of their location in the This has the identification of many STAT5 binding sites in novel target genes, and these regions regulated by STAT5 outside of traditional promoter regions. with