A Database of Transcription Factors Involved in Lymphocyte Development

SHORT COMMUNICATION LymphTF-DB: a database of transcription factors involved in lymphocyte development

PJ Childress, RL Fletcher and NB Perumal School of Informatics, Indiana University-Purdue University Indianapolis, Indianapolis, IN, USA

B and T cells develop following a similar early stepwise progression to later stages where multiple developmental options are available. These developmental regimes necessitate differential gene expression regulated by a large number of transcription factors (TFs). The resultant burgeoning amount of information has opened a knowledge gap between TF activities during lymphocyte development and a researcher’s experiments. We have created the LymphTF database (DB) to fill this gap. This DB holds interactions between individual TFs and their specific targets at a given developmental time. By storing such interactions as a function of developmental progression, we hope to advance the elucidation of regulatory networks that guide lymphocyte development. Besides queries for TF-target gene interactions in developmental stages, the DB provides a graphical representation of downloadable target gene regulatory sequences with locations of the transcriptional start sites and TF-binding sites. The LymphTF-DB can be accessed freely on the web at http://www.iupui.edu/~tfinterx/. Genes and Immunity (2007) 8, 360–365; doi:10.1038/sj.gene.6364386; published online 15 March 2007

Keywords: transcription factors; targets; lymphocyte development; database

Introduction the discovery of new targets and transcription factors important in the process of lymphocyte development.7,8 Research in mammalian lymphocyte development has Interpretation of results from these high-throughput benefited from improvements in molecular sorting and experiments is, however, not always straight forward. identification techniques as well as from the use of the Further studies are needed to describe properly a factor’s mouse as a robust model system. The result is a field of place in the transcriptional regulation hierarchy. Com- study that has produced a very large amount of data. plicating matters further is the context-specific action of Additionally, lymphocyte development from progenitor some transcription factors. For example, the zinc-finger stem cells provides a model for studying more general protein Ikaros is involved in chromatin remodeling development paradigms and has most recently been complexes and has been shown to have simultaneous shown to have clinical significance in understanding the activation and repressive function within the same cell process of terminal differentiation. The development of type and stage of development.9,10 Recently many good the major lymphocytic cell types, B and T cells, is at the reviews have been written that attempt to highlight the center of gene expression studies. In this system, many key players involved in the development of these two well-defined milestones in the development process, and cell types. However, given the breadth of information molecular pathways determining cell fate have been available these reviews typically focus on a specific part described. These cells undergo similar transitions be- of the process such as early B-cell development, a tween stages dependent upon both anatomical location particular transcription factor such as NF-kB, or include and exogenous stimulation from immune stimuli.1,2 only a portion of the TFs and targets involved.5,6,11,12 The complex lymphocyte developmental regime is The explosion of data has led to the issue of dependent upon changing patterns of gene expression information management becoming a critical concern in that are accomplished, in part, by the action of specific this field of study. The scientific community has begun transcription factors (TF) on their targets. This action is a to address this issue through a variety of ways. The dynamic and generally a conserved chorus of events.3,4 TRANSFAC (http://www.gene-regulation.com/pub/ Since the early 1990s the work in this field has produced databases.html), JASPAR (http://jaspar.cgb.ki.se/cgi-bin/ a steady stream of well-defined TF-target gene inter- jaspar_db.pl) and TESS (http://www.cbil.upenn.edu/ actions.5,6 More recently, DNA microarrays have aided tess) databases are intended for use by wide audiences for general transcriptional regulation studies. These databases mix computationally predicted transcription Correspondence: Dr NB Perumal, School of Informatics, Indiana factor-binding sites (TFBS) with biologically validated University-Purdue University Indianapolis, 719 Indiana Avenue data on TF-target interactions. Research in lymphocyte Suite 311, Indianapolis, IN 46202, USA. E-mail: [email protected] development in the post-genomics era has benefited Received 21 September 2006; revised and accepted 1 February 2007; from specialized databases. This is due, in part, to the published online 15 March 2007 numbers of transcription factors and targets involved, Database of TF activity in lymphocyte development PJ Childress et al 361 Table 1 Search results for TF-target gene interactions of Bcl6

Activity Target gene B-cell stage T-cell stage Reference(s) Evidence

Results for Bcl6 (rollover names for aliases) Represses Bcl6 Mature B Tunyaplin et al.17 tt em nb ci Represses Ccl3 Immature B Shaffer et al.7 ma em nb tt Represses Ccnd2 Immature B Shaffer et al.7 ma em nb tt Represses Cd44 Immature B Shaffer et al.7 ma nb tt Represses Cd69 Immature B Shaffer et al.7 ma em nb tt Represses Cd80 Mature B Niu et al.21 rp tt ci fc Represses Cdkn1b Immature B Shaffer et al.7 ma nb tt Represses Fcer1g Th2 Dent et al.22 em tt Represses Gata3 Th2 Kusam et al.23 ko rp wb tt ia Represses Igh-7 Lymphoblast Harris et al.24 tt em ph Represses Il13 Th2 Dent et al.22 rpa ia Represses Il4 Th2 Dent et al.22 rpa ia Represses Il5 SP (+4) Arima et al.25 em ci tt mu Represses Il5 Th1 Arima et al.25 em ci tt mu Represses Il5 Th2 Arima et al.25 em ci tt mu Represses Il6 Th2 Dent et al.22 rpa ia Represses Prdm1 Immature B Tunyaplin et al.17 tt fp ci ko rp Represses Prdm1 Mature B Tunyaplin et al.17 tt fp ci ko rp Represses Trp53 Mature B Phan et al.26 tt ci kd ts

The raw data for information in the DB comes from peer-reviewed, published scientific articles that are available from the National Library of Medicine’s PubMed DB. In an attempt to obtain current and relevant data, articles published before 1985 were excluded. Keyword searches (e.g., ‘Transcription factors and T [or B]-cell development’) were performed using the NCBI’s Entrez search engine. Relevant articles were culled from thousands obtained by initially screening titles and abstracts for their potential to be included in our database. In the second and subsequent rounds of annotation, selected articles were carefully read for collection of data regarding TFs, target genes, developmental stages of expression, and experimental evidence supporting TF–target gene interactions. We have included only murine data in the LymphTF-DB. Users can search for any transcription factor or target gene by its official name or alias from the home page of the website. It is not necessary to input literal values; the search function also accepts wild card values. The combination of a MySQL database searched via a web page front end utilizing the PHP scripting language has been used to present the information in a freely available format. and the transient nature of these interactions. Apparently logists studying lymphocyte development and develop- contradictory phenomena, such as Ikaros mentioned mental biologists exploring transcriptional regulatory above, may indicate complex modes of action for the TFs networks (TRN).15–18 involved. Researchers also need to be aware of the type of evidence associated with each interaction to fully understand the regulatory mechanisms involved. The Results and discussion HemoPDB (http://bioinformatics.med.ohio-state.edu/ HemoPDB/) contains information related to transcrip- Review of relevant literature has resulted in the collec- tional control of hematopoiesis including promoter tion and cataloging of 111 unique transcription factors sequences and cis-regulatory elements for transcription involved or implicated in B- or T-cell development at the factors.13 The Immunogenetics group at Universite´ time of publication. This number represents 55 factors for Montpellier II, Montpellier, France has assembled the which interactions with a specific target gene have been IMGT/GENE-DB (http://imgt.cines.fr/cgi-bin/GEN- shown, and an additional 56 factors that have been Elect.jv?livret ¼ 0) specific for human and mouse im- shown to be present or transcriptionally active during B- munoglobulin and T-cell receptor genes.14 These are just and T-cell development. These TFs are presented in the two examples of the community addressing the informa- DB with commonly used aliases and links to the factor’s tion management issues for specific collections of data in entry in the Gene DB at NCBI.19 The LymphTF-DB the field of lymphopoiesis. From these domain-specific contains a total of 110 target genes. As with the TFs, each databases researchers hope to piece together a complete target is linked to a Gene DB entry and common aliases picture of gene expression and molecular function. are stored in the DB. The total number of interactions in Here, we present an approach to capture published the DB is 388. Of these 388 interactions, 205 interactions molecular interactions in B-and T-cell development. The constitute unique TF-target pairs with experimental LymphTF-DB (http://www.iupui.edu/~tfinterx) is a evidence of interactions in various lymphoid develop- manually curated DB of transcription factors and their mental stages. Further, we have identified 51 TF–TF activity on specific target genes at various developmental interactions wherein one TF activates or represses stages in the mouse. To the best of our knowledge this is another TF during lymphocyte development (34 unique the only publicly available DB containing these interac- pairs regardless of a developmental stage). This data can tions. Our DB has been designed to be a repository of the be accessed from the ‘Show TF–TF interactions’ link of curated data, as well as to link users to the primary the ‘TF Activity’ page. In essence, the LymphTF-DB literature from which the interactions have been derived. focuses on a searchable repository of (1) TF-target gene The goal of this work is to organize these interactions in a interactions in lymphocyte development at various logical manner that can be easily accessed by immuno- stages and (2) the upstream regulatory sequence maps

Genes and Immunity Database of TF activity in lymphocyte development PJ Childress et al 362 of target genes showing the transcription start sites (TSS) repressor, Bcl6 (Table 1). The results show that Bcl6 is and TFBS. present in both B and T cells and acts as a repressor in all The DB’s website was designed to allow users the stages. Target genes are hyperlinked to their respective ability to search each transcription factor’s interaction Gene DB entries at NCBI, and references are hyperlinked with a target gene at the site of action and/or location to the corresponding PubMed references. A listing of (developmental stage). An example of a TF-target experimental evidence supporting each TF-target gene interaction search is shown for the transcriptional interaction is also shown. A total of 15 target genes have

a a

Figure 1 The Target Gene Map Page and associated queries. (a, i) An example of a target gene map for Rag2 is shown indicating locations of TSS and TFBS in the À8000 to þ 2000 bp region (relative to the TSS). The locations of CpG islands and MARs in the corresponding upstream regulatory sequences are generated using emboss cpgplot (http://bioweb.pasteur.fr/docs/EMBOSS/cpgplot.html) and the MAR-Wiz tool from FUTURESOFT Corporation (http://www.futuresoft.org/modules/MarFinder/index.html) respectively. This page also provides a tool to download sequences of any size (within this 10 kb region). This will be a blank page for any target gene whose TSS has not been experimentally determined. (ii) The selection menu for various queries is shown. Users have the ability to look up other target genes in the display by selecting from a pull-down menu (at the top), and to view, in a new page, the other target gene interactions of a TF (selected at the bottom) whose binding site maps in the currently displayed regulatory region, and the corresponding developmental stages. (b) Query results of binding sequences for Gata3 in Rag2 and other target genes are shown. These results were obtained by selecting Gata3 (bottom, (a, ii)), performing a search for TF-target gene interactions, and then moving the mouse over the Gata3 symbol in the resultant new page. (c) Query results of distal regulatory elements in Rag2 are shown. When a distal regulatory element (enhancer, insulator, locus control region etc) has been identified, the target gene map shows a set of double arrowheads (50 or 30 to the gene) and a ‘distal elements for gene’ button (top, (a, ii)), A click of this button opens a new page with these query results. The target gene regulatory sequences and TSS locations were obtained from the Mouse Genome Informatics website (http://www.informatics.jax.org) and were annotated to denote the location of the associated TFBS. All data are derived from the Mouse genome build 34.1 and in the next release of the website the DB will migrate to Mouse genome build 36.

Genes and Immunity Database of TF activity in lymphocyte development PJ Childress et al 363 been identified for this transcription factor in murine B elements controlling the transcription of the target gene or T cells. These results also show auto-regulation of Bcl6 (Figure 1c). expression at the mature B-cell stage (Row 1, Table 1).20 The DB can also be interrogated by specific develop- Users can access curated maps of upstream regulatory mental stage as the search criterion. Table 2 shows the sequences of target genes from the ‘Target Gene Map’ results of searches for the immature B (2a) and double page of the web site. An example of a target gene positive T cell (CD4 þ CD8 þ ) stages (2b) of development. promoter map is shown in Figure 1. This map (Figure These two stages represent immune cells that have 1a,i) shows the putative TSS and TFBS experimentally undergone germ line recombination, and are in a poised mapped, CpG island frequencies and predicted matrix state awaiting antigenic stimulation. Such developmental attachment regions (MAR). Users can also retrieve any stage queries show the various TF-target gene interac- portion of the regulatory region sequences (À8000 to tions known to be active in these cell types. Relevance of þ 2000 relative to the TSS) employing a selector box tool. the DB for lymphocyte developmental studies can be Another feature included in the promoter map is the stressed by the information provided by this query ability to query a TF’s activity for other target genes that (Table 2) showing a defined set of TFs and target genes it activates or represses and the corresponding develop- active in each stage. Our collection has brought together mental stages. This feature may be useful for immunol- and presents a knowledge base for future research ogists and developmental biologists exploring TRNs in questions. One such question could be whether the lymphocyte development. A user can also obtain the presence of a TF and a target gene in a developmental binding sequences of the selected TF on the mapped stage is biologically related (‘Guilt by association’) as in a target gene and other target genes in the DB (Figure 1b). significant number of cases the curated information Another useful feature is the query for distal regulatory shown in a developmental stage query has not been

Table 2 Search results for transcription factors and target genes active in a developmental stage

Transcription factor Activity Target gene Reference

(a) All factors active in stage: immature B Bcl6 Represses Cdkn1b Shaffer et al.7 Bcl6 Represses Prdm1 Tunyaplin et al.17 Bcl6 Represses Cc;3 Shaffer et al.7 Bcl6 Represses Ccnd2 Shaffer et al.7 Bcl6 Represses Cd69 Shaffer et al.7 Bcl6 Represses Cd44 Shaffer et al.7 Ebf1 Activates Igll1 Sigvardsson et al.27 Egr1 Activates Icam1 Maltzman(2) et al.28 Pax5 Activates Cd19 Kozmik et al.29 Pax5 Activates Igh-7 Liao et al.30 Pax5 Igh Singh31 Pax5 Igj Rinkenberger et al.32 Pax5 Activates Rag2 Kishi et al.33 Pax5 Activates Blk Zwollo et al.34 Sfpi1 Activates Grap2 Garrett-Sinha et al.35 Smad3 Activates Bcl2l11 Wildey et al.36 SpiB Activates Grap2 Garrett-Sinha et al.35

(b) All factors active in stage: DP Elf1 Activates Cd4 Siu37 Ets1 Activates Cd5 Tung et al.38 Maz Activates Cd4 Siu37 Myb Activates Cd4 Siu37 Myb Activates Tcrg Hsiang et al.39 Myb Activates Ada Ess et al.40 Notch1 Activates Bcl2 Deftos et al.41 Notch1 Activates Il7r Deftos et al.42 Notch1 Activates Cd44 Deftos et al.42 Stat5a Activates Pim1 Nosaka et al.43 Tcf12 Activates Smarcc1 Ko et al.44 Tcfe2a Activates Smarcc1 Ko et al.44 Zfpn1a1 Activates Cd8a Georgopoulos9 Zfpn1a1 Activates Dntt Trinh10

Queries to obtain the list of TFs active in various stages of B- and T-cell development can be run from the ‘Developmental Stages’ page. Such queries also provide the corresponding target genes interacting with a TF in the stage being searched for. Lymphocyte developmental stages are named and delineated in keeping with well-established conventions gleaned from the publications reviewed. Briefly, two determinants of stage differentiation, that is, cell surface markers and genomic rearrangements were collected for each (if applicable) stage of development discussed in the literature. A minimal set of cell surface markers and/or stage-specific genomic rearrangements for B and T cells were used to define each stage. In the DB, TFs and target genes were annotated to be active and/or expressed in the various stages if that information was available in the literature. The B- and T-cell staging schemes are shown as links in the ‘Developmental Stages’ page of the web site.

Genes and Immunity Database of TF activity in lymphocyte development PJ Childress et al 364 verified experimentally for direct TF-target gene inter- DB maintenance and future plans actions. The DB and website have been designed to serve as a A general observation of the extensive literature tool for researchers in immunology that is easy-to-use review performed for acquiring the TF-target interaction and capable of growing as new data is published in the data revealed an ongoing debate in the research field of lymphocyte development. We plan to update the community on the particulars of what constitutes DB once every 6 months and to include text-mining individual stages. The concept of developmental stages approaches to collect and curate newly available data. is an integral part of our DB structure, and as such we Owing to the evolutionary significance in lymphocyte employed a pragmatic approach of defining stages. This development, in the next release of the web-site, we hope was accomplished through defining cell surface markers to include TF–target interaction data from other mam- and genomic rearrangements of immunoglobulin and T- malian species, and in particular, human data. The cell receptor genes. In general, the stages were defined by information stored in our DB demonstrating TF-TF the most common techniques that have been used to interactions and auto-regulation (the latter not accessible study lymphocyte development. For example, a widely by a user query) is crucial in exploring TRNs specifically recognized stage of B-cell maturation is the pro-B-cell in lymphocyte development and in more general devel- stage. This has been further broken down into ‘early’ and opmental paradigms. Future design changes are planned ‘late’ stages. Both delineations are commonly accepted. so that more advanced queries will be available from the Alternatively, the pro-B-cell stage can also be described web interface. Examples of such queries include the as fractions A, B and C.1 As a practical matter, our auto-regulation query and a user-defined query interface database used the ‘early pro-B’ and ‘late pro-B’ designa- to allow investigators to dictate what data they choose tions as a compromise between the more general pro-B- to view. Finally, in the future, we plan to provide a cell stage and the more specific (but not as often used) mechanism for community-based input and mainte- fraction A, B and C designations. nance of data that will be added, similar to the The development of lymphocytes is a well-studied ORegAnno database.45 topic in immunology. As such there are volumes of data that describe the molecular regulation of transcription factors and their targets. Storing this information in a web accessible DB has the practical benefit of giving the Acknowledgements scientific community a quick view of the state of this work. With a list of 111 transcription factors present at We appreciate the comments and suggestions of two some point during the development of B and T cells, anonymous reviewers. assimilating all the known interactions is becoming an impossible task without a computational tool. Also, by organizing the information in a logical way that mirrors development, knowledge gaps can be explored and References possible new interactions discovered. Every attempt has been made to present only the verified or demonstrated 1 Hardy RR. B-cell commitment: deciding on the players. interactions of the transcription factors. The structure of Curr Opin in Immunol 2003; 15: 158–165. the DB schema and the data collection procedures have 2 Matthias P, Rolink AG. Transcriptional networks in develop- been designed not to ‘postulate’ or project activity based ing and mature B cells. Nat Rev Immunol 2005; 5: 497–508. on previous or subsequent activity. Any gaps in 3 Jimenez JL, Mitchell MP, Sgouros JG. Microarray analysis of information thus represent a lack of inclusion in the DB orthologous genes: conservation of the translational machin- ery across species at the sequence and expression level. due to missed literature search, lack of experimental Genome Biol 2003; 4: R4. data, or a true lack of biological activity. Currently, only 4 Mata J, Bahler J. Correlations between gene expression information for transcriptional activity in the mouse is and gene conservation in fission yeast. Genome Res 2003; 13: included. The mouse model for lymphopoiesis tracks 2686–2690. closely with the human system and has been used for 5 Kuo CT, Leiden JM. Transcriptional regulation of T lympho- some time as a model system. Further, the TF-target cyte development and function. Annu Rev Immunol 1999; 17: interactions stored in, and accessible from the LymphTF- 117–149. DB would provide an excellent knowledge base for 6 Bartholdy B, Matthias P. Transcriptional control of B cell researchers interested in assembling transcription factor development and function. Gene 2004; 327: 1–23. 7 Shaffer AL, Yu X, He Y, Boldrick J, Chan EP, Staudt LM. BCL-6 networks of activity. represses genes that function in lymphocyte differentiation, We have used rigorous criteria for inclusion into the inflammation, and cell cycle control. Immunity 2000; 13: DB to ensure the accuracy of the data. TF–target 199–212. interactions must be shown in murine cells (in vivo or 8 Tabrizifard S, Olaru A, Plotkin J, Fallahi-Sichani M, Livak F, in cell culture), and such evidence requires demonstrable Petrie HT. Analysis of transcription factor expression during TF binding to cis-regulatory elements. All interactions discrete stages of postnatal thymocyte differentiation. J must show a causal relationship (i.e. fold change up or Immunol 2004; 173: 1094–1102. down on microarray analysis does not constitute an 9 Georgopoulos K. Haematopoietic cell-fate decisions, chroma- interaction), and a discernable stage of development tin regulation and Ikaros. Nat Rev Immunol 2002; 2: 162–174. must be obvious from the published work to have that 10 Trinh LA, Ferrini R, Cobb BS et al. Downregulation of TDT transcription in CD4+ CD8+ thymocytes by Ikaros proteins in interaction defined in a developmental context. We have direct competition with an Ets activator. Genes Dev 2001; 15: also included a simple submission form for users to 1817–1832. suggest interactions to be included. Any submissions 11 Busslinger M. Transcriptional control of early B cell develop- will be evaluated on a case-by-case basis. ment. Annu Rev Immunol 2004; 22: 55–79.

Genes and Immunity Database of TF activity in lymphocyte development PJ Childress et al 365 12 Beinke S, Ley SC. Functions of NF-kappaB1 and NF-kappaB2 29 Kozmik Z, Wang S, Dorfler P, Adams B, Busslinger M. The in immune cell biology. Biochem J 2004; 382 (Part 2): 393–409. promoter of the CD19 gene is a target for the B-cell-specific 13 Pohar TT, Sun H, Davuluri RV. HemoPDB: hematopoiesis transcription factor BSAP. Mol Cell Biol 1992; 12: 2662–2672. promoter database, an information resource of transcriptional 30 Liao F, Birshtein BK, Busslinger M, Rothman P. The transcrip- regulation in blood cell development. Nucleic Acids Res 2004; tion factor BSAP (NF-HB) is essential for immunoglobulin 32: D89–D90. germ-line epsilon transcription. J Immunol 1994; 152: 14 Giudicelli V, Chaume D, Lefranc MP. IMGT/GENE-DB: a 2904–2911. comprehensive database for human and mouse immunoglo- 31 Singh H. Specific recognition site probes for isolating genes bulin and T cell receptor genes. Nucleic Acids Res 2005; 33: encoding DNA-binding proteins. Methods Enzymol 1993; 218: D256–D261. 551–567. 15 Singh H, Medina KL, Pongubala JMR. Contingent gene 32 Rinkenberger JL, Wallin JJ, Johnson KW, Koshland ME. An regulatory networks and B cell fate specification. Proc Natl interleukin-2 signal relieves BSAP (Pax5)-mediated repression Acad Sci 2005; 102: 4949–4953. of the immunoglobulin J chain gene. Immunity 1996; 5: 377–386. 16 Anderson MK, Hernandez-Hoyos G, Dionne CJ, Arias AM, 33 Kishi H, Wei XC, Jin ZX, Fujishiro Y, Nagata T, Matsuda T et al. Chen D, Rothenberg EV. Definition of regulatory network Lineage-specific regulation of the murine RAG-2 promoter: elements for T cell development by perturbation analysis with GATA-3 in T cells and Pax-5 in B cells. Blood 2000; 95: 3845–3852. PU.1 and GATA-3. Dev Biol 2002; 246: 103–121. 34 Zwollo P, Desiderio S. Specific recognition of the blk promoter 17 Tunyaplin C, Shaffer AL, Angelin-Duclos CD, Yu X, by the B-lymphoid transcription factor B-cell-specific activator Staudt LM, Calame KL. Direct repression of prdm1 by Bcl-6 protein. J Biol Chem 1994; 269: 15310–15317. inhibits plasmacytic differentiation. J Immunol 2004; 173: 35 Garrett-Sinha LA, Hou P, Wang D, Grabiner B, Araujo E, 1158–1165. Rao S et al. Spi-1 and Spi-B control the expression of the 18 Warren LA, Rothenberg EV. Regulatory coding of lymphoid Grap2 gene in B cells. Gene 2005; 353: 134–146. lineage choice by hematopoietic transcription factors. Curr 36 Wildey GM, Patil S, Howe PM. Smad3 potentiates transform- Opin Immunol 2003; 15: 166–175. ing growth factor beta (TGFbeta )-induced apoptosis and 19 Maglott D, Ostell J, Pruitt KD, Tatusova T. Entrez Gene: expression of the BH3-only protein Bim in WEHI 231 B gene-centered information at NCBI. Nucleic Acids Res 2005; 33: lymphocytes. J Biol Chem 2003; 278: 18069–18077. D54–D58. 37 Siu G. Controlling CD4 gene expression during T cell lineage 20 Chen J, Alt FW. Gene rearrangement and B-cell development. commitment. Semin Immunol 2002; 14: 441–451. Curr Opin Immunol 1993; 5: 194–200. 38 Tung JW, Kunnavatana SS, Herzenberg LA, Herzenberg LA. 21 Niu H, Cattoretti G, Dalla-Favera R. BCL6 controls the The regulation of CD5 expression in murine T cells. BMC Mol expression of the B7-1/CD80 costimulatory receptor in Biol 2001; 2:5. germinal center B cells. J Exp Med 2003; 198: 211–221. 39 Hsiang YH, Goldman JP, Raulet DH. The role of c-Myb or a 22 Dent AL, Shaffer AL, Yu X, Allman D, Staudt LM. Control of related factor in regulating the T cell receptor gamma gene inflammation, cytokine expression, and germinal center enhancer. J Immunol 1995; 154: 5195–5204. formation by BCL-6. Science 1997; 276: 589–592. 40 Ess KC, Whitaker TL, Cost GJ, Witte DP, Hutton JJ, Aronow BJ. 23 Kusam S, Toney LM, Sato H, Dent AL. Inhibition of Th2 A central role for a single c-Myb binding site in a thymic locus differentiation and GATA-3 expression by BCL6. J Immunol control region. Mol Cell Biol 1995; 15: 5707–5715. 2003; 170: 2435–2441. 41 Deftos ML, He YW, Ojala EW, Bevan MJ. Correlating 24 Harris MB, Chang CC, Beron MT, Danial NN, Zhang J, notch signaling with thymocyte maturation. Immunity 1998; Keuhner D et al. Transcriptional repression of Stat6-dependent 9: 777–786. interleukin-4-induced genes by BCL-6: specific regulation 42 Deftos ML, Huang E, Ojala EW, Forbush KA, Bevan MJ. of iepsilon transcription and immunoglobulin E switching. Notch1 signaling promotes the maturation of CD4 and CD8 SP Mol Cell Biol 1999; 19: 7264–7275. thymocytes. Immunity 2000; 13: 73–84. 25 Arima M, Toyama H, Ichii H, Kojima S, Okada S, Hatano M 43 Nosaka T, Kawashima T, Misawa K, Ikuta K, Mui AL, et al. A putative silencer element in the IL-5 gene recognized Kitamura T. STAT5 as a molecular regulator of proliferation, by Bcl6. J Immunol 2002; 169: 829–836. differentiation and apoptosis in hematopoietic cells. EMBO J 26 Phan RT, Dalla-Favera R. The BCL6 proto-oncogene sup- 1999; 18: 4754–4765. presses p53 expression in germinal-centre B cell. Nature 2004; 44 Ko M, Ahn J, Lee C, Chung H, Jeon SH, Chung HY et al. E2A/ 432: 635–639. HEB and Id3 proteins control the sensitivity to glucocorticoid- 27 Sigvardsson M, O’Riordan M, Grosschedl R. EBF and E47 induced apoptosis in thymocytes by regulating the SRG3 collaborate to induce expression of the endogenous immuno- expression. J Biol Chem 2004; 279: 21916–21923. globulin surrogate light chain genes. Immunity 1997; 7: 25–36. 45 Montgomery SB, Griffith OL, Sleumer MC, Bergman CM, 28 Maltzman JS, Carmen JA, Monroe JG. Transcriptional regula- Bilenky M, Pleasance ED et al. ORegAnno: an open access tion of the Icam-1 gene in antigen receptor- and phorbol ester- database and curation system for literature-derived promo- stimulated B lymphocytes: role for transcription factor EGR1. ters, transcription factor binding sites and regulatory varia- J Exp Med 1996; 183: 1747–1759. tion. Bioinformatics 2006; 22: 637–640.

Genes and Immunity