(The FASEB Journal. 2003;17:1228-1237.)
© 2003 FASEB
Computer modeling of promoter organization as a tool to study transcriptional coregulation
THOMAS WERNER*,
,
SABINE FESSELE
,
HOLGE MAIER* and
PETER J. NELSON,1
* GSF-National Research Center for Environment and Health, Institute of Experimental Genetics, Neuherberg, Germany;
Genomatix Software GmbH, D-80339 Munich, Germany; and
Medizinische Poliklinik, Ludwig-Maximilians-University of Munich, Germany
1Correspondence: Medizinische Poliklinik, Ludwig-Maximilians-Universität München, Schillerstrasse 42, 80336, Munich, Germany. E-mail: nelson{at}medpoli.med.uni-muenchen.de
 |
ABSTRACT
|
|---|
Understanding how the regulation of gene networks is orchestrated is an important challenge for characterizing complex biological processes. Gene transcription is regulated in part by nuclear factors that recognize short DNA sequence motifs, called transcription factor binding sites, in most cases located upstream of the gene coding sequence in promoter and enhancer regions. Genes expressed in the same tissue under similar conditions often share a common organization of at least some of these regulatory binding elements. In this way the organization of promoter motifs represents a "footprint" of the transcriptional regulatory mechanisms at work in a specific biologic context and thus provides information about signal and tissue specific control of expression. Analysis of promoters for organizational features as demonstrated here provides a crucial link between the static nucleotide sequence of the genome and the dynamic aspects of gene regulation and expression.—Werner, T., Fessele, S., Maier, H ., Nelson, P. J. Computer modeling of promoter organization as a tool to study transcriptional coregulation.
Key Words: promoter modeling • bioinformatics • module • functional context
|
INTRODUCTION
|
|---|
CONTROL OF GENE expression is achieved through the coordinated action of various cis- and trans-acting factors including matrix attachment regions (S/MAR), locus control regions (LCR), gene methylation, transcription factors, enhancers and silencers, and promoters (1
2
3)
. Promoter regions comprise the genomic DNA sequences found upstream from the transcribed sequence but often overlap with, or include, the first exon of a gene. Promoters are the central processors of transcriptional control, as the regulatory information contributed by the other elements must be integrated within the context of a promoter in order to influence gene expression (1)
.
Promoters linked by regulatory networks help orchestrate the expression of gene products expressed within the same biologic context: for example, gene products that comprise developmental or signal transduction pathways, or a set of genes linked to the response of a tissue to stress (4
, 5)
. When high throughput approaches such as DNA arrays or microchips are applied to characterize gene expression, the data gathered after clustering can reflect the effects of these regulatory networks. Understanding how networks of promoters are organized provides insight into when and how the expression of specific genes is controlled and expands the capability of target identification and characterization.
The DNA sequences that comprise promoters do not provide much direct information about regulation. Promoter function is not coupled to fixed stretches of sequence homology, but rather to highly variable elements representing individual transcription factor binding sites that act as a binding site for their cognate protein. The sites are generally composed of 10 to 30 nucleotides; of these, usually only a small core of nucleotides, often separated by nonconserved sequences, establishes the criteria for a binding site. Because of this inherent variability, transcription factor binding sites cannot be efficiently described by their individual sequence. However, the flexibility of these sites can be defined by either an IUPAC consensus sequence or by weight matrices. IUPAC consensus sequences use ambiguous symbols (e.g., B=C, G or T; R=A or G) to describe the variability of nucleotide usage. A weight matrix is represented by a 2-dimensional table of numbers that reflects the nucleotide preferences for the individual positions of the aligned transcription factor binding sites and is ideally derived from a set of functionally characterized binding sites for the given transcription factor (6
7
8
9)
. A score derived from "matches" of the weight matrix to a DNA sequence can be used as an estimate of the relative efficiency of a specific transcription factor protein binding to the sequence. Binding affinity (and thus biological significance) is estimated to occur above a certain threshold score. Software that allows detection and characterization of individual binding sites is available from several sources, including Signal Scan (http://bimas.dcrt.nih.gov/molbio/signal) (10)
, MATRIX SEARCH (6)
, or MatInspector (http://www.gsf.de/biodv) (7)
. A large collection of functional binding sites derived from the literature can be found in the TRANSFAC database (web page http://transfac.gbf.de/TRANSFAC/) (9)
.
Binding site detection is clearly important for promoter regulation, and in yeast it has been helpful in elucidating promoter function. Pipel et al. used computational approaches based on DNA array data to study genome-wide transcriptional regulation (11)
. By combining DNA array data with annotation of the yeast genome, the authors were able to identify novel transcription factor motif combinations in the promoters of coregulated genes. These motifs appear to affect specific mRNA expression patterns during the cell cycle, sporulation, and in response to various stress responses. The authors observed what appears to represent regulatory cross-talk among these processes. These results elegantly demonstrate that a relatively small number of transcription factors can be responsible for a complex set of expression patterns in diverse conditions.
In this review, we will outline how to locate mammalian promoters, detail how promoter analysis, especially definition and detection of promoter modules, can be linked to cell- and tissue-specific transcriptional regulation, and demonstrate the viability of these concepts using an example of the RANTES/CCL5 chemokine promoter organization.
|
PROMOTER IDENTIFICATION
|
|---|
Obtaining the promoter regions for a series of genes is a prerequisite for promoter analysis but still a significant problem in many cases. A series of methods has been developed to identify promoter regions. Characterized promoters can often be obtained from various databases (annotated promoters in PubMed or the "Eukaryotic Promoter Database" of the ISREC (EPD, http://www.epd.isb-sib.ch). It is also possible to obtain promoter sequence from the direct mapping of cDNA into genomic sequences. Promoters are found immediately upstream of, or overlapping with, the site of transcriptional initiation (the CAP site at far 5' end of the mRNA). Thus, when the complete 5' untranslated mRNA sequence is available, promoter sequences can be reliably obtained from genomic sequences by exon mapping. However, care should be taken when using this approach, as many cDNA sequences do not include the complete 5' untranslated region. In addition, noncoding exons can often be found within the 5' untranslated regions of genes, which means that the true promoter region may be found well upstream of a partial 5' untranslated sequence. In the past few years promoter prediction programs have been generated that appear to able to identify 
50% of promoters from genomic sequence and thus can be used to locate some promoters directly in the genomic sequence if incomplete cDNA renders mapping ineffective (Dragon promoter finder: http://sdmc.krdl.org.sg/promoter; PromoterInspector: http://www.genomatix.de) (12
, 13)
.
|
PROMOTER MODULES AND FRAMEWORKS: MEDIATORS OF TISSUE AND SIGNAL-SPECIFIC EVENTS IN HIGHER ORGANISMS
|
|---|
Although binding site detection is important in higher organisms, it is generally not sufficient to elucidate promoter function. In more complex systems the functional transcription factor binding sites within promoters are organized hierarchically (14
, 15)
(Fig. 1
A). This dramatically increases the potential specificity and selectivity available to effect gene regulation (1
, 16)
. These organizational features appear to be the mammalian functional equivalent of individual elements in yeast (5)
. Combinatorial biology appears to be the key to understanding regulation in higher organisms where promoter function is determined more by the functional context within which the binding sites are located (5)
.
View larger version (26K):
[in this window]
[in a new window]
|
Figure 1. A) Promoters in higher eukaryotes are organized hierarchically and elements that control a specific pattern of expression may also be found in other promoters expressed under similar circumstances. B) Active promoters have a unique 3-dimensional structure. Changing the order or spacing of important transcription factor binding sites can change the overall structure of the promoter and thus effect transcription.
|
|
Promoter modules represent the next level of functional organization after individual transcription factor binding sites. We define promoter modules as two or more individual elements that act in a coordinated way (either synergistically or antagonistically) with the contributing elements arranged within a defined distance and sequential order (Fig. 1A, B
) (15
, 16)
. Work to date suggests that promoter modules can be pathway or cell type specific (15)
and, in this regard, can mediate the transcriptional response to specific signal transduction pathways (17
, 18)
, cell type-specific expression, and events central to developmental regulation (19)
. A given promoter module may show a robust stimulus-specific response in one tissue but in a second cell type may not be functional. This can result, for example, from a different complement or concentration of specific transcription factors or from selective signaling events further upstream in an activation pathway. Promoter modules can also exhibit cooperative protein binding and often include one binding element that represents a "poor" binding site for a specific transcription factor (17)
. Through cooperative effects, the stronger binding protein partner can stabilize the binding of the weaker partner, but the loss of either binding site abolishes function. This highlights an important aspect of transcriptional regulation, namely, that a weak binding site embedded in the correct context can be functionally as important as a strong binding site. Thus, scanning sequences with weight matrices and selection of statistically significant correlations between pairs of binding sites cannot reliably detect modules. Detecting several elements, one of which may not represent a strong binding site, can make detection of modules within promoter sequence using simple analysis for transcription factor binding sites a difficult process.
Although much can be learned concerning tissue and signal-specific gene regulation from the identification and characterization of promoter modules, they represent only one part of promoter architecture. Additional transcription elements working in concert with modules appear to fine-tune promoter regulation (15
, 20)
. These larger structures are referred to here as promoter frameworks (15)
. Frameworks associated with a specific biologic function or gene expression pattern are often conserved in promoters whereas unrelated binding sites exhibit random occurrence within these promoter sequences.
Eukaryotic polymerase type II promoters (the promoters that control most of the transcription of translated genes in the genome) are regulated by large transcription complexes composed of an ordered sequence of transcription factors. These complexes form a unique 3-dimensional arrangement. Changing the order, orientation, or distances between the transcription factors can have a profound effect on the overall structure of the initiation complex. Likewise, the conservation of organization, or promoter frameworks, can help to orchestrate the coregulation of genes (Fig. 1B
).
|
PROMOTER FEATURES CAN PROVIDE INFORMATION ABOUT THE BIOLOGIC CONTEXT OF EXPRESSION
|
|---|
Gene function has generally been thought to reside in the protein encoded by the gene. But as gene products usually require a series of other gene products to exert their biologic function, the coregulation of genes ensures that physically or functionally interacting proteins can appear together as a functional complex. For example, enzymes within a pathway interact functionally via the substrate but do not necessarily come into physical contact with each other. Genes expressed within the same biological context often share promoter modules/frameworks (1
, 4
, 5)
. As demonstrated on the RANTES/CCL5 example below using experimentally confirmed data, bioinformatics analysis can be used to expand our understanding of the biologic context of gene expression.
|
COMPARATIVE PROMOTER ANALYSIS
|
|---|
To date, the information gained from expression arrays has been used to reconstruct regulatory networks in a Boolean style (21)
. For example, it can be determined that gene X regulates gene Y positively or gene Z negatively. Though important, these analyses do not reveal how this regulation is achieved.
DNA expression arrays provide information about the RNA expression levels of genes at the time of analysis. By applying promoter analysis procedures it is possible to group coregulated genes based on their promoter organization. Cluster analysis of DNA array experiments provides the input data (22
, 23)
. By starting with these groups of coexpressed genes it is possible to derive organizational models essentially equivalent to the modules and frameworks derived functionally (22
, 23)
. Once the corresponding promoter sequences for the clustered cDNAs have been obtained, bioinformatics tools designed to sort through matrix libraries can generate potential models of the organization of transcription factors within the various promoter regions and in this way generate candidate promoter frameworks (22
, 23)
. The end result is essentially a reverse approach to that seen in the example of the RANTES promoter; a pattern of binding sites that is common to the compared promoter sequences can be identified whereas unrelated binding sites exhibit a random occurrence within these promoters.
This approach provides a powerful method for characterizing regulatory features of genes by linking them via promoter structures. Once a module or framework has been identified, it can then be used to construct a model to search for potential coregulated genes not found on the array analogous to what is presented below for the RANTES example. This can be an informative approach when attempting to identify or characterize additional target genes that may be affected within the context of an experiment. In this regard, it is important to consider the difference between coregulated as opposed to coexpressed genes. Coexpressed genes that are not coregulated may not necessarily share these promoter features.
|
THE RANTES/CCL5 PROMOTER: AN EXAMPLE OF PROMOTER FLEXIBILITY
|
|---|
The concepts detailed above can be illustrated using the promoter for the chemokine RANTES/CCL5. RANTES is involved in inflammatory protective reactions to tissue damage, infection, or foreign substances (24
25
26)
. Inflammation is a complex biological process that leads to the coordinated regulation of diverse sets of genes depending on where and how the stress occurs. RANTES/CCL5 is a member of the -CC- subfamily of chemotactic cytokines (chemokines) (27
, 28)
. This protein is expressed in a variety of tissues in response to different stimuli and plays diverse roles in inflammatory processes (27)
. The active promoter for the human RANTES/CCL5 gene has been functionally characterized in T cells (29
30
31
32)
, monocytes (17
, 18)
, astrocytes (33)
, and mesangial cells (C. Zischek, unpublished results; 34
). Individual cell type-specific transcription factors and binding elements were determined empirically by conventional molecular analysis, including DNase I protection assays, promoter-reporter gene transfection assays, site directed mutagenesis of elements in promoter-reporter gene assays, electrophoretic mobility shift assays (EMSA), EMSA competition, and EMSA supershift analysis. Transcriptional control of human RANTES/CCL5 appears to be mediated to a significant degree through six functionally characterized short regulatory elements (4
, 31
, 34)
. Figure 2
shows a summary of the individual control elements functionally identified as important for transcriptional control of RANTES/CCL5 in the four cell types studied. The six regions are not all functional in each of the cell types analyzed and the individual elements often contain overlapping binding specificities for different classes of transcription factors. Thus, different transcription factors can bind the same region in different tissues (4)
. This differential binding can be due to the level of expression of a given factor in a given tissue, alternate signal transduction cascades, and/or the transcription factor partners available for specific interaction (e.g., module formation). Promoters are dynamic switches, and the individual modules and frameworks used in a specific promoter can differ between different cell types that express the same gene. The promoter for RANTES/CCL5 is an excellent example of the flexible and complicated nature of promoters.
View larger version (51K):
[in this window]
[in a new window]
|
Figure 2. A summary of the individual control elements identified as important for transcriptional control of the RANTES/CCL5 gene in four different cell types. Molecular characterization has shown that much of this transcriptional control is encoded in <300 nucleotides of the immediate upstream region of the gene, but the combination of control regions used varies among the four different cell types.
|
|
A series of promoter modules has been determined to be functional in specific cell types for the RANTES promoter (17
, 18
, 29
30
31
32
, 34
, 35)
(C. Zischek, unpublished results). Examples of the various empirically defined promoter modules active in the different cell types in response to specific stimulation are shown in Table 1
and Fig. 3
. Module G (PU.1-Jun/CRE) and A-B (SP1-p50/p50) are active in monocytes stimulated with lipopolysaccharide (LPS) (17
, 18)
. The D-A module (IRF-1 and NF-
B) is functional in mesangial cells in response to stimulation with 
-interferon and tumor necrosis factor 
(C. Zischek, unpublished results). In T cells 5 to 7 days after activation of resting T cells, the A-B module binds a zinc finger protein RFLAT-1 at site A and Rel p50-p50 at site B (29
, 35)
. Astrocytes appear to use a module similar to that found in mesangial cells with IRF-1 factors binding to region D, interacting with Rel p50-p65 within region A (33)
.
View this table:
[in this window]
[in a new window]
|
Table 1. Different promoter modules/frameworks that have been identified for regulation of expression in the various cell typesa
|
|
View larger version (47K):
[in this window]
[in a new window]
|
Figure 3. Models that describe specific transcription factors, their orientation, and distances were developed based on the human RANTES/CCL5 promoter. The models were then used to search the human, rodent, "other vertebrate," and "other mammalian" sections of the "EMBL Nucleotide Sequence Database" of the European Bioinformatics Institute (Release 66, http://www.ebi.ac.uk/embl) and the "Eukaryotic Promoter Database" of the ISREC (EPD, http://www.epd.isb-sib.ch).
|
|
|
MODELING AFTER IDENTIFICATION OF PROMOTER FEATURES: IDENTIFICATION OF TARGETS
|
|---|
Promoter modules and frameworks are important in mediating tissue and signal-specific transcriptional responses. They tend to be conserved through evolution and appear to act independently, i.e., several functional modules may exist within a single promoter and may physically overlap to the extent of shared binding sites. Computer models representing the specific organization, distance, and orientation of the individual elements that make up the module/framework can be constructed and used to search databases for promoter sequences that show a similar organization (36)
. In this example, individual models that use distance parameters and optimized transcription factor matrices for the functionally identified factors important for transcription based on the experimentally verified binding sites of the RANTES promoter were used to search databases to detect other promoters that share the same functional organization (4
, 18)
. The results typically yield a set of promoters, a subset of which show coregulation with the starting gene.
The groups of ordered elements shown in Table 1
were used to derive overlapping cell type-specific submodels in silico (Fig. 3)
. The first models were constructed using experimentally verified information about individual binding factors and their corresponding binding sites within the RANTES/CCL5 promoter, represented by weight matrices as defined by analysis with MatInspector (for reviews about weight matrix detection methods, see refs 37
38
39
). The relative order, distance between binding elements, and their strand orientation were determined from the original human RANTES/CCL5 promoter sequence. The scoring algorithm ModelInspector (14
, 36)
combines matrix similarity measures of individual binding sites into a summary model score that was used to search individual databases. The individual models were started with default matrix similarity thresholds that were generally lower than the scores found with the human RANTES/CCL5 promoter sequence. All promoter modeling and subsequent database searches were carried out with the GEMS Launcher 1.0 software package, Genomatix Software GmbH, Munich (for more information about the tools and availability, see http://www.genomatix.de).
The four tissue-associated promoter models were used to search the human, rodent, "other vertebrate," and "other mammalian" sections of the "EMBL Nucleotide Sequence Database" of the European Bioinformatics Institute (Release 66, http://www.ebi.ac.uk/embl) and the "Eukaryotic Promoter Database" of the ISREC (EPD, http://www.epd.isb-sib.ch) (40)
. Matches to the various models were restricted to those occurring within annotated promoters or associated with 5' untranslated mRNA regions to facilitate evaluation.
The group of promoters identified for each model was then manually inspected for genes either previously characterized as coregulated with RANTES/CCL5 or genes generally associated with inflammation. These subgroups of promoters were subsequently used as refined training sets of sequences (from 4 to 10) to increase the selectivity of the models by tuning the distance ranges and matrix thresholds to more easily recognize the promoters within the set. As all of these sequences were already identified by the model-based search, they did not change the overall search results qualitatively. The final search results using the four models identified a combined total of 48 promoter matches (Table 2
).
View this table:
[in this window]
[in a new window]
|
Table 2. A series of promoters showing a similar organization of transcriptional elements identified using the various tissue-associated modelsa
|
|
|
ACTIVATED MESANGIAL CELL MODEL
|
|---|
Mesangial cells are smooth muscle-like cells found within the glomerulus of the kidney. The results of searches for the RANTES/CCL5 promoter/mesangial cell model (Fig. 3)
showed 21 promoter matches containing a similar organization of transcriptional elements (Table 2)
. Most of these genes can be categorized as inflammatory or kidney/smooth muscle related, and many appear to be functionally related. Dipeptidyl dipeptidase type IV is an endopeptidase expressed after TNF-/-IFN stimulation of mesangial cells (41)
. This protease uses RANTES/CCL5 as a substrate and "tunes" the bioactivity of RANTES/CCL5 through post-translational processing (42)
. The cleavage product generated is a potent antagonist for several of the RANTES/CCL5 receptors (42)
. Angiotensin II induces RANTES/CCL5 expression in glomerular cells and inhibits expression of interleukin 1ß (IL-1ß) -mediated nitric oxide production in mesangial cells (43)
. IL-1ß stimulation, in turn, up-regulates nitric oxide synthase expression in vascular smooth muscle cells (44
, 45)
. Nitric oxide can suppress the angiotensin II induced migration of smooth muscle cells (46)
and superoxide dismutase will inhibit iNOS expression in mesangial cells (47)
.
Most of these genes have already been described as being expressed by renal tissues (e.g., endothelial cells, podocytes, or tubular epithelial cells) during inflammatory processes. This may represent an example of "functional context," where the functional framework identified in the RANTES/CCL5 promoter used by activated mesangial cells forms the basis for aspects of orchestrated gene regulation seen within microenvironments of the kidney (4)
.
|
T CELL-DERIVED MODELS
|
|---|
Most of the promoters identified using the RANTES/CCL5 T cell model (Fig. 3)
represent genes expressed by activated T cells (Table 2)
. Programmed death-1 gene (PD-1) is expressed upon T cell activation and is thought to play an inhibitory role in immune responses. Members of the B7 family are ligands for PD-1. This family of costimulatory and inhibitory receptors and their corresponding ligands play a critical role in the modulation of immune processes and tolerance (48)
. Members of the Bcl-2 protein family such as Bim can trigger the apoptosis of T cells. Bim is thought to be required for hematopoietic homeostasis and may act as a barrier to T cell-mediated autoimmunity (49)
. Human cationic antimicrobial protein 18 (CAP18) is expressed by T cells 5 days after stimulation with IL-2 (50)
. This antimicrobial peptide can mobilize leukocytes in host defense, and cleavage products of CAP18 show chemotactic activity for polymorphonuclear leukocytes and CD4 T lymphocytes (50)
. The urokinase-type plasminogen activator (u-PA)/plasmin system plays important roles in promoting T cell migration and invasion as well as activation and proliferation (51)
. The activation and translocation of protein kinases such as PKC are key events in the regulation of T lymphocyte activation, proliferation, and function (52)
. The suppressor of cytokine signaling 3 (SOCS-3) is a T helper lineage marker and has been implicated in the negative regulation of several cytokine pathways, particularly the receptor-associated tyrosine kinase/signal transducer and activator of transcription (AK/STAT) pathways of transcriptional activation (53)
. Serum amyloid A 1 is synthesized in response to inflammatory cytokines and represents a sensitive indicator for assessing inflammatory activity, for example, during viral infection or allograft rejection (54)
.
|
IL-1 ß-STIMULATED ASTROCYTE MODEL
|
|---|
The single largest group of promoters identified using the RANTES/CCL5 astrocyte cell model (Fig. 3)
was other chemokine promoters (Table 2)
. Chemokines represent a large family of proteins (27
, 28)
. Rather than acting as single factors, proinflammatory chemokines are thought to be regulated in coordinate groups, which in turn activate common chemokine receptors (27
, 28)
. Four chemokine genes (including RANTES) were identified in the analysis of the astrocyte model (Table 2)
. Of these, eotaxin shares a receptor (CCR3) with RANTES (27
, 28)
. The chemokine IP-10 can selectively recruit similar T cell subpopulations (27
, 28)
. Murine Gro and rat KC are CXC chemokine orthologs that are important in the control of acute inflammatory events (4
, 27)
.
Stromal interaction molecule 1 (STIM1) was also identified in this search and is a type I cell surface transmembrane glycoprotein implicated in tumor growth control and stromal-hematopoietic cell interactions (55)
. Astrocytes have been shown to express the androgen receptor (56)
.
|
MONOCYTE/LPS MODEL
|
|---|
Only four genes were identified with the monocyte/LPS model (Fig. 3
; Table 2
). One of these was a phospholipid transfer protein. Plasma lipoproteins are linked to the biologic response to bacterial LPS. The movement of LPS from leukocytes into lipoproteins is thought to help attenuate host responses to LPS in vivo (57)
. The promoter for the murine chemokine MIP-2 was also identified with this search. This chemokine is important in mediating early inflammatory events and may represent a functional ortholog to human RANTES in some biologic conditions (4
, 58
, 59)
. MIP-2 is also up-regulated in monocytes in response to LPS (4
, 60)
.
The analysis of promoters for organizational features provides a link between the genomic nucleotide sequence and important aspects of gene regulation and expression.
|
CONCLUSIONS
|
|---|
The focus of bioinformatics has begun to extend from the identification of genes toward understanding how the expression and regulation of genes is orchestrated on a genomic level. The regulation of gene transcription is based in large part on the interaction between transcription factors and the genomic regulatory regions of genes within the genome. A given transcription factor can have different effects on different promoters. The information governing how transcription factors influence gene expression is laid down in the regulatory genomic sequences, not in the proteins themselves, and thus a tremendous amount of information can be mined from regulatory regions. Linking the results of functional analysis of gene regulation can allow rapid identification of a series of potential coregulated genes and thus facilitate target gene characterization and identification. Application of these methods to the analysis of coordinated gene expression by large-scale mRNA/cDNA screening systems (DNA arrays and chips) can identify the organizational features of promoters responsible for these expression patterns and provide information about the regulatory networks at work in the cell. Computer modeling of promoter organization is an important new tool in the study of transcriptional regulation and promoter regulatory network by quickly identifying potential regulatory regions and providing information about the functional context of gene expression.
|
ACKNOWLEDGMENTS
|
|---|
Support was provided by grants to P.J.N. from the DFG Sonderforschungsbereich 571 and 469 and by DFG grant "Informatic methods for the analysis and interpretation of large amounts of genomic data" (grant #2370/1-1) to H.M. and T.W.
Received for publication November 26, 2002.
Accepted for publication March 10, 2003.
|
REFERENCES
|
|---|
- Werner, T. (1999) Models for prediction and recognition of eukaryotic promoters. Mamm. Genome 10,168-175[CrossRef][Medline]
- Frisch, M., Frech, K., Klingenhoff, A., Cartharius, K., Liebich, I., Werner, T. (2002) In silico prediction of scaffold/matrix attachment regions in large genomic sequences. Genome Res. 12,349-354[Abstract/Free Full Text]
- Robertson, K. D. (2002) DNA methylation and chromatin—unraveling the tangled web. Oncogene 21,5361-5379[CrossRef][Medline]
- Fessele, S., Maier, H., Zischek, C., Nelson, P. J., Werner, T. (2002) Regulatory context is a crucial part of gene function. Trends Genet. 18,60-63[CrossRef][Medline]
- Werner, T. (2001) The promoter connection. Nat. Genet. 29,105-106[CrossRef][Medline]
- Chen, Q. K., Hertz, G. Z., Stormo, G. D. (1995) MATRIX SEARCH 1.0: a computer program that scans DNA sequences for transcriptional elements using a database of weight matrices. Comput. Appl. Biosci. 11,563-566[Abstract/Free Full Text]
- Quandt, K., Frech, K., Karas, H., Wingender, E., Werner, T. (1995) MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res. 23,4878-4884[Abstract/Free Full Text]
- Kel, A. E., Kondrakhin, Y. V., Kolpakov Ph, A., Kel, O. V., Romashenko, A. G., Wingender, E., Milanesi, L., Kolchanov, N. A. (1995) Computer tool FUNSITE for analysis of eukaryotic regulatory genomic sequences. Proc. Int. Conf. Intell. Syst. Mol. Biol. 3,197-205[Medline]
- Heinemeyer, T., Chen, X., Karas, H., Kel, A. E., Kel, O. V., Liebich, I., Meinhardt, T., Reuter, I., Schacherer, F., Wingender, E. (1999) Expanding the TRANSFAC database towards an expert system of regulatory molecular mechanisms. Nucleic Acids Res. 27,318-322[Abstract/Free Full Text]
- Prestridge, D. S. (2000) Computer software for eukaryotic promoter analysis. Methods Mol. Biol. 130,265-295[Medline]
- Pilpel, Y., Sudarsanam, P., Church, G. M. (2001) Identifying regulatory networks by combinatorial analysis of promoter elements. Nat. Genet. 29,153-159[CrossRef][Medline]
- Scherf, M., Klingenhoff, A., Frech, K., Quandt, K., Schneider, R., Grote, K., Frisch, M., Gailus-Durner, V., Seidel, A., Brack-Werner, R., et al (2001) First pass annotation of promoters on human chromosome 22. Genome Res. 11,333-340[Abstract/Free Full Text]
- Scherf, M., Klingenhoff, A., Werner, T. (2000) Highly specific localization of promoter regions in large genomic sequences by PromoterInspector: a novel context analysis approach. J. Mol. Biol. 297,599-606[CrossRef][Medline]
- Klingenhoff, A., Frech, K., Quandt, K., Werner, T. (1999) Functional promoter modules can be detected by formal models independent of overall nucleotide sequence similarity. Bioinformatics 15,180-186[Abstract/Free Full Text]
- Klingenhoff, A., Frech, K., Werner, T. (2002) Regulatory modules shared within gene classes as well as across gene classes can be detected by the same in silico approach. In Silico Biol. 2,S17-S26[Medline]
- Firulli, A. B., Olson, E. N. (1997) Modular regulation of muscle gene transcription: a mechanism for muscle cell diversity. Trends Genet. 13,364-369[CrossRef][Medline]
- Boehlk, S., Fessele, S., Mojaat, A., Miyamoto, N. G., Werner, T., Nelson, E. L., Schlondorff, D., Nelson, P. J. (2000) ATF and Jun transcription factors, acting through an Ets/CRE promoter module, mediate lipopolysaccharide inducibility of the chemokine RANTES in monocytic Mono Mac 6 cells. Eur. J. Immunol. 30,1102-1112[CrossRef][Medline]
- Fessele, S., Boehlk, S., Mojaat, A., Miyamoto, N. G., Werner, T., Nelson, E. L., Schlondorff, D., Nelson, P. J. (2001) Molecular and in silico characterization of a promoter module and C/EBP element that mediate LPS-induced RANTES/CCL5 expression in monocytic cells. FASEB J. 15,577-579[Free Full Text]
- Wang, Q., Sigmund, C. D., Lin, J. J. (2000) Identification of cis elements in the cardiac troponin T gene conferring specific expression in cardiac muscle of transgenic mice. Circ. Res. 86,478-484[Abstract/Free Full Text]
- Brazma, A., Jonassen, I., Eidhammer, I., Gilbert, D. (1998) Approaches to the automatic discovery of patterns in biosequences. J. Comput. Biol. 5,279-305[Medline]
- D’Haeseleer, P., Liang, S., Somogyi, R. (2000) Genetic network inference: from co-expression clustering to reverse engineering. Bioinformatics 16,707-726[Abstract/Free Full Text]
- Werner, T. (2001) Target gene identification from expression array data by promoter analysis. Biomol. Eng. 17,87-94[CrossRef][Medline]
- Werner, T. (2001) Cluster analysis and promoter modelling as bioinformatics tools for the identification of target genes from expression array data. Pharmacogenomics 2,25-36[CrossRef][Medline]
- Segerer, S., Nelson, P. J., Schlondorff, D. (2000) Chemokines, chemokine receptors, and renal disease: from basic science to pathophysiologic and therapeutic studies. J. Am. Soc. Nephrol. 11,152-176[Abstract/Free Full Text]
- Koning, G. A., Schiffelers, R. M., Storm, G. (2002) Endothelial cells at inflammatory sites as target for therapeutic intervention. Endothelium 9,161-171[CrossRef][Medline]
- Bajetto, A., Bonavia, R., Barbero, S., Schettini, G. (2002) Characterization of chemokines and their receptors in the central nervous system: physiopathological implications. J. Neurochem. 82,1311-1329[CrossRef][Medline]
- Nelson, P. J., Krensky, A. M. (2001) Chemokines, chemokine receptors, and allograft rejection. Immunity 14,377-386[CrossRef][Medline]
- Murphy, P. M., Baggiolini, M., Charo, I. F., Hebert, C. A., Horuk, R., Matsushima, K., Miller, L. H., Oppenheim, J. J., Power, C. A. (2000) International union of pharmacology. XXII. Nomenclature for chemokine receptors. Pharmacol. Rev. 52,145-176[Abstract/Free Full Text]
- Nelson, P. J., Ortiz, B. D., Pattison, J. M., Krensky, A. M. (1996) Identification of a novel regulatory region critical for expression of the RANTES chemokine in activated T lymphocytes. J. Immunol. 157,1139-1148[Abstract]
- Ortiz, B. D., Krensky, A. M., Nelson, P. J. (1996) Kinetics of transcription factors regulating the RANTES chemokine gene reveal a developmental switch in nuclear events during T-lymphocyte maturation. Mol. Cell. Biol. 16,202-210[Abstract]
- Ortiz, B. D., Nelson, P. J., Krensky, A. M. (1997) Switching gears during T-cell maturation: RANTES and late transcription. Immunol. Today 18,468-471[CrossRef][Medline]
- Moriuchi, H., Moriuchi, M., Fauci, A. S. (1997) Nuclear factor-kappa B potently up-regulates the promoter activity of RANTES, a chemokine that blocks HIV infection. J. Immunol. 158,3483-3491[Abstract]
- Miyamoto, N. G., Medberry, P. S., Hesselgesser, J., Boehlk, S., Nelson, P. J., Krensky, A. M., Perez, H. D. (2000) Interleukin-1beta induction of the chemokine RANTES promoter in the human astrocytoma line CH235 requires both constitutive and inducible transcription factors. J. Neuroimmunol. 105,78-90[CrossRef][Medline]
- Nelson, P. J., Pattison, J. M., Krensky, A. M. (1997) Gene expression of RANTES. Methods Enzymol. 287,148-162[CrossRef][Medline]
- Song, A., Chen, Y. F., Thamatrakoln, K., Storm, T. A., Krensky, A. M. (1999) RFLAT-1: a new zinc finger transcription factor that activates RANTES gene expression in T lymphocytes. Immunity 10,93-103[CrossRef][Medline]
- Frech, K., Werner, T. (1997) Specific modelling of regulatory units in DNA sequences. Pac. Symp. Biocomput ,151-162
- Frech, K., Quandt, K., Werner, T. (1997) Finding protein-binding sites in DNA sequences: the next generation. Trends Biochem. Sci. 22,103-104[CrossRef][Medline]
- Frech, K., Quandt, K., Werner, T. (1997) Software for the analysis of DNA sequence elements of transcription. Comput. Appl. Biosci. 13,89-97[Abstract/Free Full Text]
- Stormo, G. D. (2000) DNA binding sites: representation and discovery. Bioinformatics 16,16-23[Abstract/Free Full Text]
- Perier, R. C., Praz, V., Junier, T., Bonnard, C., Bucher, P. (2000) The eukaryotic promoter database (EPD). Nucleic Acids Res. 28,302-303[Abstract/Free Full Text]
- Shinosaki, T., Kobayashi, T., Kimura, K., Kurihara, H. (2002) Involvement of dipeptidyl peptidase IV in immune complex-mediated glomerulonephritis. Lab. Invest. 82,505-513[CrossRef][Medline]
- Iwata, S., Yamaguchi, N., Munakata, Y., Ikushima, H., Lee, J. F., Hosono, O., Schlossman, S. F., Morimoto, C. (1999) CD26/dipeptidyl peptidase IV differentially regulates the chemotaxis of T cells and monocytes toward RANTES: possible mechanism for the switch from innate to acquired immune response. Int. Immunol. 11,417-426[Abstract/Free Full Text]
- Wolf, G. (1998) Angiotensin II is involved in the progression of renal disease: importance of non-hemodynamic mechanisms. Nephrologie 19,451-456[Medline]
- Wong, H. R., Finder, J. D., Wasserloos, K., Lowenstein, C. J., Geller, D. A., Billiar, T. R., Pitt, B. R., Davies, P. (1996) Transcriptional regulation of iNOS by IL-1 beta in cultured rat pulmonary artery smooth muscle cells. Am. J. Physiol. 271,L166-L171
- Taniuchi, M., Otani, H., Kodama, N., Tone, Y., Sakagashira, M., Yamada, Y., Mune, M., Yukawa, S. (1999) Lysophosphatidylcholine up-regulates IL-1 beta-induced iNOS expression in rat mesangial cells. Kidney Int. Suppl. 71,S156-S158[CrossRef][Medline]
- Nakayama, I., Kawahara, Y., Tsuda, T., Okuda, M., Yokoyama, M. (1994) Angiotensin II inhibits cytokine-stimulated inducible nitric oxide synthase expression in vascular smooth muscle cells. J. Biol. Chem. 269,11628-11633[Abstract/Free Full Text]
- Beck, K. F., Eberhardt, W., Walpen, S., Apel, M., Pfeilschifter, J. (1998) Potentiation of nitric oxide synthase expression by superoxide in interleukin 1 beta-stimulated rat mesangial cells. FEBS Lett. 435,35-38[CrossRef][Medline]
- Freeman, G. J., Long, A. J., Iwai, Y., Bourque, K., Chernova, T., Nishimura, H., Fitz, L. J., Malenkovich, N., Okazaki, T., Byrne, M. C., et al (2000) Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. J. Exp. Med. 192,1027-1034[Abstract/Free Full Text]
- Bouillet, P., Metcalf, D., Huang, D. C., Tarlinton, D. M., Kay, T. W., Kontgen, F., Adams, J. M., Strasser, A. (1999) Proapoptotic Bcl-2 relative Bim required for certain apoptotic responses, leukocyte homeostasis, and to preclude autoimmunity. Science 286,1735-1738[Abstract/Free Full Text]
- Agerberth, B., Charo, J., Werr, J., Olsson, B., Idali, F., Lindbom, L., Kiessling, R., Jornvall, H., Wigzell, H., Gudmundsson, G. H. (2000) The human antimicrobial and chemotactic peptides LL-37 and alpha-defensins are expressed by specific lymphocyte and monocyte populations. Blood 96,3086-3093[Abstract/Free Full Text]
- Gyetko, M. R., Libre, E. A., Fuller, J. A., Chen, G. H., Toews, G. (1999) Urokinase is required for T lymphocyte proliferation and activation in vitro. J. Lab. Clin. Med. 133,274-288[CrossRef][Medline]
- Yahata, T., Abe, N., Yahata, C., Ohmi, Y., Ohta, A., Iwakabe, K., Habu, S., Yagita, H., Kitamura, H., Matsuki, N., et al (1999) The essential role of phorbol ester-sensitive protein kinase C isoforms in activation-induced cell death of Th1 cells. Eur. J. Immunol. 29,727-732[CrossRef][Medline]
- Egwuagu, C. E., Yu, C. R., Zhang, M., Mahdi, R. M., Kim, S. J., Gery, I. (2002) Suppressors of cytokine signaling proteins are differentially expressed in Th1 and Th2 cells: implications for Th cell lineage commitment and maintenance. J. Immunol. 168,3181-3187[Abstract/Free Full Text]
- Yamada, T. (1999) Serum amyloid A (SAA): a concise review of biology, assay methods and clinical usefulness. Clin. Chem. Lab. Med. 37,381-388[CrossRef][Medline]
- Sabbioni, S., Veronese, A., Trubia, M., Taramelli, R., Barbanti-Brodano, G., Croce, C. M., Negrini, M. (1999) Exon structure and promoter identification of STIM1 (alias GOK), a human gene causing growth arrest of the human tumor cell lines G401 and RD. Cytogenet. Cell Genet. 86,214-218[CrossRef][Medline]
- Finley, S. K., Kritzer, M. F. (1999) Immunoreactivity for intracellular androgen receptors in identified subpopulations of neurons, astrocytes and oligodendrocytes in primate prefrontal cortex. J. Neurobiol. 40,446-457[CrossRef][Medline]
- Kitchens, R. L., Wolfbauer, G., Albers, J. J., Munford, R. S. (1999) Plasma lipoproteins promote the release of bacterial lipopolysaccharide from the monocyte cell surface. J. Biol. Chem. 274,34116-34122[Abstract/Free Full Text]
- Gao, J. L., Wynn, T. A., Chang, Y., Lee, E. J., Broxmeyer, H. E., Cooper, S., Tiffany, H. L., Westphal, H., Kwon-Chung, J., Murphy, P. M. (1997) Impaired host defense, hematopoiesis, granulomatous inflammation and type 1-type 2 cytokine balance in mice lacking CC chemokine receptor 1. J. Exp. Med. 185,1959-1968[Abstract/Free Full Text]
- Gao, J. L., Murphy, P. M. (1995) Cloning and differential tissue-specific expression of three mouse beta chemokine receptor-like genes, including the gene for a functional
TOP
ABSTRACT
INTRODUCTION
