Bioinformatics Advance Access originally published online on March 15, 2005
Bioinformatics 2005 21(11):2623-2628; doi:10.1093/bioinformatics/bti387
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Promoter modeling: the case study of mammalian histone promoters
1Knowledge Extraction Lab, Institute for Infocomm Research 21 Heng Mui Keng Terrace, Singapore 119613
2Department of Statistics and Applied Probability, National University of Singapore 3 Science Drive 2, Singapore 117543
3Department of Molecular Biology, Biochemistry and Molecular Cell Biology, University of Göttingen Humboldtallee 23, 37073 Göttingen, Germany
*To whom correspondence should be addressed.
 |
Abstract |
|---|
 TOP Abstract 1 INTRODUCTION2 METHODS3 RESULTS AND DISCUSSIONREFERENCES |
|---|
Motivation: Histone proteins play important roles in chromosomal functions. They are significantly evolutionarily conserved across species, which suggests similarity in their transcription regulation. The abundance of experimental data on histone promoters provides an excellent background for the evaluation of computational methods. Our study addresses the issue of how well computational analysis can contribute to unveiling the biologically relevant content of promoter regions for a large number of mammalian histone genes taken across several species, and suggests the consensus promoter models of different histone groups.
Results: This is the first study to unveil the detailed promoter structures of all five mammalian histone groups and their subgroups. This is also the most comprehensive computational analysis of histone promoters performed to date. The most exciting fact is that the results correlate very well with the biologically known facts and experimental data. Our analysis convincingly demonstrates that computational approach can significantly contribute to elucidation of promoter content (identification of biologically relevant signals) complementing tedious wet-lab experiments. We believe that this type of analysis can be easily applied to other functional gene classes, thus providing a general framework for modelling promoter groups. These results also provide the basis to hunt for genes co-regulated with histone genes across mammalian genomes.
Contact: bajicv{at}i2r.a-star.edu.sg
Supplementary information: http://research.i2r.a-star.edu.sg/promoter/histones/PubMed15769833.pdf
|
1 INTRODUCTION |
|---|
|
TOPAbstract1 INTRODUCTION2 METHODS3 RESULTS AND DISCUSSIONREFERENCES |
|---|
Histones are basic proteins present in the eukaryotic cell nucleus and are broadly divided into five classes, namely H1, H2A, H2B, H3 and H4 (Wolffe, 1998). These proteins range between 220 (H1) and 102 (H4) amino acids in length and help in packaging DNA in the chromatin complex. Chromatin has a highly organized structure with its basic structural unit of nucleosome core which consists of about 146 bp of DNA wrapped around the histone octamer containing two molecules each of H2A, H2B, H3 and H4 (Luo and Dean, 1999). H1 histone, known also as linker histone, seals the two rounds of DNA on the surface of the histone octamer (Wolffe, 1998). Nucleosome core, H1 histone and the linker DNA that connects two adjacent nucleosome cores form the nucleosome. These repeated particles form the fundamental repeating unit of chromatin as a first order of compaction. This array of nucleosomes is organized in higher-order structures as chromatin or nuclear assembly. Being associated with the chromosomal structure, histones play an essential role in chromosomal processes such as gene transcription, regulation, chromosome condenzation, recombination and replication (Doenecke et al., 1997). All histones, except H4, consist of several subtypes which differ from each other in their primary protein structure. For example, linker histone H1 has nine subtypes named H1.1 to H1.5, H1o, H1oo, H1t and H1X (Albig and Doenecke, 1997; Tanaka et al., 2003; Yamamoto and Horikoshi, 1996). Similarly, several subtypes have been reported for H2A, H2B and H3 histones (Albig and Doenecke, 1997; Doenecke et al., 1997).
Histones are evolutionarily conserved and have similar functions in all living organisms. However, the degree of conservation varies among species and within the species. Among the different histone types, the H3 and H4 histones are known to be highly conserved during evolution, while histone H1 is the least evolutionarily conserved out of all histone groups (Freeman et al., 1996; Imhof and Becker, 2001). Due to the unique functions that histone proteins have in all species, it makes sense to assume that many of their genes are expressed under similar conditions. This joint regulation of histone genes is controlled both transcriptionally and posttranscriptionally (Sanchez and Marzluff, 2002; Doenecke et al., 1994). Since part of this control is exerted at the promoters of the individual histone genes, the co-expression of histone genes implies that their promoters contain a number of common features.
To understand the mechanism and conditions of gene activation, we need to unravel the promoter content of the gene. In this study we aim at examining how well application of some bioinformatics tools can help in elucidating the promoter content of a large number of mammalian histone genes. Determining promoter content usually involves searching for and characterizing potentially functional regulatory binding sites in them. This may be done using experimental or computational methods or both. Histone gene promoters have been widely studied experimentally in both animals and plants. We will focus on mammalian histone genes from humans, mice and rats. There have been various studies in the past that have attempted to characterize individual histone gene subtype promoters in mammals through experimental approaches. These include studies on histone groups H1 (Meergans et al., 1998; Osley, 1991; Dong et al., 1995; Peretti and Khochbin, 1997; Doenecke and Alonso, 1996; Grimes et al., 2003; Wolfe and Grimes, 2003a, b), H2A (Albig et al., 1999; Trappe et al., 1999; Yagi et al., 1995; Oswald et al., 1996), H2B (Albig et al., 1999; Trappe et al., 1999), H3 (Wells et al., 1987; Witt et al., 1996, 1997, 1998; Frank et al., 2003; Koessler et al., 2003) and H4 (van Wijnen et al., 1992, 1996; Last et al., 1998; Stein et al., 1996). The experimental studies, though accurate, require lot of resources and are time consuming. Moreover, these studies on histone promoters have generally been restricted to single promoters. Little attention has been given to study all histone promoter types on a large scale. This need of studying collections of histone promoters simultaneously underlines the importance of employing the computational methods that are relatively fast and cheap, and may complement the experimental approach. Many researchers have in the past employed computational methods to analyze promoter structures of specific gene groups and derived common regulatory elements that are characteristic for these groups. For example, several gene families have been analyzed for potential promoter structures such as glucocorticoid and heat-shock responsive genes (Claverie and Sauvaget, 1985), globin genes (Staden, 1988), muscle-specific genes (Wasserman and Fickett, 1998), liver-specific genes (Krivan and Wasserman, 2001) and actin genes (Klingenhoff et al., 2002) (see also Werner, 2002, 2003). The computational analysis of mammalian histone promoters is relatively less researched and has been generally confined to the analysis of specific histone types. For example, these include studies on histone H1 promoters (Werner, 2001; Duncliffe et al., 1995), histone H1o promoters (Dong et al., 1995), histone H3 promoters (Koessler et al., 2004), histone H3.3A promoters (Frank et al., 2003) and histone H3.3B promoters (Albig et al., 1995).
We have investigated promoter structures of the region covering upstream [–250, –1] genomic segments relative to transcription start sites (TSSs) for 127 histone genes from three mammalian species (human, mouse, rat) in order to find out if they share any significant level of similarity and to identify motifs likely to be potential transcription factor binding sites (TFBSs). Our hypothesis has been that, due to specific cellular functions complemented with a high level of protein conservation, histone genes are co-regulated and, therefore, we expect promoters of different histone groups to share common regulatory components. We were able to identify across species common elements in the considered promoter regions of genes within different histone groups and across these groups. We have also tried to verify results obtained from our computational study with the promoter features known from experiments. Most of the promoter motifs and their consensus models that we have determined match fairly well with the experimental data. For example, we were able to find in histone H1 group a consensus motif pattern that represents functional TFBSs ordered in a specific manner upstream from the TSS as: TATA-box, CCAAT-box, GC-box, AC-box and E2F-box. This order of TFBSs is specific to cell cycle-dependent H1 histone genes (Meergans et al., 1998; Duncliffe et al., 1995; Werner, 2001). All the motifs that we have found generally correspond well with the known TFBS in terms of composition and position. We were able to determine the promoter consensus motif models for all five examined histone gene groups and their subgroups.
Our study addresses the issue of how well computational analysis can contribute to unveiling the biologically relevant content of promoter regions for a large number of mammalian histone genes taken across several species. To the best of our knowledge, this is the most comprehensive computational analysis of promoter structures of mammalian histone genes performed to date. Analysis of the results reveals that those motifs that are over-represented in histone groups generally tend to be detected computationally as compared to those specific motifs which occur less frequently. The most exciting fact is that the results correlate very well with the biologically known facts and experimental data. The results open a possibility to further deepen our understanding of transcriptional control of mammalian histone genes. These results also provide the basis to hunt for genes co-regulated with the histone genes across mammalian genomes. The study is not limited in any way to histone promoters and the methods used have a general applicability.
|
2 METHODS |
|---|
|
TOPAbstract1 INTRODUCTION2 METHODS3 RESULTS AND DISCUSSIONREFERENCES |
|---|
2.1 Collection and pre-processing of promoter data
Using PromoSer tool release 3.0 (Halees and Weng, 2004) and FIE2.1, a new version of FIE2 tool (Chong et al., 2003), we collected 127 histone promoter sequences covering [–250, –1] segments relative to TSS. These sequences contained 19 H1 (9 human, 8 mouse, 2 rat), 29 H2A (15 human, 12 mouse, 2 rat), 32 H2B (17 human, 13 mouse, 2 rat), 23 H3 (11 human, 11 mouse, 1 rat) and 24 H4 (13 human, 10 mouse, 1 rat) histone types. FIE2.1 was also used to find the translation initiation site (TIS) position information in the 127 histone gene sequences. (See Appendix 1 of Supplementary Material for details, and Appendix 3 for promoter data.)
2.2 Ab initio motif discovery
There are many software programs that follow an ab initio approach for DNA motif discovery (Tompa et al., 2004). Usually, such programs are applied in the analysis of orthologous or co-regulated genes because these are assumed to have common transcription control motifs, thus standing a better chance of motif detection due to presumed overabundance of these motifs in the dataset. Thus, by selecting orthologous or co-regulated genes, the signal-to-noise ratio in many cases improves. Due to its relatively flexible parameter selection procedure, we have chosen MEME/MAST (Bailey and Elkan, 1994; Bailey and Gribskov, 1998) in our study of finding potentially functional motifs in histone promoters. (See Appendix 2 of Supplementary Material for details.) The results of motif analyses are shown in Supplementary Figure 1.
|
3 RESULTS AND DISCUSSION |
|---|
|
TOPAbstract1 INTRODUCTION2 METHODS3 RESULTS AND DISCUSSIONREFERENCES |
|---|
3.1 Motifs and their distributions in histone promoters
The MEME program applied to 127 histone gene promoters detected nine statistically significant motifs. In order to verify the potential biological relevance of the motifs discovered in promoter regions [–250, –1] (Supplementary Fig. 1), we referred to experimentally proven biological data of TRANSFAC (Matys et al., 2003). Using PATCH, a program accompanying the free online version of TRANSFAC 6.0, we found that Motif 2 represents Oct-1 box (TRANSFAC site no. R00662 [GenBank] ) with associated factor OTF-1 (Fletcher et al., 1987). Motif 3 corresponds to TATA-box (TRANSFAC site no. R00770 [GenBank] ) with associated factors, TBP and TFIID (Nakajima et al., 1988). Motif 4 perfectly corresponds to E2F-binding site (TRANSFAC site no. R00798 [GenBank] ) with associated E2F-1 factor (Oswald et al., 1996). Motif 5 resembles H4-box, a H4TF2 binding site ( TRANSFAC site no. R00681 [GenBank] ) with associated factor H4TF2/HinF-P (Pauli et al., 1987; Mitra et al., 2003; La Bella and Heintz, 1991). Motif 6 resembles AC-box (TRANSFAC site no. R00658 [GenBank] ) with associated factors, H1TF1 (La Bella et al., 1989), HiNF-A (van Wijnen et al., 1988b), HiNF-D (van Wijnen et al., 1996; Grimes et al., 2003). Motif 9 corresponds to the GC-box (TRANSFAC site no. R00684 [GenBank] ) with associated factor HiNF-C (van Wijnen et al., 1989), Sp1 (Courey and Tjian, 1988) and Sp3 (Birnbaum et al., 1995; Hagen et al., 1994). Motifs 1, 7 and 8 resemble the CCAAT-box (TRANSFAC site nos. R00659 [GenBank] , R00660 [GenBank] ) with associated factors, H1TF1 and H1TF2 (La Bella et al., 1989; Martinelli and Heintz, 1994; Gallinari et al., 1989), HiNF-B (van Wijnen et al., 1988a,b), NF-Y (Mantovani, 1999) and HiNF-D (van Wijnen et al., 1996; Grimes et al., 2003). These experimental findings are supportive of the approach we have used. We summarize the motifs discovered in our analysis in Table 1 along with their annotation from TRANSFAC for putative binding sites and transcription factors they might represent.
|
In order to find the relative strength of a motif for a particular histone group, we calculated the normalized motif frequency for each motif in a group by dividing the number of sequences in a group with that motif by the total number of sequences in the group. This expression allowed us to assess the relative strength of motif conservation with respect to a particular histone group as a whole (Fig. 1). A high value of normalized motif frequency indicates that the motif is well conserved across most members of a histone group, whereas a low value might indicate that the motif is conserved only in a subgroup within a histone group. According to the expression pattern that a subgroup exhibits, they could be representing: (i) S-phase of the cell cycle/DNA-replication dependent genes that are normally active during the cell proliferating stage of development such as in fetal tissues, (ii) cell-cycle independent or basally expressed replacement histone genes that tend to express in resting, differentiated cells such as in adult tissues, and (iii) tissue-specific genes that are expressed only in particular tissues such as in germinal testis and ovary tissues. In Figure 1. we present relative strengths of individual motifs in each of the five histone groups in terms of the normalized motif frequency. We observe that there are certain motifs that are specific to a particular histone group, while there are others that are shared between different histone groups. This indicates discriminatory as well as a common nature of transcriptional regulatory elements of histone promoters. Shared motifs between groups suggest common regulatory mechanisms for genes sharing those motifs, while specific motifs within a group suggest specific regulatory channels that may be required for gene transcription. We observe, for example, that Motif 5 (H4TF2-binding site) is highly specific to histone H4 group and is present in relatively less strength in histone H1 and has almost no presence in H2A, H2B and H3 histone groups. Further, we observe that within histone H1 group, Motif 5 is exclusively present in histone H1o subgroup. These observations are well supported by experimental studies where H4TF2-binding site (H4-box) is found in H4 (La Bella and Heintz, 1991; Mitra et al., 2003) and H1o (Dong et al., 1995; Peretti and Khochbin, 1997) histone genes. H4-box in histone H1o replaces CCAAT (Dong et al., 1995) normally found in somatic H1 genes. Motif 2 (Oct-1 binding site) is another such motif which is group-specific, present mostly in H2A and H2B and to a lesser extent in H3 groups. This is consistent with the finding that Oct-1 element is present in histone H2A/H2B promoter (Albig et al., 1999; Trappe et al., 1999) and histone H3.3B promoter (Witt et al., 1997; Frank et al., 2003). All the remaining seven motifs (Motifs 1, 3, 4, 6, 7, 8 and 9) are present in all the histone groups. However their relative presence in each group varies as is evident from their normalized motif frequency charts in Figure 1. For example, while Motif 9 (GC-box) is highly specific to histone group H4 and relatively less specific to other groups, Motifs 1, 7 and 8 (CCAAT-box) are less specific to histone H4 compared to the other groups. Likewise, relative strengths of Motif 3 (TATA-box) and Motif 6 (AC-box) are skewed in favour of histone H1 and Motif 4 (E2F-box) is skewed in favour of H2A/H2B. Overall, H2A and H2B seem to follow similar patterns of motif conservation. This can be expected, as H2A and H2B are divergently transcribed genes that tend to share common functional promoters (Albig et al., 1999; Trappe et al., 1999).
|
The putative binding sites represented by all these motifs have been implicated in the regulation of histone genes. While CCAAT-box, E2F-box, AC-box, Oct-1 binding site and H4-box are generally known to regulate cell cycle-dependent expression of histone genes (Doenecke et al., 1997; Oswald et al., 1996; van Wijnen et al., 1996), TATA-box is essential for the formation of transcription machinery (Nakajima et al., 1988) and is found in many genes. GC-box is known for regulating those genes whose expressions are widespread in many differentiated cell-lines, for example in housekeeping genes (Turner and Crossley, 1999). In our analysis we observe the presence of Motif 9 (GC-box) in many cell cycle-independent replacement histone genes. This suggests that Sp1 regulates expression of these genes in the differentiated cell-lines.
Overall, the motifs obtained in our analysis match fairly well with the known binding sites. It is generally evident that our analysis succeeded in detecting over-represented TFBSs. However, we were not able to detect all the known TFBSs in histone promoters, such as TE1 and TE2 elements found in histone H1t subgroup (Grimes et al., 2003), probably because these sites, when compared with the whole 127 histone promoters, may be present in very few promoters and thus have not been found to be statistically significant by the MEME program. We also realize that as a result of the trade-off in selecting a short promoter segment [–250, –1] we were not in position to detect TFBS motifs located beyond the selected promoter region. For example, we missed motifs such as upstream element (TGTGTTA) described first by Duncliffe et al. (1995) as a motif located about 450 bp upstream of the TSS in H1 histone genes. This sequence motif is complementary to 5'TAACACA3'. Thus, it is virtually the same as the H1 box (Motif 6, Table 1; see Dalton and Wells, 1988). To date, the only protein that has been shown to bind to this sequence element is HinF-D/CDP-Cut (van den Ent et al., 1994). As shown by Meergans et al. (1998), factors binding to the upstream motif and its complementary counterpart (H1 box, Motif 6) functionally interact in regulating the S-phase-dependent H1 gene expression. Our analysis did not identify any novel putative TFBS patterns. A possible reason includes the examination of only the top nine motifs all of which satisfied a theshold of E < 0.1. Thus, there is a possibility that we might have missed out on those putative TFBS-motifs that had a lower statistical significance and thus the presence of new binding sites cannot be completely ruled out.
The successful detection of previously characterized biological motifs in histone promoters suggests that these results could be used for promoter modelling purposes. The strong presence of the nine identified motifs suggests that combinatorial effect of promoter elements occurs in histone genes
3.2 Consensus motif models for histone promoters
In order to derive consensus motif models of promoters for each of the five histone groups and their subgroups, we simultaneously aligned the respective promoter motif sequences (Supplementary Table 1). The objective of alignment was to maximize motif similarity between sequences. We then proceeded according to the following steps to build consensus motif models from the motif alignment: (i) motif frequency is counted in each column without strand information; (ii) motif is considered for the consensus definition if its frequency in the column is at least 10%; if in a column no motif qualifies, the column is discarded; (iii) motif in majority (the majority motif) in a column is taken as the consensus candidate motif for that column; (iv) if the frequency of the majority motif is at least 50% it is considered as a strong motif, else it is considered as a weak motif; (v) strand present in the majority of the majority motifs is taken as the consensus strand. Consensus motif models for five histone groups and their subgroups are shown in Supplementary Table 1).
Overall, we observe that motif patterns are conserved and consistent in most histone groups and their subgroups (Supplementary Table 1). Motif arrangement is generally conserved in terms of position, order and strand orientation. Histone promoter groups generally share certain motif patterns, while they differ on the others, suggesting that there may be channels of common and disparate regulatory mechanisms at play during transcription. The same is true for subgroups within a particular histone group. Within subgroups motif patterns seem fairly homogeneous. There are motifs that are strongly conserved in histone groups and subgroups, for example Motifs 1, 3 and 6 in histone H1, Motifs 1 and 2 in histones H2A and H2B, Motif 1 in histone H3 and Motifs 3, 5 and 9 in histone H4 group. Similarly, strong motifs may be observed in histone subgroups. Typical examples include Motifs 5 and 6 in H1o, Motifs 1, 3, 6 and 9 in H1t, Motifs 1 and 3 in H2A.X, H2A.Y and H2A.Z, and Motifs 1, 2, 6, 7 and 9 in H3.3. In all five histone groups, the majority of genes belong to the cell-cycle dependent subgroup and, therefore, this subgroup of genes has a dominant influence on promoter models of each histone group, as opposed to histone genes that belong to other subgroups. Thus, it is not uncommon to find consensus motif models for the five histone groups being fairly similar to their respective major cell-cycle dependent subgroup models. For example, the motif model for the cell-cycle dependent HIST1 cluster is nearly the same as the overall consensus model for the H1 group (Supplementary Table 1). Similar trends are observed in other histone groups. We observe that motif patterns are generally conserved across species. This seems reasonable as orthologous genes frequently have similar functions and thus may be co-regulated through the similar transcriptional regulatory motifs in their promoters.
Most of the consensus motif models of histone promoter groups and their subgroups presented in Supplementary Table 1 are generally well supported by experimental studies. For example, we identified the consensus motif model for histone H1 group composed of TATA-box, CCAAT-box, GC-box, AC-box and E2F-box organized in this order from TSS upstream. Such a model for histone H1 promoters has been documented (Meergans et al., 1998; Osley, 1991; Werner, 2001; Duncliffe et al., 1995). However, due to the short promoter segment analyzed, we did not identify the Duncliffe Box (Duncliffe et al., 1995) since it is located beyond the region we analyzed. We have obtained similar supporting results for histone H1 subgroups of H1o (Meergans et al., 1998; Dong et al., 1995; Peretti and Khochbin, 1997; Doenecke and Alonso, 1996) and testis-specific histone H1t (Grimes et al., 2003; Wolfe and Grimes, 2003a, b). Our consensus motif models for histone H2A and H2B groups are in accordance with previous experimental studies for somatic histones H2A/H2B (Oswald et al., 1996; Albig et al., 1999; Trappe et al., 1999) and replacement histones H2A.X/H2A.Z (Yagi et al., 1995; Oswald et al., 1996). We observe that the consensus motif model for histone group H2A is nearly a mirror image of that of H2B with motifs located on opposite strands. This is partly expected since the vast majority of functional H2A and H2B genes share common promoter regions on opposite strands (Albig et al., 1999; Trappe et al., 1999). Consensus motif models for replacement histone genes H3.3 and testis-specific H3t are in partial agreement with their respective known models, H3.3 (Wells et al., 1987; Witt et al., 1997, 1998; Frank et al., 2003) and H3t (Witt et al., 1996). We find that our consensus motif model for histone H4 group agrees well with those that have been experimentally determined earlier (Osley, 1991; van Wijnen et al., 1992, 1996; Last et al., 1998; Stein et al., 1996).
Positional distribution of motifs in the histone promoter groups are presented in Supplementary Figure 2. While some motifs show strong positional preference (e.g. Motifs 1, 3, 6 and 9 in H1 histone, Motifs 1 and 4 in H2A histone, Motif 1 in H3 histone, and Motifs 3 and 5 in H4 histone), there are others that are not so specific. Motifs that show positional conservation include Motifs 1, 3, 6 and 9 in H1 histone, Motifs 1 and 4 in H2A histone, Motif 1 in H3 histone, and Motifs 3 and 5 in H4 histone. Most of the motifs, however, do not show positional preference. Among the histone groups, H1 histone shows the maximum positional conservation of motifs, while H2B shows the least.
In motif discovery, we have considered at most one (the best) occurrence of motifs (of specific type) per promoter sequence in order to minimize false positive cases being reported. Due to this trade-off, however, our approach may sometimes miss out multiple repeats of the same motif and hence may not faithfully represent those histone types which contain such multiple repeats of the same motif in their promoters. As an example, our model does not recognize all multiple repeats of the CCAAT and GC-boxes of histone promoters of the H3.3 subgroup. H3.3A is known for multiple repeats of GC-boxes (Frank et al., 2003), while H3.3B contains multiple repeats of CCAAT-boxes (Witt et al., 1998).
In some cases our analysis has detected multiple repeats of the same motif as different motifs. For example, Motifs 1, 7 and 8 are detected as different motifs. However, they all seem to represent the CCAAT-box. There are cases, such as the motif model of H2A/H2B, where all three Motifs 1, 7 and 8 appear indicating the presence of multiple CCAAT-boxes. Overall, our study has generated models of promoters fairly well where motifs occur at most once.
This study represents a comprehensive analysis of the mammalian histone promoters. We have derived models of these promoters and elucidated the most common and most significant patterns detected based on pure sequence analysis. The computational method we have used works fairly well for discovering the content of histone promoters. The detected motifs were shown to be tightly related to the known experimentally proven TFBSs. Our approach has demonstrated that a significant proportion of the known biologically relevant knowledge about the promoter content can be discovered by computational methods and therefore provides a complement to experimental approaches. The advantages of our approach are that it is fast, less expensive and can include many promoters simultaneously. Since it can reveal biologically significant features, it may complement experimental approaches. This study may be extended to promoters of other gene families. Additionally, the results obtained can be used for building models representative of different histone types. These models could be applied to entire genome scans of available mammalian genomes and help discover genes that may be co-regulated with the histone genes.
Received on December 16, 2004; revised on February 24, 2005; accepted on March 9, 2005
|
REFERENCES |
|---|
|
TOPAbstract1 INTRODUCTION2 METHODS3 RESULTS AND DISCUSSIONREFERENCES |
|---|
Albig, W. and Doenecke, D. (1997) The human histone gene cluster at the D6S105 locus. Hum. Genet., 101, 284–294[CrossRef][Web of Science][Medline].
Albig, W., et al. (1995) The human replacement histone H3.3B gene (H3F3B). Genomics, 30, 264–272[CrossRef][Web of Science][Medline].
Albig, W., et al. (1999) The human H2A and H2B histone gene complement. Biol. Chem., 380, 7–18[CrossRef][Web of Science][Medline].
Bailey, T.L. and Elkan, C. (1994) Fitting a mixture model by expectation maximization to discover motifs in biopolymers. Proc. Int. Conf. Intell. Syst. Mol. Biol., 2, 28–36[Medline].
Bailey, T.L. and Gribskov, M. (1998) Combining evidence using p-values: application to sequence homology searches. Bioinformatics, 4, 48–54.
Birnbaum, M.J., et al. (1995) Sp1 trans-activation of cell cycle regulated promoters is selectively repressed by Sp3. Biochemistry, 34, 16503–16508[CrossRef][Medline].
Chong, A., et al. (2003) FIE2: a program for the extraction of genomic DNA sequences around the start and translation initiation site of human genes. Nucleic Acids Res., 31, 3546–3553
Claverie, J.M. and Sauvaget, I. (1985) Assessing the biological significance of primary structure consensus patterns using sequence databanks. I. Heat-shock and glucocorticoid control elements in eukaryotic promoters. Comput. Appl. Biosci., 1, 95–104
Courey, A.J. and Tjian, R. (1988) Analysis of Sp1 in vivo reveals multiple transcriptional domains, including a novel glutamine-rich activation motif. Cell, 55, 887–898[CrossRef][Web of Science][Medline].
Dalton, S. and Wells, J.R. (1988) A gene-specific promoter element is required for optimal expression of the histone H1 gene in S-phase. EMBO J., 7, 49–56[Web of Science][Medline].
Doenecke, D. and Alonso, A. (1996) Organization and expression of the developmentally regulated H1(o) histone gene in vertebrates. Int. J. Dev. Biol., 40, 395–401[Web of Science][Medline].
Doenecke, D., et al. (1994) Organization and expression of H1 histone and H1 replacement histone genes. J. Cell. Biochem., 54, 423–431[CrossRef][Web of Science][Medline].
Doenecke, D., et al. (1997) Histones: genetic diversity and tissue-specific gene expression, a review. Histochem. Cell Biol., 107, 1–10[CrossRef][Web of Science][Medline].
Dong, Y., et al. (1995) An unstream control region required for inducible transcription of the mouse H1(zero) histone gene during terminal differentiation. Mol. Cell Biol., 15, 1889–1900
Duncliffe, K.N., et al. (1995) A H1 histone gene-specific AC-box-related element influences transcription from a major chicken H1 promoter. Gene, 163, 227–232[CrossRef][Web of Science][Medline].
Fletcher, C., et al. (1987) Purification and characterization of OTF-1, a transcription factor regulating cell cycle expression of a human histone H2b gene. Cell, 51, 773–781[CrossRef][Web of Science][Medline].
Frank, D., et al. (2003) Differential expression of human replacement and cell cycle dependent H3 histone genes. Gene, 312, 135–143[CrossRef][Web of Science][Medline].
Freeman, L., et al. (1996) Functional domains for assembly of histones H3 and H4 into the chromatin of Xenopus embryos. Biochem. Proc. Natl Acad. Sci., 93, 12780–12785.
Gallinari, P., et al. (1989) Characterization and purification of H1TF2, a novel CCAAT-binding protein that interacts with a histone H1 subtype-specific consensus element. Mol. Cell Biol., 9, 1566–1575
Grimes, S.R., et al. (2003) Transcriptional control of the testis-specific histone H1t gene. Gene, 304, 13–21[CrossRef][Web of Science][Medline].
Hagen, G., et al. (1994) Sp1-mediated transcriptional activation is repressed by Sp3. EMBO J., 13, 3843–3851[Web of Science][Medline].
Halees, A.S. and Weng, Z. (2004) PromoSer: improvements to the algorithm, visualization and accessibility. Nucleic Acids Res., 32, W191–W194
Imhof, A. and Becker, P.B. (2001) Modifications of the histone N-terminal domains. Evidence for an ‘epigenetic code’? Mol. Biotechnol., 17, 1–13[CrossRef][Web of Science][Medline].
Klingenhoff, A., et al. (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].
Koessler, H., et al. (2003) Aberrant expression pattern of replication-dependent histone H3 subtype genes in human tumor cell lines. DNA Cell Biol., 22, 233–241[CrossRef][Web of Science][Medline].
Koessler, H., et al. (2004) Human replication dependent histone H3 genes are activated by a tandemly arranged pair of two CCAAT-boxes. Biochem. J., Aug 23, Pt [Epub ahead of print].
Krivan, W. and Wasserman, W.W. (2001) A predictive model for regulatory sequences directing liver-specific transcription. Genome Res, 11, 1559–1566
La Bella, F. and Heintz, N. (1991) Histone gene transcription factor binding in extracts of normal human cells. Mol. Cell Biol., 11, 5825–5831
La Bella, F., et al. (1989) Histone H1 subtype-specific consensus elements mediate cell cycle-regulated transcription in vitro. Genes Dev., 3, 1982–1990
Last, T.J., et al. (1998) Repressor elements in the coding region of the human histone H4 gene interact with the transcription factor CDP/cut. Gene, 221, 267–277[CrossRef][Web of Science][Medline].
Luo, R.X. and Dean, D.C. (1999) Chromatin remodeling and transcriptional regulation. J. Natl. Cancer Inst., 91, 1288–1294
Mantovani, R. (1999) The molecular biology of the CCAAT-binding factor NF-Y. Gene, 239, 15–27[CrossRef][Web of Science][Medline].
Martinelli, R. and Heintz, N. (1994) H1TF2A, the large subunit of a heterodimeric, glutamine-rich CCAAT-binding transcription factor involved in histone H1 cell cycle regulation. Mol. Cell Biol., 14, 8322–8332
Marzluff, W.F., et al. (2002) The human and mouse replication-dependent histone genes. Genomics, 80, 487–498[CrossRef][Web of Science][Medline].
Matys, V., et al. (2003) TRANSFAC: transcriptional regulation, from patterns to profiles. Nucleic Acids Res., 31, 374–378
Meergans, T., et al. (1998) Conserved sequence elements in human main type-H1 histone gene promoters: their role in H1 gene expression. Eur. J. Biochem., 256, 436–446[Web of Science][Medline].
Mitra, P., et al. Mol. Cell Biol., (2003) 23, 8110–8123
Nakajima, N., et al. (1988) Factors involved in specific transcription by mammalian RNA polymerase II: purification, genetic specificity, and TATA box-promoter interactions of TFIID. Mol. Cell Biol., 8, 4028–4040
Osley, M.A. (1991) The regulation of histone synthesis in the cell cycle. Annu. Rev. Biochem., 60, 827–861[CrossRef][Web of Science][Medline].
Oswald, F., et al. (1996) The E2F transcription factor activates a replication-dependent human H2A gene in early S phase of the cell cycle. Mol. Cell Biol., 16, 1889–1895
Pauli, U., et al. (1987) Protein–DNA interactions in vivo upstream of a cell-cycle regulated human H4 histone gene. Science, 236, 1308–1311
Peretti, M. and Khochbin, S. (1997) The evolution of the differentiation-specific histone H1 gene basal promoter. J. Mol. Evol., 44, 128–134[Medline].
Sanchez, R. and Marzluff, W. (2002) The stem-loop binding protein is required for efficient translation of histone mRNA in vivo and in vitro. Mol. Cell Biol., 22, 7093–7104
Staden, R. (1988) Methods to define and locate patterns of motifs in sequences. Comput. Appl. Biosci., 4, 53–60
Stein, G.S., et al. (1996) Transcriptional control of cell cycle progression: the histone gene is a paradigm for the G1/S phase and proliferation/differentiation transitions. Cell Biol. Int., 20, 41–49[CrossRef][Web of Science][Medline].
Tanaka, M., et al. (2003) H1oo: a pre-embryonic H1 linker histone in search of a function. Mol. Cell Endocrinol., 202, 5–9[Web of Science][Medline].
Tompa, M., et al. An assessment of computational tools for the discovery of transcription factor binding sites. Nat. Biotech., In Press.
Trappe, R., et al. (1999) The expression of human H2A–H2B histone gene pairs is regulated by multiple sequence elements in their joint promoters. Biochim. Biophys. Acta, 446, 341–351.
Turner, J. and Crossley, M. (1999) Mammalian Kruppel-like transcription factors: more than just a pretty finger. Trends Biochem. Sci., 24, 236–240[CrossRef][Web of Science][Medline].
van den Ent, F.M., et al. (1994) Cell cycle controlled histone H1, H3 and H4 genes share unusual arrangements of recognition motifs for HinF-D supporting a coodinate promoter binding mechanism. J. Cell Physiol., 159, 515–530[CrossRef][Web of Science][Medline].
van Wijnen, A.J., et al. (1988a) Human H1 histone gene promoter CCAAT box binding protein HiNF-B is a mosaic factor. Biochemistry, 27, 6534–6541[CrossRef][Medline].
van Wijnen, A.J., et al. (1988b) Two target sites for protein binding in the promoter region of a cell cycle regulated human H1 histone gene. Nucleic Acids Res., 16, 571–592
van Wijnen, A.J., et al. (1989) Human H4 histone gene transcription requires the proliferation-specific. Nuclear factor HiNF-D. J. Biol. Chem., 264, 15034–15042
van Wijnen, A.J., et al. (1992) Overlapping and CpG methylation-sensitive protein–DNA interactions at the histone H4 transcriptional cell cycle domain: distinctions between two human H4 gene promoters. Mol. Cell Biol., 12, 3273–3287
van Wijnen, A.J., et al. (1996) CDP/cut is the DNA-binding subunit of histone gene transcription factor HiNF-D: a mechanism for gene regulation at the G1/S phase cell cycle transition point independent of transcription factor E2F. Proc. Natl Acad. Sci. USA, 93, 11516–11521
Wasserman, W.W. and Fickett, J.W. (1998) Identification of regulatory regions which confer muscle-specific gene expression. J. Mol. Biol., 278, 167–181[CrossRef][Web of Science][Medline].
Wells, D., et al. (1987) Unusual structure, evolutionary conservation of non-coding sequences and numerous pseudogenes characterize the human H3.3 histone multigene family. Nucleic Acids Res., 15, 2871–2889
Werner, T. (2001) Target gene identification from expression array data by promoter analysis. Biomol. Eng., 17, 87–94[CrossRef][Web of Science][Medline].
Werner, T. (2002) Finding and decrypting of promoters contributes to the elucidation of gene function. In Silico Biol., 2, 249–255[Medline].
Werner, T. (2003) Promoters can contribute to the elucidation of protein function. Trends Biotechnol., 21, 9–13[CrossRef][Web of Science][Medline].
Witt, O., et al. (1996) Testis-specific expression of a novel human H3 histone gene. Exp. Cell Res., 229, 301–306[CrossRef][Web of Science][Medline].
Witt, O., et al. (1997) Transcriptional regulation of the human replacement histone gene H3.3B. FEBS Lett., 408, 255–260[CrossRef][Web of Science][Medline].
Witt, O., et al. (1998) cAMP/phorbol ester response element is involved in transcriptional regulationof the human replacement histone gene H3.3B. Biochem. J., 329, 609–613[Medline].
Wolffe, A. Chromatin, Structure and Function, (1998) 3rd , New York Academic Press.
Wolfe, S.A. and Grimes, S.R. (2003a) Specific binding of nuclear proteins to a bifunctional promoter element upstream of the H1/AC box of the testis-specific histone H1t gene. Biol. Reprod., 68, 2267–2273
Wolfe, S.A. and Grimes, S.R. (2003b) Transcriptional repression of the testis-specific histone H1t gene mediated by an element upstream of the H1/AC box. Gene, 308, 129–138[CrossRef][Web of Science][Medline].
Yagi, H., et al. (1995) Regulation of the mouse histone H2A.X gene promoter by the transcription factor E2F and CCAAT binding protein. J. Biol. Chem., 270, 18759–18765
Yamamoto, T. and Horikoshi, M. (1996) Cloning of the cDNA encoding a novel subtype of histone H1. Gene, 173, 281–285[CrossRef][Web of Science][Medline].
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. B. James, A.-M. Conway, and B. J. Morris Regulation of the Neuronal Proteasome by Zif268 (Egr1) J. Neurosci., February 1, 2006; 26(5): 1624 - 1634. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||