Positive and negative transcriptional regulatory elements in the early H4 histone gene of the sea urchin, Strongylocentrotus purpuratus

Down-regulation of Early Sea Urchin Histone H2A Gene Relies on cis Regulative Sequences Located in the 5′ and 3′ Regions and Including the Enhancer Blocker sns

The tandem repeated sea urchin alpha-histone genes are developmentally regulated by gene-specific promoter elements. Coordinate transcription of the five genes begins after meiotic maturation of the oocyte, continues through cleavage, and reaches its maximum at morula stage, after which these genes are shut off and maintained in a silenced state for the life cycle of the animal. Although cis regulative sequences affecting the timing and the level of expression of these genes have been characterized, much less is known about the mechanism of their repression. Here we report the results of a functional analysis that allowed the identification of the sequence elements needed for the silencing of the alpha-H2A gene at gastrula stage. We found that important negative regulative sequences are located in the 462 bp sns 5 fragment located in the 3' region. Remarkably, sns 5 contains the sns enhancer blocking element and the most 3' H2A codons. In addition, we made the striking observation that inhibition of the anti-enhancer activity of sns, by titration of the binding proteins in microinjected embryos, also affected the capability of sns 5 to down-regulate transgene expression at gastrula stage. A further sequence element essential for repression of the H2A gene was identified upstream of the enhancer, in the 5' region, and contains four GAGA repeats. Altogether these findings suggest that down-regulation of the alpha-H2A gene occurs by the functional interaction of the 5' and 3' cis sequence elements. These results demonstrate the involvement of a genomic insulator in the silencing of gene expression.

The modulator is a constitutive enhancer of a developmentally regulated sea urchin histone H2A gene

Going back to the late 1970s and early 1980s, we trace the Xenopus oocyte microinjection experiments that led to the emergence of the concept of "modulator". The finding that the modulator could transactivate transcription from far upstream and in either orientation suggested that a new genetic element, different from the classical prokaryotic promoter sequences, had been discovered. This particular enhancer transactivates transcription of the sea urchin early (alpha) histone H2A gene which is regulated in early sea urchin development. We summarise the data from sea urchin microinjection experiments that confirm and extend the results obtained with Xenopus oocytes. We conclude that the H2A enhancer is bipartite, is located approx. 100 bp upstream of the TATAAATA box in the H2A gene of two sea urchin species and enhances transcription when placed at a position far upstream or far downstream of the gene unless an insulator intervenes between enhancer and promoter. Evidence from microinjection experiments with sea urchin embryos suggests that the developmental control of H2A expression resides not with the enhancer, which is constitutively active, but with a striking chromatin structure with two positioned nucleosomes near the 3' end of the gene. Within this structure, there is an insulator element.

Regulation of the sea urchin early H2A histone gene expression depends on the modulator element and on sequences located near the 3' end

Transcription of the sea urchin early histone genes occurs transiently during early cleavage, reaching the maximum at the morula stage and declining to an undetectable level at the gastrula stage. To identify the regulatory elements responsible for the timing and the levels of transcription of the H2A gene, we used promoter binding studies in nuclear extracts and microinjection of a CAT transgene driven by the early H2A promoter. We found that morula and gastrula nuclear proteins produced indistinguishable DNase I footprint patterns on the H2A promoter. Two sites of interactions, centred on the modulator/enhancer and on the CCAAT box respectively, were detected. Deletion of the modulator or coinjection of an excess of modulator sequences severely affected the expression of two transgenes driven by the enhancer-less and modulator-containing H2A promoter. Finally, a DNA fragment containing 3' coding and post-H2A spacer sequences, where upon silencing three micrococcal nuclease hypersensitive sites were previously mapped, specifically repressed at the gastrula stage the expression of the transgene driven by the H2A promoter. These results indicate that the modulator is essential for the expression of early H2A gene and that sequences for downregulation are localized near the 3' end of the H2A gene.

Repressor elements in the coding region of the human histone H4 gene interact with the transcription factor CDP/cut

The coding region of the human histone H4 gene FO108 undergoes dynamic changes in chromatin structure that correlate with modifications in gene expression. Such structural alterations generally reflect transcription factor interactions with gene regulatory sequences. To test for regulatory elements within the coding region, we performed transient transfection experiments in HeLa cells using constructs with histone H4 sequences fused upstream of a heterologous thymidine kinase promoter and CAT reporter gene. H4 gene sequences from -10 to +210 repressed transcription 4.8-fold. Further deletion and mutational analysis delineated three repressor elements within this region. Using oligonucleotide competition analysis and specific antibody recognition in electrophoretic mobility shift assays, as well as methylation interference and DNase I footprinting analyses, we have identified the CCAAT displacement protein (CDP/cut) as the factor that interacts with these three repressor elements. CDP/cut binding to these repressor sites is proliferation-specific and cell-cycle-regulated, increasing in mid to late S phase. Our results indicate that the proximal 200 nucleotides of the histone H4-coding region contain transcriptional regulatory elements that may contribute to cell-cycle control of histone gene expression by interacting with repressor complexes containing CDP/cut homeodomain transcription factors.

Modulator factor-binding sequence of the sea urchin early histone H2A promoter acts as an enhancer element

The sea urchin early H2A histone gene, like the other four members of the repeating units, is transiently expressed during very early development. To investigate the mechanisms underlying the faithful expression of the early H2A gene, we focused our attention on the modulator element. We showed by DNase I cleavage protection patterns that the modulator includes the upstream sequence element 1 (USE1) and mapped at nucleotides -137 to -108 in the early H2A gene promoter. Functional tests conducted by microinjection into sea urchin embryos then showed that the modulator element binds the transcriptional factor called modulator-binding factor 1 (MBF-1). We found in fact that coinjection of an excess of the MBF-1-binding site, either as the modulator or as the USE1, efficiently impaired the activity of the H2A promoter. An unexpected finding was the expression of the reporter gene from the early H2A promoter at the gastrula stage of embryonic development, when the early histone genes are transcriptionally silent. In addition, we also found that the modulator element was active at the gastrula stage. The potential enhancer activity of the modulator was tested by microinjecting several constructs containing single or multiple copies of the modulator element placed 5' or 3' to a thymidine kinase gene (tk) promoter in both sea urchin embryos and Xenopus laevis oocytes and determining the expression of a reporter chloramphenicol acetyltransferase gene under the control of the linked tk promoter. We found that an oligonucleotide bearing the MBF-1-binding site activates the expression of the reporter gene independently of the position and orientation. We conclude that the modulator binds the MBF-1 activator and that it is a transcriptional enhancer of the early H2A histone gene.

Promoter analysis meets pattern formation: transcriptional regulatory genes in sea urchin embryogenesis

Analyses of spatial and temporal gene control mechanisms in the sea urchin embryo have identified several important trans-regulatory factors, including some that are related to known developmental control genes of the fly and mouse. Recent advances in gene perturbation technologies, including the use of antisense oligonucleotides to target mRNAs in early-stage embryos, as well as the injection of mRNAs into zygotes to express genes ectopically, have made it possible to test the functions of such factors directly.

Constitutive expression of a microinjected glucose-regulated protein (grp78) fusion gene during early Xenopus laevis development

In this study we have found that a rat glucose-regulated protein (grp) 78 chloramphenicol acetyltransferase (CAT) fusion gene deleted to -456 bp at the 5' end and injected into fertilized Xenopus eggs was first expressed in a constitutive manner in late blastula stage embryos and displayed increased expression as the embryos developed to the gastrula and neurula stages. Using a series of internal deletion mutants and linker-scanner mutants of the rat grp78 promoter, we have found that a CCAAT box and CCAAT-like element within the region -129 to -90 were essential for constitutive expression of the chimeric genes in neurula stage embryos. These results suggest conservation of the regulatory sequences within the grp78 promoter between rat and Xenopus. Interestingly, deletion or alteration of sequences between -130 and -149 had a dramatic stimulatory effect on basal promoter activity. This effect, which was not observed previously in rat cells, may be the result of upstream elements that are transcriptionally active in Xenopus and that can compensate for the mutated or deleted sequences. It is also possible that these results indicate the presence of a negative regulatory element that is recognized by the Xenopus transcriptional apparatus.

Sea urchin early histone H2A modulator binding factor 1 is a positive transcription factor also for the early histone H3 gene

To shed some light on the mechanisms involved in the coordinate regulation of the early histone gene set during sea urchin development, we tested the hypothesis that the upstream sequence element USE1, previously identified in the early H2A modulator, could also participate in the transcription of the early histone H3 gene. We found by DNAse I protection analysis and by competition in electrophoretic mobility-shift experiments that two sequence elements of the H3 promoter closely resembled the USE1-H2A sequence in their binding activity for nuclear factors from 64-cell stage embryos. These modulator binding factor 1 (MBF-1)-related factors seem to recognize the ACAGA motif that is conserved between the USE1-like sequences of both H2A and H3 promoters. In fact, excess oligonucleotide containing a mutated USE1-H2A element in which the ACAGA sequence was mutated to AGTCA failed to compete with the USE1 sites of both H2A and H3 genes for interaction with MBF-1. Finally, in vivo transcriptional analysis in both Xenopus and sea urchin showed that an excess of USE1-H2A element efficiently competed for the activity of the H3 promoter. From these results we conclude that MBF-1 is a transcription factor conserved between sea urchin and frog and that MBF-1 or related transcription factors are involved in the coordinate expression of both H2A and H3 early histone genes.

Basal level transcription of the histone H1(0) gene is mediated by a 80 bp promoter fragment

The replacement histone H1(0) of the H1 group, known to interact with general transcription factors, has been found associated with transcriptionally repressed chromatin. Transcription of the gene in F9 stem cells is low but can be stimulated by treating the cells with retinoic acid. Using mutant deletions, we now demonstrate that basal level transcription in F9 cells is mediated by an 80 bp DNA fragment, located 430 bp upstream of the TATA box, which does not include the retinoic acid responsive element (RARE) known to bind retinoic acid receptors and stimulate transcription from an heterologous promoter after retinoic acid treatment. By footprinting, DMS interference, site-directed mutagenesis and UV-cross linking techniques we demonstrate that at least two nuclear factors, with MW of 90,000 and 30,000, bind to the 80 bp fragment and that this binding is necessary for transcription. Furthermore, positioning of this fragment upstream of the HSV-tk gene promoter stimulates transcription 2-3 times over control values, far less than the activity observed for this fragment in the homologous promoter, indicating that full activity of this fragment requires sequences located in the proximal part of the promoter.

The arthropod initiator: the capsite consensus plays an important role in transcription

Approximately 25% of arthropod RNA polymerase II-transcribed promoters contain one or more copies of the sequence TCAGT beginning within the interval (-10, +10). The clear statistical overrepresentation of this sequence and, to a lesser extent, of its cognates ACAGT, GCAGT, and TCATT, implies that they may be significant promoter elements. Their collective sequence similarity to vertebrate initiators (Inrs) of the TdT class suggests that the vertebrate and arthropod elements are homologous. Prior work in vertebrate systems has emphasized the role of the Inr in promoters lacking TATA boxes, where it can serve as an alternate staging site for polymerase II initiation. However, it is clear that the Inr sequence is by no means restricted to TATA-deficient promoters. Functional tests using the TATA-containing Drosophila gene Eip28/29 support the idea that the Inr is a facultative promoter element, required for efficient transcription under some conditions. For example, the Inr protects basal expression of Eip28/29 from the silencing effect of ecdysone response elements. In addition, the Inr is required for the function of an enhancer of basal activity in Eip28/29. We conclude that Inrs are promoter elements found sporadically throughout the higher eukaryotes, that the requirement for an Inr depends upon the array of other promoter elements which may be present in a given gene, and that Inrs may permit enhancers to discriminate among promoters.

Formation of the 3' end of sea urchin U1 small nuclear RNA occurs independently of the conserved 3' box and on transcripts initiated from a histone promoter

The formation of the 3' end of vertebrate small nuclear RNAs (snRNAs) requires that transcription initiate from an snRNA promoter. There is a loosely conserved required box 5 to 20 nucleotides (nt) 3' of the gene. The sea urchin snRNA genes contain promoter elements different from those of the vertebrate snRNAs. They also contain a characteristic 3' 15-nt sequence which is conserved among different sea urchin snRNA genes. We used microinjection of sea urchin U1 snRNA genes into sea urchin zygotes to define the sequence requirements for U1 snRNA 3'-end formation. Surprisingly, the conserved 3' box is not required for efficient 3'-end formation in vivo. Deletion analysis reveals that the 6 nt immediately 3' of the U1 snRNA are involved in 3'-end formation. Substitution analysis revealed that either these 6 nt 3' of the U1 RNA or the conserved 3' box could direct 3'-end formation. Transcripts initiated from a histone H4 promoter formed U1 3' ends about 50% as efficiently as transcripts initiated from the U1 promoter, even when the U1 end was placed in tandem with a histone 3'-processing signal, suggesting that transcription from an snRNA promoter is not necessary for formation of the 3' end of sea urchin U1 snRNA.

Formation of the 3' end of sea urchin U1 small nuclear RNA occurs independently of the conserved 3' box and on transcripts initiated from a histone promoter

The formation of the 3' end of vertebrate small nuclear RNAs (snRNAs) requires that transcription initiate from an snRNA promoter. There is a loosely conserved required box 5 to 20 nucleotides (nt) 3' of the gene. The sea urchin snRNA genes contain promoter elements different from those of the vertebrate snRNAs. They also contain a characteristic 3' 15-nt sequence which is conserved among different sea urchin snRNA genes. We used microinjection of sea urchin U1 snRNA genes into sea urchin zygotes to define the sequence requirements for U1 snRNA 3'-end formation. Surprisingly, the conserved 3' box is not required for efficient 3'-end formation in vivo. Deletion analysis reveals that the 6 nt immediately 3' of the U1 snRNA are involved in 3'-end formation. Substitution analysis revealed that either these 6 nt 3' of the U1 RNA or the conserved 3' box could direct 3'-end formation. Transcripts initiated from a histone H4 promoter formed U1 3' ends about 50% as efficiently as transcripts initiated from the U1 promoter, even when the U1 end was placed in tandem with a histone 3'-processing signal, suggesting that transcription from an snRNA promoter is not necessary for formation of the 3' end of sea urchin U1 snRNA.

Overlapping and CpG methylation-sensitive protein-DNA interactions at the histone H4 transcriptional cell cycle domain: distinctions between two human H4 gene promoters

Transcriptional regulation of vertebrate histone genes during the cell cycle is mediated by several factors interacting with a series of cis-acting elements located in the 5' regions of these genes. The arrangement of these promoter elements is different for each gene. However, most histone H4 gene promoters contain a highly conserved sequence immediately upstream of the TATA box (H4 subtype consensus sequence), and this region in the human H4 gene FO108 is involved in cell cycle control. The sequence-specific interaction of nuclear factor HiNF-D with this key proximal promoter element of the H4-FO108 gene is cell cycle regulated in normal diploid cells (J. Holthuis, T.A. Owen, A.J. van Wijnen, K.L. Wright, A. Ramsey-Ewing, M.B. Kennedy, R. Carter, S.C. Cosenza, K.J. Soprano, J.B. Lian, J.L. Stein, and G.S. Stein, Science, 247:1454-1457, 1990). Here, we show that this region of the H4-FO108 gene represents a composite protein-DNA interaction domain for several distinct sequence-specific DNA-binding activities, including HiNF-D, HiNF-M, and HiNF-P. Factor HiNF-P is similar to H4TF-2, a DNA-binding activity that is not cell cycle regulated and that interacts with the analogous region of the H4 gene H4.A (F. LaBella and N. Heintz, Mol. Cell. Biol. 11:5825-5831, 1991). The H4.A gene fails to interact with factors HiNF-M and HiNF-D owing to two independent sets of specific nucleotide variants, indicating differences in protein-DNA interactions between these H4 genes. Cytosine methylation of a highly conserved CpG dinucleotide interferes with binding of HiNF-P/H4TF-2 to both the H4-FO108 and H4.A promoters, but no effect is observed for either HiNF-M or HiNF-D binding to the H4-FO108 gene. Thus, strong evolutionary conservation of the H4 consensus sequence may be related to combinatorial interactions involving overlapping and interdigitated recognition nucleotides for several proteins, whose activities are regulated independently. Our results also suggest molecular complexity in the transcriptional regulation of distinct human H4 genes.

Overlapping and CpG methylation-sensitive protein-DNA interactions at the histone H4 transcriptional cell cycle domain: distinctions between two human H4 gene promoters

Transcriptional regulation of vertebrate histone genes during the cell cycle is mediated by several factors interacting with a series of cis-acting elements located in the 5' regions of these genes. The arrangement of these promoter elements is different for each gene. However, most histone H4 gene promoters contain a highly conserved sequence immediately upstream of the TATA box (H4 subtype consensus sequence), and this region in the human H4 gene FO108 is involved in cell cycle control. The sequence-specific interaction of nuclear factor HiNF-D with this key proximal promoter element of the H4-FO108 gene is cell cycle regulated in normal diploid cells (J. Holthuis, T.A. Owen, A.J. van Wijnen, K.L. Wright, A. Ramsey-Ewing, M.B. Kennedy, R. Carter, S.C. Cosenza, K.J. Soprano, J.B. Lian, J.L. Stein, and G.S. Stein, Science, 247:1454-1457, 1990). Here, we show that this region of the H4-FO108 gene represents a composite protein-DNA interaction domain for several distinct sequence-specific DNA-binding activities, including HiNF-D, HiNF-M, and HiNF-P. Factor HiNF-P is similar to H4TF-2, a DNA-binding activity that is not cell cycle regulated and that interacts with the analogous region of the H4 gene H4.A (F. LaBella and N. Heintz, Mol. Cell. Biol. 11:5825-5831, 1991). The H4.A gene fails to interact with factors HiNF-M and HiNF-D owing to two independent sets of specific nucleotide variants, indicating differences in protein-DNA interactions between these H4 genes. Cytosine methylation of a highly conserved CpG dinucleotide interferes with binding of HiNF-P/H4TF-2 to both the H4-FO108 and H4.A promoters, but no effect is observed for either HiNF-M or HiNF-D binding to the H4-FO108 gene. Thus, strong evolutionary conservation of the H4 consensus sequence may be related to combinatorial interactions involving overlapping and interdigitated recognition nucleotides for several proteins, whose activities are regulated independently. Our results also suggest molecular complexity in the transcriptional regulation of distinct human H4 genes.

Protein-DNA interactions at the H4-site III upstream transcriptional element of a cell cycle regulated histone H4 gene: differences in normal versus tumor cells

Upstream sequences of the H4 histone gene FO108 located between nt -418 to -213 are stimulatory for in vivo transcription. This domain contains one protein/DNA interaction site (H4-Site III) that binds factor H4UA-1. Based on methylation interference, copper-phenanthroline protection, and competition assays, we show that H4UA-1 interacts with sequences between nt -345 to -332 containing an element displaying sequence-similarity with the thyroid hormone response element (TRE). Using gel retardation assays, we also demonstrate that H4UA-1 binding activity is abolished at low concentrations of Zn2+ (0.75 mM), a characteristic shared with the thyroid hormone (TH) receptor DNA binding protein. Interestingly, phosphatase-treatment of nuclear proteins inhibits formation of the H4UA-1 protein/DNA complex, although a complex with higher mobility (H4UA-1b) can be detected; both complexes share identical protein-DNA contacts and competition behaviors. These findings suggest that phosphorylation may be involved in the regulation of H4-Site III protein/DNA interactions by directly altering protein/protein associations. H4-Site III interactions were examined in several cell culture systems during cell growth and differentiation. We find that H4UA-1 binding activity is present during the cell cycle of both normal diploid and transformed cells. However, during differentiation of normal diploid rat calvarial osteoblasts, we observe a selective loss of the H4UA-1/H4-Site III interaction, concomitant with an increase of the H4UA-1b/H4-Site III complex, indicating modifications in the heteromeric nature of protein/DNA interactions during downregulation of transcription at the cessation of proliferation. Transformed cells have elevated levels of H4UA-1, whereas H4UA-1b is predominantly present in normal diploid cells; this alteration in the ratio of H4UA-1 and H4UA-1b binding activities may reflect deregulation of H4-Site III interactions in transformed cells. We propose that H4-Site III interactions may contribute, together with protein/DNA interactions at proximal regulatory sequences, in determining the level of H4-FO108 histone gene transcription.

A conserved region in the sea urchin U1 snRNA promoter interacts with a developmentally regulated factor

The expression of the sea urchin L. variegatus U1 snRNA gene is temporally regulated during embryogenesis. Using a microinjection assay we show that a region between 203 and 345 nts 5' of the gene is required for expression. There are four conserved regions between two sea urchin species in the 345 nts 5' to the U1 gene. One region, located at about -300, binds a protein factor which is present in blastula but not gastrula nuclei. Three other potential protein binding sites within the first 200 nts 5' to the gene have been identified using a mobility shift assay and/or DNase I footprinting. Two of these regions bind factors which are not developmentally regulated and one binds a factor which is developmentally regulated. It is likely that the factor which binds at -300 is involved in expression and developmental regulation of the sea urchin U1 snRNA gene.

Protein/DNA interactions involving ATF/AP1-, CCAAT-, and HiNF-D-related factors in the human H3-ST519 histone promoter: cross-competition with transcription regulatory sites in cell cycle controlled H4 and H1 histone genes

Protein/DNA interactions of the H3-ST519 histone gene promoter were analyzed in vitro. Using several assays for sequence specificity, we established binding sites for ATF/AP1-, CCAAT-, and HiNF-D related DNA binding proteins. These binding sites correlate with two genomic protein/DNA interaction domains previously established for this gene. We show that each of these protein/DNA interactions has a counterpart in other histone genes: H3-ST519 and H4-F0108 histone genes interact with ATF- and HiNF-D related binding activities, whereas H3-ST519 and H1-FNC16 histone genes interact with the same CCAAT-box binding activity. These factors may function in regulatory coupling of the expression of different histone gene classes. We discuss these results within the context of established and putative protein/DNA interaction sites in mammalian histone genes. This model suggests that heterogeneous permutations of protein/DNA interaction elements, which involve both general and cell cycle regulated DNA binding proteins, may govern the cellular competency to express and coordinately control multiple distinct histone genes.