SNF5 homologs in plants, yeast and invertebrates

The Drosophila snr1 gene is a potential homolog of the yeast SNF5 gene that encodes a subunit of the yeast SWI/SNF complex (Dingwall, 1995).

The multiprotein complexes involved in active disruption of chromatin structure, homologous to yeast SWI/SNF complex, have been described for human and Drosophila cells. In all SWI/SNF-class complexes characterised so far, one of the key components is the SNF5-type protein. A plant (Arabidopsis thaliana) cDNA encoding a 27 kDa protein has been isolated and named BSH. BSH has high homology to yeast SNF5p and its human (INI1) and Drosophila (SNR1) counterparts as well as to other putative SNF5-type proteins from Caenorhabditis elegans, fish and yeast. With 240 amino acids, the Arabidopsis BSH is the smallest SNF5-type protein so far identified. When expressed in Saccharomyces cerevisiae, the gene for BSH partially complements the snf5 mutation. BSH is, however, unable to activate transcription in yeast when tethered to DNA. The gene for BSH occurs in single copy in the Arabidopsis genome and is ubiquitously expressed in the plant. Analysis of the whole cell and nuclear protein extracts with antibodies against recombinant BSH indicates that the protein is localized in nuclei. Transgenic Arabidopsis plants with markedly decreased physiological levels of the BSH mRNA, resulting from the expression of antisense messenger, are viable but exhibit a distinctive phenotype characterised by bushy growth and flowers that are unable to produce seeds (Brzeski, 1999).

A comparison of Ini1 (a mammalian homolog of Drosophila SNR1) with known amino acid sequences reveals no known structural motifs indicative of its function. There are at least five proteins and ORFs that have amino acid sequence similarity to Ini1. The yeast SNF5 protein shares homology with Ini1 in its central charged region. The N-terminal domain of SNF5 is a glutamine- and proline-rich region and C-terminal domain is a proline-rich region. These two regions are absent from (1) Ini1; (2) the Drosophila protein, SNR1 (SNF5 related gene 1); (3) the Caenorhabditis elegans protein, CeSNF5, and (4) an ORF from Saccharomyces cerevisiae, SFH1 (SNF Five Homolog 1). Another ORF (C2F7.08) has recently been identified, encoding a hypothetical 71.9-kDa protein from Schizosaccharomyces pombe and bearing a close similarity to Ini1. Ini1 is more similar to SpSNF5 than it is to the two yeast homologs. The C2F7.08 protein lacks the N-terminal glutamine- and proline-rich region of SNF5. However, it has a C-terminal region that is somewhat proline-rich, containing 7% of prolines, as compared with 18% prolines present in SNF5. Alignment of the amino acid sequences of all six proteins/ORFs reveals that the C-terminal halves of Ini1, SNR1, CeSNF5, and SFH1 and the C-terminal half of the central regions of SNF5 and SpSNF5 proteins have three highly conserved regions. The first two conserved regions are imperfect repeat motifs. The degree of similarity among the six related proteins is highest in the first two regions (for example, as high as 85% identity in the first conserved region of Ini1 and SNR1). Alignment of sequences of repeats 1 and 2 from all the six proteins together reveals the presence of several invariable residues, suggesting that these residues may be functionally important. There are two invariant residues L and D that are separated by 10 amino acid residues in these repeats from all Ini1-related proteins except repeat 1 of SNF5, where these residues are separated by 12 amino acids. The third region of similarity is more conserved among Ini1, SNR1, and SpSNF5 proteins than among the remaining three proteins. Searching for the presence of coils by using the program COILS indicates that the region downstream of repeat 2, including the third conserved region in all six proteins, potentially can form coiled coil structures. No functional information is available about any of these regions (Morozov, 1998).

In the yeast Saccharomyces cerevisiae, the SWI-SNF complex has been proposed to antagonize the repressive effects of chromatin by disrupting nucleosomes. The SIN genes were identified as suppressors of defects in the SWI-SNF complex, and the SIN1 gene encodes an HMG1-like protein that has been proposed to be a component of chromatin. Specific mutations (sin mutations) in both histone H3 and H4 genes produce the same phenotypic effects as do mutations in the SIN1 gene. Sin1 and the H3 and H4 histones interact genetically; the C terminus of Sin1 is shown to physically associates with components of the SWI-SNF complex. In addition, this interaction is blocked in the full-length Sin1 protein by the N-terminal half of the protein. Based on these and additional results, it is proposed that Sin1 acts as a regulatable bridge between the SWI-SNF complex and the nucleosome (Perez-Martin, 1998).

The genetic and biochemical results described in this study all point to Sin1 as a target of the SWI-SNF complex. One plausible scenario is that nucleosome disruption by the SWI-SNF complex involves not only the removal of H2A-H2B dimers but also the specific removal of other chromatin-associated proteins, such as Sin1. The experiments described in this study suggest a provisional model for Sin1 function. It is proposed that in solution, Sin1 is folded back on itself as the result of interactions between its Nt and Ct halves. This interaction would prevent Sin1 from interacting with the SWI-SNF complex in solution. The binding of Sin1 to the nucleosome (either to DNA or to histones) would, according to this model, release the inhibition. Since Sin1 is formally a repressor of transcription, while the SWI-SNF components are formally activators, the proposal that the two function together may seem paradoxical. However, it is possible that Sin1 functions both to stabilize chromatin, perhaps by interacting with the nucleosome core, and to destabilize it by recruiting the SWI-SNF complex. According to this view, Sin1 would function to maintain the balance between chromatin assembly and disassembly (Perez-Martin, 1998).

The Swi5 zinc finger and the Pho2 homeodomain DNA-binding proteins bind cooperatively to the HO promoter. Pho2 (also known as Bas2 or Grf10) activates transcription of diverse genes, acting with multiple distinct DNA-binding proteins. A genetic screen was performed to identify amino acid residues in Swi5 that are required for synergistic transcriptional activation of a reporter construct in vivo. Nine unique amino acid substitutions within a 24-amino-acid region of Swi5, upstream of the DNA-binding domain, reduce expression of promoters that require both Swi5 and Pho2 for activation. In vitro DNA binding experiments show that the mutant Swi5 proteins bind DNA normally, but some mutant Swi5 proteins (resulting from SWI5* mutations) show reduced cooperative DNA binding with Pho2. In vivo experiments show that these SWI5* mutations sharply reduce expression of promoters that require both SWI5 and PHO2, while the expression of promoters that require SWI5 but are PHO2 independent is largely unaffected. This suggests that these SWI5* mutations do not affect the ability of Swi5 to bind DNA or activate transcription but specifically affect the region of Swi5 required for interaction with Pho2. Two-hybrid experiments show that amino acids 471 to 513 of Swi5 are necessary and sufficient for interaction with Pho2 and that the SWI5* point mutations cause a severe reduction in this two-hybrid interaction. Analysis of promoter activation by these mutants suggests that this small region of Swi5 has at least two distinct functions, conferring specificity for activation of the HO promoter and for interaction with Pho2 (Bhoite, 1998).

The HTA1-HTB1 locus is one of four cell cycle-regulated loci that encode the core histones in S. cerevisiae. The cis-acting regulatory sequences at this locus include three copies of a conserved activation (UAS) element, as well as a unique negative regulatory site that controls several key aspects of HTA1-HTB1 transcription. The negative site keeps the HTA1 and HTB1 genes repressed during the G1 and G2 periods of the cell cycle and shuts off HTA1 and HTB1 transcription when the intracellular levels of histones H2A and H2B rise. A number of trans-acting factors that act at the negative site to repress transcription have been identified through genetic screens. Two of these factors, the products of the evolutionarily conserved HIR1 and HIR2 genes, are transcriptional corepressors that do not possess intrinsic DNA binding activity. They are postulated to be recruited to the negative site by a sequence-specific binding factor, and once at this site they directly repress transcription. The current view is that the activity of these corepressors must be antagonized to allow HTA1 and HTB1 transcription to be activated to maximal levels (Dimova, 1999).

Several lines of evidence suggest that the chromatin structure of the HTA1-HTB1 locus might directly contribute to the repression of HTA1 and HTB1 transcription and that the Hir1 and Hir2 proteins are involved in this effect. As also noted for the Swi/Snf-regulated SUC2 gene, nucleosomes at the HTA1-HTB1 locus are organized into an extended array. Deletion of the HIR1 or HIR2 gene partially disrupts this array and concommitantly relieves transcriptional repression. In addition, Hir2p and its vertebrate ortholog, HIRA, bind to immobilized histones H2B, H3, and H4 in vitro, suggestive of a role for Hir proteins in chromatin assembly or structure. In support of this view, mutations in HIR1 and HIR2 have been found to act synergistically with mutations in Chromatin Assembly Factor I to decrease telomeric and HM silencing, a phenomenon strongly associated with repressive chromatin structure. It was therefore asked whether nucleosome remodeling factors were required for transcription of the HTA1-HTB1 locus by analyzing the effects of mutations in the yeast Swi/Snf complex on expression of an HTA1-lacZ reporter gene. Deletion of the SNF5 gene, which prevents complex assembly, or of the SNF2 gene, which removes the catalytic subunit, reduces lacZ mRNA levels approximately 7-fold over those found in wild-type cells. In addition, transcript levels from the HTB1 gene present at the endogenous HTA1–HTB1 locus are ~3-fold lower in each snf mutant compared to a wild-type strain. Together, the data identify the HTA1–HTB1 locus as a novel Swi/Snf-regulated locus (Dimova, 1999).

To determine whether Swi/Snf is required for HTA1-HTB1 transcription because of Hir-mediated repression at the negative site, the negative site from the promoter of an HTA1-lacZ gene was deleted, leaving the three UAS elements intact. In the absence of the negative site, HTA1-lacZ mRNA levels rose by ~2.5-fold in the wild-type strain. Identical levels of transcript were also present in the snf5delta mutant, indicating that the transcriptional defect had been bypassed. Next, it was asked whether the Hir proteins themselves were responsible for the Snf2p and Snf5p dependence of the locus. When either HIR gene was deleted in a snf5delta or snf2delta mutant, both HTA1-lacZ and HTB1 mRNAs were made at levels equivalent to those produced in hir1delta or hir2delta single mutants, where transcription was derepressed. Thus, when Hir repression is abolished by either cis- or trans-acting mutations, the Swi/Snf constituents Snf2p and Snf5p are no longer required for transcription of the HTA1-HTB1 locus. This indicates a role for the Swi/Snf complex in relieving Hir-mediated repression at the histone gene locus (Dimova, 1999).

The finding that Hir repression could be directly responsible for the Swi/Snf dependence of the HTA1-HTB1 locus suggested that components of the Swi/Snf complex might physically interact with the two Hir corepressors. To test for this association, HA-tagged Hir1 or Hir2 proteins were immunoprecipitated from whole-cell extracts using a monoclonal antibody against the HA epitope, and the coprecipitation of the Swi/Snf subunits Snf5p, Snf2p/Swi2p, and Swi3p was analyzed by Western blot analysis with polyclonal antibodies specific for each protein. All three Swi/Snf proteins coimmunoprecipitate with each Hir-HA protein. These results provide the first evidence for a physical interaction between components of the yeast Swi/Snf complex and locus-specific transcriptional repressors. Moreover, since three separate Swi/Snf subunits could be coimmunoprecipitated with Hir1p and Hir2p, this provides strong evidence that the Swi/Snf complex, or a Swi/Snf subcomplex, is associated with the two corepressors. It is also likely that this association involves only a subset of Swi/Snf complexes because a very small fraction of total cellular Snf5, Snf2, and Swi3 protein was found to interact with Hir1p and Hir2p (Dimova, 1999).

Hir1p and Hir2p play a direct role in keeping HTA1 and HTB1 transcription repressed throughout most of the cell cycle; only when Hir repression is relieved at the G1/S boundary are the two histone genes transcriptionally activated. Swi/Snf is likely to function at the HTA1-HTB1 promoter during the period of transcriptional activation, a point when Hir repression is itself inactive. This raises the question of how the Hir corepressors could be involved in the recruitment of Swi/Snf. One possibility is that Swi/Snf associates with the HTA1-HTB1 promoter as a consequence of cell cycle regulatory signals that act on Hir1p and Hir2p at the G1/S boundary. These signals might alter the ability of the Hir proteins to act as corepressors and simultaneously convert them into coactivators that recruit Swi/Snf. This predicts that Swi/Snf-Hir protein interactions or Swi/Snf association with the HTA1-HTB1 promoter might be regulated as a function of the cell cycle. The notion that transcriptional repressors could function as coactivators to target Swi/Snf to specific genes has also been suggested from other studies. RB, for example, has been postulated to recruit hSwi/Snf to genes activated by the glucocorticoid receptor and E2F1. In the latter case, however, the RB-hSwi/Snf associations result in the inhibition of E2F-mediated transactivation, and in this context RB functions as a corepressor rather than a coactivator. Genetic studies in yeast indicate that the in vivo role of Swi/Snf is to antagonize chromatin-mediated transcriptional repression. It is therefore possible that the presence of Swi/Snf at the HTA1-HTB1 promoter and its physical interactions with Hir1p and Hir2p represent a role for the remodeling complex in counteracting a repressive chromatin structure mediated by the Hir proteins. Thus, Hir1p and Hir2p could both target Swi/Snf and be antagonized by the remodeling factor. The idea that Swi/Snf might relieve repression by acting on nonhistone regulatory proteins is also indicated from other studies in yeast. Both genetic and biochemical data suggest that yeast Swi/Snf antagonizes repression mediated by the Tup1p/Ssn6p corepressors at the SUC2 locus and that it counteracts the repressive effects of the general transcriptional repressor, Sin1p. Since Tup1p/Ssn6p and Sin1p are also postulated to inhibit transcription partly through their interactions with chromatin constituents, Swi/Snf might selectively target certain classes of chromatin-associated repressors for inactivation. Whether this is part of a general mechanism by which Swi/Snf functions in vivo remains to be determined (Dimova, 1999 and references).

Activation of HO in yeast involves recruitment of transcription factors in two waves. The first is triggered by inactivation of Cdk1 (Cdc2) at the end of mitosis, which promotes import into the nucleus of the Swi5 transcription factor. Swi5 recruits the Swi/Snf chromatin-remodeling complex, which then facilitates recruitment of the SAGA histone acetylase, which in turn permits the binding of the SBF transcription factor. SBF then recruits the SRB/mediator complex and this process occurs in the absence of Cdk1 activity. The second wave is triggered by reactivation of Cdk1, which leads to recruitment of PolII, TFIIB, and TFIIH. RNA polymerase is, therefore, recruited to HO in two steps and not as a holoenzyme. A similar sequence of events occurs at other SBF-regulated promoters, such as CLN1, CLN2, and PCL1 (Cosma, 2001).

Differential gene expression forms the basis for most differences in the behavior and properties of distinct cell types within a multicellular organism. This cell type diversification usually arises due to variations between cells in the abundance of DNA sequence-specific transcription factors (STFs). Such factors promote transcription either by recruiting general transcription factors (GTFs) to TATA boxes close to transcription initiation sites or by recruiting chromatin-remodeling factors like the Swi/Snf complex and the SAGA histone acetyl transferase, which alter a promoter's chromatin structure and thereby facilitate recruitment of other STFs and GTFs. Once established, changes in a promoter's chromatin structure are sometimes self-propagating and do not require the continued presence of STFs that initiated the change. Thus, the multitudinal differences between cell types in their patterns of gene expression arise through a complex interplay between the concentration of specific transcription factors within cells at the time of transcription and the history of each of its genes' chromatin structure. The latter is itself a legacy of specific transcription factors to which the cell has been exposed in the past and stochastic events that affect the propagation of chromatin structures induced by these factors (Cosma, 2001).

Recruitment of polymerase II is a crucial event in gene transcription. In yeast, 'core' polymerase consists of twelve subunits called Rpb1-Rpb12. The C-terminal domain (CTD) of the largest of these subunits, Rpb1, contains multiple copies of the sequence YSPTSPS, which is phosphorylated by several protein kinases, including Kin28 (the yeast equivalent to Cdk7), associated with TFIIH. PolII's CTD associates with an Srb/mediator complex, which is composed of a 'head' subcomplex containing Srb proteins and an middle/tail subcomplex containing proteins such as Rgr1. Because PolII can be isolated as a holoenzyme in which core polymerase is associated with the Srb/mediator complex, it is thought that the role of many if not most STFs is to recruit such an RNA polymerase holoenzyme to promoters (Cosma, 2001).

Transcription of protein-encoding genes requires, in addition to the PolII/Srb/mediator complex, several other general transcription factors: TFIID formed by TBP and TAFs (TBP-associated factors), TFIIA, TFIIB, TFIIE, TFIIF, and TFIIH. Recruitment of holoenzyme and GTFs to the TATA box to form a preinitiation complex (PIC) is thought to precede the activation of transcription. An important issue is whether holoenzyme and general transcription factors are recruited in a defined order or whether they are all recruited simultaneously, possibly as a preassembled 'complex of complexes'. Following initiation, a scaffold composed of activator, mediator, and some GTFs remains associated with the promoter and promotes formation of a functional reinitiation complex (Cosma, 2001 and references therein).

The precise roles of general transcription factors, activators, and coactivators have been extensively studied using in vitro systems in which they can be shown to stimulate or repress transcription. Such studies assess what factors are capable of, but cannot address their actual role at real promoters in vivo. To do this, it is necessary first to know where and when chromatin remodeling has taken place and where and when site-specific and general transcription factors have bound to the promoter in question. Such information can now be gathered by measuring the abundance of individual DNA sequences that can be immunoprecipitated using antibodies directed toward specific chromatin-bound factors. This technique, known as chromatin immunoprecipitation (chIP), can be combined with the use of genetic mutations to determine the interdependency of factor recruitment and thereby build a picture of the pathway by which transcription complexes are assembled on real promoters in vivo. A major limitation associated with this approach is the potential pleiotropy of mutations. This is especially problematic when the mutations in question affect general factors needed for transcription of most if not all PolII-dependent genes. Under these circumstances, it is difficult if not impossible to attribute dependency to direct effects unless phenotypes are measured within minutes of a factor's inactivation, which is often impossible to achieve. One solution to this problem is to study promoters at which events involving general transcription factors are controlled by specific physiological signals or by transcription factors whose realm of action is far more specific than that of general transcription factors. One promoter that is particularly suitable in this regard is that of the HO gene in yeast, which encodes an endonuclease that initiates mating-type switching in a lineage- and cell cycle-dependent manner in haploid yeast gametes (Cosma, 2001).

HO transcription is activated in late G1 of the cell cycle in only one of the two progeny produced at the previous cell division, in the so-called 'mother' cell that has just given birth to a bud, or 'daughter' cell. HO is activated in a stepwise fashion, which commences with the arrival during anaphase of the Swi5 transcription factor. Swi5 then recruits the Swi/Snf nucleosome-remodeling factor, which in turn recruits the SAGA histone acetyltransferase. Both events are aborted by the arrival within daughter nuclei but not within mother nuclei of a repressor called Ash1. The remodeling of nucleosomes on the HO promoter mediated by Swi/Snf and SAGA permits the binding of a second sequence-specific transcription factor called SBF, whose activity is also essential for HO transcription (Cosma, 2001).

The entry of Swi5 into nuclei during late anaphase/telophase (and hence all early events associated with the HO promoter's activation) is thought to be triggered by inactivation of the Cdk1 protein kinase, at least partly through the proteolysis of its B-type cyclin partners by a ubiquitin protein ligase called the anaphase-promoting complex (APC). This paper investigates events associated with the onset of transcription of four SBF-regulated genes, HO, CLN1, CLN2, and PCL1, since Cdk1 is reactivated through its association with the 'G1' Cln cyclins (e.g., Cln3) during late G1. Using chIP, the arrival on the HO promoter is measured of the large subunit of PolII, four subunits of the Srb/mediator complex, TFIIB, and Kin28, which is the protein kinase associated with TFIIH implicated in CTD phosphorylation and promoter clearance. Binding of SBF to HO, which is made possible by the prior remodeling of its chromatin structure by Swi/Snf and SAGA, is crucial for the subsequent recruitment of Srb/mediator, TFIIB, and Kin28. Surprisingly, the SRB/mediator complex is recruited to HO in the absence of Cdk1 activity, whereas PolII, TFIIB, and Kin28 are only recruited in its presence. These results suggest that RNA polymerase is recruited to SBF-regulated genes in two steps and not as a holoenzyme (Cosma, 2001).

Regulation of HO gene expression in the yeast Saccharomyces cerevisiae is intricately orchestrated by an assortment of gene-specific DNA-binding and non-DNA binding regulators. Binding of the early G1 transcription factor Swi5 to the distal URS1 element of the HO promoter initiates a cascade of events through recruitment of the Swi/Snf and SAGA complexes. In late G1, binding of transcription factor SBF to promoter proximal sequences results in the timely expression of HO. An important additional layer of complexity to the current model is described by identifying a connection between Swi5 and the Mediator/RNA polymerase II holoenzyme complex. Swi5 recruits Mediator to HO by specific interaction with the Gal11 module of the Mediator complex. Importantly, binding of both the Gal11 and Srb4 mediator components to the upstream region of HO is independent of the SBF factor. Swi/Snf is required for Mediator binding, and genetic suppression experiments suggest that Swi/Snf and Mediator act in the same genetic pathway of HO activation. Experiments examining the kinetics of binding show that Mediator binds to HO promoter elements 1.5 kb upstream of the transcription start site in early G1, but this binding occurs without RNA Pol II. RNA Pol II does not bind to HO until late G1, when HO is actively transcribed, and binding occurs exclusively to the TATA region (Bhoite, 2001).

Several accessory proteins referred to as mediators are required for the full activity of the Rad51 (Rhp51 in fission yeast) recombinase. In this study, in vivo functions were examined of the recently discovered Swi5/Sfr1 complex from fission yeast. In normally growing cells, the Swi5-GFP protein localizes to the nucleus, where it forms a diffuse nuclear staining pattern with a few distinct foci. These spontaneous foci do not form in swi2Δ mutants. Upon UV irradiation, Swi5 focus formation is induced in swi2Δ mutants, a response that depends on Sfr1 function, and Sfr1 also forms foci that colocalize with damage-induced Rhp51 foci. The number of UV-induced Rhp51 foci is partially reduced in swi5Δ and rhp57Δ mutants and completely abolished in an swi5Δ rhp57Δ double mutant. An assay for products generated by HO endonuclease-induced DNA double-strand breaks (DSBs) reveals that Rhp51 and Rhp57, but not Swi5/Sfr1, are essential for crossover production. These results suggest that Swi5/Sfr1 functions as an Rhp51 mediator but processes DSBs in a manner different from that of the Rhp55/57 mediator (Akamatsu, 2007).