Imitation SWI: Biological Overview | Evolutionary Homologs | Regulation | Developmental Biology | Effects of Mutation | References

Gene name - Imitation SWI

Synonyms -

Cytological map position - 49BC

Function - transcription factor

Keywords - trithorax group

Symbol - Iswi

FlyBase ID:FBgn0011604

Genetic map position - 2-66

Classification - SWI2 homolog-ATPase

Cellular location - nuclear

NCBI links: Precomputed BLAST | Entrez Gene

Recent literature
Jain, D., Baldi, S., Zabel, A., Straub, T. and Becker, P. B. (2015). Active promoters give rise to false positive 'Phantom Peaks' in ChIP-seq experiments. Nucleic Acids Res [Epub ahead of print]. PubMed ID: 26117547
Chromatin immunoprecipitation (ChIP) is widely used to identify chromosomal binding sites. Chromatin proteins are cross-linked to their target sequences in living cells. The purified chromatin is sheared and the relevant protein is enriched by immunoprecipitation with specific antibodies. The co-purifying genomic DNA is then determined by massive parallel sequencing (ChIP-seq). This study applied ChIP-seq to map the chromosomal binding sites for two ISWI-containing nucleosome remodeling factors, ACF and RSF, in Drosophila embryos. Employing several polyclonal and monoclonal antibodies directed against their signature subunits, ACF1 and RSF-1, robust profiles were obtained indicating that both remodelers co-occupied a large set of active promoters. Further validation included controls using chromatin of mutant embryos that do not express ACF1 or RSF-1. Surprisingly, the ChIP-seq profiles were unchanged, suggesting that they were not due to specific immunoprecipitation. Conservative analysis lists about 3000 chromosomal loci, mostly active promoters that are prone to non-specific enrichment in ChIP and appear as 'Phantom Peaks'. These peaks are not obtained with pre-immune serum and are not prominent in input chromatin. Mining the modENCODE ChIP-seq profiles identifies potential Phantom Peaks in many profiles of epigenetic regulators. These profiles and other ChIP-seq data featuring prominent Phantom Peaks must be validated with chromatin from cells in which the protein of interest has been depleted.

Doyen, C. M., Chalkley, G. E., Voets, O., Bezstarosti, K., Demmers, J. A., Moshkin, Y. M. and Verrijzer, C. P. (2015). A testis-specific chaperone and the chromatin remodeler ISWI mediate repackaging of the paternal genome. Cell Rep 13: 1310-1318. PubMed ID: 26549447
During spermatogenesis, the paternal genome is repackaged into a non-nucleosomal, highly compacted chromatin structure. Bioinformatic analysis revealed that Drosophila sperm chromatin proteins are characterized by a motif related to the high-mobility group (HMG) box, which is termed male-specific transcript (MST)-HMG box. MST77F is a MST-HMG-box protein that forms an essential component of sperm chromatin. The deposition of MST77F onto the paternal genome requires the chaperone function of tNAP (CG5017, termed milkah), a testis-specific NAP protein. MST77F, in turn, enables the stable incorporation of the protamines MST35Ba and MST35Bb into sperm chromatin. Following MST-HMG-box protein deposition, the ATP-dependent chromatin remodeler ISWI mediates the appropriate organization of sperm chromatin. Conversely, at fertilization, maternal ISWI targets the paternal genome and drives its repackaging into de-condensed nucleosomal chromatin. Failure of this transition in ISWI mutant embryos is followed by mitotic defects, aneuploidy, and haploid embryonic divisions. Thus, ISWI enables bi-directional transitions between two fundamentally different forms of chromatin.

Regulatory biology is currently mesmerized (one might say the field is going gaga). The fascination is with molecular machines: protein complexes that change the conformation of chromatin, making its constituent DNA accessable to transcription factors. Iswi is a component of one of these machines. Iswi was isolated on the basis of homology to yeast transcriptional activator SWI2, also known as SNF2. Iswi differs from the yeast protein in its lack of a bromodomain (Elfring, 1994). The Iswi shares with the yeast protein a presumably non functional DNA dependent helicase domain. (Helicases unwrap the helical structure of DNA).

Iswi appears to be part of a multiprotein complex consisting of four subunits. This complex, termed NURF (nucleosome remodeling factor), is able to activate the Drosophila Heat-shock-protein-70 (Hsp70) promoter along with the GAGA transcription factor (Tsukiyama, 1995a). Iswi, the 140 kDa subunit of NURF, is localized to the cell nucleus and expressed throughout Drosophila development (Tsukiyama, 1995b).

The different structures of Brahma protein and Iswi point to the possible existence of more than one chromatin remodeling system. The Iswi component of NURF is homologous to yeast Iswi, a member of the SWI2/SNF2 family of ATPases, a transcriptional activator that counters the repressive effects on nucleosomal histones. Iswi is plentiful in Drosophila, on the order of 100,000 molecules per cell.

The ATPase activity of NURF is stimulated by chromatin (Tsukiyama, 1995b). However, the ATPase activity of a second family of activators in yeast and Drosophila, called SWI/SNF, is stimulated by naked DNA (DNA free of chromatin). The concentration of SWI/SNF in yeast is on the order of 100 molecules per cell, much lower than that of Iswi in Drosophila (Tsukiyama, 1995a). Brahma is a component of the SWI/SNF complex, and contains a bromodomain unlike Iswi. This suggests the existence of two families of ATP dependent activators; one (SWI/SNF) dependent on DNA and the another (NURF) dependent on chromatin (DNA wrapped with histones). These molecular transcription activating machines work in conjunction with transcription factors to activate or silent genes, or maintain activity of already active genes.

A chromatin-accessibility complex (CHRAC), distinct from NURF but containing Iswi, has been identified that uses energy to increase the general accessibility of DNA in chromatin. Unlike other known chromatin remodelling factors, CHRAC can also function during chromatin assembly: it uses ATP to convert irregular chromatin into a regular array of nucleosomes with even spacing. CHRAC combines enzymes that modulate nucleosome structure and DNA topology. Using mass spectrometry, two of the five CHRAC subunits have been identified as the ATPase Iswi, which is also part of NURF, and topoisomerase II. Topoisomerase II normally functions in the ATP dependent relaxation of negative supercoils, but the topoisomerase activity does not seem to be involved in CHRAC function. Topo II exists as a dimer in CHRAC. Therefore the function of Topo II in CHRAC is not yet known. No other subunit is shared by the two complexes. Two hSNF2L (the human Iswi homolog) containing complexes of 700K and 450K have been identified, corresponding to the Mr determined here for CHRAC as well as that of NURF, suggesting the existence of human counterparts for both Iswi-containing complexes. CHRAC and NURF increase the accessibility of chromatin by different mechanisms. CHRAC and NURF preparations have different activities when compared in accessibility assays: CHRAC is less effective in the GAGA-factor-dependent nucleosome remodelling assay than NURF; conversely, NURF does not stimulate the sensitivity to restriction endonucleases. NURF, but not CHRAC, perturbs histone-DNA interactions in monosomes when added at a stoichiometry of one NURF per 20 nucleosome core particles. The most striking difference between CHRAC and NURF is in their effect on nucleosomal arrays. Addition of CHRAC to chromatin assembled in the absence of ATP (and therefore lacking regular nucleosome spacing) results in an ATP-dependent alignment of nucleosomes. NURF has the opposite effect: addition of large amounts of NURF to a nucleosomal array perturbs the regular repeat pattern. The presence of Iswi in different contexts suggests that chromatin remodelling machines have a modular nature and that Iswi has a central role in different chromatin remodelling reactions (Varga-Weisz, 1997).

NURF catalyzes the bidirectional redistribution of mononucleosomes reconstituted on hsp70 promoter DNA. Within minutes, in the presence of NURF, nucleosomes adopt one predominant position from an ensemble of possible locations. Movements occur in cis, with no transfer to competing DNA. Migrating intermediates trapped by Exo III digestion reveal progressive nucleosome motion in increments of several base pairs. All four core histones are retained quantitatively during this process, indicating that the general integrity of the histone octamer is maintained. It is suggested that NURF remodels nucleosomes by transiently decreasing the activation energy for short-range sliding of the histone octamer (Hamiche, 1999).

Mononucleosome reconstitution by salt gradient dialysis on a 359 bp fragment of the Drosophila hsp 70 promoter generates a mixed nucleosome population distributed over several translational positions, whose locations are dependent on the underlying DNA sequence. These nucleosomes can be fractionated by native polyacrylamide gel electrophoresis on the basis of differences in length and spatial orientation of the linker DNA. Such position-dependent electrophoretic mobility has been described for DNA-bending proteins such as CAP. The native gel mobility is minimal for DNA molecules with centrally positioned nucleosomes; mobility increases linearly, as the nucleosome location approaches the end of the fragment. The nucleosome positions were mapped precisely by gel elution of each of the four species, termed N1-N4, followed by Exonuclease III (Exo III) digestion. The pause positions of Exo III define nucleosome boundaries, which can be displayed relative to the hsp70 promoter elements. Thus, nucleosomes N1 and N2 overlap and are located over GAGA elements of the hsp70 promoter. The N3 position is located further upstream from, and partially overlaps, N1 and N2. As is generally observed when nucleosomes are reconstituted on linear fragments, there appears to be a greater preference for nucleosome location at either fragment end (N4 and N4'). Native gel electrophoresis reveals that when NURF is added to the mixed population of N1-N4 nucleosomes in the presence of ATP, there is a clear increase in the N3 species and a corresponding loss of the N1 and N2 nucleosome positions. A subtle change in the electrophoretic mobility of the N4 species can also be observed (Hamiche, 1999).

To illustrate the effects of NURF on a single positioned species, the N1 nucleosome was purifed by native gel electrophoresis and it was eluted and incubated with NURF and ATP. Nucleosomes at the N1 position moved primarily to the N3 position, but a small, clearly defined signal is also observed at N2, and between the N3 and N4 positions.The results demonstrate that nucleosome redistribution mediated by NURF is bidirectional, repositioning nucleosomes from N1 to either upstream (N3) or downstream (N2) positions on the same DNA fragment. This can be further illustrated by incubation of gel-purified N3 nucleosomes with NURF and ATP. A fraction of the N3 species is relocated to the positions corresponding to N1, N2, N4, and between N3 and N4 nucleosomes. However, most of the nucleosomes remain at N3, confirming this location as the favored one in the ensemble of equilibrium positions. When the reaction time is shortened in a similar experiment, the presence of migrating intermediates trapped by Exo III digestion at increments of a few base pairs suggests the presence of migration intermediates. The existence of such migration intermediates suggests that migration of the histone octamer is not entirely smooth, occurring in small processive steps, rather than large jumps on DNA (Hamiche, 1999).

Remodeling mediated by NURF does not proceed by the complete dissociation of the histone octamer from DNA. Since the remodeling reactions were performed in a vast excess of carrier DNA, nucleosome transfer to an unlinked DNA fragment would readily have been detected by the production of free labeled DNA. No increase in the free 359 bp DNA is observed after remodeling by NURF. Furthermore, it is argued that NURF acts catalytically; remodeling reactions were performed at a ratio of 1 NURF:50-100 nucleosomes (Hamiche, 1999).

In principle, a histone octamer could be translocated in several ways. The octamer could be made to jump by complete detachment from one position and reassociate at a new location. The octamer could undergo displacement transfer in which a stretch of free DNA could displace another stretch of nucleosomal DNA from its interaction with the histone octamer. The nucleosome could also be relocated by a rolling or sliding of the histone octamer relative to DNA, without dissociation. Sliding is favored as a means of nucleosome mobilization by NURF because it would require only the transient alteration of individual weak interactions between histones and DNA, whereas jumping would entail the energetically unfavorable disruption of all histone octamer-DNA contacts simultaneously. The crystal structure of the histone octamer complexed with DNA indicates sites of interaction between the core histones and DNA approximately every ten base pairs over the entire 146 bp length of nucleosomal DNA, with the notable insertion of an arginine residue in the minor groove at every site. To maintain critical minor groove contacts during translational sliding, a rotation of the double helix as in the advance of a screw in its screwthread would be required (Hamiche, 1999 and references).

The experimental data obtained in this study suggest that NURF promotes nucleosome sliding. (1) Migration of the core histone octamer on the same DNA fragment is observed without displacement in trans, even in the presence of a 3000-fold excess of exogenous carrier DNA. This concentration of competing DNA is 3-fold greater than that which is effective for capturing histone octamers displaced from the 5S RNA gene by the SP6 RNA polymerase. (2) At early times in the reaction with NURF, migration intermediates trapped by Exo III digestion are observed at increments of a few base pairs. The presence of these migration intermediates is not consistent with a jumping or displacement transfer process, except over a distance of a few base pairs. Sliding has long been known to be an intrinsic property of nucleosomes. However, this passive and spontaneous sliding of the histone octamer, which occurs over a period of hours, is readily distinguished from the rapid sliding that is facilitated by NURF. Moreover, while spontaneous sliding is enhanced by elevated temperatures and ionic strengths and is suppressed by 2 mM MgCl2, sliding faciliated by NURF can occur at room temperature and in moderate concentrations of both monovalent and divalent cations (Hamiche, 1999).

ISWI regulates higher-order chromatin structure and histone H1 assembly in vivo

Imitation SWI (ISWI) and other ATP-dependent chromatin-remodeling factors play key roles in transcription and other processes by altering the structure and positioning of nucleosomes. Recent studies have also implicated ISWI in the regulation of higher-order chromatin structure, but its role in this process remains poorly understood. To clarify the role of ISWI in vivo, defects in chromosome structure and gene expression were examined resulting from the loss of Iswi function in Drosophila. Consistent with a broad role in transcriptional regulation, the expression of a large number of genes is altered in Iswi mutant larvae. The expression of a dominant-negative form of ISWI leads to dramatic alterations in higher-order chromatin structure, including the apparent decondensation of both mitotic and polytene chromosomes. The loss of ISWI function does not cause obvious defects in nucleosome assembly, but results in a significant reduction in the level of histone H1 associated with chromatin in vivo. These findings suggest that ISWI plays a global role in chromatin compaction in vivo by promoting the association of the linker histone H1 with chromatin (Corona, 2007; full text of article).

Most studies of ISWI complexes in Drosophila and other organisms have focused on their ability to alter the structure or spacing of nucleosomes, the fundamental unit of chromatin structure. These findings reveal that ISWI also plays a global role in the regulation of higher-order chromatin structure. The Iswi mutations used in this study eliminate the function of multiple chromatin-remodeling complexes, including ACF, NURF, and CHRAC. Which of these complexes are required for the formation of higher-order chromatin structure? Loss of function mutations in Acf1 -- which encodes a subunit protein shared by ACF and CHRAC -- do not cause obvious defects in higher-order chromatin structure. By contrast, loss of function mutations in E(bx) -- which encodes a subunit specific to NURF -- cause male X chromosome defects similar to those observed in Iswi mutants. These findings suggest that ISWI modulates higher-order chromatin structure within the context of NURF, as opposed to ACF or CHRAC (Corona, 2007).

A striking correlation was observed between the severity of the chromosome defects resulting from the loss of ISWI function and the loss of the linker histone H1. This correlation suggests that ISWI regulates higher-order chromatin structure by promoting the association of histone H1 with chromatin. Histone H1 and other linker histones influence higher-order chromatin structure in vitro by stabilizing interactions between nucleosomes and chromatin fibers. Although the ability of histone H1 to promote chromatin compaction in vitro is well established, its function in vivo has been a topic of considerable debate. A protein with biochemical properties reminiscent of linker histones -- HHO1 -- is present in budding yeast; surprisingly, HHO1 is not essential for viability in yeast, and hho1 mutations have little effect on either gene expression or chromatin structure. Genetic studies in Tetrahymena have suggested roles for linker histones in chromatin condensation and gene expression, but the relevance of these studies to histone H1 function in higher eukaryotes remains unclear. Studies of histone H1 function in higher eukaryotes have been complicated by the presence of redundant genes encoding histone H1 or histone H1 subtypes. In spite of these difficulties, recent studies have revealed important roles for histone H1 in chromosome compaction in Xenopus and mice. Thus, the chromosome defects observed in Iswi mutants could easily result from inefficient incorporation of histone H1 into chromatin (Corona, 2007).

How might ISWI promote the association of histone H1 with chromatin? Since ISWI is not required for histone H1 synthesis, ISWI may directly promote the assembly of chromatin containing histone H1 following DNA replication. Recent biochemical studies provide support for this possibility: ACF promotes the ATP-dependent assembly of H1-containing chromatin in vitro. Loss of ACF1 function does not cause obvious changes in chromosome structure, however, suggesting that ACF either does not regulate higher-order chromatin structure in vivo or plays a redundant role in this process. It remains possible that ISWI promotes the assembly of histone-H1-containing chromatin within the context of NURF or another chromatin-remodeling complex (Corona, 2007).

The ability to promote histone H1 assembly is not a common property of all chromatin-remodeling factors, as illustrated by recent biochemical studies of CHD1. Like ACF and other ISWI complexes, the CHD1 ATPase promotes the assembly of regularly spaced nucleosomes in vitro. By contrast, CHD1 does not promote the incorporation of histone H1 during chromatin assembly in vitro. These biochemical studies provide a plausible explanation for why the loss of ISWI function leads to the loss of histone H1 without causing dramatic changes in nucleosome assembly in vivo (Corona, 2007).

In other organisms, depletion of histone H1 leads to a significant decrease in the nucleosome repeat length, presumably because of the failure to efficiently incorporate histone H1 during replication-coupled chromatin assembly. By contrast, the loss of ISWI function in salivary gland nuclei leads to a decrease in the amount of histone H1 associated with chromatin without causing dramatic changes in nucleosome repeat length. It is therefore tempting to speculate that ISWI promotes histone H1 incorporation via a replication-independent process. The association of histone H1 with chromatin is far less stable than that of core histones; histone H1 undergoes dynamic, global exchange throughout the cell cycle. Photobleaching experiments in Tetrahymena and vertebrates have suggested that the majority of histone H1 molecules associated with chromatin are exchanged every few minutes, but little is known about the factors that regulate this process. Based on the current findings, ISWI is an excellent candidate for a factor that regulates the dynamic exchange of histone H1 in vivo. Further work will be necessary to determine whether ISWI promotes histone H1 incorporation via replication-dependent or -independent mechanisms (Corona, 2007).

These findings suggest that acetylation of H4K16 may regulate the association of linker histones with chromatin in vivo. The histone H4 tail is required for the nucleosome-stimulated ATPase activity of ISWI, and for its ability to slide nucleosomes and alter their spacing in vitro. The region of the H4 tail that is critical for ISWI function in vitro is a DNA-bound basic patch (R17H18R19) adjacent to H4K16, the residue that is acetylated by the MOF histone acetyltransferase. The acetylation of H4K16 interferes with the ability of ISWI to interact with the histone H4 tail and alter the spacing of nucleosome arrays in vitro. Consistent with these findings, dosage compensation is necessary and sufficient for the decondensation of the X chromosome in Iswi mutant larvae, and genetic studies have revealed a strong functional antagonism between ISWI and MOF. Thus, H4K16 acetylation may function as a switch that regulates the histone H1 assembly mediated by ISWI (Corona, 2007).

Microarray studies revealed that ISWI is required for the proper expression of a large number of genes. These findings are consistent with numerous studies implicating ISWI in transcriptional regulation in vitro and in vivo. Does ISWI modulate transcription by altering higher-order chromatin structure? It is suspected that ISWI regulates transcription and higher-order chromatin structure via distinct mechanisms, since no obvious correlation is observed between the magnitude of the changes in gene expression and chromosome structure observed in Iswi mutant larvae. This is consistent with genetic studies in other organisms that have revealed that the loss of histone H1 does not cause dramatic changes in gene expression. No correlation was observed between the magnitude of transcriptional derepression and gene size in Iswi mutant larvae, as would be expected if ISWI relieved a general block to transcriptional elongation by Pol II. It should be noted, however, that relatively subtle, but biologically important, changes in gene expression may have escaped detection in the microarray studies. Further work will be necessary to clarify this issue and to determine whether ISWI regulates transcription and higher-order chromatin structure via distinct or related mechanisms (Corona, 2007).


cDNA clone length - 3715

Bases in 5' UTR - 326

Bases in 3' UTR - 309


Amino Acids - 1027

Structural Domains

The 1,027-residue Iswi protein contains the DNA-dependent ATPase domain characteristic of the SWI2 protein family but lacks the three other domains common to Brahma and SWI2, including the bromodomain. In contrast, the Iswi protein is highly related (70% identical) to the human hSNF2L protein over its entire length, suggesting that they may be functional homologs (Elfring, 1994).

The 140 kDa subunit of NURF has been identified as Iswi, highly related to yeast SWI2/SNF2. The relationship extends only to the ATPase domain, as Iswi lacks a bromodomain (Tsukiyama, 1995b).

Energy-dependent nucleosome remodeling emerges as a key process endowing chromatin with dynamic properties. However, the principles by which remodeling ATPases interact with their nucleosome substrate to alter histone-DNA interactions are only poorly understood. A substrate recognition domain has been identified in the C-terminal half of the remodeling ATPase Iswi and its structure has been determined by X-ray crystallography. The structure comprises three domains, a four-helix domain with a novel fold and two alpha-helical domains related to the modules of c-Myb, SANT and SLIDE, which are linked by a long helix. An integrated structural and functional analysis of these domains provides insight into how Iswi interacts with the nucleosomal substrate (Grüne, 2003).

Because Iswi-C can bind to nucleosomes, attempts were made to model such a complex using the available nucleosome structure. Since the SLIDE domain is most important for DNA binding, the structure of Iswi-C was modeled to bind a DNA duplex by superimposing SLIDE with c-Myb repeat R3 bound to DNA. Subsequently, this complex was positioned onto the nucleosomal DNA. Most potential DNA binding sites on the nucleosome are excluded because either the major groove of the nucleosomal DNA is inaccessible or because steric clashes occur between Iswi-C and the nucleosome. However, at seven different positions at the top and bottom of the nucleosomal disk, the major groove is exposed and Iswi-C can be positioned without steric clashes and without assuming any deviations from the crystal structure. At these positions, Iswi-C can be clamped onto the nucleosome with the central cylinder functioning as a molecular ruler of about 50 Å spanning the width of the nucleosomal disk. One jaw of the clamp is formed by the recognition helix of the SLIDE domain inserted into the major groove of DNA, and the other jaw is formed by the HAND domain protruding from the opposite side of the cylinder. In these models conserved basic residues of the SLIDE domain, the spacer helix, and the HAND domain would point toward the nucleosome, whereas the negatively charged SANT domain surface would point away from it (Grüne, 2003).

Imitation SWI: Evolutionary Homologs | Regulation | Developmental Biology | Effects of Mutation | References

date revised: 25 June 1999

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.