Imitation SWI


Protein Interactions: Iswi interactions in nucleosome remodeling complexes

The SANT domain is a novel motif found in a number of eukaryotic transcriptional regulatory proteins that was identified based on its homology to the DNA binding domain of c-myb. SANT domains are found within subunits of ATP-dependent chromatin remodeling enzymes (yeast Swi3p, Rsc8p, human BAF155/170, and Drosophila Iswi), as well as histone acetyltransferase (yeast and human Ada2p) and deacetylase (coREST, Mta1-L2, and N-CoR) complexes. The SANT domain is shown to be essential for the in vivo functions of yeast Swi3p, Ada2p, and Rsc8p, subunits of three distinct chromatin remodeling complexes. The Ada2p SANT domain is essential for histone acetyltransferase activity of native, Gcn5p-containing HAT complexes. Furthermore, kinetic analyses indicate that an intact SANT domain is required for an Ada2p-dependent enhancement of histone tail binding and enzymatic catalysis by Gcn5p. These results are consistent with a general role for SANT domains in functional interactions with histone N-terminal tails (Boyer, 2002).

This analysis of the Gcn5p/Ada2p dimeric complex suggests a model in which the SANT domain mediates interactions between remodeling enzymes and their chromatin substrates. It has been shown that the catalytic domain of yeast GCN5 must undergo a conformational change in the histone binding groove before efficient binding of H3 peptide can occur. A model is favored in which the SANT domain of Ada2p stabilizes this high-affinity histone binding conformation that not only permits efficient binding of histone but that also properly aligns substrates and the catalytic residues within the active site for the general base-assisted attack of the epsilon-amine on the acetyl-CoA. In this model, the SANT domain might interact with Gcn5p and thereby alter the active site, or the SANT domain may directly interact with the histone tail in order to present it in the proper conformation for efficient binding and catalysis by Gcn5p. This model also provides an explanation for why the Gcn5p/Ada2pDeltaSANT subcomplex is less active on nucleosomal substrates than free Gcn5p. In this case, disruption of the SANT domain may lead to an improper presentation or binding of the histone H3 tail (Boyer, 2002).

A model of the SANT domain as a tail-presenting module would be in agreement with a similar role being played by other SANT-domain-containing proteins. For instance, Drosophila Iswi contains a SANT domain as well as an ATPase domain that is highly related to yeast Swi2p/Snf2p. In contrast to the DNA-stimulated ATPase activity of ySwi2p, the ATPase activity of Iswi requires the histone H4 N-terminal tail. The model predicts that the SANT domain of Iswi may function as a histone H4 tail-presenting module that facilitates the ATPase/remodeling reaction. Likewise, the SANT domains found within subunits of other ATP-dependent remodeling complexes (e.g., NURD/Mi-2 or SWI/SNF) or within subunits of histone deacetylase complexes (Co-REST) may also facilitate functional interactions with the histone tails that play key roles in the regulation of their enzymatic activities in vivo (Boyer, 2002).

The ATPase Iswi is a subunit of several distinct nucleosome remodeling complexes that increase the accessibility of DNA in chromatin. Isolated Iswi protein itself is able to carry out nucleosome remodeling, nucleosome rearrangement, and chromatin assembly reactions. The ATPase activity of Iswi is stimulated by nucleosomes but not by free DNA or free histones, indicating that Iswi recognizes a specific structural feature of nucleosomes. Nucleosome remodeling, therefore, does not require a functional interaction between Iswi and the other subunits of Iswi complexes. The role of proteins associated with Iswi may be to regulate the activity of the remodeling engine or to define the physiological context within which a nucleosome remodeling reaction occurs (Corona, 1999).

Caf1 is found in NURF complex, NURF is a protein complex of four distinct subunits that assists transcription factor-mediated chromatin remodeling. One NURF subunit, Iswi, is related to the transcriptional regulators Drosophila brahma and yeast SWI2/SNF2. A second integral subunit of NURF (the 55-kDa subunit) has been termed p55 and is generally associated with polytene chromosomes. In general, p55 is not found to be focused at specific polytene chromosome loci as has been frequently observed for sequence specific transcription factors. The predicted sequence of p55 reveals a WD repeat protein that is identical with the 55-kDa subunit of the Drosophila chromatin assembly factor (Caf1). Caf1 but not NURF, associates with histone acetyltransferase (HAT), and is found in a complex with active HAT enzyme(s) in nuclear extracts. Given that WD repeat proteins related to p55 are associated with histone deacetylase and histone acetyltransferase, these findings suggest that p55 and its homologs may function as a common platform for the assembly of protein complexes involved in chromatin metabolism (Martinez-Balbás, 1998).

ACF, an ATP-utilizing chromatin assembly and remodeling factor is a multisubunit factor that contains Iswi protein. ACF is distinct from NURF, another Iswi-containing factor. ACF contains four polypeptides with apparent molecular masses of 47, 140, 170 and 185 kDa. Iswi is the 140 kDa component of ACF. In chromatin assembly, purified ACF combined with additional core histone chaperone (such as NAP-1 or CAF-1) are sufficient for the ATP-dependent formation of periodic nucleosome arrays. In chromatin remodeling, ACF is able to modulate the internucleosomal spacing of chromatin by an ATP-dependent mechanism. ACF, acting with NAP-1 can mediate promoter-specific nucleosome reconfiguration by Gal4-VP16 in an ATP-dependent manner. ACF can act catalytically by an ATP-dependent mechanism to modulate nucleosome spacing in the absence of a core histone chaperone. These results suggest that ACF acts catalytically both in chromatin assembly and in the remodeling of nucleosomes that occurs during transcriptional activation. The reaction mixture has a core histone octamer:Iswi ratio of about 90:1. This octamer:Iswi molar ratio of 90:1 reflects a minimal nucleosome:ACF ratio in the assembly reaction. The dNAP-1 polypeptide:Iswi polypeptide molar ratio is about 830:1, while the dNAP-1 polypeptide;histone polypeptide molar ratio is roughly 1:1. Thus these data suggest a catalytic function for ACF and a stoichiometric function for dNAP-1 as a core histone chaperone. It is concluded that Iswi is contained in two or more multi-protein complexes. Iswi and other closely related proteins are thought to function as ATP-driven, DNA-translocating motors that can displace histones from DNA. It is difficult, however, to envision a specific mechanism for a DNA-translocating motor in the deposition of nucleosomes. In this context, it is useful to consider a two-step mechanism for chromatin assembly, in which histone deposition by core histone chaperones (such as NAP-1 or CAF-1) initially occurs by an ATP-independent mechanism and is then followed by the ATP-dependent modulation of the internucleosomal spacing by ACF (Ito, 1997).

Chromatin assembly is a two step process; the second step is carried out by an multiprotein complex termed ACF (for ATP-utilizing chromatin assembly and remodeling factor). Purified ACF acts catalytically (at approximately one ACF protomer per 90 core histone octamers) in the deposition of histones to yield periodic nucleosome arrays in an ATP-dependent process. This chromatin assembly reaction, which can be carried out in a purified reconstituted system, requires ACF, core histones, DNA, ATP, and a histone chaperone (NAP-1 and CAF-1 were each found to function as histone chaperones in conjunction with ACF). The most purified preparations of ACF have been observed to consist of Iswi protein in addition to three other polypeptides. Iswi, a component of multiple chromatin assembly machines, copurifies precisely with ACF. The other three polypeptides of 47, 170, and 185 kD were assigned tentatively as subunits of ACF (Ito, 1997).

GAGA factor is known to remodel the chromatin structure in concert with nucleosome-remodeling factor NURF in a Drosophila embryonic S150 extract. The promoter region of fushi tarazu carries several binding sites for GAGA factor, which triggers chromatin remodeling. It is possible that GAGA factor may activate transcription by an antirepressor mechanism. To test this possibility, experiments were carried out starting from a naked DNA template. In contrast to the preassembled chromatin, little activation (up to 1.5-fold) of ftz transcription is observed after preincubation of the naked template DNA with GAGA factor and the S150 extract. This indicates that only trace levels of activation may be caused by elimination of nonspecific DNA-binding proteins in the presence of GAGA factor. These results also suggest that the GAGA factor-mediated transcriptional activation occurs specifically on the chromatin template. These observations suggest that GAGA factor-mediated chromatin remodeling is required for the proper expression of ftz in vivo. The nucleosome structure surrounding nucleotide 350 in front of the TATA element (and the TATA element itself) of ftz is disrupted by incubation with GAGA factor and the S150 extract. Both transcriptional activation and chromatin disruption are blocked by an antiserum raised against Iswi or by base substitutions in the GAGA factor-binding sites in the ftz promoter region. These results demonstrate that GAGA factor- and Iswi-mediated disruption of the chromatin structure within the promoter region of ftz activates transcription on the chromatin template (Okada, 1998).

The Drosophila nucleosome remodeling factor (NURF) is a protein complex of four subunits that assists transcription factor-mediated perturbation of nucleosomes in an ATP-dependent manner. The role of NURF has been investigated in activating transcription from a preassembled chromatin template. NURF is able to facilitate transcription mediated by a GAL4 derivative carrying both a DNA binding and an activator domain. When preassembled chromatin templates are remodeled with GAL4 (1-147) and fused to a constitutive activating region of the heat shock transcription factor HSF (GAL4-HSF) , activation is 58-fold with freshly added ATP in the remodeling reaction and 63-fold with ATP carried over from the chromatin assembly step. This level of activation is 16% of the absolute level of transcription observed on naked DNA. The activation is ATP-dependent, since substantially reduced activation by GAL4-HSF is observed when ATP is omitted during the remodeling step. The results indicate that the repression of basal transcription imposed by nucleosomes is significantly relieved by the joint action of GAL4-HSF and an ATP-dependent nucleosome remodeling activity associated with reconstituted chromatin. Substantial remodeling occurs over the promoter, as revealed by the smearing of the oligonucleosomal DNA ladder at intermediate stages of nuclease digestion and by the decreased abundance of the 146 bp mononucleosome fragment. Interestingly, for at least 100 nucleotides once nucleosome remodeling by the DNA binding factor is accomplished, a high level of NURF activity is not continuously required for recruitment of the general transcriptional machinery and transcription. These results provide direct evidence that NURF is able to assist gene activation in a chromatin context, and identify a stage of NURF dependence early in the process leading to transcriptional initiation (Mizuguchi, 1998).

NURF can facilitate transcription promoted by at least three different DNA binding motifs (the Zn finger/based region of GAGA factor, the winged helix-turn-helix of HSF, and the zinc cluster of GAL4). These finding, together with the relative abundance of NURF and the absence of physical interactions between NURF and the GAGA factor, suggest a model whereby the transient, nonspecific action of NURF on chromatin creates a window of opportunity for nucleosome disorder that is exploited by the sequence-specific DNA binding motif. Such a model, where DNA binding motifs interacting with clustered cognate elements are assisted by the independent action of NURF to overcome nucleosome organization, is different from one involving the physical recruitment of the SWI/SNF remodeling complex. The additional recruitment of NURF by as yet unidentified DNA binding factors to increase local activity may also be accomodated by the model proposed in this paper. The requirement of an activating region in these experiments suggests that it is necessary for recruitment of the RNA polymerase II apparatus along with the potential nucleosome disordering activities of TFIID. The activating region may also be needed for recruitment of other chromatin remodeling activities, for directly remodeling nucleosome structure, or for stimulating productive transcription through downstream nucleosomes (Mizuguchi, 1997).

The Iswi ATPase of Drosophila is a molecular engine that can drive a range of nucleosome remodelling reactions in vitro. Iswi is important for cell viability, developmental gene expression and chromosome structure. It interacts with other proteins to form several distinct nucleosome remodelling machines. The chromatin accessibility complex (CHRAC) is a biochemical entity containing Iswi in association with several other proteins. The two smallest CHRAC subunits, CHRAC-14 and CHRAC-16, have been identified. They contain histone fold domains most closely related to those found in sequence-specific transcription factors NF-YB and NF-YC, respectively. CHRAC-14 and CHRAC-16 interact directly with each other as well as with Iswi, and are associated with functionally active CHRAC. The developmental expression profiles of both subunits suggest specialized roles in chromatin remodelling reactions in the early embryo for both histone fold subunits (Corona, 2000).

Interaction of Iswi with Acf1

The assembly of core histones and DNA into periodic nucleosome arrays is mediated by ACF, an Iswi-containing factor, and NAP-1, a core histone chaperone, in an ATP-dependent process. The isolation of Drosophila Acf1 cDNA, which encodes the p170 and p185 forms of the Acf1 protein in ACF, is described. Acf1 is a novel protein that contains two PHD fingers, one bromodomain, and two new conserved regions. Human WSTF, which is encoded by one of a number genes that is deleted in Williams syndrome individuals, is the only currently known mammalian protein with each of the conserved motifs in Acf1. Purification of the native form of Acf1 led to the isolation of ACF comprising Acf1 (both p170 and p185 forms) and Iswi. Native Acf1 did not copurify with components of NURF or CHRAC, which are other Iswi-containing complexes in Drosophila. Purified recombinant ACF, consisting of Acf1 (either p185 alone or both p170 and p185) and Iswi, catalyzes the deposition of histones into extended periodic nucleosome arrays. Notably, the Acf1 and Iswi subunits function synergistically in the assembly of chromatin. Iswi alone exhibits a weak activity that is approximately 3% that of ACF. These results indicate that both Acf1 and Iswi participate in the chromatin assembly process and suggest further that the Acf1 subunit confers additional functionality to the general 'motor' activity of Iswi (Ito, 1999).

The high molecular component of ACF has now been cloned. ATP-utilizing chromatin assembly and remodeling factor (Acf) encodes two varients, the 170- and 185-kD (p170 and p185) subunits of ACF (Note: the convention is to refer to the multiprotein complex as ACF and to the high molecular weight subunit as Acf). Purification of native Acf from Drosophila embryos leads to the isolation of ACF consisting of Acf (both p170 and p185 forms) and subunits. Acf does not, however, copurify with components of NURF (Tsukiyama, 1995a) or CHRAC (Varga-Weisz, 1997), which are other chromatin remodeling complexes from Drosophila that similarly contain an Iswi subunit. Studies of purified recombinant ACF reveal that the Acf and Iswi subunits function synergistically in the ATP-dependent assembly of nucleosome arrays. The purified reconstituted system requires ACF, core histones, DNA, ATP, and a histone chaperone. NAP-1 and CAF-1 were each found to function as histone chaperones in conjunction with ACF (Ito, 1999).

To investigate the role of Acf in nucleosome assembly by ACF, standard chromatin assembly reactions were carried out with recombinant Acf alone, Iswi alone, or ACF. These experiments revealed no detectable chromatin assembly with either Acf alone or Iswi alone. In contrast, chromatin assembly activity was seen with ACF and is dependent on the presence of ATP. The finding that 1 unit of ACF yields greater chromatin assembly activity than 10 units of either subunit alone indicates that the Acf and Iswi subunits of ACF function cooperatively in the assembly of periodic nucleosome arrays. It is estimated that ACF assembles chromatin ~30-fold more efficiently than Iswi alone. Because the Acf and Iswi subunits are both required for full ACF activity, purified preparations of the separate subunits were tested to determine if they could be combined to yield active ACF. Chromatin assembly reactions were carried with Acf-Flag and Flag-Iswi that had been combined after synthesis and purification (postsynthetic) as well as with Acf + Flag-Iswi that had been cosynthesized and purified as the ACF complex (cosynthesis). The activity of the postsynthetically combined ACF subunits is essentially identical to that of the cosynthesized subunits. Therefore, cosynthesis of the Acf and Iswi subunits is not necessary to obtain fully active ACF. In addition, full ACF activity can be achieved with the p185 form of Acf along with Iswi (Ito, 1999).

The chromatin accessibility complex (CHRAC) was originally defined biochemically as an ATP-dependent 'nucleosome remodelling' activity. Central to its activity is the ATPase Iswi, which catalyses the transfer of histone octamers between DNA segments in cis. In addition to Iswi, four other potential subunits were observed consistently in active CHRAC fractions. The p175 subunit of CHRAC has been identified as Acf1, a protein known to associate with Iswi in the ACF complex. Interaction of Acf1 with Iswi enhances the efficiency of nucleosome sliding by an order of magnitude. Remarkably, it also modulates the nucleosome remodelling activity of Iswi qualitatively by altering the directionality of nucleosome movements and the histone 'tail' requirements of the reaction. The Acf1-Iswi heteromer tightly interacts with the two recently identified small histone fold proteins CHRAC-14 and CHRAC-16. Whether topoisomerase II is an integral subunit has been controversial. Refined analyses now suggest that topoisomerase II should not be considered a stable subunit of CHRAC. Accordingly, CHRAC can be molecularly defined as a complex consisting of Iswi, Acf1, CHRAC-14 and CHRAC-16 (Eberharter, 2001).

A heterodimeric complex of Acf1 and Iswi previously had been termed 'ACF'. In this context, Acf1 significantly increases the activity of Iswi in chromatin assembly. Since Acf1 has been identified as a component of CHRAC, the impact of Acf1 on Iswi-induced nucleosome sliding was examined. The directionality of nucleosome sliding differs depending on whether the reaction is catalysed by Iswi alone or by CHRAC. Flag-tagged Iswi and Acf1 were expressed from baculovirus vectors in insect cells, affinity purified and assayed for nucleosome sliding. In agreement with previous results, catalytic amounts (2-3 fmol) of Iswi move a mononucleosome from the center of a 248 bp rDNA fragment to the fragment end. No mobility was observed when the end-positioned nucleosome is exposed to Iswi. In contrast to the movement generated by Iswi, CHRAC catalyses nucleosome sliding from the end to the center of the DNA fragment. Strikingly, CHRAC-type directionality of nucleosome sliding is also obtained if Acf1 is added to Iswi, either after separate expression or by co-expression of both proteins in Sf9 cells. While Acf1 alone is inactive for nucleosome sliding, it boosts Iswi activity by at least an order of magnitude such that 10-fold lower enzyme concentrations (0.3-0.5 fmol) are required for nucleosome mobilization. Most importantly, Acf1 changes the directionality of sliding such that end-positioned nucleosomes move to central positions (Eberharter, 2001).

In order to determine whether Acf1 has an additional effect on the kinetics of nucleosome mobility under these conditions, a time course of nucleosome mobility was performed. The amounts of enzymes were chosen such that complete mobilization of the nucleosome was expected after 90 min (10-fold less ACF than Iswi). At any given time point throughout the reaction, the ratio of nucleosomes that had been mobilized to those that had not moved was determined. Nucleosome movement in the two reactions proceeds with similar speed, indicating that ACF is about an order of magnitude more efficient in nucleosome mobilization than Iswi alone. This could be explained most readily if Acf1 stimulates the ATPase activity of Iswi. To determine whether this was the case, the enzymes were compared in standard ATPase assays. Iswi alone shows a robust (7-fold) nucleosome stimulation of ATPase. This response to a nucleosomal structure remains unaltered if Acf1 is added, either after separate expression or through co-expression. While Acf1 alone does not show any sign of ATPase activity, it also does not stimulate the ATPase of Iswi significantly (7-fold stimulation over the free DNA level in all cases) (Eberharter, 2001).

Deletion of the H4 N-termini completely abolishes the ability of CHRAC to slide nucleosomes, whereas removal of any other histone tail has only minor effects. In contrast, Iswi-induced sliding not only requires the histone H4 N-termini (like CHRAC), but is also impaired if any of the other tails are deleted. Since Acf1 modulates the directionality of nucleosome sliding to resemble that of CHRAC, the histone tail dependence of ACF-induced nucleosomal sliding was tested. As expected, deletion of the H4 tail completely abolishes the remodelling activity of ACF. Removal of any other histone tail, however, has only little influence on the sliding activity of ACF. This result reinforces the notion of a qualitative alteration of the nucleosome remodelling activity of Iswi by Acf1, which points to altered interaction with the nucleosomal substrate (Eberharter, 2001).

Acf1 is a member of a growing family of proteins with similar domain architecture including WSTF, one of the genes invariantly deleted in William-Beuren syndrome patients. In addition to several other sequence similarities, these factors contain a prominent C-terminal bromodomain and one or two PHD fingers. Acf1-like factors so far have been found exclusively in association with Iswi. Complexes consisting of just Iswi and an Acf1-like factor were purified from Drosophila, human and yeast cells. One of the Iswi-containing remodelling factors in Xenopus extracts contains an unknown protein (p175) in addition to Iswi and Acf1. CHRAC, defined as a four-subunit complex consisting of Iswi, Acf1, CHRAC-14 and CHRAC-16, has been identified so far in human and fly cells. Although Iswi can function as a remodelling ATPase in certain in vitro assays, it has never been purified from a native source on its own. These data suggest that Iswi and an Acf1-like factor constitute a functional core module with which other proteins may associate to generate a family of diverse remodelling machines, potentially tailored to specific functions (Eberharter, 2001).

Acf1 was first identified as a protein associated with Iswi to form the nucleosome assembly and spacing factor ACF. While either recombinant Iswi or Acf1 alone is only poorly active in an assembly system, the reconstitution of ACF from the two subunits increases the in vitro nucleosome assembly activity by some 30-fold. Iswi, expressed in a bacterial system, can, in principle, function autonomously in various cell-free remodelling assays. The direct comparison of the activity of factors expressed under similar conditions from baculovirus vectors shows that Acf1 enhances Iswi-induced nucleosome mobility by about an order of magnitude. In addition, the association of Acf1 has a striking qualitative effect as it alters the directionality of nucleosome sliding triggered by Iswi and affects the sensitivity of the ATPase towards deletion of the histone N-termini on the nucleosomal substrate. Iswi and Acf1 approach the nucleosome in a co-ordinated manner, leading to a new quality of interaction, such that Acf1 does not simply enhance the action of Iswi. Whether Acf1 interacts with DNA directly, effectively hindering the sliding of the nucleosome to the fragment end, remains to be explored. Upon association of Iswi with Acf1, 10-fold lower enzyme concentrations and correspondingly fewer ATP hydrolysis events are required to move a nucleosome as compared with free Iswi alone. It is possible that ACF has a higher affinity for the nucleosomal substrate, due to interaction domains contributed by Acf1. A decreased off-rate may lead to a higher processivity of the enzyme, converting the energy of ATP hydrolysis more effectively into directional nucleosome sliding. Testing this and alternative hypotheses will require more quantitative measurements of the parameters of the nucleosome sliding reaction (Eberharter, 2001).

The PHD finger and bromodomain are likely to be involved in Acf1 activity. PHD fingers are protein interaction surfaces found in many chromatin-bound regulators. Bromodomains are equally abundant among nuclear regulators. They are a hallmark of the remodelling ATPases of the SWI2/SNF2 type, but are absent in Iswi. Bromodomains are known interactors of acetylated histone H4 N-termini. Nucleosome remodelling by Iswi critically depends on the integrity of the H4 tail on the nucleosomal substrate. It is possible that tandem PHD fingers and a bromodomain form a cooperative interaction unit, as has been suggested recently for the KRAB proteins. It will be interesting to see whether Acf1 interacts directly with the H4 tail during nucleosome mobilization and whether histone acetylation modulates this process. The function of the N-terminal WAC domain of Acf1, which has been implicated in targeting proteins to heterochromatin, is unknown (Eberharter, 2001).

Although CHRAC was first perceived due to its activity to render nucleosomal DNA accessible, it was soon discovered that CHRAC and the related ACF may have an important role in the assembly of regular nucleosomal arrays in vitro. Nucleosome mobility is not restricted to the assembly phase, but can also be observed within an ordered nucleosomal array. CHRAC may be involved mainly in the assembly and maintenance of nucleosomal arrays with dynamic properties. Several observations are in line with such a function. (1) The in vitro phenomenology shows that CHRAC and ACF can catalyse the assembly of dynamic nucleosomal arrays. (2) The restricted expression of Iswi, Acf1 and the CHRAC-14/16 pair during Drosophila embryonic development correlates with the time of most intense nuclear division. (3) Proteins with similarity to CHRAC-14 and CHRAC-16 have been found to associate with human DNA polymerase epsilon. (4) Mutation of Iswi in male flies leads to a striking abnormality of the structure of the male X chromosome, which is marked and perhaps sensitized by specific acetylation of the histone H4 N-terminus, although the additional presence of Iswi in NURF complicates the interpretation of the mutant phenotype. The outcome of rendering nucleosomes mobile may depend on the circumstances: in vitro, CHRAC can facilitate SV40 replication by promoting the access of T antigen to a nucleosomal origin. In contrast, an ACF-like complex contributes to the targeted repression in yeast, presumably by modulating nucleosome positions in the promoters of meiosis-specific genes (Eberharter, 2001 and references therein).

The assembly of DNA into chromatin is a critical step in the replication and repair of the eukaryotic genome. Chromatin assembly is an ATP-dependent process. ATP-dependent chromatin-assembly factor (ACF) uses the energy of ATP hydrolysis for the deposition of histones into periodic nucleosome arrays, and the Iswi subunit of ACF is an ATPase that is related to helicases. ACF becomes committed to the DNA template upon initiation of chromatin assembly. ACF assembles nucleosomes in localized arrays, rather than randomly distributing them. By using a purified ACF-dependent system for chromatin assembly, it had been found that ACF hydrolyses about 2-4 molecules of ATP per base pair in the assembly of nucleosomes. This level of ATP hydrolysis is similar to that used by DNA helicases for the unwinding of DNA. These results suggest that a tracking mechanism exists in which ACF assembles chromatin as an ATP-driven DNA-translocating motor. Moreover, this proposed mechanism for ACF may be relevant to the function of other chromatin-remodelling factors that contain Iswi subunits (Fyodorov, 2002).

The results of this study, along with the presence of the helicase-related Iswi subunit in ACF, implicate a model for ACF-mediated chromatin assembly in which ACF functions as a processive ATP-driven DNA-translocating motor. By analogy to helicases and other DNA-translocating enzymes (for example, polymerases), it is reasonable to assume that the translocation of ACF along the DNA will consume one molecule of ATP per bp DNA. Additional ATP may also be required to disrupt contacts between core histones and NAP-1 as well as to establish proper histone-DNA contacts. In this manner, more than one molecule of ATP per bp might be used during the chromatin assembly process. It is also relevant to consider the potential role of superhelical tension in the DNA during chromatin assembly by ACF. For instance, Iswi and other related ATPases can generate superhelical tension in unconstrained DNA and chromatin. In fact, DNA tracking by ACF may generate transient superhelical tension in DNA. A processive ATP-driven translocation model for nucleosome mobilization has been suggested previously for polymerases and chromatin remodelling factors, and involves the formation of a loop of DNA that is propagated across the surface of the histone octamer. A DNA-tracking mechanism could also be relevant to the nucleosome sliding that has been observed with the chromatin remodelling complexes NURF25, CHRAC26, 27, and SWI/SNF28. This study provides evidence that suggests a processive DNA-translocating mechanism for ACF-mediated chromatin assembly. This process involves the use of a considerable amount of ATP. This expenditure of energy provides the cell with the packaging and organization of its genome (Fyodorov, 2002).

The nucleosome remodelling ATPase ISWI resides in several distinct protein complexes whose subunit composition reflects their functional specialization. Association of ISWI with ACF1, the largest subunit of CHRAC and ACF complexes, improves the efficiency of ISWI-induced nucleosome mobilization by an order of magnitude and also modulates the reaction qualitatively. In order to understand the principle by which ACF1 improves the efficiency of ISWI, their mutual interaction requirements were mapped and a series of ACF complexes were generated lacking conserved ACF1 domains. Deletion of the C-terminal PHD finger modules of ACF1 or their disruption by zinc chelation profoundly affects the nucleosome mobilization capability of associated ISWI in trans. Interactions of the PHD fingers with the central domains of core histones contribute significantly to the binding of ACF to the nucleosome substrate, suggesting a novel role for PHD modules as nucleosome interaction determinants. Connecting ACF to histones may be prerequisite for efficient conversion of ATP-dependent conformational changes of ISWI into translocation of DNA relative to the histones during nucleosome mobilization (Eberharter, 2004).

Therefore, binding of ACF1 to ISWI leads to a remarkable increase in energy efficiency of the nucleosome remodelling reaction catalysed by ISWI: while the amount of ATP hydrolysed in response to the nucleosome substrate remains unchanged upon ACF1 interaction, the complex moves nucleosomes an order of magnitude more efficiently than ISWI alone. This study finds that the PHD modules in the C-terminus of ACF1 are crucially involved in this activation. Previously this boost of activity has been found only in conjunction with a change of nucleosome sliding directionality. Deletion of the PHD domain selectively affects the efficiency of the sliding reaction, but does not alter the type of nucleosome movement. It appears, therefore, that structures in the N-terminus of ACF1, or simply the fact that ACF1 interacts with ISWI, determine the qualitative outcome of nucleosome mobilization (Eberharter, 2004).

PHD fingers have been proposed to serve as protein interaction surfaces, but the demonstration of contacts with the central parts of core histones in the context of a nucleosome remodelling factor is novel. Disruption of the PHD structure through zinc chelation destroys the interaction of ACF1 with histones and at the same time abolishes the stimulatory effect of ACF1 on the efficiency of nucleosome sliding. Evidently, ACF1-histone contacts are crucial for efficient nucleosome remodelling. The results are in conflict with those obtained earlier by Fyodorov (2002), who did not observe a detrimental effect of deleting the PHD or bromodomain modules in an ACF-dependent chromatin assembly system. This discrepancy may be due to the different functional assays employed (chromatin assembly versus nucleosome sliding) and perhaps also to differences in the protein expression protocol: whereas Fyodorov was unable to express soluble ACF1, the current procedure yielded soluble protein that could be functionally characterized (Eberharter, 2004).

According to the favourite model, ISWI-containing remodelling factors lift DNA off the histone surface at the edge of the nucleosome and distort it into a bulge or loop. Propagation of this distortion over the surface of the histone octamer leads to nucleosome relocation. It is considered that efficient displacement of nucleosomal DNA relative to the histone octamer requires contacts of the remodelling enzyme with both the DNA and histone moieties. ISWI alone binds nucleosomes mainly through interactions of its C-terminal SLIDE domain with nucleosomal DNA (Grüne, 2003). Although a segment of the H4 N-terminus is absolutely required for ISWI function (Clapier, 2001), no stable interactions of ISWI with histones are known. ISWI may be a relatively inefficient remodelling enzyme because it lacks a stable anchoring point on the histone body. The PHD-histone contacts documented in this study may provide such an anchoring point for the enzyme, assuring that the presumed conformational changes triggered by ATP binding and hydrolysis are efficiently converted into positional shifts of DNA relative to histones. According to a variation of this model, ACF plays an active role in the propagation of the DNA distortion ('the loop') around the histone octamer, which would necessitate changing contacts of the remodelling machinery with the histone octamer surface as it traverses around the particle. The observation that the PHD fingers of ACF1 interact with all four histones suggests that they recognize a common structural feature on the histone pairs, and is compatible with models involving multiple, different contacts of the remodeller on the histone octamer (Eberharter, 2004).

ACF1 is a prominent member of a family of related proteins in various species, which interact with ISWI to form several distinct nucleosome remodelling complexes such as ACF, NURF, WCRF and NoRC. These proteins share with ACF1 a similar domain organization including C-terminal PHD and bromodomains. A region of NURF301, the largest subunit of NURF, containing two PHD fingers and the adjacent bromodomain binds to all four core histones, but the functional consequences of these interactions have not been determined. Interestingly, human ACF1 (alias WCRF180) and the related human WSTF, which associate with the human homologue of ISWI, SNF2H, function with only one PHD module and the hSNF2H-interacting protein RSF1 does not share any sequence similarity with ACF1, except for one PHD module. It is therefore likely that one PHD module may be sufficient for function (Eberharter, 2004).

PHD fingers are diverse in sequence and may connect different proteins in various contexts. The observation that PHD modules of ACF1 serve to tether a nucleosome remodelling enzyme to its substrate adds a new function to the list that should be tested for the known or suspected modifiers of chromatin structure, such as the remodelling ATPase Mi-2, the epigenetic regulators Trithorax, Polycomb-like, Ash1, Ash2 and Lid (Eberharter, 2004).

PHD modules are frequently found in the direct neighbourhood of a bromodomain. Examples include the ACF1-related proteins discussed here, and also the histone acetyltransferases CBP and p300 and the KAP1 repressor. From their functional analysis of domains involved in KAP-1 repression, it has been concluded that both domains form an integrated, cooperative unit involved in binding the Mi2alpha subunit of the NuRD complex. In the context of histone binding, the idea of cooperativity between PHD fingers and bromodomains is attractive. While it is well established that bromodomains interact preferentially with acetylated N-termini of histones H3 and H4, this study shows that the PHD modules of ACF1 interact with the central domains of the core histones. This interaction may thus be further modulated by additional contacts of bromodomain with appropriately modified N-termini, as a means of fine-tuning nucleosome remodelling activity in response to the histone modification status. Whether this principle applies to ACF remains to be seen. However, the Bromo-PHD modules of the histone acetyltransferase p300 have been functionally characterized and both domains have been found to cooperate for preferential binding to highly acetylated nucleosomes in a stringent assay (Eberharter, 2004).

Interestingly, deletion of the ACF1 bromodomain alone does not diminish nucleosome sliding, but consistently improves it. Given the limited knowledge on the structure of the ACF1 C-terminus as an entity, it is difficult to interpret this observation. Conceivably, the interaction of bromodomain with another domain of the remodelling machinery dampens its activity until a conformational change triggered by its interaction with an acetylated histone N-terminus unleashes full remodelling potential. This scenario suggests a strategy by which histone acetylation could regulate remodelling activity other than by simply increasing the affinity of the remodelling machinery for the substrate (Eberharter, 2004).

This study mapped AID, of ISWI, required for ACF1 binding, to the very C-terminus of ISWI, directly adjacent to the SANT/SLIDE module, the nucleosome interaction determinant of ISWI. Through this interaction, ACF1 is brought close to the nucleosome surface, but this interaction also leads to the apposition of several domains on both ACF subunits: the SANT domain of ISWI, which may be involved in contacting histone tails; the SLIDE domain of ISWI, which contacts nucleosomal DNA (Grüne, 2003); the PHD fingers of ACF1, which binds the histone octamer surface; and finally bromodomain with its potential for interactions with appropriately modified histone N-termini. Unravelling the sequence and dynamics of enzyme-substrate contacts during the remodelling process remains a major challenge for future research (Eberharter, 2004).

Interaction of Iswi with NURF301

To investigate the role of NURF301 in chromatin remodeling, the entire NURF301 open reading frame was introduced into a baculovirus vector for protein expression. NURF301 and the other three NURF subunits were expressed individually and each of the epitope-tagged NURF components was immunopurified. Iswi, NURF55, and NURF38 were expressed highly in Sf9 cells. By contrast, recombinant NURF301 was poorly expressed and subject to proteolysis. Nevertheless, it was possible to reconstitute recombinant NURF complexes from individual subunits and purify the full complex from partial reconstitutes by glycerol gradient centrifugation. The reconstituted NURF complex clearly displays nucleosome-dependent ATPase activity. The reconstituted NURF complex also catalyzes nucleosome mobility. Mononucleosomes deposited by salt gradient dialysis on a 359 bp hsp70 promoter fragment adopt one of four major positions (N1-N4 nucleosomes). Native NURF catalyzes the movement of nucleosomes from the N1 and N2 positions to the preferred N3 position, and induces a slight positional change of N4 nucleosomes. NURF reconstituted from individual subunits has approximately the same activity as native NURF purified by conventional chromatography. Subcomplexes containing combinations of NURF subunits were generated and it was found that a complex reconstituted from individually purified Iswi and NURF301 could catalyze ATP-dependent nucleosome sliding. Neither NURF301 alone nor a mixture of Iswi, NURF55, and NURF38 changed nucleosome distributions when assayed at equivalent concentrations (Xiao, 2001).

To improve expression of NURF301, Sf9 cells were coinfected with recombinant baculovirus for all four NURF subunits. SDS-PAGE showed that coexpression greatly reduces proteolysis of NURF301 and increases the yield of the recombinant complex by 100-fold or greater. Recombinant NURF purified from coinfected cells showed nucleosome-stimulated ATPase activity; little or no stimulation was observed when nucleosomes were substituted by free DNA or core histones. Recombinant NURF purified from coinfected cells catalyzes nucleosome sliding in a similar manner to native and reconstituted recombinant NURF complexes (Xiao, 2001).

The Iswi ATPase by itself is capable of sliding nucleosomes, although sliding is less efficient and lacks positional specificity when compared to native Iswi complexes. Recombinant Iswi is less efficient than coexpressed, recombinant NURF in nucleosome-stimulated ATP hydrolysis. Moreover, sliding of nucleosomes to the preferred N3 position is not observed even when Iswi concentrations are increased by an order of magnitude over NURF. Taken together, these results indicate that efficient and accurate nucleosome sliding requires contributions from both Iswi and NURF301 (Xiao, 2001).

To investigate whether NURF301 provides a structural framework for assembly of the NURF complex, its interactions with the three smaller NURF subunits were analyzed, using S35-labeled NURF301 protein and Iswi, NURF55, and NURF38 proteins purified from baculovirus-infected cells. NURF301 can bind directly to each of the other NURF subunits, but no stable interactions were detected among Iswi, NURF55, and NURF38. The results indicate that NURF301 provides a scaffold to organize Iswi, NURF55, and NURF38 within the NURF complex (Xiao, 2001).

These interaction studies were extended by examining whether NURF can interact stably with its substrate -- nucleosomes. Recombinant NURF complex immobilized on beads binds to reconstituted nucleosomes (359 bp) and to nucleosome core particles (146 bp, lacking linker DNA). Nucleosome binding is ATP independent. To analyze the role of NURF301 in nucleosome binding, segments of the NURF301 coding region were fused to glutathione-S-transferase (GST). Three regions in NURF301 show prominent binding to nucleosomes or core particles, as revealed by GST-301 interactions. The results demonstrate that NURF301 is important for nucleosome binding, although a role for the other NURF subunits is by no means excluded (Xiao, 2001).

HMGA proteins can induce structural changes in nucleosomes and also alter DNAconformation to assemble stereospecific, multiprotein DNA complexes at enhancers. An N-terminal deletion of amino acids 1-121 in NURF301 was constructed that eliminates just the HMGA domain (DeltaN301). Cells were then co-infected with baculovirus expressing the four NURF subunits, NURFDeltaN301, wild-type Iswi, NURF55, and NURF38, and purified the protein complex. Deletion of the N-terminal HMGA domain does not impair assembly of NURF. However, the NURFDeltaN301 complex shows reduced binding to nucleosomes as well as nucleosome core particles. Moreover, the reduced binding to nucleosomes is correlated with significant reduction in the ATPase activity. Strikingly, the NURFDeltaN301 complex also shows impaired activity in the nucleosome sliding assay. Nucleosome movements to the N3 position are significantly reduced for NURFDeltaN301 over a 10-fold concentration range; however, movement of the N4 'end' nucleosome is not significantly affected. It is concluded that the HMGA domain of NURF301 makes an important contribution to nucleosome sliding (for nucleosomes located away DNA fragment ends) (Xiao, 2001).

The nucleosome remodeling factor (NURF) is one of several ISWI-containing protein complexes that catalyze ATP-dependent nucleosome sliding and facilitate transcription of chromatin in vitro. To establish the physiological requirements of NURF, and to distinguish NURF genetically from other ISWI-containing complexes, mutations were isolated in the gene encoding the large NURF subunit, nurf301. NURF is shown to be required for transcription activation in vivo. In animals lacking NURF301, heat-shock transcription factor binding to and transcription of the hsp70 and hsp26 genes are impaired. Additionally, NURF is shown to be required for homeotic gene expression. Consistent with this, nurf301 mutants recapitulate the phenotypes of Enhancer of bithorax, a positive regulator of the Bithorax-Complex previously localized to the same genetic interval. Finally, mutants in NURF subunits exhibit neoplastic transformation of larval blood cells that causes melanotic tumors to form (Badenhorst, 2002).

ISWI, the catalytic subunit of NURF, is required for expression of the homeotic gene engrailed (en). However, ISWI is also a component of two other chromatin remodeling complexes, ACF and CHRAC. To resolve which ISWI-containing complex is required for homeotic gene expression, expression of Ultrabithorax (Ubx) and engrailed (en) were examined in nurf301 mutant animals. When both copies of nurf301 are mutated, in homozygous mutant nurf3011 larvae, expression of the Ubx protein becomes undetectable. The normal expression of Ubx in the haltere and third leg discs of wild-type third instar larvae is absent in nurf301 mutant animals. Expression of the homeotic gene en requires nurf301. The normal expression of En in the posterior compartment of imaginal discs is abolished in nurf3012 mutants. Semiquantitative RT-PCR analysis confirms that Ubx and en transcript levels are reduced in nurf301 mutant animals. These results confirm that the defects in homeotic transcription seen in iswi mutants are caused by abrogated NURF function (Badenhorst, 2002).

A positive regulator of the Bithorax-Complex, E(bx), has been localized genetically to 61A, the same cytological interval as nurf301. However, unlike numerous regulators of the BX-C, E(bx) had not been cloned. Since NURF is required for expression of Ubx, whether nurf301 corresponds to E(bx) was tested. Both alleles of E(bx) were no longer extant, so whether the mutations that were isolated in nurf301 recapitulated the published morphological properties of E(bx) mutants was tested (Badenhorst, 2002).

nurf301 mutants, like E(bx), increase the severity of bithorax (bx) mutant phenotypes. bx is a DNA regulatory element required for correct expression of Ubx in regions that give rise to the third (T3), but not second thoracic segment (T2) of the adult fly. This expression distinguishes T3 from T2 identity. Loss or reduction of Ubx levels in bx mutant animals (Ubx6.28/bx34e and Ubx6.28/bx8 mutant combinations causes a homeotic transformation of the third thoracic segment to the anterior second thoracic segment. Thus, the third thoracic segment, which is normally vestigial and naked, is transformed into the second thoracic segment, increasing its size and causing sensory bristles to develop. Moreover, the haltere (T3) is transformed toward wing fate (T2), manifested by increases in size and the development of bristles. The strength of these transformations is increased when one copy of E(bx) also is removed. Mutation of one copy of nurf301 similarly enhances bx phenotypes. With one copy of either the nurf3011, nurf3012 or a deficiency that removes nurf301 -- Df(3L)3643 -- the strength of the transformation is enhanced. nurf301 enhances both bx34e and bx8 mutations (Badenhorst, 2002).

Although NURF is required for expression of the homeotic genes in imaginal discs, neither E(bx) nor nurf301 homozygous mutant larvae display obvious homeotic transformations of the larval cuticle. The absence of mutant larval cuticle phenotypes is likely due to the large maternal dowry of nurf301 transcript contributed to embryos. Larval cuticular patterning is established before these transcripts have dissipated. Attempts were made to generate embryos lacking the maternal nurf301 contribution through use of the dominant female sterile technique. Although germ-line clones were produced using the parental chromosome, it was not possible to recover germ-line clones using nurf3011. Like ISWI, NURF301 is required for ovary development (Badenhorst, 2002).

The defects in homeotic transcription seen in nurf301 mutant animals effectively duplicate those reported for animals lacking the catalytic ISWI subunit. However, ISWI is also required to maintain higher order chromosome structure. In iswi mutants the male X chromosome is grossly disrupted relative to autosomes as revealed by polytene chromosome preparations. In nurf301 mutants the male X chromosome is similarly affected. The male X chromosome, identified by anti-MSL2 staining, is reduced in length and breadth, as seen in iswi mutant animals. The male X chromosome in homozygous mutant nurf3012 animals is highly aberrant. Disorganized male X chromosome morphology also was observed in nurf3012/nurf3013 and hemizygous nurf3011/Df(3L)3643 mutant animals. The results demonstrate that the chromosome condensation defect caused by perturbed ISWI function is mediated through the NURF complex. Although the effects on male X chromosome structure suggest that NURF can influence global chromosome structure, NURF function is not required for heterochromatic gene silencing. Reduction in NURF301 levels has no effect on position effect variegation. This is consistent with findings that showed that iswi is not a modifier of PEV (Badenhorst, 2002).

During the course of this analysis it was noticed that nurf301 mutant animals display a high incidence of melanotic tumors. Melanotic tumors have previously been reported in a number of mutant backgrounds and are generally caused by neoplastic transformation of the larval blood cells. The circulating cells (hemocytes) of the larval blood or hemolymph provide one tier of the innate immune system of insects by encapsulating or engulfing pathogens. A number of mutations have been shown to trigger the overproliferation and premature differentiation of hemocytes. Tumors form when these cells aggregate, or invade and encapsulate normal larval tissues Badenhorst, 2002).

Melanotic tumors are observed both in EMS-induced nurf301 mutants that truncate NURF301, the P-element induced mutation that reduces nurf301 transcript levels, and allelic combinations of these mutants. Tumor penetrance is extremely high (100% for nurf3012 at 25°C). Consistent with tumor development, circulating hemocyte cell number was increased dramatically in hemolymph isolated from nurf301 mutant animals. A large percentage of animals lacking ISWI, the catalytic subunit of NURF, also displayed melanotic tumors confirming that disrupted NURF function induces tumor formation. In iswi mutant animals the number of circulating hemocytes is also increased. In both nurf301 and iswi mutant hemolymph, small aggregates of hemocytes are often observed. All hemocyte cell types are present, from small round cells (prohemocytes) to crystal cells and lamellocytes (Badenhorst, 2002).

In Drosophila, larval blood cell transformation and melanotic tumor formation can be induced by inappropriate activation of either of two distinct signaling cascades: the Toll or the JAK/STAT pathway. Inappropriate activation and nuclear-localization of the Drosophila NF-kappaB homolog Dorsal, caused either by constitutive activation of the Toll receptor or removal of the inhibitor, the Drosophila IkappaB Cactus, leads to melanotic tumors in third instar larvae. In the second pathway, gain-of-function mutations in Hopscotch (Hop), the Drosophila Janus Kinase (JAK), induce melanotic tumors. Hop gain-of-function mutants cause tumor development by triggering constitutive activation and DNA-binding by the Drosophila STAT transcription factor, STAT92E (Badenhorst, 2002).

To resolve whether the melanotic tumors seen in the nurf301 mutants were caused by misregulation of either the TOLL or HOP/STAT92E pathways, whether nurf301 mutants enhance tumor phenotypes seen in constitutively active Toll or Hop mutant lines was tested. Tumor incidence in animals carrying one copy of a gain-of-function Hop mutation -- hopTum-1 -- is increased by simultaneous reduction in NURF301 levels. In contrast, removal of one copy of NURF301 fails to enhance the Toll gain-of-function allele Tl10b. The results suggest that NURF acts as a negative regulator within the Drosophila JAK/STAT signaling pathway (Badenhorst, 2002).

Molecular signatures of both JAK and Toll activation have been defined. It is known that Hop gain-of-function mutants induce expression of a complement-like protein TEP1. Overactivation of the Toll pathway also induces TEP1 synthesis but primarily induces expression of antimicrobial peptides, including Drosomycin (Drs) and Diptericin (Dpt). Loss of nurf301 induces tep1 but fails to induce drs or dpt, demonstrating that NURF301 principally affects the Hop/STAT92E pathway. Whether nurf301 interacts genetically with other known components of the Hop/STAT92E pathway was tested. Certain mutations in unpaired (upd, also known as outstretched), which encodes a ligand for the Hop receptor, display a characteristic wings-out phenotype, due to decreased activation of Hop and consequently decreased STAT92E function. When NURF301 levels are simultaneously decreased in these mutant backgrounds, animals are mostly restored to the wild-type. These genetic interactions confirm that NURF301 acts as a negative regulator of the Hop/STAT92E pathway, at a point downstream of Hop. Hence, disruption of NURF could affect either STAT92E or the targets of STAT92E. In nurf301 mutants, levels of the STAT92E transcription factor are not elevated, suggesting that NURF acts to repress the activity of STAT92E or the expression of some STAT92E target genes (Badenhorst, 2002).

An important question is how NURF is recruited to target sites in vivo. Four genes were shown in this study to be dependent on nurf301 for expression: Ubx, en, hsp26, and hsp70. All contain multiple binding sites for the GAGA factor, which is genetically required for their correct expression. On the Drosophila hsp70 and hsp26 promoters, (GA.CT)n cognate elements (to which the GAGA factor binds) are required for HSF-binding. When these sequences are deleted, HSF-binding to transgenes in polytene chromosomes is impaired, consistent with the defects seen in nurf301 mutant animals. It is therefore compelling that recent biochemical studies show that NURF and the GAGA factor bind to each other in crude extracts, and that purified NURF301 and GAGA factor interact directly in vitro. The principal interacting domains map to an N-terminal region of NURF301 and a stretch flanking the Zn finger DNA-binding motif of GAGA factor. These data suggest that NURF is recruited by the GAGA factor through specific, direct interactions with the NURF301 subunit, to catalyze local sliding of nucleosomes at bx, en, hsp26, and hsp70 promoters, increasing accessibility to sequence-specific transcription factors and RNA polymerase II. Curiously, though, reduction of nurf301 levels fails to enhance phenotypes of mutations in Trithorax-like, the gene that encodes the GAGA factor (Badenhorst, 2002).

Conceptual models of the function of chromatin remodeling machines reinforce the view that these complexes are required exclusively during gene activation to expose or 'open-up' chromatin. However, nucleosome sliding could equally be harnessed to repress genes. Phenotypic analysis of the orthologous yeast ISW2 complex suggests that ISWI-containing complexes may function both to activate and repress genes. The yeast ISW complexes appear to be recruited to sites within the genome through direct interaction with DNA-binding proteins to activate and repress genes by repositioning nucleosomes. Drosophila ISWI has been shown to be associated with transcriptionally silent regions of chromatin in salivary gland nuclei, suggesting that it may be involved in repression in this tissue. Analysis of NURF function during larval blood cell development suggests that NURF can repress targets of the JAK/STAT signaling pathway. In nurf301 mutants the expression of the complement-like protein TEP1 is induced. It remains to be established whether tep1 is a direct target of NURF (Badenhorst, 2002).

Misregulation of TEP1 is seen in mutants that overactivate the Drosophila JAK/STAT (Hop/STAT92E) signaling cascade and induce the formation of melanotic tumors. Increased signaling through the Hop/STAT92E pathway leads to the overproliferation and aberrant differentiation of larval blood cells that subsequently invade and encapsulate normal host tissue. Animals lacking NURF301, or the catalytic ISWI subunit, exhibit an identical neoplastic transformation of larval blood cells. In the absence of NURF, the proliferation and differentiation of hemocytes, and the accumulation of lamellocytes, is triggered (Badenhorst, 2002).

The data suggests that NURF normally represses targets of the Hop/STAT92E pathway. Genetic epistasis places nurf301 downstream of Hop. Loss of NURF resembles gain-of-function mutations in hop, and targets of the Hop/STAT92E cascade are up-regulated in nurf301 mutants. Normally, Hop activation leads to the expression and posttranslational modification of STAT92E. Although nurf301 mutants activate the Hop/STAT92E pathway, the levels of the STAT92E transcription factor are unchanged in NURF mutant animals. It is suggested that NURF acts downstream of STAT92E. In resting cells, in the absence of Hop/STAT92E signaling, NURF could normally repress STAT92E target genes. When NURF is removed, repression is no longer maintained, and targets are transcribed mimicking the effects of Hop activation. However, STAT activity is also influenced by a number of inhibitory, STAT-binding proteins. Among these are the suppressor of cytokine signaling (SOCS) and protein inhibitor of activated STAT (PIAS) family of inhibitors. It is possible that NURF is required for expression of one such inhibitor, and that loss of NURF activates the Hop/STAT92E cascade by removing a STAT inhibitor (Badenhorst, 2002).

The involvement of NURF in larval blood cell development agrees well with recent literature implicating a number of chromatin-modifying or chromatin-associated complexes in hemocyte development and melanotic tumor formation. Mutations in modulo, which encodes an interacting partner of the coactivator CBP, cause melanotic tumors. Of particular significance, screens for mutations that cause hematopoietic defects identified domino (dom), which encodes a member of the SWI2/SNF2 family of DNA-dependent ATPase that is distantly related to ISWI. Mutations in domino cause overproliferation of hemocytes, like NURF mutants. However, unlike NURF mutants, hemocytes fail to enter the hemolymph and remain trapped in enlarged lymph glands that become melanized. It will be interesting to assess the relative contributions of the ISWI and DOM complexes in the regulatory hierarchy of larval blood development (Badenhorst, 2002).

A striking feature of male animals that lack either NURF301 or the catalytic subunit ISWI is the distorted, bloated morphology of the male X chromosome. This implicates NURF in the maintenance of male X chromosome morphology. In flies, X chromosome dosage compensation is achieved by up-regulating transcription from the male X chromosome. One characteristic of the male X chromosome is the specific acetylation of histone H4 at Lys 16 (H4-K16), which is believed to favor a looser chromatin structure that allows increased transcription. These patterns of acetylation are established by the male-specific expression of components of the MSL complex that are tethered on the male X chromosome and subsequently recruit the histone acetyl transferase MOF (Badenhorst, 2002).

Genetic studies demonstrate that H4-K16 acetylation antagonizes ISWI function on the X chromosome. Biochemical characterization of the ISWI-containing ACF and CHRAC complexes has revealed that they can assemble and slide nucleosomes to establish regular ordered arrays. Regular nucleosome arrays are presumed to provide better substrates for chromatin compaction and, thus, it was speculated that ACF and CHRAC might be the complexes that help compact the male X chromosome. However, NURF is the ISWI complex required for normal male X chromosome morphology. Unlike ACF or CHRAC, NURF disrupts regular, ordered arrays of nucleosomes. While it is possible that NURF is required for global aspects of higher order chromosome morphology that are needed to maintain normal male X chromosome structure, other local or transcription-based mechanisms could also account for the nurf301 and iswi phenotypes. The dosage compensation machinery is recruited to the male X chromosome at specific, high affinity sites or entry points and subsequently spreads into flanking chromatin. NURF may regulate chromatin accessibility at one or a number of these initiation sites. In the absence of NURF, entry of the dosage compensation machinery at such sites may be changed. Alternatively, NURF may control transcription of components of the sex-determination and dosage compensation pathway. Irrespective, the observed antagonistic relationship between ISWI function and H4-K16 acetylation suggests that the action of NURF on the X chromosome is correspondingly influenced by H4 lysine acetylation. This influence on NURF could be direct, as suggested by effects of acetylated H4 tail peptide on ISWI ATPase activity in vitro (Badenhorst, 2002).

ISWI is the catalytic subunit of at least three protein complexes that have demonstrated in vitro chromatin-remodeling activity: NURF, ACF, and CHRAC. Here, mutation of a NURF-specific component reproduces the published properties of mutations in iswi. Both iswi and nurf301 are required for homeotic gene expression, proper larval blood cell development, and normal male X chromosome morphology. The specific in vivo functions of the ACF and CHRAC complexes remain to be established (Badenhorst, 2002).

Clues to the function of ACF and CHRAC may be derived from studies of the human Williams Syndrome Transcription Factor-ISWI complex (WSTF-ISWI). The in vitro activities of the WSTF-ISWI complex are essentially identical to ACF. WSTF-ISWI is targeted to pericentric heterochromatin during replication and is believed to allow heterochromatin reassembly in the wake of the replication fork. It is tempting to speculate that ACF and CHRAC may have similar functions in Drosophila and could be implicated in the establishment of repressive chromatin structures after replication. As mutations that selectively compromise individual remodeling complexes become available, whole genome expression analysis will allow the relative contributions of specific complexes to gene activation and silencing events in vivo to be dissected (Badenhorst, 2002).

Properties of Iswi in promoting nucleosome remodeling

The chromatin accessibility complex (CHRAC) belongs to the class of nucleosome remodeling factors that increase the accessibility of nucleosomal DNA in an ATP-dependent manner. CHRAC induces movements of intact histone octamers to neighboring DNA segments without facilitating their displacement to competing DNA or histone chaperones in trans. CHRAC-induced energy-dependent nucleosome sliding may, in principle, explain nucleosome remodeling, nucleosome positioning, and nucleosome spacing reactions known to be catalyzed by CHRAC. The catalytic core of CHRAC, the ATPase Iswi, also mobilizes nucleosomes at the expense of energy. However, the directionality of the CHRAC- and Iswi-induced nucleosome movements differs drastically, indicating that the geometry of the native complex modulates the activity of its catalytic core (Längst, 1999).

Nucleosomes positioned either at the ends of a DNA fragment or at a central position do not move significantly from these sites even when they were incubated for extended times at 37°C. Remarkably, CHRAC is able to mobilize all nucleosomes bound to fragment ends to relocate to a more central position on the DNA. This has led to the novel conclusion that NURFs may induce energy-dependent nucleosome movements beyond the realm of diffusion. This energy-dependent nucleosome mobility also differs from the diffusion of nucleosomes within a cluster of close-by positions of identical rotational positioning by its sensitivity against histone H1. Short range, energy-independent mobility is inhibited by linker histones, while at least in crude extracts energy-dependent nucleosome mobility in chromatin that contains H1 has been observed. NURF is also able to induce nucleosome movements, again highlighting the important role of Iswi. However, NURF differs from CHRAC by its inability to function as a nucleosome spacing factor, which illustrates the influence of the molecular context on Iswi function (Längst, 1999 and references).

CHRAC catalyzes a number of reactions of seemingly different natures. It increases the accessibility of nucleosomal DNA toward restriction enzymes and eukaryotic regulators; it is able to trigger the alignment of nucleosomes with respect to a boundary in chromatin, and it is able to introduce regularity into a nucleosomal array during nucleosome assembly. The finding that CHRAC is able to trigger nucleosome movements to neighboring DNA suggests that nucleosome mobility may be involved in the above-mentioned phenomena. The outcome of nucleosome mobilization will depend on the circumstances and the assays employed. It may lead to nucleosome remodeling (nucleosomes slide off a transcription factor-binding site), nucleosome repositioning (nucleosomes move to find the optimal distance with respect to a newly introduced chromatin boundary), and chromatin regularity (nucleosomes move to align with respect to each other). Chromatin that is rendered dynamic through nucleosome mobility will be flexible to respond to changes in protein/DNA interactions by continuous energy minimizations (Längst, 1999 and references).

The ATPase Iswi can be considered the catalytic core of CHRAC and presumably also other Iswi-containing remodeling complexes, such as NURF and ACF. The finding that Iswi alone is able to mobilize nucleosomes in an energy-dependent manner lends further support to the concept of modular remodeling factors. Remarkably, the outcome of nucleosome mobilization by CHRAC and recombinant Iswi (rIswi) differ profoundly. While CHRAC only induces movements of nucleosomes from the fragment ends to center positions, rIswi catalyzes the reverse reaction. It is possible that this apparent polarity of movement reflects the particularities of the assay system more than the physiological properties of the enzymes involved. Nevertheless, the phenomenon clearly demonstrates that the subunits associated with Iswi in CHRAC modulate the outcome of the mobilization. Iswi alone may simply move nucleosomes until they reach a barrier on DNA. In artificial situations this may correspond to a fragment end or otherwise a particularly rigid DNA sequence, secondary structure, or tight DNA-binding factor. In the latter case, nucleosome mobility would lead to a statistical positioning of nucleosomes next to the bound proteins, as is indeed observed when TetR repressors interact with chromatin. CHRAC contains, in addition to Iswi, a functional dimer of topoisomerase II and may contain other DNA or histone interaction surfaces on the three as yet uncharacterized subunits. If topoisomerase II interacts with two segments of DNA during the remodeling reaction, topoII may (by analogy to the positioning of a nucleosome between two DNA-bound TetR proteins) constitute boundaries that confine nucleosome positions to a central position (Längst, 1999).

Two extreme scenarios may be envisaged to explain how Iswi induces nucleosome mobility. Iswi may affect the equilibrium of nucleosome assembly and disassembly and, depending on the precise circumstances, tilt the balance toward intact nucleosomes or toward disassembly into subnucleosomal structures. The rapid reassembly of a transiently disrupted nucleosome at a different site on DNA would be interpreted as nucleosome movement. The reassembly of disrupted octamers would not necessarily be restricted to the same DNA molecule but may also lead to a transfer of a histone octamer to a different, competing DNA molecule. Such a trans-displacement has recently been observed as a consequence of nucleosome remodeling by RSC. CHRAC-induced nucleosome displacement in trans cannot be detected under conditions that are very similar to those that allowed nucleosome remodeling by RSC and even in the presence of large amounts of the artificial histone chaperone poly(glutamic acid). These data also indicate that CHRAC leaves the histone octamer intact, while affecting histone/DNA interactions in a manner that facilitates the relocation of the octamer on DNA only in cis. Mobility is characterized by (1) the stability of the histone octamer and (2) the confinement of the nucleosome to DNA segments on the same molecule. This type of mobility can presently be best described by the term 'sliding'. It is unclear at present how Iswi is able to induce nucleosome sliding, but the availability of a quantitative assay should help to elucidate the underlying mechanism of sliding (Längst, 1999).

Short range nucleosome movements could theoretically be initiated by the thermal untwisting of a segment of DNA at the nucleosome edge, which would lead to a displacement of a DNA/histone contact by perhaps as little as one base pair. This small DNA distortion may be propagated randomly over the histone octamer surface. The accumulation of many such single base translocations would lead to a 'screwing' of DNA over the histone octamer surface and may theoretically lead to significant nucleosome movements. Nucleosome mobilization factors may enhance the screwing translocation by active twisting of the DNA or by endowing the system with directionality of propagation of a DNA distortion through the nucleosome. Alternatively, a transiently disrupted histone/DNA contact at the nucleosomal edge may initiate the release of a DNA segment from the octamer. The histone contact may then recapture DNA at a different position, creating a DNA bulge the size of which would be mainly determined by the stiffness of the DNA molecule. Movement of this bulge over the octamer surface would again lead to nucleosome translocation. Efficient release of longer DNA segments might require energy input provided by a remodeling factor. One popular model suggests that a remodeling factor might use the energy of ATP to move on DNA, peeling DNA segments off a histone octamer that can be recaptured by the same octamer at a different position. Examples for such engines that are able to disrupt histone/DNA interactions are RNA polymerases, DNA helicases, and the RuvAB motor protein that moves a Holliday junction through a nucleosome. Depending on the precise experimental conditions, the action of these enzymes may lead to either histone octamer trans-displacement or its repositioning in cis. A model has been suggested whereby RSC invades a nucleosome from the edge, creating a DNA bulge that is moved over the entire octamer surface. Such a scenario would involve major alterations of the DNA path as are indeed observed upon nucleosome remodeling by RSC and SWI/SNF complexes. By contrast, CHRAC-induced nucleosome movements do not correlate with changes in the 10 bp repeat DNase I digestion pattern, suggesting that nucleosome remodeling by SWI2/SNF2 and Iswi is based on different mechanisms (Längst, 1999 and references).

The ATPase Iswi is a subunit of several distinct nucleosome remodeling complexes that increase the accessibility of DNA in chromatin. The isolated Iswi protein itself is able to carry out nucleosome remodeling, nucleosome rearrangement, and chromatin assembly reactions. The ATPase activity of Iswi is stimulated by nucleosomes but not by free DNA or free histones, indicating that Iswi recognizes a specific structural feature of nucleosomes. Nucleosome remodeling, therefore, does not require a functional interaction between Iswi and the other subunits of Iswi complexes. The role of proteins associated with Iswi may be to regulate the activity of the remodeling engine or to define the physiological context within which a nucleosome remodeling reaction occurs (Corona, 1999).

Mi-2 and Iswi, two members of the Snf2 superfamily of ATPases, reside in separate ATP-dependent chromatin remodelling complexes. The abilities of recombinant Mi-2 and Iswi to bind two different mononucleosome species were tested in a bandshift assay. Binding to a mononucleosome (core particle) containing 146 bp of DNA was compared with binding to a nucleosome assembled on the 248 bp fragment. The primary difference between these substrates is that in the 146 bp nucleosome all DNA is entirely in contact with the histone octamer, whereas the 248 bp nucleosome contains ~100 bp of naked DNA extending from the octamer. Incubation with increasing amounts of Mi-2 results in the formation of a Mi-2-core particle complex or a complex of Mi-2 with the 248 bp nucleosome. In contrast, stable Iswi-core particle complexes could not be detected, even if higher amounts of Iswi were used. Incubation of increasing amounts of Iswi with the 248 bp nucleosome produces up to three Iswi-nucleosome complexes, suggesting that more than one Iswi molecule can simultaneously associate with this substrate (Brehm, 2000).

It is concluded that Mi-2 can stably associate with a mononucleosomal substrate lacking free DNA, in agreement with the finding that free DNA does not stimulate the Mi-2 ATPase. In marked contrast, Iswi only binds the nucleosome when free DNA is protruding from the histone octamer. These results strongly suggest that Mi-2 and Iswi recognize different features of the nucleosome (Brehm, 2000).

ATP hydrolysis by Mi-2 is believed to drive remodelling of chromatin by the Mi-2 complex. In order to characterize nucleosome remodelling directly, use was made of a recently established 'nucleosome sliding' assay. The CHRAC complex and recombinant Iswi can mobilize mononucleosomes to change their position on short DNA fragments. Nucleosome reconstitution on a 248 bp fragment derived from the mouse rDNA promoter results in the formation of two main species, which differ in the position of the nucleosome relative to the DNA ends and which can be separated by gel electrophoresis under non-denaturing conditions. Mononucleosomes positioned near the ends and at the center of the DNA fragment were used as substrates for mobilization by Mi-2 and Iswi. To avoid loss of labelled nucleosomes due to formation of stable enzyme-substrate complexes (as in the bandshift assay) the reaction was stopped by addition of excess unlabelled nucleosomes. Recombinant Iswi moves the centrally positioned nucleosome to the ends of the fragment but not vice versa. Remarkably, recombinant Mi-2 behaves in exactly the opposite way: Mi-2 fails to move the central nucleosome but mobilizes the end-positioned nucleosome. Mi-2-promoted nucleosome mobilization is ATP dependent. The effect of removal of individual histone tails on nucleosome mobilization by Mi-2 was also tested. Mi-2 is able to mobilize nucleosomes lacking individual histone N-termini. It is concluded that histone tails are dispensable for stimulation of the Mi-2 ATPase (Brehm, 2000).

The presence of the CHRAC subunits affects the direction of nucleosome sliding by Iswi. Like recombinant Mi-2, the CHRAC complex moves the mononucleosome from the end towards the center of the DNA fragment but fails to mobilize the nucleosome in the opposite direction. In order to investigate whether the presence of Mi-2-associated proteins affects the direction of Mi-2- mediated nucleosome mobilization, immunoprecipitated Mi-2 complexes were tested in the mobilization assay. The alphadMi-2-C antiserum precipitates an activity that mobilizes the end-positioned nucleosome but not the centrally positioned nucleosome. This mobilization activity is ATP dependent. Immunoprecipitated Mi-2 displays the same nucleosome mobilization activity as recombinant Mi-2. Taken together, this analysis of ATPase and nucleosome mobilization activity suggests that the properties of recombinant Mi-2 are fully preserved in a native Mi-2 complex (Brehm, 2000).

Therefore, this comparative analysis has uncovered a number of fundamental differences between these two remodelling enzymes. Both recombinant ATPases can be stimulated by nucleosomes assembled from recombinant histones. Whereas the Mi-2 ATPase does not respond to free DNA, the Iswi ATPase is stimulated by free DNA to some extent. Free DNA also appears to play a role in the interaction of Iswi with the nucleosome: Iswi binds to a nucleosome reconstituted on 248 bp of DNA, which displays free DNA, but not to the core particle consisting entirely of 146 bp of nucleosomal DNA under the stringent conditions of this bandshift assay. In striking contrast, Mi-2 interacts with both nucleosomes equally well, demonstrating that it does not require free DNA for interaction. Taken together, these observations suggest that Iswi, but not Mi-2, recognizes its chromatin substrate in part through an interaction with free DNA (Brehm, 2000).

The ability to promote the movement of a nucleosome along DNA has been demonstrated for the Iswi-containing NURF and CHRAC complexes, recombinant Iswi and the SWI/SNF complex (Hamiche, 1999; Längst, 1999; Whitehouse, 1999). It is conceivable that all remodelling ATPases mobilize nucleosomes in the same way, simply reflecting a common chromatin remodelling activity. Indeed, Iswi and Mi-2 share the intrinsic capacity to promote nucleosome mobilization in an ATP-dependent manner. Surprisingly, however, Iswi and Mi-2 move the nucleosome in opposite directions in these assays. Iswi moves nucleosomes prepositioned at the center to the ends of a DNA fragment but is not able to mobilize end-positioned nucleosomes (Längst, 1999). Mi-2 moves nucleosomes positioned at the end of the DNA fragment to a central position but fails to mobilize the central mononucleosome. Whether this observed directionality of nucleosome movement translates to a regulatory difference in a physiological chromatin context is unclear at present. Nevertheless, the observed difference in the direction of nucleosome movement within the constraints of this assay system suggests that Iswi and Mi-2 interact differently with nucleosomes and that they employ different mechanisms to mobilize them. At present the mechanisms of nucleosome mobilization are not understood. Iswi only moves the central nucleosome, which is flanked on either side by free DNA. Given that free DNA plays a role in substrate recognition by Iswi, it is speculated that Iswi needs to interact with two DNA segments extending from the nucleosome in order to mobilize the nucleosome. This scenario would predict that at least two Iswi molecules simultaneously interact with the central mononucleosome. In agreement with this hypothesis, the formation of multiple Iswi-nucleosome complexes is observed in the bandshift assay, most likely reflecting the binding of multiple Iswi molecules to the 248 bp nucleosome. In contrast, Mi-2 nucleosome binding and ATPase activity is not influenced by free DNA, pointing to a different mechanism of nucleosome mobilization. Interestingly, nucleosome mobilization by Mi-2 in this assay is similar to that of Iswi in the context of the CHRAC complex (Längst, 1999). Identification of the principle that modulates Iswi activity to change the direction of nucleosome mobilization should shed light on the mechanism of nucleosome mobilization by Mi-2 as well (Brehm, 2000).

The ATPase Iswi is the molecular motor of several remodeling factors that trigger nucleosome sliding in vitro. In search for the underlying mechanism, it was found that unilateral binding of Iswi to a model nucleosome correlates with directional movement of the nucleosome toward the enzyme. It has been proposed that Iswi might loosen histone-DNA interactions through twisting DNA. However, nucleosome sliding assays on nicked DNA substrates suggest that propagation of altered twist is not involved. Surprisingly, nicks in the linker DNA in front of the nucleosome facilitate sliding. These data suggest that the rate of nucleosome sliding is limited by a conformational change other than twisting, such as the formation of a short loop, of DNA at the entry into the nucleosome (Langst, 2001).

Nucleosome remodeling factors are complex machines composed of several subunits each contributing molecular interaction surfaces. Not surprisingly, the analysis of substrate recognition and the mechanisms of remodeling are complicated. In a reductionist's approach to analyze the function of Iswi-containing nucleosome remodeling factors, use has been made of an earlier observation that recombinant Iswi (recIswi) has nucleosome remodeling activity: recIswi responds to nucleosomal substrates with increased ATPase activity and is able to induce the sliding of mononucleosomes as well as the spacing of irregular successions of nucleosomes. Although important qualitative and quantitative differences between the recIswi and Iswi in the context of native remodeling factors have been noted, analysis of the recIswi-induced nucleosome remodeling should shed light on the activity of multi-subunit remodeling factors as well (Langst, 2001).

The outcome of Iswi-induced nucleosome sliding depends on the particular underlying DNA sequence. The observation that recIswi triggers unidirectional nucleosome sliding on a highly bent DNA fragment derived from the mouse rDNA promoter allows visualization of the specific interaction of Iswi with the nucleosomal substrate by direct footprinting. recIswi interacts with DNA at one end of the nucleosome, presumably due to particular DNA sequence/structure preferences. Binding to nucleosomal edges is likely to occur in the context of native remodeling machines as well, since interactions of purified CHRAC with DNA at these positions have been observed. DNA segments near the end of a nucleosome are inherently flexible and more accessible to DNA binding factors than more internal segments and therefore lend themselves as initiation sites for nucleosome remodeling reactions. The ATPase subunits of the yeast SWI/SNF and RSC complexes, Swi2/Snf2 and Sth1, have been crosslinked to several segments of nucleosomal DNA, including the ends, indicating that different remodeling factors may approach nucleosomes in similar ways (Langst, 2001).

Visualizing the interaction of Iswi with its nucleosome substrate permits a direct correlation with the directionality of nucleosome movement. Apparently, the nucleosome moves toward the remodeling factor. Inevitably, this means that a segment of linker DNA has to become histone-associated during the remodeling reaction. Therefore, ATP-dependent nucleosome remodeling may involve 'pushing' of DNA into the nucleosomal realm. It is likely that this reaction involves contacts of the remodeling factor not only with DNA but also with the histone moiety. Such contacts have yet to be defined; however, a histone tail requirement for nucleosome recognition by Iswi has been noted. Conformational changes of the remodeling factor driven by the ATPase/nucleotide exchange cycle may result in a relative displacement of DNA-histone contacts (Langst, 2001).

In principle, displacement of DNA relative to the histone surface could be achieved through two distinct mechanisms: the DNA could be 'screwed' over the surface driven by altered twist, or it could be translationally shifted by local looping or 'bulging'. A number of remodeling factors, including recIswi, are able to alter the superhelical torsion of linear DNA, and it has been suggested that this phenomenon could be brought about by changes in DNA twist. Twist diffusion, a prime candidate mechanism for nucleosome mobility, assumes the accumulation of torsional strain in DNA between the remodeling factor and the nucleosome. Since an interaction of recIswi at the end of nucleosomal DNA has been demonstrated, the segment within which such strain can be built up must be fairly small. Single strand DNA breaks prevent the accumulation of torsional strain in free DNA, since they allow rotation of the bases around the phosphodiester backbone. Introducing single nicks at any position in the nucleosome and in adjacent linker DNA does not impair nucleosome mobility. Although it may be argued that a combination of base stacking forces and histone interactions may impair the flexibility of nucleosomal DNA to release torsional strain at backbone breaks, nucleosomal DNA is known to be remarkably flexible, allowing even restriction enzymes to transiently gain access. Importantly, DNA is less constrained at the nucleosomal entry sites and in the linker, where torsional strain would be predicted to accumulate if nucleosome mobility involved DNA twisting. In any event, the twisting model assumes that DNA is sufficiently flexible to allow displacement of DNA segments by rotation (Langst, 2001).

The strongest argument against a model involving altered twist is the observation that a single nick at a strategic position facilitates nucleosome sliding. It is proposed that an initial DNA distortion, such as formation of a tight loop or bulge, is rate-limiting nucleosome sliding and that this distortion is facilitated if the DNA is nicked. The current model assumes that Iswi contacts the linker DNA as well as a second anchor point on the nucleosome. A conformational change leads to the formation of a local DNA loop as histone-DNA interactions are displaced by analogous ones involving linker DNA. In a subsequent step, the bulge is 'released' to propagate over the nucleosome surface. It has been proposed that the repositioning of nucleosomes without dissociation from DNA can be understood through the diffusional motion of intranucleosomal loops. The histone H4 N terminus, which is crucial for Iswi ATPase and remodeling activities, may allosterically affect the conformational transition of the enzyme. According to the model, the step length of nucleosome movement essentially depends on the dimensions, the 'molecular ruler' of the enzyme. In the reductionist assay involving DNA fragments, Iswi will continue to move linker DNA 'into the nucleosome' until, due to constraints imposed by the overall geometry of Iswi-nucleosome interaction, it is unable to grab another DNA segment. The fact that the nucleosome has been observed 'hanging over' the fragment end for some 20 bp demonstrates that Iswi is able to 'pull' essentially all linker DNA into the nucleosome. In nucleosomal arrays, neighboring nucleosomes and linker histones are likely to restrict the action of the remodeling machinery (Langst, 2001).

It is possible that local loop formation or its propagation over the octamer surface generates torsion as a secondary consequence of nucleosome mobility, which may explain the observed extrusion of hairpins. The action of the SWI/SNF-like BAF complex induces a B to Z transition of DNA structure at the CSF1 promoter in vivo and in vitro, again pointing to major distortions of DNA structure during remodeling that may destabilize nucleosomes (Langst, 2001).

Alternatively, Iswi is able to twist DNA but this activity plays no role in nucleosome sliding. It has been reported that topological constraints can restrict the remodeling activity of the SWI/SNF complex. The tight constraints of a minicircle would presumably not only prevent an untwisting of DNA but would similarily affect the ability to form a 'DNA bulge' or 'local loop.' While these data clearly indicate that DNA must be flexible for some kind of deformation during nucleosome remodeling, they do not permit distinguishing between twisting or 'looping' models (Langst, 2001).

In the context of ACF and CHRAC, Iswi moves the nucleosome to the center of the rDNA fragment. The reasons for this altered directionality of movement are unclear, but preliminary results indicate that under these circumstances, interaction of the remodeling machinery with both sides of the nucleosome can be detected. Future studies will be directed at determining whether additional contacts with the nucleosome are contributed by Acf1 or whether symmetrical interactions are due to more than one Iswi moiety in native remodeling factors (Langst, 2001).

The ATPase Iswi can be considered the catalytic core of several multiprotein nucleosome remodeling machines. Alone or in the context of nucleosome remodeling factor [the chromatin accessibility complex (CHRAC), or ACF] Iswi catalyzes a number of ATP-dependent transitions of chromatin structure that are currently best explained by its ability to induce nucleosome sliding. In addition, Iswi can function as a nucleosome spacing factor during chromatin assembly, where it will trigger the ordering of newly assembled nucleosomes into regular arrays. Both nucleosome remodeling and nucleosome spacing reactions are mechanistically unexplained. As a step toward defining the interaction of Iswi with its substrate during nucleosome remodeling and chromatin assembly a set of nucleosomes lacking individual histone N termini were generated from recombinant histones. The conserved N termini (the N-terminal tails) of histone H4 were found to be essential to stimulate Iswi ATPase activity, in contrast to other histone tails. Remarkably, the H4 N terminus, but none of the other tails, is critical for CHRAC-induced nucleosome sliding and for the generation of regularity in nucleosomal arrays by Iswi. Direct nucleosome binding studies did not reflect a dependence on the H4 tail for Iswi-nucleosome interactions. It is concluded that the H4 tail is critically required for nucleosome remodeling and spacing at a step subsequent to interaction with the substrate (Clapier, 2001).

Nucleosome Remodeling Factor (NURF) is an ATP-dependent nucleosome remodeling complex that alters chromatin structure by catalyzing nucleosome sliding, thereby exposing DNA sequences previously associated with nucleosomes. How the unstructured N-terminal residues of core histones (the N-terminal histone tails) influence nucleosome sliding has been systematically studied. Bacterially expressed Drosophila histones were used to reconstitute hybrid nucleosomes lacking one or more histone N-terminal tails. Unexpectedly, it was found that removal of the N-terminal tail of histone H2B promote uncatalyzed nucleosome sliding during native gel electrophoresis. Uncatalyzed nucleosome mobility is enhanced by additional removal of other histone tails but is not affected by hyperacetylation of core histones by p300. In addition, the N-terminal tail of the histone H4 is specifically required for ATP-dependent catalysis of nucleosome sliding by NURF. Alanine scanning mutagenesis has demonstrated that H4 residues 16-KRHR-19 are critical for the induction of nucleosome mobility, revealing a histone tail motif that regulates NURF activity. An exchange of histone tails between H4 and H3 impairs NURF-induced sliding of the mutant nucleosome, indicating that the location of the KRHR motif in relation to global nucleosome structure is functionally important. These results provide functions for the N-terminal histone tails in regulating the mobility of nucleosomes (Hamiche, 2001).

To examine how the histone N-terminal tails influence NURF-mediated nucleosome sliding, the ability of tailless nucleosomes to stimulate the ATPase activity of NURF was examined. Incubation of NURF with nucleosomes lacking histone H2A, H2B, or H3 tails individually or lacking both H2A and H2B tails shows simulation of ATPase activity like that generated by WT nucleosomes. By contrast, nucleosomes lacking the histone H4 tail alone or combined with removal of other histone tails show no stimulation of the ATPase activity of NURF. The effects of tailless nucleosomes on nucleosome sliding by NURF was investigated. For WT nucleosomes and nucleosomes lacking the histone H2A, H2B, or H3 tail, NURF-mediated nucleosome mobility similar to that reported for nucleosomes containing native full length histones was observed. Mononucleosomes reconstituted from full length recombinant Drosophila histones adopt four major positions (N1-N4) on the 359-bp hsp70 promoter when analyzed by native gel electrophoresis at low ionic strength. Nucleosomes N1 and N2 were depleted, whereas N3 was enriched, along with slower migration of the N4 'fragment end nucleosomes.' In contrast, nucleosomes lacking the H4 tail show little mobilization on reaction with NURF. N1 and N2 positions were retained, whereas N3 shows little enrichment (but the N4 nucleosomes show clear changes in migration). Taken together, the results demonstrate that the N-terminal tail of histone H4 regulates the activity of NURF (Hamiche, 2001).

An examination was carried out to see whether a glutathione S-transferase (GST)-histone H4 tail fusion could block nucleosome stimulation of the ATPase activity of NURF in trans. GST-H4 tail reduces stimulation of the ATPase activity of NURF by only 2-fold when introduced at 100-fold molar excess to nucleosomes. The ability of NURF to catalyze nucleosome sliding was not detectably affected by a 100-fold molar excess of GST-H4 tail, and higher concentrations of GST-H4 tail failed to show inhibition. It appears that the H4 tail is unable to influence NURF-induced nucleosome sliding when removed from the context of the nucleosome (Hamiche, 2001).

The x-ray crystal structure of the nucleosome core particle shows that the N-terminal tails of histones H3 and H2B exit through the aligned minor grooves of adjacent gyres of the DNA superhelix, whereas H4 and H2A tails exit in minor grooves from the top or bottom edges of the disc-like particle. Beyond the nucleosome core particle, the histone tails are disordered, having no visible interactions with the 147-bp DNA superhelix. Biophysical studies of nucleosome arrays in which the histone tails are removed by trypsinization or modified by acetylation indicate their involvement in the higher-order folding of chromatin. Moreover, contacts can be observed between the tails and the histone octamer of neighboring core particles in the x-ray crystal structure. However, there is evidence that the N-terminal histone tails also interact with core particle DNA in solution. The absence of histone N-terminal tails decreases the thermostability of the nucleosome and alters the equilibrium constants for dynamic DNA site accessibility in nucleosomes. Binding of sequence-specific transcription factors to the nucleosome is modulated by the presence of the histone tails, and photochemical and UV-laser crosslinking experiments reveal physical interactions between core histone tails and nucleosomal DNA. The histone tails not only make contact with DNA in the nucleosome core particle but also can preferentially interact with linker DNA. Stabilization of nucleosomes by histone tails is apparently effective only on intrinsically straight or bent, rather than flexible, DNA fragments (Hamiche, 2001 and references therein).

This study provides an additional perspective for the histone N-terminal tails. Deletion of the N-terminal tail of histone H2B promotes uncatalyzed nucleosome mobility when perturbed by native gel electrophoresis in Tris·glycine·EDTA (or Tris·borate·EDTA), and this effect is increased by deletion of the other histone tails. The observed changes in electrophoretic migration are likely to be caused by increased translational mobility of the histone octamer on DNA, although the additional possibility of increased conformational flexibility of the linker DNA is not excluded. Thus, uncatalyzed nucleosome positioning and mobility not only may depend on structured histone-DNA interactions in the nucleosome core particle but also could be modulated by interactions of the histone H2B tail and other histone tails with nucleosomal DNA. The proximity of the basic histone H2B tails to two adjacent DNA gyres of the nucleosome core particle may provide especially suitable interactions that restrict nucleosome mobility. In this respect, it is intriguing to recall genetic studies in which deletion of the first 20 amino acids of the H2B N-terminal tail bypassed the requirement for Swi-Snf in yeast (Hamiche, 2001).

It is of interest that quantitative hyperacetylation of core histones by p300 has no detectable effect on nucleosome positioning or nucleosome dynamics. These results, which suggest that histone acetylation by p300 has a significantly greater impact on higher-order folding of nucleosome arrays than on the positioning and mobility (or stability) of individual nucleosomes, concur with other findings of a similar nature. It is also noted that deletion of the histone H3 tail produces a slight retardation in electrophoretic migration irrespective of nucleosome positioning, raising the possibility that this histone tail affects the entry-exit angle of the linker DNA (Hamiche, 2001.

This study demonstrates the importance of the histone H4 tail in ATP-dependent nucleosome sliding catalyzed by NURF. This finding concurs with results showing that the histone H4 tail is required for induction of nucleosome sliding by the CHRAC chromatin remodeling complex and for stimulation of the ATPase activity of recombinant Iswi. Given that Iswi is a common component of NURF and CHRAC, it is likely that interactions between the H4 tail and Iswi are important for activating NURF. Full efficiency and positional specificity of nucleosome sliding require the participation of the largest NURF subunit, NURF301 (Hamiche, 2001).

The definition by alanine-scanning mutagenesis of histone H4 tail residues responsible for regulating NURF activity reveals that H4 tail residues 16-KRHR-19 are critical for the induction of ATP-dependent nucleosome sliding. The proper spatial location of this regulatory motif relative to the global structure of the nucleosome is also important, because interchanging the tails of H3 and H4 impairs nucleosome sliding by NURF. These findings, taken with the failure of a GST-H4 tail fusion protein to significantly inhibit NURF function, suggest that NURF probably interacts with H4 tail residues 16-KRHR-19 in complex with nucleosomal DNA. There is evidence that part of the N-terminal tails of histone H4 (and H3) can be organized in the nucleosome as DNA-bound polypeptide segments with alpha-helical character. It will be interesting to investigate the nature of the regulatory interaction between H4 16-KRHR-19 and NURF (Iswi). Given that H4 K16 and K20 are known sites of histone acetylation and methylation, respectively, it is possible that these modifications could influence the activities of NURF and other Iswi complexes (Hamiche, 2001.

Aside from the evident importance of histone H4 16-KRHR-19 in providing a key to the ATP-dependent catalytic activity of NURF, the involvement of the other core histone tails in catalyzed nucleosome sliding is unclear. Deletion of the histone H2B tail does not bypass the need for NURF to induce ATP-dependent nucleosome sliding under in vitro assay conditions, indicating that DNA-protein interactions within the nucleosome core particle are dominant. An attractive model for nucleosome sliding invokes the ATP-dependent induction and propagation of a DNA twist or bulge over the histone octamer, a process that necessitates the transient disruption of contacts between structured histone elements and core particle DNA. However, it is possible that the dissociation of the N-terminal tail of histone H2B and the other histone tails from nucleosomal DNA may facilitate the overall nucleosome sliding mechanism (Hamiche, 2001).

The ISWI proteins form the catalytic core of a subset of ATP-dependent chromatin-remodeling activities. The interaction of the Drosophila ISWI protein with nucleosomal substrates has been studied. The ability of nucleic acids to bind and stimulate the ATPase activity of ISWI depends on length. ISWI is able to displace triplex-forming oligonucleotides efficiently when they are introduced at sites close to a nucleosome but successively less efficiently 30 to 60 bp from its edge. The ability of ISWI to direct triplex displacement was specifically impeded by the introduction of 5- or 10-bp gaps in the 3'-5' strand between the triplex and the nucleosome. In combination, these observations suggest that ISWI is a 3'-5'-strand-specific, ATP-dependent DNA translocase that may be capable of forcing DNA over the surface of nucleosomes (Whitehouse, 2003).

TRF2 associates with DREF and Iswi and regulates the PCNA promoter

Drosophila TATA-box-binding protein (TBP)-related factor 2 (TRF2) is a member of a family of TBP-related factors present in metazoan organisms. Recent evidence suggests that TRF2s are required for proper embryonic development and differentiation. However, true target promoters and the mechanisms by which TRF2 operates to control transcription remain elusive. A Drosophila TRF2-containing complex has been purified by antibody affinity; this complex contains components of the nucleosome remodelling factor (NURF) chromatin remodelling complex as well as the DNA replication-related element (DRE)-binding factor DREF. This latter finding leads to potential target genes containing TRF2-responsive promoters. A combination of in vitro and in vivo assays has been used to show that the DREF-containing TRF2 complex directs core promoter recognition of the proliferating cell nuclear antigen (PCNA) gene. Additional TRF2-responsive target genes involved in DNA replication and cell proliferation have also been identified. These data suggest that TRF2 functions as a core promoter-selectivity factor responsible for coordinating transcription of a subset of genes in Drosophila (Hochheimer, 2002).

Metazoan organisms have evolved diverse mechanisms to control the spatial and temporal patterns of gene expression during growth, differentiation and development. It has become increasingly evident that cell-type-specific components of the general transcriptional apparatus, for example the mammalian TFIID component TAFII105 or the Drosophila TAFII80 homolog Cannonball contribute significantly to tissue-specific and gene-selective transcriptional regulation in metazoan organisms. Recent studies have also established that TBP-related factors like Drosophila TRF1 can direct transcription from an alternative core promoter and a TRF1:BRF complex is required for RNA polymerase III transcription of transfer RNA genes. TRF2 is a third member of the TBP family in Drosophila and, like TBP and TRF1, TRF2 interacts with the basal transcription factors TFIIA and TFIIB10. However, the primary amino acid sequence of the putative TRF2 DNA-binding domain has diverged from TBP and TRF1 and, not surprisingly, TRF2 fails to bind to DNA containing canonical TATA boxes. But TRF2 is associated with loci on Drosophila chromosomes that are distinct from TBP and TRF1. This suggested that TRF2 may direct promoter specificity and perhaps coordinate a subset of target genes. Size exclusion chromatography indicated that Drosophila TRF2 is likely to be part of a macromolecular complex. Unlike TRF1, Drosophila TRF2 has amino-terminal and carboxy-terminal extensions flanking the putative DNA-binding core domain. This suggested that TRF2 may be associated with a set of proteins that are distinct from TBP- and TRF1-associated factors. It was reasoned that the purification and identification of TRF2-associated factors might enable the identification of TRF2-specific promoters and reveal how TRF2 operates to execute transcriptional specificity (Hochheimer, 2002).

In the absence of a functional assay allowing conventional purification of TRF2-associated factors a panel of monoclonal antibodies was generated directed against several domains of TRF2 to affinity purify TRF2 and its putative associated subunits. Approximately 3,500 hybridomas were screened and a clone was isolated that efficiently immunoprecipitated TRF2 and its associated factors from Drosophila embryo nuclear extract. A Sarkosyl-eluted complex containing TRF2 was analysed by SDS-polyacrylamide gradient gel electrophoresis (PAGE) that revealed a set of 18 polypeptides with relative molecular mass (Mr) ranging from 300K to 29K that co-immunoprecipitated consistently with TRF2 even under very stringent conditions (Hochheimer, 2002).

An 80K protein associated with TRF2 is identical to DREF1. DREF and its corresponding response element DRE have been well documented to be important for the regulation of cell-cycle and cell-proliferation genes in Drosophila (that is, genes for PCNA and the 180K and 73K subunits of DNA polymerase). The identification of the promoter-selective DNA-binding protein DREF was intriguing because Drosophila TRF2 thus far had failed to bind to canonical TATA-box elements, which suggests that TRF2 may cooperate with DREF to execute promoter specificity and perhaps operate like a metazoan sigma factor (Hochheimer, 2002).

The 140K protein associated with TRF2 is identical to Drosophila Iswi, which is the catalytic ATPase subunit of NURF, ACF and CHRAC chromatin remodelling complexes. Moreover, the 55K and 38K proteins associated with TRF2 turned out to be NURF-55/CAF-1 and NURF-38/inorganic pyrophosphatase, respectively. Notably, the peptide sequences obtained from the three largest (300K, 250K and 230K) proteins associated with TRF2 do not match NURF-301, suggesting that the presence of some NURF subunits is not merely a result of contaminating NURF in the TRF2 complex. However, analysis of the cDNAs encoding the 190K and 160K proteins associated with TRF2 revealed that both proteins contain conserved sequence motifs for 11 and 5 zinc finger motifs (C2H2), respectively; these smaller proteins thus resemble factors like the CCCTC-binding factor CTCF that has been implicated in mediating chromatin-dependent processes such as the regulation of insulator function. It is therefore possible that the TRF2 complex encompasses both promoter-selectivity functions and NURF-like components as well as other activities with distinct subunits and specificity (Hochheimer, 2002).

Sequence analysis of the cDNA encoding the 65K protein revealed a significant similarity to the RNA-binding protein Rap55 isolated from Pleurodeles waltl and Xenopus laevis, whereas sequence analysis of the 70K, 116K and 118K proteins showed no significant similarity to known proteins in the databases. The functional relevance of ß-tubulin in the TRF2 complex is at present unclear and the 47K and the 29K polypeptides associated with TRF2 are yet to be characterized (Hochheimer, 2002).

Having identified DREF as a tightly associated component of the TRF2 complex, it was next asked whether TRF2 can function as a true core promoter recognition factor and selectively initiate transcription at a promoter that is documented to be stimulated by the DRE/DREF system. The DREF-responsive PCNA promoter, which contains at least three promoter-proximal regulatory elements including an upstream regulatory element URE, DRE and two E2F recognition sites located within 200 bp upstream of the start site, was chosen (Hochheimer, 2002).

To test the responsiveness of the PCNA promoter in vitro and to map the transcription start site(s), a -580 PCNA (-580 to +56) promoter fragment, which contains all known regulatory elements, and a -64 PCNA (-64 to +56) promoter fragment, which lacks all regulatory elements except for the E2F-binding sites were used as DNA templates for in vitro transcription. Increasing amounts of a partially purified Drosophila embryo nuclear extract (H.4) that contains all the necessary basal factors were added, as well as both the TRF2 complex and limiting amounts of TFIID to the transcription reaction. Using the -580 PCNA template two distinct transcription start sites separated by 63 nucleotides were detected. Promoter 1 (with start site at position +1) was stimulated with increasing amounts of H.4 supplemented with TFIID whereas promoter 2 (with start site at position -63) was detected only with the lowest amounts of H.4 added. Using the truncated -64 PCNA, template, transcription from promoter 2 was essentially abolished, whereas a weak activity could be detected from promoter 1 by adding the maximum amount of H.4 + TFIID. In vitro and in vivo results suggest that promoter 2 might be TRF2- and DRE-dependent, whereas promoter 1 appears to be mediated by TFIID (Hochheimer, 2002).

It was next asked whether TRF2 can contribute to the enhancer-dependent activated transcription of the PCNA promoters by E2F and DP, which cooperate in DNA-binding and transcriptional activation. The co-expression of just the trancriptional activators E2F and DP in the absence of exogenous TRF2/DREF results in a substantial transcriptional activation of the PCNA reporter. It is likely that this activation by E2F/DP is mediated by endogenous TRF2. This activation is abolished with the -64 PCNA reporter, which lacks the DRE-binding sites but still contains the E2F-binding sites. As expected, inducing the co-expression of all three promoter recognition factors -- TRF2, DREF and E2F -- results in a strong synergistic activation of the PCNA promoters (80-fold) in a DRE-dependent fashion. These results suggest that in SL2 cells TRF2 and DREF can work together to stimulate the PCNA reporter in a DRE-dependent fashion. This is consistent with the finding in vitro that the TRF2 complex can selectively initiate transcription from promoter 2 of the PCNA gene in a DRE-dependent manner (Hochheimer, 2002).

To investigate whether TRF2 is involved in the coordinate regulation of other DNA replication and cell cycle genes in the Drosophila genome, oligonucleotide-based microarrays representing 13,500 Drosophila genes were hybridized with RNA probes isolated at different time points after induction of TRF2 expression in SL2 cells. The microarray analysis revealed that only 1.9% of all genes analysed were upregulated more than 2-fold and only 1.6% downregulated by more than 2-fold. These biochemical studies and cell based assays suggested that TRF2 functions as a core promoter-selectivity factor that collaborates with DREF. It was therefore asked whether there are other TRF2-responsive genes that also contain a promoter-proximal DRE. An analysis of the distribution of DREs in the Drosophila genome revealed that about 100 genes bear a consensus DRE within 1 kb of the predicted promoter region. Microarray analysis revealed that 38 of these DRE-containing genes were also responsive to TRF2 overexpression. For example, genes encoding PCNA, the 180K subunit of DNA polymerase, the a-subunit of mitochondrial DNA polymerase, and E2F were all found to be upregulated 2-5-fold by TRF2 in the microarray analysis and confirmed by RNase protection assays. Three additional DRE-regulated genes encoding TBP, the 73K subunit of DNA polymerase, and the 50K subunit of DNA polymerase were found to be downregulated (Hochheimer, 2002).

These data suggest that in addition to the PCNA gene a number of other Drosophila DRE-containing genes may also be regulated by TRF2 and further support the model that TRF2 can function as a core promoter-selectivity factor that governs a restricted subset of genes that are coordinately regulated. A recent bioinformatics study of core promoter sequences in the Drosophila genome identified a consensus DRE as the second most frequent control core element (other than TATA and INR) providing independent evidence for DRE as a core promoter element (Hochheimer, 2002).

Because these studies have relied largely on 'gain of function' assays double stranded RNA interference (dsRNAi) was imployed in flies and SL2 cells25 to determine the consequences of ablating TRF2/DREF on transcription of putative target genes such as PCNA and DNApol 180 in cultured cells. SL2 cells were treated with in vitro synthesized dsRNA and the depletion of TRF2 and DREF proteins was monitored by immunoblot analysis. After 2 days of incubation the specifically targeted TRF2 and DREF proteins were severely depleted. After 48 h of dsRNA treatment, a significant number of cells sloughed off the plate and died. This is in accordance with previous finding that TRF2 RNAi in Drosophila embryos is lethal for embryos (Hochheimer, 2002).

However, between 24 and 48 h of dsRNAi treatment it was possible to reproducably measure the activity of transiently transfected luciferase reporters fused to either the PCNA or DNApol 180 promoters and compare them to an internal control reporter gene. The activities from both the PCNA (-580 to +56) and DNApol 180 (-620 to +20) promoters were significantly reduced in TRF2-depleted cells (4.2-fold and 3-fold, respectively) relative to a Renilla luciferase control reporter driven by the HSV TK promoter. Likewise, these two target gene promoters were downregulated in DREF-depleted cells (3.5-fold and 5-fold, respectively). These depletion experiments using dsRNAi thus support the findings that TRF2 and DREF participate in directing transcription of a select subset of genes that include PCNA and DNApol 180 (Hochheimer, 2002).

Although the dsRNAi studies provide an independent line of evidence to support the notion that TRF2/DREF play a role in promoter selectivity in vivo, they fail to provide a direct mechanistic link between TRF2/DREF and the PCNA and DNApol 180 promoters. Chromatin immunoprecipitation (ChIP) experiments were carried to determine the occupancy of TRF2/DREF at the PCNA and DNApol 180 promoters in formaldehyde-treated SL2 cells using antibodies raised against TRF2, DREF and TBP. The precipitated DNA fragments with an average length of 500-1,000 base pairs (bp) were analysed directly by polymerase chain reaction (PCR). The PCNA promoter region was specifically precipitated by anti-TRF2 and anti-DREF and to a lesser extent by anti-TBP; this is consistent with previous findings and confirms that TRF2, DREF and the TFIID subunit TBP can co-localize and directly interact with the PCNA promoter region in living cells. As further evidence for the targeting of specific promoters by TRF2 and DREF, an analysis was carried out of the DRE-containing DNApol 180 promoter region, which is selectively precipitated by the TRF2-, DREF- and TBP-specific antibodies. These ChIP experiments strongly support the in vitro transcription results as well as the cell-based dsRNAi transcription assays in establishing a core promoter selectivity for the TRF2/DREF complex (Hochheimer, 2002).

A critical epitope for substrate recognition by the nucleosome remodeling ATPase ISWI

The ATPase ISWI is the catalytic core of several nucleosome remodeling complexes that are able to alter histone-DNA interactions within nucleosomes such that the sliding of histone octamers on DNA is facilitated. Dynamic nucleosome repositioning may be involved in the assembly of chromatin with regularly spaced nucleosomes and accessible regulatory sequence elements. The mechanism that underlies nucleosome sliding is largely unresolved. The N-terminal 'tail' of histone H4 is critical for nucleosome remodeling by ISWI. If deleted, nucleosomes are no longer recognized as substrates and do not stimulate the ATPase activity of ISWI. The H4 tail is part of a more complex recognition epitope, which is destroyed by grafting the H4 N-terminus onto other histones. The H4 tail requirement has been mapped to a hydrophilic patch consisting of the amino acids R17H18R19 localized at the base of the tail. These residues have been shown to contact nucleosomal DNA, suggesting that ISWI recognizes an 'epitope' consisting of the DNA-bound H4 tail. Consistent with this hypothesis, the ISWI ATPase is stimulated by isolated H4 tail peptides ISWI only in the presence of DNA. Acetylation of the adjacent K12 and K16 residues impairs substrate recognition by ISWI (Clapier, 2002).

This study highlights the importance of the H4 N-terminal residues R17H18R19 for substrate recognition by the nucleosome remodeling ATPase ISWI. Out of context, either as isolated tail peptides or grafted onto histone H3 or H2A, H4 tails containing this sequence element will not trigger the ATPase activity of ISWI. Position-dependent functions of the H4 tail have been observed in yeast. While nucleosome assembly and structure per se is unaffected by swapping the tail domains, the regulation of specific loci through chromatin, however, requires the H4 tail at its native position. The data are consistent with an independent analysis of the histone tail requirements for nucleosome sliding by NURF (Clapier, 2002).

The critical amino acids are part of a larger, basic patch, K16R17H18R19K20, where the tail emerges from the compact globular histone domains. Early crosslinking experiments demonstrated a direct contact of histidine 18 with nucleosomal DNA. The finding that combining H4 tail peptides with DNA (but with neither of the components alone), elicits an ATPase response in ISWI that strongly supports the idea that some part of the H4 tail associates with DNA, generating an 'epitope' consisting of DNA and protein that triggers the ATPase activity of ISWI. The fact that mutagenesis of arginines 17 and 19 to alanines abolishes this effect may point to a critical role for these residues in DNA binding. However, the alternative possibility that these arginines are directly recognized by ISWI cannot be excluded (Clapier, 2002).

The crystal structure of the nucleosome yields little further insight. One H4 tail is not at all visible in the structure while the second one is ordered by extensive interactions with an acidic patch on the H2A-H2B dimer surface of an adjacent nucleosome. Since both tails are not equivalent in the crystal and the localization of the visible tail involves interactions that may be dictated by crystal packing forces (mutations in this region prevent crystallization), it is possible that the interaction with a neighboring particle leads to an artificial 'lifting' of the H4 tail base off the DNA (Clapier, 2002).

It has been suggested that interaction of the H4 tail with the minor groove of DNA may lead to the induction or stabilization of an alpha-helical conformation of the part in question. On the basis of elegant in vivo experiments using the yeast model, Ling (1996) proposed that residues 16-19 of the H4 tail might adopt an alpha-helical structure upon charge neutralization by an acidic protein or DNA. Later, Baneres (1997) concluded on the basis of CD spectroscopy and secondary structure prediction that the H4 tail sequences between 16 and 24 are in an alpha-helical conformation upon interaction with DNA. Residues 16-20, which contain the ISWI recognition epitope, are important for repression of the yeast mating type loci. In a search for extragenic suppressors of mutations in these residues SIR3 has been identifed. Direct interaction of SIR3 with this site within the H4 tail induces the heterochromatic state that characterizes the silent mating type cassettes. However, SIR3 (and its companion SIR4) are apparently able to interact with the H4 tail in the absence of DNA, although the formation of an alpha-helix in the critical region has been assumed in order to explain the results of the mutagenesis study. In contrast, the current study has been unable to identify a stable interaction between the GST-H4 tail fusion protein and ISWI in the absence of DNA (Clapier, 2002).

Several mechanisms, not mutually exclusive, may contribute to repressive function of the DNA-bound basic patch in the H4 N-terminus. Interaction of SIR3/SIR4 proteins may lead to an inaccessible higher order folding of chromatin. At the same time, binding of these proteins may stabilize the contact of the basic patch with DNA, which may restrict the flexibility of the nucleosomal fiber. It is also conceivable that the interaction of ISWI-containing nucleosome remodeling factors with the tail may be modulated by SIR protein binding (Clapier, 2002).

Most interestingly, the yeast homolog of the ATPase ISWI has recently been shown to be involved in the repression of many genes, including some that govern the meiotic programme, apparently through positioning nucleosomes in ways that are incompatible with transcription. The histone tail requirements for nucleosome remodeling by the ISW1 and ISW2 complexes and their resident ATPases have not yet been determined. If the mechanism of action of the ISWI homologs have been as conserved as the H4 tails, the deletion study would predict that the regulatory events that have been ascribed to yeast ISWI function would be compromised in strains that harbor a deletion/mutation of the critical H4 tail residues (Clapier, 2002).

In yeast and higher eukaryotes, active loci are marked by particular histone acetylation patterns whereas repressed domains are generally characterized by hypoacetylation. The reconstitution of a minimal epitope able to stimulate ISWI ATPase activity allowed addressing the question of whether acetylation of specific lysine residues within the H4 N-terminus affect ISWI function. Acetylation of those lysines that reside closest to the ISWI recognition epitope, lysines 16 and 12, interfere significantly with ISWI function. Interestingly, systematic mutational analysis of single amino acids in the H4 N-terminus in yeast has suggested that acetylation of lysine 16, but not of any of the other lysines in the H4 tail, interferes with silencing of mating type loci. Acetylation of lysine 16 is also a hallmark of the hyperactive, partially decondensed male X chromosome in Drosophila. The recent observation that the higher order chromosome structure of the male X chromosome is specifically disrupted in an ISWI homozygous mutant larvae, suggested that ISWI-containing nucleosome remodeling complexes counteract the effect of histone acetylation on higher order folding of chromatin, presumably through their nucleosome assembly/spacing functions. Such a scenario is reminiscent of the repressive functions of ISWI-containing complexes in yeast (Clapier, 2002).

Why is the activity of ISWI so sensitive to deletion and modification of the recognition epitope? Early crosslinking studies in nuclei mapped the contact of histidine 18 to a region of DNA 1.5 helical turns off the dyad axis of the nucleosome. This site is likely to be in reach for ISWI bound to DNA at the exit point from the nucleosome, suggesting that a direct interaction of ISWI may be possible. It is difficult to formulate hypotheses as to how this contact stimulates the ATPase activity of ISWI, since it is not known at which step during the nucleosome remodeling cycle ATP is hydrolyzed. ATP-dependent changes in ISWI conformation may be modulated by contact with the DNA-bound H4 motif. Conversely, disruption of the histone-DNA interactions at that site due to the remodeling process may feed back on the activity of the remodeling enzyme itself (Clapier, 2002).

Genetic identification of a network of factors that functionally interact with the nucleosome remodeling ATPase ISWI

Nucleosome remodeling and covalent modifications of histones play fundamental roles in chromatin structure and function. However, much remains to be learned about how the action of ATP-dependent chromatin remodeling factors and histone-modifying enzymes is coordinated to modulate chromatin organization and transcription. The evolutionarily conserved ATP-dependent chromatin-remodeling factor ISWI plays essential roles in chromosome organization, DNA replication, and transcription regulation. To gain insight into regulation and mechanism of action of ISWI, an unbiased genetic screen was conducted to identify factors with which it interacts in vivo. It was found that ISWI interacts with a network of factors that escaped detection in previous biochemical analyses, including the Sin3A gene. The Sin3A protein and the histone deacetylase Rpd3 are part of a conserved histone deacetylase complex involved in transcriptional repression. ISWI and the Sin3A/Rpd3 complex co-localize at specific chromosome domains. Loss of ISWI activity causes a reduction in the binding of the Sin3A/Rpd3 complex to chromatin. Biochemical analysis showed that the ISWI physically interacts with the histone deacetylase activity of the Sin3A/Rpd3 complex. Consistent with these findings, the acetylation of histone H4 is altered when ISWI activity is perturbed in vivo. These findings suggest that ISWI associates with the Sin3A/Rpd3 complex to support its function in vivo (Burgio, 2008).

This study involved an unbiased genetic screen for regulators of ISWI function in Drosophila. A screen produced the first genetic interaction map for the ATP-dependent chromatin remodeler ISWI in higher eukaryotes. Misexpression of dominant-negative alleles of chromatin-remodeling enzymes in the eye-antennal disc can compromise eye development, often causing roughness and/or reduced eye size. A single K159R amino acid substitution in Drosophila ISWI (ISWIK159R) eliminates its ATPase activity, without affecting the ability of the mutant protein to be incorporated into native complexes. The expression of a UAS-ISWIK159R transgene in the developing eye, using an ey-GAL4 driver, has strong effects on cell viability and chromosome organization and results in flies with rough and reduced eyes. It was reasoned that mutations that enhance or suppress phenotypes resulting from the expression of ISWIK159R can be used to define genes involved in the same biological process as ISWI. This approach has been successfully used to conduct a genetic screen for modifiers of phenotypes caused by loss of the chromatin-remodeling factor brm. It was found that ISWI genetically interacts with a network of cellular and nuclear factors that escaped previous biochemical analyses, indicating the participation of ISWI in variety of biological processes (Burgio, 2008).

Interestingly, unbiased genetic screens aimed at the identification of factors involved in the regulation of vulval cell fates in C.elegans and sensory neuron morphogenesis in Drosophila have identified ISWI and some of the ISWIK159R enhancers as key regulators of these biological processes (Burgio, 2008).

GO analysis indicates 'neuron differentiation' and 'cell cycle regulation' as overrepresented categories within the combined strong and medium ISWIK159R enhancers. With hindsight this result is not surprising considering that the screen targeted the eye, an organ whose development is tightly linked to nervous system differentiation and the spatial as well as temporal control of cell division. Therefore, it is likely that some of the ISWIK159R enhancers isolated could work in concert with ISWI to support the differentiation and development of the adult fly eye (Burgio, 2008).

One of the goals of this screen was to isolate factors encoding enzymatic activities that could play a role in the regulation of ISWI in vivo by modifying ISWI or chromatin components with which ISWI interacts. As expected, the screen led to the isolation of a group of genes that includes kinases (e.g. trbl, grp, snf4ag), ATPases (e.g., pont), proteins associated with deacetylases (Sin3A), methyl binding factor (mbf1) and enzymes regulating the metabolism of poly-ADP-ribose (Parp). The variety of chromatin components found in the screen indicates that it is likely that a functional cross talk exists between ISWI and other chromatin-remodeling and modifying activities working in the nucleus (Burgio, 2008).

It was found that Drosophila ISWI genetically interacts with Sin3A and with its associated histone deacetylase subunit Rpd3. This genetic interaction may reflect a physical interaction between ISWI, Sin3A and Rpd3, since the three proteins co-localizes at many, though not all, sites on polytene chomosome. Although the resolution of polytene chromosome staining is limited, these biochemical data are consistent with a physical interaction between ISWI and Sin3A/Rpd3 in embryo and larval stages. Previous biochemical studies in flies have not detected the presence of Sin3A and Rpd3 proteins as integral subunits of Drosophila ISWI complexes. Therefore, the physical interaction that found between ISWI and Sin3A/Rpd3 could be transient or indirect (Burgio, 2008).

The nucleosome stimulated ATPase activity of ISWI co-purifies with a histone deacetylase activity associated with the Sin3A/Rpd3 complex in larvae. Interestingly changes in the levels of ISWI alter the binding of Sin3A/Rpd3 to polytene chromosomes and are correlated with changes in global histone H3 and H4 acetylation. Because ISWI function can be antagonized by the site-specific acetylation of histones, it is possible that the Sin3A/Rpd3 complex positively regulates ISWI activity in vivo. Therefore, ISWI and the Sin3A/Rpd3 complex may facilitate each other's function, forming a positive feedback system for chromatin regulation (Burgio, 2008).

Genetic and biochemical studies in yeast have shown that the nucleosome spacing activity of the Isw2 complex can repress transcription in a parallel pathway with the yeast Sin3/Rpd3 histone deacetylase complex. Although, the functional organization of DNA into chromatin is conserved among eukaryotes, mutations in the two yeast counterparts of ISWI, Isw1 and Isw2, do not show any severe phenotype. Conversely, ISWI is a unique and essential gene in Drosophila highlighting a possible divergent role for ISWI in flies and a distinct mechanism of interaction with the Sin3A/Rpd3 complex in higher eukaryotes. Indeed, interactions between SNF2L, a mouse ISWI homolog, and the Sin3A/Rpd3 complex have been proposed to play a role in repressing ribosomal gene transcription in mammals. Furthermore, studies of the thymocyte-enriched chromatin factor SAT1B indicate that its ability to regulates gene expression and organize chromatin folding into loop domains at the IL-2Ra locus is dependent on the catalytic activities of Sin3A/HDAC1 (the mammalian Rpd3) and the ISWI homolog SNF2H protein (Burgio, 2008).

ISWI can also be a target of site-specific acetylation by the GCN5 histone acetyltransferase. Therefore, the functional association found between ISWI and Sin3A/Rpd3 could help regulate the acetylation state of ISWI and modulate its activity. Interestingly, it has been recently reported that ISWI genetically interact with the histone acetyltransferase GCN5. gcn5 mutations cause chromosome condensation defects very similar to the one observed in ISWI and E(bx) mutants, as well as global loss of histone H4 acetylation on lysine 12. A decrease in ISWI activity as a consequence of loss of GCN5-dependent acetylation could in theory account for the observed defects. An alternative possibility is that specific histone acetylations differently regulate ISWI function. Therefore, further studies will be necessary to clarify the roles of Sin3A, Rpd3 and other histone modifying enzymes in the regulation of ISWI-containing complexes function in vivo (Burgio, 2008).

Drosophila ISWI regulates the association of histone H1 with interphase chromosomes in vivo

Although tremendous progress has been made toward identifying factors that regulate nucleosome structure and positioning, the mechanisms that regulate higher-order chromatin structure remain poorly understood. Recent studies suggest that the ISWI chromatin-remodeling factor plays a key role in this process by promoting the assembly of chromatin containing histone H1. To test this hypothesis, the function of H1 was investigated in Drosophila. The association of H1 with salivary gland polytene chromosomes is regulated by a dynamic, ATP-dependent process. Reducing cellular ATP levels triggers the dissociation of H1 from polytene chromosomes and causes chromosome defects similar to those resulting from the loss of ISWI function. H1 knockdown causes even more severe defects in chromosome structure and a reduction in nucleosome repeat length, presumably due to the failure to incorporate H1 during replication-dependent chromatin assembly. These findings suggest that ISWI regulates higher-order chromatin structure by modulating the interaction of H1 with interphase chromosomes (Siriaco, 2009).

These findings provide direct evidence that H1 is a major determinant of interphase chromosome structure and support the proposal that ISWI regulates higher-order chromatin structure by promoting the association of H1 with chromatin. The incorporation of H1 during replication-coupled chromatin assembly has a particularly dramatic effect on chromatin compaction. After chromatin has been assembled, the continued association of H1 with chromosomes, while important, appears to have more subtle effects on chromosome structure (Siriaco, 2009).

An independent analysis of phenotypes resulting from the knockdown of Drosophila His1 by RNAi was recently reported (Lu, 2009). Consistent with the current data, the authors of this study found that histone H1 is essential for Drosophila development. However, they observed relatively mild defects in salivary gland polytene chromosome structure following H1 knockdown. These defects appear similar to the weakest phenotypes observed following H1 knockdown, that may reflect differences in the extent of H1 knockdown achieved in the current studies. On the basis of the analysis of fixed polytene chromosome squashes following H1 depletion, Lu (2009) concluded that H1 is required for the alignment of sister chromatids in polytene chromosomes. Although an even stronger disruption of the banding pattern of polytene chromosome squashes was observed following H1 knockdown, such defects were rarely observed via live analysis. The data therefore argue against a major role for H1 in sister chromatid alignment and illustrate the importance of using live analysis to study factors involved in the regulation of higher-order chromatin structure (Siriaco, 2009).

The incorporation of H1 during replication-coupled chromatin assembly increases the average distance between nucleosomes, thus leading to a decrease in genomewide nucleosome density. Accordingly, a significant decrease was observed in nucleosome repeat length (NRL) following H1 knockdown. By contrast, the loss of ISWI function leads to a dramatic reduction in the level of H1 associated with chromosomes without causing obvious changes in NRL. These data strongly suggest that ISWI promotes the association of H1 with salivary gland polytene chromosomes via a replication-independent mechanism. It remains possible that an additional role for ISWI in replication-coupled H1 assembly escaped detection in the genetic studies due to the failure to completely eliminate ISWI function during the stages of salivary gland development when the bulk of DNA replication occurs. Further experiments, including the analysis of fast-acting conditional ISWI alleles, will be required to address this issue (Siriaco, 2009).

How does ISWI promote the association of H1 with chromatin? By altering the structure, accessibility or fluidity of chromatin, ISWI may facilitate the binding of H1 to chromatin during dynamic exchange. Consistent with this possibility, it was found that inhibitors of oxidative phosphorylation lead to the dissociation of H1 from polytene chromosomes accompanied by its accumulation in the nucleoplasm. Alternatively, ISWI may stabilize the binding of H1 to chromatin by influencing its phosphorylation. H1 is phosphorylated in most organisms, including Drosophila. In both Tetrahymena and mammals, the phosphorylation of H1 weakens its association with chromatin, leading to an increased frequency of H1 exchange. Thus, ISWI may indirectly promote the association of H1 with chromatin by altering the level or activity of an H1 kinase or phosphatase (Siriaco, 2009).

The chromatin of stem cells is hyperdynamic, with both histone H1 and other chromatin-associated proteins undergoing highly elevated rates of exchange. This property of pluripotent cell types appears to be functionally important, since a mutant form of H1 that tightly binds chromatin blocks stem cell differentiation. These findings suggest that ISWI and other factors that regulate the association of H1 with chromatin may play important roles in the regulation of cellular pluripotency and differentiation. This possibility is intriguing in light of recent studies implicating ISWI in both nuclear reprogramming and stem cell self-renewal (Siriaco, 2009).

Previous studies have shown that the dosage compensation machinery antagonizes ISWI function via the acetylation of its nucleosome substrate on H4K16. Furthermore, increased linker histone exchange has been observed in active chromatin enriched in core histone acetylation. It is therefore tempting to speculate that the dynamic association of H1 with chromatin is modulated by the interplay of chromatin-remodeling and -modifying enzymes, thus providing a straightforward mechanism for creating rapid, readily reversible changes in higher-order chromatin structure and gene expression. Further work will be required to test this hypothesis and clarify the molecular mechanisms that regulate the association of H1 with chromatin in vivo (Siriaco, 2009).

Genome-wide characterization of chromatin binding and nucleosome spacing activity of the nucleosome remodelling ATPase ISWI

The evolutionarily conserved ATP-dependent nucleosome remodelling factor ISWI can space nucleosomes affecting a variety of nuclear processes. In Drosophila, loss of ISWI leads to global transcriptional defects and to dramatic alterations in higher-order chromatin structure, especially on the male X chromosome. In order to understand if chromatin condensation and gene expression defects, observed in ISWI mutants, are directly correlated with ISWI nucleosome spacing activity, a genome-wide survey was conducted of ISWI binding and nucleosome positioning in wild-type and ISWI mutant chromatin. This analysis revealed that ISWI binds both genic and intergenic regions. Remarkably, ISWI was found to bind genes near their promoters causing specific alterations in nucleosome positioning at the level of the Transcription Start Site, providing an important insights in understanding ISWI role in higher eukaryote transcriptional regulation. Interestingly, differences in nucleosome spacing, between wild-type and ISWI mutant chromatin, tend to accumulate on the X chromosome for all ISWI-bound genes analysed. This study shows how in higher eukaryotes the activity of the evolutionarily conserved nucleosome remodelling factor ISWI regulates gene expression and chromosome organization genome-wide (Sala, 2011).

The unusual sensitivity of the male X chromosome to the loss of ISWI function has suggested in the past that changes in chromatin structure that accompany dosage compensation might regulate the ability of ISWI to remodel chromatin in vivo. Indeed, dosage compensation is necessary and sufficient for the chromosome defects observed in ISWI mutant larvae. In particular, genetic studies have shown that in the absence of ISWI the decondensed male X chromosome is dosage compensated and that in vitro H4K16Ac counteracts ISWI nucleosome spacing activity by reducing ISWI binding to chromatin. However, the current data show that ISWI binds dosage compensated genes affecting nucleosome positioning near the TSS. Interestingly, it has been recently shown that the dosage compensation complex preferentially binds its target genes towards their 3'-end. When ISWI binding sites were mapped on MSL-bound genes, it was found that ISWI tends to bind the 5′-end on those genes. Therefore, ISWI and the MSL occupy complementary gene functional elements along dosage compensated genes, making the nucleosome changes observation consistent with published data (Sala, 2011).

In conclusion, it was found that the loss of ISWI globally causes subtle changes in nucleosome positioning. However, these changes are highly localized and mainly concentrated at level of the TSS, including the ones mapping the X chromosome that are particularly enriched in dosage compensated genes, probably explaining the specific sensitivity of the male X for ISWI loss. Although the data suggest that in higher eukaryotes ISWI could regulate gene expression by remodelling chromatin near gene regulatory regions, additional studies will be necessary to understand the advantage of having ISWI-sensitive-nucleosome-free TSS on dosage compensated genes. Furthermore, the genome-wide analysis of ISWI binding and nucleosome spacing activity in Drosophila also revealed that ISWI-dependent alterations in nucleosome positioning at the level of the TSS are unlikely to be correlated with global chromatin structural alterations observed when ISWI activity is lost. Collectively, these data provide new insights on the role played by the evolutionarily conserved chromatin remodelling factor ISWI in transcriptional regulation and chromosome organization in higher eukaryotes (Sala, 2011).

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

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

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