Nucleosome Core Particles Containing a Poly(dA·dT) Sequence Element Exhibit a Locally Distorted DNA Structure

doi:10.1016/j.jmb.2006.06.051

J. Mol. Biol. (2006) 361, 617–624

COMMUNICATION

Nucleosome Core Particles Containing a Poly(dA·dT) Sequence Element Exhibit a Locally Distorted DNA Structure Yunhe Bao†, Cindy L. White† and Karolin Luger⁎ Howard Hughes Medical Institute and Department of Biochemistry and Molecular Biology, Colorado State University, Fort Collins, CO 80523-1870, USA

Poly(dA·dT) DNA sequence elements are thought to promote transcription by either excluding nucleosomes or by altering their structural or dynamic properties. Here, the stability and structure of a defined nucleosome core particle containing a 16 base-pair poly(dA·dT) element (A16 NCP) was investigated. The A16 NCP requires a significantly higher temperature for histone octamer sliding in vitro compared to comparable nucleosomes that do not contain a poly(dA·dT) element. Fluorescence resonance energy transfer showed that the interactions between the nucleosomal DNA ends and the histone octamer were destabilized in A16 NCP. The crystal structure of A16 NCP was determined to a resolution of 3.2 Å. The overall structure was maintained except for local deviations in DNA conformation. These results are consistent with previous in vivo and in vitro observations that poly(dA·dT) elements cause only modest changes in DNA accessibility and modest increases in steady-state transcription levels. © 2006 Elsevier Ltd. All rights reserved.

*Corresponding author

Keywords: crystal structure; fluorescence resonance energy transfer; nucleosome; DNA structure

Poly(dA·dT) sequence elements are greatly overrepresented in all eukaryotic species examined (from yeast to human), and are particularly enriched in promoter regions.1 It has been suggested that poly (dA·dT) sequence elements promote transcription by either excluding or altering the stability of nucleosomes, thereby enhancing the binding between transcription activators and their DNA targets.2–5 Poly(dA·dT) tracts exhibit a straight and rigid DNA structure that is significantly different from canonical B-form DNA. 6 Specifically, they are characterized by a narrow minor groove, 10 bp per helical turn, and bifurcated hydrogen bonds between bases. 7–10 Because of these structural properties, it has been proposed that longer poly (dA·dT) tracts would not be able to conform well to the highly bent nucleosomal superhelix under

† Y.B. and C.L.W. contributed equally to this work. Abbreviations used: MRE, metal response element; FRET, fluorescence resonance energy transfer; rmsd, rootmean-square deviation; A16, 16 bp poly(dA·dT). E-mail address of the corresponding author: [email protected]

physiological conditions. Previous studies have shown that poly(dA·dT) elements are resistant to nucleosome formation.11,12 A study in which 177 nucleosomal DNA fragments were sequenced showed that longer adenosine tracts (containing five or more contiguous adenosine residues) were underrepresented in nucleosomal DNA in vivo.13 In yeast, many promoter regions including poly (dA·dT) elements are nucleosome-free in vivo.14–16 However, several groups have shown that poly (dA·dT) sequence elements can be incorporated into nucleosomes both in vitro17–19 and in vivo. Examples for the latter include the DNA topoisomerase I promoter in yeast, where a 29 bp poly(dA·dT) tract is located within a positioned nucleosome20; and nucleosome 1 on the ADH2 promoter region in yeast which encompasses a stretch of 20 (dA·dT).21 This nucleosome is re-arranged upon activation of the ADH2 promoter. Similarly, a 16 bp poly(dA·dT) (A16) element located adjacent to a metal response element (MRE) within the Amt1 gene promoter of Candida glabrata was incorporated into a positioned nucleosome in vivo.22,23 This element is essential for rapid activation of the Amt1 promoter in response to toxic copper levels.24 Thus, poly(dA·dT) tracts alone

0022-2836/$ - see front matter © 2006 Elsevier Ltd. All rights reserved.

618 are not sufficient to disrupt nucleosome formation, but might result in local distortions in nucleosomal DNA, thereby permitting or enhancing the binding of regulatory factors. To date, several studies have investigated the impact of poly(dA·dT) elements on histone–DNA interactions. For example, one group reported that incorporation of a 16 bp poly(dA·dT) element at the end or towards the middle of the nucleosomal DNA destabilized histone–DNA interactions, and resulted in a 1.6-fold increase in position-averaged equilibrium accessibility of nucleosomal DNA target sites.5 Another study suggested the opposite effect: a 25 bp poly(dA·dT) tract stabilized the nucleosome.25 The greatest stabilizing effect occurs if the (dA·dT) tract is placed near the nucleosomal dyad. Thus, the effects of poly(dA·dT) incorporation on histone– DNA interaction still remain uncertain. Importantly, no information is available on the effect of such extreme sequences on nucleosome structure, nor is it known how their presence might result in transcription activation. Most nucleosomes whose structures have been determined to date were reconstituted with the same palindromic DNA sequence derived from human α-satellites (reviewed by Luger26). The only nucleosome structure with a different DNA sequence uses a palindromic DNA fragment that has been derived from a different α-satellite repeat.27 To investigate the effects of the poly(dA·dT) element on the properties of a nucleosome, we have previously reconstituted nucleosomes with a 147 bp DNA fragment containing an A16 element and the sequence recognition element (MRE) for Amt1.28 We found that there was no preference for DNA se-

Structure of Nucleosomes With a Poly(dAd dT) Tract

quences with or without the A16 element in an in vitro assembly system; the DNA binding domain of Amt1 bound nucleosomes containing the A16 element with only a threefold reduced affinity compared to free DNA; and no dissociation of histones was required for Amt1 binding. Here we present a more detailed analysis of nucleosomes containing the A16 element and MRE (A16 NCP). Fluorescence resonance energy transfer (FRET) was utilized to compare the stability of NCPs reconstituted with either α-satellite DNA (α-sat DNA)29 or A16 DNA, and the crystal structure of the A16 NCP was determined. A16 NCP requires a higher temperature for sliding in vitro When nucleosomes are reconstituted by saltgradient deposition, the DNA usually adopts Figure 1. A16 NCP requires higher temperatures to shift in vitro, and exhibits destabilized DNA ends. (a) Salt gradient reconstituted A16NCP (lanes 1–3) and α-sat NCP (lanes 4 and 5), before (unshifted (US)) and after a 1 h incubation at 37 °C or 55 °C, were analyzed by 5% (w/v) native PAGE and stained with Coomassie blue. (b) The A16NCP was shifted at 55 °C for the indicated times. Samples were analyzed by 5% native PAGE and stained with Coomassie blue. (c) The salt-dependent dissociation of A16NCPs (squares) and α-sat NCPs (circles) was measured by fluorescence resonance energy transfer between labels attached to H2B T112C and the DNA ends. NaCl was added up to 1.5 M in 0.02 M steps, and donor emission (475 nm) was recorded at each step upon excitation at 385 nm. Data were corrected for dilution before plotting. Standard deviations from three separate datasets were calculated and reported as error bars. Design and preparation of 147 bp A16 and 146 bp α-sat DNA have been described.28 The 147 bp A16 sequence reads in its entirety ATCAATATTCACCTGCACATTCTACCAAAAGTGTCAAAAAAAAAAAAAAAATCATGATAAGCTAATTTGGCTGÅCTCAGCTGAAC AT G C C T T T T G AT G G A G C A G T T T C C A A A TACCCTTTTGGAGTATCTGCAGGTGGATATTGAT (central base-pair is underlined; see White and Luger28 for a sequence alignment). Histones from Xenopus laevis were purified and nucleosomes were assembled using published protocols.31,41 The purified 147 bp A16 DNA and 146 bp α-sat DNA were labeled with the donor chromophore 7-diethylamino-3-(4′-maleimidylphenyl)- 4-methylcoumarin (CPM) as described.28 X. laevis mutant H2B T112C was expressed and purified as described,41 and was labeled with the acceptor chromophore fluorescein-5maleimide (FM) as described.28 Nucleosomes containing donor (CPM)-labeled DNA and acceptor (FM)-labeled H2B (for protein-DNA FRET) were mixed with 0.1 M NaCl to a final concentration of 28 μg/ml (1.33 × 10− 7 M) and incubated for 30 min in a buffer containing 10 mM TrisCl (pH 8.0) and 0.1 mM EDTA. For stability measurements, donor emission quenching as a result of energy transfer to the nearby acceptor was monitored at 475 nm (excitation at 385 nm) in response to increased ionic strength up to 1.5 M. All experiments were carried out at 20 °C, and each data set was corrected for dilution by dividing the raw signal by [1– (injection volume/total volume)]. The data sets were then normalized to the same scale, and the averages of the normalized data sets were plotted. Error bars of the standard deviation of three data sets were calculated and reported.

619

Structure of Nucleosomes With a Poly(dAd dT) Tract

several translational positions with respect to the histone octamer, even when DNA fragments with strong positioning signals are used. The different nucleosome species are identified by differential mobility on native polyacrylamide gels,30 even on DNA fragments as short as 146 base-pairs.31 For nucleosomes reconstituted with palindromic DNA fragments such as the ones used for X-ray crystallography,26 two bands are usually observed: a lower band in which the DNA is centrally positioned, and one upper band, which represents two species in which the DNA is displaced by ten base-pairs to either side of the histone octamer dyad.29 Due to the palindromic nature of the DNA, these two latter species are equivalent. Although the A16 DNA under investigation here has been derived from the palindromic α-sat DNA, introduction of the MRE (bp 66–74) and poly(dA·dT) stretch (bp 36–51) has broken the symmetry (see White and Luger28 for sequence alignment). Consequently, three bands with different electrophoretic mobility are observed on a native gel after saltgradient reconstitution (Figure 1(a), lane 1), as has previously been observed for the asymmetric 5 S rRNA gene.31 A simple heating step allows the histone octamer to attain a thermodynamically most favorable position in a process termed nucleosome sliding.30,31 The temperature at which nucleosomes slide (or the times required at constant temperatures) has been used as a measure for nucleosome stability to compare nucleosomes with mutant histones.32 α-sat NCPs completely shift by heating at 37 °C for 30 min.33 In contrast, the two offcentered species in A16 NCP require different temperatures to attain a thermodynamically stable central position on the histone octamer. Most of the upper band was shifted to the centrally positioned nucleosome at 37 °C (Figure 1(a), lane 2). However, the middle band required heating at 55 °C (Figure 1(a), lanes 3–5). A time-course of shifting for A16 NCPs showed that all nucleosomes present in the upper band shifted within 5 min at 55 °C, whereas 10 min were required for the complete shifting of the middle band (Figure 1(b)). This indicates that the relative position of poly (dA·dT) tract (A16) with respect to the histone octamer determines the energy barrier that exists for a histone octamer to slide along the DNA in a temperature-induced manner. Our finding that higher temperatures are required for shifting is consistent with the finding that the intrinsic curvature that is characteristic for A-tracts is largely lost at higher temperatures.34 The interactions between the nucleosomal DNA ends and the histone octamer are destabilized in A16 NCP The rigid nature of poly(dA·dT) tracts gives them the potential to destabilize nucleosomes or discourage nucleosome formation at ambient temperatures.15,35 On the other hand, highly stable

nucleosomes form on a 146 bp poly(dA·dT) tract at elevated temperatures,17,25 reflecting the finding that at high temperatures and high ionic strength, such tracts behave more like B-DNA.34 To resolve this disparity, FRET was used to compare the stability of NCPs reconstituted with either α-sat 146mer or A16 147mer in response to increasing ionic strength. We have previously shown that subtle changes in nucleosome stability can be determined by judicious placement of fluorescence donors and acceptors.36 Using this method, we have consistently observed at least two distinct transitions of NCP dissociation when monitoring FRET between the DNA ends and the (H3-H4)2 tetramer. The first, fully reversible transition corresponds to the dissociation of the peripheral DNA ends. The second transition is caused by the dissociation of the histone dimer from the (H3-H4)2–DNA complex. Both transitions are also apparent when monitoring FRET between the ends of the DNA and the (H2AH2B) dimer, and can be observed for α-sat NCPs, as well as for A16 NCPs (Figure 1(c)). Importantly, the dissociation of the DNA ends from the histone octamer was initiated at a significantly lower salt concentration for A16 NCP (Figure 1(c)), and the

Table 1. Crystallographic statistics Space group Unit cell dimensions (Å) a b c Resolution range (Å) Unique reflections Redundancy (last shell) Completeness (%)(last shell) I/σ (last shell) Rmergea (last shell) Rcrystb Rfreec Number of amino acids Number of DNA bases rmsdd Bonds (Å) Angles (°) Average B-factors (Å2) Protein DNA

P212121 104.9 109.6 178.0 3.2–50.0 34,730 7.1 (6.8) 99.4 (99.8) 54.60 (10.23) 0.067 (0.226) 0.2816 0.3507 758 294 0.008 1.354 84.52 168.17

A16 NCP was crystallized as described.29 Synchrotron Data were collected at the Advanced Light Source (beamline 8.2.2). Data were indexed and scaled with DENZO and SCALEPACK.42 MoQ lecular replacement was used to obtain crystallographic phases, using Protein Data Bank ID code 1KX5 as the search model. Refinement and model building was done with CNS43 and O.44 The model was checked by using stimulated annealing omit maps and validated using PROCHECK.45 DNA geometry was analyzed by 3DNA (v1.5) [Lu, XJ and Olson, WK, Rutgers university, http://www.rutchem.rutgers.edu/~xiangjun/3DNA/]. Figures were made with PYMOL [W. L. DeLano, PYMOL Molecular Graphics System (2002), http://www.pymol.org]. Numerical values in parentheses are for highest resolution shell. a Rmerge=Σ|Ih–|/Σ Ih, where is the mean of the measurements for a single hkl. b Rcryst=Σ|Fobs–Fcalc|/ÓFobs. c Calculated using 5% data randomly chosen and omitted from the refinement. d rmsd, rms deviation from ideal geometry.

620

Structure of Nucleosomes With a Poly(dAd dT) Tract

Figure 2. The A16 element attains a defined position in the crystal lattice. (a) Schematic representation of a nucleosome crystal lattice: lack of discrimination of the two halves of the non-palindromic DNA sequence results in a pseudosymmetric structure, with the A16 element apparent in both halves. (b) Precise orientation of the non-palindromic DNA sequence in the crystal lattice leads to a unique structure in which the A16 element (shown in black) is observed only in one half of the structure. Crystal packing and unit cell (black rectangle) are schematic and are not an accurate reflection of the true crystal lattice. (c) Stereo view of a section of the composite omit map, contoured at 1.2 σ, showing base-pair 67 from within the MRE. One orientation of the DNA molecule (green) fits the map much better than the other (magenta).

midpoint of the transition is at lower ionic strength compared to α-sat NCP. In contrast, the (H2A-H2B) dimer dissociated from the DNA at similar salt concentrations in the two nucleosomes. This result suggests that the interactions between the nucleosomal DNA ends and the histone octamer are destabilized due to the presence of the A16 sequence element. We have previously shown that the ends of the DNA partially unravel from the surface of the histone octamer to allow Amt1 access to its binding site near the nucleosomal dyad, in an expansion of the transient site-exposure model proposed by Widom and colleagues.37,38 Our finding that the DNA ends are destabilized in response to increasing ionic strength in A16 NCP demonstrates that DNA sequence in regions other than those directly involved may have an effect on the breathing of the DNA ends.

The crystal structure of the A16 NCP reveals that the poly(dA·dT) tract conforms to the histone octamer In order to investigate if the incorporation of the poly(dA·dT) sequence element causes any structural changes in the nucleosome, we determined the Xray structure of the A16 NCP (Table 1). When crystallizing nucleosomes reconstituted with asymmetric DNA fragments (as opposed to the previously used palindromic DNA), the orientation of nucleosomes in the crystal lattice is a potential problem. In the worst case scenario, a lack of discrimination between the two halves of the DNA fragment could lead to a convolution of orientations, as shown schematically in Figure 2(a). Ideally, the scenario as shown in Figure 2(b) in which all poly (dA·dT) are oriented identically throughout the crystal lattice should exist.

Figure 3. Overall comparison of the DNA structure of A16 NCP and α-sat NCP. Partial regions of the DNA, that is, bp 1–74 (a) and bp 75–147 (b) are viewed down the superhelical axis. The DNA of A16 NCP and α-sat NCP are colored in red and blue, respectively. The A16 element (bp 36–51, orange) and Amt1 binding site (MRE; bp 66–74, green), and the region of highest rmsd between the A16 NCP and α-sat NCP (bp 125–135, cyan) are indicated. (c) Detailed comparison of a minor groove within the A16 element (bp 37–45) and the corresponding region in α-sat NCP. (d) Side view of the A16 NCP. The poly(dA·dT) element (bp 36–51) is colored in orange; bp 125–135 is colored in cyan. (e) Stereo view of a section of the |2Fo–Fc| electron density map, calculated at 3.2 Å and contoured at 1.2 σ, showing the part of the A16 element (bp 37–39). (f) Minor groove widths (calculated with 3DNA as described in Materials and Methods) are plotted against DNA residue number (upper panel). Values for A16 NCP model and α-sat NCP model are colored in magenta and blue, respectively. The positions for the A16 element (red bar) and MRE (green bar) are labeled. Equivalent regions on the other half of the DNA are shown by broken red and green bars, respectively. Asterisks indicate regions of largest minor groove width differences. The rmsds for the DNA in A16 NCP versus that in α-sat NCP are plotted against the DNA residue number (lower panel). Plots for the two strands (bases 2–147 and 149–294) are colored in orange and cyan, respectively.

Structure of Nucleosomes With a Poly(dAd dT) Tract

Figure 3 (legend on previous page)

621

622 To distinguish between these scenarios, we first performed identical refinement regimes with three nucleosome models in which the MRE and poly (dA·dT) sequence elements were introduced on either side or on both sides of the dyad, respectively. Free R-values of 0.364, 0.376, and 0.462 clearly indicated the superiority of one particular orientation of the MRE and poly(dA·dT). However, it is still possible that discrimination between the two possible orientations is not complete. We therefore performed refinements with combined Protein Data Bank (pdb) files in which the ratio of occupancies between the two different orientations was varied. Free R-values increased for all combinations. Finally, to exclude model bias, we compared the two orientations of the MRE and poly(dA·dT) elements using composite omit maps (Figure 2(c)). Despite the limited resolution, one solution was clearly superior in accommodating the poly-purine and poly-pyrimidine pattern of the DNA sequence (Figure 2(b)). A one-base-pair example is shown in green and magenta, respectively, in Figure 2(c). The minor groove of the poly(dA·dT) tract is narrowed within the context of the nucleosome The relatively low resolution and high B-factors that characterize the DNA in this particular structure somewhat limit the types of analyses that can be done. Despite this shortcoming, important conclusions can be drawn from a superposition of the A16 NCP and 147 bp α-sat NCP. The two refined structures superimpose with an average root mean square deviation (rmsd) of 0.4 Å for proteins and 1.3 Å for DNA. Overall there is no significant difference in the path of the DNA between the two models (Figure 3(a) and (b)). In particular, the 16 bp poly(dA·dT) DNA fragment conforms well to the DNA topology typical for nucleosomal DNA. Representative regions of the electron density map for part of the poly(dA·dT) sequence element are shown (Figure 3(e)). All the histone–DNA interactions are maintained within the A16 NCP structure. A careful inspection of the structure within the (dA·dT) sequence element reveals a narrowed minor groove (Figure 3(c)). This observation is consistent with known poly(dA·dT) DNA structures.7 A more quantitative analysis is shown in Figure 3(f) (upper panel). Overall, minor groove widths of the A16 DNA are similar to those of the α-sat 147 bp DNA, except for a significant shift at the 16 bp poly(dA·dT) region (spanning base-pairs 36–51). Minor groove widths also diverge near the ends of the DNA (marked by asterisks). Not surprisingly, the phosphate atoms in these regions exhibit a high rmsd when comparing α-sat and A16 NCP (Figure 3(f), lower panel). The region of highest rmsds is located between base-pairs 125–135, the region just beside the poly (dA·dT) DNA fragment on the other gyre of the nucleosomal DNA. This is not the result of altered groove widths, but rather stems from a movement

Structure of Nucleosomes With a Poly(dAd dT) Tract

of the DNA backbone to form a more inter-digitated “groove-in-groove” arrangement (Figure 3(d)). This observation is interesting because it appears that the two most distorted DNA regions are physically close in the nucleosome, indicating that there is structural “crosstalk” between the two gyres of nucleosomal DNA. In contrast, there is no significant structural alteration in the MRE region (Figure 3(a)). One hallmark of this particular structure is the relatively high temperature factors over the entire length of the DNA, indicative of a higher degree of static or dynamic disorder than previously observed for canonical α-sat nucleosomes. This phenomenon has been observed before for other non-palindromic DNA sequences (for example, the 5 S rRNA gene (K. L., unpublished results)), and may well be the “natural” situation for nucleosomal DNA. Additionally, a certain degree of convolution between different orientations of nucleosomes in the crystal lattice may also contribute to the increased B-factors in these structures. This study was inspired by the finding that a positioned nucleosome encompasses the poly (dA·dT) element and the binding site for the transcription factor Amt1 in vivo.22 Since an artificial DNA sequence was used in the present study, in which the poly(dA·dT) sequence element and the MRE have been grafted onto a strong positioning sequence, one must be cautious in any speculation on the effect of these sequence elements on in vivo chromatin structure. Nevertheless, our findings are consistent with the in vivo and in vitro observations that modest changes in DNA accessibility and modest increases in steady-state transcript levels are caused by poly (dA·dT) elements.4,5,22,39 Poly(dA·dT) elements in yeast are mostly nucleosome-free due to their location in promoter regions,14–16,40 however, the same is not true of poly(dA·dT) elements in higher eukaryotes.13,22 Thus, our results suggest one mechanism by which poly(dA·dT) elements may regulate transcription. Protein Data Bank accession number Structure factors and coordinates have been deposited with the RCSB Protein Data Bank with accession number 2FJ7.

Acknowledgements This work was supported by a grant from the NIH (GM61909), and by the Monfort Family Foundation. We thank Pamela N. Dyer for help with histones and DNA, and Young-Jun Park for help with FRET studies. We acknowledge with gratitude Rajeswari S. Edayathumangalam and Srinivas Chakravarthy for help in crystallographic procedures.

Structure of Nucleosomes With a Poly(dAd dT) Tract

References 1. Behe, M. J. (1995). An overabundance of long oligopurine tracts occurs in the genome of simple and complex eukaryotes. Nucl. Acids Res. 23, 689–695. 2. Chen, W., Tabor, S. & Struhl, K. (1987). Distinguishing between mechanisms of eukaryotic transcriptional activation with bacteriophage T7 RNA polymerase. Cell, 50, 1047–1055. 3. Struhl, K. (1985). Naturally occurring poly(dA-dT) sequences are upstream promoter elements for constitutive transcription in yeast. Proc. Natl Acad. Sci. USA, 82, 8419–8423. 4. Iyer, V. & Struhl, K. (1995). Poly(dA:dT), a ubiquitous promoter element that stimulates transcription via its intrinsic DNA structure. EMBO J. 14, 2570–2579. 5. Anderson, J. D. & Widom, J. (2001). Poly(dA-dT) promoter elements increase the equilibrium accessibility of nucleosomal DNA target sites. Mol. Cell. Biol. 21, 3830–3839. 6. Nelson, H. C., Finch, J. T., Luisi, B. F. & Klug, A. (1987). The structure of an oligo(dA).oligo(dT) tract and its biological implications. Nature, 330, 221–226. 7. Alexeev, D. G., Lipanov, A. A. & Skuratovskii, I. (1987). Poly(dA).poly(dT) is a B-type double helix with a distinctively narrow minor groove. Nature, 325, 821–823. 8. Peck, L. J. & Wang, J. C. (1981). Sequence dependence of the helical repeat of DNA in solution. Nature, 292, 375–378. 9. Coll, M., Frederick, C. A., Wang, A. H. & Rich, A. (1987). A bifurcated hydrogen-bonded conformation in the d(A.T) base pairs of the DNA dodecamer d (CGCAAATTTGCG) and its complex with distamycin. Proc. Natl Acad. Sci. USA, 84, 8385–8389. 10. Dickerson, R. E., Drew, H. R., Conner, B. N., Wing, R. M., Fratini, A. V. & Kopka, M. L. (1982). The anatomy of A-, B-, and Z-DNA. Science, 216, 475–485. 11. Kunkel, G. R. & Martinson, H. G. (1981). Nucleosomes will not form on double-stranded RNa or over poly (dA).poly(dT) tracts in recombinant DNA. Nucl. Acids Res. 9, 6869–6888. 12. Rhodes, D. (1979). Nucleosome cores reconstituted from poly (dA-dT) and the octamer of histones. Nucl. Acids Res. 6, 1805–1816. 13. Satchwell, S. C., Drew, H. R. & Travers, A. A. (1986). Sequence periodicities in chicken nucleosome core DNA. J. Mol. Biol. 191, 659–675. 14. Filetici, P., Aranda, C., Gonzalez, A. & Ballario, P. (1998). GCN5, a yeast transcriptional coactivator, induces chromatin reconfiguration of HIS3 promoter in vivo. Biochem. Biophys. Res. Commun. 242, 84–87. 15. Lascaris, R. F., Groot, E., Hoen, P. B., Mager, W. H. & Planta, R. J. (2000). Different roles for abf1p and a Trich promoter element in nucleosome organization of the yeast RPS28A gene. Nucl. Acids Res. 28, 1390–1396. 16. Tanaka, S., Livingstone-Zatchej, M. & Thoma, F. (1996). Chromatin structure of the yeast URA3 gene at high resolution provides insight into structure and positioning of nucleosomes in the chromosomal context. J. Mol. Biol. 257, 919–934. 17. Puhl, H. L. & Behe, M. J. (1995). Poly(dA).poly(dT) forms very stable nucleosomes at higher temperatures. J. Mol. Biol. 245, 559–567. 18. Puhl, H. L., Gudibande, S. R. & Behe, M. J. (1991). Poly [d(A.T)] and other synthetic polydeoxynucleotides containing oligoadenosine tracts form nucleosomes easily. J. Mol. Biol. 222, 1149–1160.

623 19. Losa, R., Omari, S. & Thoma, F. (1990). Poly(dA).poly (dT) rich sequences are not sufficient to exclude nucleosome formation in a constitutive yeast promoter. Nucl. Acids Res. 18, 3495–3502. 20. Rubbi, L., Camilloni, G., Caserta, M., Di Mauro, E. & Venditti, S. (1997). Chromatin structure of the Saccharomyces cerevisiae DNA topoisomerase I promoter in different growth phases. Biochem. J. 328, 401–407. 21. Verdone, L., Camilloni, G., Di Mauro, E. & Caserta, M. (1996). Chromatin remodeling during Saccharomyces cerevisiae ADH2 gene activation. Mol. Cell. Biol. 16, 1978–1988. 22. Zhu, Z. & Thiele, D. J. (1996). A specialized nucleosome modulates transcription factor access to a C. glabrata metal responsive promoter. Cell, 87, 459–470. 23. Liu, X. D. & Thiele, D. J. (1997). Yeast metallothionein gene expression in response to metals and oxidative stress. Methods, 11, 289–299. 24. Koch, K. A. & Thiele, D. J. (1996). Autoactivation by a Candida glabrata copper metalloregulatory transcription factor requires critical minor groove interactions. Mol. Cell. Biol. 16, 724–734. 25. Mahloogi, H. & Behe, M. J. (1997). Oligoadenosine tracts favor nucleosome formation. Biochem. Biophys. Res. Commun. 235, 663–668. 26. Luger, K. (2003). Structure and dynamic behavior of nucleosomes. Curr. Opin. Genet. Dev. 13, 127–135. 27. Davey, C. A. & Richmond, T. J. (2002). DNAdependent divalent cation binding in the nucleosome core particle. Proc. Natl Acad. Sci. USA, 99, 11169–11174. 28. White, C. L. & Luger, K. (2004). Defined structural changes occur in a nucleosome upon Amt1 transcription factor binding. J. Mol. Biol. 342, 1391–1402. 29. Luger, K., Maeder, A. W., Richmond, R. K., Sargent, D. F. & Richmond, T. J. (1997). Crystal structure of the nucleosome core particle at 2.8 Å resolution. Nature, 389, 251–259. 30. Pennings, S., Meersseman, G. & Bradbury, E. M. (1991). Mobility of positioned nucleosomes on 5 S rDNA. J. Mol. Biol. 220, 101–110. 31. Luger, K., Rechsteiner, T. J. & Richmond, T. J. (1999). Preparation of nucleosome core particle from recombinant histones. Methods Enzymol. 304, 3–19. 32. Flaus, A., Rencurel, C., Ferreira, H., Wiechens, N. & Owen-Hughes, T. (2004). Sin mutations alter inherent nucleosome mobility. EMBO J. 23, 343–353. 33. Muthurajan, U. M., Park, Y. J., Edayathumangalam, R. S., Suto, R. K., Chakravarthy, S., Dyer, P. N. & Luger, K. (2003). Structure and dynamics of nucleosomal DNA. Biopolymers, 68, 547–556. 34. Diekmann, S. (1987). Temperature and salt dependence of the gel migration anomaly of curved DNA fragments. Nucl. Acids Res. 15, 247–265. 35. Suter, B., Schnappauf, G. & Thoma, F. (2000). Poly(dA. dT) sequences exist as rigid DNA structures in nucleosome-free yeast promoters in vivo. Nucl. Acids Res. 28, 4083–4089. 36. Park, Y. J., Dyer, P. N., Tremethick, D. J. & Luger, K. (2004). A new fluorescence resonance energy transfer approach demonstrates that the histone variant H2AZ stabilizes the histone octamer within the nucleosome. J. Biol. Chem. 279, 24274–24282. 37. Li, G. & Widom, J. (2004). Nucleosomes facilitate their own invasion. Nature Struct. Mol. Biol. 11, 763–769. 38. Li, G., Levitus, M., Bustamante, C. & Widom, J. (2005). Rapid spontaneous accessibility of nucleosomal DNA. Nature Struct. Mol. Biol. 12, 46–53.

624

Structure of Nucleosomes With a Poly(dAd dT) Tract

39. Mai, X., Chou, S. & Struhl, K. (2000). Preferential accessibility of the yeast his3 promoter is determined by a general property of the DNA sequence, not by specific elements. Mol. Cell. Biol. 20, 6668–6676. 40. Sekinger, E. A., Moqtaderi, Z. & Struhl, K. (2005). Intrinsic histone-DNA interactions and low nucleosome density are important for preferential accessibility of promoter regions in yeast. Mol. Cell. 18, 735–748. 41. Dyer, P. N., Edayathumangalam, R. S., White, C. L., Bao, Y., Chakravarthy, S., Muthurajan, U. M. & Luger, K. (2004). Reconstitution of nucleosome core particles from recombinant histones and DNA. Methods Enzymol. 375, 23–44. 42. Otwinowski, Z. & Minor, W. (1997). Processing of Xray diffraction data collected in oscillation mode. In

Methods in Enzyomology (Carter, C. W. & Sweet, R. M., eds), Macromolecular Crystallography, part A, vol. 276, pp. 307–326, Academic Press, New York. 43. Rice, L. M., Shamoo, Y. & Brunger, A. T. (1998). Phase improvement by moluti-start simulated annealing refinement and structure-factor averaging. J. Appl. Crystallog. 31, 798–805. 44. Jones, T. A., Zou, J. Y., Cowan, S. W. & Kjelgaard, M. (1991). Improved methods for building protein models in electron density maps and the location of errors in these models. Acta Crystallog. sect. A, 47, 110–119. 45. Laskowski, R., MacArthur, M., Moss, D. & Thornton, J. (1993). PROCHECK: a program to evaluate stereochemical quality of protein structures. J. Appl. Crystallog. 26, 283–291.

Edited by J. O. Thomas (Received 10 January 2006; received in revised form 15 June 2006; accepted 21 June 2006) Available online 5 July 2006