The role of histone H1 in chromatin condensation and transcriptional ...

208 downloads 0 Views 244KB Size Report
and the linker histone H1 and its variants including H5 facilitate the folding ... globular domain of the linker histone binds to the nucleosome core bridging.
Chapter 12 The role of histone H1 in chromatin condensation and transcriptional repression Memmo Buttinelli*, Gianna Panetta, Daniela Rhodes and Andrew Travers MRC Laboratory of Molecular Biology, Hills Road,Cambridge CB2 2QH *Current address: Fondazione 'Istituto Pasteur - Fondazione Cenci-Bolognetti’, c/o Dipartimento di Genetica e Biologia Molecolare, Università di Roma "La Sapienza", Piazzale le Aldo Moro, 5, 00185 Roma, Italy

In eukaryotes transcriptional repression correlates in general with chromatin condensation. This can be either domain-wide or local, encompassing in the latter case just a few nucleosomes (1). The fundamental structural unit of condensed chromatin, in particular of the '300Å fiber' or' solenoid', is the chromatosome. This particle contains a histone octamer and one molecule of linker histone associated with 168 bp of DNA (2) in contrast to the 145 bp of DNA contained in the nucleosome core particle lacking the linker histone. Varying lengths of linker DNA separate the chromatosomes and the linker histone H1 and its variants including H5 facilitate the folding of the fiber. These histones are known to be located on the inside of the 'solenoid' structure in metazoan chromatin (3). The linker histones themselves contain basic N- and C-terminal tails flanking a central globular domain. This domain is sufficient for chromatosome formation but its position in the particle and hence the orientation of the histone octamers in the solenoid has remained controversial (4).

1

2

1. Location and orientation of the globular domain of histone H5 in the nucleosome An early model of the 300Å fiber, based on images obtained from the electron microscope, proposed a solenoidal structure in which a helical array of nucleosomes was linked by histone H1 with its globular domain binding to the dyad region located on the interior of the solenoid (5). In this model a single linker DNA connects adjacent nucleosomes in the helix. Since this proposal, various alternative structures have been proposed that differ in the path of the linker DNA, e.g. the 'zig-zag' model, or in the location of linker histone on the chromatosome (6) The interior location of H1 was experimentally confirmed by neutron diffraction studies (3) but evidence for its binding site in the chromatosome remained elusive (4). The 3-D structures of both GH1 and GH5 showed that the globular domain was a three-helix bundle with structural homology to helix-turn-helix (HTH) DNA-binding proteins. These structures revealed the existence of a secondary putative DNA binding site separated by 25 Å from the DNA recognition helix of the HTH motif (7,8), consistent with previous experimental demonstrations that this domain could simultaneously bind two DNA duplexes (9,10). This structure suggested a model in which the globular domain of the linker histone binds to the nucleosome core bridging two adjacent DNA gyres (Figure 1). Both these predictions have now been confirmed. Goytisolo et al. (11) first demonstrated that mutations in the secondary binding site abolish the ability of GH5 to bind two duplexes and to form chromatosomes. Then Zhou et al. (12) directly mapped the binding site of the globular domain of linker histone H5 on mixed sequence chicken chromatosomes. This site-specific mapping used a conjugated protein-DNA photocross-linking reagent linked to specific cysteine residues substituted for serine in the wild-type protein domain. These experiments show that the recognition helix of the HTH motif, helix III, binds in the major groove of the first helical turn of the chromatosomal DNA, i.e. close to one terminus of chromatosomal DNA, while the secondary DNA binding site on the opposite face of the globular domain of H5 contacts the nucleosomal DNA close to its midpoint. These two binding sites are separated by one superhelical turn of chromatosomal DNA. By exploiting the ability of some serine->cysteine mutants to selfdimerise Zhou et al. (12) inferred that helix I and helix II of the globular domain of H5 face respectively the solvent and the nucleosome. In bulk chromatin the globular domain of the linker histone thus forms a bridge between one terminus of chromatosomal DNA and the midpoint. This mapping is also consistent with the observed cross-linking of GH5 to one terminus of chromatosomal DNA associated with a unique sequence DNA.

12

3

Taken together the data of Zhou et al. (12) places the C-terminus of GH5 on the outside of the chromatosome particle between one terminus of the bound DNA and the central gyre and its N-terminus on the inner surface of the entering DNA. The orientation of the globular domain relative to the surface of the octamer is also consistent with the protection of lysine residues in helix II from reductive methylation in chromatin containing H5 (13)

Figure 1. Model for the binding of the globular domain of histone H5 to the chromatosome. Helix III binds in a major groove close to one terminus (blue) of chromatosomal DNA while the loop between helix I and helix II contacts the DNA close to the midpoint (vermilion) of the bound sequence.

2. Structural asymmetry of the chromatosome In mixed sequence chromatosomes GH5 binds asymmetrically with respect to the DNA sequence. Initial studies on the structure of the chromatosome proposed that the linker histone might extend the protected DNA symmetrically by approximately one turn at both ends of the nucleosome core (reviewed in refs. 1,4). This conclusion was based on the preservation of a symmetric DNase I cleavage profile within the core section of chromatosomal DNA. However, a symmetric cleavage profile derived from a population of chromatosomes can in principle be derived from the symmetrisation of an asymmetric cleavage pattern for individual particles.

4 Nevertheless, the assumption of an asymmetric cleavage pattern for individual chromatosomes also implies that the cleavage pattern of individual core particles would be similarly asymmetric. More recent micrococcal nuclease digestion studies on positioned chromatosomes provide evidence for asymmetric extension of protection on addition of linker histone, with no increase at one terminus of core DNA yet a ~20 bp increase at the other (14,15). These observations are consistent with the finding that short DNA sequences corresponding to those found preferentially at the dyad of core particles are located ~93 bp from one terminus of chromatosomal DNA (16), where the half-length of core DNA is just 72.5 bp. The available data do not resolve the fundamental structural issue of the location of the two binding sites for GH5 relative to a symmetric or asymmetric extension of protection on addition of linker histone. In the cases where asymmetric extension is observed it has not been excluded that the dyad of the core particle has moved by one turn on binding linker histone. More particularly, if extension is asymmetric, does the recognition helix of GH5 bind to the extended terminus or to the terminus that defines one border of the core particle? One additional feature of chromatosomal DNA from chicken erythrocytes is the frequent occurrence of sequences related to the tetranucleotide AGGA within half a double-helical turn of one terminus (17,18). Although this signal sequence is positioned similarly to the contact point with helix III of GH5 it could also constitute a preferred binding site for a core histone tail dislocated upon GH5 binding.

3. Alternative models for the placement of linker histone on the nucleosome Both the original model for GH5 positioning proposed by Allan et al. (19) and the more recent variant of Zhou et al. (12) argue that GH5 binds at or close to the chromatosome pseudodyad. However a radically different model was recently proposed by Pruss et al. (20) and by Hayes (21). This model is based on cross-linking (20), site-directed DNA cleavage (21) and micrococcal nuclease mapping studies (22) on a chromatosome formed on Xenopus borealis somatic 5S rDNA and posits that the globular domain of linker histone H5 binds on the inside of one DNA gyre at one internal site 65 bp from the dyad. This contact site is reportedly about two helical turns from the proximal terminus of chromatosomal DNA. In this model no contact is made with the dyad region but instead the putative secondary DNA binding site is positioned on the upper surface of the octamer close to the H2A-H2B dimer.

12

5

It remains a formal possibility that the chromatosome formed on the Xenopus borealis somatic 5S rDNA has a different organization from those formed on most other DNA sequences. This model and that proposed by Zhou et al. (12) accordingly differ significantly both in the position and the number of contacts GH5 makes with nucleosomal DNA. Not only do the two models differ fundamentally in the positioning of GH5 on the nucleosome but they also differ in another significant respect. In the Pruss et al. model the C-terminus of GH5 is directed along the upper DNA gyre towards the dyad whereas in the bulk chromatin model the C-terminus is directed towards the linker DNA. Since the C-terminal domain of H5 binds to linker DNA (23) the Pruss et al. model necessarily invokes a U-turn in the Cterminal domain (24) whereas this is unnecessary in the bulk chromatin model. In principle cross-linking and site directed DNA cleavage should accurately identify DNA sequence(s) in close proximity to GH5, but the mapping of these contacts onto the nucleosome structure requires an independent reference point, of which the most appropriate in this case is the nucleosomal pseudodyad. The derivation of the Pruss et al. model assumed that the somatic 5S rDNA chromatosome occupies a single dominant translational position on two different DNA fragments which in one case (20) included vector sequences. The actual inferred dominant dyad position differed in the two sets of experiments, as also did the length of the DNA fragments and the DNA sequences (c.f. refs 20,21). However recent studies have questioned the validity of the assumption of a single dominant translational position. An et al. (25) suggest that the original detailed micrococcal nuclease mapping of the somatic 5S chromatosome may itself be intrinsically unreliable and also show, by low resolution mapping, at least two nucleosome positions on the X. borealis somatic 5S DNA sequence. Similarly Panetta et al. (26), by mapping dyad positions directly using a sitedirected hydroxyl radical cleavage reaction developed by Flaus et al. (27), show that the population of core particles and chromatosomes formed on this DNA is a mixture of several different translational settings of similar occupancy. Although the precise number of settings available to a chromatosome will depend on the length of the DNA fragment used, even

6

Figure 2. Multiple nucleosome positions on the somatic 5S rDNA. The r.h. panel shows the preparative fractionation of free DNA (lane 1) and nucleosomal complexes (lane 2) on a 5% polyacrylamide gel. The nucleosome dyad positions present in each gel band are shown in a three dimensional representation on the left. The dyads are numbered relative to the startpoint of 5S RNA transcription and deduced from site-directed hydroxyl radical cleavage. Red circles are placed every 10 bp of the DNA helix axis to indicate positions on the 5S rDNA fragment. The region of DNA bound by the histone octamer is shown in red and the free DNA in blue. The TFIIIA binding site is indicated in green where exposed, and in black where covered by a positioned nucleosome. Figure adapted from ref. 26.

with only two equivalent settings the contacts deduced from cross-linking and directed DNA cleavage cannot be unambiguously assigned to any particular dyad position (26). Indeed, the experimental data of Pruss et al. (20) and of Hayes (21) could be entirely consistent with the bulk chromatin model if at least some of their chromatosomes occupied the positions observed by Panetta et al. The two principal methods previously used to identify translational nucleosome positions, micrococcal nuclease mapping and protein-DNA cross-linking, are both dependent on DNA sequence and thus do not necessarily identify all the positions occupied by the histone octamer on a particular DNA sequence. Micrococcal nuclease is known in some cases to cleave nucleosomal DNA internally (28) and thus the positions identified by this technique, which depends on the recovery of DNA fragments whose lengths sum to ~145 bp (29), are essentially those which are resistant to internal cleavage by the enzyme. Similarly the low efficiency DNA cross-

12

7

linking technique used to locate dyad positions requires the modification of only purine residues and a prolonged incubation at high temperatures (for example 8 hr at 42° (30)). With both techniques a positive result in principle correctly identifies a nucleosomal translational position (but see An et al. (25)) but a negative result cannot be construed as an unoccupied position. In all studied cases, with the exception of one uniquely positioned nucleosome in the MMTV promoter, the site directed hydroxyl radical cleavage method identifies more translational positions than those identified by other methods. For example, on the Xenopus laevis somatic 5S rDNA the micrococcal nuclease method identifies one (31) and three positions (32) under identical ionic conditions but on slightly different DNA fragments whereas the site-directed method identifies a minimum of seven positions on a similar DNA fragment (GP and MB, unpublished observations). Formally it might be argued that the multiple positions observed using site directed hydroxyl radical cleavage arise from a destabilization of histone octamer - DNA interactions. This possibility arises because the conservative serine -> cysteine mutation in histone H4 required for coupling of the reagent lies close to, although does not share identity with, residues whose mutation reduces the requirement for the SWI/SNF remodeling complex (33). However, no evidence can be adduced for such a destabilization from genetic, structural and biochemical observations. First, Kruger et al. (33,) using a genetic screen, obtained multiple isolates of the val43ile and arg45cys/his H4 mutations that partially complemented a swi1 mutation but failed to isolate a comparable mutation at position 47 of H4. Second there are strong structural reasons why mutations at positions 43 and 45 would be expected to destabilize the core nucleosome whereas the conservative ser47cys mutation would not. In the crystal structure of the nucleosome core particle (34) val43 at the base of the H4 loop packs against H3 via a hydrophobic interaction. Altering it to a more bulky hydrophobic sidechain could significantly alter the whole direction of the loop. Similarly arg45 at the tip of the loop points straight into the minor groove and not surprisingly makes electrostatic contacts with the DNA. Altering it to either Cys or His would significantly alter the stability of these contacts. By contrast, the ser47 sidechain is on the far side of the H4 loop and does not make any contact with any other residue. The oxygen of the sidechain is in fact over 3.5 Å distant from the nearest atom, and thus the structure is unlikely to be perturbed by the substitution of a sulphur atom for oxygen. Nor is there any direct evidence that suggests that conjugation of the reagent affects the stability of histone-DNA interactions. In particular the temperature-induced shifting from an off-center to a centered position on a 146 bp DNA fragment of the Lytechinus variegatus 5S gene is identical for nucleosomes reconstituted with wild type Xenopus octamer and with the

8 mutant octamer conjugated with the reagent (27). Equally pertinently, essentially the same pattern of multiple bands generated by multiple nucleosome positions on the Xenopus borealis 5S rDNA is observed with both conjugated and unconjugated Xenopus histones (26) and also with chicken erythrocyte histones (M. Donahue and J.O. Thomas, personal communication). The ability of the nucleosome core particle to adopt multiple positions is thus independent of the provenance of the octamer but instead must depend on an intrinsic property of the DNA sequence itself. Given that rotational positioning depends on the summation of multiple weak signals these same signals would be expected to specify a family of rotationally related positions, as is observed in vivo in yeast (35).

4. A model for the selective repression of the Xenopus borealis oocyte 5S rRNA gene by histone H1. Although H1 is thought to have a general repressive effect on transcription (36,37), the two 5S RNA multigene families of Xenopus provide an instructive example of selective repression by H1. The expression of both the somatic and oocyte 5S rDNA genes is developmentally regulated. The somatic 5S rDNA genes (400 copies per haploid genome) are active in both somatic and oocyte cells, whereas the oocyte 5S rDNA genes (20,000 copies) are only transcribed in oocytes (38). The two types of genes have virtually identical coding sequences, but differ in the sequence of their flanking DNAs: that of the somatic genes is GC rich, while that of the oocyte genes is AT rich (39). Transcriptional activation of the two genes requires the same set of transcription factors and the primary event in the formation of the transcription complex is the binding of TFIIIA to the internal control region, located between position 45 and 95 of the 120 bp genes (40). TFIIIA binds with the same affinity to the two genes (41). Two conflicting models have been proposed to explain the differential expression of the two types of 5S rDNAs. From in vitro studies it has been proposed that the differential expression could be due to differences in the stability, or in the kinetics of assembly of the transcriptional complexes on the two genes (42-44). Such differences could favor the expression of the somatic gene type as the concentrations of transcription factors become limiting during development (45,46). Contrary to this model, in vivo studies show that when the expression of H1 is abolished by a ribozyme targeted to somatic H1 mRNA, the oocyte 5S rDNA expression is activated under limiting transcription factor concentrations (47,48). This observation provides good evidence that the differential regulation of transcription of the two 5S rDNA is effected by histone H1. The expression of H1 is

12

9

developmentally regulated and the selective repression of the oocyte 5S rDNA coincides with the onset of H1 expression, without affecting transcription of the somatic 5S rDNAs (42, 49). Moreover, in vitro, removal of histone H1 by salt (50) or by ion-exchange resin (42) permits the activation of oocyte 5S rDNA expression in a somatic chromatin context. All these observations point to differences in the chromatin structure over the two 5S rDNA genes, which in some way allow H1 to selectively repress transcription of the oocyte gene without affecting the expression of the somatic type. In other words the somatic gene remains accessible to transcription factors in the presence of H1. Consistent with this, in vivo chromatin probing studies have revealed differences in chromatin structure over the two types of genes (51-53). Our results (26) show that differences in the positioning of the histone octamer with respect to the TFIIIA binding site on the Xenopus borealis somatic and oocyte 5S rDNA determine the preference of binding for both TFIIIA and H1. When H1 and TFIIIA compete for binding to somatic nucleosomes, in which the TFIIIA binding site is located at one edge of the nucleosome (54) or in the linker DNA, TFIIIA binds preferentially. Indeed TFIIIA can displace H1 from the somatic nucleosome, since TFIIIA binding is observed even when added to nucleosomes containing H1 (Figure 3).

10

Figure 3. Differential binding of H1 and TFIIIA to the somatic and oocyte nucleosomes. Somatic (lane 2) and oocyte (lane 7) nucleosomes were incubated with TFIIIA (lanes 3, and 8), with H1 (lanes 4 and 9) and with H1 plus TFIIIA (lanes 5 and 10). The positions of naked 5S rDNA (D) and the TFIIIA-DNA complex (DT) are indicated. The bracket indicates the position of the complexes formed between nucleosomes and TFIIIA or H1. The complexes were fractionated on a 5% polyacrylamide gel. Reproduced with permission from Panetta et al. (26).

These results indicate that the binding of TFIIIA to a somatic nucleosome is incompatible with H1 binding. The preferential binding of TFIIIA to the somatic nucleosome is also consistent with the constitutive expression of the somatic 5S rDNA family in the presence of elevated H1 levels in vivo (42). For the oocyte nucleosomes the binding preference for H1 and TFIIIA is reversed. H1 clearly binds preferentially to the oocyte nucleosome. This is because almost all of the nucleosome positions on the oocyte 5S rDNA incorporate the TFIIIA binding site within the nucleosome, so that without repositioning the TFIIIA binding site is masked. The consequence of this is that H1 binding is favored. Once bound, H1 locks the nucleosome and prevents repositioning by TFIIIA, thus completely inhibiting TFIIIA binding. In vivo H1 has a dominant role in the selective repression of Xenopus laevis oocyte 5S rDNA (47,48). Our results suggest a simple mechanism for this selective repression. We propose that nucleosome positioning on the oocyte gene has a key role in promoting the binding of H1 by excluding binding of TFIIIA. This mechanism would be facilitated by the

12

11

accumulation of H1 and the concomitant reduction in TFIIIA concentrations in somatic cells after the midblastula transition (46,49) The mechanism we propose for a role of nucleosome positioning in transcriptional repression is based on a model system studying mononucleosomes. However, in vivo, regulation occurs in the context of nucleosome arrays. We suggest that the ultimate effect of the competition between TFIIIA and H1 at the mononucleosomal level is to determine the higher order chromatin structure of the 5S RNA genes. Thus, the preferential binding of H1 to the oocyte nucleosomes would facilitate the assembly of oocyte nucleosome arrays into the 30 nm fiber (55) ensuring the selective repression of the oocyte 5S rDNA family. Such an effect might be enhanced by the organization of oocyte genes in clusters of tandem 200 bp repeats (56), creating an optimum spacing for contiguous H1 binding (57). By contrast, the preferential binding of TFIIIA to the somatic nucleosome inhibits H1 binding, with the consequence of disfavoring chromatin condensation, thus ensuring that transcription is maintained. Similar conclusions have been drawn for the differential regulation of the oocyte and somatic 5S rRNA genes of Xenopus laevis (31), but we note that as discussed above this conclusion is strongly dependent on the analysis of nucleosome positioning by micrococcal nuclease cleavage. References: 1. 2. 3. 4. 5.

6.

7.

8.

9.

Travers, A. (1999) Towards a higher order structure for chromatin – the location of the linker histone in the nucleosome. Trends Biochem. Sci., in press. Simpson, R.T. (1978) Structure of the chromatosome, a chromatin particle containing 160 bp of DNA and all histones. Biochemistry 17, 5524-5531. Graziano, V., Gerchmann, S.E., Schneider, D.K. and Ramakrishnan, V. (1994) Histone H1 is located in the interior of the chromatin 30-nm filament. Nature 368, 351-354. Crane-Robinson, C. (1997) Where is the globular domain of linker histone located on the nucleosome? Trends Biochem. Sci. 22, 75-77. Thoma, F., Köller, T. and Klug, A. (1979) Involvement of histone H1 in the organization of the nucleosome and of the salt dependent superstructure of chromatin. J. Cell Biol. 83, 403 Bednar, J., Horowitz, R.A., Dubochet, J. and Woodcock, C.L. (1995) Chromatin conformation and salt-induced compaction: three-dimensional structural information from cryoelectron microscopy. J. Cell. Biol. 131, 1365-1376. Ramakrishnan, V., Finch, J.T., Graziano, V., Lee, P.L. and Sweet, R.M. (1993) Crystal structure of the globular domain of histone H5 and its implications for nucleosome binding. Nature 362, 219-223. Cerf, C., Lippens, G., Ramakrishnan, V., Muyldermans, S., Segers, A., Wyns, L., Wodak, S.J. and Hallenga, K. (1994) Homo- and heteronuclear two-dimensional NMR studies of the globular domain of histone H1: full assignment, tertiary structure and comparison with the globular domain of histone H5. Biochemistry 33, 11079-11086 Thomas, J.O., Rees, C. and Finch, J.T. (1992) Cooperative binding of the globular domains of histones H1 and H5 to DNA Nucl. Acids. Res. 20, 187-194.

12 10. 11.

12.

13.

14.

15.

16. 17. 18. 19. 20.

21.

22. 23. 24. 25.

26.

27.

28.

Draves, P.H., Lowary, P.T. and Widom, J. (1992) Cooperative binding of the globular domain of histone H5 to DNA. J. Mol. Biol. 225, 1105-1121. Goytisolo, F.A. Gerchman, S.E., Yu, X., Rees, C., Graziano, V., Ramakrishnan, V. and Thomas, J.O. (1996) Identification of two DNA-binding sites on the globular domain of histone H5. EMBO J. 15, 3421-3429. Zhou, Y.-B., Gerchman, S.E., Ramakrishnan, V., Travers, A. and Muyldermans, S. (1998) Position and orientation of the globular domain of linker histone H5 on the nucleosome. Nature 395, 402-405. Thomas, J.O. and Wilson, C.M. (1986) Selective radiolabeling and identification of a strong nucleosome binding site on the globular domain of histone H5. EMBO J. 5, 3531-3537. An, W., Leuba, S.H., van Holde, K. and Zlatanova, J. (1998). Linker histone protects DNA on only one side of the core particle, in a sequence-dependent manner. Proc. Natl. Acad. Sci. USA, 95, 3396-3401. Wong, J., Li, Q., Levi, B.-Z., Shi, Y.-B. and Wolffe, A.P. (1998) Structural and functional features of a specific nucleosome containing a recognition element for the thyroid hormone receptor. EMBO J. 17, 520-534. Satchwell, S.C. and Travers, A.A. (1989) Asymmetry and polarity of nucleosomes in chicken erythrocyte chromatin. EMBO J. 8, 229-238. Muyldermans, S.V. and Travers, A.A. (1994) DNA sequence organization in chromatosomes. J. Mol. Biol. 235, 855-870. Travers, A.A. and Muyldermans, S.V. (1996) A DNA sequence for positioning chromatosomes. J. Mol. Biol. 257, 486-491. Allan, J., Crane-Robinson, C. and Aviles, F.X. (1980). The structure of histone H1 and its location in chromatin. Nature 288, 675-679. Pruss, D., Bartholomew, B., Persinger, J., Hayes, J., Arents, G., Moudrianakis, M.N. and Wolffe, A.P. (1996) An asymmetric model for the nucleosome: A binding site for linker histones inside the DNA gyres. Science 274, 614-617. Hayes, J.J. (1996) Site-directed cleavage of DNA by a linker histone-Fe(II) EDTA conjugate: Localization of a globular domain binding site within a nucleosome. Biochemistry 35, 11931-11937 Hayes, J.J., and Wolffe, A.P. (1993) Preferential and asymmetric interaction of linker histones with 5S DNA in the nucleosome. Proc. Natl. Acad. Sci. USA 90, 6415-6419. Hamiche, A., Schultz, P., Ramakrishnan V., Oudet, P. and Prunell, A. (1996) Linker histone dependent DNA structure in mononucleosomes. J. Mol. Biol. 257, 30-42. Lee, K.M. and Hayes, J.J. (1998) Linker DNA and H1-dependent reorganization of histone-DNA interactions within the nucleosome. Biochemistry 37, 8622-8628. An, W., van Holde, K. and Zlatanova, J. (1998) Linker histone protection of chromatosomes reconstituted on 5S rDNA from Xenopus borealis: a reinvestigation. Nucl. Acids Res. 26, 4042-4047. Panetta, G., Buttinelli, M., Flaus, A., Richmond, T.J. and Rhodes, D. (1998) Differential nucleosome positioning on Xenopus oocyte and somatic 5 S RNA genes determines both TFIIIA and H1 binding: a mechanism for selective H1 repression. J. Mol. Biol. 282, 683-697. Flaus, A., Luger, K., Tan, S. and Richmond, T.J. (1996) Mapping nucleosome position at single base-pair resolution by using site-directed hydroxyl radicals. Proc Natl Acad Sci U S A, 93, 1370-1375. Drew, H.R. and Calladine, C.R. (1987) Sequence-specific positioning of core histones on an 860 base-pair DNA. Experiment and theory. J Mol Biol, 195, 143-173.

12

13

29.

Dong, F., Hansen, J.C. and van Holde, K.E. (1990) DNA and protein determinants of nucleosome positioning on sea urchin 5SrRNA gene sequences in vitro. Proc. Natl. Acad. Sci. USA, 87, 5724-5728. Pruss, D. and Wolffe, A.P. (1993) Histone-DNA contacts in a nucleosome core containing a 5S rRNA gene. Sera, T. and Wolffe, A.P. (1998) Role of histone H1 as an architectural determinant of chromatin structure and as a specific repressor of transcription on Xenopus oocyte 5S rRNA genes. Mol. Cell Biol. 18, 3668-3680. Howe, L. and Ausió, J. (1998) Nucleosome translational position, not histone acetylation, determines TFIIIA binding to nucleosomal Xenopus laevis 5S rRNA genes. Mol. Cell. Biol. 18, 1156-1162. Luger, K., Mäder, A.W., Richmond, R.K., Sargent, D. L. and Richmond, T.J. (1997) Crystal structure of the nucleosome core particle at 2.8 Å resolution. Nature 389, 251260. Buttinelli, M., Di Mauro, E. and Negri, R. (1993) Multiple nucleosome positioning with unique rotational setting for the Saccharomyces cerevisiae 5S rRNA gene in vitro and in vivo. Proc Natl Acad Sci U S A, 90, 9315-9319. Laybourn, P.J. and Kadonaga, J.T. (1991) Role of nucleosomal cores and histone H1 in regulation of transcription by RNA polymerase II. Science, 254, 238-245. Juan, L.J., Utley, R.T., Vignali, M., Bohm, L. and Workman, J.L. (1997) H1-mediated repression of transcription factor binding to a stably positioned nucleosome. J Biol Chem, 272, 3635-3640. Wormington, W.M. and Brown, D.D. (1983) Onset of 5 S RNA gene regulation during Xenopus embryogenesis. Dev Biol, 99, 248-257. Peterson, R.C., Doering, J.L. and Brown, D.D. (1980) Characterization of two Xenopus somatic 5S DNAs and one minor oocyte-specific 5S DNA. Cell, 20, 131-141. Engelke, D.R., Ng, S.Y., Shastry, B.S. and Roeder, R.G. (1980) Specific interaction of a purified transcription factor with an internal control region of 5S RNA genes. Cell, 19, 717-728. McConkey, G.A. and Bogenhagen, D.F. (1988) TFIIIA binds with equal affinity to somatic and major oocyte 5S RNA genes. Genes Dev, 2, 205-214. Schlissel, M.S. and Brown, D.D. (1984) The transcriptional regulation of Xenopus 5S RNA genes in chromatin: the roles of active stable transcription complexes and histone H1. Cell, 37, 903-913. Wolffe, A.P. and Brown, D.D. (1988) Developmental regulation of two 5S ribosomal RNA genes. Science, 241, 1626-1632. Seidel, C.W. and Peck, L.J. (1992) Kinetic control of 5 S RNA gene transcription. J Mol Biol, 227, 1009-1018. Pelham, H.R., Wormington, W.M. and Brown, D.D. (1981) Related 5S RNA transcription factors in Xenopus oocytes and somatic cells. Proc Natl Acad Sci U S A, 78, 1760-1764. Shastry, B.S., Honda, B.M. and Roeder, R.G. (1984) Altered levels of a 5 S genespecific transcription factor (TFIIIA) during oogenesis and embryonic development of Xenopus laevis. J Biol Chem, 259, 11373-11382. Bouvet, P., Dimitrov, S. and Wolffe, A.P. (1994) Specific regulation of Xenopus chromosomal 5S rRNA gene transcription in vivo by histone H1. Genes Dev, 8, 11471159. Kandolf, H. (1994) The H1A histone variant is an in vivo repressor of oocyte-type 5S gene transcription in Xenopus laevis embryos. Proc Natl Acad Sci U S A, 91, 72577261.

30. 31.

32.

34.

35.

36. 37.

38. 39. 40.

41. 42.

43. 44. 45.

46.

47.

48.

14 49. 50. 51. 52. 53. 54. 55. 56.

57.

Flynn, J.M. and Woodland, H.R. (1980) The synthesis of histone H1 during early amphibian development. Dev Biol, 75, 222-230. Korn, L.J. and Gurdon, J.B. (1981) The reactivation of developmentally inert 5S genes in somatic nuclei injected into Xenopus oocytes. Nature, 289, 461-465. Gottesfeld, J.M. and Bloomer, L.S. (1980) Nonrandom alignment of nucleosomes on 5S RNA genes of X. laevis. Cell, 21, 751-760. Young, D. and Carroll, D. (1983) Regular arrangement of nucleosomes on 5S rRNA genes in Xenopus laevis. Mol Cell Biol, 3, 720-730. Chipev, C.C. and Wolffe, A.P. (1992) Chromosomal organization of Xenopus laevis oocyte and somatic 5S rRNA genes in vivo. Mol Cell Biol, 12, 45-55. Rhodes, D. (1985) Structural analysis of a triple complex between the histone octamer, a Xenopus gene for 5S RNA and transcription factor IIIA. EMBO J, 4, 3473-3482. Finch, J.T. and Klug, A. (1976) Solenoidal model for superstructure in chromatin. Proc Natl Acad Sci U S A, 73, 1897-1901. Korn, L.J. and Brown, D.D. (1978) Nucleotide sequence of Xenopus borealis oocyte 5S DNA: comparison of sequences that flank several related eucaryotic genes. Cell, 15, 1145-1156. Tomaszewski, R. and Jerzmanowski, A. (1997) The AT-rich flanks of the oocyte-type 5S RNA gene of Xenopus laevis act as a strong local signal for histone H1-mediated chromatin reorganization in vitro. Nucl. Acids Res. 25, 458-466.