Origin of DNA compaction: As a repeating unit in eukaryotic chromatin, a nucleosome wraps DNA in superhelical turns around a histone octamer. Mattiroli et al. present the crystal structure of an archaeal histone-DNA complex in which the histone-mediated DNA geometry is exactly the same as that in the nucleosome. Comparing features of archaeal and eukaryotic chromatin structures offers important insights into the evolution of eukaryotic nucleosomes. Science , this issue p. 609
1 August 1990 in Proceedings of the National Academy of Sciences
Methanothermus fervidus grows optimally at 83 degrees C. A protein designated HMf (histone M. fervidus) has been isolated from this archaeal hyperthermophile that binds to double-stranded DNA molecules and increases their resistance to thermal denaturation. HMf binding to linear double-stranded DNA molecules of greater than 2 kilobase pairs also increases their electrophoretic mobilities through agarose gels. Visualization of this compaction process by electron microscopy has demonstrated the formation of quasispherical, macromolecular HMf-DNA complexes. HMf is a mixture of approximately equal amounts of two very similar polypeptides designated HMf-1 and HMf-2. Determination of the DNA sequence of the gene encoding HMf-2 (hmfB) has revealed that over 30% of the amino acid residues in HMf-2 are conserved in the consensus sequences derived for eucaryal histones H2A, H2B, H3, and H4. These archaeal polypeptides and eucaryal histones appear therefore to have evolved from a common ancestor and are likely to have related structures and functions.
Archaeal histone–DNA complexes assembled in vivo ( Takayanagi et al. 1992 Takayanagi S. Morimura S. Kusaoke H. Yokoyama Y. Kano K. Shioda M. J. Bacteriol. 1992; 174 : 7207-7216 Crossref PubMed Scopus (22) Google Scholar ) and in vitro ( Sandman et al. 1990 Sandman K. Krzycki J.A. Dobrinski B. Lurz R. Reeve J.N. Proc. Natl. Acad. Sci. USA. 1990; 87 : 5788-5791 Crossref PubMed Scopus (146) Google Scholar ) visibly resemble nucleosomes. They protect ∼60 bp ladders of DNA from micrococcal nuclease digestion, and exposure to formaldehyde cross-links the archaeal histones within these complexes into tetramers ( 5 Grayling R.A. Sandman K. Reeve J.N. Adv. Protein. Chem. 1996; 48 : 437-467 Crossref PubMed Google Scholar , 6 Grayling R.A. Bailey K.A. Reeve J.N. Extremophiles. 1997; 1 : 79-88 Crossref PubMed Scopus (26) Google Scholar ). The archaeal nucleosome appears, therefore, to be analogous, and is possibly homologous, to the structure formed at the center of the eucaryal nucleosome by the (H3+H4)2 tetramer. The (H3+H4)2 tetramer initiates the assembly of the eucaryal nucleosome, recognizes nucleosome positioning signals, and wraps ∼120 bp in a left-handed superhelix but protects only ∼70 bp from nuclease digestion ( Hamiche et al. 1996 Hamiche A. Carot V. Alilat M. De Luca F. O'Donohue M.-F. Révet B. Prunell A. Proc. Natl. Acad. Sci. USA. 1996; 93 : 7588-7593 Crossref PubMed Scopus (105) Google Scholar ). Archaeal nucleosomes formed in vitro similarly assemble at preferred locations ( Grayling et al. 1997 Grayling R.A. Bailey K.A. Reeve J.N. Extremophiles. 1997; 1 : 79-88 Crossref PubMed Scopus (26) Google Scholar ) but wrap DNA in a right-handed superhelix ( Grayling et al. 1996 Grayling R.A. Sandman K. Reeve J.N. Adv. Protein. Chem. 1996; 48 : 437-467 Crossref PubMed Google Scholar ), which, at first sight, appears to be fundamentally different from the (H3+H4)2-based structure. A slight shift in the dimer–dimer interface may, however, be all that is needed for the (H3+H4)2 tetramer to wrap DNA in a right-handed helix ( Hamiche et al. 1996 Hamiche A. Carot V. Alilat M. De Luca F. O'Donohue M.-F. Révet B. Prunell A. Proc. Natl. Acad. Sci. USA. 1996; 93 : 7588-7593 Crossref PubMed Scopus (105) Google Scholar ), and switching from left- to right-handed wrapping has been predicted to occur rapidly and reversibly and to depend on the superhelical torsion of the DNA. The structure needed for the (H3+H4)2 tetramer to constrain DNA in a right-handed superhelix would be incompatible with the addition of (H2A+H2B) dimers and would therefore limit the complex to the presence of only a histone tetramer. The handedness of the DNA superhelix in archaeal nucleosomes in vivo is not known, but archaeal histone binding to relaxed, circular DNAs in vitro introduces both negative and positive superhelicity, depending on the protein-to-DNA ratio ( Grayling et al. 1996 Grayling R.A. Sandman K. Reeve J.N. Adv. Protein. Chem. 1996; 48 : 437-467 Crossref PubMed Google Scholar ). If the dimer–dimer interface-reorientation concept is correct ( Hamiche et al. 1996 Hamiche A. Carot V. Alilat M. De Luca F. O'Donohue M.-F. Révet B. Prunell A. Proc. Natl. Acad. Sci. USA. 1996; 93 : 7588-7593 Crossref PubMed Scopus (105) Google Scholar ), archaeal nucleosomes could constrain DNA in both left- and right-handed superhelices, and the local superhelical tension in the DNA would determine which form predominates. It should be possible to test this proposal by mutagenesis of the residues on the dimer–dimer interface that direct the formation of homotetramers of a single recombinant archaeal histone (Figure 2).