Figure 1. Mapping DNAse I-hypersensitive sites in chromatin using indirect end labelling. This technique was developed in 1980 by S. Nedospasov and G. Georgiev, and independently by C. Wu and S. Elgin. The outline of the experiment is shown on the left: nuclei are isolated from tissue culture cells or tissue, treated briefly with DNAse I, the DNA is extracted and digested to completion with a restriction endonuclease. Sites of cleavage by DNAse I in chromatin are revealed by Southern blotting with a probe that abuts one end of a chosen restriction fragment. An example of the data generated by this approach is shown on the right: the gene shown changes activity during development, and thus one can examine chromatin structure at various developmental stages. Pronounced cleavage by DNAse I (red arrowheads) is only observed at the transcription start site and at a stretch of regulatory DNA located about 500 bp upstream only when the gene is active. The ‘end’ lane shows that the tissue used contains endogenous nucleases that will also preferentially cleave chromatin within the DNAse I-hypersensitive sites, if longer incubation times are used. The model system used here is the fly Sciara (see Gerbi SA, Liang N, Wu N et al., Cold Spring Harbor Symposia of Quantitative Biology 58: 487).

Figure 2. A schematic of the core histone octamer (centre) with the DNA superhelix (blue) and the core histone tails extended to their full length. Lysine residues that can be modified by acetylation are indicated with an asterisk. From Wolffe AP and Hayes J 2000 Nucleic Acids Research 27: 711. Copyright 1999 Oxford University Press. Used by permission.

Figure 3. Revealing the binding sites of specific proteins in chromatin by chromatin immunoprecipitation (ChIP). This technique was developed in 1988 by M. Solomon and A. Varshavsky and subsequently (in 1991) by M. Gorovsky and D. Allis. The outline of the experiment is shown on the left: cells are briefly treated with formaldehyde to introduce covalent crosslinks between proteins and DNA. Chromatin is isolated and sheared into small (<1-kb) fragments, followed by an immunoprecipitation against a protein of interest (ellipse). The crosslinks are then eliminated, protein removed, and the DNA analysed by polymerase chain reaction for the presence of specific sequences. An example of such data is shown on the right. The upper right panel shows that the yeast HDAC Rpd3p is required for the transcriptional repression of the INO1 gene (second panel) – this Northern blot demonstrates that INO1 levels are low in wild-type cells, but increase dramatically if the RPD3 gene is inactivated (‘rpd3’ lane). A ChIP experiment on the INO1 locus with an antibody against hyperacetylated histones is shown on the lower right: the top line is the map of the gene, and the bottom data set is a PCR reaction performed with multiple primer pairs (these detect fragments over the entire locus). Compared with the input lanes (on the right), only a fragment over the transcription start site (‘–0.09 kb’) is not immunoprecipitated with this antibody (second lane). When the RPD3 gene is inactivated (‘rpd3’), acetylation over that stretch of DNA becomes equal to that seen over adjacent areas. Thus, transcriptional repression by Rpd3p of the INO1 gene (top right) occurs concomitant with a deacetylation of chromatin over its transcription start site (bottom right). From Rundlett SE, Carmen A, Suka N et al. (1998) Nature 392: 831. Copyright 1998 Macmillan Publishers.

Figure 4. Histone tail acetylation affects the functional and structural properties of chromatin. The panel on the left shows a transcription assay performed with hyperacetylated (‘H’) and unacetylated (‘U’) chromatin. Massive amounts of read-through transcripts are generated on the former, but not the latter template (left panel). The top right panel shows a schematic representation of shape changes that chromatin undergoes in solutions of various ionic strength. The lower right panel shows that altering the acetylation state of chromatin (‘H’ stands for ‘hyperacetylated’ and ‘M’ for ‘moderately acetylated’) changes its folding: hyperacetylated fibres are more unfolded (extended) and thus have a lower Svedberg number. From Tse C, Sera T, Wolffe AP et al. (1988) Molecular and Cellular Biology 18: 4629. Copyright 1988 American Society for Microbiology. Used by permission.

Figure 5. A hypermethylated promoter is silenced via the targeting of histone deacetylase. From Robertson KR and Wolffe AP 1996 Nature Reviews Genetics 1: 11. Copyright 2000 Macmillan Publishers.