Courtesy of Ulrich Laemmli and Armand Schrumpf.

Satellite DNA is mysterious. It comprises an array of short tandem-repeat sequences, and is associated with an inactive type of chromatin — heterochromatin. The function of satellite DNA is unclear, although there is evidence that it plays a role in the structural organization of the nucleus. To understand satellite DNA better, we need tools to manipulate it. Two papers published by Ulrich Laemmli's laboratory demonstrate how sequence-specific drugs can be used to study satellite DNA. And the effects of the drugs on the heterochromatic state of satellite DNA have some surprising phenotypic consequences.

The sequence-specific drugs are polyamides that comprise a short chain of aromatic amino acids able to bind the minor groove of DNA in a sequence-specific manner. Several such compounds were made and shown to bind specifically to different classes of Drosophila satellite DNA, depending on the sequence of the basic repeat unit. By labelling the drugs with a fluorescent tag, the different minisatellites could be visualized within cells. (For example, minisatellites are highlighted in red and cyan in the stained polytene chromosomes in the image.)

Binding of the polyamide drug to satellite DNA causes significant changes to the conformation of the heterochromatin — the chromatin 'opens up', and becomes more accessible to enzymes such as topoisomerase II and endonucleases. The authors then went on to ask whether drug treatment would reactivate a gene silenced under the influence of heterochromatin.

They did this by examining the effects of polyamides on Drosophila mutants subject to position effect variegation (PEV) — mutants caused by the juxtaposition of a gene and a heterochromatic region. In some cells, the heterochromatin spreads and silences the adjacent gene in that cell and its progeny. The resulting organism is a mosaic of wild-type and mutant tissue. Two PEV mutants were tested. When white-mottled flies were fed the polyamide drug, which had no effect on wild-type flies, the phenotype was suppressed. This is the expected result — the drug leads to a more open heterochromatin conformation next to the white gene, so the gene is less likely to be silenced by heterochromatin spreading. When the second PEV mutant was treated with a polyamide drug, the effects were more surprising.

This mutant, brown Dominant (bwD), is caused by a large insertion of heterochromatic satellite DNA (labelled red in the image) in the brown gene. The fascinating characteristic of this mutant is that it silences the wild-type allele in trans — hence the 'dominant' name. This is thought to occur because the heterochromatin causes both brown alleles to be located in a heterochromatic compartment of the nucleus, leading to stochastic silencing of the wild-type allele, and the variegated phenotype. Feeding these flies with a drug that binds specifically to the satellite repeat within the bwD allele leads to a range of defects during development. These defects did not occur when wild-type flies were treated, and so were dependent on the presence of the bwD allele. After much detective work, the explanation was found — opening of the bwD heterochromatin exposes the GAGAA satellite repeat, which can soak up a certain transcription factor (GAF, a GAGA factor encoded by a Trithorax-like gene), and produces a phenotype that resembles a hypomorphic mutation in GAF. It had been shown previously that GAF moves from a euchromatic distribution (in interphase) to centric heterochromatin during mitosis, suggesting that the reversible association with satellite DNA might be a normal form of GAF regulation.

By opening up chromatin, these new polyamide drugs render satellite DNA more open to scrutiny. Their effects on PEV mutants in Drosophila will stimulate new ideas about this and other non-coding components of the eukaryotic genome.