Drosophila ISWI, a highly conserved member of the SWI2/SNF2 family of ATPases, is the catalytic subunit of three chromatin-remodeling complexes: NURF, CHRAC, and ACF. To clarify the biological functions of ISWI, we generated and characterized null and dominant-negative ISWI mutations. We found that ISWI mutations affect both cell viability and gene expression during Drosophila development. ISWI mutations also cause striking alterations in the structure of the male X chromosome. The ISWI protein does not colocalize with RNA Pol II on salivary gland polytene chromosomes, suggesting a possible role for ISWI in transcriptional repression. These findings reveal novel functions for the ISWI ATPase and underscore its importance in chromatin remodeling in vivo.

HP1 (Heterochromatin protein 1) is a conserved, non-histone chromosomal protein that is best known for its preferential binding to pericentric heterochromatin and its role in position effect variegation in Drosophila. Using immunolocalization, we show that HP1 is a constant feature of the telomeres of interphase polytene and mitotic chromosomes. This localization does not require the presence of telomeric retrotransposons, since HP1 is also detected at the ends of terminally deleted chromosomes that lack these elements. Importantly, larvae expressing reduced or mutant versions of HP1 exhibit aberrant chromosome associations and multiple telomeric fusions in neuroblast cells, imaginal disks, and male meiotic cells. Taken together, these results provide evidence that HP1 plays a functional role in mediating normal telomere behavior in Drosophila.

We show that methylated lysine 9 of histone H3 (Me9H3) is a marker of heterochromatin in divergent animal species. It localises to both constitutive and facultative heterochromatin and replicates late in S-phase of the cell cycle. Significantly, Me9H3 is enriched in the inactive mammalian X chromosome (Xi) in female cells, as well as in the XY body during meiosis in the male, and forms a G-band pattern along the arms of the autosomes. Me9H3 is a constituent of imprinted chromosomes that are repressed. The paternal and maternal pronuclei in one-cell mouse embryos show a striking non-equivalence in Me9H3: the paternal pronucleus contains no immunocytologically detectable Me9H3. The levels of Me9H3 on the parental chromosomes only become equivalent after the two-cell stage. Finally, we provide evidence that Me9H3 is neither necessary nor sufficient for localisation of heterochromatin protein 1 (HP1) to chromosomal DNA.

The canalization concept describes the resistance of a developmental process to phenotypic variation, regardless of genetic and environmental perturbations, owing to the existence of buffering mechanisms. Severe perturbations, which overcome such buffering mechanisms, produce altered phenotypes that can be heritable and can themselves be canalized by a genetic assimilation process. An important implication of this concept is that the buffering mechanism could be genetically controlled. Recent studies on Hsp90, a protein involved in several cellular processes and development pathways, indicate that it is a possible molecular mechanism for canalization and genetic assimilation. In both flies and plants, mutations in the Hsp90-encoding gene induce a wide range of phenotypic abnormalities, which have been interpreted as an increased sensitivity of different developmental pathways to hidden genetic variability. Thus, Hsp90 chaperone machinery may be an evolutionarily conserved buffering mechanism of phenotypic variance, which provides the genetic material for natural selection. Here we offer an additional, perhaps alternative, explanation for proposals of a concrete mechanism underlying canalization. We show that, in Drosophila, functional alterations of Hsp90 affect the Piwi-interacting RNA (piRNA; a class of germ-line-specific small RNAs) silencing mechanism leading to transposon activation and the induction of morphological mutants. This indicates that Hsp90 mutations can generate new variation by transposon-mediated 'canonical' mutagenesis.

We determined the distribution of 11 different transposable elements on Drosophila melanogaster mitotic chromosomes by using high-resolution fluorescent in situ hybridization (FISH) coupled with charge-coupled device camera analysis. Nine of these transposable elements (copia, gypsy, mdg-1, blood, Doc, I, F, G, and Bari-]) are preferentially clustered into one or more discrete heterochromatic regions in chromosomes of the Oregon-R laboratory stoclk Moreover, FISH analysis of geographically distant strains revealed that the locations of these heterochromatic transposable element clusters are highly conserved. The P and hobo elements, which are likely to have invaded the D. melanogaster genome at the beginning of this century, are absent from Oregon-R heterochromatin but clearly exhibit heterochromatic clusters in certain natural populations. Together these data indicate that transposable elements are major structural components of Drosophila heterochromatin, and they change the current views on the role of transposable elements in host genome evolution.