One of the most enigmatic aspects of genome organization in multicellular eukaryotes is the regionalization of chromosomes into euchromatin and heterochromatin domains. Heterochromatin, originally named "junk DNA" because no coding function could be found for it, is now considered essential for the epigenetic maintenance of centromeric function as well as for other cellular, developmental and evolutionary processes [1–3]. In most eukaryotes, the main components of heterochromatin are families of highly tandem repeated DNA or satellite DNA organized as multiple copies of a monomer sequence arranged in a head to tail pattern over megabase-long arrays. Running from a few base pairs to more than 1 kb in length, repeated units show a wide range of sizes and complexity. Despite sequence divergence between monomers of the same family often being very low over long arrays, satellite DNA in heterochromatin can change rapidly in nucleotide sequence or copy number during evolution. As a consequence, large numbers of unrelated satellite DNA families commonly compound the profile of satellite DNA in genomes [4, 5].
Drosophila melanogaster provides a model for studies of heterochromatin. About 59 Mb of the 176 Mb female genome, including the proximal half of the X chromosome, the pericentromeric region of autosomes 2 and 3, and most of the dot-like chromosome 4 is heterochromatic. The entire male Y chromosome (41 Mb) is also heterochromatic . However, polytene chromosomes, which have proved useful in mapping euchromatin regions, provide minimal resolution in heterochromatin analyses. Due to its late replication, the heterochromatin remains immersed in the diffuse and unbanded chromocentre region. Thus, mitotic rather than polytene chromosomes are preferable for chromosome mapping of heterochromatic sequences .
Over the last three decades, different families of highly repeated tandem sequences have been molecularly characterized and physically mapped to the chromosomes of D. melanogaster using different cloning strategies. Most of the highly repeated families of the fruit fly genome are composed of short repeat units 5-12 bp long, arrayed in tandem and extended over several megabases of DNA. These highly repeated sequences are localized in specific segments of the heterochromatin, which may contain different sets of satellite DNA [8–11].
The goal of large-scale genome projects is to convert the initial draft of euchromatic sequences into a complete telomere-to-telomere sequence for each chromosome. The Drosophila Heterochromatin Genome Project has made substantial progress in identifying contiguous sequences of the heterochromatin regions containing single-copy genes and dense clusters of transposable elements. However, the sequence and organization of the highly tandem repetitive DNA fraction - the vast majority of D. melanogaster heterochromatin - is little known . The apparent absence of identified highly repeated sequences in certain heterochromatic regions and BACs covering the extensive heterochromatin gaps, suggests that unknown classes of highly repetitive DNAs must be present in this fraction of the D. melanogaster chromosomes. To determine what they are will require new technologies .
Short motifs of 1-6 bp repeated in tandem are classified as SSRs or microsatellites [14, 15]. Although the genomes of higher organisms contain some long microsatellites (up to 500 nucleotides), in general these polymorphic loci are no longer than 100 nucleotides. However, in many species, SSRs may also be organized into long stretches of nearly 100 to several thousand tandem units mainly clustered in the heterochromatin, also referred to as satellite SSRs [16, 17]. For example, in many vertebrate species, the heterochromatic sex chromosomes are rich in clusters of GATA and GACA repeats . Long arrays of AAGAG and AATAT are found in the heterochromatin of fruit fly chromosomes . Moreover, GGAAT and CATTT repeats have been found in the satellite regions of human chromosomes . There is some evidence that the origin and evolution of satellite families, including those with relatively complex and longer repeated units, can be explained by the autoreplicative properties of different SSRs .
In the present study, a recently reported, simple and highly efficient non-denaturing FISH method (ND-FISH) was used to localize the classic satellites (AAGAG, AAGAC, AATAT and AATAC) and AACAC repeats . Heterochromatic regions enriched in SSRs were also sought, and the physical distribution of 16 different SSR types (mono-, di-, tri- and tetranucleotide repeats) studied in both polytene and mitotic chromosomes. An ND-FISH analysis employing DNA fibres was also performed and the SSRs characterized by Southern hybridization.