Genetic studies on heterochromatin in Drosophila melanogaster and their implications for the functions of satellite DNA

Repetitive DNA and chromosome evolution in plants

Most higher plant genomes contain a high proportion of repeated sequences. Thus repetitive DNA is a major contributor to plant chromosome structure. The variation in total DNA content between species is due mostly to variation in repeated DNA content. Some repeats of the same family are arranged in tandem arrays, at the sites of heterochromatin. Examples from the Secale genus are described. Arrays of the same sequence are often present at many chromosomal sites. Heterochromatin often contains arrays of several unrelated sequences. The evolution of such arrays in populations is discussed. Other repeats are dispersed at many locations in the chromosomes. Many are likely to be or have evolved from transposable elements. The structures of some plant transposable elements, in particular the sequences of the terminal inverted repeats, are described. Some elements in soybean, antirrhinum and maize have the same inverted terminal repeat sequences. Other elements of maize and wheat share terminal homology with elements from yeast, Drosophila, man and mouse. The evolution of transposable elements in plant populations is discussed. The amplification, deletion and transposition of different repeated DNA sequences and the spread of the mutations in populations produces a turnover of repetitive DNA during evolution. This turnover process and the molecular mechanisms involved are discussed and shown to be responsible for divergence of chromosome structure between species. Turnover of repeated genes also occurs. The molecular processes affecting repeats imply that the older a repetitive DNA family the more likely it is to exist in different forms and in many locations within a species. Examples to support this hypothesis are provided from the Secale genus.

Intrachromosomal mapping of the nucleolar organiser region relative to three marker loci on chromosome 1B of wheat (Triticum aestivum)

Restriction enzyme digestion of the ribosomal RNA genes of the nucleolar organisers of wheat has revealed fragment length polymorphisms for the nucleolar organiser on chromosome 1B and the nucleolar organiser on 6B. Variation between genotypes for these regions has also been demonstrated. This variation has been exploited to determine the recombination frequency between the physically defined nucleolar organiser on 1B (designatedNor1) and other markers; two loci,Glu-B1 andGli-B1 which code for endosperm storage proteins andRf3, a locus restoring fertility to male sterility conditioned byT. timopheevi cytoplasm.Gli-B1 andRf3 were located on the short-arm satellite but recombine with the nucleolar organiser giving a gene order ofNor1 - Rf3 - Gli-B1. Glu-B1 is located on the long arm of 1B but shows relatively little recombination withNor1, which is, in physical distance, distal on the short arm. This illustrates the discrepancy between map distance and physical distance on wheat chromosomes due to the distal localisation of chiasmata. The recombination betweenNor1 andRf3 indicates that, contrary to previous suggestions, fertility restoration is not a property of the nucleolar organiser but of a separate locus.

Unexpectedly slow homogenisation within a repetitive DNA family shared between two subspecies of tsetse fly

Repetitive DNA families in sexual species are subject to a variety of turnover mechanisms capable of homogenising newly arising mutations. Very high levels of homogeneity in DNA families in some species of Drosophila indicate that the rate of turnover is fast relative to that of mutation. To gauge the generality of such phenomena, we cloned and sequenced individual members of homologous repetitive DNA families from two subspecies of tsetse fly, Glossina morsitans centralis and G. morsitans morsitans. Unexpectedly high levels of variation were found within each subspecies, averaging 24% and 31%, respectively. Contiguous repeats and repeats cloned at random were comparably divergent. Nevertheless, it was possible to identify three instances of apparent homogenisation, each being, remarkably, of an insertion/deletion nature. We conclude that the rate of turnover in the tsetse families is comparable to that of most mutations, and discuss the possible parameters affecting flux in these families.

The spread of sequence variants in Rattus satellite DNAs

The genus Rattus has two related families of satellite DNA: Satellite I consists of tandem arrays of a 370 base pair repeat unit which is a dimer of two 185 base pair portions (a, b) which are about 60% homologous. Satellite I' consists of tandem arrays of a 185 base pair repeat unit (a') which is about 85% homologous to a and 60% homologous to b. R. norvegicus contains only satellite I but R. rattus contains both satellites I and I'. We examined certain aspects of satellite DNA evolution by comparing the spacing at which variant repeat units of each satellite have spread among non-variant repeat units in these two species. With but one exception, in R. rattus, 15 different variant repeat units have spread among non-variant repeat units of satellite I, with a spacing equal to the length of the (a,b) dimer. Similarly, fourteen different variant repeat units of the monomeric satellite I' have mixed among non-variant repeat units with a spacing equal to the length of the (a') monomer. These results suggest that a mechanism involving homologous interaction among satellite sequences could account for the spread of variant family members. We also found that a sequence variant present in certain portions of the dimeric repeat unit of satellite I is more efficiently amplified (or less efficiently corrected) than variants occurring in other regions. This was not true for the monomeric repeat unit of satellite I'.

Structure and function of an AT-rich, interspersed repetitive sequence from Chironomus thummi: solenoidal DNA, 142 bp palindrome-frame and homologies with the sequence for site-specific recombination of bacterial transposons

Chironomus thummi thummi contains a repetitive AT-rich 118 bp sequence mainly in the centromere regions and elsewhere in the genome (1). A large cluster of repeats is regularly present in the non-transcribed spacer of rDNA. Dimers and multimers of the repeat migrate slower in small pore gels than would be expected from their size. The results indicate a solenoidal structure with a coil girth of appr. 350 bp. This structure is most probably due to a highly periodic positioning of di-nucleotides of the type purine - purine or pyrimidine-pyrimidine with distances of appr. 10 bases. In a cluster of 118 bp repeats, regions of dyad-symmetry are positioned such that a 142 +/- 2 bp palindrome-frame is generated. Evidence is presented favouring the assumption that the repeat functions primarily in sister chromatid exchange.

Premeiotic chromosome doubling after genome elimination during spermatogenesis of the species hybrid Rana esculenta

Gamete production in the hybridogenetic species hybrid Rana esculenta (Rana ridibunda X Rana lessonae) is preceded by a premeiotic elimination of the R. lessonae genome and subsequent duplication of the remaining R. ridibunda genome, so that only ridibunda chromosomes enter a quasi normal meiosis, and only ridibunda gametes are formed. This is demonstrated by differences in genome specific centromere fluorescence and electrophoretic patterns between somatic and gonadal tissue.

Nucleotide sequences of highly repeated DNAs; compilation and comments

The nucleotide sequence data from highly repeated DNAs of inverte-brates and mammals are summarized and briefly discussed. Very similar conclusions can be drawn from the two data bases. Sequence complexities can vary from 2 bp to at least 359 bp in invertebrates and from 3 bp to at least 2350 bp in mammals. The larger sequences may or may not exhibit a substructure. Significant sequence variation occurs for any given repeated array within a species, but the sources of this heterogeneity have not been systematically partitioned. The types of alterations in a basic repeating unit can involve base changes as well as deletions or additions which can vary from 1 bp to at least 98 bp in length. These changes indicate that sequenceper seis unlikely to be under significant biological constraints and may sensibly be examined by analogy to Kimura's neutral theory for allelic variation. It is not possible with the present evidence to discriminate between the roles ofneutralandselectivemechanisms in the evolution of highly repeated DNA.

Tandemly repeated arrays are constantly subjected to cycles of amplification and deletion by mechanisms for which the available data stem largely from ribosomal genes. It is a matter of conjecture whether the solutions to the mechanistic puzzles involved in amplification or rapid redeployment of satellite sequences throughout a genome will necessarily give any insight into biological functions.

The lack of significant somatic effects when the satellite DNA content of a genome is significantly perturbed indicates that the hunt for specific functions at thecellularlevel is unlikely to prove profitable.

The presence or in some cases theamountof satellite DNA on a chromosome, however, can have significant effects in the germ line. There the data show that localized condensed chromatin, rich in satellite DNA, can have the effect of rendering adjacent euchromatic regionsrec−, or of altering levels of recombination on different chromosomes. No data stemming from natural populations however are yet available to tell us if these effects are of adaptive or evolutionary significance.

Tissue specificity and organisation of CpG methylation in calf satellite DNA I

Examination of bovine satellite DNA I methylation within CpG dinucleotides has been made by restriction analysis. It is shown that variations in the methylation patterns occur between different tissues (brain, liver, thymus and sperm) . Some of the 8 Hpa II sites present per repeat are clearly undermethylated in sperm as compared to other tissues. Methylation is considered therefore, as a highly specific event. It is also shown that there is a spatial specificity in the methylation pattern of the 3 Hha I sites in all tissues. These results are discussed in the light of methylation and satellite DNA functions.

DNA methylation in eukaryotes

The DNA of higher eukaryotes contains one minor base, namely 5-methylcytosine. The distribution of this minor base between different species and different DNA fractions will be considered together with the actual sequences methylated. The properties of the enzyme responsible for DNA modification will be reviewed, particular note being paid to the efficiency of methylation of different DNA substrates. Various possible functions of the 5-methylcytosine in DNA will be considered and particular attention will be paid to the finding that specific modified bases present in DNA not undergoing transcription are absent in the same genes when these are being actively transcribed.

Male-sterilizing interactions between duplications and deficiencies for proximal X-chromosome material in Drosophila melanogaster

The genetic limits of sixty-four deficiencies in the vicinity of the euchromatic-heterochromatic junction of the X chromosome were mapped with respect to a number of proximal recessive lethal mutations. They were also tested for male fertility in combination with three Y chromosomes carrying different amounts of proximal X-chromosome-derived material (BSYy+, y+Ymal126 and y+Ymal+). All deficiencies that did not include the locus of bb and a few that did were male-fertile in all male-viable Df(1)/Dp(1;Y) combinations. Nineteen bb deficiencies fell into six different classes by virtue of their male-fertility phenotypes when combined with the duplicated Y chromosomes. The six categories of deficiencies are consistent with a formalism that invokes three factors or regions at the base of the X, one distal and two proximal to bb, which bind a substance critical for precocious inactivation of the X chromosome in the primary spermatocyte. Free duplications carrying these regions or factors compete for the substance in such a way that, in the presence of such duplications, proximally deficient X chromosomes are unable to command sufficient substance for proper control of X-chromosome gene activity preparatory to spermatogenesis. We conclude that there is no single factor at the base of the X that is required for the fertility of males whose genotype is otherwise normal.

Some calculations on the amount of selfish DNA

A population genetical theory is developed to treat the amount of selfish DNA in a genome. We assume that the selfish DNA consists of replicating units and that it evolves by multiplication within a genome, exchange between genomes, and random genetic drift at reproduction. Special reference is made to the mean and variance of the number of replicating units per genome in the population. Under the assumption of no systematic evolutionary pressure, the number of units changes randomly with time, and its variance increases by replication process. Although under certain circumstances the variance increases also by exchange process, under ordinary circumstances this process tends to decrease the variance. Random genetic drift also reduces the variance. The relationship between the mean and variance at equilibrium of the number of replicating units per genome in the population was derived. The results obtained will be useful in understanding various observations on repeated DNA which presumably does not contain genetic information and which is likely to be selectively neutral.

Similarity of centromeric heterochromatin in strains of drosophila melanogaster which interact to produce hybrid dysgenesis

Many long-established laboratory strains of D. melanogaster interact with recently-collected wild-type strains, so that the progeny show sterility, enhanced mutation, male recombination and other degenerative traits, a syndrome known collectively as "hybrid dysgenesis". Tests have been made for differences in centromeric heterochromatin between interacting strains, by comparing homologues in mitotic preparations from hybrid individuals. Differences between homologues have not been revealed, either qualitatively by C-banding or quantitatively by in situ hybridisation to RNA transcripts from satellites 1.705 and 1.686. However, the sensitivities of the techniques are such that quantitative differences of less than 50% between strains could escape detection.

Unequal crossing over in the ribosomal DNA of Saccharomyces cerevisiae

Unequal sister chromatid exchanges occur at the ribosomal DNA locus of yeast during mitotic growth. The frequency of unequal crossing over, as measured by the deletion or duplication of an inserted genetic marker (LEU2), is sufficient to maintain the sequence homogeneity of the rDNA repeat units.

Restriction endonuclease and molecular analyses of three rat genomes with special reference to chromosome rearrangement and speciation problems

When differences are found between related species of organisms, it is often assumed that the differences themselves are causal factors either in speciation itself or in processes related to speciation. Two recent proposals on the functions of satellite DNA (Hatch et al., 1976 and Fry and Salser 1977) are that (a) large amounts of satellite DNA are important in facilitating chromosome rearrangements and hence cytogenetic evolution, and (b) satellite DNA differences between homologous chromosomes lead to pairing difficulties and are important in generating infertility barriers and hence speciation. If these proposals were to have some generality, one could expect organisms with very low amounts of highly repeated DNA to exhibit few chromosome rearrangements and to be evolutionarily conservative in a cytogenetic sense.--We have chosen two very closely related species of rat which are phenotypically almost indistinguishable and which have undergone massive genome reorganization. They differ by 11 major centric rearrangements (2n = 32, 2n = 50). We have characterised their genomes by restriction endonuclease digestions, thermal denaturations, analytical ultracentrifugations and reassociation techniques, and have found that they have virtually no highly repeated DNA. Thus the 11 major chromosomal rearrangements have been fixed in present day genomes with hardly any highly repeated DNA, centric or otherwise.--It appears therefore, that a large amount of highly repeated DNA is not obligatory for the formation and fixation of chromosome rearrangements. In addition, the existing literature reveals that one can find almost any situation at all, from species groups with high amounts of satellite DNA and no gross chromosomal rearrangements, to ones such as those described here, with tiny amounts of highly repeated DNA and massive chromosomal reorganisation. Since direct experimental data indicates that satellite DNA differences per se between homologous chromosomes do not cause infertility, speculations concerning modes of speciation based on satellite DNA differences between otherwise homologous chromosomes would appear to be ill founded.

In vivo sister chromatid exchange in cells of various organs of the mouse

In vivo sister chromatid exchange (SCE) in mouse cells derived from various organs was studied by infusing BrdU from the tail vein. It was found that at BrdU concentrations ranging from 2.2-13.5 micrograms/g/h, the SCE frequency in bone marrow cells seemed to stay at a constant level (1.5-2/cell/two cell cycles) whereas it started to rise as the BrdU dose exceeded this dose range. When BrdU within this dose range was infused continuously from the tail vein for appropriate hours to label chromosomes in various organs, the average SCE frequencies per cell were found to be 1.64 in bone marrow cells, 1.82 in spermatogonia, 1.99 in splenic cells, 2.89 in intestinal cells and 3.69 in cells from adjuvant stimulated lymph nodes. It is suggested that the spontaneous level of the in vivo SCE frequency might be about 1.5-2/cell/two cell cycles in the mouse. In cells derived from intestine and adjuvant stimulated lymph node, some unknown factors might work as a inducer of SCEs resulting in a significant increase in the SCE frequency in these organs.

Preferential occurrence of sister chromatid exchanges at heterochromatin-euchromatin junctions in the wallaby and hamster chromosomes

Chromosomes of two mammalian species, the white-throated wallaby and the rat-like hamster, possessed large amounts of constitutive heterochromatin which is detectable as C bands. By making use of this character the frequency of sister chromatid exchanges (SCEs) was determined for the C band and the euchromatic regions of the chromosome. In both species, the distribution of SCEs in the euchromatin of chromosomes was found to be proportional to its metaphase length, while the number of SCEs localized in the C band regions was clearly fewer than expected on the basis of the relative length of those regions at metaphase. Many SCE's were, however, detected at the junctions between the euchromatin and the C band heterochromatin. All of these findings were consistent with previous observations on the Indian muntjac and the kangaroo rat chromosomes.

Detection and location of three simple sequence DNAs in polytene chromosomes from virilis group species of Drosophila

In vitro synthesized RNAs complementary to the three satellite DNAs of Drosophila virilis have been used in a series of in situ hybridization experiments with polytene chromosomes from virilis group species. Gall and Atherton (1974) demonstrated that each of the satellites of D. virilis is comprised of many repeats of a distinct, seven base pair long, simple sequence. With few exceptions, copies of each of these simple sequences are detected in the chromocenters of all virilis group species. This is true even in species which do not possess satellite DNAs at buoyant densities corresponding to those of the satellite DNAs of D. virilis. Small quantities of the three simple sequences are also detected in euchromatic arms of several different species. The same euchromatic location may contain detectable copies of one, two, or all three simple sequence DNAs. The amounts of simple sequences at each location in the euchromatin may vary between species, between different stocks of the same species, and even between individuals of the same stock. The simple sequences located in the euchromatin appear to undergo DNA replication during formation of polytene chromosomes unlike those in heterochromatin. The locations of the euchromatic sequences are not the results of single chromosomal inversion events involving heterochromatic and euchromatic breakpoints.

Hybrid ape offspring of a mating of gibbon and siamang

The serendipitous mating of a male gibbon, Hylobates moloch, and a female siamang, Symphalangus syndactylus, has produced two female offspring born 1 year apart. The hybrid karyotype of 47 chromosomes comprises the haploid complements of the parental species, 22 for the gibbon and 25 for the siamang. Chromosomal G and C banding comparisons revealed no clear homologies between the parental karyotypes except for the single chromosome in each species containing the nucleolus organizer region. The lack of homology suggests that the structural rearrangement of chromosomes has played a major role in the process of speciation for these lesser apes.

Cytological studies of heterochromatin function in the Drosophila melanogaster male: autosomal meiotic paring

In Drosophila melanogaster it is now documented that the different satellite DNA sequences make up the majority of the centromeric heterochromatin of all chromosomes. The most popular hypothesis on this class of DNA is that satellite DNA itself is important to the pairing processes of chromosomes. Evidence in support of such a hypothesis is, however, circumstantial. This hypothesis has been evaluated by direct cytological examination of the meiotic behaviour of heterochromatically and/or euchromatically rear-ranged autosomes in the male. It was found that neither substantial deletions nor rearrangements of the autosomal heterochromatin cause any disruption of meiotic pairing. Autosomal pairing depends on homologs retaining sufficient euchromatic homology. This is the first clear demonstration that the highly repeated satellite DNA sequences in the heterochromatin of the second, third and fourth chromosomes are not important in meiotic pairing, but rather than some euchromatic homology in the autosome is essential to ensure a regular meiotic process. These results on the autosomes, when taken in conjunction with our previous studies on sex chromosome pairing, clearly indicate that satellite DNA is not crucial for male meiotic chromosome pairing of any member of the D. melanogaster genome.

Satellite DNA and heterochromatin variants: the case for unequal mitotic crossing over

Variations of constitutive heterochromatin (heteromorphisms) appear to be a general feature of eucaryotes. A variety of molecular and cytogenetic evidence supports the hypothesis that heteromorphisms result from unequal double-strand exchanges during mitotic DNA replication. Constitutive heterochromatin consists of highly repeated DNA sequences that are not transcribed. Thus, heteromorphisms are tolerated without overt phenotypic effect. Several of the highly repeated DNAs that comprise constitutive heterochromatin have been shown to contain site-specific endonuclease recognition sequences interspersed at regular intervals dependent upon nucleosome structure. These interspersed short repeated sequences could mediate unequal crossovers, resulting in quantitative variability of constitutive heterochromatin and satellite DNA. De novo variations of constitutive heterochromatin may be useful as markers of exposure to mutagens and/or carcinogens.

Absence of satellite DNA synthesis during meiotic prophase in mouse and human spermatocytes

Mouse spermatocytes were labelled in situ with 3H-thymidine at successive stages of meiosis. Isolated mouse as well as human spermatocytes were similarly labelled under in vitro conditions. DNA synthesis was followed either by tracking radioactivities in Cs2SO4 gradients or by measuring reassociation kinetics. Mouse satellite DNA and the 3 satellites of human DNA are labelled during S-phase but not during pachytene. In the mouse genome, there is a preferential labelling of regions containing foldbacks (human spermatocytes were not analyzed in this respect). The absence of detectable pachytene synthesis in satellite DNA is consistent with genetic evidence on the absence of crossing-over in constitutive heterochromatin.

The distribution of satellite and main-band DNA components in the melanogaster species subgroup of Drosophila. I. Fractionation of DNA in actinomycin D and distamycin A density gradients

Fractionation of total adult DNA of five of the seven species of the melanogaster species sub-group of Drosophila in actinomycin D and distamycin A caesium density gradients has revealed the presence of three main-band DNA components, common to all species, and ten satellite DNAs that are distributed between the species. Satellite DNAs are either unique to a species or common to two or more species. The abundance of a common satellite DNA varies between species. There is no simple relationship between the presence of a satellite DNA and a branch point of phylogenetic divergence; nevertheless the arrangement of the species in a phylogeny that is based on the numbers of satellites held in common accurately reflects the pattern of relationships between the same species based on differences in inversions of polytene chromosomes. The species can be similarly arranged according to the compositions of their mitochondrial DNAs. It is possible that the same basic set of sequences, each of low frequency, is common to all species with arbitrary or selected amplification of particular sequences to differing extents in individual species. The conservation of satellites in the group and the close parallel between the distributions of satellites and inversions between the species suggests that either the processes that operate to change both chromosomal phenomena are similarly time-dependent and occurring at relatively low rates or that their rates of change are restricted according to some undetermined functions of these aspects of the genome.