Search

The P transposable element copy numbers and the KP/full-sized P element ratios were determined in eight Brazilian strains of Drosophila melanogaster. Strains from tropical regions showed lower overall P element copy numbers than did strains from temperate regions. Variable numbers of full-sized and defective elements were detected, but the full-sized P and KP elements were the predominant classes of elements in all strains. The full-sized P and KP element ratios were calculated and compared with latitude. The northernmost and southernmost Brazilian strains showed fewer full-sized elements than KP elements per genome, and the strains from less extreme latitudes had many more full-sized P than KP elements. However, no clinal variation was observed. Strains from different localities, previously classified as having P cytotype, displayed a higher or a lower proportion of KP elements than of full-sized P elements, as well as an equal number of the two element types, showing that the same phenotype may be produced by different underlying genomic components of the P-M system.

In order to study P element dynamics in natural populations of Drosophila melanogaster, 126 isofemale lines were examined from seven of the southernmost islands of Japan (the Sakishima Islands) and from Taiwan. Gonadal dysgenesis (GD) tests showed large divergences in the P-M phenotypes (P inducing and P repressing abilities) between the island populations. The P-M characteristics of each population, however, had not greatly changed in the past 15 years. Their genomic P element profiles are highly similar, consisting mostly of full-size P and of KP elements. We found no clear relationship between phenotype and genomic P element composition.

The P family of transposable elements in Drosophila melanogaster has, since its introduction into D. melanogaster populations in this century, diversified into a number of internally deleted forms. One of these, the KP element, is abundant in the genomes of flies from populations from Asia, Europe and Africa. There is some evidence that this sequence can act as a repressor of transposition. We have studied a mutation, signed(very weak) (snVW), in which a KP element is one of two P elements inserted into the 5' exon of the X-linked singed gene. These elements can be mobilized by a trans-acting dominant mutation, Mo, with a maternal effect, on the second chromosome. The rate of somatic reversion of snvw induced by Mo is reduced threefold if the mother herself possesses snvw on her X chromosomes. This implies that snvw may be responsible for a form of repression of P element movement. The cause of this effect may be related to transcription of the KP element in snvw. However, an effect of other genomic P elements in the repression of somatic reversion of snvw cannot be entirely excluded.

To characterize the relationship between P element activities and their structures, we cloned P elements from genomic libraries of three isogenic P and Q strains derived from natural populations in Japan. These P elements were mapped with BamHI, AvaII and PstI and were classified by their size. The majority of P elements cloned were classified as either complete or relatively small P elements rather than medium size. The numbers of full length (2.9 kb) P elements per haploid genome of NP280 (P), AK194 (weak P) and WY113 (Q) were at least four, five and one, respectively. However, the 2.9 kb P element of WY113 was thought to be defective since this strain has no transposase activity. In our previous work, we demonstrated that the ORF 3-deleted P element is essential for P cytotype determination in WY113. A similar P element also exists in NP280, and this may have an important role for P cytotype determination in this strain. Two and one copies of the KP element, a deletion derivative of the P element, were found in NP280 and AK194, respectively. One of four complete P elements in NP280 was fully sequenced, and the base sequence was completely identical to that of p pi 25.1 originally derived from the U.S.A. This result is consistent with the notion that these P elements have a relatively recent origin in Drosophila melanogaster.

In order to clarify the cause of null mutations in enzyme loci, the molecular structure of six null mutations in the Gpdh locus (encoding alpha GPDH: alpha glycerol-3-phosphate dehydrogenase (NAD+), E.C. 1.1.1.8; map position at 2-17.8) that arose in mutation accumulation experiments was examined. A restriction map analysis showed that five of the mutations are insertional mutations whereas the sixth is a deletion. The Gpdh regions of these null mutations were then cloned and sequenced. The inserted DNA fragments are all internally deleted P elements measuring 1.1 kb in length. Two are a KP element and two others are a HP element. All the insertions occur in the region near the initiation signal of transcription. The deletion encompasses the seventh and eighth exons over a length of 1.1 kb. These results therefore indicate that the null mutation rate at the Gpdh locus is largely influenced by P elements.

P elements, both complete and incomplete, contain a left and right end, normally depicted as pointing away from each other. Here, we examine the properties of P elements that may be described as ‘re-arranged elements’ or ‘inside-out elements’, containing inverted ends. Two such structures exist, having either ends pointing towards each other, ‘head-to-head or H–H’, or ends pointing in the same direction ‘head-to-tail or H–T’. We show that both structures are unstable in the presence of P element transposase. For the H–H element there is a high frequency of deletion of the intervening material and almost exact rejoining of element ends with the 4 bp CATG palindromic end sequence shared by the two element ends. This result is predicted by the Beall and Rio model of P element excision. For the H–T element there is a high frequency of exact excision of the entire inverted right-end, a result again predicted by the Beall and Rio model. Both structures lead to recombination in the way expected from a normal element. The rates of recombination are, however, much lower than might be expected from the organization of ends, a result that can be explained in terms of the low likelihood of insertion into a chromosomal region lacking another P element end. We also investigate the properties of combinations of re-arranged and normal elements, and show that there is a directionality property when left and right ends are combined in trans that can be explained in terms of strand repair.

The fruitfly, Drosophila melanogaster, can be transfected with P-elements and induced to overexpress a transfected gene whose impact on lifespan can be measured. Here, it is reported that in previous experiments a transfected gene was not expressed. This suggested a new statistical analysis indicating that (1) the size of the insert, the position and the interactions of the insert with the genetic backgrounds into which the P-element are inserted have effects on lifespan similar to those attributed to overexpression; (2) these effects occur without expression of the transfected gene; and (3) effects of interactions with backgrounds and effects of positions are as large as responses to six generations of strong directional artificial selection. Reports of effects of overexpression of transfected genes on lifespan in Drosophila melanogaster may be experimental artefacts. Credible experiments on the phenotypic effects of trans-genesis need proper controls for the effects of insert size and position and should estimate the magnitudes of interactions of treatment with genetic backgrounds.

Using data from in situ hybridisation of the giant chromosomes from salivary glands, mdg-1, I, copia, and P mobile element polymorphism was studied in 17 highly-inbred lines of Drosophila melanogaster and in the expected hybrids obtained from the theoretically-crossed inbred lines. The mean copy numbers of each element on the inbred lines were close and equal to 16·8 for mdg-1, 17·5 for I, 17·8 for copia and 18·6 for P (these values included the insertions in the centromeric regions). The P element differed from the other three for variance in copy number and distribution of the number of insertion site occurrences. A low frequency of X-linked copia element insertions (as compared with frequency of insertions in the other chromosomes) was reported, suggesting that natural selection acts against insertional deleterious mutations of this element. Such a low frequency of X-linked insertions was not observed for the other three elements. The mdg-1, I and copia element copy numbers were positively correlated, thus leading to lines with high or low copy number. No correlation was observed between the P element copy number and the numbers of mdg-1, I and copia elements. Moreover, a genomic control of copy number was detected for mdg-1, I and copia, but not for P. The number of labelling sites, the number of copies, the number of heterozygous insertion sites, and various indices of genetic variability were estimated from the expected hybrids obtained from the theoretically-crossed inbred lines; the calculated numbers of labelling sites for mdg-1 and I did not differ from those obtained directly from the initial wild-derived population (we had no estimate of the number of labelling sites available in this population for copia and P). An average proportion of heterozygous sites of 0·86, 0·92, 0·90 and 0·94, and an average number of copies of 33, 35, 38 and 37 were found in the theoretical hybrids for the mdg-1, I, copia and P elements, respectively.

Hybrid viability is correlated with the I and P mobile element copy numbers of the maternal inbred line in Drosophila melanogaster

In order to simulate the outcome of the P-M status of Eurasian populations of Drosophila melanogaster, the evolution of experimental mixed-strains was monitored for up to 50 generations. The results were compared with the evolution in natura of European populations sampled in 1981-83 and 1986-87 over a similar period of time. Different combinations of P and M' strains, Q and M' strains, M' and M' strains and M and M' strains were set up at 25 degrees C and duplicated at 13 degrees C night-21 degrees C day. The possibility of a change towards a P type only appeared with the introduction of the strong P strain Harwich into Eurasian strains. Strains with strength similar to that of Harwich are not currently found in wild populations. With the introduction of weak P and Q strains of the strength presently observed in western Europe, experimental populations evolved slowly towards a Q state or a weak M' state. M'-M' mixed populations resulted in strong M' strains, as was the case for M-M' populations. In these cases. P sequences were not eliminated. In wild populations both genetic and molecular analyses showed no significant differences, over a five year period, for GD sterility potentials, for total P copy number or for distribution of the full-sized and KP elements. Changes in Eurasian populations are probably taking place at a very low rate and may even have stopped, leading to a quasi-stable differentiation over the continent. During the different steps of progressive invasion of P transposable elements, several deleted elements developed and natural selection may have acted on them. The possibility of the selection of different types of regulatory mechanisms according to the presence of different kinds of derivative elements, leading to a world-wide differentiation between P-Q and M' strains is discussed.