Gypsy is a 7kb element with a structure similar to retroviral proviruses featuring LTRs and gag, pol and env genes, features presumably indicative of a transposition mechanism involving reverse transcription and insertion of the cDNA. Retrotransposition has been difficult to study in metazoans because of a low rate of insertion. However, progeny of females homozygous for the X-linked mutation flamenco (flam) undergo high rates of gypsy insertion which can be detected by reversions of the dominant female sterile ovoD mutation (the ovo gene has proved to be a hotspot for retrotransposon insertions) and this has permitted systematic study of the molecular events associated with gypsy insertion. Dr. Corces reports a number of interesting findings using the flam strain. Principally these are: 1) gypsy RNAs are made in follicle cells and some other somatic tissues but are not present in oocytes, suggesting that insertion requires transport or infection from soma to germ line; 2) the insertion events detected in the ovoD assay occur relatively late in the germ line of the progeny (because they are not clustered) and not in somatic cells, indicating a second level of control over processing the RNA into an insertion 3) gypsy mobility is associated with presence of env polypeptides (and a corresponding 2kb RNA) in follicle cells of mothers suggestive of a virus-like stage; 4) virus-like particles can be seen in EM preparations both in fractionated extracts of flam ovaries and in the membranes of stage 10 follicles; moreover the same extract fraction that contains particles also has measurable reverse transcriptase activity; 5) feeding extracts from active strains to flam (but not +) flies lacking endogenous active elements leads to high rates of insertional mutations in their progeny, indicating that the virus particles are infectious and are probably taken in through the gut; 6) the new insertions involve many other retrotransposons besides gypsy, indicating that gypsy infection triggers a general mobilization of retrotransposons, and suggesting that the limiting component may be common to all the retrotransposons. The model favored by Dr. Corces is that infectious particles are generated in follicle cells (in the absence of flam+ protein which probably acts by degrading the RNA); they infect oocytes, losing env protein in the process, then the RNA becomes incorporated into pole cells where it is reverse transcribed and inserted at a later stage. Several viable alternatives are also considered. The proposed experiments focus on 5 aims. The first is to assay for biological activity of the gypsy env protein by determining whether it can substitute for the corresponding protein of a retrovirus, and by engineering a mutation in the env coding region of a solo gypsy element and testing for loss of activity. In addition the env protein will be expressed from a heat-shock promoter at various stages to determine if it can overcome flam+ inhibition of native gypsy elements. If not, constructs expressing other gypsy activities such as integrase, reverse transcriptase, and RNaseH under heat shock control will be added to try to identify the limiting components in flam+ genotypes. The second is to characterize the timing of insertions by PCR and to try to correlate that information with the timing of expression of the various proteins using antibodies. The third is to assess the precise role of virus particles by using antibody staining of EM sections to search for the particles in oocytes, in association with yolk granules (to test the idea that gypsy is transferred from follicles in association with yolk) and in the perivitelline space following fertilization, and whether insertion requires yolk transfer from follicles to oocytes. The fourth is to assess the mechanism of gypsy mobilization of other retrotransposons by testing for associations of gypsy env protein with the RNAs of the other elements which would be indicative of the formation of hybrid particles and of env protein being a generally limiting factor, and to test for stimulation of insertions by overexpression of gypsy integrase which would be consistent with integrase being a common limiting function. Finally purified gypsy particles will be fed to flies of other Drosophila species as a first step toward developing gypsy as a general insect transformation vector.
