Altering Egg Morphology

In mutagenesis screens for recessive female sterile mutations on the second chromosome of Drosophila melanogaster 528 lines were isolated which allow the homozygous females to survive but cause sterility. In 62 of these lines early stages of oogenesis are affected, and these females usually do not lay any eggs. In 333 lines oogenesis proceeds apparently normally to stage 8 of oogenesis, but morphological defects become often apparent during later stages of oogenesis, and are visible in the defective eggs produced by these females whereas 133 lay eggs that appear morphologically normal, but do not support normal embryonic development. Of the lines 341 have been genetically characterized and define a total of 140 loci on the second chromosome. Not all the loci are specific for oogenesis. From the numbers obtained we estimate that the second chromosome of Drosophila contains about 13 loci that are relatively specific for early oogenesis, 70 loci that are specifically required in mid to late oogenesis, and around 30 maternal-effect lethals. 0 OGENESIS is a highly regulated developmental process. In insects, as in dany metazoans, it involves the close cooperation between the germ cells and a number of somatic cell types. Genetic and morphological studies of mutations that affect oogenesis have revealed that in Drosophila melanogaster a large number of genes are active during oogenesis to ensure the production of a normal egg capable of supporting the development of a normal embryo (GARCIA-BELLIDO and ROBBINS 1983; PERRIMON, ENCSTROM and MAHOWALD 1984). These genes can be grouped in several classes. On the one hand, all cells involved in oogenesis require the normal complement of household genes that allow those cells to grow and divide. They also require some more specialized gene functions which are also expressed in other tissues. Because of their pleiotropic effects outside the ovary, mutations in all of these genes would usually cause lethality in homozygous individuals. A smaller, third group of genes are required more specifically for processes that only occur during oogenesis. Mutations in these genes will allow the homozygous carrier female to survive, but cause defects in oogenesis. These mutations can be used to analyze various aspects of pattern formation and differentiation processes unique to oogenesis. Smaller, previous screens for such female-sterile mutations were focussed mostly on genes on the X chromosome. These screens identified classes of female-sterile mutations that seem to affect particular processes in oogenesis (GANS, AUDIT and MASON Genetics 129 1 1 19-1 136 (December, 1991) 1975; MOHLER 1977; KOMITOPOULOU et al. 1983; PERRIMON et al. 1986; ORR et al. 1989). This present work describes the results of a large-scale mutagenesis experiment on the second chromosome of Drosophila and deals specifically with female sterile mutations affecting oogenesis. The larger number of mutations obtained in this screen allows us to compare the relative number of mutations and loci affecting various developmental processes that occur in oogenesis. These comparisons suggest that only few genes in the genome are exclusively required for the survival of germline or gonadal primordia in the stages before adulthood. Another small group of genes appear to be required specifically for the initial establishment and maintenance of the 15 + 1 nurse cell-oocyte cluster. On the other hand, a larger number of genes appear to be specifically involved in follicle cell functions such as secretion of the egg shell, or the migration patterns of the follicle cells within the developing egg chamber. MATERIALS AND METHODS Drosophila stocks and genetic crosses: The female-sterile mutations described in this work were all isolated after EMS mutagenesis of males carrying isogenized second chromosomes, and testing the Fs generation for fertility, as previously described (SCHUPBACH and WIESCHAUS 1989). These screens led to the isolation of 528 mutant lines [the number of 529, as reported in SCHUPBACH and WIFSCHAUS (1989), was adjusted after one of the lines was subsequently found to be fertile]. All marker mutations, chromosomal rearrangements and balancer chromosomes used in the 1120 T. Schupbach and E. Wieschaus experiments are described and referenced in LINDSLEY and GRELL (1968) or LINDSLEY and ZIMM (1985, 1986, 1987, 1990), unless otherwise indicated. Genetic mapping of the mutations was performed using an a1 d p b pr c p x sp chromosome (“all-chromosome”) or in some cases a chromosome carrying the dominant markers S Sp B1, and using, in addition, the markers cn and bw which are present on all mutagenized chromosomes. Once a female sterile mutation had been placed between two marker mutations, a larger set of additional recombinants between the two markers was isolated and tested in a second recombination experiment. This yielded a genetic map position with an error interval of less than 5 map units (P = 5%). After a particular mutation had been mapped, all chromosomal deficiencies within 10 map units on either side were tested in t rans to the mutation (Table 1). Finally, all mutations mapping between the same two markers were complemented against each other in order to find all members of a complementation group, regardless of the female sterile phenotype. In addition, a set of 14 deficiencies (indicated with an asterisk in Table 1) were crossed to all 528 sterile lines, in order to find all the mutations in the collection that are uncovered by these deficiencies. This analysis revealed that the gene vasa was actually represented by six rather than two alleles as previously reported (SCHWPBACH and WIFSCHAUS 1989), and, given the phenotype of the additional alleles, the gene had to be placed into the group of female sterile mutations affecting mid to late oogenesis. In addition, the gene previously designated as mat(2)cellHK35 was found to be allelic to the gene squash which also affects mid to late oogenesis. The mutation designated as splicedRL3 was found to be allelic to torso by KLINGLER et al. (1 988), and by STRECKER et al. (1989). As compared to the previous report, the class of maternal effect lethals has therefore been reduced to 133 lines (64 loci) from the previously reported 136 lines (67 loci). Phenotypic descriptions: Eggs were collected, fixed, and mounted in Hoyer’s medium for inspection with a compound microscope, as described in WIFSCHAUS and NUSSLEIN-VOLHARD (1986). For mutant lines that did not lay any eggs, ovaries of several females were dissected, fixed for 5-10 min in 2% glutaraldehyde, transferred to phosphate-buffered saline (PBS) and after further dissection of ovarioles and egg chambers, the ovaries were inspected with phase contrast optics, or DIC optics. The ovaries of selected mutations were stained with Hoechst stain for 5 min (1 rg/ rnl in PBS, after fixation with glutaraldehyde) and inspected with a fluorescence microscope. Enhancer trap lines ( i e . , lines carrying insertions of engineered P elements with lac2 inserts (BIER et al. 1989) which show expression patterns in ovarian cell types were crossed to some of the mutant lines, and the resulting lines were stained for &galactosidase expression as described by FASANO and KERRIDGE (1 988)l. The enhancer trap lines were obtained in our laboratory (T. SCHUPBACH, L. J. MANSEAU and E. SHADDIX, unpublished results) after mobilization of a P element insert containing the lac-Z gene on the Cy0 chromosome, which had been provided to us by the laboratories of L. Y . JAN and N. Y. JAN.