ABSTRACT
Developing egg chambers of Drosophila melanogaster (wild-type and bobbed mutants) have been examined for their nucleic acid content by cytophotometric methods. No differences were observed in the total DNA and RNA content of the egg chambers at all stages between the bobbed mutants and the wild type. It is shown that the process of oogenesis in bobbed females is prolonged, and that this prolongation occurs at all the stages of oocyte development. Since the ovaries of the bobbed females synthesize less rRNA per unit time, it is likely that this prolongation allows the egg chambers of the bobbed females to normalize their RNA content. When they achieve a given RNA content, they proceed to the next stage of development.
INTRODUCTION
The bobbed syndrome in Drosophila (slow development, short bristles and etching of the abdominal tergites) is caused by a reduction in the amount of the DNA complementary to ribosomal RNA (Ritossa, Atwood & Spiegelman, 1966). In most cases, the severity of the bobbed phenotype is inversely proportional to the amount of DNA complementary to ribosomal RNA (rDNA). Ritossa et al. (1966) proposed that the basis of the bobbed phenotype could be due to a limitation of rRNA synthesis. Following this proposal, we have recently shown (Mohan & Ritossa, 1970) that the rate of rRNA synthesis is reduced in ovaries of bobbed flies which have rDNA content lower than 0·130 %. These flies lay eggs at a reduced rate, but of normal RNA content. Ovaries from phenotypically wild flies with rDNA contents of 0·180 %, 0·370 % and 0·580 % show, instead, the same rate of rRNA synthesis and lay eggs at a normal rate and of normal RNA content. In spite of the reduced rate of rRNA synthesis, bobbed flies (both males and females) have the same RNA/DNA ratios as the phenotypically wild flies. The normalization of the rRNA content probably occurs due to the delayed development of the bobbed flies. On the basis of these results, we proposed that the normal ribosome content has to be achieved in tissues most relevant for the development of the animal. Other less relevant tissues or cells can develop even with subnormal amounts of ribosomes; they will lead to the mutated phenotype. The purpose of the present paper is to present results concerning this proposal. To do so, we have measured cytophotometrically the RNA and DNA contents of developing egg chambers stage by stage, as classified by King, Rubinson & Smith (1956). Uridine incorporation into RNA has also been followed by autoradiography. We have been particularly interested in knowing whether or not the developing oocytes of bobbed flies can proceed to their next stage of development without having a normal content of ribosomes. Our results will show that this is clearly not the case; additional data indicates that there is a general delay of all the stages of the oogenetic process.
MATERIALS AND METHODS
(a) Drosophila stocks
Wild-type Canton S was obtained from the University of Pavia. g2tybb/C(1)DX, yf was from Pasedena; wabb1/BsY and In(l) sc4L, 8R, y sc4+8cvvf were from Oak Ridge National Laboratory. XYL. Ys (108–9), Y2su(wa)waYLYs/yvf was obtained from the University of Rome. These Drosophila stocks were kept in different genotypic combinations so that the bobbed flies could be generated in one or two crosses. The stocks were never kept in the bobbed condition as such because of the labile nature of this locus. So the bobbed flies were obtained prior to the experiments by crossing parents carrying both bb+ and bb− chromosomes. The flies which first emerged were not strongly bb and they were usually discarded. The flies emerging later were strongly bb and were employed for the experiments. Only heterozygous flies were employed in these experiments because homozygous flies of various strains have a tendency to retain mature eggs and sometimes contain non-functional egg chambers (David & Merle, 1968).
The percentage of rDNA is given in brackets after each genotype in the following description. Biochemical techniques regarding the estimation of rDNA content have already been described (Mohan & Ritossa, 1970). Flies having rDNA content lower than 0-130 % are phenotypically bobbed.
(b) Cytological procedures
Seven-day-old mated flies unless otherwise specified were employed. Ovaries were dissected in a drop of Ringer solution. When twenty-five ovaries had been collected in this way, they were fixed in Serra solution (ethyl alcohol, 6 parts; formaldehyde, 3 parts; acetic acid, 1 part) for 30 min, dehydrated in a series of alcohol and embedded in paraffin wax. Sections of 10 μ thickness were always used.
DNA staining was done with the standard Feulgen method. The acid hydrolysis was, however, carried out in 5 N-HCl at room temperature, since this gives more intense staining (De Crosse & Aiello, 1966) than hydrolysis at 55–60 °C in 1 N-HCl. RNA staining was done with the gallocyanin chrome alum method (Sandritter, Kiefer & Rick, 1966). A Barr and Stroud integrating microdensito-meter was used to measure the nucleic acids quantitatively. Control slides treated with DNase (0·25 mg/ml at pH 6·0 containing 0·03 M-SO4 Mg for 3 h at 37 °C) and RNase (0·5 mg/ml at pH 6·0 for 2 h at 37 °C) were employed for all experiments.
(c) Autoradiography
Five-day-old flies of various genotypes were fed on a radioactive medium containing 5 g dead yeast, 2 g corn, 1 g sugar and 0·2 ml of 10Ci/ml [3H]uridine (1 Ci/mM). The flies were removed after 2 days and sections of the ovaries were prepared as described before. Slides were covered with K2 Ilford emulsion and exposed for 10 days. The sections were stained with methyl green pyronin.
Whole mounts of 0–, 1–, 3– and 7–day-old ovaries were done for both Feulgen reaction and aceto-orcein squashes.
RESULTS AND DISCUSSION
The development of an egg chamber has been subdivided into a series of 14 stages by King et al. (1956). However, this development is a continuous event and its separation into a series of discrete stages was made by relying on multiple morphological criteria. The characteristics used by King et al. (1956) have been employed to allocate the egg chambers to the various stages. A chamber showing characteristics of two consecutive stages is placed at the stage which includes the majority of the morphological criteria observed for that chamber.
The DNA and RNA content of developing egg chambers was measured cytophotometrically in three genotypes and the results are shown in Table 1. To overcome technical difficulties, the DNA and RNA content was measured on whole egg chambers (only in the first six stages) and not on individual nuclei. It has not been considered worth while to measure the nucleic acid content of the egg chambers in the last four stages during which the nurse cells have started to degenerate. It is obvious from Table 1 that a gradual increase takes place until the 5th stage in the DNA content in all the three genotypes. Since the measurements were made on 10μ thick sections in all cases, the figures do not correspond exactly to the levels of ploidy. The same seems to be the case for the RNA content in the first four stages, but no difference can be seen between 5th and 6th stage. This is probably due to the fact that the volume at these stages has increased without any apparent change in the quantity of RNA per unit volume. Although these figures cannot be used for absolute purposes, they are of value for comparison between the various genotypes. Regarding the DNA content at these stages, one can say confidently that there is no difference between the various genotypes. The minor differences found could be due to the fact that, at the same stage, there are differences in size since each of these stages lasts for a long time. The differences observed between the RNA contents also are of the same order of magnitude and should thus be regarded as insignificant.
RNA synthesis in developing oocytes of all stages has also been followed by autoradiography in various genotypes. As shown by the controls with RNase digestion, most of the tracer in sections is present in the form of RNA. One can see, from Figs. 1 and 2, that RNA synthesis occurs through all stages of oocyte development, but that its intensity varies between different stages. There are more grains over the nurse-cell nuclei than over the oocyte nucleus; but the oocyte cytoplasm also contains many grains. Similar observations have already been made by King & Burnett (1959), Sirlin & Jacob (1960) and by Zalokar (1960) for wild-type Drosophila. In all the genotypes, the radioactivity of the cytoplasm, both in the nurse and egg cells, rises continuously. The label moves along the streams of the ooplasm which are continuous with the nurse cell cytoplasm, indicating a flow of RNA towards the oocyte (Figs. 1, 2). No difference was seen in the amount of grains present in the cytoplasm of oocytes having different genotypes. In some sections, more tracer is found moving towards the oocyte from the nurse cells in bobbed females, suggesting that the nurse cells contribute comparatively more to the developing oocytes in the bobbed females than in the wild type.
Autoradiographs of sections 10µ thick of bb+/bb+ (0-330) 7-day-old ovaries. (A) Three egg chambers of stages 1, 3 and 4. The grains are distributed over all areas, × 2400.
(B) An egg chamber of stage 7. Most of the grains are seen in the regions adjoining the neighbouring cells. This pattern of grain distribution is not so obvious before this stage, × 2800.
(C) An egg chamber of stage 9. The pattern of grain distribution is similar to the one observed in (B). There is a continuity of grains from the nurse cell cytoplasm to the ooplasm. Small arrow marks the nurse cell and the large one the oocyte, × 2160.
Autoradiographs of sections 10µ thick of bb+/bb+ (0-330) 7-day-old ovaries. (A) Three egg chambers of stages 1, 3 and 4. The grains are distributed over all areas, × 2400.
(B) An egg chamber of stage 7. Most of the grains are seen in the regions adjoining the neighbouring cells. This pattern of grain distribution is not so obvious before this stage, × 2800.
(C) An egg chamber of stage 9. The pattern of grain distribution is similar to the one observed in (B). There is a continuity of grains from the nurse cell cytoplasm to the ooplasm. Small arrow marks the nurse cell and the large one the oocyte, × 2160.
Autoradiographs of sections of wabb1/XYL. Ysbb (0·100) 7-day-old ovaries. (A) Left egg chamber is in stage 7 and the right one in stage 8; one can see more clearly an accumulation of grains on the adjoining parts of the nurse cells in the right egg chamber. × 3600.
(B) An egg chamber of stage 10. Continuity of grains from nurse cell to oocyte is observed as described earlier for wild type. Small arrow marks the nurse cell and the large one the oocyte. × 3190.
Autoradiographs of sections of wabb1/XYL. Ysbb (0·100) 7-day-old ovaries. (A) Left egg chamber is in stage 7 and the right one in stage 8; one can see more clearly an accumulation of grains on the adjoining parts of the nurse cells in the right egg chamber. × 3600.
(B) An egg chamber of stage 10. Continuity of grains from nurse cell to oocyte is observed as described earlier for wild type. Small arrow marks the nurse cell and the large one the oocyte. × 3190.
At all stages of egg-chamber development, no difference was observed in the amount of RNA between the bobbed and the wild type ; yet we know from our previous biochemical studies that the bobbed ovaries synthesize less rRNA per unit time. Therefore, one would like to know how the normalization of RNA content takes place. To get an insight into this, we have examined some physiological and morphological parameters both in bobbed and phenotypically wild flies. The results are presented in Table 2. There are no differences between the wild and bobbed flies in mean ovariole number and mean number of egg chambers per ovariole, but clear-cut differences are seen in other parameters. Egg production is nearly the same in the two phenotypically wild genotypes (bb+XYL. Ysbb and bb+lg2ty), while it is severely reduced in the three other genotypes (Table 2). The two other measurements (daily rate of egg chamber production and total duration of egg chamber in various stages, which are obviously related to egg production) are also affected in the bobbed genotypes (wabb1/XYL.Ysbbi, wabb1/g2y, sc4sc8/XYL. Ysbb).
Our results show that the growth duration of the egg chamber from stage 1 to 14 is prolonged in the bobbed flies and that the degree of prolongation depends upon the severity of the altered phenotype. To know whether this prolongation is caused by a delay at a particular stage or occurs in all stages, we have analysed the time duration of the egg chambers at various stages of development. The results of this analysis, carried in five genotypes, are presented in Table 3. The two phenotypically wild strains bb+/XYL.Ysbb and bb+/g2ty have a duration period of 75 and 83 h respectively; these figures are in agreement with those presented by other authors (King, 1957; David & Merle, 1968). The figures for each stage are also in agreement with those given by these authors for the phenotypically wild genotypes. For all the bobbed genotypes, the time duration of the egg chamber is prolonged in all the fourteen stages and not in any particular stage. On the basis of these results, one can predict that the egg-laying should start later in the bobbed females than in the wild ones. Such a prediction is confirmed by our results (Table 4). At the time of emergence, both wild and bobbed flies contain egg chambers in the first seven stages of development; however, their frequency varies between the different genotypes: oocytes, in all 14 stages of development, start to appear in 1-day-old females of the wild type, while in bobbed females they appear at stage 10 after 1 day, stage 12 after 3 days and at all the 14 stages only later.
Stage distribution of egg chambers as a function of age in ovaries of various genotypes of Drosophila melanogaster*

It is clear from our results that the developing egg chambers of bobbed females proceed to their next stage of development only after they have synthesized a given amount of ribosomes. This is obtained by a general delay through all stages of oocyte development in the bobbed mutants. The degree of prolongation in the oogenetic processes depends upon the severity of the altered phenotype in the bobbed females, which start to lay eggs later than the wild-type females. This retardation allows them to accumulate ribosomes and to build up an efficient machinery for egg formation.
RÉSUMÉ
Influence de la teneur en RNA sur ïoogénèse chez les mutants ‘bobbed’ de Drosophila melanogaster
La teneur en acides nucléiques des oocytes et des cellules nourricières, de Drosophila melanogaster, en développement: type sauvage et mutants ‘bobbed’, a été suivie par des méthodes cytophotométriques. On n’observe pas de différences dans la teneur en DNA ni en RNA des oocytes et des cellules nourricières à tous les stades étudiés, entre les mutants et les individus de type sauvage. On sait que le processus de l’oogénèse, chez les femelles bobbed est prolongé et que cette prolongation se manifeste à tous les stades du développement des oocytes. Comme les ovaries des femelles bobbed synthétisent moins de rRNA par unité de temps, il est vraisemblable que cette prolongation permet aux oocytes et aux cellules nourricières des femelles bobbed de normaliser leur teneur en RNA. Quand elles atteignent une teneur donnée en RNA, elles procèdent au stage suivant du développement.
Acknowledgements
I am grateful to Prof. J. Brachet for his valuable suggestions and interest in the work. I was partially supported by an International Cell Research Organisation fellowship. This work has been carried out under the Euratom Contract 007.61.10 ABIB.