Introduction

The Drosophila egg chamber consists of a cluster of 16 cells interconnected by ring canals, which are surrounded by a monolayer of somatic follicle cells. One cell in the cluster develops into the oocyte and the other 15 cells become nurse cells that support the oocyte in its development.1 The 15 nurse cells degenerate by apoptosis when the oocyte is mature.2, 3 Normally, apoptosis of the nurse cells is completed at stage 14 of oogenesis when they have transferred their contents to the oocyte.3, 4 Apoptosis of the nurse cells commences at stage 10B, when nurse cells undergo cytoskeletal rearrangement.5, 6 However, in flies that are maintained under starvation conditions or have 20-hydroxyecdysone (20E) at physiological concentrations injected into the abdomen, apoptosis is observed in the nurse cells much earlier, at stages 8 and 9.7, 8 Simultaneous application of the juvenile hormone analogue (JHA), Methoprene, inhibits this apoptosis.7, 8 The ecdysteroid concentration in starved females is higher than in females that are maintained on normal food.9 Thus it seems likely that the monitoring of nutrition and the response of the fly in terms of oogenesis will be controlled in part by ecdysone and juvenile hormone (JH).

Ecdysone functions through the Ecdysone/Ultraspiracle nuclear receptor complex (EcR/USP).10, 11 The complex directly regulates the early ecdysone response genes such as the Broad-Complex (BR-C), E74 and E75. The BR-C encodes a family of zinc finger transcription factors.12 E75 encodes three orphan members of the nuclear receptor superfamily, designated E75A, E75B and E75C.13 Early ecdysone response genes are essential for oogenesis. Transcription of E75 and E74 appeared to be upregulated during stage 8 in both the nurse cells and somatic follicle cells.14 Germline clones of E75 mutant cells result in degeneration of egg chambers after stage 8, with a phenotype similar to that of the temperature-sensitive mutant ecdysoneless1.14 The ecdysone receptor, EcR, which is crucial for regulation of expression of the early response genes, has also been shown to be crucial in the germline, as mutant germline clones arrest development during mid-oogenesis.14 In addition, we have shown that EcR isoforms are differentially suppressed, with EcRA showing higher expression and EcRB lower expression under nutritional shortage.15 BR-C expression in the somatic follicle cells regulates the establishment of dorso-ventral polarity of the eggshell later in oogenesis16 and leads to prolonged endoreplication and to additional amplification of selected genes.17 In addition, BR-C expression in the follicle cells controls the cell fate of the egg chamber, determining if it will progress to develop into an egg or induce apoptosis.8 BR-C isoforms, Z1 and Z3, affect yolk protein gene expression, while Z2 and Z3 expression in the follicle cells induce apoptosis of the stages 8 and 9 egg chambers.8 Thus ecdysone, ecdysone receptor and ecdysone response genes are clearly crucial for a number of developmental decisions in normal oogenesis.

Apoptosis in many tissues and glands in insects is induced by 20E and is regulated by the same ecdysone response genes that are crucial in oogenesis.18 Dronc is one of the apoptosis inducers in Drosophila,19 and is upregulated by reaper,19, 20 which also induce apoptosis in Drosophila.21, 22 Reaper expression is regulated by BR-C and hid expression is regulated by BR-C and E74A in Drosophila salivary glands.23 But Foley and Cooley4 have established that reaper, grim and hid do not affect the apoptosis of nurse cells that commences during normal development of an egg at stage 12 of oogenesis.4 E75 is required for the suppression of diap2, which inhibits the progression of apoptosis in the larval salivary glands during Drosophila metamorphosis.23, 24 The relationship between BR-C, E75A and E75B is described by Woodard et al.25 and White et al.26 during early metamorphosis in Drosophila. The ecdysone/EcR/USP complex activates a small set of genes (early and early-late genes), some of which encode transcription factors. These gene products in turn negatively autoregulate, and turn on a large set of effecter genes (late genes), and control the expression of the mid-prepupal factor βFTZ-F1. This factor then specifies the appropriate early gene response to the subsequent prepupal pulse of ecdysone.25, 26

In this paper, we describe the network of ecdysone response genes used in the control of apoptosis of nurse cells in the stages 8 and 9 egg chamber of Drosophila. BR-C Z2 and Z3, along with E75A and E75B in the follicle cells, control the premature apoptosis of the nurse cells. BR-C responds to the signal initiated by nutritional or hormonal conditions, and controls the expression of E75A and B in the egg chamber. E75A and B have opposite effects on apoptosis, E75A acts as an apoptosis inducer and E75B acts as an apoptosis inhibitor.

Results

E75A expression in the ovary

Ecdysone response genes have been shown to be involved in developmental decisions in many insect tissues and glands.18 We investigated the early ecdysone response genes, E75A and E75B, to establish whether or not they affect apoptosis in the nurse cells of the egg chamber at stages 8 and 9. Transgenic flies of E75A and E75B were heat shocked to induce overexpression in the egg chamber, both in the nurse cells and follicle cells. E75A overexpression induced 12- and 15-fold higher expression than in OrR under fed and starved conditions, respectively, and E75B overexpression induced 19- and 17-fold higher expression than in OrR under fed and starved conditions, respectively. E75A overexpression in the egg chamber induced apoptosis of the nurse cells at stages 8 and 9 in fed flies (F3) and the percentage of egg chambers undergoing apoptosis increased in starved flies (F3S1). E75B overexpression in the egg chamber suppressed the apoptosis of nurse cells at stages 8 and 9 in starved flies (Table 1). E75A and E75B were then investigated to see if they showed a different temporal or spatial expression pattern in the ovary when comparing apoptotic (starved and 20E injected) and nonapoptotic (fed and JH-treated) conditions.

Table 1 Percentage of apoptosis at stage 8 and 9 in the ovary of various transgenic flies
Full size table

E75A expression was observed in the germarium, and follicle cells and nurse cells at early stages, then in the nurse cells at stage 10 of oogenesis in fed flies. However, in the egg chambers at stages 8 and 9, E75A expression in the follicle cells was not observed and expression in the nurse cells was weak (Figure 1A a and f). On the other hand, E75A expression at stages 8 and 9 was observed in follicle cells of starved (S3 and F3S1) flies (Figure 1A b and g) and was also present in the follicle cells at stages 8 and 9 when 20E was injected into the abdomen of the fed flies (Figure 1A d and e). There was no obvious difference in E75A comparing levels in nurse cells between apoptotic and nonapoptotic conditions, but in the follicle cells E75A expression was significantly different (Figure 1A f and g). The expression of E75A at stages 8 and 9 in starved flies was suppressed by JHA, Methoprene-treatment of the abdomen (Figure 1A c). We carried out RT-PCR to compare E75A expression levels between apoptotic and nonapoptotic conditions, using total RNA from dissected egg chambers at stages 8 and 9 only (Figure 1B). In starved flies (S3 and F3S1), E75A expression levels at stages 8 and 9 were higher than in fed flies (F3), and JHA-application suppressed the higher levels in starved flies (F3JHS1). 20E inject ion of fed flies induced higher E75A expression at stages 8 and 9 (F3EF1).

Figure 1
figure 1

E75A expression in ovaries of Drosophila under various nutritional and hormonal conditions and expression levels in stage 8 and 9 egg chambers. (A) Results of in situ hybridization to the ovaries of females under various hormonal or nutritional conditions are shown. (a) Fed: maintained with yeast for 3 days (F3), (b) starved: starved for 1 day after maintaining with yeast for 3 days (F3S1), (c: JHA treated: starved for 1 day with JHA application to the abodomen (1 μg/100 nl acetone) after maintaining with yeast for 3 days (F3JHS1), (d) and (e) 20E treated: fed for 1 day with 20E injection into the abdomen (100 pg/50 nl Ringer's solution) after maintaining with yeast for 3 days. (f) Fed and (g) starved are enlarged to show the egg chamber at stage 8 and focus on the follicle cells of fed (F3) and starved (F3S1) flies, respectively. The number on the panel indicates the stage of oogenesis. (B)Results of RT-PCR using total RNA from the egg chamber (n=150 flies) at stages 8 and 9 only are shown. F3: with yeast for 3 days, S3: without yeast for 3 days, F3S1: starved for 1 day after maintaining with yeast for 3 days, F3AS1: starved for 1 day with acetone application (100 nl) after maintaining with yeast for 3 days, F3JHS1: starved for 1 day with JHA application (1 μg/100 nl) after maintaining with yeast for 3 days, F3RF1: fed for 1 day with Ringer's injection (50 nl) after maintaining with yeast for 3 days, F3EF1: fed for 1 day with 20E injection (100 pg/50 nl) after maintaining with yeast for 3 days. rp49 is used as a control. (C) Percentage of the egg chambers at stage 8 in which E75A expression is induced and at stage 9 in which apoptosis is induced in the nurse cells. Percentage of egg chamber expression was calculated as follows: {(number of egg chambers at stage 8 showing E75A expression in the follicle cells)/(total number of the egg chambers at stage 8)−(number of stage 8 egg chambers showing nuclear condensation or fragmentation in the nurse cells)} × 100. The percentage of egg chambers showing apoptosis in nurse cells at stage 9 was calculated as follows: {(number of egg chambers at stage 9, showing nuclear condensation and fragmentation in nurse cells)/(total number of egg chambers at stage 9)} × 100. Nuclear condensation and fragmentation were detected by staining with Hoechst. There were no significant differences between the percentage of egg chambers showing E75A expression in the follicle cells at stage 8 and the percentage showing apoptosis in the nurse cells at stage 9. *Indicates that there are significant differences (within 5%), between F3 versus S3 and F3S1, F3AS1 versus F3JHS1 and F3RF1 versus F3EF1. n=12 flies

Full size image

Apoptosis of nurse cells following starvation and 20E injection was observed at late stage 8 or stage 9. E75A expression, however, was observed at stages 6 and 7 in all egg chambers of the females under all experimental conditions. If E75A expression induces apoptosis of the nurse cells at stages 8 and 9, the expression level in the follicle cells at stage 8 was likely to be important in determining whether or not apoptosis was induced. As shown in Figure 1C, there were no significant differences between the percentage of egg chambers at stage 8 in which E75A expression in the follicle cells was observed and the percentage of the nurse cells at stage 9 in which apoptosis was induced. This was true for all experimental conditions. But when compared with the controls, fed versus starved, starved versus JHA treated and fed versus 20E injected, these percentages changed dramatically (Figure 1C). The proportion of egg chambers expressing E75A at stage 8 and undergoing apoptosis at stage 9 was higher in starved flies (S3 and F3S1) than in fed flies (F3); and this higher level in starved flies was suppressed by JHA treatment. 20E treatment increased the proportions in fed flies.

We therefore propose that E75A expression in the follicle cells at stages 8 and 9 has a role in the induction of apoptosis in the nurse cells at stages 8 and 9. In flies, under nonapoptotic conditions, it seems that E75A expression level in the follicle cells is suppressed to prevent induction of apoptosis in the nurse cells at stages 8 and 9. E75A expression levels in the egg chamber are increased under apoptotic conditions, which leads to apoptosis in the nurse cells at stages 8 and 9.

E75B expression in the ovary

E75B expression was seen in region 2B of the germarium and the expression was faint in egg chambers at early stages. E75B expression was seen in the follicle cells and nurse cells at stages 4 or 5 again and expression continued until stage 9 in the ovary of fed flies (Figure 2A a and e). The E75B expression pattern was similar under all experimental conditions. However, expression levels appeared to be different by observation under the microscope when comparing nonapoptotic, fed and JHA treated with apoptotic, starved and 20E treated, flies. E75B overexpression in the egg chamber (in both nurse cells and follicle cells) suppressed apoptosis at stages 8 and 9 in starved flies (Table 1). E75B expression at stage 8 was observed in follicle cells and nurse cells (Figure 2A e, f, g and h), but expression levels were different in the egg chamber at stages 8 and 9 when comparing fed flies (F3) and starved flies (S3 and F3S1, Figure 2B and C). E75B expression levels in the egg chamber at stages 8 and 9 of starved flies were rescued by JHA treatment of starved flies (F3JHS1, Figure 2B and C). E75B is one of the ecdysone response genes in Drosophila, but E75B expression in the egg chamber at stages 8 and 9 was suppressed by 20E injection of fed flies (F3EF1, Figure 2B and C).

Figure 2
figure 2

E75B expression in ovaries of Drosophila under various nutritional and hormonal conditions. (A)Results of in situ hybridization to the ovaries of females under various hormonal or nutritional conditions are shown. (a) Fed: maintained with yeast for 3 days (F3), (b) starved: starved for 1 day after maintaining with yeast 3 days (F3S1), (c) JHA treated: starved for 1 day with JHA application to the abdomen (1 μg/100 nl acetone) after maintaining with yeast for 3 days (F3JHS1), (d) 20E treated: fed for 1 day with 20E injection (100 pg/50 nl Ringer's solution) after maintaining with yeast 3 days. (e) Fed, (f) starved, (g) JHA treated and (h) 20E treated are enlarged the egg chamber at stage 8 to focus on the follicle cells of fed (F3), starved (F3S1), JHA treated (F3JHS1) and 20E treated (F3EF1) flies, respectively. The number on the panel indicates the stage of oogenesis. (B) Results of RT-PCR, using total RNA extracted from egg chambers (n=150 flies) at stage 8 and 9 only are shown. F3: with yeast for 3 days, S3: without yeast for 3 days, F3S1: starved for 1 day after maintaining with yeast for 3 days, F3AS1: starved for 1 day with acetone application (100 nl) after maintaining with yeast for 3 days, F3JHS1: starved for 1 day with JHA application (1 μg/100 nl) after maintaining with yeast for 3 days, F3RF1: fed for 1 day with Ringer's injection (50 nl) after maintaining with yeast for 3 days, F3EF1: fed for 1 day with 20E injection (100 pg/50 nl) after maintaining with yeast for 3 days. rp49 is used as a control. (C) The graph indicates the expression level using RT-PCR results that are shown in (B) using relative value±S.D. (measured by NIH image). *Indicates that there are significant differences between fed (F3) and starved (S3 and F3S1), starved (F3S1 and F3AS1) and JHA treated (F3JHS1) and fed (F3 and F3RF1) and 20E treated (F3EF1) flies

Full size image

E75A and C mutants show similar lethal phases, but E75B mutants are viable and fertile, with no detectable phenotype.27 We thought that perhaps E75B mutants might be sensitive to nutritional stress even though under normal conditions they are fertile. Therefore, we examined E75B mutant flies that were maintained under nutritional shortage and checked how many stages 8, 9 and 10 egg chambers were present (Figure 3). The egg chambers of wild-type starved females were reduced by approximately 40, 65 and 70% of stages for 8, 9 and 10 egg chambers, respectively, compared to wild-type fed females, but the egg chambers of E75B-mutant starved flies was reduced approximately 65, 80% and over 80% for stages 8, 9 and 10 egg chambers, respectively (Figure 3a). JHA treatment of the starved wild-type flies suppresses apoptosis in stages 8 and 9 egg chambers7, 8 and led to recovery of the number of stages 8, 9 and 10 egg chambers (Figure 3b). However, when JHA was applied to the starved E75B mutant flies, the recovery levels were lower than in starved wild-type flies (Figure 3b). These results indicate that the E75B mutant is particularly sensitive to nutritional stress that E75B overexpression and JHA treatment suppressed this apoptosis at stages 8 and 9 (Table 1) and that JHA suppresses the apoptosis through E75B. From these results, we propose that E75A and E75B have opposite roles in determining if apoptosis of nurse cells at stages 8 and 9 is induced or not in Drosophila. Further, although E75B mutants are fertile and visible under normal conditions, this mutation is detrimental to fertility under conditions of nutritional stress.

Figure 3
figure 3

E75B mutant is sensitive to starvation. We made E75B mutants (E75Δ1/E75Δ51) by crossing E75Δ1/TM6B, which is an ∼3 kb deletion, most of the first exon of E75B and E75Δ51/TM6B, which is an ∼30 kb deletion that removes the first exon of E75B, as well as the adjacent exon, shared by all three E75 isoforms, that encodes the second zinc finger of the DNA binding domain (Bialecki et al., 2002). (a) Percentage reduction of stages 8, 9 and 10 egg chambers comparing fed (F3) and starved (F3S1) flies is shown. n=12 flies. The egg chambers that indicated nurse cell apoptosis were not included in these graphs. (b) The percentage of recovery at stages 8, 9 and 10 egg chambers following JHA treatment of the starved flies (F3JHS1) is shown. n=12 flies. The egg chambers that showed nurse cell apoptosis were not included in these graphs

Full size image

BR-C Z2 and Z3 control E75A and E75B expression in egg chambers

BR-C Z2 and Z3 are not expressed in the egg chambers at stages 8 and 9 of fed females, but they are expressed in the follicle cells under apoptotic conditions. In addition, overexpression in the egg chamber induces apoptosis of that egg chamber at stages 8 and 9 in fed flies.8 We have suggested that BR-C controls the developmental checkpoint at stages 8 and 9 of oogenesis in Drosophila,28 inducing either yolk protein synthesis or inducing apoptosis in the nurse cells at stages 8 and 9.8 On the basis of these results, we investigated whether or not Z2 and Z3 overexpression in the follicle cells affects E75A and E75B expression in egg chambers at stages 8 and 9.

Figure 4A shows the E75A expression pattern in hsZ2 and Z3 transgenic flies under fed conditions (F3) with or without heat shock. E75A expression in the follicle cells at stages 8 and 9 was not seen or was very faint in hsZ2 and hsZ3 transgenic flies under fed conditions without heat shock (Figure 4A c, f and g). When Z2 and Z3 overexpression in the egg chamber was induced by heat shock, E75A was expressed in the egg chamber, in nurse cells and follicle cells at stages 8 and 9 under fed conditions (Figure 4A a, b, d and e). The E75A expression pattern in egg chambers at stages 8 and 9 thus reflects the results of in situ hybridization, as shown in Figure 3a (Figure 4B). E75A expression level at stages 8 and 9 following Z2 and Z3 overexpression was higher than the expression levels without heat-shock induction in not only fed flies but also starved flies (F3S1, Figure 4B). The percentage of egg chambers at stage 8 in which E75A expression was seen corresponded to the percentage of egg chambers at stage 9 in which apoptosis was induced in fed flies (Figure 4C). Z2 and Z3 overexpression increased the percentage of egg chambers at stage 8 in which E75A expression was observed and the percentage of nurse cells at stage 9 in which apoptosis was induced.

Figure 4
figure 4

E75A expression in the follicle cell under inducing Z2 and Z3 overexpression. (A) Results of in situ hybridization to the ovaries of hsZ2 and Z3 transgenic flies with adequate nutrition, with or without heat shock. (a, b) E75A expression in the adequate nutrition (F3) with heat shock (+hs). (c)E75A expression in the ovary of F3 female without heat shock (−hs). Number indicates the stage of oogenesis. (B) Results of RT-PCR, using total RNA, extracted from egg chambers (n=150 flies) at stages 8 and 9 only of fed (F3) and starved (F3S1) hsZ2 and hsZ3 transgenic flies with or without heat shock (39°C 30 min). (C) Percentage of stage 8 egg chambers in which E75A expression was induced in follicle cells and at stage 9 in which apoptosis was induced in the nurse cells of BR-C Z2 and Z3 transgenic flies (fed flies). The percentage was calculated as follows: {(number of egg chambers at stage 8, E75A expression in the follicle cell)/(total number of egg chambers at stage 8)−(number of stage 8 egg chambers showing nuclear condensation and fragmentation)} × 100. The percentage of the egg chambers (apoptosis in the nurse cells at stage 9) is calculated as follows: {(number of egg chambers at stage 9 indicate nuclear condensation and fragmentation in the nurse cells)/(total number of egg chambers at stage 9)} × 100. There were no significant differences between the percentage of egg chambers showing E75A expression in the follicle cells and the percentage of egg chambers showing apoptosis in nurse cells at stage 9 in each condition. *Indicates that there are significant differences (within 5%) comparing results with and without heat shock. n=12 flies

Full size image

E75B expression patterns in the ovaries of hsZ2 and hsZ3 transgenic flies under fed conditions were the same as the ovaries of OrR flies under fed conditions, and when these transgenic flies underwent heat shock, the expression patterns did not differ from the flies without heat shock (data not shown). However, E75B expression levels were different with and without the heat shock inducing Z2 and Z3 overexpression (Figure 5) and decreased in the egg chambers at stages 8 and 9 under fed and starved conditions (Figure 5). Heat shock did not affect E75B expression at stages 8 and 9 in wild-type control flies.

Figure 5
figure 5

E75B expression in the ovary following Z2 and Z3 overexpression. (a) Results of RT-PCR, demonstrating E75B expression, using total RNA from the stages 8 and 9 egg chambers of fed (F3) and starved (F3S1) hsZ2 and Z3 transgenic flies with or without heat shock (39°C 30 min). n=150 flies. (b) The graph indicates the expression levels with±S.D. (measured by NIH image). *Indicates that there is a significant difference (within 5%) comparing results with and without heat shock

Full size image

Z2 and Z3 overexpression induced E75A expression and suppressed E75B expression in the egg chambers at stages 8 and 9. BR-C controls the developmental checkpoint in oogenesis at stages 8 and 9, by inducing apoptosis and regulating yolk protein gene expression.8 Probably, E75A and E75B expression in the follicle cells and nurse cells at stages 8 and 9 are regulated also by BR-C, in this case Z2 and Z3, and thus E75A and B expression in the follicle cells may ultimately induce the apoptosis of the nurse cells at stages 8 and 9, which leads to degeneration of the egg chamber. Apoptosis must therefore involve communication between the nurse cells, oocyte and follicle cells.

E75B suppresses E75A expression

We investigated whether or not E75A and E75B affected each other in the egg chambers at stages 8 and 9. E75A overexpression did not affect the E75B expression pattern in the ovary (data not shown), but E75B overexpression affected E75A (Figure 6A). In starved flies, E75A was expressed in the egg chambers at stages 8 and 9 of hsE75B transgenic flies under starvation in the same pattern as wild-type flies under starvation (Figure 6A b and c); but when the transgenic flies were heat shocked under starvation, E75A expression at stages 8 and 9 was not seen or was very faint (Figure 6A a). E75A overexpression did not affect E75B expression at stages 8 and 9 and E75B overexpression suppressed E75A expression in the egg chamber at stages 8 and 9 (Figure 6B, C and E). In starved wild-type flies, the percentage of egg chambers in which E75A expression was induced at stage 8 and in which apoptosis was induced at stage 9 was approximately 40%, but when E75B transgenic flies underwent heat shock to induce overexpression, this percentage was decreased to approximately 20% (Figure 6D). This is a significant drop in the number of egg chambers undergoing apoptosis.

Figure 6
figure 6

E75A expression in Drosophila is inhibited by E75B overexpression. (A) Results of in situ hybridization to the RNA of ovaries of hsE75B transgenic flies (269) under nutritional shortage with or without heat shock. (a) E75A expression in the ovary under nutritional shortage (F3S1) with heat shock (+hs), resulting in E75B being expressed in nurse cells and follicle cells. (b and c) E75A expression in the ovary of S3 females without heat shock (−hs). The number on the panel indicates the stage of oogenesis. (B)Results of RT-PCR (E75A expression) using total RNA extracted from the egg chambers (n=150 flies) at stages 8 and 9 in E75B transgenic flies (fed, F3 and starved, F3S1 flies). (C) tResults of RT-PCR (E75B expression), using total RNA, extracted from the egg chambers (n=150 flies) at stages 8 and 9 in E75A transgenic flies (fed, F3 and starved, F3S1 flies). (D) Percentage of stage 8 egg chambers in which E75A expression was induced in the follicle cells and at stage 9 in which apoptosis was induced in the nurse cells of E75B transgenic flies (starved flies, line 267 and 269). There were no significant differences between the percentage of the egg chambers showing E75A expression in the follicle cells at stage 8 and the percentage of egg chambers showing E75A expression in the nurse cells at stage 9. *Means that there are significant differences (within 5%) comparing with and without heat shock. n=12 flies. (E) The graph indicates the E75B expression levels using RT-PCR, and total RNA from stages 8 and 9 egg chambers of E75B transgenic flies. The results are summarized in (C) showing relative value±S.D. (measured by NIH image). *Mean that there are significant differences between experiments with and without heat shock

Full size image

Therefore, in the egg chamber at stages 8 and 9, E75B suppressed E75A expression in the follicle cells and nurse cells under fed and starved conditions. We propose that E75A expression in the follicle cells at stages 8 and 9 induces apoptosis in the nurse cells at these stages, but if E75B expression in the follicle cells at stages 8 and 9 is above threshold level, E75B suppresses E75A in the follicle cells, which in turn suppresses apoptosis in the nurse cells at stages 8 and 9.

Discussion

Early ecdysone response genes control apoptosis in nurse cells at stages 8 and 9

Apoptosis of many insect tissues and glands are affected by 20E and JH. 20E induces apoptosis and JH III treatment inhibits apoptosis in a Drosophila cell line, (l) 2 man.29 In Maduca sexta, 20E induces and JH suppresses the apoptosis of the prothoracic gland.30, 31 It is known that the early ecdysone response genes, BR-C, E74 and E75 are essential for the regulation of apoptosis in insect tissues and glands.18 E74A and BR-C activate the apoptosis inducer, hid, in the Drosophila salivary gland and BR-C activates reaper, also an apoptosis inducer.23 E75A and B induce apoptosis of salivary glands through suppression of diap2, an apoptosis suppressor.23

We have recently established that the BR-C controls the fate of the egg chamber, by progressing either development or apoptosis.8 The number of Drosophila egg chambers is reduced under nutritional shortage8 and in response to increasing ecdysone concentration.7 Apoptosis of the nurse cells at stages 8 and 9 causes a reduction in the number of egg chambers. Degenerating stages 8 and 9 egg chambers often appear elongated and the surface of the egg chamber has a rough appearance.7 In degenerating stages 8 and 9 egg chambers, follicle cells and their nuclei increase in size.7 Follicle cells are likely to phagocytose the dying nurse cells.3 After that the follicle cells die.7 Further, the apoptosis seems to be induced by BR-C Z2 and Z3 expression at stage 8. Nutritional shortage induced increased ecdysone concentration in flies;9 thus, it seemed likely that other ecdysone response genes may also affect apoptosis in the nurse cells at stages 8 and 9.

As shown in Table 1, E75A induces apoptosis in nurse cells at stages 8 and 9 of oogenesis. Stronger E75A expression in the follicle cells at stages 8 and 9 is observed in the egg chambers of females that are starved or injected with 20E, namely under apoptotic conditions (Figure 1A). In addition, BR-C, Z2 and Z3 (which induce apoptosis at stages 8 and 9) overexpression induced E75A expression at stages 8 and 9 (Figure 4). However, E75B does not act as an apoptosis inducer in the egg chamber. E75B overexpression suppresses apoptosis in the egg chambers of starved flies (Table 1). E75B expression levels at stages 8 and 9 reflect these results, E75B expression at stages 8 and 9 in fed and JHA- treated females (namely nonapoptotic conditions) is stronger than under apoptotic conditions (Figure 2B). In addition, E75B mutant flies were sensitive to nutritional shortage (Figure 3a). E75B mutant flies do not show abnormal development,14 but the percentage of egg chamber reduction in E75B mutants was higher during starvation than in wild-type females (Figure 3a). JHA treatment of starved flies suppresses the apoptosis of stage 8 and 9 egg chambers.7, 8 This suppression induced a recovery of the numbers of stage 8, 9 and 10 egg chambers (Figure 3b). JHA treatment induced higher E75B expression in stage 8 and 9 egg chambers (Figure 2B and C). On the other hand, starvation, 20E treatment and BR-C Z2 and Z3 overexpression in the follicle cells suppressed E75B expression in stage 8 and 9 egg chambers (Figures 2B, C and 5). Starvation induces high ecdysteroid concentrations in the haemolymph and ovary,32 which in turn may induce BR-C Z2 and Z3 expression in the follicle cells and thus induces apoptosis at stages 8 and 9.8 These results indicate that the two isoforms, E75A and E75B, have opposite effects on apoptosis of nurse cells. E75A acts as an apoptosis inducer and E75B acts as an apoptosis suppressor.

In summary, during apoptosis of the nurse cells at stages 8 and 9, BR-C Z2 and Z3 and E75A act as apoptosis inducers and E75B acts as an apoptosis suppressor as their expression is controlled in the follicle cells.

Interactions between ecdysone response genes

There is a developmental checkpoint at stages 8/9 in oogenesis.28 At this checkpoint, the egg chambers undergo a developmental selection, becoming committed to produce a mature egg or to undergo apoptosis. BR-C is a key gene at this developmental checkpoint. The progression of development or the induction of apoptosis depends on which BR-C isoforms are expressed in the follicle cells at stage 8.8 When flies are maintained under starvation conditions or 20E is injected into the abdomen, BR-C Z2 and Z3 are expressed in the follicle cells of the egg chamber at stage 8.8 In addition, BR-C Z1 and Z3 regulate yolk protein gene expression that begins at stage 8 in the ovary.8

BR-C Z2 and Z3 affect E75A and E75B expression at stages 8 and 9 (Figures 4 and 5). BR-C Z2 and Z3 overexpression induces stronger E75A expression in the egg chamber, both in nurse cells and follicle cells, at stages 8 and 9 under fed and starvation conditions (Figure 4A and B). In contrast, Z2 and Z3 overexpression suppresses E75B expression under fed and starved conditions (Figure 5). E75A and E75B are therefore downstream of the BR-C isoforms Z2 and Z3. Ecdysteroid concentration in starved flies is higher than in fed flies.9 We suggest that increasing ecdysteroid concentration induces BR-C Z2 and Z3 expression in the follicle cells at stages 8 and 9 and that Z2 and Z3 control the fate of egg chambers by regulating E75A and B expression in the follicle cells. Buszczak et al.14 show that BR-C, E75 and E74 are required for egg chamber development during mid-oogenesis. These genes are needed to permit the mature egg to develop, yet we find they are also needed to induce apoptosis. Expression of early ecdysone response genes is sensitive to changes in the ovarian ecdysone titre.14 Probably, there is a threshold of ecdysteroid titre in the ovary. Carney and Bender33 established that EcR is required for normal oogenesis in Drosophila. This means ecdysteroid is required for normal oogenesis in Drosophila. In contrast, ecdysteroids induce apoptosis at mid-oogenesis in Drosophila.8 We propose that these two opposite effects of ecdysteroid depend upon a threshold level of ecdysteroid titre. If levels are below the threshold, ecdysteroid induces the appropriate gene expression profile for normal oogenesis, but if it exceeds the threshold, ecdysteroids induce an alternative gene set that executes premature apoptosis at mid-oogenesis. Starvation would therefore cause the ecdysteroid concentration to exceed the threshold. We suggest that the ovarian ecdysone titre controls the apoptosis/development decision of individual egg chambers by regulation of the patterns of BR-C isoform expression.

A model for the network of ecdysone response genes, which control apoptosis in the nurse cells of the stages 8 and 9 egg chambers

Figure 7 proposes the first steps in a model for the ecdysone response gene network regulating apoptosis in the nurse cells of the egg chamber. In fed flies, only BR-C Z1 expression is observed in the follicle cells at stages 8 and 9;8 therefore, E75B expression in the follicle cells at stages 8 and 9 is not suppressed. E75B suppresses E75A expression in the follicle cells at stages 8 and 9, which in turn results in apoptosis of the nurse cells at these stages being suppressed and thus leads to the production of many mature eggs. When 20E is injected into the abdomen of fed flies, BR-C Z2 and Z3 are expressed in the follicle cells at stages 8 and 9. In fed flies, Z2 expression is induced by Z3 expression8 and it is possible that 20E activates Z2 expression directly. Z2 expression activates E75A expression in the follicle cells and suppresses E75B expression at stages 8 and 9. Z3 overexpression also activates E75A expression and suppresses E75B expression in the egg chamber at stages 8 and 9. However, we do not know if Z3 activates E75A and suppresses E75B by inducing Z2 expression or acts directly. E75A expression is increased in the follicle cells of the egg chamber at stages 8 and 9 by Z2, and E75A may activate the apoptosis pathway in the nurse cells and/or follicle cells at stages 8 and 9.

Figure 7
figure 7

Model to explain the hierarchy of ecdysone response genes regulating apoptosis of stages 8 and 9 egg chambers. Complete nutrition induces normal development of the mature egg during oogenesis. In this case, just BR-C Z1 is expressed in the follicle cells at stage 8. E75B suppresses E75A expression in the follicle cells to prevent apoptosis. However, if 20E is injected into the abdomen of fed flies, Z2 and Z3 are expressed in the follicle cells at stage 8. In this case, Z2 expression in the follicle cells was induced by Z3 overexpression in the egg chambers. This results in a decrease in E75B and E75A induces apoptosis of the nurse cells at stages 8 and 9. Under starved conditions, Z2 and Z3 are also expressed in the follicle cells at stage 8 and suppress E75B and activate E75A expression in the follicle cells. As a result, the E75B expression level is not enough to suppress E75A expression, so E75A induces apoptosis in the nurse cells of the stages 8 and 9 egg chambers

Full size image

In starved flies, nutritional shortage induces an increase in ecdysone concentration in flies9; ecdysone activates Z2 and Z3 expression in the egg chamber at stages 8 and 9.8 In this case, Z2 and Z3 expression do not affect each other; this suggests that ecdysone activates Z2 and Z3 expression in the follicle cells independently. Z2 and Z3 expression in the follicle cells activate E75A expression in the follicle cells at stages 8 and 9 and suppress E75B expression. As a result, E75A expression in the follicle cells should activate the apoptosis pathway in the nurse cells and/or follicle cells of the egg chamber and apoptosis commences at stages 8 and 9. Follicle cell development is partly independent of germ-line cell differentiation in Drosophila oogenesis,34 but induction of apoptosis at stages 8 and 9 under nutritional stress needs interactions between follicle cells and nurse cells.

The pattern of BR-C isoform expression in the follicle cells controls the checkpoint by interacting in each egg chamber to control a developmental switch leading to the development of a mature egg or to undergo apoptosis. BR-C expression is controlled by ecdysteroid concentration, which is increased in females under nutritional shortage.32 Ecdysteroids are crucial for normal oogenesis in Drosophila.14, 33, 35 On the other hand, our results suggest that ecdysteroids are also needed to induce apoptosis of the egg chamber at stages 8 and 9. We propose that there is a threshold ecdysteroid concentration in the fly and if ecdysteroid levels are below the threshold, Normal oogenesis is induced, but if levels exceed the threshold, ecdysteroids induce apoptosis of egg chambers at stages 8 and 9. When apoptosis of a nurse cell at stages 8 and 9 is induced, the genetic pathway that is activated differs between the fed and starved conditions. However, both pathways result in the activation of E75A expression in the follicle cells at stages 8 and 9 and the suppression of E75B expression at stages 8 and 9. E75A may activate the apoptosis pathway, for example the caspase pathway, which executes apoptosis. We do not know how the apoptosis that we observe in nurse cells is affected by what is happening to gene expression in the nurse cells. It is possible that E75A activates an apoptosis inducer in the follicle cells and the inducer activates the apoptosis pathway in the adjacent nurse cells or is transported to nurse cells so that both cell types die. We have identified some candidate genes for inducing or suppressing apoptosis by a microarray analysis.15 Our results show that E75A is an apoptosis inducer and E75B is the inhibitor. E75A may control expression of apoptosis inducers, which we identified by the microarray analysis, including Dp, p53 or the caspase family. On the other hand, E75B controls expression of an apoptosis inhibitor, and we therefore suggest this may be Diap1 and Diap2. The expression of E75A and E75B are regulated by alternative splicing in the yellow fever mosquito Aedes aegypti.36 We propose that the regulation of alternative splicing of E75 in the Drosophila ovary is controlled by BR-C Z2 and Z3 expression in the follicle cells of stages 8 and 9. The network of genes involved in controlling this developmental decision, which regulates how many eggs a female will produce, is complex and further investigations on how events are coordinated in the nurse cells, oocyte and follicle cells remain to be undertaken.

Materials and Methods

Drosophila maintenance

Flies were maintained on standard yeast, maize meal, sugar and agar medium at 25°C. The wild-type strain, Oregon R (OrR) was used throughout. All the flies for each experiment were 3 days old. Flies (3 days old) were transferred from a standard diet to one of sugar and water (starved, 1% agar medium, which contains 5% sucrose and 0.005% 10% Nipagen in 95% ethanol) or one of yeast (fed, 2 g bakers yeast on approximately 50 ml 1% agar medium, which contains 2.5% cornflour, 5% sucrose, 1.75% lypophilized yeast and 0.005% Nipagen in 95% ethanol). After 3 days on sugar or yeast, flies were dissected (sugar: S3, yeast: F3), transferred to sugar for 1 day after 3 days on yeast (F3S1), or topically treated with Methoprene and maintained on sugar and water for 1 day (F3JHS1), injected with 20E and maintained on yeast for 1 day (F3EF1). We used BR-C transgenic flies (TN-Q1-Q2-Z1, Z2, Z3 and Z4, kindly provided by C Bayer), E75A transgenic flies (line 283, 286), E75B transgenic flies and E75Δ1 and E75Δ51 to make E75B−/Df (line 267 and 269, E75A, E75B transgenic and E75Δ1 and E75Δ51 flies were kindly provided by CS Thummel). Flies carrying the E75Δ51 allele were crossed to E75Δ1 mutants. The flies were maintained at 25°C for 3 days with yeast (Fed, F3) or starved for 1 day after maintaining with yeast for 3 days (Starved, F3S1), then underwent heat shock at 39°C for 30 min and were maintained at 25°C for 6 h.

Injection of 20E and application of JHA

20E (Sigma) was dissolved in Insect Ringer's solution (130 mM NaCl, 4.7 mM KCl, 1.9 mM CaCl2) and 50 nl was injected at a concentration of 2 μg/ml.37 The concentration of 20E was determined according to Bownes.9 With a haemolymph volume of approximately 1 μl/female,37 100 pg 20E/female leads to a concentration of 2 × 10−7 M. Methoprene (ZR515, Zoecon) was applied topically to the ventral abdomen in 100 nl acetone. Methoprene diluted 1 : 100 in acetone corresponds to a concentration of about 1 μg/100 nl. Controls were also undertaken injecting Ringer's only and treating flies with acetone.

Hoechst staining

Ovaries were dissected in Insect Ringer's solution and fixed in 4% paraformaldehyde. After fixation and permeabilization in 1% Triton X-100 in PBS, ovaries were stained in 1 μM Hoechst in PBS for 5 min, then washed twice in PBS for 2 h and mounted in FISH medium (220 mM 1,4-diazabicyclo [2.2.2] octane, 90% glycerol, 100 mM Tris-HCl pH 8.5) and examined immediately using fluorescent filters.

RNA in situ hybridization

The protocol is based on the procedure previously described38 and modified as follows. The ovaries were dissected in Ringer's solution and fixed for 20 min in 4% p-formaldehyde in PBS. After rinsing the tissue in PBT, it was treated for 10 min in methanol/0.5 M EGTA, pH 8 (9 : 1). The ovaries can then be stored in methanol at −20°C for several months. The stored ovaries were rehydrated in PBT. The prehybridization was carried out for 1 h at 45°C in DNA hybrid (50% deionized formamide, 5 × SSC, 100 μg/ml sonicated salmon sperm DNA, 50 μg/ml Heparin, 0.1% Tween-20). The ovaries were hybridized overnight at 45°C in DNA hybrid containing digoxigenin-labeled probe (DIG-DNA labeling and detection kit, Boehringer Mannheim). For detection, a 1 : 1000 dilution of anti-DIG-AP-conjugated Ab was used. The staining reaction was performed in 100 mM Tris pH 9.5, 50 mM MgCl, 10 mM NaCl, 0.2% Tween-20, 8 mM levamisole, 4.5 μl/ml NBT and 3.5 μl/ml X-phosphate (Boeringer Mannheim) for 5 h. Anti-DIG-AP conjugate was preabsorbed with postfixed wild-type (Oregon R) ovaries at overnight 4°C. The ovaries were mounted in a mixture of PBS/glycerol (1 : 4) for microscopy. After the ovaries had been double stained using Hoechst and in situ hybridization to RNA, they were washed in PBS and stained in 1 μM Hoechst in PBS for 5 min, then washed twice in PBS for 2 h in the dark and mounted in FISH medium.

RNA extraction and RT-PCR

Transcript levels in ovaries were detected by reverse transcriptase (RT)-PCR as described previously.39 Total RNA was extracted from egg chambers at stages 8 and 9. The egg chambers at stages 8 and 9 were isolated from the ovary after dissection. The primer sequences are as below: E75A, forward 5′-TCAAGTGTCATTTCGAAGCCA-3′ and reverse 5′-AGATTGGCGATTTCCTTGTG-3′, E75B, forward 5′-GCTCTAGACACCAAAGCCATGTGCCGATCT-3′ and reverse 5′-GGCGCAGGAGATTGGCGATT-3′. The expression levels (relative value and standard deviation) were quantified by NIH image (downloaded from http://rsb.info.nih.gov/nih-image/download.html).