All EMBO Press journals Open Access as of 1 January 2024 - read the FAQs

Article
4 May 1999
Free access

Joint action of two RNA degradation pathways controls the timing of maternal transcript elimination at the midblastula transition in Drosophila melanogaster

The EMBO Journal
(1999)
18: 2610 - 2620
Maternally synthesized RNAs program early embryonic development in many animals. These RNAs are degraded rapidly by the midblastula transition (MBT), allowing genetic control of development to pass to zygotically synthesized transcripts. Here we show that in the early embryo of Drosophila melanogaster, there are two independent RNA degradation pathways, either of which is sufficient for transcript elimination. However, only the concerted action of both pathways leads to elimination of transcripts with the correct timing, at the MBT. The first pathway is maternally encoded, is targeted to specific classes of mRNAs through cis‐acting elements in the 3′‐untranslated region and is conserved in Xenopus laevis. The second pathway is activated 2 h after fertilization and functions together with the maternal pathway to ensure that transcripts are degraded by the MBT.

Introduction

In animal embryos as diverse as echinoderms, insects, amphibians and mammals, the earliest stages of development are programed by maternally synthesized RNAs and proteins (reviewed in Davidson, 1986). Subsequent phases of embryogenesis require products encoded by zygotically synthesized transcripts. The transition from maternal to zygotic control of development is referred to as the midblastula transition (MBT) which is defined by the first developmental processes that require zygotic products. The MBT should not be confused with the onset of zygotic transcription, which initiates earlier. Prior to the MBT, a subset of the maternally synthesized transcripts is degraded. In mammals, maternal transcript degradation is complete by the two‐cell stage, while in echinoderms, amphibians and insects it occurs prior to the MBT, when either several hundred cells (echinoderms), several thousand cells (amphibians) or several thousand syncytial nuclei (insects) are present.
Genetic and molecular studies in Drosophila melanogaster have demonstrated that a subset of the maternal mRNAs encode proteins controlling the cell cycle, positional specification of cells or the morphogenetic movements that drive gastrulation. These transcripts are eliminated during the syncytial blastoderm stage, shortly after zygotic transcription commences (reviewed in Cooperstock and Lipshitz, 1997; Bashirullah et al., 1998). The subsequent cellularization of the blastoderm is the first developmental process requiring zygotic contributions and thus marks the Drosophila MBT (Edgar et al., 1986; Merrill et al., 1988; Sibon et al., 1997).
During early Drosophila embryogenesis, from egg‐laying until the MBT, there are 13 synchronous syncytial nuclear divisions (Zalokar and Erk, 1976; Foe and Alberts, 1983; Campos‐Ortega and Hartenstein, 1998). Subsequently, the nuclei enter interphase of nuclear division cycle 14 and the somatic blastoderm cellularizes by concerted invagination of cell membranes from the periphery. Blastoderm cellularization is followed by the first gastrulation movements and the first zygotically programed mitosis, mitosis 14, which is no longer synchronous (Foe, 1989). It has been shown that this transition from synchronous maternally driven mitoses to asynchronous zygotically programed mitoses requires degradation of maternal string transcripts (string encodes a homolog of the cell cycle regulator, CDC25) (Edgar and O'Farrell, 1989, 1990; Edgar and Datar, 1996).
While turnover of maternal transcripts was first reported more than 20 years ago, little is known of the degradation machinery or of the mechanisms by which specific classes of transcripts are targeted to it. Here we show that two transcript degradation pathways function in the early Drosophila embryo. The first, ‘maternal’, pathway is driven by maternally encoded factors that are recruited by cis‐acting RNA degradation elements independently of whether the transcript is translationally active or translationally repressed. This maternal degradation apparatus is conserved in Xenopus oocytes and early embryos. The second ‘zygotic’ pathway becomes active 2 h after fertilization. Either pathway acting alone is sufficient to eliminate maternal transcripts; however, the joint action of both pathways is necessary for elimination of transcripts prior to the MBT. This dual degradation system is likely to coordinate developmental events by ensuring the elimination of maternal transcripts prior to the MBT.

Results

Degradation of maternally synthesized transcripts begins shortly after fertilization in Drosophila

We analyzed the dynamics of transcript degradation in the early embryo focusing on three maternally encoded RNAs: Hsp83, string and nanos. Hsp83 transcripts are particularly abundant, representing 1% of the polyadenylated transcripts in the early embryo (Zimmerman et al., 1983). RNA tissue in situ analyses demonstrated that maternal Hsp83 and string transcripts are no longer detectable by late stage 5, after completion of blastoderm cellularization (Figure 1A–F); nanos transcripts are undetectable by stage 4 (Figure 1G–I). RNA tissue in situ analyses are useful for defining when very low levels of transcripts are reached, but they are not sufficiently quantitative to be used to define the time course of degradation. Consequently, we carried out Northern blot analyses of RNA samples prepared from embryo collections spanning 30 min intervals during the first 5 h after fertilization. These blots were probed for Hsp83, string, nanos and rpA1 transcripts. rpA1 served as an internal control since maternal rpA1 transcripts are stable beyond the MBT (Riedl and Jacobs‐Lorena, 1996) (Figure 1J). Maternal Hsp83, string and nanos transcript levels decrease throughout the pre‐MBT stages initiating within the first hour after fertilization (Figure 1K). By the MBT (2.5–3.0 h after fertilization), >96% of the maternal Hsp83 transcripts loaded into the embryo have disappeared (Figure 1K). The Hsp83 transcripts present at and after this stage comprise undegraded maternal transcripts in the pole cells and zygotic transcripts. Ninety‐five percent of long isoform string transcripts (Figure 1J and K; only the long isoform is strictly maternally synthesized and is shown in K, although quantitation was done on each isoform) and >99% of the nanos transcripts have been eliminated by the MBT (Figure 1J and K). The remaining nanos transcripts represent undegraded maternal transcripts in the pole cells (Figure 1H and I). These results demonstrate that maternal transcript degradation initiates at or shortly after fertilization, and that the vast majority of these transcripts have been eliminated by the MBT.
image
Figure 1. Degradation of maternal Hsp83, string and nanos transcripts begins at fertilization, and >95% of the transcripts are eliminated by the MBT. Whole‐mount in situ analysis of transcript distributions in wild‐type early embryos are shown in (A–I) while Northern blots and quantitative analysis are shown in (J) and (K). All three species of transcripts are present throughout the embryo at late stage 2 (A, D and G; at about nuclear cycle 8; 25 min–1 h 5 min after fertilization). By stage 4 (nuclear cycle 13, 1 h 20 min–2 h 10 min after fertilization), nanos transcripts have been degraded in the somatic region (H) while Hsp83 and string transcripts are still present (B and E). By the cellular blastoderm stage (nuclear cycle 14, 2 h 10 min–2 h 50 min after fertilization), Hsp83 and string transcripts have been degraded throughout the somatic region (C and F) while Hsp83 transcripts (arrowhead, C) and nanos transcripts (I) remain undegraded in the pole cells. The previously reported zygotic expression of Hsp83 RNA (Ding et al., 1993a) can be seen in the presumptive head region (asterisk). Stages are according to Campos‐Ortega and Hartenstein (1998). Embryos are oriented with the anterior to the left and dorsal up. This convention will be followed throughout. (J) Northern blots are shown for Hsp83, string, nanos and rpA1 transcripts in developing embryos. For quantitation (K), amounts were normalized relative to rpA1 levels since rpA1 transcripts are stable. The time points shown in (J) also apply to (K). The ordinate in (K) presents [RNA] plotted on a logarithmic scale. A total of 96% of the maternally encoded Hsp83 transcripts, 95% of the strictly maternally encoded string transcripts (‘long isoform’, upper band in J) and >99% of the nanos transcripts have disappeared by 2.5 h after fertilization (arrows). The Hsp83 transcripts present after 2.5 h in embryos comprise maternal transcripts present in pole cells and zygotically synthesized transcripts. The larger string transcript, which is strictly maternal in origin (Edgar and Datar, 1996), was quantified. Details of the number of repetitions of each Northern blot experiment are given in Materials and methods.

Maternally encoded products direct transcript degradation, even in unfertilized eggs

We used unfertilized eggs to ask whether maternally encoded products are sufficient to direct transcript degradation in the absence of both fertilization and zygotic transcription. In Drosophila, egg activation occurs when the mature oocyte passes from the ovary into the uterus. This process does not require fertilization. Egg activation releases the female pronucleus from meiotic arrest at metaphase I, results in translation of various maternally encoded transcripts deposited in the oocyte, and induces depolymerization of cortical microtubules and crosslinking of the vitelline membrane (Mahowald et al., 1983).
The transcripts which we analyzed, i.e. maternal Hsp83 (Figure 2A–C), string (Figure 2D–F), nanos (Figure 2G–I) and Polar granule component (Pgc; Figure 7G), are degraded in unfertilized eggs. Northern blot analyses (Figure 2J) indicated that 99% of the Hsp83 transcripts and 90% of the long isoform string transcripts have been degraded by 4–5 h after egg activation (Figure 2K), 1–2 h later than in developing embryos (Figure 2J and K). Over 99% of the nanos transcripts have been degraded by 3 h after egg activation (Figure 2J and K), 0.5 h later than in developing embryos but >1 h earlier than Hsp83 and string transcripts. rpA1 transcripts are stable throughout this period (Figure 2J). We conclude that maternally encoded factors are sufficient for degradation of Hsp83, string, nanos and Pgc transcripts. A second, superimposed, degradation pathway causes the more rapid elimination of transcripts in embryos than in unfertilized eggs (see below).
image
Figure 2. Hsp83, string and nanos transcripts are degraded in unfertilized eggs. Whole‐mount in situ analysis of transcript distributions in unfertilized eggs are shown in (A–I) while Northern blots and quantitative analysis are shown in (J) and (K). Hsp83, string and nanos transcripts are present throughout the egg from 0 to 2 h after egg activation (A, D and G). Between 2 and 3 h after egg activation, levels of Hsp83 and string transcripts have decreased (B and E), nanos transcripts are no longer detectable outside the posterior polar plasm (H) and Hsp83 transcripts can be seen to be concentrated in the posterior polar plasm (B). By 3–4 h, all three classes of transcripts are undetectable (C, F and I) except for the posterior‐localized Hsp83 and nanos transcripts (C and I). (J) Northern blots are shown for Hsp83, string, nanos and rpA1 transcripts in unfertilized eggs. Hsp83 and string transcripts (strictly maternal, upper band) reach minimal levels by 4–5 h, whereas nanos transcripts have disappeared by 3 h after egg activation. For quantitation (K), amounts were normalized relative to rpA1 levels since rpA1 transcripts are stable. The time points shown in (J) also apply to (K). The ordinate in (K) presents [RNA] plotted on a logarithmic scale. In unfertilized eggs, >99% of the nanos transcripts, 99% of the Hsp83 transcripts and 90% of the string transcripts (strictly maternal, upper band in J) are degraded. Transcripts from unfertilized eggs are shown in gray (●) while transcripts from embryos (taken from Figure 1) are shown in black (▪) for comparison. Details of the number of repetitions of each Northern blot experiment are given in Materials and methods.

Cis‐acting elements in the 3′‐UTR target transcripts for degradation

To identify cis‐acting sequences essential for transcript degradation, we focused on the 3′‐untranslated regions (3′‐UTRs) of Hsp83 and nanos transcripts. If an element targets transcripts for degradation, then absence of this element would be expected to stabilize the transcript. Unfertilized eggs were used for the analysis in order to identify those elements necessary for maternally encoded transcript degradation.
For Hsp83, a series of reporter transgenes was constructed that contained a fragment of the Escherichia coli lacZ gene and either the full 407 nucleotides of the Hsp83 3′‐UTR or deleted versions (see Figure 3A for sequence, Figure 3B for transgenic constructs). By this analysis, a region in the 3′‐UTR from nucleotide 253 to 349, which we refer to as the Hsp83 degradation element (HDE), proved necessary for maternally encoded Hsp83 transcript degradation (Figure 3C–F). Transcripts lacking the HDE (252Δ350) are stable in unfertilized eggs for at least 6h (Figure 3F). This contrasts with transcripts that include the HDE (endogenous Hsp83 transcripts or transgenic transcripts that include the HDE) which are fully degraded in unfertilized eggs by 4 h after activation (see Figure 2C, J and K for endogenous transcripts; data not shown for transgenic transcripts). Elimination of either the 5′ or the 3′ half of the HDE (252Δ301 or 300Δ351) results in incomplete degradation (data not shown; see Figure 3B).
image
Figure 3. Mapping of the Hsp83 degradation (HDE) and protection (HPE) elements. (A) Sequence of the Hsp83 3′‐UTR with highlighted HDE (black) and HPE (gray). (B) Transcription of the transgenes is under the control of Hsp83 regulatory sequences fused to a 603 bp lacZ tag and followed by the 407 nucleotide Hsp83 3′‐UTR (1.407) or the Hsp70 3′‐UTR (‘Hsp70’) (see Materials and methods). 3′, 5′ and internal deletions within the Hsp83 3′‐UTR are shown as a diagram. +, degradation or protection occur as in wild‐type; +*, degradation incomplete; −, degradation or protection fail; NA, protection cannot be assayed in the absence of degradation; NM, no maternal RNA present in the egg (Halsell, 1995). Black bars represent Hsp83 3′‐UTR sequences; gray bars represent Hsp70 3′‐UTR sequences. The HDE which maps from nucleotide 253 to 349 and the HPE which maps from nucleotide 350 to 407 are highlighted. (C and D) Endogenous Hsp83 transcripts visualized with an antisense Hsp83 RNA probe. (E and F) Transgenic 252Δ350 reporter transcripts visualized with an antisense β‐galactosidase RNA probe. The age of eggs or embryos in hours is shown to the left. In contrast to endogenous transcripts (C and D), the 252Δ350 transcripts persist throughout the egg until at least 6 h after activation (E and F).
The first 186 nucleotides of the nanos 3′‐UTR, designated the translation control element (TCE), have also been implicated in control of transcript stability (Dahanukar and Wharton, 1996; Smibert et al., 1996). In unfertilized eggs, this element is shown here to target the maternally encoded degradation machinery to nanos transcripts. Transcripts lacking the first 186 nucleotides (ΔTCE) are not degraded by 2–4 h after egg activation (Figure 4C and D) whereas endogenous nanos transcripts have completely disappeared (Figure 4A and B).
image
Figure 4. The Hsp83 HDE and the nanos TCE are interchangeable with respect to maternal transcript degradation, which is independent of translational repression. (A and B) Endogenous nanos transcripts are degraded by 2 h after egg deposition. (C and D) Deletion of the nanos TCE (nanos 3′‐UTR nucleotides 1–185) results in transcript stabilization. (E and F) Replacement of the nanos TCE with the Hsp83 HDE (Hsp83 3′‐UTR nucleotides 253–349) restores transcript degradation. (G and H) Replacement of the Hsp83 HDE with the nanos TCE restores transcript degradation. nanos[ΔTCE+HDE] transcripts are not translationally repressed even though they are degraded. Head skeletons are shown from wild‐type (I), embryos from nanos[ΔTCE] transgenic females (J) and embryos from nanos[ΔTCE+HDE] females (K). As previously reported (Dahanukar and Wharton, 1996; Smibert et al., 1996), there is a reduction of the dorsal bridge in nanos[ΔTCE] embryos (J). Surprisingly, although nanos[ΔTCE+HDE] transcripts are degraded (F), they are not translationally repressed, as evidenced by the severe head skeleton phenotype (K) (note the loss of dorsal bridge, dorsal arms, vertical plate and ventral arms). DBr, dorsal bridge; DA, dorsal arms; VP, vertical plate; VA, ventral arms.
To study whether the Hsp83 HDE and the nanos TCE are functionally interchangeable, two additional transgenic constructs were made. In one, the HDE was added to a nanos transgene lacking the TCE (nos[ΔTCE+HDE]) (Figure 4E and F); in the other, the nanos TCE was added to an Hsp83–lacZ reporter construct lacking the HDE (Hsp83[ΔHDE+TCE]) (Figure 4G and H). Addition of the Hsp83 HDE restores degradation to the otherwise stable nos[ΔTCE] transcript (Figure 4E and F) while addition of the nanos TCE similarly targets the stable Hsp83–lacZ[ΔHDE] transcript for degradation (Figure 4G and H). While our data demonstrate that the HDE and TCE are functionally interchangeable, we have been unable to identify any significant primary sequence or secondary structure conservation between these elements.
These data support a model in which specific cis‐acting elements recruit a maternally encoded degradation machinery to both the Hsp83 and nanos transcripts.

Relationship between translation and RNA degradation

Our next goal was to study whether translational repression is a prerequisite for transcript degradation. Such a requirement had been postulated for nanos transcripts in early embryos since relief of translational repression (by deletion of the TCE) correlated with an increase in transcript stability (Dahanukar and Wharton, 1996; Smibert et al., 1996). Translation of nanos in the head region of the embryo causes repression of bicoid translation and thus head skeleton defects (Dahanukar and Wharton, 1996; Smibert et al., 1996). The severity of the defect serves as a sensitive readout of the Nanos protein level (Wharton and Struhl, 1991; Gavis and Lehmann, 1994) in the absence of a direct assay using anti‐Nanos antibodies, which are no longer available.
We examined the head skeleton of transgenic embryos carrying nos[ΔTCE] or nos[ΔTCE+HDE] transcripts. As previously reported (Dahanukar and Wharton, 1996; Smibert et al., 1996), nos[ΔTCE] transcripts are translated and produce head skeleton defects (Figure 4J). Surprisingly, even though nos[ΔTCE+HDE] transcripts are degraded (Figure 4F), they too produce head skeleton defects (Figure 4K). Furthermore, the severity of the head skeleton defects was greater in the case of nos[ΔTCE+HDE] than nos[ΔTCE]. In five independent nos[ΔTCE] transgenic lines, an average of 72 ± 29% of the embryos exhibited defects and these were restricted to the dorsal bridge. In contrast, all four independent nos[ΔTCE+HDE] lines resulted in 100 ± 0% of the embryos exhibiting severe defects (loss of dorsal bridge, dorsal arms, vertical plates and ventral arms).
Thus, not only are the nos[ΔTCE+HDE] transcripts translated, but they are translated at higher levels than the nos[ΔTCE] transcripts, perhaps because the HDE includes a translational enhancer. We conclude that translational repression and transcript degradation are not causally linked; even highly translated transcripts are targeted for degradation. Moreover, non‐protein‐coding transcripts (e.g. Pgc) are also degraded in unfertilized eggs (Figure 7G). Together, these data indicate that the transcript degradation machinery acts on transcripts that contain a degradation element independently of whether they are untranslatable, translationally repressed or translated.

Discovery of a second, zygotically encoded or activated degradation pathway

We next addressed why Hsp83 transcripts are eliminated more rapidly in embryos than in unfertilized eggs. We asked whether embryos contain an additional degradation machinery that acts independently of the HDE, by comparing the stability of transcripts containing or lacking the HDE. Degradation of endogenous Hsp83 transcripts as well as HDE‐containing transgenic transcripts is initiated shortly after fertilization, and minimal transcript levels are reached by 2.5 h (Figure 5A, B and F for endogenous transcripts; transgenic transcripts not shown). In unfertilized eggs, transcripts lacking the HDE are stable for well over 5 h (Figures 3E and F, and 5E). HDE‐deleted (252Δ350) transcripts are stable throughout the first 2 h of embryogenesis, after which degradation initiates, and 95% of these transcripts are gone by 4 h after fertilization (Figure 5C, D, F and G). Thus, in addition to the maternally encoded degradation machinery, a zygotically synthesized or zygotically triggered degradation machinery is active starting 2 h after fertilization. Furthermore, since transcript degradation occurs at this stage in the absence of the HDE, this ‘zygotic’ degradation is independent of elements in the 3′‐UTR that mediate the maternally encoded degradation. Zygotic transcript degradation occurs for all 3′‐UTR reporter constructs listed in Figure 3.
image
Figure 5. Hsp83 transcripts lacking the HDE are degraded in embryos by a zygotically synthesized or activated degradation mechanism. (A and B) Endogenous Hsp83 transcripts visualized with an antisense Hsp83 RNA probe. (C and D) Transgenic 252Δ350 reporter transcripts visualized with an antisense β‐galactosidase RNA probe. The stage of embryos is shown to the left. In contrast to endogenous transcripts (A and B) which are degraded by the MBT and so are absent by stage 5, the 252Δ350 transcripts persist beyond stage 5 (C) and are degraded by stage 8 (D). Zygotically synthesized transcripts accumulate in the head region and in the germ band (B and D). Northern blots are shown of unfertilized eggs (‘UF’ in E) and developing embryos (‘F’ in F). Quantitation is shown in (G). In unfertilized eggs, transgenic transcripts lacking the HDE (●, solid gray line) are stable, whereas they are degraded starting after 2 h and are gone by 4 h after fertilization in embryos (▴, dashed gray line). Endogenous transcripts in embryos are shown for reference (▪, solid black line; taken from Figure 1K). For quantitation, amounts were normalized as in Figures 1 and 2. In embryos, the Hsp83 transcripts that persist after 2.5 h (black, endogenous) or after 4 h (dashed line, ΔHDE) comprise maternal transcripts present in pole cells and zygotically synthesized transcripts. Details of the number of repetitions of each Northern experiment are given in Materials and methods.
Taken together, our analyses demonstrate that the joint action of both the ‘maternal’ and the ‘zygotic’ degradation pathways is needed to eliminate the bulk of maternal transcripts by the onset of the MBT.

Elements in the 3′‐UTR protect transcripts from degradation in the pole cells

Certain maternal transcripts fail to degrade in pole cells of developing embryos or in the posterior polar plasm of unfertilized eggs. These include Hsp83 (Figures 1C and 2C), nanos (Figures 1H and I, and 2H and I) and Pgc (Figure 7G) (Wang and Lehmann, 1991; Ding et al., 1993a; Nakamura et al., 1996). Our time course analysis shows that 1% of the total maternally loaded Hsp83 transcripts and <1% of the total maternally loaded nanos transcripts remain in the posterior polar plasm of unfertilized eggs (Figure 2K). Two possible mechanisms could lead to the absence of transcript degradation at the posterior: the degradation machinery might be excluded from the posterior polar plasm and thus also from the pole cells. Alternatively, the machinery could be active in the posterior polar plasm and pole cells, but certain classes of transcripts could be masked from the machinery by components of the polar plasm.
To distinguish between these possibilities, we replaced the Hsp83 3′‐UTR in our reporter construct with the 3′‐UTR from the Hsp70 transcript (Figure 3B). Transgenic reporter transcripts carrying the Hsp83 3′‐UTR are degraded normally in the soma but are not degraded in the pole cells (Figure 6A–C), mimicking the distribution of the endogenous Hsp83 transcripts (Figure 1A–C). The reporter transcripts carrying the Hsp70 3′‐UTR are taken up into the pole cells when they bud (Figure 6D and E) but are unstable and are degraded by the cellular blastoderm stage (Figure 6F; see figure legend for details). These results demonstrate that the maternally encoded degradation machinery is present throughout the egg and embryo.
image
Figure 6. Degradation activity is present in pole cells but Hsp83 transcripts are normally protected from it. (A and B) Stage 3 pole cells; (C and D) stage 4 pole cells; (E and F) stage 5 pole cells. (A–C) Hsp83 1.407 transgenic transcripts or (D–F) transgenic transcripts that replace the Hsp83 3′‐UTR with the Hsp70 3′‐UTR (see Figure 3A), visualized with an antisense β‐galactosidase probe. Transcripts that include the Hsp83 3′‐UTR are protected from degradation in pole cells, while transcripts with the Hsp70 3′‐UTR are degraded in both the pole cells and the soma by stage 5, indicating that degradation activity is present in pole cells. At stage 5, 82% of the embryos with the Hsp70 3′‐UTR transgenic RNA had little or no detectable transcript in pole cells (n = 39). At stages 6–8, 95% of the embryos with the Hsp70 3′‐UTR transgenic RNA had little or no detectable transcript in pole cells (n = 42). In contrast, 100% of the embryos carrying full‐length Hsp83 3′‐UTR transgenic RNA had high levels of transcript in their pole cells at these stages (n = 50).
image
Figure 7. The cortex and grauzone mutants result in failure of maternally encoded transcript degradation. Assays were conducted in unfertilized eggs, aged for at least 3 h, from wild‐type females (A, D, G and J), cortex females (B, E, H and K) or grauzone females (C, F, I and L). Degradation of Hsp83 (A–C), nanos (D–F), Pgc (G–I) and string (J–L) transcripts fails in eggs produced by the mutants. The defects in transcript localization (E, F, H and I) may result from cytoskeletal abnormalities (Lieberfarb et al., 1996; Page and Orr‐Weaver, 1996).
To identify ‘protection’ elements within the Hsp83 3′‐UTR, we assayed posterior protection of our reporter transcripts in unfertilized eggs (Figure 3B). Sequences sufficient for protection reside within the 3′‐most 107 nucleotides of the Hsp83 3′‐UTR since transgenic RNA produced by the 301.407 construct is protected (Figure 3A and B). Since transgenic RNA carrying only nucleotides 351–407 is unstable during oogenesis (Halsell, 1995), it was not possible to determine whether the 3′‐most 57 nucleotides alone are sufficient for protection. However, replacement of nucleotides 351–407 with the Hsp70 3′‐UTR results in failure of protection (Figure 3B), showing that nucleotides 351–407 are necessary for protection. We refer to nucleotides 351–407 as the Hsp83 protection element or HPE. Thus elements necessary for degradation versus protection map to distinct regions within the Hsp83 3′‐UTR.

Identification of mutants that fail to undergo maternally encoded RNA degradation

Since our data demonstrated that maternal factors are sufficient for transcript degradation prior to the MBT, we tested a collection of female sterile mutant lines (Schüpbach and Wieschaus, 1989) to identify maternal effect mutants that fail to undergo degradation (see Materials and methods for list). Mutants were assayed for Hsp83 RNA degradation in unfertilized eggs aged well beyond the stage at which transcripts would have disappeared in the wild‐type. The presence of protected transcripts at the posterior served as an internal control for the in situ hybridizations (if degradation occurs, then protected Hsp83 transcripts are visible at the posterior). Mutations in two loci, cortex and grauzone, result in failure of Hsp83 transcript degradation (Figure 7B and C). Failure of degradation was not specific to Hsp83 since nanos (Figure 7E and F), Pgc (Figure 7H and I) and string (Figure 7K and L) transcripts also failed to degrade. The cortex and grauzone mutants exhibit defective egg activation and cytoplasmic polyadenylation of transcripts (Lieberfarb et al., 1996; Page and Orr‐Weaver, 1996).

The maternally encoded degradation pathway is conserved in Xenopus

Several maternal transcripts are localized during Xenopus oogenesis although it is not yet known whether the type of degradation–protection system described here for Drosophila functions in their localization (reviewed in Bashirullah et al., 1998). Furthermore, as in Drosophila, many maternal transcripts are degraded prior to the Xenopus MBT (reviewed in Davidson, 1986). Thus, it was of interest to determine whether Xenopus oocytes and early embryos have a maternal RNA degradation machinery similar to that identified here in Drosophila. We injected in vitro synthesized, digoxigenin‐labeled transcripts comprising the full‐length Drosophila Hsp83 3′‐UTR into stage 6 Xenopus oocytes or recently fertilized embryos. These transcripts are unstable and are degraded rapidly after injection (Figure 8A for oocytes; data not shown for embryos). We then injected in vitro transcribed Drosophila Hsp83 3′‐UTR ΔHDE transcripts into stage 6 oocytes; these transcripts are extremely stable and are still readily detectable 24 h after injection (Figure 8B). We conclude that the Drosophila HDE functions as a degradation element in Xenopus oocytes and embryos. Thus, Xenopus must contain a transcript degradation activity highly related to the Drosophila maternal activity and capable of recognizing the Drosophila HDE.
image
Figure 8. The maternal transcript degradation pathway is conserved in Xenopus and Drosophila. (A) In vitro synthesized Drosophila Hsp83 3′‐UTR transcripts (1.407) that carry the HDE (‘+HDE’) are highly unstable in Xenopus oocytes and are almost completely degraded within 6 h of injection. (B) In contrast, in vitro synthesized Drosophila Hsp83 3′‐UTR transcripts that lack the HDE (‘ΔHDE’; 252Δ350) are stable and can still be readily detected 24 h after injection.

Discussion

Passage from maternal to zygotic control of development is thought to depend on the elimination of a subset of maternal transcript classes from the early embryo. For example, degradation of ubiquitous maternal string transcripts is necessary to permit zygotically driven, patterned domains of cell division after the MBT (Edgar and Datar, 1996). Here we have shown that the joint action of two RNA degradation pathways controls maternal transcript degradation and its timing in the early Drosophila embryo, and that at least one of these pathways (the ‘maternal’ pathway) is conserved in Xenopus.

Two genetic pathways ensure RNA degradation prior to the Drosophila MBT

The first degradation pathway is maternally encoded and does not require fertilization or zygotic gene transcription for its function. This ‘maternal’ pathway begins to function at or shortly after egg activation, and is sufficient for degradation of even abundant maternally loaded transcripts by 4–5 h after egg activation (Figure 9). The second ‘zygotic’ pathway requires fertilization and becomes active 1.5–2 h after fertilization. Acting alone, the zygotic pathway eliminates transcripts by 4 h after fertilization. Since the zygotic pathway can remove Hsp83 transcripts in 2 h, while the maternal pathway takes 4–5 h, the zygotic machinery is approximately twice as efficient as the maternal machinery. Under normal circumstances, relatively rare transcripts are degraded almost exclusively by the maternal pathway (e.g. nanos). In contrast, more abundant transcripts require the action of both pathways in order to be eliminated by the MBT (e.g. Hsp83, string). The maternal pathway initiates transcript degradation at or shortly after egg activation, while the zygotic pathway ensures degradation of abundant transcripts prior to the MBT.
image
Figure 9. Maternal and zygotic degradation activities cooperate to eliminate maternal transcripts by the MBT. Below the time scale on the x‐axis are shown the maternally driven mitoses (1–13) followed by interphase 14 (the MBT) during which zygotic products take over control of the cell cycle before mitosis 14, the first patterned and asynchronous mitosis. The y‐axis represents the [RNA] plotted on a logarithmic scale. In unfertilized eggs, an abundant transcript (e.g. Hsp83) lacking a maternal degradation element but carrying a zygotic degradation element is stable (ΔDE, blue). In contrast, such a transcript is degraded in fertilized, developing embryos by the zygotic degradation apparatus, starting ∼2 h after fertilization, such that transcripts are fully degraded by 4 h after fertilization (ΔDE, green). In the presence of both maternal and zygotic degradation elements, transcript degradation initiates at or shortly after activation of unfertilized eggs and is complete by 4 h after fertilization (+DE, black). In fertilized, developing embryos, transcripts carrying both maternal and zygotic degradation elements initially are degraded solely by the maternal apparatus (<2 h after fertilization) and then cooperatively by both the maternal and zygotic factors (>2 h after fertilization) (+DE, red). Only in this last case do abundant transcripts disappear by the MBT (2.0–2.5 h after fertilization). Rarer transcripts (e.g. nanos) are eliminated almost exclusively by the maternal machinery (yellow and grey). ΔDE, no maternal degradation element; +DE, maternal degradation element present; UF, unfertilized egg; F, developing embryo.
Our results predict four classes of maternal RNAs in the early embryo. The first class lacks both maternal and zygotic degradation elements and is stable (e.g. rpA1). The second class (if it exists—there are no known examples) would be degraded exclusively by the maternal machinery either because these transcripts lack zygotic degradation elements or because their low abundance enables them to be fully degraded by the maternal machinery prior to activation of the zygotic factors. A third class (if it exists—there are no known examples) would be degraded exclusively by the zygotic machinery, possibly because of a lack of elements that target the maternal factors. The fourth class (e.g. Hsp83, string and nanos) is degraded by the combined action of the maternal and zygotic machinery.
The dual mechanism for transcript degradation discovered here sheds light on two previously reported observations. First, it has been shown that modulation of the nuclear:cytoplasmic ratio affects the timing of disappearance of maternally synthesized string transcripts (Yasuda et al., 1991). Changing the nuclear:cytoplasmic ratio would not affect the maternal degradation machinery since it would be pre‐loaded into the embryo and its activity/concentration should be independent of that ratio. In contrast, changes in the nuclear:cytoplasmic ratio might be expected to alter the synthesis or local concentration of the zygotic degradation machinery, particularly if activation of this machinery is dependent on zygotic transcription. Thus, altering the nuclear:cytoplasmic ratio would alter the exact timing of string transcript disappearance by modulating the zygotic pathway. Secondly, our results explain why α‐amanitin injection into embryos prior to nuclear cycle 6 causes a delay in degradation of maternal string transcripts (α‐amanitin inhibits RNA polymerase II and thus zygotic transcription) (Edgar and Datar, 1996). If production or activation of the zygotic degradation machinery requires zygotic transcription, α‐amanitin would be expected to inhibit it while leaving the maternal degradation machinery unaffected. This would result in a delay in string degradation since only the maternally encoded degradation apparatus would be functional.
In summary, while previous results suggested that turnover of transcripts at the MBT might be controlled by zygotic products (O'Farrell et al., 1989; Yasuda et al., 1991), we have uncovered a dual maternal–zygotic degradation system that is likely to coordinate developmental events by ensuring the elimination of maternal transcripts prior to the MBT.

Cis‐acting elements target transcripts for degradation or for protection

We have shown that the maternal transcript degradation activity is present throughout the embryo and unfertilized egg. Deletion of specific cis‐acting 3′‐UTR sequences results in failure of maternally encoded transcript degradation. These elements do not share any obvious primary sequence or secondary structure, nor do they include any previously defined degradation tags (reviewed in Jacobson and Peltz, 1996). Despite the lack of sequence conservation, our data demonstrate that the degradation elements are interchangeable (the Hsp83 HDE can substitute for the nanos TCE, and vice versa). The cis‐acting elements defined here are not necessary for targeting transcripts to the zygotic degradation pathway since deletion of these elements has no detectable effect on the zygotic degradation process; in fact, none of our Hsp83 3′‐UTR deletions abrogates zygotic transcript degradation. There are three plausible explanations for this result: (i) there are redundant ‘zygotic’ degradation elements in the Hsp83 3′‐UTR; (ii) such elements reside in another part of the Hsp83 transcript [given the structure of the reporter transgenes, this could be the 5′‐UTR or the first 333 nucleotides of the open reading frame (ORF)]; or (iii) zygotic degradation is not a targeted event but is the default state.
For RNAs such as rpA1, which are stable in the early embryo, we do not yet know whether there are specific stabilization elements or whether transcript stability is a default state. Indirect evidence in support of the latter possibility comes from analyses of a transcript instability element in the ftz 3′‐UTR (Riedl and Jacobs‐Lorena, 1996). Addition of this element (FIE3) to rpA1 transcripts destabilizes them, suggesting that for rpA1 transcripts, stability is either the default state or that instability elements such as FIE3 are dominant to as yet undefined stability elements.
Our data indicate that it is possible to differentially stabilize transcripts carrying degradation tags in different regions of the cytoplasm and/or in different cell types. Specifically, we have shown that the maternal degradation apparatus is active throughout the egg and embryo, but that transcripts such as Hsp83, nanos and Pgc are protected from degradation in the pole plasm and pole cells. We have also been able to map sequences in the Hsp83 3′‐UTR that are necessary for this protection. The most likely candidate organelles for protecting transcripts in the polar plasm and pole cells are the polar granules themselves. First, one of the RNAs we have studied (Pgc) has been shown to be an integral component of the polar granules (Nakamura et al., 1996). Secondly, it has been shown that disruption of the polar granules results in failure of Hsp83, nanos and Pgc transcript protection in the polar plasm and pole cells, while ectopic assembly of polar granules at the anterior of the embryo results in ectopic protection of these transcripts anteriorly (Ephrussi and Lehmann, 1992; Ding et al., 1993a; Nakamura et al., 1996).

Role of transcript degradation and protection in cytoplasmic RNA localization

Differential stabilization of transcripts in specific regions of the cytoplasm represents one of several RNA localization mechanisms (reviewed in Bashirullah et al., 1998). Hsp83 is the only example of a transcript localized by a degradation–protection mechanism (Ding et al., 1993a; St Johnston, 1995). However, we have shown that nanos, Pgc and Hsp83 transcript elimination in the somatic region of the embryo is accomplished by the same RNA degradation machinery. Further, our data suggest that 99% of the Hsp83 transcripts and >99% of the nanos transcripts reside in somatic region. This last result is very similar to the 96% value reported recently for somatic nanos transcripts (Bergsten and Gavis, 1999). We speculate that the differences in localization patterns of transcripts that are obviously localized prior to egg activation and those which only appear localized well after egg activation may be largely quantitative.

Relationship of transcript degradation to translation and polyadenylation status

The literature contains numerous examples of a correlation between translational repression of mRNAs and their degradation (reviewed in Jacobson and Peltz, 1996; Cooperstock and Lipshitz, 1997). In the early Drosophila embryo, the best studied example is nanos: nanos transcripts outside of the posterior polar plasm are translationally repressed and are also unstable (Figure 4, and see also Dahanukar and Wharton, 1996; Gavis et al., 1996; Smibert et al., 1996). Deletion or mutation of cis‐acting elements that mediate translational repression results in transcript stabilization (Dahanukar and Wharton, 1996; Smibert et al., 1996; this study). Despite this correlation, our data demonstrate that there is no obligatory link between translational repression and transcript degradation. Replacing the element from the nanos 3′‐UTR that mediates translational repression and degradation (the TCE) with the Hsp83 degradation element (the HDE) is sufficient to allow maternal degradation of nanos [ΔTCE+HDE] transcripts. However, at the same time, the presence of severe head defects demonstrates that this transgenic RNA is in fact highly translated. Thus, transcript degradation can be unlinked from translational repression.
There are several examples of a correlation between polyadenylation status and transcript stability (reviewed in Jacobson and Peltz, 1996). In particular, long poly(A) tails are associated with translational activation and transcript stabilization, while short poly(A) tails are associated with translational repression and transcript destabilization. In the Drosophila embryo, it has been shown that several classes of maternal transcripts are cytoplasmically polyadenylated after fertilization, correlating with initiation of their translation (Sallés et al., 1994). The cortex and grauzone mutations perturb cytoplasmic polyadenylation (Lieberfarb et al., 1996), but transcript stability was not assayed in those studies. We have shown here that maternal transcripts are stabilized by cortex and grauzone mutations. The primary cause of the defects in embryos produced by cortex and grauzone females is unknown; their phenotypes are complex and several aspects of egg activation are abnormal (Lieberfarb et al., 1996; Page and Orr‐Weaver, 1996). Thus it is not yet possible to draw firm conclusions from these mutants regarding a possible link between cytoplasmic polyadenylation status and transcript stability.

Evolutionary conservation of transcript degradation pathways

We have shown that Xenopus oocytes and early embryos contain an activity highly related to the maternal transcript degradation activity we have defined in Drosophila. Remarkably, the Drosophila HDE is recognized by this activity, and in vitro sythesized, injected transcripts are degraded rapidly, disappearing within a few hours. Removal of the HDE stabilizes these transcripts for at least 24 h. These results suggest that evolutionarily highly conserved pathways may control transcript degradation and localization throughout metazoa, opening up the possibility of exploiting the particular advantages of different model systems for future mechanistic and functional analysis.

Materials and methods

Drosophila whole‐mount RNA in situ hybridization

RNA in situ hybridization with DNA probes was performed as described previously, with minor modifications (Tautz and Pfeifle, 1989; Ding et al., 1993a, b). Digoxigenin‐labeled antisense RNA probes were made using the Megascript RNA transcription kit (Ambion, Inc.) according to the manufacturer's instructions, and included digoxigenin‐labeled UTP (Boehringer Mannheim). The RNA probes were subjected to limited alkaline hydrolysis by incubation in an equal volume of 100 mM NaCO3 pH 10.2 for 1 h at 60°C followed by ethanol precipitation. Pre‐hybridization and hybridization were carried out at 55°C for RNA probes and at 48°C for DNA probes, and detection of the probe was as previously described (Tautz and Pfeifle, 1989). After hybridization, eggs or embryos were cleared in 50% glycerol followed by 70% glycerol. For exact staging, nuclei were visualized with 1 μg/ml 4′,6‐diamidino‐2‐phenylindole (DAPI) in phosphate‐buffered saline (PBS) for 5 min prior to clearing of the embryos. Eggs or embryos were then individually mounted in 70% glycerol and images were captured using a Spot cooled‐CCD camera (Diagnostic Instruments, Inc.) mounted on a Zeiss Axioplan microscope. Adobe Photoshop software was used to process the images. For transgenic lines, at least two independently derived lines were analyzed for each construct.

Drosophila RNA extraction and analysis

RNA samples were isolated from staged egg collections by phenol extraction using Trizol (BRL). Equal amounts of total RNA were electrophoresed in 1.2% agarose/formaldehyde/MOPS gels (Sambrook et al., 1989), and transferred onto nylon membranes (Amersham). Filters were pre‐hybridized for at least 1 h at 63°C in hybridization buffer containing 0.5 M Na‐phosphate/7% SDS. 32P‐Labeled random primed DNA probes were added and hybridization was performed overnight at the same temperature. Filters were rinsed twice at 60°C in 30 mM Na‐phosphate/0.1% SDS, and exposed to Kodak XAR5 film at −80°C with an intensifying screen. Films were also exposed to a Molecular Dynamics phosphor screen, and quantitation was performed using ImageQuant software. Hsp83, string and nanos transcript levels were normalized to rpA1 mRNA levels to control for variations in loading. The number of repetitions of each experiment was as follows: Hsp83 endogenous transcripts, fertilized = 3; Hsp83 endogenous transcripts, unfertilized = 2; Hsp83 ΔHDE transcripts, fertilized = 2; Hsp83 ΔHDE transcripts, unfertilized = 1; string transcripts, fertilized = 2; string transcripts, unfertilized = 2; nanos transcripts, fertilized = 1; nanos transcripts, unfertilized = 1.

Construction of transgenes for Drosophila germline transformation

Standard protocols were used (Sambrook et al., 1989).
Hsp83–lacZ transgenes. The following approach was taken to generate the series of lacZ–Hsp83 3′‐UTR constructs that include the 5′‐Hsp83 enhancer sequences and promoter (Figure 3B). Previous reports (Kim‐Ha et al., 1993) suggested that a full‐length lacZ ORF might not be completely transcribed in vivo. Therefore, we inserted the Hsp83 3′‐UTR fragments downstream of a truncated lacZ tag. We accomplished this by constructing pB83Z, a Bluescript subclone that contains all of the Hsp83 5′ upstream region, the first exon, the intron and the first 111 codons of the ORF fused to 603 bp of lacZ sequence (Halsell, 1995). At the 3′ end of the lacZ sequence are AatII, HindIII and KpnI cloning sites for inserting the Hsp83 3′‐UTR fragments. The template for the PCR was Eco9, an 8 kb EcoRI fragment subcloned in Bluescript that contains the full‐length Hsp83 transcription unit. 5′ Fragments of the 3′‐UTR were PCR amplified and were flanked with AatII and HindIII restriction sites, while 3′ fragments of the 3′‐UTR were flanked with HindIII and KpnI restriction sites. The position of the HindIII restriction site within the amplified fragments allowed us to generate nested 5′ deletion constructs, the 1.350hs 3′ deletion construct and a series of internal deletions within the 3′‐UTR. 3′‐UTR constructs with internal deletions (the ‘Δ’ series) were generated by first subcloning an AatII–HindIII 5′ fragment into pB83Z and subsequently adding a HindIII–KpnI 3′ fragment. After subcloning into pB83Z, a NotI–KpnI fragment was isolated and subcloned into NotI–KpnI‐digested CaSpeR4 (Thummel and Pirrotta, 1991). As a control, the Hsp70 3′‐UTR was also cloned independently of any Hsp83 3′‐UTR sequence (Figure 3A, ‘Hsp70’). The Hsp70 fragment was generated by PCR from a CaSpeR‐hs template (Thummel and Pirrotta, 1991).
nanosHsp83 hybrid transgenes. (i) nanos[ΔTCE]: a genomic 3′ fragment (BglII–EcoRI) containing a nanos 3′‐UTR lacking the TCE was subcloned into pCaSpeR2 (Thummel and Pirrotta, 1991). Subsequently a genomic 5′ fragment (BamHI) extending from the nanos cis‐regulatory sequences to the end of the ORF was inserted. (ii) nanos[ΔTCE + HDE]: a genomic 3′ fragment (BglII–EcoRI) containing a nanos 3′‐UTR lacking the TCE was subcloned into pCaSpeR2. The Hsp83 HDE was PCR amplified incorporating 5′‐BamHI and 3′‐BglII sites. This fragment was subcloned upstream of the nanos 3′ fragment. Subsequently a genomic 5′ fragment (BamHI) extending from the nanos cis‐regulatory sequences to the end of the ORF was inserted. (iii) Hsp83[ΔHDE + TCE]: the nanos TCE was amplified by PCR incorporating HindIII sites at the ends, and inserted into lacZ‐tagged pBS Hsp83[ΔHDE] (253Δ350) at the HindIII site.

Drosophila germline transformation

Germline transformation was carried out using standard procedures (Rubin and Spradling, 1982). w1118 embryos were co‐injected with 500 μg/ml of the Hsp83–lacZ construct and 100 μg/ml of pHSπ helper plasmid (Steller and Pirrotta, 1985). All transgenic insertions were homozygosed or balanced over CyO or TM3.

Drosophila stocks and egg collection

Drosophila melanogaster Oregon‐R and w1118 stocks are described in Lindsley and Zimm (1992). The following ‘class 2’ and ‘class 3’ maternal effect lethal mutants (Schüpbach and Wieschaus, 1989) were assayed for transcript degradation in unfertilized eggs: cell‐PK42, cell‐RH36, cortex, cribble, grauzone, luckenhaft, syn‐Hi10, presto, scraps, valois. For collection of unfertilized eggs, males of genotype T(Y;2)bwDRev#11, cn bwDRev#11 mr2/SM6a were crossed to appropriate females (Reed and Orr‐Weaver, 1997). Terminal duplication segregants are male sterile (Reed and Orr‐Weaver, 1997).

Drosophila embryo cuticle preparations

For cuticle preparations, wild‐type and mutant embryos were dechorionated, devitellinized and mounted as described previously (Raz and Shilo, 1993).

RNA analysis in Xenopus

Drosophila Hsp83 full‐length 3′‐UTR (1.407, see above and Figure 3) was transcribed from a pBluescriptSK subclone using T3 RNA polymerase after linearization at the KpnI site. ΔHDE 3′‐UTR (252Δ350, see above and Figure 3) was transcribed from a PCR‐amplified fragment that included the T3 RNA polymerase promoter at the 5′ end of the 3′‐UTR sequence. Transcription was carried out in the presence of digoxigenin‐11‐UTP (Kloc et al., 1996). For oocyte injections, ovaries were surgically removed from anaesthetized X.laevis females, stage 6 oocytes were manually defolliculated in calcium‐deficient buffer (1× OR2) and were stored, injected and cultured for 0.5, 3, 6 or 24 h in 1× modified Barth's buffer. For embryo injections, recently fertilized eggs were collected, injected and allowed to develop to stage 12 (gastrula stage, ∼24 h). For the time course in stage 6 oocytes (Figure 8A), 500 ng of synthetic labeled RNA was injected into the cytoplasm whereas for comparative analysis of +HDE and ΔHDE 3′‐UTRs (Figure 8B), ∼100 ng of RNA was injected. Total RNA was prepared from injected oocytes or embryos, separated on a formaldehyde gel and blotted as previously described (Kloc et al., 1989). The blot was blocked for 1.5 h in 2% blocking solution (BMB, Genius kit) and was then incubated for 1.5 h with a 1:5000 dilution of anti‐digoxigenin–alkaline phosphatase antibody in 2% blocking buffer. Subsequently blots were washed twice for 15 min each wash in G1 buffer and stained in NBT/BCIP solution. All of the preceding used BMB Genius kit reagents and protocols.

Acknowledgements

We thank C.Thummel for transformation vectors, T.Cutforth and G.Rubin for providing the Eco9 clone and 63B‐T2 DNA sequence, B.Reed for the T(Y;2)bwDRev#11, cn bwDRev#11 mr2/SM6a flies, T.Schüpbach for mutant fly lines, C.Smibert for the nanos clone, E.Gavis for communicating unpublished results, and R.McInnes, H.Krause, M.Lamka, S.Egan, S.Lewis and C.C.Hui for critical comments on the manuscript. S.R.H. was supported in part by a National Research Service Award (T32GM07616) and graduate fellowships from the Lucille P.Markey Charitable Trust and Howard Hughes Medical Institute. J.K.H. is supported by a Medical Research Council/Pharmaceutical Manufacturers Association of Canada Health Program Fellowship. R.L.C. has been supported in part by a studentship from the Medical Research Council (Canada) and a University of Toronto Open Scholarship. This research has been supported by research grants from the National Institutes of Health, USPHS (GM50221 to L.D.E.), the National Science Foundation (IBN‐9418453 to H.D.L.) and the Medical Research Council of Canada (MT13208 and, currently, MT14409, both to H.D.L).

References

Bashirullah A, Cooperstock RL and Lipshitz HD (1998) RNA localization in development. Annu Rev Biochem, 67, 335–394.
Bergsten SE and Gavis ER (1999) Role for mRNA localization in translational activation but not spatial restriction of nanos RNA. Development, 126, 659–669.
Campos‐Ortega JA and Hartenstein V (1998) The Embryonic Development of Drosophila melanogaster. Springer‐Verlag, Heidelberg, Germany.
Cooperstock RL and Lipshitz HD (1997) Control of mRNA stability and translation during Drosophila development. Semin Cell Dev Biol, 8, 541–549.
Dahanukar A and Wharton RP (1996) The Nanos gradient in Drosophila embryos is generated by translational regulation. Genes Dev, 10, 2610–2620.
Davidson EH (1986) Gene Activity in Early Development. Academic Press, Orlando, FL.
Ding D, Parkhurst SM, Halsell SR and Lipshitz HD (1993a) Dynamic Hsp83 RNA localization during Drosophila oogenesis and embryogenesis. Mol Cell Biol, 13, 3773–3781.
Ding D, Parkhurst SM and Lipshitz HD (1993b) Different genetic requirements for anterior RNA localization revealed by the distribution of Adducin‐like transcripts during Drosophila oogenesis. Proc Natl Acad Sci USA, 90, 2512–2516.
Edgar BA and O'Farrell PH (1989) Genetic control of cell division patterns in the Drosophila embryo. Cell, 57, 177–187.
Edgar BA and O'Farrell PH (1990) The three postblastoderm cell cycles of Drosophila embryogenesis are regulated in G2 by string. Cell, 62, 469–480.
Edgar BA and Datar SA (1996) Zygotic degradation of two maternal Cdc25 mRNAs terminates Drosophila's early cell cycle program. Genes Dev, 10, 1966–1977.
Edgar BA, Kiehle CP and Schubiger G (1986) Cell cycle control by the nucleo‐cytoplasmic ratio in early Drosophila development. Cell, 44, 365–372.
Ephrussi A and Lehmann R (1992) Induction of germ cell formation by oskar. Nature, 358, 387–392.
Foe VA (1989) Mitotic domains reveal early commitment of cells in Drosophila embryos. Development, 107, 1–23.
Foe VA and Alberts BM (1983) Studies of nuclear and cytoplasmic behaviour during the five mitotic cycles that precede gastrulation in Drosophila embryogenesis. J Cell Sci, 61, 31–70.
Gavis ER and Lehmann R (1994) Translational regulation of nanos by RNA localization. Nature, 369, 315–318.
Gavis ER, Lunsford L, Bergsten SE and Lehmann R (1996) A conserved 90 nucleotide element mediates translational repression of nanos RNA. Development, 122, 2791–2800.
Halsell SR (1995) Expression and Localization of Hsp83 RNA in the Early Drosophila Embryo. Ph.D. Thesis, Division of Biology, California Institute of Technology, Pasadena, CA.
Jacobson A and Peltz SW (1996) Interrelationships of the pathways of mRNA decay and translation in eukaryotic cells. Annu Rev Biochem, 65, 693–739.
Kim‐Ha J, Webster PJ, Smith JL and Macdonald PM (1993) Multiple RNA regulatory elements mediate distinct steps in localization of oskar mRNA. Development, 119, 169–178.
Kloc M, Larabell C and Etkin LD (1996) Elaboration of the messenger transport organizer pathway for localization of RNA to the vegetal cortex of Xenopus oocytes. Dev Biol, 180, 119–130.
Kloc M, Miller M, Carrasco A, Eastman E and Etkin L (1989) The maternal store of the xlgv7 mRNA in full‐grown oocyte is not required for normal development in Xenopus. Development, 107, 899–907.
Lieberfarb ME, Chu T, Wreden C, Theurkauf W, Gergen JP and Strickland S (1996) Mutations that perturb poly(A)‐dependent maternal mRNA activation block the initiation of development. Development, 122, 579–588.
Lindsley DL and Zimm G (1992) The Genome of Drosophila melanogaster. Academic Press, San Diego, CA.
Mahowald AP, Goralski TJ and Caulton JH (1983) In vitro activation of Drosophila eggs. Dev Biol, 98, 437–445.
Merrill PT, Sweeton D and Wieschaus E (1988) Requirements for autosomal gene activity during precellular stages of Drosophila melanogaster. Development, 104, 495–509.
Nakamura A, Amikura R, Mukai M, Kobayashi S and Lasko PF (1996) Requirement for a noncoding RNA in Drosophila polar granules for germ cell establishment. Science, 274, 2075–2079.
O'Farrell PH, Edgar BA, Lakich D and Lehner CF (1989) Directing cell division during development. Science, 246, 635–640.
Page AW and Orr‐Weaver TL (1996) The Drosophila genes grauzone and cortex are necessary for proper female meiosis. J Cell Sci, 109, 1707–1715.
Raz E and Shilo B‐Z (1993) Establishment of ventral cell fates in the Drosophila embryonic ectoderm requires DER, the EGF receptor homolog. Genes Dev, 7, 1937–1948.
Reed B and Orr‐Weaver T (1997) The Drosophila gene morula inhibits mitotic functions in the endo cell cycle and the mitotic cell cycle. Development, 124, 3543–3553.
Riedl A and Jacobs‐Lorena M (1996) Determinants of Drosophila fushi tarazu mRNA instability. Mol Cell Biol, 16, 3047–3053.
Rubin GM and Spradling AC (1982) Genetic transformation of Drosophila with transposable elements. Science, 218, 348–353.
Sallés FJ, Lieberfarb ME, Wreden C, Gergen JP and Strickland S (1994) Coordinate initiation of Drosophila development by regulated polyadenylation of maternal messenger RNAs. Science, 266, 1996–1999.
Sambrook J, Fritsch EF and Maniatis T (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Schüpbach T and Wieschaus E (1989) Female sterile mutations on the second chromosome of Drosophila melanogaster. I. Maternal effect mutations. Genetics, 121, 101–117.
Sibon OCM, Stevenson VA and Theurkauf WE (1997) DNA‐replication checkpoint control at the Drosophila midblastula transition. Nature, 388, 93–97.
Smibert CA, Wilson JE, Kerr K and Macdonald PM (1996) Smaug protein represses translation of unlocalized nanos mRNA in the Drosophila embryo. Genes Dev, 10, 2600–2609.
St Johnston D (1995) The intracellular localization of messenger RNAs. Cell, 81, 161–170.
Steller H and Pirrotta V (1985) A transposable P vector that confers G418 resistance to Drosophila larvae. EMBO J, 4, 167–171.
Tautz D and Pfeifle C (1989) A non‐radioactive in situ hybridization method for the localization of specific RNAs in Drosophila embryos reveals translational control of the segmentation gene hunchback. Chromosoma, 98, 81–85.
Thummel CS and Pirrotta V (1991) New pCaSpeR P element vectors. Drosophila Information Newslett, 2, 19
Wang C and Lehmann R (1991) Nanos is the localized posterior determinant in Drosophila. Cell, 66, 637–647.
Wharton RP and Struhl G (1991) RNA regulatory elements mediate control of Drosophila body pattern by the posterior morphogen nanos. Cell, 67, 955–967.
Yasuda GK, Baker J and Schubiger G (1991) Temporal regulation of gene expression in the blastoderm Drosophila embryo. Genes Dev, 5, 1800–1812.
Zalokar M and Erk I (1976) Division and migration of nuclei during early embryogenesis of Drosophila melanogaster. J Microsc Biol Cell, 25, 97–106.
Zimmerman JL, Petri W and Meselson M (1983) Accumulation of a specific subset of D.melanogaster heat shock mRNAs in normal development without heat shock. Cell, 32, 1161–1170.

Information & Authors

Information

Published In

The EMBO Journal cover image
The EMBO Journal
Vol. 18 | No. 9
4 May 1999
Table of contents
Pages: 2610 - 2620

Article versions

Submission history

Received: 22 December 1998
Revision received: 5 March 1999
Accepted: 5 March 1999
Published online: 4 May 1999
Published in issue: 4 May 1999

Permissions

Request permissions for this article.

Keywords

  1. Drosophila
  2. midblastula transition (MBT)
  3. localization
  4. stability
  5. Xenopus

Authors

Affiliations

Arash Bashirullah
Program in Developmental Biology, Research Institute, The Hospital for Sick Children 555 University Avenue Toronto Ontario M5G 1X8 Canada
Division of Biology, California Institute of Technology Pasadena CA USA
Susan R. Halsell
Division of Biology, California Institute of Technology Pasadena CA USA
Department of Cell Biology, Duke University Medical Center, Box 3709 Durham NC 27710 USA
Ramona L. Cooperstock
Program in Developmental Biology, Research Institute, The Hospital for Sick Children 555 University Avenue Toronto Ontario M5G 1X8 Canada
Department of Molecular and Medical Genetics, University of Toronto Toronto Canada
Malgorzata Kloc
Department of Molecular Genetics, University of Texas M.D.Anderson Cancer Center Houston TX USA
Angelo Karaiskakis
Program in Developmental Biology, Research Institute, The Hospital for Sick Children 555 University Avenue Toronto Ontario M5G 1X8 Canada
William W. Fisher
Program in Developmental Biology, Research Institute, The Hospital for Sick Children 555 University Avenue Toronto Ontario M5G 1X8 Canada
Division of Biology, California Institute of Technology Pasadena CA USA
Exelixis Pharmaceuticals, Inc. 260 Littlefield Avenue South San Francisco CA 94401 USA
Weili Fu
Program in Developmental Biology, Research Institute, The Hospital for Sick Children 555 University Avenue Toronto Ontario M5G 1X8 Canada
Department of Neuroscience, University of Pennsylvania Medical School Philadelphia PA 19104 USA
Jill K. Hamilton
Program in Developmental Biology, Research Institute, The Hospital for Sick Children 555 University Avenue Toronto Ontario M5G 1X8 Canada
Laurence D. Etkin
Department of Molecular Genetics, University of Texas M.D.Anderson Cancer Center Houston TX USA
Howard D. Lipshitz* [email protected]
Program in Developmental Biology, Research Institute, The Hospital for Sick Children 555 University Avenue Toronto Ontario M5G 1X8 Canada
Division of Biology, California Institute of Technology Pasadena CA USA
Department of Molecular and Medical Genetics, University of Toronto Toronto Canada

Notes

*
Corresponding author. E-mail: [email protected]

Metrics & Citations

Metrics

Citations

Download Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Select your manager software from the list below and click Download.

Citing Literature

  • Maternal mRNA deadenylation and allocation via Rbm14 condensates facilitate vertebrate blastula development, The EMBO Journal, 10.15252/embj.2022111364, 42, 3, (2022).
  • Smaug assembles an ATP‐dependent stable complex repressing nanos mRNA translation at multiple levels, The EMBO Journal, 10.1038/emboj.2010.283, 30, 1, (90-103), (2011).
  • The TATA‐binding protein regulates maternal mRNA degradation and differential zygotic transcription in zebrafish, The EMBO Journal, 10.1038/sj.emboj.7601821, 26, 17, (3945-3956), (2007).
  • Dynamical modeling of syncytial mitotic cycles in Drosophila embryos, Molecular Systems Biology, 10.1038/msb4100171, 3, 1, (2007).
  • Dominant‐negative mutants reveal a role for the Cdk7 kinase at the mid‐blastula transition in Drosophila embryos, The EMBO Journal, 10.1093/emboj/19.7.1567, 19, 7, (1567-1575), (2000).

View Options

View options

PDF

View PDF

Get Access

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media