Activation of the Maternal-to-Zygotic Transition (MZT) in Drosophila Development

During oogenesis in all animals (including humans), maternal effect gene products are deposited into the developing oocyte, which regulate early embryonic development (Figure 1). In Drosophila, these maternal effectors include, for example, the anterior-posterior defining morphogen Bicoid (Bic) and the dorsal-ventral defining morphogen Dorsal (Dl). After the initial stages of embryogenesis the control of embryonic development is shifted from the maternal effectors to the zygotic genome in a process known as the Maternal-to-Zygotic Transition (MZT). The MZT is characterized by a flurry of activity that includes the degradation of maternal transcripts and the initiation of zygotic genome transcription.

Figure 1 | Schematic representation of early oogenesis (A) and stage 9 egg chamber (B). The germline and specified somatic cells are indicated individually. In both panels, anterior is to the left (4)

In Drosophila, the MZT is known to occur 1-3 hours post-fertilization (hpf) of the zygote, however, how this process is temporally regulated was unclear until recent papers published by Liang et al1 and Nien et al2. Liang et al identified a protein, Zelda (Zld), as a key activator of the early zygotic genome, which was the first identified zygotic-genome-activator in any organism1. Nien et al expanded upon the findings from Liang et al and showed that Zld is a far reaching transcription factor required for accurate spatial and temporal expression of numerous gene networks involved with cellularization, sex determination, neurogenesis, and pattern formation2.

Discovery of a key activator of the Drosophila zygotic genome during early embryonic development

Since the entire Drosophila genome has been sequenced, it is well known that a collection of cis-regulatory heptamers, known as TAGteam sites, are located upstream of many early zygotic genes. It has also been demonstrated (for a small set of these genes) that the TAGteam sites are necessary for timely expression of these genes1, but the activator of these sites had yet to be identified.  Liang et al used a yeast one-hybrid screen to identify the unknown activator.

A yeast one-hybrid screen involves:

  1. Constructing a plasmid containing a hybrid bait sequence::reporter sequence vector
  2. Constructing a library of activation vectors
  3. Transfecting yeast with the products from 1. and 2.
  4. Identifying the colonies of yeast that express the reporter
  5. Amplifying and sequencing the library activation vectors present in those colonies3

What results from such a screen is identification of the sequences in the activation library that potentially code the activator of the bait sequence. So in this study, the bait sequence was one that contained several TAGteam sites (the Zen enhancer sequence, Figure 2a), the activator library was Drosophila’s genome, and the activator is the unknown binder of TAGteam sites.

Figure 2 | TAGteam sites bind Zld and mediate transcriptional activation. a, DNA sequence of the 91-bp zen enhancer (upper-case) plus surrounding sequences (lower-case). Base substitutions are in purple. b, Schematic organization of the zld locus (CG12701; Flybase) with the transcription start sites for the RNAisoforms RB andRA. The P{RS3}UM8171-3 insertion site is between 2661 and 2660 bp. The nucleotides deleted in zld294 and zld681 are indicated as a blank space between solid lines. c, Zld binding to oligonucleotides containing different TAGteam sites (denoted beneath each section of the gel). The first lane in each section contains free probe, the second lane contains probe plus 10 ng GST–ZldC, and the third lane contains probe plus 30 ng GST–ZldC. d, S2 cells were transfected with 0 ng (blue bar), 50 ng (red bar) or 100 ng (yellow bar) of plasmid expressing zld under control of the inducible metallothionein promoter, the zen(91)-lacZ or zen(91m)-lacZ reporter plasmids, and the luciferase control. Error bars, s.e.m.; n53.

The red letters in the zen sequence indicate TAGteam binding sites. Identified from this screen was the gene encoding Zelda (Figure 2b). To ensure that Zelda is in fact a binder of the TAGteam sites in vitro, Liang et al conducted gel shift assays with different portions of the Zen enhancer sequence (as seen as the underlined sequences in Figure 2a), and the results of the gel shift assay are presented in Figure 2c. The results of the gel shift assay indicate convincingly that Zelda does bind the TAGteam sites, and does so with differing affinities1. Also shown in Figure 1c is that Zelda binding is highly specific for TAGteam sites in that no Zelda bound a mutated TAGteam sequence in lanes 10-12 (mutated sequences indicated by purple letters in Figure 2a). To confirm in vivo transcription activation, Liang et al co-expressed the Zelda gene with a gene fusion of the (Zen enhancer) zen(91)::LacZ reporter in Drosophila S2 cells. In Figure 1d, transcription activation increases with increasing Zelda concentration for the zen(91)::LacZ fusion, whereas the zen(91m)::LacZ fusion (mutated zen(91)) was unaffected by increases in Zelda concentration1.

To investigate Zelda’s role in early zygotic gene expression, Liang et al had to engineer a Zelda-null strain of Drosophila and this was done by inducing clones of zld294 (Figure 2b) mutant germ cells in female flies. The resulting embryos were null for maternal Zelda (M zld) and male embryos were null for maternal and zygotic Zelda (M Z zld). Zelda transcripts in embryos were visualized by in situ hybridization (Figure 3).

Figure 3 | Maternal zld transcripts are lost as zygotic zld is activated in cycle 14. a–j, Wild-type (a–d) and zld294 (e–j) ovaries (a) and embryos (b–j) were hybridized with zld (a–f and h–j) or bcd (g) RNA probes. a, Midstage (left) and late-stage (right) egg chambers with zld transcripts in the nurse cells (nc) but not the columnar follicle cells that overlay the oocyte. b, An unfertilized egg is shown. c, A cycle 14 embryo undergoing cellularization. d, A late-stage embryo is shown. e, An M1Z2 zld cycle 14 embryo showing that maternal zld transcripts have disappeared. f, An M2 zld cycle 10–11 embryo is shown. g, An M2 zld cycle 14 embryo has a normal bcd pattern. h, An M2Z2 zld late cycle 14 embryo showing anomalous distribution of cytoplasm (arrows). i, AnM2Z1 zld early cycle 14 embryo showing onset of zygotic zld expression. j, AnM2Z1 zld late cycle 14 embryo showing abnormalities (arrows).

Looking at Figure 3h, one can see that no Zelda transcripts are present in the M Z zld embryo and displays an abnormal phenotype compared to the wild-type embryo in Figure 3c. Additionally, the M Z+ zld embryos do have Zelda transcripts present, but still depict an abnormal phenotype (Figures 3i-j). These data together strongly indicate that Zelda, both maternal and zygotic, is necessary for expression of zygotic patterning genes and proper embryonic development1. To test this, Liang et al investigated the expression of a variety of zygotic genes including Sry-α, sisB, and zen. As expected, the gene expression patterns of these 3 genes were severely affect in zld (maternal and zygotic Zelda-null) embryos (Figures 4b,d,f, respectively) as compared to wild-type expression (Figures 4a,c,e, respectively).

Figure 4 | Zld plays a role in zygotic gene activation and maternal RNA degradation during the MZT. a–j, Wild-type (WT; left) and M2Z2 zld294 (zld2; right) mitotic cycle 12–14 embryos were hybridized as indicated. k, Summary of expression profiles of 1–2 h wild-type and M2 zld294 embryos. Fold ch, fold change with respect to wild type (genes absent in the array data are not included). l, Percentage of genes for which there is expression data3,5,19 described as maternal (M), zygotic (Z) or both (MZ) in the downregulated (>=2-fold; n=105) and upregulated (>=1.5-fold; n=263) gene sets.

Also shown in Figure 4g-h are the gene expression patterns of sog in wild-type and zld, respectively. The narrowing region of sog (which is acted on by both Dorsal and Zelda) indicate a combinatorial effect of Dorsal and Zelda establish the range of sog expression1.  The abnormal gene expression profiles in Figures 4a-j indicate that Zelda is a far reaching transcription factor. Liang et al tested this hypothesis by investigating the expression change of a large set of genes in wild-type and zld embryos. Most genes were found to be unaffected between the wild-type and zld, while some genes were downregulated in zld embryos and interestingly some were upregulated (Figure 4k). Categorizing the up- and downregulated genes by either maternal (M), zygotic (Z), or both maternal and zygotic (MZ) revealed that in zld embryos a majority of the upregulated genes are maternal (58%) and a majority of downregulated genes are zygotic (82%) (Figure 4l). This very strongly indicates that Zelda is important for the degradation of maternal transcripts and to a higher extent for the activation of zygotic transcription1.

Summary

  • Zelda (Zld) protein is uniformly expressed in the early Drosophila embryo1
  • Zld specifically binds a collection of cis-regulatory heptamers known as TAGteam sites1
  • Zld-null embryos have abnormal cellular blastoderm phenotypes1
  • Zld-null embryos fail to transcribe gene clusters involved with cellularization, sex determination, and pattern formation1
  • Zld is essential for initiation of zygotic genome transcription and RNA degradation during MZT1

Strengths

  • The progression of experiments showed the validity of the results as both in vitro expression studies and in vivo patterning both strongly indicated Zelda as an important transcription factor in body pattern establishment.
  • The time scale of studies indicate a strong relation between on-time developmental patterning and expression of Zelda.
  • Morphological investigation of how the cellularization process is affected by Zelda expression strongly indicates that Zelda is important to the MZT as cellularization occurs during that period.

Weaknesses

  • With Zelda being a uniform transcription factor, it would have been interesting to see what Zelda binds/affects expression of over the entire genome of Drosophila.
  • Could have done more quantitative analysis of the timing of expression of Zelda and target genes to truely understand how Zelda expression affects the temporal regulation of target genes.
  • It appears that Zelda works in tandem with other transcription factors (e.g., Dorsal), it would have been interesting to see if Zelda works in tandem with other key transcriptional regulators such as Bicoid or Hunchback (Hb).

References

  1. Liang HL, Nien CY, Liu HY, Metzstein MM, Kirov N, and Rushlow C.  The zinc-finger protein Zelda is a key activator of the early zygotic genome in Drosophila. Nature Letters 456, 400-403 (2008).  http://www.nature.com/nature/journal/v456/n7220/full/nature07388.html
  2. Nien CY, Liang HL, Butcher S, Sun Y, Fu S, Gocha T, Kirov N, Manak JR, and Rushlow C. Temporal coordination of gene networks by Zelda in the early Drosophila embryo. PLoS Genetics 7, 1-16 (2011).  http://www.plosgenetics.org/article/info%3Adoi%2F10.1371%2Fjournal.pgen.1002339
  3. Ouwerkerk P, and Meijer AH. Yeast one-hybrid screening for DNA-protein interactions. Current Protocols in Molecular Biology (2011).  http://onlinelibrary.wiley.com.prx.library.gatech.edu/doi/10.1002/0471142727.mb1212s55/full
  4. Li Q, Xin T, Chen W, Zhu M, Li M. Lethal(2)giant larvae is required in the follicle cells for formation of the initial AP asymmetry and the oocyte polarity during Drosophila oogenesis. Nature Cell Research 18, 372-384 (2008).  http://www.nature.com/cr/journal/v18/n3/full/cr200825a.html

Leave a Reply

Your email address will not be published. Required fields are marked *