Coordinating development

A key feature of animal development is that the body axes are aligned correctly with respect to each other, indicating that their formation must be tightly coordinated. The first insight into how such coordination is achieved came in 1995 from a paper describing studies in Drosophila carried out by Daniel St Johnston and colleagues.

In flies, anterior–posterior (AP) polarity originates when the oocyte moves to the posterior end of the surrounding egg chamber and signals to nearby somatic follicle cells to assume a posterior fate. Later, these follicle cells signal back to the oocyte, causing polarization of the microtubule cytoskeleton. This leads to the localization of bicoid and oskar mRNAs to opposite ends of the oocyte, where these transcripts later direct the development of anterior and posterior features, respectively.

St Johnston's group suspected that a gene called gurken might be important in this process. gurken was already known to be involved in establishing dorso–ventral (DV) polarity: localized secretion of gurken protein at the upper side of the oocyte signals to the adjacent follicle cells to assume a dorsal fate, a key step in setting up the DV axis. However, the phenotype of gurken mutants indicated that this gene has an earlier role in determining AP polarity. St Johnston and colleagues confirmed this, showing that gurken is required in the oocyte for the specification of posterior follicle cells.

Striking defects in AP polarity were also seen in oocytes in gurken mutants: the normal localization of bicoid and oskar mRNAs to opposite ends of the oocyte was lost. This indicated that gurken signalling sets up the AP polarization of the microtubule cytoskeleton that directs the correct transport of these mRNAs. To test this, the authors used a fusion of kinesin — a protein that normally moves towards the plus ends of microtubules at the posterior end of the oocyte — with β-galactosidase. In gurken mutants, this protein failed to move to the posterior pole, confirming the loss of microtubule polarity.

As well as being crucial for setting up the AP axis, microtubule polarization is also essential for DV polarity. The localized secretion of gurken that initiates dorsal follicle-cell fate depends on the movment of the oocyte nucleus along microtubules to the anterior–dorsal corner of the oocyte, leading to the accumulation of gurken mRNA close to the dorsal surface. The fact that this occurs after the posterior follicle cells have been specified indicated that the gurken-dependent microtubule polarization that initiates AP polarity might also be responsible for determining the position of the DV axis. This was confirmed when it was shown that the oocyte nucleus does not migrate to its correct position in gurken mutants. In addition, gurken mRNA fails to localize to the correct position at the anterior–dorsal corner of the oocyte when the gurken receptor in the follicle cells is absent.

As St Johnston and colleagues put it in their paper, “It is hard to imagine how two perpendicular axes can be defined unless the orientation of one depends on the other.” This is exactly what this study showed: the body axes in Drosophila are coordinated by the initial establishment of AP polarity, which then determines where the DV axis is formed.