When you get immersed in the Drosophila literature, it's easy to lose sight of an important fact: flies are weird. Honestly, every species has unique properties, so it's not that we wouldn't be saying exactly the same thing about some other species if it had become the dominant experimental model system, but it's good to constantly remind ourselves of the diversity we see in biology.

One of the most noticeable properties of early fly development is the existence of an extensive maternal contribution to the embryo's organization. Mother Fly goes to a fair amount of effort to stamp her eggs with molecular labels that say, "This End Up". Is this a common strategy? Well, yes—lots of animals give their progeny a boost in reliability of development by incorporating an asymmetric distribution of informational macromolecules. There seem to be some exceptions, though, and they happen to be of particular interest to us, because they are us: mammals. Mammalian ova lack any overt asymmetries, and their early cleavages are simple and equal, producing a clump of cells called a morula with no discernible up, down, left, right, back, or front. So how does a young mammalian zygote figure out which end is supposed to be the head and which the tail?

There are some small symmetry-breaking cues that may play a role. Two that have been suspected are the sperm entry point, and the location of the second polar body. Either or both of these could define what is called the animal-vegetal axis (which is sort of roughly dorsal/anterior-ventral/posterior, and also corresponds to some distinctions in tissue type...it's complicated. Maybe I'll explain it another day.) In some vertebrates, like the frog, we do see a correlation between where the sperm enters the egg and the polarity of the embryo.

The other potential cue is the formation of the second polar body. The egg divides meiotically, to produce a haploid cell from a diploid progenitor. One of the peculiarities of the meiotic divisions of the egg is that they are unequal: one daughter cell is very, very tiny and the other is huge. This selfish hoarding of all the cytoplasm by one daughter at the expense of the other means that the divisions will produce one large cell packed with goodies for development, while the others are stunted, useless things called polar bodies, that eventually just die. Another suggestion that has been made is that the location of the polar body correlates with the position of the first cleavage, and is therefore a marker of early polarity.

Now, though, a paper by Hiiragi and Solter is suggesting that neither of these cues make a significant contribution to polarity—the zygote really is a blank slate as far as axis specification. It's a very pretty paper, because much of what they did was good old-fashioned straightforward observation using time-lapse microscopy of the fertilized mouse embryo. They noted where sperm entered and where the polar body formed, set the cameras rolling, and later looked to see where the cleavage planes formed.

Relationship between cytoskeleton, chromosomes and first cleavage plane specification. a–f, Triple immunofluorescence staining of the embryo for microtubules (green), actin (red) and DNA (blue). Scale bar represents 20 mm. a, Late interphase. b, The end of interphase. Arrowheads in a and b denote the small foci of microtubules. c, Prophase. Paternal and maternal chromosome aggregates (large and small arrow, respectively) are surrounded by masses of microtubules (arrowheads). d, Metaphase. Metaphase plate with barrel-shaped spindle. e, Anaphase. Cleavage furrow is formed at the overlap of extended microtubules (arrowheads). f, Telophase. g, Time-lapse images of an embryo doubly recorded in DIC and fluorescence for DNA. Time is given in h:min after hCG injection. Scale bar represents 50 mm. h, Schematic view of the process, summarized from a–g. Green, red and blue bars represent microtubules, maternal and paternal chromosomes, respectively.

They saw no correlation. They also saw that the polar body tended to wander around the outside of the egg a bit and nestle down in the first cleavage furrow, which explains how prior work thought there might be a relationship between them.

The one good marker of the position of the future cleavage planes was the relationship of the pronuclei. When an egg is fertilized, the nuclei of the sperm and egg are initially separate and are called pronuclei—they then migrate towards each other and fuse. This movement is associated with an organization of the cytoskeleton, which is also going to be responsible for cytokinesis, or the physical division of the cell. The investigators tested the observed correlation by removing one pronucleus and replacing it with a transplanted one, placed in a different orientation. The result: the cleavage planes formed in a place predicted by the new orientation of the two pronuclei.

Mammalian embryos therefore start with a great deal more flexibility than insect embryos. They are going to puzzle out which end is up dynamically, as development proceeds. But of course, we also have to keep in mind that Drosophila development isn't exactly rigid, either, and they also work out the details of their organization dynamically.

Hiiragi T, Solter D (2004) First cleavage plane of the mouse egg is not predetermined but defined by the topology of the two apposing pronuclei. Nature 430:360-364.