How to make a fly or a person or a fish
Back in the last century when the molecular revolution took off, a lot of researchers decided to use banana fruit flies to study the genetics and “how-to” of development. Most people were obviously more interested in how that process is brought about in humans, but scientists had to start somewhere. Flies are easier for both ethical and practical reasons — flies are cheap to rear, have short life cycles (3 weeks or so), the eggs don’t scream because they have no nervous system (yet), and the adults are assiduous egg layers. One can get lots of eggs in a half hour timespan.
Development from embryo to larva begins when an egg is sub-divided into a bunch of cells. For a fly, the first obvious signs of an embryonic form is a one cell layer lining the inner eggshell and yolk in the center. Once there are many cells, they move among each other, change shape, divide, or adhere to other cells or “decide” to stay away from others. In short, the cells talk, form allegiances, negotiate back and forth, and sometimes are on opposing teams. Those cells who will contribute to a specific structure need to organize themselves and avoid interference from those cells that are going to have a different fate. For instance, heads do their job better without a foot being in the way of one of the eyes (one mutant gene — Antennapedia —respecifies a head antennae into a foot on the fly head, but luckily it’s a small foot). Each time groups of cells change their positions and their contacts, this causes the embryo to change its architecture — thus sets of cells lose influence with some cells and gain (or regain) influence with others. Cells encounter a succession of contacts that gradually determine what function and structure they contribute to in a larva. An embryo engages in a number of spectacular shape changes before it emerges as a larva. After three larval instars and one pupal stage, the insect finally matures to a winged adult.
Back to the single egg, laid by a busy fly. One the first orders of the day is to establish a body axis, front and back, head and back end, and eventually appendages at all the right spots. Before the egg is laid, the mother helps by pumping axis determinants into the egg interior or sticking axis identifiers on the membrane in certain regions of the egg. The actual how-to of “locking-in”, say, a posterior versus a head region happens by sophisticated molecular strategies. Timewise, just a little nudge establishes the equivalent of north, south, east and west of the egg interior — anterior, posterior, dorsal and ventral, even before the egg interior is partitioned into a bunch of cells. Neat experiments have tested whether one can get a young embryo to change its mind, and make two posteriors or two heads or a convoluted mess of middle features. It turns out that signals are needed to hold on to a particular way of life, or fate. Being told once is not enough. Mostly. This is nature; exceptions occur.
Development is a logical progression, moving from simplicity to complexity. If one is to look at the anterior-posterior axis or the dorsal-ventral one, what would be the next step to more complexity (or regional specificity, to be fancy about it)? Splitting up the body axis into sub-sectors comes to mind. Instead of two sectors per axis (posterior-anterior; dorsal-ventral), the genetic machinery of the embryo creates four sub-divisions, broadly speaking — I’ll use the anterior-posterior (AP) axis as an example: very anterior, anterior middle, posterior middle, and very posterior — plus a very-most posterior spot in flies, where they sequester their reproductive germ cells, to make the next generation.
To sub-divide the AP axis, gene products are rolled out in response to the anteriorness (axis determinants) or posteriorness of the local environment. These products come from gap genes, that are so named because a loss of any one of them makes a huge gap in the body organization later on. Maybe there is no head, or part of the trunk and arms are missing, or the embryo “decides” to duplicate parts, creating a mirror image of its midbody, or its back end or has only two heads and no body.
Gap gene products establish territories. What keeps their territorial stakes intact? Is one gap gene product not able to invade the territory of another? Gap genes products do compete with each other. Essentially, gap gene products turn off each other’s gap genes within any nuclei that these products reach — they literally sit down on their competitor’s gene and obstruct their use (nuclei hold the genetic material that has the knowledge to build a body). So if one gap gene product is missing in a bunch of nuclei, another gap gene product invades that territory. The outcome is that an alternate fate awaits that body region.
So far, the AP axis has been reframed into four sectors, more or less. That is not enough to specify a complicated body pattern of an insect larva. Multiple sectors are necessary; each separate enough that they can develop independent identities, yet still contribute to the order exigent to the whole body (consider the digestive system: pharynx ? stomach? small intestine ? large intestine — these organs better be put down in order or the recipient will be a very unhappy eater and very soon dead). What is needed is a series of equally sized segments for the sake of independent development, but where each segment “remembers” its relative position donated by the axis determinants and the gap gene products.
Does the insect embryo impose (almost) equally-sized, but separate segments upon the length of the body? Yes. The head also has segments, but it is a complicated place, as all heads are. The strategy of segmentation is also found in mammals, fish, birds, and reptiles. To run one’s finger up or down one’s spine is to recall this early, but persistent embryonic organizational plan.
Products that help partition the insect body into multiple, equally-sized sectors come from segmentation genes. Gap gene products turn on and encourage the expression of segmentation genes within their territories. Not all gap genes control similarly sized territories. There could be a problem of equal spacing for the distribution of segmentation products across the gap gene territories and boundaries. However, if a segmentation gene attracts each gap gene product under certain circumstances — where a gap gene product is at high levels, because there is peace and quiet and no competition (four places) and in boundary regions (three places) where gap gene products must coexist — and where they compete for the favor of encouraging segmentation gene products — that makes seven regions available simultaneously.
Suddenly the insect embryo can be divided into 14 sectors, because a segmentation gene product has an on-off distribution — high for a band of four cells where the gap gene product is high, nothing for another four cell band, because the concentration is not right and something interferes, and high again at the boundary region, where the environment is permissive and cooperative for the making of a segmentation gene product (on-off, on-off, seven times makes fourteen).
These early segmentation gene products encourage other segmentation gene products to be made, relative to their status within the cells. They are not beyond repressing each other (it’s called refinement). Pretty soon there are 26 sectors, in addition to the head, where things are arranged a bit differently. Basic insect lore tells us that there are three segments in the thorax and ten for the abdomen (13 segments). Each of the 26 sectors represent half a segment, one anterior half next to one posterior.
The AP axis is now represented on a much smaller scale. Thanks to the specific bouquet of segmentation genes expressed in any one cell, products are manufactured that basically specialize in “me-ness” — stronger identities both with regard to position and what that cell is willing to construct in the future. Boundaries become nice and tight. Anteriorness cannot not blend across boundaries from one segment to the next, because gene products that encourage posterior structures are in the way.
As the embryo grows, cells within these segments gain more and more specific instructions, one might say developmental knowledge, both from the genes expressed and exported from their own nuclei, and from signals from other cells with regard to position inside the sector and outside. Their “remembrances” become more complex in texture, and co-ordination more and more specialized and picky. The cells develop stronger and stronger allegiances; neurons don’t just wander anywhere they please, nor do digestive cells or muscles. The cells also have fewer and fewer options to become different. With maturity comes a certain rigidity of attitude, a fixity in their fate. The cells are not just determined; they are also committed. Developmental biologists test the degree of determination: how willing are cells to embrace another fate (think stem-cell debate) and how committed are entire regions to, say, growing just one arm or can one create a multiple limb frog? What does it take for an insect to have legs on every segment and look like a centipede? (check the Web)
So organizational hierarchies eventually produces a fly — possibly one of those buzzing around the dinner table. Do we care? Yes, because the establishment of the body plan of flies and people have a lot in common. The molecular players that set the body axis, the main rules for segmentation, the formation of organs and organ systems are the same — even muscles, blood, heart, gut, aspects of the lungs and kidneys that look so very different in arthropods and mammals. The same regulatory genes, gene products, the same control patterns for cell division and shape changes, the same signals and communication between cells. The wonder is that we don’t sprout wings and have a pupal stage. And that is another story for another day.