7.3 Selected to die: studies of the CNS
Recent evidence has revealed that during development in mice, cell death occurs in two phases. The first phase is at about E15–E17 days, during neuron proliferation, and will not be considered further here. The second phase at about P0–P5 in the mouse (where P0 is the day of birth) is during the period of innervation.
The evidence for the second phase is well established. Some 50 per cent of retinal ganglion cells die during this phase and here's what happens. Dennis O'Leary (1987) used a long-lasting retrograde tracing dye to look at changes in axonal projections from the retina to the superior colliculus during development of the rat. Axons grow from the retina to the superior colliculus, so a retrograde tracing dye injected into the superior colliculus, will be taken up at any of the axon terminals present and transported along the axon to the cell body, located in the retina in this case. When O'Leary injected the dye into the caudal region of one superior colliculus at P12, and looked for dye-filled cells at P14, he found the pattern shown in Figure 20a. Most of the dye-filled cell bodies are shown in the nasal region of the contralateral retina. This result is as he expected and reveals the normal way in which the retina and superior colliculus are connected. Figure 20b shows the distribution of cells in the retina when the injection was done at P0 (the day of birth) and the retina examined two days later (P2). There are many more filled cells than at P14, and they are scattered over the retina outside the nasal region. This distribution is more widespread than when injection took place at P12. So at P0, when the injection took place, there are retinal ganglion cells scattered throughout the retina that had contacted the caudal superior colliculus, but by P12 these scattered cells have disappeared. Where have all the neurons, shown to be present and filled at P2, gone to by P12? What happens to the missing cells?
The advantage of using a long-lasting dye in these experiments is that the animals could be injected at P0 and the distribution of dye-filled cells examined at P12, as shown in Figure 20c.
What does the distribution of dye-filled cells shown in Figure 20c suggest has happened?
The distribution looks like that seen in animals injected at P12 and examined at P14 (Figure 20a). The small number of filled cells outside the nasal region suggests that most of the retinal ganglion cells that had contacted the inappropriate region of the superior colliculus have gone missing between birth and P12.
In fact the missing cells had died. They have not, for example, become concentrated in the nasal region of the retina. Thus the localised projection from the nasal region of the retina to the caudal region of the superior colliculus is due in part to the death of non-nasal retinal neurons which projected to the caudal colliculus.
Some 25–30 per cent of retinal ganglion cells fail to reach their correct targets in the brain and hence die. This leaves the death of a further 25 per cent of retinal ganglion cells to make up the known loss of 50 per cent of retinal ganglion cells. These other neurons innervate the correct targets, so selection among them for those that should die is based on a mechanism for matching the number of neurons to the target size, such as obtaining sufficient elixir as mentioned above.
Our common experience is that healthy organisms die only if they are damaged or injured. Cells certainly do die after damage or injury, e.g. after a stroke, in a process called necrosis. However, cells also use another form of death for which our common experience is not much use. This is programmed cell death, or apoptosis, in which the cell makes the proteins that destroy it. The process is called programmed cell death only because each cell has the genes, the cellular mechanism, necessary to cause the cell to die. The signals that start the apoptosis programme are the same kind of signals that set any gene transcription in motion. The difference between these two forms of cell death has been neatly summarised by Sanes ‘There is an important difference between a cell that dies gracefully by budding off neat little packages of membrane (apoptosis), compared to one that dies violently by retching catabolic [digestive] enzymes on its neighbours (necrosis).’ (Sanes et al., 2000).