Dendritic refinement of an identified neuron in the Drosophila CNS is regulated by neuronal activity and Wnt signaling
Ajeet Pratap Singh, K. Vijay Raghavan, and Veronica Rodrigues
April 15, 2010 Development 137, 1351-1360.
This is a companion paper and looks at dendritic development in particular. This paper clears-up some confusion of structure in the first paper. There are two CNS neurons, one for each side of the body. Just beyond the soma a dendrite arbor (1) attaches to the arista, a wind direction receptor. Then the main neurite goes into the higher level neuropil where there are arbors (2) and then turns to the other side where a huge arbor (3) occurs. It then goes to the antenna lobe on the other side where it arbors into the olfactory system (4). In some individuals another branch goes across the mid line to the other side where it meets itself (5). I see no clear information on where this arbor connects to. All but (1) are axon arbors, although there is evidence of other dendrites in (2, 3) but no mention of these in this paper.
The dendritic arbor for this neuron is interesting because it is extensive in the larvae, gets severely pruned beginning metamorphosis, grows a new tree which is then pruned back to a simple dendrite. I can find no information on what the larval tree connects to as the arista appears during metamorphosis, it is an adult structure. This structure gives a way to test hypotheses about dendritic plasticity because it is relatively simple and metamorphosis encompasses all the different ways a developing dendrite grows.
There are two main mechanisms involved here, a homeostatic plasticity that is regulated by genes and physiological state and a Hebbian-like plasticity which refines the system through a form of learning. Remarkably many developmental systems work this way. The larval dentrites are first pruned by glial cells  and then a period of growth of new dendrites begins. At about 30 hours apf, growth stops and the CNS connects with the arista sensory neuron and/or its interneurons and a refinement process begins which prunes back the new dendrites. In the arista itself, a similar growth process ends at 30 hrs apf and a period of refinement by cell apoptosis begins . The authors show that CNS dendritic refinement is mediated by the neuron and its synapses and thus a Hebbian process. Hebbian learning has a rich mathematical theory with it and it might be interesting to look at the data with this in mind. Counts of puncta would replace weights in this case.
There is much more to this and to the other paper as the authors try to parse out the chemical complexities of this process. There are two things going on here which are mediated by a third. First there is the genic processes that end the life of the larval dendrites and initiate and stop growth. This is mediated by the physiological state, the environmental stresses affecting the health and fitness of the larvae. Second there is a learning period which fine tunes and tests the system, again mediated by the physiological state.
- Cellular mechanisms in the development of the Drosophila arista
Biao He and Paul N. Adler
Mechanisms of Development Volume 104, Issues 1-2
- Synaptic Pruning by Microglia Is Necessary for Normal Brain Development
Paolicelli, RC et al (2011).
Science DOI: 10.1126/science.1202529