Short term fellowships (1999-2003).
Final of Artur Luczak
Results:
In this work we try to modeled the development of dendritic geometry to investigate the mechanisms responsible for morphogenesis of neurons in central nervous system. The idea is to create plausible virtual neuron - environment interactions to reproduce different observed shapes of dendrites. In this way we can test which environmental factors are sufficient and which are not sufficient to explain the dendritic complexity and what information about shape is likely to be controlled by neuron itself. The model can generate granular cells, basal and apical dendrites of pyramidal neurons and planar oriented Purkinje cells.
Based on the model the following predictions about nature of dendritic growth can be made:
The increase of thickness of the cortical layers stretch the dendrites modifying spatial geometry of the dendritic tree. Dendritic branching can be stimulated by tension induced by strong environmental stretching. Neurotrophins are sufficient to explain distribution of segments lengths of dendritic tree but not its spatial geometry. Neuritis are growing in direction toward the higher concentration of neurotrophins, this introduce additional competition based interactions between dendrites shaping its geometry. The low concentration of neurotrophins can suppress branching and cause termination of dendrite. Shorter terminal segments in neurons with higher degree and centrifugal order indicate tendency to optimize total pathlength of dendrite. The arrangement of cell bodies in layers facilitate reproduction of the same type of neurons because in this way neurons are equally exposed to the environmental forces. This could be one of the reasons for lamination of the cortex.
Presented model simulates development of neurons and it is the first one which is able to produce different types of observed dendrites reproducing topology, lengths of segments and the spatial appearance. The model is implemented in the DendGrowth program which can generate numerous dendrites of the same type and store it in popular data format "swc". The program gives possibility to analyze and visualize reconstructed neurons.
Collaborative and/or training aspects:
The consultations with Jaap van Pelt have helped to evaluate my current results and to find the new directions to develop the project.
The participation in the Quantitative Neuroanatomy Tools Workshop was the excellent opportunity to learn details about the reconstruction of neurons structure and to have discussions about my project with the best specialists in this field.
Expected future impact of visit:
The visit has generated new ideas about modeling the dendritic development and geometry, hopefully this will be presented in the prepared publication.