Figure 5.

Quantification of VGLUT1-IR contact proximo-distal distribution in control and regenerated MNs. A, Absolute number of VGLUT1-IR contacts found at different dendrite path distances from dendrite origins. Each cell is plotted separately. Most VGLUT1-IR synapses on control MNs occur adjacent to the cell body and then decrease in number with increased distance to the cell body. In contrast, similar numbers of VGLUT1-IR synapses are found at different distances on the dendrites of regenerated MNs, up to ∼1000 μm distance from dendrite origins. In both control and regenerated MNs relatively few VGLUT1-IR synapses are found at distances beyond 1000 μm. B, Average cumulative distributions of VGLUT1-IR synapses along the dendritic arbor of control and regenerated MNs. Up to 1000 μm in the dendritic arbor (vertical line) the distribution of synapses in control MNs is best fitted by a single exponential function (R² = 0.99), suggesting a progressive decay in number of synapses with distance. In contrast, the distribution of synapses in regenerated MNs was well fitted by a linear polynomial function (R² = 0.97) with no fit improvement using exponential functions. This suggests a rather uniform distribution. C, Distribution of dendritic lengths at different distances from the cell body is best described by Gaussian curves with peaks between 500 and 700 μm path distance, where the maximum number of branches are concentrated. Curves describing regenerated MNs are always slightly below that of controls at all distances, but overall this decrease was not significant (Table 2). Insets show the cumulative distribution of total dendritic length in control versus regenerated MNs. Both distributions closely overlap indicating that the relative proportions of dendritic segments at different distances from the cell body did not vary. D, Distributions of dendrite surfaces is shifted to the left compared with dendrite length because proximal dendrites are thicker and display larger surfaces. In contrast to the distribution of dendritic length, surface curves in experimental MNs were frequently above those from controls. This was due to an increase in dendrite thickness, particularly in the more proximal regions, that was reflected in very significant increases in total dendrite volume (Table 2). However, the overall distribution of dendrite surface or their cumulative percentage distribution (inset) did not reveal differences between control and injured MNs. E, Linear density of VGLUT1-IR contacts (i.e., number of VGLUT1-IR contacts per 100 μm of dendrite length) in the three control and three regenerated MNs. In control MNs VGLUT1-IR contact density is very high adjacent to the cell body (x = 0) and then declines abruptly, becoming lower and more uniform after ∼400 μm path-distance from dendrite origins. In regenerated MNs, a lower more uniform density is registered throughout the dendritic arbor. F, Surface densities (i.e., number of VGLUT1-IR contacts per 100 μm² of dendrite area) reveal a similar pattern. In this case, the decrease in density with distance on control MNs is partially offset by dendritic tapering (which reduces available surface and thus partially compensate for the decrease in synapse number per unit length of dendrite). In regenerated MNs, the density is low and uniform throughout the dendrite arbor. Beyond 1000 μm the estimates of VGLUT1-IR contact surface densities became highly variable because the number of dendritic segments in the gray matter is reduced and the extreme tapering provokes large variations in calculated densities from just a few synapses. Inset, The average of control and experimental MN densities normalized against a uniform density calculated from the total number of synapses and dendritic lengths in each cell. Although in control MNs densities greatly differ from the uniform average density (line at y = 1), in regenerated MNs densities fluctuate around the uniform distribution throughout the total length of the dendritic arbor.