Effect of localized innervation of the dendritic trees of feline motoneurons on the amplification of synaptic input: a computational study

Abstract

Previous studies show that the activation of voltage-dependent channels is dependent on the local density of synapses in the dendritic region containing voltage-dependent channels. We hypothesized that the selective innervation of excitatory vestibulospinal (VST) neurons on the medial dendrites of contralateral splenius motoneurons is designed to enhance the activation of persistent inward currents (PICs) mediated by dendritic L-type Ca(2+) channels. Using compartmental models of splenius motoneurons we compared the synaptic current reaching the soma in response to excitatory input generated by synapses with two different distribution patterns. The medial distribution was based on the arrangement of VST synapses on the dendrites of contralateral splenius motoneurons and the uniform distribution was based on an arrangement of synapses with no particular bias to any region of the dendritic tree. The number of synapses in each distribution was designed to match estimates of the number of VST synapses activated by head movements. In the absence of PICs, the current delivered by the synapses in the uniform distribution was slightly greater. However, the maximal currents were small, < or = 4.1 nA, regardless of the distribution of synapses. In models equipped with L-type Ca(2+) channels, PIC activation was largely determined by the local density of synapses in proximity to the L-type Ca(2+) channels. In 3 of 5 cells, this led to a 2- to 4-fold increase in the current generated by synapses in the medial distribution compared to the uniform distribution. In the other two cells, the amplification bias was in favour of the medial distribution but was either small or restricted to a narrow range of frequencies. These simulations suggest that the innervation pattern of VST axons on contralateral splenius motoneurons is arranged to strengthen an otherwise weak synaptic input by increasing the likelihood of activating PICs. Additional simulations suggest that this prediction can be tested using common experimental protocols.

Figure 1. The number and location of synapses belonging to the medial and uniform distributions

A, schematic illustration of the vestibular nucleus. Shaded area indicates the region of the vestibular nucleus containing the majority of VST neurons innervating contralateral splenius motoneurons. B, estimated total number of synapses (Total N_synapse) as a function of the total surface area of the dendritic tree. C, location of 354 synapses (black dots) on DVS 25-3 arranged according to the medial distribution. D, location of 354 synapses on the same cell shown in C arranged according to the uniform distribution. E, cumulative frequency of the medial to lateral location of synapses on cell DVS 25-3. The distributions of three sets of synapses are compared. The thick black line indicates the frequency of all excitatory synapses (n = 14 396) (equivalent to the total surface area of the dendritic tree); the thin black line is the frequency of 120 VST contacts based on the data of Grande et al. (2005); the grey line indicates the frequency of synapses (n = 354) based on the medial distribution (based on the data shown in C). Note that the cumulative frequencies of the medial and VST distributions are nearly identical. F, cumulative frequencies of the medial to lateral location of all excitatory synapses (n = 14 396) on cell DVS 25-3 (thick black line) and 354 synapses arranged according to the uniform distribution (grey line, based on the data shown in D).

Figure 2. Current reaching the soma under passive conditions

A, comparison of current reaching the soma in cell DVS 25-3, during the activation of synapses belonging to the medial (black lines) and uniform (grey lines) distributions when the driving potential was constant (dashed lines) and when the driving potential was determined by the local membrane potential (continuous lines). B, amount of current lost due to reductions in driving potential caused by the activity of neighbouring synapses. Values were obtained by taking the difference between the continuous and dashed lines in A. C, average amount of current lost due to nonlinear summation in all five cells. Asterisks indicate significance of P < 0.02, Kruskal–Wallis test.

Figure 3. Current reaching the soma under active conditions

A, comparison of current reaching the soma in all five cells during the activation of synapses belonging to the medial (black lines) and uniform (grey lines) distributions. Background synaptic activity was generated by activating 8% of the maximum number of excitatory synapses and 10% of the maximum number of inhibitory synapses. B, comparison of current reaching the soma in all five as in A, but background excitation was increased in 1% increments such that the synapses belonging to the medial or uniform distribution activated PICs between 40 and 60 Hz. Amount of background excitation and inhibition for each cell is indicated in the top-right corner. C, amplification of synaptic current reaching the soma by PIC activation. Amplification was defined as the ratio of the amount of current reaching the soma under active conditions to the amount of current reaching the soma under passive conditions.

Figure 4. Comparison of the density of synapses belonging to the medial and uniform distribution in proximity to the L-type Ca²⁺ channel hotspots that were activated

A, location of active hotspots for cell DVS 25-3 when synapses from medial and uniform distributions were activated at 80 Hz. B, schematic illustration of the region used to calculate synaptic density. C, differences in synaptic density between the synapses belonging to the medial and uniform distributions in the vicinity of the active hotspots shown in A. Upward deflections indicate the synaptic density was greater for synapses in the medial distribution. D, location of active hotspots for cell DVS 14-1 when synapses from medial and uniform distributions were activated at 120 Hz. E, differences in synaptic density between the synapses belonging to the medial and uniform distributions in the vicinity of the active hotspots shown in B.

Figure 5. Comparison of the current reaching the soma under active conditions during activation of different sets of synapses selected from the medial and uniform distributions

A, DVS 25-3. B, DVS 14-1. Bottom graphs indicate the average current reaching the soma (based on the results shown here and in Fig. 3B. Asterisks indicate P < 0.05, Kruskal–Wallis test. Background synaptic activity was generated by activating 8% of the maximum number of excitatory synapses and 10% of the maximum number of inhibitory synapses.

Figure 6. Current reaching the soma under active conditions following the activation of 200 or 500 synapses belonging to the medial (black lines) and uniform (grey lines) distributions in cell DVS 22-3

Background synaptic activity was generated by activating 8% of the maximum number of excitatory synapses and 10% of the maximum number of inhibitory synapses.

Figure 7. Comparison of current–voltage relationships at the onset of PICs during activation of synapses belonging to the medial (black lines) and uniform (grey lines) distributions

A, synapses from medial and uniform distributions activated at 50 Hz with background synaptic activity generated by activating 8% of the maximum number of excitatory synapses and 10% of the maximum number of inhibitory synapses. Arrow for cell DVS 25-2 indicates peak difference. B, same condition as in A, but synapses from the medial and uniform distributions were activated at 100 Hz. C, synapses from the medial and uniform distributions were activated at 100 Hz but background excitation was adjusted to correspond to that used in the simulations shown in Fig. 3B.

Figure 8

Peak differences in current reaching the soma following the activation of synapses in the medial and uniform distributions for the three conditions described in Fig. 7