The functional properties of synapses made by regenerated axons across spinal cord lesion sites in lamprey

Abstract

While the anatomical properties of regenerated axons across spinal cord lesion sites have been studied extensively, little is known of how the functional properties of regenerated synapses compared to those in unlesioned animals. This study aims to compare the properties of synapses made by regenerated axons with unlesioned axons using the lamprey, a model system for spinal injury research, in which functional locomotor recovery after spinal cord lesions is associated with axonal regeneration across the lesion site. Regenerated synapses below the lesion site did not differ from synapses from unlesioned axons with respect to the amplitude and duration of single excitatory postsynaptic potentials. They also showed the same activity-dependent depression over spike trains. However, regenerated synapses did differ from unlesioned synapses as the estimated number of synaptic vesicles was greater and there was evidence for increased postsynaptic quantal amplitude. For axons above the lesion site, the amplitude and duration of single synaptic inputs also did not differ significantly from unlesioned animals. However, in this case, there was evidence of a reduction in release probability and inputs facilitated rather than depressed over spike trains. Synaptic inputs from single regenerated axons below the lesion site thus do not increase in amplitude to compensate for the reduced number of descending axons after functional recovery. However, the postsynaptic input was maintained at the unlesioned level using different synaptic properties. Conversely, the facilitation from the same initial amplitude above the lesion site made the synaptic input over spike trains functionally stronger. This may help to increase propriospinal activity across the lesion site to compensate for the lesion-induced reduction in supraspinal inputs. The animal experiments were approved by the Animal Ethics Committee of Cambridge University.

Keywords: electrophysiology; lamprey; plasticity; regeneration; reticulospinal axon; spinal cord; spinal injury; synapse.

Figure 1

Basic synaptic properties in unlesioned and lesioned spinal cords above and below the lesion site in lamprey. The amplitude (A), paired-pulse (PP) ratio (B), and half-width (C) of single low frequency-evoked excitatory postsynaptic potential (EPSPs). The inset in (A) shows averaged (n = 10 traces) low frequency-evoked EPSPs in the different conditions. (D) The averaged activity-dependent plasticity of all synaptic inputs in unlesioned, above, and below lesion cords over 20 Hz spike trains and the recovery of the effect after the end of stimulation (stimulus numbers 21–24). (E) Comparison of connections that only showed facilitation over spike trains. (F) Comparison of connections that only showed depression over spike trains. (G) Traces showing the average of 10 traces from single connections over spike trains in an unlesioned spinal cord, and above and below the lesion site. Data are expressed as the mean ± SEM. *P < 0.05 (one-way analysis of variance followed by Tukey’s post hoc test).

Figure 2

The correlation between the paired-pulse (PP) or Train_11–20 plasticity and the initial excitatory postsynaptic potential (EPSP) amplitude. The correlation between the PP or Train_11–20 plasticity and the initial EPSP amplitude in unlesioned spinal cords using linear regression (PP ratio (r² = 0.02), Train_2–5 plasticity (r² = 0.03), Train_11–20 plasticity (r² = 0.08); n = 62, P > 0.05; A), above the lesion site (PP ratio (r² = 0.02) and Train_2–5 (r² = 0.03), and Train_11–20 plasticity (r² = 0.07), P > 0.05; B), and below the lesion site (PP ratio r² = 0.16, P > 0.05), there were significant negative relationships with the Train_2–5 (r² = 0.23; P < 0.05 data not shown) and Train_11–20 plasticity (r² = 0.29, P < 0.05; C). Linear regression was used.

Figure 3

Estimate of the number of vesicles (N_Ves) at connections in unlesioned spinal cords and connections below the lesion site in lamprey. (A) Graph showing the extrapolated exponential decay of the excitatory postsynaptic potential (EPSP) in unlesioned (n = 20 connections) and below lesion spinal cord (n = 10 connection), used to derive parameters to estimate N_Ves (see text for details). (B) Graph showing the N_Ves in unlesioned and below lesion connections. Data are expressed as the mean ± SEM. *P < 0.05 (independent samples t-test).

Figure 4

The analysis of the plasticity over the spike train (EPSP20/EPSP1) and the inverse of the coefficient of variation (CV^–2 ₂₀/CV^–2 ₁) for unlesioned (A), above lesion (B), and below lesion connections (C). Note that the x-axis shows the plasticity over the spike train and the y-axis the change in the coefficient of variation. A presynaptic change in plasticity is indicated by values falling either on or below (for depression) or above (for facilitation) the diagonal line, a postsynaptic change when values fall on the horizontal dashed line (i.e. no change in the CV^–2 ₂₀/CV^–2 ₁ despite a change in the EPSP), and both presynaptic and postsynaptic changes for depression and facilitation when values fall above and below the diagonal line, respectively (see Faber and Korn, 1991). Data are expressed as the mean ± SEM. CV^–2: Coefficient of variation; EPSP: excitatory postsynaptic potential.

Figure 5

Analysis of synaptic release properties. A variance-mean analysis of connections in unlesioned spinal cords, and above and below the lesion site (Ai, see text for details). (Aii) Sample traces in an unlesioned spinal cord in normal and low and high calcium Ringer. (Bi–Biii) The estimated number of release sites (N_min; Bi), release probability (Bii), and quantal amplitude (Biii) of unlesioned (n = 3) and below lesion (n = 3). Data are expressed as the mean ± SEM.

Figure 6

Differences in synaptic properties in animals that recovered well or poorly in lamprey. (A) The excitatory postsynaptic potential (EPSP) amplitude below the lesion site in good and poor recovery (n = 4). (B) The activity-dependent plasticity of connections above and below the lesion site in animals that recovered well or poorly. Data are expressed as the mean ± SEM.