Synapsin utilization differs among functional classes of synapses on thalamocortical cells

Abstract

Several proteins in nerve terminals participate in synaptic transmission between neurons. The synapsins, which are synaptic vesicle-associated proteins, have widespread distribution in the brain and are assumed essential for sustained recruitment of vesicles during high rates of synaptic transmission. We compared the role of synapsins in two types of glutamatergic synapses on thalamocortical cells in the dorsal lateral geniculate nucleus of mice: retinogeniculate synapses, which transmit primary afferent input at high frequencies and show synaptic depression, and corticogeniculate synapses, which provide modulatory feedback at lower frequencies and show synaptic facilitation. We used electrophysiological methods to determine effects of gene knock-out of synapsin I and II on short-term synaptic plasticity in paired-pulse, pulse-train, and posttetanic potentiation paradigms. The gene inactivation changed the plasticity properties in corticogeniculate, but not in retinogeniculate, synapses. Immunostaining with antibodies against synapsins in wild-type mice demonstrated that neither synapsin I nor II occurred in retinogeniculate terminals, whereas both occurred in corticogeniculate terminals. In GABAergic terminals, only synapsin I occurred. In corticogeniculate terminals of knock-out mice, the density of synaptic vesicles was reduced because of increased terminal size rather than reduced number of vesicles and the intervesicle distance was increased compared with wild-type mice. In the retinogeniculate terminals, no significant morphometric differences occurred between knock-out and wild-type mice. Together, this indicates that synapsin I and II are not present in the retinogeniculate terminals and therefore are not essential for sustained, high-rate synaptic transmission.

Figure 1.

Synapsin double knock-out mice and wild-type mice had similar characteristics of short-term depression in retinogeniculate synapses. A, Responses evoked by paired-pulse stimulation of retinal afferents in wild-type (WT) and double knock-out (KO) mice. Summated data from all recordings are shown. Error bars indicate SEM. The inset shows examples of five superimposed traces with different interstimulus intervals. Top traces, Wild-type mouse; bottom traces, synapsin I and II double knock-out mouse. B, Responses to pulse-train stimulation with 300 pulses delivered at 10 Hz. Summated data from all recordings. The inset shows examples of responses to the first five pulses in the train from a cell of a wild-type mouse (top trace) and a synapsin I and II double knock-out mouse (bottom trace). Calibration: 100 ms, 500 pA.

Figure 2.

Synapsin double knock-out (KO) mice and wild-type (WT) mice had different characteristics of short-term facilitation in corticogeniculate synapses. A, Responses to paired-pulse stimulation of cortical afferents. Summated data from all recordings are shown. Error bars indicate SEM. The inset shows examples of five superimposed traces with different interstimulus intervals. Top traces, Wild-type mouse; bottom traces, synapsin I and II double knock-out mouse. B, Responses to pulse-train stimulation with 300 pulses at 10 Hz. Sum of data from all recordings. The inset shows the first three EPSCs, the 50th EPSC, and the last EPSC of single traces from a wild-type mouse (top trace) and a synapsin I and II double knock-out mouse (bottom trace). Calibration: 100 ms, 50 pA.

Figure 3.

Posttetanic potentiation occurred at corticogeniculate synapses but was less pronounced in double knock-out (KO) mice compared with wild-type (WT) mice. Responses evoked by test pulses delivered at 10 s intervals before and after application of a tetanic stimulus (100 pulses at 100 Hz). Summated data from all recordings are shown. Error bars indicate SEM. The inset shows examples of single EPSCs (before, 10, 30, 90, and 150 s after tetanization) from a cell of a wild-type mouse (top trace) and a synapsin I and II double knock-out mouse (bottom trace). Calibration: 500 ms, 500 pA.

Figure 4.

Synaptic terminals in LGN were differentially labeled by antibodies against synapsin I and synapsin II. A, Synapsin I antibody labeling of an RS terminal forming an asymmetric synapse (arrow). B, Synapsin I antibody labeling of an F terminal forming two symmetric synapses (arrows). The F terminal contains dark mitochondria (m). C, Synapsin II antibody labeling of three RS terminals, forming asymmetric synapses (arrows). Notice the unlabeled RL terminal, containing pale mitochondria (m), in the same field. This large terminal forms synapses (arrowheads) onto a dendrite. Scale bars, 500 nm. D, The proportions of the RL, RS, and F terminals among three populations of synapses: Total, all synaptic terminals found within the same regions that were examined for the presence of labeled terminals; Syn I, all terminals labeled by antibodies against synapsin I; Syn II, all terminals labeled by antibodies against synapsin II.

Figure 5.

Synapsin I and II gene inactivation reduced the density of synaptic vesicles in terminals of corticothalamic afferents but had no effect on the density in terminals of retinothalamic afferents. A, RL terminal from a double knock-out (KO) mouse, containing pale mitochondria (m), with three synapses (arrowheads) onto geniculate dendrites. B, Two RS terminals from a double knock-out mouse. C, RS terminal from a wild-type (WT) mouse. Arrows in B and C point to synapses from the postsynaptic side. Scale bar, 1 μm. D, Average density of vesicles in RL and RS terminals from wild-type (WT) and synapsin I and II knock-out (KO) mice. The error bars indicate SEM.

Figure 6.

Terminal area and intervesicle distance at corticothalamic terminals were increased in the knock-out mice. A, Frequency distribution of RS terminal area in synapsin I and II knock-out (KO) compared with wild-type (WT) mice. Increased area, rather than reduction in number of vesicles, could explain the decreased vesicle density in the knock-out mice as illustrated schematically in the inset. In each pair of bars, the black one is for wild type, and the gray one is for knock-out. Bin width, 0.065 μm². B, Average intervesicle distances distance in RL and RS terminals in wild-type and knock-out mice. Deletion of synapsins led to increased distance between synaptic vesicles as illustrated in the inset. The error bars are SEM.