Diffuse and specific tectopulvinar terminals in the tree shrew: synapses, synapsins, and synaptic potentials

Abstract

The pulvinar nucleus of the tree shrew receives both topographic (specific) and nontopographic (diffuse) projections from superior colliculus (SC), which form distinct synaptic arrangements. We characterized the physiological properties of these synapses and describe two distinct types of excitatory postsynaptic potentials (EPSPs) that correlate with structural properties of the specific and diffuse terminals. Synapses formed by specific terminals were found to be significantly longer than those formed by diffuse terminals. Stimulation of these two terminal types elicited two types of EPSPs that differed in their latency and threshold amplitudes. In addition, in response to repetitive stimulation (0.5-20 Hz) one type of EPSP displayed frequency-dependent depression whereas the amplitudes of the second type of EPSP were not changed by repetitive stimulation of up to 20 Hz. To relate these features to vesicle release, we compared the synapsin content of terminals in the pulvinar nucleus and the dorsal lateral geniculate (dLGN) by combining immunohistochemical staining for synapsin I or II with staining for the type 1 or type 2 vesicular glutamate transporters (markers for corticothalamic and tectothalamic/retinogeniculate terminals, respectively). We found that retinogeniculate terminals do not contain either synapsin I or synapsin II, corticothalamic terminals in the dLGN and pulvinar contain synapsin I, but not synapsin II, whereas tectopulvinar terminals contain both synapsin I and synapsin II. Finally, both types of EPSPs showed a graded increase in amplitude with increasing stimulation intensity, suggesting convergence; this was confirmed using a combination of anterograde tract tracing and immunocytochemistry. We suggest that the convergent synaptic arrangements, as well as the unique synapsin content of tectopulvinar terminals, allow them to relay a dynamic range of visual signals from the SC.

Figure 1. Convergence of tectopulvinar terminals.

Injections of biotinylated dextran amine (BDA) in the SC labels tectopulvinar axons that form widespread (“diffuse”) axons and boutons as well as more discrete clustered (“specific”) boutons. Representative images of BDA-labeled “diffuse” (A) and “specific” (B) axons in 50 µm thick sections are illustrated using transmitted light, a 40x objective, and Nomarski optics. C–G) Confocal images (single 0.1 µm scan with a 100x objective) illustrate tectopulvinar terminals labeled with BDA (purple) and tectopulvinar terminals labeled with antibodies against the type 2 vesicular glutamate transporter (vGLUT2, green). Terminals labeled with both BDA and vGLUT2 appear white. Clusters of vGLUT2-stained terminals contain at most 1 bouton contributed by “diffuse” axons (F), while “specific” axons contributed several boutons to each cluster (C-E, G) Scale in B = 10 µm and applies to A. Scale in D = 10 µm and applies to C, F and G. Scale in E = 2 µm.

Figure 2. “Specific” tectopulvinar synapses are longer than “diffuse” tectopulvinar synapses.

Electron micrographs illustrate examples of specific tectopulvinar terminals in the Pc (A) and diffuse tectopulvinar terminals in the Pd (B) labeled by the anterograde transport of biotinylated dextran amine (BDA) from the superior colliculus. The synapse length (arrows) of each terminal type was measured. The distribution of synapse lengths is plotted (C). As a population, the length of specific synapses was found to be significantly longer than the length of diffuse synapses (P<0.05). Scale bar  =  0.5 µm.

Figure 3. Distribution of synapsins and vGLUTs in the dLGN.

Tree shrew dLGN tissue was labeled with antibodies against synapsin I or II (purple), and vGLUT1 or 2 (green). Profiles that are labeled with two antibodies appear white. Representative confocal images (single 0.2 µm scan with a 100x objective) are illustrated. Corticogeniculate terminals (RS profiles labeled with the vGLUT1 antibody) contain synapsin I (A), but not synapsin II (C). Retinogeniculate terminals (RL profiles labeled with the vGLUT2 antibody) do not contain either synapsin I (B) or synapsin II (D). Scale  =  10 µm.

Figure 4. Distribution of synapsins and vGLUTs in the pulvinar nucleus.

Tree shrew pulvinar tissue was labeled with antibodies against synapsin I or II (purple), and vGLUT1 or 2 (green). Profiles that are labeled with two antibodies appear white. Representative confocal images of the Pd (single 0.2 µm scan with a 100x objective) are illustrated. Corticopulvinar terminals (RS profiles labeled with the vGLUT1 antibody) contain synapsin I (A), but not synapsin II (C). Tectopulvinar terminals (large clusters of RM profiles labeled with the vGLUT2 antibody) contain both synapsin I (B) and synapsin II (D). Scale  =  10 µm.

Figure 5. In vitro recording methods.

A) A parasaggital section of the tree shrew brain stained for the type 2 vesicular glutamate transporter (vGLUT2) illustrates the location of the whole cell recordings in the caudal pulvinar nucleus and the location of the 8 electrode stimulus array in the stratum griseum superficial (SGS) and stratum opticum (SO) of the superior colliculus. B, C) High magnification views of the Pc (B) and Pd (C) illustrated in panel A. Immunohistochemical staining for vGLUT2 is a marker for tectopulvinar terminals and reveals the distinct arrangements of tectal terminals in the dorsal (Pd) central (Pc) pulvinar nucleus. The Pd contains dense clusters of tectal terminals and tubular clusters line long lengths of dendrites (C). In contrast, the Pc contains smaller more sparsely distributed clusters of tectal terminals (B). D) Voltage fluctuations recorded in response to the injection of depolarizing or hyperpolarizing current steps of varying size revealed that all recorded cells fired with both tonic action potentials, and low threshold calcium bursts. E) Drawing of a biocytin-labeled cell recorded in the juvenile pulvinar, F) Drawing of a biocytin-labeled cell recorded in the adult pulvinar. Scale in A = 1 mm. Scale in C = 30 µm and also applies to B. Scales in E and F = 10 µm. dLGN, dorsal lateral geniculate nucleus.

Figure 6. Two types of EPSPs.

A) With increasing stimulation intensity, tecto-pulvinar EPSPs in the Pd show a graded increase in amplitude. B) Average first type EPSP amplitudes and latencies as a function of stimulation intensity (n = 17), graph show a graded increase in peak amplitude correlate to the increase in stimulation current but the latency of the EPSP is not relative to stimulation current. C) Average second type EPSP amplitudes and latencies as a function of stimulation intensity (n = 8), second type EPSP show a graded increase in peak amplitude and no change in latency with increasing stimulation intensity, but the threshold amplitude was smaller and latency was longer (p<0.05).

Figure 7. Two types of EPSPs display distinct frequency-dependent short-term plasticity.

A) Train stimulation consisted of 8 pulses at 1 Hz, 1.25 Hz, 2.5 Hz, 5 Hz and 10 Hz, 1st, 3rd and 5th EPSPs from the train stimulation of first type tecto-pulvinar fiber showed frequency-dependent depression, recorded at resting membrane potential of -58mV. C) 1st, 3rd and 5th EPSPs from the train stimulation of second type tecto-pulvinar fiber produced EPSPs with stable amplitudes, recorded at resting membrane potential of −59 mV. B and D) Normalized average EPSPs evoked in 14 cells by 8 pulses at 1 Hz, 1.25 Hz, 2.5 Hz, 5 Hz and 10 Hz, each point was normalized by dividing the EPSP amplitude of that pulse in the train (EPSPn) to the amplitude of the first EPSP of the train (EPSP1), B) stimulation of the first type tecto-pulvinar fiber showed change in EPSP peak amplitudes of the 8 EPSPs evoked by the stimulus train at various frequencies. D) stimulation of the second type tecto-pulvinar fiber showed stable EPSP peak amplitudes of the 8 EPSPs evoked by the stimulus train at various frequencies.

Figure 8. Two types of EPSPs display distinct short-term plasticity by paired-pulse stimulation.

A) 1st and 2nd EPSP evoked by paired pulse stimulation of first type tecto-pulvinar fiber at 200 ms time interval, traces were recorded at resting membrane potential at −58 mV (upper panel) or −70 mV under voltage-clamp mode (lower panel), both traces showed paired-pulse depression. B) 1st and 2nd EPSP evoked by paired pulse stimulation of second type tecto-pulvinar fiber at 200ms time interval, traces were recorded at resting membrane potential at -59mV (upper panel) or −70 mV under voltage-clamp mode (lower panel), both traces showed stable EPSP amplitudes. C) Plots of paired-pulse ratio (EPSP2 amplitude/EPSP1 amplitude) as a function of inter-stimulus intervals (0.1 s, 0.2 s, 0.3 s, 0.4 s, 0.6 s, 0.8 s, 1 s, 2 s) by stimulation of first type tecto-pulvinar fiber, showed gradual increase of paired-pulse ratio. D) Plots of paired-pulse ratio (EPSP2 amplitude/EPSP1 amplitude) as a function of inter-stimulus intervals (0.1 s, 0.2 s, 0.3 s, 0.4 s, 0.6 s, 0.8 s, 1 s, 2 s) by stimulation of second type tecto-pulvinar fiber, showed stable paired-pulse ratio.

Figure 9. “Mixed” EPSP show characteristics of two types of EPSPs.

A) “Mixed” EPSPs show small amplitude with small stimulation currents, followed by a large increase in amplitude and corresponding decrease in latency at larger stimulation currents. B) Plots of “mixed” EPSP amplitudes and latencies as a function of stimulation intensity, showed graded increase in amplitude with small stimulation currents followed by a large increase in amplitude and corresponding decrease in latency at larger stimulation currents.