Synaptic properties of mouse tecto-parabigeminal pathways.

The superior colliculus (SC) is a critical hub for the generation of visually-evoked orienting and defensive behaviors. Among the SC's myriad downstream targets is the parabigeminal nucleus (PBG), the mammalian homolog of the nucleus isthmi, which has been implicated in motion processing and the production of defensive behaviors. The inputs to the PBG are thought to arise exclusively from the SC but little is known regarding the precise synaptic relationships linking the SC to the PBG. In the current study, we use optogenetics as well as viral tracing and electron microscopy in mice to better characterize the anatomical and functional properties of the SC-PBG circuit, as well as the morphological and ultrastructural characteristics of neurons residing in the PBG. We characterized GABAergic SC-PBG projections (that do not contain parvalbumin) and glutamatergic SC-PBG projections (which include neurons that contain parvalbumin). These two terminal populations were found to converge on different morphological populations of PBG neurons and elicit opposing postsynaptic effects. Additionally, we identified a population of non-tectal GABAergic terminals in the PBG that partially arise from neurons in the surrounding tegmentum, as well as several organizing principles that divide the nucleus into anatomically distinct regions and preserve a coarse retinotopy inherited from its SC-derived inputs. These studies provide an essential first step toward understanding how PBG circuits contribute to the initiation of behavior in response to visual signals.

The parabigeminal nucleus is a source of "retinogeniculate replacement terminals" in mice that lack retinofugal input.

In the dorsal lateral geniculate nucleus (LGN) of mice that lack retinal input, a population of large terminals supplants the synaptic arrangements normally made by the missing retinogeniculate terminals. To identify potential sources of these "retinogeniculate replacement terminals," we used mutant mice (math5-/- ) which lack retinofugal projections due to the failure of retinal ganglion cells to develop. In this line, we labeled LGN terminals that originate from the primary visual cortex (V1) or the parabigeminal nucleus (PBG), and compared their ultrastructure to retinogeniculate, V1 or PBG terminals in the dLGN of C57Blk6 (WT) mice (schematically depicted above graph). Corticogeniculate terminals labeled in WT and math5-/- mice were similar in size and both groups were significantly smaller than WT retinogeniculate terminals. In contrast, the PBG projection in math5-/- mice was extensive and there was considerable overlap in the sizes of retinogeniculate terminals in WT mice and PBG terminals in math5-/- mice (summarized in histogram). The data indicate that V1 is not a source of "retinogeniculate replacement terminals" and suggests that large PBG terminals expand their innervation territory to replace retinogeniculate terminals in their absence.

Ultrastructure of ipsilateral and contralateral tectopulvinar projections in the mouse.

Visual pathways of the brain are organized into parallel channels that code different features of the external environment. In the current study, we investigated the anatomical organization of parallel pathways from the superior colliculus (SC) to the pulvinar nucleus in the mouse. Virus injections placed in the ipsilateral and contralateral SC to induce the expression of different fluorescent proteins define two pulvinar zones. The lateral pulvinar (Pl) receives ipsilateral SC input and the caudal medial pulvinar (Pcm) receives bilateral SC input. To examine the ultrastructure of these projections using transmission electron microscopy, we injected the SC with viruses to induce peroxidase expression within synaptic vesicles or mitochondria. We quantitatively compared the sizes of ipsilateral and contralateral tectopulvinar terminals and their postsynaptic dendrites, as well as the sizes of the overall population of synaptic terminals and their postsynaptic dendrites in the Pl and Pcm. Our ultrastructural analysis revealed that ipsilateral tectopulvinar terminals are significantly larger than contralateral tectopulvinar terminals. In particular, the ipsilateral tectopulvinar projection includes a subset of large terminals (≥ 1 µm2 ) that envelop dendritic protrusions of postsynaptic dendrites. We also found that both ipsilateral and contralateral tectopulvinar terminals are significantly larger than the overall population of synaptic terminals in both the Pl and Pcm. Thus, the ipsilateral tectopulvinar projection is structurally distinct from the bilateral tectopulvinar pathway, but both tectopulvinar channels may be considered the primary or "driving" input to the Pl and Pcm.

GABAergic cell types in the superficial layers of the mouse superior colliculus.

To begin to unravel the complexities of GABAergic circuits in the superior colliculus (SC), we utilized mouse lines that express green fluorescent protein (GFP) in cells that contain the 67 kDa isoform of glutamic acid decarboxylase (GAD67-GFP), or Cre-recombinase in cells that contain glutamic acid decarboxylase (GAD; GAD2-cre). We used Cre-dependent virus injections in GAD2-Cre mice and tracer injections in GAD67-GFP mice, as well as immunocytochemical staining for gamma amino butyric acid (GABA) and parvalbumin (PV) to characterize GABAergic cells that project to the pretectum (PT), ventral lateral geniculate nucleus (vLGN) or parabigeminal nucleus (PBG), and interneurons in the stratum griseum superficiale (SGS) that do not project outside the SC. We found that approximately 30% of SGS neurons in the mouse are GABAergic. Of these GABAergic neurons, we identified three categories of potential interneurons in the GAD67-GFP line (GABA+GFP ~45%, GABA+GFP + PV ~15%, and GABA+PV ~10%). GABAergic cells that did not contain GFP or PV were identified as potential projection neurons (GABA only ~30%). We found that GABAergic neurons that project to the PBG are primarily located in the SGS and exhibit narrow field vertical, stellate, and horizontal dendritic morphologies, while GABAergic neurons that project to the PT and vLGN are primarily located in layers ventral to the SGS. In addition, we examined GABA and GAD67-containing elements of the mouse SGS using electron microscopy to further delineate the relationship between GABAergic circuits and retinotectal input. Approximately 30% of retinotectal synaptic targets are the presynaptic dendrites of GABAergic interneurons, and GAD67-GFP interneurons are a source of these presynaptic dendrites.

Synaptic organization of striate cortex projections in the tree shrew: A comparison of the claustrum and dorsal thalamus.

The tree shrew (Tupaia belangeri) striate cortex is reciprocally connected with the dorsal lateral geniculate nucleus (dLGN), the ventral pulvinar nucleus (Pv), and the claustrum. In the Pv or the dLGN, striate cortex projections are thought to either strongly "drive", or more subtly "modulate" activity patterns respectively. To provide clues to the function of the claustrum, we compare the synaptic arrangements of striate cortex projections to the dLGN, Pv, and claustrum, using anterograde tracing and electron microscopy. Tissue was additionally stained with antibodies against γ-aminobutyric acid (GABA) to identify GABAergic interneurons and non-GABAergic projection cells. The striate cortex terminals were largest in the Pv (0.94 ± 0.08 µm2 ), intermediate in the claustrum (0.34 ± 0.02 µm2 ), and smallest in the dLGN (0.24 ± 0.01 µm2 ). Contacts on interneurons were most common in the Pv (39%), intermediate in the claustrum (15%), and least common in the dLGN (12%). In the claustrum, non-GABAergic terminals (0.34 ± 0.01 µm2 ) and striate cortex terminals were not significantly different in size. The largest terminals in the claustrum were GABAergic (0.51 ± 0.02 µm2 ), and these terminals contacted dendrites and somata that were significantly larger (1.90 ± 0.30 µm2 ) than those contacted by cortex or non-GABAergic terminals (0.28 ± 0.02 µm2 and 0.25 ± 0.02 µm2 , respectively). Our results indicate that the synaptic organization of the claustrum does not correspond to a driver/modulator framework. Instead, the circuitry of the claustrum suggests an integration of convergent cortical inputs, gated by GABAergic circuits. J. Comp. Neurol. 525:1403-1420, 2017. © 2016 Wiley Periodicals, Inc.

Ultrastructure of geniculocortical synaptic connections in the tree shrew striate cortex.

To determine whether thalamocortical synaptic circuits differ across cortical areas, we examined the ultrastructure of geniculocortical terminals in the tree shrew striate cortex to compare directly the characteristics of these terminals with those of pulvinocortical terminals (examined previously in the temporal cortex of the same species; Chomsung et al. [] Cereb Cortex 20:997-1011). Tree shrews are considered to represent a prototype of early prosimian primates but are unique in that sublaminae of striate cortex layer IV respond preferentially to light onset (IVa) or offset (IVb). We examined geniculocortical inputs to these two sublayers labeled by tracer or virus injections or an antibody against the type 2 vesicular glutamate antibody (vGLUT2). We found that layer IV geniculocortical terminals, as well as their postsynaptic targets, were significantly larger than pulvinocortical terminals and their postsynaptic targets. In addition, we found that 9-10% of geniculocortical terminals in each sublamina contacted GABAergic interneurons, whereas pulvinocortical terminals were not found to contact any interneurons. Moreover, we found that the majority of geniculocortical terminals in both IVa and IVb contained dendritic protrusions, whereas pulvinocortical terminals do not contain these structures. Finally, we found that synaptopodin, a protein uniquely associated with the spine apparatus, and telencephalin (TLCN, or intercellular adhesion molecule type 5), a protein associated with maturation of dendritic spines, are largely excluded from geniculocortical recipient layers of the striate cortex. Together our results suggest major differences in the synaptic organization of thalamocortical pathways in striate and extrastriate areas.

Synaptic organization of thalamocortical axon collaterals in the perigeniculate nucleus and dorsal lateral geniculate nucleus.

We examined the synaptic targets of large non-gamma-aminobutyric acid (GABA)-ergic profiles that contain round vesicles and dark mitochondria (RLD profiles) in the perigeniculate nucleus (PGN) and the dorsal lateral geniculate nucleus (dLGN). RLD profiles can provisionally be identified as the collaterals of thalamocortical axons, because their ultrastrucure is distinct from all other previously described dLGN inputs. We also found that RLD profiles are larger than cholinergic terminals and contain the type 2 vesicular glutamate transporter. RLD profiles are distributed throughout the PGN and are concentrated within the interlaminar zones (IZs) of the dLGN, regions distinguished by dense binding of Wisteria floribunda agglutinin (WFA). To determine the synaptic targets of thalamocortical axon collaterals, we examined RLD profiles in the PGN and dLGN in tissue stained for GABA. For the PGN, we found that all RLD profiles make synaptic contacts with GABAergic PGN somata, dendrites, and spines. In the dLGN, RLD profiles primarily synapse with GABAergic dendrites that contain vesicles (F2 profiles) and non-GABAergic dendrites in glomerular arrangements that include triads. Occasional synapses on GABAergic somata and proximal dendrites were also observed in the dLGN. These results suggest that correlated dLGN activity may be enhanced via direct synaptic contacts between thalamocortical cells, whereas noncorrelated activity (such as that occurring during binocular rivalry) could be suppressed via thalamocortical collateral input to PGN cells and dLGN interneurons.