Chang-Jin Jeon

Starburst amacrine cells express parvalbumin but not calbindin and calretinin in rabbit retina.

Calcium-binding proteins (CBPs) are important components in calcium-mediated cellular signal transduction. Among the many CBPs, at least three EF-hand CBPs, calbindin-D28K (CB), calretinin (CR), and parvalbumin (PV), have been extensively studied in the retina. In the present study, we investigated the expression patterns of these three CBPs in cholinergic starburst amacrine cells (SACs), which are the most important element for direction selectivity in the rabbit retina. Double-label immunocytochemical analysis of vibratome sections and single-cell injection after immunocytochemical analysis on whole mounts were carried out in rabbit retinas. We found that all SACs in the inner nuclear layer and the ganglion cell layer contained PV. However, none of the SACs in the inner nuclear layer or ganglion cell layer contained either CB or CR. These results suggest that PV, but not CR or CB, may act as a calcium-buffering protein in the SACs of the rabbit retina.

The pathological effects of connexin 26 variants related to hearing loss by in silico and in vitro analysis

Gap junctions (GJs) are intercellular channels associated with cell–cell communication. Connexin 26 (Cx26) encoded by the GJB2 gene forms GJs of the inner ear, and mutations of GJB2 cause congenital hearing loss that can be syndromic or non-syndromic. It is difficult to predict pathogenic effects using only genetic analysis. Using ionic and biochemical coupling tests, we evaluated the pathogenic effects of Cx26 variants using computational analyses to predict structural abnormalities. For seven out of ten variants, we predicted the variation would result in a loss of GJ function, whereas the others would completely fail to form GJs. Functional studies demonstrated that, although all variants were able to function normally as hetero-oligomeric GJ channels, six variants (p.E47K, p.E47Q, p.H100L, p.H100Y, p.R127L, and p.M195L) did not function normally as homo-oligomeric GJ channels. Interestingly, GJs composed of the Cx26 variant p.R127H were able to function normally, even as homo-oligomeric GJ channels. This study demonstrates the particular location and property of an amino acid are more important mainly than the domain where they belong in the formation and function of GJ, and will provide information that is useful for the accurate diagnosis of hearing loss.

Distribution of Parvalbumin-Immunoreactive Retinal Ganglion Cells in the Greater Horseshoe Bat, Rhinolophus ferrumequinum

Parvalbumin occurs in various types of cells in the retina. We previously reported parvalbumin distribution in the inner nuclear layer of bat retina. In the present study, we identified the parvalbumin- immunoreactive neurons in the ganglion cell layer of the retina of a bat, Rhinolophus ferrumequinum, and investigated the distribution pattern of the labeled neurons. Parvalbumin immunoreactivity was found in numerous cell bodies in the ganglion cell layer. Quantitative analysis showed that these cells had medium to large-sized somas. The soma diameter of the parvalbumin-immunoreactive cells in the ganglion cell layer ranged from 12.35 to 19.12 ㎛ (n=166). As the fibers in the nerve fiber layer were also stained, the majority of parvalbumin-immunoreactive cells in the ganglion cell layer should be medium to large-sized retinal ganglion cells. The mean nearest neighbor distance of the parvalbumin-immunoreactive cells in the ganglion cell layer of the bat retina ranged from 59.57 to 62.45 ㎛ and the average regularity index was 2.95 ± 0.3 (n=4). The present results demonstrate that parvalbumin is expressed in medium to large-sized retinal ganglion cells in bat retina, and they have a well-organized distributional pattern with regular mosaics. These results should be important as they are applicable to a better understanding of the unsolved issue of a bat vision. This data will help to provide fundamental knowledge for the better understanding of the unique behavioral aspects of bat flight maneuverability.

Immunocytochemical Localization of Calcium-binding Proteins, Calbindin D28K-, Calretinin-, and Parvalbumin-containing Neurons in the Dog Visual Cortex

Although the dog is widely used to analyze the function of the brain, it is not known whether the distribution of calcium-binding proteins reflects a specific pattern in the visual cortex. The distribution of neurons containing calcium-binding proteins, calbindin D28K, calretinin, and parvalbumin in adult dog visual cortex were studied using immunocytochemistry. We also compared this labeling to that of gamma-aminobutyric acid (GABA). Calbindin D28K-immunoreactive (IR) neurons were predominantly located in layer II/III. Calretinin- and parvalbumin-IR neurons were located throughout the layers with the highest density in layers II/III and IV. The large majority of calbindin D28K-IR neurons were multipolar stellate cells. The majority of the calretinin-IR neurons were vertical fusiform cells with long processes traveling perpendicular to the pial surface. And the large majority of parvalbumin-IR neurons were multipolar stellate and round/oval cells. More than 90% of the calretinin- and parvalbumin-IR neurons were double-labeled with GABA, while approximately 66% of the calbindin D28K-IR neurons contained GABA. This study elucidates the neurochemical structure of calcium-binding proteins. These data will be informative in appreciating the functional significance of different laminar distributions of calcium-binding proteins between species and the differential vulnerability of calcium-binding proteins-containing neurons, with regard to calcium-dependent excitotoxic procedures.

Immunocytochemical localization of the calcium-binding proteins calbindin D28K, calretinin, and parvalbumin in bat visual cortex.

It is a common misconception that bats are blind, and various studies have suggested that bats have visual abilities. The purpose of this study was to investigate the cytoarchitecture of calbindin D28K (CB)-, calretinin (CR)-, and parvalbumin (PV)-immunoreactive (IR) neurons in the bat visual cortex using immunocytochemistry. The highest density of CB- and PV-IR neurons was located in layer IV of the visual cortex. The majority of CB- and PV-IR neurons were characterized by a stellate or round/oval shape. CR-IR neurons were predominantly located in layers II/III, and the cells were principally round/oval in shape. Two-color immunofluorescence revealed that 65.96%, 24.24%, and 77.00% of the CB-, CR-, and PV-IR neurons, respectively, contained gamma-aminobutyric acid (GABA). We observed calcium-binding protein (CBP)-IR neurons in specific layers of the bat visual cortex and in specific cell types. Many of the CBP-IR neurons were GABAergic interneurons. These data provide useful clues to aid in understanding the functional aspects of the bat visual system.

Parvalbumin-Immunoreactive Cells in the Superior Colliculus in Dog: Distribution, Colocalization with GABA, and Effect of Monocular Enucleation

Parvalbumin (PV) is thought to play a major role in buffering intracellular calcium. We studied the distribution, morphology of PV-immunoreactive (IR) cells, and the effect of enucleation on the PV distribution in the superior colliculus (SC) in dog (Canis familiaris) and compared PV labeling to that of calbindin D28K (CB) and GABA. These cells formed three laminar tiers in the dog SC; 1) the upper superficial gray layer (SGL), 2) the lower optic layer (OL) and the upper intermediate gray layer, and 3) the deep layer. The third tier was not very distinct when compared with the other two tiers. The distribution of PV-IR cells is thus complementary to that of CB-IR tiers. Our present data on the distribution of PV-IR cells within the superficial layers are strikingly different from those in previously studied mammals, which show PV-IR cells within the lower SGL and upper OL. However, there were no distinct differences in distribution within the deep layers compared with that of previously studied mammals. PV-IR cells in the SC varied dramatically in morphology and size, and included round/oval, vertical fusiform, stellate, horizontal and pyriform cells. Two-color immunofluorescence revealed quantitatively that 11.67% of the PV-IR cells colocalized with GABA. Monocular enucleation appeared to have no effect on the distribution of PV-IR cells in the contralateral SC. Similar to CB, these data suggest that retinal projection may not control the expression of PV in the dog SC. These results provide important information for delineating similarities and differences in the neurochemical architecture of the visual system.

Immunocytochemical Localization of Melanopsin-immunoreactive Neurons in the Mouse Visual Cortex

Melanopsin is an opsin-like photopigment found in the small proportion of photosensitive ganglion cells of the retina. It is involved in the regulation of the synchronization of the circadian cycle as well as in the control of pupillary light reflex. The purpose of the present study is to investigate whether melanopsin is also expressed in the other areas of the central visual system outside the retina. We have studied the distribution and morphology of neurons containing melanopsin in the mouse visual cortex with antibody immunocytochemistry. Melanopsin immunoreactivity was mostly present in neuronal soma, but not in nuclei. We found that melanopsin was present in a large subset of neurons within the adult mouse visual cortex with the highest density in layer II/III. In layer I of the visual cortex, melanopsin-immunoreactive (IR) neurons were rarely encountered. In the mouse visual cortex, the majority of the melanopsin-IR neurons consisted of round/oval cells, but was varied in morphology. Vertical fusiform and pyramidal cells were also rarely labeled with the anti-melanopsin antibody. The labeled cells did not show any distinctive distributional pattern. Some melanopsin-IR neurons in mouse visual cortex co-localized with nitricoxide synthase, calbindin and parvalbumin. Our data indicate that melanopsin is located in specific neurons and surprisingly widespread in visual cortex. This finding raises the need of the functional study of melanopsin in central visual areas outside the retina.

Retrograde Tracer Studies of Tecto-Reticulospinal Pathway and Dorsal Lateral Geniculate Nucleus on GluR1- and GluR4-Immunoreactive Neurons in the Hamster Superior Colliculus

We recently reported the distributions of AMPA (α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate) receptor subtypes glutamate receptors (GluR) 1 and GluR4 in the superior colliculi (SC) of hamsters with antibody immunocytochemistry and the effect of enucleation on these distributions. We also compared these labelings to those of calcium-binding proteins calbindin D28K, calretinin, parvalbumin, and GABA. In the present study, we investigated whether the GluR1- and GluR4-immunoreactive (IR) neurons are interneurons or projection neurons by injection of the retrograde tracer horseradish peroxidase (HRP) into one of each major ascending and descending pathways of the SC. HRP injections were made into a tecto-reticulospinal pathway (TRS) and dorsal lateral geniculate nucleus (dLGN). Animals were then allowed to recover and to survive for 48 hr before perfusion. Sections containing retrograde-labeled neurons were then treated for GluR-immunoreactivity. HRP injections proved that only a small population of the GluR1-IR cells project into TRS (1.4%) and dLGN (2.6%). However, a large subpopulation of GluR4-IR cells project into TRS (32.7%). The differential compositions of inter/projection neurons, along with our previous studies on the separate distribution of the GluR subunits, its differential co-localization with calcium-binding proteins and GABA, and differential reactions to enucleations, strongly imply the functional variety of the receptor subunits in visual behavior responses.