Eun-Jin Lim

Changes in retinal neuronal populations in the DBA/2J mouse

DBA/2J (D2) mice develop a form of progressive pigmentary glaucoma with increasing age. We have compared retinal cell populations of D2 mice with those in control C57BL/6J mice to provide information on retinal histopathology in the D2 mouse. The D2 mouse retina is characterized by a reduction in retinal thickness caused mainly by a thinning of the inner retinal layers. Immunocytochemical staining for specific inner retinal neuronal markers, viz., calbindin for horizontal cells; protein kinase C (PKC) and recoverin for bipolar cells, glycine, γ-aminobutyric acid (GABA), choline acetyltransferase (ChAT), and nitric oxide synthase (NOS) for amacrine cells, and osteopontin (OPN) for ganglion cells, was performed to detect preferentially affected neurons in the D2 mouse retina. Calbindin, PKC, and recoverin immunoreactivities were not significantly altered. Amacrine cells immunoreactive for GABA, ChAT, and OPN were markedly decreased in number, whereas NOS-immunoreactive amacrine cells increased in number. However, no changes were observed in the population of glycine-immunoreactive amacrine cells. These findings indicate a significant loss of retinal ganglion and some amacrine cells, whereas glycinergic amacrine cells, horizontal, and bipolar cells are almost unaffected in the D2 mouse. The reduction in amacrine cells appears to be attributable to a loss of GABAergic and particularly cholinergic amacrine cells. The increase in nitrergic neurons with the consequent increase in NOS and NO may be important in the changes in the retinal organization that lead to glaucomain D2 mice. Thus, the D2 mouse retina represents a useful model for studying the pathogenesis of glaucoma and mechanisms of retinal neuronal death and for evaluating neuroprotection strategies.

AII amacrine cells in the mammalian retina show disabled-1 immunoreactivity.

Disabled 1 (Dab1) is an adapter molecule in a signaling pathway, stimulated by Reelin, which controls cell positioning in the developing brain. It has been localized to AII amacrine cells in the mouse and guinea pig retinas. This study was conducted to identify whether Dab1 is commonly localized to AII amacrine cells in the retinas of other mammals. We investigated Dab1‐labeled cells in human, rat, rabbit, and cat retinas in detail by immunocytochemistry with antisera against Dab1. Dab1 immunoreactivity was found in certain populations of amacrine cells, with lobular appendages in the outer half of the inner plexiform layer (IPL) and a bushy, smooth dendritic tree in the inner half of the IPL. Double‐labeling experiments demonstrated that all Dab1‐immunoreactive amacrine cells were immunoreactive to antisera against calretinin or parvalbumin (i.e., other markers for AII amacrine cells in the mammalian retina) and that they made contacts with the axon terminals of the rod bipolar cells in the IPL close to the ganglion cell layer. Furthermore, all Dab1‐labeled amacrine cells showed glycine transporter‐1 immunoreactivity, indicating that they are glycinergic. The peak density was relatively high in the human and rat retinas, moderate in the cat retina, and low in the rabbit retina. Together, these morphological and histochemical observations clearly indicate that Dab1 is commonly localized to AII amacrine cells and that antiserum against Dab1 is a reliable and specific marker for AII amacrine cells of diverse mammals. J. Comp. Neurol. 470:372–381, 2004.

AII amacrine cells in the distal inner nuclear layer of the mouse retina

We serendipitously found a distal Disabled‐1 (Dab1)‐immunoreactive cell in retina of the C57BL/6J black mouse. The somata of these cells are located in the outermost part of the inner nuclear layer (INL). Their processes extend toward the outer plexiform layer (OPL), receiving synaptic inputs from horizontal and interplexiform cells. In the current study, we name this cell the “distal Dab1‐immunoreactive cell.” Double‐labeling experiments demonstrate that the distal Dab1‐immunoreactive cell is not a horizontal cell. Rather, the distal Dab1 cell appears to be a misplaced AII cell, by being glycine transporter‐1‐immunoreactive and by resembling the latter cell in an electron microscopic analysis. A distal Dab1 cell had been reported in the FVB/N mouse retina, a model of retinitis pigmentosa (Park et al. [2004] Cell Tissue Res 315:407–412). However, here, we found this distal Dab1‐immunoreactive cell in the adult and normal developing mouse retinas. Hence, we show that such cells do not require the loss of photoreceptors as suggested previously (Park et al. [2004] Cell Tissue Res 315:407–412). Instead, two other pieces of data suggest an alternative explanation sources for distal Dab1 cells. First, we find a correlation between the number of these cells in the left and right eyes Second, developmental analysis shows that the distal Dab1‐immunoreactive cell is first observed shortly after birth. At the same time, AII cells emerge, extending their neurites into the inner retina. These data suggest that distal Dab1‐immunoreactive cells are misplaced AII amacrine cells, resulting from genetically modulated anomalies owing to migration errors. J. Comp. Neurol. 494:651–662, 2006.

Brain-derived neurotrophic factor modulates the dopaminergic network in the rat retina after axotomy

Dopaminergic cells in the retina express the receptor for brain-derived neurotrophic factor (BDNF), which is the neurotrophic factor that influences the plasticity of synapses in the central nervous system. We sought to determine whether BDNF influences the network of dopaminergic amacrine cells in the axotomized rat retina, by immunocytochemistry with an anti-tyrosine hydroxylase (TH) antiserum. In the control retina, we found two types of TH-immunoreactive amacrine cells, type I and type II, in the inner nuclear layer adjacent to the inner plexiform layer (IPL). The type I amacrine cell varicosities formed ring-like structures in contact with AII amacrine cell somata in stratum 1 of the IPL. In the axotomized retinas, TH-labeled processes formed loose networks of fibers, unlike the dense networks in the control retina, and the ring-like structures were disrupted. In the axotomized retinas treated with BDNF, strong TH-immunoreactive varicosities were present in stratum 1 of the IPL and formed ring-like structures. Our data suggest that BDNF affects the expression of TH immunoreactivity in the axotomized rat retina and may therefore influence the retinal dopaminergic system.

Ectopic localization of putative AII amacrine cells in the outer plexiform layer of the developing FVB/N mouse retina.

The FVB/N mouse is a model of retinitis pigmentosa which shows a rapid loss of photoreceptors during early postnatal (P) life. We investigated the cellular localization of glycine transporter 1 (GlyT-1) in the developing FVB/N mouse retina. In control retinas, the developmental pattern of GlyT-1-immunoreactive amacrine cells was well in accordance with a previous report. However, in the FVB/N mouse retina, some GlyT-1-labeled amacrine cells sent their processes into the outer plexiform layer (OPL) from P14 onward. From P21 onward, GlyT-1-labeled cells were visible in the OPL. These cells were further characterized by double-label immunofluorescence experiments with an antiserum against disabled 1 (Dab-1), and showed Dab-1 immunoreactivity, indicating that these cells are putative AII amacrine cells. These results clearly demonstrate that AII amacrine cells have the potential capacity to respond to photoreceptor degeneration by migrating or sprouting their processes into the OPL in the developing FVB/N mouse retina.

Identification and characterization of SMI32-immunoreactive amacrine cells in the mouse retina.

Mammalian neurons express the neural intermediate filament protein neurofilament (NF). In the retina, NFs have been detected primarily in the axons and processes of retinal ganglion and horizontal cells. We found an amacrine cell type that was immunolabeled with an antibody against SMI32, a non-phosphorylated epitope on neurofilament proteins of high molecular weight, in the mouse retina. This type of amacrine cell was non-randomly distributed, and these cells exhibited a central-peripheral density gradient. Most of these cells co-expressed GABA and ChAT, but not glycine or any other amacrine cell marker. These results suggest that some SMI32-immunoreactive amacrine cells belong to a GABAergic population, and that SMI32 can therefore be used as a marker for a subset of amacrine cells in addition to ganglion cells and horizontal cells in the mouse retina.

The absence of the clathrin-dependent endocytosis in rod bipolar cells of the FVB/N mouse retina

The high rate of exocytosis at the ribbon synapses is balanced by following compensatory endocytosis. Unlike conventional synaptic terminals where clathrin-mediated endocytosis (CME) is a predominant mechanism for membrane retrieval, CME is thought to be only a minor mechanism of endocytosis at the retinal ribbon synapses, but CME is present there and it works. We examined the clathrin expression in the FVB/N rd1 mouse, which is an animal model of retinitis pigmentosa. The broadly distributed pattern of clathrin immunoreactivity in the inner plexiform layer was similar in both the control and FVB/N mouse retinas, but the immunoreactive punta within the rod bipolar axon terminals located in the proximal IPL were decreased in number and reduced in size at postnatal days 14 and they came to disappear at postnatal days 21. This preferential decrease of the clathrin expression at ribbon synapses in the rod bipolar cell axon terminals of the FVB/N mouse retina demonstrates another plastic change after photoreceptor degeneration and this suggests that clathrin may be important for normal synaptic function at the rod bipolar ribbon synapses in the mammalian retina.