The Journal of Neurophysiology Vol. 77 No. 1 January 1997, pp. 511-516
Copyright ©1997 by the American Physiological Society

RAPID COMMUNICATION

An In Vitro Model of the Kitten Retinogeniculate Pathway

William Guido, Fu-Sun Lo, and Reha S. Erzurumlu

Department of Anatomy and Neuroscience Center of Excellence, Louisiana State University Medical Center, New Orleans, Louisiana 70112

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Guido, William, Fu-Sun Lo, and Reha S. Erzurumlu. An in vitro model of the kitten retinogeniculate pathway. J. Neurophysiol. 77: 511-516, 1997. An organotypic explant coculture method is described for the developing retinogeniculate pathway of the cat. Retinal explants and thalamic slices containing the dorsal lateral geniculate nucleus (LGN), derived from early postnatal kittens, can be grown in serum-free culture medium for several days. In such cultures, retinal ganglion cells (RGCs) and LGN neurons retained their age-specific morphological features and developed functional connections. Labeling of RGCs and their processes with DiI showed that all three major classes of RGCs (/Y, /X, /W) were present in cocultured retinal explants. Retinal axons readily regenerated into thalamic slices and, over time, developed arbors within the LGN. Retrograde labeling from the LGN traced the origin of these axons almost exclusively to -cells in the retina. In vitro intracellular recordings indicated that LGN cells maintained their basic electrophysiological properties in coculture. Current injection generated action potentials, and, at hyperpolarized levels, it led to low-threshold Ca²⁺ spiking. Regenerated retinal axons also formed functional connections with LGN neurons. Electrical stimulation of the retinal explant elicited excitatory postsynaptic responses (EPSPs) in LGN cells. Drop application of specific glutamate antagonists indicated that EPSPs had both N-methyl-D-aspartate (NMDA) and non-NMDA receptor components. The morphology of the LGN neurons was examined after intracellular injections of biocytin during recording. Labeled cells were very similar to those of early postnatal kittens. Although, in general, they had relatively small soma and simple dendritic branching patterns, a few could be recognized as X- or Y-cells. Thus the coculture model can be used to assay the regenerative propensity of different types of RGCs during development.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

The cat visual pathway from the retina through the lateral geniculate nucleus (LGN) of dorsal thalamus, to visual cortex serves as an important model for elucidating developmental mechanisms underlying the formation of ordered neural connections (see Cramer and Sur 1995; Garraghty and Sur 1993; Goodman and Shatz, 1993; Shatz 1990a,b, 1996; Shatz and Sretavan 1986). This pathway develops from a crudely wired, relatively undifferentiated network, into a highly organized sensory system. In the retina, physiologically and morphologically distinct cell types, namely-, -, and -cells, establish connections with similarly distinct relay neurons of the LGN, namely X, Y, and W-cells (see Sherman 1985). Presently, little is known about the cellular changes, particularly in the membrane properties and synaptic responses of developing LGN cells, during the establishment of parallel pathways. We have used explant cocultures of kitten retina and thalamic slices containing LGN to investigate the structural and electrophysiological features of retinogeniculate connectivity. The experimental approach described here circumvents many of the technical difficulties associated with in vivo experiments and provides a model in which morphological and electrophysiological experiments can be done in parallel with pharmacological manipulations to explore cellular mechanisms underlying the establishment and refinement of the retinogeniculate pathway.

	METHODS

Abstract Introduction Methods Results Discussion References

Explant cocultures

Kittens between 0 and 2 days of age were killed with halothane. The brains and eyes were removed and placed in sterile Gey's balanced salt solution (GIBCO) supplemented with 6.5% D(+)-Galactose (GBSS, kept at 4°C). The retina from each eye was dissected out in GBSS and divided into wedged shaped explants. The diencephalon containing the LGN was excised and 400-µm thick sections were cut in the coronal plane. Thalamic slices and retinal explants were first kept in cold GBSS and then transferred on to microporous membranes of Millicell culture inserts (Millipore). The explants were arranged such that the optic nerve side of the retina apposed the lateral edge of the LGN in a thalamic slice (Fig. 1F). The inserts then were placed in six-well culture plates. The cultures were kept in fresh serum-free medium (Brewer and Cotman 1989; Collazo et al. 1992) in a humidified CO₂ incubator at 33°C for 3-5 days.

View larger version (81K): [in this window] [in a new window]   FIG. 1. DiI labeled retinal ganglion cells (A-E) and their processes (G and H) in retinogeniculate cocultures. Arrangement of a typical retinogeniculate coculture photographed while in culture (F). Explants are aligned so that optic nerve side of retina is adjacent to lateral geniculate nucleus (LGN) of a thalamic slice. A-D: retinal ganglion cells from a retinal explant cocultured with LGN: - (A-C) and -cells (D) are present. E: DiI labeling from LGN resulted in retrograde labeling of mainly , and rarely -like cells, but not -cells. Note relatively straight trajectory of its long axon crossing retina-LGN border. G: example of a retinal axon beginning to arborize in LGN after 2 days in culture. H: example of another arbor in LGNs after 5 days in culture. This arbor has numerous branches and was ~110 µm wide. * in E and H depict retina-LGN border.

In vitro recording

After 3-5 days in culture, each coculture was lifted off the microporous membrane by immersion into oxygenated artificial cerebrospinal fluid [ACSF; containing (in mM) 124 NaCl, 2.5 KCl, 2 CaCl₂, 1 MgSO₄, 26 NaHCO₃, 1.25 NaHPO₄, and 10 dextrose; pH = 7.4] and transferred into a temperature-controlled slice recording chamber. Cultures were kept at an interface of warm (34°C) humidified air (95% O₂-5% CO₂) and ACSF. The boundaries of the LGN and retina were readily identifiable under a stereoscope with fiber optic illumination (Fig. 1F). In vitro, intracellular recordings were conducted using sharp-tipped glass electrodes filled with 4 M KAC or a 2% solution of biocytin dissolved in 2 M KAC (final impedance 80-110 M). All neuronal activity was recorded through a high-impedance amplifier with bridge balanced circuitry, then digitized (10-20 KHz), and stored directly on computer. Membrane levels were controlled by injecting DC current (continuously and/or step-wise manner) through the recording electrode. Current-voltage relationships were examined at membrane levels between 45 and 90 mV by injecting a series of square wave current pulses (100-300 ms, ±1 nA, 0.1 step) to reach steady state. Responses to these current step protocols also were used to determine the presence of any voltage dependent conductances and to explore the firing pattern of LGN cells.

To generate synaptic responses in LGN, teflon-coated bipolar stimulating electrodes (Pt-Ir) were placed in the retinal explant. Excitatory postsynaptic potentials (EPSPs) were evoked at relatively low stimulus frequencies (0.2-1 Hz) using brief (100-300 µs) current pulses (0.1-0.5 mA) over a wide range of membrane potentials. Drop application (5 µl) of selective glutamate antagonists, DL-2-amino-5-phosphonovaleric acid (DL-APV; 1 mM), and 6,7-dinitroquinoxaline-2,3-dione (DNQX; 250 µM) were used to determine whether EPSPs were comprised of N-methyl-D-aspartate (NMDA) and/or non-NMDA receptor components, respectively (c.f., Scharfman et al. 1990).

Morphological analyses

Some of the cocultures were fixed with 4% paraformaldehyde in phosphate buffer and small crystals of carbocyanine dye, DiI (Molecular Probes) were placed either in the retina or in the LGN. The first labeling approach allowed us to visualize retinal axons growing into the thalamus and to examine if all classes of retinal ganglion cells (RGCs) survived and differentiated in coculture. The second labeling approach enabled us to determine whether certain types of RGCs could regenerate their axons into the LGN. To compare the morphology of cocultured RGCs with that seen in normal retinae, we prepared retinal whole mounts from 2-day-old kittens and labeled the RGCs with DiI. All dye implanted preparations were kept at 35°C for 3-14 days, after which the labeled neuronal profiles were analyzed with a Nikon Diaphot microscope equipped with epifluorescence or by using a Leica confocal laser-scanning microscope.

During intracellular recording, some LGN cells were filled with biocytin by passing a small amount of positive and negative current through the recording electrode. Later, the explants were immersion fixed in buffered 2.5% paraformaldehyde and 0.25% glutaraldehyde. Biocytin-filled LGN cells then were visualized by using the ABC method (Horikawa and Armstrong 1988). Labeled cells were drawn using a drawing tube attached to a Nikon Labophot microscope.

	RESULTS

Abstract Introduction Methods Results Discussion References

Placement of DiI in the retina yielded retrograde labeling of a stream of retinal ganglion cells whose axons passed through the dye site. The same procedure also led to anterograde labeling of retinal axons that regenerated into the cocultured thalamic slice. Figure 1 illustrates the arrangement of retinogeniculate cocultures (Fig. 1F) as well as various labeled RGCs and their axonal processes (Fig. 1, A-E, G, and H). Within the retina, morphologically different types of ganglion cells could be visualized at similar eccentricities (Fig. 1, A-D). A comparison of DiI-labeled retinal wholemounts from 2-day-old kittens (data not shown) and retinal explants cocultured with LGN for 5 days indicated that all types of retinal ganglion cells survived and maintained class-specific morphology (Fig. 1, A-E) (c.f., Montague and Friedlander 1991). Some of these cells had medium size soma and small, bushy dendritic trees (Fig. 1D), typical of -cells. Others had large soma and widely dispersed dendritic trees characteristic of -cells (Fig. 1, A-C, and E). There were also cells with smaller soma and sparsely branched, but long dendrites resembling -cells (not shown). Another morphological characteristic of ganglion cells in coculture and in normal retinal wholemounts was the presence of extensive dendritic and somal spines, a transient feature of the developing cat retina (Ramoa et al. 1988).

It was also possible to visualize anterograde labeling of retinal axons in our coculture preparations. After 24 h in vitro, axons tipped with growth cones were observed crossing the retina-LGN border. After 3 days in culture, axons of RGCs began branching and their growth cones became simpler and smaller (Fig. 1G). With longer culture periods (5 days), retinal axons formed arbors within the borders of LGNs (Fig. 1H). Terminal branches often were beaded, each ending in a small growth cone. We were not able to determine if the terminal arbors were restricted to eye-specific layers of the LGN. During perinatal development, laminar organization can be revealed only by the arrangement of retinal arbors arising from each eye and not by the cytoarchitecture of the LGN (see Shatz 1990a).

In cocultures where the dye was placed in the LGN, it was possible to determine whether RGCs were able to regenerate their axonal processes into LGN. Often it was possible to follow single retinal ganglion cell axons from the LGN, across the retina-LGN border to the parent cell body several millimeters away from central retina (Fig. 1E). This labeling approach revealed that only specific classes of RGCs innervated LGN explants. The majority of labeled cells were of the  class (Fig. 1, A-C, and E). Rarely, isolated -cells (data not shown) were observed to cross into the LGN explant. However, we saw no evidence of retrogradely labeled -cells. DiI labeling in the retina clearly indicates that -cells are present in cocultured retinal explants (Fig. 1D), but retrograde labeling experiments failed to show axons of these cells in the LGN.

Next we examined the morphological and electrophysiological properties of cocultured LGN cells. We recorded the voltage responses of 31 LGN cells in coculture, and of these, 7 were labeled with biocytin. Examples of two labeled LGN neurons are shown in Fig. 2A. Cocultured LGN cells resembled those of early postnatal kittens (Friedlander 1982; Guido and Lo 1995; Mason 1983). Often these cells had large soma and lacked extensive dendritic branching (Fig. 2A, cell 2). Some also could be recognized as immature X and Y-cells. For example, cell 1 (Fig. 2A), resembled a Y-cell with its large soma and radially oriented dendritic field (Friedlander et al. 1981).

View larger version (25K): [in this window] [in a new window]   FIG. 2. Structure and function of developing LGN cells in coculture. A: reconstructions of 2 biocytin-filled LGN cells. Cell 1 resembles a Y-cell, having a radial dendritic field and large cell body. Cell 2 cannot be readily classified; however, it looks similar to cells from early postnatal kittens. B: voltage responses of 2 different LGN cells to intracellular current injection (250 ms, 0.1- to 0.2-nA step). Hyperpolarizing pulses produce linear responses that lack a "depolarizing sag" indicative of the activation of mixed cation conductance IH. Termination of larger hyperpolarizing pulses causes the membrane to repolarize and trigger small low-threshold Ca2+ spikes (see arrows). When depolarizing pulses reach spike threshold, they generate action potentials. C: examples of low threshold Ca2+ spikes in 2 different LGN cells activated either by depolarizing current injection (0.3 nA) from a hyperpolarized level (top), or by passive repolarization (bottom) after termination of a hyperpolarizing current pulse (0.3 nA). Riding peak of each low threshold spike is a conventional action potential. D: example of a spike train generated by depolarizing current injection (0.5 nA). Response shows modest frequency accommodation. E: excitatory postsynaptic potentials (EPSPs) trigger action potential firing during membrane depolarization. Shown are 2 responses of the same cell-to-current pulse injection. During delivery of smaller step (0.2 nA), electrical activation of retinal explant produces an EPSP below spike threshold. During larger step (0.3 nA), which led to a slightly higher level of membrane depolarization, stimulation produces an EPSP that triggers an action potential. F: EPSPs activate low threshold Ca2+ spikes at hyperpolarized membrane levels. Shown are 2 EPSPs recorded from same cell during membrane hyperpolarization. During strong hyperpolarization (0.5 nA), EPSP activates a low threshold Ca2+ spike (see arrow). G: pharmacology of EPSPs in coculture. Shown are synaptic responses from a single cell before and during drop application of glutamate antagonists. Left: stimulation of retinal explant evokes a large amplitude long-lasting EPSP. Middle: application of N-methyl-D-aspartate (NMDA) antagonist 2-amino-5-phosphonovaleric acid reduces amplitude of longer, slower component, leaving shorter, faster component intact. Right: subsequent application of the non-NMDA antagonist 6,7-dinitroquinoxaline-2,3-dione greatly reduces remaining component. Resting membrane potential, as indicated, was controlled by injecting constant DC current through recording electrode.

Examples of the voltage responses of LGN cells to intracellular current injection are shown in Fig. 2, B-D. The responses of cells in coculture were very similar to those of postnatal kittens (Guido and Lo 1995; Pirchio et al. 1990). Cells responded to depolarizing current pulses with wide, overshooting action potentials (Fig. 2B). Strong depolarization also could generate a train of spikes that showed moderate frequency accommodation (Fig. 2D). Depolarization from a hyperpolarized state triggered low threshold Ca²⁺ spikes (Steriade and Llinas 1988). Typically, these were recognized as a triangular depolarization with one to two action potentials riding their crest (Fig. 2C). Like their normal postnatal counterparts, LGN cells in coculture were missing some key membrane properties that do not emerge until postnatal day 14-21 (Guido and Lo 1995; see also Ramoa and McCormick 1994). Most notable was the lack of the mixed cation conductance I_H (McCormick and Pape 1990). Strong hyperpolarizing current pulses failed toactivate a "depolarizing sag" indicative of I_H (Fig. 2, Band C).

Finally, we tested whether the retinal axons growing into LGN formed functional connections, or merely reflected axon-substrate interactions. Electrical stimulation of the retinal explant activated EPSPs in 21 of 29 cells tested. Of our biocytin-filled cells (n = 7), three could be activated by electrical stimulation of the retinal explant and these either resembled immature Y cells or cells that could not be readily classified (see Fig. 2A). Unfortunately, our small sample of biocytin-filled cells precludes us from determining whether both X and Y cells form functional connections. EPSPs were of sufficient amplitude to generate action potential firing as well as low threshold spiking (Fig. 2, E and F). Drop application of selective glutamate antagonists also revealed that EPSPs had both NMDA and non-NMDA components (Fig. 2G). Stimulation of the retinal explant (top) evoked a large amplitude, long-lasting EPSP. Drop application of the NMDA antagonist APV (1 mM) reduced the amplitude of the late component, leaving the early, fast component intact. The late, slow component showed a strong voltage dependency (data not shown), and was greatest in amplitude at more depolarized membrane levels. Subsequent application of the non-NMDA antagonist DNQX (250 µM) diminished the remaining component.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

We present an experimentally accessible, in vitro model that captures the dynamic features of the developing retinogeniculate pathway. Both retinal ganglion and LGN cells survive and maintain their age-specific characteristics in explant cocultures. Retinal axons sprout and elaborate new terminal endings in the LGN. The electrophysiological properties of developing LGN cells also are preserved in coculture (Guido and Lo 1995). Stimulation of retinal explants reveal that LGN cells form functional connections with regenerated retinal axons. Evoked EPSPs of LGN cells have both non-NMDA and NMDA components and are of sufficient magnitude to generate activity in the form of action potentials and low threshold Ca²⁺ spikes.

Studies on the developing vertebrate sensory pathways indicate that coordinated activity between pre- and postsynaptic elements contribute to the emergence and refinement of precisely ordered neural connections (see Cramer and Sur 1995; Goodman and Shatz 1993; Shatz 1990b). Although activity plays an important role in shaping some aspects of retinogeniculate organization, little is known about underlying cellular mechanisms. NMDA receptor-mediated activity seems essential for some aspects of retinogeniculate organization such as the formation of "on/off" sublaminae (Hahm et al. 1991). However, other aspects of organization, such as the formation of eye-specific lamination (Smetters et al. 1994) and LGN dendritic morphology (Dalva et al. 1994) are not regulated by such activity. Other signals, particularly in the form of retrograde messengers such as nitric oxide (Cramer et al. 1995; Kratz et al. 1994), or metabotropic glutamate receptors (McCormick and Von Krosigk 1992), as well as the expression of neurotrophins by target cells (Allendoerfer et al. 1994; King et al. 1993) also may participate in the establishment and refinement of retinogeniculate connectivity. The coculture model described in this study offers a simple and relatively easy means to systematically manipulate these factors. Because the cultures are grown in a chemically defined, serum-free medium, various neurotrophins, agonists or antagonists to candidate transmitter substances, or neuromodulators can be added to the culture medium to assay their effects on cellular differentiation and development of retinogeniculate connectivity.

The kitten retinogeniculate pathway is assembled primarily during prenatal life. By the time of birth, RGC axons have established adult-like arbor patterns in eye-specific layers of the LGN (see Garraghty and Sur 1993; Shatz and Sretavan 1986). Thus in our cocultures using explants from postnatal kittens, the growth of retinal axons into thalamic explants and establishment of synaptic connectivity reflects a regenerative event rather than ab initio form of connectivity. Our DiI labeling experiments show that mostly -cells regenerate axons into thalamus, arborize and make functional connections with LGN cells. Several possibilities singly or in combination could account for this class-specific regeneration in culture. First, during neurogenesis, -cells differentiate earlier than -cells and their axons arrive in the LGN prior to those of -cells (see Garraghty and Sur 1993; Shatz 1990a; Shatz and Srevatan 1986). As development proceeds, later arriving -cell axons begin expanding their arbors while -cell arbors are pruned (see Sur 1988). Based on this developmental sequence, it is conceivable that the /X-cell pathway might actually loose its regenerative propensity and ability to reinnervate LGN cells around the time of birth. Because our cocultures were prepared from early postnatal tissue perhaps we simply missed the period in which -cells are capable of regenerating. Coculture experiments that use retinal explants from fetal kittens, or perhaps from newborn ferrets, would help to resolve this issue. Second, the rate of axonal regeneration and elongation may be substantially slower in -cells. Extending the culture period (beyond 5 days) might enable the axons of -cells to regenerate and grow into LGN explants. Third, these differences may reflect the more "malleable" nature of the /Y-cell pathway. Competitive interactions between  and  cell axons in exacting terminal space in the LGN are well noted (see Garraghty and Sur 1993; Shatz 1990a; Shatz and Sretavan 1986; Sur 1988). Experiments which alter these competitive interactions (e.g., monocular eye lid suture or monocular enucleation) have a profound and preferential effect on the development of the /Y cell pathway (Friedlander et al. 1982; see Garraghty and Sur 1993; Sherman and Spear 1982; Sur 1988). Within this context, it is also worth noting that mature /Y-cells seem to show a greater regenerative capacity than other retinal ganglion cell types. Experiments in which transected optic nerve is sutured to a peripheral nerve reveal that -cells more readily regenerate their axons into a peripheral nerve graft (Watanabe et al. 1993). Taken together with the present results, it would appear that the intrinsic biological properties of /Y-cells differs substantially from other ganglion cell types.

	ACKNOWLEDGEMENTS

  This work was supported by the National Science Foundation (IBN 9396270), National Institutes of Neurological Disorders and Stroke (NS-32195), and LSU Neuroscience Center Incentive Grants.

	FOOTNOTES

  Address reprint requests to W. Guido.

  Received 19 August 1996; accepted in final form 3 October 1996.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ALLENDOERFER, K. L., CABELLI, R. J., ESCANDON, E., KAPLAN, D. R., NIKOLICS, K., SHATZ, C. J. Regulation of neurotrophin receptors during the maturation of the mammalian visual system. J. Neurosci. 14: 1795-1811, 1994.[Abstract]
• BREWER, G. J., COTMAN, C. W. Survival and growth of hippocampal neurons in defined medium at low density: advantages of a sandwich culture technique or low oxygen. Brain Res. 494: 65-74, 1989.[Medline]
• COLLAZO, D., TAKAHASHI, H., MCKAY, R. D. G. Cellular targets and trophic functions of neurotrophin-3 in the developing rat hippocampus. Neuron 9: 643-656, 1992.[Medline]
• CRAMER, K. S., MOORE, C. I., SUR, M. Transient expression of NADPH-Diaphorase in the lateral geniculate nucleus of the ferret during early postnatal development. J. Comp. Neurol. 353: 306-315, 1995.[Medline]
• CRAMER, K. S., SUR, M. Activity-dependent remodeling of connections in the mammalian visual system. Curr. Opin. Neurobiol. 5: 106-111, 1995.[Medline]
• DALVA, M. B., GHOSH, A., SHATZ, C. J. Independent control of dendritic and axonal form in the developing lateral geniculate nucleus. J. Neurosci. 14: 3588-3602, 1994.[Abstract]
• FRIEDLANDER, M. J. Structure of physiologically classified neurones in the kitten dorsal lateral geniculate nucleus. Nature Lond. 300: 180-183, 1982.[Medline]
• FRIEDLANDER, M. J., LIN, C.-S., STANFORD, L. R., SHERMAN, S. M. Morphology of functionally identified neurons in lateral geniculate nucleus of the cat. J. Neurophysiol. 46: 80-129, 1981.[Free Full Text]
• FRIEDLANDER, M. J., STANFORD, L. R., SHERMAN, S. M. Effects of monocular deprivation on the structure-function relationship of individual neurons in the cat's lateral geniculate nucleus. J. Neurosci. 2: 321-330, 1982.[Abstract]
• GARRAGHTY, P. E., SUR, M. Competitive interactions influencing the development of retinal axonal arbors in the cat lateral geniculate nucleus. Physiol. Rev. 73: 529-545, 1993.[Free Full Text]
• GOODMAN, C. S., SHATZ, C. J. Developmental mechanisms that generate precise patterns of neuronal connectivity. Cell 72: 77-98, 1993.
• GUIDO, W., LO, F.-S. Membrane properties and synaptic responses of mature, immature, and visually-deprived LGN cells. Soc. Neurosci. Abstr. 21: 1306, 1995.
• HAHM, J.-O., LANGDON, R. B., SUR, M. Disruption of retinogeniculate afferent segregation by antagonists to NMDA receptors. Nature Lond. 351: 568-570, 1991.[Medline]
• HORIKAWA, K., ARMSTRONG, W. E. A versatile means of intracellular labeling injection of biocytin and its detection with avidin conjugates. J. Neurosci. Methods 25: 1-11, 1988.[Medline]
• KING, V. R., XUE, J.-T., SPEAR, P. D. Postnatal development of nerve growth factor receptor expression in the cat retina. Invest. Ophthalmol. Visual Sci. Suppl. 34: 1221, 1993.
• KRATZ, K. E., BANFRO, F., MIZE, R., SCHIENER, C., GUIDO, W. Prenatal and early postnatal staining of NADPH-diaphorase in cortical and thalamic structures of the cat. Soc. Neurosci. Abstr. 20: 134, 1994.
• MASON, C. A. Postnatal maturation of neurons in the cat's lateral geniculate nucleus. J. Comp. Neurol. 217: 458-469, 1983.[Medline]
• MCCORMICK, D. A., PAPE, H. C. Properties of a hyperpolarization-activated cation current and its role in rhythmic oscillation in thalamic relay neurones. J. Physiol. Lond. 431: 291-318, 1990.[Abstract/Free Full Text]
• MCCORMICK, D. A., VON KROSIGK, M. Corticothalamic activation modulates thalamic firing through metabotropic glutamate receptors. Proc. Natl. Acad. Sci. USA 89: 2774-2779, 1992.[Abstract/Free Full Text]
• MONTAGUE, P. R., FRIEDLANDER, M. J. Morphogenesis and territorial coverage by isolated retinal ganglion cells. J. Neurosci. 11: 1440-1457, 1991.[Abstract]
• PIRCHIO, M., LIGHTOWLER, S., CRUNELLI, V. Postnatal development of the T calcium current in cat thalamocortical cells. Neuroscience 38: 39-45, 1990.[Medline]
• RAMOA, A. S., CAMPBELL, G., SHATZ, C. J. Dendritic growth and remodeling of cat retinal ganglion cells during fetal and postnatal development. J. Neurosci. 8: 4239-4261, 1988.[Abstract]
• RAMOA, A. S., MCCORMICK, D. A. Developmental changes in electrophysiological properties of LGNd neurons during reorganization of retinogeniculate connections. J. Neurosci. 14: 2089-2097, 1994.[Abstract]
• SCHARFMAN, H. E., LU, S. M., GUIDO, W., ADAMS, P. R., SHERMAN, S. M. N-Methyl-D-aspartate receptors contribute to excitatory postsynaptic potentials of cat lateral geniculate neurons recorded in thalamic slices. Proc. Natl. Acad. Sci. USA 87: 4548-4552, 1990.[Abstract/Free Full Text]
• SHATZ, C. J. Competitive interactions between retinal ganglion cells during prenatal development. J. Neurobiol. 21: 197-211, 1990a.[Medline]
• SHATZ, C. J. Impulse activity and the patterning of connections during CNS development. Neuron 5: 745-756, 1990b.[Medline]
• SHATZ, C. J. Emergence of order in visual system development. Proc. Natl. Acad. Sci. USA 93: 602-608, 1996.[Abstract/Free Full Text]
• SHATZ, C. J., SRETAVAN, D. W. Interactions between retinal ganglion cells during the development of the mammalian visual system. Annu. Rev. Neurosci. 9: 171-207, 1986.[Medline]
• SHERMAN, S. M. Functional organization of the W-, X-, and Y-cell pathways: a review and hypothesis. Progr. Psychobiol. Physiol. Psychol. 11: 233-313, 1985.
• SHERMAN, S. M., SPEAR, P. D. Organization of visual pathways in normal and visually deprived cats. Physiol. Rev. 62: 738-855, 1982.[Free Full Text]
• SMETTERS, D. K., HAHM, J.-O., SUR, M. An N-methyl-D-aspartate receptor antagonist does not prevent eye-specific segregation in the ferret retinogeniculate pathway. Brain Res. 658: 168-178, 1994.[Medline]
• STERIADE, M., LLINAS, R. R. The functional states of the thalamus and the associated neuronal interplay. Physiol. Rev. 68: 649-742, 1988.[Free Full Text]
• SUR, M. Development and plasticity of retinal X and Y axon terminations in the cat's lateral geniculate nucleus. Brain Behav. Evol. 31: 243-251, 1988.[Medline]
• WATANABE, M., SAWAI, H., FUKUDA, Y. Number, distribution and morphology of retinal ganglion cells with axons regenerated into peripheral nerve graft in adult cats. J Neurosci. 13: 2105-2117, 1993.[Abstract]

0022-3077/97 $5.00 Copyright ©1997 The American Physiological Society