Review Of The Anatomy And Physiology Of The Vertebrate Retina

The retinas of all vertebrates follow a general anatomical pattern. Located on the rear, inside surface of an eyeball, it has ten anatomical layers. Starting from the vitreous humor, the layers are (1) internal limiting membrane; (2) optic nerve axon layer; (3) ganglion cell layer (about 5 x 105 per eye in the frog); (4) inner plexiform layer; (5) inner nuclear layer; (6) outer plexiform layer; (7) outer nuclear layer; (8) external limiting membrane; (9) sensory cells (about 106 rods and cones in the frog); (10) pigment epithelium layer. The pigment epithelial layer is responsible for eye shine in nocturnal animals. Located on the inner surface of the eyeball, it reflects light back through the photoreceptors, increasing visual efficiency in trapping photons at low light levels.

There are five major classes of neurons arranged in the ten retinal layers: Ganglion cells (GCs) and their axons are in layers 1 and 2. The inner plexiform layer contains dendrites and synapses from ganglion cells, amacrine cells (> 20

types), and bipolar cells (3 types). An inner nuclear layer contains the cell bodies (somata) from amacrine cells, bipolar cells, and horizontal cells. The outer plexiform layer contains dendrites and synapses between the bipolar cells, horizontal cells, and the rods and cones. The rods and cones are the photoreceptor (PR) cells that transduce photon energy into cell membrane hyperpolarization. Figure 6.1-1 illustrates schematically the five major types of neuron in the vertebrate retina and their basic interconnections. The ganglion cells are the spiking output neurons of the retina; their axons form the optic nerve. In the cat retina, there are some 23 subtypes of GCs (Kolb et al., 1999).

Distal

Retinal Ganglionic Cell Layer

Outer nuclear layer

Outer plexiform layer

Inner nuclear layer

Ganglion cell layer

Light

To optic nerve

FIGURE 6.1-1 The five major classes of neurons in the vertebrate retina: photoreceptors (rods and cones), horizontal, bipolar, amacrine, and ganglion cells, arranged in 5 anatomical layers. Information from the photoreceptors flows vertically and horizontally. (From Kandel, E.R. et al., Principles of Neural Science, 3rd ed., Appleton & Lange, Norwalk, CT, 1991. McGraw-Hill Companies, with permission.)

Light

Outer nuclear layer

Outer plexiform layer

Inner nuclear layer

Inner f)lexiform ayer

Ganglion cell layer

To optic nerve

Distal

FIGURE 6.1-1 The five major classes of neurons in the vertebrate retina: photoreceptors (rods and cones), horizontal, bipolar, amacrine, and ganglion cells, arranged in 5 anatomical layers. Information from the photoreceptors flows vertically and horizontally. (From Kandel, E.R. et al., Principles of Neural Science, 3rd ed., Appleton & Lange, Norwalk, CT, 1991. McGraw-Hill Companies, with permission.)

Light entering the eye must pass through the inner layers of the retina before it impinges on the outer segments of the rods and cones where transduction occurs. In the dark, the soma of a rod or cone has a resting potential of about -40 mV. Light absorbed in the outer segment causes a chemical reaction that closes cGMP-gated sodium leakage channels in the outer segment. At the same time, potassium ions leak out of the inner segment of the PR cells, unaffected by the light. Metabolically driven ion pumps in the PR cell membrane continuously pump Na+ out while K+ is brought in (Kandel et al., 1991). In very intense light, all the Na+ leakage channels in the outer segment close, and the membrane potential of the PR cell soma hyperpolarizes; it saturates near -70 mV, the K+ Nernst potential. Intermediate values of light intensity cause hyperpolarization that is closely proportional to the logarithm of the light intensity.

It is interesting to compare properties of rod and cones. Cones are associated with bright-light (photopic), high-acuity color vision. Each cone contains one of three different photopigments, each having an absorption spectrum at a different wavelength. Because cones contain less photopigment than rods, they are less sensitive, and saturate at higher light intensities. Rods, on the other hand, are adapted for low-light (scotopic) vision; they contain only one type of photopigment, hence give monochromatic vision. They have high transduction gain, responding even to single photons; they saturate in daylight. They have a slower response time than cones to sudden changes in intensity; their flicker fusion frequency is about 12 Hz compared with about 50 Hz in cones. Rods evidently trade off increased sensitivity for reduced acuity.

How the graded hyperpolarizations of PR cells leads to spike outputs of various types from the ganglion cells is now described. Consider first only cones and bipolar cells (BCs) in the outer plexiform layer. Cones make chemical synapses with two types of BC, the off-center BC and the on-center BC. In both cases, in the dark, the cone continuously releases a single neurotransmitter, probably glutamate. When light acts on the outer segment of the cone, the soma hyperpolarizes, reducing the continuous rate of release of glutamate by the cone synapses. The reduced neurotrans-mitter causes the off-center BC to hyperpolarize, and the on-center BC to depolarize (different ion channels are affected by the glutamate in the two types of BCs). Each type of BC makes an excitatory synaptic connection with a corresponding type of GC (off-center GC and on-center GC). In summary, when a flash of light is directed at a cone connected to an on-center BC and an off-center GC, the former depolarizes and the latter hyperpolarizes. The on-center GC depolarizes and generates an ON burst of spikes. The off-center GC is hyperpolarized by the off-center BC, suppressing its output spikes. It is probable that there is inhibitory cross-coupling between BCs and the corresponding GCs. That is, the on-center BC inhibits or hyperpolarizes further the off-center GC. Similarly, the off-center BC may inhibit the on-center GC (Kandel et al., 1991).

When the role of the amacrine cells in the function of the retina is examined, the picture becomes much more complex. There are over 40 morphologically and neurochemically distinguishable types of amacrine cells that use at least eight different neurotransmitters (Kolb et al., 1999). It is not difficult to speculate that amacrine cells function in the feature extraction operations found in various retinas.

There are two types of horizontal cells (HCs) found in most mammalian retinas. One is an axonless, A-type, that contacts only cones. The B-type has a soma and dendrites contacting cones, and an axon several hundred micrometers long that has terminal arborizations that contact only rods (Kolb et al., 1999).

HCs mediate an antagonistic action between neighboring cones. As has been seen, cones in the surround of the receptive field (RF), in the dark, are depolarized and continuously release glutamate neurotransmitter that acts on the connecting HCs to keep them slightly depolarized. In this state, the HCs release an inhibitory neurotransmitter that maintains cones in the center of the RF in a hyperpolarized state. Illumination of cones in the RF surround hyperpolarizes the surround cones, which in turn further hyperpolarizes the HC. This in turn reduces the rate of release of inhibitory neurotransmitter by the HC and results in the depolarization of the cones in the RF center. Now more glutamate is released by the center cone, leading to hyperpolarization of the on-center BCs, leading to a reduced firing rate of the on-center GC. Thus, light in the RF surround inhibits the on-center GC output. The net effect is a form of lateral inhibition of the feed-forward type.

The neural circuitry on the retina used for low-light vision deserves special mention. The rod photoreceptors, as discussed above, have very high transduction gain, responding at the single photon level. At moderately dim light levels (e.g., at dusk), the rods evidently transmit their light-caused hyperpolarizations to adjacent cones by fast electrical synapses (gap junctions). This means that even though the light level is too low for the cones to be effective transducers, the rods share the same high-resolution signal-processing pathways used by the cones in photopic vision. Thus, there is little lost in spatial resolution; color fades as the light dims, however. Under very low light conditions (e.g., starlight), the gap junction synapses connecting rods to cones close, blocking this pathway; instead, the rods stimulate special rod BCs through ribbon synapses. The rod BC outputs go to type A II amacrine cells, which synapse directly with off-center GCs, and indirectly with on-center GCs via cone BCs. See Figure 6.1-2 for a schematic description of these pathways.

Nes Synapse

FIGURE 6.1-2 Diagram showing the details of the synaptic connections between retinal interneurons. Rod and cones synapse on BCs and ACs. Rods and cones each synapse on different BCs; however, these pathways converge on the GC layer. Electrical synapses occur between cones and rods. Cones connect to two different classes of BCs with morphologically different synapses; basal (flat) synapses make contact with off-center BCs, ribbon synapses contact on-center BCs. The on-center BCs send their dendrites into invaginations in the cone terminal; there they form the central element of three clustered synapses, the other two of which come from dendrites of HCs. Rods synapse with one type of BC, which receives inputs only at ribbon synapses. These BCs do not synapse directly with GCs; instead, they synapse with type AII amacrine cells (ACs). The AII ACs relay their inputs by synapsing directly onto off-center GCs and onto BCs that connect to on-center GCs. (From Kandel, E.R. et al., Principles of Neural Science, 3rd ed., Appleton & Lange, Norwalk, CT, 1991. With permission from the McGraw-Hill Companies.)

FIGURE 6.1-2 Diagram showing the details of the synaptic connections between retinal interneurons. Rod and cones synapse on BCs and ACs. Rods and cones each synapse on different BCs; however, these pathways converge on the GC layer. Electrical synapses occur between cones and rods. Cones connect to two different classes of BCs with morphologically different synapses; basal (flat) synapses make contact with off-center BCs, ribbon synapses contact on-center BCs. The on-center BCs send their dendrites into invaginations in the cone terminal; there they form the central element of three clustered synapses, the other two of which come from dendrites of HCs. Rods synapse with one type of BC, which receives inputs only at ribbon synapses. These BCs do not synapse directly with GCs; instead, they synapse with type AII amacrine cells (ACs). The AII ACs relay their inputs by synapsing directly onto off-center GCs and onto BCs that connect to on-center GCs. (From Kandel, E.R. et al., Principles of Neural Science, 3rd ed., Appleton & Lange, Norwalk, CT, 1991. With permission from the McGraw-Hill Companies.)

As neural structures go, the retina at first sight appears simple. However, like many physiological systems, the closer the structure, molecular physiology, and function of the retina is observed, the more complexity is revealed in its design. Many neurophysiologists have studied the retina in a wide variety of vertebrates, here listed alphabetically: cat, frog, goldfish, ground squirrel, pigeon, rabbit, salamander, turtle, to mention a few. Obviously, space prevents a detailed discussion of the results with each animal. Described below are some of the results of the classic, early studies on retinal neurophysiology. It is evident that there are many common response properties among animals' retinas.

Was this article helpful?

0 0
Peripheral Neuropathy Natural Treatment Options

Peripheral Neuropathy Natural Treatment Options

This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.

Get My Free Ebook


Responses

  • ghenet sheshy
    How do rods and cones generate action potentials to the optic nerve?
    8 years ago

Post a comment