Study guidelines
- 1.
Define the layers of the retina and the cell types located in each.
- 2.
Contrast a rod versus a cone cell with respect to structure, function, and localisation within the retina.
- 3.
Be able to trace the pathway from retina to occipital cortex and distinguish between the function of those optic nerve fibres that will end in the midbrain and those that project to the cortex.
- 4.
Given a visual field pattern, be able to localise the site of involvement within the visual pathway.
The visual pathways are of outstanding importance in clinical neurology. They extend from the retinas of the eyes to the occipital lobes of the brain. Their great length makes them especially vulnerable to demyelinating diseases such as multiple sclerosis, to tumours of the brain or pituitary gland, to vascular lesions in the territory of the middle or posterior cerebral artery, and to head injuries.
The visual system comprises the retinas, the visual pathways from the retinas to the brainstem and visual cortex, and the cortical areas devoted to higher visual functions. The retinas and visual pathways are described in this chapter. Higher visual functions are described in Chapter 29 .
Retina
The retina and the optic nerves are part of the central nervous system . In the embryo the retina is formed by an outgrowth from the diencephalon called the optic vesicle ( Chapter 1 ). The optic vesicle is invaginated by the lens and becomes the two-layered optic cup.
The outer layer of the optic cup becomes the pigment layer of the mature retina. The inner, nervous layer of the cup gives rise to the retinal neurons.
Figure 28.1 shows the general relationships in the developing retina. The nervous layer contains three principal layers of neurons: photoreceptors , which become applied to the pigment layer when the intraretinal space is resorbed; bipolar neurons ; and ganglion cells , which give rise to the optic nerve and project to the thalamus and midbrain.
Note that the retina is inverted : light must pass through the layers of optic nerve fibres, ganglion cells, and bipolar neurons to reach the photoreceptors. The ‘rationale’ for an arrangement where the photoreceptors are ‘farthest away’ from their source of stimulation, light or photons, is multifold. First, this arrangement juxtaposes the apical end of the photoreceptors (which houses their light sensitive photopigment) against the retinal pigment layer that can absorb any scattered light or light that does not react with these photoreceptor cells. Second, the retinal pigmented epithelial cells also fulfil a phagocytic role. The light sensitive photopigment within the rod photoreceptor cells has a short half-life and needs to be continually replaced. New photopigment is generated at the base of the rod cells and migrates towards the cell apex, while the aged apical components are shed, phagocytised by the retinal pigmented epithelial cells, and the proteins recycled (cones do not shed). Finally, the photoreceptor cells have a high metabolic rate, and at this innermost retinal position, they are closest to capillaries within the choroid (which underlies this pigment epithelium) that supply their nourishment.
At the point of most acute vision, the foveola, the bipolar and ganglion cell layers lean away from a central pit (fovea), and light strikes the photoreceptors directly with minimal distortion (see Foveal Specialisation, later). In the mature eye the fovea is about 1.5 mm in diameter and occupies the centre of the 5 mm wide macula lutea (‘yellow spot’) where many of the photoreceptor cells contain yellow pigment. The fovea is the point of most acute vision and lies in the visual axis —a line passing from the centre of the visual field of the eye, through the centre of the lens, to the fovea ( Figure 28.2 ). To fixate or foveate an object is to gaze directly at it so that light reflected from its centre registers on the fovea.
The axons of the ganglion cells enter the optic nerve at the optic nerve head (optic papilla) , which is devoid of retinal neurons and constitutes the physiologic blind spot .
The visual fields of the two eyes overlap across two-thirds of the total visual field. Outside this binocular field is a monocular (temporal) crescent on each side ( Figure 28.3 ). During passage through the lens, the image of the visual field is reversed, with the result that objects in the left part of the binocular visual field register on the right half of each retina and objects in the upper part of the visual field register on the lower half. This arrangement is preserved all the way to the visual cortex in the occipital lobe.
From a clinical standpoint, it is essential to appreciate that vision is a crossed sensation . The visual field on one side of the visual axis registers on the visual cortex of the opposite side. In effect the right visual cortex ‘sees the left visual field’ or space and vice versa. Only half of the visual information from each retina crosses in the optic chiasm , for the simple reason that the other half has already crossed the midline in space.
Visual defects caused by interruption of the visual pathway are always described from the patient’s point of view , that is, in terms of the visual fields, and not in terms of retinal topography.
Structure of the retina
In addition to the serially arranged photoreceptors, bipolar cells, and ganglion cells shown in Figure 28.1 , the retina contains two sets of neurons arranged transversely: horizontal cells and amacrine cells ( Figure 28.4 ). A total of eight layers are described for the retina as a whole.
The ganglion cells generate action potentials providing the ‘requisite speed for conduction’ to the thalamus and midbrain. For the other cell types, distances are very short and passive electrical charge ( electrotonus ) or graded changes within their cell membrane potential are sufficient for intercellular communication, whether by gap-junctional contact or transmitter release.
Photoreceptors
The photoreceptor neurons comprise rods and cones .
Rods function only in dim light and are not sensitive to colour (electromagnetic wavelength energy). They are scarce in the outer part of the fovea and absent from its centre. Cones respond to bright light, are sensitive to colour and to shape, and are most numerous in the fovea. (In the human eye it is estimated that there are 130 million photoreceptor cells; rods outnumber cone cells by 20 to 1 and with the exception of the fovea, are distributed throughout the retina.)
Each photoreceptor cell has an outer and an inner segment and a synaptic end-foot. In the outer segment (light sensing ‘organelle’) there are hundreds of stacked membranous discs (rods) or membrane infoldings (cones) that incorporate a visual pigment ( rhodopsin is the photopigment that absorbs light or photons and initiates a molecular cascade that results in a change in the photoreceptor membrane potential and alters the release of neurotransmitter at its synaptic end-foot; this process is called phototransduction ); new discs are formed in the inner segment of rods and transported to the outer segment, old discs are shed from the apical portion of the outer segment. The synaptic end-foot makes contact with bipolar neurons and horizontal cell processes in the outer plexiform layer .
A surprising feature of the photoreceptors is that they are hyperpolarised by light. During darkness, sodium ion (Na + ) channels are opened, creating sufficient positive electrotonus to cause leakage of the transmitter (glutamate) from their end-feet onto their bipolar neurons. Illumination causes those Na + channels to close and this change in photoreceptor membrane potential is detected by their bipolar neurons. When the receptor becomes hyperpolarised it releases less neurotransmitter as its action was inhibitory, then the bipolar (and horizontal) cells will be depolarised (excited), but if its action was excitatory, those cells will be hyperpolarised (inhibited).
Rod cells are all hyperpolarised by light, so at high levels of illumination their membrane channels are all closed and their contribution to vision is minimal, and vision depends upon the function of the cones.
Cone and rod bipolar neurons
Cone bipolar neurons
Cone bipolar neurons are of two types. ON bipolars are switched on (depolarised) by light, being inhibited by transmitter released in the dark. They converge onto ON ganglion cells. OFF bipolars have the reverse response and converge onto OFF ganglion cells ( Figure 28.5 ). Typically, one cone cell will synapse with a few cone bipolar neurons, but at the fovea it is a one-to-one relationship; each will synapse with one ganglion cell.
Rod bipolar neurons
Rod bipolar neurons activate ON and OFF cone ganglion cells indirectly via amacrine cells ( Figure 28.5 ). One rod bipolar neuron will synapse with 15 to 30 rod cells (additional convergence will occur as the response is further transmitted centrally).
Horizontal cells
The dendrites of horizontal cells are in contact with photoreceptors. The peripheral dendritic branches give rise to axon-like processes, which make inhibitory contacts with bipolar neurons.
The function of horizontal cells is to inhibit bipolar neurons outside the immediate zone of excitation. The excited bipolars and ganglion cells are said to be ‘on-line’ and the inhibited ones ‘off-line’.
Amacrine cells
Amacrine cells have no axons. Their appearance is octopus-like; the dendrites all emerge from one side of the cell. Dendritic branches come into contact with bipolar neurons and ganglion cells.
More than a dozen different morphologic types of amacrine cells have been identified, as well as several different transmitters, including acetylcholine, dopamine, and serotonin. Possible functions include contrast enhancement and movement detection. For the rods, they convert large numbers from OFF to ON with respect to their ganglion cells.
Ganglion cells
The ganglion cells receive synaptic contacts from their bipolar neurons in the inner plexiform layer. The typical response of ganglion cells to bipolar activity is ‘centre-surround’. The centre of the receptive field represents the direct connections the ganglion cell receives from its photoreceptors; the surround of the receptive field are the connections it receives from adjacent photoreceptors via horizontal cells. An ON ganglion cell is excited by a spot of light and inhibited by a surrounding annulus (ring) of light. Horizontal cells cause the inhibition. OFF ganglion cells give the reverse response.
Coding for colour
There are three types of cone photoreceptor cells with respect to spectral sensitivity. One type is sensitive to red (also called L cones based on the ‘longer’ frequency of light they detect), one to green (M cone), and one to blue (also called S cones, but making up perhaps only 5 to 10% of the cones). The sensitivity is determined by the particular visual pigment of each cone cell type. While the light frequency that results in maximal stimulation identifies the particular cone type, the cones actually respond to a broader spectrum of light frequency, and the three cone types overlap one another. Colour is perceived not by one cone cell, but by comparing the activity of different cone cells to a particular frequency or colour of light. Groups of each type are connected to ON or OFF ganglion cells. (Processing of colour begins at the retina but continues at the lateral geniculate nucleus and cortex.)
The characteristic response of ganglion cells is one of colour opponency (one colour will activate a group of cone cells and their particular ganglion cell while its ‘opponent’ will inhibit it, or they can be considered as mutually exclusive):
- •
Ganglion cells that are on-line for green are off-line for red and ganglion cells that are on-line for red are off-line for green.
- •
Ganglion cells that are on-line for blue are off-line for yellow and ganglion cells that are on-line for green are off-line for yellow.
- •
Finally, there is a similar process for white and black or luminance.
Coding for black and white
White light is a mixture of green, red, and blue. In bright conditions it is encoded by the three corresponding cones, all of them converging onto common ganglion cells. Both ON and OFF ganglion cells are involved in black-and-white vision, just as in colour vision.
In very dim conditions, such as starlight, only rod photoreceptors are active, and objects appear in varying shades of grey. The rods are subject to the same rules as cones, showing centre-surround antagonism between white and black, and being connected to ON or OFF ganglion cells.
Most rod and cone ganglion cells are small ( parvocellular or ‘P’), having small receptive fields and being responsive to colour and shape. A minority are large ( magnocellular or ‘M’), having large receptive fields and being especially responsive to movements within the visual field.
Foveal specialisation
The relative density of cones increases progressively, and their size diminishes progressively, from the edge of the fovea inwards ( Figure 28.6 ). The central one third of the fovea, little more than 100 μm wide and known as the foveola , contains only midget cones. Two special anatomic features assist the foveal cones in general, and the midget cones in particular, in transducing the maximum amount of information concerning the form and colour values of an object under direct scrutiny. First, the more superficial layers of the retina lean outward from the centre, and their neurites are exceptionally long, with the result that the outer two thirds of the foveola are little overlapped by bipolar cell bodies and the inner third is not overlapped at all; light reflected from the object strikes the cones of the foveola without any diffraction. Second, one-to-one synaptic contact between the midget cones and midget bipolar neurons, and between these and midget ganglion cells enhance fidelity of central transmission. Outside the foveola, the amount of cone-to-bipolar-to-ganglion cell convergence increases progressively.
