Eaat4 Eaat5

Cystine-glutamate exchanger

Glutamate-aspartate transporter Müller cells and astrocytes (GLAST)

Glutamate transporter 1 (GLT1) Bipolar cells

Excitatory amino acid carrier 1 Amacrine and horizontal cells (EAAC1)

- Astrocytes

Fig. 3. A vertical section of an adult rat retina incubated in D-aspartate, fixed and immu-nolabeled for D-aspartate. D-aspartate labeling is seen labeling Müller cell somata, and processes. Abbreviations as in Fig. 1. Scale bar = 20 ||M.

sodium into the cell, while potassium is transported out. In addition to the transport of glutamate, these proteins also function as channels for anions, a property that is particular prominent in EAAT4 and EAAT5 (16).

All five EAATs are expressed within the retina. GLAST is localized within Müller cells and astrocytes (see Fig. 4) (22,23), whilst GLT1, EAAC1 and EAAT5 are all localized within specific neuronal classes including bipolar (GLT1), horizontal (EAAC1), amacrine cells (EAAC1), and photoreceptors (EAAT5, GLT1) (22-26). The neuronal expression of EAAT2 is shown in Fig. 5. Immunolabeling of two splice variants of GLT1, GLT1A, and GLTB, shows labeling of bipolar cells and processes in the inner plexiform layer. EAAT5 is also shown labeling cones and bipolar cell processes in the rat retina. It has been suggested that these neuronal glutamate transporters are important for maintaining the transmitter pools within presynaptic terminals. For example, EAAC1 is localized within GABAergic neurons in the retina and brain, and is thought to maintain the levels glutamate in these terminals so that formation of GABA can occur rapidly (27,28).

EAAT4 was originally localized within Purkinje cells of the cerebellum (21). More recently, EAAT4 has been identified in astrocytes from the spinal cord (29) and retina (see Fig. 4) (30). This implies that astrocytes within the retina express two glutamate transporters, GLAST and EAAT4. These glutamate transporters display different affinities for glutamate, and distinct chloride permeabilities. It is possible that expres-

Fig. 4. Vertical sections of control, (A) and (C), and diabetic, (B) and (D), Sprague-Dawley rat retinas labeled for the glutamate-aspartate transporter (GLAST) (A, B) and excitatory amino acid transporter 4 (EAAT4) (C, D). GLAST is highly expressed by Müller cells and astrocytes in control and diabetic retina. No difference in expression level of GLAST can be noted in the diabetic rat retina, compared with the control. EAAT4 is expressed by astrocytes in the rat retina, which are located within the nerve fibre layer. No difference in expression of EAAT4 could be detected. Abbreviations as in Fig. 1 Scale bar = 50 |M.

Fig. 4. Vertical sections of control, (A) and (C), and diabetic, (B) and (D), Sprague-Dawley rat retinas labeled for the glutamate-aspartate transporter (GLAST) (A, B) and excitatory amino acid transporter 4 (EAAT4) (C, D). GLAST is highly expressed by Müller cells and astrocytes in control and diabetic retina. No difference in expression level of GLAST can be noted in the diabetic rat retina, compared with the control. EAAT4 is expressed by astrocytes in the rat retina, which are located within the nerve fibre layer. No difference in expression of EAAT4 could be detected. Abbreviations as in Fig. 1 Scale bar = 50 |M.

sion of two glutamate transporters by the same cell occurs because of the different functions of these transporters.

Although the function of specific transporter types is contentious, it is clear that GLAST is the predominant transporter for removal of glutamate within the retina and is localized in Müller cells and astrocytes (22,23). Fig. 3 illustrates an experiment that reveals the importance of Müller cells in the removal of glutamate within the retina. D-aspartate is a nonhydrolyzable glutamate analog that is transported by all five EAATs with similar efficiency. Because D-aspartate is an amino acid, it can be fixed within tissue using conventional aldehyde fixatives and then immunocytochemically detected. Fig. 2 shows a vertical section of rat retina that has been incubated with D-aspartate, and then labeled with an antibody directed against D-aspartate. D-aspartate heavily labels Müller-cell somata and their processes. Although a variety of EAATs are known to be expressed by neurons, antibodies to D-aspartate do not label these other neurons, suggesting that Müller cells are the primary site for the removal of glutamate within the retina.

Fig. 5. Vertical sections of control (A, C, E) and diabetic (B, D, F) Sprague-Dawley rat retinae labeled for glutamate transporter (GLT)A (A, B), GLT1B (C, D) and excitatory amino acid transporter 5 (EAAT5) (E, F). GLT1A labels two prominent bands within the IPL. GLT1B, a different splice variant of GLT1, labels cone bipolar cells and their processes, as well as cones. EAAT5 labels cones on the rat retina. No differences are noted between control and diabetic retinas for GLT1A or GLT1B. However, EAAT5 did show more-prominent labeling within the OPL in the diabetic rat retina. Abbreviations as in Fig. 1. Scale bar = 50 |M.

Fig. 5. Vertical sections of control (A, C, E) and diabetic (B, D, F) Sprague-Dawley rat retinae labeled for glutamate transporter (GLT)A (A, B), GLT1B (C, D) and excitatory amino acid transporter 5 (EAAT5) (E, F). GLT1A labels two prominent bands within the IPL. GLT1B, a different splice variant of GLT1, labels cone bipolar cells and their processes, as well as cones. EAAT5 labels cones on the rat retina. No differences are noted between control and diabetic retinas for GLT1A or GLT1B. However, EAAT5 did show more-prominent labeling within the OPL in the diabetic rat retina. Abbreviations as in Fig. 1. Scale bar = 50 |M.

Several studies have shown that GLAST is the most important glutamate transporter in the retina accounting for around 50% of all glutamate uptake (23,31,32). Moreover, in mice where the gene for GLAST has been knocked out, visual function, as measured by the electroretinogram, is substantially affected, although not completely lost (31). Therefore, glutamate uptake by Müller cells is central to maintaining the normal function of the retina. However, it should also be noted that GLAST alone does not account for the total removal of glutamate in the retina.

The five EAATs constitute the high-affinity glutamate uptake systems in the retina, but other proteins present within the retina can also transport glutamate in a sodium-independent manner. Recently, a cystine-glutamate exchanger has been identified in the retina (33). This transporter is an exchanger with 1:1 stoichiometry, transporting cystine into a cell in exchange for glutamate. It is electroneutral, and uses the transmembrane gradient of glutamate as the driving force. When the extracellular level of glutamate is elevated, this exchanger can release cystine whilst taking up glutamate.

In summary, glutamate uptake into retinal Müller cells and astrocytes is dependent on the activity of GLAST and/or EAAT4. In the next section we will consider the evidence that glutamate transport is abnormal during diabetes.

0 0

Post a comment