Brain Research, 539 (1991) 337-341 Elsevier

337

BRES 24499

Cytochemical demonstration of aspartate aminotransferase activity in the rat retina Robert Gebhard Department of Anatomy, University of Warzburg, W-8700 Warzburg (Germany) (Accepted 16 October 1990) Key words: Aspartate aminotransferase; Cytochemistry; Glutamate; Rat retina

The mitochondrial (m-AAT) and the cytoplasmic (c-AAT) isoenzyme activities of the glutamate synthezising enzyme aspartate aminotransferase have been localized in the rat retina on the ultrastructurai level using enzyme histochemistry. Reaction product of c-AAT was found selectively in cone pedicles, in presynaptic terminals of a subpopulation of amacrine cells and of horizontal cell processes, which are connected to rods. Rod spherules, terminals of cone-related horizontal cells and of bipolar cells reacted negatively, as well as ganglion cells, nerve fibre layer and optic nerve, m-AAT reaction product was found in all neuronal structures, most densely in the photoreceptor inner segments. The localization of c-AAT activity is in accordance with its presumed meaning in the production of releasable glutamate. There is now a considerable amount of evidence coming from pharmacological physiological and biochemical experiments, that glutamate acts as a neurotransmitter in various neurons in the retinae of different species 7"9'23'24. There have been various attempts to identify the enzymes which are responsible for the production of transmitter glutamate in neuronal tissues and therefore could serve as markers for glutamatergic neurons. Altschuler et al. 2'3 suggested in guinea pig photoreceptors and cochlear nucleus the glutamate synthesizing enzyme aspartate aminotransferase (L-aspartate: 2-0xoglutarate aminotransferase, EC 2.6.1.1; AAT) to be such a marker, which exists in a mitochondrial (m-AAT) and a cytoplasmic (c-AAT) form 6. According to this hypothesis, neurons using glutamate as transmitter should have enriched amounts of c-AAT in presynaptic terminals because of their enhanced glutamate synthesis. On the other hand, m - A A T is correlated with functions of A A T in energy metabolism in the tricarboxylic cycle 27. The distribution of A A T in the rat retina has been studied biochemically in freeze-dried and homogenized retinal layers 25 and by immunocytochemistry at the light microscopic level 11A7. These studies reported presence of AAT in photoreceptor inner segments, in the outer plexiform layer, in cell bodies of the inner and outer nuclear layer, in the inner plexiform layer and in some cells of the ganglion cell layer, but with sometimes conflicting results. There is only one electron microscopic study on the distribution of AAT-like immunoreactivity

in the guinea pig and monkey retina 19. However, no information is available about the ultrastructural localisation of A A T in the rat retina, which is interesting because of the important species differences in retinal transmitters and our different methodological approach, which demonstrates the activity of the enzyme. Since very recently enzyme histochemical methods become increasingly important in order to study the local metabolic activity of enzymes involved in neurotransmitter metabolism 14'15, the aim of the present study is to describe the distribution of c-AAT and m-AAT-activity at the ultrastructure level in the rat retina with respect to its possible function in neurotransmitter metabolism. For this we used a lead salt procedure for a catalytic enzyme histochemical approach 21. The retinae of 12 adult Wistar rats of both sexes of our own colony were used for electron microscopy. The animals were kept in Macrolon cages at about 21 0(2 with light-dark changes every 12 h (light on 6.00 a.m.). For fixation, incubation and post-incubation treatment we used a procedure described recently (for details s e e Kugler13). Briefly, 5 animals were transcardially perfused with a fixative containing 1% paraformaldehyde and 1% glutaraldehyde for 30 min in the morning. Within o n e hour the eyes were enucleated. Seven animals were decapitated without prior perfusion and the eyes were enuclated rapidly. Cornea, iris and lens were removed carefully and the remaining eyecups were placed for 1 h in the same fixative as described before. Retinae which

Correspondence: R. Gebhard. Present address: Department of Anatomy, University of Cologne, Joseph-Stelzmann-Strasse 9, W-5000 Cologne 41, Germany. 0006-8993/91/$03.50 © 1991 Elsevier Science Publishers B.V. (Biomedical Division)

338 were fixed by perfusion or immersion didn't show different results. In both cases pieces of the central retina (3 × 3 mm) with the adherent sclera were dissected out under a stereo-microscope. After washing, the tissue was embedded in gelatine, 40-am-thick Vibratome sections were cut and washed again before incubation. The incubation medium consisted in principle of 20 mM aspartic acid (Serva, Heidelberg, Germany), 4 mM a-ketoglutarate, 6 mM lead nitrate. As controls either a-ketoglutarate was omitted, or 20 mM Glu was used instead of Asp. After a 45 rain incubation at 30 °C the sections were washed twice, treated for 60 s with 0.25% yellow ammonium sulfide (Merck, Darmstadt, Germany). The sections were rinsed again and osmified for 1 h at room temperature with 1% OsO 4. After overnight rinsing at 4 °C, the sections were dehydrated in a graded series of ethanols and embedded in Durcupan AMC (Fluka). Semithin sections (1/~m) were examined both unstained and stained with pyronin G and toluidine. Ultrathin sections were examined unstained using a Zeiss EM-109 electron microscope. The observed reaction product was related to the cytoplasmic and the mitochondrial isoenzyme based on its location, because the procedure described above covers the activity of both molecules. At the light microscopic level (Figs. 1-3) the distribution of AAT reaction product shows a characteristic pattern. Pronounced activities can be observed in the inner segments of the photoreceptors, some photoreceptor perikarya, the outer plexiform layer, in some perikarya of the inner and outer row of the inner nuclear layer, and stratified in the inner plexiform layer. The ganglion cell layer shows occasionally some positive perikarya. The nerve fibre layerand the optic nerve show a rather week dot-like reaction pattern. In the control

sections no reaction product could be observed, neither in the light (Fig. 2) nor in the electron microscope (not shown). At the electron microscopic level only the mitoehondrial isoenzyme of AAT could be observed in the pigmented epithelium, which was localized exclusively in the matrix of the mitochondria (Fig. 4). The outer segments of photoreceptors didn't show any detectable amount of AAT-reaction product (Fig. 5). In the inner segments, which are very rich of mitochondria, pronounced activities could be observed in the mitochondrial matrix (Fig. 5). In some cells of the outer nuclear layer reaction product was found in the cytoplasm of perikarya sending out positive reacting processes towards the inner segments and the outer plexiform layer. Those cells were distributed preferentially at the outer side of this layer. All terminals which could be identified as cone pedicles showed high activities of c-AAT (Fig. 6). Synaptic vesicles were free of reaction product. The elements making synaptic contact to cone pedicles (horizontal and bipolar cell processes) were always found to be negative (Figs. 6 and 7). In rod spherules no c-AAT could be detected (Figs. 8 and 9), although there were a few, which contained rather low levels (Fig. 10). The lateral elements of the characteristic ribbon synapses of rod spherules, which are known to be horizontal cell processes containing synaptic vesicles, always showed a strong cytoplasmic reaction, whereas the medial element (bipolar cell dendrite) regularly reacted negatively (Figs. 8-10). Additionally, most horizontal cells showed a strong cytoplasmic reaction in their perikarya (laying in the outermost row of the inner nuclear layer, large round nucleus) and in the cytoplasm of their processes, which were observed in the innermost region of the outer

Figs. 1-3. Semithin sections (1/~m) of the central part of the rat retina. Bars = 10/~m. Fig. 1. Section of an untreated retina stained with Toluidine blue-pyronin G. PE, pigmented epithelium; OS, outer segments; IS, inner segments; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer; NFL, nerve fibre layer. Fig. 2. Control section incubated without a-ketoglutarate. No reaction product is visible. Fig. 3. Histochemical demonstration of AAT-activity at the light microscopic level. Reaction product can be observed in the inner segments of the photoreceptors, a few perikarya in the outer nuclear layer (arrows), the outer plexiform layer, in perikarya of the inner nuclear layer preferently at the outer and inner row, in the inner plexiform layer and in some cells of the ganglion cell layer (arrowheads) and a few punctata in the nerve fibre layer. Figs. 4-13. Ultracytochemical demonstration of AAT-activity. All sections are unstained. Bars = 0.5 /~m. Fig. 4. Pigmented epithelium. Reaction product is only seen in the mitochondrial matrix. B, Bruch's membrane; N, nucleus. Fig. 5. Photoreceptors. Reaction product is localized in the mitochondrial matrix of the inner segments (IS). Outer segments (OS) are free of reaction product. Fig. 6. Cone pedicle. Reaction product is found in the cytoplasm, but not in synaptic vesicles and postsynaptic elements. Arrowheads, synaptic ribbons. Fig. 7. Detail of another cone pedicle. The postsynaptic elements are free of reaction product. Arrowheads, synaptic ribbons. Fig. 8. Rod spherule. Only the horizontal cell processes (HC) contain reaction product in their cytoplasm. The photoreceptor terminal is free of reaction product. Arrowheads, synaptic ribbon; x, protrusion of photoreceptor cytoplasm. Fig. 9. Horizontal section through a rod spherute. Only the horizontal cell processes (HC) contain reaction product. Arrowheads, synaptic ribbons, x, protrusion of photoreceptor cytoplasm. Fig. 10. Rod spherule. AAT-activity in this rod tei'minal is just above detection limit and little reaction product is visible in the cytoplasm. The bipolar cell process (BC) does not show reactivity, whereas both horizontal cell processes (HC) show strong activity in their cytoplasm. Arrowheads, synaptic ribbon. Fig. 11. Amacrine 'cell process in the middle third of the inner plexiform layer contains reaction product in the cytoplasm and the mitochondrial cristae, makes a contact to a negative profile (x). The Mueller gila cell (MC) does not show any reaction product, neither in the cytoplasm nor in the mitochondria (arrows). Inset: mitochondrium of another positive amacrine process. Reaction product is localized clearly in the cristae. Fig. 12. Amacrine cell process in the inner third of the inner plexiform layer, which contains reaction product in the cytoplasm and mitochondria, makes a synapse (arrowheads) to a negative profile. M, mitochondria. Fig. 13. Nerve fibre layer. Ganglion cell axons contain reaction product only in the mitochondrial matrix, but not in the cytoplasm.

339

340 plexiform layer extending towards the photoreceptor terminals. Some of the amacrine cell bodies (laying in the innermost row of this layer, invaginated nucleus, sending processes towards the inner plexiform layer) showed strong reaction of cytoplasmic AAT, some didn't show any reaction product. On rare occasions cells in the position of bipolar cells showed a cytoplasmic reaction in their perikarya. In the inner plexiform layer the cytoplasmic isoenzyme of AAT is exclusively restricted to a subpopulation of amacrine cells. The processes of some amacrine cells, which are distributed throughout the inner plexiform layer forming conventional synapses, showed strong cytoplasmatic reaction except in the synaptic vesicles (Figs. 11 and 12). The mitochondria which were found in the positive fibers contained reaction product only in the cristae (inset of Fig. 11). All other negatively reacting fibers contained the mitochondrial reaction product in the matrix, but also negatively reacting mitochondria could be observed occasionally. Interestingly, Mueller glia cells never contained any reaction product, neither from the cytoplasmic, nor from the mitochondrial isoenzyme (Fig. 11). In no case c-AAT-positive bipolar terminals, which form ribbon synapses, were found, but bipolar cells contained m-AAT in the mitochondrial matrix. Ganglion cells contain m-AAT in the matrix, but no c-AAT activity could be observed. The few cells in the ganglion cell layer, which showed positive reaction in the light microscope, could be identified as so called 'displaced amacrine cells' by ultrastructure characteristics (no processes towards the nerve fiber layer but towards the inner plexiform layer, which contains synaptic vesicles). The ganglion cell axons contained AAT activity only in the matrix of most of their mitochondria (Fig. 13). The present study reveals at the ultrastructural level that the activity of the enzyme aspartate aminotransferase is localized in mitochondria as well as in the cytoplasm of the rat retina in correspondence with biochemical results from other tissues 6. The two isoenzymes are present in different types of neurons, but not in Mueller glia cells. The mitochondrial isoenzyme has an ubiquitous distribution throughout the retina and can be found in all types of neurons and the pigmented epithelium. In the inner segments of photoreceptors, where energy metabolism is of great importance and neurotransmission does not occur, m-AAT is most prominent compared to all other retinal layers. So the present results are in agreement with the suggested meaning of m-AAT for energy metabolism aT. In contrast to the mitochondrial isoenzyme the reaction product of c-AAT shows a heterogenous distribution, clearly marking subpopulations of neurons. In the light microscope c-AAT-like immunoreactivity was re-

ported in the outer plexiform layer of the rat retina 11'17. At the electron microscopic level our results show a nearly exclusive localization of c-AAT reaction product in the cone pedicles, whereas rod spherules appear to be negative or just above detection limit. There is good evidence coming from physiological and pharmacological experiments, that glutamate acts as a neurotransmitter in photoreceptors of several species ~2'16'~°'~8. Brandon and Lam 5 demonstrated high-affinity uptake sites for Lglutamate in human and probably also in rat cones. It is reasonable to conclude, that the enhanced activity of c-AAT is due to the fact that rat cones may use glutamate as neurotransmitter and therefore need a high level of glutamate synthesis. In the lateral elements of the rod ribbon synapses c-AAT reaction product was found, whereas the lateral elements of cone pedicle ribbon synapses were negative. These lateral elements are known to be horizontal cell processes s. This is the first time that a clear enzymatic difference in horizontal cell processes connected to rods and cones is demonstrated, because Mosinger and Altschuler 19 did not describe AAT-like-immunoreactivity in any horizontal cell of guinea pig or monkey. At present the transmitters of rat horizontal cells are not known. Though there is good evidence that G A B A serves as transmitter in some types of horizontal cells, especially in lower vertebrates 3°, this is very unlikely for rat horizontal ceils, because it could be shown at the uttrastructural level, that rat horizontal cells do not contain glutamate decarboxylase immunoreactivity TM and G A B A concentration in the outer plexiform layer and the outer part of the inner nuclear layer is very low compared to other retinal layers 26. Finally, a subpopulation of amacrine and displaced amacrine cells showed c-AAT activity in their perikarya and terminals. Indeed there is growing evidence for a role of excitatory amino acids for neurotransmission in the inner plexiform layer 1"4'1°. Uptake of glutamate has been demonstrated in amacrine cells of rabbit retina 22 and physiological evidence has been provided for aspartate as transmitter in a class of amacrine cells in the mudpuppy 29. c-AAT reaction product clearly labels a subpopulation of neurons in the rat retina, i.e. classes of horizontal and amacrine ceils and cone pedicles. Its localization in presynaptic terminals of these cells is consistent with its suggested meaning for the production of releasable glutamate. Because glutamate is t h e precursor in the synthesis of GABA, which is also known to be a retinal transmitter, one might speculate that AAT is associated with GABAergic rather than w i ~ glutamaterglc pathways. But this does not hold true, since the distribution pattern of the GABA synthesizing enzyme L-glutamate

341 decarboxylase and c - A A T are completely different in rat retina 17. A d d i t o n a l l y c - A A T is found at least in some neurons, which might be glutamatergic by physiological evidence as discussed above. In conclusion, o u r study confirms the role of presynaptic localized c - A A T as a 1 Aizenman, E., Frosch, M.P. and Lipton, S.A., Responses mediated by excitatory amino acid receptors in solitary retinal ganglion cells from rat, J. Physiol., 369 (1988) 75-92. 2 Altschuler, R.A., Mosinger, J.L., Harmisom, G.G., Parakkal, M.H. and Wenthold, R.J., Aspartate aminotransferase-like immunoreactivity as a marker for aspartate/glutamate in guinea pig photoreceptors, Nature, 298 (1982) 657-659. 3 Altschuler, R.A., Neises, G.R., Harmison, G.G., Wenthoid, R.J. and Fex, J., Immunocytochemical localisation of aspartate aminotransferase immunoreactivity in the cochlear nucleus of the guinea pig, Proc. Natl. Acad. Sci. U.S.A., 78 (1981) 6553-6557. 4 Bloomfield, S.A. and Dowling, J.E., Roles of aspartate and glutamate in synaptic transmission in rabbit retina. II. Inner plexiform layer, J. Neurophysiol., 53 (1985) 714-725. 5 Brandon, C. and Lam, M.-K., L-Glutamic acid: a neurotransmitter candidate for cone photoreceptors in human and rat retinas, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 5117-5121. 6 Cooper, A.J., Glutamate-aspartate transaminase. In A. Meister (Ed.), Methods of Enzymology, Vol. 113, Glutamate, Glutamine, Glutathione and Related Compounds, Academic Press, Orlando, FL, 1985, pp. 66-69. 7 Dow, N.W., Brunken, W.J. and Parkinson, D., The function of synaptic transmitters in the retina, Annu. Rev. Neurosci., 12 (1989) 205-225. 8 Dowling, J.E., The Retina. An Approchable Part of the Brain, Belknap Press, Cambridge, MA, 1987. 9 Ehinger, B., Neurotransmitter systems in the retina, Retina, 2 (1982) 305-321. 10 Ikeda, H. and Sheardown, M.J., Aspartate may be an excitatory transmitter mediating visual excitation of 'sustained' but not of 'transient' cells in the cat retina: iontophoretic studies in vivo, Neuroscience, 7 (1982) 25-36. 11 Inangaki, N., Kamisaki, Y., Kiyama, H., Horio, Y., Tohyama, M. and Wada, H., Immunocytochemieal localizations of cytosolic and mitochondrial glutamic oxaloacetic transaminase isoenzymes in rat retina as markers for the glutamate-aspartate neuronal system, Brain Research, 325 (1985) 336-339. 12 Ishida, A.T. and Fain, G.L., D-Aspartate potentiates the effects of L-glutamate on horizontal cells in goldfish retina, Proc. Natl. Acad. Sci. U.S.A., 78 (1981) 5890-5894. 13 Kugler, P., Cytochemical demonstration of aspartate aminotransferase in the mossy-fibre system of the rat hippocampus, Histochemistry, 87 (1987) 623-625. 14 Kugler, P., The enzyme histoehemistry of neurotransmitter metabolism, Adv. Ant. Embryol. Cell. Biol., 111 (1988) 40-60. 15 Kugler, P., Localisation of transmitter-metabolizing enzymes by enzyme histochemistry in the rat hippocampus. In V. Chan-Palay and S.L. Palay (Eds.), The Hippocampus -- New Vistas, Neurol. Neurobiol., Alan R. Liss, New York, 1989, pp. 119-130.

m a r k e r of glutamatergic neurons good reason to speculate that glutamate as neurotransmitter. results of the few ultrastructural able so far 13'19.

in the retina and gives those cells might use This supports similar studies on A A T avail-

16 Lasater, E.M. and Dowling, J.E., Carp horizontal cells in culture respond selectively to L-glutamate and its agonists, Proc. Natl. Acad. Sci. U.S.A., 79 (1982) 936-940. 17 Lin, C.-T., Li, H.-Z. and Wu, J.-Y., Immunocytochemical localization of L-glutamate decarboxylase, gamma-aminobutyric acid transaminase, cysteine sulfinic acid decarboxylase, aspartate aminotransferase and somatostatin in rat retina, Brain Research, 270 (1983) 273-283. 18 Lin, C.-T., Song, G.-X. and Wu, J.-Y., Ultrastructural demonstration of L-glutamate deearboxylase and cysteine sulfinic acid decarboxylase in rat retina by immunocytochemistry, Brain Research, 331 (1985) 71-80. 19 Mosinger, J.L. and Altschuler, R.A., Aspartate aminotransferase-like immunoreactivity in the guinea pig and monkey retinas, J. Comp. Neurol., 233 (1985) 255-268. 20 Murakami, M., Ohtsuka, T. and Shimazaki, H., Effects of aspartate and glutamate on the bipolar cells in the carp retina, Vision Res., 15 (1975) 456--458. 21 Papadimitriou, J.M. and van Duijn, P., The ultrastructural localisation of the isoenzymes of aspartate aminotransferase in murine tissue, J. Cell. Biol., 47 (1970) 84-98. 22 Redburne, D.A., G A B A and glutamate as retina neurotransmitters in rabbit retina. In G. di Chiara and G.L. Gessa (Eds.), Glutamate as Neurotransmitter, Raven Press, New York, 1981, pp. 79-89. 23 Redburne, D.A., Neurotransmitter systems in the outer plexiform layer of mammalian retina, Neurosci. Res., Suppl. 8 (1988) S127-S136. 24 Reif-Lehrer, L., Glutamate metabolism in the retina: a larger perspective. In J.A. Zadunaisky and H. Davson (Eds.), Current Topics in Eye Research, Vol. 4, Academic Press, Orlando, FL, 1984, pp. 1-95. 25 Ross, C.D., Bowers, M. and Godfrey, D.A., Distributions of the activities of aspartate aminotransferase isoenzymes in rat retinal layers, Neurosci. Len., 74 (1987) 205-210. 26 Ross, C.D., Parli, J.A. and Godfrey, D.A., Quantitative distribution of six amino acids in rat retinal layers, Vision Res., 29 (1989) 1079-1084. 27 Salganicoff, L. and de Robertis, E., Subcellular distribution of the enzymes of the glutamic acid, glutamine and y-amino-butyric acid cycles in the rat brain, J. Neurochem., 12 (1965) 287-309. 28 Shiells, R.A., Falk, G. and Naghshineh, S., Action of glutamate and aspartate analogues on rod horizontal and bipolar cells, Nature, 294 (1981) 592-594. 29 Slaughter, M.M. and Miller, R.E, The role of excitatory amino acid transmitters in the mudpuppy retina: an analysis with kainic acid and N-methyl aspartate, J. Neurosci., 3 (1983) 1701-1711. 30 Yazulla, S., GABAergic mechanisms in the retina, Prog. Ret. Res., 5 (1986) 1-52.

Cytochemical demonstration of aspartate aminotransferase activity in the rat retina.

The mitochondrial (m-AAT) and the cytoplasmic (c-AAT) isoenzyme activities of the glutamate synthesizing enzyme aspartate aminotransferase have been l...
3MB Sizes 0 Downloads 0 Views