- Email: [email protected]
Cellular localization of N-acetylaspartylglutamate in amphibian retina and spinal sensory ganglia
Brain Research, 406 (1987) 397-401 Elsevier
397
BRE 22114
Cellular localization of N-acetylaspartylglutamate in amphibian retina and spinal sensory ganglia Michele M. Kowalski, Martha Cassidy, M.A.A. Namboodiri and Joseph H. Neale Department of Biology, Georgetown University, Washington, DC 20057 (U. S. A.) (Accepted 18 November 1986) Key words: N-Acetylaspartylglutamate; Immunohistochemistry; Neuropeptide; Retina; Spinal sensory neuron; Amphibian
Antisera were produced against N-acetylaspartylglutamate (NAAG) and were used to localize the molecule within the retina and spinal sensory ganglia of Rana pipiens. NAAG immunoreactivity (IR) in the retina was confined to a subpopulation of amacrine and bipolar cells. The dipeptide was present in the perikarya of these cells and their neurites which terminated in two discrete bands of the inner plexiform layer. Some NAAG-IR was also present in the outer plexiform layer and the inner segment layer. In spinal ganglia, a subpopulation of relatively large sensory neuron cell bodies expressed NAAG-IR. These data are consistent with the hypothesis that this dipeptide has a function which is specific to discrete subclasses of neurons. In the amphibian retina, the NAAG distribution can be related to the reported involvement of the N-methyl-D-aspartate receptor in neurotransmission at the level of amacrine and ganglion cells. The N-acetylated acidic dipeptide, N-acetylaspartylglutamate ( N A A G ) , is present specifically and in a relatively high concentration in the nervous systems of vertebrate and invertebrate organisms 6,8,1°,11. Physiological studies in the prepyriform cortex suggest that N A A G may act as an excitatory neurotransmitter upon release from mitral cell axon endings 7. Consistent with this hypothesis, we recently reported the immunohistochemical localization of N A A G in mitral cells 1. We have also described a subpopulation of spinal sensory neurons which contains immunoreactive (IR) N A A G and have demonstrated synthesis of the dipeptide by these cells 4,5. In related work, N A A G was found to activate specifically the N-methyl-D-aspartate ( N M D A ) class of glutamate receptors on spinal cord neurons, although it was considerably less potent than glutamate on these sensory neuron target cells 13. While there is evidence which is consistent with a role for N A A G in neurotransmission in the prepyriform cortex 7, hippocampus 3 and spinal cord 13, its function at identified synapses remains to be established rigorously. In a recent study of the rat visual system, we identi-
fied N A A G within retinal ganglion cells and their neurite fields in the brain using both immunohistochemistry and direct isolation of the dipeptide 1,2. Given the importance of confirming the role of N A A G in retinal physiology and the extensive use of the amphibian as a model in studies of the development and function of the visual system, we have assayed the distribution of N A A G in the frog retina. This amphibian is reported to have relatively high levels of the dipeptide in its nervous system s,l°. O u r results indicate that some amphibian amacrine and bipolar neurons possess N A A G , while the retinal ganglion cells do not. The highly ordered nature of the synaptic connections of amacrine and bipolar cells in the frog retina recommends these cells and their targets for assay of the role of the dipeptide in neuronal communication. N A A G was synthesized by acetylation of aspartylglutamate 13 (Bachem) and two New Zealand rabbits were immunized with N A A G coupled to thyroglobulin via carbodiimide as previously described 5. The antisera had several thousand-fold greater affinity for soluble N A A G than for N-acetylaspartate, as-
Correspondence: J.H. Neale, Department of Biology, Georgetown University, Washington, DC 20057, U.S.A. 0006-8993/87/$03.50 ~ 1987 Elsevier Science Publishers B.V. (Biomedical Division)
398 partylglutamate, aspartate or glutamate as determined by competitive enzyme-linked immunosorbant assay ( E L I S A ) against N A A G coupled to brain protein 5. Affinity purified antibodies were prepared by passing the antiserum over Affi-Gel 102 (BioRad) to which N A A G had been coupled using carbodiimide. The antibody was eluted using salt or acidic conditions, neutralized and assayed by E L I S A for its affinity to N A A G , N-acetylaspartate, aspartylglutamate, aspartate and glutamate, each coupled to bovine serum albumin with carbodiimide. The antisera and purified antibody preparation were similar in that they bound to the aspartylglutamate, aspartate and glutamate conjugates at levels which were similar to the binding to carbodiimide-treated bovine serum albumin alone which served as the control for these assays. Binding to the N-acetylaspartate conjugate was above this background level but was less than 10% of the level obtained against the N A A G conjugate in these solid phase immunoassays. For immunohistochemistry, 5-inch grass frogs (Carolina Biological) were anesthetized prior to transcardial perfusion with heparinized saline followed by 3% carbodiimide (1-ethyl-3-(3-diaminethylaminopropyl)carbodiimide, Sigma E 7750) with 4% freshly depolymerized paraformaldehyde in 50 mM phosphate buffer (pH 7.4). Retinas and lumbar spinal sensory ganglia were excised and fixed overnight in the same mixture. Tissues were then soaked for several days in phosphate-buffered saline with 25% sucrose. Tissue was frozen a t - 1 0 0 °C and sectioned at - 2 0 °C using a Hacker-Bright cryostat. Sections (20/~m thick) were incubated for 3 h at 23 °C with 1:250 dilution of the antiserum or an equivalent dilution of the affinity purified antibody and subsequently incubated with a 1:40 dilution of FITC-conjugated anti-rabbit serum (Jackson ImmunoResearch). The a n t i - N A A G sera revealed dipeptide I R within cell bodies and neurites of a subpopulation of amacrine and bipolar neurons in the frog retina (Fig. 1). The most intense staining was observed in two very distinct bands within the inner plexiform layer. Some cell bodies in the inner nuclear layer were intensely reactive. Other cells in this layer exhibited a moderate level of staining, while the majority of the cell bodies were unreactive. It was not possible to classify every reactive cell without equivocation. However, the IR of most highly reactive cells could be followed
Gel
iP
I
,.
IPl"lsl
os
Fig. 1. NAAG-IR is indicated by the bright fluorescence detected within amacrine and bipolar cell bodies observed within the inner nuclear layer (IN) of the frog retina. Additional specific IR is found throughout the inner plexiform layer (IP) and is concentrated in two discrete synaptic bands. Less intense, specific fluorescence was also present in the thin outer plexiform layer (P) and inner segmental layer (S). While endogenous fluorescence was sometimes detected in the rod and cone cells of the outer segmental layer (OS), NAAG-specific IR was not observed in this region. A few cell bodies in the ganglion cell layer (GC) were reactive with the anti-NAAG serum. It was not possible to unequivocally define these as ganglion cells or displaced amacrine cells. Most of the perikarya in the ganglion cell layer which were identified in these sections using phase optics were clearly unreactive. Bar = 501~m.
399 through their neurites at different optical planes and, based upon this morphology, they appeared to be either amacrine or bipolar. Some bipolar neurites were traced directly to the most proximal fluorescent band in the inner plexiform layer and others appeared to extend to the outer plexiform and inner
segment layer. Amacrine cell neurites could be traced into both fluorescent bands in the inner plexiform layer. Occasionally, and intensely fluorescent cell body was present in the ganglion cell layer with neurites which extended into the inner plexiform layer. While it is not possible to rule out the possibility
Fig. 2. Several spinal sensory neurons exhibit NAAG-IR in this section from a lumbar ganglion of the frog. Typically, the IR sensory cells were among the largest in the ganglia, while unreactive perikarya had smaller diameters. This is consistent with the size distribution observed in rat sensory ganglia. Bar --- 50/~m.
400 that these were ganglion cells, they appeared structurally similar to displaced amacrine cells. In sections where the outer fiber layer was intact, there was no evidence of immunoreactive ganglion cell neurites. Our observations of N A A G - I R in frog amacrine and bipolar cells, but not in retinal ganglion cells, contrasts with our data in the rat visual system. If, as its differential cellular distribution suggests, N A A G is serving a neurotransmitter-related function in the retina, the communication processes appear to be different between these two species. The substantial difference in concentration of N A A G between nervous systems of vertebrate species, such as Rana pipiens and Carrasius auratus 8, also suggests variable participation of the dipeptide in neuronal function. The specificity of the immunohistochemical reaction was assayed in two ways. First, IR in the diluted antiserum was eliminated with 0.0012 mg/ml of a carbodiimide-coupled NAAG-bovine serum albumin complex (50 pmol/ml of N A A G and 3 mg/ml of albumin reacted for 12 h at 23 °C in the presence of 6 mg/ml of the carbodiimide), while 0.0005 mg/ml did not block the IR. In contrast, similar conjugates prepared with glutamate or aspartate failed to decrease the IR at the highest concentration tested, 0.5 mg/ml. This concentration of aspartylglutamate conjugate also failed to block the reactivity, although the signal was reduced. The N-acetylaspartate conjugate blocked the IR at a concentration of 0.05 mg/ml and diminished the signal at 0.005 mg/ml. Additionally, we have determined in other studies that the IR was blocked when antiserum was incubated with 50 ~M NAAG. The reactivity was not eliminated by incubation with 10 mM N-acetylaspartate, gamma-glutamylglutamate, aspartylglutamate, glutamate or aspartate 5. In order to further explore the possible presence of N A A G within amphibian retinal ganglion cells or their projections, tissue sections of the frog optic tecturn were examined. Although there was some N A A G - I R present in cells within the tectum, there was no evidence of specifically elevated IR in the afferent fiber layers. The pattern of IR in the left and right tectum was similar in two animals which had the right optic nerve transected 16 days prior to fixation. Thus, in contrast to the rat we could not detect N A A G within retinal ganglion cell afferents in this
species. The N A A G - I R in the frog spinal sensory ganglia was similar to that observed in the rat. A subpopulation of relatively large neuronal cell bodies expressed NAAG-1R (Fig. 2). We previously demonstrated that this IR was not present when rat ganglia were fixed in paraformaldehyde alone or in glutaraldehyde, results which indicate that the antibody reacts with a small N-blocked peptide or amino acid 5. We have also found that inadequate perfusion of the nervous system with carbodiimide, prior to immersion fixation, reduced or eliminated the retinal IR with anti-NAAG serum. When we examined the response of spinal cord neurons to N A A G , we found that it specifically activated the N M D A class of receptors 13. While similar data are not available in the frog nervous system, there are several studies of mudpuppy retina which implicate the N M D A receptor in excitatory neurotransmission between functionally specific subtypes of amacrine cells and retinal ganglion cells. Specifically, comparison of the actions of kainic acid and Nmethyl aspartate indicated that a group of sustained 'On' amacrine cells released a neurotransmitter which activated the N M D A receptor subtype at synapses with other amacrine cells and with ganglion cells 12. Recently, it has been reported that 2-amino5-phosphonovalerate, and N M D A receptor antagonist, blocked the sustained responses in mudpuppy ganglion cells driven by the 'On' pathway but not the 'Off' pathway 9. These and other data suggest that some bipolar cells may release an N M D A receptor agonist, other than aspartate or glutamate, which acts on amacrine and ganglion cells 9. The N A A O which we have identified in frog retinal neurons is a candidate for this function. N A A G ' s role in the nervous system remains to be established. Localization of the dipeptide in discrete neuronal populations is consistent with its activity as a neurotransmitter. More significantly, these data identify appropriate cells and tissues for future studies of N A A G ' s metabolism and physiological effects which are necessary to define its function.
Research supported by a research grant, D A 02297, from the National Institute on Drug Abuse.
401 1 Anderson, K.J., Monaghan, D.T,, Cangro, C.B., Namboodiri, M.A.A., Neale, J.H. and Cotman, C.W., Localization of N-acetylaspartylglutamate-like immunoreactivity in selected areas of the rat brain, Neurosci. Lett., 72 (1986) 14-20. 2 Anderson, K.J., Borja, M., Cotman, C.W., Namboodiri, M.A.A. and Neale, J.H., N-Acetylaspartylglutamateidentified in rat retinal ganglion cells and their projections in the brain,, Brain Research, in press. 3 Bernstein, J., Fisher, R.S., Zaczek, R. and Coyle, J.T., Dipeptides of glutamate and aspartate may be neuroexcitants in the rat hippocampal slice, J. Neurosci., 5 (1985) 1429-1433. 4 Cangro, C.B., Garrison, D.E., Luongo, P.A., Truckenmiller, M.E., Namboodiri, M.A.A. and Neale, J.H., First immunohistochemical demonstration of N-acetylaspartylglutamate in specific neurons, Soc. Neurosci. Abstr., 11 (1985) 108. 5 Cangro, C.B., Namboodiri, M.A.A., Sklar, L.A. and Neale, J.H., Biosynthesis and immunohistochemistry of Nacetylaspartylglutamate in spinal sensory ganglia, J. Neurochem., submitted. 6 Curatolo, A., D'Arcangelo, P., Lino, A. and Brancati, A., Distribution of N-acetyl-aspartic and N-acetyl-aspartylglutamic acids in nervous tissue, J. Neurochem., 12 (1965) 339-342. 7 ffrench-Mullen, J.M.H., Koller, K., Zaczek, R., Coyle, J.T., Hod, N. and Carpenter, D.O., N-acetylaspartylgluta-
mate: possible role as the neurotransmitter of the lateral olfactory tract, Proc. Natl. Acad. Sci. U.S.A., 82 (1985) 3897-3900. 8 Koller, K.J., Antuono, P.G. and Coyle, J.T., The ontogeny and phylogeny of N-acetyl-aspartyl-glutamate, Soc. Neurosci. Abstr., 10 (1984) 226. 9 Lukasiewicz, P.D. and McReynolds, J.S., Synaptic transmission at N-methyl-D-aspartate receptors in the proximal retina of the mudpuppy, J. Physiol. (London), 367 (1985) 99-115. 10 Miyake, M., Kakimoto, Y. and Sorimachi, M., A gas chromatographic method for the determination of N-acetyl-Laspartic acid, N-acetyl-a-aspartylglutamic acid and fl-cytryl-L-glutamic acid and their distributions in the brain and other organs of various species of animals, J. Neurochem., 36 (1981) 804-810. 11 Miyamoto, E., Kakimoto, Y. and Sano, I., Identification of N-acetyl-a-aspartylglutamic acid in the bovine brain, J. Neurochem., 13 (1966) 999-1003. 12 Slaughter, M.M. and Miller, R.F., 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. 13 Westbrook, G.L., Mayer, M.L., Namboodiri, M.A.A. and Neale, J.H., High concentrations of N-acetylaspartylglutamate (NAAG) selectively activate NMDA receptors on mouse spinal cord neurons in cell culture, J. Neurosci., 6 (1986) 3385-3392.
397
BRE 22114
Cellular localization of N-acetylaspartylglutamate in amphibian retina and spinal sensory ganglia Michele M. Kowalski, Martha Cassidy, M.A.A. Namboodiri and Joseph H. Neale Department of Biology, Georgetown University, Washington, DC 20057 (U. S. A.) (Accepted 18 November 1986) Key words: N-Acetylaspartylglutamate; Immunohistochemistry; Neuropeptide; Retina; Spinal sensory neuron; Amphibian
Antisera were produced against N-acetylaspartylglutamate (NAAG) and were used to localize the molecule within the retina and spinal sensory ganglia of Rana pipiens. NAAG immunoreactivity (IR) in the retina was confined to a subpopulation of amacrine and bipolar cells. The dipeptide was present in the perikarya of these cells and their neurites which terminated in two discrete bands of the inner plexiform layer. Some NAAG-IR was also present in the outer plexiform layer and the inner segment layer. In spinal ganglia, a subpopulation of relatively large sensory neuron cell bodies expressed NAAG-IR. These data are consistent with the hypothesis that this dipeptide has a function which is specific to discrete subclasses of neurons. In the amphibian retina, the NAAG distribution can be related to the reported involvement of the N-methyl-D-aspartate receptor in neurotransmission at the level of amacrine and ganglion cells. The N-acetylated acidic dipeptide, N-acetylaspartylglutamate ( N A A G ) , is present specifically and in a relatively high concentration in the nervous systems of vertebrate and invertebrate organisms 6,8,1°,11. Physiological studies in the prepyriform cortex suggest that N A A G may act as an excitatory neurotransmitter upon release from mitral cell axon endings 7. Consistent with this hypothesis, we recently reported the immunohistochemical localization of N A A G in mitral cells 1. We have also described a subpopulation of spinal sensory neurons which contains immunoreactive (IR) N A A G and have demonstrated synthesis of the dipeptide by these cells 4,5. In related work, N A A G was found to activate specifically the N-methyl-D-aspartate ( N M D A ) class of glutamate receptors on spinal cord neurons, although it was considerably less potent than glutamate on these sensory neuron target cells 13. While there is evidence which is consistent with a role for N A A G in neurotransmission in the prepyriform cortex 7, hippocampus 3 and spinal cord 13, its function at identified synapses remains to be established rigorously. In a recent study of the rat visual system, we identi-
fied N A A G within retinal ganglion cells and their neurite fields in the brain using both immunohistochemistry and direct isolation of the dipeptide 1,2. Given the importance of confirming the role of N A A G in retinal physiology and the extensive use of the amphibian as a model in studies of the development and function of the visual system, we have assayed the distribution of N A A G in the frog retina. This amphibian is reported to have relatively high levels of the dipeptide in its nervous system s,l°. O u r results indicate that some amphibian amacrine and bipolar neurons possess N A A G , while the retinal ganglion cells do not. The highly ordered nature of the synaptic connections of amacrine and bipolar cells in the frog retina recommends these cells and their targets for assay of the role of the dipeptide in neuronal communication. N A A G was synthesized by acetylation of aspartylglutamate 13 (Bachem) and two New Zealand rabbits were immunized with N A A G coupled to thyroglobulin via carbodiimide as previously described 5. The antisera had several thousand-fold greater affinity for soluble N A A G than for N-acetylaspartate, as-
Correspondence: J.H. Neale, Department of Biology, Georgetown University, Washington, DC 20057, U.S.A. 0006-8993/87/$03.50 ~ 1987 Elsevier Science Publishers B.V. (Biomedical Division)
398 partylglutamate, aspartate or glutamate as determined by competitive enzyme-linked immunosorbant assay ( E L I S A ) against N A A G coupled to brain protein 5. Affinity purified antibodies were prepared by passing the antiserum over Affi-Gel 102 (BioRad) to which N A A G had been coupled using carbodiimide. The antibody was eluted using salt or acidic conditions, neutralized and assayed by E L I S A for its affinity to N A A G , N-acetylaspartate, aspartylglutamate, aspartate and glutamate, each coupled to bovine serum albumin with carbodiimide. The antisera and purified antibody preparation were similar in that they bound to the aspartylglutamate, aspartate and glutamate conjugates at levels which were similar to the binding to carbodiimide-treated bovine serum albumin alone which served as the control for these assays. Binding to the N-acetylaspartate conjugate was above this background level but was less than 10% of the level obtained against the N A A G conjugate in these solid phase immunoassays. For immunohistochemistry, 5-inch grass frogs (Carolina Biological) were anesthetized prior to transcardial perfusion with heparinized saline followed by 3% carbodiimide (1-ethyl-3-(3-diaminethylaminopropyl)carbodiimide, Sigma E 7750) with 4% freshly depolymerized paraformaldehyde in 50 mM phosphate buffer (pH 7.4). Retinas and lumbar spinal sensory ganglia were excised and fixed overnight in the same mixture. Tissues were then soaked for several days in phosphate-buffered saline with 25% sucrose. Tissue was frozen a t - 1 0 0 °C and sectioned at - 2 0 °C using a Hacker-Bright cryostat. Sections (20/~m thick) were incubated for 3 h at 23 °C with 1:250 dilution of the antiserum or an equivalent dilution of the affinity purified antibody and subsequently incubated with a 1:40 dilution of FITC-conjugated anti-rabbit serum (Jackson ImmunoResearch). The a n t i - N A A G sera revealed dipeptide I R within cell bodies and neurites of a subpopulation of amacrine and bipolar neurons in the frog retina (Fig. 1). The most intense staining was observed in two very distinct bands within the inner plexiform layer. Some cell bodies in the inner nuclear layer were intensely reactive. Other cells in this layer exhibited a moderate level of staining, while the majority of the cell bodies were unreactive. It was not possible to classify every reactive cell without equivocation. However, the IR of most highly reactive cells could be followed
Gel
iP
I
,.
IPl"lsl
os
Fig. 1. NAAG-IR is indicated by the bright fluorescence detected within amacrine and bipolar cell bodies observed within the inner nuclear layer (IN) of the frog retina. Additional specific IR is found throughout the inner plexiform layer (IP) and is concentrated in two discrete synaptic bands. Less intense, specific fluorescence was also present in the thin outer plexiform layer (P) and inner segmental layer (S). While endogenous fluorescence was sometimes detected in the rod and cone cells of the outer segmental layer (OS), NAAG-specific IR was not observed in this region. A few cell bodies in the ganglion cell layer (GC) were reactive with the anti-NAAG serum. It was not possible to unequivocally define these as ganglion cells or displaced amacrine cells. Most of the perikarya in the ganglion cell layer which were identified in these sections using phase optics were clearly unreactive. Bar = 501~m.
399 through their neurites at different optical planes and, based upon this morphology, they appeared to be either amacrine or bipolar. Some bipolar neurites were traced directly to the most proximal fluorescent band in the inner plexiform layer and others appeared to extend to the outer plexiform and inner
segment layer. Amacrine cell neurites could be traced into both fluorescent bands in the inner plexiform layer. Occasionally, and intensely fluorescent cell body was present in the ganglion cell layer with neurites which extended into the inner plexiform layer. While it is not possible to rule out the possibility
Fig. 2. Several spinal sensory neurons exhibit NAAG-IR in this section from a lumbar ganglion of the frog. Typically, the IR sensory cells were among the largest in the ganglia, while unreactive perikarya had smaller diameters. This is consistent with the size distribution observed in rat sensory ganglia. Bar --- 50/~m.
400 that these were ganglion cells, they appeared structurally similar to displaced amacrine cells. In sections where the outer fiber layer was intact, there was no evidence of immunoreactive ganglion cell neurites. Our observations of N A A G - I R in frog amacrine and bipolar cells, but not in retinal ganglion cells, contrasts with our data in the rat visual system. If, as its differential cellular distribution suggests, N A A G is serving a neurotransmitter-related function in the retina, the communication processes appear to be different between these two species. The substantial difference in concentration of N A A G between nervous systems of vertebrate species, such as Rana pipiens and Carrasius auratus 8, also suggests variable participation of the dipeptide in neuronal function. The specificity of the immunohistochemical reaction was assayed in two ways. First, IR in the diluted antiserum was eliminated with 0.0012 mg/ml of a carbodiimide-coupled NAAG-bovine serum albumin complex (50 pmol/ml of N A A G and 3 mg/ml of albumin reacted for 12 h at 23 °C in the presence of 6 mg/ml of the carbodiimide), while 0.0005 mg/ml did not block the IR. In contrast, similar conjugates prepared with glutamate or aspartate failed to decrease the IR at the highest concentration tested, 0.5 mg/ml. This concentration of aspartylglutamate conjugate also failed to block the reactivity, although the signal was reduced. The N-acetylaspartate conjugate blocked the IR at a concentration of 0.05 mg/ml and diminished the signal at 0.005 mg/ml. Additionally, we have determined in other studies that the IR was blocked when antiserum was incubated with 50 ~M NAAG. The reactivity was not eliminated by incubation with 10 mM N-acetylaspartate, gamma-glutamylglutamate, aspartylglutamate, glutamate or aspartate 5. In order to further explore the possible presence of N A A G within amphibian retinal ganglion cells or their projections, tissue sections of the frog optic tecturn were examined. Although there was some N A A G - I R present in cells within the tectum, there was no evidence of specifically elevated IR in the afferent fiber layers. The pattern of IR in the left and right tectum was similar in two animals which had the right optic nerve transected 16 days prior to fixation. Thus, in contrast to the rat we could not detect N A A G within retinal ganglion cell afferents in this
species. The N A A G - I R in the frog spinal sensory ganglia was similar to that observed in the rat. A subpopulation of relatively large neuronal cell bodies expressed NAAG-1R (Fig. 2). We previously demonstrated that this IR was not present when rat ganglia were fixed in paraformaldehyde alone or in glutaraldehyde, results which indicate that the antibody reacts with a small N-blocked peptide or amino acid 5. We have also found that inadequate perfusion of the nervous system with carbodiimide, prior to immersion fixation, reduced or eliminated the retinal IR with anti-NAAG serum. When we examined the response of spinal cord neurons to N A A G , we found that it specifically activated the N M D A class of receptors 13. While similar data are not available in the frog nervous system, there are several studies of mudpuppy retina which implicate the N M D A receptor in excitatory neurotransmission between functionally specific subtypes of amacrine cells and retinal ganglion cells. Specifically, comparison of the actions of kainic acid and Nmethyl aspartate indicated that a group of sustained 'On' amacrine cells released a neurotransmitter which activated the N M D A receptor subtype at synapses with other amacrine cells and with ganglion cells 12. Recently, it has been reported that 2-amino5-phosphonovalerate, and N M D A receptor antagonist, blocked the sustained responses in mudpuppy ganglion cells driven by the 'On' pathway but not the 'Off' pathway 9. These and other data suggest that some bipolar cells may release an N M D A receptor agonist, other than aspartate or glutamate, which acts on amacrine and ganglion cells 9. The N A A O which we have identified in frog retinal neurons is a candidate for this function. N A A G ' s role in the nervous system remains to be established. Localization of the dipeptide in discrete neuronal populations is consistent with its activity as a neurotransmitter. More significantly, these data identify appropriate cells and tissues for future studies of N A A G ' s metabolism and physiological effects which are necessary to define its function.
Research supported by a research grant, D A 02297, from the National Institute on Drug Abuse.
401 1 Anderson, K.J., Monaghan, D.T,, Cangro, C.B., Namboodiri, M.A.A., Neale, J.H. and Cotman, C.W., Localization of N-acetylaspartylglutamate-like immunoreactivity in selected areas of the rat brain, Neurosci. Lett., 72 (1986) 14-20. 2 Anderson, K.J., Borja, M., Cotman, C.W., Namboodiri, M.A.A. and Neale, J.H., N-Acetylaspartylglutamateidentified in rat retinal ganglion cells and their projections in the brain,, Brain Research, in press. 3 Bernstein, J., Fisher, R.S., Zaczek, R. and Coyle, J.T., Dipeptides of glutamate and aspartate may be neuroexcitants in the rat hippocampal slice, J. Neurosci., 5 (1985) 1429-1433. 4 Cangro, C.B., Garrison, D.E., Luongo, P.A., Truckenmiller, M.E., Namboodiri, M.A.A. and Neale, J.H., First immunohistochemical demonstration of N-acetylaspartylglutamate in specific neurons, Soc. Neurosci. Abstr., 11 (1985) 108. 5 Cangro, C.B., Namboodiri, M.A.A., Sklar, L.A. and Neale, J.H., Biosynthesis and immunohistochemistry of Nacetylaspartylglutamate in spinal sensory ganglia, J. Neurochem., submitted. 6 Curatolo, A., D'Arcangelo, P., Lino, A. and Brancati, A., Distribution of N-acetyl-aspartic and N-acetyl-aspartylglutamic acids in nervous tissue, J. Neurochem., 12 (1965) 339-342. 7 ffrench-Mullen, J.M.H., Koller, K., Zaczek, R., Coyle, J.T., Hod, N. and Carpenter, D.O., N-acetylaspartylgluta-
mate: possible role as the neurotransmitter of the lateral olfactory tract, Proc. Natl. Acad. Sci. U.S.A., 82 (1985) 3897-3900. 8 Koller, K.J., Antuono, P.G. and Coyle, J.T., The ontogeny and phylogeny of N-acetyl-aspartyl-glutamate, Soc. Neurosci. Abstr., 10 (1984) 226. 9 Lukasiewicz, P.D. and McReynolds, J.S., Synaptic transmission at N-methyl-D-aspartate receptors in the proximal retina of the mudpuppy, J. Physiol. (London), 367 (1985) 99-115. 10 Miyake, M., Kakimoto, Y. and Sorimachi, M., A gas chromatographic method for the determination of N-acetyl-Laspartic acid, N-acetyl-a-aspartylglutamic acid and fl-cytryl-L-glutamic acid and their distributions in the brain and other organs of various species of animals, J. Neurochem., 36 (1981) 804-810. 11 Miyamoto, E., Kakimoto, Y. and Sano, I., Identification of N-acetyl-a-aspartylglutamic acid in the bovine brain, J. Neurochem., 13 (1966) 999-1003. 12 Slaughter, M.M. and Miller, R.F., 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. 13 Westbrook, G.L., Mayer, M.L., Namboodiri, M.A.A. and Neale, J.H., High concentrations of N-acetylaspartylglutamate (NAAG) selectively activate NMDA receptors on mouse spinal cord neurons in cell culture, J. Neurosci., 6 (1986) 3385-3392.