Brain Research, 110 (1976) 99-114

99

© Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands

AN E V A L U A T I O N OF L - G L U T A M A T E AS T H E T R A N S M I T T E R LEASED F R O M OPTIC NERVE T E R M I N A L S OF T H E P I G E O N

RE-

P. M. BEART* M RC Neurochemical Pharmacology Unit, Department of Pharmacology, Medical School, Cambridge CB2 2QD (Great Britain)

(Accepted November 14th, 1975)

SUMMARY The possibility was investigated that L-glutamic acid is the excitatory transmitter released from the optic nerve terminals of the pigeon optic tectum. (1) Superficial layers of the tectum contained high levels of endogenous glutamate and accumulated L-[3H]glutamate by a high affinity uptake process. (2) Subcellular and autoradiographic studies indicated that 10-30~o of the exogenously accumulated L-[3H]glutamate was localized within synaptosomes, and that 11-15 ~ of the synaptosomes had been labelled. (3) The glutamate-accumulating synaptosomes sedimented to the same isopycnic density as pinched-off optic nerve terminals. (4) GABA- and noradrenaline-accumulating synaptosomes were also associated with this subcellular population. (5) Retinal ablation did not change endogenous glutamate concentrations or the high affinity uptake of glutamate. The results are discussed in relation to a possible role for L-glutamate as the 'optic nerve transmitter' and in the context of previous evidence implicating glutamate as an excitatory transmitter.

INTRODUCTION L-Glutamic acid excites a wide variety of central neurones 1~, and there is evidence to suggest that the glutamate-induced depolarization of neurones mimics that of the synaptically released excitatory transmitter 15. Unfortunately none of the anta* Present address and address for correspondence: Neucology Research, The Children's Hospital Medical Center, 300 Longwood Avenue, Boston, Mass. 02115, U.S,A.

100 gonists presently available have proved of sufficient specificity ~5 to allow the uncqui-. vocal identification of the pathways where this amino acid might be the transmitier~ Neurochemical evidence, however, suggests that L-glutamate may be the transmiltcr released by cerebellar granule cells 6;~ and by spinal primary afferent terminals of the sensory pathway "~,l~.a:~,'~°. In addition, there is evidence that L-glutamate may be an excitatory transmitter" in the invertebrate nervous system:Q The present study describes investigations into a possible role flw L-glutamate as the transmitter released from pigeon optic nerve terminals. The pigeon visual system offers certain advantages ~4 over that of the mammal; in particular, most optic nerve fibres project to the optic tectum j~, the pathways are completely crossed ~1,47 the laminated tectal anatomy has been well described 11, and the tecta are easily accessible. The approach adopted in this study was to investigate biochemical parameters which should be associated with glutamate-mediated neurotransmission, ~Jnd to look for changes upon denervation. A short time interval (7 days) was deliberately chosen for the degeneration studies, for by analogy with the autonomic nervous system extensive degenerative changes should occur within 2-5 days of denervation 6,z'','~s. Furthermore, substantial degeneration is also found at similar times after denervation in the central nervous system ~3,~4,~s. METHODS

Adult pigeons (Columba livia, body wt. 250-400 g, both sexes) were sacrificed by decapitation, the cerebral cortices removed, the optic tract transected, and the optic tecta removed and placed on an ice-cold aluminium plate. Any remaining optic tract was removed, each tectum sagitatly bisected with a sharp razor blade and most of the central white matter (below about layer 13 or 14) removed to leave two sections of superficial gray matter. Approximately 70 mg wet weight of gray matter was obtained from a single tectum (total wet wt. approximately 115 rag). lntravitreal injections and retinal ablations were performed under halothanenitrous oxide-oxygen anaesthesia. A fine Hamilton microsyringe was introduced into the superior pole of the left eye and once the location had been checked by observing the characteristic pressure transmitted to the liquid in the syringe, 10 pl of an aqueous [~H]amino acid mix (100 #Ci total, containing 48/~Ci proline, 37 #Ci lysine, 11/zCi tryptophan and 4/zCi valine) was injected over approximately 1 rain. For retinal ablation the head was held stereotaxically in position, and after the appropriate surgery the contents of the orbit including the retina were removed by suction. On conclusion of the operation a small piece of adsorbent hemostatic gauze (Surgicet, Ethicon Ltd.) was placed in the base of the orbit to reduce any bleeding, the ablated eye was closed by suturing and sprayed briefly with plastic dressing (Nobecutane Spray, British Drug Houses).

Uptake studies The superficial gray matter was homogenized in 20 vol. of icecold 0.25 M sucrose and the homogenate centrifuged at 1,000 × g for 10 rain. A 40/A aliquot of

101 this crude synaptosomal (St) preparation was added to 2 ml of Krebs-Ringer phosphate pH 7.4 and the samples preincubated for 5 min at 37 °C in a shaking water bath. I.-[3H]Glutamate (25/A, final concentration 10 nM) was then added and the incubation continued for a further 5 rain. At the end of this incubation all samples were cooled to 4 °C, and the particulate material collected by rapid vacuum filtration through membrane filters (Millipore HAWP, 25 mm diameter, 0.45/~m pore size) supported in stainless steel holders. Incubation vessels and filters were washed with cold 0.9 °/o sodium chloride solution, the filters dissolved in 2 ml of ethoxyethanol in scintillation vials and 10 ml of 0 . 4 ~ butyl PBD toluene scintillant added. All experiments were done in triplicate with appropriate 0 °C and filter blanks being run concurrently. (a) Inhibitor studies. Glutamate analogues were present at a final concentration of 100 # M during both the preincubation and incubation. Experiments were done in triplicate with appropriate controls being run concurrently. (b) Kinetic studies. Experiments were carried out as above except that the initial velocities of uptake were determined over a range of glutamate concentrations (8-250 # M ) with unlabelled glutamate (100 ktl) being added with the L-[3H]glutamate. Kinetic parameters were determined from the regression lines fitted by the method of least squares to the individual initial velocity data.

Subcellular fractionation Aliquots of the crude synaptosomal preparation (prepared from a 10 ~i homogenate of optic tectum gray matter in 0.25 M sucrose) equivalent to 20-200 mg of wet weight of tissue were added to Krebs-phosphate buffer in a final volume of 10 ml and preincubated at 37 °C for 5 min. Labelled compounds (L-[14C]- or L-[3H]glutamate, L-[14C]aspartate, [3H]GABA or [3H]noradrenaline) were added, the incubation continued for 5 min and was then cooled to 4 °C. The exact concentration of label (range 15--74 nM) added varied from one experiment to the next, but was always such that at the final concentration in the incubation mix most of the label accumulated would be taken up by the high affinity uptake system 4,37. The incubation mixture was diluted with 15 ml of cold Krebs phosphate and centrifuged at 10,000 × g for 20 rain. The resultant mitochondrial (Pz) pellet was rinsed with 1 ml of 0.25 M sucrose, resuspended in 1.3 ml of 0.25 M sucrose and carefully layered onto a linear sucrose gradient (16.7 ml, 0.3-1.6 M). After loading the gradients were centrifuged for 2 h at 100,000 "< g - - in some experiments g values and centrifugation times were varied. Fractions were obtained by sampling from the bottom of the gradient: 0.5 ml and 0.4 ml fractions were routinely taken for scintillation spectrometry and biochemical analysis, respectively. When the radioactivity in each sample was to be determined the samples were expelled directly into scintillation vials containing 1.5 ml of distilled water and 12ml of 0 . 4 ~ butyl PBD-toluene-triton (1:2) scintillant was added. Otherwise fractions were expelled into centrifuge tubes and stored on ice until required (0.5-2 h). Glutamate decarboxylase activity was determined as described previously 9 by following the formation of [14C]GABA from L-[14C]glutamate. Pseudocholinesterase activity was assayed by a procedure similar to that of Klingman et al.35; acetylthio-

102 choline iodide was employed as the substrate and acetylcholinesterase activity was inhibited with the specific inhibitor B.W. 284C51. Glutamate dehydrogenase was studied in the direction of glutamate formation by a technique similar to that of" Balfizs 3. Glutamate concentrations across the gradient were assayed by a procedure based on that of Graham and Aprison 2~. Protein was determined by the method of Lowry et al.4°; bovine serum albumin was used as the standard and appropriate corrections were made for sucrose interference after checking the linearity of the gradients.

Electron microscopy autoradiography Uptake and subcellular fractionation were carried out as described above. Pz pellets (equivalent to 100 mg wet wt. of tissue) were prepared from synaptosomal preparations incubated with L-[3H]glutamate (50 #Ci, 0.4/tM) or directly from gray matter of the tectum contralateral to the eye which had received an intravitreal injection of [3H]amino acids (100 #Ci) 24 h previously (as above). Sucrose gradient fractions selected for fixation corresponded to those previously found to be in the 'synaptosomal' peak (1.1-1.3 M) of radioactivity, and fractions either side of those fixed were taken for scintillation spectrometry to check the exact position of this peak. Fractions (0.5 ml) were fixed directly with 1.5 ml of 5 ~,,iglutaraldehyde in Krebs Ringer phosphate at room temperature for 1 h and the fixed particles were pelleted by centrifugation at 100,000 :~: g for 15rain in a 3 ~ 3 ml swingout rotor. Pellets were rinsed in Krebs-Ringer solution, post-fixed in 1 '~'0 osmium tetroxide in 0.1 M phosphate buffer, washed, dehydrated in progressively increasing concentrations of ethanol, and embedded in Epon-resin. Thin sections (approximately 60 nm) were cut and prepared for autoradiography, and after 10-35 days exposure examined under the electron microscope. In a further series of experiments autoradiography was also carried out on the following material: (a) tectal P2 pellet of a pigeon intravitreally injected with [ZH]amino acids as described, (b) pellet obtained by centrifugation (100,000 f, g, 10 rain) of a fixed synaptosomal preparation ($1, equivalent to 30 mg wet wt. of tissue), which had been previously incubated with L-[aH]glutamate (25 is,Ci, 1.8/zM). Metabolism of accumulated L-glutamate Gradient fractions in the 'synaptosomal' region of sucrose gradients or Pz pellets were extracted with 70 ~ ethanol, and the amino acid fraction separated on a short Dowex 50 W (H +) column. The concentrated amino acid eluate was subjected to descending chromatography (30 h) on Whatman No. 1 paper with 8 0 ~ aqueous phenol-ethanol-water-0.88 ammonia (150:40:10:1) as the solvent. The amino acids were located by the colour of marker unlabelled amino acids with ninhydrin, the chromatograms cut into strips, and the radioactivity eluted with 4 ml ofethoxyethanol and determined by scintillation spectrometry. Essentially all accumulated radioactivity in the synaptosomal region of the gradients was present in the amino acid fraction, distributed as glutamate 61 ~ 7 ~/,i, glutamine 31 ~: 9 ~ , GABA 4 ~ 2 ~ , aspartate l i 0.3 ~ and alanine I i 0.5"/(all n 3). Immediate analysis of P,_, pellets gave /O

103 very similar results suggesting that most glutamate metabolism occurred prior to subcellular fractionation. Ami n o acid analysis

A 10~o homogenate o f freshly dissected tissues was prepared in ice-cold 5 ~i aqueous perchloric acid, centrifuged at 2000 x g for 10 rain, and a 100 #1 aliquot o f the supernatant analyzed on a Locarte amino acid analyzer (bench model). One such 100/~1 aliquot was subjected to hydrolysis (overnight in a sealed tube and at elevated temperature), with an equal volume o f 12 M hydrochloric acid and analyzed. A medium sized peak located just before the large free a m m o n i a peak on the analyzer traces was found to possess an elution time identical to that o f an authentic sample o f L-carnosine. Acid hydrolysis produced no change in G A B A levels, while {/-alanine was produced in an a m o u n t approximately equivalent to the original carnosine concentration. RESULTS Levels and distribution o f glutamate and other amino acids

The preparation of superficial tectal layers employed in these studies was of such a thickness that it would contain the optic nerve terminals la,31,~7 (confirmed by the subcellular and autoradiographic results in intravitreally injected birds reported below), and thus should be a relatively more enriched and suitable preparation than the whole tectum for investigations into the role o f L-glutamate as the transmitter released f r o m optic nerve terminals. High concentrations o f glutamate (8 #moles/g) were found in the superficial layers of the tectum. Glutamate, however, did not appear to be concentrated in superficial tectal layers, nor was there a gradient of aspartate or G A B A concentrations. Taurine and carnosine were apparently localized in deeper tectal layers (see Table I). TABLE I A M I N O A C I D L E V E L S IN F R E S H L Y D I S S E C T E D P I G E O N O P T I C T E C T U M

Values in/Jmoles/g wet wt. of tissue are the mean ± S.E.M. of the number of experiments indicated, and were determined by conventional amino acid analysis (see Methods). Lesioned samples were from tecta contralateral to the ablated eye, while ipsilateral tecta of the same pigeons served as controls and are reported in column two. Amino acid

Whole tectum (n = 3 or 4)

Superficial l a y e r s (n = 5)

Superficial layers 7 day retinal ablated (n = 5)

Glutamate Aspartate Glycine Alanine GABA Glutamine Taurine Carnosine

8.07 ~ 0.21 3.69 ~- 0.07 0.82 ± 0.07 0.60 i 0.05 3.17 i 0.16 4.69 2_ 0.31 2.64 -3- 0.08 1.38 ± 0.10

8.03 ± 0.32 3.22 :~ 0.24 0.74 ± 0.02 0.66 ~ 0.02 3.29 ± 0.08 5.02 5- 0.18 1.95 ± 0.13 0.95 ± 0.03

7.72 ± 0.16 3.34 ± 0.16 0.79 ~ 0.03 0.69 ± 0.02 3.17 ± 0.07 4.89 ± 0.35 1.83 % 0.07 1.12 :~ 0.06

104

/ 311]Glutamate uptake Crude synaptosomal preparations ($1), prepared from superficial tectal layers. rapidly accumulated exogenous L-[;3H]glutamate from the incubation medium. Kinetic analysis of L-[:SH]glutamate uptake over a range of glutamate concentrations revealed that both high and low affinity transport systems were present, with the high affinity uptake system having an apparent h',~ and Vmax of 9.8 ~- 2.3 #M and 163 i 35 nmoles/g/min, respectively. When the substrate inhibition properties of this high affinity uptake system were examined to enable a comparison with those previously described, L-aspartate (87 ?,i inhibition of control uptake), D-aspartate (88~;), D-glu90 tamate (89'};) and L-cysteate (9,/o) were all found to be potent inhibitors at 100/+M. while L-pyroglutamate was almost ineffective (71}o). Thus, the high affinity uptake system for L-glutamate in this preparation closely resembles that previously described for both synaptosomes ~,3:~and brain slices 4,~.

Subcellular fractionation When crude mitochondrial pellets (P2) from S1 synaptosomal preparations incubated with L-[~H]- or L-[14C]glutamate were layered on to linear sucrose density gradients (0.3-1.6 M), 38 ~ 2°o (n == 18) of the total radioactivity was associated with the "synaptosomal' region. Further subcellular studies revealed that exogenously accumulated L-[14C]aspartate and L-[SH]glutamate sedimented to the same isopycnic density (Fig. 1), and thus appeared to label the same subcellular population --48 3 "i~i (n = 6)of the L-aspartate was associated with the peak of radioactivity. The synaptosomal marker enzyme glutamate decarboxylase 46,~1 and the glial marker enzyme pseudocholinesterase 1°,36,60 were also associated with this subcellular population, while the mitochondrial enzyme glutamate dehydrogenase ~1 was located at higher density (see Fig. 1). This result is thus in agreement with earlier evidence that synaptosomal fractions are likely to be contaminated with glial fragments ~0,6°. In further series of 7 experiments exogenous and endogenous glutamate were shown to share an identical subcellular localization on sucrose gradients as,61, although the amount of glutamate associated with the supernatant was extremely variable. Glutamate was not the only amino acid with such a localization, however, for aspartate, glycine, alanine, GABA, glutamine and particularly taurine were all found to be present (Beart, unpublished observations). Endogenous amino acids may therefore be associated generally with the total synaptosomal population, as previously suggested41, rather than uniquely localized.

Labelling of optic nerve terminals by intravitreal injection Since intravitreal injection of radiolabelted amino acids results in the preferential labelling by newly synthesized proteins of optic nerve terminals 12,14,54,55, their synaptosomes can be localized after this pretreatment on sucrose gradients. Furthermore, as synaptosomat fractions are maximally labelled by this rapid phase of axonal transport at 24 h 14,54, this time was employed in all studies. Following an intravitreal injection of Jail]amino acids, radioactivity was found in the crude nuclear pellets of the superficial layers of the tecta both contra- and ipsitateral to the injected eye;

105

20-

o'-

{

injected *" 15- 3N_amino

10-

3H-GLU - 3H-ASP

[

'

]

.> I1J

E N r

I

i>- 20-

I

|

I

'I '

i

I

T

GDH

o

I

I

I

II

i] "

r

J T -

~

10-

oJ 1.6

../.protein

J I

1

I

1.2

0.8

0.4

I

I

I

1.2 0.8 0.4 sucrose molarity

I

I

I

1.2

08

0.4

Fig. l. Distribution of radioactivity, enzyme activity and protein after density gradient centrifugation of mitochondrial pellets on linear sucrose gradients. Glutamate decarboxylase (GAD), pseudocholinesterase (OChE) and glutamate dehydrogenase (G DH) activities, and protein were determined on the same gradients as exogenously accumulated L-glutamate (GLU). The sedimentation characteristics of preparations additionally labelled by incubation with L-aspartate (ASP) or by intravitreal injection of [3H]amino acids are also shown. In each case results are expressed as a percentage of the total radioactivity, enzyme activity or protein recovered in all fractions. A single experiment is illustrated; similar results were obtained from at least one further experiment. Further details are given in the text. the label f o u n d ipsilaterally was a p p r o x i m a t e l y h a l f o f t h a t f o u n d in the o p p o s i t e tectum a n d was p r e s u m a b l y systemic in origin 14. W h e n the Pz pellets were subjected to density g r a d i e n t centrifugation, the g r a d i e n t s p r e p a r e d f r o m ipsilateral tecta cont a i n e d essentially no radioactivity, while in those f r o m the c o n t r a l a t e r a l tecta 75 -~ 1 ~,, (n - ; 4) o f the label was f o u n d to be s y n a p t o s o m a l l y localized (Fig. 1). These subcellular studies n o t only confirm earlier results o f this type a n d the crossed nature of pigeon r e t i n o t e c t a l fibres11, 47, b u t also d e m o n s t r a t e t h a t the p r e p a r a t i o n o f superficial layers c o n t a i n e d optic nerve t e r m i n a l s . Essentially no r a d i o a c t i v i t y was located

106 in the supernatant, and a small 'hump" of label at 0.8-.1.0 M may represent myelinassociated material 14,5~'(Fig. 1). In synaptosomal preparations (Sa) of pigeons injected with [3H]amino acids 24 h previously and incubated with L-[J~C]glutamate, both labelled synaptosomal populations were localized at identical positions on sucrose gradients (Fig. 1). Although this result suggests that the glutamate-accumulating synaptosomes may in fact be those of the optic nerve terminals, they might equally well be those of other glutamate-releasing tectal pathways which were of very similar density to the pinched-off optic nerve terminals. One possible approach to this problem was to employ the technique of incomplete equilibrium sedimentation. As high concentrations of GABA (Table I) and noradrenaline (I.2/~g/g, Moore and Beart, unpublished observations) were found in superficial tectal layers, crude synaptosomal preparations were incubated with low concentrations of [3H]noradrenaline or [3H]GABA and L-[14C]glutamate (see Methods). High affinity uptake systems were found for both 3"i, GABA and noradrenaline, with 45 :~:l o~ . ( n -. 4) of . the . GABA . and 61 (n == 4) of the noradrenaline associated with the synaptosomal region on sucrose gradients. Although a large number of experiments were performed and g values, centrifugation times and loads placed on the gradients (P2 pellets equivalent to 20200 mg of original tissue) were all varied, it was never possible to separate populations of synaptosomes accumulating glutamate, GABA or noradrenaline, in view of these results, no attempts were made to separate by incomplete equilibrium sedimentation the population(s) of synaptosomes labelled by intravitreal injection and incubation with L-glutamate.

Electron microscopy autoradiography First atttoradiographic attempts to identify the sites of labelling employed sucrose gradient fractions fixed directly in suspension. In those preparations labelled with L-[ZH]glutamate, isolated silver grains were seen throughout the sections at the electron microscope level, but large clusters of silver grains were localized only over densely staining particles. These particles were always of irregular or dumbbell shape, sometimes contained internal structure and mitochondria, although synaptic vesicles could often not be identified (Fig. 2A). It is known that synaptosomes undergo osmotic shrinkage59 and consequent distortion on sucrose gradients. Analysis of low power electron micrographs indicated that 28 ~ of the total silver grains was localized over these particles (see Table I1), although there were some difficulties in the identification of structures found in the fixed sucrose gradient fractions. These densely stained and heavily labelled particles were tentatively identified as synaptosomes (Table II). Confirmation of this localization came from those preparations labelled by intravitreal injection, in which most of the silver grains were found over essentially identical particles (Fig. 2B). However, the number of silver grains over individual synaptosomes and the number of labelled synaptosomes was much less than seen with [3H]glutamate. Silver grains were occasionally associated with myelin in intravitreally injected pigeons, in agreement with previous findingsla,55. These results were confirmed by electron microscopy autoradiography in

107 TABLE I1 LOCALIZATION OF L-[3HIGLUTAMATE IN SYNAPTOSOMAL PREPARATIONS

Values were obtained from 9 and 11 randomly selected low power electron micrographs of crude homogenates and gradient fractions, respectively. The area of section in each micrograph was 75 sq. /era. The percentage of total area occupied by various tissue components was estimated by the application of a 168 point grid; the distribution of silver grains was analyzed in the same pictures and results are expressed as the percentage of total silver grain population located over the various particulate components. Preparation

Proportion of labelled synaptosomes (%)

Crude homogenates (SD 11.5

Gradient fractions

15.0

Particulate eomponent

Proportion of total surJbce area (%)

Synaptosomes 4.6 Free mitochondria 0.3 Myelin fragments 4.2 Unidentified structures 68.4 Space between particles 22.5 Synaptosomes I. 6 Free mitochondria 3.5 Myelin fragments 7.7 Unidentified structures 36.0 Space between particles 51.2

Proportion o/ total silver grains (%)

11.4 2.3 2.3 71.5 12.5 28.4 9. l 3.4 37.5 21.6

samples not subjected to sucrose gradient centrifugation (see Methods). Synaptosomes were always easily identifiable in these sections; they generally contained one or more mitochondria, and vesicles were frequently seen 59. With [3H]glutamate, clusters of silver grains were concentrated in synaptosome-enriched regions of the sections. In these synaptosome-enriched regions, analysis o f r a n d o m low power electron micrographs indicated that 11 ~ of the silver grains was localized over identifiable synaptosomes (see Table lI, Fig. 2C). The apparent discrepancy between synaptosomally localized silver grains in the crude homogenates and gradient fractions is probably due to problems of identification o f structures seen in these preparations. However, whether the preparation was a crude homogenate or a gradient fraction, the distribution of silver grains represented an enrichment relative to the surface area occupied by synaptosomes. The results indicate that approximately 11-15~o of the terminals present in superficial layers o f the rectum accumulate glutamate, and thus may use this amino acid as their transmitter substance. Labelled synaptosomes were also f o u n d in P2 pellets o f intravitreally injected pigeons, but once again the number of labelled synaptosomes was considerably less than seen with [3H]glutamate. Degeneration studies W h e n the concentrations o f a m i n o acids in superficial layers of control and lesioned tecta were compared, no significant changes were f o u n d (Tables I and liD. There was a small but insignificant fall in glutamate levels, while the levels of aspartate, another possible excitatory transmitter 15, were also unchanged. G A B A levels were

108

Fig. 2. Electron microscope autoradiographs of synaptosomal preparations of pigeon optic rectum. Silver grains are present over densely stained, irregularly shaped particles in sucrose gradients of mitochondrial pellets incubated with L-[3H]glutamate (A) or prelabelled by intravitreal injection (B). In C silver grains are localized over a synaptosome in the pellet of a homogenate incubated with L-[3H]glutamate (not subjected to density gradient centrifugation) and processed as described. Further details are given in the text. Calibration bar represents 0.5/tin.

109 TABLE llI THE EFFECT OF RETINAL ABLATIONON TIlE HIGH AFFINITYGLUTAMATEUPTAKE AND THE CONCENTRATIONS OF GLUTAMATE AND OTHER AMINO ACIDS All values are the mean ~ S.E.M. of the n u m b e r of experiments indicated. [:3H]Glutamate uptake is given as tissue medium ratios (nmoles/g/min-nmoles/ml/min). F o r other details see Table 1.

Glutamate Aspartate GABA [:~H]Glutamate uptake

(n 5) (n 5) (n = 5) (n - 3)

Control

Retinal ablated

8.03 3.22 3.29 113

7.72 3.34 3.17 139

_.-t:0.32 ± 0.24 ± 0.08 :~ 18

i 0.16 _~ 0.16 d: 0.07 :~ 14

also not significantly altered, suggesting that transneuronal degeneration, observed upon lesioning other excitatory pathways 34,62, had not occurred at this time interval. Furthermore, the high affinity uptake of L-[3H]glutamate by crude synaptosomal preparations was also unaffected 7 days after retinal ablation. When control and lesioned synaptosomal preparations of the same pigeon were examined, the tissuemedium ratio of controls was not significantly different from that of the lesioned pigeons (see Table 1II). DISCUSSION

The results of the uptake, subcellular fractionation and autoradiographic studies reported here do not rule out a role for L-glutamate as the optic nerve transmitter. However, the failure to observe reductions in the concentration of endogenous glutamate and the high affinity uptake of [3H]glutamate in the degeneration studies indicates that glutamate may not be the transmitter involved. Denervation of both autonomic6,22,58 and central pathways 23,24,~s is certainly accompanied at time intervals similar to that employed in this study (7 days) by large reductions in those biochemical parameters associated with the particular transmitter involved. Although the degeneration of optic nerve terminals is only complete at approximately 3 weeks 11Az and there does appear to be a reduction of the high affinity uptake of glutamate at this time (Cu6nod, personal communication), there is substantial evidence to suggest that considerable degenerative changes have occurred at 7 days 1,11-13. Certainly, 2 days after optic nerve transection in the cat synaptic transmission begins to fail 19. Furthermore, spinal primary afferent terminals, which may release glutamate as their transmitter, show maximal degeneration at 2 days, have completely disappeared by the fourth day 27, and transneuronal degeneration occurs from the second day after dorsal root sectiona4. Glutamate (and aspartate) concentrations and the high affinity uptake of glutamate in spinal cord 4s and olfactory bulb 42,43 also appear to be unaffected by denervation of the appropriate excitatory pathway. In addition, glutamate concentrations are unaltered in the rabbit lateral geniculate nucleus following enucleation44, although it has recently been reported

1 I0 that glutamate levels arc reduced by 20 30[~,i in the frog tectum by' enucleation {~:'-. The restllts of this investigation in the pigeon optic tectum are thus similar io ihose obtained at other sites where glutamate may be the transmitter involved. Cuenod and coworkers have also shown in an independent study that L-glutamate and GABA are accumulated by high affinity transport systems in tectal homogenates, and that the radiolabelled synaptosomes sediment to an isopycnic density very similar to that of optic nerve terminals (Cuenod, personal communication) -these results are thus in complete agreement with the present investigation. As GABAand noradrenaline-accumulating nerve terminals are associated with the optic nerve terminals, these inhibitory neurotransmitters may be directly involved in the modulation of this primary visual pathway. The results with noradrenaline are also in close agreement with fluorescent histochemical studies demonstrating the presence of catecholamine-containing terminals in superficial tectal layers in close proximit~ to layer 5 ca. Barth and Felix: have recently presented excellent evidence for GABAmediated inhibitory pathways within the tectum, and the subcellular results conl]rm their findings and those of Neal and lversen 46 on the localization of glutamate decarboxylase, exogenous and endogenous GABA. Autoradiographic studies confirm the synaptosomal localization of the exogenously accumulated [aH]GABA (Beart, unpublished observations). The high levels of carnosine detected in the pigeon rectum are also of some interest in view of recent studies on this dipeptide in the olfactory bulb 4~. A number of points raised by this investigation warrant discussion. The apparent disparity between the numbers of synaptosomes labelled by intravitreal iniection and by incubation with [all]glutamate (see Results) was not altogether unexpected. lntravitreal injection labels only optic nerve terminals 1=',x4,5~,~a, which will form only a small percentage of the synaptosomes of superficial tectal layers, while Jail]glutamate was probably accumulated by many types of synaptosomes. In contrast to earlier reports~S,"°.a0, ae, a synaptosomal localization of exogenously accumulated glutamate is described here. Both the M~iller glial cells of the retina ts and the Bergmann gila of the cerebellum a° appear to be extraneuronal sites of glutamate uptake, as do the satellite glial cells of the dorsal root ganglion TM (although the latter contains no nerve terminals). Glutamate has also been localized over Schwann cells at the insect neuromuscular junction 20, where it has been implicated as the transmitter substance. The increased retention of glutamate by suspension fixation of homogenates relative to the more usual pellet fixation s may be one reason for the localization found in this study. Although glial-localized glutamate was not found in the present investigation, glutamate-accumulating glial fragments may represent some portion of the label associated with unidentified structures (see Table II). Schon and Kelly 5a have recently shown /5'-alanine to be a specific glial cell marker, and this label should aid the identification of glial uptake sites present in homogenates. The intimate association of the glial marker enzyme pseudocholinesterase with the synaptosomat region of the gradients and recent reports of gila possessing a high affinity glutamate uptake system 21,eg,4r~,56also suggest that glial uptake of glutamate may have occurred in the preparations employed in this study. Clearly the proportion of glutamate uptake

Ill attributable to such a glial mechanism is difficult to determine, particularly as the estimates of the Vmax of the glial transport systems vary so widely 21,49,~6. Thus, in any uptake or subcellular study with glutamate, glial uptake must be recognised as contributing, perhaps substantially, to the net observed uptake of exogenous glutamate. Indeed, as gliosis occurs in the pigeon optic rectum after denervation2,12,13, and those gila invading degenerating terminals may possess a high affinity glutamate uptake, in addition to containing endogenous glutamate 4~, degenerative changes may be somewhat masked. A further problem is the possibility that the synaptically releasable pool of glutamate represents such a small fraction of the total 8 #moles/g that degenerative changes might be obscured by much larger metabolic pools. In view of these two important considerations, a reappraisal of the neurochemical methodology generally employed to study degeneration and to establish transmitter identity may be required, particularly with respect to amino acid transmitters which are so intimately associated with intermediary metabolism. It may, for example, be useful to examine the effect of lesioning on the uptake of radiolabelled fl-alanine, the activities of glia[ versus neuronal localized enzymes, and metabolic compartmentation (e.g., specific activities of glutamate and glutamine synthesized from different labelled precursors). Although this investigation has not established glutamate as the transmitter released from optic nerve terminals, there would appear to be no definitive evidence to the contrary. However, the autoradiographic results do provide new evidence for a transmitter role for glutamic acid. As suggested in this discussion, problems have been encountered at other sites where glutamate might be the transmitter involved. It may be possible that the size of the synaptically releasable pool of glutamate varies considerably with the excitatory pathway, and that at certain sites only microelectrode studies will establish transmitter identity. To date, the most positive neurochemical evidence favouring a transmitter role for glutamate is the study of Snyder and coworkers 6~ in hamster cerebellum, where viral-induced depletion of granule cells results in a concurrent fall in endogenous glutamate concentrations and the high affinity uptake. ACKNOWLEDGEMENTS

P. M. Beart was the William Elgar Buck Postdoctoral Student at St. John's College, Cambridge and a Wellcome fellow. Special thanks are due to D. Chapman for the autoradiographic studies, and to C. Ward for the amino acid analyses. The advice and assistance of J. S. Kelly, L. L. Iversen, S. Gardiner and R. Jakes are also acknowledged.

REFERENCES ] AKERT, K., CUI~NOD, M., AND MOOR, H., Further observations on the enlargement of synaptic vesicles in degenerating optic nerve terminals of the avian rectum, Brain Research, 25 (1971) 255-263.

112 2 ALTMAN, J., AN[) ALTMAN, E., Increased utilization of an amino acid and cellular prolili~ratio~ demonstrated autoradiographically in the optic pathways of pigeons, Exp. NeuroL, 6 11962) 142-151. 3 BAL~.ZS, R., Control of glutamate oxidation in brain and silver mitochondrial systems. Biochem~ J., 95 (1965) 497 508. 4 BALCAR, V. J., AND JOHNSTON, G. A. R., Glutamate uptake by brain slices and ItS relation to depolarization of neurones by acidic amino acids, J. Neurobiol., 3 (1972) 295 30t. 5 BALCAR,V. J., AND JOHNSTON, G. A. R., High affinity uptake of transmitters : studies on the uptake of L-aspartate, GABA, L-glutamate and glycine in cat spinal cord, J. Neurochem., 20 (1973) 529539. 6 BANISTER, J., AND SCRASE, M., Acetylcholine synthesis in normal and denervated sympathetic ganglia of the cat, J. Ph.vsiol. (Lond.), I I l (1950) 437-444. 7 BARTH, R., AND FELIX, D., lnfluence of GABA and glycine and their antagonists on inhibitory mechanisms of pigeon's optic tectum, Brahz Research, 80 (1974) 532-537. 8 BEART, P. M., The autoradiographic localization of L-[3H]glutamate in synaptosomal preparations, Brain Research, 103 (1976) 350-355. 9 BEART, P. M., UHR, M. L., AND JOHNSTON, G. A. R., Inhibition of GABA transaminase activity by 4-aminotetrolic acid, J. Neurochem., 19 (1972) 1849 1854. l0 COTMAN, C., HERSCHMAN, H., AND TAYLOR, D., Subcellular fractionation of cultured glial cells, J. Neurobial., 2 ( 1971 ) 169- 180. I I COWAN, W. M., ADAMSON, L., AND POWELL, T. P. S., An experimental study of the avian visual system, J. Anat. (Lond,), 95 (1961) 545-563. 12 CUI~NOD, M., BOESCH, J., MARKO, P., PERISIC, M., SANDRI, C., AND SCHONBACH, J., Contributions of axoplasmic transport to synaptic structures and functions, Int. J. Neurosci., 4 (t972) 77-87. 13 CUFZNOD,M., SANDRI, C., AND AKERT, K., Enlarged synaptic vesicles as an early sign of secondary degeneration in the optic nerve terminals of the pigeon, J. Cell Sci., 6 (1970) 605-613. 14 CUt!NOD, M., AND SCHONBA('H, J., Synaptic proteins and axonal flow in the pigeon visual pathway, J. Nearochem., 18 (1971) 809 816. 15 CURTIS, D. R., AND JOHNSTON, G. A. R., Amino acid transmitters in the mammalian central nervous system, Ergebn. Physiol., 69 (1974) 97-188. 16 DAVIDOFF, R. A., GRAHAM, L. T., JR., SHANK, R. P., WERMAN, R., AND APRISON, M. H., Changes in amino acid concentrations with loss of spinal interneurones, J. Neurochem., 14 (1967) 10251031. 17 DUGGAN, A. W., AND JOHNSTON, G. A. R , Glutamate and related amino acids in cat spinal roots, dorsal root ganglia and peripheral nerves, J. Neurochem., 17 (1970) 1205-1208. 18 EHINGER, B., AND FALCK, B., Autoradiography of some suspected neurotransmitter substances: GABA, glycine, glutamic acid, histamine, dopamine and L-DOPA, Brain Research, 33 (1971) 157 172. 19 EYSEL, U. TH., GR(~SSrR, O.-J., AND SAAVEDRA, J. P., Correlation between ultrastructural and electrophysiological signs of degeneration in optic nerve terminals of the cat's lateral geniculate nucleus, J. Neural Transm., 35 (1974) 337--344. 20 FAEDER, I. R., AND SALPETER, M. M., Glutamate uptake by a stimulated insect nerve muscle preparation, J. Cell Biol., 46 (1970) 300-307. 21 FAIVRE-BAUMAN,A., ROSSIER, J., AND BENDA, P., Glutamate accumulation by a clone of glial cells, Brain Research, 76 (1974) 371 375. 22 FELDBERG,W., Synthesis of acetylcholine in sympathetic ganglia and cholinergic nerves, J. Physiol. (Land.), 101 (1943)432-445. 23 FONNUM, F., STORM-MATHISEN, J., AND WALBERG, F., Glutamate decarboxylase in inhibitory neurones. A study of the enzyme in Purkinje cell axons and boutons in the cat, Brah~ Research, 20 (1970) 259-275. 24 FONNUM, F., AND WALaERG, F., An estimation of the concentration of 7-aminobutyric acid and glutamate decarboxylase in the inhibitory Purkinje axon terminals in the cat, Brain Research, 54 (1973) 115 127. 25 FuxE, K., AND LJUNGGREN, L., Cellular localization of monoamines in the upper brain stem of the pigeon, J. camp. Neuro/., 125 (1965) 355-382. 26 GRAHAM, L. T., JR., AND APRISON, M. H., Fluorimetric determination of aspartate, glutamate, and 7-aminobutyrate in nerve tissue using enzymic methods, Analyt. Biochem., 15 (1966) 487-497. 27 HEIMER, L., AND WALL, P. D., The dorsal root distribution to the substantia gelatinosa of the rat with a note on the distribution in the cat, Exp. Brain Res., 6 (1968) 89-99.

113 28 HELLER, A., Neuronal control of brain serotonin, Fed. Proc., 3l (1972) 81-90. 29 HENN, F. A., GOEDSTEIN, M. N., AND HAMBERGER, A., Uptake of neurotransmitter candidate glutamate by glia, Nature (Lond.), 249 (1974) 663 664. 30 H/3KFELT, T., AND LJUNGDAHE, /~., Applications of cytochemical techniques to the study of suspected transmitter substances in the nervous system. In E. COSTA, L. L. IVERSEN AND R. PAOLETTI (Eds.), Advances in Biochemical P3yehopharmacology, Vol. 6, Raven Press, New York, 1972, pp. 1-37. 31 HOLDEN, A. L., The field potential profile during activation of the avian optic tectum, J. Physiol. (Lond.), 194 (1968) 75-90. 32 JOHNSON, J. L., Glutamic acid as a synaptic transmitter in the nervous system, Brain Research, 37 (1972) 1-19. 33 JOHNSON, J. L., ANI3 APRISON, M. H., The distribution of glutamic acid, a transmitter candidate, and other amino acids in the dorsal sensory neuron of the cat, Brain Research, 24 (1970) 285 292. 34 KEEL'C, J. S., GOTTESFELD, Z., AND SCHON, F., Reduction in G A D I activity from the dorsal lateral region of the deafferented rat spinal cord, Brain Research, 62 (1973) 581-586. 35 KUNGMAN, G. I., KLINGMAN, J . D . , AND POL1SZCZUK, A., Acetyl- and pseudocholinesterase activities in sympathetic ganglia of rats, J. Neurochem., 15 (1968) 1121-1130. 36 KOELEE, G. B. K., The histochemical localization of cholinesterases in the central nervous system of the rat, J. eomp. Neurol., 100 (1954) 211-235. 37 KUHAR, M. J., Neurotransmitter uptake: a tool in identifying neurotransmitter-specific pathways, L(Je Sci., 13 (1973) 1623-1634 38 KUHAR, M. J., AND SNVDER, S. H., The subcellular distribution of free 3H-ghltamic acid in rat cerebral cortical slices, J. Pharmacol. exp. Ther., 171 (1970) 141-152. 39 LOGAN, W. J., AND SNVDER, S. H., High affinity uptake systems for glycine, glutamic and aspartic acids in synaptosomes of rat central nervous system, Brain Research, 42 (1972) 413-431. 40 LOWRY, O. H., ROSENBROUGH, N. J., FARR, A. L., AND RAYDAEE, R. J., Protein measurement with Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 41 MANGAN, J. L., AND WHITTAKER, V. P., The distribution of free amino acids in subcellular fractions of guinea-pig brain, Biochem. J., 98 (1966) 128 137. 42 MARGOEIS, F. L., Carnosine in the primary olfactory pathway, Science, 184 (1974) 909 911. 43 MARGOLIS, F. L., ROBERTS, N., FERRIERO, D., AND FELDMAN, J., Denervation in the primary olfactory pathway of mice: biochemical and morphological effects, Brain Research, 81 (1974) 469-483. 44 MARGOEIS, R. K., HEELER, A., AND MOORE, R.Y., Effects of changes in cellular composition following neuronal degeneration on amino acids in brain, Brain Research, I I (1968) 19 31. 45 MOKRASCH, L. C., Free amino acid content of normal and neoplastic astroglia, Brain Research, 25 (1971) 672 676. 46 NEAL, M. J., AND IVERSEN, L. L., Subcellular distribution of endogenous and ['~H]v-aminobutyric acid in rat cerebral cortex, J. Neurochem., 16 (1969) 1245-1252. 47 ROBERT, F., AND CUI~NOD, M., Electrophysiology of the intertectal commissures in the pigeon. I. Analysis of the pathways, Exp. Brain Res., 9 (1969) 116 122. 48 ROBERTS, P. J., AND KEEN, P., Effect of dorsal root section cn amino acids of rat spinal cord, Brahl Research, 74 (1974) 333-337. 49 ROBERTS, P. J., AND KEEN, P., [a4C]Glutamate uptake and compartmentation in glia of rat dorsal sensory ganglion, J. Neurochem., 23 (1974) 201 209. 50 ROBERTS, P.J., KEEN, P., AND MITCHELL, J. F., The distribution and axonal transport of free amino acids and related compounds in the dorsal sensory neuron of the rat, as determined by the dansyl reaction, J. Neuroehem., 21 (1973) 199 209. 51 SALGANICOEF,L., AND DE ROBERTIS, E., Subcellular distribution of glutamic acid, glutamine and 7-aminobutyric acid cycles in rat brain, J. Neurochem., 12 (1965) 287 309. 52 SCHON, F., AND KELLY, J. S., Autoradiographic localisation of [~H]GABA and [3H]glutamate over satellite glial cells, Brain Research, 66 (1974) 275 288. 53 SCHON, F., AND KELLY, J. S., Selective uptake of [:~H]/~-alanine: association with the glial uptake system for GABA, Brain Research, 86 (1975) 243-257. 54 SCHONBACH, J., AND CU~NOD, M., Axoplasmic migration of protein. A light microscopic study in the avian retinotectal pathway, Exp. Brain Res., 12 (1971) 275 282. 55 SCHONBACH,J., SCHONBACH, CH., AND CUI~NOD, M., Rapid phase of axoplasmic flow and synaptic proteins : an electron microscopical autoradiographic study, J. comp. Neurol., 141 (1971 ) 485-498.

114

56 SN()DGRASS, R.S., AND [XlRSI-',, ].o. 1., Amino acid uptake into human brain IumotHs, /l~Ji~l Research, 76 (1974) 95 107. 57 STONE, J., ANI) FREEMAN, J. A., Synaptic organization of the pigeon's optic tectum: a Golgi and current source-density analysis, Brabl Research, 27 (1971 ) 203-22 I. 58 VON EULER, U. S., AYD P~ rKnocn, A., Effect of sympathetic denervation on the noradrenaline and adrenaline content of lhe spleen, Acta physiol, scand., 24 (1951) 212 217. 59 W~I~TrAr,eR, V. P., The morphology of fractions of rat forebrain synaptosomes separated on continuous sucrose density gradients, Biochem. J., 106 (1968) 412-417. 60 WHITTAK~I~, V. P., Subcellular localization of neurotransmitters, Advanc. Q~'topharmacot., 1 (1971) 319 330. 61 Wovsl~¥, A. R., KUHAR, M.J., AND SNYD~:R, S. H., A unique synaptosomal population which accumulates glutamic and aspartic acids in brain tissue, Proc. nat. Aead. Sci. (Wash.), 68 (1971) 1102 1106. 62 YATES, P. A., AND ROnERTS, P. J., Effects of enucleation and intraocular colchicine on the amino acids of the frog optic tectum, J. Neurochem., 23 (1974) 891-893. 63 YOUN(~, A. B., OSTER-GRANITE, M . k . , HERNDON, R. M., AND SNYDER, S. H., Glutamic acid: selective depletion by viral induced granule cell loss in hamster cerebellum, Brain Resear('h. 73 (1974) I- 13.