Neurochem. Int. Vol. 20, No. 4, pp. 473~t86, 1992 Printedin Great Britain

0197-0186/92$5.00+ 0.00 PergamonPress Ltd

STRYCHNINE-INSENSITIVE GLYCINE RECEPTORS IN EMBRYONIC CHICK RETINA: CHARACTERISTICS AND MODULATION OF N M D A NEUROTOXICITY K. M. BOJE,* P. SKOLNICK, J. RARER,~ R. T. FLETCHERt and G. CHADER 1 Laboratory of Neuroscience, NIDDK and ~Laboratory of Retinal Cell and Molecular Biology, NEI, National Institutes of Health, Bethesda, MD 20892, U.S.A. (Received 21 June 1991 ; accepted 21 October 1991)

Akstract--ln the mammalian brain, the N-methyl-o-aspartate (NMDA) subtype of glutamate receptor is coupled to a cation channel and a strychnine-insensitiveglycinereceptor. The present paper demonstrates the presence of NMDA receptor-coupled strychnine-insensitiveglycinereceptors in embryonic chick retina. Both glycine and 1-aminocyclopropanecarboxylicacid (ACPC) exhibited similar potencies (271 + 39 vs 247 _ 39 nM, respectively)as inhibitors of strychnine-insensitive[aH]glycinebinding to retinal membranes. Moreover, glycine and ACPC enhanced [3H]MK-801 binding to sites within the NMDA-coupled cation channel in retinal membranes with potencies comparable to those reported in rat brain. While the potency of ACPC was significantlyhigher than glycine (ECs0 54 + 12 vs 256 + 57 nM, P < 0.02) in this measure, there were no significant differences in the maximum enhancement (efficacy) of [3H]MK-801 binding by these compounds. Since glycine appears to be required for the operation of NMDA-coupled cation channels, we examined the effects of glycineand ACPC on NMDA-induced acute excitotoxicity in the 14day embryonic chick retina. Histological evaluation of retina revealed that either ACPC (10-100/~M) or glycine(200 #M) attenuated NMDA-induced (200 #M) retinal damage, and a combination of these agents produced an enhanced protection against acute NMDA toxicity. ACPC (100/aM), but not MK-801 (1 #M) also afforded a modest protection against kainate-induced (25 #M) retinal damage. These findings demonstrate that while strychnine-insensitive glycine receptors are present in embryonic chick retina, occupation of these sites does not augment the cytotoxic actions of NMDA. Moreover, the ability of ACPC and glycine to attenuate NMDA-induced cytotoxicity does not appear to be mediated through occupation of these sites.

The excitotoxic actions of glutamate have been implicated in the neuronal damage associated with diverse pathophysiological states including hypoglycemia (Wieloch, 1985), brain trauma (Mclntosh et al., 1989), cerebral ischemia (Simon et al., 1984), hypoxia (Goldberg et al., 1987) and epilepsy (Croucher et aL, 1982 ; Stasheff et al., 1989). Converging lines of evidence indicate that activation of the N-methyl-D-aspartate (NMDA) subtype of glutamate receptor significantly contributes to the acute excitotoxic actions of glutamate by increasing intracellular cations (Na +, Ca 2÷) and the passive diffusion of CI- and water (Olney et al., 1986a ; Choi, 1987). Electrophysiological and neurochemical evidence indicate that N M D A receptor-coupled cation channels can be modulated by glycine (Johnson and Ascher, 1987; Kleckner and Dingledine, 1988; Reynolds et al., 1987 ; Ransom and Stec, 1988). This

*Author to whom correspondence should be addressed.

effect is mediated through strychnine-insensitive glycine receptors with a neuroanatomical (Bristow et al., 1986) and pharmacological (Johnson and Ascher, 1987; Galli et al., 1988) profile distinct from strychnine-sensitive glycine receptors located primarily in brainstem and spinal cord (Zarbin et al., 1981). While glycine was initially reported to potentiate NMDAgated channel opening (Johnson and Ascher, 1987), recent findings indicate that glycine is required for activation of NMDA-gated cation channels expressed in X e n o p u s oocytes (Kleckner and Dingledine, 1988) and in primary cultures of cortical neurons (Huettner, 1989). The proposed requirement for glycine to function as a co-agonist in the operation of N M D A receptorcoupled cation channels has important ramifications for both pathophysiology and clinical therapeutics. Nonetheless, elevation of extracellular glycine does not uniformly result in an augmented responsiveness of the NMDA receptor complex (reviewed in Thomson, 1990), which has been attributed to both satu473

474

K. M, BoJI el a[,

ration o1" s t r y c h n i n e - i n s e n s i t i v e glycine r e c e p t o r s in situ a n d a differential r e g u l a t i o n o f N M D A r e c e p t o r s u b t y p e s by glycine. T h e e m b r y o n i c chick retina has been used as a m o d e l to s t u d y g l u t a m a t e - i n d u c e d acute cellular d a m age (Olney et al., 1986a,b ; R o t h m a n a n d Olney, 1987). This tissue a p p e a r s to c o n t a i n multiple g l u t a m a t e receptors since a p p l i c a t i o n o f subtype-specific a g o n ists (e.g. N M D A , kainate) results in cellular d a m a g e restricted to m o r p h o l o g i c a l l y defined areas (Olney et al., 1986a,b; R o t h m a n a n d Olney, 1987). While glycinergic n e u r o n s have been described in v e r t e b r a t e retina (reviewed in M a r c , 1988), to out" k n o w l e d g e n e i t h e r the presence o f s t r y c h n i n e - i n s e n s i t i v e glycine r e c e p t o r s n o r the effects o f glycinergic ligands o n N M D A - i n d u c e d toxicity has been d e m o n s t r a t e d in e m b r y o n i c chick retinas. T h e p r e s e n t study was initiated to d e t e r m i n e w h e t h e r s t r y c h n i n e - i n s e n s i t i v e glycine r e c e p t o r s are p r e s e n t a n d c o u p l e d to N M D A gated cation c h a n n e l s in this tissue, a n d w h e t h e r N M D A - i n d u c e d n e u r o t o x i c i t y could be m o d u l a t e d by glycinergic ligands.

EXPERIMENTAL

PROCEDURES

MaleriaL~ [3H]MK-801 (sp. act. 24.8 Ci/mmol) was purchased from Du Pont NEN (Boston, MA). [3H]glycine (sp. act. 30 Ci/mmol) was purchased from American Radiolabelled Chemicals Inc. (St Louis, MO). I-Aminocyclopropanecarboxylic acid (ACPC) was purchased from Fluka Chemical Co. (Ronkonkoma, NY). Retinas were obtained from embryonic day 14 chickens (Gallus domesticus) supplied by Truslow Farms, Chestertown, MD. Other reagents were purchased from Calbiochem (San Diego, CA) or Sigma (St Louis, MO). Neurochemixlry Radioliyand bindin,q studies. Isolated retinas were frozen on dry ice and stored at - 7 0 ' C until assayed. Retinal tissues were homogenized and extensively washed to removed endogenous glutamate and glycine. Retinas were disrupted in 25 vol of 5 mM HEPES/4.5 mM Tris buffer (pH 7.4) using a Brinkman polytron (setting 6.5). The homogenized tissue was centrifuged at 20,000g for 20 rain. The resultant pellet was resuspended in fresh HEPES/Tris buffer and recentrifuged. The pellet was subsequently washed twice in HEPES/ Tris buffer containing 1 mM EDTA, followed by two additional washes with fresh HEPES/Tris. The pellet was resuspended in 5 vol of HEPES/Tris buffer and stored at - 70~C. On the day of the assay, the tissue was washed an additional three times using fresh HEPES/Tris buffer. The effects of glycine, ACPC, glutamate or N M D A on [3H]MK-801 binding were studied under nonequilibrium conditions. Compounds were added to assay tubes containing 8 nM [*H]MK-801 in HEPES/Tris buffer. Total and nonspecific binding were determined in the absence of added drugs and presence of 200 #M phencyclidine, respectively.

Nonspecific [3H]MK-801 binding was 54% of total basal (unstimulated) binding. Glycine, ACPC, and strychnine were examined for their effects on [~H]glycine binding. Nonspecilic [3H]glycine binding was 47'7o of total basal binding. Conrpounds were added to assay tubes containing 40 nM ['H]glycine in HEPES/Tris buffer. Nonspecific binding was defined with 1 mM ACPC. Assays were initiated by addition of retinal homogenates (0.37- 0.51 mg of protein) in a final volume of 0.5 ml for 2 h at 25 C. The reactions were terminated by vacuum filtration (M-24R, Brandcl Instruments, Gaithersburg, MD) through GF/B filters (for [~tl]MK-801 binding) or GF/C filters (for [~H]glycine binding) which were presoaked in 0.3% polyethylenimine, Filters were washed twice with ice cold HEPES/Tris buffer, dried and the radioactivity retained on the filters was measured in a Beckman Instruments LS 5801 liquid scintillation counter, i'-aminobutyric acid release. The incubation protocol for 7-aminobutyric acid (GABA) release was performed as described by Zccvalk eta/. (1989). In brief, isolated retinas were preincubated in 1 ml of buffer or buffer with ( ~ ) M K 801 (10 t~M) or ACPC (5 500 izM) for 10 min. The buffer consisted of 135 mM NaCI, 5.0 mM KC1, 0.5 mM CaCI> 4.5 mM MgCI,, 5.6 mM glucose and 25 mM Tris HCI. pH 7.5. The preincubation buffer was exchanged with I ml of fresh buffer (with or without pharmacologic agentsl jusl prior to the addition of 200 itM NMDA. The retinas were then incubated for an additional 30 min. Incubations were terminated by removing the incubation buffer which was frozen on dry ice and stored at - 2 0 C for assay. Retinal tissue was resuspended in 1 ml of distilled water and frozen tot protein determination (Lowry eta/., I951). Samples of incubation buffer were deriwitized with o-phthaldialdehydc (OPA) for quantitation at 340 nm by UV-HPLC. Brietly. 900 tzl aliquots of the incubation buffer were lyophilized to a volume of 300 Id and derivatized with 500 Itl of OPA solution (prepared as a I : 4 dilution of a stock solution of 27 nag OPA, 5 I~1 2-mcrcaptoethanol in l ml methanol and 9 ml of 0.1 M sodium borate buffer, pH 9.0). Two minutes later, samples were centrifuged and then assayed via IIPLC. Samples were chromatographed on an Alltech Adsorbosphere C18 31t column, with a mobile phase of 18% acetonitrile/82% 0.1 M phosphates buffer, pH 6.0. The retention time for GABA was 17 rain. The minimum detectable amount of GABA was 1 nmol injected onto the column. Other known amino acids, including ACPC, did not chromatographically interfere with GABA. The amount ol released GABA was quantitated b~r the external standard method. Cyclic G M P levels were determined in retinas dissected and incubated in 2 ml of buffer containing ( + )MK-801 ( 1 ItM ) or ACPC (10 or 100 ItM) for 5 rain prior to the addition of N M DA (200 #M). At the end of the incubation period (I or 5 min), 0.5 ml 20% perchloric acid was added to each retina. The tissues were immediately sonicated and an aliquot reserved for protein determination (Lowry el al., [951). The homogenates were centrifuged and an aliquot of the supernatant was neutralized with 5 N KOH. The samples were centrifuged and aliquots of the supernatant lyophilizcd and assayed for cyclic G M P (Steiner et al., 1972).

Tissue morphology Retinas were obtained from the eyes of embryonic chickens (day 14) by removing the cornea and vitreous and then gently shaking the eye cups in Dulbecco's phosphate buffered

Strychnine-insensitive glycine receptors in retina saline without calcium or magnesium. Retinal tissues were incubated at 37°C under a continuous flow of 95% oxygen, 5% CO2 in an apparatus described by Wetzel et al. (1989). Retinas were preincubated in buffer or buffer containing ACPC (0.1 100/aM), glycine (200 #M), or (+)-MK-801 (1 /~M) for 5 min prior to the addition of either NMDA (200 /~M) or kainic acid (25/~M). The retinas were then incubated for an additional 30 min. The incubation was terminated by decanting the medium, flooding the retinas with 10% phosphate buffered formalin, decanting, and addition of 5 ml of the fixative. After dehydration in ethyl alcohol the tissues were embedded in glycol methacrylate, sectioned (2.0 #m) and stained with hematoxylin eosin for light microscopic evaluation. The histologic damage was qualitatively assessed by an observer blinded to the experimental protocol. The severity of retinal damage was estimated from the central-most region of the retinal cross section. The extent of damage was visually graded according to the following scale : 0, no damage ; 1 +, mild damage ; 2 + , moderate damage ; 3 + , severe damage; and 4 +, extreme damage. Evaluation of the retinal damage was based on the following criteria : (1) cytoplasmic appearance, i.e. eosinophilic staining, cytoplasmic vacuoles; (2) nuclear appearance, i.e. basophilic staining, nuclear vacuoles, chromatin clumping/pyknosis, prominent nucleoli ; and (3) degree of interstitial edema.

Data reduction and statistics The modulation of [3H]MK-801 binding under nonequilibrium conditions was expressed as a ratio of specifically bound [3H]MK-801 in the presence and absence of drug modulator (stimulated binding/basal binding). Radioligand binding data were analyzed via nonlinear regression using GRAPHPAD (ISI). Statistical significance (P < 0.05) was determined by t-test or analysis of variance (ANOVA) when appropriate. Data are expressed as mean + SEM. RESULTS

475

Table 1. Effects of glycine and ACPC on [3H]glycinedisplacement and [3H]MK-801 functional binding assays

Glycine ACPC

Enhancement of [3H]MK-801 binding

Inhibition of [3H]glycine binding IC~o(nM)

ECs0 (nM)

Stimulation ratio (Em,~/Basal)

247 _+39 271 + 39

256 _+57 54 _+12*

1.32_+0.02 1.30+ 0.05

Retinal homogenates were incubated with [3H]glycine (40 nM) or [3H]MK-801 (8 nM) at 25'C for 2 h as described in Experimental Procedures. Values are the X±SEM of 3 4 experiments. Parameter estimates were obtained with GraphPad. * P < 0.05 compared to glycine, Student's t-test.

et al., 1989; N a d l e r et al., 1988; W a t s o n a n d L a n t h o r n , 1990)] potently e n h a n c e d [3H]MK-801 binding to sites within the N M D A - c o u p l e d cation channel, a n d inhibited strychnine-insensitive [3H]glycine binding to retinal h o m o g e n a t e s (Table 1 ; Fig. 1). A C P C was significantly more p o t e n t t h a n glycine in the former measure, but exhibited a similar maxim u m e n h a n c e m e n t o f [3H]MK-801 binding (Table 1). [3H]glycine binding was inhibited by b o t h glycine a n d A C P C with similar ICs0 values (Table 1). Strychnine (100 # M ) did n o t inhibit [3H]glycine binding (data not shown). Figure 2 illustrates the e n h a n c e m e n t of [3H]MK-801 binding by a c o m b i n a t i o n of glutamate a n d glycinergic ligands. While each agent elicited a modest ( ~ 2 5 - 4 0 % ) e n h a n c e m e n t of [3H]MK-801 binding, c o m b i n a t i o n s of these agents produced less t h a n additive effects.

Radioliyand binding

G A B A release

Glycine a n d A C P C [a specific, high affinity ligand at strychnine-insensitive glycine receptors ( M a r v i z d n

I n c u b a t i o n of retinas with N M D A (200 FtM) produced a significant release of G A B A into the incu-

t.6

100

1.4

Q Q

t.3 1.2

1.1 t.0 0.9

-B

-7

n

I

-6 -5 Log [M)

-4

0

B

-9 -8 -7 -6 -5 -4 Log [14]

Fig. 1. (A) Enhancement of [3H]MK-801 binding to retinal membranes by glycine or ACPC. [3H]MK-801 binding to extensively washed retinal membranes was specifically displaced by 200 #M (+)MK-801, TCP (N-[1-(2-thienyl)cyclohexyl]piperidine) or PCP (phencyclidine). The IC50 for displacement of [3H]MK-801 by (+)MK-801 was 14+2 nM (mean+SEM, n = 2). (B) Inhibition of [3H]glycine binding by glycine or ACPC. These representative experiments were replicated 3~1 times with similar results. See Table 1 for a summary of parameter estimates. Symbols : I--1,glycine, 0 , ACPC.

476

K.M. BoJE et ul. 5

-,'4

2.0

oc=t.

4

T

,p~

T T TIIT T

1.5

~ - . -",

--r-

3

r..._

~, 2 =o. r..,3

~.~

0.5

=

0.0

ncl

ACPC GIy

G1u

ACPC GIy 1.

-I-

Glu

G1u

Fig. 2. Enhancement of [3HJMK-80! binding by glutamate and glycine: retinal membranes were incubated in the presence (stimulated) or absence (basal) of I0 #M ACPC, glycine (Gly) and/or glutamate (Glu). Data are the X+SEM of 4 7 experiments performed in duplicate.

bation medium (2.56 + 0.24 nmol/ml/mg protein, n = I1) compared with controls (GABA not detected, n = 9) (Fig. 3). ACPC (l 100 #M) did not significantly modulate NMDA-stimulated GABA release, although a very high concentration (500 #M) diminished release by 33%. This reduction was not statistically significant (ANOVA). In contrast, 10 /~M MK-801 completely prevented NMDA-induced release of GABA.

Qvclic GMP accumulation NMDA did not elevate cyclic GMP in the embryonic chick retina, either in the presence or absence of Mg 2+ (Table 2).

Histolo,qy Within 15 min after exposure to NMDA (200/~M), damage to the inner nuclear layer, outer plexiform layer, ganglion cell layer, and outer nerve fiber layer of the 14 day embryonic chick retina was observed (Fig. 4b). These changes were consistent with hydropic degeneration, with characteristic changes including: (a) swollen, abnormally pale basophilic staining nucleus (karyolysis), (b) chromatin clumping, (c) vacuolar degeneration of both the nucleus and cytoplasm, and (d) interstitial edema. The damage was limited primarily to the ganglion cell layer and the inner half of the inner nuclear layer. Interstitial edema, evidenced by characteristic pate eosinophilic staining, was observed in both the outer nerver fiber layer and inner plexiform layer. NMDA-induced

NMOA (200 W,M)

-

MK-801 (10 #M)

-

nd 1.

Jr 1.

1. .

1. .

.

t" .

1.

t"

.

ACPC (#M)

5 10 50 100 500 Fig. 3. N MDA-induced release of endogenous GABA from 14 day embryonic chick retina. Retinas were preincubated with the indicated (+) concentrations of ACPC or MK-801 for 5 rain prior to exposure to NMDA (200/iM) as described in Experimental Procedures. Values are the X_+SEM of 4 I 1 samples. Key : nd, not detectable. damage appeared to progress l¥om the inner limiting membrane toward the outer limiting membrane, with a line of demarcation within the center inner nuclear layer which runs parallel to the outer limiting membrane. Time-course studies indicated that the line of demarcation was not a diffusion artifact, since incubation periods of up to 60 min caused no further increase in morphological damage (data not shown). Preincubation with MK-801 (I /~M) blocked the cell damage induced by a 30 rain exposure of NMDA (Fig. 4f). ACPC (100 ILM; Fig. 4c) decreased the severity of the NMDA-induced damage in a concentration dependent manner (10-100 #M; Fig. 5b and c). Severe NMDA-induced cellular damage was observed at each concentration of ACPC, but qualitative differences were present in the extent of" interstitial edema present (Fig. 5). No attenuation of the histological damage produced by NMDA was evident at concentrations of ACPC < 5 FzM. In the absence of NMDA, no noticeable cellular changes were apparent in retinas incubated with ACPC (10 and 100 ~zM) for 30 min (data not shown). Preincubation of retinas with glycine (200/~M) also decreased the severity of NMDA-induced retinal damage (Fig. 4d). The reduction in damage appeared similar to that observed with ACPC (100/~M). Incubation of retinas with glycine (200 #M) and ACPC (100 #M) prior to NMDA resulted in less interstitial edema (Fig. 4e) than was observed with either glycine or ACPC (compare Fig. 4c and d ; Table 3). Incubation of embryonic chick retinas with kainic

Strychnine-insensitive glycine receptors in retina

477

Table 2. Effects of MK-801 or ACPC on NMDA-mediated cyclic GMP accumulation in the embryonic chick retina Cyclic GMP concentrations (pmol/mg protein) Length of incubation (min)

With Mg2+

Without Mg2+

Control NMDA (200 #M) NMDA (200/zM) and MK-801 (1 #M) ACPC (10 gM) ACPC (100 #M)

1 I

0.43 ± 0.05 0.46±0.11

0.35 ± 0.16 1.00_+0.56

1 1 1

0.63±0.08 1.68± 1.44 1.05 ± 0.79

0.30±0.15 0.62 ± 0.20 0.62 + 0.45

Control NMDA (200 pM) NMDA (200 pM) and MK-801 (1 #M) ACPC (10/iM) ACPC (100 #M)

5 5

0.83 ± 0.68 0.78 ± 0.06

0.41 ___0.18 0.81 ±0.33

5 5 5

0.36±0.08 1.32± 1.00 1.28_+ 1.04

0.37±0.09 0.72 ±0.25 0.51 ±0.21

Treatment

Values are the X _ SD of quadruplicate samples, No statistical significance among the treatments was detected by Student's t-test with Bonferroni's correction for multiple comparisons.

acid (25/~M) p r o d u c e d a different p a t t e r n o f d a m a g e c o m p a r e d to N M D A (Fig. 6). D a m a g e was c o n f i n e d to the i n n e r n u c l e a r layer a n d was d i s t r i b u t e d t h r o u g h o u t this layer (Fig. 6b). A p p r o x i m a t e l y 5 0 % o f the cells a p p e a r e d to be d a m a g e d by this c o n c e n t r a -

tion o f kainic acid. T h e d a m a g e was characteristic o f h y d r o p i c d e g e n e r a t i o n , b u t only mild interstitial e d e m a was observed. P r e i n c u b a t i o n with M K - 8 0 1 (1 # M ) did n o t affect k a i n a t e - i n d u c e d cellular d a m a g e while p r e i n c u b a t i o n A C P C (100 /~M) p r o d u c e d a

Fig. 4 ~ C a p t i o n on p. 480.

GO:

t

~J

J_.

~4

J~

480

K . M . BoJF et al.

,

,

Fig. 4. Effects of glycine, ACPC, and MK-801 on NMDA-induced cytotoxicity, ta) Normal 14 day embryonic chick retina. Arrow, inner limiting membrane ; O, optic nerve fiber layer : G, ganglion cell layer : IP, inner plexiform layer; IN, innm nuclear layer; OP, outer plexiform layer; ON, outer nuclear layer. (b) Hydropic degeneration in the 14 day embryonic retina incubated with NMDA (200 /tM) for 30 min. Characteristic changes included both karyolysis with chromatin clumping (solid arrow) and severe interstitial edema (open arrows). A line of demarcation was observed in the center of the inner nuclear layer (IN). (c) and (d) Although still severe in nature, the extensive damage induced by NMDA was modified by preincubation with both ACPC (100 ~tM) (c) and glycine (200 #M) (d). Milder interstitial edema was evidenced by the deeper eosinophitic staining of both the inner plexiform and ganglionic cell layers (arrows). (e) A combined preincubation with both ACPC (100 #M) and glycine (200 #M) further reduced the extent of NMDA induced retinal damage. Karyolysis and chromatin clumping (arrows) was still evident, but the degree of interstitial edema present was less in all retinal layers. (f) Preincubation with M K-801 (I /,M) prevented NMDA-induced cellular damage, x 80

modest improvement, best evidenced in a darker, more uniform basophilic staining of the inner nuclear layer (Fig. 6c).

DISCUSSION The N M D A subtype of glutamate receptor is coupled to a cation channel and a strychnine-insensitive glycine receptor in many regions of the central nervous system (Monaghan et al., 1988 ; Jansen et al., 1989; reviewed in Thomson, 1990). Neurochemical and electrophysiological evidence strongly suggest that N M D A - g a t e d cation channels can be modulated by ligands acting at a strychnine-insensitive glycine

receptor (Johnson and Ascher, 1987; Kleckner and Dingledine, 1988 ; Reynolds et al., 1987 ; Ransom and Stec, 1988). Pharmacological (Olney et al., 1986b, 1987 ; Rothman and Olney, 1987), electrophysiological (Slaughter and Miller, 1983 ; Massey and Miller, 1990), and neurochemical (Anand et al., 1985: Somohano et al., 1988) studies have demonstrated that the N M D A subtype of glutamate receptor and its associated cation channel are present in vertebrate retina. While glycine receptors have been described in the retina (reviewed in Marc, 1988), to our knowledge, the presence of a strychnine-insensitive glycine receptor functionally coupled to the " N M D A receptor complex" has not been previously described in this tissuc.

Strychnine-insensitive glycine receptors in retina

Fig. 5--Caption overleaf.

481

482

K . M . BOJE el o1.

C

Fig. 5. Effects of ACPC on NMDA-mduced cytotoxicity. ,aJ Fourteen day embryomc chick retinas were incubated with 200 I~M NMDA as described in Experimental Procedures. Severe interstitial edema was present, extending from the inner limiting membrane (solid arrow) to the center of lhe inner nuclear layer (open arrow). (b) Preincubation with 10 #M ACPC slightly diminished the extent of interstitial edema. best observed in the ganglion cell layer (G). (c) Pre-incubation with 100 #M ACPC produced only a small improvement in the reduction of interstitial edema compared to lower concentrations of ACPC. This improvement was evident in the inner nuclear layer (IN), manifested by a deeper, more uniform basophilic staining. × 80

Table 3. Effects o f MK-801, glycine and A C P C on N M D A - i n d u c e d cytotoxicity Drug MK-801 (1 a M ) Glycine (200 `aM) A C P C (t00 # M )

Retinal d a m a g e 0 0 0

N M D A (200`aM) NMDA+ACPC NMDA+glycine N M D A + A C P C + glycine N M D A + MK-801

4 ~ I ± v + ~ {~- 40

Kainic acid (25 ,uM) Kainic a c i d + M K - 8 0 1 Kainic a c i d + A C P C

~4 + + + + ~- + +. ~ +

Retinas from embryonic day 14 chicks were incubated with the indicated c o m p o u n d s as described in Experimental Procedures. The extent o f retinal d a m a g e was assessed and described by the following scale: 0, no d a m a g e : +. mild; ~ + , m o d e r a t e : f ~ ~ . severe: + ~ 4 + , extreme damage.

The current findings demonstrate the presence of strychnine-insensitive glycine receptors which are allosterically linked to N M D A - r e c e p t o r gated ion channels in embryonic chick retina. Both A C P C and glycine enhanced pH]MK-801 binding to sites presumably located within the N M D A - c o u p l e d cation channel, with potencies (Table 1 ; Fig. l a) comparable to values previously reported in extensively washed brain membranes (Marviz6n et al., 1989 ; H o o d et al., 1990). Moreover, both the potencies of A C P C and glycine to inhibit [3H]glycine binding (Fig. l b) and the lack of effect of strychnine (data not shown) are also consistent with previous studies of strychnine-insensitive glycine receptors in brain (Marviz6n et al., 1989 : Hood et al., 1990). Nonetheless, there are several differences between embryonic chick retina and brain with respect to the effects of glycinergic ligands. The maximum enhance-

Strychnine-insensitive glycine receptors in retina

Fig. ~-Caption overleaf

483

484

K . M . BoJE eta/.

Fig. 6. Effects of ACPC on kainate-induced cytotoxicity. (a) Normal 14 day embryonic chick retina. Note the uniform basophilic staining of the inner nuclear layer (IN) and the absence of nuclear vacuoles. (b) Kainic acid (25 #M) produced a pattern of damage which was confined primarily to the inner nuclear layer. The damage was characterized by mild interstitial edema (open arrow) and hydropic degeneration (solid arrow) of the layer. (c) Preincubation with ACPC (100 ~M) afforded a modest improvement, best evidenced by the darker, more uniform basophilic staining of the inner nuclear layer. Chromatin clumping and numerous nuclear vacuoles are still present, x 80 ment in [3H]MK-801 binding produced by glycine, ACPC and glutamate was significantly lower in retinal membranes than previously reported using comparably prepared adult rat brain membranes (Marviz6n et al., 1989). Moreover, combinations of glycine or ACPC plus glutamate resulted in a less than additive effect on [3H]MK-801 binding in retina (Fig. 2) but synergistically enhanced [3H]MK-801 binding in brain (Marviz6n et al., 1989 ; Ransom and Stec, 1988). Finally, ACPC appears to act as a glycine partial agonist in both neurochemical (Marviz6n et al., 1989) and electrophysiological measures (Watson and Lanthorn, 1990) in brain, but it enhanced [3H]MK-801 binding similar to that of glyeine in retina (Fig. l a ; Table 1). Several hypotheses could account for the differences observed between retinal and brain membranes in the strychnine-insensitive glycinergic modulation of the N M D A receptor complex. Regional variations in the presumed glycine and/or glutamate subunit number

or composition between brain and retina could account for these differences. Alternatively, changes in the rates of[3H]MK-801 association in the presence and/or absence of glycinergic drugs could also explain the apparent modest enhancement of [3H]MK-801 binding in retina compared to brain. [3H]MK-801 binding to extensively washed tissues is generally performed under non-equilibrium conditions due to the relatively slow basal association rate of this ligand (Kloog et al., 1988). The increase in [3H]MK-801 binding produced by glycine and glutamate is primarily reflected by an increase in the rate of ligand association (Kloog et al., 1988). Hence, differences in the kinetics of [3H]MK-801 binding in the presence or absence of glyeinergic drugs could also account for the apparent modest glycinergic enhancement of [3H]MK-801 binding in retina relative to brain. Consistent with previous findings (Otney et al., 1986a,b; Zeevalk et al., 1989), a 30 min incubation with N M D A (200/~M) produced extensive damage to

485

Strychnine-insensitive glycine receptors in retina the embryonic chick retina (Figs 4 and 5; Table 2). The damage to the inner nuclear, outer plexiform, ganglion cell and outer nerve fiber layers is compatible with a mechanism involving hydropic degeneration, as previously noted by others (Olney et al., 1986a,b). This hydropic degeneration observed during the acute phase of glutamate toxicity is a result of an unregulated passive influx of CI- ionically balanced by Na ÷ influx and accompanied by large increases in cell water (Olney et al., 1986a; Zeevalk et al., 1989). Moreover, the inability of N M D A to significantly elevate cyclic GMP levels in embryonic retina (see results) is consistent with ion substitution experiments (Olney et al., 1986a) which demonstrated that Ca 2÷ was not required for N M D A to produce histological damage in the embryonic chick retina. In addition to the histological damage evident on short term exposure to NMDA, Zeevalk et al. (1989) and Zeevalk and Nicklas (1989) reported that N M D A produced a concentration-dependent release of GABA from embryonic chick retina. Since this biochemical measure is thought to directly reflect (but is more readily quantified than) histological damage (Zeevalk et al., 1989), we examined the effects of MK801 and ACPC on NMDA-induced GABA release. A very significant reduction in both NMDA-induced histological damage and GABA release (Fig. 4f; Table 3 and Fig. 3) was observed with MK-801. While ACPC reduced NMDA-induced histological damage in a concentration dependent fashion (Figs 4c, 5b and c), a marginal decrease of NMDA-induced GABA release occurred only with a high ACPC concentration (500 /~M). It is possible that the lack of agreement between the early histopathological changes and the NMDA-induced GABA release reflects a poor correlation between early hydropic changes and terminal cytotoxicity. Since glycine can augment the excitotoxic actions of NMDA in cultured cortical neurons (Paten et al., 1990), and ACPC is as potent (and efficacious) as glycine at strychnine-insensitive glycine receptors in retina (Table 1), it was predicted that ACPC would have either augmented the histological damage elicited by NMDA in retina or have no effect if extracellular glycine concentrations were sufficient to occupy strychnine-insensitive glycine receptors. However, ACPC reduced NMDA-induced cell damage in a concentration dependent fashion (Figs 4 and 5). If the biochemical measures in retinal tissue suggest that ACPC is a glycinergic agonist, then why would ACPC paradoxically reduce NMDA-induced cytotoxicity? There are several possible explanations. ACPC has been shown to exhibit partial agonist prop-

erties with a reported efficacy of ~ 0.6-0.9 in various neurochemical and electrophysiological preparations (Marviz6n et al., 1989 ; Watson and Lanthorn, 1990). It could be reasoned that the modest maximum ACPC enhancement of [3H]MK-801 binding in 14 day embryonic chick retina ( ~ 3 0 % [Table 1 ; Figs 1 and 2] compared to 147% in adult rat forebrain under similar incubation conditions) would make it difficult to detect a statistically significant partial agonist action. Thus, the discrepancy between the biochemical and histopathological findings might be due to insufficiently sensitive biochemical measures. Alternatively, the mechanism(s) of ACPC modulation of NMDA-induced cytotoxicity may not involve strychnine-insensitive glycine receptors. This hypothesis is supported by (a) the attenuation of NMDA-induced damage by glycine (Fig. 4d), (b) the additive attenuation produced by a combination of ACPC and glycine (Table 1 ; Fig. 4e), and (c) the finding that ACPC (but not MK-801) also produced a modest improvement in kainate-induced damage (Fig. 6c). The mechanism(s) through which ACPC and glycine attenuate NMDA-induced excitotoxicity in embryonic chick retina is unknown. However, while this report was in preparation, Weinberg et al. (1990) reported that glycine protected kidney proximal tubules against hypoxic injury. Among more than 45 amino acids and analogs examined, only glycine, ACPC, and L,D-alanine were active in this measure. While the concentrations used in the former study were more than one order of magnitude higher than used here to attenuate NMDA-induced cell damage, these findings indicate the possibility of a common mechanism that would not necessarily involve an extracellular receptive process, but could nonetheless prove important in the regulation of hypoxic insult. Acknowledyement--K.M.B. was a Pharmacology Research

Associate of the National Institute of General Medical Science during these studies. REFERENCES

Anand H., Roberts P. J. and Lopez-Cotome A. M. (1985) Excitatory amino acids in the chick retina: possible involvement of cyclic guanosine monophosphate. Neurosci. Left. 58, 31-36. Bristow D. R., Browery N. G. and Woodruff G. N. (1986) Light microscopic autoradiographic localization of [3H]glycine and [3H]strychnine binding sites in rat brain. Eur. J. Pharmac. 126, 303-307. Choi D. W. (1987) Ionic dependence of glutamate neurotoxicity in cortical cell culture. J. Neurosci. 7, 369-379. Croucher M. J., Collins J. F. and Meldrum B. S. (1982) Anticonvulsant action of excitatory amino acids antagonists. Science 216, 899401.

486

K . M . BoJl- el a].

Galli A., Fenigli S., Anichini A. and Pizzighelli L. (1988) Stereoselective inhibition of [3H]glycine binding to cortical membranes of rat brain by amino acids. Pharmac. Res. Conlmun. 20, 407-408. Goldberg M. P., Weiss J. H., Pham P.-C. and Choi D. W. (19871 N-methyl-D-aspartate receptors mediate hypoxic neuronal injury in cortical culture. J. Pharmac. exp. Ther. 243, 784 791. Hood W., Compton R. and Monahan J. (1990) N-methylI)-aspartate recognition site ligands modulate activity at the coupled glycine recognition site. J. Neurochem. 54, 1046 1046. ttuettner J. E. (1989) Indole-2-carboxylic acid : a competitive antagonist of potentiation by glycine at the NMDA receptor. Science 243, 1611 1613. Jansen K. L. R., Dragunow M. and Faull R. L. M. (19891 ['H]Glycine binding sites, NM DA and PCP receptors have similar distributions in the human hippocampus : an autoradiographic study. Brain Res. 482, 174 178. Johnson J. W. and Ascher P. (1987) Glycine potentiates the NMDA response in cultured mouse brain neurons. Nature 325, 529 531. Kleckner N. W. and Dingledine R. (1988) Requirement for glycine in activation of NMDA-receptors exprcssed in .Venopus oocytes. Science 241,835 837. Kloog Y., Nadler V. and Sokolovsky M. (1988) Mode of binding of [3H]dibenzocycloalkenimine (MK-801) to the N-methyl-D-aspartate (NMDA) receptor and its therapeutic implication. FEBS Lett. 230, 167 170. Lowry O. H., Rosebrough N. J., Farr A. L. and Randall R. J. (1951) Protein measurement with the Folin phenol reagent. J. biol. Chem. 193, 265 275. Marc R. E. (19881 The role of glycine in the mammalian retina. In : Progress in Retinal Research (Osborne N. and Chader G., eds), pp. 67 107. Pergamon Press, Oxlbrd. Marvizdn J. C. G., Lewin A. and Skolnick P. (1989) lAminocyclopropane carboxylic acid : a potent and selective ligand for the glycine modulatory site of the N-methylD-aspartate receptor complex. J. Neurochem. 52, 992 994. Massey S. C. and Miller R. F. (199(/) N-methyl-D-aspartate receptors of ganglion cells in rabbit retina. J. Neurophysiol. 63, 16 30. Mclntosh T. K., Vink R., Soarcs H., Hayes R. and Simon R. (1989) Effects of the N-methyl-D-aspartate receptor blocker MK-801 on neurologic function after experimental brain injury. J. Neurotrauma 6, 247 259. Monaghan D. T., Olverman H. J., Nguyen L., Watkins J. C. and Cotman C. W. (1988) Two classes of N-methyl-oaspartate recognition sites: differential distribution and differential regulation by glycine. Proc. mtm Acad. Sci., U.S.A., 85, 9836 9840. Nadler V., Kloog Y. and Sokolovsky M. (19881 l-aminocyclopropane-l-carboxylic acid (ACC) mimics the effects of glycine on the NMDA receptor ion channel. Eur. J. Pharmac. 157, 115 116. Olney J., Price M., Samson L. and Labruyere J. (1986a) The role of specific ions in glutamate neurotoxicity. Neurosci. Left. 65, 65 71. Olney J., Price M., Fuller T., Labruyere J., Samson L., Carpenter M. and Mahan K. (1986b) The anti-excitotoxic effects of certain anesthetics, analgesics, and sedativehypnotics. Neurosci. Lett. 68, 29 35. Olney J., Price M., Salles S., kabruyere J. and Friedrich

G. (1987) MK-801 powerfully protects against N-methyl aspartate neurotoxicity. Eur. J. Pharmae. 141, 357 36t. Patel J., Zinkand W. C., Thompson C.. Keith R. and Salama A. (1990) Role of glycine in the N-methyl-o-aspartatemediated neuronal cytotoxicity. J. Neurochem. 54, 849 854. Ransom R. W. and Stec N. g. (1988) Cooperative modulation of[~H]M K-801 binding to the N-methyl-D-aspartate receptor-ion channel complex by L-glutamate, glycine and polyamines. J. Neurosci. 51,830 836. Reynolds 1. J., Murphy S. N. and Miller R. J. (19871 3Hlabelled MK-801 binding to the excitatory amino acid receptor complex from rat brain is enhanced by glycme. Proc. natn Acad. Sci., U.S.A. 84, 7744 7748, Rothman S. M. and Olney J. W. (1987) Excitotoxicity and the NMDA receptor. Trends Neurosci. 10, 299 302. Simon R. P., Swan J. H., Griffiths T. and Meldrum [3. S. (1984) Blockade of N-melhyh)-aspartate receptors may protect against ischemic damage in the brain. Science 226, 850 852. Slaughter M. M. and MiIler R. F. ([983) The role of excitatory amino acid transmitters m the mudpuppy retina: analysis with kainic acid and N-methyl aspartate. J. NeluO.~ci. 3, 1701 1711. Somohano F., Roberts P. J. and Lopez-Colome A. M. (1988) Maturational changes in retinal excitatory amino acid receptors. Devl. Brain Res. 42, 59 67. Stasheff S. F., Anderson W. W., Clark S. and Wilson W. A. (1989) NMDA antagonists differentiate epileptogenesis from seizure expression in an in citro model. Science 245, 648 6511. Steiner A., Parker C. and Kipnis D. (1972) Radio-m~munoassay for cyclic nucleotides. ,I. hiol. Chem. 247, 1106 1113. Thomson A. M. ( 19901 Glycine is a coagonist at the NMDA receptor/channel complex. Pro#. Neurobiol. 35, 53 74 Watson G. and Lanthorn T. (1990) Pharmacological characteristics of cyclic homologues of glycine at the N-methylE~-aspartate receptor associated glycine site. Neuropharmacoh)y), 29, 727 730. Wciloch T. (19851 Hypoglycemia-induced neuronal damage prevented by an N-methyl-i)-aspartate agonist. Sciettce 230, 681 683. Weinberg J.. Venkatachalam M., Garzo-Quintcro R., Roeser N. and Davis J. (1990) Structural requirements for protection by small amino acids against hypoxic injury in kidney proximal tubules. FASEB J. 4, 3347 3354. Wetzel M., Fahlman C., Alligood J., O'Brien P. and Aguirrc G. (1989) Metabolic labeling of normal canine rod outer segment phospholipids in t,h:o and in vitro. £x'p. Eve Res. 48, 149 160. Zarbin M., Wamsley J. and Kuhar M. (1981) Light microscopic autoradiographic localization with [3H]strychnine. .I. Neurosci. 1,523 547. Zeevalk G. D. and Nicklas W. J. ( 19891 Acute excito-toxicity in chick retina caused by the unusual amino acids BOAA and BMAA : effects of MK-801 and kynurenate. Neurosci. Lett. 102, 284 290. Zecvalk G., Hyndman A. and Nicklas W. (19891 Excitatory amino acid-induced toxicity in chick retina: amino acid release, histology and effects of chloride channel blockers. .I. Neurochem. 53, 1610 1619.