- Email: [email protected]
Developmental increase in the sensitivity to magnesium of NMDA receptors on CA1 hippocampal pyramidal cells
Developmental Brain Research, 56 (1990) 55-61 Elsevier
55
BRESD 51143
Developmental increase in the sensitivity to magnesium of NMDA receptors on CA1 hippocampal pyramidal cells Mark A. Bowe and J. Victor Nadler Departments of Pharmacology and Neurobiology, Duke University Medical Center, Durham, NC 27710 (U.S.A.) (Accepted 15 May 1990)
Key words: N-MethyI-D-aspartate; Glutamate; Development; Ontogeny; Magnesium; Excitatory amino acid; Hippocampus
The N-methyl-o-aspartate (NMDA) receptor is involved in processes, such as associative learning, that are particularly important during early postnatal development. It has been suggested that the activity and regulation of this receptor changes during development. Activation of the NMDA receptor is normally limited by Mg2+ present in the extracellular fluid of brain. We have found that Mg2+ less potently antagonizes the depolarizing action of NMDA in developing rats than in adults. A grease-gap method was used to record depolarizations evoked in CA1 hippocampal pyramidal cells by the excitants NMDA and AMPA (a-amino-3-hydroxy-5-methyl-4-isoxazolepropionate).In the adult CA1 area, Mg2+ shifted the NMDA concentration-response curve to the right in a manner consistent with voltage-dependent open channel block (uncompetitive antagonism) in a preparation with significant receptor reserve. The potency of Mg2÷ increased during development; a greater than two-fold change in the ECso for Mg2+ was observed between 10-15 days of age and adulthood. A concentration of 10 mM reduced the maximum response of CA1 pyramidal cells to NMDA in adult rats, but not in developing rats. In addition, Mg2+ often enhanced the maximum depolarizations evoked by NMDA in 10- to 15-day-old rats, but very seldom in adults. No significant developmental changes in AMPA-induced depolarizations were observed in the presence or absence of Mg2+. These results suggest that synaptically released glutamate will readily activate NMDA receptors during early development and that its ability to do this declines with the maturation of the brain. This may at least partially explain developmental reductions in learning ability, the ability to generate long-term potentiation, the ability to evoke kindling and the sensitivity of CNS neurons to NMDA receptor-dependent excitotoxicity,
INTRODUCTION It is b e c o m i n g increasingly a p p a r e n t that activation of the N-methyl-D-aspartate ( N M D A ) r e c e p t o r plays a crucial role in the d e v e l o p m e n t of the brain. Blockade of the N M D A r e c e p t o r during d e v e l o p m e n t has been found to p r e v e n t e x p e r i e n c e - d e p e n d e n t plasticity H'21"zs, as well as early olfactory learning 3°. Certain visual responses in neocortex of kittens are m o r e d e p e n d e n t on N M D A receptors than are similar responses in adult cats 17'49. In the rat cerebellum, there is a dramatic reduction in the sensitivity of both Purkinje and granule cells l-s't9 to N M D A as the animal matures. Trophic roles for the N M D A r e c e p t o r during brain d e v e l o p m e n t have been indicated ~'2"~m'3~'4°. Pathological processes thought to be d e p e n d e n t , in part at least, on N M D A receptor activation are frequently d e v e l o p m e n t a l l y d e p e n d e n t . For c x a m p l e , h i p p o c a m p a l kindling proceeds at a more rapid rate in rat pups than it does in adults 29 and the immature rat brain is m o r e susceptible to damage from hypoxic/ ischemic conditions 24. The developing rat brain is also more sensitive to the excitotoxic action of directly applied N M D A 24-3~.
The unique v o l t a g e - d e p e n d e n t block of the N M D A receptor/channel by physiological concentrations of Mg 2+ limits the conditions under which the N M D A r e c e p t o r will participate in n o r m a l synaptic function 12'~3'39. This is the m a j o r p r o p e r t y of the N M D A r e c e p t o r that allows it to function as a gating mechanism in long-lasting forms of plasticity ~4. Interestingly, a n u m b e r of reports suggest that changes in the density or regulation of the N M D A receptor correlate with d e v e l o p m e n t a l events. For example, a correlation was shown b e t w e e n N M D A receptor density and the 'critical p e r i o d ' of e x p e r i e n c e - d e p e n d e n t plasticity in kitten visual cortex 6. B e n - A r i et al. 5 found that N M D A - i n d u c e d currents in h i p p o c a m p a l pyramidal~ cells from 0 - 9 day old rats exhibited little sensitivity to m e m b r a n e potential, suggesting a decreased voltaged e p e n d e n t block by Mg 2÷. A recent a u t o r a d i o g r a p h i c study also indicated a possible d e v e l o p m e n t a l change in the stoichiometry of the binding sites associated with the N M D A r e c e p t o r 35. To address the issue of mechanisms by which the N M D A r e c e p t o r plays a h e i g h t e n e d role in CNS events during d e v e l o p m e n t , we have sought to characterize the p h a r m a c o l o g y of the r e c e p t o r as a function of age. In
Correspondence: J.g. Nadlcr, Department of Pharmacology, Box 3813, Duke University Medical Center, Durham, NC 27710, U.S.A.
56 particular, we have examined the blockade of the N M D A receptor by Mg 2÷ to test the hypothesis that responses to
maximal concentration o1' AMPA was 20 uM for rats oi all ages in the presence or absence of Mg2'
N M D A would be less sensitive to Mg 2÷ in the early
Data analysis
developmental period. To this end, we used a recently described grease-gap preparation 32 to record N M D A induced depolarizations of C A I hippocampal pyramidal
ECso values were estimated by a least squares regression of the linear portion of semi-logarithmic concentration-response curves'*~. Each curve included 5-9 concentrations of agonist. Concentrationratios were calculated at the ECso level. EC5, values and concentration-ratios are not normally distributed. Accordingly, all parametric statistical operations were performed on the log m transformations of these values.64s. Hill slopes were calculated by the method of Williams et al.5"; ECs, values for antagonism of NMDA by Mg~ were calculated by normalizing each response to 10 #M NMDA in the presence of Mg2~ by the corresponding response to 10 MM NMDA in the absence of Mg2 ~. The absolute magnitude of excitant-induced depolarizations varied with the preparation and with the resistance of the grease seal. Therefore, for graphical presentation, all responses were normalized to the maximal response obtained in the absence of Mgz+, defined as 100%
cells. MATERIALS AND METHODS
Grease-gap preparation Lactating female Sprague-Dawley rats with 10 9-day-old foster pups each and adult rats of either sex (60-100 days of age) were obtained from Zivic-Miller Labs. (Allison Park, PA). Hippocampal slices were prepared and maintained in a grease-gap apparatus as described in detail elsewhere 32. Briefly, rats of either sex were anesthetized with ether and decapitated. Brains were quickly removed to chilled medium consisting of (mM): HEPES 25, NaCl 122, KCI 3.1, KH2PO4 0.4, CaCl2 1.3, MgSO4 1.0 and D-glucose 10, pH 7.4. Hippocampi were dissected out and longitudinal slices of 450-/am thickness were prepared using a Mcllwain tissue chopper, Regio inferior (areas CA2, CA3 and CA4) and the fascia dentata were removed by microdissection, leaving area CA1 and the retrohippocampal area including the subiculum. The CAl-subiculum slice was transferred to a two-component superfusion chamber and a grease barrier was formed at the CAl-subicular border, such that the CA1 pyramidal cell bodies and dendrites lay in one compartment and their axons projected through the grease barrier to the subiculum in the other compartment. The compartments were independently superfused at 2 ml/min with artificial CSF, which was bubbled continuously with 95% 02/5% CO2 (medium composition same as above, with substitution of 25 mM NaHCO 3 for the HEPES and the addition of 3 #M gtycine). The temperature of the tissue compartments was maintained at 32 °C.
Pharmacological testing DC potentials were differentially recorded between the two compartments with a pair of Ag/AgCI electrodes and were continuously displayed on a chart recorder. NMDA and a-amino-3hydroxy-5-methyl-4-isoxazolepropionate (AMPA), a quisqualate receptor agonist, were tested by including the excitant in the medium superfusing the CA1 compartment for 2 min. This time was sufficient to evoke a full response. The amplitude of the deflection in the differential recording, measured from onset to peak, was taken as a measure of the agonist-induced neuronal depolarization. After an initial 30-rain equilibration period, a test concentration of agonist roughly equal to the expected ECso was applied to the CA1 pyramidal cells. If a depolarization of at least 0.1 mV was obtained and the recording returned to a stable baseline, Mg2÷ was removed from the medium superfusing the CA1 compartment and the slice was equilibrated for another 40 min. At the end of this period, test concentrations of agonist were repeatedly applied until reproducible depolarizations were obtained. Concentration-response curves were generated by applying a set of agonist concentrations in random order. At least 10 min was allowed for recovery between applications of excitant. Mg2÷ was then added to the medium and a second concentration-response curve was generated beginning 30 min later. All excitants were tested in the absence of added Mg2÷ and in one or two concentrations of Mg2÷. Because high concentrations of both excitatory amino acids irreversibly attenuated responses to all excitants, the highest concentrations used were those that evoked just-maximal depolarizations. The maximal concentrations of NMDA chosen for testing in the presence of 0, 0.1, 0.316, 1.0, 3.16 and 10.0 mM added Mg2÷ were 25, 40, 50, 60, 100, and 125 MM, respectively. For animals 10--15 days of age, however, 40 #M NMDA was needed to obtain a just-maximal response in the absence of added Mg2÷. The just-
Sources of test compounds NMDA was obtained from Tocris Neuramin (Essex, U,K.), AMPA from Research Biochemicals (Natiek, MA, U.S.A.), and picrotoxin from Sigma Chemical Co. (St. Louis~ MO, U.S.A.). RESULTS Depolarizing responses to N M D A were first studied in the presence and absence of 1 m M Mg 2÷. C A l - s u b i c u l u m slices prepared from rats at least 10 days old generated depolarizations of sufficient amplitude for this purpose. However, it was found impractical to study animals less than 10 days of age, due to the fragility of the slices. Responses to N M D A were qualitatively similar at all ages. A rapid depolarization was followed by a slow after-hyperpolarization (Fig. 1). T h e m e a n amplitudes of the depolarizations varied with age. The average maximal depolarization ( + S . E . M . ) o b t a i n e d with N M D A for the age groups 10-15, 18-27, 30-36, and 60-100 days postnatal were: 1.1 _+ 0.1, 1.5 + 0.1, 0.8 + 0.2, and 0.8 + 0.2 mV, respectively. In slices from animals older than 16 days, the addition of glycine to the superfusion m e d i u m did not affect the responses of CA1 pyramidal cells to N M D A . In preliminary experiments on slices from 10- to 15-day-old
A
10-,5 ~
B
I
i}
r -~ 41
6o+ d
m
{Wga'] (raM): [NMDA] (~M):
f
@ 0.316
O
]O
mlrl
111 rain
'!
JIr
;i I ,,/
i.(I
(I ill
f)
i[]
I0
40
Fig. 1. NMDA-induced depolarizations of CA1 hippocampal pyramidal cells recorded in the presence and absence of Mg2÷. Note the greater effect of Mg2÷ in slices from adult rats.
57
5 X
o
4
X
X
Xx
o
x
..x~ ~ ,~ × ~
3
14
X X
x I C
x
X
x
x x
x I l.Xx
x
x
x
x
t_ C
× ×
C
,
,
i0
20
A
'
Ib
30 40 60+ Postnotol oge (d) Fig. 2. Concentration-ratio for 1 mM Mgz÷ (ECso for NMDA in the presence of 1 mM Mg2+/ECs0 for NMDA in the absence of added Mg2+) as a function of age. Concentration-ratio increased significantly with postnatal age (Spearman's rank correlation, P < 0.001). Each value (x) was obtained from an individual slice. Bars represent geometric means for age groups given in Table I.
animals, however, glycine increased the amplitude of depolarizations induced by concentrations of N M D A below the ECs0 in 3 of 8 slices. For this reason, 3/~M glycine, a concentration close to that present in the extracellular fluid of brain (unpublished observations), was routinely included in the superfusion medium. Addition of 1 m M Mg 2÷ to the superfusion medium depressed responses to N M D A less in slices from young rats (particularly those aged 10-15 days) than in slices from adults (Fig. 1B). Because 1 mM Mg 2÷ did not depress the maximal response obtained with N M D A at any age, it was possible to use concentration-ratios as a measure of Mg 2÷ antagonism. Fig. 2 illustrates the significant increase in the concentration-ratio during the postnatal period, implying an age-related increase in the ability of Mg 2+ to block the N M D A channel.
For further analysis, the animals were divided into four age groups within which the pEC5o values and concentration-ratios for N M D A in the presence of 1 m M Mg 2÷ were relatively constant: 10-15, 18-27, 30-36, and 60+ days of age. Only at 10-15 days of age was 1 m M Mg 2÷ significantly less effective than in adults (Table I). Its efficacy reached adult levels by 30 days of age. In addition, the potency of N M D A in the absence of added Mg 2+ changed during postnatal development. N M D A was significantly less potent than in adults during the 10-15 day period and slightly more potent between 18 and 27 days of age. In contrast, no significant changes in the potency of A M P A were found in the absence of added Mg 2+, and 1 m M Mg 2+ did not significantly affect responses to A M P A at any age. To study the developmental change in Mg 2÷ antagonism more closely, further studies focused on the differences between slices prepared from 10- to 15-day-old rats and slices prepared from adults. A series of concentration-response curves were constructed in the absence of added Mg 2+ and in the presence of 5 Mg 2+ concentrations in the range of 0.1-10 m M (Fig. 3). In slices from adult rats, increasing the Mg 2÷ concentration progressively shifted the N M D A concentration-response curve to the right. Only 10 m M Mg 2÷ depressed the maximal response to N M D A . The concentration-response curves were not parallel, but instead became slightly steeper with increasing Mg 2+ concentration (two-way A N O V A of Hill slopes yielded P < 0.0001). We noted three differences in the actions of Mg 2÷ in the 10-15 day old group. First, Mg 2÷ was less potent than in adult animals. Concentrations of Mg 2+ as low as 0.1-0.316 m M could markedly depress responses to N M D A in slices from adults, but not in slices from 10- to 15-day-old rats (Figs. 1A and 3). In fact, the age-related
TABLE I Pharmacologieal parameters for depolarization of CA1 hippocampal pyramidal cells by NMDA and AMPA ApECs0 is the change in pECs0 upon adding 1 mM Mg2÷. pECso values are means + S.E.M. for the number of slices in parentheses. ECso and concentration-ratio values are the geometric means. Age (days)
NMDA pECso in (J Mg2+ ECs0 in 0 Mg2+ ApEC5o Concentration-ratio AMPA pECso in 0 Mg2+ ECso in 0 Mg2+ ApECso Concentration-ratio
10-15
18-27
30-36
60+
5.09 + 0.02** (21) 8.1/~M 0.33 _+0.02** 2.2 5.36 _+0.03 (10) 4.4/~M 0.03 _+0.01 1.07
5.33 + 0.02* (17) 4.7#M 0.48 _+0.02 3.0 5.40 + 0.03 (6) 4.0/~M 0.03 + 0.01 1.07
5.25 + 0.03 (11) 5.6HM 0.53 -+ 0.02 3.4 5.39 _+0.03 (6) 4.0#M 0.02 _+0.02 1.04
5.23 + 0.04 (9) 5.8HM 0.54 + 0.02 3.5 5.42 + 0.02 (9) 3.8/~M 0.06 + 0.02 1.15
**P < 0.01 or *P < 0.05 compared to 60+ days of age, Dunnett's test after 2-way ANOVA (age × M g 2+ concentration) with repeated measures (Mg2÷ concentration) yielded P < 0.0001 for age and P < 0.0001 for interaction between the variables.
58 difference in Mg 2÷ antagonism was more obvious with these concentrations than with 1 mM Mg 2÷. To quantify developmental changes in antagonist potency, two analyses were performed. The Schild analysis utilized all concentrations of Mg 2+ that did not depress the maximal response in either age group (i.e., 0.1-3.16 mM Mg2+). In both groups, Schild plots were linear and their slopes were significantly less than unity (Fig. 4), consistent with uncompetitive antagonism. The Schild slopes were not significantly different (Table II). However, the X-intercept was significantly lower in the 10- to 15-day-old group. Although the X-intercept of the Schild plot cannot be regarded as a dissociation constant in this instance, it remains useful for the purpose of comparison. The difference in these intercepts suggests a greater than 3-fold increase in the potency of Mg 2+ between 10 and 15 days of age and adulthood. Calculation of ECs0 values for Mg 2+ (against 10/~M N M D A ) supported this idea. The ECs0 increased more than 2-fold during development (Table 1I). Second, 10 mM Mg z÷ markedly depressed the maximal response to N M D A in slices from adult rats, but not in slices from 10-15 day old rats (Fig. 3, Table II). This change could signify an increase in the efficacy of Mg 2÷ with age or it could simply be another manifestation of the age-related increase in Mg 2+ potency. Third, in the 10- to 15-day-old group, about one-third
g
lOO
10-15 d
0.8
o 10-15 d • 60+ d
~
0.4
m
0.0 -0.4 -0.8 I
-4.0
h
I
-3.5 -3.0 log [M92"] (M)
J
-2.5
Fig. 4. Schild plot for Mg2÷ derived from data in Fig. 3. The plot was shifted toward higher Mg2+ concentrations for 10- to 15-day:old rats compared to adults. Schild parameters are given in Table II. Values are means + S.E.M.
of the preparations generated a maximal depolarizing response to N M D A that was at least 10% greater in the presence of Mg 2+ than in its absence (Table II). The increase occurred as often in 0.1 mM Mg 2÷ as in 3.16 mM Mg 2÷. This phenomenon was observed in fewer than 10% of the preparations from adult animals, and in these cases only with 0.1-0.316 mM Mg 2÷. No concentration of added Mg 2÷ increased the maximal response to A M P A at any age. The hyperpolarization evoked by GABAergic synaptic inhibition has been shown to augment the antagonistic action of Mg 2÷ toward N M D A receptor activation 12. Therefore one possible explanation for a developmental
TABLE II o "d g
50
Developmental changes observed in the NMDA-induced depolarizations of CA 1 pyramidal cells
I
i
i
ECs0 (against 10 aM NMDA) and Schild analyses were performed with compiled data from the number of slices shown in parentheses. Age (days)
C
1oo
60+ d
o
10-15
o/
60+
g ~ ~ ~ g
50
i
-5.5
-5.0 -4.5 1o9 [NMDA] (M)
-4.0
Fig. 3. NMDA concentration-response relationships in the presence of various concentrations of added Mg2+ (0-10 mM). (O) Without added Mg2÷; (0) with Mg2÷. Values are means + S.E.M. for the foilowing.Mg2+ concentrations and number of slices (10-15 days, 60+ days): 0 (39,22), 0.1 mM (6,6), 0.316 mM (6,6), 1 mM (21,9), 3.16 mM (7,6), and 10 mM (6,6).
Schild slope 0.75+0.07 (37) 0.60+0.05 (21) X-intercept of Schild plot -3.12+0.04 -3.63+0.05* log-1 X-intercept 759 aM 234/~M pEC5oof Mg2÷ 3.40+ 0.05 (32) 3.76+0.05** (8) ECso of Mg2+ 400/tM 170/,M Depression of maximal responses by 10 mM Mg~÷ in (n) slices 12+6% (6) 44+6% a (6) Fraction of slices in which Mg2÷ increased maximal response 12/37 2/23° *P < 0.0001, Student's t-test. **P < 0.001, Student's t-test corrected for unequal Variances. ~P < 0.005, Student's t-test. ~'P< 0.05, x2-analysis.
59 increase in the potency of Mg 2+ is that N M D A activates G A B A neurons in our preparation and this action produces a greater pyramidal cell hyperpolarization in slices from adult animals than in slices from 10-15-dayold animals. To test this possibility, we studied the effects of the ),-aminobutyric acid (GABA) antagonist picrotoxin on NMDA-induced depolarizations in the presence of 1 mM Mg 2+. Picrotoxin (50 ~M) changed the ECs0 of NMDA by less than 5% at either age. Thus we conclude that the developmental increase in Mg 2+ potency is not related to changes in GABAergic inhibition. DISCUSSION Our results demonstrate that Mg 2+ is less able to reduce NMDA-evoked depolarizing responses in CA1 hippocampal pyramidal cells of developing rats than in mature pyramidal cells. This finding is consistent with the lesser voltage-sensitivity of NMDA-induced currents during early development 5. It also agrees with reports of Brady and Swann 7'8 that, in contrast to hippocampal slices from mature animals, removing Mg 2+ from medium superfusing slices from developing rats did not relieve the voltage dependence of NMDA-induced currents. A reduced sensitivity to Mg 2+ would facilitate the participation of NMDA receptors in developmental events. One must consider the possibility that artifacts related to the grease-gap method accounted for the change in Mg2+ potency. It is often difficult to obtain a true maximum in grease-gap studies of excitatory amino acids, due to excitotoxicity, desensitization and/or incomplete equilibration 33'44. This can lead to underestimates of the true ECso values. In addition, agonist concentrationresponse curves will be influenced by any activation or deactivation of voltage-dependent ion conductances that may occur secondary to a change in membrane potential. Developmental changes in such properties of area CA1 as the percentage of extracellular space, the resistance of pyramidal cell membranes and dendritic growth could also influence agonist potency. Finally, Mg 2+ might affect agonist potency in some way other than by blocking the NMDA channel. Each of these factors would be expected to alter the potency of all excitants, however. Our findings that the potency of AMPA remained essentially constant in the same preparations where developmental changes in Mg 2+ regulation of the NMDA receptor were observed and that Mg2÷ did not alter AMPA potency at any age support the view that differences among age groups in the action of Mg2+ reflect a developmental change specific to the NMDA receptor. Another possible explanation for the enhancement of Mg 2+ potency during development is that this change
occurs secondarily to a reduction in NMDA receptor density. The unusually high potency of NMDA on CA1 hippocampal pyramidal cells has been ascribed to the presence of a substantial receptor reserve 32. It is this high receptor density, coupled with the voltage-sensitivity of Mg2+-evoked open-channel block, that explains the non-parallel displacement of the NMDA concentrationresponse curve with increasing Mg 2+ concentration 32' 39,41 In the presence of a receptor reserve, agonist potency is directly related to receptor density2v and the potency of an uncompetitive antagonist, such as Mg 2+, is inversely related to receptor density41. Thus the lesser potency of Mg 2÷ in slices from developing rats might be explained by a relatively high NMDA receptor density. Different developmental profiles of NMDA receptor density have been reported by different groups. In three reports, the overall pattern of NMDA receptor ontogeny in the rat hippocampus has been described as a transient postnatal increase followed by a decrease to adult levels. In one study, this transient increase occurred in the CA1 area so early that receptor density actually declined slightly over the postnatal period covered by the present study48. In the two other studies, however, the transient increase did not peak until the third postnatal week 25"35, a pattern inconsistent with the changes we observed in Mg2+ potency. The exact timing of the transient increase in NMDA receptor density may depend upon the strain of rat used 42. A fourth study found that NMDA receptor binding increases continuously during postnatal development 43. This study utilized Sprague-Dawley rats, the same strain used in the present study. The changes in Mg2+ potency probably cannot be explained simply by changes in NMDA receptor density, however. First, there is no evidence of a developmental decrease in NMDA receptors sufficient to account for the magnitude of the change in the ECso of Mg 2+. Second, the developmental increase in the potency of NMDA could result from an increase in receptor density during development, such as that reported by Peterson et al. 43, but increased receptor density predicts a developmental reduction in the potency of Mg 2+. Although more work is needed to determine the cellular mechanism that underlies the developmental increase in the potency of Mg 2+, we favor the possibility that Mg 2+ sensitivity is altered by changes in the subunit composition of the NMDA receptor, in the .post-translational modification of the receptor 3~3~ or in the mix of NMDA receptor subtypes ~4. A precedent exists for a change in the subunit composition of an ion channelforming receptor during development 37. Indeed the results of Ben-Ari et al. 5, Brady and Swann 7'8 and McDonald et al. 35, in conjunction with the present study, are most consistent with a developmental change in
60 receptor structure. However, we cannot rule out the possibility that developmental changes in such factors as the placement of N M D A receptors on the pyramidal cell 22 or the resting membrane potential of the pyramidal cell could also play a role. These issues cannot be addressed with the grease-gap preparation. In slices from 10-15-day-old rats, Mg 2÷ frequently increased the maximal response of CA1 pyramidal cells to NMDA. This effect was observed in only two slices from adult animals, a frequency that probably reflects the normal variability of the grease-gap method. Enhancement of the maximal response to N M D A could theoretically arise from an effect of Mg 2+ on any of the factors that determine response amplitude in grease-gap experiments 33'44. For example, Mg 2+ might increase the electrical resistance of the unmyelinated pyramidal cell axons present in the developing CA1 area more than the resistance of the myelinated axons present in the adult CA1 area. If this were the case, however, we would have expected a similar enhancement of the maximal response to A M P A . No such effect of Mg 2+ was observed. A more likely possibility is that Mg 2÷, in addition to blocking the open channel, affects N M D A receptor function in some other way. Submillimolar concentrations of Mg 2÷ have been shown to enhance the binding of glycine to the N M D A receptor 34. Mg 2+ reportedly increases both the affinity of the receptor for glycine and the maximal quantity of glycine bound. Such an effect could potentiate responses to N M D A 47, even in the presence of a saturating glycine concentration. This action of Mg 2÷ would serve to counteract its channel blocking action. It is possible that the potentiation of glycine binding by Mg 2+ is more significant in developing rats than in adults. Mg 2+ is probably the most important antagonist of N M D A receptor function present in the intact brain. Therefore a lesser effect of Mg 2+ during development
would be expected to facilitate participation of the N M D A receptor in synaptic events ~z'~3. This developmental difference might contribute to the increased magnitude of long-term potentiation observed in the CA 1 area of 15-day-old rats 23'46. It might also explain the observation that, even in the presence of a physiological Mg 2+ concentration, excitatory synaptic responses of immature cerebellar granule cells have a significant N M D A receptor component, whereas equivalent responses from adult preparations only showed an N M D A receptor component when Mg 2+ was removed 2{}. Additionally, the spontaneous activity of entorhinal cortical neurons 26 and CA3 hippocampal pyramidal cells 4 recorded in slices from immature rats continued in the presence of Mg 2÷, but was suppressed by the competitive N M D A receptor antagonist D-2-amino-5-phosphonovalerate (AP5). In contrast, Mg 2÷ suppressed all such activity in slices from adult rats. A lesser Mg 2÷ antagonism would also facilitate the trophic-like actions of N M D A receptor agonists ~'38'4°. Indeed developmental changes in N M D A receptor function may have equal or even greater implications for neuronal growth and synapse formation than for synaptic activity. For technical reasons, the present study was limited to animals 10 days of age and older. The trend apparent in our data suggests that Mg 2÷ would have been an even weaker antagonist in slices from younger animals. This possibility needs to be tested with quantitative methods of studying N M D A receptor function that are applicable to animals younger than 10 days. Further experimentation is also necessary to determine the extent to which the developmental changes we observed in responses of CA1 pyramidal cells apply to CNS neurons generally.
REFERENCES
7 Brady, R.J. and Swann, J.W., Developmental change in the magnesium sensitivity of the NMDA response in hippocampal CA3 pyramidal cells, Soc. Neurosci. Abstr., 15 (1989) 327. 8 Brady, R.J. and Swann, J.W., The effects of extracellular calcium on the epileptiform activity and NMDA responses are different in mature and immature hippocampal slices, Soc. Neurosci. Abstr., 14 (1988) 239. 9 Brenneman, D.E., Forsythe, I.D., Nicol, T. and Nelson, P.G., N-Methyl-D-aspartate receptors influence neuronal survival in developing spinal cord cultures, Dev. Brain Res., 51 (1990) 63-68. 10 Brewer, G.J. and Cotman, C.W., NMDA receptor regulation of neuronal morphology in cultured hippocampal neurons, Neurosci. Left., 99 (1989) 268-273. 11 Cline, H.T., Debski, E.A. and Constantine-Paton, M., NMethyl-D-aspartate receptor antagonist desegregates eye-specific stripes, Proc. Natl. Acad. Sci. U.S.A., 84 (1987) 4342-4345. 12 Collingridge, G.L., Herron, C.E. and Lester, R.A.J., Synaptic activation of N-methyl-i>aspartate receptors in the Schaffer collateral-commissural pathway of rat hippocampus, J. Physiol., 399 (1988) 283-300.
1 Bal~zs, R., Hack, N. and Jergensen, O.S., Stimulation of the N-methyl-D-aspartate receptor has atrophic effect on differentiating cerebellar granule cells, Neurosci. Lett., 87 (1988) 80-86. 2 Bal~izs, R., Hack, N., Jergenson, O.S. and Cotman, C.W., N-Methyl-D-aspartate promotes the survival of cerebellar granule cells: pharmacological characterization, Neurosci. Lett., 101 (1989) 241-246. 3 Bartlett, M.C., Salter, M.W. and MacDonald, J.F., Modulation of NMDA currents by intracellular phosphorylation, Soc. Neurosci. Abstr., 15 (1989) 535. 4 Ben-Ari, Y., Cherubini, E., Corradetti, R. and Gaiarsa, J.L., Giant synaptic potentials in immature rat CA3 hippocampal neurones, J. Physiol., 416 (1989) 303-325. 5 Ben-Ari, Y., Cherubini, E. and Krnjevid, K., Changes in voltage dependence of NMDA currents during development, Neurosci. Lea., 94 (1988) 88-92. 6 Bode-Greuel, K.M. and Singer, W., The development of N-methyl-D-aspartate receptors in cat visual cortex, Dev. Brain Res., 46 (1989) 197-204.
Acknowledgements. We thank Ms. C. Kelsall for secretarial assistance. This study was supported by NIH Grant NS 16064 (J.V.N.) and by an NSF predoctoral fellowship (M.A.B.).
61 13 Collingridge, G.L., Herron, C.E. and Lester, R.A.J., Frequency-dependent N-methyl-D-aspartate receptor-mediated synaptic transmission in rat hippocampus, J. Physiol., 399 (1988) 301-312. 14 Cotman, C.W., Bridges, R.J., Taube, J.S., Clark, A.S., Geddes, J.W. and Monaghan, D.T., The role of the NMDA receptor in central nervous system plasticity and pathology, J. NIH Res., 1 (1989) 65-74. 15 Dupont, J.L., Gardette, R. and Crepel, F., Postnatal development of the chemosensitivity of rat cerebellar Purkinje cells to excitatory amino acids. An in vitro study, Dev. Brain Res., 34 (1987) 59-68. 16 Fleming, W.W., Westfall, D.P., De La Lande, I.S. and Jellett, L.B., Lognormal distribution of equieffective doses of norepinephrine and acetylcholine in several tissues, J. Pharmacol. Exp. Ther., 181 (1972) 339-345. 17 Fox, K., Sato, H. and Daw, N., The location and function of NMDA receptors in cat and kitten visual cortex, J. Neurosci., 9 (1989) 2443-2454. 18 Gaddum, J.H., Lognormal distributions, Nature, 156 (1945) 463-466. 19 Garthwaite, G., Yamini, B. and Garthwaite, J., Selective loss of Purkinje and granule cell responsiveness to N-methyl-Daspartate in rat cerebellum during development, Dev. Brain Res., 36 (1987) 288-292. 20 Garthwaite, J. and Brodbelt, A.R., Synaptic activation of N-methyl-D-aspartate and non-N-methyl-D-aspartate receptors in the mossy fiber pathway in adult and immature rat cerebellar slices, Neuroscience, 29 (1989) 401-412. 21 Gu, Q., Bear, M.F. and Singer, W., Blockade of NMDAreceptors prevents ocularity changes in kitten visual cortex after reversed monocular deprivation, Dev. Brain Res., 47 (1989) 281-288. 22 Hamon, B. and Heinemann, U., Developmental changes in neuronal sensitivity to excitatory amino acids in area CA1 of the rat hippocampus, Dev. Brain Res., 38 (1988) 286-290. 23 Harris, K.M. and Teyler, T.J., Developmental onset of longterm potentiation in area CA1 of the rat hippocampus, J. Physiol., 346 (1984) 27-48. 24 Ikonomidou, C., Mosinger, J.L., Shahid Salles, K., Labruyere, J. and Olney, J.W., Sensitivity of the developing rat brain to hypobaric/ischemic damage parallels sensitivity to N-methylD-aspartate neurotoxieity, J. Neurosci., 9 (1989) 2809-2818. 25 Insel, T.R., Miller, L.F., Johnson, A.E. and Gelhard, R., Ontogeny of excitatory amino acid receptor subtypes in rat brain, Soc. Neurosci. Abstr., 14 (1988) 483. 26 Jones, R.S.G. and Heinemann, U., Spontaneous activity mediated by NMDA receptors in immature rat entorhinal cortex in vitro, Neurosci. Lett., 104 (1989) 93-98. 27 Kenakin, T.P.. The classification of drugs and drug receptors in isolated tissues, Pharmacol. Rev., 36 (1984) 165-222. 28 Kleinschmidt, A., Bear, M.E and Singer, W.E, Blockade of 'NMDA' receptors disrupts experience-dependent plasticity of kitten striate cortex, Science, 238 (1987) 355-358. 29 Lee, S.S., Murata, R. and Matsuura, S., Developmental study of hippocampal kindling, Epilepsia, 30 (1989) 266-270. 30 Lincoln, J., Coopersmith, R., Harris, E.W., Cotman, C.W. and Leon, M., NMDA receptor activation and early olfactory learning, Dev. Brain Res., 39 (1988) 309-312. 31 MacDonald, J.F., Mody, I. and Salter, M.W., Regulation of N-methyI-D-aspartate receptors revealed by intracellular dialysis of routine neurones in culture, J. Physiol., 414 (1989) 17-34. 32 Martin, D., Bowe, M.A. and Nadler, J.V., A grease-gap method for studying the excitatory amino acid pharmacology of CA1
33
34
35 36
37
38
39 40
41 42 43 44 45 46 47 48
49 50
hippocampal pyramidal cells, J. Neurosci. Methods, 29 (1989) 107-114. Martin, D. and Nadler, J.V., Grease-gap methods for studying the pharmacology of excitatory amino acids on CNS neurons. In P.M. Conn. (Ed.), Electrophysiology and Microinjection, Methods in Neuroscience, Vol. 4, Academic Press, Orlando, in press. Marviz6n, J.C.G. and Skolnick, P., [3H]Glycine binding is modulated by Mg2+ and other ligands of the NMDA receptorcation channel complex, Eur. J. Pharmacol., 151 (1988) 157158, McDonald, J.W., Johnston, M.V. and Young, A.B., Ontogeny of the receptors comprising the NMDA receptor complex, Soc. Neurosci. Abstr., 15 (1989) 198. McDonald, J.W., Silverstein, F.S. and Johnston, M.V., Neurotoxicity of N-methyl-D-aspartate is markedly enhanced in developing rat central nervous system, Brain Res., 459 (1988) 200-203. Mishina, M., Takai, T., Imoto, K., Noda, M., Takahashi, T., Numa, S., Methfessel, C. and Sakmann, B., Molecular distinction between fetal and adult forms of muscle acetylcholine receptor, Nature, 321 (1986) 406-411. Moran, J. and Patel, A.J., Stimulation of the N-methylD-aspartate receptor promotes the biochemical differentiation of cerebellar granule neurons and not astrocytes, Brain Res., 486 (1989) 15-25. Nowak, L., Bregestovski, P., Ascher, P., Herbet, A. and Prochiantz, A., Magnesium gates glutamate-activated channels in mouse central neurons, Nature, 307 (1984) 462-465. Pearce, I.A., Cambray-Deakin, M.A. and Burgoyne, R.D., Glutamate acting on NMDA receptors stimulates neurite outgrowth from cerebellar granule cells, FEBS Lett., 223 (1987) 143-147. Pennefather, P. and Quastel, D.M.J., Modification of doseresponse curves by effector blockade and uncompetitive antagonism, Mol. Pharmacol., 22 (1982) 369-380. Peterson, C. and Cotman, C.W., Strain-dependent decrease in glutamate binding to the N-metbyl-D-aspartic acid receptor during aging, Neurosci. Lett., 104 (1989) 309-313. Peterson, C., Neal, J.H. and Cotman, C.W., Development of N-methyl-D-aspartate excitotoxicity in cultured hippocampal neurons, Dev. Brain Res., 48 (1989) 187-195. Simmonds, M.A., Use of slices for quantitative pharmacology. In H. Jahnsen (Ed.), In Vitro Preparations from Vertebrate Systems, Wiley, Chicbester, in press. Tallarida, R.J. and Murray, R.B., Manual of Pharmacological Calculations with Computer Programs. 2nd edn., Springer, New York, 1987. Teyler, T.J., Perkins, A.T. and Harris, K.M., The development of longterm potentiation in hippocampus and neocortex, Neuropsychologia, 27 (1989) 31-39. Thomson, A.M., Walker, V.E. and Flynn, D.M., Glycine enchances NMDA-receptor mediated synaptic potentials in neocortical slices, Nature, 338 (1989) 422-424. Tremblay, E., Roisin, M.P., Repressa, A., Charriaut-Marlangue, C. and Ben-Ari, Y., Transient increased density of NMDA binding sites in the developing rat hippocampus, Brain Res., 461 (1988) 393-396. Tsumoto, T., Hagihara, K., Sato, H. and Hata, Y., NMDA receptors in the visual cortex of young kittens are more effective than those of adult cats, Nature, 327 (1987) 513-514. Williams, T.L., Smith, D.A.S., Burton, N.R. and Stone, T.W., Amino acid pharmacology in neocortical slices: evidence for bimolecular actions from an extension of the Hill and GaddumSchild equations, Br. J. Pharmacol., 95 (1988) 805-810,
55
BRESD 51143
Developmental increase in the sensitivity to magnesium of NMDA receptors on CA1 hippocampal pyramidal cells Mark A. Bowe and J. Victor Nadler Departments of Pharmacology and Neurobiology, Duke University Medical Center, Durham, NC 27710 (U.S.A.) (Accepted 15 May 1990)
Key words: N-MethyI-D-aspartate; Glutamate; Development; Ontogeny; Magnesium; Excitatory amino acid; Hippocampus
The N-methyl-o-aspartate (NMDA) receptor is involved in processes, such as associative learning, that are particularly important during early postnatal development. It has been suggested that the activity and regulation of this receptor changes during development. Activation of the NMDA receptor is normally limited by Mg2+ present in the extracellular fluid of brain. We have found that Mg2+ less potently antagonizes the depolarizing action of NMDA in developing rats than in adults. A grease-gap method was used to record depolarizations evoked in CA1 hippocampal pyramidal cells by the excitants NMDA and AMPA (a-amino-3-hydroxy-5-methyl-4-isoxazolepropionate).In the adult CA1 area, Mg2+ shifted the NMDA concentration-response curve to the right in a manner consistent with voltage-dependent open channel block (uncompetitive antagonism) in a preparation with significant receptor reserve. The potency of Mg2÷ increased during development; a greater than two-fold change in the ECso for Mg2+ was observed between 10-15 days of age and adulthood. A concentration of 10 mM reduced the maximum response of CA1 pyramidal cells to NMDA in adult rats, but not in developing rats. In addition, Mg2+ often enhanced the maximum depolarizations evoked by NMDA in 10- to 15-day-old rats, but very seldom in adults. No significant developmental changes in AMPA-induced depolarizations were observed in the presence or absence of Mg2+. These results suggest that synaptically released glutamate will readily activate NMDA receptors during early development and that its ability to do this declines with the maturation of the brain. This may at least partially explain developmental reductions in learning ability, the ability to generate long-term potentiation, the ability to evoke kindling and the sensitivity of CNS neurons to NMDA receptor-dependent excitotoxicity,
INTRODUCTION It is b e c o m i n g increasingly a p p a r e n t that activation of the N-methyl-D-aspartate ( N M D A ) r e c e p t o r plays a crucial role in the d e v e l o p m e n t of the brain. Blockade of the N M D A r e c e p t o r during d e v e l o p m e n t has been found to p r e v e n t e x p e r i e n c e - d e p e n d e n t plasticity H'21"zs, as well as early olfactory learning 3°. Certain visual responses in neocortex of kittens are m o r e d e p e n d e n t on N M D A receptors than are similar responses in adult cats 17'49. In the rat cerebellum, there is a dramatic reduction in the sensitivity of both Purkinje and granule cells l-s't9 to N M D A as the animal matures. Trophic roles for the N M D A r e c e p t o r during brain d e v e l o p m e n t have been indicated ~'2"~m'3~'4°. Pathological processes thought to be d e p e n d e n t , in part at least, on N M D A receptor activation are frequently d e v e l o p m e n t a l l y d e p e n d e n t . For c x a m p l e , h i p p o c a m p a l kindling proceeds at a more rapid rate in rat pups than it does in adults 29 and the immature rat brain is m o r e susceptible to damage from hypoxic/ ischemic conditions 24. The developing rat brain is also more sensitive to the excitotoxic action of directly applied N M D A 24-3~.
The unique v o l t a g e - d e p e n d e n t block of the N M D A receptor/channel by physiological concentrations of Mg 2+ limits the conditions under which the N M D A r e c e p t o r will participate in n o r m a l synaptic function 12'~3'39. This is the m a j o r p r o p e r t y of the N M D A r e c e p t o r that allows it to function as a gating mechanism in long-lasting forms of plasticity ~4. Interestingly, a n u m b e r of reports suggest that changes in the density or regulation of the N M D A receptor correlate with d e v e l o p m e n t a l events. For example, a correlation was shown b e t w e e n N M D A receptor density and the 'critical p e r i o d ' of e x p e r i e n c e - d e p e n d e n t plasticity in kitten visual cortex 6. B e n - A r i et al. 5 found that N M D A - i n d u c e d currents in h i p p o c a m p a l pyramidal~ cells from 0 - 9 day old rats exhibited little sensitivity to m e m b r a n e potential, suggesting a decreased voltaged e p e n d e n t block by Mg 2÷. A recent a u t o r a d i o g r a p h i c study also indicated a possible d e v e l o p m e n t a l change in the stoichiometry of the binding sites associated with the N M D A r e c e p t o r 35. To address the issue of mechanisms by which the N M D A r e c e p t o r plays a h e i g h t e n e d role in CNS events during d e v e l o p m e n t , we have sought to characterize the p h a r m a c o l o g y of the r e c e p t o r as a function of age. In
Correspondence: J.g. Nadlcr, Department of Pharmacology, Box 3813, Duke University Medical Center, Durham, NC 27710, U.S.A.
56 particular, we have examined the blockade of the N M D A receptor by Mg 2÷ to test the hypothesis that responses to
maximal concentration o1' AMPA was 20 uM for rats oi all ages in the presence or absence of Mg2'
N M D A would be less sensitive to Mg 2÷ in the early
Data analysis
developmental period. To this end, we used a recently described grease-gap preparation 32 to record N M D A induced depolarizations of C A I hippocampal pyramidal
ECso values were estimated by a least squares regression of the linear portion of semi-logarithmic concentration-response curves'*~. Each curve included 5-9 concentrations of agonist. Concentrationratios were calculated at the ECso level. EC5, values and concentration-ratios are not normally distributed. Accordingly, all parametric statistical operations were performed on the log m transformations of these values.64s. Hill slopes were calculated by the method of Williams et al.5"; ECs, values for antagonism of NMDA by Mg~ were calculated by normalizing each response to 10 #M NMDA in the presence of Mg2~ by the corresponding response to 10 MM NMDA in the absence of Mg2 ~. The absolute magnitude of excitant-induced depolarizations varied with the preparation and with the resistance of the grease seal. Therefore, for graphical presentation, all responses were normalized to the maximal response obtained in the absence of Mgz+, defined as 100%
cells. MATERIALS AND METHODS
Grease-gap preparation Lactating female Sprague-Dawley rats with 10 9-day-old foster pups each and adult rats of either sex (60-100 days of age) were obtained from Zivic-Miller Labs. (Allison Park, PA). Hippocampal slices were prepared and maintained in a grease-gap apparatus as described in detail elsewhere 32. Briefly, rats of either sex were anesthetized with ether and decapitated. Brains were quickly removed to chilled medium consisting of (mM): HEPES 25, NaCl 122, KCI 3.1, KH2PO4 0.4, CaCl2 1.3, MgSO4 1.0 and D-glucose 10, pH 7.4. Hippocampi were dissected out and longitudinal slices of 450-/am thickness were prepared using a Mcllwain tissue chopper, Regio inferior (areas CA2, CA3 and CA4) and the fascia dentata were removed by microdissection, leaving area CA1 and the retrohippocampal area including the subiculum. The CAl-subiculum slice was transferred to a two-component superfusion chamber and a grease barrier was formed at the CAl-subicular border, such that the CA1 pyramidal cell bodies and dendrites lay in one compartment and their axons projected through the grease barrier to the subiculum in the other compartment. The compartments were independently superfused at 2 ml/min with artificial CSF, which was bubbled continuously with 95% 02/5% CO2 (medium composition same as above, with substitution of 25 mM NaHCO 3 for the HEPES and the addition of 3 #M gtycine). The temperature of the tissue compartments was maintained at 32 °C.
Pharmacological testing DC potentials were differentially recorded between the two compartments with a pair of Ag/AgCI electrodes and were continuously displayed on a chart recorder. NMDA and a-amino-3hydroxy-5-methyl-4-isoxazolepropionate (AMPA), a quisqualate receptor agonist, were tested by including the excitant in the medium superfusing the CA1 compartment for 2 min. This time was sufficient to evoke a full response. The amplitude of the deflection in the differential recording, measured from onset to peak, was taken as a measure of the agonist-induced neuronal depolarization. After an initial 30-rain equilibration period, a test concentration of agonist roughly equal to the expected ECso was applied to the CA1 pyramidal cells. If a depolarization of at least 0.1 mV was obtained and the recording returned to a stable baseline, Mg2÷ was removed from the medium superfusing the CA1 compartment and the slice was equilibrated for another 40 min. At the end of this period, test concentrations of agonist were repeatedly applied until reproducible depolarizations were obtained. Concentration-response curves were generated by applying a set of agonist concentrations in random order. At least 10 min was allowed for recovery between applications of excitant. Mg2÷ was then added to the medium and a second concentration-response curve was generated beginning 30 min later. All excitants were tested in the absence of added Mg2÷ and in one or two concentrations of Mg2÷. Because high concentrations of both excitatory amino acids irreversibly attenuated responses to all excitants, the highest concentrations used were those that evoked just-maximal depolarizations. The maximal concentrations of NMDA chosen for testing in the presence of 0, 0.1, 0.316, 1.0, 3.16 and 10.0 mM added Mg2÷ were 25, 40, 50, 60, 100, and 125 MM, respectively. For animals 10--15 days of age, however, 40 #M NMDA was needed to obtain a just-maximal response in the absence of added Mg2÷. The just-
Sources of test compounds NMDA was obtained from Tocris Neuramin (Essex, U,K.), AMPA from Research Biochemicals (Natiek, MA, U.S.A.), and picrotoxin from Sigma Chemical Co. (St. Louis~ MO, U.S.A.). RESULTS Depolarizing responses to N M D A were first studied in the presence and absence of 1 m M Mg 2÷. C A l - s u b i c u l u m slices prepared from rats at least 10 days old generated depolarizations of sufficient amplitude for this purpose. However, it was found impractical to study animals less than 10 days of age, due to the fragility of the slices. Responses to N M D A were qualitatively similar at all ages. A rapid depolarization was followed by a slow after-hyperpolarization (Fig. 1). T h e m e a n amplitudes of the depolarizations varied with age. The average maximal depolarization ( + S . E . M . ) o b t a i n e d with N M D A for the age groups 10-15, 18-27, 30-36, and 60-100 days postnatal were: 1.1 _+ 0.1, 1.5 + 0.1, 0.8 + 0.2, and 0.8 + 0.2 mV, respectively. In slices from animals older than 16 days, the addition of glycine to the superfusion m e d i u m did not affect the responses of CA1 pyramidal cells to N M D A . In preliminary experiments on slices from 10- to 15-day-old
A
10-,5 ~
B
I
i}
r -~ 41
6o+ d
m
{Wga'] (raM): [NMDA] (~M):
f
@ 0.316
O
]O
mlrl
111 rain
'!
JIr
;i I ,,/
i.(I
(I ill
f)
i[]
I0
40
Fig. 1. NMDA-induced depolarizations of CA1 hippocampal pyramidal cells recorded in the presence and absence of Mg2÷. Note the greater effect of Mg2÷ in slices from adult rats.
57
5 X
o
4
X
X
Xx
o
x
..x~ ~ ,~ × ~
3
14
X X
x I C
x
X
x
x x
x I l.Xx
x
x
x
x
t_ C
× ×
C
,
,
i0
20
A
'
Ib
30 40 60+ Postnotol oge (d) Fig. 2. Concentration-ratio for 1 mM Mgz÷ (ECso for NMDA in the presence of 1 mM Mg2+/ECs0 for NMDA in the absence of added Mg2+) as a function of age. Concentration-ratio increased significantly with postnatal age (Spearman's rank correlation, P < 0.001). Each value (x) was obtained from an individual slice. Bars represent geometric means for age groups given in Table I.
animals, however, glycine increased the amplitude of depolarizations induced by concentrations of N M D A below the ECs0 in 3 of 8 slices. For this reason, 3/~M glycine, a concentration close to that present in the extracellular fluid of brain (unpublished observations), was routinely included in the superfusion medium. Addition of 1 m M Mg 2÷ to the superfusion medium depressed responses to N M D A less in slices from young rats (particularly those aged 10-15 days) than in slices from adults (Fig. 1B). Because 1 mM Mg 2÷ did not depress the maximal response obtained with N M D A at any age, it was possible to use concentration-ratios as a measure of Mg 2÷ antagonism. Fig. 2 illustrates the significant increase in the concentration-ratio during the postnatal period, implying an age-related increase in the ability of Mg 2+ to block the N M D A channel.
For further analysis, the animals were divided into four age groups within which the pEC5o values and concentration-ratios for N M D A in the presence of 1 m M Mg 2÷ were relatively constant: 10-15, 18-27, 30-36, and 60+ days of age. Only at 10-15 days of age was 1 m M Mg 2÷ significantly less effective than in adults (Table I). Its efficacy reached adult levels by 30 days of age. In addition, the potency of N M D A in the absence of added Mg 2+ changed during postnatal development. N M D A was significantly less potent than in adults during the 10-15 day period and slightly more potent between 18 and 27 days of age. In contrast, no significant changes in the potency of A M P A were found in the absence of added Mg 2+, and 1 m M Mg 2+ did not significantly affect responses to A M P A at any age. To study the developmental change in Mg 2÷ antagonism more closely, further studies focused on the differences between slices prepared from 10- to 15-day-old rats and slices prepared from adults. A series of concentration-response curves were constructed in the absence of added Mg 2+ and in the presence of 5 Mg 2+ concentrations in the range of 0.1-10 m M (Fig. 3). In slices from adult rats, increasing the Mg 2÷ concentration progressively shifted the N M D A concentration-response curve to the right. Only 10 m M Mg 2÷ depressed the maximal response to N M D A . The concentration-response curves were not parallel, but instead became slightly steeper with increasing Mg 2+ concentration (two-way A N O V A of Hill slopes yielded P < 0.0001). We noted three differences in the actions of Mg 2÷ in the 10-15 day old group. First, Mg 2÷ was less potent than in adult animals. Concentrations of Mg 2+ as low as 0.1-0.316 m M could markedly depress responses to N M D A in slices from adults, but not in slices from 10- to 15-day-old rats (Figs. 1A and 3). In fact, the age-related
TABLE I Pharmacologieal parameters for depolarization of CA1 hippocampal pyramidal cells by NMDA and AMPA ApECs0 is the change in pECs0 upon adding 1 mM Mg2÷. pECso values are means + S.E.M. for the number of slices in parentheses. ECso and concentration-ratio values are the geometric means. Age (days)
NMDA pECso in (J Mg2+ ECs0 in 0 Mg2+ ApEC5o Concentration-ratio AMPA pECso in 0 Mg2+ ECso in 0 Mg2+ ApECso Concentration-ratio
10-15
18-27
30-36
60+
5.09 + 0.02** (21) 8.1/~M 0.33 _+0.02** 2.2 5.36 _+0.03 (10) 4.4/~M 0.03 _+0.01 1.07
5.33 + 0.02* (17) 4.7#M 0.48 _+0.02 3.0 5.40 + 0.03 (6) 4.0/~M 0.03 + 0.01 1.07
5.25 + 0.03 (11) 5.6HM 0.53 -+ 0.02 3.4 5.39 _+0.03 (6) 4.0#M 0.02 _+0.02 1.04
5.23 + 0.04 (9) 5.8HM 0.54 + 0.02 3.5 5.42 + 0.02 (9) 3.8/~M 0.06 + 0.02 1.15
**P < 0.01 or *P < 0.05 compared to 60+ days of age, Dunnett's test after 2-way ANOVA (age × M g 2+ concentration) with repeated measures (Mg2÷ concentration) yielded P < 0.0001 for age and P < 0.0001 for interaction between the variables.
58 difference in Mg 2÷ antagonism was more obvious with these concentrations than with 1 mM Mg 2÷. To quantify developmental changes in antagonist potency, two analyses were performed. The Schild analysis utilized all concentrations of Mg 2+ that did not depress the maximal response in either age group (i.e., 0.1-3.16 mM Mg2+). In both groups, Schild plots were linear and their slopes were significantly less than unity (Fig. 4), consistent with uncompetitive antagonism. The Schild slopes were not significantly different (Table II). However, the X-intercept was significantly lower in the 10- to 15-day-old group. Although the X-intercept of the Schild plot cannot be regarded as a dissociation constant in this instance, it remains useful for the purpose of comparison. The difference in these intercepts suggests a greater than 3-fold increase in the potency of Mg 2+ between 10 and 15 days of age and adulthood. Calculation of ECs0 values for Mg 2+ (against 10/~M N M D A ) supported this idea. The ECs0 increased more than 2-fold during development (Table 1I). Second, 10 mM Mg z÷ markedly depressed the maximal response to N M D A in slices from adult rats, but not in slices from 10-15 day old rats (Fig. 3, Table II). This change could signify an increase in the efficacy of Mg 2÷ with age or it could simply be another manifestation of the age-related increase in Mg 2+ potency. Third, in the 10- to 15-day-old group, about one-third
g
lOO
10-15 d
0.8
o 10-15 d • 60+ d
~
0.4
m
0.0 -0.4 -0.8 I
-4.0
h
I
-3.5 -3.0 log [M92"] (M)
J
-2.5
Fig. 4. Schild plot for Mg2÷ derived from data in Fig. 3. The plot was shifted toward higher Mg2+ concentrations for 10- to 15-day:old rats compared to adults. Schild parameters are given in Table II. Values are means + S.E.M.
of the preparations generated a maximal depolarizing response to N M D A that was at least 10% greater in the presence of Mg 2+ than in its absence (Table II). The increase occurred as often in 0.1 mM Mg 2÷ as in 3.16 mM Mg 2÷. This phenomenon was observed in fewer than 10% of the preparations from adult animals, and in these cases only with 0.1-0.316 mM Mg 2÷. No concentration of added Mg 2÷ increased the maximal response to A M P A at any age. The hyperpolarization evoked by GABAergic synaptic inhibition has been shown to augment the antagonistic action of Mg 2÷ toward N M D A receptor activation 12. Therefore one possible explanation for a developmental
TABLE II o "d g
50
Developmental changes observed in the NMDA-induced depolarizations of CA 1 pyramidal cells
I
i
i
ECs0 (against 10 aM NMDA) and Schild analyses were performed with compiled data from the number of slices shown in parentheses. Age (days)
C
1oo
60+ d
o
10-15
o/
60+
g ~ ~ ~ g
50
i
-5.5
-5.0 -4.5 1o9 [NMDA] (M)
-4.0
Fig. 3. NMDA concentration-response relationships in the presence of various concentrations of added Mg2+ (0-10 mM). (O) Without added Mg2÷; (0) with Mg2÷. Values are means + S.E.M. for the foilowing.Mg2+ concentrations and number of slices (10-15 days, 60+ days): 0 (39,22), 0.1 mM (6,6), 0.316 mM (6,6), 1 mM (21,9), 3.16 mM (7,6), and 10 mM (6,6).
Schild slope 0.75+0.07 (37) 0.60+0.05 (21) X-intercept of Schild plot -3.12+0.04 -3.63+0.05* log-1 X-intercept 759 aM 234/~M pEC5oof Mg2÷ 3.40+ 0.05 (32) 3.76+0.05** (8) ECso of Mg2+ 400/tM 170/,M Depression of maximal responses by 10 mM Mg~÷ in (n) slices 12+6% (6) 44+6% a (6) Fraction of slices in which Mg2÷ increased maximal response 12/37 2/23° *P < 0.0001, Student's t-test. **P < 0.001, Student's t-test corrected for unequal Variances. ~P < 0.005, Student's t-test. ~'P< 0.05, x2-analysis.
59 increase in the potency of Mg 2+ is that N M D A activates G A B A neurons in our preparation and this action produces a greater pyramidal cell hyperpolarization in slices from adult animals than in slices from 10-15-dayold animals. To test this possibility, we studied the effects of the ),-aminobutyric acid (GABA) antagonist picrotoxin on NMDA-induced depolarizations in the presence of 1 mM Mg 2+. Picrotoxin (50 ~M) changed the ECs0 of NMDA by less than 5% at either age. Thus we conclude that the developmental increase in Mg 2+ potency is not related to changes in GABAergic inhibition. DISCUSSION Our results demonstrate that Mg 2+ is less able to reduce NMDA-evoked depolarizing responses in CA1 hippocampal pyramidal cells of developing rats than in mature pyramidal cells. This finding is consistent with the lesser voltage-sensitivity of NMDA-induced currents during early development 5. It also agrees with reports of Brady and Swann 7'8 that, in contrast to hippocampal slices from mature animals, removing Mg 2+ from medium superfusing slices from developing rats did not relieve the voltage dependence of NMDA-induced currents. A reduced sensitivity to Mg 2+ would facilitate the participation of NMDA receptors in developmental events. One must consider the possibility that artifacts related to the grease-gap method accounted for the change in Mg2+ potency. It is often difficult to obtain a true maximum in grease-gap studies of excitatory amino acids, due to excitotoxicity, desensitization and/or incomplete equilibration 33'44. This can lead to underestimates of the true ECso values. In addition, agonist concentrationresponse curves will be influenced by any activation or deactivation of voltage-dependent ion conductances that may occur secondary to a change in membrane potential. Developmental changes in such properties of area CA1 as the percentage of extracellular space, the resistance of pyramidal cell membranes and dendritic growth could also influence agonist potency. Finally, Mg 2+ might affect agonist potency in some way other than by blocking the NMDA channel. Each of these factors would be expected to alter the potency of all excitants, however. Our findings that the potency of AMPA remained essentially constant in the same preparations where developmental changes in Mg 2+ regulation of the NMDA receptor were observed and that Mg2÷ did not alter AMPA potency at any age support the view that differences among age groups in the action of Mg2+ reflect a developmental change specific to the NMDA receptor. Another possible explanation for the enhancement of Mg 2+ potency during development is that this change
occurs secondarily to a reduction in NMDA receptor density. The unusually high potency of NMDA on CA1 hippocampal pyramidal cells has been ascribed to the presence of a substantial receptor reserve 32. It is this high receptor density, coupled with the voltage-sensitivity of Mg2+-evoked open-channel block, that explains the non-parallel displacement of the NMDA concentrationresponse curve with increasing Mg 2+ concentration 32' 39,41 In the presence of a receptor reserve, agonist potency is directly related to receptor density2v and the potency of an uncompetitive antagonist, such as Mg 2+, is inversely related to receptor density41. Thus the lesser potency of Mg 2÷ in slices from developing rats might be explained by a relatively high NMDA receptor density. Different developmental profiles of NMDA receptor density have been reported by different groups. In three reports, the overall pattern of NMDA receptor ontogeny in the rat hippocampus has been described as a transient postnatal increase followed by a decrease to adult levels. In one study, this transient increase occurred in the CA1 area so early that receptor density actually declined slightly over the postnatal period covered by the present study48. In the two other studies, however, the transient increase did not peak until the third postnatal week 25"35, a pattern inconsistent with the changes we observed in Mg2+ potency. The exact timing of the transient increase in NMDA receptor density may depend upon the strain of rat used 42. A fourth study found that NMDA receptor binding increases continuously during postnatal development 43. This study utilized Sprague-Dawley rats, the same strain used in the present study. The changes in Mg2+ potency probably cannot be explained simply by changes in NMDA receptor density, however. First, there is no evidence of a developmental decrease in NMDA receptors sufficient to account for the magnitude of the change in the ECso of Mg 2+. Second, the developmental increase in the potency of NMDA could result from an increase in receptor density during development, such as that reported by Peterson et al. 43, but increased receptor density predicts a developmental reduction in the potency of Mg 2+. Although more work is needed to determine the cellular mechanism that underlies the developmental increase in the potency of Mg 2+, we favor the possibility that Mg 2+ sensitivity is altered by changes in the subunit composition of the NMDA receptor, in the .post-translational modification of the receptor 3~3~ or in the mix of NMDA receptor subtypes ~4. A precedent exists for a change in the subunit composition of an ion channelforming receptor during development 37. Indeed the results of Ben-Ari et al. 5, Brady and Swann 7'8 and McDonald et al. 35, in conjunction with the present study, are most consistent with a developmental change in
60 receptor structure. However, we cannot rule out the possibility that developmental changes in such factors as the placement of N M D A receptors on the pyramidal cell 22 or the resting membrane potential of the pyramidal cell could also play a role. These issues cannot be addressed with the grease-gap preparation. In slices from 10-15-day-old rats, Mg 2÷ frequently increased the maximal response of CA1 pyramidal cells to NMDA. This effect was observed in only two slices from adult animals, a frequency that probably reflects the normal variability of the grease-gap method. Enhancement of the maximal response to N M D A could theoretically arise from an effect of Mg 2+ on any of the factors that determine response amplitude in grease-gap experiments 33'44. For example, Mg 2+ might increase the electrical resistance of the unmyelinated pyramidal cell axons present in the developing CA1 area more than the resistance of the myelinated axons present in the adult CA1 area. If this were the case, however, we would have expected a similar enhancement of the maximal response to A M P A . No such effect of Mg 2+ was observed. A more likely possibility is that Mg 2÷, in addition to blocking the open channel, affects N M D A receptor function in some other way. Submillimolar concentrations of Mg 2÷ have been shown to enhance the binding of glycine to the N M D A receptor 34. Mg 2+ reportedly increases both the affinity of the receptor for glycine and the maximal quantity of glycine bound. Such an effect could potentiate responses to N M D A 47, even in the presence of a saturating glycine concentration. This action of Mg 2÷ would serve to counteract its channel blocking action. It is possible that the potentiation of glycine binding by Mg 2+ is more significant in developing rats than in adults. Mg 2+ is probably the most important antagonist of N M D A receptor function present in the intact brain. Therefore a lesser effect of Mg 2+ during development
would be expected to facilitate participation of the N M D A receptor in synaptic events ~z'~3. This developmental difference might contribute to the increased magnitude of long-term potentiation observed in the CA 1 area of 15-day-old rats 23'46. It might also explain the observation that, even in the presence of a physiological Mg 2+ concentration, excitatory synaptic responses of immature cerebellar granule cells have a significant N M D A receptor component, whereas equivalent responses from adult preparations only showed an N M D A receptor component when Mg 2+ was removed 2{}. Additionally, the spontaneous activity of entorhinal cortical neurons 26 and CA3 hippocampal pyramidal cells 4 recorded in slices from immature rats continued in the presence of Mg 2÷, but was suppressed by the competitive N M D A receptor antagonist D-2-amino-5-phosphonovalerate (AP5). In contrast, Mg 2÷ suppressed all such activity in slices from adult rats. A lesser Mg 2÷ antagonism would also facilitate the trophic-like actions of N M D A receptor agonists ~'38'4°. Indeed developmental changes in N M D A receptor function may have equal or even greater implications for neuronal growth and synapse formation than for synaptic activity. For technical reasons, the present study was limited to animals 10 days of age and older. The trend apparent in our data suggests that Mg 2÷ would have been an even weaker antagonist in slices from younger animals. This possibility needs to be tested with quantitative methods of studying N M D A receptor function that are applicable to animals younger than 10 days. Further experimentation is also necessary to determine the extent to which the developmental changes we observed in responses of CA1 pyramidal cells apply to CNS neurons generally.
REFERENCES
7 Brady, R.J. and Swann, J.W., Developmental change in the magnesium sensitivity of the NMDA response in hippocampal CA3 pyramidal cells, Soc. Neurosci. Abstr., 15 (1989) 327. 8 Brady, R.J. and Swann, J.W., The effects of extracellular calcium on the epileptiform activity and NMDA responses are different in mature and immature hippocampal slices, Soc. Neurosci. Abstr., 14 (1988) 239. 9 Brenneman, D.E., Forsythe, I.D., Nicol, T. and Nelson, P.G., N-Methyl-D-aspartate receptors influence neuronal survival in developing spinal cord cultures, Dev. Brain Res., 51 (1990) 63-68. 10 Brewer, G.J. and Cotman, C.W., NMDA receptor regulation of neuronal morphology in cultured hippocampal neurons, Neurosci. Left., 99 (1989) 268-273. 11 Cline, H.T., Debski, E.A. and Constantine-Paton, M., NMethyl-D-aspartate receptor antagonist desegregates eye-specific stripes, Proc. Natl. Acad. Sci. U.S.A., 84 (1987) 4342-4345. 12 Collingridge, G.L., Herron, C.E. and Lester, R.A.J., Synaptic activation of N-methyl-i>aspartate receptors in the Schaffer collateral-commissural pathway of rat hippocampus, J. Physiol., 399 (1988) 283-300.
1 Bal~zs, R., Hack, N. and Jergensen, O.S., Stimulation of the N-methyl-D-aspartate receptor has atrophic effect on differentiating cerebellar granule cells, Neurosci. Lett., 87 (1988) 80-86. 2 Bal~izs, R., Hack, N., Jergenson, O.S. and Cotman, C.W., N-Methyl-D-aspartate promotes the survival of cerebellar granule cells: pharmacological characterization, Neurosci. Lett., 101 (1989) 241-246. 3 Bartlett, M.C., Salter, M.W. and MacDonald, J.F., Modulation of NMDA currents by intracellular phosphorylation, Soc. Neurosci. Abstr., 15 (1989) 535. 4 Ben-Ari, Y., Cherubini, E., Corradetti, R. and Gaiarsa, J.L., Giant synaptic potentials in immature rat CA3 hippocampal neurones, J. Physiol., 416 (1989) 303-325. 5 Ben-Ari, Y., Cherubini, E. and Krnjevid, K., Changes in voltage dependence of NMDA currents during development, Neurosci. Lea., 94 (1988) 88-92. 6 Bode-Greuel, K.M. and Singer, W., The development of N-methyl-D-aspartate receptors in cat visual cortex, Dev. Brain Res., 46 (1989) 197-204.
Acknowledgements. We thank Ms. C. Kelsall for secretarial assistance. This study was supported by NIH Grant NS 16064 (J.V.N.) and by an NSF predoctoral fellowship (M.A.B.).
61 13 Collingridge, G.L., Herron, C.E. and Lester, R.A.J., Frequency-dependent N-methyl-D-aspartate receptor-mediated synaptic transmission in rat hippocampus, J. Physiol., 399 (1988) 301-312. 14 Cotman, C.W., Bridges, R.J., Taube, J.S., Clark, A.S., Geddes, J.W. and Monaghan, D.T., The role of the NMDA receptor in central nervous system plasticity and pathology, J. NIH Res., 1 (1989) 65-74. 15 Dupont, J.L., Gardette, R. and Crepel, F., Postnatal development of the chemosensitivity of rat cerebellar Purkinje cells to excitatory amino acids. An in vitro study, Dev. Brain Res., 34 (1987) 59-68. 16 Fleming, W.W., Westfall, D.P., De La Lande, I.S. and Jellett, L.B., Lognormal distribution of equieffective doses of norepinephrine and acetylcholine in several tissues, J. Pharmacol. Exp. Ther., 181 (1972) 339-345. 17 Fox, K., Sato, H. and Daw, N., The location and function of NMDA receptors in cat and kitten visual cortex, J. Neurosci., 9 (1989) 2443-2454. 18 Gaddum, J.H., Lognormal distributions, Nature, 156 (1945) 463-466. 19 Garthwaite, G., Yamini, B. and Garthwaite, J., Selective loss of Purkinje and granule cell responsiveness to N-methyl-Daspartate in rat cerebellum during development, Dev. Brain Res., 36 (1987) 288-292. 20 Garthwaite, J. and Brodbelt, A.R., Synaptic activation of N-methyl-D-aspartate and non-N-methyl-D-aspartate receptors in the mossy fiber pathway in adult and immature rat cerebellar slices, Neuroscience, 29 (1989) 401-412. 21 Gu, Q., Bear, M.F. and Singer, W., Blockade of NMDAreceptors prevents ocularity changes in kitten visual cortex after reversed monocular deprivation, Dev. Brain Res., 47 (1989) 281-288. 22 Hamon, B. and Heinemann, U., Developmental changes in neuronal sensitivity to excitatory amino acids in area CA1 of the rat hippocampus, Dev. Brain Res., 38 (1988) 286-290. 23 Harris, K.M. and Teyler, T.J., Developmental onset of longterm potentiation in area CA1 of the rat hippocampus, J. Physiol., 346 (1984) 27-48. 24 Ikonomidou, C., Mosinger, J.L., Shahid Salles, K., Labruyere, J. and Olney, J.W., Sensitivity of the developing rat brain to hypobaric/ischemic damage parallels sensitivity to N-methylD-aspartate neurotoxieity, J. Neurosci., 9 (1989) 2809-2818. 25 Insel, T.R., Miller, L.F., Johnson, A.E. and Gelhard, R., Ontogeny of excitatory amino acid receptor subtypes in rat brain, Soc. Neurosci. Abstr., 14 (1988) 483. 26 Jones, R.S.G. and Heinemann, U., Spontaneous activity mediated by NMDA receptors in immature rat entorhinal cortex in vitro, Neurosci. Lett., 104 (1989) 93-98. 27 Kenakin, T.P.. The classification of drugs and drug receptors in isolated tissues, Pharmacol. Rev., 36 (1984) 165-222. 28 Kleinschmidt, A., Bear, M.E and Singer, W.E, Blockade of 'NMDA' receptors disrupts experience-dependent plasticity of kitten striate cortex, Science, 238 (1987) 355-358. 29 Lee, S.S., Murata, R. and Matsuura, S., Developmental study of hippocampal kindling, Epilepsia, 30 (1989) 266-270. 30 Lincoln, J., Coopersmith, R., Harris, E.W., Cotman, C.W. and Leon, M., NMDA receptor activation and early olfactory learning, Dev. Brain Res., 39 (1988) 309-312. 31 MacDonald, J.F., Mody, I. and Salter, M.W., Regulation of N-methyI-D-aspartate receptors revealed by intracellular dialysis of routine neurones in culture, J. Physiol., 414 (1989) 17-34. 32 Martin, D., Bowe, M.A. and Nadler, J.V., A grease-gap method for studying the excitatory amino acid pharmacology of CA1
33
34
35 36
37
38
39 40
41 42 43 44 45 46 47 48
49 50
hippocampal pyramidal cells, J. Neurosci. Methods, 29 (1989) 107-114. Martin, D. and Nadler, J.V., Grease-gap methods for studying the pharmacology of excitatory amino acids on CNS neurons. In P.M. Conn. (Ed.), Electrophysiology and Microinjection, Methods in Neuroscience, Vol. 4, Academic Press, Orlando, in press. Marviz6n, J.C.G. and Skolnick, P., [3H]Glycine binding is modulated by Mg2+ and other ligands of the NMDA receptorcation channel complex, Eur. J. Pharmacol., 151 (1988) 157158, McDonald, J.W., Johnston, M.V. and Young, A.B., Ontogeny of the receptors comprising the NMDA receptor complex, Soc. Neurosci. Abstr., 15 (1989) 198. McDonald, J.W., Silverstein, F.S. and Johnston, M.V., Neurotoxicity of N-methyl-D-aspartate is markedly enhanced in developing rat central nervous system, Brain Res., 459 (1988) 200-203. Mishina, M., Takai, T., Imoto, K., Noda, M., Takahashi, T., Numa, S., Methfessel, C. and Sakmann, B., Molecular distinction between fetal and adult forms of muscle acetylcholine receptor, Nature, 321 (1986) 406-411. Moran, J. and Patel, A.J., Stimulation of the N-methylD-aspartate receptor promotes the biochemical differentiation of cerebellar granule neurons and not astrocytes, Brain Res., 486 (1989) 15-25. Nowak, L., Bregestovski, P., Ascher, P., Herbet, A. and Prochiantz, A., Magnesium gates glutamate-activated channels in mouse central neurons, Nature, 307 (1984) 462-465. Pearce, I.A., Cambray-Deakin, M.A. and Burgoyne, R.D., Glutamate acting on NMDA receptors stimulates neurite outgrowth from cerebellar granule cells, FEBS Lett., 223 (1987) 143-147. Pennefather, P. and Quastel, D.M.J., Modification of doseresponse curves by effector blockade and uncompetitive antagonism, Mol. Pharmacol., 22 (1982) 369-380. Peterson, C. and Cotman, C.W., Strain-dependent decrease in glutamate binding to the N-metbyl-D-aspartic acid receptor during aging, Neurosci. Lett., 104 (1989) 309-313. Peterson, C., Neal, J.H. and Cotman, C.W., Development of N-methyl-D-aspartate excitotoxicity in cultured hippocampal neurons, Dev. Brain Res., 48 (1989) 187-195. Simmonds, M.A., Use of slices for quantitative pharmacology. In H. Jahnsen (Ed.), In Vitro Preparations from Vertebrate Systems, Wiley, Chicbester, in press. Tallarida, R.J. and Murray, R.B., Manual of Pharmacological Calculations with Computer Programs. 2nd edn., Springer, New York, 1987. Teyler, T.J., Perkins, A.T. and Harris, K.M., The development of longterm potentiation in hippocampus and neocortex, Neuropsychologia, 27 (1989) 31-39. Thomson, A.M., Walker, V.E. and Flynn, D.M., Glycine enchances NMDA-receptor mediated synaptic potentials in neocortical slices, Nature, 338 (1989) 422-424. Tremblay, E., Roisin, M.P., Repressa, A., Charriaut-Marlangue, C. and Ben-Ari, Y., Transient increased density of NMDA binding sites in the developing rat hippocampus, Brain Res., 461 (1988) 393-396. Tsumoto, T., Hagihara, K., Sato, H. and Hata, Y., NMDA receptors in the visual cortex of young kittens are more effective than those of adult cats, Nature, 327 (1987) 513-514. Williams, T.L., Smith, D.A.S., Burton, N.R. and Stone, T.W., Amino acid pharmacology in neocortical slices: evidence for bimolecular actions from an extension of the Hill and GaddumSchild equations, Br. J. Pharmacol., 95 (1988) 805-810,