- Email: [email protected]
NMDA receptors mediate expression of one form of functional plasticity induced by olfactory deprivation
BRAIN RESEARCH ELSEVIER
Brain Research 677 (1995) 238-242
Research report
NMDA receptors mediate expression of one form of functional plasticity induced by olfactory deprivation Donald A. Wilson * Department of Zoology, Universityof Oklahoma, Norman, OK 73019, USA
Accepted 17 January 1995
Abstract
Unilateral olfactory deprivation for the first 3 postnatal weeks results in an enhancement of granule cell mediated, feedback inhibition of mitral cells in the rat olfactory bulb. Granule cells are excited by mitral cells by both non-NMDA and NMDA receptors. The present report describes the effect of the NMDA antagonist MK-801 (0.75 mg/kg) on the expression of deprivation enhanced inhibition. The results demonstrate that (1) enhancement of lateral olfactory tract paired-pulse inhibition of evoked potentials in deprived bulbs was stimulus intensity dependent, with the greatest difference expressed at highest stimulus intensities; (2) MK-801 reduced inhibition in both undeprived and deprived bulbs in a stimulus intensity dependent manner, with the greatest reduction occurring at highest stimulus intensities; (3) MK-801 eliminated the stimulus intensity effect on inhibition in both groups; and (4) following MK-801, deprived bulbs showed the same or less inhibition than undeprived bulbs. Keywords: Olfaction; Sensory deprivation; N-Methyl-D-aspartate; Neural plasticity; Olfactory deprivation
1. Introduction
Experience during early development can profoundly influence sensory system structural and functional ontogeny. For example, in the olfactory system of rats, unilateral olfactory deprivation, induced by closure of an external nare, produces dramatic changes in olfactory system anatomy [4,11]. The olfactory bulb, the first central relay for olfactory information, can be reduced in volume by as much as 25% following olfactory deprivation during the first few postnatal weeks, primarily due to decreases in growth rate and survival of several bulb cell types [4,10]. Most of the anatomical effects of early olfactory deprivation are on late developing interneurons. Granule cells are GABAergic interneurons controlling the excitability of bulb output neurons, the mitral cells. Most granule cell neurogenesis occurs postnatally [1], and absence of odor stimulation during this time re-
* Corresponding author. Fax: (1) (405) 325-7560. email: [email protected]. 0006-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0006-8993(95)00151-4
sults in large decreases in the number of granule cells [6,13]. Our previous work has shown that, despite the decrease in granule cell number, early olfactory deprivation enhances granule cell mediated inhibition [19,22]. Late onset deprivation, which does not affect granule cell number, similarly does not affect granule cell mediated inhibition [22]. The present report is a further investigation into the paradoxical effects of early deprivation on granule cell function. Granule cell mediated inhibition can be assayed by electrical stimulation of mitral cell axons (the lateral olfactory tract, LOT), producing antidromic activation of mitral cells and monosynaptic excitation of granule cells [12]. Current evidence strongly suggests that mitral cells are glutamatergic [15,16] and granule cell post-synaptic responses are mediated by both nonN M D A and N M D A glutamate receptors [14,17,20]. Single-pulse stimulation of the L O T produces an EPSP in granule cells recorded intracellularly, which is seen as a short latency negative evoked potential recorded extracellularly in the external plexiform layer of the bulb. The duration of the intracellular EPSP and of the extracellular evoked potential are prolonged by an
239
D.4. Wilson~Brain Research 677 (1995) 238-242
N M D A component which can be reduced by N M D A antagonists [8,14,17,20]. In turn, by reducing granule cell excitation, N M D A antagonists reduce granule cell mediated inhibition of mitral cells [8,20]. Activation of N M D A receptors, thus, can amplify granule cell mediated inhibition. It seemed possible, therefore, that in the deprived bulb the deprivation-induced decrease in granule cell number may be compensated for by an NMDA-mediated enhancement in activity of spared granule cells. This compensation, or over-compensation, then could result in the observed deprivation-induced potentiation of inhibition. The present paper examined the role of N M D A receptors in the expression of enhanced inhibition in the deprived olfactory bulb.
2. Materials and methods
Male Wistar rats, born in our colony from Charles River (St. Louis) stock were used as subjects. Mothers and litters were housed in polypropylene cages lined with wood chips. Mothers and pups were allowed ad lib access to food and water. Lights were maintained on a 12:12 h cycle with lights on at 7:00 a.m. Date of birth was considered postnatal day 0 (PN0). On PN1, pups had a single nare sealed using the method of Meisami [10]. Basically, pups were cold anesthetized and a single hare was cauterized (n = 4 from 4 different litters) or the pup was cauterized on the snout as a control (n = 5 from 4 different litters). Pups were returned to the nest until testing on P N 2 0 PN25. On the day of testing, pups were anesthetized with urethane (1.5 g/kg, i.p.) and placed in a stereotaxic apparatus. A bipolar, stainless steel stimulating electrode was placed in the LOT. Responses were recorded with glass microelectrodes filled with 2 M NaCI placed in the external plexiform layer or granule cell layer of the olfactory bulb. Responses to paired pulse (equal intensity) stimulation of the L O T (0.1 ms pulse duration, 20 ms inter-pulse interval; 10-30 s inter-pair interval) were amplified and fed to a Macintosh Quadra for analysis with Spike 2 software (Cambridge Electronic Design). Paired-pulses were delivered at each of 9 intensities (200-1000 I~A). Five paired-stimuli were delivered at each intensity and waveforms computer averaged for measurement. Response measures included: amplitude of both conditioning and test responses, and conditioning response width at half amplitude (half-width; Fig. 1). Following baseline recording, animals were injected with MK-801 (0.75 m g / k g , i.p., obtained from R.B.I.) and paired-pulse testing repeated at 15-30 min post-injection. Data were analyzed with ANOVA's and simple regression analyses of the individual data.
3. Results
Stimulation of the L O T produced characteristic responses in both deprived and undeprived olfactory bulbs. Fig. 1 shows representative responses from a deprived olfactory bulb. Evoked response amplitude varied with stimulus intensity, and did not significantly differ between groups (Group x intensity ANOVA, main effect of g r o u p , F1,63 = 3.13, N.S.; Maximal response amplitude: undeprived = 3.0 _+ 0.6 mV; deprived = 2.7 _+ 0.9 mV; unpaired t-test, t 7 = 0.37, N.S.). The N M D A antagonist MK-801 reduced response amplitude in both undeprived and deprived pups (Drug x intensity ANOVA's, main effect of drug: undeprived, F1,36 = 20.30, P < 0.01; deprived F1,27 = 118.03, P < 0.01), but amplitude still varied as a function of stimulus intensity (Fig. 2a). In addition to evoked potential amplitude, halfwidth also varied as a function of stimulus intensity (Fig. 2b). Increasing L O T stimulus intensity increased evoked response halfwidth, in an intensity dependent manner (regression analysis, intensity vs. halfwidth calculated on raw data: control, slope = 0.35, P < 0.05, t-test; deprived, slope = 0.56, P < 0.05). Halfwidths in undeprived and deprived bulbs did not significantly differ
Deprived olfactory bulb E.P.'s
Pre-drug low
_
MK-801 low --
Pre-drug high --
U
5 ms, O.5 mV
-1
MK-801
high
--
Fig. 1. Evoked potentials recorded in the external plexiformlayer in a deprived olfactory bulb (responses in undeprived bulbs were similar with exceptionsnoted in text). Low intensitypaired-pulse stimulation (20 ms IPI) of the LOT produced very little inhibition of the test response, or facilitation. Response amplitude was measured peak to peak as shown in the LOW responses. High intensity stimulation produced greater inhibition. Evoked potential halfwidth was measured as shown by the arrow in the HIGH intensity responses. The NMDA antagonist MKo801 had little effect on halfwidth or inhibition at low stimulus intensities, but greatly reduced both halfwidth and inhibition at high intensities.
D.A. Wilson/Brain Research 677 (1995) 238-242
240
A
EP A M P L I T U D E
4.o3.5"
11) C ~)
3.0-
O ~
¢- ~ 0--
1.5,
0
0.5"
Undeprivedpre Undeprivedpost Deprived pro
(.3
O.O
J-
2;o
4;o
Deprivedpost
6;0 " 8;o lO'OO
LOT stimulus intensity (IJA)
B 18.0]
EP HALFWIDTH
17-°1 ~ 16.o1
~ 1so1 8 8 14o1
~=
Undeprived Pre
~.,= 13.o.
~
~.o.
="
110O ~ 100:~ .E 9.0. .~
t~
(,3 t-5 "~
8.0. 7o 6.0 i
DeprivedPre
dependent on two opposing factors, pre-synaptic facilitation and feedback inhibition. At this short inter-pulse interval, inhibition dominates at most stimulus intensities. However, the magnitude of paired-pulse inhibition varied with stimulus intensity, with higher intensities producing greater inhibition (Group × intensity ANOVA, main effect of intensity: F 8 , 6 3 - - 4 . 4 9 , P < 0.01). Deprived pups showed significantly greater paired-pulse inhibition than undeprived (group × intensity ANOVA, main effect of g r o u p : F 1 , 6 3 - - 4 . 0 0 , P < 0.05), with this effect primarily displayed at the highest stimulus intensities (e.g., 200 /zA, post-hoc Fisher test, undeprived vs. deprived, N.S.; 1000 ,/xA, post-hoc Fisher test, undeprived vs. deprived, P < 0.05; Fig. 3B). MK-801 had three primary effects on paired-pulse inhibition (Fig. 3). First, inhibition was reduced in both undeprived and deprived pups (Drug × stimulus intensity ANOVA, main effect of drug: undeprived, F1,36 =
DepnvedPost UndepdvedPost
A PAIRED-PULSE
150.0
INHIBITION
1400
" 4ob
~
sob
~o
LOT stimulus intensity ( ~A)
Fig. 2. A: evoked potential amplitude (conditioning pulse) was significantly enhanced by increasing LOT stimulus intensity. Responses in undeprived and deprived bulbs did not significantly differ before MK-801 injection. MK-801 significantly reduced amplitude but had no effect on the intensity-amplitude relationship. Standard error bars are excluded for clarity (in all cases: 0.4 mV < S.E.M. < 0.9 mV). B: evoked potential halfwidth was also enhanced by increasing LOT stimulus intensity. Responses in undeprived and deprived bulbs did not significantly differ before MK-801 injection. MK-801 significantly reduced halfwidth and the intensity-halfwidth relationship.
~ ~~ O..¢: ~._~ ®~
~80.0 120.o 110.o . . . . . . 100.o
0 0o
90.C~
~ o~
80o:
~
_
_
_
_
~
.
~
DeprivedPost UndeprivedPost
UndeprivedPre
6oo: J.
SO.(~ 40.0 •
2;0
" 400
" 6;0
'
8;0
Deprived Pre
" 10100
LOT stimulus intensity ~ A )
140.0, ~
from each other (Group × Stimulus intensity ANOVA, main effect of group, F1,63 = 0.49, N.S.). MK-801 significantly reduced evoked potential halfwidth in both undeprived and deprived groups (Drug × stimulus intensity ANOVA, main effect of drug: undeprived, F1,36 = 31.82, P<0.001; deprived, Fl,27 = 40.48, P < 0.01). Furthermore, MK-801 significantly reduced the dependence of halfwidth on stimulus intensity in undeprived animals (regression analysis: slope = 0.06, N.S.). Increasing stimulus intensity did not affect response halfwidth after NMDA receptor blockade (Fig. 2b). In deprived animals, the relationship between stimulus intensity and halfwidth was also markedly reduced by MK-801, although the regression was statistically significant (slope = 0.29, P < 0.05). A single conditioning pulse modified the amplitude of the response to a test pulse delivered 20 ms later (Figs. 1 and 3). The amplitude of the test response is
120.0 •
._= ~.._~ lOO.O.
[] Undeprived Pre
~ s0.0.
[ ] Undeprived Post
;°00 ~®
40.0-
a.
20.0-
•
Deprived Pre
[ ] Deprived Post ,
200 uA
1000 uA
LOT stimulus intensity (vA I
Fig. 3. A: test response amplitude as a percent of conditioning response amplitude in deprived and undeprived bulbs, pre- and post-MK-801 injection. Pre-MK-801, deprived bulbs showed significantly greater inhibition that undeprived bulbs, at high stimulus intensities. MK-801 significantly reduced this inhibition in both groups, and to a greater extent in deprived than undeprived bulbs. Post-MK-801, deprived bulbs showed significantly less inhibition (greater facilitation) than undeprived bulbs. B: replot of responses to the highest and lowest stimulus intensities. Asterisks signify significant difference from same intensity post-drug and from low intensity pre-drug, P < 0.05.
D~4. Wilson~Brain Research 677 (1995) 238-242
28.46, P < 0.01; deprived, F1,27 = 187.13, P < 0.01). Second, the greatest reduction in inhibition was in the deprived animals, such that following MK-801, deprived animals showed less inhibition (greater facilitation) than undeprived (group x intensity ANOVA, main effect of group: F1,63 = 11.60, P < 0.01). Finally, in contrast to pre-drug pups, the magnitude of inhibition was no longer dependent on stimulus intensity following MK-801 (Group × intensity ANOVA, main effect of intensity: F8,63 =: 0.18, N.S.).
4. Discussion The results presented here confirm previous findings that early olfactory deprivation potentiates granule cell mediated inhibition [19,22]. The present results extend the previous work by demonstrating that expression of this enhanced inhibition is NMDA receptor dependent. Under conditions of either weak LOT input, which is insufficient for NMDA receptor activation, or NMDA receptor blockade by MK-801, deprived bulbs show the same or less inhibition than undeprived controls. Thus, based on these data it is suggested that deprivation-induced loss of granule cells may be compensated for during intense stimulation by an NMDA receptor mediated amplification in the excitation of spared granule cells. The NMDA mediated compensation for lost granule cells could occur through several mechanisms. For example, as granule cell targets are lost, mitral cells may make more synaptic contacts with fewer spared granule cells. This consolidation of more mitral cells onto fewer granule cells (or more of a single mitral cell's output onto a single granule cell) could increase the probability of NMDA receptor activation, and in turn, amplify inhibition. This is well demonstrated in cultures of olfactory bulb neurons, where, despite relatively few mitral cells synapsing onto individual granule cells, NMDA receptors are activated [14]. Thus, perhaps by increasing the proportion of inputs a granule cell receives from one or a few mitral cells, threshold for activating NMDA receptors can be reduced. Alternatively, there could be changes in NMDA receptor density or pre-synaptic changes in transmitter release. While the present data cannot discriminate between these and other mechanisms, they do suggest the changes may be s~abtle. For example, using halfwidth as a measure of NMDA receptor activity, no significant difference was found in halfwidth between non-drugged undeprived and deprived bulbs (Fig. 2).
241
local circuit function in the bulb, in vivo [8]. As LOT stimulus intensity increased, evoked potential amplitude increased. This is presumed to be due to a stimulus dependent increase in the number of LOT fibers activated, resulting in a larger number of co-active excitatory inputs to the granule cells. As expected, MK-801, while slightly reducing response amplitude, did not affect this intensity-amplitude relationship. The short-latency amplitude is primarily dependent on non-NMDA glutamate receptors [14]. However, evoked potential halfwidth is presumed to be strongly influenced by an NMDA receptor mediated depolarization. Given the voltage-gated nature of NMDA receptors, the NMDA component, and therefore response halfwidth, should also vary with stimulus intensity. At low stimulus intensities, granule cells should receive insufficient excitatory input to reach threshold for NMDA receptor activation. At higher stimulus intensities, with stronger excitatory drive from a larger number of co-active inputs, threshold for NMDA receptor activation would be reached and expression of the NMDA component of the evoked potential would become apparent (increase in halfwidth). This is precisely what was observed. Furthermore, MK-801 reduced response halfwidth at high stimulus intensities to near that observed with low stimulus intensities, clearly suggesting that response to low intensity stimulation does not involve NMDA receptors, in vivo. Olfactory deprivation induces at least two types of functional plasticity in bulb circuits. First, mitral/tufted cell odor-induced activity is enhanced following reopening of the sealed nare [7,18]. Second, more intense electrical antidromic stimulation of mitral/tufted cell axons reveals enhanced feedback inhibition in deprived bulbs, as described here and elsewhere [19,22]. These apparently contradictory findings (i.e. enhanced odor response rates and yet more feedback inhibition) may be at least partly reconciled by the dependence of enhanced granule cell inhibition on stimulus intensity. Activity induced by sensory input may normally function below threshold for NMDA receptor activation of granule cells, comparable to very low intensity LOT stimulation (Fig. 3). Perhaps only when odor input is associated with a strong centrifugal input, as in associative conditioning [3,21], are NMDA receptors activated. Thus, in the deprived bulb, the deprivation-induced loss of granule cells, combined with deprivation-induced loss of glomerular layer dopamine [2,5,9,22], may under natural circumstances, contribute to the observed heightened responsiveness to odors.
4.1. NMDA receptors and olfactory bulb function
Acknowledgements
These results further demonstrate the stimulus intensity dependence of NMDA receptor involvement in
This work was supported by NIDCD Grant DC00866 to D.A.W. and Regina M. Sullivan.
242
D.A. Wilson/Brain Research 677 (1995) 238-242
References [1] Altman, J. and Das, G.D., Autoradiographic and histological studies of postnatal neurogenesis. I, Jr. Comp. Neurol., 126 (1966) 337-390. [2] Baker, H., Unilateral, neonatal olfactory deprivation alters tyrosine hydroxylase expression but not aromatic amino acid decarboxylase or GABA immunoreactivity, Neuroscience, 36 (1990) 761-771. [3] Brennan, P., Kaba, H. and Keverne, E.B., Olfactory recognition: a simple memory system, Science, 250 (1990) 1223-1226. [4] Brunjes, P.C. and Frazier, L.L., Maturation and plasticity in the olfactory system of vertebrates, Brain Res. Rev., 11 (1986) 1-45. [5] Brunjes, P.C., Smith-Crafts, L.K. and McCarty, R., Unilateral odor deprivation: effects on the development of olfactory bulb catecholamines and behavior, Dev. Brain Res., 22 (1985) 1-6. [6] Frazier, L.L. and Brunjes, P.C., Unilateral odor deprivation: Early postnatal changes in olfactory bulb density and number, J. Comp. Neurol., 269 (1988) 355-370. [7] Guthrie, K.M., Wilson, D.A. and Leon, M., Early unilateral deprivation modifies olfactory bulb function, J. Neurosci., 10 (1990) 3402-3412. [8] Jacobson, I., Butcher, S. and Hamberger, A., An analysis of the effects of excitatory amino acid receptor antagonists on evoked field potentials in the olfactory bulb, Neuroscience, 19 (1986) 267-273. [9] Kosaka, T., Kosaka, K., Hama, K., Wu, J.-Y. and Nagatsu, I., Differential effect of functional olfactory deprivation on the GABAergic and catecholaminergic traits in the rat main olfactory bulb, Brain Res., 413 (1987) 197-203. [10] Meisami, E., Effects of olfactory deprivation on postnatal growth of the rat olfactory bulb utilizing a new method for production of neonatal unilateral anosmia, Brain Res., 107 (1976) 437-444. [11] Meisami, E. and Safari, L., A quantitative study of the effects of early unilateral olfactory deprivation on the number and distri-
bution of mitrat and tufted cells and of glomeruli in the rat olfactory bulb, Brain Res., 221 (1981) 81-107. [12] Mori, K., Membrane and synaptic properties of identified neurons in the olfactory bulb, Prog. Neurobiol., 29 (1987) 275-320. [13] Skeen, L.C., Due, B.R. and Douglas, F.E., Neonatal sensory deprivation reduces granule cell number in mouse olfactory bulb, Neurosci. Lett., 63 (1986) 5-10. [14] Trombley, P.Q. and Shepherd, G.M., Noradrenergic inhibition of synaptic transmission between mitral and granule cells in mammalian olfactory bulb cultures, J. Neurosci., 12 (1992) 3985-3991. [15] Trombley, P.Q. and Shepherd, G.M., Synaptic transmission and modulation in the olfactory bulb, Curr. Opin. Neurobiol., 3 (1993) 540-547. [16] Trombley, P.Q. and Westbrook, G.L., Excitatory synaptic transmission in cultures of rat olfactory bulb, J. Neurophysiol., 64 (1990) 598-606. [17] Wellis, D.P. and Kauer, J.S., GABAa and glutamate receptor involvement in dendrodendritic synaptic interactions from salamander olfactory bulb, J. Physiol., 469 (1993) 315-339. [18] Wilson, D.A., Effects of long-term ( > 12 months) unilateral olfactory deprivation on olfactory bulb single-unit response patterns to odors, Soc. Neurosci. Abstr., 19 (1993) 123. [19] Wilson, D.A., Guthrie, K.M. and Leon, M., Modification of olfactory bulb synaptic inhibition by early unilateral olfactory deprivation, Neurosci. Lett., 116 (1990)250-256. [20] Wilson, D.A., Guthrie, K.M., Smart, R., Gall, C.M. and Sullivan, R.M., NMDA receptor modulation of olfactory bulb inhibitory circuits, Assoc. Chemoreception Sci. Abstr., 15 (1993) 299. [21] Wilson, D.A. and Sullivan, R.M., Neurobiology of associative learning in the neonate: early olfactory learning, Behav. Neural Biol., 61 (1994) 1-18. [22] Wilson, D.A. and Wood, J.G., Functional consequences of unilateral olfactory deprivation: time-course and age sensitivity, Neuroscience, 49 (1992) 183-192.
Brain Research 677 (1995) 238-242
Research report
NMDA receptors mediate expression of one form of functional plasticity induced by olfactory deprivation Donald A. Wilson * Department of Zoology, Universityof Oklahoma, Norman, OK 73019, USA
Accepted 17 January 1995
Abstract
Unilateral olfactory deprivation for the first 3 postnatal weeks results in an enhancement of granule cell mediated, feedback inhibition of mitral cells in the rat olfactory bulb. Granule cells are excited by mitral cells by both non-NMDA and NMDA receptors. The present report describes the effect of the NMDA antagonist MK-801 (0.75 mg/kg) on the expression of deprivation enhanced inhibition. The results demonstrate that (1) enhancement of lateral olfactory tract paired-pulse inhibition of evoked potentials in deprived bulbs was stimulus intensity dependent, with the greatest difference expressed at highest stimulus intensities; (2) MK-801 reduced inhibition in both undeprived and deprived bulbs in a stimulus intensity dependent manner, with the greatest reduction occurring at highest stimulus intensities; (3) MK-801 eliminated the stimulus intensity effect on inhibition in both groups; and (4) following MK-801, deprived bulbs showed the same or less inhibition than undeprived bulbs. Keywords: Olfaction; Sensory deprivation; N-Methyl-D-aspartate; Neural plasticity; Olfactory deprivation
1. Introduction
Experience during early development can profoundly influence sensory system structural and functional ontogeny. For example, in the olfactory system of rats, unilateral olfactory deprivation, induced by closure of an external nare, produces dramatic changes in olfactory system anatomy [4,11]. The olfactory bulb, the first central relay for olfactory information, can be reduced in volume by as much as 25% following olfactory deprivation during the first few postnatal weeks, primarily due to decreases in growth rate and survival of several bulb cell types [4,10]. Most of the anatomical effects of early olfactory deprivation are on late developing interneurons. Granule cells are GABAergic interneurons controlling the excitability of bulb output neurons, the mitral cells. Most granule cell neurogenesis occurs postnatally [1], and absence of odor stimulation during this time re-
* Corresponding author. Fax: (1) (405) 325-7560. email: [email protected]. 0006-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0006-8993(95)00151-4
sults in large decreases in the number of granule cells [6,13]. Our previous work has shown that, despite the decrease in granule cell number, early olfactory deprivation enhances granule cell mediated inhibition [19,22]. Late onset deprivation, which does not affect granule cell number, similarly does not affect granule cell mediated inhibition [22]. The present report is a further investigation into the paradoxical effects of early deprivation on granule cell function. Granule cell mediated inhibition can be assayed by electrical stimulation of mitral cell axons (the lateral olfactory tract, LOT), producing antidromic activation of mitral cells and monosynaptic excitation of granule cells [12]. Current evidence strongly suggests that mitral cells are glutamatergic [15,16] and granule cell post-synaptic responses are mediated by both nonN M D A and N M D A glutamate receptors [14,17,20]. Single-pulse stimulation of the L O T produces an EPSP in granule cells recorded intracellularly, which is seen as a short latency negative evoked potential recorded extracellularly in the external plexiform layer of the bulb. The duration of the intracellular EPSP and of the extracellular evoked potential are prolonged by an
239
D.4. Wilson~Brain Research 677 (1995) 238-242
N M D A component which can be reduced by N M D A antagonists [8,14,17,20]. In turn, by reducing granule cell excitation, N M D A antagonists reduce granule cell mediated inhibition of mitral cells [8,20]. Activation of N M D A receptors, thus, can amplify granule cell mediated inhibition. It seemed possible, therefore, that in the deprived bulb the deprivation-induced decrease in granule cell number may be compensated for by an NMDA-mediated enhancement in activity of spared granule cells. This compensation, or over-compensation, then could result in the observed deprivation-induced potentiation of inhibition. The present paper examined the role of N M D A receptors in the expression of enhanced inhibition in the deprived olfactory bulb.
2. Materials and methods
Male Wistar rats, born in our colony from Charles River (St. Louis) stock were used as subjects. Mothers and litters were housed in polypropylene cages lined with wood chips. Mothers and pups were allowed ad lib access to food and water. Lights were maintained on a 12:12 h cycle with lights on at 7:00 a.m. Date of birth was considered postnatal day 0 (PN0). On PN1, pups had a single nare sealed using the method of Meisami [10]. Basically, pups were cold anesthetized and a single hare was cauterized (n = 4 from 4 different litters) or the pup was cauterized on the snout as a control (n = 5 from 4 different litters). Pups were returned to the nest until testing on P N 2 0 PN25. On the day of testing, pups were anesthetized with urethane (1.5 g/kg, i.p.) and placed in a stereotaxic apparatus. A bipolar, stainless steel stimulating electrode was placed in the LOT. Responses were recorded with glass microelectrodes filled with 2 M NaCI placed in the external plexiform layer or granule cell layer of the olfactory bulb. Responses to paired pulse (equal intensity) stimulation of the L O T (0.1 ms pulse duration, 20 ms inter-pulse interval; 10-30 s inter-pair interval) were amplified and fed to a Macintosh Quadra for analysis with Spike 2 software (Cambridge Electronic Design). Paired-pulses were delivered at each of 9 intensities (200-1000 I~A). Five paired-stimuli were delivered at each intensity and waveforms computer averaged for measurement. Response measures included: amplitude of both conditioning and test responses, and conditioning response width at half amplitude (half-width; Fig. 1). Following baseline recording, animals were injected with MK-801 (0.75 m g / k g , i.p., obtained from R.B.I.) and paired-pulse testing repeated at 15-30 min post-injection. Data were analyzed with ANOVA's and simple regression analyses of the individual data.
3. Results
Stimulation of the L O T produced characteristic responses in both deprived and undeprived olfactory bulbs. Fig. 1 shows representative responses from a deprived olfactory bulb. Evoked response amplitude varied with stimulus intensity, and did not significantly differ between groups (Group x intensity ANOVA, main effect of g r o u p , F1,63 = 3.13, N.S.; Maximal response amplitude: undeprived = 3.0 _+ 0.6 mV; deprived = 2.7 _+ 0.9 mV; unpaired t-test, t 7 = 0.37, N.S.). The N M D A antagonist MK-801 reduced response amplitude in both undeprived and deprived pups (Drug x intensity ANOVA's, main effect of drug: undeprived, F1,36 = 20.30, P < 0.01; deprived F1,27 = 118.03, P < 0.01), but amplitude still varied as a function of stimulus intensity (Fig. 2a). In addition to evoked potential amplitude, halfwidth also varied as a function of stimulus intensity (Fig. 2b). Increasing L O T stimulus intensity increased evoked response halfwidth, in an intensity dependent manner (regression analysis, intensity vs. halfwidth calculated on raw data: control, slope = 0.35, P < 0.05, t-test; deprived, slope = 0.56, P < 0.05). Halfwidths in undeprived and deprived bulbs did not significantly differ
Deprived olfactory bulb E.P.'s
Pre-drug low
_
MK-801 low --
Pre-drug high --
U
5 ms, O.5 mV
-1
MK-801
high
--
Fig. 1. Evoked potentials recorded in the external plexiformlayer in a deprived olfactory bulb (responses in undeprived bulbs were similar with exceptionsnoted in text). Low intensitypaired-pulse stimulation (20 ms IPI) of the LOT produced very little inhibition of the test response, or facilitation. Response amplitude was measured peak to peak as shown in the LOW responses. High intensity stimulation produced greater inhibition. Evoked potential halfwidth was measured as shown by the arrow in the HIGH intensity responses. The NMDA antagonist MKo801 had little effect on halfwidth or inhibition at low stimulus intensities, but greatly reduced both halfwidth and inhibition at high intensities.
D.A. Wilson/Brain Research 677 (1995) 238-242
240
A
EP A M P L I T U D E
4.o3.5"
11) C ~)
3.0-
O ~
¢- ~ 0--
1.5,
0
0.5"
Undeprivedpre Undeprivedpost Deprived pro
(.3
O.O
J-
2;o
4;o
Deprivedpost
6;0 " 8;o lO'OO
LOT stimulus intensity (IJA)
B 18.0]
EP HALFWIDTH
17-°1 ~ 16.o1
~ 1so1 8 8 14o1
~=
Undeprived Pre
~.,= 13.o.
~
~.o.
="
110O ~ 100:~ .E 9.0. .~
t~
(,3 t-5 "~
8.0. 7o 6.0 i
DeprivedPre
dependent on two opposing factors, pre-synaptic facilitation and feedback inhibition. At this short inter-pulse interval, inhibition dominates at most stimulus intensities. However, the magnitude of paired-pulse inhibition varied with stimulus intensity, with higher intensities producing greater inhibition (Group × intensity ANOVA, main effect of intensity: F 8 , 6 3 - - 4 . 4 9 , P < 0.01). Deprived pups showed significantly greater paired-pulse inhibition than undeprived (group × intensity ANOVA, main effect of g r o u p : F 1 , 6 3 - - 4 . 0 0 , P < 0.05), with this effect primarily displayed at the highest stimulus intensities (e.g., 200 /zA, post-hoc Fisher test, undeprived vs. deprived, N.S.; 1000 ,/xA, post-hoc Fisher test, undeprived vs. deprived, P < 0.05; Fig. 3B). MK-801 had three primary effects on paired-pulse inhibition (Fig. 3). First, inhibition was reduced in both undeprived and deprived pups (Drug × stimulus intensity ANOVA, main effect of drug: undeprived, F1,36 =
DepnvedPost UndepdvedPost
A PAIRED-PULSE
150.0
INHIBITION
1400
" 4ob
~
sob
~o
LOT stimulus intensity ( ~A)
Fig. 2. A: evoked potential amplitude (conditioning pulse) was significantly enhanced by increasing LOT stimulus intensity. Responses in undeprived and deprived bulbs did not significantly differ before MK-801 injection. MK-801 significantly reduced amplitude but had no effect on the intensity-amplitude relationship. Standard error bars are excluded for clarity (in all cases: 0.4 mV < S.E.M. < 0.9 mV). B: evoked potential halfwidth was also enhanced by increasing LOT stimulus intensity. Responses in undeprived and deprived bulbs did not significantly differ before MK-801 injection. MK-801 significantly reduced halfwidth and the intensity-halfwidth relationship.
~ ~~ O..¢: ~._~ ®~
~80.0 120.o 110.o . . . . . . 100.o
0 0o
90.C~
~ o~
80o:
~
_
_
_
_
~
.
~
DeprivedPost UndeprivedPost
UndeprivedPre
6oo: J.
SO.(~ 40.0 •
2;0
" 400
" 6;0
'
8;0
Deprived Pre
" 10100
LOT stimulus intensity ~ A )
140.0, ~
from each other (Group × Stimulus intensity ANOVA, main effect of group, F1,63 = 0.49, N.S.). MK-801 significantly reduced evoked potential halfwidth in both undeprived and deprived groups (Drug × stimulus intensity ANOVA, main effect of drug: undeprived, F1,36 = 31.82, P<0.001; deprived, Fl,27 = 40.48, P < 0.01). Furthermore, MK-801 significantly reduced the dependence of halfwidth on stimulus intensity in undeprived animals (regression analysis: slope = 0.06, N.S.). Increasing stimulus intensity did not affect response halfwidth after NMDA receptor blockade (Fig. 2b). In deprived animals, the relationship between stimulus intensity and halfwidth was also markedly reduced by MK-801, although the regression was statistically significant (slope = 0.29, P < 0.05). A single conditioning pulse modified the amplitude of the response to a test pulse delivered 20 ms later (Figs. 1 and 3). The amplitude of the test response is
120.0 •
._= ~.._~ lOO.O.
[] Undeprived Pre
~ s0.0.
[ ] Undeprived Post
;°00 ~®
40.0-
a.
20.0-
•
Deprived Pre
[ ] Deprived Post ,
200 uA
1000 uA
LOT stimulus intensity (vA I
Fig. 3. A: test response amplitude as a percent of conditioning response amplitude in deprived and undeprived bulbs, pre- and post-MK-801 injection. Pre-MK-801, deprived bulbs showed significantly greater inhibition that undeprived bulbs, at high stimulus intensities. MK-801 significantly reduced this inhibition in both groups, and to a greater extent in deprived than undeprived bulbs. Post-MK-801, deprived bulbs showed significantly less inhibition (greater facilitation) than undeprived bulbs. B: replot of responses to the highest and lowest stimulus intensities. Asterisks signify significant difference from same intensity post-drug and from low intensity pre-drug, P < 0.05.
D~4. Wilson~Brain Research 677 (1995) 238-242
28.46, P < 0.01; deprived, F1,27 = 187.13, P < 0.01). Second, the greatest reduction in inhibition was in the deprived animals, such that following MK-801, deprived animals showed less inhibition (greater facilitation) than undeprived (group x intensity ANOVA, main effect of group: F1,63 = 11.60, P < 0.01). Finally, in contrast to pre-drug pups, the magnitude of inhibition was no longer dependent on stimulus intensity following MK-801 (Group × intensity ANOVA, main effect of intensity: F8,63 =: 0.18, N.S.).
4. Discussion The results presented here confirm previous findings that early olfactory deprivation potentiates granule cell mediated inhibition [19,22]. The present results extend the previous work by demonstrating that expression of this enhanced inhibition is NMDA receptor dependent. Under conditions of either weak LOT input, which is insufficient for NMDA receptor activation, or NMDA receptor blockade by MK-801, deprived bulbs show the same or less inhibition than undeprived controls. Thus, based on these data it is suggested that deprivation-induced loss of granule cells may be compensated for during intense stimulation by an NMDA receptor mediated amplification in the excitation of spared granule cells. The NMDA mediated compensation for lost granule cells could occur through several mechanisms. For example, as granule cell targets are lost, mitral cells may make more synaptic contacts with fewer spared granule cells. This consolidation of more mitral cells onto fewer granule cells (or more of a single mitral cell's output onto a single granule cell) could increase the probability of NMDA receptor activation, and in turn, amplify inhibition. This is well demonstrated in cultures of olfactory bulb neurons, where, despite relatively few mitral cells synapsing onto individual granule cells, NMDA receptors are activated [14]. Thus, perhaps by increasing the proportion of inputs a granule cell receives from one or a few mitral cells, threshold for activating NMDA receptors can be reduced. Alternatively, there could be changes in NMDA receptor density or pre-synaptic changes in transmitter release. While the present data cannot discriminate between these and other mechanisms, they do suggest the changes may be s~abtle. For example, using halfwidth as a measure of NMDA receptor activity, no significant difference was found in halfwidth between non-drugged undeprived and deprived bulbs (Fig. 2).
241
local circuit function in the bulb, in vivo [8]. As LOT stimulus intensity increased, evoked potential amplitude increased. This is presumed to be due to a stimulus dependent increase in the number of LOT fibers activated, resulting in a larger number of co-active excitatory inputs to the granule cells. As expected, MK-801, while slightly reducing response amplitude, did not affect this intensity-amplitude relationship. The short-latency amplitude is primarily dependent on non-NMDA glutamate receptors [14]. However, evoked potential halfwidth is presumed to be strongly influenced by an NMDA receptor mediated depolarization. Given the voltage-gated nature of NMDA receptors, the NMDA component, and therefore response halfwidth, should also vary with stimulus intensity. At low stimulus intensities, granule cells should receive insufficient excitatory input to reach threshold for NMDA receptor activation. At higher stimulus intensities, with stronger excitatory drive from a larger number of co-active inputs, threshold for NMDA receptor activation would be reached and expression of the NMDA component of the evoked potential would become apparent (increase in halfwidth). This is precisely what was observed. Furthermore, MK-801 reduced response halfwidth at high stimulus intensities to near that observed with low stimulus intensities, clearly suggesting that response to low intensity stimulation does not involve NMDA receptors, in vivo. Olfactory deprivation induces at least two types of functional plasticity in bulb circuits. First, mitral/tufted cell odor-induced activity is enhanced following reopening of the sealed nare [7,18]. Second, more intense electrical antidromic stimulation of mitral/tufted cell axons reveals enhanced feedback inhibition in deprived bulbs, as described here and elsewhere [19,22]. These apparently contradictory findings (i.e. enhanced odor response rates and yet more feedback inhibition) may be at least partly reconciled by the dependence of enhanced granule cell inhibition on stimulus intensity. Activity induced by sensory input may normally function below threshold for NMDA receptor activation of granule cells, comparable to very low intensity LOT stimulation (Fig. 3). Perhaps only when odor input is associated with a strong centrifugal input, as in associative conditioning [3,21], are NMDA receptors activated. Thus, in the deprived bulb, the deprivation-induced loss of granule cells, combined with deprivation-induced loss of glomerular layer dopamine [2,5,9,22], may under natural circumstances, contribute to the observed heightened responsiveness to odors.
4.1. NMDA receptors and olfactory bulb function
Acknowledgements
These results further demonstrate the stimulus intensity dependence of NMDA receptor involvement in
This work was supported by NIDCD Grant DC00866 to D.A.W. and Regina M. Sullivan.
242
D.A. Wilson/Brain Research 677 (1995) 238-242
References [1] Altman, J. and Das, G.D., Autoradiographic and histological studies of postnatal neurogenesis. I, Jr. Comp. Neurol., 126 (1966) 337-390. [2] Baker, H., Unilateral, neonatal olfactory deprivation alters tyrosine hydroxylase expression but not aromatic amino acid decarboxylase or GABA immunoreactivity, Neuroscience, 36 (1990) 761-771. [3] Brennan, P., Kaba, H. and Keverne, E.B., Olfactory recognition: a simple memory system, Science, 250 (1990) 1223-1226. [4] Brunjes, P.C. and Frazier, L.L., Maturation and plasticity in the olfactory system of vertebrates, Brain Res. Rev., 11 (1986) 1-45. [5] Brunjes, P.C., Smith-Crafts, L.K. and McCarty, R., Unilateral odor deprivation: effects on the development of olfactory bulb catecholamines and behavior, Dev. Brain Res., 22 (1985) 1-6. [6] Frazier, L.L. and Brunjes, P.C., Unilateral odor deprivation: Early postnatal changes in olfactory bulb density and number, J. Comp. Neurol., 269 (1988) 355-370. [7] Guthrie, K.M., Wilson, D.A. and Leon, M., Early unilateral deprivation modifies olfactory bulb function, J. Neurosci., 10 (1990) 3402-3412. [8] Jacobson, I., Butcher, S. and Hamberger, A., An analysis of the effects of excitatory amino acid receptor antagonists on evoked field potentials in the olfactory bulb, Neuroscience, 19 (1986) 267-273. [9] Kosaka, T., Kosaka, K., Hama, K., Wu, J.-Y. and Nagatsu, I., Differential effect of functional olfactory deprivation on the GABAergic and catecholaminergic traits in the rat main olfactory bulb, Brain Res., 413 (1987) 197-203. [10] Meisami, E., Effects of olfactory deprivation on postnatal growth of the rat olfactory bulb utilizing a new method for production of neonatal unilateral anosmia, Brain Res., 107 (1976) 437-444. [11] Meisami, E. and Safari, L., A quantitative study of the effects of early unilateral olfactory deprivation on the number and distri-
bution of mitrat and tufted cells and of glomeruli in the rat olfactory bulb, Brain Res., 221 (1981) 81-107. [12] Mori, K., Membrane and synaptic properties of identified neurons in the olfactory bulb, Prog. Neurobiol., 29 (1987) 275-320. [13] Skeen, L.C., Due, B.R. and Douglas, F.E., Neonatal sensory deprivation reduces granule cell number in mouse olfactory bulb, Neurosci. Lett., 63 (1986) 5-10. [14] Trombley, P.Q. and Shepherd, G.M., Noradrenergic inhibition of synaptic transmission between mitral and granule cells in mammalian olfactory bulb cultures, J. Neurosci., 12 (1992) 3985-3991. [15] Trombley, P.Q. and Shepherd, G.M., Synaptic transmission and modulation in the olfactory bulb, Curr. Opin. Neurobiol., 3 (1993) 540-547. [16] Trombley, P.Q. and Westbrook, G.L., Excitatory synaptic transmission in cultures of rat olfactory bulb, J. Neurophysiol., 64 (1990) 598-606. [17] Wellis, D.P. and Kauer, J.S., GABAa and glutamate receptor involvement in dendrodendritic synaptic interactions from salamander olfactory bulb, J. Physiol., 469 (1993) 315-339. [18] Wilson, D.A., Effects of long-term ( > 12 months) unilateral olfactory deprivation on olfactory bulb single-unit response patterns to odors, Soc. Neurosci. Abstr., 19 (1993) 123. [19] Wilson, D.A., Guthrie, K.M. and Leon, M., Modification of olfactory bulb synaptic inhibition by early unilateral olfactory deprivation, Neurosci. Lett., 116 (1990)250-256. [20] Wilson, D.A., Guthrie, K.M., Smart, R., Gall, C.M. and Sullivan, R.M., NMDA receptor modulation of olfactory bulb inhibitory circuits, Assoc. Chemoreception Sci. Abstr., 15 (1993) 299. [21] Wilson, D.A. and Sullivan, R.M., Neurobiology of associative learning in the neonate: early olfactory learning, Behav. Neural Biol., 61 (1994) 1-18. [22] Wilson, D.A. and Wood, J.G., Functional consequences of unilateral olfactory deprivation: time-course and age sensitivity, Neuroscience, 49 (1992) 183-192.