Lactation-induced deficits in NMDA receptor-mediated cortical and hippocampal activation: changes in NMDA receptor gene expression and brainstem activation

MOLECULAR BRAIN RESEARCH ELSEVIER

Molecular Brain Research 25 (1994) 323-332

Research Report

Lactation-induced deficits in NMDA receptor-mediated cortical and hippocampal activation: changes in NMDA receptor gene expression and brainstem activation Rula Abbud 1, Gloria E. Hoffman, M. Susan Smith * Department of Neurobiology, University of Pittsburgh School of Medicine Pittsburgh, E1440 Biomedical Sciences Tower, 3500 Terrace Street, Pittsburgh PA 1526L USA

Accepted 5 April 1994

Abstract During lactation, there is an inhibition of cortical and hippocampal activation in response to N-methyl-D,L-aspartate (NMA), but not kainate, as assessed by induction of c-Fos expression. To study whether changes in NMDA receptor function may account for this inhibition, NMDA receptor subunit (NMDAR1) mRNA levels were measured by both Northern analysis and in situ hybridization. Analysis of NMDAR1 gene expression by Northern blot analysis did not reveal significant differences between cycling and lactating rats. Using in situ hybridization, NMDAR1 mRNA levels in several cortical and hippocampal areas appeared to be smaller in lactating rats, compared to cycling rats, although these differences reached significance only in the fronto-parietal cortex and piriform cortex. These subtle changes in NMDAR1 receptor subunit gene expression during lactation are not likely to account for the global lack of neuronal activation in response to NMA. However, it is possible that there may be changes in other NMDA receptor subunits that could account for the deficits in NMDA receptor activation. We also examined the activation state of afferent pathways in the brainstem that provide excitatory input to the cortex and hippocampus. During lactation, NMA induced c-Fos expression in similar areas of the brainstem as during the cycle, except in the locus coeruleus and dorsal raphe, where e-Fos expression was significantly less than that observed during the cycle. In contr~tst, no differences in the pattern of c-Fos expression in the brainstem in response to kainate were observed between cycling and lactating rats. The lack of NMA-induced activation of the locus eoeruleus and dorsal raphe may contribute to the lack of cortical activation during lactation. Key words: NMDA receptor gene expression; Cortical activation; Lactation; c-los expression; Brainstem activation; Locus coeruleus; Dorsal raphe

1. Introduction Lactation in the rat is accompanied by changes in hypothalamic and cortical function. While studying the responsiveness of the hypothalamus to stimulation by excitatory amino acids (EAAs), we observed that the behavioral changes o,-.~arring after treatment of cycling rats with the N-methyl-D-aspartate ( N M D A ) receptor

* Corresponding author. Fax: (1) (412) 648-2314. I Present address: Department of Pharmacology, Case Western Reserve University, 2109 Adelbert Rd, Cleveland, OH 44106-4965, USA. Fax: (1) (216) 368-3395 0169-328X/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0169-328X(94)00081-O

agonist, N-methyl-D,L-aspartate (NMA), were completely absent in lactating rats [1]. However, behavioral responses to another E A A acting on a different set of receptors, kainate, were similar in cycling and lactating rats. The receptor specificity of E A A effects in lactating rats was also reflected by changes in expression of the immediate early gene product, c-Fos, within the cortex and hippocampus. N M A treatment of lactating rats failed to induce the massive cortical and hippocampal c-Fos labeling seen in cycling rats, whereas labeling patterns after treatment with kainate were similar in lactating and cycling animals [1]. We further characterized this p h e n o m e n o n as being dependent on both the suckling stimulus and the high levels of progesterone associated with lactation [2]. The mecha-

324

R~ ,4bhud et ai. /~l~&'~ ular Bruin Rc~'ear~h 25 f 10~4) 323--332

nisms by which suckling and progesterone affect cortical and hippocampal function are unknown. It is possible that the deficits in cortical responsiveness to N M A during lactation are due to changes in N M D A r e c e p t o r function in the h i p p o c a m p u s and cortex. In fact, several studies have suggested that p r o g e s t e r o n e and its metabolites can act at the cell m e m b r a n e to m o d u l a t e the activity of excitatory [29,30,31] and inhibitory receptors [13,21,22]. A n o t h e r possibility is that changes in afferent input to the cortex and h i p p o c a m p u s may m e d i a t e this inhibition. T h e pathways activated by the suckling stimulus travel from the nipple up the spinal cord to the brainstem

and then project to cortical and hypothalamic areas. Suckling may activate inputs that would either inhibit an excitatory drive or excite an inhibitory drive to the cortex. In either case, N M D A receptor activation could by r e n d e r e d ineffective and result in the a p p a r e n t lack of neuronal activation in the target areas in response to NMA. In this study, we sought to elucidate possible mechanisms of suckling-induced suppression of N M D A receptor excitability by determining whether N M D A receptor gene expression in cortical and hippocampal target areas was r e d u c e d during lactation and w h e t h e r areas of the brainstem known to modulate cortical

Fig. 1. In situ hybridization for NMDAR1 mRNA using an 35S-labeled riboprobe that recognizes 1.4 kb of the 5' region of the NMDAR1 mRNA. A shows the binding of the antisense probe to the hippocampus. The binding of the probe is blocked if it is pre-incubated with an excess amount of cold sense riboprobe, demonstrating the specificity of the antisense probe (B). The bar in the lower right corner = 100/am.

R. Abbud et al. / Molecular Brain Research 25 (1994) 323-332

excitability were differentially activated to express c-Fos in lactating and cycling rats after N M A or kainate treatment.

2. Materials and methods 2.1. Animals Female Sprague-Dawley rats (Zivic-Miller, Allison, PA) were housed under constant temperature (23 + 2°C) and a 12:12 h lightdark cycle. Food and water were provided ad lib. Daily vaginal smears were performed to monitor the estrous cycle. Cycling rats on diestrus day 1 of their cycle and lactating rats suckling 8 pups on day 10 of lactation were used throughout this study.

2.2. Experimental design 2.2.1. Analysis of NMDAR1 gene expression We examined whether NMDAR1 gene expression in different areas of the rat brain, especially the cortex and hippocampus, changed during lactation. For Northern blots, the dorsal hippocampus and a piece of cortex from diestrus day 1 (n = 8) and lactating (n = 8) animals were dissected immediately after decapitation and stored in Solution D (4 M guanidinium thiocyanate buffer and 0.1 M 2mercaptoethanol) at -80"C, until further processing for Northern analysis. For in situ hybridization, animals were sacrificed by decapitation and the brain dissected out of the skull under RNAse free conditions and rapidly frozen on dry ice. Brains were stored at -80°C until ready to be cut. Brain tissue from diestrous (n = 4) and lactating rats nursing 8 pups (n = 5) was processed for in situ hybridization. 2.2.2. Study of c-Fos activation in the brainstem in response to NMA and kainate In a previous study [1], c-Fos expression in the brain was examined in cycling (n = 5) and lactating (n = 6) rats after treatment with NMA (40 mg/kg; four injections at 10 min intervals, i.v.) or kainate (2.5-3.5 mg/kg; four injections at 10 rain. intervals, i.v.). To examine changes in brainstem activation in response to kainate and NMA, the brainstems from these animals were immunocytochemically stained for c-Fos and DBH (dopamine-/3-hydroxylase). DBH was used as a marker for catecholaminergic neurons, exclusive of dopamine neurons. 2.3. Irnmunocytochemistry Brainstems from NMA- and kainate-treated animals were cut into horizontal sections (25/~m thick) using a freezing sledge microtome. Double-label immunocytochemistry was performed as described previously [18], using a c-Fos antibody (Gift from Dr. Thomas Curran) at a concentration of 1:50,000 and an antibody for DBH (1:20,000; gift from Dr. Stanley Watson, Jr.; University of Michigan). The immunoperoxidase chromogens used to visualize c-Fos and DBH were nickel-diaminobenzidine (Ni-DAB, black nuclear stain), and DAB (brown cytoplasmic stain), respectively.

2.4. Quantitative analysis of c-Fos immunostaining c-Fos-positive cells in the locus coeruleus and dorsal raphe were counted. An average of four anatomically matched sections for the locus coeruleus and two sections for the dorsal raphe were analyzed for each animal. The data were expressed as average c-Fos-positive cells per section.

325

2.5. Probes used for northern blot analysis and in situ hybridization The pN60 clone for the NMDA receptor subunit NMDAR1 was kindly provided by Dr. S. Nakanishi at Kyoto University [26]. A 1431 base pair fragment from the 5' terminal sequence was cut using EcoRV and XhoI restriction enzymes. After blunt-end XbaI linker addition, the 1.4 kb fragment was subcloned into pGEM 3Z at its multiple cloning site. For Northern blots, the fragment was labeled by the random primer method, as described by Attardi et al. [5] with [a-32p]dCTP (New England Nuclear). For in situ hybridization, the pGEM 3Z plasmid was linearized with BamHI and antisense cRNA was synthesized using SP6 polymerase. Sense RNA was prepared using HindlII for linearization and T7 for transcription. Radiolabeled antisense cRNA riboprobes were prepared by running the transcription reaction in presence of 3sS-labeled a-thio-UTP, as previously described [28]. The total concentration of a-thio-UTP was 50 p.M, of which 25% was 35S-labeled. Control experiments were performed to optimize the NMDAR1 signal and minimize the background. These conditions were determined by exposing a series of slides to different probe concentrations (0.1-0.5 ~g/ml.kb), hybridization times (5, 8, 12, and 24 h), and periods of exposure to emulsion (2, 3, 5, and 7 days). Saturating conditions were obtained at a probe concentration of 0.3 ~ g / m l ' k b with 8 h of hybridization and 5 days exposure to emulsion. The specificity of the antisense probe for the NMDAR1 mRNA was determined by preincubating the 35S-labeled antisense cRNA with an excess of cold sense riboprobe overnight before applying to tissue. The binding of the antisense probe to tissue (Fig. 1A) was completely blocked when prehybridized with cold sense riboprobe (Fig. 1B).

2.6. Northern blot analysis Total RNA was extracted by the RNAzol method [8]. Briefly, tissue samples were thawed and homogenized in RNAzol T M (2 ml/100 mg tissue). Phenol-Chloroform extraction of RNA was followed by RNA precipitation with isopropanol. The RNA pellet was then washed with 75% ethanol, dried, and dissolved in 100 tzl DEPC (diethylpyrocarbonate)-treated water. This procedure gives a high yield of pure RNA with an OD260/2a0 ratio ranging between 1.8 and 2.0. The RNA yield was calculated from the OD at 260 nm. Ten p.g of the purified RNA was fractionated by electrophoresis on a 1.2% agarose-formaldehyde gel. The gel was then stained with ethidium bromide to determine the integrity of the RNA and blotted by capillary transfer onto Genescreen membranes (New England Nuclear Research Products, Boston, MA). After complete transfer, the membranes were baked for 2 h at 80-100°C and the RNA was cross-linked to the membranes by irradiation with unfiltered UV light for 2 min. The blot was hybridized to the [32p]deoxy-CTPlabeled rat NMDAR1 cDNA probe. After autoradiography of the blot, bands corresponding to the NMDAR1 mRNA were excised and counted using a scintillation counter (Packard 4530, Packard, Downers, Downers Grove, IL). The blot was later rehybridized to a cyclophilin cDNA probe (700 bases) for normalization.

2. 7. In situ hybridization Twenty/xm brain sections from diestrous day 1 and lactating rats nursing 8 pups were cut using a cryostat and wet-mounted on subbed slides (three consecutive series with 3 sections/slide). For each animal, 6 to 8 slides from a series at the level of the hippocampus were selected for these experiments. Slides from both groups were run in the same assay. The in situ hybridization protocol was developed by Steiner and colleagues and our modifications have been previously described [9,23,28,39]. Briefly, slides were dried immediately after removal from the -80°C freezer, fixed in 4% paraformaldehyde, treated with acetic anhydride in TEA buffer,

'3

R. "lhhud cta/.

Uoh'ctdap Braut l¢c.~ear('h 25 ~lc]~)4) 323 ,432

washed with 2 × S S C (sodium chloride and sodium citrate buffer), dehydrated in a series of alcohols (70~k, 05~. 1005'/), delipidaled in chloroform, washed in another series of alcohols ( 100f~, 95¢~ ), and air dried, Rat N M D A R I 3~S-labe[ed cRNA probe in 100 #1 of hybridization buffer was applied to each slide. Paraffin coverslips and rubber cement were used to keep thc slides from drying during the 8 h of incubation at 60°C. At the end of this incubation, slides were washed in 4 × S S C containing 4.2 mM I)TT, treated with RNAse and washed with 0.1×SSC at 65°( `. After being dehydrated and dried, the slides were exposed to the Kodak NTB-2 emulsion for 5 days at 4°C. The slides were then developed using the Kodak D-ttJ developer and counterstained with Cresyl violet.

2.8. Quantitatit'e analysis Of N M D A R I in situ hybridization Several areas of the brain, such as the dentate gyrus (DG), CAI, and CA3 regions of the hippocampus, two areas in the cortex (frontal and parietal; CTX), and the piriform cortex (PIR), were analyzed. Since NMDAR1 is abundantly expressed throughout the brain, an estimate of total silver grains over a defined area was obtained using the Bioscan Optimas Image Analysis System (Redmond, WA). The Optimas Program identifies silver grains by the brightness of the image; the estimate of total silver grains is given as the area occupied by silver grains within the marked area. Each anatomical region was visualized using a lO×objective and a dark field condenser. The brightness of the image was set so that the area occupied by grains in a region of the slide devoid of tissue would read approximately zero. The area analyzed was of a constant size and was used throughout the analysis. A background reading for each section was taken from an area that lacks specific binding, such as the corpus callosum. Since the area of analysis was constant, we report the data in terms of area occupied by grains (after correction for background), which is directly correlated to the total number of silver grains.

2. 9. Statistical analysis To compare c-Fos expression in the brainstem between lactating and diestrous animals, unpaired t-tests were performed (level of significance, P < 0.05). All data are reported as mean_+ S.E.M.

DIESTRUS

For the Northern analysis, the absolute values obtained were tir~,t normalized with cyclophitin and lhen expressed as a percent ~ dicstrus. A two-sided t-test was performed to compare groups (s~g. nificance level, t' ,- 0.05). All data are reported as mean + S.E.M, For the analysis of NMDAR1 mRNA levels by m situ hybridization, repeated measures A N O V A ~as performed l irsl Io determine whether there were differences in the rostro-caudal or bilateral expression of N M D A R I mRNA within each region analyzed. Since there were no differences within a region, the data lor each region were averaged together to give one value/region/animal. After testing for homogeneity of variance, the non-parametric MannWhitney test was then used to compare expression in each region between lactating and diestrous animals. All data are reported as mean +: S.E.M.

3. Results

3.1. NMDAR1 mRNA in the cortex and hippocampus during diestrus and lactation as assessed by northern blot analysis To examine changes in N M D A R 1 gene expression, our first approach was to perform a Northern blot analysis for total R N A extracted from the cortex or hippocampus of cycling and lactating rats. Fig. 2 shows an autoradiogram of cortical (left) and hippocampat (right) samples from cycling and lactating rats. The blot was hybridized with c D N A probes for N M D A R 1 and cyclophilin. The band corresponding to N M D A R 1 m R N A runs at aproximately 4.4 kb, while the cyclophilin band was 0.7 kb. Expressing the data as percent of diestrus revealed no significant differences in N M D A R 1 gene expression between cycling and lactating rats in the cortex (lactation, 99 + 23% of

LACTATION

NMDAR1

DIESTRUS

t

WgOO

LACTATION

tt

CYC

CORTEX

HIPI~OCAMPUS

Fig. 2. Autoradiogram of a Northern blot hybridized with the :~2P-labeled NMDAR1 eDNA and the cyclophitin probe. The RNA samples loaded on the gel came from the cortex (left) and hippocampus (right) of cycling and lactating rats. The band for the N M D A R I m R N A runs slightly lower than the 28S ribosomal band at approximately 4.4 kb.

R. Abbud et al. / MolecularBrain Research 25 (1994) 323-332 diestrus) or diestrus).

hippocampus

(lactation,

84 + 6%

of

3.2. Analysis by in situ hybridization o f NMDAR1 m R N A levels during diestrus and lactation Since Northern blot analysis is a less sensitive measure of m R N A levels in any specific region of the brain, we p e r f o r m e d in situ hybridization to look for regional changes in the expression of the N M D A receptor during lactation. N M D A R 1 m R N A was very abundant throughout the brain, as silver grains were

327

observed in distinct cells in the hippocampus (Fig. 3 A - C ) , the cortex (Fig. 3 D - F ) , striatum and certain areas of the hypothalamus. The area occupied by silver grains was analyzed for sub-regions of the hippocampus (i.e., dentate gyrus, CA1, CA3) as well as the fronto-parietal cortex, and piriform cortex. Tissue sections were rostro-caudally segregated and bilateral readings were taken for each area analyzed. However, data from rostral, caudal, left, and right regions of the brain were averaged together since no obvious differences at any of these levels were observed. In general, the area occupied by grains in all the regions of the

Fig. 3. Distribution of NMDAR1 mRNA in three areas of the hippocampus: the dentate gyrus (A), CA3 (B) and the CA1 (C) regions, and three regions in the cortex: the piriform cortex (D), the frontal cortex (E) and the parietal cortex (F). These regions were analyzed for area occupied by silver grains using the Optimas Image Analysis System by Bioscan. The bar in the lower right comer = 100/~m.

R. Abbud ~'t a[. 'MMecularBrain Research 25 11994) 323 332

32~

Table 1 Area occupied by grains (mean + S.E.M.) in the cortex (CTX). piriform cortex (PIR), dentate gyrus (DG), CA t and CA~ regions of the hippocampus of diestrous (n = 4) and lactating rats (n - 5) Region Diestrus Lactation P value ('TX PIR DG CA E CA 3

61643 +_3069 92720 + 3292 110856 + 2592 72564 +_1526 96725 _+5561/

40 199 + 6175 64381 + 9533 91479 + 9764 63660 + 5354 75682 + 7317

tl.037 " 11.1137~ 0.097 0.18{1 0.067

brain analyzed was smaller in the lactating rats (Table 1). However, this difference was significant only in the fronto-parietal regions of the cortex (35% decrease) and the piriform cortex (30% decrease). T h e percent decreases in the CA1, CA3, and d e n t a t e gyrus were 12, 22, and 17%, respectively, and were not statistically significant. 3.3. Differences in brainstem actiuation by N M A and kainate during diestrus and lactation

To determine w h e t h e r the deficits in cortical activation could originate at the brainstem level, we examined the pattern of c-Fos activation in the brainstem of diestrous and lactating rats treated with N M A and

kainate. We focused our attention ~m the catccholaminergic and serotoninergic pathways, both of which project broadly to the cortex and hippocampus. Double label immunocytochemistry for c-Fos and D B H allowed us to study the pattern of activation m the catecholaminergic cell bodies. DBH-positivc cell bodies in the A 1 / C 1 (ventrolateral medulla), A 2 / C 2 (nucleus tractus solitarius, Fig. 4), and A5 regions expressed c-Fos in response to both N M A and kainate in diestrous and lactating rats. For the locus coeruleus, p h o t o m i c r o g r a p h s are shown in Fig. 5 and the quantitation of c-Fos expression is shown in Fig. 7. A l t h o u g h N M A induced c-Fos expression in many cell bodies in the locus coeruleus of diestrous rats (Fig. 5A; 47_+ 10 cells, Fig. 7), little c-Fos could be observed in the cell bodies in this area of lactating rats (Fig. 5B; 6 + 3 cells, Fig: 7). However, kainate was equally effective in inducing c-Fos expression in the locus coeruteus of diestrous (Fig. 5C; 47 +_ 4 cells, Fig. 7) and lactating rats (Fig. 5D; 59 _+ 3, Fig. 7). Similarly, the activation of c-Fos expression in the dorsal raphe in response to N M A was attenuated during lactation (Fig. 6B; 26 + 10 cells, Fig. 7) when compared to diestrus (Fig. 6A; 102 + 9 cels, Fig. 7). However, treatment with kainate induced c-Fos expression in this area to a similar degree in lactating rats (Fig.

Fig. 4. Pattern of c-Fos activation in the nucleus tractus solitarius in diestrous (A,C) and lactating rats nursing 8 pups (B,D) in response to NMA (left) and kainate (right). c-Fos is shown as the black nuclear stain. The bar in the lower right corner = 100 #m.

R. Abbud et al. / Molecular Brain Research 25 (1994) 323-332

329

Fig. 5. Pattern of c-Fos expression in the locus coeruleus during diestrus (A,C) and lactation (B,D) in response to NMA (left) and kainate (right). Tissue sections were processed for double label immunocytochemistryfor c-Fos and DBH (dopamine-/3-hydroxylase).c-Fos is shown as the black nuclear stain while DBH is the light cytoplasmicstain. The bar in the lower right corner = 100 p.m. 6D; 154 + 4 cells, Fig. 7) and diestrus rats (Fig. 6C; 127 + 11 cells, Fig. 7).

4. Discussion In this study, we have attempted to elucidate some of the mechanisms involved in the apparent lack of neuronal activation, as evidenced by the lack of c-Fos activation, in the cortex and hippocampus of lactating rats in response to NMA. Our first approach was to determine whether the ineffectiveness of the N M A stimulus was due to changes in the N M D A receptor itself, since no deficits were observed in response to kainate receptor activation [1]. Recently, the cloning of the N M D A receptor has allowed us to approach this question by looking for changes in N M D A receptor gene expression. Several subunits for the rat N M D A receptor have been cloned: N M D A R 1 [26], and NR2A, 2B, 2C, and 2D [14,25]. Moreover, several splice variants of the N M D A R 1 subunit have been reported [32]. NR2 (A, B, C, D) can form a functional receptor only when expressed in a heteromeric configuration with N M D A R 1 [26]. However, homomeric expression of N M D A R 1 in X e n o p u s oocytes can result in a func-

tional channel, but with low current amplitude. In addition, N M D A R 1 is expressed throughout the rat brain, while the other subunits showed selective expression [25,26]. In this study, we have used the clone for N M D A R 1 to generate the broadest probes for in situ hybridization and Northern blot analysis. These probes recognize most of the N M D A R 1 isoforms, as they do not distinguish between splice variants. Changes in N M D A R 1 gene expression could not be detected in the dorsal hippocampus or cortex by Northern blot analysis. This suggests that global changes in N M D A R 1 gene expression in the cortex and hippocampus are not present during lactation. Therefore, we performed in situ hybridization to determine whether changes in N M D A R 1 gene expression during lactation occur in specific regions of the cortex or hippocampus. The data suggest that the changes in N M D A R 1 gene expression during lactation are very subtle. In general, the decrease in the N M D A R 1 signal during lactation appears to be more pronounced in the fronto-parietal and piriform cortices than in the CA1, CA3, and D G regions of the hippocampus. It is not clear whether such a small decrease in steady-state N M D A R 1 m R N A can result in a significant down-regulation of N M D A receptors at the cell

331~

R. Al~ht,I ~'t a]. /~h)h,cldar BraiH Re.s't'ar~'h 25 (lg94)

323-.~:~2

C

B

:'

Q

, ,

,"

."

-

7

"

"

Fig. 6. c-Fos expression in the dorsal raphe during diestrus (A,C) and lactation (B.D) after treatment with NMA (left) or kainatc (right). The bar in the lower right corner = 100/.tm.

membrane. Recent reports suggest that administration of the N M D A receptor competitive antagonist, CPP, and to a lesser extent the non-competitive antagonist, MK-801, which alter N M D A receptor activity, are accompanied by a decrease (45 and 20% decrease, respectively) in N M D A R 1 steady-state m R N A in the frontal cortex [19]. Therefore, the small changes in N M D A receptor gene expression observed in the lactating rat could reflect more pronounced changes in

Locus Coeruleus

Dorsal Rsphe

175 C O

'~ 150 125 B

(~ 100 .m

>

75

o O.

50

o

25

&l. o

0

Fig. 7. Number of c-Fos-positive cells/section in the locus coeruleus and dorsal raphe after treatment with NMA or kainate. Data represent the mean_+S.E.M. *Significantly different than diestrus group receiving the same treatment (P < 0.05).

the activation state of the N M D A receptor at the cell membrane. However, receptor autoradiography using ~25I-MK-801 did not reveal any obvious changes in N M D A receptor binding sites during lactation (unpublished observations). Therefore, the small decreases in N M D A R 1 steady-state m R N A observed in the lactating rat probably do not totally account for the lack of c-Fos activation by exogenous administration of NMA. However, it is possible that there may be changes in other N M D A receptor subunits that could account f o r the deficits in N M D A receptor activation. Given the large number of subunits that have been identified, it is not possible at this time to embark on a large-scale study to examine changes in all receptor subunits. Another possible mechanism for the deficits in cortical and hippocampal responsiveness to N M A during lactation is that the animal's level of arousal is altered by suckling, attenuating excitatory input to the cortex and hippocampus. In this study, we report that the activation of c-Fos expression by NMA, but not kainate, is greatly attenuated in the locus coeruleus and dorsal raphe of lactating rats. However, other catechotaminergic cell bodies in the brainstem were activated to a similar degree in diestrous and lactating rats in response to NMA. Whether the lack of activation of

R. Abbud et al. / Molecular Brain Research 25 (1994) 323-332

these areas in response to NMA results from a change in NMDA receptors is currently being investigated, using in situ hybridization to measure NMDAR1 mRNA levels. The lack of activation could also result from suckling-induced inhibition of afferent pathways excited by NMA that project to these areas. The locus coeruleus and dorsal raphe both serve modulatory roles in the cortex and hippocampus and have been implicated in the regulation of cortical EEG, and therefore, the state of arousal of the animal [7,16,24,33,34,35]. The locus coeruleus has been postulated to filter spontaneous activity while enhancing evoked responses at times of high arousal or alerting [11]. Patterns of cortical EEG closely reflect the locus coeruleal activity. Moreover, the locus coeruleus neurons show little activity during sleep or drowsiness. Their function during the sleep-wake cycle appears to be to alter the scope of neuronal responsiveness of their targets from internally oriented and directed states to an externally oriented mode that promtes responses to novel stimuli. In this regard it is significant that the milk ejection reflex, which occurs in response to suckling and results in the release of oxytocin, is associated with an EEG pattern characteristic of sleep [37]. The locus coeruleus also is the exclusive source of the noradrenergic nerve terminals in the hippocampus. These fibers travel in the dorsal tegmental bundle and reach the dentate gyrus via three pathways: the ipsilateral and contralateral fasciculus cinguli, the fornix, and the ventral amygdaloid bundle-ansa peduncularis. However, the inputs to Ammon's horn travel mainly in the ventral amygdaloid bundle-ansa peduncularis [15,20]. Many electrophysiological and pharmacological studies have demonstrated that exogenous norepinephrine, locus coeruleus stimulation, or glutamate injection into the locus coeruleus [4,10,12,38] can elicit a potentiation of perforant path evoked potentials in the dentate gyrus via the activation of /3-adrenergic receptors [17]. These observations suggest that the activation of the hippocampus is not a monosynaptic event. The response to activation of the major input to the hippocampus, the perforant path, is modulated by the noradrenergic drive from the locus coeruleus. During lactation, it is not clear whether the lack of c-Fos activation by NMA in the locus coeruleus alone is the main cause for the lack of activation in the hippocampus. It is possible that changes in the glutamatergic drive coming from the entorrhinal cortex via the perforant path may also play a role. The dorsal raphe also sends extensive projections to many areas in the brain [6,27,36]. While cortical responses to raphe neuron stimulation have not been studied as extensively as those to stimulation of the locus coeruleus, cortical projections of raphe neurons may function similarly to those of the locus coeruleus.

331

The attenuation of dorsal raphe responsiveness to NMA could also contribute to the deficits in cortical and hippocampal activation during lactation. In conclusion, we have attempted to determine whether the deficits in cortical and hippocampal activation by NMA during lactation are due to changes in the NMDA receptors in the cortex and hippocampus or to deficits in brainstem activation. The changes in NMDAR1 gene expression were very subtle and probably do not account for the functional deficits in cortical activation in response to NMA during lactation. However, the deficits in the activation of the locus coeruleus and dorsal raphe in response to NMA parallel the cortical and hippocampal deficits. It is possible that suckling-induced changes in the responsiveness of the locus coeruleus and dorsal raphe to NMA may mediate the lack of cortical and hippocampal activation during lactation.

Acknowledgements We wish to thank Dr. S. Nakanishi at Kyoto University for providing the NMDAR1 cDNA and Dr. Tom Curran at Roche Institute for providing the antibody for c-Fos protein. The photographic services were provided by Thomas C. Waters. This work was supported by an NIH Grant HD 14643.

References [1] Abbud, R., Lee, W.-S., Hoffman, G.E. and Smith, M.S., Lactation inhibits hippocampal and cortical expression of c-Fos in response to NMDA but not kainate receptor agonists, Mol. Cell. Neurosci., 3 (1992) 244-250. [2] Abbud, R., Hoffman, G.E. and Smith, M.S., Cortical refractoriness to NMA stimulation in the lactating rat: recovery after pup removal and blockade of progesterone receptors, Brain Res., 604 (1993) 16-23. [3] Abbud, R., Attardi, B.A., Hoffman, G.E. and Smith, M.S., NMDA-receptor mRNA expression in the cortex and hippocampus of cycling and lactating rats, Soc. Neurosci. Abstr., Anaheim, 1992. [4] Assaf, S.Y., Mason, S.T. and Miller, J.J., Noradrenergic modulation transmission between the entorhinal cortex and the dentate gyrus of the rat [proceedings], J. Physiol., 292 (1979) 52P. [5] Attardi, B., Keeping, H.S., Winters, S.J., Kotsuji, F., Maurer, R.A. and Troen, P., Rapid and profound suppression of messenger ribonucleic acid encoding follicle-stimulating hormone beta by inhibin from primate Sertoli cells, Mol. Endocrinol., 3 (1989) 280-7. [6] Azmitia, E.C. and Segal, M., An autoradiographic analysis of the differential ascending projections of the dorsal and median raphe nuclei in the rat, J. Comp. Neurol., 179 (1978) 641-67. 17] Berridge, C.W. and Foote, S.L., Effects of locus coeruleus activation on electroencephalographic activity in neocortex and hippocampus, J. Neuroscio, 11 (1991) 3135-45. [8] Chomczynski, P. and Sacchi, N., Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction, Anal. Biochem., 162 (1987)156-9. [9] Chowen-Breed, J., Fraser, H.M., Vician, L., Damassa, D.A., Clifton, D.K. and Steiner, R.A., Testosterone regulation of

3~z2

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[2l]

[22]

[23]

R. .dbbud ~'1 at. ,' Molecular Brain Res'eareh 25 ~1994~ 323-332

proopiomelanocorlin messenger ribonucleic acid in the arcuate nucleus of the male rat, Endocrinology, 124 (1989) 1697 702. Daht, D. and Winson, J., Action of norepinephrinc in the dentate gyrus. I. Stimulation of locus coeruleus. Exp. Brain Re.s., 59 (19851 491-6. Foote, S.L., Berridge, C.W., Adams, L.M. and Pineda, J.A.. Electrophysiological evidence for the involvement of the locus coeruleus in alerting, orienting, and attending, Prog. Brain Res.. 88 (199t) 521-32. Harley, C.W. and Milway, J.S., Glutamate ejection in the locus coeruleus enhances the perforant path-evoked populatkm spike in the dentate gyrus, Exp. Brain Res., 63 (1986) 143-511. Harrison, N.L., Majewska, M.D., Harrington, J.W. and Barker, J.L, Structure-activity relationships for steroid interaction with the y-aminobutyric acid-A receptor complex, J. Pharmacol. Exp. Ther., 241 (19871 346-353. lshii, T., Moriyoshi, K., Sugihara, H., Sakurada, K., Kadotani, H., Yokoi, M., Akazawa, C., Shigemoto, R., Mizuno, N. and Masu, M., Molecular characterization of the family of the Nmethyl-D-aspartate receptor subunits, J. Biol. Chem., 268 (1993) 2836-43. Jones, B.E. and Moore, R.Y., Ascending projections of the locus coeruleus in the rat. It. Autoradiographic study, Brain Res., 127 (1977) 25-53. Kocsis, B. and Vertes, R.P., Dorsal raphe neurons: synchronous discharge with the theta rhythm of the hippocampus in the freely behaving rat, Z Neurophysiol., 68 (1992) 1463-7. Lacaille, J.C. and Harley, C.W., The action of norepinephrine in the dentate gyrus: beta-mediated facilitation of evoked potentials in vitro, Brain Res., 358 (1985) 210-20. Lee, W.S., Smith, M.S. and Hoffman, G.E., Luteinizing hormone releasing hormone neurons express Fos protein during the proestrous surge of luteinizing hormone, Proc. Natl. Aead. Sci. USA, 87 (1990) 5163-5167. Lesch, K.P., Aulakh, C.S., Wolozin, B.L., Hill, J.L. and Murphy, D.L., 3-(2-Carboxypiperazin-4-yl)propyl-l-phospbonic acid decreases NMDA receptor mRNA, Eur. J. Pharmacol., 227 (1992) 109-11. toy, R., Koziell, D.A., Lindsey, J.D. and Moore, R.Y., Noradrenergic innervation of the adult rat hippocampal formation, J. Cornp. Neurol., 189 (1980) 699-710. Maggi, A. and Perez, J., Progesterone and estrogens in rat brain: modulation of GABA (y-aminobutyric acid) receptor activity, Eur. J. Pharmacol., 103 (1984) 165-168. Majewska, M.D., Harrison, N.L, Schwartz, R.D., Barker, J.L and Paul, S.M., Steroid hormone metabolites are barbiturate-like modulators of the GABA receptor, Science, 232 (1986) 10041007. Marks, D.L, Smith, M.S., Clifton, D.K. and Steiner, R.A., Regulation of gonadotropin-releasing hormone (GnRH) and galanin gene expression in GnRH neurons during lactation in the rat, Endocrinology, 133 (1993) 1450-1458.

[24] McGinty, D.J. and Harper, R.M., Dorsal raphe neunms: depression of firing during sleep in cats, Bruilz t?e~., 101 ~,lt~7~) 569-75. [25] Monyer, H., Sprengel, R., Schoepfer, R., tlerb; ,~.., Higuchi, M.. Lomeli, H., Burnashev, N,, Sakmann, B. and Seeburg, P.t! .... Heteromeric NMDA receptors: molecular and functional distinction of subtypes, Science, 256 (19921 1217-21. [26] Moriyoshi, K., Masu, M., Ishii, T., Shigemoto, R., Mizuno, N. and Nakanishi, S., Molecular cloning and characterization of the rat NMDA receptor, Nature, 354 1199t) 31 7 [27] Newman, D.B. and Liu, R.P., Nuclear origins of brainstem reticulocortical systems in the rat, Am..1. Anat., 178 11987) 279-99. [28] Smith, M.S., Lactation alters Neuropeptide-Y and proopiomelanocortin gene expression in the arcuate nucleus of the rat, Endocrinology, 133 (1993) 1258-1265. [29] Smith, S.S., Waterhouse, B.D., Chapin, J.K. and Woodward, D.J., Progesterone alters GABA and glutamate responsiveness: a possible mechanism for its anxiolytic action, Brain Res., 400 11987) 353-359. [30] Smith, S.S.. Progesterone enhances inhibitory responses of cerebellar Purkinje cells mediated by the GABA-A receptor subtype, Bra#l Res. Bull., 23 (19891 317 322. [31] Smith, S.S., Progesterone administration attenuates excitatory amino acids responses of cerebellar Purkinje cells, Neuroscience, 42 (1991) 309-320. [32] Sugihara, tt., Moriyoshi, K., Ishii, T., Masu, M. and Nakanishi, S., Structures and properties of seven isoforms of the NMDA receptor generated by alternative splicing, Biochem. Biophys. Res. Commun., 185 (1992) 826 32. [33] Steriade, M., Basic mechanisms of sleep generation, Neurology, 4 (19921 9-17. [34] Trulson, M.E. and Jacobs, B.L., Raphe unit activity in freely moving cats: correlation with level of behavioral arousal, Brain Res., 163 (1979) 135-50. [35] Vanderwolf, C.H., Cerebral activity and behavior: control by central cholinergic and serotonergic systems, lnt. Ret,. Neurobiol., 30 (1988) 225-340. [36] Vertes, R.P., A PHA-L analysis of ascending projections of the dorsal raphe nucleus in the rat, ,L Cornp. NeuroL, 313 (1991) 643-68. [37] Wakerley, J.B., Clarke, G. and Summerlee, A.J.S., Milk ejection and its control. In E. Knobil and J.D. Neill (Eds.), The Physiology of Reproduction, Raven, New York, 1988, pp. 2283-2321. [38] Washburn, M. and Moises, H.C., Electrophysiological correlates of presynaptic alpha 2-receptor-mediated inhibition of norepinephrine release at locus coeruleus synapses in dentate gyrus, J. Neurosci., 9 (1989) 2131-40. [39] Wiemann, J.N., Clifton, D.K. and Steiner, R.A., Gonadotropinreleasing hormone messenger ribonucleic acid levels are unaltered with changes in the gonadal hormone milieu of the adult male rat, Endocrinology, 127 (1990) 523-32.