Characterization of dynorphin-containing neurons on dissociated dentate gyrus cell cultures

Brain Research, 594 (1992) 91-98 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00

91

BRES 18195

Characterization of dynorphin-containing neurons on dissociated dentate gyrus cell cultures Xiao-Ping H e

a, Paul H.K. Lee b, Keith R. Pennypacker a, Raimo K. T u o m i n e n c, E n g - C h u n M a r a, Linda Thai a and Jau-Shyong H o n g "

a Neuropharmacology Section, Laboratory of Molecular and integrath'e Neuroscience, National hlstitute of Ent'ironmental Heahh Sciences, National Institutes of Health, Research Triangle Park, NC 27709 (USA), b Dit'ision of Cell Biology, Burroughs Wellcome Company, Research Triangle Park, NC 27709 (USA) and " Department of Pharmacology, Unii'ersityof Helsinki, Helsinki (Finland) (Accepted 26 May 1992)

Key words: Hippocampus; Neuronal cell culture; Immunocytochemistry; Dentate granule cell; Dynorphin; In situ hybridization

in the dentate gyms, the synthesis of the opioid peptide, dynorphin, is modulated by a variety of stimuli, in order to elucidate the cellular and molecular mechanisms regulating the synthesis of dynorphin in the hippocampus, we have established a routine primary cell culture of dentate granule neurons and identified granule-like neurons by a characteristic marker, dynorphin, in these cultures. Cultures were prepared from 7-day-old rat pups and maintained in medium with 2% fetal bovine serum. These cultures contained approximately 20% neurons and survived for over 4 weeks. After 2 weeks in culture, neurons expressing dynorphin-A and its messenger RNA were detected using immunocytochemistry and in situ hybridization, respectively. In dentate cultures, enkephalin-, cholecystokinin-, neuropeptide Y- and substance P-positive cells were observed in addition to dynorphin.positive cells with immunocytochemistry. The results suggest that dentate gyrus cell cultures provide a valid in vitro model for studying molecular mechanisms regulating prodynorphin gene expression.

INTRODUCTION The perforant path, which originates from the entorhinal cortex, terminates in the molecular layer of the dentate gyrus (DG) and synapses upon dentate granule neurons. The axons of dentate granule neurons innervate the CA3 pyramidal cells through the mossy fiber pathway. Cytochemical studies have shown that dynorphin (DYN) is localized exclusively in DG granule neurons and not in other types of hippocampal

neurons4,2~ .22.27. Several studies have demonstrated that DYN synthesis in the dentate gyrus is altered by seizure activities. Repeated electroconvulsive shocks resulted in a decrease of DYN immunoreactivity (DYN-IR) in the mossy fibers ~8and a time-related decrease in the abundance of pro-DYN mRNA in the hippocampus 29. Systemic injection of kainic acid significantly elevated pro-DYN mRNA levels6. Kindling produced by electrical stimulation of the deep prepyriform cortex, the hippocampus or the amygdala, decreased the levels of

DYN and pro-DYN mRNA in the rat hippocampus~7.2.~.30. The synthesis of DYN in the granule cells is also regulated by other stimuli. Activation of the perforant path, where glutamate is released from the terminals of this pathway, decreased DYN levels"~1. Radioimmunoassay (RIA) and immunocytochemistry (ICC) have shown a marked decrease of DYN-IR in DG granule cells after adrenalectomy and this decreased DYN-IR was restored by dexamethasone treatment (Thai et al., paper in preparation), suggesting that adrenal steroids are essential for the normal synthesis of DYN in the hippocampus. Granule cell cultures are a useful model for analyzing the cellular and molecular mechanisms regulating DYN synthesis in the hippocampus. Although hippocampal pyramidal cell cultures have been widely used for biochemical and electrophysiological studies '~, dentate granule cells have not proven as easy to culture. Boss et al.2 have reported DG cultures using neonatal rats and identified granule cell-like neurons

Correspondence: J.-S. Hong, NIEHS, P.O. Box 12233, Research Triangle Park, NC 27709, USA. Fax: (I) (919) 541-0841.

92 by their specific cell surface marker; however, DYNpositive immunostaining was not detected in these cultures. Since DYN is expressed specifically in the dentate granule cells of the hippocamDal formation 4"s, we have cultured and identified dentate granule neurons using DYN as a marker. We report, for the first time, the culture of dentate granule neurons that express DYN peptide and mRNA. MATERIALS AND METHODS

(YI! odttlr¢' ML~I cultures were prepared from 7-day-old rat pups (Fischer-344, Charles River, Raleigh, NC), but some cultures were from en:bryonic day lg (El9), postnatal day ! (PI), P3, PS, P6, P8 and PI0 rats. The pups were decapitated, their brains rapidly removed and placed in 35 × l0 mm culture dishes containing sterile saline I (in mM: NaCI 14(}, KCI 5,3, Na,HPO 4 I.I, KH2PO 4 1.1, dextrose 22.2). The hippocampus was removed with a pair of fine forceps and the area dentata (including the dentate gyrus and hilar region) was carefully dissected from the rest of the hippocampus under a dissecting microscope according to the procedures described by Boss et al. 2. The tissues were then placed into ice-cold saline I. The dentate tissue was cut into approximately I mm square pieces and then incubated in versene buffer (Gibco, Grand lshmd, NY) containing (I.1~, ',:ypsin (Gibco) for 20 rain at room temperature followed by dissociation of cells by ~.~ntlo trituration through 2 ml sterile pipettes. The cells were suspended in 20 ml Dulhccco's modified Eagle's medium (DMEM; Sigma Chemical Company, St. [,outs, Me) containing 4r~ bovine serum albumin and centrifuged for 10 min at 2,{NXIx g, the cell pellet was t,hen rcsuspended in DMEM containinl~ I(Iq;, fetal bovine serum and counted on a hemocytometer using Trypan blue; approximately 4(X),000 cells were obtained from each animal. ('ells were diluted to a final concentration of ,t x I0 '~cells/ml and (I,~ ml per chamber was plated onto 8x2() mm poly.l-lysinecoated ~lass 4.¢hl~mher slides. Cultures were incubated at 37°C in 5¢; ('O~ at I(X}% humidity. The dissociated cells were initially plated in DMEM (pll %2) containing I()~:; fetal bovine serum, 4 mM L.glutimzine, Ill0 U/ml penicillin, IIX)#g/ml streptomycin and 50 ttg/ml i;entomycin, After 24 h, the medium was removed and replaced by fresh DMEM phJs 10% fetal bovine serum, The cultures were placed in DMEM plus 2¢~ fetal bovine serum on the 4th day after plating and, thereafter, the media were changed weekly. In some studies, 10/~M cytosine /J-t)-arah.no.furem)sido (Ara-c), a mitotic inhibitor which inhibits the growth of non-neuronal cells ~, was :~dded on the second day of culture.

Primao' untisera Six rabbit antisera were used: (a' the antiserum against DYN A(1-8) had less than {l.02% cross-reacl ivity with either DYN A( I - 13) or DYN All-17), and less than 0,00~D5% cross-reactivity with either [Leu~l-enk~:phalin (Leu-Enk) or [Met~]-enkephalin (Met-ENKP'". (b) The anti-cholecystl,kinin-8 (CCK-8) serum showed the following cmss.reactivities: CCK-8 sulfilted, I(Xl~:,; CCK-8 non-sulfated, 8,1%; ga~trin-l. I0,1¢~; CCK-7, f~.4%; and CCK-4, 0,5% ~. (c)The antiserum against substance P (SP) had the following cross-rcactivities toward other tachykinins: physalaemin, 30,6%; nenrokinin A, 12,5%; kassinin, 7.5~; eledoisin, 6.5%, and neurokinin B, 1,6%I'~, (d)The antiserum against Met-ENK was a gift from Dr, W, Hcndrin (Research Triangle Institute, Research Triangle Park, NC) and had less than IC-~ cross-reactivity with either Leu-ENK or MeI-ENK-Arg ('Phe 7, and less than 0,06~;~ cross-reactivity with DYN All-8) "~z, (el The anti-neuropeptide Y (NPY) serum was donated by Dr, H,-Y.T, Yang (St, Elizabeth's Hospital, Washington, DC), The antiserum revealed a 1{1¢7rcross-reactivity with pei)tide YY and less than 0,01 cross-reactivity with FMRF-NH :, ),- or "yl-melanocyte-stimulating hormones, avian pancreatic polypeptide, human pancreatic polypep-

tide, and SP 12. (f) The antiserum against glial fibrillary acidic protein (GFAP) was purchased from Dakopatts (Glostrup, Denmark). The mouse monoclonal antibody against GFAP was purchased from Boehringer Mannheim (Indianapolis, IN), while that against micro~lbule-associated protein-I (MAPI)was obtained from Sigma.

hnmunoeytochemistry Single labeling. The avidin-biotin-peroxidase complex (ABC) technique I~' was employed to visualize the cellular distribution of DYN-, ENK-, CCK-, NPY-, and SP-IR. Briefly, the cells were fixed in 4% paraformaldehyde (pH 7.2) for 10 min. Triton X-100 (0.2%)was used for 20 min followed by 2% normal goat serum for another 20 rain to block non-specific staining. Polyclonal antisera against DYN, ENK, CCK, NPY, and SP were diluted I:1,000 with I% normal goat serum. Cells were treated with primary antibodies for 18 h at 4°C, and incubated with biotin-coupled goat anti-rabbit IgG (Vector Laboratories, Burlingame, CA) which was diluted ! :400 for I h at room temperature, and then incubated with 1:400 avidin-coupled horseradish peroxidase (Vector) for I h. Then, 3,3'-diaminobenzidine (DAB) was used to produce a brown reaction product. Negative controls in these experiments utilized either normal rabbit serum (for polyclonal antibodies) or !% normal goat serum, the vehicle for the antibodies. The specificity of DYN All-8) immunostaining was tested by preabsorption with l0 pM DYN All-8). Doubk, la&,ling. Double-labeling fluorescence ICC was performed to determine whether the DYN-positive cells were neuronal. The indirect immunofluorescence technique was used; all reagents were diluted with I% normal goat serum, Cultures were incubated for 18 h in one of the following 3 mixtures of primary antisera at the dilutions indicated: (l) rabbit anti-DYN (!: 150) and mouse antiMAPI (I:I(X}) antisera; (2) rabbit anti-DYN (!:150) and mouse anti-GFAP (1:150) antisera; and (3) rabbit anti-GFAP (1:2(XI) and mouse anti-MAPI (l:l(X)) antisera. The cultures were then incubated for I h in a mixture of goat anti-rabbit IgG conjugated to rhodamine B (l: I(X)) and goat anti-mouse IgG conjugated to fluorescoin isothiocyanate (l:100)(Molecular Probes, Inc., Eugene, OR). After a final rinse with PBS, the ,~lid~sw~r¢ air dried, mounted using fluoromount.c (Fisher Biotech, Pittsburgh, PAl, and examined under a II|lorescence microscope,

hi sire hybridization histcMl(,mistry (ISltll) In situ hybridization histochemistryt°'lt was performed to detect pm-DYN gone expression in DG cell cultures, A cocktaii of two 33 mer synthetic oligodeoxynucleotide probes, which corresponded to bases 618-652 and 691-723 of the rat prodynorphin eDNA sequence 5, were used for hybridization. The probes were labeled with ['~sS]dATP using terminal deoxynucleotidyltransferase (Bethesda Research Laboratory, Grand Island, NY), with a specific activity of 2 x lip cpm/p.I, Cells were fixed with 4% buffered paraformaldehydo (pH "/,2) for Ill rain at roe-) temperature, pretreated with acetic anhydride, delipidated in a graded series of ethanol, and dried, Cells wore then hybridized with the "~"~S-labeled probes (5 x l0 s cpm/chambur) at 37°C for 20 h. Posthybridization included a series of washes: 4 times with I xSSC (0.15 M NaCI/0,015 M sodium citrate) with gentle shaking at 55°C, once for 15 rain, and twice in I xSSC at room temperature for 30 rain per wash. The slides were then dried overnight and dipped in Kodak NTB-2 emulsion (2:1 with water) and eXlX~sedfor 3 weeks before developing in D.19 solution.

Radiobnmun~ssay Rats were sacrificed on P3, PS, PI0, PIS, P21 and PI40, The whole brain was rapidly removed and the hippocampus was immediately dissected. Radioimmunoassay of DYN A(I-8)-IR was performed as previously described Is. In brief, tissues were homogenized in 2 M acetic acid, boiled for 5 min, centrifuged at 25,000x g for 20 rain, and then the supernatant was lyophilized. Aliquots of the reconstituted samples were used for RIA. lodinated DYN All-8) was used as the radioactive label.

93

RESULTS

5oo

D

E

O

100 0 3

5

10

15

21

140

Age(Days)

Fig. I. Hippocampal dynorphin A(l-8)-like immunoreactivity during postnatal development (P3, P5, PI0, PIS, P21 and PI40) was measured by radioimmunoassay. Each value is the mean ± S.E.M. of 7 rats. * ** P < 0.001, compared to P3 group•

Cell survival and growth To obtain a developmental profile of DYN in the rat hippocampus, the content of DYN-IR was determined by radioimmunoassay in rats at different ages (P3, P5, P10, P15, P21 and P140). This kind of information would suggest the optimal age of rats to start the dentate cultures and the intervals of culturing before the experiments begin. The levels of DYN-IR in the hippocampus were very low in P3 and P5 pups, and rose significantly by P10. Between P5 and P21, DYN-IR levels in the hippocampus increased by 580% (Fig. 1). In order to find the optimal conditions for culturing DG cells, perinatal rat pups were sacrificed at different ages [embryonic day 19 (E19), P1, P3, PS, P6, P7, P8 and P10]. About half of the DG cells cultured from E19, P1, P3, P5 and P10 rats survived during the first few days after plating and exhibited neurite outgrowth. However, only one fourth of the cells survived for more

A O

t

i?!I~/~~1( '

.

B

it~k

~

~ :,

,I.

•

~

,i

..~]"

Fig. 2. Double.labeled immunofluorescence experiments with an anti-glial fibrillau acidic protein (GFAP) polyclonal antiserum (A) and an anti-microtubule-associated protein-1 (MAP1) monoclonal antibody (B) illustrate no overlap between the two proteins in 2-week-old dentate gyrus (DG) cultures. An arrow (in A) represents a GFAP-positive cell; an arrow (in B) indicates a MAPl-positive cell. Bar ffi 100 ~m.

.,t:

Fig. 3. Dynorphin-positive cells in 2.week-old DG cultures. A: DYN immunostained cells are revealed by immunoperoxidase labeling with a rabbit antiserum against DYN A(1-8). B: pro-DYN mRNA-positive cells are illustrated with in situ hybridization using ['~-~S]dATPlabeled antisense oligonucleotide probes specific for pro-DYN cDNA (33 mer). Bar = 100/~m.

g4 than 2 weeks; less than 5% were recognized as DYNcontaining cells. The same proportion of DG cells from P6, P7 or P8 pups survived. These cultures could be maintained for at least one month and sometimes for as long as 6-8 weeks and about 20% of the surviving cells contained DYN. Newly dissociated DG cells rapidly adhered to the surface of the poly-L-lysine coated slides within the first hour after plating. Neuronal processes were extended by the third day and by the end of the first week synapse-like connections formed among cells. During the next 2 weeks, the size of the cells increased slightly while the number, length and complexity! of their neurite processes increased markedly. In cult~Jres over one month of age, the cells exhibited numerous long processes with many branches. Two major morphological cell types could be t!~ tinguished after the 4th day in culture: (1) most o~ the cells were large and flat containing large pale ~,~,~'lei and abundant cytoplasm, and usually covered tht; surface of the slides; (2) about 20-25% of the cells were

A

smaller and round with extended processes which were often long or branched, or both, and usually grew on top of the monolaver of flat cells (Fig. 2). Immunocytochemical experiracn~s revealed that the former were astrocytes, which were GFAP-positive (Fig. 2A), while most of the latter were neurons, which stained for MAP1 (Fig. 2B).

Identification of dynorphin-positive dentate gyrus cells Immunocytochemistry and ISHH revealed DYNpositive and pro-DYN mRNA-positive cells, respectively. Dynorphin-positive staining was observed in DG cultures 2 weeks following plating (Fig. 3A) and the intensity of staining increased with time. Cells exhibiting DYN-IR were small and round, while most of the DYN-negazive cells were large and flat. When the DYN A(1-8) antiserum was preabsorbed with 10/zM DYN A(1-8), no significant staining was observed. LJnder high power, it was clear that DYN-positive cells had a small unstained nucleus, indicating that DYN was localized in the cytoplasm (Fig. 4D). Both the

-%-

2"

B

i o

C

D ''

4,

Fig. 4. Double-labeling fluorescence ICC experiments using an anti-GFAP monoclonal antibody (A, C) and anti-DYN A(I-8) polyclonal antiserum (B, D) in 2-week-old DG cell cultures, There is no overlap between GFAP-positive (arrow in C) and DYN-positive (arrow in D) cells.

Bar = 50 ~m.

95 soma and neuritic processes contained fine and granular-like DAB reaction product (Fig. 3A). The [35S]dATP labelled eDNA probes revealed pro-DYN-positive cells (Fig. 3B). When the sense oligonucleotide of DYN eDNA was used in control experiments, no pro-DYNpositive cells were seen in DG cultures. Dynorphinpositive cells commonly had one or two long processes which extended from the cell body and gave off shorter processes with branches; frequently, the entire length of the neurites could be traced. The terminals of some processes of DYN-positive cells contacted each other (Fig. 5B). As the cultures became older, the processes became longer and formed intertwining fascicles. When 10 ~M Ara-c, a mitotic inhibitor, was added to the culture on the second day after plating, a decrease in the number of fiat cells with a concomitant increase in the percentage of round cells resulted. However, by the second week after plating, there were fewer living cells than before and these cells had fewer processes. Immunocytochemistry showed that Ara-c not

only inhibited cell surtival but also decreased the percentage of DYN-positive cells. In untreated DG cell cultures, about 20% of the cells were DYN-positive, but less than 5% of tt,~e cells were DYN-positive following Ara-C treatments. In order to determine whether DYN-positive cells were neuronal or glial in nature, double-labeling immunofluorescence experiments were performed, using GFAP as a marker fo~ astrocytes and MAP1 as a marker for neurons. The GFAP-positive cells had large and flat somata with ]~rge nuclei (Fig. 2A). The MAPl-positive cells were smaller and round with very long neuritic processes (Fig. 2B). The ratio of MAP1positive to GFAP-positive cells was 1:4-5, indicating that only 20-25% of the cells in DG cultures were neurons. Double staining with antibodies against GFAP and DYN showed no overlap between GFAP-positive and DYN-positive cells (Fig. 4). The second set of experiments, using antibodies against both MAPI and DYN A(1-8), showed that DYN coexists with MAP1;

A

B Y

-T-

C

T

D 1

B

J

Fig. 5. Double-labelingfluorescence ICC experiments using MAP1 monoclonal antibody (A, C) and DYN A(1-8) polyclonal antiserum (B, D) show the coincidence of the two labels (a typical example is shown by arrows in C and D). A few MAPl-positivecells (arrow in A) are not DYN-positiveceils(arrow in B). Bar -- 50/~m.

96

A

B

P

a;

o

"

.a,

-

"t" i

~a

<,

e {

D

C

"~'

,~

.

.

,

Fi8, 6, hnmunoperoxidase labeling with polyclonlll ,nIi-ENK serum (A), anti-CCK serum (B), unti-NPY serum (C) and ,nli-SP sera (D) in 2.week.old DG cultures, Bar - i00/~m,

however, approximately 80-90% of the MAPl-positive neurons also were immunopositive for DYN (Fig. 5A,B).

tmmunosta#zing of other neuropeptides in dentate gyrus cultures The localization of other neuropeptides, including ENK, CCK, NPY and SP, were also examined using ICC. In immunoperoxidase experiments, besides DYN-positive cells, ENK- (Fig. 6A), CCK- (Fig. 6B), NPY- (Fig. 6C) and SP-positive (Fig. 6D) cells were observed in DG cell cultures. These neuropeptide.containing cells were smaller, round, elliptic in shape and also grew on the top of the monolayer of the flat cells. NPY-, ENK-, and CCK-positive cells had longer and varicose processes with more neurites, while the SPpositive cells had predominantly fine fibers. The morphology of some ENK-positive cells was similar to DYN-positive neurons.

DISCUSSION The principal objective of this study was to develop an in vitro system for culturing DO granule cells which could serve as a valid model for studying the regulation of pro-DYN gene expression. In this study, a routine primary cell culture of dentate granule neurons was established and DYN-positive cells were identified based on both ICC and ISHH. Since cellular morphology in dissociated cell cultures may not always resemble that which occurs in rive cytochemical markers can be useful for the identification of specific cell types, immunocytochemistry and RIA revealed that D Y N - A and DYN-B are restricted to dentate granule cells and their axons, known as the mossy fibers 4,s, No perikarya containing either DYN-A- or DYN-B-IR were observed in the dentate hilus or in any of the CA fields 4'27. Since the major site of DYN synthesis in the hippocampus has been clearly

97

demonstrated by ISHH to occur in the granule cells 23, DYN can be a useful marker for these cells. Immunocytochemical studies by others revealed DYN-IR increased between P8 and P19 in vivo ~. Our results confirm that DYN levels in the hippocampus began to increase in P10 pups in vivo and that DYN-positive neurons were visualized in 2-weebold cultures. Thus, DYN can serve as a reliable marker for DG granule cells only at certain ages. In most neuronal cultures, such as hippocampal pyramidal cell cultures, special precautions are used to prevent the overgrowth of the non-neuronal population, including the use of special media, application of substrated coverslips and treatment with glial inhibitors 2s. However, when we tried to decrease the number of glial cells using Ara-c, the survival of the DG cultures and the number of DYN-positive cell were reduced. The developmental stage-specific changes in hippocampal astrocytes suggest that the development of astrocytes closely parallels that of the dentate granule neurons, i.e., both are generated mainly postnatally '~. It is possible that astrocytes in the DG might provide stimulating factors that are necessary, for example, for attachment, survival, neurit~ elongation, and functional expression of the granuif,; neurons. Granule cells have been shown ~,0 contain both DYN and ENK, while the non-granule cells in the hilar region contain a variety ,~f other r,europeptides in vivo including CCK, NPY, SP, ¢asoactive intestinal polypeptid¢ and sotnatostatin 7't3'14'1~7'24'2¢'.Similarly, in our DG cell cultures, ENK-, CCK-, NPY- and SP-IR cells were detected in addition to DYN-IR cells. The results suggest that DG cultures could also be used to study the regulation of these neuropeptide genes. It is worth noting that although both dynorphin and enkephalin are stored in the dentate granule cells, dynorphin is a particularly good neuropeptide marker for these cells, since enkephalin is expressed in cultured astrocytes 2s. In summary, this study demonstrates the successful culturing of DG cells by demonstrating the presence of DYN by both ICC and ISHH. The DG cell cultures should be a valid in vitro model for analyzing the molecular mechanisms underlying pro-DYN gone expression. Acknowledgements. We would like to acknowledge the excellent technical assistance of Drs. L. Tian and W. Zhang. We also thank Mrs. Loretta Moore for excellent editorial assistance. X.-P.H. is a recipient of a Fogarty International Fellowship (1 FO5 TWO4205-01 BI-5) from the Fogarty International Center of the National Institutes of Health.

REFERENCES ! Balazs, R., Jorgensen, O.S. and Hack, N., N-MethyI-D-aspartate promotes the survival of cerebeilar granule cells in culture: Neuroscience, 27 (1988) 437-451. 2 Boss, B.D., Gozes, I. and Cowan, W.M., The survival of dentate gyrus neurons in dissociated culture, Dec. Brain Res., 36 (1987) 199-218. 3 Bray, S., Bustos, G.A., Lee, P.H.K., Hong, J.S., Aitken, P.G. and Nadler, J.V., Kainic acid inhibits cholecystokinin release from rat hippocampal slices, Neurosci. Lett., 100 (1989) 313-318. 4 Chavkin, C., Shoemaker, W.J., McGinty, I.F., Bayon, A. and Bloom, F.E., Characterization of the prodynorphin and proenkephalin neuropeptide systems in rat hippocampus, J. Neurosci., 5 {1985) 808-816. 5 Civelli, O., Douglass, J., Goidstein, A. and Herbert, E., Sequence and expression of the rat prodynorphin gene, Proc. Natl. Acad. Sci. USA, 82 (1985) 4291-4295. 6 Douglass, J., Grimes, L., Shook, J., Lee, P.H.K, and Hong, J.S., Systemic administration of kainic acid differentially regulates the levels of prodynorphin and proenkephalin mRNA and peptides in the rat hippocampus, Mol. Brain Res., 9 (1991) 79-86. 7 Gall, C., Berry, L.M. and HodgsoJ~, LA., Cholecystokinin in the mouse hlppocampus: l~calization in the mossy fiber and dentate ¢o:nmissura! systems, Exp. Brain Res., 62 (1986) 431-437. 8 Gall, C., Ontogeny of dynorphin-like immunoreactivity in the hippocampal formation of the rat, Brain Res., 307 (1984) 327-331. 9 Gasser, IJ.E. and Hatten, M.E., Neuron-glia interactions of rat hippocanlpal cells in citro: glial-guided neuronal migration and neuronal regulation of glial differentiation, J. Neurosci., 10 (1990) 1276-1285. 10 Gerfen, C.R., McGinty, J.F. and Young Ill, W.S., Dopamine differentially regulates dynorphin, substance P, and enkephalin expression in striatal neurons: in situ hybridization histochemical analysis, J. NeuroscL, 11 (1991) 1016-1031. I I Harrison, P.J. and Pearson, R.C.A., In situ hybridization histochemistry and the study of gene expression in the human brain, Progr. Neurobiol., 33 (1990) 271-312. 12 I-liguchi, H. and Yang, H.-Y.T., Splanchnic nerve transeclion abolishes the age.dependent increase of neuropeptide Y.like immunoreactivity in rat adrenal gland, J. Neurochem., 46 (1986)

1658-1660, 13 ltoffman, D.W. and Zamir, N., Localization anti quantitation ol dynorphin B in the rat hippocampus, Brain Res., 324 (1984) 354-357. 14 Hoffman, D.W. and Zamir, N., Localization and quantitation of proenkephalin-derived neuropeptides in the rat hippocampus, Neuropeptides, 5 (1985) 437-440. 15 Hong, J.S., Tilson, H.A. and Yoshikawa, K., Effects of lithium and haloperidol administration on the rat brain levels of substance P, J. PharmacoL Exp. Ther., 224 (1983) 590-593. 16 Hsu, S.M., Raine, L. and Fanger, H., The use of antiavidin antibody and avidin-biotin-peroxidase complex in immunoperoxidase techniques, Am. J. Clin. Pathol., 75 (1981) 816-821. 17 ladarola, M.J., Shin, C., MeNamara, J.O. and Yang, H.-Y.T., Changes in dynorphin, enkephalin and cholecystokinin content of hippocampus and substantia nigra after amygdala kindling, Brain Res., 365 (1986) 185-191. 18 Kanamatsu, T., McGinty, J.F., Mitchell, C.L. and tiong, J.S., Dynorphin- and enkephalin-like immunoreactivity is altered in limbic-basal ganglia regions of rat brain after repeated electroconvulsive shock, J. Neurosci., 6 (1986) 644-649. 19 Kohler, C., A morphological analysis of vasoactive intestinal polypeptide (VIP)-Iike immunoreactive neurons in the area dentara of the rat brain, J. Comp. Neurol., 221 (1983) 247-262. 20 Li, S., Sivam, S.P. and Hong, J.S., Regulation of the concentration of dynorphin A(I-8) in the striatonigral pathway by the dopaminergic system. Brain Res., 398 (1986) 390-392.

98 2! McGinty, ,I.F., Henriksen, SJ., Goldstein, A., Terep;',,,,',, L. and Bloom, F.E.. Dynorphin is contained wiihin hipnoeampal mossy fibers: immunochemical alterations after kainic acid ~dministration and coichicine-induced neurotoxicity, Proc. Natl. Acad. Sci. USA, 80 (1983) 589-593. 22 Morris, B.J., Haarmann, I., Kempter, B., Hollt, V. and Herz, A., Localization of prodynorphin messenger RNA in rat brain by in situ hybridization using a synthetic oligonucleotide probe, Neurosci. Let:., 69 (1986) 104-108. 23 Morris, B.,I., Moneta, M.E., ten Bruggencate, G. and Holit, V., Levels of prodynorphin mRNA in rat dentate gyrus are decreased during hippocampal kindling, Neurosci. Left., 80 (1987) 298-302. 24 Sakamoto, N., Michel, J.-P., Kopp, N., Tohyama, M. and Pearson, J., Substance P- and enkephalin-immunoreactive neurons in the hippocampus and related areas of the human infant brain, Neuroscieace, 22 (1987) 801-81 I. 25 Seifert, W., Ranscht, B., Fink, H.J., Forster, F., Beckh, S. and Muller, H.W., Development of hippocampal neurons in cell culture: a molecular approach, in W. Seifert (Ed.), Neurobiology of the Hippocampus, Academic Press, New York, 1983, pp. 109-135. 26 Stengaard-Pederse~. K., Fredens, K. and Larsson, L.-I., Comparative localization of enkephalin and cholecystokinin immunoreactivitics and heavy metals in the hippocampus, Brain Res., 273 (1983) 81-96.

27 Van Daal, J.H.H.M., Zanderink, H.E.A., .1¢likg, B.G. aad Van Abeelen, J.H.F., Distribution of dynorphi: L ann ;nethionine-enkephalin in the mouse hippocampus: influ.~ncc ,~f genotype, Neurosci. Lett., 97 (1989) 241-244. 28 Vilijn, M.-H., Vaysse, P.J.-J., Zukin, S. and Kessler, J.A., Expression of preproenkephalin mRNA by c:dtured astro~es and neurons, Proc. Natl. Acad. Sci. USA, 85 (1988) 6551-6555. 29 Xie, C.W., Lee, P.H.K., Takeuchi, K.. Owyang, V., Li, S.J., Douglass, J. and Hong, J.S., Single or repeated electroconvulsive shocks alter the levels of prodynorr,hin and proenkephalin mRNAs in rat brain, Mol. Brain Res., 6 (1989) 11-19. 30 Xie, C.W., Lee, P.H.K., Douglass, J., Crain, B. and Hong, J.S., Deep prepyriform cortex kindling differentially alters the levels of prodynorphin mRNA in rat hippocampus and striatum, Brain Res., 495 (1989) 156-160. 31 Xie, C.W., McGinty, J.F., Lee, P.H.K., Mitchell, C.L. and Hong, J.S., A glutamate antagonist blocks perforant path stimulation-induced reduction of dynorphin peptide and prodynorphin mRNA levels in rat hippocampus, Brain Res., 562 (1991) 243-250. 32 Yoshikawa, K., Williams, C. and Sabol, S.L., Rat brain preproenkephalin mRNA, J. Biol. Chem., 259 (1984) 14301-14308.