HIPPOCAMPUS, VOL. 2, NO. 2, PAGES 107-126, APRIL 1992

Ultrastructural Localization of Neuropeptide Y-like Immunoreactivity in the Rat Hippocampal Formation Teresa A. Milner and Erol Veznedaroglu Department of Neurology and Neuroscience, Division of Neurobiology , Cornell University Medical College, New York, NY, U.S.A.

ABSTRACT Neuropeptide Y (NPY) has been implicated in the modulation of hippocampal neuronal activity and in the pathophysiology of several neurological disorders involving the hippocampal formation. Thus, this study examines the light and electron microscopic immunoperoxidase labeling of a rabbit polyclonal antibody against porcine NPY in single sections through each lamina of the CA1 and CA3 regions of the hippocampus and the dentate gyrus (DG) of normal adult rats. By light microscopy, the majority of perikarya with intense NPY-like immunoreactivity (NPYLI) were located in stratum oriens of CA1 and CA3 of the hippocampus and in the hilus of the DG. Fine varicose processes with NPY-LI were found in all layers of the hippocampal formation, but were densest in the outer third of the molecular layer of the DG. The density of NPY-labeling was greater in the ventral portion of the hippocampal formation. By electron microscopy, most NPY-containing perikarya in all three hippocampal regions were: small (8-12 pm) or medium-sized (12-18 p n ) and elongated; or medium-sized and round. A dense accumulation of NPY-LI was commonly observed within the individual saccules of Golgi complexes and some rough endoplasmic reticulum in the cytoplasm. Perikarya and dendrites with NPY-LI usually were directly apposed to other neuronal processes (mostly terminals) and lacked astrocytic appositions. The majority of terminals in contact with NPY immunoreactive neurons were unlabeled and synapsed with the shafts of large and small dendrites. In CA1 and CA3 of the hippocampus, the types of synapses formed by the unlabeled terminals were not significantly different; however, more asymmetric synapses than symmetric synapses were formed by the unlabeled terminals on the shafts of small NPY-labeled dendrites in the DG. The terminals with NPY-LI (0.25-1.2 pm) contained many small, clear vesicles and 0-2 large, dense-core vesicles. The types of synapses (i.e., asymmetric and symmetric) and distribution of NPY-labeled terminals on the targets were remarkably similar in each lamina of the hippocampal subregions. The NPY-labeled terminals usually synapsed with one unlabeled perikaryon or dendrite. However, others synapsed either (1) with two unlabeled perikarya or dendrites simultaneously or (2) with one NPY-containing perikaryon or dendrite. Most of the terminals with NPY-LI formed symmetric junctions with the shafts of small (distal) dendrites. The remaining NPY-labeled terminals either (1) formed appositions that lacked a membrane specialization, but were without apparent glial intervention in the plane of section analyzed, with NPYlabeled and unlabeled perikarya and dendrites; or (2) were closely apposed without glial intervention to unlabeled and NPY-labeled terminals. The dense input from terminals forming asymmetric (excitatory)junctions and the limited glial coverage may contribute to the vulnerability of NPY-immunoreactive neurons in disorders such as Alzheimer’s disease or epilepsy. Additionally, these results provide cellular substrates in the rat hippocampal formation for (1) the direct synaptic input of NPY-labeled terminals to nonNPY-containing hippocampal neurons; and ( 2 ) presynaptic interactions between NPY-containing terminals and terminals containing other transmitters. Key words: neuropeptides, hippocampus, dentate gyrus, electron microscopy, astrocyte

Correspondence and reprint requests to: Dr. Teresa A. Milner, Division of Neurobiology, Cornell University Medical College, 41 1 East 69th Street, New York, NY 10021 U.S.A.

107

108

HZPPOCAMPUS VOL. 2, NO. 2, APRIL 1992

Recent attention has been directed to neuropeptide Y (NPY) in the hippocampal formation, in part because patients with Alzheimer’s disease have a significant reduction in both the levels and immunocytochemical detectability of the peptide (Chan-Palay et al., 1986b; Chan-Palay, 1987) and because NPY perikarya and processes are altered in both experimentally induced seizures (Sloviter, 1989) and patients with temporal lobe damage during epilepsy (de Lanerolle et al., 1989). Additionally, NPY is thought to be important in memory retention, recall, and prevention of amnesia (Flood et al., 1987; 1989; Morley and Flood, 1990). Most physiological studies in the hippocampus proper indicate that NPY acts presynaptically to inhibit excitatory transmission of pyramidal cells (Colmers et al., 1985; 1987; 1988; Colmers and Pittman, 1989; Colmers, 1990). However, NPY may directly inhibit the firing of pyramidal cells (Haas et al., 1987) and directly excite dentate granule cells (Brooks et al., 1987). Thus, NPY may have disparate functions in the hippocampal formation depending on the region studied. However, the quantitative examination of the synaptic relations of NPY-labeled terminals within the hippocampus proper compared to the dentate gyrus that might provide an ultrastructural correlate for these disparate physiological functions has not been examined. In the present study the avidin-biotin complex (ABC) peroxidase method (Hsu et al., 1981) for the localization of antibodies to porcine NPY in the hippocampal formation was used ( I ) to examine the morphological features and synaptic associations of NPY-containing neurons; and (2) to compare the type of synaptic associations and the postsynaptic targets of NPY-labeled terminals between the CAI and CA3 regions of the hippocampus and the dentate gyrus.

of NPY and closely related peptides, peptide YY and pancreatic polypeptide, were dissolved in water to yield a concentration of lop4 M. The peptides were spotted on pieces of Whatman no. 1 filter paper held by a filtration manifold (Schleicher and Schuell). Each well in the filtration unit contained 10 FL of one of the peptides at concentrations ranging from 0 to lo-’ M. The peptides were air-dried on the paper and exposed to paraformaldehyde vapors at 80°C for 1 hour. The papers were then immunolabeled with a 1 :3,000 dilution of the polyclonal NPY antibody according to the procedure of Larsson (1981). The antibody was localized immunocytochemically in the tissue using the ABC peroxidase method (Hsu et al., 1981). In brief, the sections were incubated sequentially in: ( I ) rabbit antibody to NPY (1 :3,000 dilution) for 16-20 hours; (b) a 1:50 dilution of goat anti-rabbit immunoglobulin (IgG) conjugated to biotin (Vector) for 0.5 hours; and (c) peroxidase-avidin complex (Vectastain Elite kit at double the recommended dilution) for 0.5 hours. The ABC reaction product was visualized following incubation of the tissues with 3,3’-diaminobenzidine (DAB; Aldrich) and hydrogen peroxide. All incubations were carried out at room temperature with continuous agitation. The diluents for the incubations were prepared with 0.1 M Tris-saline (pH 7.6) containing 0.1% bovine serum albumin (BSA). Triton-X (0.25% for light microscopy; 0.05% for electron microscopy) was added to the primary antibody during the incubation step.

Processing for light and electron microscopy For light microscopy, the labeled sections were mounted on acid-cleaned slides previously coated with 1% gelatin, airdried, dehydrated, and coverslipped with Histoclad. The final

METHODS

M

Fixation and preparation of sections for immunocytochemistry Studies were conducted in six adult male Sprague-Dawley rats (250-275 g; Hilltop Lab. Animals, Inc.). The rats were deeply anesthetized with Nembutal (SO mgikg, i.p.) and perfused through the ascending aorta sequentially with ( I ) 1015 mL of normal saline (0.9%) containing 1,000 units/mL of heparin, (2) 50 mL of 3.75% acrolein (Polysciences) and 2% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4), and (3) 200 mL of 2% paraformaldehyde in 0.1 M phosphate buffer. The regions of the forebrain containing the hippocampal formation (as described by Paxinos and Watson, 1987) were removed and cut into coronal blocks 6 mm thick and stored in the latter fixative for an additional 30 minutes. Sections (40 km thick) were then cut on a Vibratome through the hippocampal formation and collected in 0. I M phosphate buffer and treated with 1% sodium borohydride in 0.1 M phosphate buffer for 30 minutes prior to immunocytochemical labeling (Milner and Bacon, 1989a, 1989b).

lmmunocytochernistry A polyclonal antibody to porcine NPY prepared in rabbits (Peninsula) was used for these studies. It was tested for specificity by immunodot-blots. NPY as well as several fragments

0 1

~

9

la8 10-7

105

1-36 13-36 22-36 DA

NPY

FA 1-24 PYY PP

I

Fig. 1 . Immunodot-blot depicting the cross-reactivity between a 1 :3,000 dilution of a polyclonal antibody to porcine neuropeptide Y and varying concentrations (0M) of human neuropeptide Y (NPY) fragments 1-36, 13-36, and 22-36, desamido-NPY (DA), free-acid-NPY (FA), NPYI-24NH2(1-24) and porcine peptide YY (PYY), and rat pancreatic polypeptide (PP).

EM OF NEUROPEPTIDE Y IN HIPPOCAMPAL FORMATION / Milner and Veznedaroglu preparations were examined using Differential Interference Contrast (DIC) optics on a Nikon Microphot microscope. For electron microscopy, the labeled sections were fixed for 1 hour in 2% osmium tetroxide in 0.1 M phosphate buffer and embedded in Epon as described previously (Milner and Bacon, 1989~).Final preparations were examined with a Philips 201 electron microscope. Tissues from the animals with the best immunocytochemical labeling and preservation of ultrastructural morphology were evaluated. At least ten grids containing three to five thin sections were collected from the surface of six or more plastic-embedded Vibratome sections of the hippocampal formation from each of the best animals. All labeled profiles within one thin section from every other grid were photographed and quantitatively assessed for both the type of la-

109

beling and relationship to neuronal and non-neuronal processes. The term “association” includes asymmetric and symmetric synapses and appositions of parallel membranes neither separated by astrocytic processes nor showing conventional synaptic clefts, intercleft material, or dense specializations. Differences in the frequencies of the types of synapses of NPY-containing terminals were calculated using a chi-square statistical analysis.

RESULTS

Antibody specificity The specificity of the NPY antibody with respect to crossreactivity with fragments of the peptide as well as closely related peptides was assessed by immunodot-blotting. As

A. CAI alv

so SP

SR

SLM fis

B. CA3 C. DG

SLM +

SR

s L;‘ SP

GCL

so

H

Fig. 2 . Light microscopic distribution of NPY-LI in a coronal section of the hippocampal formation taken from a midseptotemporal level. Boxed regions in the schematic diagram in the upper left-hand corner correspond to the camera lucida drawings that show the distribution of NPY-labeled perikarya and processes in (A) the CA1 region of the hippocampus, (B) the CA3 region of the hippocampus, and (C) the dentate gyrus (DG) and also designate the regions sampled for electron microscopy. a h , alveus; fis, hippocampa1 fissure; GCL, granule cell layer; H, hilus; ML, molecular layer; SLu, stratum lucidum; SLM, stratum lacunosum-moleculare; SO, stratum oriens; SP, stratum pyramidale; SR, stratum radiatum. Calibration bar, 100 p n .

110 HZPPOCAMPUS VOL. 2, NO. 2, APRIL 1992

Fig. 3 . Electron microscopic localization of NPY-LI in perikarya from stratum oriens of the CAI region of the hippocampus. This perikarya is medium-sized (12-18 pm) and contains an indented (curved arrow) nucleus (N) with one nucleolus (n) and a medium amount of cytoplasm. Mitochondria (m), rough endoplasmic reticulum (ER), and Golgi complexes (boxed region) are among the distinguishable organelles in the cytoplasm. Both the GoIgi complexes (inset at top right) and rough endoplasmic reticulum have a dense accumulation of ABC reaction product when compared to the mitochondria. Calibration bars, 0.5 hm.

seen in the immunodot-blot (Fig. l ) , the NPY antibody recognized all NPY fragments tested. Additionally, it cross-reacted with the highest concentrations of peptide YY (PYY) and pancreatic polypeptide (PP). Thus, labeling for NPY is referred to as NPY-like imrnunoreactivity (NPY-LI) to reflect the antigen used for production of the antibody and the crossreaction with PYY and PP.

light microscopy Perikarya with NPY-LI were observed most frequently in stratum oriens of the CAI and CA3 regions of the hippocampus and in the hilus of the dentate gyrus (Fig. 2 ) . Neuronal cell bodies were seen sometimes in the inner portion of the molecular layer of the dentate gyrus, strata pyramidale and

r

Fig. 4. Morphological features of NPY-containing perikarya in the hilus of the dentate gyrus. (A) A small (8-12 hm in

diameter), elongated NPY-labeled perikarya containing an indented (curved arrow) nucleus (N) and an abundant cytoplasm. Rough endoplasmic reticulum (ER), mitochondria (m),and dense core vesicles (dcv) are among the distinguishable organelles in the cytoplasm. (B) A medium-sized, round NPY-labeled perikarya containing an indented nucleus (curved arrow) and an abundant cytoplasm. Golgi complexes (G), mitochondria (m), multivesicular bodies (mvb), and dense-core vesicles (dcv) are found in the cytoplasm. Dense accumulations of ABC reaction product are found in the Golgi complexes when compared to other organelles. Calibration bars: A, 0.5 pm; B , 1.0 pm.

EM OF NEUROPEPTIDE Y IN HIPPOCAMPAL FORMATION / Milner and Veznedaroglu

111

1 12 HIPPOCAMPUS VOL. 2, NO. 2, APRIL 1992 radiatum of CAI, and strata pyramidale, lucidum, and radiaturn of CA3. The number of NPY-immunoreactive perikarya appeared slightly greater in the ventral aspects of the hippocampal formation, particularly in the hilus. Varicose processes with NPY-LI were found throughout all layers of the hippocampal formation (Fig. 2 ) but were densest in the outer one-third of the molecular layer of the dentate gyrus. Processes within the inner two-thirds of the molecular layer of the dentate gyrus were oriented perpendicular to the granule cell layer (Fig. 2C). Conversely, processes in stratum lacunosum-moleculare of CAI and CA3 of the hippocampus often ran parallel to the hippocampal fissure (Fig. 2A,B). Like perikarya, the number of varicose processes appeared greater in the ventral aspects of the hippocampal formation.

Table 2. Dendritic Size vs. Type of Synapse Formed by Terminals with NPY-LI in t h e Hippocampal Formation Type of Synapse Region

CA 1

Electron Microscopy The boxed regions of the hippocampal formation shown in the camera lucida drawings in Figure 2 correspond to those areas sampled for electron microscopy. All samples were taken from a midseptotemporal level of the hippocampal formation.

CA3

Morphology of NPY-labeled neurons

The morphological characteristics of NPY-containing perikarya in the hippocampus and dentate gyrus were similar. Most NPY-labeled perikarya were: (1) medium-sized (12-18 pm) and elongated (Fig. 3); (2) medium-sized and round (Fig. 10B); and (3) small (8-12 pm) and elongated (Fig. 4A). All three types of perikarya contained an unlabeled indented nucleus as well as an occasional nucleolus (Fig. 3). The quantity of cytoplasm varied with the majority of perikarya having an abundant amount. Distinguishable cytoplasmic organelles included: mitochondria, Golgi complexes, rough endoplasmic reticulum (RER), a few dense-core vesicles (dcvs), and mul-

Dentate gyms

Type of Dendrite

SO (n = 26) Large (1.5-3.5 pm) Small (0.5-1.5 pm) Spine SR (n = 93) Large Small Spine SL-M (n = 57) Large Small Spine SO (n = 22) Large (1.5-4.5 pm) Small (0.5-1.5 pm) Spine SLu/SR (n = 125) Large Small Spine SL-M (n = 113) Large Small Spine Hilus (n = 93) Large (1 5 - 3 3 pm) Small (0.5-1.5 pm) Spine M L (n = 148) Large Small Spine

Asymmetric

0 1

3 0 0 10

Symmetric

I 20 1

15 (1) 50 17

0 2 4

2 43 6

0 0 4

17 0

0 2 12 (1)

1

38 65 ( 5 )

2

1

9 88 5

2 6 (2) 5

20 (1) 48 8 (1)

3 7

2 10 12 (1)

6 80 37

Dendrites with NPY-LI are indicated in parentheses.

Table l . Types of Synapses Formed by Terminals on Neurons with NPY-LI in t h e Hippocampal Formation Type of Synapse Region Sampled

CAI (n = 59) Peri karya Large dendrites (1.5-3.5 pm) Small dendrites (0.5-1.5 pm) Spines CA3 (n = 238) Perikarya Large dendrites Small dendrites Spines DG (n = 360) Perikarya Large dendrites Small dendrites Spines

Asymmetric

Symmetric

0 10

5

4

8

I

9 87 68 12 (2)

Terminals with NPY-L.1 are indicated in parentheses.

50 69 (2) 52 9

tivesicular bodies. Compared to mitochondria in the same perikarya, the individual saccules of the Golgi complexes and some RER had a dense accumulation of ABC-reaction product (Figs. 3, 4B). Perikarya with NPY-LI usually were abutted by other neuronal processes (especially unlabeled terminals) and lacked astrocytic appositions. NPY-LI was seen in large (1.51-4.5 pm in diameter) proximal dendrites and in small (0.5-1.5 pm in diameter) distal dendrites. Some large dendrites were similar to perikarya in that they contained mitochondria, smooth and RER, large dcvs, and microtubules (Figs. 5A, 6B). Other large dendrites contained only mitochondria, smooth endoplasmic reticulum, and microtubules (Figs. 5B, 6A, 7A). Smaller dendrites usually contained only distinguishable microtubules and mitochondria (Fig. 7A). Some of the dendritic spines contained smooth endoplasmic reticulum (Fig. 7B), whereas others did not (Fig. 7C). In dendrites, as in soma, the ABC reaction product was often seen as aggregates associated with endoplasmic reticulum, large dcvs, or multivesicular bodies (Figs. 5B, 6B). The peroxidase reaction product was often light in dendritic spines (Fig. 7B). Like the perikarya, NPY-labeled dendrites often were surrounded by neuronal processes, es-

EM OF NEUROPEPTIDE Y IN HIPPOCAMPAL FORMATION / Milner and Veznedaroglu

113

Fig. 5. Associations of dendrites with NPY-LI in strata radiatum of the CA3 region of the hippocampus. (A) A large NPYlabeled dendrite with several dense-core vesicles (arrowheads) receives an asymmetric synapse (open arrow) from an unlabeled terminal (uT). The labeled dendrite is also partially invaginated (small arrows) by an unlabeled dendrite (LID,),which receives asymmetric contacts (open arrows) from numerous unlabeled terminals. The NPY-labeled dendrite apposes (single small arrow) another unlabeled dendrite (uD2) and an unlabeled dendritic spine (single arrow, top left); the dendritic spine also receives an asymmetric synapse from an unlabeled terminal. Note the absence of glial processes surrounding the labeled dendrite. (B) A mossy fiber terminal (mT) forms multiple asymmetric synapses (open arrows) with a NPY-containing dendrite (NPY-D) as well as a number of unlabeled dendrites (LID). The NPY-labeled dendrite contains a multivesicular body (arrowhead), which has a dense accumulation of ABC reaction product. Calibration bars: A, 1.0 pm; B, c, 0.5 pm.

114 HIPPOCAMPUS VOL.

2, NO. 2, APRIL 1992

Fig. 6. Associations of NPY-labeled dendrites in the hilus of the dentate gyrus. (A) Numerous unlabeled terminals (uT) form synapses (open arrows) with a large NPY-labeled dendrite. (B) A large dendrite with NPY-LI containing rough endoplasmic reticulum (ER) and a dense-core vesicle (dcv) receives only a few contacts (open arrows) from unlabeled terminals. The portions of the dendrite not contacted by the terminals are surrounded by glial processes (asterisks). Calibration bars, 0.5 pm.

pecially unlabeled terminals and dendrites, and lacked glial appositions (Fig. 5A). However, some large NPY-labeled dendrites had few neuronal appositions and were partially invaginated by glial processes (Fig. 6B). The axons with NPY-LI were small (0.1-0.2 pm diameter) and were primarily unmyelinated (Fig. 12A). Occasionally, a few axons with three to four laminae of myelin were observed (not shown). The NPY-containing terminals ranged from 0.25 to 1.2 pm in diameter, with the majority less than 0.8 pm in diameter; they contained a few mitochondria and numerous small clear vesicles (scvs) (Figs. 8-13). Sometimes, the terminals with NPY-LI also contained one to two large dcvs (Figs. IOA, 12A). The ABC reaction product usually filled the axon terminal either rimming small vesicles (Fig. 8A) or more discretely associated with dcvs, the majority of which were directed toward parts of the plasmalemma away from junctions (Fig. 10A). Associations of NPY-labeled perikarya and dendrites

Perikarya and dendrites with NPY-LI were postsynaptic to both unlabeled and NPY-labeled terminals. Of the terminals included in the quantitative analysis, the majority were unlabeled (Table 1; Figs. 4, 6). In CAI, a significant number

more of the unlabeled terminals synapsed on NPY-labeled perikarya and large dendrites, whereas in CA3 and the dentate gyrus more terminals synapsed on large and small NPYlabeled dendrites (P < .05). The distribution of the synaptic types varied depending on the target. Most synapses formed by the unlabeled terminals on perikarya were symmetric ( P < .05). In general, the types of synapses formed by the unlabeled terminals on large and small dendrites were seen with comparable frequency ( P > .05) (Figs. 5B, 6,7A). However, in the dentate gyrus, significantly more (P< .05) of the synapses on NPY-containing small dendrites were asymmetric. Although only a few synapses were observed on NPY-labeled dendritic spines, most of these were asymmetric specializations ( P < .05) (Fig. 7B,C). However, with both labeled and unlabeled terminals, the dense accumulation of ABC-reaction product near the postsynaptic site of the junction made the type of synapse difficult to differentiate. Usually the NPYimmunoreactive perikarya and dendrites were postsynaptic to more than one terminal in a single section. In addition to the synaptic contacts, NPY-labeled perikarya and dendrites were closely apposed to a number of unlabeled terminals (CAI = 56; CA3 = 149; DG = 129). These appositions were without glial intervention but lacked

EM OF NEUROPEPTIDE Y IN HIPPOCAMPAL FORMATION / Milner and Veznedaroglu

115

Fig. 7. Associations on NPY-labeled dendrites and spines in stratum oriens of the CA3 region of the hippocampus. (A) A small NPY-labeled dendrite receives a perforated asymmetric synapse (curved arrows) and a symmetric synapse (open arrow) from two unlabeled terminals (uT, and uTZ, respectively). (B, C) Unlabeled terminals form asymmetric synapses (open arrows) with NPY-containing dendritic spines. The NPY-labeled spine in D contains smooth endoplasmic reticulum (SER). Calibration bars, 0.5 pm.

a membrane specialization in the plane of section analyzed. The distribution of these terminal appositions on the NPYimmunoreactive perikarya and dendrites was similar to that observed with the synapses. Associations of NPY-labeled terminals in the hippocampus

The types of synapses formed by NPY-labeled terminals were quantitatively evaluated for each layer of the CA1 and CA3 regions of the hippocampus (Tables 2 , 3). In all layers, the majority of NPY-labeled terminals synapsed with unlabeled perikarya and dendrites. The somatic synapses (14 in CA1, 13 in CA3) formed by the NPY-containing terminals were found exclusively in strata oriens and pyramidale (not shown in Table 2) and were characterized as symmetric specializations. In all layers of the hippocampus (except for stratum pyramidale), most of the NPY-labeled terminals synapsed with dendrites. The majority of the postsynaptic dendrites were unlabeled; however, a few contained NPY-LI (Tables 2 , 3). Some of the unlabeled dendrites were of pyramidal cell origin. The type of synapse formed by the NPY-labeled terminals was compared to the size of the target dendrites for strata oriens, radiatum, and lacunosum-moleculare (Table 2). In all

three layers, more synapses ( P < .05) were formed on the shafts of small dendrites (0.5-1.5 p.m in diameter) (Figs. 9, IOB, 11A, C) than the larger (1.51-3.5 pm in diameter) dendrites and dendritic spines (Fig. 7B). Moreover, most of the synapses formed by the NPY-labeled terminals were symmetric (Figs. SA, 9, 10, 11A,C) (P< .05). However, the type of synapse formed by NPY-labeled terminals on dendritic spines varied depending on the layer examined. Sometimes the terminals with NPY-LI formed asymmetric junctions; these were usually on dendritic spines (Fig. 8B). The target dendrite of the NPY-labeled terminals was often postsynaptic to other unlabeled terminals (Figs. SC, 9B, 10B, llA,C). Usually one NPY-labeled terminal contacted a single dendrite or dendritic spine. However, a small portion (Table 3) of the terminals synapsed on more than one dendrite (Figs. SA, 9A, 1 IA). Occasionally, more than one NPY-labeled terminal contacted a single dendrite (Fig. 11A). In addition to synapses, many terminals with NPY-LI were closely apposed to unlabeled dendrites (Table 3). These appositions were without any glial intervention but lacked a distinguishable membrane specialization in the plane of section analyzed. In each hippocampal layer, the distribution of these appositions on the dendrites paralleled that observed for synapses.

116 HZPPOCAMPUS VOL. 2, NO. 2, APRIL 1992

Fig. 8. Associations of terminals with NPY-LI in stratum oriens of the CA1 region of the hippocampus. (A) NPY-labeled terminal forms a symmetric synapse (curved arrow) and an apposition (straight arrow) with two unlabeled dendrites (uD, and uD2). (B) A terminal with NPY-LI forms an asymmetric synapse (arrow) with the spine of an unlabeled dendrite (uD). (C) A NPY-containing terminal with numerous small clear vesicles (scv) directly apposes (arrowhead and small arrows, respectively) the spine of an unlabeled dendrite (uD) and to an unlabeled terminal (LIT).The unlabeled terminal also forms asymmetric synapses (open arrows) with the same dendritic spine as well as adjacent one (top right). Calibration bars, 0.25 Fm.

EM OF NEUROPEPTIDE Y IN HIPPOCAMPAL FORMATION / Milner and Veznedaroglu

Fig. 9. Associations of NPY-labeled terminals in strata radiatum and lacunosum-moleculare of the CA1 region of the hippocampus. (A) Two small unlabeled dendrites (uD1 and uDz) receive symmetric synapses (curved arrows) from the same NPY-labeled terminal. An unlabeled terminal also forms an asymmetric synapse (open arrow) with uD1. (B) A terminal with NPY-LI (NPY-TI) directly apposes (two small arrows) with three small unlabeled dendrites (uD1, uDz, and u D ~ ) .One of the unlabeled dendrites (uD2) also receives a symmetric synapse from a second NPY-labeled terminal (NPY-T2). Calibration bars: A, 0.5 pm; B, 0.25 pm.

117

Fig. 10. Associations of NPY-labeled terminals in stratum radiatum of the CA3 region of the hippocampus. (A) NPYlabeled terminal forms a symmetric synapse (curved arrow) with the shaft of large unlabeled dendrites (uD). Note that the dense-core vesicle (dcv) is away from the sight of the junction. (B) NPY-labeled terminal forms a symmetric synapse (curved arrow) on an unlabeled dendritic spine, which also receives an asymmetric synapse (open arrow) from an unlabeled terminal (uT). The labeled terminal is also in direct apposition (small arrow) to the same unlabeled terminal. Calibration bars, 0.5 pm.

118

HIPPOCAMPUS VOL. 2, NO. 2, APRIL 1992

Table 3. Relation of Terminals with NPY-LI To Other Neuronal Profiles* N py BiiZiii?. ........ ........ ......... ........ ;:.:.:.:.:.:.:.: ....... .:.:.:........... :.:.:.:.>

__

CAI

S. oriens (n = 112) S. radiatum (n = 398) S. lac.-mol. (n = 274) CA3 S. oriens (n = 70) S. luc./rad. (n = 378) S. lac.-mol. (n = 419) Dentate gyrus ML (n = 437) Hilus (n = 309)

65 (58%) 247 (62%) 164 (60%)

6 (5%) 16 (4%) 21 (7.6%)

4 (4%) 1 (0.3%) 1(0.4%)

18 (16%) 64 (16. I%) 55 (20%)

19 (17%) 70 (17.6%) 33 (12%)

42 (60%) 269 (71%) 249 (59%)

5 (7%) 29 (8%) 41 (10%)

0 9 (2%) 3 (1%)

13 (19%) 33 (9%) 70 (1 7%)

10 (14%) 38 (10%) 56 (13%)

254 (58%) 189 (61%)

48 (11%) 16 (5%)

2 (0.5%) 8 (3%)

86 (19.7%) 36 (12%)

47 (10.8%) 60 (19%)

_____________.

* Includes all synapses and appositions not separated by glia.

Of the remaining terminals or preterminal axons with NPYLI, a number were not associated with any neuronal processes in the plane of section analyzed (Table 3). These terminals usually were separated from the neuropil by glial processes (Fig. 9B). However, direct appositions between two axon terminals without any glial intervention were commonly observed (Figs. 8C; IlB,C). These appositions were with other unlabeled terminals as well as those containing NPYLI (Table 3). The apposed unlabeled terminals often formed asymmetric synapses with dendrites and dendritic spines (Fig. 8C). Associations of NPY-labeled terminals in the dentate gyms

The types of synapses formed by NP Y-labeled terminals were quantitatively evaluated for the hilus and molecular layers of the dentate gyrus (Tables 2, 3). The somatic synapses formed by the NPY-containing terminals were found exclusively in the hilus (n = 55) and the granule cell layer (n = 30) (not shown in tables); they were exclusively symmetric. In both the hilus and molecular layer, most of the NPYlabeled terminals synapsed with dendrites. The majority of the postsynaptic dendrites were unlabeled; however, a few contained NPY-Ll (Table 3). Some of the unlabeled dendrites in the inner molecular layer originated from granule cells. In both the molecular layer and hilus, more synapses ( P < .05) were formed on the shafts of smaller dendrites (0.5-1.5 pm in diameter) than the larger (1.51-3.5 pm in diameter) den-

drites and dendritic spines (Figs. 12A,B, 13B). Moreover, a significant number of the synapses (P < .05) formed by the NPY-labeled terminals on all three types of dendrites were symmetric (Figs. 12A,B, 13) rather than asymmetric synapses (P < .05). In the outer molecular layer, unlabeled dendrites and dendritic spines contacted by clusters of NPY-labeled and unlabeled terminals (Fig. 13B) were especially common. In these cases, intervening glial processes were notably absent around the labeled processes. Additionally, one NPYlabeled terminal usually contacted a single dendrite or dendritic spine; however, a small portion (Table 3) of the terminals were apposed to more than one dendrite (Figs. 12B, 13B). Like the hippocampus, a number of the terminals with NPY-LI were apposed to unlabeled dendrites without glial intervention but lacked a membrane specialization in the plane of section analyzed (Figs. 12B, 13). Additionally, many of the remaining NPY-labeled terminals or preterminal axons were not associated with any neuronal processes in the plane of section analyzed (Table 3). Direct appositions between two axon terminals without any glial intervention were seen in a number of instances (Fig. 13B).

DISCUSSION The present study demonstrated that (1) the morphology and synaptic associations of NPY-containing neurons are similar in the hippocampus and dentate gyrus; ( 2 ) the majority of NPY-labeled neurons receive a heavy synaptic input and

+

Fig. 1 1 . Associations of terminals with NPY-LI in stratum lacunosum-moleculare of the CA3 region of the hippocampus.

(A) Two NPY-labeled terminals (NPY-TI and NPY-T2) form symmetric synapses (curved arrow) and an apposition (straight arrow) on the shaft of the same unlabeled dendrite (uDZ).The unlabeled dendrite also receives an asymmetric synapse (open arrow) from an unlabeled terminal. One of the NPY-containing terminals (NPY-TI) also directly apposes (small arrows) another unlabeled dendrite (uD,). (B) NPY-labeled terminal forms a symmetric synapse (curved arrow) with a small unlabeled dendrite (uD). The labeled terminal is also in direct apposition (small arrows) to an unlabeled terminal. (C) A terminal with NPY-LI forms a symmetric synapse (curved arrow) with a small unlabeled dendrite (uD) and directly apposes (small arrows) an unlabeled terminal, which forms an asymmetric synapse (open arrow) with an unlabeled dendritic spine. The unlabeled dendrite (uD) is also postsynaptic (open arrow) to an unlabeled terminal. Calibration bars, 0.5 pm.

EM OF NEUROPEPTIDE Y IN HIPPOCAMPAL FORMATION I Milner and Veznedaroglu

119

120 HZPPOCAMPUS VOL. 2, NO. 2, APRIL 1992

Fig. 12. Associations of terminals with NPY-LI in the hilus of the dentate gyrus. (A) An NPY-labeled terminal forms a symmetric synapse (curved arrow) with a small unlabeled dendrite. The region of the terminal not involved in the junction is invaginated by a glial process (asterisks) identified by the presence of numerous glial filaments (go. Note the apposition of a dcv (arrowhead) to the glial process. The glial process also apposes (small arrows) with an NPY-containing axon (A). (B) A small unlabeled dendrite (uD) receives a symmetric synapse (curved arrow) and direct apposition (small arrows) from two NPY-labeled terminals. One of the NPY-containing terminals (top) also forms a symmetric synapse (single arrow) with the spine of an unlabeled dendrite, which receives an asymmetric junction (open arrow) from an unlabeled terminal (uT). (C) A terminal with NPY-IL forms asymmetric synapses (curved arrows) with three unlabeled dendritic spines. Calibration bars, 0.25 pm.

EM OF NEUROPEPTIDE Y IN HIPPOCAMPAL FORMATION / Milner and Veznedaroglu

121

Fig. 13. Associations of NPY-labeled terminals in the inner (A) and outer (B) molecular layer of the dentate gyrus. (A) The shaft of a large unlabeled dendrite (uD) is associated with two NPY-containing terminals; one forms a symmetric synapse (curved arrow), while the other forms an apposition (small arrows). (B) NPY-labeled terminal forms two associations [one symmetric contact (curved arrow) and one apposition (two small arrows)] with two small unlabeled dendrites (uD, and uDz, respectively). The NPY-labeled terminal is also in apposition (single small arrow) to an NPY-labeled terminal. Both NPYlabeled terminals also appose an unlabeled terminal (arrowheads). Calibration bars, 0.5 wm.

have a sparse glial investment; and (3) NPY-immunoreactive terminals form primarily symmetric synapses on the shafts of small, unlabeled dendrites and are apposed to other NPYlabeled and unlabeled terminals. Specificity of the antibody The immunodot-blot and adsorption controls in the present study demonstrated that the polyclonal antibody against porcine NPY cross-reacted extensively with all NPY fragments and with the highest concentrations of PYY and PP. Thus, we have consistently referred to the reaction product as NPYLI to account for possible cross-reactivity of the antibody with these or other unidentified NPY-like peptides. However, PYY cross-reactivity of the NPY antibody within the hippocampal formation is probably minor since the majority of PYY-immunoreactivity is found in the brainstem (Ekman et al., 1986). Distribution and origin The observed light microscopic distribution NPY-labeled perikarya is consistent with previous studies localizing NPYLI in the rat (Chronwall et al., 1985; Kohler et al., 1986; de Quidt and Emson, 1986; Deller and Leranth, 1990), hedgehog, sheep (Antonopoulos et al., 1989), monkey (Smith et al., 1985; Kohler et al., 1986a, 1986b; Beal et al., 1987), and

human (Chan-Palay et al., 1986; Lotstra et al., 1989). However, the density of perikarya and processes (especially in the CA3 region of the hippocampus) was noticeably greater in the present study than in these previous studies. This observation was unexpected since the present study did not employ colchicine to enhance perikaryal labeling (Ljungdahl et al., 1978). Several factors could account for this discrepancy. These include: (1) different antibodies; (2) different fixation conditions (i.e., paraformaldehyde vs. acrolein); and (3) different labeling techniques (i.e., peroxidase antiperoxidase vs. ABC). The latter possibility is most likely since the ABC technique is extremely sensitive in detecting small amounts of antigen (Hsu et al., 1981). There are at least three sources of NPY-immunoreactive terminals in the hippocampal formation. First, since a number of the NPY-labeled neurons in the hilus co-localize somatostatin (Kohler et al., 1987) and somatostatin neurons primarily in the ipsilateral hilus are known to project to the outer molecular layer of the dentate gyrus (Bakst et al., 1986), many of the NPY terminals in the outer molecular layer probably arise from hilar neurons as well. Second, a few of the NPYimmunoreactive neurons in the dentate gyrus and hippocampus project to the contralateral hippocampal formation (Kohler et al., 1986a). Third, at least some of the NPY-containing terminals in the ventral hippocampal formation arise from the locus coeruleus (Wilcox and Unnerstall, 1990).

122 HIPPOCAMPUS VOL. 2, NO. 2, APRIL 1992 Morphology of perikarya and subcellular localization

By electron microscopy, perikarya that exhibited NPY-LI were similar in all three regions of the hippocampal formation. Many Of the perikarya Of all three types had a dense accumulation of ABC reaction product associated with the Golgi complexes and some RER. The unequal accumulation of ABC reaction product most likely reflects selective subcellular localization of NPY-LI. Such has been proposed previously for Somatostatin in the dentate gyrus (Miiner and Bacon, 1989a). This possibility is supported by observations in other regions showing that the peroxidase localization of antisera against other peptides is the same as that demonstrated with immunogold or immunoautoradiographic techniques (Pickel et al., 1991; Velley et al., 1991). These results are consistent with numerous studies that provide evidence that secretory proteins, e.g., NPY, are contained in different compartments of the Golgi complexes (see Farqyhar, 1985, for review). Some NPY-immunoreactive perikarya and terminals contained large dcvs. In terminals, most of the dcvs were located away from the site of the synapse. In dendrites, the proximity of dcvs to the plasmalemma may suggest sites of exocytotic release (Thureson-Klein et al., 1986). This vesicular content is similar to NPY-containing terminals in the frog pituitary (Danger et al., 1990) and NPY-labeled perikarya and terminals in other brain regions, such as the nucleus of the solitary tract (NTS) (Pickel et al., 1989; Massari et al., 1990). However, perikarya in the NTS contain a much greater number of dcvs than those in the hippocampal formation. In the brainstem (NTS), the presence of dcvs is thought to reflect the coexistence of NPY with another transmitter, particularly catecholamines, or modulator (Winkler et al., 1987; Scherman and Boschi, 1988). The NPY neurons in the hippocampal formation do not co-localize catecholamines but may coexist with somatostatin, since 40-70% of the NPY-immunoreactive neurons also contain this peptide (Kohler et al., 1987). Moreover, some of the NPY-containing neurons may contain yamino butyric acid (GABA), a transmitter known to be colocalized with somatostatin in the hippocampal formation (Somogyi et al., 1984; Kosaka et al., 1988; Kunkel and Schwartzkroin, 1988), and noradrenaline, a transmitter arising from NPY-labeled afferent terminals from the noradrenergic neurons in the locus coeruleus (Wilcox and Unnerstall, 1990). Most of the NPY-labeled perikarya and dendrites received a heavy synaptic input from unlabeled terminals or were abutted by other unlabeled dendrites and thus had a sparse glial investment. These morphological features markedly contrast to those reported in the NTS (Pickel et al., 1989), the nucleus accumbens septi (Massari et aI., 1988), and the striatum (Aoki and Pickel, 1989). In these regions, the perikarya and dendrites receive a sparse synaptic input, are usually not directly apposed to dendrites, and are mostly surrounded by astrocytes. Glial processes contain receptors that may mediate changes in local uptake or metabolism of amino acid transmitters and other products (Stone and Ariano, 1989). Thus, it is interesting to speculate that if NPY neurons in the hippocampal formation are stimulated by an excitatory amino

acid, e.g., glutamate, with only a minor ability to metabolize it, excess stimulation may lead to the toxic accumulation of the excitatory amino acid, resulting in cell death. However, other properties of neurons, e.g., their calcium buffering capacity, may also determine their vulnerability to degeneration since cells immunoreactive for parValbumin are resistant in different seizure models (Sloviter, 1989). Associations on nellrOnS with N P ~ - ~ ~ Unlabeled ferminals-NPY

neurons

The majority of the terminals presynaptic to the NPY-labeled neurons were unlabeled and thus were unidentifiable with respect to both the cellular origin and/or transmitter type. Recently, Deller and Leranth (1990) have shown that NPY-containing neurons in the hilus receive afferent input from ipsilateral mossy cells, the entorhinal cortex, and contralateral hippocampal formation. Some of these terminals probably contain GABA and arise from the medial septall diagonal band nuclei, since similar contacts have been observed on somatostatin and GABAergic perikarya in the hippocampal formation (Freund and Antal, 1988; Yamano and Luiten, 1989; Guylis et al., 1990). Moreover, some of the afferent terminals have the morphological characteristics, i.e., vesicular content and type of synapse, of thalamic (Wouterlood et al., 1990) and noradrenergic afferents (Miher and Bacon, 1989b; 1989~). NPY terminals-NPY

neurons

A small percentage of the terminals that contacted NPYcontaining neurons also contained NPY-LI. These findings are in agreement with previous electron microscopic studies in the dentate gyrus (Deller and Leranth, 1990). The origin of the NPY terminals remains to be determined by future dual-labeling electron microscopic studies combining tract tracers with NPY-immunoreactivity . Associations of terminals with NPY-LI Synaptic associations

The synapses between NPY-containing terminals and unlabeled perikarya and dendrites were similar in all regions of the hippocampal formation studied. These results suggest that the disparate physiological functions of NPY observed between the hippocampus and the dentate gyrus may be due to differences either in the postsynaptic target or in interactions with other afferent systems. Future studies combining physiological and anatomical techniques would be necessary to test these possibilities. Most of the synapses formed by NPY-labeled terminals on unlabeled perikarya and dendrites were symmetric. Symmetric synapses are believed to mediate inhibition. This notion is based largely on ( I ) the paucity of populations of thickened postsynaptic densities in regions of the brain containing higher proportions of inhibitory synapses (Cohen et al., 1982) and ( 2 ) the observation that terminals containing the inhibitory neurotransmitter GABA form mostly symmetric synapses in several regions of the brain (for a review, see Peters et al., 1991). Thus, the results suggest that NPY may have an inhibitory role in the hippocampal formation. Alterna-

EM OF NEUROPEPTIDE Y IN HIPPOCAMPAL FORMATION / Milner and Veznedaroglu

tively, NPY may be colocalized with an inhibitory neurotransmitter. The latter possibility seems more likely since NPY coexists with GABA neurons in many regions of the CNS (Aoki and Pickel, 1989). Despite the abundance of symmetricjunctions, a few NPYlabeled terminals formed asymmetric synapses; the majority of these were observed on dendritic spines. This is of interest since asymmetric synapses are characteristic of terminals containing excitatory transmitters (Peters et al., 1991). Thus, the detection of a few asymmetric synapses primarily on dendritic spines suggests that NPY may also have excitatory functions at more distal sites. This could involve modulatory interactions with other hippocampal afferents converging on the same spine (Rall, 1966). Alternatively, the asymmetric synapses may reflect coexistence of NPY with an excitatory transmitter. A small number of the observed NPY-labeled terminals were observed in direct contact with two different perikarya or dendrites. Such an arrangement could allow one NPYcontaining axon to modulate more than one granule or pyramidal cell simultaneously. Microenvironment

Other profiles in direct contact with NPY-labeled terminals but lacking recognizable synapses included dendrites and terminals. The appositions observed between NPY-labeled terminals and dendrites and terminals may be conducive to preor postsynaptic modulation of hippocampal neurons. Since only single section analysis was performed in this study, some appositions may form synapses in another plane of section. Nevertheless, since astrocytes have uptake sites and metabolizing enzymes for a variety of transmitters (Schousboe et al., 1980), the proximity of the two membranes and the lack of intervening glia would facilitate physiological interactions. Although the transmitter identity of the apposed unlabeled terminals is unknown, the asymmetry of many of the junctions suggests that most of the apposed terminals may contain excitatory transmitters such as glutamate (Peters et al., 1991). Other terminals may contain GABA (particularly those forming symmetric junctions) or noradrenaline. That some of the unlabeled terminals may be noradrenergic is supported by the finding that NPY decreases noradrenaline turnover in the hippocampus (Vallejo et al., 1987) and noradrenergic terminals are observed by electron microscopy in apposition to unlabeled terminals in the hippocampal formation (Milner and Bacon, 1989b; 1989~).Moreover, electron microscopic studies of the cerebral cortex and caudate putamen demonstrate that NPY-containing terminals are sometimes apposed to terminals containing catecholamines and GABA (Aoki and Pickel, 1988; 1989).

Functional considerations The synaptic interactions between NPY-containing neurons and hippocampal neuronal populations may be important in a variety of dysfunctions, including Alzheimer’s disease and epilepsy. In the autopsied hippocampi from patients with Alzheimer-type dementia compared to aged-matched controls, the number of NPY-immunoreactive neurons and fibers are significantly reduced; this loss is most severe in the regions with the highest indices of neurofibrillary tangles and

123

neuritic plaques, including the hilus of the dentate gyrus and the CAI region of the hippocampus (Chan-Palay et al., 1986a,b; Chan-Palay, 1987). The number of NPY-immunoreactive neurons in the hilus of the dentate gyrus is decreased in rats with experimental seizures (Sloviter, 1989) and in human patients with temporal lobe epilepsy (de Lanerolle et al., 1989). Additionally, experimentally induced acute recurrent limbic seizures increase the expression of preproNPY mRNA within hippocampal interneurons followed by a novel appearance of preproNPY mRNA within hippocampal granule and pyramidal cells (Gall et al., 1990; Marksteiner et al., 1990). The observations that (1) the relationships of NPYcontaining neurons and terminals are similar in the hippocampus proper and the dentate gyrus and (2) some regions of the hippocampal formation are more vulnerable than others during certain disease states may suggest that the degeneration of NPY-neurons follows the removal of a particular afferent system or postsynaptic target.

ACKNOWLEDGMENTS The authors thank Ms. Serena Chew for technical assistance and Dr. Virginia M. Pickel for her helpful suggestions on the manuscript. Supported by NIMH Grant MH42834 and NIH Grant 18974. References Antonopoulos, J., A. N. Karamanlidis, G. C. Papadopoulos, H. Michaloudi, A. Dinopoulos, and J. G. Parnavelas (1989) Neuropeptide Y-like immunoreactive neurons in the hedgehog (Erinaceus euvopaeus) and the sheep (Ovis avies) brain. J. Hirnforsch. 30: 349-360. Aoki, C., and V. M. Pickel (1988) Neuropeptide Y-containing neurons in the rat striatum: Ultrastructure and cellular relations with tyrosine hydroxylase-containing terminals and with astrocytes. Brain Res. 459:205-225. Aoki, C., and V. M. Pickel (1989) Neuropeptide Y in the cerebral cortex and the caudate-putamen nuclei: Ultrastructural basis for interactions with GABAergic and non-GABAergic neurons. J. Neurosci. 9:4333-4354. Bakst, I., C. Avendano, J. H. Morrison, and D. G. Amaral (1986) An experimental analysis of the origins of somatostatin-like immunoreactivity in the dentate gyrus of the rat. J. Neurosci. 6: 1452- 1462. Beal, M. F., M. F. Mazurek, and J. B. Martin (1987) A comparison of somatostatin and neuropeptide Y distribution in monkey brain. Brain Res. 405213-219. Brooks, P. A., J. S. Kelly, J. M. Allen, D. A. Smith, andT. W. Stone (1987) Direct excitatory effects of neuropeptide Y (NPY) on rat hippocampal neurones in vitro. Brain Res. 408:295-298. Chan-Palay, V. (1987) Somatostatin immunoreactive neurons in the human hippocampus and cortex shown by immunogold/silver intensification on vibratome sections: Coexistence with neuropeptide Y neurons, and effects in Alzheimer-type dementia. J. Comp. Neurol. 260:20 1-223. Chan-Palay, V., C. Kohler, U. Haesler, W. Lang, and G. Yasargil ’ (1986a) Distribution of neurons and axons immunoreactive with antisera against neuropeptide Y in the normal human hippocampus. J. Comp. Neurol. 248:360-375. Chan-Palay, V., W. Lang, U. Haesler, C. Kohler, and G. Yasargil (1986b) Distribution of altered hippocampal neurons and axons immunoreactive with antisera against neuropeptide Y in Alzheimer’stype dementia. J. Comp. Neurol. 248:376-394.

124 HIPPOCAMPUS VOL. 2, NO. 2, APRIL 1992 Chronwall, B. M . , D. A. DiMaggio, V. J. Massari, V. M. Pickel, D. A. Ruggiero, and T. L. O’Donohue (1985) The anatomy of neuropeptide-Y-containing neurons in rat brain. Neuroscience 15: 1159-1 181. Cohen, R. S., R. K. Carlin, D. J. Grab, and P. Siekevitz (1982) Phosphoproteins in postsynaptic densities. Prog. Brain Res. 56:49-76. Colmers, W. F. (1990) Modulation of synaptic transmission in hippocampus by neuropeptide Y: Presynaptic actions. In J. M. Allen and J. I. Koenig, eds, Central and peripheral sign$cance of neuropeptide Yand its relatedpeptides, pp. 206-218, New York Academy of Sciences, New York. Colmers, W. F., K. Lukowiak, andQ. J. Pittman(1985) Neuropeptide Y reduces orthodromically evoked population spike in rat hippocampal CAI by a possibly presynaptic mechanism. Brain Res. 346:404-408. Colmers, W. F., K. Lukowiak, and Q. J. Pittman (1987) Presynaptic action of neuropeptide Y in area CAI of the rat hippocampal slice. J . Physiol. (Lond.) 383:285-299. Colmers, W. F., K. Lukowiak, andQ. J. Pittman (1988) Neuropeptide Y action in the rat hippocampal slice: site and mechanism of presynaptic inhibition. J. Neurosci. 8:3827-3837. Colmers, W. F., and Q. J. Pittman (1989) Presynaptic inhibition by neuropeptide Y and baclofen in hippocampus: Insensitivity to pertussis toxin treatment. Brain Res. 498:99-104. Danger, J. M., M. C. Tonon, B. G. Jenks, S. Saint Pierre, J. C. Martel, A . Fasolo, B. Breton, R. Quirion, G. Pelletier, and H. Vaudry (1990) Neuropeptide Y: Localization in the central nervous system and neuroendocrine functions. Fundam. Clin. Pharmacol. 4:307-340. de Lanerolle, N. C., J. H. Kim, R. J. Robbins, and D. D. Spencer (1989) Hippocampal interneuron loss and plasticity in human temporal lobe epilepsy. Brain Res. 495:387-395. de Quidt, M. E., and P. C. Emson (1986) Distribution of neuropeptide Y-like immunoreactivity in the rat central nervous system: 11. Immunohistochemical analysis. Neuroscience 18545-618. Deller, T., and C. Leranth (1990) Synaptic connections of neuropeptide Y (NPY) immunoreactive neurons in the hilar area of the rat hippocampus. J. Comp. Neurol. 300:433-447. Ekman, R., C. Wahlestedt, G. Bottcher, F. Sundler, R. Hakanson, and P. Panula (1986) Peptide YY-like immunoreactivity in the central nervous system of the rat. Regul. Pept. 16:157-168. Farqyhar, M. G. (1985) Progress in unraveling pathways of Golgi traffic. Annu. Rev. Cell Biol. 1:447-488. Flood, J. F., M. L. Baker, E. N. Hernandez, and J. E. Morley (1989) Modulation of memory processing by neuropeptide Y varies with brain injection site. Brain Res. 503:73-82. Flood, J. F., E. N. Hernandez, and J. E. Morley (1987) Modulation of memory processing by neuropeptide Y. Brain Res. 421:280-290. Freund, T. F., and M. Antal(l988) GABA-containing neurons in the septum control inhibitory interneurons in the hippocampus. Nature 336: 170-173. Gall, C., J. Lauterborn, P. Isackson, and J. White (1990) Seizures, neuropeptide regulation, and mRNA expression in the hippocampus. Prog. Brain Res. 83:371-390. Gulyas, A. I . , T. J. Gorcs, and T. F. Freund (1990) Innervation of different peptide-containing neurons in the hippocampus by GABAergic septa1 afferents. Neuroscience 37:31-44. Haas, H. L., A. Hermann, R. W. Greene, and V. Chan-Palay (1987) Action and location of neuropeptide tyrosine (Y) on hippocampal neurons of the rat in slice preparations. J. Comp. Neurol. 257: 208-215. Hsu, S. M . , L. Raine, and H. Fanger (1981) Use of avidin-biotinperoxidase complex (ABC) in immunoperoxidase techniques: A comparison between ABC and unlabeled antibody (PAP) procedures,. J. Histochem. Cytochem. 2937-580. Kohler, C., L . Eriksson, S. Davies, and V. Chan-Palay (1986a) Neu-

ropeptide Y innervation of the hippocampal region in the rat and monkey brain. J. Comp. Neurol. 244:384-400. Kohler, C., M. Smialowska, L. G . Eriksson, V. Chan-Palay, and S. Davies (1986b) Origin of the neuropeptide-Y innervation of the rat retrohippocampal region. Neurosci. Lett. 65:287-292. Kohler, C., L. Eriksson, S. Davies, and V. Chan-Palay (1987) Colocalization of neuropeptide tyrosine and somatostatin immunoreactivity in neurons of individual subfields of the rat hippocampal region. Neurosci. Lett. 78:l-6. Kosaka, T., J . Y. Wu, and R. Benoit (1988) GABAergic neurons containing somatostatin-like immunoreactivity in the rat hippocampus and dentate gyrus. Exp. Brain Res. 71;388-398. Kunkel, D. D., and P. A. Schwartzkroin (1988) Ultrastructural characterization and GAD co-localization of somatostatin-like immunoreactive neurons in CAI of the rabbit hippocampus. Synapse 2:371-381. Larsson, L. I. (1981) A novel immunocytochemical model system for specificity and sensitivity screening of antisera against multiple antigens. J. Histochem. Cytochem. 29:408-410. Ljungdahl, A., T. Hokfelt, and G . Nilsson (1978) Distribution of substance P-like immunoreactivity in the central nervous system of the rat. I. Cell bodies and nerve terminals. Neuroscience 3: 861-943. Lotstra, F., S. N. Schiffmann, and 3.-5. Vanderhaeghen (1989) Neuropeptide Y-containing neurons in the human infant hippocampus. Brain Res. 478:211-226. Marksteiner, J., M. Ortler, R. Bellmann, and G . Sperk (1990) Neuropeptide Y biosynthesis is markedly induced in mossy fibers during temporal lobe epilepsy of the rat. Neurosci. Lett. 112: 143-148. Massari, V. J., P. J. Hornby, E. K. Friedman, T. A. Milner, R. A. Gillis, and P. J. Gatti (1990) Distribution of neuropeptide Y-like immunoreactive perikarya and processes in the medulla of the cat. Neurosci. Lett. 115:37-42. Milner, T. A,, and C. E. Bacon (1989a) Ultrastructural localization of somatostatin-like immunoreactivity in the rat dentate gyrus. J , Comp. Neurol. 290544-560. Milner, T. A., and C. E. Bacon (1989b) Ultrastructural localization of tyrosine hydroxylase-like immunoreactivity in the rat hippocampal formation. J. Comp. Neurol. 281:479-495. Milner, T. A., and C. E. Bacon (1989~)GABA-ergic neurons in the rat hippocampal formation: Ultrastructure and synaptic relationships with catecholaminergic terminals. J. Neurosci. 9:3410-3427. Morley, J. E., and J. F. Flood (1990) Neuropeptide Y and memory processing. In J. M. Allen and J. I. Koenig, eds, Central und Peripheral Significance of Neuropeptide Y and l t s Related Peptides, pp. 226-231, New York Academy of Sciences, New York. Paxinos, G., and C. Watson (1987). The Rat Brain in Stereotoxic Coordinates, Edition 2, Academic Press, New York. Peters, A., S . L. Palay, and H. deF. Webster (1991) The Fine Structure of the Nervous System, 3rd ed., Oxford University Press, New York. Pickel, V. M., J. Chan, and V. J . Massari (1989) Neuropeptide Ylike immunoreactivity in neurons of the solitary tract nuclei: Vesicular localization and synaptic input from GABAergic terminals. Brain Res. 476:265-278. Pickel, V. M., J . Chan, and S . R. Sesack (1991) Cellular basis for interactions between catecholaminergic afferents and neurons containing leu-enkephalin like-immunoreactivity in the rat caudate-putamen nuclei. J. Neurosci. Res. (in press). Rall, W. (1966) Distinguishing theoretical synaptic potentials computed for different soma-dendritic distributions of synaptic input. J. Neurophysiol. 30: 1138-1 168. Scherman, D., and G. Boschi (1988) Time required for transmitter accumulation inside monoaminergic storage vesicles differs in peripheral and central nervous system. Neuroscience 27: 1029-1036.

EM OF NEUROPEPTIDE Y IN HIPPOCAMPAL FORMATION / Milner and Veznedaroglu Schousboe, A., J. Drejer, and I. Divac (1980) Regional heterogeneity in astroglial cells: Implications of neuron-glia interactions. Trends Neurosci. 3:XIII-XXIV. Sloviter, R. S. (1989) Chemically defined hippocampal interneurons and their possible relationship to seizure mechanisms. In: The Hippocarnpus: New Vistas V. Chan Palay, and C. Kohler, eds., pp. 443-461, Alan R. Liss, New York. Smith, Y . , A. Parent, L . Kerkerian, and G. Pelletier (1985) Distribution of neuropeptide Y immunoreactivity in the basal forebrain and upper brainstem of the squirrel monkey (Saimiri sciureus). J. Comp. Neurol. 236:71-89. Somogyi, P., A. J. Hodgson, A. D. Smith, M. G. Nunzi, A. Gorio, and J. Y. Wu (1984) Different populations of GABAergic neurons in the visual cortex and hippocampus of the cat contain somatostatin- or cholecystokinin-like immunoreactivity. J. Neurosci. 412-590-2603. Stone, E. A., and M. A. Ariano (1989) Are glial cells targets of the central noradrenergic system? A review of the evidence. Brain Res. Rev. 14:297-309. Thureson-Klein, A,, R. L. Klein, and P. C. Zhu (1986) Exocytosis from large dense-cored vesicles as a mechanism for neuropeptide release in the peripheral and central nervous system. Scand. Electron. Microsc. 179-187.

125

Vallejo, M., D. A. Carter, S . Biswas, and S. L. Lightman (1987) Neuropeptide Y alters monoamine turnover in the rat brain. Neurosci. Lett. 73:155-160. Velley, L., T. A. Milner, J. Chan, S. F. Morrison, and V. M. Pickel (1991) Relationship of met-enkephalin-like immunoreactivity to vagal afferents and motor dendrites in the nucleus of the solitary tract: A light and electron microscopic dual labeling study. Brain Res. 550:298-3 12. Wilcox, B. J., and J. R. Unnerstall (1990) Identification of a subpopuiation of neuropeptide Y-containing locus coeruleus neurons that project to the entorhinal cortex. Synapse 6:284-291. Winkler, H., M. Sietzen, and M. Schober (1987) The life cycle of catecholamine-storing vesicles. In R. G. Johnson, ed, Cellular and Molecular Biology of Hormone- and Neurotransmitter-containing Vesicles, pp. 3-19, New York Academy of Sciences, New York. Wouterlood, F. G., E. Saldana, and M. P. Witter (1990) Projection from the nucleus reuniens thalami to the hippocampal region: Light and electron microscopic tracing study in the rat with the anterograde tracer Phaseolus vulgaris-leucoagglutinin.J. Comp. Neurol. 296: 179-203. Yamano, M, and P. G. M. Luiten (1989) Direct synaptic contacts of medial septa1 efferents with somatostatin immunoreactive neurons in the rat hippocampus. Brain Res. Bull. 22:993-1001.