122

Brain Research, 560 ( 1991) 122-131 © 1991 Elsevier Science Publishers B.V. All rights reserved. (10(36-8993/91/$03.5(t A DONIS (~)0689939116979W

BRES 16979

The influence of long-term potentiation on the spatial relationship between astrocyte processes and potentiated synapses in the dentate gyrus neuropil of rat brain* Jiirgen Wenzel, Grit Lammert, Ursula Meyer and Manfred Krug Institute of Anatomy, Medical Faculty of the Humboldt University, Charitd Hospital, Berlin (E R. G.) and Institute of Pharmacology and Toxicology, Medical A cademy, Magdeburg (F.R. G.) (Accepted 30 April 1991) Key words: Long-term potentiation; Astrocytic glial cell; Synapse; Dentate gyrus; Morphometry

The influence of long-term potentiation (induced by repeated high-frequency stimulation of the perforant pathway) on the distribution pattern of astrocyte processes in the neuropil of the hippocampal dentate area containing the potentiated synapses was investigated by quantitative electronmieroscopy. It has been found that significant changes occurred in the ramification of astrocyte processes as well as in their topographic relation to synaptic complexes. When comparing the results obtained in LTP animals with active control or sham-operated animals, we found significant higher numerical density, but smaller volume, higher surface density and closer apposition of astrocyte processes to the synaptic clefts, boutons terminaux or spines in the potentiated synapses containing neuropil. The glial reaction to synaptic activation has been seen most pronounced 8 h after the LTP induction. The results are pointing to a participation of the glia cells in the maintenance of the LTP effect as well as to a metabolic coupling between synaptic transmission and glia function for equilibrating the homeostasis by clearing the extracellular space next to the transmission zones. INTRODUCTION The p h e n o m e n o n of long-term p o t e n t i a t i o n (LTP) is frequently considered to offer suitable experimental conditions in modeling basic mechanisms involved in plastic changes of synaptic functions or likewise in the processes underlying learning and m e m o r y formation 17.20,35,42,43,59.

60,63.66.

LTP firstly was described in the hippocampus by Bliss and LOmo 8 and Bliss and G a r d n e r - M e d w i n 7. It consists of a rapidly occurring synaptic facilitation induced by high frequency stimulation of monosynaptic neuronal pathways that can persist for days or weeks after stimulation and can be m e a s u r e d as an increase in the amplitude and slope function of the field EPSP as well as the amplitude of the population spike of the monosynaptically e v o k e d field potential 7"18"26. Some evidence has been o b t a i n e d pointing to a certain similarity of LTP with basic mechanisms of learning and m e m o r y formation. F o r instance, LTP can be induced by a small number of tetanic impulses, occurs very rapidly and persists for an extremely long time 4"5s. M o r e o v e r , strong support for this assumption was given by data, which

d e m o n s t r a t e that LTP and m e m o r y formation are similarly accompanied by an e n h a n c e m e n t of protein synthesis36'42 and that both the acquisition and consolidation of a learned b e h a v i o u r and the induction and maintenance of LTP can be influenced by drugs in a similar m a n n e r 21" 23,27,45,50

With respect to LTP, pre- and postsynaptic mechanisms are assumed to be involved in induction and maintenance of the LTP p h e n o m e n o n . Evidence has been accumulated for an enhanced transmitter release 9, changes in postsynaptic receptors t4, induction of protein kinase C 54, changes in the phosphoinositol metabolism 39 and participation of protein or glycoprotein synthesis 21"27"28" 36,50

Especially the latter might be necessary for conformational changes in the growth processes of the postsynaptic spines which has been frequently described 12A7'33"63" 66. In search for the morphological correlate of the LTP effect, attention was focussed mainly on the p o t e n t i a t e d neurons and changes were described in the different parts of neuronal elements. O n the other hand, considering neurons and the surrounding glial cells as a func-

* The authors dedicate this work to our respected teacher in Neuroanatomy, Professor Dr. Walter Kirsche (Director of the Institute of Anatomy of the Humboldt University from 1967 to 1980), on the occasion of his 70th birthday, June 21, 1990. Correspondence: J. Wenzel, Institute of Anatomy, Medical Faculty (Charit6), Humboldt University, Philippstrasse 12, D-1040 Berlin, F.R.G. Fax: (37) (2) 2862035.

123 tional i n t e r c o n n e c t e d

entity,

it m i g h t

also b e

asked

w h e t h e r astrocytes u n d e r g o c h a n g e s d u r i n g t h e i n d u c t i o n and m a i n t e n a n c e of LTP. T h e r e f o r e , t h e o b j e c t i v e o f the c u r r e n t study was to find p h a s e - r e l a t e d r e a c t i o n s o f glial p r o c e s s e s in t h e n e u r o p i l c o n t a i n i n g I o n - t e r m p o t e n t i a t e d synapses as well as to d e m o n s t r a t e t h e f u n c t i o n a l c o o p e r a t i o n o f a s t r o c y t e s with t h e synaptic n e u r o t r a n s m i s s i o n p o t e n t i a t e d by p r e v i o u s h i g h - f r e q u e n c y stimulation. In o r d e r to get statistical significance, m e t h o d s of quantitative electron microscopy were employed. MATERIALS AND METHODS For the experiments 3 months old Wistar rats from our breeding stock were used (body weight 200-300 g).

by repeated tetanization with monophasic square wave impulses of 0.2 ms duration and an intensity that elicited a population spike of 40% of the maximal population spike amplitude. In order to avoid paroxysmal activity, the following stimulation scheme was adapted: 4 tetanizing trains of 300 impulses were given with an interval of 30 min between them. Each train was divided into 20 groups of 15 pulses, the stimulus frequency within the group being 200 Hz and the distance between the groups 5 s. In this way a total number of 1200 pulses was applicated. In 14 animals the time course of the potentiation effect was followed by repeated recordings from the dentate molecular layer upto 7 days after tetanization. (This group was not included in the ultrastructural investigation.) As can be seen in Fig. 1, the LTP effect had been induced and lasted for about 7 days. Animals that served as active controls received the same number of 1200 pulses but continuously with a frequency of only 0.2 Hz. Passive controls were implanted as the other animals but were not stimulated.

Electronmicroscopy

Surgery In order to implant chronic electrodes for stimulation the perforant pathway and recording monosynaptically evoked field potentials in the dentate gyrus, the animals were anesthetized with a mixture of Nembutal (40 mg/kg) and urethane (500 mg/kg) and placed in a David Kopf stereotaxic apparatus (bregma 1 mm above lambda). In animals which were used for electrophysiological experiments, a monopolar recording and a bipolar stimulating electrode made of 125/~m polyurethane-covered stainless steel wire were implanted at coordinates AP 2.8 mm, lat. 1.8 mm and AP 6.9 mm, lat. 4.1 mm 56. The optimal positions of the electrodes were checked by monitoring the field potential during the implantation procedure. In animals which were used for ultrastructural analysis, only the bipolar stimulating electrodes were implanted at coordinates AP 6.9 mm, lat. 4.1 mm, depth 4.0 mm. All electrodes were connected to plastic miniature sockets which were fixed to the skull by acrylic dental cement. After surgery, the animals were housed individually in plastic cages and received water and food ad libitum.

For quantitative electronmicroscopy 13 LTP animals (LTP group), 11 active control animals (AC) and 10 sham-operated animals (passive controls (PC)) were used. We investigated tissue blocks from the hippocampal area dentata (molecular layer) 1 h after tetanization (LTP group, 6 animals; AC, 4 animals; PC, 4 animals), 8 h after tetanization (LTP group, 4 animals; AC, 4 animals; PC, 3 animals), and 48 h after tetanization (LTP group, 3 animals; AC, 3 animals; PC, 3 animals). For this the animals were sacrificed in deep hexobarbital anesthesia by transcardial perfusion with buffered paraformaldehyde (l%)-glutardiaidehyde (l%)-fixation mixture. The brains were removed, the hippocampal area dentata isolated, postfixed in 1% osmium tetroxide and embedded in micropal. The ultrathin sections were taken from the middle third of the molecular layer of the area dentata, stained with uranylacetate and lead citrate. For the observations and photographs a TESLA BS 613 electron microscope was used.

Morphometry

Electrophysiology The stimulation experiments were performed at least 10 days after surgery. In the group of stimulated animals LTP was induced

Quantitative morphological estimations were performed using random photomicrographs of the sections in the middle third of the molecular layer of the dentate area, taken at a primary magnifica-

°/o 300

1 tOO

t

100 -1

~J 15

1.

27 15

ill I 27 15

[121 I I 7 15 3 0

I

2.

3.

4. Tetanizat ion

60

I 90

I At I 120rain 1 8 0

I 240

I

I

I

300rain24 M

I

l

72

Mh

I

Fig. 1. Time course of LTP of the monosynaptically evoked field potential in the dentate gyrus. Top: increase of the population spike (control is 100%). Bottom: increase of the field EPSP. Bars indicate S.E.M.

124

tion of 1(I,(~)0 times, then enlarged to exactly 70,000 times. About 20 micrographs per animal, obtained from sections of 2 or 3 tissue blocks were included in the quantitative evaluation. In each enlarged photomicrograph the test area was indicated.

The test area was 1470 cm 2, corresponding to a neuropil area of 311 um 2 in the ultrathin section. Thereforc, per animal about 61)0/zm 3 neuropil was analysed. All glia cell processes visible in the neuropil within the test area were outlined, Special cmphasis was given

Fig. 2. Electronmicrograph from the neuropil of the area dentata of the hippocampus from a control rat showing the distribution of the glial ccll proccsses (lined). Magnification: x30,000.

125 TABLE I

Influence of LTP on astrocyte processes in the area dentata neuropil Mean -+ standard deviation of mean.

l h

8h

48h

0.0591 +- 0.0169 0.0422 - 0.0197 0.0470 - 0.0172

0.0821 -+ 0.0215 0.0770 -+ 0.0320 0.0780 -+ 0.0093

0.1048 -4- 0.0042 0.0976 -+ 0.0036 0.1006 + 0.0056

0.0156 + 0.0027 0.0136 -+ 0.0044 0.0140 -+ 0.0025

0.0275 -+ 0.0025** 0.0196 -+ 0.0045 0.0192 -+ 0.0040

0.0238 -+ 0.018" 0.0211 -+ 0.0020 0.0215 -+ 0.0027

18.1 -+ 3.9 15.9 -+ 1.9 16.6 -+ 1.3

46.5 - 5.8** 25.8 - 4.3 26.1 - 5.0

27.7 -+ 2.2* 23.1 -+ 2.3 19.3 -+ 2.9

2.63 -+ 0.56 2.96 -+ 0.41 2.79 -+ 0.17

1.10 -+ 0.14"* 1.76 -+ 0.31 1.91 -+ 0.35

1.61 -+ 0.23* 1.89 -+ 0.22 2.17 -+ 0.32

Volume density of glial processes LTP AC S-op

Surface density of glial processes LTP AC S-op

Numerical density of glial processes LTP AC S-op

Volume of a single process (percentage of the volume density) LTP AC S-op

* Significant changes (P < 0.05) in LTP group with time. ** Significance level (P < 0.01) when comparing the 1 h with the 8 h value as well as with the controls. AC, Active controls; S-op, sham-operated rats or passive controls.

to glia processes in close contact to synaptic complexes (presynaptic bouton and postsynaptic spine with visible transmission zone) or to the synaptic cleft. These contact zones were also marked. Using a square lattice test system (calibration distance of test lines a = 6 mm; test point number Pr -- 4290) according to Weibe162 the following parameters were estimated in the test area of each photomicrograph: (1) Volume density V of the astrocyte processes. (2) Numerical density N A of the astrocyte processes. (3) Surface density Sv of the astrocyte processes. (4) Volume density of an averaged single astrocyte process (Vv:

percentage of an averaged single astrocyte process contributing to the volume density of all astrocyte processes in the test area) (Palkovits sl). (5) Surface density of synaptic complexes covered by astrocyte processes. (6) Percentage of synapse surface covered by astrocyte processes. (7) Percentage of synapse with the synaptic cleft bordered by astrocyte processes. For each photomicrograph these parameters were calculated from the estimated primary data (obtained by the test point counting method) by means of the stereological formulas developed by Weibe162 .

TABLE II

Influence of LTP on the synapse--astrocyte processes relationship Mean -+ standard deviation of mean.

l h

8h

48h

17.1 -+ 2.6 13.5 -+ 4.2 14.2 -+ 6.8

26.3 -+ 5.7* 19.4 -+ 4.5 18.8 + 2.3

30.7 -+ 3.5** 24.7 - 2.6 19.7 - 2.0

35.0 - 6.9 25.4 - 9.4 27.7 - 4.1

58.0 - 10.4" 38.2 - 11.3 35.3 -+ 5.0

45.2 -+ 10.7 46.1 +_ 0.7 39.5 -+ 10.7

29.1 - 6.1 37.1 - 8.5 30.0 -4- 3.0

7.3 - 4.0* 16.0 - 4.7 18.8 - 5.9

5.5 -+ 2.0** 12.6 - 4.1 25.0 -4- 4.3

Percentage of synaptic surface density covered by astrocyte processes LTP AC S-op

Percentage of synapses with synaptic cleft bordered by astrocyte processes LTP AC S-op

Percentage of synapses without any contact to astrocyte processes LTP AC S-op

* Significant difference (P < 0.01) of LTP against active control (AC) or sham-operated (S-op) rats. ** Significant difference (P < 0.05) from LTP I h and 48 h.

126 Statistics

For the statistical calculations a personal computer PC 1715 ROBOTRON or the laboratory computer ROBOTRON K 1003 and software (ABSTAT programme) for the significance tests were used. From the collected data, the mean value with its variance was calculated for each animal. Within the animal group the normal distribution of the data was proved. The F-test and the ;(2 test was used to get information about the statistical homogeneity of the animals in each group. For the final significance testing of differences between the experimental and control groups, the mean of each animal was taken as representative for this animal. The significance was tested at the P < 0.05 level using the t-test of Student with Welch-correction for small n numbers. Additionally, all

NUMERICAL DENSITY NA

k

50-

302010. O.

lh

8h

48h

VOLUME DENSITY 300 200.

100, lh

8h

48h

SURFACE DENSITY

Sv'lO~

A

3-

2"

0

lh

~LTP

8h 48h [time] IFJ'~lActiveContro[ ~ S h a m e o p e r n t e d

Fig. 3. Quantitative estimations of astrocyte processes in the neuropil test area (mean and standard deviation of mean). Top: numerical density (NA) of the astrocyte profiles. Significant increase 8 h after the LTP experiment. Middle: volume density (Vv) of a single averaged astrocyte process in relation to the total volume density of all astrocyte profiles. The values • 10-2 correspond to the volume density of a single process in percent of the total glial volume, thus being a parameter of the reduced size of the average profile 8 h after the LTP experiment. Bottom: surface density (Sv) of the astrocyte processes. Significant increase 8 h after the experiment.

significances were again proved by the parameter-frec test of Mann and Whitney. RESULTS The astrocyte processes filling the gaps of the neuropil in the network b e t w e e n dendrites and axons forming synapses, are d e a r l y discernible in electron micrographs by their typical morphological features: irregular contours, light a p p e a r a n c e of cytoplasm with low electron density, sparsely distributed organelles, never microtubules, seldom glial fibrils (Fig. 2). The quantitative estimations revealed a significant influence of the LTP exp e r i m e n t on the stereologic p a r a m e t e r s for astrocyte profiles within the neuropil in general as well as in view of astrocytic contacts with synaptic complexes or profiles bordering the synaptic cleft. Table I summarizes the data characterizing the changes which occurred with the astrocytic profiles in the neuropil after the potentiation experiment. The volume density of the astrocyte processes was not significantly different in the 3 animal groups, thus indicating that the percentage of the glial volume in the whole neuropil was not changed by the potentiation. H o w e v e r , in all groups of animals there was a significant increase in glial volume with time. The reason might be some swelling of the glial processes due to the surgery and electrode implantation in all animal groups. The specific LTP effect was significantly d e m o n s t r a t e d when comparing the data on the numerical density, volume of an averaged single profile, or surface density (Fig. 3, Table I). Most p r o n o u n c e d in the 8 h experiment, we found a significant increase in the n u m b e r of profiles per test area, however, on average the profiles exhibited a smaller volume and, therefore, we got higher values in the surface density of the astrocyte processes. In the 48 h e x p e r i m e n t this effect was less obvious, but still significant. The topographical relationship of astrocyte processes to the synapses potentiated by the stimulation experiment is shown in Figs. 4 - 6 and Table II. In the time course after the LTP effect has been established, we have seen the synaptic complexes (bouton and spine) increasingly envelo p e d by the astrocyte processes until about 30% of the synapse surface is contacted by astrocyte cytoplasm (Figs. 4, 5). There occurs also a highly significant increase in the n u m b e r of synapses with an astrocyte process profile just bordering the synaptic cleft laterally, especially in the phase 8 h after the tetanization (Fig. 6). These changes are reflected by the result that the percentage of synapses without any contact to astrocyte processes diminished from about one third to only 5-6% in the LTP group 48 h after the experiment (Fig. 7, Table ll).

127 DISCUSSION The functions of glial cells in the CNS are by no means completely defined, with the exception of myelin

production by the oligodendrocytes 11 or the u p t a k e of amino acid synaptic transmitters 41'61 and a contribution to extracellular potassium homeostasis by astrocytes 29'32' 40,47-49,53,55

Fig. 4. Electron micrographs of axospinodendritic synapses in the neuropil of the area dentata showing the close relationship of the astrocyte processes to the synaptic cleft or surface of the pre- or postsynaptic elements. Magnification: x60,000.

128

lh offer LTP

~

SYNAPSES and GLIA 6,9'/,

lh after LTP

limited by gtia

29,1*/,

without gliacontact

8hrs after LTP 36,2% surface with gliacontact 25~3%

8 hrs after LT

48hrs after LTP

hired by gtia

without g{iacont¢ 34,7% surface with gtiacontact

7%

/.,8hrs after LTF

[imited by gtia

5,5*/°

Fig. 5. Surface density of synaptic complexes. The relation between astrocyte process-covered surface density (black area) and uncovered surface density (stippled area) is demonstrated 1, 8 or 48 h after the LTP experiment.

without gliacontact

From the data of our morphometric electron microscopic study it can be concluded that not only the target neurons of the tetanized afferent pathway undergo a variety of distinct morphological changes in the pre- and postsynaptic structures after LTP induction 63--66 but also the surrounding glial cells. Evidence has been obtained by our results that the astrocytes near the potentiated synapses develop a higher degree of ramification and an enlargement of its surface. Moreover, astrocyte processes establish a closer spatial contact to the potentiated axospinodendritic synapses and tend to cover the surface

Fig. 7. Numerical relationship between synapses and astrocyte processes. Percentage of all visible synapses in the test area with or without contact to astrocyte processes in the LTP group, 1, 8 or 48 h after the LTP experiment.

49 3% surface with gtiac0nt~ct

of synaptic boutons or spines more extensively as well as to limit the synaptic cleft laterally. As a consequence, the number of synapses without a visible relation to glial

Relationship between Synapses and Glial Processes before LTP

after LTP

°t'o d

60" 504 /,0" 30" 2010" 0

lh

1

LTP

8h ~

ActiveControt

~Shame

operated

Fig, 6. Percentage of synapses in the test area with astrocyte processes bordering the synaptic cleft (mean and S.E.M.). Significant (P < 0.01) increase in the LTP group 8 h after the experiment.

Fig. 8. Fore explanation see text. A, axon; S, spine; G, glial processes.

129 processes diminishes (Fig. 8). Most of these changes occurred highly significantly in the 8 h phase after the LTP stimulation. However, even in the 48 h post-stimulation test we still found these changes in the volume of single astrocyte processes and in the percentage of synapses without astrocyte contact (Tables I and II). These results point firstly to a phaserelated reaction of the astrocytes and secondly to a direct influence of the enhanced synaptic activity and transmission on the glia processes in the neuropil containing potentiated synapses. The intimate relation of the glial processes to the potentiated synapses should reflect the functional cooperation of neurons and glial cells. It is well known that astrocytes are necessary metabolic partners of neurons, especially for the synaptic transmission 6'24'46'55. Glial cells have receptors for neurotransmitters which may be involved in neuron-glia interactions 25'4°'57. The fine ramified cytoplasm of the astrocytes contributes to the inactivation of transmitter molecules released into the synaptic cleft and intercellular space 16. Moreover, an important function is the clearance of the extracellular space from K + ions which are released during synaptic transmission together with the ionic shift of Na +, K + and C a 2+ and whose enhanced concentration can influence the membrane potential (spreading depression) 48'55. So, in order to ensure an undisturbed neuronal transmission, a clearance mechanism for potassium is extremely important. Recently it has been shown that this clearance function can be fulfilled by the astrocyte processes which contact the neurons and synapses at one hand, but also reach the blood or liquor spaces for exchange purposes by the perivascular endfeet or by forming inner and outer limiting glial membranes. Higher extracellular concentrations of K ÷ influence not only neurons but also the membrane potential of the astrocyte processes. High-affinity low-resistance channels for K ÷ in the glial cell membrane has been demonstrated recently by Reichenbach and Eberhardt 53. The K + enriched environment of a permanently activated synapse is supported to be the stimulus for the attachment of glial processes near the transmission zone and the finer ramification of the cell processes in the synaptic neuropil. Recent studies have demonstrated voltage-gated ion channels in glial membranes. Voltage-gated Na + and K + channels, presumably located in the membranes of the astrocytes forming the glia limitans (frog optic nerve), were identified 4°. Marrero et al. 4° reported that nerve impulses in the axons of the frog optic nerve transiently alter the properties of the voltage-dependent membrane channels of the surface glial cells. So a larger arborization of the as-

trocytic processes during LTP could demonstrate a possible form of enhanced signalling between potentiated neurons or synapses and glial cells. During the induction of the LTP effect the stimulation causes an intensification of the synaptic transmission which in turn is followed by an enhanced presynaptic release of glutamate at the perforant path synapses in the dentate area 9'14 together with transmembraneous shifts of the ions. Astrocytes are also known to take up and metabolize several neurotransmitters such as glutamate, glycin, G A B A or other biogenic amines from the extracellular space to prevent a desensitization of synaptic receptors 24,46,48,61" For the establishment and maintenance of the potentiation effect, calcium ions play an important role 4'5'13' 14,38. Glial cells may also be involved in the control of the extracellular calcium concentration 4°. Therefore, it might be concluded that the changes seen in astrocytes after LTP induction reflect the reaction of the system neuron-glia to an enhancement of synaptic transmission processes which tend to ensure the homeostasis in the extracellular space of ions and transmitter molecules. The extensive ramification in the neuropil as well as the intimate relationship to the synaptic cleft enable the astrocytes for these regulatory tasks in brain homeostasis. The glial cells may also participate in the neuronal protein metabolism 31 that is highly activated during the LTP 15"19'22'36'43'52'63'66 by a transfer of synthesized molecules. These results indicate that the stimulated neuronal activation and enhanced metabolism with its intensive material exchange between intra- and extracellular compartments force the glial cells to react likewise. Morphologically, these processes can be reflected in delicate ramification of more numerous but smaller glia cell processes, enlarging their total surface as well as the contact area with the potentiated synapses. However, it remains unclear if the glia reaction occurs specifically only next to tetanized synapses or if it represents a generalized effect on the affected neuropil. At the one hand, we know from the experiments of other researchers that only the synchronous co-activation of a relatively small number of synapses (about 5%) is necessary to elicit the initiation or induction of the electrical potentiation effect 3'37"44, referred to by Levy and Steward 34 and Desmond and Levy 16 as the associative potentiation. Otherwise, the high-frequency stimulation initiates not only the synaptic potentiation of the conditioned pathway, but leads also to a long-term synaptic depression of unconditioned p a t h w a y s 1"2'1°'34'37. W h i t e et al. 67 reported recently that associative events that lead to LTP or long-term depression (LTD) can be restricted to a local dendritic domain. By the methods used in our investigation a morpho-

130 logical d i f f e r e n t i a t i o n

between

activated

synapses

in-

logical i n t e r a c t i o n with the LTP induction,

especially

v o l v e d h o m o s y n a p t i c a l l y in LTP o r h e t e r o s y n a p t i c a l l y in

with the i n d u c t i o n of the N M D A r e c e p t o r - m e d i a t e d LTP

L T D is impossible within the total p o p u l a t i o n of syn-

by application of an N M D A inhibitor.

apses n e i g h b o u r e d by glial processes. T h e r e f o r e , future investigations should be p e r f o r m e d to a n s w e r the question if the glia c h a n g e s o b s e r v e d r e p r e s e n t a g e n e r a l or specific effect of the stimulation and synaptic activation. Two aspects are to be c o n s i d e r e d in such future research: (1) labelling of the s t i m u l a t e d fibres by m a r k e r s and investigation of distinct p o t e n t i a t e d synapses as well as their glia e n v i r o n m e n t ; and (2) functional or p h a r m a c o -

REFERENCES 1 Abraham, W.C. and Goddard, G.V., Asymmetric relationship between homosynaptic long-term potentiation and heterosynaptic long-term depression, Nature, 305 (1983) 717-719. 2 Abraham, W.C., Bliss, T.V.P. and Goddard, G.V., Heterosynaptic changes accompany long-term but not short-term potentiation of the perforant path in the anaesthetized rat, J. Physiol. (Lond)., 363 (1985) 335-349. 3 Andersen, P., Sundberg, S.H., Sveen, O. and Wigstri3m, H., Specific long-lasting potentiation of synaptic transmission in hippocampal slices, Nature, 226 (1977) 736-373. 4 Andersen, P. and Hvalby, O., Long-term potentiation. Problems and possible mechanisms. In R.L., Isaacson and K.H. Pribram (Eds.), The Hippocampus, VoL 3, Plenum, New York, 1986, pp. 169-187. 5 Baimbridge, K.G. and Miller, J.J. Calcium uptake and retention during long-term potentiation of neuronal activity in the rat slice preparation, Brain Research, 221 (1981) 299-305. 6 Berl, S., Nicklas, W.J. and Clarke, D.D., Glial cells and metabolic compartmentation. In E. Schoffeniels, G. Frank, D.B. Towers and L. Hertz (Eds.), Dynamic Properties of Glia Cells, Pergamon, Oxford, 1978, pp. 143-149. 7 Bliss, T.V.P. and Gardner-Medwin, A.R., Long-lasting potentiation of synaptic transmission in the dentate area of the unanaesthetized rabbit following stimulation of the perforant pathway, J. Physiol. (Lond.), 232 (1973) 357-374. 8 Bliss, T.V.P. and Lomo, T., Long-lasting potentiation of synaptic transmission in the dentate area of the anesthetized rabbit following stimulation of the perforant path, J. Physiol. (Lond.), 232 (1973) 331-356. 9 Bliss, T.V.P., Douglas, R.M., Errington, M.L. and Lynch, M.A., Correlation between long-term potentiation and release of endogenous amino acids from dentate gyrus of anaesthetized rats, J. Physiol. (Lond.), 377 (1986) 391-408. 10 Bradler, J.E. and Barrionuevo, G., Heterosynaptic correlates of long-term potentiation induction in hippocampal CA3 neurons, Neuroscience, 35 (1990) 265-271. 11 Bunge, R.P., Cellular and non-cellular influences on myelinforming cells. In T.A. Sears (Ed.), Neuronal-Glial Cell Interrelationships, Springer Verlag, Berlin, Heidelberg, New York, 1982, pp. 115-130. 12 Chang, F.L. and Greenough, W.T., Transient and enduring morphological correlates of synaptic activity and efficacy change in the rat hippocampal slice, Brain Research, 309 (1984) 34-46. 13 Collingridge, G.L., Long-term potentiation in the hippocampus: mechanisms of initiation and modulation by neurotransmitters, Trends Pharmacol. Sci., 6 (1985) 407-411. 14 Collingridge, G.L. and Bliss, T.V.P., NMDA-receptors - their role in long-term potentiation, Trends Neurosci., 10 (1987) 288293. 15 De Graan, P.N.E., Lopes da Silva, F.H. and Gispen, W.H., The role of hippocampal phosphoproteins in long-term potentiation.

Acknowledgements.This work was supported by the Ministry for Science and Technology of the former G.D.R. The authors are indebted to Mrs. Maria Wagner and Mrs. Gisela Duwe for skiUfull help in the laboratory work, to Mrs. Ursula Ziihlke for the photographic enlargements, and to Mrs. Rosemarie Leder for the graphs. Parts of this study were published in the thesis of Grit Lammert 3°.

In R.L. Isaacson and K.H. Pribam (Eds.), The Hippocampuz', Vol.3, Plenum, New York, 1986, 187-216. 16 Desmond, N.L. and Levy, W.B., Synaptic correlates of associative potentiation/depression: an ultrastructural study in the hippocampus, Brain Research, 265 (1983) 21-30. 17 Desmond, N.L. and Levy, W.B., Anatomy of associative longterm synaptic modification. In P.W. Landfield and S.A. Deadwyler (Eds.), Alan R. Liss, New York, 1988, pp. 265-305. 18 Douglas, R.M. and Goddard, G.V., Long-term potentiation of the perforant path-granule cell synapse in the rat hippocampus, Brain Research, 86 (1975) 205-215. 19 Duffy, Ch., Teyler, T.J. and Shashoua, V.E., Long-term potentiation in the hippocampal slice: evidence for stimulated secretion of newly synthesized proteins, Science, 212 (1981) 11481151. 20 Eccles, J.C., Calcium in long-term potentiation as a model for memory, Neuroscience, 10 (1983) 1071-1081. 21 Frey, U., Krug, M., Reymann, K. and Matthies, H., Anisomycin, an inhibitor of protein synthesis, blocks late phases of homo- and heterosynaptic long-term potentiation in the hippocampal CAl-region in vitro, Brain Research, 452 (1988) 57-65. 22 Gispen, W.H. and Routtenberg, A., Phosphoproteins in neuronal function, Prog. Brain Res., 69 (1986) 211-234. 23 Grecksch, G., Ott, T. and Matthies, H., The effect of intrahippocampally applied anisomycin on the retention of brightness discrimination in rats, Behav. Neurol. Biol., 29 (1980) 281-289. 24 Hamberger, A., Cotman, C.W., Sellstrom, A. and Weiler, T., Glial cells and their relationship to transmitter glutamate. In E. Schoffeniels, G. Frank, D.B. Towers and L. Hertz (Eds.), Dynamic Properties of Glial Cells, Pergamon, Oxford, 1978, pp. 163-172. 25 Kimelberg, H.K. (Ed.), Glial Cell Receptors, Raven Press, New York, 1988. 26 Krug, M., BriSdemann, R. and Ott, T., Blockade of long-term potentiation in the dentate gyrus of freely moving rats by the glutamic acid and antagonists GDEE, Brain Research, 249 (1982) 57-62. 27 Krug, M., LOssner, B. and Ott, T., Anisomycin blocks the late phase of long-term potentiation in the dentate gyrus of freely moving rat, Brain Res. Bull., 13 (1984) 39-42. 28 Krug, M., Jork, R., Reymann, K., Wagner, M. and Matthies, H., The amnesic substance 2-deoxygalatose suppresses the maintenance of hippocampal LTP, Brain Research, (1991) in press. 29 Kuffler, S.W., Neuroglial cells: physiological properties and potassium-mediated effect of neuronal activity on the glial membrane potential, Proc. R. Soc. Lond., Ser. B. 168 (1967) 1-21. 30 Lammert, G., Ultrastrukturelle Ver~inderungen protoplasmatischer Astrocyten in der Formatio hippocampi (mittleres Drittel des Stratum moleculare der Fascia dentata) nach Langzeitpotenzierung -- eine Studie zum dynamischen Charakter der neuronalen und gliazellul~iren Plastizit~tt, Dissertation, Berlin, 1990. 31 Lasek, R.I., Gainer, H. and Barker, J.L., Cell-to-cell transfer of glial proteins to the squid giant axon, J. Cell Biol., 74 (1977)

131 501-523. 32 Latzkovits, L., Neuronal-glial interactions in potassium transport. In E. Sehoffeniels, G. Frank, D.B. Towers and L. Herts (Eds.), Dynamic Properties of Glia Cells, Pergamon, Oxford, 1978, pp. 327-336. 33 Lee, K.S., Schottler, F., Oliver, M. and Lynch, G., Brief bursts of high-frequency stimulation produce two types of structural changes in rat hippocampus, J. Neurophysiol., 44 (1980) 247258. 34 Levy, W.B. and Steward, O., Synapses as associative memory elements in the hippocampal formation, Brain Research, 175 (1979) 233-245. 35 Lopes da Silva, EH., B~ir, P.R., Tielen, A.M. and Gispen, W.H., Plasticity in synaptic transmission and changes of membrane-bound protein phosphorylation, Prog. Brain Res., 55 (1982) 369-377. 36 Li~ssner, B., Schweigert, Chr., Pchalek, V., Krug, M., Frey, S. and Matthies, H., Training and LTP-induced changes of protein synthesis in rat hippocampus, Neuroscience, Suppl. 22 (1987) 512. 37 Lynch, G.S., Dunwiddie, T. and Gribkoff, V., Heterosynaptic depression: a postsynaptic correlate of long-term potentiation, Nature, 266 (1977) 736-737. 38 Lynch, G., Halpain, S. and Baudry, M., Structural and biochemical effects of high-frequency stimulation in the hippocampus. In W. Seifert (Ed.), Neurobiology of the Hippocampus, Academic Press, 1983, pp. 251-264. 39 Lynch, M.A., Clements, M.P., Errington, M.L. and Bliss, T.V.P., Increased hydrolysis of phosphatidyl-inositol-4,5bisphosphate in long-term potentiation, Neurosci. Lett., 84 (1988) 291-296. 40 Marrero, H., Astion, M.L., Coles, I.A. and Orkand, R.K., Facilitation of voltage-gated ion channels in frog neuroglia by nerve impulses, Nature, 339 (1989) 378-380. 41 Martinez-Hernandez, A., Bell, K.P. and Norenberg, M.D., Glutamine synthetase: glial localization in brain, Science, 195 (1977) 1356-1358. 42 Matthies, H., Neurobiological aspects of learning and memory, Annu. Rev. Psychol., 40 (1989) 381-404. 43 Matthies, H., In search of the cellular mechanisms of memory, Prog. Neurobiol., 32 (1989) 277-349. 44 McNaughton, B.L., Douglas, R.M. and Goddard, G.V., Synaptic enhancement in fascia dentata: eooperativity among coactive afferents, Brain Research, 157 (1978) 277-293. 45 Morris, R.G.M., Synaptic plasticity and learning: selective impairment of learning in rats by blockade of long-term potentiation in vivo by the N-methyl-D-aspartate receptor antagonist AP5, J. Neurosci., 9 (1989) 3040-3057. 46 Murphy, S. and Pearce, B., Functional receptors for neurotransmitters on astroglial cells, Neuroscience, 22 (1987) 381-394. 47 Odette, L.L. and Neuman, E.A., Model of potassium dynamics in the central nervous system, Glia, 1 (1988) 198-210. 48 Orkand, R.K., Signalling between neuronal and glia cells. In T.A. Sears (Ed.), Neuronal-Glial Cell Interrelationships, Springer Verlag, Berlin, Heidelberg, New York, 1982, pp. 147158. 49 Orkand, R.K., Glial-interstitial fluid exchange, Ann. N. Y. Acad. Sci., 481 (1986) 269-272. 50 Otani, S., Marshall, C.J., Tate, W.P., Goddard, G.V. and Abraham, W.C., Maintenance of long-term potentiation in rat dentate gyrus requires protein synthesis but not messenger RNA synthesis immediately post-tetanization, Neuroscience, 28 (1989) 519-526. 51 Palkovits, M., Determination of axon terminal density in the central nervous system, Brain Research, 108 (1976) 413--417. 52 Pohle, W., Acosta, L., Riithrich, H., Krug, M. and Matthies,

H., Incorporation of [3H]fucose in rat hippocampal structures after conditioning by perforant path stimulation and after LTPproducing tetanization, Brain Research, 410 (1987) 245-256. 53 Reichenbach, A. and Eberhardt, W., Cytotopographical specialization of enzymatically isolated rabbit retinal Miiller (glial) cells: K + conductivity of the cell membrane, Glia, 1 (1988) 191197. 54 Reyman, K.G., Frey, U., Jork, R. and Matthies, H., Polymyxin B, an inhibitor of protein kinase C, prevents the maintenance of synaptic long-term potentiation in hippocampal CA1 neurons, Brain Research, 440 (1988) 305-314. 55 Roitbak, A.I., Neuroglia: Eigenschaften, Funktionen, Bedeutung -- Brain and Behaviour Research -- Monograph Series, Vol. 10, Gustav Fischer, Verlag, Jena, 1983. 56 Skinner, J.E., Neuroscience, a Laboratory Manual, Saunders, Philadelphia, PA, 1971. 57 Stone, E.A. and Ariano, M.A., Are glial cells targets of the central noradrenergic system? A review of the evidence, Brain Res. Rev., 14 (1989) 297-309. 58 Swanson, L.W., Teyler, T.J. and Thompson, R.F. (Eds.), Hippocampal long-term potentiation: mechanisms and implications for memory. Neurosciences Research Program Bulletin, London, Vol. 20, MIT Press, Cambridge, MA, London, 1982. 59 Teyler, T.J. and Discenna, P., Long-term potentiation as a candidate mnemonic device, Brain Res. Rev., 7 (1984) 15-28. 60 Teyler, T.J. and Discenna, P., Long-term potentiation, Annu. Rev. Neurosci., 10 (1987) 131-161. 61 Walicke, P.A., Brown, D.A., Bunge, R.P., Davison, A.N., Droz, B., Fambrough, D.M., Fischer, G., Griffin, I.W., Mirsky, R, Nicholls, I.G., Orkland, R.K., Rohrer, H. and Treherne, I.E., Maintenance. State of the report. In T.A. Sears (Ed.), Neuronal --Glial Cell Interrelationships, Springer Verlag, Berlin, Heidelberg, New York, 1982, pp. 169-202. 62 Weibel, E.R., Stereological Methods. Vol. 1: Practical Methods for Biological Morphometry. Academic Press, London, 1979, pp. 26-37. 63 Wenzel, J. and Matthies, H., Morphological changes in the hippocampal formation accompanying memory formation and longterm potentiation. In N.M. Weinberger, I.L. Mc Gaugh and G. Lynch (Eds.), Memory Systems of the Brain: Animal and Human Cognitive Processes, Guilford Publications, New York, 1985, pp. 150-170. 64 Wenzel, J., Krug, M., Ott, T. and Schuster, Th., Ultrastructural changes in neurons and synapses of the dentate area accompanying long-term potentiation induced by perforant path stimulation. In H. Matthies (Ed.), Learning and Memory. Mechanisms of Information Storage in the Nervous System --Proceedings Vllth International Neurobiology Symposium, Magdeburg, 1985, Pergamon Press, Adv. Biosci. 59 (1986) 109114. 65 Wenzel, J., Grosser, P., Krug, M. and Schuster, Th., Long-term potentiation --induced ultrastructural changes in neurons and synapses of the fascia dentata. 2nd Worm Congress of Neuroscience (IBRD), Budapest 1987 Neuroscience, Suppl. 22 (1987) 128. 66 Wenzel, J. Grosser, P., Gruew, L., Lammert, G., Meyer, U., Plaschke, M., Schmidt, K., Schuster, Th., Urbach, V., Wenzel, M., Krug, M. and Matthies, H., Morphological correlates of synaptic plasticity. In J. Wenzel and J. Del Rio (Eds.), Development and Plasticity of Brain Structures and Functions --Ergeb. Exp. Med., Vol. 51, Verlag Volk und Gesundheit, Berlin, 1990, pp. 109-133. 67 White, G., Levy, W.B., Steward, O., Evidence that associative interactions between synapses during the induction of long-term potentiation occur within local dendritic domains, Proc. Natl. Acad. Sci. U.S.A., 85 (1988) 2368-2372.