- Email: [email protected]
Optical recording of rat entorhino-hippocampal system in organotypic culture
ELSEVIER
Neuroscience Letters 216 (1996) 211-213
NEUROSCI[NC[ lETTERS
Optical recording of rat entorhino-hippocampal system in organotypic culture Y a s u h i r o N a k a g a m i , H i r o s h i Saito, N o r i o M a t s u k i * Department of Chemical Pharmacology, Faculty of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan
Received 6 July 1996; revised version received 9 August 1996; accepted 15 August 1996
Abstract
It is difficult to comprehend the entorhino-hippocampal information processing using acute transverse hippocampal slice, because the dorsally inclined connections of the entorhino-hippocampal projections can be damaged easily. Therefore, we investigated the spatialtemporal propagation in organotypic cultures of the hippocampus attaching to the entorhinal cortex using a real-time optical recording system with a voltage-sensitive dye and suitability as an in vitro model. Real-time imaging demonstrated that the stimulation of the perforant pathway induced excitatory propagation in trisynaptic pathway of the hippocampus and sequentially in the layer V from the medial to the lateral entorhinal cortex. The horizontal propagation from the lateral to the medial site was also seen after the stimulation of the lateral entorhinal cortex. The analysis of the entorhino-hippocampal organotypic culture would contribute to understanding of the mechanism of learning and memory. Keywords: Hippocampus; Entorhinal cortex; Organotypic; Optical recording; Voltage-sensitive dye
The hippocampus is a cortical structure that has been shown to relate to learning and memory. The entorhinohippocampal fiber is the main input to the hippocampus, and the dentate gyrus (DG) is the main target of the projections. The output of DG induces serial excitatory propagation in CA3 and CA1, and this intrinsic hippocampal circuit is called the trisynaptic pathway [1]. Axons of CA1 pyramidal cells project into the stratum oriens or to the alveus and bend sharply toward the subiculum. Finally, the output of the subiculum terminates in the entorhinal cortex. Although the intrinsic circuit of the entorhino-hippocampal system contributes to hippocampal information processing, only few electrophysiological studies concerning the system have been conducted. The main reason is probably that traditional electrophysiological technique is generally restricted to the analysis of information from a few electrodes placed on the tissue. Furthermore, the segment of the entorhinal cortex and the hippocampus contained in the individual transverse slices are separated from each other, because the entorhino-hippocampal fiber projections incline dorsally [16]. As a result, the connections between the two segments would be cut when * Corresponding author. Tel.: +81 3 38122111, ext. 4782; fax: +81 3 38154603.
making a slice. For the same reason, there would be no projections from CA1 to the entorhinal cortex. To overcome these problems, the analysis of the entorhino-hippocampal organotypic cultures by optical recording is a good strategy. The optical recording system has been recognized as a suitable method to analyze synaptic connections, neuronal circuit and spatial-temporal pattern of neuronal activity. The system which was employed in this study achieves high time resolution (0.6 ms) and has 128 x 128 photodiodes [6]. We applied the system to the entorhino-hippocampal organotypic culture to investigate the suitability as an in vitro model for information processing in the hippocampus. Organotypic cultures of the hippocampus were processed according to the method of Stoppini et al. [9]. The preparation cultured for 10-14 days was employed for optical recording. A slice was stained for 10 min in 0.2 mg/ml RH155 [2,4,15] solution dissolved in artificial cerebrospinal fluid (ACSF) having the following composition (in mM): NaC1 127, KC1 1.6, KH2PO4 1.24, MgSO4 1.3, CaCI2 2.4, NaHCO3 26 and glucose 10, which was continuously bubbled with a mixture of 95% 02 and 5% CO2 and then washed in ACSF at least for 15 min. The optical recording system (HR Deltaron 1700, Fuji
0304-3940/96/$12.00 © 1996 Elsevier Science Ireland Ltd. All rights reserved PII S0304-3940(96) 13001-9
212
Y. Nakagami et al. / Neuroscience Letters 216 (1996) 211-213
Photo Film, Japan)consisted of a 128 × 128 photodiode array and a processing unit [6]. The system enabled us to investigate electrical activities of neurons and its usefulness has been already demonstrated in the piriform cortex [10], visual cortex [11] and neurohypophysis [7]. Transmitted light with a wavelength of 700 + 20 nm was projected. The duration of light exposure was minimized ( 1 s) by a mechanical shutter to avoid photodynamic damage and dye bleaching. Before stimulation, a 128 frame averaged image of background light signal was stored in a reference memory. The stored reference data were continuously subtracted from real-time images and transferred to the unit sequentially at a frame rate of 0.6 ms. One trial consisted of 512 real-time images, and 16 trial images were averaged to improve the signal-to-noise ratio. Then, a slice was transferred to a recording chamber and perfused at a rate of 2.0 ml/min with ACSF at room temperature (25-27°C). Conventional electrophysiological technique of extracellular recordings was also employed to check the responses of the slice and for the adjustment of the stimulus strength during optical recording. A rectangular pulse (0.5-0.7 mA) of 100 t~s duration was delivered every 15 s with a strength that gave the maximum amplitude of field population spike. Fig. 1A indicates a 12 frame sequence of signal propagation, recorded after the stimulation of the perforant pathway. About 3.0 ms after the stimulus delivery, weak excitatory signal was detected in DG. Then, the activity invaded into CA3 (6.0 ms) and propagated to CA1 (12.0 ms). Subsequently, the excitatory propagation extended into both of the presubiculum and the parasubiculum, and into layer V. The activity in layer V probably consists of both of the activity from CA1 and from the presubiculure and the parasubiculum. Finally, the horizontal propagation from the medial entorhinal cortex to the lateral entorhinal cortex in layer V was seen, followed by the vertical propagation to layer II. This type of propagation was observed in 25 slices out of 28. The time course of the optical signals recorded from several photodiodes is shown in Fig. lB. The latency of the signals was short in DG, CA3, CA1 and layer V in ascending order. This result demonstrates that the cytoarchitectonic network including entorhino-hippocampal inputs and outputs was highly retained in the entorhino-hippocampal organotypic culture. In addition, we investigated whether the same neuronal propagation was observed in the acute entorhinohippocampal slice. The excitatory propagation of the hippocampal trisynaptic pathway and the long-lasting excitation in CA3 were observed after the stimulation of the perforant pathway (Nakagami et al., submitted), but the propagation of CA1 activity to the entorhinal cortex was not observed in the acute slices (n = 49). It is suggested that regeneration and sprouting of neurons in organotypic slices restored the neural connections. Another different excitatory propagation was shown in Fig. 2. The stimulation of the layer V in the lateral entorh-
E
3.Ores
6.0ms
.:...
9.0ms
12.0 ms
15.0 ms
18.0 ms
21.0 m s
24.0 ms
36.0 ms
4 8 0 ms
60.0 m s
Ores
V
CA1
1
i 0.03 %1 L _ _ 50 ms
Fig. 1. (A) Optical signal propagation from entorhino-hippocampal organotypic slice after the stimulation of the perforant pathway. Numerals under the images correspond to time after the stimulus delivery. The fractional absorbance change was monochrome-coded, as shown by the pseudo-monochrome scale in the bottom right comer. Dashed lines in 0 ms image indicate the approximate outline of pyramidal and granule cell layers of the hippocampus, and layers of the entorhinal cortex. An asterisk indicates the stimulus site. (B) Time course of the optical signals after the stimulation of the perforant pathway. The recording positions of the signals correspond to 0 ms image in (A). Stimulus delivery is indicated as an arrowhead.
inal cortex induced the activity along layer V to the medial site. The conduction velocity of the horizontal propagation from lateral to medial was about 70% of the reversed propagation. Then the activity reached to the presubiculum and the parasubiculum and finally to layer II. The application of 10 tzM 6-cyano-7-nitroquinoxaline (CNQX), a nonN-methyl-D-aspartate receptor antagonist, blocked the propagation of the optical signals, indicating that the optical signals were derived from the postsynaptic response, not just from action potentials of stimulated axons (data not shown). An immediate question arising may be whether signal propagation observed in the present study using organotypic culture reflects innervation manifested in vivo. However, there are many reports showing that the organotypic culture retains many characteristics of its normal cytoarch-
Y. Nakagami et al. /Neuroscience Letters 216 (1996) 211-213
, , I!
"
r'l
0 ms
6.0 ms
12.0 m s
24.0 ms
30.0 ms
36.0 m s
42.0 m s
48.0 ms
54.0 m s
60.0 ms
66.0 ms
18.0 m s
Fig. 2. Optical signal propagation from entorhino-hippocampalorganotypic slice after the stimulationof the lateral entorhinalcortex. An asterisk indicatesthe stimulus site. itectonic organization, and excitatory and inhibitory synapses function in the same way as confirmed in vivo [3,12]. It is also reported that the fiber projection between the entorhinal cortex and the hippocampus selectively regenerated in both directions and showed target specificity in the static culture of large horizontal slices [17]. As shown in Fig. 1A, the present studies have demonstrated that the entorhino-hippocampal organotypic culture retained the neuronal pattern of the excitatory propagation after the stimulation of the perforant pathway by optical recording. Histochemical studies support that the pattern is reasonable. The projection to DG arises mainly from layer II of the entorhinal cortex [8] and the activity in granule ceils in DG induce excitation in the well-known trisynaptic pathway of the hippocampus [ 1]. The CA1 pyramidal neuron projects to both the presubiculum and the parasubiculum, and also terminates mainly in layer V [13]. The excitatory propagation from the lateral to the medial entorhinal cortex was demonstrated in Fig. 2. The result also corresponded well to the histochemical studies. Axons from cells in layer Va of the entorhinal cortex terminate sparsely in layer I of the presubiculum and the parasubiculum [ 14]. The projections from the presubiculum and the parasubiculum terminate selectively in layers III and II of the entorhinal cortex, respectively [ 13,14]. Within layer II, the stellate cells project to an extensive collaterals plexus within layers I and II and to a lesser extent in layer III [5]. Furthermore, the pyramidal cells in layer III project to CA1 and the subiculum, and their collaterals are predominantly distributed to layers III and I. The intrinsic entorhinal circuit highly and systematically regulates hippocampal inputs and outputs in vivo. In conclusion, we have firstly demonstrated that the entorhino-hippocampal organotypic culture retained cytoarchitectonic organization manifested in vivo by optical recording. The analysis of acute hippocampal slice by electrophysiological technique is not probably sufficient to
213
comprehend entorhino-hippocampal neuronal network. Although further detailed studies are certainly needed, we think the analysis of the entorhino-hippocampal organotypic culture would be a suitable in vitro model for information processing in the entorhino-hippocampus to investigate the mechanism of learning and memory. [1] Amaral, D.G. and Witter, M.P., The three-dimensionalorganization of the hippocampal formation: a review of anatomical data, Neuroscience, 31 (1989) 571-591. [2] Cinelli, A.R. and Salzberg, B.M., Dendriticorigin of late events in optical recordings from salamander olfactory bulb, J. Neurophysiol., 68 (1992) 786-806. [3] Frotscher, M. and Heimrich, B., Lamina-specificsynapticconnections of hippocampal neurons in vitro, J. Neurobiol., 26 (1995) 350-359. [4] Konnerth,A., Obaid,A.L. and Salzberg,B.M., Opticalrecordingof electrical activity from parallel fibres and other cell types in skate cerebellar slices in vitro, J. Physiol., 393 (1987) 681-702. [5] Lingenhohl,K. and Finch,D.M., Morphologicalcharacterizationof rat entorhinalneuronsin vivo: soma-dendriticstructure and axonal domains, Exp. Brain Res., 84 (1991) 57-74. [6] Matsumoto, G. and Ichikawa, M., Optical system for real-time imaging of electrical activity with a 128 x 128 photopixel array, Soc. Neurosci. Abstr., 16 (1990) 490. [7] Obaid, A.L. and Salzberg, B.M., Micromolar 4-aminopyridine enhances invasion of a vertebrate neurosecretory terminal arborization, J. Gen. Physiol., 107 (1996) 353-368. [8] Ruth, R.E., Collier, T.J. and Routtenberg,A., Topographicalrelationship between the entorhinalcortex and the septotemporal axis of the dentate gyrus in rats: II. Cells projectingfrom lateral entorhinal subdivisions,J. Comp. Neurol., 270 (1988) 506-516. [9] Stoppini, L., Buchs, P.-A. and Muller, D., A simple method for organotypic cultures of nervous tissue, J. Neurosci. Methods, 37 (1991) 173-182. [10] Sugitani,M., Sugai, T., Tanifuji,M. and Onoda, N., Signal propagation from piriform cortex to the endopiriformnucleus in vitro revealed by optical imaging,Neurosci. Lett., 171 (1994) 175-178. [11] Tanifuji,M., Sugiyama,T. and Murase, K., Horizontalpropagation of excitation in rat visual cortical slices revealed by optical imaging, Science, 266 (1994) 1057-1059. [12] Torp, R., Hang, F.M., Tender, N., Zimmer, J. and Ottersen, O.P., Neuroactive amino acids in organotypic slice cultures of the rat hippocampus: an immunocytochemicalstudy of the distribution of GABA, glutamate, glutamine and taurine, Neuroscience, 46 (1992) 807-823. [13] Van Groen, T. and Wyss, J.M., Extrinsic projections from area CA1 of the rat hippocampus: olfactory, cortical, subcortical and bilateral hippocampal formation projections, J. Comp. Neurol., 302 (1990) 515-528. [14] Van Groen, T. and Wyss, J.M., The postsubicularcortex in the rat: characterization of the subicularcortex and its connections,Brain Res., 529 (1990) 165-177. [15] Vranesic,I., Iijima, T., Ichikawa, M., Matsumoto, G. and Knrpfel, T., Signal transmissionin the parallel fiber - Purkinjecell system visualizedby high-resolutionimaging,Proc. Natl. Acad. Sci. USA, 91 (1994) 13014-13017. [16] Witter,M.P., Organizationof the entorhinal-hippocampalsystem: a review of current anatomicaldata, Hippocampus,3 (Suppl.)(1993) 33-44. [17] Woodhanas,P.L., Atkinson,D.J. and Raisman,G., Rapid declinein the ability of entorhinalaxons to innervatethe dentate gyrus with increasing time in organotypic co-culture, Eur. J. Neurosci., 5 (1993) 1596-1609.
Neuroscience Letters 216 (1996) 211-213
NEUROSCI[NC[ lETTERS
Optical recording of rat entorhino-hippocampal system in organotypic culture Y a s u h i r o N a k a g a m i , H i r o s h i Saito, N o r i o M a t s u k i * Department of Chemical Pharmacology, Faculty of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan
Received 6 July 1996; revised version received 9 August 1996; accepted 15 August 1996
Abstract
It is difficult to comprehend the entorhino-hippocampal information processing using acute transverse hippocampal slice, because the dorsally inclined connections of the entorhino-hippocampal projections can be damaged easily. Therefore, we investigated the spatialtemporal propagation in organotypic cultures of the hippocampus attaching to the entorhinal cortex using a real-time optical recording system with a voltage-sensitive dye and suitability as an in vitro model. Real-time imaging demonstrated that the stimulation of the perforant pathway induced excitatory propagation in trisynaptic pathway of the hippocampus and sequentially in the layer V from the medial to the lateral entorhinal cortex. The horizontal propagation from the lateral to the medial site was also seen after the stimulation of the lateral entorhinal cortex. The analysis of the entorhino-hippocampal organotypic culture would contribute to understanding of the mechanism of learning and memory. Keywords: Hippocampus; Entorhinal cortex; Organotypic; Optical recording; Voltage-sensitive dye
The hippocampus is a cortical structure that has been shown to relate to learning and memory. The entorhinohippocampal fiber is the main input to the hippocampus, and the dentate gyrus (DG) is the main target of the projections. The output of DG induces serial excitatory propagation in CA3 and CA1, and this intrinsic hippocampal circuit is called the trisynaptic pathway [1]. Axons of CA1 pyramidal cells project into the stratum oriens or to the alveus and bend sharply toward the subiculum. Finally, the output of the subiculum terminates in the entorhinal cortex. Although the intrinsic circuit of the entorhino-hippocampal system contributes to hippocampal information processing, only few electrophysiological studies concerning the system have been conducted. The main reason is probably that traditional electrophysiological technique is generally restricted to the analysis of information from a few electrodes placed on the tissue. Furthermore, the segment of the entorhinal cortex and the hippocampus contained in the individual transverse slices are separated from each other, because the entorhino-hippocampal fiber projections incline dorsally [16]. As a result, the connections between the two segments would be cut when * Corresponding author. Tel.: +81 3 38122111, ext. 4782; fax: +81 3 38154603.
making a slice. For the same reason, there would be no projections from CA1 to the entorhinal cortex. To overcome these problems, the analysis of the entorhino-hippocampal organotypic cultures by optical recording is a good strategy. The optical recording system has been recognized as a suitable method to analyze synaptic connections, neuronal circuit and spatial-temporal pattern of neuronal activity. The system which was employed in this study achieves high time resolution (0.6 ms) and has 128 x 128 photodiodes [6]. We applied the system to the entorhino-hippocampal organotypic culture to investigate the suitability as an in vitro model for information processing in the hippocampus. Organotypic cultures of the hippocampus were processed according to the method of Stoppini et al. [9]. The preparation cultured for 10-14 days was employed for optical recording. A slice was stained for 10 min in 0.2 mg/ml RH155 [2,4,15] solution dissolved in artificial cerebrospinal fluid (ACSF) having the following composition (in mM): NaC1 127, KC1 1.6, KH2PO4 1.24, MgSO4 1.3, CaCI2 2.4, NaHCO3 26 and glucose 10, which was continuously bubbled with a mixture of 95% 02 and 5% CO2 and then washed in ACSF at least for 15 min. The optical recording system (HR Deltaron 1700, Fuji
0304-3940/96/$12.00 © 1996 Elsevier Science Ireland Ltd. All rights reserved PII S0304-3940(96) 13001-9
212
Y. Nakagami et al. / Neuroscience Letters 216 (1996) 211-213
Photo Film, Japan)consisted of a 128 × 128 photodiode array and a processing unit [6]. The system enabled us to investigate electrical activities of neurons and its usefulness has been already demonstrated in the piriform cortex [10], visual cortex [11] and neurohypophysis [7]. Transmitted light with a wavelength of 700 + 20 nm was projected. The duration of light exposure was minimized ( 1 s) by a mechanical shutter to avoid photodynamic damage and dye bleaching. Before stimulation, a 128 frame averaged image of background light signal was stored in a reference memory. The stored reference data were continuously subtracted from real-time images and transferred to the unit sequentially at a frame rate of 0.6 ms. One trial consisted of 512 real-time images, and 16 trial images were averaged to improve the signal-to-noise ratio. Then, a slice was transferred to a recording chamber and perfused at a rate of 2.0 ml/min with ACSF at room temperature (25-27°C). Conventional electrophysiological technique of extracellular recordings was also employed to check the responses of the slice and for the adjustment of the stimulus strength during optical recording. A rectangular pulse (0.5-0.7 mA) of 100 t~s duration was delivered every 15 s with a strength that gave the maximum amplitude of field population spike. Fig. 1A indicates a 12 frame sequence of signal propagation, recorded after the stimulation of the perforant pathway. About 3.0 ms after the stimulus delivery, weak excitatory signal was detected in DG. Then, the activity invaded into CA3 (6.0 ms) and propagated to CA1 (12.0 ms). Subsequently, the excitatory propagation extended into both of the presubiculum and the parasubiculum, and into layer V. The activity in layer V probably consists of both of the activity from CA1 and from the presubiculure and the parasubiculum. Finally, the horizontal propagation from the medial entorhinal cortex to the lateral entorhinal cortex in layer V was seen, followed by the vertical propagation to layer II. This type of propagation was observed in 25 slices out of 28. The time course of the optical signals recorded from several photodiodes is shown in Fig. lB. The latency of the signals was short in DG, CA3, CA1 and layer V in ascending order. This result demonstrates that the cytoarchitectonic network including entorhino-hippocampal inputs and outputs was highly retained in the entorhino-hippocampal organotypic culture. In addition, we investigated whether the same neuronal propagation was observed in the acute entorhinohippocampal slice. The excitatory propagation of the hippocampal trisynaptic pathway and the long-lasting excitation in CA3 were observed after the stimulation of the perforant pathway (Nakagami et al., submitted), but the propagation of CA1 activity to the entorhinal cortex was not observed in the acute slices (n = 49). It is suggested that regeneration and sprouting of neurons in organotypic slices restored the neural connections. Another different excitatory propagation was shown in Fig. 2. The stimulation of the layer V in the lateral entorh-
E
3.Ores
6.0ms
.:...
9.0ms
12.0 ms
15.0 ms
18.0 ms
21.0 m s
24.0 ms
36.0 ms
4 8 0 ms
60.0 m s
Ores
V
CA1
1
i 0.03 %1 L _ _ 50 ms
Fig. 1. (A) Optical signal propagation from entorhino-hippocampal organotypic slice after the stimulation of the perforant pathway. Numerals under the images correspond to time after the stimulus delivery. The fractional absorbance change was monochrome-coded, as shown by the pseudo-monochrome scale in the bottom right comer. Dashed lines in 0 ms image indicate the approximate outline of pyramidal and granule cell layers of the hippocampus, and layers of the entorhinal cortex. An asterisk indicates the stimulus site. (B) Time course of the optical signals after the stimulation of the perforant pathway. The recording positions of the signals correspond to 0 ms image in (A). Stimulus delivery is indicated as an arrowhead.
inal cortex induced the activity along layer V to the medial site. The conduction velocity of the horizontal propagation from lateral to medial was about 70% of the reversed propagation. Then the activity reached to the presubiculum and the parasubiculum and finally to layer II. The application of 10 tzM 6-cyano-7-nitroquinoxaline (CNQX), a nonN-methyl-D-aspartate receptor antagonist, blocked the propagation of the optical signals, indicating that the optical signals were derived from the postsynaptic response, not just from action potentials of stimulated axons (data not shown). An immediate question arising may be whether signal propagation observed in the present study using organotypic culture reflects innervation manifested in vivo. However, there are many reports showing that the organotypic culture retains many characteristics of its normal cytoarch-
Y. Nakagami et al. /Neuroscience Letters 216 (1996) 211-213
, , I!
"
r'l
0 ms
6.0 ms
12.0 m s
24.0 ms
30.0 ms
36.0 m s
42.0 m s
48.0 ms
54.0 m s
60.0 ms
66.0 ms
18.0 m s
Fig. 2. Optical signal propagation from entorhino-hippocampalorganotypic slice after the stimulationof the lateral entorhinalcortex. An asterisk indicatesthe stimulus site. itectonic organization, and excitatory and inhibitory synapses function in the same way as confirmed in vivo [3,12]. It is also reported that the fiber projection between the entorhinal cortex and the hippocampus selectively regenerated in both directions and showed target specificity in the static culture of large horizontal slices [17]. As shown in Fig. 1A, the present studies have demonstrated that the entorhino-hippocampal organotypic culture retained the neuronal pattern of the excitatory propagation after the stimulation of the perforant pathway by optical recording. Histochemical studies support that the pattern is reasonable. The projection to DG arises mainly from layer II of the entorhinal cortex [8] and the activity in granule ceils in DG induce excitation in the well-known trisynaptic pathway of the hippocampus [ 1]. The CA1 pyramidal neuron projects to both the presubiculum and the parasubiculum, and also terminates mainly in layer V [13]. The excitatory propagation from the lateral to the medial entorhinal cortex was demonstrated in Fig. 2. The result also corresponded well to the histochemical studies. Axons from cells in layer Va of the entorhinal cortex terminate sparsely in layer I of the presubiculum and the parasubiculum [ 14]. The projections from the presubiculum and the parasubiculum terminate selectively in layers III and II of the entorhinal cortex, respectively [ 13,14]. Within layer II, the stellate cells project to an extensive collaterals plexus within layers I and II and to a lesser extent in layer III [5]. Furthermore, the pyramidal cells in layer III project to CA1 and the subiculum, and their collaterals are predominantly distributed to layers III and I. The intrinsic entorhinal circuit highly and systematically regulates hippocampal inputs and outputs in vivo. In conclusion, we have firstly demonstrated that the entorhino-hippocampal organotypic culture retained cytoarchitectonic organization manifested in vivo by optical recording. The analysis of acute hippocampal slice by electrophysiological technique is not probably sufficient to
213
comprehend entorhino-hippocampal neuronal network. Although further detailed studies are certainly needed, we think the analysis of the entorhino-hippocampal organotypic culture would be a suitable in vitro model for information processing in the entorhino-hippocampus to investigate the mechanism of learning and memory. [1] Amaral, D.G. and Witter, M.P., The three-dimensionalorganization of the hippocampal formation: a review of anatomical data, Neuroscience, 31 (1989) 571-591. [2] Cinelli, A.R. and Salzberg, B.M., Dendriticorigin of late events in optical recordings from salamander olfactory bulb, J. Neurophysiol., 68 (1992) 786-806. [3] Frotscher, M. and Heimrich, B., Lamina-specificsynapticconnections of hippocampal neurons in vitro, J. Neurobiol., 26 (1995) 350-359. [4] Konnerth,A., Obaid,A.L. and Salzberg,B.M., Opticalrecordingof electrical activity from parallel fibres and other cell types in skate cerebellar slices in vitro, J. Physiol., 393 (1987) 681-702. [5] Lingenhohl,K. and Finch,D.M., Morphologicalcharacterizationof rat entorhinalneuronsin vivo: soma-dendriticstructure and axonal domains, Exp. Brain Res., 84 (1991) 57-74. [6] Matsumoto, G. and Ichikawa, M., Optical system for real-time imaging of electrical activity with a 128 x 128 photopixel array, Soc. Neurosci. Abstr., 16 (1990) 490. [7] Obaid, A.L. and Salzberg, B.M., Micromolar 4-aminopyridine enhances invasion of a vertebrate neurosecretory terminal arborization, J. Gen. Physiol., 107 (1996) 353-368. [8] Ruth, R.E., Collier, T.J. and Routtenberg,A., Topographicalrelationship between the entorhinalcortex and the septotemporal axis of the dentate gyrus in rats: II. Cells projectingfrom lateral entorhinal subdivisions,J. Comp. Neurol., 270 (1988) 506-516. [9] Stoppini, L., Buchs, P.-A. and Muller, D., A simple method for organotypic cultures of nervous tissue, J. Neurosci. Methods, 37 (1991) 173-182. [10] Sugitani,M., Sugai, T., Tanifuji,M. and Onoda, N., Signal propagation from piriform cortex to the endopiriformnucleus in vitro revealed by optical imaging,Neurosci. Lett., 171 (1994) 175-178. [11] Tanifuji,M., Sugiyama,T. and Murase, K., Horizontalpropagation of excitation in rat visual cortical slices revealed by optical imaging, Science, 266 (1994) 1057-1059. [12] Torp, R., Hang, F.M., Tender, N., Zimmer, J. and Ottersen, O.P., Neuroactive amino acids in organotypic slice cultures of the rat hippocampus: an immunocytochemicalstudy of the distribution of GABA, glutamate, glutamine and taurine, Neuroscience, 46 (1992) 807-823. [13] Van Groen, T. and Wyss, J.M., Extrinsic projections from area CA1 of the rat hippocampus: olfactory, cortical, subcortical and bilateral hippocampal formation projections, J. Comp. Neurol., 302 (1990) 515-528. [14] Van Groen, T. and Wyss, J.M., The postsubicularcortex in the rat: characterization of the subicularcortex and its connections,Brain Res., 529 (1990) 165-177. [15] Vranesic,I., Iijima, T., Ichikawa, M., Matsumoto, G. and Knrpfel, T., Signal transmissionin the parallel fiber - Purkinjecell system visualizedby high-resolutionimaging,Proc. Natl. Acad. Sci. USA, 91 (1994) 13014-13017. [16] Witter,M.P., Organizationof the entorhinal-hippocampalsystem: a review of current anatomicaldata, Hippocampus,3 (Suppl.)(1993) 33-44. [17] Woodhanas,P.L., Atkinson,D.J. and Raisman,G., Rapid declinein the ability of entorhinalaxons to innervatethe dentate gyrus with increasing time in organotypic co-culture, Eur. J. Neurosci., 5 (1993) 1596-1609.