Lesions of the Entorhinal Cortex Impair Acquisition of Hippocampal-Dependent Trace Conditioning

Neurobiology of Learning and Memory 75, 121–127 (2001) doi:10.1006/nlme.2000.3966, available online at http://www.idealibrary.com on

Lesions of the Entorhinal Cortex Impair Acquisition of Hippocampal-Dependent Trace Conditioning Jae-Wook Ryou, Sun-Young Cho, and Hyun-Taek Kim Department of Psychology, Korea University, Seoul 136-701, Korea

Rabbits with the electrolytic lesions of bilateral entorhinal cortex (EC) were trained with the hippocampal-dependent trace conditioning of the nictitating membrane response. The multiple-unit activity of the hippocampal CA1 region was recorded during conditioning. The conditioned stimulus was a tone (1 kHz, 85 dB, 200-ms duration), the unconditioned stimulus was a corneal air puff (3 psi, 150ms duration), and the interstimulus interval was 750 ms. The EC-lesioned animals showed only 30% conditioned response (CR) by the ninth session while the shamoperated animals showed above 80% CR. The lesioned animals did not show learning-related changes in the hippocampal activity. When the training was switched to the 300-ms interstimulus interval trace conditioning, both groups learned above 80% CR. The EC-lesioned animals, however, showed less learning-related activity in the hippocampus than the sham-operated animals. These results suggest that the development of the learning-related activity in the hippocampus depends on the intact EC, and that the EC may provide a possible pathway conveying learning information from the cerebellum or cerebral cortex to the hippocampus during the trace conditioning. 䉷 2001 Academic Press Key Words: classical conditioning; nictitating membrane response; entorhinal cortex; hippocampus; trace conditioning; hippocampal-dependent learning.

During the classical conditioning of nictitating membrane (NM)/eyelid response, the cerebellum and the hippocampus develop the patterns of neuronal activity that are related to the learning of behavioral response (Berger, Rinaldi, Weisz, & Thompson, 1983; Kim, Choi, & Kim, 1991; Solomon, Vanderschaaf, Thompson, & Weisz, 1986). Lesion of the hippocampus does not affect the learning of the conditioned response (CR) in a standard delay conditioning (Solomon & Moore, 1975; Weikart & Berger, 1986), but it prevents or impairs the acquisition of CR in trace conditioning with long (⬎500 ms) trace intervals This research was supported by Korea Science and Engineering Foundation (KOSEF 951-0707-002-2) to H. T. Kim. J. W. Ryou is now at Department of Psychology and ARL division of Neural Systems, Memory, and Aging in the University of Arizona, Tucson, AZ 85724-5115. S. Y. Cho is now at Graduate School of EastWest Medical Science in Kyunghee University, in Seoul 130-701, Korea. Address correspondence and reprint requests to Jae-Wook Ryou. Fax: (520) 626-2080. E-mail: jwryou@ u.arizona.edu. 121

1074-7427/01 $35.00 Copyright 䉷 2001 by Academic Press All rights of reproduction in any form reserved.

122

RYOU, CHO, AND KIM

(James, Hardiman, & Yeo, 1987; Moyer, Deyo, & Disterhoft, 1990; Port, Romano, Steinmetz, Mikhail, & Patterson, 1986). The hippocampus receives inputs from the medial septum (MS) and the entorhinal cortex (EC) (Brodal, 1981; Swanson, Teyler, & Thompson, 1982). While EC shows patterns of neuronal activity similar to that of the hippocampus during NM conditioning, MS does not (Berger, Clark, & Thompson, 1980). Hippocampal single- and multiple-unit activity shows the learning-related changes. The neuronal activity markedly increases in the periods of conditioned stimulus (CS) and unconditioned stimulus (US) at the beginning of training, followed by moderate increases, and then decreases when the animal achieves asymptotic levels of CRs (McEchron & Disterhoft, 1997; Weiss, Kronforst-Collins, & Disterhoft, 1996). This learning-related activity is affected by the lesion of the cerebellar interpositus nucleus or the red nucleus (Sears & Steinmetz, 1990; Ryou, Cho, & Kim, 1998). It is possible that the learning information from the cerebellum or the cerebral cortex may influence hippocampal activity through the EC. In this study, we investigated the effects of EC lesion on the acquisition of the behavioral CR and the development of learning-related activity in the CA1 region of hippocampus, using a trace conditioning of NM response which is dependent upon the hippocampus. Study animals were 12 New Zealand White male rabbits weighing 1.8–2.2 kg at surgery with free access to water and food on the natural daylight cycle. Rabbits were anesthetized with intravenous injections of pentothal sodium (26 mg/kg) and positioned in the stereotaxic headholder with bregma 1.5 mm above lambda. The recording electrodes made of 00 stainless-steel insect pins insulated with epoxylite were implanted chronically in the CA1 region (5.5 mm posterior, 4.5 mm lateral, and 5.5 ⫾ 0.5 mm ventral to bregma) of the hippocampus, contralateral to the conditioned eye. Electrolytic lesions of EC were made at six sites by passing 2 mA direct current for 60 s through each lesioning electrode with tip exposed 0.5 mm. The coordinates for the EC lesion were 5 mm posterior, 6.2 mm bilateral, and 19 mm ventral to bregma; 7.5 mm, ⫾6.6 mm, 17 mm; 9 mm, ⫾6.2 mm, 15.2 mm (Fifkova & Marsala, 1967). Rabbits in the control group received the same treatment as the EC-lesioned animals except no current was passed through the electrode. After a week of recovery, rabbits were adapted for a session without conditioning stimuli. Rabbits were then trained in a trace conditioning paradigm consisting of a 200ms tone CS (1 kHz, 85 dB) and 150 ms air puff US (3 psi) with a 750-ms interstimulus interval (ISI). That is, the trace interval was 550 ms. The CR was defined as a NM extension greater than 0.5 mm during the ISI. Each session contained six blocks, each containing 10 trials. After nine acquisition sessions, the ISI was switched to 300 ms (the trace interval is 100 ms), and training was continued until a behavioral criterion of 80% CRs on two consecutive blocks. The multiple-unit activity (MUA) of the hippocampal CA1 region was amplified (⫻1000) and filtered (300 Hz–3kHz band-pass) and then sent to a window discriminator (four to five spikes per 100 ms). The output pulses of the window discriminator were collected by a personal computer (2048/s sampling rate) generating a 4-ms time bin histogram (Fig. 1). Each trial was divided into three 750-ms periods, and each period was divided again into three 250-ms subperiods (preCS1, preCS2, preCS3; CS1, CS2, CS3; US1, US2, US3). In case of 300-ms ISI training, each trial was divided into three 300ms periods, and each was divided into three 100msec subperiods. Standard scores were

ENTORHINAL LESIONS IMPAIR TRACE CONDITIONING

123

FIG. 1. Averages of NM responses and MUA histograms for an EC-lesioned animal and a control animal (left) 700 ms ISI; (right) 300 ms ISI. In each panel, the top portion is the average NM response, the middle portion is a raster of the window discriminated-output of MUA at the last in the session, and the bottom portion is the histogram of the cumulated raster through the session. The four triangles, reading from the left to the right, indicate CS onset, CS offset, US onset, and US offset.

calculated for each CS and US subperiod for each block relative to the preCS3 subperiod activity. After a marking lesion of the hippocampus (2 mA, 10 s, direct current), animals were overdosed with chloral hydrate and perfused intracardially with saline followed by 10% formalin solution. The removed brain was sliced to 50-␮m frozen sections, and every 10th section was mounted on a gelatin-coated slide and stained with thionin. Figure 2 shows the lesion sites of EC for a rabbit in the lesioned group. In every lesioned animal, the medial regions of the bilateral EC were destroyed. The recording electrodes were placed in or very near the CA1 area of hippocampus in the lesioned (n ⫽ 6) and control animals (n ⫽ 6). Bilateral lesions of the EC impaired the acquisition of behavioral CR in the trace conditioning with a 750-ms ISI (F(1, 10) ⫽ 5.73, p ⬍ .05). The lesioned group showed less than 30% CR by the ninth session, while the control group showed more than 80% CR (Fig. 3A). The onset latency of CR was not affected significantly by the lesion (F(1, 10) ⫽ 3.96, ns). The means (⫾SEM) of the onset latency in the ninth session of the EClesioned group and the control group were 709.85 ⫾ 92.06 and 592.61 ⫾ 140.07 (ms), respectively. There were no differences in the amplitude of UR between the two groups (F(1, 10) ⫽ 0.37, ns). When ISI of trace conditioning was switched to 300 ms, the EC-lesioned animals were able to reached a behavioral criterion of above 80% CR within four to five sessions. The mean of the CR onset latency of the lesioned group (219.19 ⫾ 39.28 in the last session) was not different from the control group (224.24 ⫾ 45.18). The last two sessions to the criterion of each animal were analyzed and are presented in the Fig. 3. As control animals acquired the CRs in the 750-ms trace conditioning, the hippocampal neuronal activity showed learning-related increases in the CS2 and CS3 subperiods. Bilateral lesions of the EC prevented the increase of the learning-related activity in both the CS2 and the CS3 subperiods (F(1, 10) ⫽ 18.76, p ⬍ .01; F(1, 10) ⫽ 21.29, p ⬍ .01; Fig. 3B). During the 300-ms ISI conditioning, the standard scores of hippocampal MUA

124

RYOU, CHO, AND KIM

FIG. 2. (A) Photographs of the brain slices of an EC-lesioned animal. The arrows indicate lesion sites. (B) Line drawings showing the greatest and least extent of the lesions of the EC. Numbers indicate the number of mm posterior to bregma. EC, entorhinal cortex; RSAG, area retrosplenialis agranularis; RSG, area retrosplenialis granularis

in the CS3 subperiod of the lesioned group were lower than those of the control group (F(1, 10) ⫽ 9.62, p ⬍ .05; Fig. 3B), even though lesioned group showed behavioral CR at the asymptotic level (Fig. 3A). The basic findings of this study are that the EC lesion impaired the acquisition of hippocampal-dependent trace conditioning with an ISI of 750 ms and that it prevented the development of the learning-related activity in the hippocampus. Even when the trace interval was reduced to 300 ms (i.e., short enough so that the EC lesioned animals could learn behavioral CRs), hippocampal MUA was significantly reduced in the lesioned group compared to sham rabbits. These results extend the finding of the hippocampal lesion studies in the trace conditioning paradigm with a 500-ms trace interval (Moyer et al., 1990; Port et al., 1986). There was a controversy in the learning deficit of the hippocampal lesioned animals. It seemed

ENTORHINAL LESIONS IMPAIR TRACE CONDITIONING

125

FIG. 3. (A) The mean (⫾SEM) of CR percentages of the EC lesion group (triangle) and control group (circle) in the trace conditioning with 750 ms ISI and 300 ms ISI. After the EC lesion or sham lesion, animals received nine sessions of training with 750 ms ISI, and then switched to training with 300 ms ISI. (B) The standard scores of hippocampal MUA in the trace conditioning with 750 ms ISI and 300 ms ISI. In the case of 300 ms ISI, only the last two acquisition sessions were shown.

126

RYOU, CHO, AND KIM

to depend on the extent of hippocampal damage. Using an air puff US, they found consistently that the hippocampal lesioned animals showed “nonadaptive” short-latency CRs. In our study, the EC-lesioned rabbits with severe learning deficits showed adaptive timed CRs on the few trials in which a CR occurred with 550-ms trace interval. These functional dissonances imply the roles of each structure in temporal processing. The hippocampus modulates the timing of execution of the learned behavior, and the EC sends the information of temporal events to the hippocampus where the association of the events occurs. The hippocampus receives sensory information via the superficial layers of the EC from various structures in the brain including the cerebral cortex, thalamus, and other brain stem structures (Amral & Witter, 1989; Brodal, 1981). The MUAs of the hippocampus and EC showed similar patterns during the conditioning (Berger, et al., 1980). Taken together, our data suggest that the EC contributes to the role of the hippocampus in hippocampal-dependent conditioning. The behavioral learning and the neuronal response of the hippocampus in the trace conditioning are in part dependent on the intact connections from the EC to the hippocampus. Importantly, an intact EC is essential for the development of learning-related hippocampal activity. The learning information from the cerebellum could reach the hippocampus via the EC. REFERENCES Amral, D. G., & Witter, M. P.(1989). The three-dimensional organization of the hippocampal formation: a review of anatomical data. Neuroscience, 31(3), 571–591. Berger, J. W., Clark, G. A., & Thompson, R. F.(1980). Learning-dependent neuronal responses recorded from limbic system brain structures during classical conditioning. Physiological Psychology, 8(2), 155–167. Berger, T. W., Rinaldi, P. C., Weisz, D. J., & Thompson, R. F.(1983). Single-unit analysis of different hippocampal cell type during classical conditioning of rabbit nictitating membrane response. Journal of Neurophysiology, 50(5), 1197–1219. Brodal, A.(1981). Neurological anatomy: In relation to clinical medicine. New York: Oxford Univ. Press. Fifkova, E., & Marsala, J.(1967). Stereotaxic atlases for the cat, rabbit and rat. In J. Bures, M. Petran, and J. Zachar, Eds. Electrophysiological methods in biological research (pp. 653–731). New York: Academic Press. James, G. O., Hardiman, M. J., & Yeo, C. H.(1987). Hippocampal lesions and trace conditioning in the rabbit. Behavioral Brain Research, 23, 109–116. Kim, H. T., Choi, J. S., & Kim, K. S. (1991). Functions of red nucleus, cerebellar interpositus nucleus in the classical conditioning of the rabbit’s nictitating membrane response. The Korean Journal of Biological and Physiological Psychology, 3, 65–82. McEchron, M. D., & Disterhoft, J. F.(1997). Sequence of single neuron changes in CA1 hippocampus of rabbits during acquisition of trace eyeblink conditioned responses. The Journal of Neurophysiology, 78(2), 1030–1044. Moyer, J. R., Deyo, R. A., & Disterhoft, J. F.(1990). Hippocampectomy disrupts trace eyeblink conditioning in rabbits. Behavioral Neuroscience, 104, 243–252. Port, R. L., Romano, A. R., Steinmetz, J. E., Mikhail, A. A., & Patterson, M. M.(1986). Retention and acqusition of classical trace conditioned responses by rabbits with hippocampal lesions. Behavioral Neuroscience, 100, 745–752. Ryou J. W., Cho S. Y., & Kim H. T.(1998). Lesion of the cerebellar interpositus nucleus or the red nucleus affects classically conditioned neuronal activity in the hippocampus. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 22, 169–185. Sears, L. L., & Steinmetz, J. E.(1990). Acquisition of classically conditioned-related activity in the hippocampus is affected by lesions of the cerebellar interpositus nucleus. Behavioral Neuroscience, 104(5), 681–692.

ENTORHINAL LESIONS IMPAIR TRACE CONDITIONING

127

Solomon, P. R., & Moore, J. W.(1975). Latent inhibition and stimulus generalization of classically conditioned nictitating membrane response in rabbits (Oryctolagus cuniculuc) following dorsal hippocampal ablation. Journal of Comparative and Physiological Psychology, 89, 1192–1203. Solomon, P. R., Vanderschaaf, E. R., Thompson, R. F., & Weisz, D. J.(1986). Hippocampus and trace conditiong of the rabbit’s nictitating membrane response. Behavioral Neuroscience, 100, 729–744. Swanson, L. W., Teyler, T. J., & Thompson, R. F.(1982). Hippocampal long-term potentiation: Mechanisms and implications for memory. In Neuroscience Research Program (Vol. 20). Boston: MIT Press. Weikart, C. L., & Berger, T. W.(1986). Hippocampal lesions disrupt classical conditioning of cross-modality reversal learning of the rabbit nictitating membrane response. Behavioral Brain Research, 22, 85–89. Weiss, C., Kronforst-Collins, M. A., & Disterhoft, J. F.(1996). Activity of hippocampal pyramidal neurons during trace eyeblink conditioning. Hippocampus, 6(2), 192–209.