Downloaded from learnmem.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press Temporally Graded Retrograde Amnesia Following Separate and Combined Lesions of the Perirhinal Cortex and Fornix in the Rat Kjesten A. Wiig, 1'2 Leon N. Cooper, ~'3 and Mark F. Bear 4'5 1Institute for Brain and Neural Systems and Departments of 2Neur0science and 3Physics 4H0ward Hughes Medical Institute Brown University Providence, Rhode Island 02912 Abstract 2-week retention interval than control rats. These results suggest that medial temporal The involvement of the perirhinal structures including the perirhinal cortex cortex and the fornix in retrograde and and the fornix are involved in the anterograde amnesia in the rat was consolidation of mnemonic information investigated in this experiment. Male and that their involvement in this process Sprague-Dawley rats were trained on a occurs over a discrete period of time. series of five visual discrimination problems at distinct time intervals prior to receiving bilateral, electrolytic lesions of the perirhinal cortex or the fornix, combined Introduction lesions of both these structures, or sham operations. Following recovery from Throughout the latter half of this century, surgery, rats were retested on the there has been considerable interest in determin- preoperatively learned discrimination ing how memories are encoded and stored within problems, as well as learning a new the mammalian brain. Although the precise mech- discrimination and discrimination reversal. anisms and loci of information storage have re- Results indicated that all animals with mained elusive, some progress has been made to- lesions exhibited temporally graded ward identifying both the brain structures in- retrograde amnesia, whereby memories volved in memory consolidation and the time acquired in the recent past (1-3 weeks) course or duration of this process. were impaired, and memories acquired in Much of our knowledge regarding the organi- the remote past (6-8 weeks) were spared. zation and neural foundations of normal memory There was no difference in the magnitude processes has been obtained from the study of the of retrograde amnesia between the three human amnesic syndrome. Patients who have suf- lesion groups. Animals in the perirhinal, fered damage to the medial temporal region of the fornix, and combined lesion groups were brain, including the hippocampus, amygdala, and able to learn a new discrimination problem surrounding parahippocampal cortical regions at a rate comparable to control rats; typically exhibit severe anterograde amnesia (an however, the animals with lesions were inability to form new memories) and a more vari- impaired at learning the discrimination able retrograde amnesia (loss of memories ac- reversal. The perirhinal, fornix, and quired prior to the precipitating incident). The combined lesion animals also exhibited a phenomenon of retrograde amnesia has had a par- significantly faster forgetting rate over a ticularly large impact on ideas regarding memory consolidation processes. Specifically, retrograde amnesia is often temporally graded, in that mem- SCorresponding author. ories acquired close to the time of the amnesia- LEARNING & MEMORY 3:313-325 9 1996 by Cold Spring Harbor Laboratory Press ISSN1072-0502/96 $5.00 L E A R N I N G & M E M O R Y 313 Downloaded from learnmem.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press Wiig et al. inducing incident are lost, whereas memories formance on DNMS (Otto and Eichenbaum 1992; formed in the more distant past are spared. In hu- Mumby and Pinel 1994) and paired-associate tasks mans, memories formed several years prior to the (Bunsey and Eichenbaum 1993). onset of amnesia can be disrupted (Squire et al. Moreover, it has been suggested that the per- 1975; Corkin 1984). This observation suggests irhinal cortex alone may be responsible for much that memory is not fixed at the time of learning of the mnemonic processing required for adequate but, rather, changes and becomes more stable with performance on tasks such as the DNMS. It has the passage of time (McGaugh and Gold 1976). been demonstrated, for example, that lesions re- The development of primate and rodent mod- stricted to the perirhinal cortex in the monkey els of retrograde amnesia has helped to further result in a more severe memory impairment than delineate which medial temporal lobe structures lesions confined to the entorhinal cortex (Meu- are involved in memory consolidation processes. nier et al. 1993). Similarly, bilateral ablations of Much of this research has focused on the hippoc- the perirhinal cortex alone in the rat have been ampal formation, and despite differences in tasks shown to significantly disrupt performance on and species, similar patterns of results have been postoperatively acquired DNMS memory tasks observed across studies. In these experiments, an- (Wiig and Bilkey 1994b, 1995). imals were trained on a task at different time in- Given the importance of the perirhinal cortex tervals prior to receiving bilateral ablations of the to the retention of newly acquired information, it hippocampal formation (Squire and Spanis 1984; is of considerable interest to determine the extent Salmon et al. 1987; Wincour 1990; Zola-Morgan of retrograde amnesia exhibited by animals with and Squire 1990; Kim and Fanselow 1992; Cho et lesions of this area. The following experiment was al. 1995; Kim et al. 1995) or entorhinal cortex designed to compare the retrograde consequences (Cho et al. 1993, 1995; Cho and Kesner 1996). of lesions of the perirhinal cortex and the fornix. The results of these experiments indicated that Rats learned two-choice visual discrimination animals retained information that had been ac- problems at five different time intervals prior to quired in the remote past and lost information that receiving lesions of the perirhinal cortex, fornix, had been acquired more recently. The temporal or both these structures. Retention for the preop- gradient of the retrograde amnesia varied from 10 eratively learned discrimination problems was as- days (Wincour 1990) to several months (Salmon sessed immediately following recovery from sur- et al. 1987). Although it is generally accepted that gery. In addition, the ability of animals with lesions these results cannot be used to determine the time to learn a new discrimination problem and rever- course of memory consolidation, they do support sal and retain this information over a 2-week in- the idea of a gradually changing memory trace, terval was assessed. which is more susceptible to disruption in early phases of memory consolidation. Although the involvement of the hippocampal Materials and Methods formation in retrograde amnesia has been well characterized, little is known about the contribu- SUBJECTS tion of the surrounding medial temporal cortical Twenty-four male Sprague-Dawley rats, regions. Recently, it has been demonstrated that weighing between 150 and 180 grams at the start the rhinal cortex appears to play a particularly im- of behavioral training, served as subjects in this portant role in the retention of newly acquired experiment. Subjects were individually housed in information. It has, for example, been shown that wire-mesh cages and were maintained on a 12-hr- in the monkey, surgical ablations encompassing on, 12-hr-off light-dark cycle. Rats had free access either the perirhinal and parahippocampal corti- to water but were food deprived to 85% of their ces (Zola-Morgan et al. 1989; Suzuki et al. 1993), free-feeding body weight. or the perirhinal and entorhinal cortices (Eacott et al. 1994; Meunier et al. 1993; Murray et al. 1996) result in profound memory deficits on post- APPARATUS operatively acquired delayed-non-match-to-sample (DNMS) and complex discrimination tasks. Simi- The apparatus used in this experiment con- larly, combined lesions of the perirhinal and en- sisted of a perspex runway measuring 60 cm in torhinal cortices in the rat result in impaired per- length and 20 cm in width and with walls 50 cm in L E A R N I N G & M E M O R Y 314 Downloaded from learnmem.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press RETROGRADE AMNESIA AND PERIRHINAL CORTEX height (for a more detailed description of this ap- PROCEDURE paratus, see Mumby et al. 1990). At each end of the runway were two identical goal areas, which PREOPERATIVE TRAINING were separated from the central portion of the maze by two opaque perspex guillotine doors. The pretraining phase of the experiment con- Within each goal area were two food wells, which sisted of six 15-min sessions during which the rats were separated from each other by a short divid- were habituated to the apparatus. For the first two ing wall. Food pellets were delivered to the wells sessions, each rat was placed into the apparatus, via funnels that were mounted on the outside with both guillotine doors open, and allowed to walls of the apparatus and connected to the food explore for 15 min. Pellets were scattered liberally wells by means of plastic tubing. throughout the length of the runway and in the A collection of 10 junk objects differing from food-well area. Food wells were rebaited after the one another in size, color, shape, and texture pellets in each one had been consumed. served as stimuli. The objects were large enough For the remaining four habituation session, to cover a food well but were light enough to the rats were trained to run from end to end of the enable a rat to displace them. The food rewards apparatus. Each rat was initially placed in the cen- were 45-mg food pellets (BioServ). ter of the runway, with both guillotine doors closed. After --5 sec, one guillotine door was raised, allowing the rat access to the food wells. After the rat had eaten the pellets placed in both food wells and had begun to move toward the SURGERY center of the runway, the closed guillotine door All animals were anesthetized with sodium was raised.