Abstract

It has been proposed that isolation of the inferior temporal cortex and medial temporal lobe from their cholinergic afferents results in a severe anterograde amnesia. To test this hypothesis directly, seven rhesus monkeys received a unilateral immunotoxic lesion of the cholinergic cells of the basal forebrain with an ipsilesional section of the fornix. In a second surgery, inferior temporal cortex was ablated in the opposite hemisphere. All animals were severely impaired at learning visual scenes and object–reward associations. The impairment in learning scenes was correlated with cholinergic cell loss in the basal forebrain, but not with generalized tissue damage. Two monkeys served as surgical controls with saline injection in place of the immunotoxin, but all other procedures the same, and were not as severely impaired as those with immunotoxic lesions. Previous work has shown that monkeys with bilateral section of the anterior temporal stem (white matter of the temporal lobe), amygdala and fornix show a severe new learning impairment, and provide a model of human medial temporal lobe amnesia. One effect of this combined ablation is to isolate inferior temporal cortex and medial temporal lobe from their cholinergic afferents, possibly in addition to a direct disruption of the hippocampal system. The results of the present study, then, provide a novel link between the mechanisms of medial temporal lobe amnesia and Alzheimer's disease in which the cholinergic basal forebrain shows pathology. We propose that in both cases the mnemonic impairments result from isolating inferior temporal cortex and medial temporal lobe from their cholinergic afferents, possibly in addition to a direct disruption of the hippocampal system.

Introduction

Recent work in monkeys has shown that a severe anterograde amnesia can result from the surgical section of the anterior temporal stem (white matter of the temporal lobe dorsolateral to the amygdala), amygdala and fornix (Gaffan et al., 2001; Maclean et al., 2001). This combined damage in monkeys resembles the damage seen in patients with severe anterograde amnesia after medial temporal lobe surgery, such as patient HM (Scoville and Milner, 1957) where there is damage to both the amygdala and the hippocampal system, as well as some damage to the white matter dorsolateral to the amygdala (Corkin et al., 1997). Although monkeys with combined damage to the anterior temporal stem, amygdala and fornix are severely impaired at a variety of learning tasks, lesions of the individual components produce much milder impairments (Cirillo et al., 1989; Gaffan et al., 2001). It appears, then, that the severe learning impairments seen after combined section of the anterior temporal stem, amygdala and fornix are a result of surgical disruption to all three structures. The direct cause of the learning impairment following this combined surgery is, however, unclear. There are many fibre pathways through the anterior temporal stem (Ebeling and Cramon, 1992), and fibres of passage through the amygdala are known to underlie the learning of some associative tasks (Dunn and Everitt, 1988; Malkova et al., 1997). Similarly, the hippocampal system is known to be important for memory in both humans and monkeys (Gaffan and Gaffan, 1991; Ridley et al., 1991, 1995; Gaffan and Parker, 1996; Parker and Gaffan, 1997a,b; Aggleton and Brown, 1999; Aggleton et al., 2000).

One explanation of the effects of this combined lesion is that it interrupts the projections of the cholinergic cells of the basal forebrain to the medial temporal lobe and inferior temporal cortex, as these projections pass through the fornix, amygdala and anterior temporal stem in monkeys (Kitt et al., 1987) and humans (Seldon et al., 1998). The cholinergic projections to the cortex are widespread while the cortical inputs to the same cells are much more restricted (Mesulam and Mufson, 1984). This pattern of limited cortical afferents with extensive cortical efferents makes these cholinergic cells an ideal intermediary for cortical modulation of the entire cortical mantle by only a few cortical areas (such as the frontal cortex).

This cortico-cortical modulation via the basal forebrain would support a role for the cholinergic basal forebrain in learning, by allowing the individual's goals (which might be represented in the frontal cortex) to mediate the learning of appropriate stimuli and information. Disruption of the basal forebrain interaction with inferior temporal cortex by crossed unilateral lesions of the medial forebrain bundle and inferior temporal cortex produces learning impairments similar to those seen after section of the anterior temporal stem, amygdala and fornix (Easton and Gaffan, 2000)

In humans, damage to the basal forebrain can result in severe learning problems. Surgical resection of aneurysms of the anterior communicating artery, or their rupture, can produce widespread damage in the region of the basal forebrain and result in a severe amnesia (Damasio et al., 1985; Abe et al., 1998). The cholinergic cells of the basal forebrain are also among the first to degenerate in Alzheimer's disease (Whitehouse et al., 1982). Of a large number of transmitter systems assessed in a wide range of brain areas in Alzheimer's patients at postmortem, only the reduction in the cholinergic marker cholineacetyltransferase in the cortex is seen to correlate with the severity of dementia prior to death (Bierer et al., 1995). In cases of Alzheimer's disease or aneurysm of the anterior communicating artery, however, cell loss is not confined to the cholinergic basal forebrain, so more selective lesions in animals can clarify the importance of the cholinergic basal forebrain in learning.

Recent advances have allowed examination of the specific role of the cholinergic basal forebrain by immunolesioning the cholinergic cells of the basal forebrain alone. The ribosome-inactivating toxin, saporin, can be conjugated to an antibody that recognizes the p75 neurotrophin receptor, and thus can selectively kill the cholinergic cells of the basal forebrain that express the p75 receptor (Wiley et al., 1991). In the marmoset these immunotoxic lesions have produced evidence that the cholinergic system is essential to normal mnemonic performance in the primate (Fine et al., 1997; Ridley et al., 1999a,b; Barefoot et al., 2000).

In humans there is evidence that patients in the early stages of Alzheimer's disease not only have loss of the cholinergic cells of the basal forebrain, but also have damage to the medial temporal lobe structures including the hippocampus and entorhinal cortex (Frisoni et al., 1999). Therefore, in the present study, the effects of combined lesions of the cholinergic basal forebrain and fornix in one hemisphere and inferior temporal cortex in the opposite hemisphere were tested. A lack of impairment from this combined lesion would indicate that combined damage to both the cholinergic projections to cortex and the hippocampal system is not the cause of the severe amnesia in the monkeys with section of the temporal stem, amygdala and fornix. A severe learning impairment, however, would support the hypothesis that the cholinergic isolation of the inferior temporal cortex and medial temporal lobe (either with or without associated disruption of the hippocampal system) leads to a severe learning impairment.

Materials and Methods

Subjects

Subjects were nine rhesus monkeys (Macaca mulatta), all female. The animals weighed between 2.2 and 4.1 kg at the start of training. Animals A–G formed group ACh+fx while animals H and I formed group Con.

Surgery

All operations were performed in sterile conditions with barbiturate anaesthesia and with the aid of an operating microscope. When the lesion was complete, overlying tissue was sutured in anatomical layers. Following each surgery, 10–18 days were allowed for post-operative recovery before the resumption of training.

Unilateral Lesion of the Cholinergic Cells of the Basal Forebrain with Unilateral Fornix Section (Group ACh+fx)

The first surgery for animals in group ACh+fx was a lesion of the cholinergic cells of the basal forebrain, with unilateral transection of the fornix, in the left hemisphere. The animal was placed in a stereotaxic head holder and a D-shaped bone flap was raised at the midline extending laterally on the side of the lesion. The veins constraining access to the midline were cauterized and the hemispheres separated. The hemisphere was retracted from the falx and the fibres of the corpus callosum were cut using a glass aspirator to reveal the fornix. The fornix was retracted, and tissue dorsal to the anterior commisure removed with a glass aspirator to visualize the anterior commissure. In this exposure of the anterior commisure, the descending column of the fornix was sectioned unilaterally. A Hamilton syringe (1 ml) with a blunt needle was then guided visually onto the surface of the anterior commissure in the midline and the stereotaxic coordinates of this position were measured. The coordinates for the injections were then calculated. There were 54 injections made for one hemisphere, with 22 needle tracks. Each injection was of the immunotoxin ME20.4IgG–saporin (Advanced Targetting Systems, San Diego, CA, USA) which is the ribosome inactivating toxin saporin conjugated to the antibody ME20.4IgG which recognizes the primate (including human) low-affinity neurotrophin receptor protein p75 which exists on the cholinergic cells of the basal forebrain (Maclean et al., 1997). Each injection was at a rate of 1 μl/min, with a total volume of 1 ml per injection site for every animal except animal G that received 2 ml per injection site. At the end of each injection the needle was kept in place for a further minute, and at the end of a needle track for 2 min. The sites of the injections were calculated from a stereotaxic atlas of the macaque brain (Martin and Bowden, 1996) to include all areas of the basal forebrain that are seen to stain for cholinergic cells with immunohistochemistry using the same ME20.4IgG antibody. The most anterior needle tracks were made first as these were in the region of the anterior perforated substance. Any bleeding or oedema that resulted from damaging any of these small blood vessels could then be spotted and treated during the surgery to minimize the risk to the animal. These small bleeds were reduced by the use of the blunt needled Hamilton syringe.

Unilateral Injection of Saline to Basal Forebrain with Unilateral Fornix Section (Group Con)

The first surgery in animals in group Con were unilateral injections of saline into the left basal forebrain with section of the descending column of the fornix unilaterally on the side of the injections. The procedure was identical to group ACh+fx, but with injections of saline rather than immunotoxin.

Unilateral Ablation of the Inferior Temporal Cortex

All animals received an ablation of the right inferior temporal cortex in a second surgery within 6 weeks of the first surgery. This lesion has been described in detail elsewhere (Easton and Gaffan, 2000, 2001), and included removal of tissue extending from the fundus of the rhinal sulcus to the fundus of the superior temporal sulcus, including both banks of the anterior part of the occipitotemporal sulcus. The posterior limit of the lesion was just anterior to the inferior occipital sulcus.

Histology

At the conclusion of the behavioural experiment the animals were deeply anaesthetized, then perfused through the heart with saline followed by formol–saline solution. The brains were blocked in the coronal stereotaxic plane posterior to the lunate sulcus, removed from the skull, and allowed to sink in sucrose–formalin solution. The brains were cut in 50 mm sections on a freezing microtome. Every fifth section was retained and stained with cresyl violet. Another set of sections were immunostained with antibodies for those cells expressing the p75 neurotrophin receptor, i.e. the same cells as targeted by the immunotoxin used in the experiment, using a procedure previously used in marmosets (Maclean et al., 1997).

Immunostained sections were taken at the four levels through the injection sites shown in Figure 3. At each level a score was given to the degree of cavitation in the basal forebrain and in the basal ganglia. A score of 0 represented no cavitation, a score of 5 represented complete cavitation. The sum of the scores for the basal forebrain and basal ganglia cavitation was then calculated across all the four section levels and a final cavitation score (out of a maximum of 40) was obtained. This was not a measure of cell loss, or selectivity of the lesion, but was merely a measure of the non-selective damage produced from the cavitation.

These immunostained sections were also analysed for loss of cholinergic cells. These sections were placed under a microscope linked to a computerized cell counting program that counted the number of immunostained cells. This count was taken over the entire extent of the cholinergic basal forebrain in both the lesioned and normal hemispheres of each immunotoxic lesioned animal. The brains of the animals in group Con were not immunostained because of a subsequent ablation of the frontal cortex in the immunolesioned hemisphere for an experiment not reported here. In calculating the percent depletion of cholinergic sides in the immunolesioned hemisphere, the immunolesioned hemisphere cell count was compared with the cell count in the hemisphere with the inferior temporal cortex ablation. The inferior temporal cortex ablation, however, may have caused retrograde degeneration and cell loss in the basal forebrain. Therefore, two animals that had a unilateral lesion of the inferior temporal cortex only (for an experiment not described here) were also immunostained and cell counts were made in the intact hemisphere and the hemisphere with the inferior temporal cortex lesion. It was found on average that there was a 14.2% depletion of cholinergic cells in the basal forebrain of the hemisphere with the inferior temporal cortex lesion. In the cell counts for the animals with the immunotoxic lesions, then, the immunolesioned hemisphere cell count was compared with a cell count in the inferior temporal cortex lesioned hemisphere, which had been corrected by 14.2% to allow for this retrograde degeneration from the cortical ablation.

Apparatus

Monkeys were tested in an automated apparatus, receiving small food rewards for correct choices and a large food reward at the end of a session. The apparatus is described in detail elsewhere (Parker and Gaffan, 1998).

Procedure

Figure 1 shows a timeline for group Con and group ACh+fx showing both surgeries and behavioural tests. Before starting on the experiment, all animals were trained to touch the touchscreen. Then, pre-operatively all animals were trained on scene learning. This task is described in detail elsewhere (Gaffan, 1994), but required the animal to learn the correct object (a small two-dimensional alphanumeric character) of two presented in specific spatial locations on specific unique backgrounds. Animals were presented with eight trials of 20 such problems in any session, and were presented with a new set of 20 problems for every session. Performance was measured as mean errors after the first trial with each problem, over 10 sessions. This scene learning was then retested after unilateral surgery. After the second surgery to add the inferior temporal cortex ablation, the animals were tested again on scene learning, and then post-operatively taught and tested on visual object– reward association learning. The visual object–reward association learning task is described in detail elsewhere (Easton and Gaffan, 2001) but required the animal to learn over several sessions a set of 10 concurrent visual object–reward associations between small alphanumeric objects. Animals learnt three such sets of objects and performance was measured as mean errors to criterion (90% or greater correct responses in a single test session).

Results

Histology

Figure 2 shows the intended area removed in the unilateral inferior temporal cortex ablation. The ablation was seen to be as intended in all animals, extending from the fundus of the superior temporal sulcus to the fundus of the rhinal sulcus, and posteriorly including both banks of the anterior part of the occipitotemporal sulcus. The posterior limit of the ablation was just anterior to the inferior occipital sulcus.

The animals in group ACh+fx all had substantial loss of cholinergic cells in the hemisphere that had received the immunotoxic lesion. Examples of the histology from animal E with a near complete loss of cholinergic cells on the side of the immunotoxin injections are shown in Figures 3 and 4. Cresyl stained sections are shown in Figure 3 and immunostained sections from the same levels in Figure 4. In some animals, cavitation of the tissue around the injection sites was seen. An example from animal B is shown in Figure 3E, and this particular section scored 5 out of 5 for cavitation in both the basal forebrain and basal ganglia (it should be noted that although cavitation was severe at this level in this animal, cavitation around this section was much less severe, hence the animal's relatively low total cavitation score). A measure of cavitation (as described in Materials and Methods) for each animal is shown in Table 1, along with the percent loss of cholinergic cells at each of these four levels. All animals had unilateral damage to the tissue dorsally adjacent to the anterior commisure, including a section of the descending column of the fornix unilaterally with unilateral degeneration of the fornix (e.g. Fig. 3F). Animals F and G were seen to have enlarged third ventricles at histology, on the side of the injections.

In addition to the intended lesions, some animals showed unintended damage. Animals C, D and G showed a small degree of unilateral damage to the midline, in the cingulate cortex, from the exposure of the anterior commissure during the first surgery. Previous studies have shown that complete, bilateral, ablations of the cingulate cortex do not impair the scene memory task used in the current experiment (Parker and Gaffan, 1997a). Therefore we do not believe that this small unilateral damage to the cingulate cortex in the present animals contributed to the memory impairments in these animals. Animal G showed some degree of cell loss in the motor cortex, dorsal to the region of the injections.

Scene Learning

The animals' proficiency at scene learning was measured by the mean number of errors made in each session (after the initial presentation when the animals' choice of objects was random). These results are shown in Figure 5 along with the average learning curve for each group.

The unilateral lesions of immunotoxin and fornix section, or saline injection and fornix section had no effect on scene learning, as shown in Table 1. A one-way ANOVA of the change in performance between stages ‘Pre-op’ and ‘Post-op 1’ (Table 1) showed no significant change [F(1,7) < 1]. Similarly, previous results have shown that a unilateral lesion of the inferior temporal cortex does not result in a learning impairment in this task [(Gaffan and Parker, 1996); A. Easton and D. Gaffan, unpublished observations]. However, in the initial stages of post-operative testing, animals in group ACh+fx showed some early evidence of bias towards objects in the ipsilesional visual field. This effect recovered in no more than eight sessions, and post-op 1 performance was measured after the recovery of any side bias.

Both groups were significantly impaired following the second surgery to ablate the contralateral inferior temporal cortex. Planned comparisons within each group, using the pooled error term from a one-way ANOVA of the change in performance between Pre-op and Post-op 2 (Table 1) showed that both groups were impaired [Con: t(7) = 4.31, P < 0.005 one-tailed; ACh+fx: t(7) = 38.3, P < 0.001 one-tailed]. However, the effect of group in the one-way ANOVA showed that group ACh+fx was significantly more impaired than group Con [F(1,7) = 131.7, P < 0.001].

Concurrent Visual Object–Reward Association Learning

Object–reward association learning was measured by the mean total errors to criterion (not counting the first presentation of each problem when performance was necessarily at chance levels) for each of three sets of 10 visual object–reward association problems. These results are shown in Figure 6. A t-test of the performance by animals in each group shows that the animals in group ACh+fx were impaired, relative to those in group Con [t(7) = 2.21, P < 0.05 one-tailed]. The performance of group Con was at a similar level to that of normal animals in this task (Easton and Gaffan, 2001; Gaffan et al., 2001).

Correlation Between Scene Learning and Loss of Acetylcholine in the Basal Forebrain

A stepwise multiple regression analysis was undertaken to investigate the power of cholinergic cell loss, pre-operative learning proficiency, and cavitation as predictors of post-operative learning proficiency. The predictors are shown in Table 1 as ACh loss, Pre-op and Cavitation, respectively, and the predicted variable is Post-op 2 in Table 1. The most powerful single predictor was cholinergic cell loss, which as a single predictor was significantly related to post-operative performance [t(5) = 3.291, P = 0.022]. Adding the next most powerful predictor, preoperative performance, as a second predictor produced a model in which both cholinergic cell loss [t(4) = 4.970, P = 0.008] and pre-operative performance [t(4) = 3.071, P = 0.037] were significant predictors of postoperative performance.

Cavitation was not significantly related to post-operative performance, either when added to the model with cholinergic loss as the single predictor [t(4) < 1] or when added to the model with both cholinergic loss and pre-operative learning as the two predictors [t(3) < 1]. The results of the stepwise regression analysis show, therefore, that cholinergic cell loss, and not cavitation, predicts post-operative performance.

Discussion

Crossed unilateral lesions of the cholinergic basal forebrain and fornix in one hemisphere and inferior temporal cortex in the other hemisphere resulted in a severe impairment in both scene learning and object–reward association learning. The learning impairments in the current experiment are not the result of disconnecting the fornix in one hemisphere and the inferior temporal cortex in the opposite hemisphere. Unilateral fornix section was the same for animals in groups ACh+fx and Con, and yet group ACh+fx is much more severely impaired than the control group. This supports the hypothesis that preventing the interaction between the inferior temporal cortex and the cholinergic basal forebrain bilaterally (possibly in combination with a bilateral disruption of the hippocampal system) results in a learning impairment very similar to that following bilateral section of the anterior temporal stem, amygdala and fornix (Gaffan et al., 2001).

Relationship of Cholinergic Cell Loss and Post-operative Learning

The possibility needs to be considered that a generalized (but unilateral) cell loss in the basal forebrain following the injections of immunotoxin may be responsible for the impairment, rather than a loss of the cholinergic cells in particular. Cavitation of tissue did occur in some animals in group ACh+fx, and we did not measure the loss of transmitter systems other than acetylcholine in any animal. However, the toxin used here has previously been shown to produce selective loss of acetylcholine in rhesus monkeys (Mrzljak et al., 1998), and produces significant learning impairments in marmosets without any cavitation of tissue (Fine et al., 1997; Ridley et al., 1999a,b; Barefoot et al., 2000). The impairment in marmosets is both sensitive to exacerbation by the cholinergic antagonist scopolamine (Fine et al., 1997) and to amelioration by the cholinergic agonist pilocarpine (Ridley et al., 1999b). What can be concluded from the current experiment is that a lesion of the basal forebrain using an immunotoxin, shown in other studies to be specific for cholinergic cells, produces a severe learning impairment in rhesus monkeys.

The design of the present experiment, using a large number of monkeys and a range of doses of immunotoxin in group ACh+fx, allowed a regression analysis to determine the relationship between post-operative learning and both the loss of cholinergic cells in the basal forebrain and generalized tissue damage in the form of cavitation in these animals. The results clearly show that general cavitation of tissue does not predict the post-operative learning impairment in group ACh+fx, while the loss of cholinergic cells does. This is strong evidence in support of a cholinergic explanation of the data. It is also apparent that some animals did not have any, or only had small amounts, of cavitation of tissue, but they were still severely impaired at new learning (e.g. animal A). Of interest is the observation that neither of the two animals in group Con showed any cavitation following injections of the same volume of saline. This may suggest that cavitation is a risk of near complete cholinergic depletion throughout the basal forebrain in the macaque monkey.

Effects of Fornix Section

All animals in group ACh+fx had a unilateral section of the fornix on the side of the immunotoxic lesion. The current experiment does not allow us to examine the role of the fornix section specifically in these animals. However, previous work has shown that the cholinergic projections to both inferior temporal cortex and medial temporal lobe must be lesioned to produce a severe and persistent learning impairment in the monkey (Ridley et al., 1999a,b). It is possible, however, that the fornix section in the current experiment serves not only to isolate the hippocampal system from its cholinergic afferents, but actually has an additional and additive effect to the immunotoxic lesion.

Lesions in the hippocampal–anterior thalamic circuit produce impairments in scene learning comparable to those seen in group Con (Gaffan and Parker, 1996; Parker and Gaffan, 1997a,b). Group Con had a unilateral fornix section crossed with a lesion of the inferior temporal cortex that has previously been shown to produce a similarly sized impairment (Gaffan and Parker, 1996). However, the impairment seen in group ACh+fx is much more severe than that seen in animals with interruption of this hippocampal–anterior thalamic circuit alone. This indicates that the cholinergic cell loss in the basal forebrain is having an effect independent of, and possibly in addition to, the interruption of cholinergic projections to the hippocampus. Patients with Alzheimer's disease show pathology that includes not only degeneration of cells in the cholinergic basal forebrain, but also of pathology in the medial temporal lobe. This explanation would also provide an explanation for the smaller effect of excitotoxic lesions of the basal forebrain in monkeys (Aigner et al., 1991; Voytko et al., 1994) compared with those of the more selective immunotoxic lesions in the current experiment. If the hippocampal–anterior thalamic circuit must be damaged in addition to the cholinergic basal forebrain, then an additional fornix lesion in these earlier studies would have produced a more severe learning impairment.

Role of the Basal Forebrain in Attention or Memory?

Previous studies in rats (Dunnett et al., 1991; Baxter et al., 1995, 1997) and primates (Voytko et al., 1994) have suggested that the basal forebrain is involved in the performance of attentional tasks. This issue is not directly tested in our monkeys, although in the scene task, where correct and incorrect responses are made to small objects in a complex background, animals that were severely impaired at the task did not increase the number of inappropriate responses (i.e. responses to something other than one of the two foreground objects). This indicates that they were attending normally to the objects, and not reacting to the touchscreen in an inattentive and inappropriate manner.

The Learning Impairment Is Not Due to an Agnosia

It is extremely unlikely that the crossed unilateral lesions in this study produce their impairment through a visual agnosia, i.e. simply by reducing the general activity of inferior temporal cortex by removing the cholinergic afferents to it. Although monkeys were impaired on both of the tasks reported here, as discussed above, animals performing the scene task did not increase the number of inappropriate responses made postoperatively. This implies that the animals were able to identify and accurately respond to the small alphanumeric characters against the complex visual background. Further support against an agnosic explanation comes from monkeys that have received bilateral lesions of the fornix, amygdala and anterior temporal stem (Gaffan et al., 2001). These animals had a large lesion that was not specific to any single transmitter system or fibre pathway. However, one effect of this surgery was to isolate the medial temporal lobes and inferior temporal cortex from their cholinergic afferents, which project through these three main pathways (Kitt et al., 1987; Seldon et al., 1998). All these animals exhibited a severe anterograde amnesia. One animal from this group was tested on its retention of 100 visual discrimination problems taught preoperatively. This animal reached a 90% criterion in only three trials with each problem, despite being severely impaired at learning just 10 new problems. A similar dissociation between new learning impairments and normal retention of pre-operatively taught material can also be seen in monkeys where the modulation of the basal forebrain by the midbrain is disrupted by lesion of the medial forebrain bundle (Easton and Gaffan, 2000). This implies that the medial temporal lobe and inferior temporal cortex can be completely isolated from their cholinergic afferents without the monkey becoming visually agnosic. This contrasts sharply with the effects of direct damage to the inferior temporal cortex (Gaffan et al., 1986).

A Cholinergic Explanation of Medial Temporal Lobe Amnesia

The results support the hypothesis, outlined in the introduction, that monkeys with section of the fornix, amygdala and anterior temporal stem show a severe anterograde amnesia which is a result of isolating the inferior temporal cortex and medial temporal lobe from their cholinergic afferents. However, the present experiment is unable to dissociate between the fornix section being necessary in addition to a cholinergic deafferentation of the cortex, or as simply isolating the medial temporal lobe from its cholinergic afferents. Section of the fornix, amygdala and anterior temporal stem models the surgical damage that results in severe amnesia in humans following medial temporal lobe surgery (Horel, 1978; Gaffan et al., 2001). Medial temporal lobe surgery generally involves removal of, or damage to, the amygdala and anterior temporal stem white matter, which are sectioned in the study of Gaffan et al. In these patients, the hippocampus is also damaged, which is modelled by bilateral section of the fornix in the monkeys of Gaffan et al. In contrast to the model of medial temporal lobe amnesia, the loss of cholinergic cells in the basal forebrain in the present study parallels the pathology observed in patients with severe memory deficits in Alzheimer's disease (Whitehouse et al., 1982; Bierer et al., 1995). We put forward the hypothesis, therefore, that memory loss in both Alzheimer's disease and following medial temporal lobe surgery is a result of the same underlying isolation of the inferior temporal cortex and medial temporal lobes from their cholinergic afferents, possibly in addition to an interruption of the hippocampal–anterior thalamic circuit.

Table 1
Animal Group Dose of saporin (ng/site) Number of errors in scene learning trials 2–8 Object–reward learning ACh loss (%) Cavitation score 
   Pre-op Post-op 1 Post-op 2    
ACh+fx  50 6.4 5.7 40.4 172.3 68.7 
ACh+fx 100 3.7 5.4 41.4 111.0 95.0 14 
ACh+fx 100 4.5 7.4 37.9 206.0 66.6 
ACh+fx 200 7.6 5.4 42.9 290.5 84.8 
ACh+fx 200 8.0 9.0 45.2 212.7 97.1 
ACh+fx 300 11.4 14.5 46.4 343.5 93.8 21 
ACh+fx 500 3.1 5.0 44.4 139.3 97.3 11 
Con saline 5.4 5.4 18.6 103.7 – 
Con saline 3.0 3.9 16.5 41.3 – 
Animal Group Dose of saporin (ng/site) Number of errors in scene learning trials 2–8 Object–reward learning ACh loss (%) Cavitation score 
   Pre-op Post-op 1 Post-op 2    
ACh+fx  50 6.4 5.7 40.4 172.3 68.7 
ACh+fx 100 3.7 5.4 41.4 111.0 95.0 14 
ACh+fx 100 4.5 7.4 37.9 206.0 66.6 
ACh+fx 200 7.6 5.4 42.9 290.5 84.8 
ACh+fx 200 8.0 9.0 45.2 212.7 97.1 
ACh+fx 300 11.4 14.5 46.4 343.5 93.8 21 
ACh+fx 500 3.1 5.0 44.4 139.3 97.3 11 
Con saline 5.4 5.4 18.6 103.7 – 
Con saline 3.0 3.9 16.5 41.3 – 
Figure 1.

Timeline of experimental procedures for groups ACh+fx and Con.

Figure 1.

Timeline of experimental procedures for groups ACh+fx and Con.

Figure 2.

Vertical hatching in the two drawings shows the intended inferior temporal cortical ablation on the lateral (top) and ventral (bottom) aspects of the right temporal lobe. Within these areas the cortex was removed from sulci as well as from the surface of the hemisphere. IOS: inferior occipital sulcus; LS: lateral sulcus; OTS: occipitotemporal sulcus; RS: rhinal sulcus; STS: superior temporal sulcus.

Figure 2.

Vertical hatching in the two drawings shows the intended inferior temporal cortical ablation on the lateral (top) and ventral (bottom) aspects of the right temporal lobe. Within these areas the cortex was removed from sulci as well as from the surface of the hemisphere. IOS: inferior occipital sulcus; LS: lateral sulcus; OTS: occipitotemporal sulcus; RS: rhinal sulcus; STS: superior temporal sulcus.

Figure 3.

Cresyl-stained histological sections. (A) Section at the level of the nucleus of the diagonal band in animal E. The arrow on the right (inferior temporal cortex lesioned hemisphere) shows stained cells within the nucleus. The arrow on the left indicates the lack of stained cells in the nucleus of the diagonal band in the hemisphere with the immunotoxic lesion. (B) The arrow on the right indicates normal, stained cells within the nucleus Basalis of Meynert in animal E. The arrow on the left indicates the lack of cells in this nucleus in the hemisphere with the immunotoxic lesion, and some small degree of cavitation. (C) Section from animal E. The arrow indicates cavitation within the basal forebrain and basal ganglia in the region of the immunotoxic lesion. (D) Section from animal E. The arrow indicates the region of cavitation and cell loss within the putamen and nucleus Basalis of Meynert. (E) Section from animal B. The arrow indicates an area of extensive cavitation within the basal ganglia and the basal forebrain (this section scored a maximum cavitation score of 5 in both the basal ganglia and basal forebrain). (F) Section from animal A. The arrow indicates degeneration of the fornix in the hemisphere with the immunotoxic lesion of the basal forebrain following section of the descending column of the fornix in that hemisphere. AC: anterior commisure; CA: caudate; FX: fornix; IN: insula; OT: optic tract; PU: putamen.

Figure 3.

Cresyl-stained histological sections. (A) Section at the level of the nucleus of the diagonal band in animal E. The arrow on the right (inferior temporal cortex lesioned hemisphere) shows stained cells within the nucleus. The arrow on the left indicates the lack of stained cells in the nucleus of the diagonal band in the hemisphere with the immunotoxic lesion. (B) The arrow on the right indicates normal, stained cells within the nucleus Basalis of Meynert in animal E. The arrow on the left indicates the lack of cells in this nucleus in the hemisphere with the immunotoxic lesion, and some small degree of cavitation. (C) Section from animal E. The arrow indicates cavitation within the basal forebrain and basal ganglia in the region of the immunotoxic lesion. (D) Section from animal E. The arrow indicates the region of cavitation and cell loss within the putamen and nucleus Basalis of Meynert. (E) Section from animal B. The arrow indicates an area of extensive cavitation within the basal ganglia and the basal forebrain (this section scored a maximum cavitation score of 5 in both the basal ganglia and basal forebrain). (F) Section from animal A. The arrow indicates degeneration of the fornix in the hemisphere with the immunotoxic lesion of the basal forebrain following section of the descending column of the fornix in that hemisphere. AC: anterior commisure; CA: caudate; FX: fornix; IN: insula; OT: optic tract; PU: putamen.

Figure 5.

Performance on scene learning task. The left panel shows the average group learning curve over 10 sessions of the scene task. Open squares = combined group pre-operative performance; open circles = combined performance after one unilateral lesion; filled squares = group Con; filled circles = group ACh+fx. The right panel shows the percent error in trials 2–8 on the same task. Performance of individual animals is shown by their letter, except for filled square = H and filled circle = I (both group Con). Bars represent average group performance: PRE = combined group performance pre-operatively; unilateral = combined group performance after only a single unilateral lesion; ACh+fx = performance of group ACh+fx and Con = performance of group Con.

Figure 5.

Performance on scene learning task. The left panel shows the average group learning curve over 10 sessions of the scene task. Open squares = combined group pre-operative performance; open circles = combined performance after one unilateral lesion; filled squares = group Con; filled circles = group ACh+fx. The right panel shows the percent error in trials 2–8 on the same task. Performance of individual animals is shown by their letter, except for filled square = H and filled circle = I (both group Con). Bars represent average group performance: PRE = combined group performance pre-operatively; unilateral = combined group performance after only a single unilateral lesion; ACh+fx = performance of group ACh+fx and Con = performance of group Con.

Figure 6.

Performance on concurrent visual object–reward association learning task. The panel shows the average group performance learning three sets of 10 discrimination problems concurrently. Symbols and group names are the same as Figure 5.

Performance on concurrent visual object–reward association learning task. The panel shows the average group performance learning three sets of 10 discrimination problems concurrently. Symbols and group names are the same as Figure 5.

This research was supported by the UK Medical Research Council. We thank Judi Wakeley for help in training the monkeys.

References

Abe K, Inokawa M, Kashiwagi A, Yanagihara T (
1998
) Amnesia after a discrete basal forebrain lesion.
J Neurol Neurosurg Psychiatry
 
65
:
126
–130.
Aggleton JP, Brown MW (
1999
) Episodic memory, amnesia and the hippocampal–anterior thalamic axis.
Behav Brain Sci
 
22
:
425
–444.
Aggleton JP, McMakin D, Carpenter K, Hornak J, Kapur N, Halpin S, Wiles CM, Kamel H, Brennan P, Carton S, Gaffan D (
2000
) Differential cognitive effects of colloid cysts in the third ventricle that spare or compromise the fornix.
Brain
 
123
:
800
–815.
Aigner TG, Mitchell SJ, Aggleton JP, DeLong MR, Struble RG, Price DL, Wenk GL, Pettigrew KD, Mishkin M (
1991
) Transient impairment of recognition memory following ibotenic-acid lesions of the basal forebrain in macaques.
Exp Brain Res
 
86
:
18
–26.
Barefoot HC, Baker HF, Ridley RM (
2000
) Synergistic effects of unilateral immunolesions of the cholinergic basal forebrain and contralateral ablations of the inferior temporal cortex and hippocampus in monkeys.
Neuroscience
 
98
:
243
–251.
Baxter MG, Bucci DJ, Gorman LK, Wiley RG, Gallagher M (
1995
) Selective immunotoxic lesions of basal forebrain cholinergic cells: effects on learning and memory in rats.
Behav Neurosci
 
109
:
714
–722.
Baxter MG, Holland PC, Gallagher M (
1997
) Disruption of decrements in conditioned stimulus processing by selective removal of hippocampal cholinergic input.
J Neurosci
 
17
:
5230
–5236.
Bierer LM, Haroutunian V, Gabriel S, Knott PJ, Carlin LS, Purohit DP, Perl DP, Schmeidler J, Kanof P, Davis KL (
1995
) Neurochemical correlates of dementia severity in Alzheimer's disease: relative importance of the cholinergic deficits.
J Neurochem
 
64
:
749
–760.
Cirillo RA, Horel JA, George PJ (
1989
) Lesions of the anterior temporal stem and the performance of delayed match-to-sample and visual discriminations in monkeys.
Behav Brain Res
 
34
:
55
–69.
Corkin S, Amaral DG, Gonzalez RG, Johnson KA, Hyman BT (
1997
) H.M.'s medial temporal lobe lesion: findings from magnetic resonance imaging.
J Neurosci
 
17
:
3964
–3979.
Damasio AR, Graff-Radford NR, Eslinger PJ, Damasio H, Kassell N (
1985
) Amnesia following basal forebrain lesions.
Arch Neurol
 
42
:
263
–271.
Dunn LT, Everitt BJ (
1988
) Double dissociations of the effects of amygdala and insular cortex lesions on conditioned taste aversion, passive avoidance, and neophobia in the rat using the excitotoxin ibotenic acid.
Behav Neurosci
 
102
:
3
–23.
Dunnett SB, Everitt BJ, Robbins TW (
1991
) The basal forebrain–cortical cholinergic system: interpreting the functional consequences of excitotoxic lesions.
Trends Neurosci
 
14
:
494
–501.
Easton A, Gaffan D (
2000
) Comparison of perirhinal cortex ablation and crossed unilateral lesions of medial forebrain bundle from inferior temporal cortex in the Rhesus monkey: effects on learning and retrieval.
Behav Neurosci
 
114
:
1041
–1057.
Easton A, Gaffan D (
2001
) Crossed unilateral lesions of the medial forebrain bundle and either inferior temporal or frontal cortex impair object–reward association learning in Rhesus monkeys.
Neuropsychologia
 
39
:
71
–82.
Ebeling U, Cramon DV (
1992
) Topography of the uncinate fascicle and adjacent temporal fibre tracts.
Acta Neurochir
 
115
:
143
–148.
Fine A, Hoyle C, Maclean CJ, Levatte TL, Baker HF, Ridley RM (
1997
) Learning impairments following injection of a selective cholinergic immunotoxin, ME20.4 IgG–Saporin, into the basal nucleus of Meynert in monkeys.
Neuroscience
 
81
:
331
–343.
Frisoni GB, Laakso MP, Beltramello A, Geroldi C, Bianchetti A, Soininen H, Trabucchi M (
1999
) Hippocampal and entorhinal cortex atrophy in frontotemporal dementia and Alzheimer's disease.
Neurology
 
52
:
91
–100.
Gaffan D (
1994
) Scene-specific memory for objects: a model of episodic memory impairment in monkeys with fornix transection.
J Cogn Neurosci
 
6
:
305
–320.
Gaffan D, Gaffan EA (
1991
) Amnesia in man following transection of the fornix: a review.
Brain
 
114
:
2611
–2618.
Gaffan D, Parker A (
1996
) Interaction of perirhinal cortex with the fornix-fimbria: memory for objects and object-in-place memory.
J Neurosci
 
16
:
5864
–5869.
Gaffan D, Parker A, Easton A (
2001
) Dense amnesia in the monkey after transection of fornix, amygdala and anterior temporal stem.
Neuropsychologia
 
39
:
51
–70.
Gaffan EA, Harrison S, Gaffan D (
1986
) Single and concurrent discrimination learning by monkeys after lesions of inferotemporal cortex.
Q J Exp Psychol
 
38B
:
31
–51.
Horel JA (
1978
) The neuroanatomy of amnesia: a critique of the hippocampal memory hypothesis.
Brain
 
101
:
403
–445.
Kitt CA, Mitchell SJ, DeLong MR, Wainer BH, Price DL (
1987
) Fiber pathways of basal forebrain cholinergic neurons in monkeys.
Brain Res
 
406
:
192
–206.
Maclean CJ, Baker HF, Fine A, Ridley RM (
1997
) The distribution of p75 neurotrophin receptor-immunoreactive cells in the forebrain of the common marmoset (Callithrix jacchus).
Brain Res Bull
 
43
:
197
–208.
Maclean CJ, Gaffan D, Baker HF, Ridley RM (
2001
) Visual discrimination learning impairments produced by combined transections of the anterior temporal stem, amygdala and fornix transection in marmoset monkeys.
Brain Res
 
888
:
34
–50.
Malkova L, Gaffan D, Murray EA (
1997
) Excitotoxic lesions of the amygdala fail to produce impairment in visual learning for auditory secondary reinforcement but interfere with reinforcer devaluation effects in rhesus monkeys.
J Neurosci
 
17
:
6011
–6020.
Martin RF, Bowden DM (
1996
) A stereotaxic template atlas of the macaque brain for digital imaging and quantitative neuroanatomy.
Neuroimage
 
4
:
119
–150.
Mesulam M-M, Mufson EJ (
1984
) Neural inputs into the nucleus basalis of the substantia innominata (Ch4) in the rhesus monkey.
Brain
 
107
:
253
–274.
Mrzljak L, Levey AI, Belcher S, Goldman-Rakic PS (
1998
) Localization of the m2 muscarinic acetylcholine receptor protein and mRNA in cortical neurons of the normal and cholinergically deafferented rhesus monkey.
J Comp Neurol
 
390
:
112
–132.
Parker A, Gaffan D (
1997
) The effect of anterior thalamic and cingulate cortex lesions on object-in-place memory in monkeys.
Neuropsychologia
 
35
:
1093
–1102.
Parker A, Gaffan D (
1997
) Mamillary body lesions in monkeys impair object-in-place memory: functional unity of the fornix–mamillary system.
J Cogn Neurosci
 
9
:
512
–521.
Parker A, Gaffan D (
1998
) Interaction of frontal and perirhinal cortices in visual object recognition memory in monkeys.
Eur J Neurosci
 
10
:
3044
–3057.
Ridley RM, Thornley HD, Baker HF, Fine A (
1991
) Cholinergic neural transplants into hippocampus restore learning ability in monkeys with fornix transections.
Exp Brain Res
 
83
:
533
–538.
Ridley RM, Timothy CJ, Maclean CJ, Baker HF (
1995
) Conditional learning and memory impairments following neurotoxic lesion of the CA1 field of the hippocampus.
Neuroscience
 
67
:
263
–275.
Ridley RM, Barefoot HC, Maclean CJ, Pugh P, Baker HF (
1999
) Different effects on learning ability after injection of the cholinergic immunotoxin ME20.4IgG–Saporin into the diagonal band of Broca, basal nucleus of Meynert, or both in monkeys.
Behav Neurosci
 
113
:
303
–315.
Ridley RM, Pugh P, Maclean CJ, Baker HF (
1999
) Severe learning impairment caused by combined immunotoxic lesion of the cholinergic projections to the cortex and hippocampus in monkeys.
Brain Res
 
836
:
120
–138.
Scoville WB, Milner B (
1957
) Loss of recent memory after bilateral hippocampal lesions.
J Neurol Neurosurg Psychiatry
 
20
:
11
–21.
Seldon NR, Gitelman DR, Salamon-Murayama N, Parrish TB, Mesulam MM (
1998
) Trajectories of cholinergic pathways within the cerebral hemispheres of the human brain.
Brain
 
121
:
2249
–2257.
Voytko ML, Olton DS, Richardson RT, Gorman LK, Tobin JR, Price DL (
1994
) Basal forebrain lesions in monkeys disrupt attention but not learning and memory.
J Neurosci
 
14
:
167
–186.
Whitehouse PJ, Price DL, Struble RG, Clark AW, Coyle JT, Delon MR (
1982
) Alzheimer's disease and senile dementia: loss of neurons in the basal forebrain.
Science
 
215
:
1237
–1239.
Wiley RG, Oeltmann TN, Lappi DA (
1991
) Immunolesioning: selective destruction of neurons using immunotoxin to rat NGF receptor.
Brain Res
 
562
:
149
–153.