2 mouse model of Huntington’s disease

Available online at www.sciencedirect.com

Neurobiology of Learning and Memory 89 (2008) 533–544 www.elsevier.com/locate/ynlme

Rigidity in social and emotional memory in the R6/2 mouse model of Huntington’s disease Alessandro Ciamei, A. Jennifer Morton

*

Department of Pharmacology, University of Cambridge, Tennis Court Road, CB2 1PD Cambridge, UK Received 3 August 2007; revised 22 October 2007; accepted 26 October 2007 Available online 20 February 2008

Abstract Four experiments were conducted to examine social and emotional memory in the R6/2 transgenic mouse model of Huntington’s disease. First, R6/2 mice were tested in a social transmission of food preference task where they had to acquire a preference for a flavoured food (acquisition) and subsequently to learn a preference for a different flavour (shifted reinforcement). R6/2 mice performed well in the acquisition trial. However, they were impaired in the shifted reinforcement trial and perseverated on the first preference learned. Second, mice were trained in an inhibitory avoidance paradigm, with either one or two footshocks delivered during the training. WT mice given one footshock showed retention levels lower than those of mice trained with two footshocks. By contrast, there was no difference in retention levels of R6/2 mice given either one or two footshocks. Third, mice were tested in an active avoidance task that paired a mild footshock with a warning light. R6/2 mice had a strong age-dependent deficit in this task. Finally, mice were tested in a conditioned taste aversion task that paired a saccharine solution with a nausea-inducing agent (LiCl). R6/2 mice displayed normal aversion, however this was not extinguished following repeated exposure to saccharine solution alone. Our data show that while R6/2 mice have functional hippocampus-based memory, they have deficits in striatum-based memory skills. Further, social and emotional memories appear to be encoded in a rigid way that is not influenced by subsequent learning or by arousal levels.  2007 Elsevier Inc. All rights reserved. Keywords: Huntington’s disease; R6/2 mice; Social transmission of food preference; Inhibitory avoidance; Memory systems

1. Introduction HD is a progressive neurodegenerative disorder caused by an expanded CAG repeat in the HD gene. Symptoms in HD are associated with profound neuronal loss in the striatum and cortex (The Huntington’s Disease Collaborative Research Group, 1993). Clinical symptoms of HD include motor abnormalities, cognitive decline and emotional disturbance (Bates, Harper, & Jones, 2002). HD patients also frequently exhibit psychiatric symptoms. These include deficits related to the processing of emotion (that are mainly due to a failure in the recognition of facial expression; Sprengelmeyer, Schroeder, Young, & Epplen,

*

Corresponding author. Fax: +44 01223 334100. E-mail address: [email protected] (A.J. Morton).

1074-7427/$ - see front matter  2007 Elsevier Inc. All rights reserved. doi:10.1016/j.nlm.2007.10.009

2006), and to the ability to draw correct inferences from social situations (Snowden et al., 2003). The R6/2 transgenic mouse model of HD is widely used because these mice show a progressive deterioration of locomotor and cognitive abilities (Carter et al., 1999; Lione et al., 1999; Murphy et al., 2000). R6/2 mice express the Nterminal fraction of the human HD gene containing a highly expanded CAG repeat (Mangiarini et al., 1996). However, in contrast to what is found in human patients, neuronal loss is only observed in R6/2 mice at a very late stage of the disease (Stack et al., 2005), whereas some cognitive deficits are present as early as 4 weeks of age (Lione et al., 1999). Further, abnormalities in hippocampal synaptic plasticity are present from 3 weeks of age (Gibson, Reim, Brose, Morton, & Jones, 2005). Therefore, cognitive impairments are unlikely to be due to neuronal degeneration, but rather are due to a ‘functional’ failure in neuronal

534

A. Ciamei, A.J. Morton / Neurobiology of Learning and Memory 89 (2008) 533–544

mechanisms. The degree of cognitive decline depends upon the task. ‘Hippocampal’ tasks, such as spatial learning (Lione et al., 1999; Murphy et al., 2000) and acquisition of two-choice discrimination learning (Lione et al., 1999; Morton, Skillings, Bussey, & Saksida, 2006; Pallier et al., 2007) are impaired in mice with mid-late stage disease (older than 12 weeks), whereas reversal learning deficits that depend upon intact striatum are seen much earlier (by 6–8 weeks of age), before the development of abnormal locomotor signs. From these studies it seems likely that cognitive abilities of these mice are affected by a malfunction in frontostriatal and hippocampal memory systems. Different brain regions, such as the medial temporal lobe, the basal ganglia or the amygdaloid complex, are known to be important for the consolidation of different forms of memory (e.g. declarative, procedural and emotional memory; Gold, 2004; McGaugh, 2004; Packard & McGaugh, 1996). ‘Emotional’ memory in R6/2 mice has been investigated in a single study, with deficit observed in R6/2 mice tested in a fear-conditioning paradigm (Bolivar, Manley, & Messer, 2003). Interestingly, the deficit was seen only in the contextual version of the task, and not in the cued one, suggesting a deficit in contextual memory rather than in the formation of memories for emotionally-relevant stimuli. Social cognition has not been yet investigated in the R6/2 mouse. Therefore, in this study, we tested R6/2 mice in tasks of social and emotional memory. We first examined social memory in R6/2 mice using a social transmission of food preference (STFP) task (Eichenbaum, 2000). In this task, information about a safe food (cued food) eaten by a demonstrator is transmitted to other mice through an olfaction-based social interaction. This form of social learning is a highly-preserved biological mechanism through which a colony of mice can avoid potentially unsafe foods on the basis of the experience of a single mouse (Sa`nchez-Andrade, Bronwen, & Kendrick, 2005). It has been shown to be largely dependent on an intact functionality of structures within the medial temporal lobe, including the hippocampus, the subiculum and the parahippocampal cortex (Alvarez, Lipton, Melrose, & Eichenbaum, 2001; Alvarez, Wendelken, & Eichenbaum, 2002; Ross & Eichenbaum, 2006). As with other forms of learning relating to individual survival (e.g. conditioned taste aversion; see below), the information learned is quickly stored in long-term memory, so that a single exposure is often sufficient to achieve memory consolidation (Welzl, D’Adamo, & Lipp, 2001). In a second set of experiments, we tested R6/2 mice in a one-trial inhibitory avoidance paradigm. In this task, mice learn to avoid a place where they had previously received a footshock. Consolidation in long-term memory of this aversive experience is known to engage both hippocampus and striatum (Cammarota, Bevilaqua, Ko¨hler, Medina, & Izquierdo, 2005; Packard, Vecchioli, Schroeder, & Gasbarri, 2001; Solana-Figueroa, Salado-Castillo, Galinda, Quirarte, & Prado-Alcala`, 2002). Moreover, consolidation of this form of memory is normally modulated by the activity

of the basolateral nucleus of the amygdala (McGaugh, 2004). Through its main efferent path, the stria terminalis, the basolateral nucleus of the amygdala can influence the consolidation process in the structures in which it is taking place, according to the level of emotional arousal induced by the aversive stimulus used (McGaugh, 2004; Wilensky, Schafe, & LeDoux, 2000). In a third set of experiments, we tested R6/2 mice in a two-way active avoidance paradigm. This multi-trial paradigm is a stimulus–response (S–R) learning task in which a light is used as conditioned stimulus (CS) and a footshock as unconditioned stimulus (US). The association between CS and US is learned quickly by the mice, and, following extensive training, is turned into a conditioned habit through which mice avoid the footshock using the CS as a predictor of the US. Contrary to what happens for the other procedures previously described, this task is mainly a response learning task, and as such largely depends on an intact striatum (Packard & McGaugh, 1996; Ve´csei & Beal, 1991). Moreover, it has been shown that this form of fear-based conditioning also involves the amygdala (Roozendaal, Koolhaas, & Bohus, 1993) and requires the release of dopamine in the medial prefrontal cortex (Stark, Bischof, Wagner, & Scheich, 2001). In a fourth set of experiments, mice were tested in a conditioned taste aversion (CTA) paradigm. In this task, mice learn to reject a tastant (a solution of saccharine that represents the CS) if this is associated with subsequent malaise (US) induced through the injection of a nausea-inducing agent (LiCl). Following the initial learning trial, mice were tested in an extinction paradigm in which only the CS was presented over multiple subsequent trials. Although many brain regions are involved in CTA learning, particularly those related to the processing of visceral inputs to the brain and of gustatory information, a key role in this task is played by the amygdala (Dudai, 2002; Josselyn, Kida, & Silva, 2004; Welzl et al., 2001). Moreover, the amygdala has been shown to be involved in the formation of extinction memory for this paradigm (Bahar, Dorfman, & Dudai, 2004; Bahar, Samuel, Havzi, & Dudai, 2003), together with regions within the ventro-medial prefrontal cortex (vmPFC; Akirav et al., 2006; Mickley, Kenmuir, Yocom, Justin, & Biada, 2005). 2. Methods

2.1. Animals Mice were taken from a colony established at the University of Cambridge on a CBA · C57BL/6 F1 background as described previously (Morton et al., 2005). Mice were housed in single sex, mixed genotype groups of 10 mice. All testing was performed on male mice. WT littermates were used as controls for R6/2 mice. Means (±SEM) CAG repeat length for R6/2 mice was 280 ± 2. Genotyping and repeat length measurements were performed by Laragen, Los Angeles, USA. All studies were carried out in accordance with the UK Animals (Scientific Procedures) Act 1986. Mice were housed within a 12-h light/dark cycle (lights on at 7:30 AM and off at 7:30 PM) in a temperature-controlled (19–21 C) and

A. Ciamei, A.J. Morton / Neurobiology of Learning and Memory 89 (2008) 533–544 humidity-controlled (45 ± 10%) environment. Dry food, mash, and water were available ad libitum as described previously by Carter, Hunt, and Morton (2000). All experiments were carried out during the light phase of the cycle. In the present study, we used R6/2 mice carrying a CAG repeat length of 280. In humans, the number of CAG repeats has been shown to be inversely correlated to the age of onset and directly correlated to several pathological features, such as the degree of caudate atrophy (Bates et al., 2002). In mice models of HD such linear correlations have not been observed. Indeed, preliminary observations from our lab suggest that repeat length might affect the severity of the disease in a non-linear way, with neuropathological features becoming more aggressive as the repeat length increases up to 150–160 CAG repeats, whereas longer repeat lengths induce a less severe phenotype. By the twelfth week of age, R6/2 mice with CAG repeats of 250–300 do not show the overt symptoms normally displayed by mice with shorter CAG repeats at that age, such as hind limb grooming, muscle wasting, lordokyphosis, loss of body weight, and infertility (A.J.M., unpublished observations). Although our mice can be considered as ‘pre-symptomatic’ at 12 weeks of age, we decided to also test them at ages similar to those in which R6/2 mice carrying shorter CAG repeats are normally considered to be at ‘presymptomatic’ (8 weeks) and ‘symptomatic’ (12 weeks) stages of the disease.

2.2. Open field To measure motor activity, two groups of naı¨ve mice were tested in an open field (a white Perspex circular arena; 1 m diameter), one at 8 weeks (WT, n = 10; R6/2, n = 14) and the other at 12 weeks (WT, n = 10; R6/ 2, n = 23) of age. An HVS tracker system (HVS Image 2020, Hampton, UK) was used to monitor path length, speed, time spent in the periphery (thigmotaxis) and in the centre of the arena during 5 min. Four halogen lamps (300 W) placed on the outer walls of the arena at the floor level provided a condition of low luminosity used during the behavioural measurements.

2.3. Social transmission of food preference Two groups of naı¨ve mice were tested, one at 8 weeks (WT, n = 12; R6/2, n = 10) and the other at 12 weeks (WT, n = 16; R6/2, n = 18) of age. Mice were food deprived overnight for 18 h, singly housed in polypropylene cages (30 · 12 · 16 cm). Water was available ad libitum. During the two days preceding olfactory training, mice were trained to eat standard ground food from a Petri dish (9 cm diameter) (shaping). Preference scores for the flavours used were calculated as amount of flavoured food eaten (g) · 100/total food eaten (g). The STFP experiment was run in two parts, with an ‘acquisition’ followed by a ‘shifted reinforcement’. The acquisition protocol consisted of three phases (Kogan et al., 1997). In the first phase, a food-deprived WT mouse from each home cage (demonstrator) was allowed to eat for 1 h from a dish containing either cinnamon-flavoured (1% per weight) or cocoa-flavoured (2% per weight) food. The use of cinnamon or cocoa for the cued food was counterbalanced across demonstrators. In the second phase an interaction session between the demonstrators and their cagemates (observers) took place inside the home cages immediately after the first phase. Free interaction lasted for 30 min, after which time the demonstrator was removed from the cage. In the third phase, food preference of food-deprived observers was tested 24 h after the demonstrator/ observers interaction session. Each mouse was placed into a cage containing two dishes placed at opposite corners of an experimental cage (45 · 12 · 28 cm) with no sawdust or access to water. One dish contained the cued food and the other contained the non-cued food. Mice were allowed to eat for 30 min. After the third phase, all mice were kept singly-housed with water and food available ad libitum for 2 days. At the end of this period, all mice were returned to their original home cage for two more days after which the shifted-reinforcement trial began. The procedure for the shifted-reinforcement trial was identical to the acquisi-

535

tion trial described above, except that cumin-flavoured food (0.25% in weight) was used as cued food for all mice, while the flavour previously used as the cued food (either cocoa 2% or cinnamon 1%) was used as the non-cued food. The flavours concentrations we used have been shown to possess similar levels of palatability (Alvarez et al., 2002; Kogan et al., 1997). However, we decided to test palatability of these flavours directly in our mice using a test of natural preference. To do this, after a 2 days shaping period (as above), different groups of food-deprived 12 week old naı¨ve mice were given a choice of either cocoa and cumin (WT, n = 13; R6/2, n = 12), or cinnamon and cumin (WT, n = 12; R6/2, n = 10) or cocoa and cinnamon (WT, n = 10; R6/2, n = 13) in pre-weighed Petri dishes, and were allowed to eat for 30 min. At the end of this period, mice were returned to their home cages, dishes were weighed and the amount of food eaten calculated as amount of flavoured food eaten (g) · 100/total food eaten (g). Olfactory discrimination was measured in an acquisition trial (as described above) using reduced concentrations of the relevant flavour. Two groups of 12 week old naı¨ve mice were trained with flavours at 75% (WT n = 10; R6/2 n = 13), 50% (WT n = 10; R6/2 n = 12), 25% (WT n = 10; R6/2 n = 15) or 12.5% (WT n = 13; R6/2 n = 12) of the concentration used in the acquisition trial. In both the natural preference test and olfactory discrimination test, 12 week old mice were used on the assumption that a ‘normal’ performance of 12 week old R6/2 mice in these tasks would be preceded by a ‘normal’ performance of younger mice.

2.4. Inhibitory avoidance The inhibitory avoidance apparatus consisted of two chambers divided by a sliding door (Panlab, Cornella`, Spain). The start chamber was triangular in shape and made of white Plexiglas. The shock chamber was rectangular, made of black Plexiglas, and equipped with a removable cover. The floor of the dark chamber was a stainless steel grid through which the shock could be delivered. A tensor lamp (60 W, positioned 40 cm above the apparatus) illuminated the starting chamber. Behavioural training was performed as previously described (Ciamei, Cestari, & Castellano, 2000). Briefly, on the training day, each mouse was placed in the start chamber, facing the corner opposite to the sliding door. When the mouse turned around, the door leading to the dark chamber was opened. When the mouse had stepped with all four paws into the dark side, the door was closed, and the footshock delivered. The latency to enter the dark compartment (step-through latency) was then recorded, and the mouse removed from the apparatus. The latency to enter the dark compartment was measured again 24 h later following a procedure similar to that of training, except that no footshock was administered (retention test). Two groups of naı¨ve mice were trained, one at 8 weeks (WT, n = 9; R6/ 2, n = 8) and the other at 12 weeks (WT, n = 14; R6/2, n = 10) using a single footshock (0.1 mA, 1 s). Two further groups of mice were trained at 8 weeks (WT, n = 16; R6/2, n = 21) and 12 weeks (WT, n = 9; R6/2, n = 10) using two footshocks (0.1 mA, 1 s) administered with an inter-shock interval of 5 s. In all experiments a cut-off to step-through of 300 s was used.

2.5. Active avoidance Behavioural training was performed as previously described (Mazzucchelli et al., 2002). Active avoidance was carried out in two-way shuttle-boxes (Panlab, Cornella`, Spain) placed inside soundproof boxes in order to prevent unwanted association with external noise during the task and to carry out the sessions in a dark environment. Each shuttle-box was divided into two compartments by a partition with an opening at the floor level connecting the two compartments. Each compartment had a light bulb attached on top of one of the internal walls. The floor of each compartment was a grid through which a footshock (unconditioned stimulus, US) could be delivered. The light was switched on alternately in the two compartments and used as a conditioned stimulus (CS). During each trial, the onset of the CS preceded the onset of the US by 5 s, and overlapped it

536

A. Ciamei, A.J. Morton / Neurobiology of Learning and Memory 89 (2008) 533–544

for 25 s. At the end of the 30 s period, both CS and US were automatically terminated, and a new trial begun in the other compartment. The US was an electric shock (0.1 mA) continuously applied to the grid floor. An avoidance response was recorded when the animal avoided the US by running into the dark compartment within 5 s after the onset of the CS. Animals were subjected to 80 trials/day sessions for 5 days. Different groups of naı¨ve mice were tested at 8 weeks (WT n = 8; R6/2 n = 9), 10 weeks (WT n = 9; R6/2 n = 13) and 12 weeks of age (WT n = 9; R6/2 n = 11).

2.6. Conditioned taste aversion Four groups of 10 week old mice were tested in the CTA experiment. Groups were defined on the base of the treatment received during the conditioning phase of the experiment (see below). Control mice (WT n = 6; R6/2 n = 7) received an intra-peritoneal (i.p.) injection of saline (NaCl 0.9%; volume 2% body weight) while conditioned mice (WT n = 12; R6/ 2 n = 13) received an i.p. injection of LiCl (0.2 M; 2% b.w.). Throughout both the CTA and the extinction experiments, mice were tested in rectangular polypropylene cages (60 · 12 · 16 cm). Each cage was equipped with a metal grid that served as a lid and in which two slots were present on the shorter side (front side) for bottles allocation. Throughout both the CTA and the extinction experiments, mice were kept in water deprivation overnight for 16 h inside their own home cages. CTA procedure consisted of three distinct phases, shaping, conditioning and testing. During the shaping phase, water-deprived mice were placed in experimental cages and allowed to drink from two bottles. On day 1, mice were allowed to drink for 1 h; on day 2, mice were allowed to drink for 45 min; on days 3–6, mice were allowed to drink for 30 min. After each drinking session mice were returned to their home cages where water and food were available ad libitum. The conditioning phase took place on day 7. Water-deprived mice were placed in the experimental cages and allowed to drink for 15 min from a single bottle containing a saccharine sodium solution in water (0.1%). Bottle positions for all experimental cages were randomly assigned and were counterbalanced between the left and right slots of the lids. Fifteen minutes after the end of the drinking session, mice were removed from the cages and received an i.p. injection of either saline or a LiCl solution, according to the group to which they had been assigned. Two hours after the treatment, each mouse was placed back in the experimental cage and given 30 min access to normal water from a single bottle to prevent dehydration. After this, mice were returned to their home cages where water and food were available ad libitum. Bottles were weighed before and after the drinking session, and the amount of saccharine sodium solution consumed measured. The testing of CTA took place on day 8. Water-deprived mice were placed in the experimental cages. Each cage had two bottles, one containing the saccharine sodium solution and the other normal tap water. The bottle position in the cages was reversed respect to the one used during the conditioning phase, in order to prevent the development of unwanted associations between the US and the spatial allocation of the bottle containing the CS rather than the CS itself. During the test of CTA, mice were allowed to drink for 40 min and subsequently were returned to the home cage, where water and food were available ad libitum. Bottles were weighed before and after the drinking session, and the amount of saccharine sodium solution and water consumed were calculated as amount of liquid drunk (g) · 100/total liquid drunk (g). Extinction of the conditioned aversion to the saccharine solution was evaluated over 14 days in which mice were subjected daily to a procedure identical to that used for the testing phase. Bottle positions in the lids were always reversed during each extinction trial so that the bottle containing the saccharine solution was never in the same slot for two consecutive days. On each extinction trial, bottles were weighed before and after the drinking session, and the amount of saccharine solution and water consumed were calculated as amount of liquid drunk (g) · 100/total liquid drunk (g). Over the first few days of ‘extinction’ there was an unexpected variability in the aversion shown by both WT and R6/2 mice. This disappeared by the third day of the extinction procedure. We attribute this variability to the fact

that we used rectangular cages in which a ‘left side’ and a ‘right side’ may have been identifiable. Although we alternated the placement of the bottles subsequently, the asymmetry might have introduced an element of ‘contextual contamination’ in this task. This did not affect the outcome, which was similar to that shown by other investigators (Bahar et al., 2003). However, we would recommend using square cages for this kind of study.

2.7. Statistical analysis Two-way ANOVA (genotype, age) was used to evaluate data from the open field. One-, two- and three-way ANOVA (genotype, age, flavour) were used to evaluate data from the STFP experiments. Where appropriate, Duncan multiple range post-hoc test was used for individual between groups comparisons. Non-parametric statistics were used to evaluate the retention performance of mice tested in the inhibitory avoidance task. Independent Kruskal–Wallis analyses of variance were computed for retention latencies. When appropriate, the Mann–Whitney U test was used to make comparisons between pairs of groups. Two-way ANOVA (genotype, day) with repeated measures was used to evaluate data from the active avoidance experiment. Where appropriate, Duncan multiple range post-hoc test was used for individual between groups comparisons. Two-way ANOVA (genotype, treatment) was used to evaluate data from the CTA experiment. Where appropriate, Bonferroni post-hoc test was used for individual between-groups comparisons. Two-way ANOVA (genotype, day) with repeated measures was used to evaluate the performance of LiCl treated mice in the extinction experiment. Univariate tests were then used to break down statistically significant effects of the dependent variable considered.

3. Results 3.1. Open field No significant effects of genotype, age or the interaction between these two variables were observed between WT and R6/2 mice aged either 8 or 12 weeks in the open field (Fig. 1). There was a trend towards a decrease in path length (p = .058) and speed (p = .09) observed for 12 week old R6/2 mice, as well as a trend for 8 week old R6/2 mice to avoid the centre of the arena (p = .075), but these did not reach statistical significance. 3.2. Social transmission of food preference Both WT and R6/2 mice aged 12 weeks showed a similar preference for the flavours used in the natural preference test (Table 1). In the acquisition phase of the STFP experiment, there was no difference between WT and R6/2 mice aged either 8 or 12 weeks in their preference for the cued food (Fig. 2A). Three-way ANOVA revealed a significant effect of flavour (F1,103 = 106.23, p < .001) but not of genotype, age or the interaction between these three variables. There were significant differences between the preference scores for cued and non-cued food in all groups tested. In the shifted-reinforcement trial, WT mice shifted their preference to the novel cued food at 8 weeks and showed a trend towards the novel cued food at 12 weeks. By contrast, at both ages R6/2 mice failed to shift their preference to the novel cued food. Rather, they perseverated in their preference for the previously cued food (Fig. 2B). Threeway ANOVA revealed a significant effect of the interaction

A. Ciamei, A.J. Morton / Neurobiology of Learning and Memory 89 (2008) 533–544

537

Fig. 1. Behavioural analysis of 8 and 12 week old WT (white columns) and R6/2 mice (black columns) in the open field. (A) Path length, (B) speed, (C) time spent in thigmotaxis, (D) time spent in the central area of the arena. Bars represent means ± SEM.

Table 1 Natural preference scores of WT and R6/2 mice for different flavour pairings Total of flavoured food eaten (mean percentage ± SEM) Cocoa

Cinnamon

Cumin

WT R6/2

46.98 ± 8.23 48.29 ± 12.07

— —

53.02 ± 8.23 51.71 ± 12.07

WT R6/2

— —

52.95 ± 5.85 51.38 ± 9.73

47.05 ± 5.85 48.62 ± 9.73

WT R6/2

53.11 ± 6.91 48.93 ± 6.97

46.89 ± 6.91 51.07 ± 6.97

— —

between genotype and flavour (F1,103 = 23.53, p < .001). No main effects were revealed for genotype, age, flavour or for any other interaction between these variables. There were significant differences between the preference scores for cued and non-cued food in all groups except the 12 week old WT group, and between the preference score for the cued food of WT and R6/2 mice aged 8 weeks. The olfactory acuity test did not reveal any olfactory deficit in R6/2 mice at 12 weeks of age (Fig. 2C). However, when the concentration of flavour was decreased to 25% and 12.5% of the initial concentration, WT mice no longer showed a preference for the cued food, whereas R6/2 mice still showed a clear preference for the cued food. Two-way ANOVA revealed a significant effect of genotype

(F1,4 = 6.11, p < .05). Significant differences were detected between WT and R6/2 mice at these flavour concentrations. In the olfactory acuity test, no differences were observed between WT and R6/2 mice in the total amount of food eaten (Fig. 2D). 3.3. Inhibitory avoidance In the inhibitory avoidance experiment, no differences were observed between WT and R6/2 mice in the stepthrough latency recorded during the training day (Table 2) at both ages tested. During testing, the number of footshocks administered during training modulated the strength of retention in WT mice but not in R6/2 mice at both 8 and 12 weeks of age (Fig. 3). For the 8 week old group (Fig. 3A), Kruskal–Wallis ANOVA revealed significant differences in retention scores among the groups (H(3) = 8.99, p < .05). Mann–Whitney U test revealed a significant difference between WT mice trained with single and double footshock, while no other differences were revealed between the other groups. For the 12 week old group (Fig. 3B), Kruskal–Wallis ANOVA revealed significant differences in retention scores among the groups (H(3) = 11.55, p < .01). Mann–Whitney U test revealed a significant difference between WT mice trained with single and double footshock and between WT and R6/2 mice trained with a single footshock. No significant differences

538

A. Ciamei, A.J. Morton / Neurobiology of Learning and Memory 89 (2008) 533–544

Fig. 2. Food preference scores of WT and R6/2 mice in the STFP. In (A) the preference scores of WT and R6/2 mice for the cued (striped columns) and non-cued (grey columns) food during the acquisition phase. In (B) the preference scores of WT and R6/2 mice for the new cued food (dotted columns) and the previously cued food (striped columns) during the shifted-reinforcement trial are shown. In (C) the preference scores of 12 week old WT (white columns) and R6/2 mice (black columns) during the olfactory acuity test are shown. In (D) the total amount of food eaten by 12 week old WT (white columns) and R6/2 mice (black columns) during the olfactory acuity test is shown. *p < .05, **p < .01, ***p < .001, Duncan multiple range test. Bars represent means ± SEM.

were observed in the minimal footshock intensity required to elicit vocalization between WT and R6/2 mice (data not shown).

p < .001) and the interaction between these two variables (F4,60 = 6.46, p < .001). For the 10 week old group (Fig. 4B), two-way ANOVA revealed significant effects of genotype (F1,20 = 14.40, p < .01), day (F4,80 = 16.43, p < .001) and the interaction between these two variables (F4,80 = 7.03, p < .001). For the 12 week old group (Fig. 4C), two-way ANOVA revealed significant effects of genotype (F1,18 = 38.00, p < .001), day (F4,72 = 11.42, p < .001) and the interaction between these two variables (F4,72 = 10.72, p < .001). In the comparison between the avoidance scores obtained by WT and R6/2 mice during the fifth day of performance (Fig. 4D) at the three ages tested, two-way ANOVA revealed significant effects of genotype (F1,53 = 73.43, p < .001) and age (F2,53 = 3.77, p < .05).

3.4. Active avoidance

3.5. Conditioned taste aversion

In the active avoidance experiment (Fig. 4), R6/2 mice were impaired in acquiring a S-R habit at all the ages tested when compared with WT mice. For the 8 week old group (Fig. 4A), two-way ANOVA revealed significant effects of genotype (F1,15 = 8.87, p < .01), day (F4,60 = 26.17,

Both WT and R6/2 mice developed a conditioned aversion for the saccharine solution to a similar extent (Fig. 5A). Two-way ANOVA revealed significant effects of treatment (F1,34 = 32.98, p < .001), but not of genotype or of the interaction between these two variables. No sig-

Table 2 Step-through latencies of WT and R6/2 mice during the inhibitory avoidance training trial (before footshock was delivered) Age

Experimental group

Latency (s) (means ± SEM) Genotype WT

R6/2

8 weeks

Single footshock Double footshock

16.44 ± 1.99 19.5 ± 1.98

18.87 ± 3.22 17.61 ± 1.63

12 weeks

Single footshock Double footshock

21.06 ± 6.4 22.44 ± 3.42

21.3 ± 3.27 17.5 ± 2.61

A. Ciamei, A.J. Morton / Neurobiology of Learning and Memory 89 (2008) 533–544

539

Fig. 3. Retention scores of WT and R6/2 mice on the testing day of the inhibitory avoidance paradigm. Mice aged 8 weeks (A) or 12 weeks (B) were given either 1 footshock (white columns) or 2 footshocks (striped columns) on the training day. *p < .05, **p < .01, Mann–Whitney U test. Bars represent means ± SEM.

Fig. 4. Avoidance scores of WT (s) and R6/2 mice (d) from testing at 8 weeks (A), 10 weeks (B) and 12 weeks of age (C). In (D) comparison of the avoidance scores of WT and R6/2 mice on the fifth day of performance is shown. *p < .05; **p < .01; ***p < .001, Duncan multiple range test. Bars represent means ± SEM.

nificant effects of genotype, treatment or of the interaction between these two variables were observed for the total amount of liquid drunk during the testing trial (Fig. 5B). Saline-treated WT and R6/2 mice showed a constant preference for the saccharine solution during the extinction experiment (Fig. 5C). No differences were observed between the performances of saline-treated WT and R6/2 mice. By contrast, there was a significant aversion shown for saccharine by both WT and R6/2 mice on the day of testing (Fig. 5D, arrow). This was maintained for several days and is similar to that seen by other investigators in WT mice (Bahar et al., 2003). However, whereas in WT

mice the aversion for the saccharine solution was extinguished by day 10 of the extinction procedure, R6/2 mice were impaired in extinguishing the initial conditioning, and still showed a clear aversion for the saccharine solution after 14 days of testing (Fig. 5D). Two-way ANOVA revealed significant effects of genotype (F1,23 = 4.54, p < .05) and of the interaction between genotype and day (F14,322 = 2.86, p < .001) between the performances of LiCl-treated mice. Univariate results showed significant differences between the amounts of saccharine consumed by WT and R6/2 mice on extinction days 8, 10, 12, 13 and 14.

540

A. Ciamei, A.J. Morton / Neurobiology of Learning and Memory 89 (2008) 533–544

Fig. 5. Preference scores for the saccharine solution of 10 week old WT and R6/2 mice in the CTA and extinction experiments. In (A) the volume of saccharine solution consumed by WT (white columns) and R6/2 mice (black columns) during the CTA testing phase is shown. In (B), the total amount of liquid drunk (i.e. normal water plus saccharine solution) by WT (white columns) and R6/2 mice (black columns) during the CTA testing phase is shown. In (C), the preference scores for the saccharine solution of saline-treated WT (s) and R6/2 mice (d) during the extinction experiment are shown. In (D), the preference scores for the saccharine solution of LiCl-treated WT (s) and R6/2 mice (d) during the extinction experiment are shown. The dashed line represents the average preference for the saccharine solution of saline-treated WT and R6/2 mice across the whole extinction experiment (taken from data shown in (C)). In (A), **p < .01, ***p < .001, Duncan multiple range test. In (D), *p < .05, **p < .01, univariate test. Bars represent means ± SEM. In (C) and (D) the arrows indicate the testing day.

4. Discussion Our experiments show five important findings. First, R6/2 mice are able to form functional long-term memories when tested in tasks of social and emotional memory that are known to involve the hippocampal memory system, such as the STFP and inhibitory avoidance tasks. Second, R6/2 mice show an age-dependent impairment in learning a S-R fear-based task, such as the active avoidance task, that is known to rely mainly on the striatal memory system. Third, performance of R6/2 mice in the STFP task is affected by a form of cognitive rigidity that results in their inability to disengage their choice from a previously reinforced preference, as shown in the shifted-reinforcement experiment. Fourth, in both the STFP and inhibitory avoidance tasks, R6/2 mice cannot modulate the strength of the consolidation process according to the relevance of the stimuli used to induce learning (i.e. the intensity of the scent that was socially transmitted during the STFP task or the number of footshocks received in the dark chamber of the inhibitory avoidance apparatus). Finally, as shown in the CTA experiment, R6/2 mice developed

an aversion to the saccharine solution when this was paired with the LiCl treatment, but failed to extinguish it, as was seen in WT mice. Results from open field testing do not show significant differences in basal levels of locomotor activity between WT and R6/2 mice when tested at 8 or 12 weeks of age. These are similar to data we have published previously (Dunnett et al., 1998). However, there is a non-significant trend for 12 week old R6/2 mice to show a decrease in locomotor activity and an increase in the time spent in the central zone of the arena compared to WT mice. Two previous studies (Bolivar et al., 2003; Carter et al., 2000) showed a similar trend in R6/2 mice carrying a shorter repeat length. When tested in the acquisition trial of the STFP task or in the inhibitory avoidance task, R6/2 mice display very good learning skills and retention levels. By contrast, in the active avoidance task, performance of R6/2 mice is markedly impaired by the eighth week of age. Each of these three tasks is known to involve the hippocampal and the striatal memory systems to different degrees. For example, the acquisition phase of the STFP task is predominantly a ‘hippocampal’ task relying mainly on structures within the

A. Ciamei, A.J. Morton / Neurobiology of Learning and Memory 89 (2008) 533–544

hippocampal region, including the hippocampus, subiculum and parahippocampal cortex (Alvarez et al., 2001, 2002). The hippocampus is also known to play a crucial role in the inhibitory avoidance task (Cammarota et al., 2005; Izquierdo & Medina, 1997). By contrast, the active avoidance task is known to rely mainly on the striatum (Ve´csei & Beal, 1991). Further, an involvement of the striatum in the inhibitory avoidance task has also been shown in several studies (Cammarota et al., 2005; Chavez, SaladoCastillo, Sanchez-Alavez, Quirarte, & Prado-Alcala`, 1995; Mazzucchelli et al., 2002; Packard et al., 2001). Thus, our data would suggest that as the task requirements move from those that involve mainly the hippocampus to those involving mainly the striatum, the performance of R6/2 mice worsen significantly. We show clearly that hippocampus-based learning skills are preserved in R6/2 mice at ages when striatum-based learning skills are seriously dysfunctional. This suggests that is a disturbance in memory ‘system’, rather than a single area. The concept of ‘memory system’ stems from studies showing that a dysfunction in an anatomical system such as the hippocampal formation, abolishes specific forms of memory (i.e. declarative, spatial or contextual memory) but leaves other forms of acquisition intact (e.g. stimulus–response or motor habits) (Cohen & Squire, 1980; McDonald & White, 1993; Scoville & Milner, 1957). However, it is now widely accepted that different memory systems, in particular the hippocampal and striatal systems, are not (as their anatomy would suggest) isolated entities that work independently from each other, but can interact either cooperatively or competitively to control behaviour (Chang & Gold, 2003; Kim & Baxter, 2001; Middei, Geracitano, Caprioli, Mercuri, & Ammassari-Teule, 2004; Packard & McGaugh, 1996; Poldrack & Packard, 2003; Schroeder, Wingard, & Packard, 2002). This may be particularly important in patients with neurological diseases, where the functional integration of neural processing may be impaired. For example, a recent study showed that, when tested in a route recognition task, HD patients could reach a level of performance similar to that of healthy controls (Voermans et al., 2004). However, whereas in control subjects the task induced similar levels of neural activity in the hippocampus and the caudate nucleus, in HD subjects a decreased activity in the striatum was compensated for by increased activity in the hippocampus. That is, the decreased functionality of dorsal striatum observed in HD patients, was balanced by a potentiation of hippocampal functions that allowed HD subjects to achieve an optimal performance. The behavioural paradigms we have used in our experiments have not been designed, and cannot be used, to determine whether there is a similar compensation in the interaction between striatal and hippocampal memory systems in R6/2 mice. However we describe a model in which the marked decrease in striatal functionality observed in aged R6/2 mice becomes evident only in tasks (such as the active avoidance task) where only the striatum is thought to be

541

involved, whereas the performance of mice in tasks in which also the hippocampus is involved (such as the inhibitory avoidance task) is not affected. We conclude that, as striatum-based mechanisms decline, hippocampus-based functions emerge to control the behavioural response of R6/2 mice. A second main conclusion of our experiments is that cognitive processes in R6/2 mice are rigid. The results obtained in the shifted-reinforcement trial of the STFP task, show that the preference learned by R6/2 mice during the acquisition appears to be stored in a rigid way that does not allow R6/2 mice to modulate their responding once the task contingencies are changed. This form of rigidity, that involves the perseverance of a previously learned preference, is highly reminiscent of the cognitive effects of lesion or inactivation of the vmPFC, in particular its pre-limbic (PL) and infra-limbic (IL) regions (Ragozzino, 2002; Ragozzino, Detrick, & Kesner, 2002). Lesion or inactivation of the PL/IL system impairs the ability of rodents to perform strategy switches in tasks in which a previously learned strategy has to be suppressed to generate a new one. For example, when rodents with PL/IL lesion or inactivation are tested in a cross maze, they can learn to find a pellet in one of the side arms either by using visual cues (place strategy) or by performing always the same body response (e.g. turn left; response strategy). However, when one strategy has been learned, they fail in switching to the other one. Interestingly, inactivation of the PL/IL system impairs only the switch between different strategies, while the performance of animals remains normal when a same strategy is reversed (e.g. switching from a ‘left’ body turn to a ‘right’ one; Ragozzino et al., 2002). In humans, this form of rigidity is normally referred to as ‘mental’ or ‘cognitive inflexibility’ and is a well known feature of HD patients (Lawrence, Sahakian, Rogers, Hodges, & Robbins, 1999; Lawrence et al., 1998; Watkins et al., 2000). Moreover, some studies report descriptions of HD patients as having ‘fixed ideas’ about the social context (Snowden et al., 2003). In R6/2 mice, deficits in reversal learning have been frequently observed (Lione et al., 1999; Morton et al., 2006; Pallier et al., 2007). In contrast to what has been observed for cross-modal shifts (Ragozzino et al., 2002), reversal learning involves the dorsal striatum, in particular its medial region (Palencia & Ragozzino, 2004). These studies, together with our findings, suggest that in both HD patients and R6/2 mice a failure in key structures of the cortico-striatal loops that mediate response selection (i.e. the vmPFC and dorsal striatum) might generate a form of cognitive rigidity that also affects the domain of social cognition. Notably, an involvement of dopaminergic neurotransmission in the medial prefrontal cortex has been shown in the acquisition of an active avoidance paradigm in rodents (Stark et al., 2001), suggesting that the deficit displayed by R6/2 mice in this task might represent more than just ‘striatal’ failure. A further memory system should also be taken into account when discussing the performance of R6/2 mice.

542

A. Ciamei, A.J. Morton / Neurobiology of Learning and Memory 89 (2008) 533–544

The amygdala, in particular its basolateral nucleus (BLA), is known to be crucial for the mechanisms that modulate memory formation according to the emotional relevance of the experience that is being consolidated (McGaugh, 2004; Packard & Teather, 1998). In all the tasks that we have used in our work, memory consolidation is influenced by the activity of the BLA (McGaugh, 2004; Wang, Fontanini, & Katz, 2006). In the olfactory acuity test and in the inhibitory avoidance task, WT mice show flexible responsiveness to graded stimuli. However, R6/2 mice showed pronounced deficits in flexibility of response to graded stimuli. Whereas decreasing the concentration of flavours in the STFP, or increasing the number of footshocks administered during the training trial in the inhibitory avoidance, decreased or increased (respectively) retention level of WT mice, the same ‘amount’ of retention was seen in R6/2 mice. This was not due to a different sensitivity of R6/2 mice to the stimuli used. Both WT and R6/2 mice show a similar response to the footshock, suggesting that they did not have a lower threshold of pain. Further, R6/ 2 mice could easily distinguish the lowest concentration of food. Although an enhanced olfaction of R6/2 mice might account for the results observed in the olfactory acuity task, we considered this an unlikely possibility. At the lowest concentrations, the smell of the flavours used for both the cued and the non-cued foods was clearly perceptible by the experimenter. Thus it is unlikely that the chance-level performance of WT mice was due to a cued food whose scent was imperceptible by WT mice. We suggest that it is not the olfactory acuity that is a problem, but the cognitive component of the discrimination. Abnormalities in olfactory function have been described in HD patients (Braak & Braak, 1992; Hamilton, Murphy, & Paulsen, 1999). However, data from the R6/2 mice are more consistent with data from a recent unpublished study of HD patients where olfactory thresholds were normal but olfactory discrimination was impaired (unpublished data, AJM, RA Barker, A Goodman, University of Cambridge). Together, our data suggest that the bluntened responding of R6/2 mice does not result from a sensory dysfunction. Rather, it seems likely to result from a deficit in BLA-based modulation of memory consolidation mechanisms that, by linking the experience that is being learned to its emotional value, normally promote the consolidation of the most relevant experiences. To date, no studies have investigated the formation of emotional memory in HD patients. However, a deficit in some psychological functions that involve the amygdala, such as the recognition of emotions through facial expressions, have been shown in HD patients (Johnson et al., 2007; Sprengelmeyer et al., 2006). To have a more direct measure of amygdala function, we tested R6/2 mice in a CTA paradigm followed by an extinction procedure, during which the repeated performance of the learned aversion took place in absence of the US. In the acquisition task, R6/2 mice showed levels of conditioning similar to those displayed by WT mice. Both WT and R6/2 mice showed significant levels of aver-

sion to the CS. However, whereas WT mice extinguished the learned aversion and regained a slight preference for the saccharine solution similar to that seen in the salinetreated mice, R6/2 mice failed to extinguish the learned aversion, and maintained a marked aversion for the CS throughout the experiment. Acquisition of a CTA is known to involve multiple sub-nuclei within the amygdala together with other brain structures involved in processing of gustatory information related to the CS used (i.e. the insular cortex), and of ascending viscero-sensory information related to the US (i.e. the parabrachial nucleus; Dudai, 2002). Although the exact role played by the amygdala in the formation and expression of this form of memory is still a matter of debate (Reilly & Bornovalova, 2005), some recent studies have shown a dissociation between the involvement of its central and basolateral (BLA) nuclei in the acquisition and extinction of CTA memories (Bahar et al., 2003, 2004). In particular, the central nucleus seems to be mainly involved in the acquisition process but not in the extinction of CTA, while the opposite is the case for the BLA. Thus, results from our CTA experiment suggest that in R6/2 mice, the deficit in amygdala-based function is restricted to the BLA, since the performance of mice in the acquisition trial is unimpaired. Interestingly, a dissociation of functions has been shown also between the BLA and the lateral nucleus of the amygdala, the latter being involved in tasks, such as the fear conditioning, in which the amygdala is supposed to act as a site for plasticity and consolidation (Wilensky et al., 2000). This might explain why R6/2 mice show specific deficits in modulation of memory processes (or in extinction of a CTA), without showing deficits in other fear-based tasks, such as the fear conditioning, that do not rely on the BLA (Bolivar et al., 2003). A recent study has shown a role for the vmPFC (in particular the PL and IL regions) in the formation of extinction memory for CTA (Mickley et al., 2005). These previous observations together with our current results, support the idea that a dysfunction in both vmPFC-based mechanisms of cognitive flexibility and in emotional modulation of memory processes mediated by the BLA might generate the rigidity we have observed in tasks where the reward/punishment contingencies were changed, without affecting the acquisition process. Taken together, our results suggest that the performance of R6/2 mice, in tasks of social and emotional memory, results from a complex set of modifications that affect at the same time the different brain systems involved in their formation and the interaction between them. The progressive decline of striatal functionality can be observed only in tasks in which this structure does not interact with the hippocampus, while the performance of R6/2 mice appears to be ‘normal’ in tasks in which both these systems are involved, suggesting the existence of compensatory mechanisms through which the hippocampus balances striatal failure. Moreover, our data suggest that the cognitive performance of R6/2 mice is further compromised by an early inability to extinguish the memory of previously learned

A. Ciamei, A.J. Morton / Neurobiology of Learning and Memory 89 (2008) 533–544

experiences in order to consider new sets of information and/or to confer the appropriate emotional relevance to the memory that is being formed. As a consequence of these deficits, likely to rely on a malfunction in frontostriatal loops involving both the dorsal striatum and the vmPFC, and in BLA-based modulatory mechanisms, the performance of R6/2 mice become rigid when, after the task has been learned, the stimuli used to induce learning and/or their emotional value are changed. Acknowledgments This work was supported by a grant from The HighQ Foundation. We thank Dr. Nigel Wood for helpful discussion. References Akirav, I., Khatsrinov, V., Vouimba, R.-M., Merhav, M., Ferreira, G., Rosemblum, K., et al. (2006). Extinction of conditioned taste aversion depends on functional protein synthesis but not NMDA receptor activation in the ventromedial Prefrontal cortex. Learning & Memory, 13, 254–258. Alvarez, P., Lipton, P. A., Melrose, R., & Eichenbaum, H. B. (2001). Differential effects of damage within the hippocampal region on memory for a natural nonspatial odor-odor association. Learning & Memory, 8, 79–86. Alvarez, P., Wendelken, L., & Eichenbaum, H. B. (2002). Hippocampal formation lesions impair performance in an odor–odor association task independently of spatial context. Neurobiology of Learning and Memory, 78, 470–476. Bahar, A., Samuel, A., Havzi, S., & Dudai, Y. (2003). The amygdalar circuit that acquires taste aversion memory differs from the circuit that extinguishes it. European Journal of Neuroscience, 17, 1527–1530. Bahar, A., Dorfman, N., & Dudai, Y. (2004). Amygdalar circuits required for either consolidation or extinction of taste aversion memory are not required for reconsolidation. European Journal of Neuroscience, 19, 1115–1118. Bates, G., Harper, P. S., & Jones, L. (2002). Huntington’s disease (3rd ed.). Oxford: Oxford University Press. Bolivar, V. J., Manley, K., & Messer, A. (2003). Exploratory activity and fear conditioning abnormalities develop early in R6/2 Huntington’s disease transgenic mice. Behavioural Neuroscience, 117, 1233–1242. Braak, H., & Braak, E. (1992). Allocortical involvement in Huntington’s disease. Neuropathology and Applied Neurobiology, 18, 539–547. Cammarota, M., Bevilaqua, L. R. M., Ko¨hler, C., Medina, J. H., & Izquierdo, I. (2005). Learning twice is different from learning once and from learning more. Neuroscience, 132, 273–279. Carter, R. J., Lione, L. A., Humby, T., Mangiarini, L., Mahal, A., Bates, G. P., et al. (1999). Characterization of progressive motor deficits in mice transgenic for the human Huntington’s disease mutation. Journal of Neuroscience, 19, 3248–3257. Carter, R. J., Hunt, M. J., & Morton, A. J. (2000). Environmental stimulation increases survival in mice transgenic for the exon 1 of the Huntington’s disease gene. Movement Disorders, 15, 925–937. Chang, Q., & Gold, P. E. (2003). Intra-hippocampal lidocaine injections impair acquisition of a place task and facilitate acquisition of a response task. Behavioural Brain Research, 144, 19–24. Chavez, M. E., Salado-Castillo, R., Sanchez-Alavez, M., Quirarte, G. L., & Prado-Alcala`, R. A. (1995). Post-training injections of GABAergic antagonists into the striatum produces retrograde amnesia. Neurobiology of Learning and Memory, 63, 296–300. Ciamei, A., Cestari, V., & Castellano, C. (2000). Strain-dependent interactions between MK-801 and cocaine on retention of C57BL/6

543

and DBA/2 mice tested in a one-trial inhibitory avoidance task: Involvement of dopaminergic mechanisms. Neurobiology of Learning and Memory, 73, 188–194. Cohen, N. J., & Squire, L. R. (1980). Preserved learning and retention of pattern-analyzing skill in amnesia: Dissociation of knowing how and knowing that. Science, 210, 207–210. Dudai, Y. (2002). Memory from A to Z. Oxford: Oxford University Press. Dunnett, S. B., Carter, R. J., Watts, C., Torres, E. M., Mahal, A., Mangiarini, L., et al. (1998). Striatal transplantation in a transgenic mouse model of Huntington’s disease. Experimental Neurology, 154, 31–40. Eichenbaum, H. B. (2000). A cortical-hippocampal system for declarative memory. Nature Reviews Neuroscience, 1, 41–50. Gibson, H. E., Reim, K., Brose, N., Morton, J. A., & Jones, S. (2005). A similar impairment in CA3 mossy fibre LTP in the R6/2 mouse model of Huntington’s disease and in the complexin II knockout mouse. European Journal of Neuroscience, 22, 1701–1712. Gold, P. E. (2004). Coordination of multiple memory systems. Neurobiology of Learning and Memory, 82, 230–242. Hamilton, J. M., Murphy, C., & Paulsen, J. S. (1999). Odor detection, learning, and memory in Huntington’s disease. Journal of the International Neuropsychological Society, 5, 609–615. Izquierdo, I., & Medina, J. H. (1997). Memory formation: The sequence of biochemical events in the hippocampus and its connection to activity in other brain structures. Neurobiology of Learning and Memory, 68, 285–316. Johnson, S. A., Stout, J. C., Solomon, A. C., Langbehn, D. R., Aylward, E. H., Cruce et al. (2007). Beyond disgust: Impaired recognition of negative emotions prior to diagnosis in Huntington’s disease. Brain 130, 1715–1717. Josselyn, S. A., Kida, S., & Silva, A. J. (2004). Inducible repression of CREB functions disrupts amygdala-dependent memory. Neurobiology of Learning and Memory, 82, 159–163. Kim, J. J., & Baxter, M. G. (2001). Multiple brain-memory systems: The whole does not equal the sum of its parts. Trends in Neurosciences, 26, 324–330. Kogan, J. H., Frankland, P. W., Blendy, J. A., Coblentz, J., Marowitz, Z., Schu¨tz, G., et al. (1997). Spaced training induces normal long-term memory in CREB mutant mice. Current Biology, 7, 1–11. Lawrence, A. D., Sahakian, B. J., Rogers, R. D., Hodges, J. R., & Robbins, T. W. (1999). Discrimination, reversal, and shift learning in Huntington’s disease: Mechanisms of impaired response selection. Neuropsychologia, 37, 1359–1374. Lawrence, A. D., Weeks, R. A., Brooks, D. J., Andrews, T. C., Watkins, L. H. A., Harding, A. E., et al. (1998). The relationship between striatal dopamine receptor binding and cognitive performance in Huntington’s disease. Brain, 121, 1343–1355. Lione, L. A., Carter, R. J., Hunt, M. J., Bates, G. P., Morton, J. A., & Dunnett, S. B. (1999). Selective discrimination learning impairments in mice expressing the human Huntington’s disease mutation. Journal of Neuroscience, 19, 10428–10437. Mangiarini, L., Sathasivam, K., Seller, M., Cozens, B., Harper, A., Hetherington, C., et al. (1996). Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell, 87, 493–506. Mazzucchelli, C., Vantaggiato, C., Ciamei, A., Fasano, S., Pakhotin, P., Krezel, W., et al. (2002). Knockout of ERK1 MAP kinase enhances synaptic plasticity in striatum and facilitates striatal-mediated learning and memory. Neuron, 34, 1–20. McDonald, R. J., & White, N. M. (1993). A triple dissociation of memory systems: Hippocampus, amygdala and dorsal striatum. Behavioural Neuroscience, 107, 3–22. McGaugh, J. L. (2004). The amygdala modulates the consolidation of memories of emotionally arousing experiences. Annual Reviews in Neuroscience, 27, 1–28. Mickley, G. A., Kenmuir, C. L., Yocom, A. M., Justin, A. W., & Biada, J. M. (2005). A role for prefrontal cortex in the extinction of a conditioned taste aversion. Brain Research, 1051, 176–182.

544

A. Ciamei, A.J. Morton / Neurobiology of Learning and Memory 89 (2008) 533–544

Middei, S., Geracitano, R., Caprioli, A., Mercuri, N. B., & AmmassariTeule, M. (2004). Preserved fronto-striatal plasticity and enhanced procedural learning in a transgenic mouse model of Alzheimer’s disease overexpressing mutant hAPPswe. Learning & Memory, 11, 447–452. Morton, J. A., Hunt, M. J., Hodges, A. K., Lewis, P. D., Redfern, A. J., Dunnett, S. B., et al. (2005). A combination drug therapy improves cognition and reverses gene expression changes in a mouse model of Huntington’s disease. European Journal of Neuroscience, 21, 855–870. Morton, J. A., Skillings, E., Bussey, T. J., & Saksida, L. (2006). Measuring cognitive deficits in disabled mice using an automated interactive touchscreen system. Nature Methods, 3, 767. Murphy, K. P. S. J., Carter, R. J., Lione, L. A., Mangiarini, L., Mahal, A., Bates, G. P., et al. (2000). Abnormal synaptic plasticity and impaired spatial cognition in mice transgenic for exon 1 of the human Huntington’s disease mutation. Journal of Neuroscience, 20, 5115–5123. Packard, M. G., & McGaugh, J. L. (1996). Inactivation of hippocampus or caudate nucleus with lidocaine differentially affects expression of place and response learning. Neurobiology of Learning and Memory, 65, 65–72. Packard, M. G., & Teather, L. A. (1998). Amygdala modulation of multiple memory systems: Hippocampus and caudate-putamen. Neurobiology of Learning and Memory, 69, 163–203. Packard, M. G., Vecchioli, S., Schroeder, J. P., & Gasbarri, A. (2001). Task-dependent role of dorsal striatum metabotropic glutamate receptors in memory. Learning & Memory, 8, 96–103. Palencia, C. A., & Ragozzino, M. E. (2004). The influence of NMDA receptors in the dorsomedial striatum on response reversal learning. Neurobiology of Learning and Memory, 82, 81–89. Pallier, P. N., Maywood, E. S., Zheng, Z., Chesham, J. E., Inyushkin, A. N., Dyball, R., et al. (2007). Pharmacological imposition of sleep slows cognitive decline and reverses dysregulation of circadian gene expression in a transgenic mouse model of Huntington’s disease. Journal of Neuroscience, 18, 7869–7878. Poldrack, R. A., & Packard, M. G. (2003). Competition among multiple memory systems: Converging evidence from animal and human brain studies. Neuropsychologia, 41, 245–251. Ragozzino, M. E. (2002). The effects of dopamine D1 receptor blockade in the prelimbic-infralimbic areas on behavioral flexibility. Learning & Memory, 9, 18–28. Ragozzino, M. E., Detrick, S., & Kesner, R. P. (2002). Involvement of the prelimbic-infralimbic areas of the rodent prefrontal cortex in behavioral flexibility for place and response learning. Journal of Neuroscience, 19, 4585–4594. Reilly, S., & Bornovalova, M. A. (2005). Conditioned taste aversion and amygdala lesions in the rat: A critical review. Neuroscience and Biobehavioural Reviews, 29, 1067–1088. Roozendaal, B., Koolhaas, J. M., & Bohus, B. (1993). The central amygdala is involved in conditioning but not in retention of active and passive shock avoidance in male rats. Behavioral Neural Biology, 59, 143–149. Ross, R. S., & Eichenbaum, H. B. (2006). Dynamics of hippocampal and cortical activation during consolidation of a nonspatial memory. The Journal of Neuroscience, 26, 4852–4859.

Sa`nchez-Andrade, G., Bronwen, M. J., & Kendrick, K. M. (2005). Neural encoding of olfactory recognition memory. Journal of Reproduction and Development, 51, 547–558. Schroeder, J. P., Wingard, J. C., & Packard, M. G. (2002). Post-training reversible inactivation of hippocampus reveals interference between memory systems. Hippocampus, 12, 280–284. Scoville, W. B., & Milner, B. (1957). Loss of recent memory after bilateral hippocampal lesion. Journal of Neuropsychiatry and Clinical Neuroscience, 12, 103–113. Snowden, J. S., Gibbons, Z. C., Blackshaw, A., Doubleday, E., Thompson, J., Craufurd, D., et al. (2003). Social cognition in frontotemporal dementia and Huntington’s disease. Neuropsychologia, 41, 688–701. Solana-Figueroa, R., Salado-Castillo, R., Galinda, L. E., Quirarte, G. L., & Prado-Alcala`, R. A. (2002). Effects of pretraining intrastriatal administration of p-chloroamphetamine on inhibitory avoidance. Neurobiology of Learning and Memory, 78, 178–185. Sprengelmeyer, R., Schroeder, U., Young, A. W., & Epplen, J. T. (2006). Disgust in pre-clinical Huntington’s disease: A longitudinal study. Neuropsychologia, 44, 518–533. Stack, E. C., Kubilus, J. K., Smith, K., Cormier, K., Del Signore, S. J., Guelin, E., et al. (2005). Chronology of behavioral symptoms and neuropathological sequela in R6/2 Huntington’s disease transgenic mice. Journal of Comparative Neurology, 490, 354–370. Stark, H., Bischof, A., Wagner, T., & Scheich, H. (2001). Activation of the dopaminergic system of the medial prefrontal cortex of gerbils during formation of relevant associations for the avoidance strategy in the shuttle box. Progress in Neuropsychopharmacology & Biological Psychiatry, 25, 409–426. The Huntington’s Disease Collaborative Research Group. (1993). A novel gene containing trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes. Cell 72: 971–983. Ve´csei, L., & Beal, M. F. (1991). Comparative behavioral and neurochemical studies with striatal kaininc acid- or quinolinic acid-lesioned rats. Pharmacology Biochemistry and Behavior, 39, 473–478. Voermans, N. C., Petersson, K. M., Daudley, L., Weber, B., van Spaendonck, K. P., Kremer, H. P. H., et al. (2004). Interaction between the human hippocampus and the caudate nucleus during route recognition. Neuron, 43, 427–435. Wang, Y., Fontanini, A., & Katz, D. B. (2006). Temporary basolateral amygdala lesions disrupt acquisition of socially transmitted food preferences in rats. Learning & Memory, 13, 794–800. Watkins, L. H. A., Rogers, R. D., Lawrence, A. D., Sahakian, B. J., Rosser, A. E., & Robbins, T. W. (2000). Impaired planning but intact decision making in early Huntington’s disease: Implications for specific fronto-striatal pathology. Neuropsychologia, 38, 1112–1125. Welzl, H., D’Adamo, P., & Lipp, H.-P. (2001). Conditioned taste aversion as a learning and memory paradigm. Behavioural Brain Research, 125, 205–213. Wilensky, A. E., Schafe, G. E., & LeDoux, J. E. (2000). The amygdala modulates memory consolidation of fear motivated inhibitory avoidance learning but not classical fear conditioning. Journal of Neuroscience, 20, 7059–7066.