These authors contributed equally to this work.
Reduced expression of the vesicular acetylcholine transporter causes learning deficits in mice
Article first published online: 6 SEP 2008
© 2009 The Authors Journal compilation © 2009 Blackwell Publishing Ltd/International Behavioural and Neural Genetics Society
Genes, Brain and Behavior
Volume 8, Issue 1, pages 23–35, February 2009
How to Cite
De Castro, B. M., Pereira, G. S., Magalhães, V., Rossato, J. I., De Jaeger, X., Martins-Silva, C., Leles, B., Lima, P., Gomez, M. V., Gainetdinov, R. R., Caron, M. G., Izquierdo, I., Cammarota, M., Prado, V. F. and Prado, M. A. M. (2009), Reduced expression of the vesicular acetylcholine transporter causes learning deficits in mice. Genes, Brain and Behavior, 8: 23–35. doi: 10.1111/j.1601-183X.2008.00439.x
- Issue published online: 28 JAN 2009
- Article first published online: 6 SEP 2008
- Received 6 March 2008, revised 9 June, 21 July, 22 August 2008, accepted for publication 25 August 2008
- mutant mice;
- synaptic vesicle
- Top of page
- Material and methods
- Supporting Information
Storage of acetylcholine in synaptic vesicles plays a key role in maintaining cholinergic function. Here we used mice with a targeted mutation in the vesicular acetylcholine transporter (VAChT) gene that reduces transporter expression by 40% to investigate cognitive processing under conditions of VAChT deficiency. Motor skill learning in the rotarod revealed that VAChT mutant mice were slower to learn this task, but once they reached maximum performance they were indistinguishable from wild-type mice. Interestingly, motor skill performance maintenance after 10 days was unaffected in these mutant mice. We also tested whether reduced VAChT levels affected learning in an object recognition memory task. We found that VAChT mutant mice presented a deficit in memory encoding necessary for the temporal order version of the object recognition memory, but showed no alteration in spatial working memory, or spatial memory in general when tested in the Morris water maze test. The memory deficit in object recognition memory observed in VAChT mutant mice could be reversed by cholinesterase inhibitors, suggesting that learning deficits caused by reduced VAChT expression can be ameliorated by restoring ACh levels in the synapse. These data indicate an important role for cholinergic tone in motor learning and object recognition memory.
Maintenance of cholinergic tone is directly linked to the capacity of nerve endings to effectively synthesize and release acetylcholine (ACh) (Ribeiro et al. 2006). Synthesis of ACh depends on the activity of a high affinity choline transporter (CHT1) that is necessary to supply choline for ACh synthesis. Choline provided by CHT1 is used to generate ACh in the cytosol, in a step catalyzed by the enzyme choline acetyl transferase (ChAT) (Dobransky & Rylett 2005). ChAT is likely to be in kinetic excess, therefore moderate changes on its activity or expression may not affect ACh synthesis (Brandon et al. 2004). The final key player in cholinergic nerve-endings is the vesicular acetylcholine transporter (VAChT), a protein that takes up cytosolic ACh into synaptic vesicles and thereby provides vesicular neurotransmitter for exocytotic release in response to calcium influx (Edwards 2007; Parsons 2000).
In contrast to ChAT and CHT1, VAChT activity plays a rate-limiting role in ACh release (Prado et al. 2006). Mice lacking VAChT (VAChT knockout) do not survive (Prado, de Castro, Caron and Prado, unpublished data), however, mice with reduced levels of VAChT (VAChT knockdown, VAChT KD mice) are viable and were instrumental to demonstrate a role for this transporter ‘in vivo’. These mutant mice present distinct behavioral deficits, suggesting that the ability to efficiently pack ACh in synaptic vesicles is fundamental for certain physiological functions (Prado et al. 2006).
Our initial investigation detected several physiological and behavioral deficits in heterozygous and homozygous VAChT mutants that are consistent with decreased cholinergic tone. In particular, homozygous VAChT KD (KDHOM) mice showed severe muscle fatigue consistent with a myasthenic phenotype as well as selective cognitive alterations. In contrast, heterozygous VAChT KD (KDHET) mice displayed seemingly normal neuromuscular function, although they also showed cognitive alterations similar to those found in VAChT KDHOM mice. This suggested that central cholinergic synapses are more sensitive to decreased VAChT levels than motor end-plates. These VAChT KDHET mice showed object and social recognition memory deficits and also presented impairments in performance on the rotarod. However, the precise contribution of reduced cholinergic tone for these behavioral alterations has not been investigated. Hence, deficits in object recognition memory may potentially relate to the inability of mutant mice to acquire or to retrieve information. In addition, alteration in rotarod performance could be explained by altered gait and equilibrium or by deficits in motor learning.
Here we used VAChT KDHET mice to address the question of how reduced VAChT levels affect cognitive processing related to these previously described behavioral abnormalities. We found that most cognitive deficits observed in VAChT KDHET mice are related to deficiency in learning, rather than retrieval of information. We suggest that targeting mechanisms that can facilitate information encoding could help to ameliorate the consequences of cholinergic dysfunction because of reduced VAChT levels.
Material and methods
- Top of page
- Material and methods
- Supporting Information
VAChT mutant mice were previously described (Prado et al. 2006). These mice were generated by targeting the 5’ untranslated region of the VAChT gene by homologous recombination in a mixed 129S6/ SvEvTac (129S6) x C57BL/6J background and were backcrossed to C57BL/6/Uni [imported to Brazil from Zentralinstitut fur Versuchstierzcht, Hannover, Germany and maintained at the University of Campinas (UNICAMP)] for only three generations (N3), as further backcrossing into this strain caused infertility (data not shown). Only male mice were used in this study. Heterozygous mice were intercrossed to generate all the genotypes used in these experiments and littermate wild-type mice were used as controls.
Animals were housed in groups of three to five per cage in a temperature-controlled room with a 12:12 light-dark cycles in microisolator cages. Food and water were provided ad libitum. Mouse colonies were maintained at the Federal University of Minas Gerais, Brazil, in accordance with National Institutes of Health (NIH) guidelines for the care and use of animals. Experiments were performed accordingly to approved animal protocols from the Institutional Animal Care and Use Committees at the Federal University of Minas Gerais and PUC-RS. To minimize the number of animals used, naïve mice were used for the object recognition as well as for the Y maze task and then they were used for one of the following tasks: evaluation of rotarod performance, assessment of locomotor activity or for the Morris water maze (MWM). An interval of at least 1 week was observed between two distinct tasks.
Adult VAChT KD and wild-type (WT) mice were anesthetized with ketamine/xilazine (70/10 mg/kg) i.p. and transcardially perfused with ice-cold phosphate-buffered saline (PBS) pH 7.4 for 10 min, followed by ice-cold 4% paraformaldehyde (PFA) in PBS for 10 min. Perfused brains and spinal cords were immediately postfixed in 4% PFA in PBS overnight at 4°C. Following cryoprotection in 4% PFA with 10% sucrose, tissues were rapidly frozen in isopentane over dry ice and kept at −80°C. Serial sections (40 μm thick) were cut on a Cryostat (Micron) and immersed in ice-cold PBS. Sections from control and test mice were processed simultaneously for all experiments.
Brain and spinal cord slices were permeabilized in 1.2% Triton/PBS and rinsed in PBS. Tissues were blocked for 1 h in 10% normal goat serum/PBS and immunonstained with primary antibodies VAChT (rabbit polyclonal, 1:250, Sigma Chem. Co., São Paulo, Brazil), CHT1 (rabbit polyclonal 1:250, kindly provided by R. Jane Rylett, University of Western Ontario, London, Canada) in incubation buffer (2% normal goat serum; 0.2% Triton; PBS) for 48 h at 4°C. After washing three times with PBS for 20 min each, tissues were incubated with Alexa Fluor 488 goat anti-rabbit (1:500, Invitrogen, SP, Brazil) in incubation buffer for 1 h. Slices were rinsed again and incubated with DAPI (1:1000) for 10 min at room temperature. Tissue sections were then mounted and coverslipped using ProLong® Gold antifade reagent (Invitrogen, SP, Brazil). Images were acquired using an Axiovert 200M using the ApoTome system or a LEICA SP5 confocal microscope to obtain optical sections of the tissue. Objectives used were 20× dry, 40× water immersion (1.2 Numerical aperture [NA]) and a 63× oil immersion (1.4 Numerical aperture). Fluorescence intensity analyses were carried out with the (Image J, http://rsbweb.nih.gov/ij/). Images were obtained in grayscale and a threshold was applied based on the mean value of fluorescence intensity of the whole image for VAChT staining automatically obtained using Image J. The total fluorescence intensity for VAChT was then detected automatically by the software and normalized for the fluorescence intensity of a image from a sequential Cryostat section stained with anti-CHT1 antibody that was submitted to the same procedure (i.e. automatic threshold and detection of fluorescent intensity as well). The normalized fluorescence for VAChT staining was divided by the area of the image and data from WT mice provided values for 100%.
All experiments were conducted during the light phase of the cycle. All efforts were made to minimize any suffering and the number of animals used.
Temporal order task
There are several tasks that can be used to assess object recognition memory and here to evaluate whether information encoding is affected in these mutants, we used a temporal order task for recency memory (Bevins & Besheer 2006; Dere et al. 2005). In this task rodents need to remember the order in which two distinct objects were presented, what is shown by increased exploration of the first object presented in a sequence (the less familiar object).
All animals were given a single 10-min habituation session, with no objects in the open-field arena (50 × 30 cm) which was maintained inside an illuminated larger box with controlled illumination. Twenty-four hours later, in the first sample phase (S1), the animals were allowed to explore two copies of an identical object for a total of 10 min. In the second sample phase (S2), with a delay of 1 h, other two copies of an identical object were used and animals were let to explore these objects for 10 min. One hour after S2, in the test session (T), animals were allowed to explore, during 10 min, one object identical from S1 and another identical from S2. The objects were Lego toys that presented similar textures, colors and sizes, but distinct shapes. Between trials objects and arena were cleaned with 70% alcohol and air dried. Exploration time was defined as sniffing or touching the object with nose and/or forepaws (Prado et al. 2006). The experiments were recorded and an experimenter blind to the genotype scored time exploring the object for each mice.
If object recognition memory is intact, the subjects will spend more time exploring the object from S1 compared with the object from S2 in the test session. The results are expressed as percentage of exploration for each object and the exploration times (in seconds, mean ± SEM) in S1 (WT = 36 ± 3; KDHET = 38 ± 3; t(104) = 0.6034) and S2 (WT = 35. ± 3; KDHET = 34 ± 3; t(97) = 0.2705) between the two genotypes were not statistically different. We noticed no alteration in these times after pharmacological manipulations.
To evaluate for statistical significance for the object recognition memory, the percentage exploration time for the two objects in the test session were compared by One-sample t-test to evaluate if exploration was significantly larger than chance (50% of the time). The null hypothesis was that the mice did not remember which object was presented less recently and explored the two objects equally.
Morris water maze
The water maze was a black circular pool (120 cm in diameter) conceptually divided in four equal imaginary quadrants for the purpose of data analysis. The water was made opaque with nontoxic white tempera and the temperature was 21–23°C. One centimeter beneath the surface of the water and hidden from the mouse’s view was a circular platform (9 cm in diameter), with a rough surface, which allowed mice to climb onto it easily. The swimming path of the animals was recorded using a video camera mounted above the center of the pool and analyzed using an in house video tracking and analysis system. The water maze was located in a white room with several posters and other distal visual stimuli hanging on the walls to provide spatial cues. A curtain separated the water maze room from the room where the computer was set up and where the animals were temporarily housed during the behavioral sessions. Training in the hidden platform (spatial) version of the MWM was carried out during 5 consecutive days as previously described (Rossato et al. 2006). On each day, mice received six consecutive training trials during which the hidden platform was kept in a constant location. A different starting location was used on each trial, which consisted of a swim followed by a 30-s platform sit. Any mouse that did not find the platform within 60 s was guided to it by the experimenter. The intertrial interval (ITI) was 30 s. During the ITI, mice were carefully dried with a towel by the experimenter. Memory retention was evaluated in a 60-s probe trial carried out in the absence of the escape platform 24 h after the last training session. Difference of performance between genotypes was assessed using the t-test.
Immediate working memory performance was assessed by recording spontaneous alternation behavior during a single session in a Y maze (Hughes 2004; Pych et al. 2006). Each mouse, new to the maze (30 cm long by 6 cm wide by 20 cm high), was placed at the end of one arm and allowed to move freely through the maze during an 8-min session. The series of arm entries was recorded visually. Alternation was defined if mice entered different arms three times in succession from the results of consecutive arm entering. The number of overlapping entrance sequences (e.g., ABC, BCA) was defined as the number of alternations. The effect was calculated as percentage alternation according to the following formula: [total alternation/(total arms entered − 2)] × 100. Therefore, the following hypothetical sequence of arms entered by a mice: A, C, A, B, C, A, C, B, A, C would yield an alternation score of 75% ([6 alternations/(10-2) arms entered] × 100). A generic triad sequence has 3 × 2 × 2 = 12 possibilities, because arms cannot be repeated in adjacent observations but may appear twice on the same triad sequence. However, the number of possibilities, if arms are not allowed to repeat themselves in a triad sequence, would be 3 × 2 × 1 = 6. Thus, the index of 50% means random selection of goals arms. Animals statistically reach more than 50% alternation, indicating no random arms selection. By this criterion, we would see 100% alternation score only if the animal had run consistently clockwise or anticlockwise, however, this kind of behavior was not observed for any of the animals, independently of the genotype. To determine if alternation scores were significantly above the chance (50%), we used one-sample t-test. The differences between groups were compared using Student’s t-test.
Locomotor activity was measured using an automated activity monitor (Accuscan Instruments, Inc., Columbus, OH, USA) (Sotnikova et al. 2004). Experiments were performed between 10:00 and 16:00 h. Mice were allowed to explore the locomotor activity chamber (20 × 20 cm) for 1 h. After that, they received either saline or galantamine (1 mg/kg, i.p.) and locomotor activity was monitored for additional 2 h. Activity (converted from beam breaks to cm) was measured at 5-min interval. Measurements of total activity were obtained and statistical significance was assessed by t-test when the data passed a normal distribution test. Otherwise, we used a Mann–Whitney Rank Sum Test.
To assess procedural learning we used an accelerating rotarod and investigated motor skill learning during several days. The rotarod apparatus was made of gray plastic (rotating cylinder diameter: 5 cm, width: 8 cm, height: 20 cm) with an automatic fall recorder (Insight Equipaments, Ribeirão Preto, Brazil). To characterize improvement in rotarod performance, mice were trained during eight consecutive days. In order to evaluate retained performance, mice were re-exposed to the same task 10 days later. Before daily training, mice were allowed to accommodate on the rod for 5 min. The animals were placed on the rotating roller and the rotation was increased from 5 to 35 r.p.m. over a 5-min period. For each day, mice were subjected to ten trials with a 30-s interval between trials. Statistical analysis was performed with analysis of variance (anova) (repeated measure) when the data passed a normality test, or a Friedman repeated measures anova on Ranks otherwise.
In the pretraining galantamine dose–response study, either saline or galantamine (0.5, 1 and 3 mg/kg, i.p.) was injected 30 min prior to S1. In the second study, saline or galantamine (1 mg/kg i.p.) was injected 30 min before the test session (T). The choice of dose of galantamine (1 mg/kg) was based on the findings from the dose–response study and by the fact that at this dose there is no alteration in locomotor activity. In some experiments we injected saline or donepezil (0.5 mg/kg, i.p.) 30 min before S1.
- Top of page
- Material and methods
- Supporting Information
Expression of VAChT in nerve terminals is reduced in VAChT KD mice
To visualize how decreased VAChT expression influences transporter levels in nerve terminals, we used indirect immunofluorescence in cryostat sections obtained from cortex (M1 motor cortex region, layer V), hippocampus (CA1 region), striatum (caudate) and spinal cord (lamina IX, ventral edge of the ventral horn on the cervical region). Sections from WT mice present robust and widespread punctated labeling of nerve endings detected with a VAChT antibody and this labeling is similar to that obtained with other integral synaptic vesicle proteins (Fig. 1). Staining with this VAChT antibody is specific, as it is suppressed in sections from VAChT KO mice (data not shown). The punctated labeling was decreased both in heterozygous (VAChT KDHET) and homozygous (VAChT KDHOM) mice in different areas of the brain and also in the spinal cord (Fig. 1a–c). Quantification of immunofluorescence labeling in hippocampal sections indicates that immunoreactivity is decreased by 40% by reduction of VAChT in VAChT KDHET mice. As a control experiment we also used sections from VAChT KDHOM mice, in which we could detect even further reduction (70%). No decrease in immunoreactivity was detected in sequential cryostat sections from hippocampus stained with an antibody against CHT1 (Fig. 1d) (Ribeiro et al. 2005).
Motor skill learning is affected in VAChT KD mice
We previously found that VAChT KDHET mice present a deficit in motor skill performance, whereas VAChT KDHOM mice are unable to spend prolonged amounts of time on the rotarod, likely because of reduced physical capacity (Prado et al. 2006). Although this observation suggests decreased coordination in VAChT KDHET mice, it is also possible that they may have alterations in learning of this motor skill. Therefore, in order to test the hypothesis that motor skill learning is affected in VAChT KDHET mice, we used experimental protocol involving accelerating rotarod that allows maximum motor skill acquisition over the period of days for WT mice (Fig. 2). In this experimental protocol, both WT and VAChT KDHET mice showed an improvement in performance both within sessions (intrasession performance on day 1, a Friedman Repeated Measures Analysis of Variance on Ranks show a significant difference for WT, Chi-square = 71.850, P < 0.001 and for VAChT KDHET Chi-square = 157.650, P < 0.001) and between sessions [intersession performance, two-way repeated measure anova shows effect of day F(7,203) = 59.381, P < 0.001); Figs 2 and 3], indicating that both genotypes are able to learn the motor skill. A main effect of genotype just failed statistical significance (F(1,270) = 4.064, P = 0.05) as did the genotype × day interaction (F(7,203) = 2.020, P = 0.054). Because the data indicated a clear trend for statistical difference, we also analyzed the performance for individual days. This analysis showed a significant effect of genotype on days 2, 3 and 4 [two-way repeated measure anova genotype effect for days 2: F(1,270) = 5.019; P < 0.05; day 3: F(1,270) = 8.089; P < 0.01; and day 4: F(1,270) = 7.35; P < 0.05], suggesting that VAChT KDHET mice performed worse than WT mice on these days (Fig. 2b). VAChT KDHET mice were able to perform as well as WT mice after 5 days of training suggesting that mutant mice are able to learn the task, although they needed more training to reach maximum performance. In agreement with this observation, analysis of performance by averaging the total time that animals spent on the rod during the 10 trials of each day (Fig. 3a) revealed that WT improve their performance (one-way repeated measure anova: F(7,105) = 31.144, P < 0.001) and that maximum performance is achieved on the third day (Tukey post hoc revealed a significant difference between day 8 and day 2 P < 0.01, but not between day 8 and day 3 P = 0.999). VAChT KDHET mice are also able to improve their performance (one-way repeated measure anova: F(7,105) = 30.279, P < 0.001), but they achieved their maximum performance only on the fourth day (Tukey post hoc analyses revealed significant difference between day 8 and day 3, P < 0.01; but not between day 8 and day 4, P = 0.495). Exponential fitting of curves generated for each individual mouse for the time spent on the rod in a given day (plotted similar to the average shown on Fig. 2b) shows that maximum performance is identical for two genotypes (Fig. 3b, t(15,15) = 0.2; P = 0.843). Moreover, time constants (τ) for the learning curve indicated a τof 2 days for WT mice, whereas VAChT KDHET mice had a τof 3 days to achieve maximum learning (Fig. 3c). These values are in line with the average performance, but this difference just fails statistical significance (Fig. 3c, Mann–Whitney Rank Sum Test T = 242; P = 0.073). The fact that both genotypes achieved similar maximum performance suggests that the change in the rate of acquisition for motor learning cannot be explained by alterations in physical performance or coordination. This conclusion is supported by gait analysis (Supplemental Fig. 1), which indicates that VAChT KDHET mice present no alterations in stride length or general gait.
To test whether mutant mice have altered ability to retrieve information necessary to perform in the rotarod, we compared performance of the two genotypes in the last trial of a given day with the first trial of the next day. We found no significant differences between the two genotypes (Fig. 3d, two-way repeated measure anova shows no effect of genotype: F(1,180) = 0.0119, P = 9.17; significant effect of intersession: F(6,180) = 8.831, P < 0.001 and no interaction between genotype and intersession: F(6,180) = 0.882, P = 0.509), revealing that in the day to day basis both genotypes are able to retrieve the necessary information learned in the previous day to perform successfully in the rotarod on the next day. We also tested performance retention 10 days after the eighth trial (day 18) by comparing the last trial of day 8 with the performance in the first trial of day 18 for each genotype (Fig. 3e; t(15,15) = 0.376, P = 0.709). VAChT KDHET and WT mice showed no significant differences in performance in the first trial on day 18th, suggesting that performance maintenance was equivalent between genotypes (Fig. 2c; t(15,15) = −1.132, P = 0.267). On day 18, VAChT KDHET and WT mice improved their performance (two-way repeated anova measure shows a significant effect of genotype: F(1,270) = 4.798, P<0.05; significant effect of trial: F(9,270) = 9.698, P < 0.001 and interaction between genotype and trial: F(1,270) = 2.536, P < 0.01) and both genotypes reached similar maximum performance, although KDHET needed more trials to do so (Fig. 2c; Tukey post hoc show significant difference between the genotype only on trial 3, 4, 5, respectively, P < 0.01, P < 0.01, P < 0.001). Overall, these data indicate that procedural learning is less efficient in VAChT KDHET mice.
VAChT KDHET mice have specific alterations in object recognition memory
Cognitive processing in VAChT KDHET mice was tested using the temporal order version of the object recognition task. In agreement with published data, during the test phase of this task WT mice preferred the object presented less recently, as shown by increase exploration of object A (Fig. 4a, clear bars) over object B in the test session (t(7) = 8.056, P < 0.0001). In contrast, mutant mice were unable to distinguish between the more and less recently presented object (Fig. 4a; t(7) = 0.6705, P = 0.524).
To test if reduction of VAChT expression also affects spatial memory, we utilized the hidden platform version of the MWM. There was no effect of genotype in escape latency during training sessions (Fig. 4b; t(7,12) = 0.02010, P = 0.9842) and both VAChT KDHET and WT control littermates spent similar amount of time in the target quadrant during a probe trial carried out 24 h after the last training session (Fig. 4c; t(7,12) = 0.1531, P = 0.8800). We also investigated whether mutant VAChT mice present deficits in spatial working memory. Both genotypes presents alternation scores above chance, 50% (WT: t(17) = 8.457, P < 0.0001; KDHET: t(13) = 5.186, P < 0.0002). Spontaneous alternation in the Y maze shows no effect of genotype in the percent alternation scores (Fig. 4d; t(30) = 0.7383, P = 0.4661) or in the number of arms entered (Fig. 4e; t(30) = 0.9069, P = 0.3717). Taken together, these results suggest that reduced VAChT levels affect object recognition memory, but not spatial or working memory.
Pharmacological restoration of cholinergic tone improves memory encoding in VAChT KD mice
The present observations suggest that VAChT KDHET mice have a specific deficit in the temporal order task of the object recognition memory. Is this a consequence of an acute decrease in cholinergic tone or a general developmental effect related to the persistent reduction in cholinergic tone? If reduced acetylcholine release causes object recognition memory deficits it would be expected that restoring ACh at the synapse with cholinesterase inhibitors could result in reversal of the behavioral deficits. Injection of saline prior to the trial sessions or the test session did not alter the behavior of WT or VAChT KDHET mice (not shown). Furthermore, wild-type mice were able to discriminate against an object presented less recently and treatment with galantamine (0.5, 1.0 and 3 mg/kg) injected 30 min prior to the first sample session (Fig. a–c, respectively) did not alter their performance (t(7) = 3.034; P < 0.05; t(10) = 7.089; P < 0.0001; t(9) = 5.638; P < 0.001; respectively, for 0.5, 1 and 3 mg/kg of galantamine).
At 0.5 mg/kg galantamine did not affect VAChT KDHET performance (Fig. 5a; t(6) = 0.7007, P = 0.5097). However, as can be seen in Fig. 5b,c, treatment with 1 or 3 mg/kg of galantamine significantly improved temporal order memory in VAChT KDHET mice (t(10) = 6.784, P < 0.0001; t(8) = 7.434, P < 0.0001, respectively, for 1 and 3 mg/kg of galantamine), making their performance during the test session indistinguishable from that of WT mice.
Besides its well-known action as an acetylcholinesterase inhibitor, galantamine is also able to directly modulate nicotinic receptors (Pereira et al. 1994). Therefore, to test whether the reversal of recognition memory deficit in VAChT KDHET can also be elicited by a more selective cholinesterase inhibitor, we investigated the action of donepezil (Cummings 2003). Donepezil did not alter the performance of WT control littermates in the temporal order recognition memory task (t(7) = 2.500, P < 0.05), but in agreement with the results obtained with galantamine, donepezil was also able to reverse the deficit of mutant VAChT mice (Fig. 5d; t(7) = 7.196, P < 0.001). Taken together, these results indicate that preserving ACh during memory acquisition can reverse the deficit in the object recognition temporal order task observed in VAChT KDHET mice.
It is possible that increased cholinergic tone in VAChT KDHET mice treated with cholinesterase inhibitors might also facilitate retrieval of information. To investigate this possibility, we analyzed the effect of the pretest administration on memory retention. To do that, we injected galantamine (1 mg/kg) 30 min prior to the test session. WT control mice presented no alteration in performance when injected with galantamine prior to the test session (Fig. 6a; t(7) = 2.648, P < 0.05). Moreover, VAChT KDHET mice showed no sign of improvement in recognition memory when injected with galantamine prior to the test session; mice were unable to discriminate against the more and less recently presented object (Fig. 6a; t(7) = 0.6705, P = 0.524).
We further tested whether cholinesterase inhibition was able to alter VAChT KDHET behavior in a more general way. For this test, we analyzed general locomotor activity of wild-type and VAChT KDHET mice. There was no statistic difference in total spontaneous horizontal activity between control littermates (WT) and VAChT KDHET mice [WT (1778.5 ± 246.7 cm in 60 min, all results described bellow are the mean ± SEM), VAChT KDHET (2558.6 ± 305.3 cm in 60 min); t(7,7) = 1.987, P = 0.067)], although VAChT KDHET mice showed a tendency to higher activity. Rearing was also not statistically different between genotypes [WT (224 ± 43), VAChT KDHET (306 ± 42); t(7,7) = 1.367, P = 0.19]. These results are paralleled by the horizontal locomotor activity measured at 5 min intervals (Fig. 6b). Galantamine administration did not appear to cause any change in locomotion for both genotypes (Fig. 6b). Total horizontal activity after administration of galantamine was similar for both genotypes [WT (703.6 ± 135.5 cm in 120 min), VAChT KDHET (637.2 ± 143.7 cm in 120 min), Mann–Whitney Rank Sum Test (T = 61, P = 0.505)], as was rearing [WT (52 ± 9), VAChT KDHET (71 ± 9); t(7,7) = −1.468, P = 0.164]. Taken together, these experiments indicate that cholinesterase inhibition specifically improves acquisition but not retrieval of object recognition memory in VAChT KDHET mice and it does not change spontaneous locomotor activity in these mice.
- Top of page
- Material and methods
- Supporting Information
The present data revealed a key role of VAChT in fine tuning of processes critical for learning and memory that may have important implications for better understanding several human disorders or consequences of drug treatments that affect VAChT expression. These results provide novel insights on the contribution of vesicular ACh in cognitive processing and suggested that the ability to sustain efficient transport of neurotransmitter into synaptic vesicles is of fundamental importance for learning.
VAChT KD mice present reduced capacity to sustain ACh release (Prado et al. 2006), a consequence of having reduced expression of this transporter. In fact, immunofluorescence analysis indicates that VAChT levels in synaptic terminals are decreased to the extension previously observed in immunoblot analysis. All the central nervous system areas that are relevant for behaviors studied herein showed the same levels of decrease in VAChT immunoreactivity. In contrast, immunoreactivity for CHT1 appears preserved, supporting the notion that the alteration in VAChT expression does not alter other presynaptic cholinergic proteins.
Reduced levels of VAChT selectively affect motor skill learning
Motor learning depends on improvement of accuracy, speed and general ability to perform a task, and it is accepted that this is accompanied by substantial plasticity of cortical representations (Nudo et al. 1996). Rats treated with IgG-192 saporin, a toxin that targets and eliminates basal forebrain cholinergic neurons, have deficits in learning a forelimb reaching task and exhibit alterations in cortical map representation (Conner et al. 2003), pointing to a possible role of distinct cholinergic neuronal groups in motor learning. Interestingly, neither motor coordination and gait, nor maximum performance in rotarod were altered in VAChT KDHET mice, suggesting that motor skill learning is likely to be the only major component affected in these mutants. Our additional experiments evaluating specifically motor learning extended these previous observations and revealed that alterations in vesicular release of ACh modulate the rate of motor skill learning. Interestingly, performance maintenance in the rotarod appears to be minimally affected by reduced VAChT levels. However, although maximum performance could be achieved by VAChT KDHET mice in day 18, the mutants require longer training to achieve that. The reason for this decreased alteration in performance is unknown at the moment.
Cognitive processing in VAChT KD mice
Several studies have addressed neurochemical circuits and anatomical regions involved in novel object recognition memory (Dere et al. 2007; Winters & Bussey 2005a). This kind of memory requires judgment of the previous occurrence of stimuli made on the basis of the relative familiarity of individual objects, by integrating information concerning objects and location and by using recency information (Barker et al. 2006; reviewed by Dere et al. 2007).
The perirhinal cortex is a key player in novel object recognition memory (Barker et al. 2006; Winters & Bussey 2005b), whereas the involvement of synaptic transmission in the hippocampus is controversial (Forwood et al. 2005; but see also Dere et al. 2007). There has been also controversy on the role of cholinergic system in novel object recognition memory based on lesion studies or studies with the p75 neurotrophic receptor selective immunotoxin 192-IgG saporin. Some studies show that the basal forebrain cholinergic system is important for object recognition memory with cholinergic terminals in the perirhinal cortex having a major role (Warburton et al. 2003; Winters & Bussey 2005a). Conversely others failed to find an effect with electrolytic lesions in the medial septum (Ennaceur 1998) or found effects of 192-saporin only long after the lesions were induced (Paban et al. 2005), suggesting that the period of testing after the lesion is critical. In contrast, infusion of scopolamine provided strong evidence that cholinergic tone can have profound influence in encoding that is important for object recognition memory (Winters et al. 2006, 2007). Our results are consistent with these last observations and suggest that for object recognition recency memory, the ability to sustain cholinergic tone is also critical. Importantly, decreased cholinergic tone does not affect general behavior nor does it have unspecific effects in information processing, as spatial memory appears to be preserved in VAChT KDhet mice (close to 40% decrease in VAChT expression).
The absence of spatial memory deficits in the MWM is unexpected, as some studies investigating the consequences of cholinergic lesions have suggested a role of acetylcholine in spatial memory (Dickinson-Anson et al. 1998; LeBlanc et al. 1999; Nilsson et al. 1987, 1992). In contrast, other studies using more selective lesions of cholinergic neurons failed to detect alterations in performance in the MWM task (Baxter & Gallagher 1996; Baxter et al. 1996; Frielingsdorf et al. 2006; Leanza et al. 1995). These observations are consistent with experiments using mice null for M1 muscarinic receptor that show spared spatial memory in this task (Miyakawa et al. 2001). However, because of the limited reduction in VAChT levels in VAChT KDHET mice it is possible that more pronounced reductions in cholinergic tone could still affect spatial memory. To test this possibility directly future studies involving conditional brain specific VAChT knockout mice would be necessary.
The present experiments also addressed the question of whether decreased VAChT levels affect encoding or retrieval of object recognition memory. The ability of cholinesterase inhibitors to reverse the object recognition memory impairment in VAChT KDHET mice was strictly dependent on improving cholinergic tone during encoding of information. Indeed, mice treated with cholinesterase inhibitors before memory retrieval performed as untreated VAChT KDHET mice. The data suggest that cognitive alterations responsible for the failure of VAChT KDHET mice to perform normally in object recognition memory tasks are specifically related to learning deficits. It is important to note that these cholinesterase inhibitors did not increase the exploration time during the sample phase of the task, or change exploration in the open field, suggesting that increased cholinergic tone indeed facilitated encoding information more effectively. These results expand the role of cholinergic tone to facilitate recency memory and are consistent with recent observations that indicate a role for muscarinic receptors in encoding information related to the novel object recognition task (Winters et al. 2006, 2007).
Cholinergic tone may assist memory encoding by either facilitating excitatory input to the cortex (Gioanni et al. 1999) or by modulating synaptic plasticity (Dringenberg et al. 2007). Alternatively, cholinergic activity may decrease excitatory feedback circuitry and this interference can affect information encoding (Hasselmo 2006; Hasselmo & Bower 1992; Hasselmo et al. 1992; Winters et al. 2007). Finally, cholinergic tone may regulate electrical properties of individual cells facilitating firing (Fransen et al. 2002; Klink & Alonso 1997). All of these mechanisms can be envisioned to help recruiting neuronal assemblies to encode information. Targeted VAChT mice therefore provide the means to probe how cholinergic tone regulates cellular mechanisms of learning.
Synaptic vesicles are filled with thousand molecules of ACh, however, the turnover rate for mammalian VAChT to take up ACh into vesicles is slow (approximately 1/s) (Varoqui & Erickson 1996). Estimates of vesicle recycling indicate that reuse of vesicles occurs fast after exocytosis, in the order of 15–30 s (Betz & Bewick 1992; Ryan et al. 1993). These observations suggest that synaptic vesicle filling in cholinergic terminals may not reach electrochemical equilibrium during high levels of activity. Central nervous system synapses, with their small pool of synaptic vesicles, might be particularly sensitive to reductions in VAChT levels. This should preferentially affect functions that depend on bursts of presynaptic activity demanding effective recycling of vesicles and ACh to maintain cholinergic tone.
A decrease in VAChT has been reported in Alzheimer’s disease (Efange et al., 1997). A reduction of VAChT expression has been found in striatal tissue from patients with Huntington´s disease and is also observed in an animal model of this disorder (Smith et al. 2006). Moreover, decreased levels of VAChT were reported in rats after sepsis (Semmler et al. 2007) and treatment with antipsychotics, which seems to contribute to cognitive deficits observed in this last experimental model (Terry et al. 2007). Hence, diminished vesicular storage of ACh because of decreased VAChT expression may contribute to cognitive and neurological symptoms in these disorders. Our data indicate that the ability to sustain ACh release during neuronal activity is important for learning and raise the possibility for a conserved role of cholinergic tone in facilitating explicit and implicit memory encoding. These experiments also suggest that deficits of cholinergic tone because of decreased VAChT levels can be ameliorated by treatments that restore acetylcholine levels in the synapse.
- Top of page
- Material and methods
- Supporting Information
- 2006) The different effects on recognition memory of perirhinal kainate and NMDA glutamate receptor antagonism: implications for underlying plasticity mechanisms. J Neurosci 26, 3561–3566. , , , , , , , , & . (
- 1996) Intact spatial learning in both young and aged rats following selective removal of hippocampal cholinergic input. Behav Neurosci 110, 460–467. & . (
- 1996) Intact spatial learning following lesions of basal forebrain cholinergic neurons. Neuroreport 7, 1417–1420. , , , , & . (
- 1992) Optical analysis of synaptic vesicle recycling at the frog neuromuscular junction. Science 255, 200–203. & . (
- 2006) Object recognition in rats and mice: a one-trial non-matching-to-sample learning task to study ‘recognition memory’. Nat Protoc 1, 1306–1311. & . (
- 2004) Choline transporter 1 maintains cholinergic function in choline acetyltransferase haploinsufficiency. J Neurosci 24, 5459–5466. , , , , , , , , , , & . (
- 2003) Lesions of the basal forebrain cholinergic system impair task acquisition and abolish cortical plasticity associated with motor skill learning. Neuron 38, 819–829. , , , & . (
- 2003) Use of cholinesterase inhibitors in clinical practice: evidence-based recommendations. Am J Geriatr Psychiatry 11, 131–145. (
- 2005) Episodic-like memory in mice: simultaneous assessment of object, place and temporal order memory. Brain Res Brain Res Protoc 16, 10–19. , & . (
- 2007) The pharmacology, neuroanatomy and neurogenetics of one-trial object recognition in rodents. Neurosci Biobehav Rev 31, 673–704. , & . (
- 1998) Hippocampal grafts of acetylcholine-producing cells are sufficient to improve behavioural performance following a unilateral fimbria-fornix lesion. Neuroscience 84, 771–781. , , & . (
- 2005) A model for dynamic regulation of choline acetyltransferase by phosphorylation. J Neurochem 95, 305–313. & . (
- 2007) Heterosynaptic facilitation of in vivo thalamocortical long-term potentiation in the adult rat visual cortex by acetylcholine. Cereb Cortex 17, 839–848. , , , & . (
- 2007) The neurotransmitter cycle and quantal size. Neuron 55, 835–858. . (
- 1997) Vesicular acetylcholine transporter density and Alzheimer’s disease. Neurobiol Aging 18, 407–413. , , , & . (
- 1998) Effects of lesions of the substantia innominata/ventral pallidum, globus pallidus and medial septum on rat’s performance in object-recognition and radial-maze tasks: physostigmine and amphetamine treatments. Pharmacol Res 38, 251–263. . (
- 2005) Hippocampal lesions that abolish spatial maze performance spare object recognition memory at delays of up to 48 hours. Hippocampus 15, 347–355. , & . (
- 2002) Simulations of the role of the muscarinic-activated calcium-sensitive nonspecific cation current INCM in entorhinal neuronal activity during delayed matching tasks. J Neurosci 22, 1081–1097. , & . (
- 2006) The septohippocampal cholinergic system and spatial working memory in the Morris water maze. Behav Brain Res 168, 37–46. , & . (
- 1999) Nicotinic receptors in the rat prefrontal cortex: increase in glutamate release and facilitation of mediodorsal thalamo-cortical transmission. Eur J Neurosci 11, 18–30. , , , , & . (
- 2006) The role of acetylcholine in learning and memory. Curr Opin Neurobiol 16, 710–715. . (
- 1992) Cholinergic suppression specific to intrinsic not afferent fiber synapses in rat piriform (olfactory) cortex. J Neurophysiol 67, 1222–1229. & . (
- 1992) Cholinergic modulation of cortical associative memory function. J Neurophysiol 67, 1230–1246. , & . (
- 2004) The value of spontaneous alternation behavior (SAB) as a test of retention in pharmacological investigations of memory. Neurosci Biobehav Rev 28, 497–505. . (
- 1997) Muscarinic modulation of the oscillatory and repetitive firing properties of entorhinal cortex layer II neurons. J Neurophysiol 77, 1813–1828. & . (
- 1995) Selective lesioning of the basal forebrain cholinergic system by intraventricular 192 IgG-saporin: behavioural, biochemical and stereological studies in the rat. Eur J Neurosci 7, 329–343. , , & . (
- 1999) Morris water maze analysis of 192-IgG-saporin-lesioned rats and porcine cholinergic transplants to the hippocampus. Cell Transplant 8, 131–142. , , , , & . (
- 2001) Hyperactivity and intact hippocampus-dependent learning in mice lacking the M1 muscarinic acetylcholine receptor. J Neurosci 21, 5239–5250. , , & . (
- 1987) Spatial learning and memory following fimbria-fornix transection and grafting of fetal septal neurons to the hippocampus. Exp Brain Res 67, 195–215. , , , & . (
- 1992) Spatial learning impairments in rats with selective immunolesion of the forebrain cholinergic system. Neuroreport 3, 1005–1008. , , , , & . (
- 1996) Use-dependent alterations of movement representations in primary motor cortex of adult squirrel monkeys. J Neurosci 16, 785–807. , , & . (
- 2005) Time course of behavioral changes following basal forebrain cholinergic damage in rats: environmental enrichment as a therapeutic intervention. Neuroscience 132, 13–32. , , , & . (
- 2000) Transport mechanisms in acetylcholine and monoamine storage. FASEB J 14, 2423–2434. . (
- 1994) Physostigmine and galanthamine: probes for a novel binding site on the alpha 4 beta 2 subtype of neuronal nicotinic acetylcholine receptors stably expressed in fibroblast cells. J Pharmacol Exp Ther 270, 768–778. , , , , , , & . (
- 2006) Mice deficient for the vesicular acetylcholine transporter are myasthenic and have deficits in object and social recognition. Neuron 51, 601–612. , , . et al. (
- 2006) Effects of injections of glucose into the dorsal striatum on learning of place and response mazes. Behav Brain Res 167, 373–378. , & . (
- 2005) Constitutive high-affinity choline transporter endocytosis is determined by a carboxyl-terminal tail dileucine motif. J Neurochem 94, 86–96. , , , , , & . (
- 2006) The “ins” and “outs” of the high-affinity choline transporter CHT1. J Neurochem 97, 1–12. , , , , & . (
- 2006) Retrieval induces hippocampal-dependent reconsolidation of spatial memory. Learn Mem 13, 431–440. , , , & . (
- 1993) The kinetics of synaptic vesicle recycling measured at single presynaptic boutons. Neuron 11, 713–724. , , , , & . (
- 2007) Long-term cognitive impairment, neuronal loss and reduced cortical cholinergic innervation after recovery from sepsis in a rodent model. Exp Neurol 204, 733–740. , , , , , & . (
- 2006) Cholinergic neuronal defect without cell loss in Huntington’s disease. Hum Mol Genet 15, 3119–3131. , , , , , , , , , & . (
- 2004) Dopamine transporter-dependent and -independent actions of trace amine beta-phenylethylamine. J Neurochem 91, 362–373. , , , , & . (
- 2007) Oral haloperidol or risperidone treatment in rats: temporal effects on nerve growth factor receptors, cholinergic neurons, and memory performance. Neuroscience 146, 1316–1332. , , , , , , , & . (
- 1996) Active transport of acetylcholine by the human vesicular acetylcholine transporter. J Biol Chem 271, 27229–27232. & . (
- 2003) Cholinergic neurotransmission is essential for perirhinal cortical plasticity and recognition memory. Neuron 38, 987–996. , , , , , , , & . (
- 2005a) Removal of cholinergic input to perirhinal cortex disrupts object recognition but not spatial working memory in the rat. Eur J Neurosci 21, 2263–2270. & . (
- 2005b) Transient inactivation of perirhinal cortex disrupts encoding, retrieval, and consolidation of object recognition memory. J Neurosci 25, 52–61. & . (
- 2006) Paradoxical facilitation of object recognition memory after infusion of scopolamine into perirhinal cortex: implications for cholinergic system function. J Neurosci 26, 9520–9529. , & . (
- 2007) Scopolamine infused into perirhinal cortex improves object recognition memory by blocking the acquisition of interfering object information. Learn Mem 14, 590–596. , , & . (
- Top of page
- Material and methods
- Supporting Information
We appreciate the technical assistance and outstanding care for mouse colonies from Diogo Souza, Danuza M. Diniz, Adriane Pereira and Wendy Roberts. We also thank Ms. Greiciane P. Lages for her participation in the experiments testing motor skill leaning and Dr. Michel Cyr (Neuroscience Research Group, Université du Québec à Trois-Rivières, Canada) for advice on the immunofluorescence experiments. This work was supported by grants from the NIH-Fogarty Center (TWR03-007025 and TWR21-007800 to M. A. M. P., V. F. P., I. I., R. R. G. and M. G. C.), CNPq (Mental Health and Aging programs to I. I., M. C. and M. A. M. P), MCT-Millenium Institute, FINEP, PRONEX-MG, (FAPEMIG to M. A. M. P and V. F. P.) and fellowships from CAPES, FAPEMIG and CNPq.
- Top of page
- Material and methods
- Supporting Information
Additional Supporting Information may be found in the online version of this article.
Figure S1. Gait analysis of VAChT KD mice. (a) Stride length (mean ± SEM) for WT (white bar), VAChT KDHET (gray bar) was measured between two steps from the same leg. (b) Representative examples of foot prints for WT, VAChT KDHET. n = 8 for both genotypes.
Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author.
|GBB_439_sm_FigureS1.tif||1283K||Supporting info item|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.