the effects of calcium channel antagonists on short- and long-term retention in mice using...

Neurobiology of Learning and Memory 76, 117–124 (2001)

doi:10.1006/nlme.2000.3981, available online at http://www.idealibrary.com on

BRIEF REPORT

The Effects of Calcium Channel Antagonists onShort- and Long-Term Retention in Mice Using

Spontaneous Alternation Behavior

David Quartermain and Victoria Garcia deSoria

New York University School of Medicine, New York, New York 10016

Published online May 8, 2001

The effect of calcium channel antagonists (CCA’s) on working and referencememory in mice was studied using spontaneous alternation (SA) behavior in a Tmaze. Mice were given either one or four forced trials to either the right or theleft arm on the training session (T1) followed by a free choice test (T2) at varyingintervals after the initial trial. Untreated animals given one forced trial exhibitedsignificantly greater levels of SA than chance at all delay intervals out to 20 minbut not at 30, 60, or 180 min. Animals given four forced trials showed significantlevels of SA 24 h after exposure but not at 72 h. Additional groups of mice weretreated with amlodipine, nimodipine, diltiazem, and verapamil 1 h before T1. Micegiven one forced trial were tested 30, 60, or 180 min after T1 while mice givenfour forced trials were tested 72 h after T1. Results showed that all of the CCA’sexcept verapamil produced significant SA at the 30-min interval and nimodipineand diltiazem also significantly increased SA at the 60-min-delay interval. Nosignificant effects were observed at the 180-min test. In the four trial groups, allof the CCA’s with the exception of verapamil produced significant levels of SA atthe 72-h interval. These results indicate that representative CCA’s from both thedihydropyridine and the benzothiazapine classes can facilitate both short- and long-interval SA, thereby providing further confirmation that CCA’s can enhance memoryprocessing in young animals. q 2001 Academic Press

Key Words: spontaneous alternation; working memory; long-term memory; cal-cium channel antagonists; amlodipine; nimodipine; diltiazem; verapamil; mice.

Calcium channel antagonists (CCA’s) are a biochemically heterogeneous group of drugswhich selectively block the influx of calcium ions into cells in both the peripheral andthe central nervous systems. They are all potent vasodilators and are used clinically to

Address correspondence and reprint requests to David Quartermain, Laboratory of Behavioral Neurology,Department of Neurology, NYU School of Medicine, 550 1st Avenue, New York, NY 10016. Fax: 212-263-2880. E-mail: [email protected].

117 1074-7427/01 $35.00Copyright q 2001 by Academic Press

All rights of reproduction in any form reserved.

118 QUARTERMAIN AND GARCIA DESORIA

treat cardiovascular and neurological disorders such as hypertension, angina, migraine,and cerebrovascular disease. CCA’s have also been shown to reverse age-associatedimpairments in sensorimotor and cognitive behavior in animals (e.g., Deyo, Straube, &Disterhoft, 1989; Sandin, Jasmin, & Levere, 1990). However, their effects on cognitivebehavior in neurologically normal young adult animals have been ambiguous. Some studieshave shown that they can facilitate memory (e.g., Deyo & Hittner, 1995; Quartermain,Hawxhurst, Ermita, & Puente, 1993; McMonagle-Strucko & Fanelli, 1993), while othershave shown either no effects or a memory impairment following their administration (e.g.,Clements, Rose, & Tiunova, 1995; Maurice, Bayle, & Privat, 1995).

In an attempt to clarify the role of CCA’s in long-term retention we recently comparedthe effects of posttraining administration of three different types of CCA’s on both single-trial inhibitory avoidance and multitrial thirst-motivated maze learning in young adultmice. Results showed that all of the drugs produced significant memory enhancement ona 24-h retention test in at least one of the learning tasks (Quartermain, Garcia deSoria, &Kwan, 2001). In the current study we examined the effects of CCA’s on spontaneousalternation (SA), a procedure which allowed us to investigate the effects of the antagonistson both short- and long-term retention using a common behavioral task. SA has beenwidely used to investigate short-term memory (e.g., Isseroff, 1979; Stone, Walser, Gold, &Gold, 1991). The assumption underlying the use of this procedure is that in order toalternate in a maze where odor trails are eliminated, animals must retain in memoryrepresentations of the stimuli and responses associated with a particular arm entry so thatit can be recognized on trial 2. Evidence suggests (Douglas, 1966; Hughes, 1998) thatthe memory trace consists of response-produced spatial cues mediated by the vestibularsystem combined with relevant visual cues (e.g., in the forced trial procedure in the Tmaze, visual cues from the door blocking the other arm). After a single exposure at T1in the forced trial procedure, mice will typically alternate if the interval between T1 andT2 is less than 30 min, suggesting that short-term or working memory is maintaining therepresentation of T1. Spontaneous alternation can occur at longer intervals if the frequencyof exposure to the arm at T1 is increased (Beracochea & Jaffard, 1985). In the presentexperiment we attempted to produce a more durable memory of T1 by giving mice fourforced trials to the same arm at T1. Pilot data indicated that under these training conditionsmice could alternate reliably with a 24-h interval between T1 and T2. The objective ofthe present experiment therefore was to compare the effects of CCA’s on both short (oneforced trial at T1) and long (four forced trials at T1) term retention. CCA’s from thethree major classes (dihydropyridines, benzothiazapines, and phenylalkylamines) wereadministered 1 h prior to T1 at dose levels that we had previously shown to facilitate24-h retention of passive avoidance and linear maze learning.

Subjects were male Swiss Webster mice (Hsd: ND4; Harlan) 6–8 weeks of age andbetween 25 and 30 g body wt. Animals were housed five per cage with food and wateravailable ad libitum. Animals were brought into the experimental room 1 h before com-mencement of testing.

The apparatus was a T maze constructed from black Plexiglas with a floor made fromstainless-steel rods spaced 1 cm apart. The start box measured 9 3 9 3 6 cm and wascovered by a hinged lid constructed from clear Plexiglas. A horizontal sliding doorprovided access to the enclosed stem of the maze that measured 19 3 6 3 6 cm. Thetwo arms measured 14 3 6 3 6 cm and were covered with clear hinged lids. Access to

EFFECT OF CALCIUM CHANNEL ANTAGONISTS ON SPONTANEOUS ALTERNATION 119

the arms was controlled by horizontal sliding doors. The maze was situated on a table 1m above floor level.

The following drugs were used: amlodipine, 5 mg/kg (Pfizer); nimodipine, 0.1 mg/kg(Miles); diltiazem, 5 mg/kg (Marion), and verapamil, 2 mg/kg (Knoll). Amlodipine andnimodipine represent the dihydropyridine class, diltiazem is a benzothiazapine, and vera-pamil is a phenylalkylamine CCA. Amlodipine, diltiazem, and verapamil were dissolvedin distilled water. Nimodipine was dissolved in a solution made up of 62.7% distilledwater, 20% ethanol, 17% PEG 400, 0.2% sodium citrate, and 0.1% citric acid. Vehicle-treated animals were injected with distilled water. Drugs were prepared fresh daily. Thechoice of concentration for each drug was based on pilot data and on other evidence fromthe laboratory that these dose levels produced maximum facilitation of long-term memoryin either passive avoidance or spatial discrimination learning tasks (Quartermain et al.,2001).

Rates of alternation for the retention interval and drug groups were analyzed by the x 2

test for independent samples followed by planned comparisons using 2 3 2 x 2 tests(Douglas, 1966; Squire, 1969).

1. Determination of temporal gradients of forgetting. Mice were injected with distilledwater and 1 h later placed in the start box and one arm of the maze was blocked byclosing the sliding door. For half of the animals, the left arm was closed, and for theremaining half the right was closed. When the mouse entered the open arm, the door wasshut and after 15 s the animal was removed and returned to the home cage for the retentioninterval which was 0, 10, 15, 20, 30, 60, or 180 min in duration. For the zero-min group,animals were placed in the home cage for 5 s and then put in the start compartment ofthe maze. The floor and walls of the maze as well as the tabletop under the maze werecleaned with a damp cloth after each T1 and T2 trial. The 5 mice in each cage wererandomly assigned to a different time interval (N 5 22–24/interval group). On the testtrial, both doors were open, allowing the mice a free choice of arm entry. Additionalgroups of mice were given four exposures to either the right or the left arm on the trainingsession. The procedure for each trial was the same as that described above except thatafter 15 s in the arm the animal was returned to the start compartment and given the nexttrial. After the fourth trial the animals were returned to the home cage for the retentioninterval. To determine the temporal gradient of forgetting for four training trials, groupsof mice (N 5 17–18/group) were tested 3, 6, 24, 72, or 120 h after the training session.Each mouse in the cage of 5 was randomly assigned to one of the interval groups.

The percentage of mice alternating in each retention interval group for both the oneand the four training trial groups is shown in Fig. 1. A x 2 test for independent groupscarried out on the one-trial data revealed a significant difference among the seven intervalgroups (x 2(6) 5 28.91; p 5 , .001) Planned comparisons using a 2 3 2 x 2 test indicatedthat mice tested 30 min after one training trial alternated significantly less than micetested 0, 10, 15, and 20 min after initial arm exposure. A similar analysis applied to thefour-trial data indicated that there was a significant difference in rates of alternation amongthe five interval groups (x 2(4) 5 17. 14; p 5 .002) and that the 3-, 6-, and 24-h retentiongroups alternated at significantly higher rates than the group tested 72 h after training.The results show that mice given one forced trial at T1 exhibit alternation levels signifi-cantly different from chance with intervals between T1 and T2 as long as 20 min but not


FIG

.1.

Mea

npe

rcen

tage

ofal

tern

atio

nas

afu

ncti

onof

the

T1–

T2

inte

rval

for

mic

egi

ven

eith

eron

eor

four

forc

edtr

ials

atT

1.


at longer intervals. Mice given four forced trials at T1 show high rates of alternationwhen the T1–T2 interval is 24 h or less but not when the interval is 72 h or longer.

2. Effects of CCA’s on the duration of spontaneous alternation. The intention of thesecond part of the study was to determine whether animals treated with selected CCA’sbefore T1 would demonstrate significant levels of SA at intervals where performance wasat chance levels in untreated mice. In the one-trial procedure, different groups of mice(N 5 22–24/group) were injected with either the vehicle or one of the test CCA’s 1 hbefore acquisition and tested at 30, 60, or 180 min after training. In the four-trial procedure,groups of mice (16–19/group) were treated with either the vehicle or one of the CCA’s1 h before the acquisition session and tested 72 h later, a time at which untreated micealternate at chance levels. Results of attempts to extend the one-trial temporal gradientare shown in Fig. 2.

A x 2 test for independent groups revealed a significant difference among both the 30-min and the 60-min groups (x 2(4) 5 13.026; p 5 .001, and x 2(4) 5 10.114; p 5 .039,respectively). Post hoc 2 3 2 x 2 tests (Fig. 2) indicated that amlodipine-, nimodipine-,and diltiazem-treated groups all alternated at significantly higher levels than the vehicle-treated mice at the 30-min interval and nimodipine- and diltiazem-treated mice exhibitedsignificant levels of alternation relative to vehicle-treated animals at the 60-min interval.None of the drugs produced significant SA at the 180-min interval.

The effects of treatment with CCA’s on the gradient of SA for the four-trial groups areshown in Fig. 3. Results of a x 2 test revealed significant differences among the fivetreatment groups (x 2(4) 5 9.494; p 5 .050). Post hoc tests indicated that all of the druggroups except verapamil exhibited significantly higher rates of SA than the vehicle-treated controls.

The temporal gradient data indicate that after a single exposure to an arm mice spontane-ously alternate at levels significantly greater than chance with delays as long as 20 min.This estimate of working memory is considerably longer than has been reported for food-deprived mice rewarded with milk for alternating in a T maze using a continuous 10-trial

FIG. 2. Effect of calcium channel antagonists on spontaneous alternation tested 30, 60, or 180 min aftera single forced trial at T1. Drugs were administered 1 h before T1.


FIG. 3. Effect of calcium channel antagonists on spontaneous alternation tested 72 h after four forced trialsat T1. Drugs were administered 1 h before T1.

procedure (Ritzmann, Kling, Melchior, & Glasky, 1993). Under these conditions, adultSwiss Webster mice could alternate successfully with a 60 but not a 90-s intertrial interval.The shorter duration of working memory reported in this study is probably the result ofprocedural variables that may make recency discriminations more difficult. Factors suchas stress induced by multiple-trial handling and a stringent deprivation schedule combinedwith possible accumulation of proactive interference are likely to have resulted in a shorterworking memory duration than might occur in single-trial spontaneous alternation innondeprived animals. With increasing number of trials at T1, working memory shadesinto long-term memory. Four exposures to the arm at T1 increase the duration over whichmice can spontaneously alternate to 24 h, a time interval typically used to assess long-term memory. This is, however, a relatively weak memory since alternation performancedeclines to chance levels 72 h after exposure to T1.

The results of administration of CCA’s prior to T1 indicate that duration of both short-and long-term memory can be extended by drug treatment. All of the agents exceptverapamil produced significant levels of SA at the 30-min intertrial delay and nimodipineand diltiazem also induced significant alternation at the 60-min interval. No drugs wereable to extend the duration of working memory to 180 min. A duration of around 60 minmay be the longest time interval that the rodent working memory system is designed tohandle (Markowska, Buresova, & Bures, 1983). Since rats patrol their environmentsapproximately every 1–2 h (Barnett, Dickson, Marples, & Radha, 1978; Cowan, 1977),a duration of working memory much longer than this time would not be biologicallyadvantageous.

All CCA’s with the exception of verapamil also extended the duration of long-termmemory to at least 72 h. This finding confirms other studies from this laboratory thathave shown that nimodipine, amlodipine, and diltiazen strengthen long-term memory forother multitrial instrumental learning tasks when administered immediately after the train-ing session (Quartermain et al., 1993, 2001). The absence of significant facilitation follow-ing verapamil in both short- and long-term tests of SA is noteworthy. In our previousstudies we also observed that verapamil produced inconsistent effects over a wide dose


range. Although verapamil is a powerful vasodilator, it differs from other CCA’s inthat it depresses cardiac activity (van Zwieten & Pfaffendorf, 1993), antagonizes 5-HT2receptors, and blocks the uptake of 5-HT, DA, and NE into forebrain synaptosomes(DeFeudis, 1987). It is, however, unclear how any of these actions would reduce thememory-enhancing effects of verapamil.

The mechanisms by which CCA’s facilitate short- and long-term retention are notknown. Their potent vasodilating potential has suggested the hypothesis that facilitationof learning and memory may result from the enhancement of peripheral blood flow(Deyo & Hittner, 1995). These authors showed that while peripheral administration ofthe CCA flunarizine facilitated retention, bypassing the vascular system with centraladministration caused memory disruption. However, two recent studies have reportedmemory enhancement from central administration of CCA’s (Quevedo, Vianna, Daroit,Born, Kuyven, Roesler, & Quiltfeldt, 1998; Quartermain et al., 2001). The role of vasodila-tion in CCA-induced retention enhancement is complicated by evidence that some vasodila-tors (e.g., sodium nitrite) can induce amnesia at low doses. (Martinez, Jensen, Vasquez,Lacob, McGaugh, & Purdy, 1979). Another possibility suggested by the work of Disterhoftand his associates is that short- and long-term memory facilitation may be a consequenceof increased neuronal excitability resulting from the direct blockade of L-type channelson neurons (Thompson, Moyer, Black, & Disterhoft, 1992; Disterhoft, Moyer, & Thomp-son, 1994). We are currently investigating the role of calcium channel blockade in CCA-induced memory enhancement by the use of enantiomers of amlodipine and other CCA’swhich are inactive or less active than the racemate at calcium channels.

REFERENCES

Barnett, S., Dickson, R., Marples, T., & Radha, E. (1978). Sequences of feeding sampling and exploration inwild and laboratory rats. Behavioral Processes, 3, 29–43.

Beracochea, D., & Jaffard, R. (1985). Memory deficits subsequent to chronic consumption of alcohol in mice:An analysis based on spontaneous alternation behavior. Behavioural Brain Research, 15, 15–25.

Clements, M. P., Rose, S. P. R., & Tiunova, A. (1995). v-Conotoxin GVIA disrupts memory formation in theday-old chick. Neurobiology of Learning and Memory, 64, 276–284.

Cowan, P. (1977). Systematic patrolling and orderly behaviour of rats during recovery from deprivation. AnimalBehaviour, 25, 171–184.

DeFeudis, F. V. (1987). Interactions of calcium antagonists with neurotransmitter receptors and neurotransmitteruptake processes—Overview. Drugs of Today, 23(6), 347–361.

Deyo, R., Staube, K., & Disterhoft, J. (1989). Nimodipine facilitates associative learning in aged rabbits. Science,243, 809–811.

Deyo, R., & Hittner, J. (1995). Effects of the Ca2+ channel antagonist flunarazine on visual discriminationlearning. Neurobiology of Learning and Memory, 64, 10–16.

Disterhoft, J. F., Moyer, J. R., & Thompson, L. T. (1994). The calcium rationale in aging and Alzheimer’sdisease: Evidence from an animal model of normal aging. In J. F. Disterhoft, W. H. Gispen, J. Traber, &Z. Khachaturian (Eds), Calcium rationale for aging and dementia. Annals of New York Academy of Sciences,747, 382–406.

Douglas, R. J. (1966). Cues for spontaneous alternation. Journal of Comparative and Physiological Psychology,62(2), 171–183.

Hughes, R. N. (1998). Spontaneous alternation behavior in animals: Mechanisms, motives and applications. InR. R. Hoffman, M. F. Sherrick, & J. S. Wam (Eds.), Viewing Psychology as a whole: The integrativescience of William M. Dember (pp. 269–286). Washington, DC: American Psychological Association.


Isseroff, A. (1979). Limited recovery of spontaneous alternation after extensive hippocampal damage: Evidencefor a memory impairment. Experimental Neurology, 64, 284–294.

Markowska, A., Buresova, O., & Bures, J. (1963). An attempt to account for controversial estimates of workingmemory persistence in the radial maze. Behavioural and Neural Biology, 38, 97–112.

Martinez, J. L., Jr., Jensen, R. A., Vasquez, B. J., Lacob, J. S., McGaugh, J. L., & Purdy, R. E. (1979). Acquisitiondeficits induced by sodium nitrite in rats and mice. Psychopharmacology, 60, 221–228.

Maurice, T., Bayle, J., & Privat, A. (1995). Learning impairment following acute administration of the calciumantagonist nimodipine in mice. Behavioral Pharmacology, 6(2), 167–175.

McMonagle-Strucko, K., & Fanelli, R. (1993). Enhanced acquisition of reversal training in a spatial learningtask in rats treated with chronic nimodipine. Pharmacology, Biochemistry, and Behavior, 4, 827–835.

Quartermain, D., Garcia deSoria, V., & Kwan, A. (2001). Calcium channel antagonists enhance retention ofpassive avoidance and maze learning in mice. Neurobiology of Learning and Memory, 75, 77–90.

Quartermain, D., Hawxhurst, A., Ermita, B., & Puente, J. (1993). Effect of the calcium channel blocker amlodipineon memory in mice. Behavioral and Neural Biology, 60, 211–219.

Quevedo, J., Vianna, M., Daroit, D., Born, A., Kuyven, C., Roesler, R., & Quillfeldt, J. (1998). L-Typevoltage-dependent calcium channel blocker nifedipine enhances memory retention when infused into thehippocampus. Neurobiology of Learning and Memory, 69, 320–325.

Ritzmann, R., Kling, A., Melchior, C., & Glasky, A. (1993). Effect of age and strain on working memory inmice as measured by a win-shift paradigm. Pharmacology, Biochemistry, and Behavior, 44(4), 805–807.

Sandin, M., Jasmin, S., & Levere, T. E. (1990). Aging and cognition: Facilitation of recent memory in agednon-human primates by nimodipine. Neurobiology of Aging, 11, 573–575.

Squire, L. R. (1969). Effects of pretrial and posttrial administration of cholinergic and anticholinergic drugs onspontaneous alternation. Journal of Comparative and Physiological Psychology, 69(1), 69–75.

Stone, W. S., Walser, B., Gold, S. D., & Gold, P. E. (1991). Scopolamine and morphine-induced impairmentsof spontaneous alternation performance in mice: Reversal with glucose and with cholinergic and adrenergicagonists. Behavioral Neuroscience, 105(2), 264–271.

Thompson, L., Moyer, J., Black, J., & Disterhoft, J. (1992). Cellular mechanisms for nimodipine’s reductionof age-related learning deficits. In E. Meyer (Ed), Treatment of dementias: A new generation of progress(pp. 241–246). New York: Plenum Press.

Van Zwieten, P., & Pfaffendorf, M. (1993). Similarities and differences between calcium antagonists: Pharmaco-logical aspects. Journal of Hypertension, 11(Suppl.1), S3–S11.

the effects of calcium channel antagonists on short- and long-term retention in mice using...

Documents