BEHAVIOR ANALYSIS in NEUROSCIENCE - PART 7 doc

Although some investigators test working and reference memory on an 8-arm maze with four baited and four unbaited arms, this lessens the sensitivity of the task as the memory demands for

Trang 1

Chapter

Use of the Radial-Arm Maze to Assess Learning and Memory in Rodents

Edward D Levin

Contents

I Introduction

II Radial-Arm Maze Design

A 8-Arm Maze

B 16-Arm Maze

C Mouse Radial-Arm Maze III Procedures

A Adaptation

B Win-Shift Acquisition

C Working/Reference Memory

D Repeated Acquisition

E Delayed Matching to Position

F Non-Spatial Discrimination

IV Data Analysis References

Abstract The radial-arm maze has proven to be a very useful technique for assessing spatial learning and memory in rodents Many different sizes of radial-arm mazes have been used, with the most common being the 8-arm maze The radial maze takes advantage

of rodents’ natural tendency to explore new places for food reinforcement They

0704/C12/frame Page 189 Monday, July 17, 2000 5:31 PM

Trang 2

194 Methods of Behavior Analysis in Neuroscience

baited to unbaited Although some investigators test working and reference memory

on an 8-arm maze with four baited and four unbaited arms, this lessens the sensitivity

of the task as the memory demands for the location of four baits is low We have found that use of a 16-arm maze with 12 baited and 4 unbaited arms provides a useful measure of working vs reference memory while keeping task demands high

Figure 12.3 shows an example of working and reference memory performance on

a 16-arm radial maze with the muscarinic antagonist scopolamine causing a selective impairment in working memory performance

FIGURE 12.2

Acquisition of win-shift 8-Arm radial maze performance by mice 31

7.0

6.0 5.5 5.0 4.5 4.0 6.5

Sessions

Control for Knockout SOD Knockout Control for Overexpressers SOD Overexpressers

8-Arm Radial M aze Learning

and EC- SOD Overexpressing Mice

Trang 3

Use of the Radial-Arm Maze to Assess Learning and Memory in Rodents 197

is made Random chance performance for the entries to repeat measure on an 8-arm maze as determined by computer simulation is 3.25 The response latency measure is the total session duration divided by the number of arms entered (seconds per entry)

It is common to block sessions for analysis In this way a more stable measure

of performance can be attained and sessions in which the rat does not choose enough arms to provide a valid choice accuracy score before the maximum session length

FIGURE 12.4

Repeated acquisition in the 8-Arm radial maze 31

Repeated Acquisition in the 8-Arm Radial Effects of the Nicotinic Agonist AR-R

Trial within Session

Errors per

T rial

0.0 0.5 1.0 1.5

1

2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5

Saline AR-R 13489 (2 mg/kg)

13489 0704/C12/frame Page 197 Monday, July 17, 2000 5:31 PM

Trang 4

Use of the Radial-Arm Maze to Assess Learning and Memory in Rodents 199

15 Kolb, B., Pittman, K., Sutherland, R.J., and Wishaw, I.Q., Dissociation of the contri-butions of the prefrontal cortex and dorsomedial thalamic nucleus to spatially guided behavior in the rat Behav Brain Res., 6, 365, 1982.

16 Levin, E.D., Psychopharmacological effects in the radial-arm maze Neurosci Biobe-hav Rev., 12, 169, 1988.

17 Levin, E.D and Rose, J.E., Nicotinic and muscarinic interactions and choice accuracy

in the radial-arm maze Brain Res Bull., 27, 125, 1991.

18 Olton, D.S and Werz, M.A., Hippocampal function and behavior: Spatial discrimina-tion and response inhibidiscrimina-tion Physiol Behav., 20, 597, 1978.

19 Olton, D.S., Becker, J.T., and Handelmann, G.E., Hippocampus, space, and memory.

Behav Brain Sci., 2, 313, 1979.

20 Olton, D.S., The use of animal models to evaluate the effects of neurotoxins on cognitive processes Neurobehav Toxicol Teratol., 5, 635, 1983.

21 Olton, D.S., The radial arm maze as a tool in behavioral pharmacology Physiol Behav.,

40, 793, 1987.

22 Olton, D and Markowska, A., Mazes: Their uses in delayed conditional discriminations and place discriminations, In F van Haaren (Ed.), Methods in Behav Pharmacol., Elsevier, New York, pp 195, 1993.

23 Rawlins, J and Deacon, R., Further developments of maze procedures, In A Sahgal (Ed.), Behavioral Neuroscience: A Practical Approach, Volume 1, IRL Press at Oxford University Press, New York, pp 95, 1994.

24 Walsh, T.J and Chrobak, J.J., The use of the radial arm maze in neurotoxicology.

Physiol Behav., 40, 799, 1987.

25 Levin, E., Kim, P., and Meray, R., Chronic nicotine effects on working and reference memory in the 16-arm radial maze: Interactions with D1 agonist and antagonist drugs.

26 Levin, E., Kaplan, S., and Boardman, A., Acute nicotine interactions with nicotinic and muscarinic antagonists: Working and reference memory effects in the 16-arm radial maze Behav Pharmacol., 8, 236, 1997.

27 Levin, E.D., Bettegowda, C., Weaver, T., and Christopher, N.C., Nicotine-dizocilpine interactions and working and reference memory performance of rats in the radial-arm maze Pharmacol Biochem Behav., 61, 335, 1998.

28 Peele, D.B and Baron, S.P., Effects of scopolamine on repeated acquisition on radial-arm maze performance by rats J Exp Anal Behav., 49, 275, 1988.

29 Gray, J.A., Mitchell, S.N., Joseph, M.H., Grigoryan, G.A., Bawe, S., and Hodges, H., Neurochemical mechanisms mediating the behavioral and cognitive effects of nicotine.

Drug Dev Res., 31, 3, 1994.

30 Chambers, R.A., Moore, J., McEvoy, J.P., and Levin, E.D., Cognitive effects of neonatal hippocampal lesions in a rat model of schizophrenia Neuropsychopharmacology, 15,

587, 1996.

31 Levin, E., Bettegowda, C., Blosser, J., and Gordon, J., AR-R 17779, An α 7 nicotinic agonist improves learning and memory in rats Behav Pharmacol., 10, 675–680, 1999.

Trang 5

13

Chapter

An Operant Analysis

of Fronto-Striatal Function in the Rat

Stephen B Dunnett and Peter J Brasted

Contents

I Introduction

II The Neuropathological and Behavioural Profile of HD

A HD Pathology

B HD Symptomatology III Excitotoxic (and Other) Lesions of the Rat Striatum

IV Operant Conditioning and Operant Chambers

A Operant Chambers

B Operant Tasks to Assess Striatal Function in the Rat

V Operant Analysis of Striatal Lesions: Deficits in Motor Responding

A Operant Analysis of the Sensory and Motor Aspects of Sensorimotor Striatal Neglect

B Operant Tasks to Delimit the Specificity of Striatal Neglect

VI Operant Analysis of Striatal Lesions: Deficits in Cognitive Tasks

A Delayed Matching Tasks

B Delayed Alternation Tasks VII Operant Analysis of Striatal Lesions: Deficits in Motivational State VIII Conclusion

References

Trang 6

precisely than can be achieved in the human condition These animal models of human disease can also provide a basis for assessing potential strategies for repair (e.g., neural transplantation, neuroprotective agents, or gene therapy) by giving rise

to measurable behavioural deficits against which the functional efficacy of any particular strategy can be evaluated

Striatal function was initially studied in experimental animals with the use of basal ganglia lesions.2,26 However, there were major difficulties in interpreting the consequences of lesions made by electrolysis, radiofrequency, or direct surgical excision because of the inevitable damage of the immediately adjacent afferent and efferent fibres of the internal capsule connecting the cortex to subcortical structures including thalamus However, this changed dramatically with the introduction of excitotoxic methods of lesions in the mid 1970s, opening the way for the modern era of basal ganglia research The primary excitotoxins are amino acids, such as monosodium glutamate, N-methyl-D-aspartic acid (NMDA), and kainic acid that are glutamate agonists which are toxic when against glutamate receptor bearing neurones, a feature of most neurones of the nervous system When administered directly into the striatum, excitotoxins specifically target and kill neurones within the striatum without damaging the axons of the corticofugal and corticopetal path-ways passing through and adjacent to the striatum Furthermore, injections of exci-totoxins into the striatum produce neurochemical and pathological changes similar

to those seen in HD Initially, kainic acid was used for this purpose.27-29 However, ibotenic acid and particularly quinolinic acid have since become the toxins of choice

on account of numerous neurochemical studies that demonstrate a more selective neuronal loss within the striatum, with the medium spiny GABA neurones being particularly vulnerable and the large aspiny cholinergic, neuropeptide Y, and NADPH-diaphorase positive interneurones being relatively resisitant to these toxins, corresponding to the profile of degeneration observed in HD.30-33

While a single neurotoxic insult is able to mimic the neuropathology of HD, it cannot reproduce the slow and progressive degeneration that is a characteristic feature of the human disease These features can be better mimicked by metabolic toxins, such as 3-nitropropionic acid, which target the striatal neurones selectively, even when administered peripherally.34,35 Nevertheless, excitotoxins continue to be widely used, due to the fact that they typically produce more convenient, consistent, and reproducible lesions than appear achievable with the metabolic toxins.36 The recent development of transgenic animals that mimic the genetic abnormalities of

HD and reproduce specific aspects of the pathogenetic process of the human disease are likely to provide more powerful models for the future,37-39 but the basic behav-ioural impairments observed in transgenic mice are only just beginning to be char-acterised.40,41 Moreover, the precise mechanisms of cell death are still poorly understood and the relationship between the expression of expanded polyglutamine repeat, the formation of intranuclear inclusions, cell death, and behavioural symp-toms remains to be clarified

The destruction of striatal cells with excitotoxins not only produces some of the pathological hallmarks of HD, but also gives rise to behavioural sequelae which reflect many of the symptoms seen clinically The first use of excitotoxins to model the pathology of HD showed unilateral striatal lesions to induce a marked rotation

Trang 7

panel covering the food well, licks at a drinking tube), and to deliver reinforce-ments (e.g., by operating a dispenser to deliver food pellets) Although traditionally run by electromagnetic relays, modern operant chambers are typically under micro-computer control

The classic test apparatus for evaluating operant behaviours is the Skinner box,

an automated test apparatus first devised and developed by B.F Skinner when analysing the behaviour of rats responding to obtain food reward.48 As illustrated in

Figure 13.1, a typical Skinner box provides two levers as operanda, to which the rat may respond The timing of responding into discrete trials can be achieved by making the levers retractable and only available at discrete points within the trial Discrim-inative stimuli are provided by a variety of different lights located above the levers, above and within the food hopper, and in the roof of the test chamber A loudspeaker

in the chamber can also present auditory stimuli either as white noise or discrete tones of controlled frequency and intensity A variety of different reinforcers can be built in This is most typically either a dispenser to deliver food pellets or a liquid dipper to present water into the reward chamber These will only be effective if the animal is suitably motivated by hunger or thirst, achieved by some hours of food or water deprivation, respectively, prior to the training session However, other rein-forcers are also possible, such as presentation of a receptive female rat to a male rat

in studies of hormonal control of sexual motivation.49

An alternative type of operant chamber that has proved highly effective is the nine-hole box (Figure 13.2) The nine-hole box is conceptually similar to the standard Skinner box except that instead of levers, the box is supplied with an arc of nine holes Discriminative stimuli are provided by lights at the rear of each hole, responses (nose pokes) are monitored by infra-red beams at the entrance to each hole, and food is delivered as the reinforcer to a well that is positioned at the rear of the

FIGURE 13.1

Schematic illustration of a two retractable lever operant chamber (Skinner box).

pellet

dispenser

retractable lever

stimulus lights

panel light house light

food tray

retractable flap 0704/C13/frame Page 206 Monday, July 17, 2000 5:33 PM

Trang 8

An Operant Analysis of Fronto-Striatal Function in the Rat 209

produce a lateralised nose-poke response in order to gain food reward, but the rule defining a correct response was different for the two groups The rats of one group were required to make a nose-poke into the same response hole that had been previously lit (the “SAME” condition), whereas the rats of the second group had to respond on the side which had not previously been lit (the “OPPOSITE” condition) For animals trained in the SAME condition, because of the crossover of con-nections between the brain and the periphery, we would predict that unilateral lesions

— whether of the dopamine system or the striatum itself — would produce deficits

on the contralateral side of the body (Figure 13.4) This is equally true whether the deficit is sensory, motor, or associative in nature However, the dissociation between the location of the stimuli and the response holes in the OPPOSITE condition allows differential predictions of the outcome depending on the nature of the underlying deficit Thus, if the animals have a sensory impairment in the detection of the stimuli, then we would expect the rats with unilateral lesions to be impaired making an ipsilateral response to a contralateral stimulus, whereas a response to an ipsilateral stimulus would be unaffected (Figure 13.4) Conversely, if the deficit was primarily

in the selection or initiation of motor response, then we would expect the animal to

FIGURE 13.3

Schematic illustration of the SAME and OPPOSITE versions of the Carli choice reaction time task As well as measures of accuracy and response bias, the speed of initiating (reaction time) and executing (movement time) of correct responses to the two sides are also recorded The test is based on Carli et

al 1985 62

Move (MT) Correct

Correct OPPOSITE

SAME Error

Error

Hold

Detect

Withdraw (RT)

Trang 9

nature of this neglect has been quantified using both unilateral lesions of the

nigro-striatal dopamine neurones and excitotoxic lesions of intrinsic nigro-striatal neurones

Animals were trained to perform two discriminations, independently, on

alter-nate days As in the Carli et al.62 study, the task comprised a central hole and two

response holes However, unlike the Carli et al.62 study, both response holes were

on the same side So, on one day, animals were required to respond to the holes on

the left, and on the next day to the two holes on the right (see Figure 13.6).67 All

responses required rats to detect a stimulus light in one of the two response holes,

and to make a nose poke response in the same hole Once trained, animals received

unilateral striatal lesions with central injections of quinolinic acid

When testing resumed a week later, the lesion rats showed a severe impairment

responding on the contralateral side This impairment took the form of a marked

bias toward the near hole, i.e., the hole closer to the centre, when the holes were on

the side contralateral to the lesions This response bias was so severe that lesion

animals were rarely able to produce responses to the far hole on the contralateral

side In stark contrast to this impairment, lesion rats were able to respond efficiently

and correctly when the holes were on the side ipsilateral to the lesion It was this

distinction in responding on the ipsilateral side and contralateral side that allowed

the specific nature of the hypothesised response space to be revealed.67

FIGURE 13.6

Schematic illustration of the Brasted lateralised choice reaction time task in the nine-hole box The test

is based on Brasted et al 1997 67

Hold

Detect

Withdraw (RT)

Respond (MT) Error !!

Trang 10

An Operant Analysis of Fronto-Striatal Function in the Rat 213

The design of this operant task allowed a specific comparison to be made

between two specific hypotheses concerning the nature of a striatally mediated

response space If responses were coded relative to an external referent within the

animals’ environment (allocentric coding) then animals would be expected to always

neglect the relatively contralateral hole, regardless of in which side of space the

holes were presented (Allocentric coding is often seen in perceptual neglect, when

patients with cortical lesions neglect the contralateral side of an object, regardless

of where the object is located in space.68,69) In this task, an allocentric-based deficit

would manifest itself as a bias toward the far hole when the task is performed to

the ipsilateral side, and a bias toward the near hole when the task is performed to

the contralateral side Alternatively, if responses were coded with respect to the

subject’s body (egocentric coding), then one would predict responding to be

dis-rupted only on the contralateral side

The data clearly show that striatal neglect is not seen uniformly in all parts of

space, but is restricted to the contralateral side and thus consistent with the latter,

egocentric, hypothesis When responding to the ipsilateral holes, animals showed

no evidence of biasing their responding toward the far (i.e., relatively contralateral)

hole (Figure 13.7) In contrast, animals were markedly impaired when performing

on the contralateral side and were completely unable to select responses to the far

(i.e., relatively contralateral) hole (Figure 13.7) A similar impairment was seen in

studies which examined unilateral striatal dopamine lesions, using a between-subject

design.66

FIGURE 13.7

Unilateral striatal lesions produce a marked postoperative ipsilateral response bias which is more marked

for discriminations on the contralateral than on the ipsilateral side (Data from Brasted et al 1997 67 )

Pre-lesion

Post-lesion

Control rats

Unilateral striatal lesions

40 50 60 70 80

100

90

Định dạng
Số trang	14
Dung lượng	286,14 KB