Haloperidol blocks the acquisition but not the retrieval of a conditioned sensitization to apomorphine

631

Haloperidol blocks the acquisition but not the retrieval of a conditioned sensitization to apomorphine

M. J. Acerbo, A. M. Godoy and J. D. Delius

The dopamine agonist apomorphine (apo) elicits stereotyped pecking bouts in pigeons, a response which increases with successive apo injections. The present study sought, first, to confirm the hypothesis that this sensitization arises through a Pavlovian conditioning driven by both external and internal cues; and, secondly, to advance the hypothesis that during this learning the dopaminergic activation only initiates a process that probably ends in glutamatergic synapse modifications. The conditioned nature of the sensitization to apo was examined in two separate experiments that compared context contingent and context uncontingent apo treatments. The role of dopaminergic mechanisms in the acquisition, maintenance and retrieval of sensitization- conditioned pecking was examined by administering the dopamine antagonist haloperidol (hal) either before, during or after ape sensitization treatments. A contingency

betwe~n

context and ape was found to be essential for the acquisition and retrieval of apo-sensitized pecking. A pretreatment with hal did not curtail a subsequent sensitization to apo. When hal was co-administrated with apo it suppressed the initial pecking response to apo and blocked the acquisition of sensitized responding. A pecking response normally observed when apo-sensitized pigeons are challenged with saline (sal) in the same cage in which they were sensitized, was also absenl When hal was co-administered with apo after the sensitization was

Introduction

When administered repeatedly, psychostimulant drugs, such as amphetamine and cocaine, are known

to

yield a sensitization effect. This effect involves a progressive increase of the behavioural response elicited. by a given drug dose. Because it seems likely that the sensitization relates

to

the addiction that psychostimulant substances can cause, the phenomenon has been much researched.

Nevertheless, the details of the processes that bring about sensitization are still much debated (Hinson and Poulos, 1981; Kalant, 1989; Anagnostaras and Robinson, 1996; Adams et a/., 2000; Crombag et a/.:

2000). The lack of agreement might stem from the fact that cocaine and amphetamine are two differently acting indirect and unspecific dopamine agonists (Bedingfield et a/., 1996; Laudrup and Wallace, 1999). Some direct and specific agonists of dopamine such as quinpirole and apomorphine (apo), although not known to be addictive!

complete this led at first to an only partial apo response suppression. When treated with hal in the same cage, already sensitized pigeons responded much as if they had been challenged with sal. The sensitization i!lduced byapo was thus blocked by hal co-administered during

acquisition, but during the maintenance or retrieval phase hal did not impair a previously sensitized responding. It is concluded that when pigeons are sensitized to apo, dopaminergic mechanisms are implicated in initiating the neural modifications that underlie the conditioned

sensitization, but that they themselves are not importantly altered. Behavioural Pharmacology 14:631-640

Keywords: haloperidol, apomorphine, sensitization, conditioning, dopamine, pigeon, pecking

Allgemeine Psychologie, Universitat.Konstanz, 78434 Konstanz, Germany.

Sponsorship: This research was supported by grants from the Deutsche Forschungsgemeinschaft, Bonn, to J.D.D. MJ.A. was aided by a Landesgraduiertenfiirderung Baden-Wiirttemberg grant.

Correspondence and requests for reprints to M. J. Acerbo, Psychology Department, Biopsychology Program, ·University of Michigan, 525 E. University, E. Hall, Ann Arbor, MI 48109-1109, USA.

E-mail: mjacerbo@umich.edu

also give rise to behavioural sensitization. However, in rodents these substances elicit a number of different stereotyped reactions and locomotory responses that apparently index somewhat different sensItIzation v,ariants (Moller et a/., 1987; Mattingly and Gotsick, 1989; Willner et a/., 1992; Mattingly et af., 1997; Tirelli and Heidbreder, 1999; Battisti et a/., 1999,2000). In birds 9:Po elicits a prolonged bout of repetitive pecking and almost no other motor activities (Cheng and Long, 1974; Brunelli et a/., 1975; Machlis, 1980; Delius, 1985).

In pigeons this pecking is highly similar to forage pecking, and is in fact facilitated by food deprivation (Siemann and Delius, 1992; Wynne and Delius, 1995). Never- theless, it is not normally directed at grains, as apo has, somewhat paradoxically, a strong hunger suppre- ssing effect (Deviche, 1984), but is instead aimed at small, contrasting, inedible features (Keller and Deli us, 2001).

Konstanzer Online-Publikations-System (KOPS)

URN: http://nbn-resolving.de/urn:nbn:de:bsz:352-206685

Repeated injections of a given dose of apo yield a pecking

respons~

that increases up to a dose-dependent asymp- totic level (Delius, 1985; Godoy, 2000) . Earlier on it was considered that such sensitization might be caused directly by an increase in sensitivity of the dopaminergic mechanisms mediating the locomotor activity, or else that they were indirectly caused by a gradual habituation to (or familiarization with) the procedures and environ- ments involved (Stewart and Vezina, 1991; Stewart and Badiani, 1993). Concerning the sensitization to apo in pigeons, we have found . these explanations inadequate, and, have instead offered a classical Pavlovian condi tioning account. Apo acts as an unconditioned stimulus (US) that elicits an unconditioned peckin" g response (UR)., When the pigeons experience the apo effect repeatedly iIi a particular experimental cage, this context functions as a conditioned stimulus (CSc.agc; but see below!) that promotes the development of a conditioned incremental pecking response (IR). The waxing IR adds to the initial UR, leading to a sensitized response (UR + IR) . This agrees with the finding that the pecking IR is only expressed in the same cage environment 'in which the pigeons previously experienced the apo effects. Control treatments have indicated that this context dependency is unlikely to be due to a non-associative habituation to the relevant cage and procedures (Godoy and Delius, 1999; Keller et al, 2002; see also later in this report).

Furthermore, when pigeons that had been sensitized to apo in an experimental cage were afterwards challerrged with saline (sal) in the same cage, they showed a conditioned pecking response (CR, Lindenblatt and Delius, 1987; Wynne and Delius, 1995; Keller and Delius, 2001; see also below). However, this CR 'amounts to a fraction of the sensitization IR. Smaller CRs than IRs are also often observed in amphetamine and cocaine sensi- tization experiments (cf. Anagnostaras and Robinson, 1996) . In pigeons the difference is explained by the fact that systemically administered apo, besides functioning as an US, also acts as an interoceptive stimulus (Jarbe, 1984; Djamoz and Wagmer, 1992). We assume that this cue acts as a CSapo and the sensitization IR is really driven by a roughly muliplicative CScagc x CSapo com- pound stimulus (Delius et a!., 2003; see also Bouton, 1993). Upon sal challenges, the CSapo component is obviously missing and so the CR obtained is only a minor fraction of the IR (Godoy and Delius, 1999; Keller and Delius, 2001). Upon apo challenges in a cage other than that used for the preceding sensitization (effectively a CScaac absent condition), one might expect to obtain a corre"sponding part-response to the isolated CSapo com - ponent. However, this response would be hard ,to detect because it would only be a small addition to the far Stronger pecking UR uiggered by apo through its US quality. It is fair to mention that similar, though mostly not quite as radical, conditioning accounts have also been proposed for the context dependence of amphetamine

and cocaine sensitizations (Zavala et aI., 2000; Crombag et a!., 2001; Tirelli, 2001; Anagnostaras et a!., 2002).

We hypothesize that a neural mechanism analogous to that suggested by Wickens (1990) to explain sensory- motor learning might be involved in generating the con- text-dependent sensitization to apo. This author proposed that the dopaminergic nigro/tegmen to-striatal projections converge with glutamatergic cortico-'striatal pathways in the ventral striatum and interact synaptically, such that a nearly simultaneous activation of both these pathways strengthens the glutamate-mediated transmission . In sensitization, the dopamine agonists presumably mimic the activation of the former pathway and trigger stereo- typed responses (the US-UR link). The CS cage

^x

.po compound must activate some of the latter glutamatergic pathways. These sensory pathways are assumed to be modulated at earlier relay stages, such as the retina (Djamoz and Wagmer, 1992), by dopaminergic synapses on which apo can also act (affecting the CSapo component in our case, Delius et a!., 2002). The near-synchronous activations at the ventro-striatal level strengthen the relevant glutamatergic synapses and mediate the devel- opment of the CS-CR link. Although by no means undisputed, this kind of neural model of context- dependent sensitization has recently gained some sup- port (Kelley, 1999; Bell et a!., 2000; see also Acerbo et al, 2002). Within this model it ha,s been mostly assumed that dopamine acts solely on D1-type rec, eptors. However, that may not apply to the sensi, tization-related learning that concerns us here. Employing various ' dopamine agonists and , antagonists, it has been found that the stereotyped pecking induced by apo is mediated by activation of both Dr and Dz-type receptors (Osuide and Adejoh, 1973;

Zarrindast et a!., 1992; see also Waddington and Daly, 1993 concerning the oral apo responses of rodents). At the doses that elicit stereotypic responses reliably, apo undoubtedly a~tivates both types of receptor (Baldessar- ini eta!., 1994; see also later).

One of the two aims of the present study was to obtain confirmatory evidence for the Pavlovian condition- ing hypothesis postulated earlier, _ using a control procedure that would exclude non-associative accounts.

The' other aim was to check the adequacy of the above neural model by investigating the effects of haloperidol (hal), a dopamine antagonist mainly binding to Dz-type receptors (Sokoloff et aL, 1995), on the acquisition, maintenance and retrieval of the sensi tization to apo in pigeons. When discussing this latter issue we shall also refer to two other studies, which examined the effects of a D1 -type receptor antagonist (SCH-23390) and of an N-methyl-D-aSpartate (NMDA) glutamate receptor antagonist (MK-801; Acerbo and Delius, 2003;

Acerbo et a!., 2003).

Methods

General Subjects

Drug-naive adult pigeons (Columba livia), bred from local homing stock and weighing between 450 and 5S0 g, were used. A week before the experiments began they were moved from an outside aviary to individual 40 x 40 x 45 em stainless-steel grid cages. These home cages were located in a well-ventilated and brightly lit (12 h on , 12 h off) animal ·room. Animal maintenance and treatments complied with the standards and rules specified by German animal welfare law.

Apparatus

Two different experimental cages were used. The distinct cages used with all pigeon groups were modified standard cages with their inner back- and side-walls lined with white panels speckled with dark green dots (0.8 mm in diameter, about 10 per 100 cm

²).

The aberrant cages served· as auxiliary environments in two cases (see below).

They .were designed to be markedly different from the

distinct cages, were made of aluminium sheet and were cylindrical in shape, 45 cm in diameter and 40 cm deep.

Their insides were covered with a black lining speckled with yellow triangles (sides 10 mm, about 10 triangles per 100 cm

²).

Their fronts were made of black wire netting.

Both types of cage were located in a separate, brightly lit room equipped with a video-camera and -recorder.

Procedure

Experimental treatments were administered on a once- per-day schedule. After having been injected, the pigeons were immediately and individually placed into one or the other type of cage (see below) and videotaped for 20 min, before they were returned to their home cages. The videotapes were later reviewed and the numbers of pecks per session were counted using a tally-counter. The pecks of pigeons involve a quite distinct, easy to recognize motion pattern (Horster et aI., 2002) and permit quite accurate counts that yield inter-observer agreement coefficients of rs

~

0.85. The experimental schedules to which the various groups of pigeons were subjected consiste.d invariably of two successive phases, mostly involving two differen t treatments. The abbreviated Xxx / Yyy-type names of the groups refer to these sequential treatments.

Drugs

Apo obtained from Teclapharm as a ready clinical solution (10 mg/m!) was diluted with saline to a 1 rng/ml solution just prior to injection. Throughout this study we used 0.5 mg/kg doses of apo. In earlier studies we established that doses between 0.2 and 2 mg/kg apo yielded an orderly, increasing set of dose -dependent sensitization curves (Basten-Krefft, 1977; Godoy, 2000). Hal was obtained from Gry Pharma as a ready clinical solution

(5 mg/ml) and diluted with saline to 0.4 mg/ml prior to injection. Doses of 0.2 mg/kg (low) or 0.3 mg/kg hal (moderate) were used. These doses were chosen on the basis of earlier reportS (Clinton et a/., 1987; Zarrindast et a/., 1992, Aloe et a/., 1997; Fernandez-Espejo and Gil, 1997), and our own unpublished experiments, in which it was found that while doses of hal of 0.5 mg/kg or more had an obvious sedating effect, doses of 0.3 mg/kg or less had no significant depressing effect on the forage pecking of pigeons. Nevertheless, the low hal dose was used when we

consid~red

it essential

to

minimize an immediate interference by hal-induced sedation. The moderate hal dose was used whenever we considered it experimentally desirable to enhance the more durable effects of hal.

Since, in an unpublished earlier experiment, sal-diluted vehicles or plain sal control injections had yielded the same results, we employed equivolume doses of the latter. All injections were given intramuscularly (pectoral muscle).

Data analysis

Daily mean scores and standard errors were calculated for each of the groups (see below). Because the data were often non-normally distributed, all statistical analyses were carried out with non -parametric Mann-Whitney U (between-group comparisons) and Wilcoxon T (within- group comparisons) tests.

Sensitization acquisition

This first experiment involved three groups of pigeons:

(1) Unc/Apo pigeons (n = 8) were injected with sal for 6 days and placed each time in the distinct cages. However, for the purpose of a treatment with apo that was not contingent with the distinct cage (uncontingent, Unc), they were also injected with apo either 3.5 h before (n = 4) or 3.5 h afterwards (n = 4) and p laced in the aberrant cages. This procedure ensured that the distinct cages were already familiar to the pigeons when they were subsequently injected with apo for a further 6 days and placed into the distinct cages.

(2) Hal + Apo/Apo pigeons (n = 5) were injected 'with moderate hal and apo for 6 days and placed in the distinct cage, followed by another 6 days involving the injection of apo alone and placement into the same. cage; and (3) Sal/Apo pigeons (n = 6) received sal injections for 6 days. followed by injections of apo for 6 further days.

Sensitization retention

This experiment involved three gro' ups of subjects:

(1) Apo/Sal pigeons (n = 10) were injected with apo

and placed into the distinct cage for 6 days, and then

injected with sal for another 3

(n

= 6) or 4 days

(n= 4)

and placed into the same cage;

(2) Apo/Apo pigeons

(n

= 6) were injected with apo and placed into the distinct cage for 10 days; and

(3) Apo/Apo + Hal pigeons (n = ^{8) were treated}

identically except that for the last 4 days they received low hal additionally to apo.

Sensitization retrieval

The final experiment involved four groups of subjects:

(1) Sal/Sal pigeons (n = 10) were injected with sal and placed in the distinct cages for 9 days;

(2) Une/Sal pigeons

(n

= 10) were treated in the same way except that, for the purpose of a cage-uncontingent apo treatment, for the first 6 days they were also injected with apo either 6 h before

(n

= 5) or 6 h after

(n

= 5) the sal treatment and placed in the aberrant cages; .

(3) Hal + Apo/Sal pigeons

(n

= 6) injected with lo' w hal and apo doses and placed into the distinct cage for 6 days, and then injected with sal and placed in the distinct cages for 3 days; and

(4) Apo/Hal

p~geons

(n = ^{5) were injected} with apo and placed in the distinct cages for 6 days, and then injected with moderate hal for 3 days and placed in the same cages.

Results

Sensitization acquisition

The mean pecking scores of groups Unc/Apo, Hal + Apo/

Apo and Sal/Apo are shown in Fig. 1. Its left half

Fig. 1

4000

3000

'E

c

0 C'I 2000

...

_(J)

oX

<.l Q)

0..

1000

Sal

establishes that the repeated treatment with apo in the aberrant cage experienced by the Unc/Apo pigeons led

to

an increase from an initial response of abou t 980 pecks/

20 min to a near-asymptotic response of abou t 3630 pecks/20 min (mean of days 4, 5 and 6), signifying an IR of abou t 2650 pecks/20 min. At the same time, the pecking shown by the same group under the influence of sal in the distinct cage never exceeded 10 pecks/ZO min . Thus, as found many times before, repeated administra- tions of a constant dose of apo in a given cage induces a sizeable increase of the pecking response, while a repeated control administration of sal in a given cage results in very little pecking, which does not increase (Wynne and Delius, 1995; Godoy and Delius, 1999; Keller et aI., 2002).

When the same Unc/Apo pigeons were later exposed

to

the distinct cage after being apo injected (day 7) they again exhibited an initially low response of some 990 pecks/20 min,

~hich

afterwards increased to a near- asymptotic response of some 3680 pecks/ZO min (mean of days 11, 12 and 13; Fig. 1, right), a renewed IRofsome 2690 pecks. The course of this second sensi tization was statistically indistinguishable from that of the first sensitization. In other words, the sensitization to apo that had developed in the aberrant cage did not transfer

to

the distinct cage. One migh t have expected some

respon~e

transfer since the CS

_apo

component out of the original CS

_cage x apo.

continued to be present, but, as already explained in the Introduction, the corresponding reaction is a rather elusive affair. The response drop

o ~- .... ^{- ... -} .. ^-~- ...

I

1

hal+apo

I

6

Days I 7

I 12

Sensitization acquisition experiment, mean pecking scores

±

SE. Left, days 1-6 of the Unc/Apo group (.) while apo treated in the aberrant cage (apo*), of the same Unc/Apo group and the Sal /Apo group

CA)

while sal treated in the distinct cages, and of the Hal + Apo/Apo (_) group while hal+apo treated in the distinct cages. Right, days 7-13: of all three groups while apo treated in thedistinct cages.

Fig. 2

'E

c 0

...

C'I (/)

-'"

() Q)

0...

4000

3000

2000

1000

o

j 1

I

6

Days

i~~apo I

• '''',

, . '-Ii

.-,-,-.'

j 7

Sal

10

Sensitization maintenance experiment; mean pecking scores

±

SE. Left, days 1-6 of the Apo/Sal group (.), the Apo/Apo group (A) and the Apo/

Hal

+

Apo group (_), while all being apo treated in the distinct cages. Right, days 7-10 of the iXpo /Sal group while treated with sal, the Apo/Apo group while treated with apo, and the Apo/Hal

+

Apo group while treated with hal

+

apo, all of them in the distinct cages.

occurring between the last day in the aberrant cage (day 6) and the first session in the distinct cage (day 7) was, in any case, highly significant (Wilcoxon test, P < 0.001).

For comparison, notice that the Apo/Apo group of the next experiment, which did not undergo a cage shift (it was effectively a Con/Apo group) exhibited no compar- able response drop (Fig. 2, below) . Notice also that the response drop shown' by the Unc/Apo group can not have been due to an initial unfamiliarity with the distinct cage, since its pigeons had at least as much previous experience with this latter cage on day 7 as they had with the aberrant cage on day 6.

As to the effect of halon the acqUIsition of the sensitization to apo, the Hal + Apo/Apo group exhibited

· an only insignificantly higher responding than the Sal/Apo group during the first phase (Fig. 1, left) , but a far lesser response than the Unc/Apo group when later treated exclusively with apo (compare also with the Apo/Sal group, Fig. 2, left). The co-administration of low hal together with apo thus not only suppressed the first day, unsensitized UR pecking to apo, but also prevented the acquisition of any IR. Figure 1, right, shows that during the second phase, when the pigeons were repeatedly injected with apo, both the Hal + Apo/Apo and Sal/Apo groups exhibited a sensitization much as that shown by unpretreated pigeons (Fig. 1, right; compare with group Apo/Sal, Fig. 2, left). Regarding the Sal/Apo group, the

resu l t confirms previous findings (Wynne and Delius, 1995; Godoy and Deli us, 1999) that · repeated after-sal exposures to a given cage have no appreciable effect upon a subsequent sensitization to apo in the same cage (see also Unc/Apo group here). The equivalent result obtained with the Hal + Apo/Apo group indicates that the hal + apo co-treatment had no carry-over effect on subsequent acquisition of sensitization .

Sensitization maintenance

The mean pecking scores of groups Apo/Sal, Apo/Apo and

Apo/Hal + Apo are shown in Fig. 2. All three groups

developed a sizeable sensitization to apo, even when the

near-asymptote responding of the groups was somewhat

variable (Fig. 2, Jeft; the difference between the Apo/Sal

and Apo + Hal/Sal groups was significant, Mann-Whitney,

P < 0 .05). It is obvious though that during the second

phase of the experiment (Fig. 2, right), the Apo/Apo

group retained its near-asymptotic level of responding,

whereas the response of the Apo/Hal + Apo group fell

significantly from day 6 (about 2320 pecks/20 min) to day

7 (1460 pecks/20 min; Wilcoxon test, P < 0.01) . Notice,

though, that the respon se of the latter group on day 7 was

much stronger than that of group Apo/Sal on the same day

(Fig. 3, right ; Mann-Whitney, P < 0.001) . However, the

pecking of the Apo/Hal + Apo group declined further

over the following days (Wilcoxon test, P < 0.01). At the

end of the experiment (day 10) the responding of

the Apo/Apo + Hal group was nevertheless still signifi- cantly stronger than that of the Apo/Sal group on the same day (168 versus 31 pecks/20 min;, Mann-Whitney test, P < 0.05).

Sensitization retrieval

The mean pecking scores of groups Sal/Sal, Unc/Sal, Hal + Apo/Sal and Apo/Hal are shown in Fig. 3. The resu lts of group Apo/Sal from the previous experiment are repeated for easier comparison. Figure 3, left, shows the pecking responses during the first phase of the experi- ment. The Sal/Sal group and Unc/Sal group, when treated with sal in the distinct cage, never yielded more than a mean 10 pecks/20 min. Although not shown in Fig. 3, , left, to avoid crowding, when treated with apo in the aberrant cage the Unc/Sal group yielded a significant sensitization, producing some 850 pecks/20 min on the first day and some 3100 pecks/20 min at near-asymptote, an IR of about 2250 pecks/min (Wilcoxon test, P <' 0.001; compare also with group Unc/Apo, Fig. 1, left, apo *). The response of the Apo/Sal pigeons increased similarly, from some 700 pecks/20 min shown on day 1 to a near-asymptotic responding of about 3050 pecks/20 min, a significant IR of some 2350 pecks (Wilcoxon test, P < 0.001) . The differences between the Apo/Sal and Sal/Sal' groups during this first phase were obviously highly significant (Mann-Whitney tests, P < 0.001) . The same was the case concerning the differences in the responses under sal or

Fig. 3

4000

3000

'E

c:

o £! 2000

~ ()

a.

Q)

1000

o

Sal

under apo shown by the Unc/Sal group during the first phase (Wilcoxon tests, P < 0.001).

But as Fig. 3, right, shows, at an expanded scale, no trace of this augmented pecking exhibited by the Unc/Sal group in the aberrant cage was in evidence when this group, was subsequently challenged with sal in the distinct cage: it showed no CR-like response. This agrees with the earlier finding that the IR acquired by the Unci Apo group in the aberrant cage did not transfer to the distinctive cage (Fig. 1, right). Note, however, that the Apo/Sal group (arguablyequiviient to a Con/Sal group), which had developed an IR in the distinct cage during the first phase, evinced a partial transfer, by showing a pecking CR when sal challenged in the same cage at the beginning of the second phase (day 7). This response was significantly stronger than the responses shown by both the Unc/Sal and Sal/Sal groups on the same day (Mann- Whitney tests, P < 0.01). The responses of the Apo/Sal group decayed significantly in the course of this extinction treatment (Wilcoxon test, P < 0.01). It is obvious that, even initially, this CR was considerably smaller than the IR accumulated by the same Apo/Sal group during the first phase of the experiment (Wilcoxon test, P < 0.001).

The co-administration of a low hal dose additional to apo in the Hal + Apo/Sal group strongly suppressed pecking

I

6

Days I 7

all sal I 9

100

75 'E

c:

o

50

~

25

o

-'" ₍₎

rf

Sensitizp.tion retrieval experiment, mean pecking scores

±

SE. Left, days 1-6 of the Unc/Sal group (T) and the Sal/Sal group

CA.)

while sal treated, of th~ Apo/Sal group (.) and the Apo/Hal group (.) while apo treated, and of the Hal

+

Apo/Sal group (.) while hal

+

apo treated, all of them in the distinct ca~es, Right (note expanded scale), days 7-9 of all groups while treated in the distinct cages with sal, except the Apo/Hal group which was treated With hal.

during the first phase, much as a moderate hal dose had done in the earlier Hal + Apo/Apo group (compare Fig. 3, left and Fig. 1, left). During the second phase, when sal challenged, the Hal + Apo/Sal group continued to show very little pecking (Fig. 3, right) . In fact, its response was significantly less than that correspondingly shown by the Sal/Sal group (Mann-Whitney tests, P < 0.05). Further- more, the initial pecking (day 7) exhibited under sal by the Hal + Apo/Sal group was significantly smaller than that exhibited on the same day by the Apo/Sal group (Mann- Whitney test, P < 0.02).

However, the Apo/Hal group evinced a quite different response pattern. Figure 3, left, shows that, as expected, it developed a normal sensitization to apo during the first phase, not differing significantly from that shown by the Apo/Sal group. But more importantly, Fig. 2, right, shows that during the second phase, when challenged with the moderate hal dose, the Apo/Hal group manifested a pecking CR that did not differ from that concurrently shown by the sal-challenged Apo/Sal group. The day 7 response of the Apo/Hal group was, accordingly, signifi- cantly higher than that of the Sal/Sal group (Mann- Whitney tests, P < 0.01). Much as the Apo/Sal group's responses had done (see above), the Apo/Hal group's second phase responses exhibi· ted a progressive extinction · decrement (Wilcoxon test, P < 0.05).

Discussion

Conditioned nature of apo sensitization

In the Introduction we postulated that the sensitization to apo in pigeons does not arise from a non-associative sensitization process, but rather, that it is due to an associative Pavlovian conditioning process. The Unc/Apo group (Fig. 1) showed that the increment of the pecking response (lR) in pigeons brought about by the repeated apo administration is an effect that was dependent on the constancy of the cage context in which the drug treatment took place. A context switch, if definite enough to prevent any transfer due to simple stimulus generalization, leads to an immediate loss of the IR, even when the second . cage is equally familiar to the pigeons.

This loss is thus not due to an inhibitory novelty effect, nor is it to any degree affected" by a latent inhibition effect. This circumstance indicates that the sensitization to apo must be based on an associative conditioning process (cf. Godoy and Delius, 1999; Keller and Delius, 2001; Keller et aI., 2001). The Apo/Apo group (Fig. 2) confirms that if no cage change intervenes, the sensitized responding continues into the second phase at a near- asymptotic level.

The Apo/Sal group (Figs 2 and 3) establishes that pigeons sensitized to apo in the distinct cage showed a CR when sal challenged in the same cage. The Unc/Sal group (Fig. 2), successfully sensitized to apo in the aberrant

cage, showed no such CR in the distinct cage, where they behaved just as the exclusively sal-treated Sal/Sal group.

This response pattern occurred despite the fact that, following their sal pretreatment, the Unc/Sal group was just as familiar with the distinct cage as the Apo/Sal group when the sal challenges began. The result is again in agreement with the hypothesis that, in pigeons, the sensitization to apo is the result of a Pavlovian conditioning to the cage context in which the apo treatment takes place (cf. Godoy and Delius, 1999;

Keller and Delius, 2002). However, as already pointed out in the Introduction, the CR exhibited . by the Apo/Sal group is only a fraction of the IR that the same group accumulated during the acquisition phase. The explana- tion for 'this difference in magnitude is that the IR is controlled by a CScas"

^x

apo compound, while the CR , is controlled by the CScage component alone (Delius e! aI., 2003; see also Bouton, 1993 for the incidence of analogous CS combining in more conventional condition- ing preparations). The above compounding is enabled by the fact that the administration of apo, besides acting as an UR-triggering US, also gives rise to an interoceptive CSapo (J arbe; 1984; Djamoz and Wagmer, 1992). By the way, the slight response increase exhibited by the Unci Apo group on day 7 as compared to day 1 might possibly result from the expected but necessarily elusive condi - tioned part-response to the CSapo component alone, mentioned in the Introduction. In another experiment of a somewhat similar kind, we have, in fact, a more sizeable CSapo-driven response

Haloperidol effects on the sensitization to apomorphine Since the dopamine agonist apo triggers pecking by activating both the 01- and Dz-type dopamine receptors (Zarrindast et aI., 1992), there was interest in separating the roles that these might have on the acquisition, retention and retrieval of the sensitization to apo . The present study focused en the blocking effects of the predominantly Dz-type antagonist hal (Sokoloff e! aI., 1995) . A course of hal pretreatment (Hal/Apo group, Fig. 1) did not inhibit the pecking UR and IR to a ^subse-

quent course of apo treatment, as might perhaps be expected, because of the extended half-life that hal has in mammals (about 7 days both in blood plasma and neural tissue : Cohen et aI., 1992; Kornhuber et aI., 1999).

However, we cannot totally exclude the possibility that

an inhibitory after-effect caused by the hal pretreatment

might have been compensated by a hypersensitivity to

apo induced by the same pretreatment. In mammals, a

hal pretreatment does often lead to a regulatory

hypersensitivity to dopamine agonists (e.g. GeurtS et aI.,

1999; Quieroz et aI., 2002) . Howeve r, pigeons have not, as

yet, yielded any satisfactory evidence for a comparable

hal-occasioned hypersensitivity (Deli us et aI., unpub-

lished). Nevertheless, it is true that higher doses

of hal . than those used 'here can produce paradoxical

enhancemen ts of apo-induced pecking that migh t reflect a transi tory hypersensi tivi ty (Acerbo, 2001) .

However, hal co-administered with apo (Hal + Apo/Apo group, Fig. 1) suppressed both the initial pecking UR (day 1) and the development of an IR in response ' to apo.

As soon as the co-administration of hal ceased (day 7), the UR reappeared and a normal IR developed. The UR- suppressing effect was to be expected, inasmuch as the acute pecking caused by apo depends on the joint activat ion of DI-type and Dz-type receptors (Zarrindast et 01., 1992). The absence of an IR is likely to be simply a consequence of this UR-suppressing effect. However, it must be kept in mind that in more conventional Pavlovian paradigms, a suppression of the UR does not always prevent successful conditioning (Parisi and Matthews, 1975; Richardson and Hansen, 1980).

The Apo/Hal group, when challenged with hal, showed a CR response which was statistically equivalent to the CR shown by the Apo/Sal group when challenged with sal (Fig. 3, left). However, both groups evinced a CR that was of much smaller magnitude than the IR they had developed during their previous apo treatments . This is presumably so because, as elaborated before, the CR is merely triggered by the CScage component of the CScage

^x

apo compound that elicited the IRs. The small CR undoubtedly underwent further decline because the CScage-no US treatment that these two groups were experiencing led to response extinction. In any case, the resul t indicates that hal, by itself, can not suppress the CR elicited by CScnge ' The CR must thus have been mediated by receptors other than Dz-type receptors.

Since, in another study (AceJl'bo and Delius, 2003), it was found that the DI-type receptor antagonist SCH-23390 also did not block the CR, it seems possible that neither a D I- nor Dz-type receptor activation is involved in its retrieval (but see below). In unilaterally nigra-Iesioned rats, Carey (1990) showed that only a combined hal plus SCH-23390 administration fully blocked a turning UR induced by apo. But this combination could n'ot block the retrieval of a previously established turning CR. With regard to conditioning obtained with amphetamine and cocaine, the reports are somewhat contradictory, some authors having found that its acquisition was blocked by DI-type receptor antagonists, others by Dz-type receptor antagonists (Beninger and Hahn, 1983; Ujike et aI., 1989;

Drew and Glick, 1990; Hamamura et aI., 1991; Mazurski and Beninger, 1991; Fenu et aI., 2001). The retrieval of a response previously conditioned with amphetamine was not blocked by a Dz-type receptor antagonist (Beninger and Hahn, 1983). A cursory overview of several other such studies suggests that D I-type receptors are probably essential for conditioning with dopamine agoni sts, while the role of D z-type receptors is less certain, but this issue undoubtedly merits further research.

The sharp fall of the pecking response seen in the Apo/

Hal + Apo group imm ediately when co-administered hal (Fig. 2, right, , day 7) presumal?ly reflects the already discussed blockade of the pecking UR that apo injections would normally elicit. Nevertheless, the remaining respon' se was much stronger than the CR just discussed.

We assume that this stronger response reflects the IR triggered by the CScage

^x

apo compound. A gradual decay followed, almost certain ly because the hal co-administra- tion created a CScage

^x

apo-noUS extinction treatment.

This account assumes that the CSopo component 'was mediated by the activation of DI-type receptors which are not much blocked by hal. However, the fact that Acerbo et 01. (2003) (see also Kelley, 1999; Bell et 01., 2000) obtained a closely equivalent result as . the presen t one when using the D)-type antagonist SCH-23390 instead of hal suggests at first sight that neither the D)-type nor the Dz-type antagonist can fully block the action of the CSapo (see also Schechter and Greer, 1987;

Tang and Code, 1989) . But the circumstance that SCH- 23390 was administered intra-striatally in the study of Acerbo et of. obscures the issue . somewhat, as the striatum is unlikely to be the effective origin of the CSapo, the retina being a far more plausible source (Djamoz anCl Wagmer, 1992).

Conclusion

Some of the results concern pigeon groups that received a series of apo treatments in ei ther a target cage-contingent manner or in a target cage-uncontingent manner (the latter received their apo treatments in an alternative cage). When later tested in the 'target cage after e ither an apo or a sal challenge, only the contingently apo-treated pigeons exhibited an incremental response IR or a conditioned response CR. No such responses were shown by the uncontingently apo-treated pigeons. This outcome supports once more our hypothesis that the sensitization of the pecking response to apo in pigeons is due to an associative Pavlovian conditioning process. A non-associa- tive account is precluded, in our view, because the uncontingently treated groups were concurrently sal- pretreated in the target cage, so that when tested, this cage was just as familiar to them as it was to the contingently treated groups .

Since hal blocked the development of both the IR and the CR, dopaminergic processes mediated by D z-rype receptors appear

to

be required for the acquisition of the conditioned association underl ying the sensitization to apo. However, once the IR had been normally acquired, hal did not immediately block its retention. Similarly once the CR was acquired, hal did not block its retrieval.

Both these conditioned responses therefore appear to be

no longer dependent on processes mediated by Dz-type

receptors. These results, taken together with those of

Acerbo and Delius (2003), suggest that while the

activation of both D

1-

and Dz-type receptors is essential for their acquisition, the activation of the apo-induced pecking IR and CR, neither type of receptor is necessary for their retention and retrieval. Indeed , according to resu l ts of a further study (Acerbo et o/., 2003) it seems that the expressions of these responses depends instead on the viability of NMDA glutamate receptors, much as one would expect according to the neural model we favou r. Regardless of this, we suggest that the avian apo sensi tization/condi tioni ng preparation has poten tial for research into the dopaminergic-glutamatergic synaptic interactions thought

to

underlie most sensory-motor learning in vertebrates.

Acknowledgements

We thank Ines Krug for much technical assistance and Jennifer Lee for improvements to the' language.

References

Acerbo MJ (2001). The role of dopamine and glutamate in associative learning by the pigeon. Doctoral thesis, Konstanz University, Konstanz.

Acerbo MJ, Delius JD (2003). Acquisition blockade but no retrieval blockade by SCH·23390 of apomorphine·induced conditioning (submitted).

Acerbo MJ, Gargiulo PA, Krug I, Delius JD (2002). Behavioural consequences of nucleus accumbens dopaminergic stimulation and glutamatergic blocking in pigeons. Behav Brain Res 136:171-177.

Acerbo MJ, Lee J, Delius JD (2003). Sensitization to apomorphine, effects of dizocilpine Nmda blockades (submitted).

Adams JU, Careri JM, Efferen TR, Rotrosen J (2000). Conditioned locomotor stimulant effects of cocaine in rats do not result from interference with habituation. Psychopharmacology 151 :13-18.

Aloe L, lannitelli A, Bersani G, Alleva E, Angelucci F, Maselli P, Manni L (1997).

Haloperidol administration in humans lowers plasma nerve growth factor level: evidence that sedation induces opposite effects to arousal. Neuro' psychobiology 36:65-68.

Anagnostaras SG, Robinson TE (1996). Sensitization to the psychomotor stimulant effects of amphetamine. Modulation by associative learning. Behav Neurosci 110:1397-1414.

Anagnostaras SG, Schallert T, Robinson TE (2002). Memory processes governing amphetamine·induced psychomotor sensitization. Neuropsycho' pharmacology 26:703-715.

Baldessarini RJ, Kula NS, Zong R, Neumeyer JL (1994). Receptor affinities of apomorphine enantiomers in rat brain tissue. Eur J PharmacoI254:199-203.

Basten·Krefft A (1977). Apomorphin·induziertes Verhalten bei Tauben [Apomor·

phine-induced behavior in pigeonsl. Diplomthesis, Ruhr-University, Bochum.

Battisti JJ, Uretsky NJ, Wallace U (1999). Sensitization of apomorphine-induced stereotyped behavior in mice is context dependent. Psychopharmacology 146:42-48.

Battisti JJ, Uretsky NJ, Wallace U (2000). Importance of environmental context in the development of amphetamine- or apomorphine-induced stereotyped behavior after single and multiple doses. Pharmacol Biochem Behav 66:671-677.

Bedingfield JB, Calder LD, Karler R (1996). Comparative behavioral sensitization to stereotypy by direct and indirect dopamine agonists in CF-l mice.

Psychopharmacology 124:219-225.

Bell K, Duffy P, Kalivas PW (2000). Context-specific enhacement of glutamate transmission by cocaine. Neuropsychopharmacology 23:335-344.

Beninger RJ, Hahn BL (1983). Pimozide blocks establishment but not expression of amphetamine-produced environment-specific conditioning. Science

220:1304-1306.

Bouton ME (1993). Context, time, and memory retrieval in the interference paradigms of Pavlovian learning. Psychol BuI/114:80-99.

Brunelli M, Magni F, Moruzzi G, Musumeci 0 (1975). Apomorphine pecking in the pigeon. Arch Ital Bioi 113:303-325.

Carey RJ (1990). Dopamine receptors mediate drug-induced but not pavlovian conditioned contralateral rotation in the unilateral 6-0HDA animal model.

Brain Res 515:292-298.

Cheng HC, Long JP (1974). Dopaminergic nature of apomorphine-induced pecking in pigeons. Eur J Pharmacol 26:313-320.

Clinton JE, Sterner S, Stelmacher Z, Ruiz E (1987). Haloperidol for sedation of disruptive emergency patients. Ann Emerg Med 16:319-322.

Cohen BM, Tsuneizumi T, Baldessarini RJ, Campbell A, Babb SM (1992).

Differences between antipsychotic drugs in persistence of brain levels and behavioral effects. Psychopharmacology 108:338-344.

Crombag HS, Badiani A, Maren S, Robinson TE (2000). The role of contextual versus drug-associated cues in promoting the induction of psychomotor sensitization to intravenous amphetamine. Behav Brain Res 116 :1-22.

Crombag HS, Badiani A, Chan J, Dell'Orco J, Dineen SP, Robinson TE (2001).

The ability of environmental context to facilitate psychomotor sensitization to amphetamine can be dissociated from its effect on acute drug responsive- ness and on conditioned responding. Neuropsychopharmacology 24:

680-690.

Delius JD (1985). The pecking of the pigeon: free for all.· In: Lowe CF, Richelle M, Blackman DE, Bradshaw CM (editors): Behavior Analysis and Contemporary Psychology. New York: Erlbaum; pp. 53-81.

Delius JD, Acerbo MJ, Keller S, Godoy AM, Delius JD (2002). Drogeninduziertes Lernen: Sensitivierung bei Apomorphin [Drug-induced learning: Sensitization to apomorphinl. Neuroforum 4:261-266.

Deviche P (1984). Administration of small doses of apomorphine attenuates feeding in non-deprived pigeons. Physiol Behav 33:581-585.

Djamoz MB, Wagmer HJ (1992). Localization and function of dopamine in the adult vertebrate retina. Neurochem Int 20:139-191.

Drew KL, Glick SO (1990). Role of 01 and 02 receptor stimulation in sensitization to amphetamine-induced circling behavior and in expression and extinction of the Pavlovian conditioned response. Psychopharmacology 101 :465-471.

Fenu S, Bassarao V, Di Chiara G (2001). A role for dopamine 01 receptors in the nucleus accumbens shell. J Neurosci 17:6897-6904.

Fernandez-Espejo E, Gil E (1997). Single restraint stress sensitizes acute chewing movem~nts induced by haloperidol, but not if the 5-HTI

A

agonist 8-0H-DPAT is given prior to stress. Brain Res 755:351-355.

Geurts M, Hermans E, Maloteauy JM (1999). Enhanced striatal dopamine 0(2) receptor-induced [35SlGTPgammaS binding after haloperidol treatment. Eur J PharmacoI382:119-127.

Godoy AM (2000). Dopamine and learning in the pigeon. Doctoral thesis, Konstanz University, Konstanz

Godoy AM, Delius JD (1999). Sensitization to apomorphine in pigeons due to conditioning, subject to generalization but resistant to extinction. Behav Pharmacoll0:367-378.

Hamamura T, Akiyama K, Akimoto K, Kashihara K, Okamura K, Ujike H, Otsuki S (1991). Co-administration of either a selective 01 or 02 dopamine antagonist with methamphetamine prevents methamphetamine-induced behavioral sensitization and neurochemical change, studied by in vivo intracerebral dialysis. Brain Res 546:40-46.

Hinson RE, Poulos CX (1981). Sensitization to the behavioral effects of cocaine:

Modification by pavlovian conditioning. Pharmacol Biochem Behav 15:

559-562.

Horster W, Krumm E, Mohr C, Delius JD (2002). Conditioning the pecking motions of pigeons. Behav Processes 58:27-43.

Jiirbe. TU (1984). Discriminative stimulus properties of cocaine: effects of apomorphine, haloperidol, procaine and other drugs. Neuropharmacology 23:899-907.

Kalant H (1989). Drug tolerance and sensitization: A pharmacological overview. In: Goudie AJ (editor): Psychoactive Drugs. Clifton: Humana;

pp. 547-577.

Kebabian JW, Caine DB (1979). Multiple receptors for dopamine. Nature 277:93-96.

Keller S, Delius JD (2001). Discriminative learning occasioned by the administration of a dopamine agonist. Psychopharmacology 157:

320-323.

Keller S, Delius JD, Acerbo MJ (2002). Evoking conditions, context dependence, effect persistence and conditioned nature of apomorphine sensitization.

Behav PhamacoI13:189-201.

Kelley AE (1999). Neuronal integrative activities of nucleus accumbens subregions in relation to learning and motivation. Psychobiology 27:

198-213.

Kornhuber J, Schultz A, Wiltfang J, Meineke I, Gleiter CH, Zochling R, et al.

(1999). Persistence of haloperidol in human brain tissue. Am J Psychiatry 156:885-890.

Laudrup P, Wallace U (1999). Sensitization elicited by directly and indirectly acting dopaminergic agonists: comparison using neural network analysis.

Psychopharmacology 141 :169-174.

Lindenblatt U, Delius JD (1987). Apomorphine-induced pecking in pigeons classically conditioned to environmental cues. Psychopharmacology

93:223-225.

Machlis LM (1980). Apomorphine: EHects on the timing and sequencing of pecking behavior in chicks. Pharmacol Biochem Behav 13:331-336.

Mattingly BA, Gotsick J E (1989). Conditioning and experiential factors aHecting the development of sensitization to apomorphine. Behav Neurosci 103:1311-1317.

Mattingly BA, Koch C, Osborne FH, Gotsick JE (1997). Stimulus and response factors aHecting the development of behavioral sensitization to apomorphine.

Psychopharmacology 130:109-116.

Mazurski EJ, Beninger RJ (1991). EHects of selective drugs for dopaminergic D1 and D2 receptors on conditioned locomotion in rats. Psychopharmacology 105:107-112.

Moller HG, Nowak K, Kuschinsky K (1987). Conditioning 'of pre- and post- synaptic behavioural responses to the dopamine receptor agonist apomor- phine in rats. Psychopharmacology 91 :50-55.

Osuide G, Adejoh PO (1973). EHects of apomorphine and its interactions with other drugs in the domestic fowl. Eur J Pharmacol 23:56-66.

Parisi T, Matthews TJ (1975). Pavlovian determinants of the autoshaped keypeck responses. Bull Psychon Soc 6:527-529.

Ouieroz CMT, Alcantara FB, YagGe AML, Bibancos T, Frussa-Filho R (2002).

Acute buspirone abolishes the expression of behavioral dopaminergic supersensitivity in mice. Braz J Med Bioi Res 35:237-242.

Richardson WK, Hansen S (1980). Autopecking with gross movement restrained. Psychol Rec 30:39-46.

Saxena PN, Cha~la N, Jori MBl, Iqbal S (1977). Nature of receptors involved in apomorphine responses in pigeons. Psychopharmacology 53:89-95.

Schechter MD, {:Oreer NL (1987). Evidence that the stimulus properties of apomorphirie are mediated by both D1 and D2 receptor activation. Life Sci 40:2461-2471.

Sealfon SC, Olanow CW (2000). Dopamine receptors: from structure to behavior. Trends Neurosci 23:S34-S40:

Siemann M, Delius JD (1992a). Variability of forage peckin9 in pigeons. Ethology 92:29-50.

Siemann M, Delius JD (1992b). Apomorphine-induced behaviour in pigeons (Columba livia). In: Elsner N, Richter NR '(editors): Rythmogenesis in Neurons and Networks. Stuttgart: Thieme; p. 600.

SokoloH P, Diaz J, Levesque D, Pilon C, Dimitriadou V, GriHon N, et al (1995).

Novel dopamine receptor subtypes as targets for antipsychotic drugs. In:

Abooa LG, Lajtha A (editors): A Diversity of Interacting Receptors. New York:

New York Academy of Science; pp. 278-292.

Stewart J, Badiani A (1993). Tolerance and sensitization to the behavioral eHects of drugs. Behav Pharmaco/4:289-312.

Stewart J, Vezina P (1991). Extinction procedures abolish conditioned stimulus control but spare sensitized responding to apomorphine. Behav Pharmacol 2:65-71.

Tang AH, Code RA (1989). Studies on the discriminative stimulus proper- ties of. apomorphine in rhesus monkeys. Psychopharmacology 97:

418-422.

Tirelli E (2001). Short-term contextual sensitization and conditioned hyperkinesia produced by cocaine in suckling rats aged 4-10 days and 14-20 days.

Psychopharmacology 156:42-52.

Tirelli E, Heidbreder C (1 999). Conditioning of and contextual sensitization to apomorphine-induced climbing in mice: evidence against the habituation hypothesis. Behav Neurosci 113:368-376.

Ujike H, Onoue T, Akiyama K, Hamamura T, Otsuki S (1989). EHects of selective D1 and D2 dopamine antagonists on development of methamphetamine- induced I;>ehavioral sensitization. Psychopharmacology 98:89-92.

Waddington JL, Daly SA (1993). Regulation of unconditioned motor behaviour by 01 :D2 interactions. In: Waddington JL (editor): 01:02 Dopamine Receptor Interactions. london: Academic; pp. 51-78.

Wickens J (1990). Striatal dopamine in motor activation and reward-mediated learning: steps towards a unifying model. J Neural Transm 80:9-31.

Willner P, Papp P, Cheeta S, Muscat R (1992). Environmental influences on behavioural sensitization to the dopamine agonist quinpirole. Behav Pharmaco/3:43-50.

Wynne B, Delius JD (1995). Sensitization to apomorphine in pigeons: unaffected by latent inhibition but still due to classical conditioning. Psychopharmcology 119:414-420.

Zarrindast MR, Hajian Heydari A, Hoseini Nia T (1992). Characterization of dopamine receptors involved in apomorphine-induced pecking in pigeons.

Gen Pharmacol 23:427-430.

Zavala AR, Nazarian A, Crawford CA, McDougall SA (2000). Cocaine-induced behavioral sensitization in the 'young rat. Psychopharmacology 151:

291-298.