- Email: [email protected]
The phenomenon of one-trial tolerance to the anxiolytic effect of chlordiazepoxide in the elevated plus-maze test is abolished by previous administration of chlordiazepoxide or buspirone
Life Sciences 73 (2003) 1063 – 1074 www.elsevier.com/locate/lifescie
The phenomenon of one-trial tolerance to the anxiolytic effect of chlordiazepoxide in the elevated plus-maze test is abolished by previous administration of chlordiazepoxide or buspirone M.D. Escarabajal a,*, C. Torres a, C.F. Flaherty b A´rea de Psicobiologı´a, Universidad de Jae´n, Paraje Las Lagunillas s/n. 23071 Jae´n, Spain C.F. Psychology Department, Rutgers University, New Brunswick Campus 152 Frelinghuysen Rd., Piscataway, NJ 08854-8020, USA a
b
Received 2 December 2002; accepted 10 February 2003
Abstract It has been repeatedly reported that the anxiolytic action of benzodiazepines in the elevated plus-maze test is abolished in rats that have received a single prior experience of the test apparatus (one-trial tolerance effect). To analyze whether the one-trial tolerance effect of chlordiazepoxide can be influenced by administration of chlordiazepoxide or buspirone on trial 1, male Wistar rats received an IP injection of vehicle, chlordiazepoxide (8 mg/kg) or buspirone (2.5 mg/kg) 30 min. before testing for 5 min. in the plus-maze (trial 1). Seventy-two hours later, the rats received vehicle or chlordiazepoxide 30 min. before the re-exposure to the plus-maze for 5 min. (trial 2). Groups injected with chlordiazepoxide or buspirone on trial 1 and with chlordiazepoxide on trial 2 showed an anxiolytic effect of chlordiazepoxide on trial 2, as opposed to rats injected with vehicle on trial 1 and with chlordiazepoxide on trial 2. As opposed to previous studies, the present results suggest that the influence of prior experience with the plus-maze on the anxiolytic action of chlordiazepoxide during re-exposure seems to depend critically on the drug state in which trial 1 is experienced. These results are discussed with respect to the hypothesis proposed to explain the phenomenon of one-trial tolerance. D 2003 Elsevier Science Inc. All rights reserved. Keywords: Plus-maze; Prior experience; Drug interactions; Chlordiazepoxide; Buspirone
* Corresponding author. Tel.: +34-953-00-26-64; fax: +34-953-01-21-97. E-mail address: [email protected] (M.D. Escarabajal). 0024-3205/03/$ - see front matter D 2003 Elsevier Science Inc. All rights reserved. doi:10.1016/S0024-3205(03)00416-8
1064
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
Introduction The elevated plus-maze test has been widely used for screening of anxiolytic drugs, and for exploring neurobiological bases of anxiety. This model is based on studies conducted by Montgomery (1955), who demonstrated that rodents show a tendency to avoid the open arms of a Y-shaped maze. The standard apparatus used at present consists of two open arms and two enclosed arms arranged in the shape of a plus (with the same type of arms facing each other) and elevated approximately 50 cm above the ground (Dawson and Tricklebank, 1995). Mice and rats exposed to the elevated plus-maze tend to avoid open arms and show a strong preference to closed arms (Lister, 1987; Pellow and File, 1986). This behavioral pattern can be easily detected by using traditional dependent variables -such as the number of entries into open and closed arms, the time spent in each of them, and ethological behavioral indexes (e.g. Rodgers et al., 1997). All of these anxiety measures show a marked sensitivity to the GABA/BZ receptor anxiolytic and anxiogenic agents (Dalvi and Rodgers, 1996; Rodgers and Johnson, 1998; Schmitt et al., 2000). By contrast, the sensitivity of the model to the action of putative serotonergic anxiolytic compounds (e.g. buspirone) is far from being conclusive and seems to depend on different variables (e.g. Setem et al., 1999; Silva and Brandao, 2000). These results are consistent with clinical data showing that a history of treatment with benzodiacepines may negatively affect buspirone efficacy (Schweizer et al., 1986; Schweizer and Rickels, 1991). These interactions between psychoactive drugs have not been investigated in animals for the moment. An intriguing feature of the plus-maze model is the marked attenuation or even abolition of the anxiolytic effect of benzodiazepines by a single previous experience with the maze (Dawson et al., 1994; File, 1990). This phenomenon, termed ‘‘one-trial tolerance effect’’, has received considerable attention. The effect has been obtained both in mice (Lister, 1987) and rats (File, 1990). Temporal delay between trial 1 and trial 2 can vary from 24 hr. to 2 weeks without altering the basic effect (File et al., 1990), while an increase of duration of both trials from 5 to 10 min. reinstate the action of chlordiazepoxide during re-exposure (Holmes and Rodgers, 1999). Finally, neuroanatomical studies related to this phenomenon suggest the implication of dorsomedial hypothalamus (File et al., 1999), basolateral nucleus of amygdala (File et al., 1998) and dorsal raphe (Gonzalez and File, 1997), but not hippocampus or medial raphe (Gonzalez et al., 1998). In spite of the consistency of the one-trial tolerance effect, no conclusive explanation has been found at present for this phenomenon (Gonzalez and File, 1997; Pereira et al., 1999). In this context, several hypothesis have been suggested, based on processes such as locomotor habituation to the maze (Dawson et al., 1994), development of a learned phobic-like response (File, 1996; File and Zangrossi, 1993), or lack of motivational conflict (Pereira et al., 1999; Rodgers and Shepherd, 1993). A possible reason for these assorted explanations is connected with the fact that the relative importance of some variables for inducing this phenomenon remains unclear. For example, contrary to early findings of stable test-retest profiles in groups injected with the vehicle (File, 1990; File et al., 1990), a growing body of experiments indicate that a single prior undrugged experience of the elevated plus-maze can reduce open arm entries and total entries upon retest (Dawson et al., 1994; Rodgers et al., 1992, 1996). Similarly, even though it has been repeatedly demonstrated that handling history can modulate the anxiolytic action of drugs tested in elevated plus-maze (e.g. Andrews and File, 1993), only File et al. (1992) observed that the absence of handling abolished the lack of anxiolytic effect of chlordiazepoxide on trial 2. Finally, although the one-trial tolerance phenomenon critically depends on experience in the maze, it seems to be independent of drug condition on initial exposure (File, 1990). These
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
1065
results apparently suggest that the influence of maze experience on the anxiolytic action of benzodiazepines is not modulated by the drug state on trial 1. However, since most studies have only administered vehicle or chlordiazepoxide on trial 1, this possibility has not been extensively analyzed, by including, for example, anxiolytic drugs related to serotonergic system. The inclusion of serotonergic drugs would also allow to replicate in animals drug interactions previously described in human beings. The main aim of the present experiment was to characterize further the loss of benzodiazepine anxiolysis in rats previously exposed to the elevated plus-maze. Particularly, we tried to analyze whether the one-trial tolerance effect of chlordiazepoxide could be differentially influenced by previous administration of chlordiazepoxide or buspirone (a putative serotonergic anxiolytic drug) on trial 1. Given that anxiolytic action of buspirone seems to be more potent in animals which have not been handled (Hogg, 1996), together with the fact that the results obtained by File et al. (1992) have not been replicated, we decided to conduct this experiment using non-handled rats.
Methods Subjects Sixty-five male, They were housed Food and drinking temperature (21 F 14:00 pm.
three month old Wistar rats, weighing approximately 280–320 gr, were used. individually and mantained under a 12L:12D cycle with lights on at 8:00 am. water were freely available, and the animals were maintained under controlled 1 jC) and humidity. All testing sessions were performed between 9:00 am and
Drugs Chlordiazepoxide hydrochloride (Sigma–Aldrich, Spain) and buspirone hydrochloride (Bristol– Myers Squibb, Spain) were dissolved in saline, which was used as the vehicle control. All injections were performed intraperitoneally (IP), 30 min before testing. Apparatus The elevated plus-maze consisted of two open arms and two enclosed arms, each measuring 49.5 10 cm, with black Plexiglas floors. The open arms were bounded by 1- cm- high ledges on the sides, but there were no ledges at the ends of the arms. The closed arms had 39.5- cm- high transparent Plexiglas walls. The maze was elevated 50.5 cm above the ground. The rats were carried to the room in a clear plastic neutral box measuring 32 22 11 cm. In order to record the animals’ behavior in the maze, a behavioral recording automated system was used. This system, called ‘cyberplus’ and recently validated in our laboratory (Flaherty et al., 2000; Torres and Escarabajal, 2002), consisted of ten pairs of photoelectric cells that were located in different parts of the elevated plus-maze. A computer connected to this photocell network detected light beam breaks simultaneoulsy and converted this information into scores corresponding to locomotor and anxiety indexes described below.
1066
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
Procedure On the test day, and twenty five minutes after IP injection of vehicle, buspirone (2.5 mg/kg) or chlordiazepoxide (8 mg/kg), respectively, animals were transported to the plus-maze room and left undisturbed in a neutral box for 5 min prior to testing. Immediately after this period, the rats were placed on the central square facing an open arm and allowed to freely explore the maze for 5 min. Between subjects, the maze was thoroughly cleaned with wet and dry cloths. Following this initial exposure to the maze, animals were returned to their home-cages, and re-exposed to the maze 72 hrs. later. Experimental design is presented in Table 1. Rats were randomly assigned to six groups (n = 10 for VV, V-CDP, BUS-CDP, CDP-V and BUS-V, and n = 15 for CDP-CDP). On trial 1, rats received an IP injection of vehicle (groups V-V and V-CDP), buspirone (2.5 mg/kg; groups BUS-V and BUS-CDP) or chlordiazepoxide (8 mg/kg; groups CDP-V and CDP-CDP). Thirty minutes before the re-exposure to the elevated plus-maze (trial 2), rats were IP administered with vehicle (groups V-V, BUS-V and CDP-V) or chlordiazepoxide (8 mg/kg; groups V-CDP, CDP-CDP and BUS-CDP). The dose of buspirone was selected after conducting several pilot studies that indicated a significant action of 2.5 mg/kg, but not of 1.5 mg/kg of buspirone, on the plus-maze. Similarly, the dose of chlordiazepoxide (8 mg/kg) was used because it invariably produces anxiolytic effects in rats exposed to our particular elevated plus-maze conditions. Dependent variables Several measures of anxiety and locomotor activity were obtained by cyberplus: total, open and closed entries; %open and closed entries; total time, time and %time spent in open and closed arms; time, %time and entries into the distal portions of the open arms. In order to avoid a redundancy in the results and statistics, %open time and %open entries were selected as indicators of anxiety, and the total entries were selected as an indicator of locomotory activity. Data analysis Data were subjected to two factor analysis of variance, with factors of trial, group, and trial group. Separate ANOVAs were also conducted for data of trial 1 and trial 2, respectively. Further comparisons
Table 1 Experimental design Group
Treatment Trial 1
Treatment Trial 2
V-V V-CDP CDP-CDP CDP-V BUS-CDP BUS-V
Vehicle Vehicle Chlordiazepoxide Chlordiazepoxide Buspirone Buspirone
Vehicle Chlordiazepoxide Chlordiazepoxide Vehicle chlordiazepoxide vehicle
Chlordiazepoxide (CDP, 8 mg/kg), buspirone (BUS, 2.5 mg/kg), vehicle (saline).
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
1067
Table 2 Mean values ( F S.E.M.) obtained in groups injected with vehicle (V-V, V-CDP), chlordiazepoxide (CDP-V, CDP-CDP) and buspirone (BUS-V, BUS-CDP) on trial 1 %O.E. %O.T. T.E.
V-V
V-CDP
CDP-V
CDP-CDP
BUS-V
BUS-CDP
40.2( F 6.5) 18.22( F 4.5) 14.1( F 1.1)
39.5( F 5.3) 17.4( F 4.5) 14.66( F 1.2)
52.4( F 3.8) 36.8( F 3.9) 10.2( F 1.2)
62.9( F 5.0) 35.9( F 5.2) 11.9( F 1.6)
48.0( F 8.7) 16.1( F 5.8) 5.2( F 0.8)
36.4( F 6.9) 16.7( F 4.1) 9.4( F 1.8)
Dependent variables: % open entries, % open time, total entries (differences among groups injected with chlordiazepoxide and buspirone, and groups injected with vehicle are presented in the results section of the text).
were performed using Tukey’s Honestly Significant Difference Test. For all statistical analyses, alpha was set to 0.05.
Results Chlordiazepoxide seemed to produce an unexpected effect upon locomotor activity on trial 1 in the group CDP-CDP (but not in the group CDP-V), indicated by a reduction in the number of total entries (mean of total entries for group CDP-CDP = 7.6; mean of total entries for group V-V = 14.1). To check the reliability of this effect, n in group CDP-CDP was increased to 15. No differences were obtained between scores of these animals in the number of total entries (F1,13 = 0.054 for trial 1 and F1,13 = 1.449 for trial 2), which verified the spurious nature of the activity changes originally observed. Mean values showed by groups V-V, V-CDP, CDP-V, CDP-CDP, BUS-V and BUS-CDP on trial 1 and trial 2 are presented in Table 2 and Table 3. Overall ANOVA showed a main effect of group for %open entries (F5,58 = 5.148) and %open time (F5,58 = 6.116). In addition, a significant trial group interaction was obtained for %open time (F5,58 = 2.678). In order to analyze the action of chlordiazepoxide and buspirone on trial 1, and because of the lack of significant differences between the groups injected with vehicle, chlordiazepoxide or buspirone on trial 1, respectively, data were merged in three groups depending on the drug treatment received by rats on this trial: vehicle (groups V-V and V-CDP; n = 20), chlordiazepoxide (CDP-V and CDP-CDP; n = 25) and buspirone (BUS-V and BUS-CDP; n = 20). ANOVA showed a significant effect of drug treatment on total entries (F2,61 = 10.370), %open entries (F2,61 = 6.043) and %open time (F2,61 = 8.946). Further analyses indicated that chlordiazepoxide, but not buspirone, significantly increased %open entries and %open time, without affecting the total entries. By contrast, buspirone significantly reduced
Table 3 Mean values ( F S.E.M.) obtained on trial 2 in groups V-V, V-CDP, CDP-V, CDP-CDP, BUS-V and BUS-CDP V-V %O.E. %O.T. T.E.
33.2( F 5.29) 7.40( F 3.19) 8.9( F 1.82)
V-CDP
CDP-V
CDP-CDP
BUS-V
BUS-CDP
43.6( F 3.53) 14.6( F 3.76) 6.3( F 1.39)
51.5( F 3.43) 27.7( F 5.14) 14.3( F 1.80)
58.7( F 5.82)* 36.8( F 6.79)* 11.1( F 1.53)
40.1( F 5.52) 20.2( F 4.46) 12.9( F 0.75)
56.5( F 3.23)* 29.8( F 3.95) 13.1( F 2.01)
Dependent variables: % open entries, % closed entries, total entries, % open time, % closed time, open distal portion time. (*p < 0.05 vs. V-V, trial 2).
1068
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
Fig. 1. PANEL A.- Effect of chlordiazepoxide (8 mg/kg) and buspirone (2.5 mg/kg) on %open arm entries in trial 1.(*p < 0.05 vs. V, trial 1). PANEL B.- Effect of chlordiazepoxide (8 mg/kg) and buspirone (2.5 mg/kg) on %open arm time in trial 1.(*p < 0.05 vs. V, trial 1). PANEL C.- Effect of chlordiazepoxide (8 mg/kg) and buspirone (2.5 mg/kg) on total entries in trial 1.(*p < 0.05 vs. V, trial 1).
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
1069
the number of total entries. These results demonstrated an anxiolytic effect of chlordiazepoxide, and an interfering effect of buspirone on locomotor activity on trial 1 (see Table 2 and Fig. 1). On trial 2, ANOVA again revealed highly significant differences among groups V-V, V-CDP, CDP-V, CDP-CDP, BUS-V and BUS-CDP for total entries (F5,57 = 3.028), %open entries (F5,57 = 4.087) and %open time (F5,57 = 5.121). Post hoc comparison indicated that chlordiazepoxide significantly increased %open entries and %open time in group CDP-CDP. In addition, this drug significantly increased %open entries in group BUS-CDP. These results showed that the one-trial tolerance effect observed in group V-CDP was abolished by IP administration of chlordiazepoxide or buspirone on trial 1 (see Table 3 and Fig. 2). In order to study whether the injection of chlordiazepoxide or buspirone on trial 1 could modulate the pattern of activity of the animals in the absence of the drug on trial 2, groups CDPV and BUS-V were compared to the group V-V, and to the groups CDP-CDP and BUS-CDP, respectively. No differences were observed among these groups, with the exception that a reduction
Fig. 2. PANEL A.- Effect of chlordiazepoxide (8 mg/kg) on %open arm entries in maze-experienced rats (trial 2).(*p < 0.05 vs. V-V, trial 2). PANEL B.- Effect of chlordiazepoxide (8 mg/kg) on %open arm time in maze-experienced rats (trial 2). (*p < 0.05 vs. V-V, trial 2).
1070
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
in closed time and %closed time was observed for the group CDP-V when it was compared to the group V-V on trial 2. Finally, with the aim of analyzing whether changes in locomotor habituation to the maze could explain the differences observed among groups injected with chlordiacepoxide on trial 2, post-hoc tests comparing trial 1 versus trial 2 were conducted for the variables %open entries and total entries for the groups V-V, V-CDP, CDP-CDP and BUS-CDP. The statistical analysis showed a decrease in the number of total entries for groups V-V (F1,8 = 6.069) and V-CDP (F1,9 = 30.488), an increase for group BUSCDP (F1,9 = 13.994), and no effect for group CDP-CDP (F1,14 = 0.115). By contrast, no differences were found between trial 1 and trial 2 for the variable %open entries.
Discussion Does trial 1 drug treatment alter the effectiveness of chlordiazepoxide administered on trial 2 in an elevated plus-maze? In confirmation of earlier reports, present findings indicate that a single undrugged exposure to the elevated plus-maze can significantly reduce the anxiolytic effect of chlordiazepoxide (e.g. File and Zangrossi, 1993; Holmes and Rodgers, 1999). In addition, we found that an injection of chlordiazepoxide (8 mg/kg) or buspirone (2.5 mg/kg) on trial 1 prevented tolerance to chlordiazepoxide on trial 2 that otherwise occurred in a group that had been injected with the vehicle on trial 1. As far as the chlordiazepoxide data are concerned, these results are contrary to previous studies (Dawson et al., 1994; File, 1990), and suggest that the influence of prior experience with the plus-maze seems to depend on the drug state in which trial 1 is experienced. A secondary finding of the present investigation was the markedly different profile of action of chlordiazepoxide and buspirone on trial 1. Thus, while IP administration of chlordiazepoxide produced an anxiolytic effect, buspirone selectively reduced overall activity. These results are in accordance with previous studies, showing the sensitivity of the model to the action GABAergic drugs (e.g. Dalvi and Rodgers, 1996). On the other hand, behavioral outcomes of buspirone on the elevated plus-maze have been much more controversial, showing either anxiolytic (Griebel et al., 1997), anxiogenic (Andrews and File, 1993), null (Pellow and File, 1986) or sedative effects (Collinson and Dawson, 1997). The reduction of the anxiolytic effect of chlordiazepoxide in undrugged rats previously exposed to the elevated plus-maze agrees with other published studies. However, in contrast to the absence of one-trial tolerance effect observed by File et al. (1992) in rats which had not been handled, the present results suggest that the influence of handling experience on the one-trial phenomenon is not consistent. A possible explanation of this discrepancy has been pointed out by Andrews and File (1993), who proposed that uncontrolled differences in handling history prior to the rats’ arrival in the lab could explain the unclear influence of handling on behavioral effects of anxiolytic drugs. Several lines of evidence support the hypothesis that initial maze experience is crucial for the loss of benzodiazepine efficiency when rodents are subsequently re-exposed to the plus-maze, regardless of the drug state on which trial 1 is experienced. For example, File (1990) observed a similar performance on trial 2 between groups injected with vehicle or chlordiazepoxide (7.5 mg/kg) on trial 1. Similar results were obtained by Dawson et al. (1994). In marked contrast to these findings, present data demonstrated that the efficiency of chlordiazepoxide on maze-experienced animals depends largely on the drug state induced on trial 1, given that IP administration of chlordiazepoxide or buspirone on that trial abolished the one-trial tolerance effect observed in a group previously injected with vehicle. Furthermore, a similar
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
1071
tolerance preventing effect has been obtained when morphine (4 mg/kg) was injected on trial 1 and chlordiazepoxide on trial 2 (Leszczuk and Flaherty, 2000). Thus, drugs from three different classes administered on trial 1 offset tolerance to chlordiazepoxide on trial 2. Although these drug interactions have not been reported in the clinical literature, there is evidence that patients with generalized anxiety disorder treated first with benzodiacepines, then with buspirone have reported lack of satisfaction with the effect of buspirone (Schweizer et al., 1986). Therefore, the enhanced effects of chlordiazepoxide following chlordiazepoxide or buspirone treatment obtained in our lab could have interesting involvements for clinical practice and warrant further investigation. A theoretical explanation of the results obtained in the present study cannot be simple, given that two drugs (chlordiazepoxide and buspirone) with a quite different profile of action altered the effect of chlordiazepoxide on trial 2 in the same direction. However, several hypotheses explaining the phenomenon of one trial-tolerance can be analyzed within this context. Firstly, Dawson and colleagues (Dawson et al., 1994; Dawson and Tricklebank, 1995) have argued that the efficiency loss of benzodiazepines may be an artefact related to habituation of locomotor activity between trials, rendering animals more resistant to the action of these anxiolytic drugs during re-exposure. According to this hypothesis, groups V-V and V-CDP showed a decrease in the number of total entries during reexposure, and this decrease was not observed for groups CDP-CDP and BUS-CDP (the number of total entries actually increased in the latter group). Although these data could suggest that the injection of CDP or BUS abolished the one-trial tolerance effect by preventing the development of locomotor habituation on trial 2, the fact that there were not significant differences in %open entries, and in the number of total entries among groups on trial 2, hinders a clear explanation for our results from this perspective. Secondly, Rodgers and Shepherd (1993) suggested that the one-trial tolerance effect might reflect a relative absence of an approach/avoid conflict during re-exposure. Specifically, prior knowledge of the maze in experienced animals would reduce the tendency to explore these natural aversive areas, thereby reducing conflict and eliminating the response to benzodiazepines on trial 2. In accordance to this hypothesis, Pereira et al. (1999) observed that the introduction of a new motivational conflict on trial 2 (by rendering the enclosed arms to the apparatus aversive on trial 1) reinstated the anxiolytic action of 5 mg/kg of chlordiazepoxide on the second trial. Although this hypothesis could explain the action of chlordiazepoxide on trial 1 obtained in the present study, it is unlikely that the previous injection of this drug would have reinstated a motivational conflict on trial 2. Interestingly, although highly speculative, the influence of buspirone on locomotor activity could have interfered with the exploration of the maze on trial 1, minimizing the prior knowledge of the apparatus. Consequently, a motivational conflict would be present on trial 2, thereby reinstating the anxiolytic action of chlordiazepoxide. Thus, an interpretation of the present data in terms of motivational conflict could explain our results only partially. Finally, several lines of research strongly support the view that prior test experience fundamentally alters the nature of emotional response to the plus-maze (Holmes and Rodgers, 1999). Furthermore, File and colleagues (e.g. File and Zangrossi, 1993) have proposed that initial experience in the elevated plusmaze leads to the development of a phobic-like state (fear of heights), with the absence of benzodiazepine anxiolysis related to the insensivity of phobias to these drugs. This hypothesis suggests that maze experience could be associated with some form of learning, the nature of which has not been explicitly described. A comparable proposal has been pointed out by Graeff and colleagues (Graeff et al., 1997, 1998), who repeatedly observed that rats exposed to an elevated T-maze (that consists in an
1072
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
enclosed arm positioned perpendicularly to two opposed open arms) develop an inhibitory avoidance of the open arms, measured by recording the time taken to leave the enclosed arms during three consecutive trials. This inhibitory avoidance (representing learned fear) pharmacologically differs from the unconditioned fear evaluated by recording the time to escape from the open arm (Zangrossi and Graeff, 1997). As opposed to the lack of effect of anxiolytic drugs on the behavior of plus-maze experienced rats, inhibitory avoidance in the elevated T-maze is highly sensitive to GABAergic and serotonergic compounds (Graeff et al., 1997, 1998). Therefore, the relationship between the experience of anxiety in the elevated plus-maze (‘phobia’) and T-maze (‘generalized anxiety/panic’) remains to be determined. Even though File (1990) obtained evidence, in the elevated plus-maze, indicating the absence of influence of drug state on trial 1 on the loss of anxiolytic action of chlordiazepoxide on trial 2, phobiclike response hypothesis could explain present data, at least from a speculative perspective. Thus, maze experience may lead to the acquisition of a phobic-like state through an association between the configurational features of the maze, and the aversive reaction triggered by exposure to the maze. From this point of view, a pharmacological manipulation that would interfere with this association should affect one-trial tolerance phenomenon. According to this prediction, File et al. (1990) observed that an injection of a high dose of chlordiazepoxide (75 mg/kg) on trial 1 prevented the absence of anxiolytic action of this drug (7.5 mg/kg) during re-exposure. This pharmacological procedure probably affected the one-trial tolerance effect due to the sedative or amnesic effect of the injected high dose of chlordiazepoxide. The results obtained in the present experiment could be similarly explained by assuming that previous administration of chlordiazepoxide or buspirone could have influenced the association between the maze cues and the corresponding aversive emotional response. Specifically, the injection of chlordiazepoxide on trial 1 could have reduced the aversive reaction related to the first exposure to the apparatus, virtually ruling out an association between the maze context and a reduced or non-existent anxiety state. Therefore, given that a phobic-like response would not be present on trial 2, chlordiazepoxide could exert an anxiolytic action in group CDP-CDP, as opposed to group V-CDP. Although it is not clear how the action of buspirone during initial exposure could abolish the one-trial tolerance effect, one hypothesis could be that a sedative effect of the drug (suggested by a reduction of locomotor activity) interfered with the learning process underlying the acquisiton of a phobic-like state, enabling the anxiolysis observed during re-exposure after injection of chlordiazepoxide. In any case, the lack of an anxiolytic or anxiogenic action of buspirone on trial 1 suggests that its ability to prevent the one-trial tolerance effect is not related to any change in the aversiveness of the first exposure experience. Whether our data can be explained in terms of learning processes or not is still to be determined. Further studies are currenlty under way in our laboratory in order to analyze the accuracy of the theoretical explanation presented, and to identify the experimental variables that can modulate the influence of drug state during initial experience on the one-trial tolerance phenomenon. These studies will improve our understanding of interactions amongs drugs usually prescribed for treating anxiety disorders.
Acknowledgements This study was funded by the Direccio´n General de Ensen˜anza Superior e Investigacio´n Cientı´fica (PB98-0298), Spain.
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
1073
References Andrews, N., File, S.E., 1993. Handling history of rats modifies behavioral effects of drugs in the elevated plus-maze test of anxiety. European Journal of Pharmacology 235, 109 – 112. Collinson, N., Dawson, G.R., 1997. On the elevated plus-maze the anxiolytic-like effects of the 5-HT(1A) agonist, 8-OHDPAT, but the anxiogenic-like effects of the 5-HT(1A) partial agonist, buspirone, are blocked by the 5-HT1A antagonist WAY 100635. Psychopharmacology 132, 35 – 43. Dalvi, A., Rodgers, R.J., 1996. GABAergic influences on plus-maze behavior in mice. Psychopharmacology 128, 380 – 397. Dawson, G.R., Crawford, S.P., Stanhope, J.K., Iversen, S.D., Tricklebank, M.D., 1994. One-trial tolerance to the effects of chlordiazepoxide on the elevated plus-maze may be due to locomotor habituation, not repeated drug exposure. Psychopharmacology 113, 570 – 572. Dawson, G.R., Tricklebank, M.D., 1995. Use of elevated plus maze in the search for novel anxiolytic agents. Trends in Pharmacological Sciences 16, 33 – 36. File, S.E., 1996. Recent developments in anxiety, stress and depression. Pharmacology, Biochemistry and Behavior 54, 3 – 12. File, S.E., Andrews, N., Wu, P.Y., Zharkovsky, A., Zangrossi Jr., H., 1992. Modification of chlordiazepoxide’s behavioral and neurochemical effects by handling and plus-maze experience. European Journal of Pharmacology 218, 9 – 14. File, S.E., Gonzalez, L.E., Gallant, R., 1998. Role of the basolateral nucleus of amygdala in the formation of a phobia. Neuropsychopharmacology 19, 397 – 405. File, S.E., Gonzalez, L.E., Gallant, R., 1999. Role of the dorsomedial hypothalamus in mediating the response to benzodiazepines on trial 2 in the elevated plus-maze test of anxiety. Neuropsychopharmacology 21, 312 – 320. File, S.E., Mabbutt, P.S., Hitchcott, P.K., 1990. Characterization of the phenomenon of ‘‘one-trial tolerance’’ to the anxiolytic effect of chlordiazepoxide in the elevated plus-maze. Psychopharmacology 112, 98 – 101. File, S.E., Zangrossi Jr., H., 1993. ‘‘One-trial tolerance’’ to the anxiolytic actions of benzodiazepines in the elevated plus-maze, or the development of a phobic state? Psychopharmacology 110, 240 – 244. File, S.E., 1990. One-trial tolerance to the anxiolytic effects of chlordiazepoxide in the plus-maze. Psychopharmacology 100, 281 – 282. Flaherty, C.F., Torres, C., Escarabajal, M.D., Coppotelli, C., 2000. Anxiolytic drug interactions in the elevated-plus-maze. Society for Neuroscience’s 30th Annual Meeting. New Orleans, LA. Society for Neuroscience. Abstracts 2000 (#177.4.). Gonzalez, L.E., Ouagazzal, A.M., File, S.E., 1998. Stimulation of benzodiazepine receptors in the dorsal hippocampus and median raphe reveals differential GABAergic control in two animal models of anxiety. European Journal of Neuroscience 10, 3673 – 3680. Graeff, F.G., Netto, C.F., Zangrossi Jr., H., 1998. The elevated T-maze as an experimental model of anxiety. Neuroscience and Biobehavioral Reviews 23 (2), 237 – 246. Graeff, F.G., Viana, M.B., Mora, P.O., 1997. Dual role of 5-HT in defense and anxiety. Neuroscience and Biobehavioral Reviews 21 (6), 791 – 799. Griebel, G., Rodgers, R.J., Perrault, G., Sanger, D.J., 1997. Risk assessment behavior: evaluation of utility in the study of 5-HTrelated drugs in the rat elevated plus-maze test. Pharmacology, Biochemistry and Behavior 57, 817 – 827. Hogg, S., 1996. A review of the validity and variability of the elevated plus-maze as an animal model of anxiety. Pharmacology, Biochemistry and Behavior 54, 21 – 30. Holmes, A., Rodgers, R.J., 1999. Influence of spatial and temporal manipulations on the anxiolytic efficacy of chlordiazepoxide in mice previously exposed to the elevated plus-maze. Neuroscience and Biobehavioral Reviews 23, 971 – 980. Leszczuk, M.H., Flaherty, C.F., 2000. Interaction between the anxiolytic effect of morphine and chlordiazepoxide in the elevated plus-maze. Society for Neuroscience Abstracts 2000 (#853.2.) . Lister, R.G., 1987. The use of a plus-maze to measure anxiety in the mouse. Psychopharmacology 92, 180 – 185. Montgomery, K.C., 1955. The relationship between fear induced by novel stimulation and exploratory behavior. Journal of Comparative and Physiological Psychology 48, 255 – 260. Pellow, S., File, S.E., 1986. Anxiolytic and anxiogenic drug effects on exploratory activity in an elevated plus-maze: a novel test of anxiety in the rat. Pharmacology, Biochemistry and Behavior 24, 525 – 529. Pereira, J.D.K., Vieira, R.J., Konishi, C.T., Ribeiro, R.A., Frussa-Filho, R., 1999. The phenomenon of ‘‘one-trial tolerance’’ to the anxiolytic effect of chlordiazepoxide in the elevated plus-maze is abolished by the introduction of a motivational conflict situation. Life Sciences 65, 101 – 107.
1074
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
Rodgers, R.J., Cao, B.J., Dalvi, A., Holmes, A., 1997. Animal models of anxiety: an ethological perspective. Brazilian Journal of Medical and Biological Research 30, 289 – 304. Rodgers, R.J., Johnson, N.J., Cole, J.C., Deward, C.V., Kidd, G.R., Kimpson, P.H., 1996. Plus-maze retest profile in mice: importance of initial stages of trial 1 and response to post-trial cholinergic receptor blockade. Pharmacology, Biochemistry and Behavior 54, 41 – 50. Rodgers, R.J., Johnson, NJ., 1998. Behaviorally selective effects of neuroactive steroids on plus-maze anxiety in mice. Pharmacology, Biochemistry and Behavior 59, 221 – 232. Rodgers, R.J., Shepherd, J.K., 1993. Influence of prior maze experience on behavior and response to diazepam in the elevated plus-maze and light/dark tests of anxiety in mice. Psychopharmacology 113, 237 – 242. Rodgers, R.J., Lee, C., Shepherd, J.K., 1992. Effects of diazepam on behavioural and antinociceptive responses to the elevated plus-maze in male mice depend upon treatment regimen and prior maze experience. Psychopharmacology 106, 102 – 110. Schmitt, U., Luddens, H., Hiemke, C., 2000. Behavioral effects of GABA(A) receptor stimulation and GABA-transporter inhibition. Pharmacology, Biochemistry and Behavior 65, 351 – 356. Schweizer, E., Rickels, K., 1991. Serotonergic anxiolytics: a review of their clinical efficacy. In: Rodgers, R.J., Cooper, S.J. (Eds.), 5-HT1A agonists, 5-HT3 antagonists and benzodiacepines: their comparative behavioural pharmacology. John Wiley and Sons Ltd., Chichester, pp. 135 – 147. Schweizer, E., Rickels, K., Lucki, I., 1986. Resistance to the anti-anxiety effect of buspirone in patients with a history of benzodiacepine use. New England Journal of Medicine 314, 719 – 720. Setem, J., Pinheiro, A.P., Motta, V.A., Morato, S., Cruz, A.P.M., 1999. Ethopharmacological analysis of 5-HT ligands on the rat elevated plus-maze. Pharmacology, Biochemistry and Behavior 62, 515 – 521. Silva, R.C., Brandao, M.L., 2000. Acute and chronic effects of gepirone and fluoxetine in rats tested in the elevated plus-maze: an ethological analysis. Pharmacology, Biochemistry and Behavior 65, 209 – 216. Torres, C., Escarabajal, M.D., 2002. Validation of a behavioral recording automated system in the elevated plus-maze test. Life Sciences 70, 1751 – 1762. Zangrossi Jr., H., Graeff, F.G., 1997. Behavioral validation of the elevated T-maze, a new animal model of anxiety. Brain Research Bulletin 44 (1), 1 – 5.
The phenomenon of one-trial tolerance to the anxiolytic effect of chlordiazepoxide in the elevated plus-maze test is abolished by previous administration of chlordiazepoxide or buspirone M.D. Escarabajal a,*, C. Torres a, C.F. Flaherty b A´rea de Psicobiologı´a, Universidad de Jae´n, Paraje Las Lagunillas s/n. 23071 Jae´n, Spain C.F. Psychology Department, Rutgers University, New Brunswick Campus 152 Frelinghuysen Rd., Piscataway, NJ 08854-8020, USA a
b
Received 2 December 2002; accepted 10 February 2003
Abstract It has been repeatedly reported that the anxiolytic action of benzodiazepines in the elevated plus-maze test is abolished in rats that have received a single prior experience of the test apparatus (one-trial tolerance effect). To analyze whether the one-trial tolerance effect of chlordiazepoxide can be influenced by administration of chlordiazepoxide or buspirone on trial 1, male Wistar rats received an IP injection of vehicle, chlordiazepoxide (8 mg/kg) or buspirone (2.5 mg/kg) 30 min. before testing for 5 min. in the plus-maze (trial 1). Seventy-two hours later, the rats received vehicle or chlordiazepoxide 30 min. before the re-exposure to the plus-maze for 5 min. (trial 2). Groups injected with chlordiazepoxide or buspirone on trial 1 and with chlordiazepoxide on trial 2 showed an anxiolytic effect of chlordiazepoxide on trial 2, as opposed to rats injected with vehicle on trial 1 and with chlordiazepoxide on trial 2. As opposed to previous studies, the present results suggest that the influence of prior experience with the plus-maze on the anxiolytic action of chlordiazepoxide during re-exposure seems to depend critically on the drug state in which trial 1 is experienced. These results are discussed with respect to the hypothesis proposed to explain the phenomenon of one-trial tolerance. D 2003 Elsevier Science Inc. All rights reserved. Keywords: Plus-maze; Prior experience; Drug interactions; Chlordiazepoxide; Buspirone
* Corresponding author. Tel.: +34-953-00-26-64; fax: +34-953-01-21-97. E-mail address: [email protected] (M.D. Escarabajal). 0024-3205/03/$ - see front matter D 2003 Elsevier Science Inc. All rights reserved. doi:10.1016/S0024-3205(03)00416-8
1064
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
Introduction The elevated plus-maze test has been widely used for screening of anxiolytic drugs, and for exploring neurobiological bases of anxiety. This model is based on studies conducted by Montgomery (1955), who demonstrated that rodents show a tendency to avoid the open arms of a Y-shaped maze. The standard apparatus used at present consists of two open arms and two enclosed arms arranged in the shape of a plus (with the same type of arms facing each other) and elevated approximately 50 cm above the ground (Dawson and Tricklebank, 1995). Mice and rats exposed to the elevated plus-maze tend to avoid open arms and show a strong preference to closed arms (Lister, 1987; Pellow and File, 1986). This behavioral pattern can be easily detected by using traditional dependent variables -such as the number of entries into open and closed arms, the time spent in each of them, and ethological behavioral indexes (e.g. Rodgers et al., 1997). All of these anxiety measures show a marked sensitivity to the GABA/BZ receptor anxiolytic and anxiogenic agents (Dalvi and Rodgers, 1996; Rodgers and Johnson, 1998; Schmitt et al., 2000). By contrast, the sensitivity of the model to the action of putative serotonergic anxiolytic compounds (e.g. buspirone) is far from being conclusive and seems to depend on different variables (e.g. Setem et al., 1999; Silva and Brandao, 2000). These results are consistent with clinical data showing that a history of treatment with benzodiacepines may negatively affect buspirone efficacy (Schweizer et al., 1986; Schweizer and Rickels, 1991). These interactions between psychoactive drugs have not been investigated in animals for the moment. An intriguing feature of the plus-maze model is the marked attenuation or even abolition of the anxiolytic effect of benzodiazepines by a single previous experience with the maze (Dawson et al., 1994; File, 1990). This phenomenon, termed ‘‘one-trial tolerance effect’’, has received considerable attention. The effect has been obtained both in mice (Lister, 1987) and rats (File, 1990). Temporal delay between trial 1 and trial 2 can vary from 24 hr. to 2 weeks without altering the basic effect (File et al., 1990), while an increase of duration of both trials from 5 to 10 min. reinstate the action of chlordiazepoxide during re-exposure (Holmes and Rodgers, 1999). Finally, neuroanatomical studies related to this phenomenon suggest the implication of dorsomedial hypothalamus (File et al., 1999), basolateral nucleus of amygdala (File et al., 1998) and dorsal raphe (Gonzalez and File, 1997), but not hippocampus or medial raphe (Gonzalez et al., 1998). In spite of the consistency of the one-trial tolerance effect, no conclusive explanation has been found at present for this phenomenon (Gonzalez and File, 1997; Pereira et al., 1999). In this context, several hypothesis have been suggested, based on processes such as locomotor habituation to the maze (Dawson et al., 1994), development of a learned phobic-like response (File, 1996; File and Zangrossi, 1993), or lack of motivational conflict (Pereira et al., 1999; Rodgers and Shepherd, 1993). A possible reason for these assorted explanations is connected with the fact that the relative importance of some variables for inducing this phenomenon remains unclear. For example, contrary to early findings of stable test-retest profiles in groups injected with the vehicle (File, 1990; File et al., 1990), a growing body of experiments indicate that a single prior undrugged experience of the elevated plus-maze can reduce open arm entries and total entries upon retest (Dawson et al., 1994; Rodgers et al., 1992, 1996). Similarly, even though it has been repeatedly demonstrated that handling history can modulate the anxiolytic action of drugs tested in elevated plus-maze (e.g. Andrews and File, 1993), only File et al. (1992) observed that the absence of handling abolished the lack of anxiolytic effect of chlordiazepoxide on trial 2. Finally, although the one-trial tolerance phenomenon critically depends on experience in the maze, it seems to be independent of drug condition on initial exposure (File, 1990). These
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
1065
results apparently suggest that the influence of maze experience on the anxiolytic action of benzodiazepines is not modulated by the drug state on trial 1. However, since most studies have only administered vehicle or chlordiazepoxide on trial 1, this possibility has not been extensively analyzed, by including, for example, anxiolytic drugs related to serotonergic system. The inclusion of serotonergic drugs would also allow to replicate in animals drug interactions previously described in human beings. The main aim of the present experiment was to characterize further the loss of benzodiazepine anxiolysis in rats previously exposed to the elevated plus-maze. Particularly, we tried to analyze whether the one-trial tolerance effect of chlordiazepoxide could be differentially influenced by previous administration of chlordiazepoxide or buspirone (a putative serotonergic anxiolytic drug) on trial 1. Given that anxiolytic action of buspirone seems to be more potent in animals which have not been handled (Hogg, 1996), together with the fact that the results obtained by File et al. (1992) have not been replicated, we decided to conduct this experiment using non-handled rats.
Methods Subjects Sixty-five male, They were housed Food and drinking temperature (21 F 14:00 pm.
three month old Wistar rats, weighing approximately 280–320 gr, were used. individually and mantained under a 12L:12D cycle with lights on at 8:00 am. water were freely available, and the animals were maintained under controlled 1 jC) and humidity. All testing sessions were performed between 9:00 am and
Drugs Chlordiazepoxide hydrochloride (Sigma–Aldrich, Spain) and buspirone hydrochloride (Bristol– Myers Squibb, Spain) were dissolved in saline, which was used as the vehicle control. All injections were performed intraperitoneally (IP), 30 min before testing. Apparatus The elevated plus-maze consisted of two open arms and two enclosed arms, each measuring 49.5 10 cm, with black Plexiglas floors. The open arms were bounded by 1- cm- high ledges on the sides, but there were no ledges at the ends of the arms. The closed arms had 39.5- cm- high transparent Plexiglas walls. The maze was elevated 50.5 cm above the ground. The rats were carried to the room in a clear plastic neutral box measuring 32 22 11 cm. In order to record the animals’ behavior in the maze, a behavioral recording automated system was used. This system, called ‘cyberplus’ and recently validated in our laboratory (Flaherty et al., 2000; Torres and Escarabajal, 2002), consisted of ten pairs of photoelectric cells that were located in different parts of the elevated plus-maze. A computer connected to this photocell network detected light beam breaks simultaneoulsy and converted this information into scores corresponding to locomotor and anxiety indexes described below.
1066
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
Procedure On the test day, and twenty five minutes after IP injection of vehicle, buspirone (2.5 mg/kg) or chlordiazepoxide (8 mg/kg), respectively, animals were transported to the plus-maze room and left undisturbed in a neutral box for 5 min prior to testing. Immediately after this period, the rats were placed on the central square facing an open arm and allowed to freely explore the maze for 5 min. Between subjects, the maze was thoroughly cleaned with wet and dry cloths. Following this initial exposure to the maze, animals were returned to their home-cages, and re-exposed to the maze 72 hrs. later. Experimental design is presented in Table 1. Rats were randomly assigned to six groups (n = 10 for VV, V-CDP, BUS-CDP, CDP-V and BUS-V, and n = 15 for CDP-CDP). On trial 1, rats received an IP injection of vehicle (groups V-V and V-CDP), buspirone (2.5 mg/kg; groups BUS-V and BUS-CDP) or chlordiazepoxide (8 mg/kg; groups CDP-V and CDP-CDP). Thirty minutes before the re-exposure to the elevated plus-maze (trial 2), rats were IP administered with vehicle (groups V-V, BUS-V and CDP-V) or chlordiazepoxide (8 mg/kg; groups V-CDP, CDP-CDP and BUS-CDP). The dose of buspirone was selected after conducting several pilot studies that indicated a significant action of 2.5 mg/kg, but not of 1.5 mg/kg of buspirone, on the plus-maze. Similarly, the dose of chlordiazepoxide (8 mg/kg) was used because it invariably produces anxiolytic effects in rats exposed to our particular elevated plus-maze conditions. Dependent variables Several measures of anxiety and locomotor activity were obtained by cyberplus: total, open and closed entries; %open and closed entries; total time, time and %time spent in open and closed arms; time, %time and entries into the distal portions of the open arms. In order to avoid a redundancy in the results and statistics, %open time and %open entries were selected as indicators of anxiety, and the total entries were selected as an indicator of locomotory activity. Data analysis Data were subjected to two factor analysis of variance, with factors of trial, group, and trial group. Separate ANOVAs were also conducted for data of trial 1 and trial 2, respectively. Further comparisons
Table 1 Experimental design Group
Treatment Trial 1
Treatment Trial 2
V-V V-CDP CDP-CDP CDP-V BUS-CDP BUS-V
Vehicle Vehicle Chlordiazepoxide Chlordiazepoxide Buspirone Buspirone
Vehicle Chlordiazepoxide Chlordiazepoxide Vehicle chlordiazepoxide vehicle
Chlordiazepoxide (CDP, 8 mg/kg), buspirone (BUS, 2.5 mg/kg), vehicle (saline).
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
1067
Table 2 Mean values ( F S.E.M.) obtained in groups injected with vehicle (V-V, V-CDP), chlordiazepoxide (CDP-V, CDP-CDP) and buspirone (BUS-V, BUS-CDP) on trial 1 %O.E. %O.T. T.E.
V-V
V-CDP
CDP-V
CDP-CDP
BUS-V
BUS-CDP
40.2( F 6.5) 18.22( F 4.5) 14.1( F 1.1)
39.5( F 5.3) 17.4( F 4.5) 14.66( F 1.2)
52.4( F 3.8) 36.8( F 3.9) 10.2( F 1.2)
62.9( F 5.0) 35.9( F 5.2) 11.9( F 1.6)
48.0( F 8.7) 16.1( F 5.8) 5.2( F 0.8)
36.4( F 6.9) 16.7( F 4.1) 9.4( F 1.8)
Dependent variables: % open entries, % open time, total entries (differences among groups injected with chlordiazepoxide and buspirone, and groups injected with vehicle are presented in the results section of the text).
were performed using Tukey’s Honestly Significant Difference Test. For all statistical analyses, alpha was set to 0.05.
Results Chlordiazepoxide seemed to produce an unexpected effect upon locomotor activity on trial 1 in the group CDP-CDP (but not in the group CDP-V), indicated by a reduction in the number of total entries (mean of total entries for group CDP-CDP = 7.6; mean of total entries for group V-V = 14.1). To check the reliability of this effect, n in group CDP-CDP was increased to 15. No differences were obtained between scores of these animals in the number of total entries (F1,13 = 0.054 for trial 1 and F1,13 = 1.449 for trial 2), which verified the spurious nature of the activity changes originally observed. Mean values showed by groups V-V, V-CDP, CDP-V, CDP-CDP, BUS-V and BUS-CDP on trial 1 and trial 2 are presented in Table 2 and Table 3. Overall ANOVA showed a main effect of group for %open entries (F5,58 = 5.148) and %open time (F5,58 = 6.116). In addition, a significant trial group interaction was obtained for %open time (F5,58 = 2.678). In order to analyze the action of chlordiazepoxide and buspirone on trial 1, and because of the lack of significant differences between the groups injected with vehicle, chlordiazepoxide or buspirone on trial 1, respectively, data were merged in three groups depending on the drug treatment received by rats on this trial: vehicle (groups V-V and V-CDP; n = 20), chlordiazepoxide (CDP-V and CDP-CDP; n = 25) and buspirone (BUS-V and BUS-CDP; n = 20). ANOVA showed a significant effect of drug treatment on total entries (F2,61 = 10.370), %open entries (F2,61 = 6.043) and %open time (F2,61 = 8.946). Further analyses indicated that chlordiazepoxide, but not buspirone, significantly increased %open entries and %open time, without affecting the total entries. By contrast, buspirone significantly reduced
Table 3 Mean values ( F S.E.M.) obtained on trial 2 in groups V-V, V-CDP, CDP-V, CDP-CDP, BUS-V and BUS-CDP V-V %O.E. %O.T. T.E.
33.2( F 5.29) 7.40( F 3.19) 8.9( F 1.82)
V-CDP
CDP-V
CDP-CDP
BUS-V
BUS-CDP
43.6( F 3.53) 14.6( F 3.76) 6.3( F 1.39)
51.5( F 3.43) 27.7( F 5.14) 14.3( F 1.80)
58.7( F 5.82)* 36.8( F 6.79)* 11.1( F 1.53)
40.1( F 5.52) 20.2( F 4.46) 12.9( F 0.75)
56.5( F 3.23)* 29.8( F 3.95) 13.1( F 2.01)
Dependent variables: % open entries, % closed entries, total entries, % open time, % closed time, open distal portion time. (*p < 0.05 vs. V-V, trial 2).
1068
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
Fig. 1. PANEL A.- Effect of chlordiazepoxide (8 mg/kg) and buspirone (2.5 mg/kg) on %open arm entries in trial 1.(*p < 0.05 vs. V, trial 1). PANEL B.- Effect of chlordiazepoxide (8 mg/kg) and buspirone (2.5 mg/kg) on %open arm time in trial 1.(*p < 0.05 vs. V, trial 1). PANEL C.- Effect of chlordiazepoxide (8 mg/kg) and buspirone (2.5 mg/kg) on total entries in trial 1.(*p < 0.05 vs. V, trial 1).
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
1069
the number of total entries. These results demonstrated an anxiolytic effect of chlordiazepoxide, and an interfering effect of buspirone on locomotor activity on trial 1 (see Table 2 and Fig. 1). On trial 2, ANOVA again revealed highly significant differences among groups V-V, V-CDP, CDP-V, CDP-CDP, BUS-V and BUS-CDP for total entries (F5,57 = 3.028), %open entries (F5,57 = 4.087) and %open time (F5,57 = 5.121). Post hoc comparison indicated that chlordiazepoxide significantly increased %open entries and %open time in group CDP-CDP. In addition, this drug significantly increased %open entries in group BUS-CDP. These results showed that the one-trial tolerance effect observed in group V-CDP was abolished by IP administration of chlordiazepoxide or buspirone on trial 1 (see Table 3 and Fig. 2). In order to study whether the injection of chlordiazepoxide or buspirone on trial 1 could modulate the pattern of activity of the animals in the absence of the drug on trial 2, groups CDPV and BUS-V were compared to the group V-V, and to the groups CDP-CDP and BUS-CDP, respectively. No differences were observed among these groups, with the exception that a reduction
Fig. 2. PANEL A.- Effect of chlordiazepoxide (8 mg/kg) on %open arm entries in maze-experienced rats (trial 2).(*p < 0.05 vs. V-V, trial 2). PANEL B.- Effect of chlordiazepoxide (8 mg/kg) on %open arm time in maze-experienced rats (trial 2). (*p < 0.05 vs. V-V, trial 2).
1070
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
in closed time and %closed time was observed for the group CDP-V when it was compared to the group V-V on trial 2. Finally, with the aim of analyzing whether changes in locomotor habituation to the maze could explain the differences observed among groups injected with chlordiacepoxide on trial 2, post-hoc tests comparing trial 1 versus trial 2 were conducted for the variables %open entries and total entries for the groups V-V, V-CDP, CDP-CDP and BUS-CDP. The statistical analysis showed a decrease in the number of total entries for groups V-V (F1,8 = 6.069) and V-CDP (F1,9 = 30.488), an increase for group BUSCDP (F1,9 = 13.994), and no effect for group CDP-CDP (F1,14 = 0.115). By contrast, no differences were found between trial 1 and trial 2 for the variable %open entries.
Discussion Does trial 1 drug treatment alter the effectiveness of chlordiazepoxide administered on trial 2 in an elevated plus-maze? In confirmation of earlier reports, present findings indicate that a single undrugged exposure to the elevated plus-maze can significantly reduce the anxiolytic effect of chlordiazepoxide (e.g. File and Zangrossi, 1993; Holmes and Rodgers, 1999). In addition, we found that an injection of chlordiazepoxide (8 mg/kg) or buspirone (2.5 mg/kg) on trial 1 prevented tolerance to chlordiazepoxide on trial 2 that otherwise occurred in a group that had been injected with the vehicle on trial 1. As far as the chlordiazepoxide data are concerned, these results are contrary to previous studies (Dawson et al., 1994; File, 1990), and suggest that the influence of prior experience with the plus-maze seems to depend on the drug state in which trial 1 is experienced. A secondary finding of the present investigation was the markedly different profile of action of chlordiazepoxide and buspirone on trial 1. Thus, while IP administration of chlordiazepoxide produced an anxiolytic effect, buspirone selectively reduced overall activity. These results are in accordance with previous studies, showing the sensitivity of the model to the action GABAergic drugs (e.g. Dalvi and Rodgers, 1996). On the other hand, behavioral outcomes of buspirone on the elevated plus-maze have been much more controversial, showing either anxiolytic (Griebel et al., 1997), anxiogenic (Andrews and File, 1993), null (Pellow and File, 1986) or sedative effects (Collinson and Dawson, 1997). The reduction of the anxiolytic effect of chlordiazepoxide in undrugged rats previously exposed to the elevated plus-maze agrees with other published studies. However, in contrast to the absence of one-trial tolerance effect observed by File et al. (1992) in rats which had not been handled, the present results suggest that the influence of handling experience on the one-trial phenomenon is not consistent. A possible explanation of this discrepancy has been pointed out by Andrews and File (1993), who proposed that uncontrolled differences in handling history prior to the rats’ arrival in the lab could explain the unclear influence of handling on behavioral effects of anxiolytic drugs. Several lines of evidence support the hypothesis that initial maze experience is crucial for the loss of benzodiazepine efficiency when rodents are subsequently re-exposed to the plus-maze, regardless of the drug state on which trial 1 is experienced. For example, File (1990) observed a similar performance on trial 2 between groups injected with vehicle or chlordiazepoxide (7.5 mg/kg) on trial 1. Similar results were obtained by Dawson et al. (1994). In marked contrast to these findings, present data demonstrated that the efficiency of chlordiazepoxide on maze-experienced animals depends largely on the drug state induced on trial 1, given that IP administration of chlordiazepoxide or buspirone on that trial abolished the one-trial tolerance effect observed in a group previously injected with vehicle. Furthermore, a similar
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
1071
tolerance preventing effect has been obtained when morphine (4 mg/kg) was injected on trial 1 and chlordiazepoxide on trial 2 (Leszczuk and Flaherty, 2000). Thus, drugs from three different classes administered on trial 1 offset tolerance to chlordiazepoxide on trial 2. Although these drug interactions have not been reported in the clinical literature, there is evidence that patients with generalized anxiety disorder treated first with benzodiacepines, then with buspirone have reported lack of satisfaction with the effect of buspirone (Schweizer et al., 1986). Therefore, the enhanced effects of chlordiazepoxide following chlordiazepoxide or buspirone treatment obtained in our lab could have interesting involvements for clinical practice and warrant further investigation. A theoretical explanation of the results obtained in the present study cannot be simple, given that two drugs (chlordiazepoxide and buspirone) with a quite different profile of action altered the effect of chlordiazepoxide on trial 2 in the same direction. However, several hypotheses explaining the phenomenon of one trial-tolerance can be analyzed within this context. Firstly, Dawson and colleagues (Dawson et al., 1994; Dawson and Tricklebank, 1995) have argued that the efficiency loss of benzodiazepines may be an artefact related to habituation of locomotor activity between trials, rendering animals more resistant to the action of these anxiolytic drugs during re-exposure. According to this hypothesis, groups V-V and V-CDP showed a decrease in the number of total entries during reexposure, and this decrease was not observed for groups CDP-CDP and BUS-CDP (the number of total entries actually increased in the latter group). Although these data could suggest that the injection of CDP or BUS abolished the one-trial tolerance effect by preventing the development of locomotor habituation on trial 2, the fact that there were not significant differences in %open entries, and in the number of total entries among groups on trial 2, hinders a clear explanation for our results from this perspective. Secondly, Rodgers and Shepherd (1993) suggested that the one-trial tolerance effect might reflect a relative absence of an approach/avoid conflict during re-exposure. Specifically, prior knowledge of the maze in experienced animals would reduce the tendency to explore these natural aversive areas, thereby reducing conflict and eliminating the response to benzodiazepines on trial 2. In accordance to this hypothesis, Pereira et al. (1999) observed that the introduction of a new motivational conflict on trial 2 (by rendering the enclosed arms to the apparatus aversive on trial 1) reinstated the anxiolytic action of 5 mg/kg of chlordiazepoxide on the second trial. Although this hypothesis could explain the action of chlordiazepoxide on trial 1 obtained in the present study, it is unlikely that the previous injection of this drug would have reinstated a motivational conflict on trial 2. Interestingly, although highly speculative, the influence of buspirone on locomotor activity could have interfered with the exploration of the maze on trial 1, minimizing the prior knowledge of the apparatus. Consequently, a motivational conflict would be present on trial 2, thereby reinstating the anxiolytic action of chlordiazepoxide. Thus, an interpretation of the present data in terms of motivational conflict could explain our results only partially. Finally, several lines of research strongly support the view that prior test experience fundamentally alters the nature of emotional response to the plus-maze (Holmes and Rodgers, 1999). Furthermore, File and colleagues (e.g. File and Zangrossi, 1993) have proposed that initial experience in the elevated plusmaze leads to the development of a phobic-like state (fear of heights), with the absence of benzodiazepine anxiolysis related to the insensivity of phobias to these drugs. This hypothesis suggests that maze experience could be associated with some form of learning, the nature of which has not been explicitly described. A comparable proposal has been pointed out by Graeff and colleagues (Graeff et al., 1997, 1998), who repeatedly observed that rats exposed to an elevated T-maze (that consists in an
1072
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
enclosed arm positioned perpendicularly to two opposed open arms) develop an inhibitory avoidance of the open arms, measured by recording the time taken to leave the enclosed arms during three consecutive trials. This inhibitory avoidance (representing learned fear) pharmacologically differs from the unconditioned fear evaluated by recording the time to escape from the open arm (Zangrossi and Graeff, 1997). As opposed to the lack of effect of anxiolytic drugs on the behavior of plus-maze experienced rats, inhibitory avoidance in the elevated T-maze is highly sensitive to GABAergic and serotonergic compounds (Graeff et al., 1997, 1998). Therefore, the relationship between the experience of anxiety in the elevated plus-maze (‘phobia’) and T-maze (‘generalized anxiety/panic’) remains to be determined. Even though File (1990) obtained evidence, in the elevated plus-maze, indicating the absence of influence of drug state on trial 1 on the loss of anxiolytic action of chlordiazepoxide on trial 2, phobiclike response hypothesis could explain present data, at least from a speculative perspective. Thus, maze experience may lead to the acquisition of a phobic-like state through an association between the configurational features of the maze, and the aversive reaction triggered by exposure to the maze. From this point of view, a pharmacological manipulation that would interfere with this association should affect one-trial tolerance phenomenon. According to this prediction, File et al. (1990) observed that an injection of a high dose of chlordiazepoxide (75 mg/kg) on trial 1 prevented the absence of anxiolytic action of this drug (7.5 mg/kg) during re-exposure. This pharmacological procedure probably affected the one-trial tolerance effect due to the sedative or amnesic effect of the injected high dose of chlordiazepoxide. The results obtained in the present experiment could be similarly explained by assuming that previous administration of chlordiazepoxide or buspirone could have influenced the association between the maze cues and the corresponding aversive emotional response. Specifically, the injection of chlordiazepoxide on trial 1 could have reduced the aversive reaction related to the first exposure to the apparatus, virtually ruling out an association between the maze context and a reduced or non-existent anxiety state. Therefore, given that a phobic-like response would not be present on trial 2, chlordiazepoxide could exert an anxiolytic action in group CDP-CDP, as opposed to group V-CDP. Although it is not clear how the action of buspirone during initial exposure could abolish the one-trial tolerance effect, one hypothesis could be that a sedative effect of the drug (suggested by a reduction of locomotor activity) interfered with the learning process underlying the acquisiton of a phobic-like state, enabling the anxiolysis observed during re-exposure after injection of chlordiazepoxide. In any case, the lack of an anxiolytic or anxiogenic action of buspirone on trial 1 suggests that its ability to prevent the one-trial tolerance effect is not related to any change in the aversiveness of the first exposure experience. Whether our data can be explained in terms of learning processes or not is still to be determined. Further studies are currenlty under way in our laboratory in order to analyze the accuracy of the theoretical explanation presented, and to identify the experimental variables that can modulate the influence of drug state during initial experience on the one-trial tolerance phenomenon. These studies will improve our understanding of interactions amongs drugs usually prescribed for treating anxiety disorders.
Acknowledgements This study was funded by the Direccio´n General de Ensen˜anza Superior e Investigacio´n Cientı´fica (PB98-0298), Spain.
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
1073
References Andrews, N., File, S.E., 1993. Handling history of rats modifies behavioral effects of drugs in the elevated plus-maze test of anxiety. European Journal of Pharmacology 235, 109 – 112. Collinson, N., Dawson, G.R., 1997. On the elevated plus-maze the anxiolytic-like effects of the 5-HT(1A) agonist, 8-OHDPAT, but the anxiogenic-like effects of the 5-HT(1A) partial agonist, buspirone, are blocked by the 5-HT1A antagonist WAY 100635. Psychopharmacology 132, 35 – 43. Dalvi, A., Rodgers, R.J., 1996. GABAergic influences on plus-maze behavior in mice. Psychopharmacology 128, 380 – 397. Dawson, G.R., Crawford, S.P., Stanhope, J.K., Iversen, S.D., Tricklebank, M.D., 1994. One-trial tolerance to the effects of chlordiazepoxide on the elevated plus-maze may be due to locomotor habituation, not repeated drug exposure. Psychopharmacology 113, 570 – 572. Dawson, G.R., Tricklebank, M.D., 1995. Use of elevated plus maze in the search for novel anxiolytic agents. Trends in Pharmacological Sciences 16, 33 – 36. File, S.E., 1996. Recent developments in anxiety, stress and depression. Pharmacology, Biochemistry and Behavior 54, 3 – 12. File, S.E., Andrews, N., Wu, P.Y., Zharkovsky, A., Zangrossi Jr., H., 1992. Modification of chlordiazepoxide’s behavioral and neurochemical effects by handling and plus-maze experience. European Journal of Pharmacology 218, 9 – 14. File, S.E., Gonzalez, L.E., Gallant, R., 1998. Role of the basolateral nucleus of amygdala in the formation of a phobia. Neuropsychopharmacology 19, 397 – 405. File, S.E., Gonzalez, L.E., Gallant, R., 1999. Role of the dorsomedial hypothalamus in mediating the response to benzodiazepines on trial 2 in the elevated plus-maze test of anxiety. Neuropsychopharmacology 21, 312 – 320. File, S.E., Mabbutt, P.S., Hitchcott, P.K., 1990. Characterization of the phenomenon of ‘‘one-trial tolerance’’ to the anxiolytic effect of chlordiazepoxide in the elevated plus-maze. Psychopharmacology 112, 98 – 101. File, S.E., Zangrossi Jr., H., 1993. ‘‘One-trial tolerance’’ to the anxiolytic actions of benzodiazepines in the elevated plus-maze, or the development of a phobic state? Psychopharmacology 110, 240 – 244. File, S.E., 1990. One-trial tolerance to the anxiolytic effects of chlordiazepoxide in the plus-maze. Psychopharmacology 100, 281 – 282. Flaherty, C.F., Torres, C., Escarabajal, M.D., Coppotelli, C., 2000. Anxiolytic drug interactions in the elevated-plus-maze. Society for Neuroscience’s 30th Annual Meeting. New Orleans, LA. Society for Neuroscience. Abstracts 2000 (#177.4.). Gonzalez, L.E., Ouagazzal, A.M., File, S.E., 1998. Stimulation of benzodiazepine receptors in the dorsal hippocampus and median raphe reveals differential GABAergic control in two animal models of anxiety. European Journal of Neuroscience 10, 3673 – 3680. Graeff, F.G., Netto, C.F., Zangrossi Jr., H., 1998. The elevated T-maze as an experimental model of anxiety. Neuroscience and Biobehavioral Reviews 23 (2), 237 – 246. Graeff, F.G., Viana, M.B., Mora, P.O., 1997. Dual role of 5-HT in defense and anxiety. Neuroscience and Biobehavioral Reviews 21 (6), 791 – 799. Griebel, G., Rodgers, R.J., Perrault, G., Sanger, D.J., 1997. Risk assessment behavior: evaluation of utility in the study of 5-HTrelated drugs in the rat elevated plus-maze test. Pharmacology, Biochemistry and Behavior 57, 817 – 827. Hogg, S., 1996. A review of the validity and variability of the elevated plus-maze as an animal model of anxiety. Pharmacology, Biochemistry and Behavior 54, 21 – 30. Holmes, A., Rodgers, R.J., 1999. Influence of spatial and temporal manipulations on the anxiolytic efficacy of chlordiazepoxide in mice previously exposed to the elevated plus-maze. Neuroscience and Biobehavioral Reviews 23, 971 – 980. Leszczuk, M.H., Flaherty, C.F., 2000. Interaction between the anxiolytic effect of morphine and chlordiazepoxide in the elevated plus-maze. Society for Neuroscience Abstracts 2000 (#853.2.) . Lister, R.G., 1987. The use of a plus-maze to measure anxiety in the mouse. Psychopharmacology 92, 180 – 185. Montgomery, K.C., 1955. The relationship between fear induced by novel stimulation and exploratory behavior. Journal of Comparative and Physiological Psychology 48, 255 – 260. Pellow, S., File, S.E., 1986. Anxiolytic and anxiogenic drug effects on exploratory activity in an elevated plus-maze: a novel test of anxiety in the rat. Pharmacology, Biochemistry and Behavior 24, 525 – 529. Pereira, J.D.K., Vieira, R.J., Konishi, C.T., Ribeiro, R.A., Frussa-Filho, R., 1999. The phenomenon of ‘‘one-trial tolerance’’ to the anxiolytic effect of chlordiazepoxide in the elevated plus-maze is abolished by the introduction of a motivational conflict situation. Life Sciences 65, 101 – 107.
1074
M.D. Escarabajal et al. / Life Sciences 73 (2003) 1063–1074
Rodgers, R.J., Cao, B.J., Dalvi, A., Holmes, A., 1997. Animal models of anxiety: an ethological perspective. Brazilian Journal of Medical and Biological Research 30, 289 – 304. Rodgers, R.J., Johnson, N.J., Cole, J.C., Deward, C.V., Kidd, G.R., Kimpson, P.H., 1996. Plus-maze retest profile in mice: importance of initial stages of trial 1 and response to post-trial cholinergic receptor blockade. Pharmacology, Biochemistry and Behavior 54, 41 – 50. Rodgers, R.J., Johnson, NJ., 1998. Behaviorally selective effects of neuroactive steroids on plus-maze anxiety in mice. Pharmacology, Biochemistry and Behavior 59, 221 – 232. Rodgers, R.J., Shepherd, J.K., 1993. Influence of prior maze experience on behavior and response to diazepam in the elevated plus-maze and light/dark tests of anxiety in mice. Psychopharmacology 113, 237 – 242. Rodgers, R.J., Lee, C., Shepherd, J.K., 1992. Effects of diazepam on behavioural and antinociceptive responses to the elevated plus-maze in male mice depend upon treatment regimen and prior maze experience. Psychopharmacology 106, 102 – 110. Schmitt, U., Luddens, H., Hiemke, C., 2000. Behavioral effects of GABA(A) receptor stimulation and GABA-transporter inhibition. Pharmacology, Biochemistry and Behavior 65, 351 – 356. Schweizer, E., Rickels, K., 1991. Serotonergic anxiolytics: a review of their clinical efficacy. In: Rodgers, R.J., Cooper, S.J. (Eds.), 5-HT1A agonists, 5-HT3 antagonists and benzodiacepines: their comparative behavioural pharmacology. John Wiley and Sons Ltd., Chichester, pp. 135 – 147. Schweizer, E., Rickels, K., Lucki, I., 1986. Resistance to the anti-anxiety effect of buspirone in patients with a history of benzodiacepine use. New England Journal of Medicine 314, 719 – 720. Setem, J., Pinheiro, A.P., Motta, V.A., Morato, S., Cruz, A.P.M., 1999. Ethopharmacological analysis of 5-HT ligands on the rat elevated plus-maze. Pharmacology, Biochemistry and Behavior 62, 515 – 521. Silva, R.C., Brandao, M.L., 2000. Acute and chronic effects of gepirone and fluoxetine in rats tested in the elevated plus-maze: an ethological analysis. Pharmacology, Biochemistry and Behavior 65, 209 – 216. Torres, C., Escarabajal, M.D., 2002. Validation of a behavioral recording automated system in the elevated plus-maze test. Life Sciences 70, 1751 – 1762. Zangrossi Jr., H., Graeff, F.G., 1997. Behavioral validation of the elevated T-maze, a new animal model of anxiety. Brain Research Bulletin 44 (1), 1 – 5.