- Email: [email protected]
Behavioral Validation of the Elevated T-Maze, a New Animal Model of Anxiety
Brain Research Bulletin, Vol. 44, No. 1, pp. 1–5, 1997 Copyright q 1997 Elsevier Science Inc. Printed in the USA. All rights reserved 0361-9230/97 $17.00 / .00
PII S0361-9230(96)00381-4
Behavioral Validation of the Elevated T-Maze, a New Animal Model of Anxiety HE´LIO ZANGROSSI, JR.1 AND FREDERICO G. GRAEFF Laborato´rio de Psicofarmacologia, FFCLRP e Nu´cleo de Neurocieˆncias e Comportamento da Universidade de Sao Paulo, USP, 14040-901 Ribeirao Preto, SP, Brazil [Received 17 June 1996; Accepted 14 October 1996] ABSTRACT: The elevated T-maze test of anxiety has been used to separate in the same rat conditioned from unconditioned responses of fear/anxiety. The test apparatus consists of three elevated arms—one enclosed and two open. Inhibitory avoidance—representing learned fear—is measured by recording the time taken to leave the enclosed arm in three consecutive trials. Unconditioned fear is evaluated by recording the time to escape from the open arm. In this study we investigated procedural questions raised by the use of the elevated T-maze. Experiment 1 showed that restraining the animals at the end of the enclosed arm for 30 s did not change the first (baseline) latency to leave this arm, indicating that aversion for the hands of the experimenter is not a key motivation for this response. The results of Experiment 2 indicated that open-arm experience, but not handling stress is the main cause for inhibitory avoidance acquisition, because rats trained in a T-maze with the three arms enclosed did not show the usual increase in withdrawal latency along the three consecutive trials. The same experiment also showed that the latency to leave the open arm did not undergo habituation over five consecutive trials, evidencing an aversive motivation for this response. The importance of open-arm experience for inhibitory avoidance acquisition was further suggested by the results of Experiment 3, as the removal of a shield that prevents perception of openness tended to increase avoidance latency in the elevated T-maze from the first trial. Q 1997 Elsevier Science Inc.
[ 2,3,7,8,15,16,18 ] ) . In this regard, attempts have been made to relate certain types of experimental models to specific anxiety disorders [1,4,9 – 11,21,22 ] . Recently, Graeff and co-workers [5,20] developed an animal model of anxiety to separate, in the same rat, conditioned from unconditioned fear, which have been related to generalized anxiety and panic disorders, respectively. This test, named the elevated T-maze, is derived from the elevated plus-maze, a widely used animal model of anxiety [14]. The latter has been classified as a ‘‘mixed’’ test of anxiety, because it does not separate different types of fear. This may be related to the well documented inconsistency of the effects of 5-HT interacting drugs in the elevated plus-maze [8]. The elevated T-maze apparatus consists of three arms of equal dimension elevated 50 cm from the floor. One of these arms is enclosed by lateral walls and is perpendicular to the two opposed open arms. When the rat is placed at the end of the enclosed arm, he does not see the open arms until he pokes his head beyond the wall of the closed arm. To be on an open arm seems to be an aversive experience, because rats have an innate fear of height and openness [12,14,17,19]. This allows the animal to learn inhibitory avoidance if repeatedly placed inside the enclosed arm to explore the maze. On the other hand, when the rat is placed at the end of one of the open arms he can move towards the closed arm, presumably performing an escape response. While inhibitory avoidance of the open arm is supposed to represent learned fear, the escape response from the open arm would represent innate fear. In a validating study with anxiolytic drugs, both the benzodiazepine agonist diazepam and the 5-HT 1A ligand ipsapirone impaired inhibitory avoidance in a dose-dependent way. Therefore, inhibitory avoidance in the T-maze may be related to pathologic conditions that are responsive to 5-HT 1A agonists or low doses of benzodiazepines, such as anticipatory anxiety and generalized anxiety disorder. In contrast, one-way escape behavior was not affected by the two drugs studied. The latter result may correlate with the ineffectiveness of these drugs on panic disorder [13]. Differential effects of drugs acting on the serotonergic system arising from the dorsal raphe nucleus on inhibitory avoidance and escape behaviors in the elevated T-maze have also been reported, further indicating that different kinds of fear are being generated by this model [6].
KEY WORDS: One-way escape, Inhibitory avoidance, Types of fear/anxiety, Generalized anxiety, Panic.
INTRODUCTION The use of animal models of anxiety has been fundamental in the search for new anxiolytic compounds and for the understanding of the brain mechanisms underlying anxiety. Clinically, it has been recognized that anxiety is not a unitary phenomenon, and the existence of different types of pathologic anxiety, such as panic, phobias, and posttraumatic stress disorders has been postulated. During the last years, a growing body of evidence has been accumulated indicating that anxiety defined operationally in a given animal model may differ from that generated by other models in respect to its nature ( innate or learned ) , its susceptibility to the effects of drugs and environmental manipulations, or to its neural substrate ( e.g., 1
To whom requests for reprints should be addressed.
1 / 2a40 2322 Mp 1 Tuesday Jul 22 08:51 AM EL–BRB (v. 43, no. 6) 2322
2
ZANGROSSI AND GRAEFF
Although the results of these pharmacological studies are promising, there are procedural questions related to the use of the test that deserve further consideration. For example, during acquisition of inhibitory avoidance the rat is placed at the distal end of the enclosed arm and the latency to leave this arm is measured immediately after the animal is released from the experimenter’s hand. Therefore, the initial motivation for the animal to walk along the enclosed arm may be to escape from the hand. To assess this influence, in Experiment 1 we compared the latencies of inhibitory avoidance under two conditions. In the first, the standard procedure [20] was used, while in the second the animals were kept at the end of the enclosed arm for 30 s by a barrier held 20 cm from the arm’s end. Another question relevant for the interpretation of T-maze results is the nature of the aversive stimulus that promotes inhibitory avoidance. Treit and co-workers [17] have reported that in the elevated plus-maze—the T-maze precursor—lack of a protective wall allowing thigmotaxis in the open arms is the critical factor that motivates preference for the enclosed arms. For the same reason rats would learn to avoid the open arms of the Tmaze by remaining in the enclosed arm. However, manipulation by the experimenter may also be involved, because after each trial the animal is removed from the apparatus by the experimenter’s hand. To test this hypothesis, in Experiment 2 we compared the latency to repeatedly leave the enclosed arm in the original elevated T-maze to that in a T-maze built up with the three arms enclosed by walls. If in this enclosed T-maze the latency increases in the same way as in the elevated T-maze, the animal is either avoiding manipulation, or simply decreasing exploration, due to habituation. On the other hand, if the latency in the enclosed T-maze remains stable over trials, the crucial factor leading to the latency increase in the regular T-maze is likely to be open arm experience. Furthermore, it has been observed [20] that the latency to leave the enclosed arm of the elevated T-maze for the first time (baseline latency) is not significantly different from the latency to leave the open arm (escape latency). Thus, it is not clear whether the rat is really escaping from the open arm or just exploring the new environment. To investigate this possibility, in Experiment 2 we also measured the latency for the animals to leave one of the open arms in five consecutive trials. These latencies were compared to those taken in animals exposed to the correspondent arm of the enclosed T-maze. We expected that while the latencies in the enclosed T-maze would increase gradually over trials, reflecting habituation of exploratory activity, the latencies to leave the open arm of the elevated T-maze could either remain unchanged or even decrease, due to persistent aversive motivation. In this regard, it has been reported that relative open arm exploration in the elevated plus-maze either does not increase or significantly decreases over successive exposures to the apparatus [14,17]. As the results of Experiment 2 have indicated that the crucial factor in motivating inhibitory avoidance in the T-maze is the aversive nature of the open arms, in Experiment 3 we removed the shield placed at the border of the intersection of the closed with the open arms of the maze. This shield blocks the perception of openness until the rat pokes his head beyond the walls of the enclosed arm. As a consequence we expected that avoidance latencies would be increased by this procedure. MATERIALS AND METHODS Animals Male Wistar rats weighing 220–250 g were housed in group of five with food and water freely available in a room maintained at 22 { 17C. Lights were on from 0700 to 1900 h.
Apparatus The elevated T-maze (Fig. 1) was made of wood and had three arms of equal dimensions (50 1 12 cm). One arm, the stem of the T, was enclosed by 40-cm high walls and was perpendicular to two opposed open arms. To avoid the rats falling down, the open arms were surrounded by a Plexiglas rim 1 cm high. The whole apparatus was elevated 50 cm above the floor. The enclosed T-maze used in Experiment 2 had three arms (50 1 12 cm) all surrounded by 40-cm high walls. The experiments were performed with an observer inside the room. Procedure Experiment 1. On the third and fourth days after their arrival in the laboratory, animals were gently handled for 5 min. On the fifth day, 22 rats were randomly allocated between the normal (n Å 11) and the barrier (n Å 11) procedure groups. Two minutes before the test, each animal was put in a cage (28 1 18 cm), to which they had been previously habituated. For the normal procedure group, each rat was removed from the cage and placed at the distal end of the enclosed arm facing the intersection of the arms. The time taken by the rat to leave this arm with the four paws was recorded (baseline latency). The same measurement was repeated in two subsequent trials (Avoidance 1 and Avoidance 2) at 30-s intervals. Following avoidance training (30 s), the rat was placed at the end of the right open arm and the time to leave this arm with the four paws was recorded (escape). During the 30-s intertrial intervals the animals were put back in the cage. The animals of the barrier procedure group were placed at the end of the enclosed arm facing a barrier (10 cm large and 40 cm high) introduced 20 cm from the distal end of this arm. After 30s the barrier was removed and the time taken to leave this arm was recorded (baseline latency). The same procedure was repeated in two subsequent trials (Avoidance 1 and Avoidance 2). Immediately after Avoidance 2 the animals were put back at the
FIG. 1. Drawing of the elevated T-maze.
/ 2a40 2322 Mp 2 Tuesday Jul 22 08:51 AM EL–BRB (v. 43, no. 6) 2322
ELEVATED T-MAZE BEHAVIORAL VALIDATION end of the enclosed arm where they stayed restricted by the barrier for 30 s. After this time, they were placed at the end of the open arm for escape latency measurement. Experiment 2. The animals were handled as in Experiment 1. On the test day, 19 rats were allocated between two groups; one was tested in the elevated T-maze (n Å 9), and the other in the enclosed T-maze (n Å 10). For both groups, the latencies to leave the stem of the maze were measured as for the normal procedure group in Experiment 1. After 30 s, the animals were put at the end of the right open arm (elevated T-maze group) or at the end of the right transversal arm of the enclosed T-maze. The latencies to leave these arms were measured in five consecutive trials at intervals of 30-s. Experiment 3. The animals were handled as in Experiment 1. On the test day, 28 rats were allocated between two groups; one was tested in the regular elevated T-maze (n Å 14) an the other in an elevated T-maze without the shield (see Fig. 1) at the border of the intersection of the closed with the open arms (n Å 14). For both groups, avoidance and escape latencies were measured as for the normal procedure group described in Experiment 1. Statistics The latencies to leave the stem of the T-maze (avoidance in the regular T-maze) were analyzed by split-plot analysis of variance, with test conditions (Experiment 1—barrier or no-barrier; Experiment 2—elevated or enclosed T-maze; Experiment 3— shield or no-shield) as the independent factor and trials as the repeated measure. When appropriated, between-group comparisons within each trial were made by the unpaired Student’s ttest. Latencies to leave the transversal arm (escape in the regular T-maze) were analyzed by the unpaired Student’s t-test (Experiment 1 and 3) or by split-plot analysis of variance (Experiment 2). In the latter case, test condition (elevated or enclosed Tmaze) was the independent factor and trials the repeated measure (trials 1–5). Between-group comparisons within each trial were made by the unpaired Student’s t-test. To analyze between trials changes, a repeated measure analysis of variance followed by Duncan multiple-comparison test was also performed in Experiment 2. RESULTS
3
FIG. 2. Effect of restraining with a barrier on inhibitory avoidance and escape latencies in the elevated T-maze. Bars represent the mean and the vertical lines the SEM. In the normal procedure group (NORMAL; n Å 11), latencies (BASELINE, AVOID 1 and AVOID 2) were measured immediately after the animals being released by the experimenter’s hand. In the barrier procedure group (BARRIER; n Å 11) these measurements were performed after the animals being restricted by a barrier for 30 s. ESCAPE was measured 30 s after AVOID 2. *p õ 0.05 compared with the normal procedure group within a same trial.
over trials, the latencies in the enclosed T-maze were not [trial effect: F(2, 34) Å 8.6, p õ 0.001; procedure 1 trial interaction: F(2, 34) Å 4.0, p õ 0.05]. It can also be seen in Fig. 3 that the latencies to leave the open arm of the elevated T-maze were significantly different from those measured in a transversal arm of the enclosed T-maze [procedure effect: F (1, 17) Å 7.6, p õ 0.05]. There was also a trial effect, F (4,68) Å 2.6, p õ 0.05, and a tendency for a procedure by trial interaction, F (4, 68) Å 2.1, p Å 0.09. A repeated measure analysis of variance followed by the Duncan multiple comparison test showed that in the enclosed T-maze group the first trial latency in the transversal arm was significantly shorter (p õ 0.05) than the other trials while in the elevated T-maze there was no significant difference among trials. This result indicates that withdrawal from the open arm does not undergo habituation.
Experiment 1 Figure 2 shows that both normal and barrier procedure groups acquired inhibitory avoidance in the enclosed arm of the elevated T-maze [trial effect: F (2, 40) Å 40.5, p õ 0.0001]. There was a significant difference between the two groups in avoidance performance [procedure effect: F (1, 20) Å 8.3, p õ 0.01) and a significant procedure by trial interaction [procedure 1 trial: F (2, 40) Å 3.8, p õ 0.05], because the barrier facilitated inhibitory avoidance acquisition. Within trials comparison showed that the barrier increased Avoidance 1 and 2 latencies when compared to the normal procedure group. Nevertheless, there was no difference between the two groups in the baseline latency. It can also be seen in Fig. 2 that there was no significant difference (p ú 0.05) between the two groups in the time taken to leave the open arm of the elevated T-maze. Experiment 2 Figure 3 shows that there was a significant difference in the acquisition of inhibitory avoidance in the elevated T-maze as compared to the enclosed T-maze [procedure effect: F (1, 17) Å 7.9, p õ 0.05]. While the latencies for the animals to leave the enclosed arm of the elevated T-maze were significantly increased
Experiment 3 Figure 4 shows that the animals tested in the elevated T-maze either with or without the shield acquired inhibitory avoidance from the open arms [trial effect: F (2, 56) Å 20.2, p õ 0.0001]. Nevertheless, the absence of the shield tended to increase avoidance latencies along the trials. This tendency approached statistical significance [procedure effect: F (1, 28) Å 3.4, p Å 0.08]. However, the procedure by trial interaction was not significant [procedure 1 trial: F(2, 56) Å 0.27, p ú 0.05]. Regarding escape, there was no difference (p ú 0.05) between the two groups in the latency to leave the open arm (Fig. 4). DISCUSSION The results of Experiment 1 show that restraint with a barrier for 30 s at the end of the enclosed arm did not change the first latency to leave this arm of the elevated T-maze. Therefore, in the standard procedure the initial locomotion along the enclosed arm does not seem to constitute an escape response from the experimenter’s hand. However, Avoidance 1 and 2 latencies were longer in the barrier group. The reason for this difference is not obvious. One possibility is that restraint may generate anxiety,
/ 2a40 2322 Mp 3 Tuesday Jul 22 08:51 AM EL–BRB (v. 43, no. 6) 2322
4
ZANGROSSI AND GRAEFF
FIG. 3. Latencies to withdraw from the arms of the elevated and the enclosed T-mazes. On the left of the graph, the time taken (mean / SEM) to leave the stem of the elevated T-maze (n Å 9) or of the enclosed T-maze (n Å 10) were measured in three consecutive trials (TRIAL 1–3) at 30 s interval. Thirty seconds after trial 3, the latency to leave the right transversal arm of the elevated T-maze (open) or of the enclosed T-maze (enclosed) were measured in five consecutive trials (TRIAL 1–5; right of the graph) at 30 s interval. *p õ 0.05 compared with the enclosed T-maze group within a same trial.
facilitating avoidance acquisition. Alternatively, escape from the hand may be speeding up withdrawal from the enclosed arm. A floor effect would have effaced the difference between groups in the first trial. Still another possibility is that repeated exposure to
FIG. 4. Effect of removing the wooden shield held in front of the enclosed arm on inhibitory avoidance and escape latencies in the elevated T-maze. Bars represent the mean and the vertical lines the SEM of groups of 14 rats. Measurements were performed as for the normal procedure group described in the legend of Fig. 2.
a constrained space, when the barrier was used, might accelerate habituation to the enclosed arm. It is, nevertheless, clear that rats belonging to either the barrier or the no-barrier group show significant learning along trials. The present results with the enclosed T-maze further show that latency to leave the stem of the T remained stable along the three trials corresponding to inhibitory avoidance acquisition in the elevated T-maze (Experiment 2). This result may have several meanings: first, it indicates that habituation of exploratory activity does not contribute to the increase in latency observed in the elevated T-maze; second, it tends to rule out handling stress as a cause for inhibitory avoidance acquisition, because in both the enclosed and the elevated T-maze the rats were picked up by the experimenter soon after arm emergence; third, experience of the open arms would remain, by exclusion, the crucial motivation for avoidance acquisition, as originally assumed [5,20]. A different picture emerged in the same experiment (number 2) when the latency to leave the right transversal arm—that is, open in the elevated maze—was measured five times in succession, soon after the three above measures in the stem had been taken. In the elevated T-maze the time taken to leave the open arm remained unchanged, whereas the latency to leave the transversal arm of the enclosed T-maze significantly increased over consecutive trials. Therefore, habituation has taken place in the enclosed arm only. Its rapid onset is probably due to the previous exposure to the similarly enclosed stem of the T. Therefore, the absence of habituation in the open arm has unveiled the aversive
/ 2a40 2322 Mp 4 Tuesday Jul 22 08:51 AM EL–BRB (v. 43, no. 6) 2322
ELEVATED T-MAZE BEHAVIORAL VALIDATION motivation determined by exposure to this situation. Indeed, previously reported results strongly suggest that fear of openness does not easily habituate, probably due to its high survival value [17]. Nevertheless, because in the first escape trial—which is the only one normally used in the standard procedure—the rat leaves with the same speed as in the transversal arm of the enclosed T-maze, the associated influence of the exploratory drive cannot be excluded. The critical role of open arm experience in the elevated Tmaze is also suggested by the results of Experiment 3, showing that removal of the shield that prevents perception of openness when the rat is inside the stem of the T tended to increase avoidance latencies from the first trial (BASELINE in Fig. 4). Overall, the results of the present study indicate that openness is the main factor motivating both inhibitory avoidance and one-way escape in the elevated T-maze. The original assumption that this test generates two types of fear is thus tenable in view of the reported differential sensitivity of these behaviors to several pharmacological treatments [ 5,6,20 ] . Nevertheless, it should be kept in mind that other factors may as well explain such drug selectivity. For instance, inhibitory avoidance is due to response suppression, whereas one-way escape involves response emission, and nature of the response has been suggested as an important determinant of drug effects in animal models of anxiety [ 7 ] . ACKNOWLEDGEMENTS
This work was supported by FAPESP Grant 94/0821-1 and CNPq Grant 520247/94-9. H. Zangrossi, Jr. is recipient of a Postdoc research fellowship from CNPq. We are grateful to J. R. Stella and P. Castrechini for technical assistance.
REFERENCES 1. Adamec, R. E.; Shallow, T. Lasting effects on rodent anxiety of a single exposure to a cat. Physiol. Behav. 54:101–109; 1993. 2. Barret, J. E.; Vanover, K. E. 5-HT receptors as targets for the development of novel anxiolytic drugs: Models, mechanisms and future directions. Psychopharmacology (Berlin) 112:1–12; 1993. 3. File, S. E. Behavioural detection of anxiolytic action. In: Elliott, J. M.; Heal, D. J.; Marsden, C. A., eds. Experimental approaches to anxiety and depression. New York: John Wiley Sons; 1992:25–44. 4. Graeff, F. G. Neurotransmitters in the dorsal periaqueductal grey and animal models of panic anxiety. In: Briley, M.; File, S. E., eds. New concepts in anxiety. London: Macmillan Press; 1991:288–312.
5 5. Graeff, F. G.; Viana, M. B.; Tomaz, C. The elevated T maze, a new experimental model of anxiety and memory: Effect of diazepam. Braz. J. Med. Biol. Res. 26:67–70; 1993. 6. Graeff, F. G.; Viana, M. B.; Mora, P. O. Opposed regulation by dorsal raphe nucleus 5-HT pathways of two types of fear in the elevated T-maze. Pharmacol. Biochem. Behav. 53:171–177; 1996. 7. Handley, S. L. Serotonin in animal models of anxiety: The importance of stimulus and response. In: Idzidowzki, C.; Cowen, P. J., eds. Serotonin, sleep and mental disorder. Petersfield: Wrightson Biomedical; 1991:89–115. 8. Handley, S. L.; McBlane, J. W. 5-HT drugs in animal models of anxiety. Psychopharmacology (Berlin) 112:13–20; 1993. 9. Johnston, A. L., File, S. E. Can animal tests of anxiety detect panicpromoting agents? Hum. Psychopharmacol. 3:149–152; 1988. 10. Mineka, S.; Keir, R.; Price, V. Fear of snakes in wild- and laboratory-reared rhesus monkeys (Macaca mulatta). Anim. Learn. Behav. 8:653–663; 1980. 11. Mineka, S.; Davidson, M.; Cook, M.; Keir, R. Observational conditioning of snake fear in rhesus monkey. J. Abnorm. Psychol. 93:355–372; 1984. 12. Montgomery, K. C. The relation between fear induced by novel stimulation and exploratory behaviour. J. Comp. Physiol. Psychol. 48:254–260; 1955. 13. Nutt, D. J. Anxiety and its therapy: Today and tomorrow. In: Briley, M.; File, S. E., eds. New concepts in anxiety. London: Macmillan Press; 1991:1–12. 14. Pellow, S.; Chopin, P.; File, S. E.; Briley, M. Validation of open:closed arm entries in an elevated plus-maze as a measure of anxiety in the rat. J. Neurosci. Methods 14:149–167; 1985. 15. Pesold, C.; Treit, D. The septum and amygdala differentially mediate the anxiolytic effects of benzodiazepines. Brain Res. 638:295–301; 1994. 16. Pesold, C.; Treit, D. The central and basolateral amygdala differentially mediate the anxiolytic effects of benzodiazepines. Brain Res. 671:213–221; 1995. 17. Treit, D.; Menard, J.; Royan, C. Anxiogenic stimuli in the elevated plus-maze. Pharmacol. Biochem. Behav. 44:463–469; 1993. 18. Treit, D.; Robinson, A.; Rotzinger, S.; Pesold, C. Anxiolytic effects of serotonergic interventions in the shock-probe burying test and the elevated plus-maze test. Behav. Brain Res. 54:23–34; 1993. 19. Walk, R. D.; Gibson, E. J.; Tighe, T. J. Behavior of light- and darkreared rats on a visual cliff. Science 126:80–81; 1957. 20. Viana, M. B.; Tomaz, C.; Graeff, F. G. The elevated T-maze: An animal model of anxiety and memory. Pharmacol. Biochem. Behav. 49:549–554; 1994. 21. Zangrossi, H., Jr.; File, S. E. Chlordiazepoxide reduces the generalised anxiety, but not the direct responses, of rats exposed to cat odor. Pharmacol. Biochem. Behav. 43:1195–1200; 1992. 22. Zangrossi, H., Jr. File, S. E. Habituation and generalization of phobic responses to cat odor. Brain Res. Bull. 33:189–194; 1994.
/ 2a40 2322 Mp 5 Tuesday Jul 22 08:51 AM EL–BRB (v. 43, no. 6) 2322
PII S0361-9230(96)00381-4
Behavioral Validation of the Elevated T-Maze, a New Animal Model of Anxiety HE´LIO ZANGROSSI, JR.1 AND FREDERICO G. GRAEFF Laborato´rio de Psicofarmacologia, FFCLRP e Nu´cleo de Neurocieˆncias e Comportamento da Universidade de Sao Paulo, USP, 14040-901 Ribeirao Preto, SP, Brazil [Received 17 June 1996; Accepted 14 October 1996] ABSTRACT: The elevated T-maze test of anxiety has been used to separate in the same rat conditioned from unconditioned responses of fear/anxiety. The test apparatus consists of three elevated arms—one enclosed and two open. Inhibitory avoidance—representing learned fear—is measured by recording the time taken to leave the enclosed arm in three consecutive trials. Unconditioned fear is evaluated by recording the time to escape from the open arm. In this study we investigated procedural questions raised by the use of the elevated T-maze. Experiment 1 showed that restraining the animals at the end of the enclosed arm for 30 s did not change the first (baseline) latency to leave this arm, indicating that aversion for the hands of the experimenter is not a key motivation for this response. The results of Experiment 2 indicated that open-arm experience, but not handling stress is the main cause for inhibitory avoidance acquisition, because rats trained in a T-maze with the three arms enclosed did not show the usual increase in withdrawal latency along the three consecutive trials. The same experiment also showed that the latency to leave the open arm did not undergo habituation over five consecutive trials, evidencing an aversive motivation for this response. The importance of open-arm experience for inhibitory avoidance acquisition was further suggested by the results of Experiment 3, as the removal of a shield that prevents perception of openness tended to increase avoidance latency in the elevated T-maze from the first trial. Q 1997 Elsevier Science Inc.
[ 2,3,7,8,15,16,18 ] ) . In this regard, attempts have been made to relate certain types of experimental models to specific anxiety disorders [1,4,9 – 11,21,22 ] . Recently, Graeff and co-workers [5,20] developed an animal model of anxiety to separate, in the same rat, conditioned from unconditioned fear, which have been related to generalized anxiety and panic disorders, respectively. This test, named the elevated T-maze, is derived from the elevated plus-maze, a widely used animal model of anxiety [14]. The latter has been classified as a ‘‘mixed’’ test of anxiety, because it does not separate different types of fear. This may be related to the well documented inconsistency of the effects of 5-HT interacting drugs in the elevated plus-maze [8]. The elevated T-maze apparatus consists of three arms of equal dimension elevated 50 cm from the floor. One of these arms is enclosed by lateral walls and is perpendicular to the two opposed open arms. When the rat is placed at the end of the enclosed arm, he does not see the open arms until he pokes his head beyond the wall of the closed arm. To be on an open arm seems to be an aversive experience, because rats have an innate fear of height and openness [12,14,17,19]. This allows the animal to learn inhibitory avoidance if repeatedly placed inside the enclosed arm to explore the maze. On the other hand, when the rat is placed at the end of one of the open arms he can move towards the closed arm, presumably performing an escape response. While inhibitory avoidance of the open arm is supposed to represent learned fear, the escape response from the open arm would represent innate fear. In a validating study with anxiolytic drugs, both the benzodiazepine agonist diazepam and the 5-HT 1A ligand ipsapirone impaired inhibitory avoidance in a dose-dependent way. Therefore, inhibitory avoidance in the T-maze may be related to pathologic conditions that are responsive to 5-HT 1A agonists or low doses of benzodiazepines, such as anticipatory anxiety and generalized anxiety disorder. In contrast, one-way escape behavior was not affected by the two drugs studied. The latter result may correlate with the ineffectiveness of these drugs on panic disorder [13]. Differential effects of drugs acting on the serotonergic system arising from the dorsal raphe nucleus on inhibitory avoidance and escape behaviors in the elevated T-maze have also been reported, further indicating that different kinds of fear are being generated by this model [6].
KEY WORDS: One-way escape, Inhibitory avoidance, Types of fear/anxiety, Generalized anxiety, Panic.
INTRODUCTION The use of animal models of anxiety has been fundamental in the search for new anxiolytic compounds and for the understanding of the brain mechanisms underlying anxiety. Clinically, it has been recognized that anxiety is not a unitary phenomenon, and the existence of different types of pathologic anxiety, such as panic, phobias, and posttraumatic stress disorders has been postulated. During the last years, a growing body of evidence has been accumulated indicating that anxiety defined operationally in a given animal model may differ from that generated by other models in respect to its nature ( innate or learned ) , its susceptibility to the effects of drugs and environmental manipulations, or to its neural substrate ( e.g., 1
To whom requests for reprints should be addressed.
1 / 2a40 2322 Mp 1 Tuesday Jul 22 08:51 AM EL–BRB (v. 43, no. 6) 2322
2
ZANGROSSI AND GRAEFF
Although the results of these pharmacological studies are promising, there are procedural questions related to the use of the test that deserve further consideration. For example, during acquisition of inhibitory avoidance the rat is placed at the distal end of the enclosed arm and the latency to leave this arm is measured immediately after the animal is released from the experimenter’s hand. Therefore, the initial motivation for the animal to walk along the enclosed arm may be to escape from the hand. To assess this influence, in Experiment 1 we compared the latencies of inhibitory avoidance under two conditions. In the first, the standard procedure [20] was used, while in the second the animals were kept at the end of the enclosed arm for 30 s by a barrier held 20 cm from the arm’s end. Another question relevant for the interpretation of T-maze results is the nature of the aversive stimulus that promotes inhibitory avoidance. Treit and co-workers [17] have reported that in the elevated plus-maze—the T-maze precursor—lack of a protective wall allowing thigmotaxis in the open arms is the critical factor that motivates preference for the enclosed arms. For the same reason rats would learn to avoid the open arms of the Tmaze by remaining in the enclosed arm. However, manipulation by the experimenter may also be involved, because after each trial the animal is removed from the apparatus by the experimenter’s hand. To test this hypothesis, in Experiment 2 we compared the latency to repeatedly leave the enclosed arm in the original elevated T-maze to that in a T-maze built up with the three arms enclosed by walls. If in this enclosed T-maze the latency increases in the same way as in the elevated T-maze, the animal is either avoiding manipulation, or simply decreasing exploration, due to habituation. On the other hand, if the latency in the enclosed T-maze remains stable over trials, the crucial factor leading to the latency increase in the regular T-maze is likely to be open arm experience. Furthermore, it has been observed [20] that the latency to leave the enclosed arm of the elevated T-maze for the first time (baseline latency) is not significantly different from the latency to leave the open arm (escape latency). Thus, it is not clear whether the rat is really escaping from the open arm or just exploring the new environment. To investigate this possibility, in Experiment 2 we also measured the latency for the animals to leave one of the open arms in five consecutive trials. These latencies were compared to those taken in animals exposed to the correspondent arm of the enclosed T-maze. We expected that while the latencies in the enclosed T-maze would increase gradually over trials, reflecting habituation of exploratory activity, the latencies to leave the open arm of the elevated T-maze could either remain unchanged or even decrease, due to persistent aversive motivation. In this regard, it has been reported that relative open arm exploration in the elevated plus-maze either does not increase or significantly decreases over successive exposures to the apparatus [14,17]. As the results of Experiment 2 have indicated that the crucial factor in motivating inhibitory avoidance in the T-maze is the aversive nature of the open arms, in Experiment 3 we removed the shield placed at the border of the intersection of the closed with the open arms of the maze. This shield blocks the perception of openness until the rat pokes his head beyond the walls of the enclosed arm. As a consequence we expected that avoidance latencies would be increased by this procedure. MATERIALS AND METHODS Animals Male Wistar rats weighing 220–250 g were housed in group of five with food and water freely available in a room maintained at 22 { 17C. Lights were on from 0700 to 1900 h.
Apparatus The elevated T-maze (Fig. 1) was made of wood and had three arms of equal dimensions (50 1 12 cm). One arm, the stem of the T, was enclosed by 40-cm high walls and was perpendicular to two opposed open arms. To avoid the rats falling down, the open arms were surrounded by a Plexiglas rim 1 cm high. The whole apparatus was elevated 50 cm above the floor. The enclosed T-maze used in Experiment 2 had three arms (50 1 12 cm) all surrounded by 40-cm high walls. The experiments were performed with an observer inside the room. Procedure Experiment 1. On the third and fourth days after their arrival in the laboratory, animals were gently handled for 5 min. On the fifth day, 22 rats were randomly allocated between the normal (n Å 11) and the barrier (n Å 11) procedure groups. Two minutes before the test, each animal was put in a cage (28 1 18 cm), to which they had been previously habituated. For the normal procedure group, each rat was removed from the cage and placed at the distal end of the enclosed arm facing the intersection of the arms. The time taken by the rat to leave this arm with the four paws was recorded (baseline latency). The same measurement was repeated in two subsequent trials (Avoidance 1 and Avoidance 2) at 30-s intervals. Following avoidance training (30 s), the rat was placed at the end of the right open arm and the time to leave this arm with the four paws was recorded (escape). During the 30-s intertrial intervals the animals were put back in the cage. The animals of the barrier procedure group were placed at the end of the enclosed arm facing a barrier (10 cm large and 40 cm high) introduced 20 cm from the distal end of this arm. After 30s the barrier was removed and the time taken to leave this arm was recorded (baseline latency). The same procedure was repeated in two subsequent trials (Avoidance 1 and Avoidance 2). Immediately after Avoidance 2 the animals were put back at the
FIG. 1. Drawing of the elevated T-maze.
/ 2a40 2322 Mp 2 Tuesday Jul 22 08:51 AM EL–BRB (v. 43, no. 6) 2322
ELEVATED T-MAZE BEHAVIORAL VALIDATION end of the enclosed arm where they stayed restricted by the barrier for 30 s. After this time, they were placed at the end of the open arm for escape latency measurement. Experiment 2. The animals were handled as in Experiment 1. On the test day, 19 rats were allocated between two groups; one was tested in the elevated T-maze (n Å 9), and the other in the enclosed T-maze (n Å 10). For both groups, the latencies to leave the stem of the maze were measured as for the normal procedure group in Experiment 1. After 30 s, the animals were put at the end of the right open arm (elevated T-maze group) or at the end of the right transversal arm of the enclosed T-maze. The latencies to leave these arms were measured in five consecutive trials at intervals of 30-s. Experiment 3. The animals were handled as in Experiment 1. On the test day, 28 rats were allocated between two groups; one was tested in the regular elevated T-maze (n Å 14) an the other in an elevated T-maze without the shield (see Fig. 1) at the border of the intersection of the closed with the open arms (n Å 14). For both groups, avoidance and escape latencies were measured as for the normal procedure group described in Experiment 1. Statistics The latencies to leave the stem of the T-maze (avoidance in the regular T-maze) were analyzed by split-plot analysis of variance, with test conditions (Experiment 1—barrier or no-barrier; Experiment 2—elevated or enclosed T-maze; Experiment 3— shield or no-shield) as the independent factor and trials as the repeated measure. When appropriated, between-group comparisons within each trial were made by the unpaired Student’s ttest. Latencies to leave the transversal arm (escape in the regular T-maze) were analyzed by the unpaired Student’s t-test (Experiment 1 and 3) or by split-plot analysis of variance (Experiment 2). In the latter case, test condition (elevated or enclosed Tmaze) was the independent factor and trials the repeated measure (trials 1–5). Between-group comparisons within each trial were made by the unpaired Student’s t-test. To analyze between trials changes, a repeated measure analysis of variance followed by Duncan multiple-comparison test was also performed in Experiment 2. RESULTS
3
FIG. 2. Effect of restraining with a barrier on inhibitory avoidance and escape latencies in the elevated T-maze. Bars represent the mean and the vertical lines the SEM. In the normal procedure group (NORMAL; n Å 11), latencies (BASELINE, AVOID 1 and AVOID 2) were measured immediately after the animals being released by the experimenter’s hand. In the barrier procedure group (BARRIER; n Å 11) these measurements were performed after the animals being restricted by a barrier for 30 s. ESCAPE was measured 30 s after AVOID 2. *p õ 0.05 compared with the normal procedure group within a same trial.
over trials, the latencies in the enclosed T-maze were not [trial effect: F(2, 34) Å 8.6, p õ 0.001; procedure 1 trial interaction: F(2, 34) Å 4.0, p õ 0.05]. It can also be seen in Fig. 3 that the latencies to leave the open arm of the elevated T-maze were significantly different from those measured in a transversal arm of the enclosed T-maze [procedure effect: F (1, 17) Å 7.6, p õ 0.05]. There was also a trial effect, F (4,68) Å 2.6, p õ 0.05, and a tendency for a procedure by trial interaction, F (4, 68) Å 2.1, p Å 0.09. A repeated measure analysis of variance followed by the Duncan multiple comparison test showed that in the enclosed T-maze group the first trial latency in the transversal arm was significantly shorter (p õ 0.05) than the other trials while in the elevated T-maze there was no significant difference among trials. This result indicates that withdrawal from the open arm does not undergo habituation.
Experiment 1 Figure 2 shows that both normal and barrier procedure groups acquired inhibitory avoidance in the enclosed arm of the elevated T-maze [trial effect: F (2, 40) Å 40.5, p õ 0.0001]. There was a significant difference between the two groups in avoidance performance [procedure effect: F (1, 20) Å 8.3, p õ 0.01) and a significant procedure by trial interaction [procedure 1 trial: F (2, 40) Å 3.8, p õ 0.05], because the barrier facilitated inhibitory avoidance acquisition. Within trials comparison showed that the barrier increased Avoidance 1 and 2 latencies when compared to the normal procedure group. Nevertheless, there was no difference between the two groups in the baseline latency. It can also be seen in Fig. 2 that there was no significant difference (p ú 0.05) between the two groups in the time taken to leave the open arm of the elevated T-maze. Experiment 2 Figure 3 shows that there was a significant difference in the acquisition of inhibitory avoidance in the elevated T-maze as compared to the enclosed T-maze [procedure effect: F (1, 17) Å 7.9, p õ 0.05]. While the latencies for the animals to leave the enclosed arm of the elevated T-maze were significantly increased
Experiment 3 Figure 4 shows that the animals tested in the elevated T-maze either with or without the shield acquired inhibitory avoidance from the open arms [trial effect: F (2, 56) Å 20.2, p õ 0.0001]. Nevertheless, the absence of the shield tended to increase avoidance latencies along the trials. This tendency approached statistical significance [procedure effect: F (1, 28) Å 3.4, p Å 0.08]. However, the procedure by trial interaction was not significant [procedure 1 trial: F(2, 56) Å 0.27, p ú 0.05]. Regarding escape, there was no difference (p ú 0.05) between the two groups in the latency to leave the open arm (Fig. 4). DISCUSSION The results of Experiment 1 show that restraint with a barrier for 30 s at the end of the enclosed arm did not change the first latency to leave this arm of the elevated T-maze. Therefore, in the standard procedure the initial locomotion along the enclosed arm does not seem to constitute an escape response from the experimenter’s hand. However, Avoidance 1 and 2 latencies were longer in the barrier group. The reason for this difference is not obvious. One possibility is that restraint may generate anxiety,
/ 2a40 2322 Mp 3 Tuesday Jul 22 08:51 AM EL–BRB (v. 43, no. 6) 2322
4
ZANGROSSI AND GRAEFF
FIG. 3. Latencies to withdraw from the arms of the elevated and the enclosed T-mazes. On the left of the graph, the time taken (mean / SEM) to leave the stem of the elevated T-maze (n Å 9) or of the enclosed T-maze (n Å 10) were measured in three consecutive trials (TRIAL 1–3) at 30 s interval. Thirty seconds after trial 3, the latency to leave the right transversal arm of the elevated T-maze (open) or of the enclosed T-maze (enclosed) were measured in five consecutive trials (TRIAL 1–5; right of the graph) at 30 s interval. *p õ 0.05 compared with the enclosed T-maze group within a same trial.
facilitating avoidance acquisition. Alternatively, escape from the hand may be speeding up withdrawal from the enclosed arm. A floor effect would have effaced the difference between groups in the first trial. Still another possibility is that repeated exposure to
FIG. 4. Effect of removing the wooden shield held in front of the enclosed arm on inhibitory avoidance and escape latencies in the elevated T-maze. Bars represent the mean and the vertical lines the SEM of groups of 14 rats. Measurements were performed as for the normal procedure group described in the legend of Fig. 2.
a constrained space, when the barrier was used, might accelerate habituation to the enclosed arm. It is, nevertheless, clear that rats belonging to either the barrier or the no-barrier group show significant learning along trials. The present results with the enclosed T-maze further show that latency to leave the stem of the T remained stable along the three trials corresponding to inhibitory avoidance acquisition in the elevated T-maze (Experiment 2). This result may have several meanings: first, it indicates that habituation of exploratory activity does not contribute to the increase in latency observed in the elevated T-maze; second, it tends to rule out handling stress as a cause for inhibitory avoidance acquisition, because in both the enclosed and the elevated T-maze the rats were picked up by the experimenter soon after arm emergence; third, experience of the open arms would remain, by exclusion, the crucial motivation for avoidance acquisition, as originally assumed [5,20]. A different picture emerged in the same experiment (number 2) when the latency to leave the right transversal arm—that is, open in the elevated maze—was measured five times in succession, soon after the three above measures in the stem had been taken. In the elevated T-maze the time taken to leave the open arm remained unchanged, whereas the latency to leave the transversal arm of the enclosed T-maze significantly increased over consecutive trials. Therefore, habituation has taken place in the enclosed arm only. Its rapid onset is probably due to the previous exposure to the similarly enclosed stem of the T. Therefore, the absence of habituation in the open arm has unveiled the aversive
/ 2a40 2322 Mp 4 Tuesday Jul 22 08:51 AM EL–BRB (v. 43, no. 6) 2322
ELEVATED T-MAZE BEHAVIORAL VALIDATION motivation determined by exposure to this situation. Indeed, previously reported results strongly suggest that fear of openness does not easily habituate, probably due to its high survival value [17]. Nevertheless, because in the first escape trial—which is the only one normally used in the standard procedure—the rat leaves with the same speed as in the transversal arm of the enclosed T-maze, the associated influence of the exploratory drive cannot be excluded. The critical role of open arm experience in the elevated Tmaze is also suggested by the results of Experiment 3, showing that removal of the shield that prevents perception of openness when the rat is inside the stem of the T tended to increase avoidance latencies from the first trial (BASELINE in Fig. 4). Overall, the results of the present study indicate that openness is the main factor motivating both inhibitory avoidance and one-way escape in the elevated T-maze. The original assumption that this test generates two types of fear is thus tenable in view of the reported differential sensitivity of these behaviors to several pharmacological treatments [ 5,6,20 ] . Nevertheless, it should be kept in mind that other factors may as well explain such drug selectivity. For instance, inhibitory avoidance is due to response suppression, whereas one-way escape involves response emission, and nature of the response has been suggested as an important determinant of drug effects in animal models of anxiety [ 7 ] . ACKNOWLEDGEMENTS
This work was supported by FAPESP Grant 94/0821-1 and CNPq Grant 520247/94-9. H. Zangrossi, Jr. is recipient of a Postdoc research fellowship from CNPq. We are grateful to J. R. Stella and P. Castrechini for technical assistance.
REFERENCES 1. Adamec, R. E.; Shallow, T. Lasting effects on rodent anxiety of a single exposure to a cat. Physiol. Behav. 54:101–109; 1993. 2. Barret, J. E.; Vanover, K. E. 5-HT receptors as targets for the development of novel anxiolytic drugs: Models, mechanisms and future directions. Psychopharmacology (Berlin) 112:1–12; 1993. 3. File, S. E. Behavioural detection of anxiolytic action. In: Elliott, J. M.; Heal, D. J.; Marsden, C. A., eds. Experimental approaches to anxiety and depression. New York: John Wiley Sons; 1992:25–44. 4. Graeff, F. G. Neurotransmitters in the dorsal periaqueductal grey and animal models of panic anxiety. In: Briley, M.; File, S. E., eds. New concepts in anxiety. London: Macmillan Press; 1991:288–312.
5 5. Graeff, F. G.; Viana, M. B.; Tomaz, C. The elevated T maze, a new experimental model of anxiety and memory: Effect of diazepam. Braz. J. Med. Biol. Res. 26:67–70; 1993. 6. Graeff, F. G.; Viana, M. B.; Mora, P. O. Opposed regulation by dorsal raphe nucleus 5-HT pathways of two types of fear in the elevated T-maze. Pharmacol. Biochem. Behav. 53:171–177; 1996. 7. Handley, S. L. Serotonin in animal models of anxiety: The importance of stimulus and response. In: Idzidowzki, C.; Cowen, P. J., eds. Serotonin, sleep and mental disorder. Petersfield: Wrightson Biomedical; 1991:89–115. 8. Handley, S. L.; McBlane, J. W. 5-HT drugs in animal models of anxiety. Psychopharmacology (Berlin) 112:13–20; 1993. 9. Johnston, A. L., File, S. E. Can animal tests of anxiety detect panicpromoting agents? Hum. Psychopharmacol. 3:149–152; 1988. 10. Mineka, S.; Keir, R.; Price, V. Fear of snakes in wild- and laboratory-reared rhesus monkeys (Macaca mulatta). Anim. Learn. Behav. 8:653–663; 1980. 11. Mineka, S.; Davidson, M.; Cook, M.; Keir, R. Observational conditioning of snake fear in rhesus monkey. J. Abnorm. Psychol. 93:355–372; 1984. 12. Montgomery, K. C. The relation between fear induced by novel stimulation and exploratory behaviour. J. Comp. Physiol. Psychol. 48:254–260; 1955. 13. Nutt, D. J. Anxiety and its therapy: Today and tomorrow. In: Briley, M.; File, S. E., eds. New concepts in anxiety. London: Macmillan Press; 1991:1–12. 14. Pellow, S.; Chopin, P.; File, S. E.; Briley, M. Validation of open:closed arm entries in an elevated plus-maze as a measure of anxiety in the rat. J. Neurosci. Methods 14:149–167; 1985. 15. Pesold, C.; Treit, D. The septum and amygdala differentially mediate the anxiolytic effects of benzodiazepines. Brain Res. 638:295–301; 1994. 16. Pesold, C.; Treit, D. The central and basolateral amygdala differentially mediate the anxiolytic effects of benzodiazepines. Brain Res. 671:213–221; 1995. 17. Treit, D.; Menard, J.; Royan, C. Anxiogenic stimuli in the elevated plus-maze. Pharmacol. Biochem. Behav. 44:463–469; 1993. 18. Treit, D.; Robinson, A.; Rotzinger, S.; Pesold, C. Anxiolytic effects of serotonergic interventions in the shock-probe burying test and the elevated plus-maze test. Behav. Brain Res. 54:23–34; 1993. 19. Walk, R. D.; Gibson, E. J.; Tighe, T. J. Behavior of light- and darkreared rats on a visual cliff. Science 126:80–81; 1957. 20. Viana, M. B.; Tomaz, C.; Graeff, F. G. The elevated T-maze: An animal model of anxiety and memory. Pharmacol. Biochem. Behav. 49:549–554; 1994. 21. Zangrossi, H., Jr.; File, S. E. Chlordiazepoxide reduces the generalised anxiety, but not the direct responses, of rats exposed to cat odor. Pharmacol. Biochem. Behav. 43:1195–1200; 1992. 22. Zangrossi, H., Jr. File, S. E. Habituation and generalization of phobic responses to cat odor. Brain Res. Bull. 33:189–194; 1994.
/ 2a40 2322 Mp 5 Tuesday Jul 22 08:51 AM EL–BRB (v. 43, no. 6) 2322