Brain Research, 512 (1990) 81-88 Elsevier
81
BRES 15292
GABA receptors in the posterior hypothalamus regulate experimental anxiety in rats A. Shekhar, J.N. Hingtgen and J.A. DiMicco Departments of Pharmacology, and Toxicology and Psychiatry, and Program in Medical Neurobiology, Indiana University School of Medicine, Indianapolis, IN 46202 (U.S.A.) (Accepted 15 August 1989) Key words: y-Aminobutyric acid; Hypothalamus; Anxiety; Conflict; Muscimol; Picrotoxin; Bicuculline
Blockade of y-aminobutyric acid (GABA) function in the posterior hypothalamus of rats elicits a pattern of physiological and behavioral arousal consisting of increases in heart rate, respiration and blood pressure as well as intense locomotor stimulation and a selective enhancement of avoidance responding. The present study was conducted to assess the possibility that GABA-mediated neurotransmission in the posterior hypothalamus of the rat may regulate anxiety. Male rats were trained in a 'conflict' schedule consisting of a high and a low intensity of punishment ('high' and 'low' conflict) capable of measuring decreases and increases in the level of 'anxiety', respectively. Guide cannulae were stereotaxicaUy implanted bilaterally in the posterior hypothalamus of these rats at sites where microinjection of bicuculline methiodide (BMI) 25 ng caused increases in heart rate under anesthesia. After recovery, they were tested: (1) in the high conflict schedule after microinjection of saline and two doses of the GABAA receptor agonist muscimol; and (2) in the low conflict schedule after injecting saline, the GABAA receptor antagonists, BMI and picrotoxin, and the glycine antagonist, strychnine. Injection of muscimol caused a significant and selective anti-conflict effect while both BMI and, at appropriate doses, picrotoxin produced pro-conflict effects. Microinjection of strychnine into the posterior hypothalamus or muscimol and picrotoxin into the lateral hypothalamus did not influence conflict responding. These results suggest that endogenous GABA acts on GABAA receptors in a discrete area of the posterior hypothalamus to regulate the level of experimental anxiety in rats.
INTRODUCTION The hypothalamus is thought to play an important role in regulating the physiological and behavioral responses associated with emotional arousal. Electrical stimulation of specific hypothalamic nuclei elicits increases in heart rate, blood pressure and respiration 9 as well as behavioral changes suggesting a 'fight-or-flight' reaction 8. Electrical stimulation of the posterior hypothalamus causes a behavioral response that is predominantly 'aversive' in nature ~2. The integrity of the hypothalamus appears to be essential for emotional expression, and electrolytic lesions of the posterior hypothalamus in humans causes a loss of emotional excitability and a decrease in anxiety 1°. However, findings from electrical stimulation or ablation studies are subject to considerable interpretational ambiguity because of potential effects resulting from damage or excitation of fibers of passage as well as from destruction or stimulation of cell bodies. Furthermore, these techniques alone fail to provide any insight into the identity of the neurotransmitters involved. In recent pharmacological studies, the inhibitory neurotransmitter
~-aminobutyrate ( G A B A ) has been repeatedly implicated in the regulation of emotional expression and behavior. For example, the intensity of the 'defense' reaction elicited by electrical stimulation of the hypothalamus is decreased by systemic administration of benzodiazepines, which are thought to act through a G A B A ergic mechanism 15, and blockade of G A B A e r g i c inhibition in the hypothalamus elicits a reaction resembling escape or flight in freely moving rats 6'17. Recent studies in our laboratory have shown that microinjection of drugs that impair G A B A - m e d i a t e d neurotransmission into a discrete region of the posterior hypothalamus produces increased heart rate and respiration, and a smaller increase in blood pressure in anesthetized 4'5 as well as in conscious and freely moving rats 21. Further, G A B A blockade in the same cardiostimulatory area of the posterior hypothalamus causes 'escape'-oriented locomotor stimulation in conscious rats ~7 and selectively enhances aversively-motivated responding while having no significant effect on 'approach', or positively-reinforced, responding 18. All of these findings suggest that a G A B A e r g i c mechanism may tonically
Correspondence: J.A. DiMicco, Department of Pharmacology and Toxicology, Indiana University School of Medicine, 635 Barnhill Drive, Indianapolis, IN 46202, U.S.A. 0006-8993/90/$03.50 (~) 1990 Elsevier Science Publishers B.V. (Biomedical Division)
82 inhibit a p o p u l a t i o n of n e u r o n s in the p o s t e r i o r hypot h a l a m u s w h o s e activation is responsible for the physiological and b e h a v i o r a l r e s p o n s e s associated with states of emotional
arousal
such
as m i g h t be associated with
e x p e r i m e n t a l ' a n x i e t y ' and stress. T h e p r e s e n t study was u n d e r t a k e n to test the h y p o t h esis that G A B A e r g i c n e u r o t r a n s m i s s i o n in the p o s t e r i o r h y p o t h a l a m u s r e g u l a t e s the level of ' a n x i e t y ' in conscious and
freely
moving
animals.
To
assess
experimental
' a n x i e t y ' , we e m p l o y e d a m o d i f i e d 'conflict' p a r a d i g m in which the conflict is g e n e r a t e d positively r e i n f o r c e d demonstrated
behavior.
by p u n i s h m e n t
of a
P r e v i o u s studies
have
increases and d e c r e a s e s in the level of
' a n x i e t y ' in the p a r t i c u l a r conflict p r o c e d u r e e m p l o y e d in this study with drugs that are t h o u g h t to be anxiolytic ( b e n z o d i a z e p i n e s ) and a n x i e t y - p r o v o k i n g (fl-carbolines) in h u m a n s t9. R a t s t r a i n e d in a v a r i a b l e interval a p p r o a c h s c h e d u l e to o b t a i n s w e e t e n e d milk w e r e i n t r o d u c e d to a ' c o n f l i c t ' p a r a d i g m with two p u n i s h m e n t levels: a 'high c o n f l i c t ' p e r i o d , c a p a b l e of d e t e c t i n g 'anxiolytic' activity and
a ' l o w conflict' p e r i o d ,
subthreshold pression,
for g e n e r a t i n g
capable
using a shock
intensity
punishment-induced
of d e m o n s t r a t i n g
increases
sup-
in the
' a n x i e t y ' level of the animal. T h e p e r f o r m a n c e of these rats was assessed a f t e r m i c r o i n j e c t i n g G A B A e r g i c drugs into t h e p o s t e r i o r h y p o t h a l a m u s t h r o u g h c a n n u l a e imp l a n t e d at the c a r d i o s t i m u l a t o r y region. T h e effects of injecting
G A B A A antagonists,
picrotoxin
(PIC)
and
bicuculline m e t h i o d i d e ( B M I ) , w e r e t e s t e d in the low conflict s c h e d u l e while the effect of the G A B A A agonist, muscimol,
in the
high conflict schedule.
Strychnine, a non-GABAergic
was
tested
C N S stimulant, was in-
j e c t e d at the s a m e site to d e t e r m i n e p h a r m a c o l o g i c a l specificity of t h e effects while a n a t o m i c a l specificity was e s t a b l i s h e d by p e r f o r m i n g a parallel series of e x p e r i m e n t s with rats that had c a n n u l a e
implanted
in the lateral
h y p o t h a l a m i c area.
MATERIALS AND METHODS
Behavioral training Experimentally naive approximately 6-month-old male SpragueDawley rats (225-300 g) maintained at 80% free-feeding weights were trained in a variable interval (VII) approach schedule TM. The conditioning apparatus consisted of lever-pressing chambers made of 3 mm plastic with the inner dimensions of 21 x 16 × 14 era. These boxes had a lever and a dipper feeding device mounted on the front of each box. A noise-generating device was attached to the back of the box and the entire unit placed in a sound-insulated outer compartment. On the VI 1 schedule lever presses delivered 0.2 ml of sweetened milk on the dipper at an average rate of once per minute. Each rat was given one VI session (120 rain) each day for 5 days each week. Cumulative recorders, running time meters and counters were used to record the responding in each session. After stable VI responding was established, the conflict protocol was introduced. The protocol consisted of 10 rain of initial VI period (pre-confliet) which was followed by the conflict period signalled by
a constant white noise. During the 10 min conflict period, the dipper delivered an electric shock along with the milk on a VI schedule. All rats were trained in both 'high' and 'low' conflict periods distinguished by reinforcement-coupled punishment at two levels of intensity. During high conflict the dipper delivered a 0.4 mA shock along with the milk and during low conflict the dipper delivered a 0.1 mA shock. The conflict period was followed by another 10 min of VI schedule (post-conflict). The high conflict period was signalled by a full intensity (-14 dB, where 0 d B = 1 roW) noise and the low conflict period was signalled by half intensity (-31 dB) noise. Rats were trained in the above paradigm until they discriminated between the two levels of conflict clearly such that the high conflict schedule suppressed responding during the conflict period to less than 10% of the pre-conflict VI period while the minimal punishment delivered during the low conflict period had no significant effect on responding. These high and low punishment levels were previously shown to be effective in detecting the 'anxiolytic' and 'anxiogenic' effects of drug treatments, respectively v~
Implantation of microinjection cannulae at the cardiostimulatory site in the posterior hypothalamus Cannulae were chronically implanted at the active site in the posterior hypothalamus using a technique routinely employed in this laboratory 1a'17'21. Animals were anesthetized with pentobarbital (50 mg/kg) and fixed into a stereotaxic apparatus. Heart rate was monitored throughout the surgery by a cardiotachometer triggered by lead II of the electrocardiogram and recorded on a Beckman R511 Dynograph. Intermittent heating with an infrared lamp was used to maintain the body temperature between 36 and 38 °C. Two stainless steel guide cannulae (22-gauge, 10 mm length) were mounted on the stereotaxic apparatus on either side with the internal cannulae (28-gauge, 12 mm length) fixed in place within them. The internal cannulae were connected to 10 #1 Hamilton microinjection syringes by polyethylene tubing (PE 50) and the system was filled with a solution of BMI (25 ng//d). The microinjection syringes were placed on infusion pumps set to deliver 1 ~1 of injectate over 2 min through the internal cannulae. The internal cannula along with the guide cannula was lowered into the posterior hypothalamus on one side at an angle of 10° with respect to the vertical plane, the target coordinates with respect to bregma being 1.2 mm posterior, 0.5 mm lateral and 8.5-9.0 mm ventral according to the atlas of Pellegrino et al) 3. Bicuculline methiodide 25 ng in 1 /A was injected at this site and the resultant heart rate response was noted. An increase in heart rate of at least 50 beats/rain was the criterion used to assure that the cannulae were placed at physiologically active sites. If microinjection at these coordinates failed to elicit the expected heart rate response, the same process was repeated after repositioning the cannula 0.2 mm anteriorly, medially, posteriorly or laterally with respect to the original coordinates until an active site was found. After the heart rate had returned to baseline, the same procedure was repeated on the opposite side. The target coordinates used to implant the cannulae in the lateral hypothalamic area were 1.2 mm posterior, 2.5 mm lateral and 9 mm ventral with respect to bregma. After placing both cannulae at the physiologically active sites in the posterior hypothalamus or the inactive sites at the lateral hypothalamus, the guide cannulae were fixed at these positions with the help of 3 stainless steel screws anchored to the skull and cranioplastic cement (cannulae and cement obtained from Plastic Products Co., Roanoke, VA). The internal cannulae were then removed and the guide cannulae were sealed with steel wire dummy cannulae. The rats were allowed to recover in individual plastic cages with ad lib water and approximately 20 g of food per day.
Experimental protocol After at least 3 days of recovery, each of the rats was given additional training sessions in high and low conflict schedules daily (5 days/week). Within about one week, they usually attained their baseline rates of responding during all the 3 periods of the conflict schedule after which the following experimental protocol was begun.
83 TABLE I
Changes in heart rate elicited by unilateral injection of BM125 ng into different regions of the hypothalamus in anesthetized rats PH, posterior hypothalamus; LH, lateral hypothalamus,
Site of. in1.
Num ber of rats
Right side
Left side
Baseline heart rate (bpm)
Maximum change in HR (bpm)
Time to peak (rain)
Duration of effect (rain)
Baseline heart rate (bpm)
Maximum change in HR (bpm)
Time to peak (rain)
Duration of effect (rain)
PH LH
11 4
292 + 9 298 + 15
+91 _+9* - 2 + 15
9+ 1 -
22 + 2 -
284 + 7 303 + 11
+95 + 12' +5 _+ 12
11 + 1 -
26 + 3 -
* Significant change from baseline heart rate by paired t-test, P < 0.05.
Prior to placing each of the rats in the conflict box, microinjection cannulae were inserted into the guide cannulae and 1/A of saline or a GABAergic drug solution selected randomly was injected bilaterally into the hypothalamus over a 2 min period. The drugs (from Sigma, St. Louis, MO) used were BMI (25 ng//~l), picrotoxin (1, 4 or 100 ng//A) and muscimol (5 or 10 ng//A). One minute after completion of the injection, the microinjection cannulae were removed, the rat was placed in the conflict box and responding was recorded in either the low conflict schedule (after saline, BMI and picrotoxin injections) or the high conflict schedule (after saline and muscimol injections). After an interval of at least two days, a different reagent was injected and its effect on the animal's responding was recorded. No rat received more than 4 drug injections. Strychnine (38 ng//~l, a dose equimolar to BMI 25 ng//~l) was injected into the posterior hypothalamus of 4 rats and its effect
on their responding in the low conflict schedule was noted. In another group of 4 rats with cannulae implanted in the lateral hypothalamic area, the effects of bilateral injection of muscimol 10 ng and picrotoxin 1 ng were assessed in the high and low conflict schedules, respectively. The results of the behavioral changes were tabulated and expressed as mean +_ S.E.M. Statistical analysis was performed by paired t-test and the criterion for statistical significance was P < 0.05. After completion of the experiments, the animals were injected intraperitoneally with a lethal dose of pentobarbital. The brains were then removed and fixed in 4% formaldehyde after the microinjection sites were marked by injecting 0.25/.,1 of 1 N HCI. Brains were later sectioned into 5 0 / t m slices, stained with Cresyl violet and the exact site of injection was then determined by comparing the sections with the atlas of Pellegrino et al. 13.
'-1.0
"1.0
D1.2
-.1.2
E1.4
- - 1.4
Fig. 1. Schematic representation of the bilateral hypothalamic injection sites as confirmed by histology. The numbers represent injection sites in 14 of the 15 rats used in this study. Rats, nos. 1-10 had cannulae implanted bilaterally in the posterior hypothalamus; nos. 11-14 had the cannulae implanted bilaterally in the lateral hypothalamus. The distances of each section in mm posterior to bregma are indicated (adapted from Pellegrino et a1.13). FX, fornix; MT, mamillothalamic tract; PH, posterior hypothalamus; V, ventricle.
84
400
- KIs'L'NE I
A. CONTROL
296~
BMUSCIMOL
300
200
VI 1
(~
VI 2
100 0 --
5ng
B. MUSCIMOL 5 ng
,.~
(n=3)
q00 -56 - r - r - ~
300 2O0
VI 1 C C. MUSClMOL 10 ng
100 O
VI 2 /
lOng
Fig. 2. Summary of mean number of responses in each component of the high conflict (0.4 mA shock) schedule after bilateral injection of either saline or muscimol 5 ng (above) or 10 ng (below) into the posterior hypothalamus. Compared to control (saline), muscimol caused dose-related increases in responding only during the conflict period. * Significantly different from saline by paired t-test, P < 0.05.
RESULTS Increases in heart rate of at least 50 beats/min were o b t a i n e d by injecting 25 ng of B M I in 1 /A into the p o s t e r i o r h y p o t h a l a m u s on each side in 11 rats. The increases in heart rate, which began to develop during the injection of B M I , are given along with their time course in Table I. Microinjection of B M I 25 ng into the lateral h y p o t h a l a m u s failed to elicit any significant change in heart rate in 4 rats (Table I). The sites of injection on both sides as confirmed by histology in 14 of the 15 animals used in this study were in or immediately adjacent to the posterior hypothalamic nucleus (Fig. 1). O n e rat in which the cannulae had been i m p l a n t e d in the physiologically active sites died suddenly for unknown reasons before the site could be m a r k e d and the brain fixed. In the remaining 10 rats with the cannulae i m p l a n t e d in the cardiostimulatory area, the site of implantation was confirmed to be the posterior h y p o t h a l a m u s in or immediately adjacent to the posterior hypothalamic nucleus according to Pellegrino et al. 13. The sites of implantation in the 4 rats that had the cannulae positioned in the inactive sites were in the lateral h y p o t h a l a m u s (Fig. 1). Effect o f microinjecting drugs in the 'high' conflict schedule Total n u m b e r of responses in each c o m p o n e n t of the
Y
VI I
C
VI 2
Fig. 3. Experimental tracings depicting cumulative responding in the high conflict schedule in the same rat after microinjection of
saline (A) and muscimol 5 ng (B) or 10 ng (C) bilaterally into posterior hypothalamus. Cumulative lever presses are indicated by upward deflection of the pen and delivery of reinforcement by diagonal tics. In each case, the record represents a 30-min session, comprised of a conflict period ('C'; heavy bar), during which a shock (0.4 mA) is delivered along with the reinforcement, and which is preceded and followed by 10-min periods of variable interval unpunished reinforcement (VI~ and Vie). Numbers indicate total number of responses in each component.
'high' conflict schedule after the microinjection of saline or either of two doses of muscimol (5 and 10 ng) into the posterior h y p o t h a l a m u s are r e p r e s e n t e d in Fig. 2. A f t e r microinjection of saline vehicle, responding r e m a i n e d almost completely suppressed during the conflict period.
TABLE I1 Responding in the high conflict (0.4 mA) schedule after microinjection of saline and muscimo110 ng, and in the low conflict (0.1 mA shock) schedule after microinjection of saline and picrotoxin 1 ng into the lateral hypothalamus Data are presented as number of lever presses (mean + S.E.M.) during each 10-min period. Treatment
Shock level (mA )
Number of rats
Preconflict
Conflict
Postconflict
Saline Muscimol
0.4 0.4
4 4
190 + 27 168 _+22
24 + 3 14 -+ 6
220 + 45 204 _+48
Saline Picrotoxin
0.1 0.1
4 4
212 + 42 187 + 19
182 + 26 167 + 38
179 + 24 212 +__34
85 In contrast, injection of muscimol elicited a dosedependent increase in responding during the high conflict period. This anti-conflict effect was statistically significant with the 10 ng dose of the G A B A agonist. Muscimol had no significant effect on responding in the pre- and post-conflict VI periods. A n experimental tracing demonstrating the selective effect of muscimol on responding in the conflict period in a single rat appears in Fig 3. W h e n the same high dose of muscimol (10 ng) was injected into the lateral hypothalamus, no anti-conflict effect was obtained (Table II). Instead, muscimol tended to decrease responding in all 3 periods when injected at this site although the effect was not significant. Effects o f microinjecting drugs in the 'low' conflict schedule Fig. 4 illustrates the number of responses in each component of the 'low' conflict schedule after microinjection of either saline vehicle or 1 of 3 different doses of picrotoxin bilaterally into the posterior hypothalamus. After injection of saline, there was no significant suppression of responding during the conflict period as compared to the pre- and post-conflict VI periods. However, at the lowest dose of picrotoxin tested, this
T
D
SALINE
1
PICROTOXINI
,OOl."
,n*.,
2°°F
,oo[-
,¢, .'>' 3oo~ ~
/
I
I
o
12t ,.
VI1
C
VI2
'
~
Vl 2
D. BMI 25 r1~
B. PICROTOXIN1 ng
22q
Ill lit
1
VI1
C
,
VI2
Fig. 5. Experimental tracings depicting cumulative responding in the low conflict schedule in the same rats after microinjection of saline (A and C) and picrotoxin 1 ng (B) or bicuculline methiodide (BMI) 25 ng (D) bilaterally into posterior hypothalamus. For details, see Fig. 3. agent produced a significant suppression of responding during the conflict period without affecting responding during either of the VI periods. Experimental tracings representing performance in the high conflict schedule after injection of saline and injection of picrotoxin 1 ng appear in Fig. 5. With higher doses of picrotoxin, a trend toward suppression of responding in the VI periods was also apparent and the 100 ng dose significantly suppressed responding in all 3 periods. Microinjection of BMI 25 ng into the posterior hypothalamus also produced a significant suppression of responding only during the low conflict period (Table III). The apparent trend for suppression of responding during the pre-conflict VI period (although not significant) was most evident immediately after microinjection of this G A B A antagonist since after B M I (but not after saline) significantly fewer responses occurred in the first 5 min of the VI period (mean = 56 + 6) than in the last 5 min (mean = 83 + 3; P < 0.01 by paired t-test). A n experimental tracing demonstrating the pro-conflict el-
qng (n:3l
TABLE III
2O0
lO11-
C. CONTROL
A. CONTROL
lng (n=5)
Fig. 4. Summary of mean number of responses in each component of the low conflict (0.1 mA shock) schedule after bilateral injection of either saline or picrotoxin 100 ng (top), 4 ng (middle), or 1 ng (bottom) into the posterior hypothalamus. Injection of picrotoxin 1 ng caused a significant suppression of responding during the conflict period without affecting the pre- and post-conflict VI periods. Higher doses of picrotoxin tended to suppress responding during the VI periods as well, and after 100 ng significantly suppression was seen in all 3 periods. *P < 0.05; **P < 0.01 (significant difference from saline by paired t-test).
Responding in the low conflict (0.1 mA shock) schedule after microinjection of saline, BMI 25 ng and strychnine 38 ng into the posterior hypothalamus Data are presented as number of lever presses (mean + S.E.M.) during each 10 min period. Treatment
Number of rats
Pre-conflict Conflict
Postconflict
Saline BM125 ng
3 3
244 + 40 139 + 7
192 + 54 137 + 44*
195 + 38 229 + 28
Saline 4 Strychnine 38 ng 4
201 + 11 256 + 55
175 + 18 265 + 50
238 + 37 318 + 48
* Significantly different from saline by paired t-test, p < 0.05.
86 fect of BMI appears in Fig. 5. Microinjection of strychnine into the posterior hypothalamus failed to reduce responding in any of the periods (Table 11I). As with muscimol injection in the high conflict schedule, microinjection of picrotoxin 1 ng into the lateral hypothalamus elicited no significant effect compared to saline control in the low conflict schedule (Table II). DISCUSSION The data presented above clearly show that microinjection of G A B A antagonists picrotoxin and BMI into the posterior hypothalamus causes a pro-conflict response while injecting the GABA agonist muscimol elicits an anti-conflict effect. Microinjection of picrotoxin and muscimol into the lateral hypothalamus failed to produce significant changes at doses that had clear effects when injected into the posterior hypothalamus. This finding suggests that the anti-conflict effect of muscimol and the pro-conflict effect of the GABA antagonists is specific to the posterior hypothalamus. Our previous studies have shown that the locomotor stimulation and the selective enhancement of avoidance behavior caused by injecting GABA antagonists were also specific to the posterior hypothalamus and not elicited at the lateral hypothalamus T M and that the cardiostimulatory effects of injecting GABA antagonists were lost by moving the injection cannula away from the posterior hypothalamus 5. Taken together, all these findings indicate the presence in the posterior hypothalamus of a neuronal system under GABAergic inhibition that coordinates a 'fight-or-flight' reaction consisting of cardiorespiratory arousal, locomotor stimulation, selective enhancement of aversive responding and an increase in 'anxiety'. The pro-conflict effects of picrotoxin were dosedependent and specific at the lowest dose tested, so that 1 ng suppressed responding in the conflict period without any apparent effect on responding in the VI periods. In contrast, microinjection of the highest dose (100 ng) significantly suppressed responding in all components of the protocol. Even the 4 ng dose appeared to suppress responding in the pre- and post-conflict periods (although not significantly). This generalization of suppression of responding to the VI periods with the higher doses of 'anxiogenic' drugs than those that produce a specific pro-conflict effect has been reported in other studies as well 12'14'19. Microinjection of BMI 25 ng also suppressed responding (although not significantly) in the pre-conflict period to about 50% of the saline-treated group. However, closer examination revealed that most of the apparent suppression resulted from a reduction of responding immediately after injection of BMI, since rats
responded significantly less in the first 5 min of the VI~ period than in the second 5 min. In contrast, the response rates during both the 5 min periods were not significantly different after saline treatment. This finding suggests that the majority of suppression of responding during the pre-conflict session takes place in the first 5 min after microinjection of BM1, which we have shown previously corresponds to a period of maximal generalized locomotor stimulation and 'escape' behavior 17. Thus, the ability of the rats to respond effectively may have been disrupted immediately after injection of the GABA antagonist. Although only post-synaptic GABA A antagonists and an agonist were employed in this study, we have also shown that microinjection of inhibitors of GABA synthesis causes cardiovascular and locomotor stimulation similar to that elicited by GABA blockade at the same site in the posterior hypothalamus 5'~v. The time course of these effects was much more prolonged with the synthesis inhibitors, consistent with their mechanism of action. However, the responses caused by pre-synaptic and post-synaptic GABA blockade were qualitatively similar, implicating GABA as the specific neurotransmitter involved in the effects of these drugs. Microinjection of strychnine, a central nervous system system stimulant and a presumed glycine antagonist, failed to produce a pro-conflict effect in the present study or to cause physiological and behavioral stimulation in the previous studies. In fact, strychnine tended to increase responding in all 3 periods of the conflict schedule (Table III). Because of the small number of animals employed, an effect of strychnine on conflict behavior cannot be completely ruled out on the basis of these findings. However, the fact that CNS stimulants that are GABA antagonists suppress conflict responding while strychnine increases responding non-specifically further suggests that the effects of picrotoxin, BMI and muscimol are specifically related to their effects on GABAergic mechanisms. The conflict paradigm has been used as a test for measuring experimental anxiety in animals and has been empirically validated to have a predictive value with most drugs that are anxiolytic and anxiety provoking in humans 2"7. Systemic injections of picrotoxin and bicuculline have been shown to cause a pro-conflict effect that is reversed by diazepam 3. The particular conflict schedule used in this study has been previously shown to be a sensitive measure of both increases and decreases in the level of 'anxiety '19. However, whether anti-conflict effects are in fact obtainable and whether anti-conflict effects indicate an anxiolytic action or a reduction in tolerance to reward delay are issues of current controversy 2°. The debate is further compounded by the fact
87 that the conflict tests used by various investigators differ from one another in many key conditions that may account for conflicting results and conclusions. While more research is clearly needed to resolve these issues, the conflict test appears at least to measure the effects of drugs acting at the GABA-benzodiazepine receptor complex in a manner consistent with the anxiolytic or anxiogenic effects of these same agents in humans. Even though the effects of microinjecting GABAergic drugs into the posterior hypothalamus on conflict responding suggests changes in the level of 'anxiety', other explanations for these effects are possible. Another possibility is that the physiological changes elicited by microinjection of GABAergic drugs into the posterior hypothalamus may alter the perception by the animal of an internal aversion-inducing system. Microinjection of G A B A antagonists into the posterior hypothalamus increases while injection of muscimol decreases heart rate and blood pressure in conscious rats 21. Increases in blood pressure caused by peripheral injection of vasopressin have been shown to induce internal cues responsible for conditioned taste aversion 1. In a similar fashion, the physiological changes caused by picrotoxin, BMI and muscimol could alter the animal's perception of the conflict level. It is clear from the above results that disinhibiting neurons in the posterior hypothalamus produces proconflict effects that are commonly equated with an increase in the level of 'anxiety'. While such an equation may be perilous, some data support the view that the posterior hypothalamus is involved in regulating the level of anxiety in man. Clinically, posterior hypothalamotomies have been reported to cause a decrease in anxiety such that the procedure was termed 'sedative surgery '1°.
Schvarcz 16 has reported data from stereotaxic posterior hypothalamotomies. In 33 patients in whom the posterior hypothalamus was electrically stimulated under local anesthesia prior to lesioning, this stimulation consistently elevated the heart rates of the subjects. This increase in heart rate was, in fact, used to verify the appropriate posterior hypothalamic site for subsequent lesioning, a procedure similar to the physiological localization of the posterior hypothalamus employed in the present study. In addition to the physiological arousal, these subjects also reported intense fear and anxiety during stimulation. This striking similarity between the effects of electrical stimulation of the human posterior hypothalamus and disinhibition of the same area with G A B A antagonists in rats further supports the hypothesis that the effects of GABAergic drugs on the conflict schedule observed in this study may reflect changes in the level of 'anxiety'. In summary, microinjection of G A B A A antagonists, picrotoxin and BMI, and the G A B A A agonist, muscimol, into the cardiostimulatory region of the posterior hypothalamus of rats elicits dose-dependent pro-conflict and anti-conflict effects, respectively. These effects were not obtained when GABAergic drugs were injected into the lateral hypothalamus and microinjection of strychnine, a glycinergic CNS stimulant into the posterior hypothalamus failed to cause any significant effect. The results suggest that GABAergic inhibition controls the activity of a population of neurons in the posterior hypothalamus of rats that are capable of regulating the degree of physiological and behavioral arousal as well as the level of 'anxiety' in this species.
REFERENCES
7 Geller, I., Relative potencies of benzodiazepines as measured by their effects on conflict behavior, Arch. Int. Pharmacodyn., 149 (1964) 243-247. 8 Hess, W.R., Hypothalamus and Thalamus, Thieme, Stuttgart, 1956. 9 Hilton, S.M., The defense arousal system and its relevance for circulatory and respiratory control, J. Exp. Biol., 100 (1982) 159-174. 10 Kuhar, M.J., Neuroanatomical substrates of anxiety, Trends Neurosci., 9 (1986) 307-311. 11 Lisa, M., Marmo, E., Wible, J.H., Jr. and DiMicco, J.A., Injection of muscimol into posterior hypothalamus blocks stress-induced tachycardia, Am. J. Physiol., 257 (1989) R246R251. 12 Olds, M.E. and Olds, J., Approach-avoidance analysis of the rat diencephalon, J. Comp. Neurol., 120 (1963) 259-265. 13 Pellegrino, L.J., PeUegrino, A.S. and Cushman, A.J., A Stereotaxic Atlas of the Rat Brain, 2nd edn., Plenum, New York, 1979. 14 Quintero, S., Henney, S., Lawson, P., Mellanby, J. and Gray, J.A., The effects of compounds related to gamma-aminobutyrate and benzodiazepine receptors on behavioral responses to anxiogenic stimuli in the rat: punished barpressing, Psychophar-
1 Bluthe, R.M., Dantzer, R. and Le Moal, M., Peripheral injections of vasopressin control behavior by way of interoceptive signals for hypertension, Behav. Brain Res., 18 (1985) 31-39. 2 Corda, M.G., Blaker, W.D., Mendelson, W.B., Guidotti, A. and Costa, E., Beta-carbolines enhance shock-induced suppression of drinking in rats, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 2072-2076. 3 Corda, M.G. and Biggio, G., Proconflict effect of GABA receptor complex antagonists: reversal by diazepam, Neuropharmacology, 25 (1986) 541-544. 4 DiMicco, J.A., Abshire, V.M., Hankins, K.D., Sample, R.H.B. and Wible, J.H., Jr., Microinjection of GABA antagonists into posterior hypothalamus elevates heart rates in anesthetized rats, Neuropharmacology, 25 (1986) 1063-1066. 5 DiMicco, J.A. and Abshire, V.M., Evidence for GABAergic inhibition of a hypothalamic sympathoexcitatory mechanism in anesthetized rats, Brain Research, 402 (1987) 1-10. 6 DiScala, G., Schmitt, P. and Karli, P., Flight induced by infusion of bicuculline methiodide into periventricular structures, Brain Research, 309 (1984) 199-208.
Acknowledgements. This study was supported by USPHS Grant NS 19883 and MH 17107. We thank Robert Plass, H.F. Lemberger and Joseph Szwed for technical assistance.
88
macology, 85 (1985) 244-251. 15 Schenberg, L.C., deAguiar, J.C. and Graeff, EG., GABA modulation of the defense reaction induced by electrical stimulation, Physiol. Behav., 31 (1983) 429-437. 16 Schvarcz, J.R., Results of stimulation and destruction of the posterior hypothalamus: a long term evaluation. In Sweet, W.H., Obrador, S. and Martin-Rodriguez, J.G. (Eds.), Neurosurgical Treatment in Psychiatry, Pain and Epilepsy, University Park Press, Baltimore, 1977, pp. 429-438. 17 Shekhar, A. and DiMicco, J.A., Defense reaction elicited by injection of GABA antagonists and synthesis inhibitors into the posterior hypothalamus in rats. Neuropharmacology, 26 (1987) 407-417. 18 Shekhar. A., Hingtgen, J.N. and DiMicco, J.A., Selective
enhancement of shock avoidance resonding elicited by GABA blockade in the posterior hypothalamus of rats, Brain Research, 420 (1987) 118-128. 19 Shekhar, A., Hingtgen, J.N. and DiMicco, J.A., Anxiogenic effects of noreleagnine, a water soluble beta-carboline in rats, Neuropharmacology, 28 (1989) 539-542. 20 Thiebot, M.H., Soubrie, E and Simon, P., ls delay of reward mediated by shock-avoidance behavior a critical target for anti-punishment effects of diazepam in rats'?, Psychopharmacology, 87 (1985) 473-479. 21 Wible, J.H., Jr., Luft, EC. and DiMicco, J.A., Hypothalamic GABA suppresses sympathetic outflow to the cardiovascular system, Am. J. Physiol., 254 (1988) R680-R687.
81
BRES 15292
GABA receptors in the posterior hypothalamus regulate experimental anxiety in rats A. Shekhar, J.N. Hingtgen and J.A. DiMicco Departments of Pharmacology, and Toxicology and Psychiatry, and Program in Medical Neurobiology, Indiana University School of Medicine, Indianapolis, IN 46202 (U.S.A.) (Accepted 15 August 1989) Key words: y-Aminobutyric acid; Hypothalamus; Anxiety; Conflict; Muscimol; Picrotoxin; Bicuculline
Blockade of y-aminobutyric acid (GABA) function in the posterior hypothalamus of rats elicits a pattern of physiological and behavioral arousal consisting of increases in heart rate, respiration and blood pressure as well as intense locomotor stimulation and a selective enhancement of avoidance responding. The present study was conducted to assess the possibility that GABA-mediated neurotransmission in the posterior hypothalamus of the rat may regulate anxiety. Male rats were trained in a 'conflict' schedule consisting of a high and a low intensity of punishment ('high' and 'low' conflict) capable of measuring decreases and increases in the level of 'anxiety', respectively. Guide cannulae were stereotaxicaUy implanted bilaterally in the posterior hypothalamus of these rats at sites where microinjection of bicuculline methiodide (BMI) 25 ng caused increases in heart rate under anesthesia. After recovery, they were tested: (1) in the high conflict schedule after microinjection of saline and two doses of the GABAA receptor agonist muscimol; and (2) in the low conflict schedule after injecting saline, the GABAA receptor antagonists, BMI and picrotoxin, and the glycine antagonist, strychnine. Injection of muscimol caused a significant and selective anti-conflict effect while both BMI and, at appropriate doses, picrotoxin produced pro-conflict effects. Microinjection of strychnine into the posterior hypothalamus or muscimol and picrotoxin into the lateral hypothalamus did not influence conflict responding. These results suggest that endogenous GABA acts on GABAA receptors in a discrete area of the posterior hypothalamus to regulate the level of experimental anxiety in rats.
INTRODUCTION The hypothalamus is thought to play an important role in regulating the physiological and behavioral responses associated with emotional arousal. Electrical stimulation of specific hypothalamic nuclei elicits increases in heart rate, blood pressure and respiration 9 as well as behavioral changes suggesting a 'fight-or-flight' reaction 8. Electrical stimulation of the posterior hypothalamus causes a behavioral response that is predominantly 'aversive' in nature ~2. The integrity of the hypothalamus appears to be essential for emotional expression, and electrolytic lesions of the posterior hypothalamus in humans causes a loss of emotional excitability and a decrease in anxiety 1°. However, findings from electrical stimulation or ablation studies are subject to considerable interpretational ambiguity because of potential effects resulting from damage or excitation of fibers of passage as well as from destruction or stimulation of cell bodies. Furthermore, these techniques alone fail to provide any insight into the identity of the neurotransmitters involved. In recent pharmacological studies, the inhibitory neurotransmitter
~-aminobutyrate ( G A B A ) has been repeatedly implicated in the regulation of emotional expression and behavior. For example, the intensity of the 'defense' reaction elicited by electrical stimulation of the hypothalamus is decreased by systemic administration of benzodiazepines, which are thought to act through a G A B A ergic mechanism 15, and blockade of G A B A e r g i c inhibition in the hypothalamus elicits a reaction resembling escape or flight in freely moving rats 6'17. Recent studies in our laboratory have shown that microinjection of drugs that impair G A B A - m e d i a t e d neurotransmission into a discrete region of the posterior hypothalamus produces increased heart rate and respiration, and a smaller increase in blood pressure in anesthetized 4'5 as well as in conscious and freely moving rats 21. Further, G A B A blockade in the same cardiostimulatory area of the posterior hypothalamus causes 'escape'-oriented locomotor stimulation in conscious rats ~7 and selectively enhances aversively-motivated responding while having no significant effect on 'approach', or positively-reinforced, responding 18. All of these findings suggest that a G A B A e r g i c mechanism may tonically
Correspondence: J.A. DiMicco, Department of Pharmacology and Toxicology, Indiana University School of Medicine, 635 Barnhill Drive, Indianapolis, IN 46202, U.S.A. 0006-8993/90/$03.50 (~) 1990 Elsevier Science Publishers B.V. (Biomedical Division)
82 inhibit a p o p u l a t i o n of n e u r o n s in the p o s t e r i o r hypot h a l a m u s w h o s e activation is responsible for the physiological and b e h a v i o r a l r e s p o n s e s associated with states of emotional
arousal
such
as m i g h t be associated with
e x p e r i m e n t a l ' a n x i e t y ' and stress. T h e p r e s e n t study was u n d e r t a k e n to test the h y p o t h esis that G A B A e r g i c n e u r o t r a n s m i s s i o n in the p o s t e r i o r h y p o t h a l a m u s r e g u l a t e s the level of ' a n x i e t y ' in conscious and
freely
moving
animals.
To
assess
experimental
' a n x i e t y ' , we e m p l o y e d a m o d i f i e d 'conflict' p a r a d i g m in which the conflict is g e n e r a t e d positively r e i n f o r c e d demonstrated
behavior.
by p u n i s h m e n t
of a
P r e v i o u s studies
have
increases and d e c r e a s e s in the level of
' a n x i e t y ' in the p a r t i c u l a r conflict p r o c e d u r e e m p l o y e d in this study with drugs that are t h o u g h t to be anxiolytic ( b e n z o d i a z e p i n e s ) and a n x i e t y - p r o v o k i n g (fl-carbolines) in h u m a n s t9. R a t s t r a i n e d in a v a r i a b l e interval a p p r o a c h s c h e d u l e to o b t a i n s w e e t e n e d milk w e r e i n t r o d u c e d to a ' c o n f l i c t ' p a r a d i g m with two p u n i s h m e n t levels: a 'high c o n f l i c t ' p e r i o d , c a p a b l e of d e t e c t i n g 'anxiolytic' activity and
a ' l o w conflict' p e r i o d ,
subthreshold pression,
for g e n e r a t i n g
capable
using a shock
intensity
punishment-induced
of d e m o n s t r a t i n g
increases
sup-
in the
' a n x i e t y ' level of the animal. T h e p e r f o r m a n c e of these rats was assessed a f t e r m i c r o i n j e c t i n g G A B A e r g i c drugs into t h e p o s t e r i o r h y p o t h a l a m u s t h r o u g h c a n n u l a e imp l a n t e d at the c a r d i o s t i m u l a t o r y region. T h e effects of injecting
G A B A A antagonists,
picrotoxin
(PIC)
and
bicuculline m e t h i o d i d e ( B M I ) , w e r e t e s t e d in the low conflict s c h e d u l e while the effect of the G A B A A agonist, muscimol,
in the
high conflict schedule.
Strychnine, a non-GABAergic
was
tested
C N S stimulant, was in-
j e c t e d at the s a m e site to d e t e r m i n e p h a r m a c o l o g i c a l specificity of t h e effects while a n a t o m i c a l specificity was e s t a b l i s h e d by p e r f o r m i n g a parallel series of e x p e r i m e n t s with rats that had c a n n u l a e
implanted
in the lateral
h y p o t h a l a m i c area.
MATERIALS AND METHODS
Behavioral training Experimentally naive approximately 6-month-old male SpragueDawley rats (225-300 g) maintained at 80% free-feeding weights were trained in a variable interval (VII) approach schedule TM. The conditioning apparatus consisted of lever-pressing chambers made of 3 mm plastic with the inner dimensions of 21 x 16 × 14 era. These boxes had a lever and a dipper feeding device mounted on the front of each box. A noise-generating device was attached to the back of the box and the entire unit placed in a sound-insulated outer compartment. On the VI 1 schedule lever presses delivered 0.2 ml of sweetened milk on the dipper at an average rate of once per minute. Each rat was given one VI session (120 rain) each day for 5 days each week. Cumulative recorders, running time meters and counters were used to record the responding in each session. After stable VI responding was established, the conflict protocol was introduced. The protocol consisted of 10 rain of initial VI period (pre-confliet) which was followed by the conflict period signalled by
a constant white noise. During the 10 min conflict period, the dipper delivered an electric shock along with the milk on a VI schedule. All rats were trained in both 'high' and 'low' conflict periods distinguished by reinforcement-coupled punishment at two levels of intensity. During high conflict the dipper delivered a 0.4 mA shock along with the milk and during low conflict the dipper delivered a 0.1 mA shock. The conflict period was followed by another 10 min of VI schedule (post-conflict). The high conflict period was signalled by a full intensity (-14 dB, where 0 d B = 1 roW) noise and the low conflict period was signalled by half intensity (-31 dB) noise. Rats were trained in the above paradigm until they discriminated between the two levels of conflict clearly such that the high conflict schedule suppressed responding during the conflict period to less than 10% of the pre-conflict VI period while the minimal punishment delivered during the low conflict period had no significant effect on responding. These high and low punishment levels were previously shown to be effective in detecting the 'anxiolytic' and 'anxiogenic' effects of drug treatments, respectively v~
Implantation of microinjection cannulae at the cardiostimulatory site in the posterior hypothalamus Cannulae were chronically implanted at the active site in the posterior hypothalamus using a technique routinely employed in this laboratory 1a'17'21. Animals were anesthetized with pentobarbital (50 mg/kg) and fixed into a stereotaxic apparatus. Heart rate was monitored throughout the surgery by a cardiotachometer triggered by lead II of the electrocardiogram and recorded on a Beckman R511 Dynograph. Intermittent heating with an infrared lamp was used to maintain the body temperature between 36 and 38 °C. Two stainless steel guide cannulae (22-gauge, 10 mm length) were mounted on the stereotaxic apparatus on either side with the internal cannulae (28-gauge, 12 mm length) fixed in place within them. The internal cannulae were connected to 10 #1 Hamilton microinjection syringes by polyethylene tubing (PE 50) and the system was filled with a solution of BMI (25 ng//d). The microinjection syringes were placed on infusion pumps set to deliver 1 ~1 of injectate over 2 min through the internal cannulae. The internal cannula along with the guide cannula was lowered into the posterior hypothalamus on one side at an angle of 10° with respect to the vertical plane, the target coordinates with respect to bregma being 1.2 mm posterior, 0.5 mm lateral and 8.5-9.0 mm ventral according to the atlas of Pellegrino et al) 3. Bicuculline methiodide 25 ng in 1 /A was injected at this site and the resultant heart rate response was noted. An increase in heart rate of at least 50 beats/rain was the criterion used to assure that the cannulae were placed at physiologically active sites. If microinjection at these coordinates failed to elicit the expected heart rate response, the same process was repeated after repositioning the cannula 0.2 mm anteriorly, medially, posteriorly or laterally with respect to the original coordinates until an active site was found. After the heart rate had returned to baseline, the same procedure was repeated on the opposite side. The target coordinates used to implant the cannulae in the lateral hypothalamic area were 1.2 mm posterior, 2.5 mm lateral and 9 mm ventral with respect to bregma. After placing both cannulae at the physiologically active sites in the posterior hypothalamus or the inactive sites at the lateral hypothalamus, the guide cannulae were fixed at these positions with the help of 3 stainless steel screws anchored to the skull and cranioplastic cement (cannulae and cement obtained from Plastic Products Co., Roanoke, VA). The internal cannulae were then removed and the guide cannulae were sealed with steel wire dummy cannulae. The rats were allowed to recover in individual plastic cages with ad lib water and approximately 20 g of food per day.
Experimental protocol After at least 3 days of recovery, each of the rats was given additional training sessions in high and low conflict schedules daily (5 days/week). Within about one week, they usually attained their baseline rates of responding during all the 3 periods of the conflict schedule after which the following experimental protocol was begun.
83 TABLE I
Changes in heart rate elicited by unilateral injection of BM125 ng into different regions of the hypothalamus in anesthetized rats PH, posterior hypothalamus; LH, lateral hypothalamus,
Site of. in1.
Num ber of rats
Right side
Left side
Baseline heart rate (bpm)
Maximum change in HR (bpm)
Time to peak (rain)
Duration of effect (rain)
Baseline heart rate (bpm)
Maximum change in HR (bpm)
Time to peak (rain)
Duration of effect (rain)
PH LH
11 4
292 + 9 298 + 15
+91 _+9* - 2 + 15
9+ 1 -
22 + 2 -
284 + 7 303 + 11
+95 + 12' +5 _+ 12
11 + 1 -
26 + 3 -
* Significant change from baseline heart rate by paired t-test, P < 0.05.
Prior to placing each of the rats in the conflict box, microinjection cannulae were inserted into the guide cannulae and 1/A of saline or a GABAergic drug solution selected randomly was injected bilaterally into the hypothalamus over a 2 min period. The drugs (from Sigma, St. Louis, MO) used were BMI (25 ng//~l), picrotoxin (1, 4 or 100 ng//A) and muscimol (5 or 10 ng//A). One minute after completion of the injection, the microinjection cannulae were removed, the rat was placed in the conflict box and responding was recorded in either the low conflict schedule (after saline, BMI and picrotoxin injections) or the high conflict schedule (after saline and muscimol injections). After an interval of at least two days, a different reagent was injected and its effect on the animal's responding was recorded. No rat received more than 4 drug injections. Strychnine (38 ng//~l, a dose equimolar to BMI 25 ng//~l) was injected into the posterior hypothalamus of 4 rats and its effect
on their responding in the low conflict schedule was noted. In another group of 4 rats with cannulae implanted in the lateral hypothalamic area, the effects of bilateral injection of muscimol 10 ng and picrotoxin 1 ng were assessed in the high and low conflict schedules, respectively. The results of the behavioral changes were tabulated and expressed as mean +_ S.E.M. Statistical analysis was performed by paired t-test and the criterion for statistical significance was P < 0.05. After completion of the experiments, the animals were injected intraperitoneally with a lethal dose of pentobarbital. The brains were then removed and fixed in 4% formaldehyde after the microinjection sites were marked by injecting 0.25/.,1 of 1 N HCI. Brains were later sectioned into 5 0 / t m slices, stained with Cresyl violet and the exact site of injection was then determined by comparing the sections with the atlas of Pellegrino et al. 13.
'-1.0
"1.0
D1.2
-.1.2
E1.4
- - 1.4
Fig. 1. Schematic representation of the bilateral hypothalamic injection sites as confirmed by histology. The numbers represent injection sites in 14 of the 15 rats used in this study. Rats, nos. 1-10 had cannulae implanted bilaterally in the posterior hypothalamus; nos. 11-14 had the cannulae implanted bilaterally in the lateral hypothalamus. The distances of each section in mm posterior to bregma are indicated (adapted from Pellegrino et a1.13). FX, fornix; MT, mamillothalamic tract; PH, posterior hypothalamus; V, ventricle.
84
400
- KIs'L'NE I
A. CONTROL
296~
BMUSCIMOL
300
200
VI 1
(~
VI 2
100 0 --
5ng
B. MUSCIMOL 5 ng
,.~
(n=3)
q00 -56 - r - r - ~
300 2O0
VI 1 C C. MUSClMOL 10 ng
100 O
VI 2 /
lOng
Fig. 2. Summary of mean number of responses in each component of the high conflict (0.4 mA shock) schedule after bilateral injection of either saline or muscimol 5 ng (above) or 10 ng (below) into the posterior hypothalamus. Compared to control (saline), muscimol caused dose-related increases in responding only during the conflict period. * Significantly different from saline by paired t-test, P < 0.05.
RESULTS Increases in heart rate of at least 50 beats/min were o b t a i n e d by injecting 25 ng of B M I in 1 /A into the p o s t e r i o r h y p o t h a l a m u s on each side in 11 rats. The increases in heart rate, which began to develop during the injection of B M I , are given along with their time course in Table I. Microinjection of B M I 25 ng into the lateral h y p o t h a l a m u s failed to elicit any significant change in heart rate in 4 rats (Table I). The sites of injection on both sides as confirmed by histology in 14 of the 15 animals used in this study were in or immediately adjacent to the posterior hypothalamic nucleus (Fig. 1). O n e rat in which the cannulae had been i m p l a n t e d in the physiologically active sites died suddenly for unknown reasons before the site could be m a r k e d and the brain fixed. In the remaining 10 rats with the cannulae i m p l a n t e d in the cardiostimulatory area, the site of implantation was confirmed to be the posterior h y p o t h a l a m u s in or immediately adjacent to the posterior hypothalamic nucleus according to Pellegrino et al. 13. The sites of implantation in the 4 rats that had the cannulae positioned in the inactive sites were in the lateral h y p o t h a l a m u s (Fig. 1). Effect o f microinjecting drugs in the 'high' conflict schedule Total n u m b e r of responses in each c o m p o n e n t of the
Y
VI I
C
VI 2
Fig. 3. Experimental tracings depicting cumulative responding in the high conflict schedule in the same rat after microinjection of
saline (A) and muscimol 5 ng (B) or 10 ng (C) bilaterally into posterior hypothalamus. Cumulative lever presses are indicated by upward deflection of the pen and delivery of reinforcement by diagonal tics. In each case, the record represents a 30-min session, comprised of a conflict period ('C'; heavy bar), during which a shock (0.4 mA) is delivered along with the reinforcement, and which is preceded and followed by 10-min periods of variable interval unpunished reinforcement (VI~ and Vie). Numbers indicate total number of responses in each component.
'high' conflict schedule after the microinjection of saline or either of two doses of muscimol (5 and 10 ng) into the posterior h y p o t h a l a m u s are r e p r e s e n t e d in Fig. 2. A f t e r microinjection of saline vehicle, responding r e m a i n e d almost completely suppressed during the conflict period.
TABLE I1 Responding in the high conflict (0.4 mA) schedule after microinjection of saline and muscimo110 ng, and in the low conflict (0.1 mA shock) schedule after microinjection of saline and picrotoxin 1 ng into the lateral hypothalamus Data are presented as number of lever presses (mean + S.E.M.) during each 10-min period. Treatment
Shock level (mA )
Number of rats
Preconflict
Conflict
Postconflict
Saline Muscimol
0.4 0.4
4 4
190 + 27 168 _+22
24 + 3 14 -+ 6
220 + 45 204 _+48
Saline Picrotoxin
0.1 0.1
4 4
212 + 42 187 + 19
182 + 26 167 + 38
179 + 24 212 +__34
85 In contrast, injection of muscimol elicited a dosedependent increase in responding during the high conflict period. This anti-conflict effect was statistically significant with the 10 ng dose of the G A B A agonist. Muscimol had no significant effect on responding in the pre- and post-conflict VI periods. A n experimental tracing demonstrating the selective effect of muscimol on responding in the conflict period in a single rat appears in Fig 3. W h e n the same high dose of muscimol (10 ng) was injected into the lateral hypothalamus, no anti-conflict effect was obtained (Table II). Instead, muscimol tended to decrease responding in all 3 periods when injected at this site although the effect was not significant. Effects o f microinjecting drugs in the 'low' conflict schedule Fig. 4 illustrates the number of responses in each component of the 'low' conflict schedule after microinjection of either saline vehicle or 1 of 3 different doses of picrotoxin bilaterally into the posterior hypothalamus. After injection of saline, there was no significant suppression of responding during the conflict period as compared to the pre- and post-conflict VI periods. However, at the lowest dose of picrotoxin tested, this
T
D
SALINE
1
PICROTOXINI
,OOl."
,n*.,
2°°F
,oo[-
,¢, .'>' 3oo~ ~
/
I
I
o
12t ,.
VI1
C
VI2
'
~
Vl 2
D. BMI 25 r1~
B. PICROTOXIN1 ng
22q
Ill lit
1
VI1
C
,
VI2
Fig. 5. Experimental tracings depicting cumulative responding in the low conflict schedule in the same rats after microinjection of saline (A and C) and picrotoxin 1 ng (B) or bicuculline methiodide (BMI) 25 ng (D) bilaterally into posterior hypothalamus. For details, see Fig. 3. agent produced a significant suppression of responding during the conflict period without affecting responding during either of the VI periods. Experimental tracings representing performance in the high conflict schedule after injection of saline and injection of picrotoxin 1 ng appear in Fig. 5. With higher doses of picrotoxin, a trend toward suppression of responding in the VI periods was also apparent and the 100 ng dose significantly suppressed responding in all 3 periods. Microinjection of BMI 25 ng into the posterior hypothalamus also produced a significant suppression of responding only during the low conflict period (Table III). The apparent trend for suppression of responding during the pre-conflict VI period (although not significant) was most evident immediately after microinjection of this G A B A antagonist since after B M I (but not after saline) significantly fewer responses occurred in the first 5 min of the VI period (mean = 56 + 6) than in the last 5 min (mean = 83 + 3; P < 0.01 by paired t-test). A n experimental tracing demonstrating the pro-conflict el-
qng (n:3l
TABLE III
2O0
lO11-
C. CONTROL
A. CONTROL
lng (n=5)
Fig. 4. Summary of mean number of responses in each component of the low conflict (0.1 mA shock) schedule after bilateral injection of either saline or picrotoxin 100 ng (top), 4 ng (middle), or 1 ng (bottom) into the posterior hypothalamus. Injection of picrotoxin 1 ng caused a significant suppression of responding during the conflict period without affecting the pre- and post-conflict VI periods. Higher doses of picrotoxin tended to suppress responding during the VI periods as well, and after 100 ng significantly suppression was seen in all 3 periods. *P < 0.05; **P < 0.01 (significant difference from saline by paired t-test).
Responding in the low conflict (0.1 mA shock) schedule after microinjection of saline, BMI 25 ng and strychnine 38 ng into the posterior hypothalamus Data are presented as number of lever presses (mean + S.E.M.) during each 10 min period. Treatment
Number of rats
Pre-conflict Conflict
Postconflict
Saline BM125 ng
3 3
244 + 40 139 + 7
192 + 54 137 + 44*
195 + 38 229 + 28
Saline 4 Strychnine 38 ng 4
201 + 11 256 + 55
175 + 18 265 + 50
238 + 37 318 + 48
* Significantly different from saline by paired t-test, p < 0.05.
86 fect of BMI appears in Fig. 5. Microinjection of strychnine into the posterior hypothalamus failed to reduce responding in any of the periods (Table 11I). As with muscimol injection in the high conflict schedule, microinjection of picrotoxin 1 ng into the lateral hypothalamus elicited no significant effect compared to saline control in the low conflict schedule (Table II). DISCUSSION The data presented above clearly show that microinjection of G A B A antagonists picrotoxin and BMI into the posterior hypothalamus causes a pro-conflict response while injecting the GABA agonist muscimol elicits an anti-conflict effect. Microinjection of picrotoxin and muscimol into the lateral hypothalamus failed to produce significant changes at doses that had clear effects when injected into the posterior hypothalamus. This finding suggests that the anti-conflict effect of muscimol and the pro-conflict effect of the GABA antagonists is specific to the posterior hypothalamus. Our previous studies have shown that the locomotor stimulation and the selective enhancement of avoidance behavior caused by injecting GABA antagonists were also specific to the posterior hypothalamus and not elicited at the lateral hypothalamus T M and that the cardiostimulatory effects of injecting GABA antagonists were lost by moving the injection cannula away from the posterior hypothalamus 5. Taken together, all these findings indicate the presence in the posterior hypothalamus of a neuronal system under GABAergic inhibition that coordinates a 'fight-or-flight' reaction consisting of cardiorespiratory arousal, locomotor stimulation, selective enhancement of aversive responding and an increase in 'anxiety'. The pro-conflict effects of picrotoxin were dosedependent and specific at the lowest dose tested, so that 1 ng suppressed responding in the conflict period without any apparent effect on responding in the VI periods. In contrast, microinjection of the highest dose (100 ng) significantly suppressed responding in all components of the protocol. Even the 4 ng dose appeared to suppress responding in the pre- and post-conflict periods (although not significantly). This generalization of suppression of responding to the VI periods with the higher doses of 'anxiogenic' drugs than those that produce a specific pro-conflict effect has been reported in other studies as well 12'14'19. Microinjection of BMI 25 ng also suppressed responding (although not significantly) in the pre-conflict period to about 50% of the saline-treated group. However, closer examination revealed that most of the apparent suppression resulted from a reduction of responding immediately after injection of BMI, since rats
responded significantly less in the first 5 min of the VI~ period than in the second 5 min. In contrast, the response rates during both the 5 min periods were not significantly different after saline treatment. This finding suggests that the majority of suppression of responding during the pre-conflict session takes place in the first 5 min after microinjection of BM1, which we have shown previously corresponds to a period of maximal generalized locomotor stimulation and 'escape' behavior 17. Thus, the ability of the rats to respond effectively may have been disrupted immediately after injection of the GABA antagonist. Although only post-synaptic GABA A antagonists and an agonist were employed in this study, we have also shown that microinjection of inhibitors of GABA synthesis causes cardiovascular and locomotor stimulation similar to that elicited by GABA blockade at the same site in the posterior hypothalamus 5'~v. The time course of these effects was much more prolonged with the synthesis inhibitors, consistent with their mechanism of action. However, the responses caused by pre-synaptic and post-synaptic GABA blockade were qualitatively similar, implicating GABA as the specific neurotransmitter involved in the effects of these drugs. Microinjection of strychnine, a central nervous system system stimulant and a presumed glycine antagonist, failed to produce a pro-conflict effect in the present study or to cause physiological and behavioral stimulation in the previous studies. In fact, strychnine tended to increase responding in all 3 periods of the conflict schedule (Table III). Because of the small number of animals employed, an effect of strychnine on conflict behavior cannot be completely ruled out on the basis of these findings. However, the fact that CNS stimulants that are GABA antagonists suppress conflict responding while strychnine increases responding non-specifically further suggests that the effects of picrotoxin, BMI and muscimol are specifically related to their effects on GABAergic mechanisms. The conflict paradigm has been used as a test for measuring experimental anxiety in animals and has been empirically validated to have a predictive value with most drugs that are anxiolytic and anxiety provoking in humans 2"7. Systemic injections of picrotoxin and bicuculline have been shown to cause a pro-conflict effect that is reversed by diazepam 3. The particular conflict schedule used in this study has been previously shown to be a sensitive measure of both increases and decreases in the level of 'anxiety '19. However, whether anti-conflict effects are in fact obtainable and whether anti-conflict effects indicate an anxiolytic action or a reduction in tolerance to reward delay are issues of current controversy 2°. The debate is further compounded by the fact
87 that the conflict tests used by various investigators differ from one another in many key conditions that may account for conflicting results and conclusions. While more research is clearly needed to resolve these issues, the conflict test appears at least to measure the effects of drugs acting at the GABA-benzodiazepine receptor complex in a manner consistent with the anxiolytic or anxiogenic effects of these same agents in humans. Even though the effects of microinjecting GABAergic drugs into the posterior hypothalamus on conflict responding suggests changes in the level of 'anxiety', other explanations for these effects are possible. Another possibility is that the physiological changes elicited by microinjection of GABAergic drugs into the posterior hypothalamus may alter the perception by the animal of an internal aversion-inducing system. Microinjection of G A B A antagonists into the posterior hypothalamus increases while injection of muscimol decreases heart rate and blood pressure in conscious rats 21. Increases in blood pressure caused by peripheral injection of vasopressin have been shown to induce internal cues responsible for conditioned taste aversion 1. In a similar fashion, the physiological changes caused by picrotoxin, BMI and muscimol could alter the animal's perception of the conflict level. It is clear from the above results that disinhibiting neurons in the posterior hypothalamus produces proconflict effects that are commonly equated with an increase in the level of 'anxiety'. While such an equation may be perilous, some data support the view that the posterior hypothalamus is involved in regulating the level of anxiety in man. Clinically, posterior hypothalamotomies have been reported to cause a decrease in anxiety such that the procedure was termed 'sedative surgery '1°.
Schvarcz 16 has reported data from stereotaxic posterior hypothalamotomies. In 33 patients in whom the posterior hypothalamus was electrically stimulated under local anesthesia prior to lesioning, this stimulation consistently elevated the heart rates of the subjects. This increase in heart rate was, in fact, used to verify the appropriate posterior hypothalamic site for subsequent lesioning, a procedure similar to the physiological localization of the posterior hypothalamus employed in the present study. In addition to the physiological arousal, these subjects also reported intense fear and anxiety during stimulation. This striking similarity between the effects of electrical stimulation of the human posterior hypothalamus and disinhibition of the same area with G A B A antagonists in rats further supports the hypothesis that the effects of GABAergic drugs on the conflict schedule observed in this study may reflect changes in the level of 'anxiety'. In summary, microinjection of G A B A A antagonists, picrotoxin and BMI, and the G A B A A agonist, muscimol, into the cardiostimulatory region of the posterior hypothalamus of rats elicits dose-dependent pro-conflict and anti-conflict effects, respectively. These effects were not obtained when GABAergic drugs were injected into the lateral hypothalamus and microinjection of strychnine, a glycinergic CNS stimulant into the posterior hypothalamus failed to cause any significant effect. The results suggest that GABAergic inhibition controls the activity of a population of neurons in the posterior hypothalamus of rats that are capable of regulating the degree of physiological and behavioral arousal as well as the level of 'anxiety' in this species.
REFERENCES
7 Geller, I., Relative potencies of benzodiazepines as measured by their effects on conflict behavior, Arch. Int. Pharmacodyn., 149 (1964) 243-247. 8 Hess, W.R., Hypothalamus and Thalamus, Thieme, Stuttgart, 1956. 9 Hilton, S.M., The defense arousal system and its relevance for circulatory and respiratory control, J. Exp. Biol., 100 (1982) 159-174. 10 Kuhar, M.J., Neuroanatomical substrates of anxiety, Trends Neurosci., 9 (1986) 307-311. 11 Lisa, M., Marmo, E., Wible, J.H., Jr. and DiMicco, J.A., Injection of muscimol into posterior hypothalamus blocks stress-induced tachycardia, Am. J. Physiol., 257 (1989) R246R251. 12 Olds, M.E. and Olds, J., Approach-avoidance analysis of the rat diencephalon, J. Comp. Neurol., 120 (1963) 259-265. 13 Pellegrino, L.J., PeUegrino, A.S. and Cushman, A.J., A Stereotaxic Atlas of the Rat Brain, 2nd edn., Plenum, New York, 1979. 14 Quintero, S., Henney, S., Lawson, P., Mellanby, J. and Gray, J.A., The effects of compounds related to gamma-aminobutyrate and benzodiazepine receptors on behavioral responses to anxiogenic stimuli in the rat: punished barpressing, Psychophar-
1 Bluthe, R.M., Dantzer, R. and Le Moal, M., Peripheral injections of vasopressin control behavior by way of interoceptive signals for hypertension, Behav. Brain Res., 18 (1985) 31-39. 2 Corda, M.G., Blaker, W.D., Mendelson, W.B., Guidotti, A. and Costa, E., Beta-carbolines enhance shock-induced suppression of drinking in rats, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 2072-2076. 3 Corda, M.G. and Biggio, G., Proconflict effect of GABA receptor complex antagonists: reversal by diazepam, Neuropharmacology, 25 (1986) 541-544. 4 DiMicco, J.A., Abshire, V.M., Hankins, K.D., Sample, R.H.B. and Wible, J.H., Jr., Microinjection of GABA antagonists into posterior hypothalamus elevates heart rates in anesthetized rats, Neuropharmacology, 25 (1986) 1063-1066. 5 DiMicco, J.A. and Abshire, V.M., Evidence for GABAergic inhibition of a hypothalamic sympathoexcitatory mechanism in anesthetized rats, Brain Research, 402 (1987) 1-10. 6 DiScala, G., Schmitt, P. and Karli, P., Flight induced by infusion of bicuculline methiodide into periventricular structures, Brain Research, 309 (1984) 199-208.
Acknowledgements. This study was supported by USPHS Grant NS 19883 and MH 17107. We thank Robert Plass, H.F. Lemberger and Joseph Szwed for technical assistance.
88
macology, 85 (1985) 244-251. 15 Schenberg, L.C., deAguiar, J.C. and Graeff, EG., GABA modulation of the defense reaction induced by electrical stimulation, Physiol. Behav., 31 (1983) 429-437. 16 Schvarcz, J.R., Results of stimulation and destruction of the posterior hypothalamus: a long term evaluation. In Sweet, W.H., Obrador, S. and Martin-Rodriguez, J.G. (Eds.), Neurosurgical Treatment in Psychiatry, Pain and Epilepsy, University Park Press, Baltimore, 1977, pp. 429-438. 17 Shekhar, A. and DiMicco, J.A., Defense reaction elicited by injection of GABA antagonists and synthesis inhibitors into the posterior hypothalamus in rats. Neuropharmacology, 26 (1987) 407-417. 18 Shekhar. A., Hingtgen, J.N. and DiMicco, J.A., Selective
enhancement of shock avoidance resonding elicited by GABA blockade in the posterior hypothalamus of rats, Brain Research, 420 (1987) 118-128. 19 Shekhar, A., Hingtgen, J.N. and DiMicco, J.A., Anxiogenic effects of noreleagnine, a water soluble beta-carboline in rats, Neuropharmacology, 28 (1989) 539-542. 20 Thiebot, M.H., Soubrie, E and Simon, P., ls delay of reward mediated by shock-avoidance behavior a critical target for anti-punishment effects of diazepam in rats'?, Psychopharmacology, 87 (1985) 473-479. 21 Wible, J.H., Jr., Luft, EC. and DiMicco, J.A., Hypothalamic GABA suppresses sympathetic outflow to the cardiovascular system, Am. J. Physiol., 254 (1988) R680-R687.
