- Email: [email protected]
Enhancement of electrically elicited startle by amygdaloid stimulation
Physiology& Behavior, Vol. 48, pp. 343-349. ©Pergamon Press plc, 1990. Printed in the U.S.A.
0031-9384/90 $3.00 + .00
Enhancement of Electrically Elicited Startle by Amygdaloid Stimulation J E F F R E Y B. R O S E N 1 A N D M I C H A E L D A V I S
Department of Psychiatry, Yale University, New Haven, CT 06508 R e c e i v e d 9 January 1990
ROSEN, J. B. AND M. DAVIS. Enhancement of electrically elicited startle by amygdaloid stimulation. PHYSIOL BEHAV 48(2) 343-349, 1990.--Previous experiments showed that acoustic startle amplitude can be enhanced by electrical stimulation of the amygdala. Because the acoustic startle pathway is organized in a serial fashion, startle can be elicited electrically with progressively shorter latencies by stimulating different points along this pathway [i.e., ventral cochlear nucleus (VCN), paralemniscal zone (PLZ), nucleus reticularis pontis caudalis (RPC) or medial longitudinal fasciculus (MLF)]. The present study evaluated the temporal characteristics of the facilitatory effect of amygdaloid stimulation on startle elicited electrically from different points along the acoustic startle pathway. A single 0.1-msec pulse was delivered to the central nucleus of the amygdala at various times before the onset of a 1-msec pulse in various sites in the startle pathway. The shortest amygdaloid stimulation-startle onset interval to significantly enhance startle was 0 msec for the VCN, 2 msec for the PLZ, 3 msec for the RPC and 7 msec for the MLF. These results indicate that amygdaloid stimulation enhances electrically elicited startle in a temporal manner that is complementary to facilitation of acoustic startle. The similarity of amygdala-stimulated enhancement and fear potentiation of electrically elicited startle is also discussed. Startle
Fear-potentiated startle
Amygdala
Electrical brain stimulation
Fear
Anatomical data indicate that the central nucleus projects directly to the area of the R I ~ known to be critical for startle (23). Lesions at various points along this pathway block startle potentiated by both conditioned (12) and unconditioned fear such as footshock (13). Taken together, these data suggest that conditioned fear is mediated by activation of the amygdala which then modulates startle via its direct pathway to the RPC. Consistent with this interpretation, startle-like responses elicited from either the VCN or PLZ are increased in the presence of a conditioned fear stimulus (2) or shortly following footshock (3). In contrast, startle-like responses elicited by electrical stimulation of the RPC or the pathway from the RPC to the spinal cord are not (2,3). Thus, startle-like responses elicited by stimulation of the startle circuit afferent to the RPC are affected by events thought to activate the amygdala whereas startle-like responses elicited by stimulation of the circuit downstream from the RPC are not. If this is the case, then electrical stimulation of the amygdala should also increase startle-like responses elicited from the VCN and PLZ but not from stimulation beyond the RPC. Moreover, if maximal facilitation of startle occurs when impulses from the amygdala arrive at the RPC at the same time as impulses activated by the startle stimulus, then the optimal interval for enhancement between amygdaloid stimulation and acoustically elicited startle should shift by about 1 msec from that seen when startle is elicited electrically from the VCN and by 2-3 msec when startle is elicited electrically from the PLZ. Figure 1 shows the logic for these predictions schematically, which were tested in the present study.
ACTIVATION of the amygdala has been shown to play an important role in fear. For instance, electrical stimulation of the central nucleus of the amygdala can produce a pattern of behavioral and autonomic changes that resemble a state of fear (1, 8, 15, 16, 18, 21, 22, 25). Conversely, lesions of the amygdala block both conditioned and unconditioned fear (9-11, 13, 16, 17). Recently it was shown that lesions of the central nucleus of the amygdala block increases in the amplitude of acoustic startle when startle is elicited in the presence of a conditioned fear stimulus or following aversive stimulation such as footshock (10, 11, 13). Electrical stimulation of the amygdala also causes a marked increase in the amplitude of the acoustic startle reflex (21,22). Thus, it is possible that the increase in startle seen during electrical stimulation of the amygdala mimics facilitation of startle during a state of fear (7). The acoustic startle reflex has a very short latency [8 msec recorded electromyographically in the hindlegs (14)] and appears to be mediated by a serial pathway consisting of the ventral cochlear nucleus (VCN), an area just medial and ventral to the ventral nucleus of the lateral lemniscus [paralemniscal zone (PLZ)], an area in the nucleus reticularis pontis caudalis just dorsal to the superior olives (RPC) and motor neurons in the spinal cord (5,6). Electrical, single pulse stimulation within each of these nuclei produces startle-like responses which have a progressively shorter latency as the electrode is moved from the VCN (7 msec) to the PLZ (5--6 msec) to the RPC (4--5 msec) to the spinal cord (3 msec) (6).
tRequests for reprints should be addressed to Jeffrey B. Rosen, NIMH, Biological Psychiatry Branch, Building 10 Room 3N212, 9000 Rockville Pike, Bethesda, MD 20892.
343
344
ROSEN AND DAVIS
METHOD
STIM
,l
Subjects and Surgery A total of 35 male albino Sprague-Dawley rats (Charles River Co.) weighing 300--400 g were used. Rats were housed in group cages until surgery and individually thereafter. A 12:12 light-dark cycle was maintained in the animal room. All testing was done during the light cycle. Feeding was ad lib. All rats were anesthetized with 8% chloral hydrate (1 ml/200 g) and unipolar electrodes (00 insect pins, 0.25 mm tip diameter, insulated except for 0.5 mm of the tip) were stereotaxically aimed bilaterally at the medial portion of the central nucleus of the amygdala (2.8 mm posterior, 4.0 mm lateral and 8.5 mm ventral to bregma with the skull surface level). These same animals also were implanted with bilateral electrodes in various sites in the startle circuit. Eight were in the VCN (11.3 mm posterior, 3.8 mm lateral and 8.6 mm ventral to bregma with the skull surface level), 7 in the PLZ (8.3 mm posterior, 2.0 mm lateral and 9.6 mm ventral to bregma with the skull surface level), 12 in the RPC (10.0 or 10.5 mm posterior, 1.5 mm lateral and 9.5 mm ventral to bregma with the skull surface level) and 8 in the medial longitudinal fasciculus (MLF) (on the midline and 12.3 or 12.7 mm posterior and 9.6 mm ventral to bregma with the skull surface level) which is believed to carry the axons from the RPC to the spinal cord (6). A stainless steel screw fixed to the skull over the frontal cortex was used as an indifferent electrode. Several other stainless steel screws were used to anchor the electrode assembly to the skull. The electrodes were cemented in place and the leads from the electrodes were enclosed in a Viking (#TKR-7) socket and mounted on the skull with dental acrylic.
Apparatus A stabilimeter was used to record the amplitude of the startle responses. The stabilimeter consisted of a 12 x 20 x 18 cm Plexiglas and wire mesh cage suspended between compression springs within a steel frame. The cage had a small opening in the top to allow leads to enter the cage for electrical stimulation of the brain. The stabilimeter was housed in a lighted, ventilated soundattenuating chamber. The cage was located 10 cm from a high frequency speaker (Radio Shack Supertweeter). Cage movement resulted in displacement of an accelerometer in which the resultant voltage was proportional to the velocity of displacement. Startle amplitude was defined as the maximal accelerometer voltage that occurred during the first 200 msec after a startle stimulus was delivered. As described elsewhere, acoustically or electrically elicited startle has a latency of about 5 msec recorded electromyographically in the neck muscles, 8 msec in the hindleg (6,14). This very fast sequence of muscle contractions results in cage movement having a frequency of about 7-10 Hz in this particular test cage (4). By sampling over a 200-msec period after the onset of the startle stimulus, the peak amplitude of cage movement initiated by the very short latency startle response can be reliably recorded (4). Startle amplitude (range 0 to 88 units) was recorded by an analogue-to-digital converter interfaced to a printer. Background white noise, provided by a white noise generator, was 55 dB. Sound level measurements were made within the cage with a General Radio Model 1551-C sound level meter (A-scale). Electrical brain stimulation was delivered through separate constant current stimulators (Electronic Instruments, #BSI-1) for each electrode gated from a Grass $88 stimulator which also controlled the delivery of the acoustic startle stimulus. Stimulation current, measured across a differential amplifier, and the acoustic
STARTLE I~ RESPON~IE ENHANCI~
STIM
,J
STIM
STIM
STARTLE
FIG. 1. Schematic diagram of the logic for enhancement of electrically elicited startle by amygdaloid stimulation. (A) Because the VCN and PLZ are before the proposed point of modulation of startle by amygdaloid stimulation, startle elicited from these nuclei should be enhanced by amygdaloid stimulation. (B) Because the RPC and MLF are at or beyond the point of modulation, startle elicited from these sites should not be enhanced. The acoustic startle pathway [ear, VCN, PLZ, RPC, MLF, and SC (spinal cord)] is from Davis et al. (6). The transit times shown in A between the synapses of the startle circuit are derived from Davis et al. (6) and the present experiment. The transit time from the amygdala to the startle pathway is derived from Rosen and Davis (22) and the present study. startle stimulus, measured by a dB meter, were continuously monitored by a Tektronix 5130N oscilloscope. Procedure A week following surgery, rats were placed in the test cage, given 30-60 min for adaptation and tested for enhancement of acoustic startle by electrical stimulation of the amygdala through each electrode individually. The acoustic startle stimulus was a 20-msec burst of white noise having a rise-decay time of 5 msec. The intensity of the acoustic startle stimulus was adjusted (95-115 dB) to give a low to intermediate amplitude of response (20-30 units). To determine the amount of amygdala stimulation necessary to enhance acoustic startle, a single 0.1-msec cathodal square-wave pulse ranging between 40 and 400 I~A was delivered to the amygdala simultaneously with onset of the acoustic startle stimulus to increase startle by approximately 40 units over the response elicited by the startle stimulus alone. Simultaneous stimulation of the amygdala and onset of the acoustic stimulus was chosen because previous experiments demonstrated that this produced maximal enhancement of acoustic startle (21,22). The electrode with the lowest current to enhance acoustic startle was used throughout the remaining experiment. For the electrically elicited startle experiments, startle was elicited by a 1.0-msec cathodal square wave pulse through the bilateral electrodes in the VCN, PLZ, RPC or MLF. The current was adjusted (25-180 I~A) to give a startle amplitude similar to that elicited by the acoustic stimulus (20-30 units). Immediately following this, amygdaloid stimulation was delivered 1 msec after, simultaneously, or 1, 2, 3, 4, 5, 7, 10, or 20 msec before the onset of the startle eliciting electrical pulse in the startle circuit. Trials of electrically elicited startle without an accompanying amygdaloid stimulation were also given. Ten occurrences of each of these stimulus conditions (trials) with an intertrial interval of 15 sec were given in a randomized Latin-square design. Histology Following the experiments the animals were sacrificed by an
AMYGDALOID STIMULATION AND STARTLE
345
FIG. 2. Coronal brain section stained with cresyl violet showing electrode placements in the central nucleus of the amygdala (arrows).
overdose of chloral hydrate and perfused intracardially with saline followed by 10% formalin. The brains were removed and alternate 40 p~m coronal sections through the amygdala were mounted on microscope slides and stained with cresyl violet or an acetylcholine esterase stain (which very clearly demarcates the central nucleus of the amygdala from other amygdala nuclei). Forty p~m sections through the brain stem were mounted and stained with cresyl violet. The slices were then dehydrated, coverslipped, and the placements were determined using the atlas of Paxinos and Watson (20).
Data Analysis Startle scores for each amygdaloid stimulation-startle stimulus interval were compared by taking the mean startle amplitude of the ten trials at each time interval. A repeated measures one-way ANOVA was used for statistical group analysis followed by a Newman-Keuls post hoc test (p<0.05) for individual comparisons. Scores from VCN, PLZ, RPC and MLF elicited startle were analyzed separately. RESULTS
from electrodes placed in the VAF and increased with more lateral placements. Placements in the VCN were found within VCN and the dorsal cochlear nucleus (Fig. 3A). Currents which elicited startle at the desired amplitude ranged from 30 to 100 txA. For the PLZ group, electrodes were found in and around the paralemniscal zone of the nucleus of the ventral lateral lemniscus as well as the rostral periolivary nucleus (Fig. 3B). Currents to elicit startle ranged from 50 to 180 IxA. Electrodes in the RPC group were found in the caudal ventral pons (Fig. 3C) in an area dorsal to the superior olives [the part of the RPC which lesions are most effective in blocking acoustic startle (5,6)] and as posterior as just dorsal to the facial nucleus, which is the caudal most electrode site from where startle can be elicited by moderate electrical currents. Because latencies of startle were similar with the various placements, all animals were tested and used for the analysis. Currents to elicit startle ranged from 25 to 115 p,A. For the MLF group, six electrodes were found in the MLF and two in overlying cerebellum just dorsal to the MLF (Fig. 3D). Because latencies of startle were similar with the various placements, all animals were tested and used for the analysis. Currents ranged from 40 to 160 IxA.
Histology Figures 2 and 3 show the electrode placements in the amygdala and startle circuit. Electrode tips aimed at the amygdala were found in the central nucleus (n = 10) and the area just medial to it corresponding to the origin of the ventral amygdalofugal pathway (VAF) (n = 25). As previously reported (21,22), the lowest stimulation currents which enhanced acoustic startle were obtained
Electrically Elicited Startle and Amygdaloid Stimulation Startle-like responses were elicited electrically at all stimulation sites within the startle circuit (i.e., VCN, PLZ, RPC and MLF). There were no differences in the startle amplitude between any of the sites, F(3,31)= 1.05, n.s. Enhancement of acoustic startle by amygdaloid stimulation was possible in all animals with
346
ROSEN AND DAVIS
L_ electrodes in the startle circuit. The results of amygdaloid stimulation on startle elicited electrically from the VCN, PLZ, RPC and MLF are shown in Fig. 4. Enhancement of electrically elicited startle by amygdaloid stimulation was found in all animals with electrodes in either the VCN, F(10,70)=12.9, p<0.0001, or PLZ, F(10,60)=29.9, p<0.0001. Startle elicited through the VCN was enhanced when the amygdala was stimulated simultaneously with VCN stimulation and with amygdaloid stimulation delivered up to 10 msec
before stimulation of the VCN (p<0.05). Significant enhancement first occurred when amygdaloid stimulation was presented simultaneously with the startle stimulus and became essentially maximal with 1-5 msec lead times. Amygdaloid stimulation delivered 2 msec to 10 msec before PLZ stimulation significantly enhanced startle (p<0.05). Significant enhancement first occurred when amygdaloid stimulation preceded the startle stimulus by 2 msec, with essentially maximal facilitation with 3-7 msec lead times. Startle elicited electrically from the RPC was enhanced by
AMYGDALOID STIMULATION AND STARTLE
347
t w
L FACING PAGE AND ABOVE FIG. 3. Coronal brain sections stained with cresyl violet showing electrode placements in the startle pathway used to electrically elicit startle. In A, arrows point to ventral cochlear nucleus (VCN). In this brain the left electrode tip was in the VCN while the right electrode was located in the dorsal cochlear nucleus, just dorsal to the VCN. In B, C, and D, the arrows point to electrode placements in the paralemniscal zone, nucleus of the reticularis pontis caudalis and medial longitudinal fasciculus, respectively.
amygdaloid stimulation in 11 of 12 animals, F(10,110)= 19.7, p < 0 . 0 0 0 1 . In these animals it was first enhanced significantly when amygdaloid stimulation preceded the startle stimulus by 3
msec with near maximal facilitation with 4 - 7 msec lead times. Startle elicited at the MLF was weakly but significantly facilitated, F ( 1 0 , 7 0 ) = 4 . 4 , p < 0 . 0 0 0 1 , however, only at the 7-
348
ROSEN AND DAVIS
EAR VCN 8
:E
I-rr <[ I,(n
4o
PLZ
•
RPC
J
MLF
VCN
VLL
3O
RPC
z
...............................
_'_.~__ -1
0
1
2
3
AMYGDALA STIMULATION-STARTLE
1'o
;
, 5
, 4
, 3
, 2
MS BEFORE STARTLE AMYGDALA
, 1
, 0
, 1 MS AFTER STARTLE
STIMULATION
FIG. 4. Graph of enhancement of electrically elicited startle from the VCN, PLZ, RPC and MLF by amygdaloid stimulation at different amygdaloid stimulation-startle elicitation intervals. + Denotes the shortest interval between amygdaloid stimulation and startle stimulus onset interval to significantly enhance startle. This interval became shorter as startle was elicited from the MLF, RPC, PLZ, and VCN, respectively. The dashed line signifies the mean amplitude of startle elicited from all electrode sites without amygdaloid stimulation. A pooled amplitude score was used because there were no differences between any of the groups (range was 20 to 23 units).
msec amygdaloid stimulation-startle stimulus interval (p<0.05). The magnitude of enhancement of MLF-elicited startle was significantly less than that of the VCN, PLZ and RPC (p<0.05). DISCUSSION
The present experiments examined the temporal aspects of enhancement of the electrically elicited startle reflex by stimulation of the amygdala. In addition, an attempt was made to locate the point in the startle pathway where transmission is altered by activation of the amygdala. Davis et al. (6) previously showed that startle-like responses could be elicited electrically with progressively shorter latencies as the electrodes were moved from the sensory to motor side of the reflex arc. Thus, the latency of startle elicited by stimulation of the VCN, PLZ or RPC was approximately 1, 2 or 3 msec, respectively, less than that of acoustic startle. The temporal relationship of enhancement by amygdaloid stimulation and startle electrically elicited from the various points along the startle circuit generally agree with these data. Stimulation of the amygdala presented simultaneously with the onset of VCN stimulation first showed significant enhancement of startle, which is about a 1-msec difference from the effects of amygdaloid stimulation on acoustic startle described previously (22). Significant enhancement first occurred when the amygdala was stimulated 2 msec before startle elicited by PLZ stimulation, suggesting that transmission from the VCN to the PLZ takes about 2 msec. In 11 of 12 animals, stimulation of the amygdala 3 msec before the onset of RPC stimulation enhanced startle; a difference of about 1 msec from PLZ to RPC. These data suggest the amygdaloid stimulation can facilitate both acoustically and electrically elicited startle in a manner that indicates that the startle pathway is serial, as described by Davis et al. (6). The temporal characteristics also suggest that the transmission of the startle-enhancing signal from the amygdala takes as little as 3 msec to reach the startle circuit and enhancement
STIMULATION INTERVAL (MSEC)
FIG. 5. Summary of results from the present study with electrically elicited startle and a previous study with enhancement of acoustic startle by amygdaloid stimulation (21). The graph shows the shortest interval between stimulation of the amygdala and startle stimulus onset producing significant enhancement of startle with startle elicited at various points along the startle pathway. Acoustically elicited startle first showed enhancement when the onset of a noise burst preceded amygdaloid stimulation by 1 msec ( - 1 msec on abscissa). Electrically elicited startle from the VCN, PLZ or RPC first showed enhancement when the onset of the electrically eliciting startle stimulus was respectively 0, 2, or 3 msec after amygdaloid stimulation. reaches its maximum by 4 msec (Fig. 1), consistent with previous estimates (22). A summary of the temporal relationship between amygdala-stimulated enhancement of electrically elicited startle from this study and enhancement of acoustically elicited startle from a previous paper (21) is presented in Fig. 5. The temporal relationship of enhancement of MLF-elicited startle did not agree with those of acoustic-, VCN-, PLZ-, and RPC-elicited startle. Enhancement of MLF-elicited startle occurred only with amygdaloid stimulation given 7 msec before the startle stimulus, suggesting at least a 7-8-msec transit time. Thus, the enhancement of MLF-elicited startle seems to be through amygdala efferents which are different and slower than those which affect startle elicited acoustically or by stimulating the VCN, PLZ, or RPC. Whether this is a multisynaptic or slowly conducting monosynaptic pathway to the spinal cord is unknown, although a direct amygdala pathway to the spinal cord has been shown in the cat and monkey (19,24). Electrically elicited startle from various points along the startle pathway was also used to attempt to identify the locus where the effects of amygdaloid stimulation enter the startle circuit. It was originally proposed that if startle was elicited in the circuit before the point where the signal from the amygdala enters, then the amygdala would enhance transmission in the startle pathway. Conversely, if startle was elicited in the circuit after the point where information from the amygdala enters, then the amygdala would not enhance this transmission (Fig. 1). Startle elicited electrically from the VCN or PLZ behaved in this fashion, suggesting that the point where amygdala enters the startle circuit is beyond the PLZ. Also, a primary spinal cord locus appears unlikely because startle elicited in the MLF was not increased at the appropriate stimulation-test interval. Therefore, the point where stimulation from the amygdala alters transmission in the startle circuit is most likely the RPC. However, startle elicited at the RPC was facilitated by amygdaloid stimulation in 11 of 12 animals. There are several ways to explain the results in keeping with the hypothesized amygdala impingement on the startle pathway at the RPC. First, stimulation of the area of the RPC
AMYGDALOID STIMULATION AND STARTLE
349
could also activate axons coming from the PLZ to the RPC. They may be antidromically triggered, causing the PLZ soma to initiate an orthodromic activation of the RPC which could interact with the signal coming from amygdala activation, thus facilitating startle. Alternatively, amygdaloid stimulation may produce excitatory postsynaptic potentials in the RPC which would allow for a facilitated and greater response with stimulation in the RPC. Previous work has shown that with high enough current amygdaloid stimulation by itself can elicit a startle-like reflex (21), suggesting the amygdaioid stimulation may postsynaptically affect neurons which mediate a short latency motor reflex. The tightly time-locked procedure employed in this study, using single pulses in the amygdala paired precisely with singlepulse activation of the startle circuit, should favor the detection of these types of interactions. In contrast, less temporally specific conditions, such as presentation of a visual stimulus or footshock many seconds before electrical elicitation of startle in the RPC
(2,3), would be less sensitive in detecting these types of interactions. This may explain why these conditions facilitate startle elicited electrically from the VCN and PLZ but not the RPC. Whatever the mechanism, it seems clear that we have reached the resolution of the electrically elicited startle technique and other methodologies (e.g., single unit recording in the startle pathway) will be required to definitively determine the exact point where amygdaloid stimulation facilitates startle. ACKNOWLEDGEMENTS This research was supported by National Science Foundation Grant BNS-81-20476, National Institute of Mental Health Grants MH-25642 and MH-41298, National Institute of Neurological Disorders and Stroke Grant NC-18033, Grant AFOSR-87-0336 from the Air Force Office of Scientific Research, and Research Scientist Development Award MH-00004 to Michael Davis. Jeffrey B. Rosen was supported by postdoctoral National Research Service Award MH-09550.
REFERENCES 1. Applegate, C. D.; Kapp, B. S.; Underwood, M. D.; McNall, C. L. Autonomic and somatomotor effects of amygdala central n. stimulation in awake rabbits. Physiol. Behav. 31:353-360; 1985. 2. Berg, W. K.; Davis, M. Associative learning modifies startle reflexes at the lateral lemniscus. Behav. Neurosci. 99:191-199; 1985. 3. Boulis, N. M.; Davis, M. Footshock induces sensitization of electrically elicited startle reflexes. Behav. Neurosci. 103:504-508; 1989. 4. Cassella, J. V.; Davis, M. The design and calibration of a startle measurement system. Physiol. Behav. 36:377-383; 1986. 5. Cassella, J. V.; Davis, M. Neural structures mediating acoustic and tactile startle reflexes and the acoustically elicited pinna response in rats: electrolytic and ibotenic acid studies. Soc. Neurosci. Abstr. 12:1273; 1986. 6. Davis, M.; Gendelman, D. S.; Tischler, M. D.; Gendelman, P. M. A primary acoustic startle circuit: lesion and stimulation studies. J. Neurosci. 6:791-805; 1982. 7. Davis, M.; Hitchcock, J. M.; Rosen, J. B. Anxiety and the amygdala: pharmacological and anatomical analysis of fear-potentiated startle. In: Bower, G. H., ed. The psychology of learning and motivation: Advances in research and theory, 21. San Diego: Academic Press; 1987:263-305. 8. Gastaut, H.; Vigouroux, R.; Corriol, J.; Badier, M. Effets da la stimulation electrique (par electofes a demeure) du complexe amygdalien chez le chat non narcose. J. Physiol. (Paris) 43:74t2-746; 1951. 9. Gentile, C. G.; Jan:ell, T. W.; Teich, A.; McCabe, P. M.; Schneiderman, N. The role of amygdaloid central nucleus in the retention of differential Pavlovian conditioning of bradycardia in rabbits. Behav. Brain Res. 20:263-273; 1986. 10. Hitchcock, J. M.; Davis, M. Lesions of the amygdala, but not of the cerebellum or red nucleus, block conditioned fear as measured with the potentiated startle paradigm. Behav. Neurosci. 100:11-22; 1986. 11. Hitchcock, J. M.; Davis, M. Fear-potentiated startle using an auditory conditioned stimulus: Effect of lesions of the amygdala. Physiol. Behav. 39:403--408; 1987. 12. Hitchcock, J. M.; Davis, M. The efferent pathway of the amygdala involved in conditioned fear as measured with the fear-potentiated startle paradigm. Submitted. 13. Hitchcock, J. M.; Sananes, C. B.; Davis, M. Sensitization of the startle reflex by footshock: Blockade by lesions of the central nucleus of the amygdala or its efferent pathway to the brainstem. Behav. Neurosci. 103:509-518; 1989.
14. Ison, J. R.; McAdam, D. W.; Hammond, G. R. Latency and amplitude changes in the acoustic startle reflex of the rat produced by variation in auditory prestimulation. Physiol. Behav. 10:1035-1039; 1973. 15. Iwata, J.; Chida, K.; LeDoux, J. E. Cardiovascular responses elicited by stimulation of neurons in the central amygdaloid nucleus in awake but not anesthetized rats resemble conditioned emotional responses. Brain Res. 418:183-188; 1987. 16. Kaada, B. R. Stimulation and regional ablation of the amygdaloid complex with reference to functional representations. In: Eleftheriou, B. E., ed. The neurobiology of the amygdala. New York: Plenum Press; 1972:205-281. 17. Kapp, B. S.; Frysinger, R. C.; GaUagher, M.; Haselton, J. R. Amygdala central nucleus lesions: Effects on heart rate conditioning in the rabbit. Physiol. Behav. 23:1109-1117; 1979. 18. Kapp, B. S.; Gallagher, M.; Underwood, M. D.; McNall, C. L.; Whitehorn, D. Cardiovascular responses elicited by electrical stimulation of the amygdala central nucleus in the rabbit. Brain Res. 234:251-262; 1982. 19. Mizuno, N.; Takahashi, O.; Satoda, T.; Matsushima, R. Amygdalospinal projections in the macaque monkey. Neurosci. Lett. 53: 327-330; 1985. 20. Paxinos, G.; Watson, C. The rat brain in stereotaxic coordinates. New York: Academic Press; 1986. 21. Rosen, J. B.; Davis, M. Temporal characteristics of enhancement of startle by stimulation of the amygdala. Physiol. Behav. 44:117-123; 1988. 22. Rosen, J. R.; Davis, M. Enhancement of acoustic startle by electrical stimulation of the amygdala. Behav. Neurosci. 102:195-202; 1988. 23. Rosen, J. B.; Hitchcock, J. M.; Sananes, C. B.; Miserendino, M. J. D.; Davis, M. A direct projection from the central nucleus of the amygdala to the acoustic startle pathway: Anterograde and retrograde tracing studies. Submitted. 24. Sandrew, B. B.; Edwards, D. L.; Poletti, C. E.; Foote, W. E. Amygdalospinal projections in the cat. Brain Res. 373:235-239; 1986. 25. Zbrozyna, A. W. The organization of the defence reaction elicited from amygdala and its connections. In: Eleftheriou, B. E., ed. The neurobiology of the amygdala. New York: Plenum Press; 1972: 597--606.
0031-9384/90 $3.00 + .00
Enhancement of Electrically Elicited Startle by Amygdaloid Stimulation J E F F R E Y B. R O S E N 1 A N D M I C H A E L D A V I S
Department of Psychiatry, Yale University, New Haven, CT 06508 R e c e i v e d 9 January 1990
ROSEN, J. B. AND M. DAVIS. Enhancement of electrically elicited startle by amygdaloid stimulation. PHYSIOL BEHAV 48(2) 343-349, 1990.--Previous experiments showed that acoustic startle amplitude can be enhanced by electrical stimulation of the amygdala. Because the acoustic startle pathway is organized in a serial fashion, startle can be elicited electrically with progressively shorter latencies by stimulating different points along this pathway [i.e., ventral cochlear nucleus (VCN), paralemniscal zone (PLZ), nucleus reticularis pontis caudalis (RPC) or medial longitudinal fasciculus (MLF)]. The present study evaluated the temporal characteristics of the facilitatory effect of amygdaloid stimulation on startle elicited electrically from different points along the acoustic startle pathway. A single 0.1-msec pulse was delivered to the central nucleus of the amygdala at various times before the onset of a 1-msec pulse in various sites in the startle pathway. The shortest amygdaloid stimulation-startle onset interval to significantly enhance startle was 0 msec for the VCN, 2 msec for the PLZ, 3 msec for the RPC and 7 msec for the MLF. These results indicate that amygdaloid stimulation enhances electrically elicited startle in a temporal manner that is complementary to facilitation of acoustic startle. The similarity of amygdala-stimulated enhancement and fear potentiation of electrically elicited startle is also discussed. Startle
Fear-potentiated startle
Amygdala
Electrical brain stimulation
Fear
Anatomical data indicate that the central nucleus projects directly to the area of the R I ~ known to be critical for startle (23). Lesions at various points along this pathway block startle potentiated by both conditioned (12) and unconditioned fear such as footshock (13). Taken together, these data suggest that conditioned fear is mediated by activation of the amygdala which then modulates startle via its direct pathway to the RPC. Consistent with this interpretation, startle-like responses elicited from either the VCN or PLZ are increased in the presence of a conditioned fear stimulus (2) or shortly following footshock (3). In contrast, startle-like responses elicited by electrical stimulation of the RPC or the pathway from the RPC to the spinal cord are not (2,3). Thus, startle-like responses elicited by stimulation of the startle circuit afferent to the RPC are affected by events thought to activate the amygdala whereas startle-like responses elicited by stimulation of the circuit downstream from the RPC are not. If this is the case, then electrical stimulation of the amygdala should also increase startle-like responses elicited from the VCN and PLZ but not from stimulation beyond the RPC. Moreover, if maximal facilitation of startle occurs when impulses from the amygdala arrive at the RPC at the same time as impulses activated by the startle stimulus, then the optimal interval for enhancement between amygdaloid stimulation and acoustically elicited startle should shift by about 1 msec from that seen when startle is elicited electrically from the VCN and by 2-3 msec when startle is elicited electrically from the PLZ. Figure 1 shows the logic for these predictions schematically, which were tested in the present study.
ACTIVATION of the amygdala has been shown to play an important role in fear. For instance, electrical stimulation of the central nucleus of the amygdala can produce a pattern of behavioral and autonomic changes that resemble a state of fear (1, 8, 15, 16, 18, 21, 22, 25). Conversely, lesions of the amygdala block both conditioned and unconditioned fear (9-11, 13, 16, 17). Recently it was shown that lesions of the central nucleus of the amygdala block increases in the amplitude of acoustic startle when startle is elicited in the presence of a conditioned fear stimulus or following aversive stimulation such as footshock (10, 11, 13). Electrical stimulation of the amygdala also causes a marked increase in the amplitude of the acoustic startle reflex (21,22). Thus, it is possible that the increase in startle seen during electrical stimulation of the amygdala mimics facilitation of startle during a state of fear (7). The acoustic startle reflex has a very short latency [8 msec recorded electromyographically in the hindlegs (14)] and appears to be mediated by a serial pathway consisting of the ventral cochlear nucleus (VCN), an area just medial and ventral to the ventral nucleus of the lateral lemniscus [paralemniscal zone (PLZ)], an area in the nucleus reticularis pontis caudalis just dorsal to the superior olives (RPC) and motor neurons in the spinal cord (5,6). Electrical, single pulse stimulation within each of these nuclei produces startle-like responses which have a progressively shorter latency as the electrode is moved from the VCN (7 msec) to the PLZ (5--6 msec) to the RPC (4--5 msec) to the spinal cord (3 msec) (6).
tRequests for reprints should be addressed to Jeffrey B. Rosen, NIMH, Biological Psychiatry Branch, Building 10 Room 3N212, 9000 Rockville Pike, Bethesda, MD 20892.
343
344
ROSEN AND DAVIS
METHOD
STIM
,l
Subjects and Surgery A total of 35 male albino Sprague-Dawley rats (Charles River Co.) weighing 300--400 g were used. Rats were housed in group cages until surgery and individually thereafter. A 12:12 light-dark cycle was maintained in the animal room. All testing was done during the light cycle. Feeding was ad lib. All rats were anesthetized with 8% chloral hydrate (1 ml/200 g) and unipolar electrodes (00 insect pins, 0.25 mm tip diameter, insulated except for 0.5 mm of the tip) were stereotaxically aimed bilaterally at the medial portion of the central nucleus of the amygdala (2.8 mm posterior, 4.0 mm lateral and 8.5 mm ventral to bregma with the skull surface level). These same animals also were implanted with bilateral electrodes in various sites in the startle circuit. Eight were in the VCN (11.3 mm posterior, 3.8 mm lateral and 8.6 mm ventral to bregma with the skull surface level), 7 in the PLZ (8.3 mm posterior, 2.0 mm lateral and 9.6 mm ventral to bregma with the skull surface level), 12 in the RPC (10.0 or 10.5 mm posterior, 1.5 mm lateral and 9.5 mm ventral to bregma with the skull surface level) and 8 in the medial longitudinal fasciculus (MLF) (on the midline and 12.3 or 12.7 mm posterior and 9.6 mm ventral to bregma with the skull surface level) which is believed to carry the axons from the RPC to the spinal cord (6). A stainless steel screw fixed to the skull over the frontal cortex was used as an indifferent electrode. Several other stainless steel screws were used to anchor the electrode assembly to the skull. The electrodes were cemented in place and the leads from the electrodes were enclosed in a Viking (#TKR-7) socket and mounted on the skull with dental acrylic.
Apparatus A stabilimeter was used to record the amplitude of the startle responses. The stabilimeter consisted of a 12 x 20 x 18 cm Plexiglas and wire mesh cage suspended between compression springs within a steel frame. The cage had a small opening in the top to allow leads to enter the cage for electrical stimulation of the brain. The stabilimeter was housed in a lighted, ventilated soundattenuating chamber. The cage was located 10 cm from a high frequency speaker (Radio Shack Supertweeter). Cage movement resulted in displacement of an accelerometer in which the resultant voltage was proportional to the velocity of displacement. Startle amplitude was defined as the maximal accelerometer voltage that occurred during the first 200 msec after a startle stimulus was delivered. As described elsewhere, acoustically or electrically elicited startle has a latency of about 5 msec recorded electromyographically in the neck muscles, 8 msec in the hindleg (6,14). This very fast sequence of muscle contractions results in cage movement having a frequency of about 7-10 Hz in this particular test cage (4). By sampling over a 200-msec period after the onset of the startle stimulus, the peak amplitude of cage movement initiated by the very short latency startle response can be reliably recorded (4). Startle amplitude (range 0 to 88 units) was recorded by an analogue-to-digital converter interfaced to a printer. Background white noise, provided by a white noise generator, was 55 dB. Sound level measurements were made within the cage with a General Radio Model 1551-C sound level meter (A-scale). Electrical brain stimulation was delivered through separate constant current stimulators (Electronic Instruments, #BSI-1) for each electrode gated from a Grass $88 stimulator which also controlled the delivery of the acoustic startle stimulus. Stimulation current, measured across a differential amplifier, and the acoustic
STARTLE I~ RESPON~IE ENHANCI~
STIM
,J
STIM
STIM
STARTLE
FIG. 1. Schematic diagram of the logic for enhancement of electrically elicited startle by amygdaloid stimulation. (A) Because the VCN and PLZ are before the proposed point of modulation of startle by amygdaloid stimulation, startle elicited from these nuclei should be enhanced by amygdaloid stimulation. (B) Because the RPC and MLF are at or beyond the point of modulation, startle elicited from these sites should not be enhanced. The acoustic startle pathway [ear, VCN, PLZ, RPC, MLF, and SC (spinal cord)] is from Davis et al. (6). The transit times shown in A between the synapses of the startle circuit are derived from Davis et al. (6) and the present experiment. The transit time from the amygdala to the startle pathway is derived from Rosen and Davis (22) and the present study. startle stimulus, measured by a dB meter, were continuously monitored by a Tektronix 5130N oscilloscope. Procedure A week following surgery, rats were placed in the test cage, given 30-60 min for adaptation and tested for enhancement of acoustic startle by electrical stimulation of the amygdala through each electrode individually. The acoustic startle stimulus was a 20-msec burst of white noise having a rise-decay time of 5 msec. The intensity of the acoustic startle stimulus was adjusted (95-115 dB) to give a low to intermediate amplitude of response (20-30 units). To determine the amount of amygdala stimulation necessary to enhance acoustic startle, a single 0.1-msec cathodal square-wave pulse ranging between 40 and 400 I~A was delivered to the amygdala simultaneously with onset of the acoustic startle stimulus to increase startle by approximately 40 units over the response elicited by the startle stimulus alone. Simultaneous stimulation of the amygdala and onset of the acoustic stimulus was chosen because previous experiments demonstrated that this produced maximal enhancement of acoustic startle (21,22). The electrode with the lowest current to enhance acoustic startle was used throughout the remaining experiment. For the electrically elicited startle experiments, startle was elicited by a 1.0-msec cathodal square wave pulse through the bilateral electrodes in the VCN, PLZ, RPC or MLF. The current was adjusted (25-180 I~A) to give a startle amplitude similar to that elicited by the acoustic stimulus (20-30 units). Immediately following this, amygdaloid stimulation was delivered 1 msec after, simultaneously, or 1, 2, 3, 4, 5, 7, 10, or 20 msec before the onset of the startle eliciting electrical pulse in the startle circuit. Trials of electrically elicited startle without an accompanying amygdaloid stimulation were also given. Ten occurrences of each of these stimulus conditions (trials) with an intertrial interval of 15 sec were given in a randomized Latin-square design. Histology Following the experiments the animals were sacrificed by an
AMYGDALOID STIMULATION AND STARTLE
345
FIG. 2. Coronal brain section stained with cresyl violet showing electrode placements in the central nucleus of the amygdala (arrows).
overdose of chloral hydrate and perfused intracardially with saline followed by 10% formalin. The brains were removed and alternate 40 p~m coronal sections through the amygdala were mounted on microscope slides and stained with cresyl violet or an acetylcholine esterase stain (which very clearly demarcates the central nucleus of the amygdala from other amygdala nuclei). Forty p~m sections through the brain stem were mounted and stained with cresyl violet. The slices were then dehydrated, coverslipped, and the placements were determined using the atlas of Paxinos and Watson (20).
Data Analysis Startle scores for each amygdaloid stimulation-startle stimulus interval were compared by taking the mean startle amplitude of the ten trials at each time interval. A repeated measures one-way ANOVA was used for statistical group analysis followed by a Newman-Keuls post hoc test (p<0.05) for individual comparisons. Scores from VCN, PLZ, RPC and MLF elicited startle were analyzed separately. RESULTS
from electrodes placed in the VAF and increased with more lateral placements. Placements in the VCN were found within VCN and the dorsal cochlear nucleus (Fig. 3A). Currents which elicited startle at the desired amplitude ranged from 30 to 100 txA. For the PLZ group, electrodes were found in and around the paralemniscal zone of the nucleus of the ventral lateral lemniscus as well as the rostral periolivary nucleus (Fig. 3B). Currents to elicit startle ranged from 50 to 180 IxA. Electrodes in the RPC group were found in the caudal ventral pons (Fig. 3C) in an area dorsal to the superior olives [the part of the RPC which lesions are most effective in blocking acoustic startle (5,6)] and as posterior as just dorsal to the facial nucleus, which is the caudal most electrode site from where startle can be elicited by moderate electrical currents. Because latencies of startle were similar with the various placements, all animals were tested and used for the analysis. Currents to elicit startle ranged from 25 to 115 p,A. For the MLF group, six electrodes were found in the MLF and two in overlying cerebellum just dorsal to the MLF (Fig. 3D). Because latencies of startle were similar with the various placements, all animals were tested and used for the analysis. Currents ranged from 40 to 160 IxA.
Histology Figures 2 and 3 show the electrode placements in the amygdala and startle circuit. Electrode tips aimed at the amygdala were found in the central nucleus (n = 10) and the area just medial to it corresponding to the origin of the ventral amygdalofugal pathway (VAF) (n = 25). As previously reported (21,22), the lowest stimulation currents which enhanced acoustic startle were obtained
Electrically Elicited Startle and Amygdaloid Stimulation Startle-like responses were elicited electrically at all stimulation sites within the startle circuit (i.e., VCN, PLZ, RPC and MLF). There were no differences in the startle amplitude between any of the sites, F(3,31)= 1.05, n.s. Enhancement of acoustic startle by amygdaloid stimulation was possible in all animals with
346
ROSEN AND DAVIS
L_ electrodes in the startle circuit. The results of amygdaloid stimulation on startle elicited electrically from the VCN, PLZ, RPC and MLF are shown in Fig. 4. Enhancement of electrically elicited startle by amygdaloid stimulation was found in all animals with electrodes in either the VCN, F(10,70)=12.9, p<0.0001, or PLZ, F(10,60)=29.9, p<0.0001. Startle elicited through the VCN was enhanced when the amygdala was stimulated simultaneously with VCN stimulation and with amygdaloid stimulation delivered up to 10 msec
before stimulation of the VCN (p<0.05). Significant enhancement first occurred when amygdaloid stimulation was presented simultaneously with the startle stimulus and became essentially maximal with 1-5 msec lead times. Amygdaloid stimulation delivered 2 msec to 10 msec before PLZ stimulation significantly enhanced startle (p<0.05). Significant enhancement first occurred when amygdaloid stimulation preceded the startle stimulus by 2 msec, with essentially maximal facilitation with 3-7 msec lead times. Startle elicited electrically from the RPC was enhanced by
AMYGDALOID STIMULATION AND STARTLE
347
t w
L FACING PAGE AND ABOVE FIG. 3. Coronal brain sections stained with cresyl violet showing electrode placements in the startle pathway used to electrically elicit startle. In A, arrows point to ventral cochlear nucleus (VCN). In this brain the left electrode tip was in the VCN while the right electrode was located in the dorsal cochlear nucleus, just dorsal to the VCN. In B, C, and D, the arrows point to electrode placements in the paralemniscal zone, nucleus of the reticularis pontis caudalis and medial longitudinal fasciculus, respectively.
amygdaloid stimulation in 11 of 12 animals, F(10,110)= 19.7, p < 0 . 0 0 0 1 . In these animals it was first enhanced significantly when amygdaloid stimulation preceded the startle stimulus by 3
msec with near maximal facilitation with 4 - 7 msec lead times. Startle elicited at the MLF was weakly but significantly facilitated, F ( 1 0 , 7 0 ) = 4 . 4 , p < 0 . 0 0 0 1 , however, only at the 7-
348
ROSEN AND DAVIS
EAR VCN 8
:E
I-rr <[ I,(n
4o
PLZ
•
RPC
J
MLF
VCN
VLL
3O
RPC
z
...............................
_'_.~__ -1
0
1
2
3
AMYGDALA STIMULATION-STARTLE
1'o
;
, 5
, 4
, 3
, 2
MS BEFORE STARTLE AMYGDALA
, 1
, 0
, 1 MS AFTER STARTLE
STIMULATION
FIG. 4. Graph of enhancement of electrically elicited startle from the VCN, PLZ, RPC and MLF by amygdaloid stimulation at different amygdaloid stimulation-startle elicitation intervals. + Denotes the shortest interval between amygdaloid stimulation and startle stimulus onset interval to significantly enhance startle. This interval became shorter as startle was elicited from the MLF, RPC, PLZ, and VCN, respectively. The dashed line signifies the mean amplitude of startle elicited from all electrode sites without amygdaloid stimulation. A pooled amplitude score was used because there were no differences between any of the groups (range was 20 to 23 units).
msec amygdaloid stimulation-startle stimulus interval (p<0.05). The magnitude of enhancement of MLF-elicited startle was significantly less than that of the VCN, PLZ and RPC (p<0.05). DISCUSSION
The present experiments examined the temporal aspects of enhancement of the electrically elicited startle reflex by stimulation of the amygdala. In addition, an attempt was made to locate the point in the startle pathway where transmission is altered by activation of the amygdala. Davis et al. (6) previously showed that startle-like responses could be elicited electrically with progressively shorter latencies as the electrodes were moved from the sensory to motor side of the reflex arc. Thus, the latency of startle elicited by stimulation of the VCN, PLZ or RPC was approximately 1, 2 or 3 msec, respectively, less than that of acoustic startle. The temporal relationship of enhancement by amygdaloid stimulation and startle electrically elicited from the various points along the startle circuit generally agree with these data. Stimulation of the amygdala presented simultaneously with the onset of VCN stimulation first showed significant enhancement of startle, which is about a 1-msec difference from the effects of amygdaloid stimulation on acoustic startle described previously (22). Significant enhancement first occurred when the amygdala was stimulated 2 msec before startle elicited by PLZ stimulation, suggesting that transmission from the VCN to the PLZ takes about 2 msec. In 11 of 12 animals, stimulation of the amygdala 3 msec before the onset of RPC stimulation enhanced startle; a difference of about 1 msec from PLZ to RPC. These data suggest the amygdaloid stimulation can facilitate both acoustically and electrically elicited startle in a manner that indicates that the startle pathway is serial, as described by Davis et al. (6). The temporal characteristics also suggest that the transmission of the startle-enhancing signal from the amygdala takes as little as 3 msec to reach the startle circuit and enhancement
STIMULATION INTERVAL (MSEC)
FIG. 5. Summary of results from the present study with electrically elicited startle and a previous study with enhancement of acoustic startle by amygdaloid stimulation (21). The graph shows the shortest interval between stimulation of the amygdala and startle stimulus onset producing significant enhancement of startle with startle elicited at various points along the startle pathway. Acoustically elicited startle first showed enhancement when the onset of a noise burst preceded amygdaloid stimulation by 1 msec ( - 1 msec on abscissa). Electrically elicited startle from the VCN, PLZ or RPC first showed enhancement when the onset of the electrically eliciting startle stimulus was respectively 0, 2, or 3 msec after amygdaloid stimulation. reaches its maximum by 4 msec (Fig. 1), consistent with previous estimates (22). A summary of the temporal relationship between amygdala-stimulated enhancement of electrically elicited startle from this study and enhancement of acoustically elicited startle from a previous paper (21) is presented in Fig. 5. The temporal relationship of enhancement of MLF-elicited startle did not agree with those of acoustic-, VCN-, PLZ-, and RPC-elicited startle. Enhancement of MLF-elicited startle occurred only with amygdaloid stimulation given 7 msec before the startle stimulus, suggesting at least a 7-8-msec transit time. Thus, the enhancement of MLF-elicited startle seems to be through amygdala efferents which are different and slower than those which affect startle elicited acoustically or by stimulating the VCN, PLZ, or RPC. Whether this is a multisynaptic or slowly conducting monosynaptic pathway to the spinal cord is unknown, although a direct amygdala pathway to the spinal cord has been shown in the cat and monkey (19,24). Electrically elicited startle from various points along the startle pathway was also used to attempt to identify the locus where the effects of amygdaloid stimulation enter the startle circuit. It was originally proposed that if startle was elicited in the circuit before the point where the signal from the amygdala enters, then the amygdala would enhance transmission in the startle pathway. Conversely, if startle was elicited in the circuit after the point where information from the amygdala enters, then the amygdala would not enhance this transmission (Fig. 1). Startle elicited electrically from the VCN or PLZ behaved in this fashion, suggesting that the point where amygdala enters the startle circuit is beyond the PLZ. Also, a primary spinal cord locus appears unlikely because startle elicited in the MLF was not increased at the appropriate stimulation-test interval. Therefore, the point where stimulation from the amygdala alters transmission in the startle circuit is most likely the RPC. However, startle elicited at the RPC was facilitated by amygdaloid stimulation in 11 of 12 animals. There are several ways to explain the results in keeping with the hypothesized amygdala impingement on the startle pathway at the RPC. First, stimulation of the area of the RPC
AMYGDALOID STIMULATION AND STARTLE
349
could also activate axons coming from the PLZ to the RPC. They may be antidromically triggered, causing the PLZ soma to initiate an orthodromic activation of the RPC which could interact with the signal coming from amygdala activation, thus facilitating startle. Alternatively, amygdaloid stimulation may produce excitatory postsynaptic potentials in the RPC which would allow for a facilitated and greater response with stimulation in the RPC. Previous work has shown that with high enough current amygdaloid stimulation by itself can elicit a startle-like reflex (21), suggesting the amygdaioid stimulation may postsynaptically affect neurons which mediate a short latency motor reflex. The tightly time-locked procedure employed in this study, using single pulses in the amygdala paired precisely with singlepulse activation of the startle circuit, should favor the detection of these types of interactions. In contrast, less temporally specific conditions, such as presentation of a visual stimulus or footshock many seconds before electrical elicitation of startle in the RPC
(2,3), would be less sensitive in detecting these types of interactions. This may explain why these conditions facilitate startle elicited electrically from the VCN and PLZ but not the RPC. Whatever the mechanism, it seems clear that we have reached the resolution of the electrically elicited startle technique and other methodologies (e.g., single unit recording in the startle pathway) will be required to definitively determine the exact point where amygdaloid stimulation facilitates startle. ACKNOWLEDGEMENTS This research was supported by National Science Foundation Grant BNS-81-20476, National Institute of Mental Health Grants MH-25642 and MH-41298, National Institute of Neurological Disorders and Stroke Grant NC-18033, Grant AFOSR-87-0336 from the Air Force Office of Scientific Research, and Research Scientist Development Award MH-00004 to Michael Davis. Jeffrey B. Rosen was supported by postdoctoral National Research Service Award MH-09550.
REFERENCES 1. Applegate, C. D.; Kapp, B. S.; Underwood, M. D.; McNall, C. L. Autonomic and somatomotor effects of amygdala central n. stimulation in awake rabbits. Physiol. Behav. 31:353-360; 1985. 2. Berg, W. K.; Davis, M. Associative learning modifies startle reflexes at the lateral lemniscus. Behav. Neurosci. 99:191-199; 1985. 3. Boulis, N. M.; Davis, M. Footshock induces sensitization of electrically elicited startle reflexes. Behav. Neurosci. 103:504-508; 1989. 4. Cassella, J. V.; Davis, M. The design and calibration of a startle measurement system. Physiol. Behav. 36:377-383; 1986. 5. Cassella, J. V.; Davis, M. Neural structures mediating acoustic and tactile startle reflexes and the acoustically elicited pinna response in rats: electrolytic and ibotenic acid studies. Soc. Neurosci. Abstr. 12:1273; 1986. 6. Davis, M.; Gendelman, D. S.; Tischler, M. D.; Gendelman, P. M. A primary acoustic startle circuit: lesion and stimulation studies. J. Neurosci. 6:791-805; 1982. 7. Davis, M.; Hitchcock, J. M.; Rosen, J. B. Anxiety and the amygdala: pharmacological and anatomical analysis of fear-potentiated startle. In: Bower, G. H., ed. The psychology of learning and motivation: Advances in research and theory, 21. San Diego: Academic Press; 1987:263-305. 8. Gastaut, H.; Vigouroux, R.; Corriol, J.; Badier, M. Effets da la stimulation electrique (par electofes a demeure) du complexe amygdalien chez le chat non narcose. J. Physiol. (Paris) 43:74t2-746; 1951. 9. Gentile, C. G.; Jan:ell, T. W.; Teich, A.; McCabe, P. M.; Schneiderman, N. The role of amygdaloid central nucleus in the retention of differential Pavlovian conditioning of bradycardia in rabbits. Behav. Brain Res. 20:263-273; 1986. 10. Hitchcock, J. M.; Davis, M. Lesions of the amygdala, but not of the cerebellum or red nucleus, block conditioned fear as measured with the potentiated startle paradigm. Behav. Neurosci. 100:11-22; 1986. 11. Hitchcock, J. M.; Davis, M. Fear-potentiated startle using an auditory conditioned stimulus: Effect of lesions of the amygdala. Physiol. Behav. 39:403--408; 1987. 12. Hitchcock, J. M.; Davis, M. The efferent pathway of the amygdala involved in conditioned fear as measured with the fear-potentiated startle paradigm. Submitted. 13. Hitchcock, J. M.; Sananes, C. B.; Davis, M. Sensitization of the startle reflex by footshock: Blockade by lesions of the central nucleus of the amygdala or its efferent pathway to the brainstem. Behav. Neurosci. 103:509-518; 1989.
14. Ison, J. R.; McAdam, D. W.; Hammond, G. R. Latency and amplitude changes in the acoustic startle reflex of the rat produced by variation in auditory prestimulation. Physiol. Behav. 10:1035-1039; 1973. 15. Iwata, J.; Chida, K.; LeDoux, J. E. Cardiovascular responses elicited by stimulation of neurons in the central amygdaloid nucleus in awake but not anesthetized rats resemble conditioned emotional responses. Brain Res. 418:183-188; 1987. 16. Kaada, B. R. Stimulation and regional ablation of the amygdaloid complex with reference to functional representations. In: Eleftheriou, B. E., ed. The neurobiology of the amygdala. New York: Plenum Press; 1972:205-281. 17. Kapp, B. S.; Frysinger, R. C.; GaUagher, M.; Haselton, J. R. Amygdala central nucleus lesions: Effects on heart rate conditioning in the rabbit. Physiol. Behav. 23:1109-1117; 1979. 18. Kapp, B. S.; Gallagher, M.; Underwood, M. D.; McNall, C. L.; Whitehorn, D. Cardiovascular responses elicited by electrical stimulation of the amygdala central nucleus in the rabbit. Brain Res. 234:251-262; 1982. 19. Mizuno, N.; Takahashi, O.; Satoda, T.; Matsushima, R. Amygdalospinal projections in the macaque monkey. Neurosci. Lett. 53: 327-330; 1985. 20. Paxinos, G.; Watson, C. The rat brain in stereotaxic coordinates. New York: Academic Press; 1986. 21. Rosen, J. B.; Davis, M. Temporal characteristics of enhancement of startle by stimulation of the amygdala. Physiol. Behav. 44:117-123; 1988. 22. Rosen, J. R.; Davis, M. Enhancement of acoustic startle by electrical stimulation of the amygdala. Behav. Neurosci. 102:195-202; 1988. 23. Rosen, J. B.; Hitchcock, J. M.; Sananes, C. B.; Miserendino, M. J. D.; Davis, M. A direct projection from the central nucleus of the amygdala to the acoustic startle pathway: Anterograde and retrograde tracing studies. Submitted. 24. Sandrew, B. B.; Edwards, D. L.; Poletti, C. E.; Foote, W. E. Amygdalospinal projections in the cat. Brain Res. 373:235-239; 1986. 25. Zbrozyna, A. W. The organization of the defence reaction elicited from amygdala and its connections. In: Eleftheriou, B. E., ed. The neurobiology of the amygdala. New York: Plenum Press; 1972: 597--606.