The role of the anterior hypothalamus in affective defense behavior elicited from the ventromedial hypothalamus of the cat

Brain Research, 330 (1985) 93-107 Elsevier

93

BRE 10609

The Role of the Anterior Hypothalamus in Affective Defense Behavior Elicited from the Ventromedial Hypothalamus of the Cat SUSAN A. G. FUCHS, HENRY M. EDINGER and ALLAN SIEGEL Departments of Physiology and Neuroscience, UMDNJ-New Jersey Medical School, 1O0 Bergen Street, Newark, NJ 07103 (U. S. A.) (Accepted June 26th, 1984) Key words: ventromedial hypothalamus - - anteromedial hypothalamus - - affective defense behavior - - electrical stimulation - radiofrequency l e s i o n s - [14C]2-deoxyglucoseautoradiography

In the preceding paper a hypothalamic circuit subserving feline affective defense behavior was described. This circuit included an ascending component from the ventromedial nucleus to the anterior hypothalamus and a descending component from the anterior hypothalamus to the midbrain central gray substance. The present study was undertaken to test the hypothesis that the anterior hypothalamus plays a central role in the organization of this functional pathway. In the first part of this study, dual stimulation methods were utilized to demonstrate that concurrent stimulation of the ventromedial hypothalamus facilitates the occurrence of affective defense responses elicited from the anterior hypothalamus. In the second part of the study, lesions placed in the anterior hypothalamus significantly increased the latency and threshold current for affective defense responses elicited from the ventromedial hypothalamus. [14C]2-deoxyglucose autoradiography confirmed the fact that anterior hypothalamic lesions effective in blocking affective defense were placed in regions where the vast majority of ventromedial hypothalamic fibers terminate. In contrast, lesions which had little or no effect upon the latency or threshold for affective defense elicited from the ventromedial hypothalamus appeared to leave intact the connections from the ventromedial to the anterior hypothalamus. These findings are consistent with the proposed intrahypothalamic anatomical substrate subserving affective defense behavior described in the preceding paper. INTRODUCTION In the preceding study6, an a t t e m p t was m a d e to describe the pathways associated with the feline affective defense response utilizing techniques of [3H]leucine and [14C]2-deoxyglucose autoradiography. The results o b t a i n e d from both a u t o r a d i o g r a p h ic methods suggest that the p a t h w a y from the ventromedial h y p o t h a l a m u s to the m i d b r a i n contains at least a disynaptic circuit in which the initial limb is an ascending p r o j e c t i o n to the anterior h y p o t h a l a m u s and the s e c o n d - o r d e r n e u r o n constitutes a descending projection from the anterior h y p o t h a l a m u s to the midbrain central gray substance. That such a circuit may be potentially critical for the organization of affective defense is also suggested from several other experiments. F o r e x a m p l e , knife cuts which disrupted the connections b e t w e e n the v e n t r o m e d i a l and anterior h y p o t h a l a m u s inhibited intraspecies ag-

ression in rats 12. A d d i t i o n a l l y , medial posterior hypothalamic lesions could not abolish intraspecific attack in rats or cats in which the connections b e t w e e n the ventromedial and anterior h y p o t h a l a m u s were spared 12-14. On the basis of these reports, the present study specifically sought to test the hypothesis that the ventromedial h y p o t h a l a m u s mediates affective defense through its connections with the anterior hypothalamus. This hypothesis was examined with the use of techniques of electrical stimulation and focal brain lesions coupled with [14C]2-deoxyglucose (2-DG) autoradiography. MATERIALS AND METHODS Electrical brain stimulation The methods utilized were presented in detail in the preceding study6, and will be summarized briefly here. Eight adult cats of either sex were chosen for

Correspondence: A. Siegel, Department of Neuroscience, UMDNJ-New Jersey Medical School, 100 Bergen Street, Newark, NJ 07103, U.S.A. 0006-8993/85/$03.30 ~ 1985 Elsevier Science Publishers B.V. (Biomedical Division)

94 study which did not exhibit spontaneous aggressive responses upon placement of a rat or cat in their home cage. During aseptic surgery, 16 stainless steel guide tubes were mounted stereotaxically over holes drilled through the skull overlying the hypothalamus. After a 7-day recovery period, each cat was placed in a behavioral observation chamber containing a rat, a bowl of food and a second cat to assess the directedness of affective defense. During the initial behavioral trials, insulated monopolar electrodes were lowered through the guide tubes in the awake cat in 0.5 mm steps. At each step, the cat was stimulated with biphasic rectangular pulses (0.1-0.8 mA) until an affective defense response was obtained. Affective defense sites were identified in the ventromediai nucleus, and anterior hypothalamic region. The electrodes used to elicit the behavior were then cemented in place with dental acrylic. The threshold current for the behavior was determined by the ascending and descending Method of Limits and the latency to hiss, growl or strike, was recorded at each trial of stimulation.

Dual stimulation of the hypothalamus Sites were identified in both the anteromedial and ipsilateral ventromedial hypothalamic regions from which affective defense was elicited. The electrodes were cemented in place and the behavioral latency and threshold current values at each of these sites were determined. Paired trials involving single stimulation (A) of an individual hypothalamic site alone and dual stimulation (B) of two sites within the hypothalamus were administered to the awake cat in an A-B-B-A design. Two principal situations were examined: (1) the modulatory electrode was placed in the ventromedial nucleus and the behavior-eliciting electrode was situated in the anterior hypothalamus; and (2) the modulatory electrode was located in the anterior hypothalamus and the behavior-eliciting electrode was situated in the ventromedial nucleus. In addition, attempts were made to modulate flight behavior from either the anterior or posterior hypothalamus using concurrent subthreshold stimulation delivered to a ventromedial hypothalamic site associated with affective defense. Flight behavior in the cat was characterized by pupillary dilatation followed by an attempt to jump out of the observation cage. Current delivered to the behavioral site was supra-

threshold while that delivered to the site modulating affective defense was subthreshold and did not elicit components of the response. When both areas were stimulated concomitantly, balanced biphasic pulses were set such that stimulation of the behavioral site followed the onset of the modulatory site by 4 ms. Response latencies were recorded on each trial of stimulation. At t-test for paired observations was employed to analyze the differences in response latencies obtained on trials of single and dual stimulation. After this experiment was complete, the cats were then utilized for the lesion study described below.

Hypothalamic lesions Sites were identified within the ventromedial, anterior or posterior hypothalamus from which affective defense (or flight behavior in 3 cases) could be elicited. First, the behavioral current thresholds were determined by the Method of Limits. Then, the response latencies at fixed current values (just above threshold) were recorded in 20 trials of stimulation over a 4-6 day period in order to establish the stability of the hypothalamic responses. The threshold was defined as the lowest current value at which an affective defense response could be consistently elicited within 15 s on 100% of the trials tested. In each cat, a small radiofrequency lesion (40-60 mA for 25 s) was placed through the stimulating electrode in the anterior or posterior hypothalamus following completion of the behavioral trials described above. Small lesions were utilized in order to reduce the possibility that their placement might impinge upon any aspect of the stimulated tissue within the ventromedial nucleus. After the placement of the lesion, the cat was stimulated at currents 4 times threshold utilizing the electrode through which the lesion was made in order to confirm that affective defense or flight behavior could no longer be elicited from that site. Threshold currents and mean latency values were then determined for affective defense elicited from the ventromedial nucleus during a period of 2-3 weeks following placement of the lesion. These postlesion values were then compared with prelesion values utilizing (a) a t-test for independent observations and (b) Dunnett's test for multiple comparisons. At the completion of this experiment, the animals were prepared for 2-DG autoradiography.

95 F ~ STIMULATION OF HA ALONE

A

DUAL STIMULATION OF VM AND HA

16 14 I

A

12

I

I

t-r-

_o 10

I

>

NS

I

-1i,i

8

i

/,./1

I

NS

//A

6

I

///1

z// ///

//.4

n,i I-.-

4

¢/1 I l l Z// //Z

n

//A //A /i/I

---I

2

//A

t// /// ///

//A

p-//

///51

m

,IA ,/A .zl

I

0

//A I

2

1

3

4

HISS

B 16-

6

STRIKE

FLIGHT

[ ] STIMULATIONOF VM ALONE [ ] DUAL STIMULATIONOF HA AND VM

i4-

m

m

12--

1

n,-

o_ 1 0 .:[ -r"

i

"' 8-D

0 I---

6

//A

//.t

//.4

//.41 y/A /IA //A //.1 H ~ zzA

2

//A

//A

"//A //'A

//A

"///t

iii

2

3

5

HISS

2

6

2

STRIKE

GROWL

Fig. 1. A: ventromedial hypothalamic modulation of behaviors elicited from the anterior hypothalamus. Bar graphs indicate that concurrent stimulation of the ventromedial and anterior hypothalamus alters the latency for components of affective defense reactions but has no effect upon flight behavior elicited from the anterior hypothalamus. B: anterior hypothalamic modulation of behaviors elicited from the ventromedial hypothalamus. Bar graphs indicate that concurrent stimulation of the anterior and ventromedial hypothalamus alters the latency for components of affective defense elicited from the ventromedial hypothalamus. Numbers centered below pairs of bar graphs identify the cats used in all experiments. * P < 0.05; ** P < 0.01; *** P < 0.001, NS, not significant.

96

p4C]2-Deoxyglucose autoradiography A description of the methods, presented in the preceding study6, will be outlined briefly here. In cats anesthetized with pentobarbital (40 mg/kg, i.p.), a solution of 2-DG (150 ~Ci/3 kg cat) was injected through a cannula which had been inserted within a femoral vein. Electrical stimulation was delivered through the ventromedial hypothalamic (affective defense) electrode for 5 trials (30 s on, 30 s off) prior to the administration of the 2-DG and for 45 trials following the 2-DG injection. The current applied to the ventromedial hypothalamus was that which had been employed consistently for generating behavioral responses during the course of the lesion experiments. At the end of the stimulation period, a small radiofrequency lesion was placed (30 mA, 20 s) through the hypothalamic electrode in order to facilitate identification of the electrode tip. Then, the animal was sacrificed with an overdose of pentobarbital and perfused with 3% formalin. The histological preparation of the tissue, development of the X-ray film and analysis of 2-DG data were identical to those procedures utilized in the preceding study 5-7,23,24,26,27and are described elsewhere.

ited from the ventromedial nucleus was determined (Fig. 1B). Here again, dual stimulation of the anterior and ventromedial hypothalamic sites facilitated the occurrence of hissing responses in 4 animals (P < 0.001, Cases 1-3; P < 0.01, Case 4). In addition, the latency for the striking component of affective defense was significantly reduced in two cases (P < 0.001, Case 2; P < 0.01, Case 4) and the latency for growling responses was also reduced in one animal (P < 0.001, Case 2) during concomitant stimulation. It was further noted that subthreshold stimulation delivered to anterior hypothalamic sites, which had produced flight behavior, facilitated the occurrence of hissing (P < 0.05) and growling (P < 0.001) responses elicited from the ventromedial hypothalamus in one of two cats tested (see Cases 5 and 6).

Effects of anterior hypothalamic lesions on affective defense elicited by electrical stimulation of the ventromedial hypothalamus In 4 animals (Cases 1-4), anterior hypothalamic lesions were placed through electrodes from which affective defense was elicited (Figs. 2-4). This procedure resulted in a marked statistically significant increase in the mean latency for the hissing compo-

RESULTS

Intrahypothalamic modulation of affective defense This phase of the study was conducted to initially characterize the functional relationship between sites in the ventromedial and anterior hypothalamus. Subthreshold electrical stimulation was applied at affective defense sites within the ventromedial hypothalamic nucleus and its effect on the behavior evoked from ant'erior hypothalamic stimulation was examined (Fig. 1A). In 4 cases (Cases 1-4), the latency to hiss was significantly reduced by concurrent stimulation of the ventromedial and anterior hypothalamic regions (P < 0.001). In addition, facilitation of the occurrence of the striking component of affective defense was observed in one case (P < 0.001, Case 4). In cont.rast, the occurrence of flight responses elicited from the anterior hypothalamus in two cases was not significantly modified upon stimulation of the ventromedial hypothalamus (Cases 5, 6). Subsequently, subthreshold stimulation was applied at affective defense sites in the anterior hypothalamus and its effect upon affective defense elic-

16

[]

PRE LESION EVENTS

[]

POST LESIONEVENTS

~4 12

~ ~o

I

2

3

4

5

6

7

Fig. 2. Lesion study. Bar graphs indicate that lesions of selective regions of anterior hypothalamus can suppress affective defense reactions elicited from the ventromedial hypothalamus (Cases 1-5). Note that control lesions placed elsewhere had little effect upon affective defense reactions (Cases 6 and 7). A lesion placed within the posterior hypothalamus was also effective in suppressing affective defense reactions elicited from the ventromedial hypothalamus (Case 8). The first bar of each pair indicates the mean latency for all prelesion trials and the second bar indicates the mean latency for all postlesion trials. * P < 0.05; ** P < 0.001, NS, not significant.

97

Fig. 3. Outline of maximal extent of lesion placed in each cat described in this study. nent of affective defense elicited by stimulation of the ventromedial nucleus (P < 0.001, Case 2; P < 0.05, Cases 1, 3, 4). In another case, an anterior hypothalamic lesion was placed through an electrode from which flight behavior was elicited (Case 5). This procedure was also associated with a significant increase in the mean latency for affective defense (P < 0.05). In addition, two cases are presented below in which anterior hypothalamic lesions were not effective in modifying the latency for affective defense (Case 6, 7, Figs. 5, 6) elicited from the ventromedial hypothalamus. Special attention was also given to a case in which a posterior hypothalamic lesion was associated with an increase in the latency (P < 0.001) for affective defense elicited from the ventromedial nucleus (Case 8). Below are described the results of each of these 8 cases. Representative photographs of the stimulation sites within the ventromedial hypothalamus and corresponding lesions are presented in Fig. 7.

Anteromedial hypothalamic lesions (Cases 1-3) Case 1. The smallest of the 3 lesions (1 mm in diameter) was placed in the region of the anterior dorsal hypothalamus, situated just medial to the fornix. Tissue damage was restricted to an area within the dorsal hypothalamus but did not extend to the third ventricle or into any portion of the ventral hypothalamus. Two days following placement of this lesion, the latency for the hissing component of affective defense elicited from the ventromedial nucleus was significantly increased (P < 0.01). Additionally, an elevation in the threshold current for the response was observed on the 8th postlesion day. However, on the 10th day subsequent to the placement of the lesion, both the latency and threshold current were lowered and were not significantly different from those recorded prior to the lesion. After collection of the behavioral data, the animal

was prepared for 2-DG autoradiography to determine the pathways activated during stimulation of the ventromedial nucleus. Particular attention was placed upon the distribution of label with respect to the locus of the small anterior lesion. Results from the 2-DG procedure in this animal indicate that both the site of stimulation and resultant sphere of activation (1.5 mm in diameter) were situated in the dorsal aspect of the ventromedial nucleus. Extensive quantities of label were observed within the tissue surrounding the lesion in the anterior dorsal hypothalamus. Areas of increased optical density were also present in the anterior ventral and perifornical hypothalamus. Rostral to the level of the lesion, increases in optical density were further noted in the lateral septal nucleus, diagonal band nuclei, bed nuclei of the stria terminalis and anterior commissure and medial preoptic area. In addition, activation was observed in the medial, basal, central and cortical amygdaloid nuclei. Little or no label was observed in portions of the hypothalamus caudal to the ventromedial nucleus. Case 2. In Case 2, the lesion was located in a region of the anterior hypothalamus medial to the fornix. This lesion was larger in size than that of Case 1 and the area of tissue damaged (2 x 2 mm at widest portion) involved "both the dorsal and ventral aspects of the anterior medial hypothalamus. Following placement of the lesion, the mean behavioral latencies to hiss, growl and strike were significantly elevated (P < 0.01) and the increases in latency scores were sustained throughout all test sessions. The threshold current was also found to be elevated in the postlesion period with the greatest increase occurring on the lesion day and the second day after placement of the lesion. In the 2-DG experiment, the stimulation site and resultant sphere of activation (2.0 mm in diameter) were situated in the ventral aspect of the ventromedial nucleus. Stimulation of this ventromedial hypothalamic site resulted in the same distribution pattern of metabolic activation that was noted in the previous case. The activated regions included the lateral septum, diagonal band nuclei, bed nuclei of the stria terminalis and anterior commissure and medial preoptic area. Large quantities of label were observed in the anterior hypothalamic region adjoining the periphery of the lesion. Intense levels of activation were

98

I,:!:. I l l

J

t.,.D , ..'i [ '. ~ k d --~

'

:1: .. ";-::"": ,:.::.

\

" ,,

;<.:'-"

l

~".-.

6"

. '..,::"",

-..:~,.':..'.'i:.:: ""

|

,

'

I

,

;::" :

"~..',,~"..::'"..

IBM...:~

"

"

..:..'/ " < ~ : i " : ~ . ~ " , , J ~ ' ~ " "

,_, , - - K *

",~ 'c.o;)

-

~'.::.:.:.-<.;..

~:' .4 ..: .-

I I

i'::".\

~ ' ~' ~ _,':' L.D ~ ~".."; \_)

0 _ ~,

,,

\~llr

,,........ :<....

._.Q,

"

18 - A

v

O "r" laJ m

t.A~

~

16

8hl

1.0

~

.---1"71 X~

15"-X

]

u~-,,

I"

.,= =~=.."

"I4--~I~P

"r

< 0,01

---'"

,-'~b

2 0 ,--19

|

-

'

,, ~

,.,

'~ I~

|

'"

/) l : . f l t ; A C e ( , , ',::::;,.,..,,'..~;,'---.4";..-::.".-f:,:,.".;; J / ","H ,-'x( ' ,'~.~.:. : k "~-~<..:~~. . . . . . ~ / ,, ' . . 4 , , .. :, ..-~ , . . ~ ~ . : ~ l t _ _..:.'.-::.~,:.

..::.......~.:..

' ' : " '"' " "

Jlco

...'..~:..,

... :~;~:i.0

...:1...;...:..,/

~

--;

:.:.:--'\

°

"

.9

-.~

.8

~

~

[]

0 .7

13--

0 C

\

12-11--

.6

::0

10-9--

8-z

7

~

6

~

.3

.2 .1

II

l l l l l l l l i l l l l , t l l i l l 0

4-

B

12

16

PRELESION

20

24

28

0.0 32

Pos'rLESION TIME ( D A Y S )

Fig. 4. Anterior medial hypothalamic lesion (Case 2). A lesion placed within the anteromedial hypothalamus resulted in an increase in the latency and threshold for the occurrence of affective defense responses elicited from the ventromedial hypothalamus. A n increase in optical density was noted in areas situated rostral to the site of stimulation including the region of anterior hypothalamus circumscribing the lesion.

99

a

S

i •i;~.:i:,

, BST

,~.... ~ - ~ _ . , I;

t t

HL

~

%,,~/.%.,, i

~T~.:~]~i. ~,0

2°I Ig

[]

Hem ~'nZNCY

18 17

X :~

H~ cumem p < o.ol

~11o0 .g .8

.7

~)

13 12

tu m

10 g

z

=o

.8 .5

6 _ 5

V

~

.3

-

.2 3 .1

2

I 0

,,I,.,I,,,I, 0

2

4.

II 6

8

10

12

PRELESION

14.

16

18

I" 3.o 20

POST'LESION

~ME (DAYS) Fig. 5. Small anterior hypothalamic lesion (Case 6). This lesion was placed within the anterior hypothalamus. It included the suprachiasmatic nucleus and was largely ineffective in modifying the latency or threshold for affective defense responses elicited from the ventromedial hypothalamus.

100 also observed in the perifornical and anterior ventral hypothalamus. Further, an increase in optical density was exhibited in several nuclei of the amygdala which included the anterior, medial, central, basal and cortical nuclei, perhaps resulting from the activation of fibers of the ventral amygdalofugal pathway. Caudal to the site of stimulation, label extended only to the posterior pole of the ventromedial nucleus. The locus of the lesion, the position of the ventromedial hypothalamic electrode and resulting pattern of 2-DG labeling in the forebrain as well as graphs depicting the effects of the lesions upon affective defense are shown in Fig. 4. Case 3. In Case 3, a lesion of approximately 1.5 mm in diameter was confined to the region of the anterior ventral hypothalamus. The lateral extent of ti-e lesion approached the fornix and its medial extent was about 0.5 mm from the third ventricle. Anteriorly, the lesion extended to the level of the suprachiasmatic nucleus and posteriorly to about 1.0 mm from the rostral aspect of the ventromedial nucleus. The placement of this lesion was associated with an immediate increase in the latency for hissing behavior elicited by ventromedial hypothalamic stimulation (P < 0.05). In addition, the current threshold for hissing was elevated on days 13, 17 and 19 after placement of the lesion. Results of the 2-DG procedure revealed that the sphere of activation (1.3 mm in diameter) about the tip of the stimulating electrode was confined to a central portion of the ventromedial hypothalamus adjoining the lesion. It appeared that the presence of the lesion was effective in blocking activation in other regions of the limbic forebrain. Again, no label could be detected caudal to the site of stimulation. Anterior intermediolateral lesions (Cases 4, 5) Case 4. In Case 4, the lesion was placed in the region of the anterior hypothalamus immediately ventrolateral to the descending column of the fornix. The lesion diameter was approximately 1 mm and the small area of tissue damaged appeared to lie rostraUy within the anterior dorsal hypothalamus. Caudally, the area of tissue damage occupied a more ventral position just above the optic chiasm. A significant elevation in the latency (P < 0.01) and threshold current for hissing and striking occurred on the 5th postlesion day and this increase was sustained for the du-

ration of all postlesion test periods. In this case, the tip of the stimulating electrode and sphere of activation (1.3 mm in diameter) were both situated in the dorsal aspect of the ventromedial nucleus immediately adjacent to the fornix. The regions which exhibited an increase in optical density following 2-DG administration were similar to those described for Cases 1 and 2. In particular, label was noted in extensive quantities surrounding the lesion. Label was further observed in adjacent regions of the medial preoptic area, bed nuclei of the anterior commissure and stria terminalis, diagonal band nuclei, medial and lateral nuclei of the septum. Caudal to the level of stimulation, label could be followed to the premamillary region and posterior hypothalamus. Case 5. In Case 5, a large lesion of approximately 2 mm in diameter was placed in the anterior ventral hypothalamic-preoptic junction through an electrode from which flight behavior had previously been elicited. The lesion was situated just above the supraoptic nucleus and below the column of the fornix. The latency for the hissing response was elevated significantly on the day of the lesion and the elevation was sustained throughout the postlesion test sessions (P < 0.001). Following placement of the lesion, the threshold current for affective defense increased initially but ultimately returned to baseline levels on the 15th postlesion day. Similar to the findings observed in Case 3, the results obtained from 2-DG autoradiography suggested that this lesion was effective in limiting the extent of label seen within the limbic forebrain following stimulation of the dorsal aspect of the ventromedial nucleus (sphere of activation, 1.5 mm in diameter). Specifically, increases in optical density were restricted to the region of the anterior ventral hypothalamus adjacent to the medial border of the lesion. Control lesions (Cases 6, 7) Case 6. In Case 6, flight behavior was elicited from the anterior hypothalamic electrode. Prior to the placement of the lesion, dual stimulation procedures were applied but significant modulation of flight behavior from the ventromedial hypothalamus was not observed. However, the latencies for hissing (P < 0.05) and growling (P < 0.001), elicited from the ventromediai nucleus were decreased significantly by concurrent stimulation delivered to this anterior

101

¸

:":-,;"+{"+;"

',;.,

HD ( ~ j

+

,HP

(

k

(

"+~r+.

T

,"

+ ..,1

. I

',

,~....:~+

~+~

i i~

m

"

,

+

;-'

/ ~\!o'" X

-- ~ ,'

]

~'-'*

"-,"7-"C" 2, },' ~',

I/

~ ]+~F ~+-"~, ,\~,, ,,~ ,~.

~r-"

-

[ ] HISS LA'rI~cY

18

X

16

"+-~

1.O .g

HISS CURRENT .8

>1+:P < o.ol

.7

0 c I,I

~ ,o

.5

m

~

8

-.~ ~

z

.I

2

l,,,l,,,lilllI111111

iillil,lii

0

0

4

8

12

16

20

24

28

o.o 32

POSTLESION

pRIL-I I='RION

TIME (DAYS) Fig. 6. Anterior hypothalamic lesion (Case 7). The lesion placed within the anterior hypothalamic region was largely ineffective in modifying the latency or threshold for affective defense responses elicited from stimulation of both the ventromedial hypothalamus and adjoining tissue rostral to it in the anterior ventral hypothalamus. Note the presence of label within the midbrain central gray.

102

20

1.0

19 18

.B

17 16

-8

lb 14

.7

13

.8

12 11

.5

10 0

.4

8 7 8

a

HISSLATENCY

x

HISS CURRENT

.3

5 4

.2

3

.1

2 1 0

0.0 4

8

12

18

20

24

28

POSILESION

PRELESION TIME (DAYS)

Fig. 7. Posterior hypothalamic lesion (Case 8). A lesion placed within the posterior dorsal hypothalamus resulted in an increase in the latency and threshold for the occurrence of affective defense reactions elicited from the ventromedial hypothalamus. It is likely that this lesion disrupted fibers arising from the anterior hypothalamus which descend to the region of the midbrain central gray. Note extensive labeling at levels principally rostra1 to the site of stimulation and within the medial preopticohypothalamus.

103 site. In this case (Fig. 5), a small lesion was placed in a region of the anteromedial hypothalamus immediately above the optic chiasm. The lesion (1 mm in diameter) included the extreme ventromedial aspect of the anterior hypothalamus as well as a portion of the suprachiasmatic nucleus. With the exception of 1 day, postlesion latency and threshold current values for hissing behavior were not significantly altered from prelesion values. The pattern of activation observed with 2-DG autoradiography following stimulation of the dorsal aspect of the ventromedial nucleus (sphere of activation, 1.5 mm in diameter) involved the medial anterior preoptico-hypothalamus, including tissue situated well beyond the site of the lesion. Specifically, the distribution of label was observed rostral to the stimulating electrode. It included the medial preoptic region, bed nuclei of the stria terminalis and anterior commissure, vertical and horizontal limbs of the diagonal band and lateral septal area. Case 7. In Case 7, the tip of the stimulating electrode was located on the border of the ventromedial nucleus and anterior hypothalamus. On dual stimulation trials, concurrent stimulation of the anterior hypothalamus was observed to facilitate the occurrence of affective defense elicited from the anterior-ventromedial hypothalamic site (P < 0.001) and stimulation of the anterior-ventromedial hypothalamic site facilitated the occurrence of affective defense elicited from the anterior hypothalamus (P < 0.001). In this animal (Fig. 6), a small oblong lesion (1 x 2 mm), was placed in the extreme ventromedial aspect of the anterior hypothalamus. Again, as indicated in Case 6, postlesion latency and threshold current values for affective defense were largely unchanged with the exception of the first postlesion day. The sphere of activation around the stimulating electrode (3 mm in diameter) was somewhat larger than those observed in previous cases and included the entire ventromedial nucleus as well as the anterior hypothalamus. Large increases in optical density appeared both around the small lesion, as well as throughout the limbic forebrain as described in previous cases. These labeled regions include the preoptic area, bed nuclei of the stria terminalis and anterior commissure, ventrolateral septum and diagonal band nuclei. In contrast to the descriptions noted in earlier cases, marked increases in label were noted

caudal to the site of stimulation. Specifically, label was observed in the dorsal hypothalamus, posterior hypothalamus, premammillary region, ventral tegmental area and midbrain central gray. In this manner, these results were more typically characteristic of the 2-DG findings observed following stimulation of sites in the anteromedial preopticohypothalamus6,9.

Effect of a posterior hypothalamic lesion upon affectire defense elicited from the ventromedial hypothalamus (Case 8) Flight behavior was initially elicited from the posterior dorsal hypothalamus. However, this response was not modified by concurrent stimulation of an affective defense site in the ventromedial nucleus. Similarly, the latencies for affective defense behavior weJe also unaffected by concurrent stimulation of the

Fig. 8. Autoradiographsof stimulation and lesion sites. (A and B, Case 7): A, stimulation site (triangle) in the ventromedial and anteromedial hypothalamus with its corresponding sphere of activation and, B, the locus of a lesion (arrow) placed in the anterior hypothalamus and surrounding pattern of activation. Note that the activated tissue at B extends well beyond the limits of the lesion whichwas ineffective in blockingattack; (C and D, Case 2). C, sphere of activation around the stimulation site (triangle) in the ventromedial nucleus; D, the focus of an effective lesion (arrow) placed in the anterior medial hypothalamus with surrounding pattern of activation. Note that the activated tissue at D is limited to the region immediately surrounding the lesion, thus suggesting that this lesion had successfully damaged the overwhelming majority of fibers activated following stimulation of the ventromedial nucleus.

104 flight-eliciting electrode. In this case, a lesion was placed through the posterior dorsal hypothalamic electrode (Fig. 7). Here the lesion (2 mm in diameter) was highly effective in that the latency and threshold for hissing elicited from the ventromedial hypothalamus were significantly elevated (P < 0.001). The increase in mean latency and threshold for affective defense occurred immediately after placement of the lesion and was sustained throughout the testing period. In this case, the pattern of activation, observed with the 2-DG autoradiographic method, was typical for stimulation of the ventromedial nucleus (sphere of activation, 1.5 mm diameter). Specifically, label was noted mainly rostral to the site of stimulation - in the preopticohypothalamus, and bed nuclei of the stria terminalis and anterior commissure. In contrast, no label could be significantly followed beyond the posterior aspect of the ventromedial nucleus. It should be noted that label did not appear to surround the area occupied by the lesion. Therefore, it is likely that this lesion was effective in modifying the latency and threshold values by damaging, instead, the second-order hypothalamic-central gray fibers which arise from the anterior hypothalamus and which descend through the region occupied by the lesion in the posterior hypothalamus. DISCUSSION

Characterization of hypothalamic components subserving affective defense: dual stimulation experiments The results of stimulation of either the anterior or ventromedial hypothalamus revealed a marked similarity in the appearance of the affective defense response elicited from each of these structures. This observation reinforces the view that the ventromedial and anteromedial hypothalamus form a common link in the organization of the fiber system(s) subserving affective defense behavior. In support of this position, a reduction in the latency for components of affective defense (but not flight behavior) was consistently observed following concurrent stimulation of the anterior or ventromedial hypothalamus. One possible explanation for this observation is that dual stimulation of these two hypothalamic structures involves an increase in the total number of cell groups

and fibers activated within the system subserving affective defense.

Effect of hypothalamic lesions upon affective defense elicited from the ventromedial hypothalamus In 5 cases (Cases 1-5), anterior hypothalamic lesions significantly increased the latency and threshold for affective defense elicited from the ventromedial hypothalamus. This finding is consistent with the view that the pathway from the ventromedial nucleus to the midbrain central gray involves a synapse in the anterior hypothalamus. The region of the anterior hypothalamus damaged by the effective lesions appears to occupy a position where the greatest concentration of ventromediai hypothalamic fibers terminate 6,9,27. Additionally, it was previously established6,9, 27 that stimulation of the anterior hypothalamus is associated with the direct activation of neurons which supply the central gray - - a structure which is critical for the elicitation of affective defense behavior 2,11,13A4,20,22. Therefore, it is likely that the effective lesions disrupted both the rostrally-directed projections from the ventromedial hypothalamus as well as, the neurons in the anterior hypothalamus which innervate the midbrain central gray. Results from previous studies3,4,7-10,23-27 indicate that brain mapping involving electrical stimulation coupled with 2-DG autoradiography is generally limited to the activation of regions situated not beyond the first synapse. In cases where small effective anterior hypothalamic lesions were placed, a distribution pattern of 2-DG label was nevertheless noted within several components of the limbic forebrain, including the medial amygdala, lateral septal nucleus, diagonal band nuclei and bed nucleus of the stria terminalis. This finding suggests that the first-order projections to these structures from the ventromediai hypothalamus may not be directly involved in the initiation of affective defense behavior. Instead, such fibers may possibly contribute to part of a feedback system involved in the modulation of this behavior 3.19.22. These results further suggest that the small anterior lesions may have been effective in blocking affective defense by disrupting critical groups of fibers comprising the overall circuit from the ventromedial nucleus to the central gray substance via the anterior hypothalamus. In Cases 6 and 7, little effect upon affective de-

105 fense behavior was observed following the placement of the lesions. Specifically, in Case 6, the lesion was confined to a small area of the anterior hypothalamus which included a portion of the suprachiasmatic nucleus. Prior to the placement of this lesion, stimulation of that site elicited flight behavior which could not be modulated from the ventromedial nucleus. This observation suggests that this lesion had little or no effect upon the hypothalamic circuit mediating affective defense. In Case 7, as determined from the quantity of 2-DG label present under the stimulating electrode, the activated tissue appeared to include neurons of both the ventromedial and anterior hypothalamus. Indeed, the 2-DG results revealed that, in spite of the placement of the lesion, label could be followed from the anterior hypothalamus to the midbrain central gray. Accordingly, this finding indicates that the anterior hypothalamic lesion was ineffective in altering the behavior because the primary connections from the anterior hypothalamus to the central gray were spared. In Case 8, a lesion was placed into the posterior dorsal hypothalamus which resulted in a sustained increase in the threshold and latency for the elicitation of affective defense from the ventromedial nucleus. This observation parallels the finding described by Bergquist I who placed lesions within the dorsomedial hypothalamus of the opossum and also reported an increase in the threshold for threat behavior. From our 2-DG results, it appears that the area of tissue damaged by the lesion lies within the pathway of anterior hypothalamocentral gray fibers6,9, 27 which would thus account for the clearcut elevation in threshold observed in this case. Several other observations related to the effects of anterior hypothalamic lesions upon affective defense behavior should be noted. One issue relates to the fact that the effective lesions did not completely block affective defense responses. While it is likely that elevations in response latencies and thresholds were the direct result of a disruption of neurons associated with affective defense reactions, the overall loci and sizes of these lesions were presumably insufficient to have completely destroyed all of the neurons in the anterior hypothalamus which participate in the organization of this response. In support of this notion, our 2-DG findings indicate that ventromedial hypothalamic stimulation in most instances activated

a sizeable number of axons whose rostral course extended to sites adjacent to the peripheral limits of the lesions. Thus, the application of higher currents to sites in the ventromedial nucleus probably served to recruit additional neurons which contributed to the overall response pattern but which were unaffected by the lesions. A second issue concerns the recovery of function noted over a period of 2 - 3 weeks in several of the animals examined. It is conceivable that such recovery could have been the result of the sprouting of new functional connections form the ventromedial to anterior hypothalamus. Such a possibility is consistent with the known time course for the development of new connections within the limbic system 16. Alternatively, it is also possible that, over time, existing connections between the ventromedial and anterior hypothalamus develop a higher functional efficiency. Unfortunately, support for this view is generally lacking in the literature.

Significance of the connections of the anterior hypothalamic region mediating affective defense Inputs to the anterior hypothalamus from key limbic structures such as the amygdala and septal area 15, 17-19,21,22 have been associated with the control of affective defense behavior. By virtue of the direct connections with the brainstem central gray substance, the anterior hypothalamus provides the principal efferent pathway to the midbrain effector area for this response. In the present study, the large increases in the latency and threshold for affective defense elicited from the ventromedial hypothalamus following appropriately placed anterior hypothalamic lesions suggest that the rostral connections of the ventromedial hypothalamus are fundamental to the organization of this response. Therefore, the position of the anterior hypothalamus as an interneuronal link between the ventromedial hypothalamus, limbic forebrain, and brainstem central gray may provide the anatomical basis for its function as a central integrating region mediating affective defense behavior. ACKNOWLEDGEMENTS This study was supported by N I H Grant NS 0794115 awarded by the National Institutes of Neurological and Communicative Diseases and Stroke and by a grant from the Harry Frank Guggenheim Foundation.

106 ABBREVIATIONS A AC ACC BAC BM

BP BST CC CD CE CG CM CO CP DB F HA HD HI HL HP IC L

anterior amygdaloid nucleus anterior commissure accumbens nucleus bed nucleus of the anterior commissure basal amygdaloid nucleus, pars magnocellularis basal amygdaloid nucleus, pars parvicellularis bed nucleus of the stria terminalis corpus callosum caudate nucleus central amygdaloid nucleus central gray nucleus centralis medialis cortical amygdaloid nucleus cerebral peduncle diagonal band fornix anterior hypothalamus dorsal hypothalamus habenulo-interpeduncular tract lateral hypothalamus posterior hypothalamus internal capsule lesion

REFERENCES 1 Bergquist, E. H., Output pathways of hypothalamic mechanisms for sexual, aggressive, and other motivated behaviors in opossum, J. comp. physiol. Psychol., 70 (1973) 389-398. 2 Berntson, G. G., Blockade and release of hypothalamically and naturally elicited aggressive behaviors in cats following midbrain lesions, J. comp. physiol. Psychol., 81 (1972) 541-554. 3 Brutus, M., Shaikh, M. B., Siegel, H. E. and Siegel, A., An analysis of the mechanisms underlying septal area control of hypothalamically elicited aggression in the cat, Brain Research, 310 (1984) 235-248. 4 Brutus, M., Watson, R. E., Shaikh, M. B., Siegel, H. E., Weiner, S. and Siegel, A., A [14C]2-deoxyglucose analysis of the functional neural pathways of the limbic forebrain in the rat. IV. The mediodorsal thalamic nucleus and associated cortical structures, Brain Research, 310 (1984) 279-293. 5 Fuchs, S. A. G., Dalsass, M., Siegel, H. E. and Siegel, A., The neural pathways mediating quiet biting attack behavior from the hypothalamus in the cat: a functional autoradiographic study, Aggress. Behav., 7 (1981) 51-68. 6 Fuchs, S. A. G., Edinger, H. M. and Siegel, A., The organization of the hypothalamic pathways mediating affective defense behavior in the cat, Brain Research, 330 (1985) 77-92. 7 Fuchs, S. A. G. and Siegel, A., Neural pathways, mediating hypothalamically-elicited flight behavior in the cat, Brain Research, 306 (1984) 263-281. 8 Fuchs, S. A. G., Siegel, A. and Edinger, H. M., Functional mapping of hypothalamic pathways eliciting aggression and escape behavior in the cat: a radioautographic study, Soc. Neurosci. Abstr., 6 (1980) 420. 9 Fuchs, S. A. G., Siegel, A. and Edinger, H. M., Neural circuitry subserving affective display elicited from the feline

LH LT M MA MD MM MP MT OC OT POA PT PV RE RN S SC SD SM SN SO ST VM VT III

lateral habenula lateral amygdaloid nucleus medial amygdaloid nucleus anterior mammillary nucleus mediodorsal nucleus medial mammitlary nucleus premammillary nucleus mammillothalamic tract optic chiasm optic tract preoptic area parataenial nucleus paraventricular nucleus nucleus reuniens red nucleus septum suprachiasmatic nucleus supramammillary decussation stria medullaris substantia nigra supraoptic nucleus stria terminalis ventromedial nucleus ventral tegmentum ventricle

hypothalamus, Soc. Neurosci. Abstr., 8 (1982) 973. 10 Fuchs, S. A. G., Siegel, A. and Edinger, H. M., The neural basis of escape behavior elicited from the feline hypothalamus, Soc. Neurosci. Abstr., 7 (1981) 755. 11 Glusman, M., The hypothalamic savage syndrome. In S. Frazier (Ed.), Aggression, Res. Publ. Assoc., Nerv. Ment, Dis., 1974, pp. 52-92. 12 Grossman, S. P. and Grossman, L., Surgical interruption o1 the anterior and posterior connections of the hypothalamus: effects on aggressive and avoidance behavior, Physiol. Behav., 5 (1970) 1313-1317. 13 Hunsperger, R. W., Affektreaktionen auf electrische reizung im hirnstamm der Katze, Helv. physiol. Acta, 14 (1956) 70-92. 14 Hunsperger, R. W., Role of the substantia grisia centralis mesencephalis in electrically induced rage reactions, Progr. Neurobiol., 1 (1956) 289-294. 15 Pott, C. B., Kramer, S. Z. and Siegel, A., Inhibition and facilitation of hypothalamically-elicited hissing by central gray stimulation are differentially sensitive to naloxone, Soc. Neurosci. Abstr., 9 (1983) 278. 16 Raisman, G., Neural plasticity in the septal nuclei of the adult rat, Brain Research, 32 (1971) 201-219. 17 Shaikh, M. B., Brutus, M., Siegel, H. E. and Siegel, A., Differential control of aggression by the midbrain, Exp. Neurol., 83 (1984) 436-442. 18 Siegel, A., Anatomical and functional differentiation within the amygdala-behavioral state modulation. In R. J. Bandler, (Ed.), Modulation of Sensorimotor Activity during Altered Behavioral States, Allan Liss Inc., New York, 1984, pp. 299-323. 19 Siegel, A. and Skog, D., Effects of electrical stimulation of the septum upon attack behavior elicited from the hypothalamus in the cat, Brain Research, 23 (1970) 371-380. 20 Skultety, F. M., Stimulation of periaqueductal gray and hypothalamus, Arch. Neurol., 8 (1963) 608-620. 21 Smith, D. A. and Flynn, J. P., Afferent projections to affec-

107 tive attack sites in cat hypothalamus, Brain Research, 194 (1980) 41-51. 22 Stoddard-Apter, S. L. and MacDonnell, M. F., Septal and amygdala efferents to the hypothalamus which facilitate hypothalamically elicited intraspecific aggression and associated hissing in the cat: an autoradiographic study, Brain Research, 193 (1980) 19-32. 23 Watson, R. E., Edinger, H. M. and Siegel, A., An analysis of the mechanisms underlying hippocampal control of hypothalamically-elicited aggression in the cat, Brain Research, 269 (1983) 327-345. 24 Watson, R. E., Edinger, H. M. and Siegel, A., A [14C]2-deoxyglucose analysis of the functional neural pathways of the

limbic forebrain in the rat. III. The hippocampal formation, Brain Res. Rev., 5 (1983) 133-176. 25 Watson, R. E., Siegel, H. E. and Siegel, A., Functional correlates of the septal area in the rat, Soc. Neurosci. Abstr., 8 (1982) 970. 26 Watson, R. E., Troiano, R., Poulakos, J., Weiner, S., Block, C. H. and Siegel, A., A [14C]2-deoxyglucose analysis of the functional neural pathways of the limbic forebrain in the rat. I. The amygdala, Brain Res. Rev., 5 (1983) 1-44. 27 Watson, R. E., Troiano, R., Poulakos, J., Weiner, S. and Siegel, A., A [14C]2-deoxyglucose analysis of the neural pathways of the limbic forebrain in the rat. II. The hypothalamus, Brain Res. Bull., 8 (1982) 459-476.